- Email: [email protected]
Presynaptic muscarinic receptor mechanisms and submandibular responses to stimulation of the parasympathetic innervation in bursts in rats
Autonomic Neuroscience: Basic and Clinical 99 (2002) 111 – 118 www.elsevier.com/locate/autneu
Presynaptic muscarinic receptor mechanisms and submandibular responses to stimulation of the parasympathetic innervation in bursts in rats Gunnar Tobin * Department of Pharmacology, Go¨teborg University, Box 431, SE-405 30 Go¨teborg, Sweden Received 11 February 2002; received in revised form 14 May 2002; accepted 14 May 2002
Abstract Submandibular secretory responses to electrical stimulation of the parasympathetic innervation at variable frequencies were measured in anaesthetized rats. Selective blockade by pirenzepine and by methoctramine occurred at doses (50 and of 300 nmol kg 1, i.v., respectively) that did not inhibit the responses to exogenous acetylcholine. In the presence of methoctramine, the nerve-evoked fluid responses were increased by 100% at 1 Hz independently of the total number of impulses (10 – 300), suggesting that M2 receptor activation inhibits transmitter release. The magnitude of the increase was inversely related to frequency of stimulation. The protein concentrations in the fluid responses were not significantly affected by methoctramine. Pirenzepine had an inhibitory effect on the fluid secretory responses, which was dependent of frequency, as well as of number of impulses, suggesting that M1 receptor activation facilitates transmitter release. At 10 Hz given intermittently (for 1 s at 10-s intervals), pirenzepine reduced the fluid response by 25%. The protein release was substantially and significantly reduced by pirenzepine independent of frequency but only during long periods of stimulation (300 impulses). It is concluded that muscarinic M1 receptor activation normally has a facilitatory effect on transmitter release, and that the facilitation occurs during short, intense stimulation. Muscarinic M1 receptors are, however, likely to regulate protein secretion by other mechanisms. Muscarinic M2 receptors, on the other hand, normally inhibit cholinergic transmission at low frequencies. Similar to findings in the alimentary tract of several species, stimulation in bursts at high frequencies is a more efficient stimulation pattern than continuous low frequency stimulation. This pattern of stimulation thus takes advantage of transient facilitation and avoids the inhibition at less intense neuronal activity. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Facilitation; Inhibition; Muscarinic receptor subtypes; Parasympathetic nervous system; Submandibular gland; Rat
1. Introduction Parasympathetic neurotransmission in the rat parotid gland is modulated by presynaptic muscarinic receptors, which may facilitate or inhibit transmitter release (Tobin, 1998). In this particular gland, a substantial part of the parasympathetic nerve-evoked response is resistant to atropine and adrenoceptor antagonists, and is thus mediated via non-adrenergic, non-cholinergic (NANC) transmitters (Ekstrom et al., 1983, 1985). In the rat auriculotemporal nerve, neuropeptides seem to constitute the NANC transmitters, and presynaptic muscarinic M1 receptors normally facilitate both cholinergic and peptidergic transmission during short, intense nerve activity (Tobin, 1998). At low *
Corresponding author. Tel.: +46-31-773-3442; fax: +46-31-82-1795. E-mail address: [email protected] (G. Tobin).
frequencies, on the other hand, presynaptic muscarinic M2 receptors inhibit cholinergic transmission, while at higher frequencies, peptidergic transmission is also modulated, but only after some delay. Examination of muscarinic receptor expression in the rat parotid gland, by employing subtype antisera for each of the m1 – m5 gene products, has shown low levels of m1 and m2 gene products, which were suggested to be of neuronal origin (Dai et al., 1991). In the rat submandibular gland, however, all five muscarinic receptor subtypes have been demonstrated (Brann et al., 1993; Flynn et al., 1997), but no localization to specific glandular structures has so far been reported. In contrast to parasympathetic responses in the rat parotid gland, acetylcholine mediates almost the whole parasympathetic response in the rat submandibular gland (Ekstrom et al., 1987). The present study was conducted in order to examine whether selective blockades of prejunctional muscarinic
1566-0702/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 6 - 0 7 0 2 ( 0 2 ) 0 0 0 9 4 - 2
112
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
receptors exert similar modulator effects in the submandibular gland as in the parotid gland. Accordingly, muscarinic receptor antagonists with different selectivity profiles (atropine, nonselective; pirenzepine, ‘‘M1-selective’’; methoctramine, ‘‘M2-selective’’; Eglen and Watson, 1996) have been used to investigate the indirect effects of selective blockades of axonal receptors on chorda tympani nerve-evoked secretory responses. The importance of both the frequency and duration of nerve stimulation was explored. Furthermore, it was first described in the feline submandibular gland that stimulation of the parasympathetic innervation in a burst pattern at high frequencies causes a conspicuous enhancement of vasodilatation and secretion in comparison with continuous stimulation (Bloom and Edwards, 1979; Andersson et al., 1982b). The observation has been confirmed in salivary glands of several species (Andersson et al., 1982a,b; Tobin et al., 1990) including the rat parotid gland (Ekstrom et al., 1984; Tobin, 1998). The phenomenon has been attributed to the release of neuropeptides, which preferentially occurs at high stimulation frequencies (Bloom and Edwards, 1979; Andersson et al., 1982a). Therefore, it was currently wondered whether stimulation in a burst pattern enhances rat submandibular secretory responses, and further, if presynaptic modulation by muscarinic receptors could contribute to the enhancement. A preliminary report of some of these results has been published previously (Tobin, 1996).
2. Materials and methods 2.1. Surgical procedures Forty adult male Sprague –Dawley rats (weighing 300 – 450 g) were used in the study. Anaesthesia was induced by an intraperitoneal injection of pentobarbitone (40 mg kg 1). An intravenous route for drug administration was established via a cannula inserted into the femoral vein, which was used for all further injections. Supplementary anaesthetic was injected every second hour (10 mg kg 1, i.v.). The blood pressure was monitored continuously via a catheter inserted into the femoral artery. Body temperature was kept at 38 jC by means of a thermostatically controlled heating blanket. The preparatory surgical techniques were generally as described previously (Ekstrom et al., 1987; Tobin, 1998). Briefly, the lingual nerve was cut, proximal to the point at which the chorda tympani becomes separated, and the peripheral end was ligated for subsequent electrical stimulation; this standard procedure enables the parasympathetic nerve fibres, which pass to the submandibular gland in the chorda tympani, to be stimulated within the lingual nerve and so minimizes mechanical damage. The trachea and submandibular duct were cannulated. The free end of the submandibular cannula was positioned above a photoelectric dropcounter so that drops of saliva were recorded; all saliva appearing at the top of the cannula was collected
and weighed. No spontaneous secretion occurred. At the end of the experiments, the animals were killed with an overdose of pentobarbitone, and the submandibular glands weighed (262 F 6 mg, n = 40). 2.2. Experimental procedures Submandibular responses to electrical stimulation of the chorda tympani and to intravenous injections of acetylcholine were determined. Electrical stimulation of the cut peripheral end of the nerve was performed by employing a standard 8 V square-wave stimulus with a pulse width of 2 ms at variable frequencies. Stimulation at 1 Hz is well above the threshold frequency for a fluid response in the rat submandibular gland, whereas stimulation at 10 Hz is close to EF70 and at 20 Hz close to EFmax. Furthermore, 1, 2 and 10 Hz is below the threshold frequency for the atropineresistant secretion, while only a minute atropine-resistant response occurs at 20 Hz ( < 1% of the parasympathetic response in absence of atropine; Ekstrom et al., 1987). The cut peripheral end of the nerve was stimulated at 40 Hz (maximum for the atropine-resistant secretion) over 20 min initially in order to obtain reference values and to ensure to get stable and invariable responses. This stimulation also aimed at minimizing the atropine-resistant secretory response; when this stimulation frequency was repeated at the end of the experiments and in the presence of atropine (1.5 Amol kg 1, i.v.) no secretion occurred. In the successive experimental protocol, stimulation at 1, 2, 10 and 20 Hz was delivered in 10, 30 and 300 impulses before and after administration of pirenzepine (50 nmol kg 1, i.v.) and methoctramine (300 nmol kg 1, i.v.). Occasional observations were made at the frequencies of 4 and 40 Hz. When comparisons were made between burst and continuous stimulation, the nerve was stimulated at either a high frequency given at intervals (10, 20 and 40 Hz for 1 s at 10-s intervals), or at a low (1, 2 and 4 Hz) or at a high frequency (10, 20 and 40 Hz) given continuously. The protocol was designed so as the same total number of impulses would be delivered. Saliva was routinely collected for protein analysis in the responses, except when inadequate volumes were obtained (1 and 2 Hz for 30 s). The protein analysis also included the fluid response to exogenous acetylcholine (1 Amol kg 1, i.v.). All stimulation procedures were performed at least twice in succession in order to evacuate the dead space of the submandibular cannula (3– 4 Al) and avoid contamination with secretion elicited by some previous procedure. The second sample was analysed for protein by the method of Lowry et al. (1951). 2.3. Quantification of the experimental results The fluid responses to electrical stimulation are expressed as the volume secreted per impulse delivered (nl impulse 1) or as the total volume secreted during the whole period of stimulation. It should be noted that the perfusion pressure
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
(mean arterial pressure) was not significantly affected by the muscarinic antagonists or by nerve stimulation. BP before antagonist administration: 85 F 5 mm Hg (n = 32); after methoctramine: 88 F 7 mm Hg (n = 16); after pirenzepine: 85 F 6 mm Hg (n = 16); after atropine: 79 F 8 mm Hg (n = 32).
