Role of Thy-1 in in vivo and in vitro neural development and regeneration of dorsal root ganglionic neurons

We have examined the expression of Thy-1, an abundant glycosylphosphatidylinositol (GPI)-anchored glycoprotein, in dorsal root ganglia (DRG) and associated nerve fascicles, during postnatal development and following a nerve crush. The expression levels of Thy-1 in DRG neurons, dorsal roots, and central processes in spinal cord were rather low at postnatal day 2, and gradually increased as DRG neurons matured. During early development, the expression of Thy-1 within DRG neurons was low and equally distributed between plasma membrane and cytosol. With maturation, the staining intensities of Thy-1 in both the plasma membrane and the cytosol of DRG neurons became increased. We also studied Thy-1 expression in the regeneration of mature DRG neurons following the crush injury of sciatic nerve. Two days after the crush injury, Thy-1 expression dramatically decreased in the DRG neurons on the lesion side. Between 4 and 7 days after the injury, the expression of Thy-1 gradually increased and returned to a normal level 1 week after the sciatic nerve crush. The time course of the up-regulation of Thy-1 expression during regeneration matched that of the recovery of sensory functions, such as pain withdraw reflex, placing reflex, and the score of Basso-Beattie-Bresnahan Locomotor Rating Scale. Taken together, our results suggest that Thy-1 expression is developmentally regulated and is closely associated with the functional maturation of DRG neurons during both postnatal development and nerve regeneration. Furthermore, perturbation of Thy-1 function with anti-Thy-1 antibodies promoted neurite outgrowth from primary cultured DRG neurons, again confirming the inhibitory role of Thy-1 on neurite outgrowth.     

Neurites show pathway specificity but lack directional specificity or predetermined lengths in Xenopus embryos

The earliest outgrowth of nerve fibers from identified spinal neurons labeled with horseradish peroxidase (HRP) was traced along surgically rearranged pathways in the central nervous system (CNS) of Xenopus embryos. Parts of the CNS were misaligned or inverted rostrocaudally by grafting a segment of labeled spinal cord in place of the same or different spinal cord segment of an unlabeled embryo or by joining two rostral half embryos (head-to-head) or two caudal half embryos (tail-to-tail), one half of which was derived from a labeled embryo in each combination. Donor embryos were labeled by injection of HRP into a selected blastomere at the 16- or 32-cell stage. Host embryos were unlabeled. Grafts from labeled donors to unlabeled host embryos were made at early neural tube stages before outgrowth of any nerve fibers had started (Jacobson and Huang, 1985). Routes taken by labeled nerve fibers growing into unlabeled CNS were observed at later stages, and the rates of nerve fiber elongation were calculated. Labeled nerve fibers were normal in appearance, and elongated without branching, at normal rates (22-71 micron/h). In head-to-head and tail-to-tail embryos and in embryos with inverted spinal cord grafts, nerve fibers continued elongating without branching in the direction opposite to normal in the CNS. Many fibers reached lengths that were far greater than normal. No reorientation of such maldirected nerve fibers was seen. These results indicate that nerve fiber elongation is not guided by axially polarized pathway cues or markers and that nerve fibers do not grow to predetermined lengths. However, neurites preferred to grow along stereotyped nerve fiber pathways even when forced to grow in the wrong direction or when confronted with nonneural tissue.     

Protein composition and synthesis in the adult mouse spinal cord

Propepties of spinal cord proteins were studied in adult mice subjected to unilateral crush or electrical stimulation of sciatic nerve. The protein composition of spinal tissue was determined using SDS-polyacrylamide gel electrophoresis coupled with subcellular fractionation. Comparisons of mouse spinal cord and brain revealed similarities in the types but differences in the concentrations of myelin associated proteins, nuclear histones and other proteins. Comparisons with sciatic nerve proteins demonstrated differences in types of proteins but similarities in the concentration of myelin proteins and nuclear histones. The short term (<2 hrs.) incorporation of radioactive amino acids into spinal cord proteins revealed heterogeneous rates of incorporation. Neither nerve crush six days prior to testing nor sciatic nerve stimulation had a significant effect on the protein composition or amino acid incorporation rates of spinal cord tissue. These observations suggest that known differences in spinal cord function following alterations in nerve input may be dependent upon different mechanisms than have been found in the brain.     

Peripherally-derived BDNF promotes regeneration of ascending sensory neurons after spinal cord injury

Background

The blood brain barrier (BBB) and truncated trkB receptor on astrocytes prevent the penetration of brain derived neurotrophic factor (BDNF) applied into the peripheral (PNS) and central nervous system (CNS) thus restrict its application in the treatment of nervous diseases. As BDNF is anterogradely transported by axons, we propose that peripherally derived and/or applied BDNF may act on the regeneration of central axons of ascending sensory neurons.

Methodology/Principal Findings

The present study aimed to test the hypothesis by using conditioning lesion of the sciatic nerve as a model to increase the expression of endogenous BDNF in sensory neurons and by injecting exogenous BDNF into the peripheral nerve or tissues. Here we showed that most of regenerating sensory neurons expressed BDNF and p-CREB but not p75NTR. Conditioning-lesion induced regeneration of ascending sensory neuron and the increase in the number of p-Erk positive and GAP-43 positive neurons was blocked by the injection of the BDNF antiserum in the periphery. Enhanced neurite outgrowth of dorsal root ganglia (DRG) neurons in vitro by conditioning lesion was also inhibited by the neutralization with the BDNF antiserum. The delivery of exogenous BDNF into the sciatic nerve or the footpad significantly increased the number of regenerating DRG neurons and regenerating sensory axons in the injured spinal cord. In a contusion injury model, an injection of BDNF into the footpad promoted recovery of motor functions.

Conclusions/Significance

Our data suggest that endogenous BDNF in DRG and spinal cord is required for the enhanced regeneration of ascending sensory neurons after conditioning lesion of sciatic nerve and peripherally applied BDNF may have therapeutic effects on the spinal cord injury.     

Hepatocyte growth factor protects retinal ganglion cells by increasing neuronal survival and axonal regeneration in vitro and in vivo

Hepatocyte growth factor (HGF) is known to promote the survival and foster neuritic outgrowth of different subpopulations of CNS neurons during development. Together with its corresponding receptor c-mesenchymal-epithelial transition factor (Met), it is expressed in the developing and the adult murine, rat and human CNS. We have studied the role of HGF in paradigms of retinal ganglion cell (RGC) regeneration and cell death in vitro and in vivo. After application of recombinant HGF in vitro, survival of serum-deprived RGC-5 cells and of growth factor-deprived primary RGC was significantly increased. This was shown to be correlated to the phosphorylation of c-Met and subsequent activation of serine/threonine protein kinase Akt and MAPK downstream signalling pathways involved in neuronal survival. Furthermore, neurite outgrowth of primary RGC was stimulated by HGF. In vivo, c-Met expression in RGC was up-regulated after optic nerve axotomy lesion. Here, treatment with HGF significantly improved survival of axotomized RGC and enhanced axonal regeneration after optic nerve crush. Our data demonstrates that exogenously applied HGF has a neuroprotective and regeneration-promoting function for lesioned CNS neurons. We provide strong evidence that HGF may represent a trophic factor for adult CNS neurons, which may play a role as therapeutic target in the treatment of neurotraumatic and neurodegenerative CNS disorders.     

Studying the isolated central nervous system; a report on 35 years: more inquisitive than acquisitive

1. The CNS from invertebrate animals such as slugs, snails, leeches, and cockroaches, can be isolated and kept alive for many hours. 2. The electrical and pharmacological properties of invertebrate CNS neurons have many similarities and it is probable that the basic rules governing the CNS evolved more than 600 million years ago. 3. The nerve cells can show sodium action potentials, calcium action potentials, EPSP, IPSP, biphasic potentials, electrogenic sodium pump potentials, and a variety of potassium, sodium, calcium and chloride currents. 4. Invertebrate CNS ganglia contain identifiable individual nerve cells whose properties and responses to neurotransmitters and drugs are constant and repeatable from preparation to preparation. 5. It was possible to set up an isolated CNS-nerve trunk-muscle preparation and study the transport of radioactive material from the CNS to the muscle and from muscle to CNS. This has provided information about axoplasmic transport in both invertebrate and vertebrate preparations. 6. The methods developed from studies of invertebrate isolated CNS preparations have been applied to vertebrate isolated CNS preparations. 7. In addition to thin slices of the mammalian brain, it is possible to keep 5 cm lengths of the whole mammalian spinal cord and brain stem alive for many hours. 8. The isolated mammalian spinal cord has functional ipsilateral and contralateral reflexes, ascending and descending pathways, extensive sensory integrative local area networks, and inhibitory interneuron circuits. Much of the in vivo circuitry is functional in vitro. 9. The isolated mammalian spinal cord and brain stem can be developed to include functional higher brain circuits that will provide increased understanding of the control and integrative action of the mammalian central nervous system.     

