G-Proteins in Growth and Apoptosis: Lessons from the Heart

John W Adams1,2 and Joan Heller Brown*,1

1University of California, San Diego, Department of Pharmacology, 9500 Gilman Drive, 0636, La Jolla, CA, California 92093- 0636, USA

The acute contractile function of the heart is controlled by for proliferation of adult cardiomyocytes. In light of the eects of released nonepinephrine (NE) on cardiac these considerations, it is not immediately obvious that adrenergic receptors. NE can also act in a more chronic cardiomyocyte growth and cardiomyocyte death would fashion to induce cardiomyocyte growth, characterized by be responses critical to the normal function of the heart. cell enlargement (hypertrophy), increased protein synth- In fact, the ability of cardiomyocytes to undergo esis, alterations in gene expression and addition of hypertrophic growth, which includes an increase in cell sarcomeres. These responses enhance cardiomyocyte size, is an important adaptive response to a wide range of contractile function and thus allow the heart to conditions that require the heart to work more compensate for increased stress. The hypertrophic eects eectively. As described below, adaptive or compensa- of NE are mediated through Gq-coupled a1-adrenergic tory cardiomyocyte hypertrophy appears to be regulated receptors and are mimicked by the actions of other in large part through stimulation of G-protein coupled neurohormones (endothelin, prostaglandin F2a angiotensin receptors (GPCRs). Often, the ability of cardiomyocytes II) that also act on Gq-coupled receptors. Activation of to function at high capacity under increased workload phospholipase C by Gq is necessary for these responses, cannot be sustained and the heart transitions into a and protein kinase C and MAP kinases have also been condition in which ventricular failure develops. One implicated. Gq stimulated cardiac hypertrophy is also event associated with and suspected to be causally evident in transgenic mouse models. In contrast, stimula- related to this transition to heart failure is the apoptotic tion of Gs-coupled b-adrenergic receptors or Gi-coupled death of cardiomyocytes. An intriguing hypothesis is receptors do not directly eect cardiomyocyte hypertro- that prolonged or intensi®ed activation of GPCRs may phy. Apoptosis is also induced by G-protein-coupled be a critical factor tipping the balance between pathways receptor stimulation in cardiomyocytes. Sustained or promoting hypertrophic growth and those leading to excessive activation of either Gq- or Gs-signaling path- apoptotic death of cardiomyocytes. ways results in apoptotic loss of cardiomyocytes both in vitro and in vivo. Apoptosis is associated with decreased ventricular function in the failing heart. Cardiomyocytes In vitro and in vivo models of cardiomyocyte growth provide an ideal model system for understanding the basis for G-protein mediated hypertrophy and apoptosis, and There are no well-established continuous cell lines that the mechanisms responsible for the transition from can be used to study cardiomyocyte development and compensatory to deleterious levels of signaling. This growth. A line of atrial cells (AT-1), immortalized by information may prove critical for designing interventions expression of SV-40 large T-antigen (Delcarpio et al., that prevent the pathophysiological consequences of heart 1991) has been utilized in electrophysiolgical studies failure. Oncogene (2001) 20, 1626 ± 1634. and the HL-1 cardiac muscle cell line has been derived from them (Claycomb et al., 1998). The availability of Keywords: cardiac; Gq; heart failure; hypertrophy; these cells is limited, however, and conditions for cardiomyocyte maintaining their dierentiated properties are stringent. Some investigators have studied embryonic stem cells dierentiated into cardiomyocytes (Minamino et al., Introduction 1999) but this preparation is not routinely useful. Myocytes isolated from the ventricle of adult rat, Cardiomyocytes comprise 70 ± 80% of the mass of the rabbit, and dog heart have been extensively examined adult heart. These cells are terminally dierentiated and to delineate the acute signaling pathways by which precisely designed to rapidly alter their ionic and GPCRs control cell excitability, calcium and excita- contractile function in response to adrenergic (sympa- tion-contraction coupling. However, adult heart cells thetic) and cholinergic (parasympathetic) input. Cardiac are generally dicult to maintain in cell culture and tumors are extremely rare and there is limited evidence substantial changes in their dierentiated properties can occur. Thus adult cardiomyocytes are rarely used to examine control of cell growth. The vast majority of work carried out over the last *Correspondence: JH Brown 2Current address: Arena Pharmaceuticals, Inc., 6166 Nancy Ridge 2 ± 3 decades has utilized primary cultures of neonatal Drive, San Diego, CA 92121, USA rat ventricular myocytes (NRVM) to map the signaling G proteins and cardiac growth JW Adams and JH Brown 1627 molecules and pathways involved in hypertrophic eciency gene delivery has been achieved by infection cardiomyocyte growth. Since little cell division occurs of NRVMs with cDNAs expressed by replication postnatally in cardiac myocytes, increased cardiac mass de®cient recombinant adenoviral vectors. Because is achieved primarily through increases in cardiomyo- virtually all of the cells are infected and high-level cyte cell volume. Hypertrophic growth of myocytes is expression is obtained, biochemical and molecular accompanied by increases in protein synthesis and changes underlying cardiomyocyte hypertrophy and organization of contractile proteins, such as myosin apoptosis can be assessed. light chain-2 (MLC-2) into sarcomeric units. In The in vitro NRVM system has many advantages addition, fetal isoforms of various genes including b- including ease of culture. Importantly, it circumvents myosin heavy chain (MHC) and atrial natriuretic the complexity imposed by interactions of individual factor (ANF) are re-expressed by hypertrophied myocytes with each other, with non-myocytes, with the myocytes (Parker et al., 1990). extracellular matrix and with circulating growth factors. The initial paradigm demonstrating that neurohor- This degree of isolation provides an ideal platform for mones could increase cardiomyocyte size and their the discovery of signaling molecules through which the capacity for protein synthesis documented hypertrophy cardiomyocyte responds to extracellular stimuli or of cultured NRVM in response to norepinephrine (NE) intracellularly expressed mediators. On the other hand, treatment (Simpson et al., 1982; Simpson, 1983). Using the interactions lacking in this in vitro system could play variations of this model, others have demonstrated that a central role in the in vivo control of cardiomyocyte cardiomyocyte hypertrophy occurs in response to growth and death. This dimension is provided by stimulation by a variety of G-protein coupled receptor studies using transgenic gene expression in the myo- agonists, most of which are listed in Table 1. cardium, and/or deletion of putative signaling molecules Investigation of the signaling pathways mediating by gene knockout. The a-MHC promoter has been hypertrophic growth of NRVMs was signi®cantly commonly used to achieve cardiac speci®c gene advanced by the recognition that these primary expression in transgenic mice. For example, signaling cultured cells could be genetically manipulated. Of molecules including G-protein coupled receptors, G- critical importance was the observation that transcrip- protein alpha subunits, and kinases such as MAP tional responses could be assessed using the sensitive kinases and PKC have been selectively expressed in luciferase reporter gene, driven by promoters for the myocardium by transgenesis. Despite the caveats genes upregulated in hypertrophy (e.g. ANF, MLC-2). applicable to transgenic overexpression of signaling Thus even in the face of the relatively low transfection molecules, this paradigm has been invaluable for eciency obtained using CaPO4 or electroporation of evaluating the physiological function of candidate NRVMs, genes encoding molecules of interest could be molecules discovered using the in vitro cell system (see transfected into NRVMs and their eects on regulation Izumo and Shioi, 1998 for review). of hypertrophic responses examined. Proteins, anti- bodies or cDNAs encoding genes of interest have also been microinjected into NRVMs. While this approach G-protein involvement in cardiomyocyte hypertrophy has also provided important information, only responses that can be measured at the single cell level Gq-signaling pathways in cardiomyocyte hypertrophy (e.g. increased immunostaining for ANF protein or altered sarcomere organization visualized by staining Most stimuli that induce myocardial hypertrophy in with phalloidin) can be monitored. Most recently, high NRVMs activate G-protein coupled receptors that

