Rational proteomics V: structure-based mutagenesis has revealed key residues responsible for substrate recognition and catalysis by the dehydrogenase and isomerase activities in human 3beta-hydroxysteroid dehydrogenase/isomerase type 1

Sources and control of impurity during one-pot enzymatic production of dehydroepiandrosterone

Dehydroepiandrosterone (DHEA) has a promising market due to its capacity to regulate human hormone levels as well as preventing and treating various diseases. We have established a chemical esterification coupled biocatalytic-based scheme by lipase-catalyzed 4-androstene-3,17-dione (4-AD) hydrolysis to obtain the intermediate product 5-androstene-3,17-dione (5-AD), which was then asymmetrically reduced by a ketoreductase from Sphingomonas wittichii (SwiKR). Co-enzyme required for KR is regenerated by a glucose dehydrogenase (GDH) from Bacillus subtilis. This scheme is more environmentally friendly and more efficient than the current DHEA synthesis pathway. However, a significant amount of 4-AD as by-product was detected during the catalytic process. Focused on the control of by-products, we investigated the source of 4-AD and identified that it is mainly derived from the isomerization activity of SwiKR and GDH. Increasing the proportion of glucose in the catalytic system as well as optimizing the catalytic conditions drastically reduced 4-AD from 24.7 to 6.5% of total substrate amount, and the final yield of DHEA achieved 40.1 g/L. Furthermore, this is the first time that both SwiKR and GDH have been proved to be promiscuous enzymes with dehydrogenase and ketosteroid isomerase (KSI) activities, expanding knowledge of the substrate diversity of the short-chain dehydrogenase family enzymes. KEY POINTS: • A strategy of coupling lipase, ketoreductase, and glucose dehydrogenase in producing DHEA from 4-AD • Both SwiKR and GDH are identified with ketosteroid isomerase activity. • Development of catalytic strategy to control by-product and achieve highly selective DHEA production.

Human placental 3β-hydroxysteroid dehydrogenase/steroid Δ5,4-isomerase 1: Identity, regulation and environmental inhibitors

Human placental 3β-hydroxysteroid dehydrogenase/steroid Δ5, 4-isomerase 1 (HSD3B1), a high-affinity type I enzyme, uses pregnenolone to make progesterone, which is critical for maintenance of pregnancy. HSD3B1 is located in the mitochondrion and the smooth endoplasmic reticulum of placental cells and is encoded by HSD3B1 gene. HSD3B1 contains GATA and TEF-5 regulatory elements. Many endocrine disruptors, including phthalates, methoxychlor and its metabolite, organotins, and gossypol directly inhibit placental HSD3B1 thus blocking progesterone production. In this review, we discuss the placental HSD3B1, its gene regulation, biochemistry, subcellular location, and inhibitors from the environment.

Structure-function relationships for the selective inhibition of human 3β-hydroxysteroid dehydrogenase type 1 by a novel androgen analog

