Characterization of an efficient estrogen-degrading bacterium Stenotrophomonas maltophilia SJTH1 in saline-, alkaline-, heavy metal-contained environments or solid soil and identification of four 17ß-estradiol-oxidizing dehydrogenases.

The efficient bioremediation of estrogen contamination in complex environments is of great concern. Here the strain Stenotrophomonas maltophilia SJTH1 was found with great and stable estrogen-degradation efficiency even under stress environments. The strain could utilize 17ß-estradiol (E2) as a carbon source and degrade 90% of 10 mg/L E2 in a week; estrone (E1) was the first degrading intermediate of E2. Notably, diverse pH conditions (3.0-11.0) and supplements of 4% salinity, 6.25 mg/L of heavy metal (Cd2+ or Cu2+), or 1 CMC of surfactant (Tween 80/ Triton X-100) had little effect on its cell growth and estrogen degradation. The addition of low concentrations of copper and Tween 80 even promoted its E2 degradation. Bioaugmentation of strain SJTH1 into solid clay soil achieved over 80% removal of E2 contamination (10 mg/kg) within two weeks. Further, the whole genome sequence of S. maltophilia SJTH1 was obtained, and a series of potential genes participating in stress-tolerance and estrogen-degradation were predicted. Four dehydrogenases similar to 17ß-hydroxysteroid dehydrogenases (17ß-HSDs) were found to be induced by E2, and the four heterogenous-expressed enzymes could oxidize E2 into E1 efficiently. This work could promote bioremediation appliance potential with microorganisms and biodegradation mechanism study of estrogens in complex real environments.

Studies on the three-dimensional structure of estrogenic 17 beta-hydroxysteroid dehydrogenase.

The structure-function relationship of the estrogenic 17 beta-hydroxysteroid dehydrogenase (17 beta-HSD1), a pivotal enzyme in the synthesis of active sex hormones, has been studied via protein chemistry and crystallography. A highly active and homogeneous 17 beta-HSD1 was prepared with a rapid purification from human placenta. We then characterized the native and expressed enzyme, and concluded, for the first time, that 17 beta-HSD1 is formed by two identical subunits. The enzyme was also overproduced in insect cells with a baculovirus expression system. The highly active 17 beta-HSD1 preparation was successfully crystallized in the presence of NADP-, polyethylene glycol, beta-octylglucoside and glycerol, resulting in the first diffraction quality crystals of any steroid-converting enzyme from a human source. The three-dimensional structure of 17 beta-HSD1 was determined at 2.2 A resolution, showing that the overall structure of the enzyme is similar to the other enzymes in the short-chain dehydrogenase family, with a conserved Tyr-X-X-X-Lys sequence and a serine residue in the active site. It is distinguished from the other known structures reported for short-chain dehydrogenases by the insertion of two helix-turn-helix motifs that appear to govern membrane association and substrate specificity [corrected]. More recently, the complex of 17 beta-HSD1 with estradiol has been successfully crystallized and its structure determined. The latter demonstrates detailed information of the interactions between the substrate and residues Ser142, Tyr155, His221 and Glu282 of the enzyme. These interactions and the complementarity of the substrate with the hydrophobic binding pocket make critical contributions to the enzyme specificity. The above results provide a strong basis for the design of potent inhibitors of this pivotal steroid dehydrogenase.

Purification and properties of oestradiol 17 beta-dehydrogenase extracted from cytoplasmic vesicles of porcine endometrial cells.

Porcine endometrial oestradiol-17 beta dehydrogenase was solubilized from the particulate fraction of homogenates sedimenting between 1200 g and 10,000 g by treatment with 0.4% Brij 35 in neutral buffers. The extracts were processed by successive passage through DEAE-Sepharose, Amberlyte XAD-2 and Blue-Sepharose, and the enzyme was collected from the washed affinity matrix at 0.8 M of a 0-2 M-KCl gradient. A genuine oestrone reductase was eluted at 1.9 M-KCl. The dehydrogenase pool was resolved by butyl-Sepharose chromatography into a major (80%) peak (EDHM) eluted at 0.8 M-(NH4)2SO4 and a very hydrophobic fraction (VHF) recovered at 0.1 M. EDHM was further purified by filtration through Sephadex G-200 and cation-exchange chromatography on Mono S. Sephacryl 300 was used for VHF followed by Mono S. Enrichments from the homogenate amounted to 1074-fold for EDHM and 632-fold for VHF. A single silver-stained band at 32 kDa is seen on SDS/PAGE of EDHM, and VHF contains additional bands at 45 and 80 kDa. Polyclonal antibodies (G436) raised against EDHM and the monoclonal antibody F1 raised against VHF recognize the single 32 kDa band in EDHM and both the 32 kDa and 80 kDa bands in composite VHF. The 45 kDa band of VHF reacts with neither. Monoclonal antibody W1 raised against EDHM only recognizes the 32 kDa peptide of EDHM and VHF. The specific activity for oestradiol oxidation amounts to 4081 mu-units/mg for EDHM and to 2402 mu-units/mg for VHF. Both possess a minimal (1/260) endogenous reductase activity and are devoid of 3 beta, 3 alpha- and 20 alpha-dehydrogenases. We consider EDHM to be authentic oestradiol-17 beta dehydrogenase of porcine endometrium. The composite VHF could reflect the situation of the enzyme in vivo or result from aggregations occurring during processing.

