Mechanical stress activates protein kinase cascade of phosphorylation in neonatal rat cardiac myocytes

(1)

Mechanical stress activates protein kinase

cascade of phosphorylation in neonatal rat

cardiac myocytes.

T Yamazaki, … , K Ueki, K Tobe

J Clin Invest.

1995;

96(1)

:438-446.

https://doi.org/10.1172/JCI118054

.

We have previously shown that stretching cardiac myocytes evokes activation of protein

kinase C (PKC), mitogen-activated protein kinases (MAPKs), and 90-kD ribosomal S6

kinase (p90rsk). To clarify the signal transduction pathways from external mechanical stress

to nuclear gene expression in stretch-induced cardiac hypertrophy, we have elucidated

protein kinase cascade of phosphorylation by examining the time course of activation of

MAP kinase kinase kinases (MAPKKKs), MAP kinase kinase (MAPKK), MAPKs, and

p90rsk in neonatal rat cardiac myocytes. Mechanical stretch transiently increased the

activity of MAPKKKs. An increase in MAPKKKs activity was first detected at 1 min and

maximal activation was observed at 2 min after stretch. The activity of MAPKK was

increased by stretch from 1-2 min, with a peak at 5 min after stretch. In addition, MAPKs and

p90rsk were maximally activated at 8 min and at 10 approximately 30 min after stretch,

respectively. Raf-1 kinase (Raf-1) and (MAPK/extracellular signal-regulated kinase) kinase

kinase (MEKK), both of which have MAPKKK activity, were also activated by stretching

cardiac myocytes for 2 min. The angiotensin II receptor antagonist partially suppressed

activation of Raf-1 and MAPKs by stretch. The stretch-induced hypertrophic responses such

as activation of Raf-1 and MAPKs and an increase in amino acid uptake was partially

dependent on PKC, while a PKC inhibitor completely abolished MAPK activation by

angiotensin II. These results […]

Research Article

Find the latest version:

(2)

Mechanical Stress

Activates Protein Kinase

Cascade

of Phosphorylation

in Neonatal Rat Cardiac

Myocytes

Tsutomu Yamazaki, Issei Komuro, Sumiyo Kudoh, Yunzeng Zou, Ichiro Shiojima, Takehiko Mizuno, Hiroyuki Takano, YukioHiroi, Kohjiro Ueki, Kazuyuki Tobe, Takashi Kadowaki, Ryozo Nagai, and Yoshio Yazaki

Third Department of Medicine, Universityof Tokyo School of Medicine, Tokyo 113,Japan

Abstract

Wehavepreviously shownthatstretchingcardiac myocytes evokes activation of protein kinase C(PKC), mitogen-acti-vated protein kinases (MAPKs), and90-kD ribosomal S6 kinase

(p9OI"k).

To clarify the signal transduction pathways from external mechanical stress to nuclear gene expression instretch-induced cardiac hypertrophy, we have elucidated protein kinase cascade of phosphorylation by examining the time course of activation of MAP kinase kinase kinases

(MAPKKKs),MAPkinase kinase (MAPKK), MAPKs,and

p99OTkin neonatal ratcardiacmyocytes. Mechanical stretch transiently increased the

activity

ofMAPKKKs. Anincrease in MAPKKKsactivitywasfirstdetectedat1 min and

maxi-mal activation was observed at 2 min after stretch. The

activity of MAPKKwasincreasedbystretchfrom 1-2

min,

withapeakat5 minafterstretch. Inaddition,MAPKs and

p9Orsk were maximally activated at8 min and at 10 - 30

min after stretch, respectively. Raf-1 kinase (Raf-1) and

(MAPK/extracellular signal-regulated kinase) kinase ki-nase(MEKK), both of which have MAPKKK

activity,

were also activated by stretching cardiac myocytes for 2 min. Theangiotensin II receptorantagonist partially suppressed activation of Raf-1 and MAPKs by stretch. The stretch-inducedhypertrophicresponses suchasactivation ofRaf-1

and MAPKs and an increase in amino acid uptake was

partially dependent on PKC, while aPKC inhibitor

com-pletely abolishedMAPK activationby angiotensinII.These results suggest thatmechanical stressactivates the protein

kinase cascade of phosphorylation in cardiac myocytes in

the order of Raf-1 and MEKK, MAPKK, MAPKs and

p9oI~k,

and thatangiotensin II,which may be secreted from stretched myocytes, may be partly involved in stretch-in-duced hypertrophic responses byactivating PKC. (J. Clin. Invest. 1995.96:438-446.) Keywords: cardiachypertrophy

*signal transduction*Raf-1kinase *

mitogen-activated

pro-tein kinase kinase !mitogen-activated

protein

kinase

Introduction

Cardiachypertrophyis animportantcauseofincreased

morbid-ity and mortalmorbid-ity (1).Thehypertrophiedheartexhibitsimpaired

Addresscorrespondence to IsseiKomuro, Third Department of Medi-cine,University ofTokyoSchoolofMedicine,7-3-1 Hongo, Bunkyo-ku, Tokyo 113,Japan. Phone: 81-3-3815-5411 ext. 3127; FAX: 81-3-3815-2087.

Receivedfor publication3 November 1994 andacceptedinrevised

form22March 1995.

contraction andrelaxation, and reduced coronary reserveafter transient ischemia, both of which lead to higher mortality in heart failure and ischemic heart disease.Manylines of evidence have suggested that mechanical stress plays a key role in pro-ducing cardiac hypertrophy during hemodynamic overload (for

areview,seereference 2). Both in vivo and in vitro studies have demonstrated that myocyte hypertrophy induced by pressure overload is characterizedby the expression of some immediate early genes as an early event and of fetal type genes as a late event, as well as by an increase in protein synthesis (2-6). While the outcome of hypertrophy has been extensively charac-terized, little is known concerning the signaling pathways that link the external mechanicalstressand the cellular and nuclear

eventsduring hypertrophy. Effective prevention and regression of cardiac hypertrophy require a better understanding of the molecular mechanisms of cardiac hypertrophy.

External stimuli are generally transduced into the nucleus through the protein kinase cascade of phosphorylation. It has been reported that protein kinase C

(PKC)'

is activated by stretching cardiac myocytes (5). Our recent studies have dem-onstrated that mitogen-activatedprotein kinases (MAPKs) are activatedbymechanicalstressduringcardiac myocyte hypertro-phythrough both PKC-dependent and independent pathways (7). The upstream activator of MAPKs is reported to be MAP kinase kinase (MAPKK) (8-10), which is a dual-specificity protein kinase that phosphorylates MAPKs on both threonine and tyrosine residues within a conserved threonine-glutamic acid-tyrosine motif (11). MAPKK is also activated by

phos-phorylationoftwoserine residues (12). Current candidates for this role in mammalian cells are the protooncogene protein

kinase Raf-I kinase (Raf-I) (13) and (MAPK/extracellular signal-regulated kinase) kinase kinase (MEKK) (14). Raf-I has been reported in many cell types to be activated through

PKC-dependent andindependent pathways (15), and MEKK has beensuggestedtomediateprimary signals originatingfrom receptors that activate G proteins and PKC (14). Activated

MAPKstranslocateinto the nucleus and activate nuclear

tran-scriptionfactors suchasc-myc,NF-IL6,and Elk-I(forareview,

see reference 16). Moreover, 90-kD ribosomal S6 kinase

(p90rsk)

isreportedtobe a downstreamenzymeof MAPKs(17)

and tophosphorylate nuclear lamins (18). These observations

1. Abbreviations used in thispaper: Ang II, angiotensin II; MAPK, mitogen-activated protein kinase; MAPKK, MAP kinase kinase;

MAPKKK, MAP kinase kinase kinase; MBP, myelin basic protein;

MEKK, (MAPK/extracellular signal-regulated kinase)kinasekinase;

p90,k,90-kDribosomalS6kinase; PKC, proteinkinaseC;Raf-1, Raf-1 kinase; a-raf, anti-Raf-l kinase antibodies; rMAPK, recombinant MAPK fused to maltose binding protein; rMAPKK, recombinant MAPKK fused to glutathione S transferase; TPA, 12-0-tetradeca-noylphorbol-13-acetate.

