Central nervous system lesion triggers inappropriate pathway choice in adult vertebrate system

Survival and Axonal Outgrowth of the Mauthner Cell Following Spinal Cord Crush Does Not Drive Post-injury Startle Responses

A pair of Mauthner cells (M-cells) can be found in the hindbrain of most teleost fish, as well as amphibians and lamprey. The axons of these reticulospinal neurons cross the midline and synapse on interneurons and motoneurons as they descend the length of the spinal cord. The M-cell initiates fast C-type startle responses (fast C-starts) in goldfish and zebrafish triggered by abrupt acoustic/vibratory stimuli. Starting about 70 days after whole spinal cord crush, less robust startle responses with longer latencies manifest in adult goldfish, Carassius auratus. The morphological and electrophysiological identifiability of the M-cell provides a unique opportunity to study cellular responses to spinal cord injury and the relation of axonal regrowth to a defined behavior. After spinal cord crush at the spinomedullary junction about one-third of the damaged M-axons of adult goldfish send at least one sprout past the wound site between 56 and 85 days postoperatively. These caudally projecting sprouts follow a more lateral trajectory relative to their position in the fasciculus longitudinalis medialis of control fish. Other sprouts, some from the same axon, follow aberrant pathways that include rostral projections, reversal of direction, midline crossings, neuromas, and projection out the first ventral root. Stimulating M-axons in goldfish that had post-injury startle behavior between 198 and 468 days postoperatively resulted in no or minimal EMG activity in trunk and tail musculature as compared to control fish. Although M-cells can survive for at least 468 day (∼1.3 years) after spinal cord crush, maintain regrowth, and elicit putative trunk EMG responses, the cell does not appear to play a substantive role in the emergence of acoustic/vibratory-triggered responses. We speculate that aberrant pathway choice of this neuron may limit its role in the recovery of behavior and discuss structural and functional properties of alternative candidate neurons that may render them more supportive of post-injury startle behavior.

■ Review : The Mauthner Cell: What Has it Taught us?

The Mauthner cell (M-cell) is one of the few identifiable neurons in the vertebrate central nervous system. The ability to locate the M-cell, along with its inputs and outputs, has resulted in important findings in diverse areas of neurobiology including the molecular biology of neurons, synaptic and systems physiology, behavior, development, and neuroethology. The review provides a brief overview of the M-cell and then focuses on recent studies applying state-of-the-art techniques to address new issues and revisit old ones. One advantage of this preparation is the ability to conduct multidisciplinary studies from the subcellular to behavioral levels. For example, studies of activity-dependent changes in the strength of mixed electrotonic and chemical synapses on the M-cell's lateral dendrite in vivo have been correlated with changes in the probability of eliciting a fast startle response initiated by the M-cell and its associated circuits. Similarly, it is now possible to image the activity of the M-cell and its homologs while observing motor behavior in zebrafish larvae. These approaches will provide direct tests of the functional properties of complex neural networks. Moreover, molecular mechanisms that underlie neuronal development can be tested directly with this neuron and its segmental homologs, because these cells occur in singular pairs at defined locations. Finally, after spinal cord injury, the M-cell's axon regenerates, but does not follow its original course, and the startle response gradually recovers. The accessibility of the M-cell system offers the promise that strategies employed in restoring the function of a neural network will be revealed. Thus, we anticipate that the M-cell system will become a favored preparation for multidisciplinary studies on the neuronal basis of behavior and the recovery of behavior after injury. NEUROSCIENTIST 6:26-38, 2000

Spinal cord repair in regeneration-competent vertebrates: adult teleost fish as a model system

Spinal cord injuries in mammals, including humans, have devastating long-term consequences. Despite substantial research, therapeutic approaches developed in mammalian model systems have had limited success to date. An alternative strategy in the search for treatment of spinal cord lesions is provided by regeneration-competent vertebrates. These organisms, which include fish, urodele amphibians, and certain reptiles, have a spinal cord very similar in structure to that of mammals, but are capable of spontaneous structural and functional recovery after spinal cord injury. The present review aims to provide an overview of the current status of our knowledge of spinal cord regeneration in one of these groups, teleost fish. The findings are discussed from a comparative perspective, with reference to other taxa of regeneration-competent vertebrates, as well as to mammals.

