Lemon Balm and Its Constituent, Rosmarinic Acid, Alleviate Liver Damage in an Animal Model of Nonalcoholic Steatohepatitis

nutrients

Article Lemon Balm and Its Constituent, Rosmarinic Acid, Alleviate Liver Damage in an Animal Model of Nonalcoholic Steatohepatitis

1, 1, 2 1 1 Myungsuk Kim †, GyHye Yoo †, Ahmad Randy , Yang-Ju Son , Chi Rac Hong , Sang Min Kim 1 and Chu Won Nho 1,*

1 Smart Farm Research Center, Korea Institute of Science and Technology, Gangneung, Gangwon-do 25451, Korea; [email protected] (M.K.); [email protected] (G.Y.); [email protected] (Y.-J.S.); [email protected] (C.R.H.); [email protected] (S.M.K.) 2 Research Center for Chemistry, Indonesian Institute of Sciences (LIPI), Kawasan Puspiptek, Serpong 15314, Indonesia; [email protected] * Correspondence: [email protected]; Tel.: +82-33-650-3420 These two authors contributed equally to this work. † Received: 18 February 2020; Accepted: 16 April 2020; Published: 22 April 2020

Abstract: Nonalcoholic fatty liver disease (NAFLD) ranges in severity from hepatic steatosis to cirrhosis. Lemon balm and its major constituent, rosmarinic acid (RA), effectively improve the liver injury and obesity; however, their therapeutic effects on nonalcoholic steatohepatitis (NASH) are unknown. In this study, we investigated the effects of RA and a lemon balm extract (LBE) on NAFLD and liver fibrosis and elucidated their mechanisms. Palmitic acid (PA)-exposed HepG2 cells and db/db mice fed a methionine- and choline-deficient (MCD) diet were utilized to exhibit symptoms of human NASH. LBE and RA treatments alleviated the oxidative stress by increasing antioxidant enzymes and modulated lipid metabolism-related gene expression by the activation of adenosine monophosphate-activated protein kinase (AMPK) in vitro and in vivo. LBE and RA treatments inhibited the expression of genes involved in hepatic fibrosis and inflammation in vitro and in vivo. Together, LBE and RA could improve liver damage by non-alcoholic lipid accumulation and may be promising medications to treat NASH.

Keywords: nonalcoholic steatohepatitis (NASH); lemon balm extract (LBE); rosmarinic acid (RA); AMP-activated protein kinase (AMPK); antioxidants; anti-inﬂammation

1. Introduction Nonalcoholic steatohepatitis (NASH), one of the nonalcoholic fatty liver diseases (NAFLDs), accompanied by asymptomatic hepatic steatosis and fibrosis [1]. Although the precise mechanism underlying the progression from fatty liver to NASH has not been determined, the pathogenesis of NASH is believed to involve a multi-hit model; hepatic lipid accumulation, and inflammatory responses mediated by oxidative stress [2,3]. Dietary habits and the imbalance between energy intake and expenditure result in accumulation of lipids in the liver, which is a first hit leading to NASH. Excess hepatic lipids increase free fatty acids, which are converted to lipotoxic lipids, leading to the production of reactive oxygen species as well as inflammatory responses. These processes consequently result in NASH. The ideal treatment for NASH should not only reverse the accumulation of triglycerides (TGs) in hepatocytes but also effectively suppress hepatic inflammations [4]. No pharmacological agent has been approved by the U.S. Food and Drug Administration for the treatment of NAFLD/NASH, and therefore, there is an urgent need for the development of effective treatments.

Nutrients 2020, 12, 1166; doi:10.3390/nu12041166 www.mdpi.com/journal/nutrients Nutrients 2020, 12, 1166 2 of 15

Adenosine monophosphate-activated protein kinase (AMPK) is a serine/threonine kinase, and the AMPK pathway plays a pivotal role in regulating the energy balance in cells [5]. Under low adenosine triphosphate (ATP) -to- adenosine monophosphate (AMP) ratio, AMPK is activated by phosphorylation of its α-subunit and directs the inhibition of anabolic pathways, accompanied by the induction of catabolic pathways. AMPK activation suppresses fatty acid synthesis, enhances β-oxidation, and induces anti-inflammation by decreasing the activation of hepatic stellate cells [5,6]. In an animal model of NASH, AMPK activation decreased hepatic lipid accumulation and improved insulin resistance. Furthermore, recent reports have suggested that nuclear factor erythroid 2-related factor 2 (NRF2), a major regulator of the redox status, is a potential target of AMPK, and AMPK inhibits the inflammation [7,8]. Considering the two-hit hypothesis of NASH pathogenesis, involving lipid accumulation and inflammation with oxidative stress, the activation of AMPK is regarded as one of potential targets for NAFLD/NASH. Lemon balm (Melissa officinalis) has been used as a medicinal plant to treat anxiety, gastrointestinal disorders, and Alzheimer’s disease [9–11]. Moreover, it has an antioxidant activity and protects against oxidative stress-induced apoptosis [12]. An ethyl acetate fraction of lemon balm extract has been shown to reduce high-fat-induced NAFLD in mice [13,14]. The main active compound of lemon balm is rosmarinic acid (RA, O-caffeoyl-3,4-dihydroxyphenyl lactic acid), which possesses a multitude of biological activities; the antioxidant, anti-inflammatory, anti-fatty liver, and anti-obesity effects [15–17]. However, it remains unknown whether lemon balm extract or RA exerts beneficial effects on NASH, and their molecular mechanisms of action have not been elucidated. In this study, we evaluated whether lemon balm extract and RA could suppress the pathogenesis of NASH.

2. Materials and Methods

2.1. Plant Material and Preparation of RA Aerial parts of lemon balm were supplied by Richwood Pharma Co., Ltd. (Seoul, Korea). One kilogram of dried aerial parts of lemon balm was extracted by refluxing EtOH (0–100%) in de-ionized water for 4 h and then filtered through Whatman No. 1 filter paper (GE Healthcare Life Sciences, Pittsburgh, PA, USA). RA was identified from lemon balm (3.59% in lemon balm extract, LBE) as described previously [18].

2.2. Reagents and Materials Dulbecco’s modiﬁed Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from HyClone Laboratories (Logan, UT, USA). Palmitic acid (PA), 3-(4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide (MTT), Oil Red O, and other reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Primary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and Cell Signaling Technologies (Beverly, MA, USA) (Supplementary Table S1).

2.3. Cell Culture HepG2 hepatocellular carcinoma cells (American Type Culture Collection, Manassas, VA, USA) were grown in DMEM supplemented with 10% FBS and 100 U/mL penicillin at 37 ◦C in an atmosphere of 5% CO2. For experiments, cells were incubated in the presence or absence of PA, which induces lipid accumulation in hepatocytes, for 24 h and then in the presence or absence of LBE (50 or 100 µg/mL) or RA (20 or 40 µM) for 24 h.

