Short-term estrogen replacement increases beta-preprotachykinin mRNA levels in uninjured dorsal root ganglion neurons, but not in axotomized neurons.

Dorsal root ganglion (DRG) neurons that mediate nociception express the high affinity NGF receptor (trkA) gene and the preprotachykinin (PPT) gene. NGF has been shown to regulate both of these DRG neuronal genes. Our laboratory has shown that these genes are also regulated by estrogen. Long-term daily estrogen replacement, in adult ovariectomized (OVX) rats, causes a coordinate decline in trkA and beta-PPT mRNA levels in lumbar DRG neurons, while short-term estrogen replacement increases trkA mRNA levels in uninjured as well as in axotomized lumbar DRG neurons. The purpose of the current study was to test the hypothesis that short-term estrogen replacement increases DRG beta-PPT mRNA levels in lumbar DRG neurons of OVX rats and that the increase is dependent on target-derived NGF. Sciatic nerve transection (SNT) was used to eliminate target-derived NGF in L4 and L5 DRGs in adult OVX rats. Seven days later, one-half of the SNT and one-half of the animals that had received sham sciatic nerve transactions (SHAM) received two daily injections of estradiol benzoate (EB). The remaining rats received two daily injections of vehicle alone. Quantitative in situ hybridization analyses of sections from L4 and L5 DRGs showed that two daily injections of EB significantly increased beta-PPT mRNA levels in DRGs of SHAM animals, but had no effect on beta-PPT mRNA levels in DRGs from SNT animals. These data coupled with our earlier observations of the effect of short-term estrogen replacement on DRG trkA mRNA levels, indicate that the regulation of DRG beta-PPT mRNA levels by estrogen requires target-derived NGF.

Effects of testosterone on a cross-facial nerve graft model.

A cross-facial nerve graft (CFNG) can be used to restore the blink reflex following facial paralysis. However, the efficacy of a CFNG depends on the ability of regenerating axons to breach two nerve coaptations and reinnervate endplates in denervated muscle. The neurons involved, facial motor neurons, are androgen-dependent. Testosterone enhances facial-nerve regeneration and accelerates blink-reflex recovery following nerve crush. The current study tested the hypothesis that testosterone administered to castrated or normal adult male rats would enhance axonal regeneration through a CFNG, and thereby enhance recovery of the blink reflex. To test this hypothesis, 20 adult male rats were randomly assigned to four groups, (1) normal, intact; (2) castrated; (3) castrated with testosterone proprionate (TP) administration; and (4) normal, intact with TP Each rat underwent transection of the left facial nerve, and a CFNG using the saphenous nerve as an interpositional graft, with coaptations to the ipsilateral and contralateral zygomatic branches, was carried out. Assessments included return of blink reflex, electrophysiology, quantification of motor endplates, and axonal numbers. The data from the blink reflex evaluation, the electrophysiologic assessment, and the endplate quantification suggested that TP may have an effect on regeneration through a CFNG, but that the differences between groups were not statistically significant. In contrast, exogenously administered TP significantly enhanced the number of axons that entered the nerve graft. These data suggest that pharmacologic doses of testosterone may enhance recovery following a CFNG.

Effects of short-term estrogen replacement on trkA mRNA levels in axotomized dorsal root ganglion neurons.