Differences in the regenerative response of neuronal cell populations and indications for plasticity in intraspinal neurons after spinal cord transection in adult zebrafish

In zebrafish, the capacity to regenerate long axons varies among different populations of axotomized neurons after spinal cord transection. In specific brain nuclei, 84-92% of axotomized neurons upregulate expression of the growth-related genes GAP-43 and L1.1 and 32-51% of these neurons regrow their descending axons. In contrast, 16-31% of spinal neurons with axons ascending to the brainstem upregulate these genes and only 2-4% regrow their axons. Dorsal root ganglion (DRG) neurons were not observed to regrow their ascending axons or to increase expression of GAP-43 mRNA. Expression of L1.1 mRNA is high in unlesioned and axotomized DRG neurons. In the lesioned spinal cord, expression of growth-related molecules is increased in a substantial population of non-axotomized neurons, suggesting morphological plasticity in the spinal-intrinsic circuitry. We propose that locomotor recovery in spinal-transected adult zebrafish is influenced less by recovery of ascending pathways, but more by regrowth of descending tracts and rearrangement of intraspinal circuitry.

Recovery of C-starts, equilibrium and targeted feeding after whole spinal cord crush in the adult goldfish Carassius auratus

Central nervous system neurons of many adult teleost fish are capable of regrowth across spinal cord lesions, which may result in behavioral recovery of swimming. Since there have been few, if any, studies that examine the return of behaviors other than swimming, we provide a quantitative analysis of the recovery of C-starts that occur in adult goldfish after spinal cord injury. In addition, we include a qualitative analysis of the return of targeted feeding and equilibrium. Whole spinal cord crushes near the junction of the brain and spinal cord [spinomedullary level (SML)] were made in 45 experimental fish. Eight sham-operated goldfish served as controls for the effects of the surgery procedures alone. After spinal cord crush and recovery from the anesthetic, experimental fish lay on their sides with no movement caudal to the wound. The fish were monitored for the return of behaviors for up to 190 days postoperatively. Twenty-five fish survived the course of this study. Of these fish, 12 regained equilibrium and C-starts, two regained equilibrium but not C-starts, and 11 did not regain equilibrium (one of these did display a C-start). Twenty-two of the 25 experimental fish that survived the 190 days were able to target food from the water surface. Quantitative analysis of recovered C-starts in this study revealed that the probability of eliciting the response is reduced, that latencies from stimulus to response are longer and that movement parameters (i.e. angles, distance and velocity) are reduced compared with those of sham-operated control animals for up to 190 days postoperatively. The recovery of C-starts, equilibrium and targeted feeding was due to re-growth across the wound site, since re-crushing the spinal cord at the SML resulted in the loss of these behaviors. Mauthner cells are known to initiate C-starts in goldfish. Since the majority of M-axons that regrow across a crush wound associate with an inappropriate pathway (i.e. the first ventral root), it is unlikely that these cells play a major role in the return of C-starts. We propose that regeneration of Mauthner cell homologues across the wound site is responsible for the recovery of most C-starts. The identifiability of the M-cell and its homologues provides a unique opportunity to analyze the mechanisms underlying behavioral recovery at the cellular level.

