Search

Publisher Summary This chapter discusses how ganglioside and growth factor interactions promote mesencephalic cell survival from the excitotoxic actions of 2,4,5-trihydroxyphenylalanine. The observation that Parkinson's disease (PD) is associated with a marked reduction of dopamine content in the corpus striatum forms the therapeutic rationale for the utilization of drugs, such as levodopa (L-dopa), to affect an increase in striatal dopamine content. However, the chronic treatment of PD patients with L-dopa can also elicit negative side effects such as the on-off phenomena as well as an increase in the levels of potentially toxic oxidation metabolites of L-dopa and dopamine. In a study, the 6-hydroxylated derivative of dopa (2,4,5-trihydroxyphenylalanine or TOPA) has been shown to act as a glutamatergic agonist selective for non-N-methyl- d -aspartate (NMDA) receptors eliciting both electrophysiological and neurotoxic effects on central nervous system (CNS) neurons typical of excitatory amino acids (EAA). The results with GM1 on TOPA-treated mesencephallic cultures, together with other indications for its effective neuroplastic and/or neuroprotective actions in CNS models of dopaminergic damage, indicate that GM1 may be a possible candidate to apply in animal models exposed to acute or chronic TOPA toxicity.

Glutamate neurotoxicity is correlated with an increase of cytosolic free Ca2+. In some cell systems, activation of Ca2+ dependent endonucleases or formation of free radicals can damage DNA and activate the chromatin bound enzyme poly(ADP-ribose) polymerase (pADPRP). We have investigated whether pADPRP may be involved in glutamate neurotoxicity in vitro. Cerebellar granule cells at 12 days in culture when treated with a toxic dose of glutamate (100 μM) showed a rapid and transient increase of poly ADP-ribose immunoreactivity. Cellular immunostaining was heterogeneous and returned to control levels after washout of glutamate. In the same cell preparations glutamate elicited a marked increase in enzyme protein immunoreactivity which persisted at later times. Non-toxic doses of glutamate did not affect immunostaining. In another set of experiments, pADPRP mRNA was increased 30 min after glutamate. In order to investigate the role of pADPRP in glutamate-mediated neurotoxicity, structurally different inhibitors of pADPRP (3-aminobenzamide, benzamide, 3-aminophthalhydrazide) and their inactive analogues (benzoic acid and phthalimide) were tested in this model. Addition of the inhibitors to cultures 60 min before and during the 30 min of glutamate treatment prevented neuronal death by 60–100%, assessed 24 hr later. Glutamate-induced Ca2+ influx was not affected. Inactive analogues failed to afford neuroprotection. These data indicate that not only is pADPRP activated by the early, possibly Ca2+ -mediated mechanisms initiated by glutamate, but that it might also actively contribute to the subsequent neuronal death. © 1994 Wiley-Liss, Inc.

Rat brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3) were engineered for expression in a baculovirus-infected Spodoptera frugiperda insect cell system. The BDNF and NT-3 from the culture supernatants were purified by ion-exchange and reverse-phase chromatography to apparent homogeneity. The purification procedure yielded approximately 2 mg of pure rat BDNF or NT-3 per liter of culture supernatant. A single N-terminus only was found for either secreted molecule and was analogous to that predicted from the corresponding cDNA sequence. The recombinant neurotrophins obtained were also homogeneous with regard to molecular weight and amino acid sequence. In their native conformation, the insect cell-produced rat BDNF and NT-3 molecules were homodimers consisting of 119 amino acid polypeptide chains. Thus, although the genes transfected into the S. frugiperda cells coded for proBDNF or proNT-3, the BDNF and NT-3 recovered after purification were > 95% fully processed, mature protein. Mature recombinant rat BDNF and NT-3 were found not to be significantly glycosylated. Pure, recombinant rat BDNF and NT-3 promoted the survival of embryonic dorsal root ganglion neurons in the low picomolar range. Because recombinant rat BDNF and NT-3 can be obtained in large quantities, purified to near homogeneity, and are identical in amino acid sequence to the corresponding human proteins, they are suitable for evaluation in animal models.

Publisher Summary A large number of studies have been reported that exogenously concluded that gangliosides can have positive effects on neurite outgrowth. The molecular mechanism underlying these positive effects, however, remains unclear and it has often been difficult to determine whether gangliosides mimic the full range of trophic effects of neurotrophic factors such as nerve growth factor (NGF) or whether they have more limited effects. In dorsal root ganglion neurons—that are NGF responsive—GM1 could enhance some NGF effects but that it could not substitute for NGF by directly stimulating neuronal survival. It therefore seems likely that gangliosides act by modulating signals that lead to neurite outgrowth and do not by themselves possess full neurotrophic activity. Moreover, it is equally likely that GMl modulates integrin or cell adhesion molecules (CAM) signals rather than NGF signals per se . To analyze the possible interactions of gangliosides with CAM signaling pathways, the chapter presents a series of co-cultures of PC12 cells with monolayers of 3T3 fibroblasts or the same monolayer cells expressing transfected NCAM or N-cadherin. It is shown that a direct incorporation of GM1 into the cell membrane of PC12 cells—in a trypsin insensitive manner—is associated with a dramatic enhancement of NCAM and N-cadherin-dependent neurite outgrowth.

The semisynthetic glycosphingolipid derivative II 3 Neu5-AcGgOse 4 -2-d-erythro-1,3-dihydroxy-2-chloro-acetamide-4- trans -octadacene (LIGA20) attenuated injury induced by the excitotoxic L -dopa metabolite 2,4,5-trihydroxyphenylalanine (TOPA) in cultures of rat cerebellar granule cells when presented simultaneously with TOPA (EC 50 : 9 μM LIGA20). The natural glycosphingolipid ganglioside GM1 up to 200 μM was not neuroprotective as cotreatment, although pretreatment of cells for 2 h was efficacious. This greater potency and speed of LIGA20 action extended to limiting TOPA-induced death of cultured mesencephalic dopaminergic neurons. These data suggest that LIGA20 may have therapeutic potential for the treatment of disorders associated with excitotoxic processes.

Glucocorticoid hormones are important regulators of brain development and ageing, and can impair the capacity of hippocampal neurones to survive various neurological insults. Here we show that dexamethasone, a synthetic glucocorticoid, prevents activity-dependent increases of brain-derived neurotrophic factor (BDNF) mRNA in cultures of rat hippocampal neurones. In situ hybridization was used to evaluate the levels of BDNF mRNA. Up-regulation of BDNF mRNA triggered by depolarization with high potassium, or exposure to the glutamate receptor agonist kainic acid, resulted both from higher levels of expression in neurones and from new recruitment of cells. These data suggest that the known ability of glucocorticoids to exacerbate neuronal injury following ischaemia and other metabolic insults could be due to antagonism of regulatory mechanisms governing neurotrophin levels in the brain.

