Original Contribution

H2O2-INDUCED EGR-1, FRA-1, AND C-JUN GENE EXPRESSION ISMEDIATED BY TYROSINE KINASE IN AORTIC SMOOTH MUSCLE CELLS

NAJIA JIN,* NATHAN D. HATTON,* M AUREEN A. HARRINGTON,† XIAOLIN XIA,† STEVE H. LARSEN,‡ and

RODNEY A. RHOADES**Department of Physiology and Biophysics;†Department of Molecular Biochemistry; and‡Department of Microbiology and

Immunology, Indiana University School of Medicine, Indianapolis, IN, USA

(Received9 August1999;Revised25 April 2000;Accepted22 June2000)

Abstract—Hydrogen peroxide (H2O2) has recently been shown to have a dual effect on cell growth by stimulatingproliferation and triggering apoptosis. Apoptosis induced by H2O2 is a direct consequence of oxidant injury, while theproliferative response to H2O2 is thought to be a protective mechanism against oxidant injury. Signaling of theH2O2-induced proliferative effect has been proposed to occur via the activation of mitogen-activated protein kinase(MAPK) and increase in expression of transcription factors. In the present study, H2O2-induced mitogenic signaling inaortic smooth muscle cells (ASMC) was investigated with a specific focus on the roles of tyrosine kinase and tyrosinephosphatase in the regulation of the H2O2-stimulated egr-1, fra-1, and c-jun transcription. The results show thatH2O2-induced increases in egr-1, fra-1, and c-jun mRNA levels, as measured by Northern blot analysis, are time anddose dependent with the peak of the response within 2 h. Tyrosine kinase inhibitors (genistein, amino-genistein, andtyrphostin 51) significantly attenuated H2O2-induced expression of these genes and a tyrosine phosphatase inhibitor(perox-vanadate) stimulated their expression. H2O2 stimulated tyrosine kinase activities and caused protein tyrosinephosphorylation, which was blocked by tyrphostin 51. H2O2 also caused tyrosine phosphorylation of platelet derivedgrowth factor (PDGF) receptor. These data show that H2O2 increases egr-1, fra-1, and c-jun mRNA levels in vascularsmooth muscle cells, and the increase in expression of these genes is mediated by activation of tyrosine kinase. Our dataalso provide evidence that the H2O2-induced mitogenic response is, in part, mediated through the receptor tyrosinekinase, PDGF receptor. © 2000 Elsevier Science Inc.

Keywords—Vascular smooth muscle, Transcription factors, mRNA, Tyrosine kinase, Tyrosine phosphatase, Apoptosis,Oxidant injury, Free radicals

INTRODUCTION

For decades, H2O2 has been one of the reactive oxygenspecies that has been well investigated in inflammatoryresponses and oxidant-induced stress. Recently, convinc-ing evidence has been presented to show that intracellu-lar H2O2 can act as a signaling molecule, or as a secondmessenger, involved in many cellular functions, such asoxidant-induced stress, apoptosis, and proliferation[1–7]. H2O2 is generated when cells are stimulated withcytokines, growth factors, or vasoactivators like angio-tensin II [4,8,9]. Intracellularly generated H2O2 produces

its effects through the activation of tyrosine kinase orMAP kinase since catalase or N-acetyl-csyteine blockPDGF-induced tyrosine phosphorylation and MAPK ac-tivation [4,10].

Exposure of cells to exogenous H2O2 at pharmaco-logical levels either causes apoptosis or mimics the ef-fects of growth factors. Exogenous H2O2-induced signaltransduction events for apoptosis or for proliferation aresimilar to those initiated by necrosis factor or by growthfactor. The addition of exogenous H2O2 to cells has beenused extensively to investigate cellular signaling path-ways. In vascular smooth muscle cells, H2O2 has beenshown to stimulate the multiple signaling pathways re-lated to smooth muscle contraction [11,12], proliferation[13–15], as well as apoptosis [16].

Activation of tyrosine kinase is one of most early

Address correspondence to: Dr. Najia Jin, Indiana University Schoolof Medicine, Department of Physiology and Biophysics, 635 BarnhillDrive, Science Building 374, Indianapolis, IN 46202, USA; Tel: (317)274-7132; Fax: (317) 274-3318; E-Mail: njin@iupui.edu.

Free Radical Biology & Medicine, Vol. 29, No. 8, pp. 736–746, 2000Copyright © 2000 Elsevier Science Inc.Printed in the USA. All rights reserved

0891-5849/00/$–see front matter

PII S0891-5849(00)00376-2

736

events in response to a variety of mitogens. Inhibition oftyrosine kinase not only blocks mitogen-stimulatedgrowth response but also leads to cell apoptosis, suggest-ing tyrosine kinase plays an very important role in con-trolling cell growth and death [5,17,18]. Although H2O2

has been shown to stimulate tyrosine kinase activity[19–21], little information is known about the role thattyrosine kinase plays in the regulation of H2O2-stimu-lated gene activities, especially on the regulation of egr-1and fra-1 transcription.

The early growth response-1 transcription factor, des-ignated as egr-1, has been reported to be coexpressedwith several growth factor genes [22]. A recent studyshows that the increase in expression of egr-1 is impor-tant for the survival of cells from apoptosis [23]. Fra-1belongs to the c-fos family that forms AP-1 complexeswith c-jun and enhances cell growth and proliferation inresponse to many mitogens and growth factors [24–27].Study of the role of tyrosine kinase in the regulation ofegr-1 and fra-1 transcription will provide important in-formation for better understanding of the mechanism ofcell growth and death during the exposure to reactiveoxygen species. Therefore, in the present study H2O2-induced egr-1, fra-1, as well as c-jun expressions werecharacterized in aortic smooth muscle cells.

