Molecular modulation of nuclear receptor conformational states
Citation for published version (APA): Scheepstra, M. (2016). Molecular modulation of nuclear receptor conformational states. Technische Universiteit Eindhoven.
Document status and date: Published: 07/09/2016
Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)
Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication
General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.tue.nl/taverne
Take down policy If you believe that this document breaches copyright please contact us at: [email protected] providing details and we will investigate your claim.
Download date: 02. Oct. 2021
Molecular modulation of nuclear receptor conformational states
Proefschrift
ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus, prof.dr.ir. F.P.T. Baaijens, voor een commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op woensdag 7 september 2016 om 16:00 uur.
door
Marcel Scheepstra
geboren te Gouda
Dit proefschrift is goedgekeurd door de promotoren en de samenstelling van de promotiecommissie is als volgt:
Voorzitter: prof.dr. P.A.J. Hilbers
Promotor: prof.dr.ir. L. Brunsveld
Copromotor: dr. L-G. Milroy
Leden: prof.dr. A.R. de Lera (Universidade de Vigo)
prof.dr. C.A.A. van Boeckel (Universiteit Leiden)
prof.dr. K. De Bosscher (Universiteit Gent)
prof.dr. E.W. Meijer
dr. C. Ottmann
Het onderzoek dat in dit proefschrift wordt beschreven is uitgevoerd in overeenstemming met de TU/e gedragscode wetenschapsbeoefening
To Ayla
“Als je eenmaal op de andere oever bent aangekomen, help dan ook anderen die te bereiken.” Siddhartha Gautama
M. Scheepstra ©
Printing: Gildeprint, Enschede
A catalogue record is available from the Eindhoven University of Technology library
ISBN: 978-94-6233-342-0
This research has been financially supported by the Netherlands Organization for Scientific Research via ECHO grant 711011017 and via the Gravity program 024.001.035
Table of contents
Chapter 1
Nuclear receptor function and modulation 1
Nuclear receptor structural organization 2 The nuclear receptor timeline 6 Nuclear receptor function and modulation 10 Beyond the dose response curves 14 Mechanisms for signal transduction 15 Summary points 18 Aim and outline of the thesis 18 References 21 Chapter 2
A natural product switch for a dynamic protein interface 27
Results and discussion 28 Experimental section 37 References 60 Chapter 3
Ligand-induced helix repositioning for either selective RXR heterodimerization or coactivator binding 65
Introduction 66 Chemical synthesis 71 Biochemical evaluation 73 Discussion and concluding remarks 79 Experimental section 80 References 97 Chapter 4
Synthesis and evaluation of a bis-benzanulated spiroketal as RXRα modulator 101
Results and discussion 102 Experimental section 111 References 120 Chapter 5
RORγt reveals allosteric inverse agonism for nuclear receptors 123
Results and discussion 124 Experimental section 129 References 137 Chapter 6
Epilogue 141
Introduction 142 RXR selective dimerization 142 Future directions for RORγt 144 References 148
Summary 151
Samenvatting 153
Curriculum Vitae 155
List of publications 156
Acknowledgements 157
Chapter 1
Nuclear receptor function and modulation
Nuclear receptors are proteins which play a key role in homeostasis, metabolism and development, as well as many diseases such as cancer, obesity and neuro degenerative diseases. Therefore, nuclear receptors are major targets for drug discovery and there are many examples of clinically successful nuclear receptor modulators. A broad spectrum of full, partial or inverse agonistic or antagonistic activities has been developed, which give rise to well-defined active and inactive conformational states. Helped by a better understanding of the underlying structural mechanisms and equilibria in conformational states in nuclear receptor function, there is an emerging interest to develop ligands that stabilize alternative non-classical conformations or are targeting allosteric sites. This aspect in nuclear receptor pharmacology offers possibilities to create selective therapies, and it is an exciting prospect that organic synthesis can provide solutions to many emerging pathological disorders. This chapter provides a general overview of the nuclear receptor structural organization, a brief timeline of nuclear receptor research in the past 70 years and a description of the different classical and non-classical/alternative ways to modulate nuclear receptor function. Finally, this chapter will conclude into a discussion of the challenges of our time and the current pursuits of new ways to modulate nuclear receptors.
1
CHAPTER 1
Nuclear receptor structural organization
The ligand dependence of nuclear receptors, in combination with the many physiological roles in the human body, make them ideal drug candidates. This is reflected in the current number of marketed drugs targeting nuclear receptors.1,2 The complete sequencing of the human genome led to the identification of 48 nuclear receptors,3,4 which can be divided into six evolutionary sub-families.5 They are multi-domain transcription factors, which share an overall similar conformation, consisting of five domains (figure 1.1).
Figure 1.1 | Schematic overview of the nuclear receptor domains.6,7 A variable N-terminal domain, a highly conserved DNA binding domain, a flexible hinge region and the ligand binding domain with 12 conserved helices, the F-domain is present in only a portion of all 48 nuclear receptors.
The N-terminal domain (NTD) contains the ligand-independent activation function (AF1) and is the most variable part amongst all nuclear receptors in both size and sequence. The structure determination of this domain proved to be challenging, not only because of difficulties with actually expressing and purifying the protein, but also because of the high mobility of this domain. Due to the large diversity between nuclear receptors the contribution for activation by this domain differs. The conformation of this domain is considered to be intrinsically disordered,8 with a propensity to a more stabilized conformation upon DNA binding or binding of protein.9 This induced α-helix stabilization co-factor binding is also observed for other protein examples, such as the tumor suppressor P53 – MDM2 binding10 and the α-TIF viral protein.11 Posttranslational modifications also play an important role in determining the structural conformation and therefore activity of this domain.12
2 NUCLEAR RECEPTOR FUNCTION AND MODULATION
The DNA binding domain (DBD) is highly conserved between all the nuclear receptors,13 and is known to bind two zinc ions via two zinc fingers in a globular conformation.14,15 The DBD interacts with the DNA via two α-helices: helix-1 binds base-specific to the major groove of the DNA, while helix-2 acts as a stabilizer for this interaction.16 The C-terminus of the DBD extends into the hinge region. The hinge region is a flexible domain, flanked by the DBD and the ligand-binding domain (LBD),17 which enables different orientations of the flanking DBD and LBD in respect of each other influencing the transactivation.18,19 The orientation of the NTD containing the AF1 with respect to the LDB is determined by the hinge region. The hinge region is less conserved between nuclear receptor subtypes compared to other domains and is known to play a key role in dimerization, DNA binding,20 receptor mobility and nuclear translocation.21 As many flexible domains in proteins, the hinge region is highly decorated with posttranslational modifications (figure 1.2). The effects of these posttranslational modifications on the activation or repression of nuclear receptors are not yet fully understood, but gain in interest.19,22 The hinge region extends into the highly conserved LBD. The LBD is the most extensively researched domain of the nuclear receptor and it plays a central role in the function of the receptor. It binds endogenous ligands, but can also be targeted by drugs (see also table 1.1). The function and modulation of the LBD will be described extensively in this chapter (vide infra). The F-domain is the most C-terminal domain in the nuclear receptor, though very little is known about the biological function, or biophysical or conformational character. The F- domain is highly variable between nuclear receptors in terms of amino acid sequence and number/length. Because of its position within the protein – located after helix 12 in the LBD – the F-domain plays a role in the conformation of helix 12 and therefore in the activation and ligand response of the receptor. Furthermore, the F-domain plays a role in dimerization,23,24 protein-protein interactions with other proteins, such as 14-3-3,25 or ligand independent co-activator binding.26
3 CHAPTER 1
Figure 1.2 | Examples of posttranslational modifications in nuclear receptors. Figure simplified and modified from Le Romancer and coworkers.27 Depicted is the ERα with posttranslational modifications in the several different domains of the receptor with their pharmacological roles. P: phosphorylation, Me: methylation, SUMO: sumoylation, Ac: acetylation, Ub: ubiquitylation, green dots: transcriptional activation, red dots: transcriptional repression.28
Posttranslational modifications (PTMs) play an important role in many physiological processes, especially in signaling.29 Despite considerable efforts to understand the role of PTMs in signaling and other processes, they remain poorly understood. The PTMs differ in size and complexity, from a single functional group (methyl, phosphate, acetyl) to complex peptides (ubiquitin, small ubiquitin-like modifiers) and carbohydrates. The plethora of modifiers contributes to the complexity of PTM research. Nuclear receptors are also decorated with PTMs, for example the estrogen receptor-α (ERα) in figure 1.2, and these PTMs plays a key role in nuclear receptor functioning. PTMs are often found on flexible domains, for nuclear receptors the hinge region and the F-domain. A detailed study on the molecular effect of the phosphorylation of tyrosine-537 in ERα was recently reported, by our group.26 A semisynthetic approach was used to obtain functional, homogeneous ERα containing the phosphorylated tyrosine, demonstrating significant effects of the phosphorylation on the dynamics of the c-terminal part of the ligand binding domain and therefore structural and functional changes in the receptor, leading to subtype specific cofactor recruitment. Moreover, the dimerization of ERα could be negatively controlled by the phosphorylation of threonine-594 and interaction with the 14-3-3 protein. This
4 NUCLEAR RECEPTOR FUNCTION AND MODULATION interaction could then be stabilized with a small natural product. This stabilization results then in the functional inhibition of ERα/chromatin interactions and decreased cell proliferation.25 In the case of ERα positive breast cancer, the phosphorylation of serine-305 leads to a reduced effect of tamoxifen treatment and resistant tumors,30–32 while other PTMs sensitize the receptor.33 For the androgen receptor, which is involved in prostate cancer, PTMs can also lead to tumors which are resistant to medication and ligand- independent activation of the receptor.34,35 Lysine acetylation and deacetylation of steroid receptors has been shown to have diverse functions in signaling. The role of acetyl transferases and deacetylases in cancers and other diseases is only vaguely understood,36 but for example the depletion of deacetylates led to decreased ERα levels in ERα positive tumor cells. Hinge-region acetylation of the progesterone receptor or the androgen receptor largely impacts the translocation to the nucleus and the binding to DNA.19,22 For the estrogen receptor the influence of hinge-region acetylation on nuclear translocation has not been investigated, while for the glucocorticoid receptor acetylation impairs binding to selective genes. Although PTMs for nuclear receptors are difficult to study, a better understanding of the PTM formation, localization and functional outcome has led to significant results. For example in the treatment of metabolic disorders the use of PPARγ ligands (e.g. thiazolidinones) is limited because of the side effects they elicit. New strategies are aiming to block cyclin-dependent kinase-5 mediated phosphorylation of PPARγ in combination with non-agonists, demonstrating improved insulin sensitivity and decreased side-effects.37 Coregulator proteins play a crucial role within the structural organization of nuclear receptors. Nuclear receptor coregulators have gained interest over the last two decades in nuclear receptor research. Since the identification of nuclear receptor-associated protein in 1994, around 400 different coregulator proteins have been discovered.38 They have emerged as regulators of nuclear receptor function and modulation and alteration of expression or function can lead to disease or dysregulation of downstream pathways. The steroid receptor coactivator-1 (SRC-1) was the first coregulator to be identified. Since then, the characterization and deciphering of the function of coregulators has led to an understanding of their role in human physiology. Due to the presence or absence of certain coregulators many nuclear receptor pathways are tissue-specific.39 Coregulator proteins can be divided into two classes on the basis of their functional output. Coactivators are associated with gene expression and typically bind to agonist-bound nuclear receptors. Corepressor proteins interact with apo or antagonist-bound receptors and are associated
5 CHAPTER 1 with gene repression, vide infra for detailed mechanisms on helix 12 repositioning and coregulator recruitment.
The nuclear receptor timeline
The first half of the 20th century saw the identification of thyroid, steroid and retinoid hormones. This research led to Nobel prizes in chemistry and medicine, but more importantly these high affinity hormones served as tracking agents for their receptors.40 Nuclear receptor research started in the second half of the 20th century with the discovery of the estrogen receptor by E.V. Jensen in 1958. The receptor was first shown via the oral administration of radioactively labeled estradiol to female rats. Later experiments in cells confirmed the presence of a receptor capable of binding estradiol without chemically modifying the ligand.41 Upon dosing estradiol to cells, E.V. Jensen observed binding of estradiol in the cytosol, after which the receptor-estradiol complex transformed into a complex with a lower molecular weight, which then translocated to the nucleus.42 The identification of the estrogen receptor initiated the discovery of other nuclear receptors such as the androgen receptor in 1969 by Fang and coworkers43 and the progesterone receptor in 1970 by O’Malley and coworkers,44 all with the use of radio-labeled hormones. These receptors showed similar uptake and retention mechanisms, indicating and hinting towards a superfamily of nuclear receptors. More details about the properties and dynamics of the receptors were discovered through the protein purification methods in the 1970s. The first nuclear receptor to be purified was the glucocorticoid receptor, Wang and coworkers discovered via proteolytic methods that the nuclear receptor consists out of multiple domains.45 Despite all this research, the mechanism of gene activation by nuclear receptors still had to be elucidated. Studies focused on the mRNA levels induced by hormonal stimulation46 and detection of DNA fragments where nuclear receptors were binding, named hormone response elements (HREs). Gronemeyer and coworkers were able to show the binding of nuclear receptors on the chromosome via fluorescently labeled hormones, indicating the sites that were regulated by these hormones.47 A big step forward was the cloning of the cDNA of the glucocorticoid receptor in 1985,48 which was quickly followed by that of the estrogen receptor a year later.49 These results allowed the comparison of the sequences, deletion of the zinc fingers, and interchanging of the DNA binding domains to enhance glucocorticoid responsive genes with estradiol. Because of the high analogy between for example these two receptors, especially in the DBD, cDNA libraries were screened for similarity with the known DBD. This quickly led
6 NUCLEAR RECEPTOR FUNCTION AND MODULATION to the discovery of the retinoid acid receptor in 198750 and the retinoid X receptor in 1990.51 The complete sequencing of the human genome in the early 2000s, allowed the definition of the complete set of receptors and 48 nuclear receptors were discovered.52 The x-ray crystal structure of the retinoid X receptor-α (RXRα) LBD was published in 1995 by Moras and coworkers.53 More detailed information about ligand binding and receptor function was gained in the same year by the publication of the crystal structure of the retinoid acid receptor (RAR) LBD bound to all-trans retinoic acid54 and the thyroid hormone receptor (TR).55 In the following six years the structures of 16 nuclear receptor LBDs were solved providing structural information about this protein family. Many of these crystal structures were solved in complexes with agonistic or antagonistic ligands as well as fragments of coregulatory proteins. With many of these structures solved, it was now possible to structurally design ligands and to accelerate the possibilities in targeting these receptors, while biologists discovered more and more relevant roles in diseases for nuclear receptors (Table 1.1).
Table 1.1 | Nuclear receptor endogenous ligands, functions and associated diseases56,57
Abbreviation Full name Endogenous ligand Function / Associated disease TRα, β Thyroid hormone L-3,5,3’- Metabolism and heart rate / receptor triiodothyronine Hypothyroidism, obesity RARα, β, γ Retinoic acid receptor All-trans retinoic acid Development, immune function, vision / Inflammation skin disorders, leukemia PPARα, β, γ Peroxisome Orphan (unsaturated Cell differentiation, proliferated-activated fatty acids58) metabolism / Diabetes, receptor coronary heart disease, obesity RORα, β, γ Retinoic acid related Orphan (cholesterol, Circadian rhythm, immune orphan receptor hydroxy -cholesterols59) response / Atherosclerosis, immunological disorders LXRα, β Liver X receptor Oxysterols Glucose homeostasis / Atherosclerosis FXR Farnesoid X receptor Endogenous bile Cholesterol homeostasis / acids60 Dyslipidemia, liver disease
7 CHAPTER 1
VDR Vitamin D receptor 1,25-dihydroxy vitamin Calcium homeostasis /
D3 Osteoporosis, cancer PXR Pregnane X receptor Progesterone, 17α- Xenobiotic metabolism / hydroxyprogesterone Motion sickness, vertigo CAR Constitutive Androstenol, Xenobiotic metabolism / n.d. androstane receptor androstanes HNF4α, γ Hepatocyte nuclear Linoleic acid61 Lipid metabolism / Diabetes, factor 4 hemophilia RXRα, β, γ Retinoid X receptor Fatty acids, 9-cis-13,14- Metabolism / Leukemia, dihydroretinoic acid,62 coronary heart disease 9-cis-retinoic acid* ERα, β Estrogen receptor 17-β estradiol Development / Breast cancer, osteoporosis, atherosclerosis ERRα, β, γ Estrogen-related Orphan Early embryo development / receptor Type 2 diabetes, cancer GR Glucocorticoid Cortisol Immune response, receptor metabolism / Immunological and metabolic disorders MR Mineralocorticoid Aldosterone Salt and water balance / receptor Hypertension, myocardial hypertrophy PR Progesterone Progesterone Pregnancy maintenance / receptor Breast cancer, infertility AR Androgen receptor Testosterone Development / Prostate cancer, spinal and muscular atrophy NGFI-B Nerve growth factor- No ligand binding Development / Neurological induced-B pocket and immunological disorders Nurr1 Nuclear receptor No ligand binding Maintenance of related 1 pocket dopaminergic system / Parkinson's disease, schizophrenia, and manic depression
8 NUCLEAR RECEPTOR FUNCTION AND MODULATION
GCNF Germ cell nuclear Orphan Fertility / Contraception factor Rev-ErbAα, β Orphan (heme63) Development and circadian regulation / n.d. Bold: nuclear receptors with marketed drugs, in brackets endogenous ligands for orphan receptors, not always confirmed. n.d.: not determined. * Whether 9-cis-retinoic acid is the putative endogenous RXR ligand is a highly controversial topic in retinoid research.
The importance of nuclear receptor research over the years is reflected by the number of diseases associated with them and the amount of marketed drugs. Prior to there being detailed knowledge about the nuclear receptor family of proteins, the first drugs targeting them were already in extensive use. Experiments with natural extracts containing steroids or thyroids were the basis for discovering many new biologically active compounds, and formed the basis for modern day nuclear receptor based endocrinology.64 For example, glucocorticoid receptor drug discovery was initiated by adrenal gland extracts, which were used for glucocorticoid deficiency, but also inflammatory conditions. Cortisone was soon identified as an active steroidal compound in these extracts. Purification and later the total synthesis of cortisone acetate made testing the effects on inflammatory diseases possible in the early 1950s. Supported by these early successes, further research into steroidal glucocorticoids was conducted65 leading to drugs such as prednisolone (inflammatory, auto-immune disease and liver failure) or dexamethasone (rheumatic problems).66 Another example of a first generation drug targeting a nuclear receptor was also found from natural sources. The estrogenic hormone estrone was isolated in 1929 from the urine of pregnant women, and the structure was elucidated five years later. Four years later estradiol was synthesized from cholesterol, which was then improved to ethinylestradiol in the same year. Ethinylestradiol had an improved bioavailability and became a hallmark in modern female fertility control, and is even today a major component in many oral contraceptives. Besides contraception, estrogen research played an important role for other indications. Experiments with anti-estrogenic compounds showed beneficial biological effects in estrogen-responsive breast cancer. Screens with non-steroidal like compounds led to the discovery of for example tamoxifen, which was the first chemo-preventive agent and still is the first line of treatment in breast cancer today. Since 1958, nuclear receptor research has progressed into a high value research area with approximately 13% of all drugs approved for sale targeting nuclear receptors, and 15 nuclear receptor drugs in the top 200 prescribes medicines. The major indications for these
9 CHAPTER 1 dugs include diabetes, contraception, hormone replacement therapy, autoimmune diseases and cancer. Nevertheless, there remains a demand for alternatives in the modulation of nuclear receptors.
Nuclear receptor function and modulation
Figure 1.3 | Overview of the nuclear receptor signaling pathway. The cellular function is amended through external triggers, such as hormones or drugs. Figure modified from: Molecular biology of the cell.67 NR: nuclear receptor, HSP: heat shock protein. After diffusion of a small ligand through the cytoplasmic membrane, the ligand binds to its receptor. The change in conformation affects the loss of co-repressor proteins, dimerization and translocation to the nucleus. Co-activators are recruited and transcription takes place, altering cellular function and processes.
Only five years after the cloning of the glucocorticoid receptor a reasonable understanding of nuclear receptor functioning was established. By then, the crystal structure of many unliganded (apo) as well as liganded (holo) ligand binding domains of nuclear receptors had been solved. This provided a good understanding of ligand binding and transactivation. Small ligands for nuclear receptors, such as endogenous hormones, endocrine disruptors or therapeutic drugs are able to pass through the cytosolic membrane into the cell (figure 1.3). The small hydrophobic molecule can then bind to the ligand binding domain of the nuclear receptor (also known as the AF2) in the cytoplasm. The first signaling event in nuclear receptor modulation is thus the ligand binding to the nuclear
10 NUCLEAR RECEPTOR FUNCTION AND MODULATION receptor. Nuclear receptor research from a therapeutic viewpoint focuses significantly on the ligands and already since early 20th century extracts from adrenal glands have been used in the treatment of Addison’s disease based on glucocorticoid deficiencies. This led to the total synthesis of several steroid compounds based on the endogenous nuclear receptor ligands; see some examples in figure 1.4. In later screening efforts also non-steroidal compounds were discovered.
Figure 1.4 | Examples of endogenous ligand targeting the ligand binding pocket of nuclear receptors. From left to right agonistic ligands for the: ER, AR, PR, RXR and ROR.
The ligand binding domain consists out of twelve helices, which form a compact structure comprising a hydrophobic ligand binding pocket. The 12th helix (H12) forms a crucial and moveable “lid” over the pocket, which determines the transactivation properties of the protein. The orientation of H12 is determined by allosteric effects induced through binding of specific ligands to the pocket. In the apo conformation the nuclear receptor is typically bound to a set of corepressors and heat-shock proteins. The corepressor docks to a hydrophobic groove on the surface of the ligand binding domain between H3 and H4 (figure 1.5). Upon ligand binding H12 undergoes a change in conformation, or goes from a disordered state to an ordered state. This elicits allosteric effects that disrupt the hydrophobic groove and dissociation of corepressors and heat shock proteins.68,69 The dissociation of repressor proteins can lead to dimerization of the nuclear receptor, homodimers (type I) and heterodimers (type II). Type II nuclear receptors are retained in the nucleus, while type I typically translocates to the nucleus to exert their function after ligand binding. The repositioning or folding of H12 creates a hydrophobic groove on the AF2 domain, which allows the recruitment of coactivator proteins followed by the transcription machinery. The coactivator interacts via an LXXLL consensus motif,70,71 where L is leucine and X any other naturally occurring amino acid (figure 1.5).
11 CHAPTER 1
Figure 1.5 | Three relevant conformational states known to nuclear receptors. The canonical fold of the nuclear receptor ligand binding domain consists of 12 helices. In the apo-state the coactivator is typically not bound to the ligand binding domain (PDB: 1PRG).72 With the PPAR agonist aleglitazar bound in the ligand binding pocket, helix 12 adopts the optimal conformation for coactivator (blue) recruitment (PDB: 3G8I).73 The antagonist GW6471 prevents the folding of the optimal conformation of helix 12. This results in a larger hydrophobic cleft where the corepressor (silencing mediator for retinoid hormone receptors – SMRT) with the consensus motif LXXXIXXXL can bind (PDB: 1KKQ).74
The many crystal structures of nuclear receptor ligand binding domains with their ligands reveal a common binding mode for ligand binding. This suggests a set of general guidelines for the design of high affinity agonists and antagonists. Nuclear receptor ligands tend to be hydrophobic molecules, because there is a relative lack of polar residues in the binding pocket. Hydrogen bonding or electrostatic interactions are therefore frequently not necessary for tight binding, except in positions proximal to the end of the ligand.57 The structural information and our understanding of the requirements for nuclear receptor ligands have resulted in the generation of a considerable amount of synthetic compounds ranging from agonist, antagonists and inverse agonists. Recruitment of coactivators in general enhances gene transcription and ligands that facilitate this interaction are typically agonists. Ligands that promote corepressor complexation and in this way decrease gene transcription are named inverse agonists. Ligands that disfavor coactivator recruitment through conformational changes in the receptor, are also called inverse agonists if the nuclear receptor has a constitutive level of transcriptional activity. The degree of complexation or perturbation of the coregulator complex may differ from ligand to ligand.
12 NUCLEAR RECEPTOR FUNCTION AND MODULATION
Ligands that induce graded responses to this effect are named partial agonists or partial inverse agonists. Antagonist ligands intrinsically do not affect transcriptional activity, but instead block the effect of the agonist (see figure 1.6).75
Figure 1.6 | Graphical representation of agonistic, antagonistic and inverse agonistic response for nuclear receptor coactivator recruitment. Coactivator recruitment is correlated with downstream translation and transcription. An agonistic ligand increases the response, while an inverse agonist decreases the response. Antagonist ligands do not affect transcriptional activity, but instead block or reduce the response induced by the receptor agonist.75 Partial effects give responses less than 100%. Figure adapted and modified from Fauber and Magnuson (2014).75
When comparing agonistic and antagonistic ligands of different nuclear receptor their structures tend to share a number of common features. Antagonists can be engineered from agonists by “building out” from the core of the molecule (figure 1.7). In this case, the antagonistic property of these molecules arises from the disruption of the normal helix 12 packing through steric blockade. The extension itself has influence on the binding properties of the molecules because of extra interactions with the receptor when the extension is appropriately placed. The use of a known (endogenous) nuclear receptor agonist as scaffold structure for the development of antagonists is a general viable strategy, which may lead to the discovery of selective antagonists for numerous nuclear receptors that have yet to be assigned an antagonist ligand.
13 CHAPTER 1
Figure 1.7 | Nuclear receptor agonists versus antagonists. Comparison of nuclear receptor agonists and antagonists. The antagonists have the general structure of the agonists, plus a large extension protruding often from the center of the molecule which can disrupt the folding of helix 12 in the agonistic conformation. Top row are agonists for mentioned nuclear receptors, bottom row corresponding antagonists. Examples taken from Weatherman and coworkers57 and Nahoum et al.76
Beyond the dose response curves
Besides classical agonists and antagonists targeting the canonical ligand binding pocket there are efforts to find alternatives for modulation of the transcriptional activity of nuclear receptors. For example, cooperative binding of two ligands in the ligand binding pocket has been observed in two separate studies. The natural product magnolol was shown to bind cooperatively two times in the Y-shaped pocket of PPARγ inducing an agonistic conformation.77 This cooperativity was used in a later study to design hybrid ligands starting from a screen of combinations of ligands.78 A separate study has demonstrated a fascinating synergistic effect between two environmental factors on PXR activation.79 A pharmaceutical estrogen and an organochloride pesticide both exhibited low efficacy when studied separately. Cooperative binding of the ligands showed synergistic activation and analysis demonstrated that each ligand enhances the binding of the other.79 These examples are named ‘supramolecular ligands’ and may have broader implications in the field of toxicology. Other examples of modulation can be found in the covalent linking of ligands to the receptor.78,80 The alternatives for nuclear receptor modulation mentioned above are examples of ligands still targeting the ligand binding pockets of receptors. With our increasing overall understanding of protein-protein interactions combined with our knowledge of the multiple molecular components associated with nuclear receptor- dependent gene regulation , research has been focusing on the development of small molecule modulators of nuclear receptors acting at sites other than the ligand binding
14 NUCLEAR RECEPTOR FUNCTION AND MODULATION pocket. By looking at interaction sites, such as the coregulatory binding groove, several examples of small molecule coactivator inhibitors have been found for the ER, TR, AR and RXR.81–84 Despite these considerable efforts, following these promising strategies, what is seriously lacking for the nuclear receptor alternate-site modulators is low nanomolar potencies. The design of small molecules targeting the coregulator groove is based on the LXXLL motif found in the coregulators.85,86 Besides small molecules, peptide-based inhibitors were discovered via phage display. This is a powerful general technique to identify optimal peptide binders for specific target nuclear receptors, and was successfully applied to the ER, AR PPAR and MR.71,87–90 Despite the perfect match for these peptides their pharmaceutical utility is limited owing to their low cell permeability and proteolytic stability. To circumvent these challenges and to stabilize the conformation of the α-helix, macrocyclic peptide fragments were developed. Lactam bridges, peptide hydrocarbon staples and disulfide bridging proved to be viable strategies for the stabilization of peptide conformations and inhibition of proteolytic degradation.91 The transcriptional coregulator proteins (co-activators and corepressors) themselves have also gained a degree of attention since the discovery of the steroid receptor coativator- 1 (SRC-1).38,92 The advancement of molecular technologies has allowed better characterization of binding events in the context of protein-protein interactions and gene regulation and therefore the role of regulator proteins in diseases. It also provides opportunities to discover new strategies for therapeutic interventions, such as targeting the coactivator directly,93 or targeting the enzymes that induce posttranslational modifications on the coactivators.94 Because the coregulators are involved in downstream signaling, they can provide viable targets in drug discovery. Many regulator proteins are tissue selective, giving opportunities for selective treatments.
Mechanisms for signal transduction in nuclear receptors
The mechanisms for allosteric signaling between ligand binding and coregulator binding sites has gained interest of late,95 but remain poorly understood. The sophisticated control of nuclear receptor function is modulated by a plethora of factors including ligand binding, tissue specific coregulators, posttranslational modifications, DNA recognition or selective dimerization partners (figure 1.8). To understand the complex interplay of all these factors the individual domains are studied, providing more understanding of the energy landscape of different nuclear receptor conformations.96,97
15 CHAPTER 1
Figure 1.8 | Artistic impression of two-dimensional energy landscape for nuclear receptors. a) Nuclear receptors might sample a near native state (N*) in the apo conformation, ligand binding and formation of the complex might stabilize the nuclear receptor in a more favorable energy state. b) Interactions with partners – dimerization or coregulators – might lead to stabilization of different energy states. c) Energy barriers have impact on receptor function; barriers may be reduced by environmental factors, localization, protein-protein interactions or posttranslational modifications. Thermodynamically unfavorable states (M), may be long lived because of high barriers to favorable states (N). Figure modified from Gershenson and coworkers.96
RXR forms dimers with a subset of nuclear receptors and therefore provides a useful platform to study the allosteric signaling pathways of ligands on selective dimerization. The allosterics between RXR and partner are not well understood, but especially structural studies facilitate understanding. For example the RXR agonists 9-cis-retinoic acid and LG100268 were shown to stimulate the formation of the RXRα/PPARγ dimer, while the same agonists had no effect on the formation of VDR/RXR or TR/RXR dimers.98 Another example, which makes use of our increasing understanding of selective dimerization of nuclear receptors8,99 is the activation of the Nurr1 receptor via ligand controlled Nurr1-RXR heterodimerization.100,101 The Nurr1 ligand binding pocket, in contrast to most nuclear receptors, adopts a closed conformation and is therefore difficult to target directly with small molecules. This receptor is interesting because of its role in the proper development and survival of dopaminergic neurons according to recent work reported by Jankovic et al.102 Their proof-of-concept study may lead to the treatment of neurodegenerative diseases via this allosteric signaling approach. Studies on several RXR heterodimers have shed light on the allosteric signaling and DNA binding, but is also the subject of discussion and have not yet produced definitive rules for ligand dependent signaling pathways.103–105
16 NUCLEAR RECEPTOR FUNCTION AND MODULATION
The literature describes a number of ligand dependent mechanisms: the so-called ‘mouse trap’ model, the allosteric coregulator binding mechanism, but also association and dissociation mechanisms of ligand binding. The mouse trap model is based on early crystallization efforts and homology models of ligand binding domains of RXR and RAR. Ligand binding was suggested to induce an altered position of Helix-12. In the apo-state Helix-12 was described as a stable helix, which upon agonist binding repositions itself on the surface of the receptor thereby creating a cleft for coregulator binding, and entrapping an active conformation.97 The model suggests an on / off switch, but an alternative for this model was suggested by Schwalbe and coworkers.69,106 Fluorescence spectroscopy revealed high mobility of Helix-12, but especially a lack of structural order in the apo-state of nuclear receptors. This was supported by NMR studies for PPAR and RXR.107,108 Furthermore, the crystal structures with the quaternary architectures for RXRα-PPARγ, HNF-4α and RXR- LXR complexed on DNA showed very unique features, rather than being commonly shared. These crystallization efforts combined with EM, SAXS, NMR and biochemical techniques such as H/D exchange MS or FRET might provide more detail and insights into the molecular mechanisms of signal transduction in the nuclear receptor full length biological setting. So far, small-molecule screening and modulation efforts have been performed on isolated ligand-binding domains alone, but future work might thus focus more on this concept of allosteric signal transduction and in this way address more the key physiological questions. To expand on the single domain studies and to elucidate the collaboration of different domains in allosteric signaling, molecular dynamic simulations on the quaternary structures were performed to shed light on the role of dynamics under more physiological conditions.109 Analysis showed that distant mobile regions of the complex undergo large independent motions. For example the DNA-binding domain can undergo translations relative to the ligand binding domain, and these rotations are induced by the DNA, which supports the hypothesis that the DNA is an active player in transcriptional activation.110,111 The notion that strongly correlated regions of proteins designed for transcription are not spatially proximal and are spaced by structurally ordered domains demonstrates that the mechanism with which allosteric information is transmitted through the protein must be very complex and probably arises through cooperative effects and a network of motions in the structure.
17 CHAPTER 1
Summary points
1. Nuclear receptors are ideal drug development targets given that they are regulated by small ligands, which bind in a defined and hydrophobic pocket, and in response regulate genes involved in many processes and associated diseases. 2. Crystal structures have been solved for a large number of nuclear receptors; a large portion of this work is starting to coalesce into a general scheme. This type of information has guided our notions of how ligands can impact the receptor conformation and have accelerated ligand design and the search for alternative modulation. 3. Our understanding of the effects of posttranslational modifications, mutations and protein-protein interactions on the stabilization of nuclear receptor conformations could assist the search for new selective modulators of nuclear receptors. 4. The major goal of nuclear receptor drug discovery is to identify molecules that act in a receptor- and gene-selective fashion and do not regulate all receptor genes in the same way. This might be obtained by taking advantage of selective dimerization or selective coregulatory recruitment via partial agonism. 5. Transcriptional coregulators have emerged as regulators of gene transcription through interaction with nuclear receptors and thereby modulating their activity, they are often overexpressed and amplified in malignancy. Directly targeting the protein-protein interaction with nuclear receptors might prove a viable strategy in drug development. 6. Taking into account allosteric signaling events for nuclear receptors, our assumptions might yet be simplistic. Multidisciplinary approaches and combining biophysical and biochemical techniques can provide insights into the molecular mechanisms of signal transduction and how and when nuclear receptors form their dimerization and DNA complexes.
Aim and outline of this thesis
As described in this introductory chapter, nuclear receptor modulation is an important theme in chemical biology, chemistry and drug discovery. Nuclear receptors are involved in almost all aspects of the human physiology and when misregulated are associated with major diseases. The therapeutic potential for nuclear receptor targeting drugs is enormous, and reinforced by the fact that cancer, obesity and neurodegenerative diseases are
18 NUCLEAR RECEPTOR FUNCTION AND MODULATION becoming major health issues, which are all frequently driven by malignant nuclear receptor signaling. Many successes have been accomplished in the field, which is reflected by the large number of marketed drugs targeting nuclear receptors. Nevertheless, there is a need for alternatives in nuclear receptor modulation; building on and expanding the existing approaches. Resistance, undesired side effects, and a lack of selectivity or efficacy are the major challenges. This thesis focuses on molecular modulation of two nuclear receptors. Chapters two to four are focused on modulation of the retinoid X receptor (RXR). RXR is a very challenging member of the nuclear receptor family because it forms heterodimers with many other nuclear receptors. Chapter five focuses on the RAR-related orphan receptor gamma-t (RORγt), RORγt biology continues to evolve rapidly, which is followed by a significant effort to discover modulators for this receptor. The aim of the work described in this thesis was to study the effect of small ligands of diverse chemical, structural and functional nature on nuclear receptor conformations in relationship to protein protein interactions. The described research offers comprehensive insights in nuclear receptor modulation. It utilizes viable strategies which can be applied in nuclear receptor research combining synthetic organic chemistry, biochemical evaluation, structural biology and cellular assays. Chapter two describes the investigations into the stabilization of one type of RXR conformation, with two functional outcomes via the dual binding mode of the natural product honokiol. Natural products have been an inspiration for new molecular scaffolds for a long time. The extracts from the bark of the Magnolia officinalis contains active compounds that have been used in eastern medicine against several disorders. Investigations showed that honokiol is one of the active substrates. We used NMR spectroscopy, biochemical assays and modeling to identify the dual binding mode of honokiol to RXR. Via chemical synthesis, cellular screening and X-ray crystallography the dual binding mode of honokiol was dissected into two distinct modes of action. This led on the one hand to highly potent and molecular efficient agonists for RXR and on the other hand a first-of-kind cofactor inhibitor for RXR. These findings justify the exploration of natural products for more difficult-to-address protein-protein interactions. Chapter three focuses on the modulation of RXR and builds on the molecular scaffold described in chapter two, to modulate the receptors’ surface characteristics via controlled, small molecular modifications to the ligand. This chapter describes the synthesis, biochemical evaluation and structural elucidation of RXR ligands with rational modifications targeting selectively the dimerization properties of RXR or the AF2 site. The data provide a rationale
19 CHAPTER 1 for the design of RXR ligands, built out of a unique hydrophilic region with a conserved hydrogen bonding network and an interchangeable hydrophobic region influencing the conformation of the protein. These results reveal novel molecular entries and underlying molecular mechanism into the modulation of RXR conformation and justify further explorations into their effects on the dimerization properties of RXR. Chapter four deals with the synthesis and biochemical evaluation of a new Nuclear Receptor chemotype, a bis- benzanulated spiroketal structure as an agonist for RXR. Spiroketals are a recurring motif in biologically active natural compounds, with a very special and structured molecular backbone. Despite their biological relevance, a precise structure-activity relationship has rarely been established, with only one protein co-crystal structure reported. Furthermore, to date no studies on spiroketals as molecular scaffold for nuclear receptor ligands are reported. The project commenced with a docking study of the desired “bent” conformation of the spiroketal in the L-shaped ligand binding pocket of RXR. The synthesis involved 15 steps with the longest linear sequence of 5 steps; biochemical evaluation demonstrated an agonistic behavior for the spiroketal. X-ray data confirmed the occupancy of the ligand binding pocket, coherent with the initial docking study. This study demonstrates the potential of the spiroketal structure as ligand for nuclear receptor modulation. Chapter five is the report on the discovery of a novel drug target site for Nuclear Receptors; an allosteric binding pocket in the ligand binding domain of RORγt. RORγt is critical for the differentiation and proliferation of T-helper17 cells associated with chronic autoimmune diseases. Co-crystallization of the RORγt ligand binding domain with a series of small- molecule inverse agonists demonstrates the occupancy of a previously unreported allosteric site and stabilization of an unprecedented Nuclear Receptor conformation. Helix 12 adopts an unprecedented conformation, which results in functional inhibition as shown by biochemical and cellular studies. This brings forward an approach to target Th17-mediated chronic autoimmune diseases and establishes an alternative mechanism for nuclear receptor conformational modulation. The epilogue looks forward and describes the opportunities to expand on the work described in this thesis. A recurring theme in RXR research, is the dimerization properties of RXR with other nuclear receptors. Targeting selective dimerization or selective coregulators the full therapeutic potential of RXR could be investigated. The discovery of the allosteric pocket in RORγt established an unprecedented modality of pharmacological antagonism which can explored further in drug development.
20 NUCLEAR RECEPTOR FUNCTION AND MODULATION
References
1. Overington, J. P., Al-Lazikani, B. & Hopkins, A. L. How many drug targets are there? Nat. Rev. Drug Discov. 5, 993–996 (2006). 2. Moore, J. T., Collins, J. L. & Pearce, K. H. The Nuclear Receptor Superfamily and Drug Discovery. ChemMedChem 1, 504–523 (2006). 3. Lander, E. S. et al. Initial sequencing and analysis of the human genome. Nature 409, 860–921 (2001). 4. Venter, J. C. et al. The sequence of the human genome. Science 291, 1304–1351 (2001). 5. Nuclear Receptors Nomenclature Committee. A unified nomenclature system for the nuclear receptor superfamily. Cell 97, 161–163 (1999). 6. Kumar, V. et al. Functional domains of the human estrogen receptor. Cell 51, 941–951 (1987). 7. Peters, G. A. & Khan, S. A. Estrogen receptor domains E and F: role in dimerization and interaction with coactivator RIP-140. Mol. Endocrinol. Baltim. Md 13, 286–296 (1999). 8. Chandra, V. et al. Structure of the intact PPAR-gamma-RXR- nuclear receptor complex on DNA. Nature 456, 350–356 (2008). 9. McEwan, I. J., Lavery, D., Fischer, K. & Watt, K. Natural disordered sequences in the amino terminal domain of nuclear receptors: lessons from the androgen and glucocorticoid receptors. Nucl. Recept. Signal. 5, e001 (2007). 10.Kussie, P. H. et al. Structure of the MDM2 oncoprotein bound to the p53 tumor suppressor transactivation domain. Science 274, 948–953 (1996). 11. Uesugi, M., Nyanguile, O., Lu, H., Levine, A. J. & Verdine, G. L. Induced alpha helix in the VP16 activation domain upon binding to a human TAF. Science 277, 1310–1313 (1997). 12. Bain, D. L., Heneghan, A. F., Connaghan-Jones, K. D. & Miura, M. T. Nuclear receptor structure: implications for function. Annu. Rev. Physiol. 69, 201–220 (2007). 13. Simons, S. S., Edwards, D. P. & Kumar, R. Minireview: dynamic structures of nuclear hormone receptors: new promises and challenges. Mol. Endocrinol. Baltim. Md 28, 173–182 (2014). 14. Pan, T., Freedman, L. P. & Coleman, J. E. Cadmium-113 NMR studies of the DNA binding domain of the mammalian glucocorticoid receptor. Biochemistry (Mosc.) 29, 9218–9225 (1990). 15. Freedman, L. P. et al. The function and structure of the metal coordination sites within the glucocorticoid receptor DNA binding domain. Nature 334, 543–546 (1988). 16.Khorasanizadeh, S. & Rastinejad, F. Nuclear-receptor interactions on DNA-response elements. Trends Biochem. Sci. 26, 384–390 (2001). 17. Rastinejad, F., Huang, P., Chandra, V. & Khorasanizadeh, S. Understanding nuclear receptor form and function using structural biology. J. Mol. Endocrinol. 51, T1–T21 (2013). 18. Lavery, D. N. & McEwan, I. J. Structure and function of steroid receptor AF1 transactivation domains: induction of active conformations. Biochem. J. 391, 449–464 (2005). 19.Clinckemalie, L., Vanderschueren, D., Boonen, S. & Claessens, F. The hinge region in androgen receptor control. Mol. Cell. Endocrinol. 358, 1–8 (2012). 20.Helsen, C. et al. Structural basis for nuclear hormone receptor DNA binding. Mol. Cell. Endocrinol. 348, 411–417 (2012). 21. Tanner, T. M. et al. A 629RKLKK633 motif in the hinge region controls the androgen receptor at multiple levels. Cell. Mol. Life Sci. CMLS 67, 1919–1927 (2010). 22.Daniel, A. R. et al. The progesterone receptor hinge region regulates the kinetics of transcriptional responses through acetylation, phosphorylation, and nuclear retention. Mol. Endocrinol. Baltim. Md 24, 2126–2138 (2010). 23.Skafar, D. F. & Zhao, C. The multifunctional estrogen receptor-alpha F domain. Endocrine 33, 1– 8 (2008).
21 CHAPTER 1
24.Yang, J., Singleton, D. W., Shaughnessy, E. A. & Khan, S. A. The F-domain of estrogen receptor- alpha inhibits ligand induced receptor dimerization. Mol. Cell. Endocrinol. 295, 94–100 (2008). 25.De Vries-van Leeuwen, I. J. et al. Interaction of 14-3-3 proteins with the estrogen receptor alpha F domain provides a drug target interface. Proc. Natl. Acad. Sci. U. S. A. 110, 8894–8899 (2013). 26.Tharun, I. M. et al. Subtype-specific modulation of estrogen receptor-coactivator interaction by phosphorylation. ACS Chem. Biol. 10, 475–484 (2015). 27.Le Romancer, M. et al. Cracking the estrogen receptor’s posttranslational code in breast tumors. Endocr. Rev. 32, 597–622 (2011). 28.McDonnell, D. P. & Norris, J. D. Connections and regulation of the human estrogen receptor. Science 296, 1642–1644 (2002). 29.Zhang, X., Huang, Y. & Shi, X. Emerging roles of lysine methylation on non-histone proteins. Cell. Mol. Life Sci. 72, 4257–4272 (2015). 30.Zwart, W. et al. PKA-induced resistance to tamoxifen is associated with an altered orientation of ERα towards co-activator SRC-1. EMBO J. 26, 3534–3544 (2007). 31. Droog, M., Beelen, K., Linn, S. & Zwart, W. Tamoxifen resistance: from bench to bedside. Eur. J. Pharmacol. 717, 47–57 (2013). 32.Skliris, G. P. et al. A phosphorylation code for oestrogen receptor-alpha predicts clinical outcome to endocrine therapy in breast cancer. Endocr. Relat. Cancer 17, 589–597 (2010). 33. Generali, D. et al. Phosphorylated ERalpha, HIF-1alpha, and MAPK signaling as predictors of primary endocrine treatment response and resistance in patients with breast cancer. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 27, 227–234 (2009). 34.Cai, C. et al. Lysine-specific demethylase 1 has dual functions as a major regulator of androgen receptor transcriptional activity. Cell Rep. 9, 1618–1627 (2014). 35. Naito, S. & Shiota, M. [The concept and mechanisms of castration-resistant prostate cancer]. Nihon Rinsho Jpn. J. Clin. Med. 72, 2090–2094 (2014). 36.Kadiyala, V. & Smith, C. L. Minireview: The versatile roles of lysine deacetylases in steroid receptor signaling. Mol. Endocrinol. Baltim. Md 28, 607–621 (2014). 37. Choi, S.-S., Park, J. & Choi, J. H. Revisiting PPARγ as a target for the treatment of metabolic disorders. BMB Rep. 47, 599–608 (2014). 38.Dasgupta, S., Lonard, D. M. & O’Malley, B. W. Nuclear receptor coactivators: master regulators of human health and disease. Annu. Rev. Med. 65, 279–292 (2014). 39.Smith, C. L. & O’Malley, B. W. Coregulator Function: A Key to Understanding Tissue Specificity of Selective Receptor Modulators. Endocr. Rev. 25, 45–71 (2004). 40.Willson, T. M. & Moore, J. T. Minireview: Genomics Versus Orphan Nuclear Receptors—A Half-Time Report. Mol. Endocrinol. 16, 1135–1144 (2002). 41. Jensen, E. V., Jacobson, H. I., Walf, A. A. & Frye, C. A. Estrogen action: a historic perspective on the implications of considering alternative approaches. Physiol. Behav. 99, 151–162 (2010). 42.Jensen, E. V. et al. A two-step mechanism for the interaction of estradiol with rat uterus. Proc. Natl. Acad. Sci. U. S. A. 59, 632–638 (1968). 43.Fang, S., Anderson, K. M. & Liao, S. Receptor proteins for androgens. On the role of specific proteins in selective retention of 17-beta-hydroxy-5-alpha-androstan-3-one by rat ventral prostate in vivo and in vitro. J. Biol. Chem. 244, 6584–6595 (1969). 44.O’Malley, B. W., Sherman, M. R. & Toft, D. O. Progesterone ‘receptors’ in the cytoplasm and nucleus of chick oviduct target tissue. Proc. Natl. Acad. Sci. U. S. A. 67, 501–508 (1970). 45.Wrange, O. & Gustafsson, J. A. Separation of the hormone- and DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. J. Biol. Chem. 253, 856–865 (1978).
