Liver Biology and Pathobiology
A biphasic response of hepatobiliary cholesterol metabolism to dietary fat at the onset of obesity in the mouse†
Article first published online: 25 MAR 2005
Copyright © 2005 American Association for the Study of Liver Diseases
Volume 41, Issue 4, pages 887–895, April 2005
How to Cite
Roy, S., Hyogo, H., Yadav, S. K., Wu, M. K., Jelicks, L. A., Locker, J. D., Frank, P. G., Lisanti, M. P., Silver, D. L. and Cohen, D. E. (2005), A biphasic response of hepatobiliary cholesterol metabolism to dietary fat at the onset of obesity in the mouse. Hepatology, 41: 887–895. doi: 10.1002/hep.20643
Potential conflict of interest: Nothing to report.
- Issue published online: 25 MAR 2005
- Article first published online: 25 MAR 2005
- Manuscript Accepted: 7 JAN 2005
- Manuscript Received: 27 SEP 2004
- National Institutes of Health. Grant Numbers: DK48873, DK56626, DK20514
- Irwin M. Arias American Liver Foundation Postdoctoral Fellowship
Human obesity is associated with abnormal hepatic cholesterol homeostasis and resistance to leptin action. Because leptin administration to rodents promotes the biliary elimination of plasma cholesterol, this study was designed to elucidate a pathophysiological role for leptin during the development of obesity. We fed mice diets containing high or low saturated fat contents. Before and after the onset of obesity, we measured downstream targets of leptin action and evaluated plasma, hepatic, and biliary cholesterol metabolism. Although not obese at 28 days, mice fed a high fat diet became hyperleptinemic. Sensitivity to leptin was evidenced by downregulation of both hepatic stearoyl CoA desaturase-1 and fatty acid synthase. Due principally to upregulation of adenosine triphosphate–binding cassette proteins A1 and G5, plasma high density lipoprotein (HDL) cholesterol concentrations increased, as did relative secretion rates of biliary cholesterol. A smaller, more hydrophilic bile salt pool decreased intestinal cholesterol absorption. In this setting, hepatic cholesterol synthesis was downregulated, indicative of increased uptake of plasma cholesterol. After 56 days of high fat feeding, obesity was associated with leptin resistance, as evidenced by marked hyperleptinemia without downregulation of stearoyl CoA desaturase-1 or fatty acid synthase and by upregulation of hepatic cholesterol and bile salt synthesis. Hypercholesterolemia was attributable to overproduction and decreased clearance of large HDL1 particles. In conclusion, before the onset of obesity, preserved leptin sensitivity promotes biliary elimination of endogenous cholesterol in response to dietary fat. Leptin resistance due to obesity leads to a maladaptive response whereby newly synthesized cholesterol in the liver is eliminated via bile. (HEPATOLOGY 2005;41:887–895.)
Cholesterol elimination requires transport from peripheral tissues to the liver, principally by high density lipoproteins (HDL),1 for secretion into bile.2 Recent evidence in genetically obese rodents suggests that leptin, the protein product of the obesity gene (ob), may play a key role in promoting the biliary elimination of plasma cholesterol.
HDL cholesterol concentrations are elevated in ob/ob mice as a result of defective hepatic clearance and processing of HDL particles.3, 4 Indicative of a mechanistic role for leptin in HDL metabolism, these abnormalities are reversed by low doses of leptin, which do not reduce body weight.3, 4 Using lean (Fa/−) and obese (fa/fa) Zucker rats, we have shown that acute administration of leptin increases biliary secretion of plasma-derived cholesterol,5 whereas chronic leptin administration to ob/ob mice promotes elimination of cholesterol by regulating the enterohepatic circulation of bile salts.6 Leptin acts within the central nervous system to decrease hepatic bile salt synthesis.7 Reductions in both the size and hydrophobicity of the bile salt pool serve to decrease absorption of dietary cholesterol and reabsorption of biliary cholesterol within the small intestine.6
Although genetically obese rodents have helped to define the physiological functions of leptin, human obesity is largely due to chronic overnutrition.8 Elevated plasma leptin levels in obese individuals are correlated with body mass index, indicating resistance to leptin action.9 Certain inbred strains of mice, including C57BL/6J, become obese when challenged with a diet enriched with saturated fat.10 Under conditions of diet-induced obesity, plasma leptin concentrations in mice also vary in proportion to body fat content.11
Here we demonstrate a biphasic response of hepatobiliary lipid metabolism in C57BL/6J mice to saturated fat early in the course of diet-induced obesity. Our data indicate that before the onset of obesity, a hyperleptinemic response functions physiologically to promote biliary elimination of endogenous cholesterol. Diet-induced obesity and leptin resistance leads to a maladaptive response in which hepatic uptake of plasma cholesterol is impaired, and cholesterol synthesized de novo within the liver is used instead for biliary secretion.
