Direct measurement of acetylcholine in guinea ig trachea DAVID
G. BAKER,
HILLARY
F. DON, AND JAMES
release K. BROWN
Pulmonary and Critical Care Medicine, Medical Service, and Anesthesiology Service, Veterans Affairs Medical Center, San Francisco 94121; and Cardiovascular Research Institute and Departments of Medicine and Anesthesia, University of California, San Francisco, California 94143 Baker, David G., Hillary F. Don, and James K. Brown. Direct measurement of acetylcholine release in guinea pig trachea. Am. J. Physiol. 263 (Lung Cell. Mol. Physiol. 7): L142-L147, 1992.-In this study, we applied high-performance liquid chromatography with electrochemical detection (HPLCEC) to the measurement of acetylcholine (ACh) release from nerve endings in guinea pig tracheal smooth muscle. We also tested for muscarinic inhibitory regulation of ACh release in this species, which is widely used for studies of airway neural control. Clip-connected segments of the posterior membrane of the guinea pig trachea were mounted in organ baths between stimulating electrodes and incubated in Krebs-Henseleit buffer containing (in PM) IO indomethacin, 1 neostigmine, 1 phentolamine, and 1 propranolol. To measure ACh, the bath was emptied and aliquots of buffer were injected directly into the HPLCEC; the lower limit of detection was 1 pmol/ZOO ~1 sample. Electrical field stimulation (EFS) at 5 Hz for 10 or 30 min increased ACh release from 1.8 I~I 1.4 (SE) to 6.2 t 1.3 pmol mg protein-l min-l (n = 15). The effect of atropine was examined by comparing amounts of ACh released by EFS before and after exposure to either atropine (0.3 PM) or vehicle. Before atropine treatment, EFS-evoked ACh release was 4.9 & 0.6 pmol. mg protein-l min-I; after atropine exposure, EFS-evoked release of ACh increased significantly to 15.0 * 2.2 pmol. mg protein-l min-l (n = 11; P < 0.05). Corresponding values before and during exposure to vehicle were 9.3 t 4.4 and 10.7 t 4.7 pm01 . mg protein-l min-l, respectively. The ratios of the changes in EFS-evoked ACh release were 3.1 t 0.3 and 1.3 k 0.1 in atropine-treated and vehicle-treated groups, respectively (P c 0.05). We conclude that HPLC-EC is a reliable and sensitive technique for the detection of EFS-evoked release of ACh from clip-connected segments of guinea pig tracheal smooth muscle. airway smooth muscle; autoinhibition; presynaptic regulation; high-performance liquid chromatography with electrochemical detection; muscarinic receptors l
l
l
l
l
of presynaptic inhibitory muscarinic receptors on peripheral cholinergic nerves is well established (31). Their presence on cholinergic nerves innervating airway smooth muscle has also been suggested, although the evidence for such receptors has been based largely on comparisons of contractile responses to stimulating nerve endings vs. administering exogenous acetylcholine (ACh) (16, 24). Few studies have employed direct measurement of ACh release (1, 12, 23). Study of presynaptic muscarinic receptors is relevant to diseases of the airways because recent evidence suggests that abnormal changes in these receptors may, in part, explain the bronchial hyperresponsiveness occurTHE PRESENCE
L142
1040-0605/92
$2.00 Copyright
0
ring in asthmatic patients (3, 19). These considerations highlight the need for direct measurements of ACh release from nerve endings in airway smooth muscle and for the development of models to test for dysfunction of muscarinic autoinhibitory mechanisms in disease. We used the guinea pig because this species demonstrates bronchial hyperresponsiveness after antigen challenge (US), viral infection (9), and ozone or smoke exposure (26, 27) and thus would provide abundant opportunities to examine the role of presynaptic muscarinic receptors in different forms of hyperresponsiveness. Moreover, this species is inexpensive and readily available for experimental use. For the direct measure of ACh, we chose high-performance liquid chromatography with electrochemical detection (HPLC-EC) because alternative methods have significant limitations. Bioassays, used in the past to measure the release of an ACh-like substance from guinea pig trachea, clearly lack specificity (11). Incubation of tissue with [3H]choline, followed by the monitoring of tritium in the media to reflect ACh release, is an extremely sensitive technique and has been applied recently in rat bronchi (1) and guinea pig trachea (12, 21). An important limitation, however, is that all neural compartments containing releasable ACh are not labeled with tritium equally (4, 22). Therefore changes in the amount of radiolabel released may not accurately reflect changes in the release of all ACh. A radioenzymatic assay for ACh also has been applied successfully in airway tissues (23, 30) and has adequate specificity and sensitivity. However, this technique suffers somewhat from a requirement for extracting ACh from the media before the assay, and the assay itself is time consuming. A relatively new technique for measuring ACh is HPLC-EC (28). An important advantage of this technique is that it requires no extraction, and media surrounding a tissue may be injected directly into the HPLC-EC. Also, the technique is extremely rapid arid relatively simple, permitting the processing of multiple samples within minutes. Specificity and sensitivity are excellent. Moreover, recent reports suggest that HPLC-EC may be sufficiently sensitive to detect ACh release from nerve endings in canine tracheal smooth muscle (13-S). The first goal of our study was to evaluate whether
1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
HPLC-EC was suitable as a technique for the measurement of ACh release from guinea pig airways. If this first goal was realized, the second goal was to employ this technique to study muscarinic inhibition of ACh release. Using [3H]choline uptake to detect subsequent ACh release, previous investigators (21) already have shown that stimulus-evoked release of ACh is enhanced by muscarinic antagonists like atropine; the inhibitory neuronal muscarinic receptor involved is the M2 subtype, as in other peripheral cholinergic systems (5, 6). Therefore we evaluated the usefulness of HPLC-EC in guinea pig trachea by confirming the enhancing effect of atropine on stimulus-evoked release of ACh. MATERIALS
AND
METHODS
Tracheal smooth muscle. Male guinea pigs (n = 15) weighing 400-450 g (Simmonsen Laboratories, Gilroy, CA) were anesthetized with pentobarbital sodium (20-40 mg/kg ip), and clipconnected segments of trachea were prepared according to our modifications of the method of D’Agostino et al. (12). The trachea was removed and immersed in room-temperature Krebs-Henseleit buffer containing indomethacin (KHI j and bubbled continuously with 95% O,-5% CO,. Indomethacin was added to retard cyclooxygenase activity and to prevent the progressive decrease in the contractile response evoked by electrical field stimulation (EFS) (34). The trachea was cleaned of surrounding tissue and divided into five segments of six or seven cartilaginous rings each. Segments were cleaned of epithelium by rubbing with a pipe cleaner. The segments were placed side by side with the anterior side up and were connected together by small vascular clips (Hemoclip, Edward Week); clips were attached to the edge of the cartilaginous rings as close as possible to the posterior membrane. Sutures were looped through the two end segments, the cartilaginous portions of the three middle segments were removed, and the entire clip-connected trachea was mounted in the 2-ml organ bath. One end of the sutures was looped to the bottom of the bath and the other to a forcedisplacement transducer (Grass FT03) for measurement of tension. Tension was recorded on a polygraph (Grass model 7). The contents of the KHI buffer were (in mM) 118 NaCl, 74.6 KCl, 2.5 CaCl,, 2.4 MgS04, 1.1 KH2P04, 25.5 NaHCO,, IO glucose, and 0.01 indomethacin. A resting tension of 1 g was applied initially and readjusted as necessary during the incubation period. During this 90- to
ACH
RELEASE
L143
120-min period, the bath was maintained at 37°C bubbled continuously with 95% 02-5% C02, and emptied and refilled with fresh buffer every 15 min. ACh detection. ACh was measured by HPLC-EC with our modifications of the method first described by Potter et al. (28). The analytical column, immobilized enzyme reactor (IMER; Bioanalytical Systems, West Lafayette, IN), and electrochemical detector (EG&G Princeton Applied Research model 400 EC detector or Bioanalytical Systems LC-4C detector) were arranged as shown in Fig. IA. Mobile phase (50 mM Na,HPO,, pH 8.5) was pumped at a rate of 1 ml/min by a Rainin Rabbit high-performance solvent delivery system. The EC detector, fitted with a platinum electrode set at 500 mV, was connected to a Macintosh SE computer running a Rainin Dynamax HPLC Method Manager to acquire and analyze ACh peaks. With this technique, ACh, choline, and some choline analogues were retained by the analytical column (MF-6150) and were thereby separated from all other substances in the sample. After several minutes, the retained substances were eluted sequentially so that they were separated from each other. ACh was eluted and, in the IMER (MF-6151), converted first to choline by acetylcholinesterase and then to H,O, by choline oxidase. As H,O, flowed across the platinum electrode of the EC, it was oxidized, producing a current whose magnitude reflected the amount of ACh present in the original sample (Fig. 1B). Next, choline was eluted and converted in the IMER to H202, producing a separate peak whose magnitude reflected the amount of choline present in the sample. In some experiments a second IMER (MF-6153) was used. This second IMER, which contained catalase and choline oxidase, was placed before the analytical column; it removed choline before choline was retained by the analytical column. However, we saw no advantage to the use of this column, particularly because choline and ACh were easily distinguished, and we used the second IMER in two experiments only. Experimental procedures. After incubation of the clip-connected segment in KHI buffer, the baths were refilled with KHI buffer containing 1 ,um each of neostigmine, propranolol, and phentolamine, and incubation was continued for an additional 30 min. Neostigmine was present to prevent hydrolysis of ACh by endogenous acetylcholinesterase activity. Propranolol and phentolamine were present to inhibit any pre- or postsynaptic ,& and cw-adrenergic receptors. After its addition, each agent was present throughout the remainder of the experiment. At the end of this incubation period, the contents of the bath