The investigation emphasized effects of tyrosinephosphorylation levels, which are controlled by tyrosinekinase and tyrosine phosphatase activities, on egr-1,fra-1, and c-jun mRNA expression. Our data show thatH2O2-stimulated egr-1, fra-1, and c-jun transcription istime and dose dependent with the peak times within 1 h.Tyrosine kinase inhibitors blocked H2O2-stimulatedegr-1 and fra-1 mRNA expression. A tyrosine phospha-tase inhibitor stimulated egr-1 and fra-1 mRNA expres-sion. Western analysis demonstrated that H2O2 stimu-lated tyrosine phosphorylation of proteins, includingtyrosine kinase itself. H2O2 also caused an increase intyrosine kinase activities. These results strongly suggestthat increases in the level of protein tyrosine phosphor-ylation, a direct consequence of either activation of ty-rosine kinase or inhibition of tyrosine phosphatase, me-diates H2O2-stimulated gene expression.

METHODS

Materials and reagents

Cell culture media and reagents, except as specifiedbelow, were from Sigma Chemical Co. (St. Louis, MO,USA). All reagents used for RNA isolation were ofmolecular biology grade. The RNA extraction solution,TRI reagent, which was used to extract total RNA, waspurchased from Molecular Research Center, Inc. (Cin-cinnati, OH, USA). Tyrosine kinase inhibitors, genistein

and tyrphostin 51, were from Sigma Chemical Co. andprepared in DMSO as stock solutions. Amino-genisteinwas from Calbiochem. Tyrosine phosphatase inhibitor,phenanlhrelin 1, 10-bisperox vanadate, is a generous giftof Dr. S. Ganguli (Indiana University). The cDNA ofc-jun, egr-1, fra-1, and CHOB were from American TypeCulture Collection (Rockville, MD, USA). Antiphospho-tyrosine (anti-PY) antibody, anti-PDGF receptor B anti-body, and protein G/A beads were from Santa CruzBiotechnology, Inc. (Santa Cruz, CA, USA). Antismoothmuscle-specifica-actin antibody and 4,6-diamidino-2-phenylindole (DAPI) were from Sigma Chemical Co.32P-g-ATP (3000 Ci/mmol) was purchased from Amer-sham (Braunschweig, Germany). Raytide, the controlpeptide, and the purified p60src was purchased fromOncogene Research Products (Cambridge, MA, USA).

Cell culture

Vascular smooth muscle cells were cultured as de-scribed previously [28] by using a modification of theexplant method described by Ross [29]. Briefly, maleSprague-Dawley rats (300 –325 g) were anesthetizedintraperitoneally with sodium pentobarbital (60 mg/kg). Hearts and lungs were removed from the thoraciccavity and placed in ice-cold (4°C) Dulbecco’s mod-ified Eagle’s medium (DMEM). Rat thoracic aortawere isolated and cleaned of connective and fat tissue.Aortic small branches were cut since the endotheliumon these small branches is difficult to remove. Theaortic endothelium was removed by rotating the ves-sels on a slightly sanded surgical steel rod. Absence ofendothelium was tested by the lack of acetylcholine-induced relaxation as previously described [30]. Theendothelium-denuded aorta was placed in DMEMovernight and then cut into 2 mm2 pieces. Explantswere incubated in DMEM supplemented with 20%fetal calf serum (FCS) in a cell culture incubator (5%CO2 and 95% air, 37°C) for 7 d. Tissue explants wereremoved and fresh DMEM containing 20% FCS wasadded. After reaching confluence, cells were detachedwith 0.05% trypsin in phosphate buffer saline (PBS,pH 7.4). Cells were then counted, seeded at 2503 103

cells/mL, and grown in DMEM media containing 10%FCS. The medium was changed three times a weekuntil the cells reached confluence. Cell purity wasexamined at the first passage on the basis of lightmicroscopic morphology and immunostaining withsmooth muscle-specifica-actin antibody and DAPI.The smooth muscle cell population was determined byDAPI-stained nuclei anda-actin stained cells. Cellswith about 90% purity (passages two through five)were used.

737H2O2-induced signaling in vascular smooth muscle

H2O2 stimulation

Once cells reached confluence, they were serumstarved (0.1% FCS) for 2 d to growth arrest cells andsilence gene activity. Cells were then stimulated eitherwith 1 to 300mM H2O2 for 60 min or with 100mM H2O2

at different time points (10 to 120 min). In addition, cellswere stimulated with 100mM H2O2 for 60 min in thepresence or absence of different inhibitors. Cell viabilitybetween H2O2-stimulated (10–200mM, 60 min) andcontrol cells was compared by using trypan blue meth-ods. Fetal calf serum (10%), which is a potent stimulusfor egr-1, fra-1, and c-jun transcription, was employed asa positive control. At the end of treatment, cells werewashed twice with ice-cold phosphate buffer saline andquickly frozen in liquid nitrogen followed by Northern orWestern analysis.

Northern blot analysis

Total RNA was isolated by extraction of cells with theTRI reagent solution following the manufacturer’s pro-tocol. The isolated total RNA was quantified by spectro-photometry (Beckman, DU640) at 260 and 280 nm.RNA (15 mg) from each sample was fractionated on a1.1% agarose and 2.2 M formaldehyde gel and trans-ferred to a nylon membrane. The membrane was hybrid-ized in a solution (0.25 M Na2HPO4, pH 7.2, 50%formamide, 5% SDS, 1% BSA, 4 mM EDTA, and 0.1mg/ml tRNA) containing 23 107 cpm/ml of32P-labeledcDNA probes. The cDNA probes, c-jun, fra-1, and egr-1,were labeled with32P-a-dCTP by random priming (Am-ersham). After hybridization at 42°C overnight, blotswere washed 15 min each in 23 SSC, 0.1% SDS at roomtemperature, in 0.23 SSC, 0.1% SDS at room tempera-ture, and then washed twice in 0.23 SSC, 0.1% SDS at55°C. Signals were detected by autoradiography. Blotswere dehybridized and reprobed with a Chinese hamsterovary B (CHOB) cDNA as a marker of evenness ofloading of total RNA. The mRNA signals were quanti-fied by a densitometer (GS-670; Biorad; Hercules, CA,USA) and normalized with CHOB signals.

