Endothelial cells are crucial in maintaining physiologicalfunction of the cardiovascular system. Oxidative injury in-cluding ischemia/hypoxia, reperfusion/reoxygenation, and in-flammation can induce cardiac and endothelial cell apopto-sis.1) Recent evidences suggest that mitochondrial reactiveoxygen species are involved in apoptosis because they canreadily damage mitochondrial functions.2,3) Although hydro-gen peroxide (H2O2) itself is not radical, it can attack certainenzymes, oxidize certain keto acids, lead to depletion of ATP,and reduce glutathione and reduced nicotinamide adeninedinucleotide phosphate (NADPH). Therefore in many stud-ies, it had been demonstrated that H2O2 was a pivotal com-pound that triggered numerous signal transduction pathways,including members of the mitogen-activated protein kinasefamily, p53 transcription factors, mitochondrial depolariza-tion, and caspase-3 activation.4,5) This resulted in the typicalapoptotic morphology of cell death (i.e., chromatin condensa-tion, nuclei fragmentation, apoptotic bodies) in many celltypes, and has been used as a model of oxidative stress inneurodegenerative disorders and cerebrocardiac vascular diseases.

Tetramethylpyrazine (TMP) is a major component in thetraditional Chinese herb Chuanxiong (Ligusticum wallichiiFRANCHAT), which is used in China as a new kind of calciumantagonist and an antioxidant for the treatment of cardiovas-cular diseases and myocardial and cerebral ischemic diseasesbecause of its low toxicity and effectiveness.6,7) However,pharmacokinetics studies have shown that TMP presents low

bioavailability and is metabolized quickly in vivo with ashort half-life (T1/2) of 2.89 h.8) Furthermore, accumulatingtoxicity often appears in patients in whom TMP is adminis-tered frequently. Therefore it is necessary to develop a newgeneration of cardio-cerebral vascular medicines from mo-lecular modification of TMP. Results of structure–activity re-lationship studies have indicated that the pyrazine ring in theTMP molecule may largely be the determinant of its pharma-codynamics, while the substituted groups may primarily gov-ern its pharmacokinetics and toxicity. Therefore some drug-like groups and pharmacophores may be introduced to themethyl position of TMP to improve pharmacokinetic proper-ties of TMP.

Calcium channel antagonists such as cinnarizine, flunar-izine, and lomerizine are very important cardio-cerebral vas-cular drugs currently used clinically. The piperazine ring actsas a linker in the molecular structures of these drugs and isconsidered as a functional group of the drug activity. By hy-bridization and bioisosteric replacement, we generated a se-ries of novel ligustrazine derivatives by combination with apiperazine and some pharmacophores or drug-like groups.Previous studies have shown that ligustrazine derivatives ex-hibit lower EC50 values for protecting endothelial cells fromdamage caused by hydrogen peroxide in comparison withTMP.9) Especially, the tetramethylpyrazine diphenylmethylpiperazidine (CXC195; Fig. 1), which is synthesized throughreplacing the methyl group of TMP with (4,4�-fluorine)diphenyl-methyl-piperazidine-methylium group of flunari-

432 Vol. 33, No. 3

Mechanism of Tetramethylpyrazine Analogue CXC195 Inhibition ofHydrogen Peroxide-Induced Apoptosis in Human Endothelial Cells

Yang OU,a Xue DONG,a Xin-Yong LIU,b Xian-Chao CHENG,b Yan-Na CHENG,a Lu-Gang YU,c and Xiu-Li GUO*,a

a Department of Pharmacology, School of Pharmaceutical Sciences, Shandong University; b Institute of MedicinalChemistry, School of Pharmaceutical Sciences, Shandong University; Jinan 250012, P. R. China: and c The HenryWellcome Laboratory of Molecular and Cellular Gastroenterology, Liverpool Centre for Glycobiology, School of ClinicalSciences, University of Liverpool; Liverpool L69 3BX, United Kingdom.Received September 5, 2009; accepted November 16, 2009; published online December 9, 2009

