SEARCH

SEARCH BY CITATION

Keywords:

  • brain;
  • Erk;
  • GM1;
  • neurotrophins;
  • Trk

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We investigated the ability of GM1 to induce phosphorylation of the tyrosine kinase receptor for neurotrophins, Trk, in rat brain, and activation of possible down-stream signaling cascades. GM1 increased phosphorylated Trk (pTrk) in slices of striatum, hippocampus and frontal cortex in a concentration- and time-dependent manner, and enhanced the activity of Trk kinase resulting in receptor autophosphorylation. The ability of GM1 to induce pTrk was shared by other gangliosides, and was blocked by the selective Trk kinase inhibitors K252a and AG879. GM1 induced phosphorylation of TrkA > TrkC > TrkB in a region-specific distribution. Adding GM1 to brain slices activated extracellular-regulated protein kinases (Erks) in all three brain regions studied. In striatum, GM1 elicited activation of Erk2 > Erk1 in a time-and concentration-dependent manner. The GM1 effect on Erk2 was mimicked by other gangliosides, and was blocked by the Trk kinase inhibitors K252a and AG879. Pertussis toxin, as well as Src protein tyrosine kinase and protein kinase C inhibitors, did not prevent the GM1-induced activation of Erk2, apparently excluding the participation of Gi and Gq/11 protein-coupled receptors. Intracerebroventricular administration of GM1 induced a transient phosphorylation of TrkA and Erk1/2 in the striatum and hippocampus complementing the in situ studies. These observations support a role for GM1 in modulating Trk and Erk phosphorylation and activity in brain.

Abbreviations used:
BDNF

brain-derived neurotrophic factor

Erk

extracellular signal-regulated kinase

GPCRs

G protein-coupled receptors

LDH

lactic acid dehydrogenase

MAPK

mitogen-activated protein kinase

MBP

myelin basic protein

MEK

MAPK kinase

NGF

nerve growth factor

NT-3

neurotrophin-3

Trk

high-affinity tyrosine kinase receptor for neurotrophins.

Gangliosides are components of most cell membranes, and are particularly abundant in the brain where they represent the major lipid constituent of the neuronal surface. They are thought to play a role in development, cell differentiation and oncogenic transformation (Schengrund 1990; Zeller and Marchase 1992). The monosialic acid ganglioside GM1 promotes neuronal growth and differentiation in cell cultures, and enhances phenotypic expression and neuronal repair in animal models of neurotrauma and aging. The pleiotropic neurotrophic activity of GM1 extends to multiple neuronal phenotypes in the central nervous system, including cholinergic, dopaminergic, serotoninergic and noradrenergic neurons (Hadjiconstantinou and Neff 1998).

The mechanisms for the neurotrophic actions of GM1 are not completely understood. The wide range of neuronal populations where GM1 appears to be efficacious, suggests that it may interact with a large number of neurotrophic factors or activate common pathways used by these factors for signaling. Of interest are data indicating that GM1 might interact with neurotrophins and their receptors in vivo and in vitro. Indeed, GM1 potentiates the neurotrophic effect of NGF in vivo (Cuello et al. 1989; Fong et al. 1995; for review see Hadjiconstantinou and Neff 1998, 2000) and enhances the NGF-induced tyrosine kinase receptor for neurotrophin (Trk) phosphorylation and activation (Ferrari et al. 1995; Mutoh et al. 1995), as well as NGF-induced TrkA phosphorylation and dimerization (Rabin and Mocchetti 1995; Farooqui et al. 1997). Tyrosine phosphorylation and activation of TrkA by GM1 alone has been demonstrated in C6-glioma cells expressing TrkA (Rabin and Mocchetti 1995). As the aforementioned observations were made using tumor cell lines, often over-expressing the neurotrophin receptor, under artificial conditions, their biological relevance for the neurotrophic actions of GM1 in brain remains to be demonstrated.

Binding of neurotrophins to Trk induces activation of the receptor tyrosine kinase, dimerization and autophosphorylation, and initiates a complex cascade of signal transduction events. Tyrosine phosphorylated Trk binds to Shc, phospholipase Cγ and phosphatidyl inositol 3-kinase (PI3-kinase) through SH2 domains resulting in their phosphorylation and subsequent activation of intracellular signaling pathways and transcription factors regulating gene expression (for review see Kaplan and Stephens 1994; Klesse and Parada 1999). These pathways transduce signals independently, but also can converge on the same downstream effector. Stimulation of a variety of tyrosine kinase receptors leads to a rapid elevation of the enzymatic activity of a family of structurally related serine–threonine kinases, known as mitogen-activated protein kinases (MAPKs), which convert extracellular stimuli to intracellular signals that control gene expression (Schaeffer and Weber 1999). The Erk (extracellular signal-regulated kinase) subfamily of MAPKs, is activated in response to Trk stimulation via the small G proteins, Ras and Rap1, which are the targets of multiple second messengers and kinases (Grewal et al. 1999). Thus, Erks serve as a convergence point where signals from multiple transduction pathways are integrated and processed to the nucleus. The Erk1 and Erk2 isoforms of MAPKs are activated by neurotrophins and are thought to mediate some of their survival and differentiative actions in specific subsets of peripheral and central neurons (Klesse and Parada 1999).

Our laboratory has a longstanding interest in the neurotrophic actions of exogenous GM1 on brain after an injury and during aging (Hadjiconstantinou and Neff 1988; Hadjiconstantinou et al. 1990, 1992; Goettl et al. 1999). Given the overwhelming evidence that GM1 not only mimics but also potentiates the neurotrophic action of neurotrophins in some central neuronal systems (Hadjiconstantinou and Neff 1998, 2000), we explored whether the ganglioside was capable of inducing tyrosine phosphorylation and activation of Trk and initiating signal transduction similar to that of neurotrophins. Toward this goal, we used brain slices to investigate the ability of GM1 to induce tyrosine phosphorylation of Trk in brain slices in situ, and activate Erk1 and Erk2, a convergence point of Trk transduction signaling. Slices of striatum, frontal cortex and hippocampus of adult rats were treated with GM1, other gangliosides and neurotrophins, and phosphorylation and activation of Trks and Erks estimated. To prove the validity and applicability of our in situ system, the effect of intracerebroventricularly (ICV) administered GM1 on the phopshorylation of TrkA and Erks was studied in vivo. To our knowledge this is the first study of GM1-induced phosphorylation/activation of Trks and Erks in brain tissue.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals and treatments

Male Sprague–Dawley rats, 250–300 g (Harlan Laboratories, Indianapolis, IN, USA) were used for the studies, which were conducted in accordance with the Guide for Care and Use of Laboratory Animals as adopted by the National Institutes of Health and approved by the Ohio State University Institutional Laboratory Animal Care and Use Committee.

