Vanderbilt University
Vanderbilt Institute of Chemical Biology

home research discoveries core facilities training & research opportunities seminars & events news giving contact


Discoveries Featured

A New Class of Cyclooxygenase Inhibitors with Anti-Anxiety Activity

By: Carol A. Rouzer, VICB Communications
Published: August 7, 2013

Inhibitors that selectively block cyclooxygenase-2-dependent endocannabinoid oxygenation increase endocannabinoid tone and reduce anxiety in mice.

The discovery of the cannabinoid receptors, CB1 and CB2, provided the foundation for our understanding of the pharmacology of (-)-trans9-tetrahydrocannabinol, the major psychoactive ingredient in marijuana. This discovery also led to the identification of endogenous ligands (endocannabinoids), which are responsible for physiological signaling at the cannabinoid receptors. The two primary endocannabinoids, 2-arachidonoylglycerol (2-AG) and arachidonoyl ethanolamide (AEA), are both lipid molecules derived from the tetraunsaturated fatty acid, arachidonic acid (AA) (Figure 1). 2-AG and AEA play multiple roles in physiological and pathophysiological processes, including metabolism, cognition, stress and anxiety, tumor progression, bone remodeling, pain perception, and inflammation. For both AEA and 2-AG, signaling activity is terminated by hydrolysis of the amide and ester bond, respectively, to yield AA (Figure 1). Prior studies have shown that inhibition of hydrolysis can augment and prolong endocannabinoid signaling. Now, Vanderbilt Institute of Chemical Biology member Larry Marnett, and his collaborator, Sachin Patel, report a new way to increase the action of endocannabinoids. Their discovery shows promise as a novel approach for the treatment of anxiety [D. J. Hermanson, et al. (2013) Nat. Neurosci., published online August 4, doi:10.1038/nn.3480].

Figure 1. Structure of arachidonic acid, its glyceryl ester (2-arachidonoylglycerol, 2-AG), and its ethanolamide (arachidonoyl ethanolamide, AEA). 2-AG and AEA are endocannabinoids. Their signaling activity is terminated by hydrolysis, catalyzed by monoacylglycerol lipase (MAGL) and fatty acid amide hydrolase (FAAH), respectively. Endocannabinoid signaling may also be terminated by cyclooxygenase-2- (COX-2)-dependent oxygenation to yield prostaglandin H2-glyceryl ester and prostaglandin H2-ethanolamide, respectively. Further metabolism of these species by other enzymes yields an array of other prostaglandin derivatives.

The cyclooxygenase enzymes, COX-1 and COX-2, catalyze the oxygenation of AA to produce an array of lipid mediators known as prostaglandins (PGs). Research in the Marnett lab has shown that the COX-2 enzyme can also oxygenate 2-AG and AEA, leading to the formation of PG glyceryl esters and PG ethanolamides, respectively. In either case, oxygenation terminates endocannabinoid signaling. The COX enzymes are the targets of the nonsteroidal anti-inflammatory drugs (NSAIDs), such as aspirin, ibuprofen, and naproxen, and COX-2 is the specific target of the newer coxibs, including celecoxib (Celebrex), and rofecoxib (Vioxx). Although both NSAIDs and coxibs work primarily through the blockade of PG synthesis from AA, there are indications that some of their effects are modulated by the cannabinoid receptors. Thus, inhibition of COX-2 may function in part by blocking endocannabinoid degradation and prolonging CB receptor signaling.

All readily available methods for blocking endocannabinoid degradation suffer from lack of specificity. The enzyme primarily responsible for the hydrolysis of AEA is fatty acid amide hydrolase (FAAH), while monoacylglyerol lipase (MAGL) is the major 2-AG hydrolytic enzyme. Both of these enzymes also mediate catabolism of substrates containing fatty acids other than AA, so that inhibition of FAAH blocks the normal hydrolysis of a range of N-acyl ethanolamides (NAEs) and MAGL inhibition broadly impedes monoacylglycerol (MAG) catabolism. Similarly, inhibition of COX-2 blocks the formation of PGs from AA in addition to preventing the oxygenation of the endocannabinoids. These broad pharmacologic activities can lead to unwanted side effects. Indeed, prolonged use of NSAIDS can cause stomach ulcers, and both NSAIDs and coxibs have been associated with cardiac toxicity.

