leATiON 0F GALACTlTOL METABOLISM IN AEROBACTER AEROGENES Thesis far the Degme of M. S. MlCHIGAN STATE UNWERSITY GRAN? TSUYOSHE SfilMAMOYO 1 9'7 5 h‘ _ ;- '.,.,_§}_~ v- JHESIS f ' ~ '1‘: v1.5: L" . L ‘ ‘n . ~ ' g“...- .,40 (5 amua’v' wk dc? u 95 5. ~ n- . x. 4-. onMil-‘tnwvlm #3. '5” plasma is! " “DAG & SUNS' 500K BUNNY INC. LIBRARY BINDERS din-nu .--lll.l ADV ABSTRACT Initiation of Galactitol Metabolism in Aerobacter aerogenes by Grant Tsuyoshi Shimamoto The metabolism of galactitol in Aerobacter aerogenes is shown to be initiated by a phosphoenolpyruvate-dependent sugar phosphotransferase system. The product of the reaction cata- lyzed by this system is conclusively established as L-galactitol l-phosphate by showing that it exhibits the same physical, chemical, and enzymatic characteristics as does chemically synthesized L-galactitol l-phosphate. Furthermore, this product can be enzymatically oxidized using L-galactitol 1-phosphate dehydrogenase to yield D-tagatose 6-phosphate, which was characterized with respect to gas-liquid chromatography, acid lability, periodate oxidation, and enantiomeric configuration. INITIATION OF GALACTITOL METABOLISM IN AEROBACTER AEROGENES By Grant Tsuyoshi Shimamoto A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1975 Dedicated to my family ACKNOWLEDGEMENT I thank my major professor, R. L. Anderson, who more than fulfilled his responsibility as a teacher by in- creasing my scientific abilities. I also thank the other members of the laboratory who's suggestions have also guided in my scientific training. For her enduring patience I thank my wife, Maria. ii TABLE OF CONTENTS ACKNOWLEDGEMENTS . LIST OF TABLES . LIST OF FIGURES LIST OF SYMBOLS AND ABBREVIATIONS INTRODUCTION . LITERATURE REVIEW MATERIALS AND METHODS Bacterial Strains . Media . . Mineral Media Nutrient Broth . Nutrient Agar Growth of Cultures . . Preparation of Cell Extracts Qualitative Analysis . . Dephosphorylation of Sugar Phosphates Gas- -Liquid Chromatography of the Poly- acetate Derivatives of Various Polyols . . Gas- -Liquid Chromatography of Trimethylsilyl Sugar Derivatives . . . Combined Gas- -Liquid Chromatography- -Mass Spectrometry . . . . Determination of Acid Lability . Periodate Oxidation . Enzyme Assays . . . Assay for PTS Components . . . Mannitol l- -Phosphate Dehydrogenase Assay . L- Galactitol l- -Phosphate Dehydrogenase Assay : Phospho lycolate Phosphatase Assay . . . Partial Purification and Preparation of Enzymes . Preparation of gltl and mtl. . . E11 E11 Partial Purification of E Partial Purification of HPr. Partial Purification of Mannitol l- o-Phosphate . Dehydrogenase . iii viii TABLE OF CONTENTS (Continued . . .) Partial Purification of L- Galactitol 1- -Phosphate Dehydrogenase . . Partial Purification of Phosphoglycolate Phosphatase . . . . Quantitative Analysis . . . Inorganic Phosphate Determination . Protein Determination . . Ketohexose 6- -Phosphate Determination Determination of Reducing Sugar . . . Enzymatic Determination of L- Galactitol l- -Phosphate . . - Determination of Phosphoglycolate . Determination of Formaldehyde . Determination of Glycolate . . Phosphoglycoaldehyde Determination Determination of Carbon Dioxide . Determination of Formate . Determination of Periodate Consumption . Chemical Synthesis of L- Galactitol l- -Phosphate . Reagents .7. . . . Other Analytical Procedures RESULTS . PTS- -Catalyzed Phosphorylation of Galactitol Preparation and Isolation of the Product of the PEP- -Dependent Phosphorylation of Galactitol . Identification of the Enzymatically Synthesized Reaction Product as L-Galactitol l-Phosphate and Characterization of Chemically Synthesized L-Galactitol L -Phosphate . . Gas- -Liquid Chromatography of. Derivatives of Chemically and Enzymatically Synthesized Galactitol Phosphate Before and After Dephosphorylation . . Mass Spectrometry of the TMS Derivatives of Chemically Synthesized L- Galactitol l-Phosphate After Dephosphorylation . Mass Spectrometry of Enzymatically and Chemically Synthesized Galactitol Phosphate as the TMS Derivatives Periodate Oxidation of Chemically Synthesized L-Galactitol l-Phosphate Identification of the Phosphorylated Product as L-Galactitol l—Phosphate by Enzymatic Oxidation to D-Tagatose 6- -Phosphate . . . Preparation and isolation of D-tagatose 6- -phosphate iv 67 . 68 75 85 . 86 . 93 . 93 TABLE OF CONTENTS (Continued . . .) DISCUSSION . REFERENCES GLC of the TMS derivative of the oxidized reaction product before and after dephosphorylation . Determination of the acid lability of . the oxidation product and chemically synthesized L-galactitol l-phosphate and the determination of the keto- hexose/P. molar ratio of the oxida- tion product . Periodate oxidation of the un- identified ketohexose phosphate . Purity and enantiomeric determination of the enzymatically synthesized tagatose 6-phosphate . . 103 107 123 129 136 LIST OF TABLES Table 1. 10. 11. 12. 13. 14. 15. Validity of the L-galactitol l-phosphate de- hydrogenase assay and activity with galactitol as substrate . Cofactor specificity of L-galactitol 1-phosphate dehydrogenase and apparent inducibility Summary of purification of EI . Purification summary of HPr . Summary of purification of mannitol l-phosphate dehydrogenase . . . . . . . . . . . Summary of purification of L- galactitol 1- -phosphate dehydrogenase . . . . . . . . . . . . . Summary of purification of phosphoglycolate phosphatase . . . . . . . . PTS-dependent phosphorylation of galactitol Purity of the enzymatically synthesized galactitol phosphate and test as a substrate for L- galactitol l-phosphate dehydrogenase . Retention times of the polyacetate derivatives of the polyols analyzed in Figure (9) Periodate oxidation analysis of chemically syn- thesized L-galactitol l-phosphate . P [ketohexose molar ratio of the unidentified ketohexose phosphate . Consumption of periodate by ketohexose phosphates. Production of formaldehyde after periodate oxi- dation of the unknown ketohexose phosphate . Formation of glycolate after periodate oxida- tion of the unidentified ketohexose phosphate. vi . 22 . 23 . 36 . 38 39 . 44 . 47 . 61 . 69 . 74 . 94 .110 .113 115 116 LIST OF TABLES (Continued . . .) Table 16. Formic acid production from the unidentified ketohexose phosphate after periodate oxidation. 17. Analysis for carbon dioxide production after periodate oxidation of the unidentified ketohexose phosphate . . . . . . . . . . 18. Production of phosphoglycoaldehyde after perio- date oxidation of ketohexose phosphates . 19. Formation of phosphoglycolate after periodate oxidation of the unidentified ketohexose phosphate . 20. Summary of the periodate oxidation data . 21. Purity and enantiomeric configuration of the enzymatically synthesized tagatose 6-phosphate. vii 119 122 . 124 125 127 10. 11. 12. LIST OF FIGURES DEAE cellulose chromatography I of the 35 to 70 percent ammonium sulfate fraction from the Biogel P-6 protein pool . DEAE cellulose chromatography II of the dialyzed EI pool from DEAE cellulose chroma- tography I . . . . . . . . . . . . . . Biogel P-200 chromatography of the concentrated EI pool from DEAE cellulose chromatography II . DEAE cellulose chromatography of L-galactitol 1-phosphate dehydrogenase from the 35 to 70% ammonium sulfate fraction after desalting over Biogel P-6 . DEAE cellulose chromatography of phosphoglyco- late phOSphatase from the third 25 to 40% acetone fraction . . . . . . . . . . Sodium.borohydride reduction of commercially prepared D-galactose 6-phosphate Enzymatic synthesis of galactitol phosphate . GLC characterization of the hexaacetate deriva- tive of chemically synthesized L-galactitol l-phosphate after dephosphorylation . . GLC of some polyols as their polyacetate deriv- atives . . . . . . . . . . . . GLC analysis of the TMS derivative of enzym- atically synthesized galactitol phosphate . GLC of the TMS derivative of enzymatically synthesized galactitol phosphate after de- phosphorylation . . . . . . . GLC of the hexaacetate derivative of enzymatic- ally synthesized galactitol phosphate after dephosphorylation . . viii . 31 . 33 . 40 . 45 . 55 . 65 . 71 . 73 . 76 77‘ 78 LIST OF FIGURES (Continued . . .) Figure 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. GLC of the TMS derivative of chemically syn- thesized L-galactitol 1- -phosphate after de- phosphorylation . . . . . . . . . Mass spectrometry of the isomers of D- galactose (TMS)5 . Mass spectrometry of chemically synthesized L-galactitol l-phosphate as the TMS derivative after dephosphorylation Mass spectrometry of chemically synthesized L-galactitol l-phosphate and enzymatically synthesized galactitol phosphate as their TMS derivatives . . . . . . . . . . . . Mass spectrometry of standard D-galactose 6-phosphate (TMS)6 . Periodate oxidation scheme for L-galactitol 1-phosphate . . . . . . . . . . . . . . . Enzymatic oxidation of enzymatically syn- thesized galactitol phosphate . . . . . AGL X2 Cl- Ion-exchange chromatography of the enzymatically synthesized ketohexose phosphate . . . . . . . . GLC of the TMS derivatives of the unidentified ketohexose phosphate and authentic D-tagatose 6-phosphate . . . . . . . . . . . GLC of the TMS derivatives of authentic D- tagatose and the unidentified ketohexose after dephosphorylation Acid lability of the unidentified ketohexose phosphate . . . . . . . . . . . . . . . . Theoretical periodate oxidation scheme of ketohexose phosphates Acid lability of the organic phosphate pro- duced after periodate oxidation of the un- identified ketohexose phosphate Initiation of galactitol metabolism by the PTS . . . . . . ... . . . . . . . Possible products of the PTS- -cata1yzed .phos- phorylation of galactitol . . ix 80 81 83 .87 . 89 . 95 . 99 101 104 105 108 112 121 131 133 ATP EDTA org TMS BSA TMCS NADH NADP NADPH PEP mtl gltl E11 PTS HPr Tris TCA DEAE GLC ABBREVIATIONS adenosine triphosphate ethylenediamine tetraacetate inorganic phosphate organic phosphate trimethylsilyl bis-trimethylacetamide trimethylchlorosilane nicotinamide adenine dinucleotide reduced nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide phosphate reduced nicotinamide adenine dinucleotide phosphate phosphoenolpyruvate enzyme II mannitol enzyme II galactitol phosphoenolpyruvate-dependent sugar phosphotrans- ferase system enzyme I histidine-containing phosphotransfer protein tris(hydroxymethyl)aminomethane trichloroacetic acid diethylaminoethyl gas-liquid chromatography INTRODUCTION An interest in bacterial hexitol metabolism led to my investigating the route of galactitol dissimilation in Aerobacter aerogenes PRL-R3. Galactitol catabolism has not been elucidated for any organism except for a Pseudomonas species which was found to directly oxidize the polyol to D- tagatose by utilization of a nonspecific dehydrogenase and + . . . NAD (see L1terature Rev1ew). However, 1n A, aerogenes, the occurrence of dehydrogenases specific for the phosphate esters of the naturally occuring hexitols (mannitol, sorbitol, and galactitol) suggested that the initiation of hexitol metabolism for this organism could proceed by phosphorylation prior to oxidation. Since it was known that A, aerogenes can simul- taneously transport and phosphorylate specific sugars including sorbitol by means of the phosphoenolpyruvate-dependent sugar phosphotransferase system (PTS), it was highly plausible that galactitol biodegradation could be initiated through this system. The present work confirms that galactitol metabolism in A, aerogenes is initiated through phosphorylation of the hexitol by means of the PTS in lieu of initiation by direct oxidation of galactitol. The product of the PTS-catalyzed reaction is identified as L-galactitol l-phosphate by chemical, 1 2 physical, and enzymatic analysis. Furthermore, a dehydrogen- ase for L-galactitol 1-phosphate is shown to oxidize the product of the PTS-catalyzed reaction to D-tagatose 6-phosphate. LITERATURE REVIEW The literature is plethoric with studies concerning polyol metabolism in both bacterial and mammalian systems. There are a number of review articles available which pertain to the metabolism of these carbohydrates (1-8) and therefore a detailed account will not be given here. Instead, bacterial polyol biodegradation will be circumscribed to the naturally existing hexitols galactitol, D-glucitol (sorbitol), and mannitol, with converging interest in galactitol. These hexahydric alcohols have been recognized since the nineteenth century commencing with the discovery of mannitol in 1806 and the