2.4. Chemical Analysis The concentration of RA in lemon balm extracts was determined using an Agilent 1260 Inﬁnity Liquid Chromatograph (Agilent Technologies, Santa Clara, CA, USA), equipped with a quaternary pump delivery system (G1331B), auto-sampler (G1329B), column thermostat (G1316A), and diode array and multiple wavelength detector (G1315D). After pretreatment, an amount equivalent to 10 µL Nutrients 2020, 12, 1166 3 of 15 of each sample was injected into a Capcell Pak C18-Column (5 µm, 4.6 250 mm, Shiseido, Tokyo, × Japan) at 40 ◦C, with detection at λ = 330 nm. The mobile phase A was 0.1% (v/v) formic acid in water, while mobile phase B was acetonitrile. The separation was obtained at a ﬂow rate of 0.8 mL/min with a gradient program that allowed for 5 min at 10% B followed by a 35 min step that raised eluent B to 100%, after then holding 5 min. Total analysis time was 45 min.

2.5. Animals and Treatments Of thirty-five four-week-old male db/db mice (Central Laboratory Animal, Inc., Seoul, Korea), 28 were fed the methionine- and choline-deficient (MCD, Central Laboratory Animal, Inc.) diet, and seven were fed a regular chow (control group) for 2 weeks. The 28 mice were divided into four groups (n = 7 each) and treated with LBE (200 mg/kg daily), RA (10 or 30 mg/kg daily), or the vehicle alone (MCD group) by oral gavage for an additional 2 weeks while continuing to be fed the MCD diet. The control and MCD diet-fed mice were administered an equal volume of the vehicle (carboxymethyl cellulose). The body weight and food intake were measured twice weekly. At the end of the treatment period, all mice were fasted overnight and sacrificed by intraperitoneal injection of a Zoletil–Rompun mixture. The livers and the blood were collected and stored at 80 C. − ◦ The experimental protocol was approved by the Animal Use and Care Committee of the Korea Institute of Science and Technology (2015-012; Seoul, Korea).

2.6. Histopathological Analysis Liver tissues were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with Hematoxylin and eosin (H&E). Frozen livers embedded at optimal cutting temperature were sectioned at a thickness of 4 µm using a cryostat, fixed in 4% (v/v) formalin for 10 min, and stained with an Oil Red O working solution.

2.7. Biochemical Analysis The activities of alanine transaminase (ALT) and aspartate transaminase (AST) in the serum were determined using ALT and AST activity assay kits, respectively (BioVision, Minneapolis, MN, USA). Hepatic TG levels were measured using TG kit (Cayman Chemical, Ann Arbor, MI, USA).

2.8. Determination of the Liver Collagen Content Paraﬃn-embedded liver sections were stained for collagen with Sirius Red, as described previously [19]. Hepatic hydroxyproline was measured using a hydroxyproline colorimetric assay kit (BioVision) to quantify the liver collagen content.

2.9. Western Blot Analysis Homogenized liver specimens (approximately 100 mg each) or HepG2 cells were lysed in lysis buffer supplemented with a protease inhibitor cocktail (Sigma–Aldrich). Protein concentrations were measured by the Bradford assay. Western blot analysis was performed according to a standard procedure using the specific antibodies (Table1). Briefly, protein samples were resolved on a 10% sodium dodecyl sulfate-polyacrylamide gel. These proteins were then transferred to polyvinylidene difluoride membranes. The membranes were blocked with PBS-T containing 3% bovine serum albumin (BSA). Subsequently, the membranes were reacted with specific antibodies at 1/1000 and subsequently specific secondary antibodies at 1/3000. Proteins were detected using an enhanced chemiluminescence detection system (Amersham, Buckinghamshire, UK) and visualized using a LAS-3000 luminoimager (Fuji Film Co., Tokyo, Japan). The density of each band relative to that of the β-actin band was determined using the luminoimager (n = 3). Nutrients 2020, 12, 1166 4 of 15

Nutrients 2020, 12, x FOR PEER REVIEW 4 of 16 2.10. Quantitative Real-Time Reverse Transcription–Polymerase Chain Reaction TotalTotal ribonucleic ribonucleic acid acid (RNA)(RNA) waswas isolated isolated from from liver specimens and and HepG2 HepG2 cells cells using using the the TRIzolTRIzol reagent reagent (Invitrogen, (Invitrogen, Carlsbad, Carlsbad, CA, USA), and the RNA concentration was was quantified quantiﬁed spectrophotometrically at 260 nm. cDNA was synthesized from 2 µg of total RNA in a reaction spectrophotometrically at 260 nm. cDNA was synthesized from 2 µg of total RNA in a reaction mixture containing oligo (dT) and a reverse transcription premix (ELPIS-Biotech, Daejeon, Korea). mixture containing oligo (dT) and a reverse transcription premix (ELPIS-Biotech, Daejeon, Korea). Quantitative real-time PCR (qPCR) was performed using a SYBR Green master mix (Roche, Basel, Quantitative real-time PCR (qPCR) was performed using a SYBR Green master mix (Roche, Basel, Switzerland) in a Light Cycler 480 real-time PCR system (Roche) and the following conditions: 40 Switzerland) in a Light Cycler 480 real-time PCR system (Roche) and the following conditions: 40 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s. mRNA expression levels of mouse genes cycles of 95 C for 30 s, 55 C for 30 s, and 72 C for 30 s. mRNA expression levels of mouse genes were were determined◦ using the◦ specific primers◦ listed in Supplementary Information Table 2. determined using the speciﬁc primers listed in Supplementary Information Table S2.

2.11.2.11. Statistical Statistical AnalysisAnalysis Results are presented as the mean ± standard deviation (SD). Statistical analyses were performed Results are presented as the mean standard deviation (SD). Statistical analyses were performed using SPSS 12.0 (SPSS, Inc., Chicago, IL,± USA). Differences between groups were assessed by one- using SPSS 12.0 (SPSS, Inc., Chicago, IL, USA). Diﬀerences between groups were assessed by one-way way analysis of variance, followed by Duncan’s test. A p-value < 0.05 was considered statistically analysis of variance, followed by Duncan’s test. A p-value < 0.05 was considered statistically signiﬁcant. significant. 3. Results 3. Results 3.1. LBE and RA Regulate Lipid Metabolism and Oxidative Stress in PA-Treated HepG2 Cells 3.1. LBE and RA Regulate Lipid Metabolism and Oxidative Stress in PA-Treated HepG2 Cells To investigate whether lemon balm extract could suppress lipid accumulation, HepG2 cells were treatedTo with investigate PA alone whether or with lemon LBEs extracted balm extract by 0–100% could suppress EtOH (80 lipidµg/ mL)accumulation, for 24 h. Lemon HepG2 balm cells extract were obtainedtreated with with PA 20% alone EtOH or showedwith LBEs the extracted best activity by 0 on%– the100% suppression EtOH (80 μg/mL) of lipid for and 24 TG h. accumulation,Lemon balm andextract increase obtained of AMPK with phosphorylation 20% EtOH showed (Figure the1 A–C). best activity Therefore, on we the utilized suppression the lemon of lipid balm and extract TG obtainedaccumulation, with 20% and EtOH increase for of all AMPK subsequent phosphorylation experiments (Figure and referred 1A–C). Therefore, to it as LBE we inutilized this paper. the However,lemon balm the extract content obtained of RA, with the major 20% EtOH active for compound all subsequent in lemon experiments balm, was and not referred as high to it in as lemon LBE balmin this extract paper. obtained However, with the 20% content EtOH of as RA, it was the in major lemon active balm compound extracts obtained in lemon with balm, 40–100% was not EtOH as (Figurehigh in1 Dlemon and Tablebalm1 extract). obtained with 20% EtOH as it was in lemon balm extracts obtained with 40%–100% EtOH (Figure 1D and Table 1).

