molecular mechanisms of estrogen recognition and 17-keto reduction by human 17β-hydroxysteroid...

Chemico-Biological Interactions 130–132 (2001) 637–650

Molecular mechanisms of estrogen recognitionand 17-keto reduction by human

17b-hydroxysteroid dehydrogenase 1

Debashis Ghosh a,*, Pirkko Vihko b

a Roswell Park Cancer Institute, Molecular and Cellular Biophysics, Elm and Carlton Streets, Buffalo,NY 14263, USA

b Biocenter Oulu and WHO Collaborating Centre for Research on Reproducti6e Health,Uni6ersity of Oulu, P.O. Box 5000, FIN-90401 Oulu, Finland

Abstract

The reduction of inactive estrone (E1) to the active estrogen 17b-estradiol (E2) is catalyzedby type 1 17b-hydroxysteroid dehydrogenase (17HSD1). Crystallographic studies, modelingand activity measurement of mutants and chimeric enzymes have led to the understanding ofits mechanism of action and the molecular basis for the estrogenic specificity. An elec-trophilic attack on the C17-keto oxygen by the Tyr 155 hydroxyl is proposed for initiationof the transition state. The active site is a hydrophobic pocket with catalytic residues at oneend and the recognition machinery on the other. Tyr 155, Lys 159 and Ser 142 are essentialfor the activity. The presence of certain other amino acids near the substrate recognition endof the active site including His 152 and Pro 187 is critical to the shape complementarity ofestrogenic ligands. His 221 and Glu 282 form hydrogen bonds with 3-hydroxyl of thearomatic A-ring of the ligand. This mechanism of recognition of E1 by 17HSD1 is similar tothat of E2 by estrogen receptor a. In a ternary complex with NADP+ and equilin, an equineestrogen with C7=C8 double bond, the orientation of C17=O of equilin relative to theC4-hydride is more acute than the near normal approach of the hydride for the substrate. Inthe apo-enzyme structure, a substrate-entry loop (residues 186–201) is in an open conforma-tion. The loop is closed in this complex and Phe 192 and Met 193 make contacts with theligand. Residues of the entry loop could be partially responsible for the estrogenic specificity.© 2001 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Short-chain dehydrogenase/reductase; 17b-hydroxysteroid dehydrogenase; Structure–functionrelationship; Estrogen; Estradiol; Molecular recognition

www.elsevier.com/locate/chembiont

* Corresponding author. Hauptman-Woodward Medical Research Institute, Inc., 73 High Street,Buffalo, NY 14203, USA. Tel.: +1-716-8569600; fax: +1-716-8526086.

E-mail address: [email protected] (D. Ghosh).

0009-2797/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved.

PII: S0009 -2797 (00 )00255 -6

D. Ghosh, P. Vihko / Chemico-Biological Interactions 130–132 (2001) 637–650638

1. Introduction

17b-hydroxysteroid dehydrogenases (17HSDs) are a group of enzymes that areinvolved in interconversion of active and inactive forms of androgens and estrogens[1–7] by the NAD(P)(H)-linked oxidoreductive transfer of a hydride to and fromthe 17-position of steroid molecules (Fig. 1). Eight separate genes of 17HSDisozymes, numbered 1 through 8, have been identified and cloned [2–8]. Theseisozymes differ in substrate- and tissue-specificities and in the preferred direction ofthe reaction. In human breast tissues, the most active form of estrogen, 17b-estra-diol (E2), is formed by reduction of the inactive estrogen, estrone (E1), catalyzed bytype 1 17HSD (17HSD1), as described in Fig. 1. The estrogenic specificity of17HSD1 as well as its preference for the reduction reaction has been well estab-lished [9–11]. In contrast, type 2 17HSD predominantly catalyzes the oxidativereaction and has androgenic specificity as well, whereas type 3 17HSD exhibitsandrogenic specificity, converting androstenedione to testosterone in the reductivepathway. The physiological roles of types 4-8 are not yet fully understood, althoughthey seem to demonstrate substrate- and tissue-specificities [6,7]. Recently, a type 717HSD that also has estrogenic activity has been characterized in rat ovary [12].

17HSD1 is expressed in steroidogenic tissues including estrogen target tissues,such as normal and malignant endometrium and the breast [13–18]. Because of itsestrogenic specificity and preference for the E1-to-E2 reduction direction, theenzyme is considered to be primarily responsible for E2 biosynthesis in gonads andin peripheral tissues. This enzyme has been proposed to be involved in maintaininghigh E2 levels at or near breast tumors of postmenopausal women [[19] andreferences therein]. A direct correlation between higher concentrations of E2 andonset of breast cancer, especially in post-menopausal cases, is proposed [19] andreferences therein]. There are reports of elevated E2/E1 concentration ratios inbreast tumors in comparison with the E2/E1 ratio in circulating blood [20].Furthermore, a number of studies appear to indicate higher levels of 17HSD1activities in the outer quadrant of the breast where tumors most commonly occur[20–22].

Detailed knowledge of the biochemistry and molecular biology of 17HSD1 hasgrown rapidly in the last 10 years, culminating in the determination of thethree-dimensional structure of the human enzyme [23]. This has led to an atomiclevel description of the E2 binding pocket of the enzyme and an understanding of

Fig. 1. Stereospecific reduction of E1-to-E2 by 17b-hydroxysteroid dehydrogenase type 1.

D. Ghosh, P. Vihko / Chemico-Biological Interactions 130–132 (2001) 637–650 639

its mechanism of action, and the molecular basis for the estrogen-specificity of theenzyme [9–11]. In addition, complexes of 17HSD1 with E2 and/or NADP+

[24–26], various mutant complexes [27], and structure–function analysis throughsite-directed substitutions and enzyme chimeras [10,11] have been published, whichfurther clarify the mechanism of action and provide additional detailed insights intothe origin of estrogenic/androgenic specificities of the enzyme. 17HSD1 belongs tothe short-chain dehydrogenases/reductases family (SDR) [28], requiring a Tyr–X–X–X–Lys motif and a Tyr–Lys–Ser catalytic triad for activity [23,28–30].

The first crystal structure of human estrogenic 17HSD1 from human placenta aswell as the recombinant enzyme was determined in 1995 [23]. The 2.20 A, resolutionstructure of 17HSD1 (shown in Fig. 2) revealed a fold characteristic of SDRenzymes. The active site contains the Tyr–X–X–X–Lys sequence and a Serresidue. The structure also contains three a-helices and a helix-turn-helix motif, notobserved in other short-chain dehydrogenase structures. These helices restrict accessto the active site and appear to influence substrate specificity. Recently, a structureof the ternary complex of 17HSD1 with equilin and NADP+ has been determined[31]. These structural studies together with side-directed mutations and constructionof recombinant chimeric enzymes have led to an understanding of the molecularbasis of substrate recognition and catalytic mechanism.

2. The active site and substrate recognition

The active site cleft in 17HSD1 serves a dual purpose: substrate recognition/dis-crimination and catalysis. There are two components to the recognition of anestrogenic ligand: recognition of the 3-hydroxyl group and shape complementarityof the planar A-ring and near co-planarity of A–B rings. Fig. 3a is a schematicdiagram of the substrate binding cleft. It consists of a hydrophobic pocket with twohydrophilic ends: to the right is the catalytic machinery consisting of the so-calledcatalytic triad Tyr 155–Ser 142–Lys 159, and to the left is the 3-OH grouprecognizing His 221 and Glu 284. The 17-keto oxygen accepts protons fromcatalytic residues Tyr 155 and Ser 142 (2.7 and 2.8 A, , respectively) at the catalyticend of the steroid-binding cleft. The 3-hydroxyl group of the ligand makes abifurcated hydrogen bond to His 221 and Glu 282 (2.9 and 2.9 A, , respectively) atthe recognition end of the cleft. With the exception of these two hydrophilic ends,the rest of the steroid-binding cleft is almost exclusively hydrophobic. Residues Val143, Met 147, Leu 149, Pro 150, Asn 152 and Tyr 218 comprise the complementaryprotein surface in the vicinity of the b-face of the nearly coplanar A–B rings of theligand. Residues Val 225 and Pro 187 are within van der Waal’s contact distancesto the equilin a-face, whereas residues Leu 262, Leu 263 and Met 279 are at thefloor of the cleft. In addition to the hydrophobic environment of the cleft, Phe 192and Met 193 from the substrate-entry loop (residues 186–201) line the entry path.In the apo-enzyme structure, the substrate-entry loop adopts an open conformationproviding unrestricted access to the active site cleft [23]. Instead, the substrate-entryloop forms a closed conformation in the 17b-HSD1-equilin complex, where the


