Primary dual defect of cholesterol and bile acid metabolism in liver of patients with intrahepatic calculi

GASTROENTEROLOGY 1995;108:1534-1546

Primary Dual Defect of Cholesterol and Bile Acid Metabolism in Liver of Patients With Intrahepatic Calculi JUNICHI SHODA,* BING-FANG HE,* NAOMI TANAKA,* YASUSHI MATSUZAKI,* SHYUNJI YAMAMORI,* and TOSHIAKI OSUGA* *Department of Gastroenterology, Institute of Clinical Medicine, University of Tsukuba, Ibaraki; and *Department of Gene Analysis, Mitsubishi-Yuka Bioclinical Laboratories Inc., Tokyo, Japan

Background/Aims: Intrahepatic calculi, which are characterized by cholesterol-rich pigment stones, are highly prevalent in East Asia. Their pathogenesis remains unknown. To elucidate the etiological factors underlying the formation of cholesterol-supersaturated bile, which leads to the formation of cholesterol-rich pigment stones, cholesterol and bile acid de novo syntheses in the liver were studied. Methods: Liver specimens were assayed for the catalytic activities and steady-state messenger RNA levels of 3-hydroxy-3-methylglutarylcoenzyme A (HMG-CoA) reductase and cholesterol 7~hydroxylase. Results: The activity of HMG-CoA reductase, consistent with the messenger RNA level, was significantly higher in 13 patients with intrahepatic brown pigment stones (11.2 + 1.3 pmol-min -1.mg protein -1 [mean _+ SEM; P < 0.0001] for affected hepatic lobes and 13.4 _ 1.7 [P < 0.0001] for unaffected ones [P < 0.0001]) than in 19 control subjects (6.4 +_ 0.4) and in 29 patients with gallbladder cholesterol stones (2.1 _+ 0.1). On the other hand, the activity of 7~x-hydroxylase, consistent with the messenger RNA level, was significantly lower in patients with intrahepatic brown pigment stones (2.8 _ 0.5 pmol. m i n - l . m g protein 1 [p ~ 0 . 0 0 0 1 ] for affected lobes and 2.6 +_ 0.5 [P < 0.0001] for unaffected ones) than in control subjects (6.0 _+ 0.6) and in patients with cholesterol stones (5.1 ___0.5). Conclusions: In intrahepatic calculi, the formation of supersaturated bile and cholesterol-rich pigment stones may be attributed to the primary dual defect of up-regulated cholesterogenesis and down-regulated bile acid synthesis in the liver.

ntrahepatic calculi are prevalent in East Asia ~'2 and affect 3 % - 6 % of patients undergoing surgery to remove gallstones. 3 Most intrahepatic gallstones appear as brown pigment stones (calcium bilirubinate stones) by visual inspection and infrared analysis. 4 Brown pigment stones differ from cholesterol and black pigment stones in their composition5 and etiology.6'v However, recent reports have shown that the chemical composition of intrahepatic calculi is not identical to that of brown

pigment stones in the extrahepatic bile ducts; the former contain more cholesterol and less bilirubin and bile acid8 and also include lesser amounts of bile acids modified by bacterial metabolism. 9 In addition, primary pure cholesterol stones in the intrahepatic bile ducts have recently been reported. 1°'11 The higher cholesterol content and the higher prevalence of intrahepatic calculi in women than in men, ~2 as found in cases of cholesterol stones in the gallbladder, imply possible relationships between stone pathogenesis and metabolic factors. The factors responsible for the formation of intrahepatic calculi have not been thoroughly elucidated. The very high incidence of bacteria in bile suggests that bacteria play a significant role in the formation of brown pigment stones in most cases,13 but bacterial infection cannot explain solely why brown pigment stones in the intrahepatic bile ducts contain a higher amount of cholesterol than those in the extrahepatic ducts. 8a4 We have recently observed several alternative points of bile acid composition in the gallstones, blood, bile, and liver specimens of patients with intrahepatic calculi, 14 which may be attributed not only to secondary changes resulting from the local disturbances limited to the intrahepatic bile ducts in affected lobes but also to primary alterations of hepatocyte metabolism. Of particular interest is the possibility of increased cholesterogenesis and decreased bile acid synthesis in the liver) 4 Therefore, in this study, we aimed at achieving a better understanding of the pathogenesis of cholesterol-rich brown pigment stones in the intrahepatic bile ducts. We studied the etiologic factors underlying the formation of cholesterol supersaturated bile with special reference to alterations of cholesterogenesis and bile acid synthesis in the liver of patients with intrahepatic calculi. Abbreviations used in this paper: DTT, dithiothreitol; HMG-CoA, 3hydroxy-3-methylglutaryl-coenzyme A; LDL, low-density lipoprotein; NADPH, reduced nicotinamide adenine dinucleotide phosphate. © 1995 by the American Gastroenterological Association 0016-5085/95/$3.00

May 1 9 9 5

PRIMARY DUAL DEFECT IN INTRAHEPATIC CALCULI

1535

T a b l e 1 . D a t a o f P a t i e n t s With I n t r a h e p a t i c C a l c u l i

Liver atrophy

Bile duct

and Stone location a

Patients

Stricture

Dilatation

fibrosis b

Gallbladder stone

n

Sex (M/F)

Intra

Intra and extra

R

LR

L

+

-

+

-

+

-

With

Without

Unknown

39

19/20

21

18

6

8

25

21

18

37

2

17

22

24

8

7

4

1

1

0

4

3

2

2

3

0

5

2

2

1

With brown pigment

stones With cholesterol

stones

5

3/2

"lntra, present in intrahepatic bile duct only; intra and extra, present in both intrahepatic and extrahepatic ducts; L, present in left lobe only; LR, present in both lobes; R, present in right lobe only. The above classification of stones was performed according to the criteria described by Nakayama. 15 bFindfngs based on macroscopic examination.

Materials and Methods Subjects and Sample Collections The study was performed from 1989 to 1994 in cooperation with the study group for the National Survey Project of Intrahepatic Calculi under the Ministry of Health and Welfare in Japan. The protocol for this study was approved by the official committee of this survey. The study procedures were in accordance with the ethical standards of the Helsinki Declaration. Thirty-nine patients with intrahepatic calculi, 31 patients with only one of the two hepatic lobes affected by brown pigment stones (calcium bilirubinate stones), and 8 patients with both lobes affected by brown pigment stones were included in the study. In addition, 5 patients who had only one of the two lobes affected by pure cholesterol stones were included. Data of patients with intrahepatic calculi are summarized in Table 1. The classification of intrahepatic gallstones was based on the criteria described by Nakayama. 15 The pigment stones were classified according to the proceedings of the first National Institutes of Health-International Workshop on Pigment Gallstone Disease) 6 The number of patients with gallstones and the number and sources of biological specimens such as gallstones, blood, bile, and liver obtained from each type of patient with gallstones are summarized in Table 2. Control subjects were mostly patients with adenomatous gallbladder polyps, and the rest were patients with early gastric cancer who underwent gastrectomy. No member of either group of control subjects or those of gallstone patients was grossly obese (relative body weight of < 115 %) or had evidence of hepatic, intestinal, or renal disease, diabetes mellitus, or thyroid dysfunction. None had taken lipid-lowering drugs, bile acids, or any hormones within the last 4 weeks before the sample collections. To avoid the influence of dietary intake of cholesterol, all patients included in the study were admitted to the hospitals at least a week before their procedures and were given a regular diet (2125 kcal/day, containing 82.2 g protein, 56.7 g lipids including 0.2 g cholesterol, and 315 g carbohydrate).

The gallstones were washed with distilled water, dried to a constant weight at room temperature, and ground to powder. The composition of the gallstones, including cholesterol, calcittm bilirubinate, and fatty acid-calcium, was measured by infrared analysis. Iv Cholesterol stones were defined as stones containing more than 70% of cholesterol of total weight.