Western blot analysis and immunoprecipitation

To determine if H2O2 causes activation of tyrosinekinase, protein tyrosine phosphorylation was measuredby Western blot analysis. H2O2-stimulated cells or cellspretreated with inhibitors prior to the stimulation werequickly frozen in liquid nitrogen. Frozen cells were son-icated (10 s, 33) in an ice-cold lysis buffer composed of20 mM HEPES (pH 7.4), 80 mMb-glycerophosphate, 1mM EGTA, 2 mM EDTA, 2 mM DTT, 1 mM Na3VO4,0.1 mM PMSF, and 1% Triton X-100. The lysate was

centrifuged for 20 min at 12,0003 g at 4°C to removecellular debris, and the protein concentration of eachsupernatant was determined by the Bradford method.Equal amounts of protein (20mg) from each sample wereresolved on a 10% polyacrylamide gel by electrophore-sis. Proteins were transferred to a nitrocellulose mem-brane. The membranes were blocked for 1 h at roomtemperature with 3% casein in 20 mM Tris, 500 mMNaCl, 0.05% Tween 20. The blots were incubated with1:1000 antiphosphotyrosine (anti-PY) antibody at 4°Covernight, followed by incubation for 1 h with the sec-ondary antibody (horseradish peroxidase–conjugated an-tirabbit IgG, 1:5,000; Biorad). Tyrosine-phosphorylatedproteins were visualized by enhanced chemilumines-cence (ECL; Amersham).

Immunoprecipitation was used to determine if H2O2

activates PDGF receptors. Cell lysate was prepared asdescribed above and 200mg of each protein preparationwas incubated with 5ml of anti-PDGF receptor B anti-body for 2 h at4°C. The immunocomplex was incubatedfor another 2 h with 25 ml of protein G/A beads andwashed 2 times with the lysis buffer. The immunopre-cipitates were eluted into sample buffer by boiling for 5min, and then applied to a 6% polyacrylamide gel. TheWestern blots were probed with anti-PY, as describedabove, and the same blot was then stripped and reprobedwith anti-PDGF receptor B antibody.

Measurement of tyrosine kinase activity

Tyrosine kinase activity was assayed as described byHarvey et al. [31] using a specific exogenous substrate,raytide peptide. Treated cells were lysed and sonicated asdescribed above. Protein concentrations were measuredand 10mg of protein from each sample was incubatedwith tyrosine kinase assay buffer containing 50 mMHEPES, 0.1 mM EDTA, 0.015% Brij-35, 200mM van-adate and 10 mM MgCl, 5mg raytide, and 2mCi 32P-g-ATP in a final volume of 30ml. Reaction mixtureswere incubated for 15 min at 37°C. Another set ofsamples was treated identically except the raytide wasreplaced by control peptide to measure nonspecific phos-phorylation. Phosphorylation was terminated by additionof 15 ml of cold 10% phosphoric acid. Mixtures werecarefully blotted on Whatman P-81 cation exchange pa-per (Fisher Scientific, Pittsburgh, PA, USA). The paperwas washed twice with 0.5% phosphoric acid (10 mineach) and then washed with acetone for 2 min. Radio-activity remaining on the P-81 paper was determined byscintillation spectrometry. Purified tyrosine kinase p60src

was also applied to serve as a positive control for ourassay system. Three micrograms of p60src was used inthe same reaction mixture at the same conditions exceptwithout cell proteins. Assays were performed in dupli-

738 N. JIN et al.

cate. Results were calculated by subtracting the activitiesof the control peptide from that of raytide and the ty-rosine kinase activity is expressed as pmol/mg/min.

Data analysis

Data are expressed as the mean6 standard error fromat least three experiments. The mRNA levels from theegr-1, fra-1, and c-jun and CHOB genes were measuredby densitometry, and expressed as a fold increase in theratio of egr-1, fra-1, or c-jun mRNA versus CHOBmRNA. Tyrosine kinase activity was calculated as pmol/mg/min and the Student’st-test was used to compare thesignificance of the means of two groups.

RESULTS

H2O2 increases egr-1, fra-1, and c-jun mRNA levels incultured aortic smooth muscle cells (ASMC)

To characterize the response of cells to different dosesof H2O2, the serum-starved cells were stimulated for 60min with H2O2 at the concentrations of 1, 10, 25, 50,100, 200, and 300mM. Cells incubated in regular culturemedium (DMEM containing 0.1% FCS) for 60 minserved as control. After each stimulation, cells werewashed twice with cold PBS, quickly frozen in liquidnitrogen, and prepared for Northern blot analysis. Asshown by a typical blot in Fig. 1A, H2O2 caused dose-dependent upregulation of egr-1 mRNA in ASMC. Aconcentration as low as 25mM caused a significantincrease in egr-1 mRNA expression. The signals of egr-1increased as H2O2 concentrations increased and reachedthe maximal response at 300mM in our study.

In additional experiment, the time course of H2O2-induced egr-1 mRNA expression in ASMC was exam-ined. Cells were stimulated with 100mM H2O2 for 10,20, 40, 60, 90, and 120 min. A typical Northern blot inFig. 1B shows that H2O2-induced egr-1 mRNA expres-sion was also time dependent. H2O2-induced egr-1mRNA expression was consistently observed to bemaximal 40–60 min after exposure to H2O2 and dimin-ished thereafter. Mean data from three individual exper-iments (Fig. 1C) show about a 2-fold increase in egr-1mRNA levels at 20 min and a peak time at 40 minfollowing the addition of H2O2. The maximal increase inegr-1 mRNA in response to H2O2 was about 4-foldhigher than the control (n 5 3).

H2O2-induced fra-1 mRNA expression was found tobe dose and time dependent as shown in Fig. 2. Similarlyto its effects on the egr-1 gene, a low concentration ofH2O2 (25 mM) caused a significant increase in fra-1mRNA in ASMC (Fig. 2A). The maximal increases infra-1 mRNA occurred at 100mM H2O2 stimulation.

