Acta Physiol Scand 1991, 143, 405411
Sympathetic nerve stimulation influences mucociliary activity in the rabbit maxillary sinus A. C E R V I N , S. L I N D B E R G and U. M E R C K E Department of Oto-Rhino-Laryngology, University Hospital, S-221 85 Lund, Sweden.
CERVIN, A., LINDBERG, S. & MERCKE, U. 1991. Sympathetic nerve stimulation influences mucociliary activity in the rabbit maxillary sinus. Acta Physiol Scand 143, 405-411. Received 29 April 1991, accepted 18 July 1991. ISSN 0001-6772. Department of Oto-Rhino-Laryngology, University Hospital, Lund, Sweden. The effect of preganglionic sympathetic nerve stimulation on mucociliary activity in the rabbit maxillary sinus was investigated in vivo. Response to nerve stimulation was recorded photoelectrically and expressed as a percentage of the basal mucociliary activity prior to stimulation. Nerve stimulation (15 V, 5 ms) for 60 s at 2, 10 and 20 Hz stimulated mucociliary activity, the maximum increase being 21.1 1.3% at 20 Hz, an increase that pretreatment with the cholinergic antagonist atropine reduced to 14.5_+2.4%, suggesting that part of the response involves cholinergic mechanisms. Nerve stimulation (10 Hz) of animals pretreated with the p-adrenoceptor antagonist propranolol reversed the mucociliary response from an increase to a decrease ( - 10.6k 1.6%), indicating the involvement of /%receptors in the nerve-evoked increase. Pretreatment with the aadrenoceptor antagonist phentolamine had no effect on response to nerve stimulation. Rabbits given a combined atropine, propranolol and phentolamine blockade manifested decreased mucociliary activity in response to nerve stimulation (- 10.6f 2.1 %). Guanethidine pretreatment blocked the effect of nerve stimulation on mucociliary activity, including the observed decrease after combined blockade, indicating the effect to be mediated via sympathetic nerve fibres. The decrease in mucociliary activity in response to nerve stimulation after combined cholinergic-, p-, and a-adrenoceptor blockade suggests the presence of a nonadrenergic, non-cholinergic inhibitory mechanism. It is possible that this effect is mediated by release of neuropeptide Y, as intraarterial injections of neuropeptide Y reduce mucociliary activity in the rabbit maxillary sinus, and as neuropeptide Y is released in the upper airways upon sympathetic nerve stimulation. Key words :Atropine, guanethidine, maxillary sinus, mucociliary activity, neuropeptide Y, phentolamine, propranolol, rabbit and sympathetic nerve stimulation.
T h e nasal mucosa is richly supplied with sympathetic nerve fibres. These nerve fibres mainly surround smaller arteries but are also seen close to veins and seromucous glands (Dahlstrom & Fuxe 1965, Angglrd & Densert 1974, Lacroix et al. 1990). Sympathetic effects on the mucociliary system in the upper airways have been studied with pharmacological Correspondence : Anders Cervin, Department of Oto-Rhino-Laryngology, University Hospital, S221 85 Lund, Sweden.
methods. Noradrenaline (NA), the classic transmitter in the sympathetic nervous system, acts via a- and /3-adrenoceptors. In vitro experiments on the frog palatine mucosa have shown NA to induce a decrease in ciliary activity in concentrations below M, and to accelerate it in M (Maruyama concentrations above 2 x 1984). A recent in vivo investigation in the rabbit maxillary sinus have shown doses of Na M and above to stimulate mucociliary activity, whereas no effect is seen at lower doses (Cervin et al. unpublished observation). In the same
405
406
-4. Cewirr et
al.
experimental model, agonists with both x,- and q a d r e n e r g i c properties have been shown t o decrease mucociliary activity-, and p,-agonists t o stimulate it, whereas /J,-receptor agonists have n o mucociliary effect (Hybbinette & hlercke 1982b, Cervin et ill. 1988). I n addition to the established transmitter YA, a putative peptide transmitter has been reported t o co-exist with NA,with which it is co-released upon sympathetic nerve stimulation (Ekblad et a / . 1984, Lundberg et ( I / . 1984). This peptide, which was first described by T a t e m o t o et l i / . in 1982 and named neuropeptidc Y (NPY), is a potent vasoconstrictor and acts independently of adrenergic receptors ( L u n d b e r g et a / . 1982). I n z.iz.0 studies in the rabbit maxillary sinus have shown that intraarterial bolus injections of NPY depress mucociliary activit\- dose-dependently (Cerl-in et al. 1991). I n the nasal mucosa both of the cat a n d the pig, preganglionic sympathetic nerve stimulation produces a vasoconstrictive response partly resistant to adrenergic blockade, a response that has been attributed to the release of NPY, as the concentration of NPY in venous blood from the nasal mucosa increases upon ner\-e stimulation. T h i s release of NPY is particularly noticeable a t stimulation frequencies of 10 and 6.9 Hz respectively (Lundblad rt ril. 1987, T x r o i s et a / . 1989). K’hether preganglionic sympathetic n e r w stimulation has any effect on the mucociliary actil-it? of the upper respiratory tract is unknown. The present investigation was undertaken t o stud’ the effect of preganglionic sympathetic nervc stimulation on mucociliary activity, using a pre\iously developed animal model, and to examine whether transmitters other than X.4 might contribute to the effect of such nerve stimulation.
