Chapter 2. Review of the literature
- Table of Contents
- 2.1. Regulation of cardiac contractile function
- 2.2. Autocrine/paracrine factors
- 2.3. Changes in cardiac gene expression and structure in response to increased load
- 2.4. Natriuretic peptide system
- 2.5. Cardiac extracellular matrix
- 2.6. Genetically engineered animal models in cardiovascular research
2.1. Regulation of cardiac contractile function
Contractile function of the heart is regulated by a number of intrinsic and extrinsic mechanisms. The impact of autonomic nervous system, various hormones, such as thyroid hormone, adrenocortical steroids, insulin, glucagon, and blood concentrations of O2, CO2 and H+ on cardiac contractile function has been well established (See e.g. Berne & Levy 1993). Also autocrine/paracrine effectors synthesized and secreted by endothelial cells (EC), fibroblasts or cardiomyocytes themselves have been demonstrated to possess the ability to affect cardiac contractility. Examples of such regulators are ET-1 (Kelly et al. 1990), AM (Szokodi et al. 1998), natriuretic peptides (Yamamoto et al. 1997), nitric oxide (NO) (Prendergast et al. 1997b) and Ang II (Li et al. 1994). Intrinsic mechanisms affecting cardiac function include the Frank-Starling mechanism and the force-frequency relation. The complex interplay between all these factors is occurring continuously via both the hemodynamic state and respective feedback mechanisms, and also at the level of single cardiomyocytes. The changes in cardiac function can also be divided based on the time scale of occurrence. Acutely, within a few minutes after stimuli, changes due to posttranslational modification of proteins, such as phosphorylation, can be noted in contractile and secretory function of the heart, while the structural changes occur during a longer period as a result of altered gene expression and protein synthesis.
2.1.1. Excitation-contraction coupling
The excitation-contraction coupling (E-C coupling) includes the events which follow the wave of excitation and lead to contraction. Initially, the wave of depolarization spreads rapidly along the myocardial sarcolemma, and also into the interior of the cells via the invaginations of the sarcolemma, the T-tubules, opening the voltage dependent L-type Ca2+ channels and triggering a Ca2+ influx (Hobai & Levi 1999).
2.1.1.1. Ca2+ influx leading to contraction
The calcium entering the cell through the L-type Ca2+ channels serves as a trigger to release Ca2+ (Ca2+ induced Ca2+ release, CICR) from the sarcoplasmic reticulum (SR) through SR Ca2+ release channels known as ryanodine receptors (RyR) (Fabiato & Fabiato 1979). The RyR and L-type Ca2+ channels are located in close functional association, thus allowing rapid CICR to occur (Sham et al. 1995).
The cytosolic free Ca2+ is increased 10- to 100-fold during the E-C coupling process. High intracellular calcium concentration ([Ca2+]i) levels promote Ca2+ binding to specific site in the N-terminal domain of troponin C (TnC), resulting in a conformational change of the TnC molecule (Robertson et al. 1982, for review, see Solaro & Rarick 1998). Cardiac troponin is a heterotrimer consisting of three distinct gene products: TnC, troponin I (TnI) and troponin T (TnT). TnC acts as the Ca2+ receptor, TnI inhibits the actin-myosin reaction and shuttles between tight binding to actin and tight binding to Ca2+-TnC and TnT binds to myosin, TnI, and TnC. As a consequence of the Ca2+-signaling process and the conformational change in TnC, TnI moves from its diastolic state (tightly bound to actin) to its systolic state (tightly bound to TnC) (Tao et al. 1990, Solaro & Rarick 1998). The interaction between TnI and TnC is followed by moving of the tropomyosin molecule to allow the crossbridges to attach and to produce force (Opie 1995). Heads of myosins (the crossbridges or myosin) protruding from the thick filament then react with thin-filament actins in a reaction cycle that is powered by ATP (Rayment et al. 1993).
2.1.1.2. Factors affecting the excitation-contraction coupling
A number of factors influence the E-C coupling process. Extracellular mediators, such as ET-1, AM, Ang II, NO and catecholamines, regulate the process by activating the intracellular second messengers. Depending on the agonist, the contractile force may increase or decrease, i.e. there may be a positive or negative inotropic effect, respectively. In terms of Ca2+-contractile protein interaction, in order to a positive inotropic effect to occur, either the supply of the Ca2+ during systole must increase, or the sensitivity of the TnC for Ca2+ must be elevated, which means that the response of the myofilaments at a given level of occupancy of Ca2+ binding sites is increased (Endoh 1998, Opie 1995). The majority of the inotropic interventions (e.g. the force-frequency relationship, β -adrenergic agonists and digitalis glycosides) alter the intracellular Ca2+ transient, thus acting through an upstream mechanism to increase the contractile force. The Frank-Starling mechanism, α-adrenergic agonists, ET-1 and some novel drugs, such as EMD 57033 and levosimendan, act through a downstream mechanism by increasing the sensitivity to Ca2+ (Krämer et al. 1991, Haeusler et al. 1997, Kentish & Wrzosek 1998) (for review, see Haikala & Linden 1995).
