Supplementary Method

Protein fragments and substances. Nesfatin-2/3 and nesfatin-3 were synthesized by an in vitro cell-free translation system (Shimizu, Y., et al. Nature Biotech.19, 751-755, 2001) using a plasmid containing the full-length of rat NUCB2 constructed in pRcCMV as a template and PCR-based primers; nesfatin-2/3 encoding 942 bp, forward primer, 5’ATGCAAGAAGTAGGAAGACTGAGAA-3’ and reversed primer, 5’-TTATGTGTGTGGCTCAAACTTCA-3’; nesfatin-3 encoding 699 bp, forward primer, 5’-ATGGAATATTTAAAAACGCTGA-GTGAG-3’ and reversed primer, the same as described above. The recombinant proteins were purified, and their amino acid sequences were confirmed by Mass spectrum analysis. Rat leptin was obtained from R & D Systems Inc., Minnesota, and alpha-MSH,rat cocaine- and amphetamine-regulated transcript (CART)-55-102, NPY, melanin-concentrating hormone (MCH) and orexin-A were all purchased from Peptide Institute Inc.Osaka, Japan. SHU9119 was purchased from Phoenix Pharmaceuticals, Inc., California. Troglitazone was obtained from Sankyo Pharmaceutical Co., Tokyo, Japan.

Immunohistochemical analysis: Free-floating sections (60μm each) from brains ofperfusion-fixed rats were incubated in the Ab (1:5,000 dilution) followed by incubation with biotin-conjugated goat anti-rabbit IgG, as described elsewhere (Maruyama, M., et al.Endocrinology140, 2326-33, 1999). Abs for PCs were obtained from Affinity BioReagents, Colorado.

In situ hybridization:In situ hybridization analysis was performed on cryostat sections of rat brains using a35S-labeled or digoxigenin (DIG)-labeled antisense cRNA probe against rat NUCB2 cDNA (bp 402-967) or rat POMC cDNA (bp 194-503) as described elsewhere (Imaki, T., et al. J. Neurochem. 15, 916-921, 2003). The DIG labeled RNA was detected using alkaline phosphatase-conjugated anti-DIG IgG. The density in hypothalamic nuclei was semi-quantified using an image analyzer.

Punched-out samples from hypothalamic nuclei. Punched-out samples from bilateral hypothalamic nuclei of 2 rats were obtained as described elsewhere (Shimizu, H., et al. Neurosci. Lett.204, 81-84, 1996). Pooled samples were extracted in 0.1 N HCl for assay by nesfatin-1 ELISA, and were extracted in Isogen(Nippongene Co.) for estimation of mRNA level by real-time RT-PCR.
Supplementary Figures

Figure S 1. Northern blot of cell lysatesusing a 32P-labelled NUCB2cDNA probe (bp 402 - 967). Thirty μg of each RNA obtained from samples was loaded. The samples on a gel were stained with ethidium bromide.

Figure S 2. Effects of troglitazone (TGZ) on body weight in rats. Rats were fed a food admixture containing 0.2% TGZ for 10 days. Body weights (a) were determined, blood levels of leptin (b) were measured by a rat leptin ELISA kit(Yanaihara Institute, Inc. ShizuokaJapan), and hypothalamic NUCB2 concentrations (c) were measured by Western blot. Each group contains 6 rats. Data are expressed as the mean±s.e.m. *, p<0.05 vs. the control TGZ (-) group.

Figure S 3. Western blot of rat hypothalamic extracts using an affinity-purified NUCB2 Ab-L(1:1,600 dilution). The antibody was affinity-purified using a Sepharose column with covalently bound peptide antigen. Previously, the antibody was incubated in the presence or absence of 5μg each peptide. An arrow indicates a 47.5 kDa band corresponding to the whole NUCB2.

Figure S 4.Food intake in rats after icv injection of an anti-NUCB2 Ab-L. Rats were given icv injection of control IgG or NUCB2 Ab-L-IgG at a dose of 8μg. Data are expressed as the mean±s.e.m. Each group contains 5 - 6 rats. *, p<0.01 vs. the control IgG-injection group.

Figure S 5. Putative processing sites of the rat NUCB2/nesfatin and Western blot using affinity-purified nesfatin Ab24 and Ab301. The antibodies recognize synthetic nesfatin-1 (9.7 kDa) and nesfatin-3 (27.9 kDa), respectively.

