CREB Regulates AChE-R-Induced Proliferation of Human Glioblastoma Cells

CREB Regulates AChE-R–Induced Proliferation of HumanGlioblastoma Cells1

Chava Perry*,y, Ella H. Sklan* and Hermona Soreq*

*Department of Biological Chemistry, The Institute of Life Sciences, The Hebrew University of Jerusalem,Jerusalem 91904, Israel; yDepartment of Hematology, The Tel-Aviv Sourasky Medical Center-Tel Avivand Tel-Aviv University, Tel-Aviv 64239, Israel

Abstract

The cyclic adenosine monophosphate (AMP) re-

sponse element-binding protein, CREB, often modu-

lates stress responses. Here, we report that CREB

suppresses the glioblastoma proliferative effect of the

stress-induced acetylcholinesterase variant, AChE-R.

In human U87MG glioblastoma cells, AChE-R formed a

triple complex with protein kinase C (PKC) E and the

scaffold protein RACK1, enhanced PKCE phosphor-

ylation, and facilitated BrdU incorporation. Either over-

expressed CREB, or antisense destruction of AChE-R

mRNA, PKC, or protein kinase A (PKA) inhibitors—but

not CREB combined with PKC inhibition suppressed—

this proliferation, suggesting that CREB’s repression

of this process involves a PKC-mediated pathway,

whereas impaired CREB regulation allows AChE-R–

induced, PKA-mediated proliferation of glioblastoma

tumors.

Neoplasia (2004) 6, 279–286

Keywords: Acetylcholinesterase, antisense, CREB, glioblastoma, PKCq.

Introduction

Glioblastoma multiforme (GBM), the most common primary

brain tumor, carries a grave prognosis, despite aggressive

treatment [1]. However, the mechanisms underlying GBM

pathogenesis and poor response to conventional therapy

are yet unclear. GBMs commonly overexpress both the

platelet-derived growth factor receptor and the epidermal

growth factor receptor and their ligands [2]; the latter can

activate various signaling pathways associated with glio-

blastoma cell survival and tumor formation [3]. Also, GBMs

either overexpress or lose expression of various protein

kinase C (PKC) isoforms implicated in cell proliferation and

invasion [4]. In particular, various stress signals, growth

factors, and kinases promote phosphorylation-mediated

activation of the cyclic adenosine monophosphate (AMP)

response element-binding transcription factor, CREB, in-

volved in glial cell fate determination [5,6]. As some of these

signals also induce expression of the acetylcholinesterase

variant AChE-R [7] and because AChE-R is involved in

glioblastoma proliferation [8], we explored the possibility

that these effects are interrelated. The three 3V splice variants

of AChE have distinct noncatalytic activities (reviewed in Ref.

[7]; Figure 1A). The ‘‘synaptic’’ (tailed) isoform, AChE-S, is the

principal AChE variant in the brain and muscles and its C-

terminus is encoded by the open reading frame in exon (E) 6.

AChE-E, the ‘‘erythrocytic’’ (hydrophobic) isoform, links E4 and

E5 to encode a different 43–amino acid C-terminal peptide,

which is anchored through a glycophosphoinositide moiety to

erythrocyte surface membranes. The ‘‘readthrough’’ isoform,

AChE-R, expressed in embryonic and tumor cells possesses a

C-terminus encoded by intron 4. AChE-R is overproduced

under psychological stress in response to AChE inhibitors

and in myasthenic muscles, all of which are under cholinergic

imbalance [7–9]. We have recently shown that AChE mRNA

accumulates in primary human astrocytomas in correlation with

these tumors’ grade of aggressiveness, which further associ-

ates with an mRNA splicing shift from AChE-S to the AChE-R

transcript [10]. In the present study, we further investigate the

function of the R-splice variant in cell proliferation and the

signaling molecules that mediate AChE-R’s effect. Here, we

report that in human glioblastoma cells, CREB, a common

downstream target for multiple signaling pathways, is an intrin-

sic repressor of PKCq-mediated AChE-R–induced prolifera-

tion, and demonstrate how this function may fail under drastic

excess of AChE-R. Under PKC inhibition, which blocks CREB’s

repression, AChE-R may still promote proliferation, probably

through a protein kinase A (PKA)-mediated pathway.

Materials and Methods

Cell Cultures and Transfection

Human glioblastoma U87MG and COS1 cells were grown in

Dulbecco’s modified Eagle’s medium (Biological Industries,

Beit Ha’emek, Israel) with 10% fetal calf serum (FCS) and

Address all correspondence to: Hermona Soreq, Life Sciences Institute, Givat Ram,

Jerusalem, Israel. E-mail: [email protected] study was supported by the Israel Cancer Association and the US Army Medical

Research and Materiel Command (DAMD 17-99-1-9547).

Received 3 November 2003; Revised 9 December 2003; Accepted 11 December 2003.

Copyright D 2004 Neoplasia Press, Inc. All rights reserved 1522-8002/04/$25.00

DOI 10.1593/neo.03424

Neoplasia . Vol. 6, No. 3, May/June 2004, pp. 279 – 286 279

www.neoplasia.com

BRIEF ARTICLE

2 mM L-glutamine at 37jC and 5% CO2-humidified chamber.

