Nocon Thesis

BOSTON UNIVERSITY

SCHOOL OF MEDICINE

Thesis

RESVERATROL STIMULATION OF SIRT1 & EXOGENOUS DELIVERY

OF FGF21 MIMICS METFORMIN’S ABILITY TO ALLEVIATE

NON-ALCOHOLIC FATTY LIVER DISEASE

CAUSED BY DIET-INDUCED OBESITY

by

ALLISON LEE NOCON

B.S., University of Florida, 2012

Submitted in partial fulfillment of the

requirements for the degree of

Master of Arts

2015

© 2015 by ALLISON LEE NOCON All rights reserved

Approved by

First Reader Mengwei Zang, Ph.D. Associate Professor of Medicine

Second Reader Susan K. Fried, Ph.D. Professor of Medicine and Biochemistry

  iv  

DEDICATION

I would like to dedicate this work to my mother, Dr. Maria Elen Gajo,

who is eternally supportive and unconditionally loving.

  v  

ACKNOWLEDGMENTS

This work would not have been possible without the support and guidance provided by

Mengwei Zang, Susan Fried, and Lynn Moore. I thank them for their enthusiasm,

encouragement, and seemingly endless scientific knowledge. I would also like to thank

Richard Cohen as well as other members of the Vascular Biology Unit of Boston

University School of Medicine, Xianliang Rui, Ting Luo, and Alex Sherban for their aid

in making these animal studies possible.

  vi  

RESVERATROL STIMULATION OF SIRT1 & EXOGENOUS DELIVERY

OF FGF21 MIMICS METFORMIN’S ABILITY TO ALLEVIATE

NON-ALCOHOLIC FATTY LIVER DISEASE

CAUSED BY DIET-INDUCED OBESITY

ALLISON LEE NOCON

ABSTRACT

Metformin has been used clinically since 1957 for its efficacy and safety as

therapy for type 2 diabetes. Besides ameliorating hyperglycemia without risk of

hypoglycemia, metformin also lowers plasma triglyceride levels. Furthermore, a wealth

of data shows that metformin facilitates weight loss in mice as well as humans. Due to its

numerous metabolic benefits, researchers and clinicians are interested in the possibility of

using metformin as treatment to combat obesity and other metabolic disorders such as

non-alcoholic fatty liver disease (NAFLD). Despite being the most commonly prescribed

anti-diabetic, metformin’s complete mechanism(s) for weight loss or for lowering

glucose and lipids remains an enigma. Our studies show that metformin-treated mice

exhibited decreased caloric intake, providing a viable mechanism for metformin to bring

about weight loss. Intriguingly, we found that metformin induces PRDM16 to promote

browning of iWAT and increase expression of thermogenic genes such as UCP1 and

DIO2. However, metformin did not appear to increase energy expenditure. It’s possible

that metformin’s effect on energy expenditure was masked since energy expenditure

  vii  

measurements were taken when metformin-treated mice were still losing weight and were

in a state of negative energy balance.

Recently, there has been much attention given to AMPK activators as exercise

mimetics. Metformin is known to activate AMPK and similarly brings about many

beneficial effects as exercise such as alleviation of obesity-induced NAFLD. SIRT1

stimulation by resveratrol and delivery of exogenous FGF21 mimics metformin’s ability

to combat obesity and improve NAFLD. Collectively, these results implicate metformin,

resveratrol, and exogenous administration of FGF21 as beneficial therapies for weight

loss and amelioration of NAFLD.

  viii  

TABLE OF CONTENTS

TITLE……………………………………………………………………………………...i

COPYRIGHT PAGE……………………………………………………………………...ii

READER APPROVAL PAGE…………………………………………………………..iii

DEDICATION ...................................................................................................................iv  

ACKNOWLEDGMENTS...................................................................................................v  

ABSTRACT .......................................................................................................................vi  

TABLE OF CONTENTS ................................................................................................ viii  

LIST OF FIGURES.............................................................................................................x  

LIST OF ABBREVIATIONS ...........................................................................................xii  

INTRODUCTION...............................................................................................................1  

General Roles of the Nutrient Sensors—AMPK, SIRT1, and FGF21........................1  

Goals of Thesis ............................................................................................................8  

METHODS........................................................................................................................13  

Animal Studies with Metformin Treatment ..............................................................13  

Animal Studies with Resveratrol Treatment .............................................................13  

Animal Studies with Liver-specific SIRT1 knockout (SIRT1 LKO) Mice ..............14  

Animal Studies with FGF21 Transgenic Mice..........................................................14  

  ix  

Animal Studies with In Vivo Adenoviral Gene Transfer..........................................14  

Body Composition Analysis and Comprehensive Laboratory Animal Monitoring

System (CLAMS)—Metabolic Cages.......................................................................15  

Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT) ............................15  

Hepatic Histology and Oil Red O Staining ...............................................................16  

Adipose Histology.....................................................................................................17  

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) ............................17  

Statistical Analysis ....................................................................................................17  

RESULTS..........................................................................................................................18  

DISCUSSION ...................................................................................................................39  

APPENDIX .......................................................................................................................47  

LIST OF JOURNAL ABBREVIATIONS ........................................................................50  

REFERENCES..................................................................................................................51  

CURRICULUM VITAE ...................................................................................................56  

  x  

LIST OF FIGURES

Figure Title Page 1

Metformin alleviates hepatic lipid accumulation in HFHS-fed mice.

16

2

Metformin improves glucose tolerance in HFHS-fed mice. 17

3 Metformin treatment promotes weight loss, reduces food intake, and improves body composition in HFHS-fed mice.

19

4 Metformin reduces adiposity in iWAT of diet-induced obese mice and increases UCP1 expression in iWAT of obese mice.

20

5 Metformin increases expression of thermogenic genes such as UCP1 and DIO2, as well as the co-regulatory protein PRDM16 in subcutaneous inguinal white adipose tissue of HFHS-fed mice.

20

6 Metformin increases energy expenditure during HFHS feeding.

21

7 Fuel utilization and physical activity of CHOW, HFHS, and HFHS+Metformin mice.

22

8 Resveratrol reduces impact of diet-induced obesity and improves the fasting blood glucose of HFHS-fed mice.

23

9 Resveratrol prevents the inhibition of hepatic AMPK in HFHS diet-induced obese mice.

24

10 Resveratrol improves hepatic steatosis in HFHS diet-induced obese mice.

11 Resveratrol enhances insulin sensitivity in the liver of diet-induced obese mice.

26

12 The effect of pharmacological SIRT1 activator, 27

  xi  

resveratrol, on body weight and food intake.

13

Resveratrol stimulates hepatic AMPK activation in a SIRT1-dependent manner.

28

14

Resveratrol alleviates fatty liver through SIRT1. 29

15 FGF21 transgenic mice have decreased total body weight and improved glucose tolerance when fed a CHOW-diet.

30

16

FGF21 transgenic (FGF21 TG) mice are resistant to diet-induced obesity, as demonstrated by reduced fat mass, and have improved glucose tolerance and insulin sensitivity.

32

17 FGF21 transgenic mice are resistant to HFHS diet-induced fatty liver.

33

18 Adenovirus-mediated overexpression of FGF21 in the liver reduces body weight of SIRT1 LKO mice fed a HFHS diet.

33

19 FGF21 improves hepatic steatosis in liver specific SIRT1 deficient mice on a HFHS diet.

34

20 Overexpression of FGF21 in the liver lowers fasting blood glucose levels.

35

21 Adenovirus-mediated overexpression of FGF21 in the liver restores glucose metabolism and protects against insulin resistance in HFHS diet-induced obese SIRT1 LKO mice.

36

22 A proposed working model for the mechanisms through which AMPK, SIRT1, and FGF21 modulate the progression of obesity, insulin resistance, NALFD, and alcoholic fatty liver disease.

47

  xii  

LIST OF ABBREVIATIONS

ACC............................................................................................... Acetyl-CoA carboxylase

ADP ..................................................................................................Adenosine diphosphate

ALD.................................................................................................. Alcoholic liver disease

AMP ...........................................................................................Adenosine monophosphate

AMPK ...................................................................................AMP-activated protein kinase

ANCOVA......................................................................................... Analysis of covariance

ANOVA................................................................................................ Analysis of variance

ATP ................................................................................................. Adenosine triphosphate

AUC..................................................................................................... Area under the curve

BAT .....................................................................................................Brown adipose tissue

BW.................................................................................................................... Body weight

CaMKKβ ........................................ Ca2+/calmodulin-dependent protein kinase kinase-beta

cDNA.......................................................................Complementary deoxyribonucleic acid

CLAMS ........................................ Comprehensive Laboratory Animal Monitoring System

DIO2................................................................................. Type 2 iodothyronine deiodinase

eWAT ................................................................................ Epididymal white adipose tissue

FGF21........................................................................................ Fibroblast growth factor 21

FGF21 TG ................................................................Fibroblast growth factor 21 transgenic

FGFR1c ....................................................................... Fibroblast growth factor receptor 1c

FOXO1 ..........................................................................................Forkhead box protein O1

  xiii  

GAPDH .......................................................... Glyceraldehyde 3-phosphate dehydrogenase

GFP............................................................................................... Green fluorescent protein

GLUT4 ........................................................................................Glucose transporter type 4

GTT ....................................................................................................Glucose tolerance test

G-6-Pase ...........................................................................................Glucose 6-phosphatase

HFHS..................................................................................................High fat, high sucrose

H&E..................................................................................................Hematoxylin and eosin

IHC ................................................................................................... Immunohistochemistry

IOM ..................................................................................................... Institute of Medicine

IRS1............................................................................................Insulin receptor substrate 1

ITT........................................................................................................Insulin tolerance test

iWAT...................................................................................... Inguinal white adipose tissue

LKB1 ........................................................................................................... Liver kinase B1

mRNA .......................................................................................Messenger ribonucleic acid

mTORC1 ..........................................................Mammalian target of rapamycin complex 1

NAD+.............................................................................Nicotinamide adenine dinucleotide

NAFLD..............................................................................Non-alcoholic fatty liver disease

NIAAA ..............................................National Institute on Alcohol Abuse and Alcoholism

NIH...........................................................................................National Institutes of Health

ORO.......................................................................................................................Oil Red O

PEPCK........................................................................ Phosphoenolpyruvate carboxykinase

PGC-1α ................ Peroxisome proliferator-activated receptor gamma coactivator 1-alpha

  xiv  

PKA ............................................................................................................ Protein kinase A

PPARα .................................................... Peroxisome proliferator-activated receptor alpha

PRDM16....................................................................................... PR domain containing 16

qPCR ....................................................................Quantitative Polymerase Chain Reaction

RER ............................................................................................Respiratory exchange ratio

RNA............................................................................................................Ribonucleic acid

RSV .....................................................................................................................Resveratrol

SEM............................................................................................ Standard error of the mean

Sir2 ........................................................................................ Silent information regulator 2

SIRT1 ...................................................................................................................... Sirtuin 1

SIRT1 LKO ....................................................................... Sirtuin 1 liver-specific knockout

SREBP-1 ......................................................... Sterol regulatory element-binding protein 1

TORC2 .................................................................Transducer of regulated CREB protein 2

UCP1 ...................................................................................................Uncoupling protein 1

VO2 ..........................................................................................................Volume of oxygen

VCO2 ...........................................................................................Volume of carbon dioxide

WAT......................................................................................................white adipose tissue

WT.........................................................................................................................Wild-type

1

INTRODUCTION

Nutrients are required to maintain energy homeostasis and sustain life. The ability

to adapt and respond to various nutrient fluxes is an ancient cellular function—an ability

conserved from unicellular to complex multicellular organisms, including mammals. At

the central level, the hypothalamus is the primary component of the nervous system that

interprets nutrient-related inputs. The hypothalamus then delivers hormonal and

behavioral responses with the goal of regulating energy intake and energy consumption.