FigureFigure 1. 1. EEffectsﬀects of of lemon lemon balm balm extracts extracts (0%,(0%, 20%,20%, 4040%,%, 60 60%,%, 80 80%,%, or or 100% 100% EtOH) EtOH) on on ( (AA)) lipid lipid accumulationaccumulation byby OilOil red-Ored-O stainingstaining andand ((BB)) triglyceridetriglyceride (TG)(TG) contents in in palmitic palmitic acid acid (PA) (PA)-treated-treated HepG2HepG2 cells. cells. EEffectsﬀects ofof lemonlemon balmbalm extractsextracts onon (C)) phosphorylation of AMP AMP-activated-activated protein protein kinase kinase (AMPK)(AMPK) in in HepG2 HepG2 cells. cells. ((DD)) TheThe rosmarinicrosmarinic acidacid (RA)(RA) content in the lemon balm balm extracts extracts by by liquid liquid chromatograph.chromatograph. Results Results are are expressed expressed as as the the mean mean ± SDSD ofof threethree independentindependent experiments.experiments. #### pp << 0.010.01 ± comparedcompared with with the the control control group; group; * *p p< < 0.050.05 andand ** p < 0.010.01 compared compared with with the the PA PA-treated-treated cells. cells.

Nutrients 2020, 12, 1166 5 of 15

Nutrients 2020, 12, x FOR PEER REVIEW 5 of 16 Table 1. Diﬀerence of RA content between solvents.

SampleTable 1. Difference of RA Solvent content between solvents g/100 g. ext. Sample H OSolvent 4.476 0.041 g/100 g ext. 2 ± 20% EtOH (LBE) 4.288 0.430 H2O ± 4.476 ± 0.041 40% EtOH 4.711 0.006 Lemon balm extract 20% EtOH (LBE) ± 4.288 ± 0.430 60% EtOH 4.865 0.068 ± 80% EtOH40% EtOH 5.282 0.058 4.711 ± 0.006 Lemon balm extract ± 100%60% EtOH EtOH 5.221 0.016 4.865 ± 0.068 80% EtOH ± 5.282 ± 0.058 100% EtOH 5.221 ± 0.016 The treatment with LBE (the lemon balm extract obtained with 20% EtOH) or RA inhibited the The treatment with LBE (the lemon balm extract obtained with 20% EtOH) or RA inhibited the PA-induced upregulation of the accumulation of lipids (Figure2A,B) and cellular TGs (Figure2C,D). PA-induced upregulation of the accumulation of lipids (Figure 2A,B) and cellular TGs (Figure 2C,D).

FigureFigure 2. 2.E Effectsﬀects of of LBE LBE (lemon (lemon balm balm extract extract obtained obtained with with 20% 20% EtOH) EtOH) and RA on lipid and TG accumulationaccumulation in in palmitic palmitic acid acid (PA)-treated (PA)-treated HepG2 HepG2 cells. cells. Lipid Lipid accumulation accumulation with with (A) ( LBEA) LBE and and (B) RA,(B) andRA, the and TG the content TG content with (withC) LBE (C) andLBE ( Dand) RA (D) in RA PA-treated in PA-treated HepG2 HepG2 cells. cells. Results Results are expressedare expressed as the as meanthe meanSD of± SD three of three independent independent experiments. experiments.## p < ##0.01 p < 0.01 compared compared with with the controlthe control group; group; * p < *0.05 p < ± and0.05 ** andp < 0.01** p < compared 0.01 compared with thewith PA-treated the PA-treated cells. cells.

PAPA treatment treatment increased increased the the expression expression of of lipogenic lipogenic genes; genes; sterol sterol regulatory regulatory element-binding element-binding protein-1cprotein-1c (SREBP-1c), (SREBP-1c), fatty fatty acid acid synthase synthase (FAS), (FAS), and and stearoyl-CoA stearoyl-CoA desaturase-1 desaturase (SCD-1)-1 (SCD- (Figure1) (Figure3A–D), 3A– andD), suppressed and suppressed the mRNA the mRNA and protein and protein expression expression of lipolytic of lipolytic genes; peroxisome genes; peroxisome proliferator-activated proliferator- receptoractivatedα (PPAR receptorα), peroxisomeα (PPARα), proliferator-activatedperoxisome proliferator receptor-activatedγ coactivator receptor γ 1 αcoactivator(PGC-1α), 1 andα (PGC carnitine-1α), palmitoyland carnitine transferase palmitoyl 1L (CPT-1L) transferase (Figure 1L (CPT3E–H).-1L) Treatment (Figure 3E with–H). LBE Treatment or RA reversedwith LBE these or RA changes reversed in a dose-dependentthese changes in a manner. dose-dependent Moreover, manner. LBE and Moreover, RA increased LBE and the RA mRNA increased and the protein mRNA expression and protein of antioxidant-relatedexpression of antioxidant genes; NRF2,-related superoxide genes; NRF dismutase2, superoxide 1 (SOD1), dismutase and 1 catalase (SOD1), (Figure and catalase4A–D). (Figure 4A–D).

