Galactose metabolism in human blood cells

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content GALACTOSE METABOLISM IN HUMAN BLOOD CELLS by Won Gin Ng A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Biochemistry) June, 1964 UNIVERSITY OF S O U T H E R N CALIFORNIA THE GRADUATE S C H O O L UN IVERSITY PARK LOS ANGELES. C A LIF OR N IA 9 0 0 0 7 This dissertation, written by Won Gin i. W t under the direction of hS3....Dissertation C o m mittee, and a p p r o v e d by all its members, has been presented to and accepted by the Graduate School, m partial fulfillment of requirements for the degree of D O C T O R O F P H I L O S O P H Y Dean June , 196I4. D a te.................................. DISSERTATION COMM ITTE E X u i.il ChairmatlS ACKNOWLEDGMENTS This dissertation is dedicated to ray wife, Rose, for her continuous encouragement and understanding which made the present study possible. The work was done under the supervision of Professor Bergren, to whom I am deeply indebted for invaluable guidance and stimulation throughout the course of this study. To Professor Donnell, who provided clinical material for this work, I wish to express my appreciation for his constant interest and for many helpful suggestions. My gratitude is also extended to Professor Kushinsky for his constructive ideas. I am thankful to Professors Mehl, Marx, Visser, Saltman, Heinrich, and other members of the faculty of the Department of Biochemistry and Nutrition. Their instruc tion, friendly interest, and cooperation during my graduate years are greatly appreciated. I would like also to thank Mrs. Eleanor James for her help during the past several years. For the excellent technical assistance of Mrs. Grace Perry and Miss Betty Knoch I am especially grateful; also for Mr. Max Fields' advice regarding the use of radioisotopes; and for Mrs. Rosemary Lilly's help in the preparation of this dissertation. The Biochemistry Research Laboratory of Childrens Hospital provided the facilities that were utilized for this study. During my tenure as a predoctoral fellow I was supported on a Fellowship from the Division of General Medical Sciences, National Institutes of Health, Department of Health, Education, and Welfare. PREFACE Galactosemia is a genetically determined disease characterized by a lack of galactose-l-phosphate uridyl transferase activity. Consumption of a galactose- containing diet by affected individuals causes an accumu lation of galactose-l-phosphate in tissues, and it has been presumed that the presence of this compound results directly or indirectly in the manifestations of the disease, which include mental retardation, formation of cataracts, susceptibility to infection, amino aciduria, and various other abnormalities. Diagnosis of the disease now can be confirmed readily by measurement of galactose-l-phosphate uridyl transferase in erythrocytes. Treatment by use of a diet which excludes milk products and other galactose- containing foods prevents appearance of the manifestations of the disease and permits normal development of an affected individual. However, various questions remain to be answered. First, is there biochemical heterogeneity in the disease? Second, is the lack of transferase activity in an affected individual complete? Third, is iv V there a possibility that an affected individual can acquire a degree of tolerance to galactose with increasing age, via alternate pathways or by some other mechanism? The preseht study has been directed to an investigation of these points. It has been shown by other workers that a limited number of galactosemics can produce labeled carbon dioxide in vivo from radioactive galactose, and further that this capability is reflected by a similar ability of blood cells in vitro. Such biochemical heterogeneity was demonstrated in the group of patients employed in the present study: blood cells from three among seventeen affected individ uals metabolized galactose at a markedly higher rate than the fourteen others. Further, as determined by column chromatography, the blood cells from all of the galactosemics were able to produce small, but identifiable quantities of UDPGalactose and UDPGlucose. For the three unusual individuals, the amounts of UDPHexose were larger. Alternate pathways for the metabolism of galactose, such as by way of a UDPGalactose pyrophosphorylase or via galactose-6-phosphate, have been suggested by others. No evidence could be found to indicate that such alternatives could account for the UDPHexose found in our experiments. vi It was concluded that the lack of galactose-l-phosphate uridyl transferase activity in galactosemics is not complete. On the basis of the present data, it is not possible to state whether the enzyme activity considered to be present represents a quantitative deficiency in a normal enzyme or the production of an abnormal enzyme with different kinetic characteristics. It is considered that the results found with the three unusual galactosemics represent an inherited capability rather than an acquired one. TABLE OF CONTENTS PAGE INTRODUCTION ........................................... 1 Galactose metabolism ................................ 1 Galactosemia ...................................... 9 Alternate pathways of galactose metabolism . . . 13 Background of the investigation .................. 20 STATEMENT OF THE PROBLEM AND PLAN OF ATTACK .... 25 Statement of the problem........................... 25 Plan of attack.................................... 28 MATERIALS AND METHODS................................. 33 Materials............................................. 33 Human material.................................... 33 Reagents........................................... 34 Enzymes............................................. 34 Uridine diphosphate glucose dehydrogenase . . 35 Galactokinase .................................... 37 Preparation of galactose-l-phosphate ......... 42 Purity of galactose............................. 42 Purity of galactose-l-C^...................... 45 viii PAGE Preparation of unlabeled galactose-l- phosphate ................................... 50 Preparation of galactose-l-C^-l-phosphate . 58 Equipment........................................ 65 Methods............................................ 66 Chemical determinations ........................ 66 Reducing sugar ............................... 66 Submicro sugar determination ............... 66 Glucose oxidase determination of glucose . . 67 Inorganic phosphorus ........................ 67 Lactic acid................................... 67 Protein........................................ 68 Enzymatic Assays ............................... 68 Lactic acid................................... 68 Galactose-l-phosphate ........................ 69 Galactose-l-phosphate uridyl transferase (UDPG consumption assay) ................. 70 Galactose-l-phosphate uridyl transferase (Radioactive assay) ........................ 71 Separation of galactose-l-phosphate uridyl transferase from hemoglobin ............... 80 ix Assay of UDPGal pyrophosphorylase and of UDPG pyrophosphorylase in hemolysates ........... 82 Assay of UDPG pyrophosphorylase and of UDPGal pyrophosphorylase in human liver ........... 84 Galactokinase (yeast). . 86 Galactokinase (hemolysates) .................... 87 Assay of hexokinase (in galactokinase preparation) ................................. 90 Paper Chromatography............................. 91 Dowex-l-Formate Column Chromatography ........... 93 Labeled Carbon Dioxide from Galactose-l-C^^ . . 95 Isolation of Glycolytic Intermediates from Incubation of Blood Cells .................... 96 Identification of Metabolic Intermediates of Galactose........................................ 97 Incubation of Blood Cells for Extended Periods 100 RESULTS................................................ 102 Oxidation of Galactose-l-C^ by Whole Blood from Normal and from Galactosemic Subjects ........... 104 Carbohydrate Pathway Intermediates ............... 110 Galactose-l-Phosphate Uridyl Transferase Activity in Hemolysates................................... 124 X PAGE UDPGalactose Pyrophosphorylase Ac ivity in Hemolysates.......... 131 Subsidiary Studies of Interest ................. 145 Rate of conversion of galactose-l-C^ into glycolytic intermediates ................... 145 Galactose pathway enzymes in human liver . . . 158 Galactokinase in human erythrocytes ........... 161 Mechanisms for disposal of galactose-l- phosphate by erythrocytes in v i t r o .......... 168 Chromosome 21 and galactose-l-phosphate uridyl transferase................................... 178 DISCUSSION........................................... 181 CONCLUSIONS......................................... 194 APPENDIX A ........................................... 198 Reagents........................................... 198 APPENDIX B ........................................... 201 Equipment......................................... 201 BIBLIOGRAPHY......................................... 204 LIST OF TABLES TABLE PAGE I. Assay of Galactose-l-Phosphate by Chemical Analysis of Acid Hydrolysate............. 56 II. Recovery of UDPG after Charcoal Adsorption in Terms of Optical Density............. 74 III. Oxidation of Galactose-l-C^ by Whole Bloods from Galactosemic and Normal Individuals . 105 IV. Oxidation of Galactose-l-C^1 ^ by Galactosemic Whole Bloods with Reference to Family Studies.................................... 109 V. Effect of UDPG and UTP on the Incorporation of Gal-l-C^-l-P into UDPGal-1-C^ in a Normal Hemolysate .......................... 126 VI. Erythrocyte Galactose-l-Phosphate Uridyl Transferase Determined by Labeled Galactose Procedure ................... 129 VII. Incorporation of Gal-l-C^-l-P into labeled UDPGal in the Presence of UTP in Hemolysates................................ 134 xii TABLE PAGE VIII. Separation of Enzymatic Activity by the Use of a DEAE Cellulose Column........... 135 IX Activity of UDPG Pyrophosphorylase in Hemo lysates from a Normal and a Galactosemic Subject.................................... 138 X. Inhibition of UDPG Pyrophosphorylase by Inorganic Phosphate ...................... 140 XI. Influence of Unlabeled Glu-l-P on the Incor poration of Gal-l-C^-l-P into UDPGal-1- in the Presence of UTP and Hemolysates 141 XII. Pyrophosphorylase Assay .................... 143 XIII. Activity of Gal-l-P Uridyl Transferase, UDPGal Pyrophosphorylase and UDPG Pyro phosphorylase in Human Liver Preparations 160 XIV. Cell Viability, pH, Glucose Concentration and Lactate Concentration in a 24-Hour Incubation of Normal Whole Blood ......... 172 LIST OF FIGURES FIGURE PAGE 1. Schematic Representation of Galactose Metabolism .................................... 8 2. An Alternate Pathway of Galactose Metabolism . 16 3. Purity of Galactose-l-C^................ 46 4. Radioautogram of Galactose-l-C^ Samples after Paper Chromatography..................... 47 5. Contaminants of Commercial Galactose-l-C^ . . 49 6. Products of Incubation Mixture of Galacto- kinasejGalactose, and A T P .............. 53 7. Paper Chromatogram of Supernatant Fluid from Galactose-l-C^ Incubation.............. 60 8. Radioactive Products Isolated from Supernatant Fluid of Galactose-l-C^ Incubation .... 62 9. Glycolytic Intermediates of Galactose with a Normal Blood ................................. 112 10. Glycolytic Intermediates from a Galactosemia Heterozygote Blood Incubated with Galactose-l-C^......................... 114 xiv FIGURE PAGE 11. Glycolytic Intermediates of Galactose with a "Usual" Galactosemia Blood ............... 116 12. Glycolytic Intermediates of Galactose with an "Unusual" Galactosemia Blood ............. 117 13. Column Chromatography of Galactose Intermediates ............................... 119 14. Utilization of Galactose-l-Phosphate in a < Normal Hemolysate .......................... 128 15. Comparison of Galactose-l-Phosphate Uridyl Transferase Assays by the Two Methods Employed................................... 131 16. Glycolytic Intermediates of Glucose Incubated with Whole Blood.................. 147 17. Glycolytic Intermediates of Galactose with Galactosemia Heterozygote Washed and Unwashed Blood Cells ...................... 149 18. Glycolytic Intermediates of Glucose with Washed Blood................................. 150 19. Glycolytic Intermediates of Glucose Incu bated with Washed Blood of a Heterozygous Subject...................................... 152 XV FIGURE PAGE 20. Glycolytic Intermediates of Glucose Incu bated with Washed Blood of a Galactosemic Subject...................................... 154 21. Timed-Labeling Patterns of Galactose Inter mediates of Normal Erythrocytes ......... 156 22. Timed-Labeling Patterns of Galactose Inter mediates of Normal Erythrocytes ......... 157 23. Effect of Exogenous Mg++ on Galactose-1- Phosphate Formation ........................ 164 24. Formation of Galactose-l-Phosphate at Optimum Concentration of Mg++...................... 166 25. Influence of Age on Galactokinase Activity . 167 26. Distribution of Radioactivities During 24-Hour Incubation of a Galactosemia Blood 175 LIST OF ABBREVIATIONS ADP, adenosine diphosphate ATP, adenosine triphosphate DAP, dihydroxyacetone phosphate DPG, 2,3-diphosphoglyceric acid F-6-P, fructose-6-phosphate F-l,6-diP, fructose-1,6-diphosphate Gal, galactose Gal-l-P, galactose-l-phosphate Gal-6-P, galactose-6-phosphate Gal-1,6-diP, galactose-1,6-diphosphate Glu, glucose Glu-l-P, glucose-l-phosphate Glu-6-P, glucose-6-phosphate Glu-1,6-diP, gluccse-1, 6-diphosphate Gly-3-P, glyceraldehyde-3-phosphate MPG, monophosphoglycerate NAD, nicotinamide adenine dinucleotide NADH, nicotinamide adenine dinucleotide (reduced) NADP, nicotinamide adenine dinucleotide phosphate NADPH, nicotinamide adenine dinucleotide phosphate (reduced) PEP, phosphoenolpyruvate 6-PGL, 6-phosphogluconolactone Pi, inorganic phosphate PP, inorganic pyrophosphate T-6-P, tagatose-6-phosphate T-l,6-diP, tagatose-1,6-diphosphate TTP, thymidine triphosphate UDPG, uridine diphosphate glucose UDPGA, uridine diphosphate glucuronic acid UDPGal, uridine diphosphate galactose UDPGalA, uridine diphosphate galacturonic acid UTP, uridine triphosphate INTRODUCTION Galactosemia is a hereditary disease characterized by the inability of the organism to utilize galactose. Its early clinical manifestations include vomiting, diarrhea, jaundice, hepatomegaly, splenomegaly and poor weight gain. Cataracts often develop, and if the indi vidual survives mental retardation is a usual consequence. In retrospect, the first recognizable case was described by von Reuss in 1908 (1). Gradually, other cases with similar features began to appear in the literature, and eventually the genetic aspects were recognized. Galactosemia is an experiment of nature which can be employed to study details of galactose metabolism in the human. The availability of a large group of families in which this disease occurs has led to the present investigation. Galactose Metabolism Galactose is a stereoisomer of glucose, differing in the asymmetry of the carbon atom at the fourth position, where the hydroxyl group lies on the opposite side of the 2 plane. cH^oH e//,OV H oh tr o h cX.-D-Galactose X-D-Glucose It is a constituent of many biologically important com pounds in both the animal and the plant world, such as lactose, melibiose, raffinose, stachyose, polysaccharides, cerebrosides, glycolipids and glycoproteins. Early investigations on carbohydrate metabolism indicated that galactose could serve as an energy source (2). By man, it is consumed primarily as a constituent in lactose. liver glycogen increased when either lactose or galactose was administered to fasting animals (3). The pathway for the formation of glycogen from galactose was first explored by Kosterlitz who isolated galactose-l-phosphate from the liver of rabbits which had been fed galactose (4). He also isolated this compound from yeast which had been adapted to galactose (5). In the latter case, he sug gested that two new enzymes were formed during the period of adaptation and that they participated (a) in the forma tion of galactose-l-phosphate and (b) in the conversion of galactose-l-phosphate to glucose-6-phosphate through Experiments on rats demonstrated that the level of 3 glucose-l-phosphate as an intermediate. Under this hypothesis, glycogen would be formed by the action of phosphorylase (6) . The work of Kosterlitz was confirmed by Caputto and co-workers (7). Extracts of the galactose-fermenting yeast Saccharomyces fragilis were found to contain an enzyme (galactokinase) which catalyzed the phosphorylation of galactose in the presence of adenosine triphosphate. Further, uridine diphosphate glucose was shown to be required for the conversion of galactose-l-phosphate to glucose-l-phosphate (8). This finding paved the way for recognition of the important role played by a number of nucleotide-sugar compounds in carbohydrate metabolism <9, 10). Following the isolation of uridine diphosphate glucose, additional steps in the pathway of galactose metabolism were described. Leloir (11), upon incubation of his crude enzyme from the yeast S^. fragilis with uri dine diphosphate glucose, demonstrated the formation of uridine diphosphate galactose. Equilibrium was reached when 25% of uridine diphosphate glucose was converted to uridine diphosphate galactose. To the enzyme involved he gave the name "Waldenase," considering the change at 4 carbon 4 to be a Walden inversion. Kalckar et_ a^L. (12) found that uridine triphos phate increased the conversion of galactose-l-phosphate to glucose-l-phosphate in extracts of the same yeast. Identical metabolic sequences of galactose and its deriva tives were also demonstrated in Lactobacillus bulgaricus by Hansen and co-workers (13, 14, 15) and in Saccharomyces maxianis by Garner and Grannis (16). In recent years, metabolic conversions of galac tose and its derivatives have been carried out by more refined enzyme systems (17, 18, 19). Further, additional pathways for the utilization of uridine diphosphate galactose and related compounds ?e been demonstrated in mung bean seedlings by Hassid and co-workers (20, 21, 22, 23, 24, 25, 26, 27, 28). These include the formation and interconversions of uridine diphosphate galacturonic acid, uridine diphosphate glucose, uridine diphosphate glucuronic acid, uridine diphosphate xylose, and uridine diphosphate arabinose. All of these compounds are incorporated into pectin and hemicellulose. It has been demonstrated that the conversion of galactose to glucose in the mammalian system follows the same pathway as in £>. fragilis. An early contribution to this understanding was the work of Stetten £t al.(29, 30), who showed that the in vivo biological transformation of D-galactose to D-glucose of glycogen proceeded via a direct epimerization at carbon atom 4 without rupture of the carbon skeleton of the molecule. Schwarz et_ a_l. (31, 32, 33) showed that the erythrocytes of galactosemic subjects accumulated large amounts of galactose-l-phosphate both in vivo and in vitro. Chemical analyses of autopsy samples also revealed the accumulation of this compound in lens tissue, kidney, brain, tongue, adrenals, heart, and liver (32, 34). It was speculated that the metabolic defect must involve an enzymatic reaction in which galactose-l-phosphate is transformed to a further intermediate. The normal enzymatic conversions of galactose to glucose in yeast, bacteria, and mammalian tissues, well- documented by the work of Kosterlitz, Leloir, Hansen, Kalckar, and others (5, 8, 13, 35), may be summarized as follows: 6 Galactokinase (1) Galactose 4- ATP » Gal-l-P + ADP Gal-l-P uridyl transferase (2) Gal-l-P + UDPG <— Glu-l-P + UDPGal UDPGal-4-epimerase (3) UDPGal < > UDPG Sum: Galactose + ATP----> Glu-l-P + ADP The enzymatic defect in galactosemia was finally estab lished by Kalckar and his group to be a lack of galactose 1-phosphate uridyl transferase (step 2), in studies done on liver and erythrocytes of affected individuals (36, 37). The measurement of activity of this enzyme in hemolysates now provides the basis for the detection of the abnormal galactosemia homozygotes and for distinguishing between galactosemia heterozygotes and normal homozygotes (38, 39, 40). A schematic representation of pathways of normal galactose metabolism is shown in Figure I. The following enzymes have been detected in mammalian liver: galacto kinase (41), galactose-l-phosphate uridyl transferase (42), uridine diphosphogalactose-4'-epimerase (42), and uridine diphosphate galactose pyrophosphorylase (43), which participate in reactions 1, 2, 3, and 4 of Figure I, respectively. It had been generally accepted during the past twenty years that the formation of glycogen from galactose was carried out by way of glucose-l-phosphate and phosphorylase (6). However, more recent experimental evidence, well-summarized by Isselbacher (44), has negated this concept. The formation of glycogen from UDPG by now has been established by the work of Larner, Hastings, Leloir and their associates (45, 46, 47, 48). 8 PP UTP UDPGAL^ * i (4) GAL C ) | C aT d p G A L -I-P UDPGALA <• (2) ^►UDPG ^►GLU-I UDPGAL NAD <31 j - ' UDPG ▼ POLY- T SACCHARIDES GLYCOGEN -P G LU -6-P ► UDPGA f GLUCURONIDES Figure 1. Schematic Representation of Galactose Metabolism Galactosemia The first biochemical finding in the disease was galactosuria in all affected infants. The inability of the galactosemic individual to utilize galactose was demonstrated in patients by in vivo loading with galactose, which resulted in a sustained high level of blood galactose and in lowering of the blood glucose level (49) . It became common clinical practice to carry out the galactose tolerance test to confirm diagnosis of the disease (50, 51, 52). The measurement of erythrocyte galactose-l-phosphate uridyl transferase (transferase) now is replacing the tolerance test. The enzyme assay is convenient for the patient, as contrasted to the load test, and it avoids the potential hazard of hypoglycemia in the young infant. The mode of genetic transmission of the disease was clearly established by the work of Donnell, et al.(53) who studied 14 families in which the disease had occurred. A modification of the uridine diphosphoglucose (UDPG) con sumption assay for transferase, as first used by Anderson, et al. (38), was found to distinguish galactosemia heterozygotes from normal individuals. It was demon strated that the transmission is as an autosomal recessive character, or more properly stated, as an 10 incomplete dominant. In a separate study, no linkage with blood groups could be found (54). Although the usual experience has been that no transferase activity can be found in galactosemia homo zygotes by the UDPG consumption assay, some very low positive values, perhaps equivocal, have been interpreted by Hsia to indicate that some galactosemics may have some transferase activity (55). Employing an indirect mano- metric technique, Schwarz (5 6) has reported that 13 of a group of 23 galactosemia patients showed evidence of measurable activity, attributed by him to the presence of a minor amount of transferase activity. On the clinical side, exceptional cases of galactosemia have been reported (57, 58). In most untreated patients the manifestations are severe, but in some mild. Among treated patients some affected infants could tolerate a moderate quantity of milk, whereas others became violently ill. Also, it has been considered by some that mature galactosemics tended to manage a diet containing appreciable quantities of galactose (50). Mature galactosemics studied have been very few in number, and they may not represent the disease in general: but an impression has become current that all galactosemics will adapt with time to some intake of 11 galactose. Two obvious questions arise. First, is there a biochemical difference in the metabolism of galactose among galactosemic subjects? Second, can a galactosemic adapt with time' to utilize at least a small amount of galactose? In spite of a known enzymatic block in galacto semia, a small amount of galactose has been observed to be metabolized in some affected individuals to yield other products in addition to galactose-l-phosphate. In an attempt to assess the distribution of galactose in the body of an older galactosemic individual, Eisenberg, et al. (59) infused 5 uc of galactose-l-C^-4, 1 g. of unlabeled galactose and 1 g. of menthol. The latter compound serves as an indicator for glucuronide formation. It was found that 70% of the total radioactivity appeared in the urine as galactose, 0.5% was converted to menthylglucosiduronic acid and 0.5 to 1%, stayed in ’’cellular" galactose 1-phosphate, galactose, etc. The remaining 20 to 25%, radioactivity, which was not accounted for, was assumed to be metabolized to • Segal, et al. (60) have administered galactose-1- C14 intravenously to galactosemia patients. Two of eight affected individuals were found to exhale C^402 at a rate 12 comparable to normal controls, despite absence of erythrocyte transferase activity. Segal has suggested that some galactosemics may acquire an ability to metabolize galactose sometime during childhood. Metabolic variation among galactosemic subjects in the in vitro oxidation of galactose-1 -C^ by intact erythrocytes was demonstrated by Weinberg (61). Even though the amount of produced by one particular patient (who was one of Segal's unusual patients) is only 8% of the normals, this quantity is twelve to fifteen times more than that of other galactosemics included in this study. The ability of the same patient to oxidize galactose in vitro similar to that of iri vivo experiment is of particular interest. Clinical impressions, the suggestions of Schwarz and of Hsia based on transferase findings, the in vivo studies of Segal and the in vitro findings of Weinberg argue for the possibility of biochemical heterogeneity in galactosemia. The possibility of adaptation by galactosemic subjects to utilization of galactose needs careful evaluation. On the basis of the tolerance test, Donnell, et al. (62) in our laboratory found that no signi ficant adaptation to galactose had occurred in the 13 particular individuals studied, at least as measured by galactose-l-phosphate accumulation in erythrocytes. A number of the patients used in these studies have reached puberty, but they have been continuously treated from an early age, have not been exposed to a normal diet and thus may not have been stimulated to adapt. However, one individual, now sixteen years of age, who has not been treated on a restricted diet was included in the erythro cyte galactose-l-phosphate studies. He proved to be as sensitive to very small loads of orally administered galactose as treated galactosemics. Although Segal has suggested the possibility of acquiring a tolerance to galactose, information on management of his exceptional individuals is lacking. The question of adaptation is still open. Alternate Pathways of Galactose Metabolism Alternate pathways for galactose metabolism have been suggested as an explanation for the apparent ability of some galactosemic individuals to utilize small amounts of galactose. Isselbacher (63) proposed that UDPGal pyrophosphorylase can account for CO2 and glucuronide formation from galactose. Another possible