Summary: Researchers have identified a serotonin-driven brain circuit that may explain why some patients experience increased fear and anxiety in the early weeks after starting SSRIs.
Source: University of North Carolina
UNC School of Medicine researchers trace an anxiety circuit and use a compound to reduce fearful behavior — a common acute side effect of SSRI antidepressants.
More than 100 million people worldwide use selective serotonin reuptake inhibitors (SSRIs) such as fluoxetine (Prozac) and sertraline (Zoloft) to treat depression, anxiety and related disorders. A perplexing clinical problem is that SSRIs often produce a temporary increase in anxiety during the first few weeks of treatment. That early worsening prompts many patients to stop medication before it can take full effect. Researchers at the University of North Carolina (UNC) School of Medicine have now mapped a serotonin-sensitive circuit that promotes anxiety and fear in mice, shedding light on how SSRIs can produce these short-term adverse effects and suggesting possible strategies to prevent them.
“Our goal is to find a drug that can transiently inhibit this circuit so people don’t have to endure the heightened anxiety when they first start an SSRI,” said senior investigator Thomas L. Kash, PhD, the John Andrews Distinguished Professor of Alcohol Studies in UNC’s department of pharmacology. “More broadly, this work reveals specific brain networks that drive anxious and fearful behavior in mammals.”
Mapping the serotonin-to-anxiety pathway
In a study published in Nature, the UNC team used advanced neuroscience tools — including optogenetics and chemogenetics — to trace a pathway from serotonin-producing neurons in the dorsal raphe nucleus (DRN) to a region of the extended amygdala called the bed nucleus of the stria terminalis (BNST). The researchers first showed that a standard fear stimulus (a mild foot shock) activates DRN serotonin neurons in mice. Those DRN neurons project to the BNST, and activating the DRN→BNST pathway increased anxiety-like behaviors.
Within the BNST, serotonin acted specifically through 5-HT2C (2C) serotonin receptors on a subset of neurons that express the stress-related neuropeptide corticotropin-releasing factor (CRF). When 2C-receptor–expressing BNST neurons are activated by serotonin, they inhibit a different group of BNST neurons that normally send anxiolytic (anxiety-reducing) signals to the ventral tegmental area (VTA) and lateral hypothalamus (LH). Those VTA- and LH-projecting BNST outputs have been shown to promote reward, motivation and reduced anxiety; silencing them therefore allows anxiety and fear to increase.
SSRI exposure and acute anxiety
The team then tested how fluoxetine, a widely used SSRI, affects this circuit. Because SSRIs elevate serotonin levels at their sites of action, fluoxetine heightened the activity of 2C-receptor BNST neurons and strengthened their inhibitory influence on the VTA- and LH-projecting BNST neurons. In behavioral tests, this translated to increased fear- and anxiety-like responses following SSRI exposure — mirroring the early anxiogenic effects often seen in patients beginning SSRI therapy.
Blocking the anxiety signal
Importantly, the anxiety-promoting BNST neurons were found to express CRF. When the researchers applied a compound that blocks CRF signaling, the acute fearful behaviors induced by fluoxetine were markedly reduced. This result indicates that interfering with CRF signaling in the BNST can prevent the short-term rise in anxiety produced by SSRI exposure in mice.
These findings point to a clear mechanistic sequence: DRN serotonin engages BNST 2C-receptor neurons that release CRF, and that CRF-mediated inhibition suppresses anxiolytic BNST outputs to VTA and LH, producing increased fear and anxiety. Blocking CRF signaling in the BNST counteracts this chain and mitigates the behavioral effect.
Next steps and translational potential
Key next steps include confirming whether the same DRN→BNST pathway and CRF-dependent mechanism exist in humans. Kash noted that the pathway is plausibly conserved across mammals, given the similarity of these brain regions and the clinical observation that SSRIs can provoke early anxiety in people.
The researchers are also screening drugs — ideally compounds already approved for other indications — that could safely blunt this anxiety circuit during the initial weeks of SSRI treatment. While pharmaceutical efforts to develop CRF antagonists for psychiatric disorders have so far been challenging, ongoing work to identify other receptors or proteins expressed by these BNST neurons may reveal existing drugs that can provide transient protection against SSRI-induced anxiety.
Study authors and citation
The study was led by Thomas L. Kash, PhD, with co-authors including postdoctoral researcher Catherine A. Marcinkiewcz, PhD, and graduate student Christopher M. Mazzone. Additional contributors include Giuseppe D’Agostino, Lindsay R. Halladay, J. Andrew Hardaway, Jeffrey F. DiBerto, Montserrat Navarro, Nathan Burnham, Lora K. Heisler, Claudia Cristiano, Cayce E. Dorrier, Gregory J. Tipton, Charu Ramakrishnan, Tamas Kozicz, Karl Deisseroth, Zoe A. McElligott, Andrew Holmes and others who contributed to experiments, analysis and interpretation.
Abstract (concise)
The researchers show that serotonin from the dorsal raphe nucleus activates a CRF-expressing subpopulation of BNST neurons via 5-HT2C receptors. This CRFBNST inhibitory microcircuit suppresses BNST outputs to the ventral tegmental area and lateral hypothalamus that normally reduce anxiety. Acute SSRI exposure enhances this pathway, increasing aversive behavior in mice, and CRF1 receptor antagonism prevents this SSRI-induced anxiety effect. The results identify a 5-HTDRN→CRFBNST circuit that promotes fear and anxiety and provide a potential mechanism for early adverse responses to SSRI treatment.
Source: University of North Carolina