Cell. Signal. Vol. 10, No. 10, pp. 693–698, 1998 ISSN 0898-6568/98 $19.00Copyright 1998 Elsevier Science Inc. PII S0898-6568(98)00036-9

TOPICAL REVIEW

Signalling Pathways for Cardiac HypertrophyTsutomu Yamazaki,†‡ Issei Komuro†* and Yoshio Yazaki†

†Third Department of Internal Medicine, Faculty of Medicine, ‡HealthService Center, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan

ABSTRACT. Mechanical stretch is an initial factor for cardiac hypertrophy in response to haemodynamic over-load (high blood pressure). Stretch of cardiomyocytes activates second messengers such as phosphatidylinositol,protein kinase C, Raf-1 kinase and extracellular signal-regulated protein kinases (ERKs), which are involved inincreased protein synthesis. The cardiac renin–angiotensin system is linked to the formation of pressure-overloadhypertrophy. Angiotensin II increases the growth of cardiomyocytes by an autocrine mechanism. AngiotensinII-evoked signal transduction pathways differ among cell types. In cardiac fibroblasts, angiotensin II activatesERKs through a pathway including the Gbg subunit of Gi protein, Src family tyrosine kinases, Shc, Grb2 andRas, whereas Gq and protein kinase C are important in cardiac myocytes. In addition, mechanical stretch en-hances the endothelin-1 release from the cardiomyocytes. Further, the Na1–H1 exchanger mediates mechanicalstretch-induced Raf-1 kinase and ERK activation followed by increased protein synthesis in cardiomyocytes. Notonly mechanical stress, but also neurohumoral factors induce cardiac hypertrophy. The activation of protein ki-nase cascades by norepinephrine is induced by protein kinase A through b-adrenoceptors as well as by proteinkinase C through a-adrenoceptors. cell signal 10;10:693–698, 1998. 1998 Elsevier Science Inc.

KEY WORDS. Cardiac myocyte, Cardiac fibroblast, Renin–angiotensin system, Mechanical stretch, Proteinkinase, Norepinephrine

INTRODUCTION cal stretch itself is an initial factor for cardiac hypertrophyin response to haemodynamic overload. It was first reportedCardiac hypertrophy is not only a functionally useful adap-that stretch of quiescent papillary muscle accelerates pro-tation to an enhanced workload [1], but also one of the mosttein synthesis [3]. It was successively shown that elevationinfluential clinical complications of cardiovascular disor-of aortic pressure in beating denervated hearts increases pro-ders [2]. It is well known that a percentage of patients withtein synthesis. In addition, Cooper et al. [4] reported that, incardiac hypertrophy develop heart diseases includingaorta-constricted cats, papillary muscle whose tendon has beenischaemic heart disease, heart failure and sudden cardiaccut to release the tension does not show hypertrophy, whereasdeath and that the hypertrophy is associated with an in-neighbouring uncut papillary muscle shows marked hyper-creased mortality rate [2]. Elucidation of the initial factorstrophy, and that cardiac hypertrophy is induced by pressurethat cause cardiac hypertrophy is, therefore, extremely im-overload even under the denervation of ventricular adreno-portant. In this review, we focus attention on signallingcreceptors. Furthermore, stretching cultured cardiomyocytespathways for cardiac hypertrophy.stimulates protein synthesis and specific gene expression with-out participation of neural or humoral factors [5, 6].

HAEMODYNAMICSTRESS DIRECTLY INDUCESCARDIOMYOCYTE HYPERTROPHY PROTEIN PHOSPHORYLATION CASCADES

INDUCED BY MECHANICAL STRETCHOpinions have differed about whether a primary stimulusfor hypertrophy is a mechanical stress itself or a concurring In regard to the molecular mechanisms by which externalincrease in neural or humoral factors. Some studies have load evokes cardiac hypertrophy, protein kinase cascadesshown that the activation of adrenoreceptors is an interme- are the focus of much attention. Some groups, includingdiate of haemodynamic overload and cardiac hypertrophy. ours, have shown that mechanical stretch of neonatal ratHowever, evidence presented herein suggests that mechani- cardiomyocytes activates second messengers such as phos-

phatidylinositol, protein kinase C, Raf-1 kinase and extra-*Author to whom all correspondence should be addressed. E-mail: komuro- cellular signal-regulated protein kinases (ERKs), which aretky@umin.ac.jp

Received 5 March 1998; and accepted 23 April 1998. involved in re-expression of a number of genes including

