Short-term high glucose exposure induces monocyte-endothelial cells adhesion and transmigration by increasing VCAM-1 and MCP-1 expression in human aortic endothelial cells

A

amH(cesAbhd©

Kc

1

iacplW

0d

Atherosclerosis 193 (2007) 328–334

Short-term high glucose exposure induces monocyte-endothelial cellsadhesion and transmigration by increasing VCAM-1 and MCP-1

expression in human aortic endothelial cells

Rosaria Piga a,∗, Yuji Naito b, Satoshi Kokura c, Osamu Handa c, Toshikazu Yoshikawa a

a Department of Inflammation and Immunology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto, Japanb Department of Medical Proteomics, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto, Japan

c Department of Biomedical Safety Science, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto, Japan

Received 5 January 2006; received in revised form 12 August 2006; accepted 19 September 2006Available online 13 November 2006

bstract

Acute, short-term hyperglycemia is becoming recognized as an important risk factor for several diseases. In the present study, using humanortic endothelial cells (HAECs), we investigated whether short-term high glucose exposure, either on the scale of hours, could enhance theonocyte adhesion and migration to the subendothelium via increasing expression of adhesion molecules and release of chemotactic factors.AECs stimulated with 25 mM d(+)glucose (HG) for not more than 12 h, exhibited rapid up-regulation of vascular cell adhesion molecule-1

VCAM-1) and monocyte chemoattractant protein-1 (MCP-1) mRNA and protein. Although intercellular adhesion molecule-1 (ICAM-1) isonsidered as a marker of the activation of the atherogenic process, early up-regulation was not observed, and VCAM-1 and MCP-1 proteinnhance was sufficient to increase the adhesiveness of human monocytes U-937 to HAECs and their transmigration into the subendothelialpace after 4 h HG stimulation; both effects were prevented by interfering with monoclonal antibodies against VCAM-1, CD11b, and MCP-1.n increased intracellular oxidative stress, a translocation of NF-�B to the nucleus and a prevention of adhesion and transmigration of U-937y interfering with NF-�B inhibitors was also observed after a short HG treatment. Taken together, these results suggest that either acuteyperglycemic spikes could exert an influence on the onset of diabetic complications and on the development of the atherogenic profile on

2006 Elsevier Ireland Ltd. All rights reserved.

eywords: Atherosclerosis; Short-term high glucose; Monocyte adhesion; Monocyte migration; HAEC; Vascular cell adhesion molecule (VCAM-1); Monocytehemoattractant protein (MCP-1); Reactive oxygen species; Nuclear factor (NF)-�B

atsaal

. Introduction

There is increasing evidence that the postprandial states an important contributing factor to the development oftherosclerosis in diabetic, non-diabetic and non-diabeticritically ill subjects [1–3]. Chronic hyperglycemia as been

ith regard to infarction, a recent meta-analysis has shown

∗ Corresponding author. Tel.: +81 75 251 5508; fax: +81 75 252 3721.E-mail address: [email protected] (R. Piga).

st

tsu

021-9150/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.atherosclerosis.2006.09.016

continuous correlation between glucose serum concen-rations and severity of the prognosis even in non-diabeticubjects [6]. Rapid increases in blood glucose concentrationsre common and frequent events in the life of non-diabeticnd diabetic patients [7]. Despite that, it is surprising howittle emphasis researchers have previously put on glycemicpikes as possible contributors to diabetic complications ando complication of different nature in non-diabetic subjects.

The adhesion and migration of circulating monocytes intohe subendothelial space is one of the key events in the earlytages of atherosclerogenesis [8]. This process is in part reg-lated by the expression of some adhesion molecules on

clerosis

tccop

thg

hpcc

2

2

ao(dmsllfeamfHdcwflwR

AgCmi

puJuciwT

2

mHapHmbwa5wwpFsifwT1unp

2

d(oogkRtuoc1JirPtoRSea

R. Piga et al. / Atheros

he surface of endothelial cells (ECAMs), including vascularell adhesion molecule-1 (VCAM-1), E-selectin, and inter-ellular adhesion molecule-1 (ICAM-1) and by the releasef chemotactic factors including monocyte chemoattractantrotein-1 (MCP-1) [9,10].

Although there are numbers of evidence that underlinehe effect of chronic high glucose exposure, early damage ofuman aortic endothelial cells (HAECs) by short-term highlucose stimulation is not fully investigated.