Reference Compounds Authentic standards of lithocholate, deoxycholate, chenodeoxycholate, cholate and their corresponding glycine and taurine conjugates were obtained from Steraloids Inc. (Wilton, NH). Nordeoxycholate (24 -nor- 30~, 12 o~-dihydroxy- 5 ~-cholan23-oic acid) and 7o~-hydroxycholesterol were also obtained from Steraloids. DL-Mevalonolactone was obtained from Sigma Chemical Co. (St. Louis, MO). 7o~-Hydroxy-4-cholesten-3-one was enzymatically synthesized from 7o~-hydroxycholesterol using cholesterol oxidase. I8 Ursodeoxycholate and its glycine and taurine conjugates were kindly supplied by Tokyo Tanabe Co. (Tokyo, Japan). Deuterated [25,26,26,26,27,27,27-2Hv]cho lesterol was obtained from MSD Isotopes (Merck Co. Inc., Rahway, NJ). Deuterated compounds, DL-mevalonolactone, and 7o~-hydroxy-4-cholesten-3-one were synthesized as previously described. 19 The purities were checked by thin-layer chromatography, and each compound gave only a single spot.

Chemicals All solvents were of analytical grade and were redistilled. The cyanopropyl-bonded silica (Bond-Elut CN, 500 mg), quaternary amine-bonded silica (Bond-Elut SAX, 500 rag), octadecylsilane-bonded silica (Bond-Elut CIS, 1000 mg), unbonded silica (Bond-Elut SI, 500 rag), and aminopropylbonded silica (Bond-Elut NH2, 100 mg) cartridges were obtained from Varian Co. (Harbor City, CA). Sephadex LH20 was obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). Glucose-6-phosphate, glucose-6-phosphate dehydrogenase (EC 1.1.1.49), and reduced nicotinamide adenine dinucleotide phosphate (NADPH) were obtained from Oriental Yeast (Tokyo, Japan). Dithiothreitol (DTT) and colorimetric enzymatic assay kits (Cholesterol E-Test Wako, Phospholipid

1536

SHODA ETAL.

GASTROENTEROLOGY Vol. 108, No, 5

Table 2. Data of Biological S p e c i m e n s From Each Type of Patient With Gallstones Bile

Controls Patients with Gallbladder cholesterol stones Gallbladder brown pigment stones Common bile duct brown pigment stones

n

Liver tissue

Gallbladder

Hepatic

Gallstones

Peripheral

Portal

25

19

19

10

--

25

9

44

29

29

13

--

44

18

13

13

5

--

18

11

--

--

Affected/unaffected a Intrahepatic brown pigment stones Intrahepatic cholesterol stones

Blood

39 5

11/11 b 2/2

--

11

Affected/unaffected °

---

21/24 d 3/5

21

39

5

5

5

2

aAffected, hepatic lobes affected by gallstones; unaffected, unaffected lobes. bNine paired liver specimens from affected and unaffected hepatic lobes of 9 patients, 2 specimens from affected lobes of 2 patients, and 2 specimens from unaffected lobes of 2 patients were assayed. CAffected, bile from the hepatic ducts from lobes affected by gallstones; unaffected, bile from the ducts from unaffected lobes. ~wenty-one paired bile specimens from the respective hepatic ducts from lobes affected by gallstones and unaffected lobes, and 3 bile specimens from the ducts from unaffected lobes were assayed.

C-Test Wako, and Total Bile Acids-Test Wako) were obtained from Wako Pure Chemical Industries Ltd. (Osaka, Japan). Dimethylethylsilyl imidazole was obtained from Tokyo Kasei Kogyo (Tokyo, Japan).

Preparation of Liver Microsomes Liver microsomes were prepared from fresh liver tissue in all cases, because 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase activities are lower in microsomes from frozen liver tissue than in those prepared from fresh tissue. 2° Liver tissue, 1 0 0 - 1 5 0 mg, was placed in 9 volumes (wt/vol) of ice-cold 50 mmol/L Tris-HC1 buffer (pH 7.4) containing 0.3 mol/L sucrose, 10 mmol/L DTT, 10 mmol/L ethylenediaminetetraacetic acid (EDTA), and 50 mmol/L NaF, and the microsomal fraction was prepared as described previously] 1 The microsomes were isolated in NaF to maintain the in vivo state of phosphorylation. The resulting microsomal fraction was suspended in 50 mmol/L Tris-HC1 buffer (pH 7.4) containing 0.3 mol/L sucrose, 5 mmol/L DTT, and 1 mmol/L EDTA. The microsomal protein contents were determined by the method described by BradfordY

Assay of HMG-CoA Reductase and Cholesterol 7(x-Hydroxylase Activities in Liver Microsomes Microsomal fractions ( 5 0 - 2 0 0 btg of protein) were preincubated for 10 minutes at 37°C in a total volume of 225 btL containing 50 mmol/L Tris-HC1 buffer (pH 7.4), 0.3 moll L sucrose, 5 mmol/L DTT, 1 mmol/L of EDTA, 12 mmol/L glucose-6-phosphate, and 0.1 mmol/L [2H3]HMG-CoA. The

assay was then initiated by the addition of 1 U glucose-6phosphate dehydrogenase and 750 nmol N A D P H dissolved in 225 btL of preincubated buffer, with the incubation mixture having a total N A D P H concentration of 3 mmol/L. The incubation was performed at 37°C for 15 minutes and was stopped by the addition of 50 btL of 1 mmol/L NaOH. The purification steps for the incubation mixture and the simultaneous assay by gas-liquid chromatography-mass spectrometry were performed as described previously. 21

RNA Isolation and Complementary DNA Synthesis Total RNA was isolated from the liver biopsy specimens as described previously. 23 First-strand complementary DNAs (cDNAs) were synthesized by the random primer method using a cDNA synthesis kit (Boehringer Mannheim Yamanouchi, Tokyo, Japan).

Reverse-Transcription Polymerase Chain Reaction Polymerase chain reaction amplification was performed using 25 pmol each primer, 1 ~tL eDNA template, 1.5 U Taq polymerase, 20 mmol/L deoxynucleoside triphosphates, and 1 × reaction buffer (10 mmol/L Tris-HC1, pH 8.3, containing 50 mmol/L KC1, 1.5 mmol/L MgC12, and 0.01% gelatin) in a 50-I-tL reaction volume. 24 The reactions were subjected to each cycle (low-density lipoprotein [LDL] receptor, 25; HMGCoA reductase, 24; cholesterol 7~-hydroxylase, 30; and [~actin, 20) at 94°C for 1 minute, 55°C for 1 minute, and 72°C for 1 minute. For Southern hybridization, the amplified

May 1995

PRIMARY DUAL DEFECT IN INTRAHEPATICCALCULI 1537

Table 3. Primers and Probes Specific for LDL Receptor, HMG-CoA Reductase, Cholesterol 7c(-Hydroxylase, and [3-Actin LDL receptor Sense Antisense

Probe

HMG-CoAreductase

Cholesterol 7~-hydroxylase

5'-TGT/CTG/TCA/CCT/GCA/ 5'-GCA/GCA/CAG/AAT/GI-r/ 5'-GCA/TCA/TAG/CTC/TTT/ AAT/CC-3' GGT/AG-3' ACC/CAC-3' 5'-GAA/CAC/GTA/AAG/ACC/ 5'-GTC/CTT/TAG/AAC/CCA/ 5'-GGT/GT]/CTG/CAG/CAG/ CCT/ACA-3' ATG/CC-3' TCC/TGT/AAT-3' 5'-ACG/AGC/AAG/GCT/GTC/ 5'-CTA/GGT/G'i-I/CAA/GGA/ 5'-CTA/CAA/TAT/GTC/CTG/ CCC/CCA/AGA/CGT/GCT-3' GCA/TGC/AAA/GAT/AAT-3' GAA/GAG/AGC/TTA/TAG-3'

products were electrophoresed on a 3% agarose gel and transferred onto a nylon membrane, then hybridized using the 5'32p-labeled oligonucleotide probe. In experiments involving quantitative assessment of the hybridization signal, the relative densities of autoradiograms were determined by densitometric scanning on a FAST SCAN 300A computing densitometer (Molecular Dynamics, Sunnyvale, CA). The densitometrically measured abundance of messenger RNA (mRNA) was corrected by the recovery of ~-actin mRNA, a widely distributed hepatic protein, and the data were expressed relative to the amounts of ~-actin m R N A present in each specimen. Polymerase chain reaction primers and oligonucleotide probes were designed in the same way as in previous reports describing cDNA sequences for LDL receptor, x5 HMG-CoA teductase, 26 and cholesterol 70~-hydroxylase 2v and then synthesized using an Applied Biosystems D1NA synthesizer (model 380A; Applied Biosystems Inc., Foster City, CA). The oligonucleotide panel includes the primers and probes specific for ~actin, LDL receptor, HMG-CoA reductase, and cholesterol 70~hydroxylase gene (Table 3).

