The Incorporation of Fatty Acids of Different Chain Length into Liver and Biliary Lipids in the Perfused Rat Liver Moshe Rubina,*, Ronit Pakulab, Tuvia Gilatb, and Alisa Tietzc a

Department of Surgery “B”, Beilinson Medical Center and Felsenstein Medical Research Center, Petah-Tikva 49100, b Sackler Faculty of Medicine, Minerva Center for Cholesterol Gallstones and Lipid Metabolism in the Liver, Tel Aviv University, and Department of Gastroenterology, Tel Aviv Medical Center, Tel Aviv 64239, and cDepartment of Neurobiochemistry, Tel Aviv University, Tel Aviv, Israel

ABSTRACT: In an attempt to correlate the incorporation of fatty acids (FA) of different chain length into liver and biliary lipids, isolated rat livers were perfused for 2 h with Krebs-Ringer bicarbonate containing 1% albumin and 10 µmol of [1-14C]-labeled FA: C2, C8, C10, C12, C16, and C18:1. One to 1.36 µmol of medium-chain fatty acids (MCFA, C8, C10, and C12) and 6.6 µmol of long-chain FA (LCFA) were incorporated into liver lipids, 40% of the latter into phosphatidylcholine (PC). 14C-acetate (13 nmol) was incorporated into biliary cholesterol; 14CMCFA contributed only 3.2–5 nmol; LCFA did not lead to newly synthesized cholesterol. Newly synthesized liver PC (2.75 to 3.25%) and newly synthesized liver cholesterol (6.5 to 10%) were secreted into bile. The specific radioactivity of biliary PC after infusion of all-saturated FA was 3.8–6.8 times higher than that of liver PC; for C18:1 it was only 1.7-fold. The specific radioactivity of biliary cholesterol, as compared to liver cholesterol, was 12 times higher for C2 and five times higher for MCFA. This indicates that a considerable proportion of the newly synthesized lipids was secreted into bile prior to significant mixing with preexisting liver PC and cholesterol pools. Liver PC contained 8% of unchanged 14C-C12; while 14C-C10 was not detected. Biliary PC, in contrast, contained 18% of unchanged 14C-C12 and 3% 14C-C10. These results suggest that after prolonged infusion of medium-chain triacylglycerols/longchain triacylglycerols to patients, biliary PC may become enriched with MCFA. In addition, the oxidation of these FA may provide C-2 units which increase cholesterol synthesis. Paper no. L7865 in Lipids 34, 571–578 (June 1999).

The correlation between cholesterol solubility and the concentration of biliary bile salts, phospholipids (PL), and cholesterol were formulated by Admirand and Small (1). Subsequent research on factors affecting cholesterol gallstone formation concentrated mainly on the role of cholesterol and bile *To whom correspondence should be addressed at Department of Surgery “B”, Beilinson Medical Center, Petah Tiqva 49100, Israel. E-mail: [email protected] Abbreviations: C2, acetic acid; C8, octanoic acid; C10, decanoic acid; C12, dodecanoic acid (lauric acid); C16, hexadecanoic acid (palmitic acid); C18:1, oleic acid; FA, fatty acid; FFA, free fatty acid; GLC, gas–liquid chromatography; KRB, Krebs-Ringer bicarbonate; LCFA, long-chain fatty acids; LCT, long-chain triacylglycerols; MCFA, medium-chain fatty acids; MCT, medium-chain triacylglycerols; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PL, phospholipids; TAG, triacylglycerols; TLC, thinlayer chromatography. Copyright © 1999 by AOCS Press

salts (1,2). The importance of PL in cholesterol solubilization was appreciated only more recently (3–5). Biliary PL are generally composed of 95% phosphatidylcholine (PC) and 4% phosphatidylethanolamine (PE) (6). Biliary PC is a complex mixture of different molecular species. In all animal and human studies, PC was found to contain saturated fatty acids (FA) (mainly 16:0 and 18:0) in the sn-1 position and unsaturated FA (mainly 18:2, 18:1, and 20:4) in the sn-2 position. In man, 1-palmitoyl-2-linoleyl-glyceryl-phosphoryl choline amounted to over 40% of the PC (7,8). Addition of different PC to human bile samples in vitro (5,9) caused a marked prolongation of the nucleation time. Since in vitro addition of bile acids did not affect the cholesterol nucleation time, it was suggested that increasing PL concentrations in bile might be more effective for prevention of cholesterol crystallization (9). Experiments performed in animals and humans attempted to induce changes in the composition or concentration of PL in native bile by dietary manipulations. Nervi et al. (10) correlated the high incidence of gallstones among young Chilean men with a selective decrease in biliary PL, possibly due to a large intake of legumes. A similar abnormality was induced in rats by a high legume diet (11). Dietary supplementation of fish oil in contrast to more saturated triacylglycerols (TAG) markedly increased the percentage of 20:5 and 22:6 in biliary PC, reducing the lithogenicity and incidence of cholesterol precipitates in rats (12), African green monkeys (13), and hamsters (14–16). Likewise, addition of fish oil to the diet of gallstone patients induced an increase in 20:5 and 22:6 in biliary PC with a concomitant reduction in 18:2 concentration (17). Robins and Patton (8) pointed out that in order to understand the changes occurring in biliary PC, analysis of the molecular species is required. When rats were fed pure TAGcontaining FA which were more hydrophilic than the FA normally present in bile PC, bile became highly enriched in new molecular species containing the particular FA fed. In contrast when the fat contained less hydrophilic FA, the composition of biliary PC did not change. To further characterize these changes, rat livers were perfused with FA of different chain lengths and degree of unsaturation (18). Results indicated that whereas 18:1, 20:4, and 20:5 were enriched in the sn-2 position of biliary PC, 16:1, 17:1, and 18:2 were found