113
3.2. Effects of nerve stimulation and acetylcholine in the absence of muscarinic antagonists Electrical stimulation of the chorda lingual nerve at 1, 2 and 4 Hz produced continuous flow of saliva after a latency of 5– 10 s, while 10, 20 and 40 Hz produced almost instant and brisk secretory responses. At neither frequency or
2.4. Drugs and solutions The following drugs were used: acetylcholine chloride (Sigma, St. Louis, US), atropine sulphate (Sigma), methoctramine hydrochloride (Research Biochemicals, Natick, US), phentolamine methansulphate (Ciba-Geigy, Basel, Switzerland), pirenzepine dihydrochloride (Sigma), propranolol hydrochloride (ICI Pharmaceuticals, UK). The rats were pretreated with a- and h-adrenergic antagonists (phentolamine and propranolol; 1.6 Amol kg 1, i.v.), to exclude adrenergic effects, and a supplementary dose (0.8 Amol kg 1, i.v.) of these antagonists was injected every 90 min. 2.5. Statistical analysis Statistical significance was determined by Student’s t-test for unpaired and paired data. When multiple comparisons with the same variable were made, a t-test according to the Bonferroni method was used (Wallenstein et al., 1980). Probability values of 0.05 or less were regarded as statistically significant. Values are presented in the form of means F S.E.M.
3. Results 3.1. Effects of muscarinic antagonists on acetylcholineevoked secretion Intravenous injections of acetylcholine at 1 Amol kg 1 produced responses (24 F 4 Al) with a short delay (5 –10 s) that ceased within a minute. The protein concentration therein was 0.62 F 0.07 Ag Al 1 (n = 12). At this dose of acetylcholine, the blood pressure was reduced from 85 F 5 to 60 F 6 mm Hg and recovery occurred within 1 min. The inhibitory effects of pirenzepine and methoctramine at successively increasing doses (10 nmol kg 1 to 1 Amol kg 1, i.v.) were examined on responses to acetylcholine repeatedly administered (1 Amol kg 1, i.v). Methoctramine did not affect the acetylcholine-evoked fluid and protein secretion at neither dose. Pirenzepine, on the other hand, did not affect the fluid responses at the lower doses but caused a significant reduction at the largest dose (1 Amol kg 1: 19 F 1 Al; n = 6, p < 0.05). The protein concentration was however reduced at a lower concentration (from 0.60 F 0.04 to 0.52 F 0.03 Ag Al 1 at 100 nmol kg 1, i.v.; p < 0.05). At 1 Amol kg 1 i.v. of pirenzepine the concentration was reduced to 0.38 F 0.07 Ag Al 1; p < 0.05).
Fig. 1. Mean submandibular secretion of saliva in response to stimulation of the chorda lingual nerve for 30 s in pentobarbitone anaesthetized rats expressed as total volume secreted (panel A) and as volume secreted per impulse (panel B). The experiments were performed in the presence of aand h-adrenoceptor blockade. Each point represents a mean value of either 10 observations. Vertical bars represent S.E.M. Note that the frequency of stimulation is plotted on a logarithmic scale.
114
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
pattern of stimulation did the nerve stimulation cause any change of the mean arterial blood pressure (85 F 5 mm Hg). When the stimulation was discontinued, secretion ceased within a few seconds at all frequencies. The fluid response to stimulation at 20 Hz for 30 s was substantially larger than the responses to the other stimulation frequencies delivered for 30 s, and somewhat larger than that at 40 Hz (Fig. 1A). However, when the fluid response was expressed as volume secreted per impulse, the efficiency was as high at 2, 4 and 10 Hz as at 20 Hz, but still lower at 1 Hz and substantially lower at 40 Hz (Fig. 1B; p < 0.05 – 0.01). The protein concentrations in the saliva produced by all the stimulation frequencies and delivered in 300 impulses were almost identical (means 0.55 to 0.63 Ag Al 1). Intermittent stimulation at 10 and 20 Hz given for 1 s at 10-s intervals over a period of 30 s (3 bursts), elicited larger secretory responses than continuous stimulation for 30 s at 1 and 2 Hz, respectively (Fig. 2). Burst stimulation at 40 Hz 1:10 over 30 s, however, elicited an approximately 30% smaller response than continuous stimulation at 4 Hz over the same period of time. While a continuous stimulation at 10 Hz, giving the same total number of impulses (for 3 s), elicited a lower response than the intermittent stimulation (3.0 F 0.3 vs. 4.0 F 0.5 Al; p < 0.05), stimulation at 20 Hz for 3 s evoked a similar fluid response as at the frequency given intermittently (6.9 F 0.9 vs. 6.9 F 1.2 Al). When expressed as secretion per impulse, the best efficiency was obtained at burst stimulation at 10 Hz given for 1 s at 10-s intervals (130 F 17 nl impulse 1). While continuous stimulation at 1 and 2 Hz for 30 s resulted in stimulation efficiency of 57 F 14 and 83 F 18 nl impulse 1, respectively, continuous
Fig. 2. Mean submandibular secretion of saliva in response to stimulation of the chorda lingual nerve for 30 s in pentobarbitone anaesthetized rats. Stimulation was performed at a low frequency delivered continuously (5) and at a high frequency delivered for 1 s at 10-s intervals (n). The experiments were performed in the presence of a- and h-adrenoceptor blockade. Each column represents a mean value of either 10 observations. Vertical bars represent S.E.M. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Fig. 3. Mean fluid responses to stimulation of the chorda lingual nerve in pentobarbitone-anaesthetized rats before (5) and after (n) administration of methoctramine (300 nmol kg 1, i.v.) expressed as volume secreted per impulse delivered. The stimulations were performed at varying frequencies over such periods that the same total number of impulses was delivered (30 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each point represents a mean value of 10 observations. Vertical bars represent S.E.M. **, p < 0.01. Note that the frequency of stimulation is plotted on a logarithmic scale.
stimulation at 20 Hz for 30 s resulted in 95 F 3 nl impulse 1. Thus, intermittent stimulation at 10 Hz given for 1 s at 10-s intervals showed higher efficiency than any other design of stimulation.
Fig. 4. Mean fluid responses to stimulation of the chorda lingual nerve in pentobarbitone anaesthetized rats before (5) and after (n) administration of pirenzepine (50 nmol kg 1, i.v.) expressed as volume secreted per impulse delivered. The stimulations were performed at varying frequencies over such periods that the same total number of impulses was delivered (10 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each point represents a mean value of 10 observations. Vertical bars represent S.E.M. *, p < 0.05. Note that the frequency of stimulation is plotted on a logarithmic scale.
