University of Southern California Dissertations and Theses

Studies of phosphorylated metabolic intermediates

(USC Thesis Other)

Studies of phosphorylated metabolic intermediates

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content STUDIES OF PHOSPHORYLATED METABOLIC INTERMEDIATES by Robert Fred Brunst 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 1974 U N IV E R S IT Y O F S O U T H E R N C A LIF O R N IA T H E G R A D U A T E S C H O O L U N IV E R S IT Y PARK LO S A N G E L E S , C A L I F O R N IA 9 0 0 0 7 This dissertation, written by ................R o b e rt „F ..„.B ru n st................................. under the direction of h±s... Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillm ent of requirements of the degree of D O C T O R O F P H IL O S O P H Y Dean DISSERTATION CO M M ITTEE 'hairmin This dissertation is dedicated to my wife, Jaleh, whose patience and continuous encouragement greatly facilitated the present study. I am deeply grateful to Professor William R. Bergren for stimulation and invaluable guidance throughout the course of this study. I am also indebted to Dr. George Donnell, who provided the clinical material for this study and made many helpful suggestions. Dr. Won Ng, with his vast experience in galactosemia and friendly interest and cooperation, made this study a practical reality. I am thankful to Dr. Arvin Fluharty and to other members of the Faculty of Biochemistry for their consideration and helpfulness during my graduate years. For the excellent technical assistance and advice in the use of radioisotopes I am indebted to Max Fields, for helpful suggestions in column chromatography to Elmo Jacobs, and for help in preparation of this manuscript to Wilma Bondeson. I am grateful to the Research Program of Childrens Hospital of Los Angeles for support. The Biochemistry Research Laboratory provided the facilities that were utilized for this study. Improvement in methodology for separation and analysis of compounds in body tissues and fluids has been an important factor in advancing knowledge. The present study has involved the development of an automated anion exchange system for the separation and quantitation of phosphorylated metabolic intermediates. The capability to analyze for radioactivity was also included. The completed system can perform an analysis for phosphory lated intermediates on a very small amount of sample in less than 3 hours. The system was applied to two problems of current biochemical interest. One study demonstrated that the mechanism responsible for the decreased viability of erythrocytes metabolizing galactose as the only nutrient involves a rate limitation in the Leloir pathway. There is a steady decline in intracellular ATP and 2,3-diphos- phoglyceric acid. When glucose also is present, galactose continues to be metabolized, but the decline in ATP does not occur. This is contrary to a long-held concept that galactose-l-phosphate accumulation leads to sequestration of ATP from metabolic use. A second study has focused on nucleotide metabolism in platelets. Contrary to earlier views, a functioning salvage pathway for hypoxanthine was shown to exist in these cells. iii C O N T E N T S PAGE INTRODUCTION ..... 1 Enzymatic methods .......................... 3 Electrophoretic procedures.................. 5 Chromatographic separation.................. 7 THE PROBLEM AND THE PLAN OF ATTACK................19 AUTOMATED SYSTEM FOR ANALYSIS OF PHOSPHORYLATED INTERMEDIATES..................25 The phosphorus analyzer as a system .... 26 Comments on development .................... 39 GALACTOSE METABOLISM IN THE ERYTHROCYTE........... 74 THE SALVAGE PATHWAY FOR HYPOXANTHINE IN THE PLATELET.............................. 124 DISCUSSION......................................... 145 BIBLIOGRAPHY ..................................... 161 iv LIST OF ABBREVIATIONS ADP: AMP-S: APRT: ATP: 1.3-DPG: 2.3-DPG: Epimerase: F-l,6-diP: Gal: Gal-l-P: Gal-6-P: Gal-1,6-diP: Galactokinase: GDP: Glu: G-l-P: G-6-P: G-l,6-diP: GMP: GTP: Adenosine diphosphate Adenylosuccinic acid Adenine phosphoribosyl transferase (E.C. 2.4.2.7) Adenosine triphosphate 1.3-diphosphoglyceric acid 2.3-diphosphoglyceric acid UDPGlucose-4-epimerase (E.C. 5.1.3.2) Fructose-1,6-diphosphate Galactose Galactose-1-phosphate Galactose-6-phosphate Galactose-1,6-diphosphate ATP: Galactose 1-phosphotransferase (E.C. 2.7.1.6) Guanosine diphosphate Glucose Glucose-l-phosphate Glucose-6-phosphate Glucose-1,6-diphosphate Guanosine monophosphate Guanosine triphosphate v HGPRT: Hypoxanthine guanine phosphoribosyl transferase (E.C. 2.4.2,. 8) IDP: Inosine diphosphate IMP: Inosine monophosphate ITP: Inosine triphosphate NAD: Nicotinamide adenine dinucleotide NADH: Nicotinamide adenine dinucleotide (reduced) PGM: Phosphoglucomutase (E.C. 2.7.5.1) PK: Pyruvate kinase (E.C. 2.7.1.40) Pi: Inorganic phosphate Transferase: UDPGlucose: <* -D-galactose-l-phosphate uridyltransferase (E.C. 2.7.7.12) UDPG: Uridine diphosphate glucose UDPGA: Uridine diphosphate glucuronic acid UDPGal: Uridine diphosphate galactose UDPH: Uridine diphosphate hexose UTP: Uridine triphosphate XMP: Xanthosine monophosphate INTRODUCTION An important factor in arriving at our present knowledge of metabolic pathways has been the capability of isolating and identifying intermediary compounds of metabolism. As compounds are identified, it becomes possible to demonstrate their participation in enzymatic reactions and to postulate metabolic pathways based on the logical flow of intermediates. The advent of isotope tracer techniques has provided a powerful means of providing further insight and of gaining knowledge on the control and regulation of the reactions involved. To obtain physiologically meaningful data, it is essential that intact cell systems be employed for metabolic studies in order that the effect of compart ments and different metabolic pools on regulations be evaluated. Working with disrupted cells and homogenates has provided much of our background information on metabolic processes, but these systems do not necessarily reflect intracellular kinetics. In investigations upon intact cellular systems, particularly in studies involving human biochemistry, very often only small amounts of tissue are available. Furthermore, care must be exercised to avoid changes in the status of the various metabolites during extraction 1 and subsequent isolation of these compounds. Many of the earlier approaches to the identification and quantitation of metabolic intermediates are no longer adequate. The principal methodological approaches which are commonly employed to meet the present demands for sensi tivity and specificity in the analysis of metabolic intermediates can be classified into three broad categories: (1) enzymatic determination of these com pounds, (2) electrophoretic procedures and, (3) chromato graphic methods. All of the methods in these categories have been successfully applied to metabolic problems, and each approach has both merits and shortcomings. The purpose of the present investigation has been to develop an automated system for the analysis of phosphory- lated metabolic intermediates and to apply the system to typical metabolic problems of current interest. Ion exchange chromatography has been selected as the basis for the automated analysis because of the possibility of achieving separation of a maximum number of compounds in minimal time. In order to clarify the background for the decisions made, a brief review of the available methodology is presented. 3 Enzymatic Methods The enzymatic assay of metabolic intermediates owes much of its value and popularity to the great specificity and sensitivity which can be realized in enzymatic reactions. Compounds which have nearly identical structure often can be determined independently with great certainty by this method. The general principle involves conversion of a particular compound enzymatically to another compound. The second compound, or a cofactor in the enzymatic conversion, is coupled directly or indirectly to a system which can be monitored spectrophotometrically. Commonly employed as such spectrophotometric or fluorometric indicators are the NAD/NADH or the NADP/NADPH systems, but many other coupled systems have been used. Specificity depends to a large extent on the purity of the enzymatic preparation used in the assay. Commer cial preparations of individual enzymes are often con taminated with residual amounts of other enzymes which can detract from the specificity of the intended assay. This is particularly true when an enzymatic assay is used to determine a particular intermediate present in a cellular extract, since some of the many other compounds present in these extracts might prove to be substrates for contaminating enzymes. When purified enzyme preparations are available, the method is ideally suited for studies focusing on a particular intermediate. The conditions of the assay can be adjusted for optimal sensitivity and specificity of that one compound, and several assays can be run con currently with a considerable saving of time. However, when several compounds present in cell extracts are being determined, optimal conditions of buffer, pH, ionic strength and coupling system must be adjusted for each enzymatic assay. A whole series of incubations, each at different conditions, must be carried out. This becomes very laborious and time-consuming. Certain enzymatic procedures, directed at a few particular intermediates, have been automated (1). However, these automated methods are of practical advantage only when a large number of samples are being assayed concurrently, due to the cost and instability of the enzyme preparations. An enzymatic procedure used in assaying a particular metabolic intermediate in a study does not reveal the status of other intermediates. Significant alterations in compounds not under assay will remain undetected. This constitutes a serious drawback, particularly when information is needed about a metabolic pathway as a whole. An enzymatic assay determines the concentration of an intermediate and by itself it is not suited for a study in which the specific activity of a compound is needed. This is a typical requirement when distribution of radioactivity from a radioactive precursor into inter mediates is sought. An independent method is usually required to obtain this information. Electrophoretic Procedures Both electrophoresis and chromatography depend on physico-chemical differences between the various metabolic intermediates for their separation and quantitation. High voltage paper electrophoresis has been used by a number of workers to separate various metabolic intermediates and nucleotides (2). The separation is conducted in an acidic buffer on Whatman No. 1 or No. 3 chromatography paper at 3,000-5,000 volts. To enhance resolution, without increasing the length of the paper, two dimensional electrophoresis is generally preferred over electrophoresis in a single dimension. An advantage of this method is the short time required for the separation. Many compounds are already well resolved after one hour of electrophoresis. Selec tive resolution of certain compounds can be further enhanced by an appropriate choice of buffer pH. 6 Overall limits on the duration of the electro phoresis and hence on the degree of separation possible, are set by the building up of heat as electrophoresis proceeds. This heat generated during the runs can break down more labile compounds and lead to decreased recover ies. The heat reflects an energy loss which is propor tional to the resistance of the system and to the square of the current passing through the system. One of the problems in the design of high voltage electrophoresis equipment is the provision of a means for the adequate removal of heat. The separated compounds are visualized on the paper by UV light or are detected by staining the paper with an appropriate reagent. Sensitivity of this method is in the nanomole range. Estimates of quantity are made from the intensity of the spots. Various factors such as overlapping of spots, neighboring effects of one compound on adjacent ones and uncertainties in estimating the intensity of the spots, render the procedure at best semi-quantitative. High voltage paper electrophoresis is a useful method within its limitations for metabolic studies. It has much to offer in specific applications, and it is of particular value when used in conjunction with other methodological approaches to a problem. Chromatographic Separation Chromatography has been widely applied for the separation of metabolic intermediates. The wide latitude in the conditions possible and in the modes of method ology have prompted much work aimed at improving both resolution and sensitivity. Evaluation of the relative merits among the variety of methods available is generally a function both of the degree of separation of the various compounds achieved by a method and of the relative sensitivity of the procedure to detect these compounds once separated. The term "chromatography" denotes a procedure in which a solution of substances to be separated is passed, in a direction determined by the arrangement of the particular apparatus employed, over a more or less finely divided insoluble support material. The effect, as a result of one or more mechanisms dependent upon the conditions employed, is selective retention of the individual components to different extents. The underlying mechanisms are various and include the partitioning of the moving compounds between two liquid phases, or between a liquid and a gas phase. It may involve their being reversibly bound on the surface of the support material, and if in this case physical surface forces are mainly involved, the procedure is referred to as adsorption chromatography. In partition chromatography, a liquid phase is present on the support medium. The best known example is paper chromatography and, to a degree, thin layer chromatography. In ion exchange chromatography, true ionic chemical bonds are formed reversibly between the moving compound and a solid phase of resin or analogous material. Ion exchange chromatography has been performed on ion exchange paper, thin layer chromatography and on columns of ion exchange resins. Often in practice, as in the isolation of biological compounds, sequential combinations of adsorp tion, ion exchange and partition chromatography may be involved. Adsorption chromatography; A particular form of adsorption chromatography which has been applied to phosphorylated compounds involves using columns of activated charcoal (Norit A, Darco G60, Merck NP) to adsorb various glycolytic inter mediates which are subsequently eluted with dilute 0.01N NH4OH. A serious drawback of carbon chromatography appears to be the difficulty in obtaining good recoveries of some of the compounds (3). Carbon shows a size selectivity which has been of value in the column chroma tographic separation of mixtures of simple sugars and oligosaccharides (4). Crane and Lipmann (5) have pointed 9 out the advantage of carbon for the analytical separation of nucleotide sugars from other phosphorylated compounds, and this method has been much used, particularly in batch preparation. Carbon columns do offer certain particular advantages but do not seem to provide the reproducibility or general applicability. Gas-liquid chromatography: In gas-liquid chromatography (GLC), a stationary phase is held on an inert support material in a column. The mixture of compounds to be separated is passed over the stationary phase as a vapor in a carrier gas and separation occurs as a result of differential retention of solute by the stationary phase. As the separated compounds emerge from the column they are detected by one of several means. In applying this method to phosphorylated compounds, it is necessary to make volatile derivatives. Sweely, et. al. methylated the phosphate groups and then tri methyls ilylated the OH groups on the compounds (6). Hashizume and Sasaki (7) trimethylsilylated both free sugar phosphates and their salts and obtained volatile derivatives; and Horning, et. al. replaced the conven tional trimethylsilylating agents with bis TMS acetamide (8). The variation in preparation of volatile deriva tives has resulted from the apparent instability of some 10 of the compounds to the trimethylsilylation procedure, resulting in the loss of the phosphate moiety during separation. Often more than one peak is obtained for a single compound. For example, the trimethylsilyl derivatives of the various sugar intermediates demonstrate the alpha and beta furanose and pyranose forms (9). Detection of the various compounds separated by gas-liquid chromatography is generally done with a flame ionization detector. Since the majority of these com pounds are altered during the process of separation, identification of compounds becomes difficult and in more complex cases depends on mass spectra analysis of the column effluent. No studies are available on the systematic applica tion of this method to biological extracts. Nucleotides have not been reported as analyzed by this method. Paper chromatography: Paper chromatography is the outstanding example of partition chromatography. It was used for the separation of certain mixtures by Schonheim as early as 1861 under the name "Capillar-analyse." It was re-introduced in the modern sense of the term by Consden, Gordon and Martin in 1943. Partition occurs between two phases, the mobile solvent phase and the stationary phase which is bound to 11 the cellulose of the paper. This relatively simple technique has been accommodated to a number of variations, including ascending and descending chromatography, single and two dimensional chromatography, all aimed at improv ing resolution. Hanes and Isherwood were the first to study the separation of biologically important phosphoric esters on filter paper (10). Various solvent systems were used, such as acidic water-immiscible mixtures (e.g., t-amyl alcohol, water, formic acid), or basic water miscible systems (e.g., ethyl acetate, pyridine, water). Quanti tation was performed by spraying the paper with an acid molybdate reagent, followed by heating to develop the blue color of the phosphomolybdate complex. Spots then were cut out and the phosphorus content determined. This technique was further refined by Leuthardt and Testa (11)r Cohen and Scott (12), and Bandurski and Axelrod (13). The separation of several nucleotide sugars has been described by Paladini and Leloir (14), Kenner, et. al. (15) and Munch-Petterson, et. al. (16). Coh^n and Carter have described the paper chromatographic separa tion of various adenine nucleotides (17). Most of the techniques which have been employed depend on detection of the various compounds either by hydrolysis on the paper of the various compounds, 12 followed by staining for inorganic phosphate, or by visualization of the nucleotides with ultraviolet light. Radioactive compounds resolved by paper chromatog raphy can be located by overlaying the chromatogram with a photographic film, allowing an appropriate exposure time and then developing the film (radioautography). Also, radioactivity can be detected with a scanner type of counter. Radioactive tracer techniques have been used in conjunction with paper chromatography by Calvin and Benson to study the reactions in photosynthesis (18). The radioactive compounds were separated by a two-dimen sional chromatographic technique* and detected by over laying with a photographic film. They were able to demonstrate labelled amino acids, neutral sugars and phosphorylated compounds on a single chromatogram (19). In the opinion of these authors, however, the results obtained should be interpreted qualitatively, due to uncertainties about the purity and identity of the various radioactive spots. Ion exchange liquid chroma tography helped confirm and further resolve the various components observed in work based on paper chromatography. Paper chromatography is ideally suited when a small number of very similar compounds need to be resolved, since buffer conditions can be chosen which will maximize the chemical or structural differences between these 13 compounds. Generally such compounds are first isolated from other intermediates by electrophoresis or by ion exchange chromatography for further resolution by paper chromatography. Vanderheiden, for example, separated the various diphosphates, obtained by high voltage electrophoresis of erythrocyte extracts, with a paper chromatographic technique (2). He obtained several compounds which previously had remained undetected. These studies point to the potential sensitivity of paper chromatography under selected circumstances. It is impossible, however, to maintain this sensitivity when working in a single sample with a large group of compounds encountered in a biological extract, such as monophosphates, polyphosphates, nucleotides and nucleo tide sugars. The conditions necessary to resolve the classes of compounds are not suited to provide concurrent separation of compounds within each class. Except for special "banding" techniques, paper chromatography is generally considered an analytical and not a preparative method. Although spots can be eluted from the paper, confirmation of the identity of the compounds by organo-chemical or enzymatic means is usually difficult due to the small quantity of sample. Reliance is placed on comparison of Rf values with those of standards to identify the various metabolites. This, 14 in addition to the time usually involved in paper chromatography, make this technique less favored as a primary tool for the separation of glycolytic inter mediates and nucleotides than for confirmatory work. Thin Layer chromatography: A procedure derived from paper chromatography is thin layer chromatography. In its simplest form, a thin layer of cellulose powder is used to coat a glass or plastic backing sheet. Modified forms of cellulose can be used which confer ion exchange properties. Alumina and other adsorption media also can be used, for thin layer adsorption chromatography. In the context of the present interest, thin layer chromatography has been employed successfully in the separation of nucleotides and nucleotide sugars on layers of cellulose using distilled water as the solvent (20, 21). An advantage is the short duration of the separa tion when compared to similar separations on paper. More recently a method has been described in which increased sensitivity and resolution are obtained by thin layer procedures which employ a modified cellulose as an ion exchange medium (ECTEOLA-cellulose) (22). It can be stated in general that thin layer chromatography does not provide a comprehensive separa tion of glycolytic intermediates, nucleotide and 15 nucleotide sugars, occurring in the same sample. Ion exchange chromatography: In contrast to adsorption and partition chromatog raphy, in ion exchange chromatography true heteropolar bonds are formed between the compounds being separated and the ion exchanger. The formation of these bonds is influenced by steric factors, pH, ionic strength and the nature of the ion exchanger. Once these bonds are formed, compounds can be selectively removed from the ion exchang er by reversing those conditions which favored exchange of the compound for the counter ion on the ion exchange matrix. This technique has been applied to the separation of macromolecules such as proteins and for small molecules as well. The solid support matrix of the exchanger may be based on aluminum silicates, synthetic resins or polysaccharides, or other substances. The principles of ion exchange have been incorpora ted into thin layer (21, 22, 23, 24, 25) and paper chromatography (26, 27), differing only in the solid matrix to which the charged group is covalently bound. In general, however, the column technique has gained wider use. The method lends itself for both preparative and analytical purposes. The columns can be easily regenerated, and several successive separations can be 16 accomplished with the same resin bed. Identification of the various intermediates is facilitated since the effluent of a column can be manipulated in various ways to detect the compounds being separated without the inherent limitation of having to visualize and quantitate a compound while on the separating medium. For the separation of metabolites found in acid solution extracts of biological material, columns of anion exchange resins have been employed. These strongly basic resins are composed of quaternary ammonium exchange groups attached to a styrene-divinylbenzene polymer lattice crosslinked to different extents. Anions of acidic, basic and neutral salts are exchanged for the counterion of the resin, generally chloride, formate or hydroxide. Cohn introduced the use of a Dowex-chloride (a styrene-divinylbenzene resin) as a means of separating various phosphorylated intermediates and nucleotides (28). Further resolution of these compounds was achieved by Hurlbert (29) using a gradient elution technique to remove compounds from the resin. Other modifications have been published, relating primarily to the type of gradient used (30). Nucleotides are eluted from the Dowex resin simply by varying the ionic strength of the buffer. The order 17 of elution suggests that the number and the pK of the phosphate groups of the nucleotide governs the degree of binding of these compounds to the column. Steric effects, although secondary, must also play a role, since nucleo tides with an identical number of phosphate groups, but different base composition, can also be resolved. A Dowex-formate resin also has been employed, eluted with a gradient of ammonium formate - formic acid (3). However, nucleotides are generally identified by their characteristic absorption at 275/260 nanometers, and adsorption of formate at this same wavelength makes this form of the resin less desirable for the purpose. In the separation of phosphorylated intermediates of glycolysis by ion exchange column chromatographic tech