Therefore, our study aims to emphasize the effects of acuteyperglycemia on the development of several health com-lications, becoming advisable to consider this aspect, theontrol of hyperglycemic spikes, in the treatment of severalomplications and diseases.

. Materials and methods

.1. Cells

HAECs were obtained from Clonetics (San Diego, CA)nd routinely maintained at 37 ◦C in a humified atmospheref 95% air and 5% CO2 in endothelial cell growth mediumEBM-2, Clonetics, San Diego, CA) containing 5.5 mM(+)glucose, and supplemented with endothelial growthedium 2 (EGM-2) kit (Clonetics) containing fetal calf

erum (FCS; 2%), hydrocortisone (0.04%), human fibrob-ast growth factor B (hFGF-B; 0.4%), vascular endothe-ial growth factor (VEGF; 0.1%), R3-insulin-like growthactor 1 (R3-IGF-1; 0.1%), ascorbic acid (0.1%), humanpidermal growth factor (hEGF; 0.1%), gentamicin sulfatemphotericin-B (GA-1000; 0.1%), heparin (0.1%) (completeedium), according to manufacturer’s instructions. There-

ore, 5.5 mM d(+)glucose is considered normal glucose, andAECs growing in their EBM-2 medium containing 5.5 mM(+)glucose required for their survive, were used as controlells and considered as untreated. For all experiments, cellsere used up to the sixth passage. After detachment of con-uent HAECs from the flasks with 0.025% trypsin, the cellsere plated in 24- or 96-well tissue culture plates (Nunc,oskilde, Denmark).

Human monocytic U-937 cells were purchased from themerican Type Culture Collection (Rockville, MD) andrown in RPMI 1640 medium (Invitrogren Co., Carlsbad,A) containing 10% fetal calf serum, 100 �g/ml strepto-ycin, 100 IU/ml penicillin, 250 ng/ml fungizone at 37 ◦C

n a humified atmosphere of 95% air and 5% CO2.Cell viability was assessed by WST-1 (4-[3-(4-iodo-

henyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene dis-lfonate) assay (Cell Counting Kit, Dojindo, Kumamoto,apan) as previously described [11]. According to the man-facturer’s protocol, WST-1 solution was added to confluent

fa3a

193 (2007) 328–334 329

.2. Enzyme-linked immunosorbent assay

Enzyme immunoassay was used to assess the adhesionolecule expression on HAECs as previously described [12].AEC confluent monolayers were grown on 96-well plate

nd stimulated with 25 mM d(+)glucose (HG) for differenteriod of time (0, 2, 4, 8, and 12 h). For osmolality control,AECs were grown in 5.5 mM d(+)glucose plus 19.5 mMannitol. Subsequently, the cells were washed with Hank’s

alanced salt solution (HBSS), and VCAM-1 and MCP-1ere detected by ELISA. A monoclonal antibody directed

gainst VCAM-1 (purified mouse anti-human CD106 clone1-10C9, BD Pharmingen, BD Biosciences, San Jose, CA)as added and the cells were incubated at 37 ◦C for 60 min,ashed three times with HBSS, and then incubated for 1 h ineroxidase conjugated goat-anti-mouse IgG affinity-purified(ab′)2 fragment (Cappel, Durham, NC). After washing,ubstrate (o-phenylenediamine dihydrochloride, 0.4 mg/mln buffer, pH 5.0) was added and the cells were incubatedor 30 min at room temperature. The optical density of eachell were read at 492 nm using a Microplate reader (Tosoh,okyo) to quantify the amount of bound antibody. MCP-protein level in cell culture supernatant was examined

sing a Human MCP-1 Immunoassay kit (Biosource Inter-ational, Camarillo, CA) according to the manufacturer’srotocol.

.3. Isolation of mRNA and real time PCR

HAEC confluent monolayers were grown on 35-mmishes and stimulated with HG for different period of time0–4 h), not more than 4 h because of both protein peal levelbserved at 4 h, in the simultaneous presence or absencef 10 �M PDTC. Total RNA was isolated with the aciduanidinium phenol chloroform method using an Isogenit (Nippon Gene, Tokyo, Japan). The concentration ofNA was determined by absorbance at 260 nm in rela-