3β-Hydroxysteroid dehydrogenase type 1 (3β-HSD1) is selectively expressed in human placenta, mammary glands and breast tumors in women. Human 3β-HSD2 is selectively expressed in adrenal glands and ovaries. Based on AutoDock 3 and 4 results, we have exploited key differences in the amino acid sequences of 3β-HSD1 (Ser194, Arg195) and 3β-HSD2 (Gly194, Pro195) by designing a selective inhibitor of 3β-HSD1. 2,16-Dicyano-4,5-epoxy-androstane-3,17-dione (16-cyano-17-keto-trilostane or DiCN-AND) was synthesized in a 4-step procedure from androstenedione. In purified 3β-HSD inhibition studies, DiCN-AND competitively inhibited 3β- HSD1 with K_i=4.7μM and noncompetitively inhibited 3β-HSD2 with a 6.5-fold higher K_i=30.7μM. We previously reported similar isoenzyme-specific inhibition profiles for trilostane. Based on our docking results, we created, expressed and purified the chimeric S194G-1 mutant of 3β-HSD1. Trilostane inhibited S194G-1 (K_i=0.67μM) with a noncompetitive mode compared to its 6.7-fold higher affinity, competitive inhibition of 3β-HSD1 (K_i=0.10μM). DiCN-AND inhibited S194G-1 with a 6.3-fold higher K_i (29.5μM) than measured for 3β-HSD1 (K_i=4.7μM) but with the same competitive mode for both enzyme species. Since DiCN-AND noncompetitively inhibits 3β-HSD2, which has the Gly194 and Pro195 of 3β-HSD2 in place of the Ser194 and Arg195 in 3β-HSD1, this suggests that Arg195 alone in 3β-HSD1 or S194G-1 is required to bind DiCN-AND in the substrate binding site (competitive inhibition). However, both Ser194 and Arg195 are required to bind trilostane in the 3β-HSD1 substrate site based on its noncompetitive inhibition of S194G-1 and 3β-HSD2. In support of this hypothesis, DiCN-AND inhibited our chimeric R195P-1 mutant noncompetitively with a K_i=41.3μM (similar to the 3β-HSD2 inhibition profile). Since DiCN-AND competitively inhibited S194G-1 that still contains R195 but noncompetitively inhibited R195P-1 that still contains S194, our data provides strong evidence that the Arg195 being mutated to Pro195 (as present in 3β-HSD2) shifts the inhibition mode from competitive to noncompetitive in 3β-HSD1. This supports the key role of Arg195 in 3β-HSD1 for the high affinity, competitive binding of the trilostane analogs. Our new structure/function information for the design of targeted 3β-HSD1 inhibitors may lead to important new treatments for the prevention of spontaneous premature birth.

Endoplasmic Reticulum Stress Enhances Mitochondrial Metabolic Activity in Mammalian Adrenals and Gonads

The acute response to stress consists of a series of physiological programs to promote survival by generating glucocorticoids and activating stress response genes that increase the synthesis of many chaperone proteins specific to individual organelles. In the endoplasmic reticulum (ER), short-term stress triggers activation of the unfolded protein response (UPR) module that either leads to neutralization of the initial stress or adaptation to it; chronic stress favors cell death. UPR induces expression of the transcription factor, C/EBP homology protein (CHOP), and its deletion protects against the lethal consequences of prolonged UPR. Here, we show that stress-induced CHOP expression coincides with increased metabolic activity. During stress, the ER and mitochondria come close to each other, resulting in the formation of a complex consisting of the mitochondrial translocase, translocase of outer mitochondrial membrane 22 (Tom22), steroidogenic acute regulatory protein (StAR), and 3β-hydroxysteroid dehydrogenase type 2 (3βHSD2) via its intermembrane space (IMS)-exposed charged unstructured loop region. Stress increased the circulation of phosphates, which elevated pregnenolone synthesis by 2-fold by increasing the stability of 3βHSD2 and its association with the mitochondrion-associated ER membrane (MAM) and mitochondrial proteins. In summary, cytoplasmic CHOP plays a central role in coordinating the interaction of MAM proteins with the outer mitochondrial membrane translocase, Tom22, to activate metabolic activity in the IMS by enhanced phosphate circulation.

Regulation of human 3-beta-hydroxysteroid dehydrogenase type-2 (3βHSD2) by molecular chaperones and the mitochondrial environment affects steroidogenesis

Human 3-β-hydroxysteroid dehydrogenase/isomerase types 1 and 2 (3βHSD1 and 3βHSD2, respectively) are expressed in a tissue-specific pattern by different genes. Site-directed mutagenesis studies have confirmed the function of the catalytic amino acids (Tyr154, Lys 158, Ser124 in both isoenzymes), substrate/inhibitor isoform-specific residues (His156 and Arg195 in 3βHSD1) and cofactor binding residues (Asp36 provides NAD(+) specificity in both isoenzymes). However, detailed analysis of isoform-specific organelle localization and characterization is difficult due to the 93% amino acid identity between the two isoforms. With recent advances in the knowledge of mitochondrial architecture and localization of the various translocases, our laboratory has studied the mechanisms regulating mitochondrial 3βHSD2 localization. The mitochondrial N-terminal leader sequence of 3βHSD2 directs its entry into the mitochondria where it is localized to the intermembrane space (IMS). Unlike other mitochondrial proteins, the N-terminal signal sequence of 3βHSD2 is not cleaved upon mitochondrial import. 3βHSD2 interacts with the mitochondrial translocase, Tim50, to regulate progesterone and androstenedione formation. Our studies suggest that its activity at the IMS is facilitated in a partially unfolded "molten globule" conformation by the proton pump between the matrix and IMS. The unfolded protein is refolded by the mitochondrial chaperones. The protons at the IMS are absorbed by the lipid vesicles, to maintain the proton pump and recycle 3βHSD2. As a result, one molecule of 3βHSD2 may participate in multiple catalytic reactions. In summary, the steroidogenic cell recycles 3βHSD2 to catalyze the reactions needed to produce androstenedione, progesterone and 17α-hydroxyprogesterone on demand in coordination with the mitochondrial translocase, Tim50. This article is part of a Special Issue entitled 'Steroid/Sterol signaling'.