694 T. Yamazaki et al.

those that encode atrial natriuretic peptide, skeletal a actin pertensive cardiac hypertrophy of animals and humans [21–23]. This dissociation suggests that humoral or neural fac-and the b myosin heavy chain followed by increased protein

synthesis without DNA synthesis (hypertrophy) [5–9]. In a tors or both, as well as haemodynamic overload, are re-sponsible for the development and regression of cardiac hy-recent report, we showed that stretching of cardiac myo-

cytes sequentially activates a protein kinase cascade of pertrophy. Accumulating evidence from clinical studies asphosphorylation. Both initial and maximum activation oc- well as from in vivo animal studies and cardiac cell-culturecur in the order of Raf-1 kinase, mitogen-activated protein studies shows that the cardiac renin–angiotensin system is(MAP) kinase, ERKs and p90rsk. ERKs were reported to linked to the formation of pressure-overload hypertrophy [24].translocate into the nucleus upon activation and activate All components of the renin–angiotensin system [e.g., renin,transcription factors that contain potential ERK phosphor- angiotensinogen, angiotensin-converting enzyme (ACE) andylation sites. [10]. We previously showed that mechanical angiotensin II (Ang II) receptors] were identified in theloading induces the expression of immediate early response heart at both the mRNA and the protein levels [25]. In ad-genes such as c-myc and c-fos as an early event [5, 11, 12]. dition, the renin–angiotensin system was shown to be acti-Therefore, when mechanical stress has been received by vated in experimental left ventricular hypertrophy inducedcardiac myocytes, the signal may spread in cardiac myocytes by haemodynamic overload. Angiotensinogen, ACE andby a protein kinase cascade in the order of Raf-1 kinase, Ang II type 1 receptor mRNA levels are increased in theMAP kinase kinase, ERKs and p90rsk and may evoke the nu- hypertrophied left ventricle of rats [25, 26]. Moreover, sub-clear events (gene expression). pressor doses of ACE inhibitors and Ang II types 1 receptor

Whether a short time course of activation of the Raf-1 ki- antagonist can reduce cardiac hypertrophy without a de-nase–ERK pathway contributes to cardiac cellular hypertro- crease in systemic systolic blood pressure [23, 27]; an in-phy that is chronic remains uncertain. In this regard, how- crease in left ventricular mass produced by abdominal aorticever, many lines of evidence have suggested that ERKs are constriction can be completely prevented by an ACE inhib-key molecules in intracellular signal transduction and play itor with no change in afterload and plasma renin activityessential roles in cellular proliferation and differentiation in [28], and the treatment of spontaneously hypertensive ratsmany cell types [13, 14]. In cardiac myocytes, although a re- with an Ang II type 1 receptor antagonist CV-11974, butport showed that there are discrepancies between ERK acti- not with a vasodilator hydralazine, not only reduces thevation and hypertrophic responses and that interfering thickness of the left ventricular wall, but also decreases bothmutants of ERKs do not block phenylephrine-induced acti- the relative amount of V3 myosin heavy chain and intersti-vation of the atrial natriuretic peptide promoter [15], acti- tial fibrosis [23]. Furthermore, among many kinds of anti-vation of ERKs was reported to be necessary for phenyleph- hypertensive agents, ACE inhibitors exert the most potentrine-induced transactivation of immediate early genes such hypertrophy-reducing effects on the human heart [29], re-as c-fos and foetal-type genes such as atrial natriuretic pep- sulting in improvement of left ventricular function and pro-tide [16]. We previously showed that mechanical stretch of longation of survival. These findings suggest that the locallycardiomyocytes activates a protein kinase cascade of phos- produced renin–angiotensin system plays a pivotal role inphorylation and induces specific gene expression followed pressure-overload cardiac hypertrophy, that Ang II acts toby increased protein synthesis [6, 8, 9]. This serial activa- promote the growth of cardiomyocytes by an autocrine ortion of protein kinases and immediate early gene expression paracrine mechanism and that the regression of hypertro-would trigger the long-term events such as the increase in phy by ACE inhibitors and the Ang II receptor antagonistprotein synthesis and the expression of foetal-type genes is due not only to the decreased blood pressure, but also to[17]. ERKs have been recently reported to activate phos- inbition of the tissue renin–angiotensin system. Indeed, thephorylated heat- and acid-stable protein regulated by insu- recent in vitro studies demonstrated that mechanical stretchlin, resulting in its dissociation from a cap-binding protein, induces secretion of Ang II from cytosolic granules of neo-eIF-4E, which initiates and increases cap-dependent protein natal rat cardiomyocytes and that the secreted Ang II in-synthesis [18, 19]. Quite recently, eIF-4E was demonstrated to

duces cardiomyocyte hypertrophy [23, 30, 31]. Stretch ofbe phosphorylated by electrically stimulated contraction of

cardiomyocytes is also found to increase the level of angio-adult feline cardiomyocytes in vitro and acute pressure overloadtensinogen mRNA [32], suggesting that a positive feedbackon canine hearts in vivo [20]. Although the activation ofmechanism sustains the hypertrophic response.ERKs does not always lead to the development of cardiac

hypertrophy [15], a growing body of evidence suggests theimportant role of ERKs in cardiac hypertrophy. OTHER HUMORAL FACTORS

