Naturally occurring cell death during neural development

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.

Cell death of motoneurons in the chick embryo spinal cord. X. Synapse formation on motoneurons following the reduction of cell death by neuromuscular blockade

Chronic treatment of chick embryos with neuromuscular blocking agents, such as curare, rescues motoneurons from naturally occurring cell death. In the present study, embryos treated with curare from E6 to E9 had 35% more motoneurons than controls on E10 and 42% more than controls on E16. Previous studies have shown that several aspects of motoneuron differentiation occur normally in curare-treated embryos. We report here that dendrite growth and arborization is also unaltered on E10 and E16 following curare treatment. A quantitative analysis of afferent synapses on motoneurons shows that the packing density of both axosomatic and axodendritic synapses is also normal on E10 in curare-treated embryos, despite the greater number of motoneurons present. This indicates that the interneurons that provide presynaptic input to motoneurons are able to compensate for the increased number of synaptic sites made available by curare treatment. However, by E16 the packing density of synapses is reduced by about half. Because motoneurons and their dendrites continue to grow between E10 and E16, the further increase in synaptic sites made available in curare-treated embryos apparently exceeds the compensatory capacity of presynaptic interneurons on E16. One can conclude from these results that the increased survival of motoneurons in curare-treated embryos is not owing to an increase in afferent synapses. Motoneurons in these embryos continue to survive in the face of either no change (E10) or a reduction (E16) in the number of axodendritic and axosomatic synapses. Therefore, increased motoneuron survival in this situation is very likely regulated primarily by motoneuron-target interactions.

Endocytosis and autophagy in dying neurons: an ultrastructural study in chick embryos

In an effort to understand naturally occurring neuronal death in the developing isthmo-optic nucleus, we have accentuated one of its most probably causes, failure to receive adequate trophic maintenance from the axonal terminal zone in the retina, and have studied the dying neurons ultrastructurally. Retrograde trophic maintenance was blocked by means of intraocularly injected colchicine, which caused all the isthmo-optic neurons to die by just one of the two or more kinds of cell death that they undergo during normal development. The present paper deals with the very prominent cytoplasmic aspects of this kind of cell death, notably the uptake of exogeneous horseradish peroxidase and autophagy. There were also nuclear changes, which are dealt with mainly in the accompanying paper (Clarke and Hornung, J. Comp. Neurol. 283:438-449,'89). Numerous cytoplasmic vacuoles occurred in both soma and dendrites, and they were of three main kinds, of which the smallest (less than 0.5 microns diameter) had unstructured contents, whereas the larger two (1-2 microns and 2-7 microns) were secondary lysosomes (mostly residual bodies). Intravascularly injected horseradish peroxidase labeled all three kinds of vacuole but not the free cytoplasm, indicating that the uptake was by endocytosis rather than by leakage through holes in the membrane, as is confirmed by our failure to detect any such holes. We suspect that the smallest vacuoles are the primary endosomes, that these subsequently fuse with vacuoles of the intermediate kind, and that the largest vacuoles are formed by the fusion of these latter. The purpose of the endocytosis may be to channel the plasma membrane piecemeal into the lysosomes for destruction.

Numerical matching in the mammalian CNS: lack of a competitive advantage of early over late-generated cerebellar granule cells

In this report we use postnatal 3H-thymidine injections to test whether granule cells that are generated early in postnatal cerebellar development and whose axons have access to their Purkinje cell target beginning in the first postnatal week have an advantage over granule cells generated 9 days later in the competition for target-related stabilization. In the wild-type mouse, 3-5% of the adult granule cell population is labeled by injection of 3H-thymidine at either postnatal day 4 (P4) or P13. In the lurcher mutant, however, over 40% of the surviving granule cells are labeled by P4 injection while less than 1% are labeled after a P13 injection. Together, these results suggest that time of target contact is a critical factor in the competition for neuronal survival. The results from the lurcher chimeras, however, reveal that the situation is likely to be more complicated. In all chimeras examined, with target sizes ranging from 3 to 108% of wild type, equivalent numbers of granule cells were labeled at P4 and P13. These data lead to the contradictory conclusion that, in this experimental situation, early generated granule cells do not have a competitive advantage over later-generated granule cells. The results are discussed in terms of various models of target stabilization. We propose that, of the various hypotheses, our results are best explained by postulating two distinct mechanisms for developmental cell death. Supporting evidence for this hypothesis from other neuronal systems is also briefly reviewed.

Basic fibroblast growth factor prevents ontogenetic neuron death in vivo

Basic fibroblast growth factor (bFGF) is a mitogen and a potent neurotrophic protein for ciliary ganglionic (CG) neurons in vitro. Recombinant bFGF was administered to the chorionic-allantoic membrane of chick embryos during the period of ontogenetic neuron death in the cholinergic CG between embryonic days (E) 8 and E14. Neuronal losses in untreated chicks and in embryos that received the vehicle only (phosphate-buffered saline plus cytochrome c) amounted to 44%. Basic FGF permitted the survival of 87% of the neurons present at E8. There were no apparent differences in the size and number of non-neuronal cells in CG. These data add to the increasing body of evidence that bFGF is an important multifunctional growth factor with a neurotrophic capacity.

Early postnatal development of vasoactive intestinal polypeptide- and peptide histidine isoleucine-immunoreactive structures in the cat visual cortex

