FNDC5 Alleviates Hepatosteatosis by Restoring AMPK/Mtor-Mediated

Page 1 of 46 Diabetes

Title Page

FNDC5 alleviates hepatosteatosis by restoring AMPK/mTOR-mediated

autophagy, fatty acid oxidation and lipogenesis in mice

TongYan Liu1, XiaoQing Xiong1, XingSheng Ren1, MingXia Zhao1, ChangXiang

Shi1, JueJin Wang1, YeBo Zhou1, Feng Zhang1, Ying Han1, XingYa Gao1, Qi Chen2,

YueHua Li2,YuMing Kang3, GuoQing Zhu1,2*

1Key Laboratory of Cardiovascular Disease and Molecular Intervention, Department

of Physiology, Nanjing Medical University, Nanjing, Jiangsu 210029, China;

2Department of Pathophysiology, Nanjing Medical University, Nanjing, Jiangsu

210029, China; 3Department of Physiology and Pathophysiology, Cardiovascular

Research Center, Xi'an Jiaotong University School of Medicine, Xi'an 710061, China

Short running title: FNDC5 attenuates hepatic steatosis

Word count: 3955 excluding title page, abstract, references, figure legends.

Number of figures and tables: 8 figures and 0 table

Online supplemental Data: 11 embedded figures and 1 table

*Address for correspondence:

GuoQing Zhu, M.D., Ph.D. Professor, Chair

Key Laboratory of Cardiovascular Disease and Molecular Intervention,

Department of Physiology, Nanjing Medical University, 140 Hanzhong Road,

Nanjing 210029, China

Tel: +862586862885; Fax: +862586862885; EMail: [email protected]

Diabetes Publish Ahead of Print, published online August 8, 2016 Diabetes Page 2 of 46

ABSTRACT

Fibronectin type III domaincontaining 5 (FNDC5) protein induces browning of subcutaneous fat, and mediates beneficial effects of exercise on metabolism. However, whether FNDC5 is associated with hepatic steatosis, autophagy, fatty acid oxidation

(FAO) and lipogenesis remains unknown. Herein, we show the roles and mechanisms of FNDC5 in hepatic steatosis, autophagy and lipid metabolism. Fasted FNDC5/ mice exhibited severe steatosis, reduced autophagy and FAO, and enhanced lipogenesis in liver compared with WT mice. Energy deprivation induced autophagy,

FAO and AMPK activity were attenuated in FNDC5/ hepatocytes, which were restored by activating AMPK with AICAR. Inhibition of mTORC1 with rapamycin enhanced autophagy and FAO, attenuated lipogenesis and steatosis in FNDC5/ livers.

FNDC5 deficiency exacerbated hyperlipemia, hepatic FAO and autophagy impairment, hepatic lipogenesis and lipid accumulation in obese mice. Exogenous

FNDC5 stimulated autophagy and FAO gene expression in hepatocytes, and repaired the attenuated autophagy and palmitateinduced steatosis in FNDC5/ hepatocytes.

FNDC5 overexpression prevented hyperlipemia, hepatic FAO and autophagy impairment, hepatic lipogenesis and lipid accumulation in obese mice. These results indicate that FNDC5 deficiency impairs autophagy and FAO, and enhances lipogenesis via AMPK/mTOR pathway. FNDC5 deficiency aggravates while FNDC5 overexpression prevents the HFDinduced hyperlipemia, hepatic lipid accumulation, and impaired FAO and autophagy in liver.

Keywords: hepatic steatosis; hepatocyte; lipid mentalism; obesity; signal transduction

Page 3 of 46 Diabetes

Nonalcoholic fatty liver disease (NAFLD) is characterized by triacylglycerol (TG)

accumulation within hepatocytes (1). Hepatosteatosis is strongly associated with

obesity, and may progress to steatohepatitis and even to endstage liver disease

including liver cirrhosis and hepatocellular carcinoma (2,3). Fatty acid βoxidation

(FAO) in mitochondria is a process to shorten the fatty acids into acetylCoA, which

can be converted into ketone bodies or incorporated into tricarboxylic acid cycle for

full oxidation (4). Accumulation of lipid in liver can be traced by the impaired FAO

and increased de novo lipogenesis (5).

Autophagy is a mechanism involved in cellular homeostasis delivering

cytoplasmic content to the lysosomes for degradation to macronutrients (6). Defects in

autophagy play a major role in metabolic dysregulation (7). Although some studies

showed the lipogenic role of autophagy, most experiments supported autophagy as a

lipolytic mechanism (6). Reduced autophagic function promotes the initial

development of hepatic steatosis and progression of steatosis to liver injury, and

agents to augment hepatic autophagy may have therapeutic potential in nonalcoholic

steatohepatitis (810).

Fibronectin type III domain containing 5 (FNDC5) is a type I membrane protein

that has 209 amino acid residues. FNDC5 induces browning of subcutaneous

adipocytes, and mediates the beneficial effect of exercise on metabolism (11). Irisin, a

cleaved and secreted fragment of FNDC5, acts on white adipose cells to induce a

broad program of brownfatlike development (11). Our recent studies have shown

that FNDC5 overexpression ameliorates hyperlipemia and enhances lipolysis in

Diabetes Page 4 of 46

adipose tissues in obese mice (12), and that irisin inhibits hepatic gluconeogenesis and increases glycogen synthesis in type 2 diabetic mice and hepatocytes (13). However, whether FNDC5 could improve hepatosteatosis, autophagy and FAO remains unknown. It is known that nutrient deprivation activates adenosine monophosphateactivated protein kinase (AMPK), resulting in the inhibition of mammalian target of rapamycin complex 1 (mTORC1), which regulates lipid metabolism, cellular proliferation, and autophagy (14,15). The mTORC1 inhibits peroxisome proliferatoractivated receptor (PPAR)α activity, which regulates mitochondrial functions and FAO (16). Interestingly, PPARα acts as the downstream of FNDC5 (11). The present study is designed to investigate the roles and underlying mechanisms of FNDC5 in hepatic steatosis, autophagy, FAO and lipogenesis in

FNDC5/ mice, highfat diet (HFD)induced obese mice, and primary hepatocytes.

Moreover, the therapeutic effects of FNDC5 were investigated.

RESEARCH DESIGN AND METHODS

FNDC5-/- mice and HFD-induced obese mice

Male C57BL/6 WT and FNDC5/ mice on a C57BL/6 background (Nanjing

BioMedical Research Institute, Nanjing University, Nanjing, China) were used in the experiments. In HFDinduced obesity models, mice at the age of 4 weeks began to receive HFD (21.8 kJ/g, 60% of energy as fat) for 12 weeks. Normal chow diet (14.7 kJ/g, 13% of energy as fat) was used as control (12,17). Procedures were approved by the Experimental Animal Care and Use Committee of Nanjing Medical University and conformed to the Guide for the Care and Use of Laboratory Animal (NIH

Page 5 of 46 Diabetes

publication, 8th edition, 2011). Mice were caged in an environment under controlled

temperature and humidity with free access to water and food under a 12h light/dark

cycle. At the end of experiments, mice were fasted overnight, and then euthanized

with an overdose of pentobarbital sodium (150 mg/kg, i.v.).

FNDC5 overexpression in mice

Mice at the age of 4 weeks began to receive control diet or HFD for 12 weeks. A

single intravenous injection of recombinant lentivirus (1×108 TU/ml, 100 l)

expressing FNDC5 or EGFP vector was carried out at the end of the 6th week after

the diet application (12). Acute experiments were performed 6 weeks after the

lentivirus introduction.

Knockdown of AMPK or Atg5 by siRNA in hepatocytes

Primary hepatocytes were transfected with small interfering RNA (siRNA) for

knockdown of AMPK or autophagy protein 5 (Atg5). Scramble siRNA was used as

control. The sequences of siRNA were listed as follows. AMPK: sense

CGGGAUCCAUCAGCAACUATT, antisense UAGUUGCUGAUGGAUCCCGAT

(18). Atg5:

CCGGCCTTGGAACATCACAGTACATCTCGAGATGTACTGTGATGTTCCAAG

GTTTTTG.