niques, monophosphates are easily separated from diphos phates and polyphosphates by increasing the ionic strength of the eluting solvent. Both Dowex-chloride (31) and Dowex-formate (3) have been used. Separation of the various important monophosphates has not been equally successful, and it usually has required addition al chromatography steps to resolve these compounds. Steric differences have been exploited to affect sufficient difference in affinity for the resin to resolve these metabolites. One successful technique has been to form borate complexes of those compounds at alkaline pH 18 followed by separation of the complexes using a Dowex- chloride resin and a suitable elution gradient. The complexes appear to be sufficiently different to make this technique practical, since the steric configuration of the hydroxyl groups on these phosphorylated inter mediates dictates the kind of borate complex formed. This method has also been employed to separate neutral sugars from each other using a Dowex-borate resin (32). Since acid soluble extracts of biological fluids contain both nucleotides and phosphorylated intermediates of glycolysis, separation of all these compounds by a single chromatographic run is usually very difficult due to overlapping of the various compounds. Presently, with the availability of resins of much smaller diameter and of uniform particle size, resolution can be increased considerably, and separation of all of the compounds of interest is possible if the elution gradient is carefully designed to avoid overlapping of peaks. THE PROBLEM AND THE PLAN OF ATTACK The Problem: Biochemical investigations of human tissues have greatly enhanced our understanding of the etiology, diagnosis and management of metabolic diseases. These studies have often been hampered by the lack of method ology sufficiently sensitive to analyze effectively the small amounts of material which often can be procured from human sources. Analysis of tissue for phosphorylated intermediates of metabolism has involved difficulties due to the relatively large amounts of tissue often required. In addition, much of the methodology applied to the analysis of these intermediates has been lengthy and time con suming. For these reasons, time course experiments which would reveal the kinetics of uptake and distribu tion of substrate in these tissues have not been practical in many instances. Often, methodological shortcomings have been cir cumvented by resorting to procedures which provide only semiquantitative data or which focus on a limited number of particular metabolites exclusively, sacrificing interesting and valuable information concerning other intermediates in a pathway. The present study is directed 19 20 to the development of an analytical system which permits the analysis of phosphorylated intermediates in human tissue with sufficient sensitivity to obviate the require ment for relatively large amounts of source material. A secondary purpose, but one of significant practical value, will be efforts directed to reduce the time factor involved in the analytical procedure. To accomplish this, without sacrificing resolution or reliability of the method, the procedure will be automated. A third feature which will be incorporated into the design of the system will be the potential to simultane ously monitor radioactivity in the various phosphorylated intermediates. This will allow a more detailed explora tion of the kinetics of uptake of radioactive substrates incubated with these tissues and of the distribution of radioactivity among the intermediates. A final goal of the present study will be the application of the analytical system developed to some biochemical problems of current interest, with erythro cytes and platelets as the biological material concerned. Plan of Attack; A survey of the literature suggests that liquid ion exchange column chromatography has been an effective method for the separation of the various phosphorylated intermediates of metabolism. The merits of this approach 21 appear to be versatility and flexibility. By altering the eluting gradient, resin or other parameters, enhance ment of resolution can be achieved which are not possible with other less flexible methods. Also, since the pro cedure is nondestructive, separated compounds which can be identified by other physico-chemical criteria and by chemical or biochemical means. This general method is ideally suited for automation. The compounds separated on the ion exchange medium appear successively in an eluate stream which can be manipulated in various ways to effect the desired automated detection. In the present study, liquid ion exchange chromatog raphy will constitute the basis of the analytical method. The Dowex-1-chloride anion exchange procedure originally described by Khym and Cohn (30) will be used as the basis for column separations since it has the promise of being suitable for the separation both of glycolytic inter mediates and of nucleotide phosphates. The resin, column geometry, eluting gradient and eluting time will be varied systematically to obtain the most desirable com bination of these parameters. The gradients will be tailored by gradient elution devices suited to the particular purpose. Eluting volumes will be kept as small as possible to avoid dilution in the eluate stream, thereby enhancing the sensitivity of detection of 22 compounds. Consideration will also be given to decreasing the total time required for separation of compounds while maintaining optimal resolution. A key component of the automated analysis is an ashing mechanism originally designed by Dr. Samuel Bessman (Department of Pharmacology, School of Medicine, University of Southern California) which will be employed to continually ash the elution stream from the ion ex change column. The asher will be modified as necessary to accommodate the rates of eluate delivery, of ashing and of sample removal after the ashed material has been reconstituted. For the phosphorus determination of the ashed and reconstituted column eluate, a conventional Technicon manifold using the phosphomolybdate reaction will be employed. The most suitable reagents and conditions for this assay will be evaluated systematically using as criteria reagent stability, sensitivity, linearity of response and optimal reaction temperature and time. The colored phosphomolybdate complex will be directed to a spectrophotometer fitted with an extended path flow through cell to obtain maximum sensitivity by increasing the light path through this cell. The design will be balanced against possible losses in resolution due to backmixing within the flow cell. 23 The radioactivity of the various intermediates separated by the column will be monitored with a two- channel scintillator counter which has been redesigned to accept a flow-through scintillation cell. This cell will be interposed in the eluate stream prior to ashing of the compounds. The flow-through cell will be constructed to maximize the efficiency of radioactive detection and minimize the amount of backmixing of the eluate stream within this compartment. The output of one channel of the scintillator counter will be connected to a ratemeter to obtain a rate expression of CPM/Time, while the other channel will print out CPM on a ticker tape. Both the spectrophotometer and ratemeter will be connected to the same multichannel recorder. The proper input voltages to the recorder will be obtained with appropriate voltage dividing circuits. The systematic development of each of the components of the system may entail simultaneous modifications in other parts of the system to accommodate the variation introduced to overall performance. Unexpected logistic problems will be dealt with as they occur. It is hoped that the outcome will be a system whose individual com ponents are integrated and matched in sensitivity and resolution. Experiments with two different blood cells, as 24 representative of human tissue, will be performed to demonstrate the applicability of the automated analytical system to problems of current biochemical interest. Erythrocytes will be employed to study the in vitro effect of glucose on galactose metabolism and energy production in these cells. The extremely active nucleo tide metabolism of platelets will be examined, specifi cally with respect to the salvage pathway of hypoxan- thine. The experimental details and the results obtained will be described. AUTOMATED SYSTEM FOR ANALYSIS OF PHOSPHORYLATED METABOLIC INTERMEDIATES The methodology developed in this investigation has been for the separation and quantitation of phosphory lated intermediates produced in erythrocytes and other tissues during the course of in vitro incubation. An automated system, based on an ashing device designed by Dr. Samuel Bessman (33), was developed for this purpose. The capability for quantitation of both carbon-labelled compounds was included. Modifications were made in the automated phosphate analysis component in order to increase sensitivity and reliability. Emphasis was given to the column chromatography component of the system in order to achieve the separations required in a reasonable time. Erythrocytes were used as a tissue of choice during the development of the procedures, and elution gradients were adjusted to provide optimum conditions for separation of intermediates present in the red cell in unusual amounts (e.g., 2,3-diphosphoglycerate) from other components closely adjacent in the elution pattern. During application of the system to platelets, similar consideration was given to separation of certain nucleo tides. 25 26 The Phosphorus Analyzer as a System A flow diagram of the automated procedure is shown in Figure 1. A Dowex-l-chloride anion exchange liquid chromatographic method is used to resolve phosphorylated intermediates. Borate complexes of these compounds are separated on a column employing an elution gradient increasing in NH4CI concentration and decreasing in pH. The column effluent is passed first through a flow cell in a liquid scintillation counter and then directed to an automated asher. The asher serves both as a fraction collector and as a device for drying and ashing each fraction collected. The ash, containing inorganic phosphate derived from phosphorylated intermediates, is dissolved, and a probe sampler extracts a measured volume which is pumped by a Technicon arrangement to a manifold where it is mixed with reagents for the automated phos phorus analysis. The stream flows through a heating bath to expedite color development and then through a flow cell in a Bausch & Lomb spectrophotometer. The output of the spectrophotometer is coupled to a Honeywell multi channel recorder which plots the chromatogram in terms of optical density versus time. The scintillation counter provides a continuing printout record of radioactive counts on a selected time interval. In addition, the output of the counter is routed to the multi-channel 27 S P E C T A O P H O T O M £ T . S C IN TIL LA TIO N COUNTER. taps printout Figure 1. Flow diagram of phosphorus analyzer system. 28 recorder through a rate meter. The radioactive trace is related to the phosphate chromatogram by allowance for the time interval of processing through the automated sequence. The system was developed over a period of more than two years as the investigation proceeded. Each of the components will be described separately. A. Column and gradient; The column finally found most suitable is 6 cm. long and 7 mm. in internal diameter. As purchased from Chromatronix (Berkeley, Calif.), it is fitted with an injectable sample application device which allows loading of the sample without interrupting the run. With a Beckman Accuflo pump in use, flow rate is maintained at 0.80 ml/min. with a pressure of approximately 200 lb/sq. inch. When it became possible to substitute a Chroma tronix pump for the metering function, pressures up to 500 lbs/sq. inch and higher flow rates were obtained. The column is loaded with Aminex A25 (Biorad), a spherical Dowex polyvinyl benzene chloride anion exchange resin with a mean diameter of 17.5 microns. Since glucose and other neutral sugars are not retained by the column, they are washed off with 0.0025M NH4OH prior to starting the gradient for elution of phosphorylated compounds. Lactate also is removed from 29 the column with this preliminary NH4OH wash, but it appears in the radioactive section of the chromatogram, clearly separated from the neutral sugars. A gradient-making device was constructed from plexi glass. The cylindrical mixing vessel, which has a capacity of 100 ml., is connected to four cylindrical side vessels, each of 25 ml. capacity, staggered with respect to each other. The total side-vessel capacity then is 100 ml., deliverable in four 25 ml. segments. A controlled gradient can be produced which is concave in four segments, with changing slopes depending on choice of buffer concentrations. For many of the studies done, the starting buffer was 0.02M in NH4CI, 0.0025M in NH4OH and 0.08M in sodium borate. The terminal buffer was 0.33M in NH4CI and 0.0025M in NH4OH. Depending on the concentrations employed, the device also can produce a linear or convex gradient. B. Scintillation counter: The scintillation counter in the phosphorus analyzer system provides for detection and quantitation of radio active intermediates produced during metabolic studies with labeled substrates. In addition, it also permits detection of non-phosphorylated compounds such as lactate, pyruvate, sugars, etc. 30 A Unilux III Nuclear Chicago liquid scintillation counter is used in the present system. Anthracene-filled plastic flow-through cells are commercially available in various sizes, but, due to buildup of background, it was found necessary to replace the expensive commercial cell periodically because the anthracene could not be removed and replaced. It was found that an improvised simple glass U-tube cell could serve equally as well as the commercial device. The glass cell included a pro vision for ready replacement of the anthracene when required. In the present studies, limited to the use of C-*-^, the efficiency of the flow cell has been 30-40%. Quenching of counts has been found to be negligible with the eluting buffers used. The scintillation counter provides a tape printout of counts per minute. The output of the counter also goes to a Baird Atomic ratemeter (Model 432A), and the ratemeter in turn is connected to the multi-channel recorder which records as separate traces the phosphorus analysis data representing the phosphorylated interme diates separated on the column, and the presence of radioactivity in the various peaks. The display on the recorder chart serves to identify the radioactive com pounds: the tape printout is used for quantitation of the radioactive peaks. 31 C. Asher: This component of the analytical system serves as a fraction collector and provides for ashing of the frac tions of the column eluate. After the ashed material has been reconstituted to a specified volume, a measured aliquot is assayed, in an automated sub-system, for the phosphorus derived from phosphorylated intermediates. The particular asher used in this investigation was constructed by Hauptman Enterprises of Los Angeles, California. It was designed by Dr. Samuel Bessman (33). The principal element is a circular turntable around the perimeter of which are suspended 36 borosilicate glass cups, each of approximately 1.5 ml. capacity (Figure 2). Adjacent to the turntable for about half its diameter, but attached to the frame of the device, is a heating coil located in close proximity to the bottom half of the cups. The windings of this coil come closer and closer together as it progresses along the circumference of the turntable, thereby increasing progressively the effective heating capability. The rotation of the turntable is governed by a cam located inside the body of the device. The turntable rotates on a time schedule, placing a succeeding cup under a probe delivering the column effluent. A cam setting to provide a movement every 30 seconds proved 32 HECo n s t it u t ia is pao&b asp/aatino raoee k/A*H PKoQSi) vacuum pauses oeuvtAv or column's a* " t f l u e n t Figure 2. Top view of automatic asher. 33 optimal. At the next cup station another probe delivers perchloric/nitric acid to the cup to aid in ashing. As the turntable rotates, the cups are exposed to increasing heat from the heating coil and water is evaporated and the residue is ashed. The ashed material which contains the phosphate residue of the phosphorylated intermediates, is dissolved in 1% sulfuric acid with a wetting agent added, and a set volume is aspirated by a probe connected with a length of transmission tubing to the Technicon pump. The cups are washed after the fluid is withdrawn with a series of double probes, one probe dispensing distilled water into the cups, the other aspirating it. In the process of ashing, the carbon of the organic compounds in the column eluate fractions is converted to carbon dioxide. Because of the presence of radioactive carbon-labelled compounds in the fractions, the asher was isolated in a well-vented hood to insure that C-^C^ contamination of the room air did not occur. D. Technicon subsystem for phosphorus analysis: 1. Technicon manifold: The probe-sampled aliquot of the solution (usually in dilute acid) of the ashed material from the automatic asher is pumped by a Technicon peristaltic pump to meet a stream delivering reagents for the colorimetric deter mination of phosphorus. The optical density of the color 34 produced is measured in a spectrophotometer. The mani fold diagram illustrates the several components of the system (Figure 3). All coils, tubing, nipples and fittings were purchased from Technicon Corporation (Raritan, N.Y.). Identification of these parts by number refers to the standard designations used by this company. The flow rate of the reagent mixture stream was approximately 0.55 ml/min. This rate is governed by the size of the pump tubes delivering the reagents and air and of that aspirating sample from the asher. A rate this slow is required to maintain the final assay mixture at a minimum voluem to obtain maximum sensitivity. The Technicon heating bath was modified by removing the standard double heating coil and replacing it with a single turn coil. The standard coil was unsuitable in relation to the slow rate of the assay mixture and optimum heating time. The heating bath was filled with silicone oil (Technicon) and maintained at 70°C with the thermostat. For some modifications of the assay, lower temperatures were used, or no heating at all. The stirring mechanism of the heating bath was found unneces sary. The debubbler was a standard Technicon debubbler to which a small glass tube had been attached to direct the air from the bubbles and small carryover of reaction 35 N H a -M OU/BDATS fa tA H Q C - $LU S ) 6 / n c l c t u r n COtL. A )A ( oRa n o s - Wh ITS) HO-r A A A A r-« ./ & A M P L 6 < 9 l a c i c ) A6CO A&IC A C iO C O R A N L t- b L V t\ Con A S T U A n A R oM CoLCHOAKT&l C0LACt$_ _ _ _ _ _ _ _ 7 * 0 j. 1 /A iT S S/Asatr tiaw HCAT/A/G bAT^f D&BUB&LeZ. r-C3 TO U tA bTff GOLORtMATff V/TH M.OL3-THROUGH CELL Figure 3. Manifold diagram of phosphorus analysis. 36 mixture to a waste flask. 2. Phosphate analysis reagents: The colorimetric procedure finally selected is an adaptation of the method described by Brewer, et. al. (34), which uses ascorbic acid as the reductant. Several other procedures were tested but they either did not prove sufficiently sensitive or which posed other dis advantages . The method adopted was found to be the most sensitive of all those tried. The reagent is stable for approximately 20 hours. Ascorbic acid has been purchased from two different commercial sources, with no detectable differences in sensitivity or stability of the assay. All reagents were prepared in deionized water with commercially available analytical grade chemicals. The wetting agent employed, LEVOR V, is a non-phosphate detergent which aids in obtaining a smooth reagent stream flow and a uniform bubble pattern. The reagents for phosphate analysis were: a. 4.6% w/v ascorbic acid in 1% H^SO^: Ascorbic acid is added to 1% H2SO4 solution. When the ascorbic acid is dissolved, 1 ml. LEVOR V is added per liter of solution. This reagent must be prepared prior to each run since it is slowly oxidized and is stable for only approximately 20 hours. 37 b. 0.9% w/v ammonium molybdate. 4H20 in 1% H2SO4: The salt is added to 1% H2SO4 and, when completely dissolved, 1 ml/L LEVOR V is added. It was found to be imperative that deionized water of high purity is used when making this reagent, since small traces of metals in ordinary distilled water will slowly produce a blue color in this reagent. 3. Spectrophotometer and recorder: The intensity of the color developed is measured in a Bausch & Lomb Spectronic 70 spectrophotometer. This instrument has a number of desirable features for the present purpose. It has a sensitivity of 0.5%, voltage output of 1.2 V and a wave-length range of 19 0 to 960 mu. The far red regions of the spectrum (above 6 40 mu) are detected by a separate red sensitive phototube. There is no apparent electronic drift during 30 hours of con tinuous operation. The cell holder compartment is such that it can accommodate a variety of sizes and shapes of flow cells. The reaction stream is delivered and withdrawn from the flow cell by Technicon transmission tubing through the test tube adapter located on top of the cuvette compartment. The mechanical shutter in the light path on the spectrophotometer must be held open with a small 38 clamp when using the flow cell. After constructing a variety of flow cells made by drilling holes in blocks of black plastic, it was found that the requirements of the present investigation could best be met by the Technicon flow cells used in the Technicon SMA-12 colori meter. The spectrophotometer is connected to a Honeywell multi-channel recorder (Model Electronik-112) to obtain a permanent record of the changes in optical density versus time. This recorder requires 100 mV of input for full scale deflection while the output of the spectro photometer is 1.2 volts. Consequently, it was necessary to interpose a voltage divider (based on a 10K helipot) between the two instruments to obtain correspondence between the visual display on the spectrophotometer and the printed display of the recorder. The chart paper employed is graduated in OD units which facilitates quantitation of the peaks. The chart speed was adjusted to a speed of 4"/hour, while the pen prints every 21 seconds. A simultaneous second tracing is obtained when the scintillation counter is used in conjunction with the phosphorus analyzer. Since the output of the spectro photometer is maximal at 0 optical density while the ratemeter of the scintillation counter has a minimal output at baseline counts, the printed recorded displays from these two different instruments run in opposite directions. Radioactivity in the column eluate is detected shortly after emerging from the column, but the detection of phosphate is delayed by the time required for passage through the asher and the color development. Consequently, an adjustment for the lag period is neces sary. The time difference can be measured by injecting, prior to starting the elution, a small amount of a labelled organic phosphate (e.g., 1-C^-galactose-l- phosphate) into the stream just before it enters the scintillation counter. Comments on Development of Automated Analysis A. Extraction of metabolic intermediates: The commonly used procedures described in the liter ature for extraction of phosphorylated intermediates from tissues are: (1) the precipitation of proteins with 5% perchloric acid followed by neutralization with potassium hydroxide, (2) the precipitation of proteins with 10% trichloroacetic acid followed by removal of the TCA with ether, or (3) the precipitation of proteins by heat coagulation. Removal of proteins by dialysis or ultra filtration has also been used with certain types of tissues, but the high hemoglobin concentration of 40 erythrocytes renders this procedure impractical for some of the present studies. In the perchlorate procedure, a high salt concen tration is used to precipitate the proteins, and the metabolic intermediates are recovered in the supernatant. However, the salts tend to prevent the binding of the intermediates to the ion exchange resin, with consequent elution of compounds with the void volume of the column. Extraction of metabolic intermediates with 10% TCA is a favored method, one which has been applied to erythrocyte studies by Bartlett (3). One volume of packed red cells is added slowly to two volumes of ice cold 10% TCA. After centrifugation at 5,000 RPM for 10 minutes, the TCA is extracted from the supernatant with four washings of 3 volumes of ether. This procedure was used in preliminary experiments of the present study, but it was abandoned at the time when evaluation of lactic acid became important, since the lactic acid is removed by the ether washings (35). In an experiment designed to determine the magnitude of this loss it was found that 85% of an added known amount of lactic acid went to the ether phase. The extraction method adopted for application of the automated methodology to erythrocyte studies is a modifi cation of the procedure described by Robinson (36). The 41 packed red cells are placed in a boiling water bath for 4 minutes, cooled in ice and spun down at 6,000 RMP for 10 minutes. The supernatant is removed and the precipi tate is washed with one volume of water, respun and the supernatants pooled. Comparison of this method with the TCA procedure revealed no differences in recovery of phosphorylated intermediates, and lactic acid was not lost. B. Column and gradients: The most critical and time-consuming part of the development of the automated analyzer system was achieving optimal separation of the different phosphorylated com pounds in the acid soluble extracts of biological fluids. During this phase several variables were explored, in cluding different column geometries, different resins, different types of eluting gradients, and changes in flow rate. It was not possible to consider methodically each one of these variables independently. Some combina tions of conditions were reasonably satisfactory, but others proved incompatible with the remainder of the system. From a long series of gradually improving results, we obtained our present knowledge concerning choice of resin, column design and gradients found to give optimum separation of the various phosphorylated compounds. 