Regulation of Hepatic Low-Density Lipoprotein Receptor, 3-Hydroxy-3 ...

0013-7227/02/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 87(9):4307– 4313 Copyright © 2002 by The Endocrine Society doi: 10.1210/jc.2002-012041

Regulation of Hepatic Low-Density Lipoprotein Receptor, 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase, and Cholesterol 7␣-Hydroxylase mRNAs in Human Liver ´ R, STAFFAN SAHLIN, MATS RUDLING, BO ANGELIN, LARS STÅHLE, EVA REIHNE ¨ HANS OLIVECRONA, INGEMAR BJORKHEM, AND CURT EINARSSON Centers for Metabolism and Endocrinology (M.R., B.A., E.R.), Gastroenterology (H.O., C.E.), and Nutrition and Toxicology (M.R., B.A., H.O.), Departments of Medicine (M.R., B.A., C.E.), Clinical Pharmacology (L.S.), Surgery (E.R., H.O.), and Clinical Chemistry (I.B.), Karolinska Institute at Huddinge University Hospital, S-141 86 Stockholm, Sweden; and Department of Surgery (S.S.), Danderyd Hospital, S-182 88 Danderyd, Sweden To characterize the coordinate regulation of cholesterol metabolism in human liver, we simultaneously quantified mRNA levels of cholesterol 7␣-hydroxylase (CYP7A1), 3-hydroxy-3methylglutaryl coenzyme A reductase (HMGR), and lowdensity lipoprotein receptors (LDLRs) in liver biopsies from 76 patients undergoing cholecystectomy. The three transcript levels were not different between untreated gallstone and gallstone-free patients and not significantly altered by 10-d exclusion of dietary cholesterol. Treatment with chenodeoxycholic acid suppressed CYP7A1 and to a lesser extent HMGR mRNA levels. Cholestyramine treatment increased CYP7A1, but also HMGR and LDLR mRNA, and statins only increased HMGR mRNA. Resin ⴙ statin treatment increased all mRNA

C

HOLESTEROL IS A VITAL molecule, being a structural component of cell membranes and a precursor for steroid hormones and bile acids (1). Cells maintain an optimal cholesterol level through an intricate end-product regulation. The control of the cholesterol level in plasma is important because excess cholesterol predisposes to atherosclerosis (1). The liver is a key element in the control of plasma cholesterol, which is largely determined by the rate of removal of low-density lipoproteins (LDLs) from the circulation, regulated by hepatic LDL receptors (LDLRs) (1, 2). The liver is also an important site for cholesterol synthesis, controlled by the microsomal enzyme 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGR). Finally, the liver is the only site in which substantial amounts of cholesterol are removed from the body by excretion into bile either as free cholesterol or after conversion into bile acids, a process regulated by the microsomal enzyme cholesterol 7␣-hydroxylase (CYP7A1). Reduction of plasma LDL cholesterol prevents the development of coronary heart disease (3). An improved understanding of the physiological regulation of hepatic cholesterol metabolism is thus important for our possibilities to better intervene in this disease process. However, cholesterol

Abbreviations: CDCA, Chenodeoxycholic acid; HMGR, 3-hydroxy3-methylglutaryl coenzyme A reductase; LDL, low-density lipoprotein; LDLR, LDL receptor; SREBP, steroid regulatory element-binding protein; UDCA, ursodeoxycholic acid.

species. In untreated patients, the mRNA levels of HMGR and LDLR were more strongly correlated (r ⴝ ⴙ0.60) than those of CYP7A1 and HMGR (r ⴝ ⴙ0.49) or CYP7A1 and LDLR (r ⴝ ⴙ0.21). In the treated patients, in whom bile acid synthesis was suppressed or stimulated, mRNA levels of CYP7A1 and HMGR (r ⴝ ⴙ0.84) as well as CYP7A1 and LDLR (r ⴝ ⴙ0.62) were more strongly correlated than those of HMGR and LDLR (r ⴝ ⴙ0.59). The coordinate control of HMGR and LDLR mRNA levels reflects their common regulation by shared transcriptional activation. In contrast, following changes in bile acid flux through the liver, CYP7A1 gene expression becomes a strong modulator of hepatic cholesterol metabolism. (J Clin Endocrinol Metab 87: 4307– 4313, 2002)

