Search

Chromaffin cells and sympathetic neurons are derivatives of the sympathoadrenal cell lineage of the neural crest. Although they are similar with respect to many cell biological aspects, chromaffin cells, in contrast to sympathetic neurons, continue to synthesize DNA and proliferate through their whole life span. Large numbers of neural and hormonal signals have been implicated in the regulation of chromaffin cell proliferation based on in vitro studies. We have previously shown that chromaffin cells synthesize and release transforming growth factor-beta (TGF-beta) and that exogenously applied TGF-beta suppresses chromaffin cell proliferation in vitro. We show now that TGF-beta is also a physiologically relevant factor in the control of cell division in developing chromaffin cells. We have neutralized endogenous TGF-beta in quail embryos using a monoclonal antibody recognizing all three TGF-beta isoforms, TGF-beta1, -beta2, and -beta3. Embryos deprived of TGF-beta show a prominent increase in numbers of tyrosine hydroxylase (TH)-immunoreactive adrenal chromaffin cells and TH-positive cells incorporating 5'-bromo-2'deoxyuridine. This is the first documentation of the physiological significance of a factor that has been suggested to play a role in the regulation of chromaffin cell mitosis based on in vitro experiments.

Serotonergic neurons in the central nervous system are crucial in the control of autonomic functions and behavior. Mechanisms by which development and maintenance of the serotonergic transmitter phenotype is regulated include activation of protein kinase A (PKA). Using cultures established from the E14 rat raphe we show here that forskolin (10 microM) increases numbers of neurons expressing tryptophan hydroxylase (TpOH), the key enzyme of serotonin synthesis, and uptake of the false serotonergic transmitter 5, 7-dihydroxytryptamine (5,7-DHT). As shown by short-term treatments the effect is due to phenotype induction rather than survival. To begin to understand downstream or parallel signaling pathways required for the PKA-mediated induction of serotonergic markers, we have studied the putative implication of brain-derived neurotrophic factor (BDNF) and its receptor trkB. Treatment of raphe neurons with forskolin induced BDNF mRNA assayed by competitive RT-PCR. Moreover, trkB-IgG receptor bodies fully prevented the forskolin-induced numerical increase in TpOH- and 5,7-DHT-positive cells suggesting an implication of a TrkB-activated pathway. TrkC-IgG had no effect. K252b, a specific inhibitor of trk kinase activity likewise abolished the induction of serotonergic markers by forskolin. In turn, the inductive effect of BDNF on serotonergic markers was blocked by KT5720, a specific inhibitor of PKA. Taken together, these data suggest that co-activation of cAMP- and trkB-dependent signaling pathways plays a crucial role in the regulation of the serotonergic neuronal phenotype.

Serotonergic (5-HT) neurons of the CNS develop as two separate clusters, a rostral and a caudal group, within the brain stem raphe. We show here that the transforming growth factors -β2 and -β3 (TGF-β) and the TGF-β type II receptor are expressed in the embryonic rat raphe, when 5-HT neurons develop and differentiate. To investigate putative roles of TGF-βs in the regulation of 5-HT neuron development we have generated serum-free cultures isolated either from the rostral or the caudal embryonic rat raphe, respectively. In cultures from the caudal E14 raphe saturating concentrations (5 ng/ml) of TGF-β2 and -β3 augmented numbers of tryptophan hydroxylase (TpOH) -immunoreactive neurons and cells specifically taking up 5,7-dihydroxytryptamine (5,7-DHT) by about 1.7-fold over a period of 4 days. Treatment with TGF-βs also increased uptake of 3H- 5HT uptake about 1.7-fold. Alterations in 5-HT neuron numbers were due to the induction of serotonergic markers rather than increased survival, as shown by the efficacy of delayed short-term treatments. Comparing rostral and caudal raphe cultures from different embryonic ages suggests that distinct effects of TGF-βs reflect the responsiveness of 5-HT neurons at different ages rather than of different origins. J. Neurosci Res. 56:531–538, 1999. © 1999 Wiley-Liss, Inc.

