Respiration Physiology, 87 (1992) 131-139
131
© 1992 Elsevier Science Publishers B.V. All rights reserved. 0034-5687/92/$03.50 RESP 01854
Effects of prostaglandin E 2 o n ganglionic transmission in the guinea pig trachea Sylvain DeLisle*, David Biggs**, Annabel Wang and James G. Martin Meala'ns-Christie Laboratories, McGiil University. Montreal, Quebec H2X 2P~, Canada
(Accepted 9 September 1991) Abstract. We studied the effects of prostaglandin E2 (PGE2) on the contractile responses of in viw guinea pig tracheal preparations with intact vagal innervation. The preparation was stimulated either through the vagal nerves (NS) or with an electrical field (EFS) and trachealis response was assessed from the pressure change inside the tracheal tube. Ganglionic blockade by hexamethonium inhibited responses to NS but did not affect EFS while both responses to NS and EFS were abolished by atropine or tetrodotoxin. This indicates that responses to both stimulation modalities were mediated by cholinerglc nerves but that NS involved a ganglionic relay whereas EFS did not. Within the frequency range of 0.1-20 Hz, there was a gradual increase in the pressure generated by the trachealis muscle with increasing frequency of stimulation. The frequency-response relationship was similar for NS and EFS. Thus, the ganglion does not appear to play an important filtering or amplifying role under those conditions. PGE2 (I-50mM) produced a concentration-dependent inhibition of NS and EFS without affecting responses to exogenous acetylcholine (ACh). This suggests that the main action of PGE~ is to reduce ACh release from post-ganglionic nerve terminals. PGE2 inhibited EFS to a larger extent than NS; we postulate a possible excitatory effect of PGE2 on neurotransmission in the airway ganglia.
Airway, tracheal muscle; Mammal, guinea pig; Prostnglandin, ganglionic transmission, trachea; Tracheal muscle, contractility, PGE2
Prostaglandins are released from airways or lung parenchyma either spontaneously or following physical, chemical or electrical stimuli (Yamaguchi et al., 1976; Steel et al., 1979; Schulman et ai., 1982; Grodzinska et ai., 1975; Orehek e~ ai., 1975; htoue et aL, 1984; Yen etal., 1976). One of their possible functions is to regulate autonomic neurotransmission. In trachealis muscle, prostaglandin E2 (PGE2) has little influence on the response to exogenous acetylcholine (ACh) at concentrations that inhibit
Correspondence to: J.G. Martin, Meakins-Christie Laboratories, McOill University, 3626 St. Urbain
Street, Montreal, Quebec H2X 2P2, Canada. * Present address: Howard Hughes Medical Institute and Department of Internal Medicine, University
of Iowa Hospitals and Clinics, Iowa City, IA 82242, U.S.A. ** Present address: Faculty of Pharmacy, University of Alberta, Edmonton, Alberta T6G 2N8, Canada.
132
S. DeLISLEetal.
responses to electrical field stimulation (EFS), a neurally mediated stimulus (Nakanishi et el., 1978; lnoue et al., 1984; Waiters et al., 1984; Jones et al., 1980). These results have generally been interpreted as indicating that the predominant effect of PGE 2 is to decrease ACh release from post-ganglionic nerves. An additional relaxant effect of PGE,_ on the muscle itself has been described (Shore et al., 1987). Little is known about the effect of prostaglandins on the airway parasympathetic ganglion, a structure suspected of playing a physiological role in modulating neurotransmission (Yip et aL, 1981; Mitchell et al., 1987). We hypothesized that an inhibition of ganglionic transmission by PGE2 could account for some of its bronchodilator action. To study ganglionic transmission we used a guinea pig tracheal preparation with its intact vagal innervation. In such a preparation, stimulation of the regal nerves involves serial activation of pre-ganglionic fibers, ganglia and post-ganglionic fibers whereas stimulation via an electrical field (EFS) acts by depolarizing post-ganglionic fibers and therefore bypasses airway ganglia. A comparison between these two modalities of stimulation in the same preparation provides a way to assess the effect of PGE 2 on ganglionic transmission.
Materials and methods Female, Hartley strain guinea pigs (350-650 g) (Charles River Inc., St. Constant, Quebec) were anesthetized with pentobarbital (50 mg/kg, i.p.). Isolated innervated tracheal preparations were obtained as follow,~.The trachea was exposed via a midline incision in the neck and both cervical vagi were isolated. After removal of 10-15 ml of blood through cardiac puncture, the incision was extended over the sternum. The sternum and upper rib cage were removed and the trachea and main carina were excised along with both vagi, carotid arteries, heart and aortic arch. The preparation was put in a Petri dish containing Krebs-Henseleit solution at room temperature, indomethacin and propranolol (both at 10-6 M). The tracheal tube with both vagi and recurrent laryngeal nerves intact was dissected free. The upper and lower ends of the trachea were ¢annulated with polyethylene tubing (PE-240), the preparation was mounted in a holder between two platinum plate electrodes and placed in a 25 mi bath containing a modified Krebs-Henseleit solution bubbled with 5~o CO= and 95% 02, at 37 °C. The tracheal lumen was perfused at I ml/min (Masterflex peristaltic pump, Cole Parmer). The upper tracheal cannula was attached via a valve to a saline-filled pressure transducer (Trantec Model 60-800). For measurements of tracheal pressure the valves were closed and perfusion was stopped; contractions of the trachealis muscle were recorded as intraluminalpressure changes. The pressure transducer was connected to a pen recorder (Hewlett-Packard 8805A preamplifier, 7785A recorder) for display of the signal, EFS was applied across the plate electrodes (Grass SO, stimulator) and both vagi were stimulated through a single platinum tunnel electrode. Settings on the stimulator were 5 ms duration pulses and 10 s trains for EFS; I ms duration pulses an¢~ 10 s trains Preparation.
