Báo cáo khoa học: Adenine nucleotides inhibit proliferation of the human lung adenocarcinoma cell line LXF-289 by activation of nuclear factor jB1 and mitogen-activated protein kinase pathways doc

lung adenocarcinoma cell line LXF-289 by activation of nuclear factor jB1 and mitogen-activated protein kinase pathways Rainer Scha¨fer1, Roland Hartig2, Fariba Sedehizade1, Tobias Welte

lung adenocarcinoma cell line LXF-289 by activation of nuclear factor jB1 and mitogen-activated protein kinase pathways

Rainer Scha¨fer1, Roland Hartig2, Fariba Sedehizade1, Tobias Welte3and Georg Reiser1

1 Institut fu¨r Neurobiochemie, Otto-von-Guericke-Universita¨t, Medizinische Fakulta¨t, Magdeburg, Germany

2 Institut fu¨r Immunologie, Otto-von-Guericke-Universita¨t, Medizinische Fakulta¨t, Magdeburg, Germany

3 Klinik fu¨r Pneumologie, Medizinische Hochschule Hannover, Germany

Keywords

cell cycle progression; mitogen-activated

protein kinase; nuclear factor-jB1; P2Y

receptors; phosphatidylinositol-3-kinase;

protein kinase C

Correspondence

G Reiser, Otto-von-Guericke-Universita¨t

Magdeburg, Medizinische Fakulta¨t, Institut

fu¨r Neurobiochemie, Leipziger Str 44;

39120 Magdeburg, Germany

Fax: +49 391 6713097

Tel: +49 391 6713088

E-mail: georg.reiser@medizin.

uni-magdeburg.de

(Received 4 February 2006, revised 12 June

2006, accepted 16 June 2006)

doi:10.1111/j.1742-4658.2006.05384.x

Extracellular nucleotides have a profound role in the regulation of the pro-liferation of diseased tissue We studied how extracellular nucleotides regu-late the proliferation of LXF-289 cells, the adenocarcinoma-derived cell line from human lung bronchial tumor ATP and ADP strongly inhibited LXF-289 cell proliferation The nucleotide potency profile was ATP¼ ADP¼ ATPcS > > UTP, UDP, whereas a,b-methylene-ATP, b,c-methy-lene-ATP, 2¢,3¢-O-(4-benzoylbenzoyl)-ATP, AMP and UMP were inactive The nucleotide potency profile and the total blockade of the ATP-mediated inhibitory effect by the phospholipase C inhibitor U-73122 clearly show that P2Y receptors, but not P2X receptors, control LXF-289 cell prolifer-ation Treatment of proliferating LXF-289 cells with 100 lm ATP or ADP induced significant reduction of cell number and massive accumulation of cells in the S phase Arrest in S phase is also indicated by the enhancement

of the antiproliferative effect of ATP by coapplication of the cytostatic drugs cisplatin, paclitaxel and etoposide Inhibition of LXF-289 cell prolif-eration by ATP was completely reversed by inhibitors of extracellular sig-nal related kinase-activating kinase⁄ extracellular signal related kinase 1 ⁄ 2 (PD98059, U0126), p38 mitogen-activated protein kinase (SB203508), phos-phatidylinositol-3-kinase (wortmannin), and nuclear factor jB1 (SN50) Western blot analysis revealed transient activation of p38 mitogen-activated protein kinase, extracellular signal-related kinase 1⁄ 2, and nuclear factor jB1 and possibly new formation of p50 from its precursor p105 ATP-induced attenuation of LXF-289 cell proliferation was accompanied by transient translocation of p50 nuclear factor jB1 and extracellular signal-related kinase 1⁄ 2 to the nucleus in a similar time period In summary, inhibition of LXF-289 cell proliferation is mediated via P2Y receptors by activation of multiple mitogen-activated protein kinase pathways and nuclear factor jB1, arresting the cells in the S phase

Abbreviations

a,b-MeATP, a,b-methylene adenosine 5¢-triphosphate; b,c-MeATP, b,c-methylene adenosine 5¢-triphosphate; BrdU, bromodeoxyuridine; Bz-ATP, 2¢,3¢-O-(4-benzoylbenzoyl)adenosine 5¢-triphosphate; CaMKII, calcium ⁄ calmodulin-dependent protein kinase; ERK, extracellular signal-regulated kinase; GPCR, G protein-coupled receptor; MAPK, mitogen-activated protein kinase; MEK1 ⁄ 2, ERK-activating kinase; 2MeS-ADP, 2-methylthioadenosine 5¢-diphosphate; 2MeS-ATP, 2-methylthioadenosine 5¢-triphosphate; NF-jB, nuclear factor jB; NSCLC, nonsmall cell lung cancer; PI3K, phosphatidylinositol-3-kinase; PKC, protein kinase C; PLC, phospholipase C.

There is growing evidence that extracellular nucleotides

can regulate the proliferation of numerous tumorigenic

or nontumorigenic tissues The nucleotide-activated

purinergic P2 receptor family comprises the ionotropic

P2X receptor ion channels and the metabotropic, G

protein-coupled P2Y receptors [1–3] Seven subtypes of

the P2X receptors (P2X1)7) and eight subtypes of the

P2Y receptors (P2Y1,2,4,6,11,12,13,14) have been identified

and functionally characterized [4,5] The regulatory

function of extracellular nucleotides on cell growth,

the role of nucleotides as effectors of neoplastic

transformation and the ubiquitous expression of the

purinergic receptors led to the investigation of the

therapeutic potential of nucleotides and P2 receptors

in clinical trials targeting various diseases (for review

see [6])

