Báo cáo khoa học: Upregulation of endothelial adhesion molecules by lysophosphatidylcholine pptx

Abbreviations AC, adenylyl cyclase; ACREB, dominant-negative mutant CREB protein; AM, adhesion molecule; CAPE, caffeic acid phenethyl ester; CRE, cAMP response element; CREB, cAMP respon

by lysophosphatidylcholine

Involvement of G protein-coupled receptor GPR4

Yani Zou1, Chul H Kim1, Jae H Chung1, Ji Y Kim1, Sang W Chung1, Mi K Kim2, Dong S Im1, Jaewon Lee1, Byung P Yu2,3and Hae Y Chung1,2

1 College of Pharmacy, Pusan National University, Busan, Korea

2 Longevity Life Science & Technology Institute, Pusan National University, Busan, Korea

3 Department of Physiology, University of Texas Health Science Center, San Antonio, TX, USA

Lysophosphatidylcholine (LPC), a derivative of

phos-phatidylcholine, is generated through the hydrolytic

action of phospholipase A2 at the sn-2 position of

phosphatidylcholine In vivo, LPC is claimed to be the major effective component of oxidized low-density lipoprotein [1,2], and is found in high concentrations

Keywords

adhesion molecules; cAMP response

element-binding protein; G protein-coupled

receptor 4; lysophosphatidylcholine; nuclear

factor kappaB

Correspondence

H Y Chung, Longevity Life Science &

Technology Institute, College of Pharmacy,

Pusan National University, 30 Jangjun-dong,

Gumjung-gu, Busan 609-735, Korea

Fax: +82 51 518 2821

Tel: +82 51 510 2814

E-mail: hyjung@pusan.ac.kr

(Received 2 December 2006, revised 11

February 2007, accepted 15 March 2007)

doi:10.1111/j.1742-4658.2007.05792.x

Lysophosphatidylcholine induces expression of adhesion molecules; how-ever, the underlying molecular mechanisms of this are not well elucidated

In this study, the intracellular signaling by which lysophosphatidylcholine upregulates vascular cell adhesion molecule-1 and P-selectin was delineated using YPEN-1 and HEK293T cells The results showed that lysophos-phatidylcholine dose-dependently induced expression of vascular cell adhesion molecule-1 and P-selectin, accompanied by the activation of tran-scription factor nuclear factor jB However, the nuclear factor jB inhibitor caffeic acid phenethyl ester (CAPE) and the antioxidant N-acetylcysteine only partially blocked lysophosphatidylcholine-induced adhesion molecules Subsequently, we found that the lysophosphatidylcholine receptor G pro-tein-coupled receptor 4 (GPK4) was expressed in YPEN-1 cells and trig-gered the cAMP⁄ protein kinase A ⁄ cAMP response element-binding protein pathway, resulting in upregulation of adhesion molecules Further evidence showed that overexpression of human GPK4 enhanced lysophosphatidyl-choline-induced expression of adhesion molecules in YPEN-1 cells, and enabled HEK293T cells to express adhesion molecules in response to lysophosphatidylcholine In conclusion, the current study suggested two pathways by which lysophosphatidylcholine regulates the expression

of adhesion molecules, the lysophosphatidylcholine⁄ nuclear factor-jB ⁄ adhesion molecule and lysophosphatidylcholine⁄ GPK4 ⁄ cAMP ⁄ protein kin-ase A⁄ cAMP response element-binding protein ⁄ adhesion molecule path-ways, emphasizing the importance of the lysophosphatidylcholine receptor

in regulating endothelial cell function

Abbreviations

AC, adenylyl cyclase; ACREB, dominant-negative mutant CREB protein; AM, adhesion molecule; CAPE, caffeic acid phenethyl ester; CRE, cAMP response element; CREB, cAMP response element-binding protein; EC, endothelial cell; ERK, extracellular signal-related kinase; FSK, forskolin; G2A, G2 accumulation protein; GPR4, G protein-coupled receptor 4; GPR119, G protein-coupled receptor 119; hGPR4, human GPR4 expression vector; LPC, lysophosphatidylcholine; MDL, MDL12330A; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide; NAC, N-acetylcysteine; NF-jB, nuclear factor-kappaB; PKA, protein kinase A; PKC, protein kinase C; TNF-a, tumor necrosis factor-a; VCAM-1, vascular cell adhesion molecule-1.

