Octreotide promotes gallbladder absorption in prairie dogs: A potential cause of gallstones

Octreotide promotes gallbladder absorption in prairie dogs: A potential cause of gallstones

GASTROENTEROLOGY1995;108:1547-1555 Octreotide Promotes Gallbladder Absorption in Prairie Dogs: A Potential Cause of Gallstones A. JAMES MOSER,* MOHAM...

863KB Sizes 7 Downloads 49 Views

GASTROENTEROLOGY1995;108:1547-1555

Octreotide Promotes Gallbladder Absorption in Prairie Dogs: A Potential Cause of Gallstones A. JAMES MOSER,* MOHAMMAD Z. ABEDIN, t DAN I. N. GIURGIU, t and JOEL J. ROSLYN* *Department of Surgery, UCLASchool of Medicine, Los Angeles, and Research Services, Sepulveda Veterans Affairs Medical Center, Sepulveda, California; and *Department of Surgery, Medical College of Pennsylvania and Hahnemann University, and Philadelphia Veterans Affairs Medical Center, Philadelphia, Pennsylvania

Background/Aims: Gallstone formation during octreotide administration has been causally linked to increased biliary concentrations of calcium, protein, and total lipids, all purported prolithogenic factors. These changes may be caused by octreotide-induced gallbladder stasis or a direct effect of octreotide on gallbladder absorption. We tested the hypothesis that octreotide stimulates gallbladder ion and water transport. Methods: Prairie dog gallbladders were mounted in Ussing chambers and bathed in oxygenated Ringer's solution. Electrophysiological parameters were recorded, and unidirectional Na ÷, CI-, and H20 fluxes were measured before and after serosal exposure to 50 nmol/L octreotide. Results: Octreotide exposure caused a significant decrease in transepithelial short-circuit current and potential difference and an increase in tissue resistance compared with baseline. These alterations in electrophysiological parameters coincided with changes in ion transport. Octreotide stimulated net Na ÷ and H20 absorption and converted the gallbladder from a state of CI secretion to one of CI- absorption by increasing mucosal to serosal fluxes. Octreotide effects on ion transport were blocked by 4,4'-diisothiocynostilbene2,2'-disulfonic acid and amiloride and reversed by theophylline. Conclusions: Octreotide may promote gallstone formation by inducing gallbladder stasis and by directly increasing gallbladder absorption, which may act synergistically to increase the concentration of prolithogenic factors in bile and to facilitate nucleation and stone growth.

ctreotide, a long-acting analogue of somatostatin, has been used in various clinical settings, including the management of patients with biologically active endocrine tumors, such as carcinoids and pituitary adenomas, 1 and patients with gastric and pancreatic fistulas. 2 The therapeutic potential of octreotide is a result of a hormone-mediated blockade of secretin, gastrin, and vasoactive intestinal peptide release, similar to the effect of native somatostatin. 3'4 Nonetheless, long-term therapy of octreotide has been complicated by gallstone forma-

O

tion. 5-7 Ultrasound studies of acromegalic patients receiving octreotide indicate that cholelithiasis develops within 1 year in 2 2 % - 6 7 % of patients with previously normal gallbladders. 8 Similar findings have been reported in patients with somatostatinoma syndrome that have a 40% incidence of gallstones. 9 Although there are several studies examining the effects of octreotide on the biliary tree, *°-12 the mechanisms responsible for octreotide-induced gallstone formation are not clear. Using the prairie dog gallstone model, Ahrendt et alJ 2 have shown recently that short-term administration of octreotide increases biliary concentrations of total calcium, protein, bilirubin, and lipids, all purported nucleating agents. These changes may be caused by octreotide-induced gallbladder stasis, 1I causing bile to be concentrated abnormally by prolonged mucosal exposure. An additional explanation is that these findings result from a direct effect of octreotide on gallbladder epithelial ion transport, as suggested by the findings that native somatostatin affects ion transport in small and large intestine. In addition to down-regulating gastrointestinal hormone release, somatostatin has been shown to increase ion and water absorption in the rabbit intestine and rat colon by a direct effect on these epithelia. 13-~5 This proabsorptive stimulus occurred in vitro, suggesting a primary role for somatostatin in intestinal ion transport regulation. The role of octreotide in regulating gallbladder ion transport is not clear. The objective of the present study was to test the hypothesis that octreotide stimulates gallbladder ion and water absorption that, therefore, may increase the concentrations of nucleating factors present in gallbladder bile. We would propose that increased mucosal ion transport, in concert with gallbladder stasis, acts synergistically to promote nucleation and stone Abbreviations used in this paper: DIDS, 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid; Isc, short-circuit current; Rt, transepithelial resistance; Vt, transepithelialvoltage. © 1995 by the AmericanGastroenterologicalAssociation 0016-5085/95/$3.00

1548 MOSER ET AL.

growth. The prairie dog, a widely used model for gallstone research, ~6 was selected for its similarity to h u m a n physiology and for its unique transport properties that permit the pharmacological investigation of gallbladder ion transport. ~v

