In a well-publicized article in the February 23, 2005 issue of the Journal of Neuroscience, researchers from Spain describe changes in cannabinoid receptors in the brains of AD patients, as well as animal behavioral and in-vitro data suggesting that cannabinoid agonists can protect neurons by reducing microglial activation.

The headlines on this study were predictably tantalizing (see, for example, "Marijuana May Block Alzheimer's"), but also misleading. The mass media stories barely dug beyond the paper’s stated implication that marijuana’s active ingredients could stem the progression of neurodegeneration. In reality, however, the story is more complicated and still unfolding. To begin with, two cannabinoid receptors have been identified. CB1 is the major type in brain, expressed by all types of nervous system cells, and apparently responsible for the psychoactive effects of the drug. CB2 is mainly expressed in immune system cells, but also in microglia (Benito et al., 2003), and may mediate neuroprotective effects of the drug. To complicate matters, cannabinoids may interact with other, still uncharacterized receptors.

There are many reports, in vivo as well as in vitro, of cannabinoids protecting neurons in models of excitotoxicity, ischemia, or trauma (see ARF related news story). This protection might result from effects on glutamate transmission or calcium flux. However, Pat McGeer and his colleagues at the University of British Columbia, Vancouver, have reported that agonists with affinity for both receptors are neurotoxic, a characteristic apparently mediated by CB1, whereas CB2-selective agonists can prevent microglia-mediated cell toxicity (Kegeris et al., 2003). Moreover, a group led by Julian Romero of the Fundacion Hospital Alcorcon in Madrid has reported increased expression of CB2 in AD brain, specifically in plaque-associated microglia, with no evidence of changes in CB1 levels (Benito et al., 2003). These researchers also reported finding both receptors in amyloid plaques.

The current paper, by María de Ceballos and colleagues at the Cajal Institute and Complutense University in Madrid, expands on these findings. In contrast to the study from Romero's group, first author Belén Ramírez and colleagues report a reduction in CB1 in AD brain. Specifically, neurons in the vicinity of activated microglia and CB1-containing plaques did not express CB1 in four of six AD cases. The authors also report that four of six AD cases showed a CB2-signal in dystrophic neurites and in neurons with evidence of tangles. Both receptors showed increased nitration in AD brain relative to control.

In one set of experiments, Ramírez and colleagues studied the interactions of cannabinoids and acutely administered Aβ in rats. In behavioral assessments of spatial learning, a nonselective CB1/CB2 agonist protected rats against Aβ-induced deficits. This cannabinoid also prevented Aβ-induced microglial activation in vivo, and counteracted Aβ-induced reductions in the neuronal proteins calbindin and α-tubulin, the authors report. In another set of experiments, this time in vitro, cannabinoid agonists prevented Aβ-induced microglial activation, regardless of whether they were CB2-specific or not, the authors write. Both agonists are reported to have blocked neurotoxicity induced by these activated microglia.

The authors note that in their microglia-neuron cell culture experiments, a CB2 receptor agonist was just as effective in protecting neurons as one that targets both cannabinoid receptors. They suggest that pursuing specifically the CB2 receptor in therapeutic strategies, rather than mixed-activity agonists, might avoid the psychoactive effects mediated by the CB1 receptor.

In contrast to this, McGeer and colleagues did not find CB1/CB2 agonists to be protective, but rather the opposite (Kegeris et al., 2003). "CB1 receptor agonists like THC are toxic to neurons and it would be a mistake to leave an impression, however indirect, that there might be some benefit to cannabis in AD," McGeer told ARF. "CB2 receptor agonists are clearly antiinflammatory, but if they also stimulate CB1 receptors, as do the ingredients of cannabis, the benefit will be offset by toxicity to neurons."—Hakon Heimer


  1. This is an excellent paper and it markedly focuses on the importance of both reactive gliosis and the cannabinoid receptor involvement in Alzheimer disease. Targeting reactive gliosis may represent a new, promising approach to inhibit progression of Alzheimer disease (AD). It should be interesting to see in the future the effect of β amyloid on reactive gliosis and relative CB2 expression in specific hippocampal areas. Moreover, a possible cross-talk between specific CA1, CA2, and CA3 neurons with reactive microglia needs further investigation.

