GABAergic and Glutaminergic Modulation of Centrally Evoked Arrhythmias in Rats
Abstract A standard electrical stimulus applied to the posterior hypothalamus evoked cardiac arrhythmogenic responses in the spontaneously hypertensive rat. Isolated premature ventricular beats or doublets and nonsustained ventricular tachycardic salvos were observed. This effect was associated with a large rise in blood pressure (79±3 mm Hg). The same stimulus in normotensive Wistar-Kyoto rats produced no significant cardiac arrhythmias, and the rise in blood pressure was smaller (36±2 mm Hg). We investigated the influence of baclofen, a GABAB receptor agonist, and two N-methyl-d-aspartate receptor antagonists on the arrhythmogenic response to hypothalamic stimulation. Intravenous baclofen (3 mg/kg) had no effect in the normotensive Wistar-Kyoto rats, but in the spontaneously hypertensive rats it enhanced the adjusted mean value of the number of extrasystoles from 0.5±0.5 to 18±1 (P<.001). This value was also increased (from 3±1 to 17±1, P<.001) by an intracisternal injection of baclofen (1 μg/kg). This facilitatory effect of baclofen was prevented by treatment with atenolol (0.5 mg/kg). Two glutamate receptor antagonists, ketamine (7.5 mg/kg IV) and kynurenic acid (200 μg/kg intracerebroventricularly), prevented both the arrhythmogenic response to the hypothalamic stimulation and its facilitation by baclofen. The study confirms that hypothalamic stimulation facilitates the development of arrhythmias through a sympathetic drive and that these arrhythmias are easier to induce in spontaneously hypertensive rats than in normotensive Wistar-Kyoto rats. Both the central GABAergic and the glutamatergic systems are implicated in the development of these ventricular arrhythmias, since baclofen could disinhibit the glutamatergic central pathway. These results could account for the ability of the spontaneously hypertensive rats to develop ventricular arrhythmias of central origin.
Ventricular arrhythmias have been considered to be triggered by several factors, including sympathetic nervous system activation.1 The facilitating influence of stress is well documented.2 Tachycardia and ventricular fibrillation have also been described during epileptic seizures.3 4 Our aim was to study the role of the sympathetic nervous system and its central neurotransmitter activators in the development of ventricular arrhythmias.
First, we developed an experimental model in which ventricular tachyarrhythmias were induced by electrical stimulation of the posterior hypothalamus. The specific area chosen represents an integrative center of sympathetic regulation of both the vasomotor tone and the heart rate.5 We then studied the roles of the GABAB system and of the glutamate receptors in the modulation of this arrhythmogenic response. This experimental approach was justified by numerous reports of the high density of both GABAB and glutamate receptors in the neuronal structures involved in cardiovascular regulation.6 7 8 9 10 11 12 13
In this report, the unexpected facilitating effect of baclofen, a GABAB receptor agonist, on the arrhythmogenic response to hypothalamic stimulation in SHRs is described. This response was possibly modulated by various NMDA receptor antagonists.
Two strains of rats (Okamoto) from Janvier Laboratory were used: SHRs and their normotensive controls, WKY rats. Eleven- to 13-week-old animals (240 to 350 g) were anesthetized with sodium pentobarbitone (50 mg/kg IP). One femoral vein was catheterized to allow intravenous injections and one femoral artery to record MAP via a Statham P23Db pressure transducer connected to a pressure processor and recorder (Gould BS 271); the HR was obtained from the rapid running of the pressure recording. Animals were tracheotomized, immobilized with pancuronium bromide (1 mg/kg IV), and artificially ventilated with room air (Hugo Sachs Electronic 7025).
The electrical activity of the six standard cardiac derivations D1, D2, D3, aVR, aVL, and aVF was recorded with an electrocardiograph (Hellige EK 512).
The head of the rat was placed horizontally in a stereotaxic frame (Unimecanique). Electrical stimulation was performed with a concentric bipolar electrode (OD, 250 μm) (Phymep).
The electrode was inserted into the posterior hypothalamus at the following stereotaxic coordinates: bregma, −4 mm; lateral, 0.5 mm; and depth, −8.3 mm according to the atlas of Paxinos and Watson.14 At the end of each experiment, an electrolytic lesion was effected with a current of 1 mA during 10 seconds. After the animal was killed, the brain was removed and fixed in a 10% formol saline solution to verify the electrode position (Fig 1⇓).
Stimulation was obtained through a 2-second pulse with a current of 150 μA, 100 Hz (square current).
