Eicosapentaenoic Acid Improves Hepatic Metabolism and Reduces Inflammation Independent of Obesity in High-Fat-Fed Mice and in HepG2 Cells

The prevalence of nonalcoholic fatty liver disease (NAFLD) is increasing worldwide, concurrent with increased obesity. Thus, there is urgent need for research that can lead to effective NAFLD prevention/treatment strategies. Omega-3 polyunsaturated fatty acids (n-3 PUFAs), including eicosapentaenoic acid (EPA), improve inflammation- and dyslipidemia-related metabolic disorders; however, mechanisms mediating the benefits of n-3 PUFAs in NAFLD treatment are less understood. We previously reported that EPA reversed obesity-induced hepatic steatosis in high-fat (HF)-fed B6 mice. Utilizing a combination of biochemical analyses of liver tissues from HF and HF-EPA-fed mice and a series of in vitro studies in tumor necrosis factor-alpha (TNF-α)-stimulated HepG2 cells, we dissect the mechanistic effects of EPA in reducing hepatic steatosis, including the role of EPA-targeted microRNAs (miRNA). With EPA, hepatic lipid metabolism was improved in HF-EPA mice, as indicated by decreased protein and messenger RNA (mRNA) levels of fatty acid synthase (FASN) and acetyl-CoA carboxylase (Acaca) gene, and increased mRNA levels for the peroxisome proliferator activated receptor-α (Pparα), and carnitine palmitoyltransferase (Cpt) 1a and 2 genes in the HF-EPA mice. Additionally, inflammation was reduced, as shown by decreased tumor necrosis factor-alpha (Tnfα) gene expression. Accordingly, EPA also significantly reduced FASN and ACACA mRNAs in human HepG2 cells. Glycolysis, estimated by extracellular acidification rate, was significantly reduced in HepG2 cells treated with EPA vs. vehicle. Furthermore, we identified several miRNAs that are regulated by EPA in mouse liver, including miR-19b-3p, miR-21a-5p, and others, which target lipid metabolism and inflammatory pathways. In conclusion, our findings provide novel mechanistic evidence for beneficial effects of EPA in NAFLD, through the identification of specific genes and miRNAs, which may be further exploited as future NAFLD therapies.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is considered as the hepatic manifestation of the metabolic syndrome, due its bi-directional relationship with obesity, dyslipidemia, hypertension, and type 2 diabetes mellitus (T2DM) [1]. Prevalence of NAFLD and other metabolic consequences of obesity are increasing as obesity rates continue to increase [2].

miRNA Analyses
Preliminary global miRNA profiling on liver tissue from the HF and HF-EPA groups was performed in collaboration with the University of Houston. Methods are described elsewhere [31]. Candidate miRNAs from this profiling were further validated through real time PCR analyses. Briefly, TaqMan ® Advanced miRNA cDNA Synthesis Kit (Thermo Fisher, Pittsburgh, PA, USA) was used for miRNA cDNA synthesis using samples from liver tissue from the HF and HF-EPA groups as well as HepG2 cells collected from the described treatment. miRNA expression was detected in the HF and HF-EPA groups as well as the treated HepG2 cells utilizing RT-PCR (Bio-Rad, Hercules, CA, USA), TaqMan ® Fast Advanced Master Mix (Thermo Fisher, Pittsburgh, PA, USA) and TaqMan ® Advanced miRNA Assays, (Thermo Fisher, Pittsburgh, PA, USA). miR-191-5p was used as the housekeeping miRNA for both in vivo and in vitro studies.

Statistical Analyses
Results are presented as means ± standard error of the mean (SEM). Data were analyzed by performing t-tests when comparing two groups. Differences are considered significant at p < 0.05. Three to five replicates were used from the HF and HF-EPA groups. Cell culture experiments were repeated at least three times with triplicates within each experiment.