Organ bath Fig. 1. A: High-performance liquid chromatography with electrochemical detection (HPLC-EC) method for detection of acetylcholine (ACh). Samples of buffer from organ bath are introduced through sample injector. In analytical column, ACh, choline, and some choline analogues are retained while all other substances are eluted, forming unretained peak of chromatogram shown in B. After several minutes, retained substances are eluted. In immobilized enzyme reactor (IMER), ACh is converted by acetylcholinesterase to choline and, subsequently, by choline oxidase to H202, which is oxidized at platinum electrode of electrochemical (EC) detector, producing ACh peak of chromatogram in B. Chromatogram produced by injecting a standard sample containing 10 pmol ACh in water is shown in B.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
L144
GUINEA
PIG
TRACHEAL
were emptied, the bath was refilled with buffer (same contents), and a rest period was begun (period R; n = 15 tracheas). After 30 min, the contents of the bath were analyzed for ACh. The bath was refilled with buffer, and EFS was applied for 30 (n = 8 tracheas) or 10 (n = 7) min, after which the contents of the bath were again analyzed for ACh (period S,; n = 15). Electrical impulses consisted of square-wave pulses (Winston Applied Electronics) having a duration of 0.5 ms and a frequency of 5 Hz. Current delivered to the bath was -400 mA and was monitored continuously (Tetronix single-beam storage oscilloscope). To examine the effect of atropine, we measured EFS-evoked release of ACh before and after addition of atropine (0.3 PM) to the buffer (group 1, n = 11 tracheas). Group 1 consisted of two subgroups: in group IA (n = 4), EFS was applied for 30 min (periods S, and S,), and atropine was always added after S,; in group 1B (n = 7), the period of stimulation was reduced to 10 min, but stimulation was performed a total of four times (periods S,-S,). In group 1 B, atropine was added after S, (n = 3) or after S, (n = 4). Group 2 (n = 4) served as controls for group IA; protocols were carried out identically except that vehicle, instead of atropine, was added after S,. In all protocols, 30 min was allowed between stimulations. To measure ACh in the organ bath, we emptied the bath, and the entire Z-ml volume was filtered (0.2 PM Acrodisc). Aliquots (1 ml) were introduced into the HPLC-EC through the injector, overfilling the 200~~1 sample loop. Subsequently, the sample was placed into the flow of the mobile phase. Amounts of ACh were determined by comparing areas of unknown curves with areas of standard curves; the latter were generated by injecting ACh standards (2-20 pm) before and after each experimental run. At the end of the experiment, the posterior membrane was separated from the cartilage and dissolved in NaOH, and protein content was determined by the method of Bradford (7). Data were expressed as picomoles per milligram of protein per minute, or as grams tension and were compared by Student’s paired t test (37). In addition, ratios of ACh release [Safter (S,)] for atropine-treated (group 1) and vehicle(wl%?fbre treated (group 2) preparations were calculated and compared by analysis of variance (37). The limit of detection was defined as that amount of ACh whose peak height was equal to three times the baseline noise. Retention time was defined as the time between injection of sample and the point of maximum concentration of the eluted peak. The following pharmacological agents were obtained from Sigma Chemical: acetylcholine hydrochloride, choline chloride, atropine sulfate, sodium phosphate dibasic, indomethacin, neostigmine bromide, and propranolol hydrochloride. Phentolamine (Regitine HCl) was obtained from Ciba-Geigy.
ACH
RELEASE
KHI buffer from the organ bath produced chromatograms with large choline peaks and relatively small ACh peaks. With EFS, the ACh peak increased in magnitude but remained small in comparison with the choline peak. Nevertheless, the choline peak was clearly separated from the ACh peak (Fig. 2). Mean quantity of ACh measured at the end of period R was 1.8 t 1.4 (SE) pmol mg protein-l min-I; in 12 of the experiments no ACh was measured. It should be noted that this almost complete absence of measured ACh during period R occurred even though the addition of neostigmine, propranolol, and phentolamine was associated with an increase in tension from 1.0 t 0.1 to 2.3 t 0.2 g (P < 0.05; n = 15). With EFS (period S,), tension rose to 3.5 t 0.2 g, and ACh release was evident in all 15 experiments, averaging 6.2 t 1.3 pmol mg protein-l min? l
l
l
l
Effect of atropine (0.3 PM) on EFS-evoked release of ACh. In group 1, tension fell to 0.6 t 0.1 g with addition of atropine to the organ bath. In group lA, EFS-evoked
release of ACh was 5.0 t 1.7 pmol mg protein-’ mine1 before atropine treatment (S,); with the addition of atropine (S,), EFS-evoked release increased to a mean of 18.2 t 5.1 pmolmg protein-l emin-l (P < 0.05). The ratio of the amount of ACh evoked by EFS after atropine (S,) to the amount before (Sb) for group IA was 3.6 t 0.4. In group U?, addition of atropine between periods S1 and S2 increased EFS-evoked release of ACh from 4.2 t 0.6 to 10.8 t 3.5 pmol mg protein-l mine1 with a mean S,/Sb of 2.4 t 0.4. Th e amount of ACh released remained at this elevated level throughout the rest of the experiment (during S3 and S,). When atropine was added between periods S3 and Sq, EFS-evoked release was unchanged with successive stimulations before atropine (i.e., S1 through S,), but after atropine addition (S,), EFS-evoked release again increased, from 5.0 t 0.5 to 15.0 t 1.7 pmol mg protein-l min- l (Fig. 3), with S,& equaling 3.1 t 0.6. In group 2 (n = 4), EFS-evoked release of ACh was 9.3 t 4.4 pmolmg protein-l. mine1 before and 10.7 t 4.0 pm01 mg protein-l min-l after vehicle treatment. The ratio of ACh evoked before and after vehicle treatment was 1.3 t 0.1; thus there was a small increase in evoked release with the second stimulation in the control group. In mouw 2 the increase in tension was the same before l
l
l
l
”
l
I
RESULTS
In preliminary experiments, the addition (in PM) of 10 indomethacin, 1 propranolol, 1 phentolamine, or 1 atropine to the Krebs-Henseleit buffer did not create retention peaks. For ACh and choline dissolved in water, retention peaks were generated at 4.6 (n = 40) and 6.0 (n = 8) min, respectively. When standards containing Z-20 pmol ACh were injected, both the area and the height of the ACh peaks were related linearily to the amount injected. Injection of progressively smaller amounts of ACh showed that the limit of detection was -1 pmo1/200 ~1 sample. Sensitivity depended significantly on the background noise of the particular experimental day and on the length of previous use of that particular enzyme column. When translated into molar concentrations in the organ bath, the lower limit of detection was ~5 nM. Suitability of our technique. After period R, samples of
Fig. ‘2. Chromatogram produced by injecting sample from an organ bath after 30-min electrical field stimulation (EFS). Very large choline (Ch) peak did not interfere with much smaller ACh peak. In this chromatogram (1 nA full scale) only lower portion of Ch peak is shown. Identity of ACh and Ch peaks was determined by comparing their retention times with retention times of known standards. Very small peak, seen here at 3 min (arrow), was not identified.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
ACH
RELEASE
L145
DISCUSSION
t
1
I
' 1 min
I
1
Fig. 3. Effect of atropine on EFS-evoked release of ACh. A: clip-connected tracheal segments were stimulated for 10 min at 5 Hz; a 30-min equilibration period was allowed between stimulations. Atropine (0.3 &I) was added after period S, (squares) or after period S3 (triangles) and allowed to remain in bath thereafter. ACh values are means & SE. B: portions of individual chromatograms generated in 1 experiment after periods S3 and S,. Two chromatograms were recorded-at same sensitivity (5 nA full scale), and recordings began 3 min after injection of a sample. In this experiment, atropine was added after S3. Also, a second IMER was present, which removed choline before it could be retained by analytical column, explaining absence of choline peaks in these chromatograms.