identification of galactitol in 1850 and sorbitol in 1872. Since the last quarter of the nineteenth century, these hexitols have been instituted as an aid in differentiating various bacteria (8) such as the constituents of the Enterobacteriaceae family which have distinguishable ferment- ative properties in the presence of galactitol (9). Until 1951, the metabolism of galactitol and the other polyols has been examined utilizing acid and/or gas production (IO-13), oxygen consumption (14—17), or formation of unidentified reducing sugars (18-20) as criteria. The first hexitol dehydrogenase was isolated by Blakely (21) who demonstrated its catalytic requirement in the oxidation of different polyols 3 4 to their corresponding ketohexoses. It was later shown that NAD+ was required as a coenzyme for the reaction (22). In 1955 (23) and 1956 (24) a second route of hexitol dissimilation was reported which involved phosphorylation prior to the oxidation catalyzed by a hexitol phosphate dehydrogenase requiring NAD+ as the coenzyme. The corresponding dehydro- genase for L-galactitol l-phosphate was implicated in 1955 (25) followed by isolation of a sorbitol 6-phosphate dehydrogenase in 1962 (26). The function of these dehydrogenases in relation to carbohydrate metabolism remained obscure to the investi- gators (26) since the phosphorylative mechanism was not definitively established. The resolution to this problem began in 1964 when an attempt to ascertain the existence of a kinase for sialic acids in bacteria resulted in the serendipitous discovery of a novel phosphotransferase system in Escherichia ggli (27). In 1971 this system was further characterized by Roseman and coworkers (28-30). The system catalyzed the phosphorylation of hexoses utilizing phosphoenolpyruvate (PEP) as the phosphoryl donor and was consequently named the phosphoenolpyruvate-dependent phosphotransferase system (PTS). Concurrent with Roseman's investigations have been the sugar transport studies of Anderson and associates (31-34), Kaback and associates (35-37), Kornberg and associates (38-42), Hengstengstenberg and associates (43-46), and many others. Since many reviews on sugar transport have recently appeared (47-52), only the salient aspects will be reviewed here so 5 that a relationship of transport with bacterial hexitol metabolism may become apparent. The significance of the phosphoenolpyruvate-dependent sugar phosphotransferase system to both gram negative and gram positive bacteria is the fact that it simultaneously trans- ports and phosphorylates sugars. There are many sundry aspects to the process by which this occurs, contingent upon the organism investigated, but the basic mechanism is con- stituted by a two-step reaction sequence (48): +2 E1, M3 PEP + HPr _ Pyruvate + E~HPr +- I E11, Mg+2 BvHPr + Sugar A Sugar-P + Pyruvate PEP + Sugar ‘ Sugar-P + Pyruvate 7 The above reaction delineates the assignment of three proteins, HPr, EI’ and EII’ which effect the transfer of the phosphoryl group from phosphoenolpyruvate (PEP) to the carbohydrate. The transport of the phosphorylated carbohydrate is consum- mated by the enzyme II (EII) component. The participation of other proteins elicits variations of the reaction schematic, but only those catalysts exhibited shall be expounded upon. HPr: In E. 2211 HPr has been purified to homogeneity and the molecular weight determined to be approximately 9,500 (29). It is committed to the transfer of the phosphoryl group from a phosphoenzyme I intermediate (53) to EII' Conse- quently HPr is not a true enzyme or substrate. The amino acid 6 composition reveals that there are two moles of histidine per mole of HPr. Stoichiometric studies demonstrated that one 32F is transferred from 32P-enolpyruvate to one mole mole of of HPr and that the phosphate is bound to the N-l position of one of the histidine imidazole rings (48). HPr is a constitutive protein and is located in the periplasmic space (54). In Aerobacter aerogenes HPr has been partially purified and appears to be similar to HPr in E. 3911 (55). Recently the purification of an inducible protein has been reported which supplants HPr during the phosphorylation of low con- centrations of fructose (31, 56). It has been designated FTP and functions with the inducible EII for fructose (55). Both FTP and HPr are soluble proteins. Enzyme I (E1): EI’ like HPr, is a soluble and con- stitutive protein. It catalyzes the reversible phosphoryla- tion of HPr with PEP. In E. 2911 partially purified EI has been shown to be irreversibly inactivated with N-ethylmalaimide (28) or with vinylglycolate (57). Inhibition by sulfhydryl reagents such as mecuribenzoate or dithionitrobenzoic acid can be reversed by 2-mercaptoethanol, glutathion, or cystein (28).. This implies that sulfur-containing amino acids are present and may function in the active center of EI' In A, aerogenes EI has been partially purified to demonstrate its participation in the PTS system, but has not been charac- terized (56). As with HPr, mutants that lack EI exhibit pleiotrOpism since they are incompetent in the utilization of various carbohydrates. Pleiotropic mutants have been isolated 7 from E, aerogenes (58), E. coli (59), Salmonella typhimurium (60), and Staphylococcus aureus (61). The Salmonella typhimurium mutants cannot ferment D-glucose, D-mannose, D- fructose, maltose, melibiose, N-acetylglucosamine, and the polyols D-glucitol, mannitol, and glycerol. The Staphylococcus aureus mutants are incapable of transporting sucrose, maltose, lactose, trehalose, D-galactose, D-fructose, D-ribose, and the hexitol mannitol. The mutants of E. 3913 are defective in the utilization of D-glucose, D-fructose, D-mannose, D- galactose, maltose, lactose, melibiose, succinate, and the polyols mannitol and glycerol. The E, aerogenes mutants can- not transport D-glucose, D-fructose, D-mannose, and the hexitols D-glucitol, and mannitol. As evidenced by the listed compounds, not all of the pleiotrOpic mutants are capable of dissimilating glycerol or succinate which are not phosphorylated by the PTS. However, exogenously furnished cyclic AMP restores the capahity to flourish on these sub- stances, thus implying the existence of a regulatory respons- ibility for the PTS (62). Credible control mechanisms may be catabolite repression (63, 64), transient repression (65), and inducer exclusion (66). Enzyme II (EII): In E, aerogenes the constitutive EII's catalyze irreversible phosphorylation of D-fructose (31), and apparently D-mannose (55), D-glucose (55), and L-sorbose (34). However, the preponderant EII's are inducible sugar- specific proteins which are integrally associated with the membrane (50). A schematic model for the mechanism of action 8 of EI is delineated as a contiguous membrane constituent I which complexes with both HPr and EI such that the external sugar becomes phosphorylated causing a vectorial passage of the product through a diffusion barrier (50). It has been substantiated that membrane vesicles of E. 291; are contingent upon PEP in order to take up sugars as their phosphorylated derivatives and that treatment of the vesicles with phospholipase D inhibits vectorial phosphorylation of a- methylglucoside by phosphatidylglycerol hydrolysis (35, 36, 50). Furthermore, accumulation of sugar phosphate esters in whole cells has been attributed to the PTS (61, 67, 68, 69, 70, 71, 72, 32, 73, 74, 59, 75, 76). The constitutive EII of E, coli has been solubilized into three basic components: two pro- teins and phosphatidyl glycerol (30). One of the proteins, designated II-A, has been further resolved into three proteins which ascribe phosphorylation specificity to D-glucose, D- fructose, and D-mannose respectively. The remaining protein, II-B, has not been fractionated further but the molecular weight was determined to be 36,000 (30). The II-B protein functions as a ligand between one of the II-A proteins and phosphatidylglycerol thereby constructing the active complex. The specificity of phosphorylation usually occurs at C-6 (35). However, D-fructose (31) and L-sorbose (34) are phosphorylated at C-1 in E, aerogenes. In E. coli D-fructose is phosphory- lated at 0-1 at low concentrations and at C-1 and C-6 at high concentrations of the sugar (40). 9 The distribution of the PTS appears to be restricted to the anaerobic and facultatively anaerobic bacteria with the apparent exception of Arthrobacter pyridinolis (77). Most investigations have been concerned with the facultative anaerobes. Among the other organisms studied besides those mentioned, have been Bacillus mggaterium (78), Bacillus cereus (78), Achromobacter parvulus (78), Lactobacillus arabinosus (27), Streptococus lactis (83), Bacillus subtilis (78, 79, 35, 54, 80), Clostridium perfringes (69), Clostridium thermocellum (81, 82), and Corynebacterium ulcerans (78). The majority of these organisms have also been exam- ined in regard to the metabolism of polyols. However, the first resolute endeavor to associate hexitol catabolism with the PTS was conducted by Tanaka and coworkers in 1967 (58, 83). It was shown that an E, ggli_mutant which failed to grow on mannitol was deficient in an apparently inducible EII for mannitol. Further work showed that a mutant in either EI or HPr also failed to grow on mannitol but revertants of these components regained the wild type growth characteristics (58). In 1971, Kelker and Anderson (33) established the pathway of D-glucitol metabolism in E. aerogenes. The results were con- sonant with growth on D-glucitol as being dependent upon the PTS and militated against the initiation of glucitol metabol- ism by a hexitol dehydrogenase. A partially purified glucitol 6-phosphate dehydrogenase was used in the identifi- cation of the PTS-catalyzed reaction product, a type of 10 experiment omitted by Tanaka in his studies with E, aerogenes and E. coli. Corroborative genetic evidence provided sub- stantiation of the glucitol pathway where EI and glucitol 6-phosphate dehydrogenase mutants were unable to grow on glucitol whereas the corresponding revertants could be ex- pressed on the hexitol. The investigation of hexitol metabolism in E, aerogenes is extended in this thesis to the initiation of galactitol metabolism. As with D-glucitol catabolism, it might be expected that galactitol is also transported by the PTS in lieu of dissimilation by a galactitol dehydrogenase (85). MATERIALS AND METHODS Bacterial Strains A uracil auxotroph of Aerobacter aerogenes PRL-R3 was selected for the investigation of galactitol catabolism. The uracil genetic marker of the organism was utilized in order to detect contaminations and to be consistent with any future use of this strain as the prototroph for mutant analy- sis of the pathway. Media All media used for growth of uracil auxotrophs was supplemented with 0.005% (w/v) of uracil. Uracil and sugars were autoclaved separately from the rest of the media con- stituents. The final pH of all the media was adjusted to 7.0 before sterilization in a Castle autoclave for 15 minutes at 121°C. Mineral Medium The medium was prepared by dilution, with distilled water, of concentrated stock solutions. The constituents and their final concentrations (w/v) are listed below: 0.71% NaZHPO 0.15% KH PO 0.0005% 4’ 2 4’ FeSO4-7H20, and 0.5% of the specified sugar. 0.3% (NH4)2804, 0.01% MgSOA, 11 12 Nutrient Broth Medium Two procedures were used for preparing this medium. The first method involved dissolving 5.0 g of Bactopeptone (Difco) and 3.0 g of Bacto beef extract (Difco) in 1.0 L of distilled water. The second procedure consisted of dissolving 8.0 g Bacto nutrient broth (Difco) in 1.0 L of distilled water . Nutrient Agar Medium All strains were kept viable by transferring them to a new nutrient agar slant every three months and stored under refrigeration. The medium was prepared by dissolving 23 g of Bacto nutrient agar (Difco) in 1.0 L of distilled water. Growth of Cultures Cells utilized for enzyme purification were routinely grown in Fernbach flasks containing 1.0 L quantities of mineral medium. The cultures were incubated at 30°C on a gyrorotary shaker set at speed 3. Cells were also grown in 16 liter and 40 liter quantities of mineral media in Pyrex carboys under the same conditions except that aeration was achieved by sparging with air filtered through a cotton- containing flask. Media contained in Fernbach flasks were usually inoculated with 2.0 ml of an overnight nutrient broth culture whereas media in glass carboys were inoculated with 1.0 liter of an overnight culture grown in mineral medium. 