Primary hepatocyte isolation and cell culture

Primary hepatocytes were isolated and cultured as previously described (13,19).

Briefly, mice were anesthetized with pentobarbital (50 mg/kg, i.p.). HEPES buffer

containing collagenase II (0.66 mg/mL) was perfusion via portal vein. Livers were

Diabetes Page 6 of 46

removed and excised aseptically. Cells were dispersed and filtrated. Hepatocyte suspensions were purified by centrifugation in Percoll adjusted to a density of 1.065 g/ml for 10 min at 50 g to reduce the amount of nonparenchymal cells. With this method the nonparenchymal cells is less than 1% (20). Cell viability was determined with trypan blue dye. Plates with cell viability greater than 95% were used for experiments. The hepatocytes were maintained in low glucose DMEM containing 10%

FBS with penicillin (100 units/mL) and streptomycin (100 g/mL) at 37°C in a 5%

CO2 atmosphere.

Monitor of autophagy

Cells were transfected with tandem green fluorescent protein (GFP)red fluorescent protein (RFP)LC3 adenovirus (Hanbio, Shanghai, China) for 24 h according to the instructions. Cells were treated with amino acid starvation, rapamycin or chloroquine for 2 h to observe the autophagy flux. When autophagy inducts, both GFP and RFP are expressed as yellow dots representing autophagosomes after the images emerged.

When autophagosomes fuse with lysosomes and form autolysosomes, the GFP degrades in an acid environment, but RFP–LC3 maintains showing as red dots (21).

Oil red O staining and immunohistochemistry

Livers were fixed in 4% neutral buffered formalin phosphate and then were embedded in paraffin or OCT compound, respectively. The tissues were subsequently sliced into

5m sections. Oil Red O staining was used to detect the lipid content in liver. For immunohistochemistry evaluation, liver sections were incubated with antip62 antibody (Abcam Ltd, Cambridge, UK) or antiLC3B antibody (Cell signaling

Page 7 of 46 Diabetes

Technology, Danvers, MA, USA). The antiLC3B antibody showed stronger reactivity

with LC3BII according to the manufacturer's instructions.

Western blot

Protein extracts were electrophoresed, blotted, and then incubated with antibodies

against FNDC5, S6, PS6, AMPKα, PAMPKα ,LC3B, Raptor, PRaptor,ULK1,

PULK1, GAPDH (Cell signaling Technology, Danvers, MA, USA) and P62 (Abcam

Ltd, Cambridge, UK) with appropriate secondary HRPconjugated antibodies, and

then developed.

Quantitative real-time PCR analysis

RNA extracted from liver or hepatocyte was subjected to reverse transcription and

quantitative realtime PCR was performed using a StepOnePlus RealTime PCR

System (Applied Biosystems, Foster City, CA, USA). All genes expression levels

were normalized by GAPDH levels. The sequences of primers were listed in a table

(Supplementary Table 1).

Measurement of lipids and markers of hepatocyte injury

Enzymatic methods were used to evaluate the levels of nonestesterified fatty acid

(NEFA), TG and cholesterol (CHO), and the activity of alanine aminotransferase

(ALT) and aspartate aminotransferase (AST) in serum or liver. The kits for serum

NEFA serum were bought from Wako Pure chemical industries Ltd. (Osaka, Japan);

for serum TG, AST, ALT, and hepatic NEFA from Jiancheng Bioengineering Institute

(Nanjing, China); for hepatic TG and CHO from Applygen Technologies Inc (Beijing,

China).

Diabetes Page 8 of 46

Assessment of FAO rate in vitro

Hepatocytes isolated from WT and FNDC5/ mice were incubated with 0.05 mM of carnitine and 0.25 Ci of 14Cpalmitate (GE Healthcare Life Sciences; Pittsburgh, PA) for 24 hours. 100 L of the medium was used for measuring acidsoluble metabolites with scintillation counter. And, 800 L of medium was harvested on ice and mixed with icecold perchloric acid (70%, 200 L) to precipitate BSAfatty acid complexes.

The samples were centrifuged for 10 min at 14,000 g, and the radioactivity of the

14 supernatant was evaluated by liquid scintillation as captured CO2 (18).

Chemicals

FNDC5, LPS, palmitate, WY14643, carnitine and rapamycin were bought from

Sigma Inc. (St Louis, MO, USA); AICAR was bought from Beyotime Biotechnology

Inc. (Shanghai, China).

Statistics

Data are presented as mean±SEM, and a value of P<0.05 was considered statistically significant. Twotailed, unpaired Student’s ttests were used to compare two treatment groups. Oneway and twoway ANOVA were used for data analysis of more than two groups followed by Bonferroni’s post hoc analysis.

RESULTS

Fasting causes severe lipid accumulation in FNDC5-deleted mice

Fasting causes lipid mobilization from peripheral depots into liver (22). To determine the role of FNDC5 in hepatic lipid accumulation, the responses of lipids to fasting were compared between FNDC5/ mice and WT mice. Lipid accumulation in livers

Page 9 of 46 Diabetes

was increased in FNDC5/ mice, and became severe after fasting for 16 h compared

with WT mice (Fig. 1AB). Fasting caused more increases in serum NEFA and TG

levels in FNDC5/ mice than WT mice (Fig. 1C). The efficiency of FNDC5 gene

knockout was confirmed by serum FNDC5 levels and liver FNDC5 expressions (Fig.

1DF). These results indicate that FNDC5 prevents excessive lipid accumulation in

livers.

FNDC5 deficiency causes defects in AMPK/PPARααα-mediated FAO

Reduced FAO increases hepatic lipid accumulation, and chronic starvation increases

FAO gene expressions via transcriptional mechanisms (5). Activation of AMPK

stimulates FAO via PPARα signaling (18). Thus, the roles of FNDC5 in regulating

FAO and its downstream pathway were investigated. Basal and fastinginduced FAO

gene expressions (Pparα, Hmgcs2, Cpt1, Acox1, Sirt3 and Cyp4a10) (Fig. 2A) and

AMPK phosphorylation (Supplementary Fig. 1) were reduced in FNDC5/ mice liver.

Activation of AMPK with AICAR increased FAO gene expressions (Fig. 2B) and

reduced TG levels in FNDC5/ mice liver (Fig. 2D). Furthermore, AICAR stimulated

liver AMPK activation and the following mTORC1 inhibition in WT and FNDC5/

mice (Supplementary Fig. 2). Knockdown of AMPK with siRNA increased TG

contents in both WT and FNDC5/ hepatocytes (Fig. 2E). WY14643, a PPARα

agonist, caused a greater increase in the expressions of PPARα target genes (Hmgcs2,

Cpt1, Acox1, Ehhadh, Acsl1, Peci, Cyp4a10 and Cyp4a12) in WT hepatocytes than

those in FNDC5–/– hepatocytes (Fig. 2C). WY14643 inhibited palmitateinduced lipid

accumulation in both WT and FNDC5–/– hepatocytes (Fig. 2F and Supplementary Fig.

Diabetes Page 10 of 46

3). It induced a greater increase in FAO rate in WT hepatocytes than those in FNDC5–

/– hepatocytes (Fig. 2F). These findings indicate that FAO is reduced in FNDC5/ mice liver, and FNDC5 is important for fastinginduced FAO gene expressions, which are mediated by AMPK pathway. FNDC5 deficiency causes impairment in

AMPK/PPARαmediated FAO.

Rapamycin prevents the impaired FAO and increased lipogenesis in FNDC5 deficiency

Since nutrient deprivationinduced AMPK represses mTORC1 activity (16), we further investigated whether mTORC1 is involved in the effects of FNDC5 on FAO.

FNDC5 deficiency caused an increase in liver ribosomal S6 protein phosphorylation, a marker of mTORC1 activity (23), which was suppressed by rapamycin, a mTORC1 inhibitor (Fig. 3A). It reduced the increased liver TG contents in both fed and fasted states in FNDC5–/– mice (Fig. 3D). Rapamycin restored the reduced FAO gene expressions (Pparα, Hmgcs2, Cpt1, Sirt3) in FNDC5–/– mice under both fed and fasting conditions (Fig. 3B). The results indicate that increased mTORC1 activity contributes to the reduced FAO in mice with FNDC5 deficiency. Lipogenesis is a factor contributing hepatic lipid accumulation (24). The mRNA levels of lipogenic genes (Srebp1c, Dgat1, Fasn and Scd1) were raised in FNDC5–/– liver under both fed and fasted conditions, which were attenuated by rapamycin (Fig. 3C). These results indicate that FNDC5 deficiency potentiates lipogenesis via increased mTORC1 activity.