571

Lipids, Vol. 34, no. 6 (1999)

572

M. RUBIN ET AL.

in both positions. The above findings clearly indicate that the FA composition of biliary PC can be selectively changed. Only relatively few experiments were carried out attempting to change the head-group composition. Addition of ethanolamine and linoleic acid to the diet of rats and hamsters caused a slight increase in biliary PE; cholesterol concentration and the cholesterol saturation index were significantly lower (19). In humans addition of PE (rich in 18:2) to the diet was found to increase the percentage of 18:2 in biliary PC without changing PE concentration (20). Total parenteral nutrition may result in the development of gallstones (21,22). In our clinical studies (23) we observed marked differences in the concentration of lipids in blood and bile following short-term infusion (24 h) of TAG containing a mixture of long-chain triacylglycerols (LCT) and mediumchain triacylglycerols (MCT). To try to understand the effect of MCT, we carried out a systematic in vitro study (24) on the metabolism of FA of various chain lengths by HepG-2 cells. It was found that C8 was incorporated only after β oxidation. C10, C12, and C16 were largely incorporated unchanged into cellular TAG and PL. In the HepG-2 model, the secretion of biliary lipids could not be studied. In contrast, in the perfused rat liver model, both lipid synthesis and secretion into bile could be studied simultaneously. In the present manuscript, we report the effect of FA of different chain lengths on liver and biliary lipid composition. MATERIALS AND METHODS Materials. [1-14C]Sodium acetate (52.1 mCi/mmol), [1-14C]lauric acid (52.8 mCi/mmol) and [1-14C]palmitic acid (57.0 mCi/mmol) were obtained from Amersham (United Kingdom); [1-14C]sodium octanoate (55.0 mCi/mmol) and [1-14C]oleic acid (50.0 mCi/mmol) from NEN (Boston, MA); and [1-14C]sodium decanoate (10.6 mCi/mmol) from Sigma (St. Louis, MO). All 14C-labeled FA were diluted with nonlabeled FA to yield a final specific radioactivity of 2 mCi/mmol; 14 C-acetate was diluted to a specific radioactivity of 4 mCi/mmol. All FA were obtained from Sigma. Thin-layer Silica gel 60 plates (0.25 mm) were obtained from Merck (Darmstadt, Germany), KC 18 Silica gel plates (0.2 mm) from Whatman (Maidstone, England). All solvents were of analytical grade (Merck). Albumin, bovine-fat poor, was obtained from Sigma. Scintillation fluid-Ultima-Gold was obtained from Packard (Meriden, CT). Isolated perfused rat liver studies. Male Wistar rats weighing 300–360 g were used. They were fed Purina Rodent Chow ad libitum and maintained under a constant light cycle of 12 h. The surgical procedures were performed as described by Corasanti et al. (25). Rats were anesthetized with pentobarbitone (50 mg/kg body wt i.p.) (Ceva, Paris, France). The bile duct was cannulated with PE-10 tubing (Clay Adams, Parsippany, NJ). The portal vein was cannulated with a 16gauge Teflon intravenous catheter (BOC Ohmeda AB, Helsingborg, Sweden). The liver was then perfused at a constant flow of 20 mL/min with oxygenated Krebs-Ringer bicarLipids, Vol. 34, no. 6 (1999)

