SciELO - Scientific Electronic Library Online

vol.35 número3-4Heterologous Expression of Syntaxin 6 in Saccharomyces cerevisiaeEchinococcus granulosus protoscolex formation in natural infections índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Biological Research

versión impresa ISSN 0716-9760

Biol. Res. v.35 n.3-4 Santiago  2002 

Biol Res 35:359-364, 2002


Effects of bicarbonate buffer on acetylcholine-,
adenosine 5´triphosphate-, and cyanide-induced
responses in the cat petrosal ganglion in vitro


Laboratorio de Fisiología Celular, Departamento de Biología, Facultad de Ciencias, Universidad de Chile


Acetylcholine (ACh), adenosine 5'-triphosphate (ATP) and sodium cyanide (NaCN) activate petrosal ganglion (PG) neurons in vitro, and evoke ventilatory reflexes in situ, which are abolished after bilateral chemosensory denervation. Because in our previous experiments we superfused the isolated PG with solutions free of CO2 /HCO3¯ buffer, we studied its effects on the PG responses evoked in vitro. PGs from adult cats were superfused at a constant pH, with HEPES-supplemented (5 mM) saline with or without CO2 /HCO3¯ (5% / 26.2 mM) buffer, and carotid (sinus) nerve frequency discharge (ƒCN) recorded. Increases in ƒCN evoked by ACh, ATP and NaCN in CO2-free saline were significantly reduced (P<0.05, Wilcoxon test) when CO2 / HCO3¯ was present in the superfusion medium. Thus, the presence of CO2 / HCO3¯ buffer appears to reduce PG neurons sensitivity to ACh, ATP and NaCN, an effect that may underlie the lack of ventilatory reflexes after bilateral chemodenervation.

Key terms: Petrosal Ganglion, Carotid sinus nerve, Carbon dioxide, Bicarbonate, Acetylcholine, Adenosine 5'-
triphosphate, Cyanide

The petrosal ganglion peripheral processes provide sensory innervation to the carotid bifurcation, the pharynx, and the posterior third of the tongue. Mechanosensory activity from the pharynx and the carotid sinus arises from direct stimulation of the nerve endings located in the target tissue. On the other hand, chemosensory activity from the carotid body and the tongue is generated by synaptic activation of the nerve terminals, secondary to the transduction of stimuli in specialized receptor cells. Although the precise transmitter(s) involved in the synapses between the carotid body receptor (glomus; type I) cells and the nerve terminals has not yet been well established, many molecules have been proposed to act as transmitters and/or modulators (17). We have recently reported that application of acetylcholine (ACh) to the petrosal ganglion in vitro produces a specific increase of carotid (sinus) nerve fibers frequency discharge (ƒCN), effect mimicked by nicotine, and blocked by hexamethonium and mecamylamine (2). Similarly, adenosine 5'-triphosphate (ATP), but not adenosine 5'-monophosphate, increases ƒCN when applied to the petrosal ganglion, although a modest increase in activity is also evoked in the pharyngeal branch of the glossopharyngeal nerve (6). Moreover, the responses evoked in the petrosal ganglion by ATP are inhibited by dopamine (10), while those elicited by ACh at least appear to be modulated by nitric oxide and dopamine (3,4). Conversely, hypoxia, the natural stimulus of the carotid body, had no effect on ƒCN, but oxidative phosphorylation impairment by sodium cyanide (NaCN) or sodium azide produces a dose-related increase in ƒCN, effect that is also modulated by dopamine (5). These data obtained in vitro indicate that petrosal ganglion neurons projecting through the carotid nerve are activated by ACh, ATP, and NaCN, which contrast with the abolition of reflex ventilatory responses induced by ACh and NaCN in whole animals after the bilateral section of the carotid and aortic nerves (9,15,24). However, in our previous in vitro experiments, the petrosal ganglia were superfused with HEPES-buffered saline, containing 4.17 mM HCO3¯ and equilibrated with a gas mixture devoid of CO2, while spontaneously breathing animals extracellular media is buffered by a mixture of about 26 mM HCO3¯ equilibrated with 5% CO2. Several studies indicate that the lack of CO2 /HCO3¯ buffer (8) or HEPES-buffered solutions modify the cellular responses of neurons in the central nervous system (12,13) and in the periphery (19). The most likely effect of HEPES buffer is to reduce the membrane potential (8,11,13,14,19) and intracellular pH (8,13), while increasing spontaneous activity (8). Similarly, in CO2 /HCO3¯ buffered solutions a reduction of receptor responses (19) and hyperpolarization accompanied by decreased excitability of central neurons (18) have been described. Thus, the lack of effect of both ACh and NaCN after denervation of peripheral chemoreceptors may be the result of reduced sensitivity of the petrosal ganglion neurons in CO2 /HCO3¯ buffered medium. We studied the effects of HEPES and CO2 /HCO3¯ buffered media on the responses evoked by ACh, NaCN and ATP on the cat petrosal ganglion in vitro.

