Pulmonary Pharmacology (1990) 3 93-101 ©1990 Longman Group UK Ltd
0952-0600/90/0003-0093/$10 .00
PULMONARY PHARMACOLOGY
The Effects of Local Anaesthetic Agents upon Mucus Secretion in the Feline Trachea In Vivo M . Somerville*§, J .-A. Karlssont & P .S . Richardson* *Department of Physiology, St George's Hospital Medical School, Cranmer Terrace, London SW17 ORE, UK, §Host Defence Unit, Cardiothoracic Institute, Dovehouse St, London SW3 6LY, UK, 'Department of Pharmacology, AB Draco, S-221 00 Lund, Sweden SUMMARY. The actions of lignocaine and tetrodotoxin (TTX) in a tracheal segment of the cat were tested on secretion of mucus macromolecules radiolabelled with 35S and 3 H. Lignocaine, 4.3-43 mM, given into the segment, caused a concentration dependent increase of secretion of 3H-and 35 S-labelled macromolecules. At 43 mM, lignocaine increased secretion : A 3H = +433+191%, A 35S = +327+34.5% (n=8). This effect lessened over 15-45 min . Atropine (1 mg/kg) had little effect on these responses. All concentrations of lignocaine tested (4.3-43 mM) abolished the effect of vagus nerve stimulation on secretion and diminished the effect of a submaximal concentration of pilocarpine (5 µM) in the segment in a dose-dependent manner . TTX in the segment did not alter the resting secretion . At 50 µM it abolished, and at 10 µM diminished, vagal control of secretion without affecting the secretory response to pilocarpine . The study shows that lignocaine, in concentrations which block vagal control of secretion (>_ 4.3 mM), stimulates the release of mucus macromolecules . Resting secretion is unaltered by TTX, and so does not appear to be under neurogenic inhibition . Larger concentrations of lignocaine (>_ 13 mM) also diminish pilocarpine-induced secretion, whereas TTX may inhibit nervous control of mucus secretion selectively . The results suggest that clinical anaesthesia of the airways with lignocaine may stimulate mucus secretion .
block all motor pathways controlling airway secretion while leaving direct mechanisms intact . The neurotoxin tetrodotoxin (TTX) has been used to study airway mucus secretion in vitro, 10-12 but not in intact animals . Similarly the effects of local anaesthetics on mucus secretion have scarcely been investigated, apart from one study showing lignocaine to stimulate secretion into the goose trachea ." The effect of lignocaine on mucociliary transport has been examined, but these studies have shown either no effect or an inhibition of transport rates . 14 The experiments reported here were done both to test whether interference with neural control of airway secretion by local anaesthetics in vivo would affect resting secretion and that stimulated by a muscarinic agonist, and also to see how lignocaine, a drug commonly used to render the airways anaesthetic in man, may alter secretion in an animal model of the human airway . This paper describes the effects of local anaesthetics on the resting and stimulated outputs of mucus macromolecules radiolabelled with 3 H and 35 S in the trachea of the anaesthetised cat . In the past these have been referred to as mucus glycoproteins or mucins," but recent evidence has shown that, though much of the radioactivity is in this form, some other macromolecules are also radiolabelled ." - " In this paper we refer to the radiolabelled materials in dialysed secretions as mucus macromolecules .
