Gas1 reduces Ret tyrosine 1062 phosphorylation and alters GDNF-mediated intracellular signaling

www.elsevier.com/locate/ijdevneu

26 (2008) 497–503

Gas1 reduces Ret tyrosine 1062 phosphorylation and alters

GDNF-mediated intracellular signaling

Miguel A. Lopez-Ramırez a,1, Gabriela Domınguez-Monzon b,1,Paula Vergara a, Jose Segovia a,*

a Departamento de Fisiologıa, Biofısica y Neurociencias, Centro de Investigacion y de Estudios Avanzados del IPN,

Avenida IPN # 2508, Mexico 07360, D.F., Mexicob Seccion Externa de Farmacologıa, Centro de Investigacion y de Estudios Avanzados del IPN, Mexico 07360, D.F., Mexico

Received 22 November 2007; received in revised form 18 February 2008; accepted 18 February 2008

Abstract

The present results show that the expression of Growth Arrest Specific1 (Gas1) in SH-SY5Y neuroblastoma cells significantly inhibits the

increased phosphorylation of tyrosine 1062 of the Ret receptor tyrosine kinase induced by glial-cell-line-derived neurotrophic factor (GDNF). We

also observed that Gas1 significantly reduces the activation of Akt. GDNF and members of its family of ligands (GFLs), signal through a molecular

complex consisting of one of its receptors (GFRas) and the Ret receptor tyrosine kinase. GDNF is a key component to preserve several cell

populations in the nervous system, including dopaminergic and motor neurons, and also participates in the survival and differentiation of peripheral

neurons such as enteric, sympathetic and parasympathetic. On the other hand, Gas1 is a molecule involved in cell arrest that can induce apoptosis

when over-expressed in different cell lines, including cells of neuronal and glial origin. Although, Gas1 is widely expressed during development, its

role in vivo has not yet been clearly defined. We recently showed the structural homology between Gas1 and GFRas, thus suggesting that the

physiological role of Gas1 is that of modulating the biological responses induced by GDNF and/or other members of this family of signaling

molecules. The results of this work are consistent with the hypothesis of Gas1 acting as a negative modulator of GDNF signaling.

# 2008 ISDN. Published by Elsevier Ltd. All rights reserved.

Keywords: Akt; Neurotrophic factors; Intracellular signaling; GDNF receptor

Gas1 (Growth Arrest Specific1) was isolated by differential

screening of NIH-3T3 fibroblasts after serum withdrawal

(Schneider et al., 1988), and it is a gene that codes for a protein

linked to the cell membrane through a glycosyl-phosphatidy-

linositol (GPI) anchor at its carboxy terminus (Stebel et al.,

2000). Gas1 has been identified as a tumor suppressor (Del Sal

et al., 1992; Evdokiou and Cowled, 1998), and it induces cell

death and apoptosis (Mellstrom et al., 2002; Zamorano et al.,

2003, 2004; Benitez et al., 2007). During embryogenesis Gas1

is differentially expressed, and its expression has been

Abbreviations: Dlk1, delta-like1; Gas1, Growth Arrest Specific1; GDNF,

glial-cell-line-derived neurotrophic factor; GFLs, GDNF family ligands;

GFRas, GDNF family receptors; GPI, glycosyl-phosphatidylinositol; NCAM,

neural cell adhesion molecule; Ret, rearranged during transformation; Shh,

sonic hedgehog; TGFß, transforming growth factor ß.

* Corresponding author. Tel.: +52 55 5061 3958; fax: +52 55 5061 3754.

E-mail address: [email protected] (J. Segovia).1 These two authors contributed equally to this work.

0736-5748/$34.00 # 2008 ISDN. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.ijdevneu.2008.02.006

associated with cell death during development (Lee et al.,

2001b). On the other hand, in cerebellum Gas1 acts as a positive

growth regulator (Liu et al., 2001). Exposure to growth factors

or morphogens results in important changes in Gas1 expression

(Spagnuolo et al., 2004; Lee et al., 2001a). Nevertheless, the

molecular role of Gas1 in vivo remains elusive.

To explain the molecular mechanism of action of Gas1

different proposals, including the necessary presence of p53 for

Gas1 to arrest cell growth (Del Sal et al., 1995), its interaction

with delta-like1 (Dlk1) (Baladron et al., 2002), and its capacity

antagonizing sonic hedgehog (Shh) activity (Lee et al., 2001a)

