jaemyoung suh salk institute/hhmi la jolla, ca icdm nov.9, 2012icdm2012.diabetes.or.kr/slide/s3 jae...

A New NR-FGF Axis: Regulation of Feast and Famine

Jaemyoung SuhSalk Institute/HHMI

La Jolla, CA

ICDM Nov.9, 2012

Feast and Famine

- Feast/famine cycles are a recurrent environmental stressor

- The ability to withstand periods of limited/excess nutrientavailability is critical to animal survival

Cinti S Am J Physiol Endocrinol Metab 2009;297:E977-E986

Adipose tissues:

Central to the feast/famine response

Visceral white adipose tissue

Subcutaneous white adipose tissue

Brown fat

Adipose tissues are dynamic and

possess wide developmental potential

Feast(Adipose expansion)

Famine(Adipose contraction)

Adipose development and remodeling–

coordination of multiple cell types

Adipose tissue remodeling - adaptive response to energy stress

PPARg

- Adipose tissues expand and contract during feast/famine cycles� a process called “remodeling”

- Remodeling is a complex process that requires coordinated changes within adipocytes, immune cells, surrounding vasculature, and the extracellular matrix

- Adaptive remodeling is dynamic and critical for maintaining metabolic homeostasis

- The mechanisms underlying remodeling are poorly understood

Fibroblast growth factors

Fatty acids Bile acids Vitamin D

PPARα/RXR FXR/RXR VDR/RXR

Nuclear

receptor

Growth factor

expression

Physiological

response

FGF21 FGF15/19 FGF23

Energy

homeostasis

Bile acid

homeostasis

Phosphate / Ca+

homeostasis

The NR-FGF axis

Ligands

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Fed HFD Fast Fed HFD Fast

Relative expression

Relative expression

FGF1A in Vis WAT FGF1B in Vis WAT

FGF1 in adipose tissue is regulated by

nutritional cues

**

**

Wild type visceral fat

Chow Diet High-Fat Diet

FGF1

FGF1 protein is induced by high fat diet in

adipose tissue

Adipose tissue subcompartments –

adipocytes and stromal cells

Collagenase

Adipocytes

(float)

Stromal –Vascular-Fraction

(sink)

. . .. . . . . ..

FGF1 is expressed predominantly

in visceral adipocytes

Whole Fat Subcutan. Visceral

SC adip. SVF adip. SVF

Fgf1

ERK1/2

Visceral fat depots are linked to metabolic disease and

has more active remodeling capacity than subcutaneous fat

Vis

Fibroblast growth factors

PPARαPPARα

FXRFXRVDRVDR

NR regulation?NR regulation?

Metabolism?Metabolism?

Normal

Normal

FGF1:a “boring” FGF

30

34

38

42

46

50

0 2 4 6 8 10 12 14 16 18 20 22 24

No change in body weight gain in FGF1

KO mice on a HFD

Time (weeks)

Body weight (grams)

Still boring>>.?

WT

KO

100

200

300

400

0 15 30 45 60 75 90

60

80

100

120

140

160

0 15 30 45 60 75 90

Time (min)

Glucose (mg/dl)

Glucose (mg/dl)

Time (min)

**

**

* **

Increased insulin resistance in FGF1 KO

mice on HFD

Glucose Tolerance Test Insulin Tolerance Test

WT

KO

0

4

8

12

16

0

5

10

15

20

25

Increased peripheral and hepatic insulin

resistance in FGF1 KO mice on a HFD

WT KO WT KO

IS-GDR HGP (% suppression)

GIR (mg/kg/m

in)

GIR (mg/kg/m

in)

*

*

Skeletal muscle Liver

WT

KO

Visceral adipose tissue fails to expand in

FGF1 KO mice on a HFD

0

2

4

6

8

% BW

Control HFD

0

1

2

3

4**

**

% BW

Control HFD

Liver Visceral fat

WT

KO

Increased liver fat accumulation in FGF1

KO mice on a HFD

KO

WT KO

Non-uniform size distribution in FGF1 KO

visceral adipocytes

WT

KO

-1

-0.5

0

0.5

1

1.5

2

2.5

3

Log (KO/W

T (frequency)

Small Medium Large

Bin (square µm area)

Increased fibrosis in visceral adipose tissue of

FGF1 KO mice on a HFD

Masson’s trichrome stain – collagen stains blue

KO

KOWT

Impaired vascularity in visceral adipose of

FGF1 KO mice on a HFD

WT KO

Visceral adipose tissue can’t properly expand in response to HFD in FGF1 KO mice;;. But, can it contract?