in lesional psoriatic skin [3], asthma and rhinitis [4],

and atherosclerosis lesions [5] In addition, LPC has

been reported to induce expression of cell adhesion

molecules (AMs), which form the framework for

leu-kocyte–endothelium binding [6] and subsequent

infil-tration of leukocytes across the endothelium, the initial

step in atherosclerotic changes The available data

indicate that LPC increases the expression of vascular

cell adhesion molecule-1 (VCAM-1) [7], intercellular

adhesion molecule-1 [8] and P-selectin [9] both in vitro

and in vivo However, the intracellular signal pathway

of these biological effects is still not well characterized

Earlier studies have suggested several possible

mech-anisms for LPC’s effects LPC activates the

redox-sensitive transcription factors nuclear factor-kappaB

(NF-jB) [10] and activator protein-1 [11] through the

mitogen-activated protein kinase and protein kinase C

(PKC) pathways LPC is also known to enhance

cAMP response element-binding protein (CREB)⁄

activating transcription factor activity in endothelial

cells (ECs) [12] However, it is still unclear precisely

how LPC triggers these signals Recent studies

demon-strated that low-concentration LPC regulates

activa-tion of G protein-coupled receptors [13], including

G protein-coupled receptor G2 accumulation protein

(G2A), G protein-coupled receptor 4 (GPR4) [14], and

G protein-coupled receptor 119 (GPR119) [15], thus

providing new molecular insights into the effects of

LPC on various signaling activities

In the present study, we attempted to elucidate the

mechanism by which LPC triggers the expression of the

AMs VCAM-1 and P-selectin in the rat EC line It was

found that YPEN-1 was activated to express AMs by

both the NF-jB-mediated and GPR4-mediated

path-ways This finding highlights the biological role of LPC

and provides new molecular insights into the activation

mechanism of LPC in the cellular signaling pathway

Results

Upregulated expression of AMs by LPC

Previous reports have shown that LPC induces the

expression of AMs in ECs [7]; however, the underlying

mechanism was not known Our earlier study showed

that rat endothelial YPEN-1 cells express AMs in

response to proinflammatory cytokines, such as tumor

necrosis factor-a (TNF-a) [16] In this study, YPEN-1

was used to explore the underlying mechanisms of

LPC-induced expression of AMs The expression of

VCAM-1 and P-selectin induced by LPC in YPEN-1

cells was measured by western blot analysis (Fig 1)

When YPEN-1 was incubated with LPC at increasing

concentrations from 3.12 lm to 25 lm for 12 h, expression of VCAM-1 and P-selectin was dose-dependently upregulated, peaking at 12.5 lm (Fig 1A) Moreover, LPC triggered expression of both AMs in a time-dependent pattern (Fig 1B) When the transcrip-tional regulation of AMs by LPC was analyzed, mRNA levels of both AMs were found to be increased

in a time-dependent fashion, which is comparable to TNF-a-induced activation of AMs (Fig 1C), confirm-ing the transcriptional regulation of AMs by LPC

To discriminate the cytotoxic effects of LPC, cell viability was examined The result showed that under the present experimental conditions (i.e cells were maintained in DMEM medium containing 1% fetal bovine serum when they were challenged by LPC), no significant cell death was observed even at a high con-centration (25 lm) of LPC (Fig 1D)

Contribution of NF-jB to LPC-induced activation

of AMs

To define the molecular mechanism of LPC-induced upregulation of AMs, the influence of LPC on NF-jB activation was investigated in YPEN-1 cells This is because the predominant regulatory role of NF-jB is upregulation of VCAM-1 and P-selectin in ECs [17–19] Western blot analyses were performed to examine the nuclear translocation of NF-jB components p65 and p50 As shown in Fig 2A, p65 and p50 translocated from the cytosol into nuclei in response to LPC as early as 10 min This procedure was correlated with increased phosphorylation of cytosolic jB inhibitor (IjBa) (Fig 2A), implying the activation of NF-jB by LPC To uncover the interaction between LPC-induced activation of NF-jB and upregulation of AMs, NF-jB luciferase reporter (NF-jB-Luc) and AM gene promo-ters (VCAM-1 gene promoter and P-selectin gene promoter) were transfected into YPEN-1 cells The luciferase activity was assessed 8 h after treatment of YPEN-1 cells with LPC (12.5 lm), with or without preincubation of the NF-jB inhibitor CAPE (10 lm)

As shown in Fig 2B, LPC increased the activation of NF-jB, as well as the promoter activities of P-selectin and VCAM-1 After preincubation with CAPE (10 lm) for 1 h, activation of NF-jB signaling was significantly blocked (P£ 0.05) and the promoter activity of VCAM-1 was reduced (P£ 0.05) Although the data are not shown, we made certain that these observations did not result from cell death To verify this finding, western blot analysis was carried out, and showed that increased nuclear translocation of p65 was prevented (Fig 2C), and also that VCAM-1 expression was parti-ally reduced by preincubation with CAPE (Fig 2C)

Partial blockage of LPC-induced AMs by the

antioxidant N-acetylcysteine

Given the close association between oxidative stress and

the activation of NF-jB, inhibitory effects of

antioxi-dants on LPC-induced upregulation of AMs could be

expected Thus, the well-known antioxidant

N-acetyl-cysteine (NAC) [20] was used to treat YPEN-1 cells that

were transfected with AM promoter reporters or

NF-jB-Luc After 1 h of preincubation with NAC (1 mm), LPC was added to induce activation in YPEN-1 cells The results showed that although NAC almost completely blocked NF-jB activity (P£ 0.05), it only partially suppressed the activation of VCAM-1 promo-ter (P£ 0.05) (Fig 3) Considering these results and the above findings on the NF-jB inhibitor CAPE, it was suspected that other signal transduction pathways might participate in LPC-induced AM expression

0 2 4 6 8 12 (h) VCAM-1

P-selectin

ββ-Actin

LPC (6.25 μM ) B

D

control 6.25μM 12.5μM

25μM

0 6 12 (h)