ATP is already known to inhibit the growth of

var-ious tumors by activating specific P2 receptors

Inhibi-tion of cancer growth by adenine nucleotides was first

described by Rapaport [7] In vitro extracellular

nucle-otides exert a strong antineoplastic effect on breast,

ovarian [8], skin [9], prostate [10], endometrium [11],

esophagus [12], intestinal [13] and colorectal carcinoma

[14], suggesting that extracellular nucleotides can

sup-press tumorigenesis In view of the antineoplastic

action of extracellular nucleotides [6] and the

signifi-cant in vivo antitumor activity of intraperitoneally

injected ATP against several aggressive carcinomas in

tumor-bearing mice [7,15], it will be important to

understand how the different P2 receptor subtypes

contribute to the regulation of cancer cell

prolifer-ation

In the majority of cell lines, P2Y receptors mediate

the inhibition of growth or proliferation, whereas only

a few cases have been reported where control of

prolif-eration had been claimed to be mediated by P2X

receptors Activation of P2X5 receptors stimulated the

differentiation of skin cancer cells with subsequent

inhibition of proliferation [16], and induction of cell

death was mediated via P2X7 receptors in skin cancer

and prostate cancer cells [16,17] The formation of the

lytic pore of the P2X7receptor is dependent on

coordi-nated signaling involving the p38 mitogen-activated

protein kinase (MAPK) pathway and caspase

activa-tion [18] Several mechanisms have been found for the

inhibitory action of ATP on cancer cell growth In

tumor cell lines from colon [14], endometrium [11],

and esophagus [12], activation of P2Y receptors caused

inhibition of proliferation by induction of cell cycle

arrest and⁄ or apoptosis However, the pathways

involved in P2Y receptor-mediated attenuation of cell

proliferation are not known Therefore, we here

explored which pathways are involved in the

antineo-plastic action of extracellular nucleotides on the human lung adenocarcinoma cell line LXF-289, which

is derived from solid lung adenocarcinoma Extracellu-lar nucleotides strongly attenuated the cell cycle pro-gression of proliferating LXF-289 cells, leading to marked arrest of the cells in the S phase This inhibi-tion is mediated by P2Y receptors through signaling pathways involving activation of the transcription factor nuclear factor jB1 (NF-jB1) and the MAPKs extracellular signal-related kinase 1⁄ 2 (ERK1 ⁄ 2) and p38 Our findings underline the importance of extracel-lular nucleotides in the specific modulation of cell proliferation

Results

Extracellular nucleotides attenuate LXF-289 cell proliferation via P2Y receptors

Single pulses of ATP and ADP equipotently inhibited LXF-289 cell proliferation The effect was concen-tration-dependent, with half-maximal inhibition at 18.6 ± 1.9 lm (n¼ 5) and 19.8 ± 0.6 lm (n ¼ 5), respectively (Fig 1) Significant inhibition was already

nucleotide concentration [µM]

20 40 60 80 100

UDP 2-MeSADP Ado

Fig 1 Concentration-dependent inhibition of LXF-289 cell pro-liferation by different nuccleotides Increasing concentrations (10–100 l M ) of the different nucleotides were added to proliferating LXF-289 cells, and bromodeoxyuridine (BrdU) incorporation was measured after 12 h Values are means ± SD of n ¼ 4–7 experi-ments run in triplicate.

seen at 10 lm ATP (18.2 ± 4.9%; P < 0.01) or ADP

(18.5 ± 3.5%; P < 0.01), which exhibited a reduction

of 55% at 100 lm concentration The pyrimidine

nu-cleotides UTP and UDP reduced the proliferation at

100 lm only by 23–26% (Table 1) ATPcS inhibited

the proliferation of LXF-289 cells as potently as

ADP and ATP (Table 1) Other nucleotides, such as

a,b-methylene-ATP (a,b-MeATP) and b,c-MeATP, which preferentially activate P2X receptors, and AMP

or UMP displayed no activity (Table 1) Moreover, adenosine, which is known to tightly regulate the liferation of tumor cells [19], had no effect on the pro-liferation of LXF-289 cells up to concentrations of

100 lm (Fig 1) Likewise, 2-methylthio-ADP (2-MeS-ADP), 2-methylthio-ATP (2-MeSATP) and 2¢,3¢-O-(4-benzoylbenzoyl)-ATP (Bz-ATP), which most potently activate human P2Y1,12 and P2Y11 receptors, respect-ively [20,21], did not influence the proliferation of LXF-289 cells (Table 1) Thus, the pharmacologic pro-file of the different nucleotides or nucleotide analogs tested strongly indicates that P2Y, but not P2X, recep-tors attenuate LXF-289 cell proliferation

We next examined the expression of P2Y (P2Y1,2,4,6,11,12,13) and P2X (P2X3,4,7) receptors in LXF-289 cells by RT-PCR using primers specific for the human isoforms There was evidence for the expression of mRNA for the P2Y receptor subtypes P2Y2, P2Y6, P2Y11and P2Y13and for the P2X4 recep-tor, whereas expression of the P2Y1, P2Y4, P2Y12, P2X3 and P2X7 receptors could not be detected (Fig 2)

Inhibition of growth and cell cycle progression

of LXF-289 cells

We confirmed that the attenuation of DNA synthesis

by ATP or ADP was indeed due to growth inhibition

of LXF-289 cells Treatment of proliferating LXF-289 cells with 100 lm ATP or ADP reduced the cell num-ber We observed reductions by 19.1 ± 3.8% (± SD;

n¼ 3) and 20.8 ± 4.4% (± SD; n ¼ 3) after 24 h, and by 54.0 ± 4.3% (± SD; n¼ 3) and 54.1 ± 2.6%

Table 1 Inhibition of proliferation of LXF-289 lung tumor cells.

Effects of P2 receptor agonists The influence of the different P2

receptor agonists is shown as percentage inhibition of the

prolifer-ation of LXF-289 cells Bromodeoxyuridine (BrdU) incorporprolifer-ation into

proliferating LXF-289 cells LXF-289 cells in 96-well plates were

incubated for 1 h with 100 l M of the different nucleotides or

nuc-leotide analogs listed below before the addition of 10 l M BrdU.

BrdU incorporation was determined by chemiluminescence as

des-cribed in Experimental procedures, following incubation for 12 h.