Petrosal ganglia were obtained from 9 adult cats weighing 2.88 ± 0.22 kg (mean ± SEM) anaesthetized with sodium pentobarbitone (40 mg/kg, i.p.). The carotid bifurcation was exposed through a midline incision, and the carotid nerve was cut close to the carotid body. The glossopharyngeal nerve, cut distally to the apparent origin of the carotid nerve, was followed into the skull, the petrosal ganglion exposed, and the glossopharyngeal nerve cut near the central apparent limit of the ganglion. The tissue was placed in Hanks' solution, and the capsule and epineurium removed from over the ganglion and nerves, respectively. The ganglion was placed in a superfusion chamber, pined to the bottom of a 0.2 ml compartment, and superfused with saline solutions supplemented with 5 mM HEPES buffer, pH 7.4 at 38±0.5°C, flowing at 1.2-1.8 ml/min. Each ganglion was first superfused with 100% O2-equilibrated Hanks' solution (HCO3¯, 4.2 mM), termed here on CO2-free, and then with 5% CO2 and 95% O2-equilibrated Earle's solution (HCO3¯, 26.2 mM). The carotid nerve was placed on paired electrodes, and lifted into a compartment filled with mineral oil. The electrodes were connected in turn to an AC-preamplifier, and the resulting electroneurogram was amplified, displayed in oscilloscope, and recorded on a video-cassette recorder after digital encoding. The electroneurogram was fed to a spike amplitude discriminator, whose standardized pulses were digitally counted, to assess ƒCN, in Hz. The temperature of the chamber and the ƒCN were acquired, displayed, and recorded through a PC-based custom-made data acquisition system, at 1 Hz. ACh, ATP and NaCN, in doses of 1 to 3000 µg in 10 µl boluses, were applied 1-2 mm upstream from the ganglion. The change in frequency discharge (DƒCN) was calculated as the difference between the maximal frequency (max ƒCN) achieved during a single response and the mean basal activity (bas ƒCN), computed in a 30 s interval prior to a response. The relation between DƒCNs and the doses of any of the drugs used (X) was assessed using a simplex algorithm (23) to fit the standardized responses (R= DƒCN /max DƒCN) to a sigmoid curve (R =1/ (1+{ED50 / X}S) ). Statistical differences for dependent samples were assessed by Wilcoxon test.

The application of ACh-, ATP-, and NaCN to the petrosal ganglion superfused with CO2-free solution equilibrated with 100% O2 produced a dose-dependent increase in ƒCN. The relative magnitude of the responses evoked by the three agents was similar to that previously described (2,3,5). Switching from CO2-free Hanks' solution to 5% CO2 and 95% O2-equilibrated Earle's solution had no effect on basal ƒCN or in the magnitude of the compound action potential recorded from the carotid nerve by supra-threshold electrical stimulation of the ganglion. However, applications of ACh, ATP, and NaCN elicited increases of ƒCN that were almost always smaller in amplitude than those induced by the same doses in control, CO2-free conditions. The data obtained for each one of the agents during CO2-free and CO2 /HCO3¯ superfusion were standardized to the maximal response evoked by each applied dose. Figure 1 shows the mean standardized responses evoked by at least three different doses of ATP, ACh, and NaCN in seven, six, and four different experiments, respectively. These results indicate that the responses evoked under CO2 /HCO3¯ superfusion were statistically significantly different from those evoked in CO2-free medium (p<0.05 Wilcoxon paired two-tail test). Switching back to CO2-free solution had no effect on basal ƒCN, but ACh-, ATP-, and NaCN-elicited responses were of similar or larger amplitude than those previously recorded in control CO2- free conditions. Figure 2A shows the responses evoked by a maximal ATP dose (3 mg) during CO2-free superfusion, during CO2 /HCO3¯ superfusion, and after removing CO2 in the same preparation, showing a reduction of the maximal response during CO2 /HCO3¯ superfusion and an almost complete recovery upon returning to CO2- free condition. Figure 2B illustrates the dose-response curves fitted to the ATP-induced responses in one petrosal ganglion, showing that during CO2 /HCO3¯ superfusion ED50 increased (from 0.45 to 165.35 mg) and the slope (S) of the curve was reduced (from 1.02 to 0.27) with respect to control CO2- free conditions. This effect was partly reversed upon removal of CO2 /HCO3¯ from the medium.

