Increased spontaneous activity and reduced inotropic response to catecholamines in ventricular myocytes from footshock-stressed rats

Abstract

Exposure to stressors has been shown to change atrial responsiveness to catecholamines, but it is not clear yet how it affects the ventricular myocardium, which plays a major role in the catecholamine-stimulated increase in cardiac output. Adult male rats were submitted to restraint (RST) or footshock (FS) sessions for 3 days. Reactivity to agonists of the β-adrenergic pathway was analyzed in left ventricular myocytes isolated from stressed and control rats (CTR). Whereas no significant changes were detected after RST, enhancement of catecholamine-induced spontaneous activity, accompanied by decrease in inotropic maximal response, was observed in myocytes from FS rats. Changes were reversed by β₁-, but not by α₁- or β₂-adrenoceptor (AR) blockade. Similar alterations were seen in response to forskolin. However, responsiveness to 3-isobutyl-1-methylxanthine and CaCl₂ was comparable in control and FS groups. A significant negative correlation was observed between the maximally stimulated spontaneous activity rate and contraction amplitude. Results indicate that: (a) enhanced automatism during adrenergic stimulation of myocytes from FS rats is mediated by β₁-ARs and seems to involve post-receptor mechanisms, probably decreased cAMP degradation; (b) the exaggerated spontaneous activity, which may contribute to generation of catecholaminergic arrhythmias, might limit the development of the inotropic response.

Keywords::

• β-Adrenoceptors
• arrhythmia
• catecholamines
• footshock stress
• inotropic response
• restraint stress
• ventricular myocardium

Introduction

The hallmarks of the physiological response known as the stress reaction are enhanced sympathetic outflow and activation of the hypothalamic-hypophysial-adrenal axis, resulting in increased catecholamine release and glucocorticoid secretion. Both types of chemical signal may affect the function of several peripheral tissues, thus evoking visceral, metabolic and behavioral changes that may enable the individual to cope with the stressor (Charmandari et al. 2005).

Sympathetic efferent neural activity is of paramount importance for regulation of cardiovascular function, by both the short-term effects of adrenergic mediators on vascular tonus, heart rate and myocardial inotropism, and the long-term effects of adrenoceptor (AR) stimulation on gene transcription (Müller et al. 2000). Norepinephrine (NE) and epinephrine, released from both sympathetic nerve terminals and adrenal medulla, play a role in the sympathetic regulation of cardiovascular function. Under challenging conditions (e.g., exercise, stress), catecholamines from the adrenal gland (mostly epinephrine) seem to be particularly important for enhancement of cardiac output via increase in stroke volume, apparently due to augmenting venous return and stimulating ventricular inotropic function, whereas heart rate is increased mainly by locally released NE (e.g., Donald and Shepherd 1963; Fuji and Vatner 1985; Bao et al. 2007).

Several studies have reported changes in the responsiveness of atrial tissue to catecholamines after repeated/continuous exposure to stressors such as FS, immobilization/RST, forced swimming, parasitosis and cold (Callia and De Moraes 1984; Bassani and De Moraes 1987, 1988; Spadari and De Moraes 1988; Bassani and Bassani 1993; Silveira et al. 2003; Brotto 2003; Santos and Spadari-Bratfisch 2006). Nevertheless, in spite of the primary importance of the ventricular inotropic response for the increase in cardiac output during sympathetic activation, it has not yet been investigated how stress may affect the adrenergic reactivity of ventricular myocardium. Changes in myocardial responsiveness to catecholamines might modify the effectiveness of visceral reflexes (e.g., baroceptor reflex) in which the sympathetic efferent branch modulates cardiovascular function, which may either provide adaptive support for physiological changes associated with the stress condition, or contribute to circulatory dysfunction.

In the present study, we examined the inotropic responsiveness to catecholamines in ventricular myocytes isolated from rats repeatedly exposed to two different stress-inducing procedures: RST and FS. Stress-inducing protocols were similar to those used in previous studies in which stress was shown to modify the reactivity of atrial tissue to catecholamines (Bassani and De Moraes 1987, 1988; Bassani and Bassani 1993; Silveira et al. 2003). Isolated myocytes are especially suitable for assessment of the direct effect of adrenergic agonists on the myocardium because, by contrast with what occurs with multicellular preparations, neurotransmitter release is absent and the influence of catecholamine uptake and metabolism is negligible. Our results point out stressor-dependent changes not only in the inotropic response to stimulation of the β-AR signaling pathway, but mainly in the induction of spontaneous activity, which may have important implications in the generation of ventricular arrhythmias (ter Keurs and Boyden 2007).