When fetal calf serum (FCS) alone is used as a trophic support for cultured chicken parasympathetic ciliary ganglionic (cCG) neurons, it does not show any survival-promoting effects on these neurons. When FCS is applied to heparin-affinity chromatography, however, potent survival-promoting activity is obtained in the fraction eluted with 0.5 M NaCl. Using cCG neurons as a bioassay system, this neurotrophic activity was purified by a combination of heparin-affinity chromatography, gel filtration chromatography, and Sep-Pak C18 cartridge. The 40-50-kDa fractions from the gel filtration column with strong survival-promoting activity were shown to contain insulin-like growth factor II (IGF-II) by immunoblot analysis. By acidification, the survival-promoting activity and IGF-II were translocated together from the 40-50-kDa to the 7-10-kDa fractions, and the survival-promoting activity in the 7-10-kDa fractions was blocked by an anti-IGF-II neutralizing monoclonal antibody. These results indicate that the neurotrophic substance in 0.5 M NaCl-eluate from heparin-affinity chromatography is IGF-II and that mechanisms may exist in vivo for the activation of latent IGF-II, whose biological effects may be blocked by its specific binding proteins.

Glial cell line-derived neurotrophic factor (GDNF), a member of the transforming growth factor-beta (TGF-beta) superfamily, is a potent neurotrophic factor for several neuron populations in the central and peripheral nervous system. Members of the neurotrophin, neurokine, and TGF-beta families of growth factors can affect neurons beyond their capacity to promote survival. They can play instructive roles including the determination of a particular transmitter phenotype. Here, we show that GDNF enhances the number of calretinin (CaR)-positive neurons in serum-free cultures of striatal cells isolated from embryonic rats. The effect is dose-dependent, can be elicited with concentrations as low as 0.1 ng/ml, and is not accompanied by increased incorporation of 5-bromo-2'-desoxyuridine and appearance of glial fibrillary acidic protein-positive cells. Similar, but weaker effects can be elicited by brain-derived neurotrophic factor, neurotrophin-3 and -4, fibroblast growth factor-2. Ciliary neurotrophic factor, nerve growth factor, and TGF-beta 1 do not affect striatal CaR expression. GDNF can augment CaR-positive cells at any time point and with a minimal exposure of 18 hr, suggesting induction of the phenotype rather than increased survival. By reverse transcription polymerase chain reaction (RT-PCR), we show that GDNF is expressed in the E16 striatum and in cultures derived from this tissue. GDNF also protected striatal CaR-positive neurons against glutamate toxicity. We conclude that striatal GDNF, in addition to its retrograde trophic role for nigrostriatal dopaminergic neurons, may also act locally within the striatum (e.g., by inducing the CaR phenotype and protecting these cells against toxic insult).

Basic fibroblast growth factor (bFGF; FGF-2) has potent trophic effects on developing and toxically impaired midbrain dopaminergic (DAergic) neurons which are crucially affected in Parkinson's disease. The trophic effects of FGF-2 are largely indirect, both in vitro and in vivo, and possibly involve intermediate actions of astrocytes and other glial cells. To further investigate the cellular and molecular mechanisms underlying the restorative actions of FGF-2, and to analyse in more detail the changes within astroglial cells in the MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine)-lesioned striatum, we have studied striatal expression and regulation of connexin-43 (cx43), the principal gap junction protein of astroglial cells, along with the expression of glial fibrillary acidic protein (GFAP), FGF-2, and functional coupling. Our results show an immediate, yet transient increase in cx43 mRNA, and a sustained increase in FGF-2 mRNA, GFAP-positive cells, and cx43-immunoreactive punctata following the MPTP lesion, without any induction of functional coupling between astrocytes and other glial cells as revealed by dye coupling of patched cells. Unilateral administration of FGF-2 in a piece of gelfoam caused a further increase in cx43-positive punctata immediately adjacent to the implant, which was more pronounced than after application of a gelfoam containing the non-trophic control protein in cytochrome C. These changes were parallelled by a small increase in cx43 protein determined by Western blot, but not by alterations in the coupling state of cells in the vicinity of the gelfoam implant. Although our data indicate that MPTP and exogenous FGF-2 may alter expression and protein levels of cx43, they do not support the notion that increases in cellular coupling may underly the trophic and widespread actions of FGF-2 in the MPTP-model of Parkinson's disease. © 1996 Wiley-Liss, Inc.