Assay of Cytosolic Concentrations of Estrogen and Progesterone Receptors in Liver The cytosolic concentrations of estrogen and progesterone receptors were determined by the method of Tominaga et al. 28 using monoclonal antibodies.

Analysis of Cholesterol Concentrations in Liver Microsomes The hepatic cholesterol concentrations were determined using the method of Schaffer et al. 29 with minor modifications. Deuterated cholesterol and chloroform-methanol (2:1, vol/vol) were added to 10 [.tL liver microsomes. The chloroform phase was then evaporated, and the residue was either hydrolyzed with 0.5 mol/L KOH, extracted with nhexane, and converted into dimethylethylsilyl ether or directly converted into dimethylethylsilyl ether before analysis by gasliquid chromatography-mass spectrometry.

13-Actin 5'-CTT/CTA/CAA/TGA/ GCT/GCG/TG-3' 5'-TCA/TGA/GGT/AGT/ CAG/TCA/GG-3'

after eliminating the bilirubin by the use of a Bond-Elut NH2 cartridge. 32 The total bile acid concentration was determined by an enzymatic method using Total Bile Acids-Test Wako. 33 The cholesterol saturation of bile was calculated according to the critical tables for cholesterol saturation based on the total lipid concentration. 34 The bile acids present in bile were initially extracted by hot ethanol (at 60°C) and then purified and determined in the same way as described previously. 14 Nordeoxycholate was used as an internal standard.

Analysis of Mevalonate, 7et-Hydroxy-4cholesten-3-one, and Bile Acids in Plasma Plasma level of mevalonate was determined by a highly sensitive and specific method using an isotope dilution procedure as described previously. 35 Plasma level of 70~-hydroxy-4cholesten-3-one was also determined by a highly sensitive and specific method using an isotope dilution procedure as described previously. ~9 Bile acids in plasma were determined using high-performance liquid chromatography according to the method of Hirano et al. 36

Statistical Analysis Values are given as means + SEM. The statistical significance of differences in the values between different groups was evaluated with the M a n n - W h i t n e y test (two-tailed test). Correlation was tested (two-tailed test) by calculating Spearman's rank-order correlation coefficient (r). A P value of <0.05 was defined as statistically significant.

Results Chemical Composition of Intrahepatic Calculi Intrahepatic brown p i g m e n t stones were composed of significantly more cholesterol (P < 0.01) and less calcium bilirubinate (P < 0.005) than were brown p i g m e n t stones in the extrahepatic bile ducts. Intrahepatic pure cholesterol stones were composed of more than 99% cholesterol (Table 4).

Analysis of Biliary kipids The concentrations of cholesterol and phospholipids were determined by the enzymatic methods using Cholesterol E-Test Wako 3° and Phospholipid C-Test Wako, 31 respectively,

Biliary Lipid and Bile Acid Composition Hepatic bile from patients with intrahepatic calculi (both intrahepatic brown p i g m e n t stones and intra-

1538

SHODA ET AL.

GASTROENTEROLOGY Vol. 108, No. 5

Table 4. Composition of Brown Pigment Stones in Intrahepatic and Extrahepatic Bile Ducts Measured by Infrared Analysis

Intrahepatic brown pigment stones Intrahepatic cholesterol stones Extrahepatic brown pigment stones

n

Cholesterol

Calcium bilirubinate

Fatty acidcalcium

21

43.8 + 5.5 ~

31.4 _+ 4.5 b

25.0 +_ 4.1

>99

ND

ND

19.1 _+ 1.5

55.3 + 3.7

24.1 _+ 3.3

5 11

NOTE. Values are given as means _+ SEM. ND, nondetectable. "P < 0.05, bp < 0.0005; significantly different compared with extrahepatic brown pigment stones,

hepatic pure cholesterol stones) showed a marked increase in the molar percentage of cholesterol and a decrease in the molar percentage of bile acids compared with control subjects. The magnitude of these changes, which was larger in bile from affected hepatic lobes than in bile from unaffected ones, was even greater in patients with intrahepatic calculi than in those with cholesterol stones in the gallbladder. Hepatic bile of intrahepatic brown pigment stones had a decreased percentage of deoxycholate, which is a major bacterial metabolite of bile acid in the intestine, and, concomitantly, an increased percentage of chenodeoxycholate on both affected and unaffected sides compared with control subjects and patients with gallbladder stones. However, in the 5 cases of intrahepatic pure cholesterol stones, the percentage of deoxycholate in hepatic bile was not significantly decreased. Interestingly, the magnitude of cholesterol supersaturation of gallbladder bile correlated significantly with that of the percentage of deoxycholate in biliary bile acids (r = 0.49, P = 0.0003; n = 61) in control subjects and patients with gallbladder stones, but the supersaturation of hepatic bile from patients with intrahepatic calculi was substantially higher despite the decreased percentage of deoxycholate (Table 5).

Cholesterol Concentrations in Liver Microsomes The total and free cholesterol concentrations were significantly higher in the liver microsomes from both affected (P < 0.0001) and unaffected hepatic lobes (P < 0.0001) of patients with intrahepatic brown pigment stones than in those from control subjects and from patients with cholesterol stones (Figure 1). However, there was no significant difference in cholesterol concentration between affected and unaffected lobes. Also, in the 2 cases of intrahepatic pure cholesterol stones, the total and free cholesterol concentrations were higher in the

liver microsomes from both affected and unaffected lobes than in those from control subjects and from patients with cholesterol stones. In contrast, in patients with cholesterol stones, a significant increase was found not in the total and free cholesterol fractions but in the esterified fraction (P < 0.05).

HMG-CoA Reductase and Cholesterol 7¢xHydroxylase Activities in Liver Microsomes The catalytic activities of HMG-CoA reductase in liver microsomes are shown in Figure 2. Hepatic HMGCoA reductase activity was markedly greater in the microsomes from both affected (P < 0.0001) and unaffected hepatic lobes (P < 0.0001) of patients with intrahepatic brown pigment stones than in those from control subjects, patients with cholesterol stones, and patients with brown pigment stones. There was no significant difference in the activity between affected and unaffected lobes. Furthermore, a marked increase in activity was also found in the liver microsomes of the 2 cases of intrahepatic pure cholesterol stones. The catalytic activity of HMGCoA reductase in liver microsomes correlated inversely with the percentage of deoxycholate in biliary bile acids in control subjects and patients with gallstones, including those with intrahepatic calculi (both intrahepatic brown pigment stones and pure cholesterol stones) (r = 0.74, P = 0.0001; n = 87) (Table 5 and Figure 2). In contrast, the catalytic activity of cholesterol 70~hydroxylase (Figure 3) was found to be significantly lower in the liver microsomes from both affected (P < 0.0001) and unaffected lobes (P < 0.0001) of patients with intrahepatic brown pigment stones than in those from control subjects, patients with cholesterol stones, and patients with brown pigment stones. There was no significant difference in activity between affected and unaffected lobes. In the 2 cases of intrahepatic pure cholesterol stones, a marked decrease in activity was also found in the liver microsomes. There was no significant correlation between cholesterol 7o~-hydroxylase activity and percentage of deoxycholate in biliary bile acids in control subjects and patients with gallstones, including those with intrahepatic calculi (both intrahepatic brown pigment stones and pure cholesterol stones) (r = 0.17, P = 0.12; n = 87) (Table 5 and Figure 3).

