Obesogens, Stem Cells and the
Maternal Programming of
Obesity
Bruce Blumberg, Ph.D.
Department of Developmental and Cell Biology
Department of Pharmaceutical Sciences
Developmental Biology Center
University of California, Irvine
Main Points
• Epigenetics links environment to altered gene expression
• Obesogens exist and contribute to obesity epidemic
• Obesogen action may involve reprogramming of stem cells
What is Epigenetics ?
• Literally “on top of genetics”
– coined by C.H. Waddington in 1957
• Epigenetics involves changes in gene
expression without changes in the
DNA sequence
– Heritable, maintained
– Reversible
– Encoded in chromatin
– Cellular memory
• Examples
– X inactivation
– Genomic imprinting
– Cancer – widespread silencing
or overexpression of genes
Goldberg et al. (2007) Cell 128, 635-638
Epigenetics acts through chromatin stucture
• Methylation/demethylation of DNA, proteins
• Acetylation/deacetylation of DNA-binding proteins
• Changes can act at very long range, 100s of kb to mb
– from other chromosomes!
Epigenetics acts through chromatin stucture
• Chromatin conformation affects accessibility of DNA to transcriptional
machinery – epigenetics controls genetics
• Widespread changes in DNA methylation can be associated with
diseases, e.g., cancer
Genetics and epigenetics of disease
• Some genetic diseases
– Sickle cell anemia
– Cystic fibrosis
– Hemophilia
– Marfan syndrome
– Duchenne muscular dystrophy
– Huntington’s disease
• Diseases with an epigenetic
component (from ID twin studies)
– Fragile X syndrome
– Prader-Willi syndrome
– Scleroderma
– Autism
– Schizophrenia
– Inflammatory bowel disease
– Cancer (e.g. melanoma)
Developmental Basis of Disease
• Barker Hypothesis - gestational under-nutrition leads to a thrifty phenotype
– reduced fetal growth strongly linked with chronic conditions later in life
– Increased susceptibility results from adaptations made by the fetus in an
environment limited in its supply of nutrients
• Developmental Origins of Health and Disease (Mark Hanson) more
generally proposes that development is exquisitely sensitive to
perturbations that lead to permanent changes in disease susceptibility
– Birth defects, low birth weight, premature birth
– Functional changes –appears normal but has molecular abnormalities that
persist and lead to increased disease sensitivity later in life
• Developmental programming continues into adolescence
– Överkalix studies in Sweden linking nutrition and longevity
– Programming of adipocyte number continues into puberty
• Diseases with Developmental Origins (from animal models)
–
–
–
–
–
Cardiovascular, Pulmonary (asthma)
Neurological (ADHD, Neurodegenerative diseases),
Immune/autoimmune
Endocrine, reproductive/fertility, cancer
Obesity/diabetes
The Worldwide Obesity Epidemic
• 34% of the US population are clinically obese (BMI > 30)
– Double worldwide average (Flegal et al. JAMA 2010;303:235-241)
• 68% are overweight (BMI > 25) – 86% estimated by 2020
BMI ~32
BMI ~32
Subcutaneous obesity
adaptive
BMI = 31.5
Visceral obesity
pathological
From Lars Lind
Obesity Trends* Among U.S. Adults
BRFSS, 1990, 1999, 2008
(*BMI 30, or about 30 lbs. overweight for 5’4” person)
1999
1990
2008
~17,000
22,401
30,961
No Data
<10%
Sources:
10%–14%
15%–19%
20%–24%
25%–29%
≥30%
CDC (map), U.S. Census bureau (numbers)
The Worldwide Obesity Epidemic
• 34% of the US population are clinically obese (BMI > 30)
– Double worldwide average (Flegal et al. JAMA 2010;303:235-241)
• 68% are overweight (BMI > 25 ) – 86% estimated by 2020
• Obesity accounts for 8% of healthcare costs in Western Countries
– $75 billion annually in US (2005), $147 billion (2009)
• Obesity is associated with “metabolic syndrome” -> type 2 diabetes
and cardiovascular disease
– Central (abdominal obesity)
– Atherogenic dyslipidemia (high triglycerides, high LDL, low HDL)
– Hypertension
– Insulin resistance
– Prothrombotic state
– Pro-inflammatory state (elevated CRP)
How does obesity occur ?