have been presented. Interestingly, new data using Gas1

knockout mice indicate that Gas1 facilitates Shh activity

(Martinelli and Fan, 2007; Allen et al., 2007; Seppala et al.,

2007). Recently, we showed (Schueler-Furman et al., 2006),

and soon afterward another group reported (Cabrera et al.,

2006), that Gas1 exhibits significant structural homology with

glial-cell-line-derived neurotrophic factor (GDNF) receptors

(GFRas). Based on this data, we proposed that Gas1 interferes

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503498

with GDNF-mediated intracellular signaling cascades. The

functional similarity between GFRas and Gas1 extends to their

role in embryogenesis, differentiation and glia maintenance,

and it is substantiated by overlap in their expression profile

(Choi-Lundberg and Bohn, 1995), subcellular localization and

structural details. Based on this information, we have suggested

that the relative expression and localization of the two types of

molecules, GFRas and Gas1, on the membranes of neuronal

and glial cells determines whether these cells survive or

undergo apoptotic death (Schueler-Furman et al., 2006). The

canonical mechanism of action of GDNF, and the other

members of its family (neurturin, artemin and persephin), is

based on their binding to GFRas, and in the autopho-

sphorylation of the Ret receptor tyrosine kinase, which in turn

activates different intracellular signaling pathways, including

Akt (Airaksinen and Saarma, 2002). However, there is a report

indicating that Gas1 is capable of modifying GDNF-mediated

signaling, but that Gas1 does not affect the autophosphorylation

of Ret induced by its interaction with the GDNF–GFRa1

complex (Cabrera et al., 2006). In the present paper, we show

that Gas1 alters the signaling pathways mediated by Ret,

affecting its pattern of autophosporylation.

1. Experimental procedure

1.1. Cell culture and GDNF treatments

SH-SY5Y human neuroblastoma cells were maintained at 37 8C in 95% air,

5% CO2 in Advanced DMEM (GIBCO BRL) supplemented with 10% fetal calf

serum (GIBCO BRL), 100 IU penicillin-streptomycin (GIBCO BRL) and

200 mM glutamine (Sigma), and seeded at 70% confluence. For the present

experiments, complete medium was removed and replaced with medium

without serum and antibiotics. For the different experiments cells were main-

tained in this condition for 3 or 24 h. Cells were treated with 100 ng/ml human

recombinant GDNF (Cell Sciences Inc.) for 15 min.

1.2. mRNA determination

Total RNA was isolated using the Tripure reagent (Roche Diagnostics),

cDNA was obtained by reverse transcriptase (Invitrogen) using polydt as a

primer, and Taq polymerase (Invitrogene) was used for the PCRs. To determine

ret expression, the following primers were employed: 50-CTGGTGAGGCGG-

TACACAA-30 and 50-CAGGTGGAGAAGTTCCTGGTC-30; to determine

gfra1 expression, the primers used were: 50-CTGCAGCACCAAGTACC-

GCA-30 and 50-ACCTTGACTCTGGCTGGCAGT-30. Reactions were per-

formed as follows: an initial step at 94 8C for 2 min was followed by 40

cycles (30 s at 94 8C, 30 s at 58 8C and 35 s at 72 8C), and finally a 5 min

extension step at 72 8C. Gas1 expression was determined as previously

described (Zamorano et al., 2004), and for human ß-actin, the following

primers: 50-TGGCACCACACCTTCTACA-30 and 50-TCACGCACGATT-

TCCC-30, were used. After PCR, products were electrophoresed on agarose

gels, stained with ethidium bromide, and images recorded using a BioDoc-It

System (UVP).

1.3. Immunocytochemistry

Cells were grown on glass slides and fixed for 10 min at room temperature

with 4% paraformaldehyde. Slides were blocked with 1% BSA (Sigma) for

20 min at room temperature, and incubated in the presence of a Gas1 polyclonal

antibody (1:50) at 4 8C overnight. The Gas1 antibody recognizes the Gas1

sequence: DDGVPHPPRPGSGA, and was produced by ProSci Inc (San Diego,

CA) (Benitez et al., 2007). Cells were washed with PBS, and incubated at room

temperature for 1 h with a secondary fluorescent antibody (1:35) (Jackson

ImmnunoResearch Laboratories) and washed. To ascertain the expression of

phosphorylated Ret Y 1062, cells were grown at different conditions (absence of

serum for 3 or 24 h, and in the presence or absence of added human GDNF),

fixed, treated as previously described, and permeabilized with 0.2% Triton X-

100 (Sigma). Cells were incubated with a polyclonal antibody that recognizes

Ret phosphorylated in Y 1062, diluted 1:75 (cat, sc-20252R; Santa Cruz

Biotechnology), slides were then incubated with a secondary biotinylated

anti-rabbit (1:200) antibody (Vector Laboratories), and revealed with Texas

Red streptavidin (Vector Laboratories). Slides were observed in an Olympus

BX51 fluorescence microscope, and images captured using the Image Pro

Acquisition and Analysis Software (UVP). Cells were counterstained with

DAPI to reveal the nuclei.

1.4. Western blot assays

Total protein from SH-SY5Y cells at the different growth conditions, and

both after GDNF treatment, or in the absence of added GDNF was obtained

by scraping the cells from the dishes in a lysis solution, containing a

proteases inhibitor cocktail (Complete, Roche Diagnostics). Cells were

sonicated, centrifuged, resuspended and protein amount was determined

using the BCA assay (Pierce). Proteins were diluted in a denaturing solution,

and boiled for 3 min. Fifty micrograms of proteins were run per lane on 7%

SDS-PAGE gels and transferred onto PVDF membranes (BioRad). Mem-

branes were washed with PBS, blocked with 5% non-fat milk/TBST (0.05%

Tween-20, TBS) for 1 h, and incubated in the presence of the following

polyclonal antibodies: against Ret phosphorylated Y 1062 (1:700; cat sc-

20252R); total Ret (1:700, cat sc-13104); total Akt (1:5000, cat sc-8312),

and phosphorylated Akt (1:5000, cat sc-7985) all from Santa Cruz Bio-

technology; the antibody against Gas1 previously described (1:200, ProSci);