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
3.3. Effects of nerve stimulation and acetylcholine in the presence of methoctramine Methoctramine at 300 nmol kg 1 i.v. invariably increased the fluid responses to stimulation at 1 and 2 Hz, most conspicuously at 1 Hz (Fig. 3). At 1 Hz, the flow increased by about 100% independently of the number of impulses delivered. The increase in the presence of methoctramine during stimulation at 2 and 10 Hz was less than half that obtained at 1 Hz, but it was still independent of the duration of stimulation, while no increase occurred at 20 Hz in the presence of methoctramine. After administration of
115
methoctramine, the efficiency was almost the same at all frequencies (mean range: 150 to 170 nl impulse 1), except at 10 Hz in bursts (200 F 20 nl impulse 1), at which it was significantly greater ( p < 0.01). At 20 Hz in burst (190 F 30 nl impulse 1), no statistical significance was attained, but, if any difference occurred in comparison with that in the absence of the antagonist, the response tended to be greater. According to the protein concentration, methoctramine did not affect the protein in the saliva produced at any frequency. 3.4. Effects of nerve stimulation and acetylcholine in the presence of pirenzepine Pirenzepine (50 nmol kg 1, i.v.) significantly reduced the fluid response to stimulation at 10 and 20 Hz, but just for 10 impulses and consequently, the enhancement of burst stimulation was also eliminated, as is reflected by the change in efficiency (Fig. 4). Before pirenzepine administration, the fluid response to 10 Hz given for 1 s at 10-s intervals over 30 s amounted to 3.9 F 0.6 Al, whereas the corresponding response in the presence was 2.7 F 0.5 Al ( p < 0.01; n = 10). Pirenzepine did not affect the fluid response at either of the other frequencies and patterns. However, the protein concentration was significantly reduced in the saliva obtained at 1, 2, 10 and 20 Hz when given continuously in 300 impulses (Fig. 5) and most conspicuously at 1 Hz. At 1 Hz, the protein concentration was reduced from 0.50 F 0.05 to 0.31 F 0.03 Ag Al 1 by pirenzepine. No reduction occurred of the protein concentration in saliva secreted in response to burst stimulation. Neither caused pirenzepine any reduction at shorter stimulation durations irrespective of frequency.
4. Discussion 4.1. Burst stimulation and selective muscarinic antagonism
Fig. 5. Mean changes in protein concentrations in fluid responses to stimulation of the chorda lingual nerve at varying frequencies in pentobarbitone-anaesthetized rats before (5) and after administration of (A) methoctramine (300 nmol kg 1, i.v.; n) and (B) pirenzepine (50 nmol kg 1, i.v.; n). Stimulation was carried out at varying frequencies over such periods that the same total number of impulses was delivered (300 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each column represents a mean value of 10 observations. Vertical bars represent S.E.M. *, p < 0.05; ** p < 0.01; ***, p < 0.001.
In salivary glands of several species, stimulation in bursts at high frequencies has been found to more effectively evoke vasodilator and secretory responses than continuous low frequency stimulation (Edwards et al., 1984) and further, the burst pattern has been shown to increase the release of neuropeptides, such as VIP, from parasympathetic salivary nerves (Bloom and Edwards, 1979; Andersson et al., 1982a,b; Ekstrom et al., 1984). Since peptides may potentiate responses to classical transmitters (Ekstrom and Olgart, 1986; Ekstrom and Tobin, 1990), and since the release of peptide transmitters, occur at high stimulation frequencies (Thureson-Klein et al., 1986; Thureson-Klein and Klein, 1990), neuropeptides are most likely important contributors to the burst enhancement. The enhancement varies between species, and tentatively, the degree of the occurrence of peptidergic co-transmitters may be of importance for the efficiency. However, an interplay of presynaptic modulatory mechanisms may also contribute, and the
116
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
burst stimulation may thus take advantage of transient facilitation and avoid instantaneous inhibition which occurs at less intense neuronal activity (Tobin, 1998). In contrast to the rat parotid gland, peptidergic transmitters are of low significance for the parasympathetic response in the rat submandibular gland (Ekstrom et al., 1987). Even so, initial electrical stimulation of the chorda lingual nerve may evoke small NANC mediated responses, which vanish upon repeated stimulation, probably because of depletion of peptidergic transmitters. To ensure stable transmitter contributions throughout the experiment and avoid initial high releases of peptide, an ongoing stimulation at a high frequency was currently applied to the chorda lingual nerve in order to achieve transmitter equilibrium, which occurs after 10 – 20 min of an ongoing stimulation (Ekstrom et al., 1987). In previous experiments in the parotid gland, burst stimulation at 10 Hz (for 1 s at 10-s intervals) was found to induce a secretion of 5.4 F 0.6 Al over a period of 30 s, while continuous stimulation for 30 s at 1 Hz induced a response of 4.1 F 0.5 Al ( p < 0.05; unpublished observation). Thus, the secretion increased by about 30%. This observation is consistent with the findings previously reported (Tobin, 1998). Currently, the burst pattern appeared to be more efficient in the submandibular gland than in the parotid gland of the rat, since the increase amounted to 135% in the former gland. High contents of peptides seem therefore not to be an absolute prerequisite for high efficiency of burst stimulation. The current experiments were undertaken in order to study the effects of selective muscarinic antagonists on rat submandibular secretory activity in response to stimulation of the parasympathetic innervation and to identify mechanisms by which intermittent stimulation pattern enhances responses to electrical stimulation. Evidence that selective muscarinic antagonism was achieved is provided by the fact that the doses of the pharmacological agents (pirenzepine and methoctramine) have no effect on the postsynaptic actions of exogenous acetylcholine in the parotid gland (Tobin, 1998), which are known to be mediated by muscarinic M3 receptor (Dai et al., 1991; Bockman et al., 2001). In the submandibular gland of the rat, however, muscarinic M5 co-receptors may contribute postsynaptically to the secretory response (Flynn et al., 1997; Eglen and Nahorski, 2000; Meloy et al., 2001). Nor did the doses of pirenzepine and methoctramine that were employed have any effect postsynaptically on fluid responses to acetylcholine in the rat submandibular gland. Pirenzepine affected, however, acetylcholine induced protein secretion postjunctionally, since the antagonist substantially reduced protein concentration. Furthermore, the doses employed consistently produced opposite effects on the salivary responses to parasympathetic stimulation and have been validated in a similar way in the submandibular and urinary bladder of the rabbit (Tobin, 1995; Tobin and Sjogren, 1995). The physiological effects that were presently studied could only be attributed to the M1 and M2 subpopulations of muscarinic receptor
because of the muscarinic profiles of pirenzepine and methoctramine, in spite of all muscarinic receptor subpopulations being expressed in the submandibular gland of the rat (Flynn et al., 1997). Pirenzepine shows almost no selectivity for the muscarinic M5 over the muscarinic M3 receptor, but shows 20 times greater affinity for muscarinic M1 receptor over muscarinic M3 and M5 receptors (Eglen and Nahorski, 2000). Methoctramine, on the other hand, shows little selectivity difference between the muscarinic M4 and the M3 subpopulations (4 times), whereas the affinity for muscarinic M2 receptors is 50 times greater than for muscarinic M3 receptors (Eglen and Nahorski, 2000). 4.2. Muscarinic M2 receptor blockade and secretion The present results suggest that the transmission of chorda lingual nerve impulses in the submandibular gland of the rat is modulated by presynaptic muscarinic M2 receptors, reminiscent of the muscarinic M2 receptor inhibition of synaptic transmission in the parasympathetic innervation of the rabbit submandibular gland (Tobin, 1995) and of the rat parotid gland (Tobin, 1998). The inhibition from muscarinic M2 receptor activation seems to be fast in onset (within 1 s) and independent of the duration of stimulation. However, it grew less with more intense stimulation and disappeared altogether at 20 Hz. The protein response was, in neither case, affected by the administration of methoctramine, thus indicating a pure cholinergic response as was also supported by the fact that atropine totally abolished the nerve-evoked secretion when given at the end of the experiments. The responses to stimulation in bursts were increased by methoctramine in similar degree as continuous stimulation, which agrees with the idea that muscarinic presynaptic inhibition is not dependent of the duration of the stimulation period, but only of frequency. Thus, high frequency stimulation in bursts avoids the muscarinic M2 receptor inhibition that occurs at low frequencies. 4.3. Muscarinic M1 receptor blockade and secretion The present results also suggest that transmission in the chorda lingual nerve is under the influence of facilitatory muscarinic M1 receptors, and that this type of facilitation occurs during relatively intense stimulation for short periods of time. It only occurred at the highest frequencies employed, and then just for the shortest stimulation period (10 impulses). The muscarinic M1 receptor facilitation occurring at intense but short stimulation was also reflected by pirenzepine abolishing the higher efficiency at intermittent stimulation. In the presence of pirenzepine, no variations occurred at the frequencies independently of the duration of the stimulation period. According to the protein concentration, this was reduced by pirenzepine at the high frequencies, as well as at the low, with exception for intermittent stimulation. Since pirenzepine reduced the
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
protein concentration in the acetylcholine-evoked saliva also, without affecting the fluid response, the effect on the protein concentration is likely to be exerted postjunctionally. In the rat submandibular gland, muscarinic M1 receptors have been suggested to induce increased synthesis of nitric oxide (Mitsui et al., 1997), and in salivary glands, nitric oxide has been intimately connected to the secretion of protein (Buckle et al., 1995; Edwards et al., 1996; Tobin et al., 1997; Hanna and Edwards, 1998). Tentatively, when a stimulation pattern is applied for very brief periods, this may be insufficient to induce the indirect effects via nitric oxide. Alternatively, this way to stimulate for protein secretion is overruled at very intense stimulation. Anyway, the muscarinic M1 receptor modulation of the protein secretion engages another mechanism than that modulating the fluid secretory responses. All in all, presynaptic muscarinic M1 receptors facilitate the cholinergic transmission in the submandibular gland, and likewise, to the facilitation of cholinergic and peptidergic transmission in the rat parotid gland. This occurs at intense activity for short periods. It is concluded that the release of acetylcholine from the chorda lingual nerve in the submandibular gland of the rat is normally facilitated by axonal muscarinic receptors of the M1 subtype and inhibited by those of the M2 subtype. Whereas the evidence indicates that facilitation is preferentially activated at a transient intensive neuronal activity, the inhibition appeared to be activated almost instantaneously at low frequencies. Similar to findings in the alimentary tract of several species, where stimulation in bursts at high frequencies has been found to be more efficient than continuous low frequency stimulation (Edwards et al., 1984), such a pattern also enhances the nerve-evoked secretion from rat salivary glands. This pattern of stimulation, taking advantage of transient facilitation and avoiding inhibition at less intense neuronal activity, does not necessarily need the contribution of peptidergic transmitters to be more efficient.