• Prevailing wisdom – “couch potato syndrome”
– Positive energy balance, i.e., too much food, too little exercise
• Are there other factors in obesity ?
– Stress (elevated glucocorticoids)
– Inadequate sleep (stress?)
– “Thrifty” genes which evolved to make the most of scarce calories
– Viruses, gut microbes, SNPs
• What about role of prenatal nutrition or in utero experience?
– Southampton studies
– Maternal smoking decreases birth weight and increases obesity
• What about the role of industrial chemicals in rise of obesity?
– Baillie-Hamilton (2002) postulated a role for chemical toxins
– obesity epidemic roughly correlates with a marked increase in the
use of chemicals (plastics, pesticides, etc.)
• Many chemicals have effects on the endocrine system
Hormonal control of weight
• Hormonal control of appetite and metabolism
– Leptin, adiponectin, ghrelin are key players
– Leptin, adiponectin – adipocytes
– Grehlin – stomach
– Thyroid hormone/receptor
• Sets basal metabolic rate
• Hormonal control of fat cell
development and lipid balance
– Regulated through nuclear
hormone receptors RXR, PPARγ
– PPARγ – master regulator of
fat cell development
• increased fat cell differentiation
• Increased storage in existing cells
• Increased insulin sensitivity
From Nature Medicine 10, 355 - 361 (2004)
Endocrine Disrupting Chemicals (EDCs)
• Endocrine disrupter - a compound that mimics or blocks the action of
endocrine hormones, either directly or indirectly
– Often persistent pollutants or dietary components that disturb
development, physiology and homeostasis
• Frequently act through nuclear hormone receptors
– Environmental estrogens
– Anti-androgens
– Anti-thyroid
• Recent white paper from the Endocrine Society - DiamantiKandarakis, et al, Endocrine Reviews 30 (4): 293-342 (2009)
– Details scientific support for existence and effects of EDCs
– Endorsed by American Medical Association
– Led to H.R.4190 - Endocrine Disruption Prevention Act of 2009
– Moves responsibility for research from EPA to NIEHS
Endocrine Disrupting Chemicals (EDCs)
• Are EDC-mediated disturbances in endocrine signaling
pathways involved in adipogenesis and obesity
The Nuclear Hormone Receptor Superfamily
A/B
C
D
DNA
E
F
LIGAND
Known Receptors
Orphan Receptors
Classical receptors (from biochemistry)
Vertebrate
Drosophila
GR
MR
AR
PR
ER α,β
VDR
TR α,β
EcR
TR-2 α,β
NGFI-B α,β,γ
ROR α,β,γ
Rev-erb α,β
SF-1 α,β
COUP α,β,γ
HNF-4 α, β
Tlx α,β
DHR78
DHR38
DHR3
E75, E78
FTZ-F1 α,β
svp
HNF-4
tll
cortisol
aldosterone
testosterone
progesterone
estradiol
1,25-(OH)2 vit D3
triiodothyronine
20-OH ecdysone
Adopted (EX) Orphans
No known homologs
RAR α,β,γ
RXR α,β,γ
PPAR α,β,γ
LXR α,β
FXR α,β
BXR α,β
ERR α,β,γ
DAX-1
SHP
GCNF
knirps
knirps-related
egon
DHR96
C. elegans
D. melanogaster
H. sapiens
Arabidopsis
~250 nuclear receptors
~20 nuclear receptors
~48 genes
no family members
all-trans retinoic acid
9-cis retinoic acid
fatty acids, eicosanoids
oxy-sterols
bile acids
benzoates
Nearly adopted orphans (natural ligands?)