MEDIATE CARDIAC HYPERTROPHYINDUCED BY PRESSURE OVERLOADTHE ROLE OF THE

CARDIAC RENIN–ANGIOTENSIN An increase in tumour growth factor b (TGFb) mRNA lev-SYSTEM IN PRESSURE-OVERLOAD els was reported in rat hearts overloaded by aortic bandingCARDIAC HYPERTROPHY [33]. Because the increase is recgonised 12 h after banding,

it is unlikely that TGFb is an initial mediator for a varietyLack of a correlation between an elevated arterial pressureand an increased myocardial mass was demonstrated in hy- of events induced by mechanical stress. On the contrary, we

Signalling for Cardiac Hypertrophy 695

THE ROLE OF EXTRACELLULARrecently clarified the involvement of endothelin-1 (ET-1)MATRIX AND CYTOSKELETONin mechanical stretch-induced hypertrophic responses by

using an ET-1 receptor-specific antagonist [34]. An ET type Mechanical stress is transduced into the cell from the sitesA receptor-specific antagonist, BQ123, significantly inhib- at which cells attach to the extracellular matrix (ECM)its stretch-evoked activation of Raf-1 kinase and ERKs and [41], and transmembrane ECM receptors, such as the inte-uptake of amino acids into cells. We further demonstrated grin family adhesion molecules, are good candidates forthat ET-1 is constitutively secreted from the cultured cardi- mechnoreceptors. A large extracellular domain of integrinomyocytes of neonatal rats and that mechanical stretch en- receptor complex binds to various ECM proteins, whereas ahances the ET-1 release from the cells. Additionally, short cytoplasmic domain interacts with the cytoskeleton inmRNA levels of ET-1 were increased by stretching of cardi- the cell [42]. Integrins, which are heterodimeric proteinsomyocytes. These results suggest that not only Ang II, but composed to non-covalently associated a and b subunits,also ET-1 plays an important role in mechanical stress- can transmit signals not only by organizing the cytoskele-induced cardiac hypertrophy. ton, but also by altering biochemical properties such as the

extent of tyrosine phosphorylation of a complex of proteinsincluding pp125FAK [41]. In addition, because cytoskeletonSTRETCH-ACTIVATEDproteins can potentially regulate plasma membrane proteinsCHANNELS IN HYPERTROPHICsuch as enzymes, ion channels and antiporters, mechanicalRESPONSES IN CARDIAC MYOCYTESstress could modulate these membrane-associated proteins

Many cells respond to a variety of environmental stimuli by and stimulate second-messenger systems through the cy-ion channels in the plasma membrane. Mechanosensitive toskeleton. Furthermore, the Rho family small GTP-bind-ion channels have been observed with single-channel re- ing proteins, which play a key part in regulating the actincordings in more than 30 cell types of prokaryotes, plants, cytoskeleton and cell adhesion through integrin receptors,fungi and all animals so far examined [35]. Moreover, the are activated in stretched cardiocytes independently of au-activation of mechanosensitive ion channels was proposed tocrinely released Ang II and ET-1 (our unpublished ob-as the transduction mechanism between load and protein servation). The importance of ECM-integrin-cytoskeletonsynthesis in cardiac hypertrophy [36]. The stretch-activated on mechanotransduction was recently demonstrated by ap-channels allow the passage of the major monovalent physio- plying a new technique [43]. To directly apply mechanicallogical cations Na1 and K1 and the divalent cation Ca21. load to specific cell surface molecules without changing cellWith use of a Ca21-binding fluorescent dye and the patch- shapes, Wang and his colleagues bound spherical ferromag-clamp technique, Ca21 influx through stretch channels was netic microbeads coated with specific receptor ligands toshown to lead to waves of calcium-induced calcium release the cell surface, magnetising the beads in one direction, andin cardiac myocytes [37]. then applied a second, weaker magnetic field oriented at

