Increased Activity of Hepatic Microsomal TriglycerideTransfer Protein and Bile Acid Synthesis in

Gallstone DiseaseJuan Castro,1 Ludwig Amigo,1 Juan Francisco Miquel,1 Cecilia Galman,2 Fernando Crovari,3 Alejandro Raddatz,3

Silvana Zanlungo,1 Roberto Jalil,4 Mats Rudling,2 and Flavio Nervi1

A strong interrelationship exists between the regulation of bile acid (BA) metabolism andhepatic very low density lipoprotein (VLDL) production. We have recently shown that BAsynthesis is increased in gallstone disease. We investigated the activity of hepatic microsomaltriglyceride transfer protein (MTTP) as a surrogate of VLDL production, BA synthesis, andmRNA expression levels of proteins that regulate fatty acid (FA) metabolism in the liver ofgallstone (GS) patients compared with GS-free patients. Twenty-seven volunteers subjectedto elective surgery; 9 were GS-free and 18 with GS agreed to have a liver biopsy. Wequantified by a fluorescence assay the activity of MTTP and by quantitative reverse-tran-scription PCR (RT-PCR) the mRNA content of hepatic MTTP and genes that regulatehepatic sterol and FA metabolism. Plasma was assayed for lathosterol and 7�-hydroxy-4-cholesten-3-one. Liver histology was normal in GS and GS-free patients. Serum VLDLtriglycerides and apoB were significantly increased in GS. Hepatic triglycerides tripled in GS(P < 0.001) compared with GS-free. MTTP activity increased 70% (P < 0.001). Serumlathosterol and hepatic cholesterol concentrations, and mRNA expressions of MTTP, CD36,and FABP1 were similar in GS-free and GS patients. Hepatic mRNA expression of hydroxy-3-methylglutaryl-coenzyme A reductase (HMGR) and 3-hydroxyl-3-methyl-glutaryl-CoAsynthase (HMGS) were significantly decreased—40% and 27%, respectively—in GS. Serum7�-hydroxy-4-cholesten-3-one was 75% higher, and mRNA expression of CYP7A1 wasincreased sevenfold (P < 0.001) in GS. Conclusion: Hepatic MTTP activity and BA synthesisare increased in GS. Results suggest that hepatic VLDL production and trafficking of BA areincreased in gallstone patients. (HEPATOLOGY 2007;45:1261-1266.)

Cholesterol gallstone disease (GS) is one of themost frequent and costly chronic diseases in theWestern world.1-4 Patients with cholesterol GS

have lithogenic bile supersaturated with cholesterol as a

sine qua non condition, and this is the result of abnormalregulation of hepatic cholesterol metabolism.1,2,5 We haverecently shown that the blood plasma levels of 7�-hy-droxy-4-cholesten-3-one (C4), a marker of bile acid (BA)synthesis, is increased in GS.6 This observation suggestedthat the liver was sensing a reduced inflow of BA from theintestine and that the enzymatic activities of cholesterol7�-hydroxylase (CYP7A1), the rate-limiting enzyme ofBA synthesis, could be increased in GS.

A number of observations have demonstrated a stronginterrelationship between the regulation of BA metabo-lism and hepatic very low density lipoprotein (VLDL)production.7-9 Abnormal BA absorption has been foundin familial hypertriglyceridemia,10 a genetic abnormalityassociated with diminished gene expression of the ileal BAtransporter.11 In addition, elevated serum triglycerideconcentrations have been reported in patients withCYP7A1 deficiency12 and in patients with increased fecalloss of BA as found in Crohn’s disease, ileal resection, andcholestyramine treatment.13-15 Conversely, treatment ofcholesterol GS with chenodeoxycholic acid has been

Abbreviations: BA, bile acids; C4, 7� �-hydroxy-4-cholesten-3-one; CYP7A1,cholesterol 7�-hydroxylase; FA, fatty acids; FXR, farnesoid X-activated receptor; GS,gallstone; HMGR, hydroxy-3-methylglutaryl-coenzyme A reductase; HMGS, 3-hy-droxyl-3-methyl-glutaryl-CoA synthase; MTTP, microsomal triglyceride transferprotein; VLDL, very low density lipoprotein.