alternate 14 pathway is that via the galactose-6-phosphate pathway described by Inouye and Hsia (64). An enzymatic reaction in which galactose-l- phosphate is converted to uridine diphosphate galactose in the presence of uridine triphosphate has been described as taking place in yeast (12). The same enzyme was demon strated by Isselbacher (63) to exist in rat, pigeon, and human livers. Moreover, it was noted in this study that UDPGal pyrophosphorylase activity increased with age in both rat and human livers. It was considered that this alternate route of galactose metabolism might explain the labeling of carbon dioxide and of glucuronides noted in previous work, and that this route might provide a means for greater tolerance to galactose in older galactosemic individuals. Another alternate pathway of galactose metabolism in man has been reported by Inouye and Hsia (64). Galac tose incubated with erythrocytes of galactosemic individ uals was said to yield galactose-6-phosphate and galactose- 1, 6-diphosphate in addition to galactose-l-phosphate. The presence of both galactose-l-phosphate and galactose- 6-phosphate as galactose intermediates had been demon strated earlier in algae upon incubation in the presence 15 of polymyxin B (65). The scheme postulated for galactose metabolism of this organism under the experimental condi tion employed is illustrated in Figure 2. To-date, the findings of Inouye and Hsia in man have not been confirmed either in this laboratory or elsewhere. In a study of the specificity of crystalline phosphoglucomutase of rabbit muscle, Posternak and Rosselet (66) found that galactose-l-phosphate in the presence of either glucose-1, 6-diphosphate or mannose-1, 6-diphosphate was converted into galactose-6-phosphate. The equilibrium heavily favored galactose-6-phosphate formation, as in the case of glucose-6-phosphate. At an optimum concentration of coenzyme, the action of this enzyme on galactose-l-phosphate was four hundred times slower than on glucose-l-phosphate. The practical signi ficance of this reaction in the metabolism of galactose- l-phosphate was considered doubtful. Recently galactose-6-phosphate has been identified as an intermediate in E. coli strain A (Weigle), which lacked galactokinase and transferase (67). The metabo lism of galactosamine in L. casei by this pathway has also been reported (68). Whether galactose could be phosphory- lated directly to galactose-6-phosphate was not 16 GAL GAL-1 -P GAL-6- P A t T - 6 - P A T-l, 6-DIP G L Y -3 -P + DAP ► GLU-I-P Figure 2. An Alternate Pathway of Galactose Metabolism 17 conclusively established. It was of interest, however, that glucose-6-phosphate dehydrogenase from yeast, but not from rabbit brain nor from rabbit liver, acted on galactose-6-phosphate (69). Stimulation of galactose oxidation in vivo by progesterone and menthol observed in some galactosemic subjects might also suggest an alternate pathway of galactose metabolism (70, 71). This effect had been demonstrated in rats and isolated liver tissue of rabbits (71, 72). Although studies by the same group of investi gators localized the site of action of progesterone and other steroids at the epimerase, there is no explanation for the fact that progesterone had no effect on galactose oxidation in vivo in normal human subjects (73, 74). Other minor pathways of galactose metabolism found in biological systems are of interest. The conversion of galactose-l-phosphate to thymidine diphosphate galactose has been demonstrated in extracts of Phaseolus aureus seedlings (75) and ,S. faecalis (76) . The reduction of galactose to dulcitol without phosphorylation has been shown in several mammalian tissues (77, 78, 79, 80, 81). The amount of dulcitol accumulated in lens upon feeding of galactose to rats and rabbits exceeded that of 18 galactose-l-phosphate by 400 times. However, the existence of this pathway has not been demonstrated in man. Galactose oxidase has been isolated as a specific enzyme from a fungus (82, 83). The fungus (Polyporus circinatus Fr.)at first was considered to be the source: later another fungus associated with it, Dactylium dendroides (Bu_ll.) Fr. was found to be the true producer of the enzyme (84). It catalyses oxidation of the hydroxyl group to an aldehyde group at carbon atom #6 of galactose, instead of the aldehyde group to a carboxyl group at carbon atom #1 as in the case of glucose oxidase. This reaction is of special interest in relation to a specific analysis for galactose, but it has no known metabolic involvement in the human. The DeLey-Doudoroff pathway for the conversion of galactose to triose phosphate in Pseudomonas saccharo- philia, without the formation of galactose-l-phosphate, has been recognized for several years (85). The reaction proceeds by the formation of galactonic acid, followed by 2-keto-3-deoxy-D-galactonic acid, its phosphate, pyruvate and D-glyceraldehyde-3-phosphate. No application to the human has been shown. Of special interest, however, is the finding, in 19 studies on toxic effects of galactose-l-phosphate in in vitro systems, that galactose-l-phosphate inhibits phosphoglucomutase (86, 87). It has been generally assumed that galactose-l-phosphate, or a transformation product of it, is responsible for the toxic manifestations in galactosemia. During inhibition by galactose-l- phosphate of the conversion of glucose-l-phosphate to glucose-6-phosphate, galactose-1, 6-diphosphate was formed. The inhibition could be reversed completely by the addition of glucose-1, 6-diphosphate. Another point of interest is that phosphoglucomutase from rabbit brain, in addition to catalysing conversion of galactose-l-phosphate, was effective with nine other analogues as well, demonstrating the non-specificity of this enzyme (88). Another important aspect of galactose metabolism actively being investigated by a number of workers is that of microbial genetics. Strains of E. coli-Kl2 and Salmonella with defects in galactokinase, transferase, and/or epimerase have been identified (89, 90, 91, 92, 93, 94, 95). The mechanisms of regulating enzyme synthesis in relation to the galactose pathway have been discussed (96, 97, 98). It can be expected that the results of such studies will eventually extend understanding of the 20 processes of galactose metabolism in man. Background of the Investigation The investigation to be described has been carried out in the Biochemistry Division at Childrens Hospital, with collaboration of the Metabolic Division. The study of galactosemia in this institution started in 1949 (99). Over a period of the last 14 years over twenty-five galactosemia patients have been accumulated. This is one of the best single collections of well-studied cases of this genetic disease in the world. In addition to studies on the homozygotes, members of more than twenty affected families have been classified on a genetic basis as normal or heterozygous. Some of the galactosemic subjects were detected at birth, and clinical complications were prevented in these cases. Others had been fed on a milk diet before they were received for diagnosis. The ages of the galactosemic subjects at the start of this study ranged from birth to sixteen years. Although some had low I.Q's, most were normal. In addi tion to work on clinical aspects of the disease, there have been carried out: (1) investigations on application of the galactose tolerance test to heterozygote detection, 21 (2) modification of the UDPG consumption assay for galactose-l-phosphate uridyl transferase, (3) use of the transferase assay for establishing genetic transmission of the disease, (4) genetic linkage to blood groups, and (5) relation of erythrocyte galactose-l-phosphate levels to galactose intake. The rich source of patient material, and the vast amount of clinical information available, made the present investigation possible. The study represents a further step in employing galactosemia as an experimental tool in studying galactose metabolism. The difficulties attendant upon a detailed bio chemical investigation are obvious. The disease is a problem involving humans, and, unlike the situation in animal experiments, the range of procedures and the variety of patient material that can be used are limited. Although the liver is the major site of galactose metabolism and would be the best representative tissue for studies, liver biopsy is rarely done. Erythrocyte transferase values now have been established to be reliable as an index. There is very little justification to employ biopsies merely for the sake of scientific curiosity, and the same atti tude prevails toward administration of radioactive material to a well-controlled galactosemia patient. The 22 taking of open biopsies or the in vivo administration of radioactive isotopes is reserved, at least at Childrens Hospital, for cases in which the patient will benefit directly or in which the procedure is an essential step in advancing knowledge of the disease. Treated galactosemia patients, especially those diagnosed early, have a normal outlook for their continuing development, and undue risks are avoided. Blood has been the tissue of choice for the present study, although skin and intestinal biopsies will be employed in continuing work. The use of blood in the present experiments has, in addition to its availability, a reasonable basis. Upon exposure to galactose, galactose-l-phosphate accumulates in vivo in the tissues of galactosemics, including erythrocytes. The same phenomenon occurs in vitro as well when the erythrocytes are incubated with galactose. Within 24 hours after a galactose tolerance test in a galactosemic, the in vivo level of erythrocyte galactose- l-phosphate drops from values as high as 20 mg.% to 1-2 mg.70 (62) . It is logical to consider studying in vitro the process by which the cell disposes of galactose- l-phosphate. The work of Weinberg has shown that a small amount of galactose can be oxidized in some galactosemic 23 erythrocytes. These considerations suggest that the galactosemic erythrocyte is capable of handling galactose- l-phosphate, perhaps at a very slow rate, and that the mechanisms should be studied. The presence of a galactose-6-phosphate pathway in galactosemic erythrocytes still remains to be substan tiated, but it has been profitable to explore the possibil ity of participation of UDPGal pyrophosphorylase. Although Isselbacher stated that activity of this latter enzyme could not be detected in human erythrocytes, the experi mental data was limited (100), and further work is justified. The finding by Segal in vivo and by Weinberg in vitro of production by a few unusual galactosemic subjects from labeled galactose represented an over-all metabolic event, none of the galactose intermediates having been isolated. It has been the purpose of the present investiga tion: (1) to examine the possibility of biochemical heterogeneity among the group of galactosemics available for the study, (2) to study in detail the occurrence of intermediates if it could be established that erythrocytes from certain galactosemics have some degree of capability 24 of metabolizing galactose, (3) to consider the possibility of adaptation of galactosemics with age, and (4) to relate findings to genetic transmission of the disease. STATEMENT OF THE PROBLEM AND PLAN OF ATTACK Statement of the Problem Some degree of heterogeneity in galactosemia has been suspected on clinical grounds. It has been the primary purpose of the present investigation to explore on biochemical grounds the possibility of variation in this genetic disease. Reports by Segal, et al, (60) have suggested that biochemical heterogeneity may exist. These investigators studied in vivo utilization of injected carbon-labeled galactose. Recovery of carbon label in exhaled carbon dioxide was comparable to normal controls in two of eight galactosemic individuals. In a separate in vitro study by Weinberg (61), whole blood and leucocytes from one of these individuals was able to convert galactose-l-C^ to labeled carbon dioxide at a substantially greater rate (12-15 fold) than observed with other galactosemics in his group. The results of Weinberg, together with similar findings by Isselbacher (101) with blood from two galacto semic individuals suggested a starting point for the present study. 26 Erythrocyte transferase activity could not be demonstrated in any of the galactosemic subjects studied, and Weinberg suggested that an alternate pathway might be operative in his unusual case, perhaps the uridine diphosphogalactose pyrophosphorylase system described by Isselbacher (43). One aim of the present study has been to test for the existence of an alternate pathway or pathways in galactosemic subjects. Although the finding of Isselbacher (43) that galactose-l-phosphate could be utilized in liver tissue without the participation of transferase has aroused great interest, his conclusion regarding the existence of UDPGal pyrophosphorylase in mammalian livers has been challenged by Strominger (9). Moreover, his presumption of this pathway being a means for adaptation to galactose by galactosemic individuals has not been substantiated by direct observation. The presence of galactose-6-phosphate as an intermediate of galactose metabolism in galactosemic erythrocytes, as reported by Hsia and his associates, has not been confirmed The inter-conversion of galactose-l-phosphate and galactose-6-phosphate in galactosemic hemolysates, as claimed, requires careful re-examination. 27 The amounts of carbon dioxide produced in vitro in the exceptional cases of Weinberg and of Isselbacher, while 12-15 fold that found with the usual galactosemic subject, are still only 5-8% of that with normal controls. It is possible that the enzymatic block in galactosemia may be incomplete in some individuals. The presence of some galactose-l-phosphate uridyl transferase (transferase) activity in galactosemia homozygotes has been discussed by Schwarz, £t al. and by Hsia, at al. on the basis of measurements of enzyme activity (56, 55). Employing a manometric procedure, Schwarz and his associates found no demonstrable erythrocyte transferase activity in 10 of 23 galactosemic children, but in the remainder an activity averaging 7% of normal values was found. A further pur pose of the present study has been to look for evidence bearing upon the question of completeness of the enzymatic deficiency. The possibility of the presence of inhibitors and of co-factors in hemolysates has been examined in previous work: no evidence for their existence could be found (102). Another basis for the present study has been the need to provide an explanation for the gradual disappear ance in vivo of galactose-l-phosphate accumulated in galactosemic erythrocytes following administration of an oral galactose load (62). As much as 20 mg. per cent (packed-cell basis) of galactose-l-phosphate may appear in such erythrocytes in the hours immediately following the load, but within 24 hours the level has returned to the 3 mg.per cent or less value characteristic of the particu lar galactosemic. Whether this process could be attributed to a very low transferase activity undetectable by available methods, to the existence of alternate pathways, or to some other mechanism has been an open question. In summary, the problem under study in the present investigation involves examination for biochemical variation in galactosemia, including investigation of possible secondary pathways and study of completeness of the genetically determined metabolic block. Plan of Attack In the group of galactosemia patients available for this study, no evidence for heterogeneity on clinical grounds had been noted. There were no particular individ uals who could be singled out as the most likely subjects for initial investigation. Consideration was given to in vivo use of carbon-labeled galactose as a survey measure, but this approach to the problem was not feasible. There is an understandable reluctance to expose children who are clinically well to possible hazards. Despite calculations of the very low radiation exposure in the doses which would be necessary, it has been the opinion of the Isotopes Committee at Childrens Hospital that in vivo use of carbon-14 is not justified as long as some other approach has the potential of fulfilling the experimental need. It remained, then, to consider in vitro procedures with tissues from the patients. The only possibility of obtaining liver tissue from galactosemia patients would be under some unusual circumstance, an unreliable prospect for support of the study to be made. Accordingly, it was decided to employ blood as the tissue to be investigated: samples are readily obtainable, and repetitive studies on the same individual can be made if desirable. Some degree of phenotypic difference among tissues in a galactosemic subject have been reported (103). However, it has been established that in galactosemics, transferase activity is very low or absent in liver as well as in erythrocytes and that in normal individuals this enzyme is present in both tissues (37). It is presumed, although not proven, that the halved amount of transferase found in erythrocytes of galactosemia hetero zygotes reflects a similar reduction in liver (62). Schwarz in a limited study has found correlation between erythrocyte and liver galactose-l-phosphate (32) . While there is ground for considering that results obtained in studies on galactose metabolism in erythrocytes may have application to other tissues, caution must be exercised in interpreting experimental results. The steps in the attack on the problem may be summarized as follows: • Survey of variability in the patient group: As many as possible of the galactosemic individuals available were studied by the in vitro labeled carbon dioxide procedure of Weinberg. Normal controls were included, as were heterozygote carriers from the families of affected individuals. Blood from three of the patients was found to carry on oxidation of carbon-labeled galactose at a markedly higher rate than that from other galactosemics in the group. These three individuals were considered to be in some manner biochemically different from the other affected individuals. 31 2. Since the evolution of labeled carbon dioxide in the in vitro experiments represents an over-all meta bolic process, further information concerning the varia bility noted was gained by examination of metabolic intermediates. Accordingly, residues from the in vitro incubations were subjected to column chromatography in the Dowex-l-formate gradient elution system. Identification of the separated compounds was carried out by paper chromatography, chemical analysis and enzymatic assays. 3. The ability of blood cells of certain galactosemic individuals to carry on oxidation of galac tose to some degree suggests that some transferase activity may be present, or that secondary pathways may be utilized. In addition to indirect evidence bearing on the question of the completeness of the metabolic block, efforts were made to refine the measurement of transferase by devising a radioactive procedure for the purpose. Secondary pathways investigated included: a. The UDPGal pyrophosphorylase possibility suggested by Isselbacher, including study of the influence of concurrent presence of UDPG pyrophos phorylase activity. 32 b. The pathway for disposal of galactose-1- phosphate via galactose-6-phosphate claimed by Hsia and his associates to be a possible adjunct in galactose metabolism. 4. In addition to the direct approach as outlined, a number of subsidiary studies having bearing on the problem were made: a. Time-sequence of appearance of direct transformation products and of glycolytic pathway intermediates upon incubation of blood cells with carbon-labeled galactose. b • vit:ro investigation of the disappear ance of galactose-l-phosphate accumulated in erythrocytes from galactosemic individuals. c. Measurements upon erythrocyte galacto- kinase in relation to age, with special interest in the possible effect on accumulation of galactose-l- phosphate in both normal and galactosemic infants. d. A limited study of galactose-associated enzymes in human liver. MATERIALS AND METHODS In carrying out the experiments, most of the reagents employed were from commercial sources, but it was necessary to prepare a few. Although many of the proce dures used were conventional, a number of methods either were modified or devised to fill specific needs. Materials: 1. Human material: Most of the studies done were carried out in vitro on samples of human blood. Initially, whole blood was used, but in later stages separate experiments were done with erythrocytes and leucocytes. In a few cases, it was possible to study biopsy or autopsy samples of human liver. Since the investigation has been directed at details of galactose metabolism, primary attention was directed to patients with galactosemia and, because of the genetic origin of this disease, to parents of affected individuals and to other family members. While this disease is relatively rare, a special interest in it at Childrens Hospital of Los Angeles has resulted in the accumulation of a group of affected individuals 33 34 now numbering over 25. In 6 of the families the disease occurs in two siblings. In most instances, the members of the families of the affected children have been studied from the genetic standpoint. Normal control samples were drawn from members of the laboratory staff, from children hospitalized for elective surgery and from normal, well children. For one phase of the work, blood samples from 39 patients with Down’s Syndrome (Mongolism) were employed. In a limited study, human liver tissue was used. 2. Reagents: A list of the principal reagents used is given in Appendix A. Galactose-l-C^ was pur chased from a number of sources, but in each case purifi cation steps were required to remove contaminants. Commercially available galactose-l-phosphate was not sufficiently pure for some purposes, and galactose-1 -C^- 1-phosphate could not be purchased. The procedures employed for the preparation of these two compounds will be described. 3. Enzymes: Lactic dehydrogenase (Type I) was obtained from the Sigma Chemical Company of St. Louis. Glucose oxidase for glucose analysis was in the form of a combined reagent (Glucostat) containing glucose oxidase, a horseradish peroxidase and p-anisidine (Worthington Biochemical Corporation, Freehold, New Jersey). The uridine diphosphate glucose dehydrogenase originally employed for assay of galactose-l-phosphate uridyl transferase at Childrens Hospital had been supplied by Drs. Hansen and Bretthauer of Michigan State University. At the time the present study was starting, it was decided to prepare the enzyme (as described below). However, when a suitable UDPG dehydrogenase was offered by the Sigma Chemical Company, continued preparation of the enzyme was no longer required. In the case of galactokinase, as used in the enzymatic synthesis of galactose-l-phosphate and of galactose-l-C^-l-phosphate, our own preparation from a galactose-adapted yeast was necessary. a. Uridine Diphosphate Glucose Dehydrogenase (UDPG Dehydrogenase); The enzyme was obtained from an acetone powder of beef liver prepared by the general method of Horecker (104). All of the operations in the acetone powder procedure were carried out below 4°C. Two hundred grams of fresh liver, after removal of gross connective tissue, were cut into small pieces with a knife and with scissors. The liver pieces, together with 100 ml. of acetone, which had been 36 cooled to -14°, were placed in a Waring blender (which had been pre-cooled with dry ice) . After homogenizing for 20 seconds, a second 100 ml. of cold acetone was added and homogenization was continued for another 20 seconds. A further 300 ml. of acetone then was added, and homogenization was continued for 1 minute and 50 seconds further. (Note: to avoid possible explosion hazard, the brief homogenization steps were carried on outside the cold room.) The homogenate obtained was passed through four layers of cheese cloth. The residue on the cloth was discarded. The semi-liquid material passing through was filtered through a coarse sintered glass filter. The filtrate was discarded, and the residue washed twice with 200 ml. of acetone, then finally with 200 ml. of ether. The ether-treated powder was dried overnight at room temperature (24°) in the open air. It then was stored in the freezing room awaiting use. Purification of UDPG dehydrogenase from acetone powder was carried out in accordance with the method of Strominger, e_t <al. (105). The active fraction was that obtained by alkaline ammonium sulfate precipitation (35 to 55% saturation, at pH 8). 37 b. Galactokinase: The preparation of galacto- kinase from yeast was based on the methods of Leloir and Trucco (106) and of Heinrich (107). (1) Culture: A galactose-adapted strain of Saccharomyces fragilis was grown on sterile agar slants containing 8 ml. of Wilkinson's medium and 1.5% agar (108). The galactose used was of N.F. grade (Pfanstiehl). Slants were incubated at 37° for two days growth, and then were used for inoculation or for storage of the culture. For storage, the culture was refrigerated: transfer was made to a fresh slant every two weeks. For enzyme preparation, four liters of liquid Wilkinson's medium were prepared. One liter was put in each of four 2-liter round-bottom flasks. Each flask was plugged with cotton containing an L-shaped glass tube, itself cotton-plugged. The flask assemblies, with medium, were autoclaved at 15 pounds pressure for 20 minutes. When the temperature of the medium had cooled to 24-25°, two loopfuls of culture were transferred from the agar slant to each flask. The inoculated medium was incubated at room temperature for 48 hours, 38 with aeration through the cotton-guarded inlet tubes. Cells were harvested in 50 ml. polypropyl ene tubes by centrifugation at 2500 RPM for 10 minutes. Finally, the cells were united and washed three times with equal volume of distilled water. The washed cells were spread on Petri dishes in a 2 mm. layer and exposed to air at room temperature for 3 to 4 days. The dried cells were collected in a 250 ml. Erlenmeyer flask, ready for extraction. All steps in the succeeding processes were carried out below 4°C. (2) Extraction: To 15 g. of dried yeast was added 150 ml. of 2.2% diammonium phosphate solution. The suspension was allowed to stand overnight in a refrigerator, stirred occasionally and then centrifuged. The precipitate was dis carded. The volume of supernatant fraction obtained was usually about 125 ml. In early trials of the procedure, the supernatant fluid was divided into two equal volumes (for example, two 63 ml. portions from 126 ml.) in order to test suitability of further steps, primarily bentonite treatment and ammonium sulfate precipitation. 