SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Recent studies have reported that glycosphingolipids (GSLs) might be involved in obesity-induced insulin resistance. Those reports suggested that inhibition of GSL biosynthesis in animals ameliorated insulin resistance accompanied by improved glycemic control and decreased liver steatosis in obese mice. In addition, pharmacologic GSL depletion altered hepatic secretory function. In those studies, ubiquitously acting inhibitors for GSL biosynthesis have been used to inhibit the enzyme Ugcg (UDP-glucose:ceramide glucosyltransferase), catalyzing the first step of the glucosylceramide-based GSL-synthesis pathway. In the present study a genetic approach for selective GSL deletion in hepatocytes was chosen to achieve complete inhibition of GSL synthesis and to avoid possible adverse effects caused by Ugcg inhibitors. Using the Cre/loxP system under control of the albumin promoter, GSL biosynthesis in hepatocytes and their release into the plasma could be effectively blocked. Deletion of GSL in hepatocytes did not change the quantity of bile excretion through the biliary duct. Total bile salt content in bile, feces, and plasma from mutant mice showed no difference as compared to control animals. Cholesterol concentration in liver, bile, feces, and plasma samples remained unaffected. Lipoprotein concentrations in plasma samples in mutant animals reached similar levels as in their control littermates. No alteration in glucose tolerance after intraperitoneal application of glucose and insulin appeared in mutant animals. A preventive effect of GSL deficiency on development of liver steatosis after a high-fat diet was not observed. Conclusion: The data suggest that GSL in hepatocytes are not essential for sterol, glucose, or lipoprotein metabolism and do not prevent high-fat diet-induced liver steatosis, indicating that Ugcg inhibitors exert their effect on hepatocytes either independently of GSL or mediated by other (liver) cell types. HEPATOLOGY 2010

The liver exerts a central role in metabolic events. Its location interposed between the intestinal tract and the systemic circulation enables an exocrine secretion of bile acids and cholesterol to the intestine and release of serum proteins, coagulation factors, and lipoproteins into the blood system. The excretion and transport of bile acids and cholesterol is regulated by lipid transporters located in the canalicular membrane of hepatocytes. The role of plasma membrane glycosphingolipids (GSLs) as potential modulators of those events is poorly understood. Intracellularly, GSLs influence the transport and biosynthesis of proteins.1, 2 In the cellular membrane it has been reported that GSL can change membrane receptor function; they can negatively regulate insulin receptor activation in skeletal muscle.3 In fibroblasts lacking gangliosides, cell functions such as proliferation and migration are altered, caused by reduced epidermal growth factor (EGF) binding capacity to its receptor accompanied by reduced autophosphorylation of EGFR (receptor).4

With respect to cholesterol and bile acids whose secretion is regulated in the biliary lipid secretory network, a pivotal role of distinct genes coding for lipid transporters (ABC transporters, adenosine triphosphate [ATP] binding cassette) in hepatic canalicular membranes has been described. It was found that the excretion of bile acids is regulated by way of ABCB11. Mice lacking ABCB11 display a significantly reduced bile acid secretion.5 In humans, mutations in this gene lead to progressive familial intrahepatic cholestasis (PFIC) type 2.6, 7 Canalicular transporters ABCG5 and ABCG8 form heterodimers and are, among others, responsible for liberation and excretion of cholesterol from the lipid bilayer.8 Mutations in those genes cause a reduced biliary secretion of cholesterol and diminished neutral sterol excretion in feces paired with accumulation of high levels of sterols in plasma.9 Furthermore, an apoA-I-mediated efflux of cholesterol by way of ABCA1 has been described.10 Cells from patients with genetic defects in ABCA1 exhibit increased cellular cholesterol accumulation.11 A direct influence of GSLs affecting the ABCA1/apoA-I pathway could be demonstrated in fibroblasts derived from human inherited GSL-storage disease patients such as Fabry and Sandhoff, as well as GM1-gangliosidosis. Besides GSL storage, intracellular cholesterol accumulation was reported.12 In order to investigate the influence of GSL depletion on sterol homeostasis, the key enzyme of the GSL-synthesis pathway, UDP-glucose:ceramide glucosyltransferase (Ugcg) was functionally blocked by the inhibitor AMP-DNM (N-(5-adamantane-1-yl-methoxy-pentyl)-deoxynojirimycin) in HepG2 cells.13 Absence of GSL led to an activation of sterol regulatory element-binding protein (SREBP) 1 and 2 target genes accompanied by increased cholesterol synthesis in these cells.13 Similar results have been reported in vivo when mice were fed synthetic inhibitors for Ugcg. In addition to the increased cholesterol secretion in bile and feces, total bile flow was shown to be doubled, with similarly raised concentrations of bile salts and phospholipids.14

However, systemically applied Ugcg inhibitors exert a ubiquitous mode of action and the observed alterations might have occurred due to GSL depletion of several cell types. Therefore, a model of GSL deficiency was established in which the key enzyme of the GSL synthesis pathway, Ugcg, was genetically deleted solely in hepatocytes. To exclude GSL depletion in other epithelia but hepatocytes, such as gall bladder, bile duct, and gut epithelia, Ugcg was deleted in a cell-specific manner under control of the albumin promoter using the Cre/loxP system. The release of Cre-recombinase only in hepatocytes accompanied with highly restricted deletion of GSLs in these cells enabled a specific investigation of GSL function in the liver. In the present study, surprisingly, no functional deficits, either in cholesterol, bile acid, and glucose, or in lipoprotein-homeostasis, and food-induced liver steatosis, nor structural alterations within the liver lacking GSLs could be detected.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Transgenic Animals.

