Effect of Acetylsalicylic Acid on the Amphibian Gastric Mucosa

GASTROENTEROLOGY 73:785-789, 1977 Copyright © 1977 by the American Gastroenterological Association

Vol. 73 , No. 4, Par t 1 Printed in U.S A.

EFFECT OF ACETYLSALICYLIC ACID ON THE AMPHIBIAN GASTRIC MUCOSA

I. Electrophysiological and permeability changes JERRY

G.

SPENNEY, M.D., AND MEERA BROWN

Division ofGastroenterology , Department of Medicine, U niuersity ofAlabama in Birmingham , and Veterans Administration Hospital, Birmingham, Alabama

The effects of acetylsalicylic acid on the in vitro bullfrog gastric mucosa were defined. Mter mucosal exposure of 10 or 20 mM acetylsalicylic acid, a characteristic series of changes evolved. The potential difference and resistance rose initially. After 15 to 30 min the potential difference declined to zero. Transmucosal resistance remained elevated considerably longer but began to decline toward zero after 30 to 60 min of exposure. Mucosal permeability to mannitol increased as transmucosal resistance declined, but was not markedly elevated until the 2nd hr of exposure to acetylsalicylic acid. The initial rise of potential difference was associated with inhibition of H+ secretion whereas the later decline of potential difference was due to inhibition of Cltransport. The abolition of short circuit current could not be attributed to mucosal to serosal flux of acetylsalicylate. 1. The pH of the mucosal solution is given in the figure legends and text. Concentrations given in the text and figure legends are final concentrations in the bathing solutions. Large size bullfrogs (Rana catesbeiana ) were u sed in all studies. The bullfrogs were killed by severing the spinal cord in the neck and pithing. The abdomen was opened quickly a nd the stomach removed to a dish containing amphibian Ringer's solution bubbled with 95% 0 2-5% C02 . The external muscle layer was stripped and the mucosal tube · was opened, stretched, and mounted between half-chambers made from Lucite. The serosal and mucosal solutions (table 1) were circulated from external reservoirs by a 95% 0 2 -5% C02 gas lift system. Potential difference (PD) was measured by KCl-calomel electrodes inserted int o the tubing circulating fluid to the chambers. The calomel electrodes were connected to a n Esterline Angus recorder. The resistance was calculat ed using Ohm's law and the PD resulting when current pulses of 20 ~-tamps (bipolar) were sent through agar electrodes. Resistance measurements were controlled by an automatic pulse progra mmer10 which sent bipolar pulses of 500 msec in each direction at a frequency of 0.017 Hz. The buffering capacity of acetylsalicylic acid precluded measurement of acid secretion by titration Materials and Methods in these experiments. As an alternative to measuring H+ All reagents used in these experiments were analytical secretion, experiments were performed in Cl- -free solutions in reagent grade or the highest grade available. Acetylsalicylic which the short circuit current (I5c) approximat es the H+ secre11 acid (aspirin) was purchased from Sigma Chemical Company tory rate. The mucosa was equilibrated in the chamber for 30 to 60 (St. Louis, Mo.); histamine-2·HCl was purchased from Aldrich Chemical Company, Inc. (Milwaukee, Wis.). The composition min before additions were made. When permeability was 14 of the serosal and mucosal bathing solutions is given in table quantitated, [ C]mannitol was added t o the mucosal solution which would be used throughout the day. In order to qua ntiReceived December 27, 1976. Accepted March 28, 1977. tate the flux, 500 ~-tl samples were removed from the serosal Address requests for reprints to: J erry G. Spenney, M.D. , Divi- solution every 10 min. The specific activities in the mucosal sion of Gastroenterology, Department of Medicine, University of and serosal solutions were determined from the concentration of unlabeled mannitol in each solution and the radioactivity Alabama in Birmingham, Birmingham, Alabama 35294. This study was supported by National Institutes of Health Grant quantitated in a LS-133 liquid scintillation spectrometer AM17315 and by the Medical Research Service of the Veterans (Beckman Instruments, Fullerton, Calif.). At the beginning of the acetylsalicylic acid exposure period Administration. 785