Quantification of mRNA of LDL Receptor, HMG-CoA Reductase, and Cholesterol 7¢xHydroxylase in the Liver Parallel to the increase of HMG-CoA reductase activity, steady-state levels of hepatic LDL receptor and HMG-CoA reductase m R N A increased in patients with

May 1995

PRIMARY DUAL DEFECT IN INTRAHEPATICCALCULI 1539

Table 5. Biliary Lipid and Bile Acid Composition in Gallbladder and Hepatic Bile

Patients Controls Gallbladder cholesterol stones Gallbladder brown pigment stones

Group

n

Cholesterol

Bile acids

Phospholipids

Total biliary lipid concentration

(molar %)

(molar %)

(molar %)

(g/dL)

Cholesterol saturation

C

CDC

(%)

(%)

(%)

DC (%)

GB HB

19 10

7.3 ± 0.4 12.1 + 1,1

72.4 ± 1.2 65.7 ± 1.3

20.7 ± 0.9 21.2 ± 0.9

12.8 + 2.0 1.9 ± 0.2

98 + 3 245 ± 12

3 7 . 1 _+ 2.1 39.5 ± 2.2

47.6 ± 2.5 50.1 ± 3,2

10.3 z 0.2 10.8 ± 1.2

GB HB

29 13

10.5 ± 0,6 f 16.1 ± 0.7 ~

70.3 ± 1,7 64.8 + 2,1

19.4 ± 1.4 19.1 _~ 1.6

11.0 + 1.2 2.3 ± 0,3

159 + 6 b 313 ± 17 c

2 9 . 4 ± 1.6 e 29.0 ± 1.1 f

4 5 . 8 ± 1,6 46.7 _+ 2.4

22.1 _+ 1.6 ~ 21.6 ± 2.5 °

GB HB

13 5

6.8 ± 0.7 11.7 ± 2.7

73.4 ± 1,9 68.1 _+ 3,5

19.7 + 1.5 20.2 +_ 1.3

9.8 ± 0.8 1.2 ± 0.4

112 ± 6" 255 ± 24

31.1 ± 2,6 29.0 ± 1.3 ~

4 5 . 5 ± 2.3 48.9 ± 2.0

19.0 ± 1,8 ~ 17.0 ± 2.63

HB (+) HB ( )

21 24

20,5 + 1,1 ~] 17.9 + 0.7b]

54,8 ± 2.8 a] / 6 1 . 8 ± 1.2 aj

24.3 ± 2.0 20.2 ± 1.0

0.4 ± 0.1 ° 0.7 ± 0.2 °

458 ± 44 el / 356 ± 25 ~]

31.4 + 1.6 c 37,0 ± 2.3

55.3 _+ 2.4 51.2 ± 2.3

5.6 ± 1.1 ~ 6.7 i 1.1 ~

HB (+) HB ( )

3 5

24.8 ± 5.5 e 17.3 ± 1,6 ~

50.0 + 6.2 ~ 56.5 + 3,2 ~

25.0 + 1.9 24.7 ± 1.5

0.6 + 0.1 ° 0.8 + 0 . 0 4 c

625 + 122 ~ 520 + 76 r

36.3 ± 3.6 4 2 . 3 ± 4.1

47.8 ± 6.4 43.7 ± 3.8

Intrahepatic brown pigment stones Intraheptic cholesterol stones

8.6 ± 1.9 10.5 ± 1.7

NOTE. Values are given as means ± SEM. C, cholate; CDC, chenodeoxycholate; DC, deoxycholate; GB, gallbladder bile; HB, hepatic bile; (+), from affected lobes; ( - ) , from unaffected lobes. Brackets indicate P < 0.05. ap < 0.05, bp < 0.0001, cp < 0.01, ap < 0.001, ep < 0.005, ~P < 0.0005; significantly different from control subjects.

intrahepatic brown pigment stones for both affected and unaffected hepatic lobes despite the greater cholesterol concentrations in liver microsomes compared with control subjects (P < 0.05) and patients with cholesterol stones (P < 0.05) (Table 6 and Figure 4). In contrast, steady-state level of cholesterol 7o~-hydroxylase m R N A was substantially lower in patients with intrahepatic brown pigment stones for both lobes than in control subjects (P < 0.05) and patients with cholesterol stones (P < 0.05). The resulting ratio o f m R N A level for HMGCoA reductase/cholesterol 7oc-hydroxylase was significantly higher in patients with intrahepatic brown pigment stones than in control subjects (P < 0.0001) and patients with cholesterol stones (P < 0.0001). There was no significant difference in steady-state m R N A levels for the receptor and the two enzymes between patients with intrahepatic brown pigment stones vs. pure cholesterol stones. Cytosolic Concentrations of Estrogen and Progesterone Receptors in Liver

Only a few patients were positive for cytosolic progesterone receptor in the liver, the concentration of which was <0.1 fmol/mg protein, but all control subjects and patients with gallstones were positive for cytosolic estrogen receptor in the liver. The estrogen receptor concentration in liver cytosols was found to be significantly higher in patients with intrahepatic brown pigment stones for both affected and unaffected hepatic lobes than in control subjects (P < 0.05), patients with cholesterol stones (P < 0.0001), and patients with brown pig-

ment stones (P < 0.0001), but there was no significant difference in estrogen receptor concentration between affected and unaffected lobes. The estrogen concentration was even higher in the 2 cases of intrahepatic pure cholesterol stones for both hepatic lobes than in control subjects and patients with gallbladder stones. In contrast, the estrogen receptor concentration was significantly lower in patients with cholesterol stones (P < 0.0001) and those with brown pigment stones (P < 0.01) than in control subjects. When the degree of catalytic activity of HMG-CoA reductase in liver microsomes (Figure 2) was compared with that of estrogen receptor concentration in liver cytosols (Table 7), there was a significant correlation between them in control subjects and patients with gallstones (r = 0.67, P = 0.0001; n = 87). Mevalonate and 7e~-Hydroxy-4-cholesten-aone Levels in Peripheral Blood

Mevalonate concentration in plasma, which was found to correlate closely with HMG-CoA reductase activity (Figure 2) for control subjects and patients with gallstones (r = 0.66, P = 0.0001; n = 87), was significantly higher in patients with intrahepatic calculi (both intrahepatic brown pigment stones and pure cholesterol stones) than in control subjects (P < 0.05), whereas patients with cholesterol stones had a significant decrease in mevalonate concentration. Furthermore, 7o~-hydroxy4-cholesten-3-one concentration in plasma, which was found to correlate closely with cholesterol 70~-hydroxylase activity (Figure 3) for control subjects and patients with gallstones (r = 0.58, P = 0.0001; n = 87), was

1540

SHODA ET AL.

GASTROENTEROLOGY Vol. 108, No. 5

significantly lower in patients with intrahepatic calculi (both intrahepatic brown pigment stones and pure cholesterol stones) than in control subjects (P < 0.0001), patients with cholesterol stones (P < 0.0001), and patients with brown pigment stones (P < 0.0001). However, there was no significant difference in plasma 70~hydroxy-4-cholesten-3-one concentration between control subjects and patients with gallbladder stones.

Bile Acids in Portal Blood Total bile acid concentration in portal blood was significantly lower in patients with intrahepatic brown pigment stones than in control subjects (P < 0.01) (Table 8). For bile acid composition, the percentage of deoxycholate in portal bile acids was significantly lower in patients with intrahepatic brown pigment stones than in Q

~NS/." e-

control subjects (P < 0.0001). However, in the 2 cases of intrahepatic brown pigment stones, the percentage of deoxycholate was not significantly different from that of control subjects.

