Oxidations of 17beta-estradiol and estrone and their interconversions catalyzed by liver, mammary gland and mammary tumor after acute and chronic treatment of rats with indole-3-carbinol or beta-naphthoflavone.

Altered cytochrome P450-catalyzed metabolism of 17beta-estradiol (E2) and estrone (E1) in the liver and (or) extrahepatic tissues may affect estrogen-sensitive tumorigenesis. We examined the effects of oral treatments of (i) indole-3-carbinol (13C) at 250 or 500 mg/kg or beta-naphthoflavone (beta-NF) at 40 mg/kg of body weight (bw)/day from 51 to 54 days of age (acute regimen), and (ii) 13C at 250 mg/kg or beta-NF at 20 mg/kg bw given 3x/week from 10 to 22 weeks of age (chronic regimen) in female Sprague-Dawley rats. We determined the effects of these treatments on the P450 content and P450 (CYP)-specific activities in the liver, P450-dependent metabolism of E2 and E1 by the liver and mammary gland, and interconversion of E1 and E2 catalyzed by 17beta-hydroxysteroid dehydrogenase (17beta-HSD) in these tissues and malignant mammary tumors. 13C at the two levels of acute regimen elicited similar responses. Acute and chronic treatments with 13C, but not beta-NF, increased P450 content approximately 2-fold. 13C, and to a lesser extent beta-NF, increased CYP1A1 and CYP1A2 probe activities in liver up to 117- and 27- fold, respectively, and after acute regimens, that of CYP3A by approximately 1.8-fold. 13C also increased activity of CYP2B up to 100-fold. Overall hepatic metabolism of E2 and E1, which was approximately 2-fold greater at 55 than 155 days of age, was increased (approximately 2.8-fold) by 13C with 2-, 4-, 16alpha-, 6alpha-, 6beta-, and 15alpha-hydroxy (OH) comprising > or = 54, 3, 2, approximately 2, approximately 5, 7, and 2%, respectively, of E1 and E2 metabolites. Acute regimens of beta-NF increased 2- and 15alpha-OH-E2 (62 and 5% of total) from E2 and 2-, 4-, and 6alpha-OH-E1 + 6beta-OH-E1 (32, 13, and 4% of total) from E1. Mammary gland metabolized E2 to E1 and small amounts of 15alpha-, 4-, 16alpha-, 6beta-, and 6alpha-OH-E2. After the acute IC3 regimen, E2 was also converted to 2-OH-E2. 17Beta-HSD-catalyzed oxidation of E2 was favored in the liver and reduction of E1 was favored in mammary gland and tumor (= 1% of hepatic activity). An increased (approximately 2-fold) ratio of reductive to oxidative activities in malignant mammary tumors by chronic 13C regimen may stimulate tumor growth. This is the first report showing that after chronic oral regimens, the 13C-, but not beta-NF-, induced changes in CYP complement led to elevated E2 and E1 metabolism. The persistent effects of increased putative carcinogenic and estrogenic 4- and 16alpha-OH as well as 6alpha- and 6beta-OH-E2 and 6beta-OH-E1 might counteract those of the less estrogenic 2-OH metabolites, thus accounting for the lack of suppression of mammary tumorigenesis by 13C in our previous study.

Oral acetylsalicylic acid induces biliary cholesterol secretion in the rat.

Several agents can alter biliary cholesterol secretion, critical for cholesterol excretion and gallstone formation. Although salicylate effects on bile formation and gallstones have been studied, biliary lipid secretion has not been measured during oral aspirin treatment. We examined whether oral acetylsalicylic acid affects bile lipid secretion. Three groups of young rats were fed chow for 3 wk. Two of the groups then received aspirin at either 1.67 or 3.33 g/kg diet for 4 d. Serum, hepatic, and bile lipids were measured, as were enzymes of cholesterol synthesis and esterification. With oral aspirin, bile cholesterol secretion increased by 42% and hepatic cholesteryl ester content decreased by 40%. Serum cholesterol and hepatic free cholesterol did not change. To evaluate mechanisms of the cholesterol hypersecretion, hypothyroid animals fed low-fat or fish oil diets and repleted with triiodothyronine were also studied. Aspirin stimulated cholesterol secretion to a degree similar to triiodothyronine. An additive response was seen in fish oil-fed rats. Aspirin did not appear to have a primary action on 3-hydroxy-3-methylglutaryl-CoA reductase or acyl CoA:cholesterol acyltransferase activities, and had no direct effect on esterification of cholesterol by isolated hepatocytes. Aspirin may directly increase cholesterol transport into bile or have cell membrane effects which alter cholesterol transport. It remains to be determined whether the observed alterations in bile cholesterol secretion are specific to the rat or also apply to humans.