PROSTAGLANDIN E2 AND NEUROTRAt:SMISSION IN GUINEA PIG TRACHEA 133
for nerve stimulation. Stimulation frequencies were 0.1, 0.5, 1, 5, and 20 Hz for both modalities. A supramaximal voltage setting of 25 V was used for NS. Since it was our purpose to test for a differential effect of PGE 2 on EFS vs NS, we matched the pressure responses to both stimulation modalities at 5 Hz. This was accomplished by adjusting the voltage setting for EFS to a value at or below 50 V. A frequency of 5 Hz was chosen because it gave reproducible pressure responses to repeated stimulations.
Experimentalprotocol.
After mounting, each preparation was allowed to equilibrate for 30 min. The tracheas were then stimulated every 2 rain using each ofthe four frequencies randomly and alternating between NS and EFS. The bathing solution was then changed and the first concentration of PGE2 added to the bath. After allowing 15 min for the maximal effect to take place (this time had been determined in preliminaryexperiments), the stimulation protocol was repeated. The same procedure was followed for each concentration of PGE2 (1, 5, 10 and 50 × 10 -9 M in ascending order). Control experiments were performed in order to verify the stability of the preparation over time using the same protocol with the exception that PGE2 was not added to the bath fluid between stimulations. To quantitate the magnitude of post-synaptic effects of PGE2 on trachealis, five consecutive cumulative dose-response curves to ACh (10 - 7-10 -4 M) were established on four control preparations and in four preparations before and after PGE2 (1, 5, 10 and 50 × 10-9 M). There was a 15 rain washout before addition of each concentration of PGE2.
Solutions and drugs. The Krebs solution contained (mM): NaCI, 118.3; KCI, 4.7; CaCI2' H20, 2.5; MgSO4.7H20, 1.2; KH2PO4, 1.2; NaHCO3, 25; dextrose, 5.5. It was bubbled with 95% 02/5% CO,, which maintained the pH at 7.4 at 37 °C. Drugs used were acetyicholine chloride, atropine sulphate, hexamethonium bromide, indomethacin, prostaglandin E2, and tetrodotoxin, all from Sigma Chemicals, St. Louis, MO.
Analysis ofdata. Data were expressed as mean + SE. Differences between means were compared using Student's t-test for paired data. Linear regression was performed by the method of least squares. Significance was assumed at the 5% level.
Results The ganglionic blocker hexamethonium (5 × I0 - 5 M) completely inhibited the pressure response to NS while leaving EFS unchanged, confirming that EFS acts through depolarization of post-ganglionic nerves, in contrast to NS which involves a ganglionic relay. Atropine (5 × I0-7 M) and tetrodotoxin (I0-6 M) blocked the increase in pressure induced by both EFS and NS. No non-cholinergic contractile response was observed.
134
S. DeLISLE et al. 60 50 40
i: l.) v
-i o) 1o
30
_.•____Controls PGE2 1 nM
20
__~___ PGE2 5 nM PGF..2 10 nM PGF..2 50 nM
it_
10 0
[Acetylcholine](pM) Fig. 1. Pressure responses elicited by ACh [(0.1-100) x 10-6 M] in the presence of increasing concentrations or PGE2. Each symbol represents the mean of four preparations. The vertical bars indicate one SE.
in control preparations (n ffi 4), five consecutive cumulative concentration-response curves to ACh were not significantly different from each other. Likewise, in the range of concentrations tested [(!-50) x 10 - 9 M], PGE2 did not significantly alter the cumulative concentration-response curves to ACh ( 1 0 - 7 - 1 0 -4 M; Fig. 1). There was no difference in the pattern of response of the POE2-treated specimens to the controls. The contractile response of each preparation increased linearly with the logarithm of stimulation frequency (Fig. 2). Correlation coefficients were 0.998 and 0.985 for EFS .'rod NS, respectively. The slope of the frequency-response relationship was 8.0 for EFS as compared to 8.9 for NS but this difference was not statistically significant. The effects of PGE 2 on increases in pressure evoked by NS and EFS at a stimulation frequency of 5 Hz are shown in Fig. 3A, B. When compared to their time-matched
"t
|I "t° 0~
1
0 0 Frequency,(Ha:)
Fig, 2. Pressure responses (average for all 18 experiments + $E) as a function of the logarithm of stimulation frequency for both stimulus modalities. Ten sec trains of impulses were used with settings of S msec pulses for EFS and I msec pulses for NS, The nominal voltage setting for NS was 25 V hut for EFS voltage was adjusted so as to match contractile responses at 5 Hz. Results for NS are indicated by the open circles. EFS values are shown in closed circles.
PROSTAGLANDIN
E2
AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 135
A
20-
A
0
1
10
100
[PGF.2] (nM) Time controls PGE2.exposN
B
i
0
ii ~ . . . . . ,
.......
1
I
10
.......