Carc of the animals followd the rules issued by the Suedish National Board of .igriculture and the experiments were approved by the Hoard’s animal research ethics committee. The experiments were performed on male and female rabbits, weighing 2 . g 3 . 2 kg, (mean 2.5 kg). The animals were anaesthetized with urethane, 1 g kg- h i . as an initial dose, with an extra dose of 0 . j g kg being given i.\-. during the operation. Iktails of anaesthetic and surgical techniques ha\ e
’
been published previously (Hybbinette & Mercke 1982a). .L\n i.v. cannula was inserted into one of the ear veins and perfused with saline at approximately 4 ml h and was used for the administration of the blocking substances. The superior cervical ganglion was identified at the medial side of the carotid hifurcation, the cervical trunk being dissected free from adjacent tissue, followed in a proximal direction, and cut approximately 2 em proximally to the ganglion. T h e cut nerve was kept moist with saline, and the nerve ending \vas placed on a bipolar platinum electrode. Special care was taken to cut possible connections nith the vagal nerve. Nerve stimulation --as performed with a Grass S48D stimulator (Quincy, .\lass. LS.4) connected to a Grass Stimulus Isolation Unit, SIL 5.A (Quincy, Mass. USA). Preganglionic nerve stimulations (1 5 V, -5 ms) were performed with single impulses and at a continuous frequency of 2, 10 or 20 Hz for 60 s. The intervals between stimulation periods were at least 15 min. The mucosa in the ipsilateral maxillarq- sinus was exposed through a trepanation ofabout 2 x 8 mm, which was immediately cos-ered w-ith a heated window, sealed to the bone with hone wax (Ethicon, Edinhurgh, UK). T o test nerve function, the effect of nerve stimulation (1.i V, 5 ms, 10 Hz for 30 s) on mucosal blood flow in the maxillary sinus was investigated with laser Doppler flow-metry (LDF) (Periflux Pf2, Perimed, SM-eden), both before and after the mucociliary experiments (Cervin eta/. 1988). In all animals, blood flow measured with LDF manifested a vasoconstrictive response within 10 s upon preganglionic synpathetic nerve stimulation (with the exception of the guanethidine pretreated animals), the LDF signal being reduced by approximately 70 Oo, but returning to the basal level within 3 min. Once nerve function had been established by the presence of a proper msoconstrictive response, the experimental recording of mucociliary activity could proceed. Blood pressure \\as measured in the femoral artery ( n = 9, of which 4 were pretreated with guanethidine) with a pressure transducer (Novatrans M X 807, hledex, England) and recorded on an inkwriter. Ilucociliary activity (visible as flickering light rcflections) was observed through a binocular microscope, the criterion of a functionally satisfactory preparation being visible transportation of small particles such as mucus clumps and debris cells. One of the eyepieces was switched to a phototransducer, and the mucociliary activity recorded photoelectrically (Xlercke et a / . 1974). T h e mucociliary wave pattern was monitored continuously on an oscilloscope, and recorded on an inkwritcr during challenges. T h e recordings were analysed by a computerized frequency calculator, which computed the mucociliary wave frequency in waves min every 10 s during challengcs and at intcrvals of 1-i min otherwise. Induced frequcnc! changes were expressed as percentages of
Nerve stimulation and ciliary activity the basal mucociliary wave frequency (frequency zero level) immediately preceding challenge. ECG and rectal temperature were monitored, and body temperature was maintained at 37.0"-38.5 "C by means of a heating pad. Respiratory rate was recorded from the thorax with a tocotransducer (152 78B, HewlettPackard, Germany). The following drugs were used: atropine (ACO, Sweden), propranolol (Inderal", ICI, UK), phentolamine (Regitin@, CIBA-GEIGY, Switzerland) and guanethidine (Ismelin@, CIBA-GEIGY, Switzerland). Atropine (a muscarinic receptor antagonist) contains atropine-sulphate in saline 2 mg ml-'. Inderala (an unselective P-antagonist) contains propranololhydrochloride 1 mg ml-' in sterile water, and Regitin" (an unselective a-antagonist) contains phentolaminemethanesulphonate 10 mg ml-' in sterile water. Guanethidine (a blocker of sympathetic neurotransmission) contains guanethidine-monosulphate 10 mg ml-l in sterile water. All injections were given in an ear vein at doses of 0.1 ml s -'followed by 1 ml saline to flush the cannula. The doses of atropine, propranolol, and phentolamine were chosen from those used in previous experiments in the same rabbit model (Hybbinette & Mercke 1982 b, c). The dose of guanethidine was in the same dose-range as a previous investigation in the rabbit (Rutter et al. 1987). Experimental procedures. Nerve stimulation (15 V, 5 ms, for 60 s) was not commenced until the mucociliary activity had remained stable for approximately 30 min. The following experimental series were run: (1) the effect of nerve stimulation was investigated in 11 animals at 2 and 10 Hz, and in seven rabbits at 20 Hz; (2) the effect of nerve stimulation (10 and 20 Hz) after pretreatment with atropine was investigated in eight and six animals, respectively, atropine (1 mg kg-') being injected intravenously 5 minutes prior to nerve stimulation; (3) the effect of nerve stimulation (10 Hz) after pretreatment with propranolol was investigated in 7 rabbits, propranolol (1 mg kg-') being given intravenously 10-30 min before nerve stimulation; (4) the effect of nerve stimulation (10 Hz) after pretreatment with phentolamine was investigated in 7 animals, phentolamine (1 mg kg-l) being injected intravenously 60 s prior to nerve stimulation; (5) the effect of nerve stimulation (10 Hz) after pretreatment with a combination of atropine, propranolol and phentolamine given at the above mentioned dosages and intervals was investigated in 9 rabbits, and (6) the effect of nerve stimulation (10 Hz) after guanethidine pretreatment was investigated in 5 animals, guanethidine (10 mg kg-') being injected intravenously at least 4 h before the beginning of the experiment. The same animals were then pretreated with a combined blockade of atropine, propranolol, and phentolamine
407
and the effect of nerve stimulation (10 Hz) was examined. Time-course curves were plotted from the change of mucociliary activity during the first 5 min after each challenge. Results were expressed as means and standard errors of the means (SE), with the exception of frequency zero levels which were expressed as means and standard deviations (SD). Peak responses and area under the curve were used for statistical evaluations, the basal level prior to stimulation serving as the control value (thus all rabbits served as their own controls). The basal level was defined as the spontaneous variation in the mucociliary activitj during a 5-min period prior to the start of the nerve stimulation experiments. The results were analysed with Student's t-test for paired and unpaired data, P-value < 0.05 being significant.
RESULTS Neither intravenous injections of saline and atropine, propranolol or phentolamine, nor all 3 drugs in combination had any effect on mucociliary activity or respiratory rate, whereas propranolol lowered the pulse rate from 298 7.2 to 205 k 5.4 beats min-' (n = 11, P < 0.001). Nerve stimulation at 10 and 20 Hz increased mucociliary activity, as compared to the basal level. T h e higher the stimulation frequency, the shorter the latency and the longer the duration of response, maximal response being seen at 20 Hz (Table 1). Latency was defined as the interval between challenge and the point at which the frequency had changed by 5 yo.T w o consecutive stimulations at 10 Hz with a 20-minute interval gave an identical mucociliary response to sympathetic nerve stimulation, 19.9 f3.3 % and 19.6f 5.1 yo respectively (n = 8). Sympathetic nerve stimulation did not affect respiratory rate, but in 4 rabbits there was a small increase in pulse rate (not significant). Nerve stimulation (10 Hz) resulted in a moderate increase of blood pressure in four out of five rabbits (8.413.506, n = 5), these changes were not significant compared to basal variation in blood pressure ( P = 0.2). T h e effects on blood pressure ceased 5-10 s after the nerve stimulation had stopped. I n the guanethidine pretreated animals nerve stimulation had no effect on blood pressure or pulse rate. Pretreatment with atropine decreased the acceleration of the mucociliary response to 20 Hz nerve stimulation from 2 1 . 2 k 1.5% to 14.5f 2.4%, n = 6, P < 0.05 (Fig. 1). At 10 Hz, the
408
.4. Cerrin et al.
