Effects of epidermal growth factor on epinephrine-stimulated heart function in rodents

Department of Biochemistry and Molecular Biology, University of Barcelona, 08028 Barcelona, Spain

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Epidermal growth factor (EGF) interferes with -adrenergic receptor (-AR) signaling in adipocytes and hepatocytes, which leads to decreased lipolytic and glycogenolytic responses, respectively. We studied the effect of EGF on the heart. EGF interfered with the cAMP signal generated by -AR agonists in cardiac myocytes. In perfused hearts, EGF decreased inotropic and chronotropic responses to epinephrine but not to 8-(4-chlorophenylthio)adenosine 3',5'-cyclic monophosphate. Sustained epinephrine infusion induced heart contracture, which resulted in altered heart function as demonstrated by decreased inotropy and increased heart rate variability. EGF prevented all these alterations. In the whole animal (anesthetized mice), EGF administration reduced the rise in heart rate induced by a single epinephrine dose and the occurrence of Bezold-Jarisch reflex episodes induced by repeated doses. Sialoadenectomy enhanced the response to epinephrine, and EGF administration restored normal response. All these results suggest that, by interfering with -AR signaling, EGF protects the heart against the harmful effects of epinephrine.

adenosine 3',5'-cyclic monophosphate; heart rate; sialoadenectomy; rats; mice

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE EPIDERMAL GROWTH FACTOR (EGF) gene is expressed in several tissues (10). In rodents, the highest expression is found in submandibular salivary glands (SMG) (15), which accumulate a large amount of EGF protein. In male mice, EGF accounts for >0.5% of SMG protein (19). In target tissues, EGF is recognized by the EGF receptor (also known as ErbB1), which can dimerize with any member of the ErbB receptor family (62). The EGF receptor recognizes not only EGF but also other members of the EGF family (transforming growth factor-, epiregulin, -cellulin, amphiregulin, and heparin-binding EGF-like growth factor).

Catecholamines stimulate endocrine secretion of EGF from SMG in mice (9, 19). Accordingly, emotional stress (immobilization of the animal) causes a rapid and transient increase in plasma EGF concentration (14). Social stress (experience of defeat in intermale confrontation) induced a much higher and longer-lasting increase in plasma EGF concentration (50). Because catecholamine concentration in plasma remains high during the stress experience and for some time afterwards (52), tissues are exposed to combined stimulation by both catecholamines and EGF.

We reported previously (20, 56-58) that a high, but physiological, EGF concentration interferes with -adrenergic receptor (-AR) signaling and its metabolic consequences both in adipose tissue and in liver. Here we examined the heart because it is another target tissue of catecholamines, in which the most important effects (increase of both inotropy and chronotropy) are mediated by -AR stimulation. However, -AR hyperstimulation or overexpression may cause heart disease (49). The heart is also sensitive to EGF (37, 38, 63). In addition, some cardiovascular effects of EGF were described in several animal species (16, 26, 35). The results reported here support the view that, by interfering with -AR signaling, EGF protects the heart against the harmful effects of epinephrine.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

All experimental procedures using rats and mice were approved by the Committee on Animal Care of the University of Barcelona and by the Autonomous Government of Catalonia.

Cardiac myocyte isolation and incubation. Calcium-tolerant myocytes were isolated from adult rat hearts by previously described methods (47) with minor modifications [collagenase B (Boehringer Mannheim) concentration in perfusion and digestion buffer was 1 mg/ml]. The Joklik medium (GIBCO) used in the preparation of different solutions was supplemented with 25 mM HEPES. Isolated myocytes retained their rod shape and high viability, as determined by the exclusion of Trypan blue dye. Only preparations in which >80% of the cells excluded the dye were incubated further. After isolation, cardiac myocytes were rinsed twice in Joklik's minimal essential medium supplemented with 1.2 mM MgSO₄, 1 mM DL-carnitine, 1.5 mM CaCl₂, and 1% fatty acid-free bovine serum albumin, diluted to 0.4-0.6 × 10⁶ cells/ml, and immediately incubated in a rotating (60 cycles/min) water bath at 37°C under constant oxygenation with 95% O₂-5% CO₂. At indicated times, a sample was obtained and placed onto enough HClO₄ to give a final concentration of 3%. cAMP was determined from deproteinized samples as described previously (57).

Heart perfusion. Male Wistar rats (300-350 g) (Interfauna, Barcelona, Spain) were anesthetized (60 mg/kg pentobarbital sodium prepared in phosphate-buffered saline supplemented with 2.5 mg heparin/ml). Hearts were mounted in a Langendorff apparatus and perfused at 37°C with Krebs-Henseleit solution containing 5 mM glucose and 1 mM carnitine and maintaining a hydrostatic pressure of 60-80 mmHg. The heart was connected to an electronic tension transducer (UF1, Pioden) maintaining a diastolic tension of 1 g. After 15 min of basal recording, epinephrine or EGF was infused through the aortic cannula. The infusion rate was adjusted to obtain a final concentration (after dilution with the perfusion buffer) of 10 µM (epinephrine) or 10 nM (EGF). Some hearts were infused with EGF 5 min before, and then throughout, epinephrine infusion. From the continuous tension record, developed tension (DT) in each 20-s period was calculated (difference between mean maximal and mean minimal tension). The first derivative of the tension register was obtained, and the mean maximal value in 20-s periods was calculated (+dT/dt).