M a t e r i a l s and M e t h o d s Tissues and Solutions Gallbladder tissues were prepared according to the procedures described previously, iv Adult male prairie dogs, Cynorays ludovicianus, were obtained from Otto Marten Locke (New Braunfels, TX) and housed in thermoregulated cages. All animals were fed standard laboratory chow (Purina Laboratory Chow; Ralston-Purina, St. Louis, MO). After a 16-hour fast with water provided ad libitum, animals were anesthetized intramuscularly with ketamine (100 mg/kg body wt) and xylazine (1.5 mg/kg), and careful cholecystectomy was performed. Gallbladders were opened longitudinally, rinsed free of bile with Ringer's solution, and then mounted in a twopiece lucite Ussing chamber with a 0.67-cm 2 circular aperture similar to that described by Schultz and Zalusky. is Chamber contact and seal were maintained with high viscosity silicone grease. Mucosal and serosal surfaces were perfused with equal volumes (10 mL) of modified amphibian Ringer's solution (adjusted to pH 7.4) of the following composition (in mmol/ L): Na +, 140; CI-, 124; HCO3- , 21; K +, 5.4; Ca2+, 1.3; Mg 2÷, 1.2; HPO42-, 2.4; H2PO4-, 0.6; HEPES, 5; and glucose, 10. The chamber was maintained at 37°C and gassed with 95% 02 and 5% CO2. Bicarbonate-free solutions were prepared by replacement of HCO3- with equimolar amounts of Na2SO4 adjusted to pH 7.4 and gassed with 100% 02. Octreotide solutions were made by dissolving crystalline octreotide (gift from Sandoz Pharmaceuticals, East Hanover, NJ) in Ringer's soiution. Theophylline, amiloride, and 4,4'-diisothiocynostilbene-2,2'-disulfonic acid (DIDS) were obtained from Sigma Chemical Co. (St. Louis, MO). Radionuclides 22Na, 36C1, and 3H20 were bought from Dupont New England Nuclear (Boston, MA).

Electrophysiological M e a s u r e m e n t s Open-circuit transepithelial voltage (Vt) and short-circuit current (Isc) were measured after standard four-electrode techniques using an automatic voltage current clamp (VCC 600; Physiologic Instruments, San Diego, CA) as described previously, iv Transepithelial resistance (R~) was determined by recording the potential deflections in response to a bipolar current of + 1 0 ~A passed across the epithelium by a pulse generator via Ag-AgC1 electrodes. Compensations for fluid resistance and calomel electrode asymmetry were made before mounting tissues in the chamber. Direct current was passed continually through the tissue by means of Ag-AgC1 electrodes to nullify the spontaneous V~. Tissue Rt and I~c measurements were corrected subsequently for chamber surface area.

GASTROENTEROLOGYVol. 108, No. 5

Unidirectional Fluxes Unidirectional transepithelial fluxes of Na +, C1-, and H20 were determined by the dual isotope technique with 22Na, 36C1, and 3H20 according to the procedures described in detail previously) v Briefly, 1.5 btCi of each of a pair of isotopes (22Na and 3H20 or 36C1 and 3H20) were added to the mucosal reservoir, and unidirectional mucosa to serosa fluxes of Na + (JmN~) and H20 (J~ 2°) or of C1- (JCm~) and H20 (JmH~°) were measured from three 10-minute flux periods. Both surfaces of the tissue were then washed simultaneously with 600 mL of warm modified amphibian Ringer's solution to remove the isotope from the Ussing chamber, and unidirectional serosa to mucosa fluxes of Na + (J~d) and H20 (J~m2°) or C1- (jc~) and H20 (JsH20) were measured in a manner similar to that of mucosa to serosa. Background counts were determined in 1 mL aliquots obtained from the serosal and mucosal chambers before the addition of radioisotopes. The sequence of radioisotopes and the direction of fluxes were randomized for individual experiments. All flux measurements were performed under short-circuit conditions except for 1 0 15-second periods required to record open-circuit Vt. Radioisotopes were assayed in 10 mL of biodegradable scintillation cocktail (Research Products International Corp., Mount Prospect, IL) using a well-type liquid scintillation counter (LS8000; Beckman Instruments, Fullerton, CA). Unidirectional Na +, C1 , and H20 fluxes were calculated by a previously reported standard formula. ~s Positive net flux represents ion absorption, whereas negative net flux represents secretion.

Experimental Design After stabilization of Isc, Vt, and Rt, gallbladder tissues were exposed to octreotide in concentrations ranging from 1 - 100 nmol/L to determine dose-response plots by the method of Hanes-Woolf. 19 Subsequently, electrophysiologicai parameters and transepithelial unidirectional fluxes of Na +, C1- and H20 were measured before and after serosal exposure to 50 nmol/L octreotide. To define the mechanisms by which octreotide mediates its effects on ion transport, tissues were pretreated with ion transport inhibitors (amiloride and DIDS) or bathed in HCO 3 -free Ringer's solution, then exposed to octreotide, and flux measurement continued to be recorded. Additional tissues were exposed to 50 nmol/L octreotide followed by the addition of 10 -2 mol/L theophylline, and electrical parameters were recorded to determine whether octreotide effects are 5'-cyclic adenosine monophosphate (cAMP) dependent. Tissues were washed thoroughly between serosal octreotide exposures.

Tissue Viability Tissue viability in the present experiment was assessed according to the criteria established earlier for this model. 2° These included: (1) the ability of the tissue to maintain active transport, (2) the recovery of tissue Isc and Vt after drug expo-

May 1 9 9 5

OCTREOTIDE PROMOTES GALLBLADDER ABSORPTION

Table 1. Effect of Serosal Octreotide on Gallbladder Electrophysiology

Isc (#Eq. cm -2. h -I) Vt (mV) Rt (.Q- cm2)

Baseline (n = 11)

50 n m o l / L octreotide (n = 11)

Washout (n = 11)

7.9 + 0.5 13.1 -+- 0.7 57 _+ 10

1.9 _+ 0.5 a 6.3 + 0.7 a 77 + 10 c

4.7 + 0.4 a'° 7.8 + 0,7 a 43 +_ 10 a

NOTE. Results are expressed as means _+ SEM. ap < 0 . 0 0 1 vs. corresponding baseline. ~P < 0 . 0 0 1 vs. octreotide (ANOVA). cp < 0 . 0 0 3 vs. baseline (paired t test). ap < 0.05 vs. octreotide (ANOVA).

sure with subsequent mucosal and serosal washing with Ringer's solution, and (3) the stimulation of Isc and V~ after bilateral

exposure to 10 -2 mol/L theophylline before termination of experiments. It is known from previous experiments that 1 0 - 2 mol/L theophylline elicits a maximal stimulus to ion transport in this tissue. 2°

Statistics All data are presented as means + SEM. Comparisons between groups were performed by ANOVA. A two-tailed Student's t test for the paired variables was used when applicable.

concentration of 35 nmol/L for initial exposure and 50 nmol/L after a second exposure.

