SUPPLEMENTARY INFORMATION

Transcription

1 Figure S1. The expression of Plgf, Vegfa, Vegfc or Myh7 generates lower correlation coefficients to the mitochondrial cluster, as compared to Vegfb. The intensity of Plgf, Vegfa, Vegfc and the myocyte marker Myh7 expression (black), plotted against the mean expressional intensity of the mitochondrial cluster (red) in 708 microarrays, as described in the Methods Summary section. The Pearson correlation coefficients (r) for each comparison are listed under the respective graphs. 1

2 Figure S2. Vegfb has similar expression pattern as Cycs in an array of mouse tissues. Relative mrna expression of Vegfb and Cycs in 7 wt mouse tissues analyzed by qpcr. Data in is shown as percentage of expression compared to the ribosomal L19 gene (n=3 mice). Primer sequences are listed in Table S3. Figure S3. VEGF-B does not regulate mitochondrial gene expression or copy number. a, Relative mrna expression of Ndufa5 and Cycs in hearts from fed or fasted wt and VEGF-B -/- mice. Data is shown as percentage of expression compared to the ribosomal L19 gene (n=3 mice). c, Mitochondrial DNA copy number in hearts of wt and VEGF-B -/- mice. Total DNA from hearts of wt and VEGF-B -/- mice was phenolchloroform extracted. The final aqueous phase was used to assess mitochondrial DNA (mtdna) content relative to nuclear DNA (ndna) content by qpcr, using 2 ng of total DNA as template and primers for cytochrome c oxidase II (Cox II, mitochondrial genome) and Gapdh (nuclear genome). Primer sequences are listed in Table S3. 2

3 Figure S4. Mouse transformed ECs express combinations of Fatp1, Fatp3 and Fatp4, whereas human primary and transformed ECs express Fatp1, Fatp3, Fatp4 and Fatp5. mrna expression of FATPs, VEGFR1 and NRP1 in different mouse and human endothelial cell lines. mrna was extracted from cultured cells as described in Supplemental Methods, and analyzed by conventional RT-PCR using specific primers listed in Table S3. bend3, mouse brain-capillary derived ECs (polyoma mt immortalized); MS-1, mouse pancreatic ECs (SV40T immortalized); HDMEC, primary human dermal microvascular ECs; HUVEC, primary human umbilical vein ECs; TIME, human dermal telomerase immortalized microvascular ECs. 3

4 Figure S5. VEGF-B treatment induces Fatp1 and Fatp3 expression in human primary EC lines, and Fatp1, Fatp3 and Fatp4 expression in mouse EC lines. a, Cell line specific expression of FATPs in response to VEGF-B treatment. bend3 cells are shown in Figure 2. Data is presented as fold change versus svegfr1 treated cells set to 1 a.u. b, Immunoblot of FATP4 in bend3 ECs after VEGF-B 186 treatment. *P <0.05, **P <0.01, ***P <

5 Figure S6. VEGF-B does not induce Fatp4 expression in cultured NIH3T3 fibroblasts. a, Expression of VEGFR1, VEGFR2 and NRP1 in NIH3T3 cells. b, mrna levels of Fatp4 analyzed by qpcr. NIH3T3 cells were cultured using standard procedures, and treated as described in Supplemental Methods. Data is presented as fold change versus svegfr1 treated cells set to 1 a.u. Figure S7. VEGF-B regulation of FATPs is dependent on PI3K activity mrna expression of Fatp3 and Fatp4 in bend3 cells after co-treatment with VEGF- B and protein kinase inhibitors. Cells were treated and analyzed as described in Supplemental Methods. Data is presented as fold change versus svegfr1-treated cells set to 1 a.u. LY294002, PI3K inhibitor; PD98059, MEK1 inhibitor; U0126 MEK1 and MEK2 inhibitor. 5

6 Figure S8. mrna expression levels of Vegfb, Vegfa, Plgf and Pecam1 after adenoviral transduction a, Mouse and human Vegfb and Vegfa mrna expression levels were analyzed by qpcr using previously published species-specific primers 1. Data is presented as endogenous levels of mvegfb and mvegfa in AdhVEGF-B and AdhVEGF-A transduced mouse hearts, in relation to total Vegfb and Vegfa mrna levels by combining the results obtained with specific mouse and human primers. b, Mouse Plgf mrna expression was determined in AdLacZ and AdmPlGF transduced mouse hearts using primers specific for mplgf transcripts. Primer sequences are found in Table S3. L19 was used as normalization gene, and data is shown as fold change versus endogenous or AdLacZ levels set to 1 a.u. ±SEM. c, The induction of Fatp3 and Fatp4 mrna by AdVEGF-B transduction was not due to proliferation of ECs in mouse hearts, as the mrna levels of Pecam1 were unchanged. Gene expression was analyzed by qpcr as described in Supplemental Methods. Data is presented as fold change versus AdLacZ levels set to 1 a.u. 6

