| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 47, 421-430, February 2006
Copyright © 2006 by American Society for Biochemistry and Molecular Biology
* Division of Clinical Chemistry, Department of Laboratory Medicine, Karolinska Institutet, Karolinska University Hospital, Huddinge, Sweden
Division of Gastroenterology and Hepatology, Karolinska Institutet, Karolinska University Hospital, Huddinge, Sweden
Metabolism Unit, Center for Metabolism and Endocrinology, Department of Medicine, Karolinska Institutet, Karolinska University Hospital, Huddinge, Sweden
Published, JLR Papers in Press, November 1, 2005.
1 To whom correspondence should be addressed. e-mail: gosta.eggertsen{at}karolinska.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Supplementary key words bile acids • farnesoid X receptor • liver X receptor • cholesterol 7
-hydroxylase • sterol 12-hydroxylase
Abbreviations: CA, cholic acid; CDCA, chenodeoxycholic acid; CSI, cholesterol saturation index; DCA, deoxycholic acid; FXR, farnesoid X receptor; LXR, liver X receptor; UDCA, ursodeoxycholic acid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The major bile acids in mammals are cholic acid (CA), chenodeoxycholic acid (CDCA), and muricholic acids. The latter are more hydrophilic, and are most abundant in rodents. In humans, CA and CDCA are the dominating bile acids, together with the secondary bile acids deoxycholic acid (DCA) and lithocholic acid, which originate in microbiological actions in the large intestine (2). CA is a tricarboxylic acid, mainly produced via the neutral pathway. The enzyme sterol 12-hydroxylase (CYP8B1) is necessary for its synthesis; otherwise CDCA is produced. In the intestine, CA is a key factor, together with phospholipids, in the formation of micelles to achieve an efficient absorption of cholesterol. In the mouse, it functions as a specific ligand for the nuclear receptor farnesoid X receptor (FXR), which mediates the downregulation of CYP7A1 and CYP8B1 (2). Activated FXR is also reported to modify the expression of bile acid transporters in the liver, which are sodium-taurocholate cotransport polypeptide for bile acid uptake, bile salts export pump (BSEP) for bile acid secretion, and intestine ileal bile acid binding protein (3).
Over the last few years, research efforts in the field of cholesterol and bile acid metabolism have utilized an increasing number of genetically modified mice. A general approach in many studies is to feed high-fat, high-cholesterol diets to such experimental mouse models. Generally, normal mice do not develop atherosclerosis when put on such atherogenic diets, even if their plasma cholesterol levels could vary depending on the strain (4). However, genetically modified strains, for example, LDL receptor-deficient mice, develop advanced arteriosclerotic lesions when fed diets containing cholesterol and CA (5). Such a combined diet has also been utilized to induce formation of gallstones in susceptible mouse strains. Even if CA is a key element in these pathologic processes, its exact mode of action is not well defined.
To better understand the biological roles for sterol 12-hydroxylase/CA, we created an experimental mouse model genetically deficient for CYP8B1; these mice do not synthesize CA (2). They have an increased expression of CYP7A1, which causes elevated bile acid synthesis and also biochemical changes, suggesting that the FXR effector system is downregulated. In addition, the animals show a reduced absorption of dietary cholesterol and do not increase their storage of liver cholesterol when fed cholesterol, compared with normal mice (6). The present study characterizes the effects of the administration of cholesterol-enriched and cholesterol-plus-CA diets to CYP8B1–/– mice, with a focus on the cholesterol and bile acid metabolism. Our results indicate that CA per se is an important factor in the regulation of the endogenous cholesterol balance, affecting both degradation and synthesis of cholesterol, as well as its excretion and absorption.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
99%), cholesterol (>99%), and the liver X receptor (LXR) agonist TO-901317 were purchased from Sigma-Aldrich Corp., St. Louis, MO. The FXR agonist GW4064 was kindly supplied by F. Hoffman-La Roche Ltd, Basel, Switzerland.
Animals and procedures
Mice deficient for the sterol 12-hydroxylase by targeted disruption of the gene have previously been characterized (2). The corresponding wild-type animals were obtained by crossbreeding of heterozygous individuals. Generally, males aged 2–6 months were used. Mice of both genotypes were housed at 22–24°C at a light cycle of 8 AM to 8 PM in groups of four to six animals and kept on the following diets for 7 days: i) standard chow (0.025% cholesterol, w/w); ii) 0.5% cholesterol (w/w); iii) 0.5% cholesterol plus 0.25% sodium cholate. Controls were kept on standard chow. All diets contained 10% peanut oil (w/w). Both the LXR agonist TO-901317 and the FXR agonist GW4064 were given in gavage for 5 days in doses of 10 mg/kg/day and 50 mg/kg/day, respectively, to groups of CYP8B1+/+ and –/– mice. To control animals, 1% methylcellulose was given in the same way. The last dose was delivered 2 h before the animals were euthanized. Water was available ad libitum.