J. Clin. Invest.

(3)

suggest thatMAPKs andp9o0Tkplayvital roles intransducing signals into thenucleus.

The aim of thisstudyis toelucidate the signal transduction pathways in cardiac hypertrophy evoked by mechanicalstress.

For this purpose, we examined the activation of protein kinases such as Raf-1, MEKK, MAPKK, MAPKs, and p9o0k after stretchingmyocytes.Mechanicalstressinduces sequential phos-phorylation and activation of these protein kinases in the order

of Raf-1, MEKK, MAPKK, MAPKs, and

p9O'".

A growing body of data suggests that all components of renin-angiotensin systemexist in the heart and that angiotensin H (AngH) can

induce cardiachypertrophy (19). Moreover,recentreports sug-gestthatAngHis involved in mechanical stress-induced cardiac

hypertrophy(20-22). In the present study,wethen presentthe

data suggesting that activation of the protein kinases by

stretch-ing myocytes isinduced both through Ang H-dependent and independent pathways. While the activationofprotein kinases by mechanical stress is partially dependent on PKC, Ang H, which may be secreted by stretched myocytes into theculture

media, induces activation of protein kinases in a mostly PKC-dependent manner.

Methods

[ y 32p]ATPand[3H]phenylalanine were purchased from Du Pont-New

EnglandNuclear Co. (Boston, MA);DMEMand fetal bovine serum

(FBS) from GIBCOBRL Co.(Gaithersburg, MD);polyclonal antibody againstMEKKfromSanta CruzBiotechnology, Inc., (Santa Cruz, CA);

okadic acid from Wako Pure ChemicalIndustries, Ltd., Osaka, Japan;

and ProtoBlotimmunoblottingsystem fromPromegaBiotec(Madison,

WI). Other reagents were purchased from Sigma Chemical Co. (St.

Louis, MO).

Cell cultureandstretching ofcardiac myocytes. Primary cultures of cardiac myocytes wereprepared fromventricles of l-d-oldWistar

rats asdescribedpreviously (5), basically accordingtothe methodof Simpsonetal.(23). Stretching ofmyocyteswasconductedasdescribed previously (5, 6). In brief, cardiac myocytes wereplated at a field density of 1 x 105cells/cm2 on silicone rubber culture dishes. The

culturemedium waschanged 24hafter seedingto asolutionconsisting

of DME containing 0.1% FBS. Cardiac myocytes were stretched by 20%,andlysedonice withbufferAcontaining 25mMTris-HCl,25 mMNaCl, 1 mMsodium orthovanadate, 10 mMNaF, 10 mM sodium

pyrophosphate, 10 mM okadic acid, 0.5 mM ethylene glycol-bis

(b-aminoethyl ether)N,N,N',N'tetraaceticacid(EGTA),and 1 mM phe-nyl-methylsulfonyl fluoride. Stretchand controlexperimentswere

car-riedoutsimultaneouslywiththesamepool ofcells in eachexperiment

tomatchfortemperature,CO2content, orpH ofthe medium between

stretchedand control cells.

Assay ofMAPKKK activity ofRaf-J or MEKK. Both Raf-l and MEKK have been reported to have MAP kinase kinase kinase

(MAPKKK)activity (13, 14).MAPKKKactivitywasassayed by

mea-suringthephosphorylationof recombinant MAPKK fusedtoglutathione

S transferase(rMAPKK)(24). TherMAPKKfusion proteinwasmade

by insertingmouseMAPKK1 cDNA intopGEX-2Tvector.Totalcell

lysatesortheimmunoprecipitateswith 20

jig/ml

ofanti-Raf-lantibody

(a-raf)or 10

_Ag/ml

of anti-MEKK antibodywereincubated with the substrate(100_,grMAPKK) and 2

_{ILCi [-y-32P]ATP}

inbuffer B

con-taining25 mMTris-HCl (pH 7.4), 10mMMgCl2, 1 mMDTT,40

_,M

ATP,2

_lM

proteinkinase inhibitorpeptide,and0.5 mM EGTA for 30 minat25°C.Afterincubation,rMAPKKwascollectedusing glutathione

beads andelectrophoresed on a7%polyacrylamide gel. Thegel was

dried andsubjectedtoautoradiography.a-raf isanaffinity-purified rab-bit polyclonal antibody raised against the COOH-terminal 12 amino acid peptide (CTLTTSPRLPVF) of rat Raf-l, which recognizes c-raf, butnotB-raf and A-raf(25).Anti-MEKKantibodyis alsoan

affinity-purified rabbit polyclonal antibody against the carboxy-terminal 22 amino acids of mouse MEKK and reactswith ratMEKK by

immu-noblottingandimmunoprecipitation (datanotshown).

Assayof MAPKK activity. MAPKK activity was assayed using re-combinant MAPK fused to maltosebinding protein(rMAPK) as

de-scribedpreviously(26).Inbrief,to separateMAPKKfrom endogenous

MAPKs, cell lysates were applied to Q-sepharose column and

flow-through fractionswereimmunoprecipitatedwith25

_j.g/ml

of MAPKK

antibodywhich is anaffinity-purified polyclonal antibody raised against

theNH2-terminal 16aminoacidpeptide(PKKKPTPIQLNPNPEG) of Xenopus MAPKK and termed anMAPKK(26,27).The immunoprecip-itateswereincubated withthe substrate(100

_jsg

rMAPK)in buffer B

for 30min at25TC.rMAPKwascollectedusing amyloseresin and was

electrophoresedon a7% polyacrylamide gel. The gel wasdried and

subjectedtoautoradiography.

PhosphorylationofMAPKs. The extractsof cardiac myocytes were

separated by SDS-PAGE. After electrophoresis,theproteinswere elec-trotransferredontoImmobilon-P membraneusingaMilliBlot-SDS

sys-tem(Nihon MillporeLtd., Tokyo, Japan).Theunoccupied protein bind-ingsites on the membranewereblockedby 3% BSAinbuffercontaining

10 mMTris-HCl, pH7.5, 150 mMNaCl,and 0.1% TritonX-100for 1 h at room temperature. Themembranewasincubatedwith 10

_jig/ml

ofanti-MAPKantibody, aY91 (28), for 10 hat4°C. aY91 is alsoan

affinity-purifiedrabbitpolyclonalantibody raisedagainst the 21 amino

acid sequencededuced fromthe ratMAPK cDNA(28). After washing,

themembranewasincubatedfor 1 h withalkaline phosphatase-coupled secondantibody, anddeveloped usingthe ProtoBlotimmunoblotting

systemaccordingtothemanufacturer's instructions.