A 10-μL Hamilton microsyringe (Bioblock) was used for both the intracisternal and intracerebroventricular injections. The intracisternal injections were made in the fourth ventricle through the atlanto-occipital membrane by use of a micromanipulator (Unimecanique) to hold the microsyringe at an angle of 30° with the horizontal. The intracerebroventricular injections were made in one of the lateral ventricles at the following stereotaxic coordinates: bregma, −0.8 mm; lateral, 1.5 mm; and depth, −4.5 mm according to the atlas of Paxinos and Watson.14 A solution of Evans blue dye injected under the same conditions at the end of the experiment allowed verification of the site of injection of the drugs and their distribution in the ventricular space.
The following drugs were used: sodium pentobarbitone (Nembutal, Abbott Laboratories), pancuronium bromide (Pavulon, Organon Teknica), kynurenic acid (Tocris Neuramin), ketamine (Ketalar, Parke Davis), d,l-baclofen (Sigma Chemical Co), and atenolol (Tenormine, ICI Pharma). Kynurenic acid solutions were dissolved in saline (0.9%) and adjusted to physiological pH.
Acute Extrinsic Elevation of Blood Pressure
After a median abdominal laparotomy, the abdominal aorta was exposed and manually compressed against the vertex to a degree sufficient to increase MAP to the same kinetics and level obtained during hypothalamic stimulation. The MAP was monitored through a carotid catheter.
Quantification of Ventricular Cardiac Arrhythmias
The electrically induced PVBs were quantified according to number and coupling mode. An NSVT was defined as the occurrence of at least three successive high-frequency PVBs (grade 4 of the Lown classification of 197115 16 ). We counted the number of PVBs per stimulation (single PVB, doublet, NSVT) and the numbers of rats with arrhythmias and with NSVT (Tables 1⇓ and 2⇓).
Statistical analysis was done only on the number of PVBs per stimulation; animals with no arrhythmic response were scored 0. Thus, all the experiments were taken into account and all the results were then averaged within each group.
All experiments were analyzed in a similar manner. The baseline characteristics (HR, MAP before and during stimulation and its difference, pressor response, and number of induced arrhythmias) of the control and experimental groups of rats were compared by the Mann-Whitney U test. If a difference in the baseline characteristics was found to be significant between groups, an adjustment was made when meaningful in the analysis.
A multivariate linear model was fitted for each experiment with a downward stepwise linear regression.17 Linear regression took into account the effects of other factors that may have affected the number of recorded arrhythmias. Having “removed” the possible effects of other factors enabled us to narrow down to the single effect of the drug. We could then compute a mean value of arrhythmias for both the control and the experimental groups, adjusting for the possible effects of other factors. The values thus computed were called “the mean adjusted values.” The comparisons we then made using these adjusted means between the control and experimental groups were the reflection of the sole effect of the drug. Categorical variables included in the model used were drug, time, and their interactions. Covariates (the adjustment factors) were HR, MAP, and the intensity of the pressor response.
If the interaction term between drug and time was found to be significant, the effect of the drug was tested at each individual time to determine when the drug produced its statistically significant effect. If the interaction term was not significant but the drug effect was significant, the main effect of the drug across all the categories of time was tested and presented.
The results are reported as mean±SEM in Tables 1 through 6. The mean adjusted values, from which conclusions were drawn, are presented and discussed in the text. All tests were two-tailed, and the results were considered significant at P<.05. All analyses were conducted with systat software (SYSTAT Inc).
Cardiovascular Responses to Hypothalamic Stimulation
WKY rats (Table 1⇑). Posterior hypothalamic stimulation always induced a fast and transient marked increase in MAP without any significant change in HR. It never induced any ventricular arrhythmia.
SHRs (Table 1⇑). Posterior hypothalamic stimulation frequently triggered ventricular arrhythmias (induced arrhythmias): 6 of 10 SHRs exhibited isolated or doublet PVBs (Fig 2⇓) and in one experiment, NSVT. Neither sustained ventricular tachycardia nor ventricular fibrillation was observed. Hypothalamic stimulation also induced a large, rapid, and transient increase in MAP without a change in HR in the 4 rats without arrhythmia. In the other 6 rats, arrhythmias occurred at the maximal elevation of blood pressure. It was impossible to estimate HR during the presence of the arrhythmias; however, no rhythmic disturbances preceded the observed arrhythmias. The role of an acute increase in MAP in the triggering of arrhythmias was studied in 5 other rats by acutely raising MAP by compressing the aorta. As shown in Table 2⇑, PVBs were effectively induced at the maximal blood pressure elevation.