Octreotide Effects on Ion and H20 Transport The effects of 50 nmol/L serosal octreotide on ion and water transport are summarized in Table 2. Under basal conditions, the gallbladders showed a net Na + absorption and a net C1- secretion, similar to the values reported previously. 17 However, serosal octreotide doubled net Na + absorption and converted CI- from a state of secretion to absorption. Octreotide-induced changes in ion transport were caused by a significant stimulation of mucosa to serosa fluxes of Na + and C1- without attendant changes in serosal to mucosal transport. Time course effects of octreotide on unidirectional fluxes show that these changes in epithelial Na ÷ transport occurred rapidly after treatment and doubled mucosal to serosal Na + flux within 20 minutes (Figure 3). The effect of octreotide was particularly marked in cases of C1- transport. As shown in Figure 4, octreotide immediately converted the gallbladder from a state of low baseline net C1- secretion to one of significant C1- absorption, and the peak was reached in 10 minutes. Serosal octreotide had a rapid and selective effect on transepithelial CI- absorption independent of serosal to mucosal transport, as observed for Na

+ .

Results

1o

Electrophysiological Effects of 0ctreotide Serosal exposure of gallbladder tissue to 50 nmol/L octreotide caused a significant decrease in transepithelial Isc and Vt, but an increase in Rt compared with baseline values. A wide variance in tissue resistance was observed within a group. However, in each case, tissue resistance was increased after exposure to octreotide, and the difference in tissue resistance between the control and octreotide-treated tissues was statistically significant (P < 0.05). These effects were partially reversible after washing with Ringer's solution (Table 1). Time course effects of octreotide show that gallbladder Isc decreased within seconds of serosal exposure, and the nadir was reached within 10 minutes. The effects of octreotide were long acting, and a plateau was maintained for 40 minutes with minor spontaneous recovery in the first 20 minutes (Figure 1). Dose-dependent effects of serosal octreotide were determined over the concentration range of 1 - 1 0 0 nmol/ L and were plotted by the method of Hanes-Woolf 19 (Figure 2). The threshold for octreotide-induced inhibition of Isc was 1 nmol/L, with complete inhibition of Isc at 100 nmol/L. Extrapolation from the dose-response curve of octreotide shows a half-maximal inhibitory

1549

l

9

8 ~

7

6 ~. "~ 5 r.va ~ 4 o 3 {¢1

2 1 N=I 1

0

1

0

*p
I

I

I

I

I

I

I

I

10 20 30 40 50 60 70 80 90 Time (rain)

Figure 1. Effect of 50 n m o l / L serosal octreotide on gallbladder Isc. After the determination of baseline ion flux during the first 40 minutes, octreotide was added (arrow) and fluxes determined for another 50 minutes. Peak effects were observed at 10 minutes with subsequent minor spontaneous recovery. Results are expressed as mean _+ SEM. N, number of animals. * P < 0 . 0 0 2 vs. baseline (ANOVA).

1550 MOSER ET AL.

GASTROENTEROLOGYVo1.108, No. 5

1 O0

0 0 r-

cosa to serosa Na + and H 2 0 fluxes (Table 5), suggesting that the octreotide-induced observed stimulation of Na + absorption may be mediated through Na+/H + exchange.

80

Octreotide Effect on Theophylline-lnduced Stimulation of Ion Transport

60

.2

m

rE:

40 iCso=,:35n M 20

A (5) i

-9.0

/~(=) !

-8.5

"2 nd exposure

i

i

i

--8o0

--7,5

--7,0

--6,~

log [Ocfreofide] Figure2. Dose-response relationship of serosal octreotide concentration and gallbladder Isc inhibition by the method of Hanes-Woolf. 19 Between 3 and 11 tissues were studied at each concentration. Halfmaximal inhibitory concentration effects (ICso) were observed at 35 nmol/L for initial octreotide exposure. *Inhibitory effect of 50 nmol/ L serosal octreotide on Isc after secondary exposure. Under basal conditions, the gallbladder absorbs H20, as shown in Table 2. After octreotide treatment, H 2 0 transport was increased in the mucosal to serosal direction, similar to Na + and C1- absorption, with diminished serosal to mucosal flux. The overall effect of octreotide on water was a significant increase in net absorption.

Effects of DIDS and HCOa- Substitution on Octreotide-lnduced Stimulation of Mucosa to Serosa CI- and H20 Fluxes Pretreatment of tissues with DIDS (C1 /HCO3- inhibitor) completely blocked the octreotide effect on C1- flux (Table 3), suggesting that octreotide-induced stimulation of mucosa to serosa C1- flux is mediated through Cl-/HCO3 exchange. However, nominal substitution of HCO 3 in the bathing solution diminished the octreotide-induced stimulation of mucosa to serosa C1- flux but did not completely block the octreotide effect (Table 4). Similarly, octreotide did not have any effect on mucosa to serosa H 2 0 flux in the tissues pretreated with either DIDS or HCO3- substitution (Tables 3 and 4). Amiloride Effects on Octreotide-lnduced Stimulation of M u c o s a to Serosa Na ÷ and H20 Fluxes Pretreatment of tissues with Na+/H + exchange inhibitor amiloride blocked the octreotide effect on mu-

To determine whether octreotide mediates its effects through cAMP-dependent ion transport mechanisms, octreotide-treated tissues were exposed to 10 -2 mol/L theophylline on both mucosa and serosa, and electrical parameters were measured. As shown in Table 6, the addition of theophylline reversed the octreotide-induced decrease in Isc and Ve and the increase in Re, suggesting that the octreotide effect may be mediated through down-regulation of cAMP-dependent transport processes.