Further increase in H2O2 concentration (200–300mM)did not cause additional increases in fra-1 mRNA. In-creased fra-1 mRNA appeared as early as 10 min fol-lowing H2O2 stimulation and gradually increased untilreaching the peak at about 90 min (Fig. 2B). Figure 2Cshow mean data from three individual experiments witha more than 3-fold maximal increase in fra-1 mRNA.

Fig. 1. H2O2-induced egr-1 mRNA expression in ASMC. (A) H2O2

increases egr-1 mRNA levels in a concentration-dependent manner.Cultured ASMC were stimulated with H2O2 (1, 10, 25, 50, 100, 200,and 300mM) for 60 min. Northern blots were hybridized with c-juncDNA, and reprobed with CHOB cDNA to demonstrate that approxi-mately equal amounts of total RNA present in each lane. (B) Timecourse of H2O2-induced egr-1 expression. Cells were treated with 100mM H2O2 for 10, 20, 40, 60, 90, and 120 min. Some cells werestimulated with 10% FCS for 1 h toserve as a positive control. (C) Theresponse to 100mM H2O2-induced egr-1 mRNA. Egr-1 mRNA andCHOB mRNA was quantified by densitometry and the average resultsare expressed as a ratio of egr-1 mRNA to CHOB mRNA. Data areexpressed as mean6 SE.} 5 H2O2-stimulated egr-1 mRNA (n 5 3).{ 5 control (n 5 2).

739H2O2-induced signaling in vascular smooth muscle

The kinetics of H2O2-induced c-jun mRNA expres-sion in ASMC is similar to that of egr-1. H2O2 caused adose-dependent upregulation of c-jun mRNA as shownby a typical northern blot in Fig. 3A. A significantincrease in c-jun mRNA expression was seen at 50mMH2O2, and the maximal response appeared at 100–200mM. A blot in Fig. 3B shows that H2O2-induced c-junmRNA expression was also time dependent. H2O2-in-duced c-jun mRNA expression was observed to be max-imal 40–60 min after exposure to H2O2 and returned tothe basal levels after 120 min. Mean data from three

individual experiments (Fig. 1C) show a peak time ofH2O2-induced c-jun mRNA at 40 min. The maximalincrease in c-jun mRNA in response to H2O2 was about4-fold higher than the control (n 5 3).

Tyrosine kinase inhibitors attenuate H2O2-inducedc-jun, fra-1, and egr-1 mRNA expression

Protein tyrosine phosphorylation is an early step insignaling pathways that respond to a variety of growthstimulators. A series of experiments were designed toexamine the role that tyrosine kinases play in H2O2-

Fig. 2. H2O2-induced fra-1 mRNA expression in ASMC. (A) A typicalNorthern blot shows the effect of different concentrations of H2O2 onfra-1 mRNA leves. Cells were stimulated with H2O2 (1, 10, 25, 50, 100,200, and 300mM) for 60 min, northern blots were prepared and probedwith fra-1 and CHOB cDNA. (B) H2O2 (100 mM) induced fra-1expression over time. Cells were treated with 100mM H2O2 for 10, 20,40, 60, 90, and 120 min. Cells stimulated with 10% FCS for 1 h servedas a positive control. (C) Mean data from three individual experiments.� 5 H2O2-stimulated fra-1 mRNA (n 5 3). { 5 control (n 5 2).

Fig. 3. H2O2-induced c-jun mRNA expression in ASMC. (A) A typicalNorthern blot shows H2O2-stimulated c-jun mRNA levels. Cells werestimulated with H2O2 (1, 10, 25, 50, 100, 200, and 300mM) for 60 min.(B) Time course of 100mM H2O2-induced c-jun transcription. Cellswere treated with 100mM H2O2 for 10, 20, 40, 60, 90, and 120 min.(C) Average responses of c-jun mRNA to H2O2. ■ 5 H2O2-stimulatedc-jun mRNA (n 5 3). { 5 control (n 5 2).

740 N. JIN et al.

induced proto-oncogene expression. In these experi-ments, cells were pretreated with tyrosine kinase inhib-itors, genistein (20mM), amino-genistein (20mM), ortyrphostin 51 (20mM), for 1 h prior to the addition ofH2O2 (100 mM, 60 min). The mRNA levels of c-jun,fra-1, and egr-1 in the presence or absence of inhibitorswere analyzed. As shown in Fig. 4, the tyrosine kinaseinhibitors alone had no significant effects on the basallevels of c-jun mRNA expression. However, all threetyrosine kinase inhibitors attenuated H2O2-induced c-junmRNA expression in ASMC. Genistein produced a 70%

inhibition of c-jun mRNA levels in response to H2O2.Amino-genistein, which has a more specific inhibitoryeffect than genistein on tyrosine kinase, caused a 60%inhibition on c-jun mRNA. Tyrphostin, a member of thegenistein family of tyrosine kinase inhibitor, is specific toreceptor tyrosine kinases [32]. Pretreatment of cells withtyrphostin reduced c-jun mRNA of 45%. These datastrongly suggest that tyrosine kinase plays a role in c-junmRNA induction by H2O2.

The effects of tyrosine kinase inhibitors on H2O2-induced fra-1 expression are shown in Fig. 5. Pretreat-ment with genistein (20mM, panel A) or tyrphostin (20mM, panel B) for 1 h inhibited fra-1 gene expressioninduced by H2O2 (100 mM, 60 min). Inhibition of fra-1transcription by genistein and tyrphostin was 45% and50%, respectively. Similarly to their effects on the fra-1gene, both genistein and tyrphostin also blocked egr-1mRNA induction by H2O2 and the inhibition by genisteinand tyrphostin is 60% and 50%, respectively (Figs. 6Aand 6B).

Fig. 4. Effects of tyrosine kinase inhibitors on H2O2-induced c-junexpression. Growth-arrested ASMC were pretreated with 20mMgenistein (A), amino-genistein (B), or tyrphostin 51 (C) for 1 h prior tothe addition of 10% FCS, 5 ng/ml PDGF, or 100mM H2O2. Cellstreated with the tyrosine kinase inhibitor vehicle (DMSO, 1 h) servedas controls. Northern blots shown are representative of two to threeindependent experiments.