From the 1Department of Gastroenterology, Pontificia Universidad Catolica,Santiago, Chile; the 2Center for Metabolism and Endocrinology, Karolinska Insti-tute at Huddinge University Hospital, Stockholm, Sweden; and the 3Departamentof Digestive Surgery and the 4Departament of Nephrology, Pontificia UniversidadCatolica, Santiago, Chile.

Received August 16, 2006; accepted December 19, 2006.Supported by grant 1030744 and 1010891 from Fondecyt (Fondo Nacional de

Investigacion Cientıfica y Tecnologia), Chile.Address reprint requests to: Flavio Nervi, M.D., Professor of Medicine, Pontificia

Universidad Catolica de Chile, 367 Marcoleta, Santiago, Chile. E-mail:fnervi@med.puc.cl; fax: (56) 2-639-7780.

Copyright © 2007 by the American Association for the Study of Liver Diseases.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/hep.21616Potential conflict of interest: Nothing to report.

1261

shown to reduce hypertriglyceridemia.16 BA reduces thesecretion of VLDL by repressing microsomal triglyceridetransfer protein (MTTP)17 through activation of nuclearreceptors.18

Abetalipoproteinemia, a genetic disease caused by mu-tations in the MTTP gene,19,20 blocks the assembly andsecretion of VLDL. Pharmacological inhibition of MTTPalso impairs the hepatic secretion of VLDL, as well as theintestinal secretion of chylomicrons.21,22 Liver-specificdeletion of the MTTP gene produces a dramatic reduc-tion in plasma VLDL and virtually eliminated apoli-poprotein B100 (apoB100) secretion23 and increased thesecretion of biliary BA.24 Conversely, hepatic overexpres-sion of MTTP results in increased secretion of VLDL.25

These observations support the contention that MTTPactivity in the liver reflects VLDL production and secre-tion.

Because BA synthesis is increased in GS6 and hepaticMTTP is rate limiting for VLDL production,20-25 we ex-plored the hypothesis that hepatic MTTP activity and BAsynthesis are both increased in non-obese patients withcholesterol GS. We measured the activity of hepaticMTTP and BA synthesis and the levels of mRNA geneexpression of MTTP and of a number of proteins respon-sible for hepatic fatty acids (FA) and sterol metabolism inthe livers of GS and GS-free patients.

Patients and Methods

Patient Selection and Sample Collection. The In-stitutional Review Board for Human Studies of the Fac-ulty of Medicine at the Pontificia Universidad Catolica deChile approved this study according to the ethical guide-lines of the 1975 Declaration of Helsinki. All patientsparticipating in the study gave informed consent. Theexperimental protocol included 18 GS patients subjectedto elective surgery because of previously symptomatic GSand a control group formed by 9 GS-free individuals op-erated on because of other gastrointestinal disease. Crite-ria of selection included: (1) normal serum glucose andalbumin concentrations; (2) no antecedent of obesity norknown changes of �5% in their habitual body weightbefore surgery; (3) no antecedent of treatment with drugsaffecting lipid metabolism; (4) normal results of liver bi-opsy; (5) GS-free patients operated on because of gastro-intestinal neoplasia had their disease localized to theaffected organ.

Two fasting serum specimens, one containing buty-lated hydroxytoluene collected at the time of surgery andstored at �70°C, were thawed for general chemistries,lathosterol, and C4 measurements. GS are mainly of thecholesterol type in this population,10 and the morphology

of all patients’ gallstones was of the cholesterol type in thisstudy. Determinations of lathosterol and C4 were per-formed blindly in Stockholm. Codes were opened by theinvestigators at the end of the study. Liver biopsies, ap-proximately 100 mg tissue, were obtained from the leftlobe of the liver at the beginning of surgery. Serum spec-imens were obtained stored at �70°C after induction ofanesthesia with 2.5 mg/kg propofol and 5 �g/kg fentanyl,after approximately 12 hours of fasting.