Development of Tachyphylaxis to Octreotide Tachyphylaxis has been reported in earlier studies of somatostatin in rabbit intestine. 1~ To determine whether tachyphylaxis was a factor on repeated gallbladder exposure to 50 nmol/L octreotide, electrophysiological parameters were recorded after sequential serosal additions of octreotide. Representative tissues were also exposed to increasing concentrations of octreotide (50 nmol/L, 500 nmol/L, and 5 btmol/L) to show saturation kinetics of octreotide effect on ion transport. A comparison of the effects on gallbladder electrical properties after serial exposures to 50 nmol/L octreotide is shown in Table 7. As for the initial addition, a second addition of octreotide significantly inhibited Isc and Vt while increasing Re. However, the onset of octreotide's action was significantly delayed after the second exposure. Although electrical effects of octreotide were still large after exposure 2, the magnitude of the change in Isc, V~, and Re declined and almost reached statistical significance when compared with the first exposure, suggesting tachyphylaxis. These results are confirmed by time course plots of the serosal effects of varying concentrations of octreotide that depict the development of tachyphylaxis on repeated exposures to octreotide (Figure 5). Saturation of the octreotide effect despite increasing concentrations of the peptide is obvious.

Viability of Octreotide-Treated Tissues In all cases, the present study conformed to the criteria of tissue viability set forth in Materials and Methods. First, gallbladder epithelia maintained active ion

May 1995

OCTREOTIDE PROMOTES GALLBLADDER ABSORPTION

1551

Table 2. S e r o s a l Effect o f O c t r e o t i d e on G a l l b l a d d e r Ion and Wat er Fluxes Na +

Jms J~m

Jnet

H20

CI-

Baseline (n = 7)

50 nmol/L octreotide (n = 7)

Baseline (n = 5)

50 nmol/L octreotide (n = 5)

Baseline (n = 10)

50 nmoi/L octreotide (n = 10)

17.0 + 1.3 11.6 ± 0.9 -t-5.4 ± 1.5

22.1 --+ 1.5 ~ 11.7 ± 1.1 + 1 0 . 4 + 1.7 c

10.5 ± 1.4 12.9 ± 1.5 --2.3 ± 1.1

20.1 ± 1.8 b 12.1 ± 1.7 + 8 . 7 ± 1.3 ~

682 + 53 662 ± 44 + 2 0 ± 24

715 ± 41 626 ± 34 + 8 7 + 19 °

NOTE. Results are expressed as means ± SEM. Flux expressed in microequivalents per centimeter squared per hour of ions or microliters per centimeter squared per hour of H20. A positive (+) Jn~t indicates ion or H20 absorption, and a negative ( - ) sign indicates secretion. Jms, mucosa to serosa flux; Jnet, Jms - Jsm; Jsm, serosa to mucosa flux; n, number of tissue specimens. ~P < 0.03 vs. corresponding baseline. bp < 0.002 vs. corresponding baseline. cp < 0.03 vs. corresponding baseline. op < 0.0002 vs. corresponding baseline.

transport throughout the study as shown by steady shortcircuit current. Furthermore, the electrophysiological effects of octreotide were partially reversible after washing with Ringer's solution (Table 1). Finally, ion transport mechanisms seemed viable at the conclusion of all experiments as shown by appropriate stimulatory responses of tissues to 10 -2 mol/L theophylline exposures.

~--

These results indicate that octreotide has a direct proabsorptive effect on the gallbladder epithelium. Serosal exposure of the prairie dog gallbladder to 50 nmol/ L octreotide caused a stimulation of net Na + and C1 absorption, specifically by increasing mucosal to serosal fluxes. Overall, octreotide doubled net Na + absorption and converted the gallbladder from a state of CI- secre-

4.5 5 0 n M Octreo ~ T /i\

4.0

o

Discussion

~-.

*

$

6

o .,=,1

s

3.5 0

~ ~

5.0 2.5

1

w

T

04~

4

~

5

50nM Octreo



T

2.0 1.5 x

+

LJ_

2

x

0.5

0

z

~

1.0

0.0

N=7 I

10

*p
20

vs baseline

I

I

I

I

I

I

50

40

50

60

70

80

Time

(rain)

Figure 3. Effect of 50 nmol/L serosal octreotide (Octreo) on lO-minute unidirectional Na + flux. The rate of either mucosal to serosal Na + flux (V) or serosal to mucosal Na + flux (A) in a given lO-minute period. Arrow indicates the addition of an octreotide. By 20 minutes, octreotide increased mucosal to serosal transport significantly without attendant changes in serosal to mucosal transport. Increased net Na + absorption is apparent over baseline levels. Results are expressed as mean ± SEM. N, number of tissues. *P < 0.01 vs. baseline

(ANOVA).

T,

1

~

0

I

N=5

*p
vs baseline

I

I

I

I

I

I

I

I

10

20

30

40

50

60

70

80

Time ( m i n ) Figure 4. Effect of 50 nmol/L serosal octreotide on lO-minute unidirectional CI- flux. Results are expressed as mean ± SEM. At basal condition, low baseline net CI secretion is evident (Jsm > Jms), After octreotide exposure (arrow), mucosal to serosal Cl flux (V) is selectiveJy increased with the greatest effect observed in the first 10 minutes. Serosal to mucosal CJ flux (A) is unchanged. N, number of animals. *P < 0.03 vs. baseline JC~ols (ANOVA).