To inhibit GSL biosynthesis in hepatocytes in a cell-specific manner (Fig. 1A), mice with loxP flanked exons 6 to 8 of the Ugcg-gene locus were used as described15, 16 (Fig. 1B).

thumbnail image

Figure 1. GSL synthesis pathway, cloning strategy, and genotyping of mice. (A) Glucosylceramide-based GSL synthesis pathway in hepatocytes. (B) Cloning strategy, floxed exons 6 to 8 of Ugcgflox/flox mice were excised in Ugcgflox/flox//AlbCre mutant mice in hepatocytes in a cell-specific manner. (C) Genotyping of mutant mice by PCR. Product sizes of 383 bp and 505 bp indicated the Ugcg-floxed allele and the cell-specific AlbCre transgene, respectively. (D) Ugcg total liver mRNA was reduced in Ugcgflox/flox//AlbCre mice by ≈80%. (E) Results from the mRNA-analysis were confirmed by southern blot analysis. Appearance of a 4.3-kb band solely in liver DNA indicated the highly cell-specific manner of the Ugcg-gene deletion.

Download figure to PowerPoint

To generate mice with Ugcg-gene deletion in hepatocytes, homozygous floxed mice were bred with albumin-promoter driven constitutive-Cre-animals (AlbCre).17 In a second mating step, heterozygous floxed mice with the AlbCre transgene were mated again with homozygous floxed Ugcgflox/flox mice, resulting in hepatocyte-specific glycosylceramide synthase-deficient Ugcgflox/flox//AlbCre mice. Both mouse strains used were backcrossed to C57Bl6 background for at least seven generations.

All animal experiments were performed and approved according to federal laws (Regierungspräsidium Karlsruhe, Germany).

Genotyping of Mutant Mice.

Polymerase chain reaction (PCR) and southern blot analysis of tail biopsies were performed to confirm floxed and AlbCre mutant allele (Fig. 1C). For PCR the following primers were used: Ugcg-flox forward: 5′-GATCTAAGAGGGTGAAGGCGCA-3′; Ugcg-flox reversed: 5′-AAGCCAGTCCAGTCAAACCGAG-3′; PCR products of 383 basepairs (bp) indicated the floxed alleles (Fig. 1C, upper image).

The constitutive AlbCre transgene was determined using the primers: AlbCre forward: 5′-GGAAATGGTTTCCCGCAGAAC-3′; and AlbCre reverse: 5′-ACGGAAATCCATCGCTCGACC-3′. A 505 bp PCR-product was indicative of the AlbCre transgene (Fig. 1C, lower image).

In all experiments, homozygous floxed mice with constitutive Cre-recombinase activity in hepatocytes were compared with double-floxed mice without Cre as respective controls.

Messenger RNA (mRNA) Isolation and Gene Expression with Microarrays.

Livers were prepared from sacrificed animals. Adhering adipose tissue was removed. Total mRNA was extracted as described.18 Quantitative real-time reverse-transcription PCR (RT-PCR) was done using the LC-fast DNA Master SYBR Green I kit PCR for the LightCycler (Roche, Mannheim, Germany) as described and GAPDH was used as reference gene.16 Reverse transcription and complementary RNA (cRNA) synthesis as well as labeling and hybridization were performed as recommended by the manufacturer (Affymetrix). Three Ugcgflox/flox//AlbCre samples were compared with the respective number of controls. (For more details, see Supporting Materials and Methods.) The raw and normalized data are deposited in the Gene Expression Omnibus database (www.ncbi.nlm. nih.gov/geo/; Accession No. GSE19316).

Southern Blot Analysis.

The efficiency of Ugcg-gene deletion in liver and several other tissues from a Ugcgflox/flox//AlbCre mouse was determined by southern blot analysis after Bgl II restriction digestion (Fig. 1E). To detect Ugcg-flox- and Ugcg-null-fragments, a digoxigenin-labeled 3′-southern probe was amplified according to the protocol of the DIG Probe Synthesis Kit and DIG Luminescent Detection Kit (Roche) as described.16

Lipid Analysis by Thin Layer Chromatography (TLC).

Liver and feces were lyophilized and total lipids were isolated. (For further details, see Supporting Information.)

Triglyceride (TG) Quantification in Liver Extracts.

TGs were isolated and measured with a TG-Quantification Kit (BioVision, Mountain View, CA) according to the manufacturer's recommendations.

Plasma Sphingolipid Determination.

Blood was taken from mice by way of the retroorbital venous plexus and 10 μL of plasma were used for sphingolipid isolation and quantitation as described.19–21

Nano-Electrospray Ionization (ESI) Tandem Mass Spectrometry (MS/MS).

