Peroxisome-Derived Lipids Regulate Adipose Thermogenesis by Mediating Cold-Induced Mitochondrial Fission

Supplemental Material for:

Peroxisome-derived lipids regulate adipose thermogenesis by mediating cold-induced mitochondrial fission

Hongsuk Park, Anyuan He, Min Tan, Jordan M. Johnson, John M. Dean, Terri A. Pietka, Yali Chen, Xiangyu Zhang, Fong-Fu Hsu, Babak Razani, Katsuhiko Funai, and Irfan J. Lodhi

A Preadipocytes B Adipocytes WAT (Chow) BAT (Chow) 4 WAT (HFD) BAT (HFD) 280 * 250 * 240 BAT SVF Cells iWAT SVF cells 200 * 200 level 150 3 * 160 * *

expression 100 120 * expression

80 50 mRNA 2 40 20 *

mRNA 10 * mRNA 15 8 6 * 10 1

4 Relative 5 * * 2 Relative 0 Relative 0 0 PMP70Pex7 Pex16 aP2 PMP70Pex7 Pex16 aP2 Pex7 Pex13 Pex16 PMP70

C D Control PRDM16-AKO Chow HFD 1.5 p=0.131 PMP70 p=0.073 p=0.138 Pex16 1.0 expression Pex14

Actin mRNA 0.5 ** Relative 0.0 Prdm16 Ucp1 Pex16 Pmp70Pex7

Supplemental Figure 1. Peroxisomal biogenesis factors increase during adipogenesis and with high fat feeding. (A) mRNA levels of peroxisomal biogenesis genes and aP2 in BAT and iWAT stromal vascular fraction cells before and after adipogenesis; n=3. (B) mRNA levels of peroxisomal biogenesis genes in gWAT and BAT of wild-type C57 mice fed normal chow diet or high fat diet (HFD) for 8 weeks; n=5. (C) Western blot analysis in BAT of C57 mice fed chow or HFD for 8 weeks. Each lane represents a separate mouse. (D) Gene expression analysis in BAT of PRDM16-AKO and control mice subjected to cold exposure for 2 days; n=3-4. Data are expressed as mean±SEM and were analyzed by Student’s t-test; *p<0.05 Pex16/PMP70/DAPI/Merged Pex16/COX IV/DAPI/Merged Supplemental Figure 2. /Immuno uorescence/ / analysis of Pex16 in brown/ adipocytes./ Immortalized/ BAT SVF cells were di erentiated into adipocytes and immunostained using a rabbit polyclonal anti-Pex16 antibody together with either mouse monoclonal anti-PMP70 or mouse monoclonal anti-COX IV antibody. Following incubation with uorescently-tagged secondary antibodies, the cells were analyzed using a Zeiss Axio Imager Z2 uorescent microscope. A B C Chow diet 30 Control Pex16-AKO 2.5 ) 80 Control 25 2.0 Pex16-AKO (%BW

s 20 60 1.5 15 Gram 10 weight 1.0 40 %Cells * 5 0.5 gWAT 20 0 BW Fat mass Lean mass 0.0 ControlPex16-AKO 0 HFD <1.6 1.6-6.46.4-25.6 D 60 Adipocyte size (µm2 X 1000) Control E 5 Pex16-AKO F 4000 Control 4 Pex16-AKO 40 (g/day) s * 3 3000

2 %Cell intake 2000 20 (counts/6hr) 1 Food * 0 1000

ControlPex16-AKO Activity 0 0 <1.6 1.6-6.46.4-25.6>25.6 RT Cold Adipocyte size (µm2 X 1000) G H 500 Control Pex16-AKO 300 Control Pex16-AKO

250 400 mg/dL) mg/dL) 200 300 e(

e( C on tr ol * 150 200

Glucos C on tr ol

Glucos 100 3 00 100 50 3 00 0153045607590 105 120 0153045607590 105 120 I K Control Pex16-AKO J 3000 0.85