When the Na1 ionophore monensin or veratridine is 908, by which they twisted the beads in place and therebyadded to cultured cardiomyocytes c-fos expression is ob- exerted a controlled shear stress on the bound cell surfaceserved, possibly because of increased Ca21 uptake by the receptors. They showed that integrin b1 not only inducesNa1–Ca21 exchange mechanism [33]. However, the expres- focal adhesion formation, but also supports a force-depen-sion of foetal-type genes is not induced by Na1 increase dent stiffening response, and that an increase in the cy-(our unpublished observation). Although we cannot rule toskeletal stiffness to the applied stress requires the intactout the existence of the inhibitor-insensitive stretch chan- cytoskeleton. These findings suggest that mechanical stressnels in cardiomyocytes, gadolinium and streptomycin do is first received by integrin and that, next, interlinked actinnot inhibit immediate early gene expression and protein microfilaments transduce mechanical stress in concert withsynthesis by stretch [33, 38]. The Na1 channel blocker te- microtubles and intermediated filaments. Furthermore, a re-trodotoxin also does not affect the induction of c-fos gene cent report indicates that the integrin-linked focal adhesionexpression by stretch [33]. On the other hand, we recently kinase pp125FAK exhibits extracellular matrix-dependentindicated that the Na1–H1 exchanger mediates mechani- phosphorylation on tyrosine and physically associates withcal stretch-induced hypertrophic responses such as Raf-1 ki- non-receptor protein kinases through their Src homology 2

domains [44].nase and ERK activation followed by increased protein syn-thesis in cultured cardiomyocytes [39, 40]. Pre-treatmentwith a specific inhibitor of stretch-sensitive cation channels

NOREPINEPHRINE-INDUCED(gadolinium and streptomycin), of ATP-sensitive K1 chan-CARDIAC HYPERTROPHYnels (glibenclamide) or of hyperpolarisation-activated in-

ward channels (CsCl) does not show inhibitory effects on Not only mechanical stress, but also neurohumoral factorsERK activities increased by mechanical stretch. A specific induce cardiac hypertrophy. Norepinephrine (NE) is a po-inhibitor of the Na1–H1 exchanger, HOE 694, however, tent growth factor for cardiac myocytes [45]. Because pro-markedly attenuates stretch-induced activation of Raf-1 ki- longed infusion of subhypertensive doses of NE induces an

increase in the mass of the myocardium and the thicknessnase and ERKs and inhibits the stretch-induced increase inphenylalanine incorporation into proteins [40]. of the left ventricular wall, NE has been postulated to have

696 T. Yamazaki et al.

a direct hypertrophic effect on cardiac myocytes without af- generally activate ERKs in a synergistic manner, variouskinds of PKC activators such as Ang II, phenylephrine,fecting afterload [46]. There are two major subtypes, a and

b, in NE receptors. Among a-adrenoceptors, the a1-adreno- ET-1 and mechanical stretch were added to cultured cardiacmyocytes with forskolin. Each PKC activator increases ERKceptors, but not the a2-adrenoceptors, exist in cardiomyo-

cytes. On the other hand, both b1- and b2-adrenoceptors are activities, and the simultaneous addition of forskolin andexpressed in the human heart. Simpson [45] reported that the PKC activators provokes much greater increases in ERKNE stimulates hypertrophy of neonatal rat cardiomyocytes activities [54].in culture through a1-adrenoceptors but not through b-adre-noceptors. He also showed that an increase in beating rate

ANG II-INDUCED SIGNALrequires both a1 and b1-adrenoceptor activation [47]. FromTRANSDUCTION PATHWAYS IN CARDIACthese results, he and his colleagues argued that the hypertro-MYOCYTES AND IN CARDIAC FIBROBLASTS

phic effect of b-adrenoceptor agonists could be secondary tothe release of catecholamines that have a1-stimulatory prop- Ang II directly induces cardiomyocyte hypertrophy inde-

pendently of an increase in vascular resistance or cardiacerties [48]. Clark et al. [49] reported that b-adrenergicactivation produces cardiomyocyte hypertrophy by activat- afterload [24]. Ang II has been reported to activate Src fam-

ily tyrosine kinases in cardiac myocytes, leading to the acti-ing the beating of cardiac myocytes. In addition, Bishopricet al. [50] showed that NE-induced hypertrophy, as well as vation of small GTP-binding protein Ras through sequen-

tial activation of Shc, Grb2 and Sos [55]. In contrast, bycontractility and skeletal a-actin gene expression, is medi-ated by b-adrenoceptors in high density cultures (1 3 103 using a variety of inhibitors, we demonstrated that, in car-