13 Preparation of Cell Extracts All procedures were performed at 0°C to 4°C. Bacterial cultures grown in Fernbach flasks were harvested by centrifugation at 16,000 x g in a Sorvall refrigerated centrifuge for 15 minutes. Cells grown in glass carboys were harvested using a Sharples centrifuge at room temperature at a rate of 2.0 liters per minute. The cells were washed twice by resuspension in 200 ml of 0.85 percent NaCl using a large gauge needle and syringe. The suspended cells were harvested and resuspended in 0.02 M Tris-HCl buffer (pH 7.5) containing 0.028 M 2-mercaptoethanol at a concentration of 10 ml of buffer per liter of cells. A maximum of 35 m1 of suspension was then disrupted by ultrasonic vibration for 30 minutes in a 10-kHz, 250-watt Raytheon sonicator equipped with an ice- water cooling jacket. Cellular debris was removed from the broken-cell preparation by centrifugation for 15 minutes at 45,000 x g,in a refrigerated Sorvall centrifuge. The result- ing supernatant was carefully removed with a pipette so as not to recover any of the sedimented debris. This super- natant was then used as the crude extract. Qualitative Analyses Dephogphorylation of Sugar Phosphates Chemically and enzymatically synthesized sugar phos- phates were dephosphorylated using commercially prepared Escherichia coli alkaline phosphatase which had been desalted 14 on a Biogel P-6 column (0.5 cm x 6 cm). Controls missing either enzyme or substrate were also used in order to vali- date the reaction and account for any background in the GLC analysis. The hydrolysis was carried out in a 12-ml ground- glass, stoppered conical centrifuge tube containing the following constituents in a final volume of 0.90 ml: 1 to 3 mg of sample, 0.85 mmole of Na2C03-acetic acid buffer at pH 8.0, and 12 units of alkaline phosphatase. The reaction mixture was warmed to 30°C and then the hydrolysis was in- itiated by the addition of enzyme. The reaction was monitored for 6.5 hours by assaying for P1 liberation using the modi- fied method of Fiske and SubbaRow (86). When the production of Pi ceased, the reaction vessels were heated at 100°C for 7 minutes and then cooled on ice. The precipitate was then removed by centrifugation at 12,000 x g for 10 minutes at 4°C. The supernatant was removed and treated with 20 ml of Dowex 50W-X8 H+ to eliminate bicarbonate as C02. The resin was removed by suction filtration and then washed with 10 ml of triply distilled water. The filtrates were combined in a tared glass-stoppered 50-ml centrifuge tube and evaporated to dryness under reduced pressure at 30°C using a rotary evapora- tor. The centrifuge tube was then reweighed and the sample dissolved in triply distilled water to yield a final sugar concentration of 80 ug/ul. Gas-Liquid Chromatography of the Polyacetate Derivatives of Various‘POlyols The polyacetate derivatives of sugar alcohols were 15 prepared and analyzed by a slight modification of the proce- dure used by Jones and Albersheim (87). A 50-p1 sample of the dephosphorylated sugar was transferred to a 300-pl Kontes microflex tube and evaporated to dryness at 60°C. A few crystals of sodium acetate were then added (88) followed by 50 p1 of fresh anhydrous acetic anhydride. The reaction vessels were flushed with nitrogen, capped, and placed in a metal heating block at 121°C for one hour. The samples were then allowed to cool to room temperature and then were analyzed with a Perkin Elmer 900 Gas Chromatograph equipped with an Autolab System 4 computer. The standard mixture of polyols and all other sugars tested were prepared in the same way ex- cept the concentrations were 40 pg/50 pl of acetic anhydride. The instrument parameters and settings of the gas chromato- 40.0 ml/ graph were as follows: helium carrier flow rate minute, manifold temperature = 250°C, injector temperature = 265°C, program rate = lOC/minute, initial time = 4 minutes, final time = 12 minutes, program range = 130 to 180°C, attenuation = 3600 to 60. The glass columns employed were 6 feet x 2mm internal diameter and contained a liquid phase of 0.2% polyethyleneglycol adipate, 0.2% polyethylene glycol succinate, and 0.4% XF-1150 silicon oil on a solid support of Gas-Chrom P (100 to 200 mesh). A second identical column was also used as a balancing column in order to correct for column bleeding. l6 Gas-Liquid Chromatography of Trimethylsilyl Sugar Derivatives Sugars were derivatized to their trimethylsilyl (TMS) forms by the method of Musick and Wells (89). All sugar phosphates were converted to their acid forms by treatment with Dowex 50W-X8 H+ before derivatization. A 1.0 mg sample of an ethanol mixture of the sample was dried at 30°C under nitrogen in a 0.5-dram.via1. To the dried sample was added 80 p1 of anhydrous pyridine, 200 p1 of bis-trimethylsilyl- acetamide (BSA), and 20 pl of trimethylchlorosilane (TMCS). The reaction vessels were then flushed with nitrogen, sealed with teflon screw caps, and heated at 85°C for 45 minutes. The derivatized samples were allowed to cool to room.temperature after which they were immediately analyzed by gas chromatography. Gas chromatography was performed with a Hewlit Packard High Efficiency Gas Chromatograph utilizing a 6 foot x 2 mm internal diameter glass column containing 30/0 SE-3O on a solid support of Supelcoport 80 to 100 mesh. Instrumental settings were as follows: Nitrogen flow rate 45.5 mllminute, Flash heater temperature = 310°C, Detector temperature = 310°C, temperature program rate = 1°C/minute. The temperature program range for the TMS sugars and TMS sugar phosphates was 150°C to 200°C and 200°C to 250°C respectively unless otherwise stated. Chart recordings were made at 1 cm/2 minutes. Combined Gas-Liqgid Chromatography—-Mass Spectrometry. The TMS derivatives of sugars and sugar phosphates were l7 synthesized as described previously. Mass spectrometry of samples was performed immediately after the TMS derivitiza- tion in order to avoid any stability problems of the deriva- tives. Mass spectra were recorded at 70 eV with an LKB 9000 mass spectrometer-gas chromatograph. The relative abundance of fragments was displayed as bar graphs by means of a data aquisition and processing program utilizing a pdp8/e computer (90). The instrument settings were as follows: Source temperature = 290°C, Accelerating voltage = 3.5 kV, Ionizing current = 60 FA. The samples were gas chromatographed through a 4 foot, 2 mm internal diameter glass coiled column packed with SP-2100 on Supelcoport 100 to 120 mesh. The temperature ranges used were as described before for the TMS derivatives and the program rates were carried out at 3°Clminute. Determination of Acid Lability Acid lability was determined according to the method of Wolff and Kaplan (91). The samples, usually 10 pmoles, were first converted to their acid forms by treatment with an excess of Dowex 50W-X8 H+. The resin was washed with an equal volume of distilled water and the collected filtrates were combined. The resulting solution was then adjusted to 0.5 m1. With samples from the periodate oxidation, the excess perio- date was destroyed by the addition of 1 ml of ethylene glycol per 5'ml of sample. The solutions were then transferred to test tubes (1.1 cm x 10 cm) which were placed in a water bath thermostated at 95°C. The solutions were immediately made 18 1 N in HCl by addition of 0.5 m1 of 2 N HCl and then the test tubes covered with glass marbles. The course of hydrolysis was monitored by removing 50 p1 to 200 pl aliquots and assay- ing for Pi' Periodate Oxidation Analytical periodate oxidation was conducted in the following manner, observing the considerations of Guthrie (92) and Bobbitt (93). To the reaction vessel, a 25-ml volumetric flask, was added 12 m1 of a solution containing 45 to 66 pmoles of the salt form of the sugar phosphate. A volumetric pipette was then used to add 2.725 mmoles of Nan4 contained in 10.00 m1 of distilled water. The solution was rapidly diluted to volume and the reaction vessel immediately sealed with a rubber stopped fitted with inlet and exit ports. A blank flask without sugar was treated in the same manner. Dry COZ-free nitrogen was then continuously flushed through the inlet port allowing any evolved CO2 to be displaced through the exit port into a collection flask containing 2.5 M NaOH such that the gases slowly bubbled into the alkaline solution. The reaction vessels were placed in a thermostated water bath at 20°C and the water bath was then covered with aluminum foil in order to protect the reaction mixture from light. It was found in a preliminary experiment under the same reaction conditions that fructose 6-phosphate consumed the expected quantity of periodate within 12 hours with the consumption of the first two molecular equivalents of periodate within one 19 hour. The periodate oxidations were therefore allowed to proceed for 12 hours. At the end of the oxidation procedure, the reaction solutions were analyzed for all products. Samples which could not be immediately analyzed were freed of periodate by the addition of ethylene glycol or sodium bi- sulfite followed by freezing at -20°C. Enzyme Assays In each of the following assays the enzyme concerned was present in rate-limiting quantities such that the enzyme concentration would be proportional to the rate of the reaction catalyzed. Furthermore, each assay was dependent upon the presence of the specific enzyme and substrate necessary to initiate the reactions involved. All assays were monitored for either the reduction of NAD+ or the oxidation of NADH (except where stated) at 340 nm in micro-quartz cuvettes with l-cm light paths using a Gilford absorbance-recording spectro- photometer. A conversion factor of 0.41 absorbance units per 0.01 pmole of NADH in 0.15 ml was used in determining the number of units of enzyme. Assays for PTS Components HPr, and E?§1, activities were assayed using a EI' modification of the "mannitol continuous assay" (94). Each activity was determined in the presence of saturating con- centrations of the other two activities. The assay consisted of the following components in a final volume of 0.15 ml: 20 12 pmoles of 2-amino-2-methyl-l,3-propanediol buffer (pH 9.0), 0.2 pmole of NADP+, 0.2 pmole of NAD+, 0.01 pmole of MgClz, 1.0 pmole of PEP, 1.0 pmole of D-mannitol, 0.57 pmole of 2- mercaptoethanol, 2 units of phosphoglucoisomerase, 1 unit of D-glucose 6-phosphate dehydrogenase, 1 unit of mannitol l-phosphate dehydrogenase (partially purified as described later), and variable amounts of EI' EII’ or HPr depending upon which was being assayed. The reaction could be initiated with either PEP or one of the PTS constituents but initiation with PEP was the routine procedure. An absorbance change of 0.41 was equivalent of 0.01 pmole of NADPH formed since the EII preparation still contained NADH oxidase activity (The NADH oxidase activity was not eliminated by inhibitors such as potassium.cyanide or Quinacrine without incurring inhibition of the coupling enzymes.) In addition to the NADH oxidase activity, other competing reactions such as pyruvate kinase and phosphatases prevented assaying for the PTS constituents in crude extracts and ammonium sulfate fractionations. A unit of activity was defined as that quantity of enzyme required to produce 1 pmole of NADPH per minute. Mannitol l-Phosphate Dehydrogenase Assay Since mannitol l-phosphate was not readily available, this dehydrogenase was assayed in the reverse direction using fructose 6-phosphate and observing the decrease in absorbance at 340 nm. Interfering D-sorbitol 6-phosphate dehydrogenase was not present with the mannitol 1-phosphate dehydrogenase 21 since cells grown on mannitol due not induce the D-sorbitol 6-phosphate dehydrogenase (26). The assay consisted of the following components in a final volume of 0.15 ml: 10 pmoles of glycylglycine buffer (pH 7.5), 0.05 pmole of NADH, 1.0 pmole of fructose 6-phosphate, and limiting amounts of en- zyme. The reaction was initiated with the addition of sub- strate. The rate of the reaction was measured within the first two to three minutes due to the gradual decrease in the catalysis rate after this period as was also observed by Liss, et a1. (26). After subtracting the background rate contributed by NADH oxidase, if present, the amount of mannitol l-phosphate dehydrogenase activity was calculated. A unit of activity was defined as the amount of enzyme re- quired to catalyze oxidation of l