22 NUCLEAR RECEPTOR FUNCTION AND MODULATION
46.Ringold, G. M., Yamamoto, K. R., Bishop, J. M. & Varmus, H. E. Glucocorticoid-stimulated accumulation of mouse mammary tumor virus RNA: increased rate of synthesis of viral RNA. Proc. Natl. Acad. Sci. U. S. A. 74, 2879–2883 (1977). 47.Gronemeyer, H. & Pongs, O. Localization of ecdysterone on polytene chromosomes of Drosophila melanogaster. Proc. Natl. Acad. Sci. U. S. A. 77, 2108–2112 (1980). 48.Hollenberg, S. M. et al. Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318, 635–641 (1985). 49.Green, S. et al. Human oestrogen receptor cDNA: sequence, expression and homology to v-erb- A. Nature 320, 134–139 (1986). 50.Giguere, V., Ong, E. S., Segui, P. & Evans, R. M. Identification of a receptor for the morphogen retinoic acid. Nature 330, 624–629 (1987). 51. Mangelsdorf, D. J., Ong, E. S., Dyck, J. A. & Evans, R. M. Nuclear receptor that identifies a novel retinoic acid response pathway. Nature 345, 224–229 (1990). 52.Robinson-Rechavi, M., Carpentier, A.-S., Duffraisse, M. & Laudet, V. How many nuclear hormone receptors are there in the human genome? Trends Genet. 17, 554–556 (2001). 53. Bourguet, W., Ruff, M., Chambon, P., Gronemeyer, H. & Moras, D. Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha. Nature 375, 377–382 (1995). 54.Renaud, J. P. et al. Crystal structure of the RAR-gamma ligand-binding domain bound to all- trans retinoic acid. Nature 378, 681–689 (1995). 55. Wagner, R. L. et al. A structural role for hormone in the thyroid hormone receptor. Nature 378, 690–697 (1995). 56.Sladek, F. M. Nuclear receptors as drug targets: new developments in coregulators, orphan receptors and major therapeutic areas. Expert Opin. Ther. Targets 7, 679–684 (2003). 57. Weatherman, R. V., Fletterick, R. J. & Scanlan, T. S. Nuclear-Receptor Ligands and Ligand- Binding Domains. Annu. Rev. Biochem. 68, 559–581 (1999). 58.Schupp, M. & Lazar, M. A. Endogenous ligands for nuclear receptors: digging deeper. J. Biol. Chem. 285, 40409–40415 (2010). 59.Jin, L. et al. Structural Basis for Hydroxycholesterols as Natural Ligands of Orphan Nuclear Receptor RORγ Mol. Endocrinol. 24, 923–929 (2010). 60.Wang, H., Chen, J., Hollister, K., Sowers, L. C. & Forman, B. M. Endogenous bile acids are ligands for the nuclear receptor FXR/BAR. Mol. Cell 3, 543–553 (1999). 61.Yuan, X. et al. Identification of an endogenous ligand bound to a native orphan nuclear receptor. PloS One 4, e5609 (2009). 62.de Lera, Á. R., Krezel, W. & Rühl, R. An Endogenous Mammalian Retinoid X Receptor Ligand, At Last! ChemMedChem 11, 1027–1037 (2016). 63.Raghuram, S. et al. Identification of heme as the ligand for the orphan nuclear receptors REV- ERBα and REV-ERBβ. Nat. Struct. Mol. Biol. 14, 1207–1213 (2007). 64.Schweitzer, A., Knauer, S. K. & Stauber, R. H. Therapeutic potential of nuclear receptors. Expert Opin. Ther. Pat. 18, 861–888 (2008). 65.Herzog, H. L. et al. New Antiarthritic Steroids. Science 121, 176–176 (1955). 66.van der Goes, M. C., Jacobs, J. W. & Bijlsma, J. W. The value of glucocorticoid co-therapy in different rheumatic diseases--positive and adverse effects. Arthritis Res. Ther. 16 Suppl 2, S2 (2014). 67.Alberts, B. et al. Molecular Biology of the Cell. (Garland Science, 2002). 68.Nagy, L. & Schwabe, J. W. R. Mechanism of the nuclear receptor molecular switch. Trends Biochem. Sci. 29, 317–324 (2004). 69.Oberoi, J. et al. Structural basis for the assembly of the SMRT/NCoR core transcriptional repression machinery. Nat. Struct. Mol. Biol. 18, 177–184 (2011).
23 CHAPTER 1
70.Heery, D. M., Kalkhoven, E., Hoare, S. & Parker, M. G. A signature motif in transcriptional co- activators mediates binding to nuclear receptors. Nature 387, 733–736 (1997). 71. Fuchs, S. et al. Proline Primed Helix Length as a Modulator of the Nuclear Receptor–Coactivator Interaction. J. Am. Chem. Soc. 135, 4364–4371 (2013). 72.Nolte, R. T. et al. Ligand binding and co-activator assembly of the peroxisome proliferator- activated receptor-gamma. Nature 395, 137–143 (1998). 73. Bénardeau, A. et al. Aleglitazar, a new, potent, and balanced dual PPARalpha/gamma agonist for the treatment of type II diabetes. Bioorg. Med. Chem. Lett. 19, 2468–2473 (2009). 74.Xu, H. E. et al. Structural basis for antagonist-mediated recruitment of nuclear co-repressors by PPARalpha. Nature 415, 813–817 (2002). 75. Fauber, B. P. & Magnuson, S. Modulators of the Nuclear Receptor Retinoic Acid Receptor- Related Orphan Receptor-γ (RORγ or RORc). J. Med. Chem. 57, 5871–5892 (2014). 76.Nahoum, V. et al. Modulators of the structural dynamics of the retinoid X receptor to reveal receptor function. Proc. Natl. Acad. Sci. U. S. A. 104, 17323–17328 (2007). 77.Zhang, H. et al. Molecular determinants of magnolol targeting both RXRα and PPARγ. PloS One 6, e28253 (2011). 78.Ohtera, A. et al. Identification of a New Type of Covalent PPARγ Agonist using a Ligand-Linking Strategy. ACS Chem. Biol. 10, 2794–2804 (2015). 79.Delfosse, V. et al. Synergistic activation of human pregnane X receptor by binary cocktails of pharmaceutical and environmental compounds. Nat. Commun. 6, 8089 (2015). 80.Egawa, D., Itoh, T. & Yamamoto, K. Characterization of covalent bond formation between PPARγ and oxo-fatty acids. Bioconjug. Chem. 26, 690–698 (2015). 81. Moore, T. W., Mayne, C. G. & Katzenellenbogen, J. A. Minireview: Not picking pockets: nuclear receptor alternate-site modulators (NRAMs). Mol. Endocrinol. Baltim. Md 24, 683–695 (2010). 82.Wang, Y. et al. A second binding site for hydroxytamoxifen within the coactivator-binding groove of estrogen receptor beta. Proc. Natl. Acad. Sci. U. S. A. 103, 9908–9911 (2006). 83.Estébanez-Perpiñá, E. et al. A surface on the androgen receptor that allosterically regulates coactivator binding. Proc. Natl. Acad. Sci. U. S. A. 104, 16074–16079 (2007). 84.Scheepstra, M. et al. A Natural-Product Switch for a Dynamic Protein Interface. Angew. Chem. Int. Ed. 53, 6443–6448 (2014). 85.Rodriguez-Marin, S., Murphy, N. S., Shepherd, H. J. & Wilson, A. J. Design, synthesis and conformational analyses of bifacial benzamide based foldamers. RSC Adv. 5, 104187–104192 (2015). 86.Wilson, A. J. Helix mimetics: Recent developments. Prog. Biophys. Mol. Biol. 119, 33–40 (2015). 87.Norris, J. D. et al. Peptide antagonists of the human estrogen receptor. Science 285, 744–746 (1999). 88.Hur, E. et al. Recognition and accommodation at the androgen receptor coactivator binding interface. PLoS Biol. 2, E274 (2004). 89.Mettu, N. B. et al. The nuclear receptor-coactivator interaction surface as a target for peptide antagonists of the peroxisome proliferator-activated receptors. Mol. Endocrinol. Baltim. Md 21, 2361–2377 (2007). 90.Yang, J. et al. Identification of ligand-selective peptide antagonists of the mineralocorticoid receptor using phage display. Mol. Endocrinol. Baltim. Md 25, 32–43 (2011). 91.Milroy, L.-G., Grossmann, T. N., Hennig, S., Brunsveld, L. & Ottmann, C. Modulators of protein- protein interactions. Chem. Rev. 114, 4695–4748 (2014). 92.Oñate, S. A., Tsai, S. Y., Tsai, M. J. & O’Malley, B. W. Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270, 1354–1357 (1995).
24 NUCLEAR RECEPTOR FUNCTION AND MODULATION
93.Wang, Y. et al. Small molecule inhibition of the steroid receptor coactivators, SRC-3 and SRC-1. Mol. Endocrinol. Baltim. Md 25, 2041–2053 (2011). 94.O’Malley, B. W. & Kumar, R. Nuclear receptor coregulators in cancer biology. Cancer Res. 69, 8217–8222 (2009). 95.Gronemeyer, H. & Miturski, R. Molecular mechanisms of retinoid action. Cell. Mol. Biol. Lett. 6, 3–52 (2001). 96.Gershenson, A., Gierasch, L. M., Pastore, A. & Radford, S. E. Energy landscapes of functional proteins are inherently risky. Nat. Chem. Biol. 10, 884–891 (2014). 97.Billas, I. & Moras, D. Allosteric controls of nuclear receptor function in the regulation of transcription. J. Mol. Biol. 425, 2317–2329 (2013). 98.Dawson, M. I. & Xia, Z. The retinoid X receptors and their ligands. Biochim. Biophys. Acta 1821, 21–56 (2012). 99.Kojetin, D. J. et al. Structural mechanism for signal transduction in RXR nuclear receptor heterodimers. Nat. Commun. 6, 8013 (2015). 100.McFarland, K. et al. Low dose bexarotene treatment rescues dopamine neurons and restores behavioral function in models of Parkinson’s disease. ACS Chem. Neurosci. 4, 1430–1438 (2013). 101.Sundén, H. et al. Chiral Dihydrobenzofuran Acids Show Potent Retinoid X Receptor-Nuclear Receptor Related 1 Protein Dimer Activation. J. Med. Chem. 59, 1232–1238 (2016). 102.Jankovic, J., Chen, S. & Le, W. D. The role of Nurr1 in the development of dopaminergic neurons and Parkinson’s disease. Prog. Neurobiol. 77, 128–138 (2005). 103.Rastinejad, F., Ollendorff, V. & Polikarpov, I. Nuclear receptor full-length architectures: confronting myth and illusion with high resolution. Trends Biochem. Sci. 40, 16–24 (2015). 104.Moras, D., Billas, I. M. L., Rochel, N. & Klaholz, B. P. Structure-function relationships in nuclear receptors: the facts. Trends Biochem. Sci. 40, 287–290 (2015). 105.Rastinejad, F., Ollendorff, V. & Polikarpov, I. Response to Moras et al. Trends Biochem. Sci. 40, 290–292 (2015). 106.Kallenberger, B. C., Love, J. D., Chatterjee, V. K. K. & Schwabe, J. W. R. A dynamic mechanism of nuclear receptor activation and its perturbation in a human disease. Nat. Struct. Biol. 10, 136– 140 (2003). 107.Johnson, B. A. et al. Ligand-induced stabilization of PPARgamma monitored by NMR spectroscopy: implications for nuclear receptor activation. J. Mol. Biol. 298, 187–194 (2000). 108.Lu, J., Cistola, D. P. & Li, E. Analysis of ligand binding and protein dynamics of human retinoid X receptor alpha ligand-binding domain by nuclear magnetic resonance. Biochemistry (Mosc.) 45, 1629–1639 (2006). 109.Ricci, C. G., Silveira, R. L., Rivalta, I., Batista, V. S. & Skaf, M. S. Allosteric Pathways in the PPARγ-RXRα nuclear receptor complex. Sci. Rep. 6, 19940 (2016). 110.Lefstin, J. A. & Yamamoto, K. R. Allosteric effects of DNA on transcriptional regulators. Nature 392, 885–888 (1998). 111.Hall, J. M., McDonnell, D. P. & Korach, K. S. Allosteric regulation of estrogen receptor structure, function, and coactivator recruitment by different estrogen response elements. Mol. Endocrinol. Baltim. Md 16, 469–486 (2002).
25
26 Chapter 2
A natural-product switch for a dynamic protein interface
Small ligands are a powerful way to control the function of protein complexes via dynamic binding interfaces. The classic example is found in gene transcription where small ligands regulate nuclear receptor binding to coactivator proteins via the dynamic activation function 2 (AF2) interface. Current ligands target the ligand-binding pocket side of the AF2. Few ligands are known, which selectively target the coactivator-side of the AF2, or which can be selectively switched from one side of the interface to the other. NMR and modelling were used to identify a natural product, which targets the retinoid X receptor (RXR) at both sides of the AF2. We then use chemical synthesis, cellular screening and X-ray co-crystallography to dissociate this dual activity, leading to a potent and molecularly efficient RXR agonist, and a first-of-kind inhibitor selective for the RXR/coactivator interaction. These findings justify future exploration of natural products at dynamic protein interfaces.
This work has been published: M. Scheepstra, L. Nieto Garrido, A.K.H. Hirsch, S. Fuchs, S. Leysen, C.V. Lam, L. in het Panhuis, C.A.A. van Boeckel, H. Wienk, R. Boelens, C. Ottmann, L.G. Milroy, & L. Brunsveld, A natural-product switch for a dynamic protein interface. Angew. Chem. Int. Ed Engl. 53, 6443–6448 (2014).
27 CHAPTER 2
Small ligands are a powerful way to control the function of large protein complexes via the selective modulation of dynamic binding interfaces.1,2 A classic example of this is seen in eukaryotic gene-transcription initiation, and the protein complex formed between nuclear receptors and coactivator proteins via the dynamic activation function 2 (AF2) binding interface.3 Ligand binding to a hydrophobic pocket located at the solvent- excluded side of the AF2 (Scheme 2.1), within the nuclear receptor ligand-binding domain,4 allosterically stabilizes or destabilizes coactivator protein binding at the opposite, solvent-exposed side of the interface, which in turn determines the transcription output. Ligand binding thus functions as a molecular switch, where stabilization or destabilization of the AF2 switches gene transcription either ‘on’ or ‘off’.5 Ligands targeting the nuclear receptor ligand-binding pocket continue to be an important source of drug molecules.6 However, issues of toxicity and drug resistance mean that ligands with atypical modes-of-action are in urgent demand.7–9 For instance, ligands targeting the ligand binding pocket but with atypical partial agonist/antagonist behavior – so-called selective nuclear receptor modulators ̶ are less toxic due to tissue- selective behavior.10 Alternatively, modified peptides derived from the binding epitopes of coactivator proteins or phage peptides selectively target the coactivator side of the AF2.11,12 Small non-peptidic ligands13–20 are arguably better suited than peptides as coactivator inhibitors due to their high ligand efficiency, metabolic stability and cell permeability, and some have shown promising NR-selective behavior.15,16,19 Natural products, though well investigated at the ligand-binding pocket, have been underexplored at the coactivator-side of the AF2, and are well suited for this purpose due to their biological relevance and the unique and diverse chemical space they populate.21
28 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
Scheme 2.1 | Regulation of the dynamic nuclear-receptor interface – the activation function 2 (AF2) – using small synthetic ligands derived from the same natural product (1): stabilization of the AF2 through binding at the ligand-binding pocket (2); destabilization of the AF2 through binding at the coactivator-side of the interface (3).
Herein, the development of two different ligand types, originating from the same natural product, targeting different sides of a dynamic interface, and with opposite stabilizing/destabilizing properties were reported. Co-factor recruitment screening and a combination of STD-NMR, molecular docking, and CORCEMA-ST calculations were used to show that honokiol (1, Scheme 2.1), targets both sides of the AF2 of the retinoid X receptor (RXR).22 A rational chemical-biology approach was then applied, with an efficient synthesis protocol at its core, to dissociate the dual-binding behavior of 1 and create a potent and molecularly-efficient RXR agonist and an atypical RXR-selective antagonist (2 & 3, Scheme 2.1). The RXR-coactivator interaction is important for the development of cancer,25 metabolic disorder,26 and Alzheimer’s disease.27 Current RXR ligands target the ligand- binding pocket, for which a rigid and bulky hydrocarbon-rich moiety is typically needed for potent binding (e.g., LG100268, 4, Figure 2.1b).28 Non-peptidic ligands targeting the coactivator side of the RXR AF2 are at present non-existent. For these reasons we became interested in the atypical RXR activity of honokiol (1, Scheme 2.1).29,30 Alongside related natural products isolated from the Magnolia tree bark,31 1 displays an array of biological properties, including neurite-growth induction and anti-angiogenic effects. It is likely
32–34 21,35 that 1 targets multiple proteins, in light of its fragment-like profile (MW = 266 Da), and because of the privileged nature of the biaryl structural motif.36 Importantly, 1 has
29 CHAPTER 2
also shown evidence of partial activity in a luciferase-based screen using U2OS cells overexpressing RXR.29 The RXR-activity of 1 was profiled alongside analogous natural and synthetic biaryl ligands using a fluorescence-based co-factor-recruitment assay (Figure 2.1a), where an increase in fluorescence signal would correspond with ligand binding at the RXR ligand binding pocket.37 In contrast to the other ligands tested, and contrary to our initial expectations, 1 inhibited coactivator-binding at 100 μM ligand concentration in the presence and absence of 100 μM of a potent full agonist, 4 (Figure 2.1a). At 100 μM, 4 fully saturates the ligand binding pocket, and thus excludes the binding of 1 at the ligand binding pocket. In support of this, a separate fluorescence polarization binding assay was performed, in which increasing concentrations of 4 predictably induced recruitment of a fluorescently labelled coactivator peptide (Figure 2.1b). Repeating the assay in the presence of increasing amounts of 1 resulted in a concentration-dependent decrease in
the maximum polarization signal, but without detectable changes in the EC50 for 4
(Figure 2.1b). The Ki for 1 was determined to be 94 μM ± 9 μM (experimental section). This atypical behaviour of 1 indicated a mode-of-binding distinct from the ligand binding pocket. Despite a growing number of non-peptidic coactivator inhibitor ligands,7 only a few are reported to be selective for one nuclear receptor over others,15,16,18 of which none are natural-products and none selective for RXR. Importantly, therefore, 1 was found to be selective for RXR over the estrogen and androgen receptors (ER and AR) in a fluorescence-based co-factor-recruitment assay (experimental section). Repeating the competitive fluorescence polarization assay in the presence of detergent did not alter the binding profile (data not shown),38 which, coupled with the inactivity of 1 towards ER and AR confirms the physiological significance of the interaction between 1 and RXR. In conclusion, 1 selectively inhibits the RXR-coactivator interaction in a physiologically significant manner and via an atypical mechanism, which is independent of the ligand- binding pocket. In-depth ligand-detected NMR studies were performed to further elucidate the RXR- binding mode of 1 (Figure 2.1c, experimental section). Severe line broadening of the 1H- resonances of 1 was observed in the presence of the RXR protein, which could be explained by the moderate binding affinity of the compound, as evidenced by our fluorescence polarization data, and the rapid ligand exchange. Line broadening of 1 was also observed in the presence of protein and an excess of potent ligands 2 and 4 (Figure 2.1c, experimental section). However, signal intensity and sharpening of 1 recovered
30 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE upon addition of a coactivator-derived peptide (Figure 2.1c), indicating competitive inhibition of 1 by the peptide. STD experiments on the system 1/RXR revealed the rapid exchange between free and bound states (Figure 2.1d). These data, combined with detailed competition STD and tr-NOESY NMR experiments (experimental section) suggest a novel binding mode for 1 at the coactivator side of the AF2. Moreover, theoretical CORCEMA-ST23,24 data show that the experimental STD data collected for the 1/RXR system correspond with a dual ligand binding mode (Figure 2.1d & 2.1e, experimental section) at both the ligand-binding pocket and the coactivator side.
Figure 2.1 | Honokiol (1) binds to both sides of the dynamic AF2 interface of RXR. a) Four selected examples from a fluorescence-based co-factor-recruitment assay showing inhibition of coactivator protein binding at 100 μM of 1 in the presence of a potent full agonist, 4; b) Fluorescence polarization data showing that full agonist 4 induces binding of a fluorescently labelled coactivator peptide in a concentration-dependent manner. Repeating the assay in the presence of increasing concentrations of 1 (10, 100 μM) resulted in a progressive decrease in the maximum polarization signal, but without
changing the EC50 of 4, thus showing an alternative binding mode. c) A summary of 1D-
31 CHAPTER 2
1H NMR data revealing the atypical dual binding of 1: line broadening of the 1H- resonances of 1 in the presence of protein and agonist ligand; recovery of sharp intense signals on addition of a competitor LXXLL peptide; d) 1H-NMR spectrum of honokiol, 1 (bottom) and STD spectrum of the system 1/RXR (top) show rapid ligand exchange e) The sum of separate theoretical CORCEMA-ST23,24 values at the ligand binding pocket and coactivator side of the AF2 interface compare well with the experimental STD data for the system 1/RXR, thus indicating the dual ligand binding mode.
To capitalize on the dual-binding properties of 1, an orthogonal pair of RXR ligands was developed capable of selectively targeting opposite sides of the dynamic AF2-surface. There was reason to believe that 1 inhibits coactivator-binding by mimicking the LXXLL binding motif, which is highly-conserved throughout coactivator proteins. Indeed, an overlay of the energy-minimized state of 1 and the co-crystal structure of an α-helical coactivator peptide bound to RXR (PDB ID: 2P1T)39 identified a strong overlap of the interacting Leu residues at positions i and i+4 of the α-helix and the allylic side-chains of 1 (Figure 2.2a). Furthermore, the 4’-hydroxyl functional group (para to the biaryl bond) makes a stabilizing hydrogen-bonding interaction with Glu453 – a charge-clamp residue important for selective binding of the helical LXXLL motif – on molecular docking of the LXXLL-aligned model of 1 to the AF2.
Figure 2.2 | a) Overlay of honokiol, 1 (orange) and LXXLL mimic 3 (cyan) with the LXXLL- coactivator peptide (red, PDB ID: 2P1T) bound to the RXR AF2.39 b) Cellular activities of 1 and 3 measured in a mammalian two-hybrid luciferase assay in which increasing concentrations of the ligand are co-incubated with a fixed, 100 nM, concentration of full RXR agonist, 4.
32 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
To favor selective AF2 binding, isobutyl-analog 3 was synthesized (Scheme 2.1; 25%, 3 steps, experimental section), which initially hypothesized would serve as a better mimic than 1 of the of the LXXLL motif. Experimentally, analog 3 was inactive as agonist in a mammalian two-hybrid assay (M2H) up to 50 μM (Table 2.1). In the same agonistic assay, however, 1 alone elicited a complex response, which was explained by the dual binding properties delineated in Figure 2.1c-e. Although an EC50 value could not be determined for 1 in this case, nevertheless at concentrations between 1-25 μM, 1 induced partial activation of luciferase expression followed by inhibition at the highest 50 μM test concentration.29 The catalytic activity of the luciferase protein was unchanged, in the absence or presence of 1 and 3, thus ruling out direct inhibition of luciferase as a possible mode-of-action. Similar to 1, analog 3 also inhibited coactivator binding to the AF2 in the
fluorescence polarization assay, albeit with slightly reduced affinity (Ki = 199 μM ± 2 μM).40 Importantly though, analog 3 was more effective than 1 at supressing the full agonistic activity of 4 in the mammalian two-hybrid assay (Figure 2.2b). The improved cellular activity of 3 can be explained by an improved selectivity for the solvent-exposed coactivator side of the AF2. Indeed molecular modelling suggested that the isobutyl substituents on 3 disfavor binding at the ligand binding pocket due to additional repulsive interactions (experimental section). We conclude therefore that compared to 1 LXXLL-mimic 3 selectively inhibits the RXR-coactivator interaction via a more preferential binding at the coactivator-side of the dynamic AF2-interface. The next aim was to switch selectivity from the coactivator-side of the AF2 to the ligand-binding pocket. Potent RXR ligands targeting the ligand-binding pocket (e.g., 4, Figure 2.1b) typically require a carboxylate group, which forms a salt-bridge with residue Arg316 in the hydrophilic region of the binding pocket. The modelling data (experimental section) indicated that modifying one of the allylic side-chains of 1 would favor binding at the ligand-binding pocket. Unsure of the binding preference, Analogs 15, 17 & 18 were synthesized using an efficient palladium-catalyzed cross-coupling route (scheme 2.2) in which the key biaryl bond was formed under Buchwald-modified Suzuki conditions.41
33 CHAPTER 2
Scheme 2.2. Synthesis of ligands targeting the RXR ligand-binding pocket. Reagents and
conditions: a) methyl acrylate, [Pd(dppf)Cl2] (20 mol%), Et3N, DMF, reflux, 63–86%; b)
bis(pinacolato)diboron, [Pd(OAc)2] (4 mol%), XPhos (8 mol%), KOAc, dioxane, reflux,
77–86%; c) 2-allyl-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, [Pd(PPh3)4] (10 mol%), CsF,
THF, reflux; d) 7 or 1-allyl-3-bromobenzene, [Pd2(dba)3] (10 mol%), SPhos (30 mol%),
dioxane/H2O, 110 °C, 18 h; e) 6 or 9, [Pd2(dba)3] (10 mol%), SPhos (30 mol%),
dioxane/H2O, 110 °C, 18 h; f) BBr3, CH2Cl2, –78 °C, 7–51%; g) NaOH (aq.), MeOH/1,4- dioxane, 46%-quantitative. dppf = 1,1'-Bis(diphenylphosphino)ferrocene, dba = dibenzylideneacetone.
Table 2.1 | Summary of fluorescence polarization (FP) and mammalian two-hybrid (M2H) data for synthetic ligands vs. honokiol, 1, and LG 100268 (4).
[a] Compound FP/EC50 M2H (Luciferase)/ EC50 (μM)[b] (μM)[c]
LG100268 (4) 0.15 ± 0.04 0.0051 ± 0.002 1 inactive -* 2 0.26 ± 0.11 0.0063 ± 0.004 3 inactive inactive
34 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
15 >250 >50 17 >250 >50 18 8.3 ± 2.2 0.31 ± 0.04 19 1.2 ± 0.48 6.2 ± 1.6
[a] See Figure 3 for further synthesis details; [b] fluorescence polarization (FP) assay to determine direct binding (EC50); [c] agonistic mammalian two-hybrid (M2H) luciferase assay (EC50). See the experimental section for details about the different assay. formats. * partial activity measured.
Whereas 15 and 17 were only weakly active in both fluorescence polarization and M2H assays, analog 18 showed significant activity (EC50 (FP) = 8.3 ± 2.2 μM; EC50 (M2H) = 0.31 ±
0.04 μM, Table 1). The binding model hinted at further activity gains by removing the 4’- hydroxyl group (Figure 3, R2 = OH → H). Therefore, analog 2 (Scheme 2.2) was prepared via a similar synthetic route, and, was gratifyingly 40-fold more active than 18 (EC50 (FP) =
0.26 ± 0.11 μM; EC50 (M2H) = 0.0063 ± 0.004 μM, Table 1). The circa 20-30-fold difference between the FP and M2H data is a common phenomemon,15 which can be explained by intrinsic differences between the two quite different assay formats, in particular, the different protein and peptide concentrations used.
Figure 2.3 | Ribbon representation of the X-ray co-crystal structure (PDB ID: 4OC7) of 2 bound to the RXR ligand-binding domain: a) protein (green), TIF2-derived coactiavator peptide (red), 2 (orange). b) Zoomed-in view of the RXR ligand-binding pocket with amino acid side-chains represented as sticks, and the electron density map of 2.
35 CHAPTER 2
To gain further molecular insight at the ligand-binding pocket, the X-ray co-crystal of 2 bound to the RXR ligand-binding domain was solved at 2.6 Å resolution (Figure 2.3). The carboxylate group of 2 is seen making a canonical interaction with Arg316, while the flexible allylic side-chain occupies the lipophilic region of the binding pocket. A molecular overlay with known RXR co-crystal structures (PDB IDs: 2P1T & 4K6I) did not reveal any significant differences in global protein conformation. There was no electron density found in the X-ray structure, nor evidence from MS data (data not shown) to suggest covalent attachment of 2 to the RXR protein, thus ruling out irreversible inhibition as a possible mode-of-action. Combined with the biochemical and cellular results, this data suggests that, in contrast to current RXR ligands, a rigid and bulky hydrophobic moiety is not necessary for potent binding at the ligand-binding pocket. However, analog 19 lacking the hydroxyl groups at the 2- and 4’-positions (Scheme 2.2) was significantly less active than 2 (> 200-fold), highlighting the importance of the 2- hydroxyl group for activity,[26] by simultaneously restricting the rotational freedom about the biaryl bond and through formation of a hydrogen-bonding interaction with residue Asn306. Thus, by rational design and using a short and focused synthetic route, a selective switch was managed of the targeting properties of 1 from one side of the dynamic AF2 interface of RXR to the other – from the solvent-exposed side to the ligand- binding pocket – and most notably with improved ligand efficiency (BEI)40 compared to
known RXR ligands (2, BEI (FP) = 23.5 vs. 4, BEI (FP) = 18.8). In summary, the rational dissociation was of the dual-binding properties of a natural product at a dynamic protein interface was demonstrated. This outcome has resulted in two distinct and molecularly efficient ligand types targeting opposite sides of the activation function 2 (AF2) of the retinoid X receptor (RXR). The first ligand type, represented by 3, exhibits an atypical behavior, inhibiting the coactivator binding at the solvent-exposed side of the AF2 interface. Notably, ligand 3 is the first of its kind selective for RXR. The second type, represented by 2, potently binds to the ligand-binding pocket, thereby inducing coactivator binding via an established mechanism. Our findings justify the future exploration of natural products at dynamic protein interfaces.
36 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
Experimental section
General considerations for protein expression and purification
All solutions and equipment used in the handling of microbial cultures were autoclaved or sterile filtered. Media, plastic and glassware were autoclaved at 121 °C for 20 min prior to use. Bacterial cultures were incubated in a New Brunswick Series 25 shaker. Centrifugation was performed in a Beckman Coulter Avanti J-25 centrifuge. Microcentrifugation was performed in an Eppendorf Centrifuge 5415R or a Beckman Coulter Microfuge 18. All biological laboratory buffers and media were bought from common suppliers and used as purchased. BL21(DE3) and NovaBlue E. coli competent cells were purchased from Novagen. DNA and protein concentration was determined using a NanoDrop 1000 spectrometer from Thermo Scientific using 260 nm and 280 nm wavelength respectively. Gel electrophoresis for proteins was performed using 12% SDS-PAGE gels in running buffer and visualized using InstantBlue stain. The fluorescent D22 coactivator peptide42 was purchased from Invitrogen life technologiesTM
Protein expression and purification for cofactor recruitment assays and polarization assays
The human (GST)RXRα LBD protein was produced by expression of the pET15b (Novagen) vector in E. Coli BL21(DE3) (Novagen). TB medium in a 2 L baffled conical flask was supplemented with ampicillin (100 μg/mL) and inoculated with 25 mL overnight culture. The culture was incubated at 37 °C and 250 rpm to an A600 of 0.6-0.8, cooled to 21 °C and then protein expression was induced by adding 0.1 mM isopropyl β-D-thiogalactopyranoside (IPTG). The culture was left to grow for 16 h at 21 °C before the cell pellet was collected by centrifugation (4500 rpm, 20 min, 4 °C) and stored at -80 °C until purification.
The cell pellet was suspended in lysis PBS buffer (pH 8.0), and lysed with a microfluidizer. After centrifugation to remove the cell debris, the cleared lysate was loaded onto a 1 mL Protimo- Glutathione agarose column (GE Healthcare), the protein was eluted with the elution buffer (Tris (50 mM), Glutathione (10 mM) pH 8.0). The pure collected fractions were pooled, concentrated and stored at -80 °C until further use.
Polarization assay
GST-RXRα-LBD (1 μM), fluorescein-labeled D22 co-activator peptide (0.1 μM), and the ligand at the indicated concentration in TR-FRET co-regulator buffer E (Invitrogen life technologiesTM) were incubated for 60 minutes at 4 °C and protected from light. Fluorescent polarization signals (mP) were measured with a Tecan Safire monochromator microplate reader. Experiments were performed in triplicate.
37 CHAPTER 2
Mammalian two-hybrid assay
The mammalian two-hybrid (M2H) assays were performed in U2OS cells. The RXR LBD was cloned into the pCMV-AD vector (Aligent), fused to NF-κB and Strep-tag resulting in NF-κB- Strep-RXR-LBD. The peptide sequence was cloned into the pCMV-BD vector (Aligent) and fused to GAL4 DNA binding domain. For the direct interaction assay, ~40,000 cells/well were seeded in a 24-well plate for 24 h and transfected with 40 ng pCMV-AD, 40 ng pCMV-BD, 0.2 μg pFR- Luc, and 3.2 ng pGL-Renilla using PEI (Polysciences), before being treated with the ligands in indicated concentrations. For the competition assay, cells were co-treated with 100 nM of LG100268 and the indicated concentrations of competitor 1 or 3. After 24 h of treatment the interaction was determined with a Dual-Luciferase® Reporter Assay (Promega), according to the manufacturer’s instruction. The luminescent intensities were recorded on a Synergy HT platereader (BioTek). The FR-luciferase signal was normalized over the Renilla luciferase (pGL- Renilla) signal. pCMV-BD sequence
MKLLSSIEQACDICRLKKLKCSKEKPKCAKCLKNNWECRYSPKTKRSPLTRAHLTEVESRLER LEQLFLLIFPREDLDMILKMDSLQDIKALLTGLFVQDNVNKDAVTDRLASVETDMPLTLRQHRI SATSSSEESSNKGQRQLTVSPELGSASPGEFLTARHPLLMRLLLSPS.
Cofactor recruitment assay
Nuclear receptor profiling was measured using 10 nM concentration of GST-RXRα-LBD, 25 nM anti-GST antibody, 5 mM DTT, 2% v/v DMSO and the ligand at the indicated concentration. All assays were performed on a Pamstation® controlled by EvolveHT software at 28 °C at a rate of 2 cycles per minute. Nuclear receptor PamChip arrays® contained 53 co-regulator peptides, derived from natural occurring co-regulator proteins. Per array 25 μL of assay mixture was transferred to the chip, during the period of ligand incubation (~40 min), a solution of GST- RXRα-LBD, fluorescent anti-GST and the ligand was pumped through the porous peptide containing membrane for 81 cycles. A .tiff format image of each array was obtained after 81 cycles by a charge coupled device camera-based optical system integrated in the work station. Quantification of the pictures was performed by the integrated EvolveHT software.37
Evidence for RXR selectivity of honokiol (1)
38 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
Figure S1. Coactivator-peptide microarray profile for the retinoid X receptor (RXR), subtype α and ß, with LG 100268 (4); estrogen receptor α and ß (ERα, ERß) with estradiol (E2); androgen receptor (AR) with dihydrotestosterone (DHT). Compound 4, E2 and DHT all target the ligand- binding pocket of their respective nuclear receptors.
Influence of compound 1 and 3 on luciferase activity
Luciferase from Photinus pyralis (firefly) was obtained from Sigma-Aldrich®. The protein was diluted in passive lysis buffer (Promega). 50 μL of luciferin solution (Promega) was added to the protein solution and the luciferase activity was directly measured on a Tecan Safire monochromator microplate reader. Experiments were performed in triplicate. The influence of the ligands on the luciferase activity was measured at 1 μM of luciferase.
Ki determination for 1 and 3
Fluorescence polarization measurements were performed in black 384-well-plates on a Safire2 (Tecan) plate reader. The sample volume was 30 μL containing constant concentrations of 4 (2.4 μM), RXR (2.9 μM), fluorescent peptide (20 nM) and increasing concentrations of co-factor inhibitor. The data was fitted using eq. 1
eq.1
Where A1 is the botom asymptote, A2 is the top asymptote and p is the Hill slope. The half maximum inhibitory concentration was determined via eq. 2
10 eq.2
The Ki values for compound 1 and 3 were determined with the Kd value for the protein and peptide using eq 3 and 4.
39 CHAPTER 2
∙ ∙ eq.3 ∙ ∙
eq.4
Where Kd is the dissociation constant of the fluorescein-labeled peptide protein complex, y0 is the initial bound to free concentration ration for labeld peptide, (AB) is the concentration of protein- peptide complex, A the concentration of unbound peptide and B is the concentration of RXR.
Table S1. IC50 and Ki values for 1 and 3 meausured in the absence and presence(*) of 0.01% of Triton X-100 as mentioned by B. Y. Feng, B. K. Shoichet, Nat. Protoc., 2006, 1, 550 – 553.
Compound IC50 polarization Ki 1 287 μM +/- 18 μM 94 μM +/- 9 μM
3 606 μM +/- 5 μM 198 μM +/- 2 μM
1* 328 μM +/- 141 μM 107 μM +/- 80 μM
3* 339 μM +/- 80 μM 110 μM +/- 26 μM
Expression, purification and crystallization of the RXRα LBD.
The histidine-tagged LBD of human RXRa (in a pET15b vector) was expressed in Escherichia coli BL21(DE3). Cells were grown at 37 °C in LB medium supplemented with 100 mg mL-1 ampicillin until OD600 reached about 0.7. Expression of T7 polymerase was induced by addition of isopropyl-b-d-thiogalactoside (IPTG) to a final concentration of 0.1 mM. After an additional incubation for 15 h at 15°C, cell cultures were harvested by centrifugation at 8,000 ´ g for 20 min. The cell pellet from 2 liters of RXRa LBD was resuspended in 50 ml buffer A (20 mM Tris-HCl pH 8.0, 500 mM NaCl, 5 mM imidazole) supplemented with a protease inhibitor (PMSF) and DNAse I. The suspension was then lysed by sonication and centrifuged at 35,000 g and 4 °C for 45 min. The supernatant was loaded onto a 5 ml Ni2+-affinity column, preequilibrated with buffer A. The column was washed with 10 volumes of buffer A and 10 volumes of buffer A supplemented with 50 mM imidazole. Bound proteins were eluted with buffer A containing 200 mM imidazole. The fractions containing RXRa LBD were pooled, concentrated and desalted to buffer B (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM DTT). To remove the histidine-tag the protein was incubated for 16 h at 4°C with thrombin (1 unit/mg RXR). The protein was passed through a Ni2+ column and a superdex gel filtration column. The protein was concentrated and storred at -80°C until further use.
40 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
Before crystallization the protein was mixed with a 1.5-fold molar excess of compound 2 and a 3- fold excess of TIF2 NR2 cofactor peptide (686-KHKILHRLLQDSS-698). The complex was incubated for 1h at 4 °C. initial screening was performed using JCSG+ (Qiagen) and NR LBD screen (molecular dimensionsTM) at 4 °C using hanging drop vapor diffusion. 200 nL reservoir solution was automatically mixed with 200 nL protein solution using a pipetting robot (Mosquito® Crystal) in a 96-well plate. The drops were equilibrated against 75 μL of reservoir solution. After three weeks crystals had appeared in 14 conditions. Reproduction of the crystals were performed in 24-well plates using the hanging drop vapor difusion method. Drops with a size of 2 - 3 μL using different resevoir to protein ratio were manually mixed at 4 °C and equilibrated against resevoirs with a volume of 1 mL. Optimal crystals were grown in a week in 3 μL drops with protein solution to resevoir ratio of 2:1 with 0.1 M NaHEPES, pH 7.5 and 20% PEG 2K MME. The crystals were cryo-cooled in liquid nitrogen using sucrose as cryo-protectant for X- ray data collection.
His-RXR sequence:
10 20 30 40 50 60 GSSHHHHHHS SGLVPRGSHM TSSANEDMPV ERILEAELAV EPKTETYVEA NMGLNPSSPN 70 80 90 100 110 120 DPVTNICQAA DKQLFTLVEW AKRIPHFSEL PLDDQVILLR AGWNELLIAS FSHRSIAVKD 130 140 150 160 170 180 GILLATGLHV HRNSAHSAGV GAIFDRVLTE LVSKMRDMQM DKTELGCLRA IVLFNPDSKG 190 200 210 220 230 240 LSNPAEVEAL REKVYASLEA YCKHKYPEQP GRFAKLLLRL PALRSIGLKC LEHLFFFKLI 250 260 GDTPIDTFLM EMLEAPHQMT
Table S2. Crystallographic statistics for complex between RXR & analog 2 Data collection Wavelength (Å) 0.97794 Resolution (Å) 47.77-2.50 (2.59-2-50) Space group P43212 Cell parameters (Å) a=b=67.55, c=110.07 a,b CC1/2 (%) 100 (88.2) a,c Rsym (%) 8.3 (185) a,d Rmeas (%) 8.49 (246) a Average I/σ(I) 26.9 (2.0) Completeness (%)a 99.9 (100) No. of unique reflectionsa 9329 (897) Redundancya 23.6 (25.1) Wilson B-factor (Å2) 67.3 Mosaicity (°) 0.25
Refinement
41 CHAPTER 2
Number of protein/solvent/ligand atoms 1769 / 8 / 21 Rwork/Rfree (%) 19.77 / 24.24 No. of reflections in the 'free' set 432 R.m.s. deviations from ideal values bond lengths (Å) / bond angles (°) 0.004 / 0.75 Average protein/solvent B-factor (Å2) 67.1 / 61.2 Ramachandran plot: favoured/outlier residues (%) 98.1/0 Molprobity validation: score/percentile 1.15/99 a number in parentheses is for the highest resolution shell b 43 CC1/2 = Pearson's intra-dataset correlation coefficient, as described by Karplus and Diederichs. The values were reported by phenix.cc_star44) c th Rsym = ∑h∑l│Ihl-
Materials and methods crystallography
Diffraction data for RXR/2 were collected at the PXIII beamline (Swiss Light Source, Villigen, Switzerland). The data set was indexed and integrated using XDS45 and scaled using SCALA.46 The structure was phased by molecular replacement using PDB ID 1MVC47 as search model in Phaser.48 Coot49 and phenix.refine44 were used in alternating cycles of model building and refinement. The quality of the final model was evaluated using MolProbity. All data collection, refinement and validation statistics are shown in Table S2. The crystal structure was submitted to the PDB with identifier 4OC7.
NMR experiments
NMR experiments were recorded at 298 K on a Bruker 600 MHz spectrometer, equipped with a 5 mm triple-resonance probe. Samples were buffered in 10 mM d11-Tris-HCl, 150 mM NaCl, 1mM DTT, at pH 7.5, and containing 2.5-5% v/v DMSO for ligand solubility. Standard 1H 1D and 2D NMR spectroscopy experiments such as TOCSY and NOESY were recorded to assign the proton signals of the different ligands. For the ligand screening reporter experiments, samples contained constant concentrations of honokiol 1 (250 μM) and GST-RXRα-ligand- binding domain (LBD) (1 μM), and final concentrations of agonist ligands (analog 2 and LG100268, 4) and coactivator peptide (750 μM). Saturation transfer difference (STD) experiments were carried out at constant His-RXRα-LBD concentration (5 μM) and also constant 50:1 ligand/protein ratio. For STD experiments, saturation was achieved using a pulse train of 50 ms Gaussian shaped pulses for a 2s duration and an offset of δ=−1 ppm (on-resonance) or δ=100 ppm (off-resonance). To obtain STD spectra, a total number of 1024 scans were acquired and subtraction of the on-resonance from the off-resonance spectra yielded the STD signal. The epitope mapping of the ligands was obtained by normalizing the observed STD signal intensities
42 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE with respect to that of the highest STD response. 2D tr-NOESY experiments were recorded at constant 20 μM His-RXRα-LBD concentration, by using mixing times of 200 ms, using 25:1 ligand/protein molar ratios.
a) Honokiol a) Honokiol 250µM 250µM
b) + RXR‐LBD b) + RXR‐LBD protein protein 1µM 1µM
c) c) H2 + agonist + agonist 2 4 H2 (3:1) (3:1) H2 H4 H2 H4 H4
d) + coactivator d) + coactivator peptide peptide (3:1) (3:1) H1 H1 H1 H1
Figure S2. 1H-NMR competition experiments using honokiol as reporter ligand and full agonist ligands 2 (left panel), 4 (right panel) and a coactivator peptide as test compounds. Unambiguous protons are indicated as H1 for honokiol protons, H2 for compound 2 protons and H4 for compound 4 protons. When honokiol is bound to the protein, the corresponding 1H NMR resonances are significantly broadened. Only in the presence of a competitive binder, in this case the coactivator peptide, honokiol is displaced from the binding site into the bulk solution, leading to sharpening of the signals and recovery of the honokiol signal intensity.
A
6, 4 3’,3 B 6’,2’ 8’,8 9,9’ 7’ 7
C
> > > 80‐100% 60‐80% 40‐60% <40% STDeffect
Figure S3. STD experiment, A) 1H-NMR spectrum of honokiol, 1. B) STD spectrum of the system 1/RXR, C) Epitope-mapping obtained for 1 in presence of RXR.
43 CHAPTER 2
6,4 7’ 3 8 A 3’ 7 4’ 8’ 9’
B
H2 H2 H2 H2 H1 H1 H1
6,4 3’,3 7’ 7 6’,2’ C
8’,8 9,9’
Figure S4. STD competition experiments, A)1H NMR of 2. B) STD spectrum of the system 1/2/RXR. C) 1H-NMR spectrum of honokiol, 1. Unambiguous STD signals are indicated as H1 for honokiol protons and H2 for compound 2 protons. STD analysis shows a clear binding of 1 to RXR in presence of the agonist 2, again suggesting a binding site for 1 that is different from the ligand-binding pocket.
Corcema calculations
The theoretical STD calculations were performed using the CORCEMA-ST program23,50 and compared to the experimental data, evaluating the two binding sites for 1: at the ligand-binding pocket, and the coactivator side of the AF2 interface. 1 was docked into the ligand-binding pocket of RXR, based on the co-crystal structure of compound 2-RXR complex, and followed by a standard minimization and MD simulation using the Maestro 9.3 suite from Schrödinger. 1 was docked separately at the AF2 site, as explained (Figure S6). Only residues which were 8 Å from the ligand were considered in the matrix calculations. The overall correlation time τC for the free state was set to 0.5 ns, a typical value based on the low molecular weight of the ligand. The τC for
51 the bound state was set to 16 ns (τC calculated with HYDROPRO). For methyl groups, a value of 0.85 was used for the order parameter S2 and 10 ps for the internal motion correlation time. -1 8 -1 -1 Values for the leakage factor of 0.1 s and a kon of 10 s M were used, for diffusion-controlled binding. Other parameters, such as the saturation time, protein and ligand concentrations, methyl protons in the ligand and in the protein and magnetic field were also given as an input data for the calculation. Values of Keq were set based on measured kD results and modified within 4 ± 20% variation ranges of reference values to obtain the best R-factor values: values of Keq 4x10 M-1 were used for the complex honokiol-RXR at the ligand-binding pocket, whereas for the 3 -1 complex at the coactivator side of the AF2, a value of Keq 3.3x10 M was used in the calculations.
The theoretical STD intensities were quantitatively compared to those obtained experimentally by NMR, by the scoring function termed R-factor, as implemented in the program CORCEMA-ST.
44 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
Using the RXR-1 models and the parameters mentioned above, we were able to obtained a R- factor value of 0.37 at the ligand-binding pocket and 0.70 at the coactivator side of the AF2. However, it is worth noting that the single pose at the ligand-binding pocket was not able to reproduce the complete experimental STD profile for some of the protons, such as H8. In fact, the linear combination of STD values obtained for the two honokiol binding sites gave the best agreement with the experimental data (Figure S3), thus supporting the existence of both binding sites (Figure S5).
a) b) c)
Figure S5. Comparison between experimental and theoretical STD data (CORCEMA-ST) for honokiol in the presence of RXR, at both LBP and AF2 sites. The experimental STD effects (black) for honokiol, compared with the calculated ones (red) for honokiol at a) the LBP site, b) at the AF2 site, c) the sum of STD values at both LBP and AF2 sites.
a) b)
Figure S6. Structural representation of honokiol-bound models at a) the ligand-binding pocket and b) coactivator side of the AF-2 interface. For each site, RXR residues within 5 Å of the honokiol ligand are shown.
Molecular modeling
The X-ray crystal structure of the retinoid X receptor in complex with 3-(2'-methoxy)- tetrahydronaphthyl cinnamic acid and a fragment of the coactivator TIF-2 (PDB code: 2P1T)13 was used for the modeling. Compounds 1 and 3 were modeled so as to occupy the ligand-binding pocket or so as to be superimposed with the LXXLL motif of TIF-2 for Figure 2.2a in the main
45 CHAPTER 2 manuscript and Figure S7, respectively. Compounds 1 and 18 were modeled so as to occupy the ligand-binding pocket. The energy of the system was minimized using the MAB force field as implemented in the computer program MOLOC,14 whilst keeping the protein coordinates fixed. After energy minimization of the designed fragments, all types of interactions (hydrogen bonds and lipophilic interactions) between the fragments and the protein were analyzed in MOLOC.
a) b)
Figure S7. Structural representation of RXR modeling data (protein and ligands represented as sticks): a) molecular overlay of 1 (orange) and 3 (cyan) at the RXR ligand-binding pocket; b) molecular overlay of 1 (orange) and 18 (gray). Dashed lines indicate stabilizing (black) and repulsive (red) interactions.
General considerations synthetic procedures
All the solvents employed were obtained from Biosolve BV and used without purification unless stated otherwise. The reagents were obtained from either Sigma-Aldrich or Acros and used without purification. Honkiol (1) and magnolol were bought from Carbosynth Limited. LG 100268 (4) was bought from Tokyo Chemical Industry Co., Ltd. Water was purified using a Millipore purification train. Dry solvents were tapped off a solvent purification column (MBRAUN). All the NMR data were recorded on a Varian Gemini 400 MHz NMR or a Bruker Cryomagnet for NMR spectroscopy 400 MHz (400 MHz for 1H-NMR and 100 MHz for 13C- NMR). Proton experiments are reported in parts per million (ppm) downfield of TMS and were relative to for the residual methanol (3.31 ppm) or chloroform (7.26 ppm). All 13C spectra were reported in ppm relative to residual methanol (49.00 ppm) or chloroform (77 ppm) For the LC-
MS a C4, Jupiter SuC4300A, 150x2.00 mm column using H2O with 0.1 % F.A. and acetonitrile with 0.1% F.A. with a gradient of 5% to 100% acetonitrile in 10 min, with a flow rate of 0.2 mL/min. The preparative HP-LC was performed on a Gemini S4 110A 150x21.20 mm column using H2O with 0.1% F.A. and acetonitrile with 0.1% F.A. Gradient: 40% to 60% acetonitrile in 40 minutes.
46 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
1 Commercial 4-bromo-2-chlorophenol HNMR (400 MHz, CDCl3): δ (ppm) 7.46 (d, J = 2.4 Hz, 1H), 7.28 (dd, J = 2.0, 8.4 Hz, 1H), 6.90 (d, J = 8.8 Hz, 1H), 5.49 (s, 1H); 13C NMR (100 MHz,
CDCl3): δ 150.82, 131.59, 131.46, 120.95, 117.77, 112.48.