and a monoclonal antibody against ß-actin (1:500) (Garcia-Tovar et al.,

2001). These antibodies were developed using secondary peroxidase-

coupled anti-rabbit and anti-mouse antibodies, from Zymed. Proteins were

revealed using the enhanced chemiluminescence (ECL) detection system

(PerkinElmer). As a control for protein loading and transfer, membranes

were stripped, incubated with the monoclonal antibody against actin and

revealed as described. Images from films were digitally acquired and

analyzed with a ChemiDoc System (BioRad) and densitometry analysis

performed using the Quantity One Software.

1.5. Flow cytometry

SH-SY5Y cells were harvested using Trypsin-EDTA (0.25% GIBCO BRL),

centrifuged and washed twice with PBS, pH 7.4. In some cases, a pulse of

GDNF was applied, and others were used as controls. After treatment cells were

washed and fixed with 2% paraformaldehyde for 15 min at room temperature,

and permeabilized with 0.2% Triton X-100 (Sigma) for 25 min at room

temperature. Cells were blocked with 1% BSA (Sigma), and then incubated

with the polyclonal antibody against phosphorylated Y 1062 of Ret, previously

described (1:75), overnight at 4 8C. As a secondary antibody a goat anti-rabbit

IgG coupled with rhodamine lissamine was employed (1:50 Jackson Immu-

noResearch Laboratories), and incubated for 1 h at room temperature. Cells

were finally resuspended in 300 ml of PBS and 10,000 cells were counted per

experiment. A FACSort Becton Dickinson instrument was employed, and data

analyzed using the Cell Quest program.

1.6. Gas1 mRNA silencing

Cells were grown as previously described in the absence of serum, and for

this set of experiments in the presence of 1.4 mg of small interfering RNA

against human gas1 mRNA (siRNA-Gas1; cat sc-37435) or 1.0 mg of a negative

control siRNA (siRNA-C; cat sc-37007), which consists of a set of scrambled

sequences that will not lead to the degradation of any known cellular mRNA.

Both siRNAs were obtained from Santa Cruz Biotechnology and used according

to the manufacturer’s instructions. After 24 h in the absence of serum cells were

harvested, total protein obtained and Gas1 and Ret Y 1062 expression deter-

mined by western blot analysis as previously described.

Fig. 1. Expression of ret, gfra1 and gas1 mRNAs determined by RT-PCR in

SH-SY5Y cells. Reactions were performed 3 and 24 h after serum withdrawal.

Actin was determined as a positive control, and (�) is a negative control reaction

without DNA.

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503 499

2. Results

2.1. Expression of gas1, gfra1 and ret in SH-SY5Y cells

In order to determine the effects of Gas1 on GDNF-mediated

signaling it was necessary to first demonstrate that the

molecular elements that participate in this pathway are present

in the cells. It has been reported that SH-SY5Y cells express Ret

and GFRa1 (Fukuda et al., 2002). As can be observed in Fig. 1,

our results show the expression of the mRNAs for both ret and

gfra1, 3 and 24 h after serum withdrawal, in agreement with

Fig. 2. Expression of Gas1 in SH-SY5Y cells, determined by immunocytochemistr

control to reveal the nuclei. Calibration bar is 25 mm.

previous reports. SH-SY5Y cells were chosen for these

experiments, because they express very low or undetectable

levels of Gas1 when cultured in the presence of serum, or up to

3 h after serum withdrawal, but they express Gas1 when

cultured for 24 h in the absence of serum (Cabrera et al., 2006),

a time point at which cells do not show obvious signs of damage

or of death processes, and it has been observed that SH-SY5Y

cells remain viable for several days in the absence of serum

(Bar-Am et al., 2005). To determine whether Gas1 expression

occurred in our culture conditions, we examined gas1 mRNA

expression by RT-PCR and Gas1 protein expression by

immunocytochemistry in both growth conditions. Fig. 1 shows

that we cannot detect gas1 mRNA 3 h after serum withdrawal,

whereas it is readily observed 24 h after serum withdrawal.

Protein expression is consistent with these results, for which we

observe no or a very faint signal of Gas1 3 h after serum

withdrawal, but a strong signal is observed 24 h after serum

withdrawal (Fig. 2). These results indicate that SH-SY5Y cells

are an appropriate system to study the effect of Gas1 on GDNF-

mediated intracellular signaling.