Analysis was performed with a triple quadrupole instrument (VG micromass model Quattro II) equipped with a nano-electrospray source and gold-sputtered capillaries as described.22, 23

Isolation of Hepatocytes.

Mice were anesthetized and the livers perfused by consecutive administration of liver perfusion medium and liver digestion medium according to the manufacturers protocol (Invitrogen, Mannheim, Germany). The liver capsule was removed and the hepatocytes carefully detached by pipetting up and down in hepatocyte wash medium (Invitrogen). Cells were filtered by way of a 100-μm cell strainer (Becton Dickinson, Heidelberg, Germany). The preparation was centrifuged at 50g for 3 minutes and the cell pellet was washed twice with liver wash medium. Finally, the cell pellet was lyophilized and lipids were extracted as described before.

Determination of Blood Parameters.

Alanine transferase (ALT), cholesterol, total protein, and TG were determined with a Hitachi Autoanalyzer. Blood glucose was measured from the tail vein with a blood glucose instrument (Ascensia Contour, Bayer, Leverkusen, Germany). Bile acids were determined with a bile acid kit (BI 3863, Randox, UK), according to the manufacturer's instructions. Sodium cholate at different concentrations was used as reference.

Mouse Diets.

Mice had access to chow diet containing 5% fat (Altromin, Lage, Germany; No. 1310) or alternatively Western type diet (HF-W), containing 15% fat and 0.25% cholesterol and atherosclerotic diet (HF-Ath), containing 15% fat, 1% cholesterol, and 0.5% sodium cholate (Altromin) ad libitum. In comparison to the chow diet, in HF-diets polysaccharide proportion (corn starch) was reduced for the increased content of fat (lard), cholesterol, and sodium cholate. (For further details, see Supporting Materials and Methods.) Mice under HF-W and HF-Ath diets were fed for 7 to 10 weeks with control diet and received the special diet thereafter for 8 weeks. Female as well as male mice were used in the experiments.

Glucose Tolerance Test.

Mice (males and females receiving chow diet for 8 weeks or HF-W; males, chow for 7 weeks plus HF-W for 7 weeks) were fasted for 14 hours. A pretest value was determined from tail vein blood. Glucose (2 g/kg) was injected intraperitoneally and blood glucose levels were monitored for 80 minutes in 20-minute intervals for animals with a chow diet and 100 minutes for animals that received HF-W.

Insulin Tolerance Experiment.

Mice were fasted for 14 hours. After intraperitoneal injection of insulin (1 unit/kg) blood glucose levels were monitored as described above.

Lipoprotein Determination.

Mice (males and females) were fed for 7 to 10 weeks with chow diet or for the same duration with chow, followed by 7 weeks HF-W. The distribution of total cholesterol over the different lipoproteins in serum was analyzed by fractionation of 30 μL of pooled serum using a Superose 6 column (3.2 × 30 mm, Smart-system; Pharmacia, Uppsala, Sweden) and measuring total cholesterol using enzymatic colorimetric assays (Roche Diagnostics) as described.24, 25 Precipath l was used as an internal standard.

Bile Collection.

Mice were anesthetized and the gall bladder was ligated to avoid false results by bile secretion from the gall bladder. The bile duct was cannulated and the direct biliary secretion from the liver was collected in microcapillaries (Hirschmann, Eberstadt, Germany) for 15 minutes.

Total bile acids were determined as described above. To quantify cholesterol in bile fluid, 5 μL were treated under mild alkaline conditions as described.16 After a desalting step, an aliquot of 20% from each sample was spotted on a high-performance TLC plate (VWR, Darmstadt, Germany) and chromatographed as described above.

In addition, cholesterol was determined using a colorimetric assay as described in the legend to Supporting Fig. S4.

Lipid Quantification by Densitometry.

Lipids were separated and visualized by TLC. TLC standards of cholesterol, sphingomyelin, GM2 (Sigma, Munich, Germany) as well ceramides C16 (Biomol, Hamburg, Germany) and C24 (Cayman, Ann Arbor, MI) in different concentrations were used. Lipid bands were scanned with a Shimadzu CS-9310PC TLC-scanner (Shimadzu Europe, Duisburg, Germany) at a wavelength of 580 nm.

Histochemistry and Electron Microscopy (EM).

Hematoxylin/eosin, PAS, and Goldner Trichrome stains on 3-μm sections of paraffin-embedded liver tissue as well as transmission electron microscopy were performed as described.15

Mouse numbers used were 14 mutant mice and 10 controls from the chow diet groups; for the HF-W diet, 8 mutant mice and 10 controls; 10 mutant mice and 8 controls for the HF-Ath.

For EM, total samples of mutants (n = 11) and respective controls (n = 11) were investigated.

Statistics.

All experiments were done at least twice with a minimum of four to five animals per group and statistics were calculated using Student's two-tailed unpaired t test.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Hepatocyte-Specific Ugcg-Gene Deletion.

Mutant Ugcgflox/flox//AlbCre mice and respective Ugcgflox/flox controls were born according to Mendelian inheritance; they lived to a normal age. Livers from offspring were characterized for deletion of the Ugcg-gene by quantitative RT-PCR and southern blot analysis. Both Ugcg-mRNA and DNA were decreased by 80% to 90%, respectively, in total liver homogenates (Fig. 1D,E). Residual mRNA and DNA can be explained to be derived from cells that are not affected by the Ugcg-gene deletion such as Kupffer cells, T cells, Ito cells, endothelial cells, and bile duct epithelia. Other organs besides the liver were not affected by the Ugcg-gene deletion (Fig. 1E).

