Acetylcholine release from cat carotid bodies

Brain Research 841 Ž1999. 53–61 www.elsevier.comrlocaterbres

Research report

Acetylcholine release from cat carotid bodies Robert S. Fitzgerald a

a,b,c,)

, Machiko Shirahata

a,d

, Hay-Yan Ž Jack. Wang

a

DiÕision of Physiology, Department of EnÕironmental Health Sciences, The Johns Hopkins Medical Institutions, Baltimore, MD 21205, USA b Department of Physiology, The Johns Hopkins Medical Institutions, Baltimore, MD 21205, USA c Department of Medicine, The Johns Hopkins Medical Institutions, Baltimore, MD 21205, USA d Department of Anesthesiologyr Critical Care Medicine, The Johns Hopkins Medical Institutions, Baltimore, MD 21205, USA Accepted 22 June 1999

Abstract Hypoxia, hypercapnia and acidosis stimulate the carotid body ŽCB. sending increased neural activity via a branch of the glossopharyngeal nerve to nucleus tractus solitarius; this precipitates an impressive array of cardiopulmonary, endocrine and renal reflex responses. However, the cellular mechanisms by which these stimuli generate the increased CB neural output are only poorly understood. Central to the understanding of these mechanisms is the determination of which agents are released within the CB in response to hypoxia, and serve as the stimulating transmitterŽs. for chemosensory nerve endings. Acetylcholine ŽACh. has been proposed as such an agent from the outset, but this proposal has been, and remains, controversial. The present study tests two hypotheses: Ž1. The CB releases ACh under normoxicrnormocapnic conditions; and Ž2. The amount released increases during hypoxia and other conditions known to increase neural output from the CB. These hypotheses were tested in 12 experiments in which both CBs were removed from the anesthetized cat and incubated at 378C in a physiological salt solution while the solution was bubbled with four different concentrations of oxygen and carbon dioxide. The incubation medium was exchanged at 10 min intervals for 30 min Žthree periods of incubation.. The medium was analyzed with high performance liquid chromatography-electrochemical detection for ACh content. Normoxicrnormocapnic conditions Ž21% O 2r6% CO 2 . produced a total of 0.639 " 0.106 pmolr150 ml Žmean " S.E.M.; n s 12.. All stimulating conditions produced larger total outputs: 4% O 2r2% CO 2 produced 1.773 " 0.46 pmolr150 ml; 0% O 2r5% CO 2 , 0.868 " 0.13 pmolr150 ml; 4% O 2r10% CO 2 , 1.077 " 0.21 pmolr150 ml. These three amounts were significantly greater than the normoxicrnormocapnic condition, but indistinguishable among themselves. Further, the amount of ACh released did not diminish over the 30 min of stimulation.These data support the concept that during hypoxia ACh functions as a stimulating transmitter in the CB, and are consistent with the earlier reports of cholinergic enzymes and receptors found in the CB. q 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Acetylcholine; Carotid body; Cholinergic receptor; Cholinergic stimulation; Hypoxia; HPLC

1. Introduction Systemic cardiopulmonary, endocrine and renal reflex responses to carotid body ŽCB. stimulation by hypoxia, hypercapnia and acidosis in the arterial blood are well described w1,6,21,23,30,55,59x. However, the mechanisms operating in the CB whereby these stimuli generate the increased neural activity travelling from the CB to the nucleus tractus solitarius are not yet well understood. A currently accepted model for the stimulation due to hypoxia has the transmitter-containing glomus cells of the CB depolarized by as yet unclear mechanisms. This depolarization activates voltage gated calcium channels permit) Corresponding author. EHSrSHPHrJHU, 615 N. Wolfe Street, Baltimore, MD 21205, USA

ting the entrance of extracellular calcium. The increase in cytosolic calcium by this and probably other means is presumed to promote the release of transmitters from the CB’s glomus cells. The transmitters are thought to excite, modulate and inhibit: Ž1. the activity in those chemosensory nerve endings which abut on the glomus cells and travel through the petrosal ganglion Žwhere they have their cell bodies. to the nucleus tractus solitarius; and Ž2. the further release of transmitters from the glomus cells. They do this by binding to ‘‘postsynaptic’’ receptors on the nerve endings and autoreceptors on the glomus cells. Acetylcholine ŽACh. is a transmitter found in glomus cells. Some early studies suggested an excitatory role for it w14,42,44x, but other early studies did not support such a role w12,13,31,37,48x. In fact, C. Heymans Žwinner of the 1938 Nobel Prize in Physiology or Medicine for his dis-

0006-8993r99r$ - see front matter q 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 7 7 7 - 1

54

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

covery of the cardiopulmonary reflex responses to stimulation of carotid and aortic arch chemo- and pressoreceptors. along with Eric Neil stated: ‘‘it is the authors’ opinion that ACh has nothing whatever to do with the normal transmission of the chemoreceptor impulses’’ w31x. Later, however, Eyzaguirre et al. again presented strong evidence supporting an excitatory role for ACh w15–17,51x. Again, not all agreed w9,46,56x. The present study is the continuation of an effort to explore anew the possibility that ACh is a principal excitatory transmitter in the cat CB. Previously, we have reported studies in which we used pharmacological agents to stimulate or inhibit the production andror release of ACh, and to block presumed nicotinic and muscarinic receptors w24,25x. Using immunocytochemical techniques we have reported the presence of a4 subunit-containing and a7 subunit-containing neuronal nicotinic receptors in the cat CB w35,61x. Others w7,32x, using 3 H-QNB, have reported the presence of muscarinic receptors in the CB. Finally, we have reported that ACh or nicotine can stimulate an increase in glomus cell wCa2q x i w60x. This effect is reduced in the presence of mecamylamine. The present study tests the hypotheses: Ž1. The CBs release ACh under normoxicrnormocapnic conditions; and Ž2. The amount released increases under conditions of stimulation; i.e., when neural activity is known to increase Že.g., during hypoxia..

