Fig. 1: Anatomical and functional characterization of the bulbospongiosus motor neurons (BSM-MNs).

A Left panel: Fluorogold (FG) was injected into the bulbospongiosus muscle (BSM) of adult mice (N = 12), leading to labeling across the rostrocaudal lumbosacral spinal cord. Right panel: FG-positive cells (FG+, blue) within the lumbar spinal segment 6. Nissl stain (green) was used to identify the spinal cord segment based on the atlas⁹⁰. Scale bar 200 μm. Scale bar inset 20 μm. B Serial reconstruction of all labeled FG+ cells between lumbar segment 3 and sacral segment 2 revealed their distribution at the dorsomedial ventral spinal cord, close to the gray commissure. Upper panel: distribution of all labeled FG+ cells for one animal; middle panel: cell distribution within lumbar segment 6 (L6); lower panel: cell distribution within sacral segment 1 (S1). C Total FG+ cell numbers along the lumbosacral spinal cord. Different colored dots represent different animals (N = 12 mice). Most cells were found at L6 and S1 spinal segments (elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles). Animals that lacked cells in a given spinal segment were excluded from the violin plot for that specific segment. D Post-hoc immunohistochemical staining for the alpha MN markers osteopontin (purple), choline acetyltransferase (ChAT, orange) and the sensory input marker Vesicular Glutamate Transporter 1 (VGLUT1, green) revealed that all three markers are expressed by FG+ cells (blue) (N = 3 mice). Scale bar 200 μm. Scale bar inset 20 μm. E Percentages of FG+ cells expressing ChAT (97.7 ± 2.2%), VGLUT1 (100%) and osteopontin (81.95 ± 2.2%). F Cell size distribution of FG+ cells along the rostrocaudal lumbar spinal cord. Frequency histogram depicting number of MNs in each size bin and size of the MNs (binned in 20 μm² steps). G Left panel, experimental setup: For functional characterization, rAAV-CAG-ChR2 (ChR2, Channelrhodopsin 2) viral vectors were injected into the BSM of pups (P3-P6), later used for optogenetic stimulation when reaching adulthood. For optogenetic stimulation, a fiber was moved on top and along the rostral caudal lumbar spinal cord while monitoring muscle activity in the BSM and a leg muscle (Tibialis anterior, TA) using electromyogram (EMG). Right panel: rAAV-CAG-ChR2 expression (light blue) in L6 spinal segment (Nissl stain, purple). This experiment was performed in 10 mice. Scale bar 200 μm. Scale bar inset 20 μm. H Example BSM and TA EMG trace recorded while optogenetically stimulating above the L6 spinal segment. I Higher BSM EMG amplitudes were triggered when illuminating above the L6 spinal segment (mean amplitude 8.37 ± 1.15 mV), compared to illuminating above the L3 segment (mean amplitude 0.57 ± 0.2) (N = 10 mice). Two-tailed Student’s t-test p = 0.0000005 (elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles). Different colored dots represent different animals (I–M). J The laser power necessary to elicit BSM potentials was lower above the L6 spinal segment (mean threshold 11.2 ± 0.64 mW), compared to stimulating above the L3 segment (mean threshold 17 ± 0.97 mW) (N = 10 mice). Two-tailed Student’s t-test p = 0.00002 (elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles). K The onset of BSM activity was shorter with illumination above the L6 spinal segment (mean latency to EMG 7.16 ± 1.58 ms) when compared to the L3 segment (mean latency to EMG 23.48 ± 9.43 ms) (N = 10 mice). Two-tailed Student’s t-test p = 0.04 (elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles). L Laser stimulation (10 mW) at 5 Hz (upper panel), 10 Hz (middle panel) and 20 Hz (lower panel) reliably triggered short latency action potentials in a single BSM-MNs. Shortly after the single spikes, BSM EMG potentials were observed (insets) which themselves followed blue laser light applications. No EMG responses were observed in the TA muscle of the leg. M Latencies of triggered responses after laser stimulation are plotted for single BSM-MNs (MN; N = 7 mice; mean latency 4.6 ± 0.33 ms) and MN to EMG onset (MN-EMG; mean latency 2.25 ± 0.25 ms). Two-tailed Student’s t-test p = 0.001 (elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles). N Fidelity of spike and EMG activity are shown for the different frequencies of stimulation tested (dark gray: EMG fidelity; black: spike fidelity). Elements of violin plot: center line, median; box limits, upper (75) and lower (25) quartiles (N = 7 cells from 6 different mice).