Fig. 5. Effects of tyrosine kinase inhibitors on H2O2-induced fra-1expression. (A) Growth-arrested ASMC were pretreated with or with-out genistein (Gen, 20mM) for 1 h prior to the addition of 10% FCSor 100mM H2O2 for 1 h. (B) Cells were stimulated with 10% FCS, 5ng/ml PDGF, or 100mM H2O2 for 1 h in thepresence or absence oftyrphostin 51 (TP 51, 20mM). Northern blots are representative ofthree independent experiments.

741H2O2-induced signaling in vascular smooth muscle

Tyrosine phosphatase inhibitor stimulates c-jun, fra-1,and egr-1 mRNA expression

It is known that levels of protein tyrosine phosphor-ylation are balanced by the relative activities of tyrosinekinase and tyrosine phosphatase. Since inhibition of ty-rosine kinase attenuated H2O2-induced c-jun, fra-1, andegr-1 mRNA expression, the next experiment was de-signed to determine if inhibition of tyrosine phosphataseenhances expression of those mRNAs. ASMC were stim-ulated with 100mM H2O2 in the presence or absence ofdifferent concentrations of phenanlhrelin 1, 10-bisperoxvanadate (perox-vanadate), a potent tyrosine phosphataseinhibitor. At the dose range from 0.1 to 10mM, perox-vanadate did not cause further enhancement of expres-sions of egr-1, fra-1, and c-jun mRNA in response toH2O2 (Figs. 7A, 7B, and 7C).This could be due to amaximum stimulation of these genes by H2O2 itself.However, perox-vanadate alone markedly stimulated ex-pression of these genes (Fig. 7).

H2O2-induced increase in protein tyrosinephosphorylation and tyrosine kinase activity

Tyrosine kinase inhibitors blocked and tyrosinephosphatase inhibitors increased H2O2-induced proto-oncogene expression, suggesting that increase in ty-rosine phosphorylation levels may be involved in theevent of H2O2-induced proto-oncogene expression. Todetermine if H2O2 activates tyrosine kinase, experi-ments were designed to directly measure protein ty-rosine phosphorylation in response to H2O2. ASMCwere stimulated with different concentrations of H2O2

(0, 10, 50, 100, and 200mM) for 10 min. Western boltanalysis demonstrated that H2O2 caused tyrosine phos-phorylation of many proteins (Fig. 8A). Protein ty-rosine phosphorylation induced by H2O2 was dosedependent. When the cells were pretreated with tyr-phostin 51 (20mM) for 1 h prior to the addition ofH2O2 (100 mM, 10 min), the tyrosine phosphorylationwas inhibited (Fig. 8B). To further determine thedirect effect of H2O2 on tyrosine kinase, we measured

Fig. 6. Effects of tyrosine kinase inhibitors on H2O2-stimulated egr-1expression. (A) Growth-arrested ASMC were pretreated with genistein(Gen, 20mM) for 1 h prior to the addition of 10% FCS or 100mMH2O2 for 1 h. (B) Cells were preincubated with tyrphostin 51 (TP 51,20 mM) for 1 h and then exposed to either 10% FCS or 100mM H2O2

for 1 h. Northern blots shown are representative of three independentexperiments.

Fig. 7. Effects of tyrosine phosphatase on H2O2-stimulated c-jun, fra-1,and egr-1 expression. Cells were treated with 0.1, 1, or 10mM perox-vanadate or pretreated with 0.1, 1, or 10mM perox-vanadate for 1 hprior to the addition of H2O2 (100 mM, 1 h). Northern blots werehybridized with c-jun (A), fra-1 (B), or egr-1 (C) cDNA, respectively.The blots shown are representative of four independent experiments.

742 N. JIN et al.

tyrosine kinase activity by using a specific substrate,raytide. In our assay conditions, exogenous purifiedp60src induced raytide phosphorylation to 0.826 0.21pmol/mg/min (n 5 3). Control cells showed an averageof tyrosine kinase activity of 0.0536 0.003 pmol/mg/min (n 5 7). When cells were stimulated with H2O2

(200 mM for 10 min) or PDGF (5ng/ml for 15 min),tyrosine kinase activity increased to 0.1076 0.019(n 5 7) and 0.1146 0.011 (n 5 4) pmol/mg/min,respectively (p , .05). Pretreatment of cells withtyrphostin 51 (20mM for 2 h) decreased H2O2- andPDGF-induced activities to 0.0326 0.01 (n 5 7) and0.042 6 0.011 (n 5 4) pmol/mg/min, respectively(Fig. 9).

H2O2-induced PDGF receptor tyrosinephosphorylation

There are two classic tyrosine kinases: receptor andnonreceptor tyrosine kinases. Receptor tyrosine kinasesactually are the growth factor receptors located in theinner side of the cytoplasmic membrane and subjected todimerization and autophosphorylation upon activation.To further test if H2O2 activates tyrosine kinase, wemeasured tyrosine phosphorylation of one of the receptortyrosine kinases, the PDGF receptor, in response to H2O2

stimulation in cultured aortic smooth muscle cells.ASMC were stimulated with H2O2 (200mM) for 10 min.Cells stimulated with 5 ng/ml PDGF for 15 min served asthe positive control. As shown in Fig. 10, both H2O2 andPDGF caused tyrosine phosphorylation of the PDGFreceptor B in ASMC, and the increases in the phosphor-ylation levels were about 5-fold and 2.5-fold, respec-tively. Pretreatment of cells with tyrosine kinase inhibi-tor, tyrphostin 51, significantly inhibited both H2O2-andPDGF-induced PDGF receptor phosphorylation (Fig.10B, n 5 4).

DISCUSSION

H2O2 has been shown to have opposing effects on cellgrowth. One effect is H2O2-induced cell injury and ap-

Fig. 8. H2O2-stimulated protein tyrosine phosphorylation. (A) Cellswere stimulated with different concentrations of H2O2 (0, 10, 50, 100,and 200mM) for 10 min. Western blots were prepared and probed withantiphosphotyrosine (anti-PY). (B) Effects of tyrosine kinase inhibitoron H2O2-induced tyrosine phosphorylation. Cells were stimulated withH2O2 (100mM, 10 min) in the presence or absence of tyrphostin 51 (20mM). The blots were probed with anti-PY. The blots shown are one ofthree experiments.