bonate (KRB) buffer containing (in mM) 120 NaCl, 4.7 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.8 CaCl2, and 25 NaHCO3. Added to the KRB buffer were 5.5 mM glucose and 200 U heparin/100 mL. After cannulation of the inferior vena cava, the livers were removed from the rat, transferred into a heated perfusion chamber, and perfused in a recirculating closed system (volume: 200 mL) at a constant flow rate of 40 mL/min with KRB buffer containing 5.5 mM glucose and 1% bovine serum albumin. The KRB buffer was gased continuously with a humidified mixture of 95% O2–5% CO2 and maintained at 37 ± 0.5°C. After 10 min of equilibration, a bolus of the following lipids was added: 10 µmol of [1-14C]-labeled FA: C2, C8, C10, C12, C16, and C18:1. Taurocholic acid (30 µM) was infused continuously to stimulate bile secretion. The livers were perfused for 2 h. At the end of the experiment, the livers were perfused for 15 min with fresh KRB. Perfusion pressure was monitored continuously. Bile and perfusate samples were collected every 15 min. Bile was collected into pretared tubes which were changed every 15 min. The tubes were then weighed, and bile flow was calculated as µg bile/min/g liver. At the end of the experiments, bile samples were pooled. Livers, bile, and perfusate samples were frozen at −70°C for lipid analysis. Each perfusion experiment was done twice, and livers and biles were analyzed separately. Approximately 2 mL of bile was collected. The mean weight of the livers was 10.4 ± 1.0 g. Two 1-g samples of each liver were used for lipid extraction and analysis. The results presented in the tables are the mean of these four determinations. A sample of the perfusate was removed at the end of the experiment, and the distribution of the radioactivity among the different lipid classes was determined. The viability of the perfused liver was assessed throughout the perfusion by monitoring perfusion pressure (7–12 cm H2O), determining baseline bile flow, and measuring the release of lactate dehydrogenase as well as observing the general appearance of the liver. Liver and bile lipid extraction and analysis. Liver samples of 1 g were homogenized with 5 mL of saline using an UltraTurax (Janke and Kunkel Ika-Werk, Staufen, Germany). The homogenates were then extracted with chloroform/methanol according to Bligh and Dyer (26). Pooled bile (1 mL) from each experiment and 5 mL of the perfusate collected at the end of the experiment were similarly extracted (26). The amount of the radioactivity in the extracts was determined using a Kontron-Betamatic counter (Switzerland). Aliquots of the lipid extracts were used for the separation of neutral lipids and PL by thin-layer chromatography (TLC) as previously described (24). Neutral lipids were separated on Silica gel G plates employing hexane/diethylether/methanol/ acetic acid (90:20:3:2, by vol) as solvent. PL were separated using chloroform/methanol/acetic acid/water (100:20:12:5, by vol) as solvent. For determination of radioactivity, lipid spots were scraped into scintillation vials and suspended in 3.5 mL scintillation fluid. The overlapping spot of diglycerides and cholesterol on the TLC plates was collected and

LIVER AND BILIARY LIPIDS OF PERFUSED LIVER

saponified with 0.5 M methanolic KOH for 30 min at 50°C. Samples were than acidified and extracted with hexane. Aliquots were placed on TLC plates as described above. To determine the chain length of the FA which were incorporated into PL and TAG, the corresponding spots were collected and saponified with 0.5 M methanolic KOH. FA were recovered and separated according to their chain length by reversed-phase TLC on KC18 plates utilizing acetonitrile/methanol/water (6:3:1, by vol) as solvent (24,27). Under our experimental conditions, 18:0 and 18:1 co-migrated with 16:0. Spots were scraped directly into scintillation vials, and the radioactivity was determined immediately after addition of scintillation fluid. To determine the FA composition of PL and TAG, the FA recovered after alkaline hydrolysis were methylated with diazomethane (28). The methyl esters were separated by gas–liquid chromatography (GLC) on a 30 m PAG (poly alkylene glycol) column (0.25 µm film thickness; Supelco, Bellefonte, PA) at a temperature range of 185–220°C, employing a Hewlett-Packard 5790A gas chromatograph equipped with a flame-ionization detector. The relative composition of FA mixtures was calculated employing a Hewlett-Packard 3396 integrator. For quantitative analysis, heptadecanoic acid was added as an internal standard, assuming that the response of the detector for all methyl fatty esters was identical. Saturated and unsaturated methyl FA were separated on AgNO3-impregnated plates (29), employing hexane/diethyl ether (80:20, vol/vol) as solvent. Free cholesterol was determined directly from the chloroform/methanol extracts by an enzymatic colorimetric method using a cholesterol oxidase kit from Boehringer Mannheim GmbH (Ingelheim, Germany). RESULTS Incorporation of 14C-labeled FA into liver lipids. Addition of 10 µmol of [1-14C]-labeled FA (C2, C8, C10, C12, C16, and

573

C18:1) to the perfusion buffer resulted in the incorporation of 0.66 to 6.61 µmol (6.6 to 66.1% of the radioactivity added to the perfusate) of 14C-FA into liver lipids (Table 1). More long-chain FA (LCFA) (C16 and C18:1) were incorporated than medium-chain FA (MCFA) (C8, C10, and C12). Only acetate was extensively incorporated into cholesterol (25% of the amount incorporated into liver lipids). The other FA were incorporated mainly into PC (ca. 40%) and to a lesser extent into TAG and PE. Diacylglycerols, FA, cholesterol ester, and phosphatidylserine contained only relatively small amounts of radioactivity (not shown). LCFA did not lead to newly synthesized cholesterol. Incorporation of 14C-labeled FA into biliary lipids. Bile was collected throughout the perfusion period. Figure 1 shows that bile secretion became stable after 30 min at an average rate of 1.8 µg/min/g liver and remained constant for the rest of the perfusion period. As can be seen in Figure 2 secretion of 14C-labeled metabolites into bile was dependent on the 14 C-FA added. In most cases, maximal values were obtained 40 to 50 min after addition of the 14C-substrate. 14C-Metabolites include 14C-lipids and 14C-water-soluble compounds. For C2, C8, C10, and C12 the 14C-water-soluble compounds amounted to 65–75% of the total radioactivity secreted into bile. In the presence of C16 and C18:1, only 15% of the radioactivity was water-soluble. 14C-Free fatty acid (FFA) were not secreted into bile. 14C was not detected in bile acids (30). Attempts to identify the water-soluble compounds indicated that these were not volatile and could not be extracted into ether (after acidification). 14C-Acetoacetate was not detected after incubation with β-hydroxybutyrate dehydrogenase (31). Table 2 shows the incorporation of 14C-FA into biliary lipids. When 14C-acetate was perfused, 17 nmol were recorded in biliary lipids—75.5% of which were incorporated into cholesterol. In the presence of MCFA, 19 to 22 nmol were incorporated into biliary lipids—71–78% into PC; in the presence of C16 and C18:1, 90 and 70 nmol were incorporated into biliary lipids; over 95% of the radioactivity was detected