n
100
[PG~] (nM) Fig, 3. (A) Effect of[PGE2] on the pressure response to nerve stimulation at 5 Hz (filled circles). To control for the possibility of preparation deterioration over time, time-control experiments (open circles) were conducted using the same stimulation protocol without addition of PGE 2. The time axis is therefore common to control and PGE2-exposed experiments. Each point represents the average of nine experiments :1: SE. Note the stability of the preparation over 2.5 h. (B) Same conditions as shown in panel A but responses to electrical field stimulation (EFS) were measured,
controls (n = 9), the PGE2-exposed preparations (n = 9) showed a concentrationdependent inhibition starting at a concentration of 5 × 10-9 M. The effects of PGE2 were greater on responses to EFS than NS; for example, at 50 × l0 -9 M PGE2, the response to NS was inhibited by 65% whereas EFS was inhibited by 75~. Inhibition of the responses to l, 5 and 20 Hz EFS by PGE2 (5, 10, and 50 × 10 -9 M) was significantly greater than for NS. Qualitatively similar changes were observed for 0. l and 0.5 Hz but the responses were more variable and the changes did not reach statistical significance. The differences in the effects of PGE2 on responses to NS and EFS are more completely illustrated in Fig. 4 where the ratio of the response to NS and EFS is shown for all frequencies of stimulation at the different concentrations of PGE2. The ratio of responses to NS and EFS was close to I at the stimulation frequencies of 5 and 20 Hz, prior to exposure to PGE2. At lower stimulation frequencies the ratios ranged between 0.73 and 0.75. After exposure to PGE2, there was an increase in NS/EFS at all frequencies, reflecting the relatively greater inhibition of EFS than NS. At 50 × 10- 9 M PGE2 there was a fall in the NS/EFS ratio for stimulations at 0.1 Hz. However, the
136
S. DeLISLEet al. 1.8 1.6 1.4 ~u~ 1.2 ~'
J, t ---D----g---
1.0 0.8 0.6
(;
1
10
[pG~](rig)
.1 I'lz .5 HZ 1 Hz 5 Hz 20Hz
100
Fig. 4. Ratio of the response to N S over EFS as a function of [PGE2] for all the frequencies of stimulation studied. Within each individual experiment, the ratio of consecutive NS and EFS at a given [PGE2] was calculated. Each point represents the mean of nine experiments; error bars are omitted for the sake of clarity.
responses at this frequency were very small and more variable so that the significance of the finding is uncertain. All the PGE2-exposed preparations recovered to within 10% of their baseline responsiveness 45 rain after washing (not shown).
Discussion
in this isolated innervated guinea pig tracheal preparation, both NS and EFS responses were completely inhibited by tetrodotoxin or atropine. This indicates that the tracheal contractions elicited by both modalities of stimulation result from the release of ACh from post-ganglionic neurons. The ganglionic blocker hexamethonium did not influence the response to EFS but inhibited NS, confirming that NS involves a ganglionic relay whereas EFS does not. The tracheal tube preparation therefore is suited for the comparison of the effects of drugs on ganglia, nerve terminals of post-ganglionk neurons, and airway smooth muscle. The relationship between the pressure response of the preparation and the logarithm of the stimulation frequency was linear. No statistically significant difference was found between the slope of the frequency-response relationships ofthe two modalities, suggest= ing that the airway ganglia did not amplify or filter neural impulses, or at least if modulation of transmission occurred in the ganglia it was constant at frequencies of stimulation ranging from 0. I to 20 Hz. This is in agreement with data for the ferret ganglia where the contractile response was similar for both NS and EFS stimulation frequencies below 20-30 Hz (Skoogh, 1986). Our data do not exclude a filtering effect at higher frequencies as has been demonstrated by others (Skoogh, 1986; Yip et ai., 1981). The principal effect of PGE2 was to induce concentration-dependent inhibition of the pressure responses to both EFS and NS. At the concentrations studied, it had no effects on responses to ACh as judged by inspection of the cumulative concentration-response
PROSTAGLANDIN E2 AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 137
curves. Antagonism of exogenous ACh by PGE 2 has been demonstrated by Shore et aL (1987) in canine tracheal smooth muscle preparations. As the PGE2 concentrations used were similar to ours, interspecies differences may account for the observed discrepancies. The profound effect of PGE2 on responses to stimulation suggests that the antagonism of cholinergic responses by PGE2 is the result of decreased ACh release from post-ganglionic neurons and is unlikely to be accounted for by post-junctional actions. This is in accordance with previously reported data (Nakanishi et al., 1978; Inoue et al., 1984; Waiters et al., 1984; Jones et al., 1980). IfPGE 2 were to inhibit transmission through airway ganglia, PGE 2 should affect NS, which involves synaptic transmission in intramural ganglia, more than EFS. Our data show clearly that this is not the case. On the contrary, PGE2 produced a relatively greater inhibition of EFS than of NS. A similar pattern of response was observed at all frequencies of stimulation. However, a statistically significant difference in the degree of inhibition was present only at l, 5, and 20 Hz and not at lower frequencies. PGE2 therefore appears more efficient at decreasing ACh release from post-ganglionic nerve terminals stimulated by an electrical field than by activation in a more physiological manner through the application of current to the vagal nerve trunk itself. We think that the most likely explanation is that PGE2 enhances ganglionic neurotransmission. Although we are unaware of any previous reports of the effects of prostaglandins on airway ganglionic transmission, our results are in agreement with data regarding the gastrointestinal tract. In rat colon, PGE2 and lloprost (a stable prostacyclin analogue) increased chloride secretion by a mechanism dependent upon the presence of the submucosal plexus (Diener et aL, 1988), suggesting that they stimulated neural elements in tim plexus. More direct evidence stems from experiments in guinea pig ileum where both PGE, and PGE2 increased ACh output from myenteric ganglia (Kadlec et al., 1978; Yagasaki et al., 1981). In contrast to NS, EFS not only excites post-ganglionic parasympathetic fibers but presumably also induces sympathetic nerves to release noradrenaline. Because PGE2 enhances the smooth muscle response to noradrenaline, release of this neurotransmitter in the presence of PGE2 could potentially have resulted in an added relaxant effect with EFS and not with NS. However, we performed our experiments in the presence of a concentration ofpropranolol shown to be effective at blocking t-receptors in this system (Yip et al., 1981; Chesrown et ai., 1980; Coburn et al., 1973; Coleman et al., 1974; Richardson et al., 1975). It is also possible that EFS stimulates non-adrenergic, noncholinergic (NANC) inhibitory nerves (Yip et al., 1981; Chesrown et al., 1980; Coburn et ai., 1973; Coleman et ai., 1974; Richardson et ai., 1975) in a manner that differs from their stimulation by NS. The lack of specific blockers for the NANC nerves makes it difficult to test whether this possibility could explain our results. A weakness in the interpretation of our results is the lack of a direct demonstration of differences in the modulation of ACh release to EFS and NS by PGE2. Using a radioenzymatic technique to quantitate ACh release by field-stimulated preparations of dog trachealis, Shore et al. (1984) confirmed the ability of PGE2 to reduce ACh output from cholinergic terminals. However, the period of incubation with PGE 2 required to
138
S. DeLISLEetaL
reduce responses to EFS was shorter than that required for modulation of ACh output by PGE2. Therefore, our conclusions require confirmation by more direct experimental approaches. In summary, our data suggest that PGE2 has a dual effect on parasympathetic neurotransmission in guinea pig airways. First, it inhibits responses to electrical field and vagai nerve stimulation in a manner which seems to be best explained by postulating a decrease in ACh release from post-ganglionic nerve terminals. Second, the relatively smaller effect of PGE2 on responses to vagal nerve stimulation suggests that it may enhance neurotransmission at the level of the airway ganglia. The post-ganglionic effect clearly predominates both under our experimental conditions and when pharmacological concentrations of exogenous PGE2 are used in vivo. Although one can speculate that PGE2 participates in the fine tuning of resting airway muscle tone and may be implicated in disease processes (Barnes, 1986), the physiological relevance of its action at the airway ganglia remains to be determined.
