E ﬀ ects of Ethanol Feeding in Early-Stage NAFLD Mice Induced by Western Diet

: Background : The prevalence of metabolic liver diseases is increasing and approved pharmacological treatments are still missing. Many animal models of nonalcoholic fatty liver disease (NAFLD) show a full spectrum of ﬁbrosis, inﬂammation and steatosis, which does not reﬂect the human situation since only up to one third of the patients develop ﬁbrosis and nonalcoholic steatohepatitis (NASH). Methods : Seven week old C57Bl / J mice were treated with ethanol, Western diet (WD) or both. The animals’ liver phenotypes were determined through histology, immunohistochemistry, Western blotting, hepatic triglyceride content and gene expression levels. In a human cohort of 80 patients stratiﬁed by current alcohol misuse and body mass index, liver histology and gene expression analysis were performed. Results : WD diet and ethanol-treated animals showed severe steatosis, with high hepatic triglyceride content and upregulation of fatty acid synthesis . Mild ﬁbrosis was revealed using Sirius-red stains and gene expression levels of collagen. Inﬂammation was detected using histology, immunohistochemistry and upregulation of proinﬂammatory genes. The human cohort of obese drinkers showed similar upregulation in genes related to steatosis, ﬁbrosis and inﬂammation. Conclusions : We provide a novel murine model for early-stage fatty liver disease suitable for drug testing and investigation of pathophysiology.


Introduction
Despite public health efforts, the prevalence of chronic liver diseases (CLD) is still increasing, causing an important health care burden, especially in Europe and the United States [1,2]. Importantly, the distribution of the etiology of CLD is changing: vaccines for hepatitis B virus are widely available

Human Liver Histology
We performed a percutaneous liver biopsy with a 17-G Menghini suction needle, fixed it with formalin, embedded it in paraffin and stained it with Sirius-red, as reported previously [8]. We only included biopsies of adequate quality, defined as either containing regeneration noduli or with a length of >10 mm and including >5 portal tracts. One pathologist (SD) staged the biopsies for fibrosis according to Kleiner and also scored, according to the NAFLD activity score, for steatosis (0-3), lobular inflammation (0-3) and ballooning (0-2) [10]. Human histology data are provided in Table 1.

Animal Experimentation
Twelve-week old male wild-type (C57Bl6/J) mice were purchased from (Charles River Laboratories Research Model and Services Germany GmbH, Sulzfeld, Germany). In total, 38 mice in two different badges were used. The experiments were performed twice. Animals were provided a Western diet (WD) enriched in fat and cholesterol, or normal diet for seven weeks ( Figure 1A). Diets and water were provided ad libitum (details on provided diets can be found in Supplementary Table S1). WD and normal diet-fed mice were randomized into subgroups of five individuals, whether to add 16% of ethanol to the drinking water or not. Mice were acclimatized to ethanol by stepwise increases of ethanol concentration during three weeks, providing 4, 8 and 16%, respectively. Mice were sacrificed after seven weeks of treatment and the cadavers' bodies and livers were weighted. Age-and injury-matched controls were used in all experiments. Every experiment was performed in accordance with the German Animal-Protection Law and the Guidelines of the Animal Care facility at our University (Haus für experimentelle Therapie, Bonn, Germany), and approved by the North Rhine-Westphalian State Agency for Nature, Environment, and Consumer Protection (LANUV, Germany; File Reference LANUV NRW, 84-02.04.2015.A491).

Human Liver Histology
We performed a percutaneous liver biopsy with a 17-G Menghini suction needle, fixed it with formalin, embedded it in paraffin and stained it with Sirius-red, as reported previously [8]. We only included biopsies of adequate quality, defined as either containing regeneration noduli or with a length of >10 mm and including >5 portal tracts. One pathologist (SD) staged the biopsies for fibrosis according to Kleiner and also scored, according to the NAFLD activity score, for steatosis (0-3), lobular inflammation (0-3) and ballooning (0-2) [10]. Human histology data are provided in Table 1.