  2. This is probably the best paper on the subject I have seen yet. If CB1 receptors in the brain could be utilized in programmed cell death we could have a brilliant first step in finding a cure for AD. And since antibodies and complement proteins are involved in response to amyloid, CB2 receptors could be manipulated to downregulate cytokines MHC, HLA, and MAC (major histocompatibility molecule, human leuokocyte antigen, membrane attack complex, respectively).

    Furthermore, this could also provide a new way of changing expression of protein kinases, phosphatases, ER response to stress. THC could prove very useful in preserving, returning, and even increasing neuronal functions and thus memory and functioning in society. On a final note, glia have been found themselves to be imperative to learning and transmitting messages to neurons. Superb paper!

  3. Considering data from Ramírez et al. (2005), we can effectively conclude that the endocannabinoid system may be a promising therapeutic target in Alzheimer disease (AD). This report shows the first functional evidence on neuroprotective effects of CB1 and CB2 agonists in in vivo and in vitro models of AD. Further, these authors have found that both CB1 and CB2 agonists are capable of preventing amyloid-β-induced microglial activation and improving behavioral performance in a rat model of AD. Although difficult to connect with results obtained in rats, data from human samples showed a decrease in CB1 functional coupling. The authors also indicate that CB1 protein seems to be decreased in neuronal elements located on the vicinity of BA plaques.

    This paper contains some discrepancies with previous data reported by us and others. For instance, while we recently reported CB2 expression in microglial cells in amyloid-β plaques (Benito et al., 2003), Ramírez et al. show only neuronal staining for these receptors. Further, these new data add more controversy to the precise role(s) that CB1 and CB2 receptors may play in pathological conditions. Specifically, the pro- or antiinflammatory effect of CB2 activation is a hot issue. Cece Hillard´s (Carrier et al., 2004) and Nephi Stella´s (Walter et al., 2003) groups have reported increased in vitro proliferation and migration of microglia, respectively, while Klegeris et al. (2003) have postulated CB2 receptors as mediators of antiinflammatory events.

    One important question that should be addressed is the role of endogenous cannabinoids in AD. As we suggested in a previous paper (Pazos et al., 2004), these arachidonic acid-related compounds may have a double-sided role as, together with their recognized antiinflammatory properties, they may also be easily metabolized to arachidonic acid derivatives, well-known for their proinflammatory actions.

    Researchers working on cannabinoids are used to the amplification of experimental results by mass media. In fact, sometimes we all get benefits from it. Other times, however, misleading information may result. I agree this may be the case for a sentence like, “Marijuana may block Alzheimer´s.”


    . Prevention of Alzheimer's disease pathology by cannabinoids: neuroprotection mediated by blockade of microglial activation. J Neurosci. 2005 Feb 23;25(8):1904-13. PubMed.

    . Cannabinoid CB2 receptors and fatty acid amide hydrolase are selectively overexpressed in neuritic plaque-associated glia in Alzheimer's disease brains. J Neurosci. 2003 Dec 3;23(35):11136-41. PubMed.

    . Cultured rat microglial cells synthesize the endocannabinoid 2-arachidonylglycerol, which increases proliferation via a CB2 receptor-dependent mechanism. Mol Pharmacol. 2004 Apr;65(4):999-1007. PubMed.

    . Nonpsychotropic cannabinoid receptors regulate microglial cell migration. J Neurosci. 2003 Feb 15;23(4):1398-405. PubMed.

    . Reduction of human monocytic cell neurotoxicity and cytokine secretion by ligands of the cannabinoid-type CB2 receptor. Br J Pharmacol. 2003 Jun;139(4):775-86. PubMed.

    . Role of the endocannabinoid system in Alzheimer's disease: new perspectives. Life Sci. 2004 Sep 3;75(16):1907-15. PubMed.

  4. Comment by Pat McGeer and Andis Klegeris
    Ramirez et al., in their paper on prevention of Alzheimer disease pathology by cannabinoids, concluded that “cannabinoids succeed in preventing the neurodegenerative process occurring in the disease.” This conclusion is open to question. It is based on a series of experiments demonstrating the antiinflammatory effects of stimulating CB2 receptors. However, the authors did not investigate the effects of stimulating selectively CB1 receptors.