Discussion Cholesterol content in gallstones was much higher and bile acid content was much lower in intrahepatic brown pigment stones than in extrahepatic brown pigment stones (Table 4). This was found parallel to the changes in biliaty lipid composition of hepatic bile from patients with intrahepatic calculi (Table 5), as reported by us previously. 9'14 The present results imply that, in intrahepatic calculi, altered cholesterogenesis and bile acid synthesis in the liver may be one of the etiologic factors responsible for the formation of hepatic bile substantially supersaturated with cholesterol, which leads to subsequent formation of cholesterol-rich gallstones in the intrahepatic bile ducts. The major new findings are that (1) de novo synthesis of cholesterol is up-regulated significantly in both affected and unaffected lobes of the livers of patients with intrahepatic calculi irrespective of whether the stone category is brown pigment or pure cholesterol stones and despite the major increase of total and free cholesterol concentrations in the liver micro-

& -6

NS

E

i

i

Lo

CD

Controls

Gallbladder

Intrahepatic

cholesterol stones brown pigment stones

Intrahepatic cholesterol stones

Figure 1. Total, free, and esterified cholesterol concentrations in the liver microsomes from control subjects and patients with gallstones. Control subjects, n = 19; patients with gallbladder cholesterol stones, n = 29; patients with intrahepatic brown pigment stones, n = 13; patients with intrahepatic pure cholesterol stones, n = 2. +, hepatic lobes affected by intrahepatic gallstones; - , unaffected lobes. Nine paired liver specimens from affected and unaffected hepatic lobes of 9 patients, 2 specimens from affected lobes of 2 patients, and 2 specimens from unaffected lobes of 2 patients were assayed for microsomal cholesterol concentrations. C], free cholesterol fraction; ~, esterified cholesterol fraction. The total, free, and esterified cholesterol concentrations (nmol/mg protein; mean _+ SEM) in liver microsomes were 91.6 + 5.3, 76.2 _+ 4.6, and 15.4 + 1.8 for control subjects; 106.9 _+ 4.5, 82.4 _+ 3.7, and 24.5 + 2.4 for patients with gallbladder cholesterol stones; 166.2 _+ 16.3, 142.0 + 12.4, and 24.2 _+ 6.8 for affected hepatic lobes; 179.2 _+ 21.2,148.0 + 17.5, and 31.2 + 6.2 for unaffected lobes of patients with intrahepatic brown pigment stones; and 124.4 (mean), 116.0, and 8.4 for affected lobes and 136.0, 107.2, and 28.8 for unaffected lobes of patients with intrahepatic pure cholesterol stones. *P < 0.05; * * P < 0.01; * * * P < 0.005; * * * * P < 0.0001.

Controls

Gallbladder

Gallbladder

cholesterol stones

pigmentstones

Intrahepatic

Intrahepatic

brownpigment stones cholesterol stones

Figure 2. HMG-CoA reductase activity (pmol. min -1- mg protein -1) in the liver microsomes from control subjects and patients with gallstones. ©, control subjects (n = 19); o, patients with gallbladder cholesterol stones (n = 29); &, patients with gallbladder brown pigment stones (n = 13); m, patients with intrahepatic brown pigment stones (n = 13); 0 , patients with intrahepatic pure cholesterol stones (n = 2); +, hepatic lobes affected by intrahepatic gallstones; - , unaffected lobes. Nine paired liver specimens from affected and unaffected hepatic lobes of 9 patients, 2 specimens from affected lobes of 2 patients, and 2 specimens from unaffected lobes of 2 patients were assayed for the activity. The catalytic activity of HMG-CoA reductase in liver microsomes was 6.4 _+ 0.4 pmol. rain -~. mg protein -~ (mean _+ SEM) for controls; 2.1 _+ 0.1 for gallbladder cholesterol stones; 3.0 + 0.4 for gallbladder brown pigment stones, 11.2 +_ 1.3 for affected hepatic lobes, and 13.4 _+ 1.7 for unaffected lobes of patients with intrahepatic brown pigment stones; and 15.8 (mean) for affected lobes and 15.2 for unaffected lobes of patients with intrahepatic pure cholesterol stones. *P < 0.0001.

May 1995

PRIMARY DUAL DEFECT IN INTRAHEPATIC CALCULI

somes; (2) bile acid synthesis, which is involved in catabolism of cholesterol, is down-regulated significantly in both lobes of the liver; and (3) estrogen receptor concentration in liver cytosols, which parallels HMG-CoA reductase activity, is significantly higher in patients with intrahepatic calculi in both lobes of the liver than in control subjects and patients with gallbladder stones. Biliary cholesterol output or its relative amounts in bile may be determined by the rate of hepatic de novo synthesis of cholesterol. 37-4° There is controversy as to whether catalytic activity of HMG-CoA reductase, the rate-limiting enzyme for cholesterol de novo synthesis, is increased 37-4° or unchanged 4.'42 in patients with gallbladder cholesterol stones. In patients with intrahepatic calculi, however, the present results in terms of the activity and steady-state m R N A level of hepatic HMG-CoA reductase and the concentration of plasma mevalonate indicate that de novo synthesis of cholesterol was upregulated in the liver, irrespective of whether the stone category is brown pigment or pure cholesterol stones. Furthermore, the higher level of hepatic LDL-receptor m R N A indicated an increase in hepatic uptake of lipoprotein cholesterol via the receptor-mediated pathway. Possibly reflecting the up-regulated hepatic synthesis and uptake of cholesterol, the concentrations of total and free cholesterol in liver microsomes were considerably increased in both affected and unaffected hepatic lobes of patients with intrahepatic calculi (Figure 1). The marked elevation in relative amounts of cholesterol in bile from

J.

.C_

i

t

J=

& .g -6

Controls

Gallbladder cholesterd stones

Gallbladder pigment s%ones

Intr~hepatic brown pib~rr~ stones

fntrahepatic cholesterol stones

Figure 3. Cholesterol 7c(-hydroxylase activity (pmol. min -1. mg protein -1) in the liver microsomes from control subjects and gallstone patients. For symbols and number of patients, see Figure 2. Nine paired liver specimens from affected and unaffected hepatic lobes of 9 patients, 2 specimens from affected lobes from 2 patients, and 2 specimens from unaffected lobes of 2 patients were assayed for the activity. The catalytic activity of cholesterol 7c<-hydroxylase in liver microsomes was 6.0 ± 0.6 pmol • min -1- mg protein -~ (mean ± SEM) for controls; 5.1 + 0.5 for gallbladder cholesterol stones; 6.6 ± 0.9 for gallbladder brown pigment stones; 2.8 + 0.5 for affected hepatic lobes and 2.6 _+ 0.5 for unaffected lobes of patients with intrahepatic brown pigment stones; and 1.3 (mean) for affected lobes and 1.6 for unaffected lobes of those with intrahepatic pure cholesterol stones. • P < 0.005; * * P < 0.0001.

1541

both affected and unaffected lobes could be explained by an excessive flux of cholesterol directed into bile, the source of which was derived from up-regulated cholesterogenesis in the liver. Therefore, the elevated cholesterol concentrations in the liver may be a factor in promoting the flux of cholesterol into the bile. 37 Of particular interest, there was no significant difference in the activity and m R N A level of HMG-CoA reductase between affected and unaffected lobes of the liver; this strongly suggests that the pathogenesis of intrahepatic calculi is not necessarily due to local disturbances, such as bile stasis resuiting from bile duct strictures and/or dilatations of bile ducts or bacterial infection, but rather to primary alterations of hepatic cholesterol metabolism. The details of the initiating defect causing the metabolic changes remain unknown. Sex hormones, especially estrogen, which produces pronounced changes in hepatic cholesterogenesis by enhancing HMG-CoA reductase activity43 and LDL-receptot binding, 43 could be partly responsible for the higher prevalence of intrahepatic calculi in women than in men. I2 It is interesting that the estrogen receptor concen-

Table 6. Relative mRNA Levels f o r LDL Receptor, HMG-CoA R e d u c t a s e , and C h o l e s t e r o l 7~t-Hydroxylase in Liver

Patients

n

LDL receptors

HMG-CoA reductase

Controls Gallbladder cholesterol stones DC <15% DC > 1 5 % Intrahepatic brown pigment stones + Intrahepatic cholesterol stones + -

11

0.9 + 0.2

0.3 _+ 0.06

1.1 + 0.5

12 5 7

0.8 + 0,1 1.2 + 0.2 b 0.6 _+ 0.2 b

0.2 _+ 0.03 a 0.2 ± 0.04 0.1 + 0.04

1.7 + 0.8 0.1 ± 0.02 2.9 + 1.2

12 6 6

2.1 + 0.5 a 1.8 ± 0.6 2.5 _+ 0.9

0.8 + 0.2 a 0.8 ± 0.4 0.7 _+ 0.1

0.1 + 0.01 a 0.1 ± 0.02 0.1 ± 0.01

2.6 _+ 1.4 1.6 3.6

0.4 ± 0.1 0.4 0.6

4 2 2

Cholesterol 7c~-hydroxylase

0.1 ± 0.2 0.04 0.09

NOTE. Values are given as means _+ SEM in relative units. The ratio of mRNA levels for HMG-CoA reductase/cholesterol 7c~-hydroxylase was estimated as 1.5 + 0.6 (mean _+ SEM) in control subjects, 0.8 _+ 0.3 in patients with gallbladder cholesterol stones, 14.5 ± 1.2 in patients with intrahepatic brown pigment stones, and 4.2 _+ 0.9 in patients with intrahepatic cholesterol stones. DC <15%, subjects with 15% of deoxycholate in biliary bile acids; DC >15%, subjects with >15% of deoxycholate; +, affected lobes by gallstones; - , unaffected lobes. aSignificantiy different from control subjects; P < 0.05. bSignificantly different between the two groups; P < 0.05.