Results
As previously reported, the HF and HF-EPA groups were similar in body weight at the conclusion of the 11-week study. Mouse characteristics are presented in Supplementary Table S2. There were no significant differences in food intake between the two groups [16]. Fatty acid analysis of liver tissue from the HF and HF-EPA groups revealed enrichment with dietary EPA (Table 1). Liver tissue EPA enrichment paralleled previously reported red blood cell enrichment, while other fatty acids were not significantly different between the two groups [16].

EPA Reduces Hepatic Steatosis
Oil Red O staining was used to assess TAG accumulation in the HF and HF-EPA groups ( Figure 1). The HF-EPA group showed significantly reduced hepatic lipid accumulation as well as TAG levels (previously reported) [16].  Results are mean ± SEM. * = p <0.05. PUFA: polyunsaturated fatty acid.

EPA Regulates Hepatic Lipid Metabolism.
As EPA reduced hepatic TAG accumulation, we next wanted to determine the effects of EPA on hepatic lipid metabolism. To determine whether EPA beneficially regulates hepatic lipid metabolism, we used livers from mice fed a HF diet with or without EPA (HF-EPA). Gene expression levels of known contributors to TAG synthesis, including sterol regulatory element-binding protein-1c (Srebp-1c), fatty acid synthase (Fasn), acetyl-CoA carboxylase (Acaca), diacylglycerol O-acyltransferase 2 (Dgat2), and mechanistic target of rapamycin (Mtor) were significantly reduced in the HF-EPA group (Figure 2a). Additionally, we demonstrated that FASN protein content was significantly decreased in (HF-EPA) group compared to HF-fed mice. (Figure  2b,c). Furthermore, western blotting demonstrated significant increases in AMP-activated protein kinase (AMPK) protein content (Figure 2b,d), which negatively regulates Srebp-1c, a known up-regulator of TAG synthesis [32]. Phosphorylated-AMPK was normalized to total-AMPK.

EPA Regulates Hepatic Lipid Metabolism
As EPA reduced hepatic TAG accumulation, we next wanted to determine the effects of EPA on hepatic lipid metabolism. To determine whether EPA beneficially regulates hepatic lipid metabolism, we used livers from mice fed a HF diet with or without EPA (HF-EPA). Gene expression levels of known contributors to TAG synthesis, including sterol regulatory element-binding protein-1c (Srebp-1c), fatty acid synthase (Fasn), acetyl-CoA carboxylase (Acaca), diacylglycerol O-acyltransferase 2 (Dgat2), and mechanistic target of rapamycin (Mtor) were significantly reduced in the HF-EPA group ( Figure 2a). Additionally, we demonstrated that FASN protein content was significantly decreased in (HF-EPA) group compared to HF-fed mice. (Figure 2b,c). Furthermore, western blotting demonstrated significant increases in AMP-activated protein kinase (AMPK) protein content (Figure 2b,d), which negatively regulates Srebp-1c, a known up-regulator of TAG synthesis [32]. Phosphorylated-AMPK was normalized to total-AMPK. Since markers of lipogenesis were decreased, we then examined markers of fatty acid beta-oxidation (β-oxidation) to determine if EPA was causing a beneficial shift in hepatic lipid metabolism towards catabolism. Gene expression levels of fatty acid β-oxidation, including peroxisome proliferator-activated receptor-alpha (Pparα) and carnitine palmitoyltransferase (Cpt)1a and 2 were significantly increased with EPA ( Figure 3). Since markers of lipogenesis were decreased, we then examined markers of fatty acid beta-oxidation (β-oxidation) to determine if EPA was causing a beneficial shift in hepatic lipid metabolism towards catabolism. Gene expression levels of fatty acid β-oxidation, including peroxisome proliferator-activated receptor-alpha (Pparα) and carnitine palmitoyltransferase (Cpt)1a and 2 were significantly increased with EPA ( Figure 3).  To validate in vivo findings, we used HepG2 cells treated with 25 ng/mL TNF-α (to mimic HF diet-induced inflammation) and 50 μM EPA. Gene expression levels of FASN, ACACA, and DGAT2 were significantly reduced when EPA was added to TNF-α ( Figure 4a) compared to TNF-α treatment alone. However, CPT2 was the only fatty acid β-oxidation mRNA tested that was significantly upregulated by EPA (Figure 4b). Peroxisome proliferator-activated receptor-alpha (PPARα) and CPT1a were not significantly different between groups. Data are expressed as mean ± SEM, n = 6, * = p <0.05.