diac myocytes, Ang II activates Raf-1 kinase and ERKcells/mm2). They also reported that, when cardiomyocytesare plated at a low density (3 3 102 cells/mm2), minimising through the PKC-dependent pathway but not through the

pathway of tyrosine kinases or Ras [56]. Inhibition of PKCcell contact, NE-induced skeletal a-actin gene expressionand hypertrophy are mediated by a1-adrenoceptors, and with calphostin C or down-regulation of PKC by pre-treat-

ment with a phorbol ester abolishes Ang II-induced activa-they postulated that factors related to cell communicationinfluence the pathways mediating NE-regulated gene tran- tion of Raf-1 kinase and ERKs, and addition of a phorbol es-

ter conversely induces a marked increase in the activities ofscription during cardiomyocyte hypertrophy. On the otherhand, Dubus et al. [51] reported that protein synthesis could Raf-1 kinase and ERKs. In addition, pre-treatment with two

chemically and mechanistically dissimilar tyrosine kinasebe stimulated through b- but not a-adrenoceptors in cul-tured cardiac myocytes of adult rats, and Pinson et al. [52] inhibitors, genistein and tyrphostin, does not attenuate

Ang II-induced activation of ERKs. Over-expression ofshowed that an increase in protein synthesis is induced byboth a1 and b-adrenoceptor agonists in cardiac myocytes. C-terminal Src kinase (Csk), which inhibits the function of

Src family tyrosine kinases, also has no effect on Ang II-Thus, the pathways through which NE induces cardiomyocytehypertrophy are as yet controversial. We recently reported induced activation of ERKs in cardiac myocytes. Further-

more, pre-treatment with manumycin, a Ras farnesyl trans-that the activation of protein kinase cascades followed byincreased protein synthesis is induced by cAMP-dependent ferase inhibitor, or over-expression of a dominant negative

mutant of Ras does not affect Ang II-induced activation ofprotein kinase A (PKA) through b-adrenoceptors as well asby protein kinase C (PKC) through a1-adrenoceptors [53]. ERKs in cardiomyocytes.

Cardiac myocytes possess the type 1 and type 2 subtypesNE-induced activation of ERKs is partly inhibited by eitherthe a1-adrenoceptor blocker prazosin or the b-adrenoceptor of Ang II receptors, both of which couple to second-mes-

senger generation and gene expression. We have shownblocker propranolol and is completely abolished by bothblockers. Both the b-adrenoceptor agonist isoproterenol that the aforementioned hypertrophic responses are medi-

ated by the type 1 receptors [31, 35]. On the other hand, noand the a1-adrenoceptor agonist phenylephrine increasethe activities of Raf-1 kinase and ERKs and the incorpora- physiologic role has yet been ascribed to the type 2 recep-

tors. We recently found that the type 2 receptor antago-tion of phenylalanine into proteins. Moreover, inhibition ofPKA by RpcAMP and down-regulation of PKC by long ex- nist PD123319 augments Ang II-induced hypertrophic re-

sponses such as ERK activation (unpublished observation).posure with a phorbol ester completely abolish isoprotere-nol- and phenylephrine-induced ERK activation, respec- This finding raises the possibility that Ang II counteracts its

own hypertrophic effects on cardiac myocytes.tively. We have further shown that activation of ERKs byPKA activators is dependent on trans-sarcolemmal Ca21 in- Ang II has also been known to evoke a variety of signals

and to induce the proliferation of cardiac fibroblasts [57].flux and independent of PKC. Moreover, PKA activatorssynergistically activate Raf-1 kinase and ERKs with PKC To determine the molecular mechanism by which Ang II

displays different effects on cardiac myocytes and fibro-activators such as phorbol ester followed by increased pro-tein synthesis. The activity of ERKs is increased by forskolin blasts, we examined signal transduction pathways leading to

the activation of ERKs [56, 58]. Ang II-induced ERK acti-(approximately 4-fold) and by phorbol ester (approximately5-fold), whereas the simultaneous administration of both vation is abolished by pre-treatment with pertussis toxin

and by over-expression of the Gbg subunit-binding domainagents synergistically increases the activity by approximately12-fold. To elucidate whether PKA and PKC activators of the b-adrenergic receptor kinase 1 in cardiac fibroblasts