39 (3) Bentonite Treatment: Two gms. of bentonite were added to one 63 ml. portion of the extract; this treatment resulted in precipitation of a large amount of protein. The suspension was cen trifuged, and the supernatant liquid was kept. It was found that the bentonite treatment did not increase the specific activity of the enzyme fraction, and the bentonite step was abandoned for routine use in this enzyme preparation. Other investigators have used this step for removal of hexokinase through careful selection of the amount of bentonite used. (4) Ammonium Sulfate Precipitation (45-55% Saturation): To the other 63 ml. portion of supernatant liquid, 14.3 g. of ammonium sulfate was added. After thorough mixing, the suspension was allowed to stand for 10 minutes: a large precipitate resulted. After centrifuging, the precipitate was discarded and the supernatant liquid was kept. An additional 5.45 g. of ammo nium sulfate was added, to give approximately 55% ammonium sulfate saturation. The protein precipi tated was recovered by centrifugation. 40 The precipitate was dissolved in 25 ml. of 0.001 M EDTA. The ammonium sulfate precipitation proce dure proved suitable for routine use. (5) pH 5_ Treatment: The protein solution was adjusted to pH 5.1 with 1 M acetic acid, and the suspension resulting was centrifuged. The small amount of precipitate was discarded. The pH of the supernatant fluid immediately was brought back to pH 6.0 by careful addition of 1 N sodium hydroxide. The solution was then dialyzed for a period of 4 1/2 hours against 0.02 M sodium ace tate buffer, 0.001 M in EDTA, at pH 5.9. (6) Calcium Phosphate Gel Precipitation: Calcium phosphate gel was prepared at a concentra tion of 15 mg/ml. according to Keilen and Hartree (109) . The concentration of dialyzed protein was determined and an amount of calcium phosphate gel equal to the weight of protein was added to the solution. In a given preparation, for example, the amount of protein was found to be 115 mg., and 7.7 ml. of calcium phosphate gel was added. The suspension was mixed, centrifuged and the gel precipitate was discarded. Another equal volume 41 of gel was added. At this time the gel was kept, and eluted with 10 ml. of 0.1 M ammonium phosphate buffer, pH 7.0. This fraction then contains the active enzyme. Further purification on a DEAE cellulose column, as carried out by some investi gators, was not necessary for the present purposes. (7) Assay: Assay of galactokinase was carried out as described in a separate section. Specific activity of the galactokinase preparation was expressed as umoles galactose phosphorylated per mg protein per minute. In a particular isola tion, the specific activity of each fraction was found to be as follows: Yeast extract 0.16 u Bentonite 0.16 Ammonium sulfate 1.45 Calcium phosphate gel 2.50 There was considerable variation in specific activity among different preparations. The ammonium sulfate fraction was stable for more than one month when the enzyme was stored at -17°C in a room free of organic solvent. This fraction, used for the preparation of labeled 42 galactose-l-phosphate, also contained hexokinase. The latter activity was found to be approximately l/4th that of the galactokinase activity. 4. Preparation of Galactose-1-Phosphate: It was found early in the investigation that a galactose-l- phosphate of sufficient purity for some purposes was not available on the market, and no source of labeled galactose-l-phosphate could be located. Accordingly, it was necessary to prepare both of these compounds, from galactose and from galactose-l-C^, respectively. The method of Anderson, _et al. was followed (110). This depends upon incubation of galactose in the presence of ATP and of galactokinase. In the present work, the pro cedure was modified by using a Dowex-l=formate column for initial separation of galactose-l-phosphate, followed by lyophilization. An ammonium acetate buffer first was used in the incubation, but it was found advantageous to employ the phosphate buffer of Anderson. The preparation of the galactokinase used has been described in a previous section. -r a. Purity of Galactose; Glucose is known to be a contaminant in commercial sources of galactose. This presented a potential problem, since the galactokinase 43 preparation to be used also contained hexokinase activity. Apart from other considerations, the presence of Glu-6-P or Glu-l-P would present a major problem in the assay of UDPGal pyrophosphorylase. Paper chromatography was used to establish the purity of galactose. The requirements considered are listed as follows: (1) The ability of the solvent system chosen to clearly separate galactose from glucose. (2) Separation of minimum amounts of glucose in the presence of large amounts of galactose. (3) The possibility of estimating the amount of glucose after separation. These criteria were met by the following experi mental conditions: (1) Paper: Whatman No. 1. (2) Length of paper: 57 cm., 40 cm. from spotting position to the end of the paper in the direction of moving solvent. (3) Solvent system: pyridine, n-butanol, water (1:3:1.5) . 44 (4) Preliminary equilibration: none. (5) Duration of run: 40 hours. (6) Spraying reagent: silver nitrate- ethanol. Both aniline hydrogen phthalate and silver nitrate were tried as spraying reagents. It was found that the minimum amount of sugar detected without equivocation by the former reagent was 20 micrograms and by the latter 2 micrograms or less. It should be noted that contamination either of the solvent or of the running sample with amonium ions or chloride ions can interfere with sensitiv ity of the spray reagent. Tracings or photographs of the chromatograms were made for permanent record, due to con tinued darkening of the paper upon standing. The use of photographic hypo solution for clearing excess silver salt from the paper did not prove practical, due to the low wet strength of the paper. To test the procedure, the following mixtures of glucose and galactose (in microgram amounts) were run: galactose/glucose 10:10, 16:4, 12:8, and 18:2 glucose/galactose 10:10, 16:4, 12:8, and 18:2 All of these mixtures were well separated. It was esti mated that the standard galactose sample (Sigma, purified 45 grade) contained less than 2% of glucose. This was based on the intensity of color produced in the galactose and the glucose spots from a 20 ug. galactose run. Figure 3 A shows separation of galactose from glucose on a paper chromatogram. 14 b. Purity of Galactose-l-C : The criteria of purity of galactose-l-C^ also was based on paper chromatography. Figure 3 B shows that with small amounts of galactose-l-C^ (about 2,000 cpm; sp. act. 1.69 mc/mM; Calbiochem) applied to the paper in the chromatographic solvent system described above, only one radioactive peak corresponding to galactose was observed. There was no evidence of labeled glucose or other labeled contaminants. However, when a more substantial amount (1 uc) of the same material was applied, several contaminants were revealed by radioautography.* Jn Figure 4 is shown a radioautograph of a chromatogram on which have been run glucose-1-C^ and *Radioautography was carried out with the help of Mr. Max Fields of the Biochemistry Division at Childrens Hospital. The paper chromatogram was exposed to an x-ray film (14 x 11 in.) for 3 weeks. The film was developed after the removal of paper chromatogram in the dark room. A STANDARD B 1 K SCALE 0 .4 - 0.3 - 0.2 - 0.1 - 4 0 30 20 10 LENGTH OF PAPER CHROMATOGRAM (cm ) Figure 3. Purity of Galactose-l-C^ A. Spots correspond to 10 micrograms of glucose and 10 micrograms of galactose. B. The radioactive peak corresponds to galactose-l-C^. 47 Figure 4. Radioautogram of Galactose-l-C^ Samples after Paper Chromatography (See text for explanation) 48 14 galactose-l-C from several sources. No. 1 corre sponds to Glu-l-C^ from Calbiochem; No. 2, Gal-l-C^ (Calbiochem) ; No. 3, Gal-l-C^ (National Bureau of Standards); and No. 4 to the sample of Gal-l-C^ from Calbiochem after treatment in our laboratory. The treatment was carried out by passing the commercial sample through a Dowex-1-formate column. The filtrate was concentrated by lyophilization before paper chromatography. The amounts of sample applied were large, in order to make contaminants visible, and the separation of glucose from galactose consequently was obscured. Although the radioautograms present a striking impression concerning impurities in the com mercial samples, the combined radioactivity of these contaminants as estimated in a scanograra did not constitute more than 1% of the total. The treated sample was markedly less contaminated than the original commercial sample. The Dowex-1-formate treatment was found to accom plish the removal of volatile bicarbonate and of the contaminants which have been designated as A and B, as shown in figure 5. The presence of labeled bicarbonate or carbonare as an impurity had been established by 4 0 0 - 200 - 100 50 TUBE NUMBER Figure 5. Contaminants of Commercial Galactose-l-C^ conditions: 4 uc of commercial galactose-l-C-^ were dissolved in 40 ml the solution was passed through a Dowex-1-formate column. The adsorbed were eluted with 0 to 1 M ammonium formate in a concentration gradient in the text. Experimental of water and contaminants as described 50 treatment with acid, followed by trapping of the 14 liberated C 0^ in hyamine. The hyamine was dissolved in toluene containing scintillators, and radioactivity was measured in a liquid scintillator system. Con taminant A was found in all commercial samples tested, (Calbiochem, Volk, NBS and New England Nuclear). c. Preparation of Unlabeled Galactose-1-Phosphate: (1) Incubation: Initially, incubation mix tures consisted of 5 ml. of each of the following: ATP (110 Umoles/ml.), galactose (55 umoles/ml.), MgCl2 (0.3 M), galactokinase, and ammonium acetate buffer (1 M, pH 7.1). The ammonium sulfate frac tion (40-557o saturation) of the galactokinase preparation, described in a previous section, was employed. As experience with the procedure was gained, a phosphate buffer was found to be more advantageous than the ammonium acetate. The mixture was incubated for 3 hours at 37°C. At the end of the incubation, the proteins were denatured by heating in a boiling water bath for 4 minutes. After centrifugation, the supernatant liquid was held under refrigeration pending a further step. 51 (2) Analysis of incubate: 0.1 ml. of the supernatant liquid was spotted on Whatman #1 paper. After descending chromatography with ethanol- ammonium acetate solvent for 16 hours, the paper was dried and tested for the presence of ultra violet absorbing materials, reducing sugar, and phosphorylated compounds. Spots corresponding to standard reference materials included on the chromatograms were observed (galactose-l-phosphate, ATP, ADP and inorganic phosphate). The galactose spot from the supernatant sample was very faint. In a separate analysis, based on the Nelson- Somogyi method, it was found that 96° L of original galactose had been consumed. (3) Separation of galactose-l-phosphate: The main portion of the supernatant liquid was passed through a Dowex-1-formate column. Materials absorbed onto the column were eluted with 1 M ammonium formate in a concentration gradient manner as described in a separate section. The eluate tubes (each fraction being approximately 7 ml.) were tested for galactose-l-phosphate by the cysteine-sulfuric acid method developed by 52 Dische, ejt <al.(lll), as modified by Diedrich and Anderson (112) . The ultraviolet absorption at 260 mu of each tube was checked (Beckman Model DU spectrophotometer). Eluates giving a positive test with cysteine-sulfuric acid were composited and lyophilized. An example of the separation obtained is shown in Figure 6. (Note: In Fig ure 6, due to a technical problem, the volume of eluate per tube decreased with time after tube number 100, resulting in an apparent enlargement of the ATP peak.) (4) Barium acetate treatment: Ten ml. of distilled water was added to the dry material in the lyophilization flask. The solution was trans ferred to a 15 ml. centrifuge tube, and the pH adjusted to 8.5 (the original pH was about 5.0). Three drops of 0.5 M barium acetate were added to remove inorganic phosphate, and the mixture centrifuged. Sodium sulfate (50 mg/ml.) was added dropwise to insure complete removal of barium ions in the supernatant fluid. After a second centri fugation, the supernatant liquid (total volume usually 10.7 ml.) was stored in the frozen state. ADP ATP i w O.D. 2 6 0 10 UNKNOWN 5 G A L -I-P 50 100 170 TUBE NUMBER Figure 6. Products of Incubation Mixture of Galactokinase, Galactose, and ATP 54 (5) Analysis of galactose-l-phosphate: The final supernatant was analysed both by chemical methods and by an enzymatic procedure. A problem developed with respect to a standard. It was sus pected that the commercial galactose-l-phosphate employed as a provisional standard might not be pure. Chemical analysis confirmed the suspicion. However, results of analyses for sugar content of the galactose-l-phosphate prepared as described above were congruent with those of determinations of phosphate content, and it was decided to base the standard on analysis, rather than on weighing. The provisional galactose-l-phosphate standard was prepared as follows: 5.7 mg. of barium salt of galactose-l-phosphate (Sigma Chemical Company, 2.6 mg. of galactose equivalent) was dissolved in 1 ml. of water. One to two drops of sodium sulfate (50 mg./ml.) were added to remove excess barium ion, fol lowed by centrifugation. The supernatant liquid was transferred to a centrifuge tube and diluted to 2 ml. It was assumed that no loss of galactase-l-phosphate was involved. 55 In the chemical procedure, aliquots of galactose-l-phosphate solution (both provisional standard and the prepared samples) were hydrolyzed in 1 N HC1 for 10 minutes at 100°. The hydrolyzed solutions were adjusted to neutral pH (extremely important for sugar analysis). Aliquots of the hydrolyzed and unhydrolyzed solutions were analysed for the amount of reducing sugar (Nelson- Somogyi procedure) and of inorganic phosphate (Taussky-Shorr method). Galactose was used for the sugar standard, since galactose has about 25% less reducing power than glucose. Results of chemical analysis are shown in Table I. It will be noted that less phosphate was found in the provisional standard than would be expected on the basis of the sugar determination, and that the values found in both were markedly less than expected. However, in the case of the prepared galactose-l-phosphate, equivalent results were obtained by the two determinations. (The theoretical values shown in Table I for the pre paration, are based on a hypothetical considera tion of complete conversion of galactose to TABLE I ASSAY OF GALACTOSE-1-PHOSPHATE BY CHEMICAL ANALYSIS OF ACID HYDROLYSATE Sample Reducing Sugar (RS) Phosphorus (Pi) Ratio of RS Found Theor etical % of Theory Found Theor- % of etical Theory (molar Pi basis) Provisional Standard 1.9 mg. 2.6 mg. 70 0.27 mg. 0.45 mg. 60 1.17 Preparation 18.4 mg. 50 mg. 37 3.24 mg. 8.6 mg. 36 1.03 Recovery Experiment* 104 ug. 106 ug. 98 17 ug. 17.4 ug. 98 *0n mixed aliquots of provisional standard and of preparation, employing separate analyses above to establish theoretical value. Ln O' 57 galactose-l-phosphate, i.e., a 1007. yield.) The second section of Table I shows the results of a typical recovery experiment in which aliquots of the provisional standard and the galactose-l- phosphate preparation were mixed, hydrolyzed and analysed. The finding of 98% of theoretical value by both determinations led us to accept the deter mination of sugar and of phosphate as suitable for establishing galactose-l-phosphate concentrations. The enzymatic procedure for galactose-l- phosphate determination was a modification of the method of Kirkman and Maxwell (113), as described in another section. It depends upon the measuring UDPG consumed in reacting with galactose-l- phosphate. The presence of galactose-l-phosphate uridyl transferase in excess is insured by the addition of a normal hemolysate. Aliquots of the preparation sample, after chemical analysis by the methods described, were diluted and assayed for galactose-l-phosphate by the enzymatic procedure. Correspondence was excellent, as attested by the following results: 58 Chemical Enzymatic 12.5 mg% 11.9 mg% 6.2 3.1 1.2 0.5 8.0 10.0 5.0 5.6 3.1 0.9 0.6 8.2 10.2 4.5 d. Preparation of Galactose-l-C -1-phosphate In gaining experience with preparing unlabeled galactose-l-phosphate, it was noted that the yields were not as good as those described by Anderson _et al. (110). The possibility of using the details of the ethanol-barium precipitation of Anderson were studied, but employing glucose-l-phosphate in order to conserve existing stocks of galactose-l-phosphate. In these experiments, 54-66% (5 mg. sample) and 857. (150 mg. sample) of the original glucose-l-phosphate were recovered, based on analysis of sugar and of inorganic phosphate. The conditions of the step are demanding and difficult to maintain. On the other hand, the use of a Dowex-1-formate column and lyophilization for the recovery of glucose-l-phosphate was found to provide consistently a yield in excess of 90%. Further improvement was obtained by introducing a 59 re-incubation step. (1) Incubation: To each of the two 10 ml. lusteroid centrifuge tubes were added: ATP Galactose MgCl2 Potassium phosphate buffer 1.0 ml. (0.5 M, pH 7.0) ,14 1.0 ml. (110 umoles/ml.) 1.0 ml. (55 umoles/ml.) 1.0 ml. (0.3 M) Galactose-l-C' Galactokinase (40-55% ammonium sulfate saturation fraction) 0.3 ml. (3 uc, sp.act. 1.69 mc/mM 0.7 ml. The mixture was incubated in a 37° water bath for two and one half hours. It was then immersed in a boiling water bath for 10 minutes followed by centrifuging. A small aliquot of the supernatant fluid was tested for reducing sugar and was chroma tographed on Whatman #1 paper using an ethanol- ammonium acetate solvent system. It was found that 80-90% of the galactose had been consumed; on the chromatogram were two radioactive spots, corresponding to Gal-l-P and to galactose (Fig ure 7). To complete conversion of the residual galactose to galactose-l-phosphate, re-incubation - 0.7 G A L-I-P 0.5 IK SCALE-► GAL 4 0 20 LENGTH OF PAPER CHROMATOGRAM (cm) Figure 7. Paper Chromatogram of Supernatant Fluid from Galactose-l-C^ Incubation 61 was carried out with an additional 0.5 ml. of each of the following components (in the original con centrations) for two hours: ATP, phosphate buffer, and the enzyme. At the end of this incubation period, over 95% of the galactose had been consumed (2) Isolation: The supernatant liquid was passed through a Dowex-1-formate column, and adsorbed compounds were eluted with a gradient to 1 M ammonium formate, as described in a separate section. A 0.1 ml. aliquot of each eluate tube was checked for radioactivity on planchets. Samples of extremely high activity were suitably diluted. The pattern found is shown in Figure 8. Recovery of radioactivity was estimated to be as follows: Filtrate 4.8%, Peak I 80.0% (tubes #20-31) Peak II 6.7% (tubes #33-37) Others 6.0% Total 97.5% The composited eluates in the tubes correspond ing to Peak I and the composite of those corres ponding to Peak II were concentrated separately by 62 13 ■ TUBE NUMBER Figure 8. Radioactive Products Isolated from Supernatant Fluid of Galactose-1-C1^ Incubation 63 lyophilization. The dried material from Peak I was dissolved in 11.5 ml. of water, and the Peak II material in 2.1 ml. Recovery of radioactivity after lyophilization was 100% for Peak I and 75% for Peak II. (3) Analysis: It was thought originally that the two radioactive peaks might represent two different compounds, perhaps reflecting contamina tion by labeled glucose. The first approach to their identification was by use of paper chromato graphy, employing the system described by Bandurski for the separation of sugar esters. Both fractions yielded a single spot, corresponding to galactose- l-phosphate, with no evidence for the presence of glucose-l-phosphate. Both fractions were labile to acid hydrolysis at 100°. On paper chromato graphy, the hydrolytic products of both fractions showed one radioactive spot corresponding to galactose. Trial of enzymatic determination by glucose oxidase with the hydrolytic products from both fractions gave negative results However, the unhydrolyzed fractions were able to serve as substrate in the enzymatic method for 64 galactose-l-phosphate described previously in this section. Assuming the radioactive compound in both peaks to be galactose-l-phosphate, calculation based on radioactivity checked well with concentra tions determined by the enzymatic method: Enzymatic Dilution Radioactivity Determination Peak I 1:10 20.5 mg% 19.0 mg% Peak II 1:5 13.2 mgT, 13.0 mg70 The various points cited indicated that the parti tion of radioactivity into two peaks was fortuitous and that a single compound was involved. To provide conclusive evidence that two peaks represented a single compound, galacto.se-l- phosphate, 2 ml. aliquots from each fraction were mixed and re-chromatographed. Only one radioactive peak was obtained. The original formation of two peaks might have been due to presence of salts introduced with the sample. When the same phenomenon was observed in subsequent preparations, it was assumed that the separate peaks represented one compound. However, the nature of the product was always established by chemical, chromatographic 65 and enzymatic procedures. 5. Equipment: The present study was carried out in the Biochemistry Research Division at Childrens Hospital. Adequate laboratory facilities were available, including a cold workroom at 2°C and a deep-freeze room at -17°C. Except for construction of lyophilization apparatus and a set up for ultrafiltration, the equipment employed was commercially available. During the course of the work there was gradual improvement in the capability for radio active counting. Initially this was done entirely on planchets in a manually operated thin-window counter. Then an automatic sample changer and a strip counter were added. For a time it was necessary to do scintillation counting at the University of Southern California and at the California Institute of Technology, but during the last year of the work, a liquid scintillation spectrometer (Nuclear-Chicago) was present at Childrens Hospital. A list of the principal equipment employed is given in Appendix B. Methods 1. Chemical Determinations: a. Reducing sugar [ Nelson-Somogyi,(114)]: The determination is based on the ability of carbohydrates possessing an available reducing group to reduce cupric ions to cuprous ions at slightly alkaline pH and at boiling water temperature. The cuprous ions then react with an arsenomolybdate reagent to form a complex exhibiting a green to blue color, the depth of color being dependent upon the concentration of the reducing sugar present. The solution to be tested must not be acidic. The method is sensitive in the range of 10 to 100 micrograms of glucose. The procedure is applicable to galactose also. b. Submicro sugar determination [Park and Johnson, (115)1* This method is based on the reduction of ferricyanide to ferrocyanide by reducing sugars at alkaline pH. Ferrocyanide ions produced react with ferric ions to yield ferri-ferrocyanide, which exhibits a prussian blue color. The test is extremely sensitive to small amounts of glucose, ranging from 1 - 9 micro grams. It was found essential to measure the color within 10 minutes. 67 c. Glucose oxidase determination of glucose [Worthington Bulletin, (116)]: The method is based on the following reactions: GJ.ucose Oxidase Glucose + C>2 + 1 ^ 0 ____-> H2O2 + Gluconic acid Horseradish Peroxidase ^2^2 + Reduced Chromogen Oxidized Chromogen + H2O The oxidized chromogen has a yellow color. The method is sensitive to a glucose concentration as low as 5 mg70. The glucose oxidase and the peroxidase are commercially available (Glucostat, Worthington Biochemical Corporation). d. Inorganic phosphorus [Taussky and Shorr,(117)]: This procedure is based on the reaction of inorganic phosphate with molybdate to form phosphomolybdate, which is then reduced by ferrous sulfate to form a blue colored compound. The color formed is stable. A range 2 to 40 micrograms of phosphorus can be determined accurately. e. Lactic acid [Barker and Summerson,(118)]: Lactic acid is converted to acetaldehyde on being heated with concentrated sulfuric acid. The 68 acetaldehyde formed is reacted with p-hydroxydiphenyl. Interfering materials present in the sample (iron, for example) are removed by copper hydroxide and calcium hydroxide. The test is sensitive down to a concentra tion of 5 milligram per cent. f. Protein [Lowry, et: al_. , (119) ] : The determina tion is based on the reaction of protein with copper to form a complex, which reduces phosphomolybdic- phosphotungstic acid in the Folin-Ciocaltau reagent to produce a color. It is necessary to allow the mixture to stand for 30 minutes for full color development. The amount of color in relation to protein varies with varying tyrosine and tryptophan content. However, the method is simple and is suitable for the type of pro tein measurement required in the present study. Bovine serum albumin was used as a standard. 2. Enzymatic Assays: a. Lactic acid: Determination of lactic acid was carried out according to the method of Neilands, employing lactic dehydrogenase (120). The reaction involved is the following: 69 Lactic Dehydrogenase Lactate + NAD- * " <----* Pyruvate + NADH + H"*" The spectrophotometric measurement of NADH absorbance at 340 mu reflects the concentration of lactate. Since the equilibrium heavily favors pyruvate reduction at neutral pH, pH 10 was employed by Neilands to drive the reaction in favor of lactate oxidation. Lithium lactate was used as standard instead of sodium lactate. b. Galactose-l-phosphate; The procedure employed was a modification of that described by Kirkman and Maxwell (96). The following enzymatic reactions are involved: Gal-l-P Uridyl Transferase UDPG + Gal-l-P «-----» UDPGal + Glu-l-P UDPG Dehydrogenase UDPG + 2 NAD+ ______* UDPGA + 2 NADH + 2H+ The assay is carried out in two successive incubations corresponding to the two reactions. In the first, an excess of UDPG is added, and transferase is provided by addition of a normal hemolysate. Progress of the first reaction is stopped by heating. Precipitated protein is removed, and residual UDPG is estimated in a second incubation in which a commercial preparation of UDPG dehydrogenase and NAD are included. Spectro- photometric determination of the optical density increase at 340 mu gives the amount of NADH formed, which in turn is equivalent to the amount of UDPG consumed. With more galactose-l-phosphate in the first incubation medium, more UDPG is consumed and a smaller optical density change is obtained in the second incubation. A control containing all of the incubation components, except galactose-l-phosphate, provides the basis for determining the initial amount of UDPG. The reagents and detailed procedure were those of the modification of the Kirkman and Maxwell procedure as made by Donnell, £t a_l. (121) . c. Galactose-l-phosphate uridyl transferase (UDPG consumption assay): The procedure employed was that of Bretthauer, e_t a_l. (39), which permits detection of the heterozygote state in galactosemia. The general principles are much the same as those described above for the assay of galactose-l-phosphate, except that galactose-l-phosphate now is added in excess and the transferase is limiting. 