EPA Regulates Hepatic Carbohydrate Metabolism
Since elevated glucose and increased glycolysis can contribute to the production of acetyl-CoA, a key component in de novo lipogenesis [33], we next examined markers associated with hepatic carbohydrate metabolism. To determine whether EPA regulates carbohydrate metabolism in the To validate in vivo findings, we used HepG2 cells treated with 25 ng/mL TNF-α (to mimic HF diet-induced inflammation) and 50 µM EPA. Gene expression levels of FASN, ACACA, and DGAT2 were significantly reduced when EPA was added to TNF-α ( Figure 4a) compared to TNF-α treatment alone. However, CPT2 was the only fatty acid β-oxidation mRNA tested that was significantly upregulated by EPA (Figure 4b).  To validate in vivo findings, we used HepG2 cells treated with 25 ng/mL TNF-α (to mimic HF diet-induced inflammation) and 50 μM EPA. Gene expression levels of FASN, ACACA, and DGAT2 were significantly reduced when EPA was added to TNF-α ( Figure 4a) compared to TNF-α treatment alone. However, CPT2 was the only fatty acid β-oxidation mRNA tested that was significantly upregulated by EPA (Figure 4b). Peroxisome proliferator-activated receptor-alpha (PPARα) and CPT1a were not significantly different between groups. Data are expressed as mean ± SEM, n = 6, * = p <0.05.

EPA Regulates Hepatic Carbohydrate Metabolism
Since elevated glucose and increased glycolysis can contribute to the production of acetyl-CoA, a key component in de novo lipogenesis [33], we next examined markers associated with hepatic carbohydrate metabolism. To determine whether EPA regulates carbohydrate metabolism in the were not significantly different between groups. Data are expressed as mean ± SEM, n = 6, * = p < 0.05.

EPA Regulates Hepatic Carbohydrate Metabolism
Since elevated glucose and increased glycolysis can contribute to the production of acetyl-CoA, a key component in de novo lipogenesis [33], we next examined markers associated with hepatic carbohydrate metabolism. To determine whether EPA regulates carbohydrate metabolism in the liver, we used livers from the HF and HF-EPA mice. Gene expression levels for pyruvate dehydrogenase To validate in vivo findings, we used HepG2 cells treated with TNF-α and EPA. Gene expression levels of MLXIPL and G6PC were significantly reduced when EPA was added to TNF-α (Figure 5b). liver, we used livers from the HF and HF-EPA mice. Gene expression levels for pyruvate dehydrogenase kinase 4 (Pdk4), MLX-interacting protein-like (Mlxipl), glucose-6-phosphatase catalytic subunit (G6pc), and pyruvate kinase L/R (Pklr), were significantly reduced with EPA ( Figure 5a). To validate in vivo findings, we used HepG2 cells treated with TNF-α and EPA. Gene expression levels of MLXIPL and G6PC were significantly reduced when EPA was added to TNF-α ( Figure 5b). Utilizing the Seahorse XF Glycolytic Rate Assay kit, we assessed glycolysis in HepG2 cells treated with 50μM EPA versus BSA. We found significantly reduced compensatory glycolysis ( Figure 6), which is measured when oxidative phosphorylation is inhibited, indicating reduced glycolytic capacity in HepG2 cells treated with EPA. Utilizing the Seahorse XF Glycolytic Rate Assay kit, we assessed glycolysis in HepG2 cells treated with 50µM EPA versus BSA. We found significantly reduced compensatory glycolysis ( Figure 6), which is measured when oxidative phosphorylation is inhibited, indicating reduced glycolytic capacity in HepG2 cells treated with EPA. liver, we used livers from the HF and HF-EPA mice. Gene expression levels for pyruvate dehydrogenase kinase 4 (Pdk4), MLX-interacting protein-like (Mlxipl), glucose-6-phosphatase catalytic subunit (G6pc), and pyruvate kinase L/R (Pklr), were significantly reduced with EPA ( Figure 5a). To validate in vivo findings, we used HepG2 cells treated with TNF-α and EPA. Gene expression levels of MLXIPL and G6PC were significantly reduced when EPA was added to TNF-α ( Figure 5b). Utilizing the Seahorse XF Glycolytic Rate Assay kit, we assessed glycolysis in HepG2 cells treated with 50μM EPA versus BSA. We found significantly reduced compensatory glycolysis ( Figure 6), which is measured when oxidative phosphorylation is inhibited, indicating reduced glycolytic capacity in HepG2 cells treated with EPA.

