Introduction

Nonalcoholic fatty liver disease, which is one of the most prevalent chronic liver diseases worldwide, consists of a spectrum of conditions that includes nonalcoholic steatohepatitis (NASH).1, 2, 3 Although simple steatosis may undergo an indolent natural course, when it is accompanied by advanced liver fibrosis it often takes a similar clinical course to that of chronic hepatitis C virus infection with advanced fibrosis.4 However, with all of these possible liver-associated morbidities and mortalities, a definitive treatment that would stop or reverse the progression of the disease has yet to be discovered.

Omega-3 polyunsaturated fatty acid (omega-3) is clinically used for treating hypertriglyceridemia.5 There is evidence that omega-3 suppresses hepatic triglyceride (TG) accumulation and reduces inflammation, thereby alleviating NASH in animal models.6, 7, 8, 9 However, solid clinical evidence of the benefit of omega-3 in NASH is lacking, even though the absence of significant adverse effects of omega-3 makes it convenient to use as a supplement.10, 11, 12

Ursodeoxycholic acid (UDCA) is reported to have a beneficial role in experimental NASH because of its anti-inflammatory, hepato-protective and antioxidant effects.13, 14, 15, 16 However, clinical trials have failed to show meaningful improvement in liver inflammation or fibrosis in NASH patients treated with UDCA.17, 18 Although negative results may be attributable to improperly constructed studies, including suboptimal drug dosage and study duration, it has been suggested that a combination of UDCA with other agents, such as vitamin E, might result in clinically meaningful outcomes;19 however, a study showed that long-term consumption of high doses of vitamin E (daily dose of 400 IU) increased mortality.20 Safety should be a concern when combination therapy is tested.

Agents that demonstrate beneficial effects on experimental NASH often fail to show significant improvement in human studies. Therefore, a combination therapy that is safe and more effective than either agent alone might have a better outcome in clinical settings. Thus, this study aimed to investigate the effect of omega-3 in combination with UDCA on diet-induced NASH in mice and to compare the results with those of omega-3 alone to assess the possibility of this combination as a treatment for NASH.

Materials and methods

Ethics

The animal experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee of Gangnam Severance Hospital, Yonsei University College of Medicine (Permit Number: 0011). The study was carried out in accordance with the recommendations and restrictions of the Institutional Animal Care and Use Committee.

Animals

Male C57BL/6 mice (12 weeks of age) were obtained from the Central Lab Animal (Seoul, Korea) and housed under a 12-h light–dark cycle. Mice were fed with a diet that consisted of 15% anhydrous milk fat, 1.0% cholesterol and 50% sucrose (a high-fat diet (HFD)) that was purchased from Picolab (Bethlehem, PA, USA). These mice were fed with the HFD for 24 weeks, at which point they were divided into three groups and fed with HFD alone (HFD), HFD and 70 mg kg−1 omega-3 (Pronova Biocare, Sandefjord, Norway; HFD/Omega-3) or HFD and omega-3 in combination with 20 mg kg−1 UDCA (Daewoong Pharm., Seoul, Korea; HFD/Omega-3+UDCA) for another 24 weeks. Omega-3 and UDCA were diluted in 0.75% Tween-80 and administered orally through sonde for 24 weeks. After 48 weeks of HFD feeding with or without administering experimental agents, the mice were sacrificed and their livers were harvested. A sample of fresh liver tissue was fixed in 10% buffered formalin, and the remaining tissue was snap-frozen in liquid nitrogen and stored at −80 °C. Blood samples were collected by cardiac puncture after the mice were anesthetized and were stored at −80 °C.

Histological evaluation

Sections of liver tissue specimens, fixed in 10% formalin and embedded in paraffin wax, were stained with H&E and Sirius Red for histological evaluation. A blinded pathologist evaluated the slides for inflammation as described in previous studies with minor modifications.21, 22, 23 Inflammation was histologically quantified by counting inflammatory foci in 20 consecutive high-power fields ( × 40 objective; average histological grade, grade 0: no foci, grade 1: <2 foci per high-power field, grade 2: ⩾2 foci per high-power field). Liver fibrosis was evaluated by calculating the fibrosis ratio using an image analysis system as described in a previous study with some modifications.24 Briefly, images of Sirius Red-stained sections were captured under × 10 magnification. In the stained section, red indicated fibrosis and gray indicated parenchyma. After interactive thresholding, the image was converted into a binary image. The two-dimensional patterns were measured by direct pixel counting on the binary images. The total area was the sum of the areas of the microscopic fields that included parenchyma and fibrosis. For each slide, the area of fibrosis was evaluated in 20 consecutive high-power fields and was then averaged.