71 d. Galactose-l-phosphate uridyl transferase (Radioactive assay): The UDPG consumption assay is tedious, is subject to the inherent difficulty of bas ing the starting amount of UDPG on a control tube and is dependent upon a supply of UDPG dehydrogenase of dependable quality. Further, for routine use the cost of the dehydrogenase also represents a practical problem. In the course of the present work, it was decided to devise a more direct assay, based on earlier work on liver tissue by Kalckar and his associates (37). The principle is the measurement of the incorporation of carbon label in galactose-l-C^-l-phosphate into UDPGalactose, the rate of conversion being dependent upon the amount of transferase present. An essential feature of the assay is the adsorption of UDPHexose upon charcoal, and a study of the problems involved was made. (1) Separation of mononucleotide sugars from sugar phosphates: Crane and Lipmann have noted that charcoal adsorbed all mononucleotides but not sugars or sugar phosphates (122) . They found that mononucleotides could be eluted with ethanol. 72 (a) Efficiency of charcoal adsorption of UDPG: Preparation of charcoal for this study was begun by suspending 50 gm. of activated charcoal (Merck powder) in 500 ml. of distilled water in a beaker. After stirring, the mixture was filtered under suction on a Buchner filter. The beaker was rinsed with 50 ml. of water and an additional 200 ml. of water was passed through the filter, followed by 200 ml. of 50% ethanol. Finally, the charcoal was rinsed with 100 ml. of water and dried under vacuum over night. Lumps of charcoal were broken down into fine powder with a stirring rod before storage for use. To determine the effectiveness of the charcoal in adsorbing UDPHexose, UDPGlucose was employed as a test substance. A weighed amount of activated charcoal was added to 2 ml. of .8 x 10"*^ M. UDPG solution in a 12 ml. centri fuge tube, with water serving as control. The mixture was inverted several times and then allowed to stand for 10 minutes. After centri fuging at 2500 R.P.M. for 10 minutes, the supernatant liquid was removed. UDPG was 73 eluted from the charcoal with three successive 2.0 ml. portions of 507. ethanol containing 0.1% ammonia. The optical density at 260 mu of each fraction was read in a Beckman spectro photometer in 1.0 ml. Pyrocell cuvettes. The results are summarized in Table II. It was noted 40 mg. of charcoal, but not 20 mg., were sufficient to adsorb the UDPG present. (b) Recovery of galactose-l-C^-1- phosphate: The possibility that the charcoal preparation used might adsorb a significant amount of galactose-l-phosphate was tested by employing radioactive galactose-l-phosphate. In a typical experiment, the following mixture was made up in a 12 ml. centrifuge tube: Galactose-l-C^-l-phosphate 0.1 ml. (24,500 cpm, 0.8 umoles) Water 0.9 ml. 5%, Trichloracetic acid 1.0 ml. Washed charcoal 50 mgm. After mixing by several inversions, the con tents of the tube were centrifuged at 2500 r .p.m. for 10 minutes. The supernatant liquid was TABLE II RECOVERY OF UDPG AFTER CHARCOAL ADSORPTION IN TERMS OF OPTICAL DENSITY Tube Number 1 2 3 Added Water UDPG UDPG Mg. charcoal 20 20 40 Original solution 0 1.273 1.273 Supernatant liquid 0 0.348 0 First ethanol eluate 0 0.846 0.630 Second ethanol eluate 0.050 0.070 0.610 Third ethanol eluate 0.045 0.012 0.175 Total O.D. in eluates 0.095 0.928 1.415 Difference from water 0.833 1.225 (Per cent recovery) (65) (96) 75 kept. The charcoal was washed twice with 2 ml. of water, and the supernatant liquid from each washing was united with the first supernatant. For counting, 0.1 ml. of the total (5.9 ml.) was pipetted onto a planchet. Complete recovery of galactose-l-phosphate was obtained in the combined supernatant and washings. There was no evidence for galactose-l-phosphate being adsorbed onto the charcoal. (c) Conditions for planchet counting: An initial difficulty with radioactivity counting on planchets was overcome by employing lens paper for sample spreading. On the planchets used, more material, especially non-protein substances, tended to concentrate along the edges instead of being uniformly distributed upon drying. As a result, the efficiency of counting was decreased, and results of dupli cate samples varied greatly. The lens paper used in avoiding this problem was a soft, rough- surfaced type, cut to planchet size with a metal punch. Experience with a smooth type of lens paper showed that it tended to curl away 76 from the surface upon drying. This was not true of the type settled upon. Ethanol (50%) was used to reduce the surface tension of aqueous solutions, in order to minimize the volume necessary to cover the entire surface. When the sample volumes applied were small (0.1 ml.), counts with lens paper were decreased by 20%. On the other hand, when sample volumes were 1 ml. (the usual condition in our tests), radioactivity without lens paper was even less than when the paper spread ing technique was used, and replication was poor. The advantages of the paper led to its routine use. Count values obtained were cor rected for self-absorption. (2) Procedure for radioactive assay of galactose-l-phosphate uridyl transferase: (a) Heparinized whole blood (4-5 ml.) was washed three times with equal volumes of normal saline solution. The buffy coat was removed. Final packing of the cells was carried out in a refrigerated centrifuge at 3,000 R.P.M. (approximately 1,100 x g) for 77 15 minutes. The packed cells were hemolysed by freezing in dry ice and thawing, repeated three times. The enzymatic activity of the 100% hemolysate was found to be retained in the frozen state for at least one week. A 507, hemolysate was prepared just prior to incuba tion by addition of an equal volume of water. (b) The reagents employed for the incu bations were prepared as follows: (1) Labeled galactose-l-phosphate: Each pre paration made was adjusted to contain 0.52 mg. galactose-l-phosphate per ml. (2 x 10“^ M ). Radioactive counts varied with the particular preparation (20,000-35,000 c.p.m./umole). (2) Uridine diphosphoglucose (Sigma): 1.4 mg. per ml. (2 x 10”^ M). (3) Glycine buffer: 1 M. glycine adjusted to pH 8.1 with 10 N. NaOH. (4) Charcoal: Merck's activated charcoal washed with 50% ethanol, followed by water washing and subsequent drying under vacuum. The incubation was carried out in plastic centri fuge tubes (1 x 7.7 cm.) in a constant temperature 78 block at 37°C. After 30 minutes, the reaction was stopped by the addition of 2 ml. of 570 trichloro acetic acid (TCA). Incubation mixtures were as follows: Control Test Water 0.4 ml. 0.3 ml. Glycine Buffer (pH 8.1) 0.4 0.4 Labeled Gal-l-P (0.2 umoles) 0.1 0.1 507. hemolysate 0.1 0.1 UDPG (0.2 umoles) -- 0.1 Total volume 1.0 ml. 1.0 ml. The TCA-treated contents, after mixing by several inversions, were centrifuged at 10,000 R.P.M. (approximately 12,000 x g) for 40 minutes. The supernatant liquid was withdrawn and placed in a centrifuge tube (12 ml. capacity) containing 100 mg. of washed activated charcoal. After thorough mixing, the suspension was allowed to stand for 10 minutes and then centrifuged at 2500 R.P.M. for 10 minutes. The supernatant liquid, containing unreacted galactose-l-phosphate, was discarded, and the charcoal was washed twice by suspension and centrifugation, each time with 2 ml. of distilled water. Adsorbed UDPHexose was eluted from the charcoal residue with two successive 2.0 ml. por tions of 507c ethanol containing 0.17o ammonia. The eluates were united and centrifuged to remove small amounts of charcoal. The final volume approximated 3.8 ml. A third elution portion was found to be unnecessary, since, in preliminary trials it was found to contain less than 57> of the total adsorbed radioactivity at maximum enzyme activity. Radioactivity in UDPHexose was determined on planchets in a gas flow Geiger counter (Baird Atomic, Model 135). The amount of galactose-l- phosphate consumed in the transferase-mediated reaction was taken to be equal to the UDPHexose found, and enzymatic activity was expressed as micromoles of galactose-l-phosphate consumed per 1 ml. packed red blood cells per hour. In examining the time course of reaction under the conditions of the assay, incubations were made for 10, 20, and 30 minutes. The consumption of galactose-l-phosphate was linear during the 30-minute period. 80 e. Separation of galactose-l-phosphate uridyl transferase from hemoglobin: Early in the work, the feasibility of preparing a concentrated transferase, free of hemoglobin, was studied. Separation was obtained by the use of a DEAE cellulose column, but the loss in activity during the steps employed was suffi cient to discourage use of the procedure for quantita tive purposes. Subsequently, the use of DEAE cellulose for the separation of non-heme proteins from hemo globins has been reported by others (123). A hemolysate was prepared from 10 ml. of whole blood collected in a heparinized tube from a normal person. After washing in normal saline, the erythro cytes were lysed by freezing and thawing in the presence of an equal volume of water. The 50% hemo lysate was centrifuged in a refrigerated centrifuge (Servall) at 12,000 x g for 45 minutes to remove stroma. The stroma-free solution was dialyzed for 20 hours against 500 ml. of potassium phosphate buffer (0.005 M) at pH 7.0, containing cysteine (0.001 M). To begin preparation of the column used, DEAE cellulose was suspended in a beaker filled with 0.005 M potassium phosphate buffer, pH 7.0 for 3 hours. The 81 mixture was stirred and allowed to sediment for 15 minutes, and then the upper fraction was decanted off. The sedimented particles were packed into a column made from a 25 ml. pipette, using gentle suction. The DEAE cellulose in the column was washed with 150-200 ml. of 0.005 M potassium phosphate buffer at pH 7.0. The column was placed in the cold room for at least 1 hour before use. Seven ml. of 507, hemolysate was put on the column under gentle suction, and 30 ml. of 0.005 M potassium phosphate buffer at pH 7.0, containing 0.001 M cysteine, was passed through. Practically all of the hemoglobin was removed by the buffer wash. Washing was stopped when no more visible red or pink color was seen in the effluent. The hemoglobin eluate was concentrated in an ultrafiltration apparatus on a collodion membrane. The non-heme proteins remaining on the column were eluted with 0.2 M potassium phosphate buffer at pH 7.0, containing 0.001 M cysteine. Total volume of eluate, which had a very slight pink color, was approximately 60 ml. This non-heme protein fraction also was subjected to ultrafiltration. The volume of each of the two fractions was made equal to that of the hemolysate taken originally (7 ml.), for comparison of 82 enzymatic activities on a simple basis. Transferase activity was determined either by the labeled galactose- 1-phosphate method or by the UDPG consumption assay. Activity was found to be present in the non-heme fraction but not in the hemoglobin portion. Separation of transferase from hemoglobin had been achieved, but loss of total activity in the process was estimated to be 70%. f. Assay of UDPGal Pyrophosphorylase and of UDPG Pyrophosphorylase in Hemolysates: An enzyme effecting the conversion of glucose-l-phosphate to UDPG in the presence of UTP has been shown to be present in liver, blood cells and other tissues (37,38). A corresponding enzyme for conversion of galactose-l-phosphate to UDPGal has been described by Isselbacher in liver (43). Although it has been stated that the latter enzyme, UDPGal pyrophosphorylase, does not occur in erythro cytes, it was considered necessary in the course of the present investigation to re-examine this point. It also was desirable to study UDPG pyrophosphorylase activity in hemolysates. The reactions involved are: 83 UTP + Gal-l-P UDPGal + PP (1) UTP + Glu-l-P UDPG + PP ( 2) UDPG + Gal-l-P UDPGal + Glu-l-P (3) UDPGal UDPG (4) Reaction (1) is catalyzed by UDPGal pyrophosphorylase, (2) by UDPG pyrophosphorylase, (3) by Gal-l-P uridyl transferase and (4) by UDPGal-41-epimerase. With the use of either carbon-labeled Gal-l-P or Glu-l-P as substrate, the UDPGal or UDPG formed will also be labeled. Since charcoal adsorbs UDPGal or UDPG but not Gal-l-P or Glu-l-P, the rate of UDPGal or UDPG forma tion can be determined. Enzymatic activity then is expressed in terms of such a rate. Depending on choice of substrate, either of the two enzymes can be assayed. Hemolysates contain insufficient amounts of NAD to permit the epimerase reaction to constitute an inter ference in the assay of UDPGal pyrophosphorylase or UDPG pyrophosphorylase. The preparation of hemolysates was similar to the procedures described for the transferase assay. The composition of the incubation mixtures varied accord ing to each particular immediate purpose. The procedures for incubation, separation of UDPHexose and 84 determination of radioactivity in labeled UDPHexose were similar to those which have been described for the radioactive transferase assay. g. Assay of UDPG Pyrophosphorylase and of UDPGal Pyrophosphorylase in Human Liver: The methods of assay in liver were similar to those employed with hemoly sates. However, whole liver homogenate as such was found to be unsuitable since it contains substantial amounts of NAD, of glycogen, and of UDPG-glycogen synthetase, any of which may interfere with assay of the two pyrophosphorylases. NAD activates UDPGal-4'- epimerase. Glycogen is a potential G-l-P donor in the presence of phosphorylase. UDPG-glycogen synthetase occurring in combination with glycogen primer can lead to removal of UDPG. In experiments with whole rat liver homogenate, it was found that half of the UDPG present disappeared during the incubation period, even in the absence of galactose-l-phosphate. Attempts were made to remove these interfering substances. All operations were carried out at 4°C. Prelimi nary experience was gained with guinea pig liver. In two cases it was possible to use the human tissue at once: in two instances it was held frozen, but not for 85 more than two days. A piece of tissue (100-200 mg.) was placed in a 50 ml. beaker containing 20 ml. of 0.1 M potassium phosphate buffer at pH 7.4. After several minutes it was removed to a watch glass and cut into many small pieces with scissors, an operation especially necessary with the human tissue, which was found to be more difficult to homogenize than young guinea pig liver. The cut pieces of liver were trans ferred to a vial containing 4 ml. of the same phosphate buffer and homogenized for 30 seconds in a Vir-Tis No. 23 homogenizer. A second sample was treated identically. The two homogenates were united and cen- trifued at 10,000 RPM (12,000 x g) for 45 minutes in a refrigerated centrifuge. The supernatant liquid was withdrawn carefully, avoiding lipid contamination as much as possible. It was dialyzed for 4 hours against 500 ml. of 0.16 M potassium phosphate buffer, pH 7.4. Dialysis was carried out in a 600 ml. beaker under magnetic stirring. The dialysis sack was mounted in contact with the rotating magnetic bar to provide motion of the sack. The dialyzed fraction was used immediately for enzyme assay (UDPGal pyrophosphorylase or UDPG 86 pyrophosphorylase) in the same manner as described for hemolysates. In some cases, glucose-6-phosphate dehydrogenase activity in this fraction also was tested, employing the procedure of Zinkham (124). In the preliminary studies on guinea pig liver pre parations, UDPGal pyrophosphorylase activity was found to be three times greater in the dialyzed fraction than was apparent in the non-dialyzed homogenate, but there was no difference in the galactose-l-phosphate uridyl transferase activity. The dialyzed fraction did not consume UDPG in the absence of galactose-l- phosphate, but the homogenate did. It was found that centrifugation of the dialyzed fraction at 18,000 x g for 1 1/2 hours did not alter the ratio of transferase to UDPGal pyrophosphorylase activity. According to Claude, centrifugation at this speed and duration removes glycogen particulates (125) . h. Galactokinase (Yeast): In the course of pre paring galactokinase, as described in a separate section, each fraction was assayed by the procedure of Leloir and Trucco (106). The incubation mixture contained 0.2 ml. of each of the following: ATP (40 umoles/ml.), galactose (purified grade, 87 10 umoles/ml.), MgSO^ (0.1 M) , maleate buffer at pH 6, and enzyme solution under assay. When enzymatic activity was high, the enzyme solution was diluted. The concentration of protein was determined as a reference base for each fraction assayed. Incubation was carried out for 10 minutes at 37° in a constant temperature block. At the end of the incubation period, the enzyme was denatured by heating in a water bath at 100° for 4 minutes. After coagulated protein had been centrifuged down, the amount of reducing sugar left in the supernatant fraction was determined by the Nelson-Somogyi method (114), using galactose as standard. The specific activity of galactokinase fractions was expressed in terms of micromoles of galactose phosphorylated per minute per mg. of protein present. i. Galactokinase (Hemolysates): Galactokinase activity in hemolysates was determined by a radio active method based on the work of Sherman and Adler with Escherichia coli (126). It depends upon the con version of galactose-l-C^ to labeled galactose-l- phosphate in the presence of ATP, followed by separation of the two compounds by chromatography on 88 DEAE cellulose paper strips. (1) Reagents: Galactose-l-C14, sp.act. = 3.32 uc/mg.,1 mg/ml.(NBS) ATP, 40 umoles/ml. (22 mg/ml.) Tris buffer, pH 7.0, 0.5 M, u = 0.2 (10 ml. of 1 M Tris + 9.3 ml. I N HCl) MgS04.7H20, 8 x 10"4 M. (2) Procedure: (a) Preparation of 507, hemolysate: The cells from 3-4 ml. of fresh whole blood were washed three times with an equal volume of saline. The buffy coat was removed during the first washing. The packed erythrocytes were frozen and thawed three times, and an equal volume of water was added to make a 50% hemolysate. (b) Incubation mixture: Buffer 0.3 ml. MgS04 0.1 ATP 0.1 Gal-l-C14 0.1 Hemolysate 0.4 1.0 ml. 89 (c) Tests were run in duplicate. Mixtures were incubated for 30 minutes at 37°C in lusteroid tubes (1 x 7.7 cm.). At the end of incubation, the tube was placed in a boiling bath for 4 minutes and then immersed in a cold water bath for 1 minute. In the controls, proteins were denatured by immersing in boiling water for 4 minutes prior to the addition of radioactive galactose. The tubes were then inverted several times and centrifuged at 10,000 RPM for 30 minutes (Servall SS-34). Twenty microliters of the supernatant solution were applied in a streak at the origin point of a 3.5 x 29 cm. strip of Whatman DE-20 DEAE cellulose paper sheet. Descending chromato graphy was carried out in a closed chamber, employing water as the solvent. The progress of the solvent front was watched, and when it had reached a point 20 cm. from the origin, the chromatography was discontinued. After drying, the paper was cut, starting at the origin, into three 7 cm. sections. Each of these was inserted into a vial, scintillation 90 liquid was added, and counting was carried out in a liquid scintillation counter (Nuclear- Chicago). An appropriate blank was run. The two 7 cm. segments nearest the origin were found to contain the labeled galactose-l- phosphate, while the labeled galactose had migrated to the third segment, that furthest from the origin, j. Assay of Hexokinase (in Galactokinase Preparation): Hexokinase catalyzes the following reaction: ATP + Glucose --> ■ Glucose-6-P + ADP The measurement of its activity in the galactokinase preparation was based on the method of Leloir and Trucco (106), but with glucose replacing galactose. The concentrations of substrates (glucose and ATP) and buffer, and the procedure for incubation were similar to those described for the galactokinase assay. At the end of the incubation period, proteins were denatured by heating, and reducing sugar was determined (114). The amount of glucose which disappeared in the presence of ATP during incubation was translated into hexo kinase activity, expressed in umoles glucose 91 phosphorylated per mg. protein per minute. 3. Paper Chromatography: Unless otherwise stated, Whatman #1 filter paper and descending techniques were employed. Material was spotted with a 20-ul Sahli pipette. Spotting was done on a platform equipped with underlighting and a warm air stream for drying. The use of preliminary equilibration and the time of run depended on the particular solvent system used. At the end of the run, the paper chromatogram was air-dried. Components of interest on the paper were revealed by spraying with or dipping into a detection reagent, passing through a strip counter (for radioactivity), or by use of ultraviolet light. Heating in a 100-110° oven was used to develop color if necessary. a. Among solvent systems used were the following: (1) Methanol-formic acid-water (80:15:5, on volume basis) (127) . (2) 95% ethanol-ammonium acetate, 1 M, pH 7.5 (7.5:3 on volume basis) (128). (3) Pyridine- n-butanol-water (129) . One hundred ml. pyridine, 300 ml. n-butanol and 150 ml. water were mixed in a separatory funnel. After phase separation, the bottom layer was drained off and discarded. One hundred ml. pyridine was added 92 to the top layer, and the resulting mixture (a single phase) was used for chromatography. (4) n-propanol-ammonia-water (6:3:1 on a volume basis) (130). (5) 80% ethanol containing 0.647. boric acid (131). b. Among the detection reagents used were the following: (1) For inorganic phosphorus and phosphorylated compounds, the Hanes and Isselwood (132) reagent was employed. The reagent consists of a mixture of 5 ml. of 60% (weight by weight) perchloric acid, 25 ml. of 47. (weight by volume) ammonium molybdate, 10 ml. of 1 N HC1 and 60 ml. of water. (2) For reducing sugars, both aniline hydrogen phthalate reagent (133) and alkaline silver nitrate (134) were employed. The latter proved to be the more sensitive. (a) Aniline hydrogen phthalate: Phthalic acid (1.66 gm.) was dissolved in 5 ml. of 50% acetone-water. Aniline (1.86 gm.) was added and the solution was made up to 100 ml. with acetone. 93 (b) Alkaline silver nitrate: Two separate solutions were used. The first contained 0.1 ml. of saturated silver nitrate solution in 100 ml. of acetone. For the second, 0.5 gm. of sodium hydroxide was dissolved in 100 ml. of 95% ethanol. The paper was dipped through the silver nitrate reagent, and the acetone was allowed to evaporate from the paper. When dry, the paper was dipped through the alkaline reagent. The brown-black color of reduced silver developed immediately. (3) For organic acids, an aniline-xylose reagent was used (135): one gram of xylose was dissolved in 3 ml. of water; 1 ml. of aniline was added, and the solution was made up to 100 ml. with methanol. 4. Dowex-1-Formate Column Chromatography: The pro cedures used, although modified for specific purposes, have been derived principally from the work of Bartlett (136)and of Diedrich and Anderson (112). a. Preparation of Column: A piece of glass wool was inserted at the tip end of a 25 ml. graduate pipette (with top cut off). An aqueous suspension of Dowex-l-Cl (200-400 mesh) was added until the resin reached a settled level at the 10 ml. mark. Another small piece of glass wool was laid over the top of the resin. 0.25 M sodium formate was passed through until a negative chloride test to 0.25 M silver nitrate was obtained (112). One hundred ml. of distilled water was then passed through, and the column was ready for use. In preparation of larger columns, the process was similar. b. Concentration Gradient Elution: A solution to be examined (for example, supernatant fluid of an incubation mixture) was passed through the prepared Dowex-1-formate column. Twenty to thirty ml. of water was used to wash out the non-adsorbed materials. The adsorbed materials then were eluted by a concentration gradient formed by having 1 M ammonium formate (800 ml.) gradually run from a reservoir bottle into distilled water (600 ml.) in a mixing bottle containing a magnetic stirring bar. Fractions of eluate were collected in test tubes in a Technicon fraction col lector with timer control. Tube change was generally set for 5-10 minute intervals, depending on the rate of flow of eluate. Aliquots from each tube (containing about 7 ml.) were tested by chemical methods or for 95 radioactivity, depending on the nature of the eluted materials. 5 • Labeled Carbon Dioxide from Galactose-1 -C^: Galactose-1 -C^ was incubated iri vitro for 90 minutes with whole blood or leucocyte suspension by a procedure adapted from Weinberg (61). The incubation was carried out in a Dubnoff metabolic shaker. The incubation mixture was con tained in a 50 ml. Erlenmeyer flask, into which had been placed a vial (13 x 45 mm.) for the collection of . The flask was closed with a soft rubber serum stopper. At the end of the incubation, 1 ml. of hyamine was injected with a syringe needle through the rubber cap into the inner vial. Immediately following, 0.1 ml. of 10 N H2SO4 was injected, this time into the incubation medium itself. Shaking of the acid-denatured blood mixture in the incuba tor was continued for 1 hour to insure complete liberation of C^02 and its collection in the hyamine in the center vial. Then the vial was removed. The outside surface was carefully cleaned, and the vial and its contents were placed in a standard counting vial. Ten ml. of toluene- scintillator solution containing 50 mg. of dimethyl Popop [1, 4-bis-2-(4 - Methyl - 5 - Phenyloxazolyl) - Benzene] and 4 g. of PPO (2,5-Diphenyloxazole) per 1 liter of 96 toluene was added. The hyamine and the scintillators were well mixed by shaking. counts were measured in a Nuclear Chicago liquid scintillation spectrometer system. 6. Isolation of Glycolytic Intermediates from Incubation of Blood Cells: In the earlier experiments, the residues from labeled carbon dioxide evolution studies were the starting material for examining distribution of labeled intermediates from galactose-l-C^. Later, incuba tions were done specifically for the study of the metabolic products. The denatured incubation residue, after breaking up the coagulated mass with a stirring rod, was diluted to 60 ml. with water. The pH of the mixture was adjusted to 6.0 with 1 N NaOH. At this pH a large amount of protein precipitated. The suspension was centrifuged, and the supernatant fluid was applied to a Dowex-1-formate column as described previously. The intermediates were eluted by a gradient of 0 to 1 M ammonium formate. A volume of 0.3 ml. from each of the eluate tubes was checked for radio activity by planchet counting. When radioactive counts were plotted against tube number, the chromatographic pattern of distribution of intermediates in the eluate tubes became apparent. As appropriate, tube contents were composited for further studies. 