EPA Reduces Hepatic Inflammation
Mice fed the HF and HF-EPA diets were similar in body weight; thus, we next assessed markers of obesity-associated inflammation in the liver [3,16]. To determine whether EPA reduced inflammation directly in the liver, we used liver from mice fed a HF diet with or without EPA (HF-EPA). Gene expression analyses showed increased interleukin-10 (Il-10) expression, which is anti-inflammatory, in the HF-EPA group. Furthermore, gene expression analysis showed significantly decreased mRNA expression of monocyte chemoattractant protein 1 (Mcp-1), toll-like receptor 4 (Tlr4), TNF-α and mitogen activated protein kinase 8 (Mapk8) in the HF-EPA group (Figure 7a).
To validate in vivo findings, we used HepG2 cells treated with TNF-α and EPA. Gene expression levels of NF-κB and TLR-4 were decreased with the addition of EPA (Figure 7b). Decreases in MCP-1 were not significant and expression of MAPK8 was too low for analysis. Figure 6. HepG2 cells treated with 50 μM eicosapentaenoic acid (EPA) showed a lower glycolytic phenotype compared to cells treated with bovine serum albumin (BSA). (a) Basal glycolysis was not significantly different between the two groups. (b) After inhibition of oxidative phosphorylation, EPA resulted in decreased compensatory glycolysis. Data are expressed as mean ± SEM, n = 4, * = p <0.05.

EPA Reduces Hepatic Inflammation
Mice fed the HF and HF-EPA diets were similar in body weight; thus, we next assessed markers of obesity-associated inflammation in the liver [3,16]. To determine whether EPA reduced inflammation directly in the liver, we used liver from mice fed a HF diet with or without EPA (HF-EPA). Gene expression analyses showed increased interleukin-10 (Il-10) expression, which is anti-inflammatory, in the HF-EPA group. Furthermore, gene expression analysis showed significantly decreased mRNA expression of monocyte chemoattractant protein 1 (Mcp-1), toll-like receptor 4 (Tlr4), TNF-α and mitogen activated protein kinase 8 (Mapk8) in the HF-EPA group (Figure 7a).
To validate in vivo findings, we used HepG2 cells treated with TNF-α and EPA. Gene expression levels of NF-κB and TLR-4 were decreased with the addition of EPA (Figure 7b). Decreases in MCP-1 were not significant and expression of MAPK8 was too low for analysis.