A tetramethylpyrazine analogue, CXC195, was synthesized by the Boekelheide reaction, in which the secondmethyl group of tetramethylpyrazine (TMP) was replaced with (4,4��-fluorine) diphenyl-methyl-1-piperazidine,the active group of flunarizine. We have observed protective effects of CXC195 on vascular endothelial cell sur-vival under oxidative stress in previous study. The aim of the present study was to investigate the effects ofCXC195 against apoptosis induced by hydrogen peroxide in human umbilical vein endothelial cells (HUVECs).Accordingly, a biochemical approach to elucidate the apoptotic signal pathways was attempted. HUVECs wereexposed to 150 mmM H2O2 for 12 h, resulting in an increase of apoptotic cells assessed by the nuclear staining assayand flow cytometry. Mitochondrial membrane potential was detected by retention of rhodamine123. The concen-tration of free intracellular calcium was determined by fura-2/AM fluorometry. Co-incubation with CXC195 reduced the percentage of apoptotic cells and inhibited the loss of mitochondrial membrane potential and intra-cellular calcium overload induced by H2O2. Induction of p53, the activation of caspase-3 by H2O2 which accom-panying downregulation of bcl-2, was blocked by CXC195. In addition, CXC195 clearly improved phosphoryla-tion levels of the antiapoptotic extracellular signal-regulated kinase-1/2 (ERK1/2) in cells undergoing oxidativedamage. Moreover, CXC195 showed stronger effects on inhibition of apoptotic cells and loss of mitochondrialmembrane potential and activation of phosphorylated ERK1/2 than TMP. These results suggest that CXC195prevents reactive oxygen species-induced apoptosis through inhibition of the mitochondria-dependent caspase-3pathway and ERK pathway to show a better beneficial effect in protecting endothelial cells than TMP.

Key words tetramethylpyrazine analogue CXC195; apoptosis; human endothelial cell; hydrogen peroxide; mitochondrial mem-brane potential; extracellular signal-regulated kinase-1/2

Biol. Pharm. Bull. 33(3) 432—438 (2010)

© 2010 Pharmaceutical Society of Japan∗ To whom correspondence should be addressed. e-mail: guoxl@sdu.edu.cn

zine, showed strong protective effects on human umbilicalvein endothelial cells (HUVECs) damaged by H2O2.

In this study, we investigated the effect of CXC195 onapoptosis of endothelial cells induced by H2O2.

MATERIALS AND METHODS

Materials The TMP analogue CXC195 was synthesizedby our lab. Annexin V fluorescein isothiocyanate (AnnexinV-FITC), propidium iodide (PI) and rhodamine123 (Rh123)were purchased from Sigma (St. Louis, MO, U.S.A.). Hydro-gen peroxide (H2O2 30% solution) was obtained from Wako(Osaka, Japan). DNA ladder kit was purchased from Bey-otime Biotechnology (Beyotime, Shanghai, China). Mono-clonal rabbit anti-human extracellular signal-regulated ki-nase-1/2 (ERK1/2) and phospho-ERK1/2 (Thr202/Tyr204)antibodies and antibody against b-tubulin were purchasedfrom Cell Signaling Technology (Boston, MA, U.S.A.). Mon-oclonal rabbit anti-human bcl-2, p53, and caspase-3 antibod-ies were purchased from BD Transduction Lab (San Diego,CA, U.S.A.). Biotinylated goat anti-rabbit immunoglobulin G(IgG), nitrocellulose membranes were from Amersham(Buckinghamshire, U.K.). LumiGLO Reserve Chemilumi-nescent Substrate Kit was from KPL (Gaithersburg, MD,U.S.A.). HUVECs and endothelial cell medium (ECM) werepurchased from Sciencell Research Laboratories (San Diego,CA, U.S.A.).

Cell Culture and Treatment HUVECs were cultured inECM supplement with 10% (v/v) fetal bovine serum, 100 U/ml penicillin and 100 mg/ml streptomycin in a humidifiedatmosphere of 5% CO2/95% air at 37 °C in polylysine-coatedflask. For all experiments, cells were used at passages 4 to 6and seeded at a concentration of 1�105 cells/ml. CXC195 orTMP was freshly prepared as stock solution in dimethyl sul-foxide (DMSO), and diluted with ECM supplement (0.1%(v/v) DMSO). DMSO was present at equal concentrations(0.03%) in all groups. The H2O2 was freshly prepared beforeeach experiment. CXC195 or TMP was pretreated for 30 minbefore cells were exposed to H2O2.

Apoptosis Induction Apoptosis was induced by addi-tion of H2O2. At 70—80% confluence, cells on polylysine-coated glass coverslips or polylysine-coated 6-well plateswere washed with phosphate-buffered saline (PBS), then re-plenished with ECM supplement with 150 mM H2O2 for 12 h.In the same experiments, CXC195 or TMP was pretreatedfor 30 min before H2O2 stimulation in the apoptosis assay.