Assays ofp9Orsk activity. Theactivity ofp9Onk wasmeasured as

previously described (7, 29).Theimmunoprecipitates from the lysates of cardiacmyocytes with 10jig/mlofanti-p90'kantibody(arsk(m)c) (29)or with controlpreimmune rabbit serum wereincubatedwith50

jsg

ofS6peptide (RRLSSLRA)in25mMTris-HCl,pH 7.4, 10 mM

MgCl2, 1 mMDTT, 40

jiM

ATP, 2

_/Ci

[y-32P]ATP,2

_jiM

protein

kinase inhibitor peptide,and0.5 mM EGTA. After incubation for 10 min at 25°C, 10 1.lofstopping solution containing 0.6%HCl, 1 mM ATP, and 1% BSA was added. After centrifugation, aliquots ofthe supernatants (15 tl) were spotted on 1.5 x 1.5 cm squares ofP81

paper.Thepapers were washedfivetimes foratleast 10mineachin 0.5%phosphoricacid,then in acetone, and dried. The32puptakewas measuredby Cerenkov countingmethod. arsk(m)cis alsoan affinity-purified rabbit polyclonal antibody against the amino acid sequence deducedfromthe cDNA ofmouseS6 kinaseII(29).

Phosphorylation ofRaf-1. Phosphorylation of Raf-1 protein was

examined byWestern blotanalysis. Myocyte lysateswere

electropho-resedona7%polyacrylamide gel,blotted ontoa nitrocellulose sheet,

and incubated with a-raf. Theimmunecomplexes were detectedby the

alkaline phosphatase methodasdescribed before(7).

AssayofRaf-J autokinaseactivity.Cardiacmyocytes werelysedin

buffer A,andRaf-lwasimmunoprecipitated witha-raf. The autophos-phorylation activityofRaf-1wasassayed byincubatingthe

immunopre-cipitates with 10

_ACi

[y-32P]ATP in 25 mMTris-HCl, pH 7.4, 140 mMNaCl,5 mMMnCl2, 10%glycerol, and20

_jM

ATPfor20 min.

Sampleswereseparatedby SDS-PAGE,andactivitywasdetectedby autoradiography.

Kinase assaysof aY91-immunoprecipitates inmyelin basicprotein

(MBP)-containinggels after SDS-PAGE.MAPKactivitywasmeasured using MBP-containing gels as described previously (7). In brief,

MAPKswereimmunoprecipitatedwithaY91 in the presence of 0.15% ofSDS,andwas electrophoresedon an SDS-polyacrylamide gel

con-taining0.5mg/mlMBP. MAPKs weredenaturedby 6 Mguanidine/

HCl and renatured in 50 mM Tris-HCl (pH 8.0) containing 0.04% Tween40 and 5 mM2-mercaptoethanol. PhosphorylationofMBP was

assayedby incubatingthegelwith [y-32P]ATP. Afterincubation,the

gelwas washedextensively, dried, and then subjectedto autoradiog-raphy.

(4)

A A

205 ->

106 >p 60 - ,:

49.5

-B

B

I

'Ma. a b c d a Control 1 2 4 12(nMn)

Stretch

Figure1. MAPKKKactivation by stretchincardiac myocytes.Cardiac myocyteswerestretched by 20% for theindicatedperiods of time, and

the total celllysates were subjectedto theMAPKKKassayasdescribed

inMethods. (A)Arepresentativeautoradiogram showingMAPKKK

activation bystretch incardiac myocytes;(B)thetimecourse of MAPKKKactivation.Relativekinaseactivitywasdetermined by scan-ning eachbandwithadensitometer.Theresultswereindicatedas

mean±SE for five independent experiments.Theactivitywasexpressed relative to thatobtainedinnonstretched myocytes.

the absence orpresence of2 nMstaurosporine.Therelativeamountof

proteinsynthesis was determined by assessingtheincorporation of the radioactivity into a trichloroacetic acid-insoluble fraction. 1 ,uCi/ml

[3H ]phenylalaninewasaddedto theculturemedium 2 hbefore harvest-ing. The cells were rapidly rinsed four times with ice-cold PBS (10

mMsodiumphosphate and 0.85% NaCl, pH 7.4)and incubated for 20 min onice with 1 mlof20% trichloroaceticacid. The totalradioactivity

in each dish wasdetermined by liquid scintillation counting. We

re-peatedtheexperimentfour times induplicate.

Results

Mechanicalstressup-regulates MAPKKK activities. Wehave previously shown that mechanical stressincreases the activity

of MAPKs(7).MAPKsarereportedtobeactivatedbyanother protein kinase MAPKK(8-10), theactivity of which is

regu-latedbyMAPKKKs (14).Both Raf-I and MEKK have been

reportedtohave MAPKKKactivity (13, 14).To examinethe

upstream enzymeofMAPKs,theactivityof MAPKKKs intotal celllysateswasfirstanalyzed by examiningthephosphorylation

of rMAPKK (Fig. 1, A and B). We observed a

significant

increase in theactivity of MAPKKKs fromasearly as 1 min

after stretchcomparedwith the nonstretch control

(control, Fig.

1A, lanea), and maximal induction ofactivity was observed at 2 min after stretch, with a progressive decline in

activity

thereafter. Theslightincrease inactivitycontinueduntil 12 min,

and the levels returnedto nearcontrollevelsby 30 min(data

notshown).

Mechanical stress activates MAPKK. In many cell types,

MAPKKKs are reported to phosphorylate and activate

MAPKK. In cultured cardiac myocytes of neonatal rats, we examined whether stretching of myocytes also activates

MAPKK.Withoutstretch, MAPKKactivitywasbarely detect-able inthecardiocyte lysate (Fig. 2A, lanea). Stretching by

20%for 1-10 min, however, significantly increased MAPKK activity as shown in lane b-e. The activity of MAPKK was

quantitated by densitometric scanning and the timecourse of activation of MAPKK by stretch was shown in Fig. 2 B.

MAPKKactivitywasincreased asearly as 1 min afterstretch, maximallyactivated 5minafterstretch,andgraduallydecreased

8 D C a e Control 1 2 5 10(min)

Stretch

io II J

T-n) Figure2.MAPKKactivationby stretch. MAPKK activity was assayed usingrMAPK asa specific substrate. Toseparate MAPKK from

endoge-nousMAPKs, cell lysates were applied to Q-sepharose column and flow-through fractionswereimmunoprecipitated withanMAPKK anti-body (26). Theimmunoprecipitateswereincubatedwith 100,.g rMAPK inbuffer B for 30 min at25TC.rMAPK was collected using amyloseresin and waselectrophoresedona 7% polyacrylamide gel. (A)Arepresentative autoradiogram of stretch-induced MAPKK

activa-tion incardiac myocytes; (B) the timecourse of MAPKK activation. Each bandofthe autoradiogram was quantified by a densitometer and

theactivity relativeto that in nonstretchedmyocytes wasindicated. The averageof twoindependent experiments was shown.

thereafter. The kinaseactivity returnedtobasal levelsat30min

after stretch(Fig. 2 B).

Mechanical stressphosphorylates and activates MAPKs. aY91 whichrecognizes both 42-kD and 44-kD MAPKs in the

presenceof 0. 15%-SDS (28)wasusedtodetect MAPKs in cell lysates from cultured cardiacmyocytes.Ithas been knownthat

MAPKs are activatedby phosphorylationand that phosphory-lated MAPKs are decreased in the electrophoretic mobility (30). When cardiac myocytes were stretched by 20% for 1

min, anapparentdecrease in electrophoretic mobility of 42-kD

MAPKwas notobserved(Fig. 3, lane b). The shift of the 42-kD band was first observed at 2 min after stretch (lane c). Thebandwasmaximally shiftedat 10 min (lane e). The shift

graduallydecreasedthereafter,and returnedtocontrol levelsat

60 min after stretch (datanot shown). We also measured the MAPKactivityusing MBP-containing gels. MAPK activity was also maximally increased 8 min after stretch and returned to

control levels at 60 min after stretch (reference 7, data not

shown). This observation is consistent with the notion that

phosphorylation of MAPKs reflects their activation (11).

