INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DOPAMINE (DA) is the predominant catecholamine (CA) synthesized, taken up, and stored in dense-cored vesicles by glomus cells (8, 14) of the carotid body (CB). The observation that after previous incubation with [³H]tyrosine for 2-3 h the amount of radiolabeled DA released from CBs superfused in vitro was roughly proportional to the intensity of hypoxia (9) has contributed to the hypothesis that DA is the excitatory transmitter in the CB (14, 15). However, the radiolabeling method precludes the study of the temporal correlation between chemosensory excitation and DA efflux induced by a stimulus because this method is based on the determination of overflow fractions collected for 10 or more minutes. Improvement of temporal resolution of CA determination has been possible by using electrochemical methods that measure CA oxidation currents in real time (12, 25). Recently, Donnelly (5, 6), Buerk et al. (4), and Iturriaga et al. (16) have used amperometric methods to study DA release along with chemosensory excitation induced by hypoxia, flow interruption, and NaCN in rat and cat CBs in vitro.

Because previous studies from this laboratory (16, 37) showed that the amplitude and time course of CA efflux (CA) from cat CBs superfused in vitro were not correlated with chemosensory responses evoked by NaCN and hypoxia, we concluded that DA efflux is not essential for hypoxic-induced chemosensory excitation of the CB. However, these studies were performed with CBs superfused with Tyrode solution free of CO₂-, a condition that is known to delay the carotid chemosensory response to hypoxia in cats (17, 32), and we did not study the effect of hypercapnia on CA. Such study of the effects of CO₂- on time course and amplitude of the CA induced by hypoxia seems to be necessary because carotid chemosensory response to hypoxia interacts with CO₂-H⁺ stimuli (7, 11, 21). The study of the effects of uncompensated hypercapnia (high PCO₂ and low pH) on CA is also needed because hypercapnia is a strong chemoreceptor stimulus by itself, and it has been suggested that the hypercapnic excitatory mechanism is different from that of hypoxia (17, 22). However, not only did hypercapnia enhance the responses to hypoxia in situ (11, 21) and in vitro (7, 17, 29), but CO₂- at a constant pH of 7.40 also produced that effect. In fact, Iturriaga and Lahiri (17) and Shirahata and Fitzgerald (32) reported that, in the cat CB, the addition of CO₂- to their in vitro perfusate media, at a constant pH of 7.40, increased basal chemosensory discharges and accelerated the chemosensory responses to hypoxia.

If DA is the excitatory transmitter between the glomus cell and the chemosensory nerve endings, not only hypercapnia but also the presence of CO₂- at pH 7.40 in the in vitro medium should enhance DA efflux in close correlation with the changes in chemosensory activity observed in both conditions. To test this hypothesis, we studied the simultaneous responses of CA and chemosensory discharges to hypercapnia (10% CO₂, pH 7.10) in the superfused cat CB in vitro, as well as the simultaneous responses to the addition of CO₂- buffer to the superfusate medium at pH 7.40, on basal conditions and hypoxia-induced (PO₂ 40 Torr) chemosensory excitation. Because we (16, 37) previously have found that the application of exogenous DA to cat CB, superfused with CO₂- free medium, enhanced and accelerated the hypoxia-induced CA release without changing the speed and intensity of the chemosensory responses, we also tested the effects of intrastream injections of DA on CA and chemosensory excitation induced by hypoxia during superfusion of the CB with CO₂-, at pH 7.40.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Experiments were performed in 12 male cats (3.1 ± 0.2 kg) anesthetized with pentobarbital sodium (40 mg/kg ip). The cats were tracheotomized, and one femoral vein was cannulated for injections of additional anesthetic when necessary. The CB and the carotid sinus nerve were dissected from surrounding tissue, trimmed, placed in the channel (2.0-ml volume) of a Plexiglas chamber, and superfused with Tyrode solution (in mM: 154 Na⁺, 4.7 K⁺, 2.2 Ca²⁺, 1.1 Mg²⁺, 120.0 Cl, 42 glutamate, 5.5 D-glucose, and 0.02 tyrosine). The Tyrode solution was buffered with N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES; 5 mM) or with HEPES (5 mM) plus (21 mM, replacing the same amount of sodium glutamate and equilibrated with 5% CO₂) to pH 7.40. The CBs were superfused at flows between 1.5 and 3.8 ml/min. The effluent was pumped with a vacuum system to maintain a fluid level in the channel adequate to cover the CB. Temperature of the superfusate solution was maintained at 37.5 ± 0.5°C by a proportional temperature controller. In some experiments, the PO₂ of the superfusate medium was measured by an oxygen needle-electrode (23 gauge) connected to a polarographic device (Chemical Microsystem, Diamond General).

