Cheryl F. Dreyfus

NGF gene expression in actively growing brain glia

Previous work suggested that brain NGF acts locally on cells adjacent to sites of synthesis, in addition to any putative actions on distant, projecting perikarya. To define the basis of local action, we used a sensitive nuclease protection assay to identify cells expressing the NGF gene in vivo and in vitro. In addition to neurons, glia from a variety of developing brain areas synthesized NGF mRNA, suggesting that CNS glia exhibit a generalized capacity to express the gene. Expression was associated with active glial growth. Stimulation of growth with serum increased NGF message 2-fold in culture. Moreover, rapidly growing, low-density glial cultures exhibited 8-fold higher levels of NGF mRNA than quiescent, confluent cultures. The optic nerve, which contains all 3 major types of glia, expressed the message in vivo during neonatal development. In contrast, expression was barely detectable in the adult optic nerve. Transection, which induces glial proliferation, elicited de novo appearance of NGF mRNA in the adult nerve. Our observations suggest that active glial growth is associated with expression of the NGF gene and raise the possibility that actively growing glia in the developing or injured brain regulate neuronal growth through the elaboration of NGF.

Expression of NGF and NGF receptor mRNAs in the developing brain: evidence for local delivery and action of NGF.

Nerve growth factor (NGF) is a well-documented target-derived trophic factor in the peripheral nervous system. Recently, proteins as well as mRNAs for both NGF and its receptor (NGF-R) have been detected in diverse areas in the central nervous system (CNS). Considerable evidence suggests that NGF also functions in the target synthesis/retrograde transport mode in the brain. For example, NGF is synthesized in the target hippocampus, as indicated by the presence of NGF message, and interacts with the receptors on terminals projecting from basal forebrain, where receptor mRNA is detectable. Spatial separation of NGF and receptor gene expression is consistent with the target mechanism of action. To ascertain whether local action may also occur in the CNS, we used sensitive nuclease protection assays to study the relationship of NGF and NGF-R expression in the developing brain. Our results indicate that in some brain areas, such as diencephalon, postnatal hippocampus, and olfactory bulb, NGF message was highly expressed, while receptor mRNA was virtually undetectable, suggesting that these areas serve as target sources of NGF for distant afferent neurons. By contrast, in other brain areas, such as cerebellum, striatum, perinatal olfactory bulb, and prenatal hippocampus, NGF and NGF-R mRNAs were coexpressed and coregulated developmentally. Consequently, local delivery and action of the trophic molecule may occur in these areas during these periods. We tentatively conclude that NGF may act through both distant and local modes in the developing CNS.

Nerve growth factor selectively increases cholinergic markers but not neuropeptides in rat basal forebrain in culture

We have previously used organotypic cultures to study mechanisms regulating phenotypic expression of neurotransmitter characters in the brain. Our previous work indicated that nerve growth factor (NGF) specifically increased the activity of choline acetyltransferase (CAT) in striatal cholinergic interneurons. In the present study we examined the effect of NGF on neurons of fetal rat basal forebrain-medial septal area (BF-MS) maintained in organotypic culture. Treatment with 200 biological units/ml of NGF resulted in a 3- to 6-fold increase in the specific activity of CAT. This effect was specifically blocked by anti-NGF antiserum, whereas treatment with antiserum alone did not alter the cholinergic enzyme. NGF also elicited a marked increase in CAT staining intensity, using a monoclonal antibody directed against the enzyme. Further, the number of CAT-positive neurons appeared to increase in the NGF-treated cultures. Exposure to NGF also increased the activity of another cholinergic marker, the catabolic enzyme, acetylcholinesterase. The effect of NGF appeared to be highly selective, since substance P and somatostatin levels were unchanged by NGF treatment.

Effects of nerve growth factor on cholinergic brain neurons

Nerve growth factor (NGF) is a fully characterized molecule, well known for its actions in the differentiation and maintenance of peripheral neurons. However, recent studies suggest that its actions are not limited to the periphery, but may extend to the CNS. In particular, this trophic agent appears to affect development and survival of a variety of brain cell populations. Noteworthy are its actions on cholinergic neurons that degenerate in Alzheimers disease and Huntingtons chorea. However, studies of NGF receptor sites suggest that effects of NGF may also extend to non-cholinergic cell groups. Cheryl Dreyfus summarizes these data and points to future work necessary to define further the underlying mechanisms of action and to examine the function of NGF on diverse brain populations.