TABLE 1 Incorporation of Different 14C-FA into Liver Lipidsa

FA

14 C-FAincorporated (µmol)

TAG

C2 C8 C10 C12 C16 C18:1

0.66 ± 0.01 1.36 ± 0.41 0.99 ± 0.19 1.21 ± 0.27 6.61 ± 1.88 6.61 ± 1.78

18.18 ± 1.46 29.83 ± 7.12 21.41 ± 1.37 25.49 ± 3.19 22.77 ± 1.87 35.01 ± 4.58

Radioactivity incorporated (%) Cholesterol 25.03 ± 4.2 2.49 ± 1.47 7.59 ± 1.73 3.82 ± 0.61 Trace Trace

PC

PE

31.42 ± 3.77 39.84 ± 6.01 38.66 ± 3.97 37.7 ± 2.59 46.51 ± 1.15 33.38 ± 5.03

10.79 ± 0.77 13.58 ± 1.42 14.78 ± 1.97 13.19 ± 0.73 13.91 ± 1.65 12.60 ± 0.59

a

Lipids were extracted from liver samples and separated by thin-layer chromatography (TLC) as described in the Materials and Methods section. Hexane/diethylether/methanol/acetic acid (90:20:3:2, by vol) was used for the separation of neutral lipids, chloroform/methanol/acetic acid/water (100:20:12:4, by vol) for the separation of phospholipids (PL). Lipid spots were scraped directly into scintillation vials and radioactivity determined after addition of scintillation fluid. The incorporation of 14C-fatty acid (FA) is expressed as µmol per total liver, assuming incorporation of intact FA. Diacylglycerols, FA, cholesterol esters, and phosphatidylserine contained 1–3% each. The results given are the mean (±SD) of data obtained from two different experiments. Liver weight was in the range of 9 to 11 g. TAG, triacylglycerols; PC, phosphatidylcholine; PE, phosphatidylethanolamine.

Lipids, Vol. 34, no. 6 (1999)

574

M. RUBIN ET AL.

FIG. 1. Bile flow rate. Bile was collected into pretared tubes each 15 min. The bile flow was calculated as µg bile/min per gram wet liver. The results presented are the mean values ± SD of all bile samples collected during the study period (n = 12).

in PC. Biliary PE contained ca. 2–3%, and phosphatidylserine contained less than 1% (not shown). Comparison of FA composition of liver and biliary PL. Table 3 compares the quantities and FA composition of liver and biliary PC and PE. Since the amounts of these PL and their FA composition were not affected by the addition of the different 14C-FA to the perfusate, the mean results for all experiments are presented in the table. As can be seen from Table 3, biliary PL contained mainly PC (93%) and PE (6%); phosphatidylserine amounted to less than 1% (not shown). Comparison of the incorporation of 14C-FA into liver and biliary PC and cholesterol. Table 4 compares the specific radioactivity of liver and biliary PC and cholesterol. The specific radioactivity of biliary PC after administration of all saturated FA was 3.8 to 6.8 times higher than that of liver PC; for C18:1 it was only 1.7-fold, suggesting a higher dilution of 18:1 by liver PC. The specific radioactivity of biliary choles-

terol (as compared to liver cholesterol) was 12 times higher for C2 and five times higher for MCFA. LCFA did not lead to 14 C-cholesterol synthesis. From the data presented in Tables 1 and 2, it can be calculated that 200 nmol of acetate and 400–550 nmol of MCFA were incorporated into liver PC; 4.1 nmol acetate and 13–17 nmol of MCFA were incorporated into biliary PC. The incorporation of the LCFA into liver and biliary PC was considerably higher, 3312 and 85 nmol (into liver and bile, respectively) for 14C-C16 and 2235 and 68 nmol (into liver and bile, respectively) for C18:1. Although the incorporation of LCFA (16:0 and 18:1) was considerably higher, nevertheless in all experiments 14C-labeled biliary PC amounted to 2.75 to 3.25% of the 14C-labeled liver PC. Secretion of PL, TAG, and cholesterol into the perfusate reached 0.5–1% of that present in the liver. The distribution of radioactivity between the different lipids in the perfusate was generally similar to that de-

FIG. 2. Secretion of 14C-labeled metabolites into bile. The secretion of 14C-labeled metabolites into bile was measured for each fatty acid (FA): C2 ◆, C8 ■, C10 ▲, C12 X, C16 ✴, and C18:1 ●. Bile samples were collected as described in the legend to Figure 1. A sample of 10 µL of bile from each time point was counted to determine 14C-content.