Acknowledgements This study was supported by grants from the Swedish Dental Society, Magnus Bergvalls Stiftelse and Wilhelm and Martina Lundgrens Vetenskapsfond.
References Andersson, P.O., Bloom, S.R., Edwards, A.V., 1982a. Parotid responses to stimulation of the parasympathetic innervation in bursts in weaned lambs. J. Physiol. 330, 163 – 174. Andersson, P.O., Bloom, S.R., Edwards, A.V., Jarhult, J., 1982b. Effects of stimulation of the chorda tympani in bursts on submaxillary responses in the cat. J. Physiol. 322, 469 – 483. Bloom, S.R., Edwards, A.V., 1979. The relationship between release of vasoactive intestinal peptide in the salivary gland of the cat in response
117
to parasympathetic stimulation and the atropine resistant vasodilatation. J. Physiol. 295, 35 – 36. Bockman, C.S., Bradley, M.E., Dang, H.K., Zeng, W., Scofield, M.A., Dowd, F.J., 2001. Molecular and pharmacological characterization of muscarinic receptor subtypes in a rat parotid gland cell line: comparison with native parotid gland. J. Pharmacol. Exp. Ther. 297, 718 – 726. Brann, M.R., Jorgensen, H.B., Burstein, E.S., Spalding, T.A., Ellis, J., Jones, S.V., Hill-Eubanks, D., 1993. Studies of the pharmacology, localization, and structure of muscarinic acetylcholine receptors. Ann. N. Y. Acad. Sci. 707, 225 – 236. Buckle, A.D., Parker, S.J., Bloom, S.R., Edwards, A.V., 1995. The role of nitric oxide in the control of protein secretion in the submandibular gland of the cat. Exp. Physiol. 80, 1019 – 1030. Dai, Y.S., Ambudkar, I.S., Horn, V.J., Yeh, C.K., Kousvelari, E.E., Wall, S.J., Li, M., Yasuda, R.P., Wolfe, B.B., Baum, B.J., 1991. Evidence that M3 muscarinic receptors in rat parotid gland couple to two second messenger systems. Am. J. Physiol. 261, C1063 – C1073. Edwards, A.V., Andersson, P.O., Jarhult, J., Bloom, S.R., 1984. Studies of the importance of the pattern of autonomic stimulation in relation to alimentary effectors. Q. J. Exp. Physiol. 69, 607 – 614. Edwards, A.V., Tobin, G., Ekstrom, J., Bloom, S.R., 1996. Nitric oxide and release of the peptide VIP from parasympathetic terminals in the submandibular gland of the anaesthetized cat. Exp. Physiol. 81, 349 – 359. Eglen, R.M., Nahorski, S.R., 2000. The muscarinic M(5) receptor: a silent or emerging subtype? Br. J. Pharmacol. 130, 13 – 21. Eglen, R.M., Watson, N., 1996. Selective muscarinic receptor agonists and antagonists. Pharmacol. Toxicol. 78, 59 – 68. Ekstrom, J., Olgart, L., 1986. Complementary action of substance P and vasoactive intestinal peptide on the rat parotid secretion. Acta Physiol. Scand. 126, 25 – 31. Ekstrom, J., Tobin, G., 1990. Protein secretion in salivary glands of cats in vivo and in vitro in response to vasoactive intestinal peptide. Acta Physiol. Scand. 140, 95 – 103. Ekstrom, J., Mansson, B., Tobin, G., Garrett, J.R., Thulin, A., 1983. Atropine-resistant secretion of parotid saliva on stimulation of the auriculotemporal nerve. Acta Physiol. Scand. 119, 445 – 449. Ekstrom, J., Brodin, E., Ekman, R., Hakanson, R., Sundler, F., 1984. Vasoactive intestinal peptide and substance P in salivary glands of the rat following denervation or duct ligation. Regul. Pept. 10, 1 – 10. Ekstrom, J., Brodin, E., Ekman, R., Hakanson, R., Mansson, B., Tobin, G., 1985. Depletion of neuropeptides in rat parotid glands and declining atropine-resistant salivary secretion upon continuous parasympathetic nerve stimulation. Regul. Pept. 11, 353 – 359. Ekstrom, J., Mansson, B., Tobin, G., 1987. Non-adrenergic, non-cholinergic parasympathetic secretion in the rat submaxillary and sublingual glands. Pharmacol. Toxicol. 60, 284 – 287. Flynn, D.D., Reever, C.M., Ferrari-DiLeo, G., 1997. Pharmacological strategies to selectively label and localize muscarinic receptor subtypes. Drug Dev. Res. 40, 104 – 116. Hanna, S.J., Edwards, A.V., 1998. The role of nitric oxide in the control of protein secretion in the parotid gland of anaesthetized sheep. Exp. Physiol. 83, 533 – 544. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265 – 275. Meloy, T.D., Daniels, D.V., Hegde, S.S., Eglen, R.M., Ford, A.P., 2001. Functional characterization of rat submaxillary gland muscarinic receptors using microphysiometry. Br. J. Pharmacol. 132, 1606 – 1614. Mitsui, Y., Yasuda, N., Furuyama, S., Sugiya, H., 1997. Nitric oxide synthase activities in mammalian parotid and submandibular salivary glands. Arch. Oral Biol. 42, 621 – 624. Thureson-Klein, A.K., Klein, R.L., 1990. Exocytosis from neuronal large dense cored vesicles. Int. Rev. Cyt. 121, 67 – 126. Thureson-Klein, A., Klein, R.L., Zhu, P.C., 1986. Exocytosis from large dense cored vesicles as a mechanism for neuropeptide release in the peripheral and central nervous system. Scanning Electron Microsc., 179 – 187.
118
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
Tobin, G., 1995. Muscarinic receptor subtypes in the submandibular gland and the urinary bladder of the rabbit: in vivo and in vitro functional comparisons of receptor antagonists. J. Auton. Pharm. 15, 451 – 463. Tobin, G., 1996. Presynaptic muscarinic receptors modulating parasympathetic responses in rat salivary glands. Swed. Dent. J. 20, 232. Tobin, G., 1998. Presynaptic muscarinic M1 and M2 receptor modulation of auriculotemporal nerve transmission in the rat. J. Auton. Nerv. Syst. 72, 61 – 71. Tobin, G., Sjogren, C., 1995. In vivo and in vitro effects of muscarinic receptor antagonists on contractions and release of [3H]acetylcholine in the rabbit urinary bladder. Eur. J. Pharmacol. 281, 1 – 8.
Tobin, G., Ekstrom, J., Edwards, A.V., 1990. Submandibular responses to stimulation of the parasympathetic innervation in bursts in the anaesthetized ferret. J. Physiol. (Lond.) 431, 417 – 425. Tobin, G., Edwards, A.V., Bloom, S.R., Ekstrom, J., 1997. Nitric oxide in the control of submandibular gland function in the anaesthetized ferret. Exp. Physiol. 82, 825 – 836. Wallenstein, S., Zucker, C.L., Fleiss, J.L., 1980. Some statistical methods useful in circulation research. Circ. Res. 47, 1 – 9.