In situ studies

Slices (200 × 200 μm) prepared from striatum, hippocampus and frontal cortex were suspended in oxygenated Krebs buffer, pH 7.4, and incubated for 20 min at 37°C. One milliliter of slice aliquots, resuspended in fresh Krebs buffer, were placed into test tubes, GM1 (Fidia Res. Labs, Abanoterme, Italy), other sphingolipids (Sigma, St Louis, MO, USA) or neurotrophins (Calbiochem, San Diego, CA, USA) added and incubation continued for various times as indicated in Figures and Tables. Reactions were terminated by a 10-s centrifugation, and slices were lysed for 30 min at 4°C. For the Trk studies the lysis buffer contained 137 mm NaCl, 20 mm Tris, pH 8.0, 1% Triton X-100, 10% glycerol, 1 mm phenylmethylsulfonyl fluoride and 1 mm sodium orthovanadate; whereas, for the Erk studies lysis buffer contained 20 mm HEPES, pH 7.4, 1% Triton X-100, 30 mm sodium pyrophosphate, 50 mm sodium fluoride, 1 mm EGTA, 2 mm EDTA, 1 mm dithiothreitol (DTT), 1 mm phenylmethylsylfonyl fluoride, 1 μg/mL aprotinin and 1 mm sodium orthovanadate. Insoluble material was removed by centrifugation at 15 000 g for 10 min, and proteins were determined by the Bradford method (Bio-Rad Laboratories, Hercules, CA, USA). In some experiments, kinase inhibitors, K252a, AG879, ZM 336372, PD 98059, Ro 32–0432, herbimycin A (Calbiochem) and pertussis toxin (List Biological Labs, Campbell, CA, USA) were added prior to or with GM1 or neurotrophins. The concentration of the compounds used and the duration of the incubation were determined from the literature or initial time- and concentration–response studies. Viability of slices was determined by estimating lactic acid dehydrogenase (LDH) released during incubation (Sigma). Under the experimental conditions used, no changes in LDH were observed for up to 2 h of incubation.

In vivo studies

Following anesthesia with isofuorane GM1, 10–20 nmol, or artificial CSF was administered free hand into the left ventricle using the coordinates of 1 mm caudal to the bregma, 1 mm lateral to the sagittal suture, and 4.5 mm from the surface of the skull. The doses chosen were calculated based on an estimate of 200 μL of CSF and our observation that 50–100 μm of GM1 produce maximal phosphorylation/activation of Trks and Erks in brain slices (see Results). Animals recovered from anesthesia within 5 min and were decapitated 30 or 90 min after the injection. The injection site was verified by visual inspection and the striatum and hippocampus ipsilateral to injection were dissected and immediately lysed as described above for the brain slices.

Procedures

For the in situ studies, tyrosine phosphorylated Trks (pTrks) were routinely estimated in immunoprecipitates by western blot, and in vitro kinase assays were performed in some studies to show Trk kinase activation and receptor autophosphorylation. Erk activation was routinely studied with in vitro kinase assays after immunoprecipitation using myelin basic protein (MBP) as substrate. Some western blots were performed with antiactive Erk antibodies for comparison. For the in vivo studies, phosphorylated TrkA (pTrkA), Erk1 (pErk1) and Erk2 (pErk2) were assessed with western blots using antibodies against pTrkA and pErk1/2.

Western blot

Estimation of pTrk  Immunoprecipitation was performed by incubating equal amounts of lysate protein (2–4 mg) with 2 μg of anti-Trk (C14, Santa Cruz Laboratories, Santa Cruz, CA, USA), anti-TrkB (Transduction Laboratories, Lexington, KY, USA) or anti-TrkC (798, Santa Cruz Laboratories) antibodies, followed by Protein A Sepharose (Pharmacia Corp., Peapack, NJ, USA). Immunoprecipitated proteins were eluted by boiling for 5 min in Laemmli sample buffer and equal amounts separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). Proteins were electrotransferred to nitrocellulose membrane, and incubated with antiphosphotyrosine (4G10, 1 μg/mL, Upstate Biotechnology, Lake Placid, NY, USA) or antiphosphorylated TrkA (1 : 1000 dilution, Santa Cruz Laboratories) antibody in TBS–Tween (10 mm Tris–HCl, pH 7.5, 150 mm NaCl, 0.1% Tween-20), followed by a horseradish peroxidase-conjugated secondary antibody (Santa Cruz Laboratories). Bound antibody was detected by Enhanced Chemiluminescence (ECL, Amersham, Piscataway, NJ, USA). Membranes were stripped and reprobed with an anti-Trk antibody to evaluate the efficiency of the immunoprecipitation and the amount of precipitated Trk in each sample. For each study, phosphorylation of Trk by NGF, 100 ng/mL, at 5 min served as a control to determine the reproducibility of our experiments and the relative magnitude of the GM1, or other compounds, response. In studies with TrkB or TrkC, BDNF or NT-3, 100 ng/mL, at 5 min were used.

Estimation of pErk1 and pErk2  Equal amounts of lysate protein (50 μg) were separated by 12% SDS–PAGE and transferred to nitrocellulose membranes. Blots were incubated with antiactive Erk antibody (recognizes both pErk1 and pErk2, 1 : 5000 dilution, Promega, Madison, WI, USA), followed by horseradish peroxidase-conjugated secondary antibody and detection by ECL. After stripping, blots were reprobed with anti-Erk1 or anti-Erk2 antibody (0.5 μg/mL, Santa Cruz Laboratories) to determine total Erk protein.

In vitro kinase assay

Estimation of Trk kinase activity  Trk immunoprecipitates were incubated in 50 μL kinase buffer (20 mm Tris, pH 7.4, 10 mm magnesium acetate) with 10 μCi [32P]γATP (3000 Ci/mmol, Amersham) at 30°C for 15 min. The reaction was stopped by adding 50 μL Laemmli sample buffer, proteins eluted by boiling for 5 min, separated by 10% SDS–PAGE electrophoresis, and transferred to a polyvinylidene difluoride membrane (Amersham). Then, membranes were incubated for 1 h at 55°C in 1 m KOH, and labeled alkali-resistant phosphoproteins visualized by autoradiography.

Estimation of Erk1 and Erk2 kinase activity  Equal amounts of lysate protein (2 mg) were incubated with 2 μg of anti-Erk1 or anti-Erk2 antibodies and 30 μL of Protein A Sepharose (Pharmacia) for 3 h at 4°C. Erk1 and Erk2 immunoprecipitates were incubated with 50 μL of kinase buffer containing 20 mm HEPES, pH 7.5, 10 mm MgCl2, 0.5 mg/mL MBP (Sigma), 50 μm ATP and 10 μCi [32P]γATP (3000 Ci/mmol, Amersham) at 30°C for 30 min. The reaction was stopped by boiling for 5 min in 50 μL Laemmli sample buffer. After separation by 12% SDS–PAGE, proteins were transferred to nitrocellulose membranes, and exposed to X-ray films. Subsequently, bands corresponding to phosphorylated MBP stained with Ponceau red were excised, and radioactivity determined by liquid scintillation counting.

Data collection and statistical analysis  For all western blots the signal density of bands on X-ray film was quantified by image analysis (Universal Imaging Corporation, Downingtown, PA, USA). After correction by the amount of protein, data were expressed as percent of basal in the same blot. Erk kinase activity was estimated as dpm/mg prot and expressed as percent of basal in the same experiment. For the statistical analysis of data, a non parametric one way analysis of variance, Kruskal–Wallis test, followed by a Dunn's multiple comparisons test or an independent Mann–Whitney test was used. Analysis was performed with Graphpad Instat, version 3.05, software, and a level of p < 0.05 was considered as statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

GM1 and Trk phosphorylation and activation in brain slices

Knüsel et al. (1994, 1996) introduced slices of adult rodent brain as a useful in situ system to investigate Trk activation and phosphorylation by NGF and other neurotrophins. In these studies, we utilized brain slices to test whether GM1 was able to induce activation/phopshorylation of Trk as has been postulated from in vivo experiments (Cuello et al. 1989; Fong et al. 1995; for review see Hadjiconstantinou and Neff 1998, 2000) and demonstrated in cell lines expressing Trk (Ferrari et al. 1995; Rabin and Mocchetti 1995).