2. Methods Cats of both sexes Ž3–3.5 kg. were lightly anesthetized with ketamine Ž50–100 mgrkg, i.p... A femoral venous catheter was introduced, and sodium pentobarbital Ž5–10 mgrkg. was administered. Next, two 2 ml injections Ž100 mmolrkgrinjection. of the non-specific cholinesterase inhibitor tetraisopropylpyrophosphoramide ŽisoOMPA, Sigma T-1505. w39x were administered; 15 min intervened between the injections. After the administration of heparin Ž2000 IUrkg, i.v.. the cat was deeply anesthetized with sodium pentobarbital Ž50–100 mgrkg.. The head was removed rapidly after which each common carotid artery received an infusion of a modified Krebs Ringer bicarbonate ŽKRB. solution containing 1 mM isoOMPAq 300 mM neostigmine. The millimolar concentration of the KRB solution for the following chemicals was: NaCl, 112; KCl, 4.7; CaCl 2 , 2.2; MgCl 2 , 1.1; Glucose, 11; and NaHCO 3 , 22. Bathed in the same cholinesterase inhibitor-containing solution the two carotid bodies were cleaned of fat and connective tissue. Approximately 5 min elapsed between the start and end of CB removal. For each of the 12 experiments, the pair of CBs was then placed in an Eppendorf tube containing 700 ml of KRB containing 1 mM isoOMPAq 300 mM neostigmine at 378C and bubbled first with 95% O 2r5% CO 2 Ž20 min. and then with 21% O 2r6% CO 2 Ž40 min.. This pre-ex-

posure incubation period preceded each of the four timed exposures below. After each of the pre-exposure incubation periods the pair of CBs was washed thoroughly five times with KRB now containing only 15 mM neostigmine. This solution had previously been bubbled with one of the four challenging gas mixtures: 21% O 2r6% CO 2 Ž21r6; normoxicr normocapnic exposure. or with one of the following stimulating gas mixtures: 95% N2r5% CO 2 Ž0r5; hypoxicr normocapnic exposure., 4% O 2r10% CO 2 Ž4r10; hypoxicrhypercapnic exposure., or 4% O 2r2% CO 2 Ž4r2; hypoxicrhypocapnic exposure.. The order of exposures was varied and arbitrary. Each exposure consisted of three periods lasting 10 min each for a total of 30 min during which the carotid bodies were bathed in 70–90 ml of the solution bubbled with one of the four gas mixtures. Temperature in the bath was continuously maintained at 378C–388C and monitored with a temperature probe ŽYellowsprings, YSI 400.. The pO 2 ŽMI 730., pCO 2 ŽMI 720. and pH ŽMI 710. of the medium were measured with microelectrodes ŽMicroelectrodes, Bedford, NH.. After 10 min the bathing medium was drawn off into a microfuge tube containing a 0.22 mm filter, immediately spun down for 2 min, and either stored on ice until measurement of ACh on the same day, or stored at y808C for subsequent determinations of ACh. The bathing medium drawn off was immediately replaced with 70–90 ml of solution for the second or third period. The amount of ACh per 10-min period from the two CBs was measured with a 480 Bioanalytical Systems ŽBAS. High Performance Liquid Chromatograph ŽHPLC.. Both standard bore and microbore techniques were used. With the former, the PM80 pump Žrun at 2800–3800 psi. was followed by: Ž1. a section of PEEK back-pressure tubing ŽUpchurch Scientific., Ž2. a pre-column immobilized enzyme reactor ŽBAS IMER; MF 6153. for removing choline from the sample, Ž3. the analytical column ŽBAS MF 6150., Ž4. a post-column IMER ŽBAS MF 6151. for breaking down and converting ACh to H 2 O 2 , and Ž5. the electrochemical detector ŽMF 2061.. The amount of ACh in a sample is directly proportional to the amount of H 2 O 2 electrochemically detected by the BAS LC-44 Detector which uses a thin-layer cell of a platinum working electrode ŽMF 1012. and an auxiliary electrode. The reference electrode was standard AgrAgCl ŽBAS MF 2021.. With the standard bore determinations the IMERS and analytical column were housed in a constant temperature Ž378C. chamber ŽBAS LC-23C. located in the CC-5 unit; the pump was run at 0.7 mlrmin with a 50 ml injection loop ŽMF 4122. as part of the Rheodyne Injector component ŽMF 5026.. The more sensitive microbore system was run at room temperature housed in the same cabinet. The components after the pump were: Ž1. a section of PEEK back-pressure tubing, Ž2. pre-column IMER ŽBAS MF 8807., Ž3. analytical column ŽBAS MF 8904., Ž4. post-column IMER ŽBAS