Expression of nerve growth factor receptor mRNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons

In situ hybridization histochemistry and RNA blot analysis were used to study expression of nerve growth factor receptor (NGF-R) mRNA in rat spinal cord motoneurons. The results show that NGF-R mRNA is expressed at high levels in rat spinal cord motoneurons at the time of naturally occurring cell death. This expression is sustained, but reduced, during synapse formation and is subsequently greatly reduced in the adult spinal cord. A unilateral crush lesion of the sciatic nerve resulted in an 8-fold increase in NGF-R mRNA in adult rat spinal cord motoneurons 3 days after lesion, compared with the nonlesioned side. NGF-R mRNA induction was even more pronounced 7 and 14 days after lesion, reaching levels 12 times higher than those on the nonlesioned side. However, 6 weeks after lesion, when the motor function of the leg was largely restored, NGF-R expression had decreased to levels similar to those on the contralateral side. We therefore suggest that NGF-R mediates a trophic or axonal guidance function for developing and regenerating spinal cord motoneurons.     

Regeneration of dorsal column fibers into and beyond the lesion site following adult spinal cord injury.

Regeneration is abortive following adult mammalian CNS injury. We have investigated whether increasing the intrinsic growth state of primary sensory neurons by a conditioning peripheral nerve lesion increases regrowth of their central axons. After dorsal column lesions, all fibers stop at the injury site. Animals with a peripheral axotomy concomitant with the central lesion show axonal growth into the lesion but not into the spinal cord above the lesion. A preconditioning lesion 1 or 2 weeks prior to the dorsal column injury results in growth into the spinal cord above the lesion. In vitro, the growth capacity of DRG neurite is also increased following preconditioning lesions. The intrinsic growth state of injured neurons is, therefore, a key determinant for central regeneration.     

Permanent alterations of spinal cord reflexes following nerve lesion in newborn rats

Sciatic nerve lesion in newborn rats is known to cause degeneration of a large number of axotomized motoneurones and spinal ganglion cells. Some of the surviving motoneurones exhibit abnormal firing properties and the projection pattern of central terminals of sensory neurones is altered. We report here on long-term changes in spinal cord reflexes in adult rats following neonatal nerve crush. In acutely spinalized and anaesthetized adult rats 4-6 months old in which the sciatic nerve had been crushed on one side at birth, the tibial nerve, common peroneal nerve or sural nerve were stimulated on the reinnervated and control side and reflex responses were recorded from the L5 ventral spinal roots. Ventral root responses (VRRs) to tibial and peroneal nerve stimulation on the side of the nerve lesion were significantly smaller in amplitude representing only about 15% of the mean amplitude of VRRs on the control side. The calculated central delay of the first, presumably monosynaptic component of the VRR potential was 1.6 ms on the control side while the earliest VRR wave on the side of the nerve lesion appeared after a mean central latency of 4.0 ms that seems too long to be of monosynaptic origin. These results suggest that neonatal sciatic nerve injury markedly alters the physiological properties and synaptic connectivity in spinal cord neurones and causes a marked depression of spinal cord responses to peripheral nerve stimulation.     

Regeneration of sensory axons within the injured spinal cord induced by intraganglionic cAMP elevation

The peripheral branch of primary sensory neurons regenerates after injury, but there is no regeneration when their central branch is severed by spinal cord injury. Here we show that microinjection of a membrane-permeable analog of cAMP in lumbar dorsal root ganglia markedly increases the regeneration of injured central sensory branches. The injured axons regrow into the spinal cord lesion, often traversing the injury site. This result mimics the effect of a conditioning peripheral nerve lesion. We also demonstrate that sensory neurons exposed to cAMP in vivo, when subsequently cultured in vitro, show enhanced growth of neurites and an ability to overcome inhibition by CNS myelin. Thus, stimulating cAMP signaling increases the intrinsic growth capacity of injured sensory axons. This approach may be useful in promoting regeneration after spinal cord injury.     