(1.1 g) and after (1.2 g) addition of the vehicle. The ratios S,& of EFS-evoked ACh release before and after atropine treatment or control are pooled and presented in Fig. 4.
atropine
vehicle
Fig. 4. Enhancing effect of atropine on EFS-evoked release of ACh. These histograms summarize ratios of ACh evoked before (S,) and after (Sb) atropine treatment (group I; n = 11 tracheas) or before and after vehicle treatment (group 2; n = 4). Tracheas were incubated in 0.3 PM atropine for 30 min before second EFS, and atropine remained in the bath throughout lo- or 30-min EFS. Values are means t SIZ and are taken from stimulation periods immediately before and after atropine addition. * Statistically significant (P < 0.05).
First, this study demonstrates that HPLC-EC is suitable for measurement of EFS-evoked release of ACh from nerve endings in guinea pig tracheal smooth muscle. We found that the lower limit of detection in the organ bath was 4 nM. It is interesting to note that we could not detect significant levels of ACh in period R, even though smooth muscle tone was elevated above initial set levels, presumably as a result of the inhibition by neostigmine of acetylcholinesterase activity and perhaps also its capacity to stimulate release of ACh (10). Thus, before application of an electrical field, the quantity of ACh in the neuromuscular junction was sufficient to increase tension significantly but was still below the level of detection by our technique. Second, this study shows that the muscarinic antagonist, atropine, enhances EFS-evoked release of ACh. In the central nervous system, as well as the enteric and other peripheral cholinergic nervous systems, EFS-induced release of ACh is under autoinhibitory feedback control, mediated by muscarinic receptors on cholinergic nerve terminals (5, 6, 20, 29, 36). In all of these systems, muscarinic antagonists enhance the release of ACh, probably by blocking this negative feedback control. We suggest that a similar mechanism explains our results. Atropine inhibited muscarinic receptors on cholinergic nerve terminals, prevented the negative feedback, and allowed ACh release to escape the normal autoregulatory mechanism. Our results, therefore, confirm conclusions of previous published studies. Of significance, however, is the relatively large increase in release of ACh that we report. In the study by Aas and Fonnum (1) in rat bronchial smooth muscle, scopolamine increased the potassium-evoked release of [3H]ACh by only 33%; in the study by D’Agostino et al. (12) in the guinea pig trachea, atropine facilitated release of ACh, increasing the EFS-evoked outflow of [3H]ACh by 39%. These relatively small increases contrast with the large increases, 310 and 250%, respectively, that we and Deckers et al. (14) report. In contrast to our technique and that of Deckers et al. (14), the measurement of ACh in both of the other studies employed radiolabeling where the presence of cholinesterase inhibitors was not required. Inherent in techniques used by Deckers and us is the requirement for addition of neostigmine to the bath. It is possible that the greater degree of autoinhibition in our preparation was due to the presence of neostigmine and the consequent elevated levels of ACh in the vicinity of the neuronal receptors. It has been suggested (17) that autoinhibitory mechanisms are not operative normally inasmuch as ACh is immediately hydrolyzed by acetylcholinesterase and is not able to act at presynaptic receptors. Our results give rise to the intriguing possibility that autoinhibition becomes significant only in the presence of high levels of ACh at the neuronal junction. If it is a brake on the release of ACh, the effects of autoinhibition are only seen in the presence of a process that causes increased levels of ACh, such as during acetylcholinesterase inhibition or perhaps disease. Support for this possibility comes from
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
L146
GUINEA
PIG
TRACHEAL
studies in the guinea pig myenteric plexus, where autoinhibitory mechanisms have been investigated in the presence and absence of another inhibitor of acetylcholinesterase, physostigmine (20). In the absence of physostigmine, muscarinic inhibition increased electrically evoked release of ACh by 37% compared with a ~200% increase in the presence of physostigmine (20). Thus autoinhibitory mechanisms may be functionally operative only in the presence of increased release, or decreased breakdown, of ACh. Likewise, the recently described lateral inhibitory mechanisms operating at adjacent release sites may be functional only in the presence of elevated levels of ACh (8). It is possible that a strong muscarinic autoinhibitory mechanism may interfere with the ability to detect regulatory actions of other inhibitory presynaptic receptors. Support for this possibility comes from studies in the myenteric plexus (35). In the absence of physostigmine, the ,&adrenergic agonist, isoproterenol, inhibited the evoked released of [3H]ACh, but in the presence of physostigmine and presumably high autoinhibitory tone the addition of isoproterenol failed to inhibit ACh release. This mechanism may account for the failure of investigators to detect any effect of isoproterenol on ACh release in canine trachea (23). Such interference by a strong muscarinic autoinhibitory system may suggest common signaling systems for different presynaptic receptors. In the rat cerebral cortex, for example, presynaptic muscarinic and opiate receptors both utilize 4aminopyridinesensitive K+ channels, whose activation hyperpolarizes the membrane, terminates calcium influx, and inhibits ACh release (33). Besides muscarinic, opiate, and adrenergic receptors, cholinergic nerve terminals in airway smooth muscle probably have receptors for other agonists. In a recent study, Deckers et al. (15) used HPLC-EC to provide evidence for inhibitory prostaglandin E2 receptors on cholinergic nerve terminals in dog bronchial smooth muscle. Their results are consistent with the interpretation of previous indirect studies (34). In contrast, however, Deckers et al. (15) found no evidence for presynaptic prostaglandin D2 receptors, whose existence has been postulated in previous studies with indirect techniques (32). This inconsistency serves to underscore the importance of direct measurements of ACh release. In conclusion, our data support previous suggestions that muscarinic receptors on cholinergic nerves innervating airway smooth muscle inhibit the release of ACh (24). The significance of these muscarinic inhibitory receptors awaits further investigation, and we suggest the possibility that autoinhibition may become functionally important only in the presence of abnormal amounts of ACh at the neuronal junction. Recent indirect evidence (2, 25) suggests that the muscarinic autoinhibitory mechanism is dysfunctional in asthma. Our description of a technique for the measurement of ACh release in guinea pig trachea provides an opportunity for studying presynaptic regulatory mechanisms in a variety of animal models of bronchial hyperresponsiveness. This work was supported by Tobacco-Related Disease Research Program 2 RT-265. Grant HL-27669 from the National Heart, Lung, and
ACH
RELEASE
Blood Institute, and the Research Service of the Department of Veterans Affairs. Address for reprint requests: D. G. Baker, Pulmonary and Critical Care Medicine (11 lD), Veterans Affairs Medical Center, 4150 Clement St., San Francisco, CA 94121. Received
24 February
1992; accepted
in final
form
5 May
1992.
