February 6, 2014. The brain forms its thicket of connections early in life, but additions to its wiring diagram may be made even in adults, according to a study published in Neuron on January 8. Led by Christoph Kellendonk of Columbia University in New York City, the study reports that, in mice, axonal branching increases in response to hyperactivity in other neurons within the basal ganglia, which contain circuitry important for movement, learning, and motivation.
Of possible relevance to schizophrenia, this axon remodeling was sensitive to dopamine 2 receptor (D2R) activity in the striatum, with more connections spurred by overexpressing D2Rs, but diminished with D2R blockade. The researchers suggest that wayward axonal connections may characterize the brain in schizophrenia.
The findings suggest a conspicuous sort of plasticity in the adult brain. Typically, adult plasticity involves functional changes in the strength of existing connections, or maybe even structural expansions or shrinkages of dendrite structure. But the new study suggests a wholesale growth of new connections between neurons not usually thought to be connected. These new connections bridged the direct pathway with the indirect pathway, two parallel tracks in the basal ganglia with opposing functions. Though previous studies have found similar evidence for links between these two pathways (e.g., Fujiyama et al., 2011), the new study casts these connections as highly malleable.
The sensitivity of this axon remodeling to signaling through D2Rs suggests a link to schizophrenia. To mimic the overactive dopamine signaling found in the striatum in people with schizophrenia (see SRF related news report; see also SRF Hypothesis), Kellendonk and colleagues have developed a mouse model that overexpresses D2Rs in the striatum. These mice show both cognitive and motivational deficits (see SRF related news report), as well as changes in striatal circuitry that are readily reversible. For example, D2R-overexpressing mice have withered dendrites, but these can be restored by suppressing D2R signaling (see SRF related news report). This suggests that targeted modulation of D2Rs—or of other, related changes—in the striatum could be a worthy therapeutic strategy, particularly for negative symptoms.
A direct bridge to globus pallidus
Because their previous study (see SRF related news report) found that D2R-overexpression raised the excitability of striatal output neurons—the medium spiny neurons—first author Maxime Cazorla and colleagues started by asking whether similar increases in excitability could produce axon remodeling. Using an adenovirus vector to deliver to the striatum a dominant-negative mutant of Kir2, a potassium channel that normally works to keep neurons in a hyperpolarized state, the researchers counted more-than-normal axon terminals in the external subdivision of the globus pallidus (GPe), a nucleus within the indirect pathway.
Other experiments established the direct pathway as the source of these extra inputs, namely the D1R-expressing medium spiny neurons of the striatum. Normally, these only project to the substantia nigra pars reticulate (SNr) and the entopeduncular nucleus (EN), composing the so-called direct pathway. In contrast, the indirect pathway originates in D2R-expressing striatal neurons, which in turn inhibit the GPe. These project to the subthalamic nucleus, which only then connects to the SNr/EN.
But the separation between these two pathways blurred when excitability was increased in D2R-expressing striatal neurons of the indirect pathway: The GPe showed increased density of axon terminals from striatal neurons of the direct pathway, which maintained normal densities in their usual SNr/EN targets. While these extra “bridge collaterals” connected the direct with the indirect pathway, the density of axon terminals originating from striatal neurons of the indirect pathway went unchanged. The researchers proposed that hyperexcitable conditions induce the release of a signal in the GPe that attracts axonal offshoots only from striatal neurons in the direct pathway.
The researchers then turned to their D2R-overexpressing mice, also characterized by hyperexcitable D2R-striatal neurons. Similarly, these mice showed a 1.75-fold increase in bridging collaterals in the GPe over the amount also found in control mice. These connections seemed fairly malleable, as their numbers in GPe decreased when excitability was turned down, or when D2Rs were blocked by haloperidol, a common antipsychotic drug, or when the extra D2R transgene was turned off with doxycycline. In the last case, changes in the density of bridge collateral terminals were apparent by three days and reached their full effect at two weeks.
Putting on the brakes
These bridge collaterals made noticeable dents in GPe activity. Optogenetically activating the direct pathway striatal neurons suppressed GPe firing nearly as much as optogenetically activating the indirect pathway alone. In awake, behaving control animals, activation of the direct pathway promoted locomotion, but in D2R-overexpressing mice presumably harboring bridge collaterals, this suppressed locomotion, as though the movement-inhibiting actions of the indirect pathway had been recruited. Haloperidol treatment limited this, such that optogenetic activation of the direct pathway resumed its locomotion-producing effects. This suggests that haloperidol cut off the influence of the bridge collaterals in D2R-overexpressing mice.
These bridge collaterals, then, may provide a way for the direct pathway to elicit activity in the indirect pathway, which could oppose the direct pathway’s own function. This kind of crosstalk could have consequences not just for movement, but also for cognition and motivation. The researchers propose that bridge collaterals are at work in schizophrenia and may create an imbalance in direct and indirect pathway activity that contributes to the disorder’s symptoms. Visualizing these connections will require new imaging studies.—Michele Solis.
Cazorla M, de Carvalho FD, Chohan MO, Shegda M, Chuhma N, Rayport S, Ahmari SE, Moore H, Kellendonk C. Dopamine D2 receptors regulate the anatomical and functional balance of basal ganglia circuitry. Neuron. 2014 Jan 8;81(1):153-64. Abstract