Presynaptic muscarinic receptor mechanisms and submandibular responses to stimulation of the parasympathetic innervation in bursts in rats Gunnar Tobin * Department of Pharmacology, Go¨teborg University, Box 431, SE-405 30 Go¨teborg, Sweden Received 11 February 2002; received in revised form 14 May 2002; accepted 14 May 2002
Abstract Submandibular secretory responses to electrical stimulation of the parasympathetic innervation at variable frequencies were measured in anaesthetized rats. Selective blockade by pirenzepine and by methoctramine occurred at doses (50 and of 300 nmol kg 1, i.v., respectively) that did not inhibit the responses to exogenous acetylcholine. In the presence of methoctramine, the nerve-evoked fluid responses were increased by 100% at 1 Hz independently of the total number of impulses (10 – 300), suggesting that M2 receptor activation inhibits transmitter release. The magnitude of the increase was inversely related to frequency of stimulation. The protein concentrations in the fluid responses were not significantly affected by methoctramine. Pirenzepine had an inhibitory effect on the fluid secretory responses, which was dependent of frequency, as well as of number of impulses, suggesting that M1 receptor activation facilitates transmitter release. At 10 Hz given intermittently (for 1 s at 10-s intervals), pirenzepine reduced the fluid response by 25%. The protein release was substantially and significantly reduced by pirenzepine independent of frequency but only during long periods of stimulation (300 impulses). It is concluded that muscarinic M1 receptor activation normally has a facilitatory effect on transmitter release, and that the facilitation occurs during short, intense stimulation. Muscarinic M1 receptors are, however, likely to regulate protein secretion by other mechanisms. Muscarinic M2 receptors, on the other hand, normally inhibit cholinergic transmission at low frequencies. Similar to findings in the alimentary tract of several species, stimulation in bursts at high frequencies is a more efficient stimulation pattern than continuous low frequency stimulation. This pattern of stimulation thus takes advantage of transient facilitation and avoids the inhibition at less intense neuronal activity. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Facilitation; Inhibition; Muscarinic receptor subtypes; Parasympathetic nervous system; Submandibular gland; Rat
1. Introduction Parasympathetic neurotransmission in the rat parotid gland is modulated by presynaptic muscarinic receptors, which may facilitate or inhibit transmitter release (Tobin, 1998). In this particular gland, a substantial part of the parasympathetic nerve-evoked response is resistant to atropine and adrenoceptor antagonists, and is thus mediated via non-adrenergic, non-cholinergic (NANC) transmitters (Ekstrom et al., 1983, 1985). In the rat auriculotemporal nerve, neuropeptides seem to constitute the NANC transmitters, and presynaptic muscarinic M1 receptors normally facilitate both cholinergic and peptidergic transmission during short, intense nerve activity (Tobin, 1998). At low *
Corresponding author. Tel.: +46-31-773-3442; fax: +46-31-82-1795. E-mail address: [email protected] (G. Tobin).
frequencies, on the other hand, presynaptic muscarinic M2 receptors inhibit cholinergic transmission, while at higher frequencies, peptidergic transmission is also modulated, but only after some delay. Examination of muscarinic receptor expression in the rat parotid gland, by employing subtype antisera for each of the m1 – m5 gene products, has shown low levels of m1 and m2 gene products, which were suggested to be of neuronal origin (Dai et al., 1991). In the rat submandibular gland, however, all five muscarinic receptor subtypes have been demonstrated (Brann et al., 1993; Flynn et al., 1997), but no localization to specific glandular structures has so far been reported. In contrast to parasympathetic responses in the rat parotid gland, acetylcholine mediates almost the whole parasympathetic response in the rat submandibular gland (Ekstrom et al., 1987). The present study was conducted in order to examine whether selective blockades of prejunctional muscarinic
1566-0702/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 6 - 0 7 0 2 ( 0 2 ) 0 0 0 9 4 - 2
112
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
receptors exert similar modulator effects in the submandibular gland as in the parotid gland. Accordingly, muscarinic receptor antagonists with different selectivity profiles (atropine, nonselective; pirenzepine, ‘‘M1-selective’’; methoctramine, ‘‘M2-selective’’; Eglen and Watson, 1996) have been used to investigate the indirect effects of selective blockades of axonal receptors on chorda tympani nerve-evoked secretory responses. The importance of both the frequency and duration of nerve stimulation was explored. Furthermore, it was first described in the feline submandibular gland that stimulation of the parasympathetic innervation in a burst pattern at high frequencies causes a conspicuous enhancement of vasodilatation and secretion in comparison with continuous stimulation (Bloom and Edwards, 1979; Andersson et al., 1982b). The observation has been confirmed in salivary glands of several species (Andersson et al., 1982a,b; Tobin et al., 1990) including the rat parotid gland (Ekstrom et al., 1984; Tobin, 1998). The phenomenon has been attributed to the release of neuropeptides, which preferentially occurs at high stimulation frequencies (Bloom and Edwards, 1979; Andersson et al., 1982a). Therefore, it was currently wondered whether stimulation in a burst pattern enhances rat submandibular secretory responses, and further, if presynaptic modulation by muscarinic receptors could contribute to the enhancement. A preliminary report of some of these results has been published previously (Tobin, 1996).
2. Materials and methods 2.1. Surgical procedures Forty adult male Sprague –Dawley rats (weighing 300 – 450 g) were used in the study. Anaesthesia was induced by an intraperitoneal injection of pentobarbitone (40 mg kg 1). An intravenous route for drug administration was established via a cannula inserted into the femoral vein, which was used for all further injections. Supplementary anaesthetic was injected every second hour (10 mg kg 1, i.v.). The blood pressure was monitored continuously via a catheter inserted into the femoral artery. Body temperature was kept at 38 jC by means of a thermostatically controlled heating blanket. The preparatory surgical techniques were generally as described previously (Ekstrom et al., 1987; Tobin, 1998). Briefly, the lingual nerve was cut, proximal to the point at which the chorda tympani becomes separated, and the peripheral end was ligated for subsequent electrical stimulation; this standard procedure enables the parasympathetic nerve fibres, which pass to the submandibular gland in the chorda tympani, to be stimulated within the lingual nerve and so minimizes mechanical damage. The trachea and submandibular duct were cannulated. The free end of the submandibular cannula was positioned above a photoelectric dropcounter so that drops of saliva were recorded; all saliva appearing at the top of the cannula was collected
and weighed. No spontaneous secretion occurred. At the end of the experiments, the animals were killed with an overdose of pentobarbitone, and the submandibular glands weighed (262 F 6 mg, n = 40). 2.2. Experimental procedures Submandibular responses to electrical stimulation of the chorda tympani and to intravenous injections of acetylcholine were determined. Electrical stimulation of the cut peripheral end of the nerve was performed by employing a standard 8 V square-wave stimulus with a pulse width of 2 ms at variable frequencies. Stimulation at 1 Hz is well above the threshold frequency for a fluid response in the rat submandibular gland, whereas stimulation at 10 Hz is close to EF70 and at 20 Hz close to EFmax. Furthermore, 1, 2 and 10 Hz is below the threshold frequency for the atropineresistant secretion, while only a minute atropine-resistant response occurs at 20 Hz ( < 1% of the parasympathetic response in absence of atropine; Ekstrom et al., 1987). The cut peripheral end of the nerve was stimulated at 40 Hz (maximum for the atropine-resistant secretion) over 20 min initially in order to obtain reference values and to ensure to get stable and invariable responses. This stimulation also aimed at minimizing the atropine-resistant secretory response; when this stimulation frequency was repeated at the end of the experiments and in the presence of atropine (1.5 Amol kg 1, i.v.) no secretion occurred. In the successive experimental protocol, stimulation at 1, 2, 10 and 20 Hz was delivered in 10, 30 and 300 impulses before and after administration of pirenzepine (50 nmol kg 1, i.v.) and methoctramine (300 nmol kg 1, i.v.). Occasional observations were made at the frequencies of 4 and 40 Hz. When comparisons were made between burst and continuous stimulation, the nerve was stimulated at either a high frequency given at intervals (10, 20 and 40 Hz for 1 s at 10-s intervals), or at a low (1, 2 and 4 Hz) or at a high frequency (10, 20 and 40 Hz) given continuously. The protocol was designed so as the same total number of impulses would be delivered. Saliva was routinely collected for protein analysis in the responses, except when inadequate volumes were obtained (1 and 2 Hz for 30 s). The protein analysis also included the fluid response to exogenous acetylcholine (1 Amol kg 1, i.v.). All stimulation procedures were performed at least twice in succession in order to evacuate the dead space of the submandibular cannula (3– 4 Al) and avoid contamination with secretion elicited by some previous procedure. The second sample was analysed for protein by the method of Lowry et al. (1951). 2.3. Quantification of the experimental results The fluid responses to electrical stimulation are expressed as the volume secreted per impulse delivered (nl impulse 1) or as the total volume secreted during the whole period of stimulation. It should be noted that the perfusion pressure
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
(mean arterial pressure) was not significantly affected by the muscarinic antagonists or by nerve stimulation. BP before antagonist administration: 85 F 5 mm Hg (n = 32); after methoctramine: 88 F 7 mm Hg (n = 16); after pirenzepine: 85 F 6 mm Hg (n = 16); after atropine: 79 F 8 mm Hg (n = 32).