RNA extraction and gene expression analysis by quantitative real-time PCR

Total RNA was extracted from frozen whole liver using Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. RNA samples were quantified by spectrophotometry. RNA integrity was assessed using agarose gel electrophoresis and ethidium bromide staining. The RNA samples were then diluted in RNase-free water and stored at −70 °C until use. Five micrograms of RNA was reverse transcribed using an RNA PCR kit (version 1.2; Takara Bio Inc., Japan) according to the manufacturer’s recommendations. Oligonucleotide primers and TaqMan probes for collagen α1(I), transforming growth factor (TGF) β, tumor necrosis factor (TNF) α, sterol regulatory binding protein (SREBP)-1c, fatty acid synthase, stearoyl-carnitine palmitoyl-coenzyme A desaturase, carnitine palmitoyl-coenzyme A transferase and inducible nitric oxide synthase (iNOS) were used with 18S as an internal control. The probes were obtained from Applied Biosystems (Perkin-Elmer/PE Applied Biosystems, Foster City, CA, USA) and purchased in a ready-for-use form in Assays-on-Demand Gene Expression Products. The TaqMan probes were labeled at the 5′ end with the reporter dye fluorescein and at the 3′ end with minor groove binder nonfluorescent quencher. Quantitative PCR was performed in triplicate for each sample on a Step One Plus Real Time System (Applied Biosystems). Each 20-μl reaction contained 10 μl of TaqMan Fast Universal Master Mix (Applied Biosystems, Darmstadt, Germany), 1 μl of Gene Expression Mix (Perkin-Elmer/PE Applied Biosystem) and 2 μl of cDNA diluted in 7 μl RNase-free water. The thermal cycler conditions were 20 s at 95 °C, 40 cycles of 5 s at 95 °C and then 20 s at 60°. Fold changes in target gene messengerRNA (mRNA) relative to the endogenous 18S control were calculated as described in previous studies.25

Measurement of TG content

Hepatic TG content was quantified using a commercial kit (ab65336, Abcam, Cambridge, MA, USA) according to the manufacturer’s recommendations. Briefly, lipid extracts were prepared by the homogenization of 50 mg of frozen tissue in a solution of 5% NP-40. The homogenized tissue was heated to 80–100 °C in a water bath for 2–5 min, then cooled down to room temperature. The samples were purified and diluted (1:10) in distilled H2O to be measured at OD570 mm according to the manufacturer’s instructions.

Statistical analysis

All results are shown as the mean±(s.e.m.). Data were analyzed by nonparametric analysis (Kruskal–Wallis or Mann–Whitney test) or one-way analysis of variance with Tukey’s post-hoc analysis. P<0.05 was considered statistically significant. All calculations were performed with SPSS version 15.0 software for Windows (SPSS Inc., Chicago, IL, USA).

Results

Characteristics of mice fed an HFD with or without experimental materials

The changes in body weight during the 48 weeks of the experiment are shown in Figure 1 and serum biochemical findings are summarized in Table 1. Administering omega-3 did not induce significant changes in body weight independent of UDCA combinations. However, serum TG levels decreased significantly as a result of administering omega-3 alone (P=0.012) or the concomitant administration of omega-3 and UDCA (P<0.011). There was no difference in TG levels between the omega-3 group and omega-3 with UDCA group (P=0.995, when compared with the omega-3 only group). Furthermore, serum alanine aminotransferase decreased when mice were fed omega-3 alone (P<0.001) or in combination with UDCA (P<0.001). There was no difference in alanine aminotransferase levels between the omega-3 group and omega-3 with UDCA group (P=0.999, when compared with the omega-3 only group).

