Testosterone
Testosterone, the major androgenic hormone is synthetized and released by the Leydig cells in the testis. It also gives rise to two other potent androgens: dihydrotestosterone and 5-alfa-androstenediol. Epidemiological and clinical studies indicate that testosterone status influence cardiovascular physiology and pathophysiology (Golden et al., 2002; Er et al., 2007). The effects of testosterone on cardiac electric activity have been poorly investigated. Sanchez et al. (2009) showed that the acute administration of 5-alpha-dihydrotestosterone elicited a negative chronotropism effect and increased SA node recovery time, which could improve cardiac performance. The authors also suggested that this effect might be due to an interaction with the underlying mechanisms involved in the pacemaker activity (Mangoni and Nargeot, 2008) such as T-type Ca2+ channel and inward rectifier currents and a functional interaction with ionic pumps of plasma membranes. On the other hand, the acute treatment with testosterone enhanced the spontaneous beating frequency of cultured neonatal cardiomyocytes, which was associated with an increase in the level of expression of T-type Ca2+ channels (Michels et al., 2006). It has also been reported that androgens produce changes in the male heart phenotype and on electrophysiological properties, such as shortening of the QT interval in males after puberty (Rautaharju, 1992; Lehmann, 1997; Locati et al., 1998). These contradictory data may be related to different basal HR values among various mammalian species, and more studies are necessary to better elucidate the role of testosterone on cardiac electric activity.
Most of the research concerning the effects of gonadal hormones on the cardiovascular reflexes has focused on 17p-estradiol. However, other studies have provided evidence that androgens (including testosterone) play an important role in the control of cardiovascular function by modulation of cardiovascular reflexes (Caminiti et al., 2009). Steroids can cross the blood-brain barrier and act on the central nervous system, where androgen receptors in the central cardiovascular regulatory regions, such as NA and DMNX (Peuler et al., 1990; Pouliot et al., 1996) have been demonstrated. Therefore it is possible that androgens may act on brainstem vagal preganglionic neurons to modulate cardiomotor vagal activity. In accordance with this data, El-Mass et al. (2001) have shown that in male rats, castration caused a significant attenuation of baroreceptor control of reflex bradycardia versus no effect on reflex tachycardia. Testosterone replacement increased BRS to phenylephrine in castrated rats and restored reflex bradycardic responses to levels similar to those of sham-operated rats. The muscarinic blockade by atropine in sham-operated rats caused a substantial reduction in BRS to phenylephrine, an effect that was significantly attenuated by castration and restored to sham-operated levels after testosterone replacement, suggesting that testosterone facilitates baroreceptor control of reflex bradycardia. Moreover, the modulatory role of testosterone on baroreflex responsiveness appears to involve, at least partly, enhancement of cardiac vagal efferent activity. Corroborating these data, a long-term testosterone therapy (6 weeks) improves the baroreflex sensitivity in men with chronic heart failure (Caminiti et al., 2009). The blockade of androgen receptor with flutamide attenuates the enhancement of baroreflex bradycardia in sexually mature male rats, indicating that the effects of testosterone on BRS depend on the involvement of the androgen receptor (Ward and Abdel-Rahman, 2006).