Fig. 9. H2O2- and PDGF-induced increases in tyrosine kinase activity.Cells were stimulated with H2O2 (200 mM for 10 min) or PDGF (5ng/ml for 15 min) with or without the 2 h pretreatment of 20mMtyrphostin 51. Each sample containing 10mg of protein was assayed fortyrosine kinase activity in the presence of 5mg raytide as described inMethods. Purified p60src (3 mg, n 5 3) was used to determine raytidephosphorylation in the cell-free preparations. Data are presented asmeans6 SE (n 5 3–7). *p , .05.

743H2O2-induced signaling in vascular smooth muscle

optosis [6,7,16], and the other is H2O2-induced cell pro-liferation [13–15]. These opposing effects depend on theconcentration of H2O2 and the duration of H2O2 treat-ment. H2O2-induced mitogenic effects have been thoughtto be a protective response to oxidant injury [10,21,33].Li et al. [16] has reported that treatment of vascularsmooth muscle cells with 100mM H2O2 for 24 h inducescell apoptosis as shown by a typical “DNA ladder.” Datafrom the present study show that H2O2 treatment causesa significant increase in mRNA levels of several growth-related proto-oncogenes, including egr-1, fra-1, and c-jun(Figs. 1–3). The upregulation of these genes occursquickly and reaches peak within 1 h. Cell viability in theH2O2-stimulated samples (10–200mM) showed no dif-ference from that of controls at a 1 htime point (data notshown). These investigations demonstrate that short-termstimulation with H2O2 activates the signaling pathways

of cell mitogenic responses, while long-term exposure toH2O2 leads to apoptosis.

An important mechanism of proto-oncogene expres-sion in response to H2O2 is the activation of tyrosinekinase as shown by the results of the present study.H2O2-induced egr-1, fra-1, and c-jun mRNA expressionare inhibited by the tyrosine kinase inhibitors genistein,amino-genistein, or tyrphostin 51 (Figs. 4–6), suggest-ing that activation of tyrosine kinase, sequentially in-creasing protein tyrosine phosphorylation, modulatesgene expression in response to H2O2. The tyrosine phos-phatase inhibitor perox-vanadate, which inhibits dephos-phorylation of tyrosine, and therefore, keeps tyrosinephosphorylation at a higher level, induces proto-onco-gene expression (Fig. 7). At the present, the role oftyrosine phosphatase in the regulation of H2O2-inducedgene expression is not clear. Additional experiments onphosphatase activity and phosphatase phosphorylationwill be required to delineate the direct role of tyrosinephosphatase in H2O2-induced egr-1 and fra-1 expression.

Since most kinase or phosphatase inhibitors, includ-ing genistein and vanadate, have nonspecific effects,experiments were further carried out to measure H2O2-induced protein tyrosine phosphorylation and tyrosinekinase activity to confirm the activation of tyrosine ki-nase. By using these direct approaches, we demonstratethat H2O2 stimulates tyrosine kinase activities and causesprotein tyrosine phosphorylation (Figs. 8 and 9). H2O2-induced protein tyrosine phosphorylation is dose depen-dent and inhibited by the tyrosine kinase inhibitor, tyr-phostin 51 (Fig. 8). H2O2-stimulated tyrosine kinaseactivities as measured by tyrosine phosphorylation of aspecific substrate, raytide, is also inhibited by tyrphostin(Fig. 9). These results provide direct evidence of thecapability of H2O2 to activate tyrosine kinases.

H2O2 activates several nonreceptor tyrosine kinases,such as p56lck, p60src, and p72syk [34–36], and alsoactivates receptor tyrosine kinases, such as fibroblastgrowth factor (FGF) receptors and epidermal growthfactor (EGF) receptors [19,21,33]. Our data are the firstto show that H2O2 stimulates tyrosine phosphorylation ofplatelet-derived growth factor (PDGF) receptor B in vas-cular smooth muscle cells (Fig. 10). Our data also showthat H2O2-induced egr-1 and fra-1 expressions are inhib-ited by PDGF receptor inhibitor, tyrphostin 51 (Figs.4–6). Furthermore, our data show that activation ofPDGF receptor by PDGF rather than H2O2 also stimu-lates c-jun and fra-1 expression (Figs. 4C and 5B). PDGFalso strongly stimulated egr-1 mRNA expression (ourunpublished data). Collectively, these data indicate thatsignaling transduction occurrs between H2O2, PDGF re-ceptor, and the transcription factors. The activation ofPDGF receptor plays a role in H2O2-induced egr-1 andfra-1 expression. The level of activation of PDGF recep-

Fig. 10. H2O2- and PDGF-induced PDGF receptor B phosphorylation.Cells were stimulated with H2O2 (200 mM for 10 min) or PDGF (5ng/ml for 15 min) in the presence or absence of 20mM tyrphostin 51.Cell lysates were immunoprecipitated with anti-PDGF receptor B an-tibody. The immunocomplex was resolved on a 6% SDS gel and theWestern blots were probed with antiphosphotyrosine (anti-PY) andsequentially probed with anti-PDGF receptor B antibody after strip-ping. (A) The typical Western blots. The blots shown are one of fourexperiments. (B) The mean data from four individual experiments (n 54, *p , .05).

744 N. JIN et al.

tor by H2O2 is less than that by PDGF, which is similarto the finding of EGF by Knebel et al. [19]. However,PDGF is a potent growth factor for vascular smoothmuscle; and the activation of the PDGF receptor stronglystimulates egr-1, fra-1, and c-jun gene expression.