Cardiovascular function in mice. Male Swiss CD-1 mice (40 g; Interfauna) were anesthetized (60 mg/kg pentobarbital sodium), and a cannula (PE 10, Clay-Adams; filled with phosphate-buffered saline supplemented with 100 U heparin/ml) was introduced through the left carotid artery to the aortic arch and connected to an electronic pressure transducer (SensoNor 840). EGF and epinephrine doses were administered through a three-way valve connected to the cannula. From the continuous blood pressure recording, the mean arterial blood pressure (MABP) and heart rate (HR) were calculated for every 30-s period. In some experiments, mice were sialoadenectomized as described previously (19) 1 wk before the experiment. Sham-operated mice were used as control animals in these experiments. Plasma EGF was determined as described previously (19).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of EGF on cardiac myocytes and perfused hearts. EGF increased cAMP concentration in rat cardiac myocytes in a dose-dependent manner (Fig. 1). The effect was moderate (38% increase) compared with the effect of a maximal dose of the -AR agonist isoproterenol (240% increase). These results confirm a previous report by Nair et al. (36). In many cells in which EGF does not modify resting cAMP concentration, it is able, however, to interfere with the cAMP signal generated by several G_s-coupled receptors. This happens in some targets of catecholamine action, adipocytes (57) and hepatocytes (20). Here we show that at 0.5 µM isoproterenol (Fig. 1), the effects of EGF and isoproterenol on cAMP were not additive. At a higher isoproterenol concentration (10 µM), EGF actually decreased the cAMP signal.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Effect of epidermal growth factor (EGF) in rat cardiac myocytes. A: rat cardiac myocytes were incubated without further additions (control) or in the presence of either 100 µM isoproterenol (Iso) or the indicated EGF concentration. After 10 min, samples were obtained to determine cAMP. Results are means ± SE of values in triplicate from a representative experiment. B: cells were incubated with the indicated Iso concentration and either the absence or the presence of EGF. After 10 min, samples were obtained to determine cAMP. Significance of the difference (paired t-test) vs. corresponding no-EGF value: * P < 0.05, ** P < 0.01. Results are means ± SE of 5 experiments.

In keeping with the moderate effect of EGF on cAMP in isolated myocytes, we observed that infusion of EGF into perfused rat hearts also had a moderate and transient effect on heart function (Fig. 2). EGF had a positive inotropic effect (14% over basal +dT/dt) and a negligible effect on chronotropy (2% over basal HR). The consequence of the increased inotropy was the increase in DT (23% over basal DT) and in heart work, as calculated by the tension-rate product (TRP; 19% over basal TRP). As expected, epinephrine infusion had stronger effects than EGF on both inotropy (98% increase) and chronotropy (42% increase). The consequence of both effects was the increase in DT (57% over basal) and TRP (107% over basal).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effect of EGF in perfused rat heart. Perfused rat hearts were maintained for 15 min to record basal function parameters (referred to as 100%). Epinephrine or EGF was then (t0) infused through the aortic cannula at 40 nmol/min (epinephrine) or 40 pmol/min (EGF). From the continuous tension record, developed tension (DT) in each 20-s period was calculated (difference between mean maximal and mean minimal tension). The 1st derivative of the tension function was obtained, and the mean of maximal values in 20-s periods was calculated (+dT/dt). Number of beats in each 20-s period was used to calculate heart rate (HR). TRP, DT-HR product. Results are means ± SE of % of basal value (n = 5).

Figure 3 shows the contractile response to epinephrine of hearts infused with or without EGF. At the start of epinephrine infusion, all recorded parameters were identical in hearts infused with EGF and in control hearts. The maximal inotropic and chronotropic responses to epinephrine (achieved after 20-40 s) were reduced by EGF. Similar results were obtained when hearts were infused with isoproterenol (not shown). In these experiments we started EGF infusion 5 min before epinephrine because we had observed that simultaneous infusion of EGF and epinephrine resulted in a weaker, although still significant, effect on heart response (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effect of EGF on the response of perfused rat hearts to epinephrine. Perfused rat hearts were maintained for 15 min to record basal function parameters (referred to as 100%). EGF (40 pmol/min) or Krebs-Henseleit buffer (KHB; control) was then infused. Five minutes later, epinephrine infusion (40 nmol/min) was started. Results shown correspond to values obtained at the last minute before epinephrine infusion (Before) and at the maximal response after epinephrine infusion (After). Control and EGF group values were compared by Student's t-test. * P < 0.05, ** P < 0.01.

To determine whether the effect of EGF on the contractile response to epinephrine was due to interference with -AR signaling, we studied the effect of EGF on the response of perfused hearts to the cAMP analog 8-(4-chlorophenylthio)adenosine 3',5'-cyclic monophosphate (CPT-cAMP) (Fig. 4). The rise in HR and +dT/dt was slower (maximal response after 80-100 s) than when hearts received epinephrine (see Fig. 2). EGF did not decrease the response to CPT-cAMP. The rise in HR actually increased significantly.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Effect of EGF on the response of perfused rat hearts to 8-(4-chlorophenylthio)adenosine 3',5'-cyclic monophosphate (CPT-cAMP). Perfused rat hearts were maintained for 15 min to record basal function parameters (referred to as 100%). EGF (40 pmol/min) or KHB (control) was then infused. Five minutes later, CPT-cAMP infusion (2.4 µmol/min) was started (defined as zero time). From the continuous tension record, DT in each 20-s period was calculated. The 1st derivative of the tension function was obtained, and the mean of maximal +dT/dt in 20-s periods was calculated. The number of beats in each 20-s period was used to calculate HR. Results are means ± SE of % of basal (preinfusion) value (n = 4). Statistical differences were determined by 2-way ANOVA, and the significance of the treatment (±EGF) factor is shown.

It is known that the sympathetic nervous system is involved in arrhythmogenesis associated with ischemic heart disease (18) and that a high dose of epinephrine may induce arrhythmias in nonischemic hearts (1). To study the effect of EGF on the arrhythmogenic effect of epinephrine, we ran new experiments in which epinephrine infusion lasted for 10 min. After 4 min of epinephrine infusion, the heart became progressively contractured, indicated by a rise in end-diastolic tension (Fig. 5). Simultaneously, there was a pronounced decrease in inotropy (+dT/dt dropped to 50% of initial value) and a sudden increase in HR variability (shown by the large increase in standard error of the mean value). EGF infusion reduced the magnitude of heart contracture and stopped all other alterations in contractility.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of EGF on contractile dysfunction induced by sustained infusion of epinephrine. Perfused rat hearts were maintained for 15 min to record basal function parameters (referred to as 100%). EGF (40 pmol/min) or KHB (control) was then infused. Five minutes later (zero time), epinephrine (Epi) infusion (40 nmol/min) was started. A: end-diastolic tension (EDT) calculated as the mean of minimal tension values in each 30-s period. Results are means ± SE of % of basal (preinfusion) values (n = 5). Statistical differences were determined by 2-way ANOVA, with the significance of the treatment (±EGF) factor shown. B: maximal value of +dT/dt (left) and HR (right) at the indicated time point after epinephrine infusion. Values with and without EGF at each time point were compared by Student's t-test (* P < 0.05). The Levene test indicated no homogeneity of variance of HR in hearts infused without EGF.