CAR
SXR/PXR
androstanes, xenobiotics
steroids, xenobiotics
EDCs and the obesogen hypothesis
• Obesogens - chemicals that inappropriately stimulate adipogenesis
and fat storage, disturb adipose tissue homeostasis, or alter control
of appetite/satiety to lead to weight gain and obesity
• Pre- and postnatal exposure to EDCs such as environmental estrogens
(ER) increases weight
– DES, genistein, bisphenol A
• Thiazolidinedione anti-diabetic drugs (PPARγ)
– Increase fat storage and fat cell number at all ages in humans
• Urinary phthalates correlate with waist diameter and insulin
resistance in humans
– Many chemicals linked with obesity in epidemiological studies
• several compounds cause adipocyte differentiation in vitro (PPARγ)
– phthalates, BPA, aklylphenols, PFOA, organotins
• Existence of obesogens is plausible
Endocrine disruption by organotins
• Organotins -> imposex in
mollusks
• Sex reverses genetically female
flounder and zebrafish -> males
• Which hormone receptors
might be organotin targets?
• We found that tributyltin (TBT)
– Binds and activates at ppb (low nM)
two nuclear receptors, RXR and
PPARγ critical for adipogenesis
Cl
Sn
– TBT induced adipogenesis in cell
culture models (nM)
– Prenatal TBT exposure led to
weight gain in mice, in vivo
Tributyltin-Cl
Structures of RXR and PPARγ-specific agonists
H
N
O
S
9-cis-RA
Kd = 1 nM
EC50 = 15 nM
O
H
COOH
Cl
O
Sn
Tributyltin-Cl
Kd = 12 nM
EC50 = 5 nM
N
LG268
Kd = 3 nM;
EC50 = 3 nM
COOH
N CH3
N
Rosiglitazone
Kd =
47 nM;
EC50 = 300 nM
Organotins show strong SAR on hRXR
80
70
Fold Activation
60
LG268
Butyltin
Dibutyltin
Tributyltin
Tetrabutyltin
Butyltin Tris(2-EHA)
EC50
50
DBT
TBT
4BT
40
30
> 2800 nM
5 nM
150 nM
20
10
0
0.01
0.1
1
10
100
1000
10000
Concentration nM
Grun et al., Molec Endocrinol, 2006
TBT activates PPAR
10
9
Fold Activation
8
Troglitazone
TBT
LG268
AGN203
7
6
toxic
5
4
3
2
1
0
None
1
10
100
1000
10000 100000
Concentration (nM)
PPAR –regulates lipid metabolism and adipocyte differentiation
Grun et al., Molec Endocrinol, 2006
Nuclear receptor activation by organotins
Nuclear Receptor LBD EC50 nM
Ligand
hRXRα
hRARα
hPPARγ
2-5
na
na
AGN203
0.5-2
na
na
9-cis RA
15
all-trans RA
na
8
na
3000
na
na
Tributyltin chloride
3-8
na
20
Tetrabutyltin chloride
150
ND
ND
Di(triphenyltin) oxide
2-10
na
20
Butyltin-tris (2-ethylhexanoate)
na
ND
ND
Troglitazone
na
na
1000
LG268
Butyltin chloride
Dibutyltin chloride
na
na
na
na
RXR
PPAR
Organotins are highly potent nuclear receptor agonists
Do they bind to the receptors?
Competitive Binding of TBT
3000
his6-hRXR
2500
Specific Bound cpm
his6-hPPAR
7000
6000
2000
Kd = 12.5 nM
5000
Kd = 300 nM
4000
1500
Kd = 7.5 nM
Kd = 20 nM
3000
1000
2000
500
0
LG268
TBT
0
0
0.1
Troglitazone
TBT
1000
1
10
100
Concentration nM
103
0
1
10
100
103
104
Concentration nM
• TBT binds to and activates RXR and PPARγ with high affinity
• How does it behave in adipogenic models?
Grun et al., Molec Endocrinol, 2006
What is the effect of TBT treatment, in vivo?
Newborn Liver ± TBT (in utero)
Vehicle (corn oil)
TBT
What is the effect of prenatal TBT exposure on adult animals?
Grun et al., Molec Endocrinol, 2006
TBT increases testis fat pad weight
at 10 weeks
Weight (grams)
0.4
16% increase
p = 0.037
0.3
0.2
0.1
0.0
Control
n=9
TBT
n=10
Fat depot size increases at the expense of overall body mass
Grun et al., Molec Endocrinol, 2006
How does TBT exposure cause weight gain?
