The engineered, cytosolic form of human type I 3beta-hydroxysteroid dehydrogenase/isomerase: purification, characterization and crystallization.

Human type I 3beta-hydroxysteroid dehydrogenase/isomerase (3beta-HSD/isomerase) is an integral membrane protein of human placental trophoblast and of insect Sf9 cells transfected with recombinant baculovirus containing the cDNA encoding the enzyme. Purified native or wild-type enzyme remains in solution only in the presence of detergent that may prevent crystallization. The membrane-spanning domain (residues 283-310) of the enzyme protein was deleted in the cDNA using PCR-based mutagenesis. The modified enzyme was expressed by baculovirus in the cytosol instead of in the microsomes and mitochondria of the Sf9 cells. The cytosolic form of 3beta-HSD/isomerase was purified using affinity chromatography with Cibacron Blue 1000. The NAD(+) and NaCl used to elute the enzyme were removed by size-exclusion centrifugation. Hydroxylapatite chromatography yielded a 26-fold purification of the enzyme. SDS-PAGE revealed a single protein band for the purified cytosolic enzyme (monomeric molecular mass 38.8 kDa) that migrated just below the wild-type enzyme (monomeric molecular mass 42.0 kDa). Michaelis-Menten constants measured for 3beta-HSD substrate (dehydroepiandrosterone) utilization by the purified cytosolic enzyme (K(m)=4.5 microM, V(max)=53 nmol/min per mg) and the pure wild-type enzyme (K(m)=3.7 microM, V(max)=43 nmol/min per mg), for isomerase substrate (5-androstene-3,17-dione) conversion by the purified cytosolic (K(m)=25 microM, V(max)=576 nmol/min per mg) and wild-type (K(m)=28 microM, V(max)=598 nmol/min per mg) enzymes, and for NAD(+) reduction by the 3beta-HSD activities of the cytosolic (K(m)=35 microM, V(max)=51 nmol/min per mg) and wild-type (K(m)=34 microM, V(max)=46 nmol/min per mg) enzymes are nearly identical. The isomerase activity of the cytosolic enzyme requires allosteric activation by NADH (K(m)=4.6 microM, V(max)=538 nmol/min per mg) just like the wild-type enzyme (K(m)=4.6 microM, V(max)=536 nmol/min per mg). Crystals of the purified, cytosolic enzyme protein have been obtained. The inability to crystallize the detergent-solubilized, wild-type microsomal enzyme has been overcome by engineering a cytosolic form of this protein. Determining the tertiary structure of 3beta-HSD/isomerase will clarify the mechanistic roles of potentially critical amino acids (His(261), Tyr(253)) that have been identified in the primary structure.

Thyroid hormone regulation of perch ovarian 3beta-hydroxysteroid dehydrogenase/delta5-delta4-isomerase activity: involvement of a 52-kDa protein.

Ovarian follicles were collected from perch belonging to the prespawning (vitellogenic) stage and incubated in vitro for 5 h in the absence (control) and presence of 3, 5, 3'-triiodothyronine (T3). Addition of increasing concentrations of T3 from 12.5 to 100 ng/ml caused a linear increase of 3beta-hydroxysteroid dehydrogenase/delta5-delta4-isomerase (3 beta-HSD) activity to 50 ng and then it leveled off indicating a saturation of enzyme activity with 50 ng T3. T3 stimulation of 3beta-HSD activity could be blocked by cycloheximide indicating the involvement of T3-induced protein (TIP) isolated and purified earlier from this laboratory. Addition of fish TIP purified from perch ovarian follicle (fTIP) or rat granulosa cell TIP to ovarian follicular incubation at a dose of 5 microg/ml significantly increased (P < 0.01) 3beta-HSD activity. To observe whether TIP acts directly on the enzyme or not, 3beta-HSD from perch ovarian follicle was purified to homogeneity by the following steps: (i) Sephadex G 75 gel filtration, (ii) DEAE-Sephacel chromatography, and (iii) NAD-affinity column chromatography. Purified 3beta-HSD gave a clear single band on an SDS gel and its molecular weight is 45 kDa. Addition of fTIP to an assay mixture containing purified 3beta-HSD resulted in a fourfold increase of the enzyme activity. fTIP alone did not show enzyme activity when incubated with the radiolabeled substrate. Addition of T3 (50 ng) to the 3beta-HSD assay mixture had no effect on the enzyme activity. Determination of Vmax and Km of the purified enzyme in the absence (control) and presence of fTIP demonstrated a considerable increase of 3beta-HSD affinity and rate of enzyme reaction.

Site-directed mutagenesis identifies amino acid residues associated with the dehydrogenase and isomerase activities of human type I (placental) 3beta-hydroxysteroid dehydrogenase/isomerase.