Regenerating descending axons preferentially reroute to the gray matter in the presence of a general macrophage/microglial reaction caudal to a spinal transection in adult zebrafish

We analyzed pathway choices of regenerating, mostly supraspinal, descending axons in the spinal cord of adult zebrafish and the cellular changes in the spinal cord caudal to a lesion site after complete spinal transection. Anterograde tracing (by application of the tracer rostral to the spinal lesion site) showed that significantly more descending axons (74%) regenerated in the spinal gray matter of the caudal spinal cord than would be expected from random growth. Retrograde tracing (by application of the tracer caudal to the spinal lesion site) showed that, rostral to the lesion, most of these axons (80%) extended into the major white matter tracts. Thus, ventral descending tracts often were devoid of labeled axons caudal to a spinal lesion but contained many axons rostral to the lesion in the same animals, indicating a pathway switch of descending axons from the white matter to the gray matter. Ascending axons of spinal neurons were not observed regrowing to the rostral tracer application site; therefore, they most likely did not contribute to the axonal populations analyzed. A macrophage/microglia response within 2 days of spinal cord transection, along with phagocytosis of myelin, was observed caudal to the transection by immunohistochemistry and electron microscopy. Nevertheless, caudal to the lesion, descending tracts in the white matter were filled with myelin debris during the time of axonal regrowth, at least up to 6 weeks postlesion. We suggest that the spontaneous regeneration of axons of supraspinal origin after spinal cord transection in adult zebrafish may be due in part to the axons' ability to negotiate novel pathways in the spinal cord gray matter.

Adult opossums (Didelphis virginiana) demonstrate near normal locomotion after spinal cord transection as neonates

When the thoracic spinal cord of the North American opossum (Didelphis virginiana) is transected on postnatal day (PD) 5, the site of injury becomes bridged by histologically recognizable spinal cord and axons which form major long tracts grow through the lesion. In the present study we asked whether opossums lesioned on PD5 have normal use of the hindlimbs as adults and, if so, whether that use is dependent upon axons which grow through the lesion site. The thoracic spinal cord was transected on PD5 and 6 months later, hindlimb function was evaluated using the Basso, Beattie, and Bresnahan (BBB) locomotor scale. All animals supported their weight with the hindlimbs and used their hindlimbs normally during overground locomotion. In some cases, the spinal cord was retransected at the original lesion site or just caudal to it 6 months after the original transection and paralysis of the hindlimbs ensued. Surprisingly, however, these animals gradually recovered some ability to support their weight and to step with the hindlimbs. Similar recovery was not seen in animals transected only as adults. In order to verify that descending axons which grew through the lesion during development were still present in the adult animal, opossums subjected to transection of the thoracic cord on PD5 were reoperated and Fast blue was injected several segments caudal to the lesion. In all cases, neurons were labeled rostral to the lesion in each of the spinal and supraspinal nuclei labeled by comparable injections in unlesioned, age-matched controls. The results of orthograde tracing studies indicated that axons which grew through the lesion innervated areas that were appropriate for them.

Early development and developmental plasticity of the fasciculus gracilis in the North American opossum (Didelphis virginiana)

The first objective of the present study was to ask when axons of the fasciculus gracilis reach the nucleus gracilis in the North American opossum (Didelphis virginiana). When Fast Blue (FB) was injected into the lumbar cord on postnatal day (PD) 1 and the pups were killed 2 days later, labeled axons were present within a distinct fasciculus gracilis at thoracic and cervical levels of the cord. When comparable injections were made at PD3 or 5 and the pups were allowed to survive for the same time period, a few labeled axons could be followed to the caudal medulla where they were located dorsal to the presumptive nucleus gracilis. In order to verify these observations and to determine if any of the axons which innervate the nucleus gracilis early in development originate within dorsal root ganglia, we also employed cholera toxin conjugated to horseradish peroxidase (CT-HRP) to label dorsal root axons transganglionically. When CT-HRP was injected into the hindlimb on PD1 and the pups were maintained for 1 day prior to death and HRP histochemistry, labeled axons were present within the fasciculus gracilis at thoracic and cervical levels, but they could not be traced into the medulla. When comparable injections were made on PD3, and the pups were maintained for 2 days, labeled axons were present within the caudal medulla. Our second objective was to determine whether axons of the fasciculus gracilis grow through a lesion of their spinal pathway during early development. In one group of animals, the thoracic cord was transected at PD5, 8, 12, 20 and 26 and bilateral injections of Fast Blue (FB) were made four segments caudal to the lesion 30-40 days later. After a 3-5 day survival, the pups were killed and perfused so that the spinal cord and brainstem could be removed and sectioned for fluorescence microscopy. In all of the cases lesioned at PD5, axons of the fasciculus gracilis were labeled rostral to the site of transection and they could be followed to the nucleus gracilis. Evidence for growth of fasciculus gracilis axons into the caudal medulla was also seen in cases lesioned at PD8. In contrast, labeled axons were not observed rostral to the lesion when it was made at PD12 or at later stages of development. In order to verify that some of the axons which crossed the lesion originated within dorsal root ganglia, the thoracic cord was transected at PD5 in another group of animals and 7 days later, injections of CT-HRP were made into one of the hindlimbs. After a 3 day survival, labeled axons could be traced through the lesion site and into the caudal medulla. We conclude that axons of the fasciculus gracilis reach the nucleus gracilis by at least PD5 in the opossum and that they grow through a lesion of their spinal pathway when it is made at the same age or shortly thereafter. The critical period for such growth appears to end between PD8 and PD12.