pmole of NADH per minute under the reaction conditions employed. L-Galactitol l-Phosphate Dehydrogenase Assay L-Galactitol l-phosphate dehydrogenase was assayed using chemically synthesized L-galactitol l-phosphate as the substrate. The assay consisted of the following amounts of components in a final volume of 0.15 ml: 8 pmoles of Tris- HCl buffer (pH 9.0), 0.57 pmole of 2-mercaptoethanol, 1.0 pmole of MgClz, 0.5 pmole of NAD+, 0.5 pmole of L-galactitol l-phosphate, and limiting amounts of enzyme. The reaction was initiated by the addition of substrate. Galactitol could not substitute for L-galactitol l-phosphate as a substrate for the enzyme as shown in Table (1). Table (2) shows that 22 TABLE 1. Validity of the L-galactitol 1-phosphate dehydro- genase assay and activity with galactitol as substrate. Assay Description Absorbance Crude L-Galactitol Galactitol ggange per extract l-phosphate (pmoles) nu e (p1) (pmoles) Complete 5.0 1.2 0.00 2.4 x 10"3 (see text) Complete plus 10 1.2 0.00 4.9 x 10'3 2 x (enzyme) Complete minus L-galactitol 5.0 0.00 0.00 0.00 1-phosphate Complete minus enzyme 0.0 1.2 0.00 0.00 Complete plus galactitol 5.0 0.00 1.0 0.00 minus L- galactitol l-phosphate Complete plus galactitol 0.0 0.00 1.0 0.00 minus L- galactitol -phosphate and enzyme Complete plus _3 galactitol 5.0 1.2 1.0 2.4 x 10 23 TABLE 2. Cofactor specificity of L-galactitol l-phosphate dehydrogenase and apparent inducibility Assay Description Absorbance NADP+ NAD+ Galactitol Glucose 3233?: per (pmoles) (pmoles) extract extract (mg) (mg) Complete plus NAD 0.0 0.1 0.061 0.00 0.012 minus NADP (see text) Complete + plus NADP+ 0.1 0.0 0.061 0.00 0.00 minus NAD Complete plus NAD 0.0 0.1 0.00 0.074 0.00 minus NADP Complete + plus NADP+ 0.1 0.0 0.00 0.074 0.00 minus NAD 24 the enzyme was specific for NAD+ as the cofactor and does not use NADP+ for either L-galactitol l-phosphate or galactitol. Furthermore, the enzyme was apparently inducible since no activity could be seen in crude extracts of cells grown on D-glucose (Table 2). A unit of activity was defined as the amount of enzyme necessary to catalyze the reduction of 1 pmole of NAD+ per minute under the conditions described. Phosphoglycolate Phosphatase Assay Phosphoglycolate phosphatase activity was assayed according to the method of Christeller (95). The assay mix- ture contained the following constituents in a final volume of 0.5 ml: 20 pmoles of cacodylate buffer (pH 6.3), 25 pmoles of sodium citrate, 2 pmoles of MgClz, l pmole of phos- phoglycolate, and limiting amounts of enzyme. The reaction, equilibrated at 30°C in a thermostated water bath, was in- itiated by the addition of enzyme and terminated after 10 minutes by the addition of 0.2 ml of 10% TCA. An assay blank containing the complete reaction mixture incubated for zero time was also performed in order to account for background quantities of Pi' After centrifugation of the denatured protein in a table-top centrifuge at room temperature for 10 minutes, a 0.5-ml sample of the supernatant was assayed for release of Pi by the modified method of Fiske and SubbaRow (86) as described previously. One unit of activity was de- fined as the quantity of enzyme required to hydrolyze one pmole of phosphoglycolate into Pi and glycolate per minute at 30°C. 25 Partial PurificatiOn and Prgparation of‘Enzymes All of the following operations were performed at 0 to 5°C unless otherwise stated. All centrifugations except the ultracentrifugation were accomplished with a Servall refrig- I erated centrifuge. In the anion exchange and gel filtration steps only the fractions exhibiting the highest specific activity were pooled. mtl II gltl II Preparation of E and E E?§1 and E§%tl were prepared from crude extracts of bacteria grown on D-mannitol and galactitol respectively. The membrane fraction containing the EII activity was isolated by centrifugation at 100,000 x g for four hours in a Spinco L-2 refrigerated ultracentrifuge. The resulting pellet contain- ing the EII activity was suspended in 20 to 30 m1 of 0.02 M Tris-HCl buffer (pH 7.5) containing 0.028 M 2-mercaptoethanol by use of a syringe equipped with a 22 gauge needle. The supernatant from the ultracentrifugation was retained for purification of the soluble enzymes. The suspended membrane vesicles were sonicated for 20 minutes under the same condi- tions used for the preparation of the cell extract and then were ultracentrifuged as before at 100,000 x q, The resulting membrane pellet was allowed to drain free from the supernatant and was then resuspended in a minimum volume of the same buffer. This suspension was then applied to a small column (1.2 cm x 9 cm) of Biogel P-200 and the membrane fraction was 26 eluted in the void volume with the same buffer. The eluted membrane vesicles were then concentrated by ultracentrifuga- tion as performed before and pellet was resuspended in a minimum.volume of the same buffer after removal of the super- natant. This final suspension of membrane vesicles was then used as prepared EII' All EI and HPr activities were absent in this preparation as well as 80% of the NADH oxidase activity. When stored on ice the activity was stabile for approximately one week whereas freezing and unfreezing the preparation in- activated all of the EII activity. Partial Purification of EI The retained supernatant from the EII preparation was diluted to approximately 10 mg of protein per ml using the same buffer. The solution was then brought to 35% ammonium sulfate saturation by the slow addition of powdered enzyme- grade (NH4)2804 with constant stirring. The mixture was allowed to equilibrate for one hour followed by centrifugation for 15 minutes at 30,000 x g. The precipitate was discarded and the supernatant volume measured. The appropriate amount, calculated from a nomograph (96), of ammonium.sulfate was added to the solution in the same manner as before to increase the saturation to 70%. The precipitate, containing the EI activity, was centrifuged at 12,000 x q for 15 minutes. The resulting supernatant was saved for isolation of HPr activity and the precipitate was dissolved in a minimum of 0.02 M potassium.phosphate buffer (pH 8.0) containing 0.028 M 27 2-mercaptoethanol, and 10% (v/v) glycerol. The solution was desalted, using the same buffer, on a Biogel P-6 column (2.1 cm x 30 cm) eluting the EI activity in the void volume. The activity was pooled and diluted with the same buffer to a protein concentration of 2 mg/ml. The protein solution was then applied to a DEAE cellulose column (1.7 cm x 17 cm) which had been equilibrated with the same buffer. The column was washed with three column volumes of the same buffer and then the E1 was eluted with a 0.0 to 0.5 M linear KCl gradient developed in 10 column volumes of the same buffer. The elution profile of EI activity and protein is shown in figure (1). The activity was pooled and dialyzed for 24 hours against two changes of 20 volumes each of the same buffer. The dialyzed EI activity was then applied to a second DEAE cellulose column (1.7 cm x 17 cm) which had been equilibrated with the same buffer containing 0.1 M KCl. The column was washed with the same buffer containing 0.1 M KCl and then a 0.1 to 0.3 M linear KCl gradient in 10 column volumes of the same buffer was used to elute the EI activity. Figure (2) shows the elution of EI and protein. The EI activity was pooled and dialyzed in the same manner as before. To the dialyzed solution was added 0.10 M acetic acid drop-wise with continuous stirring until the pH had been lowered to 4.5. The turbid mixture was then centrifuged at 12,000 x‘g for 15 minutes. The resulting supernatant was poured off and immediately re- adjusted to pH 8.0 by the drop-wise addition of 0.5 M NaOH with constant stirring. The resulting solution was then concentrated 28 by pressure dialysis using an UM-lO ultrafilter membrane (Amicon) at 40 psi of nitrogen. The concentrate was then applied to a Biogel P-200 column (2.1 cm x 88 cm) and the EI activity eluted with the same buffer. Figure (3) shows the elution profile of E activity and protein. The fractions I containing activity were then pooled and frozen at -20°C. This preparation of EI was stabile for over six months. Table (3) summarizes the purification data. Calcium phosphate gel and hydroxylapetite were also used for the purification of EI but did not yield any further purification. Partial Purification of HPr The 70% ammonium sulfate saturated supernatant from the EI preparation was adjusted to 100% saturation in the same manner used in the previous ammonium sulfate fractionations. After the solution had equilibrated for one hour the precipi- tate was isolated by centrifugation at 12,000 x q for 15 minutes. The pellet, containing the HPr activity, was allowed to drain free of the supernatant and was then dissolved in a minimum of 0.02 M potassium.phosphate buffer at pH 7.5. The solution was then desalted by molecular seiving through a column of Biogel P-6 (2.1 cm x 30 cm), eluting HPr in the void volume using the same buffer. The fractions containing activity were pooled and diluted with the same buffer to a protein concentration of 2 mg/ml. Three volumes of DEAE cellulose, equilibrated in the same buffer, were then added batchwise to the solution with continuous stirring. The pH was 29 FIGURE 1. DEAE cellulose chromatography I of the 35 to 70 percent ammonium sulfate fraction from the Biogel P-6 protein pool. o—————o Protein. o—————o Mannitol l-Phosphate Dehydrogenase. o—————o Enzyme I. Calculated KCl gradient. 30 Protein Absorbance at 220 nM " Q 3 E 3 '3‘ {5' m E Molarity of KCI «:1 i I O " I _, 0 II - 0 ll L 5 S j! a nu -ugu/sa|omr/ «cuobomlqap d-Houuuaw , a 3 Immgu/Iqowr/ 1 Mug 31 FIGURE 2. DEAE cellulose chromatography II of the dialyzed EI pool from DEAE cellulose chroma- tography I. I———I EI activity. H Protein. ...... Calculated KCl gradient. 32 0.0 00m M fixo.1 ”w "a I “N mu 8 S no mm mm m.olr V .1 so 70 no W n.0L. 10X AIIHVTOW OH.O ma.o. ON.OAI mm.o.r on.o.fa 0mg ZOHHUwuoc map was mcofiufipcou can .mo>Hum>Huov ouwuoomhaom uaosu mm maomaoa mace no How .m shaman cm 0: on $832 om 2 o D F r ‘ ‘ ‘ ‘ ‘ 1" 1031533): b 0d 38 801.0 31.30 'IOIIIDV'IV’D t 'IOLINNVN ‘1 TOIIW‘: 'IOLIHVHV "‘1 101.100.! -—-==: 'IOIINWVHH -—-=: 'IOLISONI --=: 'IOIIHHOS -—-=: H- b b )- 4|- 74 TABLE 10. Retention times of the polyacetate derivatives of the polyols analyzed in Figure (9). Sugar Retention Time Alcohol in minutes Rhamnitol 16.2 Fucitol 17.9 Arabitol 24.3 Xylitol 29.9 Mannitol 41.4 Galactitol 44.4 Sorbitol 46.6 (D-Glucitol) Inositol 50.7 75 and commercial galactitol was performed (presented in the following section). The enzymatically synthesized galactitol phosphate was analyzed by gas-liquid chromatography as the TMS derivative before and after dephosphorylation. Figure (10) shows that the TMS derivatives of chemically synthesized L- galactitol l-phosphate and the enzymatically synthesized compound exhibited single peaks and had identical retention times, 44.9 minutes. This also indicated that the compounds were pure by GLC criteria. The dephosphorylation of the enzymatically synthesized galactitol phosphate and its con- version to the TMS derivative yielded a compound which clearly chromatographed apart from the TMS derivative of the phos- phorylated form under identical GLC conditions. But this compound had nearly the same retention time as the TMS derivative of commercial galactitol, 16.9 minutes and 17.0 minutes respectively (Figure 11). When the enzymatically syn- thesized galactitol phosphate was dephosphorylated and derivatized to the hexaacetate form it exhibited a single peak of essentially the same retention time as commercial galactitol, 44.3 minutes and 44.4 minutes respectively (Figure 12). The data strongly imply, therefore, that the product isolated from the PTS-catalyzed reaction was galactitol phosphate. Mass Spectrometry of the TMS Derivative of Chemically Syn- thesized L-Galactitol I-Phosphate After Dgphosphorylation. Should the mass spectrum.of enzymatically synthesized 76 m m 3 fl s 5 Di 0 E5 13 J ,3. 1‘5 50 MINUTES 15 0 III U) 5 a :3 93‘ 5' Eu 8 1+5 50 {5 6 'MINUTES Figure 10. GLC analysis of the TMS derivative of enzymatically synthesized galactitol phosphate. TMS derivatization and general GLC conditions are described in Materials and Methods. The temperature program rate was 1°C/min. from 150°C to 250°C. The attenuation was 1600 A. TMS derivative of chemically synthesized L- galactitol 1-phosphate. B. TMS derivative of enzymatically synthesized galactitol phosphate. 