Acknowledgements. Supported by the Medical Research Council of Canada, Grant No. 7852.
References Barnes, P.J. (1986). State of'art: Neural control of human airways in health and disease, Am. Ray. Respir, Dis. 134: 1289-1314, Chesrown, S, E,, C, S, Venugopalan, W, M. Gold and J, M. Drazen (1980), In vivo demonstration of nonadrenergic inhibitory innervations of'the guinea pig trachea, J. CIrri, Invest, 65: 314-320. Coburn, R, F, and T, Tomita (1973), Evidence for nonadrenergic inhibitory nerves in the guinea pig trachealis muscle, Am, J Phystol, 224: 1072-1080, Coleman, R, A, end G. P, Levy (1974), A non-adrenergic inhibitory nervous pathway in guinea pig trachea, Br, J, Pharmacol. 52: 167-174, Diener, M., R, J, Bridges, S. F, Knobloch and W. Rummel (1988). Neurally mediated and direct effects of prostaglandins on ion transport in rat colon descendens. Naunyn.$chmledeberf's Arch, PharmacoL 337: 74-78, Grodzinska, L., B. Panczenko and J. Gryglewski (1975), Generation of prostaglandin E-like material by guinea pig trachea contractility by histamine. J. Pharm. Pharmacol. 27: 88-91. lnoue, T,, Y, lto and K, Takeda (1984). Prostaglandin-induced inhibition of acetylcholine release from neuronal elements of dog tracheal tissue, J, Physiol, (Lined.) 349: 553-570. Jones, T,R,, J,T. Hamilton and N,M. Lefcoe (1980). Pharmacologic modulations of cholinergic neurotransmission in guinea pig trachea in v/tro, Can. J, Physiol. Pharmacol. 58: 810-822. Kadlcc, O., K. Masek and !, Seferna (1978). Modulation by prostaglandins of the release of acetylcholine and noradrenaline in guinea pig isolated lung, J. PharmacoL Exp. That. 205: 635-645. Mitchell, R,A,, D,A, Herbert, D,G. Baker and C,B, Basbaum (1987). In vim activity of tracheal parasympathetic ganglion cell innervating tracheal smooth muscle. Brain Res. 437: 157-160. Nakanishi, H,, H. Yashida and 1", Suzuki (1978), Inhibitory effects of prostaglandins on exitatory transmission in isolated canine tracheal muscle, Jap. J. Pharmacol. 28: 883-889. O rehek, J., J, S, Douglas and A. Bouhuys (1975). Contractile responses of the guinea pig trachea in vitro: modification by prostaglandin synthesis inhibiting drugs. J. Pharmacol. Exp. That. 194: 554-564. Richardson, J, B, and T, Bouchard (1975), Demonstration of a non-adrenergic inhibitory nervous system in the trachea of the guinea pig, J. Allergy Ciin. ImmunoL 56: 473-480.
PROSTAGLANDIN E2 AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 139 Schulman, E.S., N.F. Adkinson and A.H. Newball (1982). Cyclooxygenase metabolites in human lung anaphylaxis: Airway vs parenehyma. J. Appl. Physiol. 53: 589-595. Shore, S., B. Collier and J.G. Martin (1987). Effect of endogenous prostaglandins on acetylcholine release from dog trachealis muscle. J. Appl. Physiol. 62: 1837-1844. Skoogh, B.E. (1986). Parasympathetic ganglia in the airways. Bull. Eur. Physiopathol. Respir. 221 (Suppl. 7): 143-147. Steel, L., L. Platshon and M. Kaliner (1979). Prostaglandin generation by human and guinea pig lung tissue; comparison of parenchymal and airway responses. J. Allergy Ciin. lmmunol. 64: 28"/-293. Waiters, E. H., P.M. O'Byrne, L.M. Fabbri, P.D. Graf, M.J. Holtzman and J.A. Nadel (1984). Control of neurotransmission by prostaglandins in canine trachealis smooth muscle.J. Appl. Physiol. 57: ! 29-134. Yagasaki, O., M. Takai and !. Yanagiya (198 I). Acetylcholine release from the myenteric plexus of guinea pig ileum by prostaglandin E~. J. Pharm. Pharmacol. 33: 521-525. Yamaguchi, T., B. Hitzig and R.F. Coburn (1976). Endogenous prostaglandins and mechanical tension in canine trachealis muscle. Am. J. Physiol. 230: 1737-1743. Yen, S. S., A. A. Math~ and J.J. Dugan (1976). Release of prostaglandins from healthy and sensitized guinea pig lung and trachea by histamine. Prostaglandins ! 1: 228-239. Yip, P., B. Palombini and R.F. Coburn (1981). Inhibiting innervation to the guinea pig trachealis muscle. J. Appl. Physiol. 50: 374-382.