Hepatocyte-Specific Deletion of GSL.

The results from mRNA and DNA determination could be further confirmed by GSL analysis. In the neutral fraction of Ugcgflox/flox “wildtype” mouse total liver lipid extract solely a marginal amount of GlcCer could be detected. Up-regulation of GalCer as compensation for lack of GlcCer was not observed (Supporting Fig. S1). In Ugcg-deficient mouse liver the content of GlcCer was decreased by ≈50% (Fig. 2A, Liver), whereas in isolated hepatocytes of Ugcgflox/flox//AlbCre mice the GlcCer was completely absent (Fig. 2A, Hepatocytes). The results from the neutral fraction GSL were confirmed within the acidic fraction (Fig. 2B) in which the main GSL synthesis product in mouse liver, the ganglioside GM2 (see, GSL synthesis pathway in mouse liver, Fig. 1A) was completely absent in Ugcg-deficient animals. A faint double band migrating at the height of ganglioside GM3 was still present (Fig. 2B, Liver). Analysis of acidic GSL from isolated hepatocytes of Ugcgflox/flox//AlbCre mice resulted in a complete absence of GM2 and GM3 (Fig. 2B, Hepatocytes).

thumbnail image

Figure 2. GSL-analysis by TLC and ESI-MS/MS. (A,B) GSL extracts of total liver and isolated hepatocytes (the amount loaded on TLC corresponded to 1 mg liver dry weight and hepatocyte extract to 0.25 mg protein dry weight, for running and staining conditions, see Supporting Materials and Methods). (A) Only a marginal proportion of GlcCer was found within the neutral GSL fraction as the exclusive GSL in liver. The total content of GlcCer in Ugcg-deficient mice was approximately halved. However, hepatocytes of Ugcgflox/flox//AlbCre did not express any neutral GSL; *nonspecific, GSL negative band in hepatocytes from cell culture media (phenol red). (B) Although GSL of total liver extracts from mutant and control mice still contained ganglioside GM3, this ganglioside was depleted in isolated hepatocytes, indicating an admixture of cells others than hepatocytes in total liver extracts. (C,D) ESI-MS/MS analysis of total liver extracts. (C) Extracts from Ugcgflox/flox control liver contained GM2 as the major GSL. (D) GM2 was completely depleted in Ugcgflox/flox//AlbCre mouse liver. Remnants of GM3-derivatives might be explained by admixtures of cells in the liver which are not affected by Ugcg-gene deletion, such as T cells, Kupffer cells, vascular cells, and bile duct epithelia. (E) Plasma GSL were derived from liver as shown by their absence in mutant animals. As compensation, synthesis of SM is upregulated.

Download figure to PowerPoint

The data from TLC analysis were confirmed by mass spectrometry of total liver extracts. Here it could be seen that the GM2 of the liver is ornamented with N-glycolyl neuraminic acid (Fig. 2C), whereas ganglioside GM3, still present in Ugcg-deficient mice, contained predominantly N-acetyl neuraminic acid (Fig. 2D). These results and the fact that isolated hepatocytes are depleted from GSL are indicative that ganglioside GM3 and residual GlcCer in the neutral fraction are most likely derived from cells others than hepatocytes such as T cells, Kupffer cells, vascular cells, and bile duct epithelia.

Plasma Glycolipids Originate from Liver.

Investigation of freshly prepared plasma samples from wildtype mice showed the presence of small amounts of GlcCer. The predominant GSL could be characterized as GM2 (Fig. 2E). These results almost copy the analysis of extracts from liver and hepatocytes, suggesting that plasma GSL are derived from hepatocytes. As consequence of GSL depletion, theoretical accruing ceramides generated a shift in the GSL/SM pathway (Fig. 1A) and led to increased sphingomyelin production in mutant mice (Fig. 2E).

Liver Sphingomyelin, Ceramide, Cholesterol, and Phospholipids.

Livers of Ugcgflox/flox//AlbCre mice showed a modest but significantly increased sphingomyelin concentration compared to their respective controls (Fig. 3A,A′). Within the free ceramides, only nonhydroxylated fatty acids could be detected by TLC (Fig. 3B). No quantitative difference between Ugcgflox/flox controls and Ugcg-deficient mice were seen (Fig. 3B′). No significant differences could be seen between control liver- and GSL-depleted liver total cholesterol content (Fig. 3C,C′) and phospholipid patterns (Supporting Fig. 2).

thumbnail image

Figure 3. Sphingolipid and cholesterol analysis in liver. (A-C) TLC analysis of sphingomyelin (SM, corresponded to 1 mg dry liver per lane), ceramides (1 mg dry liver per lane), and cholesterol (0.2 mg dry liver per lane), from total liver extracts (for running and staining conditions, see Supporting Materials and Methods). (A′-C′) Quantification by densitometry. SM was significantly elevated in mutant livers, whereas ceramides and cholesterol showed no alteration in their expression (*GM3 of nonhepatic origin; shown are mean values ± SEM).