42 1. Resin; Phosphorylated intermediates are best separated by a strongly basic quaternary ammonium ion exchange resin. These are styrene divinyl benzene particles covalently linked to a quaternary ammonium group. The commonly used anions are formate and chloride ions. The Dowex-1 chloride system was adopted for the present method because the Dowex-1 formate system was found to present several undesirable characteristics. With the formate system, ammonium formate is required in a concentration of up to 4 M to elute all the compounds of interest from the column. This high molarity proved detrimental to uniform and complete ashing of the columns eluate. It was also learned from a few experiments with the formate system that it was difficult to separate the different monophosphates in a single run. Furthermore, the acidic nature of the eluate quenched the counts detected by the scintillation counter, decreasing the efficiency of the scintillator. For these reasons the Dowex-1 chloride resin came to be used exclusively in development of the system. In early phases Dowex-1 chloride 200-400 mesh, 8x cross- linked was employed. Prior to using the resin, attempts were made to obtain a more uniform particle size by suspending it in starting buffer and removing the "fines" 43 by decantation. A similar resin but with only half the crosslinkage was also explored, but this resin proved too spongy when the pressure necessary to maintain the flow rates was applied to the columns. The 8x Dowex-1 chloride resin proved to be suited to separating the different phosphorylated intermediates but not ideal for the objectives of the study. Some shortcomings in resolution were overcome by using very shallow gradients. However, certain inherent limitations could not be avoided. The long duration of the run required by the shallow gradients resulted in diffusion of compounds in the resin bed which was reflected by peak broadening as the chromatography proceeded. Also, the large volume of gradient buffer required limited overall sensitivity of the method because of dilution of the compounds emerging from the column. It is known that resolution in ion exchange chroma tography varies inversely with the particle diameter of the resin in an approximate linear manner (37) . This consideration indicated that one means to improve reso lution and decrease the time required for elution of the compounds was to use a resin with a smaller, more uniform particle diameter. A new resin, Aminex A-25 of 17.5 + 2.5 microns in diameter (Biorad) was tried. This resin is similar in composition to the Dowex-1 chloride except 44 for the approximately ten fold decrease in particle diameter and greater uniformity in particle size. The expected improvement in resolution was obtained with Aminex A-25 due to the smaller particle diameter. In addition sensitivity of detection increased greatly. Less shallow gradients were possible, and elution time was decreased similarly approximately ten fold. A number of other spherical small diameter resins were tried, but the Aminex A-2 5 proved to be the most suitable. As the use of the automated system has continued, Aminex A-25 has proven very satisfactory for the separa tion of the various phosphorylated intermediates. It has been verified experimentally that the limits of resolution of this resin, with the column finally adopted, exceed those of other components of the analytical system such as the asher, and phosphate detection subsystem. 2. Column design: During the development of the system much time was devoted to studies of columns of different geometries, guided by general concepts concerning column chromato graphy. The objectives were to achieve optimal separa tion, resolution and sensitivity. The initial experiments were conducted with columns 120 cm. long, 60 cm. of which had an internal diameter of 1.2 cm. followed by 60 cm. with an internal diameter of 45 0.8 cm. Use of this long, narrow column was prompted by the known fact that resolution of compounds is approxi mately inversely proportional to the square of the diame ter of the column. The smaller diameter of the second half of the column was an attempt to increase further the separation of compounds resolved in the broader half of the column. At a flow rate of 0.8 ml/min. (requiring approximately 400 lbs/square inch pressure), separation of phosphorylated compounds required runs of 25 hours duration. Although most compounds were clearly separated, the long duration of the run caused some spreading of the peaks due to diffusion of the compounds in the resin bed, an effect particularly noticeable with compounds eluting late in the run. Since elution was with 1.5 liters of buffer, dilution of compounds in this large amount of buffer limited the sensitivity of detection to compounds in a concentration in terms of phosphorus of 0.5 umoles or larger. Subsequent experiments showed that the degree of resolution obtained by this column was no better than that obtained by using a relatively short column made from a 1 ml. serological pipette. The pipette was fitted with a plastic sleeve with a small glass wool plug and a Technicon nipple at the tapered end to support the resin bed. Since this column was in many ways easier to handle 46 particularly with respect to changing the resin and column loading, many studies with different gradients were conducted with this column. The longer columns were abandoned since there was no advantage in using them, and there were problems resulting from the high back pressures generated and from the difficulty in filling these columns with new resin after each run. Chromatographic runs were also carried out with columns improvised from 0.5 and 2 ml. pipettes. After several trials it was realized that there was no advan tage over the 1 ml. pipette, either in enhancement of resolution or in any other aspect of separation of the phosphorylated compounds. All of these experiments were conducted with the resin Dowex-1 chloride 200-400 mesh, 8x. When the smaller diameter resin Aminex A-25 became necessary to alter completely the type of column used since high pressures were necessary to maintain a reasonable flow rate. Columns with appropriate pressure fittings were bought commercially from Chromatronix (Berkeley, Calif.). Within the limits of resolution required, column lengths were limited by the increasing back pressure generated with increasing length and, to a certain extent, by the high cost of the resin. 47 Columns of 0.3 cm. internal diameter and 10 cm. length were used initially with excellent results in terms of resolution. Overall sensitivity was also enhanced since eluting volumes were decreased from 2 liters to 200 ml. Shorter columns, 6 cm. in length with the same internal diameter, did not decrease resolution, and they were adopted for further experiments. This column has been found to be most satisfactory. A few experiments were performed with a waterjacket around the column to determine the effect of temperature on resolu tion, but no advantage was observed with column tempera tures up to 60°C. The column standardized for routine use in the system is 0.3 cm. in internal diameter and 6 cm. long. It is obtained commercially from Chromatronix (Berkeley, Calif.). It is fitted with a sample injector at the top which allows for sample loading without interrupting the run. The column is easily loaded, since all fittings are made of plastic and easily disassembled by hand. Pressures up to 1,000 pounds per square inch can be handled by this column although normal working pressures are approximately 200 pounds per square inch. 3. Gradients; Separation of compounds by ion exchange chromato graphy depends on differences in affinity and strength of 48 these compounds for sites on the column. The strength of this binding is affected by steric factors and particularly by changes in pH, which determines the ionization of these compounds, and by ionic strength. Generally, sequential elution of compounds can be effected by a gradient changing in ionic strength, pH, or both. In the present study separation of monophosphates from diphosphates and polyphosphates proved to be rela tively simple. However, separation of the different monophosphates of interest (such as G-l-P, Gal-l-P, G-6-P, F-6-P and AMP) was a more complex problem. These compounds do not differ sufficiently in pK or in binding affinity to permit a good separation by ionic strength or by pH alone. Use was made of the fact that mixtures of neutral sugars have been separated in ion exchange liquid chromatography as negatively charged complexes with borate ions (32). These complexes, which differ in configuration depending on the availability of cis hydroxyl groups on the sugars, are sufficiently distinct to permit separation by ion exchange. Similarly, mixtures of sugar monophosphates have been separated by first forming their corresponding borate complexes and then eluting with an increasing gradient of ammonium 49 borate (38). Under these conditions, however, other compounds, such as nucleotides and nucleotide sugars, are not eluted since the high pH of the eluting gradient maintains the phosphate esters doubly ionized and hence more strongly bound to the anion exchanger. Khym and Cohn (2 8) separated by ion exchange not only the various sugar monophosphates but also other compounds such as ADP, ATP and nucleotide sugars. They employed a stepwise elution technique, decreasing in borate concentration from IO-^m to 10-^M, followed by a stepwise increase in NH4CI and HC1 concentration. In the present system a segmented continuous gradient was designed, patterned after the combination of Khym and Cohn. Some variations were necessary to adapt it to an automated system. The stepwise gradient used by these workers to change the borate concentration did not lend itself to automation. The effect of pH, ionic strength and borate concen tration was determined to arrive at an optimum combina tion of these three variables. a. Effect of pH: The separation of the phosphorylated compounds proved to be most sensitive to alterations in pH. As shown in Figures 4A and 4B, an increase in starting pH from 8.5 to 9.6, but with a final pH at 5.6 and with the I IlM .h -P /'l'- .A B P PVR Figure 4A. Chromatogram of standard compounds Initial pH = 8.5 Final pH = 5.6 \ o - 7 APP PYR Figure 4B. Chromatogram of standard compounds Initial pH = 9.6 Final pH = 5.6 p H GRADIENT G R A D IE N T 51 ammonium chloride and borate concentrations unchanged, had the effect of providing a much better separation of monophosphates. However, separation of other compounds such as diphosphates and adenine nucleotides was sacrificed to a certain extent, perhaps due in part to the increased availability of borate ion at the higher pH, since the pK of boric acid is 9.2. A stronger complex is formed at higher borate concentrations, as noted by Khym and Zill (30), and binding to the column is enhanced. An additional factor may be increased ioniza tion of the phosphate groups on these compounds at higher pH which would also favor increased binding to the column. The effect of pH on elution of diphosphate and triphosphate compounds was also evident. Since raising the initial pH results in a steeper pH gradient, the polyphosphate compounds elute closer together, since ionization of the phosphate hydroxyls parallels the pH gradient. b. Effect of ammonium chloride concentration; Differences in ammonium chloride concentration determine the relative elution of monophosphates, diphos phates and triphosphate compounds. It has no effect, however, on the separation of monophosphates from each other. Figures 5A and 5B compare the chromatographic 52 6 -6 -P 0 . 1 APP U0P6 AMP Figure 5A. Chromatogram of standards. Initial NH4CI concentration = 0.03M NH4CI, Final = 0. 3M NH4CI. •3 as a ATP Figure 5B. Chromatogram of standards. Initial NH4CI concentration = 0.09M, Final = 0.3M NH4CI. M O L A R V T Y -N H4 CI SKAOIffAiT M O L A R IT Y -N K iO . G R A D IE N T 53 separation of standard compounds conducted at two different starting NH4CI concentrations. In both cases the starting pH was 8.5 and final pH was 5.6, with borate concentration the same in both cases. At a higher initial NH4CI concentration the inorganic phosphate and monophosphate peaks elute earlier in the chromatogram, but no differences in separation of the different mono phosphates is apparent between the two gradients. From these experiments it was learned that manipula tions of the gradient in terms of NH4CI concentration alone was not sufficient to achieve the desired separa tion of monophosphates. c. Effect of borate ion concentration; At an initial pH of 8.5 and final pH of 5.6, and with NH4CI gradients as above, no differences were observed in resolution, particularly of monophosphates, when the sodium borate concentration was decreasing or was maintained constant at 0.05M throughout the chromato graphic run. Even when borate was kept at 0.05M through out the gradient, it is probable that the effective borate ion decreased with decreasing pH in the gradient. d. Compromise gradient: From the studies of the effect of pH, NH4CI and borate concentration on the elution order of the various compounds, a compromise gradient was tailored to provide 54 maximum resolution of all compounds of interest. A gradient device was employed to create a concave gradient increasing in NH4CI concentration. A shallow NH4CI gradient is used to separate the monophosphates from the diphosphates and allow time for the separation of the different monophosphates complexed with borate. This portion of the gradient is followed by a steeper NH^Cl gradient to separate the remainder of the phos- phorylated compounds. The gradient device used is composed of a mixing vessel of 100 ml. capacity and 4 staggered side vessels of 25 ml. each. The mixing vessel contains 100 ml. of 0.04M NH^Cl, 0.05M sodium borate, adjusted to pH 8.9 with NH^OH. The upper 3 side vessels contain 0.2M NH4CI while the lowest side vessel, the last to be added to the mixing chamber, contains 0.33M NH4CI. The gra dient produced by this combination is diagrammed in Figure 6. A typical chromatogram of erythrocytes using this gradient is shown in Figure 7. e. Elution order of compounds: The anion exchange behavior of the compounds examined is, as expected, the resultant of the effect of the phosphate groups, bases, and the stability constant of the borate complexes. Steric hindrance of the phosphate group can make the borate complex less stable, 55 I i i i k I < < 0 5 : LACT 0.2 Y AOP * TP UOPH \ \ \ I o.\ Figure 6. Chromatogram of standards using a tailored gradient as shown. ATP AOP UDPM AMP Figure 7. Typical chromatogram of erythrocytes using the tailored gradient. 56 which is reflected in lower binding to the anion exchanger. This allows for resolution of members of the same series such as glucose-l-phosphate, galactose-1- phosphate, glucose-6-phosphate and fructose-6-phosphate. Aldoses elute ahead of ketoses, due to the participation of the OH group arising from the furanose ring formation in reactions with borate. Elution of the adenine nucleo tides AMP, ADP and ATP is found to correspond to pre dictions from considerations of the number and pKs of the phosphate groups. The nucleotides and the nucleotide sugars generally do not elute in as discrete a peak as hexose phosphates and diphosphates but present broader peaks on the chromatograms. At a flow rate of 0.9 ml. per minute the compounds are eluted from the column in less than 4 hours total time. Table I depicts the elution order of various compounds as they appear on the chromatogram. Elution order is expressed in minutes after the gradient has been started, bearing in mind that there is a 47-minute delay in the automated system between the time a compound emerges from the column and the time it is detected as phosphate by the colorimeter. Reproducibility of the elution time of compounds on successive runs is within + 3 minutes. This small deviation has been traced to dead volumes inherent in the gradient device, pump Table I Compound Glycerophosphate Glucose-l-P Inorganic Phosphate Galactose-l-P NAD Dihydroxyacetone Phosphate Glucose-6-P Fructose-6-P AMP UDPG UDPGal 3-PGA 2-PGA Elution Time (min.) 27 34 36 38 44 62 66 73 102 106 107 109 109 Compound Glucose-1, 6-diP Mannose-1,6-diP Fructose-1,6-diP 2,3-DPG ADP Elution Time (min.) 132 -134 153 172 183 UDPGA 6-Phosphogluconic Acid ATP 189 199 227 58 debubbler, pressure gauge and connecting tubing. 4. Gradient device; During the course of this study gradients of a variety of shapes and volumes have been employed. Initially, eluting volumes up to 2 liters were used. To provide flexibility in tailoring gradients, a gradient making device manufactured by the National Instrument Laboratories, Inc. (Rockville, Md.) was used (39). It consists of a cylindrical mixing vessel with a capacity of 900 ml. surrounded by 6 cylindrical side vessels of 150 ml. capacity each, connected to the bottom of the central mixing vessel and arranged in a staggered con figuration so the top of each vessel coincides with the bottom of the adjoining one. Different gradient profiles can be created by varying the buffer in each of the 6 side vessels. When smaller volumes were used in eluting the column only the lower two side vessels were employed. When the 6 era. x 0.3 cm. Aminex A-25 columns came into regular use in the system, the total elution volume became 200 ml. or less. A new gradient device was constructed, based on the design of the larger type, with 4 side vessels and with a maximum capacity of 200 ml. This was made from tubular lucite and fitted with cut-off plastic pipettes as shown in Figure 8. 59 o Figure 8. Gradient device. 60 The device had to be slightly modified from the original design since the small diameter of the plastic pipettes caused an unforeseen surface tension effect. A slight overlap of the bottom of the one side vessel with the top of the adjoining one created enough differ ence in hydrostatis pressure between vessels to oversome the surface tension in the plastic pipettes and to pro vide a smooth transition in the gradient. It was found that the most convenient way to fill the side vessels is to aspirate the requisite solution into each vessel separately from the mixing chamber, employing a tube connected to the top of the side vessel. The outflow from the vessel to the mixing chamber then is clamped shut. After all four of the side vessels have been filled in this manner, the mixing chamber is filled with starting buffer, the clamps are removed from the side vessel outflow tubes, and the device is ready for the chromatographic run. At various stages of the investigation, efforts were made to improve on the gradient-making device. A promising concept was the use of a shaped rod in the reservoir of a two-chamber gradient maker. The hypo thesis was that the relative volume of limiting buffer from the reservoir could be varied by decreasing or increasing the cross-sectional area of the reservoir by 61 the area of the rod from point to point. By shaping the rod, the relative volume of the flow from the reservoir could be controlled. In practical use the concept turned out to be less attractive. First, since the reservoir buffer necessarily was of one concentration only, flexibility in gradient formation was minimal. Second, at extreme volume differences the area of the rod approximated that of the reservoir chamber, and flow from the reservoir became irregular due to surface tension effects between the surfaces. C. Asher: Several modifications of the asher were required to obtain maximum resolution of compounds, more reproducible performance and less down time due to mechanical failure. The original mechanical shaker, designed to agitate the cups while the ashed material is being reconstituted, was removed due to difficulties in regulating the degree of agitation. Instead, the probe delivering the solution for reconstitution was placed higher above the cup, with the result that sufficient mixing occurred during delivery of reagent. The original rate of rotation of the turntable was doubled (from 60 seconds between cup changes to 30 seconds) by making a new version of the cam which activates the microswitch responsible for the turntable's 62 rotation. This was done both to reduce the volume per cup in order to avoid bumping during ashing and to improve resolution of compounds through smaller aliquots of the columns eluate being collected per cup. The water used to wash the cups between cycles is delivered to the asher from a 2 liter container located approximately 3 feet above the instrument. This con tainer is maintained at a constant level by pumping with a vibrator pump distilled water from several connected 5 gallon reservoirs located under the hood containing the asher. The wash water is removed by suction. The probes used for this purpose are connected to a series of trap bottles attached to a vacuum line. Originally this outflow went directly to the suction intake of a water pump, but this arrangement proved to be unreliable. To dispense the perchlorate nitrate mixture used in ashing, and also the 1% H2SO4 to dissolve the ashed material, two 1 ml. accupettes are employed, each mounted on a reservoir bottle having a 1 liter capacity. These automatic pipettes are calibrated to deliver the appropriate amount (0.05 ml. perchlorate nitrate or 0.35 ml. 1% H2SO4). They are attached with a spring to the arm raising the probes during rotation of the turntable. To preserve the life of these accupettes, a small plastic 63 sleeve was fitted on the metal ring which limits travel of the piston. The sleeve serves to dampen impact. D. Technicon system; Several precautions were necessary when assembling the Technicon manifold. Since the pump tubes are very narrow, it was necessary to stretch these tubes with a small surgical hemostat, prior to slipping them over the connecting nipples, to prevent splitting of the tubes. The whole manifold was constructed with a maximum of glass tubing and a minimum of plastic transmission tubing. This was found essential because the chromogens produced during the reaction adhered to and coated the plastic tubes but not the glass parts. Several different kinds of plastic tubes were tried, but all possessed this unfortunate property. The plastic pump tubes must be replaced every 300 hours of operation. To preserve the life of these tubes, water is pumped through at the conclusion of each run and they are removed from the pump and left unstretched. A variety of de-bubbler designs were tried, but the most effective proved to be the standard Technicon device. It was convenient to attach the de-bubbler to the body of the spectrophotometer as closely as possible to the flow-through cell in the light path compartment of the spectrophotometer. This proved to be an important point 64 in providing better resolution. After the air segments have been removed from the reagent stream, backmixing of the reagent stream can occur readily. E. Phosphate analysis; The Fiske and Subbarow phosphate analysis method (40) does not lend itself to methodology in which reagent stability of several hours is required, since the mixed reagents themselves develop a blue color on standing. Two other methods recommended by Technicon for single and multi-channel analyzers were tested. In one instance hydrazine sulfate is employed as the reductant (41), but this method was found to have the unfortunate feature of coating glass and plastic tubes exposed to the reagent stream with chromogens. The other method uses stannous chloride in concentrated HC1 as the reductant (42). This method was very sensitive but required that the reductant be kept at refrigerator temperatures to maintain its stability. The method finally selected is an adaptation of that of Brewer, et. al. (34). This procedure employs ammonium molybdate, with ascorbic acid as the reductant. It was found to be the most stable and the most sensitive of all of those tried. During the initial phases of development, the optical density of the reagent stream was monitored at a 65 wavelength of 620 ran. as suggested by Fiske and Subbarow (40). Bartlett described a modification designed to enhance the sensitivity of the method, consisting of heating the reagent mixture prior to measuring its absorbance at 820 mu. This improvement in sensitivity was incorporated in the present methodology. The heating bath temperature was increased from 45° to 70° and color measured at 820 nm. Sensitivity was enhanced by a factor of 2.5. This change not only made it possible to detect smaller levels of phosphorylated compounds (at no expense in resolution) but it also eliminated the unpleasant problem of reagent line coating by chromogens, since smaller amounts of chromogens were required for detection. Presently, the phosphate analyzer has a sensitivity at the nanomole level. When 1 nanomole of phosphate is added to a cup of the asher, an O.D. change of 0.06 is produced. The method gives a linear relationship of O.D. vs phosphate concentration up to a level of 15 nanomoles per asher cup, after which the linearity is lost. However, from the practical standpoint this loss in linearity is not serious since at the higher phosphate levels, the phosphate is spread over several cups, and the O.D. change is no longer represented by a spike on the chromatogram but by a triangle whose area, and not 66 height, becomes proportional to phosphate concentration. A fortunate circumstance for quantitation is the lack of baseline drift during the course of the chromatographic runs. In terms of peak area, proportionality with phos phate concentration is found with peaks of less than 0.8 O.D. at the apex. The limit of quantitation, therefore, becomes a function of the particular compound since different compounds differ in elution behavior. Inorgan ic phosphorus (P^), for instance, elutes as a very narrow peak and high O.D. changes are reached at much lower concentration than an equivalent amount of ATP, which elutes more slowly, spreading the peak over a larger portion of the chromatogram. F. Spectrophotometer and flow cell; The intensity of the color developed in the auto mated phosphate analysis is measured by a Bausch & Lomb Spectronic 70 spectrophotometer, an instrument which possesses a number of desirable features for the purpose. One very important consideration is that no apparent electronic drift could be noted during 30 hours of continuous operation. The cell holder compartment can accommodate a variety of shapes and sizes of flow cells. Flow cells were constructed out of square blocks of black plastic 67 of varying length through which a hole was drilled for the light path. The diameter of this hole was maintained as small as possible and was distated by the amount of light necessary to permit blanking the reagent baseline at 0 optical density. The light path was sealed on both ends with circles punched out of transparent plastic and glued into place. The reagent stream was delivered and withdrawn from the light path by two small orifices drilled through the top of the plastic to meet the light path. Several flow cells up to 5 cm. in length were constructed. All functioned adequately when runs of long duration were performed during the period when resolution still was governed primarily by the resin. However, when the Aminex A-25 resin was introduced, it was found that backmixing within these flow cells detracted from the resolution achieved by the resin. To overcome this problem a flow cell obtained from an SMA-12 Technicon auto analyzer was then used. This flow cell constructed entirely out of glass, with a light path of 1.8 cm. was found ideally suited to match the degree of resolution obtained from the remainder of the system. G. Metering pump: During development of the automated methodology and during much of the application of the system to 68 erythrocyte metabolism studies, a Beckman Accuflow pump was used as the metering pump to drive the gradient through the ion exchange resin. This piston type pump maintains an accurate flow rate independent of the back pressure generated by the resistance inherent in the resin. At various times in the development, this pump has been used with up to as much as 800 pounds per square inch of backpressure with no changes in flow rate. The pump is placed in the system between the gradient maker and the chromatographic column. It was found necessary to place in the line ahead of the pump a small bubble trap. This consists of a section of a glass tubing 2 cm. long and 0.8 cm. in diameter, tapered at both ends. The gradient flows in through the upper part of the bubble trap, which is taped in a vertical position to the pump housing, while the pump samples from the bottom of the trap. Any bubbles which reach this section will remain in the trap and are not intro duced into the pump chamber. This precaution was necessary since air trapped in the pump can alter the flow rate dramatically. A consistent finding with all the resins used has been the formation of a dark brown layer on the top surface of the column. A similar observation has been reported by other workers using this type of resin (43). 