131
© 1992 Elsevier Science Publishers B.V. All rights reserved. 0034-5687/92/$03.50 RESP 01854
Effects of prostaglandin E 2 o n ganglionic transmission in the guinea pig trachea Sylvain DeLisle*, David Biggs**, Annabel Wang and James G. Martin Meala'ns-Christie Laboratories, McGiil University. Montreal, Quebec H2X 2P~, Canada
(Accepted 9 September 1991) Abstract. We studied the effects of prostaglandin E2 (PGE2) on the contractile responses of in viw guinea pig tracheal preparations with intact vagal innervation. The preparation was stimulated either through the vagal nerves (NS) or with an electrical field (EFS) and trachealis response was assessed from the pressure change inside the tracheal tube. Ganglionic blockade by hexamethonium inhibited responses to NS but did not affect EFS while both responses to NS and EFS were abolished by atropine or tetrodotoxin. This indicates that responses to both stimulation modalities were mediated by cholinerglc nerves but that NS involved a ganglionic relay whereas EFS did not. Within the frequency range of 0.1-20 Hz, there was a gradual increase in the pressure generated by the trachealis muscle with increasing frequency of stimulation. The frequency-response relationship was similar for NS and EFS. Thus, the ganglion does not appear to play an important filtering or amplifying role under those conditions. PGE2 (I-50mM) produced a concentration-dependent inhibition of NS and EFS without affecting responses to exogenous acetylcholine (ACh). This suggests that the main action of PGE~ is to reduce ACh release from post-ganglionic nerve terminals. PGE2 inhibited EFS to a larger extent than NS; we postulate a possible excitatory effect of PGE2 on neurotransmission in the airway ganglia.
Airway, tracheal muscle; Mammal, guinea pig; Prostnglandin, ganglionic transmission, trachea; Tracheal muscle, contractility, PGE2
Prostaglandins are released from airways or lung parenchyma either spontaneously or following physical, chemical or electrical stimuli (Yamaguchi et al., 1976; Steel et al., 1979; Schulman et ai., 1982; Grodzinska et ai., 1975; Orehek e~ ai., 1975; htoue et aL, 1984; Yen etal., 1976). One of their possible functions is to regulate autonomic neurotransmission. In trachealis muscle, prostaglandin E2 (PGE2) has little influence on the response to exogenous acetylcholine (ACh) at concentrations that inhibit
Correspondence to: J.G. Martin, Meakins-Christie Laboratories, McOill University, 3626 St. Urbain
Street, Montreal, Quebec H2X 2P2, Canada. * Present address: Howard Hughes Medical Institute and Department of Internal Medicine, University
of Iowa Hospitals and Clinics, Iowa City, IA 82242, U.S.A. ** Present address: Faculty of Pharmacy, University of Alberta, Edmonton, Alberta T6G 2N8, Canada.
132
S. DeLISLEetal.
responses to electrical field stimulation (EFS), a neurally mediated stimulus (Nakanishi et el., 1978; lnoue et al., 1984; Waiters et al., 1984; Jones et al., 1980). These results have generally been interpreted as indicating that the predominant effect of PGE 2 is to decrease ACh release from post-ganglionic nerves. An additional relaxant effect of PGE,_ on the muscle itself has been described (Shore et al., 1987). Little is known about the effect of prostaglandins on the airway parasympathetic ganglion, a structure suspected of playing a physiological role in modulating neurotransmission (Yip et aL, 1981; Mitchell et al., 1987). We hypothesized that an inhibition of ganglionic transmission by PGE2 could account for some of its bronchodilator action. To study ganglionic transmission we used a guinea pig tracheal preparation with its intact vagal innervation. In such a preparation, stimulation of the regal nerves involves serial activation of pre-ganglionic fibers, ganglia and post-ganglionic fibers whereas stimulation via an electrical field (EFS) acts by depolarizing post-ganglionic fibers and therefore bypasses airway ganglia. A comparison between these two modalities of stimulation in the same preparation provides a way to assess the effect of PGE 2 on ganglionic transmission.
Materials and methods Female, Hartley strain guinea pigs (350-650 g) (Charles River Inc., St. Constant, Quebec) were anesthetized with pentobarbital (50 mg/kg, i.p.). Isolated innervated tracheal preparations were obtained as follow,~.The trachea was exposed via a midline incision in the neck and both cervical vagi were isolated. After removal of 10-15 ml of blood through cardiac puncture, the incision was extended over the sternum. The sternum and upper rib cage were removed and the trachea and main carina were excised along with both vagi, carotid arteries, heart and aortic arch. The preparation was put in a Petri dish containing Krebs-Henseleit solution at room temperature, indomethacin and propranolol (both at 10-6 M). The tracheal tube with both vagi and recurrent laryngeal nerves intact was dissected free. The upper and lower ends of the trachea were ¢annulated with polyethylene tubing (PE-240), the preparation was mounted in a holder between two platinum plate electrodes and placed in a 25 mi bath containing a modified Krebs-Henseleit solution bubbled with 5~o CO= and 95% 02, at 37 °C. The tracheal lumen was perfused at I ml/min (Masterflex peristaltic pump, Cole Parmer). The upper tracheal cannula was attached via a valve to a saline-filled pressure transducer (Trantec Model 60-800). For measurements of tracheal pressure the valves were closed and perfusion was stopped; contractions of the trachealis muscle were recorded as intraluminalpressure changes. The pressure transducer was connected to a pen recorder (Hewlett-Packard 8805A preamplifier, 7785A recorder) for display of the signal, EFS was applied across the plate electrodes (Grass SO, stimulator) and both vagi were stimulated through a single platinum tunnel electrode. Settings on the stimulator were 5 ms duration pulses and 10 s trains for EFS; I ms duration pulses an¢~ 10 s trains Preparation.
