J. Biochem., 81, 851-858 (1977)

Purification and Some Properties of Cyclohexylamine Oxidase

from a Pseudomonas sp.1

Toshie TOKIEDA, Toshio NIIMURA , Fuminori TAKAMURA, and Tsutomu YAMAHA

Department of Medical Chemistry , National Institute of Hygienic Sciences, 1-18-1, Kamiyoga, Setagaya-ku, Tokyo 158

Received for publication, August 25 , 1976

Cyclohexylamine oxidase was purified 90-fold from cell-free extracts of Pseudomonas sp . capa ble of assimilating sodium cyclamate. The purified enzyme was homogeneous in disc elec trophoresis, and the molecular weight was found to be approximately 80 ,000 by gel filtration. The enzyme catalyzed the following reaction:

cyclohexylamine+O2+H2O •¨ eyclohexanone+NH3+H2O2

The enzyme thus can be classified as an amine oxidase; it utilized oxygen as the ultimate electron acceptor. The pH optimum of the reaction was 6.8 and the apparent Km value for cyclohexylamine was 2.5 x 10-4 M. The enzyme was highly specific for the deamination of alicyclic primary amines such as cyclohexylamine, but was found to be inactive toward ordinary amines used as substrates for amine oxidases. The enzyme solution was yellow in color and showed a typical flavoprotein spectrum; the addition of cyclohexylamine under anaerobic conditions caused reduction of the flavin in the native enzyme. The flavin of the prosthetic group was identified as FAD by thin layer chro matography. The participation of sulfhydryl groups in the enzymic action was also suggested by the observations that the enzyme activity was inhibited in the presence of PCMB and could be recovered by the addition of glutathione.

Sodium cyclamate (CHS-Na) was widely used as a Kojima and Ichibagase (2) reported that laboratory sweetening agent for various drugs and foods, but animals and humans receiving CHS-Na excreted an was banned for general use in 1969, due to possible appreciable amount of CHA in urine, many meta carcinogenicity (1) and metabolic conversion to bolic studies of CHS have been undertaken (3-6), cyclohexylamine (CHA), a toxic substance. Since and urinary excretion of other metabolites, includ ing cyclohexanone (CHnone) and cyclohexanol (CHnol), was observed in both amimals and man. 1A part of this work was presented at the Annual Meet- A series of studies in our laboratory (7-11) ing of the Kanto Division of the Agricultural Chemical Society of Japan, Tokyo, November 4, 1975. suggested that intestinal bacteria play an important Abbreviation: CHS-Na, sodium cyclamate; CHA, role in the metabolism of CHS, and this suggestion cyclohexylamine; CHnone, cyclohexanone; PCMB, was supported by results obtained in several other p-chloromercuribenzoate. laboratories (12-14). Moreover, the present

Vol. 81, No. 4, 1977 851 852 T. TOKIEDA, T. NIIMURA, F. TAKAMURA, and T. YAMAHA authors (10) isolated cyclamate-assimilating bacte terms of the turbidity at 650 nm. Protein was ria from the feces of guinea pigs excreting CHA, determined by the method of Lowry et al. (15). and confirmed that a strain of Pseudomonas sp. Ammonia was determined colorimetrically with possessed an enzyme system forming CHnone from Nessler's reagent after microdiffusion in a Conway CHS via CHA. We also reported the partial unit (16). purification and some properties of CHS sulfamat Determination of CHA and CHnone-Two ml ase, which catalyzed the desulfation of CHS to of the reaction mixture was adjusted to pH> 13 CHA (11). with 6 N NaOH and CHA was extracted with 0.5 ml Our results on the enzymic conversion of CHS of dichloromethane containing 0.02% n-tridecane to CHnone by crude bacterial extracts led us to the as an internal standard. conclusion that amine oxidase might catalyze the Two drops of 6 N HCI and 0.5 g of NaCI were deamination of CHA to CHnone because the de added to 2 ml of the reaction mixture, then CHnone amination was dependent on oxygen (10). Al- was extracted with 0.5 ml of dichloromethane though many studies on amine oxidases have been containing 0.01 % n-dodecane as an internal stand reported, an enzyme deaminating alicyclic amines ard. such as CHA has not previously been obtained. CHA and CHnone were analyzed by gas The present paper describes the purification liquid chromatography on a Shimadzu 4APF unit and some properties of CHA oxidase, which is with dual flame ionization detectors using a 2 m x involved in the second step of the assimilation of 0.3 cm glass column packed with 10% Carbowax CHS by a Pseudomonas sp. 20 M plus 2.5 % NaOH on 60 to 80 mesh Chromo sorb G. Operating conditions were as follows;