69 It has been attributed variously to resin instability at high pressures, contamination by silicates of the eluting buffer or traces of heavy metal. It was observed in the present study that the formation of the dark brown layer is enhanced by a high pH of the eluting buffer. When a filter composed of a short column of diatomaceous earth was interposed between the resin and the pump, the contaminant was effectively trapped, and the formation of the layer on the column was prevented. This suggested that it was not related to the resin but rather a product of the pump or of the gradient. The most likely candidate for a source of the material appears to be the flouri- nated hydrocarbon packing used around the piston of the pump. This packing, when placed in a beaker containing a base, slowly releases a brown colored material. The diatomaceous earth filter did not prove practical as a permanent means to remedy the problem. With con tinued use it becomes plugged and the pressure increases drastically. Experiments with regenerated columns containing the brown top layer suggested that the contaminant interfered with binding of compounds to the column, since the chromatograms obtained typically resemble those of a column which has been overloaded. With the Beckman pump in use for metering, the only means to circumvent the problem has been to use a new column 70 with fresh resin for every chromatographic separation. After development of the automated system had been substantially completed, it became possible to substitute the Beckman pump with one manufactured by Chromatronix, Berkeley, California (Model CMP-2 Cheminert Metering Pump). This is a three chamber, triple piston pump. The third pump chamber serves to pressurize each filled chamber before it is switched to the outlet, eliminating the reverse flow surge. The exposed parts are limited to glass, teflon and Kel-F. The pump is rated for 500 pounds per square inch, with a maximum flow rate of 2 ml. per minute. Since the Cheminert pump has been in use, the problem of the brown contaminant has been practically eliminated. A given resin column can be regenerated and used a number of times before discarding. H. Scintillation counter; The radioactivity in the compounds resolved by the column is monitored prior to ashing by a Unilux III liquid scintillation counter (Nuclear Chicago) fitted with a flow-through cell. It has proved to be a very useful addition to the automated phosphorus analyzer, since it permits estimation of flow in pathways of label from the substrate and also the determination of specific activities of the various metabolites. It also permits 71 monitoring of compounds separated by the column but not detected by the phosphorus analyzer, such as lactate, pyruvate and galactonate. Since radioactivity is detected very soon after the effluent emerges the column, a comparison of the peak spread in the radioactive tracing compared with that of the phosphorus analyzer provide a comparison of the relative loss in resolution experienced during ashing and subsequent phosphate determination. Commercial flow cells constructed of plastic and filled with anthracene were purchased from Nuclear Chicago during the early experimental development. These cells had a nominal capacity of one ml. and no critical backmixing was evident with these cells under the conditions of that time. However, when it became feasible to obtain equal resolution with runs of much shorter duration, it became apparent that some loss in resolution was occurring due to the flow cells. In addition, when working with extracts of skin fibroblasts and of white blood cells incubated with a labelled carbohydrate precursor, severe contamination of the commercial flow-through cells occurred early in the chromatographic separation. Attempts to remove this contaminant with different detergents, weak acid and basic solutions, were only partially successful. The 72 anthracene could not be replaced without destroying the cell, and replacement of the cells became a significant expense. A flow-through cell was constructed from a simple U shaped glass tube filled with anthracene (J.T. Baker). This tube has a much smaller diameter and therefore minimizes the problem of backmixing. Efficiency for carbon 14 with this new type of flow cell was reduced from an original 40% with the commercial cell to 33% with the presently used model, but this change was considered tolerable in light of the better resolution obtained. With this type of cell it is a simple matter to replace any contaminated anthracene with new, thereby avoiding the efforts devoted to attempting to remove a stubborn contaminant after a particular run. The output of the scintillator counter is connected to a Bio-Rad ratemeter which is connected to the recorder. Identification of a particular radioactive peak is made from the position of the peak on the chromatogram, while an accurate record of the radio activity of this peak is obtained from the ticker-tape print-out display of the scintillator counter. The radioactive trace is related to the phosphate chromatogram by allowance for the time interval of processing through the automated sequence. Calibration 73 of the interval for a given circumstance can be done early in the run by injecting a small sample of a labelled phosphorylated sugar into the stream immediately ahead of the counter, followed by comparing the radioactive peak with the phosphate peak finally appearing. GALACTOSE METABOLISM OF THE ERYTHROCYTE Erythrocytes share with other tissues the necessary enzyme complement for the metabolism of galactose. Their availability and easy manipulation have made them a choice tissue for in vivo and in vitro studies of galactose metabolism in health and disease states. It has been observed that erythrocytes are unable to maintain a steady state level of ATP when incubated with galactose as the only energy producing substrate. Other parameters of viability such as cation fluxes have also been found impaired. Several hypotheses have been advanced to account for the observations made, but none provide an integrated explanation in terms of galactose metabolism as a whole. Much of the work that has been done has been based on experimental methods which were acceptable at the time but which could not provide a full range of information. It is the purpose of the present study to utilize the automated system for analysis of phosphorylated intermediates in an examination of the problem of main tenance of ATP level in the erythrocyte subsisted on galactose. 74 75 General Background The non-nucleated mature human erythrocyte lacks DNA, RNA, and the ability to carry out protein synthesis. It has no mitochondria, Krebs cycle or cytochrome systems and consequently no capacity for oxidative phosphorylation. On a diet of sugar, vitamins and inorganic phosphorus it provides for itself sufficient energy to meet its own needs: to resist the accumulation of methemoglobin, prevent hemoglobin denaturation and Heinz body formation in the face of oxidative stresses and to actively transport cations against electrochemical gradients. The ready availability of erythrocytes has made them objects of studies not concerned with erythro cyte pathology alone but as models of cellular events occurring in tissues which share a metabolic defect with the red blood cells. The uniqueness of the mature erythrocyte as a cell has both aided and handicapped extrapolation of in vivo or in vitro findings to other human tissues. Normally in erythrocytes about 90% of the glucose is metabolized via anaerobic glycolysis by the Embden Meyerhoff pathway with the production of lactate (Figure 9). The remaining 10% is metabolized via the hexose monophosphate shunt. 76 GALACTOSE ATP ADP c GLUCOSE ■ ATP -ADP GAL-l-P + UDPG ^ UDPGAL + G-l-P c TPN TPNH G-6-P I P-6-P ATP r ^ ADP F,-l,6cLiP DHAP <.---> GAP - DPN GPNH 1,3-DPG ADP 2,3-DPG ATP Pi 3-PG 6-PG LACTONE 6-PGA TPN *TPNH co2 RU-5P 2-PG 1 PEP c ADP ATP PYRUVATE DPNH DPN =>■ LACTATE LDH Figure 9. Pathways of Erythrocyte Glycolysis 77 The contribution of shunt activity to the erythro cyte appears to be regulated by the availability of NADP for reducing the glucose-6-phosphate to 6-phospho- gluconic acid. Activity of the shunt provides reducing equivalents essential for the prevention of methemoglobin build-up in the erythrocyte and for protection of the hemoglobin molecule against environmental oxidative challenges. Erythrocytes must be able to balance energy con sumption with energy production. The only means that the erythrocyte has for the generation of ATP is through glycolysis via the Embden-Meyerhoff pathway. The maintenance of a favorable ATP balance is reflected in a high ATP to AMP ratio in the red cell at steady state conditions. Several means whereby control of energy balance could be accomplished have been suggested: (a) control at the membrane level, (b) through potentially rate-limiting enzymes subject to regulatory control, (c) by asynchronous glycolysis and temporary deviation from steady state kinetics and (d) through alternate pathways in glycolysis. In the range of an extracellular sugar concentration of 10-100 mg%, the ratio of intracellular to extra cellular glucose or galactose concentrations is 0.6 - 0.7 corresponding to the intracellular water phase which is 78 60-70% of the total volume (44). No effect of glucose upon an independent uptake of galactose has been reported (45). With glucose, the rate of entry exceeds the rate of utilization by several fold, and equilibrium is reached within seconds (46). Within the physiological range the internal glucose concentration is not a factor in regulating the glycolytic rate of the erythrocyte. Four steps in red cell glycolysis have been proposed as potentially rate limiting and subject to regulatory control: hexokinase, phosphofructokinase, glyceraldehyde- 3-phosphate and pyruvate kinase. Since hexokinase exhibits the lowest activity of the glycolytic enzymes in human erythrocytes (47)T it has been postulated that its activity modulates the overall rate of glycolysis (48). The effect of inorganic phosphate on increasing glucose consumption can be explained in part by its effect in overcoming the G-6-P inhibition of hexokinase and the ATP inhibition of phosphofructokinase (49). The phosphate induced accumulation of fructose-1,6,-diphos phate and of triose phosphates and deficit in lactate production (50)f has been interpreted as indicating a rate-limiting function of glyceraldehyde-3-P dehydro genase (51) . A change in the ATP yielding reactions without a coordinate change in ATP consuming reactions would bring 79 about a deviation from steady state kinetics. Such a condition is exemplified by triose phosphate isomerase- deficient erythrocytes which display a decreased ATP level and an accumulation of dihydroxyacetone phosphate. A non-steady state condition can become self-limiting, and control by asynchronous glycolysis is limited by the detrimental effects on viability during prolonged periods. A unique property of the human erythrocyte is the high concentration of 2,3-DPG (52). The synthesis and breakdown of 2,3-DPG was elucidated by Rapoport and Leubering (5 3) as an adjunct to the Embden-Meyerhoff pathway in the red cell (Figure 10). The liability of this compound during substrate deprivation suggests that it is not an inert by-product of glycolysis and that it serves as a reservoir for a potential source of energy to be yielded at the pyruvate kinase step. In effect, this alternate pathway serves in the erythrocyte as a control mechanism. In non-erythrocytic cells, 1,3-DPG is converted to 3-PG by phosphoglycerokinase: 1,3-DPG + ADP »- 3-PG + ADP The activity of this pathway depends on the availability of the substrates 1,3-DPG and ADP. In erythrocytes, 1.3-DPG can be converted also to 2,3-DPG by the enzyme 1.3-DPG mutase, which irreversibly catalyses this 80 1,3-DPG 3PG Cofactor ADP Inhibitor ATP 2,3 DPG Phosphatase 2,3-DPG 3-PG Inhibitor = 3PG Pi MPG Mutase Cofactor = 2,3-DPG V 2-PG 3-DPG Figure 10. The Rapoport-Leubering Shunt 81 dismutation reaction with the participation of 3-PG as a cofactor. This enzyme is sensitive to product inhibi tion and to a lesser extent to the availability of the cofactor 3-PG. The 2,3-DPG formed is hydrolytically broken down to 3-PG by a specific 2,3-DPG phosphatase. This enzyme, whose low activity in erythrocytes has hampered its clear characterization, is a unique phos phatase, since it is strongly inhibited by the end product 3-PG and relatively insensitive to inhibition by Pi. 1,3-DPG shunted through the Rapoport-Leubering pathway by-passes the ATP generating phosphoglycerokinase step. The activity of this shunt therefore serves as a means in the erythrocyte to modulate ATP production in glycolysis. Galactose Metabolism in Erythrocytes: In erythrocytes glucose, fructose and mannose are phosphorylated at the C-6 position. A single hexokinase appears to be responsible both for glucose and fructose phosphorylation, while mannose requires a separate kinase (54). In contrast, galactose is phosphorylated at the C-l position by a specific galactokinase, the first step in the principal pathway of galactose metabolism in mammalian cells (55). This sequence, the Leloir pathway (56), is identical to that for the metabolism of 82 galactose in yeast (55, 56, 57, 58). Galactose + ATP -------Gal-l-P + ADP (1) Gal-l-P + UDPG -------*• UDPgal + G-l-P (2) UDPGal — UDPG (3) Galactose + ADP -------G-l-P + ADP Sum The enzymes involved are (1) Galactokinase (E.C. 2.7.1.6), (2) Gal-l-P uridyl transferase (E.C. 2.7.7.12) (transferase) and (3) Uridine diphosphate galactose-4- epimerase (E.C. 5.1.3.2) (epimerase). The overall result is the conversion of galactose to glucose-1- phosphate. Upon conversion of G-l-P to G-6-P by phospho- glucomutase, the products derived from galactose merge with the mainstream of glycolysis. It is of particular interest that the Leloir pathway is present intact in erythrocytes, and the ready availability of this tissue has been an important factor in facilitating studies of galactose metabolism in man. Studies in galactose metabolism in the human were stimulated by efforts to understand the basis of the genetic disease galactosemia. This condition is characterized clinically by a failure to thrive, suscepti bility to infection, jaundice, hepatomegaly, amino aciduria, cataracts and mental retardation. Untreated affected children usually die young. Treatment, which is very effective, involves avoiding galactose in the diet, 83 by eliminating the lactose found in milk and milk products. Galactosemia is inherited as a Mendelian recessive condition. The defect is an absence of activity of Gal-l-P uridyl transferase, the second step in the sequence of the Leloir pathway. The defect is found not only in liver but also in the erythrocytes of affected individuals. Assay of erythrocyte transferase activity is useful adjunct to the diagnosis of the galactosemia homozygote and in detection of the heterozygote carrier state (59) . Variant forms of this enzyme have been demonstrated by gel electrophoresis in this laboratory and by other workers (60). Erythrocyte galactokinase and transferase have been partially purified and their kinetic properties have been examined (61). Both erythrocyte transferase and epimerase have been purified and studied in this laboratory (62). An inherited defect in galactokinase activity (63) and recently a deficiency of epimerase activity in red blood cells (64) have been described. Variation in utilization of galactose has been observed. In a study by Segal, et. al. (65) two out of 14 eight galactosemic patxents given a test dose of C galactose exhaled C-^C>2 at a substantial rate compared to normal controls. In vitro studies of galactose 84 metabolism of erythrocytes by Weinberg also demonstrated variations among galactosemic patients (66). Several alternate routes in addition to the Leloir pathway have been suggested (Figure 11). Isselbacher (66) proposed that Gal-l-P is converted in liver to UDPGal in the presence of UTP by an enzyme (UDPGal pyrophosphorylase). The existence of this enzyme could not be demonstrated in erythrocytes (67). Subsequent studies have shown that this reaction at best can account for disposal of only very minor amounts of Gal-l-P in the absence of transferase (68). Inouye and Hsia (69) reported that high concentrations of galactose incubated with galactosemic erythrocytes yielded compounds identi fied as Gal-6-P and Gal-1-,6-diphosphate, but no evidence for this pathway has been published by other investiga tors . An oxidative pathway of galactose to galactonic acid on the way to CO2 and xylulose has been postulated by Segal and Cuatrecasas (70). Recently, a galactonate in substantial amount was isolated from the urine of a galactosemic patient and normal individuals after oral loading with galactose (71), but the significance of this pathway in the face of more usual galactose intake is not known. 85 GALACTONATE GALACTOSE-----=>• GALACTITOL co2 GAL-6-P ATP ‘ADP GAL-l-P £ UDPG UTP Pi * G-l-P G-6-P UDPGAl CO < ( LACTATE Figure 11. Pathways of galactose metabolism. Enzymes of principal pathways are designated: G = galactokinase, T = transferase and E = epimerase. 86 Conversion of galactose to galactitol catalyzed by aldol reductase has been described in animal tissues (72) including erythrocytes (73) . This pathway clearly does account for some galactose disposal by galactosemic patients. The accumulation of Gal-l-P in the tissues of galactosemia patients is the basis for a hypothesis that this compound is in some manner responsible for the severe clinical manifestations of untreated galactosemia. Inhibition of phosphoglucomutase by Gal-l-P was shown by Sidbury in hemolysates (74). One suggestion has been that the toxic manifestations of the disease are a consequence of this inhibition. Background of the Problem Previous investigations by Pennington and Prankerd (75), by Zipursky, et. al. (76), by Oski (77) and by Hill (78) have contributed usefully but have not resulted in a unifying explanation for the effect of galactose on cell viability. Their studies were per formed either on red cell hemolysates or, in work with intact cells, at galactose concentrations greatly exceeding physiological levels. None of these earlier experiments were time course studies. Although the procedures used were satisfactory for the limited approach in each study, the earlier workers did not have 87 available the refinement in methodology available for the present investigation. The rate of galactose metabolism in red cells has been found to be about one tenth that of glucose (62, 79, 80, 81). The relatively low rate of galactose metabolism in erythrocytes, as compared to that of glucose, was first described by Schwarz, et. al. (79) in terms of a decreased oxygen consumption when these cells were incubated with galactose. From another standpoint, Beutler and Collins (80) studied the effectiveness of galactose as a substrate for methemoglobin reduction in erythrocytes. Methemoglobin reduction was markedly accelerated by methylene blue when glucose was the substrate but was only minimally increased with galactose as substrate. This observation was explained by Oski and Rose in terms of a "slow" galactose metabolism (77). The work of Oski and Rose was directed at demonstra ting that anomerization is not rate limiting in the human erythrocyte, an explanation advanced by Beutler and Collins for the apparent lack of hexosemonophosphate shunt stimulation when galactose is used as a substrate to induce methemoglobin reduction. Beutler and Collins postulated phosphorylation to tagatose-1,6-diphosphate, with result of bypassing of G-6-P and hence the shunt. Oski and Rose demonstrated indirectly that galactose does indeed reach G-6-P, that the G-6-P derived from galactose is metabolized via the pentose phosphate pathway and the glycolytic pathway in the same propor tions as is the G-6-P derived from glucose. They calculated, on the basis of indirect evidence, that the red cells accumulated products of galactose metabolism prior to G-6-P formation (presumably Gal-l-P and/or UDPGal) to the extent of 0.2 umoles per ml. of cells in the steady state. On this basis they concluded, reinforced by a previous finding of Gal-l-P by Ng, et. al. (62), that movement through the galactose pathway is "slow." Pennington and Prankerd (75) used radioactive P^2 in their study of erythrocyte phosphate ester metabolism during galactose exposure in vitro. They demonstrated a decreased ATP synthesis and/or increased degradation of phosphate esters to Pi. They also found in some experi ments that diphosphoglycerate was reduced when galactose alone was provided to the cells. However, this was a highly variable finding, with increases noted in some cases. On the basis of knowing that Gal-l-P accumulates in galactosemia erythrocytes, these authors attributed the fall in ATP to phosphorylation of Gal-l-P and its accumulation. They extrapolated this concept to normal red cells without experimental evidence that Gal-l-P 89 accumulates in these cells. Zipursky, et. al. (76) in a limited experiment with galactosemia erythrocytes demonstrated that incubation with glucose before and during galactose exposure prevented the decline in ATP and 2,3-DPG. Furthermore, under their conditions, galactose did not seem to affect the incorporation of P ^ into the pool of ATP and 2,3-DPG. The present study has been directed at a re-examina tion of why the erythrocyte cannot subsist on galactose alone. It was hoped that application of the automated system for analysis of metabolic intermediates would provide the capability of obtaining information in greater depth than has been available to earlier investigators. 