A new study using male mice has found that stopping the SSRI antidepressant paroxetine triggers a brief surge in neural activity within two brain regions linked to anxiety, with anxiety-like behaviour peaking two days after discontinuation. Researchers also observed that the heightened activity in serotonin-producing neurons of the dorsal raphe nucleus persisted for several days, even as the animals’ anxiety-like behaviour subsided by day five.
Selective serotonin reuptake inhibitors are widely prescribed for depression and some anxiety disorders, working by increasing serotonin availability in the brain. The researchers emphasised that abrupt SSRI withdrawal can produce withdrawal-like symptoms in humans, including heightened anxiety, sleep disturbances and sensory experiences, and that animal studies have shown anxiety-like behavioural changes upon discontinuation.
The Oxford University team led by Helen M. Collins examined how stopping paroxetine affects c-Fos immunoreactivity, a marker of recent neuronal activity, across several anxiety-related brain regions. c-Fos levels rise when neurons become active, allowing scientists to map which circuits are engaged after a treatment change.
The study involved seven-week-old male mice housed in groups of three littermates. Mitting designs comprised three groups of 12 mice each: two groups received daily paroxetine injections (10 mg/kg) for 12 days, while a saline-injected control group received no active drug. After this period, the saline group and one paroxetine group continued treatment for either two or five more days, whereas the third group was switched from paroxetine to saline, constituting the discontinuation group.
All animals were tested in an elevated plus-maze, a standard assay for anxiety-like behaviour, ninety minutes later. A subset of six or seven mice per group were then euthanised and their brains analysed for c-Fos immunoreactivity to assess recent neuronal activity.
Two days after discontinuation, when anxiety-like behaviour was elevated, neural activity increased in the dorsal raphe nucleus and the ventral hippocampus compared with saline-treated controls. No similar increases were found in other examined regions such as the median raphe nucleus, prefrontal cortex, amygdala, periaqueductal grey, or the paraventricular nucleus of the hypothalamus.
Within the dorsal raphe nucleus, the heightened activity occurred in serotonin-producing neurons, including a subgroup that co-produces serotonin and glutamate. The authors propose that stopping an SSRI may abruptly release serotonin neurons from the inhibition experienced during ongoing treatment, producing a rebound in activity.
Five days after discontinuation, serotonin neurons in the dorsal raphe remained more active, but the animals no longer displayed increased anxiety-like behaviour. Activity in the ventral hippocampus and in the serotonin-glutamate neurons had also returned to baseline. The authors concluded that activation of serotonin neurons alone may not suffice to generate anxiety, while ventral hippocampus activity tracked with the observed anxiety-like behaviour.
“These findings suggest that, under the current conditions, paroxetine discontinuation is associated with a short-lasting activation of anxiety-promoting circuitry limited to DRN 5-HT neurons [dorsal raphe nucleus neurons that produce serotonin] and the hippocampus. This circuitry may contribute to symptoms such as anxiety that are a feature of SSRI discontinuation syndrome,” the study authors concluded.
The study advances understanding of brain changes following SSRI discontinuation, though the authors stressed it was conducted in mice and only in males. The small group sizes—six or seven mice per group—may have limited detection of subtler changes elsewhere, and c-Fos offers only a snapshot of recent neural activity without proving causation.
The paper, “Increased c-Fos immunoreactivity in anxiety-related brain regions following paroxetine discontinuation,” was authored by Helen M. Collins, L. Sophie Gullino, Cara Fuller, Raquel Pinacho, David M. Bannerman and Trevor Sharp.
