Elsevier

Chemico-Biological Interactions

Volumes 143–144, 1 February 2003, Pages 481-491
Chemico-Biological Interactions

17-β-Hydroxysteroid dehydrogenase type 1: computational design of active site inhibitors targeted to the Rossmann fold

https://doi.org/10.1016/S0009-2797(02)00184-9Get rights and content

Abstract

17-β-Hydroxysteroid dehydrogenase type 1 (17βHSD1), also called estradiol dehydrogenase, catalyzes the NADPH-dependent reduction of the weak estrogen, estrone, into the more potent estrogen, 17-β-estradiol. 17βHSD1 is an attractive drug target in hormone-sensitive breast cancer. Past efforts to develop selective inhibitors of 17βHSD1 have focused on design of substrate analogs. It is challenging to develop steroid analogs that are devoid of any undesired biological activity. 17βHSD1 is a member of the short-chain dehydrogenase/reductase (SDR) superfamily that includes many hydroxysteroid dehydrogenases. Members of the SDR family bind NAD(P)(H) in a motif that is a modified Rossmann fold. We demonstrated previously that the Rossmann folds of classical dehydrogenases can be selectively inhibited by derivatives and analogs of the natural product gossypol. In this study, we have addressed the question whether the modified Rossmann fold in 17βHSD1 is a target for identification of lead compounds for structure-based drug design. 17βHSD1 was purified from human placenta. 17βHSD1 is inhibited by derivatives of gossypol with dissociation constants as low as 2 μM. Inhibition is competitive with the binding of cofactor. Molecular modeling studies using the published coordinates of human 17βHSD1 suggest that these inhibitors occupy the modified Rossmann fold at the nicotinamide end of the dinucleotide-binding site, extending towards the substrate site. A computational approach was used to design potential new inhibitors of 17βHSD1. The results suggest not only that derivatives of gossypol represent attractive lead compounds for structure-based drug design but also suggest that appropriate incorporation of a substrate analog into the design of these Rossmann fold inhibitors may provide pan-active site inhibitors that span the cofactor and substrate site, potentially offering specificity and increased potency.

Introduction

17β-Estradiol (E2), the most potent of human estrogens, is known to stimulate the growth of breast cancer cells [1]. In addition, a large fraction of breast tumors are hormone-sensitive. E2 functions at the nuclear level through interaction with the estrogen receptor, leading to subsequent regulation of a battery of genes that control the proliferation of mammary epithelial cells [2]. Consequently, interfering with the mitogenic activities of E2, either through blocking its production or by inhibiting its receptor interaction, has become a major goal. Attempts to block E2-receptor interactions have led to the design of inhibitors that are steroid analogs [3]. It is a challenge, however, to create analogs that exhibit selective action against the estrogen receptor, thereby eliminating undesirable biological activities outside this pathway. Therefore, limiting E2 production may prove to be a more attractive approach to the design of new therapeutics for breast cancer.

E2 is synthesized locally in peripheral targets from its inactive precursor dehydroepiandrosterone (DHEA) or its sulfate derivative (DHEA-S). This local control of active hormone levels is unique to man and a few primates, and has been termed “intracrinology”, distinguishing it from the process by which active hormone is taken from the circulation or extracellular space [4]. In order to synthesize E2, estrone (E1) must be produced from DHEA(-S), whether it be in breast epithelia or other tissues. The final reaction, occurring in breast epithelia, reduces the weak estrogen E1 to the active estrogen E2. Inhibition of this reaction catalyzed by 17-β-hydroxysteroid dehydrogenase type 1 (17βHSD1) provides a method to lower E2 production in the target tissue.

17βHSD1 is a member of the short-chain dehydrogenase/reductase (SDR) family. A number of the members of this family utilize nicotinamide adenine dinucleotides (NAD(P)(H)) as cofactors for steroid reduction or oxidation reactions. SDR proteins bind NAD(P)(H) in a motif known as the Rossmann fold, which is the cofactor-binding site in the majority of dehydrogenases [5]. The 17βHSD1 reaction is reversible and dependent on the type of cofactor (NAD(H) or NADP(H)) [6]. In vivo, however, the enzyme acts primarily as a steroid-keto reductase [7], maintaining intracellular levels of E2. In this reaction, the pro-S hydride from the reduced nicotinamide ring is transferred to the C17 carbonyl of E1 to form the more potent E2[8]. The bisubstrate reaction is reported to occur via a random mechanism [9], providing two sites that can be targeted for inhibition, i.e., the E2-binding site and the Rossmann fold.

We have previously demonstrated that the natural product gossypol, a polyphenolic binaphthyl isolated from cottonseed, inhibits all isozymes of human lactate dehydrogenase (LDH), which also contain the Rossmann fold. Several derivatives of gossypol, along with many analogs, have been synthesized. These compounds exhibit a range of selectivities for human LDHs, with inhibition constants as low as 30 nM [10], [11], [12]. Inhibition by these compounds is consistently competitive with the binding of NADH. These data, along with the structural conservation of the Rossmann fold across many oxidoreductase enzymes, suggest that these compounds may represent lead structures for design of inhibitors of dehydrogenases that possess a Rossmann fold. In this study, we evaluated gossypol, gossypol derivatives, and gossypol analogs as inhibitors of human 17βHSD1. In addition, computational approaches were used to model 17βHSD1-ligand interactions, and to suggest a further direction for the design of new inhibitors.