At the molecular level, molecules or enzymes, referred to as nutrient/energy sensors,

mediate metabolic responses of specific tissues involved in energy balance (Toorie and

Nillni, 2014). The regulation and function of nutrient sensing pathways are vital because

the organism must maintain energy homeostasis even if the nutritional status of the

organism fluctuates between positive energy balance (overnutrition) and negative energy

balance (fasting or prolonged caloric restriction). To maintain energy homeostasis,

mammals adapt to changes in nutritional status through modulation of tissue-specific

metabolic pathways via nutrient sensors (Fulco, 2008).

General Roles of the Nutrient Sensors—AMPK, SIRT1, and FGF21

AMP-activated protein kinase (AMPK)

Nutrient sensors such as AMPK, SIRT1, and FGF21 play a major role in

maintaining energy homeostasis by assessing nutritional status and appropriately

regulating metabolic responses. AMPK is an energy-sensing protein kinase that is highly

conserved and regulates cellular metabolism. It is a heterotrimeric protein kinase

2

consisting of a catalytic α subunit and regulatory β and γ subunits. Each subunit has

multiple isoforms (α1, α2, β1, β2, γ1, γ2, γ3), but the predominant isoforms in the liver

are AMPKα1, β1, γ1, which account for over 90% of AMPK activity in hepatocytes.

Critical for the activation of this enzyme, its upstream kinases such as LKB1 and

CaMKKβ phosphorylate AMPKα subunits at Thr-172 (Kahn et al., 2005). In contrast,

phosphorylation of the α subunits at Ser-485/497 by PKA reduces α subunit

phosphorylation at Thr-172, thus decreasing AMPK enzyme activity (Xiao et al., 2007).

When cellular energy levels are low, the increase in the AMP/ATP or ADP/ATP

ratio results in a change in nucleotide binding to γ subunit and an allosteric change in

AMPK. The allosteric change leads to an increase in net phosphorylation of catalytic α

subunit at Thr-172 and subsequent AMPK activation. Activation of AMPK triggers the

repression of ATP-consuming anabolic pathways such as synthetic pathways of fatty

acid, triglyceride, cholesterol, glycogen and protein through the transcriptional regulation

of gluconeogenic and lipogenic enzymes. AMPK also elicits the activation of ATP-

producing catabolic pathways by stimulating glucose uptake and glycolysis, fatty acid

uptake and oxidation, leading to the regulation of glucose and fatty acid homeostasis

(Fulco, 2008) (Kahn et al., 2005). As a regulator of glucose metabolism, activated AMPK

acts via TORC2, a transcriptional coactivator of the cyclic AMP-regulated transcription

factor, CREB, to inhibit the transcription of key gluconeogenic enzymes such as PEPCK

and G-6-Pase (Shaw et al., 2005). As a regulator of lipid biosynthetic pathways, AMPK

phosphorylates and inactivates ACC, resulting in a drop in concentration of the ACC

product, malonyl CoA, an inhibitor of fatty acid entry into mitochondria for β-oxidation.

3

In other words, AMPK-caused inactivation of ACC leads to increased fatty acid

oxidation (Hardie et al., 2012). Furthermore, AMPK also appears to decrease lipogenesis

through downregulation of lipogenic genes, including SREBP-1 and fatty acid synthase

(Li et al., 2011). Chronic activation of AMPK may also induce the expression of muscle

hexokinase and GLUT4, resulting in enhanced glucose uptake and catabolism to generate

ATP—mimicking the effects of extensive exercise training (Zhou et al., 2001). AMPK

also plays a role in the regulation of food intake in the hypothalamus (Shaw et al., 2011).

As previously mentioned, AMPK is activated upon negative energy balance. But

a higher AMP/ATP ratio is not the only way to bring about activated AMPK’s favorable

metabolic effects. Exercise and several current anti-diabetic therapies, such as metformin,

activate AMPK in the liver and are thought to therapeutically act in part through the

stimulation of this pathway. Highlighting the importance of the AMPK pathway in the

therapeutic response to metformin in humans, investigators from the Diabetes Prevention

Program found that AMPK subunits and LKB1 genes were associated with a clinical

response to metformin (He and Wondisford, 2015). Metformin ameliorates

hyperglycemia without stimulating insulin secretion or causing hypoglycemia. It also has

beneficial effects on circulating lipids linked to increased cardiovascular risk (Zhou,

2001). Metformin is not metabolized in animals or humans and is eliminated intact

through renal excretion. Despite the fact that metformin has been used since 1957, the

mechanism(s) by which metformin lowers glucose and lipids remains an enigma (He and

Wondisford, 2015).

4

Although the fact that metformin activates AMPK is well accepted (Meng 2014,

Zang 2004, Zhou 2001), how metformin activates AMPK has remained controversial for

years. Does metformin activate AMPK directly? If so, how? Or, does metformin

indirectly activate AMPK via increasing the AMP/ATP or ADP/ATP ratio via inhibition

of complex 1 of the mitochondrial respiratory chain? (He and Wondisford, 2015)

Although many studies have confirmed that metformin’s actions are mediated through

the activation of AMPK, some metformin effects are reported to be AMPK-independent

such as inhibition of mTORC1 activity (Fulco, 2008). Varying results regarding how

metformin activates AMPK could be due to differences in metformin concentrations used

during studies. It has been suggested that lower concentrations of metformin directly

activates AMPK, while higher concentrations inhibit the mitochondrial respiratory chain

and indirectly activate AMPK. It has also been postulated that metformin may mediate

AMPK activation by promoting the binding of LKB1 to AMPK (He and Wondisford,

2015). Clearly, further research is needed to determine the molecular mechanisms

underlying the beneficial metabolic effects of metformin.

Sirtuin 1 (SIRT1)

SIRT1 is one of the seven mammalian orthologs of the yeast protein Sir2, an

NAD+-dependent protein deacetylase that extends life span. SIRT1 catalyzes a reaction

that couples lysine deactelyation to NAD+ hydrolysis. During this reaction, NAD+ is

hydrolyzed to nicotinamide and O-acetyl-ADP-ribose. Nicotinamide strongly inhibits

SIRT1 deacetylase activity (Guarente, 2006, 2012). A wealth of data has shown that

SIRT1 is a nuclear nutrient sensor that provides a molecular link between the cellular

5

energy status (via NAD+) and the adaptive transcriptional responses to nutrient status.

Accumulated evidence demonstrates that SIRT1 levels and/or activity increase with

nutrient deprivation and orchestrate the adaptive response to caloric restriction (Li 2013,

Chalkiadaki 2012)

During states of negative energy balance, the liver’s key function is to maintain

euglycemia through glucose production via the regulation of glycogenolytic and

gluconeogenic processes. To increase gluconeogenic capacity under fasting condition,

SIRT1 deacetylates PGC-1α and FOXO1, thereby increasing their ability to promote the

transcription of their gluconeogenic targets and inhibit the expression of glycolytic genes

such as glucokinase (Rodgers et al., 2005). In addition to its role in carbohydrate

metabolism, SIRT1 is also a key regulator of hepatic lipid homeostasis. SIRT1

deacetylates SREBP-1 and reduces SREBP-1’s affinity to the promoters of its lipogenic

target genes, thereby decreasing expression of lipogenic genes (Gillum, 2011). PGC-1α

is a key coactivator for PPARα signaling. By deacetylating and activating PGC-1α and

PPARα, SIRT1 is able to stimulate PPARα signaling and increase fatty acid oxidation.

Taken together, SIRT1 expression and/or activity are upregulated by fasting, and it

stimulates expression of genes involved in fatty acid oxidation and gluconeogenesis

while downregulating expression of lipogenic genes during prolonged fasting (Schug,

2011).

Resveratrol (RSV) is a natural polyphenolic compound mainly found in the skin

of grapes and is well known for its phytoestrogenic and antioxidant properties.

Resveratrol has been shown to significantly increase SIRT1 activity through an allosteric

6

interaction as well as enhance the affinity of SIRT1 to NAD+ and deacetylate its

substrate. By facilitating increased SIRT1 deacetylation, PGC-1α activity is induced and

leads to increased fatty acid oxidation (Rodgers et al., 2005). SIRT1 expression is

suppressed during insulin resistance and obesity and resveratrol treatment protects mice

against diet-induced obesity and insulin resistance (Lagouge, 2006). Since resveratrol has

beneficial metabolic changes—improved insulin resistance and lipid profiles, decreased

inflammation—SIRT1-based therapies hold promise for the treatment of insulin

resistance and type 2 diabetes (Gillum, 2011). However, there is much debate about

whether or not the metabolic effect of resveratrol acts via direct or indirect (or both)

activation of SIRT1. A moderate dose of resveratrol showed increased SIRT1 activity,

mitochondrial biogenesis and function, and AMPK activation. Meanwhile, a high dose of

resveratrol activated AMPK in a SIRT1-independent manner, demonstrating that

resveratrol dosage is a critical factor (Price, 2012). Further studies are needed to elucidate

whether resveratrol acts on SIRT1 to prevent metabolic disease.

Fibroblast Growth Factor 21 (FGF21)

The human/mouse FGF family comprises FGF1-FGF23 (Itoh 2014). Based on

their mechanisms of action, FGFs can be classified as paracrine, intracrine, and

endocrine. Endocrine FGFs comprise of FGF15/19, FGF21, and FGF23 and mainly

function as hormone-like or, in part, local signaling molecules in metabolism

(Kharitonenkov and Larsen, 2011; Kharitonenkov and Shanafelt, 2009; Kharitonenkov et

al., 2005). Although most of paracrine and endocrine FGFs have proliferative activities,

FGF21 is unique. FGF21 has metabolic, but not proliferative activities (Itoh 2014).

7

FGF21 is a secretory protein that is predominantly expressed in the liver and exerts

beneficial effects on obesity and related metabolic diseases. As an endocrine hormone,

FGF21 acts on the FGF receptor 1c (FGFR1c) with β-Klotho as a necessary cofactor.

Both FGFR1c and β-Klotho are specifically expressed in their target tissues—liver,

WAT, BAT, skeletal muscle, pancreas, heart, and brain (Ye et al., 2014).