Oestradiol synthesis from oestrone in malignant breast epithelial cells: studies on a high affinity, 80 kDa form of oestradiol dehydrogenase.

Previous studies have shown that in the breast there are multiple forms of the enzyme oestradiol dehydrogenase (E2DH), responsible for the interconversion of oestrone (E1) to oestradiol (E2). We have now re-examined oestrogen metabolism in the breast cancer cell lines (T47D and MCF-7) and have shown that steroids previously shown to inhibit the conversion of E1 to E2 in normal breast tissue failed to do so when added to growing monolayers of these malignant cells. In contrast to earlier estimates in normal breast tissues, the apparent Km for this conversion in monolayers of these malignant cells is shown here to be considerably lower, at around 50 nM. Cell free studies on these cell lines have revealed the presence of a high affinity (for E1) form of this enzyme of Mw approximately 80 kDa. The ability to detect this enzyme in soluble cell fractions appears to be critically dependent on buffer composition. Normal breast epithelial cells and adipose tissue appear to be devoid of this form of E2DH. As this form of E2DH has the highest affinity for the substrate E1 of all the forms in the breast, it is probable that this 80 kDa enzyme is responsible for the conversion of E1 to E2 in cell monolayers. If the observation holds that the 80 kDa enzyme is absent in the normal tissues, then the possibility arises that this E2DH may be linked with the neoplastic process in some breast tumours containing malignant epithelial cells of a similar type as studied here.

The membrane-bound 17ß-estradiol dehydrogenase of porcine endometrial cells: purification, characterization and subcellular localization.

The membrane-bound 17ß-estradiol dehydrogenase of porcine endometrial cells was purified to homogeneity as judged by SDS-PAGE and silver staining of a single 32 kD band. A second, more hydrophobic product of the purification protocol contained additional bands at 45 and 80 kD. The 17ß-estradiol dehydrogenase activities of both products exceeded those for 17-one reduction by more than 260-fold. Activities of 3α-, 3ß- and 20α-dehydrogenases were absent in either fraction. Polyclonal and monoclonal antibodies raised against the 32 kD protein and the more hydrophobic product precipitated the enzymatic activity and reacted with the 32 and 80 kD bands, but not with the 45 kD band in Western blots. The subcellular localization of the enzyme was studied in sections of intact cells and of isolated organelles using gold sol coated with F(ab')2 fragments of monoclonal antibody F1. Gold particles were found exclusively over cytoplasmic vesicles of 120-150 nm diameter with electron-dense contents.

Hydrogen bonding in steroidogenesis: studies on new heterocyclic analogs of estrone that inhibit human estradiol 17 beta-dehydrogenase.

New heterocyclic analogs of estrone are reported that inhibit estradiol 17 beta-dehydrogenase (E2-17 beta DH) from human placenta. The inhibitors are efficiently synthesized in two steps from estrone (or its 3-O-methyl ether), giving fully characterized analogs with pyrazole or isoxazole fused to the 16,17-position on the D ring. Dixon plots of enzyme kinetic data show the heterocyclic steroids are competitive inhibitors of E2-17 beta DH. Correlating molecular structures of the inhibitors with their Ki-values yields a pattern suggesting intermolecular hydrogen bonding stabilizes the [(pyrazole)inhibitor-E2-17 beta DH] complexes. A free energy difference of 2.74 Kcal/mol calculated from Ki-value differences between hydrogen bonded (4.08 microM) and non-bonded (425 microM) [inhibitor-E2-17 beta DH] complexes is in the range for intermolecular hydrogen bonding. We conclude that specific intermolecular hydrogen bonds stabilize [hydroxysteroid-enzyme] complexes, thereby making important contributions to the affinity between hydroxysteroids and steroid-specific enzymes of steroidogenesis.

