Cyclic nucleotides and glycoproteins during formation of cholesterol gallstones in prairie dogs

GASTROENTEROLOGY

1984:87:263-g

Cyclic Nucleotides and Glycoproteins During Formation of Cholesterol Gallstones in Prairie Dogs R. A. ZAK, P. G. FRENKIEL, J. W. MARKS, A. ALLEN, and L. J. SCHOENFIELD

G. G. BONORRIS,

Cedars-Sinai England

and Newcastle

Medical

Center

and 1JCLA. Los Angeles,

Male prairie dogs received in standard diets either 0.08% cholesterol (control, n = 30) or 1.2% cholesterol (lithogenic, n = 31). Animals were killed at days 2-4, 7. 10, 21, and 39 to determine the temporal sequence of changes in mucosal cyclic adenosine 3’ : 5’-monophosphate in the gallbladder and of cholesterol saturation, glycoproteins, cholesterol crystals, and gallstones in bile of prairie dogs fed a cholesterol-rich lithogenic diet. Glycoprotein concentration in bile in the lithogenic group was significantly elevated compared to controls on all days of death. Saturation of bile and formation of cholesterol crystals occurred only in the lithogenic group, detected first after 7 days of feeding. Gallstones were found in the lithogenic group only. Elevation of cyclic adenosine 3’ : 5’-monophosphate in the mucosa of gallbladders was found in the lithogenic group increased only, beginning at day IO. In summary, glycoproteins in bile preceded cholesterol saturation and crystallization which, in turn, preceded increased mucosal cyclic adenosine 3’ : 5’-monophosphate. Studies in animal models (l-3) and humans (4) suggest that mucous glycoproteins secreted by the gallbladder in response to lithogenic bile may func-

Received August 1, 1983. Accepted February 6, 1984. Address requests for reprints to: Leslie J. Schoenfield, M.D., Division of Gastroenterology, Cedars-Sinai Medical Center. 8700 Beverly Boulevard. Los Angeles, California 90048. This work was supported in part with federal funds from the Department of Health and Human Services under Grant No. 16531. The contents of this publication do not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the United States Government. c 1984 bv the American Gastroenterological Association 0016-X)85184/$3.00

California;

IJpon

Tynr

tion as a nucleating agent in the formation of cholesterol gallstones. Freston et al. (1) observed increased mucous substances and elevated hexosamine concentrations in the bile, gallbladders, and gallstones of rabbits fed a lithogenic diet. Womack (2) noted that in hamsters fed a lithogenic diet, hypersecretion of mucus preceded formation of gallstones. Lee et al. (3) reported elevated levels of glycoproteins in the bile and mucosal scrapings from the gallbladders of patients with gallstones compared to patients with normal gallbladders. In another study, Lee et al. (4), using in vitro [“Hlglucosamine incorporation into glycoproteins showed that explants from gallbladders of prairie dogs fed a lithogenic diet synthesized and secreted a greater amount of glycoproteins than controls. The cellular mechanisms for increased synthesis of glycoproteins by gallbladders containing supersaturated bile and gallstones are unknown. Lee et al. (5) and DeBenedetto (6) have shown inhibition of synthesis and secretion of glycoproteins in explants of gallbladders by prior feeding of aspirin or indomethacin in prairie dogs. These studies implicate prostaglandins as mediators of increased synthesis of glycoproteins. The hypothesis that cyclic adenosine 3’: 5’-monophosphate (CAMP) may regulate the synthesis of glycoproteins by the gallbladder is inferred from studies of other organs. Studies in the small intestine (7) and colon (8) indicate that CAMP may stimulate the synthesis and secretion of mucous glycoproteins. Forstner et al. (7) showed increased incorporation of [14C]glucosamine into the mucosa of the small intestine of rats by theophylline, dibutyrl CAMP, and /?-adrenergic drugs. LaMont et al. (8) demonstrated increased uptake and secretion of Abbreviations monophosphate; diesterase.

used in this paper: cAMP. cycIiL adenosine 3’:5’GLC, gas-liquid chromatographv: PDE. phospho-

264

GASTKOEN'l'EKOLOGY \'ol. 87. Ro. 2

ZAK ET AL.

