Effects of E Prostaglandins on Canine Gastric Potential Difference ESAM Z. DAJANI, PhD, DAVID A. C A L L I S O N , MS, and R A L P H E. B E R T E R M A N N
Measurement of the gastric transmucosal potential difference (PD) was used to study the effect of gastric antisecretory prostaglandins on the integrity of the gastric mucosa of the Heidenhain pouch dog. Intragastric administration ofSC-29333 [( +)-15-deoxy-16-o~,~-hydroxy-16-methyl PGEI methyl ester] slightly but significantly increased the transmucosal PD when compared to vehicle control. In addition, SC-29333 administered either intravenously or intragastrically, significantly inhibited the PD fall induced by aspirin, a well established barrier breaker. In contrast, the intragastric administration of l6,16-dimethyl PGE2 methyl ester (Me-PGEz) significantly lowered the transmucosal PD and failed to modify the actions of aspirin on the integrity of the gastric mucosa. However, the intravenous administration of either prostaglandin did not affect the basal transmucosal PD values. These studies suggest that SC-29333 may strengthen the integrity of the gastric mucosal barrier against aspirin, and this could have important therapeutic potential.
Synthetic prostaglandins (PGs) of the E type have been well established as inhibitors of gastric secretion and may be potentially useful in the treatment of peptic ulcer. The effects of these PGs on the integrity of the gastric mucosa has not been adequately defined. O'Brien and Carter (1) have recently shown that oral but not intravenous administration of 16,16-dimethyl PGE2 (Me-PGE2) significantly increased the ionic permeability of the gastric mucosa in the Heidenhain pouch dog. It is not known, however, whether the damaging effects of Me-PGE2 on the gastric mucosa is universally shared with other synthetic PGEs or if this action is limited to a specific prostaglandin. To test this hypothesis, we investigated the effects of SC-29333--a potent and a long-acting gastric antisecretory prostaglandin in animals (2, 3 ) - - o n the integrity of the canine gastric mucosal barrier. The aims of these studies were twofold: (1) to study the effects of SC-29333 and the reference standard Me-PGE2 (Figure 1) on the inFrom the Departments of Biological Research and Real Time Systems, Searle Laboratories, Chicago, Illinois. Address for reprint requests: Dr. Esam Z. Dajani, Gastrointestinal Pharmacology, Department of Biological Research, Searle Laboratories, Box 5110, Chicago, Illinois 60680.
436
tegrity of the gastric mucosal barrier as measured by the gastric transmucosal potential difference (PD) technique and (2) to investigate the interaction of PGs with known " b a r r i e r b r e a k e r s " such as aspirin on the integrity of the gastric mucosa.
MATERIALS AND METHODS Animals Six female mongrel dogs weighing between 12 and 18 kg were prepared with Heidenhain pouches drained by Gregory cannula. The dogs were trained to stand quietly in Pavlov stands and were conscious during all studies. A specially constructed cannula plug permitted instillation and withdrawal of test solutions. The cannula plug was also equipped with a side arm to permit passing of the detecting electrode to the pouch. The surgery was performed two months before these studies were started and the animals were not used more than twice per week. Drugs The prostagtandins were synthesized by P. W. Collins and M. Bruhn of the Department of Chemical Research of these laboratories. The prostaglandins were initially dissolved in absolute ethanol (10 mg/ml) and were subsequently diluted with sodium phosphate buffer (pH 7.4). The final alcohol concentrations when administered to the Digestive Diseases, Vol. 23, No. 5 (May 1978)
0002-9211/78/0500-0436505.00/1 9 1978DigestiveDiseaseSystems,Inc.
E PROSTAGLANDINS AND GASTRIC PD O
~.T...-~COOCH3 ~" ] CH 3 SC-29333 [(+)-IS-DEOXY-16 c,, ~ HYDROXY, I6-METh~L PGE1 METHYL ESTER]
~"
~
~COOCH-~
Me-PGE 2 [16, t6-Dm~m~L m;E2 METAL ESTIm]
At the beginning of each experiment the pouch was thoroughly rinsed with a basal solutionof 100 mM HCI and 54 mM NaC1. The osmolarity of this solution and all subsequent test solutions was kept constant at 285-300 mosmol/ liter by adjusting the concentration of NaC1 within the solution. The pouch was then drained and 20 ml of basal solution was introduced into the pouch. At the end of a 30-rain basal period the pouch was drained and refilled with test solution containing the basal solution plus prostaglandins (10 or 30 tzg/kg) or aspirin (10 or 20 raM) or a combination of prostaglandin and aspirin. At the end of the test period, the pouch was drained and refilled with basal solution to study the recovery pattern. A cross-over experimental design was used to quantitatively assess the effect of prostaglandins on the transmucosal PD values. The differences in the means for each 3-min interval at each test condition was determined by comparing the values obtained in treated animals at various times with those obtained from the same animals under control or placebo conditions using the paired Student's t test at P < 0.05 (6). Throughout the studies, the dogs were closely observed for gross side effects, if any.
RESULTS
Fig 1. Chemical structures of SC-29333 and Me-PGE2.
Effect of Prostaglandins on the Gastric Transmucosal Potential Difference pouch did not exceed 0.3%. Prostaglandin solutions were stored at - 10~ C when not in use. Aspirin was purchased from Sigma Chemical Company, St. Louis, Missouri.
