The neurons that restore walking after paralysis.

A spinal cord injury interrupts pathways from the brain and brainstem that project to the lumbar spinal cord, leading to paralysis. Here we show that spatiotemporal epidural electrical stimulation (EES) of the lumbar spinal cord1-3 applied during neurorehabilitation4,5 (EESREHAB) restored walking in nine individuals with chronic spinal cord injury. This recovery involved a reduction in neuronal activity in the lumbar spinal cord of humans during walking. We hypothesized that this unexpected reduction reflects activity-dependent selection of specific neuronal subpopulations that become essential for a patient to walk after spinal cord injury. To identify these putative neurons, we modelled the technological and therapeutic features underlying EESREHAB in mice. We applied single-nucleus RNA sequencing6-9 and spatial transcriptomics10,11 to the spinal cords of these mice to chart a spatially resolved molecular atlas of recovery from paralysis. We then employed cell type12,13 and spatial prioritization to identify the neurons involved in the recovery of walking. A single population of excitatory interneurons nested within intermediate laminae emerged. Although these neurons are not required for walking before spinal cord injury, we demonstrate that they are essential for the recovery of walking with EES following spinal cord injury. Augmenting the activity of these neurons phenocopied the recovery of walking enabled by EESREHAB, whereas ablating them prevented the recovery of walking that occurs spontaneously after moderate spinal cord injury. We thus identified a recovery-organizing neuronal subpopulation that is necessary and sufficient to regain walking after paralysis. Moreover, our methodology establishes a framework for using molecular cartography to identify the neurons that produce complex behaviours.

Selecting single cell clustering parameter values using subsampling-based robustness metrics.

BACKGROUND: Generating and analysing single-cell data has become a widespread approach to examine tissue heterogeneity, and numerous algorithms exist for clustering these datasets to identify putative cell types with shared transcriptomic signatures. However, many of these clustering workflows rely on user-tuned parameter values, tailored to each dataset, to identify a set of biologically relevant clusters. Whereas users often develop their own intuition as to the optimal range of parameters for clustering on each data set, the lack of systematic approaches to identify this range can be daunting to new users of any given workflow. In addition, an optimal parameter set does not guarantee that all clusters are equally well-resolved, given the heterogeneity in transcriptomic signatures in most biological systems. RESULTS: Here, we illustrate a subsampling-based approach (chooseR) that simultaneously guides parameter selection and characterizes cluster robustness. Through bootstrapped iterative clustering across a range of parameters, chooseR was used to select parameter values for two distinct clustering workflows (Seurat and scVI). In each case, chooseR identified parameters that produced biologically relevant clusters from both well-characterized (human PBMC) and complex (mouse spinal cord) datasets. Moreover, it provided a simple "robustness score" for each of these clusters, facilitating the assessment of cluster quality. CONCLUSION: chooseR is a simple, conceptually understandable tool that can be used flexibly across clustering algorithms, workflows, and datasets to guide clustering parameter selection and characterize cluster robustness.

The organization of spinal neurons: Insights from single cell sequencing.

To understand how the spinal cord enacts complex sensorimotor functions, researchers have studied, classified, and functionally probed it's many neuronal populations for over a century. Recent developments in single-cell RNA-sequencing can characterize the gene expression signatures of the entire set of spinal neuron types and can simultaneously provide an unbiased view of their relationships to each other. This approach has revealed that the location of neurons predicts transcriptomic variability, as dorsal spinal neurons become highly distinct over development as ventral spinal neurons become less so. Temporal specification is also a major source of gene expression variation, subdividing many of the canonical embryonic lineage domains. Together, birthdate and cell body location are fundamental organizing features of spinal neuron diversity.

Molecular and spatial profiling of the paraventricular nucleus of the thalamus.