Animals were euthanized by cervical dislocation following CO2 anesthesia, and the liver and gallbladder were quickly washed in physiological saline and immediately frozen in liquid nitrogen. Blood was collected by heart puncture. Animals used for sampling gallbladder bile were fasted for 12 h, and light microscopy was performed on 5 µl aliquots of bile. Serum total cholesterol was determined by the standard enzymatic procedure. Lipoprotein cholesterol profiles were obtained by fast-protein liquid chromatography. Liver total and free cholesterol were determined by the method described by Bjorkhem (7). To obtain a relative index of hepatic de novo cholesterol synthesis, the concentration of lathosterol was assayed by isotope dilution-mass spectrometry as reported (8), and the ratio of lathosterol to total cholesterol was calculated.
Determination of intestinal cholesterol absorption
Cholesterol absorption was assayed, after feeding the specified diets for 6 days, by the fecal dual-isotope method, essentially as reported by Schwarz et al. (9), except that feces were collected for 24 h after administering the radioactive gavage. Percent cholesterol reabsorbed was estimated by the standard equation. [5,6-3H]ß-sitostanol was obtained from American Radiolabel Chemicals, Inc., St. Louis, MO and [4-12C]cholesterol from Amersham Biosciences, Uppsala, Sweden.
Quantitative real-time PCR
Total RNA was isolated from 75–100 mg of liver tissue using the Quick Prep Total RNA Extraction Kit (Amersham Biosciences). Oligo-dT-primed cDNA synthesis was carried out on 3 µg of liver total RNA using Superscript III reverse transcriptase (Invitrogen Life Technologies; Carlsbad, CA). Real-time PCR was carried out on an ABI Prism 7000 Sequence Detection System (Applied Biosystems; Foster City, CA) using either TaqMan probes (labeled with the fluorochromes FAM or VIC and dark quencher or TAMRA) or SYBR Green. The following genes were analyzed: mouse HMG-CoA reductase, CYP7A1, CYP8B1, CYP27, SHP, ABCA1, ABCG5, ABCG8, BSEP, MDR2, apoA-1, LDLR, SREBP1C, and SREBP2. As an internal standard, mouse hypoxanthine phosphoribosyl transferase was utilized. Control of the PCR efficiency was done by plotting standard curves using different dilutions of cDNA. The PCR products were sequenced to confirm that they represented the correct fragments.
Determination of the output of bile acids and neutral sterols in feces
Fecal samples were collected from mice kept in individual cages for 24 h. Analysis of neutral sterols and bile acids by gas-liquid chromatography was performed essentially as reported by Miettinen, Ahrens, and Grundy (10), Miettinen (11), and Grundy, Ahrens, and Miettinen (12).
Analysis of biliary lipids in gallbladder bile
Gallbladder bile samples were examined on slides for cholesterol crystals by polarizing light microscopy. The composition of bile acids in gallbladder bile was analyzed by gas-liquid chromatography, as reported (13). Total bile acid concentration was determined by an enzymatic method (14), as was cholesterol (15). Phospholipids were assayed by the method described by Rouser, Fleischer, and Yamamoto (16). The relative concentrations of biliary lipids were expressed as molar percentages of the total biliary lipids. The cholesterol saturation was calculated according to Carey (17).
Statistical analysis
Data are presented as means ± SEM. Statistical analysis was performed with STATISTICA software. The significance of differences was tested by two-way ANOVA, followed by planned comparisons. To stabilize the variances, data were logarithmically transformed when a correlation was found (18).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
Alterations in intestinal cholesterol absorption and excretion To compensate for possible variations in food intake, fecal cholesterol was expressed as the ratio of neutral sterol levels to plant sterol levels in feces (Fig. 3A). Fecal cholesterol excretion increased strongly in animals fed the cholesterol diet, with higher levels in the CYP8B1–/– animals. As expected, the percentage of absorption of cholesterol was significantly lower in the CYP8B1–/– mice than in the CYP8B1+/+ mice (42.7% and 68.7%, respectively) and dropped in both groups after cholesterol feeding (14.8% and 31.4%, respectively) (Fig. 3B).