1542

SHODA ET AL.

I

GASTROENTEROLOGY Vol. 108, No. 5

II

III

IV

+--

-.!-- --

Table 7, Cytosolic Concentrations of Estrogen and Progesterone Receptors in Liver

bp A 1357

603 310 - 234

Progesterone

(mol. mg -I. protein ~)

Patients

n

Controls Gallbladder cholesterol stones Gallbladder pigment stones Intrahepatic brown pigment stones +

19

9.6 4- 0.8

<0.1

29

5.5 4- 0.5 a

<0.1

13

6.2 +_ 0.9 b

<0.1

22 11 11

12.3 4- 0.8 ~ 11.8 + 1.0 12.9 + 1.2

<0.1 <0.1 <0.1

4 2

11.6 +_ 0.5 11.6

<0.1 <0.1

2

11.5

<0.1

Intrahepatic cholesterol stones +

B

Estrogen

(mol. mg 1 protein -1)

-

NOTE. Values are given as means _+ SEM. +, lobes with gallstones; - , unaffected lobes. aP < 0.0001, bp < 0.01, cp < 0.05; significantly different from control subjects.

1357 ....

603 310 234

yt

C 1357-603 310 234

•

Figure 4, Poiymerase chain reaction-assisted amplifications of (A) LDL receptor, (B) HMG-CoA reductase, and (C) cholesterol 7(z-hydroxylase mRNA in the liver of control subjects and patients with gallstones. Lane/, a control subject; lane II, a patient with gallbladder cholesterol stones; lane III, paired affected (+) and unaffected ( - ) hepatic lobes of a patient with intrahepatic brown pigment stones, respectively; lane IV, paired affected (+) and unaffected ( - ) lobes of a patient with intrahepatic pure cholesterol stones, respectively.

CoA reductase in the patients. On the other hand, the lower estrogen receptor concentration in patients with gallbladder stones, coupled with the lower activity and mRNA level of hepatic HMG-CoA reductase, directs the liver to depress hepatic cholesterogenesis. The significant correlation of estrogen receptor concentration with HMG-CoA reductase activity shows that estrogen receptor concentration in liver cytosols modulates hepatic cholesterogenesis. Bile acid metabolism in patients with intrahepatic calculi has not been extensively studied. The present study shows, for the first time, depressed bile acid synthesis in the liver of patients with intrahepatic calculi, which was supported by the lower activity and mRNA level of cholesterol 7o~-hydroxylase (Figure 3 and Table 6) and the lower concentration of plasma 7o~-hydroxy-4-choles-

T a b l e 8. Bile Acids in Portal Blood

Controls (n=9)

Patients with intrahepatic brown pigment stones (n=5)

Patients with intrahepatic cholesterol stones (n=2)

Total bile acids

tration in liver cytosols was significantly higher in patients with intrahepatic calculi, irrespective of the stone category, than in control subjects and patients with gallbladder stones. This was compatible with the present results in terms of the mRNA level of hepatic LDL receptor and the activity and mRNA level of hepatic HMG-

(#mol/L) C (%) CDC (%) DC (%)

41.8 30.7 55.6 14.0

_+ 10.0 4- 2.7 4- 3.6 4- 2.9

4.5 27.2 69.0 3.8

_+ 0.8 a 4- 7.7 4- 6.8 _+ 1.0 ~

5.7 38.6 49.1 12.3

5.8 46.2 39.2 14.6

NOTE. Values are given as means _+ SEM. C, cholate; CDC, chenodeoxycholate; DC, deoxycholate. ap < 0.01, bp < 0.0001; significantly different from control subjects.

May 1995

PRIMARY DUAL DEFECT IN INTRAHEPATIC CALCULI

o

1543

g F r~***

* II

8.8

!

88

'.L ,~.

'

'

[

''

"*** I

888

~.L

1-

B

O

¢-

¢-

-

Q Controls

Gallbladder cholesterol stones

Gallbladder pigment stones

Intrahepatic brown pigment stones

Intrahepatic cholesterol stones

Controls

Gallbladder cholesterol stones

Gallbladder

Intrahepatic

pigment brownpigment stones

stones

Intrahepatic

cholesterol stones

Figure 5. Plasma concentrations (nmol/L) of (A) mevalonate and (B) 7c~-hydroxy-4-cholesten-3-one in control subjects and patients with gallstones. ©, control subjects (n = 25); O, patients with gallbladder cholesterol stones (n = 44); A, patients with gallbladder brown pigment stones (n = 18); i , patients with intrahepatic brown pigment stones (n = 39); 0 , patients with intrahepatic pure cholesterol stones (n = 5). The mevalonate concentration was 43.4 ± 1.2 (mean + SEM) nmol/L for control subjects, 28.2 ± 1.1 for patients with gallbladder cholesterol stones, 37.0 +_ 1.9 for patients with gallbladder brown pigment stones, 49.2 ± 1.9 for patients with intrahepatic calculi, 49.0 ± 2.1 for patients with intrahepatic brown pigment stones, and 50.9 _+ 3.5 for patients with intrahepatic pure cholesterol stones. The 7c~-hydroxy-4cholesten-3-one concentration was 12.4 ± 1.6 nmol/L (mean ± SEM) for control subjects, 15.4 +_ 1.3 for patients with gallbladder cholesterol stones, 16.5 + 1.8 for patients with gallbladder brown pigment stones, 6.2 _+ 0.8 for patients with intrahepatic calculi, 6.4 ± 0.9 for patients with intrahepatic brown pigment stones, and 4.8 ± 0.9 for patients with intrahepatic pure cholesterol stones. Horizontal bars indicate means within each group. *P < 0.05; * * P < 0.0005; * * * P < 0.0001.

ten-3-0ne (Figure 5). In intrahepatic calculi, the bile acid concentrations were significantly increased in the liver and decreased in the hepatic bile, on both affected and unaffected sides, than in the corresponding specimens from control subjects. 14 When the biliary dynamics of ~)gmTc-EDTA were studied in intrahepatic calculi, the time activity curves of 99mTc-EDTA were prolonged significantly in not only affected but in unaffected intrahepatic bile ducts as well compared with the curves in normal subjects, 44 suggesting that bile stasis may exist even in both hepatic lobes of patients with one of the two lobes affected by gallstones. Furthermore, the curves were prolonged in the left intrahepatic bile ducts longer than in the right ducts, even in normal subjects, 44 which may reflect the anatomic difference of the left hepatic duct coursing horizontally in relation to the common hepatic duct compared with the right hepatic duct 45 and could explain why involvements of the left hepatic lobe predominate. 2 Thus, in intrahepatic calculi, alterations in bile acid concentration and composition of the liver and bile specimens may be relevant to possible impairments of hepatic secretion and biliary transport of bile acids by unknown mechanism. Eventually, the increased hepatocyte bile acid levels could result in the changes in bile acid synthesis towards depression. There were definite differences in hepatic cholesterogenesis and bile acid synthesis between patients with intrahepatic calculi and those with gallbladder stones.