3beta-hydroxysteroid dehydrogenase/steroid delta5-->4-isomerase (3beta-HSD/isomerase) was expressed by baculovirus in Spodoptera fungiperda (Sf9) insect cells from cDNA sequences encoding human wild-type I (placental) and the human type I mutants - H261R, Y253F and Y253,254F. Western blots of SDS-polyacrylamide gels showed that the baculovirus-infected Sf9 cells expressed the immunoreactive wild-type, H261R, Y253F or Y253,254F protein that co-migrated with purified placental 3beta-HSD/isomerase (monomeric Mr=42,000 Da). The wild-type, H261R and Y253F enzymes were each purified as a single, homogeneous protein from a suspension of the Sf9 cells (5.01). In kinetic studies with purified enzyme, the H261R mutant enzyme had no 3beta-HSD activity, whereas the Km and Vmax values of the isomerase substrate were similar to the values obtained with the wild-type and native enzymes. The Vmax (88 nmol/min/mg) for the conversion of 5-androstene-3,17-dione to androstenedione by the Y253F isomerase activity was 7.0-fold less than the mean Vmax (620 nmol/min/mg) measured for the isomerase activity of the wild-type and native placental enzymes. In microsomal preparations, isomerase activity was completely abolished in the Y253,254F mutant enzyme, but Y253,254F had 45% of the 3beta-HSD activity of the wild-type enzyme. In contrast, the purified Y253F, wild-type and native enzymes had similar Vmax values for substrate oxidation by the 3beta-HSD activity. The 3beta-HSD activities of the Y253F, Y253,254F and wild-type enzymes reduced NAD+ with similar kinetic values. Although NADH activated the isomerase activities of the H261R and wild-type enzymes with similar kinetics, the activation of the isomerase activity of H261R by NAD+ was dramatically decreased. Based on these kinetic measurements, His261 appears to be a critical amino acid residue for the 3beta-HSD activity, and Tyr253 or Tyr254 participates in the isomerase activity of human type I (placental) enzyme.

Alternate splicing of the RNA for hamster type 2 3 beta-hydroxysteroid dehydrogenase/delta 5-->4isomerase.

Complementary DNAs encoding the hamster type 2 3 beta-hydroxysteroid dehydrogenase/delta 5-->4 isomerase were isolated from liver and kidney cDNA libraries. Nine clones were isolated containing identical coding and 3' untranslated sequences. However, six of the clones contained a 68-nucleotide stretch in the 5' untranslated region that was missing in the other three clones. Primers were designed to flank this region and the polymerase chain reaction (PCR) was performed on hamster liver and adrenal RNA. Two PCR products were amplified of the predicted molecular sizes and with the expected sequence. Primers were then designed to amplify sequences encompassing this region from hamster genomic DNA. Sequencing of the resultant PCR products demonstrated that the 68-nucleotide stretch missing in some transcripts corresponded exactly to the second of three exons identified. We conclude that the 5' untranslated region of this mRNA is transcribed from at least three exons, and that the sequence of the second of these exons is spliced out of some of the RNA transcripts.

Isolation of a new mouse 3beta-hydroxysteroid dehydrogenase isoform, 3beta-HSD VI, expressed during early pregnancy.

The enzyme 3beta-hydroxysteroid dehydrogenase (3beta-HSD) is a key enzyme in the biosynthesis of steroid hormones. To date, this laboratory has isolated and characterized five distinct 3beta-HSD complementary DNAs (cDNAs) in the mouse (3beta-HSD I through V). These different forms are expressed in a tissue- and developmentally-specific manner and fall into two functionally distinct enzymes. 3beta-HSD I and III, and most likely II, function as dehydrogenase/isomerases, whereas 3beta-HSD IV and V function as 3-ketosteroid reductases. This study describes the isolation, characterization, and tissue-specific expression of a sixth member of this gene family, 3beta-HSD VI. This new isoform functions as an NAD+-dependent dehydrogenase/isomerase exhibiting very low Michaelis-Menten constant (Km) values for pregnenolone (approximately 0.035 microM) and dehydroepiandrosterone (approximately 0.12 microM). 3beta-HSD VI is the earliest isoform to be expressed during embryogenesis in cells of embryonic origin at 7 and 9.5 days postcoitum (pc), and is the major isoform expressed in uterine tissue at the time of implantation (4.5 days pc) and continues to be expressed in uterine tissue at 6.5, 7.5, and 9.5 days pc. 3beta-HSD VI is expressed in giant trophoblasts at 9.5 days pc and is expressed in the placenta through day 15.5 pc. In the adult mouse, 3beta-HSD VI appears to be the only isoform expressed in the skin and also is expressed in the testis, but to a lesser extent than 3beta-HSD I. Mouse 3beta-HSD VI cDNA is orthologous to human 3beta-HSD I cDNA. Human type I 3beta-HSD has been shown to be the only isoform expressed in the placenta and skin. The demonstration that mouse 3beta-HSD VI functions as a dehydrogenase/isomerase and is the predominant isoform expressed during the first half of pregnancy in uterine tissue and in embryonic cells suggests that this isoform may be involved in local production of progesterone, which is needed for successful implantation of the blastocyst and/or maintenance of early pregnancy.