44 kD 42 kD

a b

Control 1

c d e f

2 5 10 30(min)

I Stretch

Figure3. Phosphorylationof MAPKsby stretch. Phosphorylationof MAPKs wasdemonstratedbyashift in theelectrophoretic mobility.

Cellswerestretchedby 20%for the indicatedperiodsoftime,and the celllysatesweresubjectedtoWestern blotanalysiswithananti-MAPKs

(5)

700

-:2 -.

X 600

B.

E

cl)

k2

4oo

-20

3100

0. a

n-0 5 10 20 30

so

-a b C d e f g

Control 0.5 1 2 5 10 30(min)

Stretch

Figure5. Stretch-inducedRaf-l phosphorylation.Cardiac myocytes were stretchedby 20% for the indicatedperiods of time. Total cell lysates wereelectrophoresed on a 7% polyacrylamide gelandtransferred tonitrocellulosefilters.Protein blots wereprobedwitha-raf.Essentially,

T

6me

(min) similarresultswereobtained in three

independent experiments.

Figure4.Stretch-inducedp9Or"kactivation in cardiac myocytes.Cardiac

myocytes werestretched by 20% for the indicated periods of time and

celllysateswere subjectedtoimmunoprecipitationwitharsk(m)C

(closed circles) or control serum (open circles),followedby kinase assayusing S6syntheticpeptideas asubstrate. Theresultswereshown

asmean±SE forfour independent experiments. Asterisksmean

signifi-cant(P<0.05)increase inp9O'r activity compared with0 time control.

Mechanical stress activates

p90sok.

MAPKsphosphorylate

and activate p9o0k, a cytosolickinase that can phosphorylate

the nuclear lamins (18). Our preliminary data showed that stretching of myocytes for 10 min increasesp90okactivity (7).

Inthe present study, weextensively examined the time course of stretch-induced

p9orhk

activation(Fig. 4).Theactivationwas notobservedat2 minafter stretch. The increase in activitywas

first detectable at5min, peaked at 10 min, and sustained higher levels foratleast30 min after stimulation. At the 60-min time point, the activity returned tothe basallevel.

Taken together, both initial and maximal activation of

MAPKKKs, MAPKK, MAPKs, and

p9O'k

was observed in

this orderafter stretchstimulation, suggesting that like growth factor-induced signaling cascade in noncardiac cells, the stretch-inducedproteinkinase cascade spreads in cardiac myo-cytes in the order ofMAPKKKs, MAPKK,MAPKs, andp90ok.

Mechanical stress phosphorylates and activates Raf-J.

Since activation of MAPKKKstriggers the protein kinase

cas-cade of phosphorylation of MAPKs and p90ok through

MAPKK,wefurther characterized MAPKKK activation by

me-chanical stress. Raf-1 has been reported to have a MAPKKK

activity andtobe activated by phosphorylation, which can be detectedby its shift of electrophoretic mobility (13, 15). We therefore examined hyperphosphorylation ofRaf-1 byWestern

blottingusing a-raf. Althoughtherewas nosignificant change in themobility ofRaf-1 incardiac myocytes stretched for 0.5

or 1 min (Fig. 5, lane b and c), Raf-1 molecules in stretched cells for 2-10 min (lane d-f ) migrated more slowly than the Raf-I molecules in the unstimulated cells (lane a), suggesting

thatmechanical stretchhyperphosphorylated Raf-l. The mobil-ity of

Raf-1

returnedtothe control level at 30 min after stretch (lane g).

Since the hyperphosphorylation of Raf-1 could have two

meanings, activation and feedback inhibition of Raf-1 (26), we examined whether mechanicalstress activatesRaf-1. First, we examinedRaf-1 autokinaseactivity. Raf-1was immunoprecipi-tated with a-raf from the lysates of stretched myocytes and incubated with[y-32P]ATPinvitro. Stretching for 1 to 2min

increased Raf-1 autokinase activity as shown in Fig. 6 A. Autok-inase activity of Raf-1 peaked at 1 min and returned to the control level 12 min after stretch(Fig. 6,Aand B). Secondly, weexamined Raf-1 activityusingrMAPKKasasubstrate. The immunoprecipitates with a-raf were incubated withrMAPKK

and [y-32P]ATP, and the incorporationof 32p into rMAPKK

wasexamined. A rapid increase inMAPKKKactivity was ob-served in the a-rafimmunoprecipitates from stretched cellsfor

2 min (Fig. 7, lane b), indicating that mechanical stretch in-creased the MAPKKK activity of Raf-l.

Stretch-inducedactivationofRaf-Joccurs via PKC-depen-dent and independent pathways. We have previously reported that mechanicalstressinduces MAPK activation via both PKC-dependent and inPKC-dependent pathways (7). It has been reported in other cells that there are alsotwopathways for Raf-1 activa-tion, PKC-dependent and independent pathways (15). We

therefore examined whether Raf-1 activation by stretching of cardiac myocytes is associated with the activation of PKC. When myocytes were pretreated with 10-7M 12-O-tetradeca-noylphorbol-13-acetate (TPA) for 24 h, TPA itself failed to activate Raf-1 activity, suggesting that phorbol ester-sensitive PKC wascompletely depleted (data not shown). In these pre-treatedcardiocytes, activation of Raf-1 by stretch was decreased by - 50% as compared with the kinase activity in stretched

myocytes without pretreatment (Fig. 7, lane b and c). Since the stretch-induced MAPK activation was also partially

sup-a ,.

G.

4~

2

a b C

Control 1 2(min) lI

Stretch O i is 1_"""is

Figure6.Raf-1 autokinase activation by stretch. (A) Cardiac myocytes werestretched by 20% for the indicated periods of time. The

immuno-precipitateswith a-rafwereincubated with 10

_ICi

[y-32P]ATPfor 20 minat

250C,

electrophoresedon a7%polyacrylamide gel, and thegel

wassubjectedtoautoradiography. B, the intensity of76-kDband

corre-spondingtoRaf-1 wasmeasuredby densitometricscanningof the

(6)

A

205 - o

Co

XFigure

7.Raf-l

activa-tion by stretch in cardiac myocytes. Cardiac myo-cytes werestretched for 2 minand the immuno-precipitates with a-raf

from the celllysates were incubated with rMAPKK and[y-32P]ATPunder the condition as de-scribed in Methods. Puri-fied rMAPKK was

sub-jectedtoSDS-PAGEand thegelwasexposed to x-ray filmfor24h. PKC

a b C wasdown-regulatedby

pretreating cultured introl Stretch TPA cardiocytes with TPA

+ (10-7M)for24 h Stretch (lane c).

a b

Control AngII

B

49.5

32.5 -->

_ l 4~~~-. 44kD

42kD

C d

CV-1 1974 CV-11974

+

Ang 11

pressed bythedown-regulationof PKC (7), theremaybetwo

pathways, PKC-dependent and independent pathways, in

stretch-induced signalingcascade ofproteinkinasesand thetwo pathways mayconverge atthe level of Raf-1.