FNDC5 deficiency causes attenuation in autophagy and AMPK activity

Page 11 of 46 Diabetes

Reduced p62 expression and increased LC3BII generation from LC3BI are

markers of autophagy (25). The mRNA levels of autophagy genes including

UNC51like kinase 1 (Ulk1), Ulk2, autophagyrelated protein 5 (Atg5), Atg8 and

Atg10 were downregulated in FNDC5/ mouse liver (Fig. 4A). Fasting reduced p62

expression and increased LC3BII/LC3BI in WT liver, but not in FNDC5/ liver (Fig.

4B), which were confirmed by the liver immunohistochemistry (Supplementary Fig.

4). Moreover, amino acid deprivation caused a greater enhancement in autophagy flux

in WT hepatocyte than that in FNDC5/ hepatocytes (Fig. 4C). These results indicate

that autophagy in liver is reduced in FNDC5/ mice, and FNDC5 is required for

fastinginduced autophagy response.

Nutrient deprivation activates AMPK, which phosphorylates raptor (an essential

component of mTORC1), resulting in net repression of mTORC1 signaling (26,27).

AMPK stimulates autophagy through direct ULK1 phosphorylation (28). Metformin

is a biguanide antihyperglycemic agent, and causes carbohydrate starvation and

AMPK activation (29). We found that either amino acid starvation (Fig. 4D) or

metformin (Supplementary Fig. 5) stimulated AMPK, raptor and ULK1

phosphorylation in WT hepatocytes, but not in FNDC5/ hepatocytes. Activation of

AMPK with AICAR restored the reduced autophagy in FNDC5–/– hepatocytes (Fig.

4E) and FNDC5–/– liver (Supplementary Fig. 6). Knockdown of AMPK with siRNA

attenuated autophagy in both WT and FNDC5/ hepatocytes (Fig. 4F). These findings

indicate that AMPK is an essential downstream effector of FNDC5 in mediating its

effect on autophagy.

Diabetes Page 12 of 46

Rapamycin repairs the attenuated autophagy in FNDC5 deficiency

Activation of mTORC1 is required for hepatic lipid accumulation (30). Reduced raptor phosphorylation in FNDC5/ liver (Fig. 4D) suggest a possibility that the increased mTORC1 might be involved in the signaling pathway of impaired autophagy. Immunohistochemistry analysis showed that inhibition of mTORC1 with rapamycin prevented the increased p62 expression and the reduced LC3B expression in FNDC5/ liver (Fig. 5A). Consistently, Western blot analysis showed that rapamycin promoted autophagy in both WT and FNDC5–/– livers (Fig. 5B) and

FNDC5–/– hepatocytes (Fig. 5D). Rapamycin increased autophagy flux in both WT and FNDC5–/– hepatocytes. While chloroquine, an autophagy inhibitor, blocked autophagosome fusing with lysosome to form autolysosomes as showing by increased autophagosome accumulation and decreased autolysosomes in both WT and FNDC5–

/– hepatocytes (Fig. 5E). These results indicate that increased mTORC1 activity contributes the attenuated autophagy in FNDC5–/– liver. Moreover, serum ALT and

AST levels, the liver function markers, were increased in FNDC5–/– mice, which was restored by rapamycin (Fig. 5C). Palmitateinduced lipid accumulation is generally used as a cellular steatosis model (31). In order to ascertain whether reduced autophagy in FNDC5–/– mice is involved in lipid accumulation, the effects of rapamycin and Atg5 siRNA on the palmitateinduced lipid accumulation in primary hepatocytes were investigated. Palmitate caused more lipid accumulation in FNDC5–/– hepatocytes than that in WT hepatocytes, which was inhibited by rapamycin.

Inhibition of autophagy by knockdown of Atg5, an essential autophagy gene, with

Page 13 of 46 Diabetes

Atg5 siRNA attenuated the role of rapamycin in reducing lipid accumulation in

FNDC5–/– hepatocytes (Fig. 5F). Moreover, rapamycin reduces TG content in

hepatocytes in FNDC5–/– mice (Supplementary Fig. 7). These results indicate that the

ability of rapamycin to reduce lipid accumulation in FNDC5–/– hepatocytes is largely

dependent on autophagy.

FNDC5 deficiency aggravates hepatosteatosis, lipid metabolic disturbance and

impairment of autophagy in HFD-induced obese mice

HFD is generally used to induce obesity in rodents (12,32). We investigated whether

FNDC5 deficiency causes more severe hepatosteatosis, lipid metabolic disturbance

and impairment of autophagy in mice with obesity induced by HFD for 12 weeks.

Liver weight and the ratio of liver weight to body weight were greater in FNDC5–/–

/HFD mice than those in WT/HFD mice, but there were no significant difference in

body weight and food intake between WT/HFD and FNDC5–/–/HFD mice (Fig. 6A).

FNDC5 deficiency aggravated the HFDinduced increases in NEFA, TG and

cholesterol levels in both serum and livers (Fig. 6B). FAO gene expressions (Pparα,

Hmgcs2, Cpt1, Acox1 and Sirt3) in livers were downregulated (Fig. 6E), while

lipogenic gene expressions (Srebp1c, Dgat1, Fasn and Scd1) were upregulated

(Supplementary Fig. 8) in FNDC5–/–/HFD mice compared with those in WT/HFD

mice. Hepatosteatosis was more severe in FNDC5//HFD mice than that in WT/HFD

mice (Fig. 6D). In HFD mouse, p62 protein expression was increased in FNDC5–/–

liver compared with that in WT liver (Fig. 6F), and serum ALT and AST levels were

higher in FNDC5–/– mice than those in WT mice (Fig. 6C). Moreover, deletion of

Diabetes Page 14 of 46

FNDC5 gene exacerbated the AMPK inhibition and enhanced the mTOR activation in liver from HFD mice (Fig. 6G). These findings indicate that FNDC5 deficiency causes more severe hepatosteatosis, lipid metabolic disturbance and impairment of autophagy in obese mice.

Exogenous FNDC5 enhances FAO and autophagy in vitro

Primary WT hepatocytes incubated with FNDC5 (100 nM) for 12 h or 24 h increased

FAO gene expressions (Pparα, Hmgcs2, Cpt1 and Acox1) in vitro (Fig. 7A). The effects of FNDC5 for 24 h on these FAO gene expressions almost reached its maximal at the concentration of 100 nM (Fig. 7B). Exogenous FNDC5 attenuated TG accumulation in FNDC5–/– hepatocytes (Supplementary Fig. 9). It is known that lipopolysaccharide (LPS) stimulates autophagy in hepatocytes (33). Thus, we compared the role of FNDC5 with LPS in stimulating autophagy. FNDC5 reduced p62 and increased LC3BII levels, similar to the effects of LPS (Fig. 7C). Importantly, palmitateinduced lipid accumulation in primary FNDC5–/– hepatocytes was prevented by FNDC5 (Fig. 7D), and FNDC5 potentiated autophagy in FNDC5–/– hepatocytes

(Fig. 7E). In addition, exogenous FNDC5 attenuated the AMPK inhibition and mTOR activation in FNDC5–/– hepatocytes (Fig. 7F). These results indicate that exogenous

FNDC5 promotes FAO and autophagy, and prevents the FNDC5 deficiencyinduced lipid accumulation and autophagy impairment in vitro.

