October 15, 2013. Overactive dopamine receptors during brain development can leave a lasting impression on the brain, according to a study published October 13 in Nature Neuroscience. Researchers led by Zheng Li at the National Institute of Mental Health in Bethesda, Maryland, found that activity through D2 dopamine receptors (D2Rs) in the hippocampus of mice reduced dendritic spine numbers, but only when the signaling was boosted during a stage of mouse development loosely equivalent to adolescence in humans. This overactive D2R signaling in adolescence resulted in connectivity disruptions and a working memory deficit in adulthood.
The findings point to a link between two features of schizophrenia: the overactive D2R signaling hypothesized to precede and drive psychosis (see SRF hypothesis paper) and the reduction in dendritic spines found in postmortem brain samples from people with schizophrenia (Glausier et al., 2013; see SRF related news story). Dendritic spines are the receiving end of excitatory synapses, and changes to them could significantly alter information flow through the brain. Brain imaging continues to uncover connectivity changes in schizophrenia (see SRF related news story) that may be related to cognition (see SRF related news story).
The findings also highlight adolescence as a special time in brain development. Though early brain development has captured the bulk of research interest, the teenage brain is starting to be recognized for the important maturational processes underway at that time (see SRF conference report): When derailed, these might heighten risk for schizophrenia (see SRF related news story). The adolescent brain also seems more plastic than the adult brain: In a rat model of schizophrenia, adolescent rats benefit from cognitive training that was not helpful to adult rats (see SRF news related story). The new study, too, points to enhanced plasticity in adolescence, which may guide future treatment strategies.
Using several approaches, first author Jie-Min Jia and colleagues found that D2R activity influenced dendritic spine numbers in mice at postnatal day 21—an age that might be considered comparable to early adolescence in humans—just after weaning, but before sexual maturity. Excitatory synapses are actively being made at this time. The researchers injected these mice with D2R agonists (quinpirole or bromocriptine), waited 24 hours, then made hippocampal slices from their brains and labeled CA1 neurons. This revealed a slight decrease in spine density compared to mice injected with vehicle. In contrast, injection with the D2R blocker eticlopride increased spine density relative to controls. Likewise, overexpressing D2Rs with constructs delivered to the hippocampus via a lentiviral vector resulted in lower spine density, whereas knocking down D2Rs increased spine density.
D2R activation seemed to stall the transition from immature to mature dendritic spines. Spines begin as a stringy filopodium, then fatten into stubby spines before constricting at the stalk and bloating at the tip to take on a mature mushroom shape or a thinner shape marked by a small head and a long neck. In hippocampal neurons cultured for 14 days, D2R agonists reduced the density of mushroom and thin spines, increased filopodium density, but produced no change in stubby spines. Time-lapse imaging during one hour of quinpirole treatment revealed active spine remodeling in which fewer mature spines successfully took root compared to controls.
Inside the neurons, the researchers identified some of the molecular middlepersons between D2R activation and spine reduction. The cAMP pathway downstream of D2Rs was involved, as were GluN2Bs, a subunit of the NMDA type of glutamate receptor (NMDAR) that characterizes immature synapses.
The researchers next looked at mice that overexpressed D2Rs. Called “sandy” mice, these animals carry a natural deletion in dysbindin, a gene linked to schizophrenia that is involved in trafficking D2Rs to the cell membrane (see SRF related news story; SRF news story). Consistent with the previous experiments, the sandy mice showed a reduction in spine density relative to wild-type littermates—either in slices made from P21 mice or in hippocampal cultures. This reduction could be counteracted by knockdown of D2Rs via siRNA transfection or by boosting cAMP, suggesting the same connection with D2R signaling. D2Rs were central: The selective D2R blocker loxapine normalized spine density in sandy mice, whereas the less selective clozapine did not.
But the spine reduction occurred only during the time window of mouse "early adolescence": while sandy mice showed reduced spine density in the hippocampus relative to wild-type mice between three and six weeks of age, at two, eight, and 12 weeks their spine density was normal. Even knocking down D2Rs in adult sandy mice via lentivirus vector could not force a change in dendritic spine density, as though the D2R-spine mechanism becomes uncoupled in adulthood. Similar results were found in wild-type mice, and further experiments indicated that the critical time window hinged on GluN2Bs: Mature neurons could be tricked into responding to quinpirole with a decrease in dendritic spines when GluN2Bs were overexpressed, as is found in young synapses.
Connection to working memory?
Although spine density returns to normal in adults, the researchers found that the period of D2R overactivation at P21 had lasting effects on connectivity between the entorhinal cortex and its hippocampal target. Retrograde tracers injected into the CA1 layer of the hippocampus filled neurons in both the medial and lateral parts of the entorhinal cortex in adult wild-type mice, but this connectivity was shifted toward the lateral part in adult sandy mice. Treating them with a D2R blocker at P21—but not in adulthood—prevented this shift, and similar results were obtained in wild-type mice treated with quinpirole.
Overactive dopamine signaling in early adolescence may have also spurred deficits in spatial working memory, which relies on connectivity between hippocampus and entorhinal cortex. When placed in a Y maze, mice will typically enter each of the three arms consecutively, which reflects some memory for where they have recently been. Adult sandy mice alternated between the arms a bit less than wild-type mice did; similarly, wild-type mice that received a pharmacological boost to D2R activity at P21 alternated less than sandy mice treated with D2R blockers at the same age did. Finally, sandy mice treated with D2R blockers as adults still showed deficient alternation, which suggests that the window for engaging the dopamine-dendritic spine mechanism had closed.
Although future experiments will have to probe more thoroughly the cognitive effects of such dendritic spine changes, the findings point to early adolescence as a special time for brain plasticity. Schizophrenia does not usually announce itself until late adolescence or early adulthood, but studies of the prodrome may increasingly identify signs of derailed brain maturation. If these coincide with times of enhanced adolescent plasticity, then this may provide a special opportunity for treatment, or even prevention.—Michele Solis.
Jia JM, Zhao J, Hu Z, Lindberg D, Li Z. Age-dependent regulation of synaptic connections by dopamine D2 receptors. Nature Neuroscience. 2013 Oct 13. Abstract