Nutrients 2020, 12, 1166 6 of 15

Nutrients 2020, 12, x FOR PEER REVIEW 6 of 16

FigureFigure 3. 3.E ﬀEffectsects of of LBE LBE and and RA RA on theon the expression expression of lipid of lipid metabolism metabolism genes genes and proteins and proteins in HepG2 in HepG2 cells incubatedcells incubated with LBE with or RA LBE for or 24 hRA with for or 24 without h with PA. or Protein without levels PA. of Protein sterol regulatory levels of element-bindingsterol regulatory protein-1celement-binding (SREBP-1c), protein fatty-1c acid (SREBP synthase-1c), (FAS), fatty andacid stearoyl-CoA synthase (FAS desaturase-1), and stearoyl (SCD-1)-CoA with desatura (A) LBEse-1 and(SCD (B)-1 RA;) with peroxisome (A) LBE and proliferator-activated (B) RA; peroxisome receptor proliferatorα (PPAR-activatedα), peroxisome receptor α proliferator-activated (PPARα), peroxisome receptorproliferatorγ coactivator-activated 1 α receptor(PGC-1 α γ,) coactivator and carnitine 1α palmitoyl (PGC-1α, transferase) and carnitine 1L (CPT-1L) palmitoyl with transferase (E) LBE and 1L (F()CPT RA.-1L mRNA) with levels(E) LBE of and the genes(F) RA. encoding mRNA levelsSREBP-1c, of the FAS, genesand encodingSCD-1 with SREBP (C-)1c, LBE FAS and, and (D )SCD RA;-1 PPARwithα (,C PGC-1) LBE αand, and (DCPT-1L) RA; PPARα,with ( GPGC) LBE-1α, and and ( HCPT) RA.-1Lβ with-Actin (G was) LBE used and as(H an) RA. internal β-Actin control was used for westernas an internal blotting control and qPCR for western analysis. blotting Results and are qPCR expressed analysis as. Results the mean are expressedSD of three as independentthe mean ± SD ± experiments.of three independent# p < 0.05 experiments. and ## p < 0.01 # p < compared0.05 and ## withp < 0.01 the compared control group; with the * p control< 0.05 andgroup; ** p *

Nutrients 2020, 12, 1166 7 of 15

Nutrients 2020, 12, x FOR PEER REVIEW 7 of 16

FigureFigure 4. 4.E Effectsﬀects of of LBE LBE and and RA RA on on the the expressionexpression ofof antioxidativeantioxidative stressstress genesgenes and proteins in HepG2 cellscells incubated incubated with with LBE LBE or or RA RA for for 24 24 h h with with oror withoutwithout PA.PA. ProteinProtein levelslevels ofof nuclearnuclear factor erythroid 2-related2-related factor factor 2 (NRF2),2 (NRF2 superoxide), superoxide dismutase dismutase 1 (SOD1), 1 (SOD1 and), and heme heme oxygenase oxygenase 1 (HO-1) 1 (HO with-1) with (A) LBE(A) andLBE (B and) RA. (B mRNA) RA. mRNA levels levels of the of genes the genes encoding encodingNRF2, NRF SOD1,2, SOD1,and HO-1and HOwith-1 with (C) LBE(C) LBE and and (D) RA.(D) β-ActinRA. β- wasActin used was as used an internal as an internal control for control western for blotting western andblotting qPCR and analysis. qPCR Resultsanalysis are. Results expressed are asexpressed the mean as theSD mean of three ± SD independent of three independent experiments. experiments.# p < 0.05 # comparedp < 0.05 compar withed the with control the control group; ± * pgroup;< 0.05 * and p < 0.05 ** p

3.2.3.2. LBE LBE and and RA RA Increase Increase the the Level Level of of Phosphorylated Phosphorylated AMPKAMPK inin HepG2HepG2 Cellscells ToTo elucidate elucidate the the molecularmolecular mechanism by by which which LBE LBE and and RA RA suppress suppress lipid lipid accumulation, accumulation, the thephosphorylation phosphorylation of AMPK of AMPK was evaluated was evaluated in HepG2 in HepG2cells. Treatment cells. Treatmentwith LBE or with RA significantly LBE or RA signiﬁcantlyincreased AMPK increased phosphorylation AMPK phosphorylation in a dose in- and a dose- time and-dependent time-dependent manner manner (Figure (Figure 5A–D).5A–D). The Thephosphorylation phosphorylation of acetyl of acetyl-CoA-CoA carboxylase carboxylase (ACC) (ACC) was waselevated elevated after after the treatment the treatment with withLBE LBEand andRA. RA. Among Among AMPK’s AMPK’s upstream upstream kinases, kinases, the levels the levels of liver ofliver kinase kinase B1 (LKB1) B1 (LKB1) were wereincreased increased by both by 2+ bothLBE LBE and andRA treatments; RA treatments; however, however, the phosphorylation the phosphorylation of Ca2+ of/calmodulin Ca /calmodulin-dependent-dependent protein protein kinase kinaseII (CaMKII II (CaMKII)) was elevated was elevated only onlyby LBE by LBE(Figure (Figure 5E,F).5E,F). Under Under PA treatment, PA treatment, LBE LBE also also increased increased the thephosphorylated phosphorylated AMPK, AMPK, the the phosphorylated phosphorylated ACC ACC,, and and the the phosphorylated phosphorylated CaMKII, CaMKII, not not the phosphorylatedphosphorylated LBK1 LBK1 (Figure (Figure5G,H). 5G,H).

Nutrients 2020, 12, 1166 8 of 15

Nutrients 2020, 12, x FOR PEER REVIEW 8 of 16

Figure 5. EFigureffects 5. of Effects LBE of and LBE RA and on RA AMPK on AMPK signaling signaling in HepG2 HepG2 cells cells without without PA. Activation PA. Activation of AMPK of AMPK and acetyl-CoA carboxylase (ACC) phosphorylation by (A) LBE (50 or 100 μg/mL) and (B) RA (20 or µ and acetyl-CoA40 μM), carboxylase and (C) 100 μg/mL (ACC) LBE phosphorylation for 0.5–3 hours and ( byD) (40A μ)M LBE RA (50in for or 0.5 100–3 h. gActivation/mL) and of liver (B) RA (20 or 40 µM), andkinase (C) 100B1 (LKB1µg/mL) and LBE Ca2+/calmodulin for 0.5–3 h-dependent and (D) protein 40 µM kinase RA in II for(CaMKII) 0.5–3 phosphorylation h. Activation by of ( liverE) kinase B1 (LKB1)LBE and and Ca (2F+) /RA.calmodulin-dependent With PA, (G) activation of protein AMPK and kinase ACC IIphosphorylation (CaMKII) phosphorylation, and (H) activation of by (E) LBE and (F) RA.LKB1 With and PA, CaMK2(G) phosphoryactivationlation ofAMPK by LBE (50 and or 100 ACC μg/mL phosphorylation,). β-Actin was used andas an(H) internalactivation control of LKB1 for western blotting. and CaMK2 phosphorylation by LBE (50 or 100 µg/mL). β-Actin was used as an internal control for western3.3. blotting. LBE and RA Reduce Liver Damage by Reducing Lipid Accumulation and Hepatic Fibrosis in MCD diet- Fed db/db Mice 3.3. LBE and RA Reduce Liver Damage by Reducing Lipid Accumulation and Hepatic Fibrosis in MCD Diet-Fed db /db Mice Consistent with previous reports [20,21], the MCD diet decreased body and liver weights in db/db mice because the MCD diet induced histopathological features of NASH (Figure6A). The mice fed the MCD diet had significantly elevated serum alanine transaminase (ALT) and aspartate transaminase (AST) levels but LBE and RA30 administration reduced them (Figure6B,C). H&E and Oil Red O staining showed that hepatic lipid accumulation and the hepatic TG content were significantly higher in the MCD diet-fed mice than in the control mice (Figure6D–F). Sirius Red staining and an increased level of Nutrients 2020, 12, 1166 9 of 15 hepatic hydroxyproline demonstrated the development of hepatic fibrosis. Consistently, hepatic fibrosis markers; α-smooth muscle actin (α-SMA) and collagen type I alpha 1 (COL1A1), were highly expressed in the mice fed the MCD diet (Figure6G–I). However, LBE and RA30 suppressed lipid accumulation, the TG content, and fibrosis in the liver of the db/db mice fed the MCD diet (Figure6D–I). However, RA10 showed no significant effect against liver damage induced by the MCD diet in db/db mice. Nutrients 2020, 12, x FOR PEER REVIEW 10 of 16