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

MF 8903., and Ž5. electrochemical detector which consisted of a ‘‘wired’’ peroxidase glassy carbon electrode ŽMF 1024., coated with Peroxidase Redox Polymer ŽCF 1070.; the reference electrode remained a standard AgrAgCl. With the microbore setup, injections were made into a 20 ml loop ŽBAS MF 4129. at a flow of 0.130 mlrmin. The variables used in the detection of ACh by both electrode assemblies Že.g., filter w0.1 Hzx, range w0.5–2 nArfull scalex, offsets, etc.. were adjusted with controls on the LC-44 Amperometric Detector. Data were acquired from the Amperometric Detector equipped with a DA 1 Start-Up Kit including a data acquisition board with DrA and ArD converters. The board was interfaced with a Packard Bell Force 205 PC Ž486SX; 33 MHz; 12 Mb RAM.. Data acquisition was under the direction of a BAS software package, Inject ŽVersion 1.28 a0112701.. The data so acquired were subsequently quantified with a second BAS software package ŽChromgraph.. Output from the Amperometric Detector was also simultaneously led to a BAS RYT strip chart recorder for a real-time hard copy of the experiment. The mobile phase was a 30 mM solution of Na 2 HPO4 ŽpH s 8.5 " 0.05. which contained the BAS bacteriacide,

55

Kathon ŽBAS CF-2150.. Mobile phase was discarded a week after its date of composition. ACh standards were made up from neostigmine-containing KRB and run on the same day the samples were run. Contents of ACh measured with the microbore system using the 20 ml loop were multiplied by 2.5 so that all contents were reported as being contained in 50 ml volumes. Statistical significance of the data was determined using SigmaStat ŽJandel Scientific. programs. If a comparison of the treatment groups passed the normal distribution and equal variances prerequisites, Student’s paired t-test or one-way ANOVA procedures were followed. If the groups being compared did not pass these prerequisites for parametric analysis, then non-parametric procedures ŽFriedman Repeated Measures Analysis of Variance on Ranks and Student–Newman–Keul methods. were followed.

3. Results Fig. 1 presents a representative chromatograph generated with the standard bore technique. ŽA. The 50 ml

Fig. 1. Chromatogram of ACh and choline using the standard bore technique Žcf. Section 2.. ŽA. Injection of 50 ml of a standard KRB solution containing 1 pmol of ACh and 10 pmol of choline. Peak retention time for ACh is 341 s; for choline, 436 s. The areas under the curve are in arbitrary units and indicate the amount of ACh and choline. ŽB. Injection of 50 ml of CB incubation medium from a 10 min exposure to 0% O 2r5% CO 2 Ž20–30 min interval.. Peak retention time was 340 s for ACh; for choline, 436 s. These areas corresponded to 0.520 and 55.1 pmol, respectively. The concentration of ACh was therefore 10.4 nM.

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

56

Fig. 2. Chromatogram of ACh using the microbore technique. ŽA. Injection of 20 ml of a KRB standard solution containing 0.1 pmol of ACh. Peak retention time Ž905 s. is much later than with the standard bore technique due mostly to the much slower flow. ŽB. Injection of 20 ml of the CB incubation medium Žbubbled with 4% O 2 r10% CO 2 for 10 min; the 10–20 min interval.. Peak retention time is 900 s. Sample contained 0.144 pmol of ACh; the concentration of ACh in the incubation medium was 7.2 nM.

injection of the standard solution, containing 1 pmol of ACh, has a peak retention time of 341 s; the injection also contained 10 pmol of choline which had a peak retention time of 436 s. The ACh generated an area of 111,830 arbitrary units; choline, 334,995. ŽB. The 50 ml sample injection of CB incubation medium Žthe 20–30 min interval of the hypoxicrnormocapnic exposure. exhibited a peak retention time of 340 s, and generated an area of 58,111 arbitrary units, corresponding to 0.520 pmol of ACh. The concentration of ACh in the incubation medium was therefore 10.4 nM. Choline had a peak retention time of 436 s, an area of 1,845,532 arbitrary units; this corresponded to 55.1 pmol. Fig. 2 shows a representative chromatograph generated with the microbore technique. Note the much later peak

retention times. The flow is only about 20% of the flow in the standard bore system. ŽA. Peak retention time for the standard of 0.100 pmol of AChr20 ml was 905 s. For the sample of incubation medium Ž20–30 min interval of the hypoxicrnormocapnic exposure. the retention time was 900 s. The area determined from the standard curve corresponded to 0.144 pmol of AChr20 ml. The solution was 7.2 nM. Choline peaks were absent because of the efficiency of the pre-column IMER in removing the choline before the injectate arrived at the analytical column and post-column IMER. Table 1 gives the of pO 2 , pCO 2 and pH values Žmean " S.E.M.. for the incubation media in six of the experiments. These values were not determined in every experiment since the results in the six determinations were

Table 1 Gas

Mean S.E.M.

21% O 2r6% CO 2

4% O 2r2% CO 2

0% O 2r5% CO 2

4% O 2 r10% CO 2

pH

pCO 2

pO 2

pH

pCO 2

pO 2

pH

pCO 2

pO 2

pH

pCO 2

pO 2

7.403 0.012

36.4 1.6

134 2

7.858 0.016

12.4 0.0

29 2

7.471 0.009

33.6 0.3

3 1

7.147 0.021

71.2 0.6

27 1

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

virtually the same, and the same gases were used under the same conditions in all experiments. Fig. 3 summarizes the 12 experiments by showing the total amount Žmean " S.E.M.. of ACh collected from the pair of incubated carotid bodies over the three 10 min collections in response to the four different challenges. The figure addresses both hypotheses being tested: Ž1. CBs release a detectable amount of ACh under normoxicr normocapnic conditions . . . not only after 30 min, but ACh is detectable even after 10 min; and Ž2. A greater amount of ACh is released during a hypoxic challenge than under normoxic conditions. Each stimulating challenge Žhypoxic hypocapnia, hypoxic normocapnia, hypoxic hypercapnia. produced a total amount of ACh which was significantly greater than the amount released during the normoxicr normocapnic exposure. Standard ANOVA procedures on the total amounts released during the 30 min for the four treatments revealed that the data were not normally distributed. Therefore, the Friedman Repeated Measures Analysis of Variance on Ranks was performed. There was a statistically significant difference among the median values of the four treatment groups Ž P s 0.0023.. Stu-