Figure 1. Responses induced in carotid nerve by ATP, ACh, and NaCN in seven, six, and four preparations, respectively. CO2-free, 4.2 mM HCO3¯ (empty columns), and 5% CO2 , 26.2 mM HCO3¯ (dotted columns) media. * indicates P<0.05 (Wilcoxon paired test).

Figure 2. ATP-induced responses recorded from the carotid nerve. A. Frequency discharge (ƒCN) induced by a single ATP dose (3 mg at arrowhead), during superfusion with CO2-free saline (empty bar), during 5% CO2 (filled bar), and after CO2 removal (lined bar) on the same preparation. B. Dose-response curves of the increases in frequency discharge (DƒCN) induced by ATP on the same preparation during CO2-free (filled circles, continuous line), during 5% CO2 content (empty circles, interrupted line), and after CO2 removal (empty diamonds, dotted line).

Our results indicate that the sensitivity and magnitude of the responses induced by ACh, ATP and NaCN in the superfused petrosal ganglion are largely dependent upon the nature of the buffer present in the extracellular medium, at the same pH. Responses evoked by ACh, ATP and NaCN in CO2- free conditions were reduced in amplitude when they were evoked in the presence of CO2 /HCO3¯. Moreover, the threshold to evoke the responses was increased during CO2 /HCO3¯ superfusion. This reduction in sensitivity during CO2 /HCO3¯ superfusion might be partly explained if petrosal ganglion neurons are hyperpolarized by the presence of CO2 /HCO3¯ in the extracellular medium, as pyramidal hippocampal neurons are (11,12,18). However, given that in mammalian tissue the physiological buffer is CO2 /HCO3¯, its substitution by HEPES buffer may increase the sensitivity of petrosal ganglion neurons by reducing the membrane potential (Vm) (11,12,13,14,18) and by increasing the excitability (11), as is the case with hippocampal neurons. This depolarizing effect observed by the exclusion of CO2 /HCO3¯ buffer and its substitution by HEPES buffer, are also observed in salamander retinal cells (19). The overall electrophysiological properties of cat petrosal ganglion neurons do not differ significantly when recorded in CO2 /HCO3¯ buffer (1) with respect to those recorded in HEPES buffer (7,25). However, no data are available on the responses evoked from single petrosal neurons in the presence and absence of CO2 /HCO3¯.

The maximal response evoked by ACh, ATP or NaCN could not be achieved during the CO2 /HCO3¯ superfusion, at least within the dose range tested. In pyramidal hippocampal neurons, CO2 /HCO3¯ superfusion reduces the excitability, even when Vm is held constant (18). Thus, maximal frequency discharge may not be achieved under this condition. It is noteworthy that the basal discharge, the magnitude or shape of the electrically evoked compound action potential were not apparently modified by CO2 /HCO3¯ superfusion, as expected if petrosal ganglion neurons would behave as pyramidal hippocampal neurons (8,18). However, the whole nerve recording technique used in our experiments may reduce the impact of individual changes in neuronal activity in the population of fibers within the nerve. Moreover, the effect of CO2 /HCO3¯ superfusion on the amplitude of the compound action potential would depend on the magnitude of the reduction of Vm, because at constant Vm the amplitude of the action potential is not affected by CO2 /HCO3¯ in pyramidal hippocampal neurons (18). Additionally, the effect of the lack of CO2 /HCO3¯ may also depend on the architecture of the system, because neurons located deep inside hippocampal tissue slices do hyperpolarize when HEPES buffer replaces CO2 /HCO3¯ (8). Thus, the overall effect of the absence of CO2 /HCO3¯ buffer on Vm and excitability can not be assessed in our preparation. Nonetheless, responses evoked by ACh, ATP and NaCN in CO2-free medium present lower thresholds and ED50s than their counterparts evoked in the presence of CO2 /HCO3¯.