Table 1. The effect of sympathetic nerve stimulation (1.5 V, 5 ms, 60 s) at different stimulation frequencies on mucociliary actii-it?..Frequency change and area under the curve are compared with basal level Stimulation frequencl(Hz)
N
2 10 20
Latency
Duration
P 0.5 0.1
The differences in nerve distribution to the ciliated epithelium may explain the shorter latency at the 20 Hz stimulation frequency, as cholinergic nerve fibres extend to the ciliated epithelium in the rabbit maxillary sinus whereas no adrenergic nerve fibres have been seen adjacent to the epithelium (Schindelmeiser et al. 1982). It is therefore possible that the sympathetic effects on mucociliary activity are mediated via released transmitters that must diffuse for some distance before reaching the ciliated cells, thus explaining the longer latency at the stimulation frequency of 10 Hz. However, as mucociliary response to sympathetic nerve stimulation was blocked by propranolol it would appear to be predominantly due to stimulation of ,&receptors. Moreover, adding atropine and phentolamine to the propranolol pretreatment did not further enhance the change of mucociliary activity produced after propranolol pretreatment alone, emphasizing that ,&receptors mediate the response to sympathetic nerve stimulation. Animals pretreated with propranolol responded to nerve stimulation with a decrease of mucociliary activity. This decrease may be mediated by the a-adrenergic effects of NA, as a-agonists are known to decrease mucociliary activity (Hybbinette & Mercke 1982 b, Cervin et al. 1988), and NA itself has in in vitro experiments been shown to inhibit mucociliary activity (Maruyama 1984). This explanation seems less likely, since in the present model, intra-arterial bolus injections of NA after propranolol pretreatment did not retard mucociliary activity (Cervin et al. unpublished observation). Moreover, pretreatment with the unselective a-agonist phentolamine did not affect the response of mucociliary activity to sympathetic nerve stimulation. The lack of effect by a-adrenergic blockade on the decrease of mucociliary activity observed after combined pretreatment with
410
A . Cerrin et al.
atropine and propranolol further supports the view that r-receptors d o not mediate the inhibition of mucociliarj- activity. I n the nerve stimulated animals, not only is NA released but also NPY (Lundblad et (11. 1987, Lacroix et af. 1989), which may explain the different response to nerve stimulation after propranolol blockade, as compared to exogenously administered NA given after propranolol a h e r e no retarding effect was seen on the mucociliary system. T h e combined cholinergic and adrenergic blockade produced a decrease of mucociliary activity after sympathetic nerve stimulation. This finding suggests the presence of a nonadrenergic, non-cholinergic mechanism, such as the release of NPY which has recently been shown to decrease mucociliary- activity in the rabbit maxillary sinus by approximately l j o 0 in rii.0 when injected intraarterially (Cervin et af. 1991). T h e release of NPY after sympathetic nerve stimulation has been found in the pig and in the cat nasal mucosa, where high concentrations of N P Y and NL4 were released into the afferent venous flow (Lundblad et af. 1087; Lacroix et ul. 1989). T h e stimulation frequency of 1OHz was used in the present investigation, as it has been shown to release NPY at the highest ratio rts-u-zis N.A (Pernow et a/. 1988). T h e relatively long duration (approximately 5 min) of the decrease is in accord with the prolonged vasoconstrictive response to nerve stimulation found in the pig nasal mucosa (Lacroix et ul. 1988). Moreover, SPY-like immunoreactivity in the venous blood from the pig spleen in ziro has been found to be increased between 2 and 5 min after the cessation of nerve stimulation (Lundberg et a/. 1989). Taken together with the present results, such findings suggest that the decrease of mucociliary activity in response to nerve stimulation after adrenergic and cholinergic blockade may be explained by release of XPY. I t might be speculated that the effects on mucociliary activity by sympathetic nerve stimulation are indirect, as nerve stimulation also affects blood flow and secretory glands in the respiratory mucosa. T h e state of the vascular bed is probably of less importance, as ligation of the supply artery reducing the blood flow to almost nil does not affect mucociliary activity in thc rabbit maxillary sinus (Cervin et a/. 1988). Nerve stimulation ma)- also affect the secretory glands in the respiratory mucosa, producing
changes in the amount and rheological properties of the mucus, thereby changing the mucociliary activity. Wilson & Yates (1978) thus found sympathetic nerve stimulation to have a moderate stimulatory effect on secretory glands in the cat nasal mucosa. I n contrast, neither N A challenges in the rabbit trachea (Melville et al. 1976) nor N P Y challenges in the ferret trachea (Webber 1988) were found to have any effect on amount and kiscosity of the mucus. Therefore, secretory effects of sympathetic nerve stimulation are probabl! of little importance to the mucociliary system. As the antagonists used in the present in\ estigation had no effect on mucociliary activity per se, it is unlikely that an effect on mucus production by the antagonists was in any way responsible for the present results. T h a t a reflex mechanism mediated by the vagal nerve might explain the change of mucociliary activity is unlikely, as care was taken to se\er all connection with the vagal nerve, and as there was no concomitant fall in blood pressure or pulse rate. T h e small increase of blood pressure during nerve stimulation was probably due to the release of NA. Guanethidine, which blocks sympathetic neurotransmission, inhibited the effect of nerve stimulation on blood pressure and mucociliary activity (not only the increase of mucociliary activity observed after sympathetic n e n e stimulation, but also the decrease seen after blockade of classic autonomic receptors was inhibited), confirming that the effects were mediated by sympathetic nerve fibres. T o sum up, the present findings support the notion of adrenergic acceleration of mucociliary acti\ ity mediated via P-receptors by preganglionic sympathetic nerve stimulation. At 20 Hz, the acceleration has a cholinergic component. O u r results also indicate the release of an agent inhibiting mucociliary activity, an agent which may be NPY. T h e effect attributed to N P Y was seen only after blockade of classic autonomic receptors. The technical assistance of Mrs Charlotte CervinHoberg and Mr Samuel Leino is gratefully acknowledged. This investigation was supported by the Swedish Medical Research Council (Projects nos. 17X-07940 and 17P-845l), the Torsten Snderberg and Ragnar Soderberg Foundation, the Tore Nilson Foundation, the Medical Faculty of the University of Lund, Forenade Liv Mutual Group Life Insurance Company, the Craford Foundation and The Swedish Society for Medical Research.