PC12 cells were grown in Dulbecco’s modified Eagle’s

medium with 10% FCS, 8% donor horse serum (DHS), and

2 mM L-glutamine. Transfections involved 0.4 or 12 mg of

plasmid DNA per well in 48-well plates or 25-cm2 flasks

(using Lipofectamine Plus; Gibco BRL Life Technologies,

Bethesda, MD) of either human AChE-R, AChE-S [8] or

CREB expression vectors [11], or AChE-R and CREB to-

gether. The nonrelevant pEGFP-C2 plasmid (Clontech Lab-

oratories, Palo Alto, CA) served as control.

Nuclear Protein Extractions

Nuclear and cytoplasmic protein extracts were prepared

from U87MG cells 24 hours posttransfection. Cells were

washed twice in phosphate-buffered saline (PBS), scraped,

precipitated by centrifugation (1000 rpm, 5 minutes), then

rewashed in PBS. Precipitates were incubated with com-

plete miniprotease inhibitor cocktail (Roche Diagnostics

GmbH, Mannheim, Germany) and 150 ml of buffer A

(10 mM Tris–HCl, pH 7.4, 10 mM NaCl, and 1 mM EDTA)

on ice (15 minutes), lysed through a 21-gauge needle, and

centrifuged at 14,000 rpm (4jC, 10 minutes). Supernatants

containing cytoplasmic protein extracts were removed.

One hundred microliters of buffer B (10 mM Tris–HCl,

pH 7.4, 10 mM NaCl, and 1.5 mM MgCl2), 25 ml of 5 M NaCl,

and complete miniprotease inhibitor cocktail were added

to the precipitate, incubated on ice (30 minutes), and cen-

trifuged at 14,000 rpm (4jC, 10 minutes). Supernatants

containing nuclear protein extracts were stored at �20jC

until use.

Cell Proliferation Assay

U87MG human glioblastoma cells were grown in 48-well

plates and transfected (in quadriplicates) as described

above. Cell proliferation was assessed 30 hours posttrans-

fection by measuring the incorporation of 5V-bromo-2-deox-

yuridine (BrdU; Roche Diagnostics GmbH) over 6 hours, as

previously described [12].

Oligonucleotides

Human (h)EN101, a 20-mer antisense oligonucleotide

targeted at exon 2 of human AChE mRNA, was added

to the culture medium with transfected DNA at a concentra-

tion of 2 nM, previously reported to induce preferential

Figure 1. CREB interactions with AChE splice variants affect glioblastoma proliferation. (A) Shown is a schematic presentation of the ACHE gene (as included in

the reverse sequence of the cosmid inset; accession no. AF002993) and the relevant splice variants, AChE-S and AChE-R. Exons (gray boxes) and introns (white

boxes) are marked. ACHE gene expression is regulated by a distal domain (DD), a proximal promoter (PP) that includes a CRE domain, and an intronic enhancer

(IE). In glioblastoma cells, this gene is transcribed into AChE-R and AChE-S mRNA with distinct 3 V domains [7,8]. Expression of CREB in U87MG cells cytoplasmic

(C) and nuclear (N) extracts transfected with either irrelevant DNA (Ct), AChE-R, AChE-S, or CREB expression vectors, or AChE-R and CREB together.

Transfection efficacy, assessed by GFP expression, was 25 ± 8%. Lower panel shows densitometric quantification of CREB expression in the nuclear extracts. (C)

AChE-R overexpression increases U87MG cell proliferation in an antisense suppressible manner: BrdU incorporation was assessed in U87MG cells, under various

treatments, with or without overexpressing AChE-R (black and white columns, respectively). Treatment included transfections with AChE-S or CREB expression

vectors, or treatment with the human (h) EN101 antisense oligonucleotide targeted at human AChE mRNA or with the corresponding inverse oligonucleotide

INVEN101, in both cases with or without AChE-R. Notice that both CREB and EN101 were able to suppress AChE-R – induced cell proliferation. Columns show

percent increase over control (= cells transfected with the irrelevant GFP plasmid) ± SEM; average of four duplicate transfections. *Statistically significant

difference from control (P < .005, ANOVA). Mean absorbance value (A405 /A492) for control was 0.3.

280 CREB/AChE-R Regulation of Glioblastoma Growth Perry et al.

Neoplasia . Vol. 6, No. 3, 2004

degradation of AChE-R mRNA [9,12,13]. Antisense pene-

trance into cells was previously quantified using fluorescein-

labeled EN101. At 2 nM, virtually all treated cells include

EN101 molecules [14]. EN101’s three 3V-terminal residues

(*) were substituted with oxymethyl groups at the 2V position

(5V-CTGCGATATTTTCTT GTA*C*C*-3V). Similarly pro-

tected INVEN101, an oligonucleotide of sequence inverse

to EN101, served as control [9].

Immunoprecipitation

Cellhomogenateswereprepared [13]and incubated (over-

night, 4jC) with 2 mg of goat polyclonal antibodies targeted

to the human AChE N-terminus (Santa Cruz Biotechnology,

Santa Cruz, CA), then with 50 ml of suspended four Fast Flow

proteinG Sepharose beads (1 hour, 4jC;Amersham Pharma-

cia Biotech, Upsala, Sweden). Sedimented beads were

washed three times in NET buffer (50 mM Tris–HCl, pH 7.4,

150 mM NaCl, 1 mM EDTA, 0.25% gelatin, and complete

miniprotease inhibitor cocktail; Roche Diagnostics GmbH),

suspended in sample buffer, heated (95jC, 5 minutes), and

centrifuged at maximal speed to remove beads.