3-Keto reductase activity of estradiol-17 beta dehydrogenase and its contribution to androgen metabolism.

Hepatic estradiol-17 beta dehydrogenase from chickens catalyzed the reduction of the 3-keto group of androgens such as 5 alpha-dihydrotestosterone and 5 alpha-androstane-3,17-dione as well as the 17-keto reduction of 4-androstene-3,17-dione and dehydroepiandrosterone. The reaction products from 17-ketosteroid and 3-ketosteroid substrates were identified as 17 beta-hydroxysteroids and 3 beta-hydroxysteroids, respectively, by thin layer chromatography, high performance liquid chromatography and gas chromatography. Barbital inhibited both 17 beta-estradiol dehydrogenase and 5 alpha-androstane-3,17-dione reductase activity noncompetitively giving the same kinetic constant, Ki = 50 microM. 5 alpha-Androstane-3,17-dione competitively inhibited 17 beta-estradiol dehydrogenase activity. These results indicate that chicken liver estradiol-17 beta dehydrogenase is in fact a 3 beta- and 17 beta-hydroxysteroid dehydrogenase and that both 3- and 17-ketosteroids bind to the same catalytic site.

Estradiol 17 beta-dehydrogenase: full enzymatic activity in the absence of zinc.

The precise catalytic mechanism of the steroid interconverting enzyme, human placental estradiol 17 beta-dehydrogenase (EC 1.1.1.62, estradiol-17 beta:NAD+ 17-oxidoreductase), is not known. Two general models for the catalytic mechanism of dehydrogenases have been defined. One model requires Zn2+ metal for the catalytic event, as has been shown for horse liver alcohol dehydrogenase (EC 1.1.1.1, alcohol:NAD+ oxidoreductase). Another model has been demonstrated for the 2-hydroxy acid dehydrogenases in which histidine residues are necessary for enzyme activity, without participation of a metal ion. In order to define which mechanism might be operative for the placental enzyme, it became important to determine whether Zn2+, or another metal ion, is associated with the macromolecule. Several homogeneous enzyme preparations, having protein concentrations from 5-80 microM, were extensively dialyzed in a buffer containing EDTA. Atomic absorption analysis of each sample demonstrated that no Zn2+ was present, although the enzymatic activity was maintained. In addition, there was no significant detection of Mg2+ or Mn2+ above background levels. When the isolated enzyme was dialyzed against buffer containing added 0.01-20 microM ZnCl2, no increase in specific activity of the enzyme was seen. The data indicate that the presence of zinc is not required for the catalytic event. These results, together with our previous affinity-labeling studies, which demonstrate a histidine residue in the catalytic region of the active site, allow us to propose that the catalytic mechanism of the human placental estradiol 17 beta-dehydrogenase is similar to that of the 2-hydroxy acid dehydrogenases.

Human placental 17 beta-estradiol dehydrogenase: enzymatic activity in the last weeks of pregnancy and electrophoretic data.

A specific electrophoretic method for human placental 17 beta-estradiol dehydrogenase (17-HSD; EC 1.1.1.62) has been performed and a sample of about 180 placentae from Italian women has been examined. A common phenotype and only one electrophoretic variant have been observed. Also 17-HSD activity has been tested. A statistically significant negative correlation has been found between 17-HSD activity and both gestational age and birth weight in the last weeks of gestation in a group of at term newborns with weight appropriate for date. This reduction in enzymatic activity is in good agreement with the data on estrone and estradiol which both show a declining rate of increase in the last weeks of pregnancy.

Human placental estradiol 17 beta-dehydrogenase: sequence of a histidine-bearing peptide in the catalytic region.

