Precise Control Over Using THC-Photoswitches

  • Computer-based docking of cis-azo-THC 1 and trans-azo-THC 2 into the X-ray crystal structure of CB1 exemplifies their complementary behavior concerning cis- and trans-activity. Figures adapted from [1] with permission Copyright (2017) American Chemical Society.Computer-based docking of cis-azo-THC 1 and trans-azo-THC 2 into the X-ray crystal structure of CB1 exemplifies their complementary behavior concerning cis- and trans-activity. Figures adapted from [1] with permission Copyright (2017) American Chemical Society.
  • Computer-based docking of cis-azo-THC 1 and trans-azo-THC 2 into the X-ray crystal structure of CB1 exemplifies their complementary behavior concerning cis- and trans-activity. Figures adapted from [1] with permission Copyright (2017) American Chemical Society.
  • Fig. 1: Photoswitching between trans- und cis-azo-THC 1 enables reversible optical control of CB1 receptor activity. Figures adapted from [1] with permission Copyright (2017) American Chemical Society.

Tetrahydrocannabinol (THC) has high profile in the current public awareness. It is both hailed as remedy for numerous ailments by supporters of its medicinal use as well as antagonized for its psychotropic effects by opponents of Cannabis legalization. Recent successful applications of THC formulations in the treatment of various diseases, as well as the changing political climate in the western hemisphere will likely lead to widespread legalization at least for therapeutic purposes.

A scientifically sound understanding of the molecular mechanism of action of THC in the human body is of utmost importance for the assessment of opportunities along with dangers of THC as well as Cannabis formulations and may serve as basis for decision making by government bodies setting policy.

The Endocannabinoid System

Cannabinoid receptors 1 and 2 (CB1 and CB2) were discovered as a consequence of the identification of THC as the psychotropic component in Cannabis Sativa in the middle of the 20th century. Together they form the central element of the so-called endocannabinoid system, one of the most important signaling systems in all mammals, including humans. CB1 is the most frequently expressed G protein-coupled receptor in the central nervous system and plays a key role in regulating coordination, mood and cognition [2]. The second cannabinoid receptor, CB2, is located mainly on cells of the immune system and is an attractive therapeutic target for treatment of neurodegenerative diseases such as Alzheimer’s as well as cancer [3,4]. Thus, the particular interest of the pharmaceutical industry for modulating the cannabinoid receptors is no surprise.

The understanding of the subtle changes in cellular processes induced by (de-)activation of cannabinoid receptors upon binding of endo- and exogeneous ligands is still rudimentary. A prominent challenge for such studies is the high lipophilicity of most cannabinoid ligands, leading to their uncontrolled distribution in the cell membranes. Consequently, precise investigation of the dynamics of receptor (de-)activation is difficult to conduct. A powerful approach to tackle this problem involves photopharmacology, which relies on the reversible cycling between isomers of a ligand that ideally exhibit different biological activity at the biological target.

The use of light as the stimulus for isomerization leads to unmatched spatiotemporal degree of control over biological function.

Control of CB1 Activity

The entry point for our efforts towards optical control of CB1 activity was ∆9‑THC, the main active compound in Cannabis and a known CB1 partial agonist. Gratifyingly we were able to rely on our expertise in stereodivergent synthesis of THC [5]. The challenge at hand was the modulation of the inherent THC activity at CB1 using light. In photopharmacology, azobenzenes are by far the most widespread chromophores owing to their relative ease of preparation as well as to their isomeric forms (E and Z) having both sterically and electronically very different properties. When properly implemented, these characteristics can lead to drastically different pharmacological profiles of the isomeric ligand-azobenzene molecules. Accordingly, we introduced a number of azobenzenes into the THC scaffold through the power of modern coupling methods. The resulting synthetic, light responsive THC derivatives, or azo-THCs, were then evaluated in collaboration and tested in various cell-based assays.

Dr. James Frank and Prof. Dirk Trauner from LMU Munich used patch-clamp electrophysiology in order to measure the activity of CB1 on mouse tumor  cells (AtT‑20(CB1)). These cells were transiently transfected with GIRK (G protein-coupled inwardly rectifying potassium channels), a potassium channel coupled to CB1 via Gβγ-subunit of the heterotrimeric G protein. Binding of agonists such as THC to CB1 leads to opening of GIRK and potassium ions flowing out of the cells. The resulting electrical current was recorded by help of electrodes. The first promising result was obtained using trans-azo-THC 1, which induced a weak current when applied to the cells. Isomerization to the cis-form using ultraviolet light (360 nm) potentiated this effect, and a larger current was measured. Irradiation with blue light (450 nm) restored the weak current invoked by trans-azo-THC 1. These observations could be repeated over many cycles by successive irradiation with the wavelengths noted (Fig. 1). Rimonabant, a potent cannabinoid receptor antagonist, negated any effect of both trans- and cis-azo-THC 1, showcasing that the measured currents are a consequence of azo-THC 1 binding to CB1 rather than directly to GIRK. These results represent successful optical control of CB1 activity using azo-THCs. Additionally, another promising THC derivative, azo‑THC 2, was identified. In contrast to azo‑THC 1, this compound is more active in its thermodynamically more stable trans-form and the observed current decreased upon isomerization to the cis-form.