The role of sensory input in motor neuron sprouting control

Primary sensory neurons project to motor neurons directly or through interneurons and affect their activity. In our previous paper we showed that intramuscular sprouting can be affected by changing the sensory synaptic input to motor neurons. In this work, motor axon sprouting within a peripheral nerve (extramuscular sprouting) was induced by nerve injury at such a distance from muscle so as not to allow nerve-muscle trophic interactions. Two different procedures were carried out: (1) sciatic nerve crush and (2) sciatic nerve crush with homosegmental ipsilateral L3-L5 dorsal rhizotomy. The number of regenerating motor axons innervating extensor digitorum longus muscle was determined by in vivo muscle tension recordings and an index of their individual conduction rate was obtained by in vitro intracellular recordings of excitatory postsynaptic end-plate potentials in muscle fibers. The main findings were: (1) there are more regenerated axons distally from the lesion than parent axons proximally to the lesion (sprouting at the lesion); (2) sprouting at the lesion was negatively affected by homosegmental ipsilateral dorsal rhizotomy; (3) the number of motor axons innervating extensor digitorum longus muscle extrafusal fibers counted proximally to the lesion increased following nerve injury and regeneration but this did not occur when sensory input was lost. A transient innervation of extrafusal fibers by &#110 motor neurons may explain the increase of motor axons counted proximally to the lesion.     

The role of sensory input in motor neuron sprouting control

Primary sensory neurons project to motor neurons directly or through interneurons and affect their activity. In our previous paper we showed that intramuscular sprouting can be affected by changing the sensory synaptic input to motor neurons. In this work, motor axon sprouting within a peripheral nerve (extramuscular sprouting) was induced by nerve injury at such a distance from muscle so as not to allow nerve-muscle trophic interactions. Two different procedures were carried out: (1) sciatic nerve crush and (2) sciatic nerve crush with homosegmental ipsilateral L3-L5 dorsal rhizotomy. The number of regenerating motor axons innervating extensor digitorum longus muscle was determined by in vivo muscle tension recordings and an index of their individual conduction rate was obtained by in vitro intracellular recordings of excitatory postsynaptic end-plate potentials in muscle fibers. The main findings were: (1) there are more regenerated axons distally from the lesion than parent axons proximally to the lesion (sprouting at the lesion); (2) sprouting at the lesion was negatively affected by homosegmental ipsilateral dorsal rhizotomy; (3) the number of motor axons innervating extensor digitorum longus muscle extrafusal fibers counted proximally to the lesion increased following nerve injury and regeneration but this did not occur when sensory input was lost. A transient innervation of extrafusal fibers by gamma motor neurons may explain the increase of motor axons counted proximally to the lesion.     

Inhibition of Rho kinase (ROCK) increases neurite outgrowth on chondroitin sulphate proteoglycan in vitro and axonal regeneration in the adult optic nerve in vivo

Inhibitory molecules derived from CNS myelin and glial scar tissue are major causes for insufficient functional regeneration in the mammalian CNS. A multitude of these molecules signal through the Rho/Rho kinase (ROCK) pathway. We evaluated three inhibitors of ROCK, Y- 27632, Fasudil (HA-1077), and Dimethylfasudil (H-1152), in models of neurite outgrowth in vitro. We show, that all three ROCK inhibitors partially restore neurite outgrowth of Ntera-2 neurons on the inhibitory chondroitin sulphate proteoglycan substrate. In the rat optic nerve crush model Y-27632 dose-dependently increased regeneration of retinal ganglion cell axons in vivo. Application of Dimethylfasudil showed a trend towards increased axonal regeneration in an intermediate concentration. We demonstrate that inhibition of ROCK can be an effective therapeutic approach to increase regeneration of CNS neurons. The selection of a suitable inhibitor with a broad therapeutic window, however, is crucial in order to minimize unwanted side effects and to avoid deleterious effects on nerve fiber growth.     

