Sound the alarm! Activating a subset of neurons in the medulla promotes anxiety in mice, a study published in Neuron finds.
Anxiety is typically thought of as a set of “complex, social, environmental, biological, neurobiological interactions” involving multiple regions across the brain, says Oliver Robinson, professor of neuroscience and mental health at University College London, who was not involved in this work. So he was “surprised that you can have such dramatic effects by just looking at a single cell type.”
The role of medullary neurons in anxiety adds to a growing body of work on anxiety circuitry, says study investigator Carlos Fernández-Peña, assistant professor of neurological sciences at the University of Nebraska. “It was already complex. Now let’s add some more.”
Two sets of neurons—C1 and A1—in the rostral ventrolateral medulla (RVLM) produce catecholamines that could drive anxiety, but only C1 neurons are activated when mice are stressed, Fernández-Peña says. The challenge, then, was to study only C1 neurons, which are intermingled with A1 neurons, in awake mice to uncover their role in anxiety-like behaviors, he says.
He and his colleagues employed two different genetic recombination tools—one called INTRSECT and another called ConVERGD—for the first time, according to study investigator Lindsay Schwarz, associate member of the St. Jude Children’s Research Hospital faculty, who co-developed the latter tool. This pairing enabled them to make transgenic mice in which they could optogenetically modulate only the C1 neurons and not the A1 cells, using careful genetic logic.
The specificity in targeting C1 neurons was “technically very impressive,” relying on “deceptively tricky” tools, says Alexxai Kravitz, associate professor of neuroscience and psychiatry at Washington University in St. Louis, who was not involved in this work.
O
ptogenetically activating the C1 neurons caused transgenic mice to linger near the walls more than controls did in the open field test and elevated zero maze—standard assays to study anxiety-like behaviors in mice. Conversely, eavesdropping on the C1 neurons while the mice transitioned from safe closed areas to anxiety-inducing open ones showed that stress induces C1 neuronal activity, which declined sharply once the mice returned to their safe spaces.In a looming fear test, where a dark shadow broadens above mice to simulate a predator approaching, those with optogenetically silenced C1 neurons were less likely to freeze in place than controls.
Although C1 neurons project to many brain regions, their projection to periaqueductal gray matter (PAG) is responsible for processing information about physiological changes and for modulating emotional responses, the study suggests. Optogenetic stimulation of C1 projections to ventrolateral PAG in particular made mice more likely to stick to the walls of the behavioral tests without exploring. A week later, they still crouched in the corners during a follow-up test, even though they had been undisturbed in the meantime.
A
lthough these findings indicate that C1 neurons play a part in promoting anxiety, “we don’t know where in the circuit we are right now,” Schwarz says. “Our role is to keep walking both forwards and backwards in the circuit and finding all of the relevant contributors that are multiple synapses away, and we just started.”Anxiety has “a really crazy circuit that’s hugely dispersed across the brain,” says Ciaran Murphy-Royal, associate professor of neuroscience at the Université de Montréal, who showed that amygdala astrocytes encode anxiety-like states in mice in a study published in March 2026. He says he doesn’t think this work in medullary neurons conflicts with his, though, because C1 neurons and astrocytes could be two parts of a larger circuit. In fact, this work gives him a new place to look for astrocytes that may contribute to anxiety. “There’s very few people doing astrocyte studies in the brainstem. It’s really difficult.”
Part of the difficulty of studying isolated astrocyte populations is the lack of intersectional tools like the ones this study develops, which were the “key features” that made this C1 neuron targeting possible, Murphy-Royal says.
Schwarz agrees, saying that “this whole project wouldn’t have been possible without this kind of ongoing effort to make better tools to target these neurons,” and advises researchers not to let the lack of a tool stop a project before it starts. She says that more intersectional tools need to be developed in order to target only A1 neurons, which could also contribute to the vast anxiety circuit.