90 Experimental Work Saturating galactose concentration: In order to proceed with the experiments planned for the study, it was necessary to determine the saturating concentration of galactose for the erythro cyte. Earlier estimation in this laboratory, by measuring evolution of radioactive carbon dioxide upon incubating red cells with labelled galactose, may have reflected only the condition necessary for intracellular G-6-P saturation of the hexosemonophosphate shunt. During the development of the automated analysis for phosphorylated intermediates, exemplar experiments employed erythrocytes incubated under various conditions. The data suggested that a reasonable saturating concen tration of galactose is 10 mgm%. An experiment was designed to test this assumption. One ml. of packed washed erythrocytes from a normal individual were placed in each of two 15 ml. test tubes. To each tube was added 1 ml. of pH 7.4 Krebs-Ringer bicarbonate buffer which contained galactose. The amount of galactose was adjusted to bring the final galactose concentration to 10 mg% in the mixture in the first tube and to 20 mg% in the second. The tubes were gassed with a mixture of 95% oxygen and 5% CO2/ stoppered 91 and placed in a Dubnoff metabolic shaker at 37°C for 30 minutes. At the end of this pre-incubation period, 2 0 microliters of galactose-l-C^^ (250 uc/ml., 15 mc/mM) were added to tube 1 and 40 lambda to tube 2 to bring the final galactose concentrations to 13.1 mg% and 26.2 mg%, respectively. Incubation was continued for two hours after the addition of the radioactive galactose. Then 1 ml. of incubation mixture was removed from each tube and centrifuged. The phosphorylated metabolic intermediates were extracted from the cell pellets, and the distribution of radioactivity into intermediates was determined by the automated phosphorus analysis system. The relatively low specific activity of galactose in this experiment required application of a relatively large sample to the column, sufficient to permit detection of radioactive intermediates but yielding unlabelled components in excess of the range of the phosphate analyzer. The distribution of radioactivity among intermediates of the galactose pathway, corrected for background, is shown in Table II. No discrete peak corresponding to G-6-P could be detected, probably a reflection of a low steady state concentration of this intermediate under the conditions. A peak corresponding to galactonic acid was 92 Table II Distribution of Radioactivity in Intermediates after Incubation of Erythrocytes with Labelled Galactose at Two Different Concentrations CPM Found in 0.5 ml RBC after 2 hours Incubation with Galactose Concentrations of: Intermediate 13.1 mg% 26.2 mg% Galactonate 2,710 3,160 Lactate 4,380 4,840 Pyruvate 1,250 1,390 Unknown 0 350 Gal-l-P 21,660 17,550 UDPHexose 7,480 5,660 G-l,6-diP 24,370 28,410 G-6-P 0 0 2,3-DPG 46,250 34,880 TOTAL 108,100 97,240 93 evident in both sets of incubations, supporting the concept of an alternate pathway of galactose disposal via a galactose dehydrogenase (70). The peak labelled "Unknown” could not be identified by its elution position. It was shown not to be a radioactive contaminant of the substrate. The results show that doubling the galactose concen tration does not increase incorporation. It is assumed that in continuing experiments a total final concentra tion of galactose of 13.1 mg% is sufficient to maintain the galactose pathway saturated. Effects of galactose on glucose metabolism: Galactose merges with the mainstream of glycolysis at the G-6-P level. It is possible that in the erythro cyte the metabolism of glucose would be affected by the simultaneous utilization of galactose. It is the purpose of the present experiment to examine the distribution of a carbon radioactively labelled in glucose into inter mediates, in parallel incubations of erythrocytes done with and without galactose present. The experiment was carried out in the same general fashion as that described in the preceding study of the saturating concentration of galactose. The incubation was arranged to provide a concentration of 50 mg% of glucose in one vessel and 50 mg% glucose plus 20 mg% 94 galactose in the second. After 30 minutes of pre incubation at 37°C, 50 microliters of glucose-l-C^ (50 uc/0.5 ml., 57 mc/mM) were added to each vessel, and the incubation was continued for another 60 minutes. One ml. was removed from each tube and spun down to remove the supernatant KRB (Krebs-Ringer bicarbonate) and the metabolic intermediates were extracted from the cells. Table III shows the distribution of radioactivity in the intermediary compounds. No significant difference in total radioactivity incorporated from labelled glucose is apparent in the presence or absence of galactose. The counts found in G-6-P are closely similar under both conditions. In a previous experiment (Table II), with the carbon label in galactose, and glucose not present, the counts in G-6-P were to small to be detected. These findings support the concept that the rate at which galactose flows through the Leloir pathway is slow in relation to the rate of turnover of the G-6-P pool. In this experiment radioactivity from glucose was found both in G-l,6-diP and in UDPHexose. This indicates that G-6-P was converted to G-l-P, although, since the phosphoglucomutase equilibrium favors G-6-P formation, the concentration of G-l-P was below the sensitivity of detection. However, the substantial labelling of the Table III Distribution of Radioactivity in Intermediates after Incubation of Erythrocytes with Labelled Glucose in the Absence and Presence of Galactose CPM Found in 0.5 ml RBC after 1 hour Incubation with Radioactive Glucose Intermediate With Glucose Alone Glucose Plus Galactose Lactate 7,440 8,200 Pyruvate 820 1,060 Unknown 520 960 G-6-P 1,560 1,490 UDPHexose 2,120 1,220 G-l,6-diP 6,400 4,210 F-l,6-diP 1,910 1,250 2,3-DPG 31,730 31,710 TOTAL 52,500 50,100 96 obligate cofactor of the phosphoglucomutase reaction (G-l,6-diP) serves as a means to assess distribution into this pathway. A similar observation was made by Bartlett in 10-minute incubations of red cells with labelled glucose (82). Because of the limited amounts of UTP considered to be present in the red cell, and hence a restriction on UDPG pyrophosphorylase activity, the counts in UDPH (presumably UDPG) were unexpected. No explanation was evident in the data of the experiment. In comparing the radioactive counts in G-l,6-diP (and in UDPH) between the conditions of presence or absence of galactose, the number of counts with galactose present could be considered to be substantially less. This is taken to reflect the expected dilution by the unlabelled intermediate resulting from metabolism of galactose. The results obtained represent the distribution among the intermediates at the point of time at which the incubation was stopped. Of particular interest is the flow from glucose into 2,3-DPG. The large accumula tion of counts is in accord with the concept of this intermediate serving as a reservoir for the glycolytic pathway. 97 Time course studies of galactose metabolism; The primary purpose of these experiments was to observe changes in ATP concentration during the course of a two-hour in, vitro incubation of red cells with (a) only galactose as the substrate, or (b) with galactose plus glucose. Production of lactic acid from galactose also was studied. The distribution of radioactivity from galactose into intermediary metabolites was determined, and changes in concentration of these intermediates were measured. The results presented are of an experiment carried out under the final methodolo gical conditions, but the findings are representative of those of a number of other experiments done at various times during the development of the methodology employed. Erythrocytes were obtained from a normal individual. After the removal of plasma, together with the layer of leucocytes and platelets, the packed red cells were divided in two parts. For the experiments conducted in the absence of glucose the red cells were washed in isotonic saline. Erythrocytes to be incubated with glucose and galactose were washed in isotonic saline which was made up to be 100 mgs% in glucose. Incubation and sampling procedures were as described previously. One set of samples was incubated with Krebs-Ringer bicarbonate buffer at pH 7.4 made up to be 100 mg% in 98 glucose, while a similar set was incubated with KRB buffer alone. After 30 minutes of preincubation in buffer or in buffer with glucose, radioactive galactose (20 microliters of 250 uc/ml, 15 mC/mM) and unlabelled galactose was added to bring the final galactose concen tration to 13.1 mgm% in each incubation vessel. Aliquots were removed from each incubation at 0, 5, 30, 60, 90 and 120 minutes after the addition of galactose. The phosphorylated intermediates were extracted from the cells in each aliquot, and were separated and quantitated by the automated procedure. The distribution of radio activity among the various intermediates was determined in the process. The results of the analysis of the phosphorylated intermediates during the time course experiment are presented in Figure 12. The time course of incorporation of radioactivity from C-^ galactose into the intermediary pools and lactate in the presence or absence of glucose is shown in Figure 13. Since galactose was used as the radioactive tracer, ATP was not labelled, and changes in ATP are to be seen from inspection of the data obtained from the phosphate analyzer (Figure 12). There is a steady increase in the level of ATP when red cells are metabo lizing only galactose. When glucose is present in the <&Ai.ACTO€>£ 3 0 CO G a l a c r o s e *■ g l u c o s e 3-0 a t p Co T/MB - mtslu TEO Figure 12. Phosphorylated intermediates during time course incubation of normal erythrocytes. NORMAL ■& GLUCOSE so 6 0 T/ME - N1/NUTS& 30 normal - glucose 3 0 Zo s 60 3o TiMff - A / t tfj OTES 100 Figure 13. Incorporation of radioactivity of Cl« galactose into intermediary pools. 101 incubation medium in addition to galactose, the decrease in ATP is prevented, and the ATP level remains sub stantially unchanged during the two hours of incubation. The decline in ATP was further examined in a separate experiment in which erythrocytes were incubated with galactose alone over a 4-hour-period. The results are shown in Figure 14. ATP continued to decrease in the absence of glucose and reached a very low level (20% of the starting value) after the 4 hours. The concentration of 2,3-DPG also decreases during the incubation period if galactose is the only carbohy drate source. With the addition of glucose, this decrease is prevented, and the size of this pool remains unchanged over the two-hour-period. In a separate experi ment in the absence of glucose, the concentration continued to fall over a four-hour-period (Figure 14). A comparison of the changes in 2,3-DPG and in ATP in the absence of glucose shows that the rate of 2,3-DPG decrease is greater than that of ATP (Figure 15). The rate of labelling of 2,3-DPG, in conjunction with the changes in pool size, permits a comparison of the relative specific activity of this intermediate in the presence or absence of glucose in the incubation medium. Figure 13 shows that there is a linear increase in radioactivity of this pool, both in the presence and 102 3 % I ATP TIME OP IP /C U B A T IO N - H O U H& Figure 14. Pattern of phosphorylated intermediates of erythrocytes incubated with galactose for 4 hours. 103 z.o /• o 3 o Time op mcu&A7io*/-pun Figure 15A. Changes in 2,3-DPG during the incubation in the presence and absence of glucose. 1 . 0 B o 6 0 t / m m o p I p ic u o a p t io p i ~ m /a/. Figure 15B. Changes in ATP during the incubation in the presence and absence of glucose. 104 in the absence of glucose. This rate remains constant throughout the two hours of the incubation, suggesting that even two hours is not sufficient for saturation. The presence of glucose seems to have little effect on 14 the rate of labelling of this pool from C galactose. However, since the concentration of 2,3-DPG in the absence of glucose is not constant but decreases, the specific activity of this compound, in the absence of glucose, rises much faster than when glucose is present. A speculation concerning the observations on 2,3-DPG, but one which is difficult to test experimentally, is that the results may reflect distribution between two pools of this intermediate. The 2,3-DPG bound to hemoglobin, as shown by Benesh and Benesh (83) could be thought of as an inactive pool, but in equilibrium with a small active pool such as the 2,3-DPG which serves as a cofactor at the phosphoglyceromutase step. Radioactive label then could move forward toward lactic acid on the one hand and, on the other, into the large pool via the Rapoport-Leubering shunt. Since 2,3-DPG is a very large component, it is not surprising that saturation of the hypothetical large pool is not achieved in the two-hour period. In considering the data on 2,3-DPG, the assumption has been made that the peak on the ion-exchange 105 chromatogram, and the corresponding radioactive peak, represent only 2,3-DPG. It is possible that the peak masks some other compound, perhaps an intermediate derived from galactose by some other pathway than that through G-6-P. To check this assumption, an experiment for the purpose was done. This work will be described separately. An experiment was done to measure the total lactate produced in an incubation, both that present in the cells and that in the medium. The results are shown in Figure 16. It is evident that the production of lactate from galactose alone is less than when glucose has been added to the medium. If the appearance of lactate were to be accepted as the criterion, this finding would support the view that galactose is metabolized at a slower rate than glucose. However, an inspection of the labelling pattern of lactate in the presence or absence of glucose (Figure 13) shows that more radioactivity from labelled galactose in the presence of glucose than is found in lactate in the absence of glucose. This data, in contrast to the observations on total lactate production, represents what is present in the cell at a given point in time. On superficial consideration, since the source of the carbon label is galactose, the data suggests that the presence of glucose in the medium somehow forces 106 yUvnoJis _ t - x c n r r 9 .0 o -30 * * £ M c u t j i [ n o v —*| T t f iO G O P /A tC U & A T /O * /- Figure 16. Lactate production of erythrocytes metabolizing galactose + glucose. 107 increased use of galactose. A possible explanation for this apparent inconsistency may be that in the absence of glucose some proportion of lactate is derived from 2.3-DPG as the total amount of 2,3-DPG decreases. Lactate from this source would be expected to have a lower specific activity than lactate derived from other triose phosphates, because of the large size of the 2.3-DPG pool. On this basis, in the presence of glucose, more radioactivity derived from galactose will appear in lactate than when glucose is absent. These considera tions indicate that care must be used in arriving at conclusions on metabolic events from lactate analysis alone. Figure 12 shows that the concentration of Gal-l-P both in the presence and in the absence of glucose rises steadily during the initial 30 minutes of incubation and then remains constant at this level throughout the rest of the incubation period. It was concluded from this observation that phosphorylation of galactose is not the rate-limiting step in the Leloir pathway. Further, the suggestion of Penington and Prankerd that accumula tion of Gal-l-P, and sequestration of phosphate from high energy phosphate, could explain their observed decrease in ATP is not substantiated by the ATP findings in the present experiments. 108 In erythrocytes, the concentration of UDPH (UDPG and UDPGal) is too small to permit accurate measurement of the total pool size. Incorporation of radioactivity (Figure 13) reaches a plateau, both in the presence and absence of glucose, after an initial rise during the first 30 minutes of incubation. This finding substan tiates the concept that erythrocytes possess only a limited capacity to manufacture UDPG (because of lack of UTP for the UDPG pyrophosphorylase reaction). An increase in this UDPG would reveal itself by a concomitant rise in counts in the UDPG pool, since the combined activity of transferase and epimerase would distribute the label between UDPGal and UDPG. It cannot be determined from the present data on the UDPH pool if the observed effect of glucose on the Leloir pathway is at the epimerase step. An inhibition of epimerase activity by a glucose metabolite might decrease the concentration of UDPG below the level required for optimal activity of the transferase enzyme. The concentration of G-l,6-diP remained constant throughout the two hours of incubation. No differences in concentration of this intermediate were apparent in the presence or absence of glucose. Without glucose, the radioactivity in G-l,6-diP rose steadily, particularly during the first 30 minutes of incubation (Figure 13), 109 and a steady increase in counts during the incubation suggested that even after two hours this pool had not reached a steady state. With glucose added to the galactose incubation medium, increase in specific activity was observed only during the first 30 minutes of incubation after which the incorporation of counts leveled off and a constant specific activity was main tained. In addition, with glucose present, the specific activity of this pool was consistently lower than in the absence of glucose. Labelling of G-l,6-diP occurs at the phosphogluco- mutase step, since this compound serves as a cofactor for the conversion of G-l-P to G-6-P by the following reversible mechanism: G-l-P + Enzyme-P04 -=--= » • Enzyme-OH + G-l, 6-diP G-l,6-diP + Enzyme-OH Enzyme-P04 + G-6-P The enzyme favors G-6-P formation under physiolo gical conditions by a factor of 19, accounting for the concentration of G-l-P being below the capacility of detection by the methods employed. Counts appearing in G-l,6-diP from a radioactive galattose precursor must necessarily be derived from G-l-P produced at the transferase step. In the present experiment, the inability to reach a constant specific activity in this pool in the absence of glucose probably reflects the 110 large size of this pool, together with a relatively slow rate of metabolism of galactose. The decreased rate of labelling of this pool seen in the experiment with glucose present, and the plateau in specific activity, likely represents dilution of G-l,6-diP arising from G-6-P derived from glucose. That such dilution by glucose occurs even in the presence of galactose is evident from the preliminary studies with radioactive glucose and cold galactose (see Table III) in which radioactivity from glucose was found in G-l,6-diP. An important point in the present study is that a peak for G-6-P could be found on the chromatogram of the automated system only in the incubations done in the presence of glucose. When only galactose was in the medium, no rise above baseline in the G-6-P area could be observed. This is a very clear indication that the erythrocyte utilizes glucose at a much higher rate than it does galactose. Further evidence in these experiments of the lesser ability of the red cell to utilize galactose, as compared to glucose, is the observation that any radioactivity from galactose appearing in G-6-P escaped detection by the scintillator counter whether or not glucose was present in the medium. It was clear, nevertheless, that galactose is traversing this intermediate pool at some Ill definite rate, since after 5 minutes of incubation label already had appeared in 2,3-DPG and in lactate. However, the inability to visualize G-6-P, either labelled or unlabelled, as derived from galactose left open the possibility of involvement of some other pathway which constitutes a substantial by-pass of G-6-P. An experi ment with galactose of high specific activity was done to clarify this point, and labelled G-6-P was found. This work is described separately. The primary purpose of these experiments was to observe changes in ATP concentration during incubation of erythrocytes with galactose, both in the presence and the absence of glucose. The data obtained shows that galactose by itself could not sustain the intra cellular ATP level. It is felt that the inability of the Leloir pathway to adequately provide G-6-P to feed the glycolytic pathway is a likely explanation. In the presence of glucose, galactose continues to be metabo lized, but ATP does not fall. It is clear from these experiments, in agreement with previous work, that the erythrocyte utilizes glucose at a substantially higher rate than it does galactose. Initial phosphorylation is not limiting, and consequently the impediment in the utilization of galactose must be in the Leloir pathway. The amount of 112 galactose handled through the Leloir pathway in a given time, and reaching G-6-P, is markedly less than the amount of glucose that the red cell can phosphorylate to G-6-P in the same period of time. It is in this sense that one can speak of a "slow metabolism" of galactose. Studies with erythrocytes from galactosemia: Experiments were carried out with erythrocytes from individuals known to have reduced transferase activity. The samples were from a patient homozygous for galacto semia, with essentially no transferase activity, and a heterozygous individual with half normal activity. It was hoped that the data to be obtained might be of aid in interpreting limitations in galactose handling by normal red cells. Washed erythrocytes from a galactosemia homozygote were incubated under conditions identical to those in the experiments previously described. The sample was insufficient for a time course study. After 90 minutes of incubation with labelled galactose, both in the presence or absence of glucose, the labelled intermediates were studied. The results are shown in Table IV. The large accumulation of Gal-l-P in these cells was expected because of the known enzyma tic deficiency. Residual transferase activity may Table IV Compound_________________ -glucose________ +glucose Gal-l-P 8490 11,230 G-l,6-diP 500 310 2,3-DPG 210 120 Incubation of galactosemic erythrocytes with labelled galactose for 90 minutes in the presence or absence of glucose. 114 account for the labelling of G-l,6-diP and 2,3-DPG. No other intermediates were detected when galactose was the only nutrient for the cells. When glucose also was present, the number and distribution of intermediates was similar to that found in the experiments previously described. Beyond these points, it is doubtful that further interpretation of this single experiment would be appropriate. Erythrocytes of a heterozygous individual were incubated in a time course experiment with radioactive galactose, with and without glucose. Figure 17 shows the effect of glucose on the level of phosphorylated intermediates, while Figure 18 depicts the flow of radioactivity into the different intermediary pools. Distribution of radioactivity (Figure 18) shows that Gal-l-P accumulates at a greater rate in the cells from a heterozygote than in those from a normal individual and continues to build up throughout the incubation period. This is in contrast to the apparent steady condition reached by 30 minutes for a normal individual (Figure 13). This is taken as a reflection of the reduced transferase activity in the Leloir pathway. Radioactivity in Gal-l-P in the heterozygote cells continued to rise steadily at a greater rate in the presence of glucose than in its absence. This is taken 3 .0 2.0 3 0 60 T/A7£~ M/NOT££ 115 <aAL4C70^£ • / • (aUUCO^E 3 0 ATP i - Q 3 0 T/Wjr- M 'MOTES Figure 17. Changes in phosphorylated intermediates of erythrocytes heterozygous for Gal-l-P uridyl transferase. HETEfio-ZVGOTE <5AL * < S LU . 30 >o UDPH 60 T / M E - M iN U T E S 3 0 HETEKOZVGOTE G A L — G L t J , 30 20 *0 S 30 6 0 9 0 /Z O T/ME - M//JUTEE 116 Figure 18. Incorporation of galactose into intermediary compounds in RBC ' s heterozygous for galactosemia. 117 to indicate that the presence of glucose in some way restricts activity in the Leloir pathway, presumably at the transferase level. For both G-l,6-diP and 2,3-DPG, adding glucose to the incubation resulted in a decreased accumulation of radioactivity. This was taken to indicate that if there is indeed an effect of glucose in some measure, restricting the rate of galactose metabolism in the erythrocyte, the effect must be in the Leloir pathway prior to G-l-P. Upon examining the quantitation of intermediates without consideration of radioactivity (Figure 17), the striking effect of the presence of glucose in maintaining ATP and 2,3-DPG levels is closely similar to that found in earlier experiments (Figure 15). The decline of ATP and of 2,3-DPG in the absence of glucose was at much the same rate for the heterozygote cells as found for normal cells. Identification of G-6-P: In the experiments which have described no signifi cant radioactivity was found in the G-6-P area of the chromatogram following incubation of erythrocytes with labelled galactose. The inability to visualize G-6-P left open the possibility of involvement of some other pathway bypassing this central point. A more likely 118 possibility is that this reflects a very low steady state and rapid flux through this intermediate and not an alternate pathway of galactose metabolism. An experiment was performed in which galactose of high specific activity was used. Erythrocytes were incubated for 10 minutes with 20 microliters of galactose- l-C-1-4 (250 uc/ml., 15 mc/mM) with no carrier galactose added. A TCA extract was made and analyzed with the automated ion exchange column system. A clearly identifiable peak of radioactivity corresponding to G-6-P was found in the chromatogram. To confirm the identity of the peak, a TCA extract was prepared from red cells which had been incubated with radioactive glucose. Previous analysis had shown such an extract to have a sizable radioactive G-6-P peak. This extract was mixed with that from the galactose experiment. Upon chrmatography, a single discrete peak was seen on the chromatogram, and the counts from the two experiments were additive. Identity of 2,3-DPG; In an earlier discussion of the significance of 2,3-DPG in the experiments done, it was noted that it was uncertain if the large peak for 2,3-DPG represented only that compound or if it included others. 