Drugs that Block Cannabinoid Receptors Seem Problematic


by Jim Schnabel

December 3, 2008

One of marijuana’s best known side effects is hunger, and the discovery of the brain-cell receptor that mediates this effect has led to the development of nearly a dozen drugs meant to block it and thereby treat obesity. However, it is now clear that CB1, the cannabinoid receptor targeted by these therapies, is responsible for much more than “the munchies.”

On Oct. 2, the drug company Merck announced that its CB1-blocking drug taranabant had caused unacceptable side effects in a Phase III trial in obese patients. Merck has stopped development of the drug. The chief side effects of taranabant are anxiety and depression. Merck’s action follows the FDA’s rejection in 2007, essentially for the same reasons, of Sanofi’s CB1-blocking drug rimonabant (brand name Acomplia).

“I was hoping that taranabant would do better than rimonabant, but I am not really surprised that it didn’t,” says Daniele Piomelli, a researcher at the University of California, Irvine, who has been in the forefront of research in this area. “It seems that the endogenous cannabinoid system is an important regulator of the stress-coping mechanism engaged in anxiety and depression.”

The CB1 receptor

Aside from being marijuana’s prime target, CB1 is one of the most widely expressed receptors on cells throughout the body, and it is particularly abundant in the brain. The receptor appears to have somewhat different functions in each organ system. In the brain, the binding of CB1 by its corresponding neurotransmitters, or “ligands,” also has multiple effects.

On the whole, however, CB1’s job seems to be to modulate the activity of other systems in the brain. Typically CB1 receptors stud the terminals of a neuronal output stalk, or axon, which delivers signals to another neuron across a synapse. Greater activity in the synapse tends to produce a buildup of the brain’s own cannabinoids, which bind to CB1, which in turn triggers a reduction in the flow of signals across the synapse.

Some evidence suggests, for example, that this braking or calming effect of the CB1 system is necessary to prevent the uncontrolled neuronal activity known as epilepsy. Lab tests with neuronal cultures made prone to seizure-like activity have shown that the blockade of CB1 leads to continuous firing, which stops when the blockade is lifted.

As evidence from marijuana users suggests, CB1 activation also tends to counteract the brain systems that produce anxiety, depression and pain. In one experiment in 2002, mice bred without the gene for CB1 and conditioned to associate a certain sound with pain were significantly less able to forget the fearful association than were normal mice. “Mutant mice lacking CB1 receptors are known to break down more easily when stressed, while drugs that boost endogenous cannabinoid activity strengthen the ability of animals to cope with stress,” Piomelli says.

The recent clinical trials of rimonabant and taranabant, both of which bind to CB1 in a way that reduces its activity, amount to large-scale tests of CB1’s function in humans. Both sets of trials indicate that blocking CB1 activity in humans produces harmful effects like those seen in animal experiments.

When an FDA panel unanimously rejected Sanofi’s application for rimonabant in 2007, it cited sharply increased rates of anxiety, depression, suicidal thoughts and mood disorders in patients taking the drug.

Although rimonabant was approved by the European Medicines Agency in 2006, a meta-analysis of clinical trial data, published in the Lancet in 2007, determined that people taking the drug, as opposed to a placebo, were two to three times more likely to stop taking it because of depression or anxiety. This was despite the fact that people reporting a depressed mood were not allowed into the trials.

“They have to exclude individuals with mood disorders in such studies because they don’t want the liability issues, and it enables a cleaner interpretation of the trial data,” explains Richard Deyo, a pharmacologist and cannabinoid expert at Winona State University in Minnesota. But to understand the full potential impact of side effects, he adds, “you’d have to look at the whole population.”

Better to boost CB1 than to block it?

Several other drug companies have CB1 blockers in early clinical trials, though the failure of rimonabant and taranabant casts a cloud over their further development.

Other academic and commercial researchers have sought not to block CB1 activity but to boost it, to relieve depression and anxiety. But this strategy, too, is fraught with complications.

The most obvious problem is that the CB1-binding molecule known as THC is potentially addictive, as are other cannabinoid molecules that directly bind and activate CB1. CB1 also appears to have different effects depending on which system it is modulating.

CB1 receptors are found on two broad classes of nerve terminals: excitatory, which increase the activity of target neurons; and inhibitory, which reduce activity, Piomelli says. CB1 therefore can either “put on the brakes” or “release the brakes” depending on which of these systems it modulates.

Any strategy that floods the brain with CB1-binding molecules will thus have a host of effects, depending on the systems affected, the dose that reaches each system, and the ongoing level of activity in those systems—which in turn can vary according to mood, personality and genetic differences in brain chemistry.

For example, at low to moderate levels, THC, which is the main active ingredient in marijuana, appears to have reliable pain- and anxiety-relieving effects, Deyo says. “As you increase the dose, though, you get memory blocking effects, you get a tendency towards depression and you also get the potential for hallucinations,” he adds. Furthermore, THC’s effects appear to vary according to gender.

The ‘make more of it where it’s needed’ strategy

To get past such obstacles, Piomelli and his colleagues at UC-Irvine recently devised an indirect CB1-activity boosting strategy, similar to that used for some other brain-cell receptors. The body’s main CB1-activating molecule, anandamide, is normally broken down quickly by enzymes near the synapses where it works. Piomelli reasoned that by preventing this continual degradation of anandamide, he could keep it more of it around synapses and thus maintain a higher level of CB1 activity. In this way he could also increase CB1 activity primarily where it was already occurring and needed, and to a lesser extent in other areas of the brain and body where the boosting of such activity could produce undesired side effects.

In a paper published in the journal Biological Psychiatry on Sept. 22, Piomelli and colleagues, together with a National Institutes of Health team led by Stephen Goldberg, reported on a set of experiments using this strategy in monkeys. Administering a drug that inhibits FAAH, an anandamide-eating enzyme, the researchers soon were able to measure higher levels of anandamide in the monkeys’ brains.

Using standard tests, they found that this approach did not reinforce other drug-taking behavior, as THC normally does. “Our results,” the researchers wrote, “suggest that FAAH inhibitors might be used therapeutically without risk of abuse or triggering of relapse to drug abuse.”

Piomelli explains that when the FAAH inhibitor, currently named URB597, is administered, anandamide levels rise and appear to reduce inhibitory controls on a midbrain region where neurons pump out the neurotransmitters serotonin and noradrenaline. The effect is to boost the levels of these neurotransmitters in the brain.

“This physiological effect is similar to that produced by certain antidepressant drugs,” says Piomelli, “and indeed URB597 displays profound antidepressant-like properties in rats and mice.”

Organon, a unit of pharmaceutical company Schering Plough, holds the rights to URB597 and is considering clinical trials of the drug for pain relief.