Besides the testosterone-induced effects on baroreflex, this sexual hormone may also modulate the cardiopulmonary reflex and the chemoreflex. Bissoli et al (2009) demonstrated that long-term treatment (8 weeks) with supraphysiological doses of nandrolone decanoate reduces the sensitivity of BJR control of HR in male rats. The effects of testosterone on BJR seem to be time-dependent, since the same treatment for 4 weeks had no effects on BJR nor the basal HR (Andrade et al., 2008). Pereira-Junior et al. (2006) showed that 10 weeks of high-dose nandrolone decanoate treatment leads to dysfunction in tonic cardiac autonomic regulation, with marked impairment of parasympathetic cardiac modulation and sympathetic hyperactivity. Regarding the chemoreflex, data from castrated male cats suggest that testosterone increases the hypoxic and hypercapnic ventilatory responses and augmented carotid body sensitivity to hypoxia (Behan et al., 2003). In adult rats, however, castration had no effect on the ventilatory response measured at the end of hypoxia (Joseph et al., 2002). On the other hand, Bairam et al. (2009) demonstrated that gonadectomy increased the acute breathing frequency response to hypoxia in neonatal rats. Because the rapid increase in breathing frequency is attributed to peripheral chemoreceptor activation, these data suggest that testosterone attenuates carotid body function. Although several studies demonstrated contradictory results about the benefic or malefic effects of testosterone on the modulation of cardiovascular reflexes, the characterization of the mechanisms could lead to a better understanding of the effects of testosterone in cardiovascular system and to the development of new therapies.
Nitric oxide (NO)
Since the discovery of the signaling properties of nitric oxide (NO) (Ignarro et al. 1987), it has been suggested that this important molecule may be involved in many physiological processes, such as the control of cardiovascular function. NO is a free radical synthesized from L-arginine by three isoforms of nitric oxide synthase (NOS): NOS1 (neural), NOS2 (inducible), and NOS3 (endothelial) and all three isoforms have been shown to influence autonomic neural function in some manner (Schultz, 2009). NO generated at nerve synapses diffuses in an autocrine and paracrine way to influence both presynaptic and postsynaptic events on excitatory and inhibitory synapses. NO exerts its cellular actions by binding to guanylyl cyclase to activate cGMP production, which remains the only fully recognized physiological signal transduction mechanism for NO. In central neurons, cGMP then can have diverse effects on neuronal excitability. Cyclic GMP can directly bind to and modulate cyclic nucleotide-gated ion channels, bind to phosphodiesterases to impair cAMP hydrolysis, or most prominently, activate cGMP-dependent protein kinase which can directly or indirectly leads to phosphorylation of effector proteins or ion channels (Schultz, 2009).
The effects of NO on baroreflex have been already demonstrated by several investigations. Meyrelles et al (2003) have shown that adenovirus-mediated eNOS delivery to carotid sinus adventitia leads to a diminished baroreceptor activity. NO seems to have an inhibitory effect on sodium currents in baroceptor neurons (Li et al., 1999) and activates calcium dependent potassium channels, leading to membrane hyperpolarization (Bolotina et al., 1994). Besides NO effects on baroreceptor afferents, NO also exerts effects on central nuclei regulating baroreflex function.
Intracerebroventricular injections of L-NAME (an inhibitor of NO synthases) caused an enhancement in baroreflex sensitivity, indicating that NO may exert an inhibitory effect upon baroreflex (Matsumura et al, 1998). This inhibition appears to occur in both sympathetic and parasympathetic component of baroreflex. Liu et al (1996) demonstrated that NO synthase blockade with L-NNA causes an increase in the baroreflex gain, which is prevented by L-arginine injections. This augmented sensitivity is blocked by the use of atropine, indicating an inhibitory effect of NO on the parasympathetic component of the reflex. NO also seems to exert sympatoinhibitory effects, as demonstrated by Zanzinger et al (1995) who show that L-NNA administration leads to an increased basal sympathetic tonus. On the other hand, Dias et al. (2005) demonstrated a stimulatory effect of NO in the central nuclei controlling cardiovascular function. In this study, the renal sympathoinhibition induced by activation of baroreceptors and cardiopulmonary receptors is attenuated by the microinjection of L-NAME in the NTS. The same investigators also demonstrated that NO increases the number of discharges evoked by excitatory amino acids in NTS neurons that receive vagal afferent inputs, and action potentials induced by iontophoretic application of AMPA in the NTS was reduced by L-NAME, indicating a excitatory effect of NO in this nucleus (Dias et al., 2003). Some studies also showed no effects of NO on baroreflex function. eNOS gene therapy did not alter baroreflex sensitivity and autonomic balance in C57 mice and was not able to prevent the increase in sympathetic tonus and the decrease parasympathetic activity to the heart in hypertensive mice (Gava et al., 2008).
