- Email: [email protected]
Electrophysiological effects of the aqueous extract of Averrhoa carambola L. leaves on the guinea pig heart
ARTICLE IN PRESS
Phytomedicine 13 (2006) 501–508 www.elsevier.de/phymed
Electrophysiological effects of the aqueous extract of Averrhoa carambola L. leaves on the guinea pig heart$ C.M.L. Vasconcelosa,b, M.S. Arau´joa, E.A. Conde-Garciaa, a Cardiobiophysics Research Laboratory, Department of Physiology, Center of Biological and Health Sciences, Universidade Federal de Sergipe, Aracaju, Sergipe, Brazil b Laborato´rio de Tecnologia Farmaceˆutica, Universidade Federal da Paraı´ba, Joa˜o Pessoa, Paraı´ba, Brazil
Received 17 December 2004; accepted 27 January 2005
Abstract This work aims to describe some electrophysiological changes promoted by the aqueous extract (AEx) from Averrhoa carambola leaves in guinea pig heart. The experiments were carried out on isolated heart or on right atrium–ventricle preparations. In 6 hearts, the extract induced many kinds of atrioventricular blocks (1st, 2nd, and 3rd degrees); increased the QT interval from 229723 to 264719 ms; increased the QRS complex duration from 2773.1 to 59711 ms, and depressed the cardiac rate from 136717 to 89714 bpm. Furthermore, it decreased the conduction velocity of atrial impulse (1773%); reduced the intraventricular pressure (8676%), and increased the conduction time between the right atrium and the His bundle (2776.5%). The conduction time from the His bundle to the right ventricle was not altered. Atropine sulfate did not change either the electrocardiographic parameters or the intraventricular pressure effects promoted by the A. carambola AEx. Based on these results, the popular use of such extracts should be avoided because it can promote electrical and mechanical changes in the normal heart. r 2005 Elsevier GmbH. All rights reserved. Keywords: Averrhoa carambola L.; Guinea pig atrium; Myocardium; ECG; His electrogram
Introduction Herbal medicine is an increasingly common form of alternative therapy worldwide. In 1997, it was estimated that 12.1% of adults in the United States used herbal $ Research supported by Centrais Ele´tricas Brasileiras (ELETROBRAS, Process No. 23113.009351/03-67), Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sergipe (FAP-SE/FUNTEC FNS/Process No. 01/2003), Universidade Federal de Sergipe, and by the Brazilian Conselho Nacional do Desenvolvimento Cientı´ fico e Tecnolo´gico CNPq. Corresponding author. Tel.: +55 79 212 6642; fax: +55 79 246 3377. E-mail address: [email protected] (E.A. Conde-Garcia).
0944-7113/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2005.01.013
medicine in the previous 12 months, representing an outof-pocket payment of US$ 5.1 billion (Eisenberg et al., 1998). Averrhoa carambola L., known as star fruit (‘carambola’ in Brazil), belongs to the Oxalidaceae family. According to folk medicine, it is commonly used for treating restlessness, headache, nausea, and cough (Burkhill, 1935). The following symptoms have been reported after fruit ingestion: intractable hiccups, insomnia, mental confusion, and even death in patients who suffered from renal chronic failure (Martin et al., 1993; Neto et al., 1998; Chang et al., 2000). Cecchini et al. (1999) and Carolino et al. (2001) reported the characterization of a potent neurotoxin, isolated from A. carambola fruits. It is a water-soluble molecule that
ARTICLE IN PRESS 502
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
could induce convulsions and L-glutamate release from rat synaptosomes. Furthermore, infusions prepared from dry A. carambola leaves reduced the glucose levels in Wistar rat blood (Martha et al., 2000). Nevertheless, no mention could be found in the specialized literature (Medline/Pubmed: http://www.ncbi.nih.gov/entrez/ query.fcgi and Medline/Bireme – http://www.bireme.br, covered period from 1960 to 2004, seeking date: January 7, 2005) dealing with the A. carambola myocardial effects. We observed that its leaf aqueous extract (AEx) depressed both the cardiac spontaneous rate and guinea pig left atrium contractility (Vasconcelos and CondeGarcia, 2002; Vasconcelos et al., 2001). The present paper describes some electrocardiographic and electrophysiological effects of A. carambola AEx on guinea pig heart.
Material and methods Plant characterization Leaves were collected near the Federal University of Sergipe campus (Aracaju, Sergipe, Brazil) from agrotoxic-free and fumigation-free trees, during the winter season (June–July 2001). Plant identification was made by comparson with voucher specimen no. 24720 deposited in the Herbarium of the Federal University of Pernambuco, Recife, Brazil.
Aqueous extract preparation The AEx was obtained in a Soxhlet apparatus by extracting dry leaves with the following solvents (P.A.A.C.S): hexane, chloroform, acetone, ethanol, methanol, and water. Each extract was concentrated in a rotative evaporator (BUCHII RE 111, Buchi Laboratoriums-Technik AG, Flawil, Schweiz) and stored at 2773 1C in a dry atmosphere without light protection until (storage time: 1–6 months).
Phytochemistry screening Phytochemical screening was performed on AEx according to the technique proposed by Domı´ nguez (1973).
Isolated heart experimental assembly Guinea pigs (Cavia porcellus) of both sexes (300–500 g each) were sacrificed by cervical stroke 30 min after a subcutaneous administration of heparin (1000 IU, Liquemine, Roche, Sa˜o Paulo, SP, Brazil). The animal chest was opened, the heart carefully removed and mounted on a constant-flow (Milan Peristaltic Pump,
Milan Equipamentos Cientı´ ficos Ltda., Curitiba, PR, Brazil) aortic perfusion system (Langendorff technique, Do¨ring, 1990). The heart was perfused by a modified Tyrode solution (NaCl 137.0, KCl 5.0, MgCl2 0.5, NaHCO3 12.0, CaCl2 1.8, Glucose 6.0, NaH2PO4 1.8, in mM, substances of analytical grade purchased from Merck S.A. Indu´strias Quı´ micas, Rio de Janeiro, Brazil), oxygenated and buffered by carbogen mixture (95% O2+5% CO2, error less then 0.2%, purchased from Aga S.A., Sa˜o Paulo, Brazil or White Martins Gases Industriais S.A. Sa˜o Paulo, Brazil), kept at 3470.1 1C (Haake F3, Berlin, Germany), and filtered through a cellulose acetate membrane (0.45 mm) to prevent microembolia (Harrison and Raymond, 1951). When the heart was electrically driven, suprathreshold DC current pulses, isolated from the ground were used (Anapulse Stimulator 302-T, WPI Instruments, Inc., 60 Fitch Street, POX 3110, New Haven, Connecticut 06515, USA; Digitimer D4030, Digitimer DS2, Digitimer Limited, Tewin Road, Welwyn Garden City, Hertfordshire, England). The stimuli were delivered through a pair of stainless steel electrodes connected to the right atrium appendage. The whole preparation was maintained immersed in Tyrode (50 ml) in which three electrodes (Ag/AgCl/NaCl 1 M) were placed for sensing the heart electrical signal, the electrocardiogram (ECG) to be recorded. Those signals were amplified (HP8811B, HP7754A, HP7754B, Hewlett-Packard, Chicago, IL, USA) and stored in a computer for off-line processing (DI-205, DI-400, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA). Before beginning the experimental procedures, the biological preparation was allowed to stabilize for 1–2 h. To investigate whether the parasympathetic pathways were involved in the cardiodepressor effects of AEx, some hearts were atropinized by adding atropine sulfate (10 mM) to the perfusion fluid 20 min before starting the test with the AEx.
Heart rate measurements For studying the effect of AEx on cardiac rate, the isolated heart was allowed to beat spontaneously. The ECG was recorded in a computer and the heart rate was determined, beat-to-beat, from the stored data. To accomplish that, successive R–R intervals during control, test (AEx, 1 g/l), and washout, were determined (Calculate and Windaex softwares from Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA).
Intraventricular pressure Left intraventricular pressure was measured using a water-filled balloon. This device, coupled to a pressure transducer (HP 1290A, HP8805B), sent its signals to an
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
analog-to-digital (A/D) converter running at 512 samples/s (DI-205, DI-400, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA). The pressure, recorded and stored in a computer, was processed off-line. A column of mercury was used to calibrate the system used to measure intraventricular pressure.