Assay of Endothelial Cell Apoptosis The changes innuclear morphology of apoptotic cells were investigated bylabeling the cells with the nuclear stain Hoechst 33258. Thecells on the coverslips were fixed in 4% paraformaldehyde inPBS for 30 min, then stained with 10 mg/ml Hoechst 33258under dark conditions at room temperature for 10 min. Cellswere washed with PBS for three times and observed under

the fluorescence microscope (excitation, 340 nm; emission,460 nm) (AX80; Olympus, Tokyo, Japan).

Early apoptosis and late apoptosis/necrosis induced byH2O2 were detected quantitatively by flow cytometric analy-sis using Annexin V and PI.10) Cells with each treatmentwere harvested by non-enzymatic cell dissociation and cen-trifuged (120 g, 5 min) to remove the medium, washed threetimes with binding buffer (10 mM N-(2-hydroxyethyl)piper-azine-N�-2-ethanesulfonic acid (Hepes), 140 mM NaCl, 2.5mM CaCl2) and stained with 10 m l 20-mg/ml Annexin V-FITC. After 30 min of incubation, cells were washed withbinding buffer, incubated with 5 m l PI (final concentration,3.7 mM) for 10 min, then kept on ice without exposure to lightprior to analysis by flow cytometry. Annexin V and PI emis-sions were detected in the FL1-H and FL2-H channels of aFACSVantage flow cytometer (Becton Dickinson Immunocy-tometry System, San Jose, CA, U.S.A.), using emission fil-ters of 525 and 575 nm, respectively.

Measurement of Mitochondrial Membrane PotentialMitochondrial membrane potential was assessed by retentionof Rh123, a membrane-permeable fluorescent cationic dyethat is selectively taken up by mitochondria and is propor-tional to the mitochondrial membrane potential.11) After 12-hincubation in normal medium with or without treatment, thecells were changed to serum-free medium containing 10 mM

Rh123 and incubated in dark at 37 °C for 20 min. Afterwashing with PBS, the cells were analyzed by flow cytome-try with excitation and emission wavelengths of 480 and 530nm, respectively.

Measurement of Intracellular Concentration of Ca2��

([Ca2��]i) Intracellular Ca2� levels were measured by themethod described by Kamouchi et al.12) Confluent HUVECswere dispersed with 0.25% trypsin and 0.02% ethylene-diaminetetraacetic acid (EDTA), then washed thrice withHBSS containing 0.2% BSA. Cells incubated with Fura-2acetoxymethylester (Fura 2-AM; final concentration,5 mmol/l) at 37 °C for 45 min in HBSS. After washing, thecells were incubated for 20 min in a physiological salt solu-tion containing 0.5% BSA and 10 mM glucose. Then, cellswere washed twice with D-Hank’s solution with 0.2% BSAto remove the Fura 2-AM in solution. The Fura 2-AM loadedcells were treated with CXC195 at 10, 50, or 100 mmol/l, orTMP 50 mmol/l for 30 min before incubation with H2O2 foranother 30 min. Fluorescent intensities of Fura-2-loaded sus-pended cells were measured at 37 °C using continuous rapidalternating excitation from dual monochromators (340, 380nm) with emission at 510 nm by 1420 Vitor3 MultilabelCounter. The ratio of fluorescence intensity at 340 and 380nm (F�F340/F380) was calculated. Fluorescence measure-ments were converted to the calcium concentration by deter-mining the maximal fluorescence (Fmax) with Triton X-100(final concentration 0.1%), followed by the minimal fluores-cence (Fmin) with 15 mM ethylene glycol-bis(2-aminoethylether)-N,N,N�,N�-tetraacetic acid (EGTA), pH 10.5 for 15min. The formula, [Ca2�]i�Kd[(F�Fmin)]/(Fmax�F)] wasused for conversion (assuming that Kd for Fura-2–Ca2� com-plex is 224 nm at 37 °C).

Measurement of Protein p53, bcl-2, and Caspase-3 byImmunocytochemistry Expression of protein p53, bcl-2,and caspase-3 of HUVECs was measured by immunocyto-chemistry.13) Cells on the coverslips were rinsed with PBS,

March 2010 433

Fig. 1. Chemical Structure of CXC195

fixed with 4% ice-cold formaldehyde for 30 min, and perme-abilized with 0.1% Triton X-100 in PBS for 2.5 min. Afterblocking with 5% bovine serum albumin in PBS, cells wereincubated with the monoclonal rabbit anti-human antibodiesto p53, bcl-2, and caspase-3 (1 : 100 dilution) at 4 °C overnight.Cells were washed thrice with PBS containing 0.05% Tween20 (TBS) then incubated with biotinylated goat anti-rabbitIgG (1 : 1000 dilution) in TBS as a secondary antibody atroom temperature for 20 min. The protein levels of cells weredetermined by incubation with avidin-conjugated horseradishperoxidase H complex for 20 min and diaminobenzidine andhydrogen peroxide for 3—10 min. The coverslips were thenrinsed in distilled water, counterstained with hematoxylin,and mounted. Images were obtained by an IX51 Olympus Biological Microscopes (Olympus, Tokyo, Japan). To deter-mine the percentage of positive cells, �5 random visualfields/slide were collected to record the intensity of im-munostaining using Image-pro plus image analysis soft(Media Cybernetics, Bethesda, MD, U.S.A.).