Lipids, Vol. 34, no. 6 (1999)

LIVER AND BILIARY LIPIDS OF PERFUSED LIVER TABLE 2 Incorporation of Different 14C-FA into Biliary Lipidsa 14

C-FA in bile (nmol)

C2 C8 C10 C12 C16 C18:1

17.02 ± 3.13 18.62 ± 10.24 19.00 ± 7.24 21.99 ± 1.25 89.18 ± 38.43 70.65 ± 14.07

14

DISCUSSION

C incorporated (%)

Cholesterol

PC

75.47 ± 6.8 17.54 ± 6.81 26.22 ± 12.04 21.22 ± 3.61 Trace Trace

575

23.85 ± 6.77 79.89 ± 6.99 71.34 ± 12.00 75.96 ± 3.99 95.03 ± 0.91 95.79 ± 0.39

a Lipids were extracted from 1 mL bile samples and separated by TLC as described in the legend to Table 1. The incorporation of 14C-FA is expressed as nmol per total bile sample, assuming that the 14C-FA reached the bile via the liver without processing. The results are the mean (±SD) of the two different bile samples analyzed. See Table 1 for abbreviations.

tected in the liver. Only when 14C-acetate was added, the percentage of PL in the perfusate was higher and that of cholesterol was lower than in the liver. Occurrence of C10 and C12 in liver and biliary lipids. 14CTAG and PL synthesized by the liver during perfusion with 14 C-C2, -C8 and -C16 were isolated and the 14C-FA recovered after saponification were separated by TLC. Only 14C-LCFA (14C-C16 and -C18) were identified. However, after perfusion with 14C-C10, saponification of 14C-TAG yielded ca. 10% 14C-C10 (Fig. 3) and 90% of 14C-C16 and C18; 14C-C10 was not detected in liver PL. After perfusion with 14C-C12 saponification of liver TAG yield ca. 35% 14C-C12 and 65% 14 C-C16 and -C18. Liver PC contained ca. 8% of 14C-C12. Biliary PC contained 3% 14C-C10 and 18% 14C-C12. For further identification of the newly incorporated FA, methyl-FA were separated on AgNO3-impregnated plates. Less than 1% of the radioactivity was associated with monounsaturated FA (not shown), indicating that desaturation did not occur under our experimental conditions. After 14C-C18:1 perfusion, 98% of the radioactivity of the FA isolated from liver and biliary lipids was present in oleic acid.

In the present paper the metabolism of 14C-labeled FA of different chain lengths by the isolated perfused rat liver was studied. This experimental approach permitted the simultaneous measurement of FA incorporation into liver and biliary lipids. We previously reported (23) that short-term infusion of MCT/LCT to patients resulted in significant changes in bile composition, whereas LCT alone did not elicit any changes. The aim of this and our previous studies (24) was to examine in vitro the mechanism underlying these in vivo effects. LCT (500 µL) which contained 113 µmol of soya-TAG and 0.53 µmol of (3H-oleate)-labeled trioleylglycerol were added to the perfusate. After 2 h, 50% of the remaining TAG (12 µmol) were detected as free FA, indicating extensive lipolysis of the added LCT. Although the liver was perfused for 30 min with LCT-free buffer, most of the radioactivity isolated from the liver (61.4 µmol) was detected in TAG (39.8%), FFA (22.9%), and diacylglycerols (18.5%) and only 10.6% was incorporated into PL. The occurrence of large amounts of FFA in the perfusate and the liver suggests the role of FFA uptake under our experimental conditions. Since radioactively labeled MCT was not available to us, 14C-labeled MCFA (C8, C10, and C12) and LCFA (C16 and C18:1) were added instead. 14C Acetate was used in these experiments to represent acetyl CoA, the β-oxidation product of the added FA. The major finding of the present investigation is that C2, C8, C10, and C12 were largely used for the synthesis of saturated LCFA, which were subsequently incorporated into liver PL and TAG, whereas C16 and C18:1 were used directly for PC and TAG synthesis. For C8, C10, and C12 to yield LCFA, they have to be either oxidized to C2 units prior to being used for de novo FA synthesis or elongated after internalization. The occurrence of water-soluble metabolites in the bile after administration of 14C-C2 and 14C-MCFA suggests that acetyl CoA was formed by β-oxidation of MCFA (32) and was further metabolized. We did not succeed in identifying the water-

TABLE 3 Lipid Content and PL-FA Composition of Liver and Bilea FA composition (%) b