113
3.2. Effects of nerve stimulation and acetylcholine in the absence of muscarinic antagonists Electrical stimulation of the chorda lingual nerve at 1, 2 and 4 Hz produced continuous flow of saliva after a latency of 5– 10 s, while 10, 20 and 40 Hz produced almost instant and brisk secretory responses. At neither frequency or
2.4. Drugs and solutions The following drugs were used: acetylcholine chloride (Sigma, St. Louis, US), atropine sulphate (Sigma), methoctramine hydrochloride (Research Biochemicals, Natick, US), phentolamine methansulphate (Ciba-Geigy, Basel, Switzerland), pirenzepine dihydrochloride (Sigma), propranolol hydrochloride (ICI Pharmaceuticals, UK). The rats were pretreated with a- and h-adrenergic antagonists (phentolamine and propranolol; 1.6 Amol kg 1, i.v.), to exclude adrenergic effects, and a supplementary dose (0.8 Amol kg 1, i.v.) of these antagonists was injected every 90 min. 2.5. Statistical analysis Statistical significance was determined by Student’s t-test for unpaired and paired data. When multiple comparisons with the same variable were made, a t-test according to the Bonferroni method was used (Wallenstein et al., 1980). Probability values of 0.05 or less were regarded as statistically significant. Values are presented in the form of means F S.E.M.
3. Results 3.1. Effects of muscarinic antagonists on acetylcholineevoked secretion Intravenous injections of acetylcholine at 1 Amol kg 1 produced responses (24 F 4 Al) with a short delay (5 –10 s) that ceased within a minute. The protein concentration therein was 0.62 F 0.07 Ag Al 1 (n = 12). At this dose of acetylcholine, the blood pressure was reduced from 85 F 5 to 60 F 6 mm Hg and recovery occurred within 1 min. The inhibitory effects of pirenzepine and methoctramine at successively increasing doses (10 nmol kg 1 to 1 Amol kg 1, i.v.) were examined on responses to acetylcholine repeatedly administered (1 Amol kg 1, i.v). Methoctramine did not affect the acetylcholine-evoked fluid and protein secretion at neither dose. Pirenzepine, on the other hand, did not affect the fluid responses at the lower doses but caused a significant reduction at the largest dose (1 Amol kg 1: 19 F 1 Al; n = 6, p < 0.05). The protein concentration was however reduced at a lower concentration (from 0.60 F 0.04 to 0.52 F 0.03 Ag Al 1 at 100 nmol kg 1, i.v.; p < 0.05). At 1 Amol kg 1 i.v. of pirenzepine the concentration was reduced to 0.38 F 0.07 Ag Al 1; p < 0.05).
Fig. 1. Mean submandibular secretion of saliva in response to stimulation of the chorda lingual nerve for 30 s in pentobarbitone anaesthetized rats expressed as total volume secreted (panel A) and as volume secreted per impulse (panel B). The experiments were performed in the presence of aand h-adrenoceptor blockade. Each point represents a mean value of either 10 observations. Vertical bars represent S.E.M. Note that the frequency of stimulation is plotted on a logarithmic scale.
114
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
pattern of stimulation did the nerve stimulation cause any change of the mean arterial blood pressure (85 F 5 mm Hg). When the stimulation was discontinued, secretion ceased within a few seconds at all frequencies. The fluid response to stimulation at 20 Hz for 30 s was substantially larger than the responses to the other stimulation frequencies delivered for 30 s, and somewhat larger than that at 40 Hz (Fig. 1A). However, when the fluid response was expressed as volume secreted per impulse, the efficiency was as high at 2, 4 and 10 Hz as at 20 Hz, but still lower at 1 Hz and substantially lower at 40 Hz (Fig. 1B; p < 0.05 – 0.01). The protein concentrations in the saliva produced by all the stimulation frequencies and delivered in 300 impulses were almost identical (means 0.55 to 0.63 Ag Al 1). Intermittent stimulation at 10 and 20 Hz given for 1 s at 10-s intervals over a period of 30 s (3 bursts), elicited larger secretory responses than continuous stimulation for 30 s at 1 and 2 Hz, respectively (Fig. 2). Burst stimulation at 40 Hz 1:10 over 30 s, however, elicited an approximately 30% smaller response than continuous stimulation at 4 Hz over the same period of time. While a continuous stimulation at 10 Hz, giving the same total number of impulses (for 3 s), elicited a lower response than the intermittent stimulation (3.0 F 0.3 vs. 4.0 F 0.5 Al; p < 0.05), stimulation at 20 Hz for 3 s evoked a similar fluid response as at the frequency given intermittently (6.9 F 0.9 vs. 6.9 F 1.2 Al). When expressed as secretion per impulse, the best efficiency was obtained at burst stimulation at 10 Hz given for 1 s at 10-s intervals (130 F 17 nl impulse 1). While continuous stimulation at 1 and 2 Hz for 30 s resulted in stimulation efficiency of 57 F 14 and 83 F 18 nl impulse 1, respectively, continuous
Fig. 2. Mean submandibular secretion of saliva in response to stimulation of the chorda lingual nerve for 30 s in pentobarbitone anaesthetized rats. Stimulation was performed at a low frequency delivered continuously (5) and at a high frequency delivered for 1 s at 10-s intervals (n). The experiments were performed in the presence of a- and h-adrenoceptor blockade. Each column represents a mean value of either 10 observations. Vertical bars represent S.E.M. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Fig. 3. Mean fluid responses to stimulation of the chorda lingual nerve in pentobarbitone-anaesthetized rats before (5) and after (n) administration of methoctramine (300 nmol kg 1, i.v.) expressed as volume secreted per impulse delivered. The stimulations were performed at varying frequencies over such periods that the same total number of impulses was delivered (30 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each point represents a mean value of 10 observations. Vertical bars represent S.E.M. **, p < 0.01. Note that the frequency of stimulation is plotted on a logarithmic scale.
stimulation at 20 Hz for 30 s resulted in 95 F 3 nl impulse 1. Thus, intermittent stimulation at 10 Hz given for 1 s at 10-s intervals showed higher efficiency than any other design of stimulation.
Fig. 4. Mean fluid responses to stimulation of the chorda lingual nerve in pentobarbitone anaesthetized rats before (5) and after (n) administration of pirenzepine (50 nmol kg 1, i.v.) expressed as volume secreted per impulse delivered. The stimulations were performed at varying frequencies over such periods that the same total number of impulses was delivered (10 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each point represents a mean value of 10 observations. Vertical bars represent S.E.M. *, p < 0.05. Note that the frequency of stimulation is plotted on a logarithmic scale.