The physiological roles of FGF21 include the maintenance of energy homeostasis

in conditions of metabolic or environmental stress. The expression of FGF21 is induced

in multiple major metabolic organs including the liver, WAT, BAT, and skeletal muscle,

in response to stressors—starvation, nutrient excess, exercise, and cold exposure (Kim et

al., 2014). Since FGF21 acts as a critical regulator in the metabolic adaptation to fasting,

it may seem odd that FGF21 is induced in both undernutrition and overnutrition states.

However, FGF21 signaling is impaired in the liver and WAT of obese insulin-resistant

mice, which suggests that obesity might be a state of FGF21 resistance (Kim et al., 2014).

Administration of exogenous FGF21 overcomes this resistance and leads to

improvements in obesity-related metabolic deterioration. Transgenic mice overexpressing

FGF21 exhibit resistance to diet-induced obesity and tight glycemic control (Kim et al.,

2014). Insulin resistance develops due to aberrant accumulation of intracellular lipids in

insulin-responsive tissues, including ceramide. FGF21 diminishes the accumulation of

ceramides in obese animals (Itoh 2014). WAT is the predominant site that confers the

metabolic activities of FGF21. FGF21 enhances lipolysis, glucose uptake, and fatty acid

oxidation, decreases fatty acid synthesis, and improves insulin resistance. Improved

8

metabolic parameters of obese diabetic rodents, rhesus monkeys, and human subjects

treated with FGF21 have also been observed (Kim et al., 2014).

FGF21 doesn’t only act as a hepatokine. Adipose tissue can also secrete FGF21

and as an adipokine, it plays a big role in inducing thermogenesis during cold exposure or

β-adrenergic receptor stimulation. UCP1 releases chemical energy as heat in BAT and is

a way for excess energy to be wasted. Beige adipocytes are brown adipocyte-like cells,

and are UCP1-positive adipocytes in WAT. Upon cold exposure, FGF21 acts in an

autocrine/paracrine manner to activate brown adipocytes and induce the accumulation of

beige adipocytes in WAT, thereby increasing thermogenesis and energy expenditure

(Fisher et al., 2012). FGF21 has only recently emerged as a critical metabolic regulator

and future studies are necessary to investigate its effects and mechanisms. Because the

activation of pathways via AMPK, SIRT1, and FGF21 may translate into favorable

metabolic effects, these particular nutrient sensors seem to be viable therapeutic targets.

Goals of Thesis

Identifying potential drug targets for modulation of metabolic diseases has

become a high priority because consumption of calorie-dense foods coupled with a

sedentary lifestyle has propagated obesity throughout the developed world. According to

the Institute of Medicine, adults should get 45-65% of their daily calories from

carbohydrates, 20-35% from fat, and 10-35% from protein. Unfortunately, the

consumption of a high-fat, high sucrose (HFHS) diet with ~60% of calories from fat has

become more common and has greatly contributed to the increase in the number of obese

individuals. Globally, obesity has more than doubled since 1980 (WHO). In the U.S.,

9

approximately 35% of adults are obese. An additional 34% of adults are overweight

(CDC). This is particularly alarming because in obese individuals, excess adipose tissue

releases increased levels of circulating free fatty acids. These excess fatty acids deposit in

metabolic tissues such as liver, muscle, and pancreatic β-cells, resulting in non-alcoholic

fatty liver disease (NAFLD), insulin resistance, and type 2 diabetes (Schug, 2011).

Metformin is the most commonly prescribed oral anti-diabetic agent worldwide

and is taken by over 150 million of diabetic patients each year. Numerous studies in

rodents have highlighted metformin’s ability to promote weight loss (Hu 2014, Matsui

2010, Yasuda 2004). Furthermore, systematic reviews and meta-analyses of studies

involving metformin have overwhelmingly demonstrated metformin’s weight loss

capabilities in humans (Domecq, 2015). Metformin’s powerful ability to combat obesity

and bring about weight loss has made researchers and clinicians consider the possibility

of metformin as treatment outside diabetes. Polycystic ovary syndrome is known to be

associated with significant weight gain. Patients with polycystic ovary syndrome were

given lifestyle modifications to live by plus metformin, and these patients experienced

greater weight loss than patients who only made lifestyle modifications (no metformin

treatment) (Naderpoor et al., 2015). Despite the vast amount of evidence supporting

metformin’s weight loss capabilities, the mechanism(s) in which metformin leads to

weight loss remains unknown. Some studies indicate weight loss results from decreased

caloric consumption (Matsui 2010, Yasuda 2004), while other studies report metformin

brings upon beneficial metabolic effects without influencing food intake (Oh 2013,

Tajima 2013).

10

Postulating ways in which metformin reduces body weight has led researchers to

question if metformin has effects on energy expenditure. Studies in zebrafish showed that

insulin increases metabolic rate and that metformin, being an insulin-sensitizer, increases

the metabolic response to insulin by 18.9% (Renquist et al. 2013). Sulphonylurea therapy

has been reported to cause weight gain in type 2 diabetic patients. Resting metabolic rate

of sulphonylurea-treated and metformin-treated patients was evaluated and results

revealed both groups to have similar resting metabolic rates. However, when adjusted for

fat free mass (lean mass), metformin-treated patients demonstrated an increased

metabolic rate compared to sulphonyluera-treated patients (Chong et al., 1995).

Metformin prevented olanzapine-induced weight gain in rats [olanzapine is an anti-

psychotic associated with weight gain]. But, what’s more intriguing is

olanzapine+metformin increased UCP1 gene expression in BAT compared to olanzapine

alone (Hu et al., 2014). Additional evidence suggests metformin increases the

thermogenic activity of BAT: Metformin promotes VLDL-triglyceride clearance by BAT

resulting in lower plasma triglycerides (Geerling et al., 2014). PRDM16 is involved in

the development and function of classical brown and beige adipocytes. PRDM16 was

decreased in obese subjects; but, metformin treatment increased PRDM16 mRNA and

protein levels in isolated human adipocytes and in adipose tissue (Moreno-Navarrete et

al., 2015). Collectively, these studies suggest metformin may promote weight loss by

increasing energy expenditure and burning of excess energy. I’m interested in

determining if metformin can cause weight loss in HFHS diet-induced obese mice and if

this weight loss is due to increased energy expenditure.

11

Studies revealing metformin’s numerous effects have opened the possibility for it

to treat other metabolic disorders other than diabetes. Metformin has recently been

proposed to help treat non-alcoholic fatty liver disease (NAFLD) associated with obesity

since currently, there is no treatment for NAFLD other than lifestyle modification. As of

now, the efficacy of metformin to alleviate NAFLD is difficult to evaluate since systemic

analyses of studies showing improved NAFLD combine metformin and lifestyle

modifications (Mazza et al., 2012). I aim to evaluate the efficacy of metformin to

alleviate obesity-induced NAFLD.

Exercise promotes weight loss, which improves NAFLD in obese individuals

(Mazza et al., 2012). Recently, there has been much attention given to AMPK activators

as exercise mimetics. Metformin is known to activate AMPK and similarly brings about

many beneficial effects as exercise (Meng 2014, Zang 2004, Zhou 2001). It’s been shown

that AMPK and SIRT1 have intertwined roles in regulating different metabolic processes.

Numerous studies have shown that resveratrol, a known SIRT1 activator, helps prevent

weight gain (Gillum, 2011). Transgenic mice overexpressing FGF21 exhibit resistance to

diet-induced obesity (Kim, et al., 2014). Nutrient sensors AMPK, SIRT1, and FGF21 all

have implications in weight loss and in the improvement of metabolic derangements.

Since metformin activates AMPK, and AMPK and SIRT1 have intercrossing roles to

repress lipogenesis by reducing SREBP-1 levels, I’m interested to see if SIRT1

stimulation by resveratrol is able to mimic metformin’s amelioration of fatty. FGF21 is

postulated to act downstream of SIRT1, therefore I’m curious to see if exogenous

administration of FGF21 mimics stimulated SIRT1’s effect on NAFLD.

12

This thesis aims to investigate whether:

1) metformin facilitates weight loss by increasing energy expenditure via increased

expression of UCP1, DIO2, and PRDM16

2) metformin is capable of alleviating NAFLD cause by HFHS diet-induced obesity

3) SIRT1 stimulation by resveratrol and delivery of exogenous FGF21 mimics

metformin’s ability to combat obesity and improve NAFLD

13

METHODS

[Figures of experimental designs for animal studies can be found in Appendix]

Animal Studies with Metformin Treatment

Male C57BL6 mice matched for age were divided into three groups: one group

(n = 7) of mice was fed a CHOW diet consisting of 18% of calories from fat, 58% from

carbohydrates, 24% from protein (Teklad Global #2918, Madison, WI) ad libitum for 12

weeks. The second group (n = 7) was fed a HFHS diet consisting of 59% of calories from

fat, 26% from carbohydrates, 15% from protein (BioServ #F1850, Frenchtown, NJ) ad

libitum for 12 weeks. The third group (n = 7) was fed a HFHS diet ad libitum for 8

weeks, followed by 4 weeks of HFHS feeding plus metformin treatment (250 mg/kg/day)

via drinking water. Prior to metformin administration, water intake of the metformin

treated group was measured and the proper dosage of metformin was administered based

on the average daily water intake. At the time of sacrifice, serum, liver, epididymal white

adipose tissue (eWAT), inguinal white adipose tissue (iWAT), and interscapular brown

adipose tissue (BAT) were collected. Tissues were either fixed in buffered formalin and

processed for histology or snap frozen in liquid nitrogen and stored at -80oC until RNA

and proteins were isolated.

Animal Studies with Resveratrol Treatment

Male, age-matched C57BL6 mice were divided into three groups: one group

(n = 5) was fed CHOW ad libitum for 20 weeks. The second group (n = 5) was fed a

HFHS diet for 20 weeks. The third group (n = 8) was fed a HFHS diet supplemented with

14

resveratrol for 20 weeks. The HFHS diet containing resveratrol provided a dose of

130 mg/kg/day, which was calculated based on the average daily consumption of food

(5 g/day) by a 23 g mouse.

Animal Studies with Liver-specific SIRT1 knockout (SIRT1 LKO) Mice

Male wild-type and SIRT1 LKO mice were age-matched and separated into four

groups. Wild-type (n = 5) and SIRT1 LKO mice (n = 5) were fed a HFHS diet for 12

weeks. Additional wild-type (n = 5) and SIRT1 LKO mice (n = 5) were fed a HFHS diet

supplemented with resveratrol (130 mg/kg/day) for 12 weeks.

Animal Studies with FGF21 Transgenic Mice

Male wild-type and FGF21 TG on C57BL6 background were age-matched and

separated into four groups. Wild-type (n = 5) and FGF21 TG (n = 4) were fed CHOW ad

libitum for 12 weeks. Additional wild-type (n = 5) and FGF21 TG (n = 5) were fed a

HFHS diet for 12 weeks.