Brain-derived neurotrophic factor (BDNF) supports the survival of sensory neurons as well as retinal ganglion cells, basal forebrain cholinergic neurons, and mesencephalic dopaminergic neurons in vitro. Here we examined the ability of BDNF to confer protection on cultured dopaminergic neurons against the neurotoxic effects of 6-hydroxyDOPA (TOPA or 2,4,5-trihydroxyphenylalanine), a metabolite of the dopamine pathway suggested to participate in the pathology of Parkinson's disease. Cells prepared from embryonic day 14-15 rat mesencephalon were maintained with 10-50 ng/ml BDNF for 7 days prior to addition of TOPA (10-30 microM) for 24 hr. In BDNF-treated cultures, the extensive loss (90%) of tyrosine hydroxylase immunopositive cells was virtually (10%) eliminated, while the equally drastic loss (90%) of the overall cell population was limited to only a 25-30% recovery. Furthermore, the monosialoganglioside GM1 (1-10 microM), although inactive alone, acted synergistically with subthreshold amounts of BDNF to rescue tyrosine hydroxylase-positive cells against TOPA neurotoxicity. These results add impetus to exploring the therapeutic potential of gangliosides and BDNF in Parkinson's disease.

Ciliary neurotrophic factor (CNTF) is a multifunctional protein which not only promotes neuronal survival in vitro and in vivo but also controls cell division of neuronal precursors, transmitter differentiation, and glial cell differentiation. Recent studies have indicated that neurotrophic factors can alter hippocampal neuronal threshold to excitotoxin sensitivity. To examine such a role for CNTF, cultures of rat embryonic hippocampal neurons were maintained with recombinant human CNTF for different times, prior to exposure to a toxic dose of glutamate at 5 days in vitro for a further 24 hr. The cytotoxic action of 200 microM glutamate (approximately 40% of pyramidal neurons remaining after 24 hr) was reduced in a concentration-dependent manner in cultures receiving a prior exposure to CNTF within the first 3 days of cell plating: 30 ng/ml CNTF permitted about 75% of the initial number of pyramidal neurons to survive. Presentation of CNTF less than 48 hr before glutamate challenge was ineffective at up to 100 ng/ml. When pyramidal neurons were cultured with a subthreshold concentration (2 ng/ml) of CNTF together with 10 microM of the monosialoganglioside GM1 (or its inner ester form) in the same paradigm, the resulting neuronal survival was similar to that seen with 30 ng/ml CNTF in the face of a glutamate challenge. Such low doses of either CNTF or ganglioside alone were ineffective. The ability of trophic factors to influence the threshold of neuronal sensitivity to excitatory amino acid injury suggests that these proteins could play an important role in the reparative capacity of acutely traumatized central neurons and in neurodegenerative diseases linked to an excitotoxic mechanism.(ABSTRACT TRUNCATED AT 250 WORDS)

Several derivatives of kynurenic and thiokynurenic acids were synthesized and tested for their ability to protect primary cultures of cerebellar granule cells against excitotoxic damage, and to affect the binding of [3H]glycine ([3H]Gly), [3H]alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ([3H]AMPA), [3H]3-(2-carboxypiperazine-4-yl-)propyl-1-phosphonic acid ([3H]CPP), [3H]kainic acid and [3H]N-[1-(2-thienyl)cyclohexyl]-3,4-piperidine ([3H]TCP) to rat cortical membranes. Kynurenic and thiokynurenic acid derivatives with one or two halogens in position 5 or 7 were selective glycine antagonists, failing to affect N-methyl-D-aspartate (NMDA), kainate or AMPA sites at micromolar concentrations. 7-Cl-kynurenic, 7-Cl-thiokynurenic, 5,7-diCl-kynurenic and 5,7-diCl-thiokynurenic acids had similar IC50s for displacing [3H]Gly from its strychnine-insensitive site and for reducing the stimulated (0.5 microM NMDA and 1 microM glycine) [3H]TCP binding to cortical membranes. However, 7-Cl-thiokynurenic acid was particularly potent to prevent excitotoxic neuronal death in cultured cerebellar granule cells. This action may be ascribed to inhibition of lipid peroxidation, a property which was demonstrated for the 5- or 7-Cl derivatives of thiokynurenic acid. Furthermore, 7-Cl-thiokynurenic acid reduced excitotoxic damage caused by the injection of quinolinic acid in the rat striatum. Thus, 7-Cl-thiokynurenic acid appears to be a new compound with interesting antiexcitotoxic properties both in vitro and in vivo.

Glial cells execute essential functions in central nervous system (CNS) development and are also believed to play important roles during gliosis in response to trauma or disease. These developmental and pathological states have also been associated with elevated expression of opioid genes. Because levels of the cytokine interleukin-1 beta (IL-1 beta) increase following CNS lesions, we examined the possible influence of IL-1 beta on the expression of opioid genes in astrocytes cultured from rat cortex. Proenkephalin mRNA expression was stimulated by IL-1 beta in a time- and concentration-dependent manner, being maximal with 5 U/ml IL-1 beta at 4 h. Although the beta-adrenergic agonist isoproterenol was also active, interferon, glutamate, and carbachol were not. Unlike isoproterenol, the actions of IL-1 beta were not associated with a cyclic adenosine monophosphate (AMP)-dependent pathway. Interleukin-1 beta also regulated a proenkephalin-chloramphenicol acetyltransferase fusion gene transiently transfected into astrocytes, with a dose-response similar to that active in proenkephalin mRNA. These effects of IL-1 beta were region-specific, not being observed with either cerebellar or hippocampal astrocytes; however, isoproterenol was active in the latter cell populations. Proenkephalin mRNA in cortical astrocytes was stimulated following a temperature stress. These results suggest that enhanced proenkephalin gene expression in astrocytes by IL-1 beta may be important in neuroimmune interactions and in trauma-induced CNS injury or stress.

Ciliary neurotrophic factor (CNTF) is a survival factor for avian ciliary ganglion neurons and a variety of other neuronal cell types in vitro. We report here the cloning of the entire genomic sequence encoding human CNTF and its primary structure. Biologically active CNTF has been expressed in Chinese hamster ovary cells from a human genomic DNA clone. Human CNTF has no significant sequence similarity to any previously reported protein, although approximately 84% similarity exists compared with rat and rabbit CNTF. The lack of both an N-terminal signal sequence and consensus sequences for glycosylation or hydrophobic regions, and the fact that active CNTF is expressed but not released into the culture medium of transfected cells, argue in favour of human CNTF as a cytosolic protein. These data provide a basis for understanding the role of CNTF in nervous system physiology and pathology.

The B subunit of cholera toxin, a protein which binds specifically to membrane ganglioside GM1, is known to affect cell growth and differentiation. To investigate the mechanism of these cellular responses at the nuclear level, we used the induction of c-fos in astrocytes and 3T3 fibroblasts as a model. Northern blot analysis showed that treatment with B subunit provokes a rapid and transient expression of c-fos mRNA, independent of a measurable increase in cyclic AMP. The B subunit signal, which is mediated by Ca2+, was compared to cholera toxin and other agents which increase intracellular cyclic AMP levels. In transient transfection assays of astrocytes and fibroblasts, functional analysis of c-fos promoter deletions was used to identify the elements involved in transcriptional activation by B subunit. In astrocytes, the DNA region including the serum response element and the cyclic AMP response element (CRE) are equally required, whereas 3T3 cells require only the CRE for maximal induction. A synergistic effect of signal transduction was mediated by calcium and cyclic AMP on the CRE, being positive in 3T3 cells and negative in astrocytes. Diverse regulatory elements may be thus involved in responses of different cell types to the same extracellular signal. Furthermore, a single regulatory element (CRE) can integrate both calcium and cyclic AMP signals in the control of gene expression.

