Could a puff a day keep memory loss at bay? Perhaps, suggests new work from Andreas Zimmer and colleagues at Bonn University in Germany. The researchers treated old mice with a low dose of Δ9-tetrahydrocannabinal (THC), the main psychoactive component in cannabis, and found it improved the animals’ ability to learn and remember. The investigators went on to show that THC—at levels far below those needed to get high—appeared to rewire and rejuvenate aging brains. Treatment boosted the expression of genes related to synaptic function via epigenetic mechanisms and restored synapse number in the hippocampus to levels seen in much younger animals. The results, which appeared May 8 in Nature Medicine, suggest that THC, or elements in the cannabinoid system that it targets, could present new therapeutic opportunities to slow or prevent age-related cognitive decline.

“This important study confirms something that has been suspected and talked about for while—that THC use later in life seems to protect brain function,” said cannabinoid researcher Daniele Piomelli, University of California, Irvine. Others thought the work was interesting but were puzzled that THC would have opposite effects on gene expression in old and young mice.

THC binds the cannabinoid receptors CB1 and CB2, whose physiological ligands (endocannabinoids) modulate neurotransmission, inflammation, and aging in the brain. Endocannabinoid signaling declines in older rodents, so for the study, co-first authors Andras Bilkei-Gorzo and Onder Albayram asked what would happen if they gave older mice a gentle boost. They implanted osmotic minipumps filled with THC or placebo under the skin of 2-, 12-, or 18-month old animals. The pumps delivered 3 mg/kg/day, which the authors expected to be too low to create a mouse “high.” After 28 days of treatment and a five-day washout, the researchers put the mice through their paces, testing memory in water maze, novel object, and partner-recognition tests.

As expected, untreated older mice learned the tasks more slowly and forgot them sooner than did younger mice. THC reversed those losses, making the old mice perform nearly as well as young controls in the water maze, and just as well in the recognition tests. For young mice, THC had the opposite effect, reducing their ability to navigate the water maze. This is in keeping with a wealth of data that THC suppresses learning and memory in adolescents and young adults.

Along with better test scores, THC appeared to rejuvenate the old mice’s brains by some measures. Treatment reversed the synapse loss that normally occurs with age, and changed global gene expression to a profile resembling that of young mice. “It’s like the brain goes back to the good old days, when the animals were younger and the cannabinoid system was higher in activity,” said Albayram, who now works at Harvard Medical School. Once again, they found THC had the opposite effect in young mice, inducing gene expression patterns typical of aged animals.

Judging from the known functions of THC-regulated genes, the drug appeared to impact cell plasticity and signaling in older mice. Levels of anti-aging, procognition, and neuroprotective transcripts, including klotho, transthyretin, and BDNF, dramatically increased.

When the researchers looked for signaling pathways that might underlie those expression changes, they found evidence for activation of histone acetylases, pointing to epigenetic regulation by THC. In old mice, THC treatment increased histone H3 and H4 acetylation generally, and also specifically at the klotho and BDNF promoters. Increased acetylation appears necessary for the effects of THC on both gene expression and behavior, because treating mice with the histone acetylase inhibitor anacardic acid negated THCs memory-enhancing effects. Finally, the investigators showed that THCs effects were mediated by the CB1 receptor, because mice lacking CB1 showed no improvement.

“This interesting work demonstrates that THC and cannabinoid receptor activation transduce signals all the way to chromatin and synaptic gene expression,” said Li-Huei Tsai of MIT. The results jibe with work from her and other labs indicating that blocking histone deacetylases (HDACs), particularly HDAC2, can enhance memory and cognition (see May 2009 newsMay 2010 news). The new work supports the notion that HDAC inhibition will be beneficial, Tsai said, adding that she’d like to see a direct comparison of the effects of THC and an HDAC2 inhibitor in the mice.

One roadblock to clinical use of HDAC inhibitors has been a lack of specificity, leading to side effects. Could THC provide a safer and more accessible way to modulate HDACs? Possibly, says Bilkei-Gorzo, because THC’s action will be restricted to neurons that have CB receptors, whereas HDAC inhibition potentially affects all cells. At same time, he believes that HDAC modulation only partly mediates THC’s actions. “It’s crucial, but it doesn't explain the changes in every gene,” he said.