Animal Studies with In Vivo Adenoviral Gene Transfer

Adenovirus-mediated overexpression of FGF21 in the liver of HFHS-fed SIRT1

LKO mice was accomplished through intravenous injection. Two hundred microliters of

adenovirus (5 x 109 – 1 x 1010 pfu) per mouse were injected into the tail vein using a 1mL

syringe with a 255/8-gauge needle. Wild-type and SIRT1 LKO mice were fed a HFHS

diet for a total of 12 weeks. At week 10 (two weeks prior to the end of feeding period),

wild-type (n = 5) and SIRT1 LKO mice (n = 5) were injected into the tail vein with

15

Ad-GFP, and SIRT1 LKO mice (n = 8) were injected intravenously with Ad-FGF21.

Two weeks after injection, each group of mice was sacrificed.

Body Composition Analysis and Comprehensive Laboratory Animal Monitoring

System (CLAMS)—Metabolic Cages

All mice were weighed at the beginning of the HFHS diet and weekly thereafter

until sacrificed. Body composition and CLAMS were performed in the Metabolic

Phenotyping Core at Boston University. EchoMRI700, a non-invasive quantitative

magnetic resonance system, determined body composition, including fat mass, lean mass,

free water, and total body water. To acclimatize to being alone in a cage, mice going into

the metabolic cages (CLAMS) were housed individually for one week prior to CLAMS

experiments. CLAMS was used to measure metabolic rate, fuel utilization, and physical

activity. Energy expenditure was assessed by indirect calorimetry measurements.

Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT)

GTT and ITT were performed to assess whole body glucose homeostasis and

systemic insulin sensitivity, respectively. Blood glucose concentrations were measured

using the AlphaTrak Blood Glucose Monitoring System for Animals. After 16 hours of

fasting, GTTs were performed with an intraperitoneal injection of 20% glucose solution

(2 g/kg BW) into the mice and blood glucose levels were measured at 0, 15, 30, 60, 90,

and 120 minutes. After 5 hours of fasting, ITTs were performed with an intraperitoneal

injection of human insulin—Humulin R—(0.75 U/kg BW) into mice and blood glucose

levels were measured at 0, 15, 30, 60, 90, and 120 minutes.

16

Hepatic Histology and Oil Red O Staining

The livers from the mice were fixed for 48 hours in buffered formalin, processed,

embedded in paraffin, cut into 5µm sections and placed on glass slides. Hematoxylin and

eosin (H&E) staining was conducted to demonstrate general hepatic morphology.

When animals were killed, pieces of the liver were cut and embedded in ornithine

carbamyl transferase (OCT) and frozen over dry ice. The cryosectioner was used to cut

the livers embedded in OCT blocks. This process was tedious and only improved with

practice and knowledge of optimal temperature for different liver tissues. Healthy livers

were best cut at -16 to -17oC. Fatty livers were best cut at -13oC. When the temperature

of the crysectioner is too cold, the tissue will have lots of cracks when cut. At the same

time, when the temperature of the cryosection is too warm, the OCT will revert back to

its liquid form and tissue will be susceptible to degrading and will negatively affect

cellular structures.

Hepatic steatosis was assessed using the Oil Red O Stain from American

MasterTech. Getting good staining results proved to be quite a difficult task. I tried out

different protocols using different reagents like 60% propylene glycol, however the

staining results were not good. After many times of trial and error, successful Oil Red O

staining was achieved by exposing cryosections to 60% isopropanol for 30 seconds,

stained with 0.3% Oil Red O in 60% isopropanol for 25 minutes, and quickly washed

with 60% isopropanol (~30 seconds). Sections were counterstained with Harris

Hematoxylin and mounted with aqueous solution.

17

The liver images were photographed and scanned digitally, and planimetry of Oil

Red O positive stained areas were performed on the digitized images using NIH ImageJ

software. Relative areas of steatosis were quantified using ImageJ and expressed as the

percentage of area stained with Oil Red O.

Adipose Histology

Pieces of iWAT and BAT of mice were fixed for 48 hours in buffered formalin,

processed, embedded in paraffin, cut into 7µm sections and placed on glass slides. H&E

staining was conducted to demonstrate general adipose morphology.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Interscapular BAT and inguinal WAT were collected from mice. RNA was

extracted using the RNeasy Lipid Tissue Mini Kit (QIAGEN), and cDNA was generated

with the QuantiTect Reverse Transcription Kit (QIAGEN) according to the

manufacturer’s instructions. First-strand cDNA was generated by reverse-transcription

using total RNA. RT-PCR was performed using the ViiA TM7 Real Time PCR System

with the SYBR Green Real Time PCR Master Mixes (Applied Biosystems) according to

the manufacturer’s instructions. The relative levels of mRNA were determined using the

ΔΔCT Method, normalizing to GAPDH, and expressed as relative mRNA levels.

Statistical Analysis

Data are expressed as means ± SEM. The significance of the differences in the

mean values was evaluated by using ANOVA, ANCOVA, or 2-tailed Student’s t test.

Values of P < 0.05 were considered to be statistically significant.

18

RESULTS

Metformin Improves Fatty Liver and Restores Glucose Tolerance and Glucose

Homeostasis in Diet-induced Obese Mice

Overnutrition-induced obesity is associated with the development of NAFLD.

Compared to CHOW-fed mice, HFHS-fed mice exhibited severe hepatic steatosis.

Metformin treated HFHS-fed mice displayed significant reductions in steatosis compared

to HFHS-fed mice, as demonstrated by a ~60% reduction in Oil Red O-stained areas

(Fig. 1).

To determine the effect of metformin on systemic glucose homeostasis,

intraperitoneal GTT was performed. Metformin treated HFHS-fed mice displayed an

Figure 1. Metformin alleviates hepatic lipid accumulation in HFHS-fed mice. A. 20X B. 40X C. ORO positive % area * P < 0.05 vs. CHOW, # P < 0.05 vs. HFHS  

19

~34% reduction in fasting glucose levels, compared with those of HFHS-fed mice

(105.9 ± 4.07 vs. 161.0 ± 4.08 mg/dl). In fact, such lowered glucose levels in metformin-

treated mice were comparable to those of normal control mice (Fig. 2). Consistently,

integrated blood glucose concentrations, as calculated by the area under the curve (AUC),

were significantly decreased by ~40% in metformin-treated mice (Fig. 2).

Metformin Decreases Adiposity and Improves Body Composition in HFHS-fed Mice

Metformin treatment effectively caused body weight loss (Fig. 3A). Other studies

have report that metformin suppresses food consumption in mice (Matsui 2010, Yasuda

2004). While housed individually for a week to acclimatize mice for being alone in

metabolic cages, food intake of all mice in the study was recorded to help determine

whether the weight loss of the metformin-treated mice was due to metformin’s ability to

constrain hyperphagia (thus decrease calorie intake). Food intake measurements show

Figure 2. Metformin improves glucose clearance in HFHS-fed mice. GTTs were performed on mice following a 16 hour fast. Data presented as means ± SEM (n = 7). * P < 0.05 vs. CHOW, # P < 0.05 vs. HFHS

20

that metformin-treated HFHS-fed mice consumed fewer calories daily compared to

CHOW-fed and HFHS-fed mice (Fig. 3C). It’s possible that this decreased caloric

ingestion could be responsible for the decreased body weight of the metformin-treated

mice (Fig. 3B). HFHS-induced obese mice had increased adipose tissue mass, both in

Figure 3. Metformin treatment promotes weight loss, reduces food intake, and improves body composition in HFHS-fed mice. A. Body weight change over treatment period B. Final body weight C. Food intake D. Fat and lean mass E. Percent fat and lean mass Data presented as means ± SEM (n = 7). * P < 0.05 vs. CHOW, # P < 0.05 vs. HFHS

21

terms of absolute fat mass and fat weight to body weight ratio. Along with lower body

weight, metformin-treated mice had decreased fat mass, lower percent body fat, and

higher percent lean mass compared to their HFHS-fed counterparts (Fig. 3D-E). Thus,

metformin promotes weight loss and fat mass reduction.

To elucidate the physiological mechanism underlying metformin action on fat

tissue, H&E staining and IHC analysis of inguinal white adipose tissue (iWAT) and

brown adipose tissue (BAT) were performed. Multiple slides with H&E staining of

iWAT sections showed enlarged adipocytes as well as the development of crown-like

structures in diet-induced obese mice, compared to CHOW-fed mice. Looking at stained

H&E slides, metformin treatment appears to reduce adipocyte size as well as exhibit

normal cellular structures (Fig. 4A). To ensure the conclusion that metformin reduces

adipocyte size is valid, adipocyte cell volume of HFHS-fed and HFHS+Metformin-fed

mice needs be quantified and analyzed in the future. Metformin-treated mice did not

exhibit enlarged unilocular lipid droplets in the BAT, which were observed in the BAT of

mice after HFHS feeding (Fig. 4B). Interestingly, metformin-treated ob/ob mice

exhibited adipocytes expressing the key thermogenic gene, UCP1, indicating the presence

of beige adipocytes in iWAT (Fig. 4C). Metformin might play a role in the regulation of

adipose tissue functions during obesity.

Metformin Induces the Expression of Thermogenic Genes and Promotes Browning of

iWAT by Inducing PRDM16 Expression

Since subcutaneous iWAT is particularly prone to inducing a thermogenic gene

program (browning), we characterized the effects of metformin on fat tissue function.

22

Gene expression of UCP1 was enriched in iWAT of metformin-treated mice (Fig. 5A),

consistent with the presence of beige adipocytes in these pads. Expression of DIO2,

another thermogenic gene, was also upregulated in the iWAT of mice treated with

metformin (Fig. 5B). PRDM16 is involved in the development and function of classical

brown and beige adipocytes. To determine whether metformin might be involved in the

remodeling of white adipose tissues, mRNA levels in subcutaneous iWAT were assessed.

PRDM16 was significantly elevated in the iWAT of metformin-treated mice relative to

the very low levels of expression in iWAT of HFHS-fed mice (Fig. 5C).

Metformin Does Not Seem to Increase Energy Expenditure in HFHS-fed mice

Since metformin promoted whole body weight loss, decreased lipid accumulation

in both liver and white adipose tissues, and increased expression of thermogenic genes in

iWAT, metformin’s effects on energy expenditure was also investigated. When each

mouse was plotted, energy expenditure increased as body weight increased (Fig. 6A).

Energy expenditure also increases proportionally to lean mass (Fig. 6B). ANCOVA

statistical analysis was done to investigate whether HFHS-fed mice and

HFHS+Metformin-fed mice had any differences in energy expenditure that was higher

than expected based on their body weight or lean mass. Unadjusted energy expenditure

(average over 24h) means for HFHS and HFHS+Metformin were 0.700 and 0.568 kcal/hr

per mouse, respectively. The metformin-treated group weighed ~20g less than HFHS-fed

mice (Fig. 3B). After adjusting energy intake for body weight, the energy expenditure

was similar: HFHS (0.62 kcal/hr) and HFHS+Metformin (0.64 kcal/hr), N.S, P = 0.43.