The role of gallbladder emptying in gallstone formation during diet-induced rapid weight loss.

Obese persons are at risk for cholesterol gallstones because their bile is saturated with cholesterol. The risk increases during rapid weight loss by means of certain very-low-calorie diets or gastric bypass surgery. Gallstone risk factors during rapid weight loss include increased bile cholesterol saturation index and gallbladder stasis. Obese subjects were randomized to one of two low-calorie liquid diets for rapid weight loss: a 520-kcal diet with less than 2 g fat/d, and a 900-kcal diet with 30 g fat/d (including one 10-g fat meal to stimulate maximal gallbladder emptying). Bile and blood lipids, saturation index, leukocyte 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase activity, and ultrasonographic gallbladder emptying were measured repeatedly during dietary treatment. Both diets produced comparable weight loss of 22%. Bile cholesterol saturation index increased during both diets (26%), but fell to 15% below prediet level after weight loss. Compared with subjects' maximal gallbladder emptying fraction of 66%, the 520-kcal diet provided poor gallbladder emptying (35%), whereas the 10-g fat meal of the 900-kcal diet provided maximal emptying. Gallstones developed in four of six 520-kcal subjects and none of seven 900-kcal subjects (P = .021), an unanticipated difference that resulted in premature study termination for ethical reasons. Blood lipids and HMG CoA reductase activity in mononuclear leukocytes fell at week 8 during both diets, but recovered while weight was still being lost. The findings suggest that gallstone risk during rapid weight loss may be reduced by maintenance of gallbladder emptying with a small amount of dietary fat. Ultimately, weight loss reduced bile cholesterol saturation and improved highdensity lipoprotein (HDL) levels.

Thyroid hormone is required for dietary fish oil to induce hypersecretion of biliary cholesterol in the rat.

In the rat, both fish oil diet and thyroid hormone replacement are reported to augment bile cholesterol secretion out of proportion to bile flow or secretion of other bile lipids. We sought common mechanisms for these effects and evaluated the role of phospholipid fatty acid composition in the process. Methimazole-treated hypothyroid rats were fed low-fat chow or chow supplemented with 10% corn oil or fish oil, and were studied before and after thyroid hormone treatment. Serum, hepatic, and bile lipids were measured, phospholipid fatty acid composition determined, and hepatic 3-hydroxy-3-methylglutaryl CoA reductase activity assayed. Fish oil diet stimulated cholesterol secretion into bile only after thyroid hormone was given, and this action was synergistic with that of thyroid hormone. Reduced serum cholesterol in fish oil-treated rats was associated with increased biliary cholesterol secretion and diminished hepatic cholesterol content. This suggests that augmented biliary cholesterol secretion may contribute to the fish oil-induced reduction of serum cholesterol. No definite relationship between hepatic or biliary phospholipid fatty acid composition and biliary secretion was apparent, although high bile cholesterol secretion was associated with a low percentage of hepatic and bile phospholipid linoleic acid.

Vesicles and mixed micelles in hypothyroid rat bile before and after thyroid hormone treatment: evidence for a vesicle transport system for biliary cholesterol secretion.

Hypothyroid rats show reduced secretion of biliary lipids, especially cholesterol. Secretion of biliary cholesterol is markedly augmented to levels above euthryroid beginning 12-24 h after administration of thyroid hormone. In the current studies, bile from hypothyroid and triiodothyronine-treated chronic bile-fistula rats was analyzed for vesicles and mixed micelles by metrizamide gradient ultracentrifugation. For euthryoid and hypothyroid animals, less than 12% of biliary cholesterol was in a vesicle gradient fraction. After treatment with triiodothyronine, biliary cholesterol increased markedly, and 50% of total cholesterol, 60% of excess cholesterol secreted, appeared in the vesicle fraction. Triiodothyronine stimulation of vesicle secretion resulted in cholesterol-rich vesicles (cholesterol:phospholipid ratio rose from less than 0.1 to 0.56), but no change in the distinct fatty acid composition of vesicle phospholipids. The microtubule inhibitor colchicine, given 12 h after triiodothyronine, prevented subsequent increase in cholesterol secretion in the form of vesicles. These studies, in a model that allows rapid changes in biliary lipid secretion, support the hypothesis that an important component of cholesterol and phospholipid secretion into bile involves microtubules and may involve a vesicle pathway.