97 It was found that the procedure used was by far the simplest method for removing proteins. Preparation of a protein-free filtrate by trichloracetic acid precipitation (Bartlett, 137) also was carried out, but the chromato graphic pattern obtained did not differ in any way from our procedure in which isoelectric precipitation was employed. The use of trichloracetic acid has the disadvantage of adding an ether extraction step for the removal of the tri chloracetic acid. This extraction also removes lactic acid (138), and perhaps other components as well. 7. Identification of Metabolic Intermediates of Galactose: a. Lactic Ac id: Paper chromatography was carried out in three different solvent systems: in the Bandurski acid solvent (127), in 807o ethanol containing 0.64% boric acid and in ethanol-ammonium acetate. Lactic acid was further characterized by the chemical determination of Barker and Summerson (118) and by the enzymatic determination of Neilands (120). b* UDPG and UDPGal: These compounds were adsorbed on a charcoal column. The particle size of charcoal was found to be critical. Coarse charcoal was ground to a size providing a suitable capacity for holding 98 UDPHexose and yet allowing a satisfactory rate of flow. Prior to use, the charcoal in the column was washed with water until no positive chloride test was produced with silver ion. After the solution to be examined had been applied to the column, non-adsorbed materials were washed out with water. Adsorbed materials then were eluted with 50% ethanol containing 0.1% of ammonia. Ethanol was removed from the eluate by evaporation under reduced pressure, and the aqueous solution was concentrated by lyophilization. After paper chromatography in an ethanol-ammonium acetate solvent system, the samples under test (concentrated column eluates in the UDPHexose region) showed ultra violet absorption spots on the paper exactly corres ponding to radioactivity as demonstrated by strip scanning and by radioautography. Chromatography of known UDPG, as well as a mixture of known UDPG and the solution tested, gave the same results. It was of interest, that the original UDPHexose region eluates exhibited maximum spectrophotometric absorption at 259 mu at pH 6.5 and that in this fraction the spectral changes characteristic for UDPG were observed at pH 2.0 and at pH 12.4 (8). When the fraction was subjected 99 to acid hydrolysis, the hydrolytic products, when chromatographed in pyridine-butanol-water system, showed radioactive peaks on a strip counter corresponding to galactose and glucose. c. Galactose-l-phosphate: Paper chromatography revealed that the sample and standard galactose-l- phosphate had the same Rf's in three different solvent systems. Upon acid hydrolysis, both sample and standard yielded spots corresponding to galactose. An equivalent amount of inorganic phosphate was detected. d. 2, 3-Diphosphoglyceric acid: The position in the column chromatographic pattern corresponded to that found by Bartlett (136) using the same concentration gradient elution with 0-1 M ammonium formate at pH 6.4. The compound was first precipitated with barium ion. It was then chromatographed on paper in comparison with authentic 2, 3-diphosphoglyceric acid in an ethanol- ammonium acetate solvent system. Correspondence of migration was obtained. Also, it was found that this intermediate could be readily labeled with inorganic 32 phosphate-P in red blood cell incubations, confirming the results of other investigators (139). 100 8. Incubation of Blood Cells for Extended Periods: In some of the experiments done it was necessary to carry on incubation of blood cells for as long as 24 hours. Pre cautions were taken to avoid contamination by micro organisms, and tests were done to insure metabolic viability of the cells at the end of the extended period. a. Incubation conditions: All glassware, acces sory equipment and reagents (with the exception of radioactive materials) were sterilized. Autoclaving was used for glassware, and reagents were passed through appropriate filters. Sterile techniques were applied to all handling of blood samples and of incubation medium. Whole bloods were incubated at room atmosphere in a cotton-stoppered flask (300 ml.) with 100 ml. of Hank's buffer at pH 7.4, and a glucose concentration of 105 mg%. The volume of blood used depended on the nature of the experiment. In some instances, whole bloods were preincubated with galactose-l-C^ for 90 minutes. In these experiments, prior to the 24 hour incubation, unreacted galactose- 1-C^ in the medium was removed by centrifugation, followed by washing three times with normal saline. 101 Incubations were carried out in a Dubnoff shaker at 37°C. b. Cell Viability: The metabolic viability of blood cells at the start and termination of 24 hour incubation was determined by separate experiments on the amount of produced from glucose-l-C^. RESULTS A modification of Weinberg's in vitro procedure for measuring carbon dioxide production from galactose by blood cells was applied to our own galactosemia patients, including affected sibling pairs in five families. Tue expected difference in carbon dioxide production between whole blood from galactosemics and from normals was found, but three points were of particular interest. Firot, there was variation among galactosemics in the capability of their blood cells co produce carbon dioxide from galactose. Second, three individuals differed markedly from the remaining fourteen. Third, correspondence of values in sibling pairs suggested a familial influence. To examine for individual differences in more detail, labeled intermediates in the residue after incuba tion with gala^tose-l-C^ were separated by gradient elu tion of Dowex-l-formate column^, (136) . In tne galactosemia patients, tne anticipated large amount of galactose-l- phosphate was found. Unexpectedly, a peak in the UDPHexose region also was obtained consistently. In the usual type of galactosemic subject (less tnan 200 cpm of labeled CO2), 103 the amount ot UDPHexose was very small. However, in the three unusual galactosemics studied (1306 cpm to 1829 cpm), the UDPHexose peaks were markedly larger, and an additional small peak in the diphosphoglyceric acid area was present. In attempting to explain the iormation of UDPHexose, consideration was given to the possibility that the block in galactosemia might not be complete and that existing methods of transferase measurement might not be suffi ciently sensitive to detect low levels of activity. Accordingly, the assay of transferase was re-examined, employing a procedure based on the conversion of carbon- labeled galactose-l-phosphate to labeled UDPHexose. Another hypothesis for explaining the presence of UDPHexose included the alternate pathway described by Isselbacher for galactose utilization in liver by way of UDPGdlactose pyrophosphorylase (43). Although Isselbacher as stated that this enzyme does not occur in erythrocytes (100), the possibility was re-examined in the present connection. Isselbacher's finding was confirmed. During the course of the investigation, a number of subsidiary studies were done. These included: (a) a series of experiments on the rate of conversion of 14 galactose-l-C into glycolytic intermediates, (b) work 104 with galactose pathway enzymes in human liver, (c) studies of erythrocyte galactokinase, (d) experiments on the mechanism of iju vitro uisposal of accumulated galactose-l- phosphate by galactosemic erythrocytes, ana (e) examination of the possibility tnat the transferase gene is in chromo some 21 . 1. Oxidation of Galactose-l-C^ by Whole Blood from Normal and from Galactosemic Sub j ects: The capacity of human blood cells to metabolize galactose tnrough the shunt pathway was assessed by the 14 14 production of C 0^ from galactose-l-C . The manometric method was not employea since the rate of utilization of oxygen in oxidation of galactose by erythrocytes, even by those from a normal subject, was known to be extremely small (31). Furthermore, as demonstrated previously in this laboratory, galactose by itself could not sustain adequate metabolic activity without addition of glucose to 14 the incubation medium. The collection of C 0^ was found to be a relatively simple procedure and suitable for the detection of very low metabolic activities (61) . 14 Galactose-l-C of high specific activity is readily available commercially. In Table III is shown the production of C^U2 from 105 TABLE III OXIDATION OF GALACTOSE-1-C14 BY WHOLE BLOODS FROM GALACTOSEMIC AND NORMAL INDIVIDUALS* Gal-l-C14 Normals C 02,cpm Galac tosemic s 14~ C O2,cpm Untreated M.F . 58,000 K.M. 1,155 J.L. 56,466 D.C. 1,785 T .A. 1,771 M.P. 1,123 Treated J.H. 41,631 R.O. 21 B.K. 39,500 J.D. 30 R.K. 44,358 L.M. 1,106 A.B . 34,113 J .F. 172 T.W. 48 D.C. 238 *Two ml. of whole blood were incubated with 0.9 uc of Gal-l-C^4 (sp.act. 1.69 mc/mM) for a period of 90 minutes. The collection of C- * - 402 was carried out accord ing to the method described in the text. The results shown are corrected for incubation controls done without the addition of blood. 106 14 galactose-l-C in preliminary experiments with whole bloods from normal and galactosemic individuals. At first the labeled galactose was used directly as received from the supplier ("untreated"): later the same galactose was processed on a Dowex-1-formate column to remove contami nants ("treated"). Tne contribution of contaminants is evident in Table III. Their nature was not determined, but it is clear from the data that some of them could be utilized by the cells to produce carbon dioxide. That the treatment used removed labeled bicarbonate or carbonate is indicated by the labeled carbon dioxide produced in incuba tion controls (not cited in Tctble III) of 50U-900 cpm. for the untreated galactose-l-C^ and 16-40 cpm. for the treated material. The preliminary experiments gave the large differ ence expected between carbon dioxide production by blood cells from normal individuals and by those from galacto semia patients. However, in examining the results with the 14 treated galactose-l-C , it is seen that ability of blood cells from galactosemics to metabolize galactose varies greatly from individual to individual. The variability indicates that the results are not a function of residual contaminants in the galactose-l-C^. A further point to 107 be emphasized is the relatively high carbon dioxide produc tion with blood from galactosemic subject L.M. When further experiments were done, the result was found to be a consistent one. The possibility was considered that L.M. might be similar to the galactosemic subjects reported by Segal, et al. (60) , by weinberg (61) aiid by Isselbacher (101) . Segal and his group found that two galactosemic individuals were able to produce radioactive carbon dioxide from labeled galactose in vivo in "normal" amount. Six other galactosemics studied had almost no radioactive carbon dioxide in exhaled air. Weinberg included one of Segal's subjects in his in vitro work and found that blood cells from this individual were able to oxidize galactose to produce carbon dioxide at a substantially greater rate than other galactosemics examined. This suggests that there is correlation, at least in this particular individ ual, between in vitro findings with blood and the total metabolic process in vivo. Recently, Isselbacher has had similar in vitro results with two other galactosemics among the group studied by him. It was of interest to find out if other galacto semic patients in the group available to us would give 108 results comparable to L.M. In order to examine possible genetic correlations, most attention was given to families in which affected sibling pairs were present. The results are shown in Table IV. With three of the individuals tested, blood cells were able to produce carbon dioxide from galactose at an elevated rate, in contrast to the low production rate for 14 other galactosemia patients (includ ing 6 from Table III not listed in Table IV). One of the 3 unusual patients again is L.M., while the second (A.M.) is a sibling of L.M. The third of the unusual individuals (L .N.) is a Negro. All of the other galactosemia patients included (except D.C. in Table III) are Caucasians. It is of interest that, throughout the series, results in each sibling pair are similar. The blood cells of the three unusual galactosemics produce carbon dioxide from galactose at 10 to almost 100 times the rate found with the other galactosemics studied. However, the increased rates are still only about 1/40 of those with blood cells from normal individuals. The results cited for L.N., L.M., and A.M. are not accidents of a single experiment. It has been possible to reproduce them on repeated samplings from these patients. Because of limitations of blood volumes available, 109 TABLE IV OXIDATION OF GALACTOSE-1-C14 BY GALACTOSEMIC WHOLE BLOODS WITH REFERENCE TO FAMILY STUDIES Subj ects Age Sex Race C^402,cpm L.N. 13 F N 1829 L.M. 3 F W 1724 A.M. 1 F W 1306 J.D. 2 F w 40 M.D. 4 M w 41 C .De. 12 M w 20 R.De. 11 F w 20 T.W. 2 F w 48 L.W. 6 mo. M w 0 J.B. 2 M w 21 Ja .B. 6 F w 20 (1) Conditions for the incubation and collection of CO2 were similar to those for Table III. 2 uc galactose-l-C^4 (sp.act. 1 mc/mM) per flask were used. (2) L.N. is Negro; all others are Caucasian. (3) Except for L.N., each subject is a member of a sibling pair. 110 it was not possible in most cases, to study carbon dioxide production both by whole blood and by separated leucocytes. In the three instances in which incubations were done with both preparations, the results were consistent with a body of data obtained by other workers in the same laboratory in similar experiments on normal bloods and on those from galactosemia heterozygotes. Leucocytes present in the whole blood contributed about 107o of the radioactive carbon dioxide counts. Leucocytes from normal individuals are more active than erythrocytes in the oxidation of galactose; in the case of the galactosemia patients, the same result was found. The carbon dioxide evolution experiments indicated that 3 of the galactosemics differ from the other 14 studied. Further experiments were necessary to provide a basis for explaining the differences found. As a first approach, examination was made of carbohydrate pathway intermediates present in incubation residues. 2. Carbohydrate Pathway Intermediates: Dowex-1-formate column chromatography was carried out to separate glycolytic intermediates remaining in the 14 residue after incubation of whole blood with galactose-l-C . Ill It was postulated that particular intermediates present, or their amounts, might give a clue to the mechanism by which three of the galactosemics among the group under study are able to produce carbon dioxide from galactose. A possible hypothesis is that of the presence of an alter nate pathway. For example, if the contention of Hsia and co-workers is correct, namely, that galactose-l-phosphate can be converted to galactose-1, 6-diphosphate and galactose-6-phosphate in galactosemic red blood cells, these phosphorylated intermediates should be labeled under the present experimental conditions, and should be found in the column effluents. In Figure 9 is shown the chromatographic pattern of galactose intermediates resulting from incubation of whole blood from a normal subject. Incubation conditions were similar to those used in the labeled carbon dioxide experiments described above. A similar pattern was obtained with the same person when tested on three differ ent occasions, thus substantiating the validity of the chromatographic analysis under these conditions. The first peak (X) corresponded to a contaminant found in all commer cial galactose-l-C1^ samples used. It was not metabolized by the cells, and its chemical characterization was not 112 5 0 100 TUBE NUMBER Figure 9. Glycolytic Intermediates of Galactose with a Normal Blood Experimental conditions: 4 m l . of whole blood were incu bated with 4 uc of galactose-l-C^ (0.6 mg.) and 4 m l . of K-R-B-glucose buffer, pH 7.4, for 90 minutes. 113 carried out. Eluates of peaks 1, 2, 4, 3, and 6 were con centrated by lyophilization. Chemical, cnromatographic and enzymatic analyses (carried out as described j.n the section on Methods) ohowed the presence of lactic acid, galactose-l-phosphate, UDPHexoses and diphosphoglyceric acid in peaks 1, 2, 4, and 6 respectively. Peak 3 has not been identified because of the very small activity present. Peak 5 contained at least three compounds; studies based on barium precipitation suggested the presence of hexose diphosphate and triose phosphates (140). Whole bloods from six more normal subjects were incubated with galactose-1-C^, and similar chromatographic patterns were observed. When whole blood from a galacto semia heterozygote was tested under the same experimental condition, no labeled lactic acid (Peak 1) could be detected (Figure 10). When whole bloods from six more galactosemia heterozygotes were subjected to incubation with galactose-l-C^ followed by column chromatography, the labeled lactic acid production was found to be absent or very small. The results with the heterozygote individuals at first were very puzzling. It was speculated that some other metabolic defect might be linked to the known galactose-l-phosphate uridyl transferase deficiency in 114 5 0 0 0 - WV yw vw CPM 5 0 100 TUBE NUMBER Figure 10. Glycolytic Intermediates from a Galactosemia Heterozygote Blood Incubated with Galactose-l-C 14 Experimental conditions were similar to those for Figure 9. 115 galactosemia. However, it was found that the results obtained were related to the duration of incubation and to the rates at which intermediates are formed from the galactose present. These considerations are treated in a separate section. Patterns with blood from galactosemic subjects also were examined. In six of the cases initially tested, all exhibited one major labeled peak, that corresponding to galactose-l-phosphate (Figure 11). Direct analyses of the appropriate eluate tubes confirmed the identity. A small, but noticeable peak (10-30 cpm above background per 0.3 ml. of eluate) corresponding to UDPHexose area also was present. In view of the known defect in galacto semia in galactose-l-phosphate uridyi transferase, the finding of galactose-l-phosphate accumulation was not sur prising. However, the indications for the presence of UDPHexose were quite unexpected. The seventh galactosemic in the initial group was L.M., one of the individuals whose blood cells produced labeled carbon dioxide from galactose-l-C^ at an elevated rate. The chromatographic pattern from experiments with this subject (Figure 12) contained, in addition to a heavily-labeled galactose-l-phosphate peak, a separate, 116 8000 - CPM v w V W W w v / wyw 2001- 100 TUBE NUMBER Figure 11. Glycolytic Intermediates of Galactose with a "Usual" Galactosemia Blood Experimental conditions: 4 ml. of whole blood were incu bated with 4 uc of galactose-l-C^ (0.52 mg.) and 4 ml. of K-R-B-glucose buffer, pH 7.4, for 90 minutes. Galactose- l-C-^ was treated. 117 8 0 0 0 - CPM yw v ^ v v v w w w 200 - I 00- 50 100 TUBE NUMBER Figure 12. Glycolytic Intermediates of Galactose with an "Unusual" Galactosemia Blood Experimental conditions were similar to those of Figure 11. 118 distinct peak in the UDPHexose region. The second peak was of relatively high activity, distinctly larger than for the other 6 galactosemics. The pattern was repeated twice at different times with fresh blood samples. Further studies on the same subject revealed the presence of a definite, though very small peak corresponding to 2, 3-diphosphogly- ceric acid (DPGA). As the work progressed, identical patterns, includ ing peaks for both UDPHexose and for DPGA, were found for subject A.M. (a sibling of L.M.) and for patient L.N. It appeared significant that these three were the individuals whose blood cells produced significant amounts of carbon dioxide on incubation with galactose. In contrast, experi ments with other galactosemics, finally numbering a total of 14, resulted in only a small UDPHexose peak, with no evidence for the presence of DPGA. Typical results are summarized in Figure 13. The identities of the compounds in the UDPHexose region were determined separately for normal controls and for each of the three unusual galactosemics. Column eluates from several experiments on the usual galactosemics were pooled. In all instances, upon paper chromatography with ammonium acetate-ethanol at pH 7.5 (128), the 20001- 1000? CPM 500 0 8000 CPM 1000 V w 200? 0 8000 CPM 1000 119 LA i GAL-I-P NORMAL UDPH A T -3 -P , / \ etc. ’ DPGA A "USUAL" GALACTOSEMIC G A L -I-P \A / W W W UDPH "UNUSUAL!'GALACTOSEMIC G AL-I-P w 200 UDPH DPGH 50 100 TUBE NUMBER Figure 13. Column Chromatography of Galactose Intermediates 120 radioactive material migrated to the same region as did known UDPGlucose or UDPGalactose. After acid hydrolysis and chromatography with pyridine-butanol-water (129) , both labeled galactose and labeled glucose were identified. All galactosemia bloods showed a ratio of glucose to galactose of 2 to 1: in contrast, the ratio for normal bloods was 1 to 3. The presence of carbon label in glucose and also in the DPGA region of the column effluents con firms the origin of both to be from galactose. It is considered significant that no evidence for the presence of labeled galactose-6-phosphate could be found in the experiments with blood cells from galactosem ics. Hsia and his associates (64, 130, 141), have stated that when galactose was incubated with galactosemic erythrocytes or hemolysates, galactose-6-phosphate and galactose-1, 6-diphosphate were formed as well as galactose -1-phosphate. They claimed that evidence for the presence ■ 4 of galactose-6-phosphate was found in five of six galactosemic subjects (142). The ratio of galactose-6- phosphate to glactose-1-phosphate was estimated to be 1:3. It was claimed further that galactose-6-phosphate could be oxidized by glucose-6-phosphate dehydrogenase isolated from human erythrocytes, the product being the 121 corresponding 6-phosphogalactonolactone (143). Further oxidation then would yield carbon dioxide. In considering possible participation of the path way suggested by Hsia, the behavior of galactose-6- phosphate in the Dowex-l-formate chromatographic system was studied. When known galactose-6-phosphate (Sigma, practical grade) was subjected to purification on a Dowex- l-formate column, a large peak, assumed to be galactose-6- phosphate, was followed by a small one which also reacted with cysteine sulfuric acid. The compound in the major peak was acid stable, and it reacted with Hanes-Ishwood reagent very slowly at 100°. The small peak, considered to be a contaminant, constituted about 10% of the total amount. Its nature was not established. It reacted readily with Hanes-Ishwood reagent on paper chromatogram, suggesting that it contained a labile phosphate group. In terms of elution tubes, the galactose-6-phosphate followed very closely the galactose-l-phosphate position. Labeled galactose-6-phosphate was not found in any of the incubates from galactosemic erythrocytes. Evalua tion was based on the relative ease of hydrolysis of galactose-l-phosphate compared with galactose-6-phosphate. The concentrated hexose monophosphate region from 122 Dowex-l-formate column eluates was subjected to acid hydrolysis (1 N HC1, 100°, 10 rain.), followed by paper chromatography. Only galactose was found, never any evidence for unhydrolyzed galactose-6-phosphate. It was concluded that in each case the hexose monophosphate present was galactose-l-phosphate. In a separate experiment, galactosemia hemolysate was incubated with galactose-l-C^ and ATP for the two hour period used by Hsia. By the procedure of analysis just described, only galactose-l-phosphate was found to be present. In another experiment, a 30-minute incubation was carried out, again only galactose-l-phosphate was obtained. If the contention of Hsia of 257o of the hexose phosphate being galactose-6-phosphate is correct, this amount should be detectable without any difficulty by the present methods. Hsia and his associates also reported that galactose-6-phosphate incubated with galactosemic hemoly sate was converted to galactose-l-phosphate and galactose- 1, 6-diphosphate, but that galactose-l-phosphate could not form galactose-6-phosphate or galactose-1-,6-diphosphate. They suggested that a hexokinase was present to phosphory- late galactose at the No. 6 position. In the present work, 123 the possibility has been considered that in the presence of glucose a kinase for the phosphorylation of galactose in position No. 6 might be inhibited, as in the case for fructose (144). In one experiment, glucose was excluded from the incubation medium and the galactosemia erythro cytes were pre-incubated for one hour prior to the addition of 80 mg7o galactose and 10 uc of galactose-l-C^. After further incubation for five hours, galactose intermediates were separated on a Dowex-l-formate column. Only a small peak corresponding to galactose-l-phosphate was found. However, it is possible that under the unphysiological condition of the experiment, the metabolic processes might not be fully functional. It has been noted in other work in this laboratory that metabolic viability of the red cells is markedly affected when glucose is not present or is present in insufficient amount. Under the conditions of the present experiments, galactose-6-phosphate does not seem to be a galactose intermediate in erythrocytes, at least not in detectable quantity. The basis for the discrepancy between Hsia's findings and the present ones is not known. It should be noted that Hsia and his associates employed relatively large volumes of cells (as much as 100 ml. of packed red 124 cells per experiment) and quite different conditions. Under some circumstances a galactose-6-phosphate pathway might contribute to the formation of carbon dioxide from galactose, but such a route does not appear to be a factor in the present studies. 3 . Galactose-l-phosphate Uridyl Transferase Activity in Hemolysates The possibility was considered that galactose-l- phosphate uridyl transferase might not be entirely absent from blood cells and that a very low activity might account for the small production and for the formation of UDPHexose. It was hypothesized that UDPG consumption or a manometric measurement, might not be sensitive enough for this purpose (39,40). Also, while the UDPG consumption procedure has been useful in laboratories studying galactosemia as a biochemical and a genetic entity, some aspects of the method have discouraged widespread routine application. Various steps are critical, and the cost of the test for isolated studies or for a limited series of assays also presents a problem, primarily due to the unit size of the commercially available UDPG dehydrogenase. It was considered desirable to investigate another approach 125 to the assay, with the hope of increased sensitivity and with the expectation of providing the basis for a more suitable routine procedure. An assay measuring the transfer of carbon-labeled galactose-l-phosphate to uridine diphosphogalactose would be direct, readily carried out and suited to a limited number of determinations. The procedure employed by Kalckar and his associates for the detection of transferase activity in human livers was modified for use with blood cells. The details of the procedure have been described in a previous section. Initially there was some concern that small amounts of native uridine triphosphate (UTP) might interfere with the assay according to the reaction: Gal-l-P + UTP <---* ■ UDPGal + PP When this was tested (Table V), it was found that addition of either UDPG or UTP increased the incorporation of galactose-l-phosphate into UDPGal in the presence of a normal hemolysate. However, the rate of reaction catalyzed by galactose-l-phosphate uridyl transferase according to the following equation: Gal-l-P + UDPG <— * • Glu-l-P + UDPGal was much greater than that co-factored by UTP. It became 126 TABLE V EFFECT OF UDPG AND UTP ON THE INCORPORATION OF Gal-l-C-I-P INTO UDPGal-l-C14 IN A NORMAL HEMOLYSATE Tube C.P.M. Denatured 116 Control 128 UDPG 1720 UTP 320 Incubation mixture contained 0.2 umoles Gal-l-C^-l-P (4,000 cpm) , 0.4 umoles UTP and 5 umoles MgClo £r 0.2 umoles UDPG, 0.4 mmoles glycine buffer (pH 8.1), and 0.3ml. 50% hemolysate with total volume of 1 ml. Incubation period was 30 minutes at 37°. 