WKY rats treated with intravenous baclofen. No arrhythmic response to the electrical hypothalamic stimulation was observed.
SHRs treated with intracisternal baclofen (Table 3⇓). The induction of ventricular arrhythmias was facilitated by intracisternal baclofen: all the rats in the treated group developed ventricular arrhythmias that were more frequent and complex (NSVT) than in the control rats. The adjusted mean value of the induced PVBs was increased from 3±1 in the control animals to 17±1 in the treated animals (P<.001).
SHRs treated with intravenous baclofen (Table 3⇑). Reminiscent of the results obtained with intracisternal injections, intravenous injections of baclofen facilitated the induction of ventricular arrhythmias; all the treated rats became responsive, and the arrhythmias were more frequent and complex (NSVT) (Fig 3⇓). The mean adjusted value of the induced PVBs was increased from 0.5±0.5 in the control animals to 18±1 in the treated animals (P<.001).
SHRs treated with intravenous ketamine (Table 4⇓). Ketamine had no significant effect on the induced PVBs. The only factors statistically associated with the number of PVBs were HR and MAP.
SHRs treated with intracerebroventricular kynurenic acid (Table 4⇑). Kynurenic acid (200 μg/kg) was given intracerebroventricularly because of its weak transmission across the blood-brain barrier.18 It had no significant effect on the induced PVBs. The only factor statistically associated with the number of PVBs was the intensity of the pressor response.
SHRs treated with intravenous baclofen after ketamine pretreatment (Table 5⇓). Ketamine prevented the baclofen facilitation of induced ventricular arrhythmias. The adjusted mean value of the PVBs decreased from 10±1 in the control group to 6±1 in the pretreated group (P=.04).
SHRs treated with intravenous baclofen after kynurenic acid pretreatment (Table 5⇑). Like ketamine, kynurenic acid prevented the baclofen facilitation of the induced ventricular arrhythmias. Ten minutes after baclofen administration, the adjusted mean value of the induced PVBs was decreased from 10±2 in the control animals to −2±2 in the pretreated animals (P<.001); at 20 and 30 minutes after baclofen administration, the values decreased from 20±2 to −2±2 (P<.001) and from 20±2 to −1±2 (P<.001), respectively.
Role of the Sympathetic Nervous System (Table 6⇓)
The administration of the β-blocker atenolol after the baclofen facilitation of the induced arrhythmias caused a decrease in the adjusted mean values of the PVBs from 17±2 to 7±2 (P=.001) and from 15±2 to 6±2 (P=.002) 5 and 10 minutes, respectively, after administration.
Cardiovascular Responses to Hypothalamic Stimulation
We studied the role of the sympathetic nervous system and its central neurotransmitter activators in the development of ventricular arrhythmias by examining the effects of pharmacological agents on the frequency of the development of ventricular arrhythmias provoked by the electrical stimulation of the posterior hypothalamus in pentobarbitone-anesthetized rats. The latter procedure has been used previously to induce cardiac arrhythmias and/or to lower the ventricular fibrillation threshold in various animal species.19 20 21
In our experiments, when an electrical stimulus compatible with a restricted spread around the point of stimulation to avoid tissue damage was applied to a discrete hypothalamic area, no significant or reproducible arrhythmogenic response was obtained in normotensive WKY rats, but MAP increased. In contrast, the same experimental conditions induced ventricular arrhythmias in a large proportion of the SHRs. This response was associated with larger increases of blood pressure than in WKY rats (Table 1⇑).
We suggest that the sympathetic nervous system was the principal triggering element of the induced arrhythmias in the SHRs because (1) as a sympathoexcitatory area,5 posterior hypothalamic stimulation could increase MAP and ventricular excitability, and (2) both stellectomy and β-blocker treatment could prevent ventricular arrhythmias of central origin.20 22 In addition, the left ventricular hypertrophy combined with the sympathetic hyperactivity to the cardiovascular system and the high vascular reactivity of the SHRs23 24 25 26 27 could account for the arrhythmogenic and the exaggerated pressor responses regularly observed in these animals. Furthermore, we could not eliminate the specific effect of the acute elevation of MAP as one of the triggering elements of the arrhythmias observed, because an extrinsic acute elevation of the MAP could induce ventricular arrhythmias in SHRs (Table 2⇑).