Materials and Methods
Animals and Diets.
Male 6-week-old C57BL/6J mice were obtained from The Jackson Laboratory (Bar Harbor, ME). For 1 week, mice were fed a standard chow diet (4% fat, <0.02% cholesterol, LabDiet 5001; PMI Nutrition International, Brentwood, MO). Thereafter, they were fed for up to 56 days with a diet (Research Diets, New Brunswick, NJ) that was either high (catalog #D12451: 45% kcal from lard fat, 35% kcal from carbohydrate, 20% kcal from protein, ≤0.004% cholesterol) or low (catalog #D12450B: 10% kcal from lard fat, 70% kcal from carbohydrate, 20% kcal from protein, ≤0.004% cholesterol) in saturated fat content.
Biliary Secretion and Hepatic Lipid Metabolism.
Following anesthesia, bile flow was diverted for collection by cannulation of the gallbladder with a PE-10 polyethylene catheter (Becton Dickinson Primary Care Diagnostics; Becton Dickinson, Sparks, MD).6 Hepatic bile was collected for 2 hours.6 Mice were then euthanized via cardiac puncture for collection of blood and liver.
Bile Salt Pool Size and Composition.
Cholesterol Absorption and Fecal Bile Salt Excretion.
Human HDL2 (1.063 < ρ < 1.125) was isolated via buoyant density ultracentrifugation and radiolabeled with cholesteryl [1,2-3H (N)] hexadecyl ether (Perkin Elmer Life Sciences, Boston, MA) followed by 125I (NaI) (Amersham Biosciences, Piscataway, NJ).3 Mice were injected via their femoral veins with 10 mg protein of dual-labeled HDL, and blood samples were collected via retro-orbital bleeding as functions of time. Immediately following the final bleed at 24 hours, livers were homogenized to measure hepatic contents of 125I and 3H.
Magnetic Resonance Imaging.
Imaging was conducted using a 35-mm 1H coil. Eight slices of 2 mm thickness spanning the whole body of the mouse were acquired using a 9.4-Tesla magnet (GE Omega vertical wide bore system). The percentage of abdominal fat in total tissue was calculated using a histogram analysis to separate components with different image contrast.12
All experiments were approved by the Institutional Animal Care and Use Committee of the Albert Einstein College of Medicine.
Plasma cholesterol and triglyceride concentrations were determined enzymatically.6 Plasma lipoproteins were fractionated and quantified via fast protein liquid chromatography using a Superose 6 HR10/30 column (Amersham Biosciences).5 Hepatic concentrations of triglycerides and total as well as free cholesterol were determined via enzymatic assays.6 Biliary cholesterol and bile salts were measured using standard techniques.6 Concentrations and molecular species of phosphatidylcholines in bile were quantified via mass spectrometry.13
Leptin concentrations were determined via radioimmunoassay using a reagent kit designed to detect mouse leptin (Linco Research, St. Louis, MO).7
Microsomes were used to measure activities of 3-hydroxy-3-methylglutaryl (HMG) CoA reductase (EC 18.104.22.168.4), cholesterol 7α-hydroxylase (Cyp7A1) (EC 22.214.171.124), and acyl-CoA:cholesterol acyl transferase (Acat) (E.C. 126.96.36.199).6 Activity of hepatic lipase was measured in plasma of mice following heparin injection.12 Phospholipid transfer protein (PLTP) activity was measured using an assay kit (Cardiovascular Targets, New York, NY). Activity of lecithin:cholesterol acyl transferase (LCAT) activity was measured as the decrease in plasma-free cholesterol after 1 hour of incubation at 37°C.14
Western Blot Analysis.