4-bromo-2-chloro-1-methoxybenzene (8) To a solution of 4-bromo-2-chlorophenol (4.0 g, 19.3 mmol) in dry DMF (110 mL) in an oven dried flask, K2CO3 (13.3 g, 91.5 mmol) was added. The resulting mixture was stirred for 30 min at room temperature. Iodomethane (8.2 g, 3.8 mL, 57.9 mmol) was added drop wise over 10 min and the reaction was stirred at 70 °C under positive argon pressure. After 18 h the mixture was cooled to room temperature and separated between
CH2Cl2 and H2O. The aqueous layer was washed with CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated to yield crude yellow oil. The crude material was purified via column chromatography eluting with hexane/EtOAc 1% to 5% (v/v), to yield 4-bromo-2-chloro-1-methoxybenzene 8 as a white solid, 4.15 g, 97%. Silica gel TLC
1 Rf = 0.4 (Hexane/EtOAc 5% v/v); HNMR (400 MHz, CDCl3): δ (ppm) 7.49 (d, J = 2.4 Hz, 1H),
13 7.32 (dd, J = 2.4, 8.8 Hz, 1H), 6.81 (d, J = 8.8 Hz, 1H), 3.88 (s, 1H); C NMR (100 MHz, CDCl3): δ 154.52, 132.81, 130.68, 123.76, 113.44, 112.64, 56.46.
4-allyl-2-chloro-1-methoxybenzene (10) In an oven dried schlenk flask 8 (1.10 g, 4.96 mmol), cesiumfluoride (1.51 g, 9.94 mmol), tetrakis(triphenyl-phosphine)palladium(0) (0.57 g, 0.49 mmol) were charged. The schlenk flask was capped with a rubber septum and then evacuated and backfilled with argon three times. Dry and degassed THF (40 mL) was added via a syringe trough the septum and the mixture was stirred at room temperature under positive argon pressure. After 30 minutes allylboronic acid pinacol ester (1.5 g, 8.9 mmol) in dry THF (15 mL) was added. The schlenk flask was closed with a Teflon screwcap and the reaction mixture was stirred for 21 h at 78 °C. Under positive argon pressure a second portion of cesium fluoride (1.51 g, 9.94 mmol) and tetrakis(triphenylphosphine)-palladium(0) (0.57 g, 0.49 mmol) and THF (10 mL) were added. The reaction mixture was stirred at 78 °C. After 24 h the mixture was allowed to cool to room temperature and the mixture was separated between H2O (120 mL) and pentane (120 mL). The aqueous layer was washed with pentane (2 x 100 mL). the combined organic layers were washed with brine (100 mL), dried over Na2SO4, filtered and concentrated to yield a yellow oil. The crude material was purified via column chromatography eluting with Hexane/EtOAc 1-
5% (v/v) to yield 10 as colorless oil, 813 mg, 90%. Silica gel TLC Rf = 0.5 (Hexane/EtOAc 5% v/v);
1 HNMR (400 MHz, CDCl3): δ (ppm) 7.19 (d, J = 2.4 Hz, 1H), 7.02 (d, J = 2.4, 8.4 Hz, 1H), 6.87
47 CHAPTER 2
(d, J = 8.4 Hz, 1H), 5.87-5.97 (m, 1H), 5.09 (t, J = 11.2, 2H), 3.88 (s, 3H), 3.31 (d, J = 6.8 Hz, 2H);
13 C NMR (100 MHz, CDCl3): δ 153.50, 137.18, 133.35, 130.47, 127.83, 122.38, 116.23, 112.22, 56.35, 39.14.
1 Commercial 2-bromo-4-chlorophenol HNMR (400 MHz, CDCl3): δ (ppm) 7.46 (s, 1H), 7.18 (d, J
13 = 12,0, 1H), 6.95 (d, J = 8,0, 1H), 5.45 (s, 1H); C NMR (100 MHz, CDCl3): δ 151.28, 131.47, 129.39, 117.02, 110.50, 95.13.
2-bromo-4-chloro-1-methoxybenzene (5) To a solution of NaH (696 mg, 17.4 mmol) in dry THF (29 mL) was added a solution of 2-bromo-4-chlorophenol (3.0 g, 14.5 mmol) in dry THF (10 mL) in an oven dried flask. The resulting mixture was stirred for 40 min at room temperature. Iodomethane (6.2 g, 2.7 mL, 43.5 mmol) was added drop wise over 10 min and the reaction was stirred at 76 °C under positive argon pressure. After 24 h the mixture was cooled to room temperature and separated between CH2Cl2 (50 mL) en H2O (50 mL). The aqueous layer was washed with CH2Cl2 (2 x 50 mL); the combined organic layers were washed with brine, dried over
Na2SO4, filtered and concentrated to yield crude yellow oil. The crude material was purified via column chromatography eluting with hexane/EtOAc 1% to 5% (v/v), to yield 2-bromo-4-chloro-1- methoxybenzene 5 as colorless oil, 2.31 g, 72%. Silica gel TLC Rf = 0.4 (Hexane/EtOAc 5% v/v);
1 1 HNMR (400 MHz, CDCl3): δ (ppm) H NMR (399 MHz, CDCl3) δ 7.53 (s, 1H), 7.23 (d, J = 8.0,
13 1H), 6.81 (d, J = 8.0, 1H), 3.88 (s, 3H); C NMR (100 MHz, CDCl3) δ 154.92, 132.99, 128.43, 126.13, 112.70, 112.29, 56.63.
2-allyl-4-chloro-1-methoxybenzene (7) In an oven dried schlenk flask 5 (1.03 g, 4.65 mmol), cesiumfluoride (1.37 g, 9.02 mmol), tetrakis(triphenyl-phosphine)palladium(0) (0.53 g, 0.45 mmol) were charged. The schlenk flask was capped with a rubber septum and then evacuated and backfilled with argon three times. Dry and degassed THF (30 mL) was added via a syringe trough the septum and the mixture was stirred at room temperature under positive argon pressure. After 30 minutes allylboronic acid pinacol ester (1.37 g, 8.1 mmol) in dry THF (10 mL) was added. The schlenk flask was closed with a Teflon screwcap and the reaction mixture was stirred for 25 h at 78 °C. Under positive argon pressure a second portion of cesium fluoride (1.36 g, 8.95 mmol) and tetrakis(triphenylphosphine)-palladium(0) (0.53 g, 0.46 mmol) and THF (5 mL) were added. The reaction mixture was stirred at 78 °C. After 24 h the mixture was allowed to cool to room temperature and the mixture was separated between H2O (100 mL) and pentane (100 mL). The aqueous layer was washed with pentane (2 x 80 mL). The combined organic layers
48 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
were washed with brine (100 mL), dried over Na2SO4, filtered and concentrated to yield yellow oil. The crude material was purified via column chromatography eluting with Hexane/EtOAc 3%
(v/v) to yield 7 as colorless oil, 662 mg, 78%. Silica gel TLC Rf = 0.4 (Hexane/EtOAc 5% v/v);
1 HNMR (400 MHz, CDCl3): δ (ppm) 7.12 (dd, J = 2.8, 8.4 Hz, 1H), 7.10 (d, J = 2.8 Hz, 1H), 6.75 (d, J = 8.4 Hz, 1H), 6.00-5.89 (m, 1H), 5.04 (t, J = 6.8, 17.6 Hz, 2H), 3.80 (s, 3H), 3.33 (d, J = 10.4
13 Hz, 2H); C NMR (100 MHz, CDCl3): δ 155.99, 136.18, 130.62, 129.71, 127.00, 125.44, 116.25, 111.56, 55.81, 34.09.
(E)-methyl 3-(3-chloro-4-methoxyphenyl)acrylate (20)52,53 An oven dried Schlenk flask fitted with a stirring bar was charged with 4-bromo-2-chloro-1-methoxybenzene (8) (182.0 mg, 0.82 mmol) and methyl acrylate (0.29 mL, 3.52 mmol). To this mixture dry DMF (12 mL) was added.
Pd(dppf)Cl2•CH2Cl2 (131.0 mg, 0.16 mmol) and Et3N (4 mL, 54.4 mmol) were added under a positive argon pressure and the schlenk flask was sealed with a teflon screw cap. The mixture was stirred at 110 °C. After 20 h the mixture was diluted with ethyl acetate (50 mL) when the mixture was still warm and poured into H2O (50 mL). The aqueous layer was washed with ethyl acetate (2 x 50 mL). The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated under reduced pressure. The obtained crude brown paste was purified via column chromatography eluting with hexane/EtOAc 5% to 15% (v/v) to obtain 20 as a white solid. Yield:
160.4 mg, 0.71 mmol, 86%. Silica gel TLC Rf = 0.1 (Hexane/EtOAc 4% v/v); LC-MS (ESI): calc.
1 for C11H11ClO3 [M+H]: 227.66, observed 227.1, LC, Rt=5.88. HNMR (400 MHz, CDCl3): δ (ppm) 7.60 (d, J = 12.0 Hz, 1H), 7.56 (s, 1H), 7.38 (d, J = 8.4 Hz, 1H), 6.92 (d, J = 8.8, 1H), 6.29 (d, J =
13 16.0, 1H), 3.93 (s, 3H), 3.80 (s, 3H) ); C NMR (100 MHz, CDCl3): δ 167.50, 156.68, 143.27, 129.60, 128.38, 128.11, 123.31, 116.87, 112.13, 56.42, 51.85.
(E)-methyl 3-(4-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate (9)54,55 An oven dried Schlenk tube was fitted with a stirring bar and charged with: Pd(OAc)2 (6.76 mg, 0.030 mmol), XPhos (26.64 mg, 0.056 mmol), KOAc (188.83 mg, 1.92 mmol) and bis(pinacolato)diboron (482.89 mg, 1.90 mmol). The schlenk flask was sealed with a teflon screw cap and evacuated and backfilled with argon three times. 20 (143.2 mg, 0.63 mmol) in dry and degassed 1,4-dioxane (2.5 mL) was added under positive argon pressure and the mixture was stirred at 110 °C. After 4.5 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude material was purified via column chromatography eluting with hexane/EtOAc 4% to 35% (v/v). The obtained material was
49 CHAPTER 2 recrystallized from diethyl ether to obtain known compound 9 as a white solid in 86% yield.
Silica gel TLC Rf = 0.3 (Hexane/EtOAc 30% v/v); LC-MS (ESI): calc. for C17H23BO5 [M+H]: 319.2
1 observed 319.2 LC, Rt=6.22. HNMR (400 MHz, CDCl3): δ (ppm) δ 7.87 (s, 1H), 7.64 (d, J = 16.0 Hz, 1H) 7.54 (d, J = 8.8 Hz, 1H), 6.85 (d, J = 8.8 Hz, 1H), 6.33 (d J = 16.0 Hz, 1H), 3.87 (s, 3H),
13 3.79 (s, 3H), 1.36 (s, 12H); C NMR (100 MHz, CDCl3): δ 167.94, 165.96, 144.64, 136.98, 132.86, 126.66, 115.39, 110.73, 83.89, 56.08, 51.68, 24.97.
(E)-methyl 3-(5-chloro-2-methoxyphenyl)acrylate (21) An oven dried schlenk flask fitted with a stirring bar was charged with 5 (178.0 mg, 0.80 mmol) and methyl acrylate (0.29 mL, 3.52 mmol). To this mixture dry DMF (12 mL) was added. Pd(dppf)Cl2•CH2Cl2 (134.0 mg, 0.16 mmol) and Et3N (4 mL, 54.4 mmol) were added under positive argon pressure. The schlenk flask was sealed with a teflon screw cap and the mixture was stirred at 110 °C. After 20 h the mixture was diluted with ethyl acetate (50 mL) when the mixture was still warm and poured into H2O (50 mL). The aqueous layer was washed with ethyl acetate (2 times 50 mL). the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated under reduced pressure. The obtained crude brown paste was purified via column chromatography eluting with hexane/EtOAc 5% to 10% (v/v) to obtain 21 as white solid. Yield: 114.1 mg, 0.50 mmol, 63%.
Silica gel TLC Rf = 0.1 (Hexane/EtOAc 4% v/v); LC-MS (ESI): calc. for C11H11ClO3 [M+H]: 227.66,
1 observed 227.0, LC, Rt=7.21. HNMR (400 MHz, CDCl3): δ (ppm) 7.60 (d, J = 12.4 Hz, 1H), 7.56 (s, 1H), 7.38 (d, J = 8.8 Hz, 1H), 6.92 (d, J = 10.4 Hz, 1H), 3.94 (s, 3H), 3.80 (s, 3H). 13C NMR (100
MHz, CDCl3) δ 167.62, 156.93, 138.89, 131.01, 128.36, 125.93, 125.01, 119.66, 112.58, 55.98, 51.86.
(E)-methyl 3-(2-methoxy-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate (6) An oven dried Schlenk tube fitted with a stirring bar was charged with: Pd(OAc)2 (4.32 mg, 0.019 mmol), XPhos (16.52 mg, 0.035 mmol), KOAc (124.62 mg, 1.27 mmol) and bis(pinacolato)diboron (330.31 mg, 1.30 mmol). The Schlenk flask was sealed with a teflon screw cap and evacuated and backfilled with argon three times. 21 (94.5 mg, 0.42 mmol) in dry and degassed 1,4-dioxane (1.8 mL) was added under positive argon pressure and the mixture was stirred at 110 °C. After 4.5 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The product was purified via column chromatography eluting with hexane/EtOAc 15% to 35% (v/v). The obtained material was recrystallized from pentane/diethyl ether 10% (v/v) to obtain novel compound 6 as a white solid in 77% yield. Silica gel TLC Rf = 0.4 (Hexane/EtOAc 1 35% v/v); LC-MS (ESI): calc. for C17H23BO5 [M+H]: 319.2 observed 319.2 LC, Rt=7.78. HNMR
(400 MHz, CDCl3): δ (ppm) 8.02 (d, J = 16.4 Hz, 1H), 7.97 (s. 1H), 7.78 (d, J = 8.4 Hz, 1H), 6.89
50 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
(d, J = 8.4 Hz, 1H), 6.57 (d, J = 16.4 Hz, 1H), 3.91 (s, 3H), 3.79 (s, 3H), 1.34 (s, 12H). 13C NMR (100
MHz, CDCl3): δ 168.08, 160.82, 140.24, 138.48, 135.95, 135.94, 123.02, 118.59, 110.56, 83.96, 55.62, 51.71, 25.02.
OMe OMe MeO OMe HO OH HO OH BPin Pd2(dba)3, SPhos, KF BBr ,CH Cl , Cl 3 2 2 NaOH, dioxane/MeOH, 1,4-dioxane/H2O, 110 ºC -78ºCtoRT 40 °C + 85% 51% 46%
9 10 O 14 O 22 O 15 OMe OMe OH O OMe
(E)-methyl 3-(5'-allyl-2',6-dimethoxy-[1,1'-biphenyl]-3-yl)acrylate (14) An oven dried Schlenk tube fitted with a stirring bar was charged with 9 (51.34 mg, 0.16 mmol) and 10 (27.56 mg, 0.15 mmol). The Schlenk tube was evacuated and backfilled with argon three times. A mixture of degassed 1,4-dioxane/H2O 10:1 (v/v) (0.6 mL) was added. Under a positive argon pressure KF
(39.17 mg, 0.67 mmol), Pd2(dba)3 (12.84 mg, 0.014 mmol) and SPhos (18.30 mg, 0.045 mmol) were added to this mixture. The mixture was stirred at 110 °C.5 After 18 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude was purified via column chromatography eluting with hexane/EtOAc 5% to 12% (v/v) to obtain novel compound 14 in 85% yield. Silica gel TLC Rf = 0.2 (Hexane/EtOAc 10% v/v); LC-MS (ESI):
1 calc. for C21H22O4 [M+H]: 339.40, observed 339.08, LC, Rt=7.93. HNMR (400 MHz, CDCl3): δ (ppm) 7.66 (d, J = 16 Hz, 1H), 7.47 (dd, J = 2.4, 8.4 Hz, 1H), 7.43 (d, J = 2.0 Hz, 1H), 7.15 (dd, J = 2.4, 8.4 Hz, 1H), 7.05 (d, J = 2 Hz, 1H), 6.95 (d, J = 8.4 Hz, 1H), 6.90 (d, J = 8.4 Hz, 1H), 6.30 (d, J = 16 Hz, 1H), 6.04-5.94 (m, 1H), 5.12-5.05 (m, 2H), 3.80 (s, 3H), 3.78 (s, 3H), 3.75 (s, 3H), 3.36
13 (d, J = 6.8 Hz, 2H); C NMR (100 MHz, CDCl3): δ 167.96, 159.12, 155.57, 144.86, 137.82, 132.04, 131.58, 131.31, 129.36, 129.02, 128.66, 126.98, 126.93, 115.78, 115.42, 111.27, 111.25, 56.00, 55.97, 51.69, 39.51.
(E)-methyl 3-(5'-allyl-2',6-dihydroxy-[1,1'-biphenyl]-3-yl)acrylate (22) To a solution of 14 (17.5 mg,
52.7 μmol) in dry CH2Cl2 (0.27 mL) in an oven dried flask, was added a solution of 1M BBr3 in dry
CH2Cl2 (0.15 mL, 2.5 eq.) dropwise at -78 ºC. The mixture was stirred for 60 minutes at -78 ºC, 60 min at 0 ºC and 5 min at room temperature under 1 atm. argon pressure. The reaction was quenched with H2O and stirred for 5 minutes at room temperature. The mixture was extracted with CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, passed through a paper filter and concentrated.7 The crude material was purified via column chromatography (eluent: hexane: EtOAc 10% to 40% v/v) to yield the novel compound 22 in 51% yield. Silica gel TLC Rf = 0.2 (Hexane/EtOAc 35% v/v); LC-MS (ESI): calc. for C19H18O4 [M+H]:
1 311.34, observed 311.17, LC, Rt=6.53. H NMR (400 MHz, CDCl3) δ 7.64 (d, J = 16 Hz, 1H), 7.45 (dd, J = 2.4, 8.4 Hz, 1H), 7.43 (d, J = 1.6 Hz, 1H), 7.14 (dd, J = 2.4, 8.4 Hz, 1H), 7.08 (d, J = 2 Hz, 1H), 7.02 (d, J = 8.4 Hz, 1H), 6.96 (d, J = 8.0 Hz, 1H), 6.31 (d, J = 16 Hz, 1H), 6.01-5.95 (m, 1H),
13 5.12-5.06 (m, 2H), 3.80 (s, 3H), 3.37 (d, J = 6.8 Hz, 2H); C NMR (100 MHz, CDCl3) δ 168.22,
51 CHAPTER 2
155.44, 151.09, 144.74, 137.48, 133.64, 131.86, 131.61, 130.34, 129.60, 127.89, 125.46, 123.55, 117.64, 116.88, 116.12, 115.76, 51.94, 39.46.
(E)-3-(5'-allyl-2',6-dihydroxy-[1,1'-biphenyl]-3-yl)acrylic acid (15) To a solution of 25 (8.2 mg, 26.4 μmol) in a mixture of 1,4-dioxane/MeOH 14:5 v/v (380 μL) was added 4N NaOH (3.5 eq., 26.1 μL) in one portion. The reaction was stirred at 40 ºC until the ester had been completely consumed as determined by TLC. The reaction was diluted with H2O and subsequently acidified with 1M HCl to pH 2. The mixture was concentrated and extracted with EtOAc. The obtained crude material was purified vie preparative HPLC, to yield novel compound 15 as a white solid in 46% 1 yield. LC-MS (ESI): calc. for C18H16O4 [M+H]: 297.32, observed 297.17, LC, Rt=5.40. HNMR
(400 MHz, CD3OD): δ (ppm) 7.63 (d, J = 16 Hz, 1H), 7.47 (d, J = 8.8 Hz, 1H), 7.44 (s, 1H), 7.05 (d, J = 6.0 Hz, 2H), 6.94 (d, J = 8.4 Hz, 1H), 6.85 (d, J = 8.8 Hz, 1H), 6.30 (d, J = 16 Hz, 1H), 6.04-5.93 (m, 1H), 5.05 (d, J = 18.8 Hz, 1H), 5.01 (d, J = 10.4 Hz, 1H), 3.34 (d, J = 6.4 Hz, 2H); 13C
NMR (100 MHz, CD3OD) δ 170.94, 157.96, 153.53, 146.60, 139.39, 133.28, 132.91, 132.71, 130.06, 129.68, 128.26, 127.78, 126.51, 117.77, 117.13, 116.09, 115.60, 40.39.
(E)-methyl 3-(3'-allyl-4',6-dimethoxy-[1,1'-biphenyl]-3-yl)acrylate (11) An oven dried Schlenk tube fitted with a stirring bar was charged with 9 (53.71 mg, 0.17 mmol) and 7 (27.56 mg, 0.17 mmol). The Schlenk tube was evacuated and backfilled with argon three times. A mixture of degassed
1,4-dioxane/H2O 10:1 (v/v) (0.6 mL) was added. Under a positive argon pressure KF (40.80 mg,
0.70 mmol), Pd2(dba)3 (12.30 mg, 13.4 μmol) and SPhos (17.08 mg, 41.6 μmol) were added to this mixture. The mixture was stirred at 110 °C. After 18 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude was purified via column chromatography eluting with hexane/EtOAc 5% to 12% (v/v) to obtain novel compound
11 in 84% yield. Silica gel TLC Rf = 0.2 (Hexane/EtOAc 10% v/v); LC-MS (ESI): calc. for C21H22O4
1 [M+H]: 339.40, observed 339.25, LC, Rt=7.85. HNMR (400 MHz, CDCl3): δ (ppm) 7.67 (d, J = 16.0 Hz, 1H), 7.47 (d, J = 2.4 Hz, 1H), 7.44 (dd, J = 2.4, 8.4 Hz, 1H), 7.35 (dd, J = 2.4, 8.4 Hz, 1H), 7.29 (d, J = 2.4 Hz, 1H), 6.95 (d, J = 8.4 Hz, 1H), 6.90 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 16.0 Hz, 1H), 6.08-5.98 (m, 1H), 5.12-5.04 (m, 2H), 3.87 (s, 3H), 3.84 (s, 3H), 3.80 (s, 3H), 3.42 (d, J = 6.8
13 Hz, 2H); C NMR (100 MHz, CDCl3): δ 167.91, 158.49, 156.84, 144.76, 136.99, 131.24, 131.03, 130.52, 129.88, 128.72, 128.47, 127.29, 115.72, 115.59, 111.40, 110.13, 55.88, 55.64, 51.74, 34.44.
(E)-methyl 3-(3'-allyl-4',6-dihydroxy-[1,1'-biphenyl]-3-yl)acrylate (23) To a solution of 11 (18.5 mg,
54.7 μmol) in dry CH2Cl2 (0.27 mL) in an oven dried flask, was added a solution of 1M BBr3 in dry
52 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
CH2Cl2 (0.15 mL, 2.5 eq.) dropwise at -78 ºC. The mixture was stirred for 60 minutes at -78 ºC, 60 min at 0 ºC and 5 min at room temperature under 1 atm. argon pressure. The reaction was quenched with H2O and stirred for 5 min at room temperature. The mixture was extracted with
CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated. LC-MS showed 40% conversion to the di-demethylated product and 60% mono- demethylated product. The crude mixture was re-dissolved in dry CH2Cl2 (0.30 mL), to this solution was added dropwise a 1M BBr3 in dry CH2Cl2 (0.15 mL, 2.5 eq.) at -78 ºC. The mixture was stirred at 0 ºC under 1 atm. argon pressure. After 2 h the reaction was quenched with H2O.
The mixture was extracted with CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated. LC-MS showed over 90% conversion. The product was purified via column chromatography (eluent: hexane: EtOAc 10% to 40% v/v) to yield 23 in
43% yield. Silica gel TLC Rf = 0.3 (Hexane/EtOAc 40% v/v); LC-MS (ESI): calc. for C19H18O4
1 [M+H]: 311.34, observed 311.17, LC, Rt=6.25. HNMR (400 MHz, CDCl3): δ (ppm) 7.64 (d, J = 16.0 Hz, 1H), 7.41 (dd, J = 2.4, 8.4 Hz, 1H), 7.38 (d, J = 2.4 Hz, 1H), 7.23-7.20 (m, 2H), 6.97 (d, J = 8.4 Hz, 1H), 6.94 (d, J = 8.0 Hz, 1H), 6.31 (d, J = 16.0 Hz, 1H), 6.10-5.99 (m, 1H), 5.25-5.20 (m, 2H),
3.79 (s, 3H), 3.47 (d, J = 6.4 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 167.93, 154.70, 154.52, 144.68, 135.90, 131.22, 130.45, 129.07, 128.60, 128.55, 128.50, 127.44, 126.85, 117.33, 116.96, 116.37, 115.59, 51.79, 35.30.
(E)-3-(3'-allyl-4',6-dihydroxy-[1,1'-biphenyl]-3-yl)acrylic acid (18) To a solution of 23 (7.2 mg, 23.2 μmol) in a mixture of 1,4-dioxane/MeOH 14:5 v/v (350 μL) was added a 4N NaOH (3.5 eq., 25.1 μL) in one portion. The reaction was stirred at 40 ºC until the ester had been completely consumed as determined by TLC. The reaction was diluted with H2O and subsequently acidified with 1M HCl to a pH of 2. The mixture was concentrated and extracted with EtOAc. The obtained crude material was purified via preparative HPLC, to yield novel compound 18 as a white solid in 1 50% yield. LC-MS (ESI): calc. for C18H16O4 [M+H]: 297.32, observed 297.17, LC, Rt=5.37. HNMR
(400 MHz, CD3OD): δ (ppm) 7.61 (d, J = 16 Hz, 1H), 7.43 (d, J = 2 Hz, 1H), 7.38 (dd, J = 2, 8.4 Hz, 1H), 7.28 (d, J = 2.4 Hz, 1H), 7.24 (dd, J = 2.4, 8.4 Hz, 1H), 6.89 (d, J = 8.4 Hz, 1H), 6.82 (d, J = 8.0 Hz, 1H), 6.30 (d, J = 16.0 Hz, 1H), 6.10-6.00 (m, 1H), 5.06 (d, J = 16.8 Hz, 1H), 5.01 (d, J
13 = 6.4 Hz, 1H), 3.40 (d, J = 6.4 Hz, 2H); C NMR (100 MHz, CD3OD) δ 171.06, 157.91, 155.47, 146.74, 138.45, 131.98, 131.94, 130.72, 130.50, 129.18, 128.94, 127.57, 127.41, 117.37, 115.78, 115.59, 115.43, 35.32.
53 CHAPTER 2
(E)-methyl 3-(5'-allyl-2',4-dimethoxy-[1,1'-biphenyl]-3-yl)acrylate (16) An oven dried Schlenk tube fitted with stirring bar was charged with 6 (104.5 mg, 0.32 mmol) and 10 (47.8 mg, 0.26 mmol). The Schlenk tube was evacuated and backfilled with argon three times. A mixture of degassed
1,4-dioxane/H2O 10:1 (v/v) (1.2 mL) was added. Under a positive argon pressure KF (77.62 mg,
1.34 mmol), Pd2(dba)3 (25.89 mg, 28.3 μmol) and SPhos (33.89 mg, 82.6 μmol) were added to this mixture. The mixture was stirred at 110 °C.5 After 18 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude was purified via column chromatography eluting with hexane/EtOAc 5% to 12% (v/v) to obtain novel compound
16 in 78% yield. Silica gel TLC Rf = 0.1 (Hexane/EtOAc 10% v/v); LC-MS (ESI): calc. for C21H22O4
1 [M+H]: 339.40, observed 339.08, LC, Rt=8.08. HNMR (400 MHz, CDCl3): δ (ppm) 8.01 (d, J = 16 Hz, 1H), 7.66 (s, 1H), 7.51 (d, J = 8.4 Hz, 1H), 7.11 (d, J = 8.4 Hz, 2H), 6.94 (d, J = 8.4 Hz, 1H), 6.90 (d, J = 7.6 Hz, 1H), 6.54 (d, J = 16.4 Hz, 1H), 6.03-5.93 (m, 1H), 5.12-5.05 (m, 2H), 3.92 (s,
13 3H), 3.80 (s, 3H), 3.79 (s, 3H), 3.37 (d, J = 6.4 Hz, 2H); C NMR (100 MHz, CDCl3) δ 168.11, 157.55, 154.95, 140.62, 137.78, 132.75, 132.52, 130.89, 130.19, 128.53, 123.06, 118.43, 115.80, 111.43, 110.92, 55.84, 55.72, 51.70, 39.50.
(E)-methyl 3-(5'-allyl-2',4-dihydroxy-[1,1'-biphenyl]-3-yl)acrylate (24) To a solution of 16 (18.5 mg,
54.7 μmol) in dry CH2Cl2 (0.27 mL) in an oven dried flask, was added a solution of 1M BBr3 in dry
CH2Cl2 (0.15 mL, 2.5 eq.) dropwise at -78 ºC. The mixture was stirred for 60 min at -78 ºC, 60 min at 0 ºC and 5 min at room temperature under 1 atm. argon pressure. The reaction was quenched with H2O and stirred for 5 min at room temperature. The mixture was extracted with
CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated.7 The crude material was purified via preparative HPLC, Gradient: 60% to 80% acetonitrile in 20 min. Peak collected between 10 and 11.5 minutes to obtain novel compound 24 in 7% yield. Silica gel TLC Rf = 0.3 (Hexane/EtOAc 35% v/v); LC-MS (ESI): calc. for C19H18O4
1 [M+H]: 311.34, observed 311.17, LC, Rt=6.50. HNMR (400 MHz, CDCl3): δ (ppm) 8.02 (d, J = 16.4 Hz, 1H), 7.59 (d, J = 2 Hz, 1H), 7.34 (dd, J = 2, 8.4 Hz, 1H), 7.09-7.03 (m, 2H), 6.94 (d, J = 8.4 Hz, 1H), 6.89 (d, J = 8 Hz, 1H), 6.64 (d, J = 16.4 Hz, 1H), 6.02-5.92 (m, 1H), 5.11-5.05 (m, 2H),
13 3.83 (s, 3H), 3.34 (d, J = 6.8 Hz, 2H); C NMR (100 MHz, CDCl3) δ 168.82, 155.06, 150.95, 140.48, 137.81, 132.59, 132.31, 130.36, 130.07, 129.94, 129.17, 127.29, 122.32, 118.81, 117.20, 116.03, 115.82, 52.07, 39.52.
(E)-3-(5'-allyl-2',4-dihydroxy-[1,1'-biphenyl]-3-yl)acrylic acid (17) To a solution of 24 (2.2 mg, 7.1 μmol) in a mixture of 1,4-dioxane/MeOH 14:5 v/v (100 μL) was added a 4N NaOH (3.5 eq., 15.0 μL) in one portion. The reaction was stirred at 40 ºC until the ester had been completely consumed as determined by TLC. The reaction was diluted with H2O and subsequently acidified with 1M HCl to a pH of 2. The mixture was concentrated, dissolved in H2O and extracted with
54 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
EtOAc to yield novel compound 17 as a yellow solid in quantitative yield with a purity >90%. LC- 1 MS (ESI): calc. for C18H16O4 [M+H]: 297.32, observed 297.00, LC, Rt=5.82. HNMR (400 MHz,
CD3OD): δ (ppm) 7.98 (d, J = 16.0 Hz, 1H), 7.68 (d, J = 2.4 Hz, 1H), 7.41 (dd, J = 2.4, 10.8 Hz, 1H), 7.05 (d, J = 2.4 Hz, 1H), 6.94 (dd, J = 2.4, 8.4 Hz, 1H), 6.88 (d, J = 8.4 Hz, 1H), 6.80 (d, J = 8.0 Hz, 1H), 6.56 (d, J = 16.0 Hz, 1H), 6.04-5.94 (m, 1H), 5.10-5.01 (m, 2H), 3.34 (d, J = 8.4 Hz,
13 2H); C NMR (100 MHz, CD3OD) δ 171.44, 157.11, 153.51, 142.60, 139.54, 133.61, 132.63, 131.41, 130.80, 129.17, 129.04, 122.21, 118.82, 116.96, 116.65, 115.46, 40.43.
(E)-methyl 3-(3'-allyl-6-methoxy-[1,1'-biphenyl]-3-yl)acrylate (12) An oven dried Schlenk tube fitted with a stirring bar was charged with 9 (100 mg, 0.314 mmol) and 1-allyl-3-bromobenzene (54.5 mg, 0.28 mmol). The Schlenk tube was evacuated and backfilled with argon three times.
Degassed 1,4-dioxane/H2O 10:1 (v/v) (1.2 mL) was added. Under a positive argon pressure KF
(78.5 mg, 1.35 mmol), Pd2(dba)3 (24.96 mg, 27.3 μmol) and SPhos (34.8 mg, 84.8 μmol) were added to this mixture. The tube was closed and the mixture was stirred at 110 °C. After 7 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude was purified via column chromatography eluting with hexane/EtOAc 5% to
15% (v/v) to obtain novel compound 12 in 94% yield. Silica gel TLC Rf = 0.2 (Hexane/EtOAc 10% 1 v/v); LC-MS (ESI): calc. for C20H20O3 [M+H]: 309.37, observed 309.17, LC, Rt=7.22. HNMR (400
MHz, CDCl3): δ (ppm) 7.67 (d, J = 16.0 Hz, 1H), 7.50 (s, 2H), 7.37-7.32 (m, 3H), 6.97 (d, J = 9.2 Hz, 1H), 6.33 (d, J = 16 Hz, 1H), 6.07-5.97 (m, 1H), 5.15-5.08 (m, 2H), 3.85 (s, 3H), 3.80 (s, 3H),
13 3.45 (d, J = 6.8 Hz, 2H); C NMR (100 MHz, CDCl3) δ 167.88, 158.47, 144.64, 140.07, 137.91, 137.46, 131.47, 130.69, 129.79, 129.18, 128.54, 128.27, 127.76, 127.43, 125.58, 115.72, 111.46, 55.90, 51.75, 40.41.
(E)-methyl 3-(3'-allyl-6-hydroxy-[1,1'-biphenyl]-3-yl)acrylate (25) To a solution of 12 (91.4 mg, 0.3 mmol) in dry CH2Cl2 (1.3 mL) in an oven dried flask, was added a solution of 1M BBr3 in dry
CH2Cl2 (0.6 mL, 2.0 eq.) dropwise at -78 ºC. The mixture was stirred for 60 min at -78 ºC, 60 min at 0 ºC and 5 min at room temperature under 1 atm. argon pressure. The reaction was quenched with H2O and stirred for 5 min at room temperature. The mixture was extracted with
CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated.7 The crude material was purified via column chromatography (eluent: hexane:
EtOAc 10% to 50% v/v) to yield novel compound 25 in 46% yield. Silica gel TLC Rf = 0.4
(Hexane/EtOAc 25% v/v); LC-MS (ESI): calc. for C19H18O3 [M+H]: 295.34, observed 295.17, LC,
1 Rt=6.52. HNMR (400 MHz, CDCl3): δ (ppm) 7.65 (d, J = 16.0 Hz, 1H), 7.46-7.42 (m, 4H), 7.03-
55 CHAPTER 2
7.27 (m, 2H), 7.99 (d, J = 8.4 Hz, 1H), 6.31 (d, J = 16 Hz, 1H), 6.05-5.94 (m, 1H), 5.16-5.10 (m,
13 2H), 3.79 (s, 3H), 3.45 (d, J = 6.8 Hz, 2H); C NMR (100 MHz, CDCl3) δ 167.94, 154.70, 144.65, 141.73, 136.95, 136.36, 130.48, 129.66, 129.31, 129.30, 128.83, 128.76, 127.44, 126.76, 116.58, 116.54, 115.63, 51.79, 40.30.
(E)-3-(3'-allyl-6-hydroxy-[1,1'-biphenyl]-3-yl)acrylic acid (2) To a solution of 25 (40.1 mg, 0.14 mmol) in a mixture of 1,4-dioxane/MeOH 14:5 v/v (1.9 mL) was added a 4N NaOH (3.5 eq., 119 μL) in one portion. The reaction was stirred at 40 ºC until the ester had been completely consumed as determined by TLC. The reaction was diluted with H2O and subsequently acidified with 1M HCl to a pH of 2. The mixture was concentrated, dissolved in H2O and extracted with EtOAc. The obtained crude material was purified vie preparative HPLC (product obtained between 14 to 15 minutes), to yield compound 2 as a white solid in 33% yield. LC-MS (ESI): calc.
1 for C18H16O3 [M+H]: 280.32, observed 281.17, LC, Rt=5.80. HNMR (400 MHz, CD3OD): δ (ppm) 7.63 (d, J = 16.0 Hz, 1H), 7.43 (d, J = 8.4 Hz, 1H), 7.37 (s, 2H), 7.32 (t, J = 7.2 Hz, 1H), 7.14 (d, J = 7.6 Hz, 1H), 6.92 (d, J = 8.0 Hz, 1H), 6.30 (d, J = 16.0 Hz, 1H), 6.06-5.96 (m, 1H), 5.13-5.05 (m,
13 2H), 3.42 (d, J = 6.4 Hz, 2H); C NMR (100 MHz, CD3OD) δ 170.95, 158.00, 146.58, 141.11, 139.51, 138.95, 132.21, 130.62, 129.60, 129.13, 128.30, 128.16, 127.61, 117.45, 115.99, 115.95, 41.24.
(E)-3-(3-bromophenyl)acrylic acid (commercial) Silica gel TLC Rf = 0.2 (Hexane/EtOAc 25% v/v);
1 H NMR (400 MHz, CDCl3) δ (ppm) 7.70 (d, J = 16.0 Hz, 1H), 7.70 (t, J = 1.8 Hz, 1H), 7.54 (ddd, J = 8.0, 2.0, 1.0 Hz, 1H), 7.47 (ddd, J = 7.9, 1.6, 0.8 Hz, 1H), 7.29 (t, J = 7.9 Hz, 1H), 6.45 (d, J = 16.0 Hz, 1H).
(E)-methyl 3-(3-bromophenyl)acrylate (26)56 (E)-3-(3-bromophenyl)acrylic acid (1.0 g, 4.4 mmol) and K2CO3 (608 mg, 4.4 mmol) were suspended in DMF (4.4 mL). MeI (410 μL, 6.6 mmol) was added and the reaction was stirred at room temperature until the starting material was completely consumed. The reaction was quenched with NH4Cl and H2O, the aqueous layer was extracted with Et2O three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated to yield 26 as off white crystals in quantitative yield. Silica gel TLC Rf =
1 0.8 (Hexane/EtOAc 25% v/v); H NMR (400 MHz, CDCl3) δ (ppm) 7.67 (t, J = 1.8 Hz, 1H), 7.61 (d, J = 16.0 Hz, 1H), 7.51 (ddd, J = 7.9, 2.0, 1.0 Hz, 1H), 7.44 (dt, J = 7.8, 1.3 Hz, 1H), 7.26 (t, J = 15.7 Hz, 1H), 6.43 (d, J = 16.0 Hz, 1H), 3.81 (s, 3H).
56 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
(E)-methyl 3-(3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate (27)57 An oven dried schlenk tube was charged with 26 (90 mg, 0.37 mmol, 1 eq.), KOAc (147 mg, 1.49 mmol, 4 eq.),
Bis(pinacolato)diboron (143 mg, 0.56 mmol, 1.5 eq.) and Pd(dppf)Cl2 (10.4 mg, 11.2 μmol, 3 %). The schlenk tube was evacuated and backfilled with argon three times. Dioxane (1.8 mL) was added and the reaction was stirred at 80 °C. After 16 h the mixture was allowed to cool to room temperature and passed through celite with EtOAc. The product was purified via column chromatography eluting with 10 % v/v in hexane. Compound 27 was obtained in 90 % yield.
Silica gel TLC Rf = 0.5 (Hexane/EtOAc 20% v/v); LC-MS (ESI): calc. for C16H21BO4 [M+H]:
1 288.15, observed 289.17, LC, Rt=6.40. H NMR (400 MHz, CDCl3) δ (ppm) 7.98 (t, J = 1.4 Hz, 1H), 7.81 (dd, J = 7.4, 1.3 Hz, 1H), 7.71 (d, J = 16.0 Hz, 1H), 7.60 (dt, J = 7.8, 1.6 Hz, 1H), 7.39 (t, J = 7.6 Hz, 1H), 6.50 (d, J = 16.0 Hz, 1H), 3.80 (s, 3H), 1.36 (s, 12H).
(E)-methyl 3-(3'-allyl-[1,1'-biphenyl]-3-yl)acrylate (13) An oven dried schlenk tube was charged with compound 27 (50 mg, 0.17 mmol, 1.2 eq.), 1-allyl-3-bromobenzene (34.9 mg, 0.15 mmol, 1 eq.) in dioxane/H2O (10/1 v/v, 0.8 mL) to this solution was added Pd2(dba)3 (13.2 mg, 14.5 μmol, 10%), SPhos (17.8 mg, 43.4 μmol, 30%) and KF (42.0 mg, 0.72 mmol, 5eq.). The schlenk tube was closed and the reaction was stirred at 100 °C. After 6 h the mixture was cooled to room temperature and passed through celite, eluting with EtOAc. The crude material was purified via column chromatography eluting with hexane/EtOAc 5 to 10 % v/v to yield compound 13 in 86 % yield. Silica gel TLC Rf = 0.5 (Hexane/EtOAc 15% v/v);
(E)-3-(3'-allyl-[1,1'-biphenyl]-3-yl)acrylic acid (19) 4N NaOH (3.5 eq., 109.7 μL) was added to a solution of 13 (34.9 mg, 0.125 mmol) in a mixture of 1,4-dioxane/MeOH 14:5 v/v (1.6 mL). The reaction was stirred at 40 °C until the ester had been completely consumed as determined by
TLC. The reaction was diluted with H2O and subsequently acidified with 1M HCl to pH 2. The mixture was concentrated and extracted with EtOAc. The obtained crude material was purified vie preparative HPLC. Novel compound 19 was obtained as a white solid in 75% yield. LC-MS (ESI):
1 calc. for C18H16O2 [M+H]: 265,32, observed 265.08, LC, Rt=6.37. H NMR (400 MHz, CDCl3) δ (ppm) 7.87 (d, J = 16.0 Hz, 1H), 7.75 (s, 1H), 7.63 (d, J = 7.6 Hz, 1H), 7.54 (d, J = 7.8 Hz, 1H), 7.48 (d, J = 7.6 Hz, 1H), 7.48 – 7.35 (m, 3H), 7.22 (d, J = 7.3 Hz, 1H), 6.02 (ddt, J = 16.8, 10.0, 6.7 Hz,
13 1H), 5.20 – 5.07 (m, 2H), 3.48 (d, J = 6.7 Hz, 2H); C NMR (100 MHz, CDCl3) δ 171.71, 147.16, 142.28, 140.93, 140.63, 137.31, 134.65, 129.74, 129.51, 129.14, 128.20, 127.59, 127.37, 127.19, 125.11, 117.62, 116.32, 40.42.
57 CHAPTER 2
4-chloro-2-isobutyl-1-methoxybenzene (30) In an oven dried Schlenk flask 5 (250 mg, 1.26 mmol), cesiumfluoride (343 mg, 2.26 mmol), tetrakis(triphenyl-phosphine)palladium(0) (130 mg, 0.15 mmol) were charged. The Schlenk flask was evacuated and backfilled with argon three times. Dry and degassed THF (8 mL) was added under positive argon pressure and the mixture was stirred at room temperature under positive argon pressure. After 30 minutes (2-methylpropyl)boronic acid (606 mg, 5.95 mmol) in dry THF (1 mL) was added. The schlenk flask was closed with a Teflon screwcap and the reaction mixture was stirred for 21 h at 78 °C. Under positive argon pressure a second portion of cesium fluoride (343 mg, 2.26 mmol) and tetrakis(triphenylphosphine)-palladium(0) (130 mg, 0.15 mmol) and THF (1 mL) were added. The reaction mixture was stirred at 78 °C. After 24 h the mixture was allowed to cool to room temperature and the mixture was separated between H2O (30 mL) and pentane (30 mL). The aqueous layer was washed with pentane (2 x 10 mL). the combined organic layers were washed with brine (10 mL), dried over Na2SO4, filtered and concentrated to yield a yellow oil. The crude material was purified via column chromatography eluting with Hexane/EtOAc 1.5% (v/v) to yield compound 30 as colorless oil, 149 mg, 75%. Silica gel TLC Rf = 0.5 (Hexane/EtOAc 5% v/v);
1 HNMR (400 MHz, CDCl3): δ (ppm) δ 7.12 (q, J=8.8Hz, 1H) 7.06 (d, J=2.4Hz, 1H) 6.75 (d, J=8.8Hz, 1H) 3.79 (s, 3H) 2.45 (d, J=7.2Hz, 2H) 1.90 (s, 1H) 0.89 (d, J=6.8Hz, 6H); 13C NMR
(100 MHz, CDCl3) δ 132.07, 130.44, 126.37, 111.33, 55.51, 39.17, 29.70, 28.54, 26.91, 22.45.
2-chloro-4-isobutyl-1-methoxybenzene (28) In an oven dried Schlenk flask 8 (250 mg, 1.26 mmol), cesiumfluoride (343 mg, 2.26 mmol), tetrakis(triphenyl-phosphine)palladium(0) (130 mg, 0.15 mmol) were charged. The Schlenk flask was capped with a rubber septum and then evacuated and backfilled with argon three times. Dry and degassed THF (7 mL) was added via a syringe trough the septum and the mixture was stirred at room temperature under positive argon pressure. After 30 minutes (2-methylpropyl)boronic acid (606 mg, 5.95 mmol) in dry THF (1 mL) was added. The schlenk flask was closed with a Teflon screwcap and the reaction mixture was stirred for 25 h at 78 °C. Under positive argon pressure a second portion of cesium fluoride (343 mg, 2.26 mmol) and tetrakis(triphenylphosphine)-palladium(0) (130 mg, 0.15 mmol) and THF (1 mL) were added. The reaction mixture was stirred at 78 °C. After 24 h the mixture was allowed to cool to room temperature and the mixture was separated between H2O (100 mL) and pentane
58 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
(100 mL). The aqueous layer was washed with pentane (2 x 80 mL). The combined organic layers were washed with brine (100 mL), dried over Na2SO4, filtered and concentrated to yield yellow oil. The crude material was purified via column chromatography eluting with Hexane/EtOAc 3%
(v/v) to yield 28 as colorless oil, 122 mg, 61%. Silica gel TLC Rf = 0.5 (Hexane/EtOAc 5% v/v);
1 HNMR (400 MHz, CDCl3): δ (ppm) δ 7.12 (q, J=8.4Hz, 1H) 7.05 (d, J=2.8Hz, 1H) 6.74 (d, J=8.8Hz, 1H) 3,78 (s, 3H) 2.45 (d, J=7.2Hz, 2H) 1.87 (m, 1H) 0.89 (d, J=6.4Hz, 6H); 13C NMR
(100 MHz, CDCl3) δ 156.29, 132.08, 130.43, 126.37, 124.86, 111.34, 55.52, 39.16, 28.54, 22.45.
2-(5-isobutyl-2-methoxyphenyl)-4,4,5,5,-tetramethyl-1,3,2-dioxaborolane (29) K3PO4 (190 mg, 0.90 mmol), SPhos (5 mg, 0.012 mmol), PdOAc2 (1 mg, 4.5 μmol) and Bis(pinacolato)diboron (190 mg, 0.75 mmol) were charged in an over dried schlenk tube. The tube was evacuated and backfilled three times with argon and dry 1,4-dioxane was added. The mixture was stirred for 5 min at room temperature. Compound 28 (50 mg, 0.25 mmol) in 1,4-dioxane (0.7 ml) was added. The mixture was stirred at RT for 48 h. The mixture was allowed to cool to room temperature and was passed through celite, eluting with EtOAc. The crude mixture was purified via column chromatography (eluens was Hexane:EtOAc 6% to 15% v/v) to obtain compound 29 (72 mg, 0.25
1 mmol). Silica gel TLC Rf = 0.3 (Hexane/EtOAc 10% v/v); HNMR (400 MHz, CDCl3): δ (ppm) δ 7.64 (d, J=8Hz, 1H) 7.53 (s, 1H) 6.83 (d, J=8Hz, 1H) 3.82 (s, 3H) 2.48 (d, J=6.8Hz, 2H) 1.92 (m,
13 1H) 1.33 (s, 12H) 0.88 (d, J=6,4H, 6H); C NMR (100 MHz, CDCl3) δ 137.36, 134.29, 130.43, 129.53, 126.36, 109.54, 83.47, 55.12, 39.19, 28.83, 26.90, 25.01, 24.85, 22.58, 22.45. LC-MS (ESI): calc. for C17H27BO3 [M+H]: 291.21, observed 291.17, LC, Rt=7.73
3’,5-diisobutyl-2,4’-dimethoxy-1,1’-biphenyl (31) In an oven dried Schlenk tube 29 (72 mg, 0.4 mmol) and 30 (80 mg, 0.4 mmol) were weighed. The Schlenk tube was evacuated and backfilled with argon three times. A mixture of degassed 1,4-dioxane/H2O 10:1 (v/v) (0.6 mL) was added.
Under a positive argon pressure KF (100.2 mg, 1.725 mmol), Pd2(dba)3 (35.7 mg, 0.0345 mmol) and SPhos (42.5 mg, 0,104 mmol) were added to this mixture. The mixture was stirred at 110 °C. After 18 h the mixture was allowed to cool to room temperature and passed through celite, eluting with EtOAc. The obtained crude was purified via column chromatography eluting with hexane/EtOAc 5% to 12% (v/v) to obtain novel compound 31 as an inseparable mixture with compound 29 in a molar ratio of 7:3. The material was carried forward in the synthesis. (52 mg).