77 F‘I’J CD 5 94 CD 5&3 8 E' m 5" W D 20 MINUTES 10 O [-1-] g 8 (D 2 K O E; It: E" [I] Q 0 20 mums 10 . Figure 11. GLC of the TMS derivative of enzymatically synthesized galactitol phosphate after dephosphorylation. The GLC conditions were identical to those described for Figure (10). The dephosphorylation is described in Materials and Methods. A. TMS derivative of commercial galactitol. B. TMS derivative of enzymatically synthesized galactitol phosphate after dephosphorylation. 78 It: 3 8 U) :3 as 0 E3 1:: s . J 0 10 20mm,TES 50 1+0 50 [:1] £2 0 9.. to ii a: O E; [:1 E: Q L J . .. - - L—: 0 . 10 eomms 30 1+0 50 Figure 12. GLC of the hexaacetate derivative of enzymatically syn- thesized galactitol phosphate after dephosphorylation. The GLC conditions and derivatization procedure were identical to those described in Figure (9). Dephosphor- ylation was performed as described in Materials and Methods. A. B. Standard galactitol hexaacetate. Hexaacetate derivative of enzymatically synthe- sized galactitol phosphate after dephosphory- lation. 79 galactitol phosphate be congruent with that of chemically synthesized galactitol phosphate then mass spectroscopy of the enzymatically synthesized compound after dephosphoryla- tion would not be necessitated. Therefore the mass spectrum of the TMS derivative of only the chemically and not the enzymatically synthesized galactitol phosphate after de- phosphorylation was obtained as one criterion for establish- ing the chemically synthesized compound as an L-galactitol l-phosphate standard. I The chemically synthesized galactitol phosphate was dephosphorylated as before and converted to the TMS derivative as described in Materials and Methods. Figure (13) shows that the TMS derivative yielded a single peak which chroma- tographed as standard TMS-galactitol did. Figure (14) shows the mass spectra of each of the three TMS-galactose isomers, which agree with the spectra obtained by DeJongh, et al. (116). Figure (15) reveals the mass spectra of the TMS-dephosphoryl- ated compound and the TMS-galactitol standard. Each of the three galactose isomers yielded a mass spectrum of different and salient E/E ratios than that by either the standard galactitol or the dephosphorylated compound. In all spectra the parent ions were not seen due to their low abundance. The data obtained were therefore consistent with the identifica- tion of chemically reduced D-galactose 6-phosphate as a hexitol phosphorylated at a primary alcohol position. 80 b h [”1 . 3 F3 53 g E g g g 18 MENUTES 6 6 B 3 F3 53 g E g 2 13 JL 18 MI] 1S 8 0 Figure 13. GLC of the TMS derivative of chemically synthesized L- galactitol l-phosphate after dephosphorylation. The GLC conditions and derivatization procedure are described in Materials and Methods. The temperature program rate was 1°C/min. from 150 to 200°C. A. TMS derivative of standard galactitol (attenua- tion ; 3200). B. TMS derivative of chemically synthesized L-galac- titol 1-phosphate after dephosphorylation (atten- uation = 160). FIGURE 14. 81 Mass spectrometry of the isomers of D- galactose (TMS) . The molecular weight of all isomes 13 540. Absence of parent ions was due to their low abundance. Pro- cedural details are described in Materials and Methods. A. Isomer number 1. B. Isomer number 2. C. Isomer number 3. 82 “N am w .VN- O}: oom one cos omm com 02 oow om. 8. on ...... a. .3 as a. E a. .8 m . .... 2. E ... -3 .. ... as .8 m Ammoz<¢>e u . 8.3.. 26-3: neat 30533.... omx a. row m o 2 02A 0): com one 00¢ 08 com omm com on. 8. on u bethhbnbph-bhhhphu-JAPHPLE>WM- t-PhthMFibxlbfipppk}. 1 E 8— V . .9... a. fl a. .8 n. 33.5. 5 o.» n 4 II T as... 2:05.311... .2 m 1 a Tom Ha... . at» z @355... .8 u u 56.. 26.3: 235 3951.3-.. omx m co... {E com o? co.» omm oom omm com om. 00. on ...... m .... a. . n 1 Tom u u. 1 . ., ......oetfit... .3 m . .0m .1... 2.5.. 3025...... ... .8 u . 5.3.. 26-3.. last 3053.319 ..x u. < E 2 8.1 FIGURE 15. 83 Mass spectrometry of chemically synthesized L-galactitol l-phosphate as the TMS derivative after dephosphorylation. Proce- dural details are described in Materials and Methods. Parent ions were absent due to their low abundancies. The molecular weight of galactitol (TMS)6 is 614. A. Standard galactitol (TMS)6. B. TMS derivative of chemically syn- thesized L-galactitol l-phosphate after dephosphorylation. 08 08 com o? 8.» .08 . om Pomw . . .08 h ......mmrrrhwfrrmm 5 4 E 4 a ... i Ed 1 v E on 5 2:8 3 4 E! s «2 fio.» w 4 tom . 0.x ,8 .m m E*b P b bop .Vb .P h *0? b - bommb b h oom OWN CON on“ 00" on 4? 4544 ”q >>>>>> afblh—thJ? - b by )5? fbk hmbthbbebfilfiLbfibEbhb-‘VDJ-DPFMj. .n 4 a. ..r. H a- H a. m a. ... .8 w. ... ... .3 n .8 m N W38<¢3uv .om M 8.. 5.3.. goo-ta 8...: 93:80.23 30553-9 u ...IL -00“ FIGURE 17 . (Continued . 91 NN 92 0mm 08 0mm 00m 03 03 on... com 0mm 000 on. .00.. on 1 m 1? .0m . .3 .8 Auu0x0v .0m 0~x 3.3. «00-3: 12.... 95:30:01. 30533-0 0 2 00. 5.. 08 08 0mm 0 m 03 03 08 000 0...... com on. 0.. 0m .... +5....4fi %..... . .....0 a: H. i-fls. Em. .- 5 ... .0w .3 .2353"! .3 .08 Au00zav .0m 09. 32.... «co-3a 10...: 31.80570 303.33.... 2 All." 31!“ IMLV13I IAILV'IIU A NONI-l.“ 93 summarizes the periodate oxidation data and is consistent with the identity of the compound as a hexitol phosphorylated at a primary alcohol position. Identification of the Phosphorylated Product as L-Galactitol l-Phosphate by'Enzymatic Oxidation to D-TagatoSe'6-Phospfiate. To determine the position of the phosphate function and the enantiomeric configuration of the PTS-catalyzed re- action product, it was necessary to oxidize the compound with L-galactitol l-phosphate dehydrogenase and then characterize the resulting product with respect to: GLC analysis, acid lability, periodate oxidation, and enantiomeric configuration. This characterization would also be expected to corroborate the identification of the hexitol moiety of the PTS-catalyzed reaction product as galactitol. Preparation and isolation of D-tagatose 61phosphate: The enzymatically synthesized galactitol phosphate was enzymatically oxidized to a ketohexose phosphate by using NAD+ as the cofactor and L—galactitol l-phosphate dehydro- genase as the catalyst. In order to regenerate NAD+ from the reduced form it was necessary to couple the oxidation re- action with a reduction reaction, i.e. pyruvate to lactate using lactate dehydrogenase. To insure complete conversion of the galactitol phosphate, an excess of pyruvate relative to galactitol phosphate and an excess of lactate dehydrogenase relative to galactitol phosphate dehydrogenase was used. In addition to these considerations, the reaction was buffered 94 TABLE 11. Periodate oxidation analysis of chemically syn- thesized L-galactitol l-phosphate. A sample of 65.0 pmoles was oxidized and the products pro- duced were analyzed as described in Materials and Methods. Product Total pmoles Determined Theoretical (formed from perio- of product molar ratio ratio of date oxidation) of product: product: hmutol hexitol phosphate phosphate I0; Consumption 265 4.08 4 Theoretical I0; Consumption 260 Formaldehyde formed 64.8 0.99 1 Theoretical value 65.0 Glycolate formed 0.0 0.00 0 Theoretical value 0.0 Phosphoglycolate fonmed 0.0 0.00 0 Theoretical value 0.0 Acid lability No Theoretical lability No Phosphoglycoaldehyde Formed 65.0 1.00 1 Theoretical value 65.0 Formate Formed 196 3.02 3 Theoretical value 195 Carbon dioxide formed 0. 0.01 0 00 Theoretical value 0. 95 mumsdmond-H Houwuowamw-A you «Epsom coaumpqxo ouwpowumm .QH ousmfim ..0 . 0am-0-mmw ..0 0.6 + .80.... massages.-. .ou.uou.mu-. :0000 000:0 ammo +. momwo . + + 0:... ..-0 -9. . v0. . v0. . v0. . :0. . - :0 m-w-oz - 00-0-: - mo-o-m - mo-m-m - . . 0a.. -0-m..0 ..0 0.0-0.. ..0 0.0-0.. ..0 ..-0 0.. - n n n n =0 oak-o-mwm cum-o-mmo 0n.-0-:0 . . . :0 mo mo 96 with phosphate. This was because the oxidation was carried out at pH 9.0 at which alkaline phosphatase activity, present with the partially purified galactitol phosphate dehydro- genase, is known to be inhibited by high phosphate concen- trations (118). The reaction mixture contained the following constituents in'a final volume of 3.10 ml: 20 pmoles of MgClz, 0.2 mmole of KZHPO4 (pH 9.0), 0.136 mmole of sodium.pyruvate, 2.95 units of partially purified galactitol phosphate de- hydrogenase, 56 units of lactate dehydrogenase which has been previously desalted on a small Biogel P-6 column, 50 pmoles of NAD+, and 84 pmoles of the enzymatically synthesized galactitol phosphate. The reaction mixture was warmed to 30°C in a stoppered test tube placed in a thermostated water bath. The reaction was initiated by the addition of galactitol phosphate. At specific time intervals, 5 pl to 30 pl was removed to small culture tubes and then heated for a few minutes at 90°C. After heating, the tubes were cooled on ice and then the denatured protein removed by centrifugation at 5,000 x g for 10 minutes at 0°C. The resulting supernatants were then assayed for ketohexose and galactitol phosphate. The course of the reaction is shown in Figure (19). When the presence of galactitol phosphate was no longer detectable the reaction mixture was heated for 10 minutes at 100°C and then cooled on ice. The precipitate was removed by centri- fugation at 10,000 x g for 15 minutes at 0°C. The supernatant was retained and the pellet resuspended in a minimum volume of distilled water in order to wash the product from the 97 denatured protein. The suspension was centrifuged as before and the supernatant combined with the first supernatant. The solution was then adjusted to pH 8.0 and the ketohexose phosphate product purified by the ion-exchange method of Khym (119). To a column (0.78 sq. cm. x 14 cm) containing Biorad AGl-XZ 01- was applied the adjusted solution. The column was washed with 100 ml of 0.001 M NHAOH to remove pyruvate, lac- tate, and cations. The column was then washed with 1.0 liter of 0.02 M NH4Cl containing 0.01 M K23407 in order to remove phosphate ions. The presence of borate serves to complex with the hydroxyl functions of the sugar molecule and form a negatively charged complex which will strongly bind to the resin. The column was then washed with 1.0 liter of 0.025 M 0H and 0.001 M K B 0 The pro- 4 4 2 4 7' duct was finally eluted with 2.0 liters of 0.025 M NHACl con- NH Cl containing 0.0025 M NH taining 0.0025 M NH 0H and 0.01 mM K B 0 The eluate was 4 2 4 7' collected in 10 m1 fractions using a column flow rate of 1.3 ml/ minute. A 2.0-ml aliquot was removed from the fractions and assayed for ketohexose. The elution profile of product is shown in Figure (20). The peak was asymmetrical but since it was found to consist of a single compound, the asymmetry may have been to abberations in the column packing or sample application. The fractions containing ketohexose were pooled into a large round bottom flask and then evaporated to 100 ml under reduced pressure at 30°C in a rotary evaporator. To FIGURE 19. 98 Enzymatic oxidation of enzymatically syn- thesized galactitol phosphate. The oxi- dation was performed using partially purified galactitol phosphate dehydrogenase as des- cribed in the text. O——O Galactitol phosphate H Ketohexose 99 j I 8 R HLVHJSOHJ ms smomi “NJ-OJ. 150 200 100 MINUTES 100 the concentrated solution was added an excess of Dowex 50W— X8 H+ in order to remove NH: and other cations. The resin was removed by suction filtration followed by a wash with one resin-volume of distilled water. The filtrate was then freed of HCl by evaporation to a syrup under reduced pressure at 30°C in a rotary evaporator. The resulting syrup was dis- solved in a minimum volume of distilled water and then tested for the absence of Pi and chloride ions. The solution was then freed of borate ions by the addition of anhydrous methanol and removal of the resulting methyl borate by evap- oration to dryness under reduced pressure at 30°C in a rotary evaporator. The methanol treatment was repeated three more times after which the oil was dissolved in distilled water and titrated to neutrality using 0.1 M Ba( OH)? The resulting solution was evaporated to dryness under reduced pressure as described before and the solid residue suspended in 100% ethanol. The suspension was transferred to a tared amber vial and the solvent evaporated under nitrogen at 30°C. The percent recovery of synthesized ketohexose phosphate was 86.9% with a yield of 73 pmoles of the barium salt. The an- hydrous salt was then stored desiccated at 0°C. Figure (20) reveals a one to one stoichiometry between the production of ketohexose and the loss of starting compound. It can also be seen that quantitative conversion of the re- actant to ketohexose was accomplished under the described conditions. After the ketohexose product was isolated it was FIGURE 20. 