GM1 and NGF induce tyrosine phosphorylation and activation of Trk in brain slices

When striatal slices were incubated with GM1 or NGF there was an increase in phosphotyrosine detected in Trk immunoprecipitated by an antiphosphotyrosine antibody (Table 1a). In addition, exposure to GM1 or NGF increased Trk kinase activity and resulted in Trk autophosphorylation as demonstrated by an immune complex in vitro kinase assay (Table 1b). As tyrosine phosphorylation of Trk by NGF in brain slices has been reported (Knüsel et al. 1994, 1996), further investigation of the NGF effect was not pursued, and NGF served as a control to evaluate the reproducibility of our experiments. In addition to striatum, GM1 and NGF increased Trk phosphotyrosine in frontal cortex and hippocampus (Table 1). Under the conditions used, stimulation with NGF or GM1 increased pTrk in the order, striatum > hippocampus = frontal cortex. The magnitude of response to GM1 was about twofold in frontal cortex and hippocampus and about three- to fourfold in striatum. In striatum, the response to NGF was greater but comparable to that of GM1, about four- to fivefold. The basal level of tyrosine phosphorylation of Trk varied between brain regions and with experiments. As a general observation, basal phosphorylation of Trk in striatum was relatively low, while in hippocampus and frontal cortex a variably higher level was the norm.

Table 1.  GM1- and NGF-induced Trk tyrosine phosphorylation
ConditionsStriatumHippocampusFrontal cortex
  1. Brain slices were incubated with GM1, 30 min, or NGF, 5 min. pTrk was estimated by western blot using an antiphosphotyrosine antibody after Trk immunoprecipitation, as described in methods. n = 6. ap < 0.05 compared with basal. (1a) Representative western blot from a study with striatal slices; (1b) representative image of Trk autophos- phorylation assessed with an in vitro kinase assay in striatal slices.

Basal100 ± 8100 ± 1100 ± 1
GM1 (100 µm)352 ± 59a212 ± 26a212 ± 27a
NGF (100 ng/mL)472 ± 59a258 ± 32a192 ± 15a
The GM1-induced Trk tyrosine phosphorylation is time- and concentration-dependent

GM1 induced a time- and concentration-dependent increase of pTrk in all three brain regions examined. In striatum, phosphorylation was evident in 2.5 min, reached a maximal by 20–30 min, and the signal declined thereafter returning to basal by 2 h (Fig. 1a). Concentration–response studies demonstrated that 100 μm of GM1 produced a maximal effect with increases in pTrk being evident with 5 μm of GM1. High concentrations of GM1, however, attenuated the response (Fig. 1b). Similar results were observed for hippocampus and frontal cortex (data not shown). The apparent EC50 values for GM1 ranged from 28 to 45 μm in all three regions (data not shown).

image

Figure 1. GM1-induced tyrosine phosphorylation of Trk is time-and concentration-dependent. For time-response studies (a), striatal slices were incubated with GM1, 100 μm, for the indicated time periods, and for concentration–response studies (b), slices were incubated with the indicated concentrations of GM1 for 30 min. pTrk was estimated by western blot, as in Table 1. n = 4–7. *p < 0.05 compared with basal.

Download figure to PowerPoint

Induction of Trk tyrosine phosphorylation by other gangliosides

To determine the selectivity of GM1 on Trk phosphorylation, the ability of other gangliosides and relevant compounds to induce Trk phosphorylation was investigated in striatal slices (Table 2). Under the conditions used, 100 μm ganglioside and 30 min incubation, GM2, GM3, GD1a, GD1b and GT1b were all able to increase pTrk. AsialoGM1, sialic acid, sulfatide (data not shown) and ceramide failed to increase pTrk.

Table 2.  Selectivity of GM1-induced pTrk and Erk2 activity
ConditionspTrk (% of basal ± SEM)Erk2
  1. Striatal slices were incubated with various gangliosides or C2-ceramide, 100 μm, and pTrk or Erk2 activity estimated 30 or 10 min later, respectively. pTrk was estimated as in Table 1. Erk2 activity was estimated after immunoprecipitation with an in vitro kinase assay using MBP, as described in methods. n = 6–11. ap < 0.05 compared with basal.

Basal100 ± 0.4100 ± 0.5
GM1387 ± 61a203 ± 12a
GM2203 ± 37a173 ± 8a
GM3254 ± 33a180 ± 7a
GD1a282 ± 59a170 ± 12a
GD1b278 ± 43a190 ± 9a
GT1b229 ± 36a187 ± 12a
C2-ceramide101 ± 8257 ± 9a
K252a and AG879 prevent the GM1-induced Trk tyrosine phosphorylation

To evaluate the specificity of the GM1-induced Trk phosphorylation, the selective Trk kinase inhibitors K252a (Knüsel and Hefti 1992) and AG879 (Ohmichi et al. 1993) were utilized. When slices were incubated for 30 min with K252a or AG879 and NGF added during the last 5 min of the incubation, the tyrosine phosphorylation of Trk was prevented (Table 3). Likewise, K252a or AG879 added to slices concomitantly with GM1 for 30 min blocked the increase of Trk phosphotyrosine (Tables 3, 3a and 3b). Similar results were obtained when slices prepared from hippocampus or frontal cortex were used (data not shown).

Table 3.  K252a and AG879 prevent the GM1-induced phosphorylation of Trk
ConditionspTrk (% of basal ± SEM)
  1. Striatal slices were incubated with GM1, GM1 + K252a or GM1 + AG879 for 30 min. For studies with NGF, slices were preincubated with K252a or AG879 for 25 min and NGF was added for an additional 5 min pTrk was determined as in Table 1. n = 6. ap < 0.05 compared with basal. bp < 0.05 compared with GM1 or NGF, respectively. (3a and 3b): representative western blots of the above studies.

Basal100 ± 0.4
GM1 (100 µm)286 ± 22a
NGF (100 ng/mL)443 ± 79a
K252a (250 nm)62 ± 6
AG879 (100 µm)78 ± 6
GM1 + K252a108 ± 14b
NGF + K252a136 ± 20b
GM1 + AG87985 ± 5b
NGF + AG87994 ± 9b
GM1 induces TrkA, TrkB, and TrkC tyrosine phosphorylation in brain slices

In the aforementioned studies, immunoprecipitations were performed with an antipan Trk antibody recognizing all three Trk receptors. To determine the type of Trk phosphorylated by GM1, specific antibodies for TrkA, TrkB and TrkC were used for immunoprecipitation or immunoblotting as follows: for TrkA phosphorylation, an antipan Trk antibody was used for immunoprecipitation and a specific antiphosphorylated TrkA antibody was used for immunoblotting; for TrkB and TrkC phosphorylation, anti-TrkB and TrkC-specific antibodies were used for immunoprecipitation and blots were probed with an antiphosphotyrosine antibody. Using this approach a detailed regional distribution of the GM1 effect on the Trks was undertaken. GM1 induced an impressive increase of TrkA phosphotyrosine in striatum (Fig. 2a), and a lesser increase in frontal cortex and hippocampus (data not shown), a pattern displayed by NGF as well (data not shown). The effect of GM1 on TrkA was time-dependent with phosphorylation being observed as early as 5 min after adding GM1, reaching a maximum by 20 min and remaining elevated for over 60 min. The TrkA phosphorylation elicited by GM1 was comparable to that elicited by NGF, and was blocked by the Trk kinase inhibitor K252a (data not shown). GM1 induced a variable increase of TrkB phosphotyrosine which was detected most clearly in frontal cortex (Fig. 2b). Finally, GM1 induced a strong TrkC tyrosine phosphorylation signal in striatum (Fig. 2c), but not in hippocampus or frontal cortex. Under the conditions used, 100 ng/mL and 5 min incubation, BDNF induced a rather small and variable increase of pTrkB in frontal cortex and hippocampus but not in striatum, whereas NT3 induced a rather significant phosphorylation of TrkC in striatum but had no measurable effect on hippocampus (data not shown).

image

Figure 2. GM1 induces tyrosine phosphorylation of TrkA, TrkB and TrkC. Slices were incubated with GM1, 100 μm, for 30 min. pTrkA in striatum (a) was detected with immunoprecipitation with an anti-Trk antibody followed by western blot using an antiphosphorylated TrkA antibody; pTrkB in hippocampus (b) was detected with immunoprecipitation with an anti-TrkB antibody followed by western blot using an antiphosphotyrosine antibody; pTrkC in striatum (c) was detected with immunoprecipitation with an anti-TrkC antibody followed by western blot using an antiphosphotyrosine antibody, as in methods. Representative western blots of each Trk.