Rat hippocampal pyramidal neurons in culture, exposed 30 min to Mg 2+ -free, glycine-supplemented medium undergo a selective (about 35%) degeneration over the next 24 h. This neuronal injury appeared to result from excitatory synaptic transmission and subsequent activation of N-methyl- d -aspartate (NMDA) receptors, as cell death could be blocked by tetrodotoxin and NMDA, but not non-NMDA, receptor antagonists. Ganglioside GM1, which has recently been described to protect against excitotoxin-induced damage, also prevented the death of hippocampal neurons associated with the above phenomenon. Gangliosides may be a novel therapeutic tool for brain injury associated with epileptic-like activity.

The expression of the epitope recognized by the tumor-associated antigen monoclonal antibody 217c increased on cultured rat Schwann cells with time. The same phenomenon has previously been reported for the rat nerve growth factor (NGF) receptor by using monoclonal antibody 192-IgG. The distribution of 217c antibody immunoreactivity closely paralleled that observed for NGF receptors on NGF-primed pheochromocytoma (PC12) cells in culture and a number of central neurons in vivo. Immunoprecipitation of affinity-labeled NGF receptors was used to examine whether these two antibodies shared common or unique antigens. Both the quantitative data and the electrophoretic mobility of the immunoprecipitated 125I-NGF/receptor complex indicate that the antigen recognized by the 217c mAb monoclonal antibody is, in fact, the NGF receptor. Furthermore, binding studies indicated that 192-IgG and 217c recognize different epitopes on the same molecule. The characterization of the 217c antibody should provide a valuable new tool in the study of NGF receptors.

The effects of glutamate on intracellular free Ca2+, [Ca2+]i, and neurotoxicity were compared in cerebellar granule neurons in vitro. [Ca2+]i was measured with fura-2 and digital fluorescence imaging microscopy; neurotoxicity was monitored using a vital dye and colorimetric analysis. Glutamate produced dose-dependent increases in [Ca2+]i, which tended to be transient for glutamate concentrations in a range of 0.01-0.5 microM and sustained for higher levels of glutamate. The ED50 for the [Ca2+]i response to glutamate was 6 microM. The LD50 for glutamate-induced neurotoxicity was similar, i.e., 10 microM. The effect of glutamate on [Ca2+]i was greatly diminished when external Ca2+ was removed and blocked by Mg2+ or N-methyl-D-aspartate (NMDA)-type receptor antagonists. The latter conditions as well as preloading granule neurons with the intracellular Ca2+ chelator quin2 largely prevented glutamate cytotoxicity. The neurotoxic effect of glutamate required incubations with the stimulus for 10-20 min at 25 degrees C. Withdrawal of glutamate after this period was accompanied by a prolonged alteration in [Ca2+]i. Pretreatment of the cells with the ganglioside GM1 reduced this late increase in [Ca2+]i as well as the neurotoxic effects of glutamate. This indicates that glutamate-induced neurotoxicity results from a composite of diverse temporal alterations in Ca2+ homeostasis and that blunting any of these components reduces excitotoxicity.

Today we understand the brain as a dynamic, not static, organ. Central nervous system (CNS) neurons are endowed with the capacity to react to chemical signals presented from their microenvironment with morpho-functional modifications, a process termed plasticity. This phenomenon has provided the foundation for studies directed at elucidating the pathophysiological correlates of neuronal life and death, with the ultimate objective of developing strategies to improve neurological outcome following various types of CNS insults, in particular cerebrovascular insufficiency (stroke), head and spinal trauma, and neurodegenerative diseases.

Nerve growth factor (NGF) is a protein which plays a critical role in the development and survival of not only peripheral neurons, but possibly also cholinergic brain neurons. The present study describes a procedure for large scale isolation of human NGF of placental origin, and its immunological characterization. A protein species of approximately 26 kDa was obtained, which cross-reacted with antibodies to mouse NGF. Polyclonal and monoclonal anti-mouse NGF antibodies appeared to recognize different bands within this human NGF preparation. Although these polyclonal antibodies recognized both the dimeric and monomeric forms of mouse NGF, the monoclonal antibody recognized only a band corresponding to the dimeric form of mouse NGF.

Cultures of chicken day 8 embryo retinal cells, essentially free of contaminating non-neuronal elements, were used to examine the neurotoxicity of various excitatory amino acid transmitter receptor agonists. At 7 days in vitro, N-methyl-D-aspartate (NMDA), following 24 hr exposure to 0.1-1.0 mM, destroyed 60-70% of the multipolar neurons, but apparently spared photoreceptors. The cytotoxic effect of NMDA was prevented by extracellular Mg2+ or phencyclidine, suggesting a role for the NMDA ion channel; competitive NMDA antagonists were also neuroprotective. The mixed excitatory amino acid receptor agonist glutamate (0.1-1.0 mM) was also neurotoxic (approximately 70% loss of multipolar neurons) and strongly blocked by NMDA (but weakly by non-NMDA) antagonists and Mg2+, indicating a major action at NMDA receptors. As with NMDA, glutamate did not appear to affect photoreceptors. The neurotoxic action of kainate against multipolar retinal neurons, as reported by others, was confirmed here. Kainate neuronal injury was sensitive to the quinoxalinedione non-NMDA antagonists 6,7-dinitroquinoxaline-2,3-dione (DNQX) and 6-cyanoquinoxaline-2,3-dione (CNQX), but not to Mg2+ or phencyclidine. Ibotenate and quisqualate, even at millimolar concentrations, were not neurotoxic. The monosialoganglioside GM1 was also effective in reducing NMDA and non-NMDA agonist neurotoxicity to retinal neurons. Maximal ganglioside benefit required 1-2 hr of pretreatment with 100-200 microM GM1. The percentage of multipolar neurons remaining after the neurotoxin insult approximately doubled with GM1 treatment. Gangliosides may thus have a therapeutic potential in excitatory amino acid-initiated neuropathologies.

Changes in [Ca2+]i are essential in modulating a variety of cellular functions. In no other cell type does the regulation of [Ca2+]i reach the level of sophistication observed in cells of neuronal origin. Because of its physicochemical characteristics, the fluorescent Ca2+ indicator Fura-2 has become extremely popular among neuroscientists. The use of this probe, however, has generated a number of problems, in particular, extracytosolic trapping and leakage from intact cells. In the first part of this contribution we briefly discuss the practical application of Fura-2 to the study of [Ca2+]i in primary cultures of neurons and astrocytes. In the second part, we review some recent data (mainly from our laboratories) obtained in neurons and neuroendocrine cells, concerning the regulation of different types of Ca2+ channels and the role and mechanism of intracellular Ca2+ mobilization. The experimental evidence supporting the existence of a previously unrecognised organelle, the calciosome, that we hypothesize represents the functional equivalent in non-muscle cells of sarcoplasmic reticulum, will also briefly be discussed.