Heart wall movements The inotropic effect of AEx was documented by recording both intraventricular pressure and heart wall movements. These experiments were performed on isolated guinea pig hearts (constant flow Langendorf perfusion, 3470.1 1C) artificially paced (150–176 bpm). A red laser beam (diameter: 4 2 mm) was positioned to pass tangentially to the left ventricle wall. The heart movements altered the laser beam area that emerged from the ventricular edge. This variable beam was then caught by a photoresistor (Vactec VT 732E 77 29, EG & G Vactec, 200 Orchard Ridge Drive, Suit 100, Gaithersburg, MD 20 878, USA) mounted within a plastic black cylinder (25 mm long). In the cylinder base, a small hole (diameter: 2 mm) allowed the laser beam to enter and to thus reach the photoresistor. This device controlled the current intensity of a DC circuit fed by a 9 V battery. The larger the laser spot on the hole of the cylinder base, the more intense the transmitted laser beam to the photoresistor detector. The electric current in the detection circuit was monitored by recording the voltage drop across a 100 kO resistor. The voltage changes were digitized and recorded on a computer, as described previously.
His bundle electrogram and atrial impulse velocity To study both the impulse conduction through the atrioventricular (AV) node and the atrial impulse velocity, the right atrium and the right ventricle were mounted in an organ chamber. There, the biological preparation, laid and pinned on the chamber paraffin bed with its endocardium surface facing upwards, was superfused by Tyrode solution (3470.1 1C, 15 ml) buffered and oxygenated with carbogen mixture (oxygen 95%, carbon dioxide 5%, error less than 0.2%). The preparation was stimulated electrically by means of two stainless steel electrodes (DC pulses, 40 V, 1 ms). To detect the atrial and the His bundle electrical waves, two surface electrodes (Ag/Teflon) were placed on the atrial and on the His bundle, near the coronary sinus. The distance between the two surface electrodes was determined optically with the aid of a calibrated stereomicroscope (Wild Heerbrugg AG M5A, Juerg Dedual Gaebrisstrasse 8 CH-9056, Gais, Switzerland). The electrical signals were amplified and monitored on
503
an oscilloscope screen (Tektronix D44 Dual Beam, 5A22N Differential Amplifier, 5A14N Four Channel Amplifier, 5B44 Dual Time Base, Tektronix, Inc., Beaverton, OR 97005, USA). They were also digitized by an A/D converter (DI 400, DI 205, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA) and stored in a computer (sampling rate: 2.048 kHz). The time intervals for the impulse propagation from the atrium to the His bundle (A–H), as well as from the His bundle to the right ventricle (H–V) were measured; two electronic gates were generated. Each of them triggered on and off a 100 kHz pulse train, and an electronic setup determined the gate durations by counting the pulses within them (Digitimer D4030, WPI 1830, 1831, 1832, Digitimer Limited, Tewin Road, Welwyn Garden City, Hertfordshire, England). This procedure allowed the A–H and H–V conduction times to be measured with a precision of 710 ms.
Data processing The Student’s t-test for independent or dependent samples was used to decide upon the differences between two population means by adopting a significance level of 5% (Statistica for Windows). The results presented in this paper are expressed as means7s.e.m.
Results The phytochemical screening of AEx revealed the following constituents: a saponin [foam ()]; an alkaloid [Bouchardat (), Mayer (), Dragendorff (), tungstosilicic acid ()]; steroids (+); tannins [gelatin (+), FeCl3 (+++)]; and a flavonoid [magnesium tape (++), fluorescence (+++)]. Fig. 1 (upper panel) shows the electrocardiographic changes promoted by AEx (1 g/l) on isolated guinea pig heart. AEx increased the PR interval (PRi) from 102 to 166 ms characterizing a first-degree AV block (Fig. 1B). The extract also delayed the ventricular repolarization. This could be observed by the increase of the QT interval (QTi) from 215 to 268 ms, and by the increase of the QRS complex duration from 33 to 71 ms (Fig. 1B). These changes, however, disappeared during washout (Fig. 1C). Similar results were observed in 6 other hearts (lower panel) in which the AEx (1 g/l) increased the PRi from 107712 to 173729 ms (61%, n ¼ 14 trials, po0:001), QTi from 229723 to 264719 ms (15%, n ¼ 12 trials, po0:001), and QRS complex duration from 2773.1 to 59711 ms (118%, n ¼ 11 trials, po0:001). The control values for PRi, QTi, and QRS duration could be recovered during washout. Fig. 2 shows an example of AEx effect (1 g/l) on: (a) the intraventricular pressure (upper panel), (b) the heart
ARTICLE IN PRESS 504
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Fig. 1. Effects of A. carambola aqueous extract (1 g/l) on the electrocardiogram of guinea pig isolated heart. Upper panel: (A) control; (B) test by adding extract to the organ bath; and (C) washout. The extract reversibly increased: (1) PR interval (dotted line), (2) QT interval (solid line), and (3) QRS complex duration. S: stimulus artifact. Values are in milliseconds. Lower panel: results obtained in 6 other guinea pig isolated hearts (3470.1 1C, 120–150 bpm, po0:001, stimuli: 30 V, 1 ms).
wall movement (middle panel), and (c) on the ECG (lower panel). The experiment was performed on an artificially paced (150 bpm) guinea pig heart. The figure shows recording obtained at different moments: during control (control) and after perfusing with AEx (190 and 220 s). At 190 s, the intraventricular pressure was reduced from 64 to 25 mmHg (60%) and after 220 s it declined to zero. Note that the optical signal (middle panel) disappeared at 190 s. In spite of these huge mechanical effects, even after 220 s of adding the extract to the perfusion fluid, the ECG could be recorded, but showed the following parameter increase: (1) PRi (from 96 to 220 ms), (2) QRS duration (from 33 to 108 ms), and (3) QTi (from 267 to 337 ms). During washout PRi, QRS, and QTi were 96, 33, and 268 ms, respectively. In 6 hearts, the AEx reduced about 8676% the intraventricular pressure. AEx (1 g/l) depressed impulse conduction through the AV node (Fig. 3). An example of a first-degree AV block is shown and described in Fig. 1. Second-degree AV block, characterized by an intermittent conduction through the AV node (Fig. 3A), third-degree AV block – mirroring a compete impairment of the AV nodal conduction (Fig. 3B), and the Wenckbach phenomenon, in which the PRi progressively increased until a complete AV block (Fig. 3C) could also be recorded. These effects disappeared promptly during washout. Disturbances in the AV nodal conduction could be recorded in all tested hearts.
Fig. 2. Electromechanical uncoupling promoted by A. carambola aqueous extract (AEx, 1 g/l). Upper panel: superimposed intraventricular pressure curves; middle panel: curves related to heart wall movement; lower panel: electrocardiogram of artificially paced guinea pig isolated heart (150 bpm). Numbers without units stand for time intervals, measured in seconds, after beginning extract perfusion. Horizontal bars correspond to 100 ms. Periodical waves seen at the right side of the upper and middle panels correspond to the pressure and heart movement curves observed during washout. Note that 190 s after the extract, the optical signal related to the heart wall movement ceased and 220 s after the extract was administered the intraventricular pressure drops to zero. In spite of these effects, electrocardiogram waves were not abolished, but the PRi, QRS, and QTi were increased. During washout (15 min), PRi, QRS, and QTi recovered their control values. S stands for stimulus artifact (3470.1 1C, stimulus: 40 V, 1 ms).
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
505
A-H conduction (percent of control)
200
* 100
complete AV block 0
5 min
20 min
control
washout AEx 1 g/l
Fig. 3. Cardiac rhythm disturbances promoted by the aortic perfusion of the guinea pig isolated heart with A. carambola aqueous extract (1 g/l). (A) Second-degree AV block – AVB (Mobitz type II); (B) third-degree AV block (atrioventricular dissociation); (C) Wenckebach periods (Mobitz type I). These effects disappeared (results not shown) during washout (n ¼ 6 hearts, 3470.1 1C, stimulus: 40 V, 1 ms).
140
Heart rate (%)
120
Atrial conduction velocity (percent of control)
Fig. 5. Effect of A. carambola aqueous extract (1 g/l) on the atrium–His (A–H) conduction time of guinea pig isolated heart (n ¼ 5 hearts, 3470.1 1C, stimulus: 40 V, 1 ms, po0:05).
110 100 90
*
80 70 60
100 control
*
80
AEx (1 g/l)
washout
Fig. 6. Atrial impulse conduction velocity decreases (1773%) by adding AEx (1 g/l) to the perfusion fluid (3 hearts, n ¼ 11 trials, po0:05).
60 40 20 0 control
AEx (1 g/l)
washout
Fig. 4. Effects of A. carambola aqueous extract on guinea pig spontaneous heart rate. Note that A. carambola reduced the spontaneous pacemaker activity, but this effect was promptly reverted during washout (n ¼ 4 hearts, 3470.1 1C, stimulus: 40 V, 1 ms, po0:001).