A population of adult dorsal root ganglion (DRG) neurons bind NGF with high affinity and express the trkA gene. In these cells, NGF regulates gene expression and function. Recently, a number of laboratories reported the presence of estrogen receptors in DRG neurons and profound effects of estrogen on DRG gene expression. Our laboratory, for example, has reported a significant and coordinate decrease in DRG trkA and beta-preprotachykinin (beta-PPT) mRNA levels following 90 days of daily estrogen injections to ovariectomized (OVX) rats. These data suggest, as has been suggested for medial septal cholinergic neurons, that estrogen may collaborate with NGF in the regulation of DRG neuronal gene expression and function. The current study examined further this potential collaboration in the DRG by determining the effect of short-term estrogen replacement in OVX rats on DRG trkA mRNA levels following sciatic nerve transection and the resulting removal of a vital source of NGF for those cells. In OVX rats, about 40% of lumbar DRG neurons contained trkA mRNA. Short-term estrogen replacement had no effect on the percentage of neurons containing trkA mRNA, but increased the mean trkA mRNA level in uninjured DRGs of OVX rats by 23%. Axotomy in OVX rats reduced the mean trkA mRNA level by 55% but did not significantly decrease the percentage of neurons containing the mRNA. Estrogen replacement, 7 days after axotomy, partially and significantly restored the mean trkA mRNA level. It was 49% greater than that of the untreated axotomized DRGs. It did not, however, significantly increase the percentage of DRG neurons containing trkA in axotomized DRGs. These observations show that short-term estrogen has an opposite effect on DRG neuronal trkA mRNA levels as compared to that of long-term estrogen demonstrated in our previous study. Moreover, the current data show that estrogen regulates trkA mRNA levels in the absence of target-derived NGF. These data suggest that estrogen may collaborate with NGF in the maintenance of normal adult DRG gene expression and function. Furthermore, these data suggest that loss of estrogen, such as that associated with menopause, may contribute to a decline in DRG neuronal function and an exacerbation of ongoing neuropathic processes.

End-to-side neurorrhaphy: evaluation of axonal response and upregulation of IGF-I and IGF-II in a non-injury model.

This research group has introduced a model of end-to-side neurorrhaphy, in which reinnervation occurs without frank damage to donor axons. The current study used in situ hybridization to test the hypothesis that insulin-like growth factor (IGF-I and IGF-II) mRNA levels increase at the coaptation site and grafted nerve following end-to-side repair, and that this increase is associated with axonal sprouting and growth. One week after end-to-side coaptation, IGF-I mRNA was localized predominantly on the epineurial side of the graft perineurium, while IGF-II was seen mainly on the endoneurial side. IGF-I hybridization was greatest at this time and declined by 2 weeks post-procedure. No changes in IGF mRNA levels occurred in the distal donor nerve. The increase in IGF-I mRNA at 1 week preceded the appearance of myelinated axons. The presence of myelinated axons within the graft 2 weeks after end-to-side coaptation was associated with a decline in IGF-I mRNA. These data are the first to demonstrate increased IGF mRNA levels associated with axonal sprouting and growth following end-to-side neurorrhaphy. Moreover, the findings support those of earlier studies by others implicating IGFs in axonal regeneration. The increase in IGF mRNA during sprouting and axonal growth into an end-to-side coaptation indicates that the local therapeutic augmentation of endogenous IGF levels at the coaptation site may enhance axonal sprouting from a minimally injured donor nerve, and thereby increase the number of axons that reinnervate the graft.

Long-term estrogen replacement coordinately decreases trkA and beta-PPT mRNA levels in dorsal root ganglion neurons.

Estrogen status has profound effects on cutaneous sensitivity in adult female rats. The presence of alpha-estrogen receptor mRNA and protein in NGF-dependent, adult female rat dorsal root ganglion (DRG) neurons raises the possibility that estrogen modulates cutaneous sensation by acting directly on primary afferent neurons, perhaps by altering their sensitivity to NGF. The present study examined the effect of long-term (90 days) daily injections of an estrogen preparation, Premarin (Wyeth-Ayerst, Radnor, PA), to ovariectomized adult rats on lumbar DRG high-affinity NGF receptor, trkA, mRNA levels, and on beta-preprotachykinin (beta-PPT) mRNA levels, which have been shown to be regulated by NGF. Two doses were used in the experiments, the higher dose being 10 times that of the lower dose. Such injections had an effect opposite that reported for short-term, acute estrogen treatment on DRG trkA mRNA levels. The current data show that long-term daily estrogen treatment decreases trkA mRNA levels by 36%. After 90 days of estrogen treatment, no dose effect was evident. Moreover, as would be expected if beta-PPT gene expression is regulated by NGF through the trkA receptor, long-term estrogen treatment decreased DRG neuronal beta-PPT mRNA levels by about 30%. As with trkA, there was no dose effect evident after 90 days of estrogen treatment. These data suggest the possibility that estrogen modulates DRG neuropeptide gene expression and, perhaps, cutaneous sensitivity by regulating NGF receptor gene expression.

Transient lectin binding by white matter tract border zone microglia in the foetal rabbit brain.