60

100

80

120 140 160

60 80 100 120 140 160

VCAM-1 P-selectin

(μM )

*

60 80 100 120 140

VCAM-1 P-selectin

LPC (μM )

0 3.12 6.25 12.5 25 VCAM-1

P-selectin

β-actin A

0 2 4 6 8 8 (h)

LPC (25 μM )

β-Actin

P-selectin

VCAM-1

C

TNF-α

Fig 1 LPC-induced expression of AMs in YPEN-1 cells (A) Expression of AMs induced by LPC in YPEN-1 cells at different concentrations after a 12 h challenge was examined by western blot analysis The quantitative analysis is shown (B) Expression of AMs induced by 6.25 l M LPC at different time points Cells were treated with LPC maintained in 1% fetal bovine serum-containing medium One representa-tive result is shown from three experiments that yielded similar results, and the quantitarepresenta-tive analysis is shown (C) mRNA of AMs induced

by 25 l M LPC was analyzed by RT-PCT at different time points TNF-a was used as the positive control for monitoring the expression of AMs One representative result is shown (D) Cell viability after treatment with LPC at different concentrations was measured by an MTT assay Data were generated from triplicate experiments, and the results are presented as percentage of control at 0 h Statistical signifi-cance: *P £ 0.05, **P £ 0.01 versus cells without LPC treatment.

Cont 10 20 40 60 120 240 (min)

LPC (6.25 μμM )

p65 p50

pI κBα (cytosol)

A

p65 p50 pIkB

Cont 10 20 40 60 120 240 (min)

**

LPC (6.25 μM ) Cont none CAPE VCAM-1

β-actin

p65 (nucleus)

C

UTC

TC

LPC LPC + CAPE

NF- κB luciferase reporter

P-selectin gene promoter VCAM-1 gene promoter

B

Relative RLU (% of TC)

150

*

*

Fig 2 LPC-induced activation of NF-jB in YPEN-1 cells (A) LPC-induced activation of NF-jB was detected by measuring the nuclear translo-cation of NF-jB components p65 and p50, as well as the phosphorylation of the NF-jB inhibitor, IjBa The levels of p65 and p50 in the nuc-leus were examined by western blot analysis using the cell nuclear fraction after challenge with LPC at different time points Histone H1 was used as the nuclear fraction internal calibration The phosphorylation of IjBa was monitored by western blot analysis in the cytosol frac-tion One representative result is shown, and the quantitative analysis is shown beneath it *P £ 0.05, **P £ 0.01 versus control Cont., con-trol (B) Effects of LPC on NF-jB activation, P-selectin promoter activity and VCAM-1 promoter activity in transfected YPEN-1 ECs were detected by luciferase assay Cells were transfected with NF-jB luciferase reporter or P-selectin or VCAM gene promoter plasmids, which contain tandem jB-binding site or murine P-selectin or human VCAM-1 promoters, respectively Twenty-four hours after transfection, cells were treated for 8 h with various reagents, and lysed for determination of luciferase activity UTC, untransfected control; TC, transfected and untreated cells; LPC, transfected cells challenged with 12.5 l M LPC; LPC + CAPE, transfected cells preincubated with the NF-jB inhib-itor CAPE, 10 l M , for 1 h before challenge with 12.5 l M LPC for an additional 7 h Each value is expressed as the mean ± SE from three independent experiments *P £ 0.05 versus TC in the same group RLU, relative light unit (C) LPC-induced expression of VCAM-1 through NF-jB activation was detected by applying the NF-jB inhibitor CAPE YPEN-1 cells were preincubated with CAPE (10 l M ) for 1 h before challenge with 6.25 l M LPC Western blot analysis was carried out to detect the level of VCAM-1 in the cytosol fraction after 12 h of treat-ment with LPC, and the level of p65 in the nucleus was assessed 4 h after treattreat-ment with LPC One representative result is shown None, without inhibitor.

Expression of the LPC receptor GPR4 in YPEN-1

cells

To investigate the additional signaling responsible for

LPC upregulation of AMs, the involvement of LPC

receptors was examined To date, several G

protein-coupled receptors, namely G2A, GPR4, and GPR119,

have been suggested as receptors for LPC and to elicit

various biological effects These receptors show more

specific functions and trigger biological effects at

rela-tively low concentrations of LPC

We analyzed the distribution of LPC receptor

pro-teins in YPEN-1 cells and selected rat spleen tissue as

the positive control, because it is known to produce

both LPC receptors, G2A and GPR4 [21] As shown

in Fig 4A, an interesting finding was that YPEN-1

ECs expressed GPR4 mRNA but not G2A mRNA

This selectivity for GPR4 expression by ECs is

consis-tent with previous reports [21,22] GPR4 is a GPRC

eliciting second messenger cAMP in several cell types

[23–25], so the levels of cAMP were determined in

LPC-challenged YPEN-1 cells As indicated in Fig 4B,

a 15-min incubation with LPC at increasing

concentra-tions (from 6.25 lm to 12.5 lm) caused

dose-depen-dent accumulation of cAMP in YPEN-1 cells

To clarify the downstream signaling of LPC-induced

cAMP accumulation, activation of the transcription

factor CREB was assessed It is well known that

cAMP binds to specific intracellular regulatory

pro-teins such as protein kinase A (PKA) [26], leading to activation of PKA As a consequence, activated PKA phosphorylates Ser133 of the transcription factor CREB, which then binds to the cAMP response ele-ment (CRE) sequence in targeting genes [26] In our study, as shown in Fig 5A, phosphorylation of nuclear CREB was enhanced after challenge with LPC for