PROSTAGLANDIN E2 AND NEUROTRAt:SMISSION IN GUINEA PIG TRACHEA 133
for nerve stimulation. Stimulation frequencies were 0.1, 0.5, 1, 5, and 20 Hz for both modalities. A supramaximal voltage setting of 25 V was used for NS. Since it was our purpose to test for a differential effect of PGE 2 on EFS vs NS, we matched the pressure responses to both stimulation modalities at 5 Hz. This was accomplished by adjusting the voltage setting for EFS to a value at or below 50 V. A frequency of 5 Hz was chosen because it gave reproducible pressure responses to repeated stimulations.
Experimentalprotocol.
After mounting, each preparation was allowed to equilibrate for 30 min. The tracheas were then stimulated every 2 rain using each ofthe four frequencies randomly and alternating between NS and EFS. The bathing solution was then changed and the first concentration of PGE2 added to the bath. After allowing 15 min for the maximal effect to take place (this time had been determined in preliminaryexperiments), the stimulation protocol was repeated. The same procedure was followed for each concentration of PGE2 (1, 5, 10 and 50 × 10 -9 M in ascending order). Control experiments were performed in order to verify the stability of the preparation over time using the same protocol with the exception that PGE2 was not added to the bath fluid between stimulations. To quantitate the magnitude of post-synaptic effects of PGE2 on trachealis, five consecutive cumulative dose-response curves to ACh (10 - 7-10 -4 M) were established on four control preparations and in four preparations before and after PGE2 (1, 5, 10 and 50 × 10-9 M). There was a 15 rain washout before addition of each concentration of PGE2.
Solutions and drugs. The Krebs solution contained (mM): NaCI, 118.3; KCI, 4.7; CaCI2' H20, 2.5; MgSO4.7H20, 1.2; KH2PO4, 1.2; NaHCO3, 25; dextrose, 5.5. It was bubbled with 95% 02/5% CO,, which maintained the pH at 7.4 at 37 °C. Drugs used were acetyicholine chloride, atropine sulphate, hexamethonium bromide, indomethacin, prostaglandin E2, and tetrodotoxin, all from Sigma Chemicals, St. Louis, MO.
Analysis ofdata. Data were expressed as mean + SE. Differences between means were compared using Student's t-test for paired data. Linear regression was performed by the method of least squares. Significance was assumed at the 5% level.
Results The ganglionic blocker hexamethonium (5 × I0 - 5 M) completely inhibited the pressure response to NS while leaving EFS unchanged, confirming that EFS acts through depolarization of post-ganglionic nerves, in contrast to NS which involves a ganglionic relay. Atropine (5 × I0-7 M) and tetrodotoxin (I0-6 M) blocked the increase in pressure induced by both EFS and NS. No non-cholinergic contractile response was observed.
134
S. DeLISLE et al. 60 50 40
i: l.) v
-i o) 1o
30
_.•____Controls PGE2 1 nM
20
__~___ PGE2 5 nM PGF..2 10 nM PGF..2 50 nM
it_
10 0
[Acetylcholine](pM) Fig. 1. Pressure responses elicited by ACh [(0.1-100) x 10-6 M] in the presence of increasing concentrations or PGE2. Each symbol represents the mean of four preparations. The vertical bars indicate one SE.
in control preparations (n ffi 4), five consecutive cumulative concentration-response curves to ACh were not significantly different from each other. Likewise, in the range of concentrations tested [(!-50) x 10 - 9 M], PGE2 did not significantly alter the cumulative concentration-response curves to ACh ( 1 0 - 7 - 1 0 -4 M; Fig. 1). There was no difference in the pattern of response of the POE2-treated specimens to the controls. The contractile response of each preparation increased linearly with the logarithm of stimulation frequency (Fig. 2). Correlation coefficients were 0.998 and 0.985 for EFS .'rod NS, respectively. The slope of the frequency-response relationship was 8.0 for EFS as compared to 8.9 for NS but this difference was not statistically significant. The effects of PGE 2 on increases in pressure evoked by NS and EFS at a stimulation frequency of 5 Hz are shown in Fig. 3A, B. When compared to their time-matched
"t
|I "t° 0~
1
0 0 Frequency,(Ha:)
Fig, 2. Pressure responses (average for all 18 experiments + $E) as a function of the logarithm of stimulation frequency for both stimulus modalities. Ten sec trains of impulses were used with settings of S msec pulses for EFS and I msec pulses for NS, The nominal voltage setting for NS was 25 V hut for EFS voltage was adjusted so as to match contractile responses at 5 Hz. Results for NS are indicated by the open circles. EFS values are shown in closed circles.
PROSTAGLANDIN
E2
AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 135
A
20-
A
0
1
10
100
[PGF.2] (nM) Time controls PGE2.exposN
B
i
0
ii ~ . . . . . ,
.......
1
I
10
.......