Content

16:0

18:0

18:1

18:2

20:4

22:6

Liver PC PE Cholesterol

21.1 ± 4.2 10.8 ± 2.4 5.81 ± 1.01

19.3 ± 0.9 15.3 ± 1.5

20.5 ± 1.0 18.5 ± 1.8

8.5 ± 0.5 8.2 ± 0.4

17.4 ± 1.0 17.0 ± 4.0

20.5 ± 2.1 22.5 ± 1.9

6.37 ± 0.6 10.9 ± 0.9

Bile PC PE Cholesterol

1.2 ± 0.2 0.08 ± 0.01 0.26 ± 0.07

34.7 ± 1.2 27.9 ± 2.4

6.3 ± 0.3 13.3 ± 1.0

9.4 ± 0.4 8.4 ± 0.4

30.8 ± 1.5 23.7 ±1.1

10.1 ± 1.1 13.5 ± 0.8

1.9 ± 0.3 5.6 ± 0.8

a

To determine the cholesterol and PL content and FA composition, lipids were extracted from liver and bile samples and separated by TLC as described in the Materials and Methods section and legend to Table 1. PL spots were collected and saponified with 0.5 M methanolic KOH. FA were recovered after acidification, methylated with diazomethane (Ref. 28) and separated by gas–liquid chromatography as described in the Materials and Methods section. Minor FA components such as 16:1, 18:3, 20:1, and 20:5 are not included in the table. The PL content and FA composition of liver and biliary lipids were similar for all samples analyzed. The results are the mean (±SD) of 12 independent experiments. b Lipid contents of liver and bile are presented as µmol/g and µmol/mL, respectively. See Table 1 for abbreviations.

Lipids, Vol. 34, no. 6 (1999)

576

M. RUBIN ET AL.

TABLE 4 Specific Radioactivity of Liver and Biliary PC and Cholesterola PC

C2 C8 C10 C12 C16 C18:1

Cholesterol

Liver (dpm/µmol)

Bile (dpm/µmol)

Liver (dpm/µmol)

Bile (dpm/µmol)

3813 7582 7291 6070 54521 23503

19572 48395 27815 25844 248931 39032

14980 2057 4230 1780 — —

178208 10870 23040 8414 — —

a Liver and biliary PC and cholesterol were isolated and their concentration and radioactivity determined as described in Tables 1, 2, and 3. See Table 1 for abbreviations.

soluble compounds. The most likely metabolites, 3-hydroxybutyrate and acetoacetate, were not detected. During the perfusion period, irrespective of the FA administered, 2.75 to 3.25% of the 14C-labeled PC were secreted into bile. The specific radioactivity of biliary PC was higher than that of liver PC, indicating the secretion of the 14C-labeled PC into bile prior to mixing with the liver pools of PC. Whereas after perfusion with MCFA, newly synthesized biliary PC amounted to 0.57–0.70% of total biliary PC, with LCFA newly synthesized PC reached 2.8 to 3.5%. Biliary secretion of PC by the perfused liver was previously studied by Robins et al. (18,33). In these experiments 14C-choline was continuously added. Although PC secretion was four times higher than in our experiments, secretion of newly synthesized PC amounted to 3.4% of total biliary PC.

In contrast to the results obtained with HepG-2 cells (24) in which 14C-C12 was incorporated into cellular lipids to the same extent as 14C-C16, in the perfused rat liver its incorporation was like that of C8 and C10. Furthermore, whereas in HepG-2 cells, C12 was largely incorporated unchanged (80–90% of all 14C-FA in TAG; 50–60% in PL), in the perfused rat liver 14C-C12 amounted to 30–40% of all 14C-FA in TAG and 10% in PL. Unchanged 14C-C10 was not detected in liver PL. Biliary PC contained 3% of unchanged 14C-C10 and 18% unchanged 14C-C12. Thus the percentage of C10 and C12 in biliary PC was significantly higher than in liver PC. Using GLC, we could not detect C10 and C12 in biliary PC. The calculated amount of C12 present in PC (500 ng) was only 0.04% of total biliary PC-FA and thus below our detection limit. Robins et al. (18) previously showed that there were significantly greater amounts of the administered 16:1, 17:1, and 18:2 in the PC of bile than the liver leading to the secretion of 16:1-16:1, 17:1-17:1, and 18:2-18:2 species, indicating preferential secretion of newly synthesized PC. The occurrence of 14C-C12 in biliary PC in our experiments is compatible with these observations. 14 C-Acetate, in contrast to MCFA, was preferentially incorporated into liver and biliary cholesterol (Table 2). However, since during β-oxidation of MCFA dilution of 14C-acetyl CoA occurs, the actual incorporation is probably higher. During the 2-h infusion, 6.5 to 10% of the newly synthesized cholesterol was secreted into the bile. The specific radioactivity of biliary cholesterol, like that of PC, was higher than that of liver cholesterol, indicating secretion of newly synthesized

FIG. 3. Recovery of 14C-C10 and 14C-C12 in liver and biliary lipids. To determine the occurrence of 14C-C10 and 14C-C12, lipids were extracted from liver and bile samples and separated by thin-layer chromatography as described in the Materials and Methods section. Liver triacyl||, liver phosphatidylcholine (PC) ■, and biliary PC ■ were scraped off the glycerols (TAG) |■ plate and subjected to alkaline hydrolysis. FA were recovered after acidification and separated by reversed-phase thin-layer chromatography as described in the Materials and Methods section. FA spots were scraped into scintillation vials and the radioactivity determined. Each point is the mean of two to four different analyses ± SD. See Figure 2 for abbreviation.