Effect of EGF on cardiovascular function in mice. We studied male mice to determine the physiological significance of such an interplay between catecholamines and EGF on the entire animal. Both the effect of exogenously administrated EGF and the function of endogenous EGF can be studied in male mice because they accumulate a huge amount of EGF in their SMG, which is released to both saliva and plasma on adrenergic stimulation (9).

In anesthetized mice, intravenous administration of EGF had a transient (disappeared in ~8 min) and moderate effect on MABP (increased by just 6-8 mmHg), but EGF had no effect on HR (Fig. 6). EGF administration did not alter the rise in MABP induced by epinephrine. Epinephrine increased MABP by 48 ± 4 and 44 ± 4 mmHg in control and EGF-treated mice, respectively (nonsignificant differences) but reduced the rise in HR by ~60% [73 ± 7 and 30 ± 6 beats/min (bpm) in control and EGF-injected mice, respectively (P < 0.001)].

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Effect of EGF on cardiovascular function in anesthetized mice. A cannula connected to a pressure transducer was introduced into the left carotid artery of anesthetized mice, which received a dose of EGF (12 nmol/kg) or saline (control) 20 min before the administration of epinephrine (62.5 nmol/kg). Mean arterial blood pressure (MABP; A) and HR [B; expressed as beats per min (bpm)] were calculated in every 30-s period. Results are means ± SE of 9 (control) or 8 (EGF) mice.

In a further experiment, we administered EGF 20 (as above), 40 or 60 min before epinephrine. The rise in MABP induced by epinephrine (46 ± 6 mmHg) was not significantly affected by EGF administration 20 (33 ± 5 mmHg), 40 (48 ± 5 mmHg) or 60 (45 ± 3 mmHg) min before epinephrine. The rise in HR induced by epinephrine (66 ± 5 bpm) was significantly reduced by EGF administration either 20 or 40 min before epinephrine [28 ± 6 bpm (P < 0.001) and 41 ± 8 bpm (P < 0.05), respectively] but not 60 min before epinephrine [47 ± 9 bpm (nonsignificant difference)].

Strong inotropic stimulation of the heart may induce a powerful depressor reflex originating in the heart itself, known as the Bezold-Jarisch reflex, which is characterized by sudden hypotension and bradycardia (31). We observed such a response after epinephrine administration in some mice (Fig. 7B; compare with the more common regular response shown in Fig. 7A). In the next experiment, we administered repeated and increasing doses of epinephrine to control and EGF-treated mice (Fig. 7C). At the first dose, only one of nine control mice had Bezold-Jarisch reflexes during the 5-min period after epinephrine. This frequency steadily increased with every new dose: after three 20-min-spaced administrations of 62.5 nmol/kg epinephrine, 50% of mice had Bezold-Jarisch reflex. After the fifth dose, this proportion increased to near 100%. EGF administration 20 min before the first epinephrine dose made mice more resistant to the induction of Bezold-Jarisch reflex by epinephrine administration. Thus only 15% of the animals had Bezold-Jarisch reflex episodes after the third dose and <80% after the fifth dose.

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Effect of EGF on epinephrine-induced Bezold-Jarisch reflex. Most animals in the experiment described in Fig. 6 responded to epinephrine with a regular pattern (A). The continuous arterial blood pressure (ABP) record and calculated HR are shown here. The break corresponds to the time of epinephrine administration. Some mice (B) had an irregular response to epinephrine, with brief and repeated periods of sudden hypotension and bradycardia (Bezold-Jarisch reflex). To study the effect of EGF on epinephrine-induced Bezold-Jarisch reflex, mice were treated with EGF (12 nmol/kg) or saline (control; C) (10 animals/group) (C). Twenty minutes later, mice received repeated and progressive doses of epinephrine (A-E) according to the procedure shown at bottom. Percentage of animals having Bezold-Jarisch reflex in the 5-min period after each dose is shown in C. Significance of the difference between control and EGF-treated mice was determined by 2-way ANOVA after appropriate transformation of the frequency value.

The consequence of having Bezold-Jarisch reflex episodes after epinephrine administration was a large increase in HR variability (measured by standard deviation of the HR calculated every 5 s during the 1st min after each epinephrine dose). Thus HR variability was 19 ± 5 bpm after the first dose (A) and increased to 36 ± 4 bpm after dose E in control mice. In EGF-injected mice, it was 11 ± 2 and 22 ± 3 bpm after doses A and E, respectively. HR variability before every dose ranged from 8 to 10 bpm in all animals.

In the last experiment we studied the response of the cardiovascular system to epinephrine in sialoadenectomized (Sialo) mice, with or without a previous dose of EGF, compared with control (sham operated) mice. The response to the first 62.5 nmol/kg dose of epinephrine is shown in Fig. 8A. Although the rise in MABP was similar in Sialo mice, with or without previous administration of EGF (12 nmol/kg) and in control mice, the increase in HR was enhanced in Sialo but not in Sialo+EGF mice. The percentage of mice having Bezold-Jarisch reflex episodes after repeated and increasing epinephrine doses is shown in Fig. 8B. We observed Bezold-Jarisch reflex episodes after the first epinephrine dose in 19% of Sialo mice, a proportion similar to that observed in control mice after the second dose. Administration of EGF to Sialo mice 20 min before the first epinephrine dose returned the percentage of animals that had Bezold-Jarisch reflex episodes to control values. Plasma EGF concentration in awake noncannulated mice was 0.11 ± 0.02 and 0.15 ± 0.02 nM in control and Sialo mice, respectively. At the end of the experiment (10 min after dose E), plasma EGF concentration was 5.64 ± 1.32 nM in control mice (P < 0.05 vs. corresponding awake value), 0.14 ± 0.04 nM in Sialo mice (nonsignificant differences vs. corresponding awake value), and 4.54 ± 1.77 nM in Sialo+EGF mice (P < 0.05 vs. corresponding awake value).