Procedure The technique used to measure the transmucosal PD was essentially similar to that reported by Davenport et al (4) and Cooke and Kienzle (5). The reference electrode consisted of a commercial intravenous infusion set (Butterfly 21, Abbott Laboratories, Chicago, Illinois) filled at the time of the experiment with 3% agar in saturated KC1 solution and then introduced into a peripheral leg vein. The detecting electrode consisted of a polyethylene tube (PE 60, ID = 0.030 in.) filled with 3% agar in saturated KC1 solution. The detecting electrode was inserted into the pouch through the side arm of the cannula plug. Each electrode was led through the side arm of a 25-ml suction flask containing saturated KC1 in which one of a pair of balanced calomel (Corning Glass Work, Medfield, Massachussetts) cells were placed. The calomel cells were connected to an electrometer amplifier (Burr Brown, model 3431, Tucson, Arizona), followed by several stages of amplifiers (Texas Instruments, model SN72741, Dallas, Texas). These electronic arrangements allowed zero adjustment, inversion, and buffering of the transmucosal PD signal for display on a chart recorder and digital voltmeter (Datel, model DM 2000 AR, Canton, Massachussetts). The transmucosal potential difference was measured to the nearest 0.1 millivolt (mV) at 3-rain intervals. Experiments were initiated by depriving the dogs of food, but not water, for 18 hr. Each experiment consisted of three 30-min periods: basal, test, and recovery periods. Digestive Diseases, Vol. 23, No. 5 (May 1978)
The gastric transmucosal potential difference in control studies was - 5 8 + 0.5 mV (mean -+ SE), mucosa negative with respect to serosa. The intragastric instillation of the PG vehicle, which contained no more than 0.3% ethanol (v/v) did not significantly lower the PD values. The administration of PGs into the gastric pouch produced p r o n o u n c e d changes in gastric PD values when compared to basal or control period. As shown in Figure 2, Me-PGE2 produced dose-dependent and significant reduction in PD only for one period during the test session. H o w e v e r , the PD values did not return during the recovery phase when Me-PGE2 was administered at 30/~g/kg dose, while the lower dose of 10/zg/kg showed recovery pattern similar to the control. In sharp contrast, the intragastric (pouch) administration of SC-29333 at 10 and 30/xg/kg doses significantly increased the gastric PD when compared to the vehicle (0.3% v/v ethanol). This increase in PD was maintained even during the withdrawal or recovery phase of the study. H o w e v e r , the PD values associated with SC-29333 intragastric administration during the test and recovery phase did not significantly differ from the basal phase of this study. This basal phase was consistently maintained at - 5 8 mV for the entire 90-rain duration (Figure 3).
437
DAJANI ET AL ?0
BASAL
l
TEST
l
t
RECOVERY l . 9 ,L.~%.~.-.=" ,..4P-..9"""
~
. .'. g g.-.-'...,:...,'...,'"~"~"
9 ,,....w "~_~.,~--~
_~
6C
.-~'*-~"~'~-~--*
.,r"
f.. t.. ' i
E
~
-
~
p-~,
Z
~../
,,s
J
-
r
~
"- " ,
~,-,~/ ,
~
-
_
.~--:ff' ',-
u....
'"'~...~'..-o ....
. . . o"-~' "~>...~.~.-~ 9
.tr-= 9
*
e - - - - e vehi(:le (Cor, t roll ~ , - - - - , L S C - 2 9 3 3 3 (lOmco/kg)
Z ua
e......-eS C - 2 9 3 3 3 (30rr~g/kg) ~r - - -A. Me - ~ E 2 (lOmcg/kg) a~......OMe-PC, E2 (30mcg/kg)
3C
i
,
,
,
i
,
.
.
15
.
i
.
.
.
.
30 TIME
,
.
.
.
.
45
i
60
.
.
.
.
i
75
.
.
.
.
i
90
(3 M0N INTERVALSI
Fig 2. Effects of prostaglandins on the gastric transmucosal potential difference in the Heidenhain pouch dog. The prostaglandins (10 and 30/zg/kg) were introduced into the pouch in the test period and were withdrawn in the recovery period. Values represent means obtained in six dogs. Asterisks denote periods in which the difference between control (vehicle) and treated series were significantly different.
Intravenous administration of either SC-29333 or Me-PGE2 at 1.0 and 3.0/zg/kg in two dogs did not affect PD values. Effect of Aspirin on Transmucosal Potential Difference
Intragastric administration of aspirin produced a significant and dose-dependent drop of gastric transmucosal PD values which was maintained during the test and recovery phase of the experiment (Figure 3). The test solution showed the presence of blood upon withdrawal from several dogs at the end of the test phase of the study. Effect of Prostaglandins and Aspirin on the Transmucosal Potential Difference
Intragastric administration of Me-PGE2 at 10/xg/ kg together with aspirin (20 raM) did not significantly modify the PD drop associated with aspirin (Figure 4). On the other hand, SC-29333 significantly attenuated the aspirin-induced drop in PD during both the test and recovery phases of the study (Figure 4). Intravenous administration of SC-29333 at 1.0/zg/ kg also decreased aspirin actions on the transmucosal PD (Figure 5). However, this decrease was
438
not statistically significant during the test session, but showed significant differences during the recovery phase. Side Effects
The intragastric and intravenous administration of Me-PGE2 caused retching, emesis, rhinorrhea, trembling, and diarrhea in some dogs. On the other hand, no grossly observable side effects were seen with intragastric administration of SC-29333, while retching was only occasionally observed with intravenous administration. DISCUSSION The existence of a gastric mucosal barrier is a concept that has been advanced to explain why gastric secretions do not normally attack or digest the gastric mucosa. The presence of the barrier allows the intact gastric mucosa to withstand a transmucosal concentration of hydrogen (H § ions of 106 : 1 (7, 8). The barrier to hydrogen ions back-diffusing has not been adequately characterized either physiologically or anatomically, but it is believed to be dependent on the intact gastric epithelial cells. Decreased turnover and increased loss of gastric Digestive Diseases, Vol. 23, No. 5 (May 1978)
E PROSTAGLANDINS
AND GASTRIC PD 7r
BASAL
I
TEST
I
RECOVERY
T[i :
sc
:i:::i j ....
~,~
.
~ T T r T T T W' 'L/ / / / / . J.
"
~
"'~+~'~.-" 3~
. . . .
l's
. . . .
3'o
.........
. T
";.T
0 4C
J
. . . .
,;s
. . . .
-c . . . . . i I iAsplrin(lOmM)
,----A,o,,,o~O~)
6'o
9~)
,'s . . . .
. . . .
TIME (3MIN INIERVALS) Fig 3. Effect of aspirin on gastric transmucosal potential differehce in the Heidenhain pouch dog. Aspirin was introduced into the pouch only during the test period. Values represent the mean + 1 SE obtained in Six dogs. Asterisks denote periods in which differences between control and aspirin series were statistically significant.