1552

MOSER ET AL.

GASTROENTEROLOGY Vol, 108, No. 5

Table 3. Effect o f DIDS on O c t r e o t i d e S t i m u l a t i o n o f M u c o s a t o S e r o s a CI

Cl- (#Eq. cm -2. h -1) H20 (!11" cm - 2 . h -1)

Table 5, Effect o f Amiloride on O c t r e o t i d e S t i m u l a t i o n o f M u c o s a t o S e r o s a Na + and H20 Fluxes

and H20 Fluxes

Baseline

1 mmol/L DIDS

1 mmol/L DIDS + octreotide

12.8 + 1.2 728 + 65

10.1 + 1.3" 652 _+ 52 b

10.7 4- 1.2 634 _+ 57

NOTE. Results are expressed as mean _+ SEM. Five tissue specimens were in each group. ap < 0.03 vs. corresponding baseline (paired t test). oP < 0.04 vs. corresponding baseline (paired t test).

Na ÷ (#Eq. cm -2. h 1) H=O (/~L. cm 2. h-l)

Baseline

1 mmol/L amiloride

1 mmol/L amiloride + octreotide

16.4 4- 1.1 730 4- 78

12.6 + 1.0 a 667 + 69 °

14.3 _+ 1.7 613 + 66 c'~

NOTE. Results are expressed as mean +_ SEM. Four tissue specimens were in each group. aP < 0.005 vs. corresponding baseline (paired t test). bp < 0.02 VS. corresponding baseline (paired t test). cp < 0.04 vs. corresponding baseline (paired t test). oP < 0.02 vs. amiloride (paired t test).

tion to one of significant C1- absorption. Water absorption was increased in parallel with ion transport, as expected, because water movement is coupled to salt movement. Octreotide's effects on Na ÷ and C1- fluxes may be mediated through stimulation of Na+/H + and CI-/HCO3- exchangers and down-regulation of cAMPdependent transport processes. These findings have relevance to the etiology of gallstones associated with long-term administration of octreotide because previous studies have shown that gallbladder absorption is increased during early gallstone formation. 21'22The role of gallbladder absorption in gallstone pathogenesis is further evidenced by the studies of Strichartz et al. 23 who have shown that the administration of amiloride, an inhibitor of ion transport, significantly reduces the incidence of gallstones in the animals fed a cholesterol diet. It is becoming increasingly evident that gallbladder absorption plays an important role in the cascade of events, including hepatic secretion of cholesterol-saturated bile, 24 gallbladder hypomotility, 25 nucleation defect, 26'27 and the presence of prolithogenic and pronucleating factors in the bile, 28-3° that ultimately lead to gallstone formation. Octreotide-associated gallstone formation may result from the multiple effects of this agent on biliary systems, including alterations in bile composition, gallbladder absorption, and gallbladder motility. In the prairie dog

gallstone model, Ahrendt et al. I2 have shown that octreotide increases biliary concentrations of calcium, bilirubin, protein, and total lipids, all prolithogenic factors. These changes in gallbladder bile composition occur parallel with octreotide-induced gallbladder stasis, which permits the prolonged presence of lithogenic bile in the gallbladder and may promote nucleation of cholesterol crystals. Previous speculation about the mechanism for increased concentrations of biliary solutes has focused on gallbladder stasis and prolonged mucosal exposure as an etiology. The present data are the first demonstration that octreotide has an independent, direct proabsorptive effect on gallbladder ion transport that may account for the observed changes in gallbladder bile composition in addition to its effects on gallbladder motility. This observation corroborates earlier findings in a long-term canine bile fistula model, in which somatostatin infusion decreased bile flow by enhancing the ductular reabsorption of water. 3~ This is further evidence of the proabsorprive role of somatostatin throughout the biliary tract. The dose-response data in the present study show that the gallbladder epithelium is more sensitive to octreotide exposures compared with other epithelia, suggesting a possible explanation for the prominent side effects on biliary systems observed during octreotide therapy. Previous studies show a less sensitive dose-dependent effect

Table 4. Effect o f H C 0 3 - S u b s t i t u t i o n on O c t r e o t i d e

Table 6, Effect o f T h e o p h y l l i n e on G a l l b l a d d e r Ion T r a n s p o r t

E x p o s u r e on M u c o s a to S e r o s a CI- and H20

After Addition o f O c t r e o t i d e

Fluxes

HC03- free + CI- (#Eq. cm -2. h I) H20 (#L. cm -2° h -~)

HC03- free

octreotide

9.9 4- 1.1 546 + 132

12.8 4- 1,8 a 557 4- 123

NOTE. Results are expressed as mean + SEM. Five tissue specimens were in each group. ap < 0.008 vs. HC03 free (paired t test).

Isc (#Eq. cm -2o h 1) Vt (mY) Rt (-(,2° c m 2)

Octreotide

Theophylline

2.1 4-_ 0.6 4.2 + 0.9 58 + 6

7.0 ± 0.6 ~ 12.1 ± 0.9 a 42 +_ 6 b

NOTE. Results are expressed as mean _+ SEM. Nine tissue specimens were in each group. ap < 0.001 vs. octreotide (ANOVA). ~P < 0.05 vs. octreotide (paired t test).