Stretch-induced activation ofRaf-1and MAPKs ispartially

dependent on externalAng II. Recently, it has been reported

thatAngIIis secreted from cardiac myocytesbystretch (31),

and our preliminary data suggest that the increase in MAPK

activityandc-fosgeneexpressionbystretcharedependenton

Ang 11(32). We therefore examined the role of Ang II in

the mechanical stress-inducedsignalingcascade. When cardiac

myocyteswere incubated with IO-7M Ang II for 2 min,

sig-nificant activation ofRaf-I wasobserved(Fig. 8, Ang II).This

activationwascompletely suppressed bypretreatment with 1o-6

M Ang II type 1 receptor-specific antagonist, CV-11974

(CV-11974 + Ang II), suggesting that Raf-1 is activated by AngII through type 1 receptor. Next, after pretreatment with

CV-1 1974 (10-6 M) for 30 min, cardiac myocytes were

stretched by 20% for 2 min. Stretch-induced Raf-1 activation

was partially inhibited by pretreatment with CV-1 1974

10 Figure 8. Partial

depen-dence of stretch-induced

8 Raf-1 activationonAng

_ T II.Cardiac myocytes werestimulatedby Ang

i LII (l0-M)for 2minor

X 4 * stretched for 2min

with--outorwith the

pretreat-2 ment

by Ang

II-type I

receptor antagonist,

CV-C- 11974 (10-7M)for 30

AMR CVS11974OethCV.11W4 CV.11374 min.After

immunopre-ANI _SoVth _{cipitated by a-raf,}_Raf-l

wasincubated with rMAPKKand[y-32P]ATP. Purified rMAPKKwassubjectedto SDS-PAGE and thegelwassubjectedtoautoradiography. Relative kinase

activitywasdeterminedby scanningeach band withadensitometer. The resultswereindicatedasmean±SE for fourindependent

experi-ments.Theactivitywasexpressedrelativetothat obtained innontreated

myocytes. Asterisksmean asignificant (P< 0.05)increase in MAPK

activity comparedwith control.

a b c d

ControlStretch CV-11974 CV-11974

Stretch

C

a

'

a

_"S

2

0o

+ +

Ang 11 Strtch

Figure9. Partialdependenceof stretch-induced MAPK activationon

AngH. Cardiacmyocyteswerestimulatedby AngII (10-' M)for 8

min (A)orstretched by 20% for 8 min (B) with (lane c)orwithout (lane b)the pretreatmentbyCV-1 1974(10-6M)for 30min.MAPKs wereimmunoprecipitated withaY91 andelectrophoresedon a

SDS-polyacrylamide gel containingMBP.After denaturation and

renatu-ration, the gelwasincubated with [y- 32P] ATP. Thegel waswashed, dried,andsubjectedtoautoradiography. Lanea,nonstretch;lane d

nonstretch withCV-11974.(C)The intensities of 42-kD and 44-kD bandsweremeasuredbydensitometricscanningof theautoradiogram. Values represent the mean±SE from fourindependent experiments.The

intensityof 44-kD MAPK of nonstretched myocyteswasdesignated

as 1.0.

(CV-11974+Stretch), indicatingthatapartof Raf-1 activation

by mechanical stretch isdependentonAngII.

The involvement ofAngII instretch-induced MAPK activa-tionwasalso examinedbykinaseassaysusingMBP-containing

gels (32). Incubation with

i0-'

M Ang H for 8 min evoked

maximal activationof both 42-kDand 44-kD MAPKs incardiac

106 -*

80 ->

(7)

A~~ ~~~~~4 Id E44

42kD {4

4-42k0~~~~

a b C

CoL drWurp AngI 0

+ COM suo ww A 11

MgIA I

Figure 10. Dependence of Ang11-inducedMAPKactivationonPKC. (A)Cardiac myocytes werestimulatedby 10-6 M Ang H for8min in the presence(lane b)orabsence (lane c) ofpretreatmentwith2 nM

staurosporinefor30min. Theimmunoprecipitates withananti-MAPKs antibody, aY91, weresubjectedtoMAPK assays inMBP-containing gelsasdescribedin Methods.(B)Theintensitiesof42-kDand 44-kD bands weremeasuredby densitometricscanning oftheautoradiogram.

Values represent the mean valuesfromthreeindependent experiments.

Theintensityof 44-kDMAPKof nonstretchedmyocyteswasdesignated

as 1.0.

myocytes, and activation was completely blocked by 10-6 M

CV-1 1974 (Fig. 9,A and C, AngII vs CV-11974 + Ang H).

This result suggests that like Raf-Iactivation,MAPKsare acti-vated by AngIIthrough type1receptor. Stretch-induced MAPK activationwassuppressedby only 50-70%bythe pretreatment with CV-1 1974 (10-6M) for 30 min(Fig. 9 Blaneb and c; Fig. 9 C Stretch vs CV-11974 + Stretch), which isconsistent with ourprevious results (32). Addition of CV-1 1974 didnot

activateMAPKactivity byitself(Fig.9 A, lane d;andFig. 9 B,laned).Pretreatmentwith AngIItype2-receptorantagonist,

PD123319, hadnoeffectonstretch-inducedactivation of

Raf-1 and MAPKs (data not shown). These results suggest that thereare twotransductionpathwaysfor mechanical stress, Ang

11-dependent (throughtype 1 receptor) andindependent

path-ways.

Ang II-induced MAPK activation is dependent on PKC. Wehavepreviouslyshown thataPKCinhibitor,2 nM

stauro-sporine, which completely abolished the effect of 10-6MTPA,

partially suppressedthe activationofMAPKsinducedbystretch (7). Taken together, these results suggest that both PKC and external Ang 11areinvolved in part in stretch-induced activation of thesignaling pathwayfrom Raf-ItoMAPKs.Thenwe

exam-ined the relationship between PKC and Ang 11. We analyzed theeffect of PKCon MAPKactivationbyAng 11usingaPKC

inhibitor, staurosporine. As shown inFig. 10,A and B, Ang 11

(10-6

M) increased the activity of both 42-kD and 44-kD

MAPKs, whichwasalmostcompletelyblockedbypretreatment withstaurosporine(2nMfor 30 min). Addition of TPA(

_10-6

M) didnotactivate MAPKactivity onthis condition. Similar resultswereobtainedusing PKC-down-regulatedcardiac myo-cytes (datanot shown). Taken together, these results suggest thatwhilethere are twopathways, PKC-dependent and inde-pendent, for stretch-induced protein kinase cascade of phos-phorylation, Ang II-induced signaling pathways aretriggered mainly by the activation of PKC.

Stretchcauses MEKKactivation. Recently,MEKK aswell asRaf-1 has beenreportedto activateMAPKK (13, 14).We

examined whether stretchcanalsoinduceMEKKactivation in

cardiacmyocytes.Although the nonstretched myocytes showed

106 >-80

-*

Figure 11.MEKK

acti-vation bystretch in car-diac myocytes. Carcar-diac myocytes were stretched by 20% for 2min, and

MEKKwas

immunopre-cipitated from celllysates

with aspecific anti-MEKKantibody.The

immunoprecipitateswere

incubated withaspecific

substrate(rMAPKK) and[Y_32P]ATP. rMAPKKwascollected using glutathionebeads

andelectrophoresedon a

7%polyacrylamide gel.

a b The_{exposed to X-ray}gelwasdried and_film_for

Control

Stretch

10 days.

almost no activity of MEKK (Fig. 11, lane a), stretch for 2 min dramatically increased the activity of MEKK (lane b),

suggesting that not only

Raf-1

but also MEKK is activated during mechanical stretch-inducedhypertrophy.

