dSC1 and SPINE enable cell-type-resolved connectomic reconstruction of the spinal dorsal horn, revealing sensory neuron synaptic organization and modality-specific inhibition

The work provides two connectomic tools for studying spinal dorsal horn somatosensory circuits: dSC1, a large-scale, synapse-resolved EM dataset of the mouse dorsal horn, and SPINE, a deep-learning pipeline that identifies major sensory neuron subtypes directly from EM images. How sensory neuron synapses are organized into spinal cord circuitry was previously poorly understood; this work brings cell-type-resolved connectomic reconstruction to the dorsal horn, the first site of central somatosensory processing. Based on the large-scale, synapse-resolved EM dataset and the deep-learning classification pipeline described in the abstract, with SPINE applied to the dSC1 dataset.

Applying SPINE to dSC1 revealed marked heterogeneity in synaptic architecture across physiologically distinct sensory neuron subtypes: low-threshold mechanoreceptors form substantially more synapses and receive far more GABAergic presynaptic inhibition than high-threshold mechanoreceptors, thermoreceptors and nociceptors. It links synapse counts and inhibitory input strength to specific sensory subtypes, indicating that different modalities have distinct synaptic organization already at the first central processing stage. From cell-type-resolved reconstruction and synapse comparisons on the dSC1 dataset; the abstract gives directional differences without specific values.

Cell-type-resolved circuit reconstruction identified five populations of spinal inhibitory neurons (IhNs) that form synapses onto sensory neuron axons, each preferentially driven by and presynaptically inhibiting a single sensory subtype, a homotypic, modality-specific inhibitory motif; most of these IhNs also feedforwardly inhibit other sensory pathways, a heterotypic, cross-modality motif. It proposes a dual-mode logic of sensory-evoked inhibition, combining modality-specific inhibition with cross-modality feedforward inhibition that shapes modality-specific signaling at the earliest stage of somatosensory processing. Based on cell-type-resolved circuit reconstruction on dSC1 identifying five IhN populations and their connectivity preferences; the abstract does not give quantitative proportions for each population.

This work targets neuroscientists studying spinal dorsal horn somatosensory circuits, in settings that use the mouse as a model and EM connectomics as the approach: dSC1 provides the synapse-resolved data foundation, and SPINE makes major sensory neuron subtypes identifiable directly from EM images, supporting cell-type-resolved circuit reconstruction. Its conclusions concern the dorsal horn as the first site of central somatosensory processing, describing synaptic and inhibitory differences between low-threshold mechanoreceptors and high-threshold mechanoreceptors, thermoreceptors and nociceptors, as well as homotypic and heterotypic inhibitory motifs among five IhN populations. For researchers hoping to extend similar analyses to other sensory modalities, other spinal segments, or other species, this toolset and framework provide a reusable starting point.

The currently visible text is the abstract and the competing interest statement, without figures, specific synapse counts, statistical tests, or quantitative proportions for the five IhN populations, so the effect sizes and robustness of the reported differences cannot be assessed. The specific numerical ranges behind the finding that low-threshold mechanoreceptors form more synapses and receive stronger GABAergic presynaptic inhibition, and the distribution of homotypic versus heterotypic inhibitory motifs across the five IhN populations, still need to be confirmed in the main figures. In addition, the conclusions are based on the mouse spinal dorsal horn; when extending to other species, spinal segments, or brain regions, the generalization ability of SPINE's classification and the range of subtypes covered by dSC1 are open questions worth watching.