2.2. Effect of gas1 on GDNF-mediated intracellular

signaling

Considering that Y 1062 is a key element in the functioning of

the Ret receptor, we determined the levels of phosphorylation of

this particular tyrosine using an antibody that selectively

recognizes Ret phosphorylated in this position. We observed,

by immunocytochemistry, that in the absence of Gas1 (3 h

without serum), GDNF induces a large increase of Ret

phosphorylated on Y 1062, when compared with control non-

treated cells, and that in the presence of Gas1 (24 h in the absence

of serum), there is a clear decrease of phosphorylated Ret,

y 3 and 24 h after serum withdrawal. Cells were counterstained with DAPI as a

Fig. 3. Activation of Ret on Y 1062, ascertained by immunocytochemistry, in the absence or presence of Gas1 (3 and 24 h after serum withdrawal, respectively), and

treated with 100 ng/ml of human GDNF, or non-treated (no addition of GDNF). A, C, E, and G correspond to Ret Y 1062 expression; B, D, F, and H is DAPI staining;

A–D in the absence of added GDNF, E–H in the presence of exogenous GDNF. Calibration bar is 25 mm.

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503500

furthermore, even after the addition of exogenous GDNF, levels

of phosphorylated Ret remained low in Gas1-expressing cells

(Fig. 3). In order to quantitate the effects of the presence of Gas1

on Ret activity, we performed western blot analysis. For this

series of experiments, we determined the phosphorylation of Y

1062 in control conditions (3 h in the absence of serum) and in the

presence of exogenously added GDNF. Fig. 4A and B shows

Fig. 4. Panel A shows a representative western blot of proteins obtained from SH-SY

in the presence of human recombinant GDNF (GDNF); GDNF/Gas1 indicates cells 2

right) are cells after 24 h of serum withdrawal, without exogenously applied GDNF; a

human recombinant GDNF for 15 min (p-RET1062, corresponds to Ret phosphoryla

analysis of phosphorylated Ret Y 1062, normalized with respect to total Ret, and con

without the addition of exogenous GDNF, as the control to which all other data are co

(ANOVA followed by Tukeys). Panel C shows the levels of expression of phosphoryla

the right corresponds to cells in the absence of Gas1 (3 h) and stimulated with 100 ng/

(24 h) in the presence of GDNF. M1 is the area corresponding to non-specific fluoresc

Gas1 37% of cells correspond to M2 and for cells expressing Gas1 only 8% of the cel

of Gas1 in the presence of a control siRNA molecule (siRNA-C) or siRNA-Gas1

phosphorylation of Ret Y 1062 under the same culture conditions which induce th

there is a significant increase of Y 1062 activation when GDNF is

added to the culture medium, in control conditions (161% of

control). Twenty-four hours after serum withdrawal, when Gas1

is expressed, we can observe that Y 1062 phosphorylation is

significantly lower, approximately 50% compared with the

control levels in the absence of Gas1, thus indicating that Gas1 is

capable of reducing basal levels of Y 1062 phosphorylation

5Y cells at 3 h after serum withdrawal in the absence (control non-treated, n/t) or

4 h after serum withdrawal and in the presence of GDNF; Gas1 (last lane to the

ctin expression is a positive control. Treatment was the application of 100 ng/ml

ted on Y 1062; t-RET is total Ret). Panel B shows the results of the densitometric

sidering the condition in the absence of Gas1 (3 h in the absence of serum), and

mpared to N = 6 independent experiments; *P < 0.05; **P < 0.01; ***P < 0.001

ted Ret Y 1062, determined by flow cytometry. The curve filled and displaced to

ml of GDNF; the curve displaced to the left corresponds to cells expressing Gas1

ence, and M2 corresponds to phosphorylated Y 1062 signal; for non-expressing

ls are in M2; data are from a representative experiment. Panel D shows the levels

that specifically recognizes gas1 mRNA (upper row); lower row shows the

e expression of Gas1 (24 h after serum withdrawal).

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503 501

(Fig. 4A and B). A highly significant difference is also observed

when comparing GDNF-induced Y 1062 activation in the

absence (3 h) or presence (24 h) of Gas1. Fig. 4A and B shows

that Gas1 completely prevents the GDNF-induced increase in

Ret Y 1062 phosphorylation, compared with control conditions.

There is no difference in Ret activation when Gas1 is present and

GDNF is added compared with untreated control cells; however

in the presence of Gas1 the GDNF-induced activation of Ret is

significantly lower, than that induced in the absence of Gas1, and

reaches only the level of the control, non-stimulated cells (95% of

control), although in Gas1 expressing conditions GDNF induces

an increase of Ret Y 1062 phosphorylation compared to cells in

the absence of GDNF (Fig. 4A and B). We wanted to analyze the

effect of Gas1 on Y 1062 phosphorylation using a different

technique, and determined, by flow cytometry, the activation of Y

1062. Fig. 4C shows that when GDNF is added to the culture

medium the curve is displaced to the right, indicating that there is

an increase in Y 1062 phosphorylation in SH-SY5Y cells. In

contrast, the increase of Y 1062 activation induced by GDNF is

blocked when the cells express Gas1. These results indicate that

Gas1 regulates Ret activity, by inhibiting the phosphorylation of

Y 1062. To more precisely define the effect of Gas1 on the

phosphorylation of Ret Y 1062, we used a siRNA to silence Gas1

in serum-deprived cultures. Fig. 4D shows that siRNA–Gas1

induced a decrease in the expression of Gas1 when compared

with the control siRNA, in which a strong signal is detected by

western blot. In the same panel, we show that the phosphorylation

of Ret Y 1062 is more intense when Gas1 expression is reduced

(siRNA–Gas1), than when high levels of Gas1 are present in the

serum-starved cultures. These data show that Gas1 directly

modulates the phosphorylation of Ret Y 1062.