A variety of neurotransmitters are believed to elicit effects through receptor-stimulated inositol phospholipid metabolism. It appears that most major types of retinal neurons receive a direct glutamatergic input. The aim of the present studies was to characterize excitatory amino acid (EAA) receptor-mediated breakdown of inositol phospholipids and changes in Ca2+ homeostasis in primary avian retinal cell cultures. Cell monolayers, prepared from 8-day-old chick embryo neural retina, were labelled with [3H]inositol for 48 h, and used after 7 days in vitro. Kainic acid stimulated the accumulation of inositol phosphates in a time- and dose-dependent manner (ED50 = 30 microM). The EAA receptor agonists glutamate, N-methyl-D-aspartate (NMDA), ibotenate and quisqualate were all active, with the rank order: glutamate greater than kainate greater than NMDA much greater than ibotenate approximately quisqualate. External Ca2+ was required for these effects. Agonist actions were inhibited by type-specific antagonists, and also Mg2+ in the case of glutamate and NMDA. Glutamate, NMDA and kainate also elevated cytosolic free Ca2+ in individual retinal cells loaded with the Ca2(+)-sensitive dye Fura-2, as assessed by digital fluorescence ratio imaging microscopy. The agonist-induced increases in [Ca2+]i were largely dependent on extracellular Ca2+, independent of membrane depolarization and were blocked by Mg2+ for glutamate and NMDA. These results demonstrate that vertebrate retinal cells possess EAA receptors coupled to intracellular signal transduction pathways.

The phenomenon of neuronal plasticity reflects the ability of nerve cells to modify their behaviors under the influence of extrinsic agents. As with any living system, neural tissue represents a dynamic organization, whose elements are continuously changing due to interactions with one another and with their extraneural environment. Neurons are thus subject to influences from the extracellular fluid and matrix, and the other cells with which they are in direct contact. This array of extrinsic influences impinging on the neuron constitutes, in broad terms, what can be called the microenvironment of these cells. Because of their many origins and functions, agents affecting neuronal behaviors represent a crucial and diverse element in determining how nerve cells will respond to cues from the microenvironment. As we shall see later on, these cues can carry either positive or negative signals for the neuron. Our ability to alter the response(s) of neuronal cells to such extrinsic agents will constitute a powerful tool for modulating the neuroplastic behaviors of the former — an important consideration for effecting regeneration and repair processes in the brain. Such is the subject of the present chapter.

Publisher Summary This chapter discusses procedures for preparing and culturing a variety of neural cell types. It discusses advantages and limitations as well as on some selected applications of these methods to neurobiological problems. Neural cell cultures have proved to be a basic and indispensable tool for identifying and purifying substances that influence the differentiation and maturation of the nervous system, for defining their cellular and molecular consequences, and for investigating the mechanisms eliciting these consequences. In vitro neural systems also serve to study interdependences between cell types, for example, neuronal and glial cells, associations believed to occur in vivo . The study shows that the plasticity of nerve cells is more evident in the cultures of neural tissue or cells in in vitro . Neuronal cultures provide the neuroscientist with a powerful tool for identifying and analyzing extrinsic agents affecting neuronal survival and development, and for investigating the underlying cellular and molecular events. The several types of neural cultures, such as clonal cell lines, explants, and monolayers can provide different and complementary views of neural behaviors. Monolayer neuronal cultures permit the widest opportunities for the quantitative analysis of both positive and negative microenvironmentally applied influences as well as parameters at the single-cell level.

The survival and growth of neurons within the CNS is thought to be regulated by the availability of appropriate neuronotrophic factors. Compelling evidence for this idea derives from the recent findings that Nerve Growth Factor (NGF), the classical trophic factor for neural crest-derived peripheral neurons (Levi-Montalcini, 1966; Greene and Shooter, 1980; Thoenen and Barde, 1980) is also present within the CNS, where it has been proposed to act as a neuronotrophic factor for developing and lesioned adult cholinergic neurons in the basal forebrain (Gnahn et al., 1983; Hefti, 1986; Korsching, 1986; Kromer, 1987; Williams et al., 1986). These observations have encouraged the view that neuronotrophic factors in general may play an important role, not only during development but also in the adult, in particular following brain injury. At present NGF appears to address specific and limited populations of neurons. It is thus reasonable to believe that additional trophic factors exist for yet other types of neurons.

Excitatory dicarboxylic amino acid neurotransmitters, particularly glutamate, have been implicated in mediating neuronal cell injury in brain ischemia-anoxia, epilepsy, and stroke. Glutamate neurotoxicity has been demonstrated in several in vitro models, as well as its prevention by a variety of agents, including several sialic acid-containing glycosphingolipid species, gangliosides. We have now examined ganglioside effects in anoxic exposed cultures of granule cells from Postnatal Day 8 rat cerebellum. Cells between 10 and 12 days in vitro were placed into an anoxic atmosphere or subjected to a chemical model of anoxia by a pulse exposure to rotenone. Widespread neuronal degeneration of neuronal cell bodies and their associated neurite network was seen the following day. These effects on cell vitality at the morphological level were quantitatively confirmed by measuring the photometric reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide to a blue formazan product. This neuronal injury was abolished by the specific N -methyl- d -aspartate receptor noncompetitive antagonists Mg 2+ , phencyclidine and MK-801, suggesting that this subtype of glutamate receptor is involved in the pathogenesis of anoxic granule cell injury. Pretreatment for 30 to 60 min or more or concurrent treatment with ganglioside GM1 largely prevented the ensuing neuronal death (ED 50 = 25 μ M ), even 4 days later. Degeneration induced by exogenous glutamate was equally reduced. Asialo GM1 (lacking sialic acid) was ineffective. These results are consistent with the observed beneficial effects of the gangliosides in ischemic brain injury models in vivo .

— The kinetic parameters of monoamine oxidase (MAO; E.C 1.4.3.4) and catechol-O-methyl-transferase (COMT; EC 2.1.1.6) were evaluated in extracts of adrenergic and non-adrenergic mouse neuroblastoma cells and in rat glioma cells. Using the naturally-occurring substrates tyramine, tryptamine, serotonin and norepinephrine, the affinity of MAO for a given substrate was independent of the presence of the catecholaminergic pathway or cell type used, with apparent Km values ranging from 8–14 μM for tryptamine to 510–580 μM for norepinephrine. The MAO activity in glioma cells was substantially greater than in either neuroblastoma clone, but Vmax values varied little with substrate among cell lines. Both the neuronal and glial COMT had a similar Km for I-norepinephrine (200μM); the corresponding Vmax values were also similar among the different cell lines, but represented only 2–10% of the maximal MAO activity. Neuroblastoma and glioma cells, when grown from early logarithmic to stationary phase, showed no significant changes in specific activity of either MAO or COMT. Growth of cells for 3 days with 1 mM-N6,O2′ -dibutyryl adenosine-3′,5′-cyclic monophosphate resulted in no marked change in either MAO or COMT activity. These results suggest that in neurons neither MAO nor COMT plays a major role in the type of transmitter inactivation that is analogous to that of acetylcholinesterase in cholinergic synapses. The occurrence of considerable MAO and acetylcholinesterase activities in glioma cells may indicate a role for these cells in neurotransmitter inactivation.