Intracellular signaling in response to agonist stimuli is mediated by a number of second messengers. Well characterized 3´,5´-cyclic adenosine monophosphate (cAMP) and 3´,5´-cyclic guanosine monophosphate (cGMP) mediate positive and negative inotropic responses, respectively. cAMP is generated by adenylyl cyclase (AC), which is coupled to sarcolemmal receptors, e.g. β -adrenergic receptor (β -AR) (Hajjar et al. 1998). cAMP then activates protein kinase A (PKA), which can phosphorylate e.g. L-type Ca2+ channel, phospholamban (PLB) and TnI (for review, see e.g. Walsh & Van Patten 1994, Katz & Lorell 2000). By phosphorylating TnI, PKA enhances the interaction between TnI and actin, thus decreasing the sensitivity of contractile apparatus to Ca2+, but also potentially increasing the rate of relaxation (Venema & Kuo 1993). However, the potential negative inotropic effect induced by TnI phosphorylation is normally overcome by a marked increase in [Ca2+]i due to stimulation of Ca2+ influx through L-type Ca2+ channels, as occurs in response to a β -receptor agonist.
Several independent signals affect cardiac function via the guanine nucleotide binding protein (G-protein) coupled receptors. The heterotrimeric G-proteins consist of separate Gα and Gβ γ subunits. Agonist binding to membrane bound G-protein coupled receptors catalyzes the exchange of guanosine diphosphate to guanosine triphosphate GTP on Gα subunit and subsequent dissociation of Gα from Gβ γ (for review, see Molkentin & Dorn II 2001). The Gα subunit is considered to mediate the majority of the downstream effects, but Gβ γ may also have an impact on downstream signaling through mitogen activated protein (MAP) kinases (Crespo et al. 1994). The cardiovascular G-protein coupled receptors couple to the three major classes of G-proteins, as divided by the alpha subunit: Gαs, Gαi and Gαq. Classically, Gαs mediates AC activation in response to ß-AR stimulation, Gαi mediates cholinergic inhibition of AC and Gαq has been implicated in LVH development (Molkentin & Dorn II 2001). Activation of Gq for instance by ET-1 induces phosphoinositide hydrolysis by phospholipase C (PLC). The second messengers inositol-1,4,5-triphosphate (IP3) and 1,2-diacylglycerol (DAG) induce subsequent activation of protein kinase C (PKC) and downstream effectors, such as the Na+-H+ exchanger (NHE) (Wang et al. 1993).
2.1.1.3. Removal of Ca2+ from cytoplasm during diastole
During the diastole, for relaxation and ventricular filling to occur, the Ca2+ that activated the myofilaments must be removed from the cytosol. Ca2+ is extruded from the cytoplasm via sarcoplasmic reticulum Ca2+-ATPase (SERCA), sarcolemmal Na+-Ca2+ exchanger (NCX), PMCA, and mitochondrial Ca2+ uniporter (for review, see Bers 2000). Quantitatively, SERCA and NCX are most important. In rat and mice ventricles, SERCA accounts for over 90% of the Ca2+ removal during cardiac relaxation (Hove-Madsen & Bers 1993, Li et al. 1998), while in human and rabbit ventricles SERCA removes ca. 70% of the Ca2+ from the cytosol and the NCX ca. 28%. The rest of the Ca2+ is removed by PMCA and mitochondrial Ca2+ uniporter (Pieske et al. 1999b, Bers 2000) (see Fig.1). Thus, most of the Ca2+ that activates the contractile process is released from the SR, and the SR takes up most of the released Ca2+ again during diastole. PLB is a 52-amino acid phosphoprotein found in the SR membranes also in cardiomyocytes. It binds to the SERCA, inhibiting the Ca2+ binding ability. The PLB binding to SERCA is decreased via phosphorylation in response to certain stimuli, such as β -adrenergic signaling (for review, see Kiriazis & Kranias 2000). In failing hearts, the Ca2+ loading of the SR may be impaired (see section 2.2.), increasing the role of extracellular Ca2+ in EC-coupling and the role of NCX in Ca2+ transients (Pieske et al. 1999b). This may be partially responsible for the slowing down of the relaxation process as seen in heart failure (Kiriazis & Kranias 2000).