Immunoblot Analyses

U87MG cell homogenates prepared 24 hours posttrans-

fection, protein extracts, or immunoprecipitates prepared as

described above were separated by sodium dodecyl sulfate

polyacrylamide gel electrophoresis (SDS-PAGE; BioRad,

Hercules, CA) and blotted onto nitrocellulose membranes.

Membranes were incubated in a blocking solution (3% skim

milk, 2% bovine serum albumin [or 5% bovine serum albumin

when detecting a phosphorylated protein], 0.3% Tween-20 in

Tris-buffered saline, 1 hour, room temperature). Immunode-

tection (4jC, overnight) was with either monoclonal mouse

anti-PKCb (P2584; Sigma Chemical Co., St. Louis, MO)

diluted 1:8000, mouse anti-PKCq (Transduction Laborato-

ries, San Diego, CA) diluted 1:1000, mouse anti-RACK1

(R20620; Transduction Laboratories, Lexington, KY) diluted

1:2500, rabbit anti–phosphorylated PKC-q antibodies (dilut-

ed 1:300; Santa Cruz Biotechnology), rabbit anti-CREB

diluted 1:800 (New England Biolabs, Beverly, MA), or rabbit

anti–Ser-133 phosphorylated CREB diluted 1:600 (New

England Biolabs). Development (room temperature, 2 hours)

was with horseradish peroxidase–conjugated goat anti–

mouse or anti–rabbit antibodies diluted 1:10,000 (Jackson

Laboratories, West Grove, PA) and enhanced chemilumi-

nescence (ECL) kit (Amersham Pharmacia Biotech, UK).

Specific Inhibitors

H-89, a specific PKA inhibitor, and bisindolylmaleimide

(BIM), a specific PKC inhibitor (Calbiochem, San Diego, CA),

were used at 10 mM. Diisopropylfluorophosphate (DFP), an

organophosphate inhibitor of AChE catalytic activity, was

used at 1 mM.

AChE’s catalytic activity

AChE’s catalytic activity, measured by accumulation of

hydrolyzed acetylthiocholine (ATCh), was assessed as pre-

viously described [13].

Results

To explore the potentially interrelated role(s) of CREB and

AChE-R in controlling human glioblastoma cell proliferation

and to search for the mechanism(s) underlying such prolif-

eration, we compared transfection and antisense suppres-

sion effects with those of anticholinesterases and protein

kinase inhibitors by measuring cell proliferation and protein

levels and properties.

CREB Can Suppress AChE-R–Induced Proliferation

In human glioblastoma U87MG cells, transfection with an

AChE-R, but not an AChE-S expression vector, enhanced

BrdU incorporation by 42 ± 4% (average ± SEM; Figure 1C),

as compared to control cells transfected with the nonrelevant

pEGFP-C2 plasmid (P = .001, ANOVA). This suggested that

AChE-R’s proliferative effect is independent of AChE’s cat-

alytic activity, shared by these two AChE variants.

AChE-R is a soluble protein [7]. Thus, AChE-R transfec-

tion leads to AChE-R secretion, which may spread the

proliferative effect also to cells that do not contain the

plasmid. Therefore, the increase in proliferation surpassed

the percentage of transfected cells, which was assessed by

GFP expression as 25 ± 8%.

The transcription factor, CREB, notably mediates cellular

responses to various mitogens and stressors. CREB is

expressed in glioblastoma cells; however, its modulation in

these cells under stress was not studied extensively. CREB

transfection resulted in nuclear accumulation of its protein

product in U87MG cells (Figure 1B). Also, CREB transfection

induced irreproducible changes in BrdU incorporation, with

insignificant effects on cell proliferation. AChE activity re-

mained unchanged in cells overexpressing CREB as com-

pared to controls, although the ACHE promoter includes a

potential CREB-binding site (data not shown). In contrast,

U87MG cells cotransfected with CREB and AChE-R dis-

played suppressed cell proliferation to a nonsignificant dif-

ference from control (control levels ± 5%) (Figure 1C). CREB

levels remained high in the cotransfected cells, suggesting

an antimitogenic role for CREB over AChE-R–mediated

proliferation in astrocytes. To test the alternative possibility

—namely that both AChE-R and CREB (although to a lesser

extent) each act independently to induce cell prolifera-

tion (but contact-mediated growth control masks this cu-

mulative effect)—we cultured U87MG cells at lower

densities (10,000 and 5000 cells/well). However, in these

cultures as well, co-overexpression of AChE-R and CREB

did not increase cell proliferation as compared to control

cultures (data not shown), making this possibility highly

unlikely.