Hypertrophy
• Changes in the hormonal control of
appetite and satiety?
• Altered ability of adipocytes to
process and store lipids?
adipocytes
Preadipocytes
• Increased number of adipocytes
or pre-adipocytes?
Hyperplasia
Commitment
differentiation
• Mesenchymal stem cells (MSCs) (now called multipotent stromal cells)
precursors to many lineages including bone, cartilage, and adipose.
– MSCs differentiate into adipocytes following rosiglitazone exposure
– MSCs may (or may not) home to adipose depots after induction
• Hypothesis: TBT induces adipogenesis in MSCs
TBT induces adipogenic differentiation in MSCs
hMSCs
BM
mBMSCs
WAT
+ MDII
mADSCs
+DMSO, TBT or ROSI
Adipocyte
Bone
Cartilage
differentiation
conditions
+ TBT
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
TBT induces adipogenic genes in MSCs
Induction : MDII
+0 nM
+DMSO
+100 nM
+DMSO
+100 nM
+100 nM
T0070907
Adipogenic effects of TBT and ROSI in MSCs
require PPARγ
+100 nM TBT +1000 nM ROSI
+10 nM
+100 nM
+1000 nM
T0070907
+0 nM
+1000 nM
T0070907
Induction : 100 nM ROSI + MDII
Induction : 50 nM TBT + MDII
+0 nM
+10 nM
+100 nM
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Osteogenic capacity of hADSCs
x70 x240
*** ***
Osteo
6
Alizarin Red-S + Sudan Black
ratio Target / Housekeeping
Control (-)
5
4
*
3
2
***
***
1
*
*
0
Osteo + Rosi
Osteo + TBT
OPN
OSN
aP2
LEP
TBT overrides the effects of the bone-inducing cocktail, instead
causing the cells to become adipocytes
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Effects of TBT on cultured MSCs
• TBT increases the amount of adipocyte differentiation in ADSCs
– Increased number of cells with lipid
– Increased amount of lipid stored in cells
– Decreased expression of adipgenesis inhibitor Pref-1
– Increased expression of pre- and adipocyte markers
• Adipogenic effects of TBT and ROSI require PPARγ
– TBT and ROSI rescue effects of PPARγ antagonist
– TBT acts through PPARγ
• TBT inhibits ability of osteogenic cocktail to induce ADSCs to
become adipocytes
• What is the effect of prenatal exposure on ability of ADSCs to
differentiate into adipocytes or other lineages?
In vivo assays
to assess
stem cell
commitment
E16 – chemical exposure by gavage
CMC
TBT
ROSI
C57BLK6 - Pregnant dam
=
CD-1 unexposed
surrogate
in utero exposed offspring
BM
mBMSCs
WAT
mADSCs
+DMSO, TBT or ROSI
adipocyte differentiation
conditions
BM
mBMSCs
WAT
mADSCs
+DMSO, TBT or ROSI
bone differentiation
conditions
BM
mBMSCs
WAT
mADSCs
+DMSO, TBT or ROSI
cartilage differentiation
conditions
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Prenatal TBT exposure increases MSC
differentiation into adipocytes
In utero TBT exposure inhibits osteogenesis
In utero gavage treatment
+DMSO
+TBT
O
+TBT
A
O
+DMSO
+TBT
A
O
A
0.6
Lipid accumulation
60
*****
40
***
***
0.4
***
***
***
0.2
***
***
40
***
*** 20
**
0
Fabp4
ratio Target / Housekeeping
80
0
0.8
In utero CMC
In utero ROSI
In utero TBT
100
20
1
0
Calcification
staining (%surface)
• Prenatal TBT
exposure inhibits
calcium, and
enhances lipid
deposition
+DMSO
TBT
1000
**
800
***
600
400
200
**
0
**
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
• Prenatal TBT
exposure predisposes
MSCs to become
adipocytes at the
expense of their
ability to form
osteocytes
ROSI
ratio Target / Housekeeping
CMC
OPN
Effects of prenatal TBT exposure on WAT and BMD
Ab Fat wgt (g)
0.066
1.6
Males
Females
y = -0.0065x + 0.0651
R² = 0.1826
0.064
0.062
1.4
0.06
1.2
Males
0.058
1.0
0.056
0.8
0.054
0.6
0.052
0.05
0.7
0.9
1.1
1.3
1.5
0.07
0.07
0.06
Males
Females
y = -0.0302x + 0.0877
R² = 0.703
0.07
BMD
BMD
0.08
0.06
0.06
Females
0.05
0.05
0.04
0.05
0.04
0.8
1
1.2
Ab Fat wt
• Prenatal TBT leads to increased WAT and lower BMD
• What is the mechanism?