Recently we have shown that nerve growth factor (NGF) influences the coupled movements of Na+ and K+ across the membrane of chick embryo dorsal root ganglion (DRG) and other target cells. These ionic effects of NGF are consistent with a model in which NGF acts through the classical Na+,K+-ATPase pump. Direct evidence for NGF-induced alteration of this pump has been sought in the present study through two approaches. In one approach, DRG cell dissociates were incubated for 6 hours without NGF (to allow development of the ionic defect), and [3H]ouabain binding to the cells was measured before and during a delayed administration of NGF. No differences were detected in either total binding or binding time. In the second approach, intact DRG or DRG dissociates were incubated for 6 hours with or without NGF, or received NGF after 6 hours of deprivation, and Na+,K+-ATPase (ouabain-sensitive) activity was measured in the corresponding microsomal preparations. Activity levels were found to be the same in all cases, and were unchanged by addition of NGF directly to the enzyme preparations. Different concentrations of Na+, K+, or ATP affected in identical manners the enzyme preparations from NGF-treated and NGF-deprived ganglia, speaking against NGF-imposed changes in the affinities of the corresponding enzyme sites. Also unsuccessful were attempts to reveal NGF-related differences by testing ouabain-sensitive ATPase activity (1) in the presence of varying concentrations of cyclic AMP or of Ca2+, (2) after treatment with Triton X--100 or in the presence of vanadate, or (3) on addition of a 100,000g DRG extract. These negative findings are discussed in terms of the Na+,K+-ATPase hypothesis for NGF action.

Central nervous system neurons, cultured as monolayers at low density, need exogenous pyruvate for their survival. The pyruvate concentrations required by embryonic day 8 (E8) chick forebrain neurons are reduced substantially by other low molecular weight agents present in Dulbecco's modified Eagle's medium (DMEM) and in astroglia-conditioned Eagle's basal medium (EBM). To examine the nature of these pyruvate-sparing molecules, E8 chick forebrain neurons were cultured in EBM containing a concentration of pyruvate too low to support neuronal survival, unless sparing agents were supplied. This strategy permitted the identification of the pyruvate-sparing DMEM constituents, omission of which had led to neuronal loss. Pyruvate sparing required a mixture of (i) serine, (ii) pyridoxal (or riboflavin and biotin), and (iii) six essential amino acids: arginine, cystine, leucine, isoleucine, tyrosine, and tryptophan (each replaceable by corresponding amino acids, except tryptophan). All pyruvate-sparing contributors were optimally needed at or near their DMEM concentrations.

Nerve growth factor (NGF) is required for the growth and development of sensory and sympathetic neurons. Incubation of chick dorsal root ganglionic cells without NGF resulted in a decrease of active (Na+,K+-pump-mediated) K+ influx over a period of several hours. Addition of NGF to NGF-deprived cells caused 1) a return of the active K+ influx to the values occurring in cells continuously exposed to NGF, preceded by 2) a very rapid, but transient overstimulation of the Na+,K+-pump-mediated K+ influx. Restoration of normal Na+,K+-pump activity occurred at NGF concentrations of 1 biological unit/ml or greater, whereas the NGF concentration in the 1-100 biological unit/ml range affected the rapidity with which the pump restoration took place. The transient pump behavior was only observed in NGF-deprived cells and could not be elicited in NGF-supported steady-state cells or in cells having already received delayed NGF once. This transient Na+,K+-pump behavior was exclusively displayed in conjunction with a high intracellular Na+ concentration. Decreasing the external Na+ concentration below 70 mM reduced the hyperstimulation response to NGF, until at 10 mM Na+ the delayed presentation of NGF caused no overshoot at all. The effect of NGF on the Na+,K+-pump was specific for the NGF molecule and could not be mimicked by other proteins.

This study describes the survival and neurite outgrowth behaviors of cultured adrenal medullary (chromaffin) cells obtained from postnatal rats 1 day (D1) to 100 days (D100) old in response to nerve growth factor (NGF), chick eye ciliary neuronotrophic factor (CNTF), and laminin. In the absence of trophic factors the 4-day survival of cultured chromaffin cells (relative to the number of cells attached at 2 hr) increased from one-third of the cells at D1 to 40% at D8 and 90-100% at D16 and older stages. At saturating concentrations NGF increased cell survival at D8 by 90%, but failed to support all chromaffin cells present at 2 hr. In contrast, CNTF supported the survival of all cells at D8. At D1 NGF and CNTF had only a very small effect on survival during the 4-day culture period, although both factors clearly enhanced the numbers of surviving cells after 8 days. Either NGF or CNTF also elicited neurite outgrowth from rat chromaffin cells, which amounted to approximately 15-20% at D1 and D8 and subsequently decreased to about 5-8% at D30 and virtually zero at D100. At this last age both factors applied together clearly elicited neurites. Such a potentiating effect of NGF and CNTF was also seen at earlier postnatal ages. Laminin did not affect neurite growth at D30 in the absence of trophic factors, as already described for D8 rat chromaffin cells. In the presence of NGF, however, it increased neurite length and branching during a 4-day culture period and even enhanced neurite recruitment at later culture times. These data suggest that rat chromaffin cells undergo age-related changes in their responses to NGF and CNTF and that laminin modulates their neurite outgrowth behaviors in the presence of trophic factors.

Nerve Growth Factor (NGF) is known to elicit several distinct responses from its target neurons. Three different responses were exploited as bases for an assay method for NGF: (i) the traditional neurite outgrowth from 8-day chick embryo dorsal root ganglia (DRG) in modified explant culture, (ii) survival of neonatal mouse DRG neurons in monolayer cell culture, and (iii) extrusion of 22Na+ from chick embryo DRG neurons preloaded under conditions of NGF deprivation, in cell suspension. All three procedures were able to detect as little as 0.01 to 0.03 biological unit (BU) NGF/ml (about 10−11 m ). Nerve growth factor-like activity was examined in a number of human sera samples. None of the sera tested elicited a positive response with any of the three assay methods, indicating that if NGF was present, it would be at less than 0.1 to 0.2 BU/ml of undiluted serum. Human sera did not prevent genuine NGF from producing a positive response in these assays. Sera from patients with familial dysautonomia, an inherited neurologic disease suspected of having an altered concentration of circulating NGF, were also negative.

Recently we have shown that Nerve Growth Factor (NGF) influences the movement of Na+ across the membrane of chick embryo dorsal root ganglion (DRG) cells. When cell dissociates from 8-day embryonic chick DRG, equilibrated with 86Rb+ (a K+ analog) in the presence of NGF, were transferred to NGF-free medium a marked loss of intracellular K+ occurred over several hours. The time course of K+ loss was similar to the time course of Na+ accumulation which occurs in the absence of NGF. NGF-deprived, K+-depleted DRG cells reaccumulated K+ within minutes of delayed NGF presentation, just as delayed NGF administration results in the rapid extrusion of Na+ from Na+-loaded cells. Restoration of K+ competence was dependent upon NGF concentration. The occurrence of this K+ response to exogenous NGF in other ganglionic preparations correlated with traditional responses to NGF in culture and previously observed Na+ responses. Neither the development nor the expression of the ionic defect (K+ depletion, Na+ filling) during NGF deprivation required the presence of both cations in the medium. NGF-dependent restoration of intracellular K+ in NGF-deprived chick DRG cells required the presence of intracellular Na+, and NGF-dependent extrusion of Na+ required extracellular K+. Thus NGF appears to influence the coupled (active) movements of Na+ and K+ across the membrane of its target cells, possibly by means of the classical Na+, K+-ATPase pump.