REFERENCES 1. Aas, P., and F. Fonnum. Presynaptic inhibition of acetylcholine release. Acta Physiol. Stand. 127: 335-342, 1986. 2. Ayala, L. E., and T. Ahmed. Is there loss of a protective muscarinic receptor mechanism in asthma? Chest 96: 1285-1291, 1989. 3. Barnes, P. J. Muscarinic autoreceptors in airways. Their possible role in airway disease. Chest 96: 1220-1221, 1989. 4. Beani, L., C. Bianchi, A. Siniscalchi, L. Sivilotti, S. Tanganelli, and E. Veratti. Different approaches to study acetylcholine release: endogenous ACh versus tritium efflux. NuunynSchmiedeberg’s Arch. Pharmacol. 328: 119-126, 1984. 5. Bognar, I. T., B. Beinhauer, P. Kann, and H. Fuder. Different muscarinic receptors mediate autoinhibition of acetylcholine release and vagally-induced vasoconstriction in the rat isolated perfused heart. Naunyn-Schmiedeberg’s Arch. Pharmacol. 341: 279-287, 1990. 6. Bognar, I. T., M. T. Wesner, and H. Fuder. Muscarinic receptor types mediating autoinhibition of acetylcholine release and sphincter contraction in the guinea-pig iris. Nuunyn-Schmiedeberg’s Arch. Pharmacol. 341: 22-29, 1990. 7. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein-dye binding. Anal. Biochem. 72: 248-254, 1976. 8. Brock, J. A., T. C. Cunnane, K. Starke, and C. F. Wardell. cu2-Adrenoceptor-mediated autoinhibition of sympathetic transmitter release in guinea-pig vas deferens studied by intracellular and focal extracellular recording of junction potentials and currents. Naunyn-Schmiedeberg’s Arch. Pharmacol. 342: 45-52, 1990. 9. Buckner, C. K., V. Songsiridej, E. C. Dick, and W. W. Busse. In vivo and in vitro studies on the use of the guinea pig as a model for virus-provoked airway hyperreactivity. Am. Rev. Respir. Dis. 132: 305-310, 1985. 10. Carlyle, R. F. The mode of action of neostigmine and physostigmine on the guinea-pig trachealis muscle. Br. J. PharmacoZ. 21: 137-149, 1963. 11. Carlyle, R. F. The response of the guinea-pig isolated intact tracheal to transmural stimulation and the release of an acetylcholine-like substance under condition of rest and stimulation. Br. J. Pharmacol. 22: 126-136, 1964. 12. D’Agostino, G., M. C. Chiari, E. Grana, A. Subissi, and H. Kilbinger. Muscarinic inhibition of acetylcholine release from a novel in vitro preparation of the guinea-pig trachea. NuunynSchmiedeberg’s Arch. Pharmacol. 342: 141-145, 1990. 13. Deckers, I. A., and A. G. Herman. HPLC detection of acetylcholine released from canine airway segments (Abstract). Pfluegers Arch. 412: S42, 1988. 14. Deckers, I. A., M. R. Rampart, and A. 65. Herman. Evidence for presynaptic modulation of acetylcholine release in canine airways using a sensitive HPLC method (Abstract). Arch. Int. Pharmacodyn. Ther. 298: 297, 1989. I. A., M. Rampart, H. Bult, and A. G. Herman. 15. Deckers, Evidence for the involvement of prostaglandins in modulation of acetylcholine release from canine bronchial tissue. Eur. J. Pharmacol. 167: 415-418, 1988A9. 16, Faulkner, D., A. D. Fryer, and J. Maclagan. Postganglionic muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br. J. Pharmacol. 88: 181-187, 1986. 17. Hertting, G., S. Wurster, P. Gebicke-Harter, and C. Allgaier. Participation of regulatory G-proteins and protein kinase C in the modulation of transmitter release in hippocampus. In: Modulation of Synaptic Transmission and Plasticity in Nervous Systems, edited by G. Hertting, and H. C. Spatz. Berlin: SpringerVerlag, 1988, p. 147-164. 18. Ishida, K., L. J. Kelly, R. J. Thomson, L. L. Beattie, and R. R. Schellenberg. Repeated antigen challenge induces airway hyperresponsiveness with tissue eosinophilia in guinea pigs. J. Appl.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
Physiol. 67: 1133-1139, 1989. 19. Jacoby, D. B., and A. D. Fryer. Abnormalities in neural control of smooth muscle in virus-infected airways. Trends PharmacoZ. Sci. 11: 393-395, 1990. 20. Kilbinger, H., R. Kruel, and I. Wessler. Modulation by scopolamine, acetylcholine and choline of the evoked release of acetylcholine from the guinea pig myenteric plexus: evidence for a muscarinic feedback of acetylcholine secretion. In: Cholinergic Mechanisms, edited by G. Pepeu and H. Ladinsky. New York: Plenum, 1981, p. 169-175. 21. Kilbinger, H., R. Schneider, H. Siefken, D. Wolf, and G. D’Agostino. Characterization of prejunctional muscarinic autoreceptors in the guinea-pig trachea. Br. J. PharmucoZ. 103: 1757-1763, 1991. 22. Luz, S., I. Pinchasi, and D. M. Michaelson. Differential inhibition of the release of endogenous and newly synthesized acetylcholine from Torpedo synaptosomes by presynaptic muscarinic receptors. FEBS Lett. 164: 9-12, 1983. 23. Martin, J. G., and B. Collier. Acetylcholine release from canine isolated airway is not modulated by norepinephrine. J. Appl. PhysioZ. 61: 1025-1030, 1986. 24. Minette, P. A., and P. J. Barnes. Prejunctional inhibitory muscarinic receptors on cholinergic nerves in human and guinea pig airways. J. AppZ. Physiol. 64: 2532-2537, 1988. 25. Minette, P. A. H., J.-W. J. Lammers, C. M. S. Dixon, M. T. McCusker, and P. J. Barnes. A muscarinic agonist inhibits reflex bronchoconstriction in normal but not in asthmatic subjects. J. AppZ. Physiol. 67: 2461-2465, 1989. 26. Murlas, C. G., T. P. Murphy, and V. Chodimella. OS-induced mucosa-linked airway muscle hyperresponsiveness in the guinea pig. J. AppZ. Physiol. 69: 7-13, 1990. 27. Nishikawa, M., I. Hirotada, T. Fukuda, S. Suzuki, and T. Okubo. Acute exposure to cigarette smoke induces airway hyperresponsiveness without airway inflammation in guinea pigs. Am. Rev. Respir. Dis. 142: 177-183, 1990.