Fig. 2. Structure of human 17b-hydroxysteroid dehydrogenase type 1 shown with co-enzymes NADP+

and E2. Residues Tyr 155-Lys 159-Ser 142 forming the catalytic triad are also shown. The substraterecognition region is shown in light shading [23].

polypeptide chain with residues 186–201 moves towards the catalytic cleft, restrict-ing the access to the active site, as depicted in Fig. 3a. In this closed conformationPhe 192 and Met 193 make van der Waal’s contacts (3.9 and 4.2A, , respectively) tothe ligand molecule.


Several point and multiple mutations and construction of chimeric enzymesbetween human and rat 17HSD1 have led to the discovery that the region formedby residues 148–230 is mostly responsible for estrogenic specificity of the humanenzyme [11]. While the human enzyme is specific for E1, rat 17HSD1 has nearlyequal preference for E1 and androstenedione. This mutational result is in completeagreement with the structural data that the recognition end of the active site is

Fig. 3. (a) A schematic diagram of the active site of the 17HSD1-equilin-NADP+ ternary complex.Residues belonging to the hydrophilic catalytic and recognition ends, as well as residues lining thehydrophobic surrounding are shown. (b) Ligand-binding interactions in human estrogen receptor-a,drawn after Fig. 3a of Ref. [33] describing the crystal structure of 17b-estradiol complex of estrogenreceptor’s ligand-binding domain.


Fig. 4. The interactions that could be critical to the stability of the tertiary structure of the steroidbinding cleft at the b-face of the steroid and to the shape-complementarity of the cleft. Asn 152 makestwo hydrogen bonds to carbonyl of Pro 150 and to Tyr 218 OH. Other hydrogen bonding interactionsthat E2 makes are also shown. Pro 187 packs against the a-face of C-ring.

comprised primarily of C-terminal residues. Mutational data also suggest thatAsn152His, Asp153Glu and Pro187Ala changes from human to rat enzymes alsoare greatest contributors to less estrogenic and more androgenic specificity. Theloop, comprised of Asn 152 and Asp 153, makes up the b-face of the E1 bindingcleft in the active site, near the planar A–B ring system of E1 (Fig. 4). This loop,held by two hydrogen bonds between Asn 152 sidechain and Pro 150 mainchaincarbonyl and Tyr 218 OH, is critically responsible for the shape complementarity ofthe coplanar A–B rings of E1. Modeling results demonstrate that the shapecomplementarity to the planar A–B rings is lost in rat 17HSD1 due to substitutionof these important residues [11]. The closing of the substrate-entry loop describedabove on ligand binding is hinged around Pro 187, which could modulate itsconformation depending on shape and identity of the ligand. Pro 187 sidechain isclosest to the a-face of the C-ring of ligands (Fig. 4) and is thus postulated to beresponsible for recognition of the C-ring puckering. Chimeric enzymes Rat170/Hu-man171, Rat184/Human185, Rat190/Human191 and Rat266/Human267 all showprogressively less estrogenic specificity, implicating critical roles of these residues insubstrate recognition. This result also confirms the role of His 221 in estrogenrecognition, which was previously verified by a point mutation and in vitro assay[32]. However, the role of Glu 282 could not be confirmed [10].