Lipid-mediated unfolding of 3β-hydroxysteroid dehydrogenase 2 is essential for steroidogenic activity

For inner mitochondrial membrane (IMM) proteins that do not undergo N-terminal cleavage, the activity may occur in the absence of a receptor present in the mitochondrial membrane. One such protein is human 3β-hydroxysteroid dehydrogenase 2 (3βHSD2), the IMM resident protein responsible for catalyzing two key steps in steroid metabolism: the conversion of pregnenolone to progesterone and dehydroepiandrosterone to androstenedione. Conversion requires that 3βHSD2 serve as both a dehydrogenase and an isomerase. The dual functionality of 3βHSD2 results from a conformational change, but the trigger for this change remains unknown. Using fluorescence resonance energy transfer, we found that 3βHSD2 interacted strongly with a mixture of dipalmitoylphosphatidylglycerol (DPPG) and dipalmitoylphosphatidylcholine (DPPC). 3βHSD2 became less stable when incubated with the individual lipids, as indicated by the decrease in thermal denaturation (T(m)) from 42 to 37 °C. DPPG, alone or in combination with DPPC, led to a decrease in α-helical content without an effect on the β-sheet conformation. With the exception of the 20 N-terminal amino acids, mixed vesicles protected 3βHSD2 from trypsin digestion. However, protein incubated with DPPC was only partially protected. The lipid-mediated unfolding completely supports the model in which a cavity forms between the α-helix and β-sheet. As 3βHSD2 lacks a receptor, opening the conformation may activate the protein.

Selective inhibition of human 3β-hydroxysteroid dehydrogenase type 1 as a potential treatment for breast cancer

Human 3β-hydroxysteroid dehydrogenase/isomerase type 1 (3β-HSD1) is a critical enzyme in the conversion of DHEA to estradiol in breast tumors and may be a target enzyme for inhibition in the treatment of breast cancer in postmenopausal women. Human 3β-HSD2 participates in the production of cortisol and aldosterone in the human adrenal gland in this population. In our recombinant human breast tumor MCF-7 Tet-off cells that express either 3β-HSD1 or 3β-HSD2, trilostane and epostane inhibit the DHEA-induced proliferation of MCF-7 3β-HSD1 cells with 12-16-fold lower IC(50) values compared to the MCF-7 3β-HSD2 cells. Trilostane and epostane also competitively inhibit purified human 3β-HSD1 with 12-16-fold lower K(i) values compared to the noncompetitive K(i) values measured for human 3β-HSD2. Using our structural model of 3β-HSD1, trilostane was docked in the active site of 3β-HSD1, and Arg195 in 3β-HSD1 or Pro195 in 3β-HSD2 was identified as a potentially critical residue. The R195P-1 mutant of 3β-HSD1 and the P195R-2 mutant of 3β-HSD2 were created, expressed and purified. Kinetic analyses of enzyme inhibition suggest that the high-affinity, competitive inhibition of 3β-HSD1 by trilostane may be related to the presence of Arg195 in 3β-HSD1 versus Pro195 in 3β-HSD2. In addition, His156 in 3β-HSD1 may play a role in the higher affinity of 3β-HSD1 for substrates and inhibitors compared to 3β-HSD2 containing Try156. Structural modeling of the 3β-HSD1 dimer identified a possible interaction between His156 on one subunit and Gln105 on the other. Kinetic analyses of the H156Y-1, Q105M-1 and Q105M-2 support subunit interactions that contribute to the higher affinity of 3β-HSD1 for the inhibitor, epostane, compared to 3β-HSD2. Article from the Special issue on Targeted Inhibitors.