1 Silica gel TLC Rf = 0.4 (Hexane/EtOAc 10% v/v); HNMR (400 MHz, CDCl3): δ (ppm) δ 7.34 (d, J = 9.6 Hz, 1H) 7.28 (s, 1H), 7.15 (s, 1H), 7.07-6-97 (m, 2H), 6.88-6.86 (m, 1H), 3.87 (s, 3H), 3.83 (s, 3H), 2.53 (d, J = 6.8 Hz, 2H), 2.38 (d, J = 7.2 Hz, 2H), 1.99-176 (m, 2H), 0.94-0.89 (m, 12H).
3’,5-diisobutyl-[1,1’-biphenyl]-2,4’-diol (3) To a solution of 31 (52 mg) in dry CH2Cl2 (0.82 mL) in an oven dried flask, was added a solution of 1M BBr3 in dry CH2Cl2 (0.455 mL, 2.5 eq.) dropwise at -78 ºC. The mixture was stirred for 60 minutes at -78 ºC, 60 min at 0 ºC and 5 min at room
59 CHAPTER 2
temperature under 1 atm. argon pressure. The reaction was quenched with H2O and stirred for 5 min at room temperature. The mixture was extracted with CH2Cl2; the combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated. The crude material was purified via preparative HPLC (60 to 90 % Acetonitrile) to yield 3 as a white solid (4.2 mg, 0.014 mmol, 5.6 % yield from 29). LC-MS (ESI): calc. for C20H26O2 [M+H]: 299.19, observed 299.25,
1 LC, Rt=7.22 HNMR (400 MHz, CDCl3): δ (ppm) δ 7.17 (s, 2H) 6,99 (t, J = 4.8 Hz, 2H), 6,87 (d, J=8.0 Hz, 2H) 5.08 (s, 1H) 4.77 (s, 1H) 2.53 (d, J=7.2 Hz, 2H) 2.42 (d, J=7.2 Hz, 2H) 1.97 (m, 1H)
13 1.84 (m, 1H) 0.94 (q, J = 19,8 Hz, 12H); C NMR (100 MHz, CDCl3) δ 153.39, 150.41, 133.85, 131.87, 130.63, 129.35, 129.24, 128.43, 127.76, 127.42, 116.01, 115.13, 44.58, 39.27, 30.32, 28.87, 22.56, 22.35.
References
1. Majmudar, C. Y. et al. Sekikaic acid and lobaric acid target a dynamic interface of the coactivator CBP/p300. Angew. Chem. Int. Ed Engl. 51, 11258–11262 (2012). 2. Wang, N. et al. Ordering a dynamic protein via a small-molecule stabilizer. J. Am. Chem. Soc. 135, 3363–3366 (2013). 3. Huang, P., Chandra, V. & Rastinejad, F. Structural overview of the nuclear receptor superfamily: insights into physiology and therapeutics. Annu. Rev. Physiol. 72, 247–272 (2010). 4. Wurtz, J. M. et al. A canonical structure for the ligand-binding domain of nuclear receptors. Nat. Struct. Biol. 3, 87–94 (1996). 5. Nagy, L. & Schwabe, J. W. R. Mechanism of the nuclear receptor molecular switch. Trends Biochem. Sci. 29, 317–324 (2004). 6. Moore, J. T., Collins, J. L. & Pearce, K. H. The Nuclear Receptor Superfamily and Drug Discovery. ChemMedChem 1, 504–523 (2006). 7. Moore, T. W., Mayne, C. G. & Katzenellenbogen, J. A. Minireview: Not picking pockets: nuclear receptor alternate-site modulators (NRAMs). Mol. Endocrinol. Baltim. Md 24, 683–695 (2010). 8. Choi, J. H. et al. Antidiabetic actions of a non-agonist PPARγ ligand blocking Cdk5-mediated phosphorylation. Nature 477, 477–481 (2011). 9. Sadana, P. Noncanonical mechanisms to regulate nuclear receptor signaling. Future Med. Chem. 4, 1307–1333 (2012). 10.Smith, C. L. & O’Malley, B. W. Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr. Rev. 25, 45–71 (2004). 11. Geistlinger, T. R. & Guy, R. K. Novel selective inhibitors of the interaction of individual nuclear hormone receptors with a mutually shared steroid receptor coactivator 2. J. Am. Chem. Soc. 125, 6852–6853 (2003). 12. Phillips, C. et al. Design and Structure of Stapled Peptides Binding to Estrogen Receptors. J. Am. Chem. Soc. 133, 9696–9699 (2011). 13. Arnold, L. A. et al. Discovery of small molecule inhibitors of the interaction of the thyroid hormone receptor with transcriptional coregulators. J. Biol. Chem. 280, 43048–43055 (2005). 14. Arnold, L. A., Kosinski, A., Estébanez-Perpiñá, E., Fletterick, R. J. & Guy, R. K. Inhibitors of the interaction of a thyroid hormone receptor and coactivators: preliminary structure-activity relationships. J. Med. Chem. 50, 5269–5280 (2007).
60 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
15. Gunther, J. R., Parent, A. A. & Katzenellenbogen, J. A. Alternative inhibition of androgen receptor signaling: peptidomimetic pyrimidines as direct androgen receptor/coactivator disruptors. ACS Chem. Biol. 4, 435–440 (2009). 16.Nandhikonda, P. et al. Discovery of the first irreversible small molecule inhibitors of the interaction between the vitamin D receptor and coactivators. J. Med. Chem. 55, 4640–4651 (2012). 17. Rodriguez, A. L., Tamrazi, A., Collins, M. L. & Katzenellenbogen, J. A. Design, Synthesis, and in Vitro Biological Evaluation of Small Molecule Inhibitors of Estrogen Receptor α Coactivator Binding. J. Med. Chem. 47, 600–611 (2004). 18. Estébanez-Perpiñá, E. et al. A surface on the androgen receptor that allosterically regulates coactivator binding. Proc. Natl. Acad. Sci. U. S. A. 104, 16074–16079 (2007). 19.Sun, A. et al. Discovering small-molecule estrogen receptor α/coactivator binding inhibitors: high-throughput screening, ligand development, and models for enhanced potency. ChemMedChem 6, 654–666 (2011). 20.Caboni, L. et al. ‘True’ antiandrogens-selective non-ligand-binding pocket disruptors of androgen receptor-coactivator interactions: novel tools for prostate cancer. J. Med. Chem. 55, 1635–1644 (2012). 21. Over, B. et al. Natural-product-derived fragments for fragment-based ligand discovery. Nat. Chem. 5, 21–28 (2013). 22.de Lera, A. R., Bourguet, W., Altucci, L. & Gronemeyer, H. Design of selective nuclear receptor modulators: RAR and RXR as a case study. Nat. Rev. Drug Discov. 6, 811–820 (2007). 23.Jayalakshmi, V. & Krishna, N. R. Complete relaxation and conformational exchange matrix (CORCEMA) analysis of intermolecular saturation transfer effects in reversibly forming ligand-receptor complexes. J. Magn. Reson. San Diego Calif 1997 155, 106–118 (2002). 24.Krishna, N. R. & Jayalakshmi, V. Quantitative Analysis of STD-NMR Spectra of Reversibly Forming Ligand-Receptor Complexes. Top. Curr. Chem. 273, 15–54 (2008). 25.Liby, K. T., Yore, M. M. & Sporn, M. B. Triterpenoids and rexinoids as multifunctional agents for the prevention and treatment of cancer. Nat. Rev. Cancer 7, 357–369 (2007). 26.Altucci, L., Leibowitz, M. D., Ogilvie, K. M., Lera, A. R. de & Gronemeyer, H. RAR and RXR modulation in cancer and metabolic disease. Nat. Rev. Drug Discov. 6, 793–810 (2007). 27.Cramer, P. E. et al. ApoE-Directed Therapeutics Rapidly Clear β-Amyloid and Reverse Deficits in AD Mouse Models. Science 335, 1503–1506 (2012). 28.Barnard, J. H., Collings, J. C., Whiting, A., Przyborski, S. A. & Marder, T. B. Synthetic retinoids: structure-activity relationships. Chem. Weinh. Bergstr. Ger. 15, 11430–11442 (2009). 29.Kotani, H., Tanabe, H., Mizukami, H., Makishima, M. & Inoue, M. Identification of a naturally occurring rexinoid, honokiol, that activates the retinoid X receptor. J. Nat. Prod. 73, 1332–1336 (2010). 30.Jung, C.-G. et al. Honokiol increases ABCA1 expression level by activating retinoid X receptor beta. Biol. Pharm. Bull. 33, 1105–1111 (2010). 31. Lee, Y.-J. et al. Therapeutic applications of compounds in the Magnolia family. Pharmacol. Ther. 130, 157–176 (2011). 32.Banerjee, P., Basu, A., Arbiser, J. L. & Pal, S. The natural product honokiol inhibits calcineurin inhibitor-induced and Ras-mediated tumor promoting pathways. Cancer Lett. 338, 292–299 (2013). 33. Singh, T. & Katiyar, S. K. Honokiol inhibits non-small cell lung cancer cell migration by targeting PGE₂-mediated activation of β-catenin signaling. PloS One 8, e60749 (2013).
61 CHAPTER 2
34.Kumar, A., Kumar Singh, U. & Chaudhary, A. Honokiol analogs: a novel class of anticancer agents targeting cell signaling pathways and other bioactivities. Future Med. Chem. 5, 809–829 (2013). 35. Murray, C. W. & Rees, D. C. The rise of fragment-based drug discovery. Nat. Chem. 1, 187–192 (2009). 36.Horton, D. A., Bourne, G. T. & Smythe, M. L. The combinatorial synthesis of bicyclic privileged structures or privileged substructures. Chem. Rev. 103, 893–930 (2003). 37. Koppen, A. et al. Nuclear receptor-coregulator interaction profiling identifies TRIP3 as a novel peroxisome proliferator-activated receptor gamma cofactor. Mol. Cell. Proteomics MCP 8, 2212– 2226 (2009). 38.Feng, B. Y. & Shoichet, B. K. A detergent-based assay for the detection of promiscuous inhibitors. Nat. Protoc. 1, 550–553 (2006). 39.Nahoum, V. et al. Modulators of the structural dynamics of the retinoid X receptor to reveal receptor function. Proc. Natl. Acad. Sci. 104, 17323–17328 (2007). 40.Abad-Zapatero, C. Ligand efficiency indices for effective drug discovery. Expert Opin. Drug Discov. 2, 469–488 (2007). 41. Billingsley, K. L., Barder, T. E. & Buchwald, S. L. Palladium-catalyzed borylation of aryl chlorides: scope, applications, and computational studies. Angew. Chem. Int. Ed Engl. 46, 5359–5363 (2007). 42.Fuchs, S. et al. Proline Primed Helix Length as a Modulator of the Nuclear Receptor– Coactivator Interaction. J. Am. Chem. Soc. 135, 4364–4371 (2013). 43.Karplus, P. A. & Diederichs, K. Linking Crystallographic Model and Data Quality. Science 336, 1030–1033 (2012). 44.Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213–221 (2010). 45.Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132 (2010). 46.Evans, P. Scaling and assessment of data quality. Acta Crystallogr. D Biol. Crystallogr. 62, 72– 82 (2006). 47.Egea, P. F., Mitschler, A. & Moras, D. Molecular recognition of agonist ligands by RXRs. Mol. Endocrinol. Baltim. Md 16, 987–997 (2002). 48.McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007). 49.Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132 (2004). 50.Rama Krishna, N. & Jayalakshmi, V. Complete relaxation and conformational exchange matrix analysis of STD-NMR spectra of ligand–receptor complexes. Prog. Nucl. Magn. Reson. Spectrosc. 49, 1–25 (2006). 51. Garcia De La Torre, J., Huertas, M. L. & Carrasco, B. Calculation of hydrodynamic properties of globular proteins from their atomic-level structure. Biophys. J. 78, 719–730 (2000). 52.Brunner, H., de Courcy, N. L. C. & Genêt, J.-P. Application of a new combination of palladium and CaCO3 for an aerobic Heck reaction using arenediazonium-salts. Tetrahedron Lett. 40, 4815–4818 (1999). 53. Borah, A. J., Goswami, P., Barua, N. C. & Phukan, P. Synthesis of unit-B of cryptophycin-24 via Sharpless asymmetric dihydroxylation. Tetrahedron Lett. 53, 7128–7130 (2012). 54.Billingsley, K. L. & Buchwald, S. L. An improved system for the palladium-catalyzed borylation of aryl halides with pinacol borane. J. Org. Chem. 73, 5589–5591 (2008). 55. Chang, G. et al. Dibenzyl Amine Compounds and Derivatives. (2006). 56.Ambler, B. R. & Altman, R. A. Copper-Catalyzed Decarboxylative Trifluoromethylation of Allylic Bromodifluoroacetates. Org. Lett. 15, 5578–5581 (2013).
62 A NATURAL-PRODUCT SWITCH FOR A DYNAMIC PROTEIN INTERFACE
57. Morandi, S., Morandi, F., Caselli, E., Shoichet, B. K. & Prati, F. Structure-based optimization of cephalothin-analogue boronic acids as β-lactamase inhibitors. Bioorg. Med. Chem. 16, 1195– 1205 (2008).
Lidia Nieto performed the detailed NMR experiments with assistance of Hans Wienk and Rolf Boelens modeling studies were performed by Anna K. H. Hirsch and Lech-Gustav Milroy. Seppe Leysen processed and refined the crystal data.
63
64 Chapter 3
Ligand-induced helix repositioning for either selective RXR heterodimerization or coactivator binding
The conformation of the ligand binding domain of nuclear receptors determines the functional physiological outcome. Conformational changes in this ligand binding domain can be induced via small hydrophobic ligands. The retinoid X receptor (RXR) plays a key role in many physiological processes through heterodimerization with other nuclear receptors, but a limited number of small molecules for RXR modulation is available with a restricted chemical diversity and functional outcome. Building on the knowledge and chemistry described in chapter two, this chapter describes the synthesis, biochemical evaluation and structure elucidation of RXR ligands targeting the ligand binding pocket. Targeted modifications to the biaryl scaffold cause local side-chain disturbances and correlated displacements of secondary structural elements on ligand binding, which translate into either the selective repositioning of helices in the heterodimer interface of RXR or towards the activation function 2 (AF2) leading to partial activation properties. The data provide a rationale for the design of RXR ligands built of an unique hydrophilic region with a conserved hydrogen bonding network contributing to the ligand affinity, and an interchangeable hydrophobic region to probe the other parts of the pocket influencing coregulatory recuitment or the RXR dimerization interface. Together this work demonstrates a comprehensive molecular mechanism for RXR modulation which might correspond to a more general conserved mechanisms in the nuclear receptor protein family and might aid in efforts to make selective nuclear receptor modulators.
65 CHAPTER 3
Introduction
The retinoid X receptor (RXR) is a nuclear receptor whose activity is influenced via the class of compounds closely related to 9-cis-retinoic acid (9cRA). The putative endogenous ligand for RXR 9cRA is an isomerized product of all-trans retinoic acid (ATRA) which is a metabolite from vitamin A (retinol) and is essential in embryogenesis and adult homeostasis.1 ATRA binds and activates the retinoic acid receptors (RAR), which possess a linear-shaped ligand binding pocket, while 9cRA acts as an agonist for RXR and binds in the more contorted L-shaped ligand binding pocket of RXR.2 The L-shaped ligand binding pocket of RXR is unique to RXR, and aids in the design of specific ligands for this receptor. The activity of the retinoic acids is limited by the isomerization and oxidative metabolism, which affords several several hydroxy and epoxy retinoic acids.3,4 Therefore there is an interest to develop synthetic and more stable RXR modulators. There are many comprehensive overviews of the structure, biology and therapeutic implications of RXR and their modulators.2,5–7 However, the chemical diversity and functional outcome of these modulators is arguably limited owing to the molecular constraints placed by the ligand binding pocket on ligand design and the synthetic methodology to make such ligands. Unmet biomedical challenges such a RXR heterodimerization or partial agonism still remain, both of which can in principle be addressed through ligand binding to the ligand binding pocket, which motivates further medchem efforts in this direction. Structural information derived from X-ray crystallographic data of the RXR ligand binding domain in the apo- or holo- state8,9 has helped enormously in the design of new ligands and explains some of ligand-induced conformational changes of RXR. The generic structure of retinoids consist out of three domains, i.e. a hydrophobic unit, a contorted “linker” that fit the L-shaped pocket of RXR and a polar anchor (figure 3.1).
Figure 3.1 | Generic structure of RXR ligands. Structure of 9-cis-retinoic acid and bexarotene, both approved for therapy. RXR ligands consist in general out of three
66 LIGAND BINDING POCKET RXR
domains, a hydrophobic region, a twisted linked to fill the L-shaped pocket and a polar anchor which is almost exclusively a carboxylic acid group.2
Inspiration for the design of RXR ligands has also been found in nature. For example the natural product honokiol (Figure 3.2), obtained from the bark of Magnolia obovata was found to have transcriptional activity in luciferase reported cell-based assays. Moreover, the activity in cells was demonstrated by the activation of the RXR-LXR heterodimer in a dose- dependent manner.10 Small modifications to the natural product scaffold increased the potency of the molecule for the ligand binding pocket of RXR significantly.11 Electrospray mass spectrometry identified several unsaturated fatty acids with affinity for RXR, although all the measured affinities for RXR were active in the range of 5 – 10 μM.12 The natural products danthron and rhein are found in rhubarb and exhibited antagonistic properties towards RXR.13,14 Both compounds were able to inhibit 9cRA-induced transactivation and displayed no apparent activities toward other nuclear receptors. Although the reported affinities are in the sub-micromolar range, these molecular scaffolds serve as inspiration for the design of selective RXR modulators.
Figure 3.2 | Examples of RXR ligands. Natural products identified as RXR agonists. RXR antagonists, the natural product danthron and rhein, as well as synthetic transcriptional
67 CHAPTER 3
antagonists, often derived from corresponding agonists. Examples of synthetic agonists for RXR.
Because the use of 9cRA is limited by its oxidative metabolism by cytochrome P450 enzymes,15,16 there has always been general need for more stable molecules with more metabolically stable pharmacophores. The trimethylcyclohexenylvinyl unit in 9cRA (figure 3.1) can be replaced by a similar 1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene (figure 3.2) without affecting the binding potency. In this way there are no allylic protons or alkene double bonds, reducing degradation through oxidation and epoxidation. The two C=C in the conjugated polyene are prone to photoinduced isomerization and can be replaced by an arene ring.17,18 The influence on retinoid activity through modifications to the hydrophobic unit of RXR ligands has been investigated in previous studies. The methyl substituents on the hydrophobic unit should lie out of the plane of the aromatic ring. Removing a methyl group on the C4 position leads to a drop in activation. Substitution of the C4 with heteroatoms such as sulfur or oxygen was demonstrated to reduce toxicity,19 but also reduced activity compared to the parent retinoid. The synthesis and 1,4-disila-analogues was reported (for example disila-SR11237, figure 3.2)) and the carbon/silicon affected the position of helix 11 in the ligand biding domain significantly, but did not alter biophysical or biological activities such as differentiation or apoptosis in NB4 cells.20,21 In the hydrophilic domain of the RXR-binding (and indeed RAR binding) molecules a polar anchoring group is necessary for binding to RXR (and RAR). In the X-ray crystallographic structure of RXR with 9cRA, the carboxylate makes an ion pair with Arg316, two hydrogen bonds with the amide group of Ala327 and a water molecule.22 The reduced forms of ATRA, retinal and retinol are less active, but are oxidized to the active carboxylic acid in vivo. The use of sulfonates or 2,4-thiazolidinedione as polar anchor has shown promising results, especially as it leads to selectivity for RXR over the RAR. Analogs lacking any sort of polar anchor are generally poor RXR ligands. A short linker unit is necessary to bridge the hydrophobic unit with the polar end terminus. A wide variety of functionalities has been used as linker for retinoids, focusing particularly on the selectivity of ligands for RXR compared to RAR utilizing the difference in the shape of ligand binding pockets.23,24 Also metabolic stability or photo-stability can be improved via modifications to the linker unit.25
68 LIGAND BINDING POCKET RXR
The design of RXR antagonists – with the exception of danthron and rhein, which antagonize RXR via stabilization of the RXR tetramer26 –follow a strategy to the design of RXR agonists (figure 3.2). RXR antagonists have a similar molecular skeleton as the corresponding agonist but with additional hydrophobic bulk protruding from the ligand binding pocket. The potrusion disrupts the packing of Helix 12 of the ligand binding domain and in this way coactivator recruitment.27 For example, the antagonist UVI3003 is the extended antagonistic ligand derived from the agonist CD3254.28 Another example can be found in the molecule PA452 (figure 3.2), which is the analog of the potent agonist PA024.29 RXR can interact with many coregulator proteins, several transcription factors and can form homodimers or heterodimers with the nuclear receptors: LXR, FXR, CAR, TR, RAR, PPAR, VDR, PXR, Nur77 and Nurr1. Binding of ligands to RXR heterodimers can induce transcriptional activity by RXR partner receptors.30 RXR Heterodimers can be classified as either non-permissive (TR, VDR, RAR) or permissive (Nurr1, Nur77, FXR, LXR, CARR and PPAR). A non-permissive nuclear receptor would only be activated in response to its own agonist, while the RXR agonist would not enhance the transcriptional response of the heterodimer. In permissive RXR heterodimers transactivation can occur through binding of an agonist to either partner in the heterodimer or by agonists binding to both partners. Binding by both agonists could then have a synergistic or additive effect. Therefore, ligands that are selective for certain RXR-permissive heterodimers may have therapeutical applications in combination with other nuclear receptor modulators.5,7,31 Especially the PPAR – RXR heterodimer has gained a lot of attention,32,33 and this was the first heterodimer which was crystallized.34 Two published X-ray structures of RXR heterodimers bound to DNA provide detailed information about the molecular architecture and the dimerization interfaces of the nuclear receptors,34,35 consisting predominantly of residues from H7 – H11.5 Ligand LG101506 (figure 3.2) was identified as the most potent of a series of selective RXR – PPAR heterodimer activators.36,37 The potential of this selective RXR modulator in the treatment for type 2 diabetes was shown in mice, with reduced side effects an important therapeutic outome of the selective dimerization properties. Although the same ligand activates RXR – PARRα and γ heterodimers, it binds specifically to RXR and is a RXR homodimer partial agonist. Other examples of selective heterodimer activators are benzofuran 9a (figure 3.2), which activates (or has been reported to activate) the RXR – Nurr1 heterodimer in a bioluminescence resonance energy transfer (BRET) assay,38 and HX600. HX600 and derivatives have shown selectivity for RXR –
69 CHAPTER 3
Nurr1 over a broad range of dimerization partners for RXR and in Nurr1 mediated survival of dopamine neurons.39 Relevant progress has been made in the field of selective heterodimer agonists for RXR, but there is a need for more selective modulators and research tools to study these effects and to explore pharmacological possibilities. This chapter reports the synthesis, biophysical evaluation and structural elucidation of a new class of RXR ligands. The work makes use of the knowledge derived from studies on the natural product derived RXR agonists described in chapter two of this thesis. Via small modifications to the biaryl scaffold of honokiol, the natural product demonstrated potent RXR agonistic properties with very high ligand efficiency. Exploiting new possibilities using the biaryl scaffold can lead to new insights for ligand design and new tool compounds to investigate receptor conformations. Ligands 1 – 5 were designed to probe the hydrophobic region of the RXR ligand binding pocket, whilst keeping the polar interactions intact, to access the flexibility and local displacements of side-chains of the ligand binding domain (figure 3.3). To investigate the necessary contorted conformation of RXR ligands, 6 and 7 were synthesized with envisioned hindered rotational freedom. Finally, using the strategy of “extending” the agonistic ligands, compounds 9 and 10 were made as RXR partial and full antagonists respectively. The potency of the RXR ligands was then evaluated using a fluorescence-based polarization assays and cellular mammalian two-hybrid assays demonstrating highly potent and efficient ligands. To corroborate these results and to elucidate the binding mode and conformational changes in the protein, the X-ray structure of five different ligands bound to RXR was solved.
OH HO R1 HO OR
R R2 O O O OH OH OH
1:R=CH2CH=CH2 6:R1 =H,R2 =CH3 8:R=H 2:R=Ph 7:R1 =CH3,R2 =H 9:R=n-C3H7 3:R=Bn 10: R=n-C6H13 4:R=iPr 5:R=nPr
Figure 3.3 | Proposed RXR ligands. Ligands 1 – 5 probe the hydrophobic pocket in the RXR ligand binding pocket. Ligand 6 and 7 have an induced or reduced conformational rotation. The alkyl chain in ligand 9 and 10 induce an antagonistic conformation in the receptor by displacing helix 12.
70 LIGAND BINDING POCKET RXR
Chemical synthesis of the ligands
The synthesis and evaluation of ligand 1 (allyl substitution) has already been described in previous work.11 To further explore the biaryl scaffold described in chapter two of this thesis by targeting chemical modifications at the lipophilic pocket, ligands 2 – 5 were made bearing variations in the hydrophobic unit. For the synthesis of compounds 2 – 5, efficient palladium-catalyzed cross couplings were used (scheme 3.1). The cinnamic acid was treated with thionyl chloride in methanol to obtain the methyl cinnamate 11. Several commercial boronic acids or esters were then coupled using previously explored Buchwald-modified Suzuki chemistry to provide intermediates 12 – 15 in frequently excellent yields.11,40 The biaryl products were then treated with boron tribromide for the deprotection of the aryl methyl ether and subsequently with sodium hydroxide for the hydrolysis of the methyl ester. The deprotection of the phenol using boron tribromide was comparatively low yielding, but products 2 – 5 were obtained in high purity after preparative HPLC.
Scheme 3.1 | Synthesis of RXR agonists. Conditions: (a) Thionyl chloride, MeOH, 0 °C; (b) arylboronic acid or arylboronic ester, Pd2(dba)3, SPhos, KF, dioxane/H2O (10:1 v/v), 110 °C;
(c) BBr3, CH2Cl2 -78 °C. (d) NaOH, dioxane/MeOH (14:5 v/v), 40 °C.
To access the contorted conformation necessary for ligands to fit within the L-shaped pocket of RXR, molecules 6 and 7 were synthesized. A Heck coupling was performed on the methoxymethyl ether (MOM-ether) protected phenol 22, which was followed by the Miyaura borylation reaction to enable the key palladium cross coupling with 18 or 19, which were made in a single step in a moderate yield. The biaryl products were then treated with dilute hydrochloric acid to deprotect the phenol, which proceeded in much higher yield compared to the previous described boron tribromide mediated deprotection of the OMe (ligands 2 – 5),. Finally the methyl ester was hydrolyzed using sodium hydroxide and ligands 6 and 7 were obtained with high purity and in a high yield after preparative HPLC.
71 CHAPTER 3
Scheme 3.2 | Synthesis of compound 6 and 7. Reagents and conditions: (a) phenylboronic
i acid, Pd(dppf)Cl2, KOAc, dioxane/H2O (5:1 v/v), 90 °C; (b) MOMCl, Pr2NEt, CH2Cl2, room temperature; (c) methyl acrylate, Pd(dppf)Cl2, NEt3, DMF, 110 °C; (d)
Bis(pinacolato)diboron, Pd(OAc)2, XPhos, KOAc, dioxane 110 °C; (e) 18 or 19, Pd2(dba)3,
SPhos, KF, dioxane/H2O (6:1 v/v), 110 °C; (f) HCl, THF, room temperature; (g) NaOH, dioxane/MeOH (14:5 v/v).
In our efforts to modulate the RXR conformation towards the antagonistic conformation we applied the previously validated strategy by Nahoum and coworkers describing RXR antagonists with high similarity (figure 3.4).28 The synthesis and biochemical evaluation of 8 (figure 3.4) was described in chapter two of this thesis and was found to activate cofactor recruitment in fluorescence polarization assays and M2H cell assays with a moderate affinity. Modeling demonstrated that alkylation of the appropriate phenol displaced the position of helix 12 and influenced the position of L436, which plays a determinant role in the communication with helix 12.28,29
Figure 3.4 | Structures of the RXR agonist CD3254 and the series of alkyl ether analogues UVI3002 – UVI3007. Increasing length of the alkyl chain transitions the agonist into RXR antagonists.28 The same strategy was used to convert agonist 8 into antagonists
72 LIGAND BINDING POCKET RXR
Therefore, making use of the biaryl scaffold, agonist 8 was modified with different length alkyl chains. The length of the alkyl chain was hypothesized to be directly correlated to the displacement of helix 12 and therefore the effectiveness of the antagonistic properties. The antagonists 9 and 10 were synthesized in multiple steps starting from the free phenol (scheme 3.3). Subsequent sp2-sp3 Pd-catalyzed cross couplings on the bromide were used to introduce the allyl substitutent group. Using previously explored Buchwald- modified Suzuki chemistry,11,40 28 and 29 were then coupled to the organoborane 23 in excellent yields. Intermediates 30 and 31 were first treated with hydrochloric acid in THF for the deprotection of the MOM-group and subsequently with sodium hydroxide for the hydrolysis of the methyl ester.
Scheme 3.3 | Synthesis of RXR antagonists 9 and 10. Reagents and conditions: (a) 1-
bromoalkane, K2CO3, DMF; (b) 2-allyl-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, CsF,
Pd(PPh3)4, THF; (c) Pd2(dba)3, SPhos, KF, dioxane/H2O (6:1 v/v), 110 °C; (d) HCl, THF, room temperature; (e) NaOH, dioxane/MeOH (14:5 v/v).
Biochemical evaluation and crystallization
The activity of the RXR ligands was profiled using a fluorescence-based coactivator recruitment polarization assay, where an increase in fluorescence signal corresponds to an increase in coactivator recruitment through ligand binding of agonistic ligands and stabilization of the active conformation. To corroborate the fluorescence polarization (FP) data with more biologically relevant assay conditions, the ligands were also tested in a mammalian two-hybrid (M2H) assay using human osteosarcoma (U2OS) cells.
73 CHAPTER 3
The biological evaluation of ligand 1 (allyl substitution) has already been described in
11 previous work. FP and M2H assays revealed an EC50(FP) = 260 nM and EC50(M2H) = 6.3 nM compared to the full agonist LG100268: EC50(FP) = 150 nM and EC50(M2H) = 5.1 nM (table 3.1). Ligands 1 – 5 all displayed a full agonistic profile in both assays with the
measured affinities (EC50) in the nanomolar range (table 3.1). Substitution with the closely related, but synthetically and metabolically more stable n-propyl, or i-propyl did not significantly alter the affinity for RXR compared to the proposed metabolically more labile allyl group. The benzyl or phenyl substitution display a 10-fold decrease in potency in the M2H cell based assay compared to the smaller propyl substituents, which also means a drop in ligand efficiency i.e. a higher molecular weight combined with lower potency.41 Nevertheless the ligands activate the receptor with nanomolar potencies. This demonstrates the potential for ligand optimization in this region of the molecule in terms of ligand stability or side-chain displacements influencing the dimerization interface of RXR. The hydrophobic substitution using this molecular scaffold is not crucial for ligand affinity, which is reflected in the measured cellular affinities (table 3.1). Therefore, the binding is dominated by the conserved hydrogen bonding with the protein demonstrating the potential of this molecular scaffold. The differences in measured EC50 affinities between the FP assay and the M2H assay are a common phenomenon because of intrinsic differences between the two assay formats, the protein concentrations and the coregulator peptide.42
Table 3.1 | Summary of fluorescence polarization (FP) and mammalian two-hybrid (M2H) data for the RXR agonists. EC50 values for LG100268 and ligands 1 – 7, see experimental section for experimental details of the assays.
Compound FP/EC50 (nM) M2H (Luciferase)/
EC50 (nM)
LG100268 150 ± 40 5.1 ± 2.0
1 (R=CH2CH=CH2) 260 ± 110 6.3 ± 4 2 (R=Ph) 140 ± 23 85 ± 8.9 3 (R = Bn) 142 ± 8.8 92 ± 36 4 (R = iPr) 89 ± 6.9 5.8 ± 1.8 5 (R = nPr) 170 ± 80 18 ± 10 6 9900 ± 2500 > 2500 7 1017 ± 57 14600 ± 1800
74 LIGAND BINDING POCKET RXR
The co-crystallization of ligands 3 and 4 showed the canonical interactions of the carboxylate group of the ligands with Arg316, the backbone nitrogen of Ala327 and a conserved water molecule. The hydroxy group on the ligands makes a hydrogen bond with Asn306. Together this hydrogen bonding network is conserved for all the ligands and positions the hydrophobic groups. The hydrophobic side-group occupies the lipophilic region of the ligand binding pocket, where RXR amino acid reorientation can be observed. Especially ligand 4 (i-propyl substitution) repositions Ile324, Val332, Ser336 and Val342 compared to ligands 1 and 3, creating a smaller ligand binding pocket (figure 3.5). This tighter packing of helices mostly affects the position of the end of RXR helix 11 and therefore the loop between helix 11 and 12. The carboxy-terminal part of helix 11 has been identified to play a pivotal role in the dimerization of RXR.43 These results justify the further exploration of dimerization properties of these ligands, as previous work probing the lipophilic domain of RXR modulators has been shown to influence activation of RXR dimers.44,45
Figure 3.5 | Overlay of co-crystal structures of RXR with ligands 1, 3 and 4. a) Overlay of the X-ray co-crystal structures of ligands 1, 3 and 4 bound in the ligand binding pocket of RXR in ribbon representation with the TIF2 derived coregulator peptide. 1 in pale-green, 3 in orange and 4 in cyan represented as sticks. b) Zoom-in on the ligand binding pocket of RXR with the amino acid represented as sticks showing the interactions and displacements. c) Detailed overlay of the three ligands.
75 CHAPTER 3
To investigate the rotational freedom of the biaryl scaffold, the binding affinities of 6 and 7 was investigated. Ligands 6 and 7 both displayed a full agonistic profile in both FP
assay and M2H assay albeit with moderate affinities of EC50(FP) = 9.9 μM and EC50(FP) = 1.0 μM respectively, the cellular affinities were determined to be in the micromolar range, compared to the low nanomolar affinities for ligand 2 without the methyl substituent (table 3.1). The decrease in affinity might be caused by the reduced rotational freedom of the molecules and therefore a higher entropic penalty for binding to the protein in this conformation. X-ray co-crystallization of the ligands showed minor changes in the receptor conformation compared to each other. Most significant changes were observed in helix 7 and the loop between helix 11 and 12 through reorientation of amino acid residues Ser336 and Val342, which are again correlated with heterodimerization of RXR with nuclear receptor partners. The coregulator occupies the same position as observed for ligands 1, 3 and 4. The ligand fits the canonical L-shaped ligand binding pocket making the same hydrogen bonding network, and adopts a contorted conformation. The phenyl substituent for ligand 7 is free to rotate, but for ligand 6, due to steric hindrance caused by the methyl group the phenyl rotates out of the plane and occupies the lipophilic region of the RXR ligand binding pocket in an unfavorable conformation for binding (figure 3.6).
Figure 3.6 | Co-crystal structure of RXR with ligands 6 and 7. a) Overlay of the X-ray co- crystal structures of ligands 6 and 7 bound to the ligand binding pocket of RXR in ribbon representation with the TIF2 derived coregulator peptide. 6 in blue and 7 in green represented as sticks. b) Zoom-in on the ligand binding pocket of RXR with the amino acid
76 LIGAND BINDING POCKET RXR
represented as sticks showing the interactions between the ligands and the protein. c) Detailed overlay of the two ligands.
Competition of an antagonist with the full agonist LG100268 would lead to destabilization of the active conformation of the receptor ligand binding domain and a decrease of coregulator recruitment. For this reason we studied the impact of ligands 9 and 10 on coregulator recruitment via fluorescent polarization assays in competition experiments. Ligand 8 demonstrated a full agonistic profile with a moderate affinity for RXR in FP as well as M2H assys.11 Competition experiments showed that the addition of ligand 9 or 10 to the active conformation of RXR decreased the fluorescence polarization, indicating less recruitment of the coregulator and stabilization of the inactive conformation. Moreover ligand 10 demonstrated a full antagonistic effect, while ligand 9 showed a partial effect (figure 3.7a) with the following measured affinities: 9: IC50(FP) = 48.5
± 4.6 μM and 10: IC50(FP) = 46.9 ± 5.9 μM compared to UVI3003: IC50(FP) = 1.8 ± 0.6 μM. Note that ligand 9 also displayed a partial agonistic profile in an agonistic assay format. M2H competition experiments showed a significant decrease in luciferase expression for both ligands in the 10 – 40 μM range. Ligand 10 was more effective than 9 at suppressing the full agonistic activity of LG100268 (figure 3.7b).
Figure 3.7 | Ligands 9 and 10 are antagonists for RXR. a) Fluorescence polarization data showing antagonistic behavior for 9 and 10. Ligand 9 displays partial antagonism, while 10 acts as full antagonist in a competition assay against LG100268 (50 nM). b) Cellular antagonistic activities of 9 and 10 measured in a M2H luciferase competition assay against LG100268 (10 nM).
77 CHAPTER 3
To gain structural information about the mechanism of action for the antagonists, the X-ray structure was solved of the RXR ligand binding domain in complex with ligand 9 and the coregulator peptide TIF2. The overall fold of the protein displays an active conformation, which is stabilized by interactions with the coregulator peptide. Nevertheless, a comparison of this structure with that of previously reported agonist 1 revealed significant reorientation of amino acid residues (figure 3.8a,b). The most dominant reorientation affected Leu436 in helix 11 induced by the n-propyl chain, which rotated 3.0 Å towards helix 12. Leu436 has been described as key residue for the communication between the ligand and the activation function-2 (AF2).5,28,46 In this conformation the distance between the Cδ atoms of L436 and L455 in helix 12 becomes 3.25 Å, generating repulsive interactions. These data put forward that the repositioning of Leu436 lowers the association strength between helix 12 and the ligand binding domain through a steric clash, shifting the conformation of the receptor towards an antagonistic conformation as demonstrated by the FP and M2H data. The effect might be proportional to the chain length and provides a reasonable rationale for the inhibitory effect of ligand 10, but for ligand 10 no crystal structure could be obtained showing the detailed inhibitory mechanism. Moreover, modifications at this position of the ligands do not influence the positions of helix 11, which is correlated to the dimerization interface. This result shows the potential of the biaryl scaffold for selective helix repositioning in either the dimerization interface or the AF2.
Figure 3.8 | Co-crystal structure of RXR with 9. a) Overlay of the X-ray co-crystal structures of ligands 1 and 9 bound in the ligand binding pocket of RXR in ribbon representation with the TIF2 derived coregulator peptide. 1 in pale-green and 9 in blue represented as
78 LIGAND BINDING POCKET RXR
sticks. b) Zoom-in on the ligand binding pocket of RXR with the involved helices and amino acid showing the interactions and displacement of Leu436.
Discussion and concluding remarks
Despite the fact that RXR plays a major role many biological processes through heterodimerization, a limited number of small molecule RXR modulators is available, and the repertoire is restricted in chemical diversity and biophysical properties. This work has generated a compact and focused selection of RXR modulators consisting of a biaryl scaffold which are structurally different than previous reported molecules. Fluorescence polarization and cellular mammalian two-hybrid assays were used to investigate the impact of the ligands on the coregulator recruitment and activation. The X-ray structures of five ligand were solved, providing structural information for ligand binding and receptor conformation. The ligands 1 – 5 all displayed a full agonistic effect with high potency and ligand efficiency, underlining the potential of the scaffold. The co-crystal structures revealed significant amino acid repositioning and secondary structure modifications in the ligand binding domain were observed in the carboxylate end of helix 11 and the loop between the end of helix 11 and 12. This might give rise to selectivity in dimerization properties of the receptor, as this region has been identified to have a pivotal role in dimerization properties for RXR. The restricted conformational freedom for ligands 6 and 7 was unfavorable for binding to RXR, with a decrease of potency of 10 to 100 times receptively compared to ligand 2 without the methyl substituent. However, significant changes were observed in helix 7 and the loop between helix 11 and 12 due to the induced rotation of the methyl substituent. The agonists 8 was modified with an alkyl chain protruding from the biaryl core of the molecule. Indeed, although the crystal structure of the complex with the mixed partial agonist/antagonist 9 displayed helix 12 in the holo conformation stabilized by the coregulator in the AF2, fluorescence polarization and cellular data suggest inhibition of coregulator recruitment and destabilization of the active conformation. Binding of ligand 9 demonstrated a significant impact on the position of Leu436 in the ligand binding pocket compared to the fully active conformation. Moreover, the co-crystal structure with 9 provides a rational for the inhibitory effect of the full antagonist 10. The binding of ligand 9 in the ligand binding pocket did not influence the position of helix 11 as has been observed for the other ligands.
79 CHAPTER 3
In general this work has shown the potential of the biaryl scaffold for selective helix repositioning in the dimerization interface or the AF2 of RXR. With a focused selection of biaryl ligands a wide range of conformations and functional outcome was addressed, expanding the current RXR modulator repertoire with agonistic as well as antagonistic ligands. The data provides a rationale for the design of RXR ligands build of an unique hydrophilic region with a conserved hydrogen bonding network contributing to the binding affinity, and a hydrophobic region to probe the other parts of the pocket influencing dimerization properties or coregulator recruitment These findings justify further explorations of the dimerization properties induced by the ligands. The mechanistic insights presented in this work were gained using RXR, but might correspond to more general conserved mechanisms in the nuclear receptor protein family and might aid in efforts to make selective nuclear receptor modulators.
Experimental section
General considerations for protein expression and purification
All solutions and equipment used in the handling of microbial cultures were autoclaved or sterile filtered. Media, plastic and glassware were autoclaved at 121 °C for 20 min prior to use. Bacterial cultures were incubated in a New Brunswick Series 25 shaker. Centrifugation was performed in a Beckman Coulter Avanti J-25 centrifuge. Microcentrifugation was performed in an Eppendorf Centrifuge 5415R or a Beckman Coulter Microfuge 18. All biological laboratory buffers and media were bought from common suppliers and used as purchased. BL21(DE3) and NovaBlue E. coli competent cells were purchased from Novagen, XL-10. DNA and protein concentration was determined using a NanoDrop 1000 spectrometer from Thermo Scientific using 260 nm and 280 nm wavelength respectively. Gel electrophoresis for proteins was performed using 12% SDS-PAGE gels in running buffer and visualized using InstantBlue stain. Protein concentration was determined using a NanoDrop 1000 spectrometer with a wavelength ratio of 280-260 nm. The fluorescent D22 coactivator peptide was purchased from Invitrogen life technologiesTM
Polarization assay
His6-RXRα-LBD (1 μM), fluorescein-labeled D22 co-activator peptide (0.1 μM), and the ligand at the indicated concentration in buffer containing 100 mM HEPES (pH 7.5), 100 mM NaCl, 5 mM DTT and 0.1% bovine serum albumin, were incubated for 60 minutes at 4 °C and protected from light. Conditions for the competition assay: His6-RXRα-LBD (500 nM), fluorescein-labeled D22 co- activator peptide (50 nM), LG100268 (50 nM). Fluorescent polarization signals (mP) were
80 LIGAND BINDING POCKET RXR measured with a Tecan Infinite F500 plate reader. Experiments were performed in triplicate and the data were analyzed using Origin software
Mammalian two-hybrid assay
The mammalian two-hybrid (M2H) assays were performed in U2OS cells. The RXR LBD was cloned into the pCMV-AD vector (Aligent), fused to NF-κB and Strep-tag resulting in NF-κB-Strep- RXR-LBD. The peptide sequence was cloned into the pCMV-BD vector (Aligent) and fused to GAL4 DNA binding domain. For the direct interaction assay, ~40,000 cells/well were seeded in a 24-well plate for 24 h and transfected with 40 ng pCMV-AD, 40 ng pCMV-BD, 0.2 μg pFR-Luc, and 3.2 ng pGL-Renilla using PEI (Polysciences), before being treated with the ligands in indicated concentrations. For the competition assay, cells were co-treated with 100 nM of LG100268 and the indicated concentrations of competitor. After 24 h of treatment the interaction was determined with a Dual-Luciferase® Reporter Assay (Promega), according to the manufacturer’s instruction. The luminescent intensities were recorded on a Synergy HT platereader (BioTek). The FR- luciferase signal was normalized over the Renilla luciferase (pGL-Renilla) signal. pCMV-BD sequence
MKLLSSIEQACDICRLKKLKCSKEKPKCAKCLKNNWECRYSPKTKRSPLTRAHLTEVESRLER LEQLFLLIFPREDLDMILKMDSLQDIKALLTGLFVQDNVNKDAVTDRLASVETDMPLTLRQHRI SATSSSEESSNKGQRQLTVSPELGSASPGEFLTARHPLLMRLLLSPS.
Expression, purification and crystallization of the RXRα LBD.
The histidine-tagged LBD of human RXRα (in a pET15b vector) was expressed in Escherichia coli BL21(DE3). Cells were grown at 37 °C in LB medium supplemented with 100 mg mL-1 ampicillin until OD600 reached about 0.7. Expression of T7 polymerase was induced by addition of isopropyl- b-d-thiogalactoside (IPTG) to a final concentration of 0.1 mM. After an additional incubation for 15 h at 15°C, cell cultures were harvested by centrifugation at 8,000 ´ g for 20 min. The cell pellet from 2 liters of RXRα LBD was resuspended in 50 ml buffer A (20 mM Tris-HCl pH 8.0, 500 mM NaCl, 5 mM imidazole) supplemented with a protease inhibitor (PMSF) and DNAse I. The suspension was then lysed by sonication and centrifuged at 35,000 g and 4 °C for 45 min. The supernatant was loaded onto a 5 ml Ni2+-affinity column, preequilibrated with buffer A. The column was washed with 10 volumes of buffer A and 10 volumes of buffer A supplemented with 50 mM imidazole. Bound proteins were eluted with buffer A containing 200 mM imidazole. The fractions containing RXR LBD were pooled, concentrated and desalted to buffer B (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM DTT). To remove the histidine-tag the protein was incubated for 16 h at 4°C with
81 CHAPTER 3 thrombin (1 unit/mg RXR). The protein was passed through a Ni2+ column and a superdex gel filtration column. The protein was concentrated and stored at -80°C until further use.
Before crystallization the protein was mixed with a 1.5-fold molar excess of ligand and a 3-fold excess of TIF2 NR2 cofactor peptide (686-KHKILHRLLQDSS-698). The complex was incubated for 1h at 4 °C. Drops with a size of 2 - 3 μL using different reservoir to protein ratio were manually mixed and equilibrated against reservoirs with a volume of 500 mL. Optimal crystals were grown in a week in 3 μL drops with protein solution to reservoir ratio of 2:1 with 0.1 M PIPES, pH 7.0, 0.1 M NaCl, 22% PEG 2K MME. The crystals were cryo-cooled in liquid nitrogen using sucrose as cryo- protectant for X-ray data collection. Diffraction data for RXR were collected at the DESY beamline (Deutsches Elektronen-Synchrotron, Hamburg, Germany). The data set was indexed and integrated using iMosflm and scaled using SCALA. The structure was phased by molecular replacement using PDB ID 5EC9 as search model in Phaser. Coot and phenix.refine were used in alternating cycles of model building and refinement. All data collection, refinement and validation statistics are shown in Table S3.1.
Table S3.1 Crystallographic statistics for RXR complexes Ligand 3 Ligand 4 Ligand 6 Ligand 7 Ligand 9 Data collection Resolution (Å) 46.72-1.90 40.66-1.70 67.40-2.90 59.39-2.09 64.29-2.90 (1.97-1.90) (1.73-1.70) (3.08-2.90) (2.15-2.09) (3.08-2.90)
Space group P212121 P3221 P43212 P212121 P43212 Cell parameters a=63.86, a=b=77.70, a=b=67.40, a=71.70, a=b=64.29, (Å) b=86.35, c=81.33 c=108.74 b=74.14, c=112.26 c=109.00 c=99.21
Rmerge 0.096 0.092 0.136 0.118 0.129 (0.433) (0.641) (0.541) (0.603) (0.406) a Average I/σ(I) 50.1 (3.3) 19.5 (1.80) 16.5 (6.0) 14.0 (4.1) 18.1/ (7.6) Completeness 96 (75) 98 (75) 100 (100) 100 (100) 100 (100) (%)a Redundancya 9.8 (5.4) 8.5 (3.1) 20.5 (20.0) 11.4 (10.9) 19.6 (18.9)
Refinement Number of 3507/377/86 3693/289/21 1771/2/42 3644/215/84 1744/3/46 protein/solvent/li gand atoms
Rwork/Rfree (%) 20.1/23.0 14.2/16.7 19.1/27.7 19.0/23.2 18.5/28.5 No. of reflections 36774 30949 5976 32012 5632 R.m.s. deviations 0.007 / 0.856 0.009/0.940 0.008/1.054 0.007/0.900 0.008/0.985 from ideal values
82 LIGAND BINDING POCKET RXR
bond lengths (Å) / bond angles (°) Average 26.3 / 33.2 / 16.7 / 30.7 / 48.2 / 34.1 / 32.0 / 32.7 / 32.0 / 32.7 / protein/solvent/li 18.6 11.1 44.43 25.6 26.6 gand B-factor (Å2) a number in parentheses is for the highest resolution shell
General consideration synthetic procedures
All the solvents employed were commercially available and used without purification unless stated otherwise. Water was purified using a Millipore purification train. All the reagents are commercially available and used without purification. All the NMR data were recorded on a Varian Gemini 400 MHz NMR, a Bruker Cryomagnet 400 MHz a Bruker UltraShield Magnet 400 MHz or a Varian 200 MHz (400 or 200 MHz for 1H-NMR and 100 or 50 MHz for 13C-NMR). Proton experiments are reported in parts per million (ppm) downfield of TMS. All 13C spectra were reported in ppm relative to residual chloroform (77 ppm) Analytical LC-MS was performed on a
C4, Jupiter SuC4300A, 150x2.00 mm column with a gradient 5% to 100% acetonitrile in H2O in 15 min. Silica column chromatography was performed manually using silica with particle size 60 – 200 μm. Preparative HP-LC was performed on a Gemini S4 110A 150x21.20 mm column using
H2O with 0.1% Formic Acid (F.A.) and acetonitrile with 0.1% F.A. Purity and exact mass of the compounds were determined using a High Resolution LC-MS system consisting of a Waters ACQUITY UPLC I-Class system coupled to a Xevo G2 Quadrupole Time of Flight (Q-tof). The system was comprised of a Binary Solvent Manager and a Sample Manager with Fixed-Loop (SM- FL). compounds were separated (0.3 mL min-1) by the column (Polaris C18A reverse phase column 2.0 x 100 mm, Agilent) using a 15% to 75% acetonitrile gradient in water supplemented with 0.1% v/v formic acid before analysis in positive mode in the mass spectrometer.