Axonal growth cones of developing white matter tracts are guided through the cerebrum by interactions with cell surface and extracellular matrix molecules expressed by glial cells that mediate cell adhesion and contact-dependent inhibition. Specific carbohydrates are considered essential for the proper functioning of these molecular complexes. We studied developmental aspects of complex carbohydrate expression by white matter glia in the foetal rabbit brain using the tomato lectin Lycopersicon esculentum, which has affinity for components of the extracellular matrix proteins and cell surface proteins (N-acetylglucosamine) and activated lysosomal membrane glycoproteins (N-acetyllactosamine). Concentrations of the lectin-positive glia were transiently found immediately adjacent to developing white matter tracts of the foetal rabbit brain from 22 to 32 days' gestation. The number of positive cells markedly diminished by the fourth post-natal day and in the adult brain. The lectin-positive glia did not react with antibody to glial fibrillary acidic protein. However, they did express the macrophage surface antigen, Mac-1, indicating that the lectin binding reflected the presence of microglial activated lysosomal membranes. These data suggest that, in addition to their role as central nervous system scavengers, microglia are involved in a specifically timed function in the neurodevelopmental programme of white matter tract formation.

Streptozotocin-induced diabetes mellitus causes changes in primary sensory neuronal cytoskeletal mRNA levels that mimic those caused by axotomy.

Dorsal root ganglion (DRG) sensory neurons are particularly vulnerable to diabetes mellitus. There is evidence that the disease decreases both circulating and retrogradely transported neurotrophic factors that are essential to the normal maintenance and function of these cells. A substantive loss of trophic support should cause DRG neurons to respond as though they were axotomized and, like an axotomy, cause significant changes in cytoskeletal gene expression within these cells. Such changes might contribute to the deficits in sensory neuronal function that characterize diabetic neuropathy. The current study used quantitative in situ hybridization to test the hypothesis that streptozotocin-induced diabetes, like an axotomy, increases class III beta-tubulin gene expression and decreases neurofilament 68-kDa gene expression in lumbar DRG neurons. In animals that had been diabetic for 8 weeks with mean blood glucose levels of 340 mg/dl, lumbar DRG class III beta-tubulin mRNA mean steady-state levels were twofold higher than those in age-matched nondiabetic controls. Moreover, in the same animals, diabetes decreased lumbar DRG 68-kDa neurofilament mRNA mean steady-state levels by more than half. These data show that diabetes causes changes in primary sensory neuronal cytoskeletal gene expression that mimic those caused by axotomy. Moreover, they support the idea that a loss of neurotrophic support contributes to the pathogenesis of diabetic neuropathy.

Estrogen regulates neurofilament gene expression in adult female rat dorsal root ganglion neurons.

Recently, adult female dorsal root ganglion (DRG) neurons were shown to express the estrogen receptor gene and to bind estrogen. This gene expression and binding is hormone dependent. Moreover, in a subpopulation of DRG neurons, the estrogen receptor is colocalized with high-affinity (trkA) and low-affinity (p75NGFR) receptors for nerve growth factor (NGF). In this NGF-responsive subpopulation of DRG neurons, estrogen regulates expression of the NGF receptor genes and may increase the sensitivity of these cells to the neurotrophin. The present study tested the hypothesis that neurofilament gene expression, which is regulated by NGF in these cells, is dependent on hormone status. In this study, ovariectomized (OVX) rats received either long-term physiological estrogen (conjugated estrogens; Premarin, Wyeth-Ayerst) replacement (low dose) or 10 times physiological levels (high dose). Quantitative in situ hybridization with an RNA probe for the 68-kDa neurofilament mRNA revealed a significant dose-dependent effect of Premarin on DRG neurofilament gene expression. In OVX animals receiving low-dose Premarin replacement therapy the mean steady-state 68-kDa mRNA level was as high as 4 times that of untreated OVX rats. High-dose therapy increased the mean 68-kDa neurofilament steady-state mRNA level to as much as six-fold that observed in untreated OVX animals. The estrogen-dependent upregulation of neurofilament gene expression appeared to occur in all DRG neurons, rather than in a subpopulation of those cells. These data suggest an important role for estrogen in the maintenance and function of primary sensory neurons. Whether estrogen directly regulates neurofilament gene expression or indirectly regulates it by increasing DRG neuronal sensitivity to neurotrophins or other growth factors remains to be determined.