30 min; this activation was then sustained for a longer period (Fig 5A)

To delineate LPC-activated CRE signaling, a lucife-rase reporter vector CRE-Luc, containing a CRE site from the human COX-2 gene () 124 bp ⁄ + 59 bp), was applied [27] Forskolin (FSK) (10 lm), an adenylyl cyclase (AC) activator, served as the positive control As shown in Fig 5B, FSK induced significantly high luci-ferase activity in the CRE reporter vector CRE activity was induced by LPC, and this elevation was reduced by

0 10 20 30 40 50 60 70 80 90

Ⴕ G2A (602 bp)

1 2 M 3 4

GPR4 (546 bp)

1,3 YPEN-1 cDNA; 2,4 Rat spleen cDNA;

β-actin

A

*

Control 6.25 12.5 25

B

*

Fig 4 GPR4 expression and LPC-induced cAMP in YPEN-1 cells (A) Expression patterns of the LPC receptors GPR4 and G2A in YPEN-1 cells were analyzed by RT-PCR, using rat spleen tissue as positive control Lane 1: YPEN-1 cell cDNA with GPR4 probe Lane 2: spleen cDNA with GPR4 probe Lane 3: YPEN-1 cell cDNA with G2A probe Lane 4: spleen cDNA with G2A probe M: DNA ladder marker (100 bp) One representative result is shown (B) LPC-induced accumulation of cAMP was examined by ELISA YPEN-1 cells were incubated with LPC at different concentrations for

15 min, and cells were then lysed as described in Experimental pro-cedures cAMP in cell lysate was extracted and detected with a cAMP ELISA kit Data are presented as average ± SE from tripli-cate experiments Statistical significance: *P < 0.05 versus control.

0

20

40

60

80

100

120

140

160

180

200

mp1379

VCAM

VCAM-1 promoter

P-selectin promoter

NF- κB reporter

*

Fig 3 Partial blockage of LPC-induced AMs by NAC Effects of the

antioxidant NAC on LPC-induced NF-jB activation, and P-selectin

and VCAM-1 promoter activities, in YPEN-1 cells were evaluated by

luciferase assay Cells were transfected with NF-jB-Luc or

P-selec-tin or VCAM gene promoter reporters, respectively, treated with

reagents for 8 h, and then lysed for determination of luciferase.

LPC + NAC, transfected cells preincubated with NAC (1 m M ) for

1 h before challenge with 12.5 l M LPC for an additional 7 h Each

value is expressed as the mean ± SE from six independent

experi-ments *P £ 0.05 versus TC in the same group.

the AC inhibitor MDL12330A (MDL) (10 lm) (P¼

0.069), which prevents the production of cAMP by AC

The PKA inhibitor H89 (10 lm) also reduced the

amount of LPC-activated CRE-Luc, implying the

involvement of PKA Thus, the results showed that LPC

induced activation of CRE through cAMP and PKA

Upregulation of AMs by LPC through GPR4 signal

transduction

A subsequent question is whether the GPR4⁄ cAMP ⁄

PKA⁄ CREB pathway contributes to LPC-induced AM

upregulation The effects of the AC inhibitor MDL and the PKA inhibitor H89 on the promoter activities

of VCAM-1 and P-selectin in response to LPC were assessed Pretreatment of YPEN-1 cells with MDL (10 lm) and H89 (10 lm) for 1 h suppressed LPC-induced promoter activities of VCAM-1 and P-selectin

as well as CRE signaling (Fig 6A) Consistent with this result, the expression of VCAM-1 protein was found to be blocked by MDL and H89, as compared

to cells treated only with LPC This result was in line with the reduced phosphorylation of CREB by MDL and H89 (Fig 6B)

B

0

100

200

300

400

500

600

**

*

0

20

40

60

80

100

120

140

160

Cont 10 30 60 120 240 (min)

*

Histone H1

A

Cont 10 30 60 120 240 (min)

LPC (6.25 µ M)

pCREB

(nucleus)

Fig 5 LPC-induced cAMP ⁄ PKA ⁄ CREB signaling in YPEN-1 cells.

(A) LPC-induced downstream signaling of cAMP was examined by

the phosphorylation of CREB YPEN-1 cells were incubated with

LPC (6.25 l M ) for different times, and nuclear fractions were then

used for western blot analysis The quantitative analysis is shown

under the blot Statistical significance: *P £ 0.05 versus control (B)

The effects of LPC on cAMP ⁄ PKA ⁄ CREB signaling were evaluated

by luciferase assay Cells were transfected with CRE luciferase

reporter (CRE-Luc) plasmids that contain the CREB-binding site.