View larger version (25K): [in this window] [in a new window]   Fig. 8.   Effect of sialoadenectomy on the response of the cardiovascular system to epinephrine. One week after surgery, control (sham operated; C) and sialoadenectomized (Sialo; S) mice were anesthetized. A cannula connected to a pressure transducer was then introduced into the left carotid artery to obtain a continuous recording of MABP. Mice received repeated and progressive doses of epinephrine according to the procedure shown in Fig. 7C. A: maximal effect of the 1st epinephrine dose on MABP and HR. Results are means ± SE of 5-8 animals/group. After 1-way ANOVA, comparisons vs. control value were made by Tukey's test; * P < 0.05. B: % of animals having Bezold-Jarisch reflex in the 5-min period after each dose. Significance of the differences were determined by 2-way ANOVA after appropriate transformation of the frequency value. ns, Not significant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Models. This paper combines studies in myocytes isolated from rat hearts to find the effect of EGF on cAMP, in Langendorff-perfused rat hearts to find the functional consequences of these effects, and in anesthetized mice to find the physiological relevance of the ex vivo findings. We observed previously (21, 46, 56, 57) that EGF has similar effects in both liver and adipose tissue of rats and mice. Here, we also observe close corroboration between results obtained from rat preparations and from whole mice. Thus EGF had a negligible effect on HR in Langendorff-perfused rat hearts, and we could not observe any effect on HR in whole mice. In addition, the most remarkable effect described here, the attenuation of the response to epinephrine, was clearly observed in all the systems studied.

The use of pentobarbital-anesthetized mice to study the in vivo physiological significance of the ex vivo findings requires a preliminary consideration. Pentobarbital anesthesia is known to depress heart function and, indeed, HR values in our system were lower than in urethane-anesthetized mice (43) or in conscious mice (24, 30). However, the heart was sensitive to adrenergic stimulation and retained Bezold-Jarisch reflex regulation. The retention of both Bezold-Jarisch and arterial baroreceptor reflexes in pentobarbital-anesthetized rats (33) and cats (25) has been reported. Indeed, to reach a more definitive conclusion on the physiological significance of EGF in cardiovascular regulation, further studies on the acute response to stress in conscious animals are required.

Effects of EGF on otherwise nonstimulated cardiovascular system. Confirming the results published by Patel and coworkers (37, 38, 63), we found that EGF had a moderate effect on cAMP in cardiac myocytes and on contractility in perfused hearts. Later studies from Patel's group showed that purified G_s can directly associate to the juxtamembrane region of purified EGF receptor (54, 55) and that phosphorylation of G_s in tyrosine residues by purified EGF receptor increases the ability to activate adenylyl cyclase (44, 45). The effect of EGF on cAMP accumulation requires the expression of type V adenylyl cyclase (11). Such a requirement may explain why the effect of EGF increasing cAMP is observed only in very few cell types.

We were unable to observe an increase in HR after EGF administration to anesthetized mice, which is in keeping with the results obtained in perfused rat hearts. However, Keiser and Ryan (26) observed that EGF infusion increased HR in conscious rats and monkeys. Several differences in the experimental procedure may explain the differences between our results and those of Keiser and Ryan concerning the effect of EGF on HR in the whole animal. First, although we used an EGF dose (12 nmol/kg, ~72 µg/kg) that is within the range used by Keiser and Ryan (30-300 µg/kg), we administered it in a bolus injection, whereas they infused it for 20 min. Second, and perhaps the most important difference, our experiments used pentobarbital-anesthetized mice and rats, not conscious animals.

In addition to the cAMP-mediated effect on heart function, EGF stimulates arterial contraction in vitro (6, 16, 35). Although there is some controversy concerning the mechanisms involved in the effect of EGF, no studies suggested the involvement of cAMP. We observed a transient and moderate hypertensor effect of EGF in anesthetized mice, which is in keeping with the in vitro studies mentioned above and with results obtained in conscious rats by Keiser and Ryan (26). These authors showed, however, that EGF induced a hypotensor response in conscious monkeys (26).

The physiological relevance of all these effects of EGF has yet to be established. They are observed when systems are stimulated with a much higher EGF concentration than that found in plasma of mice or any other mammal under normal conditions (15). Nevertheless, effects of EGF on the cardiovascular system will have to be considered when exploring any therapeutic use of this peptide.

Effects of EGF on catecholamine-stimulated cardiovascular system. In addition to the effect of EGF on an otherwise nonstimulated cardiovascular system, our results clearly indicate that this peptide decreases the response of both rat and mouse hearts to epinephrine. Several lines of evidence suggest that this effect of EGF is the consequence of the interference with -adrenergic-induced rise of cAMP. First, EGF decreased the effect of a maximal dose of isoproterenol on cAMP in myocytes. Second, EGF decreased the chronotropic and inotropic responses of perfused hearts to epinephrine (or isoproterenol) but not to a cAMP analog. This demonstrates that the target of EGF action in perfused hearts is located upstream of protein kinase A. Very likely, the effect observed in perfused hearts is the consequence of the interference with the cAMP signal observed in isolated myocytes.

This effect of EGF on epinephrine-stimulated hearts is in apparent contradiction with that discussed above. However, it has been shown that association of purified G_s -subunit with purified EGF receptor and phosphorylation of the former in tyrosine residues was decreased by activation of G_s with guanosine 5'-O-(3-thiotriphosphate) (44, 54). Therefore, the stimulatory effect of EGF on G_s proteins can be expected to decrease when G_s proteins are activated by other receptors. It should be noted that we observed less interference with the cAMP signal when myocytes were stimulated with 0.5 µM than with 10 µM isoproterenol.