ion to absorbance at 280 nm. RNA was stored at −70 ◦Cntil reverse transcription was performed. An aliquot (1 �g)f extracted RNA was reverse-transcribed into first strandomplementary DNA (cDNA) at 42 ◦C for 40 min, using00 U/ml reverse-transcriptase (Takara Biochemicals, Shiga,apan) and 0.1 �M of oligo (dT)-adapter primer (Takara)n a 50 �l reaction mixture. Real-time polymerase chaineaction (PCR) was carried out with a 7300 Real TimeCR system (Applied Biosystems, Foster City, CA) using

he DNA-binding dye SYBER Green I for the detectionf PCR products. The reaction mixture (RT-PCR kit, CodeRO43A, Takara) contained 12.5 �l Premix Ex Taq, 2.5 �lYBER Green I, custom-synthesized primers, ROX refer-nce dye, cDNA (equivalent to 20 ng total RNA) to givefinal reaction volume of 25 �l. Primers were as follows:

or VCAM-1, sense 5′-CCCTTGACCGGCTGGAGATT-3′,nd antisense 5′-CTGGGGGCAACATTGACATAAAGTG-′; for MCP-1, sense 5′-CGCCTCCAGCATGAAAGTCT-3′,nd antisense 5′-GGAATGAAGGTGGCTGCTATG-3′; for

3 clerosis

GaswaitMt

2

deue(Cps(iracpetpBdtOqw1aiBat

2

dcCHsohsU

pthwcflolcgabmr

2

tpweMtaa

2

pRM(fiwslaqfFaFwaplomG

30 R. Piga et al. / Atheros

APDH, sense 5′-ACCACAGTCCATGCCATCAC-3′, andntisense 5′-TCCACCACCCTGTTGCTGTA-3′. The PCRettings were as follows: initial denaturation of 15 s at 95 ◦Cas followed by 40 cycle of amplification for 3 s at 95 ◦C

nd 31 s at 60 ◦C, with subsequent melting curve analysisncreasing the temperature from 60 to 95 ◦C. In order to quan-ify VCAM-1 and MCP-1 gene expression, the VCAM-1 and

CP-1 mRNA level was normalized by the total RNA con-ent (�g/�l) and GAPDH mRNA level.

.4. Monocyte adhesion assay

The adhesion assay was performed as previouslyescribed, with minor modification [13]. HAEC conflu-nt monolayers were grown on 24-well plates and stim-lated with HG for 4 h. In some experiments cells werexposed to 10 �g/ml anti-VCAM-1, or 10 �g/ml anti-CD11bpurified mouse anti-human CD11b clone ICRF44 Ancellorporation, Bayport, MN), or 10 �M NF-�B-inhibitor,yrrolidine dithiocarbamate (PDTC; Sigma, St. Louis, MO)imultaneously to HG stimulation. Subsequently, U-9371 × 106 cells/ml, 100 �l) were added to each well andncubated at 37 ◦C for 30 min. Non-adherent U-937 wereemoved, and plates were gently washed twice with HBSSnd fixed in 1% glutaraldehyde in PBS. The adherent mono-ytes were counted in five high-power fields (HPF, 40X-hase contrast objective) per well under a microscope andxpressed as adherent monocytes per HPF. At the same time,he number of adherent U-937 was also quantified usingreviously described method with minor modification [14].riefly, U-937 were labeled with 10 �mol/L of fluorescentye 2′,7′-bis (2-carboxyethyl)-5(6)-carboxy fluorescein ace-oxymethyl ester (BCECF-AM, Molecular Probes, Eugene,R) at 37 ◦C for 1 h in RPMI-1640 medium, and subse-uently washed by centrifugation. Confluent HAECs in 24-ell plates were incubated with labeled U-937 (106 cells/ml,00 �l) at 37 ◦C for 1 h. Nonadherent U-937 were removed,nd plates were gently washed twice with HBSS and thenncubated with 1% TritonX for 2 h. After the incubation,CECF fluorescent intensity of adherent cells were measuredt excitation/emission wave length of 503/525 nm respec-ively by Bio-Rad Fluoromark (Melville, NY).