The paraventricular nucleus of the thalamus (PVT) is known to regulate various cognitive and behavioral processes. However, while functional diversity among PVT circuits has often been linked to cellular differences, the molecular identity and spatial distribution of PVT cell types remain unclear. To address this gap, here we used single nucleus RNA sequencing (snRNA-seq) and identified five molecularly distinct PVT neuronal subtypes in the mouse brain. Additionally, multiplex fluorescent in situ hybridization of top marker genes revealed that PVT subtypes are organized by a combination of previously unidentified molecular gradients. Lastly, comparing our dataset with a recently published single-cell sequencing atlas of the thalamus yielded novel insight into the PVT's connectivity with the cortex, including unexpected innervation of auditory and visual areas. This comparison also revealed that our data contains a largely non-overlapping transcriptomic map of multiple midline thalamic nuclei. Collectively, our findings uncover previously unknown features of the molecular diversity and anatomical organization of the PVT and provide a valuable resource for future investigations.

A cellular taxonomy of the adult human spinal cord.

The mammalian spinal cord functions as a community of cell types for sensory processing, autonomic control, and movement. While animal models have advanced our understanding of spinal cellular diversity, characterizing human biology directly is important to uncover specialized features of basic function and human pathology. Here, we present a cellular taxonomy of the adult human spinal cord using single-nucleus RNA sequencing with spatial transcriptomics and antibody validation. We identified 29 glial clusters and 35 neuronal clusters, organized principally by anatomical location. To demonstrate the relevance of this resource to human disease, we analyzed spinal motoneurons, which degenerate in amyotrophic lateral sclerosis (ALS) and other diseases. We found that compared with other spinal neurons, human motoneurons are defined by genes related to cell size, cytoskeletal structure, and ALS, suggesting a specialized molecular repertoire underlying their selective vulnerability. We include a web resource to facilitate further investigations into human spinal cord biology.

Intersectional genetic tools to study skilled reaching in mice.

Reaching to grasp is an evolutionarily conserved behavior and a crucial part of the motor repertoire in mammals. As it is studied in the laboratory, reaching has become the prototypical example of dexterous forelimb movements, illuminating key principles of motor control throughout the spinal cord, brain, and peripheral nervous system. Here, we (1) review the motor elements or phases that comprise the reach, grasp, and retract movements of reaching behavior, (2) highlight the role of intersectional genetic tools in linking these movements to their neuronal substrates, (3) describe spinal cord cell types and their roles in skilled reaching, and (4) how descending pathways from the brain and the sensory systems contribute to skilled reaching. We emphasize that genetic perturbation experiments can pin-point the neuronal substrates of specific phases of reaching behavior.

Single cell atlas of spinal cord injury in mice reveals a pro-regenerative signature in spinocerebellar neurons.

After spinal cord injury, tissue distal to the lesion contains undamaged cells that could support or augment recovery. Targeting these cells requires a clearer understanding of their injury responses and capacity for repair. Here, we use single nucleus RNA sequencing to profile how each cell type in the lumbar spinal cord changes after a thoracic injury in mice. We present an atlas of these dynamic responses across dozens of cell types in the acute, subacute, and chronically injured spinal cord. Using this resource, we find rare spinal neurons that express a signature of regeneration in response to injury, including a major population that represent spinocerebellar projection neurons. We characterize these cells anatomically and observed axonal sparing, outgrowth, and remodeling in the spinal cord and cerebellum. Together, this work provides a key resource for studying cellular responses to injury and uncovers the spontaneous plasticity of spinocerebellar neurons, uncovering a potential candidate for targeted therapy.

A harmonized atlas of mouse spinal cord cell types and their spatial organization.

Single-cell RNA sequencing data can unveil the molecular diversity of cell types. Cell type atlases of the mouse spinal cord have been published in recent years but have not been integrated together. Here, we generate an atlas of spinal cell types based on single-cell transcriptomic data, unifying the available datasets into a common reference framework. We report a hierarchical structure of postnatal cell type relationships, with location providing the highest level of organization, then neurotransmitter status, family, and finally, dozens of refined populations. We validate a combinatorial marker code for each neuronal cell type and map their spatial distributions in the adult spinal cord. We also show complex lineage relationships among postnatal cell types. Additionally, we develop an open-source cell type classifier, SeqSeek, to facilitate the standardization of cell type identification. This work provides an integrated view of spinal cell types, their gene expression signatures, and their molecular organization.