ACH
RELEASE
L147
28. Potter, P. E., J. L. Meek, and N. H. Neff. Acetylcholine and choline in neuronal tissue measured by HPLC with electrochemical detection. J. Neurochem. 41: 188-194, 1983. M., M. Marchi, G. Maura, and G. Bonanno. Pre29. Raiteri, synaptic regulation of acetylcholine release in the CNS. CeZZ BioZ. Int. Rep. 13: 1109-1118, 1989. 30. Shore, S., B. Collier, and J. G. Martin. Effect of endogenous prostaglandins on acetylcholine release from dog trachealis muscle. J. AppZ. Physiol. 62: 1837-1844, 1987. 31. Starke, K., M. Gothert, and H. Kilbinger. Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol. Rev. 69: 864-989, 1989. 32. Tamaoki, J., K. Sekizawa, P. D. Graf, and J. A. Nadel. Cholinergic neuromodulation by prostaglandin D2 in canine airway smooth muscle. J. AppZ. Physiol. 63: 1396-1400, 1987. 33 . Torocsik, A., and E. S. Vizi. 4-Aminopyridine interrupts the modulation of acetylcholine release mediated by muscarinic and opiate receptors. J. Neurosci. Res. 27: 228-232, 1993. E. H., P. M. O’Byrne, L. M. Fabbri, P. D. Graf, M. 34. Walters, J. Holtzman, and J. A. Nadel. Control of neurotransmission by prostaglandins in canine trachealis smooth muscle. J. AppZ. Physiol. 57: 129-134, 1984. 35. Wessler, I., V. Eschenbruch, S. Halim, and H. Kilbinger. Presynaptic effects of scopolamine, oxotremorine and morphine on [3H]acetylcholine release from the myenteric plexus at different stimulation frequencies and calcium concentrations. NuunynSchmiedeberg’s Arch. Phurmucol. 335: 597-604, 1987. 36. Wetzel, G. T., and J. H. Brown. Presynaptic modulation of acetylcholine release from cardiac parasympathetic neurons. Am. J. PhysioZ. 248 (Heart Circ. Physiol. 17): H33-H39, 1985. J. H. Biostatistical Analysis. Englewood Cliffs, NJ: Pren37. Zar, tice-Hall, 1974, p. 121-150.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
G. BAKER,
HILLARY
F. DON, AND JAMES
release K. BROWN
Pulmonary and Critical Care Medicine, Medical Service, and Anesthesiology Service, Veterans Affairs Medical Center, San Francisco 94121; and Cardiovascular Research Institute and Departments of Medicine and Anesthesia, University of California, San Francisco, California 94143 Baker, David G., Hillary F. Don, and James K. Brown. Direct measurement of acetylcholine release in guinea pig trachea. Am. J. Physiol. 263 (Lung Cell. Mol. Physiol. 7): L142-L147, 1992.-In this study, we applied high-performance liquid chromatography with electrochemical detection (HPLCEC) to the measurement of acetylcholine (ACh) release from nerve endings in guinea pig tracheal smooth muscle. We also tested for muscarinic inhibitory regulation of ACh release in this species, which is widely used for studies of airway neural control. Clip-connected segments of the posterior membrane of the guinea pig trachea were mounted in organ baths between stimulating electrodes and incubated in Krebs-Henseleit buffer containing (in PM) IO indomethacin, 1 neostigmine, 1 phentolamine, and 1 propranolol. To measure ACh, the bath was emptied and aliquots of buffer were injected directly into the HPLCEC; the lower limit of detection was 1 pmol/ZOO ~1 sample. Electrical field stimulation (EFS) at 5 Hz for 10 or 30 min increased ACh release from 1.8 I~I 1.4 (SE) to 6.2 t 1.3 pmol mg protein-l min-l (n = 15). The effect of atropine was examined by comparing amounts of ACh released by EFS before and after exposure to either atropine (0.3 PM) or vehicle. Before atropine treatment, EFS-evoked ACh release was 4.9 & 0.6 pmol. mg protein-l min-I; after atropine exposure, EFS-evoked release of ACh increased significantly to 15.0 * 2.2 pmol. mg protein-l min-l (n = 11; P < 0.05). Corresponding values before and during exposure to vehicle were 9.3 t 4.4 and 10.7 t 4.7 pm01 . mg protein-l min-l, respectively. The ratios of the changes in EFS-evoked ACh release were 3.1 t 0.3 and 1.3 k 0.1 in atropine-treated and vehicle-treated groups, respectively (P c 0.05). We conclude that HPLC-EC is a reliable and sensitive technique for the detection of EFS-evoked release of ACh from clip-connected segments of guinea pig tracheal smooth muscle. airway smooth muscle; autoinhibition; presynaptic regulation; high-performance liquid chromatography with electrochemical detection; muscarinic receptors l
l
l
l
l
of presynaptic inhibitory muscarinic receptors on peripheral cholinergic nerves is well established (31). Their presence on cholinergic nerves innervating airway smooth muscle has also been suggested, although the evidence for such receptors has been based largely on comparisons of contractile responses to stimulating nerve endings vs. administering exogenous acetylcholine (ACh) (16, 24). Few studies have employed direct measurement of ACh release (1, 12, 23). Study of presynaptic muscarinic receptors is relevant to diseases of the airways because recent evidence suggests that abnormal changes in these receptors may, in part, explain the bronchial hyperresponsiveness occurTHE PRESENCE
L142
1040-0605/92
$2.00 Copyright
0
ring in asthmatic patients (3, 19). These considerations highlight the need for direct measurements of ACh release from nerve endings in airway smooth muscle and for the development of models to test for dysfunction of muscarinic autoinhibitory mechanisms in disease. We used the guinea pig because this species demonstrates bronchial hyperresponsiveness after antigen challenge (US), viral infection (9), and ozone or smoke exposure (26, 27) and thus would provide abundant opportunities to examine the role of presynaptic muscarinic receptors in different forms of hyperresponsiveness. Moreover, this species is inexpensive and readily available for experimental use. For the direct measure of ACh, we chose high-performance liquid chromatography with electrochemical detection (HPLC-EC) because alternative methods have significant limitations. Bioassays, used in the past to measure the release of an ACh-like substance from guinea pig trachea, clearly lack specificity (11). Incubation of tissue with [3H]choline, followed by the monitoring of tritium in the media to reflect ACh release, is an extremely sensitive technique and has been applied recently in rat bronchi (1) and guinea pig trachea (12, 21). An important limitation, however, is that all neural compartments containing releasable ACh are not labeled with tritium equally (4, 22). Therefore changes in the amount of radiolabel released may not accurately reflect changes in the release of all ACh. A radioenzymatic assay for ACh also has been applied successfully in airway tissues (23, 30) and has adequate specificity and sensitivity. However, this technique suffers somewhat from a requirement for extracting ACh from the media before the assay, and the assay itself is time consuming. A relatively new technique for measuring ACh is HPLC-EC (28). An important advantage of this technique is that it requires no extraction, and media surrounding a tissue may be injected directly into the HPLC-EC. Also, the technique is extremely rapid arid relatively simple, permitting the processing of multiple samples within minutes. Specificity and sensitivity are excellent. Moreover, recent reports suggest that HPLC-EC may be sufficiently sensitive to detect ACh release from nerve endings in canine tracheal smooth muscle (13-S). The first goal of our study was to evaluate whether
1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
HPLC-EC was suitable as a technique for the measurement of ACh release from guinea pig airways. If this first goal was realized, the second goal was to employ this technique to study muscarinic inhibition of ACh release. Using [3H]choline uptake to detect subsequent ACh release, previous investigators (21) already have shown that stimulus-evoked release of ACh is enhanced by muscarinic antagonists like atropine; the inhibitory neuronal muscarinic receptor involved is the M2 subtype, as in other peripheral cholinergic systems (5, 6). Therefore we evaluated the usefulness of HPLC-EC in guinea pig trachea by confirming the enhancing effect of atropine on stimulus-evoked release of ACh. MATERIALS
AND
METHODS
Tracheal smooth muscle. Male guinea pigs (n = 15) weighing 400-450 g (Simmonsen Laboratories, Gilroy, CA) were anesthetized with pentobarbital sodium (20-40 mg/kg ip), and clipconnected segments of trachea were prepared according to our modifications of the method of D’Agostino et al. (12). The trachea was removed and immersed in room-temperature Krebs-Henseleit buffer containing indomethacin (KHI j and bubbled continuously with 95% O,-5% CO,. Indomethacin was added to retard cyclooxygenase activity and to prevent the progressive decrease in the contractile response evoked by electrical field stimulation (EFS) (34). The trachea was cleaned of surrounding tissue and divided into five segments of six or seven cartilaginous rings each. Segments were cleaned of epithelium by rubbing with a pipe cleaner. The segments were placed side by side with the anterior side up and were connected together by small vascular clips (Hemoclip, Edward Week); clips were attached to the edge of the cartilaginous rings as close as possible to the posterior membrane. Sutures were looped through the two end segments, the cartilaginous portions of the three middle segments were removed, and the entire clip-connected trachea was mounted in the 2-ml organ bath. One end of the sutures was looped to the bottom of the bath and the other to a forcedisplacement transducer (Grass FT03) for measurement of tension. Tension was recorded on a polygraph (Grass model 7). The contents of the KHI buffer were (in mM) 118 NaCl, 74.6 KCl, 2.5 CaCl,, 2.4 MgS04, 1.1 KH2P04, 25.5 NaHCO,, IO glucose, and 0.01 indomethacin. A resting tension of 1 g was applied initially and readjusted as necessary during the incubation period. During this 90- to
ACH
RELEASE
L143