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
3.3. Effects of nerve stimulation and acetylcholine in the presence of methoctramine Methoctramine at 300 nmol kg 1 i.v. invariably increased the fluid responses to stimulation at 1 and 2 Hz, most conspicuously at 1 Hz (Fig. 3). At 1 Hz, the flow increased by about 100% independently of the number of impulses delivered. The increase in the presence of methoctramine during stimulation at 2 and 10 Hz was less than half that obtained at 1 Hz, but it was still independent of the duration of stimulation, while no increase occurred at 20 Hz in the presence of methoctramine. After administration of
115
methoctramine, the efficiency was almost the same at all frequencies (mean range: 150 to 170 nl impulse 1), except at 10 Hz in bursts (200 F 20 nl impulse 1), at which it was significantly greater ( p < 0.01). At 20 Hz in burst (190 F 30 nl impulse 1), no statistical significance was attained, but, if any difference occurred in comparison with that in the absence of the antagonist, the response tended to be greater. According to the protein concentration, methoctramine did not affect the protein in the saliva produced at any frequency. 3.4. Effects of nerve stimulation and acetylcholine in the presence of pirenzepine Pirenzepine (50 nmol kg 1, i.v.) significantly reduced the fluid response to stimulation at 10 and 20 Hz, but just for 10 impulses and consequently, the enhancement of burst stimulation was also eliminated, as is reflected by the change in efficiency (Fig. 4). Before pirenzepine administration, the fluid response to 10 Hz given for 1 s at 10-s intervals over 30 s amounted to 3.9 F 0.6 Al, whereas the corresponding response in the presence was 2.7 F 0.5 Al ( p < 0.01; n = 10). Pirenzepine did not affect the fluid response at either of the other frequencies and patterns. However, the protein concentration was significantly reduced in the saliva obtained at 1, 2, 10 and 20 Hz when given continuously in 300 impulses (Fig. 5) and most conspicuously at 1 Hz. At 1 Hz, the protein concentration was reduced from 0.50 F 0.05 to 0.31 F 0.03 Ag Al 1 by pirenzepine. No reduction occurred of the protein concentration in saliva secreted in response to burst stimulation. Neither caused pirenzepine any reduction at shorter stimulation durations irrespective of frequency.
4. Discussion 4.1. Burst stimulation and selective muscarinic antagonism
Fig. 5. Mean changes in protein concentrations in fluid responses to stimulation of the chorda lingual nerve at varying frequencies in pentobarbitone-anaesthetized rats before (5) and after administration of (A) methoctramine (300 nmol kg 1, i.v.; n) and (B) pirenzepine (50 nmol kg 1, i.v.; n). Stimulation was carried out at varying frequencies over such periods that the same total number of impulses was delivered (300 impulses). The experiments were performed in the presence of a- and hadrenoceptor blockade. Each column represents a mean value of 10 observations. Vertical bars represent S.E.M. *, p < 0.05; ** p < 0.01; ***, p < 0.001.
In salivary glands of several species, stimulation in bursts at high frequencies has been found to more effectively evoke vasodilator and secretory responses than continuous low frequency stimulation (Edwards et al., 1984) and further, the burst pattern has been shown to increase the release of neuropeptides, such as VIP, from parasympathetic salivary nerves (Bloom and Edwards, 1979; Andersson et al., 1982a,b; Ekstrom et al., 1984). Since peptides may potentiate responses to classical transmitters (Ekstrom and Olgart, 1986; Ekstrom and Tobin, 1990), and since the release of peptide transmitters, occur at high stimulation frequencies (Thureson-Klein et al., 1986; Thureson-Klein and Klein, 1990), neuropeptides are most likely important contributors to the burst enhancement. The enhancement varies between species, and tentatively, the degree of the occurrence of peptidergic co-transmitters may be of importance for the efficiency. However, an interplay of presynaptic modulatory mechanisms may also contribute, and the
116
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
burst stimulation may thus take advantage of transient facilitation and avoid instantaneous inhibition which occurs at less intense neuronal activity (Tobin, 1998). In contrast to the rat parotid gland, peptidergic transmitters are of low significance for the parasympathetic response in the rat submandibular gland (Ekstrom et al., 1987). Even so, initial electrical stimulation of the chorda lingual nerve may evoke small NANC mediated responses, which vanish upon repeated stimulation, probably because of depletion of peptidergic transmitters. To ensure stable transmitter contributions throughout the experiment and avoid initial high releases of peptide, an ongoing stimulation at a high frequency was currently applied to the chorda lingual nerve in order to achieve transmitter equilibrium, which occurs after 10 – 20 min of an ongoing stimulation (Ekstrom et al., 1987). In previous experiments in the parotid gland, burst stimulation at 10 Hz (for 1 s at 10-s intervals) was found to induce a secretion of 5.4 F 0.6 Al over a period of 30 s, while continuous stimulation for 30 s at 1 Hz induced a response of 4.1 F 0.5 Al ( p < 0.05; unpublished observation). Thus, the secretion increased by about 30%. This observation is consistent with the findings previously reported (Tobin, 1998). Currently, the burst pattern appeared to be more efficient in the submandibular gland than in the parotid gland of the rat, since the increase amounted to 135% in the former gland. High contents of peptides seem therefore not to be an absolute prerequisite for high efficiency of burst stimulation. The current experiments were undertaken in order to study the effects of selective muscarinic antagonists on rat submandibular secretory activity in response to stimulation of the parasympathetic innervation and to identify mechanisms by which intermittent stimulation pattern enhances responses to electrical stimulation. Evidence that selective muscarinic antagonism was achieved is provided by the fact that the doses of the pharmacological agents (pirenzepine and methoctramine) have no effect on the postsynaptic actions of exogenous acetylcholine in the parotid gland (Tobin, 1998), which are known to be mediated by muscarinic M3 receptor (Dai et al., 1991; Bockman et al., 2001). In the submandibular gland of the rat, however, muscarinic M5 co-receptors may contribute postsynaptically to the secretory response (Flynn et al., 1997; Eglen and Nahorski, 2000; Meloy et al., 2001). Nor did the doses of pirenzepine and methoctramine that were employed have any effect postsynaptically on fluid responses to acetylcholine in the rat submandibular gland. Pirenzepine affected, however, acetylcholine induced protein secretion postjunctionally, since the antagonist substantially reduced protein concentration. Furthermore, the doses employed consistently produced opposite effects on the salivary responses to parasympathetic stimulation and have been validated in a similar way in the submandibular and urinary bladder of the rabbit (Tobin, 1995; Tobin and Sjogren, 1995). The physiological effects that were presently studied could only be attributed to the M1 and M2 subpopulations of muscarinic receptor
because of the muscarinic profiles of pirenzepine and methoctramine, in spite of all muscarinic receptor subpopulations being expressed in the submandibular gland of the rat (Flynn et al., 1997). Pirenzepine shows almost no selectivity for the muscarinic M5 over the muscarinic M3 receptor, but shows 20 times greater affinity for muscarinic M1 receptor over muscarinic M3 and M5 receptors (Eglen and Nahorski, 2000). Methoctramine, on the other hand, shows little selectivity difference between the muscarinic M4 and the M3 subpopulations (4 times), whereas the affinity for muscarinic M2 receptors is 50 times greater than for muscarinic M3 receptors (Eglen and Nahorski, 2000). 4.2. Muscarinic M2 receptor blockade and secretion The present results suggest that the transmission of chorda lingual nerve impulses in the submandibular gland of the rat is modulated by presynaptic muscarinic M2 receptors, reminiscent of the muscarinic M2 receptor inhibition of synaptic transmission in the parasympathetic innervation of the rabbit submandibular gland (Tobin, 1995) and of the rat parotid gland (Tobin, 1998). The inhibition from muscarinic M2 receptor activation seems to be fast in onset (within 1 s) and independent of the duration of stimulation. However, it grew less with more intense stimulation and disappeared altogether at 20 Hz. The protein response was, in neither case, affected by the administration of methoctramine, thus indicating a pure cholinergic response as was also supported by the fact that atropine totally abolished the nerve-evoked secretion when given at the end of the experiments. The responses to stimulation in bursts were increased by methoctramine in similar degree as continuous stimulation, which agrees with the idea that muscarinic presynaptic inhibition is not dependent of the duration of the stimulation period, but only of frequency. Thus, high frequency stimulation in bursts avoids the muscarinic M2 receptor inhibition that occurs at low frequencies. 4.3. Muscarinic M1 receptor blockade and secretion The present results also suggest that transmission in the chorda lingual nerve is under the influence of facilitatory muscarinic M1 receptors, and that this type of facilitation occurs during relatively intense stimulation for short periods of time. It only occurred at the highest frequencies employed, and then just for the shortest stimulation period (10 impulses). The muscarinic M1 receptor facilitation occurring at intense but short stimulation was also reflected by pirenzepine abolishing the higher efficiency at intermittent stimulation. In the presence of pirenzepine, no variations occurred at the frequencies independently of the duration of the stimulation period. According to the protein concentration, this was reduced by pirenzepine at the high frequencies, as well as at the low, with exception for intermittent stimulation. Since pirenzepine reduced the
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
protein concentration in the acetylcholine-evoked saliva also, without affecting the fluid response, the effect on the protein concentration is likely to be exerted postjunctionally. In the rat submandibular gland, muscarinic M1 receptors have been suggested to induce increased synthesis of nitric oxide (Mitsui et al., 1997), and in salivary glands, nitric oxide has been intimately connected to the secretion of protein (Buckle et al., 1995; Edwards et al., 1996; Tobin et al., 1997; Hanna and Edwards, 1998). Tentatively, when a stimulation pattern is applied for very brief periods, this may be insufficient to induce the indirect effects via nitric oxide. Alternatively, this way to stimulate for protein secretion is overruled at very intense stimulation. Anyway, the muscarinic M1 receptor modulation of the protein secretion engages another mechanism than that modulating the fluid secretory responses. All in all, presynaptic muscarinic M1 receptors facilitate the cholinergic transmission in the submandibular gland, and likewise, to the facilitation of cholinergic and peptidergic transmission in the rat parotid gland. This occurs at intense activity for short periods. It is concluded that the release of acetylcholine from the chorda lingual nerve in the submandibular gland of the rat is normally facilitated by axonal muscarinic receptors of the M1 subtype and inhibited by those of the M2 subtype. Whereas the evidence indicates that facilitation is preferentially activated at a transient intensive neuronal activity, the inhibition appeared to be activated almost instantaneously at low frequencies. Similar to findings in the alimentary tract of several species, where stimulation in bursts at high frequencies has been found to be more efficient than continuous low frequency stimulation (Edwards et al., 1984), such a pattern also enhances the nerve-evoked secretion from rat salivary glands. This pattern of stimulation, taking advantage of transient facilitation and avoiding inhibition at less intense neuronal activity, does not necessarily need the contribution of peptidergic transmitters to be more efficient.