Proliferation in the absence of nucleotides is taken as the reference

value Inhibition of that value in percentage is given as means ± SD

of 3–6 independent experiments run in triplicate Bz-ATP,

2¢,3¢-O-(4-benzoylbenzoyl)adenosine 5¢-triphosphate; 2MeS-ADP,

2-methyl-thioadenosine 5¢-diphosphate; 2MeS-ATP, 2-methyl2-methyl-thioadenosine

5¢-triphosphate.

P2 receptor agonist Inhibition of proliferation (%)

P2Y2 P2Y6 P2Y12 P2X3 P2X7

630

Fig 2 Expression of P2 receptors in LXF-289 cells RT-PCR analysis was performed using primers for P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2Y12 and P2Y13 receptors and the P2X receptors P2X3, P2X4 and P2X7 in LXF-289 cells PCR products were separated in a 1% agarose gel and visualized with ethidium bromide GAPDH, glyceraldehyde-3-phosphate dehydrogenase Sizes of RT-PCR products of the different P2 recep-tors are indicated.

(± SD; n¼ 3) after 48 h, respectively These results

confirm the identical potency of ATP and ADP in the

inhibition of LXF-289 cell proliferation

Cell cycle analysis revealed that inhibition of

prolif-eration of LXF-289 cells by ATP and ADP was

medi-ated by retardation of cell cycle progression Flow

cytometry analysis of subconfluent LXF-289 cell

cul-tures showed that treatment with ATP or ADP (10–

100 lm) for 24 h caused a concentration-dependent

increase of cells in the S phase With 100 lm ATP and

ADP, 74.6 ± 3.2% (n¼ 3) and 66.4 ± 1.3% (n ¼ 3)

of the cells, respectively, were arrested in S phase

(Fig 3A) Figure 3B illustrates the changes of the cell

cycle distribution induced by two concentrations of

ATP in LXF-289 cells

ATP sensitizes LXF-289 cells to the activity of

anticancer drugs

Extracellular nucleotides caused accumulation of

LXF-289 cells in the S phase, where cells are especially

sensi-tive to anticancer drugs Therefore, we used cisplatin,

etoposide and paclitaxel to investigate the effects of

these anticancer drugs on LXF-289 cells in the presence

of ATP These anticancer substances inhibit cell cycle

progression by different mechanisms and form the basis

of current chemotherapeutic regimens in lung cancer

treatment Cisplatin, etoposide and paclitaxel

dose-dependently inhibited the proliferation of LXF-289 cells

with IC50 values of 13.8 ± 1.3 lm (n¼ 3), 19.4 ±

2.2 lm (n¼ 3), and 16.2 ± 1.4 nm (n ¼ 3), respectively

(Fig 4) The simultaneous addition of 100 lm ATP

enhanced the antiproliferative potency of cisplatin

3-fold, of etoposide 2-fold and of paclitaxel 2.5-fold,

with resulting IC50 values of 4.7 ± 0.8 lm, 9.3 ±

2.3 lm and 6.8 ± 1.2 nm (Fig 4) The data suggest an

additive effect of the anticancer drugs and ATP on

pro-liferation, and support our conclusion that in LXF-289

cells the ATP-mediated arrest of cell cycle progression

targets the S phase We did not find any significant

induction of DNA fragmentation in LXF-289 cells or

an increase in the number of dead cells under our assay

conditions (data not shown) Moreover, the

concentra-tions of cisplatin and etoposide used do not induce

significant apoptosis in other human lung cancer cells

such as A549 cells, as reported by others [22,23]

Signal transduction pathways involved in

nucleotide-mediated inhibition of proliferation

of LXF-289 cells

Different mechanisms and signal transduction

path-ways transduce control of proliferation by G

protein-coupled receptors (GPCRs) We therefore investigated which of the pathways found in GPCR-mediated attenuation of cell proliferation, including phospho-lipase C (PLC), protein kinase C (PKC) and extra-cellular Ca2+ influx, are involved in the control of LXF-289 cell proliferation by P2Y receptors

At a concentration of 10 lm, the PLC inhibitor U-73122 completely reversed the effect of ATP on

0 20

A

B

40 60 80

control

G0-G1: 65.74%

S-Phase:

26.27%

100 µM ATP

G0-G1: 19.91%

S-Phase:

77.97%

20 µM ATP

G0-G1: 40.40%

S-Phase:

48.39%

0 80 16

0 80 16

0 80 16

a b c

control ATP ADP

Fig 3 Effect of ATP and ADP on cell cycle progression of LXF-289 cells (A) Accumulation of LXF-289 cells in the S phase LXF-289 cells were incubated with ATP or ADP (10–100 l M ) for 24 h, and the distribution of cells in the different cell cycle phases was ana-lyzed by fluorescence-activated cell sorting (FACS) (B) Differences

in the cell cycle distribution of LXF-289 cells Depicted are the per-centages of the cells in the G0 ⁄ G1, S and G2 ⁄ M phases of the cell cycle after treatment with 20 l M or 100 l M ATP for 24 h The gat-ing used for the quantification of the cells in the G0 ⁄ G1 (a), S (b) and G2-M (c) phases is depicted by the respective bar.