In addition to the brain, emerging evidence suggests that NO can also influence sympathovagal function at the site of the end-organ itself, acting in sympathetic ganglia or vagal neurons. Neuronal nitric oxide synthase is localized in both intrinsic cardiac vagal neurons and stellate sympathetic ganglia innervating the SA node, indicating an important role NO in modulating of peripheral neuronal function (Herring and Paterson, 2009). In cholinergic neurons, NO seems to act increasing acetylcholine release through stimulation of soluble guanylate cyclase. The resultant generation of cGMP causes phosphodiesterase-3 inhibition, increasing cAMP-PKA dependent phosphorylation of N-type calcium channel and calcium-induced exocytotic release of acethycholine (Herring & Paterson, 2001). However, in the AV nodal cells, NO regulates AV excitabilility by muscarinic cholinergic attenuation of ICa-L (L-type calcium current), the mechanism likely involves the cGMP-stimulated phosphodiesterase (Han et al., 1997). In sympathetic ganglia, NO reduces the release of noradrenaline through a soluble guanylate cyclase-cGMP dependent pathway that reduces calcium influx (Schwartz et al. 1995; Wang et al. 2007), probably via stimulation of PDE2 and/or protein kinase G (Herring and Paterson, 2009). Despite some contradictory results, the role of NO in the modulating HR it is well established and the implication of changes in the NO production and/or activity for cardiovascular disease development remains an intriguing possibility of new targets for treating arrhythmias.
Renin-angiotensin-aldosterone system (RAAS)
The RAAS is a peptidergic cascade with endocrine characteristics and is considered one of the most important systems that participate of cardiovascular control. In the classical view of RAAS, angiotensinogen, an alfa-glycoprotein, is released from the liver and is cleaved in the circulation by the enzyme renin that is secreted from the juxtaglomerular apparatus of the kidney to form the decapeptide angiotensin I (Ang I). Ang I is then transformed into to the octapeptide angiotensin II (Ang II) by angiotensin converting enzyme (ACE), a membrane-bound metalloproteinase, which is predominantly expressed in high concentrations on the surface of endothelial cells in the pulmonary circulation. Ang II, considered the main effector peptide of the RAAS, acts on specific receptors (AT1 and AT2), for example, to induce vasoconstriction on vascular smooth muscle cells or to stimulate the release of aldosterone from the adrenal cortex (Paul et al., 2006).
Several lines of evidence suggest that Ang II may exert a direct modulation on cardiac ionic channels. Experiments have shown that stimulation of AT1 receptor result in the inhibition of transient outward potassium channel in myocytes from rat or canine ventricle (Shimoni and Liu, 2003; Yu et al., 2000). Ang II also increases cardiac L-type Ca2+ current (ICaL) in isolated cat myocytes (Aiello and Cingolani, 2001). In this view, the RAAS activation may therefore significantly contribute to the pathogenesis of cardiac arrhythmias. On the other hand, Ang II decreased the current density of L-type Ca2+ current in SA node cells and reduces the auto rhythm of SA node cells via enhancing slowly activated delayed rectifier K+ currents and reducing ICaL. Therefore, the elevated levels of Ang II may be involved in the occurrence of SA node dysfunction in cardiac pathophysiology (Sheng et al., 2011).
Numerous studies already demonstrated that Ang II plays a pivotal role in the neural regulation of cardiovascular system. High concentrations of AT1 receptor and fibers with Ang II immunoreactivity have been described in the dorsomedial and ventrolateral areas of the medulla (Allen et al., 1998; Averill and Diz, 2000). It is well known that Ang II causes an increased sympathetic drive, particularly by means of central mechanisms. In dogs, acute (21 h) and chronic (5 days) infusion of Ang II caused a two- to threefold increase in Fos-Li immunoreactivity in the NTS and CVLM, leading to a baroreceptor suppression of sympathoexcitatory cells in the RVLM (Lohmeier et al, 2002). Lesions at either the area postrema or the subfornical organ attenuate angiotensin II-based hypertension, indicating a direct central sympathoexcitatory action of Ang II (Collister and Hendel, 2003; Collister and Hendel, 2005). Corroborating these data, experimental models of angiotensin II-dependent hypertension present an augmented sympathetic drive (Peotta et al., 2007) and patients with chronic angiotensin-dependent renovascular hypertension have generally demonstrated higher sympathetic levels, correlated with circulating angiotensin II concentrations (Grassi e Esler, 2002). Besides Ang II effects on sympathetic drive, this peptide also exerts effects on the parasympathetic component of the reflexes. Borges et al. (2008) demonstrated that mice with renovascular hypertension presented diminished cardiac vagal activity, and together with an enhanced cardiac sympathetic activity, contributed to a reduced baroreflex sensitivity in this animal model of hypertension. Moyses et al. (1994) also demonstrated a reduced cardiac vagal activity in renovascular hypertensive rats.
Although Ang II is considered the major effector of RAAS system, growing evidence have demonstrated an important role of angiotensin-(1-7) in cardiovascular regulation. This molecule can be formed from Ang I and Ang II fragments through an angiotensin-converting enzyme (ACE) independent pathway (Santos et al., 2007). It has been demonstrated that Ang-(1-7) actions are often contrary to those described for Ang II (Benter et al., 1993). In fact, regarding the neural control of circulation, several studies have provided evidence that endogenous Ang-(1-7) enhances the baroreceptor reflex bradycardia, while Ang II attenuates it (Campagnole-Santos, 1992; Sakima et al., 2007). The beneficial effect of Ang-(1-7) on cardiovascular reflexes was also demonstrated by Oliveira et al (1996) who showed that the central infusion of a selective Ang-(1-7) antagonist attenuates baroreflex and blocks the improvement in the reflex bradycardia produced by Ang-(1-7). The specific binding of Ang-(1-7) to its receptor (Mas receptor) seems to be a basic requirement for the maintenance of normal arterial blood pressure and cardiovascular reflex control, since Mas-knockout mice presented hypertension and altered cardiovascular reflexes (Moura et al., 2010).
Natriuretic peptides (NPs)
The NPs play an important role in the regulation of cardiovascular homeostasis maintaining blood pressure and extracellular fluid volume. There are four major natriuretic peptides (NPs) that have been isolated: atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), C-type natriuretic peptide (CNP), and Dendroaspis-type natriuretic peptide (DNP). NPs exert their biological effects by binding to three distinct cell surface receptors denoted NP receptors A, B and C (NPR-A, NPR-B and NPR-C) (Rose and Giles, 2008).
Several studies demonstrated that NPs affect the electrophysiology of the heart (Rose et al., 2004) and central nervous system (Trachte et al., 2003; Rose et al., 2005). Voltage-clamp studies demonstrated that CNP can inhibit L-type Ca2+ current (ICa-L) through NPR-C binding. This inhibition involves a decrease in adenylyl cyclase activity, which leads to reduced intracellular levels of cAMP (Rose et al., 2003). These results were also demonstrated in isolated myocytes from mouse SA node, that express several cAMP-sensitive currents, including ICa-L (DiFrancesco, 1993). Corroborating these data, inhibition of adenylyl cyclase decreases HR and increases the P-R interval, suggesting that the atrioventricular conduction system is slowed following the activation of NPR-C. These data are consistent with other studies demonstrating a key role for L-type Ca2+ channels in the intrinsic regulation of SA node function and the determination of HR (Zhang et al., 2002; Mangoni et al., 2003). The molecular mechanism(s) by which CNP-NPR-C effects are compartmentalized in animal models SA node myocytes is not clear and will require further investigation.