metabolism is subject to marked interspecies variation (4), and most of our information has been derived from animal data. It is therefore essential to obtain direct information also from experiments performed in humans. Human studies have shown that interruption of the enterohepatic circulation (by resin treatment or ileal resection) stimulates (5–7), and treatment with bile acids such as chenodeoxycholic acid (CDCA) suppresses CYP7A1 and HMGR activities (6, 8 –10). Inhibition of HMGR by statin treatment stimulates LDLR expression but has little effect on CYP7A1 activity (11, 12). Tissue culture studies have shown that specific transcription factors (steroid regulatory element-binding proteins, SREBPs) sense the level of cholesterol in the cell membrane (13). Overexpression of SREBPs in transgenic mice increases LDLR and HMGR mRNAs but not CYP7A1 mRNA (14, 15). The transcriptional control of all three of these genes is thus not fully coordinated. However, data on the integrated regulation of HMGR, LDLR, and CYP7A1 mRNA levels under more physiological conditions are sparse also in animals. In the present study, we assayed human liver specimens from 76 patients under basal conditions or under various experimental regimens affecting cholesterol and bile acid homeostasis for the mRNA content of CYP7A1, HMGR, and LDLRs. The availability of frozen tissue from several clinical studies at our unit provided a unique opportunity to assay a large number of specimens on a single occasion, with the

4307

4308

J Clin Endocrinol Metab, September 2002, 87(9):4307– 4313

Rudling et al. • Regulation of Human Hepatic Cholesterol Homeostasis

aim to characterize the expression and the degree of coordination of these mRNAs under physiological conditions, and their regulatory responses during induced metabolic perturbations.

nation of enzyme activities and cholesterol concentration as described (5, 6, 11, 16, 17). Cholecystectomy was performed without any complications.

Materials Pravastatin and cholestyramine were from Bristol-Myers Squibb Co. (New Brunswick, RI), simvastatin from MSD (White House Station, NJ). CDCA (Chenofalk) and UDCA (Ursofalk) were purchased as 250-mg capsules from Dr. Falk Pharma (Freiburg, Germany). Taq polymerase was from Perkin-Elmer (Norwalk, CT). Guanidinium thiocyanate and mercaptoethanol were from Merck KGaA (Darmstadt, Germany). 35SUTP was purchased from Amersham Pharmacia Biotech (Uppsala, Sweden). All other reagents were from previously described sources (12, 18).

Patients and Methods Patients and treatments Liver biopsies from a total of 76 patients, 51 females and 15 males with cholesterol gallstones and 10 (8 females and 2 males) gallstone-free subjects undergoing elective cholecystectomy, were studied (Table 1). The patients, who were all scheduled to undergo open cholecystectomy, had been screened before inclusion, and no one had clinical or laboratory signs of diabetes mellitus; obesity (body mass index ⬎30); or diseases affecting liver, thyroid, or kidney function (5, 6, 11, 16, 17). None of the patients were initially on drug or diet therapy for hyperlipidemia or treatments with drugs known to affect lipid metabolism. Indication for cholecystectomy in the gallstone-free patients was either roentgenographic suspicion of polyps or adenomyoma in the gallbladder or a history of biliary colic inducible by cholecystokinin but with normal oral cholecystogram and ultrasonogram. Informed consent was obtained from each patient, and the studies were approved by the Ethics Committee at Huddinge University Hospital. The patients who were asked to participate in interventions were randomly selected, but no formal matching or strict protocol for randomization was used (6, 11). Of the screened patients initially asked to participate, less that 10% did not accept participation; these subjects did not display any different characteristics, compared with those who accepted. Altogether 21 untreated gallstone patients (group A) on normal diet were studied. Group B consisted of five subjects given a standarized cholesterol-free vegetarian diet containing 25 caloric % fat (polyunsaturated/saturated ratio, 3.2%) and 67 caloric % carbohydrates for 10 d before operation. To facilitate dietary control, these patients stayed in the metabolic ward for the entire period. Patients in groups C and D received CDCA (15 mg/kg䡠d) or ursodeoxycholic acid (UDCA) (15 mg/kg䡠d) for 3 wk before surgery. The medication was well tolerated, and liver function tests (aspartate aminotransferase or alanine aminotransferase, alkaline phosphatase, and bilirubin) were normal during treatment. Group E received bile acid-binding resin (cholestyramine) at a dose of 8 g twice daily for 3 wk preoperatively, and group F was treated with statins (pravastatin 20 mg twice daily or simvastatin 10 mg twice daily) for 3 wk before surgery. Group G received a combination of cholestyramine, 8 g twice daily, and pravastatin, 20 mg twice daily, for 3 wk before operation.