Antisense Suppression Supports the Notion of a Selective

AChE-R Proliferative Effect

EN101 selectively induces destruction of the AChE-R

mRNA transcript. At 2 nM concentration, EN101 induced

an inconsistent effect on cell proliferation, which did not

reach statistical significance. However, in U87MG cells over-

expressing AChE-R, EN101 reduced proliferation by 78%

(from 42% to 11% increase in BrdU incorporation over

CREB/AChE-R Regulation of Glioblastoma Growth Perry et al. 281


control, P < .005). The effect was sequence-specific, as the

inverse sequence, INVEN101, did not suppress AChE-R–

induced proliferation (maintaining a 35 ± 4% increase in cell

proliferation over control; Figure 1C). Thus, both gain of

function (namely, the proliferative effect induced by trans-

fections with an AChE-R, but not AChE-S expression vector)

as well as loss of function (abolition of the AChE-R prolifer-

ative effect by a selective antisense agent) support the

variant specificity of AChE-R’s proliferative effect.

AChE-R Forms Triple Complexes with RACK1 and PKCq

In neuronal cells, AChE-R forms intracellular triple com-

plexes with PKCbII and its scaffold protein, RACK1 [13].

However, immunoblot analysis failed to detect PKCb in

U87MG cell extracts (data not shown). To explore the

protein–protein interactions of AChE-R in glioblastoma cells,

we used antibodies targeted at the N-terminus of AChE for

coimmunoprecipitation tests. In cell homogenates from

AChE-R–transfected U87MG cells, these antibodies again

failed to coimmunoprecipitate PKCbII, but coimmunoprecipi-

tated both RACK1 and PKCq, a novel calcium-independent

PKC isotype previously reported as involved in astrocytoma

proliferation [15] (Figure 2). Antibodies to RACK1, which

efficiently detect it in immunoblots, failed to pull down these

multiprotein complexes. Compatible with previous reports

[13], this may be due to the corresponding epitope being

located in a site masked in the AChE-R/RACK1/PKC com-

plexes. Anti–AChE antibodies failed to precipitate any of

these proteins in monkey kidney COS1 cells, which do not

express AChE, attesting to the specificity of these analyses.

Rat pheochromocytoma (PC12) cells express both PKCband PKCq; however, anti–AChE antibodies pulled down

RACK1 and PKCb, but not PKCq, from homogenates of

these cells (Figure 2), suggesting that PKCb competes

successfully with PKCq in the formation of such complexes

and that the RACK1-mediated AChE-R interaction with PKC

is cell type– and PKC isoform–specific.

AChE-R Overexpression in U87MG Cells Facilitates PKCq

Phosphorylation

To test the functional significance of the in vitro–observed

AChE-R–PKCq interaction, we studied the effect of AChE-R

overexpression on PKCq phosphorylation by using selective

antibodies (Figure 3A).

Facilitated PKCq interaction with antibodies specific for

phosphorylated PKCq was reproducibly observed in extracts

of U87MG cells overexpressing AChE-R, as compared to

control cells or cells overexpressing AChE-S. This effect, as

well, appeared to be independent of AChE’s catalytic activity.

Although AChE expression is well documented in primary

human glioblastoma tumors [16,17], U87MG cells showed

only minimal endogenous AChE catalytic activity. Also,

AChE protein levels were negligible as immunolabeling of

AChE in U87MG extracts was very faint (data not shown),

likely reflecting residual levels of the protein from the serum

component of the medium the cells were grown in. Over-

expressing the AChE variants in these cells makes this

model somewhat more biologically relevant. AChE activity

was higher in cell homogenates overexpressing AChE-S;

however, PKCq phosphorylation was much more limited, as

compared with AChE-R–overexpressing cells (Figure 3B).

Thus, these data support our hypothesis that the RACK1–

AChE-R interaction, but not acetylcholine hydrolysis, facili-

tates the AChE-R–induced proliferation.

Figure 2. PKCe, but not PKC�, forms triple complexes with AChE and RACK1 in U87MG cells. Shown are results of immunodetection of RACK1, PKCe, and PKC�

in either U87MG, PC12, or COS1 cells immunoprecipitated with antibodies targeted to the human AChE N-terminus or PC12 whole cell homogenates. Schematic

presentations of the protein complexes specific for each cell type, as suggested by the immunoprecipitations, are drawn.



AChE-R– Induced Proliferation Involves PKC and PKA

Phosphorylation, But Not Cholinergic Signaling

We treated AChE-R – transfected U87MG cells with

1 mM DFP, an inhibitor of AChE’s catalytic activity; H-89, a

selective PKA inhibitor [18]; or BIM, which inhibits PKC

activities [19]. DFP did not suppress the AChE-R–induced

proliferation of AChE-R transfected U87MG cells (data not

shown), supporting our conclusion that this effect is non-

catalytic. In contrast, at 10 mM, either H-89 or BIM completely

suppressed the AChE-R proliferative effect (Figure 4A),

suggesting that both PKC and PKA signaling pathways

are involved in the AChE-R – induced proliferation of

U87MG cells.