Nerve stimulation and ciliary activity
41 1
relation to sympathetic vasoconstriction resistant to a-adrenoceptor blockade. Acta Physiol Scand 116, 393402. ANGG~RD, A,, DENSERT, 0. 1974. Adrenergic inLUNDBERG, J.M., ANGGKRD, A,, THEODORSSONnervation of the nasal mucosa in cat. A histological NORHEIM, E., PERNOW, J. 1984. Guanethidineand physiological study. Acta Otolaryngol (Stockh) of' Neuropeptide Y-like immunosensitive release 78, 232-241. reactivity in the a t spleen by sympathetic nerve CERVIN, A,, BENDE,M., LINDBERG, S., MERCKE, U., stimulation. Neurorci Lett 52, 175-180. P. 1988. Relations between blood flow and LUNDBERG, OLSSON, J.M., RUDEHILL,A,, SOLLEVI, A. 1989. mucociliary activity in the rabbit maxillary sinus. Pharmacological characterization of neuropeptide Acta Otolaryngol (Stockh) 105, 350-356. Y and noradrenaline mechanisms in sympathetic CERVIN, A., LINDBERG, S., MERCKE, U . 1991. The control of pig spleen. Eur 3 Pharmacul, 163, effects of neuropeptide Y on mucociliary activity in 103-1 13. the rabbit maxillary sinus. Acta Otolayyngol LUNDBLAD, L., ANGGKRD, A., SARIA, A,, LUNDBERG, (Stockh) 111, 960-966. J.M. 1987. Neuropeptide Y and non-adrenergic DAHLSTROM, A,, FUXE,K. 1965. The adrenergic sympathetic vascular control of the cat nasal innervation of the nasal mucosa of certain mammals. mucosa. 3 Auton Nerv Syst 20, 189-197. MARUYAMA, I. 1984. Conflicting effects of nor.4cta Otolaryngol (Stockh) 59, 65-72. adrenaline on ciliary movement of frog palatine EKBLAD,E., EDVINSSON, L., WAHLESTEDT, C., mucosa. Eur 3 Pharmacol97, 239-245. UDDMAN, R., H~KANSON, R., SUNDLER, F. 1984. G.N., HORSTMANN, G., IRAVANI, J. 1976. Neuropeptide Y co-exists and co-operates with MELVILLE, Adrenergic compounds and the respiratory tract. A noradrenaline in perivascular nerve fibres. Regul physiological and electron-microscopical study. Pept 8, 225-235. Respiration 33, 261-269. HYBBINETTE, J.-C., MERCKE, U. 1982a. A method for MERCKE, U., H~KANSSON, C.H., TOREMALM, N.G. evaluating the effect of pharmacological substances A method for standardized studies of 1974. on mucociliary activity in vino. Acta Otolaryngol mucociliary activity. Acta Otolaryngol (Stockh) 78, (Stockh) 93, 151-159. 118-123. HYBBINETTE, J.-C., MERCKE, U. 1982b. Effects of PERNOW, J., KAHAN,T., LUNDBERG, J.M. 1988. sympathomimetic agonists and antagonists on Neuropeptide Y and reserpine-resistant vasomucociliary activity. Acta Otolaqmgol (Stockh) 94, constriction evoked by sympathetic nerve stimu121-130. lation in the dog skeletal muscle. Br 3 Pharmacol HYBBINETTE, J.-C., MERCKE, U. 1 9 8 2 ~Effects . of the 94, 952-960. parasympathomimetic drug metacholine and its RUTTER, P.C., POTOCNIK, S.J., LUDBROOK, J. 1987. antagonists atropine on mucociliary activity. Acta Sympathoadrenal mechanisms in cardiovascular Otolaryngol (Stockh) 93, 465473. responses to naloxone after hemorrhage. Am 3 LACROIX, J.S., STJARNE, P., ANGGKRD, A,, LUNDBERG, Physiol, 252, H40-46. J., ADDICKS,H.W., ADDICKS, K. J.M. 1988. Sympathetic vascular control of the pig SCHINDELMEISER, 1982. Innervation of the mucosa of rabbit maxillary nasal mucosa (2) : reserpine-resistant, nonsinus. 11. Demonstration of catecholamine fluoradrenergic nervous responses in relation to neuroescence and acetylcholinesterase activity. Acta peptide Y and ATP. Acta Physiol Scand 133, Otolaryngol (Stockh) 94, 531-536. 183-197. K., CARLQVIST, M., MUTT, V. 1982. LACROIS, J.S., STJARNE, P., ANGGKRD, A., LUNDBERG,TATEMOTO, Y - a novel brain peptide with Neuropeptide J.M. 1989. Sympathetic vascular control of the pig structural similarities to peptide YY and pancreatic nasal mucosa (111) : co-release of noradrenaline and polypeptide. Nature 296, 659-660. neuropeptide Y. Acta Physiol Scand 135, 17-28. WEBBER, S.E. 1988. The effects of peptide histidine .ACROIX, J.S., ANGGKRD, A,, HOKFELT, T., OHARE, isoleucine and neuropeptide Y on mucus volume M.M.T., FAHRENKRUG, J., LUNDBERG, J.M. 1990. output from the ferret trachea. B r J Pharmacul95, Neuropeptide Y: presence in sympathetic and 49-54. parasympathetic innervation of the nasal mucosa. WILSON, H., YATES,M.S. 1978. Sympathetic nerves Cell Tissue Res 259, 119-128. and nasal secretion in the cat. Acta Otuluryngol ~ N D B E R G ,J.M., TATEMOTO, K. 1982. Pancreatic (Stockh) 85, 426430. polypeptide family (APP, BPP, NPY, and PPY) in
REFERENCES
Sympathetic nerve stimulation influences mucociliary activity in the rabbit maxillary sinus A. C E R V I N , S. L I N D B E R G and U. M E R C K E Department of Oto-Rhino-Laryngology, University Hospital, S-221 85 Lund, Sweden.
