Trophic influence of the distal nerve segment on GABAA receptor expression in axotomized adult sensory neurons

The depolarizing action of γ-aminobutyric acid (GABA), or the GABA_A receptor agonist muscimol, on rat dorsal root (L4 and L5) fibers is attenuated following transection, but not crush, of the sciatic nerve (15). Following discrete nerve crush, axons actively regenerate and contact both the distal nerve segment and the peripheral target tissues. The aim of the present study was to distinguish between these two regions as possible sources of trophic support for retrograde maintenance of dorsal root GABA receptor sensitivity. A surgical procedure was employed to permit a delimited segment of axonal regeneration while prohibiting reestablishment of end organ innervation; the sciatic nerve was crushed and a ligature was placed 3 cm distal to the crush site. Under these conditions, the injury-induced decrement in the dorsal root GABA response, observed between 12 and 21 postoperative days, was significantly attenuated relative to that of ligated nerves, in which regeneration into the distal stump does not occur. The data suggest that nerve transection by ligation restricts trophic support for maintenance of GABA receptor expression in dorsal root ganglion (DRG) neurons. Furthermore, during regeneration the denervated distal nerve segment assumes a neurotrophic role in the maintenance of dorsal root GABA sensitivity, consistent with the hypothesis that growth factors derived from reactive Schwann cells may positively regulate the expression of receptors on axotomized sensory neurons.     

Thyroid hormone reduces the loss of axotomized sensory neurons in dorsal root ganglia after sciatic nerve transection in adult rat

We have shown that a local administration of thyroid hormones (T3) at the level of transected rat sciatic nerve induced a significant increase in the number of regenerated axons. To address the question of whether local administration of T3 rescues the axotomized sensory neurons from death, in the present study we estimated the total number of surviving neurons per dorsal root ganglion (DRG) in three experimental group animals. Forty-five days following rat sciatic nerve transection, the lumbar (L4 and L5) DRG were removed from PBS-control, T3-treated as well as from unoperated rats, and serial sections (1 microm) were cut. The physical dissector method was used to estimate the total number of sensory neurons in the DRGs. Our results revealed that in PBS-control rats transection of sciatic nerve leads to a significant (P < 0.001) decrease in the mean number of sensory neurons (8743.8 +/- 748.6) compared with the number of neurons in nontransected ganglion (mean 13,293.7 +/- 1368.4). However, administration of T3 immediately after sciatic nerve transection rescues a great number of axotomized neurons so that their mean neuron number (12,045.8 +/- 929.8) is not significantly different from the mean number of neurons in the nontransected ganglion. In addition, the volume of ganglia showed a similar tendency. These results suggest that T3 rescues a high number of axotomized sensory neurons from death and allows these cells to grow new axons. We believe that the relative preservation of neurons is important in considering future therapeutic approaches of human peripheral nerve lesion and sensory neuropathy.     

Peripheral nerve graft with immunosuppression modifies gene expression in axotomized CNS neurons

Adult central nervous system (CNS) neurons do not regenerate severed axons unaided but may regenerate axons into apposed predegenerated peripheral nerve grafts (PNGs). We examined gene expression by using microarray technology in laser-dissected lateral vestibular (LV) neurons whose axons were severed by a lateral hemisection at C3 (HX) and in lateral vestibular nucleus (LVN) neurons that were hemisected at C3 and that received immunosuppression with cyclosporine A (CsA) and a predegenerated PNG (termed I-PNG) into the lesion site. The results provide an expression analysis of temporal changes that occur in LVN neurons in nonregenerative and potentially regenerative states and over a period of 42 days. Axotomy alone resulted in a prolonged change in regulation of probe sets, with more being upregulated than downregulated. Apposition of a PNG with immunosuppression muted gene expression overall. Axotomized neurons (HX) upregulated genes commonly associated with axonal growth, whereas axotomized neurons whose axons were apposed to the PNG showed diminished expression of many of these genes but greater expression of genes related to energy production. The results suggest that axotomized LVN neurons express many genes thought to be associated with regeneration to a greater extent than LVN neurons that are apposed to a PNG. Thus the LVN neurons remain in a regenerative state following axotomy but the conditions provided by the I-PNG appear to be neuroprotective, preserving or enhancing mitochondrial activity, which may provide required energy for regeneration. We speculate that the graft also enables sufficient axonal synthesis of cytoskeletal components to allow axonal growth without marked increase in expression of genes normally associated with regeneration.     