Confronting false discoveries in single-cell differential expression.

Differential expression analysis in single-cell transcriptomics enables the dissection of cell-type-specific responses to perturbations such as disease, trauma, or experimental manipulations. While many statistical methods are available to identify differentially expressed genes, the principles that distinguish these methods and their performance remain unclear. Here, we show that the relative performance of these methods is contingent on their ability to account for variation between biological replicates. Methods that ignore this inevitable variation are biased and prone to false discoveries. Indeed, the most widely used methods can discover hundreds of differentially expressed genes in the absence of biological differences. To exemplify these principles, we exposed true and false discoveries of differentially expressed genes in the injured mouse spinal cord.

Cell type prioritization in single-cell data.

We present Augur, a method to prioritize the cell types most responsive to biological perturbations in single-cell data. Augur employs a machine-learning framework to quantify the separability of perturbed and unperturbed cells within a high-dimensional space. We validate our method on single-cell RNA sequencing, chromatin accessibility and imaging transcriptomics datasets, and show that Augur outperforms existing methods based on differential gene expression. Augur identified the neural circuits restoring locomotion in mice following spinal cord neurostimulation.

GDF3 is a BMP inhibitor that can activate Nodal signaling only at very high doses.

Within the TGF-beta superfamily, there are approximately forty ligands divided into two major branches: the TGF-beta/Activin/Nodal ligands and the BMP/GDF ligands. We studied the ligand GDF3 and found that it inhibits signaling by its co-family members, the BMPs; however, GDF3 has been described by others to have Nodal-like activity. Here, we show that GDF3 can activate Nodal signaling, but only at very high doses and only upon mRNA over-expression. In contrast, GDF3 inhibits BMP signaling upon over-expression of GDF3 mRNA, as recombinant protein, and regardless of its dose. We therefore further characterized the mechanism through which GDF3 protein acts as a specific BMP inhibitor and found that the BMP inhibitory activity of GDF3 resides redundantly in the unprocessed, predominant form and in the mature form of the protein. These results confirm and extend the activity that we described for GDF3 and illuminate the experimental basis for the different observations of others. We suggest that GDF3 is either a bi-functional TGF-beta ligand, or, more likely, that it is a BMP inhibitor that can artificially activate Nodal signaling under non-physiological conditions.

Cerebellospinal Neurons Regulate Motor Performance and Motor Learning.

To understand the neural basis of behavior, it is important to reveal how movements are planned, executed, and refined by networks of neurons distributed throughout the nervous system. Here, we report the neuroanatomical organization and behavioral roles of cerebellospinal (CeS) neurons. Using intersectional genetic techniques, we find that CeS neurons constitute a small minority of excitatory neurons in the fastigial and interpositus deep cerebellar nuclei, target pre-motor circuits in the ventral spinal cord and the brain, and control distinct aspects of movement. CeS neurons that project to the ipsilateral cervical cord are required for skilled forelimb performance, while CeS neurons that project to the contralateral cervical cord are involved in skilled locomotor learning. Together, this work establishes CeS neurons as a critical component of the neural circuitry for skilled movements and provides insights into the organizational logic of motor networks.

Decoding Cell Type Diversity Within the Spinal Cord.

To understand fundamental mechanisms of mammalian spinal cord function, it is necessary to reveal the diverse array of constituent spinal "cell types" - populations that can be consistently identified because they share a unique and cohesive set of characteristics. Many parameters can contribute to the definition of a spinal cord cell type, including location, morphology, lineage, electrophysiological properties, circuit features, gene expression patterns, and behavioral contribution. While it is not necessary for all of these features to align completely at all times to identify an individual cell type, a correlation of these characteristics paints a rich portrait of cell identity. This review will summarize recent advances in the identification of mammalian spinal cord neuronal cell types and will highlight the power of transcriptional profiling to identify and characterize the cell types of the spinal cord.