FNDC5 overexpression attenuates hepatosteatosis and the FAO and autophagy impairment in HFD-induced obese mice

The effects of lentiviral vectormediated FNDC5 overexpression in HFDinduced

Page 15 of 46 Diabetes

obese mice were investigated to determine whether FNDC5 could be used as a

therapeutic strategy for hepatosteatosis and impairment of FAO and autophagy in

obesity. FNDC5 overexpression attenuated the HFDinduced increases in NEFA, TG

and cholesterol levels in both serum and liver (Fig. 8A), and reduced the

HFDinduced lipid accumulation in liver (Fig. 8B). It restored the reduced FAO gene

expression (Fig. 8C) and the increased lipogenic gene expression (Fig. 8D) in

HFDinduced obese mouse liver. Immunohistochemistry showed that the increased

p62 expression and reduced LC3B expression in liver were prevented by FNDC5

overexpression (Fig. 8E), which were further confirmed by Western blot analysis (Fig.

8F). FNDC5 overexpression attenuated the AMPK inhibition and mTOR activation in

liver from HFD mice (Fig. 8G). Effectiveness of FNDC5 overexpression in the

experiments were confirmed by the changes of serum FNDC5 levels (Supplementary

Fig. 10). These findings indicate that longterm increased FNDC5 effectively prevents

hepatosteatosis and attenuates the FAO and autophagy impairment in HFDinduced

obese mice.

DISCUSSION

Hepatic steatosis is generally regarded as hepatic manifestation of the metabolic

syndrome in diabetes and obesity and is thought to be the initial stage in NAFLD.

NAFLD is characterized by the accumulation of TG in lipid droplets within

hepatocytes (1,2,34). The primary novel findings in the present study are that FNDC5

plays critical roles in reducing hepatic lipid accumulation by restoring

AMPK/mTORmediated autophagy, fatty acid oxidation and lipogenesis. FNDC5

Diabetes Page 16 of 46

deficiency deteriorated hepatosteatosis, FAO and autophagy impairment in obesity, while FNDC5 overexpression alleviated hepatosteatosis, improved FAO and autophagy in obesity.

Hepatic FAO are increased in response to energy demand in fasted state (24).

Fasting upregulates TG hydrolysis to supply NFFA for oxidation to meet cellular energy needs (35). An alternative energy source with respect to energy deprivation is provided by the breakdown of cellular components by autophagy (35,36). Induction of autophagy corrects hepatic lipid overaccumulation (37). Defective autophagy is involved in NAFLD (38). We found that hepatic FAO in the fed state was reduced in

FNDC5–/– mice, and the fastinginduced FAO enhancement was much weaker in

FNDC5–/– mice than that in WT mice. Consistently, FNDC5 deficiency caused a mild hepatic lipid accumulation in the fed state, but a severe hepatic lipid accumulation in the fasted state. Palmitateinduced more lipid accumulation in FNDC5–/– hepatocytes than that in WT hepatocytes in vitro. Although FNDC5 deficiency had no significant effect on hepatic autophagy in fed state, the fastinginduced autophagy enhancement response in WT mice almost disappeared in FNDC5–/– mice. Inhibition of autophagy increased lipid accumulation in both WT and FNDC5–/– hepatocytes. These findings indicate that FNDC5 is strongly associated with hepatic FAO and autophagy, which at least partially contribute to the reduction of hepatic lipid accumulation, particularly in the fasting state.

Hepatic steatosis is linked to being obese or overweight in most cases (39). We found that FNDC5 deficiency deteriorated hepatosteatosis accompanying FAO and

Page 17 of 46 Diabetes

autophagy impairment in HFDinduced obesity, while FNDC5 overexpression

alleviated hepatosteatosis and repaired the reduced FAO and autophagy in

HFDinduced obese mice. Exogenous FNDC5 stimulated FAO and autophagy,

attenuated the palmitateinduced hepatic lipid accumulation in primary hepatocytes.

These findings indicate the importance of FNDC5 in attenuating hepatic steatosis.

Administration of FNDC5 or increased FNDC5 production is expected to be a

therapeutic regimen for preventing hepatosteatosis, FAO and autophagy impairment

in obesity.

Most fatty acids in the liver are metabolized by FAO (40). AMPK represses

mTORC1 activity (16), and maintains energy homeostasis via suppressing cellular

ATPconsuming processes and stimulating ATPproducing catabolic pathways

including FAO (41). AMPK inhibits protein synthesis and mTOR signaling (42),

while mTORC1 inhibits PPARα expression and function (16). PPARα stimulates the

expression of FAO genes and implicated in nonalcoholic steatohepatitis (43). We

found that FNDC5 deficiency reduced AMPK, raptor and ULK1 phosphorylation.

Activating AMPK or PPARα partially restored the downregulation of PPARα and

FAO gene expressions. Furthermore, AMPK activation was found in association with

TG levels in FNDC5–/– hepatocytes. PPARα agonist WY14643 increased FAO rate

and reduced lipid accumulation in hepatocytes with FNDC5 deficiency. Inhibition of

mTORC1 attenuated the increased mTORC1 activity and TG levels and partially

restored the attenuated Pparα and FAO gene expressions in FNDC5–/– livers. These

data indicate that FNDC5 deficiency reduces AMPK phosphorylation, which

Diabetes Page 18 of 46

subsequently causes mTORC1 activation, and thus, inhibits PPARα target gene expressions and FAO. It has been shown that ghrelin upregulates autophagy via

AMPK/mTOR restoration (44). Blockage of the mTOR pathway restores endoplasmic reticulum stressinduced autophagy (45). We found that FNDC5 deficiencyinduced defect in autophagy was prevented by AMPK activation and deteriorated by AMPK suppression. Inhibition of mTORC1 restored autophagy impairment and attenuated the liver injury in FNDC5/ mice. Furthermore, rapamycin alleviated palmitateinduced lipid accumulation in FNDC5–/– hepatocytes, which was abolished by the knockdown of the essential autophagy gene Atg5. These results reveal that

FNDC5 deficiency causes AMPK inhibition and mTORC1 activation and autophagy defects. The beneficial effect of mTORC1 inhibition was largely dependent on restoration of autophagy, further suggesting that the autophagic defect in FNDC5 deficiency is partially responsible for hepatic steatosis. It is noted that rapamycin treatment strongly suppressed the liver S6 phosphorylation, but have modest roles in increasing FAO and reducing lipogenic gene expressions in fasting FNDC5KO mice.

This discrepancy suggest a possibility that some other signal pathways may be involved in regulating FAO and lipogenesis.

Lipogenesis is another important factor contributing to lipid accumulation in liver

(24). Decreased lipogenic gene expressions resulted from mTORC1 inhibition was found in FNDC5–/– mice. In HFDinduced obese mice, lipogenic gene expressions were increased in FNDC5–/– mice compared with WT mice. FNDC5 overexpression prevented the increased lipogenic gene expressions in HFDinduced obese mice.

Page 19 of 46 Diabetes

These results indicate that FNDC5 deficiency stimulates lipogenesis via mTOR

pathway, which is involved in lipid accumulation in liver of HFDinduced obesity. On

the other hand, serum TG and NEFA levels were increased in FNDC5–/– mice,

particularly in the fasting state. FNDC5 deficiency deteriorated hyperlipemia in

HFDinduced obese mice. Our previous study has shown that FNDC5 overexpression

prevented hyperlipemia in HFDinduced obese mice (12), which was further

confirmed in the present study. These results revealed that FNDC5 plays a critical role

in preventing hyperlipemia.

Previous study have showed that FNDC5 overexpression in HFDinduced obese

mice increases energy expenditure, attenuates hyperglycemia and insulin resistance,

and activates lipolysis in adipose tissues (12). Irisin, a cleaved and secreted fragment

of FNDC5, reduces gluconeogenesis, increases glycogenesis, and improves insulin

resistance in streptozotocin/HFDinduced type 2 diabetes (13). Strong irisin

immunoreactivity has been found in liver (46,47). Serum irisin concentrations were

inversely associated with liver TG contents in the liver in obese adults (48). A

limitation in the present study is whether the effects of FNDC5 are caused by its

cleaved fragment irisin were not investigated.

In summary, FNDC5 reduces hepatic lipid accumulation via

AMPK/mTORmediated autophagy and FAO enhancement and de novo lipogenesis

reduction. FNDC5 deficiency aggravates, while FNDC5 overexpression prevents

hepatic steatosis, hyperlipemia, impaired FAO and autophagy, and enhanced

lipogenesis in obesity (Supplementary Fig. 11). FNDC5 can be used as a therapeutic

Diabetes Page 20 of 46

regimen for preventing hepatosteatosis and impairment of FAO and autophagy in obesity.