The amino acid sequence of an octapeptide from the catalytic site of human placental estradiol 17 beta-dehydrogenase (EC 1.1.1.62) was established by affinity-labeling techniques. The enzyme was inactivated separately by 12 beta-hydroxy-4-estrene-3,17-dione 12-(bromo[2-14C]acetate) and 3-methoxyestriol 16-(bromo[2-14C]acetate) at pH 6.3. The inactivations, in both cases, followed pseudo-first-order kinetics with half-times for the 12 beta and 16 alpha derivatives being 192 and 68 h, respectively. Both derivatives are known substrates that inactivate in a time-dependent, irreversible manner and that modify cysteine residues to form (carboxymethyl)cysteine and histidine residues to form either N tau- or N pi-(carboxymethyl)histidine. The inactivated enzyme samples were separately reduced, carboxymethylated, and digested with trypsin. The tryptic digests were applied to Sephadex G-50 and the radioactive N tau- and N phi-(carboxymethyl)histidine-bearing peptides identified. The peptides were further purified by cation-exchange chromatography and gel filtration. Final purification was achieved by HPLC prior to sequencing. It was determined that both steroid derivatives modified either of the two histidine residues in the peptide Thr-Asp-Ile-His-Thr-Phe-His-Arg. These histidines are different from a histidine that was previously shown to be alkylated by estrone 3-(bromoacetate) and that was presumed to proximate the A ring of the bound steroid. It is concluded that the two histidine residues identified in the present study proximate the D ring of the steroid as it binds at the active site and may participate in the hydrogen transfer effected by human placental estradiol 17 beta-dehydrogenase.

A single step procedure for purification of estradiol 17 beta-dehydrogenase from human placenta.

A new and simple procedure for purification to homogeneity of the soluble estradiol 17 beta-dehydrogenase [E C.1.1.1.62] from human placenta has been developed. The purification is achieved in a single step by affinity-absorption of Cibachrome Blue F3G-A coupled to Sepharose and selective elution with NADP+. Homogeneous estradiol 17 beta-dehydrogenase has a specific activity of 7.2 units/mg. and has been purified 2400-fold with a 93% recovery.

Oestradiol-17 alpha dehydrogenase from chicken liver.

The NADP+-linked oestradiol-17 alpha dehydrogenase (EC 1.1.1.148) present in cell-free extracts of chicken liver was investigated with the aim of separating it from a closely related oestradiol-17 beta dehydrogenase (EC 1.1.1.62) found in the same subcellular fraction. However, its chromatographic behaviour on CM-cellulose and DEAE-cellulose was almost identical with that previously reported for the latter enzyme, including resolution into two peaks on the anion-exchanger. Both peaks contained oestradiol-17 alpha dehydrogenase and oestradiol-17 beta dehydrogenase activity. Further attempts to separate the putative enzymes by dye-ligand chromatography with the use of the dyes Procion Yellow, Reactive Red and Cibachron Blue linked to Sepharose were unsuccessful, and they behaved identically on affinity columns of adenosine 2',5'-bisphosphate-agarose and 17 beta-oestradiol 3-hemisuccinate bound to Sepharose. A previous report of partial separation on Sephadex G-200 was not confirmed. Slab gel electrophoresis of enzyme preparations after affinity chromatography on adenosine 2',5'-bisphosphate-agarose revealed multiple bands in systems containing sodium dodecyl sulphate, whereas analysis by rod gel electrophoresis gave two major and one minor bands that stained coincidently for oestradiol-17 alpha dehydrogenase, oestradiol-17 beta dehydrogenase, epitestosterone dehydrogenase and testosterone dehydrogenase activities. Isoelectric focusing gave four enzymically active peaks that each oxidized oestradiol-17 alpha and -17 beta. Apparent Km values for the two forms of oestradiol-17 alpha dehydrogenase obtained by DEAE-cellulose chromatography were 17 and 23 microM for oestradiol-17 alpha, and 8.7 and 11.0 microM for NADP+. Limited kinetic studies with oestradiol-17 alpha and -17 beta with the use of the mixed-substrate method showed that the total velocity was equal to the sum of the separate velocities. The active-site inhibitor-alkylating agent 17 beta-(1-oxoprop-2-ynyl)androst-4-en-3-one did not cause time- or temperature-dependent inhibition, in contrast with the reported case of the oestradiol-17 beta dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase activities of the human placental oestradiol dehydrogenase. NADP+ appeared to afford some protection against inhibition. Investigation of substrate specificity with a limited range of steroids suggests that the enzyme(s) from chicken liver differs substantially from the oestradiol-17 beta dehydrogenase from human placenta, and although the evidence is not conclusive it suggests the existence of one enzyme.

Dehydroepiandrosterone is a substrate for estradiol 17 beta-dehydrogenase from human placenta.

The enzyme estradiol 17 beta-dehydrogenase (17 beta-ED) [E.C.1.1.1.62] from human placenta was purified to homogeneity by the initial steps of a published procedure, followed by an affinity chromatography step in Reactive Blue 2-Sepharose, eluting with NADP. The pure enzyme is not specific for estrogenic substrates, it also catalyzes the oxidation-reduction of several androgens and progesterone (i.e. dehydroepiandrosterone, androstenedione, 5 alpha-dihydrotestosterone, and 20 alpha-dihydroprogesterone). The comparison of the kinetic parameters for these substrates, shows that dehydroepiandrosterone could be a physiological ligand of the enzyme, and consequently involved in the control of its function in estrogen metabolism.