Acknowledgements This study was supported by grants from the Swedish Dental Society, Magnus Bergvalls Stiftelse and Wilhelm and Martina Lundgrens Vetenskapsfond.
References Andersson, P.O., Bloom, S.R., Edwards, A.V., 1982a. Parotid responses to stimulation of the parasympathetic innervation in bursts in weaned lambs. J. Physiol. 330, 163 – 174. Andersson, P.O., Bloom, S.R., Edwards, A.V., Jarhult, J., 1982b. Effects of stimulation of the chorda tympani in bursts on submaxillary responses in the cat. J. Physiol. 322, 469 – 483. Bloom, S.R., Edwards, A.V., 1979. The relationship between release of vasoactive intestinal peptide in the salivary gland of the cat in response
117
to parasympathetic stimulation and the atropine resistant vasodilatation. J. Physiol. 295, 35 – 36. Bockman, C.S., Bradley, M.E., Dang, H.K., Zeng, W., Scofield, M.A., Dowd, F.J., 2001. Molecular and pharmacological characterization of muscarinic receptor subtypes in a rat parotid gland cell line: comparison with native parotid gland. J. Pharmacol. Exp. Ther. 297, 718 – 726. Brann, M.R., Jorgensen, H.B., Burstein, E.S., Spalding, T.A., Ellis, J., Jones, S.V., Hill-Eubanks, D., 1993. Studies of the pharmacology, localization, and structure of muscarinic acetylcholine receptors. Ann. N. Y. Acad. Sci. 707, 225 – 236. Buckle, A.D., Parker, S.J., Bloom, S.R., Edwards, A.V., 1995. The role of nitric oxide in the control of protein secretion in the submandibular gland of the cat. Exp. Physiol. 80, 1019 – 1030. Dai, Y.S., Ambudkar, I.S., Horn, V.J., Yeh, C.K., Kousvelari, E.E., Wall, S.J., Li, M., Yasuda, R.P., Wolfe, B.B., Baum, B.J., 1991. Evidence that M3 muscarinic receptors in rat parotid gland couple to two second messenger systems. Am. J. Physiol. 261, C1063 – C1073. Edwards, A.V., Andersson, P.O., Jarhult, J., Bloom, S.R., 1984. Studies of the importance of the pattern of autonomic stimulation in relation to alimentary effectors. Q. J. Exp. Physiol. 69, 607 – 614. Edwards, A.V., Tobin, G., Ekstrom, J., Bloom, S.R., 1996. Nitric oxide and release of the peptide VIP from parasympathetic terminals in the submandibular gland of the anaesthetized cat. Exp. Physiol. 81, 349 – 359. Eglen, R.M., Nahorski, S.R., 2000. The muscarinic M(5) receptor: a silent or emerging subtype? Br. J. Pharmacol. 130, 13 – 21. Eglen, R.M., Watson, N., 1996. Selective muscarinic receptor agonists and antagonists. Pharmacol. Toxicol. 78, 59 – 68. Ekstrom, J., Olgart, L., 1986. Complementary action of substance P and vasoactive intestinal peptide on the rat parotid secretion. Acta Physiol. Scand. 126, 25 – 31. Ekstrom, J., Tobin, G., 1990. Protein secretion in salivary glands of cats in vivo and in vitro in response to vasoactive intestinal peptide. Acta Physiol. Scand. 140, 95 – 103. Ekstrom, J., Mansson, B., Tobin, G., Garrett, J.R., Thulin, A., 1983. Atropine-resistant secretion of parotid saliva on stimulation of the auriculotemporal nerve. Acta Physiol. Scand. 119, 445 – 449. Ekstrom, J., Brodin, E., Ekman, R., Hakanson, R., Sundler, F., 1984. Vasoactive intestinal peptide and substance P in salivary glands of the rat following denervation or duct ligation. Regul. Pept. 10, 1 – 10. Ekstrom, J., Brodin, E., Ekman, R., Hakanson, R., Mansson, B., Tobin, G., 1985. Depletion of neuropeptides in rat parotid glands and declining atropine-resistant salivary secretion upon continuous parasympathetic nerve stimulation. Regul. Pept. 11, 353 – 359. Ekstrom, J., Mansson, B., Tobin, G., 1987. Non-adrenergic, non-cholinergic parasympathetic secretion in the rat submaxillary and sublingual glands. Pharmacol. Toxicol. 60, 284 – 287. Flynn, D.D., Reever, C.M., Ferrari-DiLeo, G., 1997. Pharmacological strategies to selectively label and localize muscarinic receptor subtypes. Drug Dev. Res. 40, 104 – 116. Hanna, S.J., Edwards, A.V., 1998. The role of nitric oxide in the control of protein secretion in the parotid gland of anaesthetized sheep. Exp. Physiol. 83, 533 – 544. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265 – 275. Meloy, T.D., Daniels, D.V., Hegde, S.S., Eglen, R.M., Ford, A.P., 2001. Functional characterization of rat submaxillary gland muscarinic receptors using microphysiometry. Br. J. Pharmacol. 132, 1606 – 1614. Mitsui, Y., Yasuda, N., Furuyama, S., Sugiya, H., 1997. Nitric oxide synthase activities in mammalian parotid and submandibular salivary glands. Arch. Oral Biol. 42, 621 – 624. Thureson-Klein, A.K., Klein, R.L., 1990. Exocytosis from neuronal large dense cored vesicles. Int. Rev. Cyt. 121, 67 – 126. Thureson-Klein, A., Klein, R.L., Zhu, P.C., 1986. Exocytosis from large dense cored vesicles as a mechanism for neuropeptide release in the peripheral and central nervous system. Scanning Electron Microsc., 179 – 187.
118
G. Tobin / Autonomic Neuroscience: Basic and Clinical 99 (2002) 111–118
Tobin, G., 1995. Muscarinic receptor subtypes in the submandibular gland and the urinary bladder of the rabbit: in vivo and in vitro functional comparisons of receptor antagonists. J. Auton. Pharm. 15, 451 – 463. Tobin, G., 1996. Presynaptic muscarinic receptors modulating parasympathetic responses in rat salivary glands. Swed. Dent. J. 20, 232. Tobin, G., 1998. Presynaptic muscarinic M1 and M2 receptor modulation of auriculotemporal nerve transmission in the rat. J. Auton. Nerv. Syst. 72, 61 – 71. Tobin, G., Sjogren, C., 1995. In vivo and in vitro effects of muscarinic receptor antagonists on contractions and release of [3H]acetylcholine in the rabbit urinary bladder. Eur. J. Pharmacol. 281, 1 – 8.
Tobin, G., Ekstrom, J., Edwards, A.V., 1990. Submandibular responses to stimulation of the parasympathetic innervation in bursts in the anaesthetized ferret. J. Physiol. (Lond.) 431, 417 – 425. Tobin, G., Edwards, A.V., Bloom, S.R., Ekstrom, J., 1997. Nitric oxide in the control of submandibular gland function in the anaesthetized ferret. Exp. Physiol. 82, 825 – 836. Wallenstein, S., Zucker, C.L., Fleiss, J.L., 1980. Some statistical methods useful in circulation research. Circ. Res. 47, 1 – 9.