An exciting discovery in the Marnett lab provides a way to suppress endocannabinoid degradation and prolong signaling with a high degree of specificity. They found that certain NSAID derivatives, designated substrate-selective inhibitors (SSIs), potently inhibit COX-2-dependent oxygenation of AEA and 2-AG with little to no effect on AA oxygenation. Experiments using site-directed mutations of COX-2 revealed a key ion pairing/hydrogen bonding interaction required for inhibition of AA but not endocannabinoid oxygenation by some NSAIDs. This finding suggested that tertiary amides of the NSAID indomethacin should have SSI activity. Synthesis of a series of such compounds led to LM-4131, the morpholino amide of indomethacin (Figure 2). LM-4131 inhibited the oxygenation of 2-AG by COX-2 with an IC50 (concentration that causes 50% inhibition) of 622 nM, while having no effect on the oxygenation of AA. LM-4131 also had no effect on the activity of FAAH or MAGL in vitro.


Figure 2.   Structures of indomethacin and LM-4131.


In mice at a dose of 10 mg/kg, LM-4131 significantly increased brain levels of AEA and 2-AG to 139% and 109% of control, respectively. Indomethacin similarly increased both AEA and 2-AG levels, while the selective COX-2 inhibitor NS-398 increased AEA levels only. However, consistent with their pharmacology, both indomethacin and NS-398 produced increases in AA and decreases in PGs, while LM-4131 had no effects on these species. LM-4131 also increased AEA levels in stomach, small intestine, kidney, and lung, but not heart and liver. The compound had no effect on AA or PG levels in these organs.

LM-4131 did not augment brain endocannabinoid levels in COX-2 knockout mice. Wild-type mice treated with LM-4131 did not show broad increases in NAEs or MAGs, which were seen in mice treated with an inhibitor of FAAH or MAGL, respectively. However, a combination of LM-4131 with a FAAH inhibitor produced a additive effect on the level of AEA, while addition of a MAGL inhibitor to LM-4131 had a similar effect on the level of 2-AG. Together these results confirmed that LM-4131’s ability to elevate AEA and 2-AG levels in vivo was due to selective inhibition of endocannabinoid oxygenation by COX-2 and demonstrated that LM-4131 has no direct effect on either FAAH or MAGL.

Prior studies had indicated that elevation of brain AEA levels through use of a FAAH inhibitor reduces anxiety in mice, so the Patel lab investigated the anxiolytic effects of LM-4131. In three separate mouse models of anxiety, LM-4131 was effective, with activity similar to a well-characterized FAAH inhibitor. LM-4131 had no anxiolytic activity in COX-2 knockout mice, confirming that its mechanism of action depended on inhibition of COX-2. The finding that LM-4131 was also ineffective in CB1 knockout mice or in wild-type mice treated with a CB1 antagonist confirmed that its anxiolytic activity required a functioning endocannabinoid signaling system.

Although the anxiolytic activity of LM-4131 is achieved through augmentation of endocannabinoid signaling, the compound did not produce cannabimimetic effects, including hypolocomotion, hypothermia, catalepsy, and antinociception, typically observed with the administration of exogenous cannabinoids. Furthermore, mice treated with high doses of LM-4131 did not suffer gastrointestinal hemorrhage observed at comparable doses of indomethacin.

Together, the results provide strong support for the hypothesis that substrate-selective COX-2 inhibition is an effective and highly specific method for increasing endocannabinoid tone in the brain. The result, at least in mice, is a reduction in anxiety without the strong psychotropic effects observed with administration of exogenous cannabinoids. Based on the physiological actions of endocannabinoids, other beneficial effects, such as reduction of pain and inflammation or suppression of tumorigenesis, may also be achieved with the approach. Furthermore, the fact that SSIs are derived from the familiar and widely used NSAIDs should facilitate their translation to the clinic. Clearly, the discovery of SSIs lays the foundation for an exciting new class of therapeutic agents with a wide range of potential indications.






Vanderbilt University School of Medicine | Vanderbilt University Medical Center | Vanderbilt University | Eskind Biomedical Library

The Vanderbilt Institute of Chemical Biology 896 Preston Building, Nashville, TN 37232-6304 866.303 VICB (8422) fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2013 by Vanderbilt University Medical Center