127 evident that native UTP would be insufficient to affect the transferase assay. As the work progressed, the result with added UTP was traced to the presence of small amounts of native glucose-l-phosphate (discussed in a separate section on UDPGal pyrophosphorylase). Other nucleoside triphos phates (ATP and TTP) did not affect the incorporation of galactose-l-phosphate into nucleotide-galactose. The assumption was made that UDPGal-4'-epimerase is inactive under the conditions of the assay (36). Upon this basis, the UDPHexose eluted from charcoal should be UDPGal. This was confirmed by acid hydrolysis of pooled eluates, followed by chromatography (butanol-pyridine). All the radioactivity was present in the galactose region of the chromatogram: no evidence for the presence of labeled glucose was found. In establishing conditions for the assay, a linear relationship between the product formed and the time of incubation during an experimental period of 30 minutes was found (Figure 14). Erythrocyte galactose-l-phosphate uridyl trans ferase values for 66 different individuals, as measured by the labeled galactose-l-phosphate procedure, are given in Table VI. Each value represents the average of duplicate 128 O UJ GO < * a. e o fc x q- _ Q) l i I CO £o 2 s o ° in _J a> < o o e 3 2 10 20 30 TIME (MINUTE) Figure 14 Utilization of Galactose-l-Phosphate in a Normal Hemolysate Incubation mixture contained 0.2 umoles Gal-l-C^-l-P (7,200 cpm), 0.2 umoles UDPG, 0.4 mmoles glycine buffer (pH 8.1), and 0.1 ml. 507o hemolysate with total volume of 1 ml. TABLE VI 129 ERYTHROCYTE GALACTOSE-1-PHOSPHATE URIDYL TRANSFERASE DETERMINED BY LABELLED GALACTOSE PROCEDURE Galactosemia Galactosemia Normal Down's Homozygotes Heterozygotes Controls Syndrome 0.008 0.97 1.80 2.46 -0.032 0.89 1.93 2.24 0.040 0.99 1.98 1.82 -0.008 1.03 1.90 2.07 -0.008 1.01 1.44 1.88 -0.012 1.19 1.99 1.92 0.028 1.04 1.84 1.73 0.004 1.03 2.01 2.08 -0.005 0.90 2.80 1.74 -0.002 0.84 1.78 2.27 -0.015 0.88 2.11 1.98 1.70 1.99 1.45 1.90 1.80 1.86 1.59 1.89 2.04 2.00 1.78 1.81 1.73 2.52 2.26 1.74 1.85 1.78 1.69 2.03 1.79 1.94 Number 11 11 24 20 Mean 0.002 0.98 1.89 1.98 S.E. 0.099 0.28 0.23 a. Activity expressed in terms of micromoles galactose-1- phosphate converted per ml. packed RBC per hour. b . Each value shown represents: (1) a separate individual, (2) average of duplicate determinations. c. Normal controls include 10 children and 14 adults. d. Negative values for galactosemia homozygotes represent effect of background correction. 130 determinations. Statistical evaluation showed agreement between duplicates to be within + 57. of the average value. No significant difference was found between results on normal children and normal adults, and consequently both have been included in a single control group. Both galactosemia homozygotes and galactosemia heterozygotes were examined. The group of Down's syndrom (Mongolism) patients were included in the present study because of reports suggesting that the gene concerned with galactose- l-phosphate uridyl transferase activity may be located in chromosome 21 (145, 146). The implications of the present findings will be discussed separately. Reproducibility of the assay from day to day was determined on two different samples. Hemolysates, held in frozen storage, were assayed on four separate days during a seven-day period. Variation from initial values did not exceed + 57>. A comparison of assays on the same samples between the labeled galactose-l-phosphate method and the UDPG consumption procedure is given in Figure 15. Statistical treatment of the data showed good correlation (Spearman U D PG CONSUMPTION METHOD (umoles/ml./Hr.) 131 7 6 5 4 3 2 2 3 LABELLED GALACTOSE-I-PHOSPHATE METHOD (um oles/m l./H r.) Figure 15. Comparison of Galactose-l-Phosphate Uridyl Transferase Assays by the Two Methods Employed ■ DOW N’ S SYNDROME A NORMALS • GALACTOSEMIC HETEROZYGOTES X COMPOSITES OF GALACTOSEMIC HOMOZYGOTES 132 rank correlation coefficient of 0.75, p <4 0.001).* As is evident from Table VI and Figure 15, the labeled galactose-l-phosphate procedure can discriminate between galactosemia heterozygotes and normal individuals, an important point if the assay were to be used in family studies. The radioactive method can be used interchange ably with the UDPG consumption method. The labeled galactose-l-phosphate procedure is more direct, less demanding in detail and well-suited to routine application. However, it also is evident from the results presented, that there is no increase in sensitivity with the radio active assay and that this particular objective was not attained. Of interest in this connection is that two of the unusual galactosemics (L.M. and A.M.) were included in the galactosemic group in Table VI. For both, transferase activity by the radioactive assay was zero. 4. UDPGalactose Pyrophosphorylase Activity in Hemolysates: The presence of UDPGal pyrophosphorylase in erythrocytes has not been demonstrated, although it has *The assistance of Mr. Malcolm Williamson with statistical interpretation is greatefully acknowledged. 133 been shown to be present in mammalian livers (43). The apparent UDPGal pyrophosphorylase activity noted with normal hemolysates upon addition of UTP (as shown in Table V) warranted further study. The presence of this enzyme could account for the formation of UDPHexose from galactose and for the C-^C^ produced from ga Lactose-1 -C^ by blood cells from some galactosemic individuals. Employing the methods described in a previous section, a series of experiments were carried out with hemolysates from normal individuals, galactosemia heter ozygotes and galactosemia homozygotes. As is shown in Table VII, there was some activity with all the normals tested, extremely low activity in heterozygous individuals and none in galactosemic subjects. As a matter of interest, a normal hemolysate was separated into two fractions by the use of a DEAE cellulose column. The hemoglobin came through the column: adsorbed non-heme proteins were eluted in a separate fraction. Both fractions were tested for transferase (labeled Gal-l-P method) and UDPGal pyrophosphorylase activities. Under the circumstances of this experiment, it was neces sary to store the fractions at 4°C for 10 days. In Table VIII it will be seen that almost 60% of the 134 TABLE VII INCORPORATION OF Gal-l-CI4-l-P INTO LABELED UDPGal IN THE PRESENCE OF UTP IN HEMOLYSATES* Activity in units** Normals: G.D. 0.13 M.F. 0.19 W.N. 0.15 A.B . 0.16 J.C. 0.18 J.L. 0.07 Heterozygotes: E.M. 0.01 Mrs. M. 0.02 I.M. 0 Mrs . C. 0.03 Mrs. A. 0.02 Homozygotes: B.P. 0 S.F. 0 D.C. 0 M.P. 0 C.D. 0 R.D. 0 J.B. (Jayne) 0 J .B. (Jeffrey) 0 *Conditions for the assay were similar to those described in Table V. **Activity was expressed as umoles of Gal-l-P consumed per hour per 1 ml. of red blood cells. 135 TABLE VIII SEPARATION OF ENZYMATIC ACTIVITY BY THE USE OF A DEAE CELLULOSE COLUMN* Gal-l-P Fraction uridyl transferase (C.P.M.) "UDPGal pyrophosphorylase" (C.P.M.) Fresh hemolysate 1784 272 Hemolysate (10 days at 4°) 768 56 Hemoglobin fraction (10 days at 4°) 0 0 Non-heme protein fraction (10 days at 4° 540 0 Recovery of Activity in non-heme protein fraction (%) 70 0 ^Experimental conditions were similar to those of Table V, except for the use of 0.1 ml. and 0.3 ml. of fractions for the assay of the transferase and UDPGal pyrophosphorylase, respectively. Activities are corrected for the controls. 136 transferase activity and about 80%, of the apparent UDPGal pyrophosphorylase activity were lost by the hemolysate on storage. Based on activity remaining in the hemolysate, 70% of the transferase activity was recovered in the non-heme protein fraction, but the activity exhibited in the presence of UTP was complete .y lost. The hemoglobin fraction contained neither transferase nor UDPGal pyrophos phorylase activity. The results of this experiment were thought to be significant in indicating that the apparent UDPGal pyrophosphorylase activity found in hemolysates might result from the presence of a compound (presumably glucose-1-phosphate) which segregated with the hemoglobin fraction. It has been mentioned by Isselbacher (43) that, in the presence of galactose-l-phosphate uridyl transferase, contamination with glucose-l-phosphate can lead to false conclusions regarding apparent UDPGal pyrophosphorylase activity. If UDPG pyrophosphorylase activity is present, UDPG can be formed: Glu-l-P + UTP «--» UDPG + PP By action of transferase, UDPGal can result: Gal*-1-P + UDPG*—* UDPGal* + Glu-l-P UDPG pyrophosphorylase has been demonstrated in hemolysates 137 by Kalckar and his associates, using a UDPG consumption method (38). Accordingly, tests were made for this enzyme using labeled glucose-1-phosphate* under experimental con ditions similar to those for the UDPGal pyrophosphorylase assay. In Table IX, it is shown that galactosemic and normal hemolysates had approximately equal UDPG pyrophos phorylase activity. When UDPG pyrophosphorylase activity was expressed in terms of umoles product formed per hour per 1 ml. RBC, the results found corresponded to those of Kalckar and his associates. Prior to carrying out further work on possible effects of UDPG pyrophosphorylase, the possibility was con sidered that there might be sufficient inorganic phosphate (final concentration of 1.5 x 10 ^ M) in the galactose-l- phosphate samples to inhibit UDPGal pyrophosphorylase activity. Such an inhibition had been found by Oliver in rat liver and guinea pig brain preparations (147). Similar *Glucose-l-C^-l-phosphate (specific activity, 37 mc/mM) was obtained from Dr. E. Neufeld, Department of Biochemistry, University of California, Berkeley. Paper chromatography with an ethanol-ammonium acetate solvent system showed that 107o of radioactivity was localized in the glucose region, probably as a result of decomposition. Prior to the assay, unlabeled glucose-l-phosphate was added to constitute a solution 2 x 10~3 m, radioactivity of 6,000 cpm per 0.1 ml. 138 TABLE IX ACTIVITY OF UDPG PYROPHOSPHORYLASE IN HEMOLYSATES FROM A NORMAL AND A GALACTOSEMIC SUBJECT* Cpm Normal in Ethanol Eluates Galactosemic Denatured 140 cpm - No UTP 192 cpm 188 cpm UTP 1632 cpm 1888 cpm Activity in units** 2 .06 2 .35 *Incubation mixture contained 0.2 umoles Glu-l-C^-l-P (5800 cpm), 0.4 umoles UTP, 5 umoles MgClo, 0.4 mmoles glycine buffer (pH 8.1),and 0.1 ml. 507o nemolysate with total volume of 1 ml. Incubation period was 30 minutes at 37°C. **Activity is expressed as umoles Glu-l-P consumed per hour per 1 ml. red blood cells. 139 results were found in the present study with red cell hemolysates (Table X). Increased concentrations of magne sium ions diminished the effect of phosphate, a phenomenon observed in many Mg-H- ion dependent reactions. On the basis of the data in Table X, the maximum inhibition of UDPG pyrophosphorylase by the amount of inorganic phosphate in galactose-l-phosphate samples was calculated to be less than 20%. If the effect upon a hypothetical UDPGal pyro phosphorylase were to be similar in magnitude, it is unlikely that the absence of UDPGal pyrophosphorylase activity in galactosemic hemolysates could be ascribed to phosphate inhibition alone. The effect of unlabeled Glu-l-P upon incorporation of Gal-l-C^-l-P into uridine nucleotide sugar was studied. Table Xlshows that Glu-l-P at a concentration of 2 x \0~^ M slightly increased the incorporation. At 2 x 10”^ M Glu-1- P, the increment was ten fold in a normal hemolysate. No effect was observed with a galactosemic hemolysate. On the basis of these results, it was estimated that a concen tration of 1.4 x 10 ^ M of Glu-l-P in the incubation mixture could account for the activity observed with normal hemolysate in previous experiments. For this reason, as the studies continued, controls containing added 140 TABLE X INHIBITION OF UDPG PYROPHOSPHORYLASE BY INORGANIC PHOSPHATE* Mg Pi UDPGlu-l-C14 C.P.M. % In hibition 0.005 M - 1172 - 0.005 M 0.005 M 508 56.7 0.005 M 0.0005 M 892 24.0 0.1 M - 956 - 0.1 M 0.005 M 848 11.4 0.1 M 0.0005 M 948 0.7 *Experimental conditions were similar to those of Table IX except for the addition of MgC^ and KH PO. as indicated. 2 4 141 TABLE XI INFLUENCE OF UNLABELED Glu-l-P ON THE INCORPORATION OF Gal-1-C14-1-P INTO UDPGal-l-C14 IN THE PRESENCE OF UTP AND HEMOLYSATES* Glu-l-P Cone . Incorporation Activity Normal Galactosemic 0 0.14 u 0 2 x 10'6 M 0.16 u - 2 x 10"5 M 0.33 u 0 ^Experimental conditions were similar to those for Table V except for the addition of unlabeled Glu-l-P as indicated. Activity is expressed as umoles Gal-l-P consumed per hr. per 1 ml. red blood cells. 142 glucose-1-phosphate were included (Table XII). There was no influence upon the assay with freshly prepared galacto semia hemolysates, but the addition of glucose-1-phosphate led to a marked increase of incorporation of galactose-1- C^ label into UDPHexose with freshly prepared normal and galactosemia heterozygote hemolysates at a glucose-1- phosphate concentration of 2 x 10"^M. The incorporation was greatly reduced when 2-3 day old hemolysates were used. This may account for the difference in magnitude of incor poration between the results in Table VII and Table XII. These results suggest that the incorporation found reflects the presence of galactose-1-phosphate uridyl transferase and that it does not represent a separate UDPGalactose pyrophosphorylase activity in erythrocytes. A native glucose-l-phosphate level in erythrocytes, estimated at -4 5 x 10 M, would permit the incorporation observed. It is not likely that reagents would contribute sufficient glucose-l-phosphate to be significant. With a concentra tion of galactose-l-phosphate in the incubations at 2 x 10"^ M., the material used would have to be 107c contaminated. As has been shown in a previous section, the galactose used in preparation of labeled galactose-l-phosphate con tained at most no more than 27, of glucose. Furthermore, 143 TABLE XII PYROPHOSPHORYLASE ASSAY* Hemolysate Umoles Labeled per hour per ml UDPHexose formed . packed cells** Without G-l-P G-l-P Added (2 x 10"5 m) Normals: B.K. 0.17 0.61 W.N. 0.30 0.63 M.S. 0.35 0.60 G.S. 0.25 0.65 H.F. 0.27 0.71 Heterozygotes: F.G. 0.08 0.25 M.G. 0.09 0.21 F.M. 0.16 0.35 M.D. 0.19 0.44 M.B. 0.10 0.31 Homozygotes: T.A. 0 0 B. 0 0 G. 0 0 L.M. 0 0 A.M. 0 0 *Incubation mixture contained 0.2 umoles Gal-1-C^-l-P (15,000 cpm in liquid scintillation system), 0.4 umoles UTP, 5 umoles MgCl2, 0.4 mmoles glycine buffer (pH 8.1), Glu-l-P at final concentration as indicated, and 0.3 ml. of freshly prepared 50% hemolysate with total volume of 1 ml. Incubation period was 30 minutes at 37°. **Net increase over control without UTP. 144 hexokinase present in the crude galactokinase would con vert glucose to glucose-6-phosphate. Also, equilibrium in the phosphoglucomutase reaction heavily favors glucose-6- phosphate. In one experiment the level of galactose-l- phosphate was increased to three times that in the usual transferase assay. Seven times the usual amount of hemolysate (from L.M.) was used. The incubation period was prolonged to two hours, and the amount of nucleotide used was doubled. Under these conditions, both UTP and UDPG stimulated some incorporation. However, the radio activity eluted from charcoal was only 2-3 times that of background and double that of the blank. The activity expressed in terms of umoles of galactose-l-phosphate consumed per hour per 1 ml. packed red blood cells was only 0.0025 to 0.005 umoles. This small activity was difficult to interpret. While it may represent UDPGal pyrophosphorylase activity, it is more likely that it represents the effect of a very low transferase activity in the presence of native glucose-l-phosphate and UDPG pyrophosphorylase. The present results are concordant with Isselbacher1s earlier statement concerning the lack of 145 UDPGal pyrophosphorylase activity in erythrocytes. The formation of labeled UDPHexose upon incubation of galactosemia erythrocytes with galactose-l-C^ must be upon some other basis. Since no evidence for any other possible route could be found, it was considered that the most likely possibility is that the enzymatic block in galactosemia is incomplete. 5. Subsidiary Studies of Interest 14 a. Rate of Conversion of Ga'lactose-l-C into Glycolytic Intermediates; Previously described experiments on galactose metabolism with blood from galactosemia heterozygotes (Figure 10) showed a greatly reduced formation of labeled lactic acid from galactose-1 -C^ as compared to findings with normal bloods. Since it was con sidered improbable that a second genetic defect is present in galactosemia, linked to the transferase deficiency, the possibility was entertained that the results obtained with heterozygotes might be due to the rates at which the carbon label in galactose appears in successive intermediate products. It is 146 known that galactosemia heterozygotes have approxi mately half the normal quantity of erythrocyte galactose-l-phosphate uridyl transferase (53), and it was thought that this limitation might be sufficient to delay the appearance of lactic acid. Also, the possibility considered was that production of lactic acid might be inhibited by accumulation of galactose- l-phosphate . As a first step, whole bloods from normal, heterozygous and homozygous subjects were incubated with glucose-l-C^, rather than with labeled galactose. In Figure 16 it is shown that identical chromato graphic patterns of glycolytic intermediates were for all subjects, including the heterozyotes. Due to dilution of isotope by unlabeled glucose (about 100 mgT, in final concentration) the height of the radioactive peaks was greatly decreased as compared with experi ments with galactose of equal radioactivity. The highest labeled peak was found to contain diphosphoglyceric acid. Incubations also were done with washed blood cells without plasma, a major source of glucose dilution. To check whether washing the cells would affect their 147 1000 NORMAL CPM 500 1000 HETEROZYGOUS CPM 500 0 1000 HOMOZYGOUS CPM 500 50 100 TUBE NUMBER Figure 16. Glycolytic Intermediates of Glucose Incubated with Whole Blood Experimental conditions were similar to those of Figure 9 except for the use of 4 uc of Glu-l-C14 (sp.act. 2mc/mM). 148 metabolism, patterns were compared in the following incubations: one with whole blood and the other with washed blood cells incubated in a medium containing labeled galactose and 10 mg.7o glucose. In Figure 17, it can be seen that there was no difference in glyco lytic intermediate patterns under these conditions for a heterozygous subject. Furthermore, the lack of lactic acid is apparent in both cases (peak 1). The chromatographic patterns of glucose glycolytic intermediates after incubation of washed blood cells of normal, heterozygous and homozygous persons were compared (Figure 18). The results show that there was again not much difference among these individuals. When the patterns were compared with that of galactose intermediates found in incubations of blood from normal persons, it is recognized that lactic acid (the first highly labeled peak) was present in the glucose incubation. Although the degree of radioactivity varied among different peaks, pattern of glycolytic intermediates of glucose was similar to that of galac tose in normals. This is in accord with the trans formation of galactose to enter the glucose glycolytic pathway and with the known fact that galactosemic 149 2000 WASHED RBC 1000 CPM 200 2000 WHOLE BLOOD _ 0 l 1000" CPM 50 100 TUBE NUMBER Figure 17. Glycolytic Intermediates of Galactose with Galactosemia Heterozygote Washed and Unwashed Blood Cells Experimental conditions: For whole blood incubation, condi tions were similar to those described for Figure 9. For washed RBC, the cell concentration was reconstituted to original packed cell volume prior to incubation in 10 mg.7o glucose; otherwise the conditions were the same for whole blood. 150 N O R M A L C P M 200 2000 f - 1000 " C P M 200 2000 HETEROZYGOUS I 000 C P M 200 HOMOZYGOUS 50 100 TUBE NUM BER Figure 18. Glycolytic Intermediates of Glucose with Washed Blood Experimental conditions were similar to those in Figure 17. 151 subjects have no difficulty in metabolizing glucose. The results indicated that the observations on lactic acid formation from galactose by cells from heterozy gotes could not arise from an associated defect in the glycolytic pathway. Further, it was decided that problem must be concerned with the path from galactose to glucose-6-phosphate. To examine the possibility that the failure of lactic acid to appear was due to inhibition by a galactose intermediate, presumably galactose-l- phosphate, the effect of accumulated galactose-l- phosphate in the cells on the labeling of glucose intermediates was studied. Blood from a heterozygous individual, known in separate experiments not to pro duce labeled lactic acid from galactose-l-C^ in 90 minutes, was preincubated with unlabeled galactose (7.5 mg%) for 90 minutes. The blood cells then were washed and were reincubated with 10 mg.T, glucose con taining 4 uc glucose-l-C^ for a further 90 minutes. In Figure 19 it is apparent that preincubation of cells with galactose had no effect on the appearance of lactic acid. Although, based on manometric determina tions, glucose is said to have a much higher metabolic 152 5 000- WITHOUT GALACTOSE vw I 000 200 5000 - WITH GALACTOSE VAA/ vw 1000 b w v w 200 ■ 50 100 TUBE NUMBER Figure 19. Glycolytic Intermediates of Glucose Incubated with Washed Blood of a Heterozygous Subject Experimental conditions are described in the text. 153 rate than galactose (31), it was considered that some effect of galactose-l-phosphate should have been apparent if the inhibition hypothesis was a valid one. A more stringent test was made by using blood cells from a galactosemia homozygote. It was known that such cells can accumulate galactose-l-phosphate in substan tial amount. Further, as has been shown above, glucose metabolism in galactosemia erythrocytes is unimpaired. In the experiments done, a significant amount of galactose-l-phosphate had accumulated at the end of a 90-minute pre-incubation period (more than 8 mg.70, by direct measurement). The washed cells were then re-incubated with labeled glucose. Even under this condition, no effect of galactose-l-phosphate in inhibiting lactic acid formation was observed (Figure 20) . Experiments were done to test the possibility that the absence of labeled lactic acid after a 90-minute incubation of galactosemia heterozygote blood with galactose-l-C^, might be due to a decreased rate of formation. To test this hypothesis the incubation period was prolonged to 180 minutes, employing whole blood from the same heterozygote used in previous 5 0 0 0 1 PREINCUBATED WITH GALACTOSE AND GLUCOSE 1 5 n CPM 200 5 0 0 0 PREINCUBATED WITH GLUCOSE n 1000 CPM 50 100 TUBE NUMBER Figure 20. Glycolytic Intermediates of Glucose Incubated with Washed Blood of a Galactosemic Subject Experimental conditions are described in the text. 155 experiments. The medium was similar to that described for Figure 9. A substantial amount of labeled lactic acid was found upon column chromatography in the Dowex- 1-formate system. This result led to a study of the rates of appearance of successive intermediates upon incubating blood from normal individuals with labeled galactose. Incubations were carried out as previously described, but for varying lengths of time, followed by separation of intermediates by gradient elution in the Dowex-l-formate system. In Figures 21 and 22 are shown the labeled intermediates present after 5, 15, 30, and 60 minutes of incubation. Contaminants were present in the area designated as "X." Within 5 minutes, the carbon label was present in galactose-l- phosphate and UDPHexose. A small amount of activity in the hexose diphosphate/triose phosphate also was detectable. At 15 minutes, diphosphoglycerate appeared. While the level of labeled UDPHexose increased slightly, finally reaching a steady state level, galactose-l- phosphate continued to accumulate, eventually to a level twice that present at 5 minutes. At 30 minutes, accumulation of galactose-l-phosphate still was 156 8 00 CPM 100 0 5 MINUTES 15 MINUTES 800 CPM 100 TUBE NUMBER Figure 21. Timed-Labeling Patterns of Galactose Inter mediates of Normal Erythrocytes Experimental conditions were similar to those for Figure 9 No. 1, 2, 3, 4, 5, 6 refer to the following compounds, respectively: lactic acid, Gal-l-P, not identified, UDPHexose, triose phosphate (etc.), and DPG. X denotes contaminants. 2000| - lOOOi- 30 MINUTES CPM 2 2000 157 60 MINUTES CPM 50 100 TUBE NUMBER Figure 22. Timed-Labeling Patterns of Galactose Inter mediates of Normal Erythrocytes Experimental conditions were similar to those for Figure 9 No. 1, 2, 3, 4, 5, 6 refer to the following compounds, respectively: lactic acid, Gal-l-P, not identified, UDPHexose, triose phosphate (etc.), and DPG. X denotes contaminants. 158 continuing. While radioactivity in UDPHexose remained relatively constant, radioactivity in the triose phosphate and diphosphoglycerate areas increased. At 60 minutes, galactose-l-phosphate was still accumulat ing, and labeled lactic acid began to appear. Diphosphoglycerate steadily increased in amount. This experiment was repeated several times. Although there was some variation from individual to individual, the general pattern of results was sub stantially the same. It was apparent that even with normal individuals, the rate of progress through the pathway from galactose is relatively slow in the erythrocyte. The findings have been sufficiently interesting to justify a more extensive investigation, to be undertaken separately from the present study, b. Galactose Pathway Enzymes in Human Liver: During the course of this work, it was possible to test for the presence of UDPGal pyrophosphorylase in a limited number of human liver biopsy and autopsy samples. Isselbacher (63) has suggested that activity of this enzyme in human liver might increase with age and that an increased capacity of this secondary pathway might be a mechanism for galactosemics to adapt 159 galactose. However, data in the literature on this enzyme in human liver is scanty, and the hypothesis of adaptation has not been supported on clinical grounds. The present work does not contribute further informa tion on the adaptation problem, but it lends emphasis to the value of continuing such investigations. In Table XIII are shown the results of studies on three different enzymes in human liver preparations: Gal-l-P uridyl transferase, UDPGal pyrophosphorylase, and UDPG pyrophosphorylase. The most interesting case is that of patient E.G., an 11-year old girl who died of circulatory failure resulting from an anomalous pulmonary venous return. It had never been suspected that this patient might be galactosemic, yet her liver showed very low transferase activity and no UDPGal pyrophosphorylase activity. On the other hand, UDPG pyrophosphorylase activity was normal, as was that of glucose-6-phosphate dehydrogenase (not shown). Family history revealed that one uncle is in a mental insti tution and that one sibling had died of liver disease. The present results suggest that undetected galacto semia may have been present in E.G. Unfortunately it was not possible to carry out transferase studies 160 TABLE XIII ACTIVITY OF Gal-l-P URIDYL TRANSFERASE, UDPGal PYROPHOSPHORYLASE AND UDPG PYROPHOSPHORYLASE IN HUMAN LIVER PREPARATIONS* Patient Transferase UDPGalPP UDPGPP E.G. (Autopsy) 5.7 0 65 P.R. (Biopsy) 41 30 72.6 R.W. (Biopsy) 42 40 63.5 E.J.B. (Autopsy) 30 2.6 32 ^Activity was expressed as umoles of labeled hexose phosphate incorporated into nucleotide per hour per gram protein. Conditions for incubation of liver homogenates were similar to those used for hemolysates. 