To ascertain whether the decrease caused by Gas1 in the

activation of Ret Y 1062 induced by GDNF has an impact on

the intracellular signaling pathways regulated by this neuro-

Fig. 5. A representative western blot showing the changes in the activation of

Akt in SH-SY5Y cells at 3 h after serum withdrawal in the absence (non-treated,

n/t); or in the presence (GDNF) of human recombinant GDNF; GDNF/Gas1

indicates cells 24 h after cell withdrawal and in the presence of GDNF; Gas1

(last lane to the right) are cells after 24 h of serum withdrawal, without

exogenously applied GDNF. Treatments were the application of 100 ng/ml

human recombinant GDNF for 15 min. The results presented are the densito-

metric analysis of phosphorylated Akt (p-AKT) normalized with respect to total

Akt (t-AKT) and expressed as ratio of the control (non-treated) condition. The

mean of two independent experiments is shown in the bottom lane. Actin

expression is a positive control.

trophic factor we determined one of the main downstream

events of Ret phosphorylation, namely the activation of Akt,

which is crucial for cell survival. We used western analysis to

determine the levels of phosphorylation of Akt, and normalized

them with levels of total Akt. There is a basal signal of

phosphorylated Akt in control (no Gas1, 3 h without serum),

that increases two fold when GDNF is added to the culture

medium (Fig. 5). The presence of Gas1 (24 h without serum)

reduces Akt activation by 78%, and Gas1 is capable not only of

blocking the effect of exogenous GDNF, but to reduce Akt

phosphorylation below basal levels (0.69 of control). These

results indicate that the reduction of Ret Y 1062 phosphoryla-

tion caused by Gas1 affects the cascade of intracellular

signaling regulated by GDNF, and supports the concept that the

physiological mechanism of action of Gas1 is mediated by

negatively modulating the biological responses produced by

GDNF stimulation.

3. Discussion

GDNF and members of its family of ligands (GFLs), signal

through a multistep mechanism, by which the ligand forms a

pentameric molecular complex consisting of dimers of both the

GFRas and the Ret receptor tyrosine kinase (Schlee et al.,

2006). GDNF is a key component to maintain several cell

populations in the central nervous system, including dopami-

nergic and motor neurons, and also participates in the survival

and differentiation of peripheral neurons such as enteric,

sympathetic and parasympathetic. Recently, it has been shown

that GDNF and GFRa1 induce the formation of neuronal

synapses, and presynaptic differentiation, based on a mechan-

ism in which GDNF triggers the trans-homophilic binding

between different cells expressing GFRa1 (Ledda et al., 2007).

Furthermore, GDNF is also involved in kidney morphogenesis

and spermatogenesis. GFLs and their receptors, as well as Gas1

are present in invertebrates and in all vertebrates, thus

indicating their essential role in development (Airaksinen

et al., 2006; Hatinen et al., 2007). The molecular signaling

mediated by GDNF is very complex, because it can also signal

independently of Ret, requires the presence of TGF-ß to exert

its neurotrophic effect in many cell populations, interacts with

heparin sulphate glycosaminoglycans, and can also signal

through neural cell adhesion molecule, NCAM (for a review

see: Sariola and Saarma (2003)). On the other hand, we recently

showed the homology between Gas1 and GFRas, thus

suggesting that the physiological role of Gas1 is modulating

the biological responses induced by GDNF and/or other

members of this family of signaling molecules (Schueler-

Furman et al., 2006).

Gas1 is involved in cell arrest, and can induce apoptosis

when overexpressed in different cell lines, including cells of

neuronal and glial origin (Mellstrom et al., 2002; Zamorano

et al., 2003, 2004; Benitez et al., 2007). Although, Gas1 is

widely expressed during development its role in vivo has not yet

been clearly defined for contrasting results coming from

knockout mice have been reported (Lee et al., 2001b; Liu et al.,

2001). There have also been several proposals regarding the

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503502

molecular mechanism of action of Gas1, including the

necessary presence of p53 for Gas1 to exert its effects

inhibiting cell cycle, its interaction with Dlk1, and of Gas1

acting as antagonist to Shh (Del Sal et al., 1995; Baladron et al.,

2002; Lee et al., 2001b). The demonstration of the structural

homology of Gas1 with GFRas, provides us with a rational

basis to test hypothesis referring to the molecular mechanisms

of action of Gas1, and allow us to propose the concept that the

biological effects of Gas1 on neural cells and in brain are

caused by its capacity to modulate GDNF-signaling.

To test this hypothesis, we determined the effects of the

presence of Gas1 on the activation, induced by GDNF, of the

Ret receptor tyrosine kinase, and downstream, on the activation

of Akt, a key component of the GDNF-mediated cascade of

events, that is involved in cell survival pathways. Ret can be

autophosphorylated in several different tyrosines, as well as in

serine (Santoro et al., 2004; Sariola and Saarma, 2003), but Y

1062 has been found to be critical for GDNF signaling and for

the binding of different adaptor molecules, including Shc,

FRS2, DOK4/5, IRS 1/2 and Enigma (Airaksinen and Saarma,

2002), and it has been demonstrated that this tyrosine is

indispensable for the activation of PI3K and for the recruitment

of MAPK and Akt, as well as for recruiting Ret to lipid rafts (De

Vita et al., 2000; Hayashi et al., 2000; Paratcha et al., 2001).