INTRODUCTION In the airways, the acinar glands of the submucosa and the goblet and ciliated cells of the surface epithelium secrete mucus macromolecules including glycoproteins .' These molecules constitute part of the periciliary fluid in which the cilia beat and also of the more highly visco-elastic layer of mucus which overlies the cilia and which is transported by them . Secretion of mucus macromolecules is involved in the mechanisms which defend the airways ; inhalations of irritants and some mediators have been shown to accelerate their release . 2 ' 3 Both sympathetic and parasympathetic nerves control airway mucus secretion in the cat, and antagonists to adrenoceptors and muscarinic receptors, respectively, inhibit the actions of these motor pathways . 4- ' More recently, however, it has become clear that some nerve-mediated effects survive the presence of large concentrations of muscarinic and adrenoceptor antagonists ; so non-adrenergic, non-cholinergic (NANC) nerves appear to form part of the secretomotor pathway ." - ' o It would be useful to have a drug to
Correspondence to : Dr P .S . Richardson, Department of Physiology, St . George's Hospital Medical School, London SWl7 ORE . 93
94
Pulmonary Pharmacology
METHODS Anaesthesia and preparation of cats Anaesthesia was induced in cats with pentobarbitone sodium, 42 mg/Kg, by intraperitoneal injection . Subsequent doses of pentobarbitone were given via a catheter into the femoral vein to sustain anaesthesia . Two cannulae were inserted into the cervical trachea to isolate a tracheal segment in situ, with nerve and blood supply intact, from which mucus could be flushed . The cat breathed through a third cannula, inserted into the trachea at its entry into the thorax, caudal to the isolated segment .' The tracheal segment was flushed with Krebs-Henseleit solution and then filled with 0 .5 mCi of 3H-glucose and 2 .0 mCi of sodium 35 S-sulphate in 4 .0 ml Krebs-Henseleit for 1 h to allow the secretory cells to take up these radiolabelled precursors and synthesise them into mucus macromolecules which would later be secreted into the tracheal lumen . At intervals in the course of the experiment, the tracheal segment was flushed with 10 ml Krebs-Henseleit solution . The tracheal washings were first mixed with guanidine hydrochloride, giving a 6M solution to dissolve mucus, and were then dialysed repeatedly against distilled water containing 0.1 % of both sodium azide and sodium sulphate to disperse radiolabel not covalently bound to macromolecules . After dialysis, samples were weighed and three 1 .0 ml aliquots from each were mixed with 10-12 ml of scintillation fluid (Scintran Cocktail T, BDH) and counted in a calibrated liquid scintillation spectrometer to determine the quantities of 35 S- and 3H-labels bound to macromolecules in each sample . In some experiments, where the effect of lignocaine was tested in the presence of atropine, three 1 .0 ml aliquots were also analysed for glycoconjugate content by the periodic acid Schiff (PAS) reaction ." The method was calibrated against standard concentrations of bovine submaxillary mucin (Sigma) and results are expressed in terms of this standard . Protocol The result from one experiment, involving lignocaine application to the trachea and which illustrates the protocol used, is illustrated in Figure 1 . After the period of radiolabelling had finished (time zero), mucus collections were made at intervals for 2 .25 h before any local anaesthetic was given . Then the first concentration of lignocaine hydrochloride (Astra), or control Krebs-Henseleit solution as in Figure 1, was given in Krebs-Henseleit for 2 h (Period 1), during which time the cut peripheral ends of the vagus nerves were stimulated (pulses of 10 V and 2 msec at 10 Hz, for 10 min of a 15 min collection period s and pilocarpine, 5 µM in Krebs-Henseleit, was given into the segment for another 15 min collection period . This concentration of pilocarpine was chosen to give a clear but submaximal stimulation of secretion (see Results) . In the cat trachea, pilocarpine's effect on
3H-macromolecules a C ou
PERIOD 1 (Control : Krebs-H)
$ 8-macromolecules
a
PERIOD 2 (4 .3 mid lignocaine)
C
Fig. 1 - Graphs showing the rate of output of 3 H-(upper panel) and 35 S-labelled macromolecules (lower panel) against time in one experiment . The graphs illustrate the design of experiments, with plain Krebs-Henseleit solution filling the tracheal segment for the first 2 .25 h of the experiment, then, over the next 2 .0 h, the first test period (PERIOD 1 ; in this case Krebs-Henseleit control), then 0.5 h of Krebs-Henseleit again, followed by the second test period (PERIOD 2 ; in this case 4.3 mM lignocaine in Krebs-Henseleit) . Finally there was a single control sample, for 15 min, with plain Krebs-Henseleit filling the segment again . During both Periods 1 and 2, secretion was stimulated where indicated by vagus nerve stimulation and 5 tM pilocarpine .