[“Hlglucosamine by mucosal biopsy specimens of rabbit colon in response to dibutyrl cAMP. In addition to gallbladder mucin, other factors that may be important in the pathogenesis of gallstones include increased cystic duct resistance (9) and delayed gallbladder emptying (lo), both of which produce stasis within the gallbladder. Cholesterol gallstones were prevented in prairie dogs fed a lithogenic diet both by sphincterotomy (11) and by facilitating emptying of the gallbladder in response (12) or lipoprotein gavage (10). to cholecystokinin Recently, Doty et al. (13) reported on the temporal sequence of events that leads to the formation of cholesterol gallstones in prairie dogs fed a lithogenic diet. This sequence included, in order of occurrence, hepatic secretion of lithogenic bile, increased production of mucus, formation of cholesterol crystals. and finally, stasis of the gallbladder. Our hypothesis was that in response to saturated bile, cyclic nucleotides mediate increased secretion of glycoproteins which, in turn, facilitates formation of gallstones. The aim of the present study was to determine the temporal sequence of saturation of bile, increased cyclic nucleotides in the mucosa of gallbladders, increased glycoproteins in bile, and the formation of crystals and gallstones in prairie dogs fed a cholesterol-rich diet.

Experimental Sixty-one omys

Design

black-tailed, prairie dogs (Cyweighing 0.65-1.25kg were obtained

male.

ludoviuianus)

from Otto M. Locke,

New Braunfels, Texas. The animals underwent at least a 2-wk adaptation period during which time they were fed standard Purina laboratory Chow (Ralston Purina Co., St. Louis, MO.) containing O.OF% cholesterol by weight. The animals were kept in stainless steel cages and housed 1 to 2 to a cage with access to food and water ad libitum. Temperature in the vivarium was maintained at 23"C,and a period of light from 4 PM to 4 AM was controlled by an electric timer. After the adaptation period, the prairie dogs were divided randomly into two groups. Thirty animals served as the experimental group (lithogenic) and received a diet containing 1.2”/,, cholesterol by weight (Teklad Test Diets, Division of ARSiSprague-Dawley, Madison, Wis.). Thirty animals served as controls and consumed the same diet as the lithogenic group, except that the cholesterol content remained 0.08% by weight. Both diets contained sufficient calories, essential fatty acids. and vitamins designed to maintain the health and weight of the animals. The diet contained 230/, protein, 45% carbohydrate, and 23”/, fat by weight. The diet was kept refrigerated in the dark to minimize oxidation of cholesterol, and was analyzed weekly by gas-liquid chromatography (GLC) to confirm the content of cholesterol. The animals were killed after 2-4, 7, 10, 21, and 39 days of feeding. At death, the following were determined without knowledge of the group: weight

of the animals; cholesterol monohydrate crystals and liquid crystals, mucous gel, glycoproteins. saturation index, and gallstones in bile of gallbladders: LAMP and phosphodiesterase activity in mucosa of gallbladders. Results are expressed as mean I SE,M and comparisons among groups were evaluated by the unpaired Student’s ttest.

Methods and Procedures Collection

of Bile

A predetermined number of animals from both the lithogenic and control groups were studied upon death (Table 1). The animals were killed between 9 and 11 .UI after a 12-h fast. Ketamine HCl 100 mgikg was injected i.m. to induce unconsciousness. The abdominal viscera ivere exposed by a midline incision and the gallbladders lvere identified. Gallbladder bile was aspirated lvith a 20-gauge needle and syringe prewarmed to 37°C.