|
|
|
Alterations in the metabolism of hepatic cholesterol Hepatic cholesterol levels were strongly increased in both groups, with the highest mean level observed in the CYP8B1+/+ animals (11.8 µg/mg, compared with 9.2 µg/mg in the CYP8B1–/– mice), which exceeded the value found in CYP8B1+/+ animals receiving cholesterol only (7.4 µg/mg) (Fig. 1A). The fraction of esterified cholesterol was similar in CYP8B1+/+ and CYP8B1–/– mice (74.9% and 70.8%, respectively) and close to the value found for the cholesterol-fed CYP8B1+/+ animals (74.6%). The liver cholesterol synthesis was further depressed in both groups, as estimated by the lathosterol assay (Fig. 1B). Although decreased, the mRNA quantities for HMG-CoA reductase were markedly higher in the CYP8B1–/– animals than in the CYP8B1+/+ mice, in which no changes were recorded, compared with CYP8B1+/+ mice fed cholesterol (Fig. 1C). Although the cholesterol synthesis was depressed to the same level in both CYP8B1+/+ and –/– mice, the HMG-CoA reductase mRNA levels were markedly higher in the latter, possibly related to fact that the CYP8B1+/+ mice had higher hepatic cholesterol levels than the CYP8B1–/– animals. The SREBP2 mRNA levels were nevertheless nearly equal in both groups, and on levels similar to those of CYP8B1+/+ animals receiving cholesterol-enriched diet only (Fig. 1D).
Alterations in bile acid synthesis and composition of bile
The CYP7A1 mRNA levels of the CYP8B1+/+ mice did not differ from those of cholesterol-fed CYP8B1+/+ animals (Fig. 2B). In CYP8B1–/– mice, the levels tended to be higher than in control mice, but much lower than those seen in the cholesterol-fed CYP8B1–/– mice. In both groups, no significant changes of the CYP27 mRNA were observed, although the levels of SHP mRNA showed a tendency to increase (data not shown). The levels of BSEP and MDR2 did not change significantly, compared with the controls. Interestingly, ABCG5 and ABCG8 mRNA increased 3- to 5-fold in both groups of animals (Fig. 2C, D). In the bile, both the CYP8B1+/+ and CYP8B1–/– mice showed similar values for the percentage of total bile acids. However, the fraction of hydrophobic bile acids (cholic acid and deoxycholic acid) was higher in the CYP8B1+/+ animals than in the CYP8B1–/– mice, whereas the fraction of more-hydrophilic bile acids (ß-muricholic, -muricholic, and UDCA) was lower (Table 1). No essential differences were found between the two groups in regard to cholesterol, the CSI%, and phospholipids (Table 2), although the first two analytes showed much higher values than those found in the cholesterol-fed CYP8B1–/– mice. Only the concentration of total lipids differed, being lower in the CYP8B1–/–mice. In the gallbladder of the CYP8B1+/+ animals, plenty of cholesterol crystals were observed, while the bile of the CYP8B1–/– mice was clear and devoid of solid material.
Alterations in intestinal cholesterol absorption and excretion Fecal cholesterol excretion was similar in both groups, but lower than that found in mice fed 0.5% cholesterol, indicating the enhancing effect of CA on the cholesterol absorption (Fig. 3A). The mean percentage of cholesterol resorption in the CYP8B1–/– mice was 54%, whereas that in the CYP8B1+/+ mice was 39% (Fig. 3B).
Alterations in serum lipids and lipoproteins Plasma levels of total cholesterol did not change in animals fed the combination of cholesterol and CA (Fig. 4A); the VLDL- and LDL-cholesterol levels increased to the same extent in both groups, whereas the HDL-cholesterol levels were similar to those of the control animals (Fig. 4B). The mRNA levels for ABCA1 (Fig. 5A) and LDLR (data not shown) did not differ between the groups and the values found in animals on normal chow. The mRNA levels of apoA-1 did not show any significant differences, compared with the other mouse groups (Fig. 5B).
Effects of TO-901317 and GW4064 treatment
To better define whether the effects induced by the different diet regimes (cholesterol and cholesterol/CA) were mediated via the nuclear receptors LXR and FXR, mice were treated with specific agonists to each of these receptors.