    Cannabinoids such as Δ-9 THC stimulate both CB1 and CB2 receptors. CB1 agonists are toxic to several types of neuronal cells in vitro (Blevins and Regan, 1976; Chan et al., 1998; Klegeris et al., 2003; Lew, 1996). In vivo data show that Δ-9 THC can cause neuronal death after prolonged exposure periods (Scallet, 1991). Heavy use of cannabis is also known to have deleterious effects on cognition and memory (Pope and Yurgelun-Todd, 1996; Solowij et al., 2002), despite some reports of the neuroprotective effects of cannabinoids (for review see Guzman et al., 2001).

    Ramirez et al. reported a reduction of CB1-expressing neurons in AD, which would be consistent with the previous data on the deleterious effects of CB1 stimulation. Destruction of neurons in AD could be enhanced by endogenous CB1 agonists such as anandamide.

    Ligands selective for CB2, such as JWH 015 and indomethacin morpholinylamide are a safer class to pursue in search of antiinflammatory agents for the treatment of AD. CB2 is strongly expressed by immune system cells (Cabral and Dove Pettit, 1998; Klein et al., 1998), including human and rodent microglia (Klegeris et al., 2003; Walter et al., 2003), but it is not expressed by human neuroblastoma SH SY5Y cells (Klegeris et al., 2003). Agonists of CB2, but not CB1, prevent activated microglia from secreting neurotoxic materials into culture supernatant (Klegeris et al., 2003), reinforcing previous reports of their antiinflammatory activity.

    In summary, cannabis ingredients stimulate both CB1 and CB2, so they have the potential to damage neurons directly, even though they may also protect them indirectly. The best approach is to concentrate on selective CB2 agonists.


    . delta-9-Tetrahydrocannabinol: effect on macromolecular synthesis in human and other mammalian cells. Arch Toxicol. 1976 Mar 11;35(2):127-35. PubMed.

    . Drugs and immunity: cannabinoids and their role in decreased resistance to infectious disease. J Neuroimmunol. 1998 Mar 15;83(1-2):116-23. PubMed.

    . Hippocampal neurotoxicity of Delta9-tetrahydrocannabinol. J Neurosci. 1998 Jul 15;18(14):5322-32. PubMed.

    . Control of the cell survival/death decision by cannabinoids. J Mol Med (Berl). 2001;78(11):613-25. PubMed.

    . Reduction of human monocytic cell neurotoxicity and cytokine secretion by ligands of the cannabinoid-type CB2 receptor. Br J Pharmacol. 2003 Jun;139(4):775-86. PubMed.

    . Cannabinoid receptors and immunity. Immunol Today. 1998 Aug;19(8):373-81. PubMed.

    . Tau protein after delta-9-tetrahydrocannabinol in a human neuroblastoma cell line. Gen Pharmacol. 1996 Oct;27(7):1141-3. PubMed.

    . The residual cognitive effects of heavy marijuana use in college students. JAMA. 1996 Feb 21;275(7):521-7. PubMed.

    . Neurotoxicology of cannabis and THC: a review of chronic exposure studies in animals. Pharmacol Biochem Behav. 1991 Nov;40(3):671-6. PubMed.

    . Cognitive functioning of long-term heavy cannabis users seeking treatment. JAMA. 2002 Mar 6;287(9):1123-31. PubMed.

    . Nonpsychotropic cannabinoid receptors regulate microglial cell migration. J Neurosci. 2003 Feb 15;23(4):1398-405. PubMed.

  5. Pat McGeer makes an interesting point, but further review of the research shows that agonists of CB1 receptors can in fact be neuroprotective. It is quite possible to manipulate CB1 receptors to induce apoptosis in affected neurons (i.e. those with amyloid deposits). In vivo experiments have suggested possible therapeutic treatments utilizing CB1 receptors.

    Some experiments point to damge caused by overstimulation of CB2 receptors, as well, but more research is needed to properly utilize agonist/antagonist signals to manipulate the immune response in AD patients. Check out some of the research coming out of Canada as early as 1998. Fatty acid cannabinoid-like brain chemicals show much promise in helping treat AD, slowing progression, and possibly devloping new preventive measures.

  6. This is a very well-designed study that adds up to the accumulating evidence on cannabinoid-mediated neuroprotection. It shows many aspects of their beneficial effects, in experimental models (both in-vitro and in-vivo), and highlights the effect of cannabinoids on activated microglia, namely, on inflammatory processes. The experimental findings are supported by findings from clinical material of AD patients, and the authors propose that attenuation of long-lasting inflammatory reaction by cannabinoids may prevent neurodegenerative processes.