Growth/differentiation factor 5 (GDF5) is a novel member of the transforming growth factor-beta (TGF-beta) superfamily of multifunctional cytokines. We show here that GDF5 is expresed in the developing CNS including the mesencephalon and acts as a neurotrophic, survival promoting molecule for rat dopaminergic midbrain neurons, which degenerate in Parkinson's disease. Recombinant human GDF5 supports dopaminergic neurons, dissected at embryonic day (E) 14 and cultured for 8 days under serum-free conditions, to almost the same extent as TGF-beta 3, and is as effective as glial cell line-derived neurotrophic factor (GDNF), two established trophic factors for midbrain dopaminergic neurons. In contrast to TGF-beta and GDNF, GDF5 augments numbers of astroglial cells in the cultures, suggesting that it may act indirectly and through pathways different from those triggered by TGF-beta and GDNF. GDF5 also protects dopaminergic neurons against the toxicity of N-methylpyridinium ion (MPP+), which selectively damages dopaminergic neurons through mechanisms currently debated in the etiology of Parkinson's disease (PD). GDF5 may therefore now be tested in animal models of PD and might become useful in the treatment of PD.

Growth factors can induce both proliferation or differentiation of neuroblastoma (NB) cells through interaction with specific receptors. Using two automated colorimetric assays for determinations of cell numbers, the present study demonstrates that (a) different NB and neuroepithelioma cell lines show distinct responses, both qualitatively and quantitatively, to basic FGF (bFGF), NGF, and EGF(b) even closely related NB cell lines (e.g., SK-N-SH, SH-SY5Y, and SHEP) do not respond uniformly to these factors; c) responses of the two neuroepithelioma cell lines employed (SK-N-MC and CHP-100) differ, but match those of certain NB cell lines; and d) two growth factors, bFGF and EGF, may both stimulate or inhibit proliferation, depending on the cell line studied. Specifically, IMR-32, SK-N-SH, and SH-SY5Y showed a mitogenic response to each growth factor. Maximal proliferative responses ranged from 204–355% as compared to controls (100%). GICAN was stimulated by NGF (199%), and SK-N-MC and NMB by EGF (282 and 140%, respectively), but other factors were ineffective. CHP-100 and GIMEN were inhibited by bFGF. NGF and EGF were not effective on CHP-100 cells, while EGF caused an arrest of mitogenic activity in GIMEN cells, and NGF stimulated their proliferation. Cell lines SHEP and LAN1 did not respond to any factor. To begin to analyze putative relationships of growth factor responsiveness and growth factor/growth factor receptor expressions, IMR-32, GIMEN, and LAN1 cell lines were studied for the presence of bFGF, NGF, FGF receptors (R)-1 (flg) and FGFR-4, trk, and low-affinity NGF receptor (p75) mRNAs. All three cell lines expressed bFGF and NGF mRNA, but not the FGFR-1, FGFR-4, trk, and p75 mRNAs. These results suggest extremely diverse patterns of NB/neuroepithelioma cell responsiveness to “mitogenic” growth factors and no overt correlation between such responses and growth factor/growth factor receptor expression. © 1995 Wiley-Liss, Inc.