119 In an experiment designed to confirm the identity of 2,3-DPG, red cells were incubated with radioactive galactose. The radioactive fraction corresponding to the 2,3-DPG peak on the chromatogram was collected from the system prior to ashing. It was concentrated and re-chromatographed by thin layer chromatography. The volume of the fraction was decreased under vacuum by rotary evaporation at 45°C. It was supplemented with cold ADP as a marker for the thin layer. ADP was chosen because on the liquid ion exchange system it overlaps the 2,3-DPG peak. Therefore, it could serve as a useful means to assess if better resolution of these compounds can be achieved with thin layer. The mixture was applied as a streak on Brinkman MN Polygram Cel 400, 0.1 mm. thick thin layer plates. The plates were chromatographed with anhydrous methanol to remove the salts present in the sample. After drying, the plates were re-chromatographed with a mixture of 120 parts n-propanol, 52 parts concentration NH^OH and 2 8 parts water. A band corresponding to ADP was identified with ultraviolet (254 nm). No other ultraviolet absorbing areas could be detected. The plate was cut in half and one half was assayed for radioactivity by cutting narrow bands, placing these in liquid scintillation vials with 120 toluene and counting. The radioactivity was confined to a single band on the thin layer plate. The other half of the plate was similarly cut into small bands, placed in test tubes, boiled in acid and assayed for phosphorus with the phosphomolybdate reaction. Only two widely separated areas on the thin layer plate contained phosphate. One of these corresponded to ADP, since it coincided with the ultraviolet absorbing area. The other band coincided with the area containing the radioactivity. Since no other radioactivity or phosphate containing areas were detected on the plate, it was assumed that a single compound was present. Since this compound had an Rf equivalent to that of authentic 2,3-DPG, the identity of the 2,3-DPG in the ion exchange chromatogram is confirmed. 121 Discussion The present investigation substantiates and presents in a time course version the decrease in ATP seen by Pennington and Prankerd and the effect of glucose in preventing this decline as reported by Zipursky, et. al. However, the present study shows that the reduction in ATP level is not due to phosphorylation of galactose and sequestration of Gal-l-P. Rather, it demonstrates that the decline in ATP results from a restriction in the Leloir pathway ("slow" metabolism of galactose) which fails to provide enough G-6-P for adequate functioning of the glycolytic pathway. A further consequence is an increased breakdown of 2,3-DPG without balancing inflow into this pool, as this pool is used for ATP production at the pyruvate kinase step. The amount of galactose handled through the Leloir pathway in a given time, and reaching G-6-P, is markedly less than the amount of glucose that the red cell can phosphorylate to G-6-P in the same period of time. It is in this sense that one can speak of a "slow metabolism" of galactose. That galactose metabolism does indeed go through G-6-P, and not a by-pass pathway, was demonstrated in this study. 122 Transphosphorylation of glycolytic intermediates with ADP occurs at two steps in the Embden-Meyerhoff pathway: At the phosphoglycerokinase step (1,3-DPG) and at the pyruvate kinase step (PEP). To maintain proper stoichiometry between energy production and demands, the alternate 2,3-DPG pathway serves as a means to regulate ATP production, since any 1,3-DPG handled via this pathway has bypassed the first of the ATP generating steps, the phosphoglycerokinase reaction. Hill (78) did short-term incubations to study operation of the Leloir pathway in red cells. It was shown in this work that transferase activity is inhibited by G-l-P and UDPGal, both products of the reaction. Inhibition by UDPG also was observed at concentrations to be expected in the cell. Inhibition by G-6-P, ATP or ADP was a concentration well in excess of these intracellular. Epimerase activity was unaffected by G-l-P or G-6-P. These observations tend to support that conclusion reached in the study that the limitation on the capacity of the Leloir pathway to metabolize galactose is primarily a function of the transferase step. It is evident that the rate at which the Leloir pathway operates is subject to complex regulation. A central factor undoubtedly is transferase activity being influenced by the concentration of UDPG and by its 123 products, G-l-P and UDPGal. Certainly the rate at which epimerase can constantly replenish the UDPG pool is an important factor, especially in view of the limited amounts of UTP available in the red cell for UDPG pyro- phosphorylase activity. The net effect of all of the influences upon the Leloir pathway is to limit flow of galactose metabolites, with the central problem being the limitation imposed upon handling of Gal-l-P by transferase. THE SALVAGE PATHWAY FOR HYPOXANTHINE IN THE PLATELET Platelets are a hematological cell type which constitute the end product of megakaryocyte maturation in the bone marrow. Like erythrocytes they are not nucleated, but they do possess an active Krebs cycle and form glycogen. Early work was directed at functional studies of these cells in aggregation and adhesion in the coagulation mechanism. Effective biochemical studies have been realized only in the last decade. These cells appear to be extremely sensitive to environmental changes, making them very difficult to approach experi mentally. Biochemical studies have been aimed principally at correlating intracellular events with functional platelet behavior. In order to carry out many types of biochemical studies on platelets, relatively large blood samples have been required to produce the required quantities of isolated platelets. This, together with the sensitivity of these cells to manipulation, have presented problems in biochemical investigation. The availability of the atuomated phosphorus system, with its capacility of effective analysis of phosphorylated intermediates from minimal amounts of tissue, suggested that effective studies could be carried out. 124 125 Because of a strong interest in platelets on the part of Dr. Georges Rivard of the Hematology Division at Childrens Hospital, a basis for collaborative work became a reality. Blood samples were obtained by him, and platelet-rich plasma fractions were prepared. The biochemical studies were done in the Biochemistry Research Laboratory. The question asked had to do with whether or not the hypoxanthine could be salvaged by platelets for conver sion to adenine nucleotides. The information that could be gathered from the literature indicated that a salvage pathway was not functional in the platelet. In the light of what is known about other cell types, this did not seem reasonable. Preliminary experiments showed very quickly that after incubation of platelets with radio active hypoxanthine, radioactivity was present in ATP. Subsequently, other workers stated in a published abstract that they had demonstrated in platelets the salvage enzyme, hypoxanthine-guanine phosphoribosyl transferase (HGPRT). On the basis of the preliminary findings, and encouraged by the report of the other investigators, the study was continued on a broader basis. In the course of the work, thin layer chromatography also was employed, a circumstance providing for comparison of the 126 automated phosphorus system with other methodology of similar purpose. Background Adenine nucleotides are intimately involved in the aggregation of platelets. There appear to be two pools of these nucleotides. One, which is relatively metabolically inert, is confined to storage granules and is therefore designated the storage pool. It is composed predominantly of ADP and ATP (84). This pool is depleted during platelet aggregation. A second pool, designated the metabolic pool, is not compartmentalized and is involved in the conventional reactions of glycolysis and the TCA cycle. That this pool is active metabolically has been demonstrated in studies of platelets incubated with radioactive adenine (85). It was also shown that the metabolic pool is the precursor for the larger storage pool of adenine nucleotides. Mature platelets are unable to synthesize purines de novo (86) , yet it is known that some nucleotide degradation must occur, since a constant loss to the cell of hypoxanthine has been measured (85). In the face of this constant loss and in view of the inability of the cell to manufacture purine bases de novo, it is difficult to understand how the platelet is able to maintain an adequate stock of nucleotides. Three 127 possible sources for these compounds have been postulated: a) they are an inherited gift from their magakaryocyte precursor and depend on this initial supply for their survival, b) they are salvaged from endogenous catabolic reactions of nucleic acids or c) they are obtained from exogenous dietary sources. Holmes, et. al. (87, 88) demonstrated an active incorporation of adenine and adenosine by intact platelets. However, hypoxanthine could not be shown to be incorporated (88). The uptake and incorporation of bases and nucleo sides can occur via two different enzymatic pathways. The first pathway involves phosphoribosyltransferase reactions, in which the free base condenses with 6- phosphoribosyl-l-pyrophosphate (PRPP) to form ribo nucleotides in one step. Two of these enzymes are known, one specific for hypoxanthine and guanine (HGPRT) and another specific only for adenine (APRT). These reactions are reffered to as "salvage" pathways. The second pathway requires two enzymes, a nucleo side phosphorylase which reacts the free base with ribose-l-phosphate to form ribonucleosides and a second kinase enzyme which phosphorylates the ribonucleoside to the respective ribonucleotide. 128 Since the plasma concentrations of adenine or of guanine are very low compared to that of hypoxanthine, hypoxanthine appears to have more potential significance in maintaining the pools of adenine nucleotides in platelets. Additionally, the fact that this compound is produced by these cells suggests that it is immediately available for recycling purposes. Experiments General protocol: Platelets were obtained from whole blood of healthy volunteers by differential centrifugation (88). The platelets were obtained suspended in plasma (platelet- rich plasma or PRP). It was this suspension which was used for incubation experiments. Incubation was with C ^ hypoxanthine in a 37°C metabolic shaker. At given time, aliguots were removed, transferred to ice cold buffer and centrifuged. The pellet was lysed in ice cold water by freezing and thawing and recentrifuged. The supernatant from this centrifugation was saved. The pellet was washed with 7.5% TCA, recentrifuged and the supernatant pooled with the previous one. The TCA was extracted with 5 washings of ether. The aqueous phase, the platelet extract, was lyophilyzed and used in the chromatographic determinations. 129 The platelet extract was applied to the phosphate analyzer to separate the various nucleotide intermediates. The identities of the nucleotides appearing on the chromatogram were verified against standards and for one specific purpose by a peak shift experiment. Alternatively thin layer chromatography was used to separate the various nucleotide intermediates. Experimental details: The conclusions reached were based on a number of experiments. For clarity a typical protocol is described. Blood was provided by healthy adult volunteers. By venipuncture, 20 ml. of blood was removed from each and anticoagulated with one volume of 3.8% sodium citrate per nine volumes of blood. The blood was centrifuged at 2 00 x G for 10 minutes at room temperature. The super- natnat platelet rich plasma (PRP) was removed with a plastic pipette, and the platelets were counted in a MK-4 Haema-count Platelet Counting System (General Science Corp., Bridgeport, Conn.). Erythrocyte and leukocyte contamination was evaluated by visual countering and found to be one per approximately 5000 platelets. An aliquot of platelet rich plasma (4 ml.) was transferred to a small plastic test tube, fortified with approximately 10 microliters of a solution of 8-C^ 130 hypoxanthine (specific activity 53.7 mCi/mmole, obtained from Schwarz Bio-Research) to a final concentration of 10“^M, and incubated 37°C in a Dubnoff shaking incubator. Prior to incubation, the air in the tube was replaced by a mixture of 95% oxygen, 5% CO2• At one hour, 2 hours and 4 hours, one ml. aliquots of the mixture were transferred to plastic tubes containing 4 ml. of ice cold Tyrode solution and centri fuged for 15 minutes at 1000G. The supernatant was quantitatively removed for radioactive counting. Three ml. of ice cold distilled water was added to the pellet and mixed on a vortex mixer for 1 minute followed by one cycle of freezing and thawing in a glycerol dry ice mixture. The lysate so obtained was added dropwise to 3 ml. of ice cold 15% TCA while vortexing. This mixture was then submitted to five cycles of freeze- thawing in glycerol-dry ice. After one hour on melting ice, the preparation was centrifuged at 12,000 x G for 10 minutes, the supernatant was saved and the pellet washed with 2 ml. of 7.5% TCA. This suspension was recentrifuged, and the supernatant was removed and pooled with the previous supernatant. The solid pellet was saved for counting. The TCA was removed from the pooled supernatants by extracting 5 times with three volumes of diethyl ether. The pooled platelet extracts 131 (aqueous layer) were lyophilized, re-dissolved in 0.2 ml. distilled water and chromatographed without further treatment. Ion exchange chromatography was carried out with the phosphate analyzer system. The entire sample from each incubation was applied to the column. The column was eluted with a linear gradient of NH4CI from 0.05M to 0.3M. The pH was maintained at pH 10 with NH4OH throughout the entire gradient. The eluate from the column was monitored for radioactivity with the liquid scintillator counter. Since the experiment was designed to determine if any radioactive hypoxanthine is incorporated into nucleotides, the tracing obtained from the phosphate analysis was used only to identify the various radio active peaks. In some instances known nucleotides were added to the platelet extract as markers. Figure 19 is a typical chromatogram obtained after incubating platelets for four hours with hypoxanthine. As is evident, hypoxanthine is incorporated into various intermediates. Twenty seven per cent of the total radioactivity added was recovered in the TCA extract of the platelet pellet. Of this intracellular radioactivity, 84% was found in adenine nucleotides (ATP, ADP and AMP), 8.6% in guanine nucleotides (GMP, REtAT/'JE RAO/OACT/VJTY 132 ATP H ' f P o A A h l T H I N S Figure 19. Chromatogram of platelets incubated 4 hours with 8-Cl4-hypoxanthine. 133 GDP and GTP)f 4.5% in IMP and 2.9% in hypoxanthine. No radioactivity was detected in IDP or ITP. In this column chromatographic system XMP eluted with IMP and adenylosuccinic acid (AMPS) with GDP. Inosine could not be resolved from hypoxanthine and was counted with the latter. Figure 20 shows that the total incorporation of radioactivity into the nucleotide pool is linear over a — 5 4 hour period. This suggests that 10 M hypoxanthine saturates the uptake mechanism during this time period. The relative distribution of adenine and guanine nucleotides is shown in Figure 21. The ratio of these intermediates does not change appreciably during the incubation period. The authenticity of the adenine and guanine nucleo tide peaks was confirmed with a peak shift experiment. A platelet extract prepared from an incorporation' experiment, containing the typical distribution of adenine and guanine nucleotides, was incubated with 10 microliters of 10 mM phospho-enol-pyruvate (Sigma P7252) and one unit of pyruvate kinase (Calbiochem 5506) for one hour at room temperature. Pyruvate kinase is known to be able to phosphorylate ADP to ATP and GDP to GTP (89). As shown in Figure 22 all the radioactivity originally in the GDP and ADP peaks now was found in the MMOMOLeS O F R A D IO act/ve AD EN IN S NUCL&OTtOCS 134 3 0 © TIMS OF INCUBATION Figure 20. Incorporation of radioactivity of 8-cl4-hypoxanthine in platelets over 4 hours. NANO M O LES O F RADIOACTIVE COMPOUNDS PER ‘ O P L A T E L E T S 25© l o o - ISO ■ loo - SO • AOENtue UUCLEOTtOES ATP ATP AOP ATP AOP AMP AOP AMP I h r Zhrs AMP 4hft 20 >5 \0 6UA.NINE NUCLEOT10ES GTP GDP fiMP /Hk GTP GTP GDP GOP SMP GMP 2htm 4 him Figure 21. Relative distribution of adenine and guanine nucleotides in platelets incubated with 8-C^-^ hypoxanthine. R£i*r/v£ Rao/ oact/v try 136 6 8 3 % ATP Z. > V0 AMP Figure 22. Chromatograms of platelet extract before (A) and after (B) treatment with phospho-enol pyruvate. 137 GTP and ATP peaks, respectively. In conjunction with ion exchange chromatography platelet extracts were also analyzed using thin layer chromatography. This technique was used not only to confirm the results obtained with ion exchange but also to separate nucleotides which could not be resolved or remained undetected due to masking in the ion exchange procedure. The thin layer procedures were developed and carried out by Dr. John McLaren of the Biochemistry Research Laboratory. For these studies platelets were prepared and incubated under identical conditions to those used for experiments employing ion exchange chromatography for analysis. A time course study was not done, but a single point determination was made of platelets incubated for 4 hours with hypoxanthine. Two-dimensional chromatography (90) was performed on Brinkman MON Polygram Cel 400, 0.1 mm. thick microcrystalline cellulose precoated plastic sheets (20 x 20 cm.). The plates were chromatographed in one direction with a mixture of 6 parts n-propanol, 3 parts 25% NH4OH and one part water. They were then chromato graphed in the second direction with a mixture of two parts isopropanol, 79 parts saturated ammonium sulfate and 19 parts water. The time of chromatography was 138 variable but usually required 12-16 hours for each direction. Carrier nucleotides were added to the platelet extract to visualize the intermediates with ultraviolet light (254 nm). The spots were outlined and the plate xeroxed to provide a permanent record. The spots of the known markers as well as the origin were removed, placed in 5 ml. of scintillation liquid and counted. In addition some plates were overlayed with clinical x-ray film and stored in the dark for 1 to 3 weeks before development of the film. The autoradiography did not reveal the presence of any unexplained radioactive spots. Figure 23 shows a typical thin layer chromatogram indicating the position of the nucleotides added as carriers. When the chromatograms were analyzed for radioactivity, the proportional counts in each of the nucleotide intermediates corresponded closely to the data obtained with the automated ion exchange chromato graphy. Compounds which overlap in the latter system were clearly resolved by thin layer. The counts in inosine by thin layer never exceeded 2% of the total counts recovered intracellularly, and consequently only a small error was introduced in the ion exchange chromatograms by regarding these counts as hypoxanthine. No significant counts were obtained in the 139 MeNo&ue / ' " MBNtNE 6<J*NOSlN£ *ANT>hn£ * 6UANIN£ r'-s i i * o * J W c Add £ A AMp K « AfiP o O J T O O C t M f i imp i/ip AMAiI ' A w O « > '"J o - ► s b c on o enecnaN Figure 23. Thin layer chromatogram of standard nucleotides. 140 AMP-S area of the thin layer, suggesting that all the radioactivity in the AMP-S, GDP peak was GDP. This also had been demonstrated in the peak shift experiment, since the entire AMP-S, GDP peak was converted to GTP with pyruvate kinase. The thin layer study also showed that all the counts in the XMP-IMP peak on the ion exchange chromatograms belonged to IMP. Other studies; The remaining work in this study was carried on entirely by Dr. Georges Rivard and Dr. John McLaren. The presence in platelets from normal individuals of the enzyme hypoxanthine-guanine phosphoribosyl transferase (HGPRT) was confirmed. Platelets from patients with the Lesch-Nyhan syndrome were examined. Such individuals are known to be deficient in HGPRT in other cells, and it was established to be true also of their platelets. If a basic hypothesis of this investigation indeed is correct, namely that the first step in the conversion of hypoxanthine to ATP by the platelet involves HGPRT, then repeating the experiments described with platelets from Lesch-Nyhan syndrome patients should not result in the formation of labelled ATP from labelled hypoxanthine. Such labelling of ATP did not occur. Experiments also were done to show: (1) that the 141 rate of incorporation of hypoxanthine and the distri bution of radioactivity were similar in washed and non-washed platelets, and (2) that radioactive nucleo tides were not produced by contaminating cells (erythrocytes and leukocytes) in the platelet prepara tions, and (3) that normal plasma itself was in no way responsible for production of the different metabolites. Efforts to establish in this investigation the course of the complete pathway from hypoxanthine to the adenine nucleotides were not successful. The anticipated course would be from hypoxanthine to IMP to AMP-S to AMP. The amount of AMP-S present at any one time was too small, and too uncertain, to provide a basis for conclusions. Another study is in progress on the pathway, including use of a specific inhibitor for the conversion of AMP-S to AMP and evaluation of the enzyme for the conversion of IMP to AMP-S. Discussions and Conclusions Incubation of normal washed platelets over a period of four hours with (8-^C) hypoxanthine at a concentra tion of 10“^M resulted in a linear incorporation of radioactivity into adenine and guanine nucleotides. This finding is consistent with normal platelets having the enzymes necessary for salvage of hypoxanthine and with hypoxanthine-guanine phosphoribosyl transferase (HGPRT) 142 being the obligatory first step in this pathway. The finding was further substantiated by the other investi gators in the group who showed HFPRT activity to be present in platelets. They also demonstrated that platelets from Lesch-Nyhan syndrome patients, which lack HGPRT activity, did not show any significant incorpora tion of hypoxanthine into adenine nucleotides when incubated under the same conditions as normal platelets. The relative distribution of radioactivity between adenine and guanine nucleotides was found to correspond with that of Goetz, who labelled these compounds by incubating platelets with radioactive adenine (91). A slightly greater proportion of AMP was found in the present study but this could have been due to incomplete inhibition of adenylate kinase activity during the extraction procedure. The physiological significance of these findings remains speculative at this point. It is difficult to estimate to what extent circulating platelets might depend on hypoxanthine to maintain their nucleotide pool. Platelets appear to be removed from the circulation at random, and therefore a mixture of young and old cells is present in blood. It is possible that as platelets age and exhaust their supply of nucleotides inherited from megakaryocytes precursors, increasingly more 143 reliance might be placed on hypoxanthine salvage. Ion exchange chromatography provided a very effec tive means for the purpose of demonstrating incorporation of radioactive hypoxanthine into nucleotides. Because of the availability of the automated system for analysis of phosphorylated intermediates, an extensive analysis of each period of time course experiments could be carried out in minimal time with a maximal yield of information. However, identification of compounds in this procedure depends upon position in the chromatogram as determined by runs done with standards, and one cannot be certain in experiments with biological materials if one or more unanticipated compounds may be present in the same region of a chromatogram as a recognized substance. With this in mind, thin layer chromatography was employed as a supplementary procedure. The resolutions obtained in this additional approach proved to be very useful in interpretation of the results of the automated ion exchange analytical system. Although more complete resolution of nucleotides is obtained with thin layer, there are shortcomings of this method constituting the only methodological approach. In terms of sample capacity, thin layer is limited since only a small volume can be spotted. If the entire platelet extract were to be reduced to this volume, the 144 ionic strength would rise sufficiently to interfere with resolution of the various compounds. When adjacent spots on the chromatogram are cut out for radioactive counting, more subjective judgment is required than interpretation of shoulders on ion exchange chromato grams. The development of thin layer plates, requiring 12-16 hours in each direction, greatly exceeds the approximate 3 hours required for a similar analysis by the automated ion exchange chromatography. Although thin layer chromatograms can be done in multiple, the time factor becomes a very important factor in sequential studies, when the conditions of each succeeding experi ment depend upon the outcome of the previous one. A general conclusion of this study in specific relation to methodology is that the automated ion exchange analytical system greatly facilitated the investigation. It was possible to use far less quanti ties of platelets than would be required for a more conventional ion exchange approach. The short time required for an analysis with a large information content was very favorable. However, because of some problems in resolution and of some uncertainty of doubling of compounds within a given peak, resorting to thin layer chromatography as a secondary, supplementary