Author contributions. T.Y.L., X.Q.X., X.S.R., M.X.Z., C.X.S., J.J.W., Y.B.Z. and Z.F. performed the experiments. T.Y.L., X.Q.X., Y.H., X.Y.G. and G.Q.Z. analyzed the data. T.Y.L., X.Y.G., Q.C., Y.H.L., Y.M.K. and G.Q.Z. wer involved in study design.

T.Y.L. and G.Q.Z. wrote the manuscript. T.Y.L., X.Y.G., Q.C., Y.H.L., Y.M.K. and

G.Q.Z. edited the manuscript.

Acknowledgments. We thank the generous support of the Collaborative Innovation

Center for Cardiovascular Disease Translational Medicine.

Funding. This study was supported by National Natural Science Foundation of China

(31271213, 31571167 & 91439120).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Page 21 of 46 Diabetes

References

1. Adams,LA, Ratziu,V: Nonalcoholic fatty liver Perhaps not so benign. J Hepatol

2015; 62:10021004

2. Byrne,CD, Targher,G: NAFLD: A multisystem disease. J Hepatol 2015; 62:S47S64

3. Marra,F, Lotersztajn,S: Pathophysiology of NASH: perspectives for a targeted

treatment. Curr Pharm Des 2013; 19:52505269

4. Eaton,S, Bartlett,K, Pourfarzam,M: Mammalian mitochondrial betaoxidation.

Biochem J 1996; 320 ( Pt 2):345357

5. Koo,SH: Nonalcoholic fatty liver disease: molecular mechanisms for the hepatic

steatosis. Clin Mol Hepatol 2013; 19:210215

6. Kwanten,WJ, Martinet,W, Michielsen,PP, Francque,SM: Role of autophagy in the

pathophysiology of nonalcoholic fatty liver disease: a controversial issue. World J

Gastroenterol 2014; 20:73257338

7. Codogno,P, Lotersztajn,S: When autophagy chaperones liver metabolism. Cell Metab

2014; 20:392393

8. Amir,M, Czaja,MJ: Autophagy in nonalcoholic steatohepatitis. Expert Rev

Gastroenterol Hepatol 2011; 5:159166

9. Codogno,P, Meijer,AJ: Autophagy in the liver. J Hepatol 2013; 59:389391

10. Xiao,Y, Liu,H, Yu,J, Zhao,Z, Xiao,F, Xia,T, Wang,C, Li,K, Deng,J, Guo,Y, Chen,S,

Chen,Y, Guo,F: Activation of ERK1/2 Ameliorates Liver Steatosis in Leptin

ReceptorDeficient (db/db) Mice via Stimulating ATG7Dependent Autophagy.

Diabetes 2016; 65:393405

Diabetes Page 22 of 46

11. Bostrom,P, Wu,J, Jedrychowski,MP, Korde,A, Ye,L, Lo,JC, Rasbach,KA,

Bostrom,EA, Choi,JH, Long,JZ, Kajimura,S, Zingaretti,MC, Vind,BF, Tu,H, Cinti,S,

Hojlund,K, Gygi,SP, Spiegelman,BM: A PGC1alphadependent myokine that drives

brownfatlike development of white fat and thermogenesis. Nature 2012;

481:463468

12. Xiong,XQ, Chen,D, Sun,HJ, Ding,L, Wang,JJ, Chen,Q, Li,YH, Zhou,YB, Han,Y,

Zhang,F, Gao,XY, Kang,YM, Zhu,GQ: FNDC5 overexpression and irisin ameliorates

glucose/lipid metabolic derangements and enhances lipolysis in obesity. Biochim

Biophys Acta 2015; 1852:18671875

13. Liu,TY, Shi,CX, Gao,R, Sun,HJ, Xiong,XQ, Ding,L, Chen,Q, Li,YH, Wang,JJ,

Kang,YM, Zhu,GQ: Irisin inhibits hepatic gluconeogenesis and increases glycogen

synthesis via the PI3K/Akt pathway in type 2 diabetic mice and hepatocytes. Clin Sci

(Lond) 2015; 129:839850

14. Russo,GL, Russo,M, Ungaro,P: AMPactivated protein kinase: a target for old drugs

against diabetes and cancer. Biochem Pharmacol 2013; 86:339350

15. Zoncu,R, Efeyan,A, Sabatini,DM: mTOR: from growth signal integration to cancer,

diabetes and ageing. Nat Rev Mol Cell Biol 2011; 12:2135

16. Sengupta,S, Peterson,TR, Laplante,M, Oh,S, Sabatini,DM: mTORC1 controls

fastinginduced ketogenesis and its modulation by ageing. Nature 2010;

468:11001104

17. Frances,DE, Motino,O, Agra,N, GonzalezRodriguez,A, FernandezAlvarez,A,

Cucarella,C, Mayoral,R, CastroSanchez,L, GarciaCasarrubios,E, Bosca,L,

Page 23 of 46 Diabetes

Carnovale,CE, Casado,M, Valverde,AM, MartinSanz,P: Hepatic cyclooxygenase2

expression protects against dietinduced steatosis, obesity, and insulin resistance.

Diabetes 2015; 64:15221531

18. Tateya,S, RizzoDe,LN, Handa,P, Cheng,AM, MorganStevenson,V, Ogimoto,K,

Kanter,JE, Bornfeldt,KE, Daum,G, Clowes,AW, Chait,A, Kim,F: VASP increases

hepatic fatty acid oxidation by activating AMPK in mice. Diabetes 2013;

62:19131922

19. BamjiMirza,M, Zhang,W, Yao,Z: Expression of human hepatic lipase negatively

impacts apolipoprotein AI production in primary hepatocytes from Lipcnull mice. J

Biomed Res 2014; 28:201212

20. Klingmuller,U, Bauer,A, Bohl,S, Nickel,PJ, Breitkopf,K, Dooley,S, Zellmer,S,

Kern,C, Merfort,I, Sparna,T, Donauer,J, Walz,G, Geyer,M, Kreutz,C, Hermes,M,

Gotschel,F, Hecht,A, Walter,D, Egger,L, Neubert,K, Borner,C, Brulport,M,

Schormann,W, Sauer,C, Baumann,F, Preiss,R, MacNelly,S, Godoy,P, Wiercinska,E,

Ciuclan,L, Edelmann,J, Zeilinger,K, Heinrich,M, Zanger,UM, Gebhardt,R,

Maiwald,T, Heinrich,R, Timmer,J, von,WF, Hengstler,JG: Primary mouse

hepatocytes for systems biology approaches: a standardized in vitro system for

modelling of signal transduction pathways. Syst Biol (Stevenage ) 2006; 153:433447

21. Liu,Y, Palanivel,R, Rai,E, Park,M, Gabor,TV, Scheid,MP, Xu,A, Sweeney,G:

Adiponectin stimulates autophagy and reduces oxidative stress to enhance insulin

sensitivity during highfat diet feeding in mice. Diabetes 2015; 64:3648

22. McCue,MD: Starvation physiology: reviewing the different strategies animals use to

Diabetes Page 24 of 46

survive a common challenge. Comp Biochem Physiol A Mol Integr Physiol 2010;

156:118

23. Morran,DC, Wu,J, Jamieson,NB, Mrowinska,A, Kalna,G, Karim,SA, Au,AY,

Scarlett,CJ, Chang,DK, Pajak,MZ, Oien,KA, McKay,CJ, Carter,CR, Gillen,G,

Champion,S, Pimlott,SL, Anderson,KI, Evans,TR, Grimmond,SM, Biankin,AV,

Sansom,OJ, Morton,JP: Targeting mTOR dependency in pancreatic cancer. Gut 2014;

63:14811489

24. Bechmann,LP, Hannivoort,RA, Gerken,G, Hotamisligil,GS, Trauner,M, Canbay,A:

The interaction of hepatic lipid and glucose metabolism in liver diseases. J Hepatol