The frequency of chemosensory discharges (f_x) was recorded from the whole desheathed carotid sinus nerve, which was placed on a pair of platinum-iridium electrodes and lifted into paraffin oil. The nerve signals were preamplified, amplified, filtered (10 Hz-1 kHz), selected with an electronic amplitude discriminator, electronically counted, and printed. The signals were also stored on a videocassette recorder analog-to-digital system for later analyses.

CA was measured through a high-speed computerized chronoamperometric system (IVEC-10, Medical System). Carbon multifiber microelectrodes (3 fibers, 30-µm tip diameter, Quanteon) were gently advanced into the CB. A 550-mV pulse was applied to the carbon microelectrode, with respect to a Ag/AgCl reference electrode, for 100 ms at a rate of 5 Hz. The resulting oxidation current was digitally integrated during the last 80 ms of each pulse, averaged for five cycles, and displayed at a rate of 1 Hz. The reduction current, generated when the potential returned to resting level (0 V), was assessed in the same manner. Carbon-fiber microelectrodes were coated 4-6 times with Nafion (Aldrich Chemical) to reduce their sensitivity to ascorbic and dihydroxyphenylacetic acid (12). Carbon microelectrodes were calibrated with DA hydrochloride (DA in 0.1 mM HClO₄ to prevent oxidation), at a final concentration of 2-12 µM in Tyrode solution buffered with HEPES at pH 7.40; only electrodes showing high sensitivity and linear currents (r² > 0.995) were used. We did not attempt to differentiate between DA and other CAs released from the CB. However, most of the electroactive species measured from the CBs should correspond with DA because the reduction-to-oxidation current ratios (red/ox) measured were similar to those obtained during the calibration with DA (red/ox range: 0.5-0.7). The red/ox have been used as indexes to identify the electroactive species under oxidation (see Refs. 12, 16). CA was expressed as efflux over baseline levels and was computed from the oxidation currents.