7 Figure S9. mrna expression of FATPs in wt and VEGF-B -/- tissues. a, Expression pattern of FATPs in wt mouse tissues, analyzed by conventional RT- PCR using specific primers listed in Table S3. b, mrna levels of FATPs in VEGF-B -/- mice were determined by qpcr in oxidative skeletal muscle and in ibat as described in Supplemental Methods. Data is shown as fold change versus wt levels set to 1 a.u. ibat, interscapular BAT. **P <0.01. Figure S10. Endothelial enrichment in isolated mouse EC-fractions from heart. Enrichment of isolated EC and non-ec fractions from heart analyzed by conventional RT-PCR. Pecam1 and Vegfr1 were used as marker genes for ECs, and Myh6, Tnnt2 and Vegfb for cardiomyocytes. Total cdna from heart was used as positive controls (Heart); NTC, non-template control. Myh6, cardiac myosin heavy polypeptide 6; Tnnt2, cardiac troponin T2 7

8 Figure S11. The vascular FATPs, Fatp3 and Fatp4, are downregulated in ECs isolated from BAT of VEGF-B -/- animals, but not in ECs isolated from lung. a, Relative expression of FATPs in isolated cell fractions from wt BAT (n=2-3). Data is shown as percentage of expression compared to L19 (n=2-3). b, mrna levels of FATPs in EC fractions isolated from wt and VEGF-B -/- BAT and lung, and in non-ec fractions isolated from wt and VEGF-B -/- BAT (n=3). Data is shown as fold change versus wt levels set to 1 a.u. Figure S12. FATP3 is expressed in the endothelium in oxidative skeletal muscle, whereas FATP4 is also expressed in the myocytes, especially in type I fibers. Immunohistochemical localization of FATP3, FATP4 and type I myosin in serial sections from wt soleus muscles. Asterisks indicate type I fibers. Scale bars, 50 µm (n=2-3 mice). 8

9 Figure S13. FATP3 and FATP4 can induce LCFA-uptake in cultured ECs. BODIPY-FA uptake into bend3 cells after transfection with the indicated plasmids ± a 10x molar excess of OA (lower panel). Quantification is shown in Figure 3. Figure S14. Silencing Fatp3 and Fatp4 expression by sirnas blunts both the transcriptional, and lipid-transporting effects of VEGF-B treatment. a-c, Relative mrna expression of Fatp3 (a), Fatp4 (b) and Gapdh (c) in bend3 cells that were transfected with either scrambled control sirna or the specific targeting sirnas, and treated with VEGF-B 186 or anti-vegf-b mab as described in 9

10 Supplemental Methods. Results are presented as percentage of expression relative to the normalization gene L19. d, Quantification of BODIPY-FA uptake into bend3 cells that were transfected with Fatp3 and/or Fatp4 targeting sirnas, and treated with VEGF-B 186 or anti-vegf-b mab. Representative images are shown in Figure 3. *P <0.05, **P <0.01, ***P < Figure S15. VEGF-B treatment does not increase endothelial permeability. a, Measurements of trans-endothelial resistance (TER) over membranes that were unseeded (No cells), or seeded with bend3 cells (Cells). Cells were subsequently treated with VEGFs, or mock treated, and TER was measured before treatment, as well as after 5 hrs and 30 hrs of treatment. b, Trans-endothelial leakage was assessed by adding the inert carbohydrate 14 C-inulin to bend3 monolayers after treatment with different VEGFs for 30 hrs. c, Trans-endothelial transport of LCFAs across bend3 cell layers. 14 C-labeled OA was added to cell culture inserts unseeded or seeded with bend3 cells. Transport was determined by measuring the radioactivity in the lower compartments in b-c. *P <0.05, **P <0.01, ***P <