Download figure to PowerPoint

GM1 and Erk phosphorylation and activation in brain slices

Neuronal growth and differentiation programs are determined by transcription factors, some of which are controlled by Erk1/2. The best studied Erk signaling cascade involves recruitment of Ras following receptor tyrosine kinase autophosphorylation and tyrosine phosphorylation of adaptor proteins. Activated Ras subsequently engages Raf and MEK1/2 kinases, resulting in phosphorylation/activation of Erk1/2 (Kolch 2000). These studies investigated the ability of GM1 to activate Erk1/2 in brain slices in situ, characterized the response and attempted to decipher the components of the GM1 signaling cascade.

GM1 and NGF induce phosphorylation and activation of Erk1 and Erk2 in brain slices

Incubation of slices prepared from striatum, frontal cortex and hippocampus with GM1 or NGF increased the phosphorylation of Erk1 and Erk2 estimated by western blot using antiactive Erk antibodies (Tables 4 and 4a). In vitro kinase assays performed in stiatal slices demonstrated that the GM1- and NGF-induced phosphorylation of Erk1 and Erk2 was accompanied by increased activity of the kinases using MBP as substrate (Table 4b). In general, the pErk2 response was greater than that of Erk1 in all brain regions, and the effect of GM1 was equivalent to that of NGF; except in hippocampus where NGF elicited a more robust response. For the studies to follow, we employed in vitro kinase assays to characterize the effect of GM1 on the activity of Erk1/2 in striatal slices.

Table 4.  GM1 and NGF increase pErk1 and pErk2
ConditionsStriatumHippocampusFrontal cortex
pErk1pErk2pErk1pErk2pErk1pErk2
  1. Brain slices were incubated with GM1, 10 min, or NGF, 5 min, and pErk1 and pErk2 were determined in lysates by western blot with an antiactive Erk antibody, as described in Methods. n = 6. ap < 0.05 compared with basal. (4a) Representative western blot of a study with striatal slices; (4b) representative image of an in vitro kinase assay performed with striatal slices using MBP, as described in Methods.

Basal100 ± 0.1100 ± 0.5100 ± 2100 ± 0.1100 ± 0.5100 ± 0.2
GM1 (100 μm)161 ± 8a270 ± 17a127 ± 4a214 ± 10a192 ± 20a274 ± 33a
NGF (100 ng/mL)164 ± 13a353 ± 25a121 ± 4a505 ± 35a170 ± 32a330 ± 58a
GM1-induced activation of Erk1/2 is time- and concentration-dependent

Incubation of striatal slices with GM1 at 100 μm for various time periods increased the activity of Erk1 and Erk2. Activation of Erk2 was evident by 2 min reaching a maximum by 10 min, about a twofold increase of basal (Fig. 3a). Activity declined thereafter and returned to basal levels within 1 h. At 10 min, the maximal response of Erk2 was achieved with 100 μm of GM1, but it diminished with higher GM1 concentrations (Fig. 3b). The apparent GM1 EC50 for Erk2 was about 5 μm. Erk1 activity was maximally increased by 5 min (30–50% of basal) with 50 μm of GM1. Like Erk2, the response of Erk1 was attenuated with high concentrations of GM1, and it returned to basal levels within 1 h (data not shown). As Erk1 and Erk2 displayed similar activation profiles after GM1 and the magnitude of Erk2 response was greater than that of Erk1, we focused our studies on Erk2.

image

Figure 3. GM1-induced activation of Erk2 is time- and concentration-dependent. For time–response studies (a), striatal slices were incubated with GM1, 100 μm, for the indicated time periods, and for concentration–response studies (b), slices were incubated with various concentrations of GM1 for 10 min. A representative autoradiograph from time–response studies is presented on (a, top): 1, basal; 2, 2 min; 3, 5 min; 4, 10 min; 5, 20 min; 6, 30 min; 7, 60 min. A representative autoradiograph from concentration–response studies is presented on (b, top): 1, basal; 2, 1 μm; 3, 5 μm; 4, 10 μm; 5, 50 μm; 6, 100 μm; 7, 500 μm; 8, 1000 μm. Erk2 activity was estimated in immunoprecipitates with an in vitro kinase assay using MBP, as in methods. n = 6–10. *p < 0.05 compared with basal.

Download figure to PowerPoint

Other gangliosides induce Erk2 activation

To test the selectivity of GM1 response on Erk2, other gangliosides and related compounds were examined for their ability to induce Erk2 activation (Table 2). Under the conditions used, 100 μm of ganglioside and 10 min of incubation, all the gangliosides tested enhanced Erk2 activity in striatal slices, although the magnitude of response was variable. Ceramide activated Erk2 by more than twofold. In contrast, sialic acid, sulfatide and lactosyl-ceramide were unable to activate Erk2, while asialoGM1 demonstrated some variable effect (data not shown).

Antagonists of Trk, Raf and MEK1/2 kinases prevent the GM1-induced Erk2 activation

To investigate whether the activation of Erk2 following GM1 was related to the observed phosphorylation of Trk, the Trk kinase-selective inhibitors K252a and AG879 were utilized. K252a had a small, variable and statistically non-significant effect on Erk2 activity, but it prevented the GM1- and NGF-induced activation of Erk2 (Tables 5 and 5b). Likewise, AG879 blocked the GM1- and NGF-induced activation of Erk2, but it lowered basal Erk2 activity slightly (Table 5).

Table 5.  GM1- and NGF-induced activation of Erk2: inhibition by protein kinase inhibitors
ConditionsErk2 activity (% of basal ± SEM)
  1. Striatal slices were incubated with K252a, AG879, ZM 336372 or PD 98059 for 20 min and then GM1 or NGF were added for an additional 10 min (GM1) or 5 min (NGF). The incubation time for inhibitors alone was 25 or 30 min, respectively. Erk2 activity was assayed as in Table 2. n = 5–8. ap < 0.05 compared with basal, bp < 0.05 compared with GM1 or NGF, respectively. (5a and 5b): representative images from the above studies.

Basal100 ± 0.5
GM1 (100 µm)209 ± 12a
NGF (100 ng/mL)206 ± 11a
K252a (250 nm)121 ± 9
AG879 (100 µm) 87 ± 7
ZM 336372 (1 µm)100 ± 8
PD 98059 (50 µm) 75 ± 8
GM1 + K2525a127 ± 8b
GM1 + AG879 84 ± 4b
GM1 + ZM 336372 93 ± 6b
GM1 + PD 98059 83 ± 10b
NGF + K252a102 ± 11b
NGF + AG879 86 ± 7b
NGF + ZM 336372 96 ± 10b
NGF + PD 98059 71 ± 7b

In some studies, we attempted to block the GM1-induced phosphorylation of Erk 1/2 with a pan-Trk antibody. In this set of experiments, a 30-min pre-incubation of slices with anti-Trk antibody, 2 μg, caused about a 30% reduction in the GM1- or NGF-induced phosphorylation of Erk1/2 as estimated by western blot (data not shown).

To further elucidate the signaling cascade engaged by GM1 to activate Erk2 in stiatal slices, the Raf inhibitor ZM 336372 (Hall-Jackson et al. 1999), and the MEK1/2 inhibitor PD 98059 (Alessi et al. 1995) were employed. Both inhibitors decreased basal Erk2 activity and blocked the effect of GM1 and NGF on Erk2 (Tables 5 and 5a), indicating a requirement for Raf and MEK1/2 activation.