Measurement of ERK1/2 Using Western BlottingAfter treatment with reagents, confluent monolayers of cellswere washed twice in ice-cold PBS and lysed with extractionbuffer (20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA,1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophos-phate, 1 mM b-glycerophosphate, 1 mM Na3VO4, 1 mg/ml leu-peptin, and 1 mM PMSF). Protein concentrations of cell extracts were determined by BCA assay (Hyclone-Pierce,South Logan, U.S.A.). Total cell lysate was mixed withLaemmli sample buffer and placed in a boiling water bath for5 min. Proteins were subjected to sodium dodecyl sulfate(SDS)-polyacrylamide gel electrophoresis (PAGE), trans-ferred to a nitrocellulose, and incubated with monoclonalrabbit anti-human antibodies against ERK1/2, phospho-ERK1/2 (Thr202/Tyr204), and b-tubulin. Immunoblots weredeveloped using horseradish peroxidase-conjugated second-ary antibodies.14) Immunoreactive bands were visualized byECL system (Amersham Pharmacia Biotech, Piscataway, NJ,U.S.A.) and quantified by densitometry using an elec-trophoresis image analysis system (FR980, Shanghai FuriScience & Technology, Shanghai, China).

Statistical Analysis Values are expressed as means�S.D. Statistical comparisons were performed by ANOVA fol-lowed by Fisher’s protected least significance difference(PLSD) test. A probability value 0.05 was considered sig-nificant.

RESULTS

Effect of CXC195 on Apoptosis Induced by H2O2 Toevaluate the cytoprotective action of CXC195 on apoptosisinduced by H2O2, the nuclei of HUVECs were stained withHoechst 33258 and assessed by fluorescent microscope.Cells that exhibited reduced nuclear size, chromatin conden-sation, intense fluorescence, and nuclear fragmentation wereconsidered apoptotic. The microscopic pictures in Fig. 2show that the control cells had intact nuclei, whereas theH2O2-treated cells show significant nuclear fragmentation.These changes in nuclear characteristics of apoptosis wererescued significantly in the cells pretreated with CXC195.

In addition to the morphological evaluation, the protectiveeffect of CXC195 against apoptosis was confirmed by flow

cytometric analysis using Annexin V and the PI double-stainsystem. The Annexin V�/PI� population was regarded asnormal healthy cells, while Annexin V�/PI� cells were takenas a measure of early apoptosis and Annexin V�/PI� asnecrosis/late apoptosis. Typical examples are shown in Figs.3A—F. Using this method, we found that the control grouphad 95% intact, living cells and 5% of cells in the earlyand late phases of apoptosis. An increase of apoptotic cellswas observed in the H2O2-treated group with a lower numberof living cells. CXC195 administration led to a reproduci-ble decrease in the rate of early apoptosis and necrosis/lateapoptosis in cells exposed to H2O2. This positive effect ofCXC195 was observed in a dose-dependent manner (Fig.3G). Moreover, 50 mM CXC195 showed better effects thanTMP at an equal dosage on inhibition of early apoptosis andnecrosis/late apoptosis.

Effect of CXC195 on H2O2-Induced Decrease in Mito-chondrial Membrane Potential Exposure of HUVECs toH2O2 for 12 h decreased the fluorescence intensity of Rh123staining, representing a fall in the mitochondrial membranepotential. Both 50 mM and 100 mM CXC195 pretreatment sig-

434 Vol. 33, No. 3

Fig. 2. Inhibitory Effect of CXC195 on H2O2-Induced Apoptosis in HUVECs

Cells were incubated in the presence or absence of CXC195 for 30 min, then exposedto 150 mM H2O2 for 12 h. The cells were stained with the DNA-binding fluorochromeHoechst 33258. Fluorescence micrographs of HUVEC nuclei from untreated cells (A);H2O2-treated cells (B); cells preincubated with 50 mM TMP (C), 10 mM (D), 50 mM (E),and 100 mM (F) CXC195 for 30 min. Scale bar�50 mm. Arrow indicates apoptotic cells.The percentage of apoptotic cells in 100 cells is scored in 4 random observations. Theresults shown in (G) are mean�S.D. Differences with a value of p0.05 were consid-ered statistically significant. ∗ p0.05; ∗∗ p0.01.

nificantly inhibited the fall of mitochondrial membrane po-tential caused by H2O2. Moreover, 50 mM CXC195 showedbetter effects than TMP at an equal dosage (Figs. 4A—H).

