Steroid transforming enzymes from microorganisms--III. Properties of 4-ene-3-oxosteroid-5alpha-reductase from Mycobacterium smegmatis

Aerobic catabolism of sterols by microorganisms: key enzymes that open the 3-ketosteroid nucleus

Aerobic degradation of the sterol tetracyclic nucleus by microorganisms comprises the catabolism of A/B-rings, followed by that of C/D-rings. B-ring rupture at the C9,10-position is a key step involving 3-ketosteroid Δ1-dehydrogenase (KstD) and 3-ketosteroid 9α-hydroxylase (KstH). Their activities lead to the aromatization of C4,5-en-containing A-ring causing the rupture of B-ring. C4,5α-hydrogenated 3-ketosteroid could be produced by the growing microorganism containing a 5α-reductase. In this case, the microorganism synthesizes, in addition to KstD and KstH, a 3-ketosteroid Δ4-(5α)-dehydrogenase (Kst4D) in order to produce the A-ring aromatization, and consequently B-ring rupture. KstD and Kst4D are FAD-dependent oxidoreductases. KstH is composed of a reductase and a monooxygenase. This last component is the catalytic unit; it contains a Rieske-[2Fe-2S] center with a non-haem mononuclear iron in the active site. Published data regarding these enzymes are reviewed.

Steroid 5alpha-reductase activity of Treponema denticola

INTRODUCTION

Previously we have shown that reference and freshly isolated Treponema denticola cultures possess 5alpha-reductase (5alpha-R) and 3beta- and 17beta-hydroxysteroid dehydrogenase activity. A gene matching the 3-oxo-5alpha-steroid 4-dehydrogenase family protein (gene ID: 2739284; locus tag: TDE2697) has been identified in T. denticola ATCC 35405. The aim of the work presented here was to optimize assay conditions and determine steroid substrate specificities for the 5alpha-R activity of T. denticola ATCC 33520.

METHODS

5alpha-R activity of cell-free preparations was assayed with radioactive steroid substrates. 5alpha-R-reduced products were identified using thin-layer chromatography and a radioisotope scanner. Assay conditions were optimized for co-factor, buffer and pH requirements. Apparent substrate specificities were determined for progesterone, 4-androstenedione, testosterone and corticosterone. The time-course for metabolism of radiolabelled progesterone and cholesterol substrates was investigated with anaerobic cultures.

RESULTS

The optimum pH for 5alpha-R was 5.5 and the preferred co-factor was NADPH. The order of the steroids with respect to their 5alpha-R substrate specificities was (in descending order): progesterone, 4-androstenedione, testosterone and corticosterone. There are at least two intermediates in the synthesis of 5alpha-dihydrocholesterol from cholesterol.

CONCLUSION

These results suggest that the 3-oxo-5alpha-steroid 4-dehydrogenase family protein gene of T. denticola codes for a functional protein that resembles mammalian 5alpha-R isoenzyme 2 with regard to co-factor requirement and pH optimum.

Properties of a 4-ene-3-ketosteroid-5 alpha-reductase in cell extracts of the intestinal anaerobe, Eubacterium sp. strain 144

When Eubacterium sp. 144 was grown in the presence of progesterone, extracts of these cells contained a 4-ene-3-ketosteroid-5 alpha-reductase (5 alpha-reductase). No evidence for the presence of a 5 beta-steroid-reductase or a 5 alpha to 5 beta-steroid-isomerase was found. 5 alpha-Reductase activity was dependent on reduced methyl viologen as the electron donor and this could be generated biologically by adding pyruvate or H2 to cell extracts or chemically by adding sodium dithionite. NADH or NADPH with or without flavin nucleotides were not electron donors for 5 alpha-reductase. Most of the 5 alpha-reductase activity (60-65%) of crude extracts was located in the membrane fraction and the enzyme was solubilized by treatment with 1% Triton X-100. Optimum 5 alpha-reductase activity occurred at pH 7.0-7.5 in potassium phosphate buffer but was stimulated by Tris-HCl buffer (pH 8.0-9.0). 5 alpha-Reductase activity was highest at 10% (v/v) methanol and was progressively inhibited by higher methanol concentrations. Sulfhydryl reagents strongly inhibited 5 alpha-reductase but the enzyme was not affected by other metabolic inhibitors. Extracts prepared from cells induced with 16-dehydroprogesterone and grown without hemin contained 5 alpha-reductase and 16-dehydroprogesterone-reductase activities equivalent to those found in extracts of induced cells grown with hemin. This indicates that hemin is not required for the synthesis of active steroid double bond-reductases in strain 144.

1-Ene-steroid reductase of Mycobacterium sp. NRRL B-3805

The microbial enzymatic reduction of 1,4-androstadiene-3,17-dione (ADD) to 4-androstene-3,17-dione (AD), testosterone and 1-dehydrotestosterone (DHT) is described. Two reducing activities observed in washed cell suspensions and cell free extracts of Mycobacterium sp. NRRL B-3805 were found to account for these bioconversions. One was a 1-ene-steroid reductase and the other a 17-keto steroid reductase. The first reducing activity was found to appear in the soluble cell fraction whereas the latter could be precipitated by centrifugation. Maximum 1-ene-steroid reductase specific activity was achieved during the exponential growth phase of the organism and significantly increased upon induction with ADD. The 1-ene-steroid reductase was partially purified (30-fold) by ammonium sulfate fractionation, gel-filtration and ion-exchange chromatography, and was eluted from a Sephacryl S-300 column with an Mr = 115,000. The 1-ene-steroid reductase activity was NADPH-dependent and had specificity towards steroid compounds containing C-1,2 double bond with an apparent Km for ADD of 2.2 X 10(-5) M. The reverse reaction catalyzing C-1,2 dehydrogenation could not be detected in our preparations. The results suggest that in Mycobacterium sp NRRL B-3805 and B-3683 the steroid C-1,2 dehydrogenation and 1-ene reduction are two separable activities.