Up-regulation of laminin B2 gene expression in dorsal root ganglion neurons and nonneuronal cells during sciatic nerve regeneration.

A recent study in our laboratory showed that B2 laminin gene is expressed by all L4 and L5 dorsal root ganglion (DRG) neurons as well as by satellite and Schwann cells. Because the laminin B2 subunit has a domain that supports neurite extension in culture, the present study was undertaken to test the hypothesis that laminin B2 gene expression would increase during sciatic nerve regeneration. In situ hybridization was used to examine B2 laminin gene expression in L4 and L5 DRGs 28 days after creating and bridging a 10-mm sciatic nerve gap with an impermeable silicone tube. Overall there was a nearly threefold increase in DRG B2 laminin chain mRNA at this timepoint, a time when axons are known to show vigorous regrowth. Both neurons and nonneuronal cells contributed to this increase. These data suggest that an up-regulation of B2 laminin gene expression by DRG neurons and nonneuronal cells may play a role in peripheral nerve regeneration.

Involvement of an axonal reflex in IgE-mediated inflammation in mouse skin.

A neurogenic component of IgE-mediated inflammation was demonstrated in mice by footpad denervation. Footpad swelling was reduced 26% following sciatic nerve transection, but unaffected by rhizotomy or spinal nerve transection. These data provide in vivo evidence that an axonal reflex is involved in IgE-mediated inflammation and completed distal to the cell bodies of the sensory neurons located in the lumbar spinal ganglia. Furthermore, depletion of neuropeptides with capsaicin also reduced IgE-mediated swelling by 26%, indicating that unmyelinated axons are involved in the neurogenic component of IgE-mediated inflammation.

Neuronal nitric oxide synthase is induced in spinal neurons by traumatic injury.

Nitric oxide appears to mediate the immune functions of macrophages, the influence of endothelial cells on blood vessel relaxation, and also to serve as a neurotransmitter in the central and peripheral nervous system. Macrophage nitric oxide synthase is inducible with massive increases in new nitric oxide synthase protein synthesis following immune stimulation of macrophages. By contrast, endothelial nitric oxide synthase and neuronal nitric oxide synthase are thought to be constitutive with activation induced by calcium entry into cells in the absence of new protein synthesis. Developmental studies showing the transient expression of neuronal nitric oxide synthase in embryonic and early postnatal life in rodent spinal motoneurons and cerebral cortical plate neurons (Bredt and Snyder, unpublished observations) implies inducibility of neuronal nitric oxide synthase. Moreover, neuronal nitric oxide synthase expression is greatly enhanced in sensory ganglia following peripheral axotomy. Staining for NADPH diaphorase in spinal motoneurons is greatly increased following ventral root avulsion. In many parts of the Central Nervous System NADPH diaphorase staining reflects nitric oxide synthase. In the present study, we have combined in situ hybridization for neuronal nitric oxide synthase, immunohistochemical staining of neuronal nitric oxide synthase, and NADPH diaphorase staining to establish that neuronal nitric oxide synthase expression is markedly augmented in spinal motoneurons following avulsion. The generality of this effect is evident from augmented staining in nucleus dorsalis following spinal cord transection.

Differential laminin gene expression in dorsal root ganglion neurons and nonneuronal cells.

Laminin is a substrate for cell migration and process outgrowth during development and may play an important role in peripheral nerve regeneration. While most studies have emphasized the expression and production of this glycoprotein by Schwann cells, in the current study we use in situ hybridization to examine the expression of the various laminin chain genes in neuronal as well as nonneuronal cells of normal adult rat lumbar dorsal root ganglia. Hybridization using cRNA probes derived from cDNAs encoding nonhomologous regions of each laminin chain showed differential localization of the chain messages within the ganglia. B2 chain mRNA was abundant in large and small neurons, satellite cells, and Schwann cells. B1 chain message was also localized in satellite and Schwann cells, but its neuronal expression was primarily restricted to a subpopulation of smaller neurons. Message encoding merosin M chain, an A chain variant, was localized within satellite cells and Schwann cells but was absent or of very low abundance in neurons. A chain mRNA was not detectable in any of the three nervous tissue cell types. The differential expression of the four laminin chains in the DRG is discussed in relation to the functional epitopes of the molecule.