.5. Monocyte transmigration assay

The transmigration assay was performed as previouslyescribed, with minor modification [15]. Transwells (poly-arbonate membranes with 3 �m pore sizes, Nucleoporeorp., Pleasanton, CA) in 24-well plates were used, andAEC confluent monolayers on Transwell membranes were

timulated with HG for 4 h in the simultaneous presence

tewR

193 (2007) 328–334

artment, and were incubated for 1 h at 37 ◦C. Cells that hadransmigrated to the lower compartment were counted in fiveigh-power fields (HPF, 40X-phase contrast objective) perell under a microscope and expressed as adherent mono-

ytes per HPF. At the same time, the number of migrateduorescent U-937 was also quantified as described previ-usly on adhesion assay. After the treatments of HAECs,abeled U-937 were added to the upper compartment andells were incubated for 1 h at 37 ◦C. Cells that had transmi-rated to the lower compartment were harvested in 1 ml PBSnd then incubated with 1% TritonX for 2 h. After the incu-ation, BCECF fluorescent intensity of migrated cells wereeasured at excitation/emission wave length of 503/525 nm

espectively by Bio-Rad Fluoromark (Melville, NY).

.6. NF-κB activation assay

HAEC confluent monolayers cultured in 100 mm diame-er plates were stimulated with HG for 1 h in the simultaneousresence or absence of 10 �M PDTC. 10 ng/ml TNF-� for 1 has used as positive control. Subsequently, HAECs nuclear

xtraction was performed using a Nuclear Extract Kit (Activeotif, Carlsbad, CA) according to the manufacturer’s pro-

ocol. Then, NF-�B on nuclear extracts was detected usingTransAM NF-�B p65 Kit (Active Motif, Carlsbad, CA)

ccording to the manufacturer’s protocol.

.7. Determination of intracellular ROS

Intracellular ROS were detected using the fluorescentrobes 2′,7′-dichlorofluorescin diacetate (DCFH-DA) andedoxSensor Red CC-1 (Molecular Probes, Eugene, OR).itochondria were labeled using MitoTraker Green FM

Molecular Probes, Eugene, OR). DCFH-DA assay was per-ormed as described earlier [16]. To quantify the amount ofntracellular ROS production, HAEC confluent monolayersere grown on 35-mm dishes and stimulated with HG for the

pecified periods (0, 30, 60, and 90 min). Cells were then col-ected, resuspended in PBS, and incubated with DCFH-DAt a final concentration of 5 �M, for 15 min at 37 ◦C. Subse-uently, the cells were washed once with PBS and incubatedor 1 h at 37 ◦C. The cells were excited at 490 nm by Bio-Radluoromark (Melville, NY) and DCF emission was recordedt 525 nm. The mitochondrion-selective MitoTraker GreenM stain in conjunction with RedoxSensor Red CC-1 stainas used to verify the distribution of the oxidized product,

ccording to the manufacturer’s protocol. The nonfluorescentrobe Red CC-1 passively enters live cells and then intracel-ular ROS was assessed by the fluorescence intensity of thexidized red-fluorescent product emission accumulated in theitochondria. Mitocondria were labeled using MitoTrakerreen FM that is essentially nonfluorescent in aqueous solu-

ions, and become fluorescent once it accumulates in the lipidnvironment of mitochondria. HAECs were grown on two-ell chamber glass slides (Lab-Tek Chamber Slide; Nunc,ochester, NY) until confluence. After stimulation of HAECs

clerosis

wmteGo

2

e(tSs

3

3p

wossusasOp

Mf

FHl41Ppt*

atwbt(miHgr

3m

1MIiapip1

3H

Since adhesion molecules are essential for the inter-action of endothelial cells with monocytes, U-937 wereused to determined if HG-induced VCAM-1 expression on

R. Piga et al. / Atheros

ith HG for 1 h, Red CC-1 and Green FM were added to theedium at a final concentration of 1 �M, for 10 min at room

emperature. The cells were excited at 540 and 490 nm andmission was recorded at 600 and 516 nm for Red CC-1 andreen FM, respectively, and observed using the FV1000 flu-view BX61 (Olympus, Tokyo, Japan).

.8. Data analysis

All data of at least three separate experiments werexpressed as mean ± S.D. Two-way analysis of varianceANOVA) was used to assess the differences between mul-iple groups using GraphPad Prism 4 program (GraphPadoftware, Inc., San Diego, CA). Values of p < 0.05 were con-idered to be statistically significant.