Construction of plasmids for in vitro generation of cRNA and mRNA A human 170-bp CYP7A1 gene fragment was obtained by amplification by the PCR, using native Taq polymerase, of human genomic DNA [bases 650 – 816 in the human gene (19)]. The product was cloned into the XBaI/EcoRI site of the pGEM-3Zf (⫹) vector and sequenced. A human HMGR gene fragment was obtained after amplification of a 150-bp fragment (bases ⫺6 to 143) of the human gene (20). A human LDLR 140-bp gene fragment was made by amplification of nucleotides 1204 –1344 of the human gene (21).

Preparation of hepatic RNA Frozen (⫺70 C) liver tissue was homogenized in 4 m guanidinium thiocyanate and 1% mercaptoethanol. After addition of N-lauroylsarcosine (0.5%), total RNA was isolated by ultracentrifugation on CsCl2 (18). Isolated RNA was dissolved in diethylpyrocarbonate-treated water, precipitated with ethanol, and quantitated at 260 nm assuming 1 OD ⫽ 37 ␮g RNA/ml. Hepatic mRNA levels were assayed by solution hybridization (18). Briefly, 35S-UTP-labeled cRNA probes were hybridized with four different increasing amounts of hepatic RNA at 68 C (on average from 15 to 56 ␮g total RNA per tube, range 2–110 ␮g). After Rnase treatment, hybrids were precipitated and collected on filters. The slopes of the hybridization signals were calculated by the method of least squares and compared with the slope generated from a synthetic human mRNA standard. The r values for the three standards were 0.99, 0.99, and 0.98 for HMGR, CYP7A1, and LDLR, respectively. For the samples the corresponding average r values were 0.98, 0.96, and 0.98. Each mRNA type was assayed in one large single assay. In a few cases, assays could not be completed for all three transcripts because of nonsignificant r values or inappropriate amounts of RNA. The successful numbers of assays in the 76 patients were 71 for HMGR, 67 for CYP7A1, and 59 for LDLR. The mRNA abundance was calculated assuming 5.5 pg DNA/cell and a RNA:DNA ratio of 2.7:1; this is not an absolute quantitation.

Experimental procedure All patients were admitted to the hospital in the morning the day before surgery and were given the regular hospital diet containing about 0.5 mmol cholesterol per day (except group B). After a 12-h fast, cholecystectomy was performed between 0800 and 1000 h. Standardized anesthesia was given (6, 11), and after the abdomen was opened, a liver biopsy (2– 4 g) was taken from the left lobe of the liver. One piece was rapidly frozen and stored in liquid nitrogen until analysis; the remaining part was used for immediate preparation of microsomes and determi-

Statistics Comparisons between gallstone and gallstone-free patients were made by t test. Comparisons between groups (see Fig. 2) were performed

TABLE 1. Clinical data of the patients Patients

No. of females/males

Gallstone free Gallstone treatment None Cholesterol-free diet CDCA UDCA Cholestyramine Pravastatin or simvastatin Cholestyramine ⫹ pravastatin Data are means ⫾

SEM.

Age (yr)

Body weight (kg)

Mean

Range

8/2

51 ⫾ 3

27– 69

67 ⫾ 3

16/5 5/0 10/1 8/2 1/3 5/1 6/3

53 ⫾ 3 60 ⫾ 7 42 ⫾ 3 48 ⫾ 3 40 ⫾ 6 47 ⫾ 4 57 ⫾ 4

29 –73 45 –77 25 –55 30– 61 20 –56 31 –58 43 –71

73 ⫾ 3 62 ⫾ 5 67 ⫾ 3 70 ⫾ 3 76 ⫾ 5 65 ⫾ 2 68 ⫾ 2

Total plasma cholesterol (mM) Pretreatment

At surgery

6.1 ⫾ 0.3 5.6 ⫾ 0.3 4.4 ⫾ 0.2 5.9 ⫾ 0.3 5.7 ⫾ 1.0 5.6 ⫾ 0.4 6.9 ⫾ 0.5

5.8 ⫾ 0.3 4.9 ⫾ 0.3 5.1 ⫾ 0.2 5.2 ⫾ 0.2 3.8 ⫾ 0.8 4.7 ⫾ 0.2 4.0 ⫾ 0.2


by one-way ANOVA with Dunnett’s test. Data were log transformed when there was a clear positive correlation between group means and group sd.