Experimental design. In a series of experiments, seven CBs were superfused with Tyrode solution containing HEPES- and equilibrated with 5% CO₂-20% O₂ at pH 7.40. The CBs were stimulated with hypercapnic (10% CO₂; pH 7.10) and hypoxic superfusions (PO₂ 40 Torr) for 5-8 min. In other series of experiments, five CBs were superfused first with Tyrode solution equilibrated with air in the absence of CO₂- at pH 7.40 for 60-90 min and then with the same Tyrode solution supplemented with and equilibrated with 5% CO₂-20% O₂ at pH 7.40. The CBs were exposed to hypoxic superfusion with and without CO₂- for 5-8 min. In both series, DA hydrochloride (10-100 µg, dissolved in saline containing 1 mM ascorbic acid to prevent oxidation) was injected into the superfusate channel 4 cm upstream from the CBs in boluses of 20-40 µl.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effects of hypercapnic and hypoxic superfusions on CA and chemosensory responses in the presence of CO₂-. Figure 1 shows the effect of hypercapnic and hypoxic superfusions on CA and f_x. Switching from Tyrode solution equilibrated with 5% CO₂-20% O₂ at pH 7.40 to Tyrode solution equilibrated with 10% CO₂-20% O₂ at pH 7.10 increased f_x without any noticeable change in CA. By contrast, hypoxic superfusion (PO₂ 40 Torr) increased f_x and induced a delayed increase in CA, equivalent to ~2.0 µM of DA. In this experiment, the maximal f_x achieved during hypoxia preceded the maximal value of CA by ~150 s. In the CBs studied in this series, the first hypoxic superfusion always increased CA, whereas hypercapnia failed to increase CA in five of seven CBs; in the remaining two, hypercapnia produced small increases in CA (CA < 0.5 µM). In this experimental group, the first hypoxic superfusion increased CA by 3.1 ± 0.9 µM, with a time to reach the CA peak of 458 ± 15.2 s, whereas f_x increased from a baseline of 76.6 ± 13.8 to a maximum of 173.9 ± 26.8 Hz (P < 0.01) in 142.1 ± 16.9 s. Thus the peak of the chemosensory response to hypoxia preceded the peak of the CA signal by 324.7 ± 18.7 s (P < 0.01).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effect of acid-hypercapnia and hypoxia on catecholamine efflux (CA) and chemosensory discharges (fx). Open horizontal bar, hypercapnic superfusion; solid horizontal bar, hypoxic superfusion.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effect of repeated hypercapnic and hypoxic stimuli on CA and maximal increase in fx. fx, Change in fx expressed as percentage of first hypoxic response. Open and solid bars are defined as in Fig. 1.

Effect of exogenous DA on chemosensory responses and CA induced by hypoxia and hypercapnia in the presence of CO₂ and . Figure 3 compares the effects of the administration of exogenous DA on CA induced by hypoxia and hypercapnia in two CBs. Figure 3A shows an example of the most frequently observed electrochemical response. Intrastream injections of exogenous DA (3 boluses of 10 µg each) enhanced the speed and amplitude of the subsequent CA induced by hypoxia but not the amplitude and speed of the small CA induced by hypercapnia (Fig. 3A). The enhancing effect of DA on the hypoxic-induced CA was observed in six of seven CBs studied. Conversely, the amplitude and speed of the small CA induced by hypercapnia, after DA administration, remained unchanged in five of seven CBs (Fig. 3A). In the remaining two CBs, hypercapnia increased CA after the adminis-tration of DA, as shown in Fig. 3B. In this example, immediately after the administration of the third bolus of DA (100 µg), superfusion with Tyrode solution equilibrated with 10% CO₂-20% O₂ at pH 7.10 induced an increase in CA. However, the subsequent hypercapnic stimulus performed 20 min afterward failed to evoke CA. Nonetheless, a subsequent hypoxic superfusion did increase CA.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effect of dopamine (DA) injections on CA induced by hypoxia and hypercapnia in 2 carotid bodies (CBs). A: DA (3 boluses of 10 µg at arrowheads) enhances speed and amplitude of next hypoxic-induced CA but not that induced by hypercapnia. B: DA (3 boluses of 100 µg at arrowheads) enhances CA induced by 1st hypercapnic superfusion, but a subsequent hypercapnic stimulus performed after 20 min failed to induce an appreciable CA. An hypoxic superfusion performed after hypercapnia still increased CA. Open and solid bars are defined as in Fig. 1.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of several injections of DA (10 and 100 µg at arrowheads) on CA (A) induced by hypoxic superfusions (1-3) and on chemosensory responses (B). PO2, superfusate PO2. Bars, hypoxic superfusions.