May 1995

OCTREOTIDE PROMOTES GALLBLADDER ABSORPTION 1553

of octreotide in the intestine, with concentrations of 1 10 ~mol/L required for a proabsorptive effect) 2 In the rabbit ileum, the action of somatostatin was characterized by a rapid spontaneous recovery except at 10 Bmol/L dose. 13 This pattern of octreotide action contrasts with the long-acting effect of octreotide on the prairie dog gallbladder, in which a 50-nmol/L dose caused a prolonged decrease in Isc with minimal recovery over time. Moreover, the effects on ion fluxes were observed at octreotide concentrations that were 2 0 - 2 0 0 times lower than the concentrations that caused similar effects in the intestine. Together these phenomena suggest that the gallbladder epithelium is particularly sensitive to octreotide and may account for the high rate of biliary complications reported during octreotide therapy. The cellular mechanism for the effects of octreotide on gallbladder epithelia is suggested by the change in mucosal to serosal ion flux and the increase in tissue resistance after the peptide exposure. Under basal conditions, the prairie dog gallbladder shows electrogenic ion transport, which is largely composed of uncoupled Na + absorption and CI secretion and is likely mediated through Na+/H + and CI-/HCO3- exchangers, iv In the present study, we have reproduced the earlier baseline data, and the Isc (7.9 ~tEq" cm -2" h-i; Table 1)generated by this tissue is accounted for by net Na + absorption ( + 5 . 4 b t E q ' c m - 2 " h -1) and net C1- secretion ( - 2 . 3 ~ E q " c m - 2 " h 1). After octreotide exposure, the Isc decreases to a value of 1.9 btEq" c m - 2 " h -1 that may be accounted for by net absorption of both Na ÷ (10.4 ~Eq" cm -2" h -1) and C1- (8.7 ~Eq" c m - 2 " h-i). An octreotide-induced increase in net Na + absorption and a reversal of net CI- secretion to net CI- absorption is caused by an increase in mucosa to serosa fluxes of Na ÷ and C1- without any change in serosa to mucosa fluxes.

Table 7. Electrophysiological Effect of Sequential Serosal Exposure of Octreotide Exposure 1 (n = 11) A of Isc (%) of Vt (%) z$ of Rt (%) T,a~r,(min)

--71 -61 +36 6

+ 6 _+ 9 + 5 ± 3

Exposure 2 (n = 10) --54 -37 +29 21

4- 7 _+ 9 + 6 _+ 3 ~

NOTE. Results are expressed as means + SEM. Comparison of the percent change (Z&) with respect to baseline in Isc, Vt, and Rt after initial (exposure 1) and subsequent (exposure 2) additions of 50 nmol/L serosal octreotide. A negative A indicates a decrease, and a positive A indicates an increase. n, number of tissue specimens; Tnadir, the time after drug addition at which Isc reached a minimum. ap < 0.005 vs. exposure 1 (ANOVA).

50nM Ocfreo lO

tM

i

500nM Octreo

9

5~M Oetreo

8 0 ?

:L

V

o Q

6

5

I

I

I

I

!

I

I

I

o

1o

20

ao

40

50

60

vo

Time ( m i n ) Figure 5. Representative tissue showing development of tachyphylaxis caused by repeated exposure to octreotide. Exposure to increasing doses of octreotide is indicated by the arrows. Saturation of the octreotide effect is apparent from the figure.

Because serosa to mucosa fluxes are virtually identical in both control and octreotide-treated tissues, paracellular pathways seem to be unaffected by octreotide. Furthermore, octreotide changes the relationship of Na ÷ to C1flux from 2:1 (basal, 17.0 vs. 10.5) to 1:1 (octreotide, 22.1 vs. 20.1; Table 2). W e propose that octreotide mediates its effects by stimulating Na+/H + and CI-/HCO 3parallel exchangers with a separate action on apical C1conductance, thereby increasing Na ÷ and CI- absorption and raising Rt. This is supported by the data from ion transport inhibitors (DIDS and amiloride) and HCO3 substitution experiments of octreotide effects on C1- and Na ÷ fluxes (Tables 3 - 5 ) . Pretreatment of tissues with DIDS or HCO3- substitution blocks octreotide-stimulated C1- flux. Similarly, amiloride treatment inhibits octreotide-induced stimulation of Na + flux. It is of interest to note that about 25% of C1- absorption is sensitive to DIDS or HCO3- substitution. Similarly, about 25% of Na + absorption is sensitive to amiloride. An alternative explanation for the octreotide-induced changes in Na + flux may include the reversal of apical C1-/HCO3 exchange to HCO3-/C1- exchange leading to a decrease in intracellular p H that will stimulate Na+/H + exchange. However, we do not have any evidence that suggests that reversal of C1-/HCO3 to HCO3-/C1 will lead to a decrease in intracellular p H that would increase Na +

1554

MOSER ET AL.

absorption by stimulating Na+/H + exchange. In addition, we could not find any significant changes in mucosal p H because the bathing solution was buffered with HEPES and 21 mmol/L HCO3- and constantly gassed with 95% 02 and 5% CO2. The finding that theophylline reverses the octreotideinduced decrease in Isc is very intriguing. In healthy prairie dog gallbladders, the addition of theophylline, which inhibits the enzymatic conversion of cyclic adenosine monophosphate to 5'-adenosine monophosphate, causes an immediate large increase in Isc. This increase in Isc may be, in part, caused by an increase in apical membrane C1- conductivity with an increase in C1- secretion as reported in Necturus gallbladder. 33 In contrast, octreotide causes a decrease in Isc and a reversal of net C1- secretion to net C1- absorption. One possible explanation for this finding is that the octreotide effect is not mediated through cAMP-dependent processes. If octreotide would have worked through cAMP, theophylline would not have the effect on octreotide-treated tissues that has been observed in the present study. Alternatively, octreotide may down-regulate the effects of cAMP that is overcome by theophylline. Several studies have shown that somatostatin lowers intracellular cAMP levels in isolated cell preparations. 34'35 Dharrnsathaphorn et al. have reported that somatostatin affects electrolyte and water movement by blocking the action of cAMP in the rat jejunum in which somatostatin did not affect intracellular cAMP. 36 W e have not measured the intracellular levels of cAMP in octreotide-exposed tissues. However, theophylline data suggest that an octreotideinduced decrease in Isc and a reversal of C1- secretion to C1- absorption may be mediated through the downregulation of cAMP-dependent transport processes. There are several lines of evidence to suggest that these in vitro effects of octreotide have physiological significance in vivo. This study represents the first demonstration that octreotide has a significant effect on ion transport at such low doses. Moreover, the flux effects were observed at doses close to the half-maximal inhibitory concentration in this tissue and were almost completely reversed by washing, indicating that the dose used did not have a toxic effect. Furthermore, Ho et al. 6 reported human serum levels of octreotide ranging from 6.6 to 22 nmol/L after therapeutic subcutaneous doses of 2 5 0 500 btg. Serosal 50 nmol/L octreotide used in the present study, therefore, is a concentration that correlates well with in vivo therapeutic serum levels. The serosal action of the peptide is also physiological because of the circulation of somatostatin in serosal vessels in vivo. However, we should be cautious to extrapolate these data on acute