secretion is abolished by atropine ,4' 9 evidence that this agonist acts solely via muscarinic receptors . Lignocaine was then replaced with plain Krebs-Henseleit for 30 min before the next concentration of lignocaine, or Krebs-Henseleit control in some cases, was applied for 2 h (Period 2) and the vagal stimulation and pilocarpine treatment repeated . On two occasions the pH of Krebs-Henseleit and the most concentrated lignocaine hydrochloride solution (43 mM) were measured and found to be 7 .55 and 7 .60 for the former and 7 .35 and 7 .40 for the latter. In four experiments, repeated stimulations by pilocarpine and by electrical stimulation of the peripheral ends of the vagus nerves were given in the absence of lignocaine to test the reproducibility of the responses . In the remainder, three concentrations of lignocaine hydrochloride, 4 .3 mM, 13 .0 mM and 43 mM (i .e . 0.10%, 0 .30% and 1 .0% w/v lignocaine base respectively), were used and compared to Krebs-Henseleit buffer. Identical concentrations were not repeated in
Local Anaesthesia of the Airway and Mucus Secretion
the same experiment, but otherwise all possible combinations of concentrations were tested . In experiments where the effects of lignocaine were tested in the presence of atropine, the muscarinic antagonist was injected i.v ., 1 mg/kg body weight, 45 min before Period 1, and also given at a concentration of 1 .7 gM into the tracheal wash fluid for the rest of the experiment . These concentrations of atropine are sufficient to abolish the effect of 50 tM pilocarpine on airway secretion .' Two series of experiments concerned tetrodotoxin (TTX) . The initial concentration (10 µM) was chosen as being one order of . magnitude greater than the concentration necessary to block all secretory effects of field stimulation in human bronchi in vitro 12 (see Discussion) . A preliminary set was performed in which TTX (Sigma or Cambridge Bioscience), either 10 µM or 50 µM in Krebs-Henseleit, was given into the tracheal segment and the vagus nerves stimulated (see above) after 30 min exposure to this neurotoxin, or pilocarpine (50 µM) was given at the same stage in the experiment . The design of the second set resembled that for lignocaine with TTX (50 µM) alternating with Krebs-Henseleit in the two periods . In two cats, the action of histamine hydrochloride was tested on the secretion of radiolabelled macromolecules into the trachea . The trachea was prepared and radiolabelled as described above . Then, at hourly intervals, histamine dissolved in Krebs-Henseleit solution was introduced into the tracheal segment for 15 min . The concentrations tested ranged from 0 .1 µM to 100 pM in log units. Treatment of data Secretion of the two radioisotopes were expressed as output rate in each sample (Bq min - l ) . As the secretory rates tended to diminish with time after the pulse label, we expressed the effectiveness of a stimulus in terms of the percentage increase over the output in unstimulated control periods (see Fig . 2) . In the case of vagal stimulation, which gave a discrete increase in the rate of output of radiolabelled macromolecules lasting only for the period of stimulation, the change was expressed relative to the mean rate of output in the two bracketing control periods ; with pilocarpine, where the effect of the stimulus lasted into the succeeding control period (Fig . 1), the increase was expressed relative to the single preceding control period . Statistical treatment of results Stimulated output was also expressed in proportion to the control output (i .e . stimulated output rate/control output rate) . These figures were log transformed to make their distribution more nearly normal and parametric statistical tests applied to the transformed data. For simplicity of presentation, however, results illustrated and quoted in the text are means and standard errors of percentage changes, untransformed. Where paired stimuli were compared (e .g. when the vagus nerves were
95
stimulated twice in the same cat) the resulting stimulations of radio-isotope output were compared by the paired t-test. When the stimulations produced by several concentrations of lignocaine or TTX were compared, an analysis of variance (ANOVA) was used to test the interaction between concentration and response ; Scheffe's test was then employed to test differences in response between two concentrations . 19
RESULTS Repeated stimulations in the absence of local anaesthesia In 4 experiments in which vagal stimulation and 5 µM pilocarpine were repeated in the absence of any local anaesthetic agent, the effects of the stimuli, expressed as change in stimulated secretory rates over those in the appropriate control periods, were not significantly different (Fig . 2) . This suggests the stimuli used gave reasonably repeatable effects and that it is valid to compare the results of stimuli given at the two different times employed in later experiments .