Stretch inducesanincrease inphenylalanine incorporation

intomyocytes viabothPKC-dependentandindependent

path-ways. We have previously shown that stretching of myocytes increasesphenylalanine incorporation into cells,suggestingthat mechanicalstressdirectly induces cardiac cellular hypertrophy (5, 6). To examine whether PKC is involved in stretch-induced cardiachypertrophy,wemeasured the relativeamountofprotein

synthesis after stretch in the presence or absence of

stauro-sporine. As shown in Fig. 12, stretch by 20% for 24 h stimulated

anamino acidincorporationby - 1.5-fold. This increase was

only partially suppressed by pretreatment with 2 nM

stauro-sporine. These data suggest that the cellularhypertrophyis also induced bymechanical stress both through PKC-dependent and independent pathways.

Discussion

In the present study, we investigated the signaling pathways which are evokedby mechanical stressin neonatalratcardiac myocytes.Stretching of cardiac myocytes sequentially activated

C0 Figure12.Partial

depen-dence ofstretch-evoked

8* phenylalanine

incorpo-* rationonPKC. After pre-treatment with

stauro-S

010 -

sporine

(2

nM)for30

min,cardiac myocytes

werestimulatedby

.. _ _ _ stretch for 24 h and

[3H]phenylalanine (1

= o- Z _ _ _ sCi/ml)was added 2 h

Control Stretch Staurosporine before harvest. The total

+ _{radioactivity}_of incorpo-Stretch rated [3H] phenylalanine intoproteinswasdeterminedbyliquid scintillation counting. The

rela-tivephenylalanine uptakewas shownasmean±SE from four

(8)

protein kinase cascade of phosphorylation.Both initial and max-imal activation occurred in the order ofMAPKKK, MAPKK, MAPKs, and pgork. It has been reported thatMAPKs translo-cateinto the nucleus upon activation and activate transcription factors which contain potential MAPK phosphorylation sites (16). We have previously shown that mechanical loading in-duces expression of immediate early response genes such as c-mycand c-fos genes as an early event (3, 5, 33). Therefore, once an extracellular signal, mechanical stress, is received by cardiac myocytes, the signal may spread in cardiac myo-cytes by protein kinase cascade in the order of MAPKKKs -+ MAPKK-+MAPKs +

p9Ork

and_{may evoke the nuclear}_events

(gene expression). In addition, we have previously shown that c-fos was abundantly expressed in

_cardiac,

myocyte-rich frac-tion rather than in nonmuscle-rich fracfrac-tion (5) and that stretch-ing of cardiac myocytes for 24 h increased total RNA content by45%, while RNA content in the nonmuscle-rich fraction was not increased by stretch (6). These results suggest that the stimulation of gene expression and hypertrophic responses by stretch mainly occurs in cardiac myocytes although Ang II

whose secretion is induced by stretch from cardiac myocytes mayact onnonmuscle cardiocytes byaparacrine mechanism.

MAPKKKs, in general, have been implicated as transduc-tionsignalingmolecules in thecytoplasmupstream ofMAPKs

and downstream of other molecules suchasRas, PKC, and Src (14).It has been reported that both Raf-1 andMEKKhave a MAPKKKactivity andcanphosphorylateand activate MAPKK (13, 14). In other words, signal transduction pathways from Raf-1 andMEKKconvergeat MAPKKresulting in the activa-tion of MAPKs. There have been many reports in yeast and other mammalian cells suggesting thatRaf-l may regulate the

MAPKnetworkprimarilyin responsetoreceptors thatare

asso-ciated withtyrosinekinaseactivity and thatMEKKmight

medi-ate primary signals originatingfrom receptors that activate G

proteinsand PKC(34, 35). Toclarifythe upstreamsignals,we

examined which protein kinase, Raf-1 or MEKK,is activated in stretched myocytes. Wedemonstrated for the first time that

stretching of myocytes activates both Raf-1 and MEKK in a

similar timecourse.Themaximum activation of both Raf-1and MEKKwasobservedasearlyas 2min(Figs.7and 11 ). These results mayimply thatstretchingof cardiac myocytes activates multiplesignaltransduction pathwayssuchastyrosine kinases,

G proteins, and PKC, which is consistent with the

suggesting

that p21ras exists

upstream ofRaf-I in the signal transduction pathway. Recently, p21 ' _has_{been reported to be activated in cardiac myocytes by}

mechanical stress (36). Taken together, the PKC-independent activation of Raf-1 by cardiocyte stretch may be induced through p21ras activation.

Although there is no direct evidence proving that the MAPK cascade is involved in the stretch-induced hypertrophic re-sponses, many studies in noncardiac cells as well as cardiac cells suggest that MAPK cascade play a critical role in growth responses. Thorburn et al. (40) showed that phenylephrine in-duced the gene expression of atrial natriuretic factor in the cardiocytes and this induction depends on the activation of MAPKs. MAPK has been reported to phosphorylate many pro-teins including c-myc, c-jun and p62TCF

(c-fos)

which are known to be activated during cardiac hypertrophy (2, 16). Ras which is one of upstream signaling molecules for MAPK cas-cade has been shown to be necessary and sufficient for hypertro-phic response (41). These observations raise the possibility that the MAPK cascade may be important in the hypertrophic response of the heart. Development of specific MAPK inhibitors ordominant negative MAPK mutants would elucidate this pos-sibility.

Recent evidence has suggested that Ang II plays an im-portant role in cardiac hypertrophy (20-22). Ang II shows direct stimulatory effects on increased protein synthesis in cul-tured cardiac myocytes (42, 43). AngII infusion at subpressor doses induces left ventricular hypertrophy in adult rats (44), and cardiac hypertrophy of spontaneous hypertensive rats is prevented by low doses of an AngIItype 1 receptor antagonist (32). Moreover, hypertrophic events such as specific gene ex-pression and activation of PKC and MAPKs induced by me-chanical loading resemble those induced by Ang II (5-7). It has been reported that mechanical stretch induces secretion of Ang II from cardiocytes (31). In the present study, stretch-induced activation of Raf-I and MAPKs was in part inhibited by pretreatment with a type 1 AngII receptor antagonist. Since the amount of the antagonist used in the present study was sufficient to suppress the maximal effect of Ang II, other factors different from Ang II may be involved in Raf-l and MAPK activation.

Mechanical stress induced

Raf-1

activation, which was in part suppressed by the down-regulation of PKC (Fig. 7). In the previous paper, we have shown that activation of MAPKs by stretch was also only partially suppressed by a PKC inhibitor, staurosporine (7). On the contrary, Ang II-evoked activation of

Raf-1

and MAPK, and the increase in c-fos mRNA expres-sion and protein synthesis were completely blocked by PKC

down-regulation or staurosporine (Fig. 10,A and B and refer-ence 43). Although the signals evoked by mechanical stress mimicthose by AngII,thereare some differences in the signal transduction pathways between mechanical stress and AngII.

In short, mechanical stress activates MAPK cascade both via PKC-dependent and independent pathways,whileMAPK acti-vation by Ang II ismostly dependent on a PKC pathway.

The protein kinase cascade activated by mechanical stress

is composed of PKC, Raf-l, MEKK, MAPKK, MAPKs, and

p9Orsk(Fig. 13). Stretch stimulus induces the activation of Raf-1 and MEKK in cardiac myocytes. A part of

Raf-1

activation

(9)

Figure13. A modelfor the signaltransduction

cascadeactivated by

me-chanical stressincardiac

myocytes.Mechanical stretchinducessecretion of AngHfromcardiac

myocytes.AngII

acti-vatesPKC andthen in-ducesthesequential

acti-vationofproteinkinases

(Raf-l MAPKK -+ MAPK

p90"k).