LXF-289 cell proliferation, whereas the inactive analog

U-73343 had no effect (Table 2) Blockade of

receptor-operated Ca2+channels with SKF-96365

dose-depend-ently reduced the inhibitory effect of ATP (data not

shown) A reduction of 80% was seen at 50 lm (Table 2) This result and the pharmacologic profile obtained for inhibition of proliferation clearly show that P2Y receptors mediate the attenuation of

LXF-289 cell proliferation

Preincubation of LXF-289 cells with the potent PKC inhibitor Go¨-6983 concentration-dependently abolished the antiproliferative effect of ATP (Table 2)

In addition, we investigated the involvement of the

cisplatin, etoposide (µ M )

0

20

40

60

80

100

+ ATP-Cisplatin + Etoposide + ATP-Etoposide Control + ATP

paclitaxel (n M )

0

20

40

60

80

100

+ ATP-Paclitaxel Control + ATP

A

B

Fig 4 Effect of anticancer drugs and ATP on proliferation of

LXF-289 cells Bromodeoxyuridine (BrdU) incorporation into DNA was

measured in LXF-289 cells (control) or cells incubated with different

concentrations of the anticancer drugs cisplatin or etoposide (A) or

paclitaxel (B) in the absence (solid symbols) or presence (open

symbols) of 100 l M ATP Values (means ± SD of four or five

inde-pendent experiments run in triplicate) are expressed as percentage

of BrdU incorporation into DNA in cells without the addition of

either ATP or the anticancer drugs (basal proliferation ¼ 100%

con-trol) IC50 values were calculated using the SIGMAPLOT curve-fitting

program (SPSS, Chicago, IL, USA).

Table 2 Inhibition of proliferation of LXF-289 lung tumor cells Effects of signaling pathway inhibitors on ATP-inhibited proliferation (A) and of basal proliferation (B) of LXF-289 lung tumor cells Prolif-erating LXF-289 cells in 96-well plates were incubated for 1 h with the various signal transduction pathway inhibitors listed below before the addition of 100 l M ATP (ATP) or medium (B) Bromode-oxyuridine (BrdU) incorporation was then determined by chemilumi-nescence as described in Experimental procedures following incubation for 12 h (A) Data give percentage reduction of the inhibi-tory effect exerted by ATP on BrdU incorporation The inhibition of proliferation of LXF-289 cells by 100 l M ATP (see value in Table 1)

is taken as 100% (B) Basal incorporation of BrdU into proliferating LXF-289 cells in the presence of the inhibitors for the kinases cal-cium ⁄ calmodulin-dependent protein kinase (CaMKII), extracellular signal-regulated kinase 1 ⁄ 2 (ERK1 ⁄ 2), p38 mitogen-activated pro-tein kinase (MAPK), propro-tein kinase C and PI3 kinase All values are means ± SD of 3–6 independent experiments run in triplicate (A)

Signaling pathway inhibitors Concentration

Reduction of inhibitory effect of ATP (%)

(B) Kinase

Reduction of basal proliferation (%)

calcium⁄ calmodulin-dependent kinase II (CaMKII),

which is also known to regulate cell cycle progression

and proliferation Inhibition of CaMKII activity by

KN-62 (25 lm) completely reduced the inhibition by

ATP

The inhibition of LXF-289 cell proliferation by

100 lm ATP was totally attenuated by blockers of

either ERK-activating kinase (MEK1⁄ 2) (20 lm

PD98059, 1 lm U0126), p38 MAPK (20 lm

SB203580) or phosphatidylinositol-3-kinase (PI3K)

(wortmannin), indicating that each pathway seems to

be involved in ATP-mediated inhibition of LXF-289

cell proliferation (Table 2) Inhibition of the MAPK

cascade enzymes MEK1⁄ 2 and p38 MAPK by their

selective inhibitors PD98059⁄ U0126 and SB203580,

respectively, antagonized the nucleotide-mediated

inhi-bition of cell proliferation in a

concentration-depend-ent manner (data not shown)

We further investigated whether the activation of

ERK1⁄ 2 and p38 MAPK may be part of the

antipro-liferative activity of ATP, as these kinases are

import-ant regulators of proliferation as well as apoptosis

Western blotting revealed a time-dependent increase in

both ERK1 and ERK2 phosphorylation with 100 lm

ATP (Fig 5A) ERK1⁄ 2 activation was maximal in

the cytosolic fraction after 10 min and had strongly

declined below the control level after 20 min

Concom-itant with this rapid activation, a transient

transloca-tion of activated ERK1⁄ 2 into the nuclear fraction

was induced by ATP The translocation was maximal

at 10 min and declined thereafter, with no active

ERK1⁄ 2 detectable in the nuclear fraction after 60 min

(Fig 5A)

Exposure of LXF-289 cells to ATP also caused

rapid phosphorylation of p38 MAPK (Fig 5B) Again,

as for ERK1⁄ 2, maximal activation was seen at 10 min

after addition of ATP However, in contrast to

ERK1⁄ 2, no translocation of activated p38 MAPK to

the nucleus could be detected Instead, after a rapid

decline of phosphorylated p38 MAPK down to

unde-tectable levels after 30 min, activated p38 MAPK

reap-peared at 60 min after addition of agonist (Fig 5B)

These results further indicate the involvement of both

MAPKs, ERK1⁄ 2 and p38, in the signaling pathways,

consistent with the inhibition of the antiproliferative

activity of ATP by the MEK1⁄ 2 inhibitors PD98059

and UO126 and the p38 MAPK inhibitor SB20358

(Table 1) The differences in the phosphorylation of

ERK1⁄ 2 and p38 MAPK are not due to altered

expression of ERK1⁄ 2 and p38 MAPK, as western

blotting of total cell lysates did not reveal any change

of either kinase after treatment with 100 lm ATP for

up to 60 min (Fig 5C)

Activation of the transcription factor NF-jB can lead to induced tumor proliferation or to suppression

of tumor growth, resulting in apoptosis [24] Therefore,

we also investigated the involvement of NF-jB in nucleotide-mediated inhibition of LXF-289 cell pro-liferation Preincubation of LXF-289 cells with the NF-jB1 inhibitory peptide NF-jB SN50 (20 lm) com-pletely abrogated the inhibitory effect of 100 lm ATP (Table 2) This suggests that inhibition of LXF-289 cell proliferation by ATP is mediated through NF-jB Fur-ther evidence for the involvement of NF-jB in the signaling pathway activated by P2Y receptors was obtained by the use of nonsteroidal anti-inflammatory drugs, which have been shown to inhibit activation of

kDa 0 10 20 30 60 0 10 20 30 60

Cytosol Nucleus

[min]