Several studies provide very firm evidence that H2O2

activates MAPK in a variety of cell types [37–40]. Ourprevious study showed that genistein inhibited H2O2-induced MAPK activation [28]. Studies from severalother laboratories have also reported that tyrosine kinaseinhibitors (tyrphostin and herbimycin) suppressed theactivation of MAPK induced by H2O2 [38,39,41]. In thisstudy, we found that specific tyrosine kinase inhibitorsblocked mRNA expression of c-jun, fra-1, and egr-1.Collectively, these data suggest that H2O2-induced egr-1,fra-1, and c-jun expression is a specific signaling eventcoming from the activation of the tyrosine kinase andMAPK cascade.

In summary, our data demonstrate the mitogenic ef-fect of H2O2 on tyrosine kinase activation and proto-oncogene expression in vascular smooth muscle. H2O2-induced expression of egr-1, fra-1, and c-jun mRNA ismediated by tyrosine kinases, particularly involvedPDGF receptor activation. In addition to our data, recentstudies show that treatment of cells with tyrosine kinaseinhibitors causes apoptosis in a variety of cell types[5,17,18]. These investigations provide an important in-sight into the control mechanism of cell apoptosis andgrowth. Specifically, the level of protein tyrosine phos-phorylation plays a key role in the balance between cellgrowth and death.

Acknowledgements— This work was supported by the American HeartAssociation Indiana Affiliate. The authors thank Dr. Darl R. Swartz forhelp on the cell purity study and Ming Ming Wang and Maggie Yu fortheir technical assistance.

REFERENCES

[1] Gate, L.; Paul, J.; Ba, G. N.; Tew, K. D.; Tapiero, H. Oxidativestress induced in pathologies: the role of antioxidants.Biomed.Pharmacother.53:169–80; 1999.

[2] Lander, H. M. An essential role for free radicals and derivedspecies in signal transduction.FASEB J.11:118–124; 1997.

[3] Rhee, S. G. Redox signaling: hydrogen peroxide as intracellularmessenger.Exper. Mol. Med.31:53–59; 1999.

[4] Sundaresan, M.; Yu, Z. X.; Ferrans, V. J.; Irani, K.; Finkel, T.Requirement for generation of H2O2 for platelet-derived growthfactor signal transduction.Science270:296–299; 1995.

[5] Simizu, S.; Imoto, M.; Masuda, N.; Takada, M.; Umezawa, K.Involvement of hydrogen peroxide production in erbstatin-in-duced apoptosis in human small cell lung carcinoma cells.CancerRes.56:4978–4982; 1996.

[6] Kanno, S.; Ishikawa, M.; Takayanagi, M.; Takayanagi, Y.;Sasaki, K. Characterization of hydrogen peroxide–induced apo-ptosis in mouse primary cultured hepatocytes.Biol. Pharm. Bull.23:37–42; 2000.

[7] Wei, T.; Ni, Y.; Hou, J.; Chen, C.; Zhao, B.; Xin, W. Hydrogenperoxide–induced oxidative damage and apoptosis in cerebellargranule cells: protection by ginkgo biloba extract.Pharmacol.Res.41:427–433; 2000.

[8] Kang, S. W.; Chae, H. Z.; Seo, M. S.; Kim, K.; Baines, I. C.;Rhee, S. G. Mammlia peroxiredoxin isoforms can reduce hydro-gen peroxide generation in response to growth factors and tumornecrosis factor-a. J. Biol. Chem.273:6297–6302; 1998.

[9] Rajagopalan, S.; Kurz, S.; Munzei, T.; Tarpey, M.; Freeman,B. A.; Griendling, K. K.; Harrison, D. G. Angiotensin II–medi-ated hypertension in the rat increases vascular superoxide produc-tion via membrane NADH/NAPH oxidase activation.J. Clin.Invest.97:1916–1923; 1996.

[10] Guyton, K. Z.; Liu, Y.; Gorospe, M.; Xu, Q.; Holbrook, N. J.Activation of mitogen-activated protein kinase by H2O2. Role incell survival following oxidant injury.J. Biol. Chem.271:4138–4142; 1996.

[11] Jin, N.; Rhoades, R. A. Activation of tyrosine kinases in H2O2-induced contraction in pulmonary artery.Am. J. Physiol. (HeartCirc. Physiol. 41)272:H2686–H2692; 1997.

[12] Sotnikova, R. Investigation of the mechanisms underlying H2O2-evoked contraction in the isolated rat aorta.Gen. Pharmacol.31:115–119; 1998.

[13] Baas, A. S.; Berk, B. C. Differential activation of mitogen-activated protein kinases by H2O2 and O2

2 in vascular smoothmuscle cells.Circ. Res.77:29–36; 1997.

[14] Fiorani, M.; Cantoni, O.; Tasinato, A.; Boscoboinik, D.; Azzi, A.Hydrogen peroxide and fetal bovine serum–induced DNA syn-thesis in vascular smooth muscle cells: positive and negativeregulation by protein kinase C isoforms.Biochim. Biophys. Acta1269:98–104; 1995.

[15] Rao, G. N.; Berk, B. C. Active oxygen species stimulate vascularsmooth muscle cell growth and proto-oncogene expression.Circ.Res.70:593–599; 1992.

[16] Li, P. F.; Dietz, R.; Von Harsdor, R. Differential effect of hydro-gen peroxide and superoxide anion on apoptosis and proliferationof vascular smooth muscle cells.Circulation 96:3602–3609;1997.

[17] Constantinou, A. I.; Kamath, N.; Murley, J. S. Genistein inacti-vates bcl-2, delays the G2/M phase of the cell cycle, and inducesapoptosis of human breast adenocarcinoma MCF-7 cells.Eur. J.Cancer34:1927–1934; 1998.

[18] Kyle, E.; Neckers, L.; Takimoto, C.; Curt, G.; Bergan, R.Genistein-induced apoptosis of prostate cancer cells is precededby a specific decrease in focal adhesion kinase activity.Mol.Pharmacol.51:193–200; 1997.

[19] Knebel, A.; Rahmsdorf, H. J.; Ullrich, A.; Herrlich, P. Dephos-phorylation of receptor tyrosine kinases as target of regulation byradiation, oxidants or alkylating agents.EMBO. J.15:5314–5325;1996.