LPC + MDL, transfected cells preincubated with MDL (10 l M ) for

1 h before challenge with 12.5 l M LPC for an additional 7 h;

LPC + H89, transfected cells preincubated with H89 (10 l M ) for 1 h

before challenge with 12.5 l M LPC for an additional 7 h FSK,

10 l M , was used as positive control Each value is expressed

as the mean ± SE from six independent experiments *P £ 0.05

versus TC.

ββ-Actin

pCREB (nucleus)

VCAM-1

B

TC

LPC

LPC + MDL

Relative RLU (% of TC)

LPC + H89

A

CRE luciferase reporter P-selectin gene promoter VCAM gene promoter

Fig 6 Activation of AMs by LPC through the cAMP ⁄ PKA pathway (A) Effects of LPC on CREB activation, P-selectin and VCAM-1 pro-moter activities in transfected YPEN-1 ECs were detected by lucif-erase assay Cells were transfected with CRE luciflucif-erase reporter or P-selectin gene promoter or VCAM gene promoter plasmids, which contain CREB-binding site or murine P-selectin or human VCAM-1 promoters, respectively LPC + MDL, transfected cells

preincubat-ed with MDL (10 l M ) for 1 h before challenge with 12.5 l M LPC for an additional 7 h; LPC + H89, transfected cells preincubated with H89 (10 l M ) for 1 h before challenge with 12.5 l M LPC for an additional 7 h Each value is expressed as the mean ± SE from three independent experiments (B) The contribution of cAMP and PKA activation to LPC-induced CREB activation and expression of VCAM was detected by applying the AC inhibitor MDL and the PKA inhibitor H89 YPEN-1 cells were preincubated with MDL (10 l M ) or H89 (10 l M ) for 1 h before challenge with 12.5 l M LPC Western blot analysis was carried out to detect the level of

VCAM-1 in the cytosol fraction after VCAM-12 h of treatment with LPC One rep-resentative result is shown.

The role of CREB in LPC-triggered transactivation

of AMs was then further examined by utilizing a

con-struct encoding a dominant-negative mutant CREB

protein, ACREB ACREB was constructed by fusing a

designed acidic amphipathic extension to the

N-termi-nus of the CREB leucine zipper domain [28] The

aci-dic extension of ACREB interacts with the basic

region of CREB, forming a coiled-coil extension of the

leucine zipper, and thus prevents the basic region of

wild-type CREB from binding to DNA [28,29] As

shown in Fig 7, cotransfection of ACREB with the

VCAM-Luc and P-selectin gene promoters completely

abolished LPC-induced AM promoter activation; in

contrast, the ACREB control vector (ACREB con) did

not significantly affect the activation of AMs by LPC

The involvement of GPR4 in LPC-induced AM

that were transfected with human GPR4 expression

vector (hGPR4) An expression vector for hGPR4

(NM_005282) was prepared using pcDNA3.1(+)⁄

myc-His [23] It was found that the basal level of

VCAM-1 was significantly increased after transfection

with hGPR4, and that it was further elevated by the

LPC challenge (Fig 8A) To avoid the potential effects

of endogenous GPR4, the hGPR4 expression vector was further transfected into HEK293T cells Following transfection of hGPR4, the HEK293T cells gained the ability to induce activation of CRE signaling as well as activation of the VCAM-1 gene promoter, and these activations were further enhanced by LPC challenge (Fig 8B)

LPC

ACREB

- + + +

- - +

ACREB (con)

0

40

80

120

160

P-selectin gene promoter

Fig 7 Abolition of LPC induction of AMs by cotransfection of

ACREB The involvement of CREB in LPC-induced AM activation

was evaluated by cotransfection with the CREB dominant negative

vector ACREB Cells were transfected with ACREB expression

vec-tor, ACREB control vector luciferase reporter, P-selectin gene

pro-moter or VCAM-1 gene propro-moter plasmids as indicated After

incubation for 30 h, cells were treated with LPC (12.5 l M ) for an

additional 8 h, and lysed for determination of luciferase activity.

Each value is expressed as the mean ± SE from six independent

experiments Statistical significance: *P < 0.05, **P < 0.01 versus

cells transfected with ACREB control vector and not challenged

with LPC.

hGPR4 hGPR4 (con)

A

0

#

**

200 400 600 800 1000

1200

VCAM gene promoter P-selectin gene promoter

B

hGPR4 hGPR4 (con)

##

**

**

0 50 100 150 200 250 300 350 400 450

VCAM gene promoter CRE luciferase reporter

Fig 8 Enhancement of LPC-induced AM activation by transfection

of GPR4 The effect of GPR4 on LPC-induced AM activation was evaluated by transfection of human GPR4 expression vector (hGPR4) to YPEN-1 cells (A) and HEK293T cells (B) before treat-ment with LPC (12.5 l M ) Cells were transfected with hGPR4 vec-tor, hGPR4 control vecvec-tor, P-selectin gene promoter or VCAM-1 gene promoter plasmids as indicated Thirty hours later, after transfection, cells were treated with LPC (12.5 l M ) for an addi-tional 8 h, and lysed for determination of luciferase Each value is expressed as the mean ± SE from six independent experiments Statistical significance: **P < 0.01 versus cells transfected with hGPR4 control vector without LPC challenge. #P < 0.05 versus cells transfected with hGPR4 expression vector without LPC challenge.

