Co-Transport of Axonal Sodium and Potassium Channels
Grant P. Higerd-Rusli, Matthew Alsaloum, Sidharth Tyagi, Nivedita Sarveswaran, Mark Estacion, et al.
(see pages 4794–4811)
Neuronal function, from synaptic integration to spike output, depends heavily on the type, expression level, and localization of ion channels in the plasma membrane. Axonal ion channels are synthesized in the cell body and must be delivered to appropriate locations along the axon shaft and within the axon terminal. Studying how these channels are trafficked has been difficult because they are expressed at low levels and therefore are difficult to visualize using fluorescent tags. A recently developed technique—optical pulse-chase axonal long-distance (OPAL) imaging—overcomes this problem by labeling ion channels with fluorescent tags selectively in the soma, keeping background fluorescence low in axons. This technique was previously used to track movement of vesicles containing NaV1.7 in the distal portion of sensory axons. Higerd-Rusli et al. have now used the technique to compare trafficking of additional channel types as well as other proteins.
The authors examined all the major NaV channel types that contribute to the depolarizing phase of action potentials in unmyelinated C fibers of rat dorsal root ganglion; they also examined KV7.2 channels, which contribute to the repolarizing phase. Labeled channels moved predominantly in the anterograde direction, albeit with pauses, and accumulated distally. Notably, the number of moving and paused vesicles and the rate of movement was similar regardless of which tagged ion channel was expressed. Tagging of two proteins in the same cell revealed that—as previously shown for NaV1.7—NaV1.8 and KV7.2 were carried in vesicles that contained Rab6a, a small GTPase involved in vesicle trafficking. Moreover, vesicles carrying NaV1.7 often carried NaV1.6, NaV1.8, NaV1.9, NaVβ2, KV7.2, or tumor necrosis factor receptor 1 as well. In contrast, the sodium calcium exchanger NCX2 was trafficked independently of NaV1.7. Moreover, although neuropeptide Y is carried in Rab6a-containing vesicles, it was trafficked independently of NaV1.8.
These results indicate that a variety of axonal membrane proteins with opposing or unrelated functions are transported in the same vesicles. Given that the relative levels of different ion channels strongly influence neuronal function, future work should determine whether appropriate levels are achieved before the channels are loaded into vesicles, after they reach their destination, or both.
NaV1.8 (purple) is carried in vesicles containing Rab6a (yellow, top three panels). Although neuropeptide Y (yellow, bottom three panels) is also carried in Rab6a-containing vesicles, it is trafficked independently of NaV1.8. See Higerd-Rusli et al. for details.
Changes in Interneuron Activity Across the Sleep/Wake Cycle
Aurélie Brécier, Mélodie Borel, Nadia Urbain, and Luc J. Gentet
(see pages 4852–4866)
The aggregate activity of neuronal populations gives rise to oscillations that can be measured with electroencephalography. Oscillatory frequency varies across behavioral states from active task engagement to restful wakefulness to rapid eye movement (REM) and non-REM (NREM) sleep. How the activity of different populations of interneurons changes over oscillatory cycles and across behavioral states is poorly understood. To address this, Brécier et al. recorded across the sleep/wake cycle from genetically identified parvalbumin- (PV), somatostatin- (SST), and vasoactive intestinal polypeptide- (VIP) expressing neurons, as well as from presumptive pyramidal neurons in layer 2/3 of mouse somatosensory barrel cortex.
Spiking of PV neurons was highest and most regular during REM sleep. Spiking decreased rapidly as mice awoke and remained low and irregular during wakefulness. VIP neurons were also most active during REM sleep, but their spiking was most irregular during NREM sleep. Although the average spiking of SST and pyramidal neurons did not vary across the sleep/wake cycle, a subset of pyramidal cells was highly active during REM sleep; the activity of these neurons decreased rapidly when mice awoke.
Spiking in approximately half of the neurons in each class varied across theta oscillations (5–9 Hz) with peak firing during the trough of the oscillation. In addition, spiking in a large proportion of PV and VIP neurons and about half of SST and pyramidal neurons showed modulation at delta frequencies (1–4 Hz). Finally, spiking of PV and pyramidal neurons increased, whereas spiking of SST neurons decreased, during spindles in non-REM sleep. In addition, the spiking of approximately half of VIP, PV, and pyramidal neurons was phase modulated during spindles, with VIP neurons firing later in the spindle oscillation than PV and pyramidal neurons.
These results indicate that inhibition in layer 2/3 of somatosensory cortex differs across the sleep/wake cycle. In particular, the activity of both PV neurons, which provide strong perisomatic inhibition to pyramidal cells, and VIP neurons, which target other interneurons, is highest during REM sleep. Despite the increase in average inhibition, a subset of pyramidal cells become highly active during REM sleep. Thus, VIP and PV neurons may work together to enable particular ensembles of pyramidal neurons to become active, potentially affecting memory consolidation and influencing dreams.
Footnotes
This Week in The Journal was written by Teresa Esch, Ph.D.