The group of intrahepatic calculi, irrespective of the stone category, represented an increased hepatic cholesterogenesis with a decreased bile acid synthesis, but the group of gallbladder stones represented a rather decreased cholesterogenesis with an unchanged bile acid synthesis. Some populations with gallbladder cholesterol stones (e.g., Native Americans) 46 have metabolic defects similar to those seen in patients with intrahepatic calculi; the present results do not exclude the possibility of such gallstone populations with an increased hepatic cholesterogenesis and a decreased bile acid synthesis. Deoxycholate deserves recognition as an important factor for supersaturated bile and subsequent stone formation in cholesterol gallstone disease in the gallbladder, 4v'48 because the expansion of deoxycholate pool nqay cause a hypersecretion of cholesterol into the bile. 49 In intrahepatic calculi, the hepatic bile was found substantially supersaturated with cholesterol, despite the significant decrease of deoxycholate in bile under bile static circumstances. Considering an inverse correlation between biliary percentage of deoxycholate and HMG-CoA reductase activity, relative amounts of deoxycholate in the total bile acid pool as well as cytosolic estrogen concentrations in the liver may be important factors modulating hepatic cholesterogenesis in men. This could be supported by the results on the regulation of bile acid feeding on de novo cholesterol synthesis. 5° Probably, the sources of cholesterol pool contributing to the biliary

1544

SHODA ET AL.

supersaturation are not be the same between intrahepatic and gallbladder stones. There are definite differences in relative prevalence of intrahepatic calculi between the West and the East 2 and in age-related incidence between intrahepatic and gallbladder brown pigment stones. 51'52 The incidence of intrahepatic calculi is higher in the younger generations compared with the incidence of gallbladder pigment calculi. 51'52 These differences imply that genetic factors, e.g., the specified types of HLA found in intrahepatic calculi, 53 possibly related with the observed metabolic defects, may be more closely involved in the pathogenesis of cholelithiasis in the intrahepatic bile ducts than in the gallbladder. It is of particular interest that the majority of intrahepatic calculi consists of brown pigment stones but few pure cholesterol stones in the intrahepatic bile ducts, despite the similar metabolic defects in hepatic cholesterol and bile acid metabolism. One of the reasons may be secondary bacterial infections in the affected intrahepatic bile ducts, which cause hydrolysis of bilirubin glucuronides leading to precipitation of calcium bilirubinates and subsequent formation of calcium bilirubinate stones, because colonization of bile ducts by enteric bacteria is frequent in intrahepatic calculi) 4 Thus, the presence of brown pigment as well as pure cholesterol stones among intrahepatic calculi, suggesting the complex nature of the problem, i.e., not only the formation and precipitation of calcium bilirubinate but also the solubility of cholesterol, should be considered when dealing with intrahepatic calculi. Further studies are needed to clarify the role of biliary pronucleating and/or antinucleating proteins both for cholesterol 55-5v and for calcium 58 in intrahepatic calculi. In summary, the present study concludes that (1) in intrahepatic calculi, the formation of hepatic bile substantially supersaturated with cholesterol may be attributed to both up-regulated cholesterogenesis and down-regulated bile acid synthesis in the liver; (2) these pathological changes found in both affected and unaffected lobes of the liver can be interpreted as being the primary dual defect; and (3) the pathogenetic mechanism responsible for the biliary cholesterol supersaturation and subsequent stone formation in patients with intrahepatic calculi may be different from that in patients with gallbladder cholesterol stones.

References 1. Nakayama F, Koga A. Hepatolithiasis: present status. World J Surg 1984;8:9-14. 2. Nakayama F, Soloway RD, Nakama T, Miyazaki K, Ichimiya H, Sheen PC, Ker CA, Ong GB, Choi TK, Boey J, Foong WC, Tan EC, Thung KH, Lee CN. Hepatolithiasis in East Asia: retrospective study. Dig Dis Sci 1986;31:21-26.

GASTROENTEROLOGY Vol. 108, No. 5

3. Nakayama F, Furusawa T, Nakama T. Hepatolithiasis in Japan: present status. Am J Surg 1980;139:216-220. 4. Nakayama F, Frusawa T, Nakama T. Results of questionnaire on intrahepatic gallstones in Japan. Jpn J Gastroenterol Surg 1978; 11:979-983, 5. Trotman BW, Ostrow JD, Soloway RD. Pigment vs cholesterol cholelithiasis: comparison of stone and bile composition. Am J Dig Dis 1974; 19:585-590. 6. Miyake H, Johnston CG. Etiological studies. Digestion 1968;1: 219-228. 7. Soloway RD, Trotman BW, Ostrow JD. Pigment gallstone. Gastroenterology 1977; 72:167-182. 8. Yamashita N, Yanagisawa J, Nakayama F. Composition of intrahepatic calculi--etiological significance. Dig Dis Sci 1988;33: 449-453. 9. Shoda J, Tanaka N, Matsuzaki Y, Honda A, Osuga T, Shigematsu N, Miyazaki H. Microanalysis of bile acid composition in intrahepatic calculi and its etiological significance. Gastroenterology 1991; 101:821-830. 10. Strichartz SD, Abedin MZ, Ippoliti AF, Drezin M, Roslyn JJ. Intrahepatic cholesterol stones: a rationale for dissolution therapy. Gastroenterology 1991; 100:228-232. 11. Ohta T, Nagakawa T, Takeda T, Fonseca L, Kanno M, Mori K, Kayahara M, Ueno K, Miyazaki I, Terada T. Histological evaluation of the intrahepatic biliary tree in intrahepatic cholesterol stones, including immunohistochemical staining against apoprotein A-I. Hepatology 1993; 17:531-537. 12. Nagase M, Hisaka Y, Soloway RD, Tanimura H, Setoyama M, Kato H. Gallstones in western Japan. Factors affecting the prevalence of intrahepatic gallstones. Gastroenterology 1980;78: 684-690. 13. Maki T. Pathogenesis of calcium bilirubinate gallstone: role of E. coli, ~-glucuronidase and coagulation by inorganic ions, polyelectrolytes and agitation. Ann Surg 1966; 164:90-100. 14. Shoda J, Tanaka N, He B, Matsuzaki Y, Osuga T, Miyazaki H. Alterations of bile acid composition in gallstones, bile, and liver of patients with hepatolithiasis, and their etiological significance. Dig Dig Sci 1993;38:2130-2141. 15. Nakayama F. Intrahepatic calculi: a special problem in East Asia. World J Surg 1982;6:802-804. 16. Trotman BW, Soloway RD. Pigment gallstone disease: summary of the National Institutes of Health-International Workshop. Hepatology 1982; 2:879-884. 17. Trotman BW, Morris TA, Sanchez HM, Sotoway RD, Ostrow JD. Pigment versus cholesterol cholelithiasis: identification and quantitation by infrared spectroscopy. Gastroenterology 1977; 72:495-498. 18. Brooks CJW, Cole WJ, Lawrie TDV, MacLachlan J, Borthwick JH, Barret GM. Selective reactions in the analytical characterization of steroids by gas chromatography-mass spectrometry. J Steroid Biochem 1983;19:189-201. 19. Yoshida T, Honda A, Matsuzaki Y, Tanaka N, Shoda J, Osuga T, Miyazaki T. Determination of 7c(-hydroxy-4-cholesten-3-one level in human plasma using isotope-dilution mass spectrometry and monitoring its circadian rhythm in humans as index of bile acid biosynthesis. J Chromatogr 1994;655:179-187. 20. Kwekkeboom J, Kempen HJ, van Voorthuizen EM, Griffionen M, Cohen LH. Postnatal developmental profile of 3-hydroxy-3-methylglutaryl-CoA reductase, squalene synthetase and cholesterol 7(zhydroxylase activities in the liver of domestic swine. Biochim Biophys Acta 1990;1042:146-149. 21. Honda A, Shoda J, Tanaka N, Matsuzaki Y, Osuga T, Shigematsu N, Tohma M, Miyazaki H. Simultaneous assay of the activities of two key enzymes in cholesterol and bile acid metabolism by gas chromatography-mass spectrometry. J Chromatogr 1991; 1565:53-66.