Vagal Stimulation WO 3 H-macromolecules
5
$ S-macromolecules
a 400 0 c s
U 200 C
100
T T
100-
a 0
Pilocarpine, 5µM 700
3H .macmmolecules
35
700
S-macromoleailes
aoI
500 -
C C
I
0400z
400 -
d
c 300-
300
s
U
200 -
200
100 -
100
W
C a
I
0
Fig. 2 - Bar graph summarising the percentage changes in radiolabelled mucus macromolecule outputs in response to vagal stimulation (see text) and administration of 5 µM pilocarpine in four experiments where these stimuli were repeated in the absence of local anaesthetic . Each bar represents the mean response ± s.e .m, n = 4 . Differences between first and second responses were small and did not approach statistical significance (P > 0 .2 in each case) .
96
Pulmonary Pharmacology
Lignocaine Effects of lignocaine on basal secretion
Each concentration of lignocaine stimulated the output of both 3 H- and 35 S-labelled macromolecules (n >_ 6 with each concentration) . These effects were concentration-related and the increases in the output of the two radioisotopes were approximately equal (Fig . 3), so the secretory response was clearly different from that to vagus nerve stimulation or pilocarpine where stimulation of 35 S predominated (see Discussion) . This stimulation by lignocaine was always greatest in the initial sample, and secretion normally returned to the pre-stimulus value by the second to fourth samples, despite the continued presence of lignocaine in the trachea . Washing out lignocaine with Krebs-Henseleit after 2 hours did not result in any clear change in secretory rate (Fig . 1) . The secretory response to two concentrations of lignocaine and to control Krebs-Henseleit were also tested in the presence of atropine (Fig . 3) . It is clear that the muscarinic agonist did not block the secre-
tory action of lignocaine, though it is possible that the effect of 43 mM lignocaine on the output of 3 Hlabelled macromolecules was somewhat diminished . The continued efficacy of lignocaine in the presence of atropine was verified by testing the PAS reactivity of samples in some of these experiments . With KrebsHenseleit control the increase in secretory rate was 0 .10+1 .53 µg/min ; with 13 mM lignocaine the increase was 34 .6 ± 7 .89 tg/min, and with 43 mM lignocaine the increase was 46 .7 + 13 .58 .tg/min (n = 4 in each case) . Effect of lignocaine on secretion induced by stimulation of vagus nerves
Stimulating the peripheral cut-ends of the vagus nerves in the absence of lignocaine increased the secretion of both 3 H- and 35 S-labelled macromolecules (A 3 H= +78±9 .5% ; A3-IS= +260±37 .9%, n = 16) . Figure 4 shows the effect of vagus nerve stimulation in the presence of different concentrations of lignocaine in the tracheal segment . Even the lowest concentration (4 .3 mM) abolished any statistically significant effect of nerve stimulation on output of radiolabelled macromolecules .
'H-macromolecules
The effect of lignocaine on secretion induced by pilocarpine
500 -
400 -
a 200
100
0
--mw t
0 400-
35
4 .3
13
43
13
43
S-macromolecules
0
4.3
Lignocaine Concentration : mM
Fig. 3 - Graphs showing the effect of introduction of lignocaine at different concentrations into the tracheal segment (bold lines, n >_ 6 for each point) . Upper panel shows the mean change in rate of 3 H-labelled macromolecule output+ s .e .m . ; lower panel shows the equivalent for 35 S-labelled macromolecules . Secretory effects depended on lignocaine concentration . In addition the dashed lines show the effects of two different concentrations of lignocaine in the presence of atropine, 1 mg/ml i .v . (n = 4 to 6 for each point) .