Effect of CXC195 on Intracellular Ca2�� ConcentrationThe fluorescent Ca2� indicator, fura 2-AM, was used for themeasurement of intracellular Ca2� concentration ([Ca2�]i). Inthe presence of Ca2� in the culture medium, addition ofCXC195 reduced the intracellular concentration of Ca2� inresponse to subsequent H2O2 stimulation (Fig. 5). At 10, 50,and 100 mmol/l CXC195, the intracellular concentration ofCa2� was 112.45, 102.64, and 96.64 nM, respectively, whereasin the cells with H2O2 alone the figure was 131.8 nM. Thepresence of TMP showed a similar effect, although to lessextent, on [Ca2�]i as CXC195.

Effect of CXC195 on the Expression of Protein bcl-2,p53, and Caspase-3 in H2O2-Treated HUVECs To testwhether CXC195 affected the expression levels of anti-apop-totic bcl-2, negative nuclear transcription factor p53, and theapoptosis marker caspase-3 after exposure to H2O2, specificantibodies of bcl-2, p53, and caspase-3 were used for im-munocytochemistry analysis. The effects of CXC195 at con-centrations of 10, 50 and 100 mM on the expression levels ofbcl-2, p53, and caspase-3 in the H2O2-treated cells are shown

in Figs. 6—8. In cells exposed to H2O2, expression of bcl-2was decreased whereas that of p53 and caspase-3 was en-hanced. However, in H2O2-exposed cells treated with CXC195at concentrations of 50 and 100 mM, TMP 50 mM inhibited the decrease of bcl-2 expression and increased expression ofp53 and caspase-3. These results clearly demonstrate thatCXC195 could prevent oxidant-induced apoptosis throughupregulation of anti-apoptotic bcl-2 and downregulation ofapoptotic markers, protein p53 and caspase-3.

Involvement of ERK Pathway in Anti-apoptotic Actionof CXC195 The expression levels of ERK1/2 and phos-phorylated ERK1/2 were examined because these kinaseshave been shown to regulate apoptosis. Expression of phos-phorylated ERK1/2 in cells treated with CXC195 for 12 hwas not changed. However, phosphorylation level of ERK1/2was elevated after exposure to H2O2, with no significantchange in the total protein level (Fig. 9). CXC195 signifi-cantly enhanced the phosphorylation level of ERK1/2 in cellstreated with H2O2. Moreover, 50 mM CXC195 showed bettereffects than TMP at an equal dosage.

DISCUSSION

CXC195 has previously been shown to prevent H2O2 in-jury in endothelial cells during oxidative stress by increasingthe activities of antioxidative enzymes. The present studyshows that CXC195 protects HUVECs against apoptosis in-duced by H2O2 through the mitochondria-dependent caspase-3 pathway and ERK pathway. CXC195 attenuated intracellu-lar Ca2� overload and inhibited the decrease of mitochondrialmembrane potential and activation of p53 transcription factortriggered by H2O2 injury, which included modulation of ex-pression of its downstream genes of bcl-2, a component ofthe apoptotic death cascade. This is the first report on the directsurvival-promoting effects of CXC195 on endothelial cells.

Reactive oxygen species such as superoxide anion, hy-droxyl radicals, and H2O2 are unwanted toxic substances. Ex-cessive production of reactive oxygen species in cells can ei-ther directly or indirectly lead to mitochondrial dysfunction,apoptosis, and cell death.15) During ischemia/reperfusion, mi-tochondrial dysfunction is a critical event that triggers apop-tosis, causes energy depletion, and ultimately leads to celldeath.16) The decrease in mitochondrial membrane potentialinduced by H2O2 further indicates the disruption of mito-chondrial membrane integrity, whereby reactive oxygenspecies produced in mitochondria may then leak to the cyto-plasm, lead to oxidative stress, and initiate cell death viaactivation of apoptosis signaling.17) Reactive oxygen speciescan induce the release of cytochrome c from mitochondria,stimulating proteolytic caspases. However, reactive oxygenspecies can be blocked or delayed by various antioxidants,such as hydrogen- or electron-donating compounds.18) TMPis one of the most important active ingredients of the Chi-nese herb Ligusticum wallichii FRANCHAT, which is used forthe treatment of many ischemic disorders as an antioxidantand a new kind of calcium antagonist. Previous studies haveshown that TMP significantly protects vascular endothelialcells against hypoxia and angiotensin (Ang) II-induced dam-age by scavenging intracellular reactive oxygen species for-mation.19,20) CXC195 is a novel derivative of TMP via molec-ular modification of TMP to improve its pharmacokinetic