Table 1 G-protein coupled receptors regulating cardiomyocyte growth Agonist Receptor G-protein References (selected)

Ang II AT1 Gq (Sadoshima and Izumo, 1993b)

ET-1 ETA Gq/Gi (Shubeita et al., 1990; Hilal-Dandan et al., 1992; 1997 ET-3 (Tamamori et al., 1996)

PGF2a PGF Gq (Adams et al., 1998a)

PE a1-AdrR Gq (Bishopric et al., 1987; Chien et al., 1991)

LPA Edg (2, 4 or 7 ?) Gq/Gi (Goetzl et al., 2000)

Thrombin PAR-1 Gq/Gi (Glembotski et al., 1993; Sabri et al., 2000)

NE a1-AdrR Gq (Simpson et al., 1982)

ISO b1, b2-AdrR Gs (Simpson, 1985)

SPC Edg (1 or 3 ?) Gi (Sekiguchi et al., 1999)

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1628 regulate the heterotrimeric G-protein Gq (see Table role in Gq-induced signaling in most systems. Interest- 1). Gq-coupled receptor agonists shown to induce ingly it is not clear that signi®cant Ins(1,4,5)P3 hypertrophy in NRVM include norepinephrine (NE), formation occurs in response to the G-protein angiotensin II (Ang II), endothelin-1 (ET-1), throm- activation in NRVMs (Matkovich and Woodcock, 2+ bin, lysophosphatidic acid (LPA) and prostaglandin 2000) or that Ins(1,4,5)P3 induced Ca release is of F2a (PGF2a) (Simpson et al., 1982; Sadoshima and major physiological signi®cance in the heart. The other Izumo, 1993a; Ito et al., 1991; Glembotski et al., product of PLC induced cleavage of PIP2, diacylgly- 1993; Goetzl et al., 2000; Adams et al., 1996). The cerol, likely plays a key role in hypertrophic responses activated alpha subunit of Gq directly stimulates by regulating the activity of protein kinase C (PKC). phospholipase (PLC), catalyzing the hydrolysis of The role of PKC as a growth signaling molecule in the phosphatidylinositol bisphosphate (PIP2). Several lines myocyte has been demonstrated by studies examining of evidence indicate that it is the activation of this the eects of phorbol esters (Dunnmon et al., 1990), by pathway that is responsible for induction of kinase overexpression of constitutively activated forms of cascades and transcriptional responses resulting in PKC in NRVMs (Shubeita et al., 1992; Kariya et al., hypertrophy. Amongst the earliest studies demonstrat- 1991; Takeishi et al., 2000), by studies using ing a requirement for Gq were those in which a pharmacological inhibitors (Komuro et al., 1991; speci®c antibody to the Gq alpha (Gaq) subunit was Karns et al., 1995; Yamazaki et al., 1995, 1999) and microinjected into NRVMs and shown to block PE in experiments with receptors for activated C-kinases stimulated ANF expression, cell enlargement, and (RACKS) (Mochly-Rosen et al., 2000). The role of myo®lament organization (LaMorte et al., 1994); the PKC in cardiac growth regulation is the subject of same antibody was also demonstrated to block the extensive investigation and numerous critical reviews PE and GTPgS stimulated activation of PLC in these (Steinberg et al., 1995; Puceat and Vassort, 1996; cells. Studies comparing responses to heterologously Simpson, 1999; Jalili et al., 1999). MAP Kinases are expressed muscarinic receptor subtypes and chimeras also activated in response to the Gq coupled receptor in NRVM also indicated that the ability to couple to agonists. Their role in hypertrophic cell growth is well Gq was required for activation of hypertrophic documented, albeit not always conclusive (see Sugden signals (Ramirez et al., 1995). The ability to activate and Clerk, 1998 for review). PLC also correlates with the dierential ability of a vs a AdrR to induce cardiomyocyte hyper- 1A/C 1B Gq-coupled receptors in cardiac hypertrophy trophic responses. (McWhinney et al., 1999). Finally recent experiments in our laboratory have demon- Most forms of cardiovascular stress that lead to strated that overexpression of the alpha subunit of cardiac hypertrophy activate the sympathetic nervous Gq by adenoviral infection of NRVMs results in system and elevate plasma NE levels. This observation, increased PLC activity and development of a coupled with the in vitro eects of NE, suggest a causal hypertrophic phenotype similar to that caused by relationship between adrenergic receptor activation and activation of Gq-coupled receptors (Adams et al., cardiac hypertrophy in the intact heart. Hypertrophic 1998b, 2000). The hypertrophic phenotype, induced eects of NE appear to occur predominantly through by agonist or Gaq expression are evident in Figure 1, a1-adrenergic receptor (a1-AdrR) activation. The a1- where actin ®laments are stained with ¯uorescently- AdrR subtypes that regulate cardiomyocyte function tagged phalloidin in ®xed cells. are the a1A/C and a1B (for review see Varma and Deng One of the important second messengers generated 2000). One of the earliest studies using cardiac speci®c when PIP2 is hydrolyzed by PLC is Ins (1,4,5)P3. The gene expression in transgenic mice showed that an production of Ins(1,4,5)P3 and subsequent release of activated mutant of the a1B-AdrR, expressed in the Ca2+ from endoplasmic reticular stores plays a critical mouse ventricle, resulted in mild cardiac hypertrophy