CERVIN, A., LINDBERG, S. & MERCKE, U. 1991. Sympathetic nerve stimulation influences mucociliary activity in the rabbit maxillary sinus. Acta Physiol Scand 143, 405-411. Received 29 April 1991, accepted 18 July 1991. ISSN 0001-6772. Department of Oto-Rhino-Laryngology, University Hospital, Lund, Sweden. The effect of preganglionic sympathetic nerve stimulation on mucociliary activity in the rabbit maxillary sinus was investigated in vivo. Response to nerve stimulation was recorded photoelectrically and expressed as a percentage of the basal mucociliary activity prior to stimulation. Nerve stimulation (15 V, 5 ms) for 60 s at 2, 10 and 20 Hz stimulated mucociliary activity, the maximum increase being 21.1 1.3% at 20 Hz, an increase that pretreatment with the cholinergic antagonist atropine reduced to 14.5_+2.4%, suggesting that part of the response involves cholinergic mechanisms. Nerve stimulation (10 Hz) of animals pretreated with the p-adrenoceptor antagonist propranolol reversed the mucociliary response from an increase to a decrease ( - 10.6k 1.6%), indicating the involvement of /%receptors in the nerve-evoked increase. Pretreatment with the aadrenoceptor antagonist phentolamine had no effect on response to nerve stimulation. Rabbits given a combined atropine, propranolol and phentolamine blockade manifested decreased mucociliary activity in response to nerve stimulation (- 10.6f 2.1 %). Guanethidine pretreatment blocked the effect of nerve stimulation on mucociliary activity, including the observed decrease after combined blockade, indicating the effect to be mediated via sympathetic nerve fibres. The decrease in mucociliary activity in response to nerve stimulation after combined cholinergic-, p-, and a-adrenoceptor blockade suggests the presence of a nonadrenergic, non-cholinergic inhibitory mechanism. It is possible that this effect is mediated by release of neuropeptide Y, as intraarterial injections of neuropeptide Y reduce mucociliary activity in the rabbit maxillary sinus, and as neuropeptide Y is released in the upper airways upon sympathetic nerve stimulation. Key words :Atropine, guanethidine, maxillary sinus, mucociliary activity, neuropeptide Y, phentolamine, propranolol, rabbit and sympathetic nerve stimulation.
T h e nasal mucosa is richly supplied with sympathetic nerve fibres. These nerve fibres mainly surround smaller arteries but are also seen close to veins and seromucous glands (Dahlstrom & Fuxe 1965, Angglrd & Densert 1974, Lacroix et al. 1990). Sympathetic effects on the mucociliary system in the upper airways have been studied with pharmacological Correspondence : Anders Cervin, Department of Oto-Rhino-Laryngology, University Hospital, S221 85 Lund, Sweden.
methods. Noradrenaline (NA), the classic transmitter in the sympathetic nervous system, acts via a- and /3-adrenoceptors. In vitro experiments on the frog palatine mucosa have shown NA to induce a decrease in ciliary activity in concentrations below M, and to accelerate it in M (Maruyama concentrations above 2 x 1984). A recent in vivo investigation in the rabbit maxillary sinus have shown doses of Na M and above to stimulate mucociliary activity, whereas no effect is seen at lower doses (Cervin et al. unpublished observation). In the same
405
406
-4. Cewirr et
al.
experimental model, agonists with both x,- and q a d r e n e r g i c properties have been shown t o decrease mucociliary activity-, and p,-agonists t o stimulate it, whereas /J,-receptor agonists have n o mucociliary effect (Hybbinette & hlercke 1982b, Cervin et ill. 1988). I n addition to the established transmitter YA, a putative peptide transmitter has been reported t o co-exist with NA,with which it is co-released upon sympathetic nerve stimulation (Ekblad et a / . 1984, Lundberg et ( I / . 1984). This peptide, which was first described by T a t e m o t o et l i / . in 1982 and named neuropeptidc Y (NPY), is a potent vasoconstrictor and acts independently of adrenergic receptors ( L u n d b e r g et a / . 1982). I n z.iz.0 studies in the rabbit maxillary sinus have shown that intraarterial bolus injections of NPY depress mucociliary activit\- dose-dependently (Cerl-in et al. 1991). I n the nasal mucosa both of the cat a n d the pig, preganglionic sympathetic nerve stimulation produces a vasoconstrictive response partly resistant to adrenergic blockade, a response that has been attributed to the release of NPY, as the concentration of NPY in venous blood from the nasal mucosa increases upon ner\-e stimulation. T h i s release of NPY is particularly noticeable a t stimulation frequencies of 10 and 6.9 Hz respectively (Lundblad rt ril. 1987, T x r o i s et a / . 1989). K’hether preganglionic sympathetic n e r w stimulation has any effect on the mucociliary actil-it? of the upper respiratory tract is unknown. The present investigation was undertaken t o stud’ the effect of preganglionic sympathetic nervc stimulation on mucociliary activity, using a pre\iously developed animal model, and to examine whether transmitters other than X.4 might contribute to the effect of such nerve stimulation.