Animal Experimentation
Twelve-week old male wild-type (C57Bl6/J) mice were purchased from (Charles River Laboratories Research Model and Services Germany GmbH, Sulzfeld, Germany). In total, 38 mice in two different badges were used. The experiments were performed twice. Animals were provided a Western diet (WD) enriched in fat and cholesterol, or normal diet for seven weeks ( Figure 1A). Diets and water were provided ad libitum (details on provided diets can be found in Suppl. Table S1). WD and normal diet-fed mice were randomized into subgroups of five individuals, whether to add 16% of ethanol to the drinking water or not. Mice were acclimatized to ethanol by stepwise increases of ethanol concentration during three weeks, providing 4, 8 and 16%, respectively. Mice were sacrificed after seven weeks of treatment and the cadavers' bodies and livers were weighted. Age-and injurymatched controls were used in all experiments. Every experiment was performed in accordance with the German Animal-Protection Law and the Guidelines of the Animal Care facility at our University (Haus für experimentelle Therapie, Bonn, Germany), and approved by the North Rhine-Westphalian State Agency for Nature, Environment, and Consumer Protection (LANUV, Germany; File Reference LANUV NRW, 84-02.04.2015.A491).

Real-Time PCR
Quantitative polymerase chain reaction (qPCR) in animal tissue was performed as described previously [7]. In brief, total RNA was isolated with ReliaPrep™ RNA Miniprep Systems (Promega, Madison, WI, USA) from shock-frozen liver samples following the ReliaPrep™ RNA Cell Miniprep System protocol. cDNA was synthesized by ImProm-II Reverse Transcription System (Promega, Madison, WI, USA). Amplification by qPCR was performed on the 7300 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using commercial TaqMan ® gene expression assays by Applied Biosystems according to the manufacturer's protocol. Each qPCR analysis included duplicate wells and appropriate control reactions. The expression of each gene of interest was calculated by the delta-delta Ct method according to Livak KJ and Schmittgen TD [11]. Gene amplification was normalized against 18s rRNA expression in each sample and expression gene levels are shown as relative expression units by comparing to the control group. The complete list of gene expression assays is shown in Supplementary Table S2. Total RNA from human tissue was isolated from liver biopsies embedded in Tissue-Tek ® optimal cutting temperature (O.C.T.) compound (Sakura, Alphen aan den Rijn, The Netherlands) using peqGold TriFast (peqLab, VWR, Radnor, PA, USA) following the manufacturer's instructions. RT-PCR reactions were performed with PrimeScript RT Reagent Kit with gDNA Eraser (Takara, Shimogyo-Ku, Japan). Gene-specific primers were designed for each gene using Primer3 software (Supplementary Table S3). Gene relative abundance was assessed by RT-PCR using KAPA SYBR Fast Master Mix (Kapa Biosystems, Basel, Switzerland) on a Roche (LC480) system. Again, the relative expression was calculated by the delta Ct (2 −DCt ) method [11]. All data were normalized to the 18S endogenous reference gene.

Hepatic Triglyceride Content
Hepatic triglyceride content was analyzed as described previously [12]. The amount was measured photometrically from homogenized snap-frozen liver samples using TG liquicolor mono kit (Human Diagnostics, Wiesbaden, Germany) according to the manufacturer's instructions, as published elsewhere [13].

Western Blotting
Protein levels were analyzed by Western blot as previously described [7]. Snap-frozen liver tissues were homogenized and diluted. Their protein amounts were measured with the DC assay kit (Bio-Rad, Munich, Germany). Forty micrograms of protein specimen were subjected to SDS-PAGE under reducing conditions (10% gels) and proteins were blotted on nitrocellulose membranes. The membranes were blocked and incubated with primary antibodies against either srebp-1c (ab28481, Abcam, Cambridge, United Kingdom) or scd-1 (2794, Cell Signalling Technology, Danvers, MA, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as an endogenous control (sc-47724, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Finally, membranes were incubated with the corresponding secondary antibody, and blots were developed using enhanced chemiluminescence. Protein quantification was performed by ImageJ (V.1.51j8, NIH, Bethesda, MD, USA) and the results were corrected for GAPDH levels.