Figure 1 Hypertrophic phenotype, induced by agonist or Gaq expression. Actin ®laments in dierent experimental groups are stained with phalloidin in ®xed cells (see text for detail)

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1629 (Milano et al., 1994a). No hypertrophy developed in challenged heart. However, PGF2a receptor knockouts transgenic mice expressing wild type a1B-AdrR (Akhter have not been analysed for cardiac responses and et al., 1997; Grupp et al., 1998). This may re¯ect the antagonists to PGF2a receptors have not been weak coupling of the a1B-AdrR to PLC (Theroux et al., generated, thus the role of PGF2a in in vivo 1996). Viral expression of the a1B receptor also fails to hypertrophy remains to be determined. stimulate PLC or hypertrophy in NRVMs. In contrast, the a receptor stimulates both inositol phosphate 1A/C Evidence for Gaq signaling in hypertrophic responses of production and ANF-luciferase expression (McWhin- transgenic mice ney et al., 1999). Studies using pharmacological inhibitors to block PE induced hypertrophy and As discussed above a variety of GPCRs including the a1- phosphoinositide hydrolysis in NRVM also support Adr, Ang II, PGF2a and ET1 receptors participate in the the notion that the a1A/C-AdrRs, signaling through Gq regulation of cardiac growth in NRVMs by coupling to to PLC, that mediates the hypertrophic response to Gaq. To determine whether Gaq activation is sucient catecholamines, at least in the rat heart (Simpson et al., to induce hypertrophy in vivo, transgenic mice over- 1991; Knowlton et al., 1993). Transgenic mice in which expressing Gaq in the myocardium were generated the a1A/C receptor is overexpressed and mice in which (D'Angelo et al., 1997). These mice developed cardiac the a1A/C receptor is knocked out have been generated, hypertrophy in proportion to the extent of Gaq however no data on development of hypertrophy in expression, as demonstrated by increases in cardiac these animals is yet published. mass, changes in left ventricular function and alterations It is now well established that ACE inhibitors, which in hypertrophic gene expression. Evidence that Gaqis block Ang II formation, prevent the development of sucient to induce hypertrophy in transgenic mice was pressure overload induced cardiac hypertrophy in animal complemented by several lines of evidence indicating models and in humans (Zhu et al., 1997). However, ACE that Gaq activation was necessary for hypertrophy in inhibition aects other aspects of cardiovascular regula- response to pressure overload induced by transverse tion including sympathetic tone. There is also evidence aortic constriction. In one study, a Gaq inhibitor peptide for attenuation of experimentally induced cardiac previously shown to prevent receptor coupling to Gaq hypertrophy by receptor antagonists to Ang II. The was expressed in the myocardium of transgenic mice salutary eects of AT1 receptor blockers have been (Akhter et al., 1998). Hypertrophy induced by pressure demonstrated to occur independently of the drugs eects overload was signi®cantly attenuated in these animals. A on blood pressure (Linz and Scholkens, 1992; Ehmke et separate line of evidence for the involvement of Gaq al., 1999). Studies using the AT1 antagonists losartan signaling derived from studies using transgenic expres- and valsartan demonstrate a role of AT1 receptors in sion of RGS4, which acts as a GTPase activator protein hypertrophic growth, both in vitro and in vivo (Sadoshi- (GAP) for, and attenuates activation of, Gaq. When ma and Izumo, 1993b; Thurmann et al., 1998). However, expressed in the mouse heart, RGS4 signi®cantly in vitro studies suggest that a major component of the diminished pressure overload induced hypertrophy observed hypertrophy induced by Ang II results from (Rogers et al., 1999). Finally, recent studies demonstrate activation of AT1 receptors on non-muscle cells, release complete absence of a hypertrophic response to pressure of paracrine factors, and activation of growth pathways overload in cardiac-restricted Gaq/Ga11 knockout mice in neighboring myocytes (Ito et al., 1993). Interestingly, (S Oermanns, personal communication). Taken to- transgenic mice overexpressing AT1 receptors in the gether, these results support the hypothesis that Gaq- myocardium develop cardiac hypertrophy (Paradis et al., mediated signaling is both necessary and sucient for 2000), but pressure overload and stretch induced the development of cardiac hypertrophy. hypertrophy still occur in AT1 knockout mice (Kudoh et al., 1998; Harada et al., 1998). Thus, the precise role of Gs signaling in cardiac hypertrophy Ang II system in regulating cardiac growth remains uncertain. The b-Adr receptors are powerful regulators of cardiac Other Gq-coupled receptor agonists, including ET-1 contractile function. Stimulation of b-Adr receptors and PGF2a likely regulate cardiac hypertrophy in vivo. increases heart rate and myocardial contractility by It is known that endothelin receptor blockers inhibit coupling to the heterotrimeric G-protein Gas, which, cardiac hypertrophy induced by hypertension and upon activation, stimulates adenylyl cyclase resulting in pressure overload and these agents have been used the formation of cAMP. Activation of the cAMP successfully in animal models of heart failure (Miyau- dependent protein kinase (PKA) induces the phosphor- chi and Goto, 1999). Thus probably this G-protein ylation of L-type Ca2+ channels (increasing Ca2+ entry coupled receptor agonist contributes to the develop- at each beat), of phospholamban (increasing the uptake ment of cardiac hypertrophy in vivo. PGF2a is an of Ca2+ into the sarcoplasmic reticulum), of troponin I exceedingly eective stimulus for cardiomyocyte hyper- (increasing the cross-bridge cycling rate), and of the trophy in NRVMs (Adams et al., 1996; Lai et al., ryanodine receptor (gating Ca2+ release from the SR). 1996). Notably, PGF2a is upregulated in the stressed Because the Gas-cAMP signaling pathway is rapidly myocardium (Lai et al., 1996) suggesting that it may activated and inactivated, it is well suited for also participate in the autocrine/paracrine growth instantaneous control of cardiovascular performance signaling response that underlie hypertrophy in the in response to demand.