Results

We first evaluated whether untreated patients with cholesterol gallstones had altered transcript levels, compared with untreated patients without gallstones. No significant difference in mean mRNA levels for CYP7A1, HMGR, or LDLRs was found between the two groups, however (Fig. 1). We did not observe any influence of age, gender, or body weight on mRNA levels (data not shown), and we therefore pooled the data from the two groups of untreated controls in the following comparisons. Our approach made an analysis of the correlations among the three transcripts meaningful (Fig. 1). There were significant positive correlations between the mRNA levels for CYP7A1 and HMGR (r ⫽ ⫹0.49) and

FIG. 1. Relationships between hepatic mRNA abundance for CYP7A1, HMGR, and LDLR in samples from untreated patients with (open circles) and without (filled circles) gallstones. Regression lines were calculated by the method of least squares, r ⫽ correlation coefficient.

J Clin Endocrinol Metab, September 2002, 87(9):4307– 4313 4309

between those for HMGR and LDLRs (r ⫽ ⫹0.60). However, there was no significant correlation between CYP7A1 and LDLR mRNAs (Fig. 1B). Compared with most other species, bile acid and cholesterol synthesis as well as hepatic LDL uptake are low in humans (1, 4), and it has been suggested that this may be due to the higher dietary intake of cholesterol and fat in humans, compared with laboratory animals. To evaluate a possible effect of reducing dietary cholesterol, we studied five gallstone patients after 10 d on a cholesterol-free diet (Fig. 2). There was a tendency for increased mRNA levels for HMGR (by 80%), but this change was not statistically significant.

FIG. 2. Levels of mRNA for CYP7A1 (A), HMGR (B), and the LDLR (C) in the entire series of patients. Controls, Patients with and without gallstones; cheno, chenodeoxycholic acid; urso, ursodeoxycholic acid; cholfree, patients on cholesterol-free diet; statin, patients receiving pravastatin (n ⫽ 3) or simvastatin (n ⫽ 3); quest, cholestyramine; quest⫹prava, cholestyramine and pravastatin. Bars show average and SEM. White numbers within the bars are number of assayed samples of the group. *, Statistically significant difference (P ⬍ 0.05), compared with controls (Dunnett’s test).

4310


Thus, in response to this rather extreme diet, there were no drastic alterations in hepatic cholesterol metabolism. However, our data should not be taken to reject the hypothesis that the hepatic synthesis of cholesterol and bile acids may be suppressed in humans on a Western diet. By feeding gallstone patients the primary bile acid, CDCA, a 70% suppression of CYP7A1 transcripts was achieved (Fig. 2A). This is consistent with the concept of negative feedback inhibition of transcription of the CYP7A1 gene by bile acids (22). Feeding with UDCA does not suppress CYP7A1 activity or bile acid synthesis (6, 9), and we did not detect any change in CYP7A1 mRNA levels after treatment with this bile acid (Fig. 2A). The different effects of CDCA and UDCA on CYP7A1 would be expected to differently influence HMGR and LDLR mRNAs. Thus, HMGR mRNA levels were lower in response to CDCA but not to UDCA treatment (Fig. 2B). No change in LDLR mRNA was observed during CDCA or

FIG. 3. Relationships among the mRNA abundance of CYP7A1, HMGR, and LDLR shown in Fig. 2. A–C, Averages ⫾ SEM for the seven indicated groups are shown: 1, controls, patients with and without gallstones; 2, CDCA; 3, UDCA; 4, patients on cholesterol-free diet; 5, patients receiving pravastatin (n ⫽ 3) or simvastatin (n ⫽ 3); 6, cholestyramine; 7, cholestyramine ⫹ pravastatin. D–F, All individuals on treatment regimens are shown (r ⫽ correlation coefficient).


UDCA treatment, again indicating a weaker coupling between bile acid synthesis and LDLR gene expression. Resin treatment (cholestyramine) acts by binding intestinal bile acids, and particularly dihydroxy bile acids such as CDCA (23), thereby interrupting their enterohepatic circulation and derepressing CYP7A1 activity. As seen in Fig. 2, resin therapy resulted in a pronounced increase in CYP7A1 mRNA but also in HMGR and LDLR transcripts. Competitive inhibition of HMGR by treatment with statins (pravastatin or simvastatin) increased the mRNA for this enzyme. The effects on LDLR mRNA levels were less pronounced. Combined treatment with resin and statin increased all three transcripts (Fig. 2), indicating that depletion of a regulatory pool of cholesterol by simultaneously stimulated bile acid production and inhibition of cholesterol synthesis activated transcriptional regulation of the LDLR gene. By correlating the individual responses in mRNA levels