G protein-coupled receptors and GM1-induced activation of Erk2

In addition to receptor tyrosine kinases, such as Trks, G protein-coupled receptors (GPCRs) have also been shown to be linked to the Erk cascade via multiple but distinct transduction pathways (Gudermann 2001). In most cases stimulation of Erks follows GPCR-mediated activation of Ras with a mechanism similar to that employed by the receptor tyrosine kinases (Luttrell et al. 1999). GPCR-mediated activation of Erks is complex and can occur via Gs, Gi/o or Gq/11-coupled receptors. Erk activation by Gi-coupled receptors is pertussis toxin-sensitive, and Src tyrosine protein kinases serve as proximal signal transducers (Abram and Courtneidge 2000). On the other hand, Gq/11-coupled receptors initiate a pertussis toxin-insensitive process that depends on PKC activation (Luttrell et al. 1999; Gudermann 2001). These studies were designed to explore the possibility that GPCRs, in particular Gi- and Gq/11-coupled, are involved in the GM1-induced phosphorylation of Erks in brain slices. Pre-incubation of striatal slices with pertussis toxin did not prevent the activation of Erk2 by GM1, and similar results were observed after pre-treatment with herbimycin A, a selective inhibitor of Src tyrosine protein kinase (Yang and Chang 1997; Table 6). Furthermore, PKC activity did not increase in striatal slices treated with GM1 (data not shown), and Ro 32–0432, a selective inhibitor of traditional and recently discovered PKC isoenzymes (Birchall et al. 1994), failed to prevent the GM1-induced activation of Erk2 (Table 6). We made similar observations when slices prepared from hippocampus were evaluated (data not shown).

Table 6.  Pertussis toxin, herbimycin A and Ro 32–0432 do not prevent the GM1-induced activation of Erk2
TreatmentErk2 activity (% of basal ± SEM)
  1. Striatal slices were incubated with pertussis toxin, herbimycin A or Ro 32–0432 for 20 min and then GM1 was added for an additional 10 min Erk2 activity was estimated as described in Table 2. n = 5–10. ap < 0.05 compared with basal.

Basal100 ± 3
GM1 (100 μm)228 ± 33a
Pertussis toxin (10 μm)107 ± 4
Ro 32–0432 (1 μm)113 ± 3
Herbimycin A (5 μm)110 ± 6
Pertussis toxin + GM1190 ± 15a
Ro 32–0432 + GM1204 ± 16a
Herbimycin A + GM1248 ± 15a

ICV administration of GM1 increases pTrkA and pErk2

To prove the validity and applicability of our findings from the in situ brain system to the brain in vivo, we administered GM1 ICV to a group of rats. Thirty minutes after the ICV administration of GM1, 10 or 20 nmol, there was a robust increase in pTrkA (Fig. 4a) and pErk1 and pErk2 (Fig. 4b) in striatum, and similar results were obtained in hippocampus as well (data not shown). The observed increase was about 50–60% above vehicle value, and as with the slices, response tended to decline with higher doses of the ganglioside, 100 nmol (data not shown). The GM1-induced phosphorylation was short-lasting and returned to near control values by 90 min (data not shown).

image

Figure 4. Administration of GM1 induces phosphorylation of TrkA and Erk1 and Erk2 in striatum in vivo. GM1 or CSF was administered ICV and pTrkA (a) or pErk1 and pErk2 (b) were estimated by western blot after 30 min, as in methods. Representative western blots.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Accumulated evidence suggests that GM1 ganglioside, like the neurotrophins, has the potential to protect injured and aged central neurons, up-regulate neuronal phenotypic expression, induce neuronal sprouting, and enhance neuronal function and metabolism (Hadjiconstantinou and Neff 1998, 2000). In contrast to peptide neurotrophic factors, GM1 can be administered systemically and has shown promising efficacy in some clinical trials (Nobile-Orazio et al. 1994; Hadjiconstantinou and Neff 2000). GM1 is an interesting, controversial and enigmatic compound, and has the potential to serve as a model molecule for designing neuroprotective and neurorestorative drugs. Although some steps towards understanding its mechanism(s) of action have been made, how GM1 works in the brain has largely been a matter of speculation. Because of commonalities with the neurotrophic action of NGF, interest has been focused on the interaction of GM1 with TrkA. Existing evidence, however, suggests that the profile of neurotrophic action of GM1 resembles more that of BDNF, NT-3 and NT4/5. Like these neurotrophins, GM1 has a broader phenotypic repertoire of trophic action on central neurons implying a possible interaction with TrkB and TrkC also (Hadjiconstantinou and Neff 2000). These studies, for the first time, demonstrate in situ and in vivo induction of neurotrophin receptor Trk activation/phosphorylation by GM1 in the brain, and provide evidence of GM1-initiated signal transduction pathways that involve Erks.

Using brain slices, Knüsel et al. (1994, 1996) reported that NGF was the most potent of the neurotrophins for inducing tyrosine phosphorylation of Trk, in striatum, hippocampus and frontal cortex; in contrast, BDNF and NT3 presented low efficacy. Our observations are in general agreement with those of Knüsel and colleagues. As our studies focused on GM1, we are not able to compare qualitatively or quantitatively the effect of GM1 to that of the endogenous Trk ligands. However, given that in most systems 100 ng/mL of neurotrophin produces a maximal Trk phosphorylation at 5 min, we draw attention to the following observations regarding GM1: (i) The maximal response was seen at 20 min and was comparable with that of NGF, and (ii) The effect on Trk was relatively brief by comparison to that reported for NGF (Knüsel et al. 1994, 1996). Trk phosphorylation decreased progressively after 30 min, and at 2 h was no longer detectable. As the levels of LDH remained stable over 2 h of incubation, we believe that the action of GM1 on Trk in brain slices is relatively rapid and short-lasting. The increase of pTrk after GM1 was prevented by the Trk kinase inhibitors K252a and AG879, implying that the increase in phosphotyrosine resulted from Trk kinase activation. Indeed, increased pTrk reflected enhanced receptor tyrosine kinase activity, as demonstrated by autophosphorylation of the receptor in an in vitro kinase assay. The maximal increase of pTrk was observed by 20 min with 100 μm of GM1, a concentration similar to that inducing Trk phosphorylation in cell cultures (Ferrari et al. 1995; Rabin and Moccheti 1995). The concentration response curve was bell-shaped, with response attenuation at high GM1 concentrations. This might indicate two interacting receptor sites, or alternatively, high concentrations of GM1 in the incubation mixture could promote the formation of micelles and other aggregates (Saqr et al. 1993) that interfere with the interaction of GM1 with Trk. The tyrosine phosphorylating effect of GM1 was also extended to TrkB and TrkC. In contrast to the consistent and robust phopshorylation of Trk A, the TrkB and TrkC response to GM1 appeared to be variable and region-dependent, perhaps reflecting differential neurotrophin receptor expression, localization and responsiveness.