The transforming growth factors beta (TGF-beta), a family of regulatory polypeptides, are involved in numerous vital processes including inflammation and wound healing. Since repair of a peripheral nerve lesion includes a series of well-defined steps of cellular actions possibly controlled by TGF-beta s, and since TGF-beta mRNA and immunoreactivity have been found in the normal peripheral nerve, we have examined in the lesioned peripheral nerve. Sciatic nerves of adult rats were either crushed (allowing axonal regeneration) or transfected (to prevent axonal regeneration and to induce Wallerian degeneration in the distal stump). After intervals of 6 hours, 2 and 6 days post-lesion, the rats were sacrificed and each nerve was cut into four segments, two proximal and two distal to the lesion site. TGF-beta 1-3 mRNA were determined for each segment. We demonstrate that TGF-beta 1 mRNA levels are higher than those of TGF-beta 3; the amplitude of mRNA regulation depends on time, type of lesion and localization relative to the lesion site. TGF-beta 2 mRNA could not be detected. For TGF-beta 1-3 immunocytochemistry, animals were sacrificed 12, 24, 48, 72 hours and 7 and 14 days after surgery. TGF-beta immunoreactivity (IR) was observed for all isoforms in lesioned and unlesioned nerves. In the segment directly adjacent to the lesion at its proximal side, an increase of TGF-beta-IR became apparent as soon as 12 hours after surgery; it remained elevated during the whole period observed in both models. In the segment adjoining the distal side of the lesion, an increase of TGF-beta-IR was observed after 48 hours, which was still present after 14 days. At day 7 after crush or transection, an increase of TGF-beta-IR was detected in the most distal segments, which reached its highest levels at the end of our observation period. Our results suggest that the presence of axonal contact might induce an enhancement of TGF-beta expression by Schwann cells in the distal stump of a lesioned and regenerating peripheral nerve. Since we demonstrate an increase of TGF-beta mRNA and protein expression also in the distal stump of transected nerves where Schwann cells are not able to contact sprouting axons from the proximal part, other regulatory pathways must exist. The acquisition of a "reactive" Schwann cell phenotype after peripheral nerve lesion might involve an upregulation of TGF-beta expression.

Neuroblastoma cells have been shown to contain growth factors, which may be involved in the regulation of cell proliferation and/or differentiation. We have examined whether human IMR-32 neuroblastoma cells store factors with a capacity to promote neuron survival and differentiation. Heparin affinity chromatography, cell blotting, and isoelectric focusing of IMR-32 cell extracts revealed multiple neurotrophic activities at molecular weights of 16.8, 24.1, 39.0, 45.3, 52.3 and isoelectric points of 4.0, 5.0, 8.3, 9.0. Immunocytochemistry, immunoblotting, and radio-immunoassay with specific antibodies suggests that one neurotrophic activity is immunologically identical with basic fibroblast growth factor (bFGF). This assumption is supported by Northern blot analysis, which shows a 6.0 kb bFGF transcript.

To further clarify the function of basic fibroblast growth factor (bFGF) in the nervous system, we have examined its distribution, lesion-dependent regulation, retrograde transport, and trophic roles on rat hypoglossal neurons. In adult rats, bFGF-like immunoreactivity is localized in hypoglossal motoneurons, drastically reduced 2 days after axotomy, and re-expressed by 11 days. Neuron numbers and morphology assessed by Nissl staining are not affected by the lesion. 125J bFGF is specifically retrogradely transported by hypoglossal motoneurons from their peripheral nerve terminals. Moreover, bFGF stimulates the in vitro survival of hypoglossal neurons (ED50 2 ng/ml). In vivo administration of bFGF prevents lesion-induced motoneuron death to 14% in 7 day old rats and to 60% in 18 day old rats, but not the axotomy-induced decrease of choline acetyltransferase activity in the hypoglossal nucleus of adult rats. These results are consistent with a neurotrophic role of bFGF in the hypoglossal system. © 1992 Wiley-Liss, Inc.

Alterations in neuronal morphology occur in the brain during normal aging, but vary depending on neuronal cell types and brain regions. Such alterations have been related to memory and cognitive impairment. Changes in hippocampal spine densities are thought to represent a morphological correlate of altered brain functions associated with hippocampal-dependent learning and memory. We therefore have analyzed the impact of aging on different hippocampal-dependent learning tasks and on changes in dendritic spines of CA1 hippocampal and dentate gyrus neurons by analyzing adult (6–7 months) and aged (21–22 months) C57/Bl6 mice. We found a significant decrease in spine numbers of basal CA1 dendrites and decreases in spine length of apical dendrites of CA1 and dentate gyrus neurons. Furthermore, aged mice exhibited significant deficits in hippocampus-dependent learning tasks, such as the probe trial of the Morris water maze and T maze learning. Given the fact that there is no neuronal loss in the hippocampus in aged mice (von Bohlen und Halbach and Unsicker [2002] Eur. J. Neurosci. 16:2434–2440), we suggest that the memory and cognitive decline in the context of aging may be accompanied by rather subtle anatomical changes, such as numbers and morphology of dendritic spines. © 2006 Wiley-Liss, Inc.