2.1.1.4. The role of the plasma membrane Ca2+-ATPase in heart
PMCA is a ubiquitous Ca2+-transporting enzyme extruding Ca2+ from the cell (Schatzmann 1966) (for review, see Carafoli 1992). As mentioned, in excitable cells expressing the high capacity NCX, the activity of PMCA in vitro is rather low compared with NCX (Bers 2000). In the myocardium, the expression of the PMCA isoforms 1, 2, and 4 has been shown (Stauffer et al. 1995, Hammes et al. 1994, for review, see Carafoli & Stauffer 1994), but the physiological significance has remained unknown. Due to the high affinity to Ca2+, PMCA has been suggested to play a role in fine-tuning Ca2+ in the final phase of diastole in the heart (for review, see Carafoli 1994).
PMCA is known to localize in caveolae, 50- to 100-nm plasma membrane invaginations, containing receptors for ET-1 and various other ligands. Also a number of important signaling molecules, such as Gαs, ras, PKCα, MAP kinase, AC and Src tyrosine kinase are enriched in caveolae (Fujimoto 1993, Chun et al. 1994, Hammes et al. 1998) (for reviews, see e.g. Couet et al. 1997, Smart et al. 1999). PMCA has been suggested to play a role in growth and differentiation processes in myoblasts as well as in other cell types in vitro (Hammes et al. 1996). Altered growth and differentiation responses to phenylephrine and isoproterenol were found in PMCA overexpressing neonatal cardiac myocytes in vitro (Hammes et al. 1998).
The finding that cardiac overexpression of PMCA resulted in no differences in voltage dependence, activation, and inactivation behavior of L-type Ca2+ current between TG cells and control adult cardiomyocytes confirmed the previous hypothesis that the significance of PMCA in Ca2+ extrusion is minor. Only when the SR was blocked by thapsigargin (SERCA inhibitor) and ryanodine (blocks the RyRs), a marginally different time constant of [Ca2+]i decline was seen (Hammes et al. 1998). Thus, the role of PMCA in cardiac myocytes has remained obscure.
2.1.2. The Frank-Starling mechanism
In 1895 Frank discovered that the greater the preload, the greater the force generated by frog cardiac muscle. In 1914 Starling demonstrated the same phenomenon in canine heart-lung preparation by elevating either right atrial pressure or aortic resistance (see e.g. Berne & Levy 1993, Katz & Lorell 2000).
The Frank-Starling mechanism (heterometric autoregulation) plays a major role in intrinsic regulation of cardiac function (Sarnoff & Berglund 1954; for review, see Katz & Lorell 2000). The role of the Frank-Starling response is augmented in the elderly, who have a dimished increase in the heart rate in response to physical excercise. It is also known that this response is preserved even in hypertrophied and failing hearts (Holubarsch et al. 1996). In normal subjects, the Frank-Starling response contributes to cardiac output during submaximal exercise (Plotnick et al. 1986), and changes in posture (Drake-Holland et al. 1990). An increase in ventricular end-diastolic volume, produced by increased venous return or decreased aortic outflow, leads immediately to a more powerful contraction. At the molecular basis, the mechanism of this phenomenon is not well understood. The main theory of the cellular basis of the Frank-Starling law has for long been length-dependent myofilament activation (Allen & Kentish 1985). The length dependence of myofilament activation is very prominent in normal hearts, operating at sarcomere lengths less than the optimal 2.2 µm (Solaro & Rarick 1998). The length-dependent activation has been suggested to relate to increased Ca2+ affinity of the Ca2+ -binding part of the contractile element, TnC (Kentish et al. 1986). A possible mechanism is that the change in sarcomere length involves a change in interfilament spacing that modulates the ability of crossbridges to react with thin filaments (actin) at the same Ca2+ concentration, thus increasing the rate of crossbridge formation, as suggested by studies using osmotic compression of the cardiomyocytes (McDonald & Moss 1995). However, in a recent study with x-ray diffraction analysis, the osmotic compression to achieve lattice spacing typical of longer length could not produce a change in Ca2+ sensitivity of force (Konhilas et al. 2002). Other possible cellular mechanisms explaining the Frank-Starling relationship include positive cooperativity in crossbridge binding, or strain of titin, elastic protein of the contractile element (Fitzsimons et al. 2001, Cazorla et al. 2001).