Figure 6. Effect of LBE and RA on liver damage in methionine- and choline- deficient (MCD) diet-fed Figure 6. Effect of LBE and RA on liver damage in methionine- and choline- deficient (MCD) diet-fed db/db mice. (A) Experimental design of the nonalcoholic steatohepatitis (NASH) animal study. (B) db/db mice.Serum (A alanine) Experimental transaminase design(ALT) and of (C the) aspartate nonalcoholic transaminase steatohepatitis (AST) levels. (D) (NASH)H&E staining animal of study. (B) Serumrepresentative alanine transaminase liver sections (ALT) (arrows and indicate (C) aspartate inflammatory transaminase cells) (magnification: (AST) levels.50×; scale ( Dbar:) H&E 200 staining of representativeμm). (E) Representative liver sections images (arrows of Oil indicate Red O-stained inflammatory liver sections cells) (50×; scale (magnification: bar: 100 μm). (F 50) The; scale bar: × 200 µm).liver (E) RepresentativeTG content. (G) Collagen images deposition, of Oil as Red demonstrated O-stained by Sirius liver Red sections staining (50 (50×;; scale scale bar: bar: 200 100 µm). μm). (H) The hydroxyproline content in liver sections. (I) mRNA expression levels of× fibrosis-related (F) The liver TG content. (G) Collagen deposition, as demonstrated by Sirius Red staining (50 ; genes, normalized to those of β-actin. Results are expressed as the mean ± SD (% control). # p < 0.05, ## × scale bar: 200 µm). (H) The hydroxyproline content in liver sections. (I) mRNA expression levels of fibrosis-related genes, normalized to those of β-actin. Results are expressed as the mean SD (% ± control). # p < 0.05, ## p < 0.01, and ### p < 0.001 compared with the control group; * p < 0.05 and ** p < 0.01 compared with the MCD diet-fed group (n = 5–6 per group). Nutrients 2020, 12, x FOR PEER REVIEW 11 of 16

Nutrientsp < 20200.01,, 12and, 1166 ### p < 0.001 compared with the control group; * p < 0.05 and ** p < 0.01 compared with10 of 15 the MCD diet-fed group (n = 5–6 per group).

3.4. LBE and RA30 R Reduceeduce NASH via R Regulationegulation of the AMPK Pathway,Pathway, Inflammation,Inflammation, and the NRF2NRF2 Pathway in MCD Diet-Fed db/db Mice Pathway in MCD Diet-Fed db/db Mice The protein and mRNA expression of lipogenesis genes; SREBP-1c and FAS, was reduced by The protein and mRNA expression of lipogenesis genes; SREBP-1c and FAS, was reduced by LBE and RAs (Figure7A,C). In contrast, the mRNA and protein expression of PPAR α and CPT-1L, LBE and RAs (Figure 7A,C). In contrast, the mRNA and protein expression of PPARα and CPT-1L, involved in lipolysis, was increased by LBE and RA (Figure7B,D). In particular, RA30 escalated the involved in lipolysis, was increased by LBE and RA (Figure 7B,D). In particular, RA30 escalated the mRNA expression of the PPARα, which indicated a stronger impact of RA on lipolysis compared mRNA expression of the PPARα, which indicated a stronger impact of RA on lipolysis compared with that of LBE. However, the effects of RA10 on markers of lipid metabolism, except SREBP-1c and with that of LBE. However, the effects of RA10 on markers of lipid metabolism, except SREBP-1c and PPARα, were not significant. PPARα, were not significant.

Figure 7. 7. EEﬀffectsects of of LBE LBE and and RA RA on the on expression the expression of genes of and genes proteins and proteins involved involved in fatty acid in regulation fatty acid inregulation MCD diet-fed in MCD db diet/db- mice.fed db/db Protein mice levels. Protein of (levelsA) SREBP-1c of (A) SREBP and FAS,-1c and and FAS, (B) PPARand (Bα) andPPARα CPT-1L. and TheCPT- densitometric1L. The densitometric analysis analysis of Western of Western blot of ( Cblot) SREBP-1c of (C) SREBP and FAS,-1c and and FAS, (D) and PPAR (Dα) andPPARα CPT-1L, and normalizedCPT-1L, normalized to those to of those glyceraldehyde of glyceraldehyde 3-phosphate 3-phosphate dehydrogenase dehydrogenase (GAPDH) (GAPDH) mRNA mRNA levels oflevels the genesof the encodinggenes encoding (E) SREBP-1c (E) SREBPand-1cFAS and, and FAS (F,) andPPAR (Fα) andPPARαCPT-1L and .CPTβ-Actin-1L. β was-Actin used was as anused internal as an controlinternalfor control qPCR for analysis. qPCR analysis Results. areResults expressed are expressed as the mean as the SDmean (% ± control). SD (% control).# p < 0.05, # p ##< 0.05,p < 0.01, ## p ± and< 0.01###, andp < ###0.001 p < compared0.001 compared with the with control the control group; group; * p < 0.05 * p < and 0.05 ** andp < 0.01** p < compared 0.01 compared with the with MCD the diet-fedMCD diet group-fed group (n = 5–6 (n per= 5– group).6 per group).

To investigate changes in the antioxidant defense response, we measured the mRNA and protein expression levels levels of of N NRF2RF2 and and SOD1. SOD1. The The mRNA mRNA expression expression of the of N theRF NRF2-encoding2-encoding gene genewas only was onlyincrea increasedsed by RA30, by RA30,whereas whereas the protein the proteinlevel was level elevated was elevatedby LBE (Figure by LBE 8A,B). (Figure The8 A,B).gene expression The gene expressionof the inflammatory of the inflammatory molecules; macrophage molecules; macrophage inflammatory inflammatory protein-1 alpha protein-1 (MIP-1α alpha) and (MIP-1 intercellularα) and intercellularadhesion molecule adhesion 1 molecule (ICAM-1), 1 ( wasICAM-1 examined), was examined to demonstrate to demonstrate the impacts the impacts of LBE of LBEand and RA RA on oninflammation inflammation in the in theNASH NASH model. model. The The expression expression of these of these genes genes was was escalated escalated by the by MCD the MCD diet dietbut butwas wasreduced reduced by LBE by LBE and and RA30 RA30 administration administration (Figure (Figure 8C).8C). The The relative relative phosphorylation phosphorylation of AMPKAMPK was enhanced enhanced by by both both LBE LBEand RA and (Figure RA 8 (FigureD,E). However, 8D,E). the However, e ffect of the RA30 effect on the of phosphorylation RA30 on the

Nutrients 2020, 12, 1166 11 of 15 Nutrients 2020, 12, x FOR PEER REVIEW 12 of 16 of AMPK was much stronger than that of LBE, indicating that the major role of RA in the NASH model phosphorylation of AMPK was much stronger than that of LBE, indicating that the major role of RA might be a modiﬁcation of lipid metabolism. in the NASH model might be a modification of lipid metabolism.