Persephin (PSP) is the most recently discovered member of the GDNF family of neurotrophic factors. We have used an RT-PCR approach to start addressing the putative functional significance of PSP by determining sites of its synthesis in the neonatal rat brain. Generally, two transcripts were found. Sequence analysis of the transcripts identifies an 88 bp intronic sequence. Neural tissues analysed included cortex, hippocampus, striatum, diencephalon, mesencephalon, cerebellum, hindbrain and spinal cord as well as superior cervical, dorsal root ganglia, adrenal gland, and PC12 pheochromocytoma cells. As non-neuronal tissues, sciatic nerve, optic nerve, primary astroglial, oligodendroglial, O2A progenitor, and glioma cells (C6, B49) were also included. All tissues/cells except oligodendrocytes and O2A progenitor cells were strongly positive for PSP mRNA. To test the hypothesis of whether PSP might act as a target-derived factor, as suggested for GDNF, the motoneuron-muscle axis has been analysed. PSP is synthesized in skeletal muscle and, to a higher extent, in the spinal cord. Moreover, PSP is synthesized in purified embryonic motoneurons. Together, these data do not support a role for PSP as a typical target-derived neurotrophic factor for motoneurons. We conclude that PSP is synthesized throughout the nervous system and that it is presumably of both astroglial and neuronal origin, in contrast to GDNF and neurturin, which seem to be predominantly of neuronal origin.

Chromaffin cells grafted to the brain of animals with experimental parkinsonism and patients with Parkinson's disease can restore nigrostriatal functions. Mechanisms underlying these beneficial effects are unknown, but may include growth factors rather than the minute amounts of dopamine (DA) liberated from chromaffin cells. We now report that protein from chromaffin granules, which release their contents by exocytosis, promotes survival and uptake of 3H-DA of mesencephalic DAergic neurons in vitro and protect against N-methylpyridinium ion toxicity. This neurotrophic effect is accompanied by cell proliferation and mediated by astroglial cells induced in these cultures. Inhibition of cell proliferation and concomitant astrogliosis by 5-fluorodeoxyuridine and alpha-aminoadipic acid abolishes the trophic effect. Two highly specific inhibitors of the epidermal growth factor receptor (EGFR) signal transduction pathway, 4,5-dianilinophthalimide (10 microM) and tyrphostin B56 (10 microM), selectively block the neurotrophic capacity of chromaffin granule protein. As expected, they also block the mitogenic effects of EGF and TGF-alpha. However, these two mitogens do not mimic the pronounced mitogenic and trophic actions of chromaffin granule protein. Culture medium conditioned by mesencephalic cells pretreated with chromaffin granule protein promotes survival of DAergic neurons without increasing numbers of astroglial cells. The effective molecule is unlikely to be glial cell line-derived neurotrophic factor, whose mRNA is not detectable in cultures treated with chromaffin granule protein. We conclude that chromaffin granules contain a putatively novel growth factor, which signals through the EGFR and may be responsible for the known protective and restorative actions of chromaffin cell grafts to the lesioned nigrostriatal system.

The protective role of basic fibroblast growth factor (FGF-2) for 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)- and methylpyridiniumion (MPP+)-lesioned dopaminergic (DAergic) nigrostriatal neurons was studied, using dissociated cell cultures of embryonic day (E) 14 rat mesencephalon. Cells were grown in different culture media and received FGF-2 (5 ng/ml) and/or the toxins (5 μM) at various schedules, but were consistently allowed to differentiate for 3 days prior to becoming exposed to the toxin. Survival of tyrosine hydroxylase (TH)-immunoreactive cells at 7 days was only markedly impaired by MPTP, if horse serum (HS) or bovine serum albumine (BSA) were omitted from the culture medium. FGF-2 increased the number of TH-immunoreactive cells, and this increase not diminished by MPTP under any culture condition. Uptake of 3H-DA was significantly reduced by MPTP in HS- and BSA-containing, but not in protein-less cultures. A protective effect by FGF-2 was only seen in the presence of BSA. MPP+ caused a more pronounced reduction in 3H-DA uptake than MPTP, and this effect was partially reversed by the addition of FGF-2, unless cultures contained HS. Neurofilament protein (NF), an indirect measure for the total number of neurons present in the cultures, was not significantly reduced by MPTP or MPP+ corroborating the specificity of the toxin for DAergic neurons, which constitute only a minor fraction in these cultures. In line with the wide spectrum of target neurons of FGF-2, this factor significantly increased NF contents under any culture condition. Quantification of the amounts of glial fibrillary acidic protein (GFAP) revealed stimulatory effects of FGF-2 (2.5- to 4-fold) and at least 10-fold higher levels in the presence as compared to the absence of HS. These data show that FGF-2 can protect DAergic neurons against MPTP- and MPP+-mediated damage. However, the effects of the toxins as well as of FGF-2 are partially dependent on culture conditions. Variations in the effectiveness of toxins and FGF-2 are not overtly related to the total numbers of neurons or astroglial cells, but may reflect culture type-dependent alterations of neuronal and glial metabolism. © 1993 Wiley-Liss, Inc.