50

phospho-ERK 1/2

50 37 A B

C

0 10 30 60 [min]

Cell lysate

kD

ERK 1/2

37

p38

37

RelA (p65)

50

10

NF- κB1

75 50

p105 p50 D

E

0 10 20 30 60 0 10 20 30

Cytosol Nucleus

[min]

kD

Fig 5 ATP-mediated activation and nuclear translocation of extra-cellular signal-related kinase 1 ⁄ 2 (ERK1 ⁄ 2), p38 mitogen-activated protein kinase (MAPK) and nuclear factor jB (NF-jB) LXF-289 cells were incubated with 100 l M ATP, and the incubation was stopped

by aspiration of the medium at the different time points indicated After preparation of subcellular fractions as described in Experimen-tal procedures, proteins were separated by SDS ⁄ PAGE and trans-ferred to nitrocellulose Blots from cytosolic and nuclear fractions were incubated with antibodies to (A) phospho-ERK1 ⁄ 2 (1 : 1000 dilution) and (B) phospho-p38 MAPK (1 : 1000 dilution), and with polyclonal antibodies (1 : 1000 dilution) specific for (D) NF-jB1 (p50 and p105) and (E) RelA (p65) Total cell lysate (C) was incubated with antibodies to total ERK1 ⁄ 2 and p38, respectively Primary anti-bodies were incubated at 4 C overnight, and horseradish peroxi-dase (HRP)-conjugated secondary antibody (1 : 20 000 dilution) was incubated for 1.5 h at room temperature The antibody reaction was visualized with enhanced chemiluminescence The positions of the protein molecular mass markers are indicated in kilodaltons (kDa) The experimental data shown are typical for three independ-ent experimindepend-ents performed with differindepend-ent passages of LXF-289 cells.

NF-jB and cancer cell proliferation [25–27] Curcumin

(20 lm) and sulindac sulfide (10 lm) totally abolished

the ATP-induced inhibition of LXF-289 cell

prolifer-ation (Table 2), without inhibiting LXF-289 cell

prolif-eration at this concentration in the absence of

nucleotides (data not shown) These results further

support the involvement of NF-jB in the P2Y

recep-tor-mediated control of LXF-289 cell proliferation

Western blot analysis showed that incubation of

LXF-289 cells with 100 lm ATP caused a detectable

decrease of the p50 form of NF-jB1 in the cytosol after

10 min, which is still visible after 60 min (Fig 5D)

Simultaneously with the decrease of p50 in the cytosol,

an increase of NF-jB1 p50 in the nuclear fraction

could be observed, starting after 10 min, which

persis-ted after 60 min of nucleotide exposure (Fig 5D)

Sign-aling by ATP also influenced the formation of p50

from its precursor p105 Partial degradation of the

p105 form of NF-jB1, which has an inhibitory

func-tion in the cytosol, like the inhibitor jB (IjB)s, to the

p50 form is visible after 20–30 min (Fig 5D) Thus,

ATP not only induced the translocation of NF-jB1,

but also possibly led to newly formed p50 from the

pre-cursor p105 In addition, ATP induced the

disappear-ance of p65 NF-jB from the cytosol after 20–30 min,

but, surprisingly, no translocation into the nucleus was

observed (Fig 5E) These data suggest that NF-jB1

and p65 have different roles in the cell cycle regulation

of proliferating LXF-289 cells by nucleotides

Discussion

Our data demonstrate an important role of

extracellu-lar nucleotides in the regulation of proliferation of

human lung tumor cells Thus far, it has not been

found that extracellular nucleotide P2Y receptors

downregulate cell cycle progression in epithelial cells

via activation of NF-jB Attenuation of LXF-289 cell

proliferation by ATP is mediated by the activation of

the MEK⁄ ERK1 ⁄ 2, PI3K and p38 MAPK pathways

Activation and translocation of the transcription

fac-tors NF-jB1 (p50) and RelA (p65) was also involved

These pathways lead to a massive arrest of the cells in

the S phase Activation of NF-jB by P2Y receptors

has only been found so far to be associated with the

stimulation of cell proliferation, such as in osteoclasts

[28] or A549 alveolar lung tumor cells [29]

The activity profile of the different nucleotides and

nucleotide analogs used, as well as the complete

inhibi-tion of the antiproliferative effect of ATP by the PLC

inhibitor U-73122, clearly indicate that P2Y receptors,

but not P2X receptors, mediate the inhibition of

LXF-289 lung cell proliferation The inability of

2-MeSATP and Bz-ATP to influence the proliferation

of LXF-289 cells excludes the participation of the P2Y11 receptor subtype in the inhibition Bz-ATP has been shown to activate human P2Y11 receptors, whereas 2-MeSATP activates the human P2Y11 recep-tor with similar potency as ATP [20] The phar-macologic profile seen here for the inhibition of proliferation best fits the characteristics of the subtypes P2Y2 and P2Y13, which were found to be expressed in LXF-289 cells by RT-PCR Thus, our data suggest that these P2Y receptors are involved in the down-regulation of LXF-289 cell proliferation The P2Y2 receptor has been found to be important for the proliferation of stem cells and of human melanomas, and to play a role during differentiation as well as dur-ing development and agdur-ing [9,30–32] However, our finding here that ATP was much more potent than UTP is not consistent with the equipotent activation

of the P2Y2 receptor by ATP and UTP (reviewed in [3]) Similarly, the inactivity of 2-MeSADP, which is more potent than ADP at the human P2Y13 receptor [33], creates another conundrum Similaryl, a nucleo-tide activity profile, which was not consistent with the pattern of P2Y receptors expressed, posed an enigma

in the work by Neary and coworkers, where the stimu-lation of the ERK cascade in rat cortical astrocytes by nucleotide analogs was studied [34]