GABAergic and Glutaminergic Neurotransmission
The capacity of both the central GABAergic and the glutamatergic systems to modulate the sympathetic traffic to the cardiovascular system is widely documented. Indeed, the inhibitory amino acid GABA is the likely neurotransmitter that provides, through short ascending projections to the retrofacial neurons at the ventrolateral medulla, the inhibitory mechanism essential for the negative feedback loop inherent to the arterial baroreceptor reflex. In contrast, the excitatory amino acid glutamate seems to play a key role in these descending central retrofacial pathways that activate the spinal sympathetic preganglionic neurons.28 We examined the effects of different central agonists and antagonists of the amino acid receptors in the genesis of the ventricular arrhythmias.
We studied the effects of baclofen on the induction of ventricular arrhythmias evoked by hypothalamic stimulation. Baclofen is a lipophilic analogue of GABA described as early as the 1980s as a selective agonist of GABAB receptor subtype.29 It crosses the blood-brain barrier and has been used to treat spastic syndromes for many years. Its actions are considered to be of central origin.
We previously demonstrated that baclofen induced a marked hypotensive effect originating from the rostral part of the brain in anesthetized animals.30 We also reported that in rabbits, baclofen reduced the myocardial oxygen consumption during electrical stimulation of the hypothalamic area.31
Although the administration of intravenous baclofen caused no arrhythmogenic activity on electrical stimulation in normotensive WKY rats, it unexpectedly facilitated the arrhythmogenic effect of the same stimulus in SHRs. This effect was most probably of central origin, since a very low dose (1 μg/kg) injected intracisternally facilitated the arrhythmogenic effect of the hypothalamic stimulation in the SHRs (Table 3⇑). This dose was 3.5(log) times less than that necessary to obtain the same effect after systemic injection.
Baclofen acted through an activation of the sympathetic traffic, since (1) baclofen could also increase MAP (data not shown), probably through an activation of the sympathetic pathway to the vascular tree,32 and (2) β-blockers could prevent ventricular arrhythmias induced by posterior hypothalamic stimulation and facilitated by baclofen (Table 6⇑).
As previously reported, GABA probably inhibits the central retrofacial sympathoexcitatory neurons through short ascending projections belonging to the baroreflex arc. Since baclofen is known to be a GABAB agonist, the increase in arrhythmias and MAP caused by its administration were unexpected. It could be that it acted through an inhibition of the inhibitory pathways of the baroreflex arc, since the activation of some GABAB receptors could inhibit the release of various neurotransmitters. Thus, baclofen could inhibit both the release of GABA and the release of the neurotransmitters that activate the GABA neurons.33 34 The first relay of the baroreflex arc, the NTS, could be one of the targets of the effects of baclofen in our SHRs, since (1) baclofen injected intracisternally could easily reach the NTS and (2) the effect of injecting baclofen directly into the NTS corroborated the effect of baclofen on MAP.6 34 Baclofen could thus disinhibit the central sympathoexcitatory neurons, which are partly glutamatergic.
We studied the effects of two antagonists on different sites of the NMDA receptors to verify the involvement of the glutamatergic pathways in the arrhythmogenic response to hypothalamic stimulation and its facilitation by baclofen. Ketamine, a channel blocker at the phencyclidine site, had no significant effect on the induction of ventricular arrhythmias, but it prevented baclofen facilitation (Tables 4⇑ and 5⇑). However, because of its short half-life,35 the effect was brief. Since, in our experimental system, ketamine could also act through a functional inhibition of the sympathetic activity,36 we attempted to confirm the involvement of NMDA receptors by using kynurenic acid, a selective antagonist at the glycine site of these receptors. Like ketamine, kynurenic acid had no significant effect on the induction of ventricular arrhythmias, but it prevented baclofen facilitation (Tables 4⇑ and 5⇑). Moreover, the induction of PVBs could be completely prevented by either ketamine or kynurenic acid, leading one to suspect that the number of induced PVBs was too low to be significantly reduced by these two drugs. These results strongly suggest (1) the involvement of the central glutamatergic sympathoexcitatory pathway in the arrhythmogenicity of the hypothalamic stimulation and (2) the disinhibition of this pathway by baclofen. Of interest, we did not observe a significant elevation in baseline blood pressure subsequent to intracerebroventricular injection of kynurenic acid, as Sun et al did.37 These results could be explained by the different routes of administration and also by the strain differences.
