Control of pH of airway surface liquid of the ferret trachea in vitro H. KYLE, Department
J. P. T. WARD, AND J. G. WIDDICOMBE of Physiology, St. George’s Hospital Medical School, Londun S W17 ORE, United Kingdom
KYLE, H., J. P. T. WARD, AND J. G. WIDDICOMBE. Control of pH of airway surface liquid of the ferret trachea in vitro. J. Appl. Physiol. 68(l): 135440, 1990.-We measured the pH of airway surface liquid (ASL) secreted by the ferret trachea in vitro by using a catheter-tipped pH electrode implanted in a collecting cannula close to the airway epithelium. Mucus secretion was promoted by methacholine (0.02 mmol/l) in the organ bath. The pH of the ASL was 6.85 $- 0.03 (SE) compared with a bath value of 7.39 2 0.01, when the bath was bubbled with 5.65% COz. Changing the bath COz from 0 to 20.93% CO2 altered the bath pH from 8.06 to 6.96, but the ASL pH only varied from 6.92 to 6.85. This homeostasisof ASL pH wasnot the result of the buffering powers of the ASL, becauseex situ buffer curves for secretedASL were similar to those for KrebsHenseleitsolution. Changingthe luminal COzcontent by blowing gasesthrough the trachea changed ASL pH by values similar to that ex situ. However, when external organ bath CO2 waschanged,the luminal COz changeswereproportionately far smaller. Measurement of rates of diffusion of COz acrossthe tracheal wall indicated that this was not a limiting factor in the results. Similarly, measurementof metabolic rate CO2 production in the tracheal lumen indicated that this did not significantly affect the results. We conclude that the pH of ASL is significantly on the acid side of the pH of interstitial fluid and plasmaand that it is maintained relatively constant despite large changesin external PH. homeostasis;hydrogen ions; secretion
THE LIQUID in the airways consists of periciliary fluid with superimposed mucus from submucosal glands. The combination is sometimes called airway surface liquid
(ASL). The pH of the ASL is usually assumed to be close to that of interstitial fluid and plasma, namely 7.40. Measurements of pH of sputum and nasal secretions give a wide range of values (1,2,13,19), which is not surprising in view of possible contamination and conditioning factors. Measurements on tracheobronchial mucus and ASL collected directly are few. Guerrin et al. (5, 6), in extensive studies, found that the mean pH of tracheobronchial aspirates in humans was around 6.71-6.97 and that of rabbits 6.73. For human tracheal ASL, Mentz et al. (12) give a mean value of 6.88, whereas the rat is said to have a tracheal pH of 7.42-7.57 (4). By using the ferret isolated tracheal preparation to collect unadulterated ASL, we have shown that the pH is around 7.00, and that it is slightly decreased when secretagogues are applied to the preparation (17, 24). The present paper describes the effects of changes in luminal and external CO2 concentrations on mucus pH.
An abstract describing (10) .
the method
has been published
METHODS
Preparation of ferret whole trachea in vitro. Ferrets of either sex, weighing 0.5-1.5 kg, were anesthetized with pentobarbital sodium (Sagatal, May & Baker, 50 mg/kg ip). The trachea was exposed and cannulated just below the larynx with a Perspex collecting cannula. The animal was then killed by cardiac injection of pentobarbital (Euthatal, May & Baker). The chest was opened and the trachea cleared of tissue and cut just above the carina; the carinal end was cannulated. The trachea was mounted laryngeal end down in a water-jacketed organ bath at 37”C, bathed on the submucosal side with KrebsHenseleit solution at 37”C, and gassed with 95% 02-5% COZ. The ionic concentration (mmol/l) of the KrebsHenseleit solution was 145 Na+, 5.9 K+, 2.5 Ca2+, 1.2 Mg+, 126 Cl-, 26 HCOF, 1.2 H2P04, 1.2 SO:-, and 5.6 glucose and pH was 7.40 when gassed. ASL was carried by gravity and mucociliary transport to the lower cannula where it collected in the well of the cannula and could be periodically drawn into a polyethylene catheter inserted into the cannula to form an airtight seal. The method has been described and illustrated (10, 17, 24). pH electrodes. The pH-sensitive electrodes were of plastic dip-cast type, manufactured from 1-mm-OD, 0.5 mm-ID polyvinylchloride tubing 60 mm long (Esco, London) with a plug of porous ceramic in the tip as a former. These were dip-cast with a H+-selective ligand mix consisting of 1% tri-n-dodecylamine (20) and 0.06% potassium tetraphenylborate in polyvinylchloride (26.9%) and dioctyladipate (72.0%). The pH electrodes were filled with a phosphate buffer (pH 6.0) and saturated with AgCl, and the ligand junction reference electrodes with a solution consisting of 140 mM NaCl, 1 mM CaC12, and 4 mM KC1 were saturated with AgCl. The composition of the latter was designed to minimize the effects of any contamination into the small volume of fluid being measured. Ag-AgCl wires were made from Trimel-coated fine silver wire (0.5 mm, Johnson Mathey, London) with electrochemical deposition of AgCl. The electrodes were designed to fit into depressions drilled into the well of the mucus-collecting cannula to ensure that their tips were always covered with ASL. The recording system consisted of a purpose-built high-impedance unity gain amplifier connected to a chart
0161-7567/90 $1.50 Copyright 0 1990 the American Physiological society
135
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
136
CONTROL OF AIRWAY SURFACE LIQUID PH
recorder (Philips PM8251). The whole apparatus was screened to minimize electrical interference. The pH electrodes were calibrated in situ with commercial pH standard buffers (accurate to kO.02 pH, Electronic Instruments) and exhibited a linear response over the range tested (pH 6.00-9.00). Electrodes that showed a response of 48 mV/pH unit or an appreciable voltage drift were discarded. Calibrations were performed before and after each experiment; there was never any significant difference between the two calibration curves. ExperimentaL protocols. Because the resting secretion of the ASL was zero or close to zero, to ensure that the electrodes were covered with ASL, methacholine (0.02 mmol/l) was always present in the medium bathing the trachea. This produced a continuous slow secretion from submucosal glands that lasted many hours. To prevent excessive buildup of ASL, samples were withdrawn at the end of every 30 min throughout the experiment; this also allowed us to verify that there was ASL on the electrodes. Under control conditions the Krebs-Henseleit solution in the organ bath was bubbled with 5.65% COn in 02. During experimental periods the Krebs-Henseleit solution in the organ bath was replaced with stock KrebsHenseleit solution that had been bubbled vigorously for at least 1 h before use with gases containing different proportions of 02 and COS: 100% 02, 5.65% CO2 in 02 (both from gas cylinders), and -10% C02-90% 02 and 20% CO&30% O2 (both mixtures in bags). Bubbling with the appropriate gas was continued in the organ bath. The ASL pH was measured continuously throughout the experiment. Unless otherwise stated there were two 30. min control periods and experimental periods were 30 min long. These periods were chosen to allow adequate time for stabilization of any pH changes in the ASL and of luminal changes in C02% (see RESULTS and Table 3). All values in RESULTS apply to the ends of the 30-min