Taken together, our findings show that exposure of brain slices to GM1 results in tyrosine phosphorylation of all three tyrosine kinase receptors for neurotrophins. The provided evidence is particularly strong for phosphorylation of TrkA by GM1, as has been indirectly suggested from in vivo experiments (Cuello et al. 1989; Fong et al. 1995; for review see Hadjiconstantinou and Neff 1998, 2000). The importance of our findings lies in the fact that for the first time they demonstrate an early induction of pTrk in brain tissue by GM1 alone. The validity of our conclusions from the in situ studies is strengthened by our original observation that ICV administration of GM1 induces a rapid and short-lasting tyrosine phosphorylation of TrkA in brain in vivo. Our in situ and in vivo findings point to a drug/receptor-like interaction between exogenous GM1 and Trk, suggesting a pharmacological action. In support of this interpretation we were not able to detect increased GM1 content in Trk immunoprecipitates under the conditions used to induce phosphorylation of Trk (data not shown). The notion that exogenous GM1 might have pharmacological properties does not contradict the long standing belief that incorporation of GM1 into membrane is a prerequisite for biological action (Saqr et al. 1993). The two mechanisms, pharmacological and biological, could complement each other and work in unison toward achieving and maintaining an effect. For example, a pharmacological action could initiate a rapid activation of early signaling cascades, whereas incorporation and accumulation of the ganglioside in the plasma membrane over time might be important for ensuring long-term activation of initial signaling cascades and/or recruiting new pathways with the same final destination and outcome.

How GM1 interacts with Trk is unknown. Studies in cell lines have mainly investigated the GM1/NGF synergism in phosphorylating and activating TrkA. So far, direct binding of GM1 to NGF or alteration of [125I] NGF binding by GM1 have been ruled out (Ferrari et al. 1983; Farooqui et al. 1997) and Mutoh et al. (1995) have proposed that in PC12 cells, GM1 enhances the NGF-elicited Trk kinase activity via an association with the Trk protein. Our in situ and in vivo findings, however, suggest that exogenous GM1 alone, directly or indirectly, can induce Trk kinase activation and autophosphorylation in brain tissue. Based on reports that GM1 increases the content of NGF in the brain (Duchemin et al. 1997) and elevates intracellular calcium in neurons (Hilbush and Levine 1992), the possibility that the effect of GM1 on Trk phosphorylation is indirect, through pre- or post-synaptically released neurotrophins (Blöchl and Thoenen 1996; Goodman et al. 1996) was investigated. Under the experimental conditions used in our studies, GM1 failed to release NGF (data not shown), ruling out that the phosphorylation of TrkA is due to enhanced release of its endogenous ligand. Taking advantage of recent advances of knowledge, three possible scenarios for GM1/Trk interaction can be put forward for consideration:

  • i. GM1 might mediate its effects on neurotrophin tyrosine kinases through interaction inside the membrane itself, especially in the glycolipid-enriched domains, caveolae, where both GM1 (Parton 1994) and tyrosine kinase receptors, including Trk (Wu et al. 1997; Peiróet al. 2000) are present.

  • ii. GM1 could directly interact with the extracellular portion of Trk, whose leucine-rich motifs are thought to bind neurotrophins (Windisch et al. 1995a, 1995b) and immunoglobulin-like domains to confer neurotrophin binding affinity and specificity (Pérez et al. 1995; Urfer et al. 1995). In lymphocytes, GM1 binds to and regulates CD4, a molecule containing immunoglobulin-like domains (Saggioro et al. 1993), and other gangliosides bind to proteins bearing extracellular immunoglobulin-like domains, such as myelin-associated glycoprotein and other members of the sialoadhesin family (Schnaar et al. 1998).

  • iii. GM1 via a yet to be established cell surface receptor could induce a ligand-independent activation (transactivation) of Trk (Lee and Chao 2001; for review see Luttrell et al. 1999; Guderman 2001).

Erks are emerging as important regulators of neuronal function. In addition to their well established role in regulating cell growth, proliferation and differentiation (Schaeffer and Weber 1999), this family of MAP kinases is an important player for activity-dependent processes, such as synaptic plasticity, long-term potentiation and cell survival (Grewal et al. 1999). Erk1 and Erk2 are activated by a diverse array of ligands, through tyrosine kinase receptors, GPCRs and calcium-dependent pathways, and are among the protein kinases most commonly used in signal transduction (English et al. 1999). In neurons, the Ras-MEK1/2-Erk1/2 cascade has been demonstrated to be both necessary and sufficient for NGF-induced differentiation (English et al. 1999; Klesse and Parada 1999), whereas its role for neurotrophin induced-survival is more controversial (Kaplan and Miller 2000). GM1 induced a rapid and transient activation of Erk1/2 in slices prepared from the striatum, hippocampus and frontal cortex of rat brain. As with pTrk, a maximal response was observed with 100 μm GM1, and activity diminished with high concentrations of the ganglioside. The observation that GM1-induced phosphorylation of Trk and Erk in brain slices raises the possibility of an intimate relationship between these two events. Phosphorylation of TrkA by GM1 has been linked to activation of Erks in C6-glioma cells expressing the neurotrophin receptor (Rabin and Mocchetti 1995). In our studies, a number of direct and indirect lines of evidence support a link between Trk phosphorylation and Erk activation by GM1 in brain slices: (i) the selective Trk kinase inhibitors K252a and AG879 reversed the GM1-induced activation of Erk2; (ii) pre-incubation with pan-Trk antibodies partially blocked both the GM1- and NGF-induced activation of Erk2; (iii) blockade of Gi-coupled receptors with pertussis toxin or inhibition of the intermediate Src kinase did not prevent the GM1 effect; (iv) PKC, a mediator of Gq/G11-coupled pertussis-toxin insensitive activation of Erk2, was not involved in the GM1 response; (v) notably, ICV administration of GM1 also induced phosphorylation of Erk1/2 in vivo in a pattern similar to that of Trk validating our in situ studies and providing additional evidence for a role of the signaling molecule for the neurotrophic action of the ganglioside and for an association with Trk activation/phosphorylation.

The specificity profile of the GM1 effect on Trk and Erk support a model for study whereby GM1-induced phosphorylation of Trk initiates a signaling cascade(s) that results in activation of Erk1/2 via Raf and MEK1/2. The intermediate steps and the targets of this putative signaling pathway are under investigation. That the maximal activation of Erks by GM1 in situ precedes that of Trk does not contradict this model, as both molecules might be under different phosphorylation/dephosphorylation regulation; GM1-sustained phosphorylation of Trk might be needed for transducing other signals, such as PI3-kinase (Hadjiconstantinou et al. 2000); and the net result of Erk activation by GM1 might be the sum of the ganglioside-effect on various Trks and on different neural cells. Finally, the in situ brain system and the different methodologies used to evaluate Trk phosphorylation and Erk activation might contribute to the temporal discrepancy. In PC12 cells, NGF induces both transient and sustained activation of Erks, with the latter thought to be important for cell differentiation (Marshall 1995). Accordingly, differentiated mature brain neurons would be expected to respond with transient activation of Erks. Indeed, in the brain slices, NGF induced a transient pErk2 increase that peaked by 5–10 min and lasted for about 60 min (data not shown).

The tyrosine phosphorylation of Trk and the activation of Erk2 are not an exclusive property of GM1 ganglioside. Other major gangliosides present in brain, GD1a, GD1b, GT1b as well as GM2 and GM3 displayed similar action when added to brain slices. This is not a surprising finding as, in addition to GM1, other gangliosides display neurotrophic action as well. A GD1a, GD1b, GT1b and GT1 mix enhance neuritogenesis in cell cultures and promote regenerative and behavioral responses when administered to animal models of central or peripheral neuronal injury (Schengrund 1990); GT1b potentiates the NGF-induced neurite extension in PC12 cells and promotes long-term survival after serum deprivation (Ferrari et al. 1983, 1995); and GM3 increases dopamine and GABA uptake in embryonic mesencephalic cultures (Leon et al. 1988). Our data and the literature suggest that the whole ganglioside molecule is required for Trk or Erk phosphorylation/activation. Our observation that ceramide does not affect Trk phosphorylation in brain tissue questions the suggestion of MacPhee and Baker (1999), that the GM1-dependent enhancement of TrkA activity may be due to liberated ceramide. Van Brocklyn et al. (1997) reported both GM1- and ceramide-mediated activation of Erk2 in glioma cells, but the authors concluded that the contribution of the ceramide to the GM1 effect was at most minimal. Presently, we do not know whether there is a link between the Trk and Erk activation by the various gangliosides in our system. Fukumoto et al. (2000), reported that GD3 synthase gene expression in PC12 cells results in continuous activation of TrkA and Erk1/2, probably due to the increased production of GD1b and GT1b. Furthermore, we do not know whether all gangliosides act at the same site and utilize the same mechanism(s) for initiating common signaling pathways. Recent advances in our understanding of the organization of the cell membranes have provided useful clues for future inquiry in this direction. Of particular interest is the emergence of the sphingolipid-enriched cell membrane domains, such as caveolae and glycosphinglolipid-signaling domains, as unique membrane compartments with biological function. They are thought to provide a spatial environment where various protein and lipid molecules involved in signal transduction interact, and they have been shown to partake in signal transduction upon glycosphingolipid-mediated stimulation (Masserini et al. 1999; Hakomori 2000).