Carc of the animals followd the rules issued by the Suedish National Board of .igriculture and the experiments were approved by the Hoard’s animal research ethics committee. The experiments were performed on male and female rabbits, weighing 2 . g 3 . 2 kg, (mean 2.5 kg). The animals were anaesthetized with urethane, 1 g kg- h i . as an initial dose, with an extra dose of 0 . j g kg being given i.\-. during the operation. Iktails of anaesthetic and surgical techniques ha\ e
’
been published previously (Hybbinette & Mercke 1982a). .L\n i.v. cannula was inserted into one of the ear veins and perfused with saline at approximately 4 ml h and was used for the administration of the blocking substances. The superior cervical ganglion was identified at the medial side of the carotid hifurcation, the cervical trunk being dissected free from adjacent tissue, followed in a proximal direction, and cut approximately 2 em proximally to the ganglion. T h e cut nerve was kept moist with saline, and the nerve ending \vas placed on a bipolar platinum electrode. Special care was taken to cut possible connections nith the vagal nerve. Nerve stimulation --as performed with a Grass S48D stimulator (Quincy, .\lass. LS.4) connected to a Grass Stimulus Isolation Unit, SIL 5.A (Quincy, Mass. USA). Preganglionic nerve stimulations (1 5 V, -5 ms) were performed with single impulses and at a continuous frequency of 2, 10 or 20 Hz for 60 s. The intervals between stimulation periods were at least 15 min. The mucosa in the ipsilateral maxillarq- sinus was exposed through a trepanation ofabout 2 x 8 mm, which was immediately cos-ered w-ith a heated window, sealed to the bone with hone wax (Ethicon, Edinhurgh, UK). T o test nerve function, the effect of nerve stimulation (1.i V, 5 ms, 10 Hz for 30 s) on mucosal blood flow in the maxillary sinus was investigated with laser Doppler flow-metry (LDF) (Periflux Pf2, Perimed, SM-eden), both before and after the mucociliary experiments (Cervin eta/. 1988). In all animals, blood flow measured with LDF manifested a vasoconstrictive response within 10 s upon preganglionic synpathetic nerve stimulation (with the exception of the guanethidine pretreated animals), the LDF signal being reduced by approximately 70 Oo, but returning to the basal level within 3 min. Once nerve function had been established by the presence of a proper msoconstrictive response, the experimental recording of mucociliary activity could proceed. Blood pressure \\as measured in the femoral artery ( n = 9, of which 4 were pretreated with guanethidine) with a pressure transducer (Novatrans M X 807, hledex, England) and recorded on an inkwriter. Ilucociliary activity (visible as flickering light rcflections) was observed through a binocular microscope, the criterion of a functionally satisfactory preparation being visible transportation of small particles such as mucus clumps and debris cells. One of the eyepieces was switched to a phototransducer, and the mucociliary activity recorded photoelectrically (Xlercke et a / . 1974). T h e mucociliary wave pattern was monitored continuously on an oscilloscope, and recorded on an inkwritcr during challenges. T h e recordings were analysed by a computerized frequency calculator, which computed the mucociliary wave frequency in waves min every 10 s during challengcs and at intcrvals of 1-i min otherwise. Induced frequcnc! changes were expressed as percentages of
Nerve stimulation and ciliary activity the basal mucociliary wave frequency (frequency zero level) immediately preceding challenge. ECG and rectal temperature were monitored, and body temperature was maintained at 37.0"-38.5 "C by means of a heating pad. Respiratory rate was recorded from the thorax with a tocotransducer (152 78B, HewlettPackard, Germany). The following drugs were used: atropine (ACO, Sweden), propranolol (Inderal", ICI, UK), phentolamine (Regitin@, CIBA-GEIGY, Switzerland) and guanethidine (Ismelin@, CIBA-GEIGY, Switzerland). Atropine (a muscarinic receptor antagonist) contains atropine-sulphate in saline 2 mg ml-'. Inderala (an unselective P-antagonist) contains propranololhydrochloride 1 mg ml-' in sterile water, and Regitin" (an unselective a-antagonist) contains phentolaminemethanesulphonate 10 mg ml-' in sterile water. Guanethidine (a blocker of sympathetic neurotransmission) contains guanethidine-monosulphate 10 mg ml-l in sterile water. All injections were given in an ear vein at doses of 0.1 ml s -'followed by 1 ml saline to flush the cannula. The doses of atropine, propranolol, and phentolamine were chosen from those used in previous experiments in the same rabbit model (Hybbinette & Mercke 1982 b, c). The dose of guanethidine was in the same dose-range as a previous investigation in the rabbit (Rutter et al. 1987). Experimental procedures. Nerve stimulation (15 V, 5 ms, for 60 s) was not commenced until the mucociliary activity had remained stable for approximately 30 min. The following experimental series were run: (1) the effect of nerve stimulation was investigated in 11 animals at 2 and 10 Hz, and in seven rabbits at 20 Hz; (2) the effect of nerve stimulation (10 and 20 Hz) after pretreatment with atropine was investigated in eight and six animals, respectively, atropine (1 mg kg-') being injected intravenously 5 minutes prior to nerve stimulation; (3) the effect of nerve stimulation (10 Hz) after pretreatment with propranolol was investigated in 7 rabbits, propranolol (1 mg kg-') being given intravenously 10-30 min before nerve stimulation; (4) the effect of nerve stimulation (10 Hz) after pretreatment with phentolamine was investigated in 7 animals, phentolamine (1 mg kg-l) being injected intravenously 60 s prior to nerve stimulation; (5) the effect of nerve stimulation (10 Hz) after pretreatment with a combination of atropine, propranolol and phentolamine given at the above mentioned dosages and intervals was investigated in 9 rabbits, and (6) the effect of nerve stimulation (10 Hz) after guanethidine pretreatment was investigated in 5 animals, guanethidine (10 mg kg-') being injected intravenously at least 4 h before the beginning of the experiment. The same animals were then pretreated with a combined blockade of atropine, propranolol, and phentolamine
407
and the effect of nerve stimulation (10 Hz) was examined. Time-course curves were plotted from the change of mucociliary activity during the first 5 min after each challenge. Results were expressed as means and standard errors of the means (SE), with the exception of frequency zero levels which were expressed as means and standard deviations (SD). Peak responses and area under the curve were used for statistical evaluations, the basal level prior to stimulation serving as the control value (thus all rabbits served as their own controls). The basal level was defined as the spontaneous variation in the mucociliary activitj during a 5-min period prior to the start of the nerve stimulation experiments. The results were analysed with Student's t-test for paired and unpaired data, P-value < 0.05 being significant.
RESULTS Neither intravenous injections of saline and atropine, propranolol or phentolamine, nor all 3 drugs in combination had any effect on mucociliary activity or respiratory rate, whereas propranolol lowered the pulse rate from 298 7.2 to 205 k 5.4 beats min-' (n = 11, P < 0.001). Nerve stimulation at 10 and 20 Hz increased mucociliary activity, as compared to the basal level. T h e higher the stimulation frequency, the shorter the latency and the longer the duration of response, maximal response being seen at 20 Hz (Table 1). Latency was defined as the interval between challenge and the point at which the frequency had changed by 5 yo.T w o consecutive stimulations at 10 Hz with a 20-minute interval gave an identical mucociliary response to sympathetic nerve stimulation, 19.9 f3.3 % and 19.6f 5.1 yo respectively (n = 8). Sympathetic nerve stimulation did not affect respiratory rate, but in 4 rabbits there was a small increase in pulse rate (not significant). Nerve stimulation (10 Hz) resulted in a moderate increase of blood pressure in four out of five rabbits (8.413.506, n = 5), these changes were not significant compared to basal variation in blood pressure ( P = 0.2). T h e effects on blood pressure ceased 5-10 s after the nerve stimulation had stopped. I n the guanethidine pretreated animals nerve stimulation had no effect on blood pressure or pulse rate. Pretreatment with atropine decreased the acceleration of the mucociliary response to 20 Hz nerve stimulation from 2 1 . 2 k 1.5% to 14.5f 2.4%, n = 6, P < 0.05 (Fig. 1). At 10 Hz, the
408
.4. Cerrin et al.