7G
BASAL
i
TEST
1
.RECOVERY
I
~ SO
z a.
:::I
......
irl
40
.'
1
r TT]'TI 3C
.
.
.
.
~ . 15
.
.
.
.
' 30
.
TiME
.
.
.
(3MIN
' 45
9
9
9
~
i 60
.
.
.
.
t 75
.
.
.
.
i 90
INTERVALS)
Fig 4. Effect of prostaglandins on aspirin-induced changes in gastric transmucosal potential difference in the dog. The prostaglandins (10 txg/kg) and aspirin (20 mM) were administered into the pouch during the test period and were withdrawn during the recovery period. Values represent the mean + 1 SE obtained in six dogs. Asterisks denote periods in which the difference between aspirin control and treated series were statistically significant. Digestive Diseases, Vol. 23, No. 5 (May 1978)
439
DAJANI ET AL 7C
BASAL
-"
o
l
TE~T
",t
. . . .
,~ . . . .
T!"
3b . . . .
4's . . . .
RECOV[RY
.
U'"
"--"~176
~o
1
T[
|
I
] , ~ k - a" "
r T;-
~
. . . .
,~ . . . .
9b
TIME 13 MIN INtERVAtS)
Fig 5. Effect of SC-29333 administered intravenously (1.0/xg/kg) on aspirin (20 mM) induced changes in the gastric transmucosal potential difference in the dog. Values represent the mean + 1 sE obtained in six dogs. Asterisks denote periods in which the difference between treated and control group were statistically significant.
epithelial cells disrupt the gastric mucosal barrier, leading to the formation of gastric erosions and hemorrhagic erosive gastritis (9). The functional metabolic state of the mucosal surface and oxyntic cells may play an important role in determining the quantity of hydrogen ion back-diffusion by the mucosal cells before ulceration occurs. Normal mucosal blood flow may completely clear small rates of back-diffusing H § and thereby protect the mucosa from its damaging effects, but when H § accumulates in the mucosa, ulceration may occur (8). Measurement of the gastric transmucosal potential difference is a sensitive and a convenient technique for evaluating the integrity of the gastric mucosal barrier (10, 11). Ulcerogenic drugs such as aspirin and ethanol lower gastric PD values, and studies performed in animals have demonstrated that a very good correlation exists between the changes in the ionic flux (Na § H +) and the lowering of the gastric transmucosal PD when the barrier is broken (4, 12). The present studies demonstrated that Me-PGE2 when administered locally into the pouch, but not intravenously, moderately but significantly lowered the transmucosal potential difference in the dog. This finding agrees with the work reported by
440
O'Brien and Carter (1), who have shown that MePGE2 disrupted the gastric mucosal barrier when locally but not intravenously administered in the Heidenhain pouch dog model. A moderate back-diffusion of H § and a large increase in the net movement of Na § into the pouch was observed by O'Brien and Carter (1) with local instillation of 20 /xg/kg Me-PGE2 into the pouch. However, in our studies, SC-29333, a chemically-related i6-methyl PGE1 analog did not lower the transmucosal PD value when locally administered into the pouch. In fact, the opposite was seen since SC-29333 slightly but significantly increased the gastric transmucosal PD when compared to the ethanol vehicle control series. The concentration of ethanol employed in the present study to dissolve the prostaglandins was 0.3% (v/v). This low concentration of ethanol was deliberately chosen since higher concentrations (814% v/v) were reported by Davenport (13) to disrupt the integrity of the gastric mucosa. With this low (0.3% v/v) concentration of ethanol, these studies demonstrated that the ethanol vehicle did not significantly lower the transmucosal PD values. Support for the potential protective effects of SC29333 on the canine gastric mucosa was provided by Digestive Diseases, Vol. 23, No. 5 (May 1978)
E PROSTAGLANDINS AND GASTRIC PD the interaction study with aspirin. It is well established that aspirin shares with alcohol and bile salts the capacity to disrupt the integrity of the gastric mucosal barrier, allowing back-diffusion of hydrogen ions and thereby causing acute erosive gastritis (14). In the present study the local intragastric instillation of aspirin at 10 and 20 mM concentrations produced a significant, dose-dependent drop in the gastric transmucosal potential difference. However, the concomitant local intragastric administration of SC-29333 significantly attenuated the aspirin-induced drop in transmucosal potential difference. In sharp contrast, the reference standard Me-PGE2 neither potentiated nor attenuated the effects of aspirin on the transmucosal PD in this model. It is not known whether the actions of SC-29333 on gastric mucosa are locally or systemically mediated. Therefore, we investigated the effects of SC29333, administered parenterally, on the PD drop induced by local instillation of aspirin. The experiment demonstrated that intravenous administration of SC-29333 attenuated and hastened the recovery of the transmucosal PD fall induced by aspirin. These experiments suggest that the effects of SC29333 on gastric PD has both local and systemic components. The specific contributions of each of these two mechanisms on the gastric transmucosa/ PD cannot be established from these studies. The effects of SC-29333 on the transmucosal gastric PD appear to be a property inherent for this prostaglandin and are not related to its gastric antisecretory actions. The basis for this belief is that both SC-29333 and Me-PGE2 were tested in the range of their effective gastric antisecretory doses in the dog. The intragastric (pouch) doses for SC29333 and Me-PGE2 employed in the present study were 10 and 30 /xg/kg. The oral IDs0 (a dose that reduces acid output by 50% in the histamine-stimulated Heidenhain pouch dog secretions) for MePGEz and SC-29333 were reported to be approximately 10/xg/kg (2, 3, 15). But, it should be pointed out that these doses may considerably exceed the doses essential to inhibit pouch secretion when PGs are directly administered into the pouch. Andersson and Nylander (16) reported that the intragastric instillation of Me-PGE2 at the low dose of 5/xg/dog (approximately equivalent to 0.5/xg/kg) directly into a Pavlov pouch significantly inhibited the pentagastrin-stimulated gastric secretion. Despite the use of such high doses of the prostaglandins administered locally into the pouch, Me-PGE2 slightly Digestive Diseases, Vol. 23, No. 5 (May 1978)