periods. Buffer curves for Krebs-Henseleit solution were obtained by taking the solution from the organ bath at the end of 30-min test periods, i.e., when it had been bubbled with a known gas mixture for at least 1.5 h. The pH electrodes used for the corresponding tracheal experiment were placed in the solution to measure pH. Buffer curves for ex situ ASL were obtained by leaving the pH electrodes in the collecting cannula in an empty organ bath, filling the well of the collecting cannula with ASL that had been previously collected, and passing different gas mixtures through the organ bath and over the ASL and cannula until steady pH readings were obtained. To induce luminal changes in COZ, gas mixtures were humidified by passing them through a chamber containing a soaked porous pot at 37°C and via a thermally insulated catheter through the tracheal collecting cannula. The tap at the top of the upper cannula was opened to allow the gas to escape. The pH of the ASL was measured continuously, and the ASL samples were not removed from the electrodes throughout this experiment. The gases were administered in random sequence for 30 min, adequate time for equilibration (see Table 3). Four to six gas mixtures were used in each experiment. The
external medium was always gassed with 5% C02-95% OZ. Values in RESULTS apply to the ends of the 309min periods. To estimate the CO2 content of luminal gas (when it was not being experimentally changed), samples were collected by opening the tap on the upper cannula to allow the luminal gas and air (total volume ~5 ml) to be drawn into a syringe via a small length of catheter inserted into the lower collecting cannufa. The syringe was then capped and the COZ content of the sample measured (Beckman LB2 gas analyzer). From estimates of the tracheal volume, calculated from the tracheal dimensions, plus the volume of the upper cannula, the dilution of the luminal sample was calculated. The length and diameter of the trachea were measured to the nearest 0.5 mm at the end of each experiment and gave volumes in the range 1.5-2.0 ml. The accuracy of the gas measurements from the tracheal lumen was calculated from a model trachea. A piece of rubber tubing of similar size and volume to a trachea was mounted on the collecting cannula in the organ bath with an upper cannula attached. The model was filled with a known gas and sealed by closing the taps on the upper and lower cannulas, and the measuring procedure followed. Twenty-five estimations gave a consistent underestimate of -11.65 t 0.07% (SE). A correction has not been applied to the results. RESULTS Submucosal changes in COZ. Figure 1 shows the mean buffer curve for ASL, in the collecting cannula but without a trachea present, exposed to gases of different CO2 content (see METHODS). For comparison the curve for Krebs-Henseleit solution bubbled with gases of different COZ contents in the organ bath is also shown. At high CO2 values the curves are similar, but at low CO2 values the curve for ASL is relatively more acid. Table 1 and Fig. 1 show the changes in ASL pH in situ when the bath was bubbled with different concentrations of COs, from the same experiments that gave the curve for Krebs-Henseleit solution. The ASL pH was always more acid and varied little (means from 6.85 to 6.92) compared with bath pH (means from 6.96 to 8.06). The luminal CO2 contents, initially ~5.65%, changed in the direction of the external COZ content but did not reach external values (Table 1). Table 2 shows the effect of changes in COZ content of gas bubbling the bath against ASL and bath pH and luminal COa. The changes in bath pH were always significant and in the direction expected and of appropriate size for COa changes in Krebs-Henseleit solution (Fig. 1). The mean changes in ASL pH were usually insignificant and always far smaller than the changes in bath pH. The percent changes in luminal COa were usually significant and in the expected directions but far smaller than the percent changes in external COZ. Luminal changes in COZ. To test the effect of luminal COZ on ASL pH, the bath was bubbled with 5.65% COn throughout and different gases were blown continually through the trachea, with step changes in COZ content. Figure 2 shows the relationship of the two variables,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
CONTROL
OF AIRWAY
80 l
SURFACE
LIQUID
___+__
ML ex situ
__O__
Krebs-Hmseleit
--W--
ASL in situ
137
PH
FIG. 1. A combined graph of changes in ASL pH (n = 5-51) and Krebs-Henseleit pH (n = 551) when gases with different %COz are bubbled through Krebs-Henseleit solution (Table 1). For comparison, CC&pH curve for previously collected ASL, determined without trachea present, is given (12= 3-21). Values are means t SE.
6.0
1
1
I
0
10
20
1. Measurement of ASL pH, organ bath pH, and luminal %C02 when Krebs-Henseleit solution surrounding submucosal surface of trachea is gassed with gasesof different CO2 content TABLE
. Lummal
Bath COz, % ASL
0.0 ~- 6.92&0.05* ( 11) 5.65 6.85zkO.O3* (51) 10.04kO.87 6.85+0.07* (5) 20.93kO.45 6.87t0.06 (10) Values are means 2 SE of no. theses. ASL, airway surface liquid. bath pH.
COz, %
Bath
8.06zkO.07 (8) 4.55AO.30 (10) 7.39kO.01 (45) 5.58k0.18 (47) 7.21kO.07 (5) 6.3320.47 (5) 8.0920.54 (9) 6.96kO.02 (7) of observations indicated in paren* P < 0.05, ASL pH compared with
which is generally similar to the CO,-pH curves for Krebs-Henseleit solution and for secreted ASL without the trachea present (Fig. 1). Figure 2 also includes the corresponding values for ASL pH and luminal COa when the external COa was changed (Table 1); there is little change in ASL pH with COZ, and the points are displaced considerably to the acid side of the curve based on values derived from directly changing luminal COZ. TABLE
2. Changes in ASL pH, organ bath pH, and luminal A Bath COz, %
5.65 ---) 0.00 0.00 + 5.65 5.65 ---) 10.04
%COz with changes in %CO, in organ bath
APH
A Luminal
ASL
Bath
+0.05~0.05 (11) -0.06z!z0.02* (11)
(8) (8) -0.18-t-0.05* (5) +O. 18-10.04t (5) -0.44+0.02t (7) +0.4O~o.olt (7)
-0*01t0*01
+0.02kO.03 0.00~0.03 5.65 + 20.93 20.93 -3 5.65 +0.04&0.02* Values are means t SE of no. of observations indicated 10.04 ---) 5.65
Table 3 gives the half times for changes in ASL pH caused by step changes in luminal COz content. The half times for responses to increasing CO2 content were always considerably shorter than those as the result of decreasing COz content. CO2 diffusion from tracheal lumen. To test the rate of diffusion of CO2 across the tracheal wall, in three preparations the trachea was initially filled with 21.37% CO2 in O2 and the COz remaining at various intervals was estimated. The external buffer was bubbled with 5% CO2 in Oz. Figure 3 shows that equilibrium was 87% complete in 30 min and complete within 60 min. Replotting Fig. 3 logarithmically shows that the initial decrease in CO2 percentage was considerably faster than one would expect from an exponential relationship. Because the luminal CO2 content might be affected by CO2 production by the mucosal tissue, in six experiments the lumen was initially filled with 5.16% CO2 in 02 while the external buffer was bubbled with the same mixture. The trachea was kept closed. After various intervals the luminal COz percentage was measured (Fig. 3). The luminal CO2 increased to 6.81 t 0.45% after 30 min and to 8.43 -$-0.98% after 120 min.