Effects of CO₂- at pH 7.40 and DA on the hypoxic-induced CA and chemosensory discharges. In another series of experiments, five CBs were initially superfused with CO₂--free Tyrode solution, equilibrated with 20% O₂ at pH 7.40, and then superfused with the same Tyrode solution supplemented with and equilibrated with 5% CO₂-20% O₂ at the same pH. Figure 5 shows the usual electrochemical responses induced by hypoxia during superfusion with Tyrode solution either without (Fig. 5A) or with CO₂- (Fig. 5B), before and after the administration of intrastream injections of DA (3 boluses of 10 µg). Without CO₂-, and before the administration of DA, hypoxia induced a delayed increase in CA, attaining a maximum response at the end of the 7-min period of hypoxic superfusion (H₁ in Fig. 5A). Immediately after the injection of the third bolus of DA, the same hypoxic superfusion produced a fast increase in CA (H₂ in Fig. 5A), whereas a later hypoxic superfusion period, performed after 20 min, evoked a reduced and delayed CA response (H₃ in Fig. 5A). The presence of CO₂- in the superfusate medium did not modify this pattern of the hypoxic-induced CAs, as shown in Fig. 5B. However, after the administration of DA, hypoxia (H₄ in Fig. 5B) produced a larger increase in CA than in the absence of CO₂-, whereas a hypoxic superfusion (H₅ in Fig. 5B) performed after 20 min evoked only a small and delayed CA. Figure 5C shows the CAs and chemosensory responses elicited by hypoxic stimulations H₁, H₂, and H₄. Despite large differences in amplitude and dissimilar time course of the CA induced by these hypoxic superfusions, the chemosensory responses were similar (Fig. 5C).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effects of DA (3 boluses of 10 µg at arrowheads) on CA induced by hypoxic superfusions (H1-H5; bars). A: effects during superfusion with Tyrode solution without CO2-. B: effects during superfusion of Tyrode solution with CO2-. C: effects of hypoxic superfusions H1, H2, and H4 on CA and fx.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Effects of CO2--buffered Tyrode solution on CA and fx. In normoxic condition, isohydric (pH 7.40) superfusion was switched from HEPES-buffered to CO2--buffered Tyrode solution (arrowhead). After 10 min, CB was superfused with hypoxic medium for 7 min. Bar, hypoxic superfusion.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effects of DA (3 boluses of 10-100 µg) on time to reach maximal values of CA (tCA) and fx (tfx) induced by hypoxia in 5 CBs in absence (A) and presence (B) of CO2-. Open bars, control response; hatched bars, after DA. * P < 0.05 with respect to control response.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Effects of DA (3 boluses of 10-100 µg) on amplitude of CA and fx induced by hypoxia during superfusion in 5 CBs without (A) and with (B) CO2-. Open bars, control hypoxic stimulations; hatched bars, hypoxic stimulation after DA; solid bars, hypoxic stimulation after 20 min without DA. * P < 0.05 with respect to control response.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The prevailing hypothesis of CB chemoreception proposes that the glomus (type I) cells are the primary site of transduction of O₂, CO₂ and H⁺ stimuli (8, 14), although this is not a unanimous view (26, 33). In response to hypoxia and hypercapnia, glomus cells are expected to release an excitatory transmitter, which, in turn, generates the chemosensory impulses in the afferent nerve endings. If DA is the excitatory transmitter released from the glomus cells of the CB in response to hypercapnia and hypoxia, a close relationship should exist between the amplitude of DA efflux and chemosensory excitation, even on repeated exposure to the same stimuli. Contrary to this expectation, present results show that hypoxia produced chemosensory excitation and increased CA, but hypercapnia, although evoking chemosensory excitation, did not consistently increase CA. Similarly, switching superfusion from HEPES-buffered medium to Tyrode solution containing CO₂- at pH 7.40 increased basal f_x but failed to release CAs. Moreover, either in the presence of CO₂ and or in their absence at the same pH 7.40, repeated hypoxic and hypercapnic stimulations progressively reduced the amplitude of CA from the CB, although producing similar increases in f_x. These results extend our previous observations that repeated injections of NaCN and hypoxic stimuli progressively reduced CA but still increased f_x in the cat CB superfused with Tyrode solution free of CO₂- (16, 37). A similar pattern of response was recently reported by Donnelly (6) in the rat CB, in which repetitive anoxic stimulations resulted in progressive reductions in CA, measured by amperometry, whereas the amplitude of the chemosensory responses was maintained. Furthermore, Donnelly (6) reported that the exposure of rat CB to moderate hypoxia (PO₂ 80 Torr) and repetitive anoxia produced CAs of comparable amplitude, although the maximum chemosensory response was significantly reduced in mild hypoxia. Taken together, these observations show a clear dissociation between the amplitude of chemosensory excitation and CA and do not support the hypothesis of DA being the excitatory hypercapnic or hypoxic transmitter in the CB.