The concentration of pilocarpine used, 5 µM, had a clear but submaximal effect upon secretion (compare the results of 5 gM, Fig . 5, and 50 .tM pilocarpine, Table 2) . Increasing concentrations of lignocaine progressively lessened the responses of 35 S-macromolecules to pilocarpine stimulation (Fig . 5) . ANOVA showed the interaction between lignocaine concentration and 35 S-macromolecular response to pilocarpine was highly significant (F = 42 .4, p < 0 .0001) . Even the lowest concentration of lignocaine significantly lessened the agonist's effect (p
0952-0600/90/0003-0093/$10 .00
PULMONARY PHARMACOLOGY
The Effects of Local Anaesthetic Agents upon Mucus Secretion in the Feline Trachea In Vivo M . Somerville*§, J .-A. Karlssont & P .S . Richardson* *Department of Physiology, St George's Hospital Medical School, Cranmer Terrace, London SW17 ORE, UK, §Host Defence Unit, Cardiothoracic Institute, Dovehouse St, London SW3 6LY, UK, 'Department of Pharmacology, AB Draco, S-221 00 Lund, Sweden SUMMARY. The actions of lignocaine and tetrodotoxin (TTX) in a tracheal segment of the cat were tested on secretion of mucus macromolecules radiolabelled with 35S and 3 H. Lignocaine, 4.3-43 mM, given into the segment, caused a concentration dependent increase of secretion of 3H-and 35 S-labelled macromolecules. At 43 mM, lignocaine increased secretion : A 3H = +433+191%, A 35S = +327+34.5% (n=8). This effect lessened over 15-45 min . Atropine (1 mg/kg) had little effect on these responses. All concentrations of lignocaine tested (4.3-43 mM) abolished the effect of vagus nerve stimulation on secretion and diminished the effect of a submaximal concentration of pilocarpine (5 µM) in the segment in a dose-dependent manner . TTX in the segment did not alter the resting secretion . At 50 µM it abolished, and at 10 µM diminished, vagal control of secretion without affecting the secretory response to pilocarpine . The study shows that lignocaine, in concentrations which block vagal control of secretion (>_ 4.3 mM), stimulates the release of mucus macromolecules . Resting secretion is unaltered by TTX, and so does not appear to be under neurogenic inhibition . Larger concentrations of lignocaine (>_ 13 mM) also diminish pilocarpine-induced secretion, whereas TTX may inhibit nervous control of mucus secretion selectively . The results suggest that clinical anaesthesia of the airways with lignocaine may stimulate mucus secretion .
block all motor pathways controlling airway secretion while leaving direct mechanisms intact . The neurotoxin tetrodotoxin (TTX) has been used to study airway mucus secretion in vitro, 10-12 but not in intact animals . Similarly the effects of local anaesthetics on mucus secretion have scarcely been investigated, apart from one study showing lignocaine to stimulate secretion into the goose trachea ." The effect of lignocaine on mucociliary transport has been examined, but these studies have shown either no effect or an inhibition of transport rates . 14 The experiments reported here were done both to test whether interference with neural control of airway secretion by local anaesthetics in vivo would affect resting secretion and that stimulated by a muscarinic agonist, and also to see how lignocaine, a drug commonly used to render the airways anaesthetic in man, may alter secretion in an animal model of the human airway . This paper describes the effects of local anaesthetics on the resting and stimulated outputs of mucus macromolecules radiolabelled with 3 H and 35 S in the trachea of the anaesthetised cat . In the past these have been referred to as mucus glycoproteins or mucins," but recent evidence has shown that, though much of the radioactivity is in this form, some other macromolecules are also radiolabelled ." - " In this paper we refer to the radiolabelled materials in dialysed secretions as mucus macromolecules .
INTRODUCTION In the airways, the acinar glands of the submucosa and the goblet and ciliated cells of the surface epithelium secrete mucus macromolecules including glycoproteins .' These molecules constitute part of the periciliary fluid in which the cilia beat and also of the more highly visco-elastic layer of mucus which overlies the cilia and which is transported by them . Secretion of mucus macromolecules is involved in the mechanisms which defend the airways ; inhalations of irritants and some mediators have been shown to accelerate their release . 2 ' 3 Both sympathetic and parasympathetic nerves control airway mucus secretion in the cat, and antagonists to adrenoceptors and muscarinic receptors, respectively, inhibit the actions of these motor pathways . 4- ' More recently, however, it has become clear that some nerve-mediated effects survive the presence of large concentrations of muscarinic and adrenoceptor antagonists ; so non-adrenergic, non-cholinergic (NANC) nerves appear to form part of the secretomotor pathway ." - ' o It would be useful to have a drug to
Correspondence to : Dr P .S . Richardson, Department of Physiology, St . George's Hospital Medical School, London SWl7 ORE . 93
94
Pulmonary Pharmacology
METHODS Anaesthesia and preparation of cats Anaesthesia was induced in cats with pentobarbitone sodium, 42 mg/Kg, by intraperitoneal injection . Subsequent doses of pentobarbitone were given via a catheter into the femoral vein to sustain anaesthesia . Two cannulae were inserted into the cervical trachea to isolate a tracheal segment in situ, with nerve and blood supply intact, from which mucus could be flushed . The cat breathed through a third cannula, inserted into the trachea at its entry into the thorax, caudal to the isolated segment .' The tracheal segment was flushed with Krebs-Henseleit solution and then filled with 0 .5 mCi of 3H-glucose and 2 .0 mCi of sodium 35 S-sulphate in 4 .0 ml Krebs-Henseleit for 1 h to allow the secretory cells to take up these radiolabelled precursors and synthesise them into mucus macromolecules which would later be secreted into the tracheal lumen . At intervals in the course of the experiment, the tracheal segment was flushed with 10 ml Krebs-Henseleit solution . The tracheal washings were first mixed with guanidine hydrochloride, giving a 6M solution to dissolve mucus, and were then dialysed repeatedly against distilled water containing 0.1 % of both sodium azide and sodium sulphate to disperse radiolabel not covalently bound to macromolecules . After dialysis, samples were weighed and three 1 .0 ml aliquots from each were mixed with 10-12 ml of scintillation fluid (Scintran Cocktail T, BDH) and counted in a calibrated liquid scintillation spectrometer to determine the quantities of 35 S- and 3H-labels bound to macromolecules in each sample . In some experiments, where the effect of lignocaine was tested in the presence of atropine, three 1 .0 ml aliquots were also analysed for glycoconjugate content by the periodic acid Schiff (PAS) reaction ." The method was calibrated against standard concentrations of bovine submaxillary mucin (Sigma) and results are expressed in terms of this standard . Protocol The result from one experiment, involving lignocaine application to the trachea and which illustrates the protocol used, is illustrated in Figure 1 . After the period of radiolabelling had finished (time zero), mucus collections were made at intervals for 2 .25 h before any local anaesthetic was given . Then the first concentration of lignocaine hydrochloride (Astra), or control Krebs-Henseleit solution as in Figure 1, was given in Krebs-Henseleit for 2 h (Period 1), during which time the cut peripheral ends of the vagus nerves were stimulated (pulses of 10 V and 2 msec at 10 Hz, for 10 min of a 15 min collection period s and pilocarpine, 5 µM in Krebs-Henseleit, was given into the segment for another 15 min collection period . This concentration of pilocarpine was chosen to give a clear but submaximal stimulation of secretion (see Results) . In the cat trachea, pilocarpine's effect on
3H-macromolecules a C ou
PERIOD 1 (Control : Krebs-H)
$ 8-macromolecules
a
PERIOD 2 (4 .3 mid lignocaine)
C
Fig. 1 - Graphs showing the rate of output of 3 H-(upper panel) and 35 S-labelled macromolecules (lower panel) against time in one experiment . The graphs illustrate the design of experiments, with plain Krebs-Henseleit solution filling the tracheal segment for the first 2 .25 h of the experiment, then, over the next 2 .0 h, the first test period (PERIOD 1 ; in this case Krebs-Henseleit control), then 0.5 h of Krebs-Henseleit again, followed by the second test period (PERIOD 2 ; in this case 4.3 mM lignocaine in Krebs-Henseleit) . Finally there was a single control sample, for 15 min, with plain Krebs-Henseleit filling the segment again . During both Periods 1 and 2, secretion was stimulated where indicated by vagus nerve stimulation and 5 tM pilocarpine .
secretion is abolished by atropine ,4' 9 evidence that this agonist acts solely via muscarinic receptors . Lignocaine was then replaced with plain Krebs-Henseleit for 30 min before the next concentration of lignocaine, or Krebs-Henseleit control in some cases, was applied for 2 h (Period 2) and the vagal stimulation and pilocarpine treatment repeated . On two occasions the pH of Krebs-Henseleit and the most concentrated lignocaine hydrochloride solution (43 mM) were measured and found to be 7 .55 and 7 .60 for the former and 7 .35 and 7 .40 for the latter. In four experiments, repeated stimulations by pilocarpine and by electrical stimulation of the peripheral ends of the vagus nerves were given in the absence of lignocaine to test the reproducibility of the responses . In the remainder, three concentrations of lignocaine hydrochloride, 4 .3 mM, 13 .0 mM and 43 mM (i .e . 0.10%, 0 .30% and 1 .0% w/v lignocaine base respectively), were used and compared to Krebs-Henseleit buffer. Identical concentrations were not repeated in
Local Anaesthesia of the Airway and Mucus Secretion
the same experiment, but otherwise all possible combinations of concentrations were tested . In experiments where the effects of lignocaine were tested in the presence of atropine, the muscarinic antagonist was injected i.v ., 1 mg/kg body weight, 45 min before Period 1, and also given at a concentration of 1 .7 gM into the tracheal wash fluid for the rest of the experiment . These concentrations of atropine are sufficient to abolish the effect of 50 tM pilocarpine on airway secretion .' Two series of experiments concerned tetrodotoxin (TTX) . The initial concentration (10 µM) was chosen as being one order of . magnitude greater than the concentration necessary to block all secretory effects of field stimulation in human bronchi in vitro 12 (see Discussion) . A preliminary set was performed in which TTX (Sigma or Cambridge Bioscience), either 10 µM or 50 µM in Krebs-Henseleit, was given into the tracheal segment and the vagus nerves stimulated (see above) after 30 min exposure to this neurotoxin, or pilocarpine (50 µM) was given at the same stage in the experiment . The design of the second set resembled that for lignocaine with TTX (50 µM) alternating with Krebs-Henseleit in the two periods . In two cats, the action of histamine hydrochloride was tested on the secretion of radiolabelled macromolecules into the trachea . The trachea was prepared and radiolabelled as described above . Then, at hourly intervals, histamine dissolved in Krebs-Henseleit solution was introduced into the tracheal segment for 15 min . The concentrations tested ranged from 0 .1 µM to 100 pM in log units. Treatment of data Secretion of the two radioisotopes were expressed as output rate in each sample (Bq min - l ) . As the secretory rates tended to diminish with time after the pulse label, we expressed the effectiveness of a stimulus in terms of the percentage increase over the output in unstimulated control periods (see Fig . 2) . In the case of vagal stimulation, which gave a discrete increase in the rate of output of radiolabelled macromolecules lasting only for the period of stimulation, the change was expressed relative to the mean rate of output in the two bracketing control periods ; with pilocarpine, where the effect of the stimulus lasted into the succeeding control period (Fig . 1), the increase was expressed relative to the single preceding control period . Statistical treatment of results Stimulated output was also expressed in proportion to the control output (i .e . stimulated output rate/control output rate) . These figures were log transformed to make their distribution more nearly normal and parametric statistical tests applied to the transformed data. For simplicity of presentation, however, results illustrated and quoted in the text are means and standard errors of percentage changes, untransformed. Where paired stimuli were compared (e .g. when the vagus nerves were
95
stimulated twice in the same cat) the resulting stimulations of radio-isotope output were compared by the paired t-test. When the stimulations produced by several concentrations of lignocaine or TTX were compared, an analysis of variance (ANOVA) was used to test the interaction between concentration and response ; Scheffe's test was then employed to test differences in response between two concentrations . 19
RESULTS Repeated stimulations in the absence of local anaesthesia In 4 experiments in which vagal stimulation and 5 µM pilocarpine were repeated in the absence of any local anaesthetic agent, the effects of the stimuli, expressed as change in stimulated secretory rates over those in the appropriate control periods, were not significantly different (Fig . 2) . This suggests the stimuli used gave reasonably repeatable effects and that it is valid to compare the results of stimuli given at the two different times employed in later experiments .