(5)
(5) (10) (10)
+0.66+0.07t
-0.64+0.08t
Cog, %
-1.lOk0.26t (10) -0.05~0.08 (10) +0*9340*21t
(5)
-0.58-c-0.12-f (5) +3*03+0*31t
(9)
-2.0720.18t (6) * P < 0.05, t P < 0.01 for change in value. in parentheses. ASL, airway surface liquid.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
138
CONTROL
OF AIRWAY
SURFACE
LIQUID
PH
80 *
-o0
75.
co2 chanp in lumen
CO2 changein bath
FIG. 2. Measurement of ASL pH with changes in %COz when gas is passed through lumen of the trachea (n = 3-15). Also shown are values for luminal %COz and ASL pH when gases of different %COz were bubbled through external Krebs-Henseleit solution (n = 5-47, Table 1). Values are means & SE.
X
ocr
70.
1
I
0
10
1
20
co20
TABLE 3. Half-time changes in ASL pH caused by passing gases of different %COa through tracheal lumen
trations. However, the position of the curves could not explain the relative acidity of the ASL, especially at low CO2 concentrations. In this paper ail results have been Change in COZ,% n Half Time, xnin presented and analyzed statistically as pH values, al0.00 + 5.82 though strictly one should not use normal distribution 7 1.01*0.21t 5.82 + 0.00 8 6.44*2.30* statistics for pH, which is a logarithmic function. How0.00 --+ 10.65 5 1.2O,tO.33* ever, it is conventional to do so, as in nearly all the 4 10.58 --) 0.00 9.88&4.02* papers quoted here. Converted to fH+] our results give a 0.00 + 17.24 3 0.90*0.55 value of 140 nmol/l for a pH of 6.85, over three times 17.24 -+ 0.00 3 3.73k3.33 Values are means + SE. 0.00% COz, pure 0,; 582% COz, from a gas greater than the normal value for plasma and interstitial cylinder; 10.65 k 0.65, 10.58 k 0.92, and 17.24 2 0.47% COZ, from gas fluid (40 nmol/l for a pH of 7.40). Neither body weight, mixtures. ASL, airway surface liquid. * P c 0.05; P P < 0.01. age, nor sex of the ferrets appeared to influence the results. rHSCUSSIUN Because the changes in pH of ASL when external CO2 concentration was raised were usually insignificantly The liquid in the trachea comprises periciliary fluid small, it was impossible to determine their time constant and mucus secreted from submucosal glands, sometimes after a step change in CO2 is introduced in the external referred to as “sol” and “gel” (24). The ferret trachea has buffer. However, changes in luminal CO2 as the result of relatively few goblet or serous cells (18), which presumoutward diffusion were 87% complete in 30 min, and ably contribute to the ASL. Although the mucoglycoprotein content of glandular mucus is higher than that of even if the time constant for inward diffusion of CO2 were greater, one would have expected clear and signifipericiliary fluid, small solutes can presumably equilibrate cant changes in pH of ASL within 30 min. Resting between the two. Thus the classical distinction between preparations showed only small changes in luminal CO2 the two main components is not clear. In our experiments percentage over the same time period (Fig. 3), changes mucus secretion from submucosal glands was promoted by methacholine to ensure enough ASL to cover the pH presumably as the result of metabolic production of COZ. Measurement of luminal CO2 percentages when the electrodes and allow measurements. external buffer was bubbled with different CO2 concenThe ASL is significantly more acid than the surroundtrations showed that these changed far less than one ing Krebs-Henseleit solution or normal interstitial fluid. would expect from the external values. Under these The low pH of ASL is not the result of the particular buffering powers of the secreted mucus, because both the conditions the luminal CO2 relationship to ASL pH was quite different from that when COZ-rich gas flowed C02-pH curve for secreted ASL with the trachea absent (Fig. 1) and that for ASL in situ with gas being passed through the trachea. In the latter instance the gas flow through the trachea (Fig. 2) gave very similar curves to emerging close to the pH-sensitive electrode presumably that of Krebs-Henseleit solution (Fig. 1); the former caused a C02-pH relationship as if the trachea had not curves were on the more acid side for high CO2 concenbeen present. Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
CONTROL
OF AIRWAY
SURFACE
LIQUID
PH
139
FIG. 3. Changes in luminal %CO with time. l , Lumen initially filled with 21.37% COz (r2 = 3); 0, lumen initially filled with 5.16% CO2 (n = 6).
0
0
50
Time (min) To ensure that the pH-sensitive electrodes were covered with ASL, submucosal gland secretion was promoted by a small maintained dose of methacholine. The pH of unadulterated gland secretion is not known. However, the control pH of ASL in the same model was 7.12 t 0.03 for nonstimulated secretions and was reduced to 6.97 t 0.56 wh en gland secretion was stimulated by methacholine (17); the latter value is close to those described in the present paper, Thus, although methacholine may have slightly reduced the pH of ASL, it is unlikely to have modified the main results. Published values for the pH of ASL and secreted mucus vary quite widely, presumably because of very different experimental methods of collecting mucus, the possibilities of evaporation and contamination with other liquids, and the fact that some studies with humans were on patients with respiratory disease. Thus the pH of sputum has been measured and recorded as between 6.6 (19) and 8.2 (1). Secretions from the lower respiratory tract give values from 5.9 (22) to 8.12 (9). One of the most extensive studies used in vivo measurements with a pH electrode mounted in a bronchoscope and lowered onto the mucosal surface of the trachea of healthy humans and rabbits (5). Values were 6.73 t 0.56 for the rabbit, 6.5-7.5 for healthy humans, and 6.1-7.9 for tracheostomized patients. Similar values for the rat trachea are 7.42-7,57 (4). However, electrodes placed in contact with the airway epithelium may cause local damage and thereby distort the mucus pH. By using a micropipette method, Nielson et al. (14, 15) obtained a value of 6.92 -+ 0.04 for the pH in the lining surface liquid of the alveoli of rabbits. Mentz et al. (12) collected ASL from healthy humans and found a value of 6.88 t 0.08 for trachea and 6.77 t 0.10 for bronchi. These values should
be compared with those of Guerrin et al. (6), who reported means of 6.71 -t 6.97 for the human trachea. Thus our results support the majority of publications indicating a pH well on the acid side of plasma and interstitial fluid. The pH of ASL is not only acid compared with interstitial fluid but the acidity is maintained despite large differences in external pH values. A similar result was obtained by Nielson et al. (15), who showed that the surface liquid of the alveoli of rabbits was little changed in pH in response to changes in external CO2 concentration and resultant pH alterations. Thus a homeostatic mechanism seems to apply to the ferret trachea as well as to the rabbit alveoli. The basis of this homeostasis is not clear. Mucus can act as a passive barrier to H+ diffusion (7, 11). The mucus may act as an ampholyte and thereby restrict H+ movement because of its properties as a proton donor (21). However, our results indicate that there may be an active control of the pH of ASL in addition to any passive property of secreted mucus as an ampholyte. Presumably this active control would be based on epithelial secretion of H+/HCO& and the secretory mechanism would be regulated by H+/COz concentrations at the epithelial cell membranes. To test this hypothesis one needs to study the transport systems on the apical and basolateral membranes. If ASL is maintained acid relative to interstitial fluid by a homeostatic mechanism, the functions and results of such a mechanism have to be considered. Speculative possibilities include the following: 1) there may be an optimal pH to limit bacterial adhesion and growth; 2) there will be optimal pH values for the activity of luminal enzymes and for the physicochemical properties of luminal macromolecules such as mucins; 3) release of mediators from mast cells depends on pH, being inhibited
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
140
CONTROL
OF AIRWAY
at values ~6.8 (2, 23); 4) ciliary function might be affected by pH, although published results suggest that the cilia are active over a wide range (8, 11); 5) luminal PH might affect the beha vior of migratory cells. Many of these functions have been studied, but there is little information in the literature as to their optimal pH values or the effect of changes in pH. If ASL is a barrier to H+ diffusion, then the abrupt changes in luminal CO2 percentage (from 0 to -5.5% in normal tidal breathing) might be effectively buffered from having any chemical effect on the epithelium and deeper tissues. This and other possibilities need exploration. We are grateful to Dr. S. E. Webber for valuable discussion and help with the manuscript. H. Kyle was supported by Karl Thomae. Present address of J. P. T. Ward, Depts. of Medicine and Physiology, St. Thomas’ Hospital Medical School, Lambeth Palace Rd., London SE1 7EH, UK. Address for reprint requests: J. G. Widdicombe, Dept. of Physiology, St. George’s Hospital Medical School, Cranmer Terrace, London SW17 ORE, WK. Received 10 April 1989; accepted in final form 24 August 1989.
SURFACE
9.
10.
11. 12.
13.
14. 15. 16.
REFERENCES 1. BASCH, F,, P. HOLINGER, AND H. G, PONCHER. Physical and chemical properties of sputum. 1, Factors determining variations in portions from different parts of the tracheobronchial tree. Am. J. Dis. Child. 62: 981-990,1941. 2. DIAMANT, B. Histamine release elicited by extracts of Ascaris suis-influence of oxygen lack and glucose. Acta Physiol. Stand. 52: 8-22,196l. 3. FABRICANT, N. D. Significance of the pH of nasal secretions in situ. Arch, Otolaryngol. 34: 297-301, 1941. 4. GATTO, L. A. pH of mucus in rat trachea. J. Appl. Physiol. 50: 1224-1226,198l. 5. GUERRIN, F., C. VOISIN, V. MACQUET, H. ROBIN, AND P. LEQUIEN. Apport de la pH metrie bronchique in situ. Progr. Respir. Res. 6: 372-383,197l. 6. GUERRIN, F., C. VOISIN, V. MACQUET, H. ROBIN, F. WATTEL, AND J. P. BOULANGER. Possibilites de la pH metrie bronchique in situ. In: Hypersecretion Brunchique. Clinchy: Poinsot, 1969, vol. 1, p. 249-256. 7. HOLMA, B, Influence of buffer capacity and pH-dependent rheological properties of respiratory mucus on health effects due to acidic pollution. Sci. Total Environ. 41: 101-123, 1985. 8. HOLMA, B., M. LINDCREN, AND J. MORKHOLDT ANDERSEN. pH
17. 18. 19. 20,
21.
22, 23. 24.
LIQUID
PH
effects on ciliomotility and morphology of respiratory mucus. Arch. Environ. Health 32: 216-226, 1977. KWART, H., W. W. MOSELEY, AND M. KATZ. The chemical characterization of human tracheobronchial secretion: a possible clue to the origin of fibrocystic mucus. Ann. NY Acad. Sci. 106: 709721,1963. KYLE, H., N. P. ROBINSON, J. P. T. WARD, AND J. G. WIDDICOMBE. Measurement of mucus [H+] and other ionic concentrations in the isolated trachea of the ferret (Abstract). J. Physiol. Land. 387: 15P, 1987. LUK, C. K. A., AND M. J. DULFANO. Effect of pH, viscosity and ionic-strength changes on ciliary beating frequency of human bronchial explants. Clin. Sci. Lund. 64: 449-451, 1983. MENTZ, W. M., M. R. KNOWLES, J. B. BROWN, J. T. GATZY, AND R. C. BOUCHER. Measurement of airway surface liquid (ASL) composition of normal human subjects (Abstract). Am. Rev. Respir. Dis. 129: 315, 1984. MITTEMAIER, R. Untersuchung uber die Wasserstoffronenkonzentration an Sekreten und Schlermhauten im besonderen bei chronischen Nebenhohlenkrankungen. Arch. Ohren-Nusen-KehZkopfheiZkd. 127: 1, 1930. NIELSON, D. W. Electrolyte composition of pulmonary alveolar subphase in anesthetized rabbits. J. Appl. Physiol. 60: 972-979, 1986. NIELSON, D. W., J. GOERKE, AND J. A. CLEMENTS. Alveolar subphase pH in the lungs of anesthetized rabbits. Proc. Natl. Acud. Sci. USA 78: 7119-7123,198l. PFEIFFER, C. J. Experimental analysis of hydrogen ion diffusion in gastrointestinal mucus glycoprotein. Am. J. Physiol. 240 (Gastrointest. Liver Physiol. 3): G176-G182, 1981. ROBINSON, N. P., H. KYLE, S. E. WEBBER, AND J. G. WIDDICOMBE. Electrolyte and other chemical concentrations in tracheal airway surface liquid and mucus. J. Appl. Physiol. 66: 2129-2135, 1989. ROBINSON, N. P., L. VENNING, H. KYLE, AND J. G. WIDDICOMBE. Quantitation of the secretory cells of the ferret tracheobronchial tree. J. Anut. 145: 173-188, 1986. RYLEY, H. C., AND T. D. BROGAN. Variation in the composition of sputum in chronic chest disease. Br. J. Exp. Pathol. 49: 625-633, 1968. SCHULTHESS, P., Y. SHIJO, H. V. PHAM, E. PRETCH, D. AMMANN, AND W. SIMAON. A hydrogen ion-selective liquid membrane electrode based on tri-n-dodecylamine as neutral carrier. Anal. Chim, Acta 131: ill-116,198l. SHIAU, Y.-F,, P. FERNANDEZ, M. J. JACKSON, AND S. McMONAGLE. Mechanisms maintaining a low-pH microclimate in the intestine. Am. J. Physiol. 248 (Gastrointest. Liver Physiol. 11): GBOB-G617,1985. STEINMANN, E. La secretion bronchique et le pH. Branches 6: 126129,1965. UVNAS, B., B. DIAMANT, B. HOGBERG, AND I, T. THON. Mechanism of mast cell disruption induced by a principle extracted from Ascaris suis. Am. J. Physiol. 199: 575-577, 1960. WIDDICOMBE, J. G. Airway mucus. Eur. Respir. J. 2: 107-115,1989.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
J. P. T. WARD, AND J. G. WIDDICOMBE of Physiology, St. George’s Hospital Medical School, Londun S W17 ORE, United Kingdom
KYLE, H., J. P. T. WARD, AND J. G. WIDDICOMBE. Control of pH of airway surface liquid of the ferret trachea in vitro. J. Appl. Physiol. 68(l): 135440, 1990.-We measured the pH of airway surface liquid (ASL) secreted by the ferret trachea in vitro by using a catheter-tipped pH electrode implanted in a collecting cannula close to the airway epithelium. Mucus secretion was promoted by methacholine (0.02 mmol/l) in the organ bath. The pH of the ASL was 6.85 $- 0.03 (SE) compared with a bath value of 7.39 2 0.01, when the bath was bubbled with 5.65% COz. Changing the bath COz from 0 to 20.93% CO2 altered the bath pH from 8.06 to 6.96, but the ASL pH only varied from 6.92 to 6.85. This homeostasisof ASL pH wasnot the result of the buffering powers of the ASL, becauseex situ buffer curves for secretedASL were similar to those for KrebsHenseleitsolution. Changingthe luminal COzcontent by blowing gasesthrough the trachea changed ASL pH by values similar to that ex situ. However, when external organ bath CO2 waschanged,the luminal COz changeswereproportionately far smaller. Measurement of rates of diffusion of COz acrossthe tracheal wall indicated that this was not a limiting factor in the results. Similarly, measurementof metabolic rate CO2 production in the tracheal lumen indicated that this did not significantly affect the results. We conclude that the pH of ASL is significantly on the acid side of the pH of interstitial fluid and plasmaand that it is maintained relatively constant despite large changesin external PH. homeostasis;hydrogen ions; secretion
THE LIQUID in the airways consists of periciliary fluid with superimposed mucus from submucosal glands. The combination is sometimes called airway surface liquid