Biotransformation of dehydroepiandrosterone (DHEA) with Penicillium griseopurpureum Smith and Penicillium glabrum (Wehmer) Westling

Microbial transformation of dehydroepiandrosterone (DHEA, 1) using Penicillium griseopurpureum Smith and Penicillium glabrum (Wehmer) Westling has been investigated. Neither fungi had been examined previously for steroid biotransformation. One novel metabolic product of DHEA (1) transformed with P. griseopurpureum Smith, 15α-hydroxy-17a-oxa-d-homo-androst-4-ene-3,17-dione (5), was reported for the first time. The steroid products were assigned by interpretation of their spectral data such as (1)H NMR, (13)C NMR, IR, and HR-MS spectroscopy. P. griseopurpureum Smith was proven to be remarkably efficient in oxidation of the DHEA (1) into androst-4-en-3,17-dione (2). The strain was also observed to yield different monooxygenases to introduce hydroxyl groups at C-7α, -14α, and -15α positions of steroids. Preference for Baeyer-Villiger oxidation to lactonize D ring and oxidation of the 3β-alcohol to the 3-ketone were observed in both incubations. The strain of P. glabrum (Wehmer) Westling catalyzed the steroid 1 to generate both testololactone 3, and d-lactone product with 3β-hydroxy-5-en moiety 8. In addition, the strain promoted hydrogenation of the C-5 and C-6 positions, leading to the formation of 3β-hydroxy-17a-oxa-d-homo-5α-androstan-3,17-dione (9). The biotransformation pathways of DHEA (1) with P. glabrum (Wehmer) Westling and P. griseopurpureum Smith have been investigated, respectively. Possible metabolic pathways of DHEA (1) were proposed.

The functions of key residues in the inhibitor, substrate and cofactor sites of human 3beta-hydroxysteroid dehydrogenase type 1 are validated by mutagenesis

In postmenopausal women, human 3beta-hydroxysteroid dehydrogenase type 1 (3beta-HSD1) is a critical enzyme in the conversion of DHEA to estradiol in breast tumors, while 3beta-HSD2 participates in the production of cortisol and aldosterone in the human adrenal gland. The goals of this project are to determine if Arg195 in 3beta-HSD1 vs. Pro195 in 3beta-HSD2 in the substrate/inhibitor binding site is a critical structural difference responsible for the higher affinity of 3beta-HSD1 for inhibitor and substrate steroids compared to 3beta-HSD2 and whether Asp61, Glu192 and Thr8 are fingerprint residues for cofactor and substrate binding using site-directed mutagenesis. The R195P-1 mutant of 3beta-HSD1 and the P195R-2 mutant of 3beta-HSD2 have been created, expressed, purified and characterized kinetically. Dixon analyses of the inhibition of the R195P-1 mutant, P195R-2 mutant, wild-type 3beta-HSD1 and wild-type 3beta-HSD2 by trilostane has produced kinetic profiles that show inhibition of 3beta-HSD1 by trilostane (K(i)=0.10microM, competitive) with a 16-fold lower K(i) and different mode than measured for 3beta-HSD2 (K(i)=1.60microM, noncompetitive). The R195P-1 mutation shifts the high-affinity, competitive inhibition profile of 3beta-HSD1 to a low-affinity (trilostane K(i)=2.56microM), noncompetitive inhibition profile similar to that of 3beta-HSD2 containing Pro195. The P195R-2 mutation shifts the low-affinity, noncompetitive inhibition profile of 3beta-HSD2 to a high-affinity (trilostane K(i)=0.19microM), competitive inhibition profile similar to that of 3beta-HSD1 containing Arg195. Michaelis-Menten kinetics for DHEA, 16beta-hydroxy-DHEA and 16alpha-hydroxy-DHEA substrate utilization by the R195P-1 and P195R-2 enzymes provide further validation for higher affinity binding due to Arg195 in 3beta-HSD1. Comparisons of the Michaelis-Menten values of cofactor and substrate for the targeted mutants of 3beta-HSD1 (D61N, D61V, E192A, T8A) clarify the functions of these residues as well.