2012; 56:952964

25. Kroemer,G, Marino,G, Levine,B: Autophagy and the integrated stress response. Mol

Cell 2010; 40:280293

26. Hales,EC, Taub,JW, Matherly,LH: New insights into Notch1 regulation of the

PI3KAKTmTOR1 signaling axis: targeted therapy of gammasecretase inhibitor

resistant Tcell acute lymphoblastic leukemia. Cell Signal 2014; 26:149161

27. Zhang,MZ, Wang,Y, Paueksakon,P, Harris,RC: Epidermal growth factor receptor

inhibition slows progression of diabetic nephropathy in association with a decrease in

endoplasmic reticulum stress and an increase in autophagy. Diabetes 2014;

63:20632072

28. Egan,DF, Shackelford,DB, Mihaylova,MM, Gelino,S, Kohnz,RA, Mair,W,

Vasquez,DS, Joshi,A, Gwinn,DM, Taylor,R, Asara,JM, Fitzpatrick,J, Dillin,A,

Viollet,B, Kundu,M, Hansen,M, Shaw,RJ: Phosphorylation of ULK1 (hATG1) by

Page 25 of 46 Diabetes

AMPactivated protein kinase connects energy sensing to mitophagy. Science 2011;

331:456461

29. Loos,JA, Cumino,AC: In Vitro AntiEchinococcal and Metabolic Effects of

Metformin Involve Activation of AMPActivated Protein Kinase in Larval Stages of

Echinococcus granulosus. PLoS One 2015; 10:e0126009

30. Sapp,V, Gaffney,L, EauClaire,SF, Matthews,RP: Fructose leads to hepatic steatosis

in zebrafish that is reversed by mechanistic target of rapamycin (mTOR) inhibition.

Hepatology 2014; 60:15811592

31. Park,JY, Kim,Y, Im,JA, Lee,H: Oligonol suppresses lipid accumulation and improves

insulin resistance in a palmitateinduced in HepG2 hepatocytes as a cellular steatosis

model. BMC Complement Altern Med 2015; 15:185

32. Aoun,M, Fouret,G, Michel,F, Bonafos,B, Ramos,J, Cristol,JP, Carbonneau,MA,

Coudray,C, FeilletCoudray,C: Dietary fatty acids modulate liver mitochondrial

cardiolipin content and its fatty acid composition in rats with non alcoholic fatty liver

disease. J Bioenerg Biomembr 2012; 44:439452

33. Chen,C, Deng,M, Sun,Q, Loughran,P, Billiar,TR, Scott,MJ: Lipopolysaccharide

stimulates p62dependent autophagylike aggregate clearance in hepatocytes. Biomed

Res Int 2014; 2014:267350

34. Marchesini,G, Mazzotti,A: NAFLD incidence and remission: Only a matter of weight

gain and weight loss? Journal of Hepatology 2015; 62:1517

35. Mizushima,N, Levine,B, Cuervo,AM, Klionsky,DJ: Autophagy fights disease

through cellular selfdigestion. Nature 2008; 451:10691075

Diabetes Page 26 of 46

36. Komatsu,M, Waguri,S, Ueno,T, Iwata,J, Murata,S, Tanida,I, Ezaki,J, Mizushima,N,

Ohsumi,Y, Uchiyama,Y, Kominami,E, Tanaka,K, Chiba,T: Impairment of

starvationinduced and constitutive autophagy in Atg7deficient mice. J Cell Biol

2005; 169:425434

37. Farah,BL, Landau,DJ, Sinha,RA, Brooks,ED, Wu,Y, Fung,SYS, Tanaka,T,

Hirayama,M, Bay,BH, Koeberl,DD, Yen,PM: Induction of autophagy improves

hepatic lipid metabolism in glucose6phosphatase deficiency. J Hepatol 2016;

64:370379

38. Singh,R, Kaushik,S, Wang,Y, Xiang,Y, Novak,I, Komatsu,M, Tanaka,K, Cuervo,AM,

Czaja,MJ: Autophagy regulates lipid metabolism. Nature 2009; 458:11311135

39. Arslan,N: Obesity, fatty liver disease and intestinal microbiota. World J

Gastroenterol 2014; 20:1645216463

40. Lodhi,IJ, Semenkovich,CF: Peroxisomes: a nexus for lipid metabolism and cellular

signaling. Cell Metab 2014; 19:380392

41. Kemmerer,M, Finkernagel,F, Cavalcante,MF, Abdalla,DS, Muller,R, Brune,B,

Namgaladze,D: AMPActivated Protein Kinase Interacts with the Peroxisome

ProliferatorActivated Receptor Delta to Induce Genes Affecting Fatty Acid

Oxidation in Human Macrophages. PLoS One 2015; 10:e0130893

42. Reiter,AK, Bolster,DR, Crozier,SJ, Kimball,SR, Jefferson,LS: Repression of protein

synthesis and mTOR signaling in rat liver mediated by the AMPK activator

aminoimidazole carboxamide ribonucleoside. Am J Physiol Endocrinol Metab 2005;

288:E980E988

Page 27 of 46 Diabetes

43. Pawlak,M, Lefebvre,P, Staels,B: Molecular mechanism of PPARα; action and its

impact on lipid metabolism, inflammation and fibrosis in nonalcoholic fatty liver

disease. J Hepatol 2015; 62:720733

44. Mao,Y, Cheng,J, Yu,F, Li,H, Guo,C, Fan,X: Ghrelin Attenuated Lipotoxicity via

Autophagy Induction and Nuclear FactorkappaB Inhibition. Cell Physiol Biochem

2015; 37:563576

45. Huang,H, Li,X, Zhuang,Y, Li,N, Zhu,X, Hu,J, Ben,J, Yang,Q, Bai,H, Chen,Q: Class

A scavenger receptor activation inhibits endoplasmic reticulum stressinduced

autophagy in macrophage. J Biomed Res 2014; 28:213221

46. Aydin,S: Three new players in energy regulation: Preptin, adropin and irisin. Peptides

2014; 56C:94110

47. Aydin,S, Kuloglu,T, Aydin,S, Eren,MN, Celik,A, Yilmaz,M, Kalayci,M, Sahin,I,

Gungor,O, Gurel,A, Ogeturk,M, Dabak,O: Cardiac, skeletal muscle and serum irisin

responses to with or without water exercise in young and old male rats: cardiac

muscle produces more irisin than skeletal muscle. Peptides 2014; 52:6873

48. Zhang,HJ, Zhang,XF, Ma,ZM, Pan,LL, Chen,Z, Han,HW, Han,CK, Zhuang,XJ, Lu,Y,

Li,XJ, Yang,SY, Li,XY: Irisin is inversely associated with intrahepatic triglyceride

contents in obese adults. J Hepatol 2013; 59:557562

Diabetes Page 28 of 46

Figure Legends

Figure 1—FNDC5 deficiency causes lipid accumulation in liver after fasting.

Twomonth old WT and FNDC5/ mice were fasted for 16 h. A: oil Red O staining showing lipid droplets in the liver sections. B: liver TG and NEFA levels. C: serum

TG and NEFA levels. D: serum FNDC5 levels determined with ELISA method. E: liver FNDC5 protein expression determined with Western blot. F: liver FNDC5 mRNA. *P<0.05 vs. WT. †P<0.05 vs. Fed. n=6.

Figure 2—Impaired FAO in liver of FNDC5/ mice. A: hepatic FAO gene expression after 16 hour’s starvation. B: effects of AICAR (200 mg/kg, i.p.) for 5 days on hepatic

FAO gene expression. C: effects of WY14643 on PPARα target gene expression in hepatocytes treated with palmitate (125 M) and WY14643 (30 M) for 6 h. D: effects of AICAR (200 mg/kg, i.p.) for 5 days on hepatic TG contents. E: effects of

AMPKsiRNA (5 nM) for 48 h on TG levels in hepatocytes. F: effects of WY14643 on TG levels and FAO rate in hepatocytes treated with 14Cpalmitate (0.25 Ci), carnitine (0.05 mM) and WY14643 (30 M) for 24 h. *P<0.05 vs. WT; †P<0.05 vs.

Fed; ‡P<0.05 vs. Saline or DMSO; #P<0.05 vs. CtrlsiRNA. n=3 in AC; n=6 in DF.