Protein expression in liver was determined via Western blot analysis using polyclonal antibodies against mouse stearoyl CoA desaturase-1 (SCD-1) (kindly provided by Dr. Alan Tall, Columbia University, NY),15 scavenger receptor class B type I (SR-BI), and fatty acid synthase (Novus Biologicals, Littleton, CO). Pooled fast protein liquid chromatography fractions were concentrated 25-fold using a Microcon centrifugal filter (Millipore, Bedford, MA) for detection of apolipoproteins using appropriate antibodies (Biodesign International, Saco, ME).
Northern Blot Analysis.
Messenger RNA (mRNA) expression in liver was determined via Nothern blot analysis using complementary DNA probes that were generously provided by Drs. Richard Green, Northwestern University, Chicago, IL (Cyp7A1); Xian-Cheng Jiang, State University of New York-Downstate Medical Center, Brooklyn, NY (PLTP); and Nan Wang, Columbia University (adenosine triphosphate–binding cassette protein [Abc] a1). Quantification was achieved via densitometry and normalization to β-actin expression.
Quantitative Real-Time Polymerase Chain Reaction.
Gene-specific primers were used for quantification of low-density lipoprotein (LDL) receptor mRNA.16 Polymerase chain reaction was performed with an ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA).
Data are expressed as the mean ± SEM (n = 5 mice/group). Statistical significance of the differences between the means of experimental groups was determined using Student t test. Differences were considered statistically significant for two-tailed P < .05. Assays were performed simultaneously on samples following their collection at 28 days or 56 days. Because of interassay variability, results obtained at 28 days were not directly comparable to those obtained at 56 days.
Figure 1 displays the time course of diet-induced obesity in male C57BL/6J mice. Mice fed a high fat diet gained excess weight (Fig. 1A), becoming significantly heavier than mice fed a low fat diet at 35 days. Final body weights were 28.4 ± 0.7 g for low fat fed mice and 34.1 ± 2.0 g for high fat fed mice. Consistent with obesity, high fat fed mice were 20% to 28% heavier than low fat fed mice during the period from 35 to 56 days. Using the same diet compositions as in the current study, Van Heek et al.17 observed that C57BL/6J mice fed either the low fat or high fat diet consumed calories at essentially the same rate both before and after the onset of obesity. Therefore, weight gain was highly likely to have been attributable to the composition of the diet, as opposed to differences in feeding behavior. Figure 1B shows corresponding abdominal cross-sections via magnetic resonance imaging of C57BL/6J mice at 56 days. When quantified volumetrically, the percentage of abdominal fat in high fat fed animals was 1.6-fold higher compared with their low fat fed counterparts. Figure 1C gives plasma leptin concentrations at 28 and 56 days. Although body weights at 28 days did not differ significantly, plasma leptin concentrations were elevated by 1.3-fold in high fat fed mice. This difference was increased to 2.6-fold at 56 days. Because leptin downregulates expression of hepatic SCD-118 and fatty acid synthase,19 we determined the expression levels of these proteins in livers of mice as indicators of sensitivity to endogenous leptin. Livers of mice fed a high fat diet for 28 days expressed 37% less SCD-1 than low fat fed animals (see Fig. 1C) but returned to basal levels after 56 days of high fat feeding. Furthermore, consistent with control by leptin action, there was a 33% reduction in hepatic expression of fatty acid synthase at 28 days, but not at 56 days (data not shown).
As shown in Fig. 2A, high fat feeding increased plasma total cholesterol concentrations at 28 and 56 days without affecting triglyceride concentrations. Figure 2B-C displays the distribution of cholesterol among plasma lipoproteins. At 28 days (see Fig. 2B), cholesterol concentrations (mg/dL) contained in very low density lipoproteins remained unchanged (low fat, 8.5 ± 1.3; high fat, 7.9 ± 1.3). Because there was not a discrete demarcation between LDL and HDL1, these fractions were integrated as a single peak for purposes of calculating plasma concentrations. The combined LDL and HDL1 cholesterol concentrations tended to be higher in high fat fed mice at 28 days (low fat, 23.5 ± 2.9; high fat, 37.6 ± 5.5), but this change did not achieve statistical significance. By contrast, HDL cholesterol concentrations were increased 1.8-fold in high fat fed mice (low fat, 61.8 ± 4.4; high fat, 108.3 ± 16.7). This increase was associated with an increase in hepatic expression of Abca1 mRNA (right inset in Fig. 2B).
Although very low density lipoprotein cholesterol concentrations at 56 days were not influenced by diet (low fat, 14.6 ± 1.6; high fat, 13.3 ± 1.0), they were increased at 56 days compared with 28 days of feeding for both diets. This result most likely reflected adaptation of the control animals that were fed the low fat diet, which differs in composition from standard laboratory chow. At 56 days, high fat feeding led to a marked increase in cholesterol concentrations contained in LDL and HDL1 particles (low fat, 21.7 ± 1.8; high fat, 78.1 ± 5.6). This outcome was due principally to an increase in the proportion of HDL1 particles containing apolipoprotein (apo) E and apoA-I (left inset in Fig. 2C). There were more modest (nonsignificant) increases in HDL cholesterol concentrations (low fat, 73.8 ± 6.7; high fat, 119.7 ± 19.6) and Abca1 mRNA expression (right inset in Fig. 2C). Consistent with their larger sizes, HDL1 particles contained a higher molar percentage of cholesterol in esterified form when compared with HDL (HDL1, 52.0 ± 0.5; HDL, 32.5 ± 1.2).
To explore the metabolic basis for increased steady-state concentrations of HDL1 cholesterol in high fat fed mice, we examined key regulators of HDL1 production and clearance before and following the onset of diet-induced obesity (Table 1). In plasma of mice, HDL particles become enriched with lipids principally due to the activities of PLTP and LCAT. By contrast, hepatic lipase reduces HDL size and facilitates selective uptake of HDL-derived lipids. At 28 days, PLTP activity was decreased in mice fed the high fat diet, whereas activity of hepatic lipase was slightly higher and LCAT activity was unchanged. Consistent with transcriptional downregulation of PLTP by leptin,20 the reduction in PLTP activity at 28 days was associated with a 1.6-fold reduction in hepatic PLTP mRNA (not shown). At 56 days, accumulation of HDL1 was associated with a sharp increase in PLTP activity. However, this increase was not associated with a change in hepatic PLTP mRNA. Activity of LCAT was also increased at 56 days, whereas hepatic lipase activity tended to decrease (P < .07).
|Diet||28 Days||56 Days|
|PLTP (arbitrary units)||11.3 ± 0.3||9.6 ± 0.5*||8.9 ± 1.0||13.2 ± 1.1*|
|HL (arbitrary units)||5.4 ± 0.1||6.0 ± 0.1*||14.3 ± 0.7||11.3 ± 1.5|
|LCAT (μg/mL/h)||2.8 ± 1.0||4.9 ± 1.5||0.5 ± 0.1||1.5 ± 0.5*|
|Biliary secretion (nmol/h × 10−2)|
|Cholesterol||0.81 ± 0.06||1.01 ± 0.02||0.57 ± 0.02||0.47 ± 0.03*|
|Phospholipid||3.6 ± 0.4||3.8 ± 0.3||4.6 ± 0.5||4.3 ± 0.2|
|Bile salt||10.8 ± 0.5||8.4 ± 0.5*||19.4 ± 1.3||17.9 ± 1.0|
|Bile salt pool size (μmol)||12.4 ± 0.3||10.8 ± 0.6*||7.6 ± 0.3||9.1 ± 1.0|
|Hydrophobic index||−0.28 ± 0.02||−0.36 ± 0.03*||−0.33 ± 0.01||−0.31 ± 0.04|
|Cholesterol absorption (%)||65.6 ± 2.5||53.7 ± 1.4*||75.2 ± 1.8||64.4 ± 3.1*|
|Hepatic activity (pmol/mg/min)|
|HMG-CoA reductase||36.6 ± 5.6||19.0 ± 1.6*||26.0 ± 4.0||66.4 ± 11.0*|
|Acat||100.9 ± 5.7||67.8 ± 13.9||43.8 ± 3.8||57.6 ± 7.2|
|Cyp7A1||11.1 ± 0.4||11.3 ± 0.4||6.2 ± 0.3||7.9 ± 0.2*|
|Fecal bile salt excretion (μmol/d)||2.4 ± 0.3||1.4 ± 0.2*||1.4 ± 0.1||2.0 ± 0.2*|
Plasma accumulation of HDL1 particles has been observed due to genetic disruption of SR-BI21 and in the absence of functional leptin.3 Because HDL2 is a well-characterized ligand for the principal HDL receptor, SR-BI, we used this lipoprotein to test the function of hepatic SR-BI in mice fed high fat diets for 56 days. Both [3H]cholesteryl ether and 125I-labeled protein contained in HDL2 were removed at the same rates from plasma in low and high fat fed mice. Rates of appearance of [3H]cholesteryl ether and radiolabeled protein in livers of high fat fed mice were unchanged, as was the hepatic expression of SR-BI (data not shown). By contrast, mRNA encoding the LDL receptor, which plays a key role in clearance of HDL1 particles from plasma,22 was downregulated by 30% in mice fed a high fat diet for 56 days.