Analytical Procedures. Plasma VLDL separation wasperformed by ultracentrifugation of plasma specimens.26

The Autokit ApoB was used for the quantitative determi-nation of apoB in serum VLDL (Wako Chemicals, Rich-mond, VA). Assay of lathosterol was determined byisotope dilution mass spectrometry after the addition ofdeuterium-labeled internal standard to 50 �l plasma.27

Assay of C4 was performed as described in detail else-where.28 Serum free FA determinations were performedusing a commercial kit (RANDOX Laboratories, Crum-lin, UK). Hepatic cholesterol, triglycerides, and phospho-lipids were extracted, solubilized, and measured aspreviously described.29,30

Microsomal Triglyceride Transfer Protein ActivityAssay. MTTP activity was measured by using an MTTPassay kit according to the manufacturer’s instructions(Roar Biomedical, New York, NY). This method has beenpreviously used in previously published experimentalwork.31-33 The assay is based on a transfer of fluorescenceattributable to MTTP activity, between donor and accep-tor particles. Liver samples were homogenized and soni-cated in buffer containing 15 mM Tris (pH 7.4), 40 mMNaCl, 1 mM EDTA, and protease inhibitors. The MTTPassay was performed by incubating 10 �l (75-125 �gprotein) liver homogenate with 10 �l donor and 10 �lacceptor solutions in a total volume of 250 �l buffer andincubated for 3 hours at 37°C. MTTP activity was calcu-lated by measuring fluorescence at the excitation wave-length of 465 nm and emission wavelength of 538 nmusing the Turner Biosystems fluorometer (Sunnyvale,CA).

Isolation of Total RNA and Real-Time Quantita-tive RT-PCR. Total RNA was isolated from human liverby an acidic guanidinium isothiocyanate/phenol/chloro-form extraction procedure as described by Chomczynskiand Sacchi.34 Total RNA, 1.875 �g, was reverse tran-scribed using the SuperScript First-Strand Synthesis Sys-tem for RT-PCR (Invitrogen, Bios Chile, I.G.S.A.,Santiago, Chile) and random hexamers.

Quantitative PCR was performed in a Mx3000P Real-Time PCR System (Stratagene, La Jolla, CA) using Plat-inum Quantitative PCR SuperMix-UDG (Invitrogen,Bios Chile, I.G.S.A., Santiago, Chile), 25 ng reverse-tran-

1262 CASTRO ET AL. HEPATOLOGY, May 2007

scribed RNA, and specific TaqMan Gene Expression As-says (Applied Biosystems, Foster City, CA) for a numberof genes that regulate FA and sterol metabolism, and eu-karyotic 18S rRNA. The primers sequence used in theassays is available on request. Thermal cycling conditionswere 95°C for 10 minutes followed by 40 cycles at 93°Cfor 15 seconds, and 60°C for 1 minute. Reactions wereperformed in duplicate. Relative quantification of geneexpression was performed using the comparative thresh-old (Ct) method as described by Applied Biosystems (Fos-ter City, CA). Changes in mRNA expression level werecalculated after normalization to 18S expression.

Statistics. Data are presented as mean � SEM. Statis-tical analysis was carried out by Student t test. Linearcorrelation was used to examine the relation between vari-ables, and normal probability plots were used to analyzewhether log-transformation of data was necessary.

ResultsThe clinical data, serum glucose, insulin, and lipids of

the 2 groups of patients are shown in Table 1. Age andbody mass index between subjects with and without GSwere not significantly different. Serum concentrations oftotal plasma, low-density lipoprotein or high-density li-poprotein cholesterol, total triglycerides, and FA were notincreased among subjects with GS as compared to GS-freesubjects. VLDL triglycerides and apo B concentrations

were significantly increased (P � 0.02) in GS comparedto GS-free patients

Figure 1A shows that triglyceride levels were increasedalmost 300% in the liver of GS compared to GS-freepatients (P � 0.001) in spite of absence of histologicalevidence of fat accumulation the liver. Hepatic free FAconcentration was similar in both groups of patients (Fig.1B). The activity of MTTP was significantly increased71% (P � 0.001), and the mRNA expression level wasunchanged in the liver of GS subjects compared withGS-free (Fig. 1C, D). No significant correlation wasfound between body mass index and MTTP activity, andmRNA expression level. Hepatic mRNA expression levelsof Apo B, the constitutive apoprotein of nascent VLDLparticles, FA translocase FAT/CD36, and FABP1 re-mained unchanged in the liver of GS compared with GS-free patients (results not shown).