This study investigates the survival properties and changes in the morphological phenotype of adrenal medullary (chromaffin and neuronal) cells cultured from embryonic chicks at different developmental ages (embryonic days E8 to E16) in response to nerve growth factor (NGF) and ciliary neuronotrophic factor (CNTF). The 4-day survival of medullary cells from all embryonic ages except E8 was about 80% of the seeded cells and was only slightly enhanced by the addition of saturating doses of CNTF (10 ng/ml). With no factors, after 4 days 10–30% of the surviving medullary cells extended neurites. NGF (100 ng/ml) and, even more, CNTF (10 ng/ml) and their combination substantially increased the proportions of neuritebearing cells (up to 70%). The effect of the factors were maximal at E10 and E12 and declined at older developmental ages. Neurite growth was virtually unaffected by NGF and CNTF at E8. These results show that in vitro survival and neurite growth of chick adrenal medullary cells in response to trophic factors is developmentally regulated.

Dorsal root ganglionic cells, when incubated in vitro in the absence of nerve growth factor (NGF), undergo a general metabolic degeneration which is preceded by loss of certain permettion properties. To determine in which ways an absence of NGF can also affect the capacity of these cells to take up an important energy source, namely glucose, experiments were carried out in which cells were incubated with or without NGF for varying times, and then presented with the factor and tested for the ability to take up3H-labeled 2-deoxy- d -glucose. As with exogenous uridine, hexose transport in DRG cells was reduced by NGF deprivation and restored by delayed NGF administration (up to 6 h). Both the initial rate and equilibrium level were affected in an NGF dose-dependent fashion. Calculation of apparentKt and Vmax in NGF-deprived and NGF-supported cells showed about two-fold differences between NGF-controlled and NGF-independent hexose transports, suggesting corresponding differences between NGF-dependent and other ganglionic cells. Restoration of hexose transport by delayed NGF administration took place within minutes of presentation of the factor. The delay before onset of restoration and the speed with which restoration was achieved have been found also to be dependent on the NGF concentration, suggesting that they reflect equilibration kinetics between NGF and its binding sites rather then the development of the response within the cells. Thus, NGF can rapidly modulate permeation properties which regulate the availability of major energy substrates for the cell. This effect of NGF is discussed in the context of current views on the mode of action of the factor.

Considerable variations in both neuritic and non-neuronal cell outgrowths of ganglionic expiant cultures are imposed when different substrata and medium supplements are used. Therefore, a neuritic response to exogenous agents such as gangliosides may be better, or exclusively revealed under selected combinations of medium and substratum. We have surveyed a number of different ganglia for their ability to display a neuritic response to ganglioside GM1 and the culture conditions under which such a display may take place. GM1 can recognizably promote neurite extension from embryonic day 8 chick dorsal root ganglia, day 11 sympathetic ganglia, and day 8 ciliary ganglia. Demonstration of such effects, however, requires an appropriate balance between promoting and inhibiting influences by other extrinsic influences (culture substrata, neuronotrophic factors, serum inhibitors) thereby allowing for improvement by the ganglioside. Different ganglia require different combinations of those influences. Ciliary ganglia, for example, will show a GM1 response on a polyornithine substratum in medium supplemented with the chick eye Ciliary Neuronotrophic Factor and 1% fetal calf serum, but not with 10% serum or no serum. The effective concentration range for GM1 is in accord with previously reported serum binding data. These, and other data are consistent with an action of GM1 in executing a neurite program, rather than an imposition of the program itself.

The survival of nerve cells, in vitro and in vivo , is controlled by extrinsic agents called neuronotrophic factors. The best known among them, nerve growth factor, has been shown to control the performance of the Na + , K + pump in the membrane of its target ganglionic neurons. In turn, the operation of the pump is essential for the survival of these neurons. Ionic control may be an important mechanism by which cell survival, growth and/or differentiation is regulated by certain extrinsic factors or hormones.

Previous studies with microcultures of astroglial (AG) cells from newborn rat cerebrum had shown an ability of gangliosides to interact with AG cells cultured under defined conditions. We have now investigated the capability of gangliosides to stimulate DNA synthesis and cell number increases in similar secondary microcultures of newborn rat cerebrum AG cells. At a concentration of 6 X 10(-5)M, GM1 ganglioside stimulated DNA synthesis and increased cell numbers, with DNA synthesis leading cell increases by 12-24 hr. The ganglioside-induced AG cell proliferative response occurred with GD1a, GD1b and GT1b, GT1b being the most potent at 10(-5)M--while asialo GM1 and sialic acid were without effect. In the standard test cultures, DNA synthesis declined very steeply after the first day, with cell numbers stabilizing at the level reached after 2 days. Ganglioside was not itself responsible for the restricted proliferative response, as serum produced the same behaviors.

Cell dissociates from embryonic chick dorsal root ganglia, incubated for 6 h with 22Na+, accumulated four to six times more radioactivity in the absence than in the presence of Nerve Growth Factor (NGF). The accumulation of radioactivity paralleled the external Na+ concentration, indicating that the cells may have been reaching equilibrium with the medium. Delayed presentation of NGF to 22Na+-loaded cells caused a rapid loss of radioactivity, even with extracellular 22Na+ still present, demonstrating that NGF caused an overall efflux of Na+ rather than an accelerated equilibration. The Na+ exclusion from 22Na+-loaded cells was dependent upon NGF concentration. Use of nutrient-rich medium, serum, and certain hormones and other proteins did not prevent the Na+ accumulation in the absence of NGF or its reversion by delayed NGF administration. Incubation of the ganglionic cells with ouabain or dinitrophenol during the 22Na+ loading period (no NGF) increased the rate, but not the magnitude, of loading. The same incubation carried out in a Na+-free medium and followed by 22Na+ presentation resulted in fast radioactive loading that was identical to that occurring in drug-free, NGF-deprived cells and was not prevented by presentation of NGF together with the 22Na+. These data are consistent with a model in which NGF acts through a Na+ pump rather than by restricting Na+ influxes.

The search for neuronotrophic factors addressing CNS neurons requires CNS neuronal cell cultures to quantitate putative effects on neuronal survival. Investigation of neurons dissociated from several embryonic CNS tissues have shown that their short-term survival requires supplementation of the culture medium with either pyruvate or the enzyme catalase. Pyruvate can be replaced with alpha-ketoglutarate or oxaloacetate, or with amino acids capable to transaminate to these three metabolites in the presence of exogenous alpha-ketoacid acceptors. Experiments were designed to evaluate the ability of cultured CNS neurons to utilize glucose as their primary source. We show that: (1) catalase requires the availability of glucose in the medium in order to exert its neuronal maintenance effect, (2) in the absence of catalase, the cells are unable to metabolize glucose through the tricarboxylic acid cycle, (3) catalase restores the neuronal ability to utilize glucose for oxydative metabolism, and renders redundant the use of other sources such as glutamate conversion to alpha-ketoglutarate, (4) graded concentrations of glucose in the medium affect in parallel these metabolic activities and the viability of the cultured neurons, and (5) anti-oxidant agents other than catalase mimic the catalase effects. We conclude that dissociated embryonic CNS neurons suffer from a block in glucose utilization which results from an imbalance between free radical attack and cellular defenses to it and speculate on a more general involvement of peroxidation damage in the trophic requirements for neuronal survival.