AChE-R Proliferative Effect Under PKC Inhibition of

CREB Signaling

Neither BIM nor H-89 affected U87MG proliferation under

AChE-S/CREB co-overexpression. Also, the PKA inhibitor,

H-89, had no apparent effect on U87MG cell proliferation

under AChE-R/CREB co-overexpression. In contrast, cell

proliferation increased significantly (46 ± 1.5%, P < .005)

over control in BIM-treated cells co-overexpressing AChE-R

and CREB (Figure 4A), demonstrating that BIM revoked

CREB’s suppression of AChE-R–induced proliferation, re-

trieving the full scope of the AChE-R proliferative effect. This

was compatible with the assumption that the CREB-sup-

pressive effect over AChE-R–induced U87MG cell prolifer-

ation depends on PKC activation. Indeed, CREB

phosphorylation increased in cells cotransfected with

AChE-R and CREB, suggesting an interaction between

these two signaling pathways (Figure 4B). Nevertheless,

under PKC inhibition, which prevents the suppressive effect

of CREB, AChE-R proliferative effects could be trans-

duced through the PKA-dependent pathway (Figure 4, left

upper panel ).

Discussion

Using U87MG cells, we found that the transcription factor,

CREB, and the stress-induced variant of acetylcholinester-

ase, AChE-R, contribute together to the balance between

signals promoting and suppressing the proliferation of glio-

blastoma cells. AChE-R enhances proliferation in a manner

independent of its catalytic activity, probably transduced by

either PKC- or PKA-mediated signaling pathways, and sup-

pressible by CREB as well as by an AChE-R–targeted

antisense agent. In glioblastoma cells, AChE-R interacts

with RACK1 and PKCq in a triple complex that differs from

the PKCbII-including complex of PC12 cells. Our findings are

compatible with the assumption that in glioblastoma cells

under acute situations, associated with extreme excess of

AChE-R, CREB’s regulation may fail to prevent uncontrolled

proliferation.

Transcriptional Regulation Of AChE-R–Induced Proliferation

Our findings suggest an antimitogenic role for CREB in

astrocytes and point to an intrinsic transcriptional regulation

mechanism over AChE-R–mediated proliferation. CREB

Figure 3. Enhanced PKCe phosphorylation under an excess of AChE-R, but not of AChE-S. (A) Western blot analysis showing immunodetection with anti –

phosphorylated PKCe antibodies in U87MG extracts from three different transfections with either AChE-R or AChE-S vectors. Densitometric quantification of the

p-PKCe signal is presented in the lower panel. (B) AChE’s catalytic activity (measured by accumulation of hydrolyzed acetylthiocholine, ATCh) in corresponding cell

extracts.



is a plasticity-associated transcription factor, mediating re-

sponses to various neurotransmitters, mitogenic factors,

and differentiating factors [6]. CREB promotes proliferation

and survival of neurons and glia in the injured brain [20] and

mediates cell viability during early embryonic development

[21]. However, in smooth muscle cells, CREB activation (by

Ser-133 phosphorylation) associates with suppressed ex-

pression of multiple cell cycle regulatory genes and reduced

proliferation [6,22]. Thus, CREB may operate either as an

inducer or as a suppressor of gene expression, depending

on the signal pathway promoting its activation.

Antisense Suppression of the AChE-R Proliferative Effect

EN101 is a 2V-oxymethylated antisense oligonucleotide,

which targets a common site on the exon 2–encoded part of

AChE mRNA. EN101 selectively induces destruction of the

unstable AChE-R mRNA transcript, possibly because it can

interact only with newly transcribed AChE mRNA chains.

Whereas the relatively stable AChE-S mRNA is protected

from degradation in translatable complexes, the rapidly

emerging AChE-R mRNA transcripts are destroyed before

having the chance to get protected. Selective AChE-R

mRNA destruction by EN101 was demonstrated in the

mouse [13], rat [9], and human clinical studies [23]. Nano-

molar doses of such antisense agents attenuated cell prolif-

eration in cultured osteosarcoma cells (SaOs-2) [12] and

human hematopoietic progenitor cells [24]. Here, we report

that EN101 was able to significantly suppress the AChE-R

proliferative effect in cultured glioblastoma cells, suggesting

a role for AChE in the pathogenesis of various tumors.

Although EN101 is currently being tested in the UK and

Israel for its capacity to improve neuromuscular functioning

in myasthenic patients [23], over a dozen anti– tumor anti-

sense drugs are currently being tested for treating different

tumors, at different phases of clinical trials [25]. Further re-

search will be required to test the anti–neoplastic utility of

EN101 in the treatment of glioblastoma and/or other tumors.

Attributing AChE-R–Induced Proliferation to PKCq

AChE-R interaction with PKCbII and its scaffold protein,

RACK1, was recently reported to mediate extended conflict

behavior [13]. In glioblastoma U87MG cells, PKCq, but not

PKCb, forms triple complexes with RACK1 and AChE. The

RACK1-mediated interaction between AChE and a specific

PKC isoform thus appears to be cell type–restricted. Al-

though both PKCb and PKCq were detected in PC12 cells,

only PKCb formed AChE–RACK1 complexes in these cells,

suggesting that PKCq would interact with AChE–RACK1

complexes only in the absence of PKCb. Another possibility

is that because the complexes were formed with endoge-

nous AChE, rather than with an overexpressed specific

variant, difference in the composition of AChE variants may

contribute to the formation of these triple complexes.