Considering the relevance of the phosphorylation of Ret in this

particular tyrosine, we decided to study the effect of GDNF on

the activation of Ret in Y 1062, and employed an antibody that

specifically recognizes Ret phosphorylated in this position. Our

results show that the expression of Gas1 in SH-SY5Y cells

significantly reduces the increase of GDNF-induced phosphor-

ylation of Y 1062, compared with cells in the absence of Gas1.

Furthermore, in Gas1 expressing cells, the levels of Ret

phosphorylation are significantly lower compared with control

cells. However, there is an increase on Ret Y 1062

phosphorylation, when comparing GDNF-treated and non-

treated cells expressing Gas1 (Figs. 3 and 4). Basal levels of

GDNF are low in SH-SY5Y cells, and serum starvation, does

not increase GDNF expression (Bar-Am et al., 2005). We

hypothesized that the period (24 h) in the absence of serum,

may induce a greater affinity for GDNF, and Gas1 expression

may not completely compensate for this up-regulation, so a

partial response to exogenous GDNF, occurs even in the

presence of Gas1. In any event, the response induced by GDNF

in the presence of Gas1, is much lower than that induced in the

absence of Gas1, and only reaches control levels (95% of

control). We consider that the present results are consistent with

the hypothesis of Gas1 acting as a negative modulator of GDNF

signaling. Our results also show that Gas1 is responsible for

modulating Ret Y 1062 phosphorylation, because in serum-

starved cells, knocking-down Gas1 by the use of a specific

siRNA is reflected in the levels of Ret Y 1062 phosphorylation.

This is shown in Fig. 4D, where we observe that when Gas1

expression decreases, the activation of Ret Y 1062 increases;

whereas in the presence of Gas1, the phosphorylation of Ret Y

1062 diminishes. This occurs in the same serum-deprived

culture conditions that favor the expression of Gas1. Moreover,

we also observed that Gas1 expression significantly reduced the

activation of Akt, a result that is also coherent with the effect of

Gas1 inhibiting the phosphorylation of Ret Y 1062, and with

the hypothesis of Gas1 as a modulator of GDNF.