The control of LXF-289 cell proliferation by purin-ergic metabotropic P2Y receptors involves the activa-tion of multiple pathways One pathway activates the transcription factor NF-jB, a major regulatory factor

of proliferation and apoptosis Attenuation of

LXF-289 cell proliferation by extracellular ATP is mediated

by the induction of nuclear translocation of p50 NF-jB1 and possibly new formation of p50 from its precursor p105 This unique activation of NF-jB during attenuation of cell proliferation through P2Y receptors has been found for the first time Thus far, activation of NF-jB through P2Y or P2X receptors has been associated with stimulation of proliferation of different cell types This has been reported for lung alveolar tumor cells [29], osteoclasts [28], T cells [35] and human monocytes [36] We also found a decrease

of p65 in the cytosolic fraction, but no corresponding accumulation in the nuclear fraction, suggesting a dif-ferent role and site of action for the NF-jB forms p50 and p65 in nucleotide-mediated growth control Gen-etic deletion studies have revealed an important role for NF-jB1 in the regulation of the proliferation and fate of neural progenitor cells [37] Moreover, NF-jB1 p50 plays a central role in the pathogenesis of classical Hodgkin’s lymphoma and in anaplastic large-cell lymphomas, complexed with BCL-3 [38]

In addition to the transcription factor NF-jB1,

extracellular ATP simultaneously activates the MAPK

signaling pathways of p38 MAPK and ERK1⁄ 2 The

simultaneous activation of both MAPK pathways is

indicated by the similar time dependence in the

activa-tion of ERK1⁄ 2 and p38 MAPK Both pathways seem

to be involved in the downregulation of the

prolifer-ation of LXF-289 cells This is indicated by the fact

that blockage of either pathway by the inhibitors for

p38 MAPK and ERK1⁄ 2 completely reversed the

inhibitory effect of ATP on LXF-289 cell proliferation

Inhibitor blockage of either pathway completely

reversed the inhibition of LXF-289 cell proliferation

by ATP Further indirect support for the involvement

of ERK1⁄ 2 is derived from the observed translocation

of ERK1⁄ 2 to the nucleus The translocation of

activa-ted ERK1⁄ 2 to the nucleus is a prerequisite for the

control of cell cycle progression [39,40]

The significance of ERK and p38 phosphorylation

for the activation of specific gene products under these

conditions of downregulation of cell cycle progression

via P2Y receptors still has to be identified There are

multiple targets (e.g transcription factors E2F, ETS,

ATF and AP1, p27(KIP); HBP1, pRB) that can be

phosphorylated by ERK1⁄ 2 and ⁄ or p38 MAPK and

are possibly involved in the promotion of cell cycle

progression and the regulation of the cell cycle

check-points (for reviews see [41–45]) Recent advances

indi-cate a number of links between the activation of p38

kinase and the DNA checkpoint pathways and their

possible interaction in the modulation of cell cycle

con-trol and DNA mismatch repair [46–50] Intriguingly,

another mechanism by which p38 MAPK may

negat-ively regulate the cell cycle is by activation of the

mito-tic spindle assembly checkpoint pathway that monitors

the correct formation of the spindle and attachment of

kinetochores [51,52]

There is extensive crosstalk between the ERK1⁄

2-activated, p38 MAPK-activated and NF-jB-activated

pathways The simultaneous activation of ERK1⁄ 2,

p38 kinase and NF-jB1 has been described only for

inflammatory mediators, such as lipopolysaccharide or

tumor necrosis factor-a (reviewed in [53]), but not for

extracellular nucleotides Thus, in addition to the

con-trol of lymphocyte or macrophage function, the

simul-taneous activation of these pathways may participate

in the attenuation of epithelial cell proliferation by

extracellular nucleotides Moreover, in some systems

GPCRs couple to NF-jB through sequential activation

of conventional PKC isoforms and IjB kinase, leading

to degradation of IjB by the proteasome [54] It is

possible that the P2Y receptors act through PKC or,

alternatively, through PI3K to activate NF-jB in lung

tumor cells However, we do not know whether these pathways converge at a single target, e.g NF-jB, or whether they interfere with the regulation of S phase entry, S phase progression and⁄ or exit from S phase to G2 phase

Until now, P2Y receptor-mediated inhibition of cell proliferation by control of cell cycle progression has been found only in esophageal cells [12] Activation of P2Y receptors by ATP and ADP inhibits LXF-289 cell proliferation very likely through direct attenuation of

S phase progression This is indicated by the massive increase of cells in the S phase, the reduction in cell proliferation, and the decrease in bromodeoxyuridine (BrdU) incorporation into newly synthesized DNA In addition, the additive effect of ATP with that of paclit-axel and etoposide also suggests that ATP inhibits growth by interfering with cell cycle progression at the

S phase Paclitaxel and etoposide exert a block of NSCLC cells at the G2⁄ M and G2 phase, respectively [22,23], whereas cisplatin causes an accumulation of G0⁄ G1 cells [55]

The transient translocation of ERK1⁄ 2 and of p50 NF-jB1 to the nucleus suggests a role of both proteins

in the regulation of cell cycle progression by extracellu-lar nucleotides Sustained activation of the ERK1⁄ 2 and PI3K pathways is not only necessary for cell cycle progression into S phase but also regulates progression during G2⁄ M phase by distinct cell cycle timing requirements [39,56] However, we observed only tran-sient ERK1⁄ 2 activity in the nucleus for up to 60 min, which is much shorter than the sustained activity for ERK1⁄ 2, which persists from late S phase to G2 and

M phases [39] Moreover, nothing is known about a direct link of NF-jB1 with regulation of the S phase checkpoint or S phase progression Thus, for both pro-teins, the mechanistic details of inhibition of cell cycle progression remain unclear