11 Figure S16. Normal expression of fatty acid handling proteins in the small intestine of VEGF-B -/- animals. Expression of the fatty acid handling proteins Fatp4 and Cd36 in the small intestine of wt and VEGF-B -/- mice. Male wt and VEGF-B -/- mice were sacrificed and the proximal half of the small intestine was dissected and rinsed with PBS. RNA was isolated using RNeasy kit (Qiagen) as described in Supplemental Methods. The expression of Fatp4 and Cd36 was analyzed by qpcr and the results are presented as fold change versus the level in wt mice, which was set to 1 a.u. (n=3-6 mice). Figure S17. VEGF-B -/- mice have normal basal plasma levels of triglycerides (TGs) and non-esterified fatty acids (NEFAs). Blood was collected from hearts of Avertin-anesthetized, normally fed and agematched 8-12 week old male mice using EDTA syringes, and centrifuged for 1 min at 16,000 g. TG and NEFA content in plasma was determined using the Serum Triglyceride Determination Kit (Sigma), and the Wako NEFA C test kit (Wako Diagnostics), respectively, according to the manufactures instructions. 11

12 Figure S18. Accumulation of the 14 C-OA tracer in different WAT depots. Relative radioactivity in retroperitoneal (rwat), inguinal (iwat) and subcutaneous WAT (scwat) in VEGF-B -/- mice as compared to controls 24 hrs after oral gavage, preformed as described in Supplemental Methods (n=4 mice). ewat is shown in Figure 4. *P <0.05, **P <0.01. Figure S19. VEGF-B -/- mice accumulate lipids in WAT a, Weight of epididymal (ewat), retroperitoneal (rwat), inguinal (iwat) and subcutaneous WAT (scwat) fat pads from wt and VEGF-B -/- mice normalized to body weight (n=5-7). b, Quantification of total body fat from MRI scans performed as described in Supplemental Methods. *P <0.05, **P <

13 Figure S20. Quantitation of ORO staining in heart and muscle, and mean lipid vacuole size in BAT. Quantifications of ORO- and HE-staining of sections from wt and VEGF-B -/- mice (a), and from NRP1-EC-WT and NRP1-EC-KO mice (b). Representative images are shown in Figure 4. *P <0.05, **P <0.01, ***P < Figure S21. hvegf-b content in plasma 5 days after adenoviral administration. Wt and VEGF-B -/- mice were treated, and plasma was collected and analyzed by ELISA as described in Supplemental methods (n=4-5 mice). Recombinant hvegf-b protein was used for the standard curve. 13

14 Figure S22. Administration of AdVEGF-B rescues the phenotype of the VEGFB - /- mice but not the NRP-EC-KO mice. a, Cardiac Fatp4 mrna levels after adenoviral administration. Data is presented as fold change versus wt+adcmv or NRP1-EC-WT+AdCMV hearts set to 1 a.u. b, Quantification of ORO staining of heart sections from wt and VEGF-B -/- hearts (n=4-5) and from NRP1-EC-WT and NRP1-EC-KO hearts (n=3-4) after adenoviral administration. c, Representative ORO images from NRP1-EC-WT and NRP1-EC- KO mice after adenoviral delivery. *P <0.05, **P <

15 Figure S23. VEGF-B -/- mice have normal capacity for lipid β-oxidation, and VEGF-B per se does not induce oxidation of fatty acids in cultured myotubes. a, Ex vivo β-oxidation in isolated organs from wt and VEGF-B -/- mice. Dissected organs were weighted and incubated with 14 C-OA, after which formed 14 CO 2 was collected onto alkalinized membranes and analyzed by liquid scintillation (n=5). b, Level of oxidized 14 C-OA in cultured C2C12 myotubes in response to treatment with VEGF-B 186, globular adiponectin (gadiponectin, positive control) or mock treatment. Formed 14 CO 2 was collected onto alkalinized membranes and analyzed by liquid scintillation (n=3 wells per treatment). Both methods were essentially performed as previously described 2,3. Figure S24. Direct measurements of radioactivity in dissected hearts confirm the results from PET image analysis. Dissected mouse hearts from VEGF-B -/- mice show higher cardiac radioactivity, as compared to wt hearts, 60 min after [ 18 F]FDG i.v. injection. Data shows mean radioactivity as percentage of injected [ 18 F]FDG dose (n= 3-4 mice). *P <

