Scottish Hospital Endowments Research Trust for a grant (to J. J.). Studies of the Substrate Specificity of Cholesterol Oxidase from. Nocardia erythropolis in the ...
557th MEETING, LIVERPOOL
675
going reduction. However, the carbonyl group of the larger, and more polar, 178-0(3-carboxypropanoyl)glycollyl side chain was not reduced. A 17/3-carboxyl or 178-methoxycarbonyl side chain was not reduced, probably for electronic reasons. In this respect, the reducing power of cortisone reductase resembled that of NaBH4 or lithium tri-t-butoxyaluminium hydride, and was less than that of LiAIH4. Estimates of the binding of various compounds were obtained from inhibition studies. From these it appeared likely that interactions between the enzyme-NADH complex and the 20-0x0 group contribute little to the stability of the ternary enzyme-NADH20-0x0 steroid complex. On the other hand, the 20a-hydroxy configuration (instead of the 20jLhydroxy configuration) evidently generates repulsive forces sufficient to offset the forces that normally bind the rest of the steroid molecule to the enzyme-NAD+ complex. The presence of a hydroxyl group on either of the carbon atoms (C-17 and C-21) adjacent to the reaction site (C-20) did not prevent the enzyme and NAD+ from abstracting the 20a-hydrogen, but in the presence of a hydroxyl group on both of these atoms, this reaction was undetectable. The joint presence of these neighbouring hydroxyl groups would be expected on electronic grounds to make hydride abstraction more difficult, but other enzymes (EC 1.1.1.6 and 1.1.1.94) catalyse the analogous oxidation of the 2-hydroxyl group of glycerol or glycerol phosphate. We thank Professor H. M. Keir for his interest and encouragement, Professor W. Klyne and D. N. Kirk for gifts of reference steroids from the Medical Research Council collection, and the Scottish Hospital Endowments Research Trust for a grant (to J. J.).
Studies of the Substrate Specificity of Cholesterol Oxidase from Nocardia erythropolis in the Oxidation of 3-Hydroxy Steroids ANDREW G. SMITH and CHARLES J. W. BROOKS Department of Chemistry, University of Glasgow, Glasgow G12 8 Q Q , U.K. An extracellular cholesterol oxidase (cholesterol-02 oxidoreductase, EC 1.1.3.6) has been demonstrated in cultures of Streptomyces violascens (Fukuda et al., 1973) and has been obtained in a crystalline form from the culture fluid of Brevibacterium sterolicum (Uwajima et af., 1973, 1974). The purification of cholesterol oxidase from Nocardia species has been reported (Flegg, 1973; Richmond, 1973), and this enzyme is now commercially available for the clinical estimation of cholesterol. The enzyme effects the oxidation of cholesterol to 4-cholesten-3-one in the presence of oxygen (with the concomitant formation of H202),but it is not specificfor cholesterol (Flegg, 1973;Richmond, 1973; Allain et al., 1974; Uwajima et al., 1974). We have used the enzyme in the characterization of 26-hydroxycholestero1 (Smith et al., 1974) and in the analysis of 38hydroxy steroids by g.1.c. (Smith & Brooks, 1974). A more detailed investigation of the substrate specificity of cholesterol oxidase from Nocardia erythropolis is now reported. The enzyme comprises a 38-hydroxy steroid oxidase and a A5-steroid isomerase component. To determine the rate of steroid oxidation alone, the H202 generated was measured by the oxidative coupling of an excess of horseradish peroxidase (HzOz oxidoreductase, EC 1.1.11.7) to give a quinone-imine dye (Trinder, 1969; Allain et al., 1974). were mixed with 2.75m1 of 50 ~ M - N ~ H , P O ~ - N ~ ~ H P O ~ Steroids in propan-2-01(50~1) buffer (pH7.0) containing 1mg of Triton X-IOO/ml, 0.1 ml of 4:aminoantipyrine ( 2 . 4 m ~ ) and 0.1 mI of phenol ( 0 . 4 ~in ) a cuvette of 1cm light-pathmaintained at 30°C. Incubations were initiated by the addition of peroxidase [0.18 unit (,umol/min)] and cholesterol oxidase (usually 0.02 unit) in buffer (10~1).The rate of production of H2O2was indicated by the increase in extinction at 500nm. The initial rates determined for seven VOI. 3