In human glioblastoma cells, induction of a dominant-

negative PKCq mutant blocked cell proliferation [15]. PKCq,

furthermore, contributes to tumor development and cell

invasion in prostate cancer [26] and induces glial cell metas-

tasis by activating Erk to mediate integrin-dependent

Figure 4. CREB release of AChE-R – induced proliferation under PKC

inhibition. (A) Shown is the outcome of U87MG cell transfections with CREB,

AChE-R, or AChE-S vectors, alone (top) or together (bottom), following

treatment with either 10 �M H-89, a specific PKA inhibitor (hatched bars), or

10 �M BIM, a specific PKC inhibitor (filled bars). Columns represent percent

increase of BrdU incorporation over control ± SEM; average of four duplicate

transfections. *Statistically significant difference from control (P < .05,

ANOVA). (B) CREB phosphorylation increases in U87MG cells under

cotransfection with CREB and AChE-R. Immunoblot analysis was with

antibodies specific for Ser-133 –phosphorylated CREB. A densitometric

quantification of the p-CREB signal is shown. (C) Proposed mechanism for

AChE-R – induced glioblastoma cell proliferation.



adhesion and cell migration [27]. RACK1 links PKCq to

integrin b chains [28], suggesting its involvement with these

events.

PKCq’s activation and ability to respond to secondary mes-

sengers require specific phosphorylation [29]. In this study,

we show an association between PKCq phosphorylation and

AChE-R overexpression in glioblastoma cells. The AChE-R–

PKCq interaction in glioblastoma cells, as well as PKCq’s

facilitated phosphorylation, thus highlight AChE-R’s involve-

ment in a signaling pathway associated with tumor cell

proliferation and aggressiveness, supporting our notion of

AChE-R’s role in glioblastoma tumorigenesis.

AChE-R–Induced Proliferation Involves Phosphorylation,

But Not Cholinergic Signaling

Acetylcholine stimulates proliferation of rat cortical astro-

cytes and human astrocytoma cells by activating muscarinic

acetylcholine receptors [30,31]. MAPK, PKCq, and ~ were

also implicated in this process [32,33]. However, the more

aggressive glioblastoma cells lack functioning acetylcholine

receptors [34]. Together with our finding that AChE-R–

induced proliferation is refractory to DFP inhibition of AChE’s

catalytic activity, this suggests that glioblastoma proliferation

dissociates from cholinergic pathways. Within transfected

cells, the C-terminus of AChE-R confines it to the cytoplasm

[8]. There is also recent immunocytochemical evidence

demonstrating the cytosolic localization of neuronal AChE-R

[35]. To promote cell proliferation, signal transduction

pathways are hence required for inducing nuclear activation

of cell cycle effectors. We found that AChE-R–induced

proliferation involves both PKA and PKC signals. The fact

that both PKC and PKA inhibition completely abolished

AChE-R–induced proliferation raises a possibility that calls

for further exploration—that these pathways are being acti-

vated in succession following AChE-R’s signal. Moreover, as

signaling pathways can be activated and inactivated by a

number of intermediates, other pathways and mediators

should be further investigated as well.

Suppressed AChE-R Proliferative Effect Under PKC-

Induced CREB Signaling

Various growth factors, stress signals, and kinases, in-

cluding PKA, PKC, calcium/calmodulin kinase 2, and MAPK-

activated protein (MAPKAP), promote CREB activation by

Ser-133 phosphorylation, resulting in complex, at times

diverse, cellular outcomes, including cell proliferation or

quiescence, which are context- and activator-dependent

[6,11,36]. PKC-mediated CREB activation induced prolifer-

ation of early oligodendrocytes [37]. Although PKA-depen-

dent CREB activation promotes astroglia differentiation [5], it

is required for Schwann cells proliferation [38]. CREB’s

suppression of AChE-R–induced proliferation was associat-

ed with CREB/Ser-133 phosphorylation and was revoked by

PKC—but not by PKA—inhibitors, suggesting that CREB’s

antimitogenic effects are PKC-mediated, but also that

AChE-R may induce proliferation through a PKA-dependent

pathway. This is compatible with the assumption that

AChE-R–induced PKA-mediated proliferation, which would

be redundant in the presence of effective PKC-mediated

CREB activation and phosphorylation under low endoge-

nous CREB levels, can become fully expressed under PKC

inhibition and CREB overexpression: an increase in AChE-R

levels results in a PKC-mediated (according to our data,

PKCq-mediated) cell proliferation. This PKC activation is

probably PKA-dependent, as both PKC and PKA inhibitors

abolished AChE-R’s effects, suggesting that these path-

ways are activated in succession. However, because PKC

may be activated through other pathways as well, this

calls for further investigation in order to show direct PKA-

dependent PKC activation upon AChE-R signal.

PKA may also mediate cell proliferation through a CREB-

dependent pathway, as reported by others [39,40]. Our data

suggest that when activated by PKC, CREB suppresses

AChE-R–induced cell proliferation, unless AChE-R is in

extreme excess as compared to CREB’s endogenous levels.

Under PKC inhibition, both AChE-R–PKC–mediated prolif-

eration as well as PKC-mediated CREB’s inhibitory effect

are attenuated. Yet, under AChE-R excess, PKC inhibition

results in a net proliferative effect probably because AChE-R

induces proliferation through a non–PKC-dependent—

perhaps a PKA-mediated—pathway. Nevertheless, other

signaling pathways may be involved in this effect as well

(Figure 4, scheme).