The current study revealed the molecular mechanisms

underlying the effects of LPC on the upregulation of

AMs Our data suggested the existence of two

path-ways, through NF-jB or through GPR4, emphasizing

the involvement of GPR4 in LPC-induced AMs,

thereby providing a new insight into the molecular

mechanism of LPC’s role in the expression of AMs

in ECs

The LPC-induced increase of NF-jB DNA-binding

activity was reported previously [30–32]; however, the

upstream signal is not known, although PKC [10] is

suggested to be involved From the current data, the

onset of NF-jB activation by LPC in ECs seems to be

immediate, as indicated by the rapid nuclear

transloca-tion of p65, 10 min after challenge with LPC

(Fig 2A); therefore, an early PKC (5 min) response

would be expected [33] Detailed information on the

activation of NF-jB signaling by LPC was not sought

in the present study, but we were able to confirm the

activation of NF-jB in ECs by LPC and its important

role in the induction of VCAM-1 and P-selectin by

LPC It is worth mentioning the involvement of

react-ive oxygen species in NF-jB activation NF-jB, being

redox-sensitive, has been shown to be influenced by an

increase in reactive oxygen species that would disrupt

the redox balance [34] Thus, it remains to be

deter-mined whether LPC influences NF-jB activity by

increased reactive oxygen species production, e.g

through various oxidases [35–37] or by other pathways

such as PKC [10]

What is new in the current study are our data

show-ing the involvement of LPC-activated CREB signalshow-ing

in the upregulation of AMs LPC-induced activation

of CREB [38–40] and an elevation in the level of

cAMP in neutrophils [41] have been reported Rikitake

et al suggest that both p38 and ERK may function as

upstream signaling pathways capable of activating

CREB and activating transcription factor-1 with

subse-quent induction of cyclooxygenase-2 expression by

LPC [12] However, here we did not find that the p38

inhibitor SB 203580 or the ERK inhibitor PD98059

influenced LPC-induced expression of AMs (data not

shown)

In the current study, we found that GPR4 played an

important role in LPC-induced expression of AMs in

ECs Our findings on the expression of GPR4 but not

G2A in ECs, shown in Fig 5, are in agreement with a

recent finding of a critical role of GPR4 in endothelial

cell function, reported by Kim et al [22] As shown in

Fig 8A, the overexpression of GPR4 enhanced the

response of YPEN-1 cells to LPC, leading to

addi-tional activation of VCAM-1 More convincing evi-dence shows that after transfection with the GPR4 gene, HEK293T cells gain the ability to respond to LPC by expressing VCAM-1 and activating CRE sign-aling, as shown in Fig 8B Taken together, these results strongly indicate the existence of receptor-medi-ated signaling in LPC-induced AMs In a recent study, GPR119 was shown to regulate LPC-induced upregu-lation of cAMP in pancreatic b-cells, resulting in the secretion of insulin [15] We are currently examining the possible participation of GPR119 in LPC-induced

AM expression

The data from the present study indicated that downstream signaling following LPC-induced GPR4 activation was via the cAMP⁄ PKA ⁄ CRE pathway These findings are consistent with previous reports showing that GPR4 elicits cAMP in Swiss 3T3 cells [23], activates the cAMP⁄ PKA pathway in Cryp-tococcus neoformans [25], and stimulates CRE-driven transcription [42] The role of CREB in LPC-induced AMs was previously unknown An early study repor-ted the inability of LPC to induce transcription of VCAM-1 in HUVECs [43], probably because the reporter plasmid that the authors used was encoded

(0⁄) 755 bp) without a CRE () 1420 bp) site, although it did contain NF-jB- and AP-1-binding sites In contrast, our present study clearly showed that LPC induced AM expression though CREB The expression of CREB’s dominant-negative control, ACREB, was able to abolish LPC-induced activation

of AMs (Fig 7) Therefore, the participation of LPC⁄ GPR4 ⁄ cAMP ⁄ PKA signaling in ECs may well contribute to AM expression through activation of the CRE site

One question left unresolved in the present study is whether the LPC⁄ NF-jB ⁄ AM and LPC⁄ GPR4 ⁄ cAMP⁄ PKA ⁄ CREB ⁄ AM pathways work equally, one pathway works preferentially over the other, or a there

is cross-talk between the two pathways Further studies should be able to answer this question, and elucidate other factors that may regulate the oxidative stress-induced and receptor-mediated pathways

In conclusion, from the current study, we were able to document two pathways by which LPC regulates the expression of AMs) the LPC ⁄

NF-jB⁄ AM and LPC⁄ GPR4 ⁄ cAMP ⁄ PKA ⁄ CREB ⁄ AM pathways Our new findings emphasize the import-ance of the LPC receptor in regulating EC function, and highlight the potential critical roles of LPC in modulating many pathophysiologic processes, including atherosclerosis

Experimental procedures

Materials

LPC (synthetic palmitoyl-LPC 16:0) was purchased from

Avanti Polar Lipid Inc (Alabaster, AL, USA) CAPE was

purchased from BioMol (Plymouth Meeting, PA, USA)