The results showed that the expression of CYP7A1 was increased significantly by the LXR agonist TO-901317 in both CYP8B1+/+ and CYP8B1–/– mice, compared with the controls, although the effect was much more pronounced in the latter group (Fig. 6A). Elevated mRNA levels were also found for SREBP1C (Fig. 6B) and ABCG5/G8 (Fig. 6C, D), although no effects were observed for SHP, BSEP, and MDR2 (data not shown). FXR agonist treatment significantly decreased the levels of CYP7A1 mRNA in both groups (Fig. 6A), but no changes were recorded for the mRNA levels of ABCG5/G8 (Fig. 6C, D), SREBP1C (Fig. 6B) and MDR2 (data not shown). SHP showed a tendency to increase in the CYP8B1–/– mice after the GW4064 treatment, whereas no differences were found in the CYP8B1+/+ animals (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylase when fed a cholesterol-enriched diet. The increased conversion of cholesterol into bile acids appears to be one reason these mice accumulate only minor quantities of liver cholesterol, compared with CYP8B1+/+ animals. This process appears to occur via the classical pathway as CYP27 expression did not undergo any significant changes. Another contributing factor in the CYP8B1–/– mice may be the reduced influx of cholesterol from the intestine (Fig. 3B).
The rate-limiting enzyme in the bile acid synthesis, CYP7A1, is subjected to a sophisticated regulation in which several nuclear receptors (LXR, FXR, PXR, FTF, and HNF4) (1) participate. The higher expression level of CYP7A1 in our CYP8B1–/– mice is probably due to elimination of the repressive effects of FXR, for which, in the mouse, CA represents the most important ligand (2). Several other mechanisms have also been described that downregulate CYP7A1 expression, involving PXR, the c-Jun N-terminal kinase (1), and insulin (19), all of which are independent of SHP. It seems likely, however, that under normal conditions, FXR is the most important repressor of CYP7A1 expression. That the upregulation obtained by cholesterol is mediated by LXR, presumably activated by oxysterols, is supported by our finding that treatment with an LXR agonist induces a pattern similar to the reaction following cholesterol feeding. The difference in CYP7A1 mRNA levels between cholesterol-fed CYP8B1+/+ and CYP8B1–/– mice illustrates the repressive power of the FXR-mediated mechanism, with 4-fold higher levels of CYP7A1 mRNA occurring in the CYP8B1–/– mice, although analysis of fecal bile acids indicated a doubling of the production rate. The basal transcription level of CYP7A1 seems to depend more on FXR than on LXR, because there were no alterations in the mRNA levels for CYP7A1 in LXR-deficient mice, as opposed to those seen in FXR-null animals (20, 21). The expanded bile acid pool in the CYP8B1–/– mice does not repress the bile acid synthesis, because its composition consists mainly of muricholic acids; these are more hydrophilic than CA and are poor activators of FXR. In addition, they do not promote cholesterol absorption (22). Addition of the FXR agonist GW4064 effectively depressed the transcription of CYP7A1 in both groups of mice to a greater extent than did the combined diet of cholesterol/CA. In the latter case, the CYP7A1 mRNA levels were slightly higher than those observed in the mice fed only chow, representing mixed and counteracting effects by LXR and FXR, whereby the LXR activation was mediated by the increased availability of cholesterol. The FXR effect nevertheless dominates, confirming previous reports (21, 23, 24). These results reflect the dual effect of CA in mice, as a stimulator of cholesterol absorption and as an FXR agonist. SHP mRNA levels were not significantly increased in any of our experimental regimes, but have been reported to rise after treatment using GW4064 at higher doses (100 mg/kg/day) (25) or CA (2).
In normal mice fed cholesterol, the endogenous cholesterol synthesis will be downregulated by increased hepatic cholesterol levels (26). In cholesterol-fed CYP8B1–/– mice, a higher rate of cholesterol synthesis is required to compensate for the amounts converted to bile acids, but even then, the hepatic cholesterol levels are much lower than in the CYP8B1+/+ mice. It is likely that fine-tuning of CYP7A1 expression by a careful interplay between FXR and LXR stimuli is aimed at preserving a correct cholesterol economy, a delicate balance that is impaired in CYP8B1–/– mice.
According to current concepts, the transporters ABCG5 and ABCG8 represent the main mechanism for transfer of cholesterol into bile (27, 28). Normal mice fed 2% cholesterol for 1 week increased their hepatic and intestinal ABCG5/G8 expression (primarily considered to be mediated by LXR) (29, 30) 2- to 3-fold. However, ABCG5/G8 expression nevertheless seems less responsive to cholesterol treatment than is that of CYP7A1, inasmuch as no significant changes were observed for the mRNA levels in the two groups of animals fed 0.5% cholesterol, although the cholesterol content in the bile doubled in the CYP8B1+/+ mice. Interestingly, the LXR agonist treatment induced a significant increase of the ABCG5/G8 mRNA, whereas GW4064 did not have any significant effects, compared with the control animals. The prominent increase of ABCG5/G8 mRNA in animals fed the cholesterol/CA diet could include effects mediated by LXR (triggered via the increased hepatic cholesterol) and FXR. CA has previously been reported to enhance the transcription of ABCG5/G8, possibly indirectly via FXR (29, 30). The underlying mechanisms for the regulation of ABCG5/G8 by CA/FXR or LXR are not known, and no specific responsive elements have so far been identified in the introns or in the short intergenic region between the two genes (31). However, FXR/CA may not be involved in maintaining the basal activity of ABCG5/G8, inasmuch as the CYP8B1–/– mice did not display any significant differences in their ABCG5/G8 mRNA levels, compared with the CYP8B1+/+ animals, when fed normal chow. In SHP knock-out mice, slightly increased levels of ABCG5/G8 mRNA were found, but this was ascribed to an enlarged pool of CA (32).