Our results showed that the peak chemosensory response to hypoxia preceded the corresponding CA peak by 3 min or more. The time course of the delayed CA induced by mild hypoxia in the cat in vitro CB preparation in this study is similar to that elicited by anoxia and severe hypoxia in the rat CB superfused in vitro (5). On the contrary, Buerk et al. (4), using gold microelectrodes polarized at 150 mV, obtained faster CAs evoked by interruption of the perfused flow in the perfused-superfused preparation of the cat CB. They attributed the fast and slow rates of CA to the distance between the electrode tip and the glomus cell clusters. Microelectrodes that poorly penetrate the CB exhibited slower CA. Thus a delayed CA may be explained by a lengthy diffusion distance between the source of CA (i.e., glomus cell clusters) and the tip of the carbon-fiber electrode (see Refs. 6 and 16 for discussion). However, we cannot attribute the fast and slow rates of CA found in this study to the distance of the electrode tip to the glomus cells. After administration of DA to the CB, both the speed and the amplitude of the subsequent hypoxia-induced CA were markedly enhanced, but the amplitude and rate of rise of the chemosensory response were not modified. Therefore, the normal pattern of a delayed CA induced by hypoxia that followed the rise in f_x was reversed, and after DA application the CA peak precedes the peak of the chemosensory response. Thus slow, without preloading of the CB with exogenous DA, and faster CA responses, after preloading of the CB with exogenous DA, appeared to depend on the amount of DA stored in the CB cells and are independent of the distance between the tip of the microelectrode and the glomus cells, which remained constant during the experiment.

The faster and larger CAs induced by hypoxia, and, in a few cases, by hypercapnia, after applications of exogenous DA, suggest that part of the DA, which increased CA up to 30 µM, was incorporated by the glomus cells or other CB cellular structures. This recently incorporated DA pool is released by the next hypoxic stimulation and is rapidly exhausted by repeated hypoxic stimuli, whereas the chemosensory response remains without noticeable changes. Cytotoxic-hypoxic stimulation evoked by cyanide also produced a similar reduction in the exogenously incorporated DA pool (16). A newly incorporated DA pool that is rapidly taken up and released from the glomus cells is compatible with the very fast washout of newly incorporated [³H]DA, with a half-time of 4 min, found in rabbit CBs (13). Thus it is possible that exogenous DA may increase the store of DA in the glomus cells, and after the release even from the same source, the steeper diffusion gradient may account for the earlier increase in CA. It is also plausible that exogenous DA may load the endings of sympathetic neurons or petrosal chemosensory afferent neurons. In fact, Finley et al. (10) found that 41% of the rat CB afferents expresses tyrosine hydroxylase, the rate-limiting enzyme of catecholamine synthesis, whereas 86% of these fibers also contains 3,4-dihydroxyphenylalanine decarboxylase, the DA-synthesizing enzyme. Moreover, Almaraz and Fidone (2) found evidence that selective stimulation of C-fibers in the carotid sinus nerve of cats produced the release of [³H]CA from the CB. However, neither chronic carotid sinus nerve section nor chronic sympathectomy modified the uptake of [³H]DA in the rabbit CB, suggesting that the uptake of DA by sympathetic or chemosensory afferent neurons should be small compared with that by glomus cells (13). Moreover, the synthesis and release of [³H]DA from the cat CB induced by hypoxia (30) and severe hypercapnia (31) were unaffected by chronic carotid sinus nerve section.