In situ fixation of the neonatal brain and spinal cord.

A delay in the autopsy can result in significant tissue autolysis, especially in the central nervous system. We have developed a rapid technique of in situ fixation that preserves central nervous system tissues until the formal autopsy can be performed. Through the lateral margin of the anterior fontanelle, Zamboni's solution is injected percutaneously into the lateral ventricles and allowed to exit via an intrathecal spinal needle. The choice of fixative allows a wide array of postmortem studies to be done.

Development of nitric oxide synthase expression in the superficial dorsal horn of the rat spinal cord.

Development of nitric oxide synthase (NOS) expression in the superficial dorsal horn of the rat spinal cord was studied using NADPH diaphorase histochemistry. At birth, no positive staining was seen in the superficial laminae of the cord. A week later, a few small positive neurons and fibers were seen in presumptive lamina II. The adult pattern of NOS expression was evident by the end of the third postnatal week.

Diabetic neuropathies.

Diabetic neuropathy is a common complication of diabetes that may be associated both with considerable morbidity (painful polyneuropathy, neuropathic ulceration) and mortality (autonomic neuropathy). The epidemiology and natural history of diabetic neuropathy is clouded with uncertainty, largely caused by confusion in the definition and measurement of this disorder. We have reviewed various clinical manifestations associated with somatic and autonomic neuropathy, and we herein discuss current views related to the management of the various abnormalities. Although unproven, the best evidence suggests that near-normal control of blood glucose in the early years after diabetes onset may help delay the development of clinically significant nerve impairment. Intensive therapy to achieve normalization of blood glucose also may lead to reversibility of early diabetic neuropathy, but again, this is unproven. Our ability to manage successfully the many different manifestations of diabetic neuropathy depends ultimately on our success in uncovering the pathogenic processes underlying this disorder. The recent resurgence of interest in the vascular hypothesis, for example, has opened up new avenues of investigation for therapeutic intervention. Paralleling our increased understanding of the pathogenesis of diabetic neuropathy, refinements must be made in our ability to measure quantitatively the different types of defects that occur in this disorder. These tests must be validated and standardized to allow comparability between studies and more meaningful interpretation of study results.

Axo-glial interactions at the dorsal root transitional zone regulate neurofilament protein synthesis in axotomized sensory neurons.

After dorsal root crush, dramatic ultrastructural differences are observed between regenerated dorsal root axonal endings that are physically blocked at a ligation neuroma and those that are allowed to form axo-glial endings among the astrocytes at the dorsal root transitional zone (DRTZ). Physically blocked axonal endings swell immensely with membranous organelles and neurofilaments (NFs) while axo-glial endings do not, suggesting that DRTZ astrocytes stop axonal growth by activating a physiological stop pathway within those endings. Since protease-dependent NF degradation at axonal endings is a part of this pathway, this study addresses the question of whether NF subunit synthesis in the dorsal root ganglion (DRG) is regulated by the pathway. Lumbar dorsal roots were crushed and, at various postinjury times, the attached DRGs were removed and pulse-labeled in vitro with 35S-methionine for subsequent analysis of protein synthesis by electrophoresis and fluorography. Within 24 hr of axotomy, there was a down-regulation of the 68 kDa (NF-L) and 145 kDa (NF-M) NF subunits. At 14 d postcrush, a time when most of the regenerating axons have reached and been stopped by DRTZ astrocytes, NF protein synthesis returned to control levels. By contrast, when the axons were prevented from reaching the DRTZ by ligating or removing segments of the roots, NF synthesis failed to return to normal levels. These data suggest that activation of the physiological stop pathway by DRTZ astrocytes regulates NF protein synthesis in the DRG.

Radial glial interaction with cerebral germinal matrix capillaries in the fetal baboon.