Fig. 4 deals with the AEx (1 g/l) effect on spontaneous heart rate. In 4 hearts, the cardiac rate decreased from 136717 to 89714 bpm (35%, n ¼ 8 trials, po0:001). This effect disappeared, however, around 20 min after washing the extract out (Fig. 4). Heart perfusion with atropine sulfate (10 mM, 20 min) did not prevent either the electrocardiographic changes or the left ventricular pressure decrease produced by AEx (1 g/l). AEx (1 g/l) prolonged (2776.5%, n ¼ 10 trials, 5 hearts, po0:001) the electrical wave conduction time from the right atrium to the His bundle (A–H), before
inducing a complete AV block (Fig. 5). This effect was abolished during washout (10371.6%). The H–V interval remained unchanged. In the presence of AEx (1 g/l), the conduction velocity of the atrial impulse (Fig. 6) was reversibly decreased (1773%, n ¼ 11 trials, 3 hearts, po0:05).
Discussion Myocardium tissues are built for speed in conducting electrical waves adequate to initiate contraction. To accomplish that, they use propagate action potentials elicited by changes of ionic channels conductance. Through the specific channels, sodium, potassium, and calcium ions move across the outer cellular membrane to depolarize and repolarize the cell. The electrical wave spreading through the myocardium generates an electromagnetic field that modifies the electric potentials of the entire body, allowing the ECG to be peripherally
ARTICLE IN PRESS 506
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
recorded. To analyze the effects of AEx on the myocardial electrical properties and to determine its action on the membrane channels, AEx was assayed on two kinds of preparations: (1) isolated heart (Langendorf perfusion) and (2) isolated right atrium and right ventricle. The results showed that AEx was able to: (1) increase the PR interval, (2) increase the QT interval, (3) increase the QRS complex duration, (4) decrease the cardiac rate, (5) decrease the conduction velocity of atrial impulse, (6) prolong the conduction time for the electrical impulse to travel from the right atrium to the His bundle, frequently inducing a third-degree AV block, (7) decrease the intraventricular pressure, and (8) decrease the ventricular wall mobility. Studies carried out to determine the impulse conduction time from the atrium to the His bundle have stressed the hypothesis that the increase of PR interval, promoted by AEx, is derived from the extracts effect on the AV nodal cells. It is well-established now that the PR interval is mainly dependent on the slow action potentials elicited on the AV node (Prystowsky, 1988; Golovko and Tipans, 1986; Brown et al., 1986; Gilmour and Zipes, 1985). At this site, cell depolarization occurs due to the spontaneous decrease of the membrane potassium conductance and to the slow inward currents carried by sodium and calcium ions crossing the cellular membrane through sodium slow channels and L-type calcium channels (Noma et al., 1980a, b; Mcdonald, 1982; Conde-Garcia, 1998). Two major factors contribute to make the AV nodal structure particularly vulnerable to the effect of myocardial depressant substances, e.g., calcium channel blockers and cholinergic drugs (Cranefield, 1975; Katz, 1977; Conde-Garcia, 1998). The first factor is the small action potential amplitude observed in this structure, and the second is its poor intercellular coupling due to a relatively small area of gap junctions (Challice and Viragh, 1974a, b; Sommer and Johnson, 1979). Therefore, the depressant effect of AEx on the AV node may be related either to a reduction of the membrane slow inward currents or to an increase of the intercellular pathway resistance, or even both. The former hypothesis is strengthened by the following AEx effects: (1) heart rate decrease in nonpaced hearts, (2) intraventricular pressure decrease followed by a progressive reduction of the heart wall mobility in constant paced hearts, and (3) a significant decrease of the impulse conduction velocity within the AV node, leading to several degrees of AV blocks. These findings are all dependent upon the slow inward currents and, as a consequence, drugs that can reduce them – e.g. dihydropiridines and verapamil – also reduce the spontaneous heart rate, reduce the myocardial force – explaining the intraventricular pressure decrease as well as the inhibition of the heart wall mobility seen with AEx, and delay the impulse conduction through the AV node. Furthermore, our results showed that the AEx
increased the QT interval, suggesting it is able to delay the ventricular repolarization. This important effect, which can be ascribed to a more prolonged inactivation of the membrane potassium channels, leads to the appearance of long-lasting action potentials. AEx also increased the QRS duration. Such findings point toward a reduction of the impulse conduction velocity, also at the ventricular myocardium level. In the heart, the electrical wave is slowed down when the myocardium length constant (l) is shortened or when the propagated action potential amplitude is reduced, or even in both circumstances. The length constant – a parameter derived from the Cable Theory (Kelvin, Lord, 1855, 1856; Hermann, 1905; Cole and Curtis, 1938; Hodgkin and Rushton, 1946) – p is ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi theoretically given by the following equation: l ¼ ða RmÞ=ð2 RiÞ (Jack et al., 1975, p. 37; Aidley, 1971, p. 51), where Rm stands for the membrane resistance, Ri is the intracellular pathway resistance, and a is the fiber radius. Thus, to reduce l – in order to explain the impulse velocity decrease – it would be necessary to decrease Rm or to increase Ri, or even both. Nevertheless, these hypotheses do not seem plausible because, if that were the case, the Rm decrease, during the resting membrane, would have to be dependent on the opening of potassium K1-type channels, such as is promoted by acetylcholine. In this case, the QT interval should be shortened. Instead, our results showed just the contrary: the QTi increased from 229723 to 264719 ms (15%, n ¼ 12 trials, po0:001). On the other hand, if Ri were increased by the AEx, the intercellular resistance at the next level would be the expected site for such effect, considering that the AEx does not significantly change the Tyrode electrical conductivity. It is now well established that the intercellular healing process depends upon the increase of intracellular calcium concentration (Heilbrunn, 1956; De Mello et al., 1969; De Mello and Dexter, 1970; De´le`ze, 1970). If this were the core mechanism of AEx effect, a contractile force increase would be expected to occur, but what could be seen was a huge decrease of intraventricular pressure, signalizing a decrease of the contractile myocardium force. Therefore, a new approach developed apart from the myocardial resistive components must be raised to shed light on the mechanism through which, under AEx, the electrical wave velocity decreases in guinea pig myocardium. It is known that the impulse velocity depends strongly on the fast inward sodium currents (Jack et al., 1975). Drugs or experimental maneuvers leading to the reduction of such a current contribute to reduce the first temporal derivative of the depolarization phase of propagated action potentials and this slows down the impulse conduction in both atrial and ventricular tissues. If a release of acetylcholine from the myocardium nerve endings is promoted by AEx, some effects such as force decrease and QTi duration reduction could be expected.
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Nevertheless, such release does not occur because atropine sulfate added to the organ bath did not impair AEx in promoting its myocardium effects. The flavonoids chrysanthemin and cyanin, (Gunasegaran, 1992) and isoquercitrin and rutin (Tiwari et al., 1979) were isolated from the flowers, and several terpenoids (sesquiterpenes, monoterpenes, and one triterpene lupeol) were isolated from fruit or carambola tree bark. The phytochemical analysis performed on this AEx revealed flavonoids and tannins. Unfortunately, there is no scientific report dealing with the effect of these molecules on the heart tissue. Some papers suggest the existence of a neurotoxin in star fruit juice. However, despite the efforts made by Neto et al. (1998, 2003) and Cecchini et al. (1999), the chemical nature of such a substance remains obscure. Uraemic patients with severe intoxication manifested after ingesting 25–500 ml of star fruit juice showed important evidence of cardiovascular system involvement. This was manifested by bradycardia, cardiorespiratory arrest, tachycardia, hemodynamic instability, and arterial hypotension. Such signals and symptoms could be mirroring the myocardial depressant effect. Whether the intoxication promoter is a neurotoxin remains controversial. Fang et al. (2001), Chen et al. (2002), and Tse et al. (2003) suggest that the oxalate should play an important role in the acute nephropathy induced by star fruit juice. In spite of the fact that the carambola fruit is rich in oxalates, its leaf AEx (1000 mg/ ml) contains only 21 mg/ml of oxalic acid. In our experiments (left guinea pig atria), at this concentration, neither oxalic acid nor sodium oxalate changed the atrial contractility (results not shown), suggesting that the myocardial active compound present in the AEx is not the oxalate. Contrary to popular belief, the data presented in this paper showed that herbal remedies could pose serious health risks. Therefore, any effort contributing to a better characterization of natural products bioeffects seems justified.
Acknowledgements The authors thank Professor Dr. Jose´ Maria Barbosa Filho and Mr. Raimundo Nonato Silva Filho (LTF/ UFPB, Paraı´ ba, Brazil) for the assistance to improve this paper.