The early postnatal development of neurons containing vasoactive intestinal polypeptide (VIP) and peptide histidine isoleucine (PHI) has been analyzed in visual areas 17 and 18 of cats aged from postnatal day (P) 0 to adulthood. Neuronal types are established mainly by axonal criteria. Both peptides occur in the same neuronal types and display the same postnatal chronology of appearance. Several cell types are transient, which means that they are present in the cortex only for a limited period of development. According to their chronology of appearance the VIP/PHI-immunoreactive (ir) cell types are grouped into three neuronal populations. The first population comprises six cell types which appear early in postnatal life. The pseudohorsetail cells of layer I possess a vertically descending axon which initially gives rise to recurrent collaterals, then forms a bundle passing layers III to V, and finally, horizontal terminal fibers in layer VI. The neurons differentiate at P 4 and disappear by degeneration around P 30. The neurons with columnar dendritic fields of layers IV/V are characterized by a vertical arrangement of long dendrites ascending or descending parallel to each other, thus forming an up to 600 microns long dendritic column. Their axons always descend and terminate in broad fields in layer VI. The neurons appear at P 7 and are present until P 20. The multipolar neurons of layer VI occur in isolated positions and have broad axonal territories. The neurons differentiate at P 7 and persist into adulthood. Bitufted to multipolar neurons of layers II/III have axons descending as a single fiber to layer VI, where they terminate. The neurons appear at P 12 and persist into adulthood. The four cell types described above issue a vertically oriented fiber architecture in layers II-V and a horizontal terminal plexus in layer VI which is dense during the second, third and fourth week. Concurrent with the disappearance of the two transient types the number of descending axonal bundles and the density of the layer VI plexus is reduced, but the latter is maintained during adulthood by the two persisting cell types. Two further cell types belong to the first population: The transient bipolar cells of layers IV, V, and VI have long dendrites which extend through the entire cortical width. Their axons always descend, leave the gray matter, and apparently terminate in the upper white matter. The neurons differentiate concurrently with the pseudohorsetail cells at P 4, are very frequent during the following weeks, and eventually disappear at P 30.(ABSTRACT TRUNCATED AT 400 WORDS)

Formation and growth of the cerebral convolutions. II. Cell death in the gyrus suprasylvius and adjoining sulci in the cat

Cell death, calculated by counting pyknotic nuclei to assess the number of dying cells, in the gyrus suprasylvius (GS-Syl) and adjoining sulci sulcus lateralis (SL) and sulcus suprasylvius (SS-Syl) was studied in cats aged 5, 15, 25 days and 6 months. Three patterns of cell death were characterized: (1) an ascending gradient from the inner to the upper cortical layers; (2) a lateromedial gradient from the SS-Syl towards the SL; and (3) a predominance of cell death in the sulcal zones. These patterns are in accordance with the sequence of cortical neurogenesis, the lateromedial pattern of the whole formation and growth of the GS-Syl and adjoining sulci, and the differences in the cortical thickness between the sulci and the gyral crown.

Correlated ultrastructural damage between cerebellum cells after early anticonvulsant treatment in mice

The anticonvulsants phenobarbital and diphenylhydantoin administered early in life to mice resulted in significant and long-lasting ultrastructural damage, including abnormalities of mitochondria, myelin sheaths, and lamellar inclusion bodies inside identified cells throughout the cortical layers of the cerebellum in treated vs control mice. The magnitude, distribution and duration of damage was age and treatment specific. No differences were detected in density of parallel fiber processes nor in synapse density within the molecular layer. Neuron profiles containing damaged organelles were not homogeneously distributed but made up only a small fraction of the total cell population examined. In our experiments, there was an overall within-animal correlation explaining 45% of the magnitude of damage in different cerebellar regions, but between synaptically connected cells, specifically mossy fiber axon varicosities and granule cell dendrite profiles, the subset population ratio of damaged-to-total mitochondria was highly significantly correlated (70-90%; P less than 0.001). We hypothesized that some correlated transneuronal degeneration and death in the central nervous system may have a transynaptically regulated component that first appears as correlated damage between synaptically connected cells, perhaps regardless of the degree of toxicity. The orderly cytoarchitecture and cell connections of the cerebellar cortex can be used to study these patterns of degeneration.

Regulation of granule cell number by a predetermined number of Purkinje cells in development

Development dysgenesis of Purkinje cells or granule cells was analyzed for the reciprocal effect of reduced number of each cell type on the other. A single pre- or postnatal injection of methylazoxymethanol acetate (MAM) in the rat reduces either the number of Purkinje cells or the number of granule cells when administered at the time of their respective genesis. The total number of these two types of neurons was obtained from cell density values of each layer and the total volume of the granular layer and the area of the Purkinje cell layer. The results show that Purkinje cells (targets) strictly determine the maximum number of granule cells (afferent neurons) following deficits in the number of Purkinje cells produced by prenatal MAM administration. Deficits in Purkinje cells were accompanied by a proportionally smaller number of granule cells so that the ratio remained constant. On the other hand, the reduction in the number of granule cells of the postnatal MAM model did not affect the number of Purkinje cells. These results indicate that the maximum number of these afferent neurons is constrained unidirectionally through a property defined by the number of their target neurons which develop earlier. Furthermore the number of afferent cells had no effect on the number of target cells.

Distinct roles for adhesion molecules during innervation of embryonic chick muscle

In vitro studies have suggested that the cell adhesion molecules NCAM and G4/L1 contribute to a variety of events during neural development. We have directly tested the role played by these molecules in the process of initial nerve ingrowth and ramification in the embryonic chick iliofibularis muscle by in ovo injections of specific adhesion-blocking antibodies and analysis of the resultant nerve branching pattern in muscle whole mounts. Antibodies against both molecules produced axonal defasciculation, which resulted in an enhanced transverse projection to the fast region of the muscle. In the case of anti-G4/L1, we also observed a large increase in the number of side branches that form from nerve trunks in the slow region and an enhancement of nerve branching in the fast region. Conversely, anti-NCAM produced a striking decrease in both the number and length of side branches in the slow region, and a reduction in nerve branching in the fast region. A similar reduction of nerve branching was obtained following injection of an endosialidase, which removes sialic acid from NCAM, and which was observed to enhance fiber-fiber apposition, presumably by increasing cell adhesion. Based on their biochemical properties in vitro and their in vivo distribution, both NCAM and G4/L1 are in a position to contribute to axon-axon adhesive interactions, whereas NCAM would be expected to also promote axon-myotube interactions. Our observations in fact indicate that these two adhesion molecules play different but complementary roles during muscle innervation and, specifically, that axon-axon fasciculation is influenced by both NCAM and G4/L1 in an anatomically distinct manner to regulate the overall pattern of nerve branching and that NCAM-mediated axon-myotube interactions are necessary for the attainment of the normal stereotyped pattern of nerve branching in both fast and slow regions of this muscle.