There

mightbeotherunknown pathwaysforactivating

Raf-l.MEKKisalso

ac-tivatedbystretchin

car-diacmyocytes.

withexternal AngIIorPKC.Ourfindings highlight thecomplex

andspecific signaling pathwaysofstretch-inducedcardiac myo-cyte hypertrophy. These stretch-induced signal transduction pathways incardiacmyocyteshave similarities insomeaspects

to those of cells which are exposed to hyperosmotic shock.

Hyperosmotic shock causes activation of Pbs2 (a member of

MAPKK), Hogl (45), p38 (46) and Jnk (47) (members of the MAPKfamily) in theyeasts (Pbs2, Hogi ) and in

mamma-liancells (p38, Jnk). Theseobservationssuggestthat the

signal-ing pathways evoked by mechanical stress, if hyperosmotic shock wereakind of mechanical stress, might be highly

con-served in evolution and that the transduction ofexternal

me-chanical stress into the nucleus is quite important for living

cells. Further studiesare necessarytoelucidatethemechanisms

bywhich the mechanical stimulus isconverted into these

bio-chemical events andto determine upstream molecules

includ-ing "mechanoreceptors," which activate PKC, Raf-1, and

MEKK (48).

Acknowledgments

We thank Toru Suzuki for critical reading, andFumiko Harima and

Megumi Fukuda for the excellenttechnicalassistance. Weacknowledge

Takeda Chemical Industries, Ltd. (Osaka, Japan) forproviding

CV-11974.

This work wassupported by agrant-in-aid forscientific research

and developmentalscientific researchfrom theMinistry of Education,

Science and Culture,agrantfrom the JapanCardiovascularFoundation,

theSankyo Life Science and the StudyGroupofMolecularCardiology

(toI. Komuro),Japan,and agrantfrom Japan Heart Foundation

Re-searchGrantFor 1994(toT.Yamazaki),Japan.

References

1.Levy,D., R. J.Garrison,D.D.Savage, W. B. Kannel, andW. P.Castelli. 1990.Prognostic implications ofechocardiographicallydeterminedleft ventricular

massin theFraminghamheartstudy. N.EngL. J.Med.322:1561-1566.

2. Komuro, I.,and Y.Yazaki. 1993. Control of cardiacgeneexpression by mechanicalstress.Annu.Rev.Physiol. 55:55-75.

3.Komuro,I., M.Kurabayashi,F.Takaku, and Y.Yazaki.1988. Expression ofcellularoncogenes in the myocardium during thedevelopmental stageand

pressureoverloadhypertrophy of theratheart. Circ.Res. 62:1075-1079. 4. Mulvagh, S. L., L.H.Michael, M. B. Perryman,R.Roberts,and M. D.

Schneider. 1987. Ahemodynamic load invivo induces cardiacexpressionof the

cellularoncogene,c-myc.Biochem. Biophys. Res. Commun. 147:627-36.

5.Komuro,I., T.Kaida, Y. Shibazaki,M.Kurabayashi,F.Takaku,and Y.

Yazaki.1990.Stretchingcardiacmyocytesstimulatesproto-oncogeneexpression.

J.Biol.Chem.265:3595-3598.

6. Komuro,I., Y.Katoh, T.Kaida, Y. Shibazaki,M.Kurabayashi,F.Takaku, and Y.Yazaki.1991. Mechanical loadingstimulates cellhypertrophyandspecific

geneexpression in culturedratcardiac myocytes.J. Biol. Chem.266:1265-1268.

7.Yamazaki, T., K. Tobe, E. Hoh, K. Maemura,T.Kaida, I.Komuro, H.

Tamemoto,T.Kadowaki, R. Nagai, andY. Yazaki. 1993. Mechanical loading activatesmitogen-activated protein kinaseandS6 peptidekinasein cultured rat

cardiacmyocytes.J.Biol. Chem. 268:12069-12076.

8.Gomez, N.,and P. Cohen. 1991.Dissection of theproteinkinase cascade

bywhichnervegrowth factor activates MAP kinases.Nature(Lond.).

353:170-173.

9.Ahn, N.G., R. Seger, R. L. Bratlien,C. D.Diltz,N. K.Tonks,and E.G. Krebs.1991.MultiplecomponentsinanEGF-stimulated proteinkinase cascade.

In vitroactivation ofaMBP/MAP2 kinase. J. Biol. Chem. 266:4220-4227.

10.Matsuda, S., H. Kosako, K. Takenaka,K.Moriyama,H.Sakai,T.

Aki-yama,Y.Gotoh, and E. Nishida. 1992. XenopusMAPkinase activator:

identifica-tionandfunction as akey intermediatein thephosphorylationcascade. EMBO (Eur.Mol.Biol. Org.) J. 11:973-982.

11. Nishida, E.,and Y. Gotoh. 1993.The MAP kinase cascadeis essential

fordiversesignal transduction pathways. TrendsBiochem. Sci. 18:128-131.

12. Zheng, C.-F., and K.-L. Guan. 1994. Activationof MEKfamilykinase

requiresphosphorylation oftwoconserved Ser/Thr residues. EMBO(Eur.Mol.

Biol. Org.) J. 13:1123-1131.

13. Kyriakis, J.M., H. App, X. Zhang,P.Banerjee,D. L.Brautigan, U. R. Rapp,and-J.Avruch. 1992. Raf-l activatesMAP kinase-kinase. Nature(Lond.). 358:417-421.

14. Lange-Carter, C. A., A. M. Pleiman, K. J. Blumer,and G. L. Johnson.

1993. A divergence in the MAP kinase regulatory network defined by MEK

kinaseand Raf. Science (Wash. DC). 260:315-319.

15. Wood, K. W., C. Sarnecki, T. M. Roberts, and J. Blenis. 1992. Ras

mediatesnerve growth factor receptor modulation of three signal-transducing proteinkinases: MAPkinase, Raf-l and RSK. Cell. 68:1041-1050.

16.Davis, R. J. 1993. The mitogen-activated proteinkinasesignaltransduction

pathway. J.Biol. Chem. 268:14553-14556.

17. Sturgill, T.W., L. B. Ray, E. Erickson,and J. L. Maller. 1988.

Insulin-stimulatedMAP-2 kinasephosphorylatesand activates ribosomal proteinS6

ki-naseII. Nature(Lond.). 334:715-718.

18. Ward, G. E., and M. W. Kirschner. 1990. Identification of cell cycle-regulatedphosphorylation sitesonnuclearlamin C. Cell. 61:561-577.

19. Baker, K. M., G. W. Booz, and D. E. Dostal. 1992. Cardiac actionsof

angiotensinH:role ofanintracardiac renin-angiotensinsystem. Annu. Rev. Phys-iol. 54:227-241.

20. Linz, W., B. A. Schoelkens, and D. Ganten. 1989. Convertingenzyme

inhibitorspecificallypreventsdevelopment and inducestheregressionofcardiac

hypertrophy inrats.Clin. Exp. Hypertens. 11:1325-1350.

21. Baker,K.M., M.I.Cherin, S. K. Wixon,andJ. F. Aceto. 1990.

Renin-angiotensinsysteminvolvement in pressure-overloadcardiachypertrophyin rats.

Am. J.Physiol. 259:H324-H332.

22.Schunkert, H., V. J. Dzau, S. S. Tang, A.T. Hirsch,C. S.Apstein, and

B.H.Lorell. 1990. Increasedratcardiacangiotensinconvertingenzymeactivity andmRNAexpression inpressure overload leftventricularhypertrophy: effect

on coronary resistance, contractility, and relaxation. J. Clin. Invest.