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1630 Early studies suggested that cardiac hypertrophy, ated by pertussis toxin treatment (Sekiguchi et al., induced by chronic NE infusion, was attenuated by 1999). The notion that Gi signaling alone is not treatment with b-AdrR antagonists (Zierhut and sucient for cardiomyocyte hypertrophy is consistent Zimmer, 1989). Similarly, studies in Simpson's labora- with the ®nding that Gi linked M2 muscarinic tory demonstrated a modest hypertrophic eect of the cholinergic receptors did not induce hypertrophy b-adrenergic agonist, isoproterenol, in NRVMs (Simp- whereas a chimeric M2 muscarinic receptor that son, 1985). This growth response was suggested to be coupled to Gq was eective in this regard (Ramirez dependent upon increased myocyte contractile activity, et al., 1995). and due at least in part to the release of paracrine It is known that Gi signaling pathways regulate factors by non-muscle cells (for review see Long et al., activation of the ERK family of MAP kinases. This 1992). Recently, transgenic mice overexpressing Gasin occurs through eects of released bg subunits on PI3 the myocardium have been generated and extensively kinase and/or bARK, with subsequent activation of studied (Vatner et al., 2000; Iwase et al., 1996, 1997). downstream kinase cascades (Src, Ras, Raf-1). Nota- As expected, an increase in contractile sensitivity to b- bly, ERK MAP kinase activation itself is unlikely to be adrenergic stimulation was evident. Interestingly ana- a sucient stimulus for induction of hypertrophic lysis of Gas transgenic mice expressing the transgene in cardiomyocyte growth (Post et al., 1996). In addition, a chimeric fashion indicated that hypertrophy was not there is little evidence for a critical role for bg signaling evenly distributed among the Gas expressing cells but in cardiac hypertrophy. While bg subunits released was prominent in regions where Gas expressing cells from Gq were shown to be responsible for PI3 kinase were concentrated and lacking in regions where Gas activation in response to banding (Prasad et al., 2000), expressing cells were sparse (Vatner et al., 2000). This mice expressing the bARK C-terminal domain, which suggested that local areas of enhanced contraction or binds to and blocks the function of bg subunits, had a neurohumoral signaling were involved in the develop- normal hypertrophic response to pressure overload ment of the hypertrophic phenotype. The ®ndings from (Choi et al., 1997). this in vivo model of Gas overexpression are consistent We recently observed that an activated mutant form with the in vitro data suggesting that b-AdrR/Gs of Gai2 when overexpressed via adenoviral infection is induced hypertrophy may occur as a secondary sucient to induce hypertrophic growth of NRVMs response to the well known contractile eects mediated (Meszaros et al., manuscript in preparation). This via Gas signaling. could re¯ect more promiscuous coupling of the highly The b1-AdrR is the predominant cardiac subtype. b1- expressed Ga subunit to a tyrosine kinase pathway. It AdrR transgenic mice exhibit cardiomyocyte hypertro- is perhaps noteworthy that the eectiveness of Gi phy and increased contractile function (Engelhardt et pathways in regulating hypertrophy may be greater in al., 1999; Bisognano et al., 2000). In these mice, neonatal cardiomyocytes isolated from mouse (vs rat) hypertrophy progressed to a dilated cardiomyopathy ventricle. These cells show surprisingly poor induction with loss of ventricular function, myocyte degeneration of hypertrophy by exogenous ligands, but undergo and replacement ®brosis (Bisognano et al., 2000; spontaneous hypertrophy in culture, through a pertus- Engelhardt et al., 1999). In contrast to the striking sis toxin sensitive pathway (Deng et al., 2000 and eects observed in b1-AdrR transgenics b2-AdrR P Simpson, personal communication). expression lead to enhanced contractile function with- out signi®cant cardiomyocyte hypertrophy or patho- physiology, except at extraordinarily high levels of G-protein involvement in cardiomyocyte apoptosis and receptor expression (Milano et al., 1994b; Rockman et heart failure al., 1996; Liggett et al., 2000). These data indicate that b2-AdrR eects on cardiomyocyte contractile functions Cellular growth is characterized by up-regulation of do not necessarily lead to cardiomyocyte hypertrophy. genes and transcription factors that promote cell-cycle Dierences in b2- and b1-AdrR signaling pathways progression. Since mature cardiac myocytes cannot discussed below may account for these dierences. divide, it has been suggested that stimuli that promote cardiomyocyte growth may ultimately lead to apoptosis instead of cell division (Katz, 1995). This paradigm Gi signaling pathways in cardiac hypertrophy was apparent in studies in which forced overexpression There is limited evidence supporting a role for Gi of the transcription factor E2F-1 led not only to DNA signaling in cardiomyocyte hypertrophy. Some growth synthesis, but also to widespread apoptosis (Agah et factors that stimulate hypertrophy through receptors al., 1997) of cardiac myocytes. Therefore, although coupled to Gq also induce receptor coupling to Gi (see hypertrophy may serve as a short-term adaptive Table 1). Gi activation can contribute to the hyper- process to increase the contractile mass of the heart, trophic response to these agonists. For example, the growth signals responsible for hypertrophic growth inhibition of Gi signaling with pertussis toxin partially may eventually lead to cardiomyocyte apoptosis. inhibits ET-1 stimulated hypertrophy in cultured Cardiomyocyte apoptosis has been associated with neonatal rat ventricular myocytes (Hilal-Dandan et heart failure in several animal models including al., 1997). Similarly, sphingosylphosphorylcholine pressure overload, myocardial infarction, and hyper- (SPC) induced cardiomyocyte hypertrophy is attenu- trophic and idiopathic dilated cardiomyopathies. A