Histological Staining Methodology and Quantification
Oil red O staining. For detection of fat accumulation in the liver, 10 µm thick sections of liver samples embedded in O.C.T. compound were cut using a cryostat. Sections were dehydrated overnight, washed with 60% isopropyl alcohol and stained with Oil red O (3%) followed by counterstaining with hematoxylin, as described previously [13]. To detect collagen fibers, paraffin-embedded sections (2-3 µm) were treated with 0.1% Sirius-red F3B in saturated picric acid (Chroma, Münster, Germany), as described previously [13]. Paraffin-embedded sections (2-3 µm) were also stained with hematoxylin-eosin (H&E) for detection of cellular ballooning, as described previously. Immunohistochemical staining for F4/80 were performed in paraffin-embedded sections (2-3 µm). The sections were incubated with a rat-anti-F4/80 (clone BM8; BMA Dianova, Hamburg, Germany). Thereafter, biotinylated rabbit-antirat (Biozol, Eching, Germany) secondary antibodies were used respectively. Finally, sections were counterstained with hematoxylin. Histological stains were captured with a Nikon Digital Sight DS-Vi1 (Chiyoda, Tokyo, Japan). Oil red O and Sirius red staining and F4/80 immunohistochemistry were quantified via open source ImageJ software (V.1.51j8, NIH, Bethesda, MD, USA) using macros for automatized quantification and color detection. Macros are provided in Supplementary Materials 1-3 [14]. Quantification was expressed as percentage of positively stained area. A minimum of ten high power fields were captured for analysis.

Measurement of Serum Metabolites and Enzymes
Blood was collected immediately after animal sacrifice, centrifuged and serum was stored at −80 • C until analysis. Serum levels of alanine transaminase (ALT) were measured with the Reflotron Plus blood analysis system (Roche Diagnostics, Mannheim, Germany). Serum levels of triglycerides were analyzed with the TG liquicolor mono kit (Human Diagnostics, Wiesbaden, Germany), according to the manufacturer's protocol.

Statistical Analysis
Statistical analyses were performed using Prism V.5.0 (GraphPad, San Diego, CA, USA) and SPSS V26 (IBM, Armonk, New York, NY, USA). Data were expressed as mean ± standard error of the mean, and comparisons between two groups were realized by nonparametric t-tests based on Mann-Whitney U calculation, unless otherwise specified. p-Values below 0.05 (*) were considered statistically significant.

WD and Ethanol Feeding Led to Increase of Body and Liver Weight
Twelve week-old mice were treated for seven weeks either with ethanol or Western diet or a combination of both. After seven weeks of treatment, ethanol-fed mice did not change body weight ( Table 2). WD feeding alone led to an increase of body weight, but ethanol as a cofactor decreased body weight compared to WD back to baseline ( Table 2). Neither ethanol feeding nor WD feeding altered liver weight, but the combination of ethanol and WD led to a significant increase of liver weight ( Table 2). Liver-to-body weight ratio was calculated, demonstrating that WD feeding led to a decreased liver-to-body weight ratio but, with ethanol as a cofactor, liver to-body weight ratio was significantly increased ( Figure 1B). Briefly, ethanol did decrease liver and body weight but increased liver weight in WD-fed animals.

WD and Ethanol Feeding Led to An. Increase of Serum ALT, But No Change in Triglycerides
As serological disease markers, we analyzed serum levels of transaminases. For ALT, no change was observed if ethanol was fed alone. WD treatment led to a decrease of serum ALT in our animals. If ethanol and WD were administered concomitantly, we observed a significant increase in serum ALT, indicating hepatic damage in our mice ( Figure 1C). For serum triglycerides, no differences were revealed in the serum, indicating that metabolic dysfunction was not induced within the experimental period, although an increasing trend can be identified in the combination group ( Figure 1D).