While working with Lennart Mucke at the Gladstone Institute in San Francisco, Dena Dubal reported that klotho, a protein with profound anti-aging functions, improved cognition by shoring up synapses (see May 2014 news). “It’s interesting that klotho would be elevated [by THC],” said Dubal, who now runs a lab and clinic at the University of California, San Francisco. “The work points to the importance of intersecting pathways that contribute to synaptic health: Klotho, BDNF, and organizational changes may dovetail to improve the synaptic health of the aging brain. It will be important to dig deep to understand the relationship between klotho and cognition during aging.”

As a physician, however, Dubal cautions about over-interpreting the results. “It could be easy for people to jump from this to thinking that marijuana is good for the aging brain, but I’d call for pause. The effects need to be studied very carefully. The effects of THC can differ widely by dose and age. As a neurologist, I’ve seen a lot of THC-based pathology in the brain,” Dubal said.

Going forward, Bilkei-Gorzo says his group is planning a clinical trial of low-dose THC in people with mild cognitive impairment. He is also studying THC in mouse models of AD. Previous studies of THC treatment for AD focused on agitation and aggressiveness in people with advanced disease. Bilkei-Gorzo says he wants to look at earlier stages for effects on cognition.—Pat McCaffrey


  1. This is a fascinating study which demonstrates that THC, the psychoactive ingredient of cannabis, restores cognitive function in older mice. The authors also showed that this was mediated by stimulation of cell plasticity and signalling mechanisms in older animals. The cannabinoid receptor 1 (CB1) appears to be important, and the authors suggest that the regular administration of low-dose THC normalized the weak cannabinoid signalling that occurs with aging. The authors also present data to suggest that this reverses age-related epigenetic changes, possibly through the activation of histone acetylase.

    These findings are consistent with previous knowledge.  The cannabinoid system, targeted by cannabis, is a complex network of receptors and molecules that have many functions in the brain, and help reduce neuronal damage and promote repair. Previous studies have shown that THC and cannabidiol (CBD), another alkaloid found in cannabis, reduce Alzheimer’s-type pathology in mouse models of Alzheimer’s disease. It is also known that sirtuin 1 (SIRT1), which removes acetyl groups from many histones, interacts with the cannabinoid receptor. Cannabis is therefore of interest in aging.

    However, there are many unanswered questions. Importantly, this study showed that THC had the opposite effects in young mice, worsening memory and evoking the molecular profile of older mice. There is good evidence that cannabis worsens cognition in humans, particularly young adults who abuse it over a period of time. The reasons for discrepant effects in young and old mice are not known. Are they related to an optimal level of stimulation of CB1 and CB2? If yes, how does one achieve this optimal level and at what age does the transition from deleterious to beneficial effects occur? Is the effect the same in humans? Replication of the findings in mice, and the studies of cannabinoid system in aging humans are necessary before one can consider translating any such findings to intervention studies.

    There has been some work in animal models of Alzheimer’s disease to see if cannabis could slow the progression of the pathology. The results have not been consistent. Understanding the changes in the cannabinoid system with aging may help to clarify the mechanisms. Many other questions remain. What is a low dose? The dose of 3 mg/Kg, when translated into human terms, is not necessarily a low dose. The psychoactive effects of THC may be unacceptable, and long-term effects of such interventions are unclear. It is not known if the same results can be achieved with CBD, which does not have psychoactive properties.

    In short, this report should prompt further work on the cannabinoid system with aging, and its implications for human health, both in normal aging and in age-related diseases. It is too early to regard THC, or cannabis in general, as an elixir of youth.

Make a Comment

To make a comment you must login or register.


News Citations

  1. It’s an HDAC2 Wrap— Memory-suppressing DNA Modifier Identified
  2. Histone Acetylation: Epigenetic Achilles’ Heel of Memory in Aging
  3. Longevity Gene Boosts Brainpower, Even in the Young

Further Reading

No Available Further Reading

Primary Papers

  1. . A chronic low dose of Δ(9)-tetrahydrocannabinol (THC) restores cognitive function in old mice. Nat Med. 2017 Jun;23(6):782-787. Epub 2017 May 8 PubMed.