Uptake of serotonin by intrinsic neurons of the myenteric plexus grown in organotypic tissue culture.

The myenteric plexus contains axons, not found elsewhere in the peripheral nervous system, which are distinguished by a specific, high affinity transport system for serotinin (5-HT). This study was undertaken to determine the location of the cell bodies of origin of these axons. Vagotomy decreased uptake of [3H]5-HT and tritiated norepinephrine ([3H]NE) by the myenteric plexus. However, while examination by histofluorescence revealed the presence of descending vagal adrenergic fibers, no evidence was found for the presence or accumulation of 5-HT above a vagal ligature. Vagus nerves thus contain adrenergic but not serotonergic axons. The gut was also denervated of all extrinsic axons by growth of intestinal explants in organotypic tissue culture for 3 weeks. Uptake of [3H]5-HT persisted while uptake of [3H]NE was lost. Light and quantitative electron microscopic radioautography revealed that, as in intact gut, the elements of the cultures responsible for uptake of [3H]5-HT were axons distinguished by varicosities containing large dense cored vesicles. In conclusion, these experiments establish that the mammalian gut contains intrinsic neurons which selectively take up 5-HT. The capacity of these neurons for 5-HT uptake may be influenced by the vagus nerves.

Local support cells promote survival of substantia nigra dopaminergic neurons in culture

Recent studies suggest that brain neurons require extracellular signals for continued survival during maturity as well as development. However, factors underlying the survival of specific populations of central neurons remain to be defined. To examine the regulation of neuronal survival, we have studied the substantia nigra (SN) dopaminergic (DA) system, in dissociated cell culture. DA neuron number was monitored immunocytochemically with antibody to tyrosine hydroxylase (TH), the DA biosynthetic enzyme. Initially, mixed cultures were grown at low, medium, and high densities in serum-containing media. After 7 days, the number of neuron-specific enolase (NSE)-positive cells, a measure of total neuron number, was proportional to cell plating density. In contrast, high density culture elicited a marked, disproportionate increase in TH-immunopositive cells, suggesting that high density conditions selectively enhanced the DA subpopulation. To define the role of cellular interactions in the selective increase in DA cells, virtually pure neuron cultures were compared to support cell-neuron cocultures, in fully defined medium. In support cell-neuron cocultures, SN support cells evoked a four-fold increase in TH cells, while NSE number did not differ from controls. Moreover, local support cells elicited a greater increase in TH cell number than support cells derived from other brain regions. To determine whether increased TH cell number reflected enhanced survival, or possibly expression of TH by new populations, we monitored the time course of this effect. TH cell number remained constant after 3 days in cocultures, while declining fourfold in controls. In parallel studies, support cells were added to SN dissociates at zero time or after 3 days.(ABSTRACT TRUNCATED AT 250 WORDS)

GABAergic and cholinergic neurons exhibit high-affinity nerve growth factor binding in rat basal forebrain

We have used dissociated, rat basal forebrain cultures to identify specific cell types that are potentially responsive to nerve growth factor (NGF). Expression of high-affinity NGF binding sites was examined. A subpopulation of cells containing choline acetyltransferase (CAT), the acetylcholine-synthesizing enzyme, exhibited high-affinity binding, employing combined immunocytochemistry and 125I-NGF radioautography. Unexpectedly, a gamma-aminobutyric acid (GABA)-containing cell group also expressed high-affinity binding. These cells that exhibit high-affinity binding appear to be neurons since they stain positively with the neuron marker, neuron-specific enolase, and negatively with the nonneuron marker, glial fibrillary acidic protein. Our observations suggest that NGF may regulate multiple brain systems and functions that have yet to be explored. Conversely, only subsets of cholinergic or GABA neurons expressed high-affinity binding, suggesting that these transmitter populations are composed of differentially responsive subpopulations.