16 Table S1. Vegfb is included in the mitochondrial co-expression cluster. The clustering of co-expressed genes (r>0.80) was preformed as described in Supplemental Methods. Vegfb is emphasized in bold. Sdhd succinate dehydrogenase complex, subunit D, integral membrane protein Atp5c1 ATP synthase, H+ transporting, mitochondrial F1 complex, gamma polypeptide 1 Atp5c1 ATP synthase, H+ transporting, mitochondrial F1 complex, gamma polypeptide 1 Ndufa8 NADH dehydrogenase (ubiquinone) 1 alpha subcomplex, 8 Ndufs1 NADH dehydrogenase (ubiquinone) Fe-S protein 1 Sdhc succinate dehydrogenase complex, subunit C, integral membrane protein Ndufab1 NADH dehydrogenase (ubiquinone) 1, alpha/beta subcomplex, Mus musculus cdna clone IMAGE: , partial cds Atp5o ATP synthase, H+ transporting, mitochondrial F1 complex, O subunit Ndufb10 NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 10 Ndufv3 NADH dehydrogenase (ubiquinone) flavoprotein 3 Ndufs3 NADH dehydrogenase (ubiquinone) Fe-S protein 3 Ogdh oxoglutarate dehydrogenase (lipoamide) Ndufa9 NADH dehydrogenase (ubiquinone) 1 alpha subcomplex, 9 Sdhb succinate dehydrogenase complex, subunit B, iron sulfur (Ip) Ndufa3 NADH dehydrogenase (ubiquinone) 1 alpha subcomplex, 3 Uqcr ubiquinol-cytochrome c reductase (6.4kD) subunit Ndufc1 NADH dehydrogenase (ubiquinone) 1, subcomplex unknown, 1 Ndufv1 NADH dehydrogenase (ubiquinone) flavoprotein 1 Ndufs2 NADH dehydrogenase (ubiquinone) Fe-S protein 2 Uqcrc1 ubiquinol-cytochrome c reductase core protein 1 Coq9 coenzyme Q9 homolog (yeast) Ndufb7 NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 7 Ndufb2 NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 2 Ndufb9 NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 9 Np15 nuclear protein 15.6 Ndufb2 NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 2 Atp5e ATP synthase, H+ transporting, mitochondrial F1 complex, epsilon subunit --- Mus musculus cdna clone IMAGE: , with apparent retained intron Atp5l ATP synthase, H+ transporting, mitochondrial F0 complex, subunit g Uqcrh ubiquinol-cytochrome c reductase hinge protein Cox5b cytochrome c oxidase, subunit Vb Ndufb8 NADH dehydrogenase (ubiquinone) 1 beta subcomplex 8 Ndufv2 NADH dehydrogenase (ubiquinone) flavoprotein 2 Ndufs5 NADH dehydrogenase (ubiquinone) Fe-S protein 5 Fh1 fumarate hydratase E04Rik RIKEN cdna E04 gene Ech1 enoyl coenzyme A hydratase 1, peroxisomal Etfb electron transferring flavoprotein, beta polypeptide Dci dodecenoyl-coenzyme A delta isomerase (3,2 trans-enoyl-coenyme A isomerase) Hadhb hydroxyacyl-coenzyme A dehydrogenase/3-ketoacyl-coenzyme A thiolase/enoyl- Coenzyme A hydratase (trifunctional protein), beta subunit Hsdl2 hydroxysteroid dehydrogenase like 2 Etfdh electron transferring flavoprotein, dehydrogenase Cpt2 carnitine palmitoyltransferase 2 Pdha1 pyruvate dehydrogenase E1 alpha 1 Mrps36 mitochondrial ribosomal protein S36 Cox7a1 cytochrome c oxidase, subunit VIIa 1 Idh3g isocitrate dehydrogenase 3 (NAD+), gamma Cs citrate synthase Chchd10 coiled-coil-helix-coiled-coil-helix domain containing 10 AI expressed sequence AI Vegfb vascular endothelial growth factor B Chchd3 coiled-coil-helix-coiled-coil-helix domain containing 3 Uqcrb ubiquinol-cytochrome c reductase binding protein Mcee methylmalonyl CoA epimerase Mor1 malate dehydrogenase, mitochondrial 16