FigureFigure 8. 8.E Effectsffects of of LBE LBE and and RA RA on on the the expression expression of of oxidative oxidative stress-regulating stress-regulating genes genes in MCD in MCD diet-fed diet- dbfed/db db/db mice. mice (A). ( HepaticA) Hepatic protein, protein, (B) ( theB) the densitometric densitometric analysis analysis of of WB, WB, and and (C (C) mRNA) mRNA expression expression ofof NRF2 NRF2 and and SOD1. SOD1. ((DD)) HepaticHepatic mRNA mRNA expression expression of of macrophagemacrophage inflammatory inflammatory protein-1 protein-1 alpha alpha (MIP-1(MIP-1αɑ)) and and intercellularintercellular adhesion adhesion molecule molecule 1 1 ( ICAM-1(ICAM-1).). ((EE)) PhosphorylationPhosphorylation ofof AMPK AMPK and and ( F(F)) the the densitometricdensitometric analysis analysis of of WB WB in MCD in MCD diet-fed diet- micefed mice treated treated with LBE with or LBE RA. or GAPDH RA. GAPDH was asan was internal as an controlinternal for control WB analysis for WB and analysisβ-Actin and was β- usedActin for was mRNA used expression.for mRNA Resultsexpression are. expressed Results are as expressed the mean asSD the (% mean control). ± SD *(%p < control).0.05 compared * p < 0.05 with compared the control with group; the control# p < group;0.05, ## #p p< < 0.01,0.05, and## p <### 0.01p <, and0.001 ### ± comparedp < 0.001 compared with the control with the group; control * p < group;0.05 compared * p < 0.05 with compared the MCD with diet-fed the MCD group diet (n-=fed5–6 group per group). (n = 5– 6 per group). 4. Discussion 4. DiscussionNASH, acomplex disease involved in NAFLD, is characterized by excess accumulation of lipids in hepatocytes,NASH, a complex along with disease hepatic involved inflammation in NAFLD, [1]. is In characterized the present study,by excess LBE accumulation and RA reduced of lipids the serumin hepatocytes, markers of along liver with damage hepatic and inflammation suppressed the [1] lipid. In the accumulation present study, as well LBE as and fibrosis RA reduced in the liver, the whichserum indicate markers amelioration of liver damage of liver and damage.suppressed These the datalipid suggestaccumulation that LBE as well and RAas fibrosis can modulate in the liver, the processeswhich indicate and symptoms amelioration associated of liver with damage. NASH. These data suggest that LBE and RA can modulate the proceLBEsses andand symptoms RA elevated associated the levels with ofNASH. LKB1, a kinase phosphorylating AMPK at Thr172. As a result,LBE and LBE RA and elevated RA increased the levels the of phosphorylation LKB1, a kinase phosphorylating of AMPK, leading AMPK to a decrease at Thr172. in As SREBP-1c. a result, BecauseLBE and SREBP-1c RA increased is a transcription the phosphorylation factor of FAS of AMPK, and SCD1, leading which to area decrease enzymes in of SREBP fat formation-1c. Because [22], theSREBP effects-1c ofis LBEa transcription and RA on f theactor activation of FAS and of AMPK SCD1, resultedwhich are in enzymes the reduction of fat of formation fat formation [22], viathe effects of LBE and RA on the activation of AMPK resulted in the reduction of fat formation via decreases in FAS and SCD1 levels. Furthermore, the activation of AMPK by LBE and RA increased

Nutrients 2020, 12, 1166 12 of 15 decreases in FAS and SCD1 levels. Furthermore, the activation of AMPK by LBE and RA increased PGC-1α and PPARα, resulting in an increase in CPT-1L [5,23]. Considering that these markers are involved in fatty acid oxidation, the data indicate that LBE and RA may induce fatty acid oxidation. Taken together, LBE and RA suppressed the lipid accumulation in the liver by inhibiting fatty acid synthesis and enhancing fatty acid oxidation, thus leading to the suppression of liver damage. Excess fatty acids in NAFLD elevate the levels of free fatty acids in hepatocytes, which causes the generation of lipotoxic lipids [3]. Lipotoxic lipids lead to hepatocyte injuries and inflammation, due to increased oxidative stress and stimulation of immune cells in the liver. Therefore, controlling oxidative stress and inflammation is a critical target in the treatment of NASH. NRF2 is a representative detoxifying enzyme, which acts as a transcription factor for antioxidant enzymes, such as SOD1 and heme oxygenase 1 (HO-1) [24]. The increases in NRF2 levels, caused by LBE and RA treatments, indicate their ability to enhance antioxidant mechanisms in the NASH model. Liver injuries by reactive oxygen species (ROS) and lipotoxic lipids result in the immune responses through hepatic macrophages, leading to the activation of hepatic stellate cells [25,26]. These activated hepatic stellate cells recruit circulating immune cells and are differentiated to myofibroblasts, which ultimately results in liver fibrosis. Suppression of inflammatory molecules, MIP-1α and ICAM-1 by LBE and RA indicates less activated hepatic stellate cells, following less α-SMA and COL1A1, the markers of liver fibrosis. Taken together, LBE and RA can interrupt the progression from fatty liver to NASH through their antioxidant and anti-inflammatory activities. These results confirmed previous studies of RA, in which RA had an anti-fibrotic effect under carbon tetrachloride-induced liver damage and an anti-inflammatory activity on oleic acid-induced NAFLD [27,28]. Although more RA was extracted from lemon balm at higher concentrations of EtOH, we found that the activity of lemon balm extract obtained with 20% EtOH at 200 mg/kg was similar to that of RA at 30 mg/kg in the NASH model, although the concentration of RA in LBE was approximately 4–5% (equivalent to an RA dose of ~10 mg/kg). These data indicate that other small molecules present in LBE contribute to its activity against NASH, even though their content in LBE is very low compared to that of RA. The presence of molecules such as caffeic acid, hesperidin, luteolin, etc., has been reported in LBEs [29]. A number of studies have demonstrated beneficial effects of these compounds; hesperidin and luteolin reduced hepatic steatosis, and epicatechin has antioxidant, anti-inflammatory, and anti-obesity effects [30–33]. Lithospermic acid, the secondary major compound of LBE has been reported to exhibit an antioxidant activity against oxidative stress [34], but shows no significant effects on lipid accumulation and AMPK activation. Because its concentration was independent of the concentration of water or EtOH, it is unlikely to be a critical factor in the strong effect of LBE against NASH. Therefore, the effect of LBE may be due to synergistic effects of a number of small molecules present in lemon balm, not just one compound.