Expression of high- and low-affinity nerve growth factor receptors by Purkinje cells in the developing rat cerebellum

Mounting evidence indicates that nerve growth factor plays a role in the development and function of the basal forebrain-cerebral cortical system. In addition, recent studies indicate that nerve growth factor receptor messenger RNA is transiently detectable in whole cerebellum in the neonatal rat. We now report that cerebellar Purkinje cells express high-affinity and low-affinity nerve growth factor receptor sites, identified by 125I-NGF binding, in the postnatal Day 10 rat in vivo. The expression correlates with cerebellar development, suggesting that nerve growth factor may regulate the normal ontogeny of cerebellar Purkinje cells.

Nigral type I astrocytes release a soluble factor that increases dopaminergic neuron survival through mechanisms distinct from basic fibroblast growth factor

Our studies have been directed to the identification of local, naturally-occurring molecules that support substantia nigra (SN) dopaminergic (DA) neuron survival. We have previously demonstrated that local Type I astrocytes selectively increase the dopaminergic population [30,31]. However, the mechanism of action remains to be defined. To determine whether survival is elicited through diffusible agents, Type I astrocyte conditioned medium (CM) was tested on SN dissociates. After 7 days of exposure to CM, DA neuronal integrity was monitored immunocytochemically with antibody to tyrosine hydroxylase (TH), the DA biosynthetic enzyme, or by TH catalytic assay. CM increased TH+ cell number greater than 2-fold, suggesting that a soluble factor(s) promoted neuron survival. Neurons cultured in serum free medium (SFM) are known to contain few, but detectable numbers of glia [34]. To examine whether CM affected neurons directly, or indirectly through glia, glial populations were stained with antibody against the glial marker, glial fibrillary acidic protein (GFAP). We employed several approaches to define the potential role of glia. Initially, CM was compared to basic fibroblast growth factor (bFGF), a glial mitogen that reportedly enhances nigral DA neuron survival. bFGF enhanced TH activity in our system, as well, but the effect was blocked by the mitotic inhibitor 5-fluorodeoxyuridine (FDUR), which kills dividing glia. In parallel studies CM increased enzyme activity and TH cell number in cultures exhibiting GFAP+ cells. To define the role of these glial cells in the CM effect, we completely eliminated astrocytes in CM-treated cultures employing alpha-aminoadipic acid (AA; 10-30 microM), a specific gliotoxin. At a concentration of AA that eliminated detectable GFAP+ cells, CM continued to elicit a significant increase in TH cell number. These data suggest that, in contrast to effects of bFGF, the DA neurotrophic activity in CM acts directly on nigral neurons to enhance survival.

Regeneration of adrenergic axons in rat sciatic nerve: Effect of a conditioning lesion

Abstract The rat sciatic nerve was crushed at the level of the mid-thigh (testing lesion), and the rate of outgrowth of the regenerating adrenergic axons was examined in the nerve. Two methods were used: (1) counts of axons containing norepinephrine demonstrable by histofluorescence; (2) measurement of the specific uptake of [3H] d,l -norepinephrine in vitro. In order to determine whether a prior nerve lesion had any effect on the outgrowth rate, all of the animals were subjected, two weeks prior to the testing lesion, to either a transection of the tibial branch of the sciatic nerve at the level of the ankle (conditioning lesion) or a sham operation. Counts of fluorescent axons made 3 days after the testing lesion showed that at 1.40 mm proximal to the lesion, the numbers of axons in both the conditioning-lesion group and the sham-operated group were not significantly different from non-lesioned animals. At a level 0.35 mm proximal to the testing lesion, counts in both groups were about 75% greater than at 1.40 mm, indicating that axonal sprouting had occurred. At 1.40 mm distal to the testing lesion, the numbers of axons had decreased by 9% in the sham-operated group, and by 24% in the group that had been subjected to a conditioning lesion (P The specific uptake of [3H] d,l -norepinephrine was measured 3 and 7 days after the testing lesion. The axonal component of uptake showed an approximately exponential decay with distance from the testing lesion; the leading edge of axonal uptake advanced at a rate of 3.9 ± 0.5 mm/day (S.E.) in the sham-operated group compared to 1.8 ± 0.6 mm/day (S.E.) in the conditioning-lesion group (P