In Alzheimer disease and other neurodegenerative disorders, chronic activation of glutamatergic neurons is thought to contribute to excitotoxicity and, eventually, neuronal death. But the brain is not defenseless against this onslaught, and a new study reveals the details of the brain’s strategy for tamping down overexcited hippocampal neurons. Working in a mouse seizure model, Beat Lutz and Giovanni Marsicano at the Johannes Gutenberg University, Mainz, Germany, and Max Planck Institute of Psychiatry, Munich, and colleagues show that activation of cannabinoid 1 receptors (CB1) on a subset of glutamatergic hippocampal neurons calms kainic acid-induced excitation and seizures. Their results, published in the August 17 issue of Neuron, provide a mechanism by which endocannabinoids (the endogenous counterparts of the psychoactive component of marijuana) mediate neuroprotection in epilepsy, stroke, and neurodegenerative diseases.

Seizures result from an imbalance of excitatory and inhibitory neuronal activity, and CB1 receptor activation is regarded to be inhibitory and neuroprotective. Previous work from Marsicano and Lutz (Marsicano et al., 2003) showed that kainic acid treatment of mice induced endocannabinoids, which reduced neuron death. CB1 knockout animals experienced more severe seizures in response to kainic acid than normal mice, and more cell damage. However, since the knockout lacked CB1 in many neurons throughout the brain, the target cells for the protective effects were unclear.

In the new study, the trio of first authors, Krisztina Monory and Federico Massa from Mainz and Munich, and Michaela Egertova at the University of London, used a combination of conditional knockouts and kainic acid challenge to show that hippocampal glutamatergic neurons are the focus of CB1’s protective effects.

When they knocked out CB1 only in glutamatergic cortical neurons (including hippocampus, neocortex, and amygdala), mice still experienced stronger seizures compared to their wild-type littermates. In contrast, even though GABAergic neurons (the targets of the anti-seizure medication diazepam) have abundant CB1 receptors, and are also activated during seizures, knocking out CB1 in GABAergic neurons had no effect on seizure intensity. The conclusion—that GABAergic neurons do not participate in physiological control by endocannabinoids during development of KA-induced seizures—left the glutamatergic system as the major target.

But which glutamatergic neurons? By immunohistochemistry, the researchers localized functional CB1 receptors to glutamatergic terminals in the hippocampus, and specifically to the mossy cells’ termini. These cells, which sit with their cell bodies in the hilus of the dentate gyrus, are central to epileptic seizure development in both mice and humans. To determine if the mossy cells were the target of endocannabinoids, the researchers created a highly targeted deletion by injecting an adenovirus expressing the cre recombinase into the hippocampus of CB1 floxed mice. The virally induced deletion of just hippocampal receptors caused a worsening of seizures, proving that CB1 in the hippocampal mossy cells protects against KA-induced seizures.

“Our results clearly show that hippocampal glutamatergic neurons, where CB1 receptors are present at low but detectable levels, are the central mediators of on-demand endocannabinoid-dependent protection against KA-induced acute excitotoxic seizures,” the authors conclude. In an accompanying preview, Bradley Alger of the University of Maryland in Baltimore outlines the pressing issues to be addressed next, including determining whether inhibition of glutamate release explains neuroprotection, and identifying the neuroprotective endocannabinoid and its site of production.

Early work on endocannabinoids showed that they increase after head injury (see ARF related news story) and help reduce neuronal cell death. A recent report (see ARF related news story) showed decreases in CB1 receptors in the brains of AD patients, as well as animal behavioral and in-vitro data suggesting that cannabinoid agonists can protect neurons against amyloid-induced toxicity. In both head injury and AD, the protection came from the anti-inflammatory actions of cannabinoid receptor agonists, and also involved the CB2 receptor, which is present on immune cells and microglia as well as neurons. The new results raise the possibility that CB1 receptor agonists also might act directly on neurons to quiet glutamatergic synaptic activity.

This raises the question of whether endocannabinoids could regulate Aβ production in addition to their anti-inflammatory and anti-excitotoxicity effects. People who have had surgery for intractable seizures have turned out to have had massive amyloid pathology in the removed pieces of hyperexcited brain tissue (MacKenzie and Miller, 1994), presumably because of activity-dependent production of Aβ (Cirrito et al., 2005). The work should not be taken as an endorsement of marijuana to prevent or treat AD, however. The cannabinol from smoking or eating pot floods all areas of the brain, whereas the effects of endocannabinoids are focal and controlled, as elegantly demonstrated by the current study.—Pat McCaffrey


  1. I think it would be worthwhile to look at the brain of older people who have used marijuana chronically and compare it with the brain of age-matched people who never used it. This simple comparison could throw added light on whether cannabis could help people with Alzheimer disease.

  2. The research is very interesting and important. In Los Angeles, California, use of medical cannabis is encountered working in the field with adolescents or young adults. There is indeed controversy since hidden side effects include perceptual and family disorders. Those also need consideration at the psychosocial level.

    At the neuronal level, cannabinoid research also needs to rule out an effect of cannabinoids on reducing prion fibrils (see Colin et al., 1999) or neurogenesis (La Spada, 2005). Neuroendocrine effects, prolactin release, gonadal atrophy, and tumor genesis need attention when studying cannabinoids.


    . Identification of microglial signal transduction pathways mediating a neurotoxic response to amyloidogenic fragments of beta-amyloid and prion proteins. J Neurosci. 1999 Feb 1;19(3):928-39. PubMed.

    . Huntington's disease and neurogenesis: FGF-2 to the rescue?. Proc Natl Acad Sci U S A. 2005 Dec 13;102(50):17889-90. PubMed.

  3. Cannabinoid agonist is shown to have a thermal hyperalgesia effect in inflammatory pain involving the sensory pathways and activation of the calcineurin (Nathaniel et al., 2006). The role of sensory inhibition, anhedonia, and known effects of calcineurin in psychosis also need consideration.


    . Cannabinoid WIN 55,212-2 regulates TRPV1 phosphorylation in sensory neurons. J Biol Chem. 2006 Oct 27;281(43):32879-90. Epub 2006 Sep 5 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Toward a High in Cannabinoid Research: News on Neuroprotective Effect
  2. Cannabinoid Receptors and AD: Searching Beyond the Simple Story

Paper Citations

  1. . CB1 cannabinoid receptors and on-demand defense against excitotoxicity. Science. 2003 Oct 3;302(5642):84-8. PubMed.
  2. . Senile plaques in temporal lobe epilepsy. Acta Neuropathol. 1994;87(5):504-10. PubMed.
  3. . Synaptic activity regulates interstitial fluid amyloid-beta levels in vivo. Neuron. 2005 Dec 22;48(6):913-22. PubMed.

Further Reading


  1. . AMPA receptor downscaling at the onset of Alzheimer's disease pathology in double knockin mice. Proc Natl Acad Sci U S A. 2006 Feb 28;103(9):3410-5. PubMed.
  2. . APP processing and synaptic function. Neuron. 2003 Mar 27;37(6):925-37. PubMed.

Primary Papers

  1. . The endocannabinoid system controls key epileptogenic circuits in the hippocampus. Neuron. 2006 Aug 17;51(4):455-66. PubMed.
  2. . Not too excited? Thank your endocannabinoids. Neuron. 2006 Aug 17;51(4):393-5. PubMed.