March 2010 435

Fig. 3. Inhibitory Effect of CXC195 on Cell Apoptosis and Necrosis In-duced by H2O2

Cells were incubated in drug-free medium (A) or medium containing 150 mM H2O2

(B) for 12 h; or cells were preincubated with 50 mM TMP (C) or 10 mM (D), 50 mM (E),and 100 mM (F) CXC195 for 30 min and then exposed to H2O2 for 12 h. Distinction be-tween living, early apoptotic, and late apoptotic/necrotic cells and examples of dot-plots were determined by flow cytometry following annexin V and PI double-staining.Horizontal axis represents annexin V intensity and vertical axis shows PI staining. The results shown in (G) are mean�S.D. for three independent experiments. Differenceswith a value of p0.05 were considered statistically significant. ∗∗ p0.01; ΔΔ p0.01vs. untreated cells; ## p0.01 vs. H2O2-treated cells; ξp0.05 vs. TMP-treated cells.

features and action. In the present study, CXC195 inhibitedintracellular calcium concentration increase by H2O2 simi-larly to TMP, and strongly inhibited decrease of mitochondr-ial membrane potential in response to H2O2, prevented apop-totic morphological and biochemical changes, inhibited theexpression of caspase-3 in HUVECs, and reduced the extent

of apoptotic cell death. These results suggest that CXC195may attenuate apoptosis through mitochondrial-dependentcaspase-3 pathway and the calcium antagonism. Further touncover the mechanism underlying such CXC195-mediatedcell protection during oxidative stress, the expression of p53,bcl-2, ERK1/2 and phosphorylated ERK1/2 proteins in HU-VECs was determined.

The tumor suppressor gene p53 appears to affect the mito-chondrial membrane potential through reactive oxygenspecies, and may induce apoptosis.21,22) Upon exposure toDNA-damaging agents, p53 downregulates the expression ofbcl-2, an anti-apoptotic protein.23) It has recently been shownthat p53, activated by NF-kB, is essential for H2O2-inducedapoptosis in human endothelial cells.13) Bcl-2-expressingcells have an enhanced antioxidant status that suppresses ox-idative stress signals.24) In the present study, we observed sig-nificant induction of p53 following H2O2 insult. However,CXC195 blocked activation of p53 transcriptional factor fol-lowing DNA damage caused by H2O2 injury, and upregulatedexpression of its downstream gene protein bcl-2 in HUVECs.

The results of this study confirm those of previous reportsthat treatment with H2O2 can increase phosphorylation ofERK1/2, a posttranslational modification that is associatedwith activation of ERK1/2.25) H2O2-induced transient activa-tion of the ERK pathway is thought part of the cell defense

436 Vol. 33, No. 3

Fig. 4. CXC195 Attenuated H2O2-Induced Decrease of Mitochondrial Membrane Potential

HUVECs exposed to 150 mM H2O2 with/without CXC195 for 30 min were incubated with Rh123, then the fluorescence intensity was measured. (A) Blank (cells incubated with-out Rh123); (B) H2O2-untreated cells; (C) H2O2-treated cells; (D) cells pretreated with 50 mM TMP for 30 min before H2O2; cells pretreated with 10 mM (E); 50 mM (F); 100 mM (G)CXC195 for 30 min before H2O2. FI: mean fluorescence intensity. Results of mitochondrial membrane potential are shown in (H). Mitochondrial membrane potential of H2O2-un-treated cells is defined as 100% and the data are expressed as percentage of the untreated cells, mean�S.D.; n�5. Differences with a value of p0.05 were considered statisticallysignificant. ∗ p0.05; ∗∗ p0.01.

Fig. 5. Effect of CXC195 on Intracellular Ca2� Concentration in HUVECs Injured by H2O2

The Fura 2-AM loaded cells were treated with CXC195 at 10, 50, 100 mmol/l, orTMP 50 mmol/l for 30 min followed by incubation with H2O2 for another 30 min. Con-trol group was treated with culture medium only. Fluorescence measurements wereconverted to [Ca2�]i by determining the maximal fluorescence and the minimal fluores-cence. The data are expressed as mean�S.D.; n�8. Differences with a value of p0.05were considered statistically significant. ∗ p0.05; ∗∗ p0.01.