57

dent–Newman–Keuls method for locating the differences showed that the control group median differed significantly from the other three, but the three hypoxic treatment groups had median values that were statistically indistinguishable among themselves. The same was true for the total amounts at the end of 20 min. The medians differed from each other with statistical significance Ž P s 0.0085.. The Student–Newman–Keuls method again identified the normoxicrnormocapnic exposure as being different from the other three treatment group medians, and they are not different among themselves. The figure also shows that for each of the four challenges, the amount of ACh releasedr50 ml remained virtually constant over the three 10 min intervals Žthe three boxes comprising each of the bars in the bar graph.. The only exception was during the hypoxicrhypocapnic challenge when the amount released in the 20–30 min interval was significantly greater than the amount released in the 10–20 min interval ŽFriedman Repeated Measures Analysis of Variance of Ranks: P s 0.0498..

4. Discussion To the best of our knowledge, this is the first report of a chemical technique used to measure ACh release from the CBs of any species. In summary the data from this in vitro study show that: Ž1. cat CBs release ACh continually even under normoxicrnormocapnic conditions; Ž2. hypoxia in any form will promote an increase in that release; and Ž3. release under hypoxic stimulation does not decay in the course of 30 min. 4.1. Critique of the method

Fig. 3. Bar graph indicating the amount of ACh Žmean"S.E.M.. detected in 150 ml from the three 10 min exposures to the four gas mixtures. Boxes comprising each bar represent the 10 min period’s output. Lowest box represents the 0–10 min interval; the middle box, the 10–20 min interval; the top box, the 20–30 min interval. Numbers under the bars represent the percent of O 2 rCO 2 in the mixture. Standard error bars are for the total amount: ns12 experiments. The amounts of ACh detected in the media of the three stimulating gas mixtures Ž4r2, 0r5, 4r10. are significantly greater than the amount detected during exposure to the normoxicrnormocapnic mixture Ž21r6., but they are not significantly different among themselves. The figure supports the two hypotheses being tested: Ž1. The CBs release ACh in detectable amounts under normoxicrnormocapnic Ž21r6. conditions; and Ž2. Challenged with hypoxia, the CBs release a greater amount than during normoxicrnormocapnic conditions. The figure also shows that the amount of ACh measured in three sequential 10 min intervals did not diminish during any of the exposures. U P - 0.05.

In pilot studies many in vitro CB preparations were tried; many different time periods were used; many gas mixtures were used; many protocols were followed. In by far the majority of instances, ACh could be found in the medium regardless of whether we incubated or superfused the CBs. Again in by far the majority of cases, the amount was greater when the exposure was to a stimulating composition; i.e., one known to increase the neural output of CB preparations in vitro or in vivo. Different concentrations of cholinesterase inhibitors were used. IsoOMPA was used to block the effect of any non-specific cholinesterase w3,38x. The CBs have pseudocholinesterases which can breakdown ACh if given the opportunity w38x. Ordinarily, the faster acting acetylcholinesterase, detected in the CBs of several species w38x, completes the task before the other esterases have a chance. Neostigmine was used specifically for acetylcholinesterase. The methods reported herein produced the most consistent results. There was, nonetheless, considerable variabil-

58

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

ity in the amount of ACh released. In these experiments, there were several procedures which could determine the amount of ACh the HPLC technique would report. There was inter-animal variability as in any experiment. Secondly, though the preparing of the CBs for incubation was quite consistent, it could be that some compromising factor was introduced. Perhaps some CBs became hypoxic. For this reason, we first treated the CBs with a pre-challenge incubation period, part of which used 95% O 2r5% CO 2 . However, because of the possibility that oxygen radicals might influence ACh output, we limited this to 15–20 min. This period was followed by 40 min of 21% O 2r6% CO 2 . Thirdly, there was no independent index of whether or not the cholinesterase blockers were effective. It appeared that two or three of the CB pairs exhausted their supply of ACh before completing the four-challenge protocol. These experiments were discarded. But to determine if this was a more general phenomenon we applied, in many of the experiments comprising this report, an additional period of stimulation after the completion of the four challenges. In all cases, a large amount of ACh was released. Hence, in general, the protocol followed did not seem to exhaust the supply of ACh even though we did not include choline in the recovery incubation periods between exposures. 4.2. Interpretation of the results Many investigators have observed that the flow of hypoxic arterial blood or of some hypoxic physiological salt solution through Žand over. the CB produces an increase in neural output. Hence, the data reported herein support the hypothesis that ACh is an excitatory transmitter involved in the cat CBs’ response to hypoxemia. In the past, basic criteria had been proposed which, if fulfilled, establish an agent as a synaptic neurotransmitter wRef. w15x; p. 581x. CB-located ACh had fulfilled all of these criteria except the ‘‘collectability of the transmitter’’. The present report completes for the cat CB this single as yet unfulfilled basic criterion. Since ACh excites in all species except the rabbit, it would seem appropriate, therefore, to designate ACh an excitatory transmitter. In the rabbit, ACh can inhibit w49x or stimulate w9x neural activity. Earlier studies of ACh within the CBs have determined the content of ACh within the CB. The first of these was the study of Eyzaguirre et al. w16x. Using bioassay and chromatographic techniques, they reported that ACh or an ACh-like substance was present in CB tissues of the cat, 20–30 mgrg of tissue. If the cat CB is assumed to weigh about 1 mg, then the CB in these studies contained about 137 pmol of ACh. Jones w38x reported a similar amount from 12 CBs taken from eight cats, 23.1 mgrg of tissue. Gual and Marsal w29x, using a chemiluminescent technique, reported 75 pmolrCB when the CBs were exposed to 100% oxygen at 348C. These three studies are in substantial agreement regarding the content per CB. Fidone et al. w18x, using pyrolysis gas chromatographyrmass fragmen-