The effect of acetylsalicylic acid (aspirin) injury on gastric fundic mucosa has been studied in man1• 2 and experimental animals. 3-s Baskin et al. (; has reported that early changes in man include swelling of surface epithelial cells. Hingson and Ito 7 have shown similar changes in rat stomach exposed to 10 mM acetylsalicylic acid in 0.1 M HCl. Kasbekar8 and Fromm et al. 9 have shown inhibition ofH+ secretion and reduction of potential difference by sodium salicylate in amphibian and rabbit fundic mucosa, respectively. These studies suggest that salicylates may have effects on cellular structure and function but the interrelation has not been defined. The effects on ion transport suggest that salicylates either: (1) directly inhibit transport mechanisms in the epithelium; or (2) interfere with energy production required to support transport functions of the epithelium. The studies reported here investigate electrophysiological and permeability changes after acetylsalicylate addition to the mucosal solution bathing the in vitro amphibian gastric fundic mucosa.

786

SPENNEY AND BHOWN

Vol. 73, No . 4 , Part 1

(after a control period of60 min in permeability experiments), the mucosal solution was drained and replaced with a fresh solution containing acetylsalicylic acid at the desired concentration and pH. That solution contained 10 MCi of [acety l14C)salicylic acid in experiments in which acetylsalicylic acid flux was measured. When permeability was being determined, samples were taken from the serosal solution every 10 min for another 2 hr. Student's t-test was used to evaluate the significance of differences: P < 0.025 was accepted as significant.

was added. Over a longer period (70 to 150 min) the resistance declined to near zero. The rapidity of changes decreased as the pH of the acetylsalicylic acid-containing mucosal solution increased (fig. 2). These results are shown as percentage of change referred to the control period for each mucosa since the PD in the control period (PD 0 ) was different for each mucosa. The mean PD0 was 28, 21, and 15 mv for the experiments at pH 3, 4, and 6, respectively. The changes ofPD occurred more Results rapidly at higher acetylsalicylic acid concentrations Exposure of the mucosal surface to 10 or 20 mM acetyl- (fig. 3). The magnitude of the PD rise at pH 4 and 6 salicylic acid caused a characteristic series of changes. differed significantly from pH 3 (P < 0.025). The differThe PD rose over a period of 10 min from 13.5 to 19 mv ence between pH 4 and 6 did not achieve significance. (fig. 1). During this period the transmucosal resistance The time for the peak rise of PD and its return to PD0 rose from 65 to 95 ohms per cm2 • The PD reached its differed significantly when pH 3 and 4 or 6 were comapogee after a 15-min exposure; thereafter, the PD be- pared. When pH 4 and 6 were compared, the time for gan to decline and approached zero 30 to 45 min after peak rise of PD differed significantly , but the time for acetylsalicylic acid was added. During the phase of de- return to PD 0 did not differ significantly. clining PD, the transmucosal resistance continued to rise to approximately 235 ohm·cm2 • Mter about 35 min the transmucosal resistance began to decline and reached control levels 75 min after acetylsalicylic acid TABLE 1. Composition of solutions ( mM) Cl-containing solutions

Na+ K+ Ca++ Ma++ Cl· HC03 H 2 P04

so.-

Glucose , .Sucrose · Mannitol Milliosmolar pH

·

Cl-free solutions

Serosal

Mucosal

Serosal

Mucosal

85 5 1.5 1.0 80 18.5 1.5 5 5

90 5

85 5

90 5 1.5 1.0

80

45 5

5.0

30 210 7.2

18.5 1.5 40 5

7.5 210

70 210 7.2

pH 6 n=3

47 .5 210

2 300 ~/cm

200

. R 200

FIG. 2. Bullfrog mucosae mounted in Lucite chambers were exposed to 20 mM acetylsalicylic acid at pH 3, 4, and 6 (mucosal solution). The addition indicated pH 7.0 was m ade to th e serosal solution. Serosal addition did not reduce th e potential difference or transmucosal resistance.

100 lsc 0

PD IO

' ASA

40

IOmM n =4

5

20 10

30

50

70

20mM n=4

90

minutes

FIG. 1. A representative experiment in which bullfrog mucosa , ·mounted in a Lucite chamber was exposed to 20 mM acetylsa licylic acid (arrow) at pH 4.0. Potential difference, transmucosal resistance, and ca lculated short circuit current are shown.

0

20

40 minutes

60

FIG. 3. Bullfrog mucosae mounted in Lucite chambers were exposed to either 10 or 20 mM acetylsalicylic acid, pH 4.0, in the mu cosal solution .

The difference between the magnitude of PD rise for 10 versus 20 mM acetylsalicylic acid at pH 4 was not significant. The time course compared at the peak and on return to PD11 were not significantly different, but the time required for reduction of PD to 50% PD11 differed significantly, (P < 0.005). Control mucosae studied for as long as 8 hr maintained their PD, resistance, and control level of mannitol permeability. No control mucosae spontaneously underwent changes similar to the changes seen after acetylsalicylic acid addition. Thus, factors favoring rapidity of electrophysiological changes include low pH and high acetylsalicylic acid concentration in the mucosal solutions. The changes underlying: (1) the initial rise of PD and resistance, (2) the subsequent decline of PD, and (3) the relation of transmucosal resistance to permeability of nonionic molecules required further explanation. Phase of rising PD and resistance. The initial rise of PD and resistance suggested that either H+ secretion was inhibited or net Cl- transport was increased. The buffering capacity of acetylsalicylic acid in the mucosal solution obviated meaningful quantitation of H + secretion by titration. To circumvent this problem, mucosae were bathed with solutions in which Cl- was replaced by S04=. Under these conditions the PD is oriented with the mucosal solution positive and the magnitude of the short circuit current approximates the H+ secretory rate. 11 Paired mucosae derived from a single fundus were used to compare the changes in Cl--containing and Cl--free solutions. Upon addition of acetylsalicylic acid, the PD fell to zero in 15 min in Cl--free condition whereas the PD rose when Cl- was present (fig. 4) . Thus, the initial rise of PD seen when Cl--containing solutions were used reflected inhibition ofH+ secretion.

200

C1-

R

150

100 0

787

ACETYLSALICYLIC ACID AND GASTRIC MUCOSA

October 1977

100 1sc

-20

The transmucosal resistance rose in either condition, i.e., when Cl- was present or was replaced by 804=· Stimulation of Cl- transport is not suggested by these experiments since the Isc remained unchanged or fell slightly during the rise of PD. Phase of decreasing PD. When Cl--containing solutions were used, the decreasing PD was associated with an increase in transepithelial resistance. The short circuit current fell to near zero during the phase of decreasing PD (fig. 4) and could be attributed to: (1) dissipation of the potential gradient by another anion (acetylsalicylate); (2) increased permeability of the mucosa such that the mucosal to serosal Cl- flux increased; or (3) inhibition of Cl- transport. Acetylsalicylate dissipation of the potential gradient would require 4. 7 p.moles per hr-cm2 of acetylsalicylate anion moving from the mucosal to the serosal solution without an accompanying cation; this estimate is based on the Isc shown in figure 4. To evaluate this possibility, the unidirectional flux of acetylsalicylate was measured. Figure 5 gives the results of two of these experiments which were selected because the Isc represented the high and low extremes. The results were plotted as the decrement of Isc versus the acetylsalicylate flux. If the acetylsalicylate flux were responsible for dissipation of the potential gradient, the graph would have been linear with a slope of 26.8. The graph shows neither characteristic. Relation of resistance and permeability to nonionic molecules. Permeability changes in the mucosa can range from changes in permeability to small molecules or ions to massive changes with increased permeability to molecules as large as serum proteins. The time course of resistance changes suggested that massive changes of permeability occurred relatively late in the course of acetylsalicylate injury. These changes were assessed by measuring the permeability to [14 C]mannitol. The mucosal solution was altered to contain 7.5 mM mannitol and 50 p.c of [14C]mannitol; the serosal solution contained 30 mM mannitol. Figure 6 shows a representative experiment. The [14C]mannitol permeability coefficient began to rise and exceeded control levels during the 2nd hour after acetylsalicylic acid addition. The control

mv -10

300

50

so~

100

decrement 200 of lsc ()JA!cm2 )

75 1sc 1000

50

R 500

25

0

10

20

30

100

40

minutes

FIG. 4. A representative experiment in which the fundus from a single bullfrog was divided between two chambers. One-half was exposed to acetylsalicylic acid (20 mM, pH 4.0, mucosal) in Cl- containing solutions (top panel). The other half was exposed to the same acetylsalicylic acid concentration and pH, in solutions in which Cl- was replaced by so.~.

200 400 600 800

1000

ASA flux (n moles/cm~ hr)

FIG. 5. Mucosae were exposed on the mucosal surface to 10 mM acetylsalicylic acid containing 10 fLCi of [acetyl-1 4 C]salicylic acid. Flux measurements were performed as indicated under Materials and Methods. The flux of acetylsalicylic was related to the decrement of calculated short circuit current. Mucosal solution pH was 4.0.

788

SPENNEY AND BHOWN

Vol. 73, No.4 , Part 1

permeability was 2.0 x 10- o em per sec. The permeability coefficient rose during the following 1.5 hr to 2.8, 4.2, and 6.5 x I0- 1• em per sec. Thus, the increased permeability (exceeding control levels) occurred during the 2nd hour after addition of acetylsalicylic acid. During this time transmucosal resistance was falling. The highest mannitol permeability coefficient occurred when the transmucosal resistance was below control levels. Thus, increased mannitol permeability correlated with the phase of decreasing mucosal resistance. Table 2 gives the results of a group of mucosae exposed to 20 mM acetylsalicylic acid, pH 4.0. Permeability was calculated over 30-min periods. The mannitol permeability did not increase until the resistance began to decrease; this occurred 60 to 90 min after addition of acetylsalicylic acid. 16,16-Dimethyl prostaglandin E 2, which is reported. to prevent mucosal injury due to salicylates, 12 was mcluded in the serosal solution of several experiments in a concentration of 1 JLg per ml. Inclusion of this protaglandin derivative did not prevent the permeability changes after acetylsalicylic acid addition (table 3). In addition, electrophysiological changes of PD and resistand) occurred as in other experiments in which protaghln'dins were not included.

Discussion Injury to gastric mucosa has been studied by physiological and morphological studies. In vitro physiol~g_ical studies by Kasbekar8 and Fromm et al. 9 have utihzed salicylic acid whereas those by Flemstrom and Marsden13 and Jorgensen et al. 14 have studied the more commonly used acetylsalicylic acid. These studies have documented the decrease of PD and reduction of transmucosal resistance. The mechanism underlying these changes has not been defined. The studies reported here define a characteristic set of changes after exposure of the in vitro mucosa to acetylsalicylic acid. We have arbitrarily divided the phases of injury into four phases described by the PD. Pha~e I is the period during which the PD is rising. Phase II IS the period of declining PD, and phase III is the per.iod characterized by a zero or near zero PD and decreasmg transmucosal resistance. Phase IV is the late period, usually 60 to 120 min after exposure, when transmucosal resistance has decreased to a low value. The effect of acetylsalicylic acid on mucosal ion transport encompasses both H + and Cl- . The mechanism of these changes has not been defined in previous studies but the alternatives are: (1) inhibition by direct interaction with a component of the transport enzymes; (2) inhibition of H+ by a protonophore effect of acetylsalicylic acid moving from the mucosal solution; (3) inhibi600 tion oflsc by transmucosal movement of acetylsalicylate 400 anion; or (4) inhibition of both H+ and CI- transport.by R metabolic inhibition reducing energy to support active transport. Inhibition of H + secretion. Acetylsalicylate was shown to inhibit H+ secretion as one of the earliest effects. Coincident with inhibition of H + secretion, transmucosal PD rose. The present studies indicate that especially during these early time periods very small 6 amounts of acetylsalicylic acid are moving across the mucosa. As shown in figure 5, the largest amounts of acetylsalicylic acid appearing in the serosal solution • • were only 1JLmole per cm2 -hr. Thus, mucosal to serosal 2 em/sec • movement of acetylsalicylate could not account for inhibition of H + secretion. 20 40 60 80 100 120 140 160 minutes It remains possible that acetylsalicylate could insert FIG.. 6. A representative experiment in which fundic mucosa was into the cell membrane to act as a protonophore across exposed to a 20 mM acetylsalicylic acid mucosal solution (pH 4.0) the apical membrane. Under this condition no net which contained [14 C]mannitol. Mannitol permeability was calcu- movement of acetylsalicylate from the mucosal to the lated from the flux per unit concentration gradi~nt and is expressed serosal solution would be required. Studies of N avanax . in\ entimeters per second. neurone15 h ave suggested an altered surface charge after salicylate treatment. TABLE 2. Mucosal p ermeability after mucosal exposure to The studies reported here do not investigate acetylsalacetylsalicylate icylate interaction with the H+ secretory mechanism. Mannitol permeability as Observation period n % of control (mean ± SE) Jorgensen et al.w reported that theophylline reversed the inhibition of H + secretion caused by salicylates. In Control ;:~ 0- 30

30- 60

Acetylsalicylic acid <0- 30 30-60 6o- 9o 90- 120

12 12

100% 100

12 12 10 7

121 113 154 373

± ± ± ±

TABLE

15.5 9.6 14.1" 57.1"

· ".;i>. <::o.025 for the period compared with the control period.

3. Changes of mannitol permeability after acetylsalicylate in the presence of prostaglandin E 2 Time Permeability P value min % control 0- 60 100 60-90 199 < 0.025 90- 120 222 < 0.005

October 1977

ACETYLSALICYLIC ACID AND GASTRIC MUCOSA

several experiments, we added theophylline to the serosal solution when the PD had reached near zero value. Even when salicylate was removed, theophylline did not restore H + secretion. A metabolic basis for inhibition ofH+ secretion would be consistent with the reduction of mucosal ATP after exposure to sodium salicylate. 8 However, this reduction of mucosal ATP was measured 60 min after addition of sodium salicylate-much later than the time required for inhibition of H+ secretion. Inhibition of net chloride transport. In these studies, we have used the calculated short circuit current as an indicator of net Cl- transport by the mucosa. CI- transport was inhibited at a much slower rate than H + secretion. Ignorance of the subcellular mechanism has precluded any direct in vitro assessment of acetylsalicylate interaction. Movement of acetylsalicylate anion in the absence of an accompanying cation could account for reduction of the short circuit current. Figure 5 demonstrates that this mechanism is unlikely. If all acetylsalicylate absorption occurred via a conductive pathway, the acetylsalicylate current could account for only a small fraction of the observed decrement of Isc (20% of the lower plot and 6% of the upper plot of fig. 5). Thus, acetylsalicylate absorption cannot account for dissipation of the PD. Metabolic inhibition remains a likely mechanism. The reduced mucosal ATP content 60 min after addition of sodium salicylate8 conforms more closely to the time course of inhibition ofCI- transport. A similar sequence of inhibition was noted for the amytal-treated mucosa. 17 Mucosal permeability. Maintenance of an effective "barrier" is an important characteristic of gastric mucosa; under normal conditions, this epithelium can maintain an H + gradient of 10H. In vivo salicylates and acetylsalicylate have been shown to reduce the ability of gastric mucosa to maintain these gradients and increased loss of intralumenal acid has been found. The studies reported here quantitate the changes of mucosal permeability using the nonionic hexitol, mannitol. Mucosal permeability to mannitol increased as the transepithelial resistance declined. Significant increases were achieved during the period 60 to 120 min after addition of acetylsalicylic acid. Extrapolation to other molecules is difficult, but based on ionic radius, H+ permeability would presumably increase at an earlier time and increase more markedly than mannitol permeability. Kasbekar8 measured sucrose permeability after addition of salicylic acid at pH 3.0; mucosal permeability increased more rapidly than we found for acetylsalicylic acid in these studies. In Kasbekar's studies, decline of PD and resistance also occurred more rapidly, suggesting that all mucosal changes may occur more rapidly