March 2010 437

Fig. 8. Effect of CXC195 on H2O2-Induced Elevation of ApoptosisMarker Caspase-3

HUVECs were pretreated with CXC195 for 30 min. Then, cells were stimulated withH2O2 (150 mM) for 12 h. Immunocytochemistry was performed with monoclonal rabbitanti-human antibody to p53 and a biotinylated goat anti-rabbit IgG. The protein level wasdetermined by incubation with avidin-conjugated horseradish peroxidase H complex anddiaminobenzidine and hydrogen peroxide (light brown). (A) H2O2-untreated cells; (B)H2O2-treated cells; (C) cells pretreated with 50 mM TMP for 30 min before H2O2; cellspretreated with 10 mM (D); 50 mM (E); 100 mM (F) CXC195 for 30 min before H2O2. Rep-resentative micrographs and quantitative data evaluated by densitometry (G) are shown(n�3). Scale bar�50 mm. The data are expressed as mean�S.D. Differences with a valueof p0.05 were considered statistically significant. ∗ p0.05; ∗∗ p0.01.

Fig. 9. Effect of CXC195 on H2O2-Induced Activation of ERK1/2

HUVECs were pretreated with 50 or 100 mM CXC195 or 50 mM TMP for 30 min.Then, cells were stimulated with/without H2O2 (150 mM) for 12 h for determination ofERK1/2 and phosphorylated ERK1/2 (p-ERK1/2) activity. Cells were harvested, lysed,and used for Western blot analysis. The activities of ERK1/2 and p-ERK1/2 weremeasured as described in Materials and Methods. Representative blots (A, cells treatedwithout H2O2; B, cells treated with H2O2) and quantitative data evaluated by densitom-etry (C) are shown (n�3). The data are expressed as mean�S.D. Differences with avalue of p0.05 were considered statistically significant. ∗ p0.05; ∗∗ p0.01.

Fig. 6. Effect of CXC195 on bcl-2 Protein Levels in HUVECs Damagedby H2O2

HUVECs were pretreated with CXC195 for 30 min. Then, cells were stimulated withH2O2 (150 mM) for 12 h. Immunocytochemistry was performed with monoclonal rabbitanti-human antibody to bcl-2 and a biotinylated goat anti-rabbit IgG. The protein levelwas determined by incubation with avidin-conjugated horseradish peroxidase H complexand diaminobenzidine and hydrogen peroxide (light brown). (A) H2O2-untreated cells;(B) H2O2-treated cells; (C) cells pretreated with 50 mM TMP for 30 min before H2O2;cells pretreated with 10 mM (D); 50 mM (E); 100 mM (F) CXC195 for 30 min before H2O2.Representative micrographs and quantitative data evaluated by densitometry (G) areshown (n�3). Scale bar�50 mm. The data are expressed as mean�S.D. Differences witha value of p0.05 were considered statistically significant. ∗ p0.05; ∗∗ p0.01.

Fig. 7. Effect of CXC195 on H2O2-Induced Activation of p53 in HUVECs

HUVECs were pretreated with CXC195 for 30 min. Then, cells were stimulated withH2O2 (150 mM) for 12 h. Immunocytochemistry was performed with monoclonal rabbitanti-human antibody to p53 and a biotinylated goat anti-rabbit IgG. The protein level wasdetermined by incubation with avidin-conjugated horseradish peroxidase H complex anddiaminobenzidine and hydrogen peroxide (light brown). (A) H2O2-untreated cells; (B)H2O2-treated cells; (C) cells pretreated with 50 mM TMP for 30 min before H2O2; cellspretreated with 10 mM (D); 50 mM (E); 100 mM (F) CXC195 for 30 min before H2O2. Rep-resentative micrographs and quantitative data evaluated by densitometry (G) are shown(n�3). Scale bar�50 mm. The data are expressed as mean�S.D. Differences with a valueof p0.05 were considered statistically significant. ∗ p0.05; ∗∗ p0.01.

mechanism.26) However, it has been shown that ERK’s rela-tively stable activation contributes to antioxidant defensemechanisms.27) Indeed, the authors found that CXC195 pro-duced an elevation of ERK phosphorylation. The underlyingmechanism might be related to its antioxidation. The resultsindicate that the protective effect of CXC195 in H2O2-induced apoptosis of HUVECs is associated with the ERKpathway. Further research will be performed to investigatethe role of CXC195 on intracellular anti-apoptotic PI3-kinase/Akt pathway.