tometry reported the content in 15 CBs to be about 11.7 pmolrCB. The average release of ACh from a CB in our in vitro preparation exposed to the normoxicrnormocapnic gas mixture over 30 min was 0.319 pmol, about 0.4% of the content reported by Gual and Marsal w29x, and about 2.7% of the amount reported by Fidone et al. w18x. The maximum release per CB over 30 min in our study Ž0.887 pmolrCB. was about 1.2% and 7.6% of these authors’ values for content, respectively. Metz w47x reported that when the CBs of dogs were perfused in vivo with hypoxic blood for 30 min, the CB venous outflow contained 1966 pgrml of ACh; when perfused with hypoxicrhypercapnic blood, the venous outflow contained 2807 pgrml. A 30 min perfusion produced 0.9 ml of venous outflow, amounting to 9.7 pmolr30 min and 13.9 pmolr30 min, respectively. This is more than an order of magnitude larger than our measurements of release from in vitro cat CBs, and would approach the total content measured by Fidone et al. w18x in cats. But since to our knowledge, no one has measured the ACh content of dog’s CBs, the amount released may constitute a percentage of the content not substantially different from those for the cat CB given above. The much larger release from the dog’s CB could be attributable to a difference in technique with in vivo perfusion able to liberate more measurable quantities of ACh than in vitro incubation over the same time period. Jones w38x, however, was unsuccessful in reproducing Metz’s results with dogs in 13 in vivo cat preparations regardless of whether he used either blood or a physiological salt solution as the perfusate. Our data may generate questions among those who find ACh acceptable as a candidate for an excitatory transmitter in CB chemotransduction, but who have measured neural output from the CBs during any of the forms of hypoxia we used. The neural output from the CB of a cat ventilated with a hypoxicrnormocapnic or hypoxicrhypercapnic mixture can increase two- to four-fold w19,22,40,50x. But the hypoxicrnormocapnic exposure in this study prompted a release of ACh from the CBs of only 136% of the normoxicrnormocapnic exposure and the hypoxicrhypercapnic mixture, 168% of the normoxicrnormocapnic mixture. Clearly, the amount of ACh released does not increase in a quantitatively parallel manner with the increase in neural activity reported in the other studies w19,22,40,50x. Further, the in vivo neural output from the CB of an artificially ventilated cat made hypoxic increases. If the cat is then hyperventilated, reducing the arterial pCO 2 while maintaining the hypoxic arterial pO 2 , the neural activity will fall Žunpublished observations; S. Lahiri, personal communication.. Hypercapnia increases the neural response to hypoxemia; hypocapnia reduces the neural response to hypoxemia w64x. In our present study, the amount of ACh released during exposure to hypoxic hypocapnia, though not significantly greater, shows a tendency to be greater than the other forms of hypoxia. That is, in eight of the 12 experiments the amount of ACh released during

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61

hypoxic hypocapnia was greater than during hypoxic hypercapnia, and in nine of the 12 this amount was greater than the amount released during hypoxic normocapnia. There is the possibility that oxygen andror carbon dioxide tensions Žperhaps by changing pH. may act to sensitize the receptors in addition to releasing ACh. Perhaps during the exposure to 4% O 2r2% CO 2 the alkalosis resulting from a pCO 2 of about 12 mmHg ŽpH s 7.858. might act to inhibit cholinergic receptors on the apposed sensory nerve endings and on the glomus cell Žautoreceptors. responsible for a negative feedback control of ACh release. If the inhibiting autoreceptors were inhibited, more ACh would be released. The inhibition of the postsynaptic receptor responsible for excitation of the apposed sensory ending would reduce the neural activity in that neuron. Several types of muscarinic receptors are modulated by pH w2,34x. In the smooth muscle of guinea pig ileum acidosis increases the ACh-activated inward current Žnonselective cationic current. with no significant change in reversal potential or on other ionic permeabilities. Unfortunately, these investigators did not test to see if alkalosis decreased the inward current. Nicotinic ACh receptors are also significantly modulated by pH w43,52x. Peak agonistactivated membrane currents in Xenopus oocytes induced to express Torpedo receptors through mRNA injections, were pH dependent — inhibited at acidic pH values and maximal at alkaline values. In contrast to this, however, channel kinetics were different. Mean channel open time was maximal at neutral pH and decreased at both acidic and alkaline pH. Hence, both channel permeability properties and channel gating properties are dependent on extracellular pH. Other studies showed that ACh-induced peak currents in mouse muscle nicotinic ACh receptors were reversibly decreased at both acidic and alkaline pHs. There is strong indication that the pH dependence of nicotinic ACh receptors is a function of their subunit composition. To our knowledge, the impact of pH on a4 subunits, detected on glomus cells of the cat CB w35x, or on the a7 subunits, found on the sensory nerve endings abutting on cat glomus cells w61x, has yet to be reported. Other transmitters could modulate the cholinergic receptors. Substance P has been proposed as an excitatory transmitter in the CB during hypoxia; significant amounts of data support this proposal w53,54x. Perhaps its mode of action involves preventing the desensitization of cholinergic receptors as it does in other cholinergic systems w4x. Finally, the catecholamines, specifically dopamine, have been proposed as the stimulating transmitter w26,27x. And it is clear that dopamine is indeed released during hypoxia w20,28,65,66x. But there is bountiful data demonstrating that dopamine, exogenously applied in a clinically relevant dose, hyperpolarizes chemoreceptor sensory nerve endings w58x and significantly reduces the neural response to hypoxia w8,28,33,45,57,63,65x. Further, when the dopamine D2 receptor is blocked, the CB neural response to hypoxia increases w41,62,65x. Finally, more recent studies of the rat