Recently, it has been reported that the mechanism of action of

Gas1 is caused by its capacity to bind Ret, sequester this molecule

in lipid rafts, and that it produces no effect on Ret phosphoryla-

tion (Cabrera et al., 2006). In the present paper, we have not

determined whether Ret is localized in lipid rafts, but it is widely

accepted that the function of GFRa1 is to recruit Ret to lipid rafts,

to activate it (Sariola and Saarma, 2003). It is also well

established that in lipid rafts activated Ret associates with the

adaptor protein FRS2, and outside of rafts it interacts with Shc,

however, the association of Ret with both FRS2 and Shc is via

phosphorylated Y 1062 (Paratcha et al., 2001; Sariola and

Saarma, 2003), thus indicating that inside or outside of rafts Y

1062 phosphorylation is a necessary event to activate

intracellular signaling pathways, including Akt, and MAPK/

ERK. We consider that some of the discrepancy between our data

and other previously reported (Cabrera et al., 2006) may be due to

the fact of the specificity determining the phosphorylation of Ret

in one single position (Y 1062), because when using a general

antiphosphorylation reagent, the overall phosphorylation of the

Ret receptor can be ascertained, but the phosphorylation state of a

single tyrosine residue cannot (Schlee et al., 2006). Gas1 may

interfere with GDNF-signaling at different levels, thus these data

are not necessarily contradictory. We deem that the manner by

which Gas1 negatively modulates GDNF-mediated signaling

shown in this paper, that is by reducing the levels of Ret Y 1062

phosphorylation, is a novel observation, indicating a mechanism

of action by which the effects of Gas1 can be more completely

understood. Based on these data, it is possible to explain how

Gas1 could affect most of the GDNF-induced signaling

pathways, irrespectively of whether Ret is inside or outside

lipid rafts, and could also account for Gas1 modulation of the

differential kinetics of Ret signaling when interacting with

membrane bound or soluble GFRa1 (Paratcha et al., 2001). This

proposal is also consistent with current knowledge of the

functioning of tyrosine kinase receptors, since the changes in the

pattern of phosphorylation of Ret negatively affects its activity,

and reduce the operation of the intracellular signaling cascades

induced by GDNF. Thus, we consider that an effect altering the

transducing capacities of Ret is a consistent hypothesis to explain

how Gas1 may modulate GDNF functioning. At the present time

we do not know how Gas1 inhibits Y 1062 activation, whether it

is a structural hindrance effect when it interacts with the GFL–

GFRa–Ret multiprotein complex, if another molecule is

recruited, or by a different mechanism. It is clear, however,

that further work is necessary to completely dissect the molecular

cascade modulated by Gas1, that ends inducing a caspase-3-

dependent apoptotic process (Zamorano et al., 2003, 2004). It has

been reported that Ret acts as a dependence receptor, and that it

induces caspase-3-mediated apoptosis in the absence of GDNF

(Bordeaux et al., 2000). These data are consistent with our

finding of the activation of caspase-3 in glioma cells transfected

with gas1 (Zamorano et al., 2003, 2004), and could be interpreted

as an apoptotic process triggered by the lack of GDNF activity,

caused by the interference of Gas1 on Ret activation. The recent

M.A. Lopez-Ramırez et al. / Int. J. Devl Neuroscience 26 (2008) 497–503 503

findings of a positive action of Gas1 on Shh activity during

development (Martinelli and Fan, 2007; Allen et al., 2007), as

well as its regulatory effect on GDNF suggest that Gas1 plays a

critical role in the integration of different intercellular signaling

systems, during development and in the adult brain.

The results presented in this paper are consistent with a

unifying principle to study the molecular mechanism of action

of Gas1 in the nervous system, and support a new conceptual

frame to understand how GDNF exerts its many biological

effects. On the other hand, Gas1 might be an important agent in

the treatment of tumors, particularly gliomas (Zamorano et al.,

2004; Benitez et al., 2007), so understanding its mechanism of

action will be useful in this respect, since it may help

identifying molecules and pathways of therapeutic relevance.

Acknowledgements

This work was partially supported by CONACyT grant

54756 (JS). We thank Dr. J.M. Hernandez (Cinvestav) for the

kind gift of the antibody against ß-actin, and B.E. Reyes-

Marquez for her assistance with the FACS experiments.

References

Airaksinen, M.S., Holm, L., Hatinen, T., 2006. Evolution of the GDNF family

ligands and receptors. Brain Behav. Evol. 68, 181–190.

Airaksinen, M.S., Saarma, M., 2002. The GDNF family: signalling, biological

functions and therapeutic value. Nat. Rev. Neurosci. 3, 383–394.

Allen, B.L., Tenzen, T., McMahon, A.P., 2007. The Hedgehog-binding proteins

Gas1 and Cdo cooperate to positively regulate Shh signaling during mouse

development. Genes Dev. 21, 1244–1257.

Baladron, V., Ruiz-Hidalgo, M.J., Bonvini, E., Gubina, E., Notario, V., Laborda,

J., 2002. The EGF-like homeotic protein dlk affects cell growth and

interacts with growth-modulating molecules in the yeast two-hybrid system.

Biochem. Biophys. Res. Commun. 291, 193–204.

Bar-Am, O., Weinreb, O., Amit, T., Youdim, M.B.H., 2005. Regulation of Bcl-2

family proteins, neurotrophic factors, and APP processing in the neuror-

escue activity of propargylamine. FASEB J. 19, 1899–1901.

Benitez, J.A., Arregui, L., Vergara, P., Segovia, J., 2007. Targeted-simultaneous

expression of Gas1 and p53 using a bicistronic adenoviral vector in gliomas.

Cancer Gene Ther. 14, 836–846.

Bordeaux, M.C., Forcet, C., Granger, L., Corset, V., Bidaud, C., Billaud, M.,

Bredesen, D.E., Edery, P., Mehlen, P., 2000. The RET proto-oncogene

induces apoptosis: a novel mechanism for Hirschsprung disease. EMBO J.

19, 4056–4063.

Cabrera, J.R., Sanchez-Pulido, L., Rojas, A.M., Valencia, A., Manes, S.,

Naranjo, J.R., Mellstrom, B., 2006. Gas1 is related to the glial cell-derived

neurotrophic factor family receptors a and regulates ret signaling. J. Biol.

Chem. 281, 14330–14339.

Choi-Lundberg, D.L., Bohn, M.C., 1995. Ontogeny and distribution of glial cell

line-derived neurotrophic factor (GDNF) mRNA in rat. Brain Res. Dev.

Brain Res. 85, 80–88.

De Vita, G., Melillo, R.M., Carlomagno, F., Visconti, R., Castellone, M.D.,

Bellacosa, A., Billaud, M., Fusco, A., Tsichlis, P.N., Santoro, M., 2000.

Tyrosine 1062 of RET-MEN2A mediates activation of Akt (Protein Kinase

B) and mitogen-activated protein kinase pathways leading to PC12 cell

survival. Cancer Res. 60, 3727–3731.

Del Sal, G., Ruaro, E.M., Utrera, R., Cole, C.N., Levine, A.J., Schneider, C.,

1995. Gas1-induced growth suppression requires a transactivation-indepen-

dent p53 function. Mol. Cell. Biol. 15, 7152–7160.

Del Sal, G., Ruaro, M.E., Philipson, L., Schneider, C., 1992. The growth arrest-

specific gene, gas1, is involved in growth suppression. Cell 70, 595–607.

Evdokiou, A., Cowled, P.A., 1998. Tumor-suppressive activity of the growth

arrest-specific gene GAS1 in human tumor cell lines. Int. J. Cancer 75,

568–577.

Fukuda, T., Kiuchi, K., Takahashi, M., 2002. Novel mechanism of regulation of

Rac activity and lamellipodia formation by RET tyrosine kinase. J. Biol.