for CYP7A1, HMGR, and LDLRs during these perturbations of bile acid and cholesterol metabolism, the integrated regulatory response in the transcriptional control of the three proteins in human liver could be defined (Fig. 3). Significant correlations were obtained both for the averages of all seven groups (Fig. 3, A–C) and for the individual data from all subjects on treatment regimens (Fig. 3, D–F). The correlation between HMGR and LDLR mRNAs (r ⫽ ⫹0.60) was similar to that observed in untreated subjects (Figs. 1C and 3F), again consistent with a shared transcriptional control. In the treated individuals (Fig. 3, D and E), the correlations between CYP7A1 and HMGR and between CYP7A1 and LDLR mRNAs were very strong (r ⫽ ⫹0.84 and ⫹0.62, respectively; both P ⬍ 0.001). This indicates that, in contrast to the basal state (Fig. 1), perturbations primarily affecting CYP7A1 and bile acid metabolism exert a strong control of mRNA levels for HMGR and LDLRs. The importance of CYP7A1 as a driving force for the clearance of cholesterol also could be seen from the stronger negative correlation between plasma cholesterol and CYP7A1 mRNA (r ⫽ ⫺0.25), compared with LDLR mRNA (r ⫽ ⫺0.058) in the whole series of patients. Discussion

Previous studies on enzyme activities in gallstone patients have reported increased (8, 24, 25) as well as unaltered (9, 10, 17, 26) HMGR and reduced (8, 9, 24) as well as unchanged (17, 26) CYP7A1 activity. The present data showing similar mRNA levels are in line with previous studies of Scandinavian gallstone subjects, (10, 17) in whom no major abnormality has been observed for these enzyme activities. Our new finding of similar mRNA levels for the LDLRs in patients with and without cholesterol gallstones would argue against a major contribution also of the LDLR, and thus probably LDL cholesterol, in the pathogenesis of gallstone disease in the Scandinavian population. In agreement with this contention, plasma LDL cholesterol has not been reported to differ between patients with and without gallstones (27). Bile acid synthesis and LDL clearance have both been found to be reduced with age in normal individuals (28, 29). The age range of the controls was relatively limited in the present study, which may explain why we did not observe an influence of that parameter. Furthermore, although morbid obesity has been shown to be associated with increased mRNA levels of CYP7A1 and HMGR (30), there was no influence of body weight in the present work. However, obese patients (body mass index ⬎30) were not included in the present study. Under physiological conditions, HMGR and LDLR mRNAs were closely coregulated, probably because of their common transcriptional activation by SREBPs (13–15). A similar positive correlation (r ⫽ ⫹0.69) between HMGR and LDLR mRNA levels has been observed by Powell and Kroon (31) in liver samples from untreated humans. The apparent lack of correlation between CYP7A1 and LDLR mRNAs (Fig. 1B) suggests that normal variation of CYP7A1 activity may not be important for hepatic LDLR expression in man. Recent studies have indicated a possible influence of genetic variation at the CYP7A1 gene locus on plasma LDL cholesterol levels (32); however, we did not observe any correlation


between plasma cholesterol level and CYP7A1 mRNA in our untreated patients (data not shown). The nuclear receptor FXR has recently been shown to be involved in the transcriptional regulation of bile acid synthesis (33). After binding bile acids, FXR reduces CYP7A1 gene transcription and CDCA has the highest affinity toward FXR of all bile acids tested (34). CDCA also seems to be more effective than other di- or trihydroxylated bile acids to suppress CYP7A1 activity and in vivo bile acid production in man (6, 8, 9, 35). The lack of change in LDLR mRNA in CDCA-treated patients was somewhat unexpected because CDCA treatment, in contrast to UDCA, has been reported to increase plasma LDL cholesterol levels moderately (23). It may be mentioned that another hydrophobic bile acid, deoxycholic acid, was recently found not to influence mRNA levels (or activities) of CYP7A1, HMGR, or LDLRs (35). We have previously demonstrated that statins have no effect on CYP7A1 activity (11), and we now show also that CYP7A1 mRNA levels were unaffected. The failure to clearly demonstrate a significant increase in LDLR mRNA levels after treatment with statins alone was somewhat surprising. Although the number of subjects was limited, a 2- to 3-fold increase of the hepatic LDLR protein was found in two previous studies with a similar design (11,12). Although no final explanation to this discrepancy can be given, one possibility could be posttranscriptional regulation of the LDLRs in human liver. Such a phenomenon has indeed been observed in vitro (36), but further studies are needed to elucidate this question. It should be noted that CYP7A1 is not the only enzyme for conversion of cholesterol into bile acids in man. In addition to the classical neutral pathway involving CYP7A1 as the first rate-limiting step, there is an alternative acidic pathway in which the mitochondrial sterol 27-hydroxylase (CYP27) initiates the first step (37). However, in a recently reported study, the variation of CYP27 mRNA between patients treated with cholestyramine and CDCA was only about 1.5fold (38), compared with about 20-fold for CYP7A1 mRNA (Fig. 2A). This suggests that CYP27 may not be of major importance in the regulation of bile acid production in man. The thinking that the alternative acidic pathway via CYP27 is of little or no regulatory importance was recently confirmed in mice by Schwartz et al. (39). It is of interest to relate the degree of changes induced at the mRNA level to previously reported changes in enzyme activity in response to the perturbations studied here. In Fig. 4, we have compared the percentage changes in mRNA levels and enzyme activities of CYP7A1 and HMGR. It is clear that the changes in CYP7A1 activity parallel those in mRNA under most conditions, consistent with regulation of CYP7A1 at the transcriptional level. However, under some conditions, notably combined statin and cholestyramine treatment, the enzyme activity was lower than predicted by its mRNA level (Fig. 4A). This probably is due to the reduced availability of substrate for the enzyme, microsomal free cholesterol, in this situation (Fig. 4C). The activity of HMGR increased far more than mRNA levels in response to statins and statins ⫹ cholestyramine. This is expected since the statin is bound the enzyme in vivo thereby inhibiting its activity. During microsome isolation, the inactive enzyme is