The presence of fibroblast growth factor (FGF) was investigated by immunocytochemistry in cultured neuronal cells derived from the peripheral (PNS) and the central nervous system (CNS) of chick and rat embryos. Polyclonal antimouse FGF antibodies, which cross-react with basic and acidic FGF, were used in the peroxidase immunocytochemical staining method. FGF immunoreactivity was found in neurons. Staining intensity in chick and rat brain neuronal cells increased during the culture period, reached a maximum after 6-8 days, and subsequently declined. Embryonic chick ciliary and dorsal root ganglionic (DRG) neurons as well as DRG neurons from newborn rat displayed intense FGF immunoreactivity at 12 and 48 hr in culture. Nonneuronal cells were not stained. The data demonstrate for the first time that cultured neurons from the CNS and PNS contain FGF.

In a search for neurotrophic factors (NTFs) regulating retinal ganglion cell (RGC) death in the chick embryo we have used purified and cultured RGCs. Purification of RGCs from embryonic day 10 was achieved by employing the "panning" method (Silverstein and Chun: Soc Neurosci Abstr 13:1054, 1987). The obtained neuron population consisted of 97% RGCs as demonstrated by retrograde labeling with a fluorescence dye. RGCs were cultured at low density in a chemically defined medium and short-term survival (24 hr) was determined. In the absence of NTFs, less than 3% of the RGCs survived. In the presence of various crude or purified NTFs (eye, brain, and tectum extracts; glial-conditioned medium; ciliary neurotrophic factor [CNTF]; nerve growth factor [NGF]) 31% to 52% of the RGCs were maintained. The effects of NGF and CNTF were not additive. Neither acidic nor basic fibroblast growth factor was able to maintain RGCs in culture. Our results, obtained with a culture system which allowed the analysis of direct trophic actions, suggest that NGF and CNTF may be NTFs for overlapping subpopulations of chick RGCs.

Basic fibroblast growth factor (bFGF) recently has been established as a survival- and transmitter-promoting neurotrophic agent for embryonic neurons in vitro. Its local application to lesioned adult optic and sciatic nerves has been shown to rescue axotomized retinal and sensory neurons that otherwise die. Following transection of the fimbria fornix pathway connecting the medial septum (MS) to the hippocampus, MS neurons undergo severe cell death, which can be prevented partially by infusion of nerve growth factor (NGF). In the same lesion paradigm, we find that 87% of these neurons visualized by cresyl-violet staining have disappeared by 4 weeks after unilateral fimbria fornix transection in adult rats. Implantation of gel foam soaked with 8 micrograms bFGF reduced neuron death to 68%. A similar rescue effect was seen with 0.3 microgram NGF. NGF administered at 20 micrograms reduced cell losses to 54%. Thus, bFGF rescued 22% and NGF at 20 micrograms 38% of the neurons that otherwise would have died. Choline acetyltransferase immunocytochemistry revealed dramatic losses of cholinergic neurons on the lesioned, compared with the unlesioned, side. Cholinergic neuron death was clearly reduced by the bFGF and NGF treatments. Basic FGF, in contrast to NGF, did not prevent a reduction in size of surviving neuronal cell bodies. Considered in the context of FGF being present in brain and hippocampal neurons, our results suggest a possible role for FGF as a neurotrophic factor for CNS neurons in vivo.