lowered the PD values while SC-29333 did not. This argument suggests that the actions of SC-29333 on the integrity of the gastric mucosal barrier may not be related to its gastric antisecretory effects. The use of the PD technique for investigation of the protective effects on gastric mucosa has been recently criticized (17, 18). Many of the problems associated with the interpretation of changes in gastric transmucosal PD reflect the assumption that change in PD always indicates change in the "barrier" properties of the gastric mucosa (19). However, many other factors also influence PD and include (20): gastric acid secretion, ionic fluxes, liquid junction potential, gastrointestinal hormones, cellular damage, and active transport. Thus, it is important that the contributions of these factors be considered in the final analysis of the results. Ivey et al (21) have reported that cimetidine, a histam i n e - H 2 - r e c e p t o r antagonist, significantly increased the gastric potential difference in adult volunteers. The increase in the PD observed in the presence of cimetidine was probably due to the inhibition of hydrogen ion secretion rather than an actual tightening of the barrier integrity (17). One could overcome this deficiency by supplying a high quantity of exogenous acid to negate the antisecretory component of the drug. In the presence of high exogenous acid concentrations, it has been reported that the administration of metiamide--another histamine-H2-receptor antagonist--at an effective gastric antisecretory dose did not alter the gastric mucosal barrier integrity in the canine Heidenhain pouch irrigated with 80 mM HCI solution (1). In our experiments with the dog, a high concentration of the exogenous acid (100 mM HC1) was provided to rule out any potential gastric antisecretory effects by the prostaglandins. The activity of SC-29333 in a high intragastric acid environment further supports the idea that its effect on the gastric mucosal barrier is independent of its gastric antisecretory property. The implication of the results obtained with MePGE2 is that high doses of this prostaglandin might cause gastritis. However, the presence of gastric acid seems to be necessary for the occurrence of acute ulcer but, since this prostaglandin is in fact an inhibitor of gastric acid secretion, the potential over-all adverse capacity is only a matter of speculation at this time. On the other hand an agent like SC-29333 may have important therapeutic potential both in the healing and, perhaps, in the prevention of benign gastric ulcer, gastritis, and stress ulcer.
441
DAJANI ET AL
ACKNOWLEDGMENTS The authors wish to thank Professor Paul Bass (University of Wisconsin) for reviewing the manuscript, Mr. Howard Mickley for technical assistance, Ms D. Rogan for secretarial services, and Ms Gerianne V~argason for editorial assistance.
REFERENCES 1. O'Brien PE, Carter DC: Effect of gastric secretory inhibitors on the gastric mucosal barrier. Gut 16:437-442, 1975 2. Dajani EZ, I)riskiI1 DR, Bianchi RG, Collins PW, Pappo R: SC-29333: A potent inhibitor of canine gastric secretion. Am J Dig Dis 21:1049-1057, 1976 3. Dajani EZ, Driskill DR, Bianchi RG, Collins PW, Pappo R: Studies on the biological properties of a pr0staglandin analog, SC-29333, a potent inhibitor of gastric secretion. Advances in Prostaglandin and Thrombox~:ne Research, Vol 2. B S~an~uelson, R Paoletti (eds). New York, R~aven Press, 1976, 944 4. Davenport HW, Warner HA, Code CF: Functional s~gnificance of gastric mucosal barrier to sodium. Gastroenterology 47:142-152, 1964 5. Cooke AR, Kienzle MG: Studie s of anti-inflammatory drugs and aliphatic alcohols on antral mucosa. Gastroenterology 66:56-62, 1974 6. Snedecor GW, Cochran WG: Statistical Methods, 6th ed, Ames, Iowa, Iowa State University, 1967 7. Davenport HW: Salicylate damage to the gastric mucosal barriei'. N Engl J Med 276:1307-1312, 1967 8. Skillman JJ, Silen ~: Stress ulceration in the acutely ill. Annu Rev Med 2719-22, 1976
442
9. Croft DN: Cell turnover and loss and the gastric mucosal barrier. Am J Dig Dis 22:383-386, 1977 10. Geall MG, Phillips SF, Summerskill WHJ: Profile of gastric potential difference in man. Effects of aspirin, a~lcohol, bile and endogenous acid. Gastroenterology 58:437-443 , 1970 11. Murray HS, Strottman MP, Cooke AR: Effect of several drugs on gastriC potential difference in man. Br Med J 1:1921, 1974 12. Chavsta TE, Cooke AR: The effect of several ulcerogenic drugs on the canine gastric mucosal barrier, J Lab Clin Med 79:302-315, 1972 13. Davenport HW: Ethanol damage to canine oxyntic glandular mucosa. Proc Soc Exp Biol Med 126:657-662, 1967 14. Ivey KJ: Gastric mucosal barrier. Gastroenterology 61:247257, 1957 15. Robert A, Lancaster C, Nezamis JE, Badalament JN: A gastric antisecretory and anti-ulcer prostaglandin with oral and long-acting activity. GaStroenterology 64:790, 1973 16. Andersson S, Nylander B: Local inhibitory action of 16,16dimethy! PGE2 on gastric acid secretion i n t h e dog. Advances in Prostag!andin and Thromboxane Research, Vol 2. B Samuelson, R Paoletti (eds). New york, Raven Press, !976, 943 17. Silen W: Cimetidine and gastric potential difference. Lancet 2:1314, 1975 18. Read NW, Levin RJ: Cimetidine and gastric potential difference. Lancet 2:1314, 1975 19. Thjodleifsson B, Wormsley KG: Back diffusion--fact or fiction? Digestion 15:53-72, 1977 20. Tarnawsk i A, Krause W, Ivey KJ: Drop in gastric potential difference after pentagastrin unrelated to oxyntic cell stimulation. Abstract presented at the 1977 American Gastroenterological Association, Toronto, Canada 21. Ivey KJ, Baskin W, Jeffrey G: Effect of cimetidine on gastric potential difference in man. Lancet 2:1072-1073, 1975
Digestive Diseases, Vol. 23, No. 5 (May 1978)
Measurement of the gastric transmucosal potential difference (PD) was used to study the effect of gastric antisecretory prostaglandins on the integrity of the gastric mucosa of the Heidenhain pouch dog. Intragastric administration ofSC-29333 [( +)-15-deoxy-16-o~,~-hydroxy-16-methyl PGEI methyl ester] slightly but significantly increased the transmucosal PD when compared to vehicle control. In addition, SC-29333 administered either intravenously or intragastrically, significantly inhibited the PD fall induced by aspirin, a well established barrier breaker. In contrast, the intragastric administration of l6,16-dimethyl PGE2 methyl ester (Me-PGEz) significantly lowered the transmucosal PD and failed to modify the actions of aspirin on the integrity of the gastric mucosa. However, the intravenous administration of either prostaglandin did not affect the basal transmucosal PD values. These studies suggest that SC-29333 may strengthen the integrity of the gastric mucosal barrier against aspirin, and this could have important therapeutic potential.
Synthetic prostaglandins (PGs) of the E type have been well established as inhibitors of gastric secretion and may be potentially useful in the treatment of peptic ulcer. The effects of these PGs on the integrity of the gastric mucosa has not been adequately defined. O'Brien and Carter (1) have recently shown that oral but not intravenous administration of 16,16-dimethyl PGE2 (Me-PGE2) significantly increased the ionic permeability of the gastric mucosa in the Heidenhain pouch dog. It is not known, however, whether the damaging effects of Me-PGE2 on the gastric mucosa is universally shared with other synthetic PGEs or if this action is limited to a specific prostaglandin. To test this hypothesis, we investigated the effects of SC-29333--a potent and a long-acting gastric antisecretory prostaglandin in animals (2, 3 ) - - o n the integrity of the canine gastric mucosal barrier. The aims of these studies were twofold: (1) to study the effects of SC-29333 and the reference standard Me-PGE2 (Figure 1) on the inFrom the Departments of Biological Research and Real Time Systems, Searle Laboratories, Chicago, Illinois. Address for reprint requests: Dr. Esam Z. Dajani, Gastrointestinal Pharmacology, Department of Biological Research, Searle Laboratories, Box 5110, Chicago, Illinois 60680.
436
tegrity of the gastric mucosal barrier as measured by the gastric transmucosal potential difference (PD) technique and (2) to investigate the interaction of PGs with known " b a r r i e r b r e a k e r s " such as aspirin on the integrity of the gastric mucosa.
MATERIALS AND METHODS Animals Six female mongrel dogs weighing between 12 and 18 kg were prepared with Heidenhain pouches drained by Gregory cannula. The dogs were trained to stand quietly in Pavlov stands and were conscious during all studies. A specially constructed cannula plug permitted instillation and withdrawal of test solutions. The cannula plug was also equipped with a side arm to permit passing of the detecting electrode to the pouch. The surgery was performed two months before these studies were started and the animals were not used more than twice per week. Drugs The prostagtandins were synthesized by P. W. Collins and M. Bruhn of the Department of Chemical Research of these laboratories. The prostaglandins were initially dissolved in absolute ethanol (10 mg/ml) and were subsequently diluted with sodium phosphate buffer (pH 7.4). The final alcohol concentrations when administered to the Digestive Diseases, Vol. 23, No. 5 (May 1978)
0002-9211/78/0500-0436505.00/1 9 1978DigestiveDiseaseSystems,Inc.
E PROSTAGLANDINS AND GASTRIC PD O
~.T...-~COOCH3 ~" ] CH 3 SC-29333 [(+)-IS-DEOXY-16 c,, ~ HYDROXY, I6-METh~L PGE1 METHYL ESTER]
~"
~
~COOCH-~
Me-PGE 2 [16, t6-Dm~m~L m;E2 METAL ESTIm]
At the beginning of each experiment the pouch was thoroughly rinsed with a basal solutionof 100 mM HCI and 54 mM NaC1. The osmolarity of this solution and all subsequent test solutions was kept constant at 285-300 mosmol/ liter by adjusting the concentration of NaC1 within the solution. The pouch was then drained and 20 ml of basal solution was introduced into the pouch. At the end of a 30-rain basal period the pouch was drained and refilled with test solution containing the basal solution plus prostaglandins (10 or 30 tzg/kg) or aspirin (10 or 20 raM) or a combination of prostaglandin and aspirin. At the end of the test period, the pouch was drained and refilled with basal solution to study the recovery pattern. A cross-over experimental design was used to quantitatively assess the effect of prostaglandins on the transmucosal PD values. The differences in the means for each 3-min interval at each test condition was determined by comparing the values obtained in treated animals at various times with those obtained from the same animals under control or placebo conditions using the paired Student's t test at P < 0.05 (6). Throughout the studies, the dogs were closely observed for gross side effects, if any.
RESULTS
Fig 1. Chemical structures of SC-29333 and Me-PGE2.
Effect of Prostaglandins on the Gastric Transmucosal Potential Difference pouch did not exceed 0.3%. Prostaglandin solutions were stored at - 10~ C when not in use. Aspirin was purchased from Sigma Chemical Company, St. Louis, Missouri.
Procedure The technique used to measure the transmucosal PD was essentially similar to that reported by Davenport et al (4) and Cooke and Kienzle (5). The reference electrode consisted of a commercial intravenous infusion set (Butterfly 21, Abbott Laboratories, Chicago, Illinois) filled at the time of the experiment with 3% agar in saturated KC1 solution and then introduced into a peripheral leg vein. The detecting electrode consisted of a polyethylene tube (PE 60, ID = 0.030 in.) filled with 3% agar in saturated KC1 solution. The detecting electrode was inserted into the pouch through the side arm of the cannula plug. Each electrode was led through the side arm of a 25-ml suction flask containing saturated KC1 in which one of a pair of balanced calomel (Corning Glass Work, Medfield, Massachussetts) cells were placed. The calomel cells were connected to an electrometer amplifier (Burr Brown, model 3431, Tucson, Arizona), followed by several stages of amplifiers (Texas Instruments, model SN72741, Dallas, Texas). These electronic arrangements allowed zero adjustment, inversion, and buffering of the transmucosal PD signal for display on a chart recorder and digital voltmeter (Datel, model DM 2000 AR, Canton, Massachussetts). The transmucosal potential difference was measured to the nearest 0.1 millivolt (mV) at 3-rain intervals. Experiments were initiated by depriving the dogs of food, but not water, for 18 hr. Each experiment consisted of three 30-min periods: basal, test, and recovery periods. Digestive Diseases, Vol. 23, No. 5 (May 1978)
The gastric transmucosal potential difference in control studies was - 5 8 + 0.5 mV (mean -+ SE), mucosa negative with respect to serosa. The intragastric instillation of the PG vehicle, which contained no more than 0.3% ethanol (v/v) did not significantly lower the PD values. The administration of PGs into the gastric pouch produced p r o n o u n c e d changes in gastric PD values when compared to basal or control period. As shown in Figure 2, Me-PGE2 produced dose-dependent and significant reduction in PD only for one period during the test session. H o w e v e r , the PD values did not return during the recovery phase when Me-PGE2 was administered at 30/~g/kg dose, while the lower dose of 10/zg/kg showed recovery pattern similar to the control. In sharp contrast, the intragastric (pouch) administration of SC-29333 at 10 and 30/xg/kg doses significantly increased the gastric PD when compared to the vehicle (0.3% v/v ethanol). This increase in PD was maintained even during the withdrawal or recovery phase of the study. H o w e v e r , the PD values associated with SC-29333 intragastric administration during the test and recovery phase did not significantly differ from the basal phase of this study. This basal phase was consistently maintained at - 5 8 mV for the entire 90-rain duration (Figure 3).
437
DAJANI ET AL ?0
BASAL
l
TEST
l
t
RECOVERY l . 9 ,L.~%.~.-.=" ,..4P-..9"""
~
. .'. g g.-.-'...,:...,'...,'"~"~"
9 ,,....w "~_~.,~--~
_~
6C
.-~'*-~"~'~-~--*
.,r"
f.. t.. ' i
E
~
-
~
p-~,
Z
~../
,,s
J
-
r
~
"- " ,
~,-,~/ ,
~
-
_
.~--:ff' ',-
u....
'"'~...~'..-o ....
. . . o"-~' "~>...~.~.-~ 9
.tr-= 9
*
e - - - - e vehi(:le (Cor, t roll ~ , - - - - , L S C - 2 9 3 3 3 (lOmco/kg)
Z ua
e......-eS C - 2 9 3 3 3 (30rr~g/kg) ~r - - -A. Me - ~ E 2 (lOmcg/kg) a~......OMe-PC, E2 (30mcg/kg)
3C
i
,
,
,
i
,
.
.
15
.
i
.
.
.
.
30 TIME
,
.
.
.
.
45
i
60
.
.
.
.
i
75
.
.
.
.
i
90
(3 M0N INTERVALSI
Fig 2. Effects of prostaglandins on the gastric transmucosal potential difference in the Heidenhain pouch dog. The prostaglandins (10 and 30/zg/kg) were introduced into the pouch in the test period and were withdrawn in the recovery period. Values represent means obtained in six dogs. Asterisks denote periods in which the difference between control (vehicle) and treated series were significantly different.
Intravenous administration of either SC-29333 or Me-PGE2 at 1.0 and 3.0/zg/kg in two dogs did not affect PD values. Effect of Aspirin on Transmucosal Potential Difference
Intragastric administration of aspirin produced a significant and dose-dependent drop of gastric transmucosal PD values which was maintained during the test and recovery phase of the experiment (Figure 3). The test solution showed the presence of blood upon withdrawal from several dogs at the end of the test phase of the study. Effect of Prostaglandins and Aspirin on the Transmucosal Potential Difference
Intragastric administration of Me-PGE2 at 10/xg/ kg together with aspirin (20 raM) did not significantly modify the PD drop associated with aspirin (Figure 4). On the other hand, SC-29333 significantly attenuated the aspirin-induced drop in PD during both the test and recovery phases of the study (Figure 4). Intravenous administration of SC-29333 at 1.0/zg/ kg also decreased aspirin actions on the transmucosal PD (Figure 5). However, this decrease was
438
not statistically significant during the test session, but showed significant differences during the recovery phase. Side Effects
The intragastric and intravenous administration of Me-PGE2 caused retching, emesis, rhinorrhea, trembling, and diarrhea in some dogs. On the other hand, no grossly observable side effects were seen with intragastric administration of SC-29333, while retching was only occasionally observed with intravenous administration. DISCUSSION The existence of a gastric mucosal barrier is a concept that has been advanced to explain why gastric secretions do not normally attack or digest the gastric mucosa. The presence of the barrier allows the intact gastric mucosa to withstand a transmucosal concentration of hydrogen (H § ions of 106 : 1 (7, 8). The barrier to hydrogen ions back-diffusing has not been adequately characterized either physiologically or anatomically, but it is believed to be dependent on the intact gastric epithelial cells. Decreased turnover and increased loss of gastric Digestive Diseases, Vol. 23, No. 5 (May 1978)
E PROSTAGLANDINS
AND GASTRIC PD 7r
BASAL
I
TEST
I
RECOVERY
T[i :
sc
:i:::i j ....
~,~
.
~ T T r T T T W' 'L/ / / / / . J.
"
~
"'~+~'~.-" 3~
. . . .
l's
. . . .
3'o
.........
. T
";.T
0 4C
J
. . . .
,;s
. . . .
-c . . . . . i I iAsplrin(lOmM)
,----A,o,,,o~O~)
6'o
9~)
,'s . . . .
. . . .
TIME (3MIN INIERVALS) Fig 3. Effect of aspirin on gastric transmucosal potential differehce in the Heidenhain pouch dog. Aspirin was introduced into the pouch only during the test period. Values represent the mean + 1 SE obtained in Six dogs. Asterisks denote periods in which differences between control and aspirin series were statistically significant.
7G
BASAL
i
TEST
1
.RECOVERY
I
~ SO
z a.
:::I
......
irl
40
.'
1
r TT]'TI 3C
.
.
.
.
~ . 15
.
.
.
.
' 30
.
TiME
.
.
.
(3MIN
' 45
9
9
9
~
i 60
.
.
.
.
t 75
.
.
.
.
i 90
INTERVALS)
Fig 4. Effect of prostaglandins on aspirin-induced changes in gastric transmucosal potential difference in the dog. The prostaglandins (10 txg/kg) and aspirin (20 mM) were administered into the pouch during the test period and were withdrawn during the recovery period. Values represent the mean + 1 SE obtained in six dogs. Asterisks denote periods in which the difference between aspirin control and treated series were statistically significant. Digestive Diseases, Vol. 23, No. 5 (May 1978)
439
DAJANI ET AL 7C
BASAL
-"
o
l
TE~T
",t
. . . .
,~ . . . .
T!"
3b . . . .
4's . . . .
RECOV[RY
.
U'"
"--"~176
~o
1
T[
|
I
] , ~ k - a" "
r T;-
~
. . . .
,~ . . . .
9b
TIME 13 MIN INtERVAtS)
Fig 5. Effect of SC-29333 administered intravenously (1.0/xg/kg) on aspirin (20 mM) induced changes in the gastric transmucosal potential difference in the dog. Values represent the mean + 1 sE obtained in six dogs. Asterisks denote periods in which the difference between treated and control group were statistically significant.
epithelial cells disrupt the gastric mucosal barrier, leading to the formation of gastric erosions and hemorrhagic erosive gastritis (9). The functional metabolic state of the mucosal surface and oxyntic cells may play an important role in determining the quantity of hydrogen ion back-diffusion by the mucosal cells before ulceration occurs. Normal mucosal blood flow may completely clear small rates of back-diffusing H § and thereby protect the mucosa from its damaging effects, but when H § accumulates in the mucosa, ulceration may occur (8). Measurement of the gastric transmucosal potential difference is a sensitive and a convenient technique for evaluating the integrity of the gastric mucosal barrier (10, 11). Ulcerogenic drugs such as aspirin and ethanol lower gastric PD values, and studies performed in animals have demonstrated that a very good correlation exists between the changes in the ionic flux (Na § H +) and the lowering of the gastric transmucosal PD when the barrier is broken (4, 12). The present studies demonstrated that Me-PGE2 when administered locally into the pouch, but not intravenously, moderately but significantly lowered the transmucosal potential difference in the dog. This finding agrees with the work reported by
440
O'Brien and Carter (1), who have shown that MePGE2 disrupted the gastric mucosal barrier when locally but not intravenously administered in the Heidenhain pouch dog model. A moderate back-diffusion of H § and a large increase in the net movement of Na § into the pouch was observed by O'Brien and Carter (1) with local instillation of 20 /xg/kg Me-PGE2 into the pouch. However, in our studies, SC-29333, a chemically-related i6-methyl PGE1 analog did not lower the transmucosal PD value when locally administered into the pouch. In fact, the opposite was seen since SC-29333 slightly but significantly increased the gastric transmucosal PD when compared to the ethanol vehicle control series. The concentration of ethanol employed in the present study to dissolve the prostaglandins was 0.3% (v/v). This low concentration of ethanol was deliberately chosen since higher concentrations (814% v/v) were reported by Davenport (13) to disrupt the integrity of the gastric mucosa. With this low (0.3% v/v) concentration of ethanol, these studies demonstrated that the ethanol vehicle did not significantly lower the transmucosal PD values. Support for the potential protective effects of SC29333 on the canine gastric mucosa was provided by Digestive Diseases, Vol. 23, No. 5 (May 1978)
E PROSTAGLANDINS AND GASTRIC PD the interaction study with aspirin. It is well established that aspirin shares with alcohol and bile salts the capacity to disrupt the integrity of the gastric mucosal barrier, allowing back-diffusion of hydrogen ions and thereby causing acute erosive gastritis (14). In the present study the local intragastric instillation of aspirin at 10 and 20 mM concentrations produced a significant, dose-dependent drop in the gastric transmucosal potential difference. However, the concomitant local intragastric administration of SC-29333 significantly attenuated the aspirin-induced drop in transmucosal potential difference. In sharp contrast, the reference standard Me-PGE2 neither potentiated nor attenuated the effects of aspirin on the transmucosal PD in this model. It is not known whether the actions of SC-29333 on gastric mucosa are locally or systemically mediated. Therefore, we investigated the effects of SC29333, administered parenterally, on the PD drop induced by local instillation of aspirin. The experiment demonstrated that intravenous administration of SC-29333 attenuated and hastened the recovery of the transmucosal PD fall induced by aspirin. These experiments suggest that the effects of SC29333 on gastric PD has both local and systemic components. The specific contributions of each of these two mechanisms on the gastric transmucosa/ PD cannot be established from these studies. The effects of SC-29333 on the transmucosal gastric PD appear to be a property inherent for this prostaglandin and are not related to its gastric antisecretory actions. The basis for this belief is that both SC-29333 and Me-PGE2 were tested in the range of their effective gastric antisecretory doses in the dog. The intragastric (pouch) doses for SC29333 and Me-PGE2 employed in the present study were 10 and 30 /xg/kg. The oral IDs0 (a dose that reduces acid output by 50% in the histamine-stimulated Heidenhain pouch dog secretions) for MePGEz and SC-29333 were reported to be approximately 10/xg/kg (2, 3, 15). But, it should be pointed out that these doses may considerably exceed the doses essential to inhibit pouch secretion when PGs are directly administered into the pouch. Andersson and Nylander (16) reported that the intragastric instillation of Me-PGE2 at the low dose of 5/xg/dog (approximately equivalent to 0.5/xg/kg) directly into a Pavlov pouch significantly inhibited the pentagastrin-stimulated gastric secretion. Despite the use of such high doses of the prostaglandins administered locally into the pouch, Me-PGE2 slightly Digestive Diseases, Vol. 23, No. 5 (May 1978)