101 AGl-XZ Cl- Ion-exchange chromatography of the enzymatically synthesized ketohexose phosphate. The pmoles of ketohexose are expressed in terms of the total amount present in the fraction (10 ml) assayed. Other details are given in the text. 102 000 8000. 8. 0... .00.. 00.. 00 20.8.3... 00 0: 00 0 00 r 01.80000030000 ...-“00 ..0.. mo+mz..m.. x 80.0. . F N0......«.. 2 8000.0 _- ...-0mm 2 80.0 _ PK _ ..on x 300.0 4 8:2 .. 800.044 .033. z 8.0— 8...... 2.80.0 8...... ... 30.0 8...... 2 80.0 asoxsnom SH’IOH” 103 characterized with respect to GLC analysis as the TMS derivative before and after dephosphorylation,acid lability, ketohexose/Pi ratio, periodate oxidation, and purity and enantiomeric configuration. GLC of the TMS derivative of the oxidized reaction product before and after dephosphorylation: The unidentified ketohexose phosphate was derivatized to the TMS form and analyzed by GLC utilizing D-tagatose 6-phosphate as a stand- ard. Figure (21) shows that the TMS derivative of chemically synthesized D-tagatose 6-phosphate had the same retention time as the TMS derivative of the unidentified ketohexose phosphate (i.e. 3.13 minutes). The shoulder seen after the peak was presumably the a or B isomer of the derivatized com- pounds. When the sample was dephosphorylated and derivatized to the TMS form it exhibited only a single peak which had the same retention time as authentic D-tagatose (Figure 22). It had been previously established that tagatose is separable from the other ketohexoses under the same GLC conditions as used in this analysis (120). It was therefore concluded that the product of the L-galactitol l-phosphate dehydrogenase- catalyzed oxidation was tagatose phosphate, probably tagatose 6-phosphate. Determination of the acid lability of the oxidation product and chemically synthesized Legalactitol l-phosphate and the determination of the ketohexose/Pi molar ratio of the oxidationgproduct: The acid lability of the unidentified 104- DETECTOR RESPONSE o 2 l- 6 MINUTES n: 2 8 VJ :3 8 83 an F3 :- o 2 .1- 6 IMINUTES Figure 21. GLC of TMS derivatives of the unidentified ketohexose phosphate and authentic D-tagatose 6-phosphate. The analysis was conducted as described in.Materials and Methods. The GLC temperature was set to isothermal at 200°C and the attenuation at 320. A. Chemically sythesized D-tagatose 6-phosphate. B. Enzymatically synthesized ketohexose phosphate. 105 DETECTOR RESPONSE 0 h 6 {0 MINUTES 51.] CD 2 ° N a- ‘CD 2 a: c> b‘ E- lid :- J o l- 6 10 MENUTES Figure 22. GLC of TMS derivatives of authentic D-tagatose and the unidentified ketohexose phosphate after dephosphoryla- tion. The dephosphorylation and analysis was accomp- lished as described in Materials and Methods. The GLC was set to isothermal at 155°C and the attenuation at 80. A. Standard D-tagatose TMS derivative. 3. TMS derivative of the unidentified ketohexose phosphate after dephosphorylation. 106 ketohexose phosphate was differentiated from that of fructose l-phosphate, fructose 6-phosphate, and chemically synthesized L-galactitol l-phosphate according to the procedure in Materials and Methods. Figure (23) delineates the hydrolysis rates of each compound. The rate of liberation of Pi from the unidentified ketohexose phosphate closely paralleled the rate of fructose 6-phosphate, which was comparable to that established previously (24), while the hydrolysis rates of fructose 1-phosphate and chemically synthesized L-galactitol l-phosphate were the most labile and stabile respectively. The acid stability of the chemically synthesized L-galactitol l-phosphate was as predicted for a hexitol phosphorylated at a primary alcohol position. This was consistent with the other characterization data of the chemically synthesized L-galactitol l-phosphate (shown previously). The unidentified ketohexose phosphate was determined to be acid-stabile and the number one position was eliminated as a possible location of the phosphate group since the hydrolysis rate was signifi- cantly less than that of fructose l-phoSphate. However, the position of the phosphate function cannot be conclusively assigned to the C-6 position but can be bonded only at an acid-stabile location. The position was definitively estab- lished through periodate oxidation analysis of the compound (see below). The ketohexose/Pi molar ratio was determined by use of the Roe assay and total Pi analysis as described in Materials and Methods. Table (12) shows the results for the unidentified 107 ketohexose phosphate as well as the values for fructose 6- phosphate and fructose 1-phosphate. The values were essentially those expected, 1:1. Therefore the unidentified ketohexose phosphate was monophosphorylated. Periodate oxidation of the unidentified ketohexose phosphate: Periodate oxidation of the unidentified ketohexose phosphate was conducted in the classical manner as described in Materials and Methods. Both fructose 6-phosphate and fructose l-phosphate were also used in the same analysis as standards. The expected reaction scheme is delineated in Figure (24). The ketohexose phosphates are shown to react in their ring forms since this has been established previously (121). The rate-limiting step in the degradation of these sugar phosphates is the hydrolysis of the ester bond generated from the consumption of the first two molar equivalents of periodate. The hydrolysis rate is slow due to the mild acid pH of the reaction and production of additional acid equiv- alents (122). From Figure (24) it is evident that a keto- hexose l-phosphate would yield two different products and different molar ratios of products than would a ketohexose 6-phosphate. This difference in periodate oxidation charac- teristics was therefore used as a criterion for identifying the ketohexose phosphate. Table (13) reveals the total amount of periodate consumed by each compound as determined by titration with sodium thiosulfate. The empirical values agree very closely FIGURE 23. 108 Acid Lability of the unidentified keto- hexose phosphate. Acid hydrolysis of 0 samples was conducted in l N HCl at 95 C and the Pi liberated was determined as described in Materials and Methods. H D-Fructose l-phosphate. O—l D-Fructose 6-phoSphate . o---0Unidentified ketohexose phosphate. 0-13Chemically synthesized L-galactitol l-phosphate. 109 100'” 1’ 60“ S I SXIOHCILH INHDHHJ so“ 100 MINUTES 110 TABLE 12. Pi/ketohexose molar ratio of the unidentified ketohexose phOSphate. Details of the analysis are given in Materials and Methods. Ketohexose Pi Ketohexose P./Ketohexose Theoretical phosphate (pmoles) 1 . Pi/Ketohexose (pmoles) (molar ratio) (molar ratio) Fructose-6-P 0.098 0.100 0.98 1.00 Fructose-l—P 0.098 0.098 1.00 1.00 Unidentified ketohexose 0.102 0.098 1.04 1.00 phosphate 111 with the theoretical values as determined from.the reaction scheme in Figure (24). For each mole of ketohexose 6- phosphate three moles of periodate were reduced to 103 and every mole of ketohexose l-phosphate consumed four moles of periodate. The unidentified ketohexose 6-phosphate con- sumed three moles of periodate for each mole of the keto- hexose phosphate suggesting that it was very similar in structure to D-fructose 6-phosphate. The quantity of formaldehyde produced by each sample was quantitatively analyzed using chromotropic acid (Table 14). As expected, D-fructose l-phosphate yielded a stoichio- metric amount of formaldehyde, one mole of formaldehyde for every mole of sample, while D-fructose 6-phosphate and the unidentified ketohexose phosphate produced no detectable amounts of formaldehyde. The results were in concert with the phosphorylation of the unidentified ketohexose phosphate at position number six because only D-fructose l-phosphate can furnish an oxidizable primary alcohol function during the course of periodate oxidation. D-Fructose 6-phosphate as the dialdehyde intermediate hydrolyzes yielding a two-carbon moiety (glycolate) containing a primary alcohol group which cannot be oxidized to formaldehyde due to the adjacent carboxyl function (123). Glycolate was colorimetrically determined utilizing 2,7-dihydroxynaphthalene. Table (15) shows that D-fructose 6-phosphate and the unidentified ketohexose phosphate pro- duced the expected amount of glycolate while D-fructose 112 .AHNHV moumnmmosa omoxmsoumx mo 080500 Gowumpwxo mumpOHumm Hmofiuouomna .qm mmanm 00000 . ma meow w-m ouwsdmoud-H uncuosum 0am-0-m00 0-0.0 N 00000 0m 00 0m +_ .+ r-I0 --0-0 N . 0 00 0w .u0 0-0-0 N I .- IN A . + Ale-II- 00000 Alwlll- 000-.w AI-mlllo-mo 00 AIHIII 00-00 -00 . -0H 00-0-0 0 0 0... + -0H m _ _ . 00000 0u0-0 . 0c -00 I . - 0-0-0 00000 0n0-0-m00 .... ouosmmoud-w omouosum 00 00 00 mo . N . N . N . 0H0- - 00 um- - 0 um-o- 00 0am-0-m00 . 0 . 0 . 0 w 0 . . - D ' U I - 00 0u0-0 00 00 m 0 00 . 0 00 . 0 ... _ ouw-0 008 4 00000 _ 00-...70 W00 0N0 0 N00 N 0-0 -00 - moomo mooo _IIIOuo —|Iliu-Qm - . . 00000 00m00 00.000 113 TABLE 13. Consumption of periodate by ketohexose phosphates. The reaction conditions and titrimetric determina- tion of 10% consumption are described in Materials and Method Ketohexose Total Total I0; Theoretical IQ; phosphate (saggiz) consumption consumption Pm (pmoles) (pmoleS) Fructose-6-P 65.2 195 195 Fructose-l-P 65.2 256 260 Unidentified ketohexose 45.5 139 136 phosphate 114 l-phosphate did not produce this end product. Examining the reaction scheme outlined in Figure (24), fructose-l-P cannot produce glycolate since carbon atoms one and two are con- stituents of phosphoglycoaldehyde while the remaining carbons are further oxidized to one carbon units. The glycolate produced from ketohexose 6-phosphates is not further oxidized due to the poor complexing behavior of the carboxyl group with the periodate ion (93). The amount of formic acid produced was determined by titration with standardized sodium hydroxide. In this case it was expected that all of the compounds would produce formic acid. However, fructose l-phosphate would produce three moles of formic acid per mole of compound and the ketohexose 6- phosphates would produce two moles of formic acid per mole of compound. Table (16) shows that the observed values agreed very closely with the theoretical values (see Figure 24). The three-carbon moiety of fructose l-phosphate is degraded completely into three one-carbon units whereas the analogous moiety from.fructose 6-phosphate can only be broken down to formic acid and phosphoglycoaldehyde due to the pre- sence of the phosphate group at carbon number six. The production of carbon dioxide was determined since a possibility existed that the number one position of the ketohexose 6-phosphates could be oxidized first leaving carbon number two available for oxidation to C02. This poss- ibility was not likely since the data accumulated indicated 115 TABLE 14. Production of formaldehyde after periodate oxida- tion of the unknown ketohexose phosphate. Details of the determination are given in Materials and Methods. Ketohexose Total Total Theoretical phosphate sample formaldehyde formaldehyde (pmoles) produced produced (pmoles) (pmoles) Fructose-6-P 65.2 0.0 0.0 Fructose-l-P 65.2 65.7 65.2 Unidentified ketohexose 45.5 0.0 0.0 phosphate 116 TABLE 15. Formation of glycolate after periodate oxidation of the unidentified ketohexose phosphate. Details of the determination are described in Materials and Methods. Ketohexose Total Total Theoretical phosphate sample glycolate glycolate (pmoles) produced produced (pmoleS) (pmoles) Fructose-6-P 65.2 60.0 65.2 Fructose-l-P 65.2 0.0 0.0 Unidentified ketohexose 45.5 45.6 45.5 phosphate 117 TABLE 16. Formic acid production from the unidentified keto- hexose phosphate after periodate oxidation. Details of the determination are described in Materials and Methods. Ketohexose Total Total Theoretical phosphate sample formic acid formic acid (pmoles) produced produced (pmoles) (pmoles) Fructose-6-P 65.2 135 130 Fructose—l-P 65.2 195 195 Unidentified ketohexose 45.5 91.4 91.0 phosphate 118 initial cleavage between carbons two and three or carbons three and four leaving no intermediate which would be oxidiz- able to C02. The formation of CO2 was determined using D- fructose as a standard. Table (17) shows that fructose pro- duced carbon dioxide in the expected quantity while the