(ASL). The pH of the ASL is usually assumed to be close to that of interstitial fluid and plasma, namely 7.40. Measurements of pH of sputum and nasal secretions give a wide range of values (1,2,13,19), which is not surprising in view of possible contamination and conditioning factors. Measurements on tracheobronchial mucus and ASL collected directly are few. Guerrin et al. (5, 6), in extensive studies, found that the mean pH of tracheobronchial aspirates in humans was around 6.71-6.97 and that of rabbits 6.73. For human tracheal ASL, Mentz et al. (12) give a mean value of 6.88, whereas the rat is said to have a tracheal pH of 7.42-7.57 (4). By using the ferret isolated tracheal preparation to collect unadulterated ASL, we have shown that the pH is around 7.00, and that it is slightly decreased when secretagogues are applied to the preparation (17, 24). The present paper describes the effects of changes in luminal and external CO2 concentrations on mucus pH.
An abstract describing (10) .
the method
has been published
METHODS
Preparation of ferret whole trachea in vitro. Ferrets of either sex, weighing 0.5-1.5 kg, were anesthetized with pentobarbital sodium (Sagatal, May & Baker, 50 mg/kg ip). The trachea was exposed and cannulated just below the larynx with a Perspex collecting cannula. The animal was then killed by cardiac injection of pentobarbital (Euthatal, May & Baker). The chest was opened and the trachea cleared of tissue and cut just above the carina; the carinal end was cannulated. The trachea was mounted laryngeal end down in a water-jacketed organ bath at 37”C, bathed on the submucosal side with KrebsHenseleit solution at 37”C, and gassed with 95% 02-5% COZ. The ionic concentration (mmol/l) of the KrebsHenseleit solution was 145 Na+, 5.9 K+, 2.5 Ca2+, 1.2 Mg+, 126 Cl-, 26 HCOF, 1.2 H2P04, 1.2 SO:-, and 5.6 glucose and pH was 7.40 when gassed. ASL was carried by gravity and mucociliary transport to the lower cannula where it collected in the well of the cannula and could be periodically drawn into a polyethylene catheter inserted into the cannula to form an airtight seal. The method has been described and illustrated (10, 17, 24). pH electrodes. The pH-sensitive electrodes were of plastic dip-cast type, manufactured from 1-mm-OD, 0.5 mm-ID polyvinylchloride tubing 60 mm long (Esco, London) with a plug of porous ceramic in the tip as a former. These were dip-cast with a H+-selective ligand mix consisting of 1% tri-n-dodecylamine (20) and 0.06% potassium tetraphenylborate in polyvinylchloride (26.9%) and dioctyladipate (72.0%). The pH electrodes were filled with a phosphate buffer (pH 6.0) and saturated with AgCl, and the ligand junction reference electrodes with a solution consisting of 140 mM NaCl, 1 mM CaC12, and 4 mM KC1 were saturated with AgCl. The composition of the latter was designed to minimize the effects of any contamination into the small volume of fluid being measured. Ag-AgCl wires were made from Trimel-coated fine silver wire (0.5 mm, Johnson Mathey, London) with electrochemical deposition of AgCl. The electrodes were designed to fit into depressions drilled into the well of the mucus-collecting cannula to ensure that their tips were always covered with ASL. The recording system consisted of a purpose-built high-impedance unity gain amplifier connected to a chart
0161-7567/90 $1.50 Copyright 0 1990 the American Physiological society
135
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
136
CONTROL OF AIRWAY SURFACE LIQUID PH
recorder (Philips PM8251). The whole apparatus was screened to minimize electrical interference. The pH electrodes were calibrated in situ with commercial pH standard buffers (accurate to kO.02 pH, Electronic Instruments) and exhibited a linear response over the range tested (pH 6.00-9.00). Electrodes that showed a response of 48 mV/pH unit or an appreciable voltage drift were discarded. Calibrations were performed before and after each experiment; there was never any significant difference between the two calibration curves. ExperimentaL protocols. Because the resting secretion of the ASL was zero or close to zero, to ensure that the electrodes were covered with ASL, methacholine (0.02 mmol/l) was always present in the medium bathing the trachea. This produced a continuous slow secretion from submucosal glands that lasted many hours. To prevent excessive buildup of ASL, samples were withdrawn at the end of every 30 min throughout the experiment; this also allowed us to verify that there was ASL on the electrodes. Under control conditions the Krebs-Henseleit solution in the organ bath was bubbled with 5.65% COn in 02. During experimental periods the Krebs-Henseleit solution in the organ bath was replaced with stock KrebsHenseleit solution that had been bubbled vigorously for at least 1 h before use with gases containing different proportions of 02 and COS: 100% 02, 5.65% CO2 in 02 (both from gas cylinders), and -10% C02-90% 02 and 20% CO&30% O2 (both mixtures in bags). Bubbling with the appropriate gas was continued in the organ bath. The ASL pH was measured continuously throughout the experiment. Unless otherwise stated there were two 30. min control periods and experimental periods were 30 min long. These periods were chosen to allow adequate time for stabilization of any pH changes in the ASL and of luminal changes in C02% (see RESULTS and Table 3). All values in RESULTS apply to the ends of the 30-min periods. Buffer curves for Krebs-Henseleit solution were obtained by taking the solution from the organ bath at the end of 30-min test periods, i.e., when it had been bubbled with a known gas mixture for at least 1.5 h. The pH electrodes used for the corresponding tracheal experiment were placed in the solution to measure pH. Buffer curves for ex situ ASL were obtained by leaving the pH electrodes in the collecting cannula in an empty organ bath, filling the well of the collecting cannula with ASL that had been previously collected, and passing different gas mixtures through the organ bath and over the ASL and cannula until steady pH readings were obtained. To induce luminal changes in COZ, gas mixtures were humidified by passing them through a chamber containing a soaked porous pot at 37°C and via a thermally insulated catheter through the tracheal collecting cannula. The tap at the top of the upper cannula was opened to allow the gas to escape. The pH of the ASL was measured continuously, and the ASL samples were not removed from the electrodes throughout this experiment. The gases were administered in random sequence for 30 min, adequate time for equilibration (see Table 3). Four to six gas mixtures were used in each experiment. The
external medium was always gassed with 5% C02-95% OZ. Values in RESULTS apply to the ends of the 309min periods. To estimate the CO2 content of luminal gas (when it was not being experimentally changed), samples were collected by opening the tap on the upper cannula to allow the luminal gas and air (total volume ~5 ml) to be drawn into a syringe via a small length of catheter inserted into the lower collecting cannufa. The syringe was then capped and the COZ content of the sample measured (Beckman LB2 gas analyzer). From estimates of the tracheal volume, calculated from the tracheal dimensions, plus the volume of the upper cannula, the dilution of the luminal sample was calculated. The length and diameter of the trachea were measured to the nearest 0.5 mm at the end of each experiment and gave volumes in the range 1.5-2.0 ml. The accuracy of the gas measurements from the tracheal lumen was calculated from a model trachea. A piece of rubber tubing of similar size and volume to a trachea was mounted on the collecting cannula in the organ bath with an upper cannula attached. The model was filled with a known gas and sealed by closing the taps on the upper and lower cannulas, and the measuring procedure followed. Twenty-five estimations gave a consistent underestimate of -11.65 t 0.07% (SE). A correction has not been applied to the results. RESULTS Submucosal changes in COZ. Figure 1 shows the mean buffer curve for ASL, in the collecting cannula but without a trachea present, exposed to gases of different CO2 content (see METHODS). For comparison the curve for Krebs-Henseleit solution bubbled with gases of different COZ contents in the organ bath is also shown. At high CO2 values the curves are similar, but at low CO2 values the curve for ASL is relatively more acid. Table 1 and Fig. 1 show the changes in ASL pH in situ when the bath was bubbled with different concentrations of COs, from the same experiments that gave the curve for Krebs-Henseleit solution. The ASL pH was always more acid and varied little (means from 6.85 to 6.92) compared with bath pH (means from 6.96 to 8.06). The luminal CO2 contents, initially ~5.65%, changed in the direction of the external COZ content but did not reach external values (Table 1). Table 2 shows the effect of changes in COZ content of gas bubbling the bath against ASL and bath pH and luminal COa. The changes in bath pH were always significant and in the direction expected and of appropriate size for COa changes in Krebs-Henseleit solution (Fig. 1). The mean changes in ASL pH were usually insignificant and always far smaller than the changes in bath pH. The percent changes in luminal COa were usually significant and in the expected directions but far smaller than the percent changes in external COZ. Luminal changes in COZ. To test the effect of luminal COZ on ASL pH, the bath was bubbled with 5.65% COn throughout and different gases were blown continually through the trachea, with step changes in COZ content. Figure 2 shows the relationship of the two variables,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
CONTROL
OF AIRWAY
80 l
SURFACE
LIQUID
___+__
ML ex situ
__O__
Krebs-Hmseleit
--W--
ASL in situ
137
PH
FIG. 1. A combined graph of changes in ASL pH (n = 5-51) and Krebs-Henseleit pH (n = 551) when gases with different %COz are bubbled through Krebs-Henseleit solution (Table 1). For comparison, CC&pH curve for previously collected ASL, determined without trachea present, is given (12= 3-21). Values are means t SE.
6.0
1
1
I
0
10
20
1. Measurement of ASL pH, organ bath pH, and luminal %C02 when Krebs-Henseleit solution surrounding submucosal surface of trachea is gassed with gasesof different CO2 content TABLE
. Lummal
Bath COz, % ASL
0.0 ~- 6.92&0.05* ( 11) 5.65 6.85zkO.O3* (51) 10.04kO.87 6.85+0.07* (5) 20.93kO.45 6.87t0.06 (10) Values are means 2 SE of no. theses. ASL, airway surface liquid. bath pH.
COz, %
Bath
8.06zkO.07 (8) 4.55AO.30 (10) 7.39kO.01 (45) 5.58k0.18 (47) 7.21kO.07 (5) 6.3320.47 (5) 8.0920.54 (9) 6.96kO.02 (7) of observations indicated in paren* P < 0.05, ASL pH compared with
which is generally similar to the CO,-pH curves for Krebs-Henseleit solution and for secreted ASL without the trachea present (Fig. 1). Figure 2 also includes the corresponding values for ASL pH and luminal COa when the external COa was changed (Table 1); there is little change in ASL pH with COZ, and the points are displaced considerably to the acid side of the curve based on values derived from directly changing luminal COZ. TABLE
2. Changes in ASL pH, organ bath pH, and luminal A Bath COz, %
5.65 ---) 0.00 0.00 + 5.65 5.65 ---) 10.04
%COz with changes in %CO, in organ bath
APH
A Luminal
ASL
Bath
+0.05~0.05 (11) -0.06z!z0.02* (11)
(8) (8) -0.18-t-0.05* (5) +O. 18-10.04t (5) -0.44+0.02t (7) +0.4O~o.olt (7)
-0*01t0*01
+0.02kO.03 0.00~0.03 5.65 + 20.93 20.93 -3 5.65 +0.04&0.02* Values are means t SE of no. of observations indicated 10.04 ---) 5.65
Table 3 gives the half times for changes in ASL pH caused by step changes in luminal COz content. The half times for responses to increasing CO2 content were always considerably shorter than those as the result of decreasing COz content. CO2 diffusion from tracheal lumen. To test the rate of diffusion of CO2 across the tracheal wall, in three preparations the trachea was initially filled with 21.37% CO2 in O2 and the COz remaining at various intervals was estimated. The external buffer was bubbled with 5% CO2 in Oz. Figure 3 shows that equilibrium was 87% complete in 30 min and complete within 60 min. Replotting Fig. 3 logarithmically shows that the initial decrease in CO2 percentage was considerably faster than one would expect from an exponential relationship. Because the luminal CO2 content might be affected by CO2 production by the mucosal tissue, in six experiments the lumen was initially filled with 5.16% CO2 in 02 while the external buffer was bubbled with the same mixture. The trachea was kept closed. After various intervals the luminal COz percentage was measured (Fig. 3). The luminal CO2 increased to 6.81 t 0.45% after 30 min and to 8.43 -$-0.98% after 120 min.
(5)
(5) (10) (10)
+0.66+0.07t
-0.64+0.08t
Cog, %
-1.lOk0.26t (10) -0.05~0.08 (10) +0*9340*21t
(5)
-0.58-c-0.12-f (5) +3*03+0*31t
(9)
-2.0720.18t (6) * P < 0.05, t P < 0.01 for change in value. in parentheses. ASL, airway surface liquid.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
138
CONTROL
OF AIRWAY
SURFACE
LIQUID
PH
80 *
-o0
75.
co2 chanp in lumen
CO2 changein bath
FIG. 2. Measurement of ASL pH with changes in %COz when gas is passed through lumen of the trachea (n = 3-15). Also shown are values for luminal %COz and ASL pH when gases of different %COz were bubbled through external Krebs-Henseleit solution (n = 5-47, Table 1). Values are means & SE.
X
ocr
70.
1
I
0
10
1
20
co20
TABLE 3. Half-time changes in ASL pH caused by passing gases of different %COa through tracheal lumen
trations. However, the position of the curves could not explain the relative acidity of the ASL, especially at low CO2 concentrations. In this paper ail results have been Change in COZ,% n Half Time, xnin presented and analyzed statistically as pH values, al0.00 + 5.82 though strictly one should not use normal distribution 7 1.01*0.21t 5.82 + 0.00 8 6.44*2.30* statistics for pH, which is a logarithmic function. How0.00 --+ 10.65 5 1.2O,tO.33* ever, it is conventional to do so, as in nearly all the 4 10.58 --) 0.00 9.88&4.02* papers quoted here. Converted to fH+] our results give a 0.00 + 17.24 3 0.90*0.55 value of 140 nmol/l for a pH of 6.85, over three times 17.24 -+ 0.00 3 3.73k3.33 Values are means + SE. 0.00% COz, pure 0,; 582% COz, from a gas greater than the normal value for plasma and interstitial cylinder; 10.65 k 0.65, 10.58 k 0.92, and 17.24 2 0.47% COZ, from gas fluid (40 nmol/l for a pH of 7.40). Neither body weight, mixtures. ASL, airway surface liquid. * P c 0.05; P P < 0.01. age, nor sex of the ferrets appeared to influence the results. rHSCUSSIUN Because the changes in pH of ASL when external CO2 concentration was raised were usually insignificantly The liquid in the trachea comprises periciliary fluid small, it was impossible to determine their time constant and mucus secreted from submucosal glands, sometimes after a step change in CO2 is introduced in the external referred to as “sol” and “gel” (24). The ferret trachea has buffer. However, changes in luminal CO2 as the result of relatively few goblet or serous cells (18), which presumoutward diffusion were 87% complete in 30 min, and ably contribute to the ASL. Although the mucoglycoprotein content of glandular mucus is higher than that of even if the time constant for inward diffusion of CO2 were greater, one would have expected clear and signifipericiliary fluid, small solutes can presumably equilibrate cant changes in pH of ASL within 30 min. Resting between the two. Thus the classical distinction between preparations showed only small changes in luminal CO2 the two main components is not clear. In our experiments percentage over the same time period (Fig. 3), changes mucus secretion from submucosal glands was promoted by methacholine to ensure enough ASL to cover the pH presumably as the result of metabolic production of COZ. Measurement of luminal CO2 percentages when the electrodes and allow measurements. external buffer was bubbled with different CO2 concenThe ASL is significantly more acid than the surroundtrations showed that these changed far less than one ing Krebs-Henseleit solution or normal interstitial fluid. would expect from the external values. Under these The low pH of ASL is not the result of the particular buffering powers of the secreted mucus, because both the conditions the luminal CO2 relationship to ASL pH was quite different from that when COZ-rich gas flowed C02-pH curve for secreted ASL with the trachea absent (Fig. 1) and that for ASL in situ with gas being passed through the trachea. In the latter instance the gas flow through the trachea (Fig. 2) gave very similar curves to emerging close to the pH-sensitive electrode presumably that of Krebs-Henseleit solution (Fig. 1); the former caused a C02-pH relationship as if the trachea had not curves were on the more acid side for high CO2 concenbeen present. Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
CONTROL
OF AIRWAY
SURFACE
LIQUID
PH
139
FIG. 3. Changes in luminal %CO with time. l , Lumen initially filled with 21.37% COz (r2 = 3); 0, lumen initially filled with 5.16% CO2 (n = 6).
0
0
50
Time (min) To ensure that the pH-sensitive electrodes were covered with ASL, submucosal gland secretion was promoted by a small maintained dose of methacholine. The pH of unadulterated gland secretion is not known. However, the control pH of ASL in the same model was 7.12 t 0.03 for nonstimulated secretions and was reduced to 6.97 t 0.56 wh en gland secretion was stimulated by methacholine (17); the latter value is close to those described in the present paper, Thus, although methacholine may have slightly reduced the pH of ASL, it is unlikely to have modified the main results. Published values for the pH of ASL and secreted mucus vary quite widely, presumably because of very different experimental methods of collecting mucus, the possibilities of evaporation and contamination with other liquids, and the fact that some studies with humans were on patients with respiratory disease. Thus the pH of sputum has been measured and recorded as between 6.6 (19) and 8.2 (1). Secretions from the lower respiratory tract give values from 5.9 (22) to 8.12 (9). One of the most extensive studies used in vivo measurements with a pH electrode mounted in a bronchoscope and lowered onto the mucosal surface of the trachea of healthy humans and rabbits (5). Values were 6.73 t 0.56 for the rabbit, 6.5-7.5 for healthy humans, and 6.1-7.9 for tracheostomized patients. Similar values for the rat trachea are 7.42-7,57 (4). However, electrodes placed in contact with the airway epithelium may cause local damage and thereby distort the mucus pH. By using a micropipette method, Nielson et al. (14, 15) obtained a value of 6.92 -+ 0.04 for the pH in the lining surface liquid of the alveoli of rabbits. Mentz et al. (12) collected ASL from healthy humans and found a value of 6.88 t 0.08 for trachea and 6.77 t 0.10 for bronchi. These values should
be compared with those of Guerrin et al. (6), who reported means of 6.71 -t 6.97 for the human trachea. Thus our results support the majority of publications indicating a pH well on the acid side of plasma and interstitial fluid. The pH of ASL is not only acid compared with interstitial fluid but the acidity is maintained despite large differences in external pH values. A similar result was obtained by Nielson et al. (15), who showed that the surface liquid of the alveoli of rabbits was little changed in pH in response to changes in external CO2 concentration and resultant pH alterations. Thus a homeostatic mechanism seems to apply to the ferret trachea as well as to the rabbit alveoli. The basis of this homeostasis is not clear. Mucus can act as a passive barrier to H+ diffusion (7, 11). The mucus may act as an ampholyte and thereby restrict H+ movement because of its properties as a proton donor (21). However, our results indicate that there may be an active control of the pH of ASL in addition to any passive property of secreted mucus as an ampholyte. Presumably this active control would be based on epithelial secretion of H+/HCO& and the secretory mechanism would be regulated by H+/COz concentrations at the epithelial cell membranes. To test this hypothesis one needs to study the transport systems on the apical and basolateral membranes. If ASL is maintained acid relative to interstitial fluid by a homeostatic mechanism, the functions and results of such a mechanism have to be considered. Speculative possibilities include the following: 1) there may be an optimal pH to limit bacterial adhesion and growth; 2) there will be optimal pH values for the activity of luminal enzymes and for the physicochemical properties of luminal macromolecules such as mucins; 3) release of mediators from mast cells depends on pH, being inhibited
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
140
CONTROL
OF AIRWAY
at values ~6.8 (2, 23); 4) ciliary function might be affected by pH, although published results suggest that the cilia are active over a wide range (8, 11); 5) luminal PH might affect the beha vior of migratory cells. Many of these functions have been studied, but there is little information in the literature as to their optimal pH values or the effect of changes in pH. If ASL is a barrier to H+ diffusion, then the abrupt changes in luminal CO2 percentage (from 0 to -5.5% in normal tidal breathing) might be effectively buffered from having any chemical effect on the epithelium and deeper tissues. This and other possibilities need exploration. We are grateful to Dr. S. E. Webber for valuable discussion and help with the manuscript. H. Kyle was supported by Karl Thomae. Present address of J. P. T. Ward, Depts. of Medicine and Physiology, St. Thomas’ Hospital Medical School, Lambeth Palace Rd., London SE1 7EH, UK. Address for reprint requests: J. G. Widdicombe, Dept. of Physiology, St. George’s Hospital Medical School, Cranmer Terrace, London SW17 ORE, WK. Received 10 April 1989; accepted in final form 24 August 1989.