References Aidley, D.J., 1971. The Physiology of Excitable Cells. Cambridge University Press, England, p. 467. Brown, H.F., Noble, D., Noble, S.J., Taupignon, A.I., 1986. The relationship between the transient inward current (TI)
507
and other components of slow inward current in mammalian cardiac muscle. Jpn. Heart J. 27 (Suppl. 1), 127–142. Burkhill, I.H., 1935. A dictionary of the economic products of the Malay peninsula (Ministry of Agriculture, Kuala Lumpur, 1935). In: Muir, C.K., Lam, C.K. (Eds.), Med J Malaysia 34, 279–281. Carolino, R.O., Beleboni, R.O., Cecchini, A.L., Maurı´ cio, V.B., Fontana, A.C.K., Neto, M.M., Coutinho-Netto, J., 2001. Isolamento e identificac¸a˜o da neurotoxina da carambola (Averrhoa carambola, Oxalidaceae). Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No. 01.301. Cecchini, A.L., Fontana, A.C.K., Beleboni, R.O., Pizzo, A.B., Santos, W.F., Thomazini, J.A., Neto, M.M., CoutinhoNeto, J., 1999. Purification and biochemical characterization of a convulsant neurotoxin from star fruit (Averrhoa carambola): specific neurotransmitter release on rat cerebral synaptosomes. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No. 12.193. Challice, C.E., Viragh, S., 1974a. The architectural development of the early mammalian heart. Tissue Cell 6, 447–462. Challice, C.E., Viragh, S., 1974b. Ultrastructure of the Mammalian Heart. New York, Academic Press. Chang, J.M., Hwang, S.J., Kuo, H.T., Tsai, J.C., Guh, J.Y., Chen, H.C., Tsai, J.H., Lai, Y.H., 2000. Fatal outcome after ingestion of star fruit (Averrhoa carambola) in uremic patients. Am. J. Kidney Dis. 35, 189–193. Chen, C.L., Chou, K.J., Wang, J.S., Yeh, J.H., Fang, H.C., Chung, H.M., 2002. Neurotoxic effects of carambola in rats: the role of oxalate. J. Formos Med. Assoc. 101, 337–341. Cole, K.S., Curtis, H.J., 1938. Electric impedance of ‘‘Nitella’’ during activity. J. Gen. Physiol. 22, 37–64. Conde-Garcia, E.A., 1998. Biofı´ sica, second ed. Editora SARVIER, Sa˜o Paulo, Brazil, p. 387. Cranefield, P.F., 1975. The Conduction of Cardiac Impulse. Futura Publishing Company, New York, p. 404. De Mello, W.C., Dexter, D., 1970. Increased rate of sealing in beating heart muscle of the toad. Circ. Res. 26, 481–489. De Mello, W.C., Motta, G., Chapeau, M., 1969. A study on the healing-over of myocardial cells of toads. Circ. Res. 24, 475–487. De´le`ze, J., 1970. The recovery of resting potential and input resistance in sheep heart injured by knife or laser. J. Physiol. 208, 547–562. Domı´ nguez, X.A., 1973. Me´todos de Investigacio´n Fitoquı´ mica, vol. 1. Editorial Limusa, Me´xico, p. 97. Do¨ring, H.J., 1990. The isolated perfused heart according to Langendorff technique – function – application. Physiol. Bohemoslovaca 39, 481–505. Eisenberg, D.M., Davis, R.B., Ettner, S.K., 1998. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. J. Am. Med. Assoc. 286, 1569–1575. Fang, H.C., Chen, C.L., Wang, J.S., Chou, K.J., Chiou, Y.S., Lee, P.T., Yeh, J.H., Yeh, M.Y., Chung, H.M., 2001. Acute oxalate nephropathy induced by star fruit in rats. Am. J. Kidney Dis. 38, 876–880. Gilmour, R.F., Zipes, D.P., 1985. Slow inward current and cardiac arrhythmias. Am. J. Cardiol. 55, 89B–101B.
ARTICLE IN PRESS 508
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Golovko, V.A., Tipans, I., 1986. Modelling of the effects of Na+ and Ca2+ ions on the generation of action potentials by rabbit heart true pacemaker cells. Gen. Physiol. Biophys. 5, 617–624. Gunasegaran, R., 1992. Flavonoids and anthocyanins of three oxalidaceae. Fitoterapia 63, 89–90. Harrison, G.E., Raymond, W.H.A., 1951. Preparation of clean chemical solutions. Br. Med. J. 2, 930–932. Heilbrunn, L.V., 1956. In: Heilbrunn, L.V. (Ed.), The Dynamics of Living Protoplasm. Academic Press, New York. Hermann, L., 1905. Beitra¨ge zur Physiologie und Physik des Nerven. Pflu¨gers Arch Ges. Physiol. 109, 95–144. Hodgkin, A.L., Rushton, W.A.H., 1946. The electrical constants of a crustacean nerve fiber. Proc. R. Soc. 133, 444–479. Jack, J.J.B., Noble, D., Tsien, R.W., 1975. Electric Current Flow in Excitable Cells. Clarendon Press, Oxford, p. 502. Katz, A.M., 1977. Physiology of the Heart. Raven Press, New York, pp. 355–379. Kelvin, Lord, (Thompson W.), 1855. On the theory of electric telegraph. Proc. R. Soc. 1, 328–399. Kelvin, Lord, (Thompson W.), 1856. On the theory of electric telegraph. Philos. Mag. 11, 146–160. Martha, R.C.D., Poubel, J., Ferreira, L.C.L., Lima, R.S., Borra´s, M.R., Costa, P.R.C., Roland, I.A., 2000. Atividade hipogliceˆmica de Averrhoa carambola L. usada em Manaus como antidiabe´tico. NewsLab 38, 142–148. Martin, L.C., Caramori, J.S.T., Barreti, P., Soares, V.A., 1993. Soluc¸o intrata´vel desencadeado por ingesta˜o de carambola (‘‘Averrhoa carambola’’) em portadores de insuficieˆncia renal croˆnica. J. Bras. Nefrol. 15, 92–94. McDonald, T.F., 1982. The slow inward calcium current in the heart. Annu. Rev. Physiol. 44, 425–434. Neto, M.M., Robl, F., Coutinho-Netto, J., 1998. Intoxication by star fruit (Averrhoa carambola) in six dialysis patients? Nephrol. Dial. Transplant. 13, 570–572.
Neto, M.M., Costa, J.A., Garcia-Cairasco, N., Netto, J.C., Nakagawa, B., Dantas, M., 2003. Intoxication by star fruit (Averrhoa carambola) in 32 uraemic patients: treatment and outcome. Nephrol. Dial. Transplant. 18, 120–125. Noma, A., Kotake, H., Irisawa, H., 1980a. Slow inward current and its role mediating the chronotropic effect of epinephrine in the rabbit sinoatrial node. Pflu¨ger Arch. 388, 1–9. Noma, A., Irisawa, H., Kokobun, S., Kotake, H., Nishimura, M., Watanabe, Y., 1980b. Slow current systems in A–V node of the rabbit heart. Nature 285, 228–229. Prystowsky, E.N., 1988. The effects of slow channel blockers and beta blockers on atrioventricular nodal conduction. J. Clin. Pharmacol. 28, 6–21. Sommer, J.R., Johnson, E.A., 1979. Ultrastructure of cardiac muscle. In: Berne, R.M., Sperelakis, N., Geiger, S.R. (Eds.), Handbook of Physiology – The Cardiovascular System, vol. I, Maryland, USA. pp. 113–186, 168. Tiwari, K.P., Masood, M., Minocha, P.K., 1979. Chemical constituents of Gemlina phillipinensis, Adenocalymna nitida, Allamanda cathartica, Averrhoa carambola and Maba buxifolia. J. Indian Chem. Soc. 56, 944. Tse, K.-C., Yip, P.-S., Lam, M.-F., Choy, B.-Y., Li, F.-K., Lui, S.-L., Lo, W.-K., Chan, T.-M., Lai, K.-N., 2003. Star fruit intoxication in uraemic patients: case series and review of the literature. Intern. Med. J. 33, 314–316. Vasconcelos, C.M.L., Conde-Garcia, E.A., 2002. Alterac¸o˜es eletrofisiolo´gicas produzidas pela Averrhoa carambola L em corac¸a˜o de cobaia. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No.11.029 Vasconcelos, C.M.L., Dias, C.S., Barbosa, J.M., CondeGarcia, E.A., 2001. Averrhoa carambola L. – Estudo do seu componente cardiodepressor. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No.03.002.