phosphorylated sugars produced essentially no C02. The small amount produced by fructose 1-phosphate and the unidentified ketohexose phosphate represented only 5% and 3% of the total pmoles of the compounds respectively. These values most likely rose from errors in titration and/or absorption of C02 from the air. The results were those expected since the phosphorylated sugars react as though they are in the ring form and the free ketohexoses react as though they are in the open chain form and are able to yield carbon dioxide (123). The absence of CO2 production, furthermore, was not due to absorption of CO2 into the reaction mixture as carbonic acid since the solution was acidic and consequently caused the carbonic acid to be broken down to CO2 and H20. Phosphosphoglycoaldehyde was determined quantitativ- ely as acid-labile phosphate since no convenient and direct assay of this compound was available. The acid-labile keto- hexose l-phosphate, which is expected to yield phosphoglyco- late after periodate oxidation, would produce an acid-stabile phosphate compound while the acidvstable fructose 6-phosphate and the unidentified ketohexose phosphate would yield the acid-labile phosphoglycoaldehyde. Thus, the analysis would 119 TABLE 17. Analysis for carbon dioxide production after periodate oxidation of the unidentified keto- hexose phosphate. The method of collecting C0 and the titrimetric determinations are des- cribed in Materials and Methods. Total Total Theoretical Sample sample carbon dioxide carbon dioxide (pmoles) produced produced (pmoles) (pmoles) Fructose-6-P 65.2 0.0 0.0 Fructose-l-P 65.2 3.5 0.0 D-Fructose 65.0 61.5 65.0 Unidentified ketohexose 45.5 1.5 0.0 phosphate 120 be diagnostic for phosphoglycoaldehyde under the conditions described in Materials and Methods. Figure (25) shows that the acid lability of the phosphate compounds from fructose 6-phosphate and the unidentified ketohexose phosphate closely paralleled the hydrolysis rate of authentic phosphoglyco— aldehyde. The acid stability of the phosphoglycolate from fructose 1-phosphate was nearly identical with that of com- mercial phosphoglycolate. Therefore, the qualitative analy- sis was consistent with the evidence that the unidentified ketohexose phosphate must be phosphorylated at position number six. Table (18) represents the quantitative data obtained from the analysis. The total amount of acid-stabile phosphate, as determined by complete acid hydrolysis, for phosphoglycolate from fructose l-phosphate also agreed very well with the theoretical value. Phosphoglycolate was quantitatively analyzed utiliz- ing highly purified phosphoglycolate phosphatase. Table (19) shows that only fructose l-phosphate produced phosphoglyco- late as predicted while the other two compounds yielded no such product. The quantity of phosphoglycolate detected agreed very closely with the theoretical value. The reaction scheme in Figure (24) predicts that ketohexose 6-phosphates cannot form phosphoglycolate but rather phosphoglycoaldehyde and therefore the unidentified ketohexose phosphate was deter- 'mined to be phosphorylated at carbon atom number six. The periodate oxidation data are summarized in Table (20). It can be seen that the experimentally obtained values lOO - 90" 60* PERCENT HYDROLYSIS Figure 25. ‘d—b L t 25 50 75 ZMINUTES~ Acid lability of the organic phosphate produced after periodate oxidation of the unidentified ketohexose phosphate. The acid lability was determined as was done before periodate oxidation. o—o Organic phosphate from the unidentified ketohexose phosphate. H Organic phosphate from the chemically syn- thesized L-galactitol l-phosphate. 0——oOrganic phosphate from fructose 6-phos- phate. D-—n Phosphoglycoaldehyde. H Phosphoglycolate. HOrganic phosphate from fructose l-phos- phate. 122 TABLE 18. Production of phosphoglycoaldehyde after period- ate oxidation of ketohexose phosphates. The quantity of phosphoglycoaldehyde was determined as the amount of acid-labile phosphate formed as calculated from Figure (25). P. determinations are described in Materials and Methods. Total Total Theoretical Total Theoretical Ketohexose sample acid- acid-labile acid- acid-stabile phosphate (pmoles) labile P i produced stabile Pi produced Pi Pi (pmdhafl {momxxfl. gnnbafi pnthed (pmoles?) (pmoles) Fructose-6-P 65.2 66.0 65.2 0.0 0.0 Fructose-l-P 65.2 0.0 0.0 64.7 65.2 Unidentified ketohexose 45.5 45.0 45.5 0.0 0.0 phosphate 123 are nearly identical with the theoretical figures. Since the data for the unidentified ketohexose phosphate yielded essentially the same results as the D-fructose 6-phosphate standard, it was concluded that the phosphate function was located in the number six position of the molecule and that there was no doubt of its identity as a ketohexose 6-phosphate. With the identification of the sugar moiety as tagatose by previous GLC analysis (Figure 22), it was now concluded that enzymatic oxidation of the galactitol phosphate from the PTS- catalyzed synthesis must have yielded tagatose 6-phosphate. The above data afford no information as to the enantiomeric form of the enzymatically synthesized tagatose 6-phosphate. It was therefore necessary to determine the enantiomeric configuration in order to establish the position of the phosphate function on the galactitol phosphate molecule. The appropriate experiment was then conducted as described in the following section. Purity_and enantiomeric determination of the enzymatic- ally synthesized tagatose 6-phosphate: Both the purity and the enantiomer configuration were established simultaneously by use of a coupled enzyme assay. The reaction mixture contained the following components in a micro-quartz cuvette (l-cm light path) in a final volume of 0.15 ml: 25 pmoles of glycylglycine buffer (pH 7.5), 4.0 pmoles of MgClz, 1.5 pmoles of ATP, 0.06 pmole of NADH, 13 units of rabbit muscle a-glycerophosphate dehydrogenase containing triose phosphate isomerase, 0.075 unit 124 TABLE 19. Formation of phosphoglycolate after periodate oxi- dation of the unidentified ketohexose phosphate. Assay of phosphoglycolate using phosphoglycolate phosphatase is described in Materials and Methods. Total Total Theoretical Ketohexose sample phosphoglycolate phosphoglycolate phosphate (pmoles) produced produced (pmoleS) (pmoles) Fructose 6-P 65.2 0.0 0.0 Fructose l-P 65.2 63.3 0.0 Unidentified ketohexose 45.5 0.0 0.0 phosphate 125 TABLE 20. Summary of the periodate oxidation data. All emperical values were calculated from the preced- ing tables and the theoretical values determined from the reaction scheme of Figure (24). pmoles Oxidation product/pmoles Sample Oxidation Fructose Fructose Unidentified product 6-phosphate l-phosphate ketohexose phosphate 10; Consumed 2.97 3.93 3.04 TheoreticalRatk> 3.00 4.00 3.00 Formaldehyde 0.00 1.01 0.00 Theoretical Ratu>0.00 1.00 0.00 Glycolate 0.92 0.00 1.02 Theoretical Ratio 1. 00 0.00 1.00 Phosphoglycolate 0.00 0.97 0.00 Theoretical Ratio 0.00 1.00 0.00 Acid Lability Labile Stable Labile Theoretical Lability Labile Stable Labile Formate 1.99 2.99 2.00 Theoretical Ratk>2.00 3.00 2.00 Carbon dioxide 0.00 0.04 0.03 Theoretical Ratio 0.00 0.00 0.00 126 of tagatose 1,6-diphosphate aldolase from Staphylococcus aureus, 10 units of rabbit muscle fructose 6-phosphate kinase, and 0.010 or 0.025 pmole of the tagatose 6- phosphate. The reaction was initiated by the addition of the substrate and the decrease in absorbance at 340 nm was monitored at 30°C with a Gilford 2400 recording spectrophoto- meter. The total absorbance change was then determined after correcting for that change due to dilution from addition of the substrate. The pmoles of sample or NADH oxidized was then calculated using the conversion factor given in the enzyme assay section (see Materials and Methods). If the product was D-tagatose 6-phosphate then it was expected that the coupled enzyme assay would demonstrate a loss of two moles of NADH for every mole of D-tagatose 6- phosphate present. The L-isomer would yield a one to one stoichiometry between the ketohexose phosphate and loss of NADH due to the D-isomer specificity of triose phosphate isomerase (124). Table (21) shows that a one to two stoichio- metry, moles ketohexose to moles NADH oxidized, exists rather than a one to one mole ratio. The analysis was per- formed at two different concentrations of the tagatose 6- phosphate and also in the absence of either fructose 6- phosphate kinase, tagatose 1,6-phosphate aldolase, or the tagatose 6-phosphate in order to insure the validity of the determination. From the results it was concluded that only the D-isomer was formed from the oxidation of the PTS- catalyzed reaction product. It was also concluded, then, that 127 TABLE 21. Purity and enantiomeric configuration of the enzymatically synthesized tagatose 6-phosphate. The details of the analysis are described in the text. NADH Expected Assay Description oxi- NADH dized oxidized Tagatose-6-P Total (pmoles) (pmoles) (pmoles) absorbance change at 340 nm Complete 0.025 2.050 0.050 0.050 (see text) Complete 0.010 0.800 0.019 0.020 (see text) Complete minus 0.025 0.000 0.000 0.000 fructose-6-P kinase Complete minus 0.025 0.000 0.000 0.000 tagatose-1,6- diphosphate aldolase Complete minus 0.000 0.000 0.000 0.000 tagatose-6-P 128 L-galactitol l-phosphate must have been formed from the PTS- catalyzed reaction. Calculation of the percent purity of the isolated oxidation product from Table (21) yielded a value of 100%. DISCUSSION The pathway by which certain hexitols such as mannitol and sorbitol are metabolized have been established previously (see Literature Review) but the pathway of galactitol cata- bolism has not been elucidated for any organism except for a Pseudomonas species (see Introduction). The work reported in this thesis conclusively demonstrated that an initiation of galactitol metabolism in Aerobacter aerogenes was dependent upon the phosphoenolpyruvate-dependent phosphotransferase sys- tem. Furthermore, the product of the PTS-catalyzed reaction was established as L-galactitol l-phosphate by chemical, physical, and enzymatic methods. Figure (26) delineates the reaction scheme by which galactitol is presumably phosphorylated by the PTS as based upon PTS-catalyzed phosphorylation of other sugars (35). Each of the participating catalysts involved was required for the biosynthesis of the phosphorylated product. In addition, the rate of synthesis of product was dependent upon the con— centrations of the enzymatic constituents. The figure also reveals that PEP is the phosphoryl donor which is contrary, assuming that the same system is functional in both E, 2213 and A, aerogenes, to the implications of WOlff and Kaplan (25) who suggested that a kinase, and therefore ATP, was com- mitted to the phosphorylation of galactitol. Their experiments 129 130 'with crude extracts of 2, 2222 grown on galactitol demon— strated a dependence upon ATP for the formation of a sub- strate utilized by partially purified L-galactitol l-phosphate dehydrogenase. However, the experiments did not preclude the possibility that the crude extract catalyzed formation of PEP from ATP and endogenes levels of pyruvate present and thereby allowing the PEP to serve as the true phosphoryl donor. It is not surprising that Wolff and Kaplan implicated the role of a kinase for the formation of galactitol phosphate since they were not cognizant of the PTS which was discovered 8 years later in 1964 by Roseman and coworkers (27). The position of the phosphoryl function on the reaction product was also of importance in this study since the PTS is known to phosphorylate certain sugars in the number one or number six positions (31, 34, 40). If the possibilities are circumscribed to the terminal positions of the hexitol then only two contingencies prevail for the identity of the PTS- catalyzed reaction product as shown in Figure (27). The figure also presents four possible products that can arise after subsequent oxidation of galactitol phosphate. By using product characterization methods, the identities of the pro- ducts from the PTS-catalyzed reaction and the L-galactitol l-phosphate dehydrogenase-catalyzed reaction.were unequivac- ably established. Chemically synthesized L-galactitol l-phosphate from sodium borohydride reduction of commercially prepared D- galactose 6-phosphate exhibited the same physical, chemical, 131 Pyruvate ‘ HPnNP galactitol 2 ltl 42 RI, Mg+ Efl, Mg PEP HPr L-galactitol l-phosphate Figure 26. Initiation of galactitol metabolism by the PTS. 132 and enzymatic properties as did the enzymatically synthesized galactitol phosphate. Since both components yielded the same mass spectrum, had identical retention times in the GLC analysis, had the same retention time by GLC as did authentic galactitol after their dephosphorylation with alkaline phosphatase, and served as substrates for partially purified L-galactitol l-phosphate dehydrogenase, it was almost con- clusive that the product of the PTS-catalyzed reaction was L- galactitol l-phosphate. When the compound was enzymatically oxidized utilizing L-galactitol l-phosphate dehydrogenase, a ketohexose phosphate was formed which had the D configura- tion. When the ketohexose phosphate was analyzed by GLC it had the same retention time as did authentic D—tagatose 6- phosphate or D-tagatose upon dephosphorylation. This elim- inated the possibilities of L-tagatose l-phosphate and L- ‘ tagatose 6-phosphate as oxidation products. Upon periodate oxidation of the acid—stabile D-ketohexose phosphate, quantitative amounts of products were formed which were con- sistent with the periodate oxidation of a ketohexose 6- phosphate. The periodate oxidation also caused the formation of an acid-labile phosphate and therefore supported the con- clusion that the phosphate function was located at C-6. The identity of the oxidized compound must, therefore, be D- tagatose 6-phosphate, which in turn proves that the PTS catalyzed phosphorylation of galactitol to produce L-galactitol l-phosphate and not L-galactitol 6-phosphate. FIGURE 27. 