The occurrence of intraventricular hemorrhage (IVH) is a frequent problem in premature infants delivered between 24 (60%) and 32 weeks (80%) of gestation, a time during which the germinal matrix (GM) is prominent over the head of the caudate nucleus. Most IVH arises from the GM and it has been proposed that an important factor in the pathogenesis of IVH is a weakness of GM capillary walls due to deficient support by surrounding immature glial cells. The purpose of this study was to examine the glial-capillary interaction in the GM of fetal baboons sacrificed at 100 days (54%) gestation, a stage of GM development comparable to that during which human neonatal IVH occurs. Brains from a later gestational stage (162 days, 88%), after GM involution, were also examined. At 100 days of gestational age, the GM was prominent over the head of the caudate and contained vimentin positive, but not glial fibrillary acidic protein positive, radial glial cells which formed endfeet on capillaries in the region. Ultrastructurally, all the GM capillaries examined from this gestational time had complete, continuous endothelia marked by few pinocytotic vesicles and prominent tight junctions. The endothelial cells rested upon uninterrupted basement membranes which were contacted by clearly identifiable glial endfeet. These data show that GM capillaries have morphologies typical of CNS capillaries and suggest that the capillary immaturity within the GM is not a major contributing factor to IVH.

Axotomized frog sciatic nerve releases diffusible neurite-promoting factors.

Using the bullfrog (Rana catesbeiana) dorsal root ganglia (DRG) and its sciatic nerve (ScN) as a model system, we have previously described neuronal and non-neuronal molecular changes associated with the early regenerative response of DRG neurons to axotomy. Since diffusible molecular factors, released by axotomized ScN, might function to stimulate axon regrowth, we have assayed the ability of ScN-conditioned bath to promote in vitro neurite outgrowth from PC-12 cells. Diffusible ScN proteins were collected by incubating segments of normal or axotomized ScN in a small volume of RPMI media for 4 h (nerve bath). The nerve baths, supplemented with serum, were then added to PC-12 cell cultures to assay for the presence of neurite growth factors released by ScN. Results showed that nerve baths, collected from sham-operated or axotomized ScN, could not induce the differentiation of PC-12 into neurite-bearing cells. Therefore, in all subsequent neurite growth assay experiments, an exogenous source of nerve growth factor (NGF) (50 ng/ml) was added to the nerve baths or unconditioned media to generate and maintain PC-12 neuritic structure. We found that nerve baths, collected from previously axotomized (at least 3 days post-injury) nerve, contained diffusible factors which enhanced PC-12 neurite growth, relative to unconditioned media. No neurite growth factors were observed to be released by sham-operated ScN or 1-day post-axotomized ScN. Further experiments were conducted to identify the diffusible neurite growth factors released from axotomized ScN. We showed that the release (if any) of endogenous diffusible NGF or laminin from axotomized nerve could not have accounted for the facilitation of neurite growth. Analysis of radiolabelled ScN proteins by two-dimensional polyacrylamide gel could not have accounted for the facilitation of neurite growth. Analysis of radiolabelled ScN proteins by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) showed that the relative abundance of two diffusible proteins (M(r) approximately 35 and 70 kDa) in the nerve bath was directly correlated with the ability of the nerve bath to facilitate PC-12 neurite growth.

Is CNS trauma a prerequisite for the elongation of CNS axons into denervated peripheral nerve?

The demonstration that some central nervous system (CNS) axons can regenerate when provided with a suitable environment raises the possibility of new treatments for CNS injury. However, at present the conditions for optimal regeneration are not well understood. For example, the methods used in previous studies have entailed CNS trauma as part of the research protocol (e.g. that resulting from the implantation of peripheral nerve grafts), and so the role of neuronal or axonal injury in the regrowth observed has been difficult to establish. To determine whether such injury is necessary for the central reinnervation of denervated peripheral nerve, the L5 dorsal root has been chronically denervated in rats by freeze-thawing its dorsal root ganglion (DRG), and the root has been left attached to either traumatized or non-traumatized spinal cord. The trauma induced was quite mild, and resulted from several vertical insertions of a fine needle. Two to 4 months later, retrogradely transported horseradish peroxidase (HRP) was used to label spinal neurons which sent axons into the denervated roots. HRP-labelled neurons were found in each of the spinal cords subjected to trauma, but no labelled neurons were observed in any of the non-traumatized cords. The number of HRP-labelled neurons in individual spinal cords was positively correlated with the degree of spinal cord trauma. We conclude first that the chronic and intimate presence of a denervated PNS tissue in continuity with the spinal cord is not, in itself, a sufficient stimulus to induce its reinnervation by CNS axons. Second, we conclude that under the conditions of this experiment CNS trauma is a prerequisite for the reinnervation of denervated peripheral nervous tissue by CNS axons.