Download figure to PowerPoint

Liver Histochemistry, Weight Relations, and Lipid Content Under Special Diets.

Liver structures and hepatocyte shape were unaltered in GSL-deficient liver (Fig. 4A). After treatment of animals with HF-W for 8 weeks, steatosis was seen in both the mutant and control animals (Fig. 4A). Steatosis and TG storage was further elevated after prolonged HF-W application (Supporting Fig. S3). Feeding of HF-Ath led to extreme lipid storage, but again with no difference comparing control with mutant livers (Fig. 4A). Mice that received HF-Ath showed a severe steatosis that was predominantly caused by accumulation of cholesterol (Fig. 4B). Liver TG-concentrations were increased as well under this diet (Fig. 4C) without apparent differences between Ugcgflox/flox//AlbCre mutants and controls.

thumbnail image

Figure 4. Liver morphology, weight relation, and lipid content under special animal diets. (A) Mice were fed ad libitum for 7 to 10 weeks with chow-diet (males), HF-W (9 to 10 weeks chow and 8 weeks HF-W, females), or HF-Ath (7 to 8 weeks chow plus 8 weeks HF-Ath, females). Animals under a chow diet showed a normal liver morphology. Liver steatosis occurred after treatment with HF-W and increased drastically in animals that received HF-Ath. However, with hematoxylin/eosin staining no difference could be observed within each feeding group between control and mutant mice. (B) Mice fed with HF-Ath displayed a severe steatosis predominantly caused by accumulating cholesterol in hepatocytes of mutant animals and to a similar extent also in controls. (C) Liver TG concentrations increased in both in the HF-W- and even stronger in the HF-Ath-fed group when compared with chow-fed animals. No significant differences between liver GSL-depleted mice and controls were seen. (D,E) Total BW and liver-to-BW ratios appeared unaffected comparing mutant with control mice for all three diets applied. In animals fed with HF-Ath, both mutants and controls showed a drastic increase in the percentage of liver to BW, supporting the histochemistry data.

Download figure to PowerPoint

Total body weight (BW) and liver-to-BW ratios were unaffected for all three diets applied comparing mutant with control mice (Fig. 4D,E; the time course of the weight gain of animals under HF-W can be seen in Supporting Fig. S3). In mutant and control animals with HF-Ath, a drastic increase in the ratio of liver to BW was seen supporting the histologic results shown in Fig. 4A.

EM Depicts Normal Liver Structures.

Membrane and microvilli of liver canaliculi showed an inconspicuous structure in control as well as in mutant mice (Fig. 5A,B). In addition, hepatocyte organelle arrangement and appearance remained unaltered with respect to number and structure in GSL-deficient animals.

thumbnail image

Figure 5. EM of liver mice with a chow diet. (A) Section of a control liver. (B) Ugcgflox/flox//AlbCre mutant liver. The hepatic canaliculi (secretory microchannels of the liver) were regularly formed in the mutant animals (white arrows). Cell organelles (ER, endoplasmic reticulum; M, mitochondria) were inconspicuously distributed within the cytoplasm of hepatocytes. Glycogen, G, peroxisomes, P, and fat-droplets, F, could be found in control and mutant mice in similar manner.

Download figure to PowerPoint

Blood Chemistry Did Not Change in GSL-Deficient Mice.

As investigations of blood parameters of animals under chow diet reflected no significant differences (Fig. 6A-F), mice were adjusted to a high fat/high cholesterol-containing with and without sodium cholate (HF-W and HF-Ath). GSL-deficient animals under HF-W and HF-Ath had no significantly different plasma cholesterol concentrations (Fig. 6C), triglycerides (Fig. 6D), and blood glucose levels (Fig. 6A) when compared to their respective controls. Mice that received HF-Ath became hypercholesterolemic (Fig. 6C). Plasma ALT was drastically increased in the HF-Ath-fed mouse group without showing differences between mutant and control animals (Fig. 6B). This result correlated well with liver histochemistry (Fig. 4). Bile acids showed a slight but significant increase within the HF-Ath treated mutant Ugcgflox/flox//AlbCre mice (Fig. 6E). Plasma protein content remained unchanged (Fig. 6F), implying that the production of the main plasma protein albumin (≈60% of total proteins) by the liver did not differ between controls and mutant animals in all diets applied.

thumbnail image

Figure 6. Blood chemistry. Plasma parameters were investigated in control and mutant mice (males and females) at the age of 8 to 10 weeks (chow diet, [controls] n = 11; [Ugcgflox/flox//AlbCre] n = 10) or 16 to 18 weeks when animals had received HF-W for an additional 8 weeks ([controls] n = 13; [Ugcgflox/flox//AlbCre] n = 11) or HF-Ath ([controls] n = 14; [Ugcgflox/flox//AlbCre] n = 15). None of the investigated blood parameters such as glucose (A), alanine transferase ALT (B), cholesterol (C), triglycerides (D), and total plasma protein values (F) showed significant differences between Ugcgflox/flox controls and Ugcgflox/flox//AlbCre mutants. Solely plasma bile acid concentrations in mutant animals were slightly but significantly elevated (E) (shown are mean values ± SEM).