1.4
• TBT exposure biases the MSC compartment toward adipocytes
– 7-15% more pre-adipocytes in TBT-treated than control animals
• Increased expression of adipocyte markers -> more pre-adipocytes
– Decreased potential to form osteoblasts
• TBT exposure may have altered setpoint for adipocyte number
– Permanent?
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Effects of prenatal TBT on MSC pool
How does prenatal TBT exposure promote
adipocyte differentiation?
Effects of in utero TBT
exposure on adipogenic
pathway genes
uninduced + TBT 14D
ADIPOQ
LEP
Resistin
IRS-2
PPARγ2+/-
PPARγ2+
Fabp4+
Fabp4+
LEP+
LEP+
Pref1-
Pref1-
GyK+
GyK+
PEPCK+
/
/
LPL+
/
ADIPOQ+
Kirchner et al., 2010 Molec Endocrinol, 24, 526-539
Epigenetic effects of prenatal TBT exposure on
promoter methylation of PPARγ target genes
How does TBT affect PPARγ regulators?
Ezh2
KLF4
Zfp423
BMP
SIRT1
SMART
NCoR
RIP140
CBP/p300
SRC
PGC-1α
ADIPOQ
• Zfp423 regulates PPARγ expression (Gupta et al. 2010)
• BMP4 activates C/EBPβ (Bowers et al. 2007)
• Ezh2 represses Wnt during adipogenesis by methylating H3K27 at its
promoter (Wang et al. 2010)
LEP
Resistin
• Wnt10b represses adipogenesis, repressed by Ezh2 (Wang
et al. 2010)
IRS-2
• Wnt5b promotes adipogenesis by inhibiting Wnt/ β-catenin pathway
(Christodoulides et al. 2008)
• Sox9 represses C/EBPβ/δ activity (Wang et al. 2009)
Obesogen exposure and development
• Organotins are exceptionally potent agonists of RXR and PPAR at
environmentally-relevant levels (ppb)
– ~5 nM EC50, 12.5 nM Kd on RXR
– ~20 nM EC50 and Kd on PPAR
• TBT drives adipocyte differentiation in cell culture models
• TBT exposure during development induces adipogenesis in two
vertebrate species: mouse and Xenopus
– Inhibits bone formation in culture and in females
• The effects of maternal TBT exposure are multi-generational in
females and fully trans-generational in males
– Fat depot size, gene expression but little effect on total weight
• Multiple potential modes of action
– PPARγ-RXR
– Aromatase expression/function – estradiol levels
– Glucocorticoid levels
– Other stressors?
Conclusions – organotins and obesity
• Is organotin exposure a contributing factor for obesity?
– Adult exposure rapidly induces adipogenic genes
• Drugs that activate PPARγ increase obesity
– Prenatal TBT exposure permanently alters adult phenotype
– Prenatal TBT exposure recruits MSCs to adipocyte lineage and
diverts them from bone lineage
• Are humans exposed to sufficient levels of TBT for concern?
– PVC is up to 3% w/w (0.1 M) organotins
– Prevalent contaminants in dietary sources
– Fungicide on high value crops, used in water systems
– Average blood level of 27 nM in 32 random people tested
– TPT levels from ~0.5–2 nM in Finnish fishermen
• Human exposure to organotins may reach levels sufficient to
activate high affinity receptors
– 1000 x lower dose than natural dietary RXR and PPAR ligands
Is the environment making us fat?