120-min period, the bath was maintained at 37°C bubbled continuously with 95% 02-5% C02, and emptied and refilled with fresh buffer every 15 min. ACh detection. ACh was measured by HPLC-EC with our modifications of the method first described by Potter et al. (28). The analytical column, immobilized enzyme reactor (IMER; Bioanalytical Systems, West Lafayette, IN), and electrochemical detector (EG&G Princeton Applied Research model 400 EC detector or Bioanalytical Systems LC-4C detector) were arranged as shown in Fig. IA. Mobile phase (50 mM Na,HPO,, pH 8.5) was pumped at a rate of 1 ml/min by a Rainin Rabbit high-performance solvent delivery system. The EC detector, fitted with a platinum electrode set at 500 mV, was connected to a Macintosh SE computer running a Rainin Dynamax HPLC Method Manager to acquire and analyze ACh peaks. With this technique, ACh, choline, and some choline analogues were retained by the analytical column (MF-6150) and were thereby separated from all other substances in the sample. After several minutes, the retained substances were eluted sequentially so that they were separated from each other. ACh was eluted and, in the IMER (MF-6151), converted first to choline by acetylcholinesterase and then to H,O, by choline oxidase. As H,O, flowed across the platinum electrode of the EC, it was oxidized, producing a current whose magnitude reflected the amount of ACh present in the original sample (Fig. 1B). Next, choline was eluted and converted in the IMER to H202, producing a separate peak whose magnitude reflected the amount of choline present in the sample. In some experiments a second IMER (MF-6153) was used. This second IMER, which contained catalase and choline oxidase, was placed before the analytical column; it removed choline before choline was retained by the analytical column. However, we saw no advantage to the use of this column, particularly because choline and ACh were easily distinguished, and we used the second IMER in two experiments only. Experimental procedures. After incubation of the clip-connected segment in KHI buffer, the baths were refilled with KHI buffer containing 1 ,um each of neostigmine, propranolol, and phentolamine, and incubation was continued for an additional 30 min. Neostigmine was present to prevent hydrolysis of ACh by endogenous acetylcholinesterase activity. Propranolol and phentolamine were present to inhibit any pre- or postsynaptic ,& and cw-adrenergic receptors. After its addition, each agent was present throughout the remainder of the experiment. At the end of this incubation period, the contents of the bath
Organ bath Fig. 1. A: High-performance liquid chromatography with electrochemical detection (HPLC-EC) method for detection of acetylcholine (ACh). Samples of buffer from organ bath are introduced through sample injector. In analytical column, ACh, choline, and some choline analogues are retained while all other substances are eluted, forming unretained peak of chromatogram shown in B. After several minutes, retained substances are eluted. In immobilized enzyme reactor (IMER), ACh is converted by acetylcholinesterase to choline and, subsequently, by choline oxidase to H202, which is oxidized at platinum electrode of electrochemical (EC) detector, producing ACh peak of chromatogram in B. Chromatogram produced by injecting a standard sample containing 10 pmol ACh in water is shown in B.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
L144
GUINEA
PIG
TRACHEAL
were emptied, the bath was refilled with buffer (same contents), and a rest period was begun (period R; n = 15 tracheas). After 30 min, the contents of the bath were analyzed for ACh. The bath was refilled with buffer, and EFS was applied for 30 (n = 8 tracheas) or 10 (n = 7) min, after which the contents of the bath were again analyzed for ACh (period S,; n = 15). Electrical impulses consisted of square-wave pulses (Winston Applied Electronics) having a duration of 0.5 ms and a frequency of 5 Hz. Current delivered to the bath was -400 mA and was monitored continuously (Tetronix single-beam storage oscilloscope). To examine the effect of atropine, we measured EFS-evoked release of ACh before and after addition of atropine (0.3 PM) to the buffer (group 1, n = 11 tracheas). Group 1 consisted of two subgroups: in group IA (n = 4), EFS was applied for 30 min (periods S, and S,), and atropine was always added after S,; in group 1B (n = 7), the period of stimulation was reduced to 10 min, but stimulation was performed a total of four times (periods S,-S,). In group 1 B, atropine was added after S, (n = 3) or after S, (n = 4). Group 2 (n = 4) served as controls for group IA; protocols were carried out identically except that vehicle, instead of atropine, was added after S,. In all protocols, 30 min was allowed between stimulations. To measure ACh in the organ bath, we emptied the bath, and the entire Z-ml volume was filtered (0.2 PM Acrodisc). Aliquots (1 ml) were introduced into the HPLC-EC through the injector, overfilling the 200~~1 sample loop. Subsequently, the sample was placed into the flow of the mobile phase. Amounts of ACh were determined by comparing areas of unknown curves with areas of standard curves; the latter were generated by injecting ACh standards (2-20 pm) before and after each experimental run. At the end of the experiment, the posterior membrane was separated from the cartilage and dissolved in NaOH, and protein content was determined by the method of Bradford (7). Data were expressed as picomoles per milligram of protein per minute, or as grams tension and were compared by Student’s paired t test (37). In addition, ratios of ACh release [Safter (S,)] for atropine-treated (group 1) and vehicle(wl%?fbre treated (group 2) preparations were calculated and compared by analysis of variance (37). The limit of detection was defined as that amount of ACh whose peak height was equal to three times the baseline noise. Retention time was defined as the time between injection of sample and the point of maximum concentration of the eluted peak. The following pharmacological agents were obtained from Sigma Chemical: acetylcholine hydrochloride, choline chloride, atropine sulfate, sodium phosphate dibasic, indomethacin, neostigmine bromide, and propranolol hydrochloride. Phentolamine (Regitine HCl) was obtained from Ciba-Geigy.
ACH
RELEASE
KHI buffer from the organ bath produced chromatograms with large choline peaks and relatively small ACh peaks. With EFS, the ACh peak increased in magnitude but remained small in comparison with the choline peak. Nevertheless, the choline peak was clearly separated from the ACh peak (Fig. 2). Mean quantity of ACh measured at the end of period R was 1.8 t 1.4 (SE) pmol mg protein-l min-I; in 12 of the experiments no ACh was measured. It should be noted that this almost complete absence of measured ACh during period R occurred even though the addition of neostigmine, propranolol, and phentolamine was associated with an increase in tension from 1.0 t 0.1 to 2.3 t 0.2 g (P < 0.05; n = 15). With EFS (period S,), tension rose to 3.5 t 0.2 g, and ACh release was evident in all 15 experiments, averaging 6.2 t 1.3 pmol mg protein-l min? l
l
l
l
Effect of atropine (0.3 PM) on EFS-evoked release of ACh. In group 1, tension fell to 0.6 t 0.1 g with addition of atropine to the organ bath. In group lA, EFS-evoked
release of ACh was 5.0 t 1.7 pmol mg protein-’ mine1 before atropine treatment (S,); with the addition of atropine (S,), EFS-evoked release increased to a mean of 18.2 t 5.1 pmolmg protein-l emin-l (P < 0.05). The ratio of the amount of ACh evoked by EFS after atropine (S,) to the amount before (Sb) for group IA was 3.6 t 0.4. In group U?, addition of atropine between periods S1 and S2 increased EFS-evoked release of ACh from 4.2 t 0.6 to 10.8 t 3.5 pmol mg protein-l mine1 with a mean S,/Sb of 2.4 t 0.4. Th e amount of ACh released remained at this elevated level throughout the rest of the experiment (during S3 and S,). When atropine was added between periods S3 and Sq, EFS-evoked release was unchanged with successive stimulations before atropine (i.e., S1 through S,), but after atropine addition (S,), EFS-evoked release again increased, from 5.0 t 0.5 to 15.0 t 1.7 pmol mg protein-l min- l (Fig. 3), with S,& equaling 3.1 t 0.6. In group 2 (n = 4), EFS-evoked release of ACh was 9.3 t 4.4 pmolmg protein-l. mine1 before and 10.7 t 4.0 pm01 mg protein-l min-l after vehicle treatment. The ratio of ACh evoked before and after vehicle treatment was 1.3 t 0.1; thus there was a small increase in evoked release with the second stimulation in the control group. In mouw 2 the increase in tension was the same before l
l
l
l
”
l
I
RESULTS
In preliminary experiments, the addition (in PM) of 10 indomethacin, 1 propranolol, 1 phentolamine, or 1 atropine to the Krebs-Henseleit buffer did not create retention peaks. For ACh and choline dissolved in water, retention peaks were generated at 4.6 (n = 40) and 6.0 (n = 8) min, respectively. When standards containing Z-20 pmol ACh were injected, both the area and the height of the ACh peaks were related linearily to the amount injected. Injection of progressively smaller amounts of ACh showed that the limit of detection was -1 pmo1/200 ~1 sample. Sensitivity depended significantly on the background noise of the particular experimental day and on the length of previous use of that particular enzyme column. When translated into molar concentrations in the organ bath, the lower limit of detection was ~5 nM. Suitability of our technique. After period R, samples of
Fig. ‘2. Chromatogram produced by injecting sample from an organ bath after 30-min electrical field stimulation (EFS). Very large choline (Ch) peak did not interfere with much smaller ACh peak. In this chromatogram (1 nA full scale) only lower portion of Ch peak is shown. Identity of ACh and Ch peaks was determined by comparing their retention times with retention times of known standards. Very small peak, seen here at 3 min (arrow), was not identified.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
ACH
RELEASE
L145
DISCUSSION
t
1
I
' 1 min
I
1
Fig. 3. Effect of atropine on EFS-evoked release of ACh. A: clip-connected tracheal segments were stimulated for 10 min at 5 Hz; a 30-min equilibration period was allowed between stimulations. Atropine (0.3 &I) was added after period S, (squares) or after period S3 (triangles) and allowed to remain in bath thereafter. ACh values are means & SE. B: portions of individual chromatograms generated in 1 experiment after periods S3 and S,. Two chromatograms were recorded-at same sensitivity (5 nA full scale), and recordings began 3 min after injection of a sample. In this experiment, atropine was added after S3. Also, a second IMER was present, which removed choline before it could be retained by analytical column, explaining absence of choline peaks in these chromatograms.