Studies on Baeyer-Villiger oxidation of steroids: DHEA and pregnenolone D-lactonization pathways in Penicillium camemberti AM83

Penicillium camemberti AM83 strain is able to carry out effective Baeyer-Villiger type oxidation of DHEA, pregnenolone, androstenedione and progesterone to testololactone. Pregnenolone and DHEA underwent oxidation to testololactone via two routes: through 4-en-3-ketones (progesterone and/or androstenedione respectively) or through 3beta-hydroxy-17a-oxa-d-homo-androst-5-en-17-one. Analysis of transformation progress of studied substrates as function of time indicates that the 17beta-side chain cleavage and oxidation of 17-ketones to d-lactones are catalyzed by two different, substrate-induced, BVMOs. In the presence of a C-21 substrate (pregnenolone or progesterone) induction of the enzyme catalyzing cleavage at 17beta-acetyl chain was observed, whereas DHEA and androstenedione induced activity of the BVMO responsible for the ring-D oxidation; 5-en-3beta-alcohol was a more effective inducer that the respective 4-en-3-ketone.

Structural basis for the selective inhibition of human 3beta-hydroxysteroid dehydrogenase 1 in human breast tumor MCF-7 cells

Human 3beta-hydroxysteroid dehydrogenase/isomerase type 1 (3beta-HSD1) is a critical enzyme in the conversion of DHEA to estradiol in breast tumors and may be a target enzyme for inhibition in the treatment of breast cancer in postmenopausal women. Human 3beta-HSD2 participates in the production of cortisol and aldosterone in the human adrenal gland in this population. In our recombinant human breast tumor MCF-7 Tet-off cells that express either 3beta-HSD1 or 3beta-HSD2, trilostane and epostane inhibit the DHEA-induced proliferation of MCF-7 3beta-HSD1 cells with 12- to 16-fold lower IC(50) values compared to the MCF-7 3beta-HSD2 cells. The compounds also competitively inhibit purified human 3beta-HSD1 with 12- to 16-fold lower K(i) values compared to the noncompetitive K(i) values measured for human 3beta-HSD2. Using our structural model of 3beta-HSD1, trilostane or 17beta-acetoxy-trilostane was docked in the active site of 3beta-HSD1, and Arg195 in 3beta-HSD1 or Pro195 in 3beta-HSD2 was identified as a potentially critical residue (one of 23 non-identical residues in the two isoenzymes). The P195R mutant of 3beta-HSD2 were created, expressed and purified. Kinetic analyses of enzyme inhibition suggest that the high affinity, competitive inhibition of 3beta-HSD1 by trilostane and epostane may be related to the presence of Arg195 in 3beta-HSD1 vs. Pro195 in 3beta-HSD2.

Transformation of 5-ene steroids by the fungus Aspergillus tamarii KITA: mixed molecular fate in lactonization and hydroxylation pathways with identification of a putative 3beta-hydroxy-steroid dehydrogenase/Delta5-Delta4 isomerase pathway

The fungus Aspergillus tamarii metabolizes progesterone to testololactone in high yield through a sequential four step enzymatic pathway which, has demonstrated flexibility in handling a range of steroidal probes. These substrates have revealed that subtle changes in the molecular structure of the steroid lead to significant changes in route of metabolism. It was therefore of interest to determine the metabolism of a range of 5-ene containing steroidal substrates. Remarkably the primary route of 5-ene steroid metabolism involved a 3beta-hydroxy-steroid dehydrogenase/Delta(5)-Delta(4) isomerase (3beta-HSD/isomerase) enzyme(s), generating 3-one-4-ene functionality and identified for the first time in a fungus with the ability to handle both dehydroepiansdrosterone (DHEA) as well as C-17 side-chain containing compounds such as pregnenolone and 3beta-hydroxy-16alpha,17alpha-epoxypregn-5-en-20-one. Uniquely in all the steroids tested, 3beta-HSD/isomerase activity only occurred following lactonization of the steroidal ring-D. Presence of C-7 allylic hydroxylation, in either epimeric form, inhibited 3beta-HSD/isomerase activity and of the substrates tested, was only observed with DHEA and its 13alpha-methyl analogue. In contrast to previous studies of fungi with 3beta-HSD/isomerase activity DHEA could also enter a minor hydroxylation pathway. Pregnenolone and 3beta-hydroxy-16alpha,17alpha-epoxypregn-5-en-20-one were metabolized solely through the putative 3beta-HSD/isomerase pathway, indicating that a 17beta-methyl ketone functionality inhibits allylic oxidation at C-7. The presence of the 3beta-HSD/isomerase in A. tamarii and the transformation results obtained in this study highlight an important potential role that fungi may have in the generation of environmental androgens.

Structure/function of the inhibition of human 3beta-hydroxysteroid dehydrogenase type 1 and type 2 by trilostane

The human type 1 (placenta, breast tumors) and type 2 (gonads, adrenals) isoforms of 3beta-hydroxysteroid dehydrogenase/isomerase (3beta-HSD) are key enzymes in biosynthesis of all active steroid hormones. Human 3beta-HSD1 is a critical enzyme in the conversion of DHEA to estradiol in breast tumors and may be a major target enzyme for the treatment of breast cancer. 3beta-HSD2 participates in the production of cortisol and aldosterone in the human adrenal gland. The goals of this project are to evaluate the role of the 2alpha-cyano group on trilostane (2alpha-cyano-4alpha,5alpha-epoxy-17beta-ol-androstane-3-one) and determine which amino acids may be critical for 3beta-HSD1 specificity. Trilostane without the 2alpha-cyano group, 4alpha,5alpha-epoxy-testosterone, was synthesized. Using our structural model of 3beta-HSD1, trilostane or 4alpha,5alpha-epoxy-testosterone was docked in the active site using Autodock 3.0, and the potentially critical residues (Met187 and Ser124) were identified. The M187T and S124T mutants of 3beta-HSD1 were created, expressed and purified. Dixon analyses of the inhibition of wild-type 3beta-HSD1, 3beta-HSD2, M187T and S124T by trilostane and 4alpha,5alpha-epoxy-testosterone suggest that the 2alpha-cyano group of trilostane is anchored by Ser124 in both isoenzymes. Kinetic analyses of cofactor and substrate utilization as well as the inhibition kinetics of M187T and the wild-type enzymes suggest that the 16-fold higher-affinity inhibition of 3beta-HSD1 by trilostane may be related to the presence of Met187 in 3beta-HSD1 and Thr187 in 3beta-HSD2. This structure/function information may lead to the production of more highly specific inhibitors of 3beta-HSD1 to block the hormone-dependent growth of breast tumors.

Structure/function of human type 1 3beta-hydroxysteroid dehydrogenase: An intrasubunit disulfide bond in the Rossmann-fold domain and a Cys residue in the active site are critical for substrate and coenzyme utilization

The human type 1 (placenta, breast tumors) and type 2 (gonads, adrenals) isoforms of 3beta-hydroxysteroid dehydrogenase/isomerase (3beta-HSD) are key enzymes in steroidogenic pathways leading to the production of all active steroid hormones. Kinetic analyses of purified 3beta-HSD1 show that the Michaelis-Menten constants (Km) for substrates and cofactor are decreased dramatically (three- to eight-fold) by the addition of beta-mercaptoethanol (BME), which suggest that a disulfide bond may be critical to ligand utilization. Western immunoblots and SDS-PAGE of purified 3beta-HSD1 in the presence or absence of BME showed a lack of intersubunit disulfide bonds in the dimeric enzyme. The Rossmann-fold domain of 3beta-HSD1 contains two Cys residues, Cys72 and Cys111, which are capable of forming an intrasubunit disulfide bond based on their proximity in our structural model. Our structural model also predicts that Cys83 may affect the orientation of substrate and cofactor. To test these predictions, the C72S, C72F, C111S, C111A, C83S and C83A mutants of 3beta-HSD1 were produced, expressed, and purified. BME failed to diminish the Km values of substrate and cofactor for C72S, C72F, C111S and C111A but produced a 2.5 decrease in Km values for C83A ligands similar to wild-type 3beta-HSD. Thus, our results support the presence of an intrasubunit disulfide bond between Cys72 and Cys111 that participates in the tertiary structure of the Rossmann-fold domain. Although C83S had no enzyme activity, the C83A mutant enzyme exhibited two- to five-fold higher Km values for substrate and cofactor but had similar K(cat) values compared to wild-type 3beta-HSD. These data characterize the roles of Cys residues in 3beta-HSD and validate the predictions of our structural model.