The authors have declared that no competing interests exist.
Conceived and designed the experiments: EL AKG AJV JMA FB. Performed the experiments: EL CPAAR GHP RO FB. Analyzed the data: EL GHP ME JA JK HSO AKG AJV JMA FB. Contributed reagents/materials/analysis tools: JA GHP JK HSO. Wrote the paper: EL ME AJV JMA FB.
The iminosugar N-(5′-adamantane-1′-yl-methoxy)-pentyl-1-deoxynoijirimycin (AMP-DNM), an inhibitor of glycosphingolipid (GSL) biosynthesis is known to ameliorate diabetes, insulin sensitivity and to prevent liver steatosis in ob/ob mice. Thus far the effect of GSL synthesis inhibition on pre-existing NASH has not yet been assessed. To investigate it, LDLR(−/−) mice were kept on a western-type diet for 12 weeks to induce NASH. Next, the diet was continued for 6 weeks in presence or not of AMP-DNM in the diet. AMP-DNM treated mice showed less liver steatosis, inflammation and fibrosis. Induction of fatty acid beta-oxydation was observed, as well as a reduction of plasma lipids. Our study demonstrates that AMP-DNM treatment is able to significantly correct pre-existing NASH, suggesting that inhibiting GSL synthesis may represent a novel strategy for the treatment of this pathology.
The metabolic syndrome represents a combination of health risk factors including abdominal obesity, insulin resistance, dyslipidemia and hypertension. Non Alcoholic Fatty Liver Disease (NAFLD) is the hepatic manifestation of the metabolic syndrome. NALFD includes a large variety of liver derangements ranging from simple fat accumulation in the parenchymal cells (steatosis) to non-alcoholic steatohepatitis (NASH) including inflammation and varying degrees of fibrosis. NAFLD is estimated to affect at least 20% of the general adult population and over 50% of the obese population
We and others have previously shown that two distinct classes of inhibitors of glucosylceramide (GlcCer) synthase, the rate limiting enzyme involved in glycosphingolipid (GSL) biosynthesis, improved glycemic control, decreased insulin resistance and reduced fatty liver development in animal models of obesity i.e. diet-induced obesity (DIO) mice and ob/ob mice
In the present study, LDLR(−/−) and APOE*3 Leiden mice, two models sensitive to liver steatosis were allowed to develop NASH for 12 weeks on high fat-high cholesterol diet and were subsequently treated with AMP-DNM for 6 weeks. We observed that despite the maintenance of the animals on a high fat-high cholesterol diet, AMP-DNM treatment reduced plasma lipids and that the steatosis, the inflammatory and fibrotic status were profoundly improved.
In the present study, 40 LDLR(−/−) mice were fed a western-type diet to induce NASH and were subsequently treated with two different doses of AMP-DNM to achieve the dosing level of 50 and 100 mg AMP-DNM. kg bw−1.day−1. AMP-DNM supplementation did not affect the behaviour of the animals. At the dose of 100 mg AMP-DNM, the bodyweight of the animals was reduced compared to the control group and food consumption slightly decreased after the switch of diet (
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. Plasma concentration of triglycerides (A), free fatty acids (FFA) (B) and cholesterol (C) after 18 weeks of chow diet (chow), 12 weeks of western-type diet (12w) or 18 weeks of western-type diet with or without AMP-DNM treatment at the indicated doses. *p<0.05; **p<0.01.
plasma (nmol.ml−1) | liver (nmol.g−1 liver) | |||
glccer | cer | glccer | cer | |
|
11.5±0.3** | 6.7±0.4** | 38.5±7.0 | 123.1±11.2 |
|
81.4±4.6 | 45.2±5.2 | 46.1±3.8 | 131.9±5.8 |
|
85.5±6.7 | 45.1±2.6 | 62.5±5.5 | 150.8±7.7 |
|
32.0±3.2** | 35.3±4.7 | 33.4±3.9 |
150.1±11 |
|
10.1±0.8** | 17.7±0.7** | 21.7±1.3 |
161.7±8 |
Data are expressed as mean ± SEM of 5 (plasma) and 10 (liver) mice.
, p<0.001.
liver (nmol.g−1 liver) | ||||
Laccer | Gb3 | GM2-gl | GM3 | |
|
16.4±1.3 | 7.2±0.6 | 255.9±19.6 | 19.8±8.5 |
|
22.8±1.2 | 6.3±0.8 | 308.7±22.5 | 23.2±3.4 |
|
16.5±1.1 |
2.8±0.3 |
338.8±14.5 | 24.6±4.1 |
|
14.6±1.5 |
0.9.±0.1 |
173.4±13.8 |
16.7±2.6 |
Data are expressed as mean ± SEM of 5 mice.
, p<0.01;
, p<0.001.