Cholesterol

CrysluJs

and

Gallstones

in Bile

One drop of bile was placed on a glass slide tor immediate light and polarizing microscopy. Presence of cholesterol crystals was defined as one or more crystals per 100x power field. Cholesterol monohydrate crystals were identified by birefringence and notched rhomboidal shapes. Liquid crystals, which appear as birefringent maltese crosses, were sought but not seen, and mucous gel was identified as nonbirefringent amorphous strands. The remaining bile was used for measurement of glycoproteins and lipids. After aspiration of the bile, the gallbladders were removed, opened by a longitudinal incision, and examined for stones. Stones also were sought in the common duct. The stones were removed, rinsed with water, frozen, and stored for subsequent analysis. Histology and Preparation Gallbladder

of Mucosa

of

A section of intact gallbladder was sent to pathology for histologic staining for mucus using periodic acid& Schiff and Alcian Blue, examination of goblet cells. and evidence of cholecystitis using H & E. The pathologist had no knowledge of which group of animals the sample came from. The mucosa of the remaining gallbladder was separated from the muscularis by scraping with a scalpel. To verify that complete separation of the mucosa had been acconplished, the denuded gallbladder also was examined b! microscopy. ‘J‘able I.

Number of i\nimuls Studied in the ~ontrof und Lithogenic Groups nt Death

August 198.I

Analysis

CAMP AND MLJCIJS IN GALLSTONE:

of BiJiary

Determination

Lipids

of Saturation

und

Index

Bile samples were analyzed for cholesterol and individual bile acids by GLC (14), for phospholipids by a modification of the calorimetric method of Fiske and Subbarow (15). and for total bile acids by the steroid dehydrogenase method of Talalay (16). The cholesterol saturation index was calculated according to the formula of Thomas and Hofmann (17), based on the cholesterol saturation equilibrium of Hegardt and Dam (18). The calculations are applicable because the total lipid concentration in the bile in the gallbladder of the prairie dog was 8.3 -+ 0.8 [controls) and 8.2 -t 1.0 (experimental). range 6.6-10.3

g’%,.

GJ!roproteins

in Bile

Samples of bile (l-2 cm,’ each) were pipetted into dialysis bags (Sigma Chemical Co., St. Louis, MO.) and dialyzed in distilled water for 5 days to remove pigments that could potentially interfere with the calorimetric determination of glycoproteins. These samples then underwent column chromatography on 16 x l-cm columns of Sepharose 4B-Cl (Sigma). The column was eluted with 0.2 M NaCUO.O2% wtivol sodium azide. The Sepharose 4B chromatography separated the glycoprotein from free protein and residual bile salts and bile pigments as shown by Pearson and colleagues (19). The void volume was not tested for carbohydrate and protein content. The void volume was collected and used for subsequent quantification of glycoproteins in bile. The method of Mantle and Allen (20) was used for the determination of glycoprotein in bile. It is a sensitive calorimetric method for measuring polysaccharides that are oxidized by periodate. The Schiff reagent was prepared and the oxidized glycoprotein was coupled to the Schiff base. Schiff reagent was prepared by dissolving 1 g of basic fuschin (Eastman Kodak Co., Rochester, N.Y.) in 100 ml of boiling water and by then adding 20 ml of 1 M HCl to the solution cooled to 50°C. The solution was mixed twice with 300 mg of activated charcoal shaken for 5 min. and filtered to remove the charcoal. The resulting solution was then stored in an amber glass bottle at room temperature. Directly before use. 0.150 g of sodium metabisulfite [Sigma) was added to every 6 ml of Schiff reagent required and the solution was incubated at 37°C until it was colorless or just pale vellow (-1.5 h). Lyophilized mucin from bovine submaxillary glands (Sigma) was used to develop a standard curve. Between 25 and 400 pg of bovine submaxillary mucin in 2 ml of water was incubated for 2 h at 37°C with 0.2 ml of freshly made solution of periodic acid. A l-cm” sample of biliary glycoprotein was similarly incubated. The solution of periodic acid was prepared by adding 10 ~1 of periodic acid (50% solution) to 10 ml of 7% acetic acid. After periodate oxidation, 0.2 ml of Schiff solution (containing the metabisulfite) was added to the solution of glycoprotein. To allow for full development of color, the resulting solution was left for 30 min at room temperature before absorbance was read at 555. Samples of bile that

l-‘OKMATION

265

had undergone dialysis and column chromatography also were analyzed by GLC for constituent sugar residues. Trimethylsilyl ether derivatives were chromatographed on SE-30 columns as described by Bhatti and colleagues (21). The patterns found on GLC were compared to those of standards which included N-ac:etylglucosamine. fucose. galactose, and mannose (Sigma).