EPA Regulates Hepatic miRNA Involved in Lipid Metabolism and Inflammation
MiRNA regulate various genes related to metabolism and inflammation [34], with few studies indicating n-3 PUFA to target miRNAs [35,36]. Therefore, we sought to determine differences in miRNA regulation by EPA in the livers of mice fed a HF diet. Global miRNA profiling in liver tissues of mice fed with or without EPA revealed about 30 miRNAs that were significantly different between the HF and HF-EPA samples (data not shown). Targets were selected based on >1.5-fold

EPA Regulates Hepatic miRNA Involved in Lipid Metabolism and Inflammation
MiRNA regulate various genes related to metabolism and inflammation [34], with few studies indicating n-3 PUFA to target miRNAs [35,36]. Therefore, we sought to determine differences in miRNA regulation by EPA in the livers of mice fed a HF diet. Global miRNA profiling in liver tissues of mice fed with or without EPA revealed about 30 miRNAs that were significantly different between the HF and HF-EPA samples (data not shown). Targets were selected based on >1.5-fold expression, statistical significance of p ≤ 0.05, and false discovery rate (FDR) ≤5%. miR-19b-3p, -21a-5p, and -101b-3p were identified for validation as they were documented to have a role in NAFLD [37][38][39]. Furthermore, miR-let7a-5p was chosen for validation for its known role in inflammation [40,41] and miR-455-5p was chosen for its novelty related to liver function.
To further validate the liver specific function of these miRNAs, we used HepG2 cells treated with TNF-α and EPA in order to mimic in vivo studies. Utilizing RT-qPCR, we found significant decreases in miR-21a-5p, miR-101b-3p and miR-455-5p (Figure 9). miR-let-7a-5p expression was too low for detection.