5. Conclusions We found that LBE and RA exhibited suppressive activities on NASH. They can modulate lipid metabolism via AMPK activation and suppress inﬂammation via changes in NRF2 signaling. Importantly, the extract of lemon balm obtained with 20% EtOH showed eﬀectiveness similar to that of RA at high concentrations. Therefore, considering the cost of developing and making drugs, LBE may be a good candidate for the treatment and prevention of NASH.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/12/4/1166/s1, Table S1: Antibodies used for western blotting, Table S2: Primers used for real-time qRT-PCR. Author Contributions: M.K., S.M.K., and C.W.N. planned and designed the study. M.K., G.Y., A.R., and Y.-J.S. carried out the experiments. C.R.H. analyzed chemical proﬁling of plant extracts. M.K. and G.Y. drafted the manuscript. All authors have read and approved the ﬁnal manuscript prior to submission. Nutrients 2020, 12, 1166 13 of 15

Funding: This work was supported by the National Research Council of Science and Technology (NST) funded by the Korean Government (Ministry of Science and ICT) (grant No. CRC-15-01-KIST) and Korea Institute of Science and Technology internal grant (2Z06110). Conﬂicts of Interest: The authors declare no conﬂict of interest.

Abbreviations

ACC acetyl-CoA carboxylase ALT alanine aminotransferase AMPK AMP-activated protein kinase AST aspartate aminotransferase CAMKII Ca2+/calmodulin-dependent protein kinase II CoA coenzyme A COL1A1 collagen type I alpha 1 CPT-1 carnitine palmitoyl transferase I CRTC CREB-regulated transcription coactivator 1 FAS fatty acid synthase HO-1 heme oxygenase-1 LBE lemon balm extract extracted by 20% ethanol LKB-1 liver kinase B1 MCD diet methionine- and choline-deﬁcient diet NAFLD nonalcoholic fatty liver disease NASH nonalcoholic steatohepatitis NRF2 nuclear factor erythroid-derived 2-related factor 2 PA palmitic acid PGC-1 peroxisome proliferator-activated receptor γ coactivator 1 PPARα peroxisome proliferator-activated receptor α RA rosmarinic acid SCD-1 stearoyl-CoA desaturase-1 α-SMA α -smooth muscle actin SOD1 superoxide dismutase 1 SREBP-1 sterol regulatory element-binding protein-1 TG triglyceride

References

1. Anderson, N.; Borlak, J. Molecular mechanisms and therapeutic targets in steatosis and steatohepatitis. Pharmacol. Rev. 2008, 60, 311–357. [CrossRef][PubMed] 2. Day, C.P.; James, O.F. Steatohepatitis: A tale of two “hits”? Gastroenterology 1998, 114, 842–845. [CrossRef] 3. Brunt, E.M.; Wong, V.W.; Nobili, V.; Day, C.P.; Sookoian, S.; Maher, J.J.; Bugianesi, E.; Sirlin, C.B.; Neuschwander-Tetri, B.A.; Rinella, M.E. Nonalcoholic fatty liver disease. Nat. Rev. Dis. Primers 2015, 1, 15080. [CrossRef][PubMed] 4. Sahebkar, A.; Chew, G.T.; Watts, G.F. New peroxisome proliferator-activated receptor agonists: Potential treatments for atherogenic dyslipidemia and non-alcoholic fatty liver disease. Expert Opin. Pharmacother. 2014, 15, 493–503. [CrossRef][PubMed] 5. Herzig, S.; Shaw, R.J. AMPK: Guardian of metabolism and mitochondrial homeostasis. Nat. Rev. Mol. Cell Biol. 2018, 19, 121–135. [CrossRef] 6. Weng, S.Y.; Schuppan, D. AMPK regulates macrophage polarization in adipose tissue inﬂammation and NASH. J. Hepatol. 2013, 58, 619–621. [CrossRef] 7. Joo, M.S.; Kim, W.D.; Lee, K.Y.; Kim, J.H.; Koo, J.H.; Kim, S.G. AMPK Facilitates Nuclear Accumulation of Nrf2 by Phosphorylating at Serine 550. Mol. Cell. Biol. 2016, 36, 1931–1942. [CrossRef] 8. Salminen, A.; Hyttinen, J.M.; Kaarniranta, K. AMP-activated protein kinase inhibits NF-kappaB signaling and inﬂammation: Impact on healthspan and lifespan. J. Mol. Med. 2011, 89, 667–676. [CrossRef] Nutrients 2020, 12, 1166 14 of 15