In addition, mechanisms involved in EGF action may be more complex in whole cells than in reconstituted systems. A variety of effects on cAMP signal generated by several hormones was described in whole cell systems. In A-431 human epidermoid carcinoma cells, in which EGF inhibits cAMP accumulation induced by bradykinin (27), phosphorylation of the G_s -subunit in tyrosine residues by activated EGF receptor leads to reduced guanosine nucleotide exchange (27) and hence inactivation of G_s protein (i.e., the opposite of the effect in the reconstituted system). In other cells, such as gastric mucosa parietal cells, in which EGF inhibits glucagon-like peptide-1-induced acid production, the effect appears to involve the activation of a G_i protein (51). A somewhat different effect of EGF on G_i protein function was described by Tebar et al. (57, 58) in rat adipocytes. In this system EGF appeared to increase the sensitivity of G_s-stimulated adenylate cyclase to the inhibitory effect of G_i proteins.

Studies of mouse hepatocytes, although they do not rule out an effect of EGF at the G protein level, strongly suggest that a phosphodiesterase is the main target of EGF action in this cell type (20). Recent studies from Houslay's lab (3, 22, 29) demonstrate that extracellular signal-regulated kinase (ERK)2, which is activated by the EGF receptor signaling cascade, phosphorylates several isoforms of type 4 phosphodiesterase (PDE4). Furthermore, phosphorylation may increase (PDE4D3 or PDE4D5 isoforms) or decrease (PDE4D1 or PDE4D2 isoforms) phosphodiesterase activity. Thus, depending on the abundance of these isoforms, EGF may have the opposite effect on cAMP in different cell types.

Therefore, it is conceivable that stimulation of G_s proteins by -AR (or other G_s-coupled receptors) abrogates the stimulatory effect of EGF on G_s protein and allows for other effects on G_s or G_i proteins and/or on phosphodiesterases. This hypothesis explains why EGF did not inhibit, but rather enhanced, some of the effects of CPT-cAMP in perfused hearts. Because CPT-cAMP directly activates protein kinase A, any effect of EGF on phosphodiesterases or on the signal transduction from -AR to adenylyl cyclase would not operate. Under these conditions, we would expect EGF actually to increase intracellular cAMP levels and so enhance the effect of CPT-cAMP.

The mechanisms involved in the cross talk between signaling systems have received much attention for more than a decade. cAMP-elevating agents interfere with EGF effects in a variety of cells including several fibroblast cell lines and smooth muscle cells (see Ref. 7 for review). In keeping with this, it was shown that purified catalytic subunit of protein kinase A phosphorylates EGF receptor and this is accompanied by decreased autophosphorylation and a diminished tyrosine kinase activity of the receptor (5). Furthermore, cAMP may interfere with the perhaps most important signaling cascade of the EGF receptor, the activation of ERK1 and ERK2 (12, 28, 60). However, in some cell systems cAMP-elevating agents do not interfere with EGF-induced activation of ERK1 and ERK2 (4, 32) or they can even enhance EGF-dependent activation of these mitogen-activated protein kinases (48, 61). As discussed above, recent studies that show that isoforms with different domain composition may even have an opposite response to certain stimulus illustrate how the cross talk between signaling systems can be cell type specific. This was reviewed recently by Houslay and Kolch (23).

We have not further explored the mechanisms involved in the effect of EGF on cAMP. Rather, we studied the physiological significance of such an effect. It should be noted that the effect of EGF was observed at 10⁸ M. This concentration is within the physiological range of variation of plasma EGF in mice under several stress conditions (14, 39, 50) and similar to that obtained on adrenergic stimulation of SMG (9, 19).

We conclude that the effect of EGF on heart response to epinephrine is physiologically relevant in mice, not only because exogenous EGF administration decreased the effect of epinephrine on HR but also because the lack of endogenous secretion of EGF from SMG (due to sialoadenectomy) enhanced such an effect of epinephrine, which was normalized by previous administration of EGF. These studies were performed in anesthetized mice. New experiments will be required to determine whether the effect of EGF persists in conscious mice. Nevertheless, the results presented here may explain an apparent paradox: catecholamines, through ₁-adrenergic receptors, stimulate the release of EGF to the bloodstream (9, 19), which in turn interferes with -AR mediated responses in heart and other tissues (20, 57). We suggest that the biological significance of such an interference is to protect the heart against harmful effects of intense adrenergic stimulation. We observed in perfused hearts that EGF reduced the contracture and the arrhythmogenic effect of sustained epinephrine infusion. In anesthetized mice, EGF prevented the induction of Bezold-Jarisch reflex by repeated epinephrine doses. Although the physiological significance of the Bezold-Jarisch reflex is not well understood, epinephrine-induced sudden hypotension and bradycardia is indeed a harmful effect of this hormone. In humans this has been linked to syncope in several clinical situations (53).

Many studies have looked at physiological functions of EGF in adult animals, mostly rats or mice. Conclusions were obtained through the administration of EGF or arose from the analysis of the consequences of sialoadenectomy. Thus it was established that EGF accelerates wound repair (8), is involved in male reproductive function (59), and is also involved in female mammary gland development during gestation (40). EGF helps control liver regeneration (34) and protects gastric mucosa against ulcerogenic agents (42). These reports, and many others, tell us about quite long-term (several days) functions of circulating EGF. Therefore, none of them gives a satisfactory explanation of one the most important facts concerning accumulation of EGF in mice SMG, acute secretion on adrenergic stimulation, which results in a rapid and transient rise in plasma EGF concentration (19).

It can be argued that acute increase in plasma EGF concentration after adrenergic stimulation is not a general phenomenon. For instance, it does not occur in female mice (19) and male rats (41). However, the EGF receptor (ErbB1) binds other members of the EGF family (TGF-, amphiregulin, HB-EGF), which act mainly through autocrine and paracrine mechanisms (62). Therefore, it cannot be excluded that the EGF receptor (not because of stimulation with plasma EGF but with other locally produced ligands) may be involved in heart protection in species in which SMG accumulate much less EGF and there is no rise in plasma EGF on adrenergic stimulation. In fact, some roles for HB-EGF in the heart have been reported (2, 17). Very recently, mice with ventricular restricted deletion of Erbb2 (a member of the Erbb gene family that can heterodimerize with ErbB1; Ref. 62) developed dilated cardiomyopathy (13). All these recent reports indicate that ErbB receptors and their ligands have a function in long-term protection of the heart. Our results indicate that one of these ligands can also protect the heart against the harmful effects of catecholamines in the short term. Therefore, it will be worth exploring the usefulness of EGF, or other Erb1 ligands, as protective agents against ischemia-induced heart disease, given that local release of catecholamines is involved in ischemia-induced arrhythmias and heart failure (18).

	ACKNOWLEDGEMENTS

The authors express gratitude to Robin Rycroft for editorial help.

	FOOTNOTES

^* J. Lorita and N. Escalona contributed equally to this work.

This study was supported by Grant PB97-0936 from the Dirección General de Enseñanza Superior e Investigación Científica, Ministerio de Educación y Ciencia, Spain.

Address for reprint requests and other correspondence: I. Ramírez, Dept. of Biochemistry and Molecular Biology, Univ. of Barcelona, Avda. Diagonal 645, 08028 Barcelona, Spain (E-mail: sunyer{at}bio.ub.es).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

July 18, 2002;10.1152/ajpheart.00217.2002

Received 13 March 2002; accepted in final form 8 July 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

1.   Antalóczy, Z. Role of catecholamines in stress-induced heart disease. In: Pathogenesis of Stress-Induced Heart Disease, edited by Beamish RE, Panagia V, and Dhalla NS.. Boston, MA: Nijhoff, 1985, p. 213-227.

2.   Asakura, M, Kitakaze M, Takashima S, Liao Y, Ishikura F, Yoshinaka T, Ohmoto H, Node K, Yoshino K, Ishiguro H, Asanuma H, Sanada S, Matsumura Y, Takeda H, Beppu S, Tada M, Hori M, and Higashiyama S. Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metalloproteinase inhibitors as a new therapy. Nat Med 8: 35-40, 2002[ISI][Medline].

3.   Baillie, GS, MacKenzie SJ, McPhee I, and Houslay MD. Sub-family selective actions in the ability of Erk2 MAP kinase to phosphorylate and regulate the activity of PDE4 cyclic AMP-specific phosphodiesterases. Br J Pharmacol 131: 811-819, 2000[ISI][Medline].

4.   Baptist, M, Dumont JE, and Roger PP. Intercellular heterogeneity of early mitogenic events: cAMP generalizes the EGF effect on c-Fos protein appearance but not on MAP kinase phosphorylation and nuclear translocation in dog thyroid epithelial cells. Exp Cell Res 221: 160-171, 1995[ISI][Medline].

5.   Barbier, AJ, Poppleton HM, Yigzaw Y, Mullenix JB, Wiepz GJ, Bertics PJ, and Patel TB. Transmodulation of epidermal growth factor receptor function by cyclic AMP-dependent protein kinase. J Biol Chem 274: 14067-14073, 1999[Abstract/Free Full Text].

6.   Berk, BC, Brock TA, Webb RC, Taubman MB, Atkinson WJ, Gimbrone MA, Jr, and Alexander RW. Epidermal growth factor, a vascular smooth muscle mitogen, induces rat aortic contraction. J Clin Invest 75: 1083-1086, 1985[ISI][Medline].

7.   Bornfeldt, KE, and Krebs EG. Crosstalk between protein kinase A and growth factor receptor signaling pathways in arterial smooth muscle. Cell Signal 11: 465-477, 1999[ISI][Medline].

8.   Buckley, A, Davidson JM, Kamerath CD, Wolt TB, and Woodward SC. Sustained release of epidermal growth factor accelerates wound repair. Proc Natl Acad Sci USA 82: 7340-7344, 1985[Abstract/Free Full Text].

9.   Byyny, RL, Orth DN, Cohen S, and Doyne ES. Epidermal growth factor: effects of androgen and adrenergic agents. Endocrinology 95: 776-782, 1974[ISI][Medline].

10.   Carpenter, G, and Wahl MI. The epidermal growth factor family. In: Peptide Growth Factors and Their Receptors, edited by Sporn MB, and Roberts AB.. Berlin: Springer, 1990, vol. I, p. 69-171.

11.   Chen, ZT, Nield HS, Sun H, Barbier A, and Patel TB. Expression of type V adenylyl cyclase is required for epidermal growth factor-mediated stimulation of cAMP accumulation. J Biol Chem 270: 27525-27530, 1995[Abstract/Free Full Text].

12.   Cook, SJ, and McCormick F. Inhibition by cAMP of Ras-dependent activation of Raf. Science 262: 1069-1072, 1993[Abstract/Free Full Text].

13.   Crone, SA, Zhao YY, Fan L, Gu Y, Minamisawa S, Liu Y, Peterson KL, Chen J, Kahn R, Condorelli G, Ross J, Jr, Chien KR, and Lee KF. ErbB2 is essential in the prevention of dilated cardiomyopathy. Nat Med 8: 459-465, 2002[ISI][Medline].

14.   Fernández-Varó, G, Mena MP, Arnau A, Sánchez O, Soley M, and Ramírez I. Immobilization stress induces c-Fos accumulation in liver. Cell Stress Chaperones 5: 306-312, 2000[ISI][Medline].

15.   Fisher, DA, and Lakshmanan J. Metabolism and effects of epidermal growth factor and related growth factor in mammals. Endocr Rev 11: 418-442, 1990[Abstract].

16.   Florian, JA, and Watts SW. Epidermal growth factor: a potent vasoconstrictor in experimental hypertension. Am J Physiol Heart Circ Physiol 276: H976-H983, 1999[Abstract/Free Full Text].

17.   Fujino, T, Hasebe N, Fujita M, Takeuchi K, Kawabe J, Tobise K, Higashiyama S, Taniguchi N, and Kikuchi K. Enhanced expression of heparin-binding EGF-like growth factor and its receptor in hypertrophied left ventricle of spontaneously hypertensive rats. Cardiovasc Res 38: 365-374, 1998[Abstract/Free Full Text].

18.   Goldstein, DS. Stress, Catecholamines, and Cardiovascular Disease. Oxford, UK: Oxford Univ. Press, 1995, p. 366-370.

19.   Grau, M, Rodríguez C, Soley M, and Ramírez I. Relationship between epidermal growth factor in mouse submandibular glands, plasma, and bile: effects of catecholamines and fasting. Endocrinology 135: 1854-1862, 1994[Abstract].

20.   Grau, M, Soley M, and Ramírez I. Interaction between adrenaline and epidermal growth factor in the control of liver glycogenolysis in mouse. Endocrinology 138: 2601-2609, 1997[Abstract/Free Full Text].

21.   Grau, M, Tebar F, Ramírez I, and Soley M. Epidermal growth factor administration decreases liver glycogen and causes mild hyperglycaemia in mice. Biochem J 315: 289-293, 1996[Medline].

22.   Hoffmann, R, Baillie GS, MacKenzie SJ, Yarwood SJ, and Houslay MD. The MAP kinase ERK2 inhibits the cyclic AMP-specific phosphodiesterase HSPDE4D3 by phosphorylating it at Ser579. EMBO J 18: 893-903, 1999[ISI][Medline].

23.   Houslay, MD, and Kolch W. Cell-type specific integration of cross-talk between extracellular signal-regulated kinase and cAMP signaling. Mol Pharmacol 58: 659-668, 2000[Free Full Text].

24.   Janssen, BJA, Leenders PJA, and Smits JFM Short-term and long-term blood pressure and heart rate variability in the mouse. Am J Physiol Regul Integr Comp Physiol 278: R215-R225, 2000[Abstract/Free Full Text].

25.   Kawada, T, Yamazaki T, Akiyama T, Shishido T, Inagaki M, Uemura K, Miyamoto T, Sugimachi M, Takaki H, and Sunagawa K. In vivo assessment of acetylcholine-releasing function at cardiac vagal nerve terminals. Am J Physiol Heart Circ Physiol 281: H139-H145, 2001[Abstract/Free Full Text].

26.   Keiser, JA, and Ryan MJ. Hemodynamic effects of epidermal growth factor in conscious rats and monkeys. Proc Natl Acad Sci USA 93: 4957-4961, 1996[Abstract/Free Full Text].

27.   Liebmann, C, Graness A, Boehmer A, Kovalenko M, Adomeit A, Steinmetzer T, Nürnberg B, Wetzker R, and Boehmer FD. Tyrosine phosphorylation of G_s and inhibition of bradykinin-induced activation of the cyclic AMP pathway in A431 cells by epidermal growth factor receptor. J Biol Chem 271: 31098-31105, 1996[Abstract/Free Full Text].

28.   Liu, YF, Jakobs KH, Rasenick MM, and Albert PR. G protein specificity in receptor-effector coupling. Analysis of the roles of G_o and G_i2 in GH4C1 pituitary cells. J Biol Chem 269: 13880-13886, 1994[Abstract/Free Full Text].

29.   MacKenzie, SJ, Baillie GS, McPhee I, Bolger GB, and Houslay MD. ERK2 mitogen-activated protein kinase binding, phosphorylation, and regulation of the PDE4D cAMP-specific phosphodiesterasesthe involvement of COOH-terminal docking sites and NH₂-terminal UCR regions. J Biol Chem 275: 16609-16617, 2000[Abstract/Free Full Text].

30.   Madeddu, P, Salis MB, Milia AF, Emanueli C, Guerrini R, Regoli D, and Calò G. Cardiovascular effects of nociceptin in unanesthetized mice. Hypertension 33: 914-919, 1999[Abstract/Free Full Text].

31.   Mark, AL. The Bezold-Jarisch reflex revisited: clinical implications of inhibitory reflexes originating in the heart. J Am Coll Cardiol 1: 90-102, 1983[Abstract].

32.   McCawley, LJ, Li SN, Benavidez M, Halbleib J, Wattenberg EV, and Hudson LG. Elevation of intracellular cAMP inhibits growth factor-mediated matrix metalloproteinase-9 induction and keratinocyte migration. Mol Pharmacol 58: 145-151, 2000[Abstract/Free Full Text].

33.   McQueen, DS, Bond SM, Moores C, Chessell I, Humphrey PPA, and Dowd E. Activation of P2X receptors for adenosine triphosphate evokes cardiorespiratory reflexes in anaesthetized rats. J Physiol 507: 843-855, 1998[Abstract/Free Full Text].

34.   Michalopoulos, GK, and Defrances MC. Liver regeneration. Science 276: 60-66, 1997[Abstract/Free Full Text].

35.   Muramatsu, I, Hollenberg MD, and Lederis K. Vascular actions of epidermal growth factor-urogastrone: possible relationship to prostaglandin production. Can J Physiol Pharmacol 63: 994-999, 1985[Medline].

36.   Nair, BG, Parikh B, Milligan G, and Patel TB. Gs mediates epidermal growth factor-elicited stimulation of rat cardiac adenylate cyclase. J Biol Chem 265: 21317-21322, 1990[Abstract/Free Full Text].

37.   Nair, BG, Rashed HM, and Patel TB. Epidermal growth factor stimulates rat cardiac adenylate cyclase through a GTP-binding regulatory protein. Biochem J 264: 563-571, 1989[ISI][Medline].

38.   Nair, BG, Rashed HM, and Patel TB. Epidermal growth factor produces inotropic and chronotropic effects in rat hearts by increasing cyclic AMP accumulation. Growth Factors 8: 41-48, 1993[ISI][Medline].

39.   Nexo, E, Hollenberg MD, and Bing J. Aggressive behaviour in mice provokes a marked increase in both plasma epidermal growth factor and renin. Acta Physiol Scand 111: 367-371, 1981[ISI][Medline].

40.   Okamoto, S, and Oka T. Evidence for physiological function of epidermal growth factor: pregestational sialoadenectomy of mice decreases milk production and increases offspring mortality during lactation period. Proc Natl Acad Sci USA 81: 6059-6063, 1984[Abstract/Free Full Text].