Contrary to what was found during hypoxia, our results show that hypercapnia did not consistently increase CA but produced chemosensory excitation. These results agree with the preliminary statement of Buerk et al. (4) that hypercapnia produced chemosensory excitation but did not cause significant DA release from the cat CB perfused in vitro. Rigual el al. (31) reported that extremely hypercapnic and acid stimuli (pH 6.6) produced more consistent releases of [³H]DA in the cat CB. They found that the most effective stimulus was 20% CO₂ at pH 6.8 (31). However, this stimulus produced an increase in [³H]DA of only ~2-3 times the baseline, smaller than the 10-20 times over the baseline increase in [³H]DA induced by hypoxia reported by the same researchers (see Ref. 14 for review). The major difference between their cat CB preparation and ours is the duration of the effects of stimuli. They reported that 10 min of acid superfusion produced prolonged f_x and [³H]DA increases, lasting 30 and 45 min, respectively (31). On the contrary, in our preparation f_x and CA remained elevated for <5 and 10 min, respectively, after cessation of the 5- to 8-min period of acid hypercapnic or hypoxic superfusion.

Present results confirm that hypoxia stimulates f_x in the absence of exogenous CO₂- (7, 17, 18). Hypoxia reduced the PO₂-dependent K⁺ current in isolated glomus cells bathed in a CO₂--free medium (23), increased intracellular [Ca²⁺] (35), and evoked the release of CA from isolated glomus cells (35), even though the pH of glomus cells is expected to be alkaline (up to pH 7.8) in the absence of CO₂- (3). These results agree with our observation that O₂ chemoreception does not require the participation of exogenous CO₂-. However, the presence of CO₂- in the superfusate medium, at pH 7.40, raised baseline f_x and accelerated the chemosensory responses to hypoxia. However, the effects of CO₂- on baseline and hypoxic-induced rise of f_x found in the superfused preparation were less marked than in the perfused preparations of the cat CB (17, 32). The different effects of CO₂- on the speed of chemosensory responses to hypoxia could be attributed to the characteristics of both in vitro preparations. In the superfused CB, all the substrates, including CO₂ and H⁺ produced within the CB, must diffuse along the CB tissue, resulting in a gradient from the core to the periphery of the organ. Thus endogenous production of CO₂ and H⁺ in the chemoreceptor cells of the superfused CB could be enough to maintain an acid intracellular pH to keep fast and suitable O₂ chemoreception for several hours (7, 16). On the other hand, perfusion of the CB with saline solutions that allow the exchange of substrates between the vessels and the glomus cells seems to be ideal, but the flow through the CB should be higher than the normal flow due to the reduced viscosity of the saline solutions compared with blood. Consequently, endogenous CO₂ production in the perfused CB is expected to be rapidly washed out by the high saline flow and may be not enough to maintain an acid intracellular milieu. Thus the presence of CO₂- at pH 7.40 in the perfusate medium is expected to produce a large effect on the intracellular pH of glomus cells due to CO₂ hydration mediated by carbonic anhydrase (18), increasing not only basal f_x but also hypoxic chemoreception (17, 32). Indeed, the presence of CO₂- at pH 7.40 in the superfusate medium reduced the intracellular pH of isolated glomus cells from 7.8 to 7.2 (3). The lack of effect of CO₂- on the amplitude of hypoxic chemosensory responses found here agrees with previous observations in the cat CB perfused in vitro (16) and resembles the recent finding that the presence of CO₂- in the in vitro medium did not increase the amplitude of anoxic chemosensory responses in a superfused rat CB preparation (28).