59

and cat CBs, using an electrode residing within the CB itself w5,10,11,36,66x, have suggested a dissociation of dopamine from CB neural output in response to hypoxia. The precise mechanisms operating which explain how during hypoxia the final output from the CB — the neural activity — increases two- to four-fold while perhaps the most likely candidate for excitatory transmitter in the process — ACh — increases significantly less than that await further investigation. It seems quite probable that the output will result from an interplay of several factors and not just the quantity of the excitatory transmitter. This report strengthens the role of ACh as an essential excitatory transmitter in the CB’s process of chemotransducing hypoxia into increased neural output.

Acknowledgements Supported by HL 50712. We express sincere gratitude to Dr. Ralph Lydic, Professor and Director, Division of Anesthesia and Neuroscience Research, Department of Anesthesia, College of Medicine, The Milton S. Hershey Medical Center, Pennsylvania State University, Hershey, PA 17033, who was so generous with his time, resources and expertise in both instructing us in the HPLC analysis of ACh and in analyzing our earlier samples.

References w1x M. Al-Obaidi, F. Karim, Primary effects of carotid chemoreceptor stimulation on gracilis muscle and renal blood flow and renal function in dogs, J. Physiol. ŽLondon. Ž1993. 73–88. w2x J. Asselin, M. Waebroeck, P. Robberecht, J. Christophe, Effects of pH on binding of agonists and antagonists to rat heart muscarinic receptor, Biochem. J. 216 Ž1983. 11–19. w3x T.J. Biscoe, A. Silver, The distribution of cholinesterases in the cat carotid body, J. Physiol. ŽLondon. 183 Ž1966. 501–512. w4x P. Boska, B. Livett, Substance P protects against desensitization of the nicotinic response in isolated adrenal chromaffin cells, J. Neurochem. 42 Ž1984. 618–627. w5x D. Buerk, S. Lahiri, D. Chugh, A. Mokashi, Electrochemical detection of rapid DA release kinetics during hypoxia in perfused-superfused cat CB, J. Appl. Physiol. 78 Ž1995. 830–837. w6x M. De Burgh Daly, Interactions between respiration and circulation, in: N. Cherniack, J. Widdicombe ŽEds.., Handbook of Physiology, Section 3: The Respiratory System, Vol. 2, Control of Breathing, Part 2, American Physiological Society, Bethesda, MD, 1986, pp. 529–594. w7x B. Dinger, T. Hirano, S.J. Fidone, Autoradiographic localization of muscarinic receptors in rabbit carotid body, Brain Res. 367 Ž1986. 328–331. w8x R.J. Docherty, D.S. McQueen, Inhibitory action of dopamine on cat carotid chemoreceptors, J. Physiol. ŽLondon. 279 Ž1978. 425–436. w9x R.J. Docherty, D.S. McQueen, The effects of acetylcholine and dopamine on carotid chemosensory activity in the rabbit, J. Physiol. ŽLondon. 288 Ž1979. 411–423. w10x D.F. Donnelly, Does catecholamine secretion mediate the hypoxiainduced increase in nerve activity?, Biol. Signals 4 Ž1995. 304–309. w11x D.F. Donnelly, Chemoreceptor nerve excitation may not be propor-