Phytomedicine 13 (2006) 501–508 www.elsevier.de/phymed
Electrophysiological effects of the aqueous extract of Averrhoa carambola L. leaves on the guinea pig heart$ C.M.L. Vasconcelosa,b, M.S. Arau´joa, E.A. Conde-Garciaa, a Cardiobiophysics Research Laboratory, Department of Physiology, Center of Biological and Health Sciences, Universidade Federal de Sergipe, Aracaju, Sergipe, Brazil b Laborato´rio de Tecnologia Farmaceˆutica, Universidade Federal da Paraı´ba, Joa˜o Pessoa, Paraı´ba, Brazil
Received 17 December 2004; accepted 27 January 2005
Abstract This work aims to describe some electrophysiological changes promoted by the aqueous extract (AEx) from Averrhoa carambola leaves in guinea pig heart. The experiments were carried out on isolated heart or on right atrium–ventricle preparations. In 6 hearts, the extract induced many kinds of atrioventricular blocks (1st, 2nd, and 3rd degrees); increased the QT interval from 229723 to 264719 ms; increased the QRS complex duration from 2773.1 to 59711 ms, and depressed the cardiac rate from 136717 to 89714 bpm. Furthermore, it decreased the conduction velocity of atrial impulse (1773%); reduced the intraventricular pressure (8676%), and increased the conduction time between the right atrium and the His bundle (2776.5%). The conduction time from the His bundle to the right ventricle was not altered. Atropine sulfate did not change either the electrocardiographic parameters or the intraventricular pressure effects promoted by the A. carambola AEx. Based on these results, the popular use of such extracts should be avoided because it can promote electrical and mechanical changes in the normal heart. r 2005 Elsevier GmbH. All rights reserved. Keywords: Averrhoa carambola L.; Guinea pig atrium; Myocardium; ECG; His electrogram
Introduction Herbal medicine is an increasingly common form of alternative therapy worldwide. In 1997, it was estimated that 12.1% of adults in the United States used herbal $ Research supported by Centrais Ele´tricas Brasileiras (ELETROBRAS, Process No. 23113.009351/03-67), Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sergipe (FAP-SE/FUNTEC FNS/Process No. 01/2003), Universidade Federal de Sergipe, and by the Brazilian Conselho Nacional do Desenvolvimento Cientı´ fico e Tecnolo´gico CNPq. Corresponding author. Tel.: +55 79 212 6642; fax: +55 79 246 3377. E-mail address: [email protected] (E.A. Conde-Garcia).
0944-7113/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2005.01.013
medicine in the previous 12 months, representing an outof-pocket payment of US$ 5.1 billion (Eisenberg et al., 1998). Averrhoa carambola L., known as star fruit (‘carambola’ in Brazil), belongs to the Oxalidaceae family. According to folk medicine, it is commonly used for treating restlessness, headache, nausea, and cough (Burkhill, 1935). The following symptoms have been reported after fruit ingestion: intractable hiccups, insomnia, mental confusion, and even death in patients who suffered from renal chronic failure (Martin et al., 1993; Neto et al., 1998; Chang et al., 2000). Cecchini et al. (1999) and Carolino et al. (2001) reported the characterization of a potent neurotoxin, isolated from A. carambola fruits. It is a water-soluble molecule that
ARTICLE IN PRESS 502
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
could induce convulsions and L-glutamate release from rat synaptosomes. Furthermore, infusions prepared from dry A. carambola leaves reduced the glucose levels in Wistar rat blood (Martha et al., 2000). Nevertheless, no mention could be found in the specialized literature (Medline/Pubmed: http://www.ncbi.nih.gov/entrez/ query.fcgi and Medline/Bireme – http://www.bireme.br, covered period from 1960 to 2004, seeking date: January 7, 2005) dealing with the A. carambola myocardial effects. We observed that its leaf aqueous extract (AEx) depressed both the cardiac spontaneous rate and guinea pig left atrium contractility (Vasconcelos and CondeGarcia, 2002; Vasconcelos et al., 2001). The present paper describes some electrocardiographic and electrophysiological effects of A. carambola AEx on guinea pig heart.
Material and methods Plant characterization Leaves were collected near the Federal University of Sergipe campus (Aracaju, Sergipe, Brazil) from agrotoxic-free and fumigation-free trees, during the winter season (June–July 2001). Plant identification was made by comparson with voucher specimen no. 24720 deposited in the Herbarium of the Federal University of Pernambuco, Recife, Brazil.
Aqueous extract preparation The AEx was obtained in a Soxhlet apparatus by extracting dry leaves with the following solvents (P.A.A.C.S): hexane, chloroform, acetone, ethanol, methanol, and water. Each extract was concentrated in a rotative evaporator (BUCHII RE 111, Buchi Laboratoriums-Technik AG, Flawil, Schweiz) and stored at 2773 1C in a dry atmosphere without light protection until (storage time: 1–6 months).
Phytochemistry screening Phytochemical screening was performed on AEx according to the technique proposed by Domı´ nguez (1973).
Isolated heart experimental assembly Guinea pigs (Cavia porcellus) of both sexes (300–500 g each) were sacrificed by cervical stroke 30 min after a subcutaneous administration of heparin (1000 IU, Liquemine, Roche, Sa˜o Paulo, SP, Brazil). The animal chest was opened, the heart carefully removed and mounted on a constant-flow (Milan Peristaltic Pump,
Milan Equipamentos Cientı´ ficos Ltda., Curitiba, PR, Brazil) aortic perfusion system (Langendorff technique, Do¨ring, 1990). The heart was perfused by a modified Tyrode solution (NaCl 137.0, KCl 5.0, MgCl2 0.5, NaHCO3 12.0, CaCl2 1.8, Glucose 6.0, NaH2PO4 1.8, in mM, substances of analytical grade purchased from Merck S.A. Indu´strias Quı´ micas, Rio de Janeiro, Brazil), oxygenated and buffered by carbogen mixture (95% O2+5% CO2, error less then 0.2%, purchased from Aga S.A., Sa˜o Paulo, Brazil or White Martins Gases Industriais S.A. Sa˜o Paulo, Brazil), kept at 3470.1 1C (Haake F3, Berlin, Germany), and filtered through a cellulose acetate membrane (0.45 mm) to prevent microembolia (Harrison and Raymond, 1951). When the heart was electrically driven, suprathreshold DC current pulses, isolated from the ground were used (Anapulse Stimulator 302-T, WPI Instruments, Inc., 60 Fitch Street, POX 3110, New Haven, Connecticut 06515, USA; Digitimer D4030, Digitimer DS2, Digitimer Limited, Tewin Road, Welwyn Garden City, Hertfordshire, England). The stimuli were delivered through a pair of stainless steel electrodes connected to the right atrium appendage. The whole preparation was maintained immersed in Tyrode (50 ml) in which three electrodes (Ag/AgCl/NaCl 1 M) were placed for sensing the heart electrical signal, the electrocardiogram (ECG) to be recorded. Those signals were amplified (HP8811B, HP7754A, HP7754B, Hewlett-Packard, Chicago, IL, USA) and stored in a computer for off-line processing (DI-205, DI-400, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA). Before beginning the experimental procedures, the biological preparation was allowed to stabilize for 1–2 h. To investigate whether the parasympathetic pathways were involved in the cardiodepressor effects of AEx, some hearts were atropinized by adding atropine sulfate (10 mM) to the perfusion fluid 20 min before starting the test with the AEx.
Heart rate measurements For studying the effect of AEx on cardiac rate, the isolated heart was allowed to beat spontaneously. The ECG was recorded in a computer and the heart rate was determined, beat-to-beat, from the stored data. To accomplish that, successive R–R intervals during control, test (AEx, 1 g/l), and washout, were determined (Calculate and Windaex softwares from Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA).
Intraventricular pressure Left intraventricular pressure was measured using a water-filled balloon. This device, coupled to a pressure transducer (HP 1290A, HP8805B), sent its signals to an
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
analog-to-digital (A/D) converter running at 512 samples/s (DI-205, DI-400, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA). The pressure, recorded and stored in a computer, was processed off-line. A column of mercury was used to calibrate the system used to measure intraventricular pressure.
Heart wall movements The inotropic effect of AEx was documented by recording both intraventricular pressure and heart wall movements. These experiments were performed on isolated guinea pig hearts (constant flow Langendorf perfusion, 3470.1 1C) artificially paced (150–176 bpm). A red laser beam (diameter: 4 2 mm) was positioned to pass tangentially to the left ventricle wall. The heart movements altered the laser beam area that emerged from the ventricular edge. This variable beam was then caught by a photoresistor (Vactec VT 732E 77 29, EG & G Vactec, 200 Orchard Ridge Drive, Suit 100, Gaithersburg, MD 20 878, USA) mounted within a plastic black cylinder (25 mm long). In the cylinder base, a small hole (diameter: 2 mm) allowed the laser beam to enter and to thus reach the photoresistor. This device controlled the current intensity of a DC circuit fed by a 9 V battery. The larger the laser spot on the hole of the cylinder base, the more intense the transmitted laser beam to the photoresistor detector. The electric current in the detection circuit was monitored by recording the voltage drop across a 100 kO resistor. The voltage changes were digitized and recorded on a computer, as described previously.