Synthetic procedures
(E)-methyl 3-(3'-benzyl-6-methoxy-[1,1'-biphenyl]-3-yl)acrylate In an oven-dried Schlenk tube were weighed (E)-methyl 3-(3-chloro-4-methoxyphenyl)acrylate (97.3 mg, 0.43 mmol), 2-(3- benzylphenyl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (152.5 mg, 0.52 mmol), KF (124.9 mg, 2.15 mmol), SPhos (54.0 mg, 0.13 mmol) and Pd2(dba)3 (38.9 mg, 0.013 mmol). The Schlenk tube was
83 CHAPTER 3
evacuated and backfilled with argon three times. Dioxane/H2O (10:1 v/v, 2.25 mL) was addded under a positive argon flow and the reaction was stirred at 110 °C for 18 h. The mixture was then allowed to cool to room temperature and passed through celite®, eluting with EtOAc. The product was purified via column chromatography eluting with 17% EtOAc in heptane to yield a colorless oil. 146 mg, 0.41 mmol, 95%. Silica gel TLC Rf = 0.28 (17% v/v EtOAc in heptane); LC-MS (ESI): 1 calc. for C24H22O3 [M+H]: 359.16 observed 359.08, LC, Rt=8.68 min; H NMR (400 MHz, CDCl3) δ 7.73 (d, J = 16.0 Hz, 1H), 7.53 – 7.48 (m, 2H), 7.40 – 7.20 (m, 9H), 6.98 (d, J = 8.4 Hz, 1H), 6.39
13 (d, J = 16.0 Hz, 1H), 4.08 (s, 2H), 3.83 (s, 3H), 3.83 (s, 3H); C NMR (100 MHz, CDCl3) δ 167.80, 158.38, 144.56, 141.07, 140.98, 137.83, 131.30, 130.61, 130.15, 129.08, 128.54, 128.28, 128.02, 127.31, 127.22, 126.17, 115.64, 111.41, 55.75, 51.67, 42.05.
(E)-3-(3'-benzyl-6-hydroxy-[1,1'-biphenyl]-3-yl)acrylic acid (E)-methyl 3-(3'-benzyl-6-methoxy-[1,1'- biphenyl]-3-yl)acrylate (60 mg, 0.17 mmol) was dissolved in dry CH2Cl2 (0.73 mL). The mixture was cooled to -78 °C, and BBr3 (1 M in CH2Cl2, 335 μL 0.34 mmol) was added and the reaction was stirred for 1 h at -78 °C and 1 h at 0 °C. the reaction was then allowed to warm to room temperature and was quenched with H2O. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo. LC-MS (ESI): calc. for C22H20O3 [M+H]: 345.15 observed 345.17, LC, Rt=7.70 min.
The resulting crude mixture was dissolved in dioxane/MeOH (14:5 v/v, 2.4 mL) and 4 N NaOH (350 μL) andthe reaction was stirred for 20 h at room temperature. The solvent was evaporated and the residue was separated between H2O and CH2Cl2. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo. The product was purified via preparative HPLC. LC-MS (ESI): calc. for
1 C22H18O3 [M+H]: 331.14 observed 331.00, LC, Rt=6.65 min; H NMR (400 MHz, CDCl3) δ 7.75 (d, J = 15.9 Hz, 1H), 7.49 – 7.40 (m, 3H), 7.33 – 7.19 (m, 8H), 6.99 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 15.9
13 Hz, 1H), 4.05 (s, 2H); C NMR (100 MHz, CDCl3) δ 172.48, 154.98, 146.74, 142.96, 140.63, 136.23, 130.81, 129.80, 129.66, 129.57, 129.15, 129.05, 128.84, 128.78, 127.17, 126.75, 126.50, 116.63, 115.08, 42.09; HRMS (m/z): [M + H]+ calcd 331.1334, found 331.1319.
(E)-methyl 3-(6-methoxy-[1,1':3',1''-terphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed (E)-methyl 3-(3-chloro-4-methoxyphenyl)acrylate (119.8 mg, 0.53 mmol), [1,1'-biphenyl]-3- ylboronic acid (130.2 mg, 0.66 mmol), KF (154.5 mg, 2.66 mmol), SPhos (66.5 mg, 0.16 mmol)
84 LIGAND BINDING POCKET RXR
and Pd2(dba)3 (48.2 mg, 0.053 mmol). The schlenk tube was evacuated and backfilled with argon three times. Dioxane/H2O (10:1 v/v, 2.8 mL) was addded under a positive argon flow and the reaction was stirred at 110 °C for 18 h. The mixture was then allowed to cool to room temperature and was passed through celite, eluting with EtOAc. The product was purified via column chromatography eluting with 17% EtOAc in heptane to yield a colorless oil. 180 mg, 0.52 mmol,
98%. Silica gel TLC Rf = 0.25 (17% v/v EtOAc in heptane); LC-MS (ESI): calc. for C23H20O3 [M+H]:
1 345.15 observed 345.08, LC, Rt=8.52 min; H NMR (400 MHz, CDCl3) δ 7.77 – 7.70 (m, 2H), 7.68 – 7.64 (m, 2H), 7.62 – 7.57 (m, 2H), 7.54 – 7.35 (m, 6H), 7.01 (d, J = 8.5 Hz, 1H), 6.39 (d, J = 16.0
13 Hz, 1H), 3.87 (s, 3H), 3.82 (s, 3H); C NMR (100 MHz, CDCl3) δ 167.75, 158.35, 144.45, 141.17, 138.16, 131.16, 130.55, 129.25, 128.78, 128.53, 128.41, 128.36, 127.33, 127.28, 127.25, 126.21, 115.67, 111.39, 55.79, 51.63.
(E)-3-(6-hydroxy-[1,1':3',1''-terphenyl]-3-yl)acrylic acid (E)-methyl 3-(6-methoxy-[1,1':3',1''-terphenyl]-
3-yl)acrylate (100 mg, 0.29 mmol) was dissolved in dry CH2Cl2 (1.26 mL). The mixture was cooled to -78 °C, and BBr3 (1 M in CH2Cl2, 580 μL 0.58 mmol) was added and the reaction was stirred for 1 h at -78 °C and 1 h at 0 °C. the reaction was then allowed to warm to room temperature and was quenched with H2O. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo to obtain
(E)-methyl 3-(6-hydroxy-[1,1':3',1''-terphenyl]-3-yl)acrylate. LC-MS (ESI): calc. for C22H18O3 [M+H]: 331.13 observed 331.17, LC, Rt=8.25 min.
The resulting crude mixture was dissolved in dioxane/MeOH (14:5 v/v, 1.6 mL) and 4 N NaOH (250 μL). The reaction was stirred for 20 h at room temperature. The solvent was evaporated and the residue was separated between H2O and CH2Cl2. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo. The product was purified via preparative HPLC. LC-MS (ESI): calc. for
1 C21H16O3 [M+H]: 317.12 observed 317.17, LC, Rt=6.52 min; H NMR (400 MHz, CDCl3) δ 7.77 (d, J = 15.9 Hz, 1H), 7.69 – 7.56 (m, 5H), 7.53 – 7.36 (m, 6H), 7.04 (d, J = 7.9 Hz, 1H), 6.36 (d, J = 15.9
13 Hz, 1H); C NMR (100 MHz, CDCl3) δ 171.82, 155.03, 146.63, 142.75, 140.50, 136.63, 130.84, 130.08, 129.81, 129.08, 128.83, 127.97, 127.92, 127.85, 127.35, 127.31, 116.75, 115.09; HRMS (m/z): [M + H]+ calcd 317.1178, found 317.1179.
85 CHAPTER 3
(E)-methyl 3-(3'-isopropyl-6-methoxy-[1,1'-biphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed (E)-methyl 3-(3-chloro-4-methoxyphenyl)acrylate (144.9 mg, 0.64 mmol), (3- isopropylphenyl)boronic acid (128.2 mg, 0.78 mmol), KF (187.2 mg, 3.22 mmol), SPhos (80.7 mg,
0.20 mmol) and Pd2(dba)3 (60.7 mg, 0.066 mmol). The schlenk tube was evacuated and backfilled with argon three times. Dioxane/H2O (10:1 v/v, 3.4 mL) was addded under a positive argon flow and the reaction was stirred at 110 °C for 18 h. The mixture was then allowed to cool to room temperature and was passed through celite, eluting with EtOAc. The product was purified via column chromatography eluting with 15% EtOAc in heptane to yield (E)-methyl 3-(3'-isopropyl-6- methoxy-[1,1'-biphenyl]-3-yl)acrylate as colorless oil in a quantitative yield. Silica gel TLC Rf = 0.35
(15% v/v EtOAc in heptane); LC-MS (ESI): calc. for C20H22O3 [M+H]: 311.16 observed 311.08, LC,
1 Rt=8.52 min; H NMR (399 MHz, CDCl3) δ 7.70 (d, J = 16.0 Hz, 1H), 7.52 – 7.45 (m, 2H), 7.41 – 7.31 (m, 4H), 6.96 (d, J = 8.5 Hz, 1H), 6.35 (d, J = 16.0 Hz, 1H), 3.83 (s, 3H), 3.79 (s, 3H), 2.96
13 (hept, J = 6.9 Hz, 1H), 1.29 (d, J = 6.9 Hz, 6H); C NMR (100 MHz, CDCl3) δ 167.86, 158.44, 148.73, 144.66, 137.65, 131.66, 130.68, 129.06, 128.49, 128.09, 127.75, 127.05, 125.56, 115.60, 111.42, 55.81, 51.69, 34.25, 24.14.
(E)-3-(6-hydroxy-3'-isopropyl-[1,1'-biphenyl]-3-yl)acrylic acid (E)-methyl 3-(3'-isopropyl-6-methoxy-
[1,1'-biphenyl]-3-yl)acrylate (100 mg, 0.32 mmol) was dissolved in dry CH2Cl2 (1.4 mL). The mixture was cooled to -78 °C, and BBr3 (1 M in CH2Cl2, 644 μL 0.64 mmol) was added and the reaction was stirred for 1 h at -78 °C and 1 h at 0 °C. the reaction was then allowed to warm to room temperature and was quenched with H2O. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo to obtain (E)-methyl 3-(6-hydroxy-3'-isopropyl-[1,1'-biphenyl]-3-yl)acrylate. LC-MS (ESI): calc. for C19H19O3 [M+H]: 297.15 observed 297.17, LC, Rt=8.40 min.
The resulting crude mixture was dissolved in dioxane/MeOH (14:5 v/v, 4.6 mL) and 4 N NaOH (564 μL). The reaction was stirred for 20 h at room temperature. The solvent was evaporated and the residue was separated between H2O and CH2Cl2. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated in vacuo. The product was purified via preparative HPLC. LC-MS (ESI): calc. for
1 C18H18O3 [M+H]: 283.13 observed 283.17, LC, Rt=6.43 min; H NMR (400 MHz, CDCl3) δ 7.77 (d, J = 15.9 Hz, 1H), 7.51 – 7.42 (m, 3H), 7.34 – 7.27 (m, 3H), 7.02 (d, J = 8.3 Hz, 1H), 6.34 (d, J = 15.9
13 Hz, 1H), 2.98 (p, J = 6.9 Hz, 1H), 1.30 (d, J = 6.9 Hz, 6H); C NMR (100 MHz, CDCl3) δ 172.43, 155.05, 150.66, 146.82, 135.97, 130.79, 129.68, 129.65, 129.17, 127.26, 127.14, 126.80, 126.38, 116.57, 115.00, 34.34, 24.14; HRMS (m/z): [M + H]+ calcd 283.1334, found 283.1334.
86 LIGAND BINDING POCKET RXR
(E)-methyl 3-(4-methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate An oven dried Schlenk flask with a stirring bar was charged with (E)-methyl 3-(3-chloro-4- methoxyphenyl)acrylate (250 mg, 1.103 mmol), Pd(OAc)2 (5.8 mg, 0.026 mmol), XPhos (21.4 mg, 0.045 mmol), KOAc (321.87 mg, 3.28 mmol) and bis(pinacolato)diboron (841.0 mg, 3.31 mmol). The schlenk flask was sealed with a teflon screw cap and evacuated and backfilled with argon three times. Dioxane (2.6 mL) was added and the reaction mixture was stirred at 110 °C. After 4.5h the mixture was allowed to cool to room temperature and was diluted with ethyl acetate and passed through Celite, eluting with EtOAc. The product was purified via column chromatography eluting with hexane/EtOAc 10% to 25% (v/v) to obtain an off white solid (304.5 mg, 0.96 mmol), 87% 1 yield. HNMR ( 400 MHz, CDCl3): 7.88 (d, J = 4.0 Hz, 1H), 7.66 (d, J = 16.0 Hz, 1H), 7.56 (qt, J = 12.0 Hz, 1H), 6.86 (d, J = 8.0 Hz, 1H), 6.35 (d, J = 16.0 Hz, 1H), 3.86 (s, 3H), 3.79 (s, 3H), 1.36 (s, 13 12H). CNMR ( 100 MHz, CDCl3): 167.6, 165.95, 144.65, 136.99, 132.89, 126.64, 115.37, 110.72, 83.90, 56.08, 51.71, 24.97
1-bromo-3-propylbenzene To a solution of 3-bromopropiophenone (2.98 g, 14 mmol) in TFA (30 mL) was added dropwise triethylsilane (11.5 mL) at 0 °C in 5 minutes and the mixture was stirred for additional 20 minutes. The reaction mixture was heated to 80 °C and stirred overnight. The reaction mixture was allowed to cool to RT and concentrated. Toluene was added and the mixture was again concentrated to obtain the crude material. The product was purified via column chromatography eluting with hexane to obtain 1-bromo-3-propylbenzene (78.2 mg, 0.39 mmol).
1 3% yield. HNMR (400MHz, CDCl3): δ (ppm) 7.33-7.29 (m, 2H), 7.16-7.08 (m, 2H), 2.55 ( t, J = 7.6, 13 2H), 1.63 (sext, J = 7.4 Hz, 2H), 0.93 (t, J = 7.3, 3H); CNMR (100 MHz, CDCl3): 144.99, 131.48, 129.74, 128.70, 127.11, 122.28, 37.64, 24.34, 13.71.
(E)-methyl 3-(6-methoxy-3'-propyl-[1,1'-biphenyl]-3-yl)acrylate An oven dried Schlenck flask with a stirring bar was charged with 1-bromo-3-propylbenzene (78.2 mg, 0.39 mmol) and (E)-methyl 3-(4- methoxy-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate ( 152 mg, 0.477 mmol). The
Schlenck flask was evacuated and backfilled with argon three times. Pd2(dba)3 (37.7 mg, 0.0412 mmol), SPhos (49.63 mg, 0.121 mmol) and KF (114.69 mg, 1.974 mmol) were added under positive argon pressure. Degassed 1,4-dioxane/H2O (10:1 v/v, 1.8 mL) was added and the reaction mixture was stirred at 100 °C. After 7h the reaction mixture was allowed to cool to room temperature and passed through celite®, eluting with EtOAc. The product was purified via column
87 CHAPTER 3 chromatography eluting with hexane/EtOAc 5%-10% to obtain the pure (E)-methyl 3-(6-methoxy- 3'-propyl-[1,1'-biphenyl]-3-yl)acrylate as a dark yellow oil (95 mg ; 0.31 mmol), 78% yield. 1HNMR
(400 MHz, CDCl3): 7.69 (d, J = 16.0 Hz, 1H), 7.49 (m, 2H), 7.33 (m, 3H), 7.18 (m, 1H), 6.98 (d, J = 8.0 Hz, 1h), 6.35 (d, J = 16.0 Hz, 1H), 3.85 (s, 3H), 3.80 (s, 3H), 2.64 (t, J = 16, 2H), 1.69 (m, 2H), 13 0.98 (t, J = 7.6 Hz, 3H); CNMR ( 100 MHz, CDCl3): 168.10, 158.70, 144.88, 142.86, 137.85, 131.84, 130.89, 129.93, 129.33, 128.74, 128.28, 127.88, 127.50, 127.14, 115.88, 111.65, 56.09, 51.95, 38.47, 24.93, 14.27.
(E)-methyl 3-(6-hydroxy-3'-propyl-[1,1'-biphenyl]-3-yl)acrylate In an oven dried round bottom flask with a stirring bar was weighed (E)-methyl 3-(6-methoxy-3'-propyl-[1,1'-biphenyl]-3-yl)acrylate (63.2 mg, 0.204 mmol). The flask was evacuated and backfilled with argon three times. CH2Cl2 (0.91 mL ) was added and the solution was cooled to -78 °C. 1M BBr3 in CH2Cl2 (0.4 mL, 0.41 mmol) was added dropwise to the mixture and the mixture was stirred for 1h at -78 °C, 1h at 0 °C and 15 minutes at RT. Subsequently the reaction was quenched with H2O (1.04 mL) stirred for 5 minutes and extracted with CH2Cl2 (3 x 2 mL). The organic layers were washed with brine and dried over Na2SO4, filtered and concentrated. The product was purified via column chromatography eluting with EtOAc/Hexane (5%-10% ; v/v) and subsequently concentrated to yield (E)-methyl 3-(6-hydroxy-3'- propyl-[1,1'-biphenyl]-3-yl)acrylate as a yellow oil. Yield: 25.1 mg ; 0.085 mmol, 42% yield. 1HNMR
( 400 MHz, CDCl3): 7.67 (d, J = 16.0 Hz, 1H), 7.43 (m, 2H), 7.26 (m, 3H), 7.00 (d, J = 8.0 Hz, 1H), 6.33 (d, J = 8.0 Hz, 1H), 3.79 (s, 3H), 2.66 (t, J = 6.0 Hz, 2H), 1.69 (m, 2H), 0.98 (t, J = 6.0 Hz, 13 3H); CNMR ( 100 MHz, CDCl3): 167.77, 154.56, 144.52, 144.20, 135.93, 130.29, 129.35, 129.10, 128.53, 127.27, 126.11, 116.34, 115.45, 51.61, 38.00, 24.52, 13.58.
(E)-3-(6-hydroxy-3'-propyl-[1,1'-biphenyl]-3-yl)acrylic acid 4M NaOH (181μL, 0.168 mmol) was added to a solution of (E)-methyl 3-(6-hydroxy-3'-propyl-[1,1'-biphenyl]-3-yl)acrylate (25.1 mg, 0.084 mmol) in 1,4-dioxane/MeOH (14:5 ; v/v) (1.4 mL). The reaction mixture was stirred at 40 °C until the ester had been completely consumed, determined by TLC. The reaction mixture was diluted with H2O and subsequently acidified with 1M HCl to a pH of 2. The mixture was concentrated and extracted with EtOAc. The product was purified via preperative HPLC (8.8 mg, 0.03 mmol). 37% yield after preperative HPLC. LC-MS (ESI): calc. for C18H18O3 [M+H]: 283.13, observed 283.17, LC, Rt = 6.47.
1 H NMR (400 MHz, CDCl3) δ 7.75 (d, J = 15.9 Hz, 1H), 7.50 – 7.39 (m, 3H), 7.29 – 7.22 (m, 3H), 7.01 (d, J = 8.3 Hz, 1H), 6.33 (d, J = 15.9 Hz, 1H), 2.66 (t, J = 7.6 Hz, 2H), 1.69 (sext, J = 7.5 Hz,
13 2H), 0.97 (t, J = 7.3 Hz, 3H); C NMR (100 MHz, CDCl3) δ 172.22, 155.07, 146.83, 144.45, 135.93, 130.77, 129.64, 129.58, 129.19, 129.08, 128.78, 127.14, 126.26, 116.58, 114.95, 38.16, 24.69, 14.02.
88 LIGAND BINDING POCKET RXR
4-bromo-2-chloro-1-propoxybenzene 4-bromo-2-chlorophenol (4 g, 19.28 mmol) was dissolved in
DMF (100 mL). To this solution K2CO3 (8.0 g, 57.9 mmol) and 1-bromopropane (8.8 mL, 96 mmol) were added. The reaction was stirred at 70 °C for 22 hour. The reaction was then quenched with
H2O and extracted with CH2Cl2 (3x). The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. the product was purified via flash column chromatography, eluting with heptane/EtOAc 95/5 v/v to yield 4-bromo-2-chloro-1-propoxybenzene as a colorless oil
(4.7 g, 19 mmol, 98%). Silica gel TLC Rf = 0.46 (Heptane/EtOAc 5% v/v); GC-MS (ESI) m/z calc. 1 for C9H10BrClO: 249.53, most abundant peaks observed: 250, 210, 208, Rt = 5.04; HNMR (400
MHz, CDCl3): δ (ppm) 7.47 (d, J = 2.4 Hz, 1H), 7.27 (dd, J = 8.8, 2.4 Hz, 1H), 6.75 (d, J = 8.8 Hz, 1H), 3.93 (t, J = 6.5 Hz, 2H), 1.84 (h, J = 7.4 Hz, 2H), 1.06 (t, J = 7.4 Hz, 3H); 13C NMR (50 MHz,
CDCl3) δ 153.99, 132.62, 130.48, 124.07, 114.49, 112.26, 70.84, 22.52, 10.54.
4-allyl-2-chloro-1-propoxybenzene An over dried schlenk flask was charged with 4-bromo-2-chloro-
1-propoxybenzene (1.02 g, 4.08 mmol), CsF (1.31 g, 8.62 mmol) and Pd(PPh3)4 (467 mg, 0.404 mmol). The flask was then evacuated and backfilled with argon three times and THF (34 mL) was added. The mixture was stirred for 30 minutes at room temperature before 2-allyl-4,4,5,5- tetramethyl-1,3,2-dioxaborolane (1.35 mL, 7.21 mmol) was added. The reaction was stirred for 22 h at 78 °C. Another portion of CsF (1.26 g, 8.29 mmol), Pd(PPh3)4 (467 mg, 0.404 mmol) and THF (30 mL) were added and the reaction was stirred at 78 °C. After 24 h the reaction was allowed to cool to room temperature and was separated between pentane and H2O. The aqueous layer was washed with pentane (2x). The combined organic layers were washed with brine, dried over
Na2SO4, filtered twice and evaporated. The product was purified via column chromatography eluting with heptane/EtOAc 5% v/v to yield 4-allyl-2-chloro-1-propoxybenzene as colorless oil (690 mg, 3.27 mmol, 82%). Silica gel TLC Rf = 0.58 (Heptane/EtOAc 5% v/v); GC-MS (ESI) m/z calc. 1 for C12H15ClO: 210.70, most abundant peaks observed: 210, 168, 133, Rt = 5.02; H NMR (400 MHz,
CDCl3) δ 7.20 (d, J = 2.1 Hz, 1H), 7.01 (dd, J = 8.9, 2.2 Hz, 1H), 6.85 (d, J = 8.4 Hz, 1H), 5.99 - 5.88 (m, 1H), 5.12 - 5.05 (m, 2H), 3.98 (t, J = 6.5 Hz, 2H), 3.31 (d, J = 6.7 Hz, 2H), 1.86 (sext, J = 7.1 Hz,
13 2H), 1.08 (t, J = 7.4 Hz, 3H); C NMR (50 MHz, CDCl3) δ 153.03, 137.21, 133.11, 130.36, 127.70, 122.87, 116.11, 113.56, 70.81, 39.13, 22.66, 10.61.
89 CHAPTER 3
4-bromo-2-chloro-1-(hexyloxy)benzene 4-bromo-2-chlorophenol (4 g, 19.3mmol) was dissolved in dry DMF (100 mL) in an oven dried flask. To this solution was added K2CO3 (8 g, 57.9 mmol) and 1-bromohexane (14 mL, 100 mmol). The reaction was stirred at 70 °C for 18 hours. The reaction was separated between H2O and CH2Cl2. The aqueous layer was washed with CH2Cl2 (2 times).
The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated to yield yellow oil. The product was purified via flash column chromatography eluting with hexane/EtOAc 95/5 v/v to yield 4-bromo-2-chloro-1-(hexyloxy)benzene as a colorless oil (5.5 g, 18.9 mmol, 98%). Silica gel TLC Rf = 0.70 (Hexane/EtOAc 5% v/v). GC-MS (ESI) m/z calc. for 1 C12H16BrClO: 291.6, most abundant peaks observed: 292, 210, 208, Rt = 6.25; HNMR (400 MHz,
CDCl3): δ (ppm) 7.45 (d, J = 2.4 Hz, 1H), 7.25 (dd, J = 8.8, 2.4 Hz, 1H), 6.72 (d, J = 8.8 Hz, 1H), 3.94 (t, J = 6.8 Hz, 2H), 1.83 - 1.76 (m, 2H), 1.50 - 1.42 (m, 2H), 1.36 - 1.30 (m, 4H), 0.90 (t, J = 6.8
13 Hz, 3H); C NMR (100 MHz, CDCl3) δ 154.04, 132.57, 130.41, 124.10, 114.47, 112.26, 69.44, 31.60, 29.08, 25.69, 22.68, 14.11.
4-allyl-2-chloro-1-(hexyloxy)benzene In an oven dried schlenk flask were weighed: 4-bromo-2- chloro-1-(hexyloxy)benzene (1.01 g, 3.46 mmol), CsF (1.12 g, 7.37 mmol) and Pd(PPh3)4 (397 mg, 0.344 mmol). The flask was evacuated and backfilled with argon three times and THF (22 mL) was added. The mixture was stirred for 30 minutes at room temperature, then 2-allyl-4,4,5,5- tetramethyl-1,3,2-dioxaborolane (1.16 mL, 6.17 mmol) and THF (7.5 mL) were added. The reaction was stirred at 78 °C for 21 h. Another portion of CsF (1.12 g, 7.37 mmol), Pd(PPh3)4 (401 mg, 0.347 mmol) and THF (30 mL) were added and the mixture was stirred for 24 h at 78 °C. The mixture was allowed to cool to room temperature and was separated between pentane and H2O. The aq. layer was washed with pentane (2x). The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography eluting with heptane/EtOAc 3% v/v to yield 4-allyl-2-chloro-1-(hexyloxy)benzene as a colorless oil (789 mg,
3.12 mmol, 91%). Silica gel TLC Rf = 0.38 (Heptane/EtOAc 3% v/v); GC-MS (ESI) m/z calc. for 1 C15H21ClO: 252.78, most abundant peaks observed: 252, 168, 133, Rt = 6.21; H NMR (400 MHz,
CDCl3) δ 7.19 (d, J = 2.2 Hz, 1H), 7.00 (dd, J = 8.4, 2.2 Hz, 1H), 6.84 (d, J = 8.4 Hz, 1H), 6.01 – 5.82 (m, 1H), 5.12 – 5.01 (m, 2H), 4.00 (t, J = 6.6 Hz, 2H), 3.30 (d, J = 6.7 Hz, 2H), 1.86 – 1.77 (m,
13 2H), 1.56 – 1.43 (m, 2H), 1.39 – 1.29 (m, 4H), 0.97 – 0.83 (m, 3H); C NMR (100 MHz, CDCl3) δ 153.11, 137.26, 133.16, 130.41, 127.71, 122.94, 116.15, 113.64, 69.47, 39.18, 31.69, 29.27, 25.79, 22.74, 14.17.
90 LIGAND BINDING POCKET RXR
4-bromo-2-chloro-1-(methoxymethoxy)benzene To a solution of 4-bromo-2-chlorophenol (7.35 g,
35.4 mmol) in CH2Cl2 (50 mL) was added N,N-Diisopropylethylamine (18.5 mL, 106 mmol) and MOMCl (5.38 mL, 70.9 mL). The reaction was stirred at room temperature for 17 h. The reaction mixture was separated between CH2Cl2 and H2O. The aqueous layer was extracted twice with
CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via flash column chromatography eluting with
CH2Cl2/heptane 1:1 v/v to yield 4-bromo-2-chloro-1-(methoxymethoxy)-benzene as colorless oil. 8.9 1 g, 35.2 mmol, 99%. Silica gel TLC Rf = 0.51 (CH2Cl2 / Heptane 50% v/v); H NMR (400 MHz,
CDCl3) δ 7.51 (d, J = 2.4 Hz, 1H), 7.30 (dd, J = 8.8, 2.4 Hz, 1H), 7.05 (d, J = 8.8 Hz, 1H), 5.22 (s,
13 2H), 3.50 (s, 3H); C NMR (100 MHz, CDCl3) δ 152.30, 132.84, 130.72, 124.88, 117.74, 114.19, 95.36, 56.54.
(E)-methyl 3-(3-chloro-4-(methoxymethoxy)phenyl)acrylate In an oven dried schlenk tube were weighed: 4-bromo-2-chloro-1-(methoxymethoxy)benzene (2.0 g, 8.0 mmol), methyl acrylate (3.0 mL, 33.1 mmol), Pd(dppf)Cl2 (0.94 g, 1.3 mmol), NEt3 (35 mL, 254 mmol) and DMF (20 mL) and the reaction was stirred at 110 °C. After 20 h the reaction mixture was poured into H2O and extracted with CH2Cl2 (3 times). The combined organic layers were washed with brine, dried over
Na2SO4, filtered and evaporated in vacuo. The product was purified via column chromatography, eluting with heptane/EtOAc 20% v/v to yield (E)-methyl 3-(3-chloro-4-
(methoxymethoxy)phenyl)acrylate as white solid. 1.6 g, 6.24 mmol, 78 %. Silica gel TLC Rf = 0.34
(heptane / EtOAc 20% v/v); GC-MS (ESI) m/z calc. for C12H13ClO4: 256.05, most abundant peaks
1 observed: 256, 226, 195; H NMR (400 MHz, CDCl3) δ 7.61 – 7.55 (m, 2H), 7.36 (dd, J = 8.6, 2.0 Hz, 1H), 7.18 (d, J = 8.6 Hz, 1H), 6.33 (d, J = 15.9 Hz, 1H), 5.28 (s, 2H), 3.80 (s, 3H), 3.52 (s, 3H);
13 C NMR (100 MHz, CDCl3) δ 167.41, 154.45, 143.15, 129.75, 129.27, 128.04, 124.24, 117.37, 116.14, 95.08, 56.63, 51.87.
(E)-methyl 3-(4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)acrylate In an oven dried schlenk tube were weighed: (E)-methyl 3-(3-chloro-4- (methoxymethoxy)phenyl)acrylate (489 mg, 1.9 mmol) KOAc (573 mg, 5.84 mmol),
Bis(pinacolato)diboron (1.34 g, 5.28 mmol), Xphos (77.5 mg, 0.16 mmol) and Pd(OAc)2 (21.5 mg, 0.096 mmol). The schlenk tube was evacuated and backfilled with argon three times. Dioxane (6.5 mL) was added under a positive argon flow and the reaction was stirred at 110 °C for 5 h. The reaction mixture was then allowed to cool to room temperature and passed through celite, eluting
91 CHAPTER 3 with EtOAc. The product was purified via column chromatography eluting with 25% EtOAc in heptane. 524 mg, 1.51 mmol, 80% yield Silica gel TLC Rf = 0.25 (heptane / EtOAc 25% v/v); LC-MS 1 (ESI): calc. for C18H25BO6 [M+H]: 349.18, observed 348.92, LC, Rt = 7.42; H NMR (400 MHz,
CDCl3) δ 7.87 (d, J = 2.3 Hz, 1H), 7.65 (d, J = 16.0 Hz, 1H), 7.53 (dd, J = 8.6, 2.4 Hz, 1H), 7.03 (d, J = 8.6 Hz, 1H), 6.36 (d, J = 16.2 Hz, 1H), 5.22 (s, 2H), 3.78 (s, 3H), 3.49 (s, 3H), 1.35 (s, 12H); 13C
NMR (100 MHz, CDCl3) δ 167.82, 163.36, 144.49, 136.92, 132.38, 128.00, 116.05, 115.18, 94.78, 83.89, 56.35, 51.71, 24.97.
(E)-methyl 3-(5'-allyl-6-(methoxymethoxy)-2'-propoxy-[1,1'-biphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed: (E)-methyl 3-(4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2- dioxaborolan-2-yl)phenyl)acrylate (71.5 mg, 0.21 mmol), 4-allyl-2-chloro-1-propoxybenzene (39.5 mg, 0.18 mmol), KF (52.3 mg, 0.90 mmol), Sphos (22.2 mg, 54 μmol) and Pd2(dba)3 (16.5 mg, 18
μmol). The tube was evacuated and backfilled with argon three times. Dioxane/ H2O (5:1 v/v, 1.4 mL) was added under positive argon flow and the reaction mixture was stirred for 21 h at 110 °C. The mixture was then allowed to cool to room temperature and was passed through celite® eluting with EtOAc. The product was purified via column chromatography eluting with 17% EtOAc in heptane. 67.1 mg, 0.17 mmol, 94% yield. Silica gel TLC Rf = 0.29 (heptane / EtOAc 17% v/v); LC- 1 MS (ESI): calc. for C24H28O5 [M+H]: 397.20, observed 397.00, LC, Rt = 8.80; H NMR (400 MHz, Chloroform-d) δ 7.67 (d, J = 16.0 Hz, 1H), 7.46 – 7.42 (m, 2H), 7.19 (d, J = 9.1 Hz, 1H), 7.13 (dd, J = 8.3, 2.4 Hz, 1H), 7.06 (d, J = 2.3 Hz, 1H), 6.90 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 15.9 Hz, 1H), 5.98 (ddt, J = 16.9, 10.0, 6.7 Hz, 1H), 5.14 – 5.01 (m, 4H), 3.85 (t, J = 6.5 Hz, 2H), 3.78 (s, 3H), 3.37 (m,
13 5H), 1.62 (h, J = 7.1 Hz, 2H), 0.83 (t, J = 7.4 Hz, 3H); C NMR (100 MHz, CDCl3) δ 167.90, 156.96, 155.02, 144.81, 137.88, 131.81, 131.62, 131.60, 129.72, 128.93, 128.92, 127.90, 127.45, 115.84, 115.72, 115.18, 112.60, 94.95, 70.35, 56.17, 51.71, 39.53, 22.73, 10.57.
(E)-3-(5'-allyl-6-hydroxy-2'-propoxy-[1,1'-biphenyl]-3-yl)acrylic acid 6 M HCl (0.4 mL) was added to a solution of methyl (E)-3-(5'-allyl-6-(methoxymethoxy)-2'-propoxy-[1,1'-biphenyl]-3-yl)acrylate (30.6 mg, 77 μmol) in THF (0.4 mL) and the reaction was stirred at room temperature. After 18 h. the reaction mixture was diluted with H2O and extracted with Et2O three times. The combined organic layers were washed with sat. aq. NaHCO3 and brine, dried over Na2SO4, filtered and evaporated in vacuo to obtain methyl (E)-3-(5'-allyl-6-hydroxy-2'-propoxy-[1,1'-biphenyl]-3-yl)acrylate. LC-MS (ESI): calc. for C22H24O4 [M+H]: 353.18 observed 353.08, LC, Rt=8.05 min. The material was redissolved in dioxane/MeOH (14:5 v/v, 1.1 mL) and 4 M NaOH (135 μL, 0.54 mmol) was added. The reaction
92 LIGAND BINDING POCKET RXR was stirred at room temperature for 22 h. The solvent was evaporated and the product was purified via preparative HPLC. LC-MS (ESI): calc. for C21H22O4 [M+H]: 339.16 observed 339.0, LC, Rt=6.82
1 min. H NMR (400 MHz, CDCl3) δ 7.79 (d, J = 15.9 Hz, 1H), 7.52 (dd, J = 8.4, 2.2 Hz, 1H), 7.47 (d, J = 2.2 Hz, 1H), 7.22 (dd, J = 8.4, 2.3 Hz, 1H), 7.16 (d, J = 2.2 Hz, 1H), 7.04 (d, J = 8.4 Hz, 1H), 7.00 (d, J = 8.4 Hz, 1H), 6.35 (d, J = 15.9 Hz, 1H), 5.98 (ddt, J = 16.8, 10.1, 6.7 Hz, 1H), 5.15 – 5.07 (m, 2H), 4.02 (t, J = 6.5 Hz, 2H), 3.41 (d, J = 6.6 Hz, 2H), 1.76 (h, J = 7.4 Hz, 2H), 0.94 (t, J = 7.4
13 Hz, 3H); C NMR (100 MHz, CDCl3) δ 171.78, 156.66, 153.39, 147.04, 137.37, 134.45, 132.68, 132.27, 129.84, 129.43, 127.28, 127.21, 126.82, 118.47, 116.25, 114.65, 113.69, 71.77, 39.50, 22.62, 10.52; + HRMS (m/z): [M + H] calcd for C21H22O4, 339.1596, found 339.1592.
(E)-methyl 3-(5'-allyl-2'-(hexyloxy)-6-(methoxymethoxy)-[1,1'-biphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed: (E)-methyl 3-(4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2- dioxaborolan-2-yl)phenyl)acrylate (66.1 mg, 0.19 mmol), 4-allyl-2-chloro-1-(hexyloxy)benzene (40.0 mg, 0.16 mmol), KF (48.6 mg, 0.84 mmol), Sphos (19.7 mg, 48 μmol) and Pd2(dba)3 (15.9 mg, 17
μmol). The tube was evacuated and backfilled with argon three times. Dioxane/ H2O (5:1 v/v, 1.2 mL) was added under positive argon flow and the reaction mixture was stirred for 21 h at 110 °C. The mixture was then allowed to cool to room temperature and was passed through celite® eluting with EtOAc. The product was purified via column chromatography eluting with 17% EtOAc in heptane. 67.5 mg, 0.15 mmol, 98% yield. Silica gel TLC Rf = 0.39 (heptane / EtOAc 17% v/v); LC- 1 MS (ESI): calc. for C27H34O5 [M+H]: 439.25, observed 439.08, LC, Rt = 9.63 min.; H NMR (399 MHz, Chloroform-d) δ 7.79 – 7.59 (m, 2H), 7.48 – 7.38 (m, 3H), 7.23 – 7.04 (m, 3H), 6.90 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 16.0 Hz, 1H), 5.98 (ddt, J = 16.8, 10.0, 6.7 Hz, 1H), 5.13 – 5.01 (m, 4H), 3.88 (t, J = 6.5 Hz, 2H), 3.78 (s, 3H), 3.37 – 3.34 (m, 5H), 1.57 (q, J = 6.7 Hz, 1H), 1.27 – 1.16 (m,
13 6H), 0.84 – 0.78 (m, 3H); C NMR (100 MHz, cdcl3) δ 167.88, 156.92, 155.02, 144.80, 137.86, 131.77, 131.61, 131.58, 129.10, 128.92, 128.53, 127.86, 127.41, 115.80, 115.72, 115.15, 112.61, 94.89, 68.87, 56.16, 51.70, 39.52, 31.58, 29.32, 25.77, 22.68, 14.11.
(E)-3-(5'-allyl-2'-(hexyloxy)-6-hydroxy-[1,1'-biphenyl]-3-yl)acrylic acid 6 M HCl (0.2 mL) was added to a solution of methyl (E)-3-(5'-allyl-2'-(hexyloxy)-6-(methoxymethoxy)-[1,1'-biphenyl]-3-yl)acrylate (17 mg, 39 μmol) in THF (0.2 mL) and the reaction was stirred at room temperature. After 18 h. the reaction mixture was diluted with H2O and extracted with Et2O three times. The combined organic layers were washed with sat. aq. NaHCO3 and brine, dried over Na2SO4, filtered and evaporated in
93 CHAPTER 3 vacuo to obtain methyl (E)-3-(5'-allyl-2'-(hexyloxy)-6-hydroxy-[1,1'-biphenyl]-3-yl)acrylate. LC-MS
(ESI): calc. for C25H30O4 [M+H]: 395.22 observed 395.17, LC, Rt=8.97 min. The material was redissolved in dioxane/MeOH (14:5 v/v, 0.6 mL) and 4 M NaOH (69 μL, 0.27 mmol) was added. The reaction was stirred at room temperature for 40 h. The solvent was evaporated and the product was purified via preparative HPLC. LC-MS (ESI): calc. for C24H28O4 [M+H]: 381.21 observed 381.00,
1 LC, Rt=7.83 min; H NMR (400 MHz, CDCl3) δ 7.79 (d, J = 15.9 Hz, 1H), 7.54 – 7.45 (m, 2H), 7.24 – 7.15 (m, 2H), 7.04 (d, J = 8.4 Hz, 1H), 7.00 (d, J = 8.4 Hz, 1H), 6.35 (d, J = 15.9 Hz, 1H), 5.98 (ddt, J = 16.8, 10.0, 6.7 Hz, 1H), 5.15 – 5.07 (m, 2H), 4.04 (t, J = 6.6 Hz, 2H), 3.41 (d, J = 6.7 Hz, 2H), 13 1.72 (p, J = 6.8 Hz, 2H), 1.38 – 1.20 (m, 6H), 0.83 (t, J = 6.8 Hz, 3H); C NMR (100 MHz, CDCl3) δ 172.27, 156.68, 153.45, 147.06, 137.38, 134.46, 132.64, 132.26, 129.84, 129.44, 127.30, 127.22, 126.88, 118.49, 116.24, 114.76, 113.78, 70.35, 39.51, 31.48, 29.13, 25.57, 22.59, 14.05; HRMS (m/z): + [M + H] calcd for C24H28O4, 381.2066, found 381.2063.
3-chloro-4-methyl-1,1'-biphenyl In an oven dried schlenk tube were weighed: KOAc (354 mg, 3.61 mmol), 2-chloro-4-iodo-1-methylbenzene (321 mg, 1.27 mmol), phenylboronic acid (226 mg, 1.85 mmol) and Pd(dppf)Cl2 (95 mg, 0.13 mmol). The tube was evacuated and backfilled with argon three times and dioxane/H2O (5:1 v/v, 3 mL) was added under positive argon flow. The reaction was stirred at 90 °C for 3.5 h. The reaction mixture was allowed to cool to room temperature and was separated between CH2Cl2 and H2O. The aqueous layer was extracted twice with CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatograpphy eluting with heptane to afford 3-chloro-4- methyl-1,1'-biphenyl as colorless oil, 124 mg, 0.61 mmol, 51% yield. Silica gel TLC Rf = 0.49
(heptane); GC-MS (ESI) m/z calc. for C13H11Cl: 202.05, most abundant peaks observed: 202, 167;
1 13 H NMR (400 MHz, CDCl3) δ 7.59 – 7.21 (m, 8H), 2.39 (s, 3H); C NMR (100 MHz, CDCl3) δ 140.56, 139.88, 134.98, 134.87, 131.35, 128.96, 127.70, 127.67, 127.03, 125.37, 19.85.
(E)-methyl 3-(6-(methoxymethoxy)-6'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed: (E)-methyl 3-(4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2- dioxaborolan-2-yl)phenyl)acrylate (102.7 mg, 0.30 mmol), 3-chloro-4-methyl-1,1'-biphenyl (47.8 mg, 0.24 mmol), KF (71.9 mg, 1.24 mmol), Sphos (32.2 mg, 78 μmol) and Pd2(dba)3 (22.9 mg, 25
μmol). The tube was evacuated and backfilled with argon three times. Dioxane/H2O (6:1 v/v, 2 mL) was added under a positive argon flow and the reaction was stirred for 21 h at 110 °C. The mixture
94 LIGAND BINDING POCKET RXR was then allowed to cool to room temperature and was passed through celite® eluting with EtOAc. The product was purified via column chromatography eluting with 17 % v/v EtOAc in heptane.
Yield: 70%, 64.6 mg, 0.17 mmol. Silica gel TLC Rf = 0.29 (Heptane/EtOAc 17% v/v); LC-MS (ESI): 1 calc. for C25H24O4 [M+H]: 389.18 observed 389.08, LC, Rt=8.65 min. H-NMR (400 MHz, CDCl3): δ (ppm) 7.69 (d, J = 16.0 Hz, 1H), 7.63 – 7.59 (m, 2H), 7.57 – 7.32 (m, 9H), 6.36 (d, J = 16.0 Hz,
13 1H), 5.13 (s, 2H), 3.79 (s, 3H), 3.36 (s, 3H), 2.21 (s, 3H); C NMR (100 MHz, CDCl3) δ 167.76, 156.42, 144.42, 140.90, 138.60, 138.39, 135.96, 132.41, 130.98, 130.30, 129.20, 128.87, 128.73, 128.30, 127.22, 127.08, 126.37, 116.30, 115.18, 94.80, 56.35, 51.75, 19.79.
(E)-3-(6-hydroxy-6'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylic acid To a solution of (E)-methyl 3-(6- (methoxymethoxy)-6'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate (32.3 mg, 83 μmol) in THF (0.43 mL) was added 6M HCl (0.42 mL, 2.5 mmol). The reaction was stirred at room temperature. After 18 h the mixture was diluted with H2O and extracted with Et2O three times. The combined organic layers werewashed with sat. aq. NaHCO3 and brine, dried over Na2SO4, filtered and evaporated in vacuo to yield methyl (E)-3-(6-hydroxy-6'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate. LC-MS (ESI): calc. for C23H20O3 [M+H]: 354.15 observed 354.17, LC, Rt=7.78 min.
The material was redissolved in dioxane/MeOH (14:1 v/v, 1.3 mL) and 4M NaOH (150 μL) was added. The reaction was stirred overnight at 40 °C. The solvent was evaporated and the product was purified via preparative HPLC. LC-MS (ESI): calc. for C22H18O3 [M+H]: 331.13 observed 331.17,
1 LC, Rt=6.67 min; H NMR (400 MHz, CDCl3) δ 7.76 (d, J = 15.9 Hz, 1H), 7.63 – 7.57 (m, 3H), 7.52 (dd, J = 8.5, 2.2 Hz, 1H), 7.50 – 7.32 (m, 6H), 7.04 (d, J = 8.4 Hz, 1H), 6.33 (d, J = 15.9 Hz, 1H),
13 2.21 (s, 3H); C NMR (100 MHz, CDCl3) δ 172.26, 155.19, 146.75, 140.26, 139.90, 136.44, 135.16, 131.56, 130.85, 129.92, 129.15, 129.02, 128.52, 127.73, 127.63, 127.10, 127.06, 116.30, 115.05, 19.54. + HRMS (m/z): [M + H] calcd for C22H18O3, 331.1334, found 331.1328.
3-chloro-2-methyl-1,1'-biphenyl In an oven dried schlenk tube were weighed: KOAc (584 mg, 5.95 mmol), 1-chloro-3-iodo-2-methylbenzene (275 μL, 1.98 mmol), phenylboronic acid (364 mg, 2.99 mmol) and Pd(dppf)Cl2 (148 mg, 0.2 mmol). The tube was evacuated and backfilled with argon three times and dioxane/H2O (5:1 v/v, 5 mL) was added under positive argon flow. The reaction was stirred at 90 °C for 5.5 h. The reaction mixture was allowed to cool to room temperature and was separated between CH2Cl2 and H2O. The aqueous layer was extracted twice with CH2Cl2. The
95 CHAPTER 3
combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatograpphy eluting with heptane to afford 3-chloro-2- methyl-1,1'-biphenyl as colorless oil, 229 mg, 1.13 mmol, 57% yield. Silica gel TLC Rf = 0.56
(heptane); GC-MS (ESI) m/z calc. for C13H11Cl: 202.05, most abundant peaks observed: 202, 167;
1 13 H NMR (400 MHz, CDCl3) δ 7.46 – 7.08 (m, 8H), 2.27 (s, 3H); C NMR (100 MHz, CDCl3) δ 144.11, 141.57, 135.40, 133.84, 129.27, 128.42, 128.29, 127.28, 126.49, 17.99.
(E)-methyl 3-(6-(methoxymethoxy)-2'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate In an oven dried schlenk tube were weighed: (E)-methyl 3-(4-(methoxymethoxy)-3-(4,4,5,5-tetramethyl-1,3,2- dioxaborolan-2-yl)phenyl)acrylate (103.6 mg, 0.30 mmol), 3-chloro-2-methyl-1,1'-biphenyl (51.6 mg,
0.26 mmol), KF (71.8 mg, 1.24 mmol), Sphos (31.7 mg, 77 μmol) and Pd2(dba)3 (24.8 mg, 27 μmol).
The tube was evacuated and backfilled with argon three times. Dioxane/H2O (6:1 v/v, 2 mL) was added under a positive argon flow and the reaction was stirred for 21 h at 110 °C. The mixture was then allowed to cool to room temperature and was passed through celite® eluting with EtOAc. The product was purified via column chromatography eluting with 17 % v/v EtOAc in heptane, (E)- methyl 3-(6-(methoxymethoxy)-2'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate was obtained as a colorless oil in a quantitative yield. Silica gel TLC Rf = 0.31 (heptane/EtOAc 17% v/v); LC-MS (ESI): 1 calc. for C25H24O4 [M+H]: 389.18 observed 389.08, LC, Rt=8.63 min. H-NMR (400 MHz, CDCl3): δ (ppm) 7.68 (d, J = 16.0 Hz, 1H), 7.49 (dd, J = 8.6, 2.3 Hz, 1H), 7.44 – 7.33 (m, 6H), 7.30 – 7.21 (m, 3H), 7.18 (dd, J = 6.9, 2.1 Hz, 1H), 6.35 (d, J = 16.0 Hz, 1H), 5.14 (s, 2H), 3.79 (s, 3H), 3.37 (s,
13 3H), 2.02 (s, 3H); C NMR (100 MHz, CDCl3) δ 167.77, 156.38, 144.46, 142.45, 142.43, 138.75, 134.27, 132.90, 131.03, 129.50, 129.16, 129.10, 128.26, 128.19, 126.92, 125.33, 116.25, 115.08, 94.72, 56.33, 51.74, 18.18.