Mechanisms that promote adipose development
(and where EDCs can potentially act)
EDCs can
promote
the adipogenic
lineage at the
expense of
other lineages
brown
fat
bone
Me
Adipogenic genes ON
PPAR, fatty acid
binding protein 4,
lipoprotein lipase
(activating
enhancer
mark)
cartilage
muscle
CD24+ SMA
CD29+
PPAR+
CD34+
PDGFR+
Sca1+
lin- NG2+
White
adipose
precursors
BMP
WNT
Ac
Osterogenic, Chondrogenic,
Myogenic genes OFF
preadipocyte
Zfp423
EDCs can regulate major
signaling pathways (e.g.,
BMP, WNT) that commit
MSCs or progenitor cells
to the adipogenic lineage
EDCs can alter
the chromatin
landscape to favor
the transcriptional
activation of
adipogenic genes
osteopontin,
type II collagen,
myosin heavy chain
activation of nuclear
receptor PPAR
Upregulation
of PPAR
Expression
EDCs can promote the differentiation
of pre-adipocytes via direct activation
of PPAR (e.g., phthalates, organotins)
mature
adipocyte
EDCs can induce
lipogenesis and
inhibit lipolysis
in the adipocyte,
increasing the
storage of fat
Implications For Human Health
• Diet and exercise are insufficient to explain obesity epidemic
particularly in the very young
• Obesogens inappropriately stimulate adipogenesis and fat storage
– Prescription drugs
• Thiazolidinediones, atypical antipsychotics, anti-depressants
– Environmental contaminants
• organotins, estrogens (BPA, DEHP), PFOS, DDE, POPs
• Prenatal obesogen exposure reprograms exposed animals to be fat
– Epigenetic changes alter fate of stem cell compartment -> more
preadipocytes and more adipocyte progenitors
– Effects can be trans-generational
• Obesogens shift paradigm from treatment to prevention during
pregnancy, childhood and puberty
– Reduced exposure to obesogens, optimized nutrition
– Obesity is intractable once established
• UCI - Blumberg Lab
Rachelle Abbey
Sathya Balanchadr
Stephanie Casey
Raquel ChamorroGarcia
Connie Chung
Amanda Janesick
Jasmine Li
Hang Pham
Peggy Saha
• Former lab members
Connie Chow
Felix Grün
Tiffany Kieu
Séverine Kirchner
Sophia Liu
Lauren Maeda
Michelle Tabb
Gina Turco
Zamaneh Zamanian
Changcheng Zhou
• NINS – Okazaki, Japan
Taisen Iguchi
Hajime Watanabe
• NIHS - Tokyo, Japan
Jun Kanno
• University of Tokyo
Satoshi Inoue
Kotaro Azuma
• UCI collaborators
Olivier Cinquin
David Fruman
Matt Janes
Ed Nelson
Eric Potma
Pathik Wadhwa
Funding from NIEHS, US-EPA, UC-TSR&TP
Human Studies Supporting the Obesogen Hypothesis
• Prenatal & early life exposures to low levels of PCBs and DDE are
associated with increased weight in boys and girls at puberty (Gladen
et al, J. Pediatr., 2000).
• Childhood obesity is associated with maternal smoking in pregnancy
(Toschke et al, Eur J Pediatr 2002)
• Soy-based formula in infancy is a potential risk factor for overweight
later in life (Strom et al., JAMA, 2001; Stettler et al., 2005).
• Concentrations of urinary phthalate metabolites are associated with
increased waist circumference and insulin resistance in adult US males
(Stahlhut et al, EHP, 2007)
• Exposure to HCB during pregnancy increases the risk of overweight in
children aged 6 years ( Smink et al, Acta Paediatrica, 2008)
• Intrauterine exposure to environmental pollutants (POPs) increases
body mass during the first 3 years of life (Verhulst et al EHP, 2009)
• Prenatal exposure to DDE is associated with rapid weight gain in the
first 6 months and elevated BMI later (Mendez et al EHP, 2011)