161 on the parents of this child. P.R. and R.W. are patients with glycogen-storage disease. Activity for each of the three enzymes was demonstrated. So far as is known, persons having this metabolic disorder are normal with respect to galactose metabolism. The other case is that of E.J.B., a 4-year old boy who died of a ventricular septal defect. He had been seen in the Child Development Clinic at Childrens Hospital on two occasions, and he was regarded as mentally retarded. While there was a substantial amount of transferase and UDPG pyrophosphorylase activity in the liver, UDPGal pyrophosphorylase activ ity was very low. No logical explanation of this finding can be made. c. Galactokinase in Human Erythrocytes: Galactokinase plays an important role in the metabolism of galactose. It mediates the phosphoryla tion of galactose, a process considered by some to be the rate-limiting step in the overall conversion of galactose to carbon dioxide. It has been speculated that differences in galactokinase activity may be responsible for variation in clinical findings in 162 galactosemia, particularly with respect to milk intolerance. These unsettled questions led to the present study on galactokinase activity in human hemolysates. A further stimulus was a report by Segal e_t al. (148), who demonstrated that the galactose consumption by liver from newborn rats is much higher than that by adult liver. This is directly in contrast to the hypothesis of Isselbacher that the ability to metabo lize galactose increases with age (63). Segal and his associates speculated that the increased ability in the newborn rat might be parallel to galactokinase activity. (1) Standardization of the Assay Conditions: Since hemolysates contain a relatively high adenosine triphosphatase (ATPase) activity, the spectrophotometric method used by Heinrich (107) was found to be unsuitable. Also, the large amount of reducing substances present in hemoly sates, even after incubation with galactose, prevented use of the method of Leloir and Trucco (106). The galactokinase assay used was a modification of the procedure of Sherman and Adler 163 (126), employing radioactive galactose. In the chromatography used, galactose was separated readily from galactose-l-phosphate. Since the test is not carried out with an isolated enzyme system, the main disadvantage of the assay is the possibility of confusing galactokinase activity with phosphatase splitting of galactose-l- phosphate. However, it was found that the activity observed under the conditions adopted was due primarily to the action of galactokinase. This consideration was supported by the finding that additions of ATP to the incubation medium markedly influenced the yield of galactose-l-phosphate. The optimum concentration of added Mg++ was found to be 8 x 10“-* M (Figure 23) . This is very close to the concentration in intact red cells (1 x 10-<^ M) . However, when the concentration was increased ten to one hundred times, the amount of galactose-l-phosphate formed decreased. This phenomenon possibly may be explained by the activation of phosphatase in the presence of the higher Mg++ concentration. In a study using galactose-l-phosphate as substrate, a Mg++ GALACTOSE-I-PHOSPHATE o CD tr T3 Q) E CO O E 3 0 .3 0 0.20 • 0.10 8xl0"6 M 8xlCT5M 8xI0‘4M 8x|0“3M Mg++ CONCENTRATION Figure 23. Effect of Exogenous Mg-H- on Galactose-l-Phosphate Formation Experimental conditions are described in the Methods section. 164 165 _ 3 concentration of 8 x 10 M (final concentration) increased the hydrolysis of galactose-l-phosphate six times. At the optimum concentration of Mg++ (8 x 10”^ M), the rate of formation of galactose-l- phosphate was linear during an incubation period of 90 minutes with a freshly prepared hemolysate from a normal adult (Figure 24). Galactokinase in hemolysates was found to be very unstable: 90% of the original activity was lost after 6 days of storage in the frozen state. It was necessary to carry out the galactokinase assays on freshly prepared hemolysates. (2) Galactokinase Activity in Relation to Age: Persons of different ages were tested for hemolysate galactokinase activity. Bloods from newborn infants were obtained from Los Angeles County Hospital. In Figure 25 it is shown that erythrocyte galactokinase activity was found to be very high in the first day or two of life and that it decreases markedly within the first year. Beyond this age the activity reaches a plateau region. While the values found in newborns varied GALACTQSE-I-PHOSPHATE (umoles formed/ml. RBC) 0.4 0.3 0.2 10 20 30 40 50 60 70 80 90 TIME (MINUTE) Figure 24. Formation of Galactose-l-Phosphate at Optimum Concentration of Mg-H- 166 GALACTOKINOSE ACTIVITY (umoles GAL-I-P formed/hr./ml. RBC) 1.2 - NEW BORNS l.l • 1.0 .• • 0.9 • • • • 0.8 • a 0.7 • •••• • • • • 0.6 . • • • • 0.5 • • # - • • 0.4 - • • • 0.3 - _ • • • • • _ • m • • • • • • . . • 0.2 - 0,1 " 1 DAY IYR. 0 4 -------» ^ ^ ------------1 ------------1 -----------1 ---------- 1 ------- i 10 20 Figure 25. 30 40 50 AGE (YEARS) Influence of Age on Galactokinase Activity 167 greatly among themselves, the mean value was 2.5 to 3 times higher than that for adults or for children. No difference in erythrocyte galacto kinase activity was observed between galactosemic and normal subjects of the same ages. It is planned to extend the erythrocyte galactokinase studies to include many more values, especially during the first year of life, and to measure correlated parameters. Of importance will be Gal-l-P uridyl transferase activity, of Gal-l-P 14 accumulation and production of C O2 from labeled galactose. Parallel studies on leucocytes will be of interest. d. Mechanisms for Disposal of Galactose-1- Phosphate by Erythrocytes in Vitro: It is known that when galactose is administered in a single dose of 35 grams per square meter to galacto semic individuals, erythrocyte galactose-l-phosphate reaches levels of up to 26 mg.70 in five hours (62). However, at the end of 24 hours, the levels have dropped to 2-3 mg.%. It is evident that erythrocytes have some mechanism for the disposal of accumulated galactose-l-phosphate in vivo. Based on experience 169 14 with C 0^ production in previously described experi ments, it was estimated that glycolysis could not account for this phenomenon. A search was undertaken to determine the mechanisms operating. The approach planned was a simple one. Erythrocytes from a galacto semia patient would be pre-incubated with galactose-1- under conditions favoring accumulation of labeled galactose-l-phosphate within the cells. After washing to remove external carbon label, the cells would be incubated in a suitable medium for 24 hours. At inter vals, samples would be withdrawn and appropriate analyses would be made. The in vitro experimental plan proved to present a number of practical difficulties. In one of the early experiments, there was no decrease in intracellular galactose-l-phosphate. It was found that the pH had dropped from 7.4 to 6.8. As a result, Hank's buffer (pH 7.4) was substituted for the Krebs Ringer phosphate buffer first used. Another facet of the pH problem was the accumulation of lactic acid during the 24 hours. Since the necessity of maintaining isotonicity limited the buffer strength that could be used, the buffer volume was made very large (100 ml.) in relation to 170 that of the cells (about 2 ml.)' In another early experiment, pH was maintained, but again there was no decrease in intracellular galactose-l-phosphate after 24 hours. Upon checking 14 - 1 A C 0^ production by incubation with glucose-l-C , it was found that a very substantial fraction of metabolic capability had been lost. When analysis was made of the glucose content of the external medium, it was realized that the sugar provided had been consumed well before the end of the 24 hour period. The conditions finally adopted embodied the find ings of the early experiments. Pre-incubation with galactose-1-C^ was carried out under conditions 14 previously described for the C 0^ experiments. The cells then were washed in normal saline. In some experiments, the cells were reconstituted to the original blood volume (4 ml.) with their own plasma. The glassware employed was autoclaved, and the Hank's buffer for the 24 hour incubation was sterilized by filtration. The technique for viability testing of cells at the end of 24 hours was essentially that used in the studies, but employing glucose-l-C^ as substrate,; culturing of the medium was done at the end 171 of most of the experiments. No evidence for bacterial contamination was found. In general, hemolysis during the 24 hours was negligible. Observations made on the conditions in a typical experiment are summarized in Table XIV. It will be noted that after an initial rise, the pH of the exter nal medium was near starting level after 24 hours. Metabolic viability of the cells was maintained. Two separate types of experiments were carried out to determine whether or not significant amounts of accumulated galactose-l-phosphate would disappear from the cells under the conditions adopted. In a non labeled experiment, whole blood from galactosemic patient D. C. was pre-incubated with 80 mg.7o galactose. It was found that 12 mg.% of erythrocyte galactose-l- phosphate had accumulated. After 24 hours the level decreased to 4.4 mg.7c. In a labeled experiment, cells 14 were pre-incubated with 7 mg.% galactose-l-C (sp.act. 1.32 mc/mM) without additional carrier. During 24 hour incubation, one-half of the original erythrocyte radioactivity (corresponding to galactose-l-phosphate as determined by column chromatography) disappeared. A corresponding amount of radioactivity appeared in the 172 TABLE XIV CELL VIABILITY, pH, GLUCOSE CONCENTRATION AND LACTATE CONCENTRATION IN A 24-HOUR INCUBATION OF NORMAL WHOLE BLOOD* Time Viability**,C^^O^ pH *** Glucose*** Lactate*** Ohr s. 15,955 cpm 7.4 98 mg .7, 1 mg.7> 4 7.8 99 4 8 7.8 92 7 24 16,675 7.3 90 15 *Conditions for incubation are described in the text. **Cell viability was based on the amounts of C O2 produced from galactose-l-C^ at zero time and at the end of 24 hours. Two ml. of reconstituted blood suspension corresponding to the original packed cell volume was incubated with Gal-l-C^ (sp. act. 1.32 mc/mM) and galactose carrier in a total weight of 0.3 mg. and 2 ml. of K-R-B-glucose solution. The collection of CO2 was carried out as described in the section on Methods. Activity was expressed as cpm/3 x 10^ RBC/mm3. ***pH, glucose and lactate concentrations were determined in the supernatant fraction. 173 external medium. The radioactive material in the external medium was determined to be principally galactose. Practi cally all of the radioactivity went through a Dowex-1- formate column into the filtrate. The filtrate was then passed through a Dowex-50 column (H+ form). Again the filtrate contained all the radioactivity. The Dowex-50-treated filtrate was neutralized with ammonia and then was lyophilized. The concentrate was chromatographed on paper using a butanol-pyridine- water solvent system. More than 90% of the radio activity was located in a zone corresponding to galactose. One minor peak (not glucose), which migrated much slower than would be expected for galac tose in the same solvent, was not identified. Upon treatment of the concentrate with baker’s yeast, the radioactive spot corresponding to galactose remained: a control experiment showed that added glucose (100 mg.%) was consumed, but not galactose (100 mg.7o). Three hypotheses were considered for the finding of galactose-l-C^ in the external medium: (1) accumulated galactose-l-phosphate might diffuse out of the cell and then become hydrolyzed by plasma 174 phosphatase, (2) a reversal of the galactokinase reaction in the cell membrane might be possible, (3) a non-specific or specific phosphatase in the cell might hydrolyze galactose-l-phosphate to galactose, with free galactose then diffusing out of the cell. The presence of a plasma phosphatase, capable of hydrolyzing galactose-l-phosphate, was demonstrated. Labeled galactose-l-phosphate was incubated with plasma. Upon paper chromatography in an ethanol- ammonium acetate solvent system, one half of the radio activity was localized in the galactose region. How ever, in experiments in which plasma had been excluded from the 24 hour incubation medium, galactose was isolated from the external medium, demonstrating that plasma was not necessary for the appearance of galac to se-l-C^. As shown in Figure 26, the decrease of radio activity from the cells corresponded to the rise of radioactivity in the external medium, with a plateau being reached at about 8 hours. Only about half of the intracellular galactose-l-phosphate was converted into extracellular galactose. The mechanism could not be attributed to simple diffusion, since disappearance of 175 800 700 HEMOLYSATE 600 ■ x x § PLASMA • • c PLASMA 500- CPM 400 300 0 4 8 12 16 20 24 HOURS 800 SUPERNATE 700 - 600 • 500 CPM 400 - 300 200- 100 - 0 4 8 12 16 20 24 HOURS Figure 26. Distribution of Radioactivities During 24-Hour Incubation of a Galactosemia Blood Experimental conditions: 8 ml. of galactosemia whole blood were pre-incubated with 16 uc Gal-l-C^-^ (sp.act. 1.32 mc/mM) without any carrier and 8 ml. Hank's glucose buffer,pH 7.4, for 90 minutes. Washed cells were then reincubated with 100 ml. Hank's glucose buffer with or without 2 ml. plasma. 176 galactose-l-phosphate took place only when the cells were metabolically active. It was concluded that an enzymatic reaction or other active process must be involved. If intracellular galactose-l-phosphate splitting were to take place by a reversal of the galactokinase reaction, one would expect the disappearance of labeled galactose-l-phosphate to be slowed down by re incubation in a galactose-containing medium (149). To test this hypothesis, galactosemic whole bloods were pre-incubated under the conditions described for the experiment of Figure 26. The cells, containing about 10 mg.% galactose-l-phosphate, were washed and divided into two portions. One was re-incubated in 100 ml. of Hank's buffer containing 105 mg.% glucose. For the other, 80 mg.% unlabeled galactose also was included. Radioactivity in the external medium was determined at 2 hour intervals. It was found that radioactivity was released at an equal rate in the galactose-containing medium as compared to the unsupplemented one. It was decided that galactokinase does not play an important part in the disposal of accumulated intracellular galactose-l-phosphate. 177 The possibility was tested that a phosphatase in erythrocytes may cause hydrolysis of galactose-l- phosphate. The presence of such an enzyme was demon strated in hemolysates. Labeled galactose-l-phosphate (0.0001 M) was incubated with crude hemolysate and Mg-H- ions (0.01 M). At the end of a 3-hour incubation period, labeled galactose was separated from galactose- l-phosphate by the use of DEAE cellulose paper (126). A substantial amount of radioactivity corresponding to galactose was found. If projection from the experi mental conditions mentioned to the intracellular state could be considered valid, activity of the phosphatase would be sufficient to account for the galactose released to the external medium in the 24-hour incubations. The present work on disposition of accumulated erythrocyte galactose-l-phosphate has been exploratory in nature. The in vivo findings were not duplicated completely in vitro, and it is evident that processes other than simple enzymatic hydrolysis also are involved. The findings described provide a starting point for a more thorough study of the problem. 178 e. Chromosome 21 and Galactose-1-Phosphate Uridyl Transferase: Reports of alterations in enzyme activity in Down's syndrom (mongolism) patients by a number of investigators have suggested possible relationships to chromosome 21. Brandt (145, 146) has stated that erythrocyte levels of galactose-l-phosphate uridyl transferase in galactosemia heterozygotes, normal individuals and Down's syndrome patients stand in a 1: 2: 3 relationship to each other. On this basis he has suggested that the gene for this particular enzyme may be located in chromosome 21. In view of interest of workers at Childrens Hospital in galactosemia, it seemed appropriate to attempt to confirm Brandt's findings. Measurement of transferase levels by the radioactive galactose-l-phosphate method was contrib uted to this study (Table VI). Blood samples were studied from 11 galactosemia heterozygotes, 46 normal individuals and 42 patients with Down's syndrome. Erythrocyte galactose-l- phosphate uridyl transferase values were measured by the UDPG consumption method on all samples and by the radioactive galactose-l-phosphate procedure in many. 179 Both methods of measurement gave congruent results. It was found that the heterozygotes were clearly dis tinguishable from normal individuals on the basis of these measurements. As expected, the galactosemia heterozygotes were found to have values approximating half of those for normal controls (53) . According to Brandt's hypothesis, levels in Down's syndrome should be elevated above normal in the same proportion. The mean of erythrocyte transferase values on Down's syndrom patients was significantly higher than that for normal controls, but there was marked overlap between the two groups. The mean values were 0.98 units for galactosemia heterozygotes, 1.89 units for normals, and 1.98 units for Down's syndrome patients. The present results do not bear out the proposal of Brandt of a 1: 2: 3 gene dosage hypothesis, but at the same time they do not exclude the possibility of some other relationship of the galactose-l-phosphate uridyl trans ferase gene to chromosome 21. Possible modifying influences on gene expression must be considered. A number of samples from each group also were incubated with galactose-l-C^, and the evolution of C^-^02 was measured. The values found showed 180 relationships among the three groups similar to those in the transferase assays, although the separation between the galactosemia heterozygote and normal groups was not as marked. Here again, the average value for Down’s syndrome patients was higher than for normals, but not sufficiently increased to be consistent with the chromosome 21 hypothesis. DISCUSSION In vitro oxidation of galactose-l-C^ by blood cells from seventeen galactosemic subjects was examined. Three of these individuals produced much more labeled carbon dioxide than the remaining fourteen. This finding is similar to those of Weinberg (61) and Isselbacher (101) , and is consistent with the existence of biochemical hetero geneity in galactosemia as demonstrated Iai vivo by Segal and associates (60). It indicates that blood cells from some galactosemic individuals are capable of metabolizing galactose to carbon dioxide despite apparent absence of galactose-l-phosphate uridyl transferase activity. The results also are suggestive of familial influence. The five sibling pairs studied have correspond ing values in carbon dioxide production. Segal (150) has observed that members of one sibling pair oxidized galactose in equal amount in vivo. The present results on metabolic variation in vitro seem to be independent of age. This finding is in con trast to a generally-held impression, based on limited 181 182 results of Isselbacher on the UDPGal pyrophosphorylase (63), that all galactosemics should develop some tolerance to galactose with increasing age. It is of interest in this connection that direct studies on patients by Donnell do not support the increased tolerance hypothesis (62). The present work suggests that the variability observed is more likely to have a genetic basis than to reflect an acquired capability. To further explore differences in galactose meta bolism among galactosemic subjects, galactose intermediates from the blood incubations were separated by Dowex-l-formate column chromatography by a procedure adapted from Bartlett (136). This work represented use of labeled galactose and of a small quantity of blood cells for the first time in experiments of this type. The anticipated large amount of galactose-l-phosphate was found in all galactosemic patients, but labeled UDPHexose (both UDPGal and UDPG) also was obtained consistently. With the three unusual galactosemics, those whose blood cells produced substantial amounts of labeled carbon dioxide from galactose-l-C^, the UDPHexose peaks were markedly larger and an additional small peak in the 2,3-diphosphoglyceric acid area was present. These observations add further support to the 183 concept that metabolic variability is present in galactosemia. The fact that two of the unusual galacto semics are siblings strengthens the view of a genetic basis for biochemical variability in galactosemia. Three possibilities were considered for the forma tion of labeled UDPHexose and labeled carbon dioxide from galactose-l-C^ in the apparent absence of transferase: (1) By the established path via galactose-l-phosphate uridyl transferase, (2) Through an alternate path mediated by UDPGal pyrophosphorylase, (3) By some mechanism as yet unrecognized. The amounts of UDPHexose formed upon incubating galactose in the presence of galactosemia blood cells are very small. If UDPHexose were to arise as the result of a very low transferase activity, it is possible that the activity might not be detectable by the UDPG consumption assay. The hope of increasing the sensitivity of measure ment, an assay measuring the transfer of carbon-labeled galactose-l-phosphate to UDPGal was developed. However, even with this procedure, no transferase activity could be demonstrated in hemolysates from any of the galactosemics examined, including 2 of the 3 unusual subjects. While the labeled galactose-l-phosphate assay did not provide 184 increased sensitivity, a useful routine method was produced. It has a number of advantages over the UDPG consumption procedure. It is more direct and less demand ing in detail. It also has the capability of detecting heterozygotes in galactosemia, an important point for family studies. The inability to demonstrate transferase activity by direct assay did not exclude the possibility of the presence of such activity in low degree. However, it was necessary to examine other routes by which UDPHexose might be formed from Gal-l-P. A pathway by way of galactose-6-phosphate could not result in the formation of UDPHexose, but it could account for production of carbon dioxide from galactose. Hsia and his co-workers have reported that galactose-l- phosphate can be converted to galactose-6-phosphate by galactosemia erythrocytes (64). They have stated that galactose-6-phosphate was found in erythrocytes from five of six galactosemic subjects, the ratio of galactose-6- phosphate to galactose-l-phosphate being estimated at 1:3 (142). They have shown that galactose-6-phosphate can be oxidized in vitro by glucose-6-phosphate dehydrogenase isolated from human erythrocytes (143). Hypothetically, 185 further oxidation of the product (6-phosphogalactonolac- tone) in the cell then would yield carbon dioxide. In the present experiments, both with intact erythrocytes and with hemolysates, no evidence for the presence of labeled galactose-6-phosphate could be found. No other investigators have reported confirmation of the Hsia findings. A secondary pathway that could account for the formation of UDPHexose from galactose is that mediated by UDPGal pyrophosphorylase, an enzyme demonstrated by Isselbacher to be present in liver. Isselbacher has stated that UDPGal pyrophosphorylase does not occur in erythrocytes (100), but it was considered essential to re-examine the possibility in the present connection. In the experiments done, there was no transfer of carbon label from galactose-l-phosphate to UDPGal, in the presence of UTP, with any of the galactosemic hemolysates studied, including those from 2 of the 3 unusual subjects. Addi tion of UTP did increase incorporation of labeled galactose-l-phosphate into UDPGal in the presence of both normal and galactosemia heterozygote hemolysates. As pointed out by Isselbacher and by Strominger (43, 9), in an incubation system which also contains UDPG 186 pyrophosphorylase and glucose-l-phosphate, erroneous con clusions concerning UDPGal pyrophosphorylase activity can be reached. This view assumes that sufficient galactose-l- phosphate transferase activity is present to provide for formation of UDPGal from UDPG and Gal-l-P. In view of the stimulation of apparent UDPGal pyrophosphorylase activity obtained in the present experi ments by the addition of glucose-l-phosphate, it was concluded that the results obtained with normal and with galactosemia heterozygote hemolysates reflected the presence of transferase activity and that they did not represent the effects of a separate UDPGal pyrophosphorylase activity in erythrocytes. A concentration of native glucose-l-phosphate in intact erythrocytes of 5 x 10"^ M, as reported by Gourley (151), would be sufficient to provide enough UDPG formation to account for the apparent activity observed. UDPGal pyrophosphorylase could not be demonstrated either in normal or in galactosemia hemolysates. In none of the experiments could evidence be found for the exist ence of any other alternate route from Gal-l-P to UDPHexose. The conclusion was reached that the most probable explanation for the formation of UDPHexose in the present experiments with galactosemia blood cells is the 187 presence of a minimal transferase activity. That this activity is very low is evident from the inability to demonstrate it in the conventional UDPG assay or in the labeled Gal-l-P procedure devised in the course of this work. It is considered, also, that the activity is low enough not to be manifested in the relatively short incuba tion times employed in the UDPGal pyrophosphorylase assay. The 90 minute incubations employed in the experiments on galactose intermediates allowed a detectable amount of UDPHexose to accumulate. In this connection, it is inter esting to recall that Schwarz et_ ad., using a micro- manometric technique, found