Download figure to PowerPoint

GSL-Deficient Mice Had Normal Glucose Tolerance.

In order to investigate whether liver devoid of GSL would influence the metabolism of intraperitoneally applied glucose, blood-glucose levels were monitored at various timepoints after glucose challenge. Neither in the chow diet group (Fig. 7A) nor within the HF-W animals (Fig. 7B) were differences in the time-dependent blood glucose concentrations seen between wildtype- and Ugcg-deficient mice. However, in the HF-W group both Ugcg-deficient and control mice had higher blood glucose concentration and a slower rate of decrease. After intraperitoneal insulin injection no significant difference of blood glucose levels between either group could be seen (data not shown).

thumbnail image

Figure 7. Glucose tolerance and lipoprotein determinations. (A,B) Blood glucose measurement after intraperitoneal injection of glucose (2 g/kg). In both groups chow diet-fed mice (females and males) (A) and animals that received HF-W (males) (B) did not demonstrate altered sensitivity to glucose (mutant mice n = 6; controls n = 8). However, the blood glucose concentration in HF-W-fed mice increased in both mutants and controls, indicating a lowered glucose tolerance under high-fat diet treatment. (C,D) Measurements of lipoprotein profiles in plasma did not reflect a significant difference in HDL, LDL, and VLDL lipoprotein levels in plasma from GSL-deficient mice and controls with chow (C; controls/mutant n = 4, males) or HF-W (D; controls/mutant n = 8, males) (shown are mean values ± SEM).

Download figure to PowerPoint

One major function of hepatocytes is the synthesis and secretion of lipoproteins in which cholesterol is packed and transported from the liver into the blood stream or taken up as very low density (VLDL), low density (LDL), and high density (HDL) lipoproteins. A significant increase of lipoprotein levels could not be seen in the GSL-deficient mice as compared to their respective controls in either the chow-diet and the HF-W groups (Fig. 7C,7D).

Bile Flow, Bile Acid, and Cholesterol Secretion Remains Unaltered in GSL-Depleted Hepatocytes.

Because it has been reported that systemically applied synthetic Ugcg-inhibitors had a significant effect on the secretion of bile and bile salts, our aim was to verify whether a depletion of GSL in hepatocytes could be responsible for the observed phenomenon. Unexpectedly, cholesterol levels secreted in bile and excreted in feces (Fig. 8A,A′; Supporting Figs. S4 and 8B,8B′, Supporting Fig. S5) were unaltered in Ugcg-deficient mice as compared to controls. Determination of bile flow, bile salts in plasma, and feces from Ugcgflox/flox controls and Ugcgflox/flox//AlbCre animals did not demonstrate significant differences between either group (Fig. 8C-E).

thumbnail image

Figure 8. Sterol secretion in bile and feces. (A/A′,B/B′) Determination of total cholesterol concentration in bile (A, TLC; A′, densitometry; amount loaded on TLC 1 μL bile) and feces (B, TLC; B′, densitometry; corresponding to 1 mg feces dry weight) did not reveal significantly increased levels in mutant mice. (C-E) Also, measurements of bile flow (C), bile acids in bile (D), and bile acids in feces resulted only in marginal changes in mice with GSL-depleted hepatocytes (TLCs show one representative experiment from two; quantifications in graphs include all animals; controls n = 7, mutant n = 8; male mice were used in the experiment shown).

Download figure to PowerPoint

Affymetrix Gene Array.

Gene array data suggested that there were no significant alterations in the expression of genes in the liver, including ceramide transport protein (Cert) and SM-synthase when mRNA of mutant mice was compared with respective controls at a significance level of P ≤ 0.05 (Supporting Fig. S6).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

GSLs, in particular gangliosides such as GM3, are known to directly influence membrane receptor activation3 or might be binding partners for glycoproteins on the cellular membrane to stabilize cell-cell interactions.26 Cholesterol and bile acid homeostasis are regulated by lipid transporters located in the hepatocyte canalicular membrane.27, 28 One focus of our investigation was to see whether depletion of membrane integrated GSL by deletion of the enzyme Ugcg, decisively involved in the initial GSL synthesis step, would influence the concentration of sterol derivatives such as cholesterol and bile salts in bile fluid, feces, and plasma.

In a previous study the use of the synthetic inhibitor of Ugcg AMP-DNM in liver derived HepG2 cells led to a decrease of GSL-production of ≈75% and concomitantly to an increased cholesterol production rate in these cells.13 Application of AMP-DNM on these cells activated SREBP 1 and 2, triggering key genes in the sterol synthesis pathway, and consecutively resulted in elevated sterol synthesis.13

In a second approach, AMP-DNM was applied to mice in vivo.14 In this study an increased biliary secretion of bile salts, cholesterol, and phospholipids could be shown. In addition, in feces elevated levels of cholesterol could be found dependent on the amount of inhibitor used for GSL synthesis blockage, although the reduction of glucosylceramide synthesis was ≈50%.29