Figure 1
figure 1

Changes in bodyweight during 48 weeks of the experiment. All the mice were fed a high fat diet (HFD) for 24 weeks and were then divided into three groups. One group (n=8) was fed an HFD only (HFD), another group (n=8) was fed an HFD and administered omega-3 (HFD/Omega-3) and the other group (n=8) was fed an HFD and administered omega-3 in combination with UDCA (HFD/Omega-3+UDCA) for another 24 weeks.

Full size image
Table 1 Characterization of mice fed an HFD and administered omega-3 with or without UDCA
Full size table

Omega-3-attenuated NASH-associated liver fibrosis and adding UDCA further enhanced its anti-fibrotic effect

Histological assessment of liver fibrosis was performed using an image analysis system after Sirius Red staining. Administering omega-3 at a daily dose of 70 mg kg−1 with an HFD-improved histologically evaluated liver fibrosis (P<0.001; Figures 2a and b). Administration of UDCA at a daily dose of 20 mg kg−1 in combination with omega-3 further attenuated HFD-induced liver fibrosis (P=0.026, compared with the omega-3 only group; Figure 2b).

Figure 2
figure 2

Histological assessment of fibrosis and inflammation in diet-induced NASH. Histological analysis of the liver to evaluate inflammation and fibrosis was performed after H&E and Sirius Red staining. Mice were fed with a high fat diet (HFD) for 24 weeks and were then divided into three groups. One group (n=8) was fed an HFD only (HFD), another group (n=8) was fed an HFD and administered omega-3 (HFD/Omega-3) and the other group (n=8) was fed an HFD and administered omega-3 in combination with UDCA (HFD/Omega-3+UDCA) for 24 weeks. (a) Representative histological evaluation using H&E ( × 100) and Sirius Red ( × 100). (b) Percent region of fibrosis. Liver fibrosis was evaluated by calculating the fibrosis ratio using an image analysis system. The total area was the sum of the area of microscopic fields including parenchyma and fibrosis. (c) Inflammation was histologically quantified by counting inflammatory foci in 20 consecutive high-power fields ( × 40 objective). The averaged histological grades are expressed as the following; grade 0: no foci, grade 1: <2 foci per high-power field, grade 2: ⩾2 foci per high-power field. *P<0.05 when compared with that of the HFD group. **P<0.05 when compared with that of HFD/Omega-3 group.

Full size image

The effect of omega-3 and UDCA on NASH-associated liver fibrosis was also investigated by quantitative analyses of collagen α1(I) and TGFβ mRNA expression. Omega-3 alleviated HFD-induced collagen deposition (P=0.002) and administering UDCA in combination with omega-3 further decreased Colα1(I) expression (P=0.021, compared with the omega-3 only group; Figure 3a). In addition, administering omega-3 as a supplement to the HFD diminished the mRNA expression of the pro-fibrotic gene TGFβ (P=0.012), and administering UDCA in combination with omega-3 further accentuated this effect (P=0.029, compared with the omega-3 only group; Figure 3a).

Figure 3
figure 3

Changes in gene expression related to fibrosis, inflammation, steatosis and oxidative stress. Mice were fed a high fat diet (HFD) for 24 weeks and were then divided into three groups. One group (n=8) was fed an HFD only (HFD), another group (n=8) was fed an HFD and administered omega-3 (HFD/Omega-3) and the other group (n=8) was fed an HFD and administered omega-3 in combination with UDCA (HFD/Omega-3+UDCA) for 24 weeks. (a) Expression of fibrosis related genes in the liver. Quantitative analysis of liver Colα(I) and TGFβ mRNA, markers of fibrosis, decreased when omega-3 was added to HFD. The HFD/Omega-3+UDCA group exhibited decreased Colα1(I) and TGFβ expression when compared with that of the HFD/Omega-3 group. (b) Expression of inflammation-related TNFα in the liver. Quantitative analysis of liver TNFα mRNA was decreased in the HFD/Omega-3 and HFD/Omega-3+UDCA groups when compared with the HFD group. The HFD/Omega-3+UDCA group had decreased TNFα expression when compared with that of the HFD/Omega-3 group. (c) Expression of lipogenic genes. mRNA expression of SREBP-1c, the lipogenic gene regulator, was decreased in the HFD/Omega-3 and HFD/Omega-3+UDCA groups when compared with the HFD group. The HFD/Omega-3+UDCA group had decreased SREBP-1c expression when compared with that of the HFD/Omega-3 group. mRNA expression of fatty acid synthase and stearoyl-carnitine palmitoyl-coenzyme A desaturase, genes involved in de novo lipogenesis, were significantly decreased when omega-3 was administered. However, no additive effect was found with UDCA. CPA, a fatty acid oxidation related gene, was diminished by administering omega-3 with an HFD, but no further reduction was noticed by administering UDCA in combination with omega-3. (d) Expression of an oxidative stress-related gene. Decreased liver iNOS mRNA expression was observed with omega-3 as a supplement to HFD and adding UDCA further augmented this effect. *P<0.05 when compared with that of the HFD group. **P<0.05 when compared with that of HFD/Omega-3 group.