ANCOVA analysis resulted in an aggregate correlation within samples of r2 = 0.83,

23

Figure 5. Metformin increases expression of thermogenic genes such as UCP1 (A) and DIO2 (B) as well as the co-regulatory protein PRDM16 (C) in subcutaneous inguinal white adipose tissue of HFHS-fed mice. Thermogenic gene expression normalized to CHOW-fed mice. (n = 5-7) # P-value < 0.05 vs. CHOW

Figure 4. Metformin reduces adiposity in iWAT of diet-induced obese mice and increases UCP1 expression in iWAT of obese mice. Representative H&E staining of A. iWAT and B. BAT in mice fed a CHOW, HFHS, and HFHS+Metformin diet. C. Metformin increases UCP1 expression in iWAT of ob/ob mice.  

24

meaning HFHS feeding, metformin, and body weight together explains 83% of the

variability in energy expenditure. When energy expenditure was adjusted for lean mass,

ANCOVA analysis revealed that the adjusted means were slightly higher in the HFHS

(0.67 kcal/hr) than the HFHS+Metformin (0.6 kcal/hr) group (P = 0.03). Aggregate

correlation within samples was r2 = 0.48, meaning HFHS feeding, metformin, and lean

mass together explains 48% of the variability in energy expenditure. The fact that body

weight better predicts energy expenditure than lean mass implies that fat mass only

contributes to total energy expenditure.

Graphs depicting physical activity of the three groups (Fig. 7B-D) shows that

physical activity was similar between the three groups and that weight loss observed in

metformin-treated mice is not due to increased physical activity. Consistent with what is

already known about mice, physical activity graphs show that mice are more active

during the hours of 7pm and 7am (dark cycle).

Figure 6. Energy expenditure is proportional to body weight. A-B. Data depicted in terms of means per mouse.

25

RER measurements resulted in the following means: CHOW (0.922), HFHS

(0.795), and HFHS+Metformin (0.783). ANOVA analysis of RER revealed no significant

difference in fuel utilization (fat vs. carbohydrate) between HFHS-fed and

HFHS+Metformin-fed mice but both were significantly different from the higher RER

observed in CHOW-fed mice, as expected because CHOW is higher in carbohydrate

(P = 0.01) (Fig. 7A).

Figure 7. Fuel utilization and physical activity of CHOW, HFHS, and HFHS+Metformin mice. A-D. Data depicted in terms of means per group.

26

Resveratrol Attenuates Diet-induced Obesity in Mice While Improving Fasting Blood

Glucose Levels

Overnutrition leads to obesity and along with obesity is insulin resistance. This is

evident in the increased body weight and increased fasting blood glucose levels of the

HFHS-fed group. Mice treated with resveratrol had a modest decrease in total body

weight (Fig. 8A). Interestingly, resveratrol lowered fasting blood glucose levels nearly to

normal levels (Fig. 8B).

Resveratrol Stimulates Hepatic AMPK and Alleviates Fatty Liver in Obese Mice

AMPK also exerts effects on lipid metabolism, largely via ACC. Activated

AMPK triggers phosphorylation of ACC, which in turn deactivates ACC and fatty acid

synthesis while promoting fatty acid oxidation (Li et al., 2011). To investigate the role of

resveratrol and activated SIRT1 on AMPK activity and AMPK’s downstream targets

(specifically ACC), mice were fed CHOW, HFHS, or HFHS+RSV (130mg/kg/day) for

20 weeks. Consistent with our previous studies (Li et al., 2011), phosphorylation of

Figure 8. Resveratrol reduces impact of diet-induced obesity and improves the fasting blood glucose of HFHS-fed mice. A. Body weight B. Fasting blood glucose (md/dL). Data presented as means ± SEM (n = 5-8). * P-value < 0.05 vs. CHOW, # P-value < 0.05 vs. HFHS

27

AMPK at the Thr-172 site and consequently phosphorylated ACC were decreased in

HFHS-fed mice (Fig. 9A-B). Increased hepatic lipid accumulation was evidenced by

                                 Figure 9. Resveratrol prevents the inhibition of hepatic AMPK in HFHS diet-induced obese mice. A. Hepatic AMPK and ACC activity as measured by the amount of pAMPK and pACC, respectively B. Bar graphs showing pAMPK and pACC levels * P < 0.05 compared to CHOW, # P < 0.05 compared to HFHS

Figure 10. Resveratrol improves hepatic steatosis in HFHS diet-induced obese mice. A. 20X B. 40X C. ORO positive % area * P < 0.05 compared to CHOW, # P < 0.05 compared to HFHS  

28

increased positive Oil Red O-staining (Fig. 10A-C). Consistent with increased AMPK

activity, hepatic steatosis was attenuated by resveratrol (Fig. 10A-C).

Resveratrol-Treatment Enhances Insulin Signaling in the Liver of HFHS-fed Mice

Insulin acts via phosphorylation regulation of Insulin Receptor Substrate 1 (IRS1)

and Akt, aka Protein Kinase B (Gual, 2005). Since resveratrol increased hepatic AMPK

signaling, it was postulated that impaired insulin signaling of HFHS-fed mice would also

be attenuated by resveratrol. Compared to the HFHS-fed mice, the resveratrol-treatment

decreased IRS1 serum phosphorylation and increased Akt phosphorylation in the liver

(Fig. 11), which is indicative of enhanced hepatic insulin signaling.

Figure 11. Resveratrol enhances insulin sensitivity in the liver of diet-induced obese mice. A-B. Protein expression of hepatic phosphorylation of Akt and IRS1 B-C. Bar graphs show changes in phosphorylated Akt and phosphorylated IRS1 * P-value < 0.05 compared to CHOW, # P-value < 0.05 compared to HFHS

29

Resveratrol Stimulates Hepatic AMPK Activity and Alleviates Fatty Liver in HFHS

Diet-induced Obese Mice through SIRT1

The functional importance of SIRT1 for resveratrol metabolic action has not been

examined. To investigate whether resveratrol stimulates of AMPK through SIRT1, wild

type and SIRT1 LKO mice were fed either a HFHS or HFHS+RSV (130mg/kg/day) diet

for 12 weeks. HFHS-fed SIRT1 LKO mice have pronounced weight gain compared to

HFHS-fed wildtype (Fig. 12A-B). Food intake measurements were done to assess if this

difference in body weight is due to differences in caloric consumption. However, food

Figure 12. The effect of pharmacological SIRT1 activator, resveratrol, on body weight and food intake. A. Body weight gain B. Final body weight. C. Food intake

30

intake was not significantly different among the four groups (Fig. 12C). When fed a

HFHS diet, SIRT1 LKO mice demonstrate decreased p-AMPK and p-ACC compared to

wild-type mice (Fig. 13). SIRT1 LKO’s decrease in AMPK activity and increase in ACC

activity coincides with SIRT1 LKO’s worsened hepatic steatosis when compared to wild-

type (Fig. 14). Together, these results suggest that SIRT1 helps reduce the impact of

HFHS diet-induced obesity on the fatty liver of mice.

Resveratrol treatment increased phosphorylation of AMPK and in turn, increased

phosphorylation of ACC in wild type mice fed on the HFHS diet (Fig. 13). In line with

this, resveratrol-induced increase in AMPK activity was associated with improved

hepatic steatosis in these mice, as reflected by decreased Oil Red O stained areas

(Fig. 14). However, the ability of resveratrol to simulate AMPK and reduce fatty liver

Figure 13. Resveratrol stimulates hepatic AMPK activation in a SIRT1-dependent manner. A. Protein expression of phosphorylated AMPK & ACC B-C. Bar graphs shows changes in AMPK and ACC phosphorylation. * P < 0.05 vs. WT HFHS, # P < 0.05 vs. WT HFHS+RSV  

31

was diminished in SIRT1 LKO mice. This indicates that SIRT1 is at least partially

necessary for the beneficial effect of resveratrol on obesity-induced fatty liver.

Overexpression of FGF21 in Mice Mimics Resveratrol’s Ability to Improve Fatty Liver

Figure 14. Resveratrol alleviates fatty liver through SIRT1. A. 20X B. 40X C. ORO positive % area * P < 0.05 vs. HFHS-fed wild-type mice, # P < 0.05 vs. resveratrol treated wild-type mice

32

As shown above, resveratrol, a SIRT1 activator, alleviates HFHS diet-induced

obesity and fatty liver disease. Recent studies implicate FGF21 in the regulation of body

weight and lipid metabolism (Coskun 2008, Xu 2009). To test the hypothesis that FGF21

functions downstream from SIRT1 as a modulator of obesity- induced NAFLD and

insulin resistance, wild type and FGF21 transgenic mice were fed either a CHOW or

HFHS diet for 12 weeks.

Figure 15. FGF21 transgenic mice have decreased total body weight and improved glucose tolerance when fed a CHOW-diet. A. Body weight B. Fat and lean mass C. Percent fat and lean mass D. GTTs were performed following a 16 h fast. E) AUC for GTT Data presented as means ± SEM (n = 4-5). * P < 0.05 vs. WT group

33

Figure 16. FGF21 transgenic (FGF21 TG) mice are resistant to HFHS diet-induced obesity, as demonstrated by reduced fat mass, improved glucose tolerance, and insulin sensitivity. GTTs were performed following a 16-hour fast. ITTs were performed after a 5-hour fast. Data presented as means +- SEM (n = 5 per group). * P < 0.05 vs. WT mice

34

When fed a CHOW diet, FGF21 transgenic mice have a lower total body weight

but their percent body fat and percent lean mass are similar to that of wild-type mice

(Fig. 15A-C). Intraperitoneal GTTs demonstrated that CHOW-fed FGF21 transgenic

mice had slightly improved glucose tolerance (Fig. 15D-E). Interestingly, HFHS feeding

revealed major differences between the two genotypes. FGF21 transgenic mice were

resistant to diet-induced obesity and insulin resistance, and remained lean throughout the

whole 12 week HFHS feeding. Body composition measurements revealed that FGF21

transgenic mice have a dramatically decreased fat mass with a significantly reduced

percent body fat (Fig. 16A-C).

Intraperitoneal GTTs showed that HFHS-fed FGF21 transgenic mice have

significantly lower glucose levels and improved glucose tolerance, compared to

HFHS-fed wild-type mice (Fig. 16D). Consistently, the AUC for GTT was significantly

decreased in mice overexpressing FGF21 in the liver (Fig. 16E). FGF21 transgenic mice

Figure 17. FGF21 transgenic mice are resistant to HFHS diet-induced fatty liver. A. 20X B. ORO positive % area * P-value < 0.05 compared to wild-type CHOW, # P-value < 0.05 compared to wild-type HFHS

35

have a significantly lower AUC for ITT compared to HFHS-fed wild-type, showing

evidence for improved insulin resistance in FGF21 transgenic mice (Fig. 16F-G).

In line with improved glucose tolerance and insulin resistance, FGF21 transgenic

mice have alleviated hepatic steatosis despite being on a HFHS diet. There is a dramatic

reduction in Oil Red O-stained areas in FGF21 transgenic mice (Fig. 17).