Lipids, Vol. 34, no. 6 (1999)

LIVER AND BILIARY LIPIDS OF PERFUSED LIVER

cholesterol prior to its mixing with liver pools. The preferential secretion of newly synthesized cholesterol was reported previously (34,35). The percentage of newly synthesized cholesterol of the total biliary cholesterol reached 12.8% when acetate was administered and 3.2 to 5% in the presence of MCFA. Our present data on the secretion of 14C-PC and 14Ccholesterol and those previously reported by Robins et al. (18,33,35) indicate that most of the biliary PC and cholesterol were recruited from preexisting liver-lipid pools. The effect of MCFA on the composition of biliary PC was not described previously. Feeding coconut oil (rich in 12:0 and 14:0) to hamsters (15,16) resulted in a reduction of biliary PL secretion and an increase in the lithogenic index. Occurrence of 12:0 and 14:0 in biliary PC was not recorded. Furthermore, the contribution of MCFA to biliary cholesterol and PC in the perfused rat liver as shown in the present communication may explain the mechanisms responsible for the different effects of MCFA and LCFA on bile composition as observed in our clinical studies as well others. Further studies are being conducted to clarify this issue.

11.

12.

13.

14. 15.

16.

ACKNOWLEDGMENTS This work was performed in partial fulfillment of the requirements for a Ph.D. degree of Ronit Pakula, Sackler Faculty of Medicine, Tel Aviv University, Israel. This work was partially supported by a research grant from the chief scientist, Israel Ministry of Health, Jerusalem, and Minerva Center for Cholesterol Gallstones and Lipid Metabolism in the Liver, Tel Aviv University.

17.

18.

REFERENCES 1. Admirand, W.H., and Small, D.M. (1968) The Physicochemical Basis of Cholesterol Gallstone Formation in Man, J. Clin. Invest. 47, 1043–1052. 2. Holzbach, R.T., Marsh, M., Olszewski, M.F., and Holan, K. (1973) Cholesterol Solubility in Bile. Evidence That Supersaturated Bile Is Frequent in Healthy Man, J. Clin. Invest. 52, 1467–1479. 3. Grundy, S.M. (1989) Role of Biliary Phospholipids in Cholesterol Gallstone Formation, Gastroenterology 96, 942–943. 4. Somjen, G.J., and Gilat, T. (1983) A Nonmicellar Mode of Cholesterol Transport in Human Bile, FEBS Lett. 156, 265–268. 5. Halpern, Z., Moshkowitz, M., Laufer, H., Peled, Y., and Gilat, T. (1993) Effect of Phospholipids and Their Molecular Species on Cholesterol Solubility and Nucleation in Human and Model Biles, Gut 34, 110–115. 6. Hay, D.W., and Carey, M.C. (1990) Chemical Species of Lipids in Bile, Hepatology 12, 6S–16S. 7. Hay, D.W., Cahalane, M.J., Timofeyeva, N., and Carey, M.C. (1993) Molecular Species of Lecithins in Human Gallbladder Bile, J. Lipid Res. 34, 759–768. 8. Robins, S.J., and Patton, G.M. (1986) Separation of Phospholipid Molecular Species by High-Performance Liquid Chromatography: Potentials for Use in Metabolic Studies, J. Lipid Res. 27, 131–139. 9. Jungst, D., Lang, T., Huber, P., Lange, V., and Paumgartner, G. (1993) Effect of Phospholipids and Bile Acids on Cholesterol Nucleation Time and Vesicular/Micellar Cholesterol in Gallbladder Bile of Patients with Cholesterol Stones, J. Lipid Res. 34, 1457–1464. 10. Nervi, F., Covarrubias, C., Bravo, P., Velasco, N., Ulloa, N., Cruz, F., Fava, M., Severin, C., Del Pozo, R., Antezana, C., Val-

19. 20.

21. 22. 23.

24. 25.

26. 27. 28.