His bundle electrogram and atrial impulse velocity To study both the impulse conduction through the atrioventricular (AV) node and the atrial impulse velocity, the right atrium and the right ventricle were mounted in an organ chamber. There, the biological preparation, laid and pinned on the chamber paraffin bed with its endocardium surface facing upwards, was superfused by Tyrode solution (3470.1 1C, 15 ml) buffered and oxygenated with carbogen mixture (oxygen 95%, carbon dioxide 5%, error less than 0.2%). The preparation was stimulated electrically by means of two stainless steel electrodes (DC pulses, 40 V, 1 ms). To detect the atrial and the His bundle electrical waves, two surface electrodes (Ag/Teflon) were placed on the atrial and on the His bundle, near the coronary sinus. The distance between the two surface electrodes was determined optically with the aid of a calibrated stereomicroscope (Wild Heerbrugg AG M5A, Juerg Dedual Gaebrisstrasse 8 CH-9056, Gais, Switzerland). The electrical signals were amplified and monitored on
503
an oscilloscope screen (Tektronix D44 Dual Beam, 5A22N Differential Amplifier, 5A14N Four Channel Amplifier, 5B44 Dual Time Base, Tektronix, Inc., Beaverton, OR 97005, USA). They were also digitized by an A/D converter (DI 400, DI 205, Windaq Pro, Dataq, 241 Springside Drive Suite 200, Akron, OH 44 333, USA) and stored in a computer (sampling rate: 2.048 kHz). The time intervals for the impulse propagation from the atrium to the His bundle (A–H), as well as from the His bundle to the right ventricle (H–V) were measured; two electronic gates were generated. Each of them triggered on and off a 100 kHz pulse train, and an electronic setup determined the gate durations by counting the pulses within them (Digitimer D4030, WPI 1830, 1831, 1832, Digitimer Limited, Tewin Road, Welwyn Garden City, Hertfordshire, England). This procedure allowed the A–H and H–V conduction times to be measured with a precision of 710 ms.
Data processing The Student’s t-test for independent or dependent samples was used to decide upon the differences between two population means by adopting a significance level of 5% (Statistica for Windows). The results presented in this paper are expressed as means7s.e.m.
Results The phytochemical screening of AEx revealed the following constituents: a saponin [foam ()]; an alkaloid [Bouchardat (), Mayer (), Dragendorff (), tungstosilicic acid ()]; steroids (+); tannins [gelatin (+), FeCl3 (+++)]; and a flavonoid [magnesium tape (++), fluorescence (+++)]. Fig. 1 (upper panel) shows the electrocardiographic changes promoted by AEx (1 g/l) on isolated guinea pig heart. AEx increased the PR interval (PRi) from 102 to 166 ms characterizing a first-degree AV block (Fig. 1B). The extract also delayed the ventricular repolarization. This could be observed by the increase of the QT interval (QTi) from 215 to 268 ms, and by the increase of the QRS complex duration from 33 to 71 ms (Fig. 1B). These changes, however, disappeared during washout (Fig. 1C). Similar results were observed in 6 other hearts (lower panel) in which the AEx (1 g/l) increased the PRi from 107712 to 173729 ms (61%, n ¼ 14 trials, po0:001), QTi from 229723 to 264719 ms (15%, n ¼ 12 trials, po0:001), and QRS complex duration from 2773.1 to 59711 ms (118%, n ¼ 11 trials, po0:001). The control values for PRi, QTi, and QRS duration could be recovered during washout. Fig. 2 shows an example of AEx effect (1 g/l) on: (a) the intraventricular pressure (upper panel), (b) the heart
ARTICLE IN PRESS 504
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Fig. 1. Effects of A. carambola aqueous extract (1 g/l) on the electrocardiogram of guinea pig isolated heart. Upper panel: (A) control; (B) test by adding extract to the organ bath; and (C) washout. The extract reversibly increased: (1) PR interval (dotted line), (2) QT interval (solid line), and (3) QRS complex duration. S: stimulus artifact. Values are in milliseconds. Lower panel: results obtained in 6 other guinea pig isolated hearts (3470.1 1C, 120–150 bpm, po0:001, stimuli: 30 V, 1 ms).
wall movement (middle panel), and (c) on the ECG (lower panel). The experiment was performed on an artificially paced (150 bpm) guinea pig heart. The figure shows recording obtained at different moments: during control (control) and after perfusing with AEx (190 and 220 s). At 190 s, the intraventricular pressure was reduced from 64 to 25 mmHg (60%) and after 220 s it declined to zero. Note that the optical signal (middle panel) disappeared at 190 s. In spite of these huge mechanical effects, even after 220 s of adding the extract to the perfusion fluid, the ECG could be recorded, but showed the following parameter increase: (1) PRi (from 96 to 220 ms), (2) QRS duration (from 33 to 108 ms), and (3) QTi (from 267 to 337 ms). During washout PRi, QRS, and QTi were 96, 33, and 268 ms, respectively. In 6 hearts, the AEx reduced about 8676% the intraventricular pressure. AEx (1 g/l) depressed impulse conduction through the AV node (Fig. 3). An example of a first-degree AV block is shown and described in Fig. 1. Second-degree AV block, characterized by an intermittent conduction through the AV node (Fig. 3A), third-degree AV block – mirroring a compete impairment of the AV nodal conduction (Fig. 3B), and the Wenckbach phenomenon, in which the PRi progressively increased until a complete AV block (Fig. 3C) could also be recorded. These effects disappeared promptly during washout. Disturbances in the AV nodal conduction could be recorded in all tested hearts.
Fig. 2. Electromechanical uncoupling promoted by A. carambola aqueous extract (AEx, 1 g/l). Upper panel: superimposed intraventricular pressure curves; middle panel: curves related to heart wall movement; lower panel: electrocardiogram of artificially paced guinea pig isolated heart (150 bpm). Numbers without units stand for time intervals, measured in seconds, after beginning extract perfusion. Horizontal bars correspond to 100 ms. Periodical waves seen at the right side of the upper and middle panels correspond to the pressure and heart movement curves observed during washout. Note that 190 s after the extract, the optical signal related to the heart wall movement ceased and 220 s after the extract was administered the intraventricular pressure drops to zero. In spite of these effects, electrocardiogram waves were not abolished, but the PRi, QRS, and QTi were increased. During washout (15 min), PRi, QRS, and QTi recovered their control values. S stands for stimulus artifact (3470.1 1C, stimulus: 40 V, 1 ms).
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
505
A-H conduction (percent of control)
200
* 100
complete AV block 0
5 min
20 min
control
washout AEx 1 g/l
Fig. 3. Cardiac rhythm disturbances promoted by the aortic perfusion of the guinea pig isolated heart with A. carambola aqueous extract (1 g/l). (A) Second-degree AV block – AVB (Mobitz type II); (B) third-degree AV block (atrioventricular dissociation); (C) Wenckebach periods (Mobitz type I). These effects disappeared (results not shown) during washout (n ¼ 6 hearts, 3470.1 1C, stimulus: 40 V, 1 ms).
140
Heart rate (%)
120
Atrial conduction velocity (percent of control)
Fig. 5. Effect of A. carambola aqueous extract (1 g/l) on the atrium–His (A–H) conduction time of guinea pig isolated heart (n ¼ 5 hearts, 3470.1 1C, stimulus: 40 V, 1 ms, po0:05).
110 100 90
*
80 70 60
100 control
*
80
AEx (1 g/l)
washout
Fig. 6. Atrial impulse conduction velocity decreases (1773%) by adding AEx (1 g/l) to the perfusion fluid (3 hearts, n ¼ 11 trials, po0:05).
60 40 20 0 control
AEx (1 g/l)
washout
Fig. 4. Effects of A. carambola aqueous extract on guinea pig spontaneous heart rate. Note that A. carambola reduced the spontaneous pacemaker activity, but this effect was promptly reverted during washout (n ¼ 4 hearts, 3470.1 1C, stimulus: 40 V, 1 ms, po0:001).