) 2500 ) 2000 0.80 2000 1500 R 0.75 RE (ml/kg/hr (ml/kg/hr 1000 2

2 1000

O 0.70 500 VO VC 0 0 0.65 Light cycle Dark cycle Light cycle Dark cycle Light cycle Dark cycle Supplemental Figure 3. Characterization of Pex16-AKO mice. (A) Body composition analysis in chow fed control and Pex16-AKO mice; n=5. (B) Weight of gWAT depots in chow fed mice; n=3. (C-D) Adipocyte size distribution in chow or HFD fed mice based on sizing data obtained using a Beckman Coulter Multisizer III; n=3. (E) Food intake in chow fed mice; n=5. (F) Physical activity measured at normal room temperature and in the cold using an InfraMot TSE system; 2 5n=3.0 (G) GTT in mice fed normal chow diet; n=11. Error bars in Pex16-AKO mice are smallerC than the symbols.on (H) GTTt inr ol mice fed HFD diet for 14 weeks; n=10-11. (I-K) Measurement of VO2, VCO2 and respiratory exchange ratio (RER) in 2 5Pex16-AKO0 and control mice fed HFD for 15 weeks; n=3-4. Data are expressed as mean±SEM and were analyzed 3 0by Student’s0 t-test; *p<0.05. e

B ) o A 1.5 SC 22 C o

m 14 14 4 C

Pex16 KD µ (

r 12 12 e t e 10 10 * fluorescenc 1.0 m ) a i Ψ 8 8 d m

t ∆ ( e

l 6 6 0.5 p o

r 4 4 d

i 2 2 p i TMRE/Mito-Green 0.0 L 0 0 VehicleFCCP Control Pex16-AKO Control Pex16-AKO

C Control Pex16-AKO D kDa Control Pex16-AKO UCP1 33 Drp1 17 COX IV UCP1

37 25 20 15

Supplemental Figure 4. E ects of Pex16 inactivation on mitochondria and lipid droplets in brown and beige adipocytes. (A) Mitochondrial membrane potential measured using TMRE in di erentiated BAT SVF cells treated with scrambled or Pex16 shRNA, followed by incubation with FCCP or vehicle control. TMRE uorescence was normalized to MitoTracker Green uorescence; n=16 per condition. (B) Diameter of lipid droplets measured in TEM images of BAT from control and Pex16-AKO mice kept at room temperature or subjected to cold exposure for 2 days. (C) Western blot analysis of UCP1 and COX IV protein expression in control and Pex16-AKO mice subjected to cold exposure for 2 days. Each lane represents a separate mouse. (D) Western blot analysis of Drp1 and UCP1 levels in the mitochondrial fractions isolated from BAT of control and Pex16-AKO mice subjected to cold exposure for 2 days. Each lane represents a separate mouse. Data in panels A and B are expressed as mean±SEM and were analyzed by Student’s t-test. *P<0.05. A. C.

Pex16 KD SC GNPAT KD SC 1.5 1.25 level level 1.00 1.0 0.75 mRNA mRNA 0.50 0.5

0.25 * Relative * Relative 0.0 0.00 GNPATaP2 CD36 Pex16 aP2 CD36 B. D. SC Pex16 KD SC GNPAT KD

Supplemental Figure 5. E ect of Pex16 or GNPAT knockdown on adipogenesis. (A and C) Immortalized BAT SVF cells were subjected to adipogenesis for 4 days and then treated with the indicated shRNA and analyzed for gene expression 5 days later. (B and D) Oil red O staining in the cells in panels A and C. Data in panels A and C are expressed as mean±SEM and were analyzed by Student’s t-test. *P<0.05. A B Palmitate oxidation normalized to CS activity Citrate synthase Activity

y * 2000 p = 0.17 800 activit rotein 1500 600 mp

1000 400 palmitate/ synthase

500

nmol 200 itrate mol/min/gra µ tc 0 0 uni ControlPex16-AKO ControlPex16-AKO Supplemental Figure 6. Measurement of citrate synthase activity and palmitate oxidation. (A) Citrate synthase activity measured in homogenates of BAT from Pex16-AKO and control mice; n=7. (B) Palmitate oxidation normalized to citrate synthase activity in BAT; n=6-7. Data are expressed as mean±SEM and were analyzed by Student’s t-test; *p<0.05. A B C D BAT iWAT 2400 200 2000 * SC 1600 150 2.0 Acox2 KD