This study has shown that it is easier to induce ventricular arrhythmias by posterior hypothalamic stimulation in SHRs than in WKY rats. The principal triggering element was the hyperactivation of the sympathetic pathway to the heart. The peripheral vascular tree might also play a role, since an acute elevation of the MAP could by itself induce ventricular arrhythmias.
Both the central NMDA (excitatory) and GABAB (inhibitory) receptor pathways were involved in the induction of ventricular arrhythmias in the SHRs. The unexpected effect of baclofen observed in SHRs suggested that this drug acted mainly through an inhibition of the inhibitory pathways of the baroreflex arc in this strain of rats. We suggest that this disinhibition could partly explain the sympathetic hyperactivity of the SHRs and their ability to develop ventricular arrhythmias of central origin.
Selected Abbreviations and Acronyms
|MAP||=||mean arterial blood pressure|
|NSVT||=||nonsustained ventricular tachycardia|
|NTS||=||nucleus tractus solitarius|
|PVB||=||premature ventricular beat|
|SHR||=||spontaneously hypertensive rat(s)|
Dr Crambes was supported by la Fondation pour la Recherche Médicale and la Fédération Française de Cardiologie. The help of Dr Daniel Bichet and Hilda Warwick in rewriting the manuscript was most valuable.
Reprint requests to Pascal Bousquet, Laboratoire de Pharmacologie Cardiovasculaire et Rénale, CNRS URA 589, Faculté de Médecine, Université Louis Pasteur, 11 rue Humann, 67000 Strasbourg, France.
- Received April 5, 1994.
- Revision received June 1, 1994.
- Accepted September 6, 1995.
Verrier RL. Behavioral stress, myocardial ischemia, and arrhythmias. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders Co; 1990:343-352.
Achrafi H. La mort subite épileptique: manifestations cardiaques. Act Méd Int. 1991;8:1807-1809.
Kubo T, Nagura J, Kihara M, Misu Y. Cardiovascular effects of L-glutamate and γ-aminobutyric acid injected into the rostral ventrolateral medulla in normotensive and spontaneously hypertensive rats. Arch Int Pharmacodyn. 1986;279:150-161.
Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 2nd ed. London, UK: Academic Press; 1986:(Fig 34).
Lown B, Wolf M. Approaches to sudden death from coronary heart disease. Circulation. 1971;44:130-142.
Podrid PJ, Lown B, Graboys TB, Lampert S. Use of short-term drug testing as part of a systematic approach for evaluation of antiarrhythmic drugs. Circulation. 1986;73(suppl II):II-81-II-91.
Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and Other Multivariable Methods. 2nd ed. Boston, Mass: PWS-Kent; 1988:416-444.
Stone TW. Neuropharmacology of quinolinic and kynurenic acids. Pharmacol Rev. 1993;45:309-379.
Korteweg GCJ, Boeles JTF, Ten Cate J. Influence of stimulation of some sub-cortical areas on electrocardiogram. J Neurophysiol. 1957;20:100-107.
Manning JW, Cotten, M De V. Mechanism of cardiac arrhythmias induced by diencephalic stimulation. Am J Physiol. 1962;203:1120-1124.
Verrier RL, Calvert A, Lown B. Effect of posterior hypothalamic stimulation on ventricular fibrillation threshold. Am J Physiol. 1975;228:923-927.
Adams MA, Bobik A, Korner PI. Differential development of vascular and cardiac hypertrophy in genetic hypertension: relation to sympathetic function. Hypertension. 1989;14:191-202.
Clark DWJ, Jones DR, Phelan EL, Devine CE. Blood pressure and vascular resistance in genetically hypertensive rats treated at birth with 6-hydroxydopamine. Circ Res. 1978;43:293-300.
Provoost AP. Sympathectomy and the development of hypertension in rats. In: De Jong W, ed. Handbook of Hypertension, vol 4: Experimental and Genetic Models of Hypertension. Amsterdam, Netherlands: Elsevier Science Publishers; 1984:495-517.
Takeda K, Bunag RD. Sympathetic hyperactivity during hypothalamic stimulation in spontaneously hypertensive rats. J Clin Invest. 1978;62:642-648.
Chalmers J, Arnolda L, Llewellyn-Smith I, Minson J, Pilowsky P. Central nervous control of blood pressure. In: Swales JD, ed. Textbook of Hypertension. Boston, Mass: Blackwell; 1994:409-426.
Persson B, Henning M. Central cardiovascular and biochemical effects of baclofen in the conscious rat. Pharm Pharmacol. 1980;32:417-422.