. Results

.1. Effects of short-term HG on VCAM-1 and MCP-1rotein expression in HAECs

Our initial experiments were carried out to establishhether shot-term HG exposure could induce the expressionf adhesion molecules on HAECs. As shown in Fig. 1, HGignificantly caused a transient increase in the surface expres-ion of VCAM-1 protein beginning from 4 h and decreasingp to 12 h even if still significant at 8 h. VCAM-1 expres-ion, as measured by ELISA, increased and decreased intime-dependent manner and the resulting highest expres-

ion was approximately four-fold higher (0.393 VCAM-1.D./�g protein) than control cells (0.101 VCAM-1 O.D./�g

Cells treated with short-term HG also increased significantCP-1 protein release in cell culture supernatant beginning

rom 4 h and persisting for 12 h (Fig. 1) with a protein level of

ig. 1. Effects of short-term HG on VCAM-1 and MCP-1 protein in HAECs.AECs were cultured in EBM-2 medium in 96-well plate. Confluent mono-

ayers were then stimulated with HG for different incubation times (0, 2,, 8, and 12 h), and VCAM-1 protein expression (closed circle) and MCP-release (open circle) were detected by ELISA as described in Section 2.rotein expression was normalized to total cell protein (VCAM-1 O.D./�grotein and ng MCP-1/�g protein) and results were expressed as fold induc-ion respect time 0. Means and S.D. of at least three experiments are shown.*p < 0.001, ***p < 0.0001, when compared to time 0.

H

FispmgIaat*

a

193 (2007) 328–334 331

pproximately three-fold higher (36.2 ng MCP-1/�g protein)han control cells (11.6 ng MCP-1/�g protein). Treatmentith HG for 12 h did not affect cell viability determinedy WST-1 assay; percentage of optical density versus con-rol HAECs was approximately 95% in HG-treated HAECsdata not shown). Unlike d-glucose, the addition of 19.5 mMannitol to the media containing 5.5 mM d-glucose did not

ncrease the expression of VCAM-1 and MCP-1 proteins inAECs compared with that in control (data not shown), sug-esting that HG-induced VCAM-1 and MCP-1 is not theesult of high osmolality (per se).

.2. Effects of short-term HG on VCAM-1 and MCP-1RNA expression in HAECs

To further analyze the effect of short-term HG on VCAM-and MCP-1, we assessed mRNA expression for VCAM-1,CP-1 and GAPDH genes in HAECs by real time PCR.

ncubation of HAECs with HG from 1 to 4 h significantlyncreased the VCAM-1 and MCP-1 mRNA expression ofpproximately 3-fold and 2.5-fold, respectively, when com-ared to control (Fig. 2). Simultaneous treatment with HGn the presence of NF-�B inhibitor PDTC (10 �M), com-letely abolished mRNA expression for VCAM-1 and MCP-(Fig. 2).

.3. Monocyte adhesion to short-term HG-stimulatedAECs

AECs was enough to increase adhesiveness of monocytes.

ig. 2. Effects of short-term HG on VCAM-1 and MCP-1 mRNA expressionn HAECs. HAEC confluent monolayers were grown in 35-mm dishes andtimulated with HG for different period of time (0–4 h) in the simultaneousresence (discontinuous line) or absence (continuous line) of 10 �M PDTC.RNA expression for VCAM-1 (circle), MCP-1 (squares) and GAPDH

enes in HAECs was assessed by real time PCR as described in Section 2.n order to quantify VCAM-1 and MCP-1 gene expression, the VCAM-1nd MCP-1 mRNA level was normalized by the total RNA content (�g/�l)nd GAPDH mRNA level. Results were expressed as fold induction respectime 0. Means and S.D. of at least three experiments are shown. *p < 0.05,*p < 0.001, when compared to time 0, aap < 0.001, when compared to HGlone.

332 R. Piga et al. / Atherosclerosis 193 (2007) 328–334

Fig. 3. Monocyte adhesion to short-term HG-stimulated HAECs. HAECswere cultured in EBM-2 medium in 24-well plate. Confluent monolayerswere grown on 24-well plates and stimulated with HG for 4 h. In someexperiments cells were exposed to 10 �g/ml anti-VCAM-1, or 10 �g/mlanti-CD11b, or 10 �M PDTC simultaneously to HG stimulation. Subse-quently, BCECF-labeled U-937 (1 × 106 cells/ml, 100 �l) were added toeach well as described in Section 2 and BCECF fluorescent intensity of adher-ent cells were measured at excitation/emission wave length of 503/525 nmrfle*