Thyroid hormone differentially augments biliary sterol secretion in the rat. II. The chronic bile fistula model.

To further define thyroid hormone effects on bile acid synthesis and biliary lipid secretion, studies were done in chronic bile fistula rats. Euthyroid and methimazole-hypothyroid rats, with and without triiodothyronine (T3) injection, had total bile diversion for timed bile collections. With interrupted enterohepatic circulation, cholesterol absorption is negligible and bile acid secretion equals bile acid synthesis rate. Hypothyroid rats had diminished levels of bile acid synthesis and biliary secretion of cholesterol and phospholipid. Single dose T3 injection produced a 13-fold increase in bile cholesterol secretion and a 3-fold increase in phospholipid secretion, both initiated 12 h after T3. Bile acid synthesis increased by 50%, but the increase did not begin until 24 h after T3. Neither hypothyroidism nor T3 treatment abolished diurnal rhythms of bile acid synthesis and biliary lipid secretion. Inhibition of cholesterol synthesis with lovastatin resulted in a persistent 33% decrease in bile acid synthesis in euthyroid and hypothyroid rats, while bile cholesterol secretion only transiently decreased. Inhibition of cholesterol synthesis did not alter T3-induced bile cholesterol secretion, with a 10-fold increase seen. However, bile acid synthesis was not stimulated by T3 in the presence of lovastatin. We conclude that facilitated bile acid synthesis and biliary cholesterol secretion are early effects of T3 and may account for the hypocholesterolemia of T3. Cholesterol synthesis does not appear to be required for the T3-induced bile cholesterol secretion.

Effect of 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibition on human gut mucosa.

Mevalonic acid is an important biochemical intermediate in cholesterol synthesis and other processes involved in cell replication. 3-Hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase is the enzyme which catalyzes mevalonic acid synthesis. To determine whether a potent competitive inhibitor of this enzyme, the drug simvastatin, may have an adverse effect on enterocyte cell replication and cholesterol metabolism, small intestinal biopsies from nine hypercholesterolemic subjects were obtained before and during treatment with simvastatin as a lipid-lowering agent. Histologic review of biopsies in a blinded manner detected no change in ratio of villous length to crypt length or in mitotic index which might indicate altered cell replication. Similarly, no significant change in measured activity of HMG-CoA reductase activity was observed. In spite of the high exposure of jejunal mucosal cells to this potent competitive inhibitor of a key enzyme, no adverse effect on growth could be detected.

Effect of probucol on blood cholesterol and basal and lovastatin-induced 3-hydroxy-3-methylglutaryl coenzyme A reductase activity in mice.

The drug probucol is known to reduce levels of blood cholesterol and to have antioxidant effects on lipoproteins that may alter their metabolism. While studying probucol feeding in mice, we observed that the drug lowered total hepatic, but not gut, 3-hydroxyl-3-methylglutaryl coenzyme A (HMG CoA) reductase activities during the diurnal cycle. Hepatic fatty acyl:cholesterol acyl transferase activity and cholesterol content were not measurably affected by probucol. Probucol also abolished the induction of HMG CoA reductase activity that resulted from feeding of lovastatin, when activity was measured in microsomes washed free of drugs. These effects are consistent with previous reports that probucol increases fractional clearance of lipoprotein cholesterol by the liver. The findings raise the possibility that some patients with hypercholesterolemia may benefit from combined therapy with lovastatin plus probucol.

Lovastatin treatment inhibits sterol synthesis and induces HMG-CoA reductase activity in mononuclear leukocytes of normal subjects.