In addition to their effects on cardiac electric activity, NPs also exert effects on cardiovascular reflexes. Thomas et al. (2001) showed that ANP, BNP and CNP enhance bradycardic responses to cardiopulmonary chemoreceptor activation in conscious sheep. On the other hand, Tallarida et al. (1991) demonstrated that intravenous infusion of ANP did not substantially change the baroreflex cardiocirculatory responses to loading and unloading carotid and aortic baroreceptors. Some of the reported discrepancies may be attributed to the dose of ANP, preparation (e.g., synthetic peptide vs. atrial extract) or to experimental conditions (e.g., anaesthetized vs. conscious). The target site(s) for the NPs action on cardio-cardiac vagal reflexes is not clear and more studies are necessary to better elucidate the mechanisms involved in NP-induced changes in cardiovascular reflexes.
Thyroid hormones (TH)
Variations from euthyroid status affect virtually all physiological systems and the effects on the cardiovascular system are particularly pronounced (Levey and Klein, 1990). Hyperthyroidism causes tachycardia and cardiac arrhythmias whereas bradycardia, reduced cardiac output, and slowed relaxation result from hypothyroidism (Klein and Ojamaa, 2001). The actions of TH are mediated by two nuclear TH receptors (TRs)- a and- p, encoded by two separate genes (Yen, 2001). TR-a1 isoform represents 70% of the TRs and serves an important role in cardiac development (Mai et al., 2004) and the regulation of heart rate and contractility (Dilmann, 2010; Macchia et al., 2001). Corroborating these data, Wikstrom et al. (1998) demonstrated that TR-a1 knockout mice presented a 20% reduction in HR and a prolonged relaxation time. The molecular explanation for these results includes a diminished expression of the hyperpolarization activated cyclic nucleotide-gated potassium channel 2, which plays a pivotal role for pacemaking (Macchia et al., 2001). Changes in thyroid status are associated with changes not only in cardiac and vascular function but also in autonomic regulation of the cardiovascular system (Levey and Klein, 1990). In example, Foley et al. (2001) evaluated the effect of thyroid status on arterial baroreflex control of lumbar sympathetic nerve activity (LSNA) and HR in conscious rats. The authors report that rats with hypothyroidism exhibit blunted baroreflex mediated increases in LSNA and HR and a downward shift in baroreflex control of HR compared with euthyroid rats. On the other hand, rats with hyperthyroidism presented normal baroreflex function and sympathetic tone to the vasculature. Although hypothyroidism has been associated with sympathovagal imbalance, current literature shows conflicting results with either increased sympathetic activity (Cacciatori et al., 2000), decreased sympathetic modulation (Gallet et al., 2008) or an increased vagal tone (Xing et al., 2001).
As observed, there is a complex relationship between humoral factors, neural systems (CNS and autonomic nervous system) and cardiac electric activity (Figure 5) and disturbances in these interactions may be related with the development of arrhythmias.
Fig. 5. Schematic diagram showing the interactions between humoral factors, cardiac electric activity, autonomic balance and central nervous system and their role in arrhythmias generation.
Perspectives
As observed, the normal control of HR depends on a complex interaction between neural and humoral factors and disturbances on these systems are strongly related with arrhythmias generation. The formation of an action potential in the SA node and its propagation throughout the heart involves several ion channels, mainly Na+, K+ and Ca+2, and can be modulated by sympathetic and parasympathetic activation. The central outflow of autonomic nervous system is generated mainly in the brainstem and it involves the participation of diverse nuclei, such as NTS, CVLM and RVLM. The neuronal activity of these structures can be modulated by several hormones, including estrogen, testosterone, nitric oxide, angiotensin II, angiotensin (1-7), natriuretic peptides and thyroid hormones. Besides its effects on CNS, hormones can also regulate the release of neurotransmitters, the expression of ion channels and the activity of membrane transporters. Taken together, these data demonstrate the importance of neural and humoral systems in controlling cardiovascular function and brings out the possibility of new drug targets to treat arrhythmias.