60

w12x

w13x w14x

w15x

w16x

w17x

w18x

w19x

w20x

w21x

w22x

w23x

w24x

w25x

w26x

w27x

w28x

w29x

w30x

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61 tional to catecholamine secretion, J. Appl. Physiol. 81 Ž1996. 657– 664. W.W. Douglas, The effect of a ganglion-blocking drug, hexamethonium, on the response of the cat’s carotid body, J. Physiol. ŽLondon. 118 Ž1953. 373–383. W.W. Douglas, Is there chemical transmission at chemoreceptors?, Pharmacol. Rev. 6 Ž1954. 81–83. U.S. von Euler, G. Liljestrand, Y. Zotterman, The excitation mechanism on the chemoreceptors of the carotid body, Skand. Arch. Physiol. 83 Ž1939. 132–152. C. Eyzaguirre, R.S. Fitzgerald, S. Lahiri, P. Zapata, Arterial chemoreceptors, in: J.T. Shepherd, F.M. Abboud ŽEds.., Handbook of Physiology. The Cardiovascular System. Peripheral Circulation and Organ Blood Flow, Section 2, Vol. 3, Part 2, American Physiological Society, Bethesda, MD, 1983, pp. 557–662. C. Eyzaguirre, H. Koyano, J.R. Taylor, Presence of acetylcholine and transmitter release from carotid body chemoreceptors, J. Physiol. ŽLondon. 178 Ž1965. 463–476. C. Eyzaguirre, P. Zapata, The release of acetylcholine from the carotid body tissues. Further study on the effects of acetylcholine and cholinergic blocking agents on the chemosensory discharge, J. Physiol. ŽLondon. 195 Ž1968. 589–607. S.J. Fidone, S.T. Weintraub, W.B. Stavinoha, Acetylcholine content of normal and denervated cat carotid bodies measured by pyrolysis gas chromatographyrmass fragmentometry, J. Neurochem. 26 Ž1976. 1047–1049. R.S. Fitzgerald, G.A. Dehghani, Neural responses of the cat carotid and aortic bodies to hypercapnia and hypoxia, J. Appl. Physiol. 52 Ž1982. 596–601. R.S. Fitzgerald, P. Garger, M.C. Hauer, H. Raff, L. Fechter, Effect of hypoxia and hypercapnia on catecholamine content in cat carotid body, J. Appl. Physiol. 54 Ž1983. 1408–1413. R.S. Fitzgerald, S. Lahiri, Reflex responses to chemoreceptor stimulation, in: N. Cherniack, J. Widdicombe ŽEds.., Handbook of Physiology, Section 3: The Respiratory System, Vol. 2. Control of Breathing, Part 1, American Physiological Society, Bethesda, MD, 1986, pp. 313–362. R.S. Fitzgerald, D. Parks, Effect of hypoxia on carotid chemoreceptor response to carbon dioxide in cats, Respir. Physiol. 12 Ž1971. 218–229. R.S. Fitzgerald, M. Shirahata, Systemic responses elicited by stimulating the carotid body: primary and secondary mechanisms, in: C. Gonzalez ´ ŽEd.., The Carotid Body Chemoreceptors, Landes Bioscience, Austin, TX, 1997, pp. 171–191. R.S. Fitzgerald, M. Shirahata, Acetylcholine and carotid body excitation during hypoxia in the cat, J. Appl. Physiol. 76 Ž1994. 1566– 1574. R.S. Fitzgerald, M. Shirahata, T. Ide, Further cholinergic aspects of carotid body chemotransduction of hypoxia in the cat, J. Appl. Physiol. 82 Ž1997. 829–898. C. Gonzalez, L. Almaraz, A. Obeso, R. Rigual, Carotid body ´ chemoreceptors: from natural stimuli to sensory discharges, Physiol. Rev. 74 Ž1994. 829–898. C. Gonzalez, B.G. Dinger, S.J. Fidone, Mechanisms of carotid body ´ chemoreception, in: J.A. Dempsey, A.I. Pack ŽEds.., Regulation of Breathing, Marcel Dekker, New York, 1995, pp. 391–471. C. Gonzalez, B. Dinger, S.J. Fidone, Functional significance of ´ chemoreceptor cell neurotransmitters, in: C. Gonzalez ´ ŽEd.., The Carotid Body Chemoreceptors, Landes Bioscience, Austin, TX, 1997, pp. 47–64. A. Gual, J. Marsal, Application of the chemiluminescent method to carotid body for detecting choline and acetylcholine, in: J.A. Ribeiro, D.J. Pallot ŽEds.., Chemoreceptors in Respiratory Control, Croom Helm, London, 1987, pp. 108–113. D. Heath, P. Smith, Diseases of the Human Carotid Body, SpringerVerlag, London, 1992, pp. 127–131.

w31x C. Heymans, E. Neil, Reflexogenic Areas of the Cardiovascular System, Churchill, London, 1958, p. 191. w32x T. Hirano, B. Dinger, K. Yoshizaki, C. Gonzalez, S. Fidone, Nico´ tinic versus muscarinic binding sites in cat and rabbit carotid body, Biol. Signals 1 Ž1992. 143–149. w33x T. Ide, M. Shirahata, C.-L. Chou, R.S. Fitzgerald, Effects of a continuous infusion of dopamine on the ventilatory and carotid body responses to hypoxia in cats, Clin. Exp. Pharmacol. Physiol. 22 Ž1995. 658–664. w34x R. Inoue, Y. Waniishi, Y. Ito, Extracellular Hq modulates acetylcholine-activated nonselective cation channels in guinea pig ileum, Am. J. Physiol. 268 Ž1995. C162–C170. w35x Y. Ishizawa, R.S. Fitzgerald, M. Shirahata, B. Schofield, Localization of nicotinic acetylcholine receptors in cat carotid body and petrosal ganglion, Adv. Exp. Med. Biol. 410 Ž1996. 253–256. w36x R. Iturriaga, J. Alcayaga, P. Zapata, Dissociation of hypoxia-induced chemosensory responses and catecholamine efflux in cat carotid body superfused in vitro, J. Physiol. ŽLondon. 497 Ž1996. 551–564. w37x N. Joels, E. Neil, The idea of a sensory transmitter, in: R.W. Torrance ŽEd.., Arterial Chemoreceptors, Blackwell, Oxford, 1968, pp. 153–178. w38x J.V. Jones, Localization and quantitation of carotid body enzymes: their relevance to the cholinergic transmitter hypothesis, in: M.J. Purves ŽEd.., The Peripheral Arterial Chemoreceptors, Cambridge Univ. Press, London, 1975, pp. 143–144. w39x G. Koelle, R. Davis, E. Diliberto Jr., W. Koelle, Selective, near-total, irreversible inactivation of peripheral pseudocholinesterase and acetycholinesterase in cats in vivo, Biochem. Pharmacol. 23 Ž1974. 175–188. w40x S. Lahiri, R.G. DeLaney, Stimulus interaction in the responses of carotid body chemoreceptor single afferent fibers, Respir. Physiol. 24 Ž1975. 249–266. w41x S. Lahiri, T. Nishino, A. Mokashi, E. Mulligan, Interaction of dopamine and haloperidol with O 2 and CO 2 chemoreception in the carotid body, J. Appl. Physiol. 49 Ž1980. 45–51. w42x S. Landgren, G. Liljestrand, Y. Zotterman, The effects of certain autonomic drugs on the action potential of the sinus nerve, Acta Physiol. Scand. 26 Ž1952. 264–290. w43x L. Li, M.G. McNamee, Modulation of nicotinic acetylcholine receptor channel by pH: a difference in pH sensitivity of Torpedo and mouse receptors expressed in Xenopus oocytes, Cell. Mol. Neurobiol. 12 Ž1992. 83–93. w44x G. Liljestrand, The problem of transmission at chemoreceptors, Pharmacol. Rev. 6 Ž1954. 73–76. w45x F. Llados, P. Zapata, Effects of dopamine analogues and antagonists on carotid body chemosensors in situ, J. Physiol. ŽLondon. 274 Ž1978. 501–509. w46x D.S. McQueen, A quantitative study of the effects of cholinergic drugs on carotid chemoreceptors in the cat, J. Physiol. ŽLondon. 273 Ž1977. 515–532. w47x B. Metz, Release of acetylcholine from the carotid body by hypoxia and hypoxia plus hypercapnia, Respir. Physiol. 6 Ž1969. 386–394. w48x G. Moe, L. Capo, B. Peralta, Action of tetraethylammonium on chemoreceptor and stretch receptor mechanism, Am. J. Physiol. 153 Ž1948. 601–605. w49x L. Monti-Bloch, C. Eyzaguirre, A comparative physiological and pharmacological study of cat and rabbit carotid body chemoreceptors, Brain Res. 193 Ž1980. 449–470. w50x E. Neil, N. Joels, The carotid glomus sensory mechanism, in: D.J.C. Cunningham, B.B. Lloyd ŽEds.., The Regulation of Human Respiration, F.A. Davis, Philadelphia, 1963, pp. 163–172. w51x K. Nishi, C. Eyzaguirre, The action of some cholinergic blockers on carotid body chemoreceptors in vivo, Brain Res. 33 Ž1971. 37–56. w52x A. Palma, L. Li, X. Chen, P. Pappone, M.G. McNamee, Effects of pH on acetylcholine receptor function, J. Membr. Biol. 120 Ž1991. 67–73.