GASTROENTEROLOGYVol. 108, No. 5

administration of octreotide in vitro to long-term administration in vivo, especially because the experiments show tachyphylaxis to repetitive doses. Further studies are needed to evaluate the absorptive function of the gallbladder in vitro and in vivo during long-term administration of octreotide. In conclusion, octreotide exerts a direct proabsorptive effect on gallbladder mucosal ion transport in addition to its effect of causing gallbladder stasis. The combination of increased gallbladder absorption and stasis may act synergistically to increase the concentration of prolith0genic factors in bile and facilitate nucleation and ultimately stone growth.

References 1. Comi RJ. Pharmacology and use in pituitary tumors (pp 3 6 - 4 1 ) . In: Gorden P, moderator. Somatostatin and somatostatin analogue (SMS 201-995) in treatment of hormone secreting tumors of the pituitary and gastrointestinal tract and non-neoplastic disease of the gut. Ann Intern Med 1 9 8 9 ; 1 1 0 : 3 5 - 5 0 . 2. O'Donnell LJD, Farthing MGJ. Therapeutic potential of long acting somatostatin analogue in gastrointestinal diseases. Gut 1989; 3 0 : 1 1 6 5 - 1 1 7 2 . 3. Krulich L, Dhariwal APS, McCann SM. Stimulatory and inhibitory effects of purified hypothalamic extracts on growth hormone release from rat pituitary in vitro. Endocrinology 1 9 6 8 ; 8 3 : 7 8 3 790. 4. Dollinger HC, Raptis S, Pfeiffer EF. Effects of somatostatin on exocrine and endocrine pancreatic function stimulated by intestinal hormones in man. Horm Metab Res 1 9 7 6 ; 8 : 7 4 - 7 8 . 5. Buscail L, Tauber JP, Escorrou J, Harris AG, Bousquet-Pue] C, Delvaux M, Bayard F, Ribet A. Gallstone formation and occurence of cholesterol monohydrate crystals in gallbladder bile of patients with long-term sandostatin treatment (abstr). Gastroenterology 1989; 96:A580. 6. Ho KY, Weissberger AJ, Marbach P, Lazarus L. Therapeutic efficacy of the somatostatin analog SMS 201-995 (octreotide) in acromegaly. Ann Int Med 1 9 9 0 ; 1 1 2 : 1 7 3 - 1 8 1 , 7. PI6ckinger U, Dienemann D, Quabbe HJ. Gastrointestinal sideeffects of octreotide during long-term treatment of acromegaly. J Clin Endocrinol Metab 1 9 9 0 ; 7 1 : 1 6 5 8 - 1 6 6 2 . 8. Marteau P, Chretien Y, Calmus J, Pare R, Poupon R. Pharmacological effect of somatostatin on bile secretion in man. Digestion 1989; 4 2 : 1 6 - 21. 9. Marcial MA, Pinkus GS, Skarin A, Hinrichs A, Warhol MJ. Ampullary somatostatinoma: psammomatous variant of gastrointestinal carcinoid t u m o r - - a n immunohistochemical and ultrastructural study. Report of a case and review of the literature. Am J Clin Pathol 1 9 8 3 ; 8 0 : 7 5 5 - 7 6 1 . 10. Hopman WPM, van Liessum PA, Pieters GFFM, Smals AGH, Tangerman A, Jansen JBMJ, Rosenbausch G, Lamers CBHW, Kloppenborg PWC. Pancreatic exocrine and gallbladder function during long-term treatment with octreotide (SMS 201-995). Digestion 1990;45(suppl 1):72-76. 11. Lembcke B, Creutzfeldt W, Schleser S, Ebert R, Shaw C, Koop I. Effect of somatostatin analogue sandostatin (SMS 201-995) on gastrointestinal, pancreatic, and biliary function and hormone release in normal men. Digestion 1 9 8 7 ; 3 6 : 1 0 8 - 1 2 4 . 12. Ahrendt SA, McGuire GE, Pitt HA, Lillemoe KD. Why does somatostatin cause gallstones? Am J Surg 1991; 1 6 1 : 1 7 7 - 1 8 3 . 13. Dharmsathaphorn K, Binder HJ, Dobbins JW. Somatostatin stim-

May 1995

14.

15.

16. 17.

18.

19. 20.

21.

22.

23.

24.

25.

26.