Section snippets

Synthesis of gossypol analogs and derivatives

Derivatives and analogs of gossypol were prepared as described previously [10], [13], [14], [15].

Protein purification

The protocol for purification of 17βHSD1 was a modification of that reported by Yang et al. [16]. Fresh human placenta, 250 g, was cubed and homogenized. The 100,000×g supernatant fraction was purified on a Blue Sepharose CL-6B column. The eluent was concentrated by pressure filtration, desalted on a PD-10 column and chromatofocused. Enzyme purity was examined using SDS-PAGE electrophoresis on a 20%

Compound screening

Seven gossypol-related compounds were screened against 17βHSD1 at pH 9.2. All inhibitors were tested at a concentration of 25 μM with 0.5 mM NAD and 25 μM E2. Addition of gossypol resulted in only a slight reduction in enzyme activity (Fig. 1). Four gossypol derivatives, in which the aldehyde functional group is modified, were tested. The peri-acylated nitriles, gossylic nitrile 1,1′-diacetate (GNDA) and gossylic nitrile 1,1′-divalerate (GNDV), represent compounds in which the aldehyde group is

Gossypol-related compounds as lead structures for the inhibition of Rossmann folds

Early characterization of the dinucleotide-binding sites of a series of dehydrogenases was reported by Rossmann et al. [20], who defined the structural conservation in the cofactor-binding sites of LDH, alcohol dehydrogenase, glyceraldehyde-3P dehydrogenase, and malate dehydrogenase. Subsequent characterization of the structures of other dinucleotide-binding proteins resulted in an extended definition of the Rossmann fold by Wierenga et al. [21] and Bellamacina [22]. The “classical” Rossmann

Acknowledgements

This work was supported by US Army/DOD Breast Cancer Program grants DAMD17-00-0372 (DLVJ) and predoctoral awards DAMD17-00-1-0368 (WMB) and DAMD17-00-0369 (JPB).

References (29)

  • H. Peltoketo et al.

    Regulation of estrogen action: role of 17 beta-hydroxysteroid dehydrogenases

    Vit. Horm.

    (1999)
  • T.M. Penning

    Molecular endocrinology of hydroxysteroid dehydrogenases

    Endocr. Rev.

    (1997)
  • H. Jornvall et al.

    Short-chain dehydrogenases/reductases (SDR)

    Biochemistry

    (1995)
  • L.M. Deck et al.

    Selective inhibitors of human lactate dehydrogenases and lactate dehydrogenase from the malarial parasite Plasmodium falciparum

    J. Med. Chem.

    (1998)
  • Cited by (29)

    • Virtual screening applications in short-chain dehydrogenase/reductase research

      2017, Journal of Steroid Biochemistry and Molecular Biology
      Citation Excerpt :

      Therefore, applying nonrestrictive pharmacophore models of related enzymes as additional filter tools can assist the selection of virtual hits for biological testing by eliminating promiscuous inhibitors. Regarding the structural similarity of different proteins, Brown et al. demonstrated inhibition of human lactate dehydrogenase (LDH) and of 17β-HSD1 by binding of gossypol derivatives to the Rossmann fold [132]. Thus, on one hand structural conservation provides a basis for lead compounds targeting several related proteins but on the other hand it raises concerns about their selectivity.

    • Environmental Pollutants and Hydroxysteroid Dehydrogenases

      2014, Vitamins and Hormones
      Citation Excerpt :

      Soy isoflavone genistein inhibited human 17β-HSD1 with IC50 value of 50 μM (Brooks & Thompson, 2005). Cottonseed flavonoid gossypol competitively inhibited human 17β-HSD1 with Ki as low as 2 μM (Brown et al., 2003). Genkwanin and diosmetin are o-methylated flavones, isolated from the seeds of Alnus glutinosa and plants of Caucasian vetch, respectively.

    • Design and synthesis of bisubstrate inhibitors of type 1 17β-hydroxysteroid dehydrogenase: Overview and perspectives

      2008, European Journal of Medicinal Chemistry
      Citation Excerpt :

      Another way would be to use a more efficient cofactor mimic such as the one in the compound 10 design (Fig. 7), in which the bothersome phosphate group is adequately replaced with a more chemically manageable carboxylic acid group. Other reported Rossman-fold-targeting compounds, such as gossypol derivatives, have also been tested for 17β-HSD1 inhibition [41]. They have been shown to be competitive against the oxidative cofactor NAD+ with a Ki value as low as 2.2 μM [41], as compared to 250 μM for the adenosine esterified with the alkyl side chain spacer of EM-1745 [22].

    View all citing articles on Scopus
    View full text