Purification and characterization of epimeric estradiol dehydrogenases (17 alpha and 17 beta) from equine placenta.

Estradiol 17 alpha-dehydrogenase (EC 1.1.1.148) and estradiol 17 beta-dehydrogenase (EC 1.1.1.62) from horse placenta have been purified to homogeneity. Both enzymes are localized in the microsomal fraction and are solubilized in 1.5% sodium cholate. The 17 alpha- and 17 beta-dehydrogenases are separated by selective elution from hydroxylapatite with 0.5 and 1.0 M potassium phosphate, respectively. Subsequent purification is achieved by two affinity-absorption steps using reactive blue 2-agarose and estriol hemisuccinate-Sepharose. Homogeneous estradiol 17 alpha-dehydrogenase has a specific activity of 10 IU/mg and has been purified 5900-fold with an 87% recovery. Homogeneous estradiol 17 beta-dehydrogenase has a specific activity of 10.6 IU/mg and has been purified 15 000-fold with an apparent recovery of 100%. Each enzyme exhibits a single band on polyacrylamide disc gel electrophoresis and on sodium dodecyl sulfate (SDS) gel electrophoresis. The mobilities of the two on SDS gels are identical and correspond to subunit molecular weights of 33000. The apparent molecular weight of the undenatured, active enzyme, as determined by gel filtration on Sephacryl S-300, is 52000 in the case of the 17 alpha-dehydrogenase and 68500 for the 17 beta-dehydrogenase. Both enzymes exhibit pH optima at 9.0-9.5; both prefer NAD+ over NADP+ but utilize both cofactors. Both are highly specific for their respective epimers of estradiol with apparent Km values of 1.7 microM at pH 9.5.

Study of human placental estradiol-17 beta dehydrogenase/20 alpha-hydroxysteroid dehydrogenase by preparative disc-gel electrophoresis.

The soluble enzyme, estradiol-17 beta dehydrogenase from human term placenta, appears to co-purify with a second soluble enzyme, 20 alpha-hydroxysteroid dehydrogenase. The enzyme, which had been partially purified by affinity chromatography, fractionated on a preparative electrophoresis gel to a homogeneous preparation containing both estradiol-17 beta dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase activities in a ratio of approximately 100:1. Analytical polyacrylamide disc-gels resolved this homogeneous preparation as a single band by both protein and activity staining techniques. Homogeneous enzyme inactivated and affinity-radioalkylated by 16 alpha-[2'-14C]bromoacetoxyprogesterone or 16 alpha-[2'-14C] bromoacetoxyestradiol 3-methyl ether, and when analyzed by SDS disc-gel electrophoresis, gave a single protein band which corresponded identically to the radioactivity peaks. These observations support the hypothesis that estradiol-17 beta dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase represent dual oxidoreductase activity in one enzyme. Preparative disc-gel electrophoresis, a technique which has not been previously adapted to purification of these human placental enzyme activities, was useful to rapidly (3 days) effect a 15-fold enrichment of the estradiol-17 beta dehydrogenase specific activity from "heat-treated cytosol". Thus, laboratory-scale preparative disc-gel electrophoresis is useful for rapid, small-scale enrichment of this soluble enzyme.

Estradiol-17 beta dehydrogenase from chicken liver.

NADP+-linked estradiol-17 beta dehydrogenase has been purified 300- to 400-fold from cell-free extracts of chicken liver in a 20 to 30% yield by ammonium sulfate precipitation, ion exchange chromatography, and gel filtration. The enzyme is stable for at least 3 months when stored at -20 degrees C in buffer containing glycerol (50%, v/v). Two forms, with molecular weights of 43,000 and 97,000 are present; these show one major band (Rm = 0.27) and one minor band (Rm = 0.25) on polyacrylamide disc gel electrophoresis. (Rm is defined as the ratio of the distance migrated by the protein band to that of the tracking dye.) The species of lower molecular weight is the more active, with apparent Km values for estradiol-17 beta of 25 and 17.3 microM in the presence and absence, respectively, of bovine serum albumin in the assay medium. The apparent Km for NADP+ is 7.7 microM, and the optimum pH for dehydrogenation is 9.9. The lower molecular weight form has a lambda max at 280 nm, a shoulder at 290 nm, and an A 1% 1 cm of 12.1 at 280 nm. The fluorescence spectrum corresponds to that of a tryptophan-containing protein with lambda max at 288 nm. Isoelectric focusing in gel at pH 5 to 8 shows three major bands of pI 6.9, 6.8, and 6.0. Cross-linking with dimethyl suberimidate followed by electrophoresis reveals five bands. The enzyme is affected by thio reagents and possesses no associated estradiol-sensitive transhydrogenase activity.