Different mechanisms can be targeted for attenuation

of cell cycle progression Putative targets for nucleotide-regulated cell cycle progression are either (a) the activa-tion of cell cycle-regulated kinases, Cdc7, and cyclin dependent kinase, which are required for the initiation

of DNA replication during S phase, or (b) the recruit-ment of the initiation factor Cdc45, which is required for the elongation phase of replication, or (c) the replica-tion forks (reviewed in [57]) Especially in the last case, extracellular nucleotides may perhaps act as a signal causing replication forks to stall and halt cell cycle pro-gression, similar to energy depletion or DNA damage (see review [57]) However, this needs to be investigated for extracellular nucleotides This is very important to understand, because the stabilization of stalled replica-tion forks is crucial for maintaining cell viability [57]

In summary, our results reveal that extracellular

nucleotides inhibit proliferation of LXF-289 cells by

regulation of cell cycle progression through activation

of several signal transduction pathways The delay in

cell cyle progression through activation of P2Y

recep-tors is possibly mediated by activation of the NF-jB1

isoform of transcription factor NF-jB, the MAPKs

ERK1⁄ 2 and p38 and PI3K Thus, extracellular

nucle-otides need to be added to the class of the critical

reg-ulators of cell cycle progression, so far consisting of

hormones, growth factors, and cytokines Further

detailed investigations are needed to unravel the

sub-type of P2Y receptor mediating the observed response

and the functionality of the proteins in the signaling

mechanisms that lead to the attenuation of cell cycle

progression

Experimental procedures

Cell culture and reagents

The human lung adenocarcinoma cell line LXF-289,

estab-lished from the solid lung tumor of a 63-year-old man, was

obtained from the German Collection of Microorganisms

and Cell Cultures (DSMZ, Braunschweig, Germany)

LXF-289 cells were cultured in Ham’s F-10 (Biochrom, Berlin,

Germany) supplemented with 10% FBS, 100 IUÆmL)1

peni-cillin, and 100 lgÆmL)1streptomycin in a 5% CO2

incuba-tor at 37C Paclitaxel, etoposide and cisplatin were

purchased from Calbiochem (Bad Soden, Germany) The

PKC inhibitors Go¨6983 (Alexis, Gru¨nberg, Germany),

U-73122, U-73343, SB-203580, wortmannin (Sigma-Aldrich,

Taufkirchen, Germany), KN-62, PD-908059 (Bio-Trend,

Ko¨ln, Germany), SKF-96365 and NF-jB SN-50

(Calbio-chem) were dissolved in dimethyl sulfoxide or in NaCl⁄ Pi

to give stock concentrations of 10 or 100 mm

Measurement of cell proliferation

Cell proliferation was measured by luminometric

immuno-assay based on BrdU incorporation during DNA synthesis

using a cell proliferation ELISA BrdU Kit (Roche,

Mann-heim, Germany) according to the manufacturer’s protocol

The effect of extracellular nucleotides on BrdU

incorpor-ation was measured in LXF-289 cells, as described [29]

Cells were seeded on 96-well plates (5· 103cells per well)

and then incubated for 24 h Cells were then incubated with

100 lm (standard conditions) of the different nucleotides or

nucleotide analogs for 1 h in a final volume of 100 lL per

well Antagonists were added to the cells 1 h before the

addition of the nucleotides Subsequently, 10 lL of

BrdU-labeling solution was added to each well and the cells were

incubated again for 12 h BrdU labeling was determined as

previously described [29]

Cell growth was determined in subconfluent LXF-289 cells seeded into 12-well plates (5· 104cells per well;

500 lL per well) At 24 h after seeding, cells were treated with 100 lm ATP or ADP for 1 or 2 days Antagonists were added to the cells 1 h before the addition of the nucleo-tides Cells were washed twice with NaCl⁄ Pi, and dispersed using trypsin-EDTA, and suspended cells were counted using a hemacytometer The percentage inhibition of cell proliferation by nucleotide treatment was calculated by comparison with control cells Trypan blue was used to determine cell viability

We ascertained in all cases that the inhibitor concentra-tions used did not affect basal DNA synthesis Therefore,

we tested the effect of the compounds U-73122, U-73343, SB-203580, KN-62, PD-908059, SKF-96365 and NF-jB SN-50 in a concentration range from 0.1 to 100 lm and the inhibitors Go¨6983, U0126 and wortmannin in a concentra-tion range from 0.01 to 10 lm Maximal concentraconcentra-tions, which did not influence basal DNA synthesis, were used in the proliferation tests with the nucleotides The tion-dependent inhibition (1–100-fold of initial concentra-tion) of the antiproliferative effect of ATP was tested for the inhibitors U-73122, PD98059, SB203580, SKF-96365 and Go¨6983 (data not shown)

Cell cycle analysis

LXF-289 cells (5· 104cells per well) were seeded in 12-well plates, cultured for 24 h and exposed to the respect-ive nucleotide for another 24 h or 48 h Trypsinized cells were pelleted at 300 g, washed once with 0.5 mL of ice-cold NaCl⁄ Pi and fixed in 75% ethanol at ) 20 C for

60 min Cells were treated with RNaseA (200 lgÆmL)1) and propidium iodide (50 lgÆmL)1) in NaCl⁄ Pi at 25C for 30 min, and this followed by flow cytometry (FAC-Scan; Becton Dickinson, Franklin Lakes, NJ) Distribu-tion of the cells in G1, S and G2⁄ M phases and apoptotic populations (sub-G1 phase) were analyzed by the modfit program (Verity Software House Inc., Topsham, ME, USA)

Determination of P2 receptor expression

Expression of mRNA for different P2Y (P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2Y12, and P2Y13) and P2X (P2X3, P2X4, and P2X7) receptors in LXF-289 cells was determined by RT-PCR Total RNA was isolated from LXF-289 cells, as previ-ously described [29], with the RNeasy kit (Qiagen, Hilden, Germany) Possible genomic DNA contamination was excluded in experiments by omitting the reverse transcrip-tase in the PCR Sets of specific oligonucleotide primers were synthesized based on the published sequences for the different P2Y receptors (see below) Primer sequences for P2Y1, P2Y2, P2Y4 and P2Y6 receptors were as described