CB1-GIRK coupling is only one among many possibilities for the evaluation of CB1 optical control. Dr. Amey Dhopeswarkar and Prof. Ken Mackie from the University of Indiana were able to control cellular cAMP levels using azo-THCs. CB1 is negatively coupled to adenylyl cyclase via Gαi-subunit, and activation leads to reduced cAMP concentration. As in the case of electrophysiology, azo-THC 1 was more active in its cis-form, azo-THC 2 in its trans-form. This complementary behavior of the two azo-THCs was rationalized using computer-based modeling performed by Dr. Jessica Grandner and Prof. Vsevolod Katritch from the University of Southern California. These calculations showed that cis-azo-THC 1 and trans-azo-THC 2 could bind with high affinity to the model generated from the recently published CB1 X-ray structure (Fig. 2). Their respective isomers exhibited unfavorable interactions with the receptor, preventing efficient binding.

Conclusion and Outlook

azo-THC 1 and 2 represent the first tool compounds for reversible optical control of CB1 receptor activity. Moreover, they offer flexibility for biological experiments since due to their complementary behavior either may be used depending on whether higher activity is desired in the thermodynamically more stable trans-form or after photoisomerization to the cis-form. Meanwhile there is still room for improvement. azo-THC 1 and 2 do not exert control of CB1 activity completely in an ON/OFF fashion, because the less active isomer exhibits some basal activity. Efforts towards increasing the activity difference between the cis- and trans-isomer of azo-THCs are currently in progress. Another goal is the development of red-shifted azo-THCs which isomerize upon irradiation with red or even infrared light. In contrast to ultraviolet and blue, red light penetrates deeper into tissue and is thus more suitable for photopharmacological experiments in live tissue and in-vivo. Eventually, translation of the results of this study to the therapeutically promising CB2 receptor is highly desired.

azo-THCs constitute an exciting new possibility to decisively expand the understanding of the signaling cascades controlled by CB1 in high precision and in various cell types. Ultimately, enriched knowledge on the dynamics of CB1- and CB2 (de-)activation may be invaluable for the development of new therapies against neurodegenerative diseases such as Alz-heimer’s and Multiple Sclerosis, as well as against cancer.

Roman C.Roman C. Sarott1 and Prof. Erick M. Carreira1

1Departement Chemistry and Applied Biosciences, Laboratory of Organic Chemistry, ETH Zürich, Zürich, Switzerland

Prof. Erick M. Carreira

ETH Zürich
Laboratorium für Organische Chemie
Zürich, Switzerland


[1] Westphal, M. V.; Schafroth, M. A.; Sarott, R. C.; Imhof, M. A.; Bold, C. P.; Leippe, P.; Dhopeshwarkar, A.; Grandner, J.; Katritch, V.; Mackie, K.; Trauner, D.; Carreira, E. M.; Frank, J.; "Synthesis of Photoswitchable Δ9-Tetrahydrocannabinol Derivatives Enables Optical Control of Cannabinoid Receptor 1 Signaling" A. J. Am. Chem. Soc., 2017, 139, 18206. DOI:10.1021/jacs.7b06456

[2] Svíženská, I.; Dubový, P.; Šulcová, A. Pharmacol.; "Cannabinoid receptors 1 and 2 (CB1 and CB2), their distribution, ligands and functional involvement in nervous system structures" Biochem. Behav. 2008, 90, 501. DOI:10.1016/j.pbb.2008.05.010

[3] Ranieri R.; Laezza C.; Bifulco M.; Marasco D.; Malfitano AM; "Endocannabinoid System in Neurological Disorders" Recent Pat CNS Drug Discov. 2016, 10, 90. DOI:10.2174/1574889810999160719105433

[4] Marino S.; Idris AI.; "Emerging therapeutic targets in cancer induced bone disease: A focus on the peripheral type 2 cannabinoid receptor" Pharmacol Res. 2017 119, 391. DOI:10.1016/j.phrs.2017.02.023

[5] Schafroth, M. A.; Zuccarello, G.; Krautwald, S.; Sarlah, D.; Carreira, E. M.; "Stereodivergent Total Synthesis of Δ9-Tetrahydrocannabinols" Angew. Chem., Int. Ed. 2014, 53, 13898. DOI:10.1002/anie.201408380

More on Cannabinoids:


ETH Zürich
Vladimir-Prelog-Weg 1-5/10
8093 Zürich

Register now!

The latest information directly via newsletter.

To prevent automated spam submissions leave this field empty.