Cell death delineates axon pathways in the hindlimb and does so independently of neurite outgrowth

We wished to know whether the cell death and phagocytosis seen near the outgrowing nerve front in the hindlimb delineate axon pathways and, if so, whether the cells died only in the presence of growth cones. We unilaterally deleted the lumbosacral neural tube and reconstructed the patterns of neurite outgrowth and phagocytes during the stage when neurites first begin to colonize the thigh. In the control limbs, sensory and motor nerve pathways coincided with sites of phagocytosis, including those pathways that had yet to be colonized by growth cones. For instance, phagocytes were clustered at foci within the muscle masses where muscle nerves form a day later. However, they were not seen in adjacent, nonpathway regions such as posterior sclerotome or dorsal and ventral to the region of the plexus in which axons extend only posteriorly. Phagocytes were also seen in defined regions that are probably inaccessible to growth cones because they are too distant from pathways (i.e., subjacent to the apical ectodermal ridge) or express substances that are typical of precartilagenous tissues which may prohibit axon advance. In the experimental limbs, we conservatively estimated that neurite outgrowth was reduced to less than one-tenth (neurites were visible only with electron microscopy) or less than one-third of normal. Outgrowth extended less far distally and, in half the cases, motor innervation was completely abolished. Despite the extensive reduction in neurite outgrowth, the distribution of phagocytes was indistinguishable from that of the control side. Furthermore, the number of phagocytes did not differ significantly. We conclude that cell death delineates axon pathways remarkably well and does so without an interaction with growth cones; it is an independent characteristic of the axonal pathways and may be directly or indirectly important to axonal pathfinding. This is the first identification of a feature that characterizes prospective nerve pathways in the hindlimb.

Afferent control of neuron numbers in the developing brain

In this study we tested whether the quantitative matching of developing neuronal populations may depend on the size of the afferent supply. Partial deafferentation of the middle division of the parabigeminal nucleus (PBm) was produced before the period of naturally occurring cell death, by reducing the neuronal population of the superior colliculus following partial lesions or eye removal. The number of neurons surviving cell death in the PBm was linearly related to the number of its afferent neurons. This result supports the hypothesis that neurotrophic control by the afferent supply during the period of natural neuronal death is a major determinant of the number of neurons in the developing brain.

Promotion of survival and neurite outgrowth of cultured peripheral neurons by exogenous lipids and detergents

Gangliosides, in particular the monosialoglycosphingolipids Gtet 1 (GM1), have previously been implicated in the mediation of neuronal rescue and restitutional axonal growth, both in vitro and subsequent to brain and peripheral nerve lesions. In the present study it is shown that the bis-sialosyl gangliosides Gtet2b and Gtet3b, but not the gangliosides Gtet2a and Gtet1, promote the survival of dissociated dorsal root ganglion (DRG) neurons cultured from Embryonic Day (E) 8 chicks (DRG8) almost to the same extent as nerve growth factor (NGF). Ciliary ganglion (CG) neurons from E8 chicks (CG8) and DRG10 neurons were virtually not supported suggesting considerable specificity in terms of neuronal targets and developmental stages being addressed. Moreover, a variety of other lipids including cerebroside (Cb), dipalmitoylphosphatidylcholine (DPPC) and -serine (DPPS), sulfatide (Sf), and sphingomyelin (Sm) were tested for putative survival promoting activity toward chick CG, DRG, and lumbar sympathetic ganglion (SG11) neurons. At the highest concentration employed (2.5 x 10(-5) M), Sm, DPPC, and DPPS maintained between 45 and 65% of the plateau survival with CG8 (maximally supported by ciliary neuronotrophic factor (CNTF], DRG8, and DRG10 neurons, and 30 to 40% with SG11 neurons. Cb supported CG8 neurons at about 55% of the plateau value achieved with CNTF, but had hardly any effect on the other neuron populations tested. Control experiments using highly enriched neurons and serum-free conditions assured that the effects were unlikely to be mediated by serum components or nonneuronal cells. A variety of detergents, in particular Triton X-100, also promoted the survival of CG8 and DRG10 neurons. Ganglioside Gtet1, Sm, and Triton X-100 shifted the NGF titration curve for DRG10 neurons between 6- and 15-fold in a dose-dependent manner suggesting synergisms between NGF and lipids for neuronal maintenance. These results document the neuronotrophic potency of certain gangliosides, a heterogeneous group of structurally unrelated lipids, and detergents. The mechanisms by which these agents modulate neuronal survival still await clarification.

Transient tyrosine hydroxylase-like immunoreactive neurons contain somatostatin and substance P in the developing amygdala and bed nucleus of the stria terminalis of the rat

Tyrosine hydroxylase-like immunoreactive (TH-IR) neurons were observed from the embryonic day 17 (E17) to 6 weeks postnatally in two closely related nuclei of the limbic system, the bed nucleus of the stria terminalis (BNST) and the central nucleus of the amygdala (CNA) where they were restricted to circumscribed zones. These cells were scarce with an immature morphological aspect at E17. They progressively differentiated and increased in number until postnatal day 5 (P5), when their maximal density was reached. They were characterized as neurons by their ultrastructural appearance and the presence of both axo-somatic and axo-dendritic synaptic junctions. Moreover, TH-IR axons could be followed in the stria terminalis leaving the CNA, suggesting that part of TH-IR cells could be long projecting neurons rather than interneurons. A gradual decrease in the intensity of TH-IR and in density of labeled neurons was noted from P15 on, in both nuclei, (-50% at 4 weeks) until their total disappearance at 7 weeks. The significance of this TH-IR labeling regarding the catecholaminergic transmission remains unclear since these neurons did not contain the other catecholaminergic synthetic enzymes (DOPA-decarboxylase, dopamine-beta-hydroxylase, phenylethanolamine-N-methyl transferase) nor endogenous catecholamines. Double-labeling immunocytochemical methods, indicated that almost all the TH-IR neurons were colocalized with somatostatin 28 (SST) and with substance P (SP). Therefore these neurons expressed simultaneously 3 phenotypes, TH, SST and SP. This observation brings forth the notion of multiple neurotransmitter expression in transient neuronal populations and raises the question of neurotransmitter plasticity in the late postnatal development of the central nervous system (CNS). These neurons which were observed in two closely interconnected structures could be involved in early limbic circuits.