Table 1. The effect of sympathetic nerve stimulation (1.5 V, 5 ms, 60 s) at different stimulation frequencies on mucociliary actii-it?..Frequency change and area under the curve are compared with basal level Stimulation frequencl(Hz)
N
2 10 20
Latency
Duration
P 0.5 0.1
The differences in nerve distribution to the ciliated epithelium may explain the shorter latency at the 20 Hz stimulation frequency, as cholinergic nerve fibres extend to the ciliated epithelium in the rabbit maxillary sinus whereas no adrenergic nerve fibres have been seen adjacent to the epithelium (Schindelmeiser et al. 1982). It is therefore possible that the sympathetic effects on mucociliary activity are mediated via released transmitters that must diffuse for some distance before reaching the ciliated cells, thus explaining the longer latency at the stimulation frequency of 10 Hz. However, as mucociliary response to sympathetic nerve stimulation was blocked by propranolol it would appear to be predominantly due to stimulation of ,&receptors. Moreover, adding atropine and phentolamine to the propranolol pretreatment did not further enhance the change of mucociliary activity produced after propranolol pretreatment alone, emphasizing that ,&receptors mediate the response to sympathetic nerve stimulation. Animals pretreated with propranolol responded to nerve stimulation with a decrease of mucociliary activity. This decrease may be mediated by the a-adrenergic effects of NA, as a-agonists are known to decrease mucociliary activity (Hybbinette & Mercke 1982 b, Cervin et al. 1988), and NA itself has in in vitro experiments been shown to inhibit mucociliary activity (Maruyama 1984). This explanation seems less likely, since in the present model, intra-arterial bolus injections of NA after propranolol pretreatment did not retard mucociliary activity (Cervin et al. unpublished observation). Moreover, pretreatment with the unselective a-agonist phentolamine did not affect the response of mucociliary activity to sympathetic nerve stimulation. The lack of effect by a-adrenergic blockade on the decrease of mucociliary activity observed after combined pretreatment with
410
A . Cerrin et al.
atropine and propranolol further supports the view that r-receptors d o not mediate the inhibition of mucociliarj- activity. I n the nerve stimulated animals, not only is NA released but also NPY (Lundblad et (11. 1987, Lacroix et af. 1989), which may explain the different response to nerve stimulation after propranolol blockade, as compared to exogenously administered NA given after propranolol a h e r e no retarding effect was seen on the mucociliary system. T h e combined cholinergic and adrenergic blockade produced a decrease of mucociliary activity after sympathetic nerve stimulation. This finding suggests the presence of a nonadrenergic, non-cholinergic mechanism, such as the release of NPY which has recently been shown to decrease mucociliary- activity in the rabbit maxillary sinus by approximately l j o 0 in rii.0 when injected intraarterially (Cervin et af. 1991). T h e release of NPY after sympathetic nerve stimulation has been found in the pig and in the cat nasal mucosa, where high concentrations of N P Y and NL4 were released into the afferent venous flow (Lundblad et af. 1087; Lacroix et ul. 1989). T h e stimulation frequency of 1OHz was used in the present investigation, as it has been shown to release NPY at the highest ratio rts-u-zis N.A (Pernow et a/. 1988). T h e relatively long duration (approximately 5 min) of the decrease is in accord with the prolonged vasoconstrictive response to nerve stimulation found in the pig nasal mucosa (Lacroix et ul. 1988). Moreover, SPY-like immunoreactivity in the venous blood from the pig spleen in ziro has been found to be increased between 2 and 5 min after the cessation of nerve stimulation (Lundberg et a/. 1989). Taken together with the present results, such findings suggest that the decrease of mucociliary activity in response to nerve stimulation after adrenergic and cholinergic blockade may be explained by release of XPY. I t might be speculated that the effects on mucociliary activity by sympathetic nerve stimulation are indirect, as nerve stimulation also affects blood flow and secretory glands in the respiratory mucosa. T h e state of the vascular bed is probably of less importance, as ligation of the supply artery reducing the blood flow to almost nil does not affect mucociliary activity in thc rabbit maxillary sinus (Cervin et a/. 1988). Nerve stimulation ma)- also affect the secretory glands in the respiratory mucosa, producing
changes in the amount and rheological properties of the mucus, thereby changing the mucociliary activity. Wilson & Yates (1978) thus found sympathetic nerve stimulation to have a moderate stimulatory effect on secretory glands in the cat nasal mucosa. I n contrast, neither N A challenges in the rabbit trachea (Melville et al. 1976) nor N P Y challenges in the ferret trachea (Webber 1988) were found to have any effect on amount and kiscosity of the mucus. Therefore, secretory effects of sympathetic nerve stimulation are probabl! of little importance to the mucociliary system. As the antagonists used in the present in\ estigation had no effect on mucociliary activity per se, it is unlikely that an effect on mucus production by the antagonists was in any way responsible for the present results. T h a t a reflex mechanism mediated by the vagal nerve might explain the change of mucociliary activity is unlikely, as care was taken to se\er all connection with the vagal nerve, and as there was no concomitant fall in blood pressure or pulse rate. T h e small increase of blood pressure during nerve stimulation was probably due to the release of NA. Guanethidine, which blocks sympathetic neurotransmission, inhibited the effect of nerve stimulation on blood pressure and mucociliary activity (not only the increase of mucociliary activity observed after sympathetic n e n e stimulation, but also the decrease seen after blockade of classic autonomic receptors was inhibited), confirming that the effects were mediated by sympathetic nerve fibres. T o sum up, the present findings support the notion of adrenergic acceleration of mucociliary acti\ ity mediated via P-receptors by preganglionic sympathetic nerve stimulation. At 20 Hz, the acceleration has a cholinergic component. O u r results also indicate the release of an agent inhibiting mucociliary activity, an agent which may be NPY. T h e effect attributed to N P Y was seen only after blockade of classic autonomic receptors. The technical assistance of Mrs Charlotte CervinHoberg and Mr Samuel Leino is gratefully acknowledged. This investigation was supported by the Swedish Medical Research Council (Projects nos. 17X-07940 and 17P-845l), the Torsten Snderberg and Ragnar Soderberg Foundation, the Tore Nilson Foundation, the Medical Faculty of the University of Lund, Forenade Liv Mutual Group Life Insurance Company, the Craford Foundation and The Swedish Society for Medical Research.
