Spinal dI3 Circuits in the Modulation of Skilled and Corrective Locomotor Behaviours

Abstract

Many spinal neurons integrate sensory input to adapt a wide array of movements. Prior research has shown that spinal dorsal interneuron 3 (dI3) neurons receive cutaneous and proprioceptive sensory input and that they project to motor circuits, most notably through excitation of motoneurons. We sought to better delineate the spinal circuits within which dI3 neurons operated. To reach this objective, we tested motor performance of transgenic mice where dI3s could be chemogenetically silenced by the DREADD receptor, hM4Di. Chemogenetic inhibition of dI3s was concomitant with drug treatments that putatively inhibited (mecamylamine) or enhanced (L-acetyl carnitine) recurrent motoneuron excitation of central targets such as Renshaw cells. In addition to these two cholinergic modulators, saline was also administered as a separate treatment to act as a control. Each of these three separate treatments were administered in combination with either JHU37160, an hM4Di agonist to silence dI3s, or a saline control to maintain dI3 activity. These mice were then tested on three behavioural tasks. For performance related to fine motor control, balance, and coordination, mice were tested on a vibrating balance beam and horizontal ladder walking task. In both ladder and beam tasks, mice demonstrated an up to three-fold increase in the incidence of foot falls after dI3 inhibition with either L-acetyl carnitine or saline as a combined treatment. We also observed two- to three-fold increases in footfalls on the ladder and beam after mecamylamine administration, an antagonist of Renshaw cell excitation, in combination with dI3 silencing but also without dI3 silencing. Interestingly, ladder and beam tasks performed after a combined treatment that silenced both dI3s and Renshaw cell excitation did not appear to have a compounding effect in the observed motor deficits on these apparatuses; the increase in foot falls remained two to three-fold higher when comparing trials with and without dI3 silencing as combined with mecamylamine. Overall, we observed an increase in motor deficits and difficulty when crossing both the ladder and beam after dI3 silencing and mecamylamine administration. Additionally, mice were also tested on a treadmill apparatus wherein the hindlimb paw was either mechanically or electrically perturbed to attempt to trigger the stumbling corrective reaction. With L-acetyl carnitine and saline trials, dI3 silencing resulted in a significant reduction in the peak step height during a stumbling corrective response in comparison to trials without dI3 silencing. Conversely, the effects of dI3 inhibition and mecamylamine administration yielded a similar trend to the previous behavioural task, wherein step heights during the peak of the reflex were significantly reduced after dI3 silencing and/or mecamylamine administration compared to saline controls but were not statistically different when comparing trials with and without dI3 silencing if mecamylamine was also administered. While this reflex was not completely abolished, we observed significantly increased instances of failed attempts to clear the perturbation resulting in dragging and a significant blunting of the stumbling corrective response with both mecamylamine administration and dI3 silencing, both separately and together. Overall, our findings support the hypothesis that dI3s are involved in the activation of the stumbling corrective response, wherein proprioceptive and cutaneous signals may be used to modulate the activation of muscles during locomotion. While there remains much to be understood about dI3 spinal circuits, it appears that a lack of dI3 signaling inhibits skilled locomotor behaviours. Our results suggest that dI3s are more involved in skilled and corrective motor behaviours than previously understood, as we suggest that Renshaw cells and dI3s may operate together within a type of neural comparator module to actively monitor sensory signals and rapidly adjust motor outputs. Our experiments begin to identify the spinal circuits within which dI3s operate to shape motor activity.