level 1200 Acox3 KD level 2.0 n.s. 800 100 level 1.5 * 400 * 50 * 1.5 mRNA

200 mRNA

25 mRNA 1.0 n.s. 150 20 1.0 100 15 0.5 Relative

Relative 10 0.5 50 Relative 5 mitoDNA/NucDNA 0 0.0 0.0 0 Acox2Acox3 SC Acox2 KD Acox3 KD Acox1Acox2 Acox3 Acox1Acox2 Acox3

Supplemental Figure 7. Relative expression levels of Acox2 and Acox3 and e ects of their knockdown on mtDNA content. (A-B) Relative mRNA levels of Acox1, Acox2, and Acox3 in BAT and iWAT of wild-type C57 mice; n=3. (C) mRNA levels of Acox2 or Acox3 in BAT SVF cells following treatment with SC, Acox2, or Acox3 shRNA.; n=3-4. (D) mtDNA content normalized to nuclear DNA in the cells in panel C; n=3-4. Data are expressed as mean±SEM and were analyzed by Student’s t-test. *P<0.05; n.s., not signi cant. A B

3000 * 0.25 )

o 0.20

2000 rati 0.15 fluorescence 0.10 1000

* Mito-RoGFP 0.05 (Oxidative/Reductive

Mito-RoGFP 0.00 0 SC Pex16 KD SC Pex16 KD SC Pex16 KD Oxidative state Reductive state

Supplemental Figure 8. Measurement of mitochondrial ROS production using a ratiometric redox sensor. (A) Immortalized BAT SVF cells stably expressing lentiviral-encoded mito-RoGFP were di erentiated into adipocytes and treated with SC or Pex16 shRNA. GFP uorescence was measured following excitation at 400 nm (oxidative state) or at 490 nm (reductive state). (B) Ratio of the uorescence measured in panel A. Data are expressed as mean±SEM and were analyzed by Student’s t-test; *p<0.05; n.s., not signicant. A B Control Pex16-AKO WTL Mito 12 COX IV rotein 9 PMP70 gp

Tomm20 6

Catalase cardiolipin/m 3 nmol

0 18:2/18:1/ 18:2/18:2/ 18:2/18:2/ 18:2/18:2/ C 18:1/18:1 18:1/18:1 18:2/18:1 18:2/18:2 60 * 50

rotein 40 gp 30 * PC/m 20

nmol * * 10 *

0 0 1 1 2 3 4 6 1 2 1 4 5 6 1 4 4 4 1 4 4 6: 6: 8: 8: 0: 0: 2: 6: 8: 8: 0: 2: 2: 8: 0: /1 /1 /1 /1 /2 /2 /2 /1 /1 /1 /2 /2 /2 /1 /2 /20: /20: /18: /20: /20: :0 :0 :0 :0 :0 :0 :0 :1 :1 :0 :0 :0 :0 :1 :1 0 0 0 0 0 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 6: 8: 6: 6: 8: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a a p p p

Supplemental Figure 9. Mass spectrometric analysis of mitochondrial phospholipids. (A) Western blot analysis of whole tissue lysate (WTL) and mitochondrial fraction from BAT of control and Pex16 AKO mice. (B-C) Levels of various cardiolipin (B) and phosphatidylcholine (PC) (C) species in the mitochondrial fraction of BAT from control and Pex16-AKO mice subjected to cold exposure for 2 days. The pre xes “a” and “p” in PC side chains denote alkyl ether and plasmalogen species, respectively; n=9-10. Data are expressed as mean±SEM and were analyzed by Student’s t-test; *p<0.05. A B Control Pex16-AKO Control+AG Pex16-AKO+AG Control 2.0 175 Control+AG 150 e 125 Pex16-AKO 1.5

100 Pex16-AKO+AG expression

E 75 n.s.

lP 50 1.0 abundanc 20

* ** mRNA Tota 15 0.5 * 10 * (relative 5 Relative 0 0.0 Diacyl PEPlasmalogen PE Pex16 Pgc1α Nrf1 Tfam Drp1 Fis1 Mff

C Control Pex16-AKO Control Pex16-AKO D Pex16-AKO+AG 15.0 40 Control+AG Control+AG Pex16-AKO+AG n.s. C)

o 38 12.5 ** e( )

r 36 h / 10.0

g 34 / l

m 7.5 n.s.

