19 January 2013. Stressful events contribute to the development of psychiatric illnesses, but what are the transducers of this connection? Two studies published January 18 in Science uncover actions by stress hormones on dopamine signaling in mice. One study, led by François Tronche of Université Pierre et Marie Curie in Paris, France, finds that stress-induced glucocorticoids act on neurons receiving dopamine messages. The second study, led by Akira Sawa of Johns Hopkins University in Baltimore, Maryland, examines stress-induced consequences in mice engineered to model a genetic predisposition for psychiatric illness—disruption of the DISC1 gene. In these mice, stress-induced glucocorticoids tamped down dopamine release through epigenetic mechanisms. Both studies illuminate circuit-specific consequences of the body-wide release of glucocorticoids during stressful events.
Although these studies do not set out to model a specific psychiatric disorder, they grapple with mechanisms behind the adolescence or early adult onset of many of these, including schizophrenia. Because stress can precipitate illness onset, researchers have been studying the cumulative effects of stress on mouse behavior. Different stress regimens in mice can produce various adverse outcomes, including immobility in the forced swim test (taken as a sign of despair), anxiety-like behaviors, or impaired social interactions (Russo et al., 2012). Recent optogenetic studies in rodents have linked stress-induced depression-like behaviors to dopamine signaling (see SRF related news story), but the new studies address how dopamine systems become dysregulated by stress in the first place.
Glucocorticoids are steroid hormones released by the adrenal glands during stress. Glucocorticoid receptors are found throughout the brain, and when activated, they move to the cell’s nucleus to work as transcription factors, where they can fine-tune gene expression. But that isn’t their only mode of action, as laid out in an accompanying perspective piece by Bruce McEwen (McEwen, 2013). Though the precise steps between glucocorticoid receptor activation and changes to dopamine systems remain murky, the results suggest that glucocorticoids may offer some therapeutic clues to psychiatric disorders.
The agony of social defeat
In the first study, first author Jacques Barik and colleagues use the social defeat paradigm, in which mice are repeatedly attacked by another mouse. After 10 days of this, these “socially defeated” mice show signs of social aversion, spending less time interacting with a novel mouse. To figure out where in the brain glucocorticoids were acting to produce this behavior, the researchers engineered mice to lack glucocorticoid receptors in dopamine-containing neurons of the ventral tegmental area (VTA); these mice, however, still developed social aversion after social defeat. But when the researchers tested a different group of mice—missing glucocorticoid receptors in dopamine-receiving neurons (specifically, those with D1a receptors), these mice did not develop social aversion after social defeat, looking very much like “undefeated” mice. Interestingly, anxiety-like and despair-like behaviors remained in these mice, suggesting they are mediated by separate neural circuitry.
The ablation of glucocorticoid receptors from dopamine-receiving neurons trickled back to the dopamine-containing neurons of the VTA, too. The researchers found signs of abnormal dopamine neuron activity, with less burst-firing recorded in the VTA, and less dopamine release measured by microdialysis in the nucleus accumbens, a target of the VTA. The researchers suggest that this might involve a feedback loop from dopamine-receiving neurons in the nucleus accumbens that project back to the VTA. Next, the researchers tested whether suppressing dopamine release in control mice could protect them from developing social aversion in response to social defeat. It did: while saline-injected mice exhibited the expected social aversion, mice injected with quinpirole (a D2R agonist that lowers dopamine neuron activity through autoreceptors) spent more than twice the amount of time interacting with another mouse.
In the second study, first author Minae Niwa and colleagues focused on the lingering effects of an episode of stress in adolescence in a mouse model of susceptibility to mental illness. Mice were engineered to carry a dominant-negative form of human DISC1, a gene fingered by a chromosomal translocation found in a family beset by mental illness (see SRF related news story). Sawa and colleagues have proposed that the DISC1 disruption produces a truncated protein that binds to normal copies of the protein, thus interfering with their usual function. Sawa’s group has found schizophrenia-related phenotypes in mice expressing this dominant-negative DISC1 (DN-DISC1) in hippocampus and cortex (see SRF related news story), but the new study expresses the construct more widely throughout the brain via a prion protein promoter.
Because brain maturation continues into the teenage years (see SRF related news story), this is considered a vulnerable time for brain development, when stressful events may have lasting consequences. The researchers found that housing the DN-DISC1 mice alone for three weeks during adolescence produced behavioral abnormalities likened to symptoms of psychiatric disorders: impaired paired pulse inhibition, longer immobility in the forced swim test, and increased locomotion (including after an acute injection of methamphetamine). These outcomes required a combination of the genetic risk and stressful environment: controls, control mice with the same schedule of social isolation, or DN-DISC1 mice without social isolation did not develop these behaviors.
Looking for neural changes that might account for these behaviors, the researchers found evidence for disturbed dopamine signaling in one pathway: the frontal cortex contained decreased dopamine levels and decreased tyrosine hydroxylase (TH), an enzyme needed to make dopamine; these changes were not apparent, however, in the caudate putamen or nucleus accumbens, which are subcortical destinations for dopamine-releasing projections. These disturbances were put right by a dose of mifepristone, a glucocorticoid receptor blocker, thus implicating the stress hormone in creating the dopamine disturbances. Further experiments found that DN-DISC1 mice receiving a stint of social isolation had elevated methylation of the TH gene—an epigenetic mark that suppresses transcription—but only in those VTA neurons projecting to the cortex, not those headed to the nucleus accumbens.
This methylation pattern was not easily undone, lasting well into mouse adulthood. But a dose of mifepristone normalized methylation of the TH gene, indicating regulation by glucocorticoids. The researchers suggest that their mice may model psychotic depression, an illness reported to respond to glucocorticoid receptor blockers (Flores et al., 2006).
Together, the studies offer up some tangible leads on where stress acts in the brain to produce adverse outcomes, for mice at least. Given the complex interactions of glucocorticoids in the brain, more remains to be discovered, but the circuit-specificity of glucocorticoid action can already inform ideas about how individuals may cope with stressful events.—Michele Solis.
Barik J, Marti F, Morel C, Fernandez SP, Lanteri C, Godeheu G, Tassin JP, Mombereau C, Faure P, Tronche F. Chronic stress triggers social aversion via glucocorticoid receptor in dopaminoceptive neurons. Science. 2013 Jan 18. Abstract
Niwa M, Jaaro-Peled H, Tankou S, Seshadri S, Hikida T, Matsumoto Y, Cascella N.G. Kano S, Ozaki N, Nabeshima T, Sawa A. Adolescent Stress–Induced Epigenetic Control of Dopaminergic Neurons via Glucocorticoids. Science. 2013 Jan 18. Abstract