Fig. 4 deals with the AEx (1 g/l) effect on spontaneous heart rate. In 4 hearts, the cardiac rate decreased from 136717 to 89714 bpm (35%, n ¼ 8 trials, po0:001). This effect disappeared, however, around 20 min after washing the extract out (Fig. 4). Heart perfusion with atropine sulfate (10 mM, 20 min) did not prevent either the electrocardiographic changes or the left ventricular pressure decrease produced by AEx (1 g/l). AEx (1 g/l) prolonged (2776.5%, n ¼ 10 trials, 5 hearts, po0:001) the electrical wave conduction time from the right atrium to the His bundle (A–H), before
inducing a complete AV block (Fig. 5). This effect was abolished during washout (10371.6%). The H–V interval remained unchanged. In the presence of AEx (1 g/l), the conduction velocity of the atrial impulse (Fig. 6) was reversibly decreased (1773%, n ¼ 11 trials, 3 hearts, po0:05).
Discussion Myocardium tissues are built for speed in conducting electrical waves adequate to initiate contraction. To accomplish that, they use propagate action potentials elicited by changes of ionic channels conductance. Through the specific channels, sodium, potassium, and calcium ions move across the outer cellular membrane to depolarize and repolarize the cell. The electrical wave spreading through the myocardium generates an electromagnetic field that modifies the electric potentials of the entire body, allowing the ECG to be peripherally
ARTICLE IN PRESS 506
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
recorded. To analyze the effects of AEx on the myocardial electrical properties and to determine its action on the membrane channels, AEx was assayed on two kinds of preparations: (1) isolated heart (Langendorf perfusion) and (2) isolated right atrium and right ventricle. The results showed that AEx was able to: (1) increase the PR interval, (2) increase the QT interval, (3) increase the QRS complex duration, (4) decrease the cardiac rate, (5) decrease the conduction velocity of atrial impulse, (6) prolong the conduction time for the electrical impulse to travel from the right atrium to the His bundle, frequently inducing a third-degree AV block, (7) decrease the intraventricular pressure, and (8) decrease the ventricular wall mobility. Studies carried out to determine the impulse conduction time from the atrium to the His bundle have stressed the hypothesis that the increase of PR interval, promoted by AEx, is derived from the extracts effect on the AV nodal cells. It is well-established now that the PR interval is mainly dependent on the slow action potentials elicited on the AV node (Prystowsky, 1988; Golovko and Tipans, 1986; Brown et al., 1986; Gilmour and Zipes, 1985). At this site, cell depolarization occurs due to the spontaneous decrease of the membrane potassium conductance and to the slow inward currents carried by sodium and calcium ions crossing the cellular membrane through sodium slow channels and L-type calcium channels (Noma et al., 1980a, b; Mcdonald, 1982; Conde-Garcia, 1998). Two major factors contribute to make the AV nodal structure particularly vulnerable to the effect of myocardial depressant substances, e.g., calcium channel blockers and cholinergic drugs (Cranefield, 1975; Katz, 1977; Conde-Garcia, 1998). The first factor is the small action potential amplitude observed in this structure, and the second is its poor intercellular coupling due to a relatively small area of gap junctions (Challice and Viragh, 1974a, b; Sommer and Johnson, 1979). Therefore, the depressant effect of AEx on the AV node may be related either to a reduction of the membrane slow inward currents or to an increase of the intercellular pathway resistance, or even both. The former hypothesis is strengthened by the following AEx effects: (1) heart rate decrease in nonpaced hearts, (2) intraventricular pressure decrease followed by a progressive reduction of the heart wall mobility in constant paced hearts, and (3) a significant decrease of the impulse conduction velocity within the AV node, leading to several degrees of AV blocks. These findings are all dependent upon the slow inward currents and, as a consequence, drugs that can reduce them – e.g. dihydropiridines and verapamil – also reduce the spontaneous heart rate, reduce the myocardial force – explaining the intraventricular pressure decrease as well as the inhibition of the heart wall mobility seen with AEx, and delay the impulse conduction through the AV node. Furthermore, our results showed that the AEx
increased the QT interval, suggesting it is able to delay the ventricular repolarization. This important effect, which can be ascribed to a more prolonged inactivation of the membrane potassium channels, leads to the appearance of long-lasting action potentials. AEx also increased the QRS duration. Such findings point toward a reduction of the impulse conduction velocity, also at the ventricular myocardium level. In the heart, the electrical wave is slowed down when the myocardium length constant (l) is shortened or when the propagated action potential amplitude is reduced, or even in both circumstances. The length constant – a parameter derived from the Cable Theory (Kelvin, Lord, 1855, 1856; Hermann, 1905; Cole and Curtis, 1938; Hodgkin and Rushton, 1946) – p is ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi theoretically given by the following equation: l ¼ ða RmÞ=ð2 RiÞ (Jack et al., 1975, p. 37; Aidley, 1971, p. 51), where Rm stands for the membrane resistance, Ri is the intracellular pathway resistance, and a is the fiber radius. Thus, to reduce l – in order to explain the impulse velocity decrease – it would be necessary to decrease Rm or to increase Ri, or even both. Nevertheless, these hypotheses do not seem plausible because, if that were the case, the Rm decrease, during the resting membrane, would have to be dependent on the opening of potassium K1-type channels, such as is promoted by acetylcholine. In this case, the QT interval should be shortened. Instead, our results showed just the contrary: the QTi increased from 229723 to 264719 ms (15%, n ¼ 12 trials, po0:001). On the other hand, if Ri were increased by the AEx, the intercellular resistance at the next level would be the expected site for such effect, considering that the AEx does not significantly change the Tyrode electrical conductivity. It is now well established that the intercellular healing process depends upon the increase of intracellular calcium concentration (Heilbrunn, 1956; De Mello et al., 1969; De Mello and Dexter, 1970; De´le`ze, 1970). If this were the core mechanism of AEx effect, a contractile force increase would be expected to occur, but what could be seen was a huge decrease of intraventricular pressure, signalizing a decrease of the contractile myocardium force. Therefore, a new approach developed apart from the myocardial resistive components must be raised to shed light on the mechanism through which, under AEx, the electrical wave velocity decreases in guinea pig myocardium. It is known that the impulse velocity depends strongly on the fast inward sodium currents (Jack et al., 1975). Drugs or experimental maneuvers leading to the reduction of such a current contribute to reduce the first temporal derivative of the depolarization phase of propagated action potentials and this slows down the impulse conduction in both atrial and ventricular tissues. If a release of acetylcholine from the myocardium nerve endings is promoted by AEx, some effects such as force decrease and QTi duration reduction could be expected.
ARTICLE IN PRESS C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Nevertheless, such release does not occur because atropine sulfate added to the organ bath did not impair AEx in promoting its myocardium effects. The flavonoids chrysanthemin and cyanin, (Gunasegaran, 1992) and isoquercitrin and rutin (Tiwari et al., 1979) were isolated from the flowers, and several terpenoids (sesquiterpenes, monoterpenes, and one triterpene lupeol) were isolated from fruit or carambola tree bark. The phytochemical analysis performed on this AEx revealed flavonoids and tannins. Unfortunately, there is no scientific report dealing with the effect of these molecules on the heart tissue. Some papers suggest the existence of a neurotoxin in star fruit juice. However, despite the efforts made by Neto et al. (1998, 2003) and Cecchini et al. (1999), the chemical nature of such a substance remains obscure. Uraemic patients with severe intoxication manifested after ingesting 25–500 ml of star fruit juice showed important evidence of cardiovascular system involvement. This was manifested by bradycardia, cardiorespiratory arrest, tachycardia, hemodynamic instability, and arterial hypotension. Such signals and symptoms could be mirroring the myocardial depressant effect. Whether the intoxication promoter is a neurotoxin remains controversial. Fang et al. (2001), Chen et al. (2002), and Tse et al. (2003) suggest that the oxalate should play an important role in the acute nephropathy induced by star fruit juice. In spite of the fact that the carambola fruit is rich in oxalates, its leaf AEx (1000 mg/ ml) contains only 21 mg/ml of oxalic acid. In our experiments (left guinea pig atria), at this concentration, neither oxalic acid nor sodium oxalate changed the atrial contractility (results not shown), suggesting that the myocardial active compound present in the AEx is not the oxalate. Contrary to popular belief, the data presented in this paper showed that herbal remedies could pose serious health risks. Therefore, any effort contributing to a better characterization of natural products bioeffects seems justified.
Acknowledgements The authors thank Professor Dr. Jose´ Maria Barbosa Filho and Mr. Raimundo Nonato Silva Filho (LTF/ UFPB, Paraı´ ba, Brazil) for the assistance to improve this paper.