WD and Ethanol Feeding Led to Significant Liver Steatosis Comparable to Humans
We then assessed steatosis through histological staining and biochemical measurement of hepatic triglycerides. Oil red O staining revealed that both ethanol and WD treatment alone only induced mild hepatic steatosis, but the combination of WD and ethanol led to severe macrovesicular steatosis ( Figure 2A). This finding was confirmed through quantification of the positively stained area ( Figure 2B). Measurement of triglycerides in mice livers revealed that the highest triglyceride levels were found in WD with ethanol-treated mice ( Figure 2C). Treatment with either ethanol or WD alone led to no statistically significant increase of hepatic triglycerides ( Figure 2C). Gene expression analysis of sterol regulatory element-binding protein 1 (srebp-1c) showed upregulation in the WD-fed group, but not if ethanol was fed alone. The highest expression levels were observed in the WD and ethanol group ( Figure 2D). Interestingly, mRNA expression levels of stearoyl-CoA desaturase-1 (scd-1) showed an upregulation in all animals with the highest levels in animals fed with either ethanol or both ethanol and WD, indicating different mechanisms of steatosis development ( Figure 2E). Gene expression analysis of fatty acid synthase (fasn) revealed a downregulation in ethanol-fed animals and no alteration in WD-fed animals ( Figure 2F). The highest and most significantly increased gene expression levels of fasn were measured in WD and ethanol-fed mice ( Figure 2F). Severe hepatic steatosis was induced in WD and ethanol-treated mice after seven weeks, whereas WD feeding and ethanol treatment only induced mild steatosis. This pattern was also observed in the human cohort. The group of lean drinker and obese abstinent patients showed less elevation of hepatic fasn levels, while patients in the obese drinker group did express significantly elevated levels ( Figure 2G). Finally, we confirmed our findings from the gene expression analysis in mice through protein level analyses. In our Western blots of srebp-1c and scd-1, we confirmed the trends identified in the gene expression analysis ( Figure 2H). We saw an upregulation for ethanol and ethanol with WD-fed mice. Highest protein levels were observed in the WD group ( Figure 2I). For scd-1, we saw higher levels for ethanol-fed mice. Again, highest protein levels were identified in the ethanol + WD group ( Figure 2J).

WD and Ethanol Feeding Promoted Fibrosis Similar to Human Disease
Sirius-red staining was used to investigate fibrosis development. Ethanol alone did not induce liver fibrosis, but WD with ethanol administration induced the strongest increase in the amount of perisinusoidal collagen in the stains ( Figure 3A,B). We then performed RT-PCR for collagen type I alpha 1 (col1a1). We observed an upregulation in both the WD and the ethanol-fed group, but the highest values were again observed in the group of ethanol and WD-fed animals ( Figure 3C). In our human cohorts, col1a1 was increased in the drinker groups, with the clearest difference in the obese drinker group ( Figure 3D). Hepatic gene expression of actin alpha 2 (acta2) showed a not significant, increasing trend in the WD and ethanol group ( Figure 3E), which was confirmed in the human cohort ( Figure 3F  alpha 1 (col1a1). We observed an upregulation in both the WD and the ethanol-fed group, but the highest values were again observed in the group of ethanol and WD-fed animals ( Figure 3C). In our human cohorts, col1a1 was increased in the drinker groups, with the clearest difference in the obese drinker group ( Figure 3D). Hepatic gene expression of actin alpha 2 (acta2) showed a not significant, increasing trend in the WD and ethanol group ( Figure 3E), which was confirmed in the human cohort ( Figure 3F).

WD and Ethanol Feeding Increased Liver Inflammation
To fully assess the animals' liver phenotypes, we then checked for liver inflammation in our models. Firstly, F4/80 immunohistochemistry was performed in murine livers. Higher abundance of positively stained cells were observed in treated animals compared to controls ( Figure 4A). This was confirmed through the staining quantification, which showed a significant upregulation in all treated animals, with the highest values in mice treated with the combination of ethanol and WD ( Figure 4B). We investigated gene expression levels of cytokines in our models. When ethanol or WD were provided alone, interleukin 1 beta (il1b) remained unchanged, with a trend towards increase in WD and ethanol cofed mice ( Figure 4C). While WD and ethanol feeding alone did not lead to a change of expression of C-C chemokine receptor type 2 (ccr2), its combination led to a significant increase ( Figure 4D). Gene expression of CC-Chemokin-Ligand-2 (ccl2) was upregulated in ethanol and WDfed mice, and upregulation was strongly increased in mice treated with both WD and ethanol. ( Figure  4E). Briefly, WD was not able to induce upregulation of hepatic inflammatory marker mRNA levels,