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1631 causal link between apoptotic myocyte death and the stable form of cardiac hypertrophy (D'Angelo et al., transition from compensatory hypertrophy to cardiac 1997). However, when these mice were subjected to the dysfunction and failure is rapidly gaining acceptance. neurohumoral or hemodynamic stresses associated with Studies using isolated cells or transgenic animals in parturition (Adams et al., 1998b) or to chronic pressure which GPCR signaling pathways are highly activated overload (Sakata et al., 1997; Adams et al., manuscript suggest that G-protein signaling can cause cardiomyo- in preparation) cardiac failure rapidly developed in cyte apoptosis. This would provide a plausible association with increased cardiomyocyte apoptosis. pathogenic link between elevated levels of neuro- Another line of transgenic mice expressing an activated hormones that activate G-protein coupled receptors, rather than wild type form of Gaq in the myocardium the altered expression of G-proteins and their eectors, rapidly developed marked hypertrophy and a dilated and the development of overt heart failure. Signi®- cardiomyopathy (Mende et al., 1998). Preliminary cantly, if cardiomyocyte apoptosis contributes to loss studies provide evidence for marked apoptosis in the of contractile cells and to cardiac failure in the hearts of these animals (U Mende and J Adams, in pathologically stressed heart, interventions limiting G- preparation). The pathways responsible for Gaq protein signaling should be eective strategies for induced apoptosis are under investigation. One path- preventing these pathophysiological events. way that has been implicated is the activation of the stress-activated kinases, p38 and JNK, which are enhanced in cells undergoing Gaq-induced apoptosis Gaq signaling in apoptosis and failure (Adams et al., 1998b). In addition, mitochondrial One of the ®rst lines of evidence for the involvement of permeability is increased in GaqQ209L expressing cells Gq signaling in cardiomyocyte apoptosis came from which represents a critical target for rescue of Gaq experiments carried out in Anversa's laboratory. In induced apoptosis (Adams et al., 2000). these studies Ang II was demonstrated to induce apoptosis in neonatal and adult myocytes via PKC Gas signaling in apoptosis and failure and calcium dependent pathways (Kajstura et al., 1997; Cigola et al., 1997). The AT1 receptor antagonist A number of studies have documented that catechola- losartan blocked Ang II-induced apoptosis, while an mines are elevated in the plasma of heart failure AT2 receptor blocker was ineective. Interestingly, patients (Hasking et al., 1986). An association between losartan also blocked stretch-induced apoptosis in activation of the sympathetic nervous system, elevation cultured adult cardiomyocytes (Leri et al., 1998) of circulating catecholamines, and the development of suggesting that the Gq-coupled AT1 receptor mediates heart failure is well documented. Chronically elevated activation of apoptotic pathways by mechanical levels of Gas signaling, as would be imposed by stretch. Several studies have shown that ACE inhibi- increased catecholamines, leads to a gradual decrease tors or AT receptor antagonists also block the in responsiveness (desensitization) to inotropic stimula- cardiomyocyte apoptosis observed in spontaneously tion by b-AdrR agonists. The mechanisms involved in hypertensive rats (Li et al., 1997; Goussev et al., 1998; b-AdrR desensitization include a decrease in b1-AdrR FortunÄ o et al., 1998). receptor number, an uncoupling of the b-AdrR from While the initial hypertrophic response to Gq- Gas, an increase in Gai, a decrease in adenylyl cyclase coupled receptor stimulation in the stressed heart is activity, and an increase in the receptor kinase, adaptive, we have demonstrated that prolonged (bARK), which phosphorylates and down-regulates activation of these pathways may become maladaptive the b-AdrR. The global loss in b-AdrR sensitivity as a result of cumulative loss of myocytes through imposed at these various levels was thought for many apoptosis (Adams et al., 1998b; Dorn and Brown, years to be the cause of cardiac failure. On the other 1999). Direct evidence for the activation of apoptotic hand, the down regulation of the b-AdrR pathway pathways by Gq signaling derives from in vitro studies could be viewed as an adaptation to, rather than a using recombinant adenovirus to express a constitu- cause of, heart failure. In this regard, ameliorative tively activated mutant form of the alpha subunit of eects of beta (b-AdrR) blockers in animal models of Gq (GaqQ209L). These studies showed that sustained heart failure and in clinical trials with heart failure activation of Gaq and concomitant stimulation of PLC patients are now well documented (for review see induced marked apoptosis of cultured neonatal rat Sabbah, 1999). The possibility that excessive b-AdrR ventricular myocytes (Adams et al., 1998b). More signaling predisposes to heart failure provides a new recently we demonstrated that addition of Gq-coupled focal point for studying the role of b-AdrR signaling in receptor agonists (PGF2a or PE) to hypertrophied models of heart failure. cardiomyocytes overexpressing wild type Gaq caused NE was shown to induce apoptosis in isolated adult robust PLC activation, comparable to that seen with rat ventricular cardiomyocytes, through stimulation of expression of GaqQ209L, and likewise induced apop- the b-AdrR (Communal et al., 1998). In a subsequent tosis (Adams et al., 2000). These paradigms were study apoptosis was shown to be mediated through recapitulated in transgenic mice generated in Dorn's the b1-AdrR, via eects on cyclic AMP (Communal et laboratory. Initial studies with these transgenic mice al., 1999). Zaugg et al. (2000) also demonstrated that showed that a modest (four-fold) overexpression of it is the b1-AdrR which mediates the apoptotic Gaq in the myocardium lead to development of a response to NE. This ®nding is particularly relevant

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1632 in light of the observation that transgenic animals Gai signaling in cardiac apoptosis and failure expressing b1-AdrR, even at relatively low levels, develop heart failure associated with apoptosis (En- Increased expression of Gai protein (speci®cally Gai2) gelhardt et al., 1999; Bisognano et al., 2000). The is observed in models of dilated cardiomyopathy notion that increased Gs signaling can lead to (Bohm et al., 1990, 1994). Cardiomyocytes isolated apoptosis is further supported by studies demonstrat- from failing hearts show depressed inotropic responses ing that heart failure and apoptosis develop in older to isoproterenol, which can be restored by inactivating transgenic mice overexpressing the alpha subunit of Gs Gi with pertussis toxin (Brown and Harding, 1992). in the myocardium (Geng et al., 1999). When These ®ndings suggested a causal relationship between activation of endogenous b-AdrR in Gas transgenic increased Gai expression and the decreased contractile mice was blocked with propranolol, cardiomyocyte function associated with heart failure. More recently apoptosis and cardiac dysfunction were prevented Conklin's group demonstrated that myocardial expres- (Akai et al., 1999). These studies suggest that chronic sion of a conditionally activated Gai-linked receptor sympathetic stimulation over an extended period of resulted not only in decreased heart rate and time can contribute to the development of apoptotic contractility but in development of a lethal dilated heart failure. cardiomyopathy (Redman et al., 2000), mirroring the In contrast to what is observed for b1-AdrR or Gs physiological response associated with increased Gi overepression, b2 AdrR expression results in the expression. There is no data indicating that increased development of cardiomyopathy only when receptor Gi expression causes heart failure through induction of levels are extremely high (Liggett et al., 2000). It has apoptosis. In fact, as discussed above, increased Gi

