A synthetic cannabinoid may trigger anxiety by switching off a brake on the brain’s fear neurons, a mouse study found

Freezing, retreating and steering clear of a fox-urine odor: that was the behavior of mice dosed with a synthetic cannabinoid called CP55940, and it tracked the activity of a small group of somatostatin neurons in the central amygdala, the brain’s fear and stress hub. A Northwestern University team led by psychiatrist Sachin Patel reported in Nature Communications on Oct. 2, 2026 that the drug appears to remove local inhibition from those cells.

Patel chairs psychiatry and behavioral sciences at Northwestern’s Feinberg School of Medicine. Farhana Yasmin is first author, and funding came from the National Institutes of Health (grants MH100785 and K08 MH126166) and the Brain & Behavior Research Foundation’s Young Investigator Awards.

CP55940 and the predator-odor test

Mice received either a placebo or CP55940 before exposure to 2MT, a predator-odor analog. As the Nature Communications paper reports, the drug increased defensive responses in a dose-dependent way, raising both freezing and avoidance of the odor. Northwestern’s own summary describes treated mice freezing more often and spending less time investigating the scent.

Northwestern’s account adds the design details that make the result readable: the team tested several doses, tracked how long each mouse stayed near the scent and whether it froze or fled, and compared the drug with placebo in the same test. The paper’s title frames the task as threat investigation, and the imaging captured how the neurons represented the threat’s location and the animal’s behavior as the mouse approached or backed away.

A miniature microscope implanted in each animal recorded calcium signals from somatostatin, or SOM, neurons in the central amygdala while the mice faced the threat. CP55940 increased the activity of those neurons, though only the highest dose significantly increased the number of active cells. It also changed how the population behaved, increasing antagonistic response ensembles and strengthening the way the cells jointly represented the threat’s location and the animal’s behavior.

Patel gave the practical reading in Northwestern’s release: higher doses and environmental stress worked together to release the brake, an interaction he linked to how “a good trip can turn bad pretty quickly.” The release dates the work to October 2026, and the paper itself went online on Oct. 2.

GABA release and the lifted brake

The brake is inhibitory signaling. In brain-slice experiments, CP55940 reduced both excitatory and inhibitory currents onto SOM neurons, and the authors propose that preferential suppression of GABA release from local inputs lifts inhibition on those cells, which explains their greater activity. The study focuses on CB1 receptors, which are Gi/o-coupled and widely expressed in the brain, and frames the mechanism as presynaptic: less GABA released onto the cell, so less restraint on its firing.

A 2023 Northwestern release described Patel’s group’s earlier work in mice showing that activating cannabinoid receptors increased the activity of somatostatin neurons in the nucleus accumbens, a reward-circuit region, and that stress made those neurons more excitable through endocannabinoid release. The 2026 paper carries the same cell class into the central amygdala and tests a synthetic drug, which brings the work closer to the fear circuitry that anxiety research targets and further from the reward circuit of the earlier mouse study, where the focus was motivation.

Avoidance and freezing

The causal test was selective. Blocking synaptic output from SOM neurons prevented CP55940 from increasing avoidance of the odor, yet it did not block the drug’s effect on freezing. According to the paper, SOM activity is therefore needed for the avoidance response but not for freezing, so the cells explain part of the anxious behavior rather than all of it. Northwestern’s release states more broadly that genetically silencing the neurons made drugged mice avoid the odor less.

Patel suggested that suppressing SOM activity in the central amygdala “could represent a final pathway” for reducing anxiety more broadly, not only the anxiety that follows cannabinoids. The release adds context: U.S. cannabis use and cannabis-related emergency department visits have risen, and that earlier work has linked cannabis use to a higher long-term risk of anxiety.

The National Institute on Drug Abuse lists anxiety, fear, distrust and panic among cannabis’s short-term effects and says they are more common when a person takes a large amount, uses a strong product or has little experience. It adds that THC levels in cannabis products seized by law enforcement rose substantially between 1995 and 2022, and that dispensary flower and concentrates can exceed 40% THC.

The distance from CP55940 in a mouse to a cannabis user is the study’s main limit. The compound is a laboratory-made drug, the syndicated summary of the work reports no human data, and the doses that moved the neurons most were the highest ones tested. Even the circuit result is partial, since silencing SOM output left the drug’s effect on freezing intact, which leaves open what drives that half of the anxious response.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview