Oximes short-acting CB1 receptor agonists

https://doi.org/10.1016/j.bmc.2018.08.003Get rights and content

Abstract

New oximes short-acting CB1 agonists were explored by the introduction of an internal oxime and polar groups at the C3 alkyl tail of Δ8-THC. The scope of the research was to drastically alter two important physicochemical properties hydrophobicity (log P) and topological surface area (tPSA) of the compound, which play a critical role in tissue distribution and sequestration (depot effect). Key synthesized analogs demonstrated sub-nanomolar affinity for CB1, marked reduction in hydrophobicity (ClogP∼2.5–3.5 vs 9.09 of Δ8-THC-DMH), and found to function as either agonists (trans-oximes) or neutral antagonists (cis-oximes) in a cAMP functional assay. All oxime analogs showed comparable affinity at the CB2 receptor, but surprisingly they were found to function as inverse agonists for CB2. In behavioral studies (i.e. analgesia, hypothermia) trans-oxime 8a exhibited a predictable fast onset (∼20 min) and short duration of pharmacological action (∼180 min), in contrast to the very prolonged duration of Δ8-THC-DMH (>24 h), thus limiting the potential for severe psychotropic side-effects associated with persistent activation of the CB1 receptor. We have conducted 100 ns molecular dynamic (MD) simulations of CB1 complexes with AM11542 (CB1 agonist) and both trans-8a and cis-8b isomeric oximes. These studies revealed that the C3 alkyl tail of cis-8b orientated within the CB1 binding pocket in a manner that triggered a conformational change that stabilized the CB1 receptor at its inactive-state (antagonistic functional effect). In contrast, the trans-8a isomer’s conformation was coincided with that of the AM11542 CB1 agonist-bound structure, stabilizing the CB1 receptor at the active-state (agonistic functional effect). We have selected oxime trans-8a based on its potency for CB1, and favorable pharmacodynamic profile, such as fast onset and predictable duration of pharmacological action, for evaluation in pre-clinical models of anorexia nervosa.

Introduction

The endocannabinoid system consists of cannabinoid receptors CB1 and CB2,[1], [2] the endogenous lipid ligands arachidonoylethanolamine (anandamide, AEA)3 and 2-arachidonoylglycerol (2-AG),4 as well as their primary hydrolytic enzymes fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL). The cannabinoid receptors belong to the GPCR family and have very different functions and distributions. The CB1 receptors are widely distributed throughout the brain and mediate the majority of the cannabinoid effects on the central nervous system (CNS).5 In contrast, the CB2 receptors are primarily found in cells of immune origin,6 including microglia, although low-level expression has been reported in healthy neurons.7 The limited successes of cannabinoids as medications in humans relates to the severe psychotropic side-effects which are associated with persistent activation of CB1 receptors in the human brain. Δ9-THC and related synthetic cannabinoids are highly lipophilic molecules, which have the tendency to accumulate in fatty tissues (depot effect) and then being released slowly back into circulation and other body compartments, including the brain.8 As a result, compound accumulation in the lipid-rich environment of the nervous system prolongs the elimination phase of the drug, causing a prolonged activation of CB1 receptors in central compartments with precipitous deleterious side-effects.

Recently, a few examples with synthetic CB1 agonists have been explored to limit the sequestration of the drug in fatty tissues by incorporating hydrophilic esters groups on the molecule. Also, the ester functionality could undergo plasma hydrolysis and generate inactive hydrolytic products, thus shortening the compound’s effect.[9], [10], [11] This approach aimed to limit the prolonged activation of CB1 receptors and minimize the CB1-associated psychotropic side-effects.

In this paper, we have explored two important physicochemical properties hydrophobicity (log P) and topological surface area (tPSA) in drug action, aiming to lower the lipophilic nature of the administered drug and decrease its partition rate in areas rich in fat, especially adipose tissues. We have anticipated our approach to generate new cannabinoids with predictable pharmacokinetics and pharmacodynamic effects, such as onset and duration of drug action. To that end, we have introduced polar moieties at the C3 alkyl tail of Δ8-THC. We have selected the Δ8-THC isomer for our studies based on its superior chemical and thermal stability than Δ9-THC, the principal psychoactive constituent of cannabis. First, we have introduced an “internal” polar oxime moiety at the C3 alkyl tail of Δ8-THC to examine its effect on binding affinity for the cannabinoid receptors CB1 and CB2, and secondly, we have additionally incorporated polar groups at the terminal carbon of the alkyl tail (Fig. 1) to further lower the lipophilic character of the molecule. This design strategy offered the potential to influence the mode of drug action (absorption, distribution and sequestration) and lead to an estimated drug pharmacodynamic effect, comprising of fast onset and projected duration of drug action. We have evaluated selected CB1 agonists in biochemical and in vivo assays, as well as pharmacokinetics studies to assess potency, selectivity and druggabilty.

Section snippets

Chemistry

The reported oximes in this paper were prepared according to synthetic methods A and B.

Results and discussion

All synthesized compounds were tested for their ability to bind to CB1 and CB2 receptors using rat brain17 or HEK293 cell membranes expressing mouse CB2 (mCB2) or human CB2 (hCB2),[18], [19], [20] respectively, via competition-equilibrium binding using [3H]CP-55,940.[19], [20], [21] We used both species (human and mouse) of the CB2 receptors to assess potential affinity differences, which have been previously observed by similar structural chemotypes.22 Our structure-activity relationship (SAR)

Conclusions

In this paper, we have described the synthesis of new short-acting CB1 agonists based on the tricycle-core structure of Δ8-THC in an attempt to “drastically” reduce its lipophilic character, and restrict its tissue sequestration (depot effect), which contributes to a prolonged activation of the CB1 receptor with precipitous psychotropic effects. To that end, we have explored new analogs at the C3-alkyl tail of THC by introducing an “internal” polar oxime and polar groups at the terminal carbon

Funding

This work was supported by The Klarman Family Foundation, Boston, MA [grant number KFF # 3237].

References (29)

  • R. Mechoulam et al.

    Biochem Pharmacol

    (1995)
  • T.W. Klein et al.

    Immunol Today

    (1998)
  • J. Fernandez-Ruiz et al.

    Trends Pharmacol Sci

    (2007)
  • A. Weickgenannt et al.

    Tetrahedron

    (2012)
  • P.R. Dodd et al.

    Brain Res

    (1981)
  • K.L. Morse et al.

    Life Sci

    (1995)
  • J.Y. Shim et al.

    Methods Enzymol

    (2013)
  • T. Hua et al.

    Cell

    (2016)
  • W.A. Devane et al.

    Mol Pharmacol

    (1988)
  • S. Munro et al.

    Nature

    (1993)
  • W.A. Devane et al.

    Science

    (1992)
  • M. Herkenham et al.

    J Neurosci

    (1991)
  • P. Sharma et al.

    Iran J Psychiatry

    (2012)
  • S. Kulkarni et al.

    J Med Chem

    (2016)
  • Cited by (9)

    • Improved cyclobutyl nabilone analogs as potent CB1 receptor agonists

      2022, European Journal of Medicinal Chemistry
      Citation Excerpt :

      Additionally, its physicochemical properties (lower clog and higher tPSA) suggest a more favorable pharmacokinetic profile than its precursors. To this end, AM8936 shows a trend for faster onset of action and peak effects in vivo, which may be attributed to a lower depot distribution, a phenomenon we have observed with CB1 agonists having more polar characteristics [28,29]. Such properties will render AM8936 a valuable tool for further in vivo studies of CB1 and a lead analog suitable for further optimization.

    View all citing articles on Scopus
    View full text