[20] Natarajan, V. Oxidants and signal transduction in vascular endo-thelium.J. Lab. Clin. Med.125:26–37; 1995.

[21] Rao, G. N. Protein tyrosine kinase activity is required for oxidant-induced extracellular signal-regulated protein kinase activationand c-fos and c-jun expression.Cell Signal9:181–187; 1997.

[22] Kahchigian, L. M.; Lindner, V.; Williams, A. J.; Collins, T.Egr-1-induced endothelial gene expression: a common theme invascular injury.Science271:1427–1431; 1996.

[23] Muthukkumar, S.; Han, S. S.; Rangnekar, V. M.; Bondada, S.Role of egr-1 gene expression in B cell receptor-induced apopto-sis in an immature B cell lymphoma.J. Biol. Chem.272:27987–27993; 1997.

[24] Angel, P.; Karin, M. The role of jun, fos and the AP-1 complexin cell proliferation and transformation.Biochim. Biophys. Acta1072:129–157; 1991.

[25] Karin, M. The regulation of AP-1 activity by mitogen-activatedprotein kinases.Phil. Trans. Royal Soc. London, Series B: Bio-logical Sciences.351(1336):127–134; 1996.

[26] Karin, M.; Liu, Zg.; Zandi, E. AP-1 function and regulation.Curr.Opin. Cell Biol.9:240–246; 1997.

[27] Treisman, R. Regulation of transcription by MAP kinases cas-cades.Curr. Opin. Cell Biol.8:205–215; 1996.

[28] Zhang, J. H.; Jin, N.; Liu, Y.; Rhoades, R. A. Hydrogen peroxidestimulates extracellular signal-regulated protein kinases in pulmo-

745H2O2-induced signaling in vascular smooth muscle

nary arterial smooth muscle cells.Am. J. Respir. Cell Mol. Biol.19:324–332; 1998.

[29] Ross, R. The smooth muscle cell. II. Growth of smooth muscle inculture and formation of elastic fibers.J. Cell Biol. 50:172–186;1971.

[30] Jin, N.; Packer, C. S.; Rhoades, R. A. Pulmonary arterial hypoxiccontraction: signal transduction.Am. J. Physiol. (Lung Cell Mol.Physiol. 7)263:L73–78; 1992.

[31] Harvey, K. A.; Siddiqui, R. A.; Jansen, J.; Akard, L. P.; Thomp-son, J. M.; Cui, Y.; Chang, Q.; English, D. Growth factor induc-tion of cytosolic protein tyrosine kinase activity in human hemo-poietic progenitor cells.Br. J. Hematol.93:515–526; 1996.

[32] Majumdar, A. P.; Fligiel, S. E.; Jaszewski, R.; Tureaud, J.; Dutta,S.; Chelluderai, B. Inhibition of gastric mucosal regeneration bytyrphostin: evaluation of the role of epidermal growth factorreceptor tyrosine kinase.J. Lab. Clin. Med.128:173–180; 1996.

[33] Goldkorn, T.; Balaban, N.; Matsukuma, K.; Chea, V.; Gould, R.;Last, J.; Chan, C.; Chavez, C. EGF-receptor phosphorylation andsignaling are targeted by H2O2 redox stress.Am. J. Respir. Cell.Mol. Biol. 19:786–798; 1998.

[34] Hardwick, J. S.; Sefton, B. M. Activation of the lck proteintyrosine kinase by hydrogen peroxide requires the phosphoryla-tion of Tyr-394.Proc. Natl. Acad. Sci. USA92:4527–4531; 1995.

[35] Lee, S. F.; Huang, Y. T.; Wu, W. S.; Lin, J. K. Induction of c-junproto-oncogene expression by hydrogen peroxide through hy-droxyl radical generation and p60src tyrosine kinase activation.Free Radic. Biol. Med.21:437–448; 1996.

[36] Schieven, G. L.; Kirihara, J. M.; Burg, D. L.; Geahlen, R. L.;Ledbetter, J. A. P72 syk tyrosine kinase is activated by oxidizingconditions that induce lymphocyte tyrosine phosphorylation andCa21 signal.J. Biol. Chem.268:16688–16692; 1993.

[37] Abe, J.; Kusuhara, M.; Ulevitch, R. J.; Berk, B. C.; Lee, J. D. Big

mitogen-activated protein kinase 1 is a redox-sensitive kinase.J. Biol. Chem.271:16585–16590; 1996.

[38] Abe, M. K.; Kartha, S.; Karpova, A. Y.; Li, J.; Liu, P. T.; Kuo,W. L.; Hershenson, M. B. Hydrogen peroxide activates extracel-lular signal-regulated kinase via protein kinase C, Raf-1, andMEK1. Am. J. Respir. Cell. Mol. Biol.18:562–569; 1998.

[39] Cantoni, O.; Boscobinik, D.; Fiorani, M.; Stauble, B.; Azzi, A.The phosphorylation state of MAP-kinases modulate the cyto-toxic response of smooth muscle cells to hydrogen peroxide.FEBS Lett.389:285–288; 1996.

[40] Stevenson, M. A.; Pollock, S. S.; Coleman, C. N.; Calderwood,S. K. X-irradiation, phorbol esters, and H2O2 stimulate mitogen-activated protein kinase activity in NIH-3T3 cells through theformation of reactive oxygen intermediates.Cancer Res.54:12–15; 1994.

[41] Whisler, R. L.; Goyette, M. A.; Grants, I. S.; Newhouse, Y. G.Sublethal levels of oxidant stress stimulate multiple serine/threo-nine kinases and suppress protein phosphatases in Jurkat T cells.Arch. Biochem. Biophys.319:23–35; 1995.

ABBREVIATIONS

ASMC—aortic smooth muscle cellsCHOB—Chinese hamster ovary BDMEM—Dulbecco’s modified Eagle’s mediumFCS—fetal calf serumH2O2—hydrogen peroxideMAPK—mitogen-activated protein kinasePDGF—platelet-derived growth factor

746 N. JIN et al.