(1.1 g) and after (1.2 g) addition of the vehicle. The ratios S,& of EFS-evoked ACh release before and after atropine treatment or control are pooled and presented in Fig. 4.
atropine
vehicle
Fig. 4. Enhancing effect of atropine on EFS-evoked release of ACh. These histograms summarize ratios of ACh evoked before (S,) and after (Sb) atropine treatment (group I; n = 11 tracheas) or before and after vehicle treatment (group 2; n = 4). Tracheas were incubated in 0.3 PM atropine for 30 min before second EFS, and atropine remained in the bath throughout lo- or 30-min EFS. Values are means t SIZ and are taken from stimulation periods immediately before and after atropine addition. * Statistically significant (P < 0.05).
First, this study demonstrates that HPLC-EC is suitable for measurement of EFS-evoked release of ACh from nerve endings in guinea pig tracheal smooth muscle. We found that the lower limit of detection in the organ bath was 4 nM. It is interesting to note that we could not detect significant levels of ACh in period R, even though smooth muscle tone was elevated above initial set levels, presumably as a result of the inhibition by neostigmine of acetylcholinesterase activity and perhaps also its capacity to stimulate release of ACh (10). Thus, before application of an electrical field, the quantity of ACh in the neuromuscular junction was sufficient to increase tension significantly but was still below the level of detection by our technique. Second, this study shows that the muscarinic antagonist, atropine, enhances EFS-evoked release of ACh. In the central nervous system, as well as the enteric and other peripheral cholinergic nervous systems, EFS-induced release of ACh is under autoinhibitory feedback control, mediated by muscarinic receptors on cholinergic nerve terminals (5, 6, 20, 29, 36). In all of these systems, muscarinic antagonists enhance the release of ACh, probably by blocking this negative feedback control. We suggest that a similar mechanism explains our results. Atropine inhibited muscarinic receptors on cholinergic nerve terminals, prevented the negative feedback, and allowed ACh release to escape the normal autoregulatory mechanism. Our results, therefore, confirm conclusions of previous published studies. Of significance, however, is the relatively large increase in release of ACh that we report. In the study by Aas and Fonnum (1) in rat bronchial smooth muscle, scopolamine increased the potassium-evoked release of [3H]ACh by only 33%; in the study by D’Agostino et al. (12) in the guinea pig trachea, atropine facilitated release of ACh, increasing the EFS-evoked outflow of [3H]ACh by 39%. These relatively small increases contrast with the large increases, 310 and 250%, respectively, that we and Deckers et al. (14) report. In contrast to our technique and that of Deckers et al. (14), the measurement of ACh in both of the other studies employed radiolabeling where the presence of cholinesterase inhibitors was not required. Inherent in techniques used by Deckers and us is the requirement for addition of neostigmine to the bath. It is possible that the greater degree of autoinhibition in our preparation was due to the presence of neostigmine and the consequent elevated levels of ACh in the vicinity of the neuronal receptors. It has been suggested (17) that autoinhibitory mechanisms are not operative normally inasmuch as ACh is immediately hydrolyzed by acetylcholinesterase and is not able to act at presynaptic receptors. Our results give rise to the intriguing possibility that autoinhibition becomes significant only in the presence of high levels of ACh at the neuronal junction. If it is a brake on the release of ACh, the effects of autoinhibition are only seen in the presence of a process that causes increased levels of ACh, such as during acetylcholinesterase inhibition or perhaps disease. Support for this possibility comes from
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
L146
GUINEA
PIG
TRACHEAL
studies in the guinea pig myenteric plexus, where autoinhibitory mechanisms have been investigated in the presence and absence of another inhibitor of acetylcholinesterase, physostigmine (20). In the absence of physostigmine, muscarinic inhibition increased electrically evoked release of ACh by 37% compared with a ~200% increase in the presence of physostigmine (20). Thus autoinhibitory mechanisms may be functionally operative only in the presence of increased release, or decreased breakdown, of ACh. Likewise, the recently described lateral inhibitory mechanisms operating at adjacent release sites may be functional only in the presence of elevated levels of ACh (8). It is possible that a strong muscarinic autoinhibitory mechanism may interfere with the ability to detect regulatory actions of other inhibitory presynaptic receptors. Support for this possibility comes from studies in the myenteric plexus (35). In the absence of physostigmine, the ,&adrenergic agonist, isoproterenol, inhibited the evoked released of [3H]ACh, but in the presence of physostigmine and presumably high autoinhibitory tone the addition of isoproterenol failed to inhibit ACh release. This mechanism may account for the failure of investigators to detect any effect of isoproterenol on ACh release in canine trachea (23). Such interference by a strong muscarinic autoinhibitory system may suggest common signaling systems for different presynaptic receptors. In the rat cerebral cortex, for example, presynaptic muscarinic and opiate receptors both utilize 4aminopyridinesensitive K+ channels, whose activation hyperpolarizes the membrane, terminates calcium influx, and inhibits ACh release (33). Besides muscarinic, opiate, and adrenergic receptors, cholinergic nerve terminals in airway smooth muscle probably have receptors for other agonists. In a recent study, Deckers et al. (15) used HPLC-EC to provide evidence for inhibitory prostaglandin E2 receptors on cholinergic nerve terminals in dog bronchial smooth muscle. Their results are consistent with the interpretation of previous indirect studies (34). In contrast, however, Deckers et al. (15) found no evidence for presynaptic prostaglandin D2 receptors, whose existence has been postulated in previous studies with indirect techniques (32). This inconsistency serves to underscore the importance of direct measurements of ACh release. In conclusion, our data support previous suggestions that muscarinic receptors on cholinergic nerves innervating airway smooth muscle inhibit the release of ACh (24). The significance of these muscarinic inhibitory receptors awaits further investigation, and we suggest the possibility that autoinhibition may become functionally important only in the presence of abnormal amounts of ACh at the neuronal junction. Recent indirect evidence (2, 25) suggests that the muscarinic autoinhibitory mechanism is dysfunctional in asthma. Our description of a technique for the measurement of ACh release in guinea pig trachea provides an opportunity for studying presynaptic regulatory mechanisms in a variety of animal models of bronchial hyperresponsiveness. This work was supported by Tobacco-Related Disease Research Program 2 RT-265. Grant HL-27669 from the National Heart, Lung, and
ACH
RELEASE
Blood Institute, and the Research Service of the Department of Veterans Affairs. Address for reprint requests: D. G. Baker, Pulmonary and Critical Care Medicine (11 lD), Veterans Affairs Medical Center, 4150 Clement St., San Francisco, CA 94121. Received
24 February
1992; accepted
in final
form
5 May
1992.