method was useful and informative. DISCUSSION The automated analytical method which has been described was developed for the separation and quantita tion of metabolic intermediates present in acid soluble extracts of biological tissue, including intermediates of glycolysis, nucleotides and nucleotide sugars and other related phosphate-containing compounds. A con venient means to assess the concentration of these various metabolites is to determine the phosphorus content. A liquid chromatographic procedure using ion exchange resin was considered to be the most suitable method for the separation of the compounds from each other. Preliminary trials designed to determine the com patibility of the various components making up the system were encouraging, but two major general problems became apparent when use of the system was applied to biological extracts. Necessarily of concern was the wide range in concentration among the various phosphory- lated intermediates present in biological fluids and extracts of tissues. In particular tissues, a 100 fold range in concentration may be encountered. The second problem related to the marked chemical and structural similarity of many of the metabolites of interest. The 145 146 specificity of enzymatic reactions at the cellular level results in recognition of small differences in these compounds. It is highly desirable that an analytical method be able to achieve as nearly as possible a similar discrimination among the various intermediates if a meaningful profile of cellular events is to be obtained. Range of measurement; In approaching the first general problem, attempts were made to expand the sensitivity of the detector, i.e., phosphorus analysis, in order to encompass as large a range in concentration as possible. The ideal would be to provide for all compounds of interest in a single chromatographic separation. In practice the ideal was not attainable. As the concentration of any compound exceeded some critical level, the base width of the peak on the chromatogram also increased. The result was overlap with adjacent compounds on the chromatogram and reduced reliability in the quantitation and identifi cation of the compounds. For example, in the case of phosphorylated intermediates of erythrocytes, if an extract equivalent to 100 microliters of red cells is applied to the column, the peak corresponding to 2,3-DPG is broad enough to partially mask the nearby peak corresponding to ADP. This can be avoided by applying less sample. Both peaks become clearly identifiable, 147 but some of the other peaks then may become too small to measure accurately. Depending upon the need, it may be necessary to run two separate chromatograms to obtain the information required. Together with losses in resolution at high phosphate concentrations due to the peak broadening, another technical difficulty was encountered, one which proved to be very difficult to overcome. The intense blue color developed at higher phosphate concentrations coated the plastic transmission tubing and flow cell causing an unpredictable baseline deviation. With time this chromogen is washed into the flowing analytical stream, producing tailing and other irregularities interfering with the quantitation and identification of the various compounds. Considerable attention was given to find plastic tubing which would be better suited to carry the reaction mixture. It finally was realized that glass tubing was superior to plastic, and in the final system adopted the reaction mixture, after the addition of the reductant, is exposed only to glass. At extremely high phosphate concentrations, such as those encountered through inadvertent contamination of the system wifh phosphate, even glass became coated with a glue layer. However, at ordinary working concentrations there was no substantial problem. The use of a wetting 148 agent added to the reagent stream has also been of aid in preventing coating. The time and degree of heating of the reagent stream to enhance color development also was found to bear on the resolution of compounds and the carryover of chromo gens. Ideally, heating for 6 minutes of 100°C is desirable. In practice under these conditions, there was an increase in retention of chromogens by the glass tubing and consequently a prolonged washout time between phosphate peaks. Instead, heating at 75°C for 10 minutes proved to be satisfactory without sacrifice of sensi tivity. A particulate precipitate in the reagent stream also presented a problem. Most of the particles were small enough to move through the flow cell with the reagent stream without effect on the analysis. Occasion ally larger clump of precipitate formed and settled in the light path of the spectrophotometer flow cell causing aberrations in the baseline of the chromatogram. Much effort was devoted to designing different types with membrane or glass wool filters to trap the preci pitate, none proved practical. They had to be replaced too frequently, and they tended to disrupt the smooth flow of the reagent stream and bubble pattern. In the final system, the flow of the reagent stream has been 149 made fast enough to prevent the accumulation of precipi tate anywhere in the colorimetric manifold at ordinary phosphate concentrations. The modifications made were for the purpose of increasing the upper limit of phosphate concentration compatible with trouble-free performance. However, even at lower phosphate concentrations, with time the various components of the colorimetric manifold progressively became affected by the blue chromogen, as evidenced eventually by a steady rise in the baseline of the chromatograms. It was found necessary after every 10 chromatographic separations to flush the manifold with IN NaOH to remove this color. Further, the reagent stream is never allowed to remain in the tubing over night but is followed by a rinse with distilled water before shutdown. Resolution between compounds: The second challenging aspect of the development of the automated system related to the resolution of compounds. Many of the phosphorylated intermediates are very similar in structure and are difficult to separate. In general, resolution is a function both of the resin used and the elution gradient. The detector system also is a factor. If the detector system can accurately reflect every change in the eluate 150 from the column, the resin and gradient determine the ultimate resolution attainable. This is the case for gas chromatography and ion exchange chromatography of amino acids in which very little or no manipulation of the eluate from the column is required to detect the compounds being separated. In the present system the situation is different. The compounds must be ashed, the ash reconstituted and the phosphate measured by a colorimetric reaction. These operations, taken together, constitute the detector subsystem. These necessary manipulations of the column effluent can reduce the resolution produced by the resin. At the stage of development when Dowex resin of 200-400 mesh was used and large volumes were employed to elute the long column, overall resolution was clearly governed by the resin. It was feasible to compensate for the poor sensitivity of the colorimetric method then in use by employing flow cells of long optical path. Because of the large eluate volumes, the flow cells were not serious enough to detract from the resolution provided by the resin. However, with the change to smaller diameter resins, smaller eluting volumes and shorter time of chromatographic runs, any loss in resolution during ashing or in the colorimetric assay became a critical factor in the overall resolution 151 of the automated system. It became apparent that loss of resolution was occurring in the detector system when the scintillation counter was added to the system. Radioactive counting occurs immediately after the eluate stream leaves the column and just before it enters the asher. The chromatographic separation of radioactive compounds as detected by the scintillator flow cell were noted to be superior to that of the same compounds as determined by phosphate analysis. A number of modifications were made in the detector system to improve resolution by the asher and the colori metric assay. The tubing carrying the column effluent to the asher, and that carrying the reconstituted ashed material to the colorimetric manifold, was made as small as possible without unduly increasing back pressure in the system. The liquid scintillator flow cell obtained from a commercial source was substituted by one constructed and designed specifically to minimize backmixing of the eluate stream traversing the solid anthracene matrix of the cell. The rate of rotation of the asher has increased in order to double the number of aliquots collected in a given time. This also served to decrease bumping and splashing during ashing, since less volume then was present in each asher cup. 152 Variations in the time of reagent addition to the ashed material and modifications in the bubble pattern resulted in noticeable improvement in resolution. The flow cell finally used in the spectrophotometer has a light path of 1.8 cm. It was adapted from that of a Technicon SMA-12 autoanalyzer. In addition to having the desirable feature of being constructed entirely of glass, it also has the smallest diameter tolerable with the light transmittance requirements of the spectro photometer. With this flow cell, small air bubbles or particulate matter in the reagent stream are not trapped in the light path but flow through with the reagent stream. Contrary to the practice with conventional Technicon manifolds, in the present system the reagent stream is pumped both in and out of the flow cell, insuring a steady rate of flow. Despite all the modifications, some peak spreading still may occur during ashing and the subsequent phosphate assay. A convenient way to measure this is to place a phosphate standard in a single asher cup. Ideally the photometric indication should appear on the chromatogram as a spike of 30 seconds duration. Usually, however, this spike is distributed over a 2 minute interval on the chromatogram. 153 Elution gradient: The possibility of attaining better separation of compounds by manipulating the elution gradient was also explored. It was realized early that when the column was eluted with a linear gradient, the various compounds emerged broadly as monophosphates, diphosphates, and triphosphates. The stress in development was on displacing the various compounds within each class to vacant areas on the chromatogram by using gradients suitable in profile and in composition with respect to pH and concentrations of borate and of ammonium chloride. The gradients were formed with one of several gradient devices made for the purpose. Volumes were kept as small as possible. It was found desirable to decrease the flow rate of the buffer to allow sufficient time for equilibrium, and thus to increase the number of theore tical plates in the resin bed. However, the limitation was the need to limit the time required for the chromatographic separation. The optimum volume was determined empirically from such considerations as pressure, diffusion of compounds in the tubing carrying the column eluate to the asher and, above all, the capacity of the asher. Different types of gradient devices were studied in attempts to find an arrangement which would give the most reproducible eluting profiles. 154 Of all tested, the device finally used appeard to give the most reproducible results and to be the most convenient to handle. With the small eluting volumes finally used, the limits of reproducibility are defined by surface effects on the sides of the buffer vessels balancing with the hydrostatic pressure during elution. A small percentage of uncertainty in reproducibility has been found unavoidable. The information gained from studies with different shapes of gradients and changes in composition of the eluting buffer have been incorporated in the version of the gradient finally adopted. Sufficient insight was gained concerning gradient parameters to provide a basis for special conditions applicable in studies when the emphasis is on a particular group of compounds In such instances, the gradient conditions can be tailored to maximize resolution of the compounds of interest at the expense of those not of concern for the particular purpose. Sensitivity: In contrast to making modifications to enhance the resolution of the method or to expand the range of measurement, there was less experimental latitude in attempts to obtain improvements in absolute sensitivity 155 Some progress was made simply by decreasing the volume of buffer used to elute the column. The initial requirement of two liters was decreased during develop ment to the point that only 200 ml. is needed. This in itself provides for a 10 fold increase in sensitivity, since the phosphorylated compounds are less diluted by the elution stream. The colorimetric assay for phosphate was also made more sensitive by heating the phospho- molybdate complex and by monitoring the reagent stream at 820 nanometers in the spectrophotometer at the traditional 660 nanometers. An increase in sensitivity by a factor of 2.5 resulted from this modification. The sensitivity of the method came to be such that nanomole quantities of the various metabolic intermediates can be accurately measured. It constitutes one of the important methodological advantages of the system which was developed. It permits studies of tissues which previously were difficult to subject to detailed biochemical investigation because of small amount of sample available. Much of the previous data on phos phorylated intermediates has been obtained using relatively large quantities of blood or of tissue from human sources at autopsy. The methodology developed facilitates the in vivo or vitro biochemical studies of cultured skin fibroblasts, 156 platelets, leukocytes and erythrocytes. The quantities required are small enough that application of the method to studies of biopsy samples should be possible. General view: In addition to the high sensitivity of this method, the separation of phosphorylated intermediates in a single chromatographic separation constitutes a major improvement over existing methodology. Previously, in a comprehensive study of phosphorylated intermediates, fractions of the column's eluate had to be re-chromato graphed in different systems to obtain the same degree of resolution presently achieved in a single run on the automated system. Much of this improvement has been possible because of the availability of superior ion exchange resins which permit faster chromatographic separation with equal or better resolution than that obtained with older resins. It is not claimed that every possible separation desired can be achieved in a single run. In the work with erythrocytes, for example, it became apparent that the large amount of 2,3-DPG present required the use of two different elution gradient profiles for reliable resolution of the compounds present. In addition, in this application it was found that the UDPHexose peak could not be separated into UDPG and UDPGal without the 157 application of an additional method. Undoubtedly there are other areas in which thin layer or paper chromato graphy could provide supplementary information. However, such considerations do not in any way detract from the value of the automated analytical system as a primary tool in metabolic studies. The usefulness of the system was evident in its application to metabolic studies in erythrocytes. The capacility of working with small amounts of sample, as opposed to the large volumes of cells used in the earlier work of Bartlett and others, was in itself an advantage. The short time of chromatography made possible time course experiments which provided a great deal of information, as opposed to the single point incubations that earlier workers usually employed. The speed with which an analysis could be done, and the resolution obtained, provided the basis for exploratory studies in which the results of one experiment form the basis to establish the conditions for the next. A further advantage of the system has been its ability to provide data both on the concentration of phosphorylated inter mediates and on their specific activities following administration of a radioactive precursor. Such information provided valuable insight in the metabolic turnover of glycolytic intermediates in the application 158 to erythrocytes. The application of the automated system to nucleotide metabolism in platelets also proved its value in practice. It was possible to determine with ease in an exploratory experiment that platelets can convert radioactive hypoxanthine to radioactive ATP, contrary to existing views. The automated analysis then provided the principal means of continuing with a systematic investigation. However, as was found in the erythrocyte studies, supplementary use of thin layer chromatography was a useful adjunct. In the course of the platelet studies it became apparent that the addition of on-stream ultraviolet sensing for nucleotides, and related compounds, would increase the usefulness of the automated system. Such a provision was not possible during the course of the present work, but it is planned for the near future. The development of the automated analysis was carried to the point at which the time for a given analysis, and the resolution obtained would be acceptable for application to metabolic studies. However, further improvements are possible. For example, increased resolution and shorter chromatography times could be achieved by using high pressure liquid chromatography. This technique has given surprisingly good results in 159 very short times when applied to analytical problems such as those encountered with amino acids. However, the rapidity with which these separations are achieved, often in a matter of minutes, requires that the detector system be able to discriminate at an equally fast rate. In the present detector subsystem, the rate of ashing governs the overall rate of the phosphate analysis, and it therefore is the component which seriously limits further improvement. One possibility for improvement which could be explored is ashing by a different technique. One such possibility is using a carbon rod mini-furnace although any advantages to be gained might well be offset by difficulties to be encountered in adapting such an ashing method to continuous flow. A wet ash digestion technique has been considered, but it presents problems in adaption to rapid continuous flow and in the need to neutralize the strong acid generally used. Until there is a radical new approach to determin ation of the organic phosphorylated compounds other than analysis of phosphorus, the inpediment of the relatively slow detector subsystem remains as a hindrance to the benefits of high pressure liquid chromatography. It is possible, however, that improvements in the mechanical operation of the present 160 asher could be made to provide somewhat greater speed of analysis than is possible in the system as it is now constituted. BIBLIOGRAPHY 1. Schneider, A.S., Dunn, I., Ibsen, K.H. and Weinstein, I. Clin. Res. lji, 282 (1965). 2. Vanderheiden, B.S. Analytical Biochem. £, 1 (1964). 3. Bartlett, G.R. J. Biol. Chem. 234, 459 (1958). 4. Binkley, W.W. Adv. Carbohyd. Chem. 1_0, 55 (1955). 5. Crane, R.K. and Lipmann, F. J. Biol. Chem. 201, 235 (1953). 6. Sweeley, C.C., Bentley, R., Makita, M. and Wells, W.W. J. Am. Chem. Soc. £5^, 2497 (1953) . 7. Hashizume, T. and Sasaki, Y. Analytical Biochem. 15, 199 (1966). 8. Horning, M.G., Boucher, A.E. and Moss, A.N. J. Gas Chromatog. 6_, 297 (1967) . 9. Harvey, D.J. and Horning, M.G. J. Chromatog. 76, 51 (1973). 10. Hanes, C.S. and Isherwood, F.A. Nature 164, 1107 (1949) . 11. Leuthardt, F. and Testa, E. Helv. Chim. Acta 34, 931 (1951). 12. Cohen, S.S. and Scott, D.B.M. Science 111, 543 (1950). 13. Bandurski, R.S. and Axelrod, B. J. Biol. Chem. 193, 405 (1951). 14. Caputto, R., Leloir, L.F., Cardini, C.E. and Paladini, A.C. J. Biol. Chem. 184, 333 (1950). 15. Kenner, G.W., Todd, A.R. and Webb, R.F. J. Chem. Soc. 2843 (1954). 16. Munch-Peterson, A., Kalckar, H.M., Cutolo, E. and Smith, E.E.B. Nature 172, 1036 (1952). 161 162 17. Cohn, W.E. and Carter, C.E. J. Am. Chem. Soc. 72, 4273 (1950). 18. Calvin, M. and Benson, A.A. Science 109, 140 (1949). 19. Goodman, M., Benson, A.A. and Calvin, M. J. Am. Chem. Soc. 11_, 4257 (1955) . 20. Randerath, K. and Struck, H. J. Chromatog. 6_, 365 (1961). 21. Randerath, K. Biochem. Biophys. Res. Commun. < 6, 452 (1961). 22. Peterson, E.A. and Sober, H.A. J. Am. Chem. Soc. 78, 751 (1956). 23. Randerath, K. and Randerath, E. J. Chromatog. 16, 111 (1964). 24. Obreiter, J.B. and Stowe, B.B. J. Chromatog. 16, 226 (1964). 25. Dyer, T.A. J. Chromatog. 11, 414 (1963). 26. Jermyn, M.A. and Thomas, R. Nature 172, 728 (1953). 27. Sober, H.A. and Peterson, A.E. J. Am. Chem. Soc. 76, 1711 (1954). 28. Khym, J.X., Roberty, D.G. and Cohn, W.E. J. Am. Chem. Soc. 76. r 5523 (1954) . 29. Hurlbert, R.B., Schmitz, H., Brumm, A.F. and Potter, V.R. J. Biol. Chem. 209, 23 (1954). 30. Khym, J.X. and Cohn, W.E. J. Am. Chem. Soc. 75, 1153 (1953). 31. Bedetti, G., D'Agnolo, G. and Pocchiari, F. J. Chromatog. £9, 53 (1970) . 32. Khym, J.X. and Zill, L.P. J. Am. Chem. Soc. 74, 2090 (1952). 163 33. Bessman, S. Personal Communication. 34. Brewer, P.G., Chan, K.M. and Riley, J.B. in Automation in Analytical Chemistry, Technicon Symposium, p. 245, 1965. 35. Annison, E.F., Lindsay, D.B. and White, R.R. Biochem. J. 8^, 243 (1963). 36. Robinson, A. J. Exp. Med. 118, 359 (1963). 37. Hamilton, P.B. Analytical Biochem. 3_0, 914 (1958). 38. Lefebore, M.J., Gonzales, N.S. and Pontis, H.G. J. Chromatog. _15, 495 (1964). 39. Bessman, S.P. Analytical Biochem. 1J8, 256 (1967). 40. Fiske, C.H. and Subba Row, Y.J. J. Biol. Chem. 66, 375 (1925). 41. Hirsch, C., Kupfer, S. and Oreskes, I. Clin. Chim. Acta 18, 322 (1967). 42. Kraml, M. Clin. Chim. Acta 1^3, 442 (1966). 43. Hobbs, J.S. and Lawrence, J.G. J. Chromatog. 72, 311 (1972). 44. Murphy, J.E. J. Clin. Lab. Invest. 55_, 281 (1960) . 45. Hjelm, M. Scand. J. Clin. Lab. Invest. 18.' 95 (1966). 46. Widdas, W.F. J. Physiol. 127, 318 (1955) . 47. Rapaport, S., Hinterberger, V. and Hoffman, E.C.G. Naturwiss. j48, 501 (1961). 48. Grignani, F. and Lohr, G.W. Klin. WSCHR. 38_, 396 (1960) . 49. Rose, I.A., Warms, T.V.B. and O'Connell, E.L. Biochem. Biophys. Res. Commun. 15_, 33 (1964). 50. Tsuboi, K.K. and Fukunaga, K. J. Biol. Chem. 240, 2806 (1965). 164 51. Rose, I.A. and Warms, T.V.B. J. Biol. Chem. 241, 4848 (1966). 52. Rapoport, S. and Guest, G.M. J. Biol. Chem. 138, 153 (1949). 53. Rapoport, S. and Leubering, T. J. Biol. Chem. 183, 507 (1950) . 54. Valentine, N.W. , Oski, F.A., Paglia, D.E. , Baugham, M., Schneider, A.S. and Naiman, J.L. in Hereditary Diseases of Erythrocyte Metabolism, p. 296 (1968) (E. Beutler, ed.) New York: Grune & Stratton 55. Isselbacher, K.J., Anderson, E.P., Kurahashi, K. and Kalckar, H. Science 123, 635 (1956). 56. Caputto, R., Leloir, L.F., Cardini, C.E. and Paladini, A.C. J. Biol. Chem. 179, 497 (1949). 57. Leloir, L.F. Arch. Biochem. 3_3, 186 (1951). 58. Kalckar, H.M., Bragama, B. and Munch-Peterson, A. Nature 172, 1038 (1953). 59. Ng, W.G., Bergren, W.R. and Donnell, G.N. Clin. Chim. Acta 1!5 , 489 (1967) . 60. Beutler, E., Baluda, M.C., Sturgeon, P. and Day, R.W. J. Lab. Clin. Med. 6_8, 646 (1966) . 61. Mathai, C.K. and Beutler, E. Enzymologia 33, 22 4 (1967). 62. Ng, Won G. Exp. Eye Res. 11, 402 (1971). 63. Gitzelmann, R. Lancet II, 670 (1965). 64. Gitzelmann, R. Helv. Paed. Acta 2_7, 125 (1972). 65. Segal, S., Blair, A. and Topper, Y.J. Science 136, 150 (1962). 66. Weinberg, A.N. Metabolism 13, 728 (1961). 165 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. Isselbacher, K.J. Science 126, 652 (1957). Gitzelmann, R. Pediatric Res. 3^, 279 (1969). Inouye, T. and Hsia, D.Y.Y. J. Lab. Clin. Med. 58, 829 (1961). Segal, S. and Cuatrecasas, P. Am. J. Med. 44, 340 (1968). Bergren, W.R., Ng, W.G. and Donnell, G.N. Science 176, 683 (1972) . Van Heynigen, R. Biochem. J. 12_, 197 (1959) . Chacko, C.M., Christian, J.C, and Nadler, H.L. Am. Ped. Soc., Soc. Ped. Res. (abstract) p. 25 (1970) . Sidbury, J.B., Jr. in Molecular Genetics and Human Disease (L.I. Gardner, ed.) Springfield, Illinois: Charles C. Thomas, 1961. Pennington, J.S. and Prankerd, T.A.J. Clin. Sci. 17, 385 (1958) . Zipursky, A., Rowland, M., Ford, J.D., Haworth, J.C. and Israels, L.G. Pediatrics 3_5, 127 (1965). Oski, A.F. and Rose, I.A. J. Lab. Clin. Med. 70, 535 (1967) . Hill, H.Z. J. Cell. Physiol. 78, 419 (1971). Schwarz, V., Goldberg, L., Komrower, G.M. and Holzel, A. Biochem. J. 6_2, 34 (1956). Beutler, E. and Collins, F. Scand. J. Haemat. 2 _ , 343 (1965). Betke, K., Baltz, A. and Maas, U. Ztschr. Kinderh. 84, 226 (1960) . Bartlett, G.R. Biochim. Biophys. Acta 156, 231 (1968) . 166 83. Benesch, R. and Benesch, R.E. Biochem. Biophys. Res. Commun. 26_, 162 (1967). 84. Holmsen, H. Biochemistry of Platelets (F. Kowalski and S. Niewiarowski, eds.) New York: Academic Press, 1967. 85. Holmsen, H., Day, H.J. and Storm, E. Biochim. Biophys. Acta 186, 254 (1969). 86. Holmsen, H. and Rozenberg, M.C. Biochim. Biophys. Acta 157, 266 (1968). 87. Holmsen, H. and Rozenberg, M.C. Biochim. Biophys. Acta 155, 326 (1968). 88. Holmsen, H. and Rozenberg, M.C. Biochim. Biophys. Acta 159, 311 (1968). 89. Barman, T.E. Enzyme Handbook. Vol. 1. New York: Springer-Verlag, Inc., 1969. 90. Pataki, G. J. Chromatog. 29_, 126 (1967). 91. Goetz, U., Da Prada, M. and Pletscher, A. J. Pharmacol. Exp. Ther. 178, 210 (19 71). INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 BRUNST, Robert Fred, 1942- STUDIES OF PHOSPHORYIATED METABOLIC INTERMEDIATES. University of Southern California, Ph.D., 1974 Chemistry, biological I University Microfilms, A XEROX Company, Ann Arbor, Michigan THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Structure and biochemical function of the photosynthetic apparatus