Discussion
NAFLD is projected to be the leading cause of liver related morbidity and mortality within the next 20 years [42]. Therefore, therapeutic strategies are necessary to prevent metabolic derangements associated with obesity-related NAFLD since it is known to instigate and exacerbate insulin resistance and systemic inflammation [43].
In this study, we report (1) reduced hepatic TAG accumulation due to decreased fatty acid synthesis and upregulated β-oxidation, (2) reduced hepatic carbohydrate metabolism, and (3) reduced hepatic inflammation with EPA supplementation independent of body weight in C57BL6 mice fed a HF diet. Additionally, we were able to validate these findings using a HepG2 cell culture model of NAFLD and show that EPA has liver specific benefits in an inflammatory environment. To our knowledge, this is the first study to indicate that EPA targets hepatic miRNA involved in NAFLD pathways in order to improve metabolism and reduce inflammation.
Excessive hepatic accumulation of TAG (≥5%) is the hallmark of NAFLD [7]. Increased dietary intake, hyperglycemia and hyperinsulinemia largely influence hepatic de novo lipogenesis [44]. Insulin upregulates liver X receptor (LXR), sterol regulatory element binding protein-1c (SREBP-1c) and carbohydrate response element binding protein (ChREBP), all of which are transcription factors that control glycolytic and lipogenic genes [5,45,46]. It is known that n-3 PUFAs alter expression and nuclear localization of these transcription factors and thus reduce TAG synthesis [47,48]. Accordingly, we have shown decreased Srebp-1c and ChREBP (Mlxipl) expression as well as decreases in lipogenic genes, including Acaca and Fasn and glycolytic genes, including pyruvate kinase L/R (Pklr) in the HF-EPA mice. Additionally, we also found decreases in Dgat2 expression, the enzyme responsible for TAG synthesis [49]. Interestingly, Calo et al. demonstrated reduced expression of glycolytic and gluconeogenic genes in miR-21 knockout mice [50]. Therefore, our report of EPA-mediated decreases in miR-21a-5p in the HF-EPA group could be associated with decreases in the glycolytic and gluconeogenic genes seen in our study. In addition, metabolism studies from HepG2 cells treated with EPA indicated reductions in glycolysis.
Furthermore, we also report increased AMPK in the livers of mice fed a HF diet supplemented with EPA. AMPK is thought of as a "metabolic master switch" due to its concurrent inhibition of fatty acid synthesis, through phosphorylation of SREBP-1c [32] or raptor of the mTOR complex [51], and activation of catabolic pathways, including β-oxidation [52]. Indeed, activation of AMPK prevents steatosis [53,54]. Accordingly, we also found decreases in Mtor gene expression in the HF-EPA group. It has been suggested that mTOR activation is required for SREBP-1c activity related to lipogenesis, particularly related to insulin-mediated activity in the case of insulin resistance [55]. Our findings on hepatic lipogenesis and reduced TAG accumulation with EPA are particularly interesting since we initially reported reduced plasma insulin but unimproved glucose tolerance in the HF-EPA group [16]. Therefore, we suggest that EPA regulates lipogenesis by controlling genes involved in these pathways, independent of glycemic control but aided by reductions in insulin.
Although there is debate on the role of miR-21 in relation to NAFLD [34,50,56], our findings indicate a beneficial decrease in miR-21a-5p with regard to NAFLD and agree with a recent study showing that miR-21 knockout mice had reduced hepatic steatosis and lipogenesis [50]. Indeed, serum levels of miR-21 are higher in individuals with NAFLD [38,57]. PPARα, a key transcriptional regulator of hepatic energy homeostasis, is a suggested target of miR-21, thus indicating an inverse relationship between the two [34,58]. It is well established that PUFAs can act as ligands for PPARs, including PPARα [59] and thus upregulate β-oxidation [60,61] and related mitochondrial enzymes [47]. In agreement, our study showed an increase in Pparα gene expression and related increases in mitochondrial enzymes Cpt1a and Cpt2, indicating a body weight-independent increase in hepatic β-oxidation in mice fed a HF diet supplemented with EPA. Moreover, we suggest that by targeting miR-21a-5p, EPA could have indirectly increased Pparα expression and increased hepatic fatty acid transport and β-oxidation. Others have also demonstrated that miR-21 is targeted by n-3 PUFA [35,36], but no other study has reported miR-21 to be targeted by EPA in relation to hepatic lipid metabolism.
Inflammation can occur in the liver via (1) increased adipocyte release of pro-inflammatory cytokines that are delivered to the liver (i.e., as part of obesity-associated systemic inflammation) which may be worsened by insulin resistance; and/or (2) the influx of fatty acids from adipocytes overwhelms adaptive mechanisms causing hepatocyte dysfunction and injury in a process known as lipotoxicity [62]. Interestingly, we report reduced hepatic inflammation in the HF-EPA group despite similarities in body weight with mice fed HF diet. Reductions in NF-κB and pro-inflammatory cytokines, such as Tnf-α and Mcp-1, as well as increases in anti-inflammatory cytokines, such as Il-10, are unsurprising since n-3 PUFAs are well-established anti-inflammatory agents [63]; however, reporting improvements in inflammation independent of obesity is noteworthy. Furthermore, we were able to validate liver specific reductions in inflammation with EPA, via reductions in NF-κB, and TLR4.
Global miRNA profiling in livers from the HF and HF-EPA groups revealed significant increases in the miR-let-7 family, including miR-let7a-5p, -let7b-3p, -let7c-5p, and -let7d-5p (data not shown) with EPA. Of which, we selected and were able to validate significant increases in miR-let7a-5p with qPCR. While no study has indicated an n-3 mediated increase in the miR-let-7 family or a relationship with NAFLD, specifically, few studies have reported that miR-let-7 plays a role in the regulation of inflammatory pathways in cancer and atherosclerosis [40,41,64]. A feedback loop between miR-let-7 and transcriptional regulators such as NF-κB, as well as receptors including TNF-α and Tlr4, is consistently reported and indicate that an increases in NF-κB, for example, reduces miR-let-7, which results in the increased production of pro-inflammatory cytokines, such as IL-6 [40,41]. Therefore, increases in miR-let-7 would decrease production of pro-inflammatory cytokines. Furthermore, Xu et al. demonstrated a role for miR-let-7 in decreasing Ras activity in both NF-κB and MAPK pathways [64], thereby decreasing inflammation. In the MAPK/ERK/JNK pathway, JNK1 (also known as MAPK8) is known to promote the development of steatosis and inflammation [65]. Taken together with our gene expression studies, our findings indicate that EPA beneficially upregulates miR-let-7 and decreases the production of pro-inflammatory cytokines.
The one study that has examined the effect of n-3 PUFA intake on miRNA regulation reported n-3 PUFA mediated decreases in miR-19b-3p paired with decreases in markers of inflammation, including Tlr4 and Mapk [18]. Our findings are in agreement with these findings; however, we cannot limit the activity of miR-19b-3p to the regulation of inflammation given other studies indicating negative regulation of PPARα and AMPK [37], as well as our findings that also suggest a role in increasing β-oxidation and reducing lipid synthesis. Collectively, this highlights the potential role of n-3 mediated miR-19b-3p regulation of inflammatory and lipid metabolism pathways related to NAFLD and should be investigated further.
Lastly, we utilized a diet containing 36 g/kg EPA, which accounts for approximately 6.75% of total energy intake in the HF-fed mice [16]. This dosage exceeds the current recommendation of 1-4 g/day of n-3 PUFA in humans. However, human studies have been performed with doses up to 15g/day [15,66]. Thus, future studies are needed to examine the dose-dependent effects of EPA in obesity-related NAFLD. There are a few limitations to the current study. Only male mice were used in the current study, as they are more susceptible to diet-induced obesity and metabolic dysfunctions than females in this mouse strain. Thus, this limits translation to females [67] and warrants future studies. Another limitation to the current study is the lack of a low-fat (LF) group and comparison to LF-EPA-fed mice, which also requires future studies.