As shown in Table 1, biliary cholesterol secretion rates at 28 days were unchanged in the setting of a significant 1.3-fold increase in expression of Abcg5 mRNA in high fat fed mice (data not shown). Biliary secretion rates of phospholipid were not influenced by the high fat diet. A decrease in biliary bile salt secretion rates was observed in the absence of a reduction in Abcb11 mRNA expression levels. At 56 days, a decrease in the biliary secretion rate of cholesterol due to high fat feeding occurred without changes in Abcg5 mRNA expression, bile salt or phospholipid secretion rate, or Abcb11 mRNA expression (data not shown).
Five major bile salt species comprised more than 90 mol% of biliary bile salts of low fat fed mice at 28 days (tauromuricholates, 42.0 ± 2.1; tauroursodeoxycholate, 3.0 ± 0.3; taurocholate, 33.3 ± 2.0; taurochenodeoxycholate; 6.9 ± 2.2; taurodeoxycholate 4.4 ± 0.3). High fat feeding was associated with an increase in the proportion of the relatively hydrophilic bile salt tauroursodeoxycholate (4.4 ± 0.3) and a decrease in the proportion of the relatively hydrophobic bile salt taurodeoxycholate (2.5 ± 0.3). The only difference observed between mice fed low and high fat diets for 56 days was a reduced proportion of taurochenodeoxycholate (low fat, 4.2 ± 0.2; high fat, 2.2 ± 0.6). High fat feeding did not appreciably alter the fatty acyl compositions of phosphatidylcholines in bile.
Table 1 shows that there was a decrease in bile salt pool size in mice fed a high fat diet for 28 days. The hydrophobic index is a concentration-weighted average of HPLC-determined hydrophobicities of individual bile salts present in a mixture, which allows the overall hydrophobicity of a mixture of bile salts in the bile salt pool to be represented by a single value.23 Bile salt hydrophobic index was also decreased. There was no difference between the bile salt pool size or the hydrophobic index of mice fed the high fat and low fat diet for 56 days. Intestinal cholesterol absorption, which is regulated in part by bile salt pool size and hydrophobicity, was decreased by high fat feeding at both 28 and 56 days.
High fat feeding did not change the hepatic total or free or esterified cholesterol concentrations in livers at 28 or 56 days (data not shown). Whereas hepatic triglyceride concentrations (mg/g liver) were increased in mice fed a high fat diet for 28 days (low fat, 17.2 ± 2.4; high fat, 24.5 ± 2.0), there was no difference at 56 days (low fat, 23.0 ± 2.5; high fat, 17.3 ± 1.7). Table 1 shows hepatic activities of HMG-CoA reductase, Acat, and Cyp7A1. At 28 days, HMG-CoA reductase activity was reduced by 48%, and activity of Acat tended to decrease by 33% (P = .056). Whereas the acute effect of leptin administration is to downregulate Cyp7A124 through a central nervous system–mediated mechanism,7 the activity of Cyp7A1 was unchanged. However, rates of fecal bile salt excretion, which reflect steady-state bile salt synthesis, were decreased by high fat feeding at 28 days (see Table 1). High fat feeding for 56 days increased activities of HMG-CoA reductase and Cyp7A1, whereas an increase in Acat activity did not achieve statistical significance (P = .076). Consistent with upregulation of Cyp7A1, rates of fecal bile salt excretion were increased at 56 days.
Previous studies have characterized the development of obesity induced by dietary fat in mice according to responsiveness to exogenously administered leptin.17, 25, 26 Before becoming obese, mice fed a high fat diet remain fully responsive to exogenous leptin. The onset of obesity is associated with resistance to peripherally administered leptin, but retained sensitivity to leptin introduced directly into the central nervous system, apparently because high fat feeding reduces leptin transport across the blood–brain barrier.27
This study was designed to investigate the impact of dietary saturated fat on the biliary elimination of plasma cholesterol in C57BL/6J mice. As illustrated schematically in Fig. 3, the central finding was that the metabolic response to a diet rich in saturated fats depended on whether the animal had become obese. Multiple lines of evidence indicate that the key regulatory determinant was sensitivity or resistance to elevated plasma concentrations of endogenous leptin. Because our study builds on a model of diet-induced obesity in male C57BL/6J mice,17 the applicability of our findings to female mice is not known.
Despite the absence of excess weight gain at 28 days, plasma leptin concentrations were elevated. As measures of leptin sensitivity, we determined hepatic SCD-1 and fatty acid synthase expression, both of which are downstream targets of leptin action. Downregulation of these proteins extends the findings of Van Heek et al.,17 who reported that sensitivity to exogenous leptin was preserved in male C57BL/6J mice before the onset of diet-induced obesity. The responsiveness of target genes in this study indicates that sensitivity to endogenous leptin also remained intact.
Before the onset of obesity (Fig. 3A), high fat feeding increased plasma HDL cholesterol concentrations. In this setting, there was upregulation of Abca1 expression in the liver, which is a key determinant of plasma HDL cholesterol concentrations.28 Because saturated dietary fat and leptin administration also increase hepatic apoA-I synthesis,3, 29 leptin-induced production of HDL particles was the likely mechanism for the higher HDL cholesterol concentrations in high fat fed mice at 28 days.
Downregulation of hepatic HMG-CoA reductase suggests that leptin promoted hepatic uptake of plasma cholesterol and thereby decreased cholesterol biosynthesis. This result is consistent with previous findings that HMG-CoA reductase activity was decreased when leptin was administered via the periphery5, 6 or the central nervous system,7 and that leptin promotes hepatic clearance of HDL cholesterol via SR-BI.3, 24, 30 Possibly contributing to increased hepatic HDL clearance was a modest increase in hepatic lipase activity, which facilitates uptake of HDL-derived cholesterol into the liver.31 Consistent with preserved leptin sensitivity, PLTP, which increases HDL size by transferring phospholipids to the surface coat of HDL from remnant particles,32 was downregulated by a transcriptional mechanism. Downregulation of PLTP may have served to inhibit the formation of HDL1 particles and thereby promote cholesterol delivery to the liver. Taken together, these observations suggest that the hyperleptinemic response to 28 days of high fat feeding functioned to promote delivery of cholesterol to the liver by increasing both rates of HDL production and hepatic clearance. However, proof of this concept will require studies that demonstrate increased mass uptake of HDL cholesterol into the liver in response to leptin action.
The liver disposes of plasma cholesterol by secretion into bile unmodified or following conversion to bile salts. Whereas leptin per se downregulates Cyp7A1,7, 24 the lack of a decline in Cyp7A1 activity at 28 days most likely reflected a balance between leptin-mediated downregulation and upregulation due to activation of liver X receptor,which was evidenced by increased expression of the liver X receptor target genes, Abca1 and Abcg5.33 Notwithstanding this observation at a single time point, decreases in the bile salt pool size and the fecal bile salt excretion rate were indicative of an overall reduction rate of bile salt synthesis and suggest that the inhibitory effect of leptin predominated before the onset of obesity.
Under physiological conditions, biliary cholesterol secretion varies in proportion to the secretion rate and the hydrophobic index of secreted bile salts. Despite decreases in both bile salt secretion rate and hydrophobicity in mice fed a high fat diet for 28 days, the secretion rate of biliary cholesterol did not decline. This relative increase in cholesterol secretion was likely explained by upregulation of the Abcg5/Abcg8 canalicular cholesterol transporter,34 as reflected by increased Abcg5 mRNA. In addition, Acat activity tended to decrease in high fat fed mice at 28 days. Although downregulation by leptin has not previously been reported, inhibition of Acat has been shown to increase biliary cholesterol secretion.35
Consistent with a smaller, more hydrophilic bile salt pool6 and upregulation of Abcg5,36 cholesterol absorption was decreased in mice fed a high fat diet for 28 days. Collectively, these observations are consistent with the model in Fig. 3A, which proposes that before the onset of obesity in C57BL/6J mice, high fat feeding elicits a coordinated response that increases elimination of HDL-derived plasma cholesterol via bile. This response may be mechanistically attributed to sensitivity to elevated plasma leptin concentrations.