Effects of nerve growth factor on the survival and synaptic function of Ia sensory neurons axotomized in neonatal rats

Sensory neurons with small diameters (A delta and C cells) are known to be responsive to exogenous NGF even at postnatal stages. We have examined whether large Group Ia sensory neurons (A alpha cells) arising from muscle spindles are also responsive to NGF in neonatal rats. For this purpose, monosynaptic excitatory postsynaptic potentials (EPSPs) were evoked in spinal motoneurons by Group Ia muscle afferent volleys. When a muscle nerve was crushed on the day after birth, the monosynaptic EPSPs elicited by afferent volleys from the muscle were depressed within several weeks. This synaptic depression was partially reversed by daily treatment with NGF. NGF treatment also enhanced the EPSPs evoked by stimulation of intact muscle nerves, but this effect was less marked than that on the EPSPs produced by stimulation of the previously crushed muscle nerve. Exogenous NGF was effective for the EPSPs when the treatment began on the day after birth but not when the treatment began 4 d after birth. Following crush of a muscle nerve on the day after birth, about 45% of the sensory neurons derived from the muscle were lost. The cell death of small sensory neurons was prevented by daily treatment with NGF, whereas the NGF treatment was ineffective in preventing the cell death of large sensory neurons. The results indicate that Group Ia sensory neurons are responsive to NGF during early postnatal life.     

NGF rescues substance P expression but not neurofilament or tubulin gene expression in axotomized sensory neurons.

A reduction in the supply of retrogradely transported NGF has been proposed as a possible signal for the axotomy response in dorsal root ganglion (DRG) neurons. Components of the axotomy response that have previously been well characterized in axotomized DRG cells include changes in cytoskeletal gene expression and changes in the expression of neurotransmitters/neuromodulators such as substance P. In this study, we examined the role of NGF in the axotomy response by examining protein synthesis and mRNA levels of the low-MW neurofilament protein (NF-L) and beta-tubulin in DRG cells at 1, 7, and 12 d after axotomy with and without continuous administration of exogenous NGF. We also examined substance P levels in the DRG by immunocytochemistry under the same experimental conditions. Sciatic nerves of adult male rats were unilaterally transected at the midthigh level, and the proximal nerve stumps were placed into Silastic tubes connected to osmotic minipumps that were filled with biologically active NGF. NGF (0.5 mg/ml in saline) was continuously infused (0.5 microliter/hr) onto the proximal stumps of transected sciatic nerves for 1-12 d. Control animals were prepared in an identical fashion except that the nerves were treated with saline alone. At death, DRGs were removed from the animals; the L4 experimental DRGs (axotomized) and contralateral L4 DRGs (uninjured) were used immediately for protein synthesis experiments, while the experimental and contralateral L5 DRGs were fixed in 4% paraformaldehyde and subsequently used for in situ hybridization and immunocytochemistry. From another set of experimental animals, the L4 and L5 DRGs were harvested and used for total RNA isolation and RNA blotting experiments. Immunocytochemical studies using a polyclonal antibody to substance P showed that the immunodetectable levels of this peptide decreased to undetectable levels in DRG neurons after axotomy and saline administration. However, in axotomized neurons treated with NGF, the level of immunodetectable substance P did not decrease, but instead, increased over even that present in normal DRG neurons. Pulse labeling of DRGs with 35S-methionine:cysteine followed by 2-dimensional (2D) gel electrophoresis and fluorography revealed that the synthesis of neurofilament (NF) proteins was decreased, while that of tubulin was increased, 12 d after sciatic nerve transection. NGF administration to axotomized neurons did not alter this pattern. Quantitative analysis of in situ hybridizations of DRG neurons and RNA blot analysis with cDNA probes specific for NF-L and beta-tubulin mRNAs showed that NGF treatment of axotomized DRGs did not significantly affect cytoskeletal gene expression at the mRNA level.(ABSTRACT TRUNCATED AT 400 WORDS)     

Reducing p75 nerve growth factor receptor levels using antisense oligonucleotides prevents the loss of axotomized sensory neurons in the dorsal root ganglia of newborn rats