REFERENCES 1. Aas, P., and F. Fonnum. Presynaptic inhibition of acetylcholine release. Acta Physiol. Stand. 127: 335-342, 1986. 2. Ayala, L. E., and T. Ahmed. Is there loss of a protective muscarinic receptor mechanism in asthma? Chest 96: 1285-1291, 1989. 3. Barnes, P. J. Muscarinic autoreceptors in airways. Their possible role in airway disease. Chest 96: 1220-1221, 1989. 4. Beani, L., C. Bianchi, A. Siniscalchi, L. Sivilotti, S. Tanganelli, and E. Veratti. Different approaches to study acetylcholine release: endogenous ACh versus tritium efflux. NuunynSchmiedeberg’s Arch. Pharmacol. 328: 119-126, 1984. 5. Bognar, I. T., B. Beinhauer, P. Kann, and H. Fuder. Different muscarinic receptors mediate autoinhibition of acetylcholine release and vagally-induced vasoconstriction in the rat isolated perfused heart. Naunyn-Schmiedeberg’s Arch. Pharmacol. 341: 279-287, 1990. 6. Bognar, I. T., M. T. Wesner, and H. Fuder. Muscarinic receptor types mediating autoinhibition of acetylcholine release and sphincter contraction in the guinea-pig iris. Nuunyn-Schmiedeberg’s Arch. Pharmacol. 341: 22-29, 1990. 7. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein-dye binding. Anal. Biochem. 72: 248-254, 1976. 8. Brock, J. A., T. C. Cunnane, K. Starke, and C. F. Wardell. cu2-Adrenoceptor-mediated autoinhibition of sympathetic transmitter release in guinea-pig vas deferens studied by intracellular and focal extracellular recording of junction potentials and currents. Naunyn-Schmiedeberg’s Arch. Pharmacol. 342: 45-52, 1990. 9. Buckner, C. K., V. Songsiridej, E. C. Dick, and W. W. Busse. In vivo and in vitro studies on the use of the guinea pig as a model for virus-provoked airway hyperreactivity. Am. Rev. Respir. Dis. 132: 305-310, 1985. 10. Carlyle, R. F. The mode of action of neostigmine and physostigmine on the guinea-pig trachealis muscle. Br. J. PharmacoZ. 21: 137-149, 1963. 11. Carlyle, R. F. The response of the guinea-pig isolated intact tracheal to transmural stimulation and the release of an acetylcholine-like substance under condition of rest and stimulation. Br. J. Pharmacol. 22: 126-136, 1964. 12. D’Agostino, G., M. C. Chiari, E. Grana, A. Subissi, and H. Kilbinger. Muscarinic inhibition of acetylcholine release from a novel in vitro preparation of the guinea-pig trachea. NuunynSchmiedeberg’s Arch. Pharmacol. 342: 141-145, 1990. 13. Deckers, I. A., and A. G. Herman. HPLC detection of acetylcholine released from canine airway segments (Abstract). Pfluegers Arch. 412: S42, 1988. 14. Deckers, I. A., M. R. Rampart, and A. 65. Herman. Evidence for presynaptic modulation of acetylcholine release in canine airways using a sensitive HPLC method (Abstract). Arch. Int. Pharmacodyn. Ther. 298: 297, 1989. I. A., M. Rampart, H. Bult, and A. G. Herman. 15. Deckers, Evidence for the involvement of prostaglandins in modulation of acetylcholine release from canine bronchial tissue. Eur. J. Pharmacol. 167: 415-418, 1988A9. 16, Faulkner, D., A. D. Fryer, and J. Maclagan. Postganglionic muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br. J. Pharmacol. 88: 181-187, 1986. 17. Hertting, G., S. Wurster, P. Gebicke-Harter, and C. Allgaier. Participation of regulatory G-proteins and protein kinase C in the modulation of transmitter release in hippocampus. In: Modulation of Synaptic Transmission and Plasticity in Nervous Systems, edited by G. Hertting, and H. C. Spatz. Berlin: SpringerVerlag, 1988, p. 147-164. 18. Ishida, K., L. J. Kelly, R. J. Thomson, L. L. Beattie, and R. R. Schellenberg. Repeated antigen challenge induces airway hyperresponsiveness with tissue eosinophilia in guinea pigs. J. Appl.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
GUINEA
PIG
TRACHEAL
Physiol. 67: 1133-1139, 1989. 19. Jacoby, D. B., and A. D. Fryer. Abnormalities in neural control of smooth muscle in virus-infected airways. Trends PharmacoZ. Sci. 11: 393-395, 1990. 20. Kilbinger, H., R. Kruel, and I. Wessler. Modulation by scopolamine, acetylcholine and choline of the evoked release of acetylcholine from the guinea pig myenteric plexus: evidence for a muscarinic feedback of acetylcholine secretion. In: Cholinergic Mechanisms, edited by G. Pepeu and H. Ladinsky. New York: Plenum, 1981, p. 169-175. 21. Kilbinger, H., R. Schneider, H. Siefken, D. Wolf, and G. D’Agostino. Characterization of prejunctional muscarinic autoreceptors in the guinea-pig trachea. Br. J. PharmucoZ. 103: 1757-1763, 1991. 22. Luz, S., I. Pinchasi, and D. M. Michaelson. Differential inhibition of the release of endogenous and newly synthesized acetylcholine from Torpedo synaptosomes by presynaptic muscarinic receptors. FEBS Lett. 164: 9-12, 1983. 23. Martin, J. G., and B. Collier. Acetylcholine release from canine isolated airway is not modulated by norepinephrine. J. Appl. PhysioZ. 61: 1025-1030, 1986. 24. Minette, P. A., and P. J. Barnes. Prejunctional inhibitory muscarinic receptors on cholinergic nerves in human and guinea pig airways. J. AppZ. Physiol. 64: 2532-2537, 1988. 25. Minette, P. A. H., J.-W. J. Lammers, C. M. S. Dixon, M. T. McCusker, and P. J. Barnes. A muscarinic agonist inhibits reflex bronchoconstriction in normal but not in asthmatic subjects. J. AppZ. Physiol. 67: 2461-2465, 1989. 26. Murlas, C. G., T. P. Murphy, and V. Chodimella. OS-induced mucosa-linked airway muscle hyperresponsiveness in the guinea pig. J. AppZ. Physiol. 69: 7-13, 1990. 27. Nishikawa, M., I. Hirotada, T. Fukuda, S. Suzuki, and T. Okubo. Acute exposure to cigarette smoke induces airway hyperresponsiveness without airway inflammation in guinea pigs. Am. Rev. Respir. Dis. 142: 177-183, 1990.
ACH
RELEASE
L147
28. Potter, P. E., J. L. Meek, and N. H. Neff. Acetylcholine and choline in neuronal tissue measured by HPLC with electrochemical detection. J. Neurochem. 41: 188-194, 1983. M., M. Marchi, G. Maura, and G. Bonanno. Pre29. Raiteri, synaptic regulation of acetylcholine release in the CNS. CeZZ BioZ. Int. Rep. 13: 1109-1118, 1989. 30. Shore, S., B. Collier, and J. G. Martin. Effect of endogenous prostaglandins on acetylcholine release from dog trachealis muscle. J. AppZ. Physiol. 62: 1837-1844, 1987. 31. Starke, K., M. Gothert, and H. Kilbinger. Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol. Rev. 69: 864-989, 1989. 32. Tamaoki, J., K. Sekizawa, P. D. Graf, and J. A. Nadel. Cholinergic neuromodulation by prostaglandin D2 in canine airway smooth muscle. J. AppZ. Physiol. 63: 1396-1400, 1987. 33 . Torocsik, A., and E. S. Vizi. 4-Aminopyridine interrupts the modulation of acetylcholine release mediated by muscarinic and opiate receptors. J. Neurosci. Res. 27: 228-232, 1993. E. H., P. M. O’Byrne, L. M. Fabbri, P. D. Graf, M. 34. Walters, J. Holtzman, and J. A. Nadel. Control of neurotransmission by prostaglandins in canine trachealis smooth muscle. J. AppZ. Physiol. 57: 129-134, 1984. 35. Wessler, I., V. Eschenbruch, S. Halim, and H. Kilbinger. Presynaptic effects of scopolamine, oxotremorine and morphine on [3H]acetylcholine release from the myenteric plexus at different stimulation frequencies and calcium concentrations. NuunynSchmiedeberg’s Arch. Phurmucol. 335: 597-604, 1987. 36. Wetzel, G. T., and J. H. Brown. Presynaptic modulation of acetylcholine release from cardiac parasympathetic neurons. Am. J. PhysioZ. 248 (Heart Circ. Physiol. 17): H33-H39, 1985. J. H. Biostatistical Analysis. Englewood Cliffs, NJ: Pren37. Zar, tice-Hall, 1974, p. 121-150.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (144.173.006.094) on January 9, 2019.
