A new study published in Molecular Psychiatry has found that specific immune cells within the mouse brain are a direct cause of chronic anxiety and compulsive grooming. The research demonstrates that two distinct lineages of these cells, known as microglia, function in opposition to one another, with one group promoting these behaviors and the other acting to suppress them. These findings shift attention from neurons to the brain’s immune system as a potential regulator of certain psychiatric conditions.
Microglia are the resident immune cells of the central nervous system, responsible for maintaining brain health by removing debris and responding to injury or infection. It has become clear that their role extends beyond simple housekeeping. In mice, the microglial population is composed of two separate lineages that arise at different times during embryonic development. The majority are a type known as canonical non-Hoxb8 microglia, while a smaller subset of about 25 percent are called Hoxb8 microglia, named for a developmental gene they express.
A team of researchers at the University of Utah School of Medicine, led by Distinguished Professor Mario R. Capecchi, previously established a link between the Hoxb8 gene and specific behaviors. Mice with a disrupted Hoxb8 gene exhibit chronic anxiety and pathological overgrooming, a behavior that resembles trichotillomania, an obsessive-compulsive spectrum disorder in humans. Since the only cells in the brain that express the Hoxb8 gene are this specific subset of microglia, the team formulated a direct question: are defective Hoxb8 microglia the cause of these behaviors?
To answer this, the scientists performed a series of cell transplantation experiments. The first step required creating recipient mice with brains that were essentially a blank slate, devoid of their own native microglia. This was achieved by genetically engineering mice so that a gene called Csf1r, which is essential for microglia survival, was disabled only in those cells. These recipient mice provided a unique environment to test the function of transplanted cells in isolation.
The researchers then isolated microglia progenitor cells, an early-stage cell that develops into mature microglia, from the fetal livers of mouse embryos. They collected these progenitors from two different sources: from normal, healthy mice and from mice with the defective Hoxb8 gene. These purified cells were then injected into the brains of the microglia-less newborn recipient mice. This procedure created two experimental groups: one populated with healthy Hoxb8 microglia and another populated solely with defective Hoxb8 microglia.
The behavioral outcomes of these two groups were distinctly different. Mice that received the defective Hoxb8 microglia grew up to display the same pathological behaviors seen in the original mutant mice. They groomed themselves compulsively, leading to significant hair loss, and exhibited signs of heightened anxiety in standardized tests, such as avoiding open spaces in a maze.
In contrast, mice that received healthy Hoxb8 microglia from normal donors behaved just like typical mice, showing no signs of overgrooming or elevated anxiety. This experiment established a direct causal link, demonstrating that the defective Hoxb8 microglia were themselves sufficient to produce both behaviors.
This finding raised a more complex question about how the two different microglial populations work together. In previous work, the team proposed what they call the “Accelerator/Brake” model. This model suggests the two microglia lineages have opposing functions. The non-Hoxb8 microglia are thought to act as an “accelerator,” promoting anxiety and grooming. The Hoxb8 microglia are thought to function as a “brake,” downregulating these same behaviors to maintain equilibrium.
The model made a powerful prediction: a mouse containing only the accelerator cells, the non-Hoxb8 microglia, should exhibit abnormally high levels of anxiety and grooming because the braking system would be absent. To test this, the team performed a second set of transplantations. This time, they used a different type of recipient mouse that is born completely without any microglia, providing an even cleaner experimental system.
From the brains of healthy newborn mice, the scientists isolated and purified both types of microglia, separating the Hoxb8 from the non-Hoxb8 populations. They then created several groups of recipient mice. Some received only non-Hoxb8 microglia (the accelerator). Others received only Hoxb8 microglia (the brake). A third group received a mixture of both cell types that mimicked the brain’s natural 75-to-25 percent ratio.
The results strongly supported the Accelerator/Brake model. Mice populated exclusively with non-Hoxb8 microglia developed pathological grooming habits and showed increased anxiety. The “accelerator” was effectively stuck on. Mice that received only Hoxb8 microglia, or those that received the correct mixture of both populations, showed normal, low levels of grooming and anxiety. This suggests that in a healthy brain, the two types of microglia work in concert to fine-tune behavioral responses to the environment.
A final, subtle observation from the experiments added another layer of complexity. The original mice with the Hoxb8 gene disruption showed even more severe grooming and anxiety than the mice transplanted with only non-Hoxb8 microglia. A mouse with only non-Hoxb8 microglia represents a pure “loss of function” for the Hoxb8 gene, as the braking cells are simply absent.
The more severe symptoms in the original mutant mice imply that their defective Hoxb8 microglia are not just failing to apply the brakes; they are doing something actively detrimental. This phenomenon, known as a “gain of function,” suggests the mutated cells may be sending their own incorrect signals that further contribute to the behaviors.
These experiments are based on mouse models, and any direct application to human anxiety disorders requires further investigation. “Humans also have two populations of microglia that function similarly,” Dr. Capecchi notes. He suggests this work could reframe how researchers approach psychiatric conditions, which have historically been viewed almost exclusively through the lens of neurons. “This knowledge will provide the means for patients who have lost their ability to control their levels of anxiety to regain it,” he adds.
Future research will likely explore the molecular mechanisms through which these microglia influence neuronal circuits to control such complex behaviors. Understanding this cellular dialogue could open new therapeutic avenues. “We’re far from the therapeutic side,” says Donn Van Deren, the study’s first author, now a postdoctoral fellow at the University of Pennsylvania. “But in the future, one could probably target very specific immune cell populations in the brain and correct them through pharmacological or immunotherapeutic approaches. This would be a major shift in how to treat neuropsychiatric disorders.”
The study, “Defective Hoxb8 microglia are causative for both chronic anxiety and pathological overgrooming in mice,” was authored by Donn A. Van Deren, Ben Xu, Naveen Nagarajan, Anne M. Boulet, Shuhua Zhang & Mario R. Capecchi.