High fat feeding for 56 days was sufficient to induce obesity and was associated with hyperleptinemia and increased visceral fat deposition. Consistent with leptin resistance, hepatic SCD-1 and fatty acid synthase expression were no longer suppressed by high endogenous leptin levels. At this point in time (Fig. 3B), increases in plasma cholesterol concentrations were primarily attributable to a marked accumulation of HDL1 particles.
HDL1 is a large, lipid-rich subfraction of HDL that contains apoE as its major protein constituent.37 Increased concentrations of HDL1 particles in mouse plasma occur under conditions that include upregulation of PLTP activity38 and overexpression of LCAT.39 Indeed, we observed that a sharp increase in PLTP activity accompanied the onset of diet-induced obesity in mice, which was not attributable to transcriptional upregulation in liver. This finding is in agreement with the observation that plasma activity of PLTP in obese humans is regulated primarily by body fat mass.40 Although the mechanism is unknown, increased LCAT activity has also been associated with obesity.41, 42 Taken together, these data strongly suggest that increased rates of HDL1 formation contributed to hypercholesterolemia in obese mice.
In addition to increased production, decreased clearance appeared to play an important role in the accumulation of HDL1 particles. Targeted inactivation of the hepatic lipase gene increases HDL1 concentrations,43 and hepatic lipase is markedly downregulated in C57BL/6J ob/ob mice.44 In keeping with these observations, the activity of hepatic lipase tended to decline in high fat fed mice at 56 days. Sérougne et al.22 have demonstrated that HDL1 is bound and cleared by the LDL receptor and that SR-BI does not play an appreciable role. Downregulation of the LDL receptor provides additional mechanistic evidence that decreased clearance as well as overproduction contributed to hypercholesterolemia in obese mice.
Human obesity is associated with increased hepatic activities of HMG-CoA reductase, Cyp7A1, and Acat.45 High fat feeding for 56 days produced the same features in mice. Whereas obesity in humans is also characterized by hypersecretion of cholesterol into bile,2 we observed a modest reduction in biliary cholesterol secretion in obese mice. This outcome was most likely due to reductions in hepatic clearance of HDL-derived cholesterol and cholesterol absorption, which together comprise the principal extrahepatic sources of biliary cholesterol.2 The decrease in cholesterol absorption that occurred at 56 days was not attributable to reduced bile salt pool size or hydrophobicity and is not mechanistically explained by our current data. As illustrated in Fig. 3B, these findings suggest that diet-induced obesity was accompanied by a maladaptive response in which hepatic uptake of plasma cholesterol was impaired, and cholesterol synthesized de novo within the liver was eliminated instead.
In conclusion, whereas efficient assimilation and storage of triglycerides has contributed evolutionarily to survival and reproduction,46 it is attractive to speculate that homeostatic mechanisms also evolved to prevent cholesterol accumulation during fat consumption. Our findings suggest that leptin promotes cholesterol elimination in response to dietary fat and that obesity-associated increases in hepatic cholesterol and bile salt synthesis represent manifestations of resistance to leptin action.
- 2Pathogenesis of gallstones. In: ZakimD, BoyerTD, eds. Hepatology: A Textbook of Liver Disease. 4th ed. Philadelphia: W.B. Saunders, 2002: 1713–1743..
- 19Transcriptional regulation of fatty acid synthase gene by insulin/glucose, polyunsaturated fatty acid and leptin in hepatocytes and adipocytes in normal and genetically obese rats. Eur J Biochem 1999; 260: 505–511., , , .
- 46Gallstones. In: KingRA, RotterJI, MotulskyAG, eds. The Genetic Basis of Common Diseases. 2nd ed. Oxford, UK: Oxford University Press, 2002: 298–335., .