9. Kennedy, D.O.; Scholey, A.B.; Tildesley, N.T.; Perry, E.K.; Wesnes, K.A. Modulation of mood and cognitive performance following acute administration of Melissa officinalis (lemon balm). Pharmacol. Biochem. Behav. 2002, 72, 953–964. [CrossRef] 10. Scholey, A.; Gibbs, A.; Neale, C.; Perry, N.; Ossoukhova, A.; Bilog, V.; Kras, M.; Scholz, C.; Sass, M.; Buchwald-Werner, S. Anti-stress effects of lemon balm-containing foods. Nutrients 2014, 6, 4805–4821. [CrossRef] 11. Akhondzadeh, S.; Noroozian, M.; Mohammadi, M.; Ohadinia, S.; Jamshidi, A.H.; Khani, M. Melissa officinalis extract in the treatment of patients with mild to moderate Alzheimer’s disease: A double blind, randomised, placebo controlled trial. J. Neurol. Neurosurg. Psychiatry 2003, 74, 863–866. [CrossRef][PubMed] 12. Jeung, I.C.; Jee, D.; Rho, C.R.; Kang, S. Melissa officinalis L. Extracts Protect Human Retinal Pigment Epithelial Cells against Oxidative Stress-Induced Apoptosis. Int. J. Med. Sci. 2016, 13, 139–146. [CrossRef][PubMed] 13. Kim, J.; Lee, H.; Lim, J.; Oh, J.; Shin, S.S.; Yoon, M. The Angiogenesis Inhibitor ALS-L1023 from Lemon-Balm Leaves Attenuates High-Fat Diet-Induced Nonalcoholic Fatty Liver Disease through Regulating the Visceral Adipose-Tissue Function. Int. J. Mol. Sci. 2017, 18, 846. [CrossRef][PubMed] 14. Kim, J.; Lee, H.; Lim, J.; Lee, H.; Yoon, S.; Shin, S.S.; Yoon, M. The lemon balm extract ALS-L1023 inhibits obesity and nonalcoholic fatty liver disease in female ovariectomized mice. Food Chem. Toxicol. 2017, 106, 292–305. [CrossRef] 15. Al-Dhabi, N.A.; Arasu, M.V.; Park, C.H.; Park, S.U. Recent studies on rosmarinic acid and its biological and pharmacological activities. EXCLI J. 2014, 13, 1192–1195. 16. Govindaraj, J.; Sorimuthu Pillai, S. Rosmarinic acid modulates the antioxidant status and protects pancreatic tissues from glucolipotoxicity mediated oxidative stress in high-fat diet: Streptozotocin-induced diabetic rats. Mol. Cell. Biochem. 2015, 404, 143–159. [CrossRef] 17. Wang, S.J.; Chen, Q.; Liu, M.Y.; Yu, H.Y.; Xu, J.Q.; Wu, J.Q.; Zhang, Y.; Wang, T. Regulation effects of rosemary (Rosmarinus officinalis Linn.) on hepatic lipid metabolism in OA induced NAFLD rats. Food Funct. 2019, 10, 7356–7365. [CrossRef] 18. Noguchi-Shinohara, M.; Ono, K.; Hamaguchi, T.; Iwasa, K.; Nagai, T.; Kobayashi, S.; Nakamura, H.; Yamada, M. Pharmacokinetics, Safety and Tolerability of Melissa officinalis Extract which Contained Rosmarinic Acid in Healthy Individuals: A Randomized Controlled Trial. PLoS ONE 2015, 10, e0126422. [CrossRef] 19. Yu, J.; Zhang, S.; Chu, E.S.; Go, M.Y.; Lau, R.H.; Zhao, J.; Wu, C.W.; Tong, L.; Zhao, J.; Poon, T.C.; et al. Peroxisome proliferator-activated receptors gamma reverses hepatic nutritional fibrosis in mice and suppresses activation of hepatic stellate cells in vitro. Int. J. Biochem. Cell Biol. 2010, 42, 948–957. [CrossRef] 20. Ip, E.; Farrell, G.C.; Robertson, G.; Hall, P.; Kirsch, R.; Leclercq, I. Central role of PPARalpha-dependent hepatic lipid turnover in dietary steatohepatitis in mice. Hepatology 2003, 38, 123–132. [CrossRef] 21. Leclercq, I.A.; Farrell, G.C.; Field, J.; Bell, D.R.; Gonzalez, F.J.; Robertson, G.R. CYP2E1 and CYP4A as microsomal catalysts of lipid peroxides in murine nonalcoholic steatohepatitis. J. Clin. Investig. 2000, 105, 1067–1075. [CrossRef][PubMed] 22. Horton, J.D.; Goldstein, J.L.; Brown, M.S. SREBPs: Activators of the complete program of cholesterol and fatty acid synthesis in the liver. J. Clin. Investig. 2002, 109, 1125–1131. [CrossRef][PubMed] 23. Song, S.; Attia, R.R.; Connaughton, S.; Niesen, M.I.; Ness, G.C.; Elam, M.B.; Hori, R.T.; Cook, G.A.; Park, E.A. Peroxisome proliferator activated receptor alpha (PPARalpha) and PPAR gamma coactivator (PGC-1alpha) induce carnitine palmitoyltransferase IA (CPT-1A) via independent gene elements. Mol. Cell. Endocrinol. 2010, 325, 54–63. [CrossRef][PubMed] 24. Lee, L.Y.; Kohler, U.A.; Zhang, L.; Roenneburg, D.; Werner, S.; Johnson, J.A.; Foley, D.P. Activation of the Nrf2-ARE pathway in hepatocytes protects against steatosis in nutritionally induced non-alcoholic steatohepatitis in mice. Toxicol. Sci. 2014, 142, 361–374. [CrossRef][PubMed] 25. Koyama, Y.; Brenner, D.A. Liver inflammation and fibrosis. J. Clin. Investig. 2017, 127, 55–64. [CrossRef] 26. Zhang, C.Y.; Yuan, W.G.; He, P.; Lei, J.H.; Wang, C.X. Liver fibrosis and hepatic stellate cells: Etiology, pathological hallmarks and therapeutic targets. World J. Gastroenterol. 2016, 22, 10512–10522. [CrossRef] 27. Li, G.S.; Jiang, W.L.; Tian, J.W.; Qu, G.W.; Zhu, H.B.; Fu, F.H. In vitro and in vivo antifibrotic effects of rosmarinic acid on experimental liver fibrosis. Phytomedicine 2010, 17, 282–288. [CrossRef] 28. Balachander, G.J.; Subramanian, S.; Ilango, K. Rosmarinic acid attenuates hepatic steatosis by modulating ER stress and autophagy in oleic acid-induced HepG2 cells. RSC Adv. 2018, 8, 26656–26663. [CrossRef] Nutrients 2020, 12, 1166 15 of 15

29. Dastmalchi, K.; Dorman, H.T.D.; Oinonen, P.P.; Darwis, Y.; Laakso, I.; Hiltunen, R. Chemical composition and in vitro antioxidative activity of a lemon balm (Melissa officinalis L.) extract. LWT Food Sci. Technol. 2008, 41, 391–400. [CrossRef] 30. Wang, X.; Hasegawa, J.; Kitamura, Y.; Wang, Z.; Matsuda, A.; Shinoda, W.; Miura, N.; Kimura, K. Effects of hesperidin on the progression of hypercholesterolemia and fatty liver induced by high-cholesterol diet in rats. J. Pharmacol. Sci. 2011, 117, 129–138. [CrossRef] 31. Liu, J.F.; Ma, Y.; Wang, Y.; Du, Z.Y.; Shen, J.K.; Peng, H.L. Reduction of lipid accumulation in HepG2 cells by luteolin is associated with activation of AMPK and mitigation of oxidative stress. Phytother. Res. 2011, 25, 588–596. [CrossRef][PubMed] 32. Friedrich, M.; Petzke, K.J.; Raederstorff, D.; Wolfram, S.; Klaus, S. Acute effects of epigallocatechin gallate from green tea on oxidation and tissue incorporation of dietary lipids in mice fed a high-fat diet. Int. J. Obes. 2012, 36, 735–743. [CrossRef][PubMed] 33. Kuzu, N.; Bahcecioglu, I.H.; Dagli, A.F.; Ozercan, I.H.; Ustundag, B.; Sahin, K. Epigallocatechin gallate attenuates experimental non-alcoholic steatohepatitis induced by high fat diet. J. Gastroenterol. Hepatol. 2008, 23, e465–e470. [CrossRef][PubMed] 34. Chan, K.W.; Ho, W.S. Anti-oxidative and hepatoprotective effects of lithospermic acid against carbon tetrachloride-induced liver oxidative damage in vitro and in vivo. Oncol. Rep. 2015, 34, 673–680. [CrossRef] [PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).