The mechanism by which competitive inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase decrease serum cholesterol is incompletely understood. The few available data in humans suggest that chronic administration of the competitive inhibitor, lovastatin, decreases serum cholesterol with little or no change in total body sterol synthesis. To further define the effect of lovastatin on cholesterol synthesis in normal subjects, we investigated the effect of a single oral dose of lovastatin and a 4-week treatment period of lovastatin on mononuclear leukocyte (ML) sterol synthesis as a reflection of total body sterol synthesis. In parallel, we measured serum lipid profiles and HMG-CoA reductase activity in ML microsomes that had been washed free of lovastatin. ML sterol synthesis did not significantly decrease (23 +/- 5%, mean +/- SEM) at 3 h after a single 40-mg dose of lovastatin. With a single oral 80-mg dose, ML sterol synthesis decreased by 57 +/- 10% (P less than 0.05) and remained low for the subsequent 6 h. With both doses, total HMG-CoA reductase enzyme activity in microsomes prepared from harvested mononuclear leukocytes was induced 4.8-fold (P less than 0.01) over baseline values. Both the 20-mg bid dose and the 40-mg bid dose of lovastatin administered for a 4-week period decreased serum cholesterol by 25-34%. Lovastatin at 20 mg bid decreased ML sterol synthesis by 23 +/- 6% (P less than 0.02) and increased ML HMG-CoA reductase 3.8 times (P less than 0.001) the baseline values. Twenty four hours after stopping lovastatin, ML sterol synthesis and HMG-CoA reductase enzyme activity had returned to the baseline values. The higher dose of lovastatin (40 mg bid) decreased ML sterol synthesis by 16 +/- 3% (P less than 0.05) and induced HMG-CoA reductase to 53.7 times (P less than 0.01) the baseline value at 4 weeks. Stopping this higher dose effected a rebound in ML sterol synthesis to 140 +/- 11% of baseline (P less than 0.01), while HMG-CoA reductase remained 12.5 times baseline (P less than 0.01) over the next 3 days. No rebound in serum cholesterol was observed. From these data we conclude that in normal subjects lovastatin lowers serum cholesterol with only a modest effect on sterol synthesis. The effect of lovastatin on sterol synthesis in mononuclear leukocytes is tempered by an induction of HMG-CoA reductase enzyme quantity, balancing the enzyme inhibition by lovastatin.(ABSTRACT TRUNCATED AT 400 WORDS)

Intestinal HMG-CoA reductase activity is low in hypercholesterolemic patients and is further decreased with lovastatin therapy.

Significant cholesterol synthesis occurs in gut mucosa of animals and humans. However, the role of gut synthesis in hypercholesterolemia and the effect of drugs on this function have not been defined. We obtained jejunal biopsies and bile samples from 21 Type II hypercholesterolemic subjects (mean serum cholesterol = 331 mg/dl) on a low fat diet after an over-night fast. Whole gut mucosal homogenate was assayed for activity of 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase, the rate-determining enzyme of cholesterol synthesis. Mean reductase activity (pmol/mg per min) was 5.5 +/- 1.0 (n = 21) in hypercholesterolemic subjects versus 11.3 +/- 1.0 in 52 normal subjects (P less than 0.01). This is consistent with the hypothesis that the primary defect in these patients is not excessive cholesterol synthesis but decreased low density lipoprotein (LDL) clearance. It implies that high LDL levels down-regulate gut reductase activity. After treatment of 7 patients with lovastatin (40-80 mg/day for at least 6-13 weeks), gut reductase activity decreased from 7.7 +/- 2.6 to 3.6 +/- 0.5 (P less than 0.05), in biopsies obtained 12 hr after the last drug dose. Inhibition of reductase activity by this drug was detected 12 hr after a dose, and the enzyme was not measurably induced during 6-13 weeks of therapy. In keeping with the decrease in serum cholesterol (332----239 mg/dl) and mucosal reductase activity during lovastatin therapy, mean gallbladder bile cholesterol saturation index also decreased (1.045 +/- 0.112 vs. 0.883 +/- 0.109, n = 7).(ABSTRACT TRUNCATED AT 250 WORDS)

Abnormal cholesterol metabolism in renal clear cell carcinoma.

The clear cell form of renal cell carcinoma is known to derive its histologic appearance from accumulations of glycogen and lipid. We have found that the most consistently stored lipid form is cholesteryl ester. Clear cell cancer tissue contained 8-fold more total cholesterol and 35-fold more esterified cholesterol than found in normal kidney. Cholesteryl ester appeared to be formed intracellularly since it was not membrane-bound and since oleate was the predominant form, as opposed to linoleate in lipoprotein cholesteryl esters. The cholesterol in clear cell tumors did not appear to be a result of excessive synthesis from acetate since HMG-CoA reductase (EC 1.1.1.34) activity was lower in cancer tissue than in normal kidney (2.9 +/- 0.8 vs. 7.2 +/- 1.2 pmol/mg of protein per min). In contrast, intracellular activity of fatty acyl-coenzyme A:cholesterol acyl transferase (ACAT, EC 2.3.1.26) was higher in tumor tissue than in normal kidney (2405 +/- 546 vs. 1326 +/- 301 pmol/mg of protein per 20 min) while cytosolic cholesteryl ester hydrolase activity appeared normal. Cholesteryl ester storage in clear cell renal cancer may be a result of a primary abnormality in ACAT activity or it may be a result of reduced release of free cholesterol (relative to cell content) with a secondary elevation in ACAT activity.