133 Possible products of the PTS-catalyzed phosphorylation of galactitol. The res- pective products of the oxidation of each possible galactitol phosphate are also shown. 134 L-Tagatose l-Phosphate @o .- NAD+ L-Galactitol l-Phosphate ‘ -+ NAD PTS D-Tagatose 6-Phosphate p Galactitol b PTS D-Tagatose l-Phosphate @0 «L / NAD D-Galactitol l-Phosphate - d L- } NAD... L-Tagatose 6-Phosphate H—r'r-r-T'J—L'In r-rJ-J-r—L'I—u-r- 135 The data have implied that the second step in the bio- degradation of galactitol is the oxidation of_the phosphory- lated derivative to D-tagatose 6-phosphate. This reaction would be catalyzed by L-galactitol l-phosphate dehydrogenase as was shown to occur in E, 2221 (25). This hypothesis can be substantiated by examining the phenotypic characteristics of mutants in this enzyme and then determine whether the appro- priate metabolite accumulates. Such genetic investigations were out of the scope of this study. But the implications stated are criteria for asking how galactitol is further metabolized in Aerobacter aerogenes. 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES Carr, C.J., and Krantz, Jr., J.C., Adv. in Carb. Chem., 2, 175 (1945). Lohmar, R., Adv. in Carb. Chem., 2, 211 (1949). Fischer, H.O.L., and MacDonald, D.L., Ann. Rev. Biochem., 22, 43 (1951). Horecker, B.L., and Mahler, A.H., Ann. Rev. Biochem., 22, 207 (1955). Korkes, S., Ann. Rev. Biochem., 22, 685 (1956). DeDuve, C., and Hers, H.G., Ann. Rev. Biochem., 22, 149 (1957). Massey, V., and Veeger, C., Ann. Rev. Biochem., 22, 579 (1963). Eitel, E.H., Ind. Eng. Chem., 22, 1202 (1920). Shaw, D.R.D., Biochem. J., 22, 394 (1956). Dozois, K.P., Hachtel, F., Carr, C.J., and Kranz, Jr., J.C., J. Bacteriol., 22, 189 (1935). Hajna, A.A., J. Bacteriol., 22, 339 (1939). Bradley, B., Proc. Roy. Soc. Med., 2, 43 (1907). Teague, 0., and Morishima, K., J. Infect. Dis., 22, 52 (1920). Neal, O.R., and Walker, R.H., J. Bacteriol., 22, 173 (1935). King, T.E., and Cheldelin, V.H., J. Biol. Chem., 198, 135 (1952). King, T.E., and Cheldelin, V.H., J. Bacteriol., 22, 581 (1953). Sebek, O.R., and Randles, C.I., Bacteriol. Proc. 137 (1951). 136 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 137 Bertrand, C., Compt. rend. Acad. Sci., Paris, 126, 762 (1898). Sebek, O.K., and Randles, 0.1., J. Bacteriol., _2, 693 (1952). Hermann, S., and Neuschul, P., Biochem. 2., 233, 129 (1931). Blakely, R.L. Biochem. J., 33, 257 (1951). McCorkindale, J., and Edson, N.L., Biochem. J., _2, 518 (1954). Marmur, J., and Hotchkiss, R.D., J. Biol. Chem., 222, 383 (1955). Wolff, J.B., and Kaplan, N.O., J. Biol. Chem., 222, 849 (1956). Wolff, J.B., and Kaplan, N.O., J. Bacteriol., 22, 557 (1956). Liss, M}, Horwitz, S.B., and Kaplan, N.O., J. Biol. Chem., 221, 1342 (1962). Kundig, W., Gosh, S., and Roseman, S., Proc. Natl. Acad. Sci. U.S.A., 22, 1067 (1964). Kundig, W., and Roseman, S., J. Biol. Chem., 222, 1393 (1971). Anderson, B., Weigel, N., Kundig, W., and Roseman, S., J. Biol. Chem., 222, 7023 (1971). ' Kundig, W., and Roseman, S., J. Biol. Chem., 222, 1407 (1971). Hanson, T.E. and Anderson, R.L., Proc. Natl. Acad. Sci. U.S.A., 22, 269 (1968). Kelker, N.E., Hanson, T.E., and Anderson, R.L., J. Biol. Chem. 245, 2060 (1970). Kelker, N.E., and Anderson, R.L., J. Bacteriol., 105, 160 (1970). Kelker, N.E., Simkins, R.A., and Anderson, R.L., J. Biol. Chem., 222, 1479 (1972). Kaback, H.R., J. Biol. Chem., 222, 3711 (1968). 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 138 Kaback, H.R., Proc. Natl. Acad. Sci. U.S.A., 22, 724 (1969). Kaback, H.R., Biochim. Biophys. Actap_265, 367 (1972). Kornberg, H.L., and Reeves, R.E., Biochem. J., 222, 1339 (1972). Kornberg, H.L., in Woessner, J.K., and Huijing, F., Eds., The Molecular Basis of Biological TransportL Miami Winter Symposium, 2, Academic Press, New York (1972). Ferenci, T., and Kornberg, H.L., Fed. Eur. Biochem. Soc. Lett., 22, 360 (1971). Kornberg, H.L., and Reeves, R.E., Biochem. J., 126, 1241 (1972). Kornberg, H.L., and Smith, J., Fed. Eur. Biochem. Soc. Lett., 22, 270 (1972). Korte, T., and Hengstenberg, W., Eur. J. Biochem., 231, 295 (1971). Schrecker, 0., and Hengstenberg, W., Fed. Eur. Biochem. Soc. Lett., 22, 209 (1971). Hengstenberg, W., Fed. Eur. Biochem. Soc. Lett., 2, 277 (1970). Hengstenberg, W., Pernberthy, W.K., Hill, K.L., and Morse, M.L., J. Bacteriol., 22, 383 (1969). Anderson, R.L., and Wood, W.A., Annu. Rev. Microbiol., 22, 539 (1969). Kaback, H.R., Ann. Rev. Biochem., 2_, 561 (1970). Kaback, H.R., in Kleinzeller, A., and Bonner, F., Eds., Current Tppics in Membranes and Transport, Academic Press,7NeW'Yorki(1970). Kaback, H.R., in Tosteeson, 0., Ed., The Molecular Basis of Membrane Transport, Prentice-Hall, Inc., Engléwood Cliffs, New Jérsey, p. 421 (1969). Pardee, A.B., Science, 112, 632 (1968). Roseman, S., J. Gen. Physiol., 22, 138 (1969). Stein, R., Schrecker, 0., Lauppe, H.F., and Hengstenberg, H., Fed. Eur. Biochem. Soc. Lettpp 22, 98 (1974). 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 139 Kundig, W., Kundig, F.D., Anderson, B., and Roseman, S., J. Biol. Chem., 241, 3243 (1966). Walter, R.W., Purification and Role of the Inducible Cytoplasmic PfoteinICOmponent of the Phosphoenolpyruvate: D-Fructose l-Phosphotransferase System of Aerobacter aerogenes, (Ph.D. thesis, MiEhigan State“Uhiversity, EastLansing, 1972). Walter, R.W., and Anderson, R.L., Biochem. Biophys. Res. Commun., 22, 93 (1973). Kaback, H.R., and Walsh, C.T., J. Biol. Chem., 222, 5456 (1973). Tanaka, S., and Linn, E.C.C., Proc. Natl. Acad. Sci. U.S.A., 57, 913 (1967). Tanaka, S., Fraenkel, D.G., and Linn, E.C.C., Biochem. Biophys. Res. Commun., 22, 63 (1967). Simoni, R.D., Levinthal, M., Kundig, F.D., Kundig, W., Anderson, B., Hartman, P.E., and Roseman, S., Proc. Natl. Acad. Sci. U.S.A., 22, 1963 (1967). Egan, J.B., and Morse, M.L., Biochim. BioPhys. Acta., 109, 172 (1965). Berman, M., and Linn, E.C.C., J. Bacteriol., 105, 113 (1971). deCrombrugghe, B., Perlman, R.L., Varmus, R.E., and Pastan, I., J. Biol. Chem., 222, 5828 (1969). Pastan, I., and Perlman, R.L., J. Biol. Chem., 222, 5836 (1969). Tyler, B., and Magasanik, B., J. Bacteriol.,222, 411, (1970). Saier, M.H., and Roseman, S., J. Biol. Chem., 247, 972 (1972). Gershanovich, V.N., Yurovitskaya, N.V., and Burd, C.I., Mel. Biol. 2, 534 (1970). Ghosh, S., and Ghosh, D., Indian J. Biochem., 2, 49 (1968). Groves, D.J., and Gronlund, A.F., J. Bacteriol., 222, 1256 (1969). 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 140 Hengstenberg, W., Egan, J.B., and Morse, M.L., Proc. Natl. Acad. Sci. U.S.A., 22, 274 (1967). Hengstenberg, W., Egan, J.B., and Morse, M.L., J. Biol. Chem., 243, 1881 (1968). Hengstenberg, W., Pernberthy, W.K., Hill, K.L., and Morse, M.L., J. Bacteriol., 22, 2187 (1968). Kennedy, E.P., and Scarborough, G.A., Proc. Natl. Acad. Sci. U.S.A., 22, 225 (1967). Shablenko, V.P., Chem. Abstracts, 22, 110 (1971). Teuber, M., J. Bacteriol., 100, 1417 (1969). Wang, R.J., and Morse, M.L., J. Mol. Biol., 22, 59 (1968). Krulwich, T.A., Sobel, M.E., and Wolfrom, E.B., Biochem. Biophys. Res. Commun., 22, 258 (1973). Romano, A.H., Eberhard, S.J., Dingle, S.L., and McDowell, T.D., J. Bacteriol., 104, 808 (1970). Gay, P., and Rapoport, 0., Chem. Abstracts, 22, 117 (1970). Kundig, W., and Roseman, S., Fed. Proc., 22, 463 (1969). Patni, N.J., and Alexander, J.K., J. Bacteriol., 105, 226 (1971). Patni, N.J., and Alexander, J.K., J. Bacteriol., 105, 220 (1971). McKay, L.L., Walter, L.A., Sandine, W.E., and Elliker, P.K., J. Bacteriol., 22, 603 (1969). Tanaka, S., Lerner, S.A., and Linn, E.C.C., J. Bacteriol., 22, 642 (1967). Shaw, D.R.D., Methods in Enzymology, 2, 323 (1962). Fiske, C., and SubbaRow, Y., J. Biol. Chem., 22, 375 (1925). Jones, T.M., and Albersheim, P., Plant Physiol., 22, 926 (1972). Albersheim, P., Carb. Res., 2, 340 (1967). Musick, W.D., and Wells, W.W., Archives Biochem. Biophys., 165, 217 (1974). 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 141 Holland, J.F., Sweeley, C.C., Thrush, K.E., Teets, R.E., and Bieber, M.A.,‘Ana1.'Chem., 22, 308 (1973). Wolff, J.B., and Kaplan, N.O., J. Biol. Chem., 222, 849 (1956). Guthrie, R.D., Methods in Carb. Chem., 2, 432 (1965). Bobbitt, J.M., Adv. in Carb. Chem., 22, 1 (1956). Hanson, T.E., The Metabolism of D-Fructose in Aerobacter aerogenes, (Ph.DI thesis, Michigan State University, EiétLansing, 1969). Christeller, J.T., Phosphoglycolate Phosphatase: Purifi- cation and Properties, (Ph.D. thesis, Michigan State University, East Lansing, 1974). Green, A.A., and Hughes, W.L., Methods in Enzymology, 2, 76 (1955). Clark, Jr., J.U., Experimental Biochemistr‘, p. 32, W.H. Freeman and Co., SanFrancisco,Ca1:fornia. Clark, Jr., J.U., Experimental Biochemistri p. 37, W.H. Freeman and Co., san Francisco,—Ca11f0rnia. Tombs, M.P., Souter, F., and MacLagan, N.F., Biochem. J., 22, 167 (1959). Lowry, 0.H., Rosebrough, N.J., Parr, A.L., and Randall, R.J., J. Biol. Chem., 22;, 265 (1951). Roe, H.J., J. Biol. Chem., 222, 15 (1934). Ashwell, 0., Methods in Enzymology, 2, 75 (1957). Park, J.T., and Johnson, M.J., J. Biol. Chem., 222, 149 (1949). Frisell, W.R., Meech, L.A., and Mackenzie, C.C., J. Biol. Chem., 207, 709 (1954). Horning, E.C., and Horning, M.G., J. Org, Chem., 22, 95 (1946). Lewis, K.P., and Weinhouse, S., Methods in Enzymology, 2, 272 (1957). Bernstein, I.A., and Wood, H.G., Methods in Enzymology,2, 561 (1957). 142 108. Pierce, C., Haenisch, E.L., and Sawyer, D.T., Q2antita- tive Anal Sis fourth ed., p. 256, John Wiley an ons, Inc., New Yort. 109. Clark, Jr., J.U., Experimental Biochemistry, p. 15, W.H. Freeman and Co., San’Francisco, CaIifErnia. 110. Pierce, C., Haenisch, E.L., and Sawyer, D.T., Qpantita- tive Ana2ysis, fourth ed., p. 294, John Wiley andESons, Inc., Newaork. 111. Keilin, D., and Hartree, E.P., Proc. Roy.'Soc. (London) B 124I 397 (1938). 112. Peterson, E.A., and Sober, H.A., Methods in Enzymology, 2, 3 (1962). 113. Bio-Rad Laboratories, Gel Chromatography,_p. 32 (1971). 114. Torriani, A., Biochim. Biophys. Acta. 22, 460 (1960). 115. Sloneker, J.H., Biomedical Applications of Gas Chroma- tography, 2, 87 (1968). 116. DeJongh, D.C., Radford, T., Hribar, J.D., Hanessian, S., Bieber, M., Dawson, 0., and Sweeley, C.C., J. American Chem. Soc., 22, 1728 (1969). 117. Harvey, D.J., and Horning, M.G., J. of Chromatography, Z2, 51 (1973). 118. Garen, A., and Levinthal, C., Biochim. Biophys. Acta, 22, 470 (1960). 119. Khym, J.X., and Cohn, W.E., J. American Chem. Soc., 22, 1728 (1969). 120. Bissett, D., and Anderson, R.L., Biochem. Biophys. Res. Commun., 22, 641 (1973). 121. Morrison, M., Rouser, C., and Stotz, E., J. American Chem. Soc., 22, 5156 (1955). 122. ‘Marinetti, G.V., and Rouser, G., J. American Chem. Soc., 22, 5345 (1955). 123. Clark, Jr., J.U., Experimental Biochemistry, p. 7, W.H. Freeman and Co., San Francisco, CéIifbrnIa. 124. Meyerhof, 0., and Junowicz-Kocholatz, R., J. Biol. Chem., 222, 71 (1943). G T “WI (@1711, fig Mill! [fllflflfilflflflyfliflfliflflfi