n
100
[PG~] (nM) Fig, 3. (A) Effect of[PGE2] on the pressure response to nerve stimulation at 5 Hz (filled circles). To control for the possibility of preparation deterioration over time, time-control experiments (open circles) were conducted using the same stimulation protocol without addition of PGE 2. The time axis is therefore common to control and PGE2-exposed experiments. Each point represents the average of nine experiments :1: SE. Note the stability of the preparation over 2.5 h. (B) Same conditions as shown in panel A but responses to electrical field stimulation (EFS) were measured,
controls (n = 9), the PGE2-exposed preparations (n = 9) showed a concentrationdependent inhibition starting at a concentration of 5 × 10-9 M. The effects of PGE2 were greater on responses to EFS than NS; for example, at 50 × l0 -9 M PGE2, the response to NS was inhibited by 65% whereas EFS was inhibited by 75~. Inhibition of the responses to l, 5 and 20 Hz EFS by PGE2 (5, 10, and 50 × 10 -9 M) was significantly greater than for NS. Qualitatively similar changes were observed for 0. l and 0.5 Hz but the responses were more variable and the changes did not reach statistical significance. The differences in the effects of PGE2 on responses to NS and EFS are more completely illustrated in Fig. 4 where the ratio of the response to NS and EFS is shown for all frequencies of stimulation at the different concentrations of PGE2. The ratio of responses to NS and EFS was close to I at the stimulation frequencies of 5 and 20 Hz, prior to exposure to PGE2. At lower stimulation frequencies the ratios ranged between 0.73 and 0.75. After exposure to PGE2, there was an increase in NS/EFS at all frequencies, reflecting the relatively greater inhibition of EFS than NS. At 50 × 10- 9 M PGE2 there was a fall in the NS/EFS ratio for stimulations at 0.1 Hz. However, the
136
S. DeLISLEet al. 1.8 1.6 1.4 ~u~ 1.2 ~'
J, t ---D----g---
1.0 0.8 0.6
(;
1
10
[pG~](rig)
.1 I'lz .5 HZ 1 Hz 5 Hz 20Hz
100
Fig. 4. Ratio of the response to N S over EFS as a function of [PGE2] for all the frequencies of stimulation studied. Within each individual experiment, the ratio of consecutive NS and EFS at a given [PGE2] was calculated. Each point represents the mean of nine experiments; error bars are omitted for the sake of clarity.
responses at this frequency were very small and more variable so that the significance of the finding is uncertain. All the PGE2-exposed preparations recovered to within 10% of their baseline responsiveness 45 rain after washing (not shown).
Discussion
in this isolated innervated guinea pig tracheal preparation, both NS and EFS responses were completely inhibited by tetrodotoxin or atropine. This indicates that the tracheal contractions elicited by both modalities of stimulation result from the release of ACh from post-ganglionic neurons. The ganglionic blocker hexamethonium did not influence the response to EFS but inhibited NS, confirming that NS involves a ganglionic relay whereas EFS does not. The tracheal tube preparation therefore is suited for the comparison of the effects of drugs on ganglia, nerve terminals of post-ganglionk neurons, and airway smooth muscle. The relationship between the pressure response of the preparation and the logarithm of the stimulation frequency was linear. No statistically significant difference was found between the slope of the frequency-response relationships ofthe two modalities, suggest= ing that the airway ganglia did not amplify or filter neural impulses, or at least if modulation of transmission occurred in the ganglia it was constant at frequencies of stimulation ranging from 0. I to 20 Hz. This is in agreement with data for the ferret ganglia where the contractile response was similar for both NS and EFS stimulation frequencies below 20-30 Hz (Skoogh, 1986). Our data do not exclude a filtering effect at higher frequencies as has been demonstrated by others (Skoogh, 1986; Yip et ai., 1981). The principal effect of PGE2 was to induce concentration-dependent inhibition of the pressure responses to both EFS and NS. At the concentrations studied, it had no effects on responses to ACh as judged by inspection of the cumulative concentration-response
PROSTAGLANDIN E2 AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 137
curves. Antagonism of exogenous ACh by PGE 2 has been demonstrated by Shore et aL (1987) in canine tracheal smooth muscle preparations. As the PGE2 concentrations used were similar to ours, interspecies differences may account for the observed discrepancies. The profound effect of PGE2 on responses to stimulation suggests that the antagonism of cholinergic responses by PGE2 is the result of decreased ACh release from post-ganglionic neurons and is unlikely to be accounted for by post-junctional actions. This is in accordance with previously reported data (Nakanishi et al., 1978; Inoue et al., 1984; Waiters et al., 1984; Jones et al., 1980). IfPGE 2 were to inhibit transmission through airway ganglia, PGE 2 should affect NS, which involves synaptic transmission in intramural ganglia, more than EFS. Our data show clearly that this is not the case. On the contrary, PGE2 produced a relatively greater inhibition of EFS than of NS. A similar pattern of response was observed at all frequencies of stimulation. However, a statistically significant difference in the degree of inhibition was present only at l, 5, and 20 Hz and not at lower frequencies. PGE2 therefore appears more efficient at decreasing ACh release from post-ganglionic nerve terminals stimulated by an electrical field than by activation in a more physiological manner through the application of current to the vagal nerve trunk itself. We think that the most likely explanation is that PGE2 enhances ganglionic neurotransmission. Although we are unaware of any previous reports of the effects of prostaglandins on airway ganglionic transmission, our results are in agreement with data regarding the gastrointestinal tract. In rat colon, PGE2 and lloprost (a stable prostacyclin analogue) increased chloride secretion by a mechanism dependent upon the presence of the submucosal plexus (Diener et aL, 1988), suggesting that they stimulated neural elements in tim plexus. More direct evidence stems from experiments in guinea pig ileum where both PGE, and PGE2 increased ACh output from myenteric ganglia (Kadlec et al., 1978; Yagasaki et al., 1981). In contrast to NS, EFS not only excites post-ganglionic parasympathetic fibers but presumably also induces sympathetic nerves to release noradrenaline. Because PGE2 enhances the smooth muscle response to noradrenaline, release of this neurotransmitter in the presence of PGE2 could potentially have resulted in an added relaxant effect with EFS and not with NS. However, we performed our experiments in the presence of a concentration ofpropranolol shown to be effective at blocking t-receptors in this system (Yip et al., 1981; Chesrown et ai., 1980; Coburn et al., 1973; Coleman et al., 1974; Richardson et al., 1975). It is also possible that EFS stimulates non-adrenergic, noncholinergic (NANC) inhibitory nerves (Yip et al., 1981; Chesrown et al., 1980; Coburn et ai., 1973; Coleman et ai., 1974; Richardson et ai., 1975) in a manner that differs from their stimulation by NS. The lack of specific blockers for the NANC nerves makes it difficult to test whether this possibility could explain our results. A weakness in the interpretation of our results is the lack of a direct demonstration of differences in the modulation of ACh release to EFS and NS by PGE2. Using a radioenzymatic technique to quantitate ACh release by field-stimulated preparations of dog trachealis, Shore et al. (1984) confirmed the ability of PGE2 to reduce ACh output from cholinergic terminals. However, the period of incubation with PGE 2 required to
138
S. DeLISLEetaL
reduce responses to EFS was shorter than that required for modulation of ACh output by PGE2. Therefore, our conclusions require confirmation by more direct experimental approaches. In summary, our data suggest that PGE2 has a dual effect on parasympathetic neurotransmission in guinea pig airways. First, it inhibits responses to electrical field and vagai nerve stimulation in a manner which seems to be best explained by postulating a decrease in ACh release from post-ganglionic nerve terminals. Second, the relatively smaller effect of PGE2 on responses to vagal nerve stimulation suggests that it may enhance neurotransmission at the level of the airway ganglia. The post-ganglionic effect clearly predominates both under our experimental conditions and when pharmacological concentrations of exogenous PGE2 are used in vivo. Although one can speculate that PGE2 participates in the fine tuning of resting airway muscle tone and may be implicated in disease processes (Barnes, 1986), the physiological relevance of its action at the airway ganglia remains to be determined.