Preparation Adenosine

of Tissue

for Analvsis

of‘ Cyclic

3’: 5’-Monophosphute

The mucosa of gallbladders ~‘as separated from the muscularis by gentle scraping Lvith the edge of a scalpel. and immediately placed in ice-cold buffer (0.05 M TrisHCl, pH 7.5). It then was rapidly minced and homogenized by six strokes of an ice-cold, hand-driven conical Kontes glass tissue homogenizer (Kontes Co., \:ineland. N.J.) and strained through gauze. The time from initiation of mucosal stripping to homogenization was 5 min.

Assay

of Cyclic

Adenosint:

3’ : s’-

Monophosphate Cyclic adenosine 3’ : 5’-monophosl)hate was determined by a competitive protein-binding assay (22) cammercially supplied as a kit from Amersham-Searle Corp., Arlington Heights, Ill. The assay was performed as we have done previously in the mucosa of the small intestine of the rabbit (23) at 37°C and conditions of time (30 min) and concentration of CAMP (4-8 pmol,‘mg of protein] shown to be optimal in this laboratory for the mucosa of the gallbladder in the prairie dog. Duplicate determinations of cAMP agreed within 5%. Fresh mucosa was homogenized immediately in an icecold 0.5 M Tris buffer, pH 7.5. The cold homogenate was deproteinized with 0.1 M ZnSO., and 0.1 bl Ba(OH)L and centrifuged at 10,000 rpm for 15 min. The supernatant was lyophylized and the residue was redissolved in 0.05 Iv1 Tris buffer for the CAMP determination. This method is based on the competition between unlabeled cXMP (Calbiochem, San Diego, Calif.) and a fixed quantity of [“HIcAMP (New England Nuclear, Boston, Mass.) for binding to a protein that has a high specificity and affinity for CAMP. The amount of labeled protein-(AMP complex formed is inversely related to the amount of unlabeled cAMP present in the assay. The concentration of cAMP in the unknown was determined by comparison with a linear standard curve.

Assay

of Activity

of Phosphodiesternse

The activity of phosphodiesterase (PDE) was determined by the method of Brooker et al. (24) as modified b> Kantor et al. (25). The method depends on hydrolysis of [“Hladenosine (New England Nuclear) after phosphorolysis by 5’-nucleotidase (Sigma) in the incubation mixture. The assay was performed at conditions of time (10 min) and protein concentration (0.4 mg) shown to be optimal in our laboratory for prairie dog mucosa of gallbladder. Cyclic adenosine 3’: 5’.monophosphate degradation was

266

GASTROENTEROLOGY

ZAK ET AL

linear up to 30 min using 0.1-0.7 mg of protein per assay. Fifty microliters of the supernatant obtained for PDE assay containing 0.3-0.4 mg of protein were added to 150 ~1 of incubation mixture. The final composition of the 150-~1 incubation mixture was 30 mM Tris-HCl (pH 7.5), 10 mM MgClz, 0.01 mM CAMP, 0.67 mgiml of bovine serum albumin (Sigma), 0.33 mg/ml of 5’-nucleotidase (cobra venom), and 0.07 yCi of [“HIcAMP, ammonium salt (sp act 38.4 cpm). The reaction was carried out at 37°C in polyethylene scintillation vials for 10 min and stopped by the addition of 1 ml of a6-l-x2 resin slurry (Bio-Rad Labs, Richmond, Calif.) in water (1: 1). The mixture was allowed to equilibrate for 10 min at room temperature. Then, 10 ml of scintillation fluid [New England Nuclear) were added and the activity of [3H]adenosine was counted in a Searie Mark III model 6880 liquid scintillation counter (Searle Pharmaceuticals, Inc., Chicago, Ill.). A reaction blank for each experiment was determined in the absence of homogenate. The activity of PDE corrected for the experimental blank was expressed as the amount of CAMP hydrolyzed in 5’AMP per milligram of protein per minute and as a percentage of the control value. Determinations were performed in triplicate.