MATERIALS AND METHODS column temperature 110°C, injector temperature 200°C, detector temperature 230°C, and carrier gas Chemicals-FAD was obtained from Calbio (nitrogen) flow rate 50 ml/min. The retention times chem; FMN was from Tokyo Slats Inc.; crystalline of CHA, n-dodecane, n-tridecane, and CHnone bovine liver catalase [EC 1.11.1.6] was from Sigma were 3.5, 4.2, 7.0, and 7.2 min, respectively. Chemical Co. ; horse heart cytochrome c, bovine Enzyme Assay-Enzyme activity was de albumin, human gamma globulins and horse apo termined by measuring oxygen consumption mano ferritin were from Schwarz-Mann Co. Other metrically with a conventional Warburg apparatus chemicals were of the highest purity commercially at 30°C in air. The incubation mixture usually available. contained an appropriate amount of enzyme, Maintenance and Growth of the Organism- 120 ƒÊmol of Tris-maleate, pH 6.8, and 30 ƒÊmol of Pseudomonas sp. strain K was used in this work. CHA•EHCl in a total volume of 3 ml. The center This strain was isolated from the feces of guinea well contained 0.2 nil of 20 % (w/v) KOH. The pigs which had been given drinking water containing reaction was started by tipping in the substrate CHS-Na ad libitum, and which excreted CHA and from a side arm. One unit of the enzyme was CHnone in the urine (10). Stock cultures were defined as that amount which consumed 1 ƒÊmol of maintained in a CHS-Na-inorganic salts-vita oxygen per min under the standard assay condi mins medium (medium A) as described previously tions. Specific activity was expressed as units per (10). In order to obtain large quantities of cells, mg of protein. heavy inocula of the cells were transferred to 100 ml Preparation of Cellfree Extracts-In a typical of medium A containing 0.5 % CHS-Na in a 500 ml preparation, 5 g of dried cells was suspended in shaking flask. Sixty flasks were cultured for 7 to 0.04 MTris-HCI, pH 7.2, to a level of 20 mg per ml, 9 days at 30°C with shaking. The cells were and disrupted with an ultrasonic disintegrator collected by centrifugation, washed three times with (Ohtake Sonic 5202) for periods of 2 min (total,- 0.02 M phosphate buffer, pH 7.0, and dried over exposure, 10 min). Unbroken cells and debris P2O5 in vacuo. The dried cells were stored at were removed by centrifugation at 20,000 x g for -20°C until use. 20 min, and the supernatant was used as the crude Determination of Bacterial Growth, Protein, and extract. Ammonia-Bacterial growth was measured in Purification Procedure-Step 1. Streptomycin

J. Biochem. PURIFICATION AND PROPERTIES OF CYCLOHEXYLAMINE OXIDASE 853

sulfate treatment: Five percent streptomycin ƒÊ mol of CHA, 0.5 nil of the purified enzyme, 160

sulfate solution in 0.04 M Tris-HCI, pH 7.2, was ƒÊ mol of Tris-maleate, pH 6.8, and various electron

added slowly to the supernatant with stiring to give acceptors in a total volume of 4 ml. The reaction

a final concentration of 1 %. The mixture was was carried out at 30•Ž for 1 h in a Thunberg tube

allowed to stand for 10 min, centrifuged at 20,000 under N2, then the reduction of the acceptor was

x g for 20 min, and the supernatant was collected. determined spectrophotometrically.