Feeding the combined diet of cholesterol and CA "normalized" most of the effects observed in the CYP8B1–/– mice. Due to improved absorption, fecal cholesterol excretion was reduced to similar levels in both CYP8B1–/– and CYP8B1+/+ mice. In the CYP8B1–/– mice, cholesterol accumulated in the liver, and synthesis of cholesterol and bile acids were depressed. Increased quantities of cholesterol shuttled into the bile elevated the CSI % in both animal groups, but cholesterol crystals were only observed in the +/+ mice. However, bile from these animals had higher concentrations of total lipids and a more hydrophobic bile acid composition, two factors of importance for rapid cholesterol crystallization (33). Recently, it was shown that gallstone formation in mice could be prevented by GW4064 treatment (25), an effect assumed to depend on upregulation of the hepatic transporters BSEP and MDR2, which would increase the flow of bile acids and phospholipids, respectively, into the bile, thus improving the solubility of cholesterol. However, no significant differences in the mRNA levels of these transporters were observed by us in the CYP8B1–/– and CYP8B1+/+ mice fed the CA/cholesterol diet or treated with GW4064, nor did the concentration of phospholipids in the bile differ significantly. With regard to GW4064, the discrepancies might be dependent on the lower doses used in our experiments.
In CYP8B1–/– mice treated with a cholesterol diet, no significant changes were observed in total plasma cholesterol levels, but HDL-cholesterol was increased in parallel with the mRNA levels of the hepatic transporter ABCA1. Absence of CA thus appears to facilitate production of HDL, thereby enhancing efflux of cholesterol from the liver. The human ABCA1 gene contains LXR sites (34), but whether CA directly affects the ABCA1 expression in mice is an open question. Even if supplementation of the diet with cholate reduced the plasma apoA-1 by downregulating its transcription (35), we did not find any significant differences in the mRNA levels of apoA-1 between the CYP8B1+/+ and –/– mice kept on normal chow (Fig. 5B), indicating that the basic transcriptional rate of this gene is not influenced by physiological levels of CA. The HDL-cholesterol levels in cholesterol-fed CYP8B1–/– mice might then depend on the ABCA1 expression, as has been shown in mice overexpressing ABCA1 (36), thereby proposing a novel role for CYP8B1 in the plasma lipoprotein metabolism in mice.
Our results illustrate the key role for CYP8B1 and its effector molecule, CA, in cholesterol homeostasis in mouse. In parallel with the stimulatory effect on cholesterol absorption, CA markedly reduces degradation of cholesterol into bile acids and indirectly affects the rate of cholesterol synthesis and the liver cholesterol content. An optimal cholesterol balance appears to be tightly linked to the CA pool size, which may be one reason why CYP8B1 is subjected to such sophisticated regulation. On the other hand, shortage of CA could protect against effects from excess cholesterol intake. In humans, the situation is somewhat different, inasmuch as CDCA is not converted to muricholic acids, and furthermore, appears to be the most efficient ligand for human FXR (37, 38).
|
ACKNOWLEDGMENTS |
|---|
Manuscript received October 6, 2005 and in revised form November 1, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) regulates bile acid biosynthesis. J. Biol. Chem. 275: 28947–28953. and ß. J. Biol. Chem. 277: 18793–18800.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  W. O. Ward, D. A. Delker, S. D. Hester, S.-F. Thai, D. C. Wolf, J. W. Allen, and S. Nesnow Transcriptional Profiles in Liver from Mice Treated with Hepatotumorigenic and Nonhepatotumorigenic Triazole Conazole Fungicides: Propiconazole, Triadimefon, and Myclobutanil Toxicol Pathol, December 1, 2006; 34(7): 863 - 878. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Journal of Biological Chemistry |
| Molecular and Cellular Proteomics | ASBMB Today |