[29] Amplification was performed with 1 lL of cDNA,

for 30 cycles The sequences for the primers were

5¢-CGA GGT GCC AAG TCC TGC CCT-3¢ (forward,

posi-tions 7–27) and 5¢-CGC CGA GCA TCC ACG TTG

AGC-3¢ (reverse, positions 798–818) with 812 bp for

hP2Y11 (accession no AF030335), 5¢-CCA GTC TGT

GCA CCA GAG ACT-3¢ (forward, positions 115–135) and

5¢-ATG CCA GACTAG ACC GAA CTC-3¢ (reverse,

posi-tions 615–635) with 520 bp for hP2Y12 (accession no

AF313449), and 5¢-GGT GAC ACT GGA AGC AAT

GAA-3¢ (forward, positions 67–78) and 5¢-GAT GAT

CTT GAG GAA TCT GTC-3¢ (reverse, positions 437–457)

with 391 bp for hP2Y13 (accession no NM176894) The

primers for the P2X3,4,7 were 5¢-AGT CGG TGG TTG

TGA AGA GCT-3¢ (forward, positions 41–61) and 5¢-AAG

TTC TCA GCT TCC ATC ATG-3¢ (reverse, positions 492–

512) with 472 bp for hP2X3 (accession no AB016608),

5¢-GCC TTC CTG TTC GAG TAC GAC-3¢ (forward,

posi-tions 1951–71) and 5¢-CGC ACC TGC CTG TTG AGA

CTC-3¢ (reverse, positions 2351–2371) with 421 bp for

hP2X4 (accession no AF191093), and 5¢-GTC ACT CGG

ATC CAG AGC ATG-3¢ (forward, positions 148–168)

and 5¢-TTG TTC TTG ATG AGC ACA GTG-3¢ (reverse,

positions 660–680) with 533 bp for hP2X7 (accession no

NM002562)

Preparation of cell extracts and western blot

analysis

LXF-289 cells, seeded in 55 mm dishes, were incubated

with 100 lm ATP at 37C for various times Cells were

lysed with hypotonic buffer (20 mm Tris⁄ HCl, pH 8.1,

1 mm Mg2+, 0.05 mm Ca2+) containing an inhibitory

cock-tail for proteases and phosphatases (Complete EDTA;

Roche) and 2 mm NaVO4(hereafter named lysis buffer) by

incubation on ice for 15 min, dispersed by sonication and

then centrifuged at 1000 g for 10 min For all

centrifuga-tions an Avanti 30 centrifuge from Beckman Coulter

(Kre-feld, Germany) was used The resulting pellet, containing

nuclei and mitochondria, was resuspended in lysis buffer

containing 400 mm sucrose, layered on a sucrose cushion of

1.2 m sucrose in lysis buffer and centrifuged for 15 min at

5000 g to purify the nuclei Purified nuclei were

resuspend-ed in lysis buffer containing 0.1% (v⁄ v) Nonidet P-40,

2 mm dithiothreitol and 0.4 m NaCl, incubated on ice for

30 min to extract the proteins, and then centrifuged at

12 000 g for 10 min to remove nonsolubilized debris The

resulting supernatant (nuclear fraction) was stored at

) 80 C for further investigations The supernatant from

the first centrifugation (1000 g) was then centrifuged at

50 000 g for 45 min to prepare the cytosolic fraction

(resul-tant superna(resul-tant) The protein content was determined

using the BCA protein assay kit (Pierce, Rockford, IL,

USA) Protein fractions were subjected to SDS gel

electro-phoresis on 10% acrylamide gels and transferred to PVDF

membranes To control the complete and even transfer of proteins to the blotting membranes, membranes were immersed in a Ponceau Red-containing solution and the stained protein bands were scanned densitometrically before processing with antibodies Blots were incubated at 4C overnight with anti-human phospho-ERK1⁄ 2 (Thr183, Tyr185), anti-phospho-p38 MAPK (Cell Signaling Technol-ogy, Beverley, MA, USA) or antibodies for NF-jB1 (p50, p105) or RelA (p65) After incubation with horseradish per-oxidase-conjugated anti-rabbit IgG (Dianova, Hamburg, Germany) for 1.5 h at room temperature, the antibody reaction was visualized by enhanced chemiluminescence (Promega, Madison, WI, USA)

Statistical analyses

Experiments were conducted with cultures from different seedings The number of replica experiment is given in the figure legends Data were analyzed by Student’s t-tests for two groups or anova for multiple groups

Acknowledgements

We thank Annette Schulze for skillful technical assist-ance in the experiments The work was supported by Deutsche Krebshilfe (project 10-1754) and Land Sach-sen-Anhalt

References

1 Abbracchio MP & Burnstock G (1998) Purinergic sig-nalling: pathophysiological roles Jpn J Pharmacol 78, 113–145

2 Burnstock G & Williams M (2000) P2 purinergic recep-tors: modulation of cell function and therapeutic poten-tial J Pharmacol Exp Ther 295, 862–869

3 von Ku¨gelgen I & Wetter A (2000) Molecular pharma-cology of P2Y-receptors Naunyn Schmiedebergs Arch Pharmacol 362, 310–323

4 Dubyak GR (2003) Knock-out mice reveal

tissue-speci-fic roles of P2Y receptor subtypes in different epithelia Mol Pharmacol 63, 773–776

5 Khakh BS (2001) Molecular physiology of P2X recep-tors and ATP signalling at synapses Nat Rev Neurosci

2, 165–174

6 Burnstock G (2002) Potential therapeutic targets in the rapidly expanding field of purinergic signalling Clin Med 2, 45–53

7 Rapaport E (1983) Treatment of human tumor cells with ADP or ATP yields arrest of growth in the S phase

of the cell cycle J Cell Physiol 114, 279–283

8 Popper LD & Batra S (1993) Calcium mobilization and cell proliferation activated by extracellular ATP in human ovarian tumour cells Cell Calcium 14, 209–218