The enzyme activity was quite stable during storage Measurement of the Absorption Spectrum of

for at least 3 months at -20•Ž. CHA Oxidase Reduced with CHA-Thunberg

Step 2. (NH4)2SO4 fractionation: The cuvettes (Fujiwara Factory Co.) containing 6.5 mg

supernatant was made up to 0.26 saturation by the of enzyme (specific activity 4,500 milliunits/mg),

slow addition of solid (NH4)2SO4, then allowed to 120 ƒÊmol of KH2PO4-Na2HPO4, pH 6.8, and 30

stand at 5•Ž for 2 h. The precipitate formed was ƒÊ mol of CHA •E HCl in a total volume of 3 ml, were

centrifuged off. The supernatant was then made made anaerobic by degassing the solution with a

up to 0.84 saturation by further addition of solid vacuum pump and flushing with nitrogen. This

(NH4)2SO4. The resulting precipitate, after stand- procedure was repeated 5 times, then the reaction ing as before, was collected by centrifugation and was started by the addition of CHA from a side

redissolved in an appropriate amount of 0.04 M arm. After standing at room temperature for 5

Tris-HCI, pH 7.2. min, the absorption spectrum was measured in the Step 3. Gel filtration: The redissolved range of 340 to 659 nm. (NH4)2SO4 precipitate was applied to a column Isolation of Flavin in CHA Oxidase-Flavin (2.6 x 100 cm) of Sephadex G-200. The column was isolated according to the procedures of Tipton was eluted with 0.04 M Tris-HCl, pH 7.2, and 5-ml (19) and Burton (20). One ml of 50% (w/v) tri fractions were collected. The most active Sephadex chloroacetic acid solution was added to 4.0 ml of fractions (fractions 34-42) were combined. an ice-cold solution of CHA oxidase (1.3 mg Step 4. Ion-exchange chromatography and protein/ml). After standing on ice in the dark for

dialysis: The combined fractions were dialyzed 15 min, 5 ml of saturated (NH4)2SO4solution was

overnight at 5•Ž against 5 mm Tris-HCl, pH 7.4, added and the mixture was extracted twice with 1.2

The dialyzed fractions were applied to a column ml portions of phenol liquefied by adding a small

(2.6 x 13 cm) of DEAE-cellulose equilibrated with amount of water. The combined phenol extracts 5 mm Tris-HCl, pH 7.4. The protein was eluted were shaken with a mixture of 0.6 ml of water and

with a linear gradient of sodium chloride from an 8 ml of chloroform, and centrifuged. The upper

initial concentration of zero to a final concentration (aqueous) layer was neutralized with 1 M K2HPO4.

of 1.5 M. The activity was completely eluted between 0.5 M and 0.65 M NaCl in 5 mm Tris-HCl, RESULTS AND DISCUSSION

pH 7.4. The combined fractions were dialyzed overnight at 5•Ž against 0.04 M Tris-maleate, pH Purification of CHA Oxidase-Summary data

6.8, to remove NaCl. on the enzyme purification are shown in Table I.

Disc Electrophoresis-Disc electrophoresis The enzyme was purified approximately 90-fold

was carried out according to the method of Davis over the crude extracts with an overall yield of 34 %. The results in Table I were obtained using phos (17) using 7 % polyacrylamide gel at pH 8.9. Determination of Molecular Weight-The phate buffer in the purification procedure, but Tris

method of Andrews (18) was employed for the buffer has been used, as described in " MATERIALS AND METHODS," since it was observed that estimation of molecular weight by gel filtration. A the purified enzyme was more stable in Tris buffer column of Sephadex G-200 (2.6 x 90 cm) equi than in phosphate buffer. librated with 0.04 M Tris-HCl buffer, pH 7.4, The purified enzyme was homogeneous when containing 0.1 M KCl was used. Standard pro checked by disc electrophoresis on polyacrylamide teins used were cytochrome c, bovine albumin, ƒÁ-globulins, and apoferritin. gel, pH 8.9, and the molecular weight was estimated to be about 80,000 by the gel filtration method (18). Spectrophotometric Measurements of Electron Properties of CHA Oxidase-The optimum pH Acceptors-The reaction mixture contained 40