Isolation of Adult Spinal Cord Nuclei for Massively Parallel Single-nucleus RNA Sequencing.

Probing an individual cell's gene expression enables the identification of cell type and cell state. Single-cell RNA sequencing has emerged as a powerful tool for studying transcriptional profiles of cells, particularly in heterogeneous tissues such as the central nervous system. However, dissociation methods required for single cell sequencing can lead to experimental changes in the gene expression and cell death. Furthermore, these methods are generally restricted to fresh tissue, thus limiting studies on archival and bio-bank material. Single nucleus RNA sequencing (snRNA-Seq) is an appealing alternative for transcriptional studies, given that it accurately identifies cell types, permits the study of tissue that is frozen or difficult to dissociate, and reduces dissociation-induced transcription. Here, we present a high-throughput protocol for rapid isolation of nuclei for downstream snRNA-Seq. This method enables isolation of nuclei from fresh or frozen spinal cord samples and can be combined with two massively parallel droplet encapsulation platforms.

Massively Parallel Single Nucleus Transcriptional Profiling Defines Spinal Cord Neurons and Their Activity during Behavior.

To understand the cellular basis of behavior, it is necessary to know the cell types that exist in the nervous system and their contributions to function. Spinal networks are essential for sensory processing and motor behavior and provide a powerful system for identifying the cellular correlates of behavior. Here, we used massively parallel single nucleus RNA sequencing (snRNA-seq) to create an atlas of the adult mouse lumbar spinal cord. We identified and molecularly characterized 43 neuronal populations. Next, we leveraged the snRNA-seq approach to provide unbiased identification of neuronal populations that were active following a sensory and a motor behavior, using a transcriptional signature of neuronal activity. This approach can be used in the future to link single nucleus gene expression data with dynamic biological responses to behavior, injury, and disease.

Graded Arrays of Spinal and Supraspinal V2a Interneuron Subtypes Underlie Forelimb and Hindlimb Motor Control.

The spinal cord contains neural networks that enable regionally distinct motor outputs along the body axis. Nevertheless, it remains unclear how segment-specific motor computations are processed because the cardinal interneuron classes that control motor neurons appear uniform at each level of the spinal cord. V2a interneurons are essential to both forelimb and hindlimb movements, and here we identify two major types that emerge during development: type I neurons marked by high Chx10 form recurrent networks with neighboring spinal neurons and type II neurons that downregulate Chx10 and project to supraspinal structures. Types I and II V2a interneurons are arrayed in counter-gradients, and this network activates different patterns of motor output at cervical and lumbar levels. Single-cell RNA sequencing (RNA-seq) revealed type I and II V2a neurons are each comprised of multiple subtypes. Our findings uncover a molecular and anatomical organization of V2a interneurons reminiscent of the orderly way motor neurons are divided into columns and pools.

Satb2 Is Required for the Development of a Spinal Exteroceptive Microcircuit that Modulates Limb Position.

Motor behaviors such as walking or withdrawing the limb from a painful stimulus rely upon integrative multimodal sensory circuitry to generate appropriate muscle activation patterns. Both the cellular components and the molecular mechanisms that instruct the assembly of the spinal sensorimotor system are poorly understood. Here we characterize the connectivity pattern of a sub-population of lamina V inhibitory sensory relay neurons marked during development by the nuclear matrix and DNA binding factor Satb2 (ISR(Satb2)). ISR(Satb2) neurons receive inputs from multiple streams of sensory information and relay their outputs to motor command layers of the spinal cord. Deletion of the Satb2 transcription factor from ISR(Satb2) neurons perturbs their cellular position, molecular profile, and pre- and post-synaptic connectivity. These alterations are accompanied by abnormal limb hyperflexion responses to mechanical and thermal stimuli and during walking. Thus, Satb2 is a genetic determinant that mediates proper circuit development in a core sensory-to-motor spinal network.