577

divieso, V., and Arteaga, A. (1989) Influence of Legume Intake on Biliary Lipids and Cholesterol Saturation in Young Chilean Men. Identification of a Dietary Risk Factor for Cholesterol Gallstone Formation in a Highly Prevalent Area, Gastroenterology 96, 825–830. Rigotti, A., Marzolo, M.P., Ulloa, N., Gonzalez, O., and Nervi, F. (1989) Effect of Bean Intake on Biliary Lipid Secretion and on Hepatic Cholesterol Metabolism in the Rat, J. Lipid Res. 30, 1041–1048. Balasubramaniam, S., Simons, L.A., Chang, S., and Hickie, J.B. (1985) Reduction in Plasma Cholesterol and Increase in Biliary Cholesterol by a Diet Rich in n-3 Fatty Acids in the Rat, J. Lipid Res. 26, 684–689. Scobey, M.W., Johnson, F.L., Parks, J.S., and Rudel, L.L. (1991) Dietary Fish Oil Effects on Biliary Lipid Secretion and Cholesterol Gallstone Formation in the African Green Monkey, Hepatology 14, 679–684. Cohen, B.I., Mosbach, E.H., Ayyad, N., Miki, S., and McSherry, C.K. (1992) Dietary Fat and Fatty Acids Modulate Cholesterol Cholelithiasis in the Hamster, Lipids 27, 526–532. Berr, F., Goetz, A., Schreiber, E., and Paumgartner, G. (1993) Effect of Dietary n-3 Versus n-6 Polyunsaturated Fatty Acids on Hepatic Excretion of Cholesterol in the Hamster, J. Lipid Res. 34, 1275–1284. Jonnalagadda, S.S., Trautwein, E.A., and Hayes, K.C. (1995) Dietary Fats Rich in Saturated Fatty Acids (12:0, 14:0, and 16:0) Enhance Gallstone Formation Relative to Monounsaturated Fat (18:1) in Cholesterol-Fed Hamsters, Lipids 30, 415–424. Berr, F., Holl, J., Jungst, D., Fischer, S., Richter, W.O., Seiffert, B., and Paumgartner, G. (1992) Dietary n-3 Polyunsaturated Fatty Acids Decrease Biliary Cholesterol Saturation in Gallstone Disease, Hepatology 16, 960–967. Robins, S.J., Fasulo, J.M., Robins, V.F., and Patton, G.M. (1991) Utilization of Different Fatty Acids for Hepatic and Biliary Phosphatidylcholine Formation and the Effect of Changes in Phosphatidylcholine Molecular Species on Biliary Lipid Secretion, J. Lipid Res. 32, 985–992. Peled, Y., and Gilat, T. (1994) Effect of Dietary Phospholipids and Their Constituents on Bile Composition in Rats and Hamsters, Hepatology 19, 708–713. Pakula, R., Konikoff, F.M., Rubin, M., Ringel, Y., Peled, Y., Tietz, A., and Gilat, T. (1996) The Effect of Dietary Phospholipids Enriched with Phosphatidylethanolamine on Bile and Red Cell Membrane Lipids in Humans, Lipids 31, 295–303. Messing, B., Bories, C., Kunstlinger, F., and Bernier, J.J. (1983) Does Total Parenteral Nutrition Induce Gallbladder Sludge Formation and Lithiasis? Gastroenterology 84, 1012–1019. Roslyn, J.J., Pitt, H.A., Mann, L.L., Ament, M.E., and DenBesten, L. (1983) Gallbladder Disease in Patients on LongTerm Parenteral Nutrition, Gastroenterology 84, 148–154. Rubin, M., Halpern, Z., Charach, G., Devir, A., Antebi, A., Gilat, T., and Lichtenberg, D. (1992) The Effect of Lipid Infusion on Bile Composition and Lithogenicity in Patients Without Cholesterol Gallstones, Gut 33, 1400–1403. Pakula, R., Rubin, M., Moser, A.M., Lichtenberg, D., and Tietz, A. (1997) Biosynthesis of Medium-Chain Triacylglycerols and Phospholipids by HepG-2 cells, Lipids 32, 489–495. Corasanti, J.G., Smith, N.D., Gordon, E.R., and Boyer, J.L. (1989) Protein Kinase C Agonists Inhibit Bile Secretion Independently of Effects on the Microcirculation in the Isolated Perfused Rat Liver, Hepatology 10, 8–13. Bligh, E.G., and Dyer, W.J. (1959) A Rapid Method of Total Lipid Extraction and Purification, Can. J. Biochem. Physiol. 37, 911–917. Christie, W.W. (1982) Lipid Analysis, 2nd edn., pp. 78–79, Pergamon Press, United Kingdom. Schlenk, H., and Gellerman, J.L. (1960) Esterification of Fatty

Lipids, Vol. 34, no. 6 (1999)

578

29. 30.

31. 32.

M. RUBIN ET AL.

Acids with Diazomethane on Small Scale, Anal. Chem. 32, 1412–1414. Kates, M. (1986) Techniques of Lipidology, 2nd edn., pp. 235–236, Elsevier, The Netherlands. DeMark, B.R., Everson, G.T., Klein, P.D., Showalter, R.B., and Kern, F., Jr. (1982) A Method for the Determination of Isotope Ratios of Chenodeoxycholic and Cholic Acids in Serum, J. Lipid Res. 23, 204–210. Olsen, C. (1971) An Enzymatic Fluorimetric Micromethod for the Determination of Acetoacetate, β-Hydroxybutyrate, Pyruvate, and Lactate, Clin. Chim. Acta 33, 293–300. McGarry, J.D., and Foster, D.W. (1980) Regulation of Hepatic Fatty Acid Oxidation and Ketone Body Production, Ann. Rev. Biochem. 49, 395–420.

Lipids, Vol. 34, no. 6 (1999)

33. Robins, S.J., and Brunengraber, H. (1982) Origin of Biliary Cholesterol and Lecithin in the Rat: Contribution of New Synthesis and Preformed Hepatic Stores, J. Lipid Res. 23, 604–608. 34. Turley, S.D., and Dietschy, J.M. (1981) The Contribution of Newly Synthesized Cholesterol to Biliary Cholesterol in the Rat, J. Biol. Chem. 256, 2438–2446. 35. Robins, S.J., Fasulo, J.M., Collins, M.A., and Patton, G.M. (1985) Evidence for Separate Pathways of Transport of Newly Synthesized and Preformed Cholesterol into Bile, J. Biol. Chem. 260, 6511–6513. [Received February 3, 1998; and in final revised form April 23, 1999; revision accepted April 29, 1999]