(E)-3-(6-hydroxy-2'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylic acid To a solution of (E)-methyl 3-(6- (methoxymethoxy)-2'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate (50.8 mg, 0.13 mmol) in THF (0.7 mL) was added 6M HCl (0.65 mL, 3.9 mmol). The reaction was stirred at room temperature. After
18 h the mixture was diluted with H2O and extracted with Et2O three times. The combined organic layers werewashed with sat. aq. NaHCO3 and brine, dried over Na2SO4, filtered and evaporated in vacuo to yield methyl (E)-3-(6-hydroxy-6'-methyl-[1,1':3',1''-terphenyl]-3-yl)acrylate. LC-MS (ESI): calc. for C23H20O3 [M+H]: 354.15 observed 354.17, LC, Rt=7.78 min.
The material was redissolved in dioxane/MeOH (14:1 v/v, 1.3 mL) and 4M NaOH (150 μL) was added. The reaction was stirred overnight at 40 °C. The solvent was evaporated and the product was purified via preparative HPLC. LC-MS (ESI): calc. for C22H18O3 [M+H]: 331.13 observed 331.17,
1 LC, Rt=6.67 min. H NMR (400 MHz, CDCl3) δ 7.76 (d, J = 15.9 Hz, 1H), 7.51 (dd, J = 8.5, 2.2 Hz, 1H), 7.47 – 7.32 (m, 8H), 7.25 – 7.23 (m, 1H), 7.04 (d, J = 8.4 Hz, 1H), 6.34 (d, J = 15.9 Hz, 1H),
13 2.04 (s, 3H); C NMR (100 MHz, CDCl3) δ 172.46, 155.20, 146.81, 143.84, 141.79, 135.45, 135.00,
96 LIGAND BINDING POCKET RXR
130.90, 130.83, 129.85, 129.67, 129.37, 129.05, 128.37, 127.26, 127.03, 126.53, 116.25, 115.05, 17.94. + HRMS (m/z): [M + H] calcd for C22H18O3, 331.1334, found 331.1331.
References
1. Ross, S. A., McCaffery, P. J., Drager, U. C. & De Luca, L. M. Retinoids in embryonal development. Physiol. Rev. 80, 1021–1054 (2000). 2. Barnard, J. H., Collings, J. C., Whiting, A., Przyborski, S. A. & Marder, T. B. Synthetic Retinoids: Structure–Activity Relationships. Chem. – Eur. J. 15, 11430–11442 (2009). 3. Vieira, A. V., Schneider, W. J. & Vieira, P. M. Retinoids: transport, metabolism, and mechanisms of action. J. Endocrinol. 146, 201–207 (1995). 4. Muindi, J. R., Young, C. W. & Warrell, R. P. Clinical pharmacology of all-trans retinoic acid. Leukemia 8, 1807–1812 (1994). 5. Dawson, M. I. & Xia, Z. The retinoid X receptors and their ligands. Biochim. Biophys. Acta 1821, 21–56 (2012). 6. Vaz, B. & de Lera, Á. R. Advances in drug design with RXR modulators. Expert Opin. Drug Discov. 7, 1003–1016 (2012). 7. Pérez, E., Bourguet, W., Gronemeyer, H. & de Lera, A. R. Modulation of RXR function through ligand design. Biochim. Biophys. Acta 1821, 57–69 (2012). 8. Bourguet, W., Ruff, M., Chambon, P., Gronemeyer, H. & Moras, D. Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha. Nature 375, 377–382 (1995). 9. Egea, P. F. et al. Crystal structure of the human RXRalpha ligand-binding domain bound to its natural ligand: 9-cis retinoic acid. EMBO J. 19, 2592–2601 (2000). 10.Kotani, H., Tanabe, H., Mizukami, H., Amagaya, S. & Inoue, M. A naturally occurring rexinoid, honokiol, can serve as a regulator of various retinoid x receptor heterodimers. Biol. Pharm. Bull. 35, 1–9 (2012). 11. Scheepstra, M. et al. A Natural-Product Switch for a Dynamic Protein Interface. Angew. Chem. Int. Ed. 53, 6443–6448 (2014). 12. Lengqvist, J. et al. Polyunsaturated fatty acids including docosahexaenoic and arachidonic acid bind to the retinoid X receptor alpha ligand-binding domain. Mol. Cell. Proteomics MCP 3, 692– 703 (2004). 13. Zhang, H. et al. Danthron functions as a retinoic X receptor antagonist by stabilizing tetramers of the receptor. J. Biol. Chem. 286, 1868–1875 (2011). 14. Gampe, R. T., Jr et al. Structural basis for autorepression of retinoid X receptor by tetramer formation and the AF-2 helix. Genes Dev. 14, 2229–2241 (2000). 15. Roberts, A. B., Nichols, M. D., Newton, D. L. & Sporn, M. B. In vitro metabolism of retinoic acid in hamster intestine and liver. J. Biol. Chem. 254, 6296–6302 (1979). 16.Napoli, J. L., McCormick, A. M., Schnoes, H. K. & DeLuca, H. F. Identification of 5,8-oxyretinoic acid isolated from small intestine of vitamin A-deficient rats dosed with retinoic acid. Proc. Natl. Acad. Sci. U. S. A. 75, 2603–2605 (1978). 17. Pogenberg, V. et al. Characterization of the interaction between retinoic acid receptor/retinoid X receptor (RAR/RXR) heterodimers and transcriptional coactivators through structural and fluorescence anisotropy studies. J. Biol. Chem. 280, 1625–1633 (2005). 18. Simoni, D. et al. Studies on the apoptotic activity of natural and synthetic retinoids: discovery of a new class of synthetic terphenyls that potently support cell growth and inhibit apoptosis in neuronal and HL-60 cells. J. Med. Chem. 48, 4293–4299 (2005).
97 CHAPTER 3
19.Benbrook, D. M. et al. Biologically Active Heteroarotinoids Exhibiting Anticancer Activity and Decreased Toxicity. J. Med. Chem. 40, 3567–3583 (1997). 20.Büttner, M. W. et al. Silicon Analogues of the Retinoid Agonists TTNPB and 3-Methyl-TTNPB, Disila-TTNPB and Disila-3-methyl-TTNPB: Chemistry and Biology. ChemBioChem 8, 1688– 1699 (2007). 21. Lippert, W. P. et al. Silicon analogues of the RXR-selective retinoid agonist SR11237 (BMS649): chemistry and biology. ChemMedChem 4, 1143–1152 (2009). 22.Sussman, F. & de Lera, A. R. Ligand recognition by RAR and RXR receptors: binding and selectivity. J. Med. Chem. 48, 6212–6219 (2005). 23.Dawson, M. I. et al. Conformationally restricted retinoids. J. Med. Chem. 27, 1516–1531 (1984). 24.Dawson, M. I. et al. Conformational Effects on Retinoid Receptor Selectivity. 2. Effects of Retinoid Bridging Group on Retinoid X Receptor Activity and Selectivity. J. Med. Chem. 38, 3368–3383 (1995). 25.Christie, V. B. et al. Synthesis and evaluation of synthetic retinoid derivatives as inducers of stem cell differentiation. Org. Biomol. Chem. 6, 3497–3507 (2008). 26.Zhang, H., Chen, L., Chen, J., Jiang, H. & Shen, X. Structural basis for retinoic X receptor repression on the tetramer. J. Biol. Chem. 286, 24593–24598 (2011). 27.Nakayama, M. et al. Discovery of a Potent Retinoid X Receptor Antagonist Structurally Closely Related to RXR Agonist NEt-3IB. ACS Med. Chem. Lett. 2, 896–900 (2011). 28.Nahoum, V. et al. Modulators of the structural dynamics of the retinoid X receptor to reveal receptor function. Proc. Natl. Acad. Sci. 104, 17323–17328 (2007). 29.Takahashi, B. et al. Novel Retinoid X Receptor Antagonists: Specific Inhibition of Retinoid Synergism in RXR−RAR Heterodimer Actions. J. Med. Chem. 45, 3327–3330 (2002). 30.Lefebvre, P., Benomar, Y. & Staels, B. Retinoid X receptors: common heterodimerization partners with distinct functions. Trends Endocrinol. Metab. TEM 21, 676–683 (2010). 31. de Lera, Á. R., Krezel, W. & Rühl, R. An Endogenous Mammalian Retinoid X Receptor Ligand, At Last! ChemMedChem 11, 1027–1037 (2016). 32.Chan, L. S. A. & Wells, R. A. Cross-Talk between PPARs and the Partners of RXR: A Molecular Perspective. PPAR Res. 2009, 925309 (2009). 33. le Maire, A. et al. Activation of RXR-PPAR heterodimers by organotin environmental endocrine disruptors. EMBO Rep. 10, 367–373 (2009). 34.Chandra, V. et al. Structure of the intact PPAR-gamma-RXR- nuclear receptor complex on DNA. Nature 456, 350–356 (2008). 35. Lou, X. et al. Structure of the retinoid X receptor α–liver X receptor β (RXRα–LXRβ) heterodimer on DNA. Nat. Struct. Mol. Biol. 21, 277–281 (2014). 36.Michellys, P.-Y. et al. Design, Synthesis, and Structure−Activity Relationship Studies of Novel 6,7-Locked-[7-(2-alkoxy-3,5-dialkylbenzene)-3-methylocta]-2,4,6-trienoic Acids. J. Med. Chem. 46, 4087–4103 (2003). 37. Leibowitz, M. D. et al. Biological characterization of a heterodimer-selective retinoid X receptor modulator: potential benefits for the treatment of type 2 diabetes. Endocrinology 147, 1044–1053 (2006). 38.Sundén, H. et al. Chiral Dihydrobenzofuran Acids Show Potent Retinoid X Receptor-Nuclear Receptor Related 1 Protein Dimer Activation. J. Med. Chem. 59, 1232–1238 (2016). 39.Morita, K. et al. Selective allosteric ligand activation of the retinoid X receptor heterodimers of NGFI-B and Nurr1. Biochem. Pharmacol. 71, 98–107 (2005). 40.Billingsley, K. L., Barder, T. E. & Buchwald, S. L. Palladium-catalyzed borylation of aryl chlorides: scope, applications, and computational studies. Angew. Chem. Int. Ed Engl. 46, 5359– 5363 (2007).
98 LIGAND BINDING POCKET RXR
41. Abad-Zapatero, C. Ligand efficiency indices for effective drug discovery. Expert Opin. Drug Discov. 2, 469–488 (2007). 42.Gunther, J. R., Parent, A. A. & Katzenellenbogen, J. A. Alternative inhibition of androgen receptor signaling: peptidomimetic pyrimidines as direct androgen receptor/coactivator disruptors. ACS Chem. Biol. 4, 435–440 (2009). 43.Kojetin, D. J. et al. Structural mechanism for signal transduction in RXR nuclear receptor heterodimers. Nat. Commun. 6, 8013 (2015). 44.Ohsawa, F., Morishita, K.-I., Yamada, S., Makishima, M. & Kakuta, H. Modification at the Lipophilic Domain of RXR Agonists Differentially Influences Activation of RXR Heterodimers. ACS Med. Chem. Lett. 1, 521–525 (2010). 45.Kawata, K. et al. RXR partial agonist produced by side chain repositioning of alkoxy RXR full agonist retains antitype 2 diabetes activity without the adverse effects. J. Med. Chem. 58, 912– 926 (2015). 46.Ito, M., Fukuzawa, K., Mochizuki, Y., Nakano, T. & Tanaka, S. Ab initio fragment molecular orbital study of molecular interactions between liganded retinoid X receptor and its coactivator: roles of helix 12 in the coactivator binding mechanism. J. Phys. Chem. B 111, 3525–3533 (2007).
Processing and refinement of the crystallographic data were done by Sebastian Andrei.
99
100
Chapter 4
Synthesis and evaluation of a spiroketal as RXRα modulator
The retinoid X receptor (RXR) belongs to the superfamily of nuclear receptor gene transcription factors, and is involved in many biological processes such as for example metabolism and development. Ligand dependency is a characteristic of this protein family that can be exploited for the selective modulation of their transcriptional activity by small molecules. Previous work has highlighted the capability of bis-benzannulated 6,6- spiroketals to adopt an L-shaped conformation, thereby predisposing itself to RXR binding. Spiroketals are found abundantly in polyketide-derived nature products, some of which exhibit useful antibiotic properties, a precise structure-activity relationship for these compounds has rarely been established. Furthermore, spiroketals have to the best of our knowledge never been explored as nuclear receptor modulators. Despite the biological prevalidation of spiroketals, structural evidence for spiroketals as ligands for proteins is scarce. This chapter presents the rational design, chemical synthesis, biological evaluation and structure elucidation of a spiroketal-derived agonist of RXR. The key steps in the synthesis include a Grignard reaction to forge the hydrocarbon backbone of the ligand, and thermodynamically controlled spirocyclisation to form the racemic bisbenzannulated [6,6]spiroketal ring structure. Results from biochemical cellular studies concluded that the racemic spiroketal 1 functioned as a partial agonist of RXRα. Subsequent X-ray co- crystallization clearly showed the occupancy of the ligand binding pocket with the R- enantiomer of the spiroketal, which was coherent with the initial binding model. These findings would represent the first spiroketal-derived modulator of any nuclear receptor. The structural flexibility of the spiroketal scaffold justifies further exploration of this molecular scaffold for the selective modulation of nuclear receptors other than RXR
101 CHAPTER 4
Spiroketals are a recurring structural element in many biological active products ranging from insect pheromones to complex antibiotics (figure 4.1).1–3 The pheromones such as conophthorin and chalcogran play a key role in the communication between plants, insects and microbes. These natural chiral spiroketal skeletons have gained interest in drug discovery, however the methods for enantioselective syntheses from achiral building blocks are limited.4 The more complex rubromycin family (figure 4.1) is a class of antibiotics isolated from Streptomyces collinus and the first members were isolated in 1953 by Brockmann and Renneberg.5–7 This class of natural products display attractive biological activities, They display potent antimicrobial properties and β- and γ-rubromycin can inhibit growth of various bacteria and fungi at nanomolar concentrations. More interestingly, recent discoveries have demonstrated that these molecules can arrest the cell cycle via inhibitions of telomerases, which play a critical role in reconstructing the chromosome termini during cell division. Inhibition of this reconstruction leads ultimately to apoptosis, combined with the high telomerase activity in cancer cells makes telomerases potential drug targets. The most important pharmacophore in the interaction with human telomerase proved to be the spiroketal moiety in the ligands as α-rubromicin without the spiroketal moiety did not show any activity.2 Therefore, because of their biological relevance the spiroketal scaffold remains structurally highly intriguing and could be used for potential therapeutic agents.
Figure 4.1 | Spiroketal structures. Structures of the insect pheromones conophthorin and chalcogran.8 Structures of the rubromycin family of antibiotics, β γ-rubromycin.2 The 5,6- or 6,6-spiroketal core is often disconnected to a dihydroxy ketone and formed under acidic conditions into a racemic mixture.9 Spiroketal stereoisomers have a bis-axial conformation and benefit from conformational anomeric effects.10
102 SPIROKETALS FOR NUCLEAR RECEPTORS
Although the crystal structure of several spiroketal-derived compounds have been reported, the co-crystal structure of bistramide A, bound to actin is the only one in complex with a protein.11 In addition to antiproliferative properties, bistramide A also alters the voltage dependence of the muscle twitch tension and inhibits Na+ conductance when bound to actin.12,13 The spiroketal subunit in bistramide A is crucial for binding to actin and occupies a hydrophobic pocket, the observed conformation of the spiroketal moiety in bistramide A was also bis-axial (figure 4.2). The often potent and selective biological properties of spiroketal-derived natural products has stimulated interest in their investigation as drug compounds. However, difficulties with producing sufficient quantities via synthetic means or through isolation from host sources has prohibited the clinic evaluation of spiroketal-derived natural products and instead triggered an interest in the synthesis of spiroketal analogs.14,15 The synthesis of rubromycin analogs in particular has gained significant attention, with varying degrees of success.2,3 In attempts of their synthesis, the 5,6-spiroketal core of these natural products is often disconnected to a dihydroxy ketone (figure 4.1).16,17 However, the synthesis of 6,6-bis-benzannulated spiroketals structures is less common and only a few methods have been reported.1,10 An interesting racemic synthesis of asymmetrically substituted 6,6-bisbenzanulated spiroketals was reported by Brimble and coworkers,18 where the crystal structure of the molecule revealed an “L”-shaped structure in a bis-axial conformation (figure 4.2).
Figure 4.2 | Two X-ray crystal structures of spiroketals. Compound-11 was crystallized by Brimble and coworkers.18 Bistramide A was co-crystallized with actin by Kozmin and coworkers, PDB entry 2FXU.11 Both crystal structures show the spiroketal in the thermodynamically most stable bis-axial conformation.
Nuclear receptors are transcription factors of which the function and activity can be modulated via small lipophilic ligands. Ligands that target the retinoid-X receptor (RXR)
103 CHAPTER 4 are found to have potential therapeutic properties for cancer, diabetes and Alzheimer’s disease.19,20 The RXR LBP adopts an L-shaped conformation unique to RXR over other nuclear receptors, especially the homologous retinoid acid receptor (RAR), which aids in the design of novel ligands specific for RXR. RXR is functional as either a homodimer or heterodimer with other nuclear receptors, which makes RXR unique among the nuclear receptor family and accounts for their many diverse signaling pathways. It is generally held that 9-cis-retinoic acid is the endogenous ligand for RXR. There are several reports on synthetic ligands for RXR,21,22 but the chemical diversity and functional outcome of these ligands is arguably limited as evidence by the fact that selective RXR heterodimerization and partial RXR agonism – both in principle addressable by ligand binding to the ligand binding pocket – remain unmet biomedical challenges. New molecular scaffolds expanding on the current repertoire for RXR modulation are sought to meet these evident challenges. In our efforts to make selective nuclear receptor modulators23,24 and to identify the potential of spiroketal molecules we herein present the design, synthesis, biochemical evaluation and the co-crystallization of the spiroketal structure 1 with RXRα. Spiroketals have been found in natural products with many biological properties, but they would represent a new molecular scaffold for nuclear receptor ligands. This represents the first report of a bis-benzannulated spiroketal molecule bound to a protein.
Figure 4.3 | Docking of spiroketal (±)-1 in the RXR ligand binding pocket. Spiroketal 1 adopts an “L”-shaped conformation that fits in the L-shaped pocket of RXR. Spiroketal 1 (green) and SR11237 (yellow) display a good overlay with the carboxylic acid as anchoring point revealing the potential of (±)-1 as an RXR modulator. 9-cis-retinoic acid is the putative endogenous ligand for RXR, SR11237 is a synthetic agonist for RXR,25 and the (±)-6,6- spiroacetal crystallized by Brimble and coworkers.18
104 SPIROKETALS FOR NUCLEAR RECEPTORS
Initial docking of spiroketal 1 in the L-shaped hydrophobic ligand binding pocket of RXRα in MOLOC (experimental section) revealed a good fit of the spiroketal to the L-shaped ligand binding pocket of RXR (figure 4.3). The benzoic acid is shown to make a canonical interaction with Arg316 and the backbone of Ala327, while the elongated hydrophobic ring structure occupies the lipophilic L-shaped region of the binding pocket and is in close contact to Val265, Val342, Ile345 and Phe349. To evaluate this binding hypothesis, the bisbenzannulated spiroketal 1 (figure 4.3) was chemically synthesized.
Scheme 4.1 | Explored synthesis route for the synthesis of a bis-benzannulated spiroketal.
Reagents and conditions: (a) ethyntrimethylsilane, n-Buli, THF, -78 °C; (b) Pd(PPh3)2Cl2,
CuI, NEt3, THF, rt; (c) with benzyl 3-(benzyloxy)-4-iodobenzoate, Pd(PPh3)2Cl2, CuI, K2CO3, MeOH, DMF, rt.
The first strategy explored (scheme 4.1) for the preparation of the spiroketal (±)-I relied on two sequential Sonogashira couplings. The asymmetric dialkyne 3 allows for the first regioselective Sonogashira coupling with the aryliodide. The asymmetric dialkyne was prepared from propynal and ethyntrimethylsilane in a moderate yield, and then followed by the Sonogashira coupling using standard conditions resulting in compound 4 bearing a protected alkyne. The deprotection and the second coupling were then subsequently performed without purification of the deprotected alkyne intermediate to afford 5 in a moderate yield. Recently described Pd0/CuI and Pd0/AgI sila-Sonogashira coupling conditions26–28 were also tried for this reaction, but unfortunately did not yield product. Most disappointingly, the catalytic reduction of the two alkyne groups in 5 with simultaneous deprotection of the benzyl groups did not proceed in various attempts (Pd/C
105 CHAPTER 4
or Pd(OH)2/C, H2-atm. or 70 psi pressure, acetic acid or TFA). Performing the oxidation first would yield a highly conjugated system, which might also be more difficult to reduce. Although the initial synthesis plan in general would allow for easier modifications in later stages of the synthesis, because of major challenges and stubbornness of intermediate 5 in various attempts for the catalytic reductions this strategy was discontinued. The reported synthesis for fused aromatic spiroacetals by Brimble and co-workers18 provided a straightforward alternative synthetic route to targeted bisbenzannulated spiroketals, including I and relied on the pivotal reaction of the aldehyde 15 with alkyne 20 (scheme 4.2).
Scheme 4.2 | Retrosynthetic scheme for the synthesis of the bis-benzannulated spiroketal. The synthesis relies on the key disconnection between the aldehyde and the alkyne and the deprotection reaction followed by cyclisation.
The synthesis of the aryl aldehyde 15 (Scheme 4.3) started from the methyl protected benzoate 10 via a Claisen rearrangement of the derived allylphenol 11. Protection of the Claisen rearrangement product as methoxymethyl ether (MOM) was subsequently followed by hydroboration using borane dimethylsulfide complex. The oxidation of the primary alcohol to the aldehyde was performed using a catalytic amount of TEMPO and 1 molar equivalent of trichloroisocyanuric acid29 in an overall yield of 39% over the five steps.
Scheme 4.3 | Synthesis of the aldehyde. Reagents and conditions: (a) allyl bromide, K2CO3, acetone, 65 °C; (b) chlorobenzene, microwave, 245 °C, 10 bar; (c) MOMCl, iPr2NEt, CH2Cl2,
0 °C to rt.; (d) BH3•SMe2, NaOH, H2O2, THF, 0 °C to rt.; (e) TEMPO, trichloroisocyanuric acid, CH2Cl2, rt.
106 SPIROKETALS FOR NUCLEAR RECEPTORS
The acetylene 20 was prepared starting from the double alkylation reaction of the phenol with 2,5-dichloro-2,5-dimethylhexane. Iodination of the substituted phenol 16 was then followed by the protection of the phenol with a MOM-group. The Sonogashira reaction provided the protected alkyne 19 which was deprotected via a pyrolysis reaction under basic conditions (scheme 4.4) in an overall yield of 30% over the five steps.
Scheme 4.4 | Synthesis of the acetylene. Reagents and conditions: (a) AlCl3, CH2Cl2, 40 °C;
(b) NaOH, NaI, NaClO, Na2S2O2, MeOH, 0 °C; (c) MOMCl, iPr2NEt, CH2Cl2, 0 °C to rt.;
(d) PdCl2(PPh3)2, CuI, 2-methylbut-3-yn-2-ol, NEt3, 80 °C; (e) NaOH, toluene, reflux.
alkyne 20 and aldehyde 15 were then coupled to construct the carbon skeleton of the desired spiroketal (scheme 4.5). For this reaction the lithium acetylide was generated by treatment of the alkyne with n-butyllithium. After completion of the formation of the acetylide, the aldehyde 15 was added resulting in a reasonable 58% yield of alkynol 21, which underwent hydrogenation under catalytic conditions to obtain the secondary alcohol 22, followed by oxidation to the ketone (23). The deprotection and subsequent cyclisation of 23 was performed using the Lewis acid trimethylsilyl bromide to bisbenzannulated [6,6]- spiroketal 24 in a yield of 64%. The 13C NMR spectrum showed the characteristic
1 quaternary carbon signal for 24 at δc 97.03 ppm, while the H NMR spectrum showed the characteristic signals for the protons of the methylene groups in the ring system (figure 4.5).18
107 CHAPTER 4
Figure 4.5 | NMR data of the characteristic 1H and 13C NMR chemical shifts. Comparison of NMR of the reported spiroketal by brimble and coworkers18 and compound (±)-24.
The synthesis was then completed with hydrolysis of the methyl ester group under basic conditions. Analysis of the product via chiral column chromatography showed as expected no enantiomeric excess and a 1:1 mixture of R/S-enantiomers was obtained. The lack of asymmetric substituent groups on the sp3-hybridized ring carbons favor a 1:1 mixture under the thermodynamically driven acidic conditions used for the cyclisation reaction.30 Isolation of the two enantiomers was performed via semi-preparative chiral column chromatography, but only small amounts were obtained. Furthermore, 1 displayed stability to the acidic conditions needed for the analysis and purification (hexane/2-propanol with 0.1% formic acid) as confirmed via LC-MS. Spiroketals can in principle undergo epimerization under acidic conditions via ring opening and recyclisation. Nevertheless, the rubromycins are isolated from nature as single isomers and demonstrate remarkable stability. Furthermore, unexpected resilience against acid-catalyzed epimerization (p- toluenesulfonic acid, acetic acid, camphorsulfonic acid) for spiroketal centers has been reported previously.30 This stability is beneficial in the potential use of this molecular scaffold.
Scheme 4.5 | Synthesis of spiroketal (±)-1. Reagents and conditions: (a) n-buthyllithium,
THF, -78 °C to rt; (b) 10% Pd/C, KHCO3, EtOAc, rt.; (c) Dess-Martin periodiane, CH2Cl2, rt.; (d) TMSBr. CH2Cl2, -30 °C to rt.; (e) NaOH, dioxane/MeOH, 40 °C.
108 SPIROKETALS FOR NUCLEAR RECEPTORS
With spiroketal molecule 1 in hand the RXR-activity was profiled alongside the synthetic ligand LG100268,21 using a fluorescence-based co-factor recruitment assay where an increase in fluorescence polarization would correspond with ligand binding and stabilization of an agonistic conformation of the ligand binding domain. For spiroketal (±)-
1 an EC50(FP) = 0.54 ± 0.07 μM was measured compared to LG100268 EC50(FP) = 0.18 ± 0.02 μM, albeit with a clear partial agonistic profile for (±)-1 (figure 4.6a). To support these results, the activity of both compounds was profiled in a mammalian two-hybrid luciferase assay (figure 4.6b). Although the partial effect for (±)-1 was not observed in this assay, a concentration dependent increase in activity was measured. Moreover, the difference in potency between spiroketal (±)-1 and LG100268 is circa 100 fold using this assay format. The difference in potency for the observed overall activation between the fluorescence polarization assay and the mammalian two-hybrid assay might be explained by the use of different co-regulator peptides for the two assays. Differences between these different assays is a common phenomemon.31 Unfortunately, the separation of the two enantiomers yielded insufficient amounts of material for thorough biochemical evaluation.
Figure 4.6 | Biochemical and cellular evaluation of (±)-1 a) Fluorescence polarization assay data showing full agonist LG100268 induces binding of a fluorescently labelled coactivator peptide in a concentration-dependent manner, while spiroketal (±)-1 induces exhibits a partial agonistic behavior. b) Cellular activities of LG100268 and spiroketal (±)-1 measured in a mammalian two-hybrid luciferase assay. Error bars denote s.d. (n = 3)
To elucidate the binding mode for spiroketal (±)-1, validate the docking study and to gain insight in the binding properties, the co-crystal structure of RXR with (±)-1 was solved at 2.35 Å resolution. The X-ray structure of the RXR ligand binding domain is shown in figure 4.7a; the fold consists of the canonical 12 helices. The carboxylate group of (±)-1 is seen making a canonical interaction with Arg316 and the backbone of Ala327 (figure 4.7b). The
109 CHAPTER 4 spiroketal moiety and the tetramethylcyclohexane occupy the lipophilic region of the binding pocket. The bent conformation of the spiroketal group fits the L-shaped pocket of RXR. The overlay of the X-ray data with the initial modeling displayed a good correlation for the spiroketal (±)-1 ligand (experimental section). The aromatic substitutions in the molecule cause a flattening of the cyclohexane ring structures giving an envelope-envelope like arrangement with a bis-axial conformation which is stabilized by the anomeric effect.1 We could not find electron density in the X-ray structure for two molecules, and only the R-enantiomer of the bis-axial conformer fitted unambiguously in the electron density (figure 4.7c). Compared to the natural-derived agonists for RXR23 the hydrogen bonding interaction with Asn306 is lacking for spiroketal (±)-1. The overlay of the RXR co-crystal structure with spiroketal (±)-1 with the structure of LG10026832 did not evidently show significant changes in the overall fold of the protein as a way to explain the difference in potency for the molecules.
Figure 4.7 | Co-crystal structure of RXR with spiroketal 1. a) Ribbon representation of the X-ray co-crystal structure of (±)-1 (orange) bound to the ligand binding pocket of RXRα (green) with the cofactor derived TIF2 peptide (blue). b) Zoomed-in view of the RXRα ligand binding pocket with the amino acid side chains represented as sticks. c) Zoom in on spiroketal (±)-1, only the R-enantiomer fits in the electron density (see experimental section for the Cahn–Ingold–Prelog ordering).
In summary, the design, synthesis, biochemical evaluation and structure elucidation of the bis-benzanulated spiroketal molecule with agonistic properties for the nuclear receptor RXR has been reported. The achiral synthesis involved 15 steps with a longest linear
110 SPIROKETALS FOR NUCLEAR RECEPTORS sequence of five steps with the key steps including the Grignard reaction and the thermodynamically controlled spirocyclisation. Separation of the two enantiomers via semi preparative chiral column chromatography was successful and the spiroketal products were shown to be resilient against epimerization. The initial modeling in was validated by the X- ray structure which to the best of our knowledge represents the first report of a bis- benzanulated spiroketal bound to a protein. The biochemical evaluation revealed agonist properties for (±)-1, and these data were corroborated by a cellular mammalian two-hybrid luciferase assay. The separation of the two enantiomers unfortunately yielded insufficient material and a thorough evaluation of the two enantiomers yet needs to be performed. This represents the first report of the use of a spiroketal ligand for nuclear receptor modulation and this study justifies the further exploration of spiroketal structures for the modulation of the nuclear receptor family.
Experimental section
Molecular modeling The X-ray crystal structure of the retinoid X receptor in complex with SR11237 and a fragment of the coactivator (PDB code: 2ZXZ) was used for the modeling.25 spiroketal (±)-1 was modeled so as to occupy the ligand-binding pocket. The energy of the system was minimized using the MAB force field as implemented in the computer program MOLOC, whilst keeping the protein coordinates fixed. After energy minimization of the designed fragments, all types of interactions (hydrogen bonds and lipophilic interactions) between the fragments and the protein were analyzed in MOLOC.
Polarization assay. His-RXRα-LBD (1 μM), fluorescein-labeled D22 co-activator peptide (0.1 μM), and the ligand at the indicated concentration in the assay buffer: 10 mM HEPES, pH 7.5, 150 mM NaCl, 5 mM DTT, 0.1% v/v bovine serum albumin were incubated for 60 minutes at 4 °C and protected from light. Fluorescent polarization signals (mP) were measured with a Tecan Infinite F500 microplate reader. Experiments were performed in triplicate.
Co-crystallization of the RXR ligand binding domain. The histidine-tagged LBD of human RXRα (in a pET15b vector) was expressed in Escherichia coli BL21(DE3). Cells were grown at 37 °C in LB medium supplemented with 100 mg mL-1 ampicillin until OD600 reached about 0.7. Expression of T7 polymerase was induced by addition of isopropyl-b-d-thiogalactoside (IPTG) to a final concentration of 0.1 mM. After an additional incubation for 15 h at 15°C, cell cultures were harvested by centrifugation at 8,000 ´ g for 20 min. The cell pellet from 2 liters of RXRα LBD was resuspended in 50 ml buffer A (20 mM Tris-HCl pH 8.0, 500 mM NaCl, 5 mM imidazole) supplemented with a protease inhibitor (PMSF) and DNAse I. The suspension was then lysed by
111 CHAPTER 4 sonication and centrifuged at 35,000 g and 4 °C for 45 min. The supernatant was loaded onto a 5 ml Ni2+-affinity column, preequilibrated with buffer A. The column was washed with 10 volumes of buffer A and 10 volumes of buffer A supplemented with 50 mM imidazole. Bound proteins were eluted with buffer A containing 200 mM imidazole. The fractions containing RXRα LBD were pooled, concentrated and desalted to buffer B (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM DTT). To remove the histidine-tag the protein was incubated for 16 h at 4°C with thrombin (1 unit/mg RXR). The protein was passed through a Ni2+ column and a superdex gel filtration column. The protein was concentrated and stored at -80°C until further use.
Before crystallization the protein was mixed with a 1.5-fold molar excess of compound 1 and a 3-fold excess of TIF2 NR2 cofactor peptide (686-KHKILHRLLQDSS-698). The complex was incubated for 1h at 4 °C. Initial screening was performed using JCSG+ (Qiagen), NR LBD screen (molecular dimensionsTM) and PACT (Quigen®) at 4 °C using sitting drop vapor diffusion. 200 nL reservoir solution was automatically mixed with 200 nL protein solution using a pipetting robot (Mosquito® Crystal) in a 96-well plate. The drops were equilibrated against 75 μL of reservoir solution. After two days crystals had appeared in 6 conditions. Reproduction of the crystals were performed in 15- well plates using the hanging drop vapor diffusion method. Drops with a size of 2 - 3 μL using different reservoir to protein ratio were manually mixed and equilibrated against reservoirs with a volume of 500 μL. Optimal crystals were grown in two days in 3 μL drops with protein solution to reservoir ratio of 1:2 with: 1 M MIB buffer (Malonic acid, Imidazole, Boric acid), pH 8 25% (w/v) PEG 1500. The crystals were cryo-cooled in liquid nitrogen using sucrose as cryo-protectant for X- ray data collections.
Table S4.1 Crystallographic statistics for Spiroketal 1
Spiroketal 1 Data collection Resolution (Å) 35.59-2.35 (2.44-2.35)
Space group P43212 Cell parameters (Å) a=b=64.70, c=113.25 a,b CC1/2 (%) 99.3
Rmerge 0.11 (0.436)
a Average I/σ(I) 15.0 (4.9) Completeness (%)a 100 (100) Redundancya 10.1 (9.8)
Refinement Number of protein/solvent/ligand atoms 1775/123/30
Rwork/Rfree (%) 20.2/25.6
112 SPIROKETALS FOR NUCLEAR RECEPTORS
No. of reflections 10212 R.m.s. deviations from ideal values bond lengths 0.012 / 1.416 (Å) / bond angles (°) Average protein/solvent/ligand B-factor (Å2) 30.59/33.94/27.87 a number in parentheses is for the highest resolution shell b CC1/2 = Pearson's intra-dataset correlation coefficient, as described by Karplus and Diederichs. The values were reported by phenix.cc_star
Luciferase assays with U2OS cells. The mammalian two-hybrid (M2H) assays were performed in U2OS cells. The RXR LBD was cloned into the pCMV-AD vector (Aligent), fused to NF-κB and Strep-tag resulting in NF-κB-Strep-RXR-LBD. The peptide sequence was cloned into the pCMV-BD vector (Aligent) and fused to GAL4 DNA binding domain. For the direct interaction assay, ~40,000 cells/well were seeded in a 24-well plate for 24 h and transfected with 40 ng pCMV-AD, 40 ng pCMV-BD, 0.2 μg pFR-Luc, and 3.2 ng pGL-Renilla using PEI (Polysciences), before being treated with the ligands in indicated concentrations. After 24 h of treatment the interaction was determined with a Dual-Luciferase® Reporter Assay (Promega), according to the manufacturer’s instruction. The luminescent intensities were recorded on a Synergy HT platereader (BioTek). The FR-luciferase signal was normalized over the Renilla luciferase (pGL-Renilla) signal.
Figure S4.1 | The overlay of the model with the crystallographic data. The overlay of the docking compared to the X-ray structure displays a good reveals a good correlation.
Chiral column chromatography. Chiral column chromatography was performed on a Chiralpak IA- 3 column with a particle size of 3 μm, dimensions 2.1 mm x 100 mm, amylose immobilized on silica gel. Eluting with hexane/2-propanol (95:5 v/v) with 0.1% formic acid at a flowrate of 0.15 mL/min. Semi-preparative column chromatography was performed on a ChiralCel®OD cellulose column, 0.46x25 cm with hexane/2-propanol (98:2 v/v) with 0.1% formic acid and a flowrate of 1 mL/min. The UV signal was recorded with a PDA SPD-M10Avp between 200-350 nm.
113 CHAPTER 4
Cahn–Ingold–Prelog priority for spiroketal (±)-1
Figure S4.2 | The Cahn-Ingold-Prelog ordering for spiroketal (±)-1. The priority was determined by following the CIP priority rules. Numbers of the corresponding carbons are depicted, C0 and O0 correspond to “ghost” carbons or oxygens.
Synthetic procedures
General considerations synthetic procedures. Water was purified using a Millipore purification train. All the reagents are commercially available and used without purification. All the NMR data were recorded on a Varian Gemini 200 or 400 MHz NMR or a Bruker Cryomagnet for NMR spectroscopy 400 MHz (400 or 200 MHz for 1H-NMR and 100 or 50 MHz for 13C-NMR). Proton experiments are reported in parts per million (ppm) downfield of TMS and were relative to the residual chloroform (7.26 ppm). All 13C spectra were reported in ppm relative to residual chloroform (77 ppm) Analytical LC-MS was performed on a C4, Jupiter SuC4300A, 150x2.00 mm column with a gradient 5% to 100% acetonitrile in H2O in 15 min. Silica column chromatography was performed manually using silica with particle size 60 – 200 μm. Preparative HP-LC was performed on a
Gemini S4 110A 150x21.20 mm column using H2O with 0.1% Formic Acid (F.A.) and acetonitrile with 0.1% F.A. Purity and exact mass of the final compound was determined using a High Resolution LC-MS system consisting of a Waters ACQUITY UPLC I-Class system coupled to a Xevo G2 Quadrupole Time of Flight (Q-tof). The system was comprised of a Binary Solvent Manager and a Sample Manager with Fixed-Loop (SM-FL). compounds were separated (0.3 mL min-1) by the column (Polaris C18A reverse phase column 2.0 x 100 mm, Agilent) using a 15% to 75% acetonitrile
114 SPIROKETALS FOR NUCLEAR RECEPTORS gradient in water supplemented with 0.1% v/v formic acid before analysis in positive mode in the mass spectrometer. Compound LG100268 was commercially obtained from Sigma Aldrich used without further purification. methyl 4-(allyloxy)benzoate (11). methyl 4-hydroxy-benzoate (6.0 g, 39.4 mmol) was dissolved in acetone (44 mL). To this solution were added K2CO3 (16.3 g, 118 mmol) and 3-bromoprop-1-ene (3.5 mL, 40.2 mmol). The reaction was stirred at 65 °C for 20 h. The solvent was evaporated and the residue separated between CH2Cl2 and H2O. The aqueous layer was extracted twice with CH2Cl2.
The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated in vacuo. The product was purified via column chromatography, eluting with CH2Cl2 to obtain methyl 4-(allyloxy)benzoate as colorless oil in quantitative yield. 7.56 g, 39.3 mmol. Silica gel TLC 1 Rf = 0.53 (CH2Cl2). GCMS (EI) expected m/z: 192, most abundant peaks observed: 192, 161, 133; H
NMR (400 MHz, CDCl3) δ 7.98 (d, J = 8.8 Hz, 2H), 6.92 (d, J = 8.7 Hz, 2H), 6.09 – 6.00 (m, 1H), 5.42 (d, J = 17.3 Hz, 1H), 5.31 (d, J = 10.6 Hz, 1H), 4.58 (d, J = 5.1 Hz, 2H), 3.88 (s, 3H); 13C NMR (50
MHz, CDCl3) δ 166.92, 162.42, 132.67, 131.67, 122.82, 118.22, 114.41, 68.95, 51.97. methyl 3-allyl-4-hydroxybenzoate (12). methyl 4-(allyloxy)benzoate (1.1 g, 5.9 mmol) was dissolved in chlorobenzene (2.0 mL) in a microwave reaction vessel equipped with stirring bar. After the vessel was sealed, the sample was irradiated for 240 min at 245 °C. The reaction mixture was subsequently cooled to room temperature and the solvent was evaporated to yield methyl 3-allyl-4- hydroxybenzoate as off white solid, 1.1 g, 5.9 mmol. The material was used without further purification. Silica gel TLC Rf = 0.18 (CH2Cl2); GCMS (EI) expected m/z: 192, most abundant peaks observed: 192, 161. methyl 3-allyl-4-(methoxymethoxy)benzoate (13). methyl 3-allyl-4-hydroxybenzoate (1.1 g, 5.9 mmol)
i was dissolved in CH2Cl2 (11.5 mL). The solution was cooled to 0 °C and Pr2NEt (1.9 mL, 10.63 mmol) was added. Finally chloromethyl methyl ether (0.63 mL, 8.3 mmol) was added and the mixture was stirred at 0 °C for 30 min. The reaction was then allowed to warm to room temperature and was stirred until all the starting material was consumed. After 4 hours the reaction mixture was separated between CH2Cl2 and H2O. The aqueous layer was extracted twice with CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography eluting with CH2Cl2 to yield methyl 3-allyl-4- (methoxymethoxy)benzoate as colorless oil in 80% yield over two steps. 1.1 g, 4.74 mmol. Silica gel
TLC Rf = 0.46 (CH2Cl2); GCMS (EI) expected m/z: 236, most abundant peaks observed: 236, 191;
1 H NMR (200 MHz, CDCl3) δ 7.92 – 7.83 (m, 2H), 7.09 (d, J = 8.2 Hz, 1H), 6.10 – 5.88 (m, 1H), 5.26 (s, 2H), 5.12 – 5.00 (m, 2H), 3.88 (s, 3H), 3.48 (s, 3H), 3.42 (d, J = 6.5 Hz, 2H); 13C NMR (50
MHz, CDCl3) δ 167.07, 158.72, 136.29, 131.70, 129.73, 129.22, 123.43, 116.07, 113.04, 94.12, 56.40, 52.02, 34.46.
115 CHAPTER 4 methyl 3-(3-hydroxypropyl)-4-(methoxymethoxy) benzoate (14). methyl 3-allyl-4- (methoxymethoxy)benzoate (330.76 mg, 1.4 mmol) was dissolved in THF (15 mL) and cooled to 0 °C. Borane dimethyl sulfide complex (266 μL, 2.8 mmol) was added drop wise and the reaction was stirred for 5 h at 0 °C. The reaction mixture was allowed to warm to room temperature and was stirred overnight. The excess of borane complex was quenched with MeOH (1.5 mL). NaOH (1 M,
1.55 mL) was added carefully and finally H2O2 (35% 1.7 mL) was added drop wise. The reaction was stirred for 24 h. The reaction mixture was separated between H2O (15 mL) and Et2O (15 mL), the aqueous layer was extracted with Et2O (twice 15 mL). The combined organic layers were dried over
MgSO4, filtered and evaporated. The product was purified via column chromatography eluting with 40 % v/v EtOAC in heptane to obtain methyl 3-(3-hydroxypropyl)-4-(methoxymethoxy)benzoate as colorless oil, 190 mg, 0.75 mmol 54%. Silica gel TLC Rf = 0.19 (Hexane/EtOAc 40% v/v); LC-MS 1 (ESI): calc. for C13H18O5 [M+H]: 255.28, observed 254.83, LC, Rt=5.10 min; HNMR (400 MHz,
CDCl3): δ (ppm) 7.90 – 7.81 (m, 2H), 7.09 (d, J = 9.2 Hz, 1H), 5.26 (s, 2H), 3.88 (s, 3H), 3.87 (s, 1H), 3.66 (t, J = 6.3 Hz, 2H), 3.49 (s, 3H), 2.77 (t, J = 14.4 Hz, 2H), 1.89 (dd, J = 8.5, 6.4 Hz, 2H);
13 C NMR (50 MHz, CDCl3) δ 167.16, 158.97, 131.81, 130.70, 129.55, 123.46, 113.13, 94.34, 62.20, 56.48, 52.07, 32.77, 26.35. methyl 4-(methoxymethoxy)-3-(3-oxopropyl)benzoate (15). Methyl 3-(3-hydroxypropyl)-4-
(methoxymethoxy) benzoate (127.5 mg, 0.50 mmol) was dissolved in CH2Cl2 (1.1 mL). Trichloroisocyanuric acid (117.8 mg, 0.5 mmol) was added and the mixture was cooled to 0 °C. (2,2,6,6-Tetramethylpiperidin-1-yl)oxyl (0.8 mg, 5,0 μmol) was added and the mixture was stirred at 0 °C for 2 minutes. The mixture was then allowed to warm to room temperature and stirred for ® 20 minutes, passed through celite eluting with CH2Cl2. The organic phase was washed with sat. aq. Na2CO3, 1 M HCl, dried with Na2SO4, filtered and evaporated. The product was purified via column chomatography eluting with 35% v/v EtOAc in heptane to yield methyl 4- (methoxymethoxy)-3-(3-oxopropyl)benzoate as a white solid. 114 mg, 0.45 mmol, 90%. Silica gel
TLC Rf = 0.30 (Heptane/EtOAc 35% v/v); GC-MS (EI): calc. for: C13H16O5: 252.26, observed most
1 abundant peaks: 252, 190, 176; H NMR (400 MHz, CDCl3): δ 9.84 (t, J = 1.4 Hz, 1H), 7.91 - 7.84 (m, 2H), 7.10 (d, J = 8.5 Hz, 1H), 5.26 (s, 2H), 3.88 (s, 3H), 3.48 (s, 3H), 3.00 (t, J = 7.5 Hz, 2H),
13 2.76 (dd, J = 15.8, 8.5 Hz, 2H); C NMR (50 MHz, CDCl3) δ 201.85, 166.90, 158.87, 131.61, 130.04, 129.23, 123.52, 113.14, 94.22, 56.49, 52.07, 43.78, 23.39.
5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ol (16). 2,5-dichloro-2,5-dimethylhexane (4.0 g,
21.84 mmol) and phenol (2.1 g, 21.84 mmol) were dissolved in dry CH2Cl2 (24 mL) in an oven dried flask. To this solution ALCl3 (988 mg, 0.33 mmol) was added. The reaction was stirred at room temperature for 2 hours and then at 40 °C for 1.5 hours. The reaction mixture was cooled to 0 °C and 2 M HCl (24 mL) was poured into the mixture. The aqueous layers was extracted with Et2O
116 SPIROKETALS FOR NUCLEAR RECEPTORS
three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography eluting with 7% EtOAc in pentane to yield 5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ol as white solid. 3.2 g, 15.7 mmol, 72%. Silica gel TLC Rf = 0.37 (pentane/EtOAc 7% v/v); LC-MS (ESI): calc. for C14H20O
1 [M+H]: 205.32, observed 205.25, LC, Rt=8.20 min; H NMR (400 MHz, CDCl3) δ 7.18 (d, J = 8.5 Hz, 1H), 6.77 (d, J = 2.8 Hz, 1H), 6.64 (dd, J = 8.5, 2.8 Hz, 1H), 4.68 (s, 1H), 1.67 (s, 4H), 1.26 (s,
13 6H), 1.25 (s, 6H); C NMR (100 MHz, CDCl3) δ 153.11, 146.71, 137.45, 127.88, 113.22, 112.80, 35.28, 35.21, 34.51, 33.80, 32.14, 31.93.
3-iodo-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen -2-ol (17). To a solution of 5,5,8,8- tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ol (3.18 g, 15.56 mmol) in MeOH (37.5 mL) were added sodium iodide (2.3 g, 15.56 mmol) and sodium hydroxide (627 mg, 15.56 mmol). The mixture was cooled to 0 °C and sodium hypochlorite (5%, 21.53 mL, 15.56 mmol) was added dropwise. The reaction was stirred for 2 hours at 0 °C and then quenched with 10% w/v Na2S2O3. The pH was adjusted to 6 with 1 M HCl and the aqueous layer was extracted with Et2O three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography, eluting with 7% v/v EtOAc in heptane to yield 3- iodo-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ol as white solid. 4.21 g, 12.76 mmol,
1 82%.Silica gel TLC Rf = 0.21 (heptane/EtOAc 7% v/v); H NMR (400 MHz, CDCl3) δ 7.54 (s, 1H),
13 6.94 (s, 1H), 5.06 (s, 1H), 1.66 (s, 4H), 1.26 (s, 6H), 1.25 (s, 6H); C NMR (100 MHz, CDCl3) δ 152.49, 147.96, 140.03, 136.27, 112.82, 83.19, 77.16, 35.10, 34.96, 34.51, 33.86, 32.10, 31.83.