The plasma concentration of C4 was significantly in-creased 75% in GS patients as compared with GS-freeindividuals (P � 0.001) as shown in Fig. 2A. The mRNAexpressions of 7�-hydroxylase, the rate-limiting enzymeof the neutral pathway of BA biosynthesis (CYP7A1),increased almost 7-fold (P � 0.001) in GS compared withGS-free (Fig. 2C). Hepatic mRNA expression levels ofCYP27A1 (27�-hydroxylation of cholesterol in the“acidic” pathway of BA biosynthesis) was decreased 25%in GS, whereas the mRNA expression levels of ABCB11(canalicular transporter of bile acids) were similar in GS-

Table 1. Clinical Characteristics, Serum Glucose, Insulinand Lipids, and HOMA-IR of Patients

Characteristic GS-Free (6F/3M) GS (13F/5M)

Age 49 � 18 49 � 14Body mass index 24.3 � 3.1 25.1 � 3.2Serum glucose (mmol/l) 5.5 � 0.6 5.1 � 0.3Serum insulin (pmol/l) 45.5 (20.5-70.8) 50.7 (43.1-60.1)HOMA-IR 1.4 (0.9-2.6) 1.0 (0.4-1.2)Total cholesterol (mmol/l) 4.2 � 0.9 4.5 � 0.9LDL cholesterol (mmol/l) 3.0 � 0.8 2.7 � 0.7HDL cholesterol (mmol/l) 1.0 � 0.2 1.2 � 0.3TG (mmol/l) 1.1 � 0.4 1.4 � 0.5VLDL TG (mmol/l) † 0.3 � 0.06 0.6 � 0.08*VLDL apoB (mg/l) † 133 � 21 195 � 12*Serum-free FA (mmol/l) 1.2 � 0.2 1.1 � 0.1

NOTE. Serum specimens for determination of FAA were obtained after 12-14hrs of fasting and induction of anesthesia with Propofol 2, 5 mg/kg and Fentanyl5 mcg/kg. GS—free group was formed by patients operated on with the diagnosisof: uncomplicated hydatic cyst of the liver (1), peritoneal sarcoma (1), gastriccancer (3), colon cancer (3), and hepatoma without cirrhosis (1). All surgicalprocedures began between 8–10 AM. The morphology of gallstones was of thecholesterol type in all cholecystectomized patients. Liver biopsies were obtainedat the beginning of the surgical procedure. Values represent the mean � SE,excepting the values for the serum levels of serum insulin and HOMA–IR thatcorrespond to the medians and interquartile range of values shown in brackets.HOMA – IR, homeostasis model assessment – insulin resistance. †Values corre-spond to 6 GS-free and 6 GS patients. *P � 0.02.

Fig. 1. Hepatic triglyceride and FA concentration, MTTP activity andmRNA expression level in GS and GS-free patients. Data from GS-free(white bars) and GS (black bars) patients represent the average � SEM.*P � 0.001.

HEPATOLOGY, Vol. 45, No. 5, 2007 CASTRO ET AL. 1263

free and GS (Fig. 2B,D). Hepatic mRNA expression lev-els of the nuclear receptors including liver X receptor �and farnesoid X-activated receptor (FXR) were similar inGS-free and GS patients (results not shown).

Serum lathosterol concentrations were similar in GSpatients compared with GS-free individuals, whereas he-patic HMGR and HMGS mRNA level decreased 40%and 27%, respectively, in GS patients (P � 0.001) (Fig.3). Hepatic cholesterol concentration was similar in theGS-free and GS groups (results not shown).