Acknowledgements. Supported by the Medical Research Council of Canada, Grant No. 7852.
References Barnes, P.J. (1986). State of'art: Neural control of human airways in health and disease, Am. Ray. Respir, Dis. 134: 1289-1314, Chesrown, S, E,, C, S, Venugopalan, W, M. Gold and J, M. Drazen (1980), In vivo demonstration of nonadrenergic inhibitory innervations of'the guinea pig trachea, J. CIrri, Invest, 65: 314-320. Coburn, R, F, and T, Tomita (1973), Evidence for nonadrenergic inhibitory nerves in the guinea pig trachealis muscle, Am, J Phystol, 224: 1072-1080, Coleman, R, A, end G. P, Levy (1974), A non-adrenergic inhibitory nervous pathway in guinea pig trachea, Br, J, Pharmacol. 52: 167-174, Diener, M., R, J, Bridges, S. F, Knobloch and W. Rummel (1988). Neurally mediated and direct effects of prostaglandins on ion transport in rat colon descendens. Naunyn.$chmledeberf's Arch, PharmacoL 337: 74-78, Grodzinska, L., B. Panczenko and J. Gryglewski (1975), Generation of prostaglandin E-like material by guinea pig trachea contractility by histamine. J. Pharm. Pharmacol. 27: 88-91. lnoue, T,, Y, lto and K, Takeda (1984). Prostaglandin-induced inhibition of acetylcholine release from neuronal elements of dog tracheal tissue, J, Physiol, (Lined.) 349: 553-570. Jones, T,R,, J,T. Hamilton and N,M. Lefcoe (1980). Pharmacologic modulations of cholinergic neurotransmission in guinea pig trachea in v/tro, Can. J, Physiol. Pharmacol. 58: 810-822. Kadlcc, O., K. Masek and !, Seferna (1978). Modulation by prostaglandins of the release of acetylcholine and noradrenaline in guinea pig isolated lung, J. PharmacoL Exp. That. 205: 635-645. Mitchell, R,A,, D,A, Herbert, D,G. Baker and C,B, Basbaum (1987). In vim activity of tracheal parasympathetic ganglion cell innervating tracheal smooth muscle. Brain Res. 437: 157-160. Nakanishi, H,, H. Yashida and 1", Suzuki (1978), Inhibitory effects of prostaglandins on exitatory transmission in isolated canine tracheal muscle, Jap. J. Pharmacol. 28: 883-889. O rehek, J., J, S, Douglas and A. Bouhuys (1975). Contractile responses of the guinea pig trachea in vitro: modification by prostaglandin synthesis inhibiting drugs. J. Pharmacol. Exp. That. 194: 554-564. Richardson, J, B, and T, Bouchard (1975), Demonstration of a non-adrenergic inhibitory nervous system in the trachea of the guinea pig, J. Allergy Ciin. ImmunoL 56: 473-480.
PROSTAGLANDIN E2 AND NEUROTRANSMISSION IN GUINEA PIG TRACHEA 139 Schulman, E.S., N.F. Adkinson and A.H. Newball (1982). Cyclooxygenase metabolites in human lung anaphylaxis: Airway vs parenehyma. J. Appl. Physiol. 53: 589-595. Shore, S., B. Collier and J.G. Martin (1987). Effect of endogenous prostaglandins on acetylcholine release from dog trachealis muscle. J. Appl. Physiol. 62: 1837-1844. Skoogh, B.E. (1986). Parasympathetic ganglia in the airways. Bull. Eur. Physiopathol. Respir. 221 (Suppl. 7): 143-147. Steel, L., L. Platshon and M. Kaliner (1979). Prostaglandin generation by human and guinea pig lung tissue; comparison of parenchymal and airway responses. J. Allergy Ciin. lmmunol. 64: 28"/-293. Waiters, E. H., P.M. O'Byrne, L.M. Fabbri, P.D. Graf, M.J. Holtzman and J.A. Nadel (1984). Control of neurotransmission by prostaglandins in canine trachealis smooth muscle.J. Appl. Physiol. 57: ! 29-134. Yagasaki, O., M. Takai and !. Yanagiya (198 I). Acetylcholine release from the myenteric plexus of guinea pig ileum by prostaglandin E~. J. Pharm. Pharmacol. 33: 521-525. Yamaguchi, T., B. Hitzig and R.F. Coburn (1976). Endogenous prostaglandins and mechanical tension in canine trachealis muscle. Am. J. Physiol. 230: 1737-1743. Yen, S. S., A. A. Math~ and J.J. Dugan (1976). Release of prostaglandins from healthy and sensitized guinea pig lung and trachea by histamine. Prostaglandins ! 1: 228-239. Yip, P., B. Palombini and R.F. Coburn (1981). Inhibiting innervation to the guinea pig trachealis muscle. J. Appl. Physiol. 50: 374-382.