Results Weight

No significant differences were found in the mean body weights of the animals between the lithogenie and control groups either before institution of the experimental diets or at the time of death. The mean body weight after the adaptation period was 924 5 142 g in the lithogenic group and 955.83 + 111.21 g in controls. The mean gain in weight at the time of death was 125 2 57 g in the lithogenic group and 109 ? 68 g in the control group; this difference also was not significant.

10

7 DAYS

Figure

87. No.

2

Index

Figure 1 illustrates the mean saturation index of bile in the control and lithogenic groups at the time of death. At no time did bile become saturated in the control group, whereas in the lithogenic group, bile became saturated at 7 days of feeding, and bile also was saturated at 10, 21, and 39 days of feeding. In the lithogenic group, the saturation index was higher on day 7 vs. days 2-4 (p < 0.02) and on day 21 vs. day 7 (p < 0.02). No significant difference was found between the control and lithogenic groups at 2-4 days of feeding, but at 7, 10, 21, and 39 days the saturation index was higher (p < 0.01)in the litho-

genie group. The higher saturation index was due to a larger molar percent cholesterol and a lesser molar percent phospholipid. Cholesterol

Crystals and Gallstones

Cholesterol monohydrate crystals were not found in the control group at any time during the study. In the lithogenic group, cholesterol monohydrate crystals were identified in all 5 animals at 7 days, 5 of the 6 animals at 10 and 21 days, and all 6 animals at 39 days of feeding. Gallstones were found in the bile of the gallbladder only in the lithogenic group, in 1 animal after 10 days of cholesterol

of Animals

2-4

Saturation

Vol.

21

OF FEEDING

1 Mean saturation index in control (C) (open bars) compared to lithogenic (L) (hatched bars) after Z-4, 7, IO, 21, and 39 days of feeding. Vertical lines represent mean 5 SEM. The numbers of observations at each time-point in this and in each of the other figures are as shown in Table 1.

feeding, and in 1 animal after 21 days. Amorphous strands, similar to those previously described and illustrated (8), were found only in the lithogenic group and probably represent mucous glycoprotein. Glycoprotein Identification of the constituent sugar moieties of the glycoproteins by GLC showed them to contain N-acetylglucosamine, fucose, and galactose, as would be expected in mucous glycoprotein. The substance contained no mannose, indicating no contamination by serum glycoproteins. Problems exist and assumptions are made with all of the methods that are available for the estimation of glycoprotein, so that the measurements are semiquantitative at best. In the method used, steps were not taken to remove noncovalently bound small molecular weight glycoproteins and nucleic acids. Furthermore, the samples were not analyzed for N-acetyl galactosamine or sialic acid that might have provided further evidence for the suggested presence of mucous glycoproteins. Accordingly, in this paper, reference will be made to glycoproteins rather than mucous glycoproteins. Duplicate and triplicate determinations of glycoproteins agreed within 5%. Figure 2 illustrates the mean concentration of glycoproteins measured in the bile in the control and

August

1984

cAMP

55 -I

z 45 EB

Lltho6enic

‘a

40

s

35

z

30

8 ’

25

8

20

g

15

d

10

ps.02 *

IL)

IN GALLSTONE

FORMATION

267

WC

PC.01 I

Phosphodiesterase

5 2-4

7

10 DAYS

Figure

MUCUS

day 21 than 39 (p < 0.05). Also in the lithogenic group mucosal CAMP levels at days 10, 21, and 39 were significantly greater than at days 2-4 and 7 (p < 0.01).