DiscussionIn agreement with a study from Sunzel et al.,35 we

found that triglyceride concentration in the liver of GSwas significantly increased compared with GS-free pa-tients in the absence of histological evidence of fatty liver.Also these studies confirmed that BA synthesis was in-creased in GS and showed for the first time that serumVLDL triglycerides and apoB concentrations, and the ac-tivity of MTTP were significantly increased in the liver of

GS. These observations suggested, but did not prove, thathepatic VLDL production was increased in this disease.Our findings were consistent with epidemiological obser-vations showing that GS is associated with hypertriglyc-eridemia in some population-based studies.36-38 Wepostulate that the enhanced activity of MTTP by the liverand the increased serum VLDL triglyceride concentrationis a coordinate response to the increased level of BA syn-thesis in GS, a secondary result secondary to decreasedileal absorption of BA. This contention is supported by anumber of observations showing a strong interrelation-ship between the regulation of bile acid metabolism andhepatic VLDL production.7-9

We corroborated here by 2 different methods (serumdetermination of C4 and mRNA CYP7A1 expression lev-els) that the synthesis of BA was increased in GS, as pre-viously shown.6 Total body cholesterol synthesis assessedhere by determination of serum lathosterol concentrationremained similar in GS and GS-free patients, as previ-ously found in a Caucasian population.39 Hepatic choles-terogenesis indirectly assessed by the determination ofHMGR and HMGS mRNA expression levels wereslightly but significantly decreased in the liver of GS com-pared with GS-free patients. These results suggested thatthe liver of GS was sensing increased concentration of freecholesterol in the metabolically active pool of the hepato-cyte, in spite of the increased level of BA synthesis. Thesupply of cholesterol for the metabolic needs of the hepa-tocytes of GS patients, including VLDL production, bil-iary cholesterol secretion, and BA synthesis, may haveoriginated from increased uptake of serum lipoproteincholesterol, including VLDL remnants, low-density li-poprotein, and/or high-density lipoprotein.

The increased BA synthesis associated with increasedplasma VLDL triglyceride and apoB concentrations, andhepatic MTTP activity found in GS, indicates that crosstalk exists between these metabolic pathways. A candidateregulatory factor is the BA-activated nuclear orphan re-ceptor FXR that regulates BA synthesis and transport.40-42

FXR has been shown to repress BA biosynthesis by induc-

Fig. 2. Serum concentrations of 7�-hydroxy-4-cholesten-3-one (C4)and mRNA expression levels of CYPP7A1, CYP27A1, and ANCB11. Datafrom GS-free (white bars) and GS (black bars) patients represent theaverage � SEM. *P � 0.001. �P � .01.

Fig. 3. Serum lathosterol concentration and mRNAexpression levels of 3-hydroxyl-3-methyl-glutaryl-CoAreductase (HMGR) and 3-hydroxyl-3-methyl-glutaryl-CoA synthase (HMGS). Data from GS-free (white bars)and GS (black bars) patients represent the average �SEM. *P � 0.001.

1264 CASTRO ET AL. HEPATOLOGY, May 2007

ing the expression of short heterodimer partner 1, whichfunctions to inhibit liver receptor homologue 1, a com-petence factor essential for cholesterol 7�-hydroxylasegene transcription.40,41 BA may act similarly throughFXR to repress hepatic VLDL production. Several recentstudies indirectly support this model. Administration torats of a non-BA FXR agonist, GW4064, resulted in adose-dependent decrease in serum triglyceride levels sim-ilar to previous BA feeding studies in rats and humans.43

Conversely, disruption of BA signaling in the liver bytargeted disruption of FXR in mice resulted in increasedhepatic triglyceride contents and elevated serum triglyc-eride levels,44 analogous to BA malabsorption states.

The origin of the increased concentration of triglycer-ides in the liver of GS was not apparent in the currentstudy. The most likely hypothesis is that the increasedtriglyceride concentration in the liver of GS was the resultof increased flux of serum free FA or VLDL remnants intothe hepatocytes, because the hepatic mRNA expressionlevels of enzymes that play key roles in the regulation ofFA synthesis and oxidation were similar in GS-free andGS (unpublished results from this laboratory).

