Now and then: Development of spinal Shox2 neurons

Ever since the conception of neurophysiology as a field, scientists have been fascinated to uncover what makes neurons exhibit their characteristic patterns of activity. This question is especially interesting if the neurons can produce self-sustaining bursts of action potentials even when isolated from the network, a property called rhythmogenesis. Such rhythmogenic neurons are thought to drive essential body functions like breathing, chewing and digestion that require the activation of muscles in a persistent cycle. Rhythmogenic neurons are also thought to control repetitive, voluntary movements like walking, running, flying or swimming (locomotion). Rhythmogenic neurons are present in many different animals, and seminal work in invertebrates has provided significant insight into the ionic mechanisms that sustain oscillatory activity (Marder & Bucher, 2001). However rhythmogenesis in vertebrates, especially during locomotion, remains more elusive. In vertebrates networks controlling voluntary movements are housed in the spinal cord. In 2013 a pioneering study from the Kiehn Lab demonstrated the presence of rhythmogenic neurons in the lumbar spinal cord in mice that could potentially drive rhythmic leg movements. These neurons expressed the transcription factor Shox2, exhibited rhythmic bursting activity (Fig. 1A) during fictive locomotion and when silenced decreased the stepping rhythm, showing a direct effect on locomotor control (Dougherty et al., 2013). Since then extensive work from the Dougherty Lab has characterized these neurons and their connections, especially at neonatal stages (Ha & Dougherty, 2018). Circuits, at these stages, though present, are still immature. It is not known whether they undergo significant modifications during development to support the full repertoire of motor behaviours in adults. In both neonates and adults Shox2 neurons are rhythmogenic and exhibit bursting activity. However in neonates Shox2 neurons are heavily interconnected via direct contacts (gap junctions) that facilitate oscillatory activity. These contacts are progressively lost with age, implying that, in adults, Shox2 neurons are more likely to be reliant on cell-intrinsic mechanisms to sustain rhythmic bursts of activity. To investigate whether this is indeed the case and how adult Shox2 neurons maintain rhythmogenesis, Singh et al. performed a thorough developmental profile of cellular properties of Shox2 neurons at different stages in mice. The authors used intracellular electrodes to record ionic currents from fluorescently labelled Shox2 neurons (Fig. 1A) in the lumbar spinal cord slice preparation at three different developmental stages – neonatal, juvenile and adult. First the authors compared the properties of these neurons at rest like cell sizes and activation thresholds. Although there were subtle differences in activation threshold between neonates and adults, most other properties were comparable, suggesting these neurons overall do not undergo substantial transformations and additionally may be modifying very specific ion channels. Therefore next the authors used clever voltage manipulations and pharmacological drugs to isolate distinct ionic currents and compared their strength across development. The bursting activity consists of up and down states (Fig. 1A, right). Previous studies have shown that successful bursting requires both inward and outward currents (Harris-Warrick, 2010). Positive ionic currents flowing into the neuron help it stay in the up state, whereas outward currents help terminate the burst, allowing a fresh one to start. Consistent with this the authors observed an increase in strength for both inward (persistent inward currents, PIC) and outward (M current) currents in adults compared to neonates. To validate these findings the authors also performed in situ hybridization for ion channel components that contribute to PIC. Indeed they detected the robust expression of voltage-gated sodium and calcium channel components in Shox2 neurons at all developmental stages, confirming their physiological findings. Together these results indicate that adult Shox2 neurons significantly increased distinct ionic currents that could potentially help maintain bursting activity in the absence of direct connections. To test whether these currents indeed help bursting, the authors next turned to computational tools. First the authors combined the total dataset of resting cell properties (size, activation thresholds, etc.) and active currents (PIC, M current, etc.) across different stages and fed it to an unbiased clustering analysis to check how different neonatal and adult neurons were. Based only on differences in cell properties, the program yielded eight distinct neuron clusters, the first three of which had only neonatal neurons and the remaining five clusters comprised juvenile and adult neurons. This indicated that neonatal neurons were significantly different from juvenile and adult neurons. Furthermore the authors then built a computational model reproducing the electrophysiology results at different stages. The model predicted that specifically increasing the inward current, PIC, would result in bursting in adult neurons. The authors experimentally supported this prediction, thereby confirming that, indeed, this current is a critical contributor for rhythmogenesis. Together these results show that adult Shox2 neurons, in the absence of electrical coupling, increase distinct cell-intrinsic properties to support rhythmic bursting (Fig. 1B). This is a unique example of how the same circuit can utilize different mechanisms to achieve the same end goal (bursting) at different developmental stages. Interestingly oscillatory activity is not special to motor networks but is also observed in various brain regions that participate in fundamentally different functions. These findings therefore can be highly insightful in understanding rhythmogenesis anywhere in the nervous system and propel future enquiry to compare and contrast these different solutions. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None declared. M.S.: conception or design of the work, drafting the work or revising it critically for important intellectual content, final approval of the version to be published and agreement to be accountable for all aspects of the work. None.