4312



levels. These responses may be mediated by the nuclear receptor FXR (33). The most effective pharmacological strategy to reduce circulating cholesterol levels is presently to combine a drug that up-regulates CYP7A1 (resin) with a drug that reduces cholesterol synthesis (statin). In accordance with this, such a combination also gave the most dramatic increase in the mRNA concentrations measured. In view of the apparent key role of CYP7A1 for cholesterol metabolism in human liver, the search for novel pharmacological ways to stimulate this enzyme (40) should be promising. Acknowledgments We thank Ingela Arvidsson for expert technical assistance. Received December 19, 2001. Accepted June 7, 2002. Address all correspondence and requests for reprints to: Mats Rudling, M.D., Ph.D., CME, M63, Huddinge University Hospital, S-141 86 Stockholm, Sweden. E-mail: [email protected]. This work was supported by grants from Swedish Medical Research Council (03X-4793, 03X-7137, 03X-3141, 32X-14053, 14GX-13571); Swedish Society for Medical Research; Swedish Foundation for Strategic Research; Nordic Insulin Fund; Widengren, Thuring, Osterman, Jeansson, Axson Johnson, Ruth, and Richard Julin and Lundstro¨ m Foundations; Foundation of Old Female Servants; and Swedish Heart-Lung Foundation, “Fo¨ renade Liv” Mutual Group Life Insurance Co., Stockholm, Sweden; and Karolinska Institute.

References

FIG. 4. Relations between the means for the currently measured mRNA levels for CYP7A1 (A), HMGR (B), and previously assayed mean enzyme activities (Refs. 5, 6, 11, 16, and 17, and unpublished data). CYP 7A1 activity was determined with a method based on isotope-dilution-mass spectrometry, using deutering-labeled 7␣hydroxycholesterol as internal standard and endogenous microsomal cholesterol as substrate (41). The activity of microsomal HMGR was assayed by measuring the conversion of 14C-HMG-CoA to mevalonate, with tritium-labeled mevalonic acid as internal standard (42). C, Respective relative values for microsomal free cholesterol are shown. Data are expressed as percentage of controls (untreated gallstone patients). CON, Untreated gallstone and gallstone-free patients; S, statin treated; R, cholestyramine treated; R⫹S, cholestyramine and pravastatin-treated patients.

activated because the statin goes off from the enzyme. Thus, the high HMGR activity measured in vitro after in vivo inhibition by a statin reflects the large inactivated enzyme mass present in vivo. In conclusion, the present work shows that changes in the bile acid (and particularly CDCA) flux through the human liver are followed by marked responses in CYP7A1 mRNA levels, with less marked changes in HMGR and LDLR mRNA