NAC, MDL, H89, isobutylmethylxanthine and FSK were

obtained from Sigma Chemical Co (St Louis, MO, USA)

Cells

Cell culture

The rat endothelial cell line YPEN-1 was obtained from the

American Tissue Culture Collection (ATCC, Manassas, VA,

USA) Cells were grown in DMEM (Gibco, Grand Island,

NY, USA) containing 10% heat-inactive fetal bovine serum

(Mediatech, Herndon, VA, USA), glutamine 233.6 mgÆmL)1,

penicillin–streptomycin 72 mgÆmL)1, and amphotericin B

0.25 mgÆmL)1, and were adjusted to pH 7.4–7.6 with

NaHCO3in a CO2incubator with an atmosphere of 5% CO2

at 37C Before treatment, cells were exchanged into fresh

DMEM containing 1% fetal bovine serum

For transfection experiments with hGPR4 expression

vec-tor, HEK293T cells were obtained from the ATCC and

maintained in DMEM supplemented with 10% fetal bovine

serum and other supplements as described above

Cell viability assay

Cell viability was tested using a

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay, which is a

sensitive, quantitative colorimetric assay that measures cell

viability on the basis of the ability of mitochondrial succinyl

dehydrogenase in living cells to convert the yellow substrate

MTT into a dark blue formazan product For the assay, the

medium was removed, and a solution containing 0.01%

MTT was added to each well; this was followed by

incuba-tion at 37C for 4 h, and the formazan was dissolved in

ethanol⁄ dimethylsulfoxide (1 : 1, v ⁄ v) The plate was shaken

for 5 min, and the absorbance was measured at 560 nm

Cell sample preparation

Following incubation for 12 h or where indicated, cells

were harvested and lysed in lysis buffer A (10 mm Tris,

pH 8.0, 1.5 mm MgCl2, 1 mm dithiothreitol, 0.1% Nonidet

P-40, and protease inhibitors) After centrifugation at

13 000 g for 15 min (Mega 17R, Hanil Science Industrial

Co Ltd., Inchon, Korea, A1.55-24), the supernatant was

regarded as the cytosol fraction for assays The pellets were

resuspended in 10 mm Tris (pH 8.0), with 50 mm KCl,

100 mm NaCl, and protease inhibitors, incubated on ice for

30 min, and then centrifuged at 13 000 g at 4C for 30 min

(Mega 17R, Hanil, A1.55-24) After that, supernatants were used as the nuclear fraction

Western blot For western blotting analysis, cell samples were separated

on an SDS⁄ PAGE mini-gel The antibodies used in this study were as follows: anti-(phosphorylated CREB) (anti-pCREB) was from Upstate (Charlottesville, VA, USA), and all other antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA) Antibody labeling was detected using an ECL detection kit (Amersham Life Science, Inc., Arlington Heights, IL, USA) A prestained blue protein marker was used for molecular weight determination The amount of protein was measured with a Sigma protein assay reagent kit containing bicinchoninic acid

RT-PCR Total RNA from cells or tissue was isolated by a previously described method [44] Briefly, after treatment, cells were lysed in the presence of RNAzolB (1 mL per 100 mm dish) Chloroform (0.1 mL per 1 mL of lysate) was added to each dish The samples were shaken vigorously, and then placed

on ice for 5 min After centrifugation twice at 12 000 g at

4C for 15 min (Mega 17R, Hanil, A1.55-24), the superna-tant was removed The RNA pellet was washed with 75% ethanol, dried, and redissolved in diethylpyrocarbonate-trea-ted water cDNA was synthesized using ImProm-II reverse transcriptase (Promega, Madison, WI) PCR was carried out using a standard protocol Primers were designed for AMs as follows: P-selectin, sense strand 5¢-CGA CGT GGA CCT ATA ACT AC-3¢ and antisense strand 5¢-CCA CAC

5¢-CTT GGA GAA CCC AGA TAG AC-3¢ and antisense strand 5¢-CAG AAA ATC TCA GGA GCT GG-3¢; GPR4, sense strand 5¢-AGC ATT GCA GAC CTG CTG TA-3¢

GCA GGA AAC ACC A-3¢ and antisense strand 5¢-AAG CCA AAG GTG AAA CGC AGG T-3¢ Glyceraldehyde-3-phosphate dehydrogenase was used as internal control

Transient transfection and luciferase reporter assay

VCAM-1 and P-selectin promoter activity induced by LPC was examined using luciferase plasmid VCAM-Luc (provi-ded by W Aird, Harvard Medical School, MA, USA), which contains a human VCAM-1 promoter region spanning ) 1716 to + 119 bp, and P-selectin -Luc (R P McEver, Uni-versity of Oklahoma Health Sciences Center, OK, USA) containing wild-type murine P-selectin promoter Activation

of NF-jB and CRE was measured using NF-jB luciferase

reporter (Invitrogen, Carlsbad, CA, USA) and CRE

luci-ferase reporter (H Inoue, National Cardiovascular Center

Research Institute, Osaka, Japan), respectively Expression

vector ACREB (C Vinson, National Cancer Institute, NIH,

Rockville, MD, USA) and hGPR4 (K Seuwen, Novartis

Institutes for BioMedical Research, Basel, Switzerland) and

their controls were used as described below

Transient transfection was carried out using FuGene 6

(Roche Molecular Biochemicals, Indianapolis, IN, USA)