PDF

Studies on the transport, metabolism and chemistry of iron-sugar chelates

PDF

Enzymatic dithiol criteria

PDF

Kinetic investigation of the hydrolysis of aryl β-D-glucopyranosiduronic acids by β-glucuronidase

PDF

Studies on acyl transfer reactions on n-acylimidazoles

PDF

Galactose metabolism in human blood cells

PDF

Substrate and inhibitor interactions of glyceraldehyde 3-phosphate dehydrogenase and intramolecular general base-catalyzed alcoholysis of amides and esters

PDF

The separation of complete and incomplete blood group antibodies

PDF

Water relationships of feral house mice, Mus musculus, Linnaeus, from a coastal salt-marsh habitat

PDF

Studies in growth factor requirements and niacin metabolism of germinating orchid seeds and young tissues

PDF

Metabolism of 5-hydroxyuracil compounds

PDF

Studies on the pulmonary surface-active lipoprotein: isolation and characterization of its peptide moiety

PDF

Studies On Micellization

PDF

The ferric complexes of citrate, ETDA and NTA: their nature and the ligand exchange reaction with human transferrin

PDF

Studies of trypsin-binding ɑ₂ macroglobulin of human plasma

PDF

Effects of sex hormones on cholesterol metabolism

PDF

Steric effects in the acid, amine, and glyceraldehyde-3-phosphate dehydrogenase catalyzed hydrolysis of acyl phosphates

PDF

Studies on the biosynthesis of glycoprotein in reticulocyte stroma

PDF

The Carbohydrate Moiety Of The Tamm And Horsfall Urinary Glycoprotein Andits Significance In Cystic Fibrosis Of The Pancreas

PDF

Steric effects in imidazole and ɑ-chymotrypsin catalyzed ester hydrolysis

Asset Metadata

Creator Brunst, Robert Fred, 1942- (author)

Core Title Studies of phosphorylated metabolic intermediates

School Graduate School

Degree Doctor of Philosophy

Degree Program Biochemistry

Degree Conferral Date 1974-06

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

Tag chemistry, biochemistry,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Bergren, William R. (committee chair), Donnell, George N. (committee member), Fluharty, Arvin L. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-849074

Unique identifier UC11356495

Identifier 7423574.pdf (filename),usctheses-c18-849074 (legacy record id)

Legacy Identifier 7423574.pdf

Dmrecord 849074

Document Type Dissertation

Rights Brunst, Robert Fred

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