Vol. 81, No. 4, 1977 854 T. TOKIEDA, T. NIIMURA, F. TAKAMURA, and T. YAMAHA

TABLE I. Purification of CHA oxidase. Dried cells (0.56 g) of Pseudomonas sp. were suspended in 28 ml of 0.04 M phosphate buffer, and were then disrupted according to the procedure given in "MATERIALS AND METHODS."

a One unit: the amount of enzyme which consumed 1 ƒÊmol of oxygen per min.

and optimum temperature were about 6.8 and 35•Ž, the purified enzyme was examined. The reaction respectively. The initial velocity was measured at was carried out in a Warburg flask until CHA was

various concentrations of CHA under optimal completely consumed because the presence of CHA

conditions by determining CHnone formed, and disturbed the measurement of ammonia with Nes

the apparent Km for CHA was calculated to be sler's reagent. An aliquot of the reaction mixture

2.5 x 10-4 M by means of Lineweaver-Burk plots. was removed, and both ammonia and CHnone In the previous paper (10), the authors sug produced were determined. gested that CHS sulfamatase was a key enzyme in Table ‡U shows that the deamination of each the assimilation of CHS-Na because the conversion ƒÊ mol of CHA consumed 0.90 ƒÊmol of oxygen and rate of CHA to CHnone by crude extracts was produced 0.84 ƒÊmol of ammonia and 1.06 ƒÊmol of about 15 times greater than that of CHS-Na to CHnone. Therefore, it appeared that deamination CHA. This suggestion was further supported by of CHA in the absence of catalase proceeded in the difference of Km values between CHS sulfa accordance with the following equation: matase (5 x 10-3 M) (11) and CHA oxidase (2.5 x CHA+02+H2O •¨ CHnone +NH3+H2O2 10-4 M). Stoichiometry of the Reaction of CHA-The The addition of catalase reduced the oxygen stoichiometry of oxygen uptake, substrate con consumption by approximately half, as shown in sumption, and production of both ammonia and Fig. 1, and was found to have no effect on the con CHnone in the deamination of CHA catalyzed by sumption of CHA when the remaining CHA was

TABLE ‡U. Stoichiometry of deamination of CHA by CHA oxidase. The reaction mixture contained 4.04, 7.16 or 10.4 ƒÊmol of CHA, purified enzyme and 120 ƒÊmol of Tris-maleate, pH 6.8, in a total volume of 3 ml. The reaction was carried out at 30•Ž in a Warburg flask until CHA was completely consumed, because the

presence of CHA disturbed the measurement of ammonia with Nessler's reagent. CHA, CHnone and ammonia were determined by the methods given in " MATERIALS AND METHODS."

J. Biochem. PURIFICATION AND PROPERTIES OF CYCLOHEXYLAMINE OXIDASE 855

determined at the end of the reaction. This result as follows; 2.0 x 10-5 M sodium 2, 6-dichloroindo-

indicated the formation of H202 during the de phenol (DCPIP), 3.3 •~ 10-1 M phenazine metho amination of CHA, supporting the above equation. sulfate (+DCPIP), 1.0 •~ 10-3 M ferricyanide, 1.6 •~

Requirements for Electron Acceptors-The 10-7 M cytochrome c, 3.8 •~ 10-4 M 2, 3, 5-triphenyl- requirements of the enzyme for primary electron tetrazolium chloride, 7.5 •~ 10-4 M neotetrazolium acceptors were investigated spectrophotometrically. chloride, 2.0 •~ 10-6 M p-benzoquinone, 2.5 •~ 10-5 M The final concentrations of electron acceptors were methylene blue, 8.0 •~ 10-5 M NAD+, and 8.0 •~ 10-5