( 32 2

O 5.0 30 temperatur V 28 2.5 NE 26 Body 0.0 24 o o 0 5 0 -5 0 5 0 5 0 5 0 5 0 5 0 5 0 -2 -1 -1 1 1 2 2 3 3 4 4 5 5 6 22 C4C Time (min) Ambient Temperature Supplemental Figure 10. E ects of AG on mitochondrial lipids, gene expression, oxygen consumption rate and cold tolerance. (A) Total diacyl and plamalogen PE levels in the mitochondria from BAT of control and Pex16-AKO mice treated with or without AG; n=5. (B) Gene expression analysis in BAT of cold-treated mice. (C) Oxygen consumption rate (VO2) measured using indirect calorimetry before and after intraperitoneal norepinephrine (NE) injection in control and Pex16-AKO mice treated with or without AG diet while housed at thermoneutrality for 8 weeks. n=3-4. (D) Cold tolerance was determined by measuring rectal temperature prior to and after 6 hrs of cold exposure in control and Pex16-AKO mice treated with or without AG for 1 week; n=4-6. Data are expressed as mean±SEM and were analyzed by one-way ANOVA, followed by Fisher’s LSD test (panels A and D); Student’s t-test (panel B); or two-way ANOVA with Bonferroni post test (panel C); *p<0.05; n.s., not signi cant. A B C D

Muscle palmitate oxidation Muscle pyruvate oxidation Liver palmitate oxidation Liver pyruvate oxidation 15 40 *** 4 Control 8 n.s. Pex16-AKO *** 30 3 6 10 issue/hour issue/hour issue/hour issue/hour ft ft ft 20 ft 4 2 mo 5 10 2 1 nmol/gra mo nmol/gra mo nmol/gra nmol/gra mo 0 0 0 0 Chow dietChow diet + AG Chow dietChow diet + AG Chow dietChow diet + AG Chow dietChow + AG E ) 60 SC Pex16 KD cells SC f Basal %o 40 n(

Pex16 KD Pex16 20 ranslocatio 4t Glut Insulin SC 0 Basal Insulin Pex16 KD Pex16

mCherry Myc DAPI Merged

Supplemental Figure 11. E ect of AG on palmitate and pyruvate oxidation in skeletal muscle and liver. (A-D) Palmitate and pyruvate oxidation in skeletal muscle (A-B) and liver (C-D) of control and Pex16-AKO treated with or without AG diet for 8 weeks; n=3-4. (E) BAT SVF cells stably expressing lentivirus-encoded myc-Glut4-mCherry were di erentiated into adipocytes and then treated with scrambled or Pex16 shRNA. Five days later, the cells were treated with or without insulin after serum starvation and analyzed for Glut4 translocation using confocal microscopy. Exofacial localization of the Myc tag was detected by staining unpermeabilized cells with a rabbit anti-Myc antibody, followed by Alexa Fluor 488 goat anti-rabbit secondary antibody. The uorescence of mCherry was directly visualized. Quantication of Glut4 translocation is shown to the right of the images. Data are expressed as mean±SEM and were analyzed by one-way ANOVA, followed by Fisher’s LSD test (panels A-D) or by Student’s t-test (panel E); ***p<0.001.

Supplemental Table 1. Sequences of primers of used in quantitative real-time PCR.