R.S. Fitzgerald et al.r Brain Research 841 (1999) 53–61 w53x N. Prabhakar, J. Mitra, N. Cherniack, Role of substance P in hypercapnic excitation of carotid chemoreceptors, J. Appl. Physiol. 63 Ž1987. 2418–2425. w54x N. Prabhakar, M. Runold, Y. Yamamoto, H. Lagercrantz, C. von Euler, Effect of substance P antagonist on the hypoxia-induced carotid chemoreceptor activity, Acta Physiol. Scand. 121 Ž1984. 301–303. w55x H. Raff, S. Tzankoff, R.S. Fitzgerald, Chemoreceptor involvement in cortisol responses to hypoxia in ventilated dogs, J. Appl. Physiol. 52 Ž1982. 1092–1096. w56x S.R. Sampson, Effects of mecamylamine in responses of carotid body chemoreceptors in vivo to physiological and pharmacological stimuli, J. Physiol. ŽLondon. 212 Ž1971. 655–666. w57x S.R. Sampson, M.J. Aminoff, R.A. Jaffe, E.H. Vidruk, Analysis of inhibitory effect of dopamine on carotid body chemoreceptors in cats, Am. J. Physiol. 230 Ž1976. 1494–1498. w58x S.R. Sampson, E.H. Vidruk, Hyperpolarizing effects of dopamine on chemoreceptor nerve endings from cat and rabbit carotid bodies in vitro, J. Physiol. ŽLondon. 268 Ž1977. 211–221. w59x L. Share, M. Levy, Effect of carotid chemoreceptor stimulation on plasma antidiuretic titer, Am. J. Physiol. 210 Ž1966. 157–161.

61

w60x M. Shirahata, R.S. Fitzgerald, J.S.K. Sham, Acetylcholine increases intracellular calcium of arterial chemoreceptor cells from adult cats, J. Neurophysiol. 78 Ž1997. 2388–2395. w61x M. Shirahata, Y. Ishizawa, M. Rudisill, B. Schofield, R.S. Fitzgerald, Presence of nicotinic acetylcholine receptors in cat carotid body afferent system, Brain Res. 814 Ž1998. 213–217. w62x S. Tomares, O.S. Bamford, L.M. Sterni, R.S. Fitzgerald, J.L. Carroll, Effects of domperidone on neonatal and adult carotid chemoreceptors in the cat, J. Appl. Physiol. 77 Ž1994. 1274–1280. w63x P. Zapata, Effects of dopamine on carotid chemo- and baroreceptors in vitro, J. Physiol. ŽLondon. 244 Ž1975. 235–251. w64x P. Zapata, Chemosensory activity in the carotid nerve: effects of physiological variables, in: C. Gonzalez ´ ŽEd.., The Carotid Body Chemoreceptors, Landes Bioscience, Austin, TX, 1997, pp. 97–117. w65x P. Zapata, Chemosensory activity in the carotid nerve: effects of pharmacological agents, in: C. Gonzalez ´ ŽEd.., The Carotid Body Chemoreceptors, Landes Bioscience, Austin, TX, 1997, pp. 119–146. w66x P. Zapata, R. Iturriaga, J. Alcayaga, Dopamine efflux from the carotid during hypoxic stimulation, Adv. Exp. Med. Biol. 410 Ž1996. 261–266.