In conclusion, the antioxidant CXC195 diminishes cellapoptosis by promoting H2O2-induced mitochondrial mem-brane potential reduction and inhibiting caspase-3 activation.Moreover, here we show that CXC195 treatment of cellsgrown in H2O2 reduces expression of p53 and increases ex-pression of bcl-2 and the level of ERK phosphorylation.These findings suggest that CXC195 may have important po-tential for development of new agents for effective treatmentof vascular diseases.

Acknowledgement This work was funded by the Na-tional Science Foundation of China Grants (30672451).

REFERENCES

1) Aoki M., Nata T., Morishita R., Matsushita H., Nakagami H., Ya-mamoto K., Yamazaki K., Nakabayashi M., Ogihara T., Kaneda Y.,Hypertension, 38, 48—55 (2001).

2) Orrenius S., Drug Metab. Rev., 39, 443—455 (2007).3) Ozben T., J. Pharm. Sci., 96, 2181—2196 (2007).4) Rácz B., Gasz B., Borsiczky B., Gallyas F. Jr., Tamás A., Józsa R., Lu-

bics A., Kiss P., Röth E., Ferencz A., Tóth G., Hegyi O., Wittmann I.,Lengvári I., Somogyvári-Vigh A., Reglödi D., Gen. Comp. En-docrinol., 153, 115—123 (2007).

5) Wang B., Peng L., Zhu L., Ren P., Colloids Surf. B Biointerfaces, 60,36—40 (2007).

6) Xu H., Shi D. Z., Guan C. Y., Zhongguo Zhong Xi Yi Jie He Za Zhi,23, 376—379 (2003).

7) Li L. L., Zhang Z. R., Gong T., He L. L., Deng L., J. Pharm. Biomed.Anal., 41, 1083—1087 (2006).

8) Ye Y., Wang S., Jiang J., Zhongguo Yi Xue Ke Xue Yuan Xue Bao, 18,288—291 (1996).

9) Cheng X. C., Liu X. Y., Xu W. F., Guo X. L., Ou Y., Bioorg. Med.Chem., 15, 3315—3320 (2007).

10) Guo S., Bezard E., Zhao B., Free Radic. Biol. Med., 39, 682—695(2005).

11) Yu Z., Wang R., Xu L., Dong J., Jing Y., Cancer Lett., 269, 165—173(2008).

12) Kamouchi M., Kitazono T., Ago T., Wakisaka M., Kuroda J., Naka-mura K., Hagiwara N., Ooboshi H., Ibayashi S., Iida M., Neurosci.Lett., 416, 12—16 (2007).

13) Lee Y. J., Kang I. J., Bünger R., Kang Y. H., Microvasc. Res., 66, 91—101 (2003).

14) Yu J., Eto M., Kozaki K., Akishita M., Okabe T., Ouchi Y., Eur. J.Pharmacol., 589, 32—36 (2008).

15) Sugawara T., Chan P. H., Antioxid. Redox Signal., 5, 597—607 (2003).16) Jones S. P., Teshima Y., Akao M., Marbán E., Circ. Res., 93, 697—699

(2003).17) Budihardjo I., Oliver H., Lutter M., Luo X., Wang X., Annu. Rev. Cell

Dev. Biol., 15, 269—290 (1999).18) Carmody R. J., Cotter T. G., Redox Rep., 6, 77—90 (2001).19) Lin R., Liu J. T., Gan W. J., Wang W. R., Zhongguo Zhong Yao Za Zhi,

29, 462—465 (2004).20) Lee W. S., Yang H. Y., Kao P. F., Liu J. C., Chen C. H., Cheng T. H.,

Chan P., Clin. Exp. Pharmacol. Physiol., 32, 845—850 (2005).21) Datta K., Babbar P., Srivastava T., Sinha S., Chattopadhyay P., Int. J.

Biochem. Cell Biol., 34, 148—157 (2002).22) Phillips D. C., Allen K., Griffiths H. R., Arch. Biochem. Biophys., 407,

15—24 (2002).23) Li P. F., Dietz R., von Harsdorf R., EMBO J., 18, 6027—6036 (1999).24) Voehringer D. W., Meyn R. E., Antioxid. Redox Signal., 2, 537—550

(2000).25) Zhang L., Jope R. S., Neurobiol. Aging, 20, 271—278 (1999).26) Guyton K. Z., Liu Y., Gorospe M., Xu Q., Holbrook N. J., J. Biol.

Chem., 271, 4138—4142 (1996).27) Dugan L. L., Creedon D. J., Johnson E. M. Jr., Holtzman D. M., Proc.

Natl. Acad. Sci. U.S.A., 94, 4086—4091 (1997).

438 Vol. 33, No. 3