Axonal regrowth after spinal cord transection in adult zebrafish

Using axonal tracers, we characterized the neurons projecting from the brain to the spinal cord as well as the terminal fields of ascending spinal projections in the brain of adult zebrafish with unlesioned or transected spinal cords. Twenty distinct brain nuclei were found to project to the spinal cord. These nuclei were similar to those found in the closely related goldfish, except that additionally the parvocellular preoptic nucleus, the medial octavolateralis nucleus, and the nucleus tangentialis, but not the facial lobe, projected to the spinal cord in zebrafish. Terminal fields of axons, visualized by anterograde tracing, were seen in the telencephalon, the diencephalon, the torus semicircularis, the optic tectum, the eminentia granularis, and throughout the ventral brainstem in unlesioned animals. Following spinal cord transection at a level approximately 3.5 mm caudal to the brainstem/spinal cord transition zone, neurons in most brain nuclei grew axons beyond the transection site into the distal spinal cord to the level of retrograde tracer application within 6 weeks. However, the individually identifiable Mauthner cells were never seen to do so up to 15 weeks after spinal cord transection. Nearly all neurons survived axotomy, and the vast majority of axons that had grown beyond the transection site belonged to previously axotomized neurons as shown by double tracing. Terminal fields were not re-established in the torus semicircularis and the eminentia granularis following spinal cord transection.

Evidence for growth of supraspinal axons through the lesion after transection of the thoracic spinal cord in the developing opossum Didelphis virginiana

In the present study, we asked whether supraspinal axons grow through a complete transection of the spinal cord in the developing opossum Didelphis virginiana. When the thoracic cord was transected at postnatal day (PD) 5 and bilateral injections of Fast Blue (FB) were made four segments caudal to the lesion 30-40 days later, FB-containing neurons were found in each of the supraspinal nuclei labeled by comparable injections in age-matched unlesioned controls. Continuity between the cut ends of the cord was obviously gross when the animals were killed, and histologically recognizable spinal cord was present at the lesion site. When the same procedure was followed on pups subjected to transection at PD12, FB-containing neurons were still present at supraspinal levels, but they appeared to be fewer in number than in the PD5 cases or the age-matched controls, and none were found within the medial pontine reticular and lateral vestibular nuclei. When the lesion was made at PD20, labeled neurons were even fewer in number, and when it was made at PD26, they were restricted to the medullary raphe and the red nuclei. There was no evidence for growth of supraspinal axons across lesions made at PD33. We conclude that supraspinal axons grow through the lesion after transection of the spinal cord in neonatal opossums and that the critical period for growth of reticulospinal and vestibulospinal axons through the lesion ends earlier than that for comparable growth of raphespinal and rubrospinal axons.