May 1995

22. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976;72:248-254. 23. Chomczynski P, Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanaic-phenol-chloroform extraction. Anal Biochem 1987;162:156-159. 24. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988; 239:487 -491. 25. Yamamoto T, Davis CG, Brown MS, Schneider WJ, Casey ML, Goldstein JL, Russell DW. The human LDL receptor: a cysteinerich protein with multiple Alu sequences in its mRNA. Cell 1984;39:27-38. 26. Luskey KL, Stevens B. Human 3-hydroxy-3-methylglutaryl coenzyme A reductase: conserved domains responsible for catalytic activity and sterol-regulated degradation. J Biol Chem 1985; 260:10271-10277. 27. Noshiro M, Okuda K. Molecular cloning and sequence analysis of cDNA encoding human cholesterol 7c~-hydroxylase. FEBS Lett 1990;268:137-140. 28. Tominaga T, Yoshida Y, Kitamura M, Kosaki G. Comparative studies of estrogen receptor determinations by enzyme immunoassay using the monoclonal antibody, dextran-coated charcoal and sucrose density gradient methods. Jpn J Cancer Chemother 1985; 12:1782-1786. 29. Schaffer R, Sniegoski LT, Welch MJ, White E, Cohen VA, Hertz HS, Mandel J, Paul TC, Svensson L, BjSrkhem I, Btomstrand R. Comparison of two isotope dilution mass spectrometric methods for determination of total serum cholesterol. Clin Chem 1982; 28:5-8. 30. Ailain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem 1974;20:470-475. 31. Takayama M, Itoh S, Nagasaki T, Tanimizu I. A new enzymatic method for determination of serum choline-containing phospholipids. Clin Chim Acta 1977; 79:93-98. 32. Aufenanger M, Kattermann R. Enzymatic determination of lipids in human bile without bilirubin interference: reliable assessment of the cholesterol saturation index (CSI). J Clin Chem Clin Biochem 1989;27:605-611. 33. Mashige F, Tanaka N, Maki A, Kamei S, Yamanaka M. Direct spectrophotometry of total bile acids in serum. Clin Chim Acta 1981; 27:1352-1356. 34. Carey M. Critical tables for calculating the cholesterol saturation of native bile. J Lipid Res 1978; 19:945-955. 35. Yoshida T, Honda A, Tanaka N, Matsuzaki Y, He B, Osuga T, Kobayashi N, Ozawa K, Miyazaki H. Simultaneous determination of mevalonate and 7c~-hydroxycholesterol in human plasma by gas chromatography-mass spectrometry as the indices of cholesterol and bile acid biosynthesis. J Chromatogr 1993; 613:185-193. 36. Hirano Y, Miyazaki H, Higashidate S, Nakayama F. Analysis of 3sulfated and nonsulfated bile acids by one-step solvolysis and high-performance liquid chromatography. J Lipid Res 1987; 28:1524-1529. 37. Salen G, Nicolau G, Shefer S, Mosbach EH. Hepatic cholesterol metabolism in patients with gallstones. Gastroenterology 1975; 69:676-684. 38. Coyne MJ, Bonorris GG, Goldstein L J, Schoenfietd LJ. 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 1976;87:281-291. 39. Key PH, Bonorris GG, Marks JW, Chung A, Schoenfield LJ. Biliary lipid synthesis and secretion in gallstone patients before and

PRIMARY DUAL DEFECT IN INTRAHEPATIC CALCULI

40.

41.

42.

43.

44.

45. 46.

47. 48. 49.

50.

51. 52. 53.

54. 55.

56.

57.

58.

1545

during treatment with chenodeoxycholic acid. J Lab Clin Med 1980; 95:816-826. Maton PN, Ellis HJ, Higgins MJP, Dowling RH. Hepatic HMG-CoA reductase in human cholelithiasis: effect of chenodeoxycholic and ursodeoxycholic acids. Eur J Clin Invest 1980;10: 325-332. Carulii N, Ponz de Leon M, Zironi F, Pinetti A, Smerieri A, Iori R, Loria P. Hepatic cholesterol and bile acid metabolism in subjects with gallstones: comparative effects of short-term feeding of chenodeoxycholic and ursodeoxycholic acid. J Lipid Res 1980; 21:35-43. Reihn~r E, Angelin B, Bj6rkhem I, Einarsson K. Hepatic cholesterol metabolism in cholesterol gallstone disease. J Lipid Res 1991; 32:469-475. Angelin B, Olivecrona H, Reihn~r E, Rudling M, St&hlberg D, Eriksson M, Ewerth S, Henriksson P, Einarsson K. Hepatic cholesterol metabolism in estrogen-treated men. Gastroenterology 1992; 103:1657-1663. Takahashi K, Narumi T, Fukushima N, Momota Y, Matsumoto M, Endo M, Suzuki H, Sasaki M, Ono K. Biliary dynamics in patients with intrahepatic gallstones. J Bil Tract Pancreas 1986;10: 1317-1322. Simi M, Loriga P, Basoli A, Leardi S, Speranza V. Intrahepatic lithiasis. Am J Surg 1979;137:317-322. Grundy SM, Metzger AL, Adler RD. Mechanism of lithogenic bile formation in American Indian women with cholesterol gallstones. J Clin Invest 1972;51:3026-3043. van der Linden W. Bile acid pattern of patients with and without gallstones. Gastroenterology 1971;60:1144-1145. Marcus SN, Heaton KW. Deoxycholic acid and the pathogenesis of gallstones. Gut 1988;29:522-533. Berr F, Pratschke E, Fischer S, Paumgartner G. Disorders of bile acid metabolism in cholesterol gallstone disease. J Clin Invest 1992; 90:859-868. Heuman DM, Vlahcevic ZR, Bailey ML, Hylemon PB, Regulation of bile acid synthesis. I1. Effect of bile acid feeding on enzymes regulating hepatic cholesterol and bile acid synthesis in the rat. Hepatology 1988;8:892-897. Tanimura H, Uchiyama K, Ishimoto K, Wu S. Epidemiology of hepatolithiasis. J Bil Tract Pancreas 1994; 15:401-408. Kameda H. Present status the types of gallstones by classification in Japan. J Bil Tract Pancreas 1991;12:1179-1183. Furukawa M, Kusano T, Ohtsubo M, Shirahama S. The survey of hepatolithiasis. In: Annual reports of the Japanese Ministry of Health and Welfare. Tokyo, Japan: Japanese Government, 1991:11-16. Tabata M, Nakayama F. Bacteria and gallstones: etiological significance. Dig Dis Sci 1981;26:218-224. Gallinger S, Taylor RD, Harvey PRC, Petrunka CN, Strasberg SM. Effect of mucous glycoprotein on nucleation time of human bile. Gastroenterology 1985; 89:648-658. Groen AK, Noordam C, Drapers JAG, Egbers P, Jansen PLM, Tytgat GNJ. Isolation of a potent cholesterol nucleation-promoting activity from human gallbladder bile: role in the pathogenesis of gallstone disease. Hepatology 1990;11:525-533. Kibe A, Holzbach RT, LaRusso NF, Mao SJT. Inhibition of cholesterol crystal formation by apolipoproteins in supersaturated model bile. Science 1984; 25:514-516. Shimizu S, Sabsay B, Veis A, Ostrow JD, Rege RV, Dawes LG. Isolation of an acidic protein from cholesterol gallstones which inhibits the precipitation of calcium carbonate in vitro. J Clin Invest 1989;84:1990-1996.

Received May 19, 1994. Accepted January 6, 1995. Address requests for reprints to: Junichi 5hoda, M.D., Ph.D., De-

1546

SHODA ET AL.

partment of Gastroenterology, Institute of Clinical Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba-shi, Ibaraki 305, Japan. Supported in part by Grant-in-Aid 05670443 for Intrahepatic Calculi from the Ministry of Health and Welfare and Grant-in-Aid 05770344 for Scientific Research from the Ministry of Education, Japan. Parts of this work have been published in abstract form (Gastroenterology 1994; 106:A985). The authors thank Professor N. Kobayashi (First Department of Surgery, Ehime University School of Medicine, Ehime, Japan) and

GASTROENTEROLOGY Vol. 108, No. 5

Drs. T. Tsukui and Y. Niizuma (Department of Surgery, Mito-Kyodo Hospital, Ibaraki, Japan) for help and support throughout this work; Dr. M. Noshiro (Department of Biochemistry, School of Dentistry, Hiroshima University, Hiroshima, Japan) for complementary DNA of human cholesterol 7~-hydroxylase; and Japanese Cancer Research Resources Bank (Tokyo, Japan) for supplying complementary DNA of human low-density lipoprotein receptor and complementary DNA of human 3-hydroxy-3-methylglutaryl-coenzyme A reductase.