Peripheral Nerve Lesion Induces an Up-regulation of Spy1 in Rat Spinal Cord

Spy1, as a member of the Speedy/RINGO family and a novel activator of cyclin-dependent kinases, was shown to promote cell cycle progression and cell survival in response to DNA damage. While its expression and roles in nervous system lesion and repair were still unknown. Here, we performed an acute sciatic nerve injury model in adult rats and studied the dynamic changes of Spy1 expression in lumbar spinal cord. Temporally, Spy1 expression was increased shortly after sciatic nerve crush and peaked at day 2. Spatially, Spy1 was widely expressed in the lumbar spinal cord including neurons and glial cells. While after injury, Spy1 expression was increased predominantly in astrocytes and microglia, which were largely proliferated. Moreover, there was a concomitant up-regulation of CDK2 activity and down-regulation of p27. Collectively, we hypothesized peripheral nerve injury induced an up-regulation of Spy1 in lumbar spinal cord, which was associated with glial proliferation. Ye Huang and Yonghua Liu contributed equally to this work.     

Inhibition of murine retrovirus-induced neurodegeneration in the spinal cord by explant culture.

The neurovirulent chimeric mouse ecotropic retrovirus FrCasE causes a rapid neurodegenerative disease of the central nervous system (CNS) characterized by the appearance of spongiform lesions in motor areas 10 days after neonatal inoculation. To study the details of the pathogenic process, we examined the ability of an ex vivo spinal cord model to recapitulate disease. Organotypic spinal cord slice cultures were established from IRW mice 7 days after neonatal inoculation. This corresponds to a time when virus expression in the CNS is first detectable but spongiform changes have yet to evolve. Infectivity associated with these cultures peaked at 7 days in vitro and persisted at this level for 6 weeks. FrCasE infection of the spinal cord slices was primarily found associated with microglial cells. Infection of neurons, astrocytes, oligodendroglia, and endothelial cells was not observed; however, significant astrogliosis was found. Despite the presence of extensive microglial infection in close association with spinal motor neurons in organotypic cultures, no virus-specific spongiform degenerative changes were observed. These results suggest that removal of motor neurons from the developing CNS, despite maintaining the local cytoarchitectural relationships, prevents the virus from eliciting its pathological effects. Possible reasons for the interruption of lesion development are discussed.     

Neurite outgrowth traced by means of horseradish peroxidase inherited from neuronal ancestral cells in frog embryos

Outgrowing neurites in Xenopus embryos were labeled with horseradish peroxidase which had been injected into a single blastomere at the 32-cell stage and had been inherited by all the descendants, including neurons. Neurite outgrowth was traced from labeled trigeminal ganglion cells and most or all types of neurons present in the spinal cord at embryonic stages 20-30: primary motoneurons, commissural, dorsal longitudinal, ventral longitudinal, and Rohon-Beard neurons. All types of nerve fibers grew by the most direct pathway, apparently without errors of initial outgrowth, pathway selection, or target selection. An initial transient phase of outgrowth of filopodial processes from neuronal cell bodies and shafts of short neurites was observed which disappeared after further elongation of the neurites. The first pioneer fibers grew out from all types in a 2-hr period, from stage 20 to 22, and these fibers arrived at the targets within 3.5 hr after initial outgrowth. Additional fibers grew later in contact with the pioneers to form fascicles. Nerve fibers elongated without branching until they neared or contacted their targets. The rate of elongation at 20 degrees C was 30-75 micron/hr. The rapid, unbranched, error-free initial outgrowth and elongation of neurites to their targets is discussed in relation to theories of development of nerve pathways.     

PARP inhibition alleviates diabetes-induced systemic oxidative stress and neural tissue 4-hydroxynonenal adduct accumulation: correlation with peripheral nerve function

This study evaluated the role of poly(ADP-ribose) polymerase (PARP) in systemic oxidative stress and 4-hydoxynonenal adduct accumulation in diabetic peripheral neuropathy. Control and streptozotocin-diabetic rats were maintained with or without treatment with the PARP inhibitor, 1,5-isoquinolinediol, 3 mg kg(-1) day(-1), for 10 weeks after an initial 2 weeks. Treatment efficacy was evaluated by poly(ADP-ribosyl)ated protein content in peripheral nerve and spinal cord (Western blot analysis) and dorsal root ganglion neurons and nonneuronal cells (fluorescence immunohistochemistry), as well as by indices of peripheral nerve function. Diabetic rats displayed increased urinary isoprostane and 8-hydroxy-2'-deoxyguanosine excretion (ELISA) and 4-hydroxynonenal adduct accumulation in endothelial and Schwann cells of the peripheral nerve, neurons, astrocytes, and oligodendrocytes of the spinal cord and neurons and glial cells of the dorsal root ganglia (double-label fluorescence immunohistochemistry), as well as motor and sensory nerve conduction velocity deficits, thermal hypoalgesia, and tactile allodynia. PARP inhibition counteracted diabetes-induced systemic oxidative stress and 4-hydroxynonenal adduct accumulation in peripheral nerve and spinal cord (Western blot analysis) and dorsal root ganglion neurons (perikarya, fluorescence immunohistochemistry), which correlated with improvement of large and small nerve fiber function. The findings reveal the important role of PARP activation in systemic oxidative stress and 4-hydroxynonenal adduct accumulation in diabetic peripheral neuropathy.     