Development of 'non-specific' cholinesterase-containing neurons in the dorsal thalamus of the rat

In adult rats, neurons displaying histochemical staining for 'non-specific' cholinesterase (ChE) are found 3 distinct regions of the dorsal thalamus: the thalamic reuniens nucleus (Re), the anterior dorsal nucleus (AD), and a region that includes the lateral part of the central lateral nucleus (CL) and the ventral portion of the lateral dorsal nucleus (LD). Normal development of ChE-positive neurons was studied with cholinesterase histochemical techniques in postnatal infant rats. Although ChE staining of capillary endothelium is detectable shortly after birth, ChE staining of neurons first occurs at about postnatal day 5 (PND 5) with light staining of AD and CL-LD. At PND 7, staining in AD and CL-LD has increased in intensity and staining also is present in neurons of the anterior ventral (AV) and ventral anterior (VA) nuclei. ChE staining of neurons in Re first appears at PND 10. The number of neurons staining for ChE in each of these nuclei, and also the intensity of staining in individual neurons, appear to increase during the next several days until about PND 14. After PND 14, ChE staining intensity in neurons of AD, Re, and CL-LD appears to plateau and the pattern of staining continues into adulthood. In contrast, ChE staining of neurons in VA declines markedly and only a very few neurons in the dorsal part of VA remain ChE-positive after PND 21. ChE staining of neuropil in AV increases markedly, obscuring somatal staining in this nucleus. These results are discussed in regard to transient and continued expression of ChE activity in the dorsal thalamus and possible functional roles of ChE.

Excessive numbers of axons after early enucleation and blockade of metamorphosis in the oculomotor nerve of Xenopus laevis

The number of axons in the oculomotor (OM) nerve of Xenopus laevis tadpoles was counted in unoperated and in unilaterally enucleated animals, raised in 0.4 g/l thiourea (TU), a thyroid-blocking agent, which arrested their development at premetamorphosis. In unoperated animals the number of axons starts to decrease with metamorphosis. When raised in TU, the tadpoles do not metamorphose and show no axon loss; rather, there is a moderate increase in axon number (13%) after 6 months of thiourea-treatment. Thus metamorphosis is necessary for the initiation of axon loss. In unilaterally enucleated tadpoles, increased axon loss occurs at metamorphosis in the OM nerve of the operated side, but the contralateral OM nerve shows no loss at all. When these animals are treated with TU, there is, as compared with the effects of the TU-treatment in unoperated animals, an increase of 7% in the ipsilateral side and of 28% in the contralateral one. Thus thyroxine blockade prevents the effects of unilateral enucleation and induces an excessive number of axons during the period observed. We postulate that through blockade of metamorphosis, axon elimination is arrested, while their production continues, and that these effects are accentuated by manipulation of the axonal target.

Reduction of naturally occurring motoneuron death in vivo by a target-derived neurotrophic factor

Treatment of chick embryos in ovo with crude and partially purified extracts from embryonic hindlimbs (days 8 to 9) during the normal cell death period (days 5 to 10) rescues a significant number of motoneurons from degeneration. The survival activity of partially purified extract was dose-dependent and developmentally regulated. The survival of sensory, sympathetic, parasympathetic, and a population of cholinergic sympathetic preganglionic neurons was unaffected by treatment with hindlimb extract. The massive motoneuron death that occurs after early target (hindlimb) removal was partially ameliorated by daily treatment with the hindlimb extract. These results indicate that a target-derived neurotrophic factor is involved in the regulation of motoneuron survival in vivo.

Development of somatostatin immunoreactive neurons in the rat occipital cortex: a combined immunocytochemical-autoradiographic study

The postnatal development of somatostatin (SRIF)-immunoreactive neurons, previously labeled with [3H]thymidine on embryonic days E14-E22, has been studied in the rat occipital cortex. Immunocytochemistry combined with autoradiography showed an "inside-out" maturation pattern. Only SRIF neurons generated at E14 were present in layer VI in newborn rats. Later generated SRIF neurons appeared progressively higher in the cortex until about postnatal day 12 when SRIF neurons from E21 appeared in layer II. At 2 weeks of age, therefore, all SRIF neurons from E14-E21 were present. Most of these had been generated between E15 and E17 with a moderate number at E14 and rapidly diminishing numbers from E18 to E21. Although an overall layered distribution was apparent at peak production, there was a tendency for diffuse distribution most noticeable at E17. Diffusely distributed neurons were more likely to be below their appropriate layer than above it, thus contributing extra SRIF neurons to layer VI. At 3, 4, and 5 weeks, progressively fewer SRIF neurons were seen with a consequent reduction in the number of double-labeled neurons. It is suggested that the transient population of SRIF neurons thus revealed plays a role in cortical development.

Developmental changes in density and distribution of serotoninergic fibers in the chick spinal cord

Developmental changes of serotoninergic innervation in the chick spinal cord (third lumbosacral segment) were examined with an immunohistochemical technique using an antiserum to serotonin. In the 1-day-old hatched chick, serotoninergic fibers were located in laminae I, II, VII, IX, and X. A large number of serotonin-positive fibers and terminals were found around somal profiles of large neurons and in the neuropil of the medial and lateral parts of the lateral motor column (LMC). In the 1-week-old chick, the density of serotoninergic fibers was greatly increased in the posterior columns, and serotoninergic fibers were most densely aggregated in the dorsolateral part of the LMC. In the 2-week-old chick, a considerable decrease in the density of serotoninergic fibers was observed in the lateral funiculus and the gray matter (laminae I, II, VII, IX, and X). In the LMC, serotonin-positive fibers and terminals were largely absent from the neuropil, but were found preferentially around the somal profiles of large neurons. Between 1 and 2 weeks after hatching the density of varicosities and terminals in the neuropil of the dorsolateral and medial parts of the LMC decreased by 33% and 56%, respectively. In the 3-month-old chick, the density of serotoninergic fibers in laminae I, II, V, VII, and X had increased compared to younger ages. Serotonin-positive fibers were not evenly distributed in the LMC of the adult chicken; rather, they were densely aggregated around the soma and proximal dendrites of motoneurons in the dorsolateral LMC. Many neuronal soma in the medial and intermediate regions of the LMC lacked serotoninergic fibers.