The buffer present in the medium has been reported to modify the responses evoked in the carotid nerve by stimulation of the carotid body. Carotid chemosensory responses present enhanced kinetics and maximal response when evoked by hypoxia, cyanide and hypercapnia in the presence of CO2 /HCO3¯ buffer with respect to HEPES-buffered medium (22). Similarly, the hypoxia-induced efflux of catecholamines from the carotid body was enhanced by CO2 /HCO3¯ buffer with respect to CO2-free medium (21). However, the magnitude of the responses elicited by nicotine remained almost unaltered by the nature of the buffer, although the sensitivity of the response appears slightly increased by CO2 /HCO3¯ buffer (22). Thus, responses elicited from the carotid body by natural stimuli and pharmacological agents are enhanced by the presence of CO2 /HCO3¯, while the responses elicited by ACh, ATP and cyanide in the soma of petrosal ganglion neurons projecting through the carotid nerve behave in the opposite way. In our preparation basal ƒCN was not modified by switching to CO2 /HCO3¯ superfusion and viceversa, while the basal chemosensory discharge increases with increasing CO2 levels (see 16,20). Although basal ƒCN was not modified by the presence of CO2 / HCO3¯, the modification of intracellular pH may be partly responsible for the reduced response in our preparation. Switching from CO2 /HCO3¯- to HEPES-buffered solution produced a sustained acidification of hippocampal (8) and vagal (13) neurons, accompanied by reduced spontaneous activity (8) and increased excitability (11). If HEPES buffer accumulates within the cell, it may modify the cellular pH regulatory mechanisms (26), but HEPES concentration was low (5 mM) and was always present in our preparation. Thus, more than an enhancement of the responses by HEPES buffer per-se, the lack of CO2 /HCO3¯ in the medium appears to account for the enhancement of ACh-, ATP- and NaCN-induced reponses in the petrosal ganglion in vitro.

Reflex vascular or ventilatory responses evoked by intravascular applications of ACh, ATP and NaCN are abolished after peripheral chemodenervation (9,15,24). Although petrosal ganglion neurons are sensitive to ACh, ATP and NaCN, in physiological conditions the neurons would present a reduced sensitivity. Thus, low or no afferent activity may be evoked within the ganglion by intra-vascular administration of these drugs within the experimental range used.


We would like to thank Dr. Rodrigo Iturriaga for his criticism on this manuscript, and Mrs. Carolina Larraín for her assistance in the preparation of this manuscript. This work was supported by grant 199-0030 from the National Fund for Scientific and Technological Development of Chile (FONDECYT).


(1) ALCAYAGA J, ARROYO J (1996) Responses of cat petrosal ganglion neurons are modified by the presence of carotid body cells in tissue cultures. Adv Exp Med Biol 410: 195-201         [ Links ]

(2) ALCAYAGA J, ITURRIAGA R, VARAS R, ARROYO J, ZAPATA P (1998) Selective activation of carotid nerve fibers by acetylcholine applied to the cat petrosal ganglion in vitro. Brain Res 786: 47-54         [ Links ]

(3) ALCAYAGA J, BARRIOS M, BUSTOS F, MIRANDA G, MOLINA MJ, ITURRIAGA R (1999a) Modulatory effect of nitric oxide on acetylcholine-induced activation of cat petrosal ganglion neurons in vitro. Brain Res 825: 194-198         [ Links ]

(4) ALCAYAGA J, VARAS R, ARROYO J, ITURRIAGA R, ZAPATA P (1999b) Dopamine modulates carotid nerve responses induced by acetylcholine on the cat petrosal ganglion in vitro. Brain Res 831: 97-103         [ Links ]

(5) ALCAYAGA J, VARAS R, ARROYO J, ITURRIAGA R, ZAPATA P (1999c) Responses to hypoxia of petrosal ganglia in vitro. Brain Res 845: 28-34         [ Links ]

(6) ALCAYAGA J, CERPA V, RETAMAL M, ARROYO J, ITURRIAGA R, ZAPATA P (2000) Adenosine triphosphate-induced peripheral nerve discharges generated from the cat petrosal ganglion in vitro. Neurosci Lett 282: 185-188         [ Links ]

(7) BELMONTE C, GALLEGO R (1983) Membrane properties of cat sensory neurons with chemoreceptors and baroreceptor endings. J Physiol, London 342: 603-614         [ Links ]