Methods

Stress induction

Adult male Wistar rats (4–5 months old) were maintained under controlled temperature (23 ± 2°C) and light–dark cycle (12:12 h; lights on at 6 AM), receiving pelleted chow and water ad libitum. Rats were randomly assigned to one of the following groups: (a) controls (75 rats): rats were not manipulated, except for the routine care procedures; (b) RST stress (31 rats): rats were physically restrained in a conical plexiglas tube (75 and 25 mm major and minor inner diameters) for 1 h (Silveira et al. 2003); (c) FS stress (FS; 57 rats): rats received electric shocks (1 mA, 1 s duration) applied through the cage floor at pseudo-random intervals (15 s average) for 30 min (Bassani and De Moraes 1987, 1988; Bassani and Bassani 1993).

RST and FS sessions started at 8:30–9:30 AM and were applied once a day for 3 days. Approximately 30 min after the last session, the rat was euthanized, and the heart was excised for myocyte isolation. Animal care and stress-inducing protocols were approved by the Committee of Ethics in Animal Research of the University of Campinas (CEEA/IB/UNICAMP. docs. No. 636-1, 799-1, 951-1, and 1184-1).

Isolated ventricular myocytes

Rats were rendered unconscious by cerebral concussion, and euthanized by rapid exsanguination, after which the heart was immediately removed. Myocytes were isolated from the left ventricle by coronary perfusion at 37°C with collagenase (Carvalho et al. 2006). Briefly, hearts were perfused with modified Krebs-Henseleit solution (composition in mM: 115 NaCl; 4.6 KCl; 1.2 KH₂PO₄; 25 NaHCO₃; 2.4 MgSO₄; 11 glucose; pH 7.4, gassed with 95%O₂/5% CO₂) for 5 min, and then with the same solution containing 0.8–1 mg/ml collagenase I (Worthington Biochem., Lakewood, NJ, USA) for approximately 15 min, and again with the enzyme-free solution for 5 min. After the left ventricle was excised, cells were dissociated and kept at 4°C in the storage solution (composition in mM: 30 KCl; 70 glutamic acid; 10 KH₂PO₄; 5 4-(2-hydroxyethyl)-l-piperazine-ethanesulfonic acid (HEPES); 1 MgCl₂; glucose 11; taurine 20; pH 7.4).

Cells were plated on a collagen-coated perfusion chamber, and perfused at 23–24°C with modified Tyrode's solution with the following composition (mM): 140 NaCl, 6 KCl, 1.5 MgCl₂, 1 CaCl₂; 5 HEPES; 11 glucose; pH 7.4. Myocytes were field-stimulated at 1 Hz (biphasic voltage pulses, amplitude 1.2X the threshold, 5 ms duration).

Cell shortening was recorded with a video-edge detector (Crescent Instruments, Sandy, UT, USA; and Center for Biomedical Engineering/UNICAMP, Campinas, SP, Brazil). Twitch amplitude (ΔL), considered as the peak cell shortening evoked by an electrical stimulus, was averaged from three twitches not preceded by extrasystolic contractions (ES), and expressed as percentage of the resting cell length (RL).

Some cells developed ES during electric stimulation when exposed to high catecholamine concentrations (>0.1 μM), but not in the absence of the agonists. In all in vitro protocols, spontaneous activity (automatism) was quantitated as the rate of spontaneous contractions (SC) at rest (Boer and Bassani 2004; Carvalho et al. 2006), which significantly correlates with the rate of extrasystolic activity during electrical stimulation (Pearson correlation coefficient = 0.701; N = 261; p < 0.001). The total number of SCs was computed during two 30 s-long rest intervals separated by a 30 s electrical stimulation. In some experiments, the rate of ES developed under pacing was computed during 20–40 s periods at each agonist concentration. Irreversible inhibition of the sarcoplasmic reticulum (SR) Ca²⁺ ATPase by treatment with 10 μM thapsigargin for 5 min (Bassani et al. 1995a,b) totally abolished spontaneous activity during both rest and electrical stimulation, but not ΔL increase, in the presence of catecholamines (not shown).

In vitro experiments

Concentration-effect curves were determined for isoproterenol (ISO) and NE, as well as for 3-isobutyl-1-methylxanthine (IBMX) and CaCl₂. Some cells were challenged with 3 μM forskolin (FSK). For each cell, SC rate and ΔL were determined at baseline and at different agonist concentrations. Inotropic and automatic responses were recorded after ΔL stabilization following agonist addition (2–4 min for catecholamines and CaCl₂; 10–15 min for IBMX and FSK). When used, competitive AR antagonists were present in the perfusate for at least 30 min before the agonist was added. At the concentrations employed in this study, none of the antagonists produced significant change in ΔL or SC rate (not shown). In some experiments with control myocytes, concentration-effect curves to ISO were determined also in the presence of 1 mM caffeine.

The pA₂ value of the selective β₁-AR antagonist metoprolol (MET; 30, 60, 100, and 300 nM; N = 5–9 for each MET concentration) was determined according to Arunlakshana and Schild (1959), using ISO as the agonist.

