3’UTRTTSpTSS

dTSS

5’UTRncRNAExon

Intron

Intergenic

Replicate1

Rep

licat

e 2

0

5

10

15

0 5 10 15

Biological Replicates

WTPr

dm16

KO

0

5

10

15

0 5 10 15

WT/Prdm16 KOA B C

00

5

10

15

1 2 3 4 5

Pol

II R

PM

/ kb

Pol II in Gene Body

BAT-selective

WAT-selectiveCommon

# of

PR

DM

16 S

ites

/ 1M

bp

150

100

50

0

BAT-selective

WAT-selectiveCommon

H3K27-Ac.E F

10 kb3

0

4

0

PRDM16 BAT

H3K27-Ac. WAT

4

0

H3K27-Ac. BAT

Agt Retn

3

0

4

0

4

0

10 kb

D

Figure S1. PRDM16 is preferentially bound to BAT-selective genes in BAT

(A) Scatter plot comparing ChIP-seq binding profiles of PRDM16 from independent biological replicates of BAT (B) Scatter plot showing PRDM16 ChIP-seq signal in wildtype (WT) and Prdm16 KO BAT using pooled replicates.(C) Pie chart showing the relative enrichment of PRDM16 in different genomic regions (D) ChIP-seq stack-height profiles in reads per million (RPM) for PRDM16 in BAT and H3K27-Ac. in BAT and WAT at the Agt and Retn loci. (E) Box plot comparing Pol II (RPM/kb) levels within gene bodies of BAT-selective, common and WAT-selective genes with varying number of proximal PRDM16 binding sites(F) Average number of PRDM16 binding sites (per 1Mbp) at BAT-selective, common and WAT-selective regions of H3K27-Ac. enrichment

Number of PRDM16 Sites within 100kb

B

Ins1

Ucp1 (

2.5)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα

(1)

Pparα

(11)

Ppargc

1a (3

8)

Ppargc

1a (4

2)0.000

0.002

0.004

0.006 C/EBPβ > IgGC/EBPβ > PPARγ

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025 C/EBPβ > IgG

C/EBPβ > PRDM16

Ins1

Ucp1 (

2.5)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα

(1)

Pparα

(11)

Ppargc

1a (3

8)

Ppargc

1a (4

2)

Enr

ichm

ent (

% in

put)

0.000

0.002

0.004

0.006

0.008 C/EBPβ > IgGC/EBPβ > MED1

Ins1

Ucp1 (

2.5)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα

(1)

Pparα

(11)

Ppargc

1a (3

8)

Ppargc

1a (4

2)

Enr

ichm

ent (

% in

put)

Enr

ichm

ent (

% in

put)

Figure S2. C/EBPβ, PPARγ, PRDM16 and MED1 are co-localized at BAT-selective loci

(A) De novo motif analysis at regions of PRDM16 enrichment within 100kb of genes positively or negatively regualted by PRDM16(B) Sequential ChIP analysis. Chromatin was first immunoprecipitated using an anti-C/EBPβ antibody. Sequential immunoprecipitation was performed comparing IgG to PPARγ, MED1 and PRDM16. Experiments were performed a minimum of 3 times and representative data is shown.

De Novo Motif P-Value Hit %

10−118

10−57

10−41

10−31

55.6

30.4

49.3

21.7

TF

EBF

C/EBP

NF1-half

DR1-half

PRDM16 - activated genes

De Novo Motif P-Value Hit %

10−97

10−71

10−44

43.5

47.5

61.3

TF

C/EBP

EBF

NF1-half

DR1 10−17 4.2

PRDM16 - repressed genes

A

AC/EBPβ ChIP

Ins1

Ucp

1 (2.5

)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα (1

)

Pparα (1

1)

Pparα (1

4)

Ppargc

1a (3

8)

Ppargc

1a (4

2)0.0

0.5

1.0

1.5PPARγ ChIP

Ins1

Ucp

1 (2.5

)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα (1

)

Pparα (1

1)

Pparα (1

4)

Ppargc

1a (3

8)

Ppargc

1a (4

2)0.0

0.5

1.0

1.5

2.0

WTKO

Enr

ichm

ent (

% in

put)

E

**** **

****

Ins1

Ucp1 (

2.5)

Ucp1 (

4.7)

Ucp1 (

5.7)

Pparα

(1)

Pparα

(14)

Cidea (

13)

Ppargc

1a (3

8)

Ppargc

1a (4

2)

Pparα

(11)

*Enr

ichm

ent (

% in

put)

Pparγ

(122)

*

*

MED1 ChIP

0.00

0.05

0.10

0.15

0.20

0.25

Figure S3. Loss of PRDM16 does not affect PPARγ or C/EBPβ binding levels at BAT-selective genes