The B subunit of cholera toxin, which is multivalent and binds specifically to GM1 ganglioside on the cell surface, has previously been used as a ganglioside-specific probe to regulate DNA synthesis in thymocytes and fibroblasts. To explore in more detail this growth-regulatory action of gangliosides, C6 glioma cells (which are GM1 ganglioside deficient) were used as a model system. When cultures of C6 cells were first treated with GM1, followed by exposure to the B subunit, proliferation was inhibited, as measured by 3H-labeled thymidine incorporation into DNA. Pretreatment of the cells with 50 microM GM1 for 15 min (followed by washing with fetal calf serum) and incubation with 1 microgram/ml of B subunit for 21 h was sufficient to reduce DNA synthesis to 15% of control values (and confirmed by autoradiographic analysis), although maximal inhibition could be achieved with as little as 30 min exposure to B, followed by washing. Furthermore, the B subunit inhibited the response of the C6 cells to basic fibroblast growth factor only following GM1 pretreatment. The B subunit-induced inhibition of DNA synthesis was specific for the ganglioside GM1, and was unrelated to increases of cyclic AMP. These results demonstrate that cell-incorporated GM1 ganglioside may act as a receptor capable of undergoing a specific ligand interaction, subsequently affecting molecular processes at the nuclear level.

— The intracellular concentrations of a number of amino acids were measured in a normal clone of rat glioma cells, and in several independently derived clones selected for gross deficiency of the enzyme hypoxanthine-guanine phosphoribosyltransferase (HGPRT). A significant, approx 2-fold increase in the concentration of free glycine was observed in both mutagenized and non-mutagenized HGPRT deficient clones. The increase in glycine was independent of the phase of cell growth. A similar increase did not occur in HGPRT deficient lymphoblasts.

We had previously reported that in vitro survival of chick embryo ciliary ganglionic neurons can be assured by the addition to the culture medium of appropriate amounts of soluble macromolecular agents termed ciliary neuronotrophic factors. Particularly rich sources of one such factor are aqueous extracts from chick embryo intraocular tissues that include the smooth and striated musculature innervated by ciliary ganglionic neurons. We report here that this eye extract also contains agents that we term ganglionic neuronotrophic factors that support the survival of 11-day chick embryo sympathetic and neonatal mouse dorsal root ganglionic neurons, two traditional targets of nerve growth factor (NGF). Using a recently developed microassay procedure we found that these ganglionic activities are not inactivated by rabbit, rat, or guinea pig antisera raised against the 2.5S (beta) subunit of male mouse submaxillary NGF, rabbit antisera against 7S NGF, or quail antisera against cobra venom NGF. Both the ciliary and ganglionic activities can be quantitated simultaneously by using 24-h in vitro microassays, thus permitting a direct comparison of their respective properties. Both activities were found to (a) adsorb to DE52 cellulose and coelute at a similar salt concentration, (b) focus and be recovered from isoelectric polyacrylamide gels at exactly the same pH region, (c) be heat-and partially acid-labile, but base-stable, and (d) be inactivated by exposure to trypsin. These results suggest that the ciliary and ganglionic neuronotrophic activities are associated with the same protein.

An aqueous extract derived from selected intraocular tissues of 15-day chick embryos contains a soluble macromolecular agent which is capable of ensuring the survival of 8-day chick embryonic ciliary ganglionic neurons in monolayer culture. When this ciliary neuronotrophic factor (CNTF) was concentrated using ultrafiltration and subjected to Sephadex G100 and G200 chromatography, activity was detected in most of the eluted fractions. A peak of the most active fractions was eluted in a region corresponding to a molecular weight of 35-40 X 10(3) and contained about 20-30% of the applied protein. CNTF activity bound readily to DE-52 cellulose resin at neutral pH and was eluted with NaCl in a narrow region containing about 20-40% of the applied protein. Gel electrophoretic staining profiles of the active DE52 fraction indicated considerable (but still only partial) simplification in protein composition. While significant CNTF activity losses were incurred in response to each of the above treatments, an active material could be conveniently generated in one working day in milligram amounts having a specific activity of 60,000 trophic units/mg protein. This trophic activity is in the same range as that of the only other known neuronotrophic factor, Nerve Growth Factor.

A promising advance in our understanding of the mode of action of nerve growth factor (NGF) has been provided by our recent finding that dissociated cells from 8-day embryonic chick dirsal root ganglia (DRG) lose their competence to maintain a low intracellular Na + when incubated without NGF for 6 h, and promptly recover it on delayed presentation of NGF. To ascertain whether control of intracellular Na + is a general feature of the NGF action on its target neurons, we have now tested several NGF-sensitive tissues, both as intact ganglia and dissociates, for their ability to (i) accumulate large amounts of 22 Na + over a period of hours in the absence of NGF, and (ii) rapidly extrude the accumulated radioactivity upon delayed presentation of NGF. Intact and dissociated chick embryo DRG and sympathetic ganglia, as well as dissociated mouse DRG, displayed the expected Na + responses to the lack or the administration of NGF. No such responses were observed with intact mouse DRG, which do not require NGF in explant cultures for neuritic outgrowth, or with intact and dissociated chick embryo ciliary ganglia, which are not sensitive to NGF in either explant or dissociated cell cultures. Thus, in all ganglionic preparations examined, the occurrence of Na + responses to exogenous NGF correlated with traditional responses to NGF in culture. This, together with other recent data, reinforces the view that ionic control may be a critical element in the mechanism of action of NGF.

Secondary microcultures of newborn rat cerebrum astroglial (AG) cells, maintained in a serum-free, chemically defined medium, were treated with various agents known to elevate intracellular cyclic AMP (cAMP) levels. Earlier studies had shown these drugs to induce a process-bearing (stellate) morphology in the AG cells, a response that was antagonized by the presence of gangliosides. One millimolar dibutyryl cyclic AMP (dBcAMP), 10 microM forskolin, 12 nM cholera toxin, and 30 microM isoproterenol all raised intracellular cAMP levels, from basal values of 3 pmol/10(6) cells to 30-30,000 pmol/10(6) cells, depending on the agent tested. dBcAMP caused the greatest elevation, and forskolin the least. The timing and/or the level of the AMP response did not precisely correlate with those of the stellation response. Values of ED50 with the four agents, as determined for the cAMP response, were always higher than stellation ED50 values in all treatments, and ED50 did not correlate with the maximal levels of cyclic AMP induced by the four agents. The capacity of ganglioside GM1 to block the stellation response to the four agents was not accompanied by a similar capacity to block the cAMP responses. Lysophosphatidylserine (lysoPS) had the capacity to induce AG cell stellation as well, without altering the basal level of cAMP. Both lysoPS and gangliosides, therefore, may act directly on the cellular machinery underlying the stellation response without involving changes in intracellular AMP.

