Cellular/Molecular
Noradrenergic Terminals Supply Dopamine to Hippocampus
Caroline C. Smith and Robert W. Greene
(see pages 6072–6080)
Dopaminergic signaling from the ventral tegmental area (VTA) to nucleus accumbens, prefrontal cortex, and limbic areas is essential for reward-based learning and motivation. Many drugs of abuse enhance dopaminergic signaling, and the resulting learning likely contributes to addiction. In the hippocampus, activation of D1 dopamine receptors enhances excitatory transmission and long-term potentiation, and this is thought to contribute to reinstatement of drug seeking by strengthening contextual memories associated with drug rewards. But the D1 receptors underlying these effects are primarily located in dorsal hippocampus, where innervation by VTA afferents is sparse. In contrast, noradrenergic afferents from locus ceruleus (LC) are dense in dorsal hippocampus. Notably, dopamine is packaged into synaptic vesicles before being converted to norepinephrine in noradrenergic terminals. Smith and Greene present evidence that such terminals are the primary source of dopamine in dorsal hippocampus. Preventing neurotransmitter synthesis in LC, but not in VTA, blocked the effects of hippocampal dopamine release.
Development/Plasticity/Repair
Thalamocortical Activity Drives Barrel Formation
Nicolas Narboux-Nême, Alexis Evrard, Isabelle Ferezou, Reha S. Erzurumlu, Pascal S. Kaeser, et al.
(see pages 6183–6196)
In rodents, thalamocortical axons conveying information about whisker movements terminate in whisker-specific patches in somatosensory cortex. The somata of postsynaptic spiny stellate neurons form barrels around these patches, with their dendrites projecting inward. Barrel development requires both input from whiskers and glutamatergic transmission in the cortex, but whether activity-dependent glutamate release is required was unknown. To investigate this question, Narboux-Nême et al. knocked out Rab3-interacting molecules (RIMs), which are required for calcium-triggered vesicle release, selectively in mouse thalamic neurons. As expected, this reduced neurotransmitter release probability at thalamocortical synapses and reduced cortical responses to whisker stimulation. It also disrupted the formation of barrel walls. Although RIM-deficient axons innervated the correct cortical layer with normal topographical patterning, spiny stellate neurons were evenly distributed in barrel walls and hollows, and their dendrites were longer, spinier, and oriented more symmetrically than normal. Knocking out RIMs selectively in excitatory cortical neurons did not significantly alter barrel architecture.
In wild-type mice (left panels) thalamocortical axons terminate in patches in somatosensory cortex (top) and postsynaptic neurons form rings around these patches (bottom). Thalamocortical axons lacking RIMs (top right) form patches, but postsynaptic neurons fail to form barrels (bottom right). See the article by Narboux-Nême et al. for details.
Behavioral/Systems/Cognitive
Monetary Loss Alters Perceptual Discrimination
Offir Laufer and Rony Paz
(see pages 6304–6311)
When a cue is associated with a negative outcome, an animal alters its response not only to that cue, but also to similar cues, to avoid the negative event. What is the neural basis for this generalized response? One might expect that animals can distinguish the stimuli, but choose to respond similarly to minimize risk. Surprisingly, however, generalization also happens at the perceptual level. This occurs not only for primary reinforcers, but also, as Laufer and Paz report, for secondary reinforcers, e.g., money. In humans, tone discrimination (i.e., the smallest difference in tone that can be detected) improves with repeated presentation, even when no reward is given. Associating a tone with monetary gain did not further enhance discrimination, but associating a tone with a monetary loss decreased discrimination ability, even if subsequent discrimination was rewarded. Such alterations at the perceptual level may be beneficial, because they allow quicker responses to potentially harmful stimuli.
Neurobiology of Disease
Reactive Astrogliosis Varies across Disease States
Jennifer L. Zamanian, Lijun Xu, Lynette C. Foo, Navid Nouri, Lu Zhou, et al.
(see pages 6391–6410)
CNS injury causes reactive astrogliosis, in which astrocyte gene expression and morphology are dramatically altered. Although different types of injury trigger astrogliosis via different signals, whether the resulting reactive states are the same is unknown. Whether reactive astrocytes are helpful or harmful also remains unclear. To address these questions, Zamanian et al. analyzed gene expression profiles of astrocytes isolated from mouse brain after transient cerebral artery occlusion (modeling ischemia) or peripheral injection of a bacterial endotoxin (modeling inflammation). Unsurprisingly, two established markers of reactive astrocytes, GFAP and vimentin, were strongly upregulated in both injury models. But two other established markers were induced only after ischemia. In all, only 56 of 263 strongly induced genes were upregulated in both injury models, whereas 57 were induced only after inflammation and 150 were induced only after ischemia. Interestingly, the pattern of induced genes suggests that reactive astrogliosis is protective after stroke, but might be harmful after infection.