Except for FSK and thapsigargin (Calbiochem, La Jolla, CA, USA), all chemicals were from Sigma Chem Co. (St Louis, MO, USA). Most stock solutions were kept at − 20°C, and work solutions were prepared daily. Dimethylsulfoxide was used as the solvent for hydrophobic drugs, and its concentration in the perfusate was ≤ 0.1%. Tyrode's solution was prepared with salts of analytical grade and type I water.

Data analysis

The concentration-effect curve parameters maximum response (R_max) and pD₂ (the negative logarithm of the agonist molar concentration that produces half-maximum response) were obtained by non-linear curve fitting, except for CaCl₂, in which case fitting of a sigmoid function was not reliable. Linear regression was used for pA₂ determination.

Data are expressed as mean ± SEM. N refers to the number of studied cells, isolated from 4 to 16 different rats of a given group, for each in vitro protocol. Statistical comparisons were made by one or two-way analysis of variance followed by post-hoc Bonferroni t-test, or by Student's t-test for paired or unpaired samples, taking p = 0.05 as the limit of statistical significance. pA₂ values were compared by superposition of the 95% confidence intervals. Prism (4.0, GraphPad Software) was used for curve fitting and statistical analysis.

Results

Stressor-dependent changes in ventricular myocyte responsiveness to catecholamines

RL was comparable in all groups (control: 130 ± 2, N = 91; RST: 133 ± 3, N = 28; FS: 126 ± 2 μm, N = 85; p>0.14). Under electric stimulation at 1 Hz, ΔL and SC rate were similar for control, RST and FS groups (ΔL: 8.2 ± 0.7, 7.2 ± 0.6, and 8.0 ± 0.6% of RL, respectively, p>0.55; SC rate: 0.6 ± 0.1, 0.2 ± 0.1, and 0.7 ± 0.2 SC/min, respectively; p>0.10).

The analysis of variance revealed a significant (p ≤ 0.01) group-dependent influence on both inotropic and automatic R_max to NE, the main sympathetic neurotransmitter. Whereas responsiveness to the agonist was comparable in RST and control groups, a significant decrease in inotropic R_max, accompanied by marked increase in automatic R_max, was observed in myocytes from the FS group (Table I, Figure 1). Accordingly, 62% of the FS group myocytes developed ES (vs. 24% in the control group) when exposed to high NE concentrations. NE pD₂ values for both responses were also affected by the in vivo experimental protocol (p < 0.03), as they were significantly increased in the FS group versus RST and/or control groups (Table I). For ISO, the responsiveness pattern was similar to that observed with NE, except for the change in the pD₂ value for the inotropic response. In short, these results indicate that FS, but not RST, was effective at modifying the responsiveness of ventricular myocytes to catecholamines, decreasing the apparent inotropic efficacy of the agonists, while enhancing catecholamine-evoked automatism.

Table I.  Inotropic and automatic responses to isoproterenol (ISO) and norepinephrine (NE) in ventricular myocytes from control rats (CTR) and rats subjected to restraint (RST) and footshock (FS) for 3 days.

	ISO			NE
	CTR (17)	RST (13)	FS (17)	CTR (27)	RST (15)	FS (27)
Inotropic response
Rmax^† (% of RL)	11.2 ± 1.1	14.0 ± 2.0	5.4 ± 0.9^*,^*^*	14.8 ± 1.3	15.6 ± 1.7	8.4 ± 0.8^*,^*^*
pD2^‡	8.37 ± 0.07	8.28 ± 0.05	8.34 ± 0.14	7.55 ± 0.07	7.54 ± 0.08	7.85 ± 0.09^*,^*^*
Automatic response
Rmax^§ (SC/min)	4.4 ± 0.7	4.5 ± 0.7	10.3 ± 1.2^*,^*^*	5.3 ± 0.5	4.7 ± 0.6	10.2 ± 1.2^*,^*^*
pD2^‡	7.64 ± 0.08	7.42 ± 0.09	7.94 ± 0.18^*^*	6.77 ± 0.11	6.52 ± 0.10	6.92 ± 0.07^*^*

Data are presented as mean ± SEM. N (the number of studied cells) is shown in parentheses

^†maximum inotropic response (twitch amplitude as percentage of the resting cell length, RL)

^‡negative logarithm of the molar agonist concentration that evoked half-maximum response

^§maximum automatic response (rate of spontaneous contractions, SC, at rest)

^*p < 0.05 vs. CTR

^**p < 0.05 vs. RST (Bonferroni t-test).