Incorporation of bovine adrenal 3 beta-hydroxy-5-ene steroid dehydrogenase/delta 5-delta 4 isomerase into phospholipid vesicles.

3 beta-Hydroxysteroid dehydrogenase/delta 5-delta 4 isomerase (3 beta-HSD/I) and cytochrome P-450C21 were co-purified from bovine adrenocortical microsomes by an improved method. The 3 beta-HSD/I was successfully incorporated into liposomal membranes in which the enzyme activity was greatly stabilized. Enzymatic activities and kinetic parameters of the 3 beta-HSD/I proteoliposomes were almost the same as those of the solubilized form.

Inhibitory effect of synthetic progestins, 4-MA and cyanoketone on human placental 3 beta-hydroxysteroid dehydrogenase/5----4-ene-isomerase activity.

Human placental 3 beta-hydroxysteroid dehydrogenase/5----4-ene isomerase (3 beta-HSD) purified from human placenta transforms C-21 (pregnenolone and 17 alpha-hydroxy pregnenolone) as well as C-19 (dehydroepiandrosterone and androst-5-ene-3 beta, 17 beta-diol) steroids into the corresponding 3-keto-4-ene-steroids and is thus involved in the biosynthesis of all classes of hormonal steroids. Trilostane, epostane and cyanoketone are potent inhibitors of 3 beta-HSD with Ki values of approximately 50 nM. 4-MA, a well known 5 alpha-reductase inhibitor, is also a potent inhibitor of 3 beta-HSD with a Ki value of 56 nM. Synthetic progestin compounds such as promegestone and RU2323 show relatively strong inhibitory effects with Ki values of 110 and 190 nM, respectively. Cyproterone acetate, a progestin used in the treatment of hirsutism, acne and prostate cancer as well as norgestrel and norethindrone that are widely used as oral contraceptives also inhibit 3 beta-HSD activity at Ki values of 1.5, 1.7 and 2.5 microM, respectively.

Partial characterization of the microsomal and solubilized hypothalamic progesterone 5 alpha-reductase.

Microsomal progesterone 5 alpha-reductase activity from female rat hypothalamus has been solubilized and partially characterized in terms of kinetic and physical properties. The solubilization of progesterone 5 alpha-reductase has been accomplished through the use of a digitonin/KCL-extraction. Both the microsomal and solubilized enzyme activities exhibit similar kinetic and physical characteristics. These include their apparent Km for progesterone (Microsomal Km = 113 +/- 11 nM; solubilized Km = 144 +/- 20 nM) and their affinity (approximately 7 nM) for the 5 alpha-dihydroprogesterone analog, 4-aza-4-methyl-5 alpha-pregnane-3,20-dione, which is a potent inhibitor of progesterone 5 alpha-reduction. Both activities are inhibited by divalent cations (Zn2+ and Cu2+) and the sulfhydryl-blocking agent p-chloromercuribenzoic acid. Studies aimed at optimizing isolation and assay conditions for the hypothalamic progesterone 5 alpha-reductase indicate that the microsomal activity is enhanced in the presence of monovalent cations (particularly K+ and Li+) and the metal chelator EDTA, but is unaffected by the sulfhydryl reducing agent dithiothreitol. The activity of the solubilized enzyme is also enhanced by EDTA but slightly stimulated by dithiothreitol. Analysis of hypothalamic progesterone 5 alpha-reduction for possible flavin involvement (as a hydride carrier between NADPH and the steroid) indicates that the enzyme activity is decreased by high levels of flavins, flavin analogs and riboflavin deficiency.

Pregnenolone-binding components in the porcine adrenal gland.

The object of this study was to determine whether binding components for pregnenolone, analogous to those described in the adrenal cortex of guinea pigs and rats, were present in the porcine adrenal. A binding component for pregnenolone in the cytosolic fraction of porcine adrenal was demonstrated by sucrose density gradient centrifugation. It banded maximally at 9.6% sucrose (w/w) compared to 12.2% and 12.4% sucrose (w/w) for the plasma-binding component and serum albumin, respectively. At a pregnenolone concentration of 1 X 10(-5) M, specific cytosolic binding of 1 X 10(-8) M [3H]pregnenolone was decreased by 42%. The fractions from sucrose gradients which bound pregnenolone maximally contained 3 beta-hydroxysteroid dehydrogenase-isomerase. The cytosolic supernatant of porcine adrenal gland was resolved by chromatography on hydroxyapatite into eleven fractions, four of which bound added pregnenolone and three of which displayed enzymatic activity. Electrophoretic analysis of the enzymatically active fractions in polyacrylamide gel showed that two of them were of heterogeneous composition, whereas the third, most enzymatically active, fraction consisted principally of one band of high molecular weight.