We then determined the effect of AMP-DNM on liver steatosis. First liver weight was significantly reduced in a dose-dependent fashion (25% reduction with 50 mg AMP-DNM and 32% reduction with 100 mg AMP-DNM, p<0.01). Analyses of morphology and fat content on liver sections stained with H&E (
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. (A) Representative photomicrographs of hematoxylin-eosin staining and Oil red O staining (B) of livers section after 18 weeks of chow diet (chow), 12 weeks of western-type diet (12w) or 18 weeks of western-type diet with or without AMP-DNM treatment at the indicated doses (original magnification ×10). (C) Triglyceride content and (D) cholesterol content in liver of animals for the indicated groups. *p<0.05; **p<0.01. #p<0.01 for esterified cholesterol.
AMP-DNM ameliorates glucose homeostasis and insulin sensitivity in obese rodents
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. (A) Glucose concentration and (B) insulin concentration in plasma of fasted animals. (C) Homeostasis model assessment (HOMA) index and (D) percentage of glycated heamoglobin 1c (HbA1c). *p<0.05; **p<0.01.
Sterol regulatory element binding protein 1c (SREBP1c) is a transcription factor activated by insulin and responsible for the transcription of numerous genes involved in fatty acid synthesis. Despite insulin resistance for the glucose homeostasis pathway, it has been reported that hyperinsulinemia triggers the activation of SREBP-1c and consequently fatty acid synthesis
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. Expression levels normalized to Acidic ribosomal phosphoprotein (36B4). *p<0.05; **p<0.01.
Next, we examined the effect of AMP-DNM treatment on hepatic inflammation (
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. (A) Representative photomicrographs of immunohistochemical staining of CD68 of the indicated mice (original magnification ×10). (B) Gene expression of inflammation markers. *p<0.05; **p<0.01.
Gene expression profiles of inflammatory markers were also monitored (
We also analysed expression levels of Glycoprotein Nonmetastatic Melanoma Protein B (GPNMB) also called osteoactivin, a new marker of monocyte to macrophage differentiation
An important feature in NASH is the progression towards liver fibrosis. To monitor this aspect, we stained sections with Sirius-red. In all groups of animals, positive collagen staining was observed and was localized near vessels of periportal and centrolobular regions. After 12 and 18 weeks of western-type diet, collagen staining in these two areas was higher than in animals on chow diet. Collagen content was slightly reduced as a consequence of AMP-DNM treatment (
Mice were fed a western-type diet for 18 weeks, and received in the last 6 weeks either 0, 50 or 100 mg AMP-DNM. (A) Representative photomicrographs of Sirius-red staining of liver sections of the indicated mice (original magnification ×10). (B) Representative photomicrographs of immunohistochemical staining of activated stellate cells (original magnification ×10). (C) Gene expression of fibrosis marker. *p<0.05; **p<0.01.
We examined also the impact of AMP-DNM treatment on fatty liver of APOE*3 Leiden mice that were exposed for 12 weeks to a 1% cholesterol, 15% fat diet. The switch of diet induced a slight reduction of the animal bodyweights treated with either 50 or 100 mg AMP-DNM. No major changes were observed concerning the food consumption (
Mice were fed a high cholesterol-high fat diet (1% cholesterol, 15% fat) for 12 weeks and were fed for 6 weeks more a western-type diet (0.25% cholesterol, 15% fat) supplemented with either 0, 50 or 100 mg AMP-DNM. Representative photomicrographs of hematoxylin-eosin staining (A) and Oil red O staining (B) of livers section after 12 weeks of high cholesterol-high fat diet (12w) or after 12 weeks of high fat-high cholesterol diet followed by 6 weeks of western-type diet (18w) with or without AMP-DNM treatment at the indicated doses (original magnification ×10). (C) Triglycerides content and (D) cholesterol content in liver of animals for the indicated groups. *p<0.05; **p<0.01.
We have previously demonstrated in leptin-deficient ob/ob mice that AMP-DNM treatment reduced liver steatosis associated with a restoration of liver insulin signaling
A limitation of the leptin-deficient ob/ob and DIO models is that the mice do not develop steatohepatitis or liver fibrosis as is observed in humans
The underlying mechanism by which AMP-DNM treatment is able to correct NASH manifestation is intriguing. In the LDLR(−/−) mice the most prominent ganglioside in liver is GM2-glycol, rather than GM3. AMP-DNM treatment led to a major reduction in GM2-glycol, accompany the correction of liver steatosis. It may be speculated that GM2-glycol, rather than GM3, causes insulin resistance in the LDLR(−/−) mice liver. Hyperinsulinemia has been proposed to promote liver steatosis
Of note, lowering of glycosphingolipids by specific genetic knock-down of GlcCer synthase in hepatocytes does not reduce liver triglycerides in mice exposed to a high fat diet
In the past years, several studies showed that ER stress response plays an important role in lipid metabolism and is linked to fatty liver disease
Liver natural killer T (NKT) cells are thought to be able to ameliorate steatosis and NKT cells numbers are reduced in steatotic livers of human subjects and ob/ob mice
AMP-DNM exerts a positive effect on cholesterol homeostasis by promoting biliary sterol secretion and fecal excretion
In conclusion, our study shows clearly that drug treatment is sufficient to swiftly correct the liver manifestations of NASH in two different animal models with improved insulin sensitivity, promotion of fatty acid beta-oxidation and reduction of inflammation.