1. Myant NB 1990 Cholesterol metabolism, LDL, and the LDL receptor. San Diego: Academic Press 2. Brown MS, Goldstein JL 1986 A receptor-mediated pathway for cholesterol homeostasis. Science 232:34 – 47 3. Levine GN, Keaney Jr JF, Vita JA 1995 Cholesterol reduction in cardiovascular disease. Clinical benefits and possible mechanisms. N Engl J Med 332:512–521 4. Dietschy JM, Turley SD, Spady DK 1993 Role of liver in the maintenance of cholesterol and low density lipoprotein homeostasis in different animal species, including humans. J Lipid Res 34:1637–1659 5. Reihne´r E, Angelin B, Rudling M, Ewerth S, Bjorkhem I, Einarsson K 1990 Regulation of hepatic cholesterol metabolism in humans: stimulatory effects of cholestyramine on HMG-CoA reductase activity and low density lipoprotein receptor expression in gallstone patients. J Lipid Res 31:2219 –2226 6. Reihne´r E, Bjo¨rkhem I, Angelin B, Ewerth S, Einarsson K 1989 Bile acid synthesis in humans: regulation of hepatic microsomal cholesterol 7␣-hydroxylase activity. Gastroenterology 97:1498 –1505 7. Åkerlund JE, Reihne´r E, Angelin B, Rudling M, Ewerth S, Bjorkhem I, Einarsson K 1991 Hepatic metabolism of cholesterol in Crohn’s disease. Effect of partial resection of ileum. Gastroenterology 100:1046 –1053 8. Coyne MJ, Bonorris GG, Goldstein LI, Schoenfield LJ 1976 Effect of chenodeoxycholic acid and phenobarbital on the rate-limiting enzymes of hepatic cholesterol and bile acid synthesis in patients with gallstones. J Lab Clin Med 87:281–291 9. Carulli N, Ponz De Leon M, Zironi F, Pinetti A, Smerieri A, Iori R, Loria P 1980 Hepatic cholesterol and bile acid metabolism in subjects with gallstones: comparative effects of short term feeding of chenodeoxycholic and ursodeoxycholic acid. J Lipid Res 21:35– 43 10. Ahlberg J, Angelin B, Einarsson K 1981 Hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase activity and biliary lipid composition in man: relation to cholesterol gallstone disease and effects of cholic acid and chenodeoxycholic acid treatment. J Lipid Res 22:410 – 422 11. Reihne´r E, Rudling M, Stahlberg D, Berglund L, Ewerth S, Bjorkhem I, Einarsson K, Angelin B 1980 Influence of pravastatin, a specific inhibitor of HMG-CoA reductase, on hepatic metabolism of cholesterol. N Engl J Med 323:224 –228 12. Rudling M, Norstedt G, Olivecrona H, Reihne´r E, Gustafsson JA, Angelin B 1992 Importance of growth hormone for the induction of hepatic low density lipoprotein receptors. Proc Natl Acad Sci USA 89:6983– 6987 13. Brown MS, Goldstein JL 1999 A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA 96:11041–11048 14. Shimano H, Horton JD, Hammer RE, Shimomura I, Brown MS, Goldstein JL 1996 Overproduction of cholesterol and fatty acids causes massive liver


15.

16.

17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

enlargement in transgenic mice expressing truncated SREBP-1␣. J Clin Invest 98:1575–1584 Horton JD, Shimomura I, Brown MS, Hammer RE, Goldstein JL, Shimano H 1998 Activation of cholesterol synthesis in preference to fatty acid synthesis in liver and adipose tissue of transgenic mice overproducing SREBP-2. J Clin Invest 101:2331–2339 Sahlin S, Ahlberg J, Reihne´ r E, Ståhlberg D, Einarsson K 1992 Cholesterol metabolism in human gallbladder mucosa: relationship to cholesterol gallstone disease and effects of chenodeoxycholic acid and ursodeoxycholic acid treatment. Hepatology 16:320 –326 Reihne´ r E, Angelin B, Bjo¨ rkhem I, Einarsson K 1991 Hepatic cholesterol metabolism in cholesterol gallstone disease. J Lipid Res 32:469 – 475 Rudling M 1992 Hepatic mRNA levels for the LDL receptor and HMG-CoA reductase show coordinate regulation in vivo. J Lipid Res 33:493–501 Nishimoto M, Noshiro M, Okuda K 1993 Structure of the gene encoding human cholesterol 7␣-hydroxylase. Biochim Biophys Acta 1171:147–150 Luskey KL, Stevens B 1985 Human 3-hydroxy-3-methylglutaryl coenzyme A reductase. Conserved domains responsible for catalytic activity and sterolregulated degradation. J Biol Chem 260:10271–10277 Yamamoto T, Davis CG, Brown MS, Schneider WJ, Casey ML, Goldstein JL, Russell DW 1984 The human LDL receptor: a cysteine-rich protein with multiple Alu sequences in its mRNA. Cell 39:27–38 Vlahcevic ZR, Heuman DM, Hylemon PB 1991 Regulation of bile acid synthesis. Hepatology 13:590 – 600 Angelin B, Einarsson K 1986 Bile acids and lipoprotein metabolism. Atheroscler Rev 15:41– 66 Salen G, Nicolau G, Shefer S, Mosbach EH 1975 Hepatic cholesterol metabolism in patients with gallstones. Gastroenterology 69:676 – 684 Maton PN, Ellis HJ, Higgins MJP, Dowling RH 1980 Hepatic HMGCoA reductase in human cholelithiasis: effects of chenodeoxycholic and ursodeoxycholic acids. Eur J Clin Invest 10:325–332 Ito T, Kawata S, Imai Y, Kakimoto H, Trzaskos JM, Matsuzawa Y 1996 Hepatic cholesterol metabolism in patients with cholesterol gallstones: enhanced intracellular transport of cholesterol. Gastroenterology 110:1619 –1627 Bennion LJ, Grundy SM 1978 Risk factors for the development of cholelithiasis in man. N Engl J Med 299:1221–1227 Einarsson K, Nilsell K, Leijd B, Angelin B 1985 Influence of age on secretion of cholesterol and synthesis of bile acids by the liver. N Engl J Med 313:277–282 Ericsson S, Eriksson M, Vitols S, Einarsson K, Berglund L, Angelin B 1991 Influence of age on the metabolism of plasma low density lipoproteins in healthy males. J Clin Invest 87:591–596