Present results showed that the presence of CO₂- did not enhance the amplitude of hypoxic-induced CA. In fact, the initial hypoxic test performed in the presence of CO₂- at pH 7.40 increased CA by 3.10 ± 0.9 µM (n = 7). This value is not significantly different from the increase of 3.16 ± 0.8 µM (P > 0.05, n = 8) produced by the first hypoxic stimulation in previous experimental series done with CO₂- - free medium (16). However, present results show that the presence of CO₂- in the superfusate medium increased the amplitude of the hypoxia-induced CA after DA administration compared with the CA evoked in the same conditions but in the absence of CO₂-. This result suggests that the incorporation of exogenous DA by the CB cells is enhanced by the presence of CO₂-. Panisello and Donnelly (28) found that the presence of CO₂- enhanced the hypoxic-induced CA in the rat CB but that acid superfusion (pH 6.5) did not mimic the enhancing effect of CO₂- on CA, which was reduced by the anion channel blocker 9-anthracene carboxylic acid. These results suggest that the effect of CO₂- on the CA release induced by hypoxia seems to be related to transport rather than to intracellular acidification. However, it is still possible that the enhancing effect of CO₂- on CA secretion induced by hypoxia may be explained by a large amount of DA reuptake and storage in dense cored vesicles due to the intracellular acidification induced by CO₂-. It is well known that accumulation of CA in chromaffin granules is related to the magnitude of the H⁺ concentration gradient across the membrane granule (20). In addition, it is conceivable that the acidosis induced by high CO₂ may directly excite the nerve endings of the chemosensory afferent neurons. However, Alcayaga and Arroyo (1) recently reported that local acidification of cat petrosal neurons in primary culture had no effect on the electrically evoked action potentials in ~20% of the recorded neurons. In the remaining 80% of the recorded neurons, they found that acidification blocked the evoked action potentials. When petrosal neurons were cocultured with glomus cells, 10% of the recorded neurons responded to acidification with a train of action potentials, suggesting that CB tissue is necessary to develop that response in the sensory neurons that innervated the glomus cells.

Our results show that the initial injection of DA (10-100 µg) produced in some preparations transient inhibition of f_x, after which the CB became nonreactive to DA. This confirms previous observations of Zapata (36) that intrastream injections of DA applied to cat CBs superfused in vitro produced transient inhibition of f_x but that repeated intrastream injections of DA resulted in desensitization of the inhibitory actions, and even produced late excitatory effects in response to large doses. In experiments in situ, intracarotid and intravenous injections of exogenous DA produced chemosensory inhibition, excitation, or dual effects in CBs from different species (see Refs. 8 and 14 for review). In the cat, intracarotid and intravenous injections of DA always produced transient inhibition of f_x (18, 24), whereas continuous intravenous infusion of DA decreased the chemosensory responsiveness to hypoxia and hypercapnia (22), suggesting that DA release is not responsible for O₂ and CO₂ chemosensory responses in the cat CB (19, 22, 24). This inhibitory effect is reversed into excitation by large DA doses after blockade of D₂ DA receptors (19, 24). Domperidone, an antagonist of D₂ receptors, produces a maintained increase in f_x and enhances chemosensory responses to hyperoxia (100% O₂), hypoxia (100% N₂), and NaCN (19). Similarly, the dopaminergic D₂ antagonist haloperidol potentiates chemosensory responses to hypoxia and hypercapnia in cats (22). Thus the effects of DA and D₂ antagonist on cat CB chemoreception suggest a modulatory role for DA within the cat CB but do not support the hypothesis of DA acting as an excitatory transmitter.

In summary, our results show that, under present experimental conditions, the presence of CO₂- in the superfused medium has no effect on hypoxia-induced CA from the CB, although it may enhance the uptake of exogenously applied DA. The time course and amplitude of chemosensory responses induced by hypercapnia and mild hypoxia in the presence of CO₂- are not related to the changes in CA, suggesting that CA release is not essential for hypercapnia- or hypoxia- induced chemosensory excitation in the cat CB.