Interestingly, this mechanism of recognition of estrogenic ligands by hydrogenbonding with their characteristic 3-hydroxyl groups is utilized in estrogen receptoras well. In the crystal structure of the complex of E2 with human estrogenreceptor-a ligand-binding domain shown schematically in Fig. 3b [33], the 3-OHgroup is hydrogen bonded to a glutamic acid sidechain and a water molecule whilea histidine sidechain hydrogen bonds to 17b-OH. In 17HSD1, a deprotonatedhistidine sidechain, His 221, is critical to the recognition of the 3-OH group of thesubstrate estrone by accepting a proton from it. As E2 is more potent ligand fortranscriptional activation than E1, the histidine sidechain is used to discriminate ahydroxyl from a ketone. A histidine sidechain, His 524, positioned near the D-ringaccepts a proton from17b-OH while a glutamic acid accepts a proton from the3-OH. Thus, a mechanism similar to recognition of the estrogenic 3-OH of thesubstrate in 17HSD1 is employed by the estrogen receptor for recognition of itsnatural agonist E2.

3. Catalysis at the active site

Based on structural and mutational studies, several groups have proposedputative mechanisms of reaction for various SDR enzymes, all of which involveconserved catalytic Tyr, Lys and Ser sidechains. Specifically for 17HSD1, a putativemodel for E1-to-E2 transition, in which residues Tyr 155, Lys 159, and Ser 142have functional roles in the catalysis, has been proposed (Fig. 5). It has been shownby site directed mutational studies that Tyr 155, Lys 159, Ser 142 and His 221 areessential for the activity [32]. In the proposed structure-based hypothesis for themechanism of hydride-transfer [23], the role of Lys 159 sidechain is conceived

Fig. 5. Proposed mechanism of reaction for line stereospecific reduction of E1 to E2 by 17HSD1 [23].In the putative transition state the pro-S hydride is transferred to the a-face of steroid at C17 after anelectrophilic attack by Tyr 155 proton on C17 keto oxygen.


primarily to lower the pKa of Tyr 155 hydroxyl proton. Although in several SDRstructures, including 3a,20b-HSD and 17HSD1, the conserved Lys sidechain inter-acts with 2% and 3% hydroxyl groups of the nicotinamide ribose moiety of thecoenzyme, such interactions might only be necessary during the coenzyme bindingphase, if at all, and/or be present in the crystal structure only. A closer interactionthan what is observed in crystal structures between Tyr 155 OH and Lys 159 NZgroups (about 4–4.5A, in crystal structures) is a distinct possibility at the initiationof the transition state. This could be easily achieved if the Lys 159 sidechain adoptsa second preferred conformation (unpublished modeling result). The presence of apositive charge at such a close proximity could lower the pKa of the Tyr 155hydroxyl proton to the physiological pH. The following electrophilic attack on theC17-keto oxygen by the hydroxyl group from Tyr 155 through a strong hydrogenbonding interaction, as well as correct orientation and proximity of C4 pro-Shydride of the nicotinamide ring at the a-face of estrone could complete theinitiation of the transition state. Modeling of the transition state with estronesuggests that the transfer of HS

− should occur through a short distance of about 2A,or less between C4 of the nicotinamide ring and C17 of estrone. The approach ofthe hydride to the D-ring should occur in a direction nearly perpendicular to theplanar C17=O group of the substrate. The Ser 142 hydroxyl could as well play therole of Tyr 155 and be actually the proton donor in the catalysis. However, havinga higher intrinsic pKa than a Tyr and being further away (�7A, ) from the positivecharge of Lys 159, this seems less likely than the former scenario. Rather, a role forthe Ser 142 OH group in stabilization of the negatively charged oxianionicintermediate of Tyr 155 seems more plausible.

Finally, the proton loss must be replenished by the solvent network surroundingthe active site cavity. One of these three sidechains could, therefore, be exchangingproton with solvent molecules in the cavity. Bound water molecules, including theones bound to the catalytic Lys, have been reported in crystal structures of SDRs[23,30].

4. What initiates the hydride transfer to and from the coenzyme?

It is conceivable that one or more functional groups of a SDR enzyme isresponsible for the creation of the electromotive force that drives the hydridetransfer. This could be achieved by either a nucleophile or an electrophile ap-proaching the atom C4 of the nicotinamide ring from the pro-R (A) face,depending on whether the enzyme is catalyzing a reduction or an oxidationreaction, respectively. In nearly all SDRs studied structurally, especially the shortchain steroid dehydrogenases, such as in 3a,20b-HSD, 17b-HSD1, 7a-HSD, 20b-HSD, there is a conserved glycine (Fig. 6) which also assumes a spatially conservedposition in the proximity of the A-face of the nicotinamide ring. In nearly all of theshort-chain steroid dehydrogenase structures, including 3a,20b-HSD, 17b-HSD1,7a-HSD and 20b-HSD, the carbonyl oxygen of this glycine is pointed at the C4 HR

group at an O….C distance of about 3.5A, , as shown in Fig. 7 for 17b-HSD1. Such


Fig. 6. A comparison of amino acid sequences in SDRs in the bF strand near the binding site of thenicotinamide ring of coenzyme. Conservation of a glycine residue that may have a catalytic role isshown.

a nucleophilic attack on the C4 atom could facilitate the transfer of a pro-S hydridefrom a reduced coenzyme for a reduction reaction. This glycine is often from aconserved Gly–Pro or Pro–Gly pair at the flexible C-terminal end of the b-strandbF (Fig. 6). It is conceivable that, for a dehydrogenation reaction, an electrophilicmain-chain amide hydrogen, possibly of the same conserved glycine, points at C4instead, while the carbonyl points away from it, thereby facilitating a hydridetransfer to an oxidized coenzyme. Glycine is an ideal amino acid to provide thebackbone flexibility in order for an enzyme to be functionally active both inreductive as well as in oxidative directions.

Fig. 7. A putative catalytic role for Gly 186. Carbonyl of Gly 186 points toward C4HR of thenicotinamide ring of coenzyme in the active site of the ternary complex of 17HSD1. Bound equilin andcatalytic Tyr 155 are also partially visible.


Fig. 8. Differences in three-dimensional structures of C and D rings of E1 and equilin are demonstratedby superimposing their A and B rings.

5. Catalysis versus inhibition

Equilin (3-hydroxyestra-1,2,5,7-tetraen-17-one) is one of the major componentsof estrogens used in estrogen replacement therapy, in conjunction with estrone and17a-dihydroequilin. These conjugated estrogens are administered under the com-mercial name Premarin (physicians desk reference, 1996) as salts of their sulfateesters, which are subsequently hydrolyzed to free estrogens. The 17HSD1–equilin–NADP+ complex crystallized with a dimer in the asymmetric unit. This is incontrast to the apo- and wild-type 17HSD1–E2 complexes [23–25] in which thedimer was crystallographically related. A homodimer is known to be the functionalunit of the enzyme. This holo-form represents a true ternary complex of the wildtype enzyme, with the cofactor and a steroidal ligand. Seventy seven percentinhibition of 17HSD1 enzyme activity was achieved with a 1mM concentration ofequilin, establishing it as a potent inhibitor of E1-to-E2 reduction [31].

Both equilin and NADP+ have well defined electron density in the A subunit ofthe dimeric enzyme. However, the ligand density in the B subunit is poorly defined.As in the E2-complex, the equilin molecule makes four hydrogen bond formingcontacts with protein atoms, two with Tyr 155 and Ser 142 at the catalytic end anda bifurcated hydrogen bond to His 221 and Glu 282 at the recognition end (Fig.3a). The chemical structure of equilin differs from the E1 only by the presence ofa C7=C8 double bond. However, three-dimensional structures of the substrate E1and equilin are strikingly different at the C–D ring systems, because of the presenceof the C7=C8 double-bond in equilin, as shown in Fig. 8. The difference in torsionangle C7–C8–C9–C11 (−179° for estrone and 121° for equilin [34]) caused by theC7=C8 double bond results in a 0.9A, displacement between the C17 carbonatoms. Higher isotropic temperature factors of the atoms of the D-ring of equilin(�40 A, 2) in comparison with those of A–B rings (�30 A, 2) are also suggestive of


higher thermal motion at this end. Ideally, for a reduction reaction, it is postulatedthat the hydride should approach the carbon atom roughly normal to the carbonylplane. The orientation of C17=O of equilin relative to the C4-hydride is moreacute (52.7°) than in the above scenario, owing to the differences in puckering ofthe C–D ring system. The distance between the C4 atom of the nicotinamide ringof the cofactor and C17 of equilin is 3.4 A, . However, in order to evaluate thestructural origin of inhibition by equilin, the active site structure of the E1–17HSD1 complex should ideally be compared with that of the 17–HSD1–equilincomplex, since E2 is not a substrate, but the product of catalysis. No such complexstructure is available to date. The origin of the inhibitory property of equilin, fromthe structural perspective, is, therefore, a manifestation of its altered C–D-ringstructure, and location and orientation of its C17 keto group, with respect to thecatalytic machinery at the catalytic end of the active site.

6. Ligand-entry loop interactions and possible co-operativity

The open and closed conformations of the substrate-entry loop are illustrated inFig. 9. The overall tertiary structures of the apo-enzyme and the equilin-complexare nearly identical, except for the substrate-entry loop (shown as thicker cross-sec-tion of the backbone) between the strand bF and the helix aG. The root mean

Fig. 9. Comparison of apo–17HSD1 and 17HSD1–Equilin–NADP+ complex structures, showing openand closed forms of the substrate-entry loop.


square deviation for 284 Ca-atoms of the 17HSD1–equilin complex compared tothe apo form is 0.7A, ; the root mean squared deviation for the substrate-entry loopis 3.6A, , with a maximum of 6.8A, at Lys 195. The loop in the A subunit has welldefined electron density except for Lys 195 sidechain and packs against both equilinand NADP+ via Phe 192 and Met 193. Both Phe 192, and Met 193 line thesubstrate-entry path and have van der Waals contacts with the D ring of equilinand the nicotinamide head group of NADP+. This closing of the substrate-entryloop effectively traps the ligand in the steroid binding cleft by occluding the entrypath shown in Fig. 3a. This combination of an ordered loop and the presence ofboth the cofactor and ligand are unique features of the A-subunit of the 17b–HSD1–equilin complex. In contrast, the electron density for the loop in the Bsubunit is poorly defined after Thr 190. This result may be a direct consequence ofcooperativity that may exist within the dimeric molecule of the enzyme.

In all SDR enzyme structures, the dimer interface made of the four-helix bundleaE’s and aF’s is strictly conserved. Two catalytic sites of the dimer are adjacent tothis helix bundle and the catalytic motif Tyr–X–X–X–Lys resides on aF’s. It isconceivable that structural perturbations that follow substrate and coenzymebinding in one of the sites modify interfacial interactions among the helices and aretransmitted across the interface to the other site. However, such structural differ-ences, if any, between the two subunits of the apo-enzyme dimer and its ternarycomplexes have not yet been detected.

Understanding of the structural basis for estrogen recognition and catalysis hasled to new ideas for further research towards design of specific, high affinityinhibitors of the enzyme that someday could be used as agents for lowering levelsof active estrogen.

Acknowledgements

This research is partially supported by grant no. DK26546 from NIH and a grantfrom Roswell Alliance Foundation.

References

[1] K.J. Ryan, L.L. Engel, The interconversion of estrone and estradiol by human tissue slices,Endocrinology 52 (1953) 287–291.

[2] H. Peltoketo, V. Isomaa, O. Maentausta, R. Vihko, Complete amino acid sequence of humanplacental 17b-hydroxysteroid dehydrogenase deduced from cDNA, FEBS Lett. 239 (1988) 73–77.

[3] V. Luu-The, et al., Characterization of cDNAs for human estradiol 17b-dehydrogenase andassignment of the gene to chromosome 17: evidence to two mRNA species with distinct 5%-terminiiin human placenta, Mol. Endocrinol. 3 (1989) 1301–1309.

[4] L. Wu, et al., Expression, cloning and characterization of human 17b-hydroxysteroid dehydroge-nase, a microsomal enzyme possessing 20b-hydroxysteroid dehydrogenase activity, J. Biol. Chem.268 (1993) 12964–12969.

[5] W.M. Geissler, et al., Male pseudohermaphroditism caused by mutations of testicular 17b-hydrox-ysteroid dehydrogenase 3, Nature Genet. 7 (1994) 34–39.


[6] J. Adamski, et al., Molecular cloning of a novel widely expressed human 80 kDa 17b-hydroxys-teroid dehydrogenase IV, Biochem. J. 311 (1995) 437–443.

[7] Y. Deyashiki, et al., Molecular cloning and characterization of mouse estradiol 17b-dehydrogenase(A-specific), a member of the aldoketoreductase family, J. Biol. Chem. 270 (1995) 10461–10467.

[8] H. Peltoketo, V. Luu-The, J. Simard, J. Adamski, 17b-Hydroxysteroid dehydrogenase (HSD)/17-ketosteroid reductase (KSR) family; nomenclature and main characteristics of the 17HSD/KSRenzymes, J. Mol. Endocrinol. 23 (1999) 1–11.

[9] M. Poutanen, M. Miettinen, R. Vihko, Differential estrogen specificities for transiently expressedhuman placental 17b-hydroxysteroid dehydrogenase and an endogenous enzyme expressed incultured COS-m6 cells, Endocrinology 133 (1993) 2639–2644.

[10] T. Puranen, M. Poutanen, D. Ghosh, P. Vihko, R. Vihko, Characterization of structural andfunctional properties of human 17b-hydroxysteroid dehydrogenase type 1 using recombinantenzymes and site-directed mutants, Mol. Endocrinol. 11 (1997) 77–86.

[11] T. Puranen, M. Poutanen, D. Ghosh, R. Vihko, P. Vihko, Origin of substrate specificityof humanand rat 17b-hydroxysteroid dehydrogenase type 1, using chimeric enzymes and site-directedsubstitutions, Endocrinology 138 (1997) 3532–3539.

[12] H. Peltoketo, P. Nokelainen, Y.S. Piao, R. Vihko, P. Vihko, Two 17b-hydroxysteroid dehydroge-nases of estradiol biosynthesis: type 1 and type 7, J. Steroid. Biochem. 69 (1999) 431–439.

[13] N. Fournet-Dulguerov, et al., Immunohistochemical localization of aromatase cytochrome P-450and estradiol dehydrogenase in the syncytiotrophoblast of the human placenta, J. Clin. Endocrinol.Metab. 65 (1987) 757–764.

[14] O. Maentausta, et al., Immunohistochemical localization of 17b-hydroxysteroid dehydrogenase inthe human endometrium during the menstrual cycle, Lab. Invest. 65 (1991) 582–587.

[15] E. Dupont, F. Labrie, V. Luu-The, G. Pelletier, Localization of 17b-hydroxysteroid dehydrogenasethroughout gestation in human placenta, J. Histochem. Cytochem. 39 (1991) 1403–1407.

[16] S.A. Ghersevich, M.H. Poutanen, H.K. Martikainen, R. Vihko, Expression of 17b-hydroxysteroiddehydrogenase in human granulosa cells: correlation with follicular size, cytochrome P450 aro-matase activity and oestradiol production, J. Endocrinol. 143 (1994) 139–150.

[17] O. Maentausta, et al., Immunohistochemical study of the human 17b-hydroxysteroid dehydroge-nase and steroid receptors in endometrial adenocarcinoma, Cancer 70 (1992) 1551–1555.

[18] M. Poutanen, V. Isomaa, V.-P. Lehto, R. Vihko, Immunological analysis of 17b-hydroxysteroiddehydrogenase in benign and malignant human breast tissue, Int. J. Cancer 50 (1992) 386–390.