Cell death in the inner and outer nuclear layers of the developing cat retina

The occurrence and distribution of pyknotic profiles (dying cells) in the inner (INL) and outer nuclear layers (ONL) were mapped from radial sections through 13 kitten retinae aged from the 30th postconceptional day (30PCD) to the 81PCD (17th postnatal day). Cell death in the INL occurs in two consecutive waves, each beginning at the area centralis and proceeding toward the retinal edge. The first wave involves cells located in the inner part of the INL, adjacent to the inner plexiform layer. It begins on the 50PCD and lasts until the 66PCD, with a peak of 72,000 profiles on the 58PCD. This wave of cell death seems to be associated with the formation of the inner plexiform layer and may be due to the elimination of amacrine cells which have failed to establish sustaining connections with their appropriate target cells. The second wave involves cells located in the outer part of the INL, adjacent to the outer plexiform layer. It begins on the 60PCD and lasts until after the 81PCD, with a peak of 65,000 on the 71PCD. This wave appears to be due to the elimination of ectopic photoreceptor cells displaced by the formation of the OPL and presumably prevented from establishing sustaining connections with their appropriate target cells. A phase of cell death in the ONL begins on about the 42PCD and lasts until after the 81PCD, with a peak of 10,300 pyknotic profiles on the 60PCD.(ABSTRACT TRUNCATED AT 250 WORDS)

Cytoarchitecture in cultured rat neocortex explants

Neocortex explants obtained from 6-day-old rat pups and cultured in a serum-free medium from 5 hr to 13 days in vitro (DIV) show preservation of cytoarchitectural characteristics. Major changes in the size of the explants and their layers occur during the first 2 DIV. A radial arrangement of neurons within layer 2-3-4, which becomes apparent between 2 and 10 DIV, suggests an advance in maturation in culture. In contrast to the situation in vivo, a distinct layer 4 cannot be consistently identified. During the first 2 DIV, a transposition of cells into the pial direction can be seen. Individual degenerated cells are spotted especially in layer 5. The presence of these cells amidst healthy neurons suggests that their death has a functional cause. Neurons at the ventricular border of layer 6 become relatively large in comparison with other neurons. Within the cultured tissue there is a marked increase in GFA reactivity compared to the situation in vivo. The described results clearly indicate that in these cultured explants, both similarities and differences are of interest for studies on the formation of neuronal circuitry within the cerebral cortex.

Diffusible proteins prolong survival of dorsal lateral geniculate neurons following occipital cortex lesions in newborn rats

Removal of the occipital cortex in newborn rats results in the rapid and nearly complete degeneration of the dorsal lateral geniculate nucleus (dLGN) in 5 days. In previous studies we have shown that transplants of embryonic posterior cortex neurons, which are allowed to develop in culture for 5 days prior to transplantation into the site of the lesion, prolong the survival of a particular population of host dLGN neurons for an additional week. In this study we tested the possibility that the transplant cells synthesize diffusible proteins which are responsible for this neurotrophic effect. Culture medium conditioned by explants of embryonic occipital cortex and diencephalon was concentrated by vacuum dialysis or ultrafiltration through membranes with at least a 10-kDa cut-off. This concentrated medium was loaded into polyacrylamide or sodium alginate gels which were then implanted into the cavity of the lesion. Five days after implantation, the alginate-conditioned-medium implants result in a 3-fold increase in dLGN survival compared to unconditioned medium controls, while a two-fold increase in survival of the nucleus is found with the polyacrylamide-conditioned-medium implants. Proteolysis of the conditioned medium eliminates all neurotrophic activity. The results suggest that the death of dLGN neurons following the cortical lesion is due to the loss of diffusible proteinaceous neurotrophic factors--factors that may operate during normal in vivo development of the geniculocortical pathway.

Chronic application of curare does not increase the level of motoneuron survival-promoting activity in limb muscle extracts during the naturally occurring motoneuron cell death period

Naturally occurring motoneuron cell death during development is a well-described phenomenon and the existence of a survival factor provided by target muscles has been postulated. Blockade of activity by chronic application of a neuromuscular junction blocker rescues almost all motoneurons from cell death. The present study was conducted in order to examine the possibility that the motoneuron survival-promoting activity in muscles increases following activity blockade. Cell culture was used to assess the degree of motoneuron survival-promoting activity present in muscle extracts. Embryonic chick motoneurons were labeled by injecting the water-insoluble fluorescent dye, DiI (Molecular Probes, Inc.) into the spinal nerves both before and during the cell death period. The labeled cells extending long neurites were counted after 2 days of culture as viable motoneurons in low-density heterogeneous cell cultures. The culture medium, Ham F12/DMEM (1:1 mixture) supplemented with 10% horse serum, 5% chick serum, and 5% fetal calf serum, was employed as a basic culture medium for assessing motoneuron survival factor, since it supported the survival of a significantly higher number of motoneurons derived from embryos before cell death than those during the cell death period, thus representing the motoneuron's requirement for survival factor in vivo. The number of surviving motoneurons clearly increased in proportion to the amount of muscle extract added to the culture medium. In comparison with control chick embryos, the dose-response relation between the number of surviving motoneurons and the amount of muscle extract added did not change when embryos were used after chronic application of curare. These results therefore indicate that survival factor derived from target muscle is crucial to the in vitro motoneurons during the cell death period, but do not support the idea that inactive muscle contains a higher amount of the survival factor.

Neuronal loss in the spiral ganglion of young rats

A quantitative study of spiral ganglion neurones was performed in rats during postnatal days 4, 5, 6, 30 and 60. There are 25,194 +/- 462 ganglion cells on postnatal day 4, abruptly falling to 18,809 +/- 514 on the 6th postnatal day. This neuronal loss accounts for the 22% of the overall ganglion cell population. The number of neurones remains almost unchanged from the 6th to the 60th postnatal day. This numerical variation in the neuronal population of the spiral ganglion seems to be related to the changes that take place during cochlear synaptogenesis, at the end of the first postnatal week, on the base of the outer hair cells. These changes involve competition among efferent endings approaching the cell and some afferents connected with it at birth, that disappear as a result of such a competition.