We have recently reported the existence within chick embryo heart cell conditioned medium (HCM) of two distinct and independently assayable factors. One agent, ciliary neuronotrophic factor (CNTF), supports the in vitro survival of 8-day chick embryo ciliary ganglionic (CG) neurons. The other factor, polyornithine-attachable neurite promoting factor (PNPF) is required for extensive neuritic growth from these same CNTF-supported CG neurons. In the present study we have examined the occurrence of PNPF activity within nearly 100 different conditioned media using our previously described chick CG bioassay system. From this screening we conclude that: (1) PNPF production is a rather widespread property of cultured neural as well as non-neural cells; and (2) the chick bioassay is sensitive to PNPF activity from all the species examined, including mouse, rat, human and chick cells. We next examined the effects of 3 representative PNPF-containing conditioned media (from chick heart, mouse Schwann and rat Schwannoma) on neurite production from 3 other peripheral ganglionic neuronal cultures (8-day chick dorsal root, 11-day chick sympathetic, and neonatal mouse dorsal root ganglia) as well as 4 central neuronal cultures (8-day chick embryo telencephalon, optic lobe and spinal cord and neonatal mouse cerebellum). The results of these studies indicate: (1) that the peripheral neurons exhibit a dramatic increase in neurite production in response to PNPF which can be easily recognized both qualitatively and quantitatively; whereas (2) the CNS neurons showed essentially no PNPF-induced increase in neurite production. The sole exception to the latter was the appearance within the chick spinal cord cultures of a neuronal population which extended very long neurites in response to PNPF.

Cultures of PC12 pheochromocytoma cells were established on a polyornithine substratum in medium supplemented with the chemically defined N1 mixture in the presence or absence of Nerve Growth Factor (NGF). Normal cell proliferation in the absence of NGF was equally competent when fetal calf serum (FCS) was replaced with N1-supplemented medium. The differentiation of PC12 cells, which occurs upon NGF treatment, ultimately results in cell death without the addition of 0.1% FCS to the N1-supplemented medium. The combination of N1, 0.1% FCS, and NGF permits the PC12 cells to develop a neuritic outgrowth much earlier than when higher (1-10%) FCS levels are used. Neurite retraction is caused in a dose-dependent manner by a delayed presentation of FCS. Within 2 days of serum presentation, however, neurites regrow to achieve that percentage of neurite-bearing cells which is seen without a serum challenge. Moreover, the retraction response becomes less pronounced with time over the 8-day culture period for any given serum concentration. Among the N1 ingredients, only insulin and transferrin are needed by PC12 cells for survival whether in the dividing state or not. Neurite growth was not dependent on any of the N1 components.

Secondary cultures of neonatal rat astroglial cells, maintained in a serum-free, chemically defined medium were treated with several agents thought to activate cyclic AMP-synthesizing systems. Dibutyryl cyclic AMP (dBcAMP), forskolin and cholera toxin promoted, within 2 h, the near-complete conversion of 1-day-old (D1) astroglial cells from a flat, epithelioid morphology to a stellate (star-shaped) morphology. With all 3 agents, cell susceptibility to morphological change declined with culture age, 5-day-old cultures failing to respond altogether. D1 cultures, after 48 h of treatment, had reverted to the flat morphology. Gangliosides reported to stimulate adenylate cyclase were also tested, using purified GM1 X GM1 failed to stimulate the conversion to stellate morphologies. GM1, however, did affect these astroglial cells by causing a block or reversal of their morphological response to dBcAMP, forskolin or cholera toxin. The GM1 response was specific for the intact ganglioside molecule, asialo GM1 and sialic acid having no effect. Gangliosides GD1a, GD1b and GT1b were also active, being effective at ca. 4-fold lower concentrations. The response to GM1 appeared to involve a direct interaction with the astroglial cell, rather than influencing either substratum or medium components.

Chick embryo dorsal root ganglia display a rapid and transient rise in their cyclic AMP content when presented with nerve growth factor. These ganglia also depend on nerve growth factor for control of their intracellular Na+ and K+ levels. A sequential relationship between the cyclic AMP and Na+ responses is not readily apparent. Incubation of chick sensory ganglia in a sodium-free medium does not prevent the cyclic AMP response to nerve growth factor from occurring. When ganglia are first incubated with ouabain for 6 h, presentation of nerve growth factor elicits a cyclic AMP response, but no Na+ response. The cyclic AMP response therefore does not depend on the Na+ environment. An initial presentation of nerve growth factor to the ganglia for 30 min, followed by its withdrawal and subsequent re-administration at different intervals over several hours failed to result in a second cyclic AMP response. Nevertheless, the expected Na+ behaviors were still observed. Dibutyryl cyclic AMP is capable of eliciting a cyclic AMP response in chick sensory ganglia after 6 h of nerve growth factor deprivation. When both agents were presented simultaneously to the ganglia, only a single cyclic AMP response was obtained, corresponding in time to the response elicited by dibutyryl cyclic AMP alone-indicating that this drug acts on the NGF-sensitive cells. At the same time dibutyryl cyclic AMP alone failed to result in a Na+ response, leading one to conclude that the cyclic AMP response to nerve growth factor is truly not mediating the Na+ response. Additional support for the mutual independence of these two short-latency responses is provided by the apparent inability of nerve growth factor to cause a cyclic AMP response in chick embryo sympathetic ganglia, another traditional target for the factor, which is capable of displaying a Na+ response.

Adrenal chromaffin cells from early postnatal rats maintained in culture have previously been shown to grow neuritic processes and survive better in the presence of nerve growth factor (NGF). In the present study we have quantitated the effects on chromaffin cell (postnatal day (D) 8) survival and neurite outgrowth of: NGF, ciliary neuronotrophic factor (CNTF), activities contained in various types of conditioned media (CM), and various substrata (laminin, fibronectin and polyornithine-binding neurite-promoting factor from RN 22 Schwannoma cells - PNPF). At saturating concentrations CNTF (50 ng/ml) and C6 glioma cell CM, (50-fold concentrated) supported survival over the 4-day culture period of all the chromaffin cells present in culture 2 h after seeding. NGF (50 ng/ml) and the non-concentrated CMs from primary Schwann cell and astrocytes as well as Schwannoma and C6 glioma cell cultures, achieved the maintenance of only about half the number of cells above the baseline survival as compared to CNTF and the concentrated C6-CM. These results are compatible with two subsets of D8 chromaffin cells, one only supported by CNTF and the concentrated CM and the other supported by either NGF or CNTF. Either NGF or CNTF elicited neurite outgrowth from 15-20% of the surviving cells. Combination of maximal doses of NGF and CNTF caused a small increase in neurite recruitment beyond that elicited by either factor alone. Low doses of CNTF added to the effect of NGF, shifting the NGF titration curve by about 4-fold. Neurite outgrowth was also induced by the concentrated, but not the unconcentrated C6-CM. Laminin, fibronectin and PNPF did not affect the fibronectin and PNPF did not affect the recruitment of neurites as compared to a polyornithine substratum unless the cultures were supplemented with a neuronotrophic factor and carried for 7 days. However, even before showing effects on neurite recruitment these substrata affected various neuritic performances, such as length, neurite numbers and endings per cell.