Subscriber access provided by University of Florida | Smathers Libraries
Article 9
Synthesis of Photoswitchable #-Tetrahydrocannabinol Derivatives Enables Optical Control of Cannabinoid Receptor 1 Signaling Matthias Westphal, Michael Andreas Schafroth, Roman Sarott, Michael Imhof, Christian Bold, Philipp Leippe, Amey Dhopeshwarkar, Jessica Grandner, Vsevolod Katritch, Ken Mackie, Dirk Trauner, Erick M Carreira, and James Allen Frank J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b06456 • Publication Date (Web): 21 Nov 2017 Downloaded from http://pubs.acs.org on November 21, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Synthesis of Photoswitchable 9-Tetrahydrocannabinol Derivatives Enables Optical Control of Cannabinoid Receptor 1 Signaling Matthias V. Westphal,∫,† Michael A. Schafroth, ∫,† Roman C. Sarott,† Michael A. Imhof,† Christian P. Bold,† Philipp Leippe,‡ Amey Dhopeshwarkar,# Jessica M. Grandner,┴ Vsevolod Katritch,┴ Ken Mackie,# Dirk Trauner,*,‡,§ Erick M. Carreira,*† and James A. Frank*‡Ψ † Laboratorium für Organische Chemie, Eidgenössische Technische Hochschule Zürich, Vladimir-Prelog-Weg 3, 8093 Zürich, Switzerland ‡ Department of Chemistry and Center for Integrated Protein Science, Ludwig Maximilians University Munich, Butenandtstraße 5-13, 81377 Munich, Germany # Department of Psychological and Brain Sciences and the Gill Center, Indiana University, Bloomington, IN 47405, USA ┴ Department of Biological Sciences and Department of Chemistry, Bridge Institute, University of Southern California, Los Angeles, California 90089, USA § Department of Chemistry, New York University, 100 Washington Square East, New York, NY 10003-6699, USA Ψ present address: Research Laboratory of Electronics, Massachusetts Institute of Technology, 50 Vassar St., Cambridge, Massachusetts 02139, USA KEYWORDS tetrahydrocannabinol, photopharmacology, CB1, GIRK channel, inhibitory GPCR ABSTRACT: The cannabinoid receptor 1 (CB1) is an inhibitory G protein-coupled receptor abundantly expressed in the central nervous system. It has rich pharmacology and largely accounts for the recreational use of cannabis. We describe efficient asymmetric syntheses of four photoswitchable 9-tetrahydrocannabinol derivatives (azo-THCs) from a central building block 3-Br-THC. Using electrophysiology and a FRET-based cAMP assay, two compounds are identified as potent CB1 agonists that change their effect upon illumination. As such, azo-THCs enable CB1-mediated optical control of inwardly-rectifying potassium channels, as well as adenylyl cyclase.
INTRODUCTION Preparations of Cannabis sativa have long been used for recreational and therapeutic purposes across many cultures.1,2 The identification of 9-tetrahydrocannabinol (9-THC) as the major psychoactive constituent in Cannabis3 led to the discovery of the endocannabinoid system, which consists of cannabinoid receptors (CBRs),4,5 endogenous lipophilic ligands (endocannabinoids)6,7 and the enzymes regulating their synthesis and degradation.8 Cannabinoid receptor 1 (CB1) is one of the most widely expressed G protein-coupled receptors (GPCRs) in the central nervous system, and its activation has been associated with mood, motor coordination, memory, cognition and other physiological processes.9 In contrast, cannabinoid receptor 2 (CB2) is expressed primarily in the periphery, especially in cells of the immune system,10 and is a potential therapeutic target for the treatment of autoimmune diseases, neuropathic pain, and cancer.11,12 Together, the CBRs are inhibitory receptors expressed almost ubiquitously throughout the human body, and are crucial to signal transduction, hormone secretion9 and energy metabolism.8,13 Therefore, tools which enable
precise control of cannabinoid receptors are vitally important to understand their diverse functions and relation to signal transduction and disease states. Pharmacological manipulation of CBRs through the use of classical and non-classical cannabinoids14 is often limited by their hydrophobic nature. Ligand activation/inactivation kinetics are notoriously difficult to control, as these compounds remain primarily distributed within cell membranes. Since light can be applied with unmatched degree of spatial and temporal precision, photochromic ligands offer an attractive solution to this problem.15 In one approach, a pharmacophore is masked by a photolabile protecting group which can be removed upon a flash of light, enabling precise control of ligand release.16-18 This approach has been applied to the endocannabinoid anandamide.19 However, ligand release is an irreversible process and subsequent ligand deactivation relies on metabolism or active transport. To address this limitation, photoswitchable small molecules can be used to enable reversible optical control over biological function,20 a strategy known as photopharmacology.21 In this approach, a photoswitch is incorporated into the ligand and allows for reversible cycling between
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
distinct conformations which exhibit different activities at their target. This strategy has been applied to achieve reversible and time-controlled modulation of ion channels, GPCRs, enzymes, and transcription factors.22 Currently, no tools exist which enable reversible optical control over CBR function. Herein, we describe the synthesis of 3-Br-THC and its elaboration into several azobenzene-containing photoswitchable 9-THC derivatives
Page 2 of 7
(azo-THCs). Using whole-cell electrophysiology and a FRET-based cAMP assay, we demonstrate that the efficacy of two azo-THCs can be modulated using light to facilitate optical control of CB1 and its downstream signaling pathways. These include both the activation of G protein-coupled inwardly rectifying potassium (GIRK) channels and the inhibition of adenylyl cyclase.