Distinct neurotrophic factors from skeletal muscle and the central nervous system interact synergistically to support the survival of cultured embryonic spinal motor neurons

Motor neurons isolated from 6-day-old embryonic chick spinal cords require muscle extract for survival in culture; however, it was found, that some motor neurons, identified by retrograde labeling with rhodamine, will survive in mixed spinal cell cultures in the absence of the extract. The motor neuron survival-promoting activity produced by spinal cells is soluble and differs from the factor present in muscle extract, the two activities acting in a synergistic manner: the spinal cell activity potentiated that of muscle to decrease its ED50 by an order of magnitude, the motor neuronal survival (30%) seen in the presence of both factors being more than the sum of their individual activities. This synergism was shown to be restricted to the action of the spinal cell factor on motor neurons, no effect of the factor being noted with sympathetic neurons. As a series of defined growth and survival factors present in the central nervous system (nerve growth factor, brain-derived neurotrophic factor, acidic and basic fibroblast growth factors) had no effect on motor neuron survival, we conclude that the molecule responsible for the motor neuron survival-promoting activity of the spinal cells is a previously undefined factor.

Programmed death of peripheral pioneer neurons in the grasshopper embryo

The Ti1 pioneer neurons arise at the distal tip of the metathoracic leg in the grasshopper embryo, and are the first neurons in the limb bud to extend axons to the central nervous system (C. M. Bate (1976) Nature (London) 260, 54-56; H. Keshishian (1980) Dev. Biol. 80, 388-397). By providing a neural pathway along which growth cones of later arising neurons migrate, these pioneer axons establish the route of one of the major nerve trunks in the leg (Keshishian, 1980; R. K. Ho and C. S. Goodman (1982) Nature (London) 297, 404-406; D. Bentley and H. Keshishian (1982) Science 218, 1082-1088). Here, we demonstrate that at the 55-59% stage of development, the two Ti1 pioneer neurons undergo programmed death. The role which these pioneers serve in establishing a nerve route appears to be their only function, and may be important for the normal development of the peripheral nervous system. The Ti1 pioneers provide an example of a previously hypothesized class (J. W. Truman (1984) Annu. Rev. Neurosci. 7, 171-188) of programmed neuron death: obsolete neurons whose function was developmental rather than behavioral.

Cell death in the developing retinal ganglion cell layer of the wallaby Setonix brachyurus

The distribution and number of dying cells in the developing retinal ganglion cell layer of the wallaby Setonix brachyurus were assessed by using cresyl violet stained tissue. The density of dying cells has been expressed per 100 live cells for the entire retinal surface, data being presented as a grid of 500 micron squares. For statistical analysis, retinae were divided into 8 regions; dorsal, ventral, nasal, and temporal quadrants, each further divided into center and periphery. This method allowed comparison of the extent of cell death at different retinal locations as the high density area centralis of live cells developed temporal to the optic disk from 60 days onward. Between 30 and 70 days, dying cells were seen across the entire retina; beyond 100 days very few were seen. Initially, there was a significantly higher incidence of dying cells in the central retina compared to the periphery, whereas from 50 days this situation was reversed. Analysis of the central retina before and during area centralis formation consistently indicated a significantly lower number of dying cells per 100 live cells in temporal compared to other retinal quadrants. This differential pattern suggests that cell death lowers live cell densities less in the emerging area centralis than elsewhere, and therefore must play a part in establishing live cell density gradients. However, we cannot exclude the possibility that other factors are also instrumental. Indeed, factors such as areal growth (Beazley et al., in press) presumably operate at later stages since live cell density gradients continue to be accentuated even after cell death is complete. Numbers of dying cells peaked by 50 days, reaching approximately 1% of the live cell population. At this stage, counts were also maximal for live cells with values up to 30% above the adult range.

Changes in the numbers of retinal ganglion cells and optic nerve axons in the developing albino rabbit

In albino rabbits aged from the 16th postconceptional day (16PCD) to adulthood, the number of axons in the optic nerves were estimated from sample areas totalling 1-12% of the cross-sectional area of the nerve. On the 16PCD there are about 20,000 axons in the optic stalk. The number of axons in the retrobulbar part of the optic nerve reaches a peak value of 766,000 on the 23PCD, and then decreases to about 350,000 by the 32PCD (the day of birth). The number of axons does not change between the 32PCD and 50PCD, but thereafter it slowly decreases, reaching the adult number (294,000) by the 84PCD. A similar trend is apparent in pigmented animals. Thus, on the 25PCD there are 736,000 axons in the retrobulbar part of the optic nerve and the number decreases to 428,000 by the 31PCD. In the adult pigmented rabbit there are 280,000 axons in the optic nerve. In animals younger than the 32PCD, growth cones are present, and the number of axons in the prechiasmal part of the optic nerve was 8-22% lower than in the retrobulbar part of the same nerve. These observations suggest that there is a continued outgrowth of axons from the eye towards the target nuclei. By the 32PCD, the numbers of axons in the retrobulbar and prechiasmal parts of the nerve were very similar, suggesting that by this age all axons had reached the chiasm. The numbers of retinal ganglion cells (RGCs) labelled by massive injections of horseradish peroxidase into the retino-recipient nuclei were estimated in albino rabbits aged from the 24PCD to adulthood. RGCs were counted in evenly spaced sample areas totalling 4-11% of the retinal area. On the 24PCD, the number of labelled RGCs (500,000) was lower than the number of axons in the optic nerve (probably because not all RGC axons had reached their target nuclei by this age). However, by the 27PCD the number of labelled RGCs (550,000) was very similar to the number of prechiasmal axons (568,000). At all ages thereafter, the numbers of both RGCs and axons were very similar, with adult RGC numbers (about 291,000) being reached by the 85PCD. We conclude that axon loss in the rabbit optic nerve after the 27PCD is almost certainly due to the elimination (presumably death) of the parent RGCs, and we suggest that RGC death is also the most likely cause of axon loss prior to the 27PCD.(ABSTRACT TRUNCATED AT 400 WORDS)

Enhanced Purkinje cell survival in granuloprival cerebellar cultures

The number of large cortical neurons that survived in cerebellar cultures in which granule cells had been destroyed by exposure to cytosine arabinoside was 3-4 times the number in normal cultures. Transplantation of granuloprival cerebellar cultures with granule cells and glia resulted in a reduction of the large cortical neuron population (predominantly Purkinje cells) to normal, while the number of such neurons remained elevated after transplantation with glia alone. These results indicated that granule cells were critical for the reduction of large cortical neurons. The rescue of large cortical neurons in granuloprival cultures was attributed to an expanded target field for Purkinje cell axon collateral projections.