In the present work a genetic GSL-depletion was performed by the use of the Cre/loxP-system under control of the albumin promoter. Consequently, a complete depletion of GSL in hepatocytes was achieved, whereby other cells were not affected by the Ugcg-gene-deletion. However, in contrast to the two previous reports, neither the quantity of bile secretion nor bile, fecal, and plasma sterol concentrations were altered. Two possibilities may explain the discrepancy between both: on the one hand the pharmacologic “inhibitor-model,” and on the other hand the “genetic GSL-deficiency-model.” Systemic application of AMP-DNM silences GSL-biosynthesis ubiquitously and does not solely affect hepatocytes. Because Ugcg is expressed in almost all cells, a variety of processes may have accompanied the effect that finally led to an increased sterol synthesis and secretion by the hepatocytes in vitro and in vivo. Second, as the depletion of GSL appears to be quite low (≈50%) in the in vivo model, it may be speculated that AMP-DNM per se and not the moderate reduction of GSLs has been responsible for the effects observed.

Recently, the role of GSL in the development of liver steatosis in obese mice was investigated. By use of two different types of inhibitors for Ugcg, a reduction in glucose sensitivity, and improved liver steatosis in ob/ob mice was shown.30, 31 A similar phenotypic difference could not be observed in this liver-specific genetic Ugcg deletion model. In Ugcgflox/flox//AlbCre animals, neither liver morphology nor glucose tolerance was altered, as compared to their Ugcgflox/flox litters. By feeding animals with high fat / high cholesterol-containing diets, with and without sodium cholate, increased liver steatosis occurred to a similar extent in both the hepatocyte GSL-deficient as well as control animals independent from the animal diet applied. The property of Ugcg inhibitors to act ubiquitously on GSL biosynthesis obviously targeted several cell types that may then have interacted to achieve a reduction of liver steatosis. A direct GSL-independent influence of those inhibitors, however, may have to be considered.