Gene Name Forward (5' to 3') Reverse (5' to 3') Acaa1 GTGGCATCAGAAATGGGTCT GGCTTTCTCACTCTCCAGCA Acadm GCCCAGAGAGCTCTAGACGA CAACCTTCATCGCCATTTCT Acox1 GGATGGTAGTCCGGAGAACA AGTCTGGATCGTTCAGAATCAAG Acox2 CCTTCCTAGACCTGCTTCCC TGTCCGTCATAACAGCCAAG Acox3 CTTCTGAGAAACGGGGACAA GCTCGGTAGGCACTAAGAGG Agps GTACCAATGAGTGCAAAGCG CCCCATCCATTCCATTTCAT aP2 GCGTGGAATTCGATGAAATCA CCCGCCATCTAGGGTTATGA CD36 GCGACATGATTAATGGCACA CCTGCAAATGTCAGAGGAAA Cpt1 CCTGGGCATGATTGCAAAG ACGCCACTCACGATGTTCTTC CS TGCAGCTGTAGCTCTCTCCC AAGGACGAGGCAGGATGAG CtsB GTTTGGCTTGGTGGCTGC CAGCAGGCAAGAAAGAAGGA Cyp4a10 CCCTGATGGACGCTCTTTAC AGAGTCTGGTGCAAACCTGG Cyp4a14 GCCATGCTTATCTGCCATTC ATCTGGCAGCAATTCAAAGC Drp1 AGATCGTCGTAGTGGGAACG AGAATGAGAGGTCTCCGGGT Fis1 ATTTGAATATGCCTGGTGCC CATAGTCCCGCTGTTCCTCT Gamt GCAGCCACATAAGGTTGTTCC CTCTTCAGACAGCGGGTACG Gatm GACCTGGTCTTGTGCTCTCC GGGATGACTGGTGTTGGAGG Gnpat CATGGACGTTCCTAGCTCCT CCCACTTTTTGAGGTCTTTCG Hsd17b4 AGGCTAGACTCATGGCTTCG TGAAGTCCCCTCCTAAGTCG L32 TTCCTGGTCCACAATGTCAA GGCTTTTCGGTTCTTAGAGGA LipA AGTATTCACCGAATCCCTCG CTAGAATCTGCCAGCAAGCC Mff GGAGTTCCAAATGCCAGTGTGATA TGGATAAGGTCAAGATCTGCTGGT Mfn1 TGAATAACCGTTGGGATGCT GCTTCCGACGGACTTACAAC Mfn2 GTCCTGGACGTCAAAGGGTA ATTGATCACGGTGCTCTTCC MT-CO1 TCCTACCACCATCATTTCTCC CTGATGCTCCTGCATGGG MT-CO2 CATCAAACCGACCAGGGTT AATTATTGAAGCAGATCAGTTTTCG MT-ND6 CCAACATAACTCCAACATCATCA GTATTGGGGGTGATTATAGAGGTTT Nrf1 TAGTCCTGTCTGGGGAAACC CTGGTACATGCTCACAGGGA Opa1 TTTGCAGAAGACGGTGAGAA TCCTGGTGAAGAGCTTCAATG Pex13 CCCAAACCCTGGGAGTCT GGGCACTCTTGTAAGTGTTGG Pex16 CCGTTCTATGACCGCTTCTC GGAGGGCAAGTAGTCCATGA Pex16 (-4) GGAGCTCCCGGTCTCGGGAG CTAAGCCCGCCTCCCGGTTG Pex16 (-601) GGAGGCCAGCTTAGAACCTG CGATCTGCACCTCGGGTTTC Pex16 (-1228) GGGTTTCTCTGTGTACCTCTGG GCTCGTTTAACCGGGCAGTG Pex16 (-1790) CCTAAGCCAGAGGCAGCTGG CCTTATCGCTTGGATTAGAGTCCTG Pex16 (-2372) TCTGCTCTTTGTCTCTGGAGCC CCAACAGTGAAGAGTCATCTCTCTAGG Pex19 AGCATCATGCAGAACCTCCT TGCTGCTGCTGGTACTTCTC Pex7 GAAATTTTCACCGTTCCACG GTGATGCTCCACTGTTTCCA Pgc1a TGTAGCGACCAATCGGAAAT TGAGGACCGCTAGCAAGTTT Pmp70 TCTGCCTACTCCATAAGCGG CACCACAGCTCGCTCTTTCT PPARa AGTTCGGGAACAAGACGTTG CAGTGGGGAGAGAGGACAGA Prdm16 CAGAGGTGTCATCCCAGGAG ACGGATGTACTTGAGCCAGC RPL30 Cell Signaling Technology, #7015 Cell Signaling Technology, #7015 Ryr2 ATGGCTTTAAGGCACAGCG CAGAGCCCGAATCATCCAGC Gene Name Forward (5' to 3') Reverse (5' to 3') Serca2b ACCTTTGCCGCTCATTTTCCAG AGGCTGCACACACTCTTTACC Slc6a8 TGCATATCTCCAAGGTGGCAG CTACAAACTGGCTGTCCAGA Tfam CCAAAAAGACCTCGTTCAGC GACAGATTTTTCCAAGCCTCA Tomm20 ACTGCATCTACTTCGACCGC CAGGTAACTTGGAAAGCCCA Ucp1 GTACCAAGCTGTGCGATGTC TCAGCTGTTCAAAGCACACA