Transport of circulating serum cholesterol by human renal cell carcinoma.

Clear cell renal cancer contains a large quantity of cholesterol ester (300-mg./gm. protein). To determine whether abnormalities in cholesterol transport could account for this sterol accumulation, the uptake, release, and imaging capabilities of intravenously injected 131I-6-iodomethyl-29-norcholesterol, a cholesterol analogue, were studied preoperatively in five patients with clear cell renal cancer. At surgery, samples of the liver, tumor, adrenal, and non-tumor kidney were obtained for analysis. 131I-sterol uptake by the tumor, when normalized for cholesterol content, was less than for adrenal, liver or kidney. In contrast, release of preloaded 131I-sterol from the human tumors was consistently slower than for normal kidney. The reduced release of free cholesterol from renal cancer cells may, in part, be responsible for the accumulation of cholesterol in human renal cancer.

Diurnal rhythm of HMG CoA reductase activity in canine intestine is independent of luminal contents.

Activity of the enzyme 3-hydroxy-3-methylglutaryl coenzyme A reductase (EC 1.1.1.34) measured in isolated segments of canine intestinal mucosa showed a distinct diurnal rhythm. Total activity changed over a twofold range with a peak occurring during midday, shortly after feeding. Since the isolated segments had no contact with luminal contents, the rhythm was not directly related to food components or bile salts. Humoral or neural influences must mediate the rhythm. The diurnal rhythm persisted for at least 3-5 mo, but was lost by 10 mo following formation of the isolated segment, possibly because of mucosal involution.

Abnormal intestinal permeability to sugars in diabetes mellitus.

A test of intestinal mucosal function which utilizes the differential permeability of L-rhamnose and lactulose has been reported to be helpful in the diagnosis of gluten-sensitive enteropathy. We have applied this test to 48 male subjects with diabetes mellitus to evaluate its usefulness as a screening test in diabetic patients and to further study sugar absorption in these individuals. Total urinary lactulose excretion in the 13 healthy control subjects was 54.5 +/- 8.5 mg/5 h, while excretion by diabetic patients was increased at 116.1 +/- 15.7 mg/5 h (p less than 0.01). Similarly, total L-rhamnose excretion by diabetic patients was significantly higher (139.7 +/- 14.3 mg/5 h vs 84.3 +/- 18.4 mg/5 h, p less than 0.05). The ratio of percent urinary excretion for lactulose/L-rhamnose (L/R ratio) for diabetic patients (0.197 +/- 0.024) was not different from the control subjects (0.151 +/- 0.2). Nine out of 48 diabetic patients studied had lactulose/L-rhamnose ratios higher than the mean plus two standard deviations of the control group, which might lead to the diagnosis of small bowel mucosal disease. Although we may have been detecting subclinical mucosal disease or gluten sensitive enteropathy in a subgroup, it appears that this test of intestinal mucosal function should be interpreted with caution in diabetic patients.

3-Hydroxy-3-methylglutaryl coenzyme A reductase activity in the human gastrointestinal tract.

Activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase (EC 1.1.1.34) was measured in intestinal mucosa of the human gastrointestinal tract. Activity was highest in gastric mucosa (18.2 pmol per mg per min) and there was a constant low level in the small bowel and colon (approximately 10 pmol per mg per min). Phosphorylation/dephosphorylation modulation of intestinal reductase activity was demonstrated in normal mucosa. Expressed jejunal reductase activity was significantly higher in celiac sprue mucosa and mucosa from defunctionalized intestine of jejunoileal bypass patients. Enzyme activity also increased during 24-hr mucosal organ culture in the absence of exogenous cholesterol. Addition to the culture medium of pure cholesterol or 25-hydroxycholesterol dissolved in a small volume of ethanol suppressed the culture-induced increase to 86 +/- 3% and 69 +/- 5% of paired controls, respectively. This evidence suggests that a moderate degree of feedback regulation of intestinal cholesterol synthesis by luminal sterol occurs in man. Mucosal HMG-CoA reductase activity was also measured in patients with hyperlipoproteinemia. Patients with either predominant hypercholesterolemia or predominant hypertriglyceridemia lipid profiles had "normal" expressed reductase activity, but feedback regulation by free cholesterol could not be demonstrated in either group under these conditions.