23
676
BIOCHEMICAL SOCIETY TRANSACTIONS
Table 1. Apparent K,,, and V values for the oxidation of some 3-hydroxy steroids by cholesterol oxidase The assay procedures at 30°C and pH7.0 were as described in the text. S.D.values are given in parentheses. K U l V KmI V 3-Hydroxy steroid 01M) (pmol. min-' * mg-') Cholesterol 2.9 (0.2) 18.5 (0.4) 0.16 Side-chain variants 26-Hydroxycholesterol 1.2 (0.05) 10.4 (0.1) 0.12 25-Hydroxy-27-norcholesterol 1.4 (0.1) 11.6 (0.1) 0.12 5-Cholene-38-24-diol 4.8 (0.7) 15.8 (0.5) 0.30 23,24-Dinor-5-cholene-3~,22-diol 9.3 (0.7) 13.5 (0.5) 0.69 0.90 S-Pregnene-3B,208-diol 3.7 (0.5) 4.1 (0.1) 17B-(Hydroxymethyl)-5androsten3/3-01 3.1 (0.3) 0.32 (0.02) 9.7 5-Androstene-3j?,178-diol 1.2 (0.2) 0.1 1 (0.01) 10.9 20,25-Diazacholesterol 124 (4.3) 1.45 (0.03) 86 Nuclear variants Sa-Cholestan-3~-01 2.3 (0.1) 14.5 (0.2) 0.16 4-Cholesten-38-01 3.4 (0.3) 20.6 (0.5) 0.17 5a-Cholest-7-en-38-01 10.7 (0.4) 11.8 (0.2) 0.91 5a-Cholest-8(14)-en-38-01 2.8 (0.3) 3.1 (0.1) 0.90 5,7-Cholestadien-38-ol 5.5 (0.5) 5.9 (0.2) 0.93 4,4-Dimethylcholesterol 7.5 (0.5) 0.028 (0.001) 268 4a-Methylcholestero1 12.6 (1.3) 0.038 (0.002) 332 4jl-Methylcholesterol 28.4 (3.8) 0.34 (0.03) 83.5 * SB-Cholestan-38-01 * 5-Cholesten-3a-01 * 4-Cholesten-3a-ol 58-Cholestan-3a-ol t Sa-Cholestan-28-01 t Cholecalciferol t
* Detectable rate but too low for kinetic measurements. t No detectable oxidation. substrate concentrations in duplicate, or five in triplicate, were used to determine kinetic constants. Provisional values of K,,, @M) and V (pmol-min-' .mg-' of enzyme) were obtained by the graphical method of Eisenthal & Cornish-Bowden (1974). These were then estimated more precisely by using a computer program for a least-squares adjustment of the Michaelis-Menten curve (Wentworth, 1965) on a Digico Micro 16P computer. The products of enzymic oxidations were identified by g.1.c.-mass spectrometry. A series of A5-38-hydroxy steroids with hydroxylated side chains was used to explore the requirements of cholesterol for side chain length (Table 1). 17~-(Hydroxymethyl)-5with no side chain, androsten-38-01, with a C1 side chain, and S-androstene-3/?,17jl-diol were only slowly oxidized by the enzyme, and the K,,,/Vvalues (which may be broadly regarded as an inverse measure of enzyme efficiency) were approx. 60 times as high as for (C, and cholesterol. With S-pregnene-38,208-dioland 23,24-dinor-5-cholene-38,22-diol C3 side chains respectively) oxidation proceeded at a relatively high rate. Though 25-hydroxy-27-norcholesteroland 26-hydroxycholestero1,with C, and C8 side chains respectively, were oxidized more slowly than cholesterol, the three Km/V values were comparable. The changes in K,,, values in this series were only eightfold, as compared with a 58-fold change in V, suggesting that the G17 side chain does not greatly affect the
1975
557th MEETING, LIVERPOOL
677