86:1913-1920.

23. Simpson, P., and S. Savion. 1982. Differentiation ofrat myocytes in

singlecell cultures with and without proliferating nonmyocardialcells. Circ. Res.

50:101-116.

24. Crews, C. M., A. Alessandrini, and R. L. Erikson. 1992. The primary

structureof MEK, aproteinkinase thatphosphorylates the ERK geneproduct. Science(Wash. DC). 258:478-480.

25.Izumi,T., H. Tamemoto,M. Nagao, T. Kadowaki, F.Takaku, and M.

Kasuga.1991.Insulin and platelet derived growthfactor stimulatephosphorylation of thec-raf productatserine andthreonineresidues in intact cells. J.Biol.Chem.

266:7933-7939.

26. Ueki, K., S. Matsuda, K. Tobe,Y. Gotoh,H.Tamemoto,M. Yachi, Y.

Akanuma, Y. Yazaki, E.Nishida, and T. Kadowaki. 1994. Feedbackregulation ofmitogen-activated proteinkinase kinase kinaseactivityof c-Raf-l insulinand

phorbolesterstimulation. J. Biol. Chem. 269:15756-15761.

27. Kosako, H., E. Nishida, and Y. Gotoh. 1993. cDNA cloning of MAP kinasekinaserevealskinase cascade pathways inyeasts tovertebrates. EMBO (Eur.Mol. BioLOrg.) J. 12:787-794.

Stretch

Ang

11 PC

,

_/"O~

MEKK

_Raf-Il

MAPKK

MAPK

(10)

28. Tobe, K., T. Kadowaki, H. Tamemoto, K. Ueki, K. Hara, 0. Koshio, K. Momomura, Y. Gotoh, E. Nishida, Y. Akanuma, et al. 1991. Insulin and 12-0-tetradecanoylphorbol-13-acetateactivationoftwoimmunologically distinct my-elin basicprotein/microtubule-associatedprotein 2(MBP/MAP2)kinasesvia de novo phosphorylation of threonine and tyrosine residues. J. Biol. Chem.

266:24793-24803.

29. Tobe, K., T. Kadowaki, K. Hara, Y. Gotoh, H. Kosako, S. Matsuda, H. Tamemoto, K. Ueki, Y. Akanuma, E. Nishida, and Y. Yazaki. 1992.Sequential activationof MAP kinaseactivator,MAPkinases,and S6peptidekinase in intact ratliverfollowinginsulininjection.J. Biol. Chem.267:21089-21097.

30. Boulton, T. G., S. H. Nye, D. J. Robbins, N. Y. Ip, E. Radziejewska, S. D.Morgenbesser, R. A. DePinho, N. Panayotatos, M. H. Cobb, and G. D.

Yanco-poulos.1991. ERKs: afamilyofprotein-serine/threoninekinases that areactivated andtyrosine phosphorylatedin response to insulin and NGF. Cell. 65:663-675. 31. Sadoshima, J., Y. Xu, H. S.Slayter,and S. Izumo. 1993.Autocrinerelease ofangiotensin II mediates stretch-induced hypertrophyof cardiac myocytes in vitro. Cell. 75:977-984.

32.Kojima,M.,I.Shiojima,T.Yamazaki,I. Komuro, Y.Zou,Y. Wang, T.

Mizuno,K. Ueki, K.Tobe,T.Kadowaki, etal. 1994.AngiotensinII receptor

antagonistTCV-1 16 inducesregressionofhypertensiveleft ventricular

hypertro-phy in vivo and inhibits intracellularsignaling pathwayofstretch-mediated cardio-myocytehypertrophyin vitro.Circulation. 89:2204-2211.

33.Komuro, I.,Y.Shibazaki, M.Kurabayashi, F.Takaku, and Y. Yazaki.

1990.Molecularcloning ofgene sequences from rat heartrapidly responsiveto pressure overload. Circ. Res. 66:979-985.

34.Pelech,S. L., and J. S.Sanghera.1992. MAP kinases:chartingthe

regula-torypathways.Science(Wash.DC).257:1355-1356.

35.Leberer, E.,D.Dignard, D.Harcus, D. Y.Thomas,and M. Whiteway. 1992.Theproteinkinasehomologue Ste2Opisrequiredto link theyeast

phero-moneresponseG-proteinbeta gamma subunitstodownstreamsignalling compo-nents.EMBO(Eur. Mol.Biol. Org.)J. 11:4815-4824.

36.Sadoshima, J.,andS.Izumo. 1993. Mechanical stretchrapidlyactivates multiplesignaltransductionpathwaysin cardiac myocytes:potentialinvolvement of anautocrine/paracrinemechanism. EMBO(Eur. Mol.Biol.Org.)J.

12:1681-1692.

37. Kolch, W., G. Heidecker, G. Kochs, R. Hummel, H. Vahidi, H. Mischak, G. Finkenzeller, D. Marme, and U. R. Rapp. 1993. Protein kinase Ca activates RAF-1 by direct phosphorylation. Nature (Lond.). 364:249-252.

38. Lissoos, T. W., D. W. A. Beno, and B. H. Davis. 1993. 1, 25-dihydroxyvi-tamin D3 activates Raf kinase and Raf perinuclear translocation via a protein kinaseC-dependent pathway. J. Biol. Chem. 268:25132-25138.

39. Zhang, X., J. Settleman, J. M. Kyriakis, E. Takeuchi-Suzuki, S. J. Elledge, M. S. Marshall, J. T. Bruder, U. R. Rapp, and J. Avruch. 1993. Normal and oncogenicp21 proteinsbind to the amino-terminal regulatory domain of c-Raf-1. Nature(Lond.). 364:308-313.

40.Thorburn,J., J. A.Frost,and A. Thorburn.1994.Mitogen-activatedprotein kinases mediate changes in gene expression, but not cytoskeletal organization associated with cardiac muscle cellhypertrophy. J. Cell Biol. 126:1565-1572.

41. Thorburn, A., J. Thorburn, S.-Y. Chen, S. Powers, H. E. Shubeita, J. R. Feramisco, and K. R. Chien. 1993. HRas-dependent pathways can activate morphological andgeneticmarkers of cardiac muscle cell hypertrophy. J. Biol. Chem. 268:2244-2249.

42. Baker, K. M., and J. F. Aceto. 1987. Angiotensin II stimulation of protein

synthesisand cellgrowthin chick heart cells. Am. J. Physiol. 259:H610-618. 43. Katoh, Y.,I.Komuro, Y.Shibazaki,H.Yamaguchi, and Y. Yazaki. 1989. Angiotensin II induces hypertrophy and oncogene expression in cultured cardiac myocytes.Circulation. 80(Suppl. II) :II-450.

44.Dostal, D. E., and K. M. Baker. 1992. Angiotensin II stimulation of left ventricularhypertrophy in adult rat heart. Am. J. Hypertens. 5:276-280.

45. Brewster, J. L., T. de Valoir, N. D. Dwyer, E. Winter, and M. C. Gustin, 1993. An osmosensing signal transduction pathway in yeast. Science (Wash. DC).

259:1760-1763.

46. Han, J., J.-D. Lee, L.Ribbs, and R. J. Ulevitch. 1994. A MAP kinase

targeted byendotoxin andhyperosmolarityin mammalian cells. Science (Wash. DC). 265:808-811.

47.Galcheva-Gargova, Z., B.Derijard,I.-H. Wu, and R. J. Davis. 1994. An

osmosensing signaltransduction pathway in mammalian cells. Science (Wash. DC). 265:806-808.