Figure 1  Concentration-effect curves for norepinephrine (NE) in ventricular myocytes isolated from control rats (CTR; N = 27), as well as rats submitted to restraint (RST; N = 15) and footshock (FS; N = 27) sessions for 3 days. Inotropic response (panel A) was taken as the increase in peak twitch cell shortening (ΔL), expressed as percent of the resting cell length (RL). Automatic response (panel B) was taken as the increase in the rate of spontaneous contractions (SC) in the absence of electric stimulation. Symbols and bars indicate mean and SEM values, respectively. Curve parameters are presented in Table 1.

The AR signaling pathway and FS-induced increase in spontaneous activity

The selective α₁-AR blocker prazosin (PRZ, 0.1 μM) did not affect significantly inotropic or automatic response to NE (which acts at both α₁- and β₁-AR) in either control or FS group (Tables I and II). The lack of significant effects of PRZ on the responses of control cells is consistent with previous observation that α₁-AR contribution to NE effects seems to be small in rat atrium and ventricular myocytes (Boer et al. 2004). On the other hand, the persistence of the FS-induced alterations in the presence of PRZ indicates that the α₁-AR signaling cascade does not seem to play an important role in such alterations.

Table II.  Inotropic and automatic responses in ventricular myocytes from control (CTR) and footshock-stressed rats (FS): influence of β-adrenoceptor antagonists and response to 3-isobutyl-1-methylxanthine (IBMX).

Agon	Antag	CTR		FS
		Rmax^†	pD2	Rmax^†	pD2
Inotropic response (% of RL)
NE	PRZ	12.6 ± 1.9 (8)	7.45 ± 0.20	7.9 ± 1.1 (9)^*	7.92 ± 0.21^*
ISO	MET	10.5 ± 1.2 (6)	6.56 ± 0.11^*^*	9.5 ± 0.9 (10)^*^*	6.58 ± 0.13^*^*
ISO	BUT	7.9 ± 1.4 (16)	8.22 ± 0.15	5.8 ± 1.6 (10)	7.98 ± 0.20
ISO	ICI	8.1 ± 1.3 (22)	8.28 ± 0.13	7.2 ± 1.4 (15)	8.29 ± 0.11
IBMX	–	6.7 ± 0.9 (9)	5.04 ± 0.23	7.3 ± 0.7 (12)	4.93 ± 0.18
Automatic response (SC/min)
NE	PRZ	5.2 ± 0.8 (8)	6.72 ± 0.08	9.3 ± 0.9 (9)*	6.83 ± 0.25
ISO	MET	6.2 ± 1.2 (6)	5.87 ± 0.18^*^*	5.2 ± 0.6 (10)^*^*	6.20 ± 0.15^*^*
ISO	BUT	8.7 ± 0.8 (16)^*^*	7.48 ± 0.19	10.8 ± 2.4 (10)	7.49 ± 0.10
ISO	ICI	12.6 ± 1.0 (22)^*^*	7.53 ± 0.17	12.3 ± 1.1 (15)	7.42 ± 0.13
IBMX	–	6.8 ± 1.1 (9)	4.53 ± 0.20	6.4 ± 0.9 (10)	4.26 ± 0.11

Data are mean ± SEM. N (the number of studied cells) is shown in parenthesis

^‡Maximum inotropic (increase in twitch peak shortening, as percentage of the rest cell length, RL) and automatic responses (rate of spontaneous contractions, SC at rest). Agonists (Agon): norepinephrine (NE); isoproterenol (ISO). Antagonists (Antag): prazosin (PRZ, 0.1 μM); metoprolol (MET, 0.1 μM); butoxamine (BUT, 0.3 μM); and ICI118551 (ICI, 0.1 μM). For NE and ISO, data were also compared by two-way analysis of variance with those obtained in the absence of antagonists (shown in Table 1)

*p < 0.05 vs. CTR

^†p < 0.05 vs. in the absence of antagonist in the same group (Bonferroni t-test).

To identify the role of β-AR subtypes involved in the altered responsiveness to ISO seen in FS group myocytes, curves for the agonist were determined during blockade of β₁-AR (MET, 0.1 μM) or β₂-AR (butoxamine, BUT, 0.3 μM; and ICI118551, 0.1 μM). β₁-AR antagonism by MET markedly decreased ISO pD₂ for both responses in control cells (p < 0.001), without affecting significantly R_max values. In FS group myocytes, however, the rightward shift in both curves was accompanied by a paradoxical enhancement of the inotropic R_max and concomitant depression of the automatic R_max (p < 0.04 for group × MET interaction), so that both R_max values were comparable with those in the control group (Table II). Blockade of β₂-AR by BUT or ICI118551 did not significantly change R_max values to ISO in FS cells. In control myocytes, however, the automatic R_max was significantly elevated by both antagonists (p < 0.05; Figure 2(A),(B); Table II), whereas the inotropic R_max was ∼30% lower, although this difference did not reach statistical significance (p>0.05 for antagonist and group × antagonist interaction). A trend to lower pD₂ values was observed, which was marginally significant for ICI118551 (automatic response only, 0.05>p>0.04), but not for BUT (0.08>p>0.05). Thus, the changes in ISO responsiveness evoked in control cells by β₂-AR blockade resembled those of FS stress, while β₁-AR antagonism abolished the alterations in ISO apparent efficacy observed in FS myocytes.

Figure 2  Concentration-effect curves for isoproterenol (ISO) in ventricular myocytes isolated from control (CTR) and footshock-stressed rats (FS), determined in the presence of 0.1 μM ICI118551 (ICI). Dashed gray and black lines indicate the curve to ISO obtained in the absence of the antagonist in CTR and FS, respectively. Twitch shortening peak (panel A) and rate of spontaneous contractions during rest (panel B) are expressed as in Figure 1. Panel C shows the rate of extrasystolic contractions (ES) recorded during pacing as a function of ISO concentration. Symbols and bars indicate mean and SEM values, respectively. Curve parameters are presented in Tables I and II.

Figure 2(C) shows that ES induction by ISO in control myocytes requires high agonist concentrations ( ≥ 0.1 μM), with attainment of a maximum average rate of ∼5 ES/min. This was partially due to the low incidence of ES in these cells (36% at 0.3–1 μM ISO). In contrast, in cells from the FS group, ES could be observed at a lower agonist concentration (30 nM), and reached a rate fourfold higher at 0.3–1 μM ISO, at which ES occurred in 89% of the cells. Similarly to SC, ES rate in FS in myocytes was little affected by β₂-AR blockade by ICI118551, which, however, greatly enhanced the response in control cells, reproducing the pattern observed in FS myocytes. Comparison of ES rates at 1 μM ISO by two-way analysis of variance revealed significant influence of group and ICI118551 (p < 0.04), while the post-hoc test detected significant effect of the antagonist only in the control group (p < 0.01).

Concentration-effect curves for the partial β-AR agonist CGP12177 were determined in a few cells. In both control and FS groups, inotropic R_max was very low ( < 20% of that to NE), and CGP12177 concentrations up to 10 μM were unable to evoke significant spontaneous activity (data not shown).

Schild plot for the antagonism of ISO inotropic effect by MET in control ventricular myocytes resulted in a line with slope not significantly different from 1.0 (mean and 95% confidence interval were 0.90 ± 0.06; R² = 0.99). MET pA₂ value (mean and 95% confidence interval: 8.71 [8.34–9.08]) was not statistically different from that previously determined in rat atria (Bassani and De Moraes 1988; Bassani and Bassani 1993), in which the contribution of the β₂ subtype to the response to ISO is undetectable (Juberg et al. 1985; Bassani and De Moraes 1988). The increase in contractile R_max observed during MET exposure precluded the pA₂ determination in FS cells.

To investigate post-receptor mechanisms, we initially determined the response to the adenylyl-cyclase stimulator FSK (3 μM). The positive inotropic effect of FSK on rat ventricular myocytes is completely abolished by H-89 (Bassani et al. 1995b), which indicates a crucial role of cAMP-dependent protein kinase (PKA) activation in its induction. The inotropic response to FSK was comparable to the R_max to ISO, although the increase in spontaneous activity was 25% lower than the automatic R_max to ISO in control cells. As shown in Figure 3, FS group cells developed greater automatic response and lower inotropic response to FSK in comparison with control (p < 0.03), which resembled the changes in the R_max to catecholamines. This result could not be explained if changes in β-AR coupling to G_αs proteins and adenylyl-cyclase stimulation upon receptor occupation were relevant in the determination of the altered responsiveness to catecholamines in FS group myocytes.

Figure 3  Inotropic (left bars: increase in twitch peak shortening, ΔL, as percent of resting cell length, RL) and automatic (right bars: rate of spontaneous contractions, SC, at rest) responses to 3 μM forskolin (FSK) in ventricular myocytes isolated from control (CTR) and footshock-stressed (FS) rats. Bars are mean and SEM. Baseline ΔL was 8.0 ± 1.1 in CTR (N = 19) and 8.9 ± 0.8% of RL in FS (N = 20). Baseline SC rate was 0.3 ± 0.2 and 0.8 ± 0.3 SC/min in CTR and FS, respectively (p>0.10). ^*p < 0.05 vs CTR (t-test for unpaired samples).

The R_max and pD₂ values for the non-selective phosphodiesterase (PDE) inhibitor IBMX in producing positive inotropic and automatic responses were similar in control and FS group cells (p>0.22; Figure 4, Table II). Likewise, comparable responses (p>0.38) to the increase in extracellular Ca²⁺ concentration ([Ca²⁺]_o (from 0.25 to 6 mM) were observed in control and FS groups (Figure 5).

Figure 4  Concentration-effect curves for 3-isobutyl-1-methylxanthine (IBMX) in ventricular myocytes isolated from control (CTR; N = 9) and footshock-stressed rats (FS; N = 12). Inotropic (panel A) and automatic (panel B) responses are expressed as in Figure 1. Symbols and bars indicate mean and SEM values, respectively. Baseline ΔL and SC rate were not significantly different in CTR and FS (p>0.36). Curve parameters are presented in Table II.

Figure 5  Inotropic (panel A) and automatic (panel B) responses to the increase in extracellular CaCl₂ in isolated ventricular myocytes from control (CTR; N = 8) and footshock-stressed rats (FS; N = 8). Symbols and bars indicate mean and SEM values, respectively.

Because FS-dependent differences were not present when IBMX was the agonist, we further investigated whether the altered responsiveness to ISO observed in FS group cells could be reproduced in control myocytes exposed to 1 mM caffeine, which causes both PDE inhibition and facilitation of diastolic SR Ca²⁺ release (Daly 2000). In the absence of ISO, caffeine did not change ΔL significantly (10.8 ± 1.3 vs. 10.8 ± 1.4% of RL, p>0.98). Basal SC rate was increased, but the difference did not achieve statistical significance (3.8 ± 0.9 vs. 1.0 ± 0.7 SC/min, p>0.05). With caffeine in the perfusate, the inotropic curve was markedly flattened, as shown in Figure 6(A) (R_max = 3.9 ± 0.8 vs. 11.2 ± 1.1% of RL in the absence of caffeine, p < 0.01), whereas the automatic R_max was nearly threefold greater (13.0 ± 1.2 vs. 4.4 ± 0.1 SC/min in the absence of caffeine; p < 0.01). Caffeine also enhanced the sensitivity to ISO (pD₂ for inotropic effect: 9.10 ± 0.15 vs. 8.37 ± 0.07 in the absence of caffeine; pD₂ for automatic effect: 8.74 ± 0.33 vs. 7.64 ± 0.08 in the absence of caffeine, p < 0.001), probably due to PDE inhibition, which is expected to potentiate ISO-stimulated cAMP accumulation. Thus, regarding the R_max values, caffeine reproduced in control cells the same qualitative changes in the responsiveness to ISO seen in FS group myocytes, i.e., an enhanced automatic response paralleled by decreased positive inotropic response. The possibility of a causal relationship between these alterations is suggested by the significant negative correlation between inotropic and automatic R_max values (Pearson correlation coefficient = − 0.355; N = 205; p < 0.001), when considering the data of the curves in all three experimental groups (Figure 6(B)).

Figure 6  A: Concentration-effect curve for isoproterenol (ISO) determined in the presence of 1 mM caffeine (caff). Inotropic and automatic responses were determined in the same set of control ventricular myocytes (N = 8) and are expressed as in Figure 1. Data are mean and SEM. Gray and black dashed lines indicate the curves for the inotropic and automatic effects of ISO, respectively, in the absence of caffeine. B: Relationship between automatic and inotropic maximum responses (R_max) to agonists determined in a set of 205 myocytes.

Discussion

This study reports for the first time that short, intermittent exposure to a noxious stressor FS may enhance catecholaminergic stimulation of spontaneous activity in rat ventricular myocytes. This change was significantly correlated with decreased inotropic response to stimulation of the β₁-AR signaling pathway. The results indicate involvement of post-receptor mechanisms in the genesis of this alteration, which may limit inotropic responsiveness at high catecholamine levels and increase the susceptibility to catecholamine-dependent ventricular arrhythmias.

In mammalian ventricle, most of the contraction-activating Ca²⁺ pool is released from the SR due to activation of SR Ca²⁺ channels (ryanodine receptors, RyR) by Ca²⁺ that enters the cell during the action potential. It is generally accepted that the positive inotropic effect of β-AR stimulation relies on increased transmembrane Ca²⁺ influx and Ca²⁺ uptake by the SR Ca²⁺-ATPase due to PKA phosphorylation of sarcolemmal L-type Ca²⁺ channels and phospholamban (PLB, a negative regulator of the SR Ca²⁺-ATPase), respectively (ter Keurs and Boyden 2007). The resulting elevation in SR Ca²⁺ content and in L-type Ca²⁺ current peak contributes to increase the fraction of the SR Ca²⁺ content released at a twitch, thus enhancing Ca²⁺ transient and contraction amplitude (Bassani et al. 1995a; Ginsburg and Bers 2004). In addition, positive RyR regulation by PKA and/or Ca²⁺-calmodulin-dependent protein kinase II has also been reported (e.g., Zalk et al. 2007). On the other hand, SR Ca²⁺ overload and RyR hyperphosphorylation also enhance the rate of diastolic SR Ca²⁺ release, which may give rise to spontaneous Ca²⁺ transients and contractions (Bassani et al. 1997; Lukyanenko et al. 1999; ter Keurs and Boyden 2007; Zalk et al. 2007). Because SR Ca²⁺ loading seems to underlie both effects of catecholamines, it is plausible to consider that stimulation of inotropism and spontaneous activity might be mechanistically related. The potency of catecholamines for enhancing spontaneous activity is lower than for increasing inotropism (Boer and Bassani 2004; Carvalho et al. 2006; present results), which suggests that manifestation of automatism requires a certain degree of inotropic stimulation. Accordingly, we observed that the partial β-AR agonist CGP12177 failed at both evoking a substantial inotropic response and stimulating automatism. The crucial role of enhanced SR-cytosol Ca²⁺ cycling in the development of spontaneous activity was confirmed by total ES and SC suppression after the SR function was disabled by thapsigargin, whereas increase of ΔL by catecholamines, albeit smaller, was still visible.

The β₁ subtype comprises 90–95% of the total β-AR population in rat ventricular myocytes (Kitagawa et al. 1995; Ranu et al. 2000). β₁-ARs mediate most of the catecholamine-stimulated cAMP generation in rodent ventricle, producing global increase in cytosolic cAMP concentration, whereas the cAMP elevation evoked by β₂-AR stimulation, when detectable, is much smaller and spatially confined to the subsarcolemmal region (Kitagawa et al. 1995; Nikolaev et al. 2006; Xiao et al. 2006). Accordingly, PLB phosphorylation by PKA at Ser-16 takes place during β₁-AR but not β₂-AR stimulation of mammalian ventricle (Jo et al. 2002; Xiao et al. 2006; DeSantiago et al. 2008). PLB phosphorylation results in enhanced SR Ca²⁺ uptake and consequent increase in SR Ca²⁺ content, which plays an important role in the inotropic response (Li et al. 2002) and seems to be essential for the development of spontaneous activity. Galindo-Tovar and Kaumann (2008) recently showed that the β₁-AR antagonist CGP20712A could suppress extrasystolic activity induced by epinephrine in mouse ventricular myocardium. In line with these observations, our present results show that β₁-AR blockade was able to reverse the increase in automatic R_max to ISO in FS cells. It should be observed that the latter effect was accompanied by elevation of the inotropic R_max, so that, in the presence of MET, inotropic and automatic responsiveness to ISO was comparable in control and FS group myocytes. These results not only indicate that altered responses in FS myocytes are mediated by β₁-ARs, but also imply that these cells are able to develop normal contractile response to catecholamines, provided that the exaggerated spontaneous activity is suppressed.

Depressed inotropic stimulation associated with enhanced spontaneous activity by catecholamines was previously observed by Steinberg et al. (2002) in ventricular myocytes from arrhythmia-prone canines. Venetucci et al. (2006) reported that partial RyR inhibition by tetracaine suppresses spontaneous Ca²⁺ transients at high ISO concentrations, while increasing the amplitude of systolic transients. This may occur because spontaneous release causes transient, partial SR Ca²⁺ depletion (Díaz et al. 1997), which would contribute to decrease systolic SR Ca²⁺ release. In the present study, a significant, negative correlation was revealed between inotropic and automatic maximal responses. It should be noticed that depressed inotropic responses and enhanced automatism were also seen in FS myocytes challenged with FSK, even though these cells could produce “normal” (i.e., similar to control) responses under partial β₁-AR blockade. An implication of these findings is that, if spontaneous activity may limit the inotropic response, the function of the β-AR signaling pathway may be underestimated in conditions associated with enhanced diastolic SR Ca²⁺ leak (e.g., Ca²⁺ overload, ventricular hypertrophy, heart failure; Bassani et al. 1997; Pogwizd et al. 2001; Carvalho et al. 2006; ter Keurs and Boyden 2007; Zalk et al. 2007). Additionally, these results suggest that the depressed contractile response could not be attributed to impaired coupling between β₁-AR and G_s, or of the latter to adenylyl-cyclase.

It has been reported that FS enhances functional β₂-AR expression in the mediation of positive chronotropic and inotropic responses to ISO and epinephrine in rat atria (Bassani and De Moraes 1988; Bassani and Bassani 1993; Santos and Spadari-Bratfisch 2006; A.L. Moura and R.C. Spadari, personal communication). However, we could not detect major effects of β₂-AR blockade on the inotropic or automatic responsiveness to ISO in ventricular myocytes from rats submitted to the same FS protocol as in these previous studies. It thus appears that the FS-induced functional changes in the myocardial β-AR population are chamber-specific. This also seems to be the case for repeated RST, as similar experimental protocols were reported to cause supersensitivity and subsensitivity to chronotropic and inotropic effects of catecholamines, respectively, in rat atria (Brotto 2003; Silveira et al. 2003), but no detectable effects on ventricular myocytes (present results).

An intriguing result was observed in control myocytes exposed to BUT and ICI118551: automatic R_max to ISO was markedly increased, whereas inotropic R_max was decreased by 30%, suggesting that β₂-AR blockade may unmask a strong β₁-AR facilitation of spontaneous activity. If so, β₂-AR occupation might activate signaling pathways functionally antagonistic to the β₁-AR-cAMP-PKA cascade. We have indeed observed attenuation of NE-stimulated spontaneous activity by exposure to a low concentration of β₂-AR agonist (D.C. Boer, J.W.M. Bassani, R.A. Bassani, unpublished observations). It has been proposed that, in addition to the classical G_αs-cAMP-PKA pathway, myocardial β₂-ARs (and possibly β₁-ARs) may be coupled to alternative signaling pathways that lead to phosphatidylinositol-3-kinase (PI3K) activation. Evidence is available that PI3K is involved in the spatial restriction of β₂-AR-PKA signaling to the subsarcolemmal region (which precludes PLB phosphorylation) and inhibition of β₁-AR stimulated enhancement of L-type Ca²⁺ current (Jo et al. 2002; Leblais et al. 2004). Moreover, cAMP compartmentalization and PI3K activation have been associated with cAMP degradation by PDEs, particularly PDE4 with regard to catecholamine inotropic effect and regulation of the SR Ca²⁺-ATPase function (e.g., Patrucco et al. 2004; Xiao et al. 2006; Kerfant et al. 2007; Christ et al. 2009).

FS group myocytes showed altered automatic and inotropic responses to FSK and catecholamines, but not to the non-selective PDE inhibitor IBMX or to increasing [Ca²⁺]_o, which suggests that PKA activation by cAMP, myofilament responsiveness to Ca²⁺ and intrinsic cellular mechanisms involved in enhanced inotropism and spontaneous activity during increased Ca²⁺ availability (e.g., SR Ca²⁺ uptake and release) do not seem to be markedly changed by FS. It is tempting to speculate that FS stress might cause decrease in myocardial PDE activity, possibly involving loss of a putative β₂-AR-mediated PDE stimulation, particularly in the SR Ca²⁺ pump compartment. This would potentiate local cAMP accumulation during β₁-AR stimulation. As a result of greater PKA activation and PLB phosphorylation, increased SR Ca²⁺ loading and leak would lead to heightened spontaneous activity, which would limit further increase in inotropism. This speculation is based on the observation that the development of catecholamine-dependent automatism at both rest and pacing in FS myocytes was similar to that observed in control cells exposed to β₂-AR blockers, and that β₂-AR antagonism failed to further enhance automatic responses in the FS group. It has been shown that PDE4 strongly modulates the inotropic response to β₁-AR activation in rodent ventricle, and that its inhibition potentiates catecholamine-induced contractile and spontaneous activity (Vargas et al. 2006; Galindo-Tovar and Kaumann 2008). Moreover, we were able to reproduce in control cells the qualitative changes in the magnitude of the responses to ISO seen in FS group myocytes by either β₂-AR blockade or PDE inhibition with caffeine. Nevertheless, further experiments are required to investigate this possibility, as well as the possible involvement of PI3K.

Increased diastolic SR Ca²⁺ release during catecholamine stimulation may activate depolarizing Ca²⁺-dependent currents, especially via the Na⁺/Ca²⁺ exchanger, generating abnormal electrical activity that may be propagated and lead to arrhythmias (Fujiwara et al. 2008). This mechanism has been considered to underlie stress- and/or exercise-triggered catecholaminergic polymorphic ventricular tachycardia of genetic etiology, as well as increased susceptibility to sudden death in heart failure and other cardiovascular pathophysiological conditions (Pogwizd et al. 2001; Kontula et al. 2005; ter Keurs and Boyden 2007; Zalk et al. 2007). However, regardless of genetic alterations, a strong association between stress and cardiac arrhythmias has been also recognized under less severe conditions (e.g., Brodsky et al. 1987; Lane et al. 2005). Although more investigation is required to elucidate the underlying mechanisms, our present results are suggestive that the increased automatism in FS myocytes during β-adrenergic stimulation may affect negatively ventricular pumping function by both blunting the inotropic response to catecholamines and increasing the susceptibility to ventricular arrhythmias.

Acknowledgements

This work received partial support from CNPq (Proc N. 300632/2005-3 and 141175/2002-8). We are grateful to Ms. Elizângela S. Oliveira for excellent technical support, to Dr. Pedro X. Oliveira for developing the software used for analysis of cell shortening data, to Dr. Regina C. Spadari for kindly providing CGP12177, and to Dr. José W.M. Bassani for his critical comments on the manuscript.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.