References Aidley, D.J., 1971. The Physiology of Excitable Cells. Cambridge University Press, England, p. 467. Brown, H.F., Noble, D., Noble, S.J., Taupignon, A.I., 1986. The relationship between the transient inward current (TI)
507
and other components of slow inward current in mammalian cardiac muscle. Jpn. Heart J. 27 (Suppl. 1), 127–142. Burkhill, I.H., 1935. A dictionary of the economic products of the Malay peninsula (Ministry of Agriculture, Kuala Lumpur, 1935). In: Muir, C.K., Lam, C.K. (Eds.), Med J Malaysia 34, 279–281. Carolino, R.O., Beleboni, R.O., Cecchini, A.L., Maurı´ cio, V.B., Fontana, A.C.K., Neto, M.M., Coutinho-Netto, J., 2001. Isolamento e identificac¸a˜o da neurotoxina da carambola (Averrhoa carambola, Oxalidaceae). Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No. 01.301. Cecchini, A.L., Fontana, A.C.K., Beleboni, R.O., Pizzo, A.B., Santos, W.F., Thomazini, J.A., Neto, M.M., CoutinhoNeto, J., 1999. Purification and biochemical characterization of a convulsant neurotoxin from star fruit (Averrhoa carambola): specific neurotransmitter release on rat cerebral synaptosomes. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No. 12.193. Challice, C.E., Viragh, S., 1974a. The architectural development of the early mammalian heart. Tissue Cell 6, 447–462. Challice, C.E., Viragh, S., 1974b. Ultrastructure of the Mammalian Heart. New York, Academic Press. Chang, J.M., Hwang, S.J., Kuo, H.T., Tsai, J.C., Guh, J.Y., Chen, H.C., Tsai, J.H., Lai, Y.H., 2000. Fatal outcome after ingestion of star fruit (Averrhoa carambola) in uremic patients. Am. J. Kidney Dis. 35, 189–193. Chen, C.L., Chou, K.J., Wang, J.S., Yeh, J.H., Fang, H.C., Chung, H.M., 2002. Neurotoxic effects of carambola in rats: the role of oxalate. J. Formos Med. Assoc. 101, 337–341. Cole, K.S., Curtis, H.J., 1938. Electric impedance of ‘‘Nitella’’ during activity. J. Gen. Physiol. 22, 37–64. Conde-Garcia, E.A., 1998. Biofı´ sica, second ed. Editora SARVIER, Sa˜o Paulo, Brazil, p. 387. Cranefield, P.F., 1975. The Conduction of Cardiac Impulse. Futura Publishing Company, New York, p. 404. De Mello, W.C., Dexter, D., 1970. Increased rate of sealing in beating heart muscle of the toad. Circ. Res. 26, 481–489. De Mello, W.C., Motta, G., Chapeau, M., 1969. A study on the healing-over of myocardial cells of toads. Circ. Res. 24, 475–487. De´le`ze, J., 1970. The recovery of resting potential and input resistance in sheep heart injured by knife or laser. J. Physiol. 208, 547–562. Domı´ nguez, X.A., 1973. Me´todos de Investigacio´n Fitoquı´ mica, vol. 1. Editorial Limusa, Me´xico, p. 97. Do¨ring, H.J., 1990. The isolated perfused heart according to Langendorff technique – function – application. Physiol. Bohemoslovaca 39, 481–505. Eisenberg, D.M., Davis, R.B., Ettner, S.K., 1998. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. J. Am. Med. Assoc. 286, 1569–1575. Fang, H.C., Chen, C.L., Wang, J.S., Chou, K.J., Chiou, Y.S., Lee, P.T., Yeh, J.H., Yeh, M.Y., Chung, H.M., 2001. Acute oxalate nephropathy induced by star fruit in rats. Am. J. Kidney Dis. 38, 876–880. Gilmour, R.F., Zipes, D.P., 1985. Slow inward current and cardiac arrhythmias. Am. J. Cardiol. 55, 89B–101B.
ARTICLE IN PRESS 508
C.M.L. Vasconcelos et al. / Phytomedicine 13 (2006) 501–508
Golovko, V.A., Tipans, I., 1986. Modelling of the effects of Na+ and Ca2+ ions on the generation of action potentials by rabbit heart true pacemaker cells. Gen. Physiol. Biophys. 5, 617–624. Gunasegaran, R., 1992. Flavonoids and anthocyanins of three oxalidaceae. Fitoterapia 63, 89–90. Harrison, G.E., Raymond, W.H.A., 1951. Preparation of clean chemical solutions. Br. Med. J. 2, 930–932. Heilbrunn, L.V., 1956. In: Heilbrunn, L.V. (Ed.), The Dynamics of Living Protoplasm. Academic Press, New York. Hermann, L., 1905. Beitra¨ge zur Physiologie und Physik des Nerven. Pflu¨gers Arch Ges. Physiol. 109, 95–144. Hodgkin, A.L., Rushton, W.A.H., 1946. The electrical constants of a crustacean nerve fiber. Proc. R. Soc. 133, 444–479. Jack, J.J.B., Noble, D., Tsien, R.W., 1975. Electric Current Flow in Excitable Cells. Clarendon Press, Oxford, p. 502. Katz, A.M., 1977. Physiology of the Heart. Raven Press, New York, pp. 355–379. Kelvin, Lord, (Thompson W.), 1855. On the theory of electric telegraph. Proc. R. Soc. 1, 328–399. Kelvin, Lord, (Thompson W.), 1856. On the theory of electric telegraph. Philos. Mag. 11, 146–160. Martha, R.C.D., Poubel, J., Ferreira, L.C.L., Lima, R.S., Borra´s, M.R., Costa, P.R.C., Roland, I.A., 2000. Atividade hipogliceˆmica de Averrhoa carambola L. usada em Manaus como antidiabe´tico. NewsLab 38, 142–148. Martin, L.C., Caramori, J.S.T., Barreti, P., Soares, V.A., 1993. Soluc¸o intrata´vel desencadeado por ingesta˜o de carambola (‘‘Averrhoa carambola’’) em portadores de insuficieˆncia renal croˆnica. J. Bras. Nefrol. 15, 92–94. McDonald, T.F., 1982. The slow inward calcium current in the heart. Annu. Rev. Physiol. 44, 425–434. Neto, M.M., Robl, F., Coutinho-Netto, J., 1998. Intoxication by star fruit (Averrhoa carambola) in six dialysis patients? Nephrol. Dial. Transplant. 13, 570–572.
Neto, M.M., Costa, J.A., Garcia-Cairasco, N., Netto, J.C., Nakagawa, B., Dantas, M., 2003. Intoxication by star fruit (Averrhoa carambola) in 32 uraemic patients: treatment and outcome. Nephrol. Dial. Transplant. 18, 120–125. Noma, A., Kotake, H., Irisawa, H., 1980a. Slow inward current and its role mediating the chronotropic effect of epinephrine in the rabbit sinoatrial node. Pflu¨ger Arch. 388, 1–9. Noma, A., Irisawa, H., Kokobun, S., Kotake, H., Nishimura, M., Watanabe, Y., 1980b. Slow current systems in A–V node of the rabbit heart. Nature 285, 228–229. Prystowsky, E.N., 1988. The effects of slow channel blockers and beta blockers on atrioventricular nodal conduction. J. Clin. Pharmacol. 28, 6–21. Sommer, J.R., Johnson, E.A., 1979. Ultrastructure of cardiac muscle. In: Berne, R.M., Sperelakis, N., Geiger, S.R. (Eds.), Handbook of Physiology – The Cardiovascular System, vol. I, Maryland, USA. pp. 113–186, 168. Tiwari, K.P., Masood, M., Minocha, P.K., 1979. Chemical constituents of Gemlina phillipinensis, Adenocalymna nitida, Allamanda cathartica, Averrhoa carambola and Maba buxifolia. J. Indian Chem. Soc. 56, 944. Tse, K.-C., Yip, P.-S., Lam, M.-F., Choy, B.-Y., Li, F.-K., Lui, S.-L., Lo, W.-K., Chan, T.-M., Lai, K.-N., 2003. Star fruit intoxication in uraemic patients: case series and review of the literature. Intern. Med. J. 33, 314–316. Vasconcelos, C.M.L., Conde-Garcia, E.A., 2002. Alterac¸o˜es eletrofisiolo´gicas produzidas pela Averrhoa carambola L em corac¸a˜o de cobaia. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No.11.029 Vasconcelos, C.M.L., Dias, C.S., Barbosa, J.M., CondeGarcia, E.A., 2001. Averrhoa carambola L. – Estudo do seu componente cardiodepressor. Federac¸a˜o de Sociedades de Biologia Experimental (FeSBE, Brazil), Abstract No.03.002.