Positron emission tomography study of human brain functional development

From over 100 children studied with 2-deoxy-2[18F]fluoro-D-glucose and positron emission tomography we selected 29 children (aged 5 days to 15.1 years) who had suffered transient neurological events not significantly affecting normal neurodevelopment. These 29 children were reasonably representative of normal children and provided an otherwise unobtainable population in which to study developmental changes in local cerebral metabolic rates for glucose (lCMRGlc). In infants less than 5 weeks old lCMRGlc was highest in sensorimotor cortex, thalamus, brainstem, and cerebellar vermis. By 3 months, lCMRGlc had increased in parietal, temporal, and occipital cortices; basal ganglia; and cerebellar cortex. Frontal and dorsolateral occipital cortical regions displayed a maturational rise in lCMRGlc by approximately 6 to 8 months. Absolute values of lCMRGlc for various grey matter regions were low at birth (13 to 25 mumol/min/100 gm), and rapidly rose to reach adult values (19 to 33 mumol/min/100 gm) by 2 years. lCMRGlc continued to rise until, by 3 to 4 years, it reached values of 49 to 65 mumol/min/100 gm in most regions. These high rates were maintained until approximately 9 years, when they began to decline, and reached adult rates again by the latter part of the second decade. The highest increases of lCMRGlc over adult values occurred in cerebral cortical structures; lesser increases were seen in subcortical structures and in the cerebellum. This time course of lCMRGlc changes matches that describing the process of initial overproduction and subsequent elimination of excessive neurons, synapses, and dendritic spines known to occur in the developing brain. The determination of changing metabolic patterns accompanying normal brain development is a necessary prelude to the study of abnormal brain development with positron emission tomography.

Development of the brachial lateral motor column in the wingless mutant chick embryo: motoneuron survival under varying degrees of peripheral load

Survival of motoneurons in the lateral motor column (LMC) of the chick embryo is known to depend on the periphery. How this dependence relates to the normally occurring death of motoneurons is unknown. Analysis of the time course of LMC cell loss in the absence of varying amounts of limb musculature could help bring about an understanding of this relationship. We undertook this analysis by studying LMC development in wingless chick embryos. Grossly these embryos lack wings, but we have reported that some of them possess more than 40% of the normal volume of wing bud-derived muscles (M.E. Lanser and J.F. Fallon, Anat. Rec. 217:61-78, 1987). In the present work we compared the time course of LMC development in wingless embryos that possessed varying amounts of wing bud-derived musculature with that in normal embryos. In normal embryos little cell loss occurs from the brachial LMC prior to day 8 (15% of the total cell loss). Most of the normal cell loss occurs between 8 and 10 days (62% of the total cell loss). In the wingless LMC, anywhere from 55% to 70% of the total cell loss occurs before day 8. The death of motoneurons prior to day 8 is proportional to the amount of wing bud musculature eliminated by the mutation. Cell loss after day 8 is proportional to the amount of wing bud musculature spared by the mutation. Therefore, when the limb is missing, most motoneurons die before the major period of cell loss even begins in the normal LMC. Counts of dead cells in the LMC also support this conclusion. In addition, curves plotting the rates at which cells are lost from the brachial LMC provide a suggestion that normal cell loss is biphasic and that limb removal enhances primarily the first phase of cell loss. These data suggest that the majority of motoneurons may die for different reasons in the normal and the limb-deprived LMCs. Overall, the number of motoneurons surviving in the brachial LMC is proportional to the volume of wing bud-derived muscle present. However, as the muscle volume approaches zero, motoneuron number does not. This suggests that most, but not all, motoneurons depend on limb bud-derived muscles for survival. Finally, the decreased motoneuron number in the wingless LMC, when compared to normal after the cell death period, cannot be totally accounted for by the additional loss of cells that occurred during the cell death period in the wingless LMC.(ABSTRACT TRUNCATED AT 400 WORDS)

Morphology and quantitative changes of transient NPY-ir neuronal populations during early postnatal development of the cat visual cortex

The early postnatal development of neuropeptide Y-containing neurons in the visual cortex of the cat was analyzed. Immunohistochemistry reveals several stages of morphological differentiation and degeneration. Completely undifferentiated neurons have very small somata with nuclei surrounded by a thin rim of cytoplasm and processes unclearly differentiated into dendrites and axons. Processes bear growth cones. Differentiating neurons show an increase in soma size and complexity of processes. Axons are recognizable. Fully differentiated neurons have well-defined axonal and dendritic patterns. Degenerating neurons are identified by thick, heavily beaded processes covered by hairy appendages and vacuolar inclusions in the somata. Cell death is expressed by shrunken somata and lysed, fragmented processes. According to their postnatal time course of differentiation and/or degeneration, NPY-immunoreactive neurons, which form several morphologically distinct cell types, are grouped into 3 neuronal populations. (1) Pseudopyramidal cells, bitufted "rectangular" cells with wide dendritic fields, unitufted cells, and small multipolar cells are located in the gray matter and have a rather primitive morphology resembling cell types found in lower vertebrate cortex and tectum. They constitute a first transient neuronal population, because all neurons are fully differentiated at birth and become largely eliminated by postnatal day (P) 12. (2) Axonal loop cells are mainly located in the white matter. Their most prominent feature is an often long hairpin loop formed by either the main axon itself or by a major collateral. The axonal branches pass through the cortex to connect the white matter and layer I. Axons do not form local plexusses and terminal elements in the gray matter. Neurons differentiate perinatally, form a first peak from P6 to P10, followed by a decrease in cell number and innervation density at P12, followed by a second peak from P15 to P20. After P20 the number of axonal loop cells steadily decreases, and they become eliminated by P48. (3) A third population consists of neurons with a higher degree of axonal ramification and a variety of axonal patterns. Early members are located mainly at the layer VI/white matter border, differentiate during the first postnatal week, and give rise to a diffuse innervation of the gray matter without forming specific terminal elements. Some of the early axonal patterns persist into adulthood, whereas others are not found in the adult brain.(ABSTRACT TRUNCATED AT 400 WORDS)

Intrinsically directed pruning as a mechanism regulating the elimination of transient collateral pathways