6-iodo-7-(methoxymethoxy)-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene (18). To a solution of
3-iodo-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ol (3.97 g, 12.02 mmol) in CH2Cl2 (23 mL) was added diisopropylethylamine (3.8 mL, 21.64 mmol) and chloro(methoxy)methane (1.2 mL, 16.83 mmol) carefully. The cooling bath was removed after 30 min and the reaction was stirred at room temperature for 5 hours. The reaction mixture was separated between CH2Cl2 and H2O. The aqeous layer was extracted twice with CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography, eluting with 7% EtOAc in heptane to obtain 6-iodo-7-(methoxymethoxy)-1,1,4,4-tetramethyl-1,2,3,4- tetra hydronaphthalene. 4.1 g, 10.97 mmol, 91%. Silica gel TLC Rf = 0.48 (heptane/EtOAc 7% v/v); 1 GC-MS (EI): calc. for C16H23IO2: 374.26, observed most abundant peaks: 374, 359, 329; H NMR
(400 MHz, CDCl3) δ 7.65 (s, 1H), 6.99 (s, 1H), 5.20 (s, 2H), 3.53 (s, 3H), 1.65 (d, J = 1.6 Hz, 4H),
13 1.26 (s, 6H), 1.25 (s, 6H); C NMR (100 MHz, CDCl3) δ 154.05, 147.00, 141.22, 137.52, 113.30, 95.51, 84.66, 56.58, 35.06, 35.04, 34.68, 33.92, 31.97, 31.84
4-(3-(methoxymethoxy)-5,5,8,8-tetramethyl-5,6,7,8-tetrahy-dronaphthalen-2-yl)-2-methylbut-3-yn-2- ol (19). An oven dried schlenk tube was charged with 6-iodo-7-(methoxymethoxy)-1,1,4,4-
117 CHAPTER 4
tetramethyl-1,2,3,4-tetrahydronaphthalene (904 mg, 2.4 mmol), PdCl2(PPh3)2 (214 mg, 0.31 mmol) and CuI (60.1 mg, 0.32 mmol). The flask was evacuated and backfilled wit argon three times, then
NEt3 (35 mL) and 2-methylbut-3-yn-2-ol (290 μL, 2.9 mmol) were added under a positive argon flow. The reaction was stirred at 80 °C for 20 hours. The reaction mixture was allowed to cool to room temperature and was passed through celite, eluting with EtOAc and concentrated under reduced pressure. The product was purified via column chromatography eluting with heptane/EtOAc 80/20 v/v to yield a colorless oil, 68.7 mg, 87%. Silica gel TLC Rf = 0.22 (20% EtOAc/heptane); GCMS
1 (EI) expected m/z: 330.2, observed [M+]: 330; HNMR (400 MHz, CDCl3): δ (ppm) 7.30 (s, 1H), 6.97 (s, 1H), 5.20 (s, 2H), 3.53 (s, 3H), 1.64 (s, 4H). 1.62 (s, 6H), 1.242 (s, 6H), 1.237 (s, 6H); 13C NMR
(100 MHz, CDCl3) δ 155.41, 147.26, 139.04, 131.71, 114.16, 111.28, 96.81, 95.78, 78.94, 65.81, 56.41, 35.07, 35.02, 34.73, 33.82, 31.93, 31.79, 31.67.
6-ethynyl-7-(methoxymethoxy)-1,1,4,4-tetramethyl-1,2, 3,4 –tetrahydronaphthalene (20). 4-(3- (methoxymethoxy) -5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-2-methylbut-3-yn-2-ol (221.44 mg, 0.67 mmol) was disolved in dry toluene (17.6 mL). NaOH (137 mg, 3.43 mmol) was added and the reaction was refluxed at 120 °C for 3.5 hour. The reaction mixture was allowed to cool to room temperature and was quenched with sat. aq. NH4Cl, the mixture was diluted with H2O and extracted twice with Et2O. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography eluting with 17% v/v EtOAc in heptane to yield 6-ethynyl-7-(methoxymethoxy)-1,1,4,4-tetramethyl-1,2,3,4- tetrahydronaphthalene as a white solid. 164 mg, 0.6 mmol, 90%. Silica gel TLC Rf = 0.30
(heptane/EtOAc 7% v/v); GC-MS (EI): calc. for: C18H24O2: 272,38, observed most abundant peaks:
1 272, 257, 227, 225; H NMR (400 MHz, CDCl3) δ 7.39 (s, 1H), 7.04 (s, 1H), 5.23 (s, 2H), 3.53 (s, 3H),
13 3.22 (s, 1H), 1.66 (s, 4H), 1.26 (s, 6H), 1.25 (s, 6H); C NMR (100 MHz, CDCl3) δ 156.12, 147.98, 138.89, 132.41, 113.35, 110.22, 95.47, 80.76, 79.90, 56.39, 35.03, 34.86, 33.83, 31.94, 31.80. methyl4-(3-hydroxy-5-(3-(methoxymethoxy)-5,5,8,8-tetra methyl-5,6,7,8-tetrahydronaphthalen-2- yl)pent-4-yn-1-yl)-3-(methoxymethoxy)benzoate (21). n-butyllithium (152 μL, 1.1 M in hexane, 0.17 mmol) was added dropwise to a stirred solution of 6-ethynyl-7-(methoxymethoxy)-1,1,4,4- tetramethyl-1,2,3,4-tetrahydronaphthalene (52.1 mg, 0.19 mmol) in THF (600 μL) at -78 °C. The mixture was stirred at -78 °C for 40 minutes. methyl 4-(methoxymethoxy)-3-(3-oxopropyl)benzoate (38,6 mg, 153 mmol) in THF (300 μL) was added dropwise and the reaction was stirred for 60 minutes at -78 °C. The reaction mixture was then allowed to warm to room temperature and was stirred for 2.5 hour. The reaction was then quenched with H2O and extracted with EtOAc three times. The combined organic layers were washed with brine and evaporated. The product was purified via column chromatography, eluting with 30 % EtOAc in heptane, to yield a pale yellow oil.
47 mg, 58%. Silica gel TLC Rf = 0.25 (heptane/EtOAc 30% v/v); LC-MS (ESI): calc. for C31H40O7
118 SPIROKETALS FOR NUCLEAR RECEPTORS
1 [M+Na]: 547.63, observed 547.50, LC, Rt=9.02 min; H NMR (200 MHz, CDCl3) δ 7.94 – 7.83 (m, 1H), 7.33 (s, 1H), 7.09 (d, J = 8.5 Hz, 1H), 7.01 (s, 1H), 5.26 (s, 2H), 5.21 (s, 2H), 4.64 (t, J = 6.4 Hz, 1H), 3.87 (s, 3H), 3.51 (s, 3H), 3.48 (s, 3H), 2.98 – 2.87 (m, 2H), 2.24 – 2.04 (m, 2H), 1.65 (s, 4H),
13 1.25 (s, 6H), 1.24 (s, 6H); C NMR (50 MHz, CDCl3) δ 167.06, 159.04, 155.64, 147.54, 138.94, 131.86, 130.51, 129.67, 123.48, 113.68, 113.17, 110.92, 95.59, 94.28, 92.77, 82.08, 62.76, 56.46, 56.37, 52.01, 37.93, 35.07, 34.81, 33.84, 31.95, 31.81, 26.28. methyl 4-(3-hydroxy-5-(3-(methoxymethoxy)-5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl) pentyl)-3-(methoxymethoxy)benzoate (22). methyl 4-(3-hydroxy-5-(3-(methoxymethoxy)-5,5,8,8- tetramethyl-5,6,7,8-tetra hydronaphthalen-2-yl)pent-4-yn-1-yl)-3-(methoxymethoxy) benzoate (23.7 mg, 0.045 mmol) was dissolved in EtOAc (0.6 mL). KHCO3 (16.8 mg, 0.17 mmol) and 10 % palladium on carbon (18 mg, 0.17 mmol) were added and the reaction was stirred at room temperature under an atmosphere of hydrogen. After 2 hours the mixture was diluted with EtOAc and passed through celite, to obtain the title compound as colorless oil. 21.7 mg, 0.041 mmol, 91%.
Silica gel TLC Rf = 0.21 (heptane/EtOAc 30% v/v); LC-MS (ESI): calc. for C31H44O7 [M+Na]: 551.67,
1 observed 551.42, LC, Rt=9.15 min; H NMR (200 MHz, CDCl3) δ 7.91 – 7.79 (m, 2H), 7.08 – 6.92 (m, 3H), 5.25 (s, 2H), 5.17 (s, 2H), 3.88 (s, 3H), 3.68 – 3.56 (m, 1H), 3.49 (s, 3H), 3.47 (s, 3H), 2.82 – 2.62 (m, 4H), 1.85 – 1.69 (m, 4H), 1.65 (s, 4H), 1.26 (s, 6H), 1.23 (s, 6H); 13C NMR (50 MHz, cdcl3) δ 167.10, 158.90, 153.22, 144.00, 138.38, 131.79, 131.26, 129.43, 128.44, 128.09, 123.50, 113.13, 111.80, 95.05, 94.28, 70.71, 56.44, 56.25, 52.02, 38.17, 37.52, 35.30, 34.44, 33.80, 32.05, 31.99, 26.52, 26.37. methyl 3-(methoxymethoxy)-4-(5-(3-(methoxymethoxy)-5,5,8,8-tetramethyl-5,6,7,8-tetrahydro- naphthalen-2-yl)-3-oxopentyl)benzoate (23). methyl 4-(3-hydroxy-5-(3-(methoxymethoxy)-5,5,8,8- tetramethyl-5,6,7,8-tetrahydro naphthalen-2-yl)pentyl)-3-(methoxymethoxy)benzoate (21.7 mg,
0.041 mmol) was dissolved in CH2Cl2 (2.4 mL). Dess–Martin periodinane (38.4 mg, 0.091 mmol) was added and the reaction was stirred for 18 hours at room temperature. The reaction was quenched with sat aq. KHCO3 (5 mL) and 10 % Na2S2O3 (5 mL) solution. The aqueous layer was extracted with CH2Cl2 three times. The combined organic layers were washed with brine, dried over
Na2SO4, filtered and evaporated to yield the title compound. 18.5 mg, 0.035 mmol, 86%. Silica gel
TLC Rf = 0.41 (heptane/EtOAc 30% v/v); LC-MS (ESI): calc. for C31H42O7 [M+Na]: 549.65, observed:
1 549.42; LC, rt 9.50 min; H NMR (200 MHz, CDCl3) δ 7.92 – 7.80 (m, 2H), 7.12 – 6.93 (m, 3H), 5.25 (s, 2H), 5.15 (s, 2H), 3.88 (s, 3H), 3.47 (s, 6H), 3.01 – 2.65 (m, 8H), 1.65 (s, 4H), 1.25 (s, 6H),
13 1.23 (s, 6H); C NMR (50 MHz, CDCl3) δ 210.04, 167.00, 158.92, 153.29, 144.35, 138.25, 131.59, 130.07, 129.77, 128.05, 127.43, 123.47, 113.11, 111.72, 94.83, 94.17, 56.44, 56.20, 52.05, 43.36, 42.63, 35.29, 34.47, 33.82, 32.05, 31.96, 25.11, 24.82.
119 CHAPTER 4 methyl 6,6,9,9-tetramethyl-3,4,6,7,8,9-hexahydrospiro [benzo[g]chromene-2,2'-chroman]-7'- carboxylate (24). To a solution of methyl 3-(methoxymethoxy)-4-(5-(3-(methoxymethoxy)-5,5,8,8- tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-3-oxopentyl)benzoate (18.5 mg, 35 μmol) in CH2Cl2 (350 μL) was added trimethylsilyl bromide (46 μL, 351 μmol) at -30 °C. The reaction was stirred for
1 hour at -30 °C and 1 hour at 0 °C. the reaction was then quenched with H2O and extracted with
EtOAc three times. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via column chromatography eluting with 17% EtOAc in heptane. to yield methyl 6,6,9,9-tetramethyl-3,4,6,7,8,9- hexahydrospiro[benzo[g]chromene-2,2'-chroman]-7'-carboxylate as white solid, 9,5 mg, 23 μmol,
64%. Silica gel TLC Rf = 0.41 (heptane/EtOAc 17% v/v); LC-MS (ESI): calc. for C27H32O4 [M+H]:
1 421.24, observed 421.25, LC, Rt=10.10 min; H NMR (400 MHz, CDCl3) δ 7.86 (d, J = 2.1 Hz, 1H), 7.76 (dd, J = 8.6, 2.3 Hz, 1H), 7.02 (s, 1H), 6.78 (d, J = 8.5 Hz, 1H), 6.64 (s, 1H), 3.87 (s, 3H), 3.39 – 3.13 (m, 2H), 2.81 – 2.66 (m, 2H), 2.28 – 2.15 (m, 2H), 2.00 – 1.90 (m, 2H), 1.63 (s, 4H), 1.26 (s,
13 3H), 1.26 (s, 3H), 1.21 (s, 3H), 1.17 (s, 3H); C NMR (126 MHz, CDCl3) δ 167.19, 156.64, 149.90, 144.67, 137.73, 131.26, 129.13, 126.94, 122.70, 122.26, 119.43, 117.45, 114.17, 97.03, 51.97, 35.36, 35.27, 34.30, 33.79, 32.30, 32.23, 32.05, 31.95, 31.45, 31.37, 21.02, 20.73.
6,6,9,9-tetramethyl-3,4,6,7,8,9-hexahydrospiro[benzo[g] chromene-2,2'-chroman]-7'-carboxylic acid (1). The carboxylate (9 mg, 21.4 μmol) was dissolved in 1,4-dioxane/MeOH (14:5 v/v, 0.5 mL). NaOH (4 M, 50 μL) was added and the reaction was stirred at 40 °C. After 4 h the solvent was evaporated and the residue was redissoved in H2O. The aqueous solution was acidified to pH 4 and extracted three times with CH2Cl2, the combined organic layers were washed with brine, dried over
Na2SO4, filtered and evaporated to obtain compound 1 in quantitative yield. LC-MS (ESI): calc. for
1 C26H30O4 [M+H]: 407.22 observed 407.25, LC, Rt=8.98 min; H NMR (400 MHz, Chloroform-d) δ 7.92 (s, 1H), 7.84 (d, J = 8.6 Hz, 1H), 7.03 (s, 1H), 6.81 (d, J = 8.6 Hz, 1H), 6.65 (s, 1H), 3.33 (ddd, J = 17.3, 12.8, 5.8 Hz, 1H), 3.22 (ddd, J = 16.1, 12.9, 5.7 Hz, 1H), 2.80 (ddd, J = 16.5, 5.8, 2.8 Hz, 1H), 2.70 (ddd, J = 16.1, 5.8, 2.5 Hz, 1H), 2.30 – 2.17 (m, 2H), 2.04 – 1.89 (m, 2H)., 1.64 (s, 4H), 1.27 (s,
13 3H), 1.26 (s, 3H), 1.22 (s, 3H), 1.17 (s, 3H); C NMR (101 MHz, CDCl3) δ 171.72, 157.43, 149.86, 144.71, 137.79, 132.02, 129.89, 126.96, 122.43, 121.66, 119.40, 117.63, 114.18, 97.14, 35.36, 35.27, 34.30, 33.80, 32.29, 32.23, 32.05, 31.95, 31.44, 31.33, 21.00, 20.71; HRMS (m/z): [M + H]+ calcd 407.2222, found 407.2224.
References
1. Perron, F. & Albizati, K. F. Chemistry of spiroketals. Chem. Rev. 89, 1617–1661 (1989).
120 SPIROKETALS FOR NUCLEAR RECEPTORS
2. Brasholz, M., Sörgel, S., Azap, C. & Reißig, H.-U. Rubromycins: Structurally Intriguing, Biologically Valuable, Synthetically Challenging Antitumour Antibiotics. Eur. J. Org. Chem. 2007, 3801–3814 (2007). 3. Sperry, J., Wilson, Z. E., Rathwell, D. C. K. & Brimble, M. A. Isolation, biological activity and synthesis of benzannulated spiroketal natural products. Nat. Prod. Rep. 27, 1117–1137 (2010). 4. Yoneda, N., Fukata, Y., Asano, K. & Matsubara, S. Asymmetric Synthesis of Spiroketals with Aminothiourea Catalysts. Angew. Chem. Int. Ed Engl. 54, 15497–15500 (2015). 5. Brockmann, H. & Renneberg, K.-H. Rubromycin, ein rotes Antibiotikum aus Actinomyceten. Naturwissenschaften 40, 59–60 (1953). 6. Brockmann, H. & Renneberg, K.-H. Collinomycin, ein gelbes Antibiotikum aus Actinomyceten. Naturwissenschaften 40, 166–167 (1953). 7. Brockmann, H., Lenk, W., Schwantje, G. & Zeeck, A. Zur kenntnis der rubromycine. Tetrahedron Lett. 7, 3525–3530 (1966). 8. Beck, J. J., Baig, N., Cook, D., Mahoney, N. E. & Marsico, T. D. Semiochemicals from ex situ abiotically stressed cactus tissue: a contributing role of fungal spores? J. Agric. Food Chem. 62, 12273–12276 (2014). 9. Raju, B. R. & Saikia, A. K. Asymmetric synthesis of naturally occurring spiroketals. Mol. Basel Switz. 13, 1942–2038 (2008). 10.Favre, S., Vogel, P. & Gerber-Lemaire, S. Recent synthetic approaches toward non-anomeric spiroketals in natural products. Mol. Basel Switz. 13, 2570–2600 (2008). 11. Rizvi, S. A., Tereshko, V., Kossiakoff, A. A. & Kozmin, S. A. Structure of Bistramide A−Actin Complex at a 1.35 Å Resolution. J. Am. Chem. Soc. 128, 3882–3883 (2006). 12. Biard, J. F. et al. Bistramides A, B, C, D, and K: a new class of bioactive cyclic polyethers from Lissoclinum bistratum. J. Nat. Prod. 57, 1336–1345 (1994). 13. Statsuk, A. V. et al. Actin is the primary cellular receptor of bistramide A. Nat. Chem. Biol. 1, 383– 388 (2005). 14. Zinzalla, G., Milroy, L.-G. & Ley, S. V. Chemical variation of natural product-like scaffolds: design and synthesis of spiroketal derivatives. Org. Biomol. Chem. 4, 1977–2002 (2006). 15. Milroy, L.-G. et al. Natural-product-like spiroketals and fused bicyclic acetals as potential therapeutic agents for B-cell chronic lymphocytic leukaemia. ChemMedChem 3, 1922–1935 (2008). 16.Capecchi, T., Koning, C. B. de & Michael, J. P. Synthesis of the bisbenzannelated spiroketal core of the γ-rubromycins. The use of a novel Nef-type reaction mediated by Pearlman’s catalyst. J. Chem. Soc. [Perkin 1] 2681–2688 (2000). doi:10.1039/B002984J 17. Waters, S. P., Fennie, M. W. & Kozlowski, M. C. Investigation of a convergent route to purpuromycin: benzofuran formation vs spiroketalization. Org. Lett. 8, 3243–3246 (2006). 18. Brimble, M. A., Flowers, C. L., Trzoss, M. & Tsang, K. Y. A facile synthesis of fused aromatic spiroacetals based on the 3,4,3′,4′-tetrahydro-2,2′-spirobis(2H-1-benzopyran) skeleton. Tetrahedron 62, 5883–5896 (2006). 19.Altucci, L., Leibowitz, M. D., Ogilvie, K. M., Lera, A. R. de & Gronemeyer, H. RAR and RXR modulation in cancer and metabolic disease. Nat. Rev. Drug Discov. 6, 793–810 (2007). 20.Cramer, P. E. et al. ApoE-Directed Therapeutics Rapidly Clear β-Amyloid and Reverse Deficits in AD Mouse Models. Science 335, 1503–1506 (2012). 21. Barnard, J. H., Collings, J. C., Whiting, A., Przyborski, S. A. & Marder, T. B. Synthetic Retinoids: Structure–Activity Relationships. Chem. – Eur. J. 15, 11430–11442 (2009). 22.Dawson, M. I. & Xia, Z. The retinoid X receptors and their ligands. Biochim. Biophys. Acta 1821, 21–56 (2012).
121 CHAPTER 4
23.Scheepstra, M. et al. A Natural-Product Switch for a Dynamic Protein Interface. Angew. Chem. Int. Ed. 53, 6443–6448 (2014). 24.Sundén, H. et al. Chiral Dihydrobenzofuran Acids Show Potent Retinoid X Receptor-Nuclear Receptor Related 1 Protein Dimer Activation. J. Med. Chem. 59, 1232–1238 (2016). 25.Lippert, W. P. et al. Silicon analogues of the RXR-selective retinoid agonist SR11237 (BMS649): chemistry and biology. ChemMedChem 4, 1143–1152 (2009). 26.Chinchilla, R. & Nájera, C. Recent advances in Sonogashira reactions. Chem. Soc. Rev. 40, 5084– 5121 (2011). 27.Nishihara, Y. et al. Synthesis of unsymmetrically disubstituted ethynes by the palladium/copper(I)-cocatalyzed sila-Sonogashira–Hagihara coupling reactions of alkynylsilanes with aryl iodides, bromides, and chlorides through a direct activation of a carbon– silicon bond. Tetrahedron 68, 4869–4881 (2012). 28.Nishihara, Y. et al. Palladium/copper-catalyzed sila-Sonogashira reactions of aryl iodides with alkynylsilanes via a direct C–Si bond activation. Tetrahedron Lett. 50, 4643–4646 (2009). 29.De Luca, L., Giacomelli, G. & Porcheddu, A. A Very Mild and Chemoselective Oxidation of Alcohols to Carbonyl Compounds. Org. Lett. 3, 3041–3043 (2001). 30.Aitken, H. R. M., Furkert, D. P., Hubert, J. G., Wood, J. M. & Brimble, M. A. Enantioselective access to benzannulated spiroketals using a chiral sulfoxide auxiliary. Org. Biomol. Chem. 11, 5147–5155 (2013). 31. Gunther, J. R., Parent, A. A. & Katzenellenbogen, J. A. Alternative inhibition of androgen receptor signaling: peptidomimetic pyrimidines as direct androgen receptor/coactivator disruptors. ACS Chem. Biol. 4, 435–440 (2009). 32.Love, J. D. et al. The structural basis for the specificity of retinoid-X receptor-selective agonists: new insights into the role of helix H12. J. Biol. Chem. 277, 11385–11391 (2002).
Modeling studies were performed by Anna K. H. Hirsch and Lech Milroy. Initial synthesis explorations were undertaken by Sebastian Andrei. Crystallographic data was collected, processed and refined by Sebastian Andrei and Seppe Leysen. Joost van Dongen helped with the chiral column chromatography.
122 Chapter 5
RORγt reveals allosteric inverse agonism for nuclear receptors
A hallmark of the Nuclear Receptor superfamily of transcription factors is their ability to bind small ligands at a highly conserved hydrophobic pocket located within the protein’s ligand binding domain (LBD). The nuclear receptor RORγt is critical for the differentiation and proliferation of Th17 cells associated with several chronic autoimmune diseases. This work reports the discovery of a novel allosteric binding site on the nuclear receptor RORγt. Co-crystallization of the ligand binding domain of RORγt with a series of small-molecule antagonists demonstrates occupancy of a previously unreported allosteric binding pocket. Binding at this non-canonical site induces an unprecedented conformational reorientation of helix 12 in the RORγt LBD, which blocks cofactor binding. The functional consequence of this allosteric ligand-mediated conformation is inhibition of function as evidenced by both biochemical and cellular studies. RORγt function is thus antagonized in a manner molecularly distinct from that of previously described orthosteric RORγt ligands. This brings forward an approach to target RORγt for the treatment of Th17-mediated autoimmune diseases. The elucidation of an unprecedented modality of pharmacological antagonism establishes a mechanism for modulation of nuclear receptors.
This work has been published: M. Scheepstra, S. Leysen, G. van Almen, J.R. Miller, J. Piesvaux, V. Kutilek, H. van Eenennaam, H. Zhang, K. Barr, S. Nagpal, S.M. Soisson, M. Kornienko, K. Wiley, N. Elsen, S. Sharma, C.C. Correll, B.W. Trotter, M. van der Stelt, A. Oubrie, C. Ottmann, G. Parthasarathy, & L. Brunsveld, Identification of an allosteric binding site for RORγt inhibition. Nature Communications, 6:8833 (2015).
123 CHAPTER 5
Nuclear receptors modulate gene transcription in response to small lipophilic ligands, regulate metabolism, the reproductive system, development, and are involved in many pathological processes.1 Approximately 13% of all approved drugs target the nuclear receptor superfamily and some of the most used drugs, for example those against auto- immune diseases, modulate nuclear receptor function.2 The ROR nuclear receptor subclass is an example of an orphan nuclear receptor for which responsiveness to endogenous natural products, such a cholesterol,3,4 as well as small synthetic ligands5 has been shown. In contrast to the multifunctional profile role of subtypes RORα and RORβ, subtype RORγt (encoded by RORc), which is critically located in the thymus, mainly
6,7 regulates the development of TH17 cells. TH17 cells are key modulators in autoimmune
8 9 diseases, including multiple sclerosis and Crohn’s disease. TH17 cells produce the cytokine IL17, which is responsible for enhanced inflammatory processes.10 Active RORγt
11,12 is a prerequisite for the differentiation of TH17 cells. Inhibition of the function of these cells by inhibiting RORγt with synthetic ligands constitutes a novel, highly sought after, strategy for the treatment of autoimmune diseases.5,10 It has been suggested that the RORγt LBD is locked in a holo-conformation, resulting in constitutive activity.13 The co-crystal structures of RORγt bound to hydroxycholesterols,14 and synthetic ligands such as T090131715 (Fig. 1a) have shown that RORγt is still responsive to ligands. Spurred by the gain in biological knowledge about RORs and the link to interleukin production, there has been an intense interest in discovering synthetic ligands for the modulation of this nuclear receptor. The large structural diversity for reported RORγt ligands and the limited structural information has given rise to questions about the mode of action of these molecules. In the nuclear receptor drug discovery field there is a strong quest to elucidate novel modes of action, for example of compounds targeting alternative sites on the LBD.16– 18 For these reasons an investigation was started to gain more structural insight into RORγt ligand binding. Given the constitutive activity of RORγt, potent inhibitory ligands with an undisclosed mode of action provide an excellent starting point. Herein, RORγt modulation via allosteric inverse agonism is reported, unprecedented for nuclear receptors.
124 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
Figure 1 | Co-crystal structure of RORγt with 1. a, Structures of T0901317,15 indazoles 1 – 3,19 and cholesterol. b, Zoomed-in view of the novel allosteric binding pocket of RORγt with indazole 1. Indazole 1 is shown as orange sticks in the electron density. The RORγt residues involved in hydrophobic interactions and hydrogen bonding are shown as white and green sticks respectively. c, Structure superposition of RORγt (blue) in complex with a classical ligand, T0901317 shown in red (PDB ID 4NB620) and crystal structure of RORγt (green) in complex with indazole 1, shown in orange. Indazole 1 (orange) sits in a new allosteric pocket in direct contact with helix 12 inducing its repositioning. d, A 2D plot showing the interactions between indazole 1 and the surrounding amino acids.
Indazole 1 (Fig. 1a) belongs to a class of compounds patented by Merck & Co as RORγt inhibitors,19 with a novel RORγt chemotype and potent activity. Co-crystallization of the RORγt with indazole 1, revealed a binding site different from the conventional ligand
125 CHAPTER 5
binding site, as for example addressed by the classical inverse agonist T0901317 (Fig. 1c). Indazole 1 binds instead to an allosteric pocket in the RORγt LBD, located more outward from the classical binding site. Crystal structure analysis shows that this allosteric pocket, absent in the classical folding motif, is newly formed by a number of hydrophobic residues on helices 4, 5, 11, and the reoriented, flexible helix 12 (Fig. 1b). Hydrogen bonding interactions exist between the molecule´s carboxylic acid group and the side chain of RORγt residue Q329 as well as the main-chain amide hydrogen atoms of residues A497 and F498. Additionally, the ortho-substituted trifluoromethyl moiety is close to the polar hydrogen of T325 and can perform additional H-F bonding (Fig. 1b, d). Apart from these distinct polar interactions, the newly generated allosteric pocket is predominantly hydrophobic, because of the amino acids side chains of residues on helices 4, 5, 11 and notably, helix 12 and the loop between helix 11 and 12. This arrangement thus generates a druggable molecular binding site critically regulated by the helix 12 positioning. Previously reported crystal structures with inverse agonists binding to the canonical site of RORγt have lacked structural data for the position of the helix 12 and the resulting cofactor binding site (e.g. Fig. 1c with T0901317). The recently reported crystal structure of RORγt with a potent tertiary amine agonist revealed the coactivator binding site on the LBD, in a classical nuclear receptor helix 12 switch agonist mode and bound to a cofactor peptide motif (Fig. 2a).21 The new crystal structure of RORγt with indazole 1 reveals that this ligand binds to a receptor surface position, normally occupied by helix 12 in a holo- or agonist liganded conformation. Whereas nuclear receptor inverse agonists or antagonists normally lead to a destabilization of the helix 12 folding, binding of indazole 1 induces a fixed conformational change in the positioning of helix 12 to interact with the ligand (Fig. 2b). For this, helix 11' 22 partially unfolds to span the distance to the displaced helix 12 N- terminus (Fig. 2a, b). The final orientation of helix 12 is as such, that the classical binding surface for the cofactor LXXLL motif is not only modified, but actively blocked. This overall orientation would effectively antagonize cofactor binding (Fig. 2c) to the RORγt with resulting decrease in RORγt activity.
126 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
Figure 2 | Allosteric mechanism of RORγt with 1. a, Agonistic conformation of RORγt published by Yang and coworkers21 (PDB entry 4NIE) showing the position of the ligand (red) in the ligand binding site, helix 12 of the LBD (yellow) and the bound cofactor (blue). b, Antagonistic conformation of RORγt, with indazole 1 (orange) making contact with helix 12 (brown) at the allosteric site in the RORγt LBD. c, Superposition of the agonistic and allosteric antagonistic RORγt conformations, with colors corresponding to those in panels 2a and 2b. The position of helix 12 has shifted to the position of the co-factor peptide, demonstrating the allosteric inverse agonistic mechanism. For a structural comparison with hydroxycholesterol liganded RORγt, see Supporting Figure 1.
To determine the functional effect, a small set of indazoles 1 – 3 was prepared and tested alongside an agonist and canonical inverse agonist in an AlphaScreen® assay. Cholesterol functioned as a weak classical agonistic, further enhancing the intrinsic activity of RORγt.14,23 Indazoles 1 – 3 inhibited coactivator binding in a dosed-dependent manner (Fig.
3a) with half-maximum inhibitory concentration (IC50) values of 7 ± 1 nM (1), 98 ± 23 nM
(2) and 280 ± 117 nM (3). The known inverse agonist T0901319 measured an IC50 value of 24 ± 13 nM in this assay, which is in good agreement with previously reported values.15,20 These data thus show that the allosteric binding mode of ligands 1 – 3 translates in potent inhibition of co-activator binding to RORγt. Furthermore, these data also suggest that the bulky ortho substituted side groups on the phenyl ketone moiety enhance RORγt binding, in line with the binding modus observed in the crystal structure. To corroborate our structural findings and gain more insight in the mode of action for these compounds, competitive binding assays for 1 (Fig. 3c) and T0901317 (Fig. 3d) were performed against fixed concentrations of cholesterol. Based on the structural data, 1 should not compete with cholesterol for the traditional ligand binding pocket. Rather, 1 and cholesterol would
address two different, independent binding pockets. As such, the IC50 value of 1 for
127 CHAPTER 5
inhibition of the RORγt coactivator interaction should be independent of cholesterol concentration. Cholesterol increases the RORγt coactivator interactions in the absence or at low concentrations of 1, in line with the agonistic properties of cholesterol, but does not affect the binding affinity of indazole 1. Indeed, the competitive assay for indazole 1 showed
no significant change in IC50 value when performed at different concentrations of cholesterol (Fig. 3c, Table 1). In contrast, the inverse agonist T0901317 competes for the same binding site as cholesterol, resulting in a competitive displacement and a cholesterol
concentration-dependent increase of the IC50 value for T0901317 (Fig. 3d, Table 1). Together, these data demonstrate that the potent allosteric inhibition of co-activator binding to RORγt by indazole 1 is independent of agonist binding.
Figure 3 | Allosteric inverse agonism. a, TR-FRET assay showing the effect of T0901317 and indazoles 1 – 3 on cofactor recruitment to the RORγt LBD in a dose dependent manner. Indazoles 1 – 3 function as inhibitors, while cholesterol promotes cofactor binding to RORγt. b. IL17a mRNA expression in EL4 cells treated with 1 – 3 (10 μM, 24 hours) or DMSO (n = 6). Error bars denote s.e.m. **, P < 0.01. c, Inhibition of cofactor binding by the allosteric inverse agonist 1 is independent of cholesterol concentration. d. Inhibition of cofactor binding by ligand T0901317 depends on cholesterol concentration.
128 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
Table 1 | IC50 values for the competition assay.
Concentration IC50: 1 (nM) IC50: T0901317
cholesterol (μM) (nM)
0 4.7 ± 1.4 11 ± 3.4 1 2.5 ± 0.9 18 ± 4.9 10 1.8 ± 1.5 195 ± 59 25 2.3 ± 1.8 548 ± 152
RORγt mediates IL17a gene expression in EL4 cells.6 The EL4 murine lymphoblast constitutively express RORγt and IL17a.24 EL4 cells were treated with 10 μM of ligands 1 – 3 for 24 hours and IL17a mRNA levels were measured by quantitative reverse transcriptase PCR (Fig. 3b). Treatment of EL4 cells with the more potent ligands 1 and 2 significantly reduced the IL17a mRNA levels, while the weaker ligand 3 was not able to reduce the mRNA levels, in line with the lower biochemical activity observed for this compound. This result thus proves the functionality of the novel mode of allosteric inverse agonism on receptor activity and control over gene transcription. In summary, the structural and functional elucidation of an unprecedented nuclear receptor allosteric inverse agonism was demonstrated. RORγt regulates a variety of physiological processes and has emerged as a highly promising drug target for autoimmune diseases. The discovery of this novel allosteric nuclear receptor pocket offers a new strategic opportunity to develop therapeutics targeting autoimmune disorders via inhibition of TH17 cell differentiation through RORγt. The elucidation of an alternative mode-of-action of pharmacological nuclear receptor modulation, independent of agonist binding, moreover provides entry into novel pharmacological concepts for this highly relevant class of drug targets.
Experimental section
Synthetic procedures
All the solvents employed were commercially available and used without purification unless stated otherwise. Water was purified using a Millipore purification train. All the reagents are commercially available and used without purification. All the NMR data were recorded on a Varian Gemini 400 MHz NMR or a Bruker Cryomagnet for NMR spectroscopy 400 MHz (400 MHz for
129 CHAPTER 5
1H-NMR and 100 MHz for 13C-NMR). Proton experiments are reported in parts per million (ppm) downfield of TMS and were relative to the residual methanol (3.31 ppm) or chloroform (7.26 ppm). All 13C spectra were reported in ppm relative to residual methanol (49.00 ppm) or chloroform (77 ppm) Analytical LC-MS was performed on a C4, Jupiter SuC4300A, 150x2.00 mm column with a gradient 5% to 100% acetonitrile in H2O in 15 min. Silica column chromatography was performed manually using silica with particle size 60 – 200 μm. Preparative HP-LC was performed on a
Gemini S4 110A 150x21.20 mm column using H2O with 0.1% Formic Acid (F.A.) and acetonitrile with 0.1% F.A. Gradient: 40% to 60% acetonitrile in 20 minutes. Purity and exact mass of the compounds were determined using a High Resolution LC-MS system consisting of a Waters ACQUITY UPLC I-Class system coupled to a Xevo G2 Quadrupole Time of Flight (Q-tof). The system was comprised of a Binary Solvent Manager and a Sample Manager with Fixed-Loop (SM- FL). compounds were separated (0.3 mL min-1) by the column (Polaris C18A reverse phase column 2.0 x 100 mm, Agilent) using a 15% to 75% acetonitrile gradient in water supplemented with 0.1% v/v formic acid before analysis in positive mode in the mass spectrometer. Synthesis was performed following and addapting procedures by Merck & Co.19 Compound T0901317 was commercially obtained from Sigma Aldrich and used without further purification.
(2-chloro-6-(trifluoromethyl)phenyl)(3-iodo-1H-indazol-1-yl)methanone Thionyl chloride (1 mL, 13.8 mmol) was added dropwise to 2-chloro-6-(trifluoromethyl)benzoic acid (100 mg, 0.45 mmol) in an oven dried flask. The reaction mixture was stirred at 75 °C overnight. After removal of the thionyl chloride under reduced pressure, the benzoyl chloride was dissolved in dry CH2Cl2 (1 mL). To this solution 3-iodo-1H-indazole (108.5 mg, 0.45 mmol) and DMAP (54.4 mg, 0.46 mmol) were added. Finally, TEA (130 μl, 0.9 mmol) was added dropwise and the mixture was stirred at room temperature.
After 24 h the reaction mixture was diluted with H2O (8 mL) and CH2Cl2 (8 mL). The layers were separated and the aqueous layer was washed with CH2Cl2 (2 x 8 mL). The combined organic layers were washed with brine, dried over Na2SO4, filtered and the solvent evaporated. The crude material was purified via silica column chromatography eluting with pentane/EtOAc 5% to 8% (v/v) to yield 183.3 mg (2-chloro-6-(trifluoromethyl)phenyl)(3-iodo-1H-indazol-1-yl)methanone as white solid, 90.4%. LC-MS (ESI): calc. for C15H7ClF3IN2O [M+H]: 451.6 observed 451.1, LC, Rt=7.17
130 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
1 min. H-NMR (400 MHz, CDCl3): δ (ppm) 8.53 (d, J = 8.4 Hz, 1H), 7.74-7.67 (m, 3H), 7.61-7.49
13 (m, 3H); C-NMR (100 MHz, CDCl3): δ 164.02, 138.95, 133.12, 132.88, 131.86, 131.10, 131.03, 126.09, 124.94, 124.89, 124.55, 124.35, 122.20, 115.61, 105.63. methyl 4-(1-(2-chloro-6-(trifluoromethyl)benzoyl)-1H-indazol-3-yl)benzoate (2-chloro-6-(trifluoro- methyl)phenyl)(3-iodo-1H-indazol-1-yl)methanone (115.7 mg, 0.26 mmol), (4-(methoxy- carbonyl)phenyl)boronic acid (71.9 mg, 0.4 mmol), Pd(PPh3)4 (19.4 mg, 26.5 μmol) and CH3CO2K
(80.5 mg, 0.82 mmol) were dissolved in dioxane/H2O (5:1 v/v, 5 mL) in a schlenk tube under argon. The reaction was stirred at 90 ºC. After 3 hours the mixture was allowed to cool to room temperature and diluted with CH2Cl2 (25 mL) and H2O (25 mL). The organic layer was washed with brine, dried over Na2SO4, filtered and concentrated in vacuo. The obtained crude material was purified via silica column chromatography (eluent: pentane/EtOAc, 4% to 10% v/v) to yield compound methyl 4-(1-(2-chloro-6-(trifluoromethyl)benzoyl)-1H-indazol-3-yl)benzoate as a pale yellow solid, 109 mg, 92%. LC-MS (ESI): calc. for C23H14ClF3N2O3 [M+H]: 459.8 observed 459.3
1 LC, Rt=7.53 min. H-NMR (400 MHz, CDCl3): δ (ppm) 8.69 (d, J = 8.4 Hz, 1H), 8.13 (d, J = 8.4 Hz, 2H), 8.02 (d, J = 8.0 Hz, 1H), 7.90 (d, J = 8.4 Hz, 2H), 7.74-7.69 (m, 3H), 7.62-7.52 (m, 2H), 3.95
13 (s, 3H); C-NMR (100 MHz, CDCl3): δ 166.71, 164.91, 150.54, 140.61, 135.84, 133.05, 132.86, 131.17, 130.87, 130.19, 130.14, 128.31, 126.12, 125.13, 124.92, 124.88, 121.44, 116.14, 52.46.
4-(1-(2-chloro-6-(trifluoromethyl)benzoyl)-1H-indazol-3-yl)benzoic acid (1) LiOH · H2O (31.2 mg, 0.74 mmol) was added to a solution of methyl 4-(1-(2-chloro-6-(trifluoromethyl)benzoyl)-1H- indazol-3-yl)benzoate (107.1 mg, 0.23 mmol) in THF/H2O (0.9 mL). The reaction was stirred at room temperature. After 24 h the reaction was diluted with H2O, neutralized with acetic acid (~pH
4) and extracted with CH2Cl2. The combined organic layers were washed with brine, dried over
Na2SO4, filtered and concentrated in vacuo. The product was purified via preparative LCMS. LC- 1 MS (ESI): calc. for C22H12ClF3N2O3 [M+H]: 445.8 observed 445.3 LC, Rt=6.83 min. H-NMR (400
MHz, CD3OD): δ (ppm) 8.63 (d, J = 8.4 Hz, 1H), 8.20 (d, J = 8.0 Hz, 1H), 8.16 (d, J = 8.4 Hz, 2H),
13 7.97 (d, J = 8.4 Hz, 2H), 7.89-7.75 (m, 4H), 7.64 (t, J = 7.6 1H); C-NMR (100 MHz, CD3OD): δ 169.10, 162.83, 151.81, 141.68, 136.61, 134.43, 133.63, 133.18, 132.69, 131.39, 131.33, 129.11, 127.45,
19 + 126.08, 122.91, 116.45. F-NMR (376 MHz, CD3OD): δ -61.15. HRMS (m/z): [M + H] calcd for
C22H12ClF3N2O3, 445.0566, found 445.0577. MS/MS (m/z): most abundant peaks: 364.0823, 209.9834, 208.9799, 206.9828 and 178.9877.
131 CHAPTER 5
(3-iodo-1H-indazol-1-yl)(2-(trifluoromethyl)phenyl)methanone Thionyl chloride (1 mL, 13.8 mmol) was added dropwise to 2-(trifluoromethyl)benzoic acid (85.5 mg, 0.46 mmol) in an oven dried flask. The reaction mixture was stirred overnight at 75 °C. After removal of the thionyl chloride under reduced pressure, the benzoyl chloride was dissolved in dry CH2Cl2 (1 mL). To this solution 3-iodo- 1H-indazole (113 mg, 0.46 mmol), DMAP (56 mg, 0.46 mmol) were added. Finally, TEA (130 μl, 0.9 mmol) was added dropwise and the mixture was stirred at room temperature.
After 24 h the reaction mixture was separated between H2O (10 mL) and CH2Cl2 (10 mL). The aqueous layer was extracted with CH2Cl2 (2 x 10 mL). The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated in vacuo. The crude material was purified via silica column chromatography eluting with pentane/EtOAc 5% (v/v) to yield 146 mg (3-iodo- 1H-indazol-1-yl)(2-(trifluoromethyl)phenyl)methanone as white solid, 90%. GC-MS (EI) expected +• 1 for C15H8F3IN2O: 415.96, observed [M ]: 416 with fragments: 347, 173, 145; H-NMR (400 MHz,
CDCl3): δ (ppm) 8.53 (d, J = 8.4 Hz, 1H), 7.81-7.79 (m, 1H), 7.73-7.66 (m, 3H), 7.60-7.84 (m, 3H);
13 C-NMR (100 MHz, CDCl3): δ 166.86, 139.44, 132.88, 131.53, 131.02, 130.98, 130.59, 129.16, 126.75, 126.71, 125.92, 124.98, 122.14, 115.75, 105.07. tert-butyl 4-(1-benzoyl-1H-indazol-3-yl)benzoate In an oven dried schlenk tube were weighed: (4- (tert-butoxycarbonyl)phenyl)boronic acid (117 mg, 0.53 mmol), KOAc (106 mg, 1.08 mmol) and
Pd(dppf)Cl2 (28.4 mg, 39 μmol). Finally (3-iodo-1H-indazol-1-yl)(2-
(trifluoromethyl)phenyl)methanone (144 mg, 0.35 mmol) in dioxane/H2O (5:1 v/v, 5.2 mL) was added and the reaction was stirred for 3.5 h at 90 °C. The reaction mixture was allowed to cool to room temperature and was then separated between CH2Cl2 and H2O. The aqueous layer was extracted with CH2Cl2 twice. The combined organic layers were washed with brine, dried over
Na2SO4, filtered and evaporated. The product was purified via silica column chromatography eluting with 45% pentane in CH2Cl2 (Rf = 0.35) to yield tert-butyl 4-(1-benzoyl-1H-indazol-3- yl)benzoate as an off-white solid. Yield: 80%, 129 mg, 0.28 mmol. LC-MS (ESI): calc. for
1 C26H21F3N2O3 [M+H]: 467.16 observed 467.17, LC, Rt=9.67 min. H-NMR (200 MHz, CDCl3): δ (ppm) 8.67 (d, J = 8.4 Hz, 1H), 8.12-7.99 (m, 3H), 7.94-7.63 (m, 7H), 7.61-7.43 (m, 1H), 1.60 (s,
13 9H); C-NMR (50 MHz, CDCl3): δ 167.92, 165.36, 150.09, 141.06, 135.39, 133.56, 132.95, 131.54,
132 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
130.39, 130.03, 129.03, 128.05, 126.68, 126.59, 125.90, 125.01, 121.42, 116.24, 81.52, 28.32; 19F-
NMR (188 MHz, CDCl3) δ -59.29.
4-(1-(2-(trifluoromethyl)benzoyl)-1H-indazol-3-yl)benzoic acid (2) tert-butyl 4-(1-(2-(trifluoro- methyl)benzoyl)-1H-indazol-3-yl)benzoate (128 mg, 0.27 mmol) was disolved in a mixture of
CH2Cl2/TFA/H2O (6.5/3/0.5, 3 mL). The reaction was stirred at room temperature. After 2.5 h the solvent was evaporated and the product was purified via preparative LCMS. LC-MS (ESI): calc. for 1 C22H13F3N2O3 [M+H]: 411.10 observed 411.17, LC, Rt=7.58 min. H-NMR (400 MHz, CDCl3 with
10% CD3OD) δ 8.65 (d, J = 8.4 Hz, 1H), 8.16 (d, J = 8.1 Hz, 2H), 8.07 (d, J = 8.1 Hz, 1H), 7.94 (d, J 13 = 8.0 Hz, 2H), 7.86 – 7.82 (m, 1H), 7.77 – 7.64 (m, 4H), 7.56 (m, 1H); C-NMR (100 MHz, CDCl3 with 10% CD3OD): δ 168.19, 167.95, 150.05, 140.80, 133.17, 133.15, 130.28, 130.17, 129.87, 128.72, 19 127.93, 127.66, 126.39, 126.35, 125.80, 124.89, 124.77, 122.17, 121.28; F-NMR (188 MHz, CDCl3
+ with 10% CD3OD) δ -55.63; HRMS (m/z): [M + H] calcd for C22H13F3N2O3, 411.0956, found 411.0957. MS/MS (m/z): most abundant peaks: 391.0890, 174.0244, 173.0212 and 145.0264.
(3-iodo-1H-indazol-1-yl)(phenyl)methanone 3-iodo-1H-indazole (100 mg, 0.42 mmol) and DMAP
(53 mg, 0.44 mmol) were dissolved in CH2Cl2 (1.5 mL). Benzoyl chloride (48 μL, 0.42 mmol) and
NEt3 (114 μL, 0.82 mmol) were added and the reaction was stirred for 24 h at room temperature.
The reaction mixture was then separated between CH2Cl2 and H2O. The aqueous layer was extracted with CH2Cl2 twice. The combined organic layers were washed with brine, dried over
Na2SO4, filtered and evaporated. The product was purified via silica column chromatography to yield (3-iodo-1H-indazol-1-yl)(phenyl)methanone as a pale yellow solid. 130 mg, 0.37 mmol, 89%. +• 1 GC-MS (EI) expected for C14H9IN2O: 347.98, observed [M ]: 348 with fragments: 105, 77; H-NMR
(400 MHz, CDCl3): δ (ppm) 8.53 (d, J = 8.4 Hz, 1H), 8.11 - 8.09 (m, 2H), 7.69 (ddd, J = 8.4, 7.0, 1.3 13 Hz, 1H), 7.64 - 7.59 (m, 1H), 7.57 - 7.46 (m, 4H); C-NMR (100 MHz, CDCl3): 167.43, 140.24, 132.73, 132.69, 131.45, 130.72, 130.51, 128.22, 125.47, 121.95, 116.03, 104.01. tert-butyl 4-(1-benzoyl-1H-indazol-3-yl)benzoate In an oven dried schlenk tube were weighed: (4- (tert-butoxycarbonyl)phenyl)boronic acid (123 mg, 0.55 mmol), KOAc (107 mg, 1.09 mmol) and
Pd(dppf)Cl2 (26 mg, 36 μmol). Finally (3-iodo-1H-indazol-1-yl)(phenyl)methanone (126 mg, 0.36 mmol) in dioxane/H2O (5:1 v/v, 5.3 mL) was added and the reaction was stirred for 3 h at 90 °C. The reaction mixture was allowed to cool to room temperature and was then separated between
133 CHAPTER 5
CH2Cl2 and H2O. The aqueous layer was extracted with CH2Cl2 twice. The combined organic layers were washed with brine, dried over Na2SO4, filtered and evaporated. The product was purified via silica column chromatography eluting with 40% v/v in CH2Cl2 (Rf = 0.25) to yield tert-butyl 4-(1- benzoyl-1H-indazol-3-yl)benzoate as a pale yellow paste. LC-MS (ESI): calc. for C25H22N2O3 [M+H]:
1 399.46 observed 399.08, LC, Rt=9.85 min. H-NMR (400 MHz, CDCl3): δ (ppm) 8.67 (d, J = 8.4 Hz, 1H), 8.22 – 8.11 (m, 4H), 8.07 – 8.00 (m, 3H), 7.70 – 7.59 (m, 2H), 7.57 – 7.46 (m, 3H), 1.63
13 (s, 9H); C-NMR (50 MHz, CDCl3) δ 168.36, 165.41, 149.38, 141.88, 135.80, 133.29, 132.84, 132.49, 131.51, 130.11, 129.70, 128.09, 128.03, 127.17, 125.46, 124.48, 121.20, 116.52, 81.52, 28.34.