Figure 1. Design of photoswitchable 9-THC derivatives, azo-THCs. (A) Structures and CB1 Ki values of -THC and derivatives bearing bulky C-3 substituents as assessed by radioligand binding assay using rat brain preparations. (B) 3-Br-THC as the central building block for the preparation of azo-THCs.
RESULTS Design, synthesis and characterization of photoswitchable 9-THC derivatives The design of photoswitchable 9-THCs was partially guided by the recently elucidated structure of the CB1 receptor.23 In these studies, in silico docking suggested that 9THC could bind within the transmembrane helices with the C-3-alkyl chain extending towards Trp356. This is a conserved tryptophan residue among class A GPCRs and is thought to be an important toggle-switch for CB1 activation.24 In addition, the precedence of high-affinity THC derivatives bearing bulky C-3-substituents, such as 1,1-dimethylheptyl (1),25 adamantyl (2) and naphthyl (3) groups (Fig. 1A),26,27 strongly suggested that incorporation of sterically demanding residues could be tolerated by the receptor. In photopharmacology, azobenzenes are commonly used chromophores.20,21 Accordingly, four probes containing this photoswitch at C-3 were designed (Fig. 1B). azo-THC-1 possesses the diazene directly attached to the aromatic 9-THC core, while azo-THC-2–4 contain a phenyl spacer with diazenes in the ortho-, meta- and para-positions, respectively. The synthesis of azo-THC-1–4 relied on the novel building block termed 3-Br-THC as a convenient intermediate for late-stage diversification through cross-coupling (Fig. 1B). The synthesis of 3-Br-THC demanded alteration of our previously published route (harsh deprotection conditions: neat MeMgI at 160 °C),28 in particular for the construction of the chromane motif. We subjected 4 to Ir-catalyzed allylation to afford 5 in 52% yield (dr > 20:1, ee > 99%). Ring-
closing metathesis using Grubbs II catalyst followed by iodine mediated oxidative esterificationN29 and double Grignard addition afforded tertiary alcohol 7 in 76% yield from 5. Chromane 8 was then constructed by KHMDS mediated intramolecular SNAr displacement30 in 90% yield. Demethylation using conditions commonly employed (NaSEt, DMF, 120 °C)31,32 yielded 3-Br-THC (Scheme 1). Scheme 1. Synthesis of 3-Br-THC
Reagents and conditions: (a) [Ir(COD)Cl]2 (3 mol%), (S)-PN-olefin (12 mol%), (S)-Jørgensen–Hayashi catalyst (15 mol%), Zn(OTf)2 (5 mol%), 5-methylhex-5-enal, DCE, rt, 52%. (b) HG-II (3 mol%), CH2Cl2, rt, 91%. (c) I2, KOH, MeOH, 0 °C, 88%. (d) MeMgI, Et2O, 0 °C, 95%. (e) KHMDS, THF–PhMe, 65 °C, 90%. (f) NaSEt, DMF, 140 °C, 86%.