789

upon exposure to salicylic acid. Flemstrom and Marsden13 found dextran permeability increased 60 to 180 min after addition of acetylsalicylic acid to gastric mucosa fromRana temporaria. The timing of these permeability changes is quite similar to those reported here. The magnitude of dextran permeability was much lower based on the greater molecular size. Thus, the effects of acetylsalicylic acid on the in vitro amphibian mucosa include inhibition of H+ secretion, inhibition of net Cl- transport, and increased permeability to nonionic molecules. These effects were reflected by changes of the transmucosal PD, short circuit current, and transepithelial resistance. REFERENCES 1. Ivey KJ, Morrison S, Gray C: Effect of intravenous salicylate on gastric mucosal barrier in man. Am J Dig Dis 17:1055-1064, 1972 2. Smith BM, Skillman JJ, Edwards BG, et al: Permeability of the human gastric mucosa: alteration by acetylsalicylic acid and ethanol. N Eng! J Med 285:716-721, 1971 3. Davenport HW: Damage to the gastric mucosa: effects of salicylates and stimulation. Gastroenterology 49:189-196, 1965 4. Davenport HW: Gastric mucosal injury by fatty acids and acetylsalicylic acids. Gastroenterology 46:245-253, 1964 5. Davenport HW: Gastric mucosal hemorrhage in dogs: effects of acid, aspirin and alcohol. Gastroenterology 56:539-449, 1969 6. Baskin WN, Ivey KJ, Krause WJ, et al: Aspirin induced ultrastructural changes in human gastric mucosa: correlation with potential difference. Ann Intern Med 85:299-303, 1976 7. Hingson DJ, Ito S: Effect of aspirin and related compounds on the fine structure of mouse gastric mucosa. Gastroenterology 61:156-177 , 1971 8. Kasbekar D: Effects of salicylate and related compounds on gastric HCI secretion. Am J Physiol 225:521-527, 1973 9. Fromm D, Schwartz JH, Ouijano R: The effects of salicylate and bile salt on ion transport by isolated gastric mucosa of the rabbit. Am J Physiol 230:319-326, 1976 10. Spenney JG, Ostroy F: An adjustable constant current stimulator for electrophysiologic experiments. IEEE Trans Biomed Eng 24:67-69, 1977 11. HeinzE, Durbin R: Evidence for an independent hydrogen ion pump in the stomach. Biochim Biophys Acta 31:246-247, 1959 12. Cohen MM: Prostaglandin E 2 prevents gastric mucosal barrier damage (abstr) . Gastroenterology 68:876, 1975 13. Flemstrom G, Marsden NVB: Dextran permeability, electrical properties , and H + secretion in isolated frog gastric mucosa after acetylsalicylic acid. Gastroenterology 64:278-284, 1973 14. Jorgensen TG, Kaplan EL, Peskin GW: The mechanisms of action of salicylates on gastric acid secretion in vitro. Surg Forum 22:315-316, 1971 15. Barker JS, Levitan H: The antagonism between salicylate-induced and pH-induced changes in the membrane conductance of molluscan neurones. Biochim Biophys Acta 274:638-643, 1972 16. Jorgensen TG, Kaplan EL, Peskin GW: Salicylate effects on gastric secretion. Scand J Clin Lab Invest 33:31-38, 1974 17. Sachs G, Shoemaker R, Hirschowi tz Bl: The action of amytal on frog gastric mucosa. Biochim Biophys Acta 143:522-531, 1967