In conclusion, our findings are compatible with the hy-

pothesis that CREB’s basal levels are insufficient to block the

AChE-R proliferative effect in cells with extreme excess of

AChE-R compared to CREB (e.g., under AChE-R transfec-

tion). This may increase the risk for glial tumor growth in

individuals exposed to anticholinesterases or head trauma,

both shown to induce massive AChE-R overexpression [7].

Acknowledgements

We are grateful to Alexander Honigman and Alexander

Levitzki (Jerusalem) for the CREB plasmid and U87MG

glioblastoma cells. Chava Perry, MD, is the incumbent of a

basic research fellowship from the Israel Ministry of Health.

References[1] De Angelis LM (2001). Brain tumors. N Engl J Med 344, 114 – 123.

[2] Mischel PS and Cloughesy TF (2003). Targeted molecular therapy of

GBM. Brain Pathol 13, 52 –61.

[3] Lokker NA, Sullivan CM, Hollenbach SJ, Israel MA, and Giese NA

(2002). Platelet-derived growth factor (PDGF) autocrine signaling reg-

ulates survival and mitogenic pathways in glioblastoma cells: evidence

that the novel PDGF-C and PDGF-D ligands may play a role in the

development of brain tumors. Cancer Res 62, 3729 –3735.

[4] Mandil R, Ashkenazi E, Blass M, Kronfeld I, Kazimirsky G,

Rosenthal G, Umansky F, Lorenzo PS, Blumberg PM, and Brodie

C (2001). Protein kinase C alpha and protein kinase C delta play

opposite roles in the proliferation and apoptosis of glioma cells.

Cancer Res 61, 4612 –4619.

[5] Bayatti N and Engele J (2001). Cyclic AMP modulates the response of

central nervous system glia to fibroblast growth factor-2 by redirecting

signalling pathways. J Neurochem 78, 972 – 980.

[6] Harris AL (2002). Hypoxia—a key regulatory factor in tumour growth.

Nat Rev Cancer 2, 38 –47.

[7] Soreq H and Seidman S (2001). Acetylcholinesterase—new roles for

an old actor. Nat Rev Neurosci 2, 294 – 302.

[8] Brenner T, Hamra-Amitay Y, Evron T, Boneva N, Seidman S, and



Soreq H (2003). The role of readthrough acetylcholinesterase in the

pathophysiology of myasthenia gravis. FASEB J 17, 214 – 222.

[9] Kaufer D, Friedman A, Seidman S, and Soreq H (1998). Acute stress

facilitates long-lasting changes in cholinergic gene expression [see

comments]. Nature 393, 373 –377.

[10] Perry C, Sklan EH, Birikh K, Shapira M, Trejo L, Eldor A, and Soreq H

(2002). Complex regulation of acetylcholinesterase gene expression in

human brain tumors. Oncogene 21, 8428 – 8441.

[11] Goren I, Tavor E, Goldblum A, and Honigman A (2001). Two cysteine

residues in the DNA-binding domain of CREB control binding to CRE

and CREB-mediated gene expression. J Mol Biol 313, 695 –709.

[12] Grisaru D, Lev-Lehman E, Shapira M, Chaikin E, Lessing JB, Eldor A,

Eckstein F, and Soreq H (1999). Human osteogenesis involves differ-

entiation-dependent increases in the morphogenically active 3V alterna-

tive splicing variant of acetylcholinesterase. Mol Cell Biol 19, 788 –795.

[13] Birikh KR, Sklan EH, Shoham S, and Soreq H (2003). Interaction of

‘‘readthrough’’ acetylcholinesterase with RACK1 and PKCbeta II corre-

lates with intensified fear-induced conflict behavior. Proc Natl Acad Sci

USA 100, 283 –288.

[14] Galyam N, Grisaru D, Grifman M, Melamed-Book N, Eckstein F,

Seidman S, Eldor A, and Soreq H (2001). Complex host cell re-

sponses to antisense suppression of ACHE gene expression. Anti-

sense Nucleic Acid Drug Dev 11, 51 –57.

[15] Sharif TR, Sasakawa N, and Sharif M (2001). Regulated expression

of a dominant negative protein kinase C epsilon mutant inhibits the

proliferation of U-373MG human astrocytoma cells. Int J Mol Med 7,

373 – 380.

[16] Perry C, Eldor A, and Soreq H (2002). Runx1/AML1 in leukemia:

disrupted association with diverse protein partners. Leuk Res 26,

221 – 228.

[17] Razon N, Soreq H, Roth E, Bartal A, and Silman I (1984). Character-

ization of activities and forms of cholinesterases in human primary brain

tumors. Exp Neurol 84, 681 – 695.

[18] Kawasaki H, Springett GM, Mochizuki N, Toki S, Nakaya M,

Matsuda M, Housman DE, and Graybiel AM (1998). A family of

cAMP-binding proteins that directly activate Rap1. Science 282,

2275 – 2279.

[19] Gekeler V, Boer R, Uberall F, Ise W, Schubert C, Utz I, Hofmann J,

Sanders KH, Schachtele C, and Klemm K (1996). Effects of the

selective bisindolylmaleimide protein kinase C inhibitor GF 109203X

on P-glycoprotein – mediated multidrug resistance. Br J Cancer 74,

897 – 905.