SURFACE
9.
10.
11. 12.
13.
14. 15. 16.
REFERENCES 1. BASCH, F,, P. HOLINGER, AND H. G, PONCHER. Physical and chemical properties of sputum. 1, Factors determining variations in portions from different parts of the tracheobronchial tree. Am. J. Dis. Child. 62: 981-990,1941. 2. DIAMANT, B. Histamine release elicited by extracts of Ascaris suis-influence of oxygen lack and glucose. Acta Physiol. Stand. 52: 8-22,196l. 3. FABRICANT, N. D. Significance of the pH of nasal secretions in situ. Arch, Otolaryngol. 34: 297-301, 1941. 4. GATTO, L. A. pH of mucus in rat trachea. J. Appl. Physiol. 50: 1224-1226,198l. 5. GUERRIN, F., C. VOISIN, V. MACQUET, H. ROBIN, AND P. LEQUIEN. Apport de la pH metrie bronchique in situ. Progr. Respir. Res. 6: 372-383,197l. 6. GUERRIN, F., C. VOISIN, V. MACQUET, H. ROBIN, F. WATTEL, AND J. P. BOULANGER. Possibilites de la pH metrie bronchique in situ. In: Hypersecretion Brunchique. Clinchy: Poinsot, 1969, vol. 1, p. 249-256. 7. HOLMA, B, Influence of buffer capacity and pH-dependent rheological properties of respiratory mucus on health effects due to acidic pollution. Sci. Total Environ. 41: 101-123, 1985. 8. HOLMA, B., M. LINDCREN, AND J. MORKHOLDT ANDERSEN. pH
17. 18. 19. 20,
21.
22, 23. 24.
LIQUID
PH
effects on ciliomotility and morphology of respiratory mucus. Arch. Environ. Health 32: 216-226, 1977. KWART, H., W. W. MOSELEY, AND M. KATZ. The chemical characterization of human tracheobronchial secretion: a possible clue to the origin of fibrocystic mucus. Ann. NY Acad. Sci. 106: 709721,1963. KYLE, H., N. P. ROBINSON, J. P. T. WARD, AND J. G. WIDDICOMBE. Measurement of mucus [H+] and other ionic concentrations in the isolated trachea of the ferret (Abstract). J. Physiol. Land. 387: 15P, 1987. LUK, C. K. A., AND M. J. DULFANO. Effect of pH, viscosity and ionic-strength changes on ciliary beating frequency of human bronchial explants. Clin. Sci. Lund. 64: 449-451, 1983. MENTZ, W. M., M. R. KNOWLES, J. B. BROWN, J. T. GATZY, AND R. C. BOUCHER. Measurement of airway surface liquid (ASL) composition of normal human subjects (Abstract). Am. Rev. Respir. Dis. 129: 315, 1984. MITTEMAIER, R. Untersuchung uber die Wasserstoffronenkonzentration an Sekreten und Schlermhauten im besonderen bei chronischen Nebenhohlenkrankungen. Arch. Ohren-Nusen-KehZkopfheiZkd. 127: 1, 1930. NIELSON, D. W. Electrolyte composition of pulmonary alveolar subphase in anesthetized rabbits. J. Appl. Physiol. 60: 972-979, 1986. NIELSON, D. W., J. GOERKE, AND J. A. CLEMENTS. Alveolar subphase pH in the lungs of anesthetized rabbits. Proc. Natl. Acud. Sci. USA 78: 7119-7123,198l. PFEIFFER, C. J. Experimental analysis of hydrogen ion diffusion in gastrointestinal mucus glycoprotein. Am. J. Physiol. 240 (Gastrointest. Liver Physiol. 3): G176-G182, 1981. ROBINSON, N. P., H. KYLE, S. E. WEBBER, AND J. G. WIDDICOMBE. Electrolyte and other chemical concentrations in tracheal airway surface liquid and mucus. J. Appl. Physiol. 66: 2129-2135, 1989. ROBINSON, N. P., L. VENNING, H. KYLE, AND J. G. WIDDICOMBE. Quantitation of the secretory cells of the ferret tracheobronchial tree. J. Anut. 145: 173-188, 1986. RYLEY, H. C., AND T. D. BROGAN. Variation in the composition of sputum in chronic chest disease. Br. J. Exp. Pathol. 49: 625-633, 1968. SCHULTHESS, P., Y. SHIJO, H. V. PHAM, E. PRETCH, D. AMMANN, AND W. SIMAON. A hydrogen ion-selective liquid membrane electrode based on tri-n-dodecylamine as neutral carrier. Anal. Chim, Acta 131: ill-116,198l. SHIAU, Y.-F,, P. FERNANDEZ, M. J. JACKSON, AND S. McMONAGLE. Mechanisms maintaining a low-pH microclimate in the intestine. Am. J. Physiol. 248 (Gastrointest. Liver Physiol. 11): GBOB-G617,1985. STEINMANN, E. La secretion bronchique et le pH. Branches 6: 126129,1965. UVNAS, B., B. DIAMANT, B. HOGBERG, AND I, T. THON. Mechanism of mast cell disruption induced by a principle extracted from Ascaris suis. Am. J. Physiol. 199: 575-577, 1960. WIDDICOMBE, J. G. Airway mucus. Eur. Respir. J. 2: 107-115,1989.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 20, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