    Studies from our own (Panikashvili et al. 2001; 2005) and many other laboratories well agree with this concept. Yet, in contrast to the large body of evidence suggesting a neuroprotective role of cannabinoids, another paper should be cited that, like Pat McGeer’s group, reports the toxic effect of anandamide {“The dark side of endocannabinoids” : A neurotoxic role for anandamide, (Cernak et al., 2004)). I am not sure yet how to reconcile between these conflicting data.

    Our group is interested in the pathophysiology of traumatic brain injury (TBI) and we study the role of cannabinoids, both endogenous and exogenous, in closed head injury, a mouse model of TBI. We have reported that the levels of the endocannabinoid 2-arachidonoylglycerol (2-AG) increased 10-fold within 4 hours after injury in mice, and that when injected after injury, the synthetic 2-AG improved outcome (Panikashvili et al., 2001). We extended our studies and showed that this effect is indeed CB1 receptor mediated by using CB1 knockout (KO) mice. The recovery of the KO mice was slower than that of the wild-types, and they did not respond to treatment with exogenous 2-AG (Panikashvili et al. 2005 ). The mechanism proposed in our recent study (Panikashvili et al. 2005) is that 2-AG inhibits the activation (nuclear translocation) of the transcription factor NF-kappaB, which is a “master” regulator of the inflammatory response. NF-kappaB has been shown to be involved in brain damage and to promote neuronal cell death in in-vitro and in in-vivo models of ischemic and neurodegenerative neurological diseases. Thus, the inhibition of TNFαrelease reported by Ramirez et al. is well explained by the inhibition of NF-kappaB, as shown for 2-AG (our own findings) and HU-211 (dexanabinol) by Juttler et al. (2004). However, the latter authors show that inhibition of NF-kappaB by cannabinoids is not mediated via the CB1 or CB2 receptors. In addition, we found (Panikashvili et al., abstract presented at 2004 Neuroscience meeting, and full paper submitted for publication) that 2-AG inhibited the acute (1-4 h) post traumatic expression of the main proinflammatory cytokines: TNF-α, IL-1β and IL-6.

    In conclusion, the study by Ramirez et al is an important contribution to the field of cannabinoids as neuroprotective agents, and, supported by others, strongly favors its anti-inflammatory effects as one of its mechanism of action.


    . The "dark side" of endocannabinoids: a neurotoxic role for anandamide. J Cereb Blood Flow Metab. 2004 May;24(5):564-78. PubMed.

    . The cannabinoid dexanabinol is an inhibitor of the nuclear factor-kappa B (NF-kappa B). Neuropharmacology. 2004 Sep;47(4):580-92. PubMed.

    . An endogenous cannabinoid (2-AG) is neuroprotective after brain injury. Nature. 2001 Oct 4;413(6855):527-31. PubMed.

    . CB1 cannabinoid receptors are involved in neuroprotection via NF-kappa B inhibition. J Cereb Blood Flow Metab. 2005 Apr;25(4):477-84. 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

Paper Citations

  1. . Cannabinoid CB2 receptors and fatty acid amide hydrolase are selectively overexpressed in neuritic plaque-associated glia in Alzheimer's disease brains. J Neurosci. 2003 Dec 3;23(35):11136-41. PubMed.
  2. . Reduction of human monocytic cell neurotoxicity and cytokine secretion by ligands of the cannabinoid-type CB2 receptor. Br J Pharmacol. 2003 Jun;139(4):775-86. PubMed.

External Citations

  1. "Marijuana May Block Alzheimer's"

Further Reading


  1. . Involvement of brain-derived neurotrophic factor in cannabinoid receptor-dependent protection against excitotoxicity. Eur J Neurosci. 2004 Apr;19(7):1691-8. PubMed.
  2. . CB1 cannabinoid receptors and on-demand defense against excitotoxicity. Science. 2003 Oct 3;302(5642):84-8. PubMed.

Primary Papers

  1. . Prevention of Alzheimer's disease pathology by cannabinoids: neuroprotection mediated by blockade of microglial activation. J Neurosci. 2005 Feb 23;25(8):1904-13. PubMed.