Heart perfusion of fluorescent spheres – vascular space in red, nuclei in blue

KO

36 weeks high fat diet

6 weekschow diet

Modern-day yoyo diet

WT

KO

Further impairment of FGF KO visceral

adipose tissue upon HFD withdrawal

Failure of FGF1 KO visceral WAT to properly contract upon HFD withdrawal

WT KO

Fat necrosis in FGF1 KO mice after

removal of HFD

Fatty

acids

Bile acids Vitamin D

NR PPARα/RXR FXR/RXR VDR/RXR

Nuclear

receptor

Growth factor

expression

Physiological

response

FGF1 FGF21 FGF15/19 FGF23

Fat

remodeling

Energy

homeostasis

Bile acid

homeostasis

NR-FGF Axis

?

?

?

Phosphate / Ca+

homeostasis

0

4

8

12

16

0

4

8

12

16

Regulation of FGF1 by PPARγ and TZDs

Luc/LacZ

PPARα

Ligand - + - + - +

PPARγ PPARδ PPARα

Ligand - + - + - +

PPARγ PPARδ

FGF1A FGF1B

**

**

*

0.00

0.05

0.10

0.15

0.20

0.25

PPARγ binds to the FGF1 promoter

FGF1

Input chromatin (%)

36B4

** Adip.

WT KO

FGF1

ERK1/2

SVF

WT KO

FGF2

FGF1 levels reduced in

PPARγ adipose KO mice

IgG

PPARγ

A new NR-FGF axis

Acknowledgements

Ron Evans Hans JonkerMichael DownesAnn AtkinsMaryam AhmadianRuth Yu

Jerry OlefskyPingping Li

Future directions

• Some major questions ;

- Why is only visceral fat affected by loss of FGF1?

- Are adipocytes solely responsible for the whole body KO phenotype?

- What are the target cells of FGF1 action?

- Is there any therapeutic relevance for FGF1 in treating metabolic disease?

Fasting glucose (mg/dl)

104

108

112

116

120

124

1.0 1.5 2.0

Fasting blood glucose

FGF1 (fold induction)

*

FGF1 (fold induction)

0.0

0.5

1.0

1.5

2.0

Induction of FGF1 by TZD

ctrl TZD

FGF1 link to insulin sensitization

- Is there a human connection?

FGF1 gain-of-function studies- exploring therapeutic possibilities

• Pharmacological

- Parental delivery of recombinant FGF1 protein into animal models of obesity/diabetes.

• Genetic

- Tissue-specific transgenic overexpression of FGF1

- Viral overexpression of FGF1

Experimental approaches

Anti-diabeticInsulin sensitization

TZD

PPARγ (master regulator)

Cardiovascular toxicityWeight gainEdemaLiver toxicityBone loss

FGF1

ActosAvandia

$ 4 bn

>100 target genes

FGF1 : therapeutic possiblities ?

?Time (days)

Recombinant FGF1 has potent glucose lowering effects in diabetic ob/ob mice

?

Adipose stem cell(CD34, CD29, Sca-1, CD24,

PPARγ, Zfp423)

1

2

3

Mature adipocyte(aP2, Glut4, Perilipin,

LPL, Leptin, Resistin)

4

Mural cell(SMA, NG2, PDGFRβ)

Endothelial cell(PECAM-1, VE-Cadherin,

VEGFR2,Tie2)

Hypertropic adipocyte(TNF-α, ROS, FFA)

Dying adipocyte(Perilipin-)

The adipocyte life cycle

DIET &NUTRITION

DRUGS

AGING

DEVELOPMENT

DISEASE

Biology of adipose tissues

Feast and Famine

VDR

PPARg

-Feast/famine cycles have occurred throughout time

-The ability to withstand periods of limited/excess nutrientavailability is a critical aspect of survival

Why would alleles predisposing to obesity

exist in natural populations?

“Thrifty allele hypothesis”

James V. Neel, M.D.,Ph.D.

(1960s)

“Genes associated with common modern diseases like diabetes,

hypertension and obesity are part of the human gene pool,

because they helped our early ancestors survive when calories

and salt were less abundant.”

Fatty

acids

Fatty

acids

Bile acids Vitamin D

PPARγ/RXR PPARα/RXR FXR/RXR VDR/RXR

Nuclear

receptor

Growth factor

expression

Physiological

response

FGF1 FGF21 FGF15/19 FGF23

Fat

remodeling

Energy

homeostasis

Bile acid

homeostasis

Phosphate

and calcium

homeostasis

A new member of the FGF-NR Axis

Acknowledgements

FGF1 is NOT boring!

+

BlastocystZygote

Ectoderm

Endoderm

Mesoderm

Development: Lineage RestrictionBrain Skin

Liver Pancreas

Lung ThyroidGI Tract

Bone Muscle

Blood Kidney HeartPluripotent

Stem Cells( )

Topological distribution of adipose

organs – developmentally regulated