M NADP+. None of the artificial electron accep

tor tested was active as a primary electron accep

tors for the enzyme, and the reaction proceeded

only when molecular oxygen was present. Conse

quently, the enzyme appeared to be a typical amine oxidase. Substrate Specificity-The relative activities toward various amines were determined at a con centration of 10 mm, and the results are summarized in Table 111. Only a few alicyclic amines among these com- pounds were significantly attacked by the enzyme. CHA was deaminated at the highest rate, and the relative activities of the enzyme toward cyclo Fig. 1. Formation of H2O2 during deamination of heptylamine, 1, 2, 3, 4-tetrahydro-2-naphthylamine, CHA. Enzyme activity was measured under the assay and 1, 2-cyclohexanediamine were 42, 23, and 18 conditions given in "MATERIALS AND METHODS" except that the reaction was carried out in the presence of that of CHA, respectively. The enzyme was

of 60 ƒÊg of catalase. 0, no catalase; •~, catalase. found to be specific for the deamination of alicyclic

TABLE ‡V. Substrate specificity of CHA oxidase. Enzyme activity was determined by measuring oxygen consumption under the assay conditions given in " MATERIALS AND METHODS." All the substrates were added at a final concentration of 10 mm.

Vol. 81, No. 4, 1977 856 T. TOKIEDA, T. NIIMURA, F. TAKAMURA, and T. YAMAHA

primary amines such as CHA, since N-methyl- TABLE ‡W. Effect of glutathione on the inhibition cyclohexylamine was not deaminated at all. In of CHA oxidase activity by PCMB. Enzyme activity addition, the hydrogen atom at C-1, together with was measured under the assay conditions given in "MATERIALS AND METHODS the amino group, was essential for activity, since ." The purified 1-aminocyclohexane-l-carboxylic acid was not enzyme was incubated in 0.04 M Tris-maleate, pH 6.8, deaminated. containing 2•~10-5 at PCMB (final concentration) at Effects of Metal Ions-The effects of various 30•Ž for 10 min, and then divided into two portions. metal ions (10-2 M) on the enzyme activity were Each of them was incubated at 30•Ž for 10 min with or without 2 x 10-2 M glutathione (final concentration), tested by manometric assay at pH 6 .2. The extents dialyzed against 0.04 M Tris-maleate, pH 6.8, at 5•Ž of inhibition of CuSO4 , HgC12, ZnSO4, CdC12, for 48 h, and the enzyme activities of the dialyzed

CaC12,MnSO4, and MgSO4 for the enzyme activity solutions determined. The conditions of incubation were 100, 99, 48, 38, 21, 15, and 7%, respectively. of non-treated enzyme were as described above, except As indicated above, the enzyme activity was that neither PCMB nor glutathione was added. strongly inhibited by Cu2+ and Hg2+. It is well known that these metal ions react with SH groups and that Cult combines with flavins, so the partici pation of such groups in the reaction seems likely. Inhibitory Effects of Various Reagents-The enzyme was inhibited by some copper chelating reagents such as KCN, sodium diethyldithiocarba mate, cuprizone, potassium xanthogenate, and salicylaldoxime, but other chelating reagents such as EDTA, cupferron, and isobutylxanthoic acid hydroxylamine hydrochloride, and semicarbazide showed no inhibitory effect at 10-2 M. Thus , it is hydrochloride (10-3 M) showed no inhibitory effect. not clear whether Cue} participates in the reaction However, flavin enzyme inhibitors such as quina or not. crine (5 x 10-3 M) (21), quinine (5 x 10-3 at) (21), and Further experiments were undertaken to in methylene blue (10-4 M) (22) showed more than vestigate the restorative effects of Cult and Mos+ 85 % inhibition. on enzyme activity which had been decreased by Prosthetic Group of CHA Oxidase-As shown dialysis against solutions of KCN, salicylaldoxime in Table V, quinacrine inhibition was partially or cuprizone, but neither Cult nor Most restored released by FAD or FMN. This suggests that the the activity. Erwin and Hellerman (26) pointed enzyme is a flavoprotein. out that their mitochondria) monoamine oxidase Moreover, treatment with acid ammonium dose not require copper, in spite of the marked sulfate (23) or CaC12 (24) and dialysis against inhibition by 8-hydroxyquinoline. quinacrine or KBr solution (25) were carried out The enzyme activity was inhibited by sulfhy in an attempt to separate the active apoenzyme dryl reagents such as PCMB (5 •~ 10-5 M, 91 from CHA oxidase, but the decrease in enzyme inhibition) and N-ethylmaleimide (5 •~ 10-2 M, 27 activity caused by these treatments was not re- inhibition), but not by iodoacetic acid (5 •~ 10-2 M). covered by the addition of FAD . The inhibition by PCMB could be partially released Fluorescent material could be liberated from by glutathione (Table ‡W). It was thus presumed the enzyme by treatment with trichloroacetic acid that sulfhydryl groups in CHA oxidase participated as described in " MATERIALS AND METHODS ." in the enzyme activity, as in mitochondria) mono- The yellow color and yellow-green fluorescence of amine oxidase. However, the present enzyme is the extracts were bleached by the addition of completely different from the latter enzyme with Na2S2O4, and were restored again by shaking the respect to substrate specificity because it only mixture in air. catalyzes the oxidative deamination of alicyclic Figure 2 shows the absorption spectra of FAD , primary amines. In this respect, CHA oxidase is CHA oxidase, CHA oxidase reduced with CHA , unique. and phenol extracts of CHA oxidase denatured by Carbonyl reagents such as hydrazine sulfate, trichloroacetic acid. The absorption spectrum of