[19] M. Poutanen, V. Isomaa, H. Peltoketo, R. Vihko, Role of 17b-hydroxysteroid dehydrogenase type1 in endocrine and intracrine estradiol biosynthesis, J. Steroid Biochem. Mol. Biol. 55 (1995)525–532.

[20] V.H.T. James, J.M. McNeil, P.A. Beranek, R.C. Bonney, M.J. Reed, The role of tissue steroids inregulating aromatase and oestradiol 17b-hydroxysteroid dehydrogenase in breast and endometrialcancer, J. Steroid Biochem. 25 (1986) 787–790.

[21] E.F. Adams, et al., Paracrine influence of human breast stromal fibroblastson breast epithelial cells:secretion of a polypeptide which stimulates reductive 17b-oestradiol dehydrogenaseactivity, Int. J.Cancer 42 (1990) 119–122.

[22] V.H.T. James, et al., Regulation of estrogen concentrations in human breast tissues, Annals N.Y.Acad. Sci. 595 (1990) 227–235.

[23] D. Ghosh, et al., Structure of human estrogenic 17b-hydroxysteroid dehydrogenase at 2.20A,resolution, Structure 3 (1995) 503–513.

[24] A. Azzi, P.H. Rehse, D.W. Zhu, R.L. Campabell, F. Labrie, S.X. Lin, Crystal structure of humanestrogenic 17b-hydroxysteroid dehydrogenase complexed with 17b-estradiol, Nat. Struct. Biol. 3 (8)(1996) 665–668.

[25] R. Breton, D. Housset, C. Mazza, J.C. Fontecill-Camps, The structure of a complex of human17b-hydroxysteroid dehydrogenase with estradiol and NADP+ identifies two principal targets forthe design of inhibitors, Structure 4 (1996) 905–915.

[26] S.-X. Lin, et al., Studies on the three-dimensional structure of estrogenic 17b-hydroxysteroiddehydrogenase, J. Endocrinol. 150 (1996) S13–S20.


[27] C. Mazza, R. Breton, D. Housset, J.C. Fontecilla-Camps, Unusual charge stabilization of NADP+

in 17b-hydroxysteroid dehydrogenase, J. Biol. Chem. 273 (1998) 8145–8152.[28] H. Jornvall, et al., Short-chain dehydrogenases/reductases (SDR), Biochemistry 34 (1995) 6003–

6013.[29] D. Ghosh, et al., Three-dimensional structure of holo 3a,20b-hydroxysteroid dehydrogenase: a

member of the short-chain dehydrogenase family, Proc. Natl. Acad. Sci. USA 88 (1991) 10064–10068.

[30] D. Ghosh, Z. Wawrzak, C.M. Weeks, W.L. Duax, M. Erman, The refined three-dimensionalstructure of 3a,20b-hydroxysteroid dehydrogenase and possible roles of the residues conserved inshort-chain dehydrogenases, Structure 2 (1994) 629–640.

[31] M.W. Sawicki, M. Erman, T. Puranen, P. Vihko, D. Ghosh, Structure of the ternary complex ofhuman 17b-hydroxysteroid dehydrogenase type 1 with 3-hydroxyestra-1,3,5,7-tetraen-17-one (equi-lin) and NADP+, Proc. Natl. Acad. Sci. USA 96 (1999) 840–845.

[32] T.J. Puranen, M.H. Poutanen, H.E. Peltoketo, P.T. Vihko, R.K. Vihko, Site-directed mutagenesisof the putative active site of human 17b-hydroxysteroid dehydrogenase type 1, Biochem. J. 304(1994) 289–293.

[33] D.M. Tanenbaum, Y. Wang, S.P. Williams, P.B. Sigler, Crystallographic comparison of theestrogen and progesterone receptors’ ligand binding domains, Proc. Natl. Acad. Sci. USA 95 (1998)5998–6003.

[34] M.W Sawicki, N. Li, D. Ghosh, Equilin. Acta Crystallogr. C55 (1999) 425–427.

.

molecular mechanisms of estrogen recognition and 17-keto reduction by human 17β-hydroxysteroid...

Documents