AMP-DNM was synthesized as previously described
Experiments were performed with the approval of the local Ethical Committee for Animal Experiments.
40 female LDLR(−/−) mice (8–12 weeks-old) were fed a western-type diet (0.25% cholesterol, 15% fatty acids; Arie Blok, Woerden, the Netherlands) for 12 weeks to induce NASH and 10 mice were sacrificed. The 30 remaining mice were subdivided into 3 groups of 10 animals, one group continued the same diet for another 6 weeks, the second group received the same diet supplemented with 0.3 g AMP-DNM per kg diet to obtain the calculated dose of 50 mg AMP-DNM. kg bw−1. day−1 and the third group received 0.6 g AMP-DNM per kg diet to obtain 100 mg AMP-DNM. kg bw−1. day−1. In parallel, one group of 5 LDLR(−/−) mice was kept on chow diet. In a second experiment, 40 female APOE*3 Leiden (8–12 weeks-old) were fed a high fat-high cholesterol diet (1% cholesterol, 15% fatty acids) for 12 weeks and 10 mice were sacrificed. The 30 remaining mice continued with a western-type diet (0.25% cholesterol, 15% fatty acids) supplemented or not with two different doses of AMP-DNM to obtain the calculated dose of 50 mg and 100 mg AMP-DNM. kg bw−1. day−1.
Blood samples of non-fasted animals were collected via the tail vein at the time points indicated in the figures. At the end of the experiments, blood samples were collected by abdominal aorta puncture and plasma samples were stored at −20°C. Livers were quickly excised and weighed, and parts were snap-frozen in liquid nitrogen and stored at −80°C, or fixed in 10% buffered formalin and embedded in paraffin.
Blood glucose levels were determined in plasma of fasted animals using a hand-held Glucometer (Ascensia Elite; Bayer A.G., Leverkusen, Germany) one week before the sacrifice of the animals. Hemoglobin A1c levels were measured using a single measurement A1C now device (Metrika, Sunnyvale, CA). Insulin levels were determined by enzyme-linked immunosorbent assay (Crystal Chem Inc., Downers Grove, IL). Free cholesterol, total cholesterol, and triglycerides in liver samples were determined after lipid extraction according to Folch
Paraffin embedded liver sections (7 µM) were de-waxed and stained with hematoxylin-eosin for general histology, or with 0.2% picro-sirius red to detect fibrillar collagen deposits. To detect neutral lipids, cryostat sections of 7 µM were stained with 0.3% Oil Red O. For detection of macrophages/monocytes, a rat polyclonal anti-CD68 was used (Serotec, Oxford UK) and an anti-alpha smooth muscle actin (SMA) antibody to visualize activated stellate cells (1A4, Abcam, Cambridge, UK). Frozen sections of livers fixed in cold acetone were incubated one hour at room temperature with CD68 (1/100 dilution) or SMA antibody (1/100 dilution) followed by incubation for 30 min at room temperature with the corresponding secondary antibodies. Visualization of the complex was done with 3,3′-diaminobenzidine tetrahydrochloride (DAB, Immunologic, Duiven, The Netherlands) for 5 min. For all stainings, haematoxylin (Sigma-Aldrich, Zwijndrecht, the Netherlands) was used to counterstain. Primary antibodies were omitted in negative control samples. Images were captured with a Leica DFC 420 video camera.
Total RNA was isolated from 50–100 mg of liver tissue using Trizol reagent (Invitrogen, Breda, The Netherlands) and reverse transcribed with SuperScript II Reverse Transcriptase and random hexamers (Invitrogen, Breda, The Netherlands) after treatment with RQ1 Rnase-free Dnase (1 units/2 µg of total RNA, Promega, Leiden, The Netherlands). Gene expression analysis was performed on a Bio-Rad MyIQ Single-color Real-Time PCR Detection System by using the Bio-Rad IQ SYBR Green Supermix (Bio-Rad Laboratories Inc., Hercules, CA). Expression levels were normalized to Acidic ribosomal phosphoprotein (36B4).
Values presented in figures concerning the LDLR(−/−) mice represent mean ± SEM. Dunnett's comparison test was performed between 18 weeks controls and each other group using GraphPad Prism software. Values presented in figures concerning the APOE*3 Leiden mice represent mean ± SEM. Dunnett's comparison test was performed between baseline at 12 weeks and each other group using GraphPad Prism software *, p<0.05; **, p<0.01; ***, P<0.001.
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
(DOC)
We are grateful to Han Levels for helping with the lipoprotein profile; Nike Claessen for support and guidance.