Conclusions
In conclusion, our findings demonstrate that EPA exhibits protective effects in liver steatosis and inflammation that may be mediated by genes and miRNAs in lipid and carbohydrate metabolism as well as inflammatory pathways ( Figure 10).
Given the side effects and limitations of pharmacological therapies, these findings provide the foundation for novel therapeutics related to n-3 PUFAs in improving metabolism and inflammation in the liver. However, future studies are necessary to understand the role of these n-3 PUFA-targeted miRNAs as novel therapies in the liver and other tissues in relation to metabolic diseases. Additionally, EPA reduces hepatic inflammation by reducing expression of toll-like receptor 4 (TLR4), as well as the production of pro-inflammatory cytokines, monocyte chemoattractant protein-1 (MCP-1), nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB), tumor necrosis factor-alpha (TNF-α), and mitogen activated protein kinase 8 (Mapk8). MicroRNA, including miR-21 and miR-let7a-5p are targeted by EPA and may contribute to improved hepatic metabolism and reduced inflammation.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Table S1: Diet Composition, Table S2. Characteristics in C57BL/6J HF and HF-EPA Mice.   and gluconeogenesis and thus the amount of glucose available for de novo lipogenesis. Further, TAG synthesis is reduced with EPA, through activation of AMP-activated protein kinase (AMPK), resulting inhibition of sterol regulatory element-binding protein-1c (SREBP-1c) and reduction of enzymes, fatty acid synthase (FASN), acetyl-CoA carboxylase (ACC), and diglyceride acetyltransferase (DGAT2). Utilization of free fatty acids (FFA) for adenosine triphosphate (ATP) generation is upregulated with EPA through induced expression of peroxisome proliferator-activated receptor-alpha (PPARα) and carnitine palmitoyltransferase (CPT)-1a and CPT2. Additionally, EPA reduces hepatic inflammation by reducing expression of toll-like receptor 4 (TLR4), as well as the production of pro-inflammatory cytokines, monocyte chemoattractant protein-1 (MCP-1), nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB), tumor necrosis factor-alpha (TNF-α), and mitogen activated protein kinase 8 (Mapk8). MicroRNA, including miR-21 and miR-let7a-5p are targeted by EPA and may contribute to improved hepatic metabolism and reduced inflammation.