4-(1-benzoyl-1H-indazol-3-yl)benzoic acid (3) tert-butyl 4-(1-benzoyl-1H-indazol-3-yl)benzoate (102 mg, 0.26 mmol) was was disolved in a mixture of CH2Cl2/TFA/H2O (6.5/3/0.5, 2.5 mL). The reaction was stirred at room temperature. After 2.5 h the solvent was evaporated and the product was purified via preparative LCMS. LC-MS (ESI): calc. for C21H14N2O3 [M+H]: 343.36 observed
1 343.08, LC, Rt=7.53 min. H-NMR (400 MHz, CDCl3 with 10% CD3OD) δ (ppm) 8.65 (d, J = 8.5 Hz, 1H), 8.26 – 8.14 (m, 4H), 8.13 – 8.05 (m, 3H), 7.74 – 7.63 (m, 2H), 7.60 – 7.53 (m, 3H). 13C-
NMR (100 MHz, CDCl3 with 10% CD3OD) δ 168.55, 168.47, 149.44, 141.66, 135.88, 133.01, 132.39, 131.18, 130.31, 129.62, 127.96, 127.93, 125.41, 124.32, 121.11, 116.19. HRMS (m/z): [M + H]+ calcd for C21H16N2O3, 343.1083, found 343.1078. MS/MS (m/z): most abundant peaks: 399.0266, 325.0970, 105.0337 and 77.0389.
Protein preparation for the HTRF assay. The human RORγt LBD used for the HTRF assay was expressed as a His6-tag fusion protein from the pET15b expression vector in Escherichia coli BL21(DE3) cells. Cells transformed with this vector were grown in 2x YT medium supplemented with ampicillin until an OD600 = 0.7 was reached. Protein expression was then induced with 0.1 mM isopropyl-b-d-thiogalactoside (IPTG). After incubation for 16 h at 16°C, cell cultures were harvested by centrifugation. The cells were lysed via sonication and the protein was purified via Ni2+-affinity column chromatography.
HTRF assay. The homogeneous time resolved FRET assays were performed in triplicate with 20 nM His6-RORγt and 100 nM biotin labeled co-factor peptide. The biotin labeled peptide was made via Fmoc solid phase peptide synthesis on a MultiPep Rsi system (Intavis). Terbium labeled anti- His antibody and D2 labeled streptavidin were used at recommended concentrations by the supplier Cisbio Bioassays. Assay buffer contained: 100 mM HEPES (pH 7.5), 100 mM NaCl, 5 mM DTT, 0.1% BSA. The plates (white 384-well plates Greiner Bio-One) were incubated for 2 h at 4 °C before the FRET ratio was measured in a Tecan Infinite® F500 plate reader. The data was analyzed using Origin software.
134 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
Protein expression, purification, crystallization, data collection and structure determination. A pCDFDuet-1 expression vector encoding the RORγt ligand binding domain (LBD, residues 265- 507) with an N-terminal StrepII-SUMO-tag was transformed by heatshock into Nico21(DE3) E. coli cells (NEB). Single colonies were used to inoculate three pre-cultures of 30 ml LB containing 100 μg/ml streptomycin. After overnight incubation at 37 ºC, each pre-culture was transferred to 1.5 L of ZYP5052 medium25 without lactose (making ZYP505), but with 100 μg/ml streptomycin and 0.1 % antifoam SE-15 (Sigma-Aldrich). These cultures were incubated further until they reached an
OD600nm=4. At that point, the temperature was decreased to 18 ºC, protein expression was induced by adding IPTG to a final concentration of 200 μM and the cultures were grown for 24 hours more. The cells were harvested by centrifugation and were dissolved in lysis buffer (20 mM Tris pH 7.5,
300 mM NaCl, 5 mM MgCl2, 2 mM 2-mercapto-ethanol and 10 μg/ml DNase I (Sigma-Aldrich)) at 10 g/100 ml. After cell lysis using an Emulsiflex-C3 homogeniser (Avestin), the cell lysate was cleared by centrifugation (at 4 ºC) and the supernatant was loaded on a column consisting of 10 ml Strep-Tactin Superflow High Capacity resin (IBA). The fusion protein was eluted from the resin with 3 column volumes elution buffer (20 mM Tris, 300 mM NaCl, 2 mM 2-mercapto-ethanol and 2.5 mM Desthiobiotin at pH 7.5) and the StrepII-SUMO-tag was removed by adding dtUD1 SUMO protease.26 Next, the protein mixture was concentrated using an Amicon ultra-centrifugation device with a 3 kDa cutoff (Millipore) and loaded on a Superdex 75 pg 16/60 size-exclusion column (GE Life Sciences). Here, the cleaved StrepII-SUMO-tag and RORγt LBD co-eluted. To separate them, the SEC fractions were combined and loaded for a second time over the regenerated Strep-Tactin column. Finally, pure RORγt was collected in the flow-through and concentrated in presence of ligand to 8 mg/ml.
Crystallization, data collection and structure determination. Before crystallization, RORγt was mixed with an additional 1 equivalent of ligand. The crystals were grown at room temperature using the sitting drop vapor diffusion method. Optimal crystals were grown in a week by mixing 1 μL protein solution with the same volume of a crystallization condition containing 0.2 M MgCl2, 0.1 M TRIS, pH 8.5 and 7 % w/v PEG 6K. The well reservoir was filled with 1 ml of the crystallization condition. Crystals were cryoprotected in the mother liquor supplemented with 20 % glycerol and flash-cooled in liquid nitrogen.
Diffraction data were collected at 100 K at the PX II beamline at the SLS synchrotron (Villigen, Switzerland). iMOSFLM and AIMLESS of the CCP4 suite were used for integration and scaling respectively. The structure was phased by molecular replacement using PDB ID 4NIE21 as search model in Phaser.27 Ligand restraints were generated using Grade (Global Phasing, version 1.2.8) and CCDC Mogul.28 Coot and phenix.refine were used in alternating cycles of model building and refinement. The quality of the final model was evaluated using MolProbity.29 All data collection,
135 CHAPTER 5 refinement and validation statistics are shown in Table S1. The crystal structure was submitted to the PDB with identifier 4YPQ. Figures were created using PyMOL (The PyMOL Molecular Graphics System, Version 1.7. Schrödinger, LLC) and LigPlot+.
Table s1 | Crystallographic statistics for PDB ID 4YPQ
Data collection Wavelength (Å) 1.50 Å Resolution range (Å)a 35.47-2.32 (2.40-2.32) Space group R32:H Cell parameters a=b=173.81 Å, c= 67.22 Å Solvent content (%) 48.10 a,b CC1/2 0.999 (0.680) a,c Rsym 0.121 (0.886) a,d Rmeas 0.132 (0.972) 12.1 (2.7) Completeness (%)a 100.0 (100.0) No. of unique reflections 16884 (1672) Redundancya 11.7 (11.6) Wilson B-factor (Å2) 42.08
Refinement Resolution range used 35.47-2.32 Total no. of reflections used 16882 No. of reflections in the 'free' set 841 Number of non-hydrogen protein / solvent atoms 3874 / 560 Rwork 0.175
Rfree 0.229 R.m.s. deviations from ideal valuesf bond lengths (Å) / bond angles (°) 0.008 / 1.06 Average protein/solvent B-factor (Å2) 46.3 / 48.10 Ramachandran plot: favoured / outlier residuesf (%) 97.9 / 0 Molprobity score / percentilef 1.05 / 100 Clash score / percentilef 2.49 / 99 Good rotamers (%)f 99.54
a Number in parentheses is for the highest resolution shell b CC1/2 = Pearson's intra-dataset correlation coefficient as described in (Karplus & Diederichs, 2012) c th Rsym = ∑h∑l│Ihl-
136 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
Figure S1 | Allosteric mechanism of RORγt with indazole 1. Agonistic conformation of RORγt published by martynowski and coworkers (PDB entry 3KYT) showing the position of the 20α hydroxycholesterol (red ) in the ligand binding site, helix 12 of the LBD (yellow) and the bound cofactor (blue). b, Antagonistic conformation of RORγt, with indazole 1 (orange) making contact with helix 12 (brown) at the allosteric site in the RORγt LBD. c, Superposition of the agonistic and allosteric antagonistic RORγt conformations, with colors corresponding to those in panels 2a and 2b. The position of helix 12 has shifted to the position of the co-factor peptide, demonstrating the allosteric inverse agonistic mechanism.
Quantitative RT–PCR. EL4 cells were grown in Dulbecco's Modified Eagle Medium (Gibco®). 24 hours after the cells were seeded onto a 12-well plate, the cells were treated with ligands 1 – 3 (from 10 mM stock in DMSO) or DMSO. After 24 hours the cells were harvested and RNA was isolated using a RNeasy Plus Micro Kit (Qiagen) and reverse transcribed using the iScrip cDNA biosynthesis kit (Bio-Rad). Quantitative reverse transcriptase PCR was performed to analyze mRNA levels of mouse IL17a levels using SYBR green technology (Bio-Rad) on a CFX™ Real-Time System (Bio-Rad). Primer sequences used for IL17a30: Fw: ctccagaaggccctcagactac, Rev: ctgtgtcaatgcggagggaaagct and Gapdh: Fw: ggtggacctcatggcctaca, Rev: ctctcttgctcagtgtccttgct. The level of IL17a mRNA expression was normalized to that of Gapdh expression. All data are expressed as the mean ± s.e.m. (n = 6). Statistical analysis was performed using an one way anova comparing against the DMSO control following Dunnett post hoc test.
References
1. Gronemeyer, H., Gustafsson, J.-A. & Laudet, V. Principles for modulation of the nuclear receptor superfamily. Nat. Rev. Drug Discov. 3, 950–964 (2004). 2. Overington, J. P., Al-Lazikani, B. & Hopkins, A. L. How many drug targets are there? Nat. Rev. Drug Discov. 5, 993–996 (2006). 3. Xu, T. et al. Ursolic acid suppresses interleukin-17 (IL-17) production by selectively antagonizing the function of RORgamma t protein. J. Biol. Chem. 286, 22707–22710 (2011).
137 CHAPTER 5
4. Fujita-Sato, S. et al. Structural basis of digoxin that antagonizes RORgamma t receptor activity and suppresses Th17 cell differentiation and interleukin (IL)-17 production. J. Biol. Chem. 286, 31409–31417 (2011). 5. Fauber, B. P. & Magnuson, S. Modulators of the Nuclear Receptor Retinoic Acid Receptor- Related Orphan Receptor-γ (RORγ or RORc). J. Med. Chem. 57, 5871–5892 (2014). 6. Solt, L. A. et al. Suppression of TH17 differentiation and autoimmunity by a synthetic ROR ligand. Nature 472, 491–494 (2011). 7. Solt, L. A. & Burris, T. P. Action of RORs and their ligands in (patho)physiology. Trends Endocrinol. Metab. TEM 23, 619–627 (2012). 8. Yang, Y. et al. Impact of suppressing retinoic acid-related orphan receptor gamma t (ROR)γt in ameliorating central nervous system autoimmunity. Clin. Exp. Immunol. 179, 108–118 (2015). 9. Kanai, T., Mikami, Y., Sujino, T., Hisamatsu, T. & Hibi, T. RORγt-dependent IL-17A-producing cells in the pathogenesis of intestinal inflammation. Mucosal Immunol. 5, 240–247 (2012). 10.Huang, Z., Xie, H., Wang, R. & Sun, Z. Retinoid-related orphan receptor γt is a potential therapeutic target for controlling inflammatory autoimmunity. Expert Opin. Ther. Targets 11, 737–743 (2007). 11. He, Y. W. et al. Down-regulation of the orphan nuclear receptor ROR gamma t is essential for T lymphocyte maturation. J. Immunol. Baltim. Md 1950 164, 5668–5674 (2000). 12. Yang, X. O. et al. T Helper 17 Lineage Differentiation Is Programmed by Orphan Nuclear Receptors RORα and RORγ. Immunity 28, 29–39 (2008). 13. Harris, J. M., Lau, P., Chen, S. L. & Muscat, G. E. O. Characterization of the retinoid orphan- related receptor-alpha coactivator binding interface: a structural basis for ligand-independent transcription. Mol. Endocrinol. Baltim. Md 16, 998–1012 (2002). 14. Jin, L. et al. Structural Basis for Hydroxycholesterols as Natural Ligands of Orphan Nuclear Receptor ROR? Mol. Endocrinol. 24, 923–929 (2010). 15. Kumar, N. et al. The Benzenesulfoamide T0901317 [N-(2,2,2-Trifluoroethyl)-N-[4-[2,2,2- trifluoro-1-hydroxy-1-(trifluoromethyl)ethyl]phenyl]-benzenesulfonamide] Is a Novel Retinoic Acid Receptor-Related Orphan Receptor-α/γ Inverse Agonist. Mol. Pharmacol. 77, 228–236 (2010). 16.Moore, T. W., Mayne, C. G. & Katzenellenbogen, J. A. Minireview: Not picking pockets: nuclear receptor alternate-site modulators (NRAMs). Mol. Endocrinol. Baltim. Md 24, 683–695 (2010). 17. Hughes, T. S. et al. An alternate binding site for PPARγ ligands. Nat. Commun. 5, (2014). 18. Scheepstra, M. et al. A natural-product switch for a dynamic protein interface. Angew. Chem. Int. Ed Engl. 53, 6443–6448 (2014). 19.Karstens, W. F. J. et al. Rorgammat Inhibitors. (2012). 20.Fauber, B. P. et al. Structure-based design of substituted hexafluoroisopropanol- arylsulfonamides as modulators of RORc. Bioorg. Med. Chem. Lett. 23, 6604–6609 (2013). 21. Yang, T. et al. Discovery of Tertiary Amine and Indole Derivatives as Potent RORγt Inverse Agonists. ACS Med. Chem. Lett. 5, 65–68 (2014). 22.Stehlin, C. et al. X-ray structure of the orphan nuclear receptor RORbeta ligand-binding domain in the active conformation. EMBO J. 20, 5822–5831 (2001). 23.Wang, Y., Kumar, N., Crumbley, C., Griffin, P. R. & Burris, T. P. A second class of nuclear receptors for oxysterols: Regulation of RORalpha and RORgamma activity by 24S- hydroxycholesterol (cerebrosterol). Biochim. Biophys. Acta 1801, 917–923 (2010). 24.Ichiyama, K. et al. Foxp3 Inhibits RORγt-mediated IL-17A mRNA Transcription through Direct Interaction with RORγt. J. Biol. Chem. 283, 17003–17008 (2008).
138 RORγT REVEALS ALLOSTERIC INVERSE AGONISM
25.Studier, F. W. Protein production by auto-induction in high density shaking cultures. Protein Expr. Purif. 41, 207–234 (2005). 26.Weeks, S. D., Drinker, M. & Loll, P. J. Ligation Independent Cloning Vectors for Expression of SUMO Fusions. Protein Expr. Purif. 53, 40–50 (2007). 27.McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007). 28.Bruno, I. J. et al. Retrieval of crystallographically-derived molecular geometry information. J. Chem. Inf. Comput. Sci. 44, 2133–2144 (2004). 29.Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12–21 (2010). 30.Ivanov, I. I. et al. The Orphan Nuclear Receptor RORγt Directs the Differentiation Program of Proinflammatory IL-17+ T Helper Cells. Cell 126, 1121–1133 (2006).
Geert van Almen assisted with the cellular experiments. Seppe leysen processed and refined the crystal data.
139
140
Chapter 6
Epilogue
For the last few decades nuclear receptor modulation via small molecules has been investigated and a relative good understanding of ligand design targeting the classical binding pocket and the underlying transcriptional mechanisms is established. The work in this thesis focusses on alternatives of RXR and RORγt modulation and shows the potential of nuclear receptor modulation via the orthosteric pocket, the coregulator pocket or an allosteric site. The reported strategies combine biochemical evaluation with cellular and extensive structural data. Looking forward however, there are challenges and questions that remain. RXR plays many biological roles in the human body via the dimerization with a large number of other nuclear receptors. The recurring theme for RXR research is therefore to explore the therapeutical potential of RXR in the selective dimerization of the receptor with these other nuclear receptors. The identification of the previously unreported allosteric pocket for RORγt inhibition offers opportunities for selective modulation of RORγt. Modeling has identified several different molecular scaffolds targeting the allosteric pocket. Furthermore, current efforts focus on the development of screening methods to identify new ligands for this pocket. Lastly, sequence and structural homology showed that the principle of an allosteric pocket might be translated to other members of the nuclear receptor family. This chapter looks forward and describes the opportunities to expand on the work described in this thesis.
141 CHAPTER 6
Introduction
The work described in this thesis showed different approaches of nuclear receptor modulation with small molecules. For example, the work described in chapters two and three showed that only minor modifications to the ligands can have different functional outcomes and selective amino acid reorientation leading to displacements of secondary structural elements in the ligand binding domain. Small molecules can thus be powerful tools to modulate the activity of the receptor in a medicinal chemistry setting, but they can also serve as tools to answer more fundamental questions.1,2 Selective nuclear receptor dimerization, coregulatory recruitment, or partial agonism are challenges in nuclear receptor research which in principle could be addressed by ligand binding. The described allosteric pocket in chapter five demonstrates inhibition of RORγt independent of agonist binding. This provides an entry into novel pharmacological concepts for this nuclear receptor, but more importantly might be translated to other members of the nuclear receptor family. Opportunities for the development of novel allosteric modulators and assay development for the allosteric pocket of this highly relevant drug target3–5 are currently under investigation.
RXR selective dimerization
A recurring theme for RXR research is the selective dimerization of the receptor with other nuclear receptors. RXR forms heterodimers with a large number of nuclear receptor partners6 and although there are some successes targeting RXR directly,1,7 the real potential of RXR drug design might be in inducing selective dimerization. The X-ray crystal structures of two RXR full length heterodimers (RXR-PPAR and RXR-LXR) have been solved8,9 providing some insights in nuclear receptor signaling and allosteric mechanisms. These results have identified dimerization interfaces and key residues of RXR, playing a pivotal role in dimerization.10 Helix 5, together with the C-terminus of helix 11, plays a central role in the allosteric control between the dimer interface, helix 12, the ligand binding pocket and the AF2 surface. This then can provide guidelines in our efforts to make ligands inducing selective dimerization. However, the picture is far from finished, for example structural positioning of each domain within the full length receptors, or organization of the receptors in cells remain elusive.11,12 To investigate selective dimerization, typically bioluminescence resonance energy transfer-based (BRET) assays are used,13,14 or direct functional assays.15 One might also think of developing a homogeneous time resolve FRET assay, where the dimerization partner of RXR can be exchanged for all the relevant partners.
142 EPILOGUE
Future work regarding the RXR modulators reported in chapter 2 – 4 in this thesis might focus on the biophysical evaluation of the dimerization inducing properties of the molecules. A pharmacological relevant dimerization partner for RXR is Nurr1, for example in targeting neurodegenerative diseases.15–17 Although this heterodimer has gained a lot of attention, structural information and molecular details of this heterodimer for the selective dimerization are lacking. The X-ray structure of this heterodimer would be a big step forward in the design of selective modulators for this dimer pair, as it would identify the key residues involved in the dimer formation and allow for rational design of modulators of this protein-protein interaction. Other, pharmacological relevant, nuclear receptor dimer partners are PPARγ, for its roles in metabolic disorders, and RAR, for certain types of cancer.1
Figure 6.1 | Ligand design for RXR. Rational for the RXR ligand design utilizing the biaryl scaffold to selectively target the coregulator recruitment of the dimerization interface.
The results described in chapter three of this thesis provide a rational for the design of RXR ligands probing the coregulator interactions or the dimerization interface selectively. Using the biaryl scaffold and keeping the hydrogen bonding network with Asn306 and Arg316 in place – which was identified to be key – modifications in the hydrophobic region are allowed without a loss in binding affinity. The biaryl scaffold could then be utilized to probe different regions of the RXR ligand binding pocket selectively (figure 6.1). In principle, the developed chemistry (chapter 2 and 3) allows for relative straightforward modifications to the ligands expanding the chemical diversity. Some examples of ligands probing the dimerization interface have been shown for RXR, but given the complexity and the number of dimerization partners for RXR, diversification of the compound library
143 CHAPTER 6 together with crystallization could shed more light on selective dimerization properties or selective coregulatory recruitment.
Future directions for RORγt
The discovery of the previously unreported allosteric binding site for RORγt inhibition has opened up possibilities for selective modulation of RORγt independent of endogenous ligands. RORγt is located in the thymus and mainly regulates the development of TH17 cells which are key regulators in autoimmune diseases such as multiple sclerosis and Crohn’s disease through the cytokine interleukin-17.18–20 For this reason the receptor has gained a lot of interest; especially ligands that inhibit the receptor are highly sought after. In silico screening has identified several molecular scaffolds such as the isoquinolines (figure 6.2), or isoxazoles capable of binding in the allosteric pocket. To investigate binding properties and to validate the modeling studies, the isoquinoline scaffold ligands 4 – 11 were readily prepared (scheme 6.1).
Figure 6.2 | Proposed allosteric inverse agonists for RORγt. Structures of indazoles 1 – 3 which were identified as allosteric inverse agonists for RORγt.21 Ligands 4 – 11 with the isoquinoline scaffold are proposed allosteric inverse agonists for RORγt identified in modeling studies.
For the synthesis of isoquinoline ligands 4 to 11 identical strategies were used except for the first synthesis step. The retrosynthetic scheme starting from 3- or 4-bromoisoquinoline is shown in scheme 6.1. The synthesis started with cross-coupling reactions on either the 3-bromoisoquinoline or the 4-bromoisoquinoline. The benzoic acid, the furan or the pyrrole were coupled in good yields using palladium or copper as catalyst. The nitrile was then installed as handle for the Grignard reaction in a two-step fashion without purification of the intermediate N-oxide. The Grignard reactions with the bromobenzene or the 1- bromo-2-(trifluoromethyl)benzene were performed in good yields. Unfortunately,
144 EPILOGUE
Grignard reactions with bis-substituted arenes – which showed the most potential in the modelling studies – were not yet successful. The syntheses of ligand 4 to 11 were then finished with the deprotection of the carboxylic acids.
Scheme 6.1 | Retrosynthetic scheme for the synthesis of 4 to 11. Identical synthesis strategy was applied for the synthesis of ligands 4 – 11 except for minor modification in the first cross-coupling step.
The half maximal inhibitory concentrations (IC50) were determined using a HTRF assay as described in chapter five of this thesis. The IC50 values were determined to be in the range of 20 – 50 μM for all ligands. Ligands with the trifluoromethyl substituent showed higher affinity across all the compounds compared to the non-substituted. Additionally, the ligands with the pyrrole substituent (8 and 9) displayed a weak agonistic profile for RORγt. Unfortunately, no structural data could yet be obtained for these ligands and although modeled for the allosteric pocket, the binding mode for these ligands is yet unknown.
a b
Figure 6.3 | Design of a fluorescent probe for RORγt. a) Substitutions on the C-6 position of the indazole structure would protrude though a gap in the structure between helices 5, 11 and 12. b) Design of the probe molecule includes a fluorescent label, a linker and a RORγt binding region.
145 CHAPTER 6
The x-ray structure reported in chapter five of this thesis showed that the allosteric ligand binding site is situated in close proximity to helix 12 on “the outside” of the protein with a clear opening between the helices 5, 11 and 12 (figure 6.3a). Substitution on the C6 position of the indazole scaffold with a linker protruding through this gap would allow for attachment of fluorescent dyes for detection (figure 6.3b). This in principle could then be used as biomarker in cellular assays or in the development of a high throughput screening method to identify novel compounds for the allosteric ligand binding pocket. Tracer molecule 19 was synthesized in eight steps (scheme 6.2), following and adapting reported work from Merck & Co,22 facilitating late stage modifications in linker units or fluorescent labels. The indazole scaffold was used for its high binding properties to RORγt in the allosteric pocket. Several linker units, with varying length and hydrophobicity were investigated, demonstrating only a minor loss in binding affinity for the hexane linker unit.
Scheme 6.2 | Synthesis of the fluorescent probe. Reagents and conditions: (a) NaNO2,
AcOH, H2O, room temp; (b) I2, KOH, DMAC 0 °C; (c) 2-chloro-6-(trifluoromethyl)benzoyl chloride, Et3N, DMAP, CH2Cl2, room temp; (d) (4-(tert-butoxycarbonyl)phenyl)boronic acid, Pd(dppf)Cl2, KOAc, dioxane/H2O 5:1 v/v, 90 °C; (e) LiOH, THF/H2O 1:1 v/v, room temp; (f) tert-butyl (6-aminohexyl)carbamate, PyBOB, DIPEA, CH2Cl2, room temp; (g)
CH2Cl2, TFA, H2O, room temp; (h) AlexaFluor-647 NHS-ester, DIPEA, DMSO, room temp.
As the AlexaFluor-647 dye makes for a good FRET pair with europium cryptates,23 the direct binding of 19 was measured via an HTRF assay using an α-His6-antibody labeled with europium chelates as FRET donor, where an increase in FRET ratio would correspond to binding of the labeled ligand to the protein. The Kd for 19 was determined to be 46 ± 12 nanomolar (figure 6.4a). The labeled ligand 8 was then used as a proof of principle in
146 EPILOGUE
competition assays (figure 6.4b) for ligands 1 – 3 The determined IC50 values were 1: 2.7 ± 2.1 μM, 2: 40 ± 3 nM and 3: 3.2 ± 0.2 nM, which is in good agreement with the previously reported values.21
Figure 6.4 | HTRF assays with the fluorescent probe 19 for His6-RORγt. a) Direct binding of the fluorescent probe molecule 19 to RORγt. Assay conditions: [RORγt]: 50 nM, [Eu- labeled anti-His6 antibody]: 1.2 nM and 19 at indicated concentrations. The Kd for 19 to RORγt was determined to be 46 ± 12 nM. b) HTRF competition assay with 19. Conditions:
[RORγt]: 50 nM, [Eu-labeled anti-His6 antibody]: 1.2 nM, [19]: 200 nM and indicated concentrations of ligands.
First steps towards the development of a high throughput assay were taken. To screen for allosteric modulators for RORγt blocking of the orthosteric ligand binding pocket might be a necessity. This might be achieved using a covalent ligand utilizing Cys299 in the ligand binding pocket. For the assay development a fluorescent label was used, other tags such as radiolabels or tags for, for example cell localization or proteolytic degradation might be explored. To explore whether the principle of allosteric modulation of nuclear receptors could be applied to other members of the protein family, in silico sequence and structural homology studies were performed. PPARγ was identified as a nuclear receptor which in principle is also capable of forming an allosteric pocket close to helix 12. In summary, this epilogue provides some entry points to expand on the work described in this thesis. Most importantly addressing the dimerization properties of RXR and the assay development for the allosteric pocket of RORγt as well as some insights into novel RORγt modulators.
147 CHAPTER 6
References
1. Dawson, M. I. & Xia, Z. The retinoid X receptors and their ligands. Biochim. Biophys. Acta 1821, 21–56 (2012). 2. Rastinejad, F., Huang, P., Chandra, V. & Khorasanizadeh, S. Understanding nuclear receptor form and function using structural biology. J. Mol. Endocrinol. 51, T1–T21 (2013). 3. Solt, L. A. & Burris, T. P. Action of RORs and their ligands in (patho)physiology. Trends Endocrinol. Metab. TEM 23, 619–627 (2012). 4. Fauber, B. P. & Magnuson, S. Modulators of the Nuclear Receptor Retinoic Acid Receptor- Related Orphan Receptor-γ (RORγ or RORc). J. Med. Chem. 57, 5871–5892 (2014). 5. Zhang, Y., Luo, X., Wu, D. & Xu, Y. ROR nuclear receptors: structures, related diseases, and drug discovery. Acta Pharmacol. Sin. 36, 71–87 (2015). 6. Lefebvre, P., Benomar, Y. & Staels, B. Retinoid X receptors: common heterodimerization partners with distinct functions. Trends Endocrinol. Metab. TEM 21, 676–683 (2010). 7. Barnard, J. H., Collings, J. C., Whiting, A., Przyborski, S. A. & Marder, T. B. Synthetic Retinoids: Structure–Activity Relationships. Chem. – Eur. J. 15, 11430–11442 (2009). 8. Chandra, V. et al. Structure of the intact PPAR-gamma-RXR- nuclear receptor complex on DNA. Nature 456, 350–356 (2008). 9. Lou, X. et al. Structure of the retinoid X receptor α–liver X receptor β (RXRα–LXRβ) heterodimer on DNA. Nat. Struct. Mol. Biol. 21, 277–281 (2014). 10.Kojetin, D. J. et al. Structural mechanism for signal transduction in RXR nuclear receptor heterodimers. Nat. Commun. 6, 8013 (2015). 11. Rastinejad, F., Ollendorff, V. & Polikarpov, I. Nuclear receptor full-length architectures: confronting myth and illusion with high resolution. Trends Biochem. Sci. 40, 16–24 (2015). 12. Moras, D., Billas, I. M. L., Rochel, N. & Klaholz, B. P. Structure-function relationships in nuclear receptors: the facts. Trends Biochem. Sci. 40, 287–290 (2015). 13. Giner, X. C., Cotnoir-White, D., Mader, S. & Lévesque, D. Selective ligand activity at Nur/retinoid X receptor complexes revealed by dimer-specific bioluminescence resonance energy transfer- based sensors. FASEB J. Off. Publ. Fed. Am. Soc. Exp. Biol. 29, 4256–4267 (2015). 14. Sundén, H. et al. Chiral Dihydrobenzofuran Acids Show Potent Retinoid X Receptor-Nuclear Receptor Related 1 Protein Dimer Activation. J. Med. Chem. 59, 1232–1238 (2016). 15. McFarland, K. et al. Low dose bexarotene treatment rescues dopamine neurons and restores behavioral function in models of Parkinson’s disease. ACS Chem. Neurosci. 4, 1430–1438 (2013). 16.Volakakis, N. et al. Nurr1 and Retinoid X Receptor Ligands Stimulate Ret Signaling in Dopamine Neurons and Can Alleviate α-Synuclein Disrupted Gene Expression. J. Neurosci. Off. J. Soc. Neurosci. 35, 14370–14385 (2015). 17. Wang, J. et al. Selective brain penetrable Nurr1 transactivator for treating Parkinson’s disease. Oncotarget 7, 7469–7479 (2016). 18. Huang, Z., Xie, H., Wang, R. & Sun, Z. Retinoid-related orphan receptor γt is a potential therapeutic target for controlling inflammatory autoimmunity. Expert Opin. Ther. Targets 11, 737– 743 (2007). 19.Kanai, T., Mikami, Y., Sujino, T., Hisamatsu, T. & Hibi, T. RORγt-dependent IL-17A-producing cells in the pathogenesis of intestinal inflammation. Mucosal Immunol. 5, 240–247 (2012). 20.Yang, Y. et al. Impact of suppressing retinoic acid-related orphan receptor gamma t (ROR)γt in ameliorating central nervous system autoimmunity. Clin. Exp. Immunol. 179, 108–118 (2015). 21. Scheepstra, M. et al. Identification of an allosteric binding site for RORγt inhibition. Nat. Commun. 6, 8833 (2015). 22.Karstens, W. F. J. et al. Rorgammat Inhibitors. PCT Int. Appl. WO 2012/106995 (2012).
148 EPILOGUE
23.Degorce, F. et al. HTRF: A technology tailored for drug discovery - a review of theoretical aspects and recent applications. Curr. Chem. Genomics 3, 22–32 (2009).
149
150 SUMMARY
Molecular modulation of nuclear receptor conformational states
Modulation of nuclear receptors remains an important theme in the chemical biology and medicinal chemistry. Nuclear receptors are involved in almost every aspect of human physiology such as development, metabolism or homeostasis. Malfunctioning of the delicate equilibria or misregulation of these receptors often lead to serious diseases such as cancer, obesity, neurological or autoimmune disorders. Since the discovery of nuclear receptors, major successes have been achieved in the development of small molecules for the modulation of the receptors, which is reflected in the large amount of marketed drugs for this class of receptors. Nevertheless, there remains a serious demand to expand on current ligand repertoire and to develop novel strategies to modulate nuclear receptors. Major challenges in the field are drug resistance and side-effects caused by lack of selectivity and efficacy. The work in this thesis focusses on two nuclear receptors: the retinoid X receptor (RXR) plays a major role in the human body through its dimerization properties with other nuclear receptors. For the second receptor described in this thesis named RAR-related orphan receptor gamma-t (RORγt), a role in the immune system was discovered recently. Nuclear receptor biology discovers new roles for the receptors and continues to evolve, this is then followed by significant efforts to discover new modulators for this class of receptors. Chapter two of this thesis describes the investigations into the dual binding mode of the natural product honokiol. Natural products have been an inspiration for new molecular structures for a long time and the extracts from the bark of the Magnolia officinalis contains active compounds that have been used in eastern medicine against several disorders. Honokiol is one of the biological active compounds and was identified to bind to RXR at two relevant sites in the receptor. Chemical modifications to the natural product allowed us to dissect the dual binding mode in either highly potent agonists for RXR or on the other hand a first-of-kind cofactor inhibitor for RXR. These findings justify the exploration of natural products for more difficult-to-address protein-protein interactions. The work described in chapter three continues on the natural product molecular scaffold. By introducing small modifications to the ligands local side-chain disturbances and correlated displacements of secondary structural elements on ligand binding were introduced in RXR. The potential of the biaryl scaffold was underlined by the relevant potencies found for RXR modulation. The data provide a rational for the design of RXR ligands to probe the ligand binding pocket influencing either coregulatory recruitment or
151 SUMMARY the RXR dimerization interface. Together this work demonstrates a comprehensive molecular mechanism for RXR modulation which might correspond to more general conserved mechanisms in the nuclear receptor protein family and might aid in efforts to make selective nuclear receptor modulators. Chapter four describes the extensive synthesis of a bis-banzanulated spiroketal molecule as ligand for RXR. Spiroketals are found abundantly in polyketide-derived nature products, some of which exhibit useful antibiotic properties, a precise structure-activity relationship for these compounds has rarely been established and they have never been explored as nuclear receptor modulators. The chapter presents the rational design, chemical synthesis, biological evaluation and structure elucidation of a spiroketal-derived agonist of RXR. These findings represent the first spiroketal-derived modulator of any nuclear receptor. The structural flexibility of the spiroketal scaffold justifies further exploration of this molecular scaffold for the selective modulation of nuclear receptors other than RXR. Chapter five is the report on the discovery of an allosteric binding pocket in the ligand binding domain of RORγt. RORγt is critical for the differentiation and proliferation of T- helper17 cells which are associated with chronic autoimmune diseases. Co-crystallization of the RORγt ligand binding domain with a series of small-molecule inverse agonists demonstrates the occupancy of a previously unreported allosteric site. The structural result of this is an unprecedented conformation for helix 12. Biochemical and cellular studies demonstrate the functional inhibition of RORγt. This brings forward an approach to target Th17-mediated chronic autoimmune diseases and establishes an alternative mechanism for nuclear receptor modulation. The epilogue of this thesis describes suggestions for future directions. Building on the natural product derived biaryl scaffold the dimerization properties and partial agonism for RXR can be investigated. Modulation of selective dimerization properties with permissive dimerization partners for RXR might offer new therapeutically relevant applications. The discovery of an allosteric pocket for RORγt might be further explored through different molecular scaffolds. Assay development using fluorescent probes might also allow high throughput screening to identify novel allosteric modulators for RORγt modulation. In conclusion, the work presented in this thesis demonstrates different ways to modulate nuclear receptor structural conformations using small molecules. Via rational design and targeted modifications to the ligands the functional outcome of the receptors can be modulated. This is relevant for our understanding of nuclear receptor functioning, but also holds potential for drug development.
152 SAMENVATTING
Moleculaire modulatie van conformaties in kernreceptoren
Modulatie van kern receptoren of ook wel hormoon receptoren is een belangrijk thema in de chemische biologie, de chemie en medicijn ontwikkeling. De receptor is betrokken bij bijna alle aspecten van de menselijke fysiologie zoals metabolisme, ontwikkeling en homestase. Misregulatie van deze receptoren leidt dan ook vaak tot aandoeningen, dit wordt onderstreept door het feit dat ziekten zoals kanker, obesitas of neurologische aandoeningen steeds grotere problemen worden. Sinds de ontdekking van de kernreceptoren zijn er al veel successen geboekt in het onderzoeksveld wat gereflecteerd wordt in de grote hoeveelheid medicijnen die werken via deze receptoren. Desalniettemin, is de vraag naar nieuwe manieren om de activiteit van de kernreceptoren te beïnvloeden, gebruikmakend van de opgebouwde kennis over de jaren. Grote vraagstukken blijven dan ook resistentie, ongewenste bijeffecten of een gebrek aan selectiviteit en effectiviteit. Het werk in dit proefschrift richt zich op twee kernreceptoren, als eerste de retinoïde X receptor (RXR), welke een grote rol speelt in het menselijk lichaam door de dimeer formatie met andere kernreceptoren. Voor de tweede de kernreceptor beschreven in dit proefschrift genaamd RORγt; is recent een rol ontdekt in auto-immuun ziekten. Het doel van het werk beschreven in dit proefschrift is het bestuderen van relevante conformaties van kernreceptoren die geïnduceerd worden door kleine moleculen. Voor dit doel is er een selectie van moleculen gesynthetiseerd met verschillende functionele uitkomsten zoals activatie of onderdrukking van transcriptionele activiteit. Naast de traditionele modulatie van de receptoren werden ook alternatieven gevonden. Hoofdstuk twee beschrijft het onderzoek naar de werking van de natuurstof honokiol, gevonden in de magnolia boom. Honokiol grijpt aan op twee relevante bindingsplaatsen in RXR, in eerste instantie als activator, maar in tweede instantie als onderdrukker van de receptor. Door chemische modificaties te maken aan het molecuul is de dubbele werking gesplitst in potente activatoren en onderdrukkers. Het onderzoek rechtvaardigt verder onderzoek naar natuurstoffen als inspiratie voor eiwit-eiwit interacties. Hoofdstuk drie richt zich op de traditionele modulatie van RXR bouwt op de voorgaande opgedane kennis. Door kleine verandering aan te brengen in de moleculen, gebruikmakend van hetzelfde moleculaire motief, kan er een verscheidenheid aan conformaties worden geïnduceerd in de receptor. Dit heeft gevolgen voor de dimeer formatie en activatie van de receptor en geeft een rationeel voor ligand design en ontwikkeling. Hoofdstuk vier beschrijft de uitgebreide synthese van een compleet nieuw ligand design voor RXR, namelijk spiroketalen.
153 SAMENVATTING
Spiroketalen worden veel gevonden in de natuur en vertonen veel biologische activiteit, echter deze klasse is nooit onderzocht als liganden voor kernreceptoren. Vooral de moeizame isolatie en de uitdagende synthese remmen het onderzoek naar deze klasse van moleculen. Het ligand bindende domein van RXR is uniek vergeleken met de andere kernreceptoren, door gebruik te maken van de 3-dimensionale structuur van spiroketalen kan selectiviteit voor dit RXR worden verkregen. De succesvolle synthese en biochemische karakterisatie laten zien dat deze klasse van liganden kan worden gebruikt voor de modulatie van RXR. De ontdekking van een alternatieve ligand bindende plaats in RORγt is beschreven in hoofdstuk vijf van dit proefschrift. Deze onverwachte zij-ingang voor deze kernreceptor maakt het mogelijk om de transcriptionele functie te blokkeren onafhankelijk van de lichaamseigen stoffen en wordt daardoor aantrekkelijk voor medicijnontwikkeling. De ontdekking maakt het mogelijk om gericht onderzoek te doen naar deze alternatieve ligand bindende plaats om zo potentiele stoffen te vinden in de behandeling van auto- immuun ziekten. In conclusie, het werk dat gepresenteerd is in dit proefschrift laat zien dat we controle kunnen krijgen over kernreceptoren gebruikmakend van kleine moleculen. Door gerichte verandering aan te brengen in de liganden kunnen we de conformatie en daardoor de functionele uitkomst bepalen voor deze eiwitten. Dit is relevant voor het verkrijgen van inzichten in de werking kernreceptoren, maar heeft ook potentie in de ontwikkeling van nieuwe medicijnen.
154 CURRICULUM VITAE
Curriculum Vitae
Marcel Scheepstra werd geboren op 7 november 1984 te Gouda. Na het behalen van zijn VWO diploma op de Goudse Scholen Gemeenschap – Leo Vroman, begon hij een jaar later de studie Biomedische Technologie aan de Technische Universiteit Eindhoven. Tijdens deze opleiding heeft hij stage gelopen in de groep van Professor B.G Davis aan de universiteit van Oxford, waar hij onderzoek deed naar chemische manieren om post-translationele modificaties te introduceren in eiwitten. Zijn afstudeerwerk werd uitgevoerd in de groep van Professor Luc Brunsveld en omvatte de ontwikkeling van kleine moleculen als agonisten voor kernreceptoren gebaseerd op natuurstoffen. Na het behalen van zijn masterdiploma in 2012 begon hij zijn promotieonderzoek onder leiding van Professor Luc Brunsveld. Het doel van zijn onderzoek was het bestuderen van alternatieve conformaties van kernreceptoren en het moduleren van de transcriptionele activiteit via kleine moleculen. De belangrijkste resultaten van dit onderzoek staan beschreven in dit proefschrift.
Marcel Scheepstra was born on 7 November 1984 in Gouda, The Netherlands. After the completion of his pre-university education at the Goudse Scholen Gemeenschap – Leo Vroman, he started studying Biomedical engineering at the Eindhoven University of Technology a year later. During his studies he performed an internship in the group of Professor B.G Davis at the University of Oxford, where he studied chemical approaches to introduce posttranslational modifications in proteins. His master project was completed in the group of Professor Luc Brunsveld and involved small molecules based on natural products for nuclear receptor modulation. In 2012 he started as a PhD candidate in the chemical biology group under supervision of Professor Luc Brunsveld at the Eindhoven University of Technology. The goal of his research was to study alternative conformations in nuclear receptors. The most important results of this research are presented in this thesis.
155
List of publications
H. Sundén, A. Schäfer, M. Scheepstra, S. Leysen, M. Malo, J. Ma, E.S. Burstein, C. Ottmann, L. Brunsveld, R. Olsson, Chiral Dihydrobenzofuran Acids Show Potent Retinoid X Receptor–Nuclear Receptor Related 1 Protein Dimer Activation, J. Med. Chem., 2016, 59, 1232–1238. I.M. Tharun, L. Nieto Garrido, C. Haase, M. Scheepstra, M. Balk, S. Moecklinghoff, W. Adriaens, S.A. Dames, L. Brunsveld, Subtype-specific modulation of estrogen receptor-coactivator interaction by phosphorylation. ACS Chem. Biol., 2015, 10, 475- 484. M. Scheepstra, S. Leysen, G.C. van Almen, J.R. Miller, J. Piesvaux, V. Kutilek, H. van Eenennaam, H. Zhang, K. Barr, S. Nagpal, S.M. Soisson, M. Kornienko, K. Wiley, N. Elsen, S. Sharma, C.C. Correll, B.W. Trotter, M. van der Stelt, A. Oubrie, C. Ottmann, G. Parthasarathy, L. Brunsveld, Identification of an allosteric binding site for RORγt inhibition. Nat. Commun. 2015, 6:8833 doi: 10.1038/ncomms8833. This article was picked up by 6 news outlets and featured in the Faculty of 1000 M. Scheepstra, L. Nieto Garrido, A.K.H. Hirsch, S. Fuchs, S. Leysen, C.V. Lam, L. in het Panhuis, C.A.A. van Boeckel, H. Wienk, R. Boelens, C. Ottmann, L.G. Milroy, L. Brunsveld, A natural-product switch for a dynamic protein interface. Angew. Chem. Int. Ed., 2014, 53, 6443-6448. This article was selected as a 'Hot Paper' by the referees at Angewandte Chemie D.A. Uhlenheuer, J.F. Young, H. Nguyen, M. Scheepstra, L. Brunsveld, Cucurbit[8]uril induced heterodimerization of methylviologen and naphthalene functionalized proteins. Chem. Commun., 2011, 47, 6798-6800.
156 ACKNOWLEDGEMENTS
Acknowledgements
De afgelopen jaren heb ik het geluk gehad omringd te zijn door vele goede collega’s, vrienden en familie. Een promotieonderzoek doe je dan ook niet alleen en via dit dankwoord wil ik mijn waardering en dank uitspreken aan iedereen die geholpen heeft. Luc, als promotor heb ik je manier van begeleiden altijd zeer gewaardeerd. De sfeer en het onderzoek in de groep is naar mijn mening zeer goed en dat is aan jou te danken. Je zegt zelf altijd dat de nadruk moet liggen op het onderwijs en persoonlijke ontwikkeling, ik heb dan ook altijd alle kansen gehad om mezelf te ontwikkelen op wetenschappelijk en persoonlijk vlak. Ik ben je dan ook zeer dankbaar voor de mogelijkheid die je me geboden hebt om in de Chemische Biologie groep te promoveren en alle begeleiding die je me daarbij hebt gegeven. Lech, as co-promotor I would like to thank you for critically reading my thesis and the supervision in the many projects that we worked on the past four years and as my supervisor during my master project. I hope you will enjoy your scientific career. I would also like to thank prof.dr. A.R. de Lera, prof.dr. C.A.A. van Boeckel, prof.dr. K. De Bosscher, prof.dr. E.W. Meijer and dr. C. Ottmann for critically reading my thesis and taking part in my defense committee The scientific contributions to the work in this thesis are mentioned in each chapter, nevertheless I would like to thank once more the people I worked with. My chemical biology experience started with an internship with Dana Uhlenheuer. Dana, thanks for the good start you gave me in the lab. The work described in chapter 2 could not have been done without: Lidia Nieto, Anna Hirsch, Sascha Fuchs, Seppe Leysen, Chan Vinh Lam. For the work described in Chapter 5 I would like to thank: Seppe Leysen, Geert van Almen, Mario van der Stelt, Arthur Oubrie and all the people from Merck Sharp & Dohme. For the joint scientific projects: Inga Tharun, Christian Haase, Mark Balk, Wencke Adriaens, Henrik Sundén, Anja Schäfer, prof. Roger Olsson. Also the other former group members: Katja Petkau - Milroy, Pauline Neirynck, Ralph Bosmans, Maria Bartel, Michael Sonntag, Dung Dang, Matthew Burton, Ingrid de Vries - van Leeuwen. Current group members: Richard, Sebastian, Jeroen, Sam, Anniek, Eva, Thuur, Bas, Jurgen, Ellen, Eline, Loes, for all the helpful discussions during meetings, the lunches, drinks, parties, retreats and the nice working atmosphere. Also the Protein Engineering group of Maarten Merkx, thanks for all the fruitful discussions and suggestion in the Friday morning meeting. Best of luck with finishing your PhD’s or postdocs and all the best in your future careers.
157 ACKNOWLEDGEMENTS
Speciale dank ook aan mijn paranimfen Remco en Sebastian. Ik wens jullie veel succes met het afmaken van jullie PhD’s en verdere carrière. Tegen de studenten die ik heb mogen begeleiden (master of andere projecten): Nicky Hoek, Sebastian Andrei, Eline Sijbesma, Rowin de Visser, Pim de Vink, Lenne Lemmens, Mark van den Bosch, Joris Adriaans, Femke Meijer: dank voor jullie hulp en veel succes in jullie verdere carrière, wetenschappelijk of anderszins. Heel veel dank ook aan alle mensen die heel veel werk verrichten om alles goed te laten verlopen. Jolanda Spiering, dank je voor de solide basis voor het synthese werk. Peggy de Graaf – Heuvelmans voor de organisatie van het bio-lab. Henk Eding voor de straffe koffie, Kantoorartikelen, schoonmaakspullen, taart, borrels en de vele andere dingen die je regelt. Hans Damen voor het afhandelen van bestellingen en het in goede banen leiden van alle chemicaliën in het gebouw. Het analytisch team, in het bijzonder Joost van Dongen voor de hulp met de karakterisatie van moleculen en chirale kolom chromatografie. Marjo van Hoof en Joke Rediker voor het uit handen nemen van administratieve taken. Alle mensen die naast hun onderzoek een hoop tijd investeren in het draaiende houden van de NMR apparaten. Voor de goede sfeer op het kantoor wil ik graag alle voorgaande, maar ook huidige kantoorgenoten bedanken. Het nadeel van het dankwoord schrijven na vier jaar onderzoek is het risico dat je iemand vergeet. Mocht dat het geval zijn, mijn excuses hiervoor en toch heel erg bedankt. Tegen de gasten van Meteoor zeg ik alleen maar: daar gaat ie… tot straks. Als laatste wil ik mijn familie bedanken. Pap, mam, dank voor alle steun tijdens mijn studie en promotie. Leren doe je voor jezelf was altijd het advies. Lieve Ilse, Martijn, Luuk en Feie, dank voor de vele gezellige momenten, leuke bbq’s en het goede voorbeeld met jullie gezellige gezin. Lieve Mariola, dank voor alle lieve steun die ik van jou krijg! Nu weer een nieuw avontuur en een nieuwe stap in ons leven met ons gezin.
Marcel
158