according to the manufacturer’s instructions After

transfec-tion, cells were treated with reagents as per the experimental

design Briefly, YPEN-1 cells were seeded into 48-well plates

(1· 105cellsÆmL)1, and 250 lL per well), and cultured in

DMEM containing 10% fetal bovine serum overnight For

transfection, the cells should be over 90% confluent For

single transfection, plasmid (0.1 lg per well) was used, and

for cotransfection, plasmids were mixed in a 1 : 1 ratio to a

total amount of 0.1 lg per well Following transfection, cells

were cultured for 24 h, and then exposed to DMEM

con-taining 1% fetal bovine serum with⁄ without designated

rea-gents for an additional 8 h Luciferase activity was measured

with the Steady-Glo Luciferase Assay System (Promega)

and detected by luminometer GENios Plus (Tecan Group

Ltd, Salzburg, Austria) The obtained raw luciferase

activit-ies were normalized by protein concentration per well

Measurement of cAMP

The accumulation of cAMP in the YPEN-1 cells was

induced by LPC at different concentrations in serum-free

medium at 37C for 15 min in the presence of 1 mm

iso-butylmethylxanthine The cells were then lysed with 1 m

HCl Quantitative determination of the cAMP

concentra-tion in cell lysates was performed using a cAMP (low pH)

immunoassay kit (R&D Systems, Minneapolis, MN, USA)

All procedures followed the user’s manual provided with

the enzyme immunoassay kit

Statistical analysis

For western blot analysis or RT-PCR, one representative

blot was used from three independent experiments Image

analysis was performed using the imagej (NIH, Bethesda,

MD, USA) program anova was conducted to analyze

sig-nificant differences among all groups Differences among

the means of individual groups were assessed by Fischer’s

Protected LSD post hoc test Values of P < 0.05 were

con-sidered to be statistically significant

Acknowledgements

This study was supported by a grant from the Korea

Health R&D project, Ministry of Health and Welfare,

Republic of Korea (A050166) We thank the Aging

Tissue Bank for distributing the aged tissue (R21-2005-000-10008-0)

References

1 Aiyar N, Disa J, Ao Z, Ju H, Nerurkar S, Willette RN, Macphee CH, Johns DG & Douglas SA (2007) Lyso-phosphatidylcholine induces inflammatory activation of human coronary artery smooth muscle cells Mol Cell Biochem 295, 113–120

2 Kougias P, Chai H, Lin PH, Lumsden AB, Yao Q & Chen C (2006) Lysophosphatidylcholine and secretory phospholipase A2 in vascular disease: mediators of endothelial dysfunction and atherosclerosis Med Sci Monit 12, RA5–RA16

3 Ryborg AK, Gron B & Kragballe K (1995) Increased lysophosphatidylcholine content in lesional psoriatic skin Br J Dermatol 133, 398–402

4 Mehta D, Gupta S, Gaur SN, Gangal SV & Agrawal

KP (1990) Increased leukocyte phospholipase A2 activ-ity and plasma lysophosphatidylcholine levels in asthma and rhinitis and their relationship to airway sensitivity

to histamine Am Rev Respir Dis 142, 157–161

5 Portman OW & Alexander M (1969) Lysophosphatidyl-choline concentrations and metabolism in aortic intima plus inner media: effect of nutritionally induced athero-sclerosis J Lipid Res 10, 158–165

6 Vainer B (1997) Role of cell adhesion molecules in inflammatory bowel diseases Scand J Gastroenterol

32, 401–410

7 Kume N, Cybulsky MI & Gimbrone MA Jr (1992) Lysophosphatidylcholine, a component of atherogenic lipoproteins, induces mononuclear leukocyte adhesion molecules in cultured human and rabbit arterial endothelial cells J Clin Invest 90, 1138–1144

8 Kita T, Kume N, Ochi H, Nishi E, Sakai A, Ishii K, Nagano Y & Yokode M (1997) Induction of endothelial platelet-derived growth factor-B-chain and intercellular adhesion molecule-1 by lysophosphatidylcholine Ann

NY Acad Sci 811, 70–75

9 Thukkani AK, McHowat J, Hsu FF, Brennan ML, Hazen SL & Ford DA (2003) Identification of alpha-chloro fatty aldehydes and unsaturated lysophosphati-dylcholine molecular species in human atherosclerotic lesions Circulation 108, 3128–3133

10 Sugiyama S, Kugiyama K, Ogata N, Doi H, Ota Y, Ohgushi M, Matsumura T, Oka H & Yasue H (1998) Biphasic regulation of transcription factor nuclear factor-kappaB activity in human endothelial cells by lysophosphatidylcholine through protein kinase C-mediated pathway Arterioscler Thromb Vasc Biol 18, 568–576

11 Fang X, Gibson S, Flowers M, Furui T, Bast RC Jr & Mills GB (1997) Lysophosphatidylcholine stimulates