The low-affinity p75 receptor for nerve growth factor (p75NGFR) has been implicated in mediating neuronal cell death in vitro. A recent in vitro study from our laboratory showed that the death of sensory neurons can be prevented by reducing the levels of p75NGFR with antisense oligonucleotides. To determine if p75NGFR also functions as a death signal in vivo, we have attempted to reduce its expression in peripheral sensory neurons by applying antisense oligonucleotides to the proximal end of the transected sciatic or median and ulnar nerves. We report here that antisense oligonucleotides, when applied to the proximal stump of a transected peripheral nerve, are retrogradely transported and effectively reduce p75NGFR protein levels in sensory neurons located in the dorsal root ganglia. Furthermore, treatment of the proximal nerve stump with antisense p75NGFR oligonucleotides significantly reduced the loss of these axotomized sensory neurons. These findings further support the view that p75NGFR is a death signaling molecule and that it signals death in axotomized neurons in the neonatal sensory nervous system. © 1996 Wiley-Liss, Inc.     

Discharge characteristics of axotomized abducens internuclear neurons in the adult cat

The aim of the present work was to characterize the axotomy-induced changes in the discharge properties of central nervous system neurons recorded in the alert behaving animal. The abducens internuclear neurons of the adult cat were the chosen model. The axons of these neurons course through the contralateral medial longitudinal fascicle and contact the medial rectus motoneurons of the oculomotor nucleus. Axotomy was carried out by the unilateral transection of this fascicle (right side) and produced immediate oculomotor deficits, mainly the incapacity of the right eye to adduct across the midline. Extracellular single-unit recording of abducens neurons was carried out simultaneously with eye movements. The main alteration observed in the firing of these axotomized neurons was the overall decrease in firing rate. During eye fixations, the tonic signal was reduced, and, on occasion, a progressive decay in firing rate was observed. On-directed saccades were not accompanied by the high-frequency spike burst typical of controls; instead, there was a moderate increase in firing. Similarly, during the vestibular nystagmus, neurons hardly modulated during both the slow and the fast phases. Linear regression analysis between firing rate and eye movement parameters showed a significant reduction in eye position and velocity sensitivities with respect to controls, during both spontaneous and vestibularly induced eye movements. These firing alterations were observed during the 3 month period of study after lesion, with no sign of recovery. Conversely, abducens motoneurons showed no significant alteration in their firing pattern. Therefore, axotomy produced long-lasting changes in the discharge characteristics of abducens internuclear neurons that presumably reflected the loss of afferent oculomotor signals. These alterations might be due to the absence of trophic influences derived from the target.     

Galanin expression in sensory neurons after nerve compression or transection

Galanin is probably involved in nociceptive sensory processing in spinal cord. We investigated whether a common injury, peripheral nerve compression, induced up-regulation of galanin (immunocytochemistry) in sensory neurons in rats 6 or 14 days post-injury and compared the response with other nerve injuries. Sciatic nerve compression increased the number of galanin positive sensory neurons as compared to uninjured and contralateral dorsal root ganglia. Complete transection was more efficient than a partial transection and a slight compression injury as an inducer of galanin. Mainly small diameter sensory neurons became positive but also some large diameter neurons. We conclude that nerve compression up-regulates galanin in sensory neurons. The extent of the induction could be related to the severity of nerve injury.     

Effects of nerve growth factor on TTX- and capsaicin-sensitivity in adult rat sensory neurons

We have investigated the effects of nerve growth factor (NGF, 2.5 ng/ml for 1–2 weeks) on enriched adult rat dorsal root ganglion (DRG) neurons maintained in cell culture in defined media. Whole-cell recordings in cells cultured in the absence and presence of NGF revealed no significant difference in resting membrane potential and input resistance. However, the threshold for spike generation was significantly lower in untreated cells than in treated cells; −25 ± 1.1mV vs−19 ± 2.2mV, respectively. The sensitivity of the Na⁺ spike to tetrodotoxin (TTX, 1 μM) was different in cells cultured in the absence or presence of NGF. For example, spikes were abolished by TTX in 100% of untreated cells, while in NGF-treated cells the spike was abolished in only 41% of the neurons. Chemosensitivity of DRG neurons was also different in the absence and presence of NGF. For example, the percent of neurons in which a current activated by 8-methyl-N-vanillyl-6-nonenamide (capsaicin, 500 nM) was detected, increased from 18% in untreated cells to 55% in NGF-treated cells. NGF did not influence the number of cells surviving. The results indicate that NGF can regulate TTX and capsaicin sensitivity in these adult rat sensory neurons. Our experimental protocol indicates that this effect is not mediated by a factor in the serum or released from non-neuronal cells.     