WD and Ethanol Feeding Increased Liver Inflammation
To fully assess the animals' liver phenotypes, we then checked for liver inflammation in our models. Firstly, F4/80 immunohistochemistry was performed in murine livers. Higher abundance of positively stained cells were observed in treated animals compared to controls ( Figure 4A). This was confirmed through the staining quantification, which showed a significant upregulation in all treated animals, with the highest values in mice treated with the combination of ethanol and WD ( Figure 4B). We investigated gene expression levels of cytokines in our models. When ethanol or WD were provided alone, interleukin 1 beta (il1b) remained unchanged, with a trend towards increase in WD and ethanol cofed mice ( Figure 4C). While WD and ethanol feeding alone did not lead to a change of expression of C-C chemokine receptor type 2 (ccr2), its combination led to a significant increase ( Figure 4D). Gene expression of CC-Chemokin-Ligand-2 (ccl2) was upregulated in ethanol and WD-fed mice, and upregulation was strongly increased in mice treated with both WD and ethanol. ( Figure 4E). Briefly, WD was not able to induce upregulation of hepatic inflammatory marker mRNA levels, but ethanol as a cofactor did increase its expression ( Figure 4C-E). In our human cohort, il1b expression levels were increased in ethanol-related groups lean drinker and obese drinker, but not in the obese abstinent group ( Figure 4F). For ccr2, lean abstinent and obese abstinent groups showed higher expressions compared to the control group, but the highest values were observed in the obese drinker group ( Figure 4G). For liver patients presenting obesity and alcohol misuse in their medical history, inflammatory genes were upregulated compared to the other group of patients ( Figure 4F,G). expression levels were increased in ethanol-related groups lean drinker and obese drinker, but not in the obese abstinent group ( Figure 4F). For ccr2, lean abstinent and obese abstinent groups showed higher expressions compared to the control group, but the highest values were observed in the obese drinker group ( Figure 4G). For liver patients presenting obesity and alcohol misuse in their medical history, inflammatory genes were upregulated compared to the other group of patients ( Figure 4F,G).

Discussion
In this study, we analyzed effects of ethanol in WD-fed mice and characterized a new murine early-stage fatty liver disease model that developed histological and gene expression features of early metabolic liver disease via simultaneous ethanol and WD feeding for only seven weeks. Mice showed hepatomegaly, severe steatosis and mild inflammation and fibrosis. Our model further showed a similar gene expression profile in humans affected by early-stage liver disease. These features potentially render the model suitable to investigate pathogenic mechanisms and for testing biomarkers and novel therapeutic strategies.
Recently, awareness has emerged that the steatosis component of metabolic liver disease is mostly driven by both ethanol consumption and diet [15,16]. This group was probably underrepresented in previous models because of the distinct attribution into models for alcoholic and nonalcoholic fatty liver disease [17]. In daily clinical practice, alcohol consumption drives disease progression in obese patients with fatty liver disease [18]. This fact is often neglected in animal models exclusively fed with WD or similar formulas. Large cohort analyses revealed that alcohol consumption, even within the acceptable limit for the diagnosis of NAFLD, negatively affects the severity of chronic liver disease in patients with metabolic syndrome [19]. Overall, data on alcohol consumption mostly rely on anamnestic data, but are probably underestimated in cohort analyses [20]. We therefore believe that our model mirrors a realistic setting, since ethanol was coadministered within the experimental period. Moreover, the use of wild-type mice in our model contributed to a natural injury development since genetic modifications were not applied.
Chronic ethanol consumption is likely to drive hepatic steatosis. This was demonstrated in several previous studies. It has been recently reported that ethanol drives liver steatosis in mice