IaltwaaeTa(twnsi

3H

ngaowtpcHs9a

Fig. 4. Monocyte transmigration to short-term HG-stimulated HAECs.Transwells in 24-well plates were used, and HAEC confluent monolayers ontranswell membranes were stimulated with HG for 4 h. In some experimentscells were exposed to 10 �g/ml anti-MCP-1, or 10 �M PDTC simultaneouslyto HG stimulation. Subsequently, BCECF-labeled U-937 (1 × 106 cells/ml,100 �l) were added to the upper compartment as described in Section 2.After the incubation, BCECF fluorescent intensity of migrated cells weremeasured at excitation/emission wave length of 503/525 nm respectively byBio-Rad Fluoromark. Results were expressed as BCECF fluorescence foldist

(sts

3

p1 h and NF-�B binding activity [Rel A (p65)] was evaluatedby ELISA as described in Section 2. As shown in Fig. 5,HG induced a significant increase of the NF-�B binding

Fig. 5. Effect of short-term HG on NF-�B activation. HAEC confluentmonolayers were cultured in 100 mm diameter plates and stimulated withHG for 1 h in the simultaneous presence or absence of 10 PDTC. 10 ng/mlTNF-� for 1 h was used as positive control. Subsequently, HAECs nuclearextraction was performed using a Nuclear Extract Kit and NF-�B binding

espectively by Bio-Rad Fluoromark. Results were expressed as BCECFuorescence fold induction respect control. Means and S.D. of at least threexperiments are shown. aaap < 0.0001, when compared to control. *p < 0.05,**p < 0.0001, when compared to HG alone.

ncubation of confluent HAECs with HG for 4 h causedn approximate three-fold increase in adhesion of BCECF-abeled U-937 compared with the adhesion of U-937 to con-rol (Fig. 3). Furthermore, simultaneous treatment of HAECsith 10 �g/ml anti-VCAM-1 and anti-CD11b monoclonal

ntibodies resulted in a significant decrease of HG-induceddhesion of U-937 to HAECs, reaching the highest inhibitoryffect when a combination of both antibodies was used.his HG-induced adhesiveness effect on U-937 adhesion waslso significantly inhibited by the NF-�B inhibitor PDTC10 �M). In these experiments, all inhibitors at the concen-ration used did not affect the viability of HAECs during thehole 4 h treatment, as measured by the WST-1 assay (dataot shown). A similar pattern was obtained performing theame experiments using nonlabeled U-937 and counting themn five high-power fields (data not shown).

.4. Monocyte transmigration to short-termG-stimulated HAECs

To determine whether MCP-1 release in HAEC super-atants was sufficient to increase the monocyte transmi-ration activity, the transmigration assay was performed intranswell system as described in Section 2. Incubation

f confluent HAEC monolayers on transwell membranesith HG for 4 h caused an almost three-fold increase in

ransmigration of BCECF-labeled U-937 to the lower com-artment compared with the transmigration of U-937 toontrol (Fig. 4). Furthermore, simultaneous treatment of

ignificant suppression of HG-induced transmigration of U-37. This HG-induced effect on monocyte migration waslso significantly inhibited by the NF-�B inhibitor PDTC

aafaw

nduction respect control. Means and S.D. of at least three experiments arehown. aap < 0.001, when compared to control. **p < 0.001, when comparedo HG alone.

10 �M). A similar pattern was obtained performing theame experiments using nonlabeled U-937 and counting theransmigrated monocytes in five high-power fields (data nothown).

.5. Effect of short-term HG on NF-κB activation

HAECs were stimulated with HG in the simultaneousresence or absence of NF-�B inhibitor PDTC (10 �M) for

ctivity on nuclear extracts was examined using a TransAM NF-�B p65 Kits described in Section 2. Results were expressed as NF-�B O.D./�g proteinold induction respect control. Means and S.D. of at least three experimentsre shown. aaap < 0.0001, aap < 0.001, when compared to control. **p < 0.001,hen compared to HG alone.

R. Piga et al. / Atherosclerosis

Fig. 6. Effect of short-term HG on ROS production. Intracellular ROS pro-duction was determined using the fluorescence probe DCFH-DA. HAECconfluent monolayers were grown on 35-mm dishes and stimulated withHG for the specified periods (0, 30, 60, and 90 min) (close circle). Controlcells are also shown (open circle). Cells were then collected, resuspended inPBS, and incubated with DCFH-DA as described in Section 2 and intracellu-lar ROS was assessed by the fluorescence intensity of DCF emission. Resultswac

aTP

3H

flw5itisStmj

4

tiasa

dpwp

fdacc

bptwi

ttfihVetanE

eeoamwcisrVtoabtt

tgbtriNe

aserum glucose concentrations cause alterations in both nor-

ere expressed as DCF fluorescence fold induction respect time 0. Meansnd S.D. of at least three experiments are shown. *p < 0.05, **p < 0.001, whenompared to control.

ctivity of approximately 2.5-fold when compared to control.his HG-induced effect was prevented by NF-�B inhibitorDTC.

.6. Evaluation of intracellular ROS on short-termG-treated HAECs

Intracellular ROS production was determined using theuorescence probe DCFH-DA. After treatment of HAECsith HG for the indicated times, the cells were treated with�M membrane-permeant nonfluorescent DCFH-DA, and

ntracellular ROS was subsequently assessed by measuringhe fluorescence intensity of the DCF emission. The HG-nduced NF-�B activation on HAECs was preceded by aignificant increase in the intracellular ROS levels (Fig. 6).imilar pattern was obtained measuring the distribution of

he oxidized products by confocal microscopy using theitochondrion-selective MitoTraker Green FM stain in con-

unction with RedoxSensor Red CC-1 stain (data not shown).

. Discussion

In the present study, we discuss the importance of short-erm HG exposure, either on the scale of hours, on its ability toncrease the expression of VCAM-1 and MCP-1 in HAECs,nd the power of these two proteins to exert monocyte adhe-ion and migration even in a short time expression, and withshort increased protein level.

Acute hyperglycemia during myocardial infarction pre-icts adverse short-term outcomes and mortality in diabetic

mtg

193 (2007) 328–334 333

By the evidence that the postprandial state is a contributingactor to the development of atherosclerosis in diabetic, non-iabetic and non-diabetic critically ill subjects [1–3], it seemsdvisable to consider this new aspect of hyperglycemia, theontrol of hyperglycemic spikes, in the treatment of relatesomplications and diseases in both healthy and ill patients.

Adhesion and migration molecules regulate the interactionetween endothelium and leukocyte, resulting involved in therocess of atherogenesis. An increase in their expression onhe endothelial surface and increased release in the vesselall cause increased adhesion and migration of leukocytes,

n particular monocytes [8–10].Although ICAM-1 is considered as a marker of the activa-

ion of the atherogenic process [9], our data clearly showedhat the treatment of HAECs with short-term HG was suf-cient to induce VCAM-1 and the consequent binding ofuman monocytes, leading to the conclusion that increasedCAM-1 protein expression plays an important role on the

arly stage of the adhesion event. This speculation was fur-her supported by the finding that using a specific monoclonalntibody against VCAM-1, the monocyte adhesion was sig-ificantly inhibited, even if not totally because of multipleCAMs involvement in HG-induced adhesion.

Despite ICAM-1 expression was not significantlynhanced by short-term HG, treatment of HAECs, the pres-nce of monoclonal antibody against CD11b, mainly ligandf ICAM-1, also showed a inhibitory effect on monocytedhesion, suggesting the involvement of ICAM-1 in HG-ediated binding, whereas combination of both antibodiesas maximally effective, even if not totally inhibiting mono-

yte adhesion, showing the evidence of multiple ECAMsnvolvement in HG-induced monocyte-endothelial cell adhe-ion. These findings were in agreement with our previouseports [18], and moreover suggest the principal role ofCAM-1 in the early stage of monocyte adhesion. Short-

erm HG exposure is also effective on the increasing releasef MCP-1 in the culture supernatant that was sufficient ton early transmigration of monocytes, which was neutralizedy specific monoclonal antibody against MCP-1, suggestinghat early increased transmigration activity was mainly dueo MCP-1 release.

There are several evidences supporting the concepthat cronic continuous and/or intermittent-prolonged hyper-lycemia works through the production of oxidative stressy the production of free radicals, that leads to the activa-ion of NF-�B and consequent gene expression [19]. Ouresults suggest that either acute hyperglycemic spikes couldmmediately increase ROS production and consequent earlyF-�B activation, leading to the VCAM-1 and MCP-1 gene

xpression.In conclusion, hyperglycemia can acutely induce alter-

tions of normal human homeostasis. Acute increase in

oglycemic subjects and in diabetic subjects. On the basis ofhis evidence it can be hypothesised that the acute effects oflucose concentrations can add to those produced by chronic

3 clerosis

hps

tdast

R

[

[

[

[

[

[

[

[

[

34 R. Piga et al. / Atheros

yperglycemia, thus contributing to the final picture of com-licated diabetes and its consequences, such as atherosclero-is and cardiac stroke.

Our observations offer an explanation for the predisposi-ion of hyperglycemia to atherosclerosis although in a con-ition of high glucose short exposure of the arterial tree,nd they may provide important suggestions on therapeutictrategies to apply, aimed at avoiding hyperglycemic spikeso prevent its complications and consequences.

eferences

[1] Ceriello A. The post-prandial state and cardiovascular disease: rele-vance to diabetes mellitus. Review. Diabet Metab Res Rev 2000;16:125–32.

[2] Finney SJ, Zekveld C, Elia A, Evans TW. Glucose control and mortalityin critically ill patients. JAMA 2003;290:2041–7.

[3] Laakso M. Hyperglycemia and cardiovascular disease in type 2 dia-betes. Review. Diabetes 1999;48:937–42.

[4] Nomura S, Suzuki M, Katsura K, et al. Platelet-derived microparticlesmay influence the development of atherosclerosis in diabetes mellitus.Atherosclerosis 1995;116:235–40.

[5] Tkac I, Kimball BP, Lewis G, Uffelman K, Steiner G. The severityof coronary atherosclerosis in type 2 diabetes mellitus is related tothe number of circulating triglyceride-rich lipoprotein particles. Arte-rioscler Thromb Vasc Biol 1997;17:3633–8.

[6] Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hyperglycaemiaand increased risk of death after myocardial infarction in patientswith and without diabetes: a systematic overview. Review. Lancet2000;355:773–8.

[7] Ceriello A. The possible role of postprandial hyperglycaemia in

[8] Gerrity RG. The role of the monocyte in atherogenesis. I. Transitionof blood-borne monocytes into foam cells in fatty lesions. Am J Pathol1981;103:181–90.

[

193 (2007) 328–334

[9] Albelda SM, Smith CW, Ward PA. Adhesion molecules and inflamma-tory injury. FASEB J 1994;8:504–12.

10] Gerszten RE, Garcia-Zepeda EA, Lim YC, et al. MCP-1 and IL-8 trig-ger firm adhesion of monocytes to vascular endothelium under flowconditions. Nature 1999;398:718–23.

11] Ishiyama M, Tominaga H, Shiga M, et al. A combined assay of cell via-bility and in vitro cytotoxicity with a highly water-soluble tetrazoliumsalt, neutral red and crystal violet. Biol Pharm Bull 1996;19:1518–20.

12] Yoshida N, Manabe H, Terasawa Y, et al. Inhibitory effects of vita-min E on endothelial-dependent adhesive interactions with leuko-cytes induced by oxidized low density lipoprotein. Biofactors 2000;13:279–88.

13] Reddy S, Hama S, Grijalva V, et al. Mitogen-activated protein kinasephosphatase 1 activity is necessary for oxidized phospholipids to inducemonocyte chemotactic activity in human aortic endothelial cells. J BiolChem 2001;276:17030–5.

14] Chen JW, Chen YH, Lin FY, Chen YL, Lin SJ. Ginkgo biloba extractinhibits tumor necrosis factor-alpha-induced reactive oxygen speciesgeneration, transcription factor activation, and cell adhesion moleculeexpression in human aortic endothelial cells. Arterioscler Thromb VascBiol 2003;23:1559–66.

15] Ryu JW, Hong KH, Maeng JH, et al. Overexpression of uncouplingprotein 2 in THP1 monocytes inhibits beta2 integrin-mediated firmadhesion and transendothelial migration. Arterioscler Thromb VascBiol 2004;24:864–70.

16] Takahashi M, Shibata M, Niki E. Estimation of lipid peroxidation oflive cells using a fluorescent probe, diphenyl-1-pyrenylphosphine. FreeRadic Biol Med 2001;31:164–74.

17] Cao JJ, Hudson M, Jankowski M, Whitehouse F, Weaver WD. Relationof chronic and acute glycemic control on mortality in acute myocardialinfarction with diabetes mellitus. Am J Cardiol 2005;96:183–6.

18] Naito Y, Shimozawa M, Kuroda M, et al. Tocotrienols reduce 25-hydroxycholesterol-induced monocyte-endothelial cell interaction by

19] Nishikawa T, Edelstein D, Du XL, et al. Normalizing mitochondrialsuperoxide production blocks three pathways of hyperglycaemic dam-age. Nature 2000;404:787–90.