erythrocyte transferase activity averaging 7% of normal values in 13 of 23 galactosemic children (56). The presence of a minimal galactose-l-phosphate uridyl transferase activity in galactosemia erythrocytes also would account for the labeled carbon dioxide production found. This requires a supply of UDPG, as of course is true in normal tissue, and consequently it would be expected that factors concerned in UDPG formation might affect galactose metabolism in the galactosemia patient. Related to this viewpoint is the finding of Tada and co workers (152, 153) that oral administration of orotic acid 188 to galactosemic patients resulted in a remarkable decrease of urinary galactose and in a change in galactose tolerance. Addition of uridine or of orotic acid, also decreased the in vitro accumulation of hexose phosphate in galactosemia erythrocytes under anaerobic incubation with galactose. Prior to the present work, there has been no detailed study of intermediates of galactose metabolism in erythrocytes from galactosemia heterozygotes. The results are in accord with the view that the Leloir pathway through UDPHexose, and finally to the glucose pathway at glucose-l- phosphate, is the major route for the utilization of galactose. However, it was found that erythrocytes from some heterozygotes did not produce lactic acid from galactose in 90 minutes of incubation, as was the case with normal erythrocytes. This phenomenon was shown to be due to a decreased rate of formation rather than a second genetic defect in galactosemia. It is known that galacto semia heterozygotes have approximately half the normal quantity of erythrocyte transferase, and this limitation might be sufficient to delay the appearance of lactic acid. Experiments in which the incubation time was increased did result in the finding of lactic acid. The time course study of the glycolytic 189 intermediates of galactose, as shown in Figure 21, provided interesting comparisons. The chromatographic patterns clearly showed the appearance of intermediates in sequence. Time course of the pathway proved to be a slower process than had been expected. As much as 60 minutes were required for erythrocytes from some individuals to show a detectable amount of labeled lactic acid. A more extensive study is in progress. Of interest, is the finding of Bartlett et_ a_l. (154, 155) that only half of the lactic acid could be accounted for from the labeled glucose disappearing during four hours of incubation; the rest of lactic acid came from intracellular reserve intermediates. It was noted that the radioactivity in UDPHexose stayed practically constant at incubation times of from 15 to 60 minutes, while galactose-l-phosphate continued to build up within this period. This observation may bear upon the in vivo finding of Donnell, et_ al_. that, despite the presence of transferase in normal individuals, there often is a detectable amount of erythrocyte galactose-l- phosphate in an oral galactose load test at the 1/2 or 1 hour points. The results of another group in this laboratory on C^02 production by erythrocytes from galactose-l-C^ did 190 not show as marked a separation of galactosemia hetero zygotes from normal homozygotes as might be expected on the basis of relative transferase values. It was considered that some other steps might be limiting in the production 14 of C O2, and perhaps also of lactic acid. When UDPHexose was isolated from normal bloods incubated with galactose-1-C^, about 80% of the radio activity was found in UDPGal and 20%, in UDPG. On the contrary, in UDPHexose isolated from galactosemia bloods incubated under the same conditions, 20-30% of the radio activity was in UDPGal, with 70-80% in UDPG. In studies by others at the equilibrium state in an isolated system, 257o UDPGal and 75% UDPG were obtained (11) . The present finding suggests that in erythrocytes the transferase reaction forms UDPGal at a greater rate than UDPGal is transformed to UDPG by the epimerase step. In this circum stance the epimerase is rate-limiting. This concept is in accord with the results of Isselbacher e_t a_l. (36) and of Segal et al. (156) in hemolysate studies. In galactosemia erythrocytes, with the minimal transferase activity postulated to be present, the transferase becomes rate- limiting, and the epimerase reaction is practically at the equilibrium state. 191 The liver studies were too limited to contribute information on adaptation. Although UTP increased the incorporation of galactose-l-phosphate into UDPGal in liver samples from two patients with glycogen storage disease, the existence of a separate UDPGal pyrophosphory lase could not be fully substantiated. It is possible that the incubation medium might have contained a source of glucose-l-phosphate, particularly in view of the glycogen present. For one of the other individuals studied, the liver transferase activity was very low. No clinical symptoms of galactosemia had been manifested prior to death, but this may have been due to avoidance of galactose-containing foods. Such cases have been known in clinical practice. The liver is considered to be the primary organ for the metabolism of galactose, and the importance of continuing studies on liver tissue is evident. While biopsy samples from galactosemia patients are not justified in general, opportunities for obtaining liver samples at appendectomy or at surgery for other abdominal conditions should be anticipated. Also, with suitable consent, it is possible to obtain biopsy specimens from normal individuals under such circumstances. 192 As part of the present studies, galactokinase activity was found to be elevated in hemolysates from newborn infants, to levels as much as three times those found for older children and for adults. A gradual decrease was noted within the first year of life, and beyond this age the activity continued at approximately the same level. A parallel study of galactose-l-phosphate uridyl transferase and of the epimerase reaction is in progress. On the basis of finding that the rate of oxida tion of galactose is much higher in newborn rat liver than in adult rat liver, Segal and his associates (148) have suggested that galactokinase may be the rate-determining step. The problem of the disposal of galactose-l- phosphate in galactosemia red cells accumulated during galactose loading is an intriguing one. In the course of the present work it was shown that the process is carried on in part by an intracellular phosphatase. The in vivo findings were not completely duplicated in vitro, and pertinent questions remain to be answered, in particular concerning the problem of transport of galactose, and possibly of its derivatives. Since chromosomal analysis shows that most Down's 193 syndrome (mongolism) patients are trisomic for chromosome 21, there has been interest in attempting to correlate enzymes involved in genetic disorders with mongolism. Of particular interest is the claim of Brandt that erythrocyte galactose-l-phosphate uridyl transferase values in galactosemia heterozygotes, normal individuals, and Down's syndrome patients stand in a 1:2:3 relationship to each other (145, 146). He suggested that the structural gene for transferase is located in chromosome 21, the increased value in Down's syndrome being due to the extra chromosome present. In the present study it was found that the mean erythrocyte transferase activity in Down's syndrome patients is significantly higher, but not sufficiently to support the hypothesis of a 1:2:3 gene dosage relationship. Of interest is the finding, by other members of the group involved in the Down's syndrome study, of elevated erythrocyte glucose-6-phosphate dehydrogenase activity in mongolism. Since this enzyme is considered to be sex- linked, further doubt is thrown on Brandt's hypothesis. It is possible that chromosome 21 may contain genes providing for regulation of a variety of metabolic processes. Undoubtedly, this question will be the subject of much study by many investigators. CONCLUSIONS In galactosemia, a disease known to be genetic in origin, there is a defect in the enzyme galactose-l- phosphate uridyl transferase. The present study demon strated that biochemical heterogeneity exists among individuals affected with this disease. Blood cells from three of seventeen homozygotes produced substantially greater amounts of labeled carbon dioxide upon in vitro incubation with galactose-l-C^ than did erythrocytes from the remaining fourteen. In a study of galactose intermediates in red cells, galactose-l-phosphate was found to accumulate as antici pated, but uridine diphosphate hexose (UDPHexose) unexpectedly was found to be present in small amounts for all of the galactosemics under examination. With each of the three unusual individuals, there was an elevation of UDPHexose, and a small amount of labeled 2,3-diphospho- glyceric acid also was found. Five sibling pairs were shown to have corresponding values for C^^O^ production and similar patterns of labeled intermediates, suggesting a familial influence in the biochemical heterogeneity. 19 4 The presence of carbon label in the UDPHexose and in 2,3-diphosphoglyceric acid indicated that these com pounds were formed from galactose. Either galactose-l- phosphate uridyl transferase activity may not be absent completely in galactosemia or some other pathway for the conversion of galactose to UDPHexose may be present. A logical possibility would be one involving the enzyme UDPGal pyrophosphorylase, but no evidence for its presence in erythrocytes could be found. The experimental work done provided no evidence for the existence of any other pathway to explain the conversion of galactose-l-phosphate into UDPHexose. The conclusion was reached that a minimal trans ferase activity is present in all galactosemic individuals, at least in erythrocytes, and that some of these individu als have more activity than others. Whether the minimal activity reflects a quantitative difference or a different enzyme could not be determined. Evidence from family studies indicated that the minimal ability of galactosemic subjects to metabolize galactose is inherited rather than acquired. In the course of the investigation, a number of supplementary studies were made, as follows: 196 1. A routine method was developed for the assay of galactose-l-phosphate uridyl transferase, using galactose-l-C^-l-phosphate. 2. The rate of conversion of galactose to lactic acid was found to be relatively slow in erythrocytes, even for normal individuals, compared with glucose. 3. Galactokinase activity was found to be con siderably elevated in hemolysates from newborn infants as compared to those of older children and adults. 4. The in vitro disposal of accumulated galactose- l-phosphate in galactosemic erythrocytes was found to be carried out in part by an intracellular phosphatase. 5. The mean galactose-l-phosphate uridyl trans ferase activity in patients with Down's syndrom (of the type trisomic in chromosome 21) was found to be signifi cantly higher than in normal individuals, but not sufficiently so to support a hypothesis of a gene dosage relationship and consequent localization of the gene for transferase in chromosome 21. APPENDIXES APPENDIX A REAGENTS Adenosine-5 1-triphosphate, Na (Sigma Chemical Company, St. Louis) Bentonite (Braun Corporation, Los Angeles) Charcoal, activated (Merck powder; 30-50 mesh, Barnebey- Cheney) Cysteine Hydrochloride (Sigma) Diethylamino ethaiiol (DEAE) cellulose (California Corpora tion for Biochemical Research) (Calbiochem) 2,3-Diphosphoglycerate, Barium Salt (Sigma) B-Diphosphopyridine Nucleotide, oxidized (Sigma) B-Diphosphopyridine Nucleotide, reduced (Sigma) Dowex-l-chloride, 200-400 mesh (Calbiochem) Dowex-50W, 200-400 mesh (Baker Chemical Company, Phillipsburg, New Jersey) Fructose-1,6-diphosphate, Barium Salt (Calbiochem) Galactose, purified grade (Sigma) Galactose, N. F. (Pfanstiehl Chemical Company, Waukegan, Illinois) Galactose-l-C^ (Calbiochem, Los Angeles; Volk, Chicago; National Bureau of Standards, Washington, D.C.; New England Nuclear, Boston) Galactose-l-phosphate, Barium Salt (Sigma) 198 199 Galactose-6-phosphate, Barium Salt, practical (Sigma) Glucose, C. P. (Bjaker) Glucose-l-C^ (Calbiochem) Glucose-l-phosphate, Potassium Salt (Calbiochem) Glucose-l-C-^-1-phosphate (Gift of reference sample from Dr. E. Neufeld, Department of Biochemistry, University of California, Berkeley, Calif.) Glucose-6-phosphate, Potassium Salt (Sigma) Glycine, C. P. (Pfanstiehl) Lactate, Lithium (Hartman-Leddon Company, Philadelphia, Pennsylvania) DL-Lactate-l-C^, Zinc (Volk, Chicago) Phenol reagent (Folin-Ciocateau) (Western Surgical Supply Company, Los Angeles) Phosphoenolpyruvate, tricyclohexyammonium (Sigma) D-3-Phosphoglyceric acid, Barium Salt (Sigma) Pyruvate, Sodium (Sigma) 14 Pyruvate-3-C , Sodium (Calbiochem) Thymidine triphosphate (Calbiochem) Triphosphopyridine Nucleotide, oxidized (Sigma) Triphosphopyridine Nucleotide, reduced (Sigma) Uridine diphosphate glucose, Sodium Salt (Sigma) Uridine-5 1-triphosphate, Sodium Salt (Sigma) 200 The more usual reagents (e.g., hydrochloric acid, sodium hydroxide, phosphates, etc.) were obtained in suitably pure grade through local supply houses. APPENDIX B EQUIPMENT 1• Radioactive Counting: Scaler-Timer (Baird-Atomic, Model 135) Automatic Sample Changer (Baird-Atomic, Model 755) Calculator (Victor Digit-matic) Strip Scanner Ratemeter (Baird-Atomic, Model 432A) Linear Recorder (Texas Instruments, Inc.) Liquid Scintillation Spectrometer (Nuclear-Chicago; Packard Tricarb) 2. General Equipment: Centrifuge, Refrigerated (Servall) Centrifuge, Preparative Ultracentrifuge (Spinco) Chromatographic Tanks Colorineter (Beckman Model C) Constant Temperature Blocks (Roeco, Los Angeles) Evaporator, Vacuum, Rotating (Rinco) Fraction Collectors (Technicon) Gas Mixtures (Geiger gas, Matheson; 95%02 - 57.C02, Calox Company) 201 202 Homogenizers (Potter-Elvejhem; Waring Blender; Vir-Tis "23") Lyophilization Equipment (constructed) Metabolic Shakers, Dubnoff (Precision Scientific Co., American Instrument Co.) Spectrophotometer (Beckman DU) Ultrafiltration Apparatus (constructed) Ultraviolet Lamps (Ultra-Violet Products, Inc.) BI BL I OG R AP H Y BIBLIOGRAPHY 1. von Reuss, A. Wien, med. Wehnschr. 5J3, 799 (1908). 2. Carpenter, T. M. and Lee, R. C. Am. J. Physiol. 102 635 (1932). 3. Deuel, H J., MacKay, E. M., and Gulick, M. Proc. Soc. Exptl. Biol. Med., 3(), 24 (1932-33). 4. Kosterlitz, H. W. J. Physiol. 9_3, 34p (1938). 5. Kosterlitz, H. W. Biochem. J. .37., 322 (1943). 6. Cori, G. T. and Cori, C. F. Proc. Soc. Exptl. Biol. Med., 36, 119 (1937) . 7. Caputto, R., Leloir, L F., Trucco, R. E., Cardini, C. E., and Paladini, A. C. J. Biol. Chem. 179, 497 (1949). 8. Caputto, R., Leloir, L. F., Cardini, C. E. and Paladini, A. C. J. Biol. Chem. 184, 333 (1950). 9. Strominger, J. L. Physiol. Rev., 4-C), 55 (1960). 10. Leloir, L. F., The Harvey Lecture. 56., 23 (1960-61). 11. Leloir, L. F. Arch. Biochem. 3_3, 186 (1951). 12. Kalckar, H. M., Braganca, B. and Munch-Petersen, A. Nature, 172, 1038 (1953). 13. Rutter, W. J. and Hansen, R. G., J. Biol. Chem. 202 323 (1953). 14. Hansen, R. G. and Craine, E. M., J. Biol. Chem. 208 293 (1954). 15. Hansen, R. G. and Freedland, R. A., J. Biol. Chem. 216, 303 (1955) . 204 16. 17 . 18. 19. 20. 21. 2 2. 23. 24. 23 . 26 . 27 . 28. 29. 30. 205 Garner, R. L. and Grannis, G. F. Science. 114, 501 (1951). Kurahashi, K. and Anderson, E. P. Biochim. Biophys. Acta. 29, 498 (1958). Maxwell, E. S. and de Robichon-Szulmajster, H., J.Biol. Chem., 235, 308 (1960). Maxwell, E. S., de Robichon-Szulmajster, H. and Kalckar, H. M. Arch. Biochem. Biophys., 7_8, 407 (1958). Ginsburg, V., Stumpf, P. K. and Hassid, W. Z., J.Biol. Chem. 223., 977 (1956). Ginsburg, V., J. Biol. Chem., 232, 55 (1958). Kessler, G., Neufeld, E. F., Feingold, D. S. and Hassid, W. Z., J. Biol. Chem. 236, 308 (1961). Neufeld, E. F., Feingold, D. S., Kessler, G. A., lives, S. M. and Hassid, W. Z. Fed. Proc. 20, 79 (1961). Feingold, D. S., Neufeld, E. F. and Hassid, W. Z. Arch. Biochem. Biophys. 78., 401 (1958). Neufeld, E. F., Feingold, D. S. and Hassid, W. Z. J. Am. Chem. Soc. 8(3, 4430 (1958). Feingold, D. S., Neufeld, E. F. and Hassid, W. Z., J. Biol. Chem. 235, 910 (1960). Neufeld, E. F., Feingold, D. S. and Hassid, W. Z. Arch. Biochem. Biophys. 8_3, 96 (1959). Neufeld, E. F., Feingold, D. S. and Hassid, W. Z. J. Biol. Chem. 235, 906 (1960). Stetten, D., Jr., J. Biol. Chem. 159, 123 (1945). Topper, Y. J. and Stetten, D., Jr. J. Biol. Chem. 193, 149 (1951). 206 31. 32. 33. 34. 35. 36. 37 . 38. 39. 40. 41. 42. 43. 44. 45 . 46. Schwarz, V., Goldberg, L., Komrower, G. M. and Holzel, A. Biochem. J. 62^ 34 (1956). Schwarz, V. Arch. Dis. Child. 35, 428 (1960). Schwarz, V., Holzel, A. and Komrower, G. M. Lancet,I_, 24 (1958). Schwarz, V. and Goldberg, L. Biochim. Biophys. Acta. 18, 310 (1955) . Kalckar, H. M., Anderson, E. P. and Isselbacher, K. J. Biochim. Biophys. Acta. 20i, 262 (1956). Isselbacher, K. J., Anderson, E. P., Kurahashi, K. and Kalckar, H. M., Science. 123, 635 (1956). Anderson, E. P., Kalckar, H. M. and Isselbacher, K. J. Science. 125, 113 (1957) . Anderson, E. P., Kalckar, H. M., Kurahashi, K. and Isselbacher, K. J., J. Lab. Clin. Med., .50, 469 (1957). Bretthauer, R. K., Hansen, R. G., Donnell, G. N. and Bergren, W. R. Proc. Nat. Acad. Sci. 45_, 328 (1959) . Kirkman, H. N., and Bynum, E. Ann. Human Genetics. 23, 117 (1959) . Cardini, C. E. and Leloir, L. F. Arch. Biochem. 45, 55 (1953). Maxwell, E. S., J. Biol. Chem. 229, 139 (1957). Isselbacher, K. J., J. Biol. Chem. 232, 429 (1958) Isselbacher, K. J., Prog, in Liver Dis., _1, 202 (1961). Larner, J., Villar-Palasi, C. and Richman, D. J. Ann. New York Acad. Sci. 82_, 345 (1959). Hastings, A. B., Ashmore, J. and Cahill, G. E., Jr. Arch. Biochem. Biophys. 65, 78 (1956). 207 47 . 48. 49. 50. 51. 52 . 53. 54. 55 . 56. 57 . 58. 59. 60. Leloir, L. F. and Cardini, C. E., J. Am. Chem. Soc. 79, 6340 (1957). Leloir, L. F., Olavarria, J. M., Goldemberg, S. H. and Carminatti, H. Arch. Biochem. Biophys. 81, 508 (1959) . Mason, H H. and Turner, M. E. A.M.A. Am. J. Dis. Child. 5j0, 359 (1935) . Townsend, E. H., Jr., Mason, H H. and Strong, P. S. Pediatrics. 1_, 760 (1951). Holzel, A. and Komrower, G. M. Arch. Dis. Child. 30, 155 (1955). Donnell, G. N., Bergren, W. R. and Roldan, M. Pediatrics. 2_4, 418 (1959). Donnell, G. N., Bergren, W. R., Bretthauer, R. K. and Hansen, R. G. Pediatrics. 25_, 572 (1960). Donnell, G. N., Koler, R. and Smith, S. M. Ann.Human Genetics. 2_7, 125 (1963). Haia, D. Y.-Y., Tannenbaum, M., Schneider, J. A., Huang, I. and Simpson, K., J. Lab. Clin. Med. 56, 368 (1960). Schwarz, V., Wells, A. R., Holzel, A. and Komrower, G. M. Ann. Human Genetics. 25_, 179 (1961). Gitzelmann, R. and Hadorn, b. Helv. Paediat. Acta. 16, 1 (1961). Walker, F. A., Hsia, D. Y.-Y., Slatis, H. M. and Steinberg, A. G. Ann. Human Genetics. 2_5, 287 (1962). Eisenberg, F., Jr., Isselbacher, K. J., and Kalckar, H. M. Science. 126, 116 (1957). Segal, S., Blair, A., and Topper, Y. J. Science. 136, 150 (1962). 61. 62. 63 . 64. 65 . 66. 67 . 68. 69. 70. 71. 72 . 73. 74. 75. 208 Weinberg, A. N. Metabolism. ljO, 728 (1961). Donnell, G. N. , Bergren, W. R. Perry, G. and Koch, R. Pediatrics. 3JL, 802 (1963). Isselbacher, K. J. Science. 126, 652 (1957). Inouye, T. and Hsia, D. Y.-Y., J. Lab. Clin. Med. 58, 829 (1961) . Galloway, R. A. and Krauss, R. W. Plant Physiol. 34, 380 (1959) . Posternak, Th. and Rosselet, J. P. Helv. Chim. Acta. 27, 246 (1954). Rogers, D. and Yu, S.-H. J. Bacteriol. 84, 877 (1962). Shiota, T., Blumenthal, H., Disraely, M. N. and McCann, M. P. Arch. Biochem. Biophys. 96_, 143 (1962). Egyud, L. G. and Whelen, W. J. The 423rd Meeting of Biochem. Soc., Dec., 1962. Elder, T. D., Segal, S., Maxwell, E. S. and Topper, Y. J. Science. 132^, 225 (1960). Pesch, L. A., Segal, S. and Topper, Y. J., J. Clin. Invest. 39_, 178 (1960). Pesch, L. A. and Topper, Y. J. Biochim. Biophys. Acta. 30, 206 (1958). Simon, E. R., Pesch, L. A. and Topper, Y. J. Biochem. Biophys. Res. Comm. 1_, 6 (1959). Segal, S. and Blair, A. J., J. Clin. Invest. 40, 2016 (1961). Neufeld, E. F., Biochem. Biophys. Res. Comm. 7_, 461 (1962). 76. 77 . 78. 79 . 80. 81. 82. 83. 84. 85. 86. 87 . 88. 89. 90. 91. 209 Pazur, J. H , Kleppe, K. and Cepure, A. Biochem. Biophys. Res. Comm. 7j_ 157 (1962). van Heyningen, R. Biochem. J. 7_3, 197 (1959). van Heyningen, R. Biochem. J. j69, 481 (1958). van Heyningen, R. Nature. 184, 194 (1959). Kinoshita, J. H., Merola, L. 0. and Dikmak, E. Biochim. Biophys. Acta. 62_, 176 (1962). Kinoshita, J. H., Merola, L. 0., Saton, E. and Dikmak, E. Nature. 194, 1085 (1962). Cooper, J. A. D., Smith, W., Bacila, M. and Medina, H. J. Biol. Chem. 234, 445 (1959). Avigad, G., Amaral, D., Asensio, C. and Horecker, B. L. J. Biol. Chem. 237, 2736 (1962). Noble, M. K. and Madhosingh, C. Biochem. Biophys. Res. Comm. 12_, 146 (1963). DeLey, J. and Doudoroff, M., J. Biol. Chem., 227, 745 (1957). Sidbury, J. B., Jr., Abs. of 132nd Meeting of Am. Chem. Soc. 27c, Sept., 1957. Ginsburg, V. and Neufeld, E. F. Abs. of 132nd Meeting of Am. Chem. Soc., 27c, Sept., 1957. Egyud, L. G. and Whelen, W. J. The 423rd Meeting of Biochem. Soc. Dec., 1962. Morse, M. L., Lederberg, E. M. and Lederberg, J. Genetics. 4T, 142 (1956). Kurahashi, K. Science, 125, 114 (1957). Fukasawa, T. and Nikaido, H. Nature. 184, 1168 (1959). 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 210 Nikaido, H Biochim. Biophys. Acta. 48, 460 (1961). Soffer, R. L., J. Bacteriol. 82, 471 (1961). Spyrides, G. and Kalckar, H. M. Biochem. Biophys. Res. Comm. _3, 306 (1960). Fukasawa, T., Jokura, K. and Kurahashi, K. Biochim. Biophys. Res. Comm. 7_, 121 (1962). Buttin, G. Symp. on Quant. Biol. 26, 213 (1961). Yarmolinsky, M. B., Jordan, E. and Wiesmeyer, H. Symp. on Quant. Biol. 26_, 217 (1961). Kalckar, H. M. and Sundararagan, T. A. Symp. on Quant. Biol. 2j6, 227 (1961). Donnell, G. N. and Lann, S. H. Pediatrics. ]_, 503 (1951). Isselbacher, K. J. in The Metabolic Basis of Inherited Disease. (J. B. Stanbury, J. B. Wyngaarden and D. S. Fredrickson, eds.) New York: McGraw-Hill, 1960. Isselbacher, K. J. Personal Communication. Bergren, W. R. and Donnell, G. N. Ann. Human Genetics. 2_6, 135 (1962). Topper, Y. J., Laster, L. and Segal, S. Nature, 196, 1006 (1962) . Horecker, B. L., J. Biol. Chem., 183, 593 (1950). Strominger, J. L., Maxwell, E. S., Axelrod, J. and Kalckar, H. M. J. Biol. Chem. 224, 79 (1957). Leloir, L. F. and Trucco, R. E. Methods in Enzymology. 1_, 290 (1955) (S. P. Colowick and N. 0. Kaplan, eds.) New York: Academic Press, Inc. 211 107 . 108. 109. 110. Ill. 112 . 113. 114. 115 . 116. 117 . 118. 119. 120. 121. 122. Heinrich, MR., J. Biol. Chem. 239, 50 (1964). Wilkinson, J. F. Biochem. J., 44, 460 (1949). Keilen, D. and Hartree, E. F. Proc. Royal Soc. B 124, 397 (1938). Anderson, E. P., Maxwell, E. S. and Burton, R. M. J. Am. Chem. Soc., 8_1, 6514 (1959). Dische, Z., Shettles, L. B. and Osnos, M. Arch. Biochem. , 22_, 169 (1959) . Diedrich, D. F. and Anderson, L. Anal. Biochem. 22, 189 (1959). Kirkman, H. N. and Maxwell, E. S., J. Lab. Clin. Med. 56, 161 (1960). Nelson, N., J. Biol. Chem. 153, 375 (1944). Park, J. T. and Johnson, M. J., J. Biol. Chem. 181, 149 (1949). Glucostat Bulletin. Worthington Biochemical Corp., Freehold, New Jersey, 1958. Taussky, H. H. and Shorr, E., J. Biol. Chem. 202, 675 (1953). Barker, S, B. and Summerson, W. H.,J. Biol. Chem. 138, 535 (1941). Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., J. Biol. Chem., 193, 265 (1951). Neilands, J. B. Methods in Enzymology. 1_, 449 (1955). (S. P. Colowick and N. 0. Kaplan, eds.) New York: Academic Press, Inc. Donnell, G. N., Bergren, W. R. and Perry G. Personal communication. Crane, R. K. and Lipmann, F., J. Biol. Chem. 201, 235 (1953). 212 123. 124. 125 . 126. 127. 128. 129. 130. 131. 132. 133. 134. 135 . 136. 137. 138. Hennessey, M. A., Waltersdorph, A. M., Huennekens, F. M. and Gabrio, B. W., J. Clin. Invest. 41, 1257 (1962). Zinkham, W. H., Johns Hopkins Hosp. Bull. 102, 169 (1959). Claude, A., J. Exptl. Med. 84, 61 (1946). Sherman, J. R. and Alder, J., J. Biol. Chem., 238, 873 (1963). Bandurski, R. S. and Axelrod, B., J. Biol. Chem. 193, 405 (1951). Paladini, A. C. and Leloir, L. F., Biochem. J., 51, 426 (1952). Chargaff, E., Levine, C. and Green, C., J. Biol. Chem. , 175., 67 (1948) . Inouye, T., Tannenbaum, M. and Hsia, D. Y.-Y. Nature, 193, 67 (1962). Cohen, S. S. and Scott, D. B. Science. Ill, 543 (1950). Hanes, C. S. and Isherwood, F. A. Nature. 164, 1107 (1949). Partridge, S. M. Nature. 164, 443 (1949). Trevelyan, W. E., Procter, D. P. and Harrison, J S. Nature. 166, 144 (1950). Smith, I. Chromatographic and Electrophoretic Techniques. Vol. .1, 279; New York: Interscience Pub1i sher s, Inc. 1960. Bartlett, G. R., J. Biol. Chem. 234, 459 (1959). Bartlett, G. R., J. Biol. Chem. 234, 449 (1959). Annison, E. F., Lindsay, D. B. and White, R. R. Biochem. J. 88_, 243 (1963). 139. 140. 141. 142. 143. 144. 145 . 146. 147 . 148. 149. 150. 151. 152 . 153. 154. 213 Gourley, D. R. H. Arch. Biochem. Biophys. 40, 13 (1952). Umbreit, W. W., Burris, R. H. and Stauffer, J. F. Manometric Techniques; Minneapolis, Minn.: Burgess Publishing Co., p. 270 (1957). Hsia, D. Y.-Y., Inouye, T. and Walker, F. A., J. Am. Med. Assoc. 178, 944 (1961). Hsia, D. Y.-Y. Personal communication. Schneider, J. A., Inouye, T. and Hsia, D. Y.-Y. Clin. Res. 1_1, 227 (1963). Bonsignore, A., Fornaini, G., Leoncini, G., Segni,P. and Fantoni, A. It. J. Biochem. L2, 1 (1963). Brandt, N. J. Lancet, 11, 837 (1962). Brandt, N. J., Froland, A., Mikkelsen, M., Nielsen,A. and Tolstrup, N. Lancet, yi, 700 (1963). Oliver, I. T. Biochim. Biophys. Acta. 5_2, 75 (1961). Segal, S., Roth, H. and Bertoli, D. Science. 142, 1311 (1963). Atkinson, M. R. Burton, R. M. and Morton, R. K. Biochem. J. TQ, 813 (1961). Segal, S. Personal Communication. Gourley, D. R. H. Arch. Biochem. Biophys. 40, 1 (1952) Tada, K., Tohoku J. Exp. Med. 77_, 340 (1962) Tada, K., Akabane, J. and Yokoyama, M. Tohoku J. Exp. Med. 77, 400 (1962). Bartlett, G. R. and Marlow, A. A., J. Lab. Clin. Med. 42, 188 (1953) . 214 155. Bartlett, G. R. and Marlow, A. A., J. Lab. Clin. Med. 42, 178 (1953). 156. Segal, S., Elder, T. D. and Topper, Y. J., J. Clin. Invest. 40, 1081 (1961). This d isse rta tio n has been 64— 12,481 m ic ro film ed exactly .as receiv ed NG, Won Gin, 1934- GALACTOSE METABOLISM IN HUMAN BLOOD CELLS. U n iversity of Southern C alifornia Ph.D ., 1964 C h e m istry , biological University Microfilms, Inc., Ann Arbor, Michigan

Asset Metadata

Creator Ng, Won Gin, 1934- (author)

Core Title Galactose metabolism in human blood cells

Contributor Digitized by ProQuest (provenance)

School Graduate School

Degree Doctor of Philosophy

Degree Program Biochemistry

Degree Conferral Date 1964-06

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, biochemistry,OAI-PMH Harvest

Language English

Advisor Bergren, William R. (committee chair), [illegible] (committee member), Mehl, John W. (committee member), Saltman, Paul (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-339968

Unique identifier UC11359105

Identifier 6412481.pdf (filename),usctheses-c18-339968 (legacy record id)

Legacy Identifier 6412481.pdf

Dmrecord 339968

Document Type Dissertation

Rights Ng, Won Gin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA