- Email: [email protected]
Effect of Propranolol on Ricinoleic Acid- and Deoxycholic Acid-Induced Changes of Intestinal Electrolyte Movement and Mucosal Permeability
00 16-fiOH'ii7H/7 'i0·1-0fi6&j;O~. 00/0 GASTitOf:NTf:HOI.OGY 7fi :66H-67:l. l!J7H Copyright ..., 197H hy thl' American Gastroenlerological Association
Vol. 75 , No.4
Printed in U.SA.
EFFECT OF PROPRANOLOL ON RICINOLEIC ACID- AND DEOXYCHOLIC ACID-INDUCED CHANGES OF INTESTINAL ELECTROLYTE MOVEMENT AND MUCOSAL PERMEABILITY
Evidence against the importance of altered permeability in the production of fluid and electrolyte accumulation HENRY J . BINDER, M.D., JoHN W. DoBBINS, M.D., LoRRAINE AND DIANNE
S.
C.
RAcusEN, M.D.,
WHITING
Department of' Internal Medicine , Yale University , New Haven , Connecticut
Hydroxy fatty acids and bile acids produce both intestinal fluid and electr?l_Yte accumulation and increases in inulin clearance, a parameter of mucosal permeab1hty. The relationship of the changes in mucosal permeability to the production of fluid and electrolyte accumulation is uncertain. These experiments were designed to determine whether the alterations of mucosal permeability produced by ricinoleic acid and deoxycholic acid were related to the production of hydroxy fatty acid- and bile acidinduced fluid and electrolyte accumulation in the rat colon. Propranolol (1 mg per 100 g of body weight) administered daily for 3 days inhibited ricinoleic acid- and deoxycholic acid-induced Na and water accumulation. In contrast, propranolol did not affect either the increase in inulin clearance or the decrease in electrical potential difference produced by ricinoleic acid and deoxycholic acid. Further, amphotericin B increased inulin clearance by the colon and also increased water and N a absorption. These studies suggest that changes in mucosal permeability are not primarily responsible for hydroxy fatty acid- and bile acid-induced fluid and Na accumulation. Bile acids and hydroxy fatty acids significantly affect intestinal function in several different ways. 1- 21 Perfusion of either the small or large intestine with an isosmolar solution containing either a dihydroxy bile acid or a hydroxy fatty acid results in a reversal of net fluid and electrolyte absorption to that of net fluid and electrolyte accumulation. l-a. 12- 14 (The term fluid and electrolyte accumulation rather than secretion is used to describe the addition of water and electrolyte to a solution perfusing a segment of intestine. The term secretion is restricted to designate active transport processes (~.g., that which is stimulated by cholera enterotoxin.)) During such perfusions several other alterations of intestinal function occur: an increase in mucosal permeability, an alteration in mucosal morphology, a change in intestinal motor activity, an alterReceived November 28, 1977. Accepted April 24 , 1978. Address requests for reprints to: Henry J Binder, M.D. , Department oflnternal Medicine, Yale University School of Medicine, New Haven, Connecticut 06510. This study was supported by Research Grant AM-14669 from the National Institute of Arthritis, Metabolism and Digestive Diseases, United States Public Health Service, and a grant from The John A. Hartford Foundation, Inc. Dr. Dobbins is the recipient of Clinical Investigator Award AM00232 and Dr. Racusen of National Research Award AM-05389 both from the National Institute of Arthritis, Metabolism and Digestive Diseases.
ation in active ion transport processes, and an increase in mucosal cyclic AMP.4- 11 • 1;;...21 The relative contributions of these alterations of function to the genesis of fluid and electrolyte transport changes have yet to be defined. Bile acids and hydroxy fatty acids increase intestinal content and adenylate cyclase cyclic AMP activity .10 • 1 1• 19 • 20 • 22 Although it is generally accepted that cholera enterotoxin produces fluid and electrolyte accumulation by stimulating mucosal adenylate cyclase and the resulting increase in mucosal cyclic AMP activates active ion secretion by mechanisms that are, as yet, unknown, 23 it is less certain whether similar mechanisms explain the fluid and electrolyte accumulation produced by bile acids and hydroxy fatty acids. A series of recent experiments with propranolol has provided additional evidence to suggest that bile acids and hydroxy fatty acids induce luminal fluid and electrolyte accumulation secondary to increased mucosal cyclic AMP.24 • 25 In these studies propranolol inhibited bile acid and hydroxy fatty acid stimulation of both fluid accumulation and mucosal adenylate cyclase activity. It is unknown whether other alterations of intestinal function produced by bile acids and hydroxy fatty acids are also inhibited by propranolol. Existing evidence clearly indicates that bile acids and hydroxy fatty acids produce alterations in mucosal permeability. 6 • 7 ' 12 ' 18 • 21 Increased clearance of inulin,
668
October 1978
FATTY ACIDS. BILE ACIDS . AND MUCOSAL PERMEABILITY
a measure of mucosal permeability, occurs during perfusion of either the colon or jejunum with ricinoleic acid, a hydroxy fatty acid, and an increase in urea and oxalate absorption is also observed during perfusion with both deoxycholic acid and ricinoleic acid. 1;· 1H Finally, an increase in absorption oflow molecular weight polyethylene glycol has been reported during perfusion of the rabbit large intestine with ricinoleic acid. 21 It has recently been suggested that the fluid and electrolyte accumulation produced by hydroxy fatty acids and bile acids is mediated by these changes in mucosal permeability,18 but it is uncertain whether an increase in mucosal permeability alone can ever provide a driving force for fluid and electrolyte accumulation. These present studies, designed to evaluate the possible importance of alteration in mucosal permeability in the genesis of bile acid- and hydroxy fatty acidinduced fluid and electrolyte accumulation, were divided into two portions. First, the effect of propranolol on bile acid- and hydroxy fatty acid-induced changes in mucosal permeability was assessed. Second, the effect on colonic fluid and electrolyte movement during perfusion with amphotericin B, a compound which in other tissues alters mucosal permeability, was also ascertained. 21;-28 Our results demonstrated that propranolol administration completely inhibits sodium accumulation produced by bile acids and hydroxy fatty acids, but that propranolol did not alter the associated increase in mucosal permeability; we also observed an increase in fluid and electrolyte absorption during amphotericin B perfusion. These studies suggest that the alterations in mucosal permeability produced by bile acids and hydroxy fatty acids are independent of their effects on fluid and electrolyte movement. Materials and Methods Nonfasting male Sprague Dawley nits weighing between 225 and 300 g were used in all experiments. Fluid and electrolyte movement in the entire colon was determined by a continuous perfusion technique as previously described.<;, 12 In all experiments fluid and electrolyte movement or inulin clearance was determined during an initial control (saline) period which was then followed by an experimental period. The composition of the control solution (in millimoles per liter) was Na , 140; K, 5; Cl , 120; HCOa, 25. Labeled and unlabeled polyethylene glycol (approximately 10,000 counts per min per ml and 5 g per liter, respectively) was added to all solutions as a nonabsorbable marker. The infusion rate was approximately 0.54 ml per min in all experiments. An equilibrium period of 80 min was followed by four 20-min study periods. The experimental period in which 6 mM ricinoleic acid, 3 mM deoxycholic acid, or amphotericin B (100 tJ-g per ml) was added to the perfusing solution was then performed in an identical fashion . At the conclusion of the experiment the colon was removed, placed in an oven at 105°C for 18 hr, and then weighed. Positive numbers represent net absorption; negative ones represent net accumulation. Water and Na movement are expressed as microliters and microequivalents per minute per gram of dry tissue weight, respectively. In the inulin clearance studies 25 tJ-Ci of l:'H]inulin was injected intravenously after ligation of both renal pedicles, and the concentration of inulin in plasma and colonic effiuent was then determined. Colonic clearance was calculated by standard formulae and expressed as microliters per minute
669
per gram of dry tissue weight. Transmural electrical potential difference (PD) was measured during the clearance experiments. In these studies PE 205 bridges containing 4 g of agar per dl of perfusion solution were placed in the rectum 4 em above the anus and in the peritoneal cavity and attached to balanced calomel half cells. The PD was measured by a high impedance potentiometer. In all of these studies the effect of the experimental agent (amphotericin B, deoxycholic acid, or ricinoleic acid) on fluid and electrolyte movement, inulin cl earance, and PD was compared to the control period in the same animal. In the deoxycholic acid and the ricinoleic acid studies the effect of propranolol on water , Na, and K movement, inulin clearance, and PD was also measured . Propranolol (1 mg per 100 g of body weight) was injected intraperitoneally daily for 3 days. The control animals received an equivalent volume of saline. The perfusion and clearance studies were then performed 24 hr after the last injection. All results were expressed as mean ± SEM. Paired and unpaired Student's t-test was employed as indicated. 2n
Results Ricinoleic acid studies (table 1). During perfusion with the saline solution in the control animals there was water and sodium absorption and potassium accumulation. The administration of propranolol (1 mg per 100 g of body weight for 3 days) did not result in any change in water, sodium, and potassium movement during the saline control period (table 1). In the control animals perfusion with 6 mM ricinoleic acid resulted in net fluid accumulation ( -36.7 ± 2.1 ILl per min per g of dry tissue weight) and in net sodium accumulation ( -4.8 ± 2.6 ILEq per min per g of dry tissue weight) . In contrast, perfusion of the colon of animals treated with propranolol with the 6 mM ricinoleic acid solution did not result in either water or sodium accumulation (table 1). Sodium movement was identical in the propranololtreated animals during both the saline and ricinoleic acid perfusions (18.4 ± 0.9 versus 16.4 ± 1.9 ILEq per min per g of dry tissue weight; P > 0.4). Propranolol partially reversed the water accumulation produced by ricinoleic acid so that water movement in the propranolol-treated animals during perfusion with the hydroxy fatty acid was significantly different from water movement both during the saline period and during ricinoleic acid pefusion in the control animals. In the control animals ricinoleic acid altered water and Na movement by 122 ± 19 ILl per min per g of dry tissue weight and 22.8 ± 3.5 ILEq per min per g of dry tissue weight. In contrast, the decrease in water and N a movement induced by ricinoleic acid in the propranolol-treated animals was significantly less (63 ± 10 ILl per min per g of dry tissue weight (P < 0.02) and 2.1 ± 0.9 ILEq per min per g of dry tissue weight (P < 0.001)). Potassium accumulation increased during perfusion with 6 mM ricinoleic acid concomitant with fluid movement into the lumen (table 1). Administration of propranolol resulted in an inhibition of this increased potassium accumulation so that potassium movement during perfusion with ricinoleic acid was similar to that observed during the saline perfusions. Determination of inulin clearance by the colon and measurement of the PD was also performed in the control and propranolol-treated animals. Perfusion with
670
V ol . 75 , No.4
BINDER ET AL.
ricinoleic acid resulted in a significant increase in inulin clearance in the control animals from 5.1 ± 0.9 to 19.2 ± 5.6 f.Ll per min per g of dry tissue weight (P < 0.01) (table 1). There was also a significant decrease in the mean PD during the ricinoleic acid perfusions ( -19.0 ± 1.5 compared to -11.9 ± 1.4 mv (Iuman negative) (P < 0.05)) . These ricinoleic acid-induced changes in inulin cl earance and in PD persisted unchanged in the propranolol-treated animals (table 1), a contrast to the results of fluid and electrolyte movement in th e propra nolol-treated animals. Thus, perfusion with ricinoleic acid resulted in an increase in inulin cl earance and a decrease in PD in the propranololtreated animals, changes that were identical to those observed in the control animals . Deoxycholic acid studies (table 2) . Deoxycholic acid also resulted in a reversal of water and sodium absorption to net water and sodium accumulation in control animals. Propranolol decreased water and sodium abTAB LE
sorption during the saline perfusion in the deoxycholic acid studies although propranolol did not affect fluid and sodium movement during the saline perfusion in the ricinoleic acid studies. In a manner identical to the ricinoleic acid studies, propranolol inhibited water and sodium accumulation during perfusion with 3 mM deoxycholic acid. In the propranolol-treated animals sodium absorption was identical during the control and the deoxycholic acid perfusion (12.5 ± 2.6 versus 13.2 ± 3.2 fl-Eq per min per g of dry tissue weight; P > 0.4). Water movement during bile acid perfusion in the propranolol animals was significantly less than that observed during the saline perfusion but was also significantly different from the deoxycholic acid study in the control animals. Potassium accumulation increased during the bil.e acid perfusions in the control but not in the propranolol-treated animals (table 2). Inulin clearance increased and PD decreased in the control animals during perfusion with deoxycholic acid
1. Effect of propranolol on 6 mM ricinoleic acid-induced changes of the movement of water , sod ium and potassium , inulin clearance, and potential difference (PD) " Wa ter
Na
JJ.l/min·g dry wt
Control (8 , 7)" Sa line Ri cinoleic acid p Propra nolol (8, 7)1' Sa line Ricinoleic acid p
85. 2 ± 2.1 -36.7 ± 2.1 < 0.001
70.5 ± 17. 0 7.0 ± 9.5'' < 0.01
K p.Eq /min ·g dry wt
Inulin clearance
PD
JJ.ilmin ·g dry wt
mu
18.0 ± 2.1 -4.8 ± 2.6 < 0.001
-1.1±0.2 -3.5 ± 0.5 < 0.001
5.1 ± 0.9 19.2 ± 5.6 < 0.05
-19.0 ± 1.6 -11.9±1.4 < 0.001
16.4 ± 2.0 18.4 ± 1.0 11 NS'
-1.1±0.2 -1.3 ± 0.1" NS
7.6 ± 0.6 20.9 ± 3.7c < 0.01
- 20 .9 ± 1.0 -14.6 ± 0.9" < 0.001
" Mean ± SEM. Propranolol-trea ted anima ls received 1 mg per 100 g of body weight daily for 3 days and control animals received saline injections. All anima ls were initia lly perfused with a Ringer-like solution (saline) followed by the ricinoleic acid solution as described in the text . Studi es of fluid a nd electrolyte movem ent were performed in one group of animals and those of inulin clearance and PD in another . Positive numbers represent net absorption; negative ones represent net secretion. PD is expressed as lumen negative with respective to peritoneum . " First number in parenthesis represents number of animals in the studies of fluid and electrolyte movement; second number represents number of animals in the inulin cleara nce and PD studies. " P < 0.001 compared to ricinoleic acid period in the control a nimals. 1 ' P < 0.005 compa r ed to ricinoleic acid per iod in the control a nimals. ,. Not signifi cantly different from the ricinoleic a cid peri od in the control animals. I NS, not significa nt. TABLE
2. Effect of propranolol on 3 mM d eoxycholic acid-induced changes of the movement of water , sodium , and potassium, inulin
clearance, and potential difference (PD)'' Water p.l/mi n ·g dry wt
Control (8, 7)1' Saline Deoxycholic acid p P ropranolol (8, 7) 1' Sa line Deoxycholic acid p
105.3 ± 30.9 - 79. 7 ± 13.4 < 0.001
56.3 ± 10.6 8. 5 ± 18.8C < 0.05
Na
K p.Eq/m in ·g dry wt
20.7 ± 4.5 -15.3 ± 1.8 < 0.001
12.5 ± 2.6 13. 2 ± 3.3" NS'
Inulin clearance
PD
JJ.i lmin ·g dry wt
mu
-1.0±0.1 -3 .5 ± 0.2 < 0.001
6.2 ± 0.5 17.3 ± 3. 5 < 0.01
-19.1±1.3 -12.3 ± 0.7 < 0.001
-1.2 ± 0.2 -1.5 ± 0.1d NS
9.9 ± 1.0 20.4 ± 3.5" < 0.05
-20.8 ± 1.0 -13.2 ± 0.7' < 0.001
:. · " See footnote a and b of table 1 :or details: Perfusion with Ringer-like solution was followed by deoxycholic acid perfusion. P < 0.005 compared to deoxycholic ac1d penod in the control animals . " P < 0.001 compar ed to deoxycholic acid period in the control anima ls. ,. Not significantly differen t from ricinoleic acid period in the control a nimals. 1 NS , not significant.
671
FATTY ACIDS, BILE ACIDS, AND MUCOSAL PERMEABILITY
October 1978
steatorrhea and ileal disease respectively.:n-3 a These present studies confirm that perfusion of the colon with hydroxy fatty acids or bile acids results in a reversal of fluid and electrolyte movement from net absorption to net accumulation. 1-a, 12- 14 There is disagreement regarding the mechanism by which the alteration of fluid and electrolyte movement by bile acids and hydroxy fatty acids occurs in vivo. Stimulation of ion secretory processes, possibly mediated by cyclic AMP, or alteration of mucosal permeability have recently been considered as the two most likely explanations for the observed changes in fluid and electrolyte movement. Several recent studies clearly indicate that mucosal permeabilty is altered during perfusion with either hydroxy fatty acids or bile acids. For example, when the rat colon is perfused with ricinoleic acid, an increase in inulin clearance has been observed, 12 and perfusion of the hamster jejunum with ricinoleic acid is associated with increased clearance of both inulin and dextran. 18 Further, perfusion of the rabbit colon with ricinoleic acid results in an increase in absorption of low molecular weight polyethylene glycol. 21 Exposure of the colon to deoxycholic acid also results in increased mucosal permeabilty: an increase in urea absorption and increased oxalate clearance occurs. 6 Our present and previous experiments demonstrate that hydroxy fatty acids and bile acids increase mucosal permeability. It has not been established if an increase in mucosal permeabilty alone can alter net fluid and electrolyte movement. These studies provide compelling evidence that altered mucosal permeability is not solely
(table 2) and this was not altered by pretreatment with propranolol. Inulin clearance increased in the propranolol-treated animals by 10.5 ± 2.9 p,l per min per g of dry tissue weight compared to an increase of 11.0 ± 3.4 p,l per min per g of dry tissue weight in the control animals. Deoxycholic acid produced similar decreases in PD in both the propranolol and the control animals (7.6 ± 0.8 versus 6.8 ± 1.2 mv). Amphotericin B studies. Figure 1 demonstrates that both inulin clearance and urea clearance by the colon were significantly increased during perfusion with amphotericin B compared to perfusion with the saline solution. Urea clearance was greater than inulin clearance during both the control and the amphotericin B periods. This is consistent with previous observations demonstrating that clearance by the intestine of substances with different molecular weights is inversely related to their size. 30 Parallel studies assessed the effect of amphotericin B on fluid and electrolyte movement. In these experiments a significant increase in both water (139.5 ± 21.6 p,l per min per g of dry tissue weight; P < 0.01) and sodium absorption (15.2 ± 4.0 p,Eq per min per g of dry tissue weight; P < 0.01) occurred during the amphotericin B perfusion period compared to the control period. Discussion Both hydroxy fatty acids and bile acids are the active ingredients of several laxative products and have been implicated as mediators for the diarrhea associated with 50
UREA CLEARANCE
INULIN CLEARANCE
40
100
>- 30 ....
75
:J "C
co ....... c
E ....... 20
50
:1..
p
p <0.02 N = 6
10
Control
N
25
Ampholericin
8
Control
=8
Amphotericin
8
FIG. 1. Effect of amphotericin B on mucosal permeability. Significant increase in both inulin clearance and urea clearance by the colon occurred during perfusion with amphotericin B (100 J.l.g perm!) .
672
Vol. 75 , No.4
BINDER ET AL.
re,;ponsible for the changes in fluid and electrolyte movement associated with exposure of the intestinal epithelium to either hydroxy fatty acids or bile acids. Propranolol has been previously used to inhibit bile acid- and hydroxy fatty acid-induced fluid and electrolyte movement in the rabit colon.~-1. 2 '' We have confirmed these previous observations and have extended these findings to delineate the effect of this agent on the associated changes in inulin clearance (i.e., mucosal permeability). Administration of propranolol for 3 days inhibited water and sodium accumulation but did not change the increase in inulin clearance produced by either ricinoleic acid or deoxycholic acid. The changes in permeability are therefore not primarily responsible for the subsequent changes in electrolyte transport because the inhibition in fluid and electrolyte accumulation by propranolol is not associated with a parallel inhibition of either bile acid- or hydroxy fatty acidmediated increases in inulin clearance. Although a decrease in PD may represent one of several events, an increase in permeability may be reflected by a decrease in PD. Propranolol also did not prevent the decrease in PD induced by either ricinoleic acid or deoxycholic acid. These studies do not exclude the possibility that increased permeability may contribute to the fluid accumulation. It is unlikely that increased mucosal permeability is the primary or sole factor responsible for the observed changes in electrolyte movement. Propranolol completely inhibited Na secretion induced by ricinoleic acid and deoxycholic acid but only partially inhibited the changes in water movement. There is no adequate explanation of the partial reversal of water movement. It appears that in the propranololtreated animals ricinoleic acid is associated with nonisosmolar transport. We cannot exclude the possibility that increased mucosal permeability may in part be responsible for this decrease in net water absorption. Propranolol may have an unusual effect in the secreting intestine especially inasmuch as we do not know the mechanism of propranolol's action in these experiments. Although propranolol is a f3 adrenergic antagonist, it is doubtful that the mechanism by which it inhibits fluid and electrolyte accumulation is related to f3 adrenergic blockade. f3 agonists in the colon do not stimulate ion secretion,a4 and epinephrine, a mixed adrenergic agonist, neither increases adenylate cyclase activity nor cyclic AMP levels in colonic mucosa. 3 5 Both deoxycholic acid and ricinoleic acid increased potassium accumulation in the control animals but not in the propranolol-treated animals. Changes in potassium movement are often linked to increases in PD .3 H The observed alterations in potassium transport in these studies are most likely not related to the PD. Potassium accumulation increased during bile acid and hydroxy fatty acid perfusions which were associated with a decrease in PD (tables 1 and 2). Propranolol inhibited these changes in potassium movement but did not prevent the decrease in PD produced by deoxycholic acid or ricinoleic acid. It is likely that the increase in potassium accumulation was secondary to the changes in water and sodium movement because the changes in
potassium movement paralleled the changes in both sodium and water movement. Osmotically induced fluid accumulation is also associated with increased movement of potassium into the colonic lumen (D. S. Whiting and H. J. Binder, unpublished observations). In the amphotericin B studies mucosal permeability was altered by a drug which is known to affect epithelial permeability of the toad bladder, renal tubule, and jejunum. 21r- 2H Although the changes in permeability produced by amphotericin B are most likely dissimilar to the changes in permeability produced by detergents such as ricinoleic acid and deoxycholic acid, perfusion with amphotericin B is also associated with increases in intestinal clearance. Similar increases in intestinal clearance were observed by Rhode and Chen in studies of the canine jejunum using amphotericin B. 28 During amphotericin B perfusion a change in fluid and electrolyte movement was also observed, but this was an increase not a decrease in net water and sodium absorption. Therefore, these amphotericin B experiments permit us to suggest that changes in permeabilty, at least as shown by an increase in inulin and urea clearance, are not necessarily associated with net fluid accumulation. In conclusion, perfusion of the intestine with either ricinoleic acid or deoxycholic acid is associated with fluid and electrolyte accumulation and increased mucosal permeability. These experiments indicate that the observed increases in mucosal permeabilty are hot directly nor solely responsible for the changes in fluid and electrolyte movement associated with bile acids and hydroxy fatty acids. REFERENCES 1. Mekhjian HS , Phillips SF, Hofmann AF: Colonic secretion of
water and electrolytes induced by bile acids: perfusion studies in man. J Clin Invest 50:1569-1577 , 1971 2. KragE, Phillips SF: Active and passive bile acid absorption in man. Perfusion studies of the ileum and jejunum. J Clin Invest 53:1686-1694, 1974
3. Wingate DL, Phillips SF, Hofmann AF: Effect of glycine-conjugated bile acids with and without lecithin on water and glucose absorption in perfused human jejunum. J Clin Invest 52:12301236, 1973
4. Teem MW, Phillips SF: Perfusion of the hamster jejunum with conjugated and unconjugated bile acids: inhibition of water absorption and effects on morphology. Gastroenterology 62:261267, 1972
5. Saunders DR, Hedges JR, Sillery J, et a!: Morphological and functional effects of bile salts on rat colon. Gastroenterology 68:1236-1245, 1975
6. Dobbins JW, Binder HJ: Effect of bile salts and fatty acids on the colonic absorption of oxalate. Gastroenterology 70:1096-1100, 1976
7. Rummel W, Nell G , Wanitschke R: Action mechanisms of antiabsorptive and hydragogue drugs. In Intestinal Absorption and Malabsorption . Edited by TZ Czaky. New York, Raven Press, 1975, p 209-227 8. Kirwan WO, Smith AN, Mitchell WD, et a!: Bile acids and colonic motility in the rabbit and the human. I. The rabbit. Gut 16:894-902, 1975
9. Binder HJ, Rawlins CL: Effect of conjugated dihydroxy bile salts on electrolyte transport in rat colon. J Clin Invest 52:1460-1466, 1973
Octoberl.978
FATTY ACIDS. BILE ACIDS, AND MUCOSAL PERMEABILITY
10. Binder HJ, Filburn C, Volpe BT: Bile salt alteration of colonic
electrolyte transport: role ol' cyclic adenosine monophosphate. Gastroenterology 68:503-508, 1975 11. Conl ey DR, Coyne MJ, Bonorri s GG, et al: Bile acid stimulation of colonic adenylate cyclase and secretion in the rabbit. Am J Dig Dis 21:453-458, 1976 12. Bright-Asa re P, Binder HJ : Stimulation of colonic secretion of water and electrolytes by hydroxy fatty acids. Gastroenterology 64:81-88 , 1973 13. Ammon HV, Phillips SF: Inhibition of il eal water absorption by
intraluminal fatty acids. Influence of chain length, hydroxylation, and conjugation of fatty acids. J Clin Invest 53:205-210, 1974 14. Ammon HV , Thomas PH , Phillips SF: Effects of oleic and
15.
16.
17.
18.
19.
20.
21.
ricinoleic acid on net jejuna l water and electrolyte movement. J Clin Invest 53:374-379, 1974 Stewart JJ , Gaginella TS, Olsen WA , et a l: Inhibitory actions of laxatives on motility and water and electrolyte transport in the gastrointestinal tract. J Pharmacal Exp Ther 192:458-467 , 1975 Stewart JJ , Gaginella TS, Bass P: Actions of ricinoleic acid and structurally related fatty acids on the gastrointestinal tract. I. Effects on the smooth muscle contractility in vitro. J Pharmacal Exp Ther 195:347-354, 1975 Christensen J, Freeman BW: Circular muscle electromyogram in the rat colon: local effect of sodium ricinoleate. Gastroenterology 63:1011-1015, 1972 Cline WS, Lorenzsonn V, Benz L, et al: The effects of sodium ricinoleate on small intestinal function and structure. J Clin Invest 58:380-390, 1976 Binder HJ: Cyclic adenosine monophosphate controls bile salt and hydroxy fatty acid-induced colonic electrolyte secretion (abstr). J Clin Invest 53:7-8, 1974 Binder HJ, Dobbins JW, Whiting DS: Evidence against importance of altered mucosal permeability in ricinoleic acid induced fluid secretion (abstr). Gastroenterology 72:1029, 1977 Gaginella TS, Chadwick VS, Debongnie JC, et a!: Perfusion of rabbit colon with ricinoleic acid: dose-related mucosal injury, fluid secretion, and increased permeability. Gastroenterology
673
73:95-101 , 1977 22. Mertens RB , Mayer SE , Wheeler HO: Effect of conjugated bile
acids on cyclic AMP levels in rabbit ileal mucosa (abstr). Gastroenterology 70:919, 1976 23. Finkelstein RA: Cholera. CRC Crit Rev Microbial 2:552-623, 1973 24 . Conley DR, Coyne MJ , Chung A, et a!: Propranolol inhibits
adenylate cyclase and secretion stimulated by DCA in the rabbit colon. Gastroenterology 71:72-75, 1976 25. Feinstein JD , Coyne MJ , Bonorris GG, et al: Bile acid and fatty acid effect on intestinal cyclic-AMP, Na+-K +-ATPase and secretion (abstr). Gastroenterology 72:1183, 1977 26. Lichtenstein NS, Leaf A: Effect of amphotericin B on the permeability of the toad bladder. J Clin Invest 44:1328-1342 , 1965 27. Stroup RF , Weinman E, Hayslett JP, et a!: Effect of luminal
28. 29. 30.
31. 32. 33. 34. 35.
36.
permeability on net transport across the amphibian proximal tubule. Am J Physiol 226:1110-1116, 1974 Rohde JE, Chen LC: Permeabilty and selectivity of canine and human jejunum during cholera. Gut 13:191-196, 1972 Snedecor GW, Cochran WG: Statistical Methods. Sixth edition. Ames, Iowa State University Press, 1967 Loehry CA, Kingham J , Baker J : Small intestinal permeability in a nimals and man. Gut 14:683-688, 1973 Binder HJ: Fecal fatty acids- mediators of diarrhea? Gastroent erology 65:847-850, 1973 Binder HJ, Donowitz M: A new look a t laxative action. Gastroenterology 69:1001-1005, 1975 Binder HJ: Pharmacology of laxatives. Annu Rev Pharmacal Toxicol 17:355-367 , 1977 Racusen L, Binder HJ: Catecholamines alter colonic ion transport (abstr) . Gastroenterology 72:1115, 1977 Coyne MJ, Bonorris GG, Chung A, et al: Inhibition by propranolol of bile acid stimulation of rabbit colonic adenylate cyclase in vitro. Gastroenterology 71:68-71 , 1976 Fisher KA , Binder HJ, Hayslett JP: Potassium secretion by colonic mucosal cells after potassium adaptation. Am J Physiol 231:987-994, 1976
Vol. 75 , No.4
Printed in U.SA.
EFFECT OF PROPRANOLOL ON RICINOLEIC ACID- AND DEOXYCHOLIC ACID-INDUCED CHANGES OF INTESTINAL ELECTROLYTE MOVEMENT AND MUCOSAL PERMEABILITY
Evidence against the importance of altered permeability in the production of fluid and electrolyte accumulation HENRY J . BINDER, M.D., JoHN W. DoBBINS, M.D., LoRRAINE AND DIANNE
S.
C.
RAcusEN, M.D.,
WHITING
Department of' Internal Medicine , Yale University , New Haven , Connecticut
Hydroxy fatty acids and bile acids produce both intestinal fluid and electr?l_Yte accumulation and increases in inulin clearance, a parameter of mucosal permeab1hty. The relationship of the changes in mucosal permeability to the production of fluid and electrolyte accumulation is uncertain. These experiments were designed to determine whether the alterations of mucosal permeability produced by ricinoleic acid and deoxycholic acid were related to the production of hydroxy fatty acid- and bile acidinduced fluid and electrolyte accumulation in the rat colon. Propranolol (1 mg per 100 g of body weight) administered daily for 3 days inhibited ricinoleic acid- and deoxycholic acid-induced Na and water accumulation. In contrast, propranolol did not affect either the increase in inulin clearance or the decrease in electrical potential difference produced by ricinoleic acid and deoxycholic acid. Further, amphotericin B increased inulin clearance by the colon and also increased water and N a absorption. These studies suggest that changes in mucosal permeability are not primarily responsible for hydroxy fatty acid- and bile acid-induced fluid and Na accumulation. Bile acids and hydroxy fatty acids significantly affect intestinal function in several different ways. 1- 21 Perfusion of either the small or large intestine with an isosmolar solution containing either a dihydroxy bile acid or a hydroxy fatty acid results in a reversal of net fluid and electrolyte absorption to that of net fluid and electrolyte accumulation. l-a. 12- 14 (The term fluid and electrolyte accumulation rather than secretion is used to describe the addition of water and electrolyte to a solution perfusing a segment of intestine. The term secretion is restricted to designate active transport processes (~.g., that which is stimulated by cholera enterotoxin.)) During such perfusions several other alterations of intestinal function occur: an increase in mucosal permeability, an alteration in mucosal morphology, a change in intestinal motor activity, an alterReceived November 28, 1977. Accepted April 24 , 1978. Address requests for reprints to: Henry J Binder, M.D. , Department oflnternal Medicine, Yale University School of Medicine, New Haven, Connecticut 06510. This study was supported by Research Grant AM-14669 from the National Institute of Arthritis, Metabolism and Digestive Diseases, United States Public Health Service, and a grant from The John A. Hartford Foundation, Inc. Dr. Dobbins is the recipient of Clinical Investigator Award AM00232 and Dr. Racusen of National Research Award AM-05389 both from the National Institute of Arthritis, Metabolism and Digestive Diseases.
ation in active ion transport processes, and an increase in mucosal cyclic AMP.4- 11 • 1;;...21 The relative contributions of these alterations of function to the genesis of fluid and electrolyte transport changes have yet to be defined. Bile acids and hydroxy fatty acids increase intestinal content and adenylate cyclase cyclic AMP activity .10 • 1 1• 19 • 20 • 22 Although it is generally accepted that cholera enterotoxin produces fluid and electrolyte accumulation by stimulating mucosal adenylate cyclase and the resulting increase in mucosal cyclic AMP activates active ion secretion by mechanisms that are, as yet, unknown, 23 it is less certain whether similar mechanisms explain the fluid and electrolyte accumulation produced by bile acids and hydroxy fatty acids. A series of recent experiments with propranolol has provided additional evidence to suggest that bile acids and hydroxy fatty acids induce luminal fluid and electrolyte accumulation secondary to increased mucosal cyclic AMP.24 • 25 In these studies propranolol inhibited bile acid and hydroxy fatty acid stimulation of both fluid accumulation and mucosal adenylate cyclase activity. It is unknown whether other alterations of intestinal function produced by bile acids and hydroxy fatty acids are also inhibited by propranolol. Existing evidence clearly indicates that bile acids and hydroxy fatty acids produce alterations in mucosal permeability. 6 • 7 ' 12 ' 18 • 21 Increased clearance of inulin,
668
October 1978
FATTY ACIDS. BILE ACIDS . AND MUCOSAL PERMEABILITY
a measure of mucosal permeability, occurs during perfusion of either the colon or jejunum with ricinoleic acid, a hydroxy fatty acid, and an increase in urea and oxalate absorption is also observed during perfusion with both deoxycholic acid and ricinoleic acid. 1;· 1H Finally, an increase in absorption oflow molecular weight polyethylene glycol has been reported during perfusion of the rabbit large intestine with ricinoleic acid. 21 It has recently been suggested that the fluid and electrolyte accumulation produced by hydroxy fatty acids and bile acids is mediated by these changes in mucosal permeability,18 but it is uncertain whether an increase in mucosal permeability alone can ever provide a driving force for fluid and electrolyte accumulation. These present studies, designed to evaluate the possible importance of alteration in mucosal permeability in the genesis of bile acid- and hydroxy fatty acidinduced fluid and electrolyte accumulation, were divided into two portions. First, the effect of propranolol on bile acid- and hydroxy fatty acid-induced changes in mucosal permeability was assessed. Second, the effect on colonic fluid and electrolyte movement during perfusion with amphotericin B, a compound which in other tissues alters mucosal permeability, was also ascertained. 21;-28 Our results demonstrated that propranolol administration completely inhibits sodium accumulation produced by bile acids and hydroxy fatty acids, but that propranolol did not alter the associated increase in mucosal permeability; we also observed an increase in fluid and electrolyte absorption during amphotericin B perfusion. These studies suggest that the alterations in mucosal permeability produced by bile acids and hydroxy fatty acids are independent of their effects on fluid and electrolyte movement. Materials and Methods Nonfasting male Sprague Dawley nits weighing between 225 and 300 g were used in all experiments. Fluid and electrolyte movement in the entire colon was determined by a continuous perfusion technique as previously described.<;, 12 In all experiments fluid and electrolyte movement or inulin clearance was determined during an initial control (saline) period which was then followed by an experimental period. The composition of the control solution (in millimoles per liter) was Na , 140; K, 5; Cl , 120; HCOa, 25. Labeled and unlabeled polyethylene glycol (approximately 10,000 counts per min per ml and 5 g per liter, respectively) was added to all solutions as a nonabsorbable marker. The infusion rate was approximately 0.54 ml per min in all experiments. An equilibrium period of 80 min was followed by four 20-min study periods. The experimental period in which 6 mM ricinoleic acid, 3 mM deoxycholic acid, or amphotericin B (100 tJ-g per ml) was added to the perfusing solution was then performed in an identical fashion . At the conclusion of the experiment the colon was removed, placed in an oven at 105°C for 18 hr, and then weighed. Positive numbers represent net absorption; negative ones represent net accumulation. Water and Na movement are expressed as microliters and microequivalents per minute per gram of dry tissue weight, respectively. In the inulin clearance studies 25 tJ-Ci of l:'H]inulin was injected intravenously after ligation of both renal pedicles, and the concentration of inulin in plasma and colonic effiuent was then determined. Colonic clearance was calculated by standard formulae and expressed as microliters per minute
669
per gram of dry tissue weight. Transmural electrical potential difference (PD) was measured during the clearance experiments. In these studies PE 205 bridges containing 4 g of agar per dl of perfusion solution were placed in the rectum 4 em above the anus and in the peritoneal cavity and attached to balanced calomel half cells. The PD was measured by a high impedance potentiometer. In all of these studies the effect of the experimental agent (amphotericin B, deoxycholic acid, or ricinoleic acid) on fluid and electrolyte movement, inulin cl earance, and PD was compared to the control period in the same animal. In the deoxycholic acid and the ricinoleic acid studies the effect of propranolol on water , Na, and K movement, inulin clearance, and PD was also measured . Propranolol (1 mg per 100 g of body weight) was injected intraperitoneally daily for 3 days. The control animals received an equivalent volume of saline. The perfusion and clearance studies were then performed 24 hr after the last injection. All results were expressed as mean ± SEM. Paired and unpaired Student's t-test was employed as indicated. 2n
Results Ricinoleic acid studies (table 1). During perfusion with the saline solution in the control animals there was water and sodium absorption and potassium accumulation. The administration of propranolol (1 mg per 100 g of body weight for 3 days) did not result in any change in water, sodium, and potassium movement during the saline control period (table 1). In the control animals perfusion with 6 mM ricinoleic acid resulted in net fluid accumulation ( -36.7 ± 2.1 ILl per min per g of dry tissue weight) and in net sodium accumulation ( -4.8 ± 2.6 ILEq per min per g of dry tissue weight) . In contrast, perfusion of the colon of animals treated with propranolol with the 6 mM ricinoleic acid solution did not result in either water or sodium accumulation (table 1). Sodium movement was identical in the propranololtreated animals during both the saline and ricinoleic acid perfusions (18.4 ± 0.9 versus 16.4 ± 1.9 ILEq per min per g of dry tissue weight; P > 0.4). Propranolol partially reversed the water accumulation produced by ricinoleic acid so that water movement in the propranolol-treated animals during perfusion with the hydroxy fatty acid was significantly different from water movement both during the saline period and during ricinoleic acid pefusion in the control animals. In the control animals ricinoleic acid altered water and Na movement by 122 ± 19 ILl per min per g of dry tissue weight and 22.8 ± 3.5 ILEq per min per g of dry tissue weight. In contrast, the decrease in water and N a movement induced by ricinoleic acid in the propranolol-treated animals was significantly less (63 ± 10 ILl per min per g of dry tissue weight (P < 0.02) and 2.1 ± 0.9 ILEq per min per g of dry tissue weight (P < 0.001)). Potassium accumulation increased during perfusion with 6 mM ricinoleic acid concomitant with fluid movement into the lumen (table 1). Administration of propranolol resulted in an inhibition of this increased potassium accumulation so that potassium movement during perfusion with ricinoleic acid was similar to that observed during the saline perfusions. Determination of inulin clearance by the colon and measurement of the PD was also performed in the control and propranolol-treated animals. Perfusion with
670
V ol . 75 , No.4
BINDER ET AL.
ricinoleic acid resulted in a significant increase in inulin clearance in the control animals from 5.1 ± 0.9 to 19.2 ± 5.6 f.Ll per min per g of dry tissue weight (P < 0.01) (table 1). There was also a significant decrease in the mean PD during the ricinoleic acid perfusions ( -19.0 ± 1.5 compared to -11.9 ± 1.4 mv (Iuman negative) (P < 0.05)) . These ricinoleic acid-induced changes in inulin cl earance and in PD persisted unchanged in the propranolol-treated animals (table 1), a contrast to the results of fluid and electrolyte movement in th e propra nolol-treated animals. Thus, perfusion with ricinoleic acid resulted in an increase in inulin cl earance and a decrease in PD in the propranololtreated animals, changes that were identical to those observed in the control animals . Deoxycholic acid studies (table 2) . Deoxycholic acid also resulted in a reversal of water and sodium absorption to net water and sodium accumulation in control animals. Propranolol decreased water and sodium abTAB LE
sorption during the saline perfusion in the deoxycholic acid studies although propranolol did not affect fluid and sodium movement during the saline perfusion in the ricinoleic acid studies. In a manner identical to the ricinoleic acid studies, propranolol inhibited water and sodium accumulation during perfusion with 3 mM deoxycholic acid. In the propranolol-treated animals sodium absorption was identical during the control and the deoxycholic acid perfusion (12.5 ± 2.6 versus 13.2 ± 3.2 fl-Eq per min per g of dry tissue weight; P > 0.4). Water movement during bile acid perfusion in the propranolol animals was significantly less than that observed during the saline perfusion but was also significantly different from the deoxycholic acid study in the control animals. Potassium accumulation increased during the bil.e acid perfusions in the control but not in the propranolol-treated animals (table 2). Inulin clearance increased and PD decreased in the control animals during perfusion with deoxycholic acid
1. Effect of propranolol on 6 mM ricinoleic acid-induced changes of the movement of water , sod ium and potassium , inulin clearance, and potential difference (PD) " Wa ter
Na
JJ.l/min·g dry wt
Control (8 , 7)" Sa line Ri cinoleic acid p Propra nolol (8, 7)1' Sa line Ricinoleic acid p
85. 2 ± 2.1 -36.7 ± 2.1 < 0.001
70.5 ± 17. 0 7.0 ± 9.5'' < 0.01
K p.Eq /min ·g dry wt
Inulin clearance
PD
JJ.ilmin ·g dry wt
mu
18.0 ± 2.1 -4.8 ± 2.6 < 0.001
-1.1±0.2 -3.5 ± 0.5 < 0.001
5.1 ± 0.9 19.2 ± 5.6 < 0.05
-19.0 ± 1.6 -11.9±1.4 < 0.001
16.4 ± 2.0 18.4 ± 1.0 11 NS'
-1.1±0.2 -1.3 ± 0.1" NS
7.6 ± 0.6 20.9 ± 3.7c < 0.01
- 20 .9 ± 1.0 -14.6 ± 0.9" < 0.001
" Mean ± SEM. Propranolol-trea ted anima ls received 1 mg per 100 g of body weight daily for 3 days and control animals received saline injections. All anima ls were initia lly perfused with a Ringer-like solution (saline) followed by the ricinoleic acid solution as described in the text . Studi es of fluid a nd electrolyte movem ent were performed in one group of animals and those of inulin clearance and PD in another . Positive numbers represent net absorption; negative ones represent net secretion. PD is expressed as lumen negative with respective to peritoneum . " First number in parenthesis represents number of animals in the studies of fluid and electrolyte movement; second number represents number of animals in the inulin cleara nce and PD studies. " P < 0.001 compared to ricinoleic acid period in the control a nimals. 1 ' P < 0.005 compa r ed to ricinoleic acid per iod in the control a nimals. ,. Not signifi cantly different from the ricinoleic a cid peri od in the control animals. I NS, not significa nt. TABLE
2. Effect of propranolol on 3 mM d eoxycholic acid-induced changes of the movement of water , sodium , and potassium, inulin
clearance, and potential difference (PD)'' Water p.l/mi n ·g dry wt
Control (8, 7)1' Saline Deoxycholic acid p P ropranolol (8, 7) 1' Sa line Deoxycholic acid p
105.3 ± 30.9 - 79. 7 ± 13.4 < 0.001
56.3 ± 10.6 8. 5 ± 18.8C < 0.05
Na
K p.Eq/m in ·g dry wt
20.7 ± 4.5 -15.3 ± 1.8 < 0.001
12.5 ± 2.6 13. 2 ± 3.3" NS'
Inulin clearance
PD
JJ.i lmin ·g dry wt
mu
-1.0±0.1 -3 .5 ± 0.2 < 0.001
6.2 ± 0.5 17.3 ± 3. 5 < 0.01
-19.1±1.3 -12.3 ± 0.7 < 0.001
-1.2 ± 0.2 -1.5 ± 0.1d NS
9.9 ± 1.0 20.4 ± 3.5" < 0.05
-20.8 ± 1.0 -13.2 ± 0.7' < 0.001
:. · " See footnote a and b of table 1 :or details: Perfusion with Ringer-like solution was followed by deoxycholic acid perfusion. P < 0.005 compared to deoxycholic ac1d penod in the control animals . " P < 0.001 compar ed to deoxycholic acid period in the control anima ls. ,. Not significantly differen t from ricinoleic acid period in the control a nimals. 1 NS , not significant.
671
FATTY ACIDS, BILE ACIDS, AND MUCOSAL PERMEABILITY
October 1978
steatorrhea and ileal disease respectively.:n-3 a These present studies confirm that perfusion of the colon with hydroxy fatty acids or bile acids results in a reversal of fluid and electrolyte movement from net absorption to net accumulation. 1-a, 12- 14 There is disagreement regarding the mechanism by which the alteration of fluid and electrolyte movement by bile acids and hydroxy fatty acids occurs in vivo. Stimulation of ion secretory processes, possibly mediated by cyclic AMP, or alteration of mucosal permeability have recently been considered as the two most likely explanations for the observed changes in fluid and electrolyte movement. Several recent studies clearly indicate that mucosal permeabilty is altered during perfusion with either hydroxy fatty acids or bile acids. For example, when the rat colon is perfused with ricinoleic acid, an increase in inulin clearance has been observed, 12 and perfusion of the hamster jejunum with ricinoleic acid is associated with increased clearance of both inulin and dextran. 18 Further, perfusion of the rabbit colon with ricinoleic acid results in an increase in absorption of low molecular weight polyethylene glycol. 21 Exposure of the colon to deoxycholic acid also results in increased mucosal permeabilty: an increase in urea absorption and increased oxalate clearance occurs. 6 Our present and previous experiments demonstrate that hydroxy fatty acids and bile acids increase mucosal permeability. It has not been established if an increase in mucosal permeabilty alone can alter net fluid and electrolyte movement. These studies provide compelling evidence that altered mucosal permeability is not solely
(table 2) and this was not altered by pretreatment with propranolol. Inulin clearance increased in the propranolol-treated animals by 10.5 ± 2.9 p,l per min per g of dry tissue weight compared to an increase of 11.0 ± 3.4 p,l per min per g of dry tissue weight in the control animals. Deoxycholic acid produced similar decreases in PD in both the propranolol and the control animals (7.6 ± 0.8 versus 6.8 ± 1.2 mv). Amphotericin B studies. Figure 1 demonstrates that both inulin clearance and urea clearance by the colon were significantly increased during perfusion with amphotericin B compared to perfusion with the saline solution. Urea clearance was greater than inulin clearance during both the control and the amphotericin B periods. This is consistent with previous observations demonstrating that clearance by the intestine of substances with different molecular weights is inversely related to their size. 30 Parallel studies assessed the effect of amphotericin B on fluid and electrolyte movement. In these experiments a significant increase in both water (139.5 ± 21.6 p,l per min per g of dry tissue weight; P < 0.01) and sodium absorption (15.2 ± 4.0 p,Eq per min per g of dry tissue weight; P < 0.01) occurred during the amphotericin B perfusion period compared to the control period. Discussion Both hydroxy fatty acids and bile acids are the active ingredients of several laxative products and have been implicated as mediators for the diarrhea associated with 50
UREA CLEARANCE
INULIN CLEARANCE
40
100
>- 30 ....
75
:J "C
co ....... c
E ....... 20
50
:1..
p
p <0.02 N = 6
10
Control
N
25
Ampholericin
8
Control
=8
Amphotericin
8
FIG. 1. Effect of amphotericin B on mucosal permeability. Significant increase in both inulin clearance and urea clearance by the colon occurred during perfusion with amphotericin B (100 J.l.g perm!) .
672
Vol. 75 , No.4
BINDER ET AL.
re,;ponsible for the changes in fluid and electrolyte movement associated with exposure of the intestinal epithelium to either hydroxy fatty acids or bile acids. Propranolol has been previously used to inhibit bile acid- and hydroxy fatty acid-induced fluid and electrolyte movement in the rabit colon.~-1. 2 '' We have confirmed these previous observations and have extended these findings to delineate the effect of this agent on the associated changes in inulin clearance (i.e., mucosal permeability). Administration of propranolol for 3 days inhibited water and sodium accumulation but did not change the increase in inulin clearance produced by either ricinoleic acid or deoxycholic acid. The changes in permeability are therefore not primarily responsible for the subsequent changes in electrolyte transport because the inhibition in fluid and electrolyte accumulation by propranolol is not associated with a parallel inhibition of either bile acid- or hydroxy fatty acidmediated increases in inulin clearance. Although a decrease in PD may represent one of several events, an increase in permeability may be reflected by a decrease in PD. Propranolol also did not prevent the decrease in PD induced by either ricinoleic acid or deoxycholic acid. These studies do not exclude the possibility that increased permeability may contribute to the fluid accumulation. It is unlikely that increased mucosal permeability is the primary or sole factor responsible for the observed changes in electrolyte movement. Propranolol completely inhibited Na secretion induced by ricinoleic acid and deoxycholic acid but only partially inhibited the changes in water movement. There is no adequate explanation of the partial reversal of water movement. It appears that in the propranololtreated animals ricinoleic acid is associated with nonisosmolar transport. We cannot exclude the possibility that increased mucosal permeability may in part be responsible for this decrease in net water absorption. Propranolol may have an unusual effect in the secreting intestine especially inasmuch as we do not know the mechanism of propranolol's action in these experiments. Although propranolol is a f3 adrenergic antagonist, it is doubtful that the mechanism by which it inhibits fluid and electrolyte accumulation is related to f3 adrenergic blockade. f3 agonists in the colon do not stimulate ion secretion,a4 and epinephrine, a mixed adrenergic agonist, neither increases adenylate cyclase activity nor cyclic AMP levels in colonic mucosa. 3 5 Both deoxycholic acid and ricinoleic acid increased potassium accumulation in the control animals but not in the propranolol-treated animals. Changes in potassium movement are often linked to increases in PD .3 H The observed alterations in potassium transport in these studies are most likely not related to the PD. Potassium accumulation increased during bile acid and hydroxy fatty acid perfusions which were associated with a decrease in PD (tables 1 and 2). Propranolol inhibited these changes in potassium movement but did not prevent the decrease in PD produced by deoxycholic acid or ricinoleic acid. It is likely that the increase in potassium accumulation was secondary to the changes in water and sodium movement because the changes in
potassium movement paralleled the changes in both sodium and water movement. Osmotically induced fluid accumulation is also associated with increased movement of potassium into the colonic lumen (D. S. Whiting and H. J. Binder, unpublished observations). In the amphotericin B studies mucosal permeability was altered by a drug which is known to affect epithelial permeability of the toad bladder, renal tubule, and jejunum. 21r- 2H Although the changes in permeability produced by amphotericin B are most likely dissimilar to the changes in permeability produced by detergents such as ricinoleic acid and deoxycholic acid, perfusion with amphotericin B is also associated with increases in intestinal clearance. Similar increases in intestinal clearance were observed by Rhode and Chen in studies of the canine jejunum using amphotericin B. 28 During amphotericin B perfusion a change in fluid and electrolyte movement was also observed, but this was an increase not a decrease in net water and sodium absorption. Therefore, these amphotericin B experiments permit us to suggest that changes in permeabilty, at least as shown by an increase in inulin and urea clearance, are not necessarily associated with net fluid accumulation. In conclusion, perfusion of the intestine with either ricinoleic acid or deoxycholic acid is associated with fluid and electrolyte accumulation and increased mucosal permeability. These experiments indicate that the observed increases in mucosal permeabilty are hot directly nor solely responsible for the changes in fluid and electrolyte movement associated with bile acids and hydroxy fatty acids. REFERENCES 1. Mekhjian HS , Phillips SF, Hofmann AF: Colonic secretion of
water and electrolytes induced by bile acids: perfusion studies in man. J Clin Invest 50:1569-1577 , 1971 2. KragE, Phillips SF: Active and passive bile acid absorption in man. Perfusion studies of the ileum and jejunum. J Clin Invest 53:1686-1694, 1974
3. Wingate DL, Phillips SF, Hofmann AF: Effect of glycine-conjugated bile acids with and without lecithin on water and glucose absorption in perfused human jejunum. J Clin Invest 52:12301236, 1973
4. Teem MW, Phillips SF: Perfusion of the hamster jejunum with conjugated and unconjugated bile acids: inhibition of water absorption and effects on morphology. Gastroenterology 62:261267, 1972
5. Saunders DR, Hedges JR, Sillery J, et a!: Morphological and functional effects of bile salts on rat colon. Gastroenterology 68:1236-1245, 1975
6. Dobbins JW, Binder HJ: Effect of bile salts and fatty acids on the colonic absorption of oxalate. Gastroenterology 70:1096-1100, 1976
7. Rummel W, Nell G , Wanitschke R: Action mechanisms of antiabsorptive and hydragogue drugs. In Intestinal Absorption and Malabsorption . Edited by TZ Czaky. New York, Raven Press, 1975, p 209-227 8. Kirwan WO, Smith AN, Mitchell WD, et a!: Bile acids and colonic motility in the rabbit and the human. I. The rabbit. Gut 16:894-902, 1975
9. Binder HJ, Rawlins CL: Effect of conjugated dihydroxy bile salts on electrolyte transport in rat colon. J Clin Invest 52:1460-1466, 1973
Octoberl.978
FATTY ACIDS. BILE ACIDS, AND MUCOSAL PERMEABILITY
10. Binder HJ, Filburn C, Volpe BT: Bile salt alteration of colonic
electrolyte transport: role ol' cyclic adenosine monophosphate. Gastroenterology 68:503-508, 1975 11. Conl ey DR, Coyne MJ, Bonorri s GG, et al: Bile acid stimulation of colonic adenylate cyclase and secretion in the rabbit. Am J Dig Dis 21:453-458, 1976 12. Bright-Asa re P, Binder HJ : Stimulation of colonic secretion of water and electrolytes by hydroxy fatty acids. Gastroenterology 64:81-88 , 1973 13. Ammon HV, Phillips SF: Inhibition of il eal water absorption by
intraluminal fatty acids. Influence of chain length, hydroxylation, and conjugation of fatty acids. J Clin Invest 53:205-210, 1974 14. Ammon HV , Thomas PH , Phillips SF: Effects of oleic and
15.
16.
17.
18.
19.
20.
21.
ricinoleic acid on net jejuna l water and electrolyte movement. J Clin Invest 53:374-379, 1974 Stewart JJ , Gaginella TS, Olsen WA , et a l: Inhibitory actions of laxatives on motility and water and electrolyte transport in the gastrointestinal tract. J Pharmacal Exp Ther 192:458-467 , 1975 Stewart JJ , Gaginella TS, Bass P: Actions of ricinoleic acid and structurally related fatty acids on the gastrointestinal tract. I. Effects on the smooth muscle contractility in vitro. J Pharmacal Exp Ther 195:347-354, 1975 Christensen J, Freeman BW: Circular muscle electromyogram in the rat colon: local effect of sodium ricinoleate. Gastroenterology 63:1011-1015, 1972 Cline WS, Lorenzsonn V, Benz L, et al: The effects of sodium ricinoleate on small intestinal function and structure. J Clin Invest 58:380-390, 1976 Binder HJ: Cyclic adenosine monophosphate controls bile salt and hydroxy fatty acid-induced colonic electrolyte secretion (abstr). J Clin Invest 53:7-8, 1974 Binder HJ, Dobbins JW, Whiting DS: Evidence against importance of altered mucosal permeability in ricinoleic acid induced fluid secretion (abstr). Gastroenterology 72:1029, 1977 Gaginella TS, Chadwick VS, Debongnie JC, et a!: Perfusion of rabbit colon with ricinoleic acid: dose-related mucosal injury, fluid secretion, and increased permeability. Gastroenterology
673
73:95-101 , 1977 22. Mertens RB , Mayer SE , Wheeler HO: Effect of conjugated bile
acids on cyclic AMP levels in rabbit ileal mucosa (abstr). Gastroenterology 70:919, 1976 23. Finkelstein RA: Cholera. CRC Crit Rev Microbial 2:552-623, 1973 24 . Conley DR, Coyne MJ , Chung A, et a!: Propranolol inhibits
adenylate cyclase and secretion stimulated by DCA in the rabbit colon. Gastroenterology 71:72-75, 1976 25. Feinstein JD , Coyne MJ , Bonorris GG, et al: Bile acid and fatty acid effect on intestinal cyclic-AMP, Na+-K +-ATPase and secretion (abstr). Gastroenterology 72:1183, 1977 26. Lichtenstein NS, Leaf A: Effect of amphotericin B on the permeability of the toad bladder. J Clin Invest 44:1328-1342 , 1965 27. Stroup RF , Weinman E, Hayslett JP, et a!: Effect of luminal
28. 29. 30.
31. 32. 33. 34. 35.
36.
permeability on net transport across the amphibian proximal tubule. Am J Physiol 226:1110-1116, 1974 Rohde JE, Chen LC: Permeabilty and selectivity of canine and human jejunum during cholera. Gut 13:191-196, 1972 Snedecor GW, Cochran WG: Statistical Methods. Sixth edition. Ames, Iowa State University Press, 1967 Loehry CA, Kingham J , Baker J : Small intestinal permeability in a nimals and man. Gut 14:683-688, 1973 Binder HJ: Fecal fatty acids- mediators of diarrhea? Gastroent erology 65:847-850, 1973 Binder HJ, Donowitz M: A new look a t laxative action. Gastroenterology 69:1001-1005, 1975 Binder HJ: Pharmacology of laxatives. Annu Rev Pharmacal Toxicol 17:355-367 , 1977 Racusen L, Binder HJ: Catecholamines alter colonic ion transport (abstr) . Gastroenterology 72:1115, 1977 Coyne MJ, Bonorris GG, Chung A, et al: Inhibition by propranolol of bile acid stimulation of rabbit colonic adenylate cyclase in vitro. Gastroenterology 71:68-71 , 1976 Fisher KA , Binder HJ, Hayslett JP: Potassium secretion by colonic mucosal cells after potassium adaptation. Am J Physiol 231:987-994, 1976