The increased activity of MTTP in the liver of GS wasnot related to higher than normal mRNA expression lev-els of the MTTP gene, suggesting a posttranscriptionalstimulation of MTTP activity in GS. Although we did notfind differences in hepatic FA concentrations in thisstudy, we postulate that the flux of free FA, and otherlipids that form the nascent VLDL particles in the endo-plasmic reticulum, was increased in our GS patients. Infact, substrate stimulation of MTTP activity occurs byincreasing the supply of free FA in Hela cells45 and byglucose in the rat liver.46,47 Increased supply of glucoseand free FA can occur in insulin resistance and diabetes,48

conditions associated with GS in this population.49 Onemay be tempted to speculate that hepatic insulin resis-tance not detected by the HOMA method used in thisstudy could be involved in the metabolic abnormalitiesfound in our GS patients.

In conclusion, our current results showing an increasedconcentration of triglycerides and activity of hepaticMTTP in the liver of GS do not prove but do support thehypothesis that VLDL production is increased in GS.This study showing increased synthesis of BA in GS sug-gests an increased fecal loss of bile acids, a situation con-sistent with a decreased BA pool as found in GS.1-4 Thiscondition should determine an increased demand forcholesterol in the liver. Thus, an important driving forcestimulating BA synthesis, and presumably hepatic VLDLproduction in GS, should simply be a reduced level of BAin the enterohepatic circulation as the primary patho-physiological mechanism. An important issue therefore

would be to discover whether GS subjects from this pop-ulation present a primary defect in the ileum character-ized by impaired absorption and increased fecal loss of BAand a secondary defect in the liver characterized by in-creased BA synthesis and VLDL production.

References1. Shaffer EA. Epidemiology and risk factors for gallstone disease: has the

paradigm changed in the 21st century? Curr Gastroenterol Rep 2005;7:132-140.

2. Dowling RH. Review: pathogenesis of gallstones. Aliment Pharmacol Ther2000;14:39-47.

3. Lee SP, Ko CW. Gallstones. In: Yamada T, Alpers DH, Laine L, OwyangC, Powell DW, eds. Textbook of Gastroenterology. Philadelphia: Lippin-cott,1999; 2258-2280.

4. Bennion LJ, Grundy SM. Risk factors for the development of cholelithiasisin man. N Engl J Med 1978;299:1161-1167; 1221-1227.

5. Zanlungo S, Rigotti A, Nervi F. Hepatic cholesterol transport from plasmainto bile: implications for gallstone disease. Curr Opin Lipidol 2004;15:279-286.

6. Galman C, Miquel JF, Perez RM, Einarsson C, Ståhle L, Marshall G, et al. Bileacid synthesis is increased in Chilean Hispanics with gallstones and in gallstonehigh-risk Mapuche Indians. Gastroenterology 2004;126:741-748.

7. Kang S, Davis RA. Cholesterol and hepatic lipoprotein assembly and se-cretion. Biochim Biophys Acta 2000;1529:223-230.

8. Mason TM. The role of factors that regulate the synthesis and secretion ofvery-low-density lipoprotein by hepatocytes. Crit Rev Clin Lab Sci 1998;35:461-487.

9. Farkkila M, Miettinen TA. Lipid metabolism in bile acid malabsorption.Ann Med 1990;22:5-13.

10. Duane WC. Abnormal bile acid absorption in familial hypertriglyceride-mia. J Lipid Res 1995;36:96-107.

11. Duane WC, Hartich LA, Bartman AE, Ho SB. Diminished gene expres-sion of ileal apical sodium bile acid transporter explains impaired absorp-tion of bile acid in patients with hypertriglyceridemia. J Lipid Res 2000;41:1384-1389.

12. Pullinger CR, Eng C, Salen G, Shefer S, Batta AK, Erickson SK, et al.Human cholesterol 7alpha-hydroxylase (CYP7A1) deficiency has a hyper-cholesterolemic phenotype. J Clin Invest 2002;110:109-117.

13. Buchwald H, Varco RL, Matts, JP, Long JM, Fitch LL, Campbell GS, et al.Effect of partial ileal bypass surgery on mortality and morbidity fromcoronary heart disease in patients with hypercholesterolemia. N EnglJ Med 1990;323:946-955.