After the rapid increase in contractile force, there is a further increase in myocardial performance during the next few minutes of stretch. In vivo this allows the end-diastolic volume to return toward its original value (von Anrep 1912, Parmley & Chuck 1973). This slow rise in contractile strength, also known as Anrep effect or homeometric autoregulation, accounts for < 10% to 25% of the overall contractile response to load in physiological temperatures (Tucci et al. 1984, Perez et al. 2001).
In isolated, blood perfused canine hearts as well as in isolated ferret papillary muscle increased intracellular Ca2+ and also cAMP concentrations have been shown to parallel alterations in contractile force in response to an increase in end diastolic pressure (Todaka et al. 1998, Calaghan et al. 1999). In contrast, in rat atrial preparation, stretching did not change the production of cAMP or cGMP (Tavi et al. 2000). Furthermore, if cAMP would mediate the slow force response, the resulting PKA activation would also lead to phosphorylation of TnI, decreasing the Ca2+ sensitivity of the contractile element. This hypothesis contrasts with the finding that Ca2+ sensitivity of the contractile elements at the beginning of the stretch is increased (Kentish & Wrzosek 1998). Alvarez et al. (1999) suggested that intracellular alkalinization by ET-1 and Ang II induced NHE activation accounts for the mechanism. Indeed, it seems that this mechanism might play a role in hypertrophied, failing or especially in ischemic hearts (Krämer et al. 1991, Perez et al. 1995, Tavi et al. 1999). However, in a further study in normal cat papillary muscle it was shown that intracellular alkalinization is not occurring in the presence of bicarbonate buffered medium (Perez et al. 2001). Furthermore, the NHE activation was suggested to induce a slight increase in [Na+]i, leading to activation of NCX in reverse mode (Na+ out, Ca2+ in). To confirm this it was shown that intracellular Na+ replacement by lithium or by blocking the reverse mode of NCX prevented the development of the slow force response (Perez et al. 2001). This mechanism would also explain the increase in [Ca2+]i. In a study by another group (Calaghan & White 2001), the pivotal role of the endocardial endothelium in the slow force response was confirmed, ET-1 being the key mediator, independently of Ang II.
Recent evidence suggests that the Frank-Starling mechanism is subject to paracrine regulation. Basal release of NO attenuates diastolic stiffness and thus augments the Frank-Starling response (Prendergast et al. 1997b). The slow phase response is regulated via stretch induced release of ET-1 and Ang II. However, at present the role of these mediators in the complete Frank-Starling response in whole organ level is unclear.
2.1.3. The force-frequency relationship
When the contractile frequency is increased, cardiac output is elevated through an increased number of beats per minute, as during exercise. In most species, including nonfailing human myocardium, increased frequency also leads to elevation of contractile force, an event also known as the Treppe phenomenon or the Bowditch effect (Miura et al. 1992). However, in the failing myocardium, frequency potentiation of contractile force is inverse, decreasing contractile force. The Treppe phenomenon has been suggested to result from increased transsarcolemmal Ca2+ influx leading to greater filling of the SR and therefore, a higher amount of Ca2+ available for release during systole (Pieske et al. 1995). This positive inotropic effect can be further augmented with ß-AR agonist dobutamine under resting conditions, when the heart rate is modulated by pacing (Kambayashi et al. 1992). In failing hearts, SR Ca2+ uptake was significantly reduced, suggesting a possible mechanism for inverse force-frequency relationship in CHF (Pieske et al. 1995). Altered Ca2+ handling could be explained by a depressed role of SERCA combined with enhanced cytosolic Ca2+ extrusion via NCX (Pieske et al. 1999b).
2.1.4. The adrenergic system
The effectors of the sympathetic nervous system, i.e. epinephrine and norepinephrine, act on cardiac myocytes via both α- and β - adrenergic receptors. Currently, three ß-AR subtypes, designated ß1-AR, ß2-AR, and ß3-AR, have been cloned and pharmacologically characterized. A fourth subtype (ß4-AR) may also exist, but it is not well characterized (for review, see e.g. Post et al. 1999). All three of the cloned ß-AR subtypes belong to the large family of seven membrane-spanning GPCRs. While β 1-AR is the predominant subtype on cardiac myocytes (66% in mouse and 80% in rat), also ß2-ARs are present (34% in mouse and 20% in rat cardiac myocytes) and capable of mediating positive inotropic responses (Bristow et al. 1986, Hilal-Dandan et al. 2000). Activation of ß1- and ß2-ARs results in an increase in intracellular cAMP after AC stimulation through Gs proteins (Post et al. 1999). The increase in cAMP leads to phosphorylation of PLB, calcium channels and contractile element proteins via PKA. Phosphorylation of these proteins alters their activity and leads to a functional response including positive inotropic effect. In contrast to other ß-ARs, ß3-AR activation leads to negative inotropic response. Inhibitors of NO synthase successfully blocked the negative inotropism of ß3-AR stimulation (Gauthier et al. 1998).