been suggested that the b27-AdrR couples not only to expression might be expected to decrease apoptotic Gs but also to Gi. The b2-AdrR has been shown to signaling, e.g. via protective eects of Gi signaling on functionally antagonize b1-AdrR responses through a activation of p38 kinase or PI3 kinase. Thus, while PTX-sensitive pathway (Xiao et al., 1995). Likewise, changes in Ga1 function can clearly aect both b2-AdrR coupling to a PTX sensitive protein (pre- hypertrophic and apoptotic responses, the role of Gi sumably Gi) also acts to suppress apoptotic responses signaling in cardiomyocyte growth appears at present to b-AdrR stimulation (Communal et al., 1999) or to to be largely modulatory. ischemic stress (Chesley et al., 2000). These data suggest that activation of b2-AdrR could serve an important function by limiting the apoptotic response Abbreviations to, and development of heart failure following ischemia a-AR, a-AdrR: a-adrenergic receptor; b-AR, b-AdrR: b- or prolonged b-AdrR stimulation. adrenergic receptor; b-ARK; b-adrenergic receptor kinase; The mechanism of b1-AdrR induced apoptosis is a-MHC, b-MHC: a-myosin heavy chain, b-myosin heavy under investigation. A pathway requiring protein chain; ACE: angiotensin-converting enzyme; ANF: atrial kinase A was implicated in the apoptosis induced by natriuretic factor; Ang II: angiotensin II; cAMP: cyclic chronic infusion of isoproterenol in vivo (Shizukuda et AMP; ET-1: endothelin-1; GAP: GTPase activator protein; al., 1998). In other studies isoproterenol-induced GPCR: G protein-coupled receptor; ISO: isoproterenol; cardiomyocyte apoptosis was demonstrated to be LPA: lysophosphatidic acid; MAP kinase: mitogen-acti- mediated by a calcium-calcineurin pathway (Saito et vated protein kinase; MLC-2: myosin light chain-2; NE: 2+ norepinephrine; NRVM: neonatal rat ventricular myocyte; al., 2000). Thus, cyclic AMP and Ca would appear to PAR: protein-activated receptor; PE: phenylephrine; be second messengers in these pathways. Interestingly, PGF2a: prostaglandin F2a;PIP2: phosphatidylinositol the inhibitory eect of b2-AdrR and Gi signaling does bisphosphate; PKA: cAMP-dependent protein kinase; not appear to be mediated by decreasing cyclic AMP PKC: protein kinase C; PLC: phospholipase C; RACKS: but instead via p38 MAP kinase (Communal et al., receptors for activated C-kinases; RGS4: regulator of G 2000) or PI3 kinase (Chesley et al., 2000). protein signaling 4; SPC: sphingosylphosphorylcholine.

References

Adams JW, Migita DS, Yu MK, Young R, Hellickson MS, Akai K, Yang G-P, Geng Y-J, Takagi G, Bishop S, Ishikawa Castro-Vargas FE, Domingo JD, Lee PH, Bui JS and Y, Shannon RP, Wagner TE, Vatner DE, Homcy CJ and Henderson SA. (1996). J. Biol. Chem., 271, 1179 ± 1186. Vatner SF. (1999). J. Clin. Invest., 104, 551 ± 558. Adams JW, Pagel AL, Means CK, Oksenberg D, Armstrong Akhter SA, Luttrell LM, Rockman HA, Iaccarino G, RC and Brown JH. (2000). Circ. Res., 87, 1180 ± 1187. Lefkowitz RJ and Koch WJ. (1998). Science, 280, 574 ± Adams JW, Sah VP, Henderson SA and Brown JH. (1998a). 577. Circ. Res., 83, 167 ± 178. Akhter SA, Milano CA, Shotwell KF, Cho M-C, Rockman Adams JW, Sakata Y, Davis MG, Sah VP, Wang Y, Liggett HA, Lefkowitz RJ and Koch WJ. (1997). J. Biol. Chem., SB, Chien KR, Brown JH and Dorn GW. (1998b). Proc. 272, 21253 ± 21259. Natl. Acad. Sci. USA, 95, 10140 ± 10145. Bishopric NH, Simpson PC and Ordahl CP. (1987). J. Clin. Agah R, Kirshenbaum LA, Abdellatif M, Truong LD, Invest., 80, 1194 ± 1199. Chakraborty S, Michael LH and Schneider MD. (1997). J. Clin. Invest., 100, 2722 ± 2728.