Nerve stimulation and ciliary activity
41 1
relation to sympathetic vasoconstriction resistant to a-adrenoceptor blockade. Acta Physiol Scand 116, 393402. ANGG~RD, A,, DENSERT, 0. 1974. Adrenergic inLUNDBERG, J.M., ANGGKRD, A,, THEODORSSONnervation of the nasal mucosa in cat. A histological NORHEIM, E., PERNOW, J. 1984. Guanethidineand physiological study. Acta Otolaryngol (Stockh) of' Neuropeptide Y-like immunosensitive release 78, 232-241. reactivity in the a t spleen by sympathetic nerve CERVIN, A,, BENDE,M., LINDBERG, S., MERCKE, U., stimulation. Neurorci Lett 52, 175-180. P. 1988. Relations between blood flow and LUNDBERG, OLSSON, J.M., RUDEHILL,A,, SOLLEVI, A. 1989. mucociliary activity in the rabbit maxillary sinus. Pharmacological characterization of neuropeptide Acta Otolaryngol (Stockh) 105, 350-356. Y and noradrenaline mechanisms in sympathetic CERVIN, A., LINDBERG, S., MERCKE, U . 1991. The control of pig spleen. Eur 3 Pharmacul, 163, effects of neuropeptide Y on mucociliary activity in 103-1 13. the rabbit maxillary sinus. Acta Otolayyngol LUNDBLAD, L., ANGGKRD, A., SARIA, A,, LUNDBERG, (Stockh) 111, 960-966. J.M. 1987. Neuropeptide Y and non-adrenergic DAHLSTROM, A,, FUXE,K. 1965. The adrenergic sympathetic vascular control of the cat nasal innervation of the nasal mucosa of certain mammals. mucosa. 3 Auton Nerv Syst 20, 189-197. MARUYAMA, I. 1984. Conflicting effects of nor.4cta Otolaryngol (Stockh) 59, 65-72. adrenaline on ciliary movement of frog palatine EKBLAD,E., EDVINSSON, L., WAHLESTEDT, C., mucosa. Eur 3 Pharmacol97, 239-245. UDDMAN, R., H~KANSON, R., SUNDLER, F. 1984. G.N., HORSTMANN, G., IRAVANI, J. 1976. Neuropeptide Y co-exists and co-operates with MELVILLE, Adrenergic compounds and the respiratory tract. A noradrenaline in perivascular nerve fibres. Regul physiological and electron-microscopical study. Pept 8, 225-235. Respiration 33, 261-269. HYBBINETTE, J.-C., MERCKE, U. 1982a. A method for MERCKE, U., H~KANSSON, C.H., TOREMALM, N.G. evaluating the effect of pharmacological substances A method for standardized studies of 1974. on mucociliary activity in vino. Acta Otolaryngol mucociliary activity. Acta Otolaryngol (Stockh) 78, (Stockh) 93, 151-159. 118-123. HYBBINETTE, J.-C., MERCKE, U. 1982b. Effects of PERNOW, J., KAHAN,T., LUNDBERG, J.M. 1988. sympathomimetic agonists and antagonists on Neuropeptide Y and reserpine-resistant vasomucociliary activity. Acta Otolaqmgol (Stockh) 94, constriction evoked by sympathetic nerve stimu121-130. lation in the dog skeletal muscle. Br 3 Pharmacol HYBBINETTE, J.-C., MERCKE, U. 1 9 8 2 ~Effects . of the 94, 952-960. parasympathomimetic drug metacholine and its RUTTER, P.C., POTOCNIK, S.J., LUDBROOK, J. 1987. antagonists atropine on mucociliary activity. Acta Sympathoadrenal mechanisms in cardiovascular Otolaryngol (Stockh) 93, 465473. responses to naloxone after hemorrhage. Am 3 LACROIX, J.S., STJARNE, P., ANGGKRD, A,, LUNDBERG, Physiol, 252, H40-46. J., ADDICKS,H.W., ADDICKS, K. J.M. 1988. Sympathetic vascular control of the pig SCHINDELMEISER, 1982. Innervation of the mucosa of rabbit maxillary nasal mucosa (2) : reserpine-resistant, nonsinus. 11. Demonstration of catecholamine fluoradrenergic nervous responses in relation to neuroescence and acetylcholinesterase activity. Acta peptide Y and ATP. Acta Physiol Scand 133, Otolaryngol (Stockh) 94, 531-536. 183-197. K., CARLQVIST, M., MUTT, V. 1982. LACROIS, J.S., STJARNE, P., ANGGKRD, A., LUNDBERG,TATEMOTO, Y - a novel brain peptide with Neuropeptide J.M. 1989. Sympathetic vascular control of the pig structural similarities to peptide YY and pancreatic nasal mucosa (111) : co-release of noradrenaline and polypeptide. Nature 296, 659-660. neuropeptide Y. Acta Physiol Scand 135, 17-28. WEBBER, S.E. 1988. The effects of peptide histidine .ACROIX, J.S., ANGGKRD, A,, HOKFELT, T., OHARE, isoleucine and neuropeptide Y on mucus volume M.M.T., FAHRENKRUG, J., LUNDBERG, J.M. 1990. output from the ferret trachea. B r J Pharmacul95, Neuropeptide Y: presence in sympathetic and 49-54. parasympathetic innervation of the nasal mucosa. WILSON, H., YATES,M.S. 1978. Sympathetic nerves Cell Tissue Res 259, 119-128. and nasal secretion in the cat. Acta Otuluryngol ~ N D B E R G ,J.M., TATEMOTO, K. 1982. Pancreatic (Stockh) 85, 426430. polypeptide family (APP, BPP, NPY, and PPY) in
REFERENCES