Signalling for Cardiac Hypertrophy 697

but not in cardiac myocytes. Inhibition of PKC strongly in- highly divergent among cell types. PKC, but not tyrosine ki-activates ERKs activated by Ang II in cardiac myocytes, nases or Ras, are critical for Ang II-induced hypertrophic re-whereas inhibitors of tyrosine kinases but not of PKC abol- sponses in cardiac myocytes. In contrast, Ang II induces theish Ang II-induced ERK activation in cardiac fibroblasts. proliferation of cardiac fibroblasts through the bg-subunitsOver-expression of Csk suppresses the activation of trans- of Gi proteins-Src-Ras-dependent pathway. The presence offected ERKs in cardiac fibroblasts. Ang II rapidly induces multiple and cell-type-specific signal transduction pathwaysthe phosphorylation of Shc and the association of Shc with may be favourable for cardiomyocytes to quickly and effi-Grb2. Co-transfection of the dominant-negative mutant of ciently respond to external stimuli such as neurohumoralRas or of Raf-1 kinase abolishes Ang II-induced ERK acti- factors and mechanical overload. Moreover, still to be es-vation in cardiac fibroblasts. In contrast, over-expression of tablished are the physiologic and the potential pathophysi-Csk or of dominant-negative Ras has no effect on Ang II- ologic roles of the Ang II type 2 receptor. The gene knock-induced ERK activation in cardiac myocytes. Collectively, out and over-expression techniques are powerful ways toAng II-evoked signal transduction pathways differ among study gene functions, and the loss-of-function and gain-of-cell types. In cardiac fibroblasts, Ang II activates ERKs function animals would be sources of new insight into thethrough a pathway including the Gbg subunit of Gi pro- functional importance of the renin–angiotensin system intein, tyrosine kinases including Src family tyrosine kinases, cardiac hypertrophy and, as a consequence, various heartShc, Grb2, Ras and Raf-1 kinase, whereas Gq and PKC are diseases.important in cardiac myocytes.

This work was supported by a Grant-in-Aid from the Ministry of Edu-cation, Science, Sports, and Culture, Japan. We thank M. Ono, K.

CONCLUSIONS Abe and A. Kondoh for the excellent secretarial assistance.

A major intriguing question remains unanswered. How ismechanical stress converted into biochemical signals? In Referencesother words, what is the mechanoreceptor or the transducer 1. Morgan H. E., Gordon E. E., Kita Y., Chua B. H. L., Russo

L. A., Peterson C. J., McDermott P. J. and Watson P. A.for mechanical stress in cardiomyocytes? Mechanical stress(1987) Annu. Rev. Physiol. 49, 533–543.is assumed to directly change the conformations of the func-

2. Levy D., Garrison R. J., Savage D. D., Kannel W. B. andtional proteins such as enzymes or GTP-binding proteins orCastelli W. P. (1990) N. Engl. J. Med. 322, 1561–1566.

to directly activate enzymes such as phospholipase by plac- 3. Peterson M. B. and Lesch M. (1972) Circ. Res. 31, 317–327.ing the enzymes close to their phospholipid substances in 4. Cooper G., Kent R. L., Uboh C. E., Thompson E. W. and Ma-the plasma membrane. As mentioned earlier, the Na1–H1 rino T. A. (1985)J. Clin. Invest. 75, 1403–1414.

5. Komuro I., Kaida T., Shibazaki Y., Kurabayashi M., Takaku F.exchanger or the integrin-cytoskeleton complex may be anand Yazaki Y. (1990) J. Biol. Chem. 265, 3595–3598.alternative candidate structure for a mechanoreceptor and

6. Komuro I., Katoh Y., Kaida T., Shibazaki Y., Kurabayashi M.,a transducer. Moreover, our preliminary study suggests that Takaku F. and Yazaki Y. (1991) J. Biol. Chem. 266, 1265–mechanical stretch induces activation of Rho family GTP- 1268.binding proteins, which in turn triggers the release of Ang 7. Sadoshima J. and Izumo S. (1993) EMBO J. 12, 1681–1692.

8. Yamazaki T., Tobe K., Hoh E., Maemura K., Kaida T., Ko-II and ET-1 from cardiac myocytes. Further studies are re-muro I., Tamemoto H., Kadowaki T., Nagai R. and Yazaki Y.quired to identify specific signalling molecules, including(1993) J. Biol. Chem. 268, 12069–12076.mechanoreceptors and mechanotransducers.