17 Table S2. Transcriptional analyses of β-oxidation related genes revealed no major changes in VEGF-B -/- organs. mrna levels of genes in lipid metabolism determined by qpcr in tissues from VEGF-B -/- mice, normalized by the mean value of wt mice set to 1 unit (n=3 mice). *P <0.05, **P <0.01. ns= non significant. Cpt1/2, carnitine palmitoyltransferase, Slc25a20, solute carrier family 25 (mitochondrial carnitine/acylcarnitine translocase), member 20; Pgc1a, PPAR gamma coactivator 1α; Acsl, acyl-coa synthetase long-chain; Acadm, acyl-coenzyme A dehydrogenase medium chain; Acox1, acyl-coenzyme A oxidase; Ucp1/3, Uncoupling protein; Srebp, sterol regulatory element binding transcription factor; Acc, acetyl-coa carboxylase; Glut4, facilitated glucose transporter 4. Gene/Organ Heart Muscle ibat Cpt1a/b 0.57±0.07 * 0.61±0.15 * 0.72±0.04 * Cpt2 0.76±0.06 ns 0.75±0.05 * 0.75±0.11 ns Slc25a ±0.13 ns 0.71±0.12 * 0.63±0.07 ** Pgc1a 0.98±0.12 ns 0.92±0.07 ns 1.08±0.1 ns Ppara 1.13±0.08 ns 0.81±0.08 ns - Pparg ±0.12 ns Acsl 0.98±0.12 ns 0.95±0.15 ns 0.85±0.07 ns Mcad (Acadm) 1.01±0.08 ns 0.93±0.02 ns 1.14±0.12 ns Acox1 0.86±0.11 ns 0.78±0.09 ns 1.00±0.11 ns Ucp ±0.16 ns Ucp ±0.37 ns Srebp1c 1.27±0.06 ns 1.02±0.08 ns 2.13±0.20 ** Srebp2 1.44±0.03 * 1.21±0.05 * 1.28±0.09 * Acc 1.58±0.09 * 1.65±0.12 * 1.68±0.13 ** Glut4 1.79±0.12 ** 1.12±0.7 ns 1.42±0.07 ** 17