(A) ChIP-qPCR analysis of PPARy and C/EBPβ binding at BAT-selective genes in WT and Prdm16 KO BAT(mean ± SEM; n = 3; *p < 0.05)(B) Relative Med1 mRNA levels in epididymal WAT, inguuinal WAT and BAT (mean ± SEM; n = 3; *p<0.05; **p<0.01)(C) Relative Med1 mRNA levels in differentiating brown adipocytes (mean ± SEM; n = 3)(D) Box plot comparing MED1 occupancy changes upon Prdm16 KO in BAT at sites that possess or lack PRDM16 binding (***p < 10−10)(E) ChIP-qPCR analysis of MED1 levels at BAT-sel. genes in WT and Prdm16 KO BAT (mean ± SEM; n = 3; *p < 0.05, **p<0.01

−4

−2

0

2

4

Log 2F

C K

O/W

T

MED1 enrichment Prdm16 KO/WT

Without PRDM16With PRDM16

D

***

0.0

0.2

0.4

0.6

0.8

D0 D2 D4 D6 D8

Med1 mRNAC

0.00

0.05

0.10

0.15

0.20

0.25

Med1 mRNA

Rel

ativ

e m

RN

A le

vels

B

*

**

eWAT

iWAT

BAT

Rel

ativ

e m

RN

A le

vels

Enr

ichm

ent (

% in

put)

Figure S4. PRDM3 binds and recruits MED1 to BAT-selective loci

(A) Relative mRNA expression levels of Prdm16 and Med1 in immortalized brown adipocytes, primary brown adipocytes, and Prdm16-deficient brown adipocytes that have been transduced with Prdm16 (mean ± stdev; n = 3; ***p<0.001)(B)MED1 protein levels from WT and Prdm16 KO adipocytes(C) Relative mRNA levels of BAT-sel. genes in PRDM16-deficient brown adipocytes that were transduced with control or PRDM16-expressing virus. (mean ± Stdev; n = 3; *p < 0.05, **p<0.01, ***p<0.001 )(D) Relative mRNA levels of BAT-selective, common and WAT-selective genes from differentiated primary brown adipocytes that were transfected with siScr or siMed1. (mean ± Stdev; n = 3; *p < 0.05, **p<0.01, ***p<0.001 )(E) ChIP-qPCR analysis of PRDM16 enrichment after acute MED1 deletion at BAT-sel. genes (mean ± Stdev; n = 3). Inset - Western blot analysis of MED1 levels after acute MED1 deletion(F) Commassie blue staining of GST and GST-PRDM16 fusion proteins

GST22

4-454

455-6

80

1-223

681-8

80

881-1

038

1039

-1176

C

2000

4000

6000

8000

10000 ***

Rela

tive

mRN

A le

vel

F

Prdm16

E siScr siMed1

MED1

ACTIN

0.0

0.1

0.2

0.3

0.4

0.51.01.52.0

Ins1

Ucp1 (

2.5)

Ucp1 (

4.7)

Ucp1 (

5.7)

Cidea (

13)

Pparα

(1)

Pparα

(11)

Ppargc

1a (3

8)

Ppargc

1a (4

2)

PR

DM

16 C

hIP

Enr

ichm

ent

(% In

put)

0123

10

20

30

40

50

****

**

**

**

***

**

Pparα

Ppargc

1aCide

aUcp

1Cox

5b

Cox7a

1Cyc

sMed

1

DAcute Knockdown of MED1

0.0

0.5

1.0

1.5

Med1Fabp4

Pparγ

C/ebpβ

Prdm16Pparα

Ppargc1a

CideaUcp

1 AgtTle3

Gpr64Fads

Morc4 Npr3Ccn

d2

*** ** *

**

*** ****

*

0.0

0.5

1.0

1.5

Rel

ativ

e m

RN

A le

vel

Immortalized Cells

0.0

0.5

1.0

1.5

2.0

Primary CellsA

******

MED1

ACTIN

WT KO

B

Prdm16 Med1 Prdm16 Med1

WTKO

WTKO

Rel

ativ

e m

RN

A le

vel

Si-Scr

Si-Med1

Ctl

PRDM16

Si-Scr

Si-Med1

Rank Term P-Value1 brown fat cell dif ferentiation 1.92E-132 fat cell dif ferentiation 2.22E-103 oxidation reduction 4.86E-094 fatty acid metabolic process 7.50E-09

5generation of precursor metabolites and energy 4.84E-08

6energy derivation by oxidation of organic compounds 5.25E-08

7 organic ether metabolic process 6.73E-078 glycerol ether metabolic process 2.94E-069 glycerolipid metabolic process 2.96E-0610 coenzyme metabolic process 2.97E-06

−0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

Pol

II R

PM

2.5

2.0

1.5

1.0

0.5

0

0

1

2

3

4

5

6

None TE SE

Pol

II e

nric

hmen

t (R

PM

/ kb

)SETENone

Relative Position in the gene body

***

*** GO analysis for genes near SE

A B

Ucp1

Cidea

Pparα

Ntrk3

Pank1

Aspg

Kcnk3

Cpn3

Cox7a

1Cox

5bCyc

s

0.5

1.0

1.5

2.0

VehicleJQ1JQ1 (-)