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1633 Bisognano JD, Weinberger HD, Bohlmeyer TJ, Pende A, Hilal-Dandan R, Ramirez MT, Villegas S, Gonzalez A, Raynolds MV, Sastravaha A, Roden R, Asano K, Blaxall Endo-Mochizuki Y, Brown JH and Brunton LL. (1997). BC, Wu SC, Communal C, Singh K, Colucci W, Bristow Am. J. Physiol., 272, H130 ± H137. MR and Port JD. (2000). J. Mol. Cell. Cardiol., 32, 817 ± Hilal-Dandan R, Urasawa K and Brunton LL. (1992). J. 830. Biol. Chem., 267, 10620 ± 10624. Bohm M, Eschenhagen T, Gierschik P, Larisch K, Lensche ItoH,HirataY,AdachiS,TanakaM,TsujinoM,KoikeA, H, Mende U, Schmitz W, Schnabel P, Scholz H, Steinfath Nogami A, Murumo F and Hiroe M. (1993). J. Clin. M and Erdmann E. (1994). J. Mol. Cell. Cardiol., 26, 133 ± Invest., 92, 398 ± 403. 149. ItoH,HirataY,HiroeM,TsujinoM,AdachiS,Takamoto Bohm M, Gierschik P, Jakobs KH, Pieske B, Schnabel P, T, Nitta M, Taniguchi K and Marumo F. (1991). Circ. Ungerer M and Erdmann E. (1990). Circulation, 82, Res., 69, 209 ± 215. 1249 ± 1265. IwaseM,BishopSP,UechiM,VatnerDE,ShannonRP, Brown LA and Harding SE. (1992). Br.J.Pharmacol.,106, Kudej RK, Wight DC, Wagner TE, Ishikawa Y, Homcy 115 ± 122. CJ and Vatner SF. (1996). Circ. Res., 78, 517 ± 524. Chesley A, Lundberg MS, Asai T, Xiao RP, Ohtani S, IwaseM,UechiM,VatnerDE,AsaiK,ShannonRP,Kudej Lakatta EG and Crow MT. (2000). Circ. Res., 87, 1172 ± RK, Wagner TE, Wight DC, Patrick TA, Ishikawa Y, 1179. Homcy CJ and Vatner SF. (1997). Am. J. Physiol., 272, Chien KR, Knowlton KU, Zhu H and Chien S. (1991). H585 ± H589. FASEB J., 5, 3037 ± 3046. Izumo S and Shioi T. (1998). J. Card. Fail., 4, 263 ± 270. Choi D-J, Koch WJ, Hunter JJ and Rockman HA. (1997). J. Jalili T, Takeishi Y and Walsh RA. (1999). Cardiovasc Res., Biol. Chem., 272, 17223 ± 17229. 44, 5±9. Cigola E, Kajstura J, Li B, Meggs LG and Anversa P. (1997). Kajstura J, Cigola E, Malhotra A, Li P, Cheng W, Meggs LG Exp. Cell. Res., 231, 363 ± 371. and Anversa P. (1997). J. Mol. Cell. Cardiol., 29, 859 ± Claycomb WC, Lanson Jr NA, Stallworth BS, Egeland DB, 870. Delcarpio JB, Bahinski A and Izzo Jr NJ. (1998). Proc. Kariya K, Karns LR and Simpson PC. (1991). J. Biol. Chem., Natl. Acad. Sci. USA, 95, 2979 ± 2984. 266, 10023 ± 10026. Communal C, Colucci WS and Singh K. (2000). J. Biol. Karns LR, Kariya K and Simpson PC. (1995). J. Biol. Chem., Chem., 275, 19395 ± 19400. 270, 410 ± 417. Communal C, Singh K, Pimentel DR and Colucci WS. Katz AM. (1995). Eur. Heart J., 16, 110 ± 114. (1998). Circulation, 98, 1329 ± 1334. Knowlton KU, Michel MC, Itani M, Shubeita HE, Ishihara Communal C, Singh K, Sawyer DB and Colucci WS. (1999). K, Brown JH and Chien KR. (1993). J. Biol. Chem., 268, Circulation, 100, 2210 ± 2212. 15374 ± 15380. D'Angelo DD, Sakata Y, Lorenz JN, Boivin GP, Walsh RA, Komuro I, Katoh Y, Kaida T, Shibazaki Y, Kurabayashi M, Liggett SB and Dorn GW. (1997). Proc. Natl. Acad. Sci. Hoh E, Takaku F and Yazaki Y. (1991). J. Biol. Chem., USA, 94, 8121 ± 8126. 266, 1265 ± 1268. Delcarpio JB, Lanson Jr NA, Field LJ and Claycomb WC. Kudoh S, Komuro I, Hiro Y, Zou Y, Harada K, Sugaya T, (1991). Circ. Res., 69, 1591 ± 1600. Takekoshi N, Murakami K, Kadowaki T and Yazaki Y. Deng X-F, Rokosh DG and Simpson PC. (2000). Circ. Res., (1998). J. Biol. Chem., 273, 24037 ± 24043. 87, 781 ± 788. LaiJ,JinH,YangR,WinerJ,LiW,YenR,KingKL, Dorn GW and Brown JH. (1999). Trends Cardiovasc. Med., Zeigler F, Ko A, Cheng J, Bunting S and Paoni NF. 9, 26 ± 34. (1996). Am. J. Physiol., 271, H2197 ± H2208. DunnmonPM,IwakiK,HendersonSA,SenAandChien LaMorte VJ, Thorburn J, Absher D, Spiegel AM, Brown JH, KR. (1990). J. Mol. Cell. Cardiol., 22, 901 ± 910. Chien KR, Feramisco JR and Knowlton KU. (1994). J. Ehmke H, Faulhaber J, Munter K, Kirchengast M and Biol. Chem., 269, 13490 ± 13496. Wiesner RJ. (1999). Hypertension, 33, 954 ± 960. Leri A, Claudio PP, Li Q, Wang X, Reiss K, Wang S, Engelhardt S, Hein L, Wiesmann F and Lohse MJ. (1999). Malhotra A, Kajstura J and Anversa P. (1998). J. Clin. Proc. Natl. Acad. Sci. USA, 96, 7059 ± 7064. Invest., 101, 1326 ± 1342. FortunÄ o MA, Ravassa S, Etayo JC and Diez J. (1998). Li Z, Bing OH, Long X, Robinson KG and Lakatta EG. Hypertension, 32, 280 ± 286. (1997). Am. J. Physiol., 272, H2313 ± H2319. Geng Y-J, Ishikawa Y, Vatner DE, Wagner TE, Bishop SP, Liggett SB, Tepe NM, Lorenz JN, Canning AM, Jantz TD, Vatner SF and Homcy CJ. (1999). Circ. Res., 84, 34 ± 42. Mitarai S, Yatani A and Dorn GW. (2000). Circulation, Glembotski CC, Irons CE, Krown KA, Murray SF, Sprenkle 101, 1707 ± 1714. AB and Sei CA. (1993). J. Biol. Chem., 268, 20646 ± 20652. Linz W and Scholkens BA. (1992). J. Cardiovasc. Pharma- Goetzl EJ, Lee H, Azuma T, Stossel TP, Turck CW and col., 20, S83±S90. Karliner JS. (2000). J. Biol. Chem., 275, 14573 ± 14578. Long CS, Kariya K, Karns LR and Simpson PC. (1992). GoussevA,SharovVG,ShimoyamaH,TanimuraM,Lesch Basic Res. Cardiol., 87, 19 ± 31. M, Goldstein S and Sabbah HN. (1998). Am.J.Physiol., Matkovich SJ and Woodcock EA. (2000). J. Biol. Chem., 275, H626 ± H631. 275, 10845 ± 10850. Grupp IL, Lorenz JN, Walsh RA, Boivin GP and Rindt H. McWhinneyC,WenhamD,KanwalS,KalmanV,HansenC (1998). Am. J. Physiol., 275, H1338 ± H1350. and Robishaw JD. (1999). J. Biol. Chem., 275, 2087 ± 2097. Harada K, Komuro I, Zou Y, Kudoh S, Kijima K, Mende U, Kagen A, Cohen A, Aramburu J, Schoen FJ and Matsubara H, Sugaya T, Murakami K and Yazaki Y. Neer EJ. (1998). Proc. Natl. Acad. Sci. USA, 95, 13893 ± (1998). Circ. Res., 82, 779 ± 785. 13898. Hasking GJ, Esler MD, Jennings GL, Burton D, Johns JA and Korner PI. (1986). Circulation, 73, 615 ± 621.