The 1H NMR spectra of prepared material suggested the presence of a minor co-product. This was most evident by a characteristic resonance signal (=3.51 ppm) shifted downfield relative to the one assigned to the benzylic proton in 3-Br-THC (=3.15 ppm). The benzylic proton of independently synthesized 9-THC with cis-configuration of the vicinal stereocenters resonates at =3.56 ppm (compared
ACS Paragon Plus Environment
Page 3 of 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
to =3.20 ppm for the trans-diastereoisomer).28 This comparison led to the conclusion that during demethylation, epimerization occurred to a minor extent (trans to cis at ring fusion > 10:1). During the early stages of this project, Dethe et al. reported the nominal synthesis of 3-Br-THC by BF3·Et2O-mediated Friedel-Crafts reaction of 1-bromo-3,5-dihydroxybenzene with a monoterpenoid derivative.33 We note, however, that the NMR spectra of material prepared as described in Scheme 1 differs significantly from the published data. We confirmed our structure by elaborating 8 into known 9-tetrahydrocannabinol methyl ether (see SI).34 Scheme 2. Synthesis of azo-THC-1–4
Reagents and conditions: (a) TBSOTf, 2,6-lutidine, rt, 74%. (b) Pd(OAc)2 (10 mol%), P(tBu)3·HBF4 (10 mol%), Cs2CO3, PhMe, 110 °C, 89%. (c) 2,6-lutidine, TMSOTf, CH2Cl2, 0 °C; NaHCO3, air, MeOH, rt, 52%. (d) TBAF, THF, 0 °C, 93%. (e) azo-THC-2/3: PdCl2(dppf) (5 mol%), Cs2CO3, 10/11, MeOH, 65 °C, 53%/70%. azo-THC-4: PdCl2(dppf) (5 mol%), 12, DMF–H2O, 65 °C, 38%.
TBS-protection of 3-Br-THC and Buchwald-Hartwig coupling with N-Boc-phenylhydrazine afforded 9 in 66% yield (Scheme 2). Trimethylsilyl triflate mediated Boc-removal35 gave an intermediate hydrazine, which after workup was oxidized by ambient atmosphere to furnished the silyl-protected photoswitch. Deprotection with TBAF then afforded azo-THC-1 in 48% yield from 9. For the synthesis of azo-THC-2 and azo-THC-3, unprotected 3-Br-THC was directly derivatized by Suzuki cross-coupling with aryltrifluoroborates 10 and 11 (meta and para)36 to afford the desired products in 53% and 70% yield, respectively. However, the synthesis of ortho-substituted azo-THC-4 proved more challenging. As our attempts to prepare the ortho-substituted potassium trifluoroborate from the respective pinacol ester failed, we turned our attention to lithium triolborate 12.37 Although 3-Br-THC was entirely consumed during the attempted reaction with 12 under the conditions employed for potassium trifluoroborate cross-coupling (PdCl2(dppf), MeOH, Cs2CO3), no product could be isolated, likely due to base-mediated decomposition. Triolborates are known to readily undergo cross-coupling reactions without additional base, rendering them ideal reagents for sensitive substrates. Indeed, in the absence of additional base azo-THC-4 was isolated upon PdCl2(dppf)-catalyzed cross-coupling in 38% yield. Using UV-Vis spectroscopy, we determined that all azo-THCs behaved as conventional azobenzenes (Fig. 2, Fig. S1). In the dark, they existed in the thermodynamically favored trans-configuration. Isomerization to the cis-form
was triggered on UV-A irradiation (365 nm), and blue light (450 nm) reversed this process. The compounds were also bi-stable. For example, cis-azo-THC-3 underwent thermal relaxation to the trans-isomer with a -value of 216 min (3.60 h) in water.