Vagal Stimulation WO 3 H-macromolecules
5
$ S-macromolecules
a 400 0 c s
U 200 C
100
T T
100-
a 0
Pilocarpine, 5µM 700
3H .macmmolecules
35
700
S-macromoleailes
aoI
500 -
C C
I
0400z
400 -
d
c 300-
300
s
U
200 -
200
100 -
100
W
C a
I
0
Fig. 2 - Bar graph summarising the percentage changes in radiolabelled mucus macromolecule outputs in response to vagal stimulation (see text) and administration of 5 µM pilocarpine in four experiments where these stimuli were repeated in the absence of local anaesthetic . Each bar represents the mean response ± s.e .m, n = 4 . Differences between first and second responses were small and did not approach statistical significance (P > 0 .2 in each case) .
96
Pulmonary Pharmacology
Lignocaine Effects of lignocaine on basal secretion
Each concentration of lignocaine stimulated the output of both 3 H- and 35 S-labelled macromolecules (n >_ 6 with each concentration) . These effects were concentration-related and the increases in the output of the two radioisotopes were approximately equal (Fig . 3), so the secretory response was clearly different from that to vagus nerve stimulation or pilocarpine where stimulation of 35 S predominated (see Discussion) . This stimulation by lignocaine was always greatest in the initial sample, and secretion normally returned to the pre-stimulus value by the second to fourth samples, despite the continued presence of lignocaine in the trachea . Washing out lignocaine with Krebs-Henseleit after 2 hours did not result in any clear change in secretory rate (Fig . 1) . The secretory response to two concentrations of lignocaine and to control Krebs-Henseleit were also tested in the presence of atropine (Fig . 3) . It is clear that the muscarinic agonist did not block the secre-
tory action of lignocaine, though it is possible that the effect of 43 mM lignocaine on the output of 3 Hlabelled macromolecules was somewhat diminished . The continued efficacy of lignocaine in the presence of atropine was verified by testing the PAS reactivity of samples in some of these experiments . With KrebsHenseleit control the increase in secretory rate was 0 .10+1 .53 µg/min ; with 13 mM lignocaine the increase was 34 .6 ± 7 .89 tg/min, and with 43 mM lignocaine the increase was 46 .7 + 13 .58 .tg/min (n = 4 in each case) . Effect of lignocaine on secretion induced by stimulation of vagus nerves
Stimulating the peripheral cut-ends of the vagus nerves in the absence of lignocaine increased the secretion of both 3 H- and 35 S-labelled macromolecules (A 3 H= +78±9 .5% ; A3-IS= +260±37 .9%, n = 16) . Figure 4 shows the effect of vagus nerve stimulation in the presence of different concentrations of lignocaine in the tracheal segment . Even the lowest concentration (4 .3 mM) abolished any statistically significant effect of nerve stimulation on output of radiolabelled macromolecules .
'H-macromolecules
The effect of lignocaine on secretion induced by pilocarpine
500 -
400 -
a 200
100
0
--mw t
0 400-
35
4 .3
13
43
13
43
S-macromolecules
0
4.3
Lignocaine Concentration : mM
Fig. 3 - Graphs showing the effect of introduction of lignocaine at different concentrations into the tracheal segment (bold lines, n >_ 6 for each point) . Upper panel shows the mean change in rate of 3 H-labelled macromolecule output+ s .e .m . ; lower panel shows the equivalent for 35 S-labelled macromolecules . Secretory effects depended on lignocaine concentration . In addition the dashed lines show the effects of two different concentrations of lignocaine in the presence of atropine, 1 mg/ml i .v . (n = 4 to 6 for each point) .
The concentration of pilocarpine used, 5 µM, had a clear but submaximal effect upon secretion (compare the results of 5 gM, Fig . 5, and 50 .tM pilocarpine, Table 2) . Increasing concentrations of lignocaine progressively lessened the responses of 35 S-macromolecules to pilocarpine stimulation (Fig . 5) . ANOVA showed the interaction between lignocaine concentration and 35 S-macromolecular response to pilocarpine was highly significant (F = 42 .4, p < 0 .0001) . Even the lowest concentration of lignocaine significantly lessened the agonist's effect (p