FGF21 Rescues the Increased Susceptibility to Weight Gain and Fatty Liver of

Hepatocyte-Specific Deletion of SIRT1 in Mice

The HFHS diet-induced obesity of the SIRT1LKO mouse model is characterized

by increased fat mass, insulin resistance, and hepatic steatosis even worse than its wild

Figure 18. Adenovirus-mediated overexpression of FGF21 in the liver reduces body weight and fat mass in SIRT1 LKO mice fed a HFHS diet. A. Body weight B. Fat and lean mass C. Percent fat and lean mass Data presented as means ± SEM (n=5-7) * P-value < 0.05 compared to WT+Ad-GFP, # P-value < 0.05 compared to SIRT1 LKO+Ad-GFP

36

type counterparts. FGF21 has emerged as important regulator of body weight and given

that fatty liver is associated with FGF21 insufficiency in SIRT1 LKO mice (Li 2014,

Kahn 2005), rescue experiments with adenovirus encoding FGF21 (Ad-FGF21) were

performed to determine the effect of FGF21 gain-of-function on the body weight and

steatotic phenotype of SIRT1 LKO mice.

Ad-FGF21 injected SIRT1 LKO mice had a modest weight loss resulting in a

final body weight similar to wild-type Ad-GFP injected mice (Fig. 18A). Besides

improvement in whole body weight, SIRT1 LKO mice injected with Ad-FGF21 exhibited

improvements on body composition. Ad-FGF21-injected mice had decreased percent fat

Figure 19. FGF21 improves hepatic steatosis in SIRT1 LKO mice on a HFHS diet. A. 20X B. 40X C. ORO positive % area * P < 0.05 compared to WT+Ad-GFP, # P < 0.05 compared to SIRT1 LKO+Ad-GFP

37

mass (Fig. 18C). Fatty liver in SIRT1LKO mice was attenuated by FGF21

overexpression, as reflected by a reduction in Oil Red O-stained areas (Fig. 19).

Overexpression of FGF21 in the Liver Ameliorates Glucose Tolerance and Restores

Glucose Homeostasis in SIRT1 LKO Mice Fed a HFHS Diet

To determine in vivo function of hepatic FGF21 gain-of-function on glucose

homeostasis and insulin sensitivity, GTT and ITT were performed on the HFHS-fed

SIRT1 LKO mice that were injected with either the GFP or FGF21 adenovirus. Two

weeks after the injection of the adenovirus, SIRT1LKO mice expressing FGF21

displayed a ~13% reduction in fasting glucose levels, compared with those of Ad-GFP-

injected SIRT1LKO mice (130.3 + 8.9 vs. 167.0 + 11.7 mg/dl) (Fig. 20).

In intraperitoneal GTTs, SIRT1 LKO mice expressing FGF21 showed

significantly lower glucose levels and improved glucose tolerance, compared to Ad-GFP-

injected SIRT1 LKO mice (Fig. 21A). Consistently, integrated glucose concentrations

calculated as AUC were significantly decreased in SIRT1LKO mice expressing FGF21

(Fig. 21B). In intraperitoneal ITTs, SIRT1LKO mice expressing FGF21 showed

Figure 20. Overexpression of hepatic FGF21 lowers fasting blood glucose levels. Data presented as means ± SEM (n = 5-7). * P < 0.05 vs. WT+Ad-GFP, # P < 0.05 vs. SIRT1 LKO+Ad-GFP  

38

significantly lower insulin levels and improved insulin sensitivity, compared to

SIRT1LKO mice expressing control GFP (Fig. 21C). Consistently, integrated insulin

concentrations calculared as AUC was significantly decreased in SIRT1 LKO mice

expressing FGF21 (Fig. 21D). These results demonstrate that the defective metabolic

phenotype of HFHS-fed SIRT1LKO mice is partially reversed by FGF21.

Figure 21. Adenovirus-mediated overexpression of FGF21 in the liver restores glucose metabolism and protects against insulin resistance in HFHS diet-induced obese SIRT1 LKO mice. A) GTTs were performed on mice fasted for 16 h B) AUC for GTT C) ITTs were performed on mice fasted for 5 h D) AUC for ITT Data presented as means ± SEM (n = 5-7). * P < 0.05 vs. WT+Ad-GFP, # P < 0.05 vs. SIRT1 LKO+Ad-GFP

39

DISCUSSION

This thesis intended to investigate if metformin’s weight loss capability is due to

increased energy expenditure via increased expression of UCP1, DIO2, and PRDM16.

Additionally, I aimed to evaluate metformin’s efficacy in alleviating NAFLD cause by

HFHS diet-induced obesity and if SIRT1 stimulation by resveratrol and delivery of

exogenous FGF21 mimics metformin’s ability to combat obesity and improve NAFLD.

Metformin treatment in diet-induced obese mice caused weight loss and improved

body composition. In line with this, systematic reviews and meta-analyses of studies

involving metformin have overwhelmingly demonstrated metformin’s weight loss

capabilities in humans (Domecq, 2015). The weight loss and decreased fat mass seen in

metformin-treated HFHS-fed mice may be due to the slight decrease in food intake

(Fig. 3C). Other studies have reported that metformin decreases food consumption in

mice (Matsui 2010, Yasuda 2004). Studies looking into the effect of metformin on energy

intake and satiety in obese children also show that metformin decreased caloric

consumption and lead to significant weight loss (Adeyemo, 2015). For future studies, it

may be a good idea to pair-feed the HFHS+Metformin group with the HFHS group so

that both groups have the same caloric intake. With both groups consuming the same

amount of calories, it would be easier to determine whether the differences in body

weight was due to metformin’s ability to reduce food intake or via another mechanism.

Because metformin promotes weight loss and fat mass reduction, we were curious

whether metformin plays a role in the regulation of thermogenesis during obesity.

40

Metformin was found to induce the expression of thermogenic genes, UCP1 and DIO2,

and promote browning of iWAT by inducing PRDM16 expression. It was hypothesized

that metformin increases whole body energy expenditure and via increased

thermogenesis, metformin is able to foster weight loss and improved body composition.

Adjusted energy expenditure means: HFHS (0.62 kcal/hr) and HFHS+Metformin

(0.64 kcal/hr) turned out to be contrary to my hypothesis. If one simply looks at the

unadjusted means of HFHS (0.700 kcal/hr) and HFHS+Metformin (0.568 kcal/hr) and is

not careful, one could misinterpret results. HFHS+Metformin-fed mice do have lower

energy expenditure but this is due to their smaller body weight compared to the HFHS-

fed mice, not due to metformin. HFHS-fed mice have higher energy expenditure due their

increased weight (both increased fat mass and increased lean mass to be able to hull

around the increased fat mass).

Similar misinterpretations regarding energy expenditure can occur when data is

expressed as kcal/kg/hr instead of the correct way—kcal/hr/mouse. One cannot simply

divide energy expenditure by kg of body weight because HFHS-fed mice have higher

body weights and thus would be divided by a larger number. Doing this would result in

the misinterpretation that HFHS-fed mice have decreased energy expenditure. This is a

concern because most of this excess weight is composed of fat, which does not have a

high metabolic rate. Likewise, energy expenditure of HFHS+Metformin-fed mice would

be divided by a smaller number (their kg of body weight is lower and less of it is fat) and

this would result in the misinterpretation that HFHS+Metformin mice have increased

energy expenditure compared to HFHS-fed mice (Tschop 2011). To correctly analyze

41

energy expenditure, one must use the data as kcal/hr/mouse, and then use ANCOVA

statistical analysis to adjust for body weight or lean mass. Analyses can also enter fat

mass, which requires energy to move, and this should be done in future analyses of this

data. It would also be important to separately analyze day and night values, because mice

are more active at night, when they also eat.

When energy expenditure was adjusted for lean mass, ANCOVA analysis resulted

in the following adjusted means: HFHS (0.67 kcal/hr) and HFHS+Metformin

(0.6 kcal/hr). These results indicate that when adjusted for lean mass, metformin-treated

mice have an 11% decrease in energy expenditure. This was opposite of what we

postulated. However, these energy expenditure measurements were taken while the

metformin-treated mice were still losing weight and in a state of negative energy balance.

Associated with being in a state of negative energy balance is decreased energy

expenditure, which could explain the results above. It’s quite probable that a metformin

effect on energy expenditure could have been masked. What should have been done/what

should be done in a future study is wait to get energy expenditure measurements once the

weight loss of the metformin-treated group has plateaued and are no longer in a state of

negative energy balance. Supporting the need for future investigation, aggregate

correlation within samples when adjusted for lean mass was r2 = 0.48. This means HFHS

feeding, metformin, and lean mass together explain 48% of the variability in energy

expenditure, leaving over half of variability unexplained. Further considerable research is

necessary to determine whether the unexplained variability can be explained if energy

expenditure of HFHS+Metformin-fed mice is measured at a more appropriate time.

42

Evidence that the search for metformin’s effect on energy expenditure shouldn’t

be abandoned, metformin-induced inhibition of the mitochondrial respiratory chain

increases FGF21 expression, thereby increasing activation of brown adipocytes and

promoting browning of iWAT (Kim Lee 2013). Resveratrol has also recently been shown

to induce browning of iWAT via AMPK activation (Wang 2015). It would be interesting

to investigate how resveratrol and delivery of exogenous FGF21 affects energy

expenditure.

Oil Red O staining and GTTs demonstrate metformin’s ability to alleviate

NAFLD and restore glucose homeostasis in HFHS diet-induced obese mice. Performing

ITTs would help delineate whether or not metformin improves insulin sensitivity and to

gain more knowledge, this should be done in the future. Resveratrol stimulation of SIRT1

attenuated weight gain and was able to sustain healthy fasting blood glucose levels in

mice (Fig. 8). Consistent with other studies (Lagouge 2006), our studies indicate that

resveratrol protects against diet-induced obesity and hyperglycemia. Furthermore, our

studies determined that SIRT1 is at least partially needed for resveratrol to carry out its

metabolic benefits.

It’s been shown that AMPK and SIRT1 have multiple and intercrossing roles in

regulating different metabolic processes. SIRT1 can activate AMPK, likely via directly

deacetylating the AMPK protein kinase, LKB1, increasing the phosphorylation and

activity of AMPK (Hou et al., 2008; Lan et al., 2008). SIRT1 and AMPK may be part of

an autoregulatory loop where activation of AMPK somehow increases cellular

NAD+/NADH ratio, in turn activating SIRT1. SIRT1 and AMPK do not function

43

independently or linearly and considerable research efforts are still needed to better

understand the intertwined roles of AMPK and SIRT1 (Price 2012). Resveratrol has been

described to stimulate SIRT1 and repress lipogenesis. Highlighting the intertwined roles

of AMPK and SIRT1, resveratrol increased AMPK’s activity, which suppressed fatty

acid synthesis by phosphorylating and inactivating ACC. Resveratrol’s ability to promote

fatty acid oxidation may explain its attenuation in hepatic steatosis (Fig. 9C-E).