Chem. 277, 19114–19121.

Garcia-Tovar, C.G., Perez, A., Luna, J., Mena, R., Osorio, B., Aleman, V.,

Mondragon, R., Mornet, D., Rendon, A., Hernandez, J.M., 2001. Biochem-

ical and histochemical analysis of 71 kDa dystrophin isoform (Dp71f) in rat

brain. Acta Histochem. 103 (2), 209–224.

Hatinen, T., Holm, L., Airaksinen, M.S., 2007. Loss of neurturin in frog—

comparative genomics study of GDNF family ligand-receptor pairs. Mol.

Cell Neurosci. 34, 155–167.

Hayashi, H., Ichihara, M., Iwashita, T., Murakami, H., Shimono, Y., Kawai, K.,

Kurokawa, K., Murakumo, Y., Imai, T., Funahashi, H., Nakao, A., Takaha-

shi, M., 2000. Characterization of intracellular signals via tyrosine 1062 in

RET activated by glial cell line-derived neurotrophic factor. Oncogene 19,

4469–4475.

Ledda, F., Paratcha, G., Sandoval-Guzman, T., Ibanez, C.F., 2007. GDNF and

GFRalpha1 promote formation of neuronal synapses by ligand-induced cell

adhesion. Nat. Neurosci. 10, 293–300.

Lee, C.S., Buttitta, L., Fan, C.-M., 2001a. Evidence that the WNT-inducible

growth arrest-specific gene 1 encodes an antagonist of sonic hedgehog

signaling in the somite. Proc. Natl. Acad. Sci. USA 98, 11347–11352.

Lee, C.S., May, N.R., Fan, C.M., 2001b. Transdifferentiation of the ventral

retinal pigmented epithelium to neural retina in the growth arrest specific

gene 1 mutant. Dev. Biol. 236, 17–29.

Liu, Y., May, N.R., Fan, C.M., 2001. Growth arrest specific gene 1 is a positive

growth regulator for the cerebellum. Dev. Biol. 236, 30–45.

Martinelli, D.C., Fan, C.-M., 2007. Gas1 extends the range of Hedgehog action

by facilitating its signaling. Genes Dev. 21, 1231–1243.

Mellstrom, B., Cena, V., Lamas, M., Perales, C., Gonzalez, C., Naranjo, J.R.,

2002. Gas1 is induced during and participates in excitotoxic neuronal death.

Mol. Cell Neurosci. 19, 417–429.

Paratcha, G., Ledda, F., Baars, L., Coulpier, M., Besset, V., Anders, J., Scott, R.,

Ibanez, C.F., 2001. Released GFRalpha1 potentiates downstream signaling,

neuronal survival, and differentiation via a novel mechanism of recruitment

of c-Ret to lipid rafts. Neuron 29, 171–184.

Santoro, M., Carlomagno, F., Melillo, R.M., Fusco, A., 2004. Dysfunction of the

RET receptor in human cancer. Cell. Mol. Life Sci. 61, 2954–2964.

Sariola, H., Saarma, M., 2003. Novel functions and signalling pathways for

GDNF. J. Cell Sci. 116, 3855–3862.

Schlee, S., Carmillo, P., Whitty, A., 2006. Quantitative analysis of the activation

mechanism of the multicomponent growth-factor receptor Ret. Nat. Chem.

Biol. 2, 636–644.

Schneider, C., King, R.M., Philipson, L., 1988. Genes specifically expressed at

growth arrest of mammalian cells. Cell 54, 787–793.

Schueler-Furman, O., Glick, E., Segovia, J., Linial, M., 2006. Is GAS1 a co-

receptor for the GDNF family of ligands? Trends Pharmacol. Sci. 27, 72–

77.

Seppala, M., Depew, M., Martinelli, D.C., Fan, C.M., Sharpe, P., Cobourne, M.,

2007. Gas1 is a modifier for holoprosencephaly an genetically interacts with

sonic hedgehog. J. Clin. Invest. 117, 1575–1584.

Spagnuolo, R., Corada, M., Orsenigo, F., Zanetta, L., Deuschle, U., Sandy, P.,

Schneider, C., Drake, C.J., Breviario, F., Dejana, E., 2004. Gas1 is induced

by VE-cadherin and vascular endothelial growth factor and inhibits

endothelial cell apoptosis. Blood 103, 3005–3012.

Stebel, M., Vatta, P., Ruaro, M.E., Del Sal, G., Parton, R.G., Schneider, C., 2000.

The growth suppressing gas1 product is a GPI-linked protein. FEBS Lett.

481, 152–158.

Zamorano, A., Lamas, M., Vergara, P., Naranjo, J.R., Segovia, J., 2003.

Transcriptionally mediated gene targeting of gas1 to glioma cells elicits

growth arrest and apoptosis. J. Neurosci. Res. 71, 256–263.

Zamorano, A., Mellstrom, B., Vergara, P., Naranjo, J.R., Segovia, J., 2004.

Glial-specific retrovirally mediated gas1 gene expression induces glioma

cell apoptosis and inhibits tumor growth in vivo. Neurobiol. Dis. 15,

483–491.