Supplemental Methods

Virus Packaging. Plasmids encoding shRNA for mouse Pex16 (18633) and GNPAT

(TRCN0000193539) were purchased from Sigma-Aldrich (Saint Louis, MO). Packaging vector psPAX2 (#12260), envelope vector pMD2.G (#12259), and scrambled shRNA plasmid (#1864) were obtained from Addgene. 293T cells in 10 cm dishes were transfected with 2.66 g psPAX2,

0.75 g pMD2.G, and 3 g shRNA plasmid. After 48 hr, media were collected, filtered using

0.45 m syringe filters, and used to treat cells. After 36 hr, cells were selected with puromycin and knockdown was assessed after an additional 48 hr or as indicated. To make retroviral particles encoding Cre recombinase, 293T cells in 10 cm dishes were transfected with 3 g pBABE-Cre plasmid and 3 g A helper plasmid. After 48 hr, media were collected, filtered using 0.45 M syringe filters, and used to treat 60% confluent iWAT stromal vascular cells.

After 36 hr, the cells were selected with 2 g/ml puromycin.

Pex16 Promoter Reporter Assays. COS7 cells were co-transfected with either Pex16-eGFP and empty plasmid or Pex16-eGFP and HA-PRDM16 using Lipofectamine 2000 (Invitrogen). Three days after transfection, the transfected cells were imaged with a Leica DMI4000B inverted fluorescence microscope. For luciferase reporter assays, COS7 cells were co-transfected with

Pex16-Luciferase and Renilla luciferase plasmids together with either an empty vector or HA-

PRDM16. Cells were harvested 3 days later and luciferase reporter assays were performed using the Dual-Glo® Luciferase Assay (Promega, Madison, WI). Firefly luciferase signals were normalized to the Renilla luciferase signals.

Gene Expression Analysis by Quantitative RT-PCR. Total RNA from mouse tissues or cells was isolated with PureLink RNA Mini Kit (Life Technologies) and reverse transcribed using

High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). For quantitative real- time PCR analysis, PowerUP SYBR Green reagent (Applied Biosystems) was used along with the primers spanning exon-exon boundaries. Relative mRNA expression level was determined using the 2(-Delta Delta CT) method with ribosomal protein L32 as the internal reference control.

Mitochondrial Redox State Assay. BAT SVF cells stably expressing lentiviral-encoded Mito- roGFP were grown to confluence in a 96-well plate before treatment with differentiation media, as previously described (1). At day 4, the cells were treated with scrambled or Pex16 shRNA lentivirus in the maintenance media. After 24 hr treatment, the viral media were removed and replaced with fresh maintenance media. At day 9 of differentiation, the media were removed and cells were washed once with pre-warmed PBS. The fluorescence of GFP was measured using a

525 nm wavelength emission filter following excitation at 400 nm (oxidative state) or 484 nm

(reductive state) using a Tecan Infinite M200 plate reader.