In conclusion, our studies for the first time demonstrate that GM1 induces tyrosine phosphorylation of TrkA, TrkB and TrkC high-affinity receptors for neurotrophins, and initiates signal transduction resulting in activation of the MEK1/2/Erk1/2 pathway in brain slices. More importantly, our studies provide evidence for in vivo tyrosine phosphorylation of TrkA and Erk1/2 in brain tissue after GM1 ICV administration. The prevention of the GM1 effect on Trks and Erks by selective Trk kinase inhibitors suggests a possible link between the two events. This is supported further by the exclusion of Gi-or Gq/11-coupled receptors as possible sites for the GM1 effect. In neurons, the MEK1/2/Erk1/2 pathway coordinates complex cellular responses such as differentiation, survival, synaptic plasticity, long-term potentiation, and learning and memory (Grewal et al. 1999; Sweatt 2001). Usage of the MEK1/2/Erk1/2 pathway by GM1 might explain the diverse actions of the ganglioside on immature and mature neurons. Finally, the observed tyrosine phosphorylation of Trks in brain tissue after GM1, in situ and in vivo, confirms the long-standing speculation that GM1 acts, in part, as a neurotrophin mimetic.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported, in part, by NIA Grant AG10530, and a grant by AFAR – OA. The expert technical assistance of Ms. Amy E. Colvin is greatly appreciated.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Abram C. L. and Courtneidge S. A. (2000) Src family tyrosine kinases and growth factor signaling. Exp. Cell Res. 254, 113.DOI: 10.1006/excr.1999.4732
  • Alessi D. R., Cuenda A., Cohen P., Dudley D. T. and Saltiel A. R. (1995) PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J. Biol. Chem. 270, 2748927494.
  • Birchall A. M., Bishop J., Bradshaw D., Cline A., Coffey J., Elliot L. H., Gibson V. M., Greenham A., Hallam T. J., Harris W., Hill C. H., Hutchings A., Lamont A. G., Lawton G., Lewis E. J., Maw A., Nixon J. S., Pole D., Wadsworth J. and Wilkinson S. E. (1994) Ro 32–0432, a selective and orally active inhibitor of protein kinase C prevents T-cell activation. J. Pharmacol. Exp. Ther. 268, 922929.
  • Blöchl A. and Thoenen H. (1996) Localization of cellular storage compartments and sites of constitutive and activity-dependent release of nerve growth factor (NGF) in primary cultures of hippocampal neurons. Mol. Cell. Neurosci. 7, 173190.
  • Cuello A. C., Garofalo L., Kenigsberg R. L. and Maysinger D. (1989) Gangliosides potentiate in vivo and in vitro effects of nerve growth factor on central cholinergic neurons. Proc. Natl. Acad. Sci. USA 86, 20562060.
  • Duchemin A. M., Neff N. H. and Hadjiconstantinou M. (1997) GM1 increases the content and mRNA of NGF in the brain of aged rats. Neuroreport 8, 38233827.
  • English J., Pearson G., Wilsbacher J., Swantek J., Karandikar M., Xu S. and Cobb M. H. (1999) New insights into the control of MAP kinase pathways. Exp. Cell Res. 253, 255270.DOI: 10.1006/excr.1999.4687
  • Farooqui T., Franklin T., Pearl D. K. and Yates A. J. (1997) Ganglioside GM1 enhances induction by nerve growth factor of a putative dimer of Trk A. J. Neurochem. 68, 23482355.
  • Ferrari G., Anderson B. L., Stephens R. M., Kaplan D. R. and Greene L. A. (1995) Prevention of apoptotic neuronal death by GM1 ganglioside. Involvement of Trk neurotrophin receptors. J. Biol. Chem. 270, 30743080.
  • Ferrari G., Fabris M. and Gorio A. (1983) Gangliosides enhance neurite outgrowth in PC12 cells. Brain Res. 8, 215221.
  • Fong T. G., Neff N. H. and Hadjiconstantinou M. (1995) GM1 and NGF synergism on choline acetyltransferase and choline uptake in aged brain. Neurobiol. Aging 16, 917923.
  • Fukumoto S., Mutoh T., Hasewaga T., Miyazaki H., Okada M., Goto G., Furukawa K. and Urano T. (2000) GD3 synthase gene expression in PC12 cells results in the continuous activation of TrkA and Erk1/2 and enhanced proliferation. J. Biol. Chem. 275, 58325838.
  • Goettl V. M., Wemlinger T. A., Duchemin A. M., Neff N. H. and Hadjiconstantinou M. (1999) GM1 ganglioside restores dopaminergic neurochemical and morphological markers in aged rats. Neuroscience 92, 9911000.
  • Goodman L. J., Valverde J., Lim F., Geschwind M. D., Federoff H. J., Geller A. I. and Hefti F. (1996) Regulated release and polarized localization of brain-derived neurotrophic factor in hippocampal neurons. Mol. Cell. Neurosci. 7, 222238.DOI: 10.1006/mcne.1996.0017
  • Grewal S. S., York R. D. and Stork P. J. S. (1999) Extracellular-signal-regulated kinase signalling in neurons. Curr. Opin. Neurobiol. 9, 544553.
  • Gudermann T. (2001) Multiple pathways of ERK activation by G protein-coupled receptors. In: Complexity in Biological Information Processing (Bock, G. and Goode, J., eds), pp. 6884. John Wiley, Winchester.
  • Hadjiconstantinou M. and Neff N. H. (1988) Treatment with GM1 ganglioside restores striatal dopamine in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated mouse. J. Neurochem. 51, 11901196.
  • Hadjiconstantinou M. and Neff N. H. (1998) GM1 ganglioside: in vivo and in vitro trophic actions on central neurotransmitter systems. J. Neurochem. 70, 133513445.
  • Hadjiconstantinou M. and Neff N. H. (2000) GM1 ganglioside: Commonalities of action with neurotrophins. In: Neurobiology of the Neurotrophins (Mocchetti, I., ed.), pp. 427453. FP Graham Publishing Co, Johnson City, TN.
  • Hadjiconstantinou M., Karadesheh N. S., Rattan A. K., Tejwani G. A., Fitkin J. G. and Neff N. H. (1992) GM1 ganglioside enhances cholinergic parameters in the brain of senescent rats. Neuroscience 46, 681686.
  • Hadjiconstantinou M., Ren Q., Duchemin A.-M. and Neff N. H. (2000) Activation of brain PI3-kinase by GM1 ganglioside. J. Neurochem. 74, S37.
  • Hadjiconstantinou M., Yates A. J. and Neff N. H. (1990) Hypoxia-induced neurotransmitter deficits in neonatal rats are partially corrected by exogenous GM1 ganglioside. J. Neurochem. 55, 864869.
  • Hakomori S. (2000) Cell adhesion/recognition and signal transduction through glycosphingolipid microdomain. Glycoconj. J. 17, 143151.