(8) BONNET U, WIEMANN M, BINGMANN D (1998) CO2/HCO3¯ -withdrawal from the bath medium of hippocampal slices: biphasic effect on intracellular pH and bioelectric activity of CA3-neurons. Brain Res 796: 161-170         [ Links ]

(9) BOUVEROT P, FLANDROIS R, PUCCINELLI R, DEJOURS P (1965) Étude de rôle des chémorécepteurs artérieles dans la régulation de la respiration pulmonaire chez le chien éveillé. Arch Int Pharmacodyn Ther 157: 253-271         [ Links ]

(10) CERPA V, ALCAYAGA J (2000) Dopamine inhibits ATP-induced responses in the cat petrosal ganglion in vitro. Biol Res 33: R-33         [ Links ]

(11) CHURCH J (1992) A change from HCO3¯- to HEPES-buffered medium modifies membrane properties of rat CA1 pyramidal neurones in vitro. J Physiol, London 455: 51-71         [ Links ]

(12) CHURCH J, MCLENNAN H (1989) Electrophysiological properties of rat CA1 pyramidal neurones in vitro modified by changes in extracellular bicarbonate. J Physiol, London 415: 85-108         [ Links ]

(13) COWAN AI, MARTIN RL (1995) Simultaneous measurement of pH and membrane potential in rat dorsal vagal motoneurons during normoxia and hypoxia: a comparison in bicarbonate and HEPES buffers. J Neurophysiol 74: 2713-2721         [ Links ]

(14) COWAN AI, MARTIN RL (1996) Inonic basis of the membrane potential responses of rat dorsal vagal motoneurons to HEPES buffer. Brain Res 717: 69-75         [ Links ]

(15) EUGENIN J, LARRAIN C, ZAPATA P (1989) Correlative contribution of carotid and aortic afferences to the ventilatory chemosensory drive in steady-state normoxia and to the ventilatory chemoreflexes induced by transient hypoxia. Arch Biol Med Exp 22: 395-408         [ Links ]

(16) EYZAGUIRRE C, ZAPATA P (1984) Perspectives in carotid body research. J Appl Physiol 57: 931-957         [ Links ]

(17) GONZALEZ C, ALMARAZ L, OBESO A, RIGUAL R (1994) Carotid body chemoreceptors: from natural stimuli to sensory discharges. Physiol Rev 74: 829-898         [ Links ]

(18) GU XQ, YAO H, HADDAD GG (2000) Effect of extracellular HCO3¯ on Na+ channel characteristics in hippocampal CA1 neurons. J Neurophysiol 84: 2477-2483         [ Links ]

(19) HARE WA, OWEN WG (1998) Effects of bicarbonate versus HEPES buffering on measured properties of neurons in the salamander retina. Vis Neurosci 15: 263-271         [ Links ]

(20) HEMPLEMAN SC, POWELL FL, PRISK GK (1992) Avian arterial chemoreceptor responses to steps of CO2 and O2. Respir Physiol 90: 325-340         [ Links ]

(21) ITURRIAGA R, ALCAYAGA J (1998) Effects of CO2/HCO3¯ on catecholamine efflux from the cat carotid body. J Appl Physiol 84: 60-68         [ Links ]

(22) ITURRIAGA R, LAHIRI S (1991) Carotid body chemoreception in the absence and presence of CO2/HCO3¯ . Brain Res 568: 253-260         [ Links ]

(23) JOHNSTON A (1985) SIMP: a computer program in BASIC for nonlinear curve fitting. J Pharmacol Meth 14: 323-329         [ Links ]

(24) SERANI A, ZAPATA P (1981) Relative contribution of carotid and aortic bodies to cyanide-induced ventilatory responses in the cat. Arch Int Pharmacodyn Ther 252: 284-297         [ Links ]

(25) VARAS R, ALCAYAGA J, ZAPATA P (2000) Acetylcholine sensitivity in primary sensory neurons dissociated from the cat petrosal ganglion. Brain Res 882: 201-205         [ Links ]

(26) WALSH PJ (1990) Fish hepatocytes accumulate HEPES: a potential source of error in studies of intracellular pH regulation. J Exp Biol 148: 495-499         [ Links ]

Correspondence to:.Julio Alcayaga. Laboratorio de Fisiología Celular. Departamento de Biología. Facultad de Ciencias, Universidad de Chile. Casilla 653. Santiago. Chile. Phone: (56-2) 678-7254. FAX: (56-2) 271-2983. email:

Received: June 25, 2002. Accepted: July 30, 2002

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons