The Dorsal Column Lesion Model of Spinal Cord Injury and Its Use in Deciphering the Neuron-Intrinsic Injury Response

Abstract

The neuron-intrinsic response to axonal injury differs markedly between neurons of the peripheral and central nervous system. Following a peripheral lesion, a robust axonal growth program is initiated, whereas neurons of the central nervous system do not mount an effective regenerative response. Increasing the neuron-intrinsic regenerative response would therefore be one way to promote axonal regeneration in the injured central nervous system. The large-diameter sensory neurons located in the dorsal root ganglia are pseudo-unipolar neurons that project one axon branch into the spinal cord, and, via the dorsal column to the brain stem, and a peripheral process to the muscles and skin. Dorsal root ganglion neurons are ideally suited to study the neuron-intrinsic injury response because they exhibit a successful growth response following peripheral axotomy, while they fail to do so after a lesion of the central branch in the dorsal column. The dorsal column injury model allows the neuron-intrinsic regeneration response to be studied in the context of a spinal cord injury. Here we will discuss the advantages and disadvantages of this model. We describe the surgical methods used to implement a lesion of the ascending fibers, the anatomy of the sensory afferent pathways and anatomical, electrophysiological, and behavioral techniques to quantify regeneration and functional recovery. Subsequently we review the results of experimental interventions in the dorsal column lesion model, with an emphasis on the molecular mechanisms that govern the neuron-intrinsic injury response and manipulations of these after central axotomy. Finally, we highlight a number of recent advances that will have an impact on the design of future studies in this spinal cord injury model, including the continued development of adeno-associated viral vectors likely to improve the genetic manipulation of dorsal root ganglion neurons and the use of tissue clearing techniques enabling 3D reconstruction of regenerating axon tracts. © 2018 The Authors. Developmental Neurobiology Published by Wiley Periodicals, Inc. Develop Neurobiol 00: 000-000, 2018.

Keywords: conditioning lesion; dorsal column lesion; dorsal root ganglia; regeneration-associated gene program; spinal cord injury.

Figure 1

Commonly used rodent DC lesion models. A schematic diagram of the rat spinal cord and common DC lesion paradigms. Injured areas are depicted with striped lines, together with instruments commonly used to perform the lesion. (A–C) Transection injuries of the spinal cord, illustrating in (A) lateral hemisection of the spinal cord, (B) dorsal hemisection of the spinal cord, and (C) bilateral transection of the DC (microscissors and scalpel depicted, other instruments are also used as summarized in Table 1 and Supporting Information, Table 2). Besides transection, the DC lesion can be implemented using forceps creating a crush injury. (D) Contusion or compression injury by dropping or placing a weight on the spinal cord in a controlled manner. (E) Severing individual superficial DC axons using a laser. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

Detailed illustration of DC lesion models for SCI. Schematic drawings of transverse sections of adult rat cervical (C7) spinal cords (modified from Watson et al., 2008) depicted in the striped areas the injuries to the spinal cord with (A) dorsal hemisection of the spinal cord, (B) lateral hemisection of the spinal cord, (C) complete bilateral DC transection, (D) bilateral DC aspiration, (E) spinal cord contusion, and (F) single DC axon transection injuries.

Figure 3

The anatomy of the dorsal root ganglia, the spinal cord and targets for neuron‐intrinsic experimental intervention strategies. Axons of DRG neurons bifurcate into two branches, one going into the periphery and the other going into the spinal cord. These axons relay information including heat, pain, and position from the body and project via the DC to the brainstem directly (the gracile nucleus, cuneate nucleus and the external cuneate nucleus) or indirectly via spinal neurons (not shown). Not shown: collaterals also innervate spinal cord grey matter in the segments around where they enter, and some collaterals descend caudally. Axon collaterals also innervate Clarke's nucleus in the thoracic cord. The arrows indicate targets for neuron‐intrinsic intervention to promote axonal growth and plasticity, with (A) introduction of pharmacological agents by injection into the SN, (B) CL of the SN, (C) viral vector delivery by injection into the DRG, (D) using transgenic animals, and (E) subcutaneous or intrathecal injection to deliver pharmacological agents (not illustrated). Key: gn = gracile nucleus, cn = cuneate nucleus, ecn = external cuneate nucleus, dc = dorsal column, cst = corticospinal tract, d = dorsal nucleus (Clarke's nucleus), rs = rubrospinal tract, lf = lateral funiculus, vf = ventral funiculus. [Color figure can be viewed at http://wileyonlinelibrary.com]