* * * p<.o5 **

AND

21

39

OF FEEDING

2. Mean (open

concentrations of glycoproteins in control (C) bars) vs. lithogenic (L) (hatched bars) after 2-4. 7. 10, 21. and 39 days of feeding. Vertical lines represent mean + SEM.

lithogenic groups at death. Glycoprotein was present in the bile of controls at death, without significant differences among different times of death. Mean glycoprotein concentrations in the lithogenic group were significantly higher than in controls at death. In the lithogenic group, the concentration of glycoproteins at day 10 was significantly greater than those at days 2-4, 7, 21, and 39 (p < 0.01). Cyclic Adenosine

ActivjtJ

Figure 4 illustrates the mean activity of PDE in the control and lithogenic groups at death. The PDE activity did not change with time in either the control or lithogenic group and no significant differences between these groups were found at any time of death. This suggests that the rise in CAMP was stimulated by increased activity of adenylate cyclase rather than by inhibition of the activity of phosphodiesterase. Morphology

of Gallbladders

No gallbladder tissue showed histologic evidence of cholecystitis. Histologic examination confirmed that only mucosa had been removed from the gallbladder in the scraping procedure. No differences between the control and lithogenic groups in mucous staining of the goblet cells of the gallbladder were noted.

3’: 5’-Monophosphate

Figure 3 illustrates the mean quantity of mucosal CAMP in the control and lithogenic groups at death. Cyclic adenosine 3’: 5’-monophosphate was present in the mucosa of gallbladders of controls at death, without significant differences among the different times of death. No significant differences were found in mucosal CAMP between the control and lithogenic groups at days 2-4 and 7, but CAMP was significantly higher in the lithogenic group at days 10, 21, and 39 (p < 0.01).In the lithogenic group, no significant difference was found between days 10 and 21,but CAMP was significantly higher at

Discussion As found in previous studies, this current study in prairie dogs has demonstrated increased concentrations of glycoproteins in the bile of animals fed a cholesterol-rich lithogenic diet (4,13). The factors in lithogenic bile that are responsible for the stimulation of increased secretion of glycoprotein are unknown. Substances that have been implicated include cholesterol (4,13), lysolecithin (26),and calcium (27,28). In explants of gallbladders from prairie dogs fed a lithogenic diet, prior feeding of indomethacin inhibited the synthesis and secre-

40 36

2 &

32

&

26

cl ??

Control

(CI

Lithogenic

* p<.Ol

(L)

vrc

g

24

5

150

;

20

e5

140

f

16

c

130

e

12

120

i

6

5 F $j

o

4

g 0

100

2-4

10

7 DAYS

Figure

21

110

2-4

39

OF FEEDING

3. Mean quantity of mucosal cyclic adenosine 3’:5’-monophosphate in control (C) [open bars] vs. lithogenic (L) (hatched bars] after 2-4. 7, 10. 21. and 39 days of feeding. Vertical lines represent mean 2 SEM.

7

10 DAYS

Figure

4. Mean (open

21

39

OF FEEDING

phosphodiesterase activity (PDE) in control (C) bars) vs. lithogenic (L) (hatched bars] after 2-4. 7. 10, 21, and 39 days of feeding. Vertical fines represent mean t SEM.

268

ZAK ET AL.