Citrate Synthase Activity. Citrate synthase activity was measured in BAT homogenates using a commercially available kit from Sigma (CS0720). The citrate synthase activity assay was performed in accordance to the manufacturer’s instructions. Raw values for citrate synthase activity were normalized to protein content for each sample.

Mitochondrial Membrane Potential Assay. Mitochondrial membrane potential in Pex16 knockdown and control BAT SVF cells was measured using a TMRE Mitochondria Membrane Potential Assay Kit obtained from Cayman Chemicals (Ann Arbor, MI) according to the manufacturer’s instructions. Briefly, BAT SVF cells cultured in 96-well plates were grown to confluence before treatment with differentiation medium, as previously described (1). On 4 days following the initiation of differentiation, the cells were infected with scrambled or Pex16 shRNA lentivirus in the differentiation medium. After 24 hrs, the viral medium was removed and fresh differentiation medium was added. On day 9, the membrane potential-dependent fluorescence of TMRE was measured and normalized to the fluorescence of MitoTracker Green using a Tecan Infinite M200 plate reader. As a positive control for membrane depolarization, the mitochondrial uncoupler carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP) was added.

Mitochondria Isolation. Mitochondria were isolated from BAT as previously reported (2).

Briefly, after PBS perfusion, BAT was collected from mice, transferred to 1.5 ml tubes, and minced thoroughly with scissors. The tissue was then transferred to a homogenization tube containing 100 l of homogenization buffer (0.25 M sucrose, 20 mM HEPES in distilled H2O) and homogenized for 10 strokes. The pestle was washed with 800 l of homogenization buffer to remove any residual homogenate. The homogenates were transferred to 1.5 ml tubes and vortexed thoroughly before centrifugation at 800 g for 10 min at 4oC. The upper fat cake was removed by forceps and the supernatant (~800 l) was transferred to a new tube and centrifuged at 10,000 g for 10 min at 4oC. The supernatant was discarded and the pellet (mitochondrial fraction) was used immediately for lipid extraction or stored at -80oC for future use.

Electrospray Ionization Mass Spectrometry. Lipids were extracted from adipocytes as previously described (3) and analyzed by mass spectrometry. Structural characterization and quantitation of lipids by low-energy CAD tandem mass spectrometry was conducted on a

Thermo Scientific (San Jose, CA) TSQ Vantage Triple Stage Quadrupole mass spectrometer equipped with Xcalibur operating system as previously described (4, 5).

Histology. For immunofluorescence analysis of PMP70 in Pex16-AKO and control mice, BAT was embedded in Tissue-Tek OCT compound (Sakura Finetek), frozen at -80oC and then 10 µm sections were cut using a Leica cryostat. Sections were placed on glass slides and immunostained using a rabbit anti-PMP70-Atto-488 antibody and images were captured using a Nikon A1Rsi scanning confocal microscope.

Supplemental References 1. Lodhi IJ, Dean JM, He A, Park H, Tan M, Feng C, et al. PexRAP Inhibits PRDM16- Mediated Thermogenic Gene Expression. Cell Rep. 2017;20(12):2766-74. 2. Vernochet C, Mourier A, Bezy O, Macotela Y, Boucher J, Rardin MJ, et al. Adipose- specific deletion of TFAM increases mitochondrial oxidation and protects mice against obesity and insulin resistance. Cell Metab. 2012;16(6):765-76. 3. Bligh EG, and Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959;37(8):911-7. 4. Brugger B, Erben G, Sandhoff R, Wieland FT, and Lehmann WD. Quantitative analysis of biological membrane lipids at the low picomole level by nano-electrospray ionization tandem mass spectrometry. Proc Natl Acad Sci U S A. 1997;94(6):2339-44. 5. Hsu FF, and Turk J. Electrospray ionization with low-energy collisionally activated dissociation tandem mass spectrometry of glycerophospholipids: mechanisms of fragmentation and structural characterization. J Chromatogr B Analyt Technol Biomed Life Sci. 2009;877(26):2673-95.