Chu and others have shown that the endocannabinoid 2-arachidonoylglycerol (2-AG) protects brain neurons from inflammation and Aβ-induced cell death (see Zhang and Chen, 2008; Chen et al., 2011). Breakdown of 2-AG generates arachidonic acid, the precursor to prostaglandins and leukotrienes that mediate neuroinflammation. Other recent work identified monoacylglycerol lipase (MAGL) as the primary enzyme that converts 2-AG into arachidonic acid (Nomura et al., 2011). “Our strategy was to block this enzyme (in AD mice) to increase the neuroprotective and anti-inflammatory effects of endogenous 2-AG,” Chen told Alzforum.

First author Rongqing Chen and colleagues treated 5xFAD mice, an aggressive AD amyloid model, with the MAGL inhibitor JZL184. Developed by scientists at Scripps Research Institute in La Jolla, California, the compound shrank tumor growth in cultured human cancer cells (Nomura et al., 2010) and suppressed neuroinflammation and cell loss in a parkinsonian mouse model. In the current study, Chen’s team injected the MAGL inhibitor into the peritoneum of presymptomatic 5xFAD mice three times a week under a 16-week prevention regimen starting at two months, or an eight-week treatment starting at four months of age. In this AD strain, brain deposits start appearing at two months, plaques form by four months, and cognitive deficits show up between five and six months of age. By immunoblotting and immunohistochemistry, both treatment groups had less brain Aβ (total and Aβ42) than did vehicle-treated controls. Brain levels of BACE1, the enzyme that initiates cleavage of Aβ from amyloid precursor protein (APP), were also down. Fewer neurons were dying in these animals relative to controls, as shown in brain slices stained with the FluoroJade C neurodegenerative marker.

To examine effects on spine density, the scientists crossed 5xFAD mice with a transgenic line expressing neuron-specific green fluorescent protein (GFP). Two-photon imaging of hippocampal CA1 and dentate granule neurons revealed lower spine density in the AD/GFP mice—largely preventable by eight-week JZL184 treatment, the scientists report. AD mice downregulate glutamate AMPA and NMDA receptor subunits; again, the MAGL inhibitor prevented these deficits.

The compound also delivered functional benefits, alleviating problems with neurotransmission and long-term potentiation that ravage the hippocampal CA1 region of AD mice. More importantly, both eight- and 16-week treatments prevented memory deterioration as measured in the Morris water maze.

Interestingly, the MAGL inhibitor boosted cognition and increased spine density even in normal mice. That jibes with a recent study showing enhanced spatial learning and memory in MAGL knockout mice (Pan et al., 2009).

Scientists contacted for comment on this study agreed that the mouse data look promising, but said more work is needed to flesh out molecular details before considering the compound for human testing. Blocking MAGL will likely affect many downstream pathways. “You’re going to take out good pathways and bad pathways at once,” said Katrin Andreasson of Stanford University School of Medicine, Palo Alto, California. “I think you need to get downstream of MAGL to identify which are the toxic prostaglandin signaling pathways. Then you can design more specific approaches with fewer side effects.”

Tibor Harkany of Karolinska Institute, Stockholm, Sweden, expressed concern about side effects. “If you consider that MAGL inhibition would increase brain endocannabinoid levels six- to sevenfold (as per the data from this paper), and endocannabinoids affect synaptic neurotransmission globally, then one unresolved question is whether MAGL inhibition would exert undesired effects on many higher cognitive functions and/or make people drowsy or affect their motor coordination,” Harkany noted in an e-mail to Alzforum. “The cost-benefit relationship of using MAGL inhibitors must carefully be weighed, and in more sophisticated models, before increasing hope for a new treatment.” Persistent cannabis use, for example, is associated with neuropsychological decline (see Meier et al., 2012).

Grace Sun of the University of Missouri, Columbia, extended the specificity argument in the other direction. She suggested that MAGL inhibition may be insufficient for mopping up the pool of arachidonic acid (AA) that gets broken down into harmful prostaglandins and leukotrienes. Since release of arachidonic acid (AA) “can be mediated by other enzymes, including phospholipases, which are probably more abundant than MAGL, it is not clear how this enzyme is linked specifically to AA-derived prostaglandins in the AD brain,” she suggested.

On the whole, the present paper offers proof of principle that lipid cascades can be pharmaceutically targeted, said Steffany Bennett of the University of Ottawa, Canada. She said the new work supports the idea that tweaking lipid pathways “to promote a ‘normal biology’ may represent a new frontier for AD research.” (See full comment below.)

On the neuroinflammation angle, human data paint a confusing picture. Past clinical trials of non-steroidal anti-inflammatory drugs (NSAIDs) showed no benefit in people with mild to moderate AD (see ARF related news story) or in at-risk seniors on the verge of dementia (see ARF related news story). However, observational studies have suggested that NSAIDs do protect against AD (see Vlad et al., 2008), leading scientists to hold out hope that hitting neuroinflammatory pathways might work as a preventive therapy. Furthermore, Andreasson noted, MAGL inhibitors are different from NSAIDs because “they are a one-two punch.” In addition to diminishing downstream prostaglandin effects, they boost 2-AG.

Chen and colleagues are beginning to unravel the molecular underpinnings behind the synaptic and cognitive improvement tied to MAGL inhibition. The scientists reported that 2-AG may protect neurons via PPAR-γ (Du et al., 2011), and they have National Institutes of Health funding to extend preliminary findings suggesting that epigenetic mechanisms such as noncoding microRNAs are involved in the neuroprotection.

Enhancing 2-AG signaling may eventually turn out to be helpful for other brain disorders as well. In a recent study, treatment with another MAGL inhibitor normalized synaptic defects and corrected behavioral abnormalities in a mouse model of fragile X syndrome, the most common genetic cause of autism (Jung et al., 2012).—Esther Landhuis.

Reference:
Chen R, Zhang J, Wu Y, Wang D, Feng G, Tang YP, Teng Z, Chen C. Monoacylglycerol Lipase is a Therapeutic Target for Alzheimer’s Disease. Cell Reports. 31 Oct 2012;2:1-11. Abstract