Astroglia demonstrate regional differences in their ability to maintain primary dendritic outgrowth from mouse cortical neurons in vitro

To determine whether glia from different regions of the central nervous system (CNS) initiate or maintain primary dendritic growth, embryonic day 18 mouse cortical neurons were co-cultured with rat (postnatal day 4) astroglial cells derived from retina, spinal cord, mesencephalon, striatum, olfactory bulb, retina, and cortex. Axon and dendrite outgrowth from isolated neurons was quantified using morphological and immunohistochemical techniques at 18 h and 1, 3, and 5 days in vitro. Neurons initially extend the same number of neurites, regardless of the source of glial monolayer; however, glial cells differ in their ability to maintain primary dendrites. Homotypic cortical astrocytes maintain the greatest number of primary dendrites. Glia derived from the olfactory bulb and retina maintained intermediate numbers of dendrites, whereas only a small number of primary dendrites were maintained by glia derived from striatum, spinal cord, or mesencephalon. Longer axons were initially observed from neurons grown on glia that did not maintain dendrite number. Axonal length, however, was similar on the various monolayers after 5 days in vitro. Neurons that were grown in media conditioned by either mesencephalic or cortical glia for the first 24 h followed by culture media from glia of the alternate source for 4 days in vitro confirmed that glia maintained, rather than initiated, the outgrowth of the primary dendritic arbor. These results indicate that glial cells derived from various CNS regions differ in their ability to maintain the primary dendritic arbor from mouse cortical neurons in vitro. © 1995 John Wiley & Sons, Inc.     

Oncostatin M Reduces Lesion Size and Promotes Functional Recovery and Neurite Outgrowth After Spinal Cord Injury

The family of interleukin (IL)-6 like cytokines plays an important role in the neuroinflammatory response to injury by regulating both neural as well as immune responses. Here, we show that expression of the IL-6 family member oncostatin M (OSM) and its receptor is upregulated after spinal cord injury (SCI). To reveal the relevance of increased OSM signaling in the pathophysiology of SCI, OSM was applied locally after spinal cord hemisection in mice. OSM treatment significantly improved locomotor recovery after mild and severe SCI. Improved recovery in OSM-treated mice was associated with a reduced lesion size. OSM significantly diminished astrogliosis and immune cell infiltration. Thus, OSM limits secondary damage after CNS trauma. In vitro viability assays demonstrated that OSM protects primary neurons in culture from cell death, suggesting that the underlying mechanism involves direct neuroprotective effects of OSM. Furthermore, OSM dose-dependently promoted neurite outgrowth in cultured neurons, indicating that the cytokine plays an additional role in CNS repair. Indeed, our in vivo experiments demonstrate that OSM treatment increases plasticity of serotonergic fibers after SCI. Together, our data show that OSM is produced at the lesion site, where it protects the CNS from further damage and promotes recovery.     

Axonal labyrinths in the rat spinal cord: a consequence of degenerative atrophy.

Transection, crush or local colchicine treatment of a peripheral nerve induces degenerative atrophy of central terminals of primary sensory neurons in the Rolando substance of the rat spinal cord. In addition to osmiophilic alterations that occur in the course of degenerative processes in general, degenerative atrophy is characterized by the appearance of spectacular labyrinthine formations. Electron-microscopic analysis reveals that these consist of flattened axonal profiles. Axonal labyrinths are interpreted as signs of futile regenerative efforts of axon terminals undergoining degenerative atrophy. Labyrinths disappear from the Rolando substance several months after peripheral nerve injury, when degenerative atrophy of the central terminal is replaced by regenerative proliferation.