J. Biochem. PURIFICATION AND PROPERTIES OF CYCLOHEXYLAMINE OXIDASE 857

TABLE V. Effects of FAD and FMN on the inhibition of CHA oxidase activity by quinacrine. Enzyme activity was measured under the assay conditions given in "MATERIALS AND METHODS" except for the use of 0.04 M phosphate, pH 6.8, in place of Tris-maleate buffer to prevent the precipitation of quinacrine. The reaction mixtures, containing the purified enzyme, 2 x 10-3 M quinacrine and either FAD or FMN each at 2x10-3 M, were incubated at 30°C for 10 min before starting the reaction by the addition of substrate.

TABLE ‡Y. Thin-layer chromatography of the flavin from CHA oxidase. The purified enzyme was dena tured with trichloroacetic acid as described in " MATE RIALS AND METHODS," and the phenol extracts were spotted on a silica gel film (Tokyo Kasei Co.)- Development was carried out in the dark, and the spots were detected under ultraviolet light.

Fig. 2. Absorption spectra of FAD, CHA oxidase, CHA oxidase reduced with CHA, and the phenol extract of CHA oxidase. The purified enzyme (2.6 mg/ml a Solvent system: A, 5% Na2HPO4; B, tert-amyl protein) in 0.04 M phosphate, pH 6.8, was used. Anaer alcohol: formic acid: H2O (3: 1: 1); C, n-butanol: acetic obic reaction and the preparation of the phenol extract acid : H2O (4: 1: 5) ; D, n-butanol : ethanol: 5% Na2HPOa from CHA oxidase were carried out as described in (10:5:6). "MATERIALS AND METHODS ." ---, FAD; - , CHA oxidase; ---, phenol extract of CHA oxidase; ----, CHA oxidase reduced with CHA. In order to identify the flavin in the enzyme,. phenol extracts of CHA oxidase were examined by extracts of CHA oxidase showed absorption ?? thin-layer chromatography using four solventt at around 450 nm and 375 nm which were systems. The results are shown in Table VI. The ?? FAD . However the flavin nucleotide enzyme extracts gave a single major fluorescent spot ?? shifted to 460 nm in CHA which was identical with authentic FAD. No, ?? ds longer wavelength is spots corresponding to FMN and riboflavin were ??pectra (25). Moreover, detected. These results suggest that CHA oxidase ??C HA oxidase under ??an is a flavoprotein with FAD as the prosthetic group. an apparent reduction of Renwick and Williams (6) suggested that the enzyme. Therefore, the gut flora are responsible for the conversion of CHS ??be an electron acceptor into CHA but not for the further metabolism of ??tivity. this amine. However, we recently observed that :858 T . TOKIEDA, T. NIIMURA, F. TAKAMURA, and T. YAMAHA intestinal microorganisms played an important role 9. Asahina, M. (1972) J. Food Hyg. Soc. Japan in the deamination of CHA in a rabbit receiving (in Japanese) 13, 133 CHA orally. 