tion of glycoproteins. This also was associated with FI-a, a decreased release of 6-keto-prostaglandin product of the breakdown of prostacyclin, implicating prostaglandins as mediators of the increased synthesis and secretion of glycoproteins. In the present study, the hypothesis was tested that CAMP is a cellular mediator of increased secretion of glycoproteins by gallbladders of prairie dogs fed a lithogenic diet. Cyclic adenosine monophosphate has been shown to mediate secretion of mucin in other gastrointestinal organs (7,8), but this has not been previously evaluated in the gallbladder. The concentration of glycoproteins in the bile was significantly elevated as early as days 2-4 after the initiation of cholesterol feeding, whereas significant elevation of mucosal CAMP was not found until death at 10 days of feeding. This finding is not consistent with CAMP mediation of glycoprotein synthesis and secretion; however, the initial elevation of glycoproteins in the bile (days 2-4 and 7) could represent release of stored glycoproteins and the augmented concentration at day 10 could represent secretion of newly formed glycoproteins. If this were so, the temporal sequence would be compatible with CAMP as the mediator of the secretion of glycoprotein. If the control of the synthesis and secretion of glycoprotein were through a feedback mechanism governed by the concentration of stored intracellular glycoprotein, however, release of stored mucus could be the sitmulus for increased production of mucus. Bile did not become saturated and crystals did not form until after 7 days of feeding the lithogenic diet. The finding of an elevation of the glycoproteins at days 2-4, therefore, does not support the hypothesis that lithogenic bile stimulates production of mucus, unless, as discussed above, stimulation of mucus did not occur until day 10. Additional observations on day 5 or 6 might have provided finer discrimination of the sequence of events. Also, raised serum cholesterol in response to the diet might have mediated all subsequent changes directly or indirectly. The elevation of glycoproteins in the bile before saturation of bile found in this study is in contrast to the temporal sequence of events reported by Doty et al. (13) in which lithogenic bile preceded formation of mucus. The difference could be explained by the fact that in their study, the first death was after 2.5 wk of feeding the lithogenic diet, whereas in the present study, death occurred at earlier times, possibly increasing the sensitivity to detect an early change. Furthermore, our data are compatible with that of Lee et al. (4) who demonstrated, in vitro, increased incorporation of [“Hlglucosamine into mucous glycoproteins in explants of gallbladders of prairie dogs as early as 3 days after initiating a

lithogenic diet, at a time when bile was not saturated with cholesterol. In contrast to other reports in prairie dogs (4,11,29), this study did not achieve consistent formation of cholesterol gallstones. The difference may be explained by differences in the adaptation period or the source and composition of the diets. Lee (4) described formation of cholesterol gallstones in 5 of 11 animals after 14 days of feeding a lithogenic diet containing 1.2% cholesterol. Hutton (11) similarly reported formation of cholesterol gallstones in 11 of 12 animals after 6 wk of feeding a 1.2% cholesterol diet. Doty (13) found gallstones in 7 of 15 animals fed a diet containing only 0.4% cholesterol for up to 16 wk. The possibility exists that the temporal appearance of saturated bile and crystallization on the one hand and changes in the content of glycoprotein in bile on the other might have been different if the animals in our study had responded to the lithogenic diet with a greater frequency of gallstones. In conclusion, prairie dogs fed a cholesterol-rich diet demonstrate the following: (a) increased glycoproteins in bile preceded cholesterol saturation and crystallization which in turn preceded increased mucosal CAMP; (b) the appearance of glycoproteins in bile prior to formation of crystals supported their role as nucleating agents; and (c) no evidence was found to support the hypothesis that saturated bile or CAMP promoted formation of glycoproteins.

References 1. Freston JW. Bouchier IAD. Newman J. Biliary mucous substances in dihydrocholesterol-induced cholelithiasis. Gastroenterology 1969:57:670-8. 2. Womack NA. The development of gallstones. Surg Gyncxol Obstet 1971;133:937-45. 3. Lee SP. Lim TH, Scott AJ. Carbohydrate moieties of glycoproteins in human and gallbladder bile. gallbladder mucosa, and gallstones. Clin Sci Mol Med 1979;56:533-8. 4. Lee SP, LaMont JT, Carey MC. Role of gallbladder mucus hypersecretion in the evolution of cholesterol gallstones. J Clin Invest 1981:67:1712-23. 5. Lee SP. Carey MC, LaMont JT. Aspirin prevention of cholesterol gallstone formation in prairie dogs. Science 1981: 211:1429-32. 6. DeBenedetto D, Turner B, Handin RI, LaMont JT. Indomethacin inhibits gallbladder secretion and prostaglandin release (abstr). Gastroenterology 1981;80:1133. 7. Forstner GG. Shih M, Lukie B. Cyclic AMP and intestinal glycoprotein synthesis. The effect of beta-adrenergic agents, theophylline, and dibutyrl CAMP. Can J Physiol Pharmacol 1973;51:122-9. 6. LaMont JT. Ventola A. Stimulation of colonic glycoprotein synthesis by dibutryl CAMP and theophylline. Gastroenterology 1977;72:82-6. 9. Pitt HA, Roslyn JJ. Kuchenbecker AL, et al. The role of cystic duct resistance in the pathogenesis of cholesterol gallstones. 1 Surg Res 1981;30:508-14. IO. Roslyn JJ, DenBesten L, Thompson JE. Effects of periodic:

August

1984

emptying 11.