SIRT1 LKO mice are more susceptible to obesity-induced NAFLD and the

defective AMPK activity may explain profound fatty liver in obese SIRT1 LKO mice

(Fig. 13). Further demonstrating the importance of SIRT1, metformin action is impaired

in cells lacking SIRT1 (Canto 2009). Fatty liver is associated with FGF21 insufficiency

in SIRT1 LKO mice (Li et al., 2014). Ablation of SIRT1 in the liver increases the

propensity for weight gain and hepatic steatosis; but intriguingly, this can be rescued via

adenoviral overexpression of FGF21. Furthermore, GTTs and ITTs showed that

exogenous FGF21 delivery can reverse defective glucose homeostasis and improve

insulin sensitivity.

Evaluation of Diets Used in Animal Studies

The CHOW diet used in these experiments is the 2918 Irradiated Teklad Global

18% Protein Rodent Diet from Harlan Laboratories. This CHOW is comprised of 24% of

calories from protein, 18% from fat, and 58% from carbohydrates. The IOM recommends

adults to get 10-35% of their calories from protein, 20-35% from fat, and 45-65% from

carbohydrates. CHOW could be considered “low fat” seeing that its fat content is slightly

lower than IOM recommendations. CHOW has an energy density of 3.1 kcal/gram.

44

CHOW’s source of protein is soybean meal and its source of fat is soybean oil. CHOW’s

source of carbohydrate is ground wheat, ground corn, and wheat midds. The HFHS diet

used in these experiments is the F1850 High fat mouse diet paste from BioServ. This

HFHS diet is comprised of 15% of calories from protein, 59% from fat, and 26% from

carbohydrates. The HFHS diet is 5.51 kcal/gram. The source of protein for this diet is

casein and its source of fat is lard. This HFHS diet’s source of carbohydrate is

maltodextrin.

Harlan Laboratories’ CHOW and BioServ’s HFHS diet have completely different

sources of protein, fat, and carbohydrate. If one chooses to use these two diets in an

experiment to compare low fat and high fat diets, the different sources of protein, fat, and

carbohydrate can lead to confounding results. For instance, CHOW’s source of fat is

soybean oil while the source of fat for the HFHS diet is lard. It has been shown that

replacing lard with soybean oil in high-fat diet alleviates obesity-related inflammation

and insulin resistance (Wang, Yang 2013). The difference in the source of fat would

corrupt the results gathered from that study. This CHOW would not be a good control

diet for this HFHS diet. Appropriate control diets should have the same source of protein,

fat, and carbohydrate.

Since our studies evaluated metformin and resveratrol treatment in HFHS-fed

mice we avoided the problem of having an improper control diet. However, the HFHS

diet from BioServ used in all our studies does not contain fiber. Fiber helps regulate the

body’s use of sugars, helping keep hunger and blood sugar in check. High fiber intake

has been linked to a lower risk of type 2 diabetes and heart disease. It’s possible that this

45

lack in fiber confounded some of our results or contributed to variation in our results. It’s

also important to mention that this HFHS diet is comprised of 60% of calories from fat,

which is quite high. Although it’s true that Western societies typically consume diets high

in fat, 60% is not representative of a typical high-fat diet in humans.

Conclusion

Overnutrition from consumption of diets high in fat and sugar has propagated the

incidence of obesity. Excess adipose tissue of obese individuals releases increased levels

of circulating free fatty acids and these excess fatty acids overload metabolic tissues such

as liver, muscle, and pancreatic β-cells, resulting in NAFLD, insulin resistance, and type

2 diabetes. Metformin, the most commonly prescribed anti-diabetic drug, facilitates

weight loss likely via decreasing caloric intake. Metformin increased expression of

UCP1, DIO2, and PRDM16; but did not seem to increase energy expenditure. However,

energy expenditure was measured during a state of negative energy balance, which may

have masked an effect of metformin on metabolic rate. Further studies are needed to

determine whether metformin-mediated increased expression of thermogenic genes

translates into increased whole body energy expenditure. Our studies demonstrate that

metformin is capable of alleviating NAFLD caused by HFHS diet-induced obesity and

that SIRT1 stimulation by resveratrol and exogenous administration of FGF21 mimics

metformin’s ability to combat obesity and improve hepatic steatosis. Collectively, these

results implicate metformin, resveratrol, and exogenous administration of FGF21 as

beneficial therapies for weight loss and amelioration of NAFLD.

46

Figure 22. A proposed working model for the mechanisms through which AMPK, SIRT1, and FGF21 modulate the progression of obesity, NAFLD, and insulin resistance.

47

APPENDIX

Experimental design of animal studies with metformin treatment.

Experimental design of animal studies with resveratrol treatment.

48

Experimental design of wild-type and SIRT1 LKO mice with resveratrol treatment. .

Experimental design of wild-type and FGF21 transgenic mice fed CHOW and a HFHS diet.

49

50

 LIST OF JOURNAL ABBREVIATIONS

Acta Biochem Biophys Sin ...................................... Acta Biochimica et Biophysica Sinica

Am J Physio Endo Metab..... American Journal of Physiology Endocrinology Metabolism

Ann Med................................................................................................ Annals of Medicine

Biochem Biophys Res Commun ...Biochemical & Biophysical Research Communications

Cell Metab .................................................................................................. Cell Metabolism

Eur J Nutr ..............................................................................European Journal of Nutrition

Exp Diab Res.....................................................................Experimental Diabetes Research

Exp Mol Med........................................................... Experimental and Molecular Medicine

Hum Reprod Update...............................................................Human Reproduction Update

Int J of Obesity ................................................................... International Journal of Obesity

J Biol Chem ........................................................................Journal of Biological Chemistry

J Clin Endo Metab................................Journal of Clinical Endocrinology and Metabolism

J Clin Invest....................................................................... Journal of Clinical Investigation

Mol Cell Endo ......................................................... Molecular and Cellular Endocrinology

Nat Rev Mol Cell Biol.......................................... Nature Reviews Molecular Cell Biology

Trends Endocrinol.Metab................................... Trends in Endocrinology and Metabolism

Trends in Mol Med............................................................... Trends in Molecular Medicine

51

REFERENCES

1. “Adult Obesity Facts”. Division of Nutrition, Physical Activity, and Obesity. Centers

for Disease Control and Prevention. http://www.cdc.gov/obesity/data/adult.html

2. Canto, C., Gerhart-Hines, Z., Feige, J.N., Lagouge, M., Noriega, L., Milne, J.C., Elliott, P.J., et al. (2009) AMPK regulates energy expenditure by modulating NAD+ metabolism and SIRT1 activity. Nature 458, 1056-1060.

3. Coskun, T., Bina, H., Schneider, M., Dunbar, J., et al. (2008). Fibroblast growth factor 21 corrects obesity in mice. Endocrinology 149(12): 6018-6027.

4. Chalkiadaki, A., Guarente, L. (2012). Sirtuins mediate mammalian metabolic responses to nutrient availability. Nature Reviews 8, 287-296.

5. Chong, P.K., Jung, R.T., Rennie, M.J., Scrimgeous, C.M. (1995). Energy expenditure in type 2 diabetic patients on metformin and sulphonylurea therapy. Diabetic Medicine 12(5), 401-408.

6. “Dietary Reference Intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids.” Institute of Medicine. http://iom.nationalacademies.org/Reports/2002/Dietary-Reference-Intakes-for-Energy-Carbohydrate-Fiber-Fat-Fatty-Acids-Cholesterol-Protein-and-Amino-Acids.aspx

7. Domecq, J.P., Prutsky, G., Leppin, A., Sonbol, M.B., Altayar, O., et al. (2015) Clinical review: Drugs commonly associated with weight change: a systematic review and meta-analysis. J Clin Endo Metab 100(2), 363-370.

8. “Fiber”. The Nutrition Source. Harvard School of Public Health. http://www.hsph.harvard.edu/nutritionsource/carbohydrates/fiber/

9. Fisher, M., Kleiner, S., Douris, N., Fox, E.C., Mepani, E.J., et al. (2012) FGF21 regulates PG1-alpha and browing of white adipose tissues in adaptive thermogenesis. Genes and Development 26:271-281.

10. Fulco, M., Sartorelli, V. (2008). Comparing and Contrasting the Roles of AMPK and SIRT1 in Metabolic Tissues. Cell Cycle 7(23), 3669-3679.

11. Geerling, J.J., Boon, M.R., van der Zon, G.C., van den Berg, S.A., et al. (2014). Metformin lowers plasma triglycerides by promoting VLDL-triglyceride clearance by brown adipose tissue in mice. Diabetes 63.

52

12. Gillum, M.P., Erion, D.M., Shulman, G.I. (2011). Sirtuin-1 regulation of mammalian metabolism. Trends in Mol Med 17(1).

13. Gual, P., Le Marchand-Brustel, Y., Tanti, J.F. (2005). Positive and negative

regulation of insulin signaling through IRS-1 phosphorylation. Biochimie 87(1).

14. Guarente, L. (2006). Sirtuins as potential targets for metabolic syndrome. Nature 444, 868-874.

15. Guarente, L. (2012). Sirtuins and calorie restriction. Nat Rev Mol Cell Biol. 13, 207. 16. Hardie, D.G., Ross, F.A., Hawley, S.A. (2012). AMPK: a nutrient and energy sensor

that maintains energy homeostasis. Nature Reviews 13, 251-262. 17. He, L., Wondisford, F.E. (2015). Metformin action: Concentrations matter. Cell

Metab 21. 18. Hou, X., Xu, S., Maitland-Toolan, K.A., Sato, K., Jiang, B., Ido, Y., Lan, F., Walsh,

K., Wierzbicki, M., Verbeuren, T.J., et al. (2008). SIRT1 regulates hepatocyte lipid metabolism through activating AMP-activated protein kinase. J Biol Chem 283, 20015-20026.

19. Hu, Y., Young, A.J., Ehli, E.A., Nowotny, D., Davies, P.S., et al. (2014) Metformin

and berbine prevent olanzapine-induced weight gain in rats. PLOS ONE 9(3). 20. Itoh, N. (2014). FGF21 as a hepatokine, adipokine, and myokine in metabolism and

diseases. Frontiers in Endocrinology 5(107). 21. Kahn, B.B., Alquier, T., Carling, D., and Hardie, D.G. (2005). AMP-activated protein

kinase: Ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab 1, 15-25.

22. Kharitonenkov, A., and Larsen, P. (2011). FGF21 reloaded: challenges of a rapidly

growing field. Trends Endocrinol.Metab 22, 81-86. 23. Kharitonenkov, A., and Adams, A.C. (2014). Inventing new medicines: The FGF21

story. Molecular Metabolism 3, 212-229. 24. Kharitonenkov, A., Shiyanova, T.L., Koester, A., Ford, A.M., Micanovic, R.,

Galbreath, E.J., Sandusky, G.E., Hammond, L.J., Moyers, J.S., Owens, R.A., et al. (2005). FGF-21 as a novel metabolic regulator. J Clin Invest 115, 1627-1635.

25. Kim, K.H., Myung-Shik, L. (2014). FGF21 as a stress hormone: The roles of FGF21

in stress adaptation and the treatement of metabolic diseases.

53

26. Kim, K.H., Yeon, T.J., Seong, H.K., Hye, S.J., Kyong, S.P., Hae-Youn, L.,

Myung0Shik, L. (2013). Metformin-induced inhibition of the mitochondrial respiratory chain increases FGF21 expression via ATF4 activation. Biochem Biophys Res Commun. 440, 76-81.