Optical control of CB1-GIRK channel coupling CB1 is coupled via G subunits of heterotrimeric G proteins to GIRK channels,38 which hyperpolarize the cell on activation. To screen and compare the applicability of the four azo-THCs to control CB1 activation in living cells, we utilized whole-cell
Figure 2. Photoisomerization of azo-THCs. (A) azo-THC-3 could be isomerized between its cis- and trans-configurations with UV-A and blue irradiation, respectively. (B) The UV-Vis spectra of azo-THC-3 (20 M in ddH2O) in its dark-adapted (black, trans), UV-A-adapted (grey, cis isomer) and blue-adapted (blue, trans isomer) photostationary states.
patch clamp electrophysiology in a mouse tumor cell line which stably express CB1 receptors (AtT-20(CB1) cells).39 These cells were transiently transfected with the GIRK1 and GIRK2 K+ channel subunits,40 which form GIRK channels and produce an inwardly rectifying K+ (KIR) current in response to CB1 activation.41 As validation for the CB1-GIRK channel coupling in this heterologous expression system, the CB1 agonist CP-55940 (100 nM) produced an inward current when holding the cellular membrane potential at 60 mV (Figure S2A). Increased KIR currents were observed in both voltage ramps (140 to 20 mV) and voltage steps (120 to 20 mV, 20 mV steps) (Fig. S2B,C). We then applied each azo-THC ligand (at 2 M), and measured KIR currents under blue (450 nm, trans isomer) and UV-A (360 nm, cis isomer) irradiation (Fig. 3A). To compensate for heterogeneous GIRK channel expression levels between individual cells, the measured currents were normalized to the maximum current evoked by CP-55940 (100 nM), which was applied at the end of each experiment (Fig. 3B). In this case, azo-THC-1 and azo-THC-2 were inactive when compared to the vehicle control and did not affect KIR currents in the trans- or cis-configurations (Fig. 3A, Fig. S3B-D). However, azo-THC-3 and azo-THC-4 emerged as photoswitchable ligands for CB1. When holding the cellular membrane potential at 60 mV, application of trans-azo-THC-3 (2 M) produced a small inward current that was potentiated by isomerization to the cis-form upon 360 nm irradiation (Fig. 3C). This effect could be immediately reversed with blue light, and the change in efficacy could be repeated over multiple cycles. This demonstrates that azo-THC-3 acti-
ACS Paragon Plus Environment
Journal of the American Chemical Society
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
vates CB1 more strongly in the cis-configuration. In contrast, azo-THC-4 was more potent in the trans-configuration. Application of azo-THC-4 (2 M) caused a robust inward current that was sharply reduced on isomerization to the cis-isomer (Fig. 3D). Again, this effect was reversed on blue irradiation. These results were confirmed by measuring current-voltage (IV)-steps between 120 to 20 mV. Greater KIR currents were observed in the presence of cis-azo-THC-3 (Fig. 3E) and similarly trans-azo-THC-4 (Fig. 3F). We sought to explore the effects of azo-THC-3 further, as it can be considered a “reversibly caged” ligand due to its decreased efficacy in the dark-adapted trans-form. Voltage ramps in the presence of trans- and cis-azo-THC-3 confirmed that the largest current difference was observed at
Page 4 of 7
negative membrane potentials (inward current) (Fig. 4A). Above 0 mV (outward current), essentially no photocurrent was observed, confirming that the measured light-dependency of KIR was due to the activation of GIRK channels. Over many photoswitching cycles, the photocurrent induced by isomerization to the cis-form gradually decreased (Fig. 4B), an effect which is likely attributed to CB1 desensitization.42 This effect could also be observed by exploiting the bi-stable nature of azo-THC-3 (Fig. 4C). After photoactivation to the cis isomer, the resulting current slowly diminished over time in the dark, and the initial cis-current was not fully restored upon another isomerization cycle. The CBR antagonist rimonabant43 (RIM, 2 M) entirely abolished the effects induced by trans- or cis-azo-THC-3 (Fig. 4D), suggesting that the recorded KIR currents were CB1-dependent and not a direct effect on the GIRK channels.
Figure 3. Optical control of CB1-GIRK coupling. azo-THCs were evaluated using whole-cell electrophysiology in AtT-20(CB1) cells transiently expressing GIRK channels. (A) After application of azo-THC-3 (2 M), IV-steps were recorded between 120 to 20 mV under 450 and then 360 nm irradiation. Currents were normalized to the maximum current evoked by CP-55940 (100 nM), displayed as overlaid IV-steps from a representative cell. (B) KIR currents for azo-THC-1–4 (2 M) normalized to the maximum current evoked by CP-55940. (C, D) KIR currents for azo-THC-3 and azo-THC-4 upon irradiation at 450 and 360 nm. (E, F) IV-relationships from multiple cells for azo-THC-3 and azo-THC-4. Figures in brackets indicate the number of independent experiments. ns = not significant, P>0.05, *P