Super Enhancers Typical Enhancers

*** *** *** *** *** ******

***

***

**

Rel

ativ

e m

RN

A le

vels

**

CJQ1- treatment

Figure S5. PRDM16 regulates Super Enhancers in BAT

(A) Average Pol II profile (RPM) around gene bodies of genes associated with super enhancers (red), typical enhancers (black), or genes without proximal MED1 occupancy (grey). Insert – box plot representation of Pol II signal (RPM/kb) of same data (***p < 10−50) (B) DAVID gene ontology (GO) analysis for super enhancer-associated genes in WT BAT(C) Relative mRNA levels of genes marked by SE or TE in differentiated primary brown adipocytes that were treated with either vehicle, JQ1 or an inert JQ1 analog (JQ1(-)) (mean ± Stdev; n = 6; *p < 0.05, **p<0.01, ***p<0.001)

Table S1 Primers used for real-time qPCR and ChIP-qPCR analysis

mRNA Fwd Rev

Agt AAGACCCTGCATGATCAGCTC CTTCCTGCCTCATTCAGCATC

Aspg ATTGCCTTCAGG GGCTGTGAC CTGGCCCAGCACATAGGACAGT

C/ebpβ ACGACTTCCTCTCCGACCTCT CGAGGCTCACGTAACCGTAGT

Ccnd2 AGCTGTCCCTGATCCGCAAGC AAGCCCACAGATGGCTGCTCC

Cidea TGCTCTTCTGTATCGCCCAGT GCCGTGTTAAGGAATCTGCTG

Cox5b GCTGCATCTGTGAAGAGGACAAC CAGCTTGTAATGGGTTCCACAGT

Cox7a1 CAGCGTCATGGTCAGTCTGT AGAAAACCGTGTGGCAGAGA

Cpn3 GCCGACATCCCTCCGGACATC TCCAGGTGACGGACCTGAGTGT

Cycs GCAAGCATAAGACTGGACCAAA TTGTTGGCATCTGTGTAAGAGAATC

Fabp4 ACACCGAGATTTCCTTCAAACTG CCATCTAGGGTTATGATGCTCTTCA

Fads TGCGCAAGTTCCTGAAACCCCT CATGTCTTCAGCTGCCTGGCG

Gpr64 CCACACCAGCCCCATCTGTCC TCCATCTGGGATACTTGGGCTTCC

Kcnk3 GCTTCGCCG GCTCCTTCTACTT CTAGTGTGAGCGGGATGCCCAG

Med1 TGCTTGGAAAATTCCTCAAAA GATGTCAAAGTGGCTCACCA

Morc4 GTATCCAGGCGCAGGCGGTTA AGATGGCTTCTAGGCTGGGCA

Npr3 GTGCGCTACATCCAAGGCAGC TCCACTGGTCATGCCGTGTCTG

Ntrk3 ATGCGAGCCCTACACCTCCTA GACTGCTATGGACACCCCAAA

Pank1 CGCTGTTCGCCCAGCATGATTC CAGCTTAACCAGGGTTCCACCGA

Ppargc1a CCCTGCCATTGTTAAGACC TGCTGCTGTTCCTGTTTTC

Pparα GCGTACGGCAATGGCTTTAT GAACGGCTTCCTCAGGTTCTT

Pparγ2 TGGCATCTCTGTGTCAACCATG GCATGGTGCCTTCGCTGA

Prdm16 CAGCACGGTGAAGCCATTC GCGTGCATCCGCTTGTG

Tbp GAAGCTGCGGTACAATTCCAG CCCCTTGTACCCTTCACCAAT

Tle3 CTCACCCACACCCCGCATCAA AAACCGCAATGTTCCCGTCGC

Ucp1 ACTGCCACACCTCCAGTCATT CTTTGCCTCACTCAGGATTGG

ChIP Fwd Rev

18s AGTCCCTGCCCTTTGTACACA CGATCCGAGGGCCTCACT

Cidea (13) GGCCACTTGAGGAGCCAACCA TGGGCACTGGCCTTGTAGCTG

Fabp4 GACAAAGGCAGAAATGCACA AATGTCAGGCATCTGGGAAC

Ins1 GGACCCACAAGTGGAACAAC GTGCAGCACTGATCCACAAT

Ppargc1a (38) TCCGAGTTTCCCTGCTGTGGC AGGGACTTGCAGCTGTGGTGG

Ppargc1a (42) GAGGTGGCACCAGGACACCAG CCCAAGCTCGAGACTCCGCTC

Pparα (1) GGGGCATGTGCATTCCGTGAC CACTGGGGCTCTGCCAACTGA

Pparα (11) AAGAGCATGGGACAGTGGCCG TGGCCAGCTGAAGGTCACCAC

Pparα (14) CCTGCCCCATAGGCAGTATGGTC ACAGGGGCAGAAGCCAAGCTG

Pparγ (122) AGCTTTGCTGGCTAGAGGTG TTTCGCAGAACTGAGGTTGA

Ucp1 (2.5) CAAATGGTGACCGGGTGCCCT GGGTGACTGACCCTCTGTGACG

Ucp1 (4.7) CCCCACTGCCTGTCACGTTCA GAAGCTGCCGAATGGTGCGTC

Ucp1 (5.7) ACCACACCATTTGGAGCCTGAC TGAGTTTGCAGGGAGGATGGGC