14. Grundy SM, Ahrens EH Jr., Salen G. Interruption of the enterohepaticcirculation of bile acids in man: comparative effects of cholestyramine andileal exclusion on cholesterol metabolism. J Lab Clin Med 1971;78:94-121.

15. Einarsson K, Ericsson S, Ewerth S, Reihner E, Rudling M, Stahlberg D, etal. Bile acid sequestrants: mechanisms of action on bile acid and cholesterolmetabolism. Eur J Clin Pharmacol 1991;40(Suppl 1):S53-S58.

16. Miller NE, Nestel P J. Triglyceride-lowering effect of chenodeoxycholicacid in patients with endogenous hypertriglyceridemia. Lancet 1974;2:929-931.

17. Hirokane H, Nakahara M, Tachibana S, Shimizu M, Sato R. Bile acidreduces the secretion of very low density lipoprotein by repressing micro-somal triglyceride transfer protein gene expression mediated by hepatocytenuclear factor-4. J Biol Chem 2004;279:45685-45692.

18. Watanabe M, Houten SM, Wang L, Moschetta A, Mangelsdorf DJ, Hey-man RA, et al. Bile acids lower triglyceride levels via a pathway involvingFXR, SHP, and SREBP-1c. J Clin Invest 2004;113:1408-1418.

19. Wetterau JR, Aggerbeck LP, Bouma ME, Eisenberg C, Munck A, HermierM, et al. Absence of microsomal triglyceride transfer protein in individualswith abetalipoproteinemia. Science 1992;258:999-1001.

20. Gregg RE, Wetterau JR. The molecular basis of abetalipoproteinemia.Curr Opin Lipidol 1994;5:81-86.

HEPATOLOGY, Vol. 45, No. 5, 2007 CASTRO ET AL. 1265

21. Gordon DA. Recent advances in elucidating the role of the microsomaltriglyceride transfer protein in apolipoprotein B lipoprotein assembly.Curr Opin Lipidol 1997;8:131-137.

22. Shelness GS, Sellers JA. Very-low-density lipoprotein assembly and secre-tion. Curr Opin Lipidol 2001;12:151-157.

23. Raabe M, Veniant MM, Sullivan MA, Zlot CH, Bjorkegren J, Nielsen LB,et al. Analysis of the role of microsomal triglyceride transfer protein in theliver of tissue-specific knockout mice. J Clin Invest 1999;103:1287-1298.

24. Amigo L, Castro J, Miquel JF, Zanlungo S, Young S, Nervi F. Inactivationof hepatic microsomal triglyceride transfer protein protects mice fromdiet-induced gallstones. Gastroenterology 2006;131:1870-1878.

25. Tietge UJ, Bakillah A, Maugeais C, Tsukamoto K, Hussain M, RaderDJ. Hepatic overexpression of microsomal triglyceride transfer protein(MTP) results in increased in vivo secretion of VLDL triglycerides andapolipoprotein B. J Lipid Res 1999;40:2134-2139.

26. Chung BH, Wilkinson T, Geer JC, Segrest JP. Preparative and quantita-tive isolation of plasma lipoproteins: rapid, single discontinous densitygradient ultracentrifugation in a vertical rotor. J Lipid Res 1980;21:284-291.

27. Lund E, Sisfontes L, Reihner E, Bjorkhem I. Determination of serum levelsof unesterified lathosterol by isotope dilution-mass spectrometry. ScandJ Clin Lab Invest 1989;49:165-171.

28. Pettersson L, Eriksson CG. Reversed-phase high-performance liquid chro-matographic determination of 7 alpha-hydroxy-4-cholesten-3-one in hu-man serum. J Chromatogr B Biomed Appl 1994;657:31-36.

29. Carr TP, Andersen CJ, Rudel LL. Enzymatic determination of triglyceride,free cholesterol, and total cholesterol in tissue lipid extracts. Clin Biochem1993;26:39-42.

30. Zanlungo S, Amigo L, Mendoza H, Miquel JF, Vio C, Glick JM, et al.Sterol carrier protein 2 gene transfer changes lipid metabolism andenterohepatic sterol circulation in mice. Gastroenterology 2000;119:1708-1719.