Full size image

Omega-3 and UDCA-attenuated NASH-associated liver inflammation

Histological analysis of liver inflammation was carried out after H&E staining. Administering UDCA in combination with omega-3 ameliorated histologically assessed liver inflammation when compared with that of mice fed an HFD only (P=0.036; Figures 2a and c). In addition, when liver expression of TNFα, an inflammatory cytokine, was assessed, administering omega-3 attenuated the TNFα mRNA increase (P<0.001) and adding UDCA further decreased TNFα expression (P=0.008, compared with the omega-3 only group; Figure 3b).

Omega-3 and UDCA-attenuated hepatic TG accumulation

The effect of omega-3 and UDCA on HFD-induced hepatic TG accumulation was evaluated by measuring the tissue TG content. Administering omega-3 with the HFD resulted in decreased hepatic TG content compared with that of mice fed the HFD only (P=0.029). In addition, the combination of UDCA and omega-3 further attenuated hepatic TG accumulation (P=0.021; Figure 4).

Figure 4
figure 4

Hepatic triglyceride content. Hepatic triglyceride content was measured in frozen liver tissue using a commercial kit. The mice were fed a high fat diet (HFD) for 24 weeks and were then divided into three groups. One group (n=8) was fed an HFD only (HFD), another group (n=8) was fed an HFD and administered omega-3 (HFD/Omega-3) and the other group (n=8) was fed an HFD and administered omega-3 in combination with UDCA (HFD/Omega-3+UDCA) for another 24 weeks. *P<0.05 when compared with that of the HFD group. **P<0.05 when compared with that of the HFD/Omega-3 group.

Full size image

Omega-3 suppressed the expression of lipogenesis-associated genes and adding UDCA augmented this effect

The mRNA expression of SREBP-1c, which is known to regulate lipogenic enzymes,26 was decreased after omega-3 administration in HFD-induced NASH (P=0.014). Administering UDCA in combination with omega-3 further diminished SREBP-1c expression (P=0.001, when compared with the omega-3 only group; Figure 3c). The mRNA expression of fatty acid synthase, a gene involved in de novo lipogenesis, decreased when omega-3 was administered (P=0.035), but UDCA did not show an additive effect (P=0.343, when compared with the omega-3 only group). The mRNA expression of stearoyl-carnitine palmitoyl-coenzyme A desaturase, another gene involved in de novo lipogenesis, was also reduced when omega-3 was administered (P=0.022), but there was no significant difference when UDCA was administered in combination with omega-3 (P=0.083 when compared with the omega-3 only group). The mRNA expression of carnitine palmitoyl-coenzyme A transferase, a mitochondrial β-oxidation enzyme, decreased with omega-3 supplement ingestion (P=0.045), but adding UDCA did not produce a further reduction in carnitine palmitoyl-coenzyme A transferase mRNA expression (P=0.686, when compared with the omega-3 only group).

The effect of UDCA on NASH was associated with decreased iNOS mRNA expression

The upregulation of iNOS is known to be associated with liver inflammation and tissue damage,27, 28 and UDCA has been reported to reduce iNOS in several inflammatory diseases.29, 30 We evaluated whether the beneficial effect of UDCA on NASH in our study was associated with decreased iNOS mRNA expression. Administration of omega-3 with the HFD resulted in diminished iNOS mRNA (P=0.012), and adding UDCA in combination with omega-3 further suppressed iNOS mRNA expression (P=0.029, when compared with the omega-3 only group; Figure 3d).