It can be concluded that hepatocytes depleted of GSL accompanied by loss of plasma-GSL do not change basic liver function with respect to sterol, glucose, and lipo-(protein) homeostasis. In addition, GSL-deficiency in hepatocytes does not prevent high fat-induced liver steatosis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We thank Professor G. Schütz for providing the AlbCre mice, and Gabi Schmidt, Claudia Schmidt, and Martina Volz for excellent technical support.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Sprong H, Degroote S, Claessens T, van Drunen J, Oorschot V, Westerink BH, et al. Glycosphingolipids are required for sorting melanosomal proteins in the Golgi complex. J Cell Biol 2001; 155: 369-380.
  • 2
    Tamboli IY, Prager K, Barth E, Heneka M, Sandhoff K, Walter J. Inhibition of glycosphingolipid biosynthesis reduces secretion of the beta-amyloid precursor protein and amyloid beta-peptide. J Biol Chem 2005; 280: 28110-28117.
  • 3
    Yamashita T, Hashiramoto A, Haluzik M, Mizukami H, Beck S, Norton A, et al. Enhanced insulin sensitivity in mice lacking ganglioside GM3. Proc Natl Acad Sci U S A 2003; 100: 3445-3449.
  • 4
    Liu Y, Su Y, Wiznitzer M, Epifano O, Ladisch S. Ganglioside depletion and EGF responses of human GM3 synthase-deficient fibroblasts. Glycobiology 2008; 18: 593-601.
  • 5
    Wang R, Salem M, Yousef IM, Tuchweber B, Lam P, Childs SJ, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad Sci U S A 2001; 98: 2011-2016.
  • 6
    Gerloff T, Stieger B, Hagenbuch B, Madon J, Landmann L, Roth J, et al. The sister of P-glycoprotein represents the canalicular bile salt export pump of mammalian liver. J Biol Chem 1998; 273: 10046-10050.
  • 7
    Strautnieks SS, Bull LN, Knisely AS, Kocoshis SA, Dahl N, Arnell H, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 1998; 20: 233-238.
  • 8
    Small DM. Role of ABC transporters in secretion of cholesterol from liver into bile. Proc Natl Acad Sci U S A 2003; 100: 4-6.
  • 9
    Yu L, Hammer RE, Li-Hawkins J, Von Bergmann K, Lutjohann D, Cohen JC, et al. Disruption of Abcg5 and Abcg8 in mice reveals their crucial role in biliary cholesterol secretion. Proc Natl Acad Sci U S A 2002; 99: 16237-16242.
  • 10
    Oram JF, Lawn RM, Garvin MR, Wade DP. ABCA1 is the cAMP-inducible apolipoprotein receptor that mediates cholesterol secretion from macrophages. J Biol Chem 2000; 275: 34508-34511.
  • 11
    Bodzioch M, Orso E, Klucken J, Langmann T, Bottcher A, Diederich W, et al. The gene encoding ATP-binding cassette transporter 1 is mutated in Tangier disease. Nat Genet 1999; 22: 347-351.
  • 12
    Glaros EN, Kim WS, Quinn CM, Wong J, Gelissen I, Jessup W, et al. Glycosphingolipid accumulation inhibits cholesterol efflux via the ABCA1/apolipoprotein A-I pathway: 1-phenyl-2-decanoylamino-3-morpholino-1-propanol is a novel cholesterol efflux accelerator. J Biol Chem 2005; 280: 24515-24523.
  • 13
    Bijl N, Scheij S, Houten S, Boot RG, Groen AK, Aerts JM. The glucosylceramide synthase inhibitor N-(5-adamantane-1-yl-methoxy-pentyl)-deoxynojirimycin induces sterol regulatory element-binding protein-regulated gene expression and cholesterol synthesis in HepG2 cells. J Pharmacol Exp Ther 2008; 326: 849-855.
  • 14
    Bijl N, van Roomen CP, Triantis V, Sokolovic M, Ottenhoff R, Scheij S, et al. Reduction of glycosphingolipid biosynthesis stimulates biliary lipid secretion in mice. HEPATOLOGY 2009; 49: 637-645.
  • 15
    Jennemann R, Sandhoff R, Langbein L, Kaden S, Rothermel U, Gallala H, et al. Integrity and barrier function of the epidermis critically depend on glucosylceramide synthesis. J Biol Chem 2007; 282: 3083-3094.
  • 16
    Jennemann R, Sandhoff R, Wang S, Kiss E, Gretz N, Zuliani C, et al. Cell-specific deletion of glucosylceramide synthase in brain leads to severe neural defects after birth. Proc Natl Acad Sci U S A 2005; 102: 12459-12464.
  • 17
    Postic C, Magnuson MA. [Role of glucokinase (GK) in the maintenance of glucose homeostasis. Specific disruption of the gene by the Cre-loxP technique]. J Annu Diabetol Hotel Dieu 1999: 115-124.
  • 18
    Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156-159.
  • 19
    Ghauharali-van der Vlugt K, Langeveld M, Poppema A, Kuiper S, Hollak CE, Aerts JM, et al. Prominent increase in plasma ganglioside GM3 is associated with clinical manifestations of type I Gaucher disease. Clin Chim Acta 2008; 389: 109-113.
  • 20
    Groener JE, Poorthuis BJ, Kuiper S, Helmond MT, Hollak CE, Aerts JM. HPLC for simultaneous quantification of total ceramide, glucosylceramide, and ceramide trihexoside concentrations in plasma. Clin Chem 2007; 53: 742-747.
  • 21
    Neville DC, Coquard V, Priestman DA, te Vruchte DJ, Sillence DJ, Dwek RA, et al. Analysis of fluorescently labeled glycosphingolipid-derived oligosaccharides following ceramide glycanase digestion and anthranilic acid labeling. Anal Biochem 2004; 331: 275-282.
  • 22
    Sandhoff R, Geyer R, Jennemann R, Paret C, Kiss E, Yamashita T, et al. Novel class of glycosphingolipids involved in male fertility. J Biol Chem 2005; 280: 27310-27318.
  • 23
    Sandhoff R, Hepbildikler ST, Jennemann R, Geyer R, Gieselmann V, Proia RL, et al. Kidney sulfatides in mouse models of inherited glycosphingolipid disorders: determination by nano-electrospray ionization tandem mass spectrometry. J Biol Chem 2002; 277: 20386-20398.
  • 24
    Out R, Hoekstra M, Habets K, Meurs I, de Waard V, Hildebrand RB, et al. Combined deletion of macrophage ABCA1 and ABCG1 leads to massive lipid accumulation in tissue macrophages and distinct atherosclerosis at relatively low plasma cholesterol levels. Arterioscler Thromb Vasc Biol 2008; 28: 258-264.
  • 25
    Out R, Jessup W, Le Goff W, Hoekstra M, Gelissen IC, Zhao Y, et al. Coexistence of foam cells and hypocholesterolemia in mice lacking the ABC transporters A1 and G1. Circ Res 2008; 102: 113-120.
  • 26
    Pan B, Fromholt SE, Hess EJ, Crawford TO, Griffin JW, Sheikh KA, et al. Myelin-associated glycoprotein and complementary axonal ligands, gangliosides, mediate axon stability in the CNS and PNS: neuropathology and behavioral deficits in single- and double-null mice. Exp Neurol 2005; 195: 208-217.
  • 27
    Ikonen E. Cellular cholesterol trafficking and compartmentalization. Nat Rev Mol Cell Biol 2008; 9: 125-138.
  • 28
    Lo Sasso G, Petruzzelli M, Moschetta A. A translational view on the biliary lipid secretory network. Biochim Biophys Acta 2008; 1781: 79-96.
  • 29
    Aerts JM, Ottenhoff R, Powlson AS, Grefhorst A, van Eijk M, Dubbelhuis PF, et al. Pharmacological inhibition of glucosylceramide synthase enhances insulin sensitivity. Diabetes 2007; 56: 1341-1349.
  • 30
    Zhao H, Przybylska M, Wu IH, Zhang J, Maniatis P, Pacheco J, et al. Inhibiting glycosphingolipid synthesis ameliorates hepatic steatosis in obese mice. HEPATOLOGY 2009; 50: 85-93.
  • 31
    Bijl N, Sokolovic M, Vrins C, Langeveld M, Moerland PD, Ottenhoff R, et al. Modulation of glycosphingolipid metabolism significantly improves hepatic insulin sensitivity and reverses hepatic steatosis in mice. HEPATOLOGY 2009; 50: 1431-1441.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_23545_sm_suppinfo.doc3925KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.