This Week in The Journal

Cellular/Molecular

Zinc Regulates Release of Some Synaptic Vesicles

Nathalie Lavoie, Danny V. Jeyaraju, Modesto R. Peralta III, László Seress, Luca Pellegrini, et al.

Some miniature EPSCs (mEPSCs) recorded in hippocampal CA3 pyramidal cells have unusually large amplitudes. These mEPSCs are thought to result from unusually high glutamate content in a subset of synaptic vesicles in mossy fiber terminals. This might occur because zinc, which is loaded into mossy fiber synaptic vesicles by zinc transporter 3 (ZnT3), increases activity of a vesicular glutamate transporter. Consistent with this, Lavoie et al. found that chelating zinc or knocking out ZnT3 reduced the number of large-amplitude mEPSCs in mouse hippocampal slices. Eliminating vesicular zinc also slowed release of a subset of vesicles during high-frequency stimulation and altered the calcium dependence of this release. Zinc was present in less than 20% of mossy fiber synaptic vesicles, and after prolonged activity, zinc labeling moved from vesicles to endosome-like structures. These data suggest that during high-frequency stimulation, zinc speeds the release of vesicles having high glutamate content, and these vesicles are recycled through the endosomal pathway.

Development/Plasticity/Repair

Synapsin Upregulation Contributes to Long-Term Facilitation

Anne K. Hart, Diasinou Fioravante, Rong-Yu Liu, Gregg A. Phares, Leonard J. Cleary, et al.

(see pages 18401–18411)

Long-term sensitization (LTS) of the gill-and-siphon withdrawal reflex in Aplysia has been a valuable model for investigating the neural bases of memory formation. LTS results from long-term facilitation (LTF) of sensory–motor neuron synapses. Many of the cellular and molecular mechanisms that underlie this LTF—e.g., activation of the cAMP response element-binding protein CREB-1, subsequent transcriptional activation and protein synthesis, and structural modification of synapses—are conserved across species. Of the hundreds of CREB-regulated genes, however, few have been proven necessary for long-term plasticity. Hart et al. suggest that one such gene encodes the synaptic vesicle protein synapsin, which is involved in vesicle docking, fusion, and recycling. Serotonin, which induces LTF in Aplysia sensory–motor neuron cultures, increased binding of CREB to the synapsin promoter and increased synapsin expression. Moreover, when synapsin upregulation was blocked via RNA interference, serotonin no longer induced LTF, suggesting synapsin upregulation was required for LTF.

Behavioral/Systems/Cognitive

ACC Neurons May Signal New Reward Contingencies

Daniel W. Bryden, Emily E. Johnson, Steven C. Tobia, Vadim Kashtelyan, and Matthew R. Roesch

(see pages 18266–18274)

According to reinforcement learning theories, after animals associate a cue–action pair with reward, the cognitive effort required to obtain the reward decreases; if the learned action then fails to produce reward—or if reward is obtained unexpectedly—cognitive processes are re-engaged so new reward contingencies can be learned. Although neurons that respond to positive and/or negative unexpected outcomes have been identified in several brain areas, neurons that re-engage cognitive processes have not been identified. Bryden et al. hypothesized that such neurons are located in the anterior cingulate cortex (ACC), an area previously linked to arousal and attention, predicting action outcomes, and signaling surprising outcomes. Consistent with their hypothesis, neurons in rat ACC increased firing—particularly around the time of cue presentation—on trials following reward contingency shifts. Firing rates were highest on trials in which rats responded most quickly to the start signal, consistent with this firing being associated with increased attention.

A neuron in ACC fired more in the first 10 trials after the reward contingency changed (red), particularly around the time an odor cue was presented (dashed line), than after many trials in which the expected reward was received (blue). See the article by Bryden et al. for details.

Neurobiology of Disease

Melatonin MT₂ Receptors Promote Non-REM Sleep

Rafael Ochoa-Sanchez, Stefano Comai, Baptiste Lacoste, Francis Rodriguez Bambico, Sergio Dominguez-Lopez, et al.

(see pages 18439–18452)

Non-rapid eye movement (NREM) sleep is characterized by EEG slow waves and spindles. Spindles are produced by rhythmically active GABAergic neurons in nucleus reticularis (Rt). These neurons phasically hyperpolarize thalamocortical neurons, causing postinhibitory spike bursts that drive cortical oscillations. NREM sleep is initiated by GABAergic neurons in the preoptic region of the hypothalamus, which are activated by neurochemicals that accumulate during wakefulness. The light–dark cycle also influences sleep onset, partly via melatonin, which is produced in the dark. Ochoa-Sanchez et al. report that melatonin promotes NREM sleep via type 2 receptors (MT₂Rs) in Rt. MT₂Rs were highly expressed in rodent Rt, and an MT₂R agonist increased spikes per burst in Rt neurons. The agonist also decreased time spent awake and latency to sleep, and increased NREM sleep duration and spindle frequency without affecting REM sleep. These effects were absent in MT₂R-null mice, which spent more time awake and less time in NREM sleep than wild-type mice.