10. Asahina, M., Niimura, T., Yamaha, T., & Taka Though it is known that the gut contents - hashi, T. (1972) Agr. Biol. Chem. 36, 711 contain oxygen up to a maximum partial pressure 11. Niimura, T., Tokieda, T., & Yamaha, T. (1974) of 40-50 mm Hg (27-29), it is not cleat whether J. Biochem. 75, 407 12. Drasar, B.S., Renwick, A.G., & Williams, R.T. CHnone or CHnol excreted in the urine of mammals (1972) Biochem. J. 129, 881 given CHA is produced only by this enzyme. In 13. Hayashi, C., Iwahara, S., Tanimura, A., Kawamata, the intestinal tract, there is a further possibility K., & Kaneko, T. (1973) Eisei Shikenjo Hokoku that CHA may be deaminated reductively by (in Japanese) 91, 11 hydrogenase or oxidatively by dehydrogenase 14. Bickel, M.H., Burkard, B., Meler-Strasser, E., & independently of oxygen, and a study of bacterial Van Den Broek-Boot, M. (1974) Xenobiotica 4, 425 deaminating enzymes other than CHA oxidase is 15. Lowry, O.H., Rosebrough, N.J., Farr, A.L., & in progress in our laboratory. Randall, R.J. (1951) J. Biol. Chem. 193,265 16. Thomson, J.F. & Morrison, G.R. (1951) Anal. Chem. The authors wish to thank Misses M. Honma, T. Kan 23,1153 bara, and H. Iwahara for technical assistance. 17. Davis, B.J. (1964) Ann. N. Y. Acad. Sci. 121, 404 18. Andrews, P. (1964) Biochem. J. 91, 222 19. Tipton, K.F. (1968) Biochim. Biophys. Acta 159,451 REFERENCES 20. Burton, K. (1951) Biochem. J. 50, 258 1. Price, J.M., Biava, C.G., Oser, B.L., & Vogin, E.-E. 21. Hellerman, L., Lindsay, A., & Bovarnick, M.R. (1970) Science 167, 1131 (1964) J. Biol. Chem. 163, 553 2. Kojima, S. & Ichibagase, H. (1966) Chem. Pharm. 22. Yasunobu, K.T., Igaue, I., & Gomes, B. (1968) Bull. 14, 971 Ad. in Pharmacol. 6A, 43 3. Kojima, S. & Ichibagase, H. (1968) Chem. Pharm. 23. Warburg, O. & Christian, W. (1938) Biochem. Z. Bull. 16, 1851; (1969) 17, 2620 298, 150 4. Davis, T.R.A., Adler, N., & Opsahi, J.C. (1969) 24. Morell, D.B. (1952) Biochem. J. 51, 657 Toxicol. Appl. Pharmacol. 15, 106 25. Massey, V. & Curti, B. (1966) J. Biol. Chem. 241, 5. Prosky, L. & O'Dell, R.G. (1971) J. Pharm. Sci. 60, 3417 1341 26. Erwin, V.G. & Hellerman, L. (1967) J. Biol. Chem. 6. Renwick, A.G. & Williams, R.T. (1972) Biochem. J. 242, 4230 129, 869 27. Crompton, D.W.T. (1965) J. Exp . Biol. 43, 473 7. Asahina, M., Yamaha, T., Watanabe, K. , & 28. Smith, M.H. (1969) Nature 223, 1129 Sarrazine, G. (1971) Chem. Pharm. Bull. 19, 628 29. Soleim, H.A. & Scheline, R.R. (1972) Acta Pharma 8. Asahina, M., Yamaha, T., Sarrazin, G., & Wata col. et Toxicol. 31, 471 nabe, K. (1972) Chem. Pharm. Bull. 20,102

J. Biochem.