31. Letteron P, Sutton A, Mansouri A, Fromenty B, Pessayre D. Inhibition ofmicrosomal triglyceride transfer protein: another mechanism for drug-induced steatosis in mice. HEPATOLOGY 2003;38:133-140.

32. Fisher EA, Pan M, Chen X, Wu X, Wang H, Jamil H, et al.The triplethreat to nascent apolipoprotein B: evidence for multiple distinct degrada-tive pathways. J Biol Chem 2001;276:27855-27863.

33. Funatsu T, Suzuki K, Goto M, Arai Y, Kakuta H, Tanaka H, et al. Pro-longed inhibition of cholesterol synthesis by atorvastatin inhibits apoB-100 and triglyceride secretion from HepG2 cells. Atherosclerosis 2001;157:107-115.

34. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem1987;162:156-159.

35. Sunzel H, Martinssin A, Hood B. The lipid content of the human liver: acomparison between patients with gallstone disease and peptic ulcer dis-ease. Metabolism 1964;13:1469-1474.

36. Einarsson K, Hellstrom K, Kallner M. Gallbladder disease in hyperli-poproteinaemia. Lancet 1975;1:484-487.

37. Attili AF, Capocaccia R, Carulli N, Festi D, Roda E, Barbara L, et al.Factors associated with gallstone disease in the MICOL experience. Mul-ticenter Italian Study on Epidemiology of Cholelithiasis. HEPATOLOGY

1997;26:809-818.38. Loria P, Lonardo A, Lombardini S, Carulli L, Verrone A, Ganazzi D, et al.

Gallstone disease in non-alcoholic fatty liver: prevalence and associatedfactors. J Gastroenterol Hepatol 2005;20:1176-1184.

39. Muhrbeck O, Wang FH, Bjorkhem I, Axelson M, Einarsson K. Circulat-ing markers for biosynthesis of cholesterol and bile acids are not depressedin asymptomatic gallstone subjects. J Hepatol 1997;27:150-155.

40. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, et al.Molecular basis for feedback regulation of bile acid synthesis by nuclearreceptors. Mol Cell 2000;6:507-515.

41. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, etal. A regulatory cascade of nuclear receptors FXR, SHP-1, and LRH-1represses bile acid biosynthesis. Mol Cell 2000;6:517-526.

42. Lu TT, Repa JJ, Mangelsdorf DJ. Orphan nuclear receptors as eLiXiRs andFiXeRs of sterol metabolism. J Biol Chem 2001;276:37735-37738.

43. Maloney PR, Parks DJ, Haffner CD, Fivush AM, Chandra G, PlunketKD, et al. Identification of a chemical tool for the orphan nuclear receptorFXR. J Med Chem 2000;43:2971-2974

44. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G, Gonzalez FJ.Targeted disruption of the nuclear receptor FXR/BAR impairs bile acidand lipid homeostasis. Cell 2000;102:731-744.

45. Gordon DA, Jamil H, Sharp D, Mullaney D, Yao Z, Gregg RE, et al.Secretion of apolipoprotein B-containing lipoproteins from HeLa cells isdependent on expression of the microsomal triglyceride transfer proteinand is regulated by lipid availability. Proc Natl Acad Sci U S A 1994;91:7628-7632.

46. Cho H, Kang H, Choi S, Ju Y, Lee H, Park H. The possible role of Ca2�on the activation of microsomal triglyceride transfer protein in rat hepato-cytes. Biol Pharm Bull 2005;28:1418-1423.

47. Cho H, Kang H, Ju Y, Lee H, Kim H, Park H. Augmentation of Ca2�-induced microsomal triglyceride transfer protein activity by glucose supplyenhances hypertriglyceridemia in vivo. Biol Pharm Bull 2006;29:889-895.

48. Ginsberg HN, Zhang YL, Hernandez-Ono A. Regulation of plasma trig-lycerides in insulin resistance and diabetes. Arch Med Res 2005;36:232-240.

1266 CASTRO ET AL. HEPATOLOGY, May 2007