Hepatic and intestinal 3-hydroxy-3-methylglutaryl coenzyme A reductase activity in genetically diabetic mice.

Cholesterol synthesis rate, as determined by 3-hydroxy-3-methylglutaryl coenzyme A reductase activity, is characterized in the major organs of genetically diabetic mice. Both C57BL/Ks db+/db+ and C57BL/6 ob+/ob+ mice are hyperinsulinemic and insulin-resistant. These animals demonstrate loss of the circadian rhythm of hepatic reductase activity and a tendency for increased intestinal activity. As a result, proportionally more endogenous cholesterol synthesis occurs in intestinal mucosa than liver in genetically diabetic animals. Thus, the alterations in activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase which are observed in animal models of diabetes are the result of diminished insulin effect rather than insulin level.

Effect of exorphins on gastrointestinal function, hormonal release, and appetite.

Peptic digestion of gluten results in the production of substances having opiatelike activity in bio- and receptor assays. These substances have been termed exorphins. In this study, we determined the effect of gluten, hydrolyzed gluten, and hydrolyzed gluten plus the opiate blocker naloxone on a variety of hormonal parameters, gastrointestinal transit time, small bowel mucosal integrity, and satiety. Hydrolyzed gluten prolonged intestinal transit time, and this effect was reversed by concomitant administration of naloxone. Hydrolyzed gluten also produced a naloxone-reversible increase in plasma somatostatinlike activity, which may have been responsible for the delayed transit time. No effects of the "exorphins" could be demonstrated on serum gastrin, cortisol, carbohydrate metabolism, or small bowel mucosal integrity. Although a number of studies have suggested a role for endogenous opiates in appetite regulation, we could not demonstrate any effect of "exorphins" on the amount of calories ingested nor on the perception of satiety. This study defines a potential role for small opiatelike peptides in foodstuffs in the regulation of intestinal function.

The effect of severe zinc deficiency on activity of intestinal disaccharidases and 3-hydroxy-3-methylglutaryl coenzyme A reductase in the rat.

Specific activities of five enzymes were measured in intestinal mucosa of zinc-deficient rats and compared to activities in appropriate zinc-sufficient controls. Three disaccharidases were found to be significantly reduced in zinc deficiency. Alkaline phosphatase, a zinc metalloenzyme, also showed reduced activity. Activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase, the rate-limiting enzyme of cholesterol synthesis, was elevated. It is possible that impaired carbohydrate digestion (reflected in disaccharidase activity) and even defective lipid absorption (reflected in reductase activity) may contribute to the poor nutrition and diarrhea of zinc deficiency.

In vivo regulation of canine intestinal 3-hydroxy-3-methylglutaryl coenzyme A reductase by cholesterol, lipoprotein, and fatty acids.

Thirty-Vella-isolated ileal segments in dogs were used to study the regulation of intestinal cholesterol synthesis. This excluded fistula enabled independent in vivo manipulation of luminal and vascular influences on mucosal cells. Segments were studied repeatedly and each animal served as its own control. Cholesterol synthesis rate was assessed by measuring mucosal activity of the rate-limiting enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase. Luminal cholesterol and 25-hydroxycholesterol were shown to reduce HMG CoA reductase activity to 64 +/- 7% and 42 +/- 4%, respectively, of control within 4 hr. Reductase activity in the excluded segment also responded to alterations in serum cholesterol produced by cholesterol or cholestyramine feeding. Similarly, in vitro studies showed that lipoprotein cholesterol inhibited HMG CoA reductase in mucosa from the excluded segment but not in mucosa from intact bowel. In contrast to sterols, fatty acids stimulated HMG CoA reductase activity by luminal contact. These findings suggest that the cholesterol needs of canine intestinal epithelial cells are acutely balanced by absorption and synthesis of cholesterol. Mucosal cells may also utilize lipoprotein cholesterol under certain conditions, perhaps via low density lipoprotein receptors. Fatty acid absorption stimulated cholesterol synthesis in the absence of luminal cholesterol, perhaps to facilitate chylomicron formation.