18 Table S3. Primer sequences. All sequences are written 5 3 Fwd primer Rev primer mvegfb TCTGAGCATGGAACTCATGG TCTGCATTCACATTGGCTGT mndufa5 ATCACCTTCGAGAAGCTGGA ACTTCACCACCCTGAAGCAA mcycs CCAAATCTCCACGGTCTGTT CCAGGTGATGCCTTTGTTCT mplgf CCCACACCCAGCTCACGTATTTA TCCCCTCTACATGCCTTCAATGC mvegfa CAGGCTGCTGTAACGATGAA TATGTGCTGGCTTTGGTGAT ml19 GGTGACCTGGATGAGAAGGA TTCAGCTTGTGGATGTGCTC mbeta actin ACTCTTCCAGCCTTCCTTC ATCTCCTTCTGCATCCTGTC mb2m CTGACCGGCCTGTATGCTAT CCGTTCTTCAGCATTTGGAT mcpt1a CATGTCAAGCCAGAGGAAGA TGGTAGGAGAGCAGCACCTT mcpt1b CCCATGTGCTCCTACCAGAT CCTTGAAGAAGCGACCTTTG mcpt2 CAGCATATGATGGCTGAGTG GTGGTTTATCCGCTGGTATG mslc25a20 TTTGCAGGGATCTTCAACTG CCCTTTGTACAAGGAGGTGA mpgc1a GGAGCCGTGACCACTGACA TGGTTTGCTGCATGGTTCTG mppara GACAAGGCCTCAGGGTACCA GCCGAATAGTTCGCCGAAA mpparg CCATTCTGGCCCACCAAC AATGCGAGTGGTCTTCCATCA macsl ATATCTACCTGCGGAGTGAAG CCTTCCCAAGTTTCAACAAGTC macadm TCGGAGGCTATGGATTCAAC CAGCCTCTGAATTTGTGCAG macox1 ACTTGTTTGAGTGGGCCAAG AGAGATTCGGCCTCTCTGTG mhsd17b4 CGGTTTTGAGAAGCCCATATT ATTCTGTTTCCTTCCTTCCACA mhmgcs1 CAGTACTCACCTCAGCAGTTG CACACAAGTTCTCGAGTCAAG mucp1 CTGCCAGGACAGTACCCAAG GCCACAAACCCTTTGAAAAA mucp2 GCCACAAACCCTTTGAAAAA TACAAGGGGTTCATGCCTTC mucp3 AGCCCTCTGCACTGTATGCT CAGAAAGGAGGGCACAAATC msrebp1c* GGCACTAAGTGCCCTCAACCT GCCACATAGATCTCTGCCAGTGT msrebp2** GGATCATCCAGCAGCCTTTGA ACCGGGACCTGCACCTGT macc* CCCAGCAGAATAAAGCTACTTTGG TCCTTTTGTGCAACTAGGAACGT mglut4 ACTCTTGCCACACAGGCTCT CCTTGCCCTGTCAGGTATGT mfatp1 TCAATGTACCAGGAATTACAGAAGG GAGTGAGAAGTCGCCTGCAC mfatp2 CATGAGCTAAACCACCAGGG TTCCTGAGGATACAAGATACCATTG mfatp3 CGCAGGCTCTGAACCTGG TCGAAGGTCTCCAGACAGGAG or CCTCGGTTTCTCAGGCTCCA CTGTACCGGGCAGGTGTGA mfatp4 GCAAGTCCCATCAGCAACTG GGGGGAAATCACAGCTTCTC mfatp5 GTGGTCAGAGATTCCAGGTTCC GCTATACCAGCATGTCCGCTC mfatp6 TACAACCAAGTGGTGACATCTCTG AATCTCTTCGGTCAATGGGAC mcd36 GATGAGCATAGGACATACTTAGATGTG CACCACTCCAATCCCAAGTAAG mfabp1 CCAGAAAGGGAAGGACATCA GTCTCCAGTTCGCACTCCTC mfabp3 TTCAGCTGGGAATAGAGTTCG CTGCACATGGATGAGTTTGC mfabp4 GATGGTGACAAGCTGGTGGT AATTTCCATCCAGGCCTCTT mfabp5 GGAAGATGGCGCCTGGTGGA CCGAGTACAGGTGACATTGT mpecam1 AGAGACGGTCTTGTCGCAGT TACTGGGCTTCGAGAGCATT mvegfr1 GGAGGAGTACAACACCACGG TTGAGGAGCTTTCACCGAAC mmyh6 AGCTGGAGAATGAGCTGGAG GCAGCCGCATTAAGTTCTTC mtnnt2 CGTGAGGAGGAGGAGAACAG TCCTCTCTGCCAGGATCTTC mvegfr2 AGCACCTCTCTCGTGATTTCC AGTAAAAGCAGGGAGTCTGTGG mnrp1 GGAGCTACTGGGCTGTGAAG CCTCCTGTGAGCTGGAAGTC CoxII*** TCTCCCCTCTCTACGCATTCTA ACGGATTGGAAGTTCTATTGGC Gapdh TGCGACTTCAACAGCAACTC GCCTCTCTTGCTCAGTGTCC hl19 GGCACATGGGCATAGGTAAG CCATGAGAATCCGCTTGTTT hvegfr1 CTTCACCTGGACTGACAGCAA CTCAGCGTGGTCGTAGGT hnrp1 CGCTACCAGAAGCCAGAGGA CATCCACAGCAATCCCACCAA hfatp-1 TCTATGGGGTGGCTGTTCCA TCAAACCCTCTCGCTGCA hfatp-2 AGGGGAAAATGTGGCCACCA TTAGAAACCGGGGCCTTGCA hfatp-3 CAGAGGTCTTCGAGGCCCTA AAGGTCTCTGTGGTGGCCAA hfatp-4 CTTTTCCAGCCGCTTCCACA TGGCTGGCAGGGAATGCA hfatp-5 TGTGCGTGCCAGGTTGTGA ATCCCCACATTGAAGCCCTCA hfatp-6 GCTGGGCCTTATAAGCACACA CAACCTCAGTGGTTGCGACA The following primer sequneces were taken from these publications: * Primer sequence copied from reference 4 ** Primer sequence copied from reference 5 *** Primer sequence copied from reference

19 References 1. Thijssen, V.L., Brandwijk, R.J., Dings, R.P. & Griffioen, A.W. Angiogenesis gene expression profiling in xenograft models to study cellular interactions. Exp Cell Res 299, (2004). 2. Fruebis, J., et al. Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci U S A 98, (2001). 3. Mao, X., et al. APPL1 binds to adiponectin receptors and mediates adiponectin signalling and function. Nature cell biology 8, (2006). 4. Jiang, T., et al. Diet-induced obesity in C57BL/6J mice causes increased renal lipid accumulation and glomerulosclerosis via a sterol regulatory elementbinding protein-1c-dependent pathway. J Biol Chem 280, (2005). 5. Zhou, R.H., et al. Vascular endothelial growth factor activation of sterol regulatory element binding protein: a potential role in angiogenesis. Circ Res 95, (2004). 6. Villena, J.A., et al. Orphan nuclear receptor estrogen-related receptor alpha is essential for adaptive thermogenesis. Proc Natl Acad Sci U S A 104, (2007). 19