27.

ulates sodium and chloride absorption in the rabbit ileum. Gastroenterology 1980; 7 8 : 1 5 5 9 - 1 5 6 5 . Dharmsathaphorn K, Racusen L, Dobbins JW. Effect of somatostatin on ion transport in the rat colon. J Clin Invest 1980; 53:813-820. Guandalini S, Kachur JF, Smith PL, Miller RJ, Field F. In vitro effects of somatostatin on ion transport in rabbit intestine. Am J Physiol 1 9 8 0 ; 2 3 8 : G 6 7 - G 7 4 Gurll N, DenBesten L. Animal models of human cholesterol gallstone disease: a review. Lab Anim Sci 1 9 7 8 ; 2 8 : 4 2 8 - 4 3 2 . Roslyn JJ, Abedin MZ, Saunders KD, Cates JA, Strichartz SD, Alpirin M, Fromm M, Palant CE. Uncoupled basal sodium absorption and chloride secretion in prairie dog gallbladder. Comp Biochem Physiol 1 9 9 1 ; 1 0 0 A : 3 3 5 - 3 4 1 . Schultz SG, Zalusky R. Ion transport in isolated rabbit ileum. I. Short circuit current and Na + fluxes. J Gen Physio11964; 4 7 : 5 6 7 584. Segel IH. Biochemical calculations. New York: Wiley, 1976:236. Saunders KD, Cates JA, Abedin MZ, Kleinman R, Roslyn JJ. Calcium channel blockade inhibits gallbladder ion transport. J Surg Res 1 9 9 0 ; 4 9 : 3 0 6 - 3 1 0 . Lee SP. Enhanced fluid transport across gallbladder mucosa in experimental cholelithiasis. Am J Physiol 1 9 7 8 ; 2 3 4 : E 5 7 5 E578. Conter RL, Roslyn JJ, Porter-Fink V, DenBesten L. Gallbladder absorption increases during early cholesterol gallstone formation. Am J Surg 1 9 8 6 ; 1 5 1 : 1 8 4 - 1 9 1 . Strichartz SD, Abedin MZ, Abdou MS, Roslyn JJ. The effects of amiloride on biliary calcium and cholesterol gallstone formation. Ann Surg 1 9 8 9 ; 2 0 9 : 1 5 2 - 1 5 6 . Admirand WH, Small DM. The physicochemical basis of cholesterol gallstone formation in man. J Clin Invest 1 9 6 8 ; 4 7 : 1 0 4 3 1052. Roslyn JJ, DenBesten L, Pitt HA, Kuchenbecker S, Polarek JW. Effects of cholecystokinin on gallbladder stasis and cholesterol gallstone formation. J Surg Res 1 9 8 1 ; 3 0 : 2 0 0 - 2 0 6 . Holtzbach RT, Marsh M, Olszewiski M, Holan K. Cholesterol solubility in bile: evidence that supersaturated bile is frequent in healthy man. J Clin Invest 1 9 7 3 ; 5 2 : 1 4 6 7 - 1 4 7 9 . Gollish SH, Burnstein JM, Ilson RG, Petrunka CN, Strasberg SM.

OCTREOTIDE PROMOTES GALLBLADDER ABSORPTION 1555

28.

29.

30.

31.

32.

33.

34.

35.

36.

Nucleation of cholesterol monohydrate crystals from hepatic and gallbladder bile of patients with cholesterol gallstones. Gut 1983; 2 4 : 8 3 6 - 8 4 4 . Strichartz SD, Abedin MZ, Abdou MS, Roslyn JJ. Increased biliary calcium in cholesterol gallstone formation. Am J Surg 1988; 155:131-137. Magnuson TH, Lillemoe KD, Scheers DE, Pitt HA. Altered bile composition during cholesterol gallstone formation: cause or effect? J Surg Res 1 9 9 0 ; 4 8 : 5 8 4 - 5 8 9 . Kibe A, Dudley MA, Halpern Z, Lynn MP, Breuer AC, Holzbach RT. Factors affecting cholesterol monohydrate nucleation time in model bile solutions of supersaturated bile. J Lipid Res 1985; 2 6 : 1 1 0 2 - 1 1 1 1 . Ren~ E, Danzinger RG, Hofmann AF, Nakagaki M. Pharmacologic effect of somatostatin on bile formation in the dog. Gastroenterology 1983; 8 4 : 2 0 - 2 9 . Anthone GJ, Bastidas JA, Orandle MS, Yeo CJ. Direct proabsorptive effect of octreotide on ionic transport in the small intestine. Surgery 1990; 1 0 8 : 1 1 3 6 - 1 1 4 2 . Petersen KU, Reuss L. Cyclic AMP-induced chloride permeability in the apical membrane of Necturus gallbladder epithelium. J Gen Physiol 1 9 8 3 ; 8 1 : 7 0 5 - 7 2 9 . Hausdorff WP, Aguilera G, Catt KJ. Inhibitory actions on somatostatin on cyclic AMP and aldosterone production in agonist-stimulated adrenal glomerulosa cells. Cell Signal 1989; 1 : 3 7 7 - 3 8 6 . Zink A, Scherubl H, Kleinman D, Hoflich M, Zeigler R, Raue F. Inhibitory effect of somatostatin on cAMP accumulation and calcitonin secretion in C-cells. Involvement of pertussis toxin-sensitive G-proteins. Mol Cell Endocrinol 1 9 9 2 ; 8 6 : 2 1 3 - 2 1 9 . Dharmsathaphom K, Sherwin RS, Dobbins JW. Somatostatin inhibits fluid secretion in the rat jejunum. Gastroenterology 1980; 7 8 : 1 5 5 4 - 1 5 5 8 .

Received January 18, 1994. Accepted January 13, 1995. Address requests for reprints to: Joel J. Roslyn, M.D., Department of Surgery, Medical College of Pennsylvania and Hahnemann University, 3300 Henry Avenue, Philadelphia, Pennsylvania 19129. Fax: (215) 843-1095. Supported by Merit Review from the Department of Veterans Affairs.