Oncogene G proteins and cardiac growth JW Adams and JH Brown 1634 MilanoCA,AllenLF,RockmanHA,DolberPC,McMinn Sekiguchi K, Yokoyama T, Kurabayashi M, Okajima F and TR, Chien KR, Johnson TD, Bond RA and Lefkowitz RJ. Nagai R. (1999). Circ. Res., 85, 1000 ± 1008. (1994a). Science, 264, 5582 ± 5586. Shizukuda Y, Buttrick PM, Geenen DL, Borczuk AC, Kitsis Milano CA, Dolber PC, Rockman HA, Bond RA, Venable RN and Sonnenblick EH. (1998). Am.J.Physiol.,275, ME, Allen LF and Lefkowitz RJ. (1994b). Proc. Natl. H961 ± H968. Acad. Sci. USA., 91, 10109 ± 10113. Shubeita HE, Martinson EA, Van Bilsen M, Chien KR and Minamino T, Yujiri T, Papst PJ, Chan ED, Johnson GL and Brown JH. (1992). Proc. Natl. Acad. Sci. USA, 89, 1305 ± Terada N. (1999). Proc. Natl. Acad. Sci. USA., 96, 15127 ± 1309. 15132. Shubeita HE, McDonough PM, Harris AN, Knowlton KU, Miyauchi T and Goto K. (1999). Trends Pharmacol. Sci., 20, Glembotski CC, Brown JH and Chien KR. (1990). J. Biol. 210 ± 217. Chem., 265, 20555 ± 20562. Mochly-Rosen D, Wu G, Hahn H, Osinska H, Liron T, Simpson P. (1983). J. Clin. Invest., 72, 732 ± 738. Lorenz JN, Yatani A, Robbins J and Dorn GW. (2000). Simpson P. (1985). Circ. Res., 56, 884 ± 894. Circ. Res., 86, 1173 ± 1179. Simpson P, McGrath A and Savion S. (1982). Circ. Res., 51, Paradis P, Dali-Youcel N, Paradis FW, Thibault G and 787 ± 801. Nemer M. (2000). Proc. Natl. Acad. Sci. USA, 97, 931 ± Simpson PC. (1999). Circulation, 99, 334 ± 337. 936. Simpson PC, Kariya K, Karns LR, Long CS and Karliner JS. Parker TG, Packer SE and Schneider MD. (1990). J. Clin. (1991). Mol. Cell. Biol., 12, 35 ± 43. Invest., 85, 507 ± 514. Steinberg SF, Goldberg M and Rybin VO. (1995). J. Mol. Post GR, Goldstein D, Thuerauf D, Glembotski CC and Cell. Cardiol., 27, 141 ± 153. Brown JH. (1996). J. Biol. Chem., 271, 8452 ± 8457. Sugden PH and Clerk A. (1998). Circ. Res., 83, 345 ± 352. Prasad SVN, Esposito G, Mao L, Koch WJ and Rockman Takeishi Y, Ping P, Bolli R, Kirkpatrick DL, Hoit BD and HA. (2000). J. Biol. Chem., 275, 4693 ± 4698. Walsh RA. (2000). Circ. Res., 86, 1218 ± 1223. Puceat M and Vassort G. (1996). Mol. Cell. Biol., 157, 65 ± Tamamori M, Ito H, Adachi S, Akimoto H, Marumo F and 72. Hiroe M. (1996). J. Clin. Invest., 97, 366 ± 372. Ramirez MT, Post GR, Sulakhe PV and Brown JH. (1995). Theroux TL, Esbenshade TA, Peavy RD and Minneman KP. J. Biol. Chem., 270, 8446 ± 8451. (1996). Mol. Pharmacol., 50, 1376 ± 1387. Redman CM, Degtyarev MY, Kwa AT, Salomonis N, Cotte Thurmann PA, Kenedi P, Schmidt A, Harder S and N, Nanevicz T, Fidelman N, Desai K, Vranizan K, Lee Rietbrock N. (1998). Circulation, 98, 2037 ± 2042. EK, Coward P, Shah N, Warrington JA, Fishman GI, Varma DR and Deng X. (2000). Can. J. Physiol. Pharmacol., Bernstein D, Baker AJ and Conklin BR. (2000). Proc. 78, 267 ± 292. Natl. Acad. Sci. USA, 97, 4826 ± 4831. VatnerDE,YangG-P,GengY-J,AsaiK,YunJS,Wagner Rockman HA, Hamilton RA, Jones LR, Milano CA, Mao L TE, Ishikawa Y, Bishop SP, Homcy CJ and Vatner SF. and Lefkowitz RJ. (1996). J. Clin. Invest., 97, 1618 ± 1623. (2000). Circ. Res., 86, 802 ± 806. RogersJH,TamirisaP,KovacsA,WeinheimerC,Courtois Xiao R-P, Ji X and Lakatta EG. (1995). Mol. Pharmacol., 47, M, Blumer KJ, Kelly DP and Muslin AJ. (1999). J. Clin. 322 ± 329. Invest., 104, 567 ± 576. Yamazaki T, Komuro I, Kudoh S, Zou Y, Shiojima I, Sabbah HN. (1999). Clin. Cardiol., 22, V16 ± V20. MizunoT,TakanoH,HiroiY,UekiK,TobeK, Sabri A, Muske G, Zhang H, Pak E, Darrow A, Andrade- Kadowaki T, Nagai R and Yazaki Y. (1995). J. Clin. Gordon P and Steinberg SF. (2000). Circ. Res., 86, 1054 ± Invest., 96, 438 ± 446. 1061. Yamazaki T, Komuro I, Zou Y and Yazaki Y. (1999). Sadoshima J-I and Izumo S. (1993a). Circ. Res., 73, 424 ± Hypertens. Res., 22, 113 ± 119. 438. ZauggM,XuW,LucchinettiE,Sha®qSA,JamaliNZand Sadoshima J-I and Izumo S. (1993b). Circ. Res., 73, 413 ± Siddiqui MAQ. (2000). Circulation, 102, 344 ± 350. 423. ZhuYC,ZhuYZ,GohlkeP,StaussHMandUngerT. Saito S, Hiroi Y, Zou Y, Aikawa R, Haruhiro T, Sibasaki F, (1997). Am.J.Cardiol.,80, 110A ± 117A. Yazaki Y, Nagai R and Komuro I. (2000). J. Biol. Chem., Zierhut W and Zimmer HG. (1989). Circ. Res., 65, 1417 ± 275, 34528 ± 34533. 1425. Sakata Y, Hoit BD, Liggett SB, Walsh RA and Dorn GW. (1997). Circulation, 97, 1488 ± 1495.

Oncogene