degree of binding but that it may serve to orientate the steroid correctly relative to the active site concerned with oxidation. The very poor binding of 20,25-diazacholesterol to the enzyme is attributed to an adverse effect of the two basic centres: we have also noticed that the alkaloid solasodine is virtually unattacked by the enzyme, whereas its oxygen analogue, diosgenin, is a moderately good substrate. 5a-Cholestan-38-01 and 4-cholesten-3~-ol were comparable with cholesterol in their susceptibility to oxidation, but 5a-cholest-7-en-3/3-ol, 5acholest-8( 14)-en-38-01 and 5,7-cholestadien-3~-01were poorer substrates. 4,CDimethyl groups greatly decreased the rate of oxidation, and 4a-methylcholesterol was oxidized only oneeighth as fast as the 4 8 isomer, suggesting steric interference, particularly by the 4a-methyl group, in the removal of the 3a-hydrogen. (Relative-rate studies also showed that 48-hydroxycholestero1 was oxidized five times faster than the 4a-isomer.) Little or no oxidation was observed with 5~-cholestan-3~-ol or with 3a-hydroxy steroids. Sa-Cholestan-2&01 was not a substrate. Cholecalciferol has been reported to be a substrate for certain enzymes acting on cholesterol (Fraser & Kodicek, 1968), but was not oxidized by cholesterol oxidase. We thank the Medical Research Council for financial support (to C. J. W. B. and Professor W. A. Harland), Mr. D. Giles (Boehringer, London W.5, U.K.) for the gift ofcholesteroloxidase, Professor R. P. Cook, Dr. J. Gilmore,Dr. R.M. Moriarty, Dr. H. H. Rees, Dr. G. F. Woods and the M.R.C. Steroid Reference Collection for the gift of steroids, and Professor B. Capon for the use of his computer program. Allain, C. C., Poon, L. S., Chan, C. S. G., Richmond,W. & Fu, P. C. (1974) Clin.Chem. (WinsfonSalem, N.C.) 20,470475
Eisenthal, R. & Cornish-Bowden, A. (1974) Biochem. J. 139,715-720 Flegg, H. M. (1973) Ann. Clin.Biochem. 10,79-94 Fraser, D. R. & Kodicek, E. (1968) Nafure (London) 220,1031-1032 Fukuda, H., Kawakami, Y. & Nakamura, S. (1973) Chem. Pharm. Bull. 21,2057-2060 Richmond, W. (1973) Clin.Chem. ( Winston-Salem,N.C.) 19,1350-1356 Smith, A. G. & Brooks, C. J. W. (1974) J. Chromatog. 101,373-378 Smith, A. G., Gilbert, J. D., Harland, W. A. & Brooks, C. J. W. (1974) Biochem. J. 139,793795
Tinder, P. (1969) Ann. Clin.Biochem. 6 , 2 4 2 7 Uwajima, T., Yagi, H., Nakamura, S. & Terada, 0. (1973) Agri. Biol. Chem. 37,2345-2350 Uwajima, T., Yagi, H. & Terada, 0. (1974) Agri. Biol. Chem. 38,1149-1156 Wentworth, W. E. (1965) J. Chem. Educ. 42,96-103
Testosterone Metabolism in Superfused Human Hyperplastic Prostate GRAHAM H. BEASTALL University Department of Steroid Biochemistry, Royal Infirmary, Glasgow G4 OSF, U.K. Since the observation by Bruchovsky & Wilson (1968) that testosterone was metabolized into dihydrotestosterone (178-hydroxy-5a-androstan-3-one)by rat ventral prostate, considerable attention has been focused on the metabolism of testosterone in the male accessory sex glands that are known target organs for the hormone. A wide distribution of testosterone 5a-reductase has been established (King & Mainwaring, 1974), and the almost universal presence of dihydrotestosterone in these tissues has led to the view that it is the active hormone for promoting androgen-induced growth (Robel, 1971; Wilson, 1972). However, the different techniques and experimental conditions used to study testosterone metabolism have led to inconsistencies as to the quantitative importance of dihydrotestosterone and the presence, or absence, of minor testosterone metabolites. Workers in this laboratory have used a superfusion technique to study steroid dynamics in slices of prostate, and conditions have been established, in which the tissue is in a steady state with respect to steroid uptake and metabolism
Vol. 3