Also α1-AR activation may mediate the positive inotropic responses to catecholamines or adrenergic agonists. There are three subtypes of α1-ARs (α1A-, α1B-, and α1D-ARs), all of which are encoded by distinct genes (for review, see e.g. Brodde & Michel 1999). All α1-AR subtypes are GPCRs, and most commonly the intracellular second messengers are IP3 and DAG formed by PLC activation. IP3 mediates the Ca2+ release from intracellular stores, while also an increase in Ca2+ sensitivity of the myofilaments and NHE activation contribute to positive inotropic effect. Interestingly, in rat heart, α1B-AR is the predominant subtype, while in human and also in mice hearts the α1A- subtype is present in highest amounts (Brodde & Michel 1999). There are also differences in coupling of α1-ARs to downstream effectors in mouse and rat cardiomyocytes, since no α1-adrenergic stimulation of phosphoinositide turnover could be detected in mouse cardiomyocytes (Hilal-Dandan et al. 2000).
CHF is associated with a number of alterations in the activation and deactivation of beta-adrenergic receptor pathways. Continuous adrenergic stimulus results in uncoupling of β -ARs (Post et al. 1999). Activation of the sympathetic nervous system is considered to be one of the major pathophysiological abnormalities in patients with heart failure (Cohn et al. 1984). Elevated circulating norepinephrine and epinephrine have been implicated in contributing to the ß-AR down regulation in both protein and messenger ribonucleic acid (mRNA) level and receptor uncoupling that are characteristic of end-stage heart failure, resulting in subsensitivity to ß-agonist stimulation (Bristow et al. 1982, Fowler et al. 1986) (for reviews, see e.g. Dzimiri 1999). An important mechanism for rapidly regulating ß-AR function is agonist-stimulated receptor phosphorylation by G-protein –coupled receptor kinases (GRKs), resulting in decreased sensitivity to subsequent catecholamine stimulation. ß-AR kinase (ßARK) is a member of this family of GRKs that phosphorylate and regulate a wide variety of receptors that couple to heterotrimeric-G proteins (Pitcher et al. 1998). Interestingly, mice that lack the ability to generate norepinephrine or epinephrine due to genetic disruption of dopamine ß-hydroxylase show increased cardiac contractility associated with a decrease in the level of ßARK1 protein and kinase activity (Cho et al. 1999).
β -AR antagonists were previously considered contraindicated in heart failure due to their negative inotropic effect. Thereafter, an increasing body of evidence has shown that many of the neurohumoral compensatory mechanisms that are activated during CHF are actually deleterious. During the past decade large clinical trials have provided the proof that β -AR antagonists bisoprolol, carvedilol and metoprolol are valuable drugs in treatment of CHF (CIBIS Investigators and Committees 1994, Packer et al 1996, MERIT-HF Study Group 1999), pointing out the importance of understanding the pathophysiological mechanisms behind complex diseases such as CHF.
2.1.5. Circulating hormones
A number of circulating hormones have an impact on the myocardial performance. The adrenomedullary hormone epinephrine exerts its effects on cardiac myocytes through α- and ß-ARs. It is likely that under normal conditions the circulating catecholamines have only minor effects on cardiac contractility compared with the influence of the sympathetic nervous system, which is usually activated in parallel with the adrenomedullary hormone release (Berne & Levy 1993). In rat, hyperthyroidism increases the cardiac contractility (Kolar et al. 1992), and acutely thyroid hormones exert positive inotropic effects in isolated rat heart (Segal et al. 1996). Interestingly, thyroid hormone seems to regulate the expression pattern of central Ca2+ handling proteins, SERCA and NCX, in rat heart during postnatal development inducing an increase in the SERCA mRNA, while decreasing NCX mRNA levels (Rohrer & Dillmann 1988, Arai et al. 1991). Hypothyroid rats have been utilized as a model of decreased contractile function and heart failure (Ng et al. 1991), and all three subtypes of thyroid hormone receptors in cardiomyocytes were downregulated by phenylephrine treatment in cell culture and pressure overload in vivo. Insulin and glucagon (Farah 1983) as well as circulating insulin-like growth factors (Cittadini et al. 1998) have a direct positive inotropic effect on myocardial contraction, although the physiological significance of these effects is not known. Also natriuretic peptides have direct cardiac effects (see section 2.4).