  • Hall-Jackson C. A., Eyers P. A., Cohen P., Goedert M., Boyle F. T., Hewitt N., Plant H. and Hedge P. (1999) Paradoxical activation of Raf by a novel Raf inhibitor. Chem. Biol. 6, 559568.
  • Hilbush B. S. and Levine J. M. (1992) Modulation of Ca2+ signaling pathway by GM1 ganglioside in PC12 cells. J. Biol. Chem. 34, 2478924795.
  • Kaplan D. R. and Miller F. D. (2000) Neurotrophin signal transduction in the nervous system. Curr. Opin. Neurobiol. 10, 381391.
  • Kaplan D. R. and Stephens R. M. (1994) Neurotrophin signal transduction by the Trk receptor. J. Neurobiol. 25, 14041417.
  • Klesse L. J. and Parada L. F. (1999) Trks: signal transduction and intracellular pathways. Micr. Res. Techn. 45, 210216.
  • Knüsel B. and Hefti F. (1992) K-252 compounds: modulators of neuro- trophin signal transduction. J. Neurochem. 59, 19871996.
  • Knüsel B., Kaplan D. R. and Hefti F. (1996) Intraparenchymal NGF injections in adult and aged rats induce long-lasting Trk tyrosine phosphorylation. Experim. Neurol. 139, 121130.
  • Knüsel B., Rabin S. J., Hefti F. and Kaplan D. R. (1994) Regulated neurotrophin receptor responsiveness during neuronal migration and early differentiation. J. Neurosci. 14, 15421554.
  • Kolch W. (2000) Meaningful relationships: the regulation of the Ras/Raf/MEK/ERK pathway by protein interactions. Biochem. J. 351, 289305.DOI: 10.1042/0264-6021:3510289
  • Lee F. S. and Chao M. V. (2001) Activation of Trk neurotrophin receptors in the absence of neurotrophins. Proc. Natl. Acad. Sci. USA 98, 35553560.
  • Leon A., Toso R. D., Tresti D., Benvegnu D., Facci L., Krishner G., Tettamanti G. and Toffano G. (1988) Development and survival of neurons in dissociated fetal mesencephalic serum-free cultures. II. Modulatory effects of gangliosides. J. Neurosci. 8, 746753.
  • Luttrell L. M., Daaka Y. and Lefkowitz R. L. (1999) Regulation of tyrosine kinase cascades by G protein-coupled receptors. Curr. Opin. Cell Biol. 11, 177183.
  • MacPhee I. and Baker P. A. (1999) Extended ceramide exposure activates the trkA receptor by increasing receptor homodimer formation. J. Neurochem. 72, 14231430.DOI: 10.1046/j.1471-4159.1999.721423.x
  • Marshall C. J. (1995) Specificity of receptor tyrosine kinase signalling: transient versus sustained extracellular signal-regulated kinase activation. Cell 80, 179185.
  • Masserini M., Palestini P. and Pitto M. (1999) Glycolipid-enriched caveolae and caveolae-like domains in the nervous system. J. Neurochem. 73, 111.DOI: 10.1046/j.1471-4159.1999.0730001.x
  • Mutoh T., Tokuda A., Miyadai T., Hamaguchi M. and Fujiki N. (1995) Ganglioside GM1 binds to the Trk protein and regulates receptor function. Proc. Natl. Acad. Sci. USA 92, 50875091.
  • Nobile-Orazio E., Carpo M. and Scarlato G. (1994) Gangliosides: Their role in clinical neurology. Drugs 47, 576585.
  • Ohmichi M., Pang L., Ribon V., Gazit A., Levitzki A. and Saltiel A. R. (1993) Tyrosine kinase inhibitor tyrphostin blocks the cellular actions of nerve growth factor. Biochemistry 32, 46504658.
  • Parton R. G. (1994) Ultrastructural localization of gangliosides; GM1 is concentrated in caveolae. J. Histochem. Cytochem. 42, 155166.
  • Peiró S., Comella J. X., Enrich C., Martin-Zanca D. and Rocamora N. (2000) PC12 cells have caveolae that contain TrkA. J. Biol. Chem. 275, 3794637852.
  • Pérez P., Coll P. M., Hempstead B. L., Martin-Zanca D. and Chao M. V. (1995) NGF binding to the trk tyrosine kinase receptor requires the extracellular immunoglobulin-like domains. Mol. Cell. Neurosci. 6, 97105.
  • Rabin S. J. and Mocchetti I. (1995) GM1 ganglioside activates the high-affinity nerve growth factor receptor trkA. J. Neurochem. 65, 347354.
  • Saggioro D., Sorio C., Calderazzo F., Callegaro L., Panozzo M., Berton G. and Chieco-Binachi L. (1993) Mechanism of action of monosialoganglioside GM1 as a modulator of CD4 expression. Evidence that GM1–CD4 interaction triggers dissociation of p56lck from CD4 and CD4 internalization and degradation. J. Biol. Chem. 268, 13681375.
  • Saqr H. E., Pearl D. K. and Yates A. J. (1993) A review and predictive models of ganglioside uptake by biological membranes. J. Neuro- chem. 61, 395411.
  • Schaeffer H. J. and Weber M. J. (1999) Mitogen-activated protein kinases; specific messages from ubiquitous messengers. Mol. Cell. Biol. 19, 24352444.
  • Schengrund C.-L. (1990) The role(s) of gangliosides in neural differentiation and repair: a perspective. Brain Res. Bull. 24, 131141.
  • Schnaar R. L., Collins B. E., Wright L. P., Kiso M., Tropak M. B., Roder J. C. and Crocker P. R. (1998) Myelin-associated glycoprotein binding to gangliosides. Structural specificity and functional implications. Ann. NY Acad. Sci. 845, 92105.
  • Sweatt J. D. (2001) The neuronal MAP kinase cascade: a biochemical signal integration system subserving synaptic plasticity and memory. J. Neurochem. 76, 110.
  • Urfer R., Tsouflas P., O'Connel L., Shelton D. L., Rarada L. F. and Presta L. G. (1995) An immunoglobulin like domain determines the specificity of neurotrophin receptors. EMBO J. 15, 27952805.
  • Van Brocklyn J. R., Vandenheede J. R., Fertel R., Yates A. J. and Rampersaud A. A. (1997) Ganglioside GM1 activates the mitogen-activated protein kinase Erk2 and p70, S6 kinase in U-1242 MG human glioma cells. J. Neurochem. 69, 116125.
  • Windisch J. M., Marksteiner R., Lang M. E., Auer B. and Schneider R. (1995a) Brain-derived neurotrophic factor, neurotrophin-3, and neurotrophin-4 bind to a single leucine-rich motif of TrkB. Biochemistry 34, 1125611263.
  • Windisch J. M., Marksteiner R. and Schneider R. (1995b) Nerve growth factor binding site on TrkA mapped to a single 24-amino acid leucine-rich motif. J. Biol. Chem. 47, 2813328138.
  • Wu C., Butz S., Ying Y.-S. and Anderson G. W. (1997) Tyrosine kinase receptors concentrated in caveolae-like domains from neuronal plasma membrane. J. Biol. Chem. 272, 35543559.
  • Yang Y.-W. and Chang Y.-H. (1997) Inhibition of pp60c–src protein kinase by herbimycin A in polyomavirus middle tumor antigen-transformed cells. Anticancer Res. 17, 32373280.
  • Zeller C. B. and Marchase R. (1992) Gangliosides as modulators of cell function. Am. J. Physiol. 262 (Cell Physiol.) 31, C1341C1355.