BDNF regulation of androgen receptor expression in axotomized SNB motoneurons of adult male rats

Brain-derived neurotrophic factor (BDNF) prevents the axotomy-induced loss of androgen receptor-like immunoreactivity (AR-LI) in the spinal nucleus of the bulbocavernosus (SNB) motoneurons of adult male rats. In this report, we investigated the dose-response effect of BDNF on androgen receptor expression in axotomized SNB motoneurons, and examined whether delayed application of BDNF to the cut SNB axons can completely reverse the axotomy-induced loss of androgen receptor expression. We also used autoradiography to test whether axotomy decreases the ability of SNB motoneurons to accumulate androgens. SNB motoneurons were axotomized bilaterally and BDNF or PBS was applied to the proximal ends of the axons. The percentage of SNB motoneurons expressing medium or high AR-LI was the major measure of androgen receptor expression. AR-LI was significantly higher on the BDNF-treated side than on the contralateral side treated with phosphate-buffered saline (PBS) for all three doses of BDNF (1.45, 2.9, and 5.8 mg/ml) and was higher than in rats treated bilaterally with PBS. Moreover, AR-LI at the highest dose of BDNF was not different from that in intact SNB motoneurons. Delayed application of BDNF to the axotomized SNB motoneurons restored the AR-LI to the intact level. The AR-LI decreased by axotomy started to increase significantly 4 days after BDNF application and returned to the intact level by 10 days. Furthermore, axotomy significantly decreased the percentage of SNB motoneurons to accumulate tritiated testosterone or its metabolites. In conclusion, our data demonstrate that BDNF completely prevents and reverses the axotomy-induced loss of AR-LI. Moreover, decrease of AR-LI by axotomy reflects the decrease in the ability of SNB motoneurons to accumulate androgens.     

Neuroprotective effect of deprenyl in sensory neurons of axotomized dorsal root ganglion

Spinal motoneuron neuroprotection by deprenyl was previously reported; the present study was carried out to evaluate neuroprotectivity in the dorsal root ganglion sensory neuron. The total neuron counts were calculated, and the axotomized sensory neurons of the dorsal root ganglion were significantly lower than those of the unaxotomized sides. Three secondary and three tertiary parameters were used. The secondary parameters were: the percentages of sensory neuron increase at the axotomized side (PNIA) and at the unaxotomized side (PNIU), and the percentage of neuronal response (PNR). The tertiary parameters were: the percentages of maximal response at the axotomized side (PMRA) and at the unaxotomized side (PMRU), and the percentage of maximal relative response (PMRR). Nonlinear statistical analysis using Gaussian, quadratic and logistic models of the tertiary parameters suggested that the data were bell-shape, which indicated that the data were biphasic. The data were divided into ascending and descending sets, and linear regression. They were analyzed according to Bent-hyperbola model and the ascending set was considered as a neurotrophic phase, while the descending one as a neurotoxic phase. The slops of PMRA were higher than that of PMRU, which indicates that the axotomized neurons were more sensitive than the unaxotomized neurons to the protective and neurotoxic effect of deprenyl. Moreover, the results showed that deprenyl had a proliferative effect on the dorsal root ganglion sensory neuron.     

Upregulation of bradykinin B2 receptor expression by neurotrophic factors and nerve injury in mouse sensory neurons

Bradykinin B2 receptor mRNA was detected at low levels, both by RT-PCR and by in situ hybridization, in freshly isolated dorsal root ganglia (DRG) and in ganglia cultured in the absence of neurotrophic factors, but was strongly upregulated by culture in the presence of nerve growth factor (NGF). The effect of NGF is mediated via TrkA receptors. The related neurotrophins, brain-derived neurotrophic factor, neurotrophin-3, and neurotrophin-4, were ineffective in upregulating B2 mRNA, but a small upregulation was seen with the unrelated neurotrophin glial cell line-derived neurotrophic factor (GDNF). Surface membrane B2 receptor expression, detected by immunofluorescence using a B2-specific antibody, was low in outgrowing axons cultured in the absence of neurotrophic factors, but was elevated by addition of NGF or GDNF. Conditioned media prepared by incubating injured nerve, skin, or muscle had a similar effect to NGF in upregulating B2 mRNA and protein expression, and the activity was largely removed by neutralization of NGF in the conditioned medium with an anti-NGF antibody. After nerve crush injury in vivo an enhancement in B2 mRNA expression was seen, peaking after 7 days and returning to precrush levels after 14 days. In all conditions tested, the proportion of neurons expressing B2 mRNA remained the same at around 23% of small neurons, suggesting that upregulation only occurs in the B2-positive neurons. These experiments show that NGF, and to a lesser extent GDNF, upregulates the expression of bradykinin B2 mRNA and B2 receptor protein in the surface membrane of DRG neurons and that NGF is an important factor responsible for upregulating bradykinin B2 receptor expression after nerve crush injury in vivo.     