12.

13.

14.

15. 16. 17.

18.

cAMP

of gallbladder

on gallbladder

function

and

forma-

tion of cholesterol gallstones. Surg Forum 1979;30:403-4. Hutton SLV. Sievert CE, Vennes JA, Duane WC. The effect of sphincterotomy on gallstone formation in the prairie dog. Gastroenterology 1981;81:663-7. Roslyn JJ. DenBesten L. Pitt HA, et al. Effects of cholecystokinin on gallbladder stasis and cholesterol gallstone formation. 1Surg Kes 1981;30:200-4. Doty JE. Pitt HA, Kuchenbecker SL, Porter-Fink V, DenBesten LM’. Kale of gallbladder mucus in the pathogenssis of cholesterol gallstones. Am J Surg 1983;145:54&61. Schoenfield LJ, Bonorris GG, Ganz P. Induced alterations in the rate-limiting enzymes of hepatic cholesterol and bile acid synthesis in the hamster. J Lab Clin Med 1973;82:858-68. Fiske CH. Subbarow 1’. The calorimetric determination of phosphorus. J Biol Chem 1925;66:375-400. Talalay P. Enzymatic analysis of steroid hormones. Meth Biol Chem Anal 1960:8:119-43. Thomas PJ. Hofmann AF. A simple calculation of lithogenic index of bile: expressing biliary lipid composition on rectangular coordinates. Gastroenterology 1973;65:698-700. Hegardt FG. Dan H. The solubility of cholesterol in aqueous solution of bile salt and lecithin. Z Ernahrungswiss 1971: 10:239.

19. f’earson JF. Kaura K, Taylor W. Allen A. The composition and polymeric structure of mucus glycoprotein from human gallbladder bile. Biochim Biophys Acta 1982;706:221-8. 20. Mantle M, Allen A. A colorometric assay for glycoproteins based on the periodic acid/Schiff stain. Biochem Sot: Tram 1978;6:607-9.

AND

MUCUS

IN GALLSTONE

E’OKMATION

269

21. Bhatti T. Chambers RE, Clamp JR. The gas chromatographic properties of biologically important n-acetylglucosamine derivatives, monosaccharides, trisaccharides. tetrasaccharides, and pentasaccharides. Biochim Biophys Acta 1!170:222:33947. 22. Gilman AG. A protein binding assay for c-AMP. Froc Nat1 Acad Sci USA 1970;67:305-12. 23. Taub M, Bonorris GG, Chung A. Coyne MJ, Schoenfield LJ. Effect of propranolol on bile acid- and cholera enterotoxinstimulated cAMP and secretion in rabbit intestine. Gastroenterology 1977;72:101-5. 24. Brooker G, Thomas LJ. Applemsn MM. The assay of cyclic AMP and cyclic GMP in biological materials by enzymatic radioisotope displacement. Biochemistry 1968;7:4177-81. 25. Kantor HS, Tao P, Gorbach SL. Stimulation of intestinal adenylate cylase by E. coli enterotoxin: comparison of strains from an infant and an adult with diarrhea. J Infect Dis 1974;12Y:l-9. 26. Neiderhiser D, Thornell E. Bjorck S. Svanik J. Effect of lysophosatidyl-choline on gallbladder function in the cat (abstr). Gastroenterology 1981;80:1239. 27. Malet FF, Soloway RD, Trotman BW. Calcium alters gallbladder glycoprotein secretion (abstr). Clin Res 1981;29:668A. 28. Doty JE. Roslyn JJ, Pitt HA, et al. Alterations in gallbladder calcium and protein concentration during cholesterol gallstone formation (abstr). Clin Res 1981:29:31A. 29. Doty JE, Porter-Fink V, DenBesten L. et al. Cholesterol saturated bile increases gallbladder blood flow (abstr). Clin Kes 1982;30:34A.