[20] Ong WY, Lim HM, Lim TM, and Lutz B (2000). Kainate-induced neuro-

nal injury leads to persistent phosphorylation of cAMP response ele-

ment-binding protein in glial and endothelial cells in the hippocampus.

Exp Brain Res 131, 178 – 186.

[21] Bleckmann SC, Blendy JA, Rudolph D, Monaghan AP, Schmid W, and

Schutz G (2002). Activating transcription factor 1 and CREB are im-

portant for cell survival during early mouse development. Mol Cell Biol

22, 1919 – 1925.

[22] Klemm DJ, Watson PA, Frid MG, Dempsey EC, Schaack J, Colton LA,

Nesterova A, Stenmark KR, and Reusch JE (2001). cAMP response

element-binding protein content is a molecular determinant of

smooth muscle cell proliferation and migration. J Biol Chem 276,

46132 – 46141.

[23] Argov Z, McKee D, Agus S, Soreq H, Ben-Yoseph O, Brawer S, and

Sussman JD (2003). EN101: a novel antisense therapeutic strategy for

myasthenia gravis (MG) Neurology. In press.

[24] Grisaru D, Deutch V, Shapira M, Pick M, Sternfeld M, Melamed-

Book N, Kaufer D, Galyam N, Gait M, and Owen D (2001). ARP,

a peptide derived from the stress-associated acetylcholinesterase

variant has hematopoietic growth promoting activities. Mol Med 7,

93– 105.

[25] Tamm I, Dorken B, and Hartmann G (2001). Antisense therapy in on-

cology: new hope for an old idea? Lancet 358, 489 – 497.

[26] Wu D, Foreman TL, Gregory CW, McJilton MA, Wescott GG, Ford OH,

Alvey RF, Mohler JL, and Terrian DM (2002). Protein kinase C epsilon

has the potential to advance the recurrence of human prostate cancer.

Cancer Res 62, 2423 – 2429.

[27] Besson A, Davy A, Robbins SM, and Yong VW (2001). Differential

activation of ERKs to focal adhesions by PKC epsilon is required for

PMA-induced adhesion and migration of human glioma cells. Onco-

gene 20, 7398 – 7407.

[28] Besson A, Wilson TL, and Yong VW (2002). The anchoring protein

RACK1 links protein kinase C epsilon to integrin beta chains. Require-

ments for adhesion and motility. J Biol Chem 277, 22073 – 22084.

[29] Keranen LM, Dutil EM, and Newton AC (1995). Protein kinase C is

regulated in vivo by three functionally distinct phosphorylations. Curr

Biol 5, 1394 – 1403.

[30] Guizzetti M, Costa P, Peters J, and Costa LG (1996). Acetylcholine as

a mitogen: muscarinic receptor– mediated proliferation of rat astrocytes

and human astrocytoma cells. Eur J Pharmacol 297, 265 –273.

[31] Gurwitz D, Razon N, Sokolovsky M, and Soreq H (1984). Expression of

muscarinic binding sites in primary human brain tumors. Brain Res 316,

61– 70.

[32] Guizzetti M and Costa LG (2002). Effect of ethanol on protein kinase C

zeta and p70S6 kinase activation by carbachol: a possible mechanism

for ethanol-induced inhibition of glial cell proliferation. J Neurochem 82,

38– 46.

[33] Yagle K, Lu H, Guizzetti M, Moller T, and Costa LG (2001). Activation

of mitogen-activated protein kinase by muscarinic receptors in

astroglial cells: role in DNA synthesis and effect of ethanol. Glia 35,

111 – 120.

[34] Matute C, Arellano RO, Conde-Guerri B, and Miledi R (1992). mRNA

coding for neurotransmitter receptors in a human astrocytoma. Proc

Natl Acad Sci USA 89, 3399 – 3403.

[35] Nijholt I, Farchi N, Kye M, Sklan EH, Shoham S, Verbeure B, Owen D,

Hochner B, Spiess J, and Soreq H (2003). Stress-induced alternative

splicing of acetylcholinesterase results in enhanced fear memory and

long-term potentiation. Mol Psychiatry. In press.

[36] Goodman RH and Smolik S (2000). CBP/p300 in cell growth, trans-

formation, and development. Genes Dev 14, 1553 – 1577.

[37] Afshari FS, Chu AK, and Sato-Bigbee C (2001). Effect of cyclic AMP on

the expression of myelin basic protein species and myelin proteolipid

protein in committed oligodendrocytes: differential involvement of the

transcription factor CREB. J Neurosci Res 66, 37 – 45.

[38] Lee MM, Badache A, and DeVries GH (1999). Phosphorylation of

CREB in axon-induced Schwann cell proliferation. J Neurosci Res 55,

702 – 712.

[39] Leicht M, Greipel N, and Zimmer H (2000). Comitogenic effect of cat-

echolamines on rat cardiac fibroblasts in culture. Cardiovasc Res 48,

274 – 284.

[40] Rosenberg D, Groussin L, Jullian E, Perlemoine K, Bertagna X, and

Bertherat J (2002). Role of the PKA-regulated transcription factor

CREB in development and tumorigenesis of endocrine tissues. Ann

NY Acad Sci 968, 65 – 74.



CREB Regulates AChE-R-Induced Proliferation of Human Glioblastoma Cells

Documents