Figure 3—Rapamycin attenuates the impaired FAO and enhanced lipogenesis in the liver of FNDC5–/– mice. Mice treated with rapamycin (5 mg/kg) for 3 days followed by fasting for 16 h. A: S6 phosphorylation. B: FAO related gene expressions. C: lipogenic gene expressions. D: TG levels. *P<0.05 vs. WT; †P<0.05 vs. Saline;

‡P<0.05 vs. Fed. n=3 in AC; n=6 in D.

Figure 4—Reduced autophagy in liver of FNDC5/ mice. A: autophagy gene

Page 29 of 46 Diabetes

expressions in liver. B: p62 and LC3B expressions in liver with or without fasting for

16 h. C: images showing LC3 staining in GFPRFPLC3 adenovirus infected

hepatocytes with or without amino acid deprivation for 2 h. Green dots,

autophagosomes; red dots, autolysosomes; yellow dots, autophagosomes. D:

expressions of FNDC5, phosphorylation of AMPK, Raptor and ULK1 in hepatocytes,

which were subjected to serum starvation for 8 h followed by amino acid starvation. E:

effects of AICAR (1 mM) for 24 h on AMPK phosphorylation, p62 and LC3B

expressions in hepatocytes. F: effects of AMPKsiRNA (5 nM) for 48 h on AMPK

phosphorylation, p62 and LC3B expressions in hepatocytes. *P<0.05 vs. WT.

†P<0.05 vs Fed or saline or controlsiRNA. n=4.

Figure 5—Rapamycin repairs the attenuated autophagy in liver of FNDC5–/– mice.

AC, Mice treated with rapamycin (5 mg/kg) for 3 days followed by fasting for 16 h.

A: hepatic immunohistochemistry for p62 and LC3B. B: hepatic p62 and LC3B

protein expressions. C: serum ALT and AST. D: effects of rapamycin (5 M for 2 h)

on p62 and LC3B expressions in hepatocytes. E: images showing the effects of

rapamycin (50 nM) or chloroquine (10 M) on autophagy in GFPRFPLC3

adenovirus infected hepatocytes. Green dots, autophagosomes; red dots,

autolysosomes; yellow dots, autophagosomes. F: Oilred O staining showing lipid

accumulation. Hepatocytes were incubated with controlsiRNA and Atg5siRNA

followed by treatment of palmitate (250 M) with or without rapamycin (5 M) for

16 h. *P<0.05 vs. WT. †P<0.05 vs. Saline. n=3.

Figure 6—FNDC5 deficiency exacerbates lipid accumulation, attenuated FAO and

Diabetes Page 30 of 46

autophagy in liver due to 12week HFD in mice. A: body weight, liver weight and livertobody weight ratio, average food intake. B: NEFA, TG and cholesterol levels in serum and liver. C: serum ALT and AST levels. D: Oil Red O staining showing lipid droplets in the liver sections. E: FAOrelated gene expression in liver. F: p62 protein expression. G: phosphorylation of AMPK, raptor and S6. *P<0.05 vs. WT.

†P<0.05 vs. Ctrl. n=7 in AC, n=3 in DG.

FNDC5–/– hepatocytes. E: effects of FNDC5 (100 nM) on p62 and LC3B expressions in FNDC5–/– hepatocytes. F: effects of FNDC5 (100 nM) on the phosphorylation of

AMPK, raptor and S6 in WT and FNDC5–/– hepatocytes.*P<0.05 vs. PBS. †P<0.05 vs.

WT. n=3.

Figure 8—FNDC5 overexpression repairs attenuated FAO and autophagy in livers in

HFDinduced obese mice. A: NEFA, TG and CHO levels in serum and liver. B:

Oilred O staining showing the lipid accumulation in liver. C: hepatic FAO gene expressions. D: hepatic lipogenic gene expressions. E: immunohistochemistry of liver sections for p62 and LC3B. F: hepatic p62 and LC3B protein expressions. G: phosphorylation of AMPK, raptor and S6. *P<0.05 vs. Vector. †P<0.05 vs. Ctrl. n=6

Page 31 of 46 Diabetes

in A. n=3 in BG.

Diabetes Page 32 of 46

Figure 1—FNDC5 deficiency causes lipid accumulation in liver after fasting. Two-month old WT and FNDC5-/- mice were fasted for 16 h. A: oil Red O staining showing lipid droplets in the liver sections. B: liver TG and NEFA levels. C: serum TG and NEFA levels. D: serum FNDC5 levels determined with ELISA method. E: liver FNDC5 protein expression determined with Western blot. F: liver FNDC5 mRNA. *P<0.05 vs. WT. †P<0.05 vs. Fed. n=6. Figure 1 52x36mm (300 x 300 DPI)

Page 33 of 46 Diabetes

Figure 2—Impaired FAO in liver of FNDC5/ mice. A: hepatic FAO gene expression after 16 hour’s starvation. B: effects of AICAR (200 mg/kg, i.p.) for 5 days on hepatic FAO gene expression. C: effects of WY14643 on PPARα target gene expression in hepatocytes treated with palmitate (125 M) and WY14643 (30 M) for 6 h. D: effects of AICAR (200 mg/kg, i.p.) for 5 days on hepatic TG contents. E: effects of AMPKsiRNA (5 nM) for 48 h on TG levels in hepatocytes. F: effects of WY14643 on TG levels and FAO rate in hepatocytes treated with 14Cpalmitate (0.25 Ci), carnitine (0.05 mM) and WY14643 (30 M) for 24 h. *P<0.05 vs. WT; †P<0.05 vs. Fed; ‡P<0.05 vs. Saline or DMSO; #P<0.05 vs. CtrlsiRNA. n=3 in AC; n=6 in DF. Figure 2 74x72mm (300 x 300 DPI)

Diabetes Page 34 of 46

Page 35 of 46 Diabetes

Figure 4—Reduced autophagy in liver of FNDC5-/- mice. A: autophagy gene expressions in liver. B: p62 and LC3B expressions in liver with or without fasting for 16 h. C: images showing LC3 staining in GFP-RFP-LC3 adenovirus infected hepatocytes with or without amino acid deprivation for 2 h. Green dots, autophagosomes; red dots, autolysosomes; yellow dots, autophagosomes. D: expressions of FNDC5, phosphorylation of AMPK, Raptor and ULK1 in hepatocytes, which were subjected to serum starvation for 8 h followed by amino acid starvation. E: effects of AICAR (1 mM) for 24 h on AMPK phosphorylation, p62 and LC3B expressions in hepatocytes. F: effects of AMPK-siRNA (5 nM) for 48 h on AMPK phosphorylation, p62 and LC3B expressions in hepatocytes. *P<0.05 vs. WT. †P<0.05 vs Fed or saline or control-siRNA. n=4. Figure 4 102x137mm (300 x 300 DPI)

Diabetes Page 36 of 46

Figure 5—Rapamycin repairs the attenuated autophagy in liver of FNDC5–/– mice. A-C, Mice treated with rapamycin (5 mg/kg) for 3 days followed by fasting for 16 h. A: hepatic immunohistochemistry for p62 and LC3B. B: hepatic p62 and LC3B protein expressions. C: serum ALT and AST. D: effects of rapamycin (5 µM for 2 h) on p62 and LC3B expressions in hepatocytes. E: images showing the effects of rapamycin (50 nM) or chloroquine (10 µM) on autophagy in GFP-RFP-LC3 adenovirus infected hepatocytes. Green dots, autophagosomes; red dots, autolysosomes; yellow dots, autophagosomes. F: Oil-red O staining showing lipid accumulation. Hepatocytes were incubated with control-siRNA and Atg5-siRNA followed by treatment of palmitate (250 µM) with or without rapamycin (5 µM) for 16 h. *P<0.05 vs. WT. †P<0.05 vs. Saline. n=3. Figure 5 112x166mm (300 x 300 DPI)

Page 37 of 46 Diabetes

Figure 6—FNDC5 deficiency exacerbates lipid accumulation, attenuated FAO and autophagy in liver due to 12-week HFD in mice. A: body weight, liver weight and liver-to-body weight ratio, average food intake. B: NEFA, TG and cholesterol levels in serum and liver. C: serum ALT and AST levels. D: Oil Red O staining showing lipid droplets in the liver sections. E: FAO-related gene expression in liver. F: p62 protein expression. G: phosphorylation of AMPK, raptor and S6. *P<0.05 vs. WT. †P<0.05 vs. Ctrl. n=7 in A-C, n=3 in D-G. Figure 6 80x85mm (300 x 300 DPI)