9. Yamazaki T., Komuro I., Kudoh S., Zou Y., Shiojima I., Mi-In the last part of this review, we concentrated on NE- zuno T., Takano H., Hiroi Y., Ueki K., Tobe K., Kadowaki T.,

and Ang II-induced hypertrophic responses. NE activates Nagai R. and Yazaki Y. (1995) J. Clin. Invest. 96, 438–446.10. Davis R. J. (1993) J. Biol. Chem. 268, 14553–14556.protein kinases such as Raf-1 kinase and ERKs and increases11. Komuro I., Kurabayashi M., Takaku F. and Yazaki Y.protein synthesis in cardiomyocytes of neonatal rats. It is

(1988)Circ. Res. 62, 1075–1079.noteworthy that, in contrast with other cell types, not only12. Komuro I., Shibazaki Y., Kurabayashi M., Takaku F. and Ya-

PKC activation (by a1-adrenoceptor stimulation), but also zaki Y. (1990)Circ. Res. 66, 979–985.PKA activation (by b-adrenoceptor stimulation) increases 13. Pages G., Lenormand P., L’allemain G., Chambard J.-C., Mel-the activities of Raf-1 kinase and ERKs in cardiac myocytes oche S. and Pouyssegur J. (1993) Proc. Natl. Acad. Sci. USA

90, 8319–8323.and induces hypertrophy. Moreover, PKA-activating agents14. Cowley S., Paterson H., Kemp P. and Marshall C. J. (1994)synergistically activate Raf-1 kinase and ERKs and induce

Cell 77, 841–852.cardiomyocyte hypertrophy with PKC-activating agents. 15. Post G. R., Goldstein D., Thuerauf D. J., Glembotski C. C.Hypertrophic signals evoked by PKA and PKC activation and Brown J. H. (1996) J. Biol. Chem. 271, 8452–8457.may converge at the Raf-1 kinase–ERK cascade and may in- 16. Thorburn J., Frost J. A. and Thorburn A. (1994) J. Cell Biol.

126, 1565–1572.duce a variety of hypertrophic responses. In this regard, the17. Komuro I. and Yazaki Y. (1993) Annu. Rev. Physiol. 55,recent report of Daaka et al. [59] indicates that PKA acti-

55–75.vated by b2-adrenergic stimulation phosphorylates the b2- 18. Haystead T. A., Haystead C. M., Hu C., Lin T. A. and Law-adrenergic receptor and that the phosphorylation results in rence J. J. (1994) J. Biol. Chem. 269, 23185–23191.ERK activation. On the other hand, Ang II-induced signal 19. Pause A., Belsham G. J., Gingras A. C., Donze O., Lin T. A.,

Lawrence J. J. and Sonenberg N. (1994) Nature 371, 762-767.transduction pathways through the type 1 receptors are

698 T. Yamazaki et al.

20. Wada H., Ivester C. T., Carabello B. A., CooperIV G. and 40. Yamazaki T., Komuro I., Kudoh S., Zou Y., Nagai R., AikawaR., Uozumi H. and Yazaki Y. (1998) Role of ion channels andMcDermott P. J. (1996) J. Biol. Chem. 271(14), 8359–8364.exchangers in mechanical stretch-induced cardiomyocyte hy-21. Sen S., Tarazi R. C., Khairallah P. A. and Bumpus F. M.pertrophy. Circ Res 82, 430–437.(1974) Circ. Res. 35, 775–781.

41. Juliano R. L. and Haskill S. (1993) J. Cell Biol. 120, 577–585.22. Frohlich E. D. and Tarazi R. C. (1979) Am. J. Cardiol. 44,42. Hynes R. O. (1992) Cell 69, 11–25.959–963.43. Wang N., Butler J. P. and Ingber D. E. (1993) Science 260,23. Kojima M., Shiojima I., Yamazaki T., Komuro I., Zou Y.,

1124–1127.Wang Y., Mizuno T., Ueki K., Tobe K., Kadowaki T., Nagai44. Schaller M. D., Hildebrand J. D., Shammon J. D., Fox J. W.,R. and Yazaki Y. (1994) Circulation 89, 2204–2211.

Vines R. R. and Parsons J. T. (1994) Mol. Cell. Biol. 14,24. Baker K. M., Booz G. W. and Dostal D. E. (1992) Annu. Rev.1680–1688.Physiol. 54, 227–241.

45. Simpson P. (1983) J. Clin. Invest. 72, 732–738.25. Suzuki J., Matsubara H., Urakami M. and Inada M. (1993)46. Laks M. M., Morady F. and Swan H. J. C. (1973)Chest 64,Circ. Res. 73, 439–447. 75–78.26. Schunkert H., Dzau V. J., Tang S. S., Hirsch A. T., Apstein 47. Simpson R. (1985) Circ. Res. 56, 884–889.