Discussion
In this study, we analyzed effects of ethanol in WD-fed mice and characterized a new murine early-stage fatty liver disease model that developed histological and gene expression features of early metabolic liver disease via simultaneous ethanol and WD feeding for only seven weeks. Mice showed hepatomegaly, severe steatosis and mild inflammation and fibrosis. Our model further showed a similar gene expression profile in humans affected by early-stage liver disease. These features potentially render the model suitable to investigate pathogenic mechanisms and for testing biomarkers and novel therapeutic strategies.
Recently, awareness has emerged that the steatosis component of metabolic liver disease is mostly driven by both ethanol consumption and diet [15,16]. This group was probably underrepresented in previous models because of the distinct attribution into models for alcoholic and nonalcoholic fatty liver disease [17]. In daily clinical practice, alcohol consumption drives disease progression in obese patients with fatty liver disease [18]. This fact is often neglected in animal models exclusively fed with WD or similar formulas. Large cohort analyses revealed that alcohol consumption, even within the acceptable limit for the diagnosis of NAFLD, negatively affects the severity of chronic liver disease in patients with metabolic syndrome [19]. Overall, data on alcohol consumption mostly rely on anamnestic data, but are probably underestimated in cohort analyses [20]. We therefore believe that our model mirrors a realistic setting, since ethanol was coadministered within the experimental period. Moreover, the use of wild-type mice in our model contributed to a natural injury development since genetic modifications were not applied.
Chronic ethanol consumption is likely to drive hepatic steatosis. This was demonstrated in several previous studies. It has been recently reported that ethanol drives liver steatosis in mice suffering from hepatitis B virus infection [21]. The role of ethanol within the pathophysiology of NAFLD is currently poorly understood and seems to induce different dose-dependent effects. Bucher et al. recently showed that low-dose ethanol-feeding (2%) in high fat diet-treated mice may attenuate NAFLD through downregulation of proinflammatory and profibrotic genes, as well as restoration of mitochondrial function [22]. Our data indicate that higher amounts of ethanol coadministration (16%) in WD-fed animals led to synergistic development of steatosis as shown by Oil Red O staining and measurement of hepatic triglycerides. In our WD + ethanol-fed mice, we could show upregulation of scd-1, a key enzyme of lipid metabolism, which seems to modulate susceptibility to obesity, insulin resistance, diabetes and hyperlipidemia [23]. A recent study similarly showed increased hepatic steatosis and inflammatory markers in high fat diet and ethanol cofeeding [24]. Further studies are urgently needed to elucidate the pathogenic mechanisms displayed by ethanol in metabolic liver disease.
Our data showed increase of liver fat content if ethanol was coadministered. This finding supports clinical evidence, since steatosis is one of the hallmarks of ALD [25]. A widely used model for chronic ALD is the Lieber-DeCarli model, which has been known for over 50 years [26]. However, it only induces mild steatosis and fibrosis without inflammation. Many of the currently existing murine ALD models reflect advanced disease stages. Our model might fill in the gap between models for simple steatosis and models including ethanol combined with CCl 4 mimicking alcoholic steatohepatitis with severe fibrosis [27][28][29][30]. The present model acknowledges the fact that many patients suffering from metabolic liver disease do not develop significant fibrosis, which is only observed in around 30% [4].
Our model further included perisinusoidal fibrosis, which is another advantage since many models lack this hallmark of metabolic liver disease [26,31,32]. The development of F1 fibrosis within a short time frame of seven weeks reflects the usual progression of human early stage of liver disease. Patients with mere steatosis can develop fibrosis, which justifies the need for monitoring [33,34]. To date, laboratory tests show the poorest accuracy of noninvasive diagnostic methods for early-stage liver fibrosis [35]. For further strategies, a model with beginning fibrosis may be helpful in developing and testing novel biomarkers for liver fibrosis. The computer-based detection of fibrosis we used in this study reduces the influence of interobserver variation when classifying fibrosis to distinct stages using semiquantitative scales (F1-F4).
Applicability to the human situation is crucial when using animal models. Therefore, cross-validation in human disease of the animal model is required [36]. We confirmed the phenotype of our animal model in a patient cohort of obese drinkers on a gene transcript level, using well-established and validated key genes for alcoholic and metabolic liver disease within the human ALD cohort [37][38][39]. Nowadays, an increasing number of publications is emerging, indicating the need of a new nomenclature for metabolic liver disease, as the term NAFLD might be misleading. Suggested names like "fatty liver disease" and "metabolic dysfunction-associated fatty liver disease" (MAFLD) have recently been proposed in order to better illustrate the phenotype of these patients [40][41][42]. The here-proposed WD-ethanol model attempts to mimic the pathophysiological complexity of the disease.
In summary, the present study offers a new protocol for inducing metabolic liver disease in mice, including elevated transaminases, severe hepatic steatosis and mild fibrosis induced by WD and ethanol feeding. Our data indicates different mechanisms of steatosis development depending on the provided etiology. To date, this is the first animal study that focuses on the newly defined entity of MAFLD. The main advantages of this model are the wildtype-based approach, technically easy performance and rapid induction within a time period of only seven weeks, making it resource saving. It represents an early-stage liver disease model covering the two most abundant etiologies in humans, namely overnutrition and alcohol consumption. Based on the proposed protocol, WD and ethanol-fed mice could provide a helpful murine model for fatty liver disease research.

Conclusions
This study provides a novel murine model for early-stage liver disease including severe liver steatosis, mild fibrosis and inflammation based on the combination of Western diet and ethanol feeding. Herein, we studied effects of ethanol-feeding on the development of hepatic steatosis. We compared the liver phenotype of our animals to a human cohort of early-stage fatty liver disease, stratified via body mass index (BMI) and current alcohol consumption.