In neonatal cats, neurons in frontoparietal areas of the cerebral cortex have axons which branch, some collaterals project transiently to the cerebellum, whereas others project by way of the pyramidal tract to the brainstem and spinal cord and persist into the adult. If cerebrocerebellar collaterals are eliminated simply because they are exuberant, then experimentally removing the collaterals in the pyramidal tract should cause the normally ephemeral projections to the cerebellum to persist. To test this hypothesis, the pyramidal tract was cut unilaterally at the pontomedullary junction in 5-9-postnatal-day-old (PND) cats, and 35-68 days later the frontoparietal cortex ipsilateral to the pyramidotomy was injected with tritiated amino acids. From the end of the lesioned pyramidal tract, labeled axons were traced into pathways that descended aberrantly into the caudal medulla and spinal cord, but there was never any transported label in the cerebellum. In a second series of experiments, the fluorescent dye Fast blue (FB) was injected into the spinal cord (2-5 PND) prior to cutting the contralateral pyramidal tract (9-12 PND) to determine if the pyramidotomy caused the axotomized cortical neurons to die. There were no neurons labeled with FB in the frontoparietal cortex on the side of the pyramidotomy, but many retrogradely labeled neurons were present contralaterally in the cortex, suggesting that the pyramidotomy caused the death of all axotomized cortical neurons. In a final set of experiments, FB was injected into the spinal cord and the cerebellar cortex was ablated (2-3 PND) prior to cutting the pyramidal tract (9-72 PND). Cerebellar decortication results in the persistence of cerebrocerebral projections to the partially deafferented deep nuclei, therefore injections of Nuclear yellow (NY) or Diamidino yellow (DY) were made later (32-86 PND) into the cerebellar nuclei on the side of the decortication to determine if these projections persist in pyramidotomized cats. After pyramidotomies at 9 PND, there were no neurons labeled with fluorescent dyes in the ipsilateral frontoparietal cortex, indicating that the cerebrocerebellar collaterals, even under experimental conditions which normally cause them to persist, could not sustain the axotomized cortical neurons. Pyramidotomies at 24 PND or later did not cause all axotomized neurons to die since neurons labeled with FB were present in the ipsilateral cortex. These findings suggest that during development of corticosubcortical pathways there is a hierarchical.(ABSTRACT TRUNCATED AT 400 WORDS)

Presynaptic terminals persist following degeneration of "flight" muscle during development of a flightless grasshopper

We have studied the development of a neuromuscular system for which mature function has been lost through evolution in the grasshopper, Barytettix psolus (Cohn and Cantrall, 1974). Barytettix is flightless throughout life and has only vestigial wings that are incapable of active movement. Adult Barytettix lack muscles homologous to the indirect flight muscles of locusts and grasshoppers that fly, while other thoracic muscles are similar. We have found, using light and electron microscopic examination of tissues from various developmental stages, that the metathoracic dorsal longitudinal muscle is present and is innervated during nymphal life but is absent in adults. Yet its nerve persists and, in the adult, contains axonal presynaptic specializations opposite inappropriate targets such as glial processes and basal lamina. Our findings indicate that selective muscle death during development is one mechanism underlying the reduction of the flight system of Barytettix through evolution. The finding that presynaptic terminals persist in the absence of the muscle indicates that the muscle and its innervation follow programs of development that are at least partially independent and reinforces the concept that in insects motorneurons, and perhaps neurons in general, are not dependent upon trophic influences from their targets for survival and maintenance of their differentiated phenotype.

Muscle-derived factors that support survival and promote fiber outgrowth from embryonic chick spinal motor neurons in culture

The purposes of the experiments reported is to provide an unambiguous demonstration that embryonic skeletal muscle contains factors that act directly on embryonic spinal motor neurons both to support their survival and to stimulate the outgrowth of neurites. Cells of lumbar and brachial ventral spinal cords from 6-day-old chick embryos were separated by centrifugation in a two-step metrizamide gradient, and a motor neuron enriched fraction was obtained. Motor neurons were identified by retrogradely labeling with rhodamine isothiocyanate, and were enriched fourfold in the motor neuron fraction relative to unfractionated cells. In culture, the isolated motor neurons died within 3-4 days unless they were supplemented with embryonic chick skeletal muscle extract. Two functionally distinct entities separable by ammonium sulfate precipitation were responsible for the effects of muscle extracts on motor neurons. The 0-25% ammonium sulfate precipitate contained molecules that alone had no effect on neuronal survival but when bound to polyornithine-coated culture substrata, stimulated neurite outgrowth and potentiated the survival activity present in muscle. Most of this activity was due to a laminin-like molecule being immunoprecipitated with antisera against laminin, and immunoblotting demonstrated the presence of both the A and B chains of laminin. A long-term survival activity resided in the 25-70% ammonium sulfate fraction, and its apparent total and specific activities were strongly dependent on the culture substrate. In contrast to the motor neurons, the cells from the other metrizamide fraction (including neuronal cells) could be kept in culture for a prolonged time without addition of exogenous factor(s).

The development of motoneurons in the embryonic spinal cord of the mouse mutant, muscular dysgenesis (mdg/mdg): survival, morphology, and biochemical differentiation

Motoneuron development was studied in the spinal cord of the mouse mutant, muscular dysgenesis, between embryonic days (E) 13 and 18. Dysgenic embryos are characterized by the absence of neuromuscular activity (motility) and exhibit a number of other striking changes in neuromuscular development. Many of these changes have also been observed in chick embryos chronically treated with neuromuscular blocking agents that suppress motility. Motoneuron survival, as well as several other aspects of neuronal development, was examined in the thoracic and lumbar spinal cords of mutant and control embryos. There was a significant decrease in motoneuron numbers in control embryos indicating the presence of naturally occurring cell death in the mouse spinal cord. At all ages examined, the dysgenic embryos had significantly more healthy and significantly fewer degenerating motoneurons than controls. There were no differences in the number of dorsal root ganglion neurons or in any of the other morphometric parameters examined between mutant and control embryos. Creatine kinase activity, a marker for myofiber maturation, was significantly reduced in the limb musculature of mutant embryos. Choline acetyltransferase activity was significantly increased in the spinal cord of mutant embryos. No significant differences were observed in spinal cord levels of acetylcholinesterase activity between control and mutant embryos. The absence of muscle contractions in the dysgenic mouse is associated with a number of changes in neuromuscular development, including a substantial reduction of naturally occurring motoneuron death.