Human placental 17 beta-estradiol dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase. Two activities at a single enzyme active site.

Two soluble enzyme activities, 17 beta-estradiol dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase, present in the cytosol fraction of term human placenta, were co-purified with a constant ratio of specific activities, approximating 100:1, respectively. The "pure enzyme" is a single band on sodium dodecyl sulfate disc gel electrophoresis. To evaluate whether catalysis of the estrogen and progestin substrates occurs at a single active site, alkylation studies using 16 alpha-bromoacetoxyprogesterone were designed. This affinity alkylating steroid binds at the enzyme-active site (km 256 microM; Vmax = 0.012 mumol/min/mg), inactivates the enzyme in an irreversible and time-dependent manner which follows pseudo-first order kinetics, and causes coincident loss of both the 17 beta- and 20 alpha-activities. Affinity radioalkylation studies using 16 alpha-[2'-3H]bromoacetoxyprogesterone indicate that 2 mol of steroid bind per mol of inactivated enzyme dimer (Mr = 68,000). Amino acid analyses of the acid hydrolysate of radioalkylated enzyme show that 16 alpha-bromoacetoxyprogesterone dicarboxymethylates a histidyl residue in the active site. These results are identical with those reported for 16 alpha-[2'-3H]bromoacetoxyestradiol 3-methyl ether inactivation and radioalkylation of identically purified "17 beta-estradiol dehydrogenase." Computer graphics were used to construct a model in which: 1) binding of estrogen and progestin substrates at one active site permits stereospecific catalysis; 2) the estrogen and progestin analogs' alkylating side arms have access to a common histidine residue. These observations clearly demonstrate that the catalysis of estrogen and progestin substrates can occur at a single active site of one enzyme.

Estradiol 17 beta-dehydrogenase and 20 alpha-hydroxysteroid dehydrogenase from human placental cytosol: one enzyme with two activities?

The soluble enzyme estradiol 17 beta-dehydrogenase (17 beta-ED) from human term placental cytosol is reported to be a stereospecific oxidoreductase for estrogen substrates. A published purification scheme (heat treatment and affinity chromatography) yielded a homogeneous protein which had the reported characteristics of pure 17 beta-ED and also had 20 alpha-hydroxysteroid dehydrogenase (20 alpha-HSD) activity. Spectrophotometric assay when the buffer contained albumin, 8 mg/ml, masked the 20 alpha-HSD activity observed in albumin-free conditions and may explain why this bifunctional activity has gone unrecognized. In human placenta, one enzyme may catalyze stereospecific oxidation/reduction of both estrogen and progesterone.

"Affinity" chromatography of steroid-transforming enzymes with a non-steroidal ligand.

The chromatographic behaviour of an avian oestradiol-17 beta dehydrogenase, the 3(17) beta-hydroxy steroid dehydrogenase from Pseudomonas testosteroni and cortisone reductase from Streptomyces dehydrogenans was studied on columns of p-(phenoxypropoxy)aniline attached to CNBr-activated Sepharose. The ligand was effective in adsorbing the oestradiol dehydrogenase from a partially purified extract of chicken liver, and the cortisone reductase was perferentially retained when mixtures of the three dehydrogenases were applied to columns in 10mM-buffer. Under these conditions the 3(17)beta-hydroxy steroid dehydrogenase was eluted in the front, but was adsorbed in the presence of 3 M-KCl. beta-N-Acetylglucosaminidase present in the liver preparation was not retained by the ligand, whereas lactate dehydrogenase from rabbit muscle was adsorbed in a manner similar to the retention pattern found on affinity chromatography with 2',5'-ADP--Sepharose. The mean overall purification of the oestradiol dehydrogenase was 13-fold, with a mean recovery of 53%. p-(Phenoxypropoxy)aniline offers promise for the purification of steroid-transforming enzymes where elution with substrate or cofactor is not wanted. It is also suggested that the ligand may be of service in the purification of receptors of hormonal steroids.