30. Ståhlberg D, Rudling M, Angelin B, Bjorkhem I, Forsell P, Nilsell K, Einarsson K 1997 Hepatic cholesterol metabolism in human obesity. Hepatology 25:1447–1450 31. Powell EE, Kroon PA 1994 Low density lipoprotein receptor and 3-hydroxy3-methylglutaryl coenzyme A reductase gene expression in human mononuclear leukocytes is regulated coordinately and parallels gene expression in human liver. J Clin Invest 93:2168 –2174 32. Wang J, Freeman DJ, Grundy SM, Levine DM, Guerra R, Cohen JC 1998 Linkage between cholesterol 7␣-hydroxylase and high plasma low density lipoprotein cholesterol concentrations. J Clin Invest 101:1283–1291 33. Russell DW 1999 Nuclear orphan receptors control cholesterol catabolism. Cell 97:539 –542 34. Wang H, Chen J, Hollister K, Sowers LC, Forman BM 1999 Endogenous bile acids are ligands for the nuclear receptor FXR/BAR. Mol Cell 3:543–553 35. Hillebrant C-G, Nyberg B, Angelin B, Axelson M, Bjorkhem I, Rudling M, Einarsson C 1999 Deoxycholic acid treatment in patients with cholesterol gallstones: failure to detect a suppression of cholesterol 7␣-hydroxylase activity. J Intern Med 246:399 – 407 36. Sharkey MF, Miyanohara A, Elam RL, Friedman T, Witztum JL 1990 Posttranscriptional regulation of retroviral vector-transduced low density lipoprotein receptor activity. J Lipid Res 31:2167–2178 37. Princen HMG, Post SM, Twisk J 1997 Regulation of bile acid biosynthesis. Curr Pharmaceut Design 3:59 – 84 38. Bjo¨ rkhem I, Araya Z, Rudling M, Angelin B, Einarsson C, Wikvall K 2002 Differences in the regulation of the classical and the alternative pathway for bile acid synthesis in human liver: NO COORDINATE REGULATION OF CYP7A1 AND CYP27A1. J Biol Chem 277:26804 –26807 39. Schwartz M, Russell DW, Dietschy JM, Turley SD 2001 Alternative pathways of bile acid synthesis in the cholesterol 7a-hydroxylase knockout mouse are not upregulated by either cholesterol or cholestyramine feeding. J Lipid Res 42: 1594 –1603 40. Angelin B, Eriksson M, Rudling M 1999 Bile acids and lipoprotein metabolism: a renaissance for bile acids in the post-statin era? Curr Opin Lipidol 10:269 –274 41. Einarsson K, Angelin B, Ewerth S, Nilsell K, Bjorkhem I 1986 Bile acid synthesis in man: assay of hepatic microsomal cholesterol 7 alpha-hydroxylase activity by isotope dilution-mass spectrometry. J Lipid Res 27:82– 88 42. Angelin B, Einarsson K, Liljeqvist L, Nilsell K, Heller RA 1984 3-hydroxy3-methylglutaryl coenzyme A reductase in human liver microsomes: active and inactive forms and cross-reactivity with antibody against rat liver enzyme. J Lipid Res 25:1159 –1166