27. Lagouge, M., Argmann, C., Gerhart-Hines, Z., Mezaine, H., Lerin, C., et al. (2006).

Cell 127, 1109-1112. 28. Lan, F., Cacicedo, J.M., Ruderman, N., and Ido, Y. (2008). SIRT1 modulation of the

acetylation status, cytosolic localization, and activity of LKB1. Possible role in AMP-activated protein kinase activation. J Biol Chem. 283, 27628-27635.

29. Li, X. (2013). SIRT1 and energy metabolism. Acta Biochem Biophys Sin 30. Li, Y., Wong, K., Giles, A., Jiang, J., Lee, J.W., Adams, A.C., Kharitonenkov, A.,

Yang, Q., Gao, B., Guarente, L., et al. (2014). Hepatic SIRT1 Attenuates Hepatic Steatosis and Controls Energy Balance in Mice by Inducing Fibroblast Growth Factor 21. Gastroenterology 146, 539-549 e537.

31. Li, Y., Xu, S.Q., Mihaylova, M.M., Zheng, B., Hou, X.Y., Jiang, B.B., Park, O., Luo,

Z.J., Lefai, E., Shyy, J.Y.J., et al. (2011). AMPK Phosphorylates and Inhibits SREBP Activity to Attenuate Hepatic Steatosis and Atherosclerosis in Diet-Induced Insulin-Resistant Mice. Cell Metab 13, 376-388.

32. Mazza, A., Fruci, B., Garinis, G.A., Giuliano, S., Malaguarnera, R., Belfiore, A.

(2012) The Role of Metformin in the Management of NAFLD. Exp Diab Res. 33. Moreno-Navarrete, J.M., Ortega, F., Moreno, M., Xifra, G., et al. (2015). PRDM16

sustains white fat gene expression profile in human adipocytes in direct relation with insulin action. Mol Cell Endo.

34. Meng, S., Cao, J., He, Q., Xiong, L., Change, E., et. al. (2015). Metformin activates

AMP-activated protein kinase by promoting formation of the αβα heterotrimeric complex. J Biol Chem. 290(6).

35. Naderpoor, N., Shorakae, S., de Courten, B., Misso, M.L., Moran, L.J., Teede, H.J.

(2015). Hum Reprod Update. 36. “Obesity and overweight”. World Health Organization.

http://www.who.int/mediacentre/factsheets/fs311/en/

54

37. Oh, T.J., Shin, J.Y., Kang, G.H., Park, K.S., Cho, Y.M. (2013). Effect of the combination of metformin and fenofibrate on glucose homeostasis in diabetic Goto-Kakizaki rats. Exp Mol Med 45, e30.

38. Price, N.L., Gomes, A.P., Ling, A.J.Y., Duarte, F.V., Martin-Montalvo, A.M., et al.

(2012). SIRT1 is Required for AMPK Activation and the Beneficial Effects of Resveratrol on Mitochondrial Function. Cell Metab 15, 675-690.

39. Renquist, B.J., Zhang, C., Williams, S.Y., Cone, R.D. (2013). Development of an

Assay for High-Throughput Energy Expenditure Monitoring in the Zebrafish. Zebrafish 10(3), 343-352.

40. Rodgers, J.T., Lerin, C., Haas, W., Gygi, S.P., Spiegelman, B.M., and Puigserver, P.

(2005). Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113-118.

41. Schug, T.T., Li, X. (2011). Sirtuin 1 in lipid metabolism and obesity. Ann Med 43(3),

198-211. 42. Shaw, R.J., Lamia, K.A., Vasquez, D., Koo, S.H., Bardeesy, N., DePinho, R.A.,

Montminy, M., and Cantley, L.C. (2005). The Kinase LKB1 Mediates Glucose Homeostasis in Liver and Therapeutic Effects of Metformin. Science.

43. Tajima, K., Nakamura, A., Shirakawa, J., Togashi, Y., Orime, K., et al. (2013).

Metformin prevents liver tumorigenesis induced by high-fat diet in C57BL6 mice. Am J Physio Endo Metab 305: E987-E998.

44. Toorie, A.M., Nillni, E.A. (2014). Minireview: Central Sirt1 Regulates Energy

Balance via the Melanocortin System and Alternate Pathways. Molecular Endocrinology 28:9, 1423-1434.

45. Tschop, M.H., Speakman, J.R., Arch, J.R, Auwerx, J., Bruning, J.C., Chan, L., et al. (2011) A guide to analysis of mouse energy metabolism. Nature Methods 9(1), 57-63.

46. Wang, S., Liang, X., Yang, Q., Fu, X., Rogers, C.J., et al. Resveratrol induces brown-

like adipocyte formation in white fat through activation of AMP-activated protein kinase (AMPK)α1. Int J of Obesity 39, 967-976.

47. Wang, X., Cheng, M., Zhao M., Guo, F., et al. (2013) Differential effects of high-fat-diet rich in lard oil or soybean oil on osteopontin expression and inflammation of adipose tissue in diet-induced obese rats. Eur J Nutr 52(3), 1181-1189.

55

48. Xu, J., Lloyd, D.J., Hale, C., Stanislaus, S., Chen, M., Sivits, G., et al. (2009). Fibroblast growth factor 21 reverse hepatic steatosis, increases energy expenditure, and improves insulin sensitivity in diet-induced obese mice. Diabetes 58, 250-259.

49. Zang, M., Xu, S., Maitland-Toolan, K.A., Zuccollo, A., et al. (2006). Polyphenols stimulate AMP-activated protein kinase, lower lipids, and inhibit accelerated artherosclerosis in diabetic LDL receptor-deficient mice. Diabetes 55.

50. Zang, M., Zuccoll, A., Hou, X., Nagata, D., Walsh, K., et al. (2004). AMP-activated protein kinase is required for the lipid-lowering effect of metformin in insulin-resistant human Hep G2 cells. J Biol Chem 276(46).

51. Zhou, G., Myers, R., Ying, L., Chen, Y., Shen, X., Fenyk-Melody, J., et al. (2001). Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 108(8).

56

CURRICULUM VITAE

 ALLISON LEE NOCON

[email protected] (850) 384-1210

DOB: 1990 EDUCATION Boston University School of Medicine—Boston, MA Expected Sept 2015 Division of Graduate Medical Sciences M.A. Nutrition & Metabolism University of Florida—Gainesville, FL May 2012 B.S. Food Science & Human Nutrition RESEARCH EXPERIENCE Graduate Research Assistant—Boston, MA Jan 2014 – Present PI: Mengwei Zang, PhD; Boston University School of Medicine, Vascular Biology • Investigate the benefits of nutrient sensors—AMPK, SIRT1, and FGF21—on obesity

and metabolic derangements including non-alcoholic fatty liver disease, insulin resistance, and alcoholic liver disease

• Investigate the effect of anti-diabetic agent, metformin, on weight loss, improvement of body composition and glucose clearance, browning of iWAT for increased thermogenesis and increased energy expenditure

• Carry out animal studies including treatment feeding, resection of organs, oral gavage, tail vein injections, intraperitoneal injections, glucose tolerance tests, insulin tolerance tests, RNA extraction, qPCR, H&E staining, and Oil Red O staining

Undergraduate Research Assistant—Gainesville, FL Aug 2010 - April 2012 PI: Nicholas Simpson, PhD; University of FL School of Medicine, Endocrinology • Surgically implanted bio-artificial pancreas into mice under anesthesia with goals to

develop a bio-artificial pancreas that will benefit those with diabetes or pancreatitis • Executed daily blood glucose measurements of mice and cultured cells Undergraduate Research Assistant—Gainesville, FL Oct 2009 - May 2010 • Electrically stimulated gonads of horseshoe crabs to collect sperm as data investigating

sperm competition in horseshoe crabs, tagged and took measurements of wild horseshoe crabs, helped raise developing horseshoe crabs starting from birth (Day 0)

57

OBSERVATIONAL EXPOSURE IN CLINICS / SHADOWING Boston Medical Center—Boston, MA Feb – March 2015 Dr. Brian Jacobson, Gastroenterology • Shadowed in on hospital rounds and regular patient consultations ranging from GERD

to hepatitis to anal fissures

Gastroenterology Associates, Sacred Heart Hospital—Pensacola, FL Dec 2013 • Spent over 50 hours observing procedures—endoscopies, esophagus dilations,

colonoscopies, polyp removals, and a laparoscopic ERCP—and listening in on hospital rounds and regular patient consultations

Okaloosa Public Health Department—Fort Walton Beach, FL June - July 2010 Women, Infants, and Children program (WIC) • Taught about-to-be mothers the necessity of prenatal vitamins and benefits of breast

feeding as well as explained which beneficial food products WIC coupons could buy Environmental Health • Post BP oil spill in the Gulf of Mexico, assisted in taking samples of the water to

determine when it was safe for the public to swim and fish

UF Health Shands Hospital—Gainesville, FL Jan - April 2010 Dr. Kenneth Heilman, Neurology • Spent over 40 hours (3 h/wk) observing routine clinical tests such as reflex checks,

motor skills assessment, and memory recall

Project Peru with FIMRC—Medical Mission Trip May 4-18, 2009 • Observed intramedullary nailing to treat femoral shaft fracture, hysterectomy, natural

and C-section birth, newborn screenings, provided basic dental health to children, and taught basic English to children in the Peruvian cities Trujillo, Agallpampa, and Cusco

COMMUNITY SERVICE & INVOLVEMENT Boston Healthcare for the Homeless Program—Boston, MA Feb 2015 - Present • As a volunteer, helped put together and deliver meals to patients especially for those

with dietary restrictions due to medical conditions (3 h/wk)

North Florida Regional Hospital—Gainesville FL Aug - Dec 2012 • Volunteered over 70 hours (4 h/wk) as a patient ambassador and doing clerical work After School Gators—Gainesville, FL Sept 2008-March 2012 • Volunteered aiding with homework, promoting good study habits, and formed bonds

with students in elementary school an hour weekly

58

Co-educational Chair—Gainesville, FL Nov 2010-Nov 2011 Kappa Kappa Gamma, Epsilon Phi Chapter • Kept track of members GPAs; when needed, advised individuals under academic probation University of Florida’s Young Leaders Conference (YLC)—Gainesville, FL Sept 2009-Jan 2010, Sept 2010-Jan 2011 • Created, taught, and lead workshops for a leadership conference for exceptional Florida high school students International Student Volunteers (ISV)—New Zealand June 2009 • Planted trees and removed invasive plant species to improve Motuihe Island's habitat so

that endangered kiwi birds could be reintroduced to the island EMPLOYMENT Earth Origins --Gainesville, FL Aug 2012-Dec 2012 • Full-time in supplements department providing nutritional information and customer service Dr. Gajo and Associates—Fort Walton Beach, FL Summers 2005-2011 • Checked patients in and out, collected co-pays, data entry, got authorizations with

patients’ insurance, updated medication logs, and performed other necessary office routines