Diabetes Page 38 of 46

Figure 7—Exogenous FNDC5 enhances FAO and autophagy in hepatocytes. A: time effects of FNDC5 (100 nM) on FAO related gene expressions in WT hepatocytes. B: dose effects of FNDC5 (20, 100 and 200 nM) on FAO related gene expressions in WT hepatocytes. C: effects of LPS (100 ng/mL) or FNDC5 (100 nM) on p62 and LC3B expressions in WT hepatocytes. D: Oil-red O staining showing that FNDC5 (100 nM) prevented palmitate (250 µM)-induced lipid accumulation (red color) in WT and FNDC5–/– hepatocytes. E: effects of FNDC5 (100 nM) on p62 and LC3B expressions in FNDC5–/– hepatocytes. F: effects of FNDC5 (100 nM) on the phosphorylation of AMPK, raptor and S6 in WT and FNDC5–/– hepatocytes.*P<0.05 vs. PBS. †P<0.05 vs. WT. n=3. Figure 7 86x97mm (300 x 300 DPI)

Page 39 of 46 Diabetes

Figure 8—FNDC5 overexpression repairs attenuated FAO and autophagy in livers in HFD-induced obese mice. A: NEFA, TG and CHO levels in serum and liver. B: Oil-red O staining showing the lipid accumulation in liver. C: hepatic FAO gene expressions. D: hepatic lipogenic gene expressions. E: immunohistochemistry of liver sections for p62 and LC3B. F: hepatic p62 and LC3B protein expressions. G: phosphorylation of AMPK, raptor and S6. *P<0.05 vs. Vector. †P<0.05 vs. Ctrl. n=6 in A. n=3 in B-G. Figure 8 95x120mm (300 x 300 DPI)

Diabetes Page 40 of 46

Online Data Supplement

FNDC5 alleviates hepatosteatosis by restoring AMPK/mTOR mediated autophagy, fatty acid oxidation and lipogenesis in mice

TongYan Liu1, XiaoQing Xiong1, XingSheng Ren1, MingXia Zhao1, ChangXiang Shi1, JueJin Wang1, YeBo Zhou1, Feng Zhang1, Ying Han1, XingYa Gao1, Qi Chen2, YueHua Li2,YuMing Kang3, GuoQing Zhu1,*

1Key Laboratory of Cardiovascular Disease and Molecular Intervention, Department of Physiology, Nanjing Medical University, Nanjing, Jiangsu 210029, China; 2Department of Pathophysiology, Nanjing Medical University, Nanjing, Jiangsu 210029, China; 3Department of Physiology and Pathophysiology, Cardiovascular Research Center, Xi'an Jiaotong University School of Medicine, Xi'an 710061, China

Page 41 of 46 Diabetes

Supplemental Figures

Supplementary Fig. 1 Effects of fasting for 16 h on liver AMPK phosphorylation in WT and FNDC5/ mice. *P<0.05 vs. WT; †P<0.05 vs. Fed. n=3.

Supplementary Fig. 2 Effects of an AMPK activator AICAR (200 mg/kg, i.p.) for 5 days on the liver AMPK, raptor and S6 phosphorylation in WT and FNDC5/ mice. *P<0.05 vs. WT; †P<0.05 vs. Saline. n=3.

Diabetes Page 42 of 46

Supplementary Fig. 3 Effects of WY14643, a PPARα agonist, on lipid accumulation in hepatocytes (Oil Red O staining, ×200). Hepatocytes were incubated with palmitate (125 M) with WY14643 (30 M) or vehicle for 6 h.

Supplementary Fig. 4 Immunohistochemistry of liver sections for p62 and LC3B in WT and FNDC5/ mice with or without fasting for 16 h.

Page 43 of 46 Diabetes

Supplementary Fig. 5 Expression of FNDC5, and phosphorylation of AMPK, Raptor and ULK1 in primary WT and FNDC5/ mouse hepatocytes treated with metformin. The hepatocytes were serum starved for 8 hours and then stimulated by metformin (2 mM). Values are mean±SE. * P<0.05 vs. WT. n=4.

Supplementary Fig. 6 Effects of an AMPK activator AICAR (200 mg/kg, i.p.) for 5 days on autophagy in WT and FNDC5/ mice. Liver p62 expression and LC3BII/LC3BI were used as an autophagy flux marker and an autophagosome marker, respectively. *P<0.05 vs. WT; †P<0.05 vs. Saline. n=3.

Diabetes Page 44 of 46

Supplementary Fig. 7 Rapamycin reduces TG content in hepatocytes in FNDC5–/– mice. Values are mean±SE *P<0.05 vs. WT. †P<0.05 vs. Saline. n=6.

Supplementary Fig. 8 FNDC5 deficiency increases the expressions of lipogenic genes (Srebp1c, Dgat1, Fasn and Scd1) in mice fed with highfat diet (HFD). Values are mean±SE. *P<0.05 vs. WT. †P<0.05 vs. Ctrl. n=3.

Supplementary Fig. 9 Exogenous FNDC5 attenuates palmitateinduced lipid accumulation in FNDC5deficient hepatocytes. Values are mean±SE. *P<0.05 vs. PBS. †P<0.05 vs. WT. n=3.

Page 45 of 46 Diabetes

Supplementary Fig. 10 FNDC5 overexpression on serum FNDC5 levels control mice and HFDinduced obese mice. Values are mean±SE. *P<0.05 vs. Vector. n=6 for each group.

Supplementary Fig. 11 Schematic diagram showing the roles of FNDC5 and its signaling. A, physiological significance of FNDC5 and its downstream signal mechanism; B, FNDC5 deficiency induced changes; C, relationship between HFDinduced obesity and FNDC5. Solid arrow, stimulation; dashed arrow, inhibition.

Diabetes Page 46 of 46

Supplemental Tables Supplementary Table 1 Primers for Realtime quantitative PCR analysis in mice Genes Forward Reverse Pparα CAGTGGGGAGAGAGGACAGA AGTTCGGGAACAAGACGTT Hmgcs2 ATACCACCAACGCCTGTTATG CAATGTCACCACAGACCACCA Cpt1 AGTGGCCTCACAGACTCCAG GCCCATGTTGTACAGCTTCC Acox1 CCTGATTCAGCAAGGTAGGG TCGCAGACCCTGAAGAAATC Sirt3 ATAGAATTCATGGCGCTTGAC ATAGAATTCTCTGTCCTGTCC Cyp4a10 AAGGGTCAAACACCTCTGGA GATGGACGCTCTTTACCCAA Cyp4a12 GGTGTCCAAGGTCATCAAGG TGGCAGACTCTGTTCGTGTC Ehhadh TGGCTCTAACCGTATGGTCC CTATGATCCGCCTCTGCAA Acs1 ACCATCAGTGGTACCCGCTA CGCTCACCACCTTCTGGTAT Peci CGAGTTGGCTGAATGGAGTA CCAGCTGTGGGAATCTCTGT Srebp1c TGGTTGTTGATGAGCTGGAG GGCTCTGGAACAGACACTGG Dgat1 TCACCACACACCAATTCAGG GACGGCTACTGGGATCTGA Fasn GTTGGCCCAGAACTCCTGTA GTCGTCTGCCTCCAGAGC Scd1 CACCTGCCTCTTCGGGATTT TCTGAGAACTTGTGGTGGGC Ulk1 TTACCAGCGCATCGAGCA TGGGGAGAAGGTGTGTAGGG Ulk2 GGATTAAAACCGGTGAATGG TGATGGGAGTTCCTACATGAAA Atg5 ACAGCTTCTGGATGAAAGGC TGGGACTGCAGAATGACAGA Atg8 GCTGCTTCTCCCCCTTGTAT CCGAGAAGACCTTCAAGCAG Atg10 TTCTGAAGTGACGAGACCTGC AGCCTCGGCTTATAGCACTCA