41.   Olsen, PS, Kirkegaard P, Poulsen SS, and Nexo E. Adrenergic effects on renal secretion of epidermal growth factor in the rat. Regul Pept 11: 17-25, 1985[ISI][Medline].

42.   Olsen, PS, Poulsen SS, Kirkegaard P, and Nexo E. Role of submandibular saliva and epidermal growth factor in gastric cytoprotection. Gastroenterology 87: 103-108, 1984[ISI][Medline].

43.   Peotta, VA, Vasquez EC, and Meyrelles SS. Cardiovascular neural reflexes in L-NAME-induced hypertension in mice. Hypertension 38: 555-559, 2001[Abstract/Free Full Text].

44.   Poppleton, H, Sun H, Fulgham D, Bertics P, and Patel TB. Activation of Gs by the epidermal growth factor receptor involves phosphorylation. J Biol Chem 271: 6947-6951, 1996[Abstract/Free Full Text].

45.   Poppleton, HM, Sun H, Mullenix JB, Wiepz GJ, Bertics PJ, and Patel TB. The juxtamembrane region of the epidermal growth factor receptor is required for phosphorylation of G_s. Arch Biochem Biophys 383: 309-317, 2000[ISI][Medline].

46.   Quintana, I, Grau M, Moreno F, Soler C, Ramírez I, and Soley M. The early stimulation of glycolysis by epidermal growth factor in isolated rat hepatocytes is secondary to the glycogenolytic effect. Biochem J 308: 889-894, 1995[ISI][Medline].

47.   Ramírez, I, Kryski AJ, Ben-Zeev O, Schotz MC, and Severson DL. Characterization of triacylglycerol hydrolase activities in isolated myocardial cells from rat heart. Biochem J 232: 229-236, 1985[ISI][Medline].

48.   Roberson, MS, Ban M, Zhang T, and Mulvaney JM. Role of the cyclic AMP response element binding complex and activation of mitogen-activated protein kinases in synergistic activation of the glycoprotein hormone  subunit gene by epidermal growth factor and forskolin. Mol Cell Biol 20: 3331-3344, 2000[Abstract/Free Full Text].

49.   Rockman, HA, Koch WJ, and Lefkowitz RJ. Seven-transmembrane-spanning receptors and heart function. Nature 415: 206-212, 2002[Medline].

50.   Sánchez, O, Arnau A, Pareja M, Poch E, Ramírez I, and Soley M. Acute stress-induced tissue injury in mice: differences between emotional and social stress. Cell Stress Chaperones 7: 36-46, 2002[ISI][Medline].

51.   Schmidtler, J, Dehne K, Schusdziarra V, Classen M, and Schepp W. Pertussis toxin-sensitive inhibition of glucagon-like peptide 1-stimulated acid production by epidermal growth factor and transforming growth factor  in rat parietal cells. Eur J Pharmacol 246: 59-66, 1993[ISI][Medline].

52.   Sgoifo, A, De Boer SF, Westenbroek C, Maes FW, Beldhuis H, Suzuki T, and Koolhaas JM. Incidence of arrhythmias and heart rate variability in wild-type rats exposed to social stress. Am J Physiol Heart Circ Physiol 273: H1754-H1760, 1997[Abstract/Free Full Text].

53.   Somers, VK, and Abboud FM. Neurocardiogenic syncope. Adv Intern Med 41: 399-435, 1996[ISI][Medline].

54.   Sun, H, Chen ZT, Poppleton H, Scholich K, Mullenix J, Weipz GJ, Fulgham DL, Bertics PJ, and Patel TB. The juxtamembrane, cytosolic region of the epidermal growth factor receptor is involved in association with -subunit of G_s. J Biol Chem 272: 5413-5420, 1997[Abstract/Free Full Text].

55.   Sun, H, Seyer JM, and Patel TB. A region in the cytosolic domain of the epidermal growth factor receptor antithetically regulates the stimulatory and inhibitory guanine nucleotide-binding regulatory proteins of adenylyl cyclase. Proc Natl Acad Sci USA 92: 2229-2233, 1995[Abstract/Free Full Text].

56.   Tebar, F, Grau M, Mena MP, Arnau A, Soley M, and Ramírez I. Epidermal growth factor secreted from submandibular salivary glands interferes with the lipolytic effect of adrenaline in mice. Endocrinology 141: 876-882, 2000[Abstract/Free Full Text].

57.   Tebar, F, Ramírez I, and Soley M. Epidermal growth factor modulates the lipolytic action of catecholamines in rat adipocytes. Involvement of a Gi protein. J Biol Chem 268: 17199-17204, 1993[Abstract/Free Full Text].

58.   Tebar, F, Soley M, and Ramírez I. The antilipolytic effects of insulin and epidermal growth factor in rat adipocytes are mediated by different mechanisms. Endocrinology 137: 4181-4188, 1996[Abstract].

59.   Tsutsumi, O, Kurachi H, and Oka T. A physiological role of epidermal growth factor in male reproductive function. Science 233: 975-977, 1986[Abstract/Free Full Text].

60.   Wu, J, Dent P, Jelinek T, Wolfman A, Weber MJ, and Sturgill TW. Inhibition of the EGF-activated MAP kinase signalling pathway by adenosine 3'-5'-monophosphate. Science 262: 1065-1069, 1993[Abstract/Free Full Text].

61.   Yao, H, Labudda K, Rim C, Capodieci P, Loda M, and Stork PJS Cyclic adenosine monophosphate can convert epidermal growth factor into a differentiating factor in neuronal cells. J Biol Chem 270: 20748-20753, 1995[Abstract/Free Full Text].

62.   Yarden, Y, and Sliwkowski MX. Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2: 127-137, 2001[ISI][Medline].

63.   Yu, YM, Nair BG, and Patel TB. Epidermal growth factor stimulates cAMP accumulation in cultured rat cardiac myocytes. J Cell Physiol 150: 559-567, 1992[ISI][Medline].