Molecular Logic of Spinocerebellar Tract Neuron Diversity and Connectivity

Myungin Baek, Vilas Menon, Thomas Jessell, Adam Hantman, Jeremy Dasen

Preprint posted on January 09, 2019

Article now published in Cell Reports at

What cultivates the diversity of spinocerebellar tract neurons? Transcriptomic and genetic study suggests Hox genes to be the key.

Selected by Yen-Chung Chen

Background and context

Coordination of motions depends on communication between the peripheral sensory system and the central nervous system, and spinocerebellar tract neurons are central to this communication. Spinocerebellar tract neurons relay proprioceptive sensory information about muscle tension and position from the body to the cerebellum. When this communication is defective, although motions are maintained, the ability to predict and correct errors is damaged.

Previous retrograde tracing studies have shown that different subpopulations of spinocerebellar tract neurons exist along the rostrocaudal axis, which receive inputs from specific muscle groups that are often functionally antagonistic. This observation leads to the hypothesis that spinocerebellar tract neurons might integrate the information from each muscle pair and transmit the information to the cerebellum for further processing.

While the anatomy and connectivity of spinocerebellar tract neurons have been described, the underlying molecular basis of their diversity and connectivity remains largely unknown. Understanding how spinocerebellar tract neurons are specified and form synapses with different subsets of peripheral sensory neurons could tell us more about how neurons find the right target in a distant and crowded neural nucleus or column and form synapses with the right partner. Furthermore, identification of critical developmental genes may provide a more precise way to manipulate spinocerebellar tract neurons to reveal their role in the motor circuit.

Key findings

In this study, Baek et al. first systematically characterized the anatomy of spinocerebellar tract neurons projecting to the cerebellum by retrograde labeling, identifying eight spatially distinct subpopulations, and then profiled the transcriptomic difference between cervical and thoracic spinocerebellar tract neurons. Transcriptomic analysis revealed differences in the expression of ion channels, neurotransmitter transporters, and transcription factors, suggesting that the spinocerebellar tract neurons in different segments obtain distinct transcriptional programs. The segmental difference is consistent with the early morphogen patterning along the rostrocaudal axis and reminiscent of subtype specification of spinal motor neurons. To better characterize the heterogeneity of spinocerebellar tract neurons within each segment, the authors performed single cell RNA-seq and identified eight transcriptomically distinct clusters, among which one novel subtype, marked by Shox2, Scip, and Fam19A4, was examined in vivo and detected as a subpopulation in the caudal cervical segment.

The authors then asked how these molecularly distinct subtypes of spinocerebellar tract neurons are specified. Inspired by the prominent segmental difference, the authors hypothesized that Hox genes might be the key regulators of neuronal subtype determination, similar to the case of spinal motor neurons and interneurons [1][2]. Indeed, different spinocerebellar tract neurons express a distinct combination of Hox genes. With the loss of function of Hoxc9, a critical regulator of thoracic segment for motor and interneurons, Clarke’s column, a subpopulation of spinocerebellar tract neurons, failed to form in the thoracic segment. On the other hand, the other remaining spinocerebellar tract neurons in the thoracic segment stopped expressing their marker genes and started to express a combination similar to caudal cervical spinocerebellar tract neurons (Figure 1). Regarding circuit formation, thoracic spinocerebellar tract neurons received peripheral sensory innervations that are otherwise exclusive for cervical neurons in the lack of Hoxc9. The changes in gene expression and connectivity upon Hoxc9 loss of function are consistent with a conversion of cell identity from thoracic to cervical spinocerebellar tract neurons.

Figure 1. Spinocerebellar tract neurons within different rostrocaudal segments differ in spatial organization and molecular signature, which is regulated by the expression of Hox genes.


In short, Baek et al. systematically characterized spinocerebellar tract neurons with a transcriptomic approach and identified segmentally distinct subtypes that are governed by Hox gene expression.

Why I like this preprint

Hox genes are critical for spinal motor neurons to further specify into different subpopulations and achieve precise connection with individual muscle groups. Recently, a similar paradigm has been shown to be adopted by some spinal interneurons, making it tempting to speculate that Hox genes are the general hub to integrate information from different morphogen gradients and to define the neural diversity in the spinal cord. If that is the case, how the seemingly ubiquitous expression of Hox genes within a segment can specify different neural types becomes an intriguing question, and this question might be answered by interrogating the crosstalk between cell type-specifying transcription factors and Hox genes. In this preprint, Baek et al. elegantly depicted how the spatial and molecular identities of spinocerebellar tract neurons are governed by Hox genes, demonstrating that Hox regulation is indeed a shared regulatory feature within the spinal cord. Additionally, the characterization of spinocerebellar tract neurons provides a novel context for future study of how fate specifying programs and Hox genes synergistically give rise to neural diversity.

Open question

  1. The spatial distribution of spinocerebellar tract neurons differs segmentally on both ventrodorsal and mediolateral axis, do they come from the same progenitor zone? If not, can single cell RNA-seq give us a hint about which layers spinocerebellar tract neurons come from?
  2. In contrast to spinal motor neurons [3] and V1 interneurons [2], where no prominent change in neuron number was reported with Hoxc9 loss of function, the decrease of thoracic spinocerebellar tract number is intriguing. Is the cell number change as prominent before developmental programmed cell death as it is postnatally? Or alternatively, do thoracic spinocerebellar tract neurons switch to other fates in the absence of Hoxc9?


  1. Dasen, J. S. & Jessell, T. M. Chapter Six Hox Networks and the Origins of Motor Neuron Diversity. Current Topics in Developmental Biology 88, (Elsevier Inc., 2009).
  2. Sweeney, L. B. et al. Origin and Segmental Diversity of Spinal Inhibitory Interneurons. Neuron 97, 341–355.e3 (2018).
  3. Dasen, J. S., Liu, J. P. & Jessell, T. M. Motor neuron columnar fate imposed by sequential phases of Hox-c activity. Nature 425, 926–933 (2003).

Tags: scrna-seq

Posted on: 8th April 2019 , updated on: 11th April 2019

Read preprint (No Ratings Yet)

  • Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here