C. S. and Lorell B. H. (1990) J. Clin. Invest. 86, 1913–1920. 48. Simpson P. C., Kariya K., Karns L. R., Long C. S. and Karliner27. Linz W., Schoelkens B. A. and Ganten D. (1989) Clin. Exp. J. S. (1991) Mol. Cell. Biochem. 104, 35–43.

Hypertens. 11, 1325–1350. 49. Clark W. A., Rudnick S. J., Lapres J. J., Lesch M. and Decker28. Baker K. M., Cherin M. I., Wixon S. K. and Aceto J. F. (1990) R. S. (1991) Am. J. Physiol. 261, C530–C542.

Am. J. Physiol. 259, H324–H332. 50. Bishopric N. H. and Kedes L. (1991) Proc. Natl. Acad. Sci.29. Dahlof B., Pennert K. and Hansson L. (1992) Am. J. Hyper- USA 88, 2132–2136.

tens. 5, 95–110. 51. Dubus I., Samuel J. L., Marotte F., Delcayre C. and Rappaport30. Sadoshima J., Xu Y., Slayter H. S. and Izumo S. (1993) Cell L. (1990) Circ. Res. 66, 867–874.

75, 977–984. 52. Pinson A., Schluter K. D., Zhou X. J., Schwartz P., Kessler31. Yamazaki T., Komuro I., Kudoh S., Zou Y., Shiojima I., Mi- I. G. and Piper H. M. (1993) J. Mol. Cell. Cardiol. 25, 477–

490.zuno T., Takano H., Hiroi Y., Ueki K., Tobe K., Kadowaki T.,53. Yamazaki T., Komuro I., Zou Y., Kudoh S., Shiojima I., HiroiNagai R. and Yazaki Y. (1995) Circ. Res. 77, 258–265.

Y., Mizuno T., Aikawa R., Takano H. and Yazaki Y. (1997)32. Shyu K. G., Chen J. J., Shih N. L., Chang H., Wang D. L.,Circulation 95, 1260–1268.Lien W. P. and Liew C. C. (1995) Biochem. Biophys. Res.

54. Yamazaki T., Komuro I., Zou Y., Kudoh S., Mizuno T., HiroiCommun. 211, 241–248.Y., Shiojima I., Takano H., Kinugawa K. I., Kohmoto O., Ta-33. Komuro I., Katoh Y., Hoh E., Takaku F. and Yazaki Y. (1991)kahashi T. and Yazaki Y. (1997) J. Mol. Cell. Cardiol. 29,Jpn. Circ. J. 55, 1149–1157.2491–2501.34. Yamazaki T., Komuro I., Kudoh S., Zou Y., Shiojima I., Hiroi

55. Sadoshima J. and Izumo S. (1996) EMBO J. 15, 775–787.Y., Mizuno T., Maemura K., Kurihara H., Aikawa R., Takano56. Zou Y., Komuro I., Yamazaki T., Aikawa R., Kudoh S., Shio-H. and Yazaki Y. (1996) J. Biol. Chem. 271, 3221–3228. jima I., Hiroi Y., Mizuno T. and Yazaki Y. (1996) J. Biol.

35. Morris C. E. (1990) J. Membr. Biol. 113, 93–107. Chem. 271, 33592–33597.36. Kent R. L., Hoober K. and Cooper G, IV, (1989) Circ. Res. 57. Schorb W., Booz G. W., Dostal D. E., Conrad K. M., Chang

64, 74–85. K. C. and Baker K. M. (1993) Circ. Res. 72, 1245–1254.37. Sigurdson W., Ruknudin A. and Sachs F. (1992) Am. J. Phys- 58. Zou Y., Komuro I., Yamazaki T., Kudoh S., Aikawa R., Uo-

iol. 262, H1110–H1115. zumi H., Zhu W., Shiojima I., Hiroi Y., Mizuno T. and Yazaki38. Sadoshima J., Takahashi T., Jahn L. and Izumo S. (1992) Y. (1998) Cell type specific angiotensin II-evoked signal

Proc. Natl. Acad. Sci. USA 89, 9905–9909. transduction pathways: Critical role of Gbg subunit, Src fam-39. Takewaki S., Kuro-o M., Hiroi Y., Yamazaki T., Noguchi T., ily and Ras in cardiac fibroblasts. Circ Res 82, 337–345.

Miyagishi A., Nakahara K., Aikawa M., Manabe I., Yazaki Y. 59. Daaka Y., Luttrell L. M. and Lefkowitz R. J. (1997) Nature390, 88–91.and Nagai R. (1995) J. Mol. Cell. Cardiol. 27, 729–742.