lowered the PD values while SC-29333 did not. This argument suggests that the actions of SC-29333 on the integrity of the gastric mucosal barrier may not be related to its gastric antisecretory effects. The use of the PD technique for investigation of the protective effects on gastric mucosa has been recently criticized (17, 18). Many of the problems associated with the interpretation of changes in gastric transmucosal PD reflect the assumption that change in PD always indicates change in the "barrier" properties of the gastric mucosa (19). However, many other factors also influence PD and include (20): gastric acid secretion, ionic fluxes, liquid junction potential, gastrointestinal hormones, cellular damage, and active transport. Thus, it is important that the contributions of these factors be considered in the final analysis of the results. Ivey et al (21) have reported that cimetidine, a histam i n e - H 2 - r e c e p t o r antagonist, significantly increased the gastric potential difference in adult volunteers. The increase in the PD observed in the presence of cimetidine was probably due to the inhibition of hydrogen ion secretion rather than an actual tightening of the barrier integrity (17). One could overcome this deficiency by supplying a high quantity of exogenous acid to negate the antisecretory component of the drug. In the presence of high exogenous acid concentrations, it has been reported that the administration of metiamide--another histamine-H2-receptor antagonist--at an effective gastric antisecretory dose did not alter the gastric mucosal barrier integrity in the canine Heidenhain pouch irrigated with 80 mM HCI solution (1). In our experiments with the dog, a high concentration of the exogenous acid (100 mM HC1) was provided to rule out any potential gastric antisecretory effects by the prostaglandins. The activity of SC-29333 in a high intragastric acid environment further supports the idea that its effect on the gastric mucosal barrier is independent of its gastric antisecretory property. The implication of the results obtained with MePGE2 is that high doses of this prostaglandin might cause gastritis. However, the presence of gastric acid seems to be necessary for the occurrence of acute ulcer but, since this prostaglandin is in fact an inhibitor of gastric acid secretion, the potential over-all adverse capacity is only a matter of speculation at this time. On the other hand an agent like SC-29333 may have important therapeutic potential both in the healing and, perhaps, in the prevention of benign gastric ulcer, gastritis, and stress ulcer.
441
DAJANI ET AL
ACKNOWLEDGMENTS The authors wish to thank Professor Paul Bass (University of Wisconsin) for reviewing the manuscript, Mr. Howard Mickley for technical assistance, Ms D. Rogan for secretarial services, and Ms Gerianne V~argason for editorial assistance.
REFERENCES 1. O'Brien PE, Carter DC: Effect of gastric secretory inhibitors on the gastric mucosal barrier. Gut 16:437-442, 1975 2. Dajani EZ, I)riskiI1 DR, Bianchi RG, Collins PW, Pappo R: SC-29333: A potent inhibitor of canine gastric secretion. Am J Dig Dis 21:1049-1057, 1976 3. Dajani EZ, Driskill DR, Bianchi RG, Collins PW, Pappo R: Studies on the biological properties of a pr0staglandin analog, SC-29333, a potent inhibitor of gastric secretion. Advances in Prostaglandin and Thrombox~:ne Research, Vol 2. B S~an~uelson, R Paoletti (eds). New York, R~aven Press, 1976, 944 4. Davenport HW, Warner HA, Code CF: Functional s~gnificance of gastric mucosal barrier to sodium. Gastroenterology 47:142-152, 1964 5. Cooke AR, Kienzle MG: Studie s of anti-inflammatory drugs and aliphatic alcohols on antral mucosa. Gastroenterology 66:56-62, 1974 6. Snedecor GW, Cochran WG: Statistical Methods, 6th ed, Ames, Iowa, Iowa State University, 1967 7. Davenport HW: Salicylate damage to the gastric mucosal barriei'. N Engl J Med 276:1307-1312, 1967 8. Skillman JJ, Silen ~: Stress ulceration in the acutely ill. Annu Rev Med 2719-22, 1976
442
9. Croft DN: Cell turnover and loss and the gastric mucosal barrier. Am J Dig Dis 22:383-386, 1977 10. Geall MG, Phillips SF, Summerskill WHJ: Profile of gastric potential difference in man. Effects of aspirin, a~lcohol, bile and endogenous acid. Gastroenterology 58:437-443 , 1970 11. Murray HS, Strottman MP, Cooke AR: Effect of several drugs on gastriC potential difference in man. Br Med J 1:1921, 1974 12. Chavsta TE, Cooke AR: The effect of several ulcerogenic drugs on the canine gastric mucosal barrier, J Lab Clin Med 79:302-315, 1972 13. Davenport HW: Ethanol damage to canine oxyntic glandular mucosa. Proc Soc Exp Biol Med 126:657-662, 1967 14. Ivey KJ: Gastric mucosal barrier. Gastroenterology 61:247257, 1957 15. Robert A, Lancaster C, Nezamis JE, Badalament JN: A gastric antisecretory and anti-ulcer prostaglandin with oral and long-acting activity. GaStroenterology 64:790, 1973 16. Andersson S, Nylander B: Local inhibitory action of 16,16dimethy! PGE2 on gastric acid secretion i n t h e dog. Advances in Prostag!andin and Thromboxane Research, Vol 2. B Samuelson, R Paoletti (eds). New york, Raven Press, !976, 943 17. Silen W: Cimetidine and gastric potential difference. Lancet 2:1314, 1975 18. Read NW, Levin RJ: Cimetidine and gastric potential difference. Lancet 2:1314, 1975 19. Thjodleifsson B, Wormsley KG: Back diffusion--fact or fiction? Digestion 15:53-72, 1977 20. Tarnawsk i A, Krause W, Ivey KJ: Drop in gastric potential difference after pentagastrin unrelated to oxyntic cell stimulation. Abstract presented at the 1977 American Gastroenterological Association, Toronto, Canada 21. Ivey KJ, Baskin W, Jeffrey G: Effect of cimetidine on gastric potential difference in man. Lancet 2:1072-1073, 1975
Digestive Diseases, Vol. 23, No. 5 (May 1978)