Morphological response of axotomized septal neurons to nerve growth factor

Septal efferent fibers from the neurons in the medial septal nucleus are destroyed by fimbria-fornix aspirative lesion. In the present study we used quantitative morphometric techniques to evaluate the response of these axotomized septal neurons to a constant infusion of nerve growth factor (NGF). By 2 weeks following the lesion, approximately 75% of the cholinergic neurons had degenerated in the untreated rats. The remaining cholinergic neurons showed few signs of the effect of the lesion when stained for a polyclonal antibody to ChAT and examined in 40-micron-thick sections. In 1-micron-thick sections the remaining ChAT-immunoreactive (IR) neurons also appeared no different from the intact ChAT neurons. However, non-ChAT-IR neurons had a shrunken nucleus, while all other morphometric parameters appeared normal. NGF infusion protected most of the ChAT-IR neurons from degenerating. The saved neurons had the same parameters as the undamaged ChAT-IR neurons when examined in either 40-micron- or 1-micron-thick sections. In addition, the shrunken appearance of the non-ChAT-IR neurons' nuclei was avoided by the NGF infusions. Enlarged ChAT-IR processes were evident in the dorsolateral quadrant of the septum following damage to the fimbria-fornix. NGF-infusions prevented the formation of these processes. Instead, in the treated animals the dorsal lateral quadrant contained a dense plexus of fine ChAT-IR varicosities. Taken together these results demonstrate that NGF not only can protect the cholinergic neurons from axotomy-induced degeneration but can also cause the saved neurons to maintain the same morphometric appearance as intact ChAT-IR neurons.(ABSTRACT TRUNCATED AT 250 WORDS)     

A novel 7-transmembrane receptor expressed in nerve growth factor-dependent sensory neurons

This study reports on the full-length cDNA cloning of a gene identified on the basis of its preferential expression in nerve growth factor, compared with neurotrophin-3-dependent neurons. It encodes a putative 7-transmembrane polypeptide that is distantly related to other members of the G protein-coupled receptor superfamily. Unique features of this receptor include a very long carboxy-terminal tail of 360 amino acids and a specific expression pattern in the chick peripheral nervous system, including nerve growth factor-dependent sensory and sympathetic neurons, as well as enteric neurons. In the central nervous system, the receptor is strongly developmentally regulated and is expressed at high levels in the external granule cell layer of the cerebellum, as well as in motoneurons of the spinal cord, and in retinal ganglion cells.     

Influence of nerve growth factor on neurofilament gene expression in mature primary sensory neurons

To analyze the possible influence of nerve growth factor (NGF) on neurofilament synthesis in primary sensory neurons, adjacent cryostat sections of lumbar dorsal root ganglia (DRG) from adult rats were processed for either NGF-receptor radioautography or in situ hybridization with a neurofilament cDNA probe. Labeling by both procedures was quantified with computer assistance for approximately 300 neurons in each of selected ganglia. For uninjured neurons, no correction was detected between NGF binding and neurofilament mRNA, even after infusion of NGF into the lumbar subarachnoid space for 1 week. One or 3 weeks after sciatic nerve transection, neurofilament labeling densities in large DRG neurons were sharply reduced and the normal bimodal pattern in frequency histograms had become unimodal. Intrathecal infusion of NGF counteracted this injury-induced reduction of neurofilament mRNA but only in neurons with high-affinity NGF receptors. To explain the effects of NGF on axotomized neurons and the normal diversity of neurofilament gene expression among neurons with NGF receptors, we postulate that NGF permits NGF-sensitive DRG neurons to respond differentially to a second factor stimulating neurofilament synthesis.