Discussion

Although several previous studies have reported that omega-3 attenuates diet-induced NASH,31, 32, 33 to our knowledge, this is the first study to report an additive effect of omega-3 and UDCA on NASH. In accordance with other studies, our study also demonstrated that omega-3 alone could alleviate liver fibrosis identified by decreased collagen α1(I) mRNA expression. However, administering UDCA in combination with omega-3 further reduced fibrosis at the mRNA level and also further attenuated histologically assessed liver fibrosis when compared with the group to which omega-3 was administered alone. It has been repeatedly suggested that liver fibrosis is one of the most important factors in forecasting the progression of NASH,34, 35, 36 and an additive effect of omega-3 and UDCA in diminishing NASH-associated fibrosis makes this combination a possible candidate for the treatment of NASH.

TGFβ has been identified as a key regulator of liver fibrosis that exerts its profibrogenic effect through hepatic stellate cells.37 The role of TGFβ signaling in the progression of NASH-associated fibrosis has also been reported in several studies, making it an attractive target for treatment.38, 39 Our study demonstrated that omega-3 and UDCA had an additive effect in suppressing TGFβ mRNA expression, although TGFβ is not suggested as a target for this combination therapy, according to our results.

Even with the promising results of either omega-3 or UDCA on experimentally induced NASH, neither omega-3 nor UDCA has succeeded in showing a significant therapeutic effect when tested in NASH patients. Randomized controlled clinical trials investigating the effect of either omega-3 or UDCA on NASH in patients have failed to show significant improvement when compared with a placebo.17, 18 However, a systemic review has suggested that negative results on the beneficial role of these agents might be attributed to improperly constructed study designs including suboptimal drug dosage and study duration.19 Meanwhile, several studies testing UDCA-based combination regimens, such as UDCA with vitamin E, resulted in biochemical or histological improvements in NASH patients.40, 41 However, as described earlier, there has been a report showing the adverse effects of long-term vitamin E consumption,20 and safety should be the utmost concern in medical treatments of NASH. No serious side-effects have been recorded with UDCA treatment, and a human study using 25 mg kg−1 of UDCA, which was higher than the dose used in our study, has been carried out without reporting any serious adverse events.42 As for omega-3, the currently recommended maximal dose is 4000 mg per day, which is equivalent to the dose used in this study.

In addition to liver fibrosis, inflammation is also known to have a major role in the development and progression of NASH.23, 43 Although administering omega-3 alone failed to show significant histologic changes, administering UDCA in combination with omega-3 resulted in histologically improved inflammation in diet-induced NASH, suggesting an additive effect. A previous study also demonstrated decreased expression of the proinflammatory cytokine TNFα after omega-3 supplementation,44 which is in accordance with our study results.

TG accumulation in the liver is known to be controlled by SREBP-1c, a transcriptional factor that regulates lipogenic enzymes.45, 46 Several studies have reported that omega-3 supplementation downregulated SREBP-1c gene expression,6, 7, 8, 9 and a recent study demonstrated that UDCA mediated lipogenesis by repressing SREBP-1c expression.47 Our study suggested an additive effect of omega-3 and UDCA for suppressing SREBP-1c, although this study could not reveal an additive effect of UDCA for regulating genes related to de novo lipogenesis or fatty acid oxidation.

Although our study does not delineate the mechanistic role of omega-3 and UDCA-combined treatment for improving NASH, suppressed iNOS expression suggests that reduced liver damage from oxidative stress may be a possible explanation. Increased iNOS production has been demonstrated in fructose-induced hepatic steatosis,48 and suppressed iNOS production has been associated with the use of anti-inflammatory drugs.49 Studies investigating more detailed mechanisms are needed in the future.

In conclusion, our study demonstrated an additive effect of omega-3 and UDCA for alleviating fibrosis and inflammation in diet-induced NASH. Our study also demonstrated the additive effect of omega-3 and UDCA for suppressing the lipogenic gene SREBP-1c and the oxidative stress related gene iNOS. These results suggest that the combination of omega-3 and UDCA may be a possible treatment for NASH, although validation using different NASH models and human studies are necessary.