Figure S 6. Elution profiles of rat hypothalamic extracts (a) and cerebrospinal fluid extracts (b) on HPLC. Eight rat hypothalamic extracts by homogenization in 0.1% trifluoroacetic acid(TFA) and incubation at 650C for 15 min and 600μl cerebrospinal fluid extracts were each applied on a Millipore Sep-Pak C18 cartridge and extensively washed with 0.1% TFA followed by elution with 60% acetonitrile. The obtained samples were analyzed on reverse-phased C18 HPLC (4.6x250 mm,GL Sciences Inc. Japan; elution speed=1.0 ml/min, acetonitrile gradient=0 -60% in 0.1% TFA). All fractions of each sample on HPLC weremonitored by acompetitive ELISA for nesfatin-1 using biotin-labeled nesfatin-1 peptide and nesfatin Ab24-derived IgG. The assay detected as little as 4.6 ng nesfatin-1 per tube, and the coefficient of intra-assay variation was 2.8%. In this assay, nesfatin-1 had less than 0.01% cross-reactivity against leptin, alpha-MSH, CART, NPY, MCH and orexin-A.

Figure S 7. Western blot of fractionated samples of cerebrospinal fluid extracts on HPLC using an affinity-purified nesfatin Ab24. An Arrow indicates a 9.7 kDa band corresponding to nesfatrin-1 peptide.

Figure S 8. Immunohistochemical analysisof the rat hypothalamic Arc and PVN using a nesfatin Ab24. In the absorption study, the antibody was previously incubated with5 μgeach ofindicated peptides. The bottom figure shows Western blot of the rat hypothalamic extracts using an affinity-purified nesfatin Ab24.

Figure S 9. Changes in the fat weights after continuous icv injection of nesfatin-1. Rats received chronic icv injection of vehicle or nesfatin-1 (daily 5 pmol) for 10 days. (a), Subcutaneous fat; (b), Epididymal fat; (c), Mesenteric fat; (d), Retroperitoneal fat; (e), Brown adipose tissue; (f), Gastrocnemius. *, p<0.01 vs. vehicle injection. Each group contains 4〜5 rats. Data are expressed as the mean±s.e.m. *, p<0.01 vs. the vehicle-injection group.

Figure S 10. Hypothalamic NUCB2 concentrations (estimated by Western blot) 10 days after chronic icv injection of missense and antisense MON (daily 40 μg). Each group contains 4〜5 rats. Data are expressed as the mean±s.e.m. *, p<0.01 vs. the missense MON-injection group.

Figure S 11. Effects of alpha-MSH on NUCB2 gene expression in the hypothalamic PVN. Rats were given icv injection of 0.9 nmol alpha-MSH, andfour hrs latter,the samples were obtained. The mRNA expression in the hypothalami was estimated using a digoxigenin-labeled NUCB2 antisense cRNA probe. (a), representative figures in vehicle injection; (b), representative figures in alpha-MSH injection; (c), NUCB2 mRNA expression in the PVN with the mean±s.e.m. Each group contains four rats. *, p<0.01 vs. the vehicle-injection group.

Figure S 12. Effects of nesfatin-1 on expression of POMC, AgRP, NPY and CRH genes in the hypothalamic Arc and PVN. Rats were given icv injection of 25 pmol nesfatn-1, and one and four hrs latter, the samples were obtained. The upper panels show representative figures (a-d) of insitu hybridization analysis of the hypothalami four hrs after injection, using a digoxigenin-labeled POMC antisense cRNA probe. Each group contains 3 rats, and data are expressed as the mean±s.e.m. The lower panels show mRNA levels (e-h) measured by real-time RT-PCR in the hypothalami one hr after injection. Each group contains 6 rats, and data are expressed as a ratio of each gene expression against GAPDHexpression with the mean±s.e.m. TaqMan probes for real-time RT-PCR were all purchased from Applied Biosystems, California; POMC, Rn00595020-m1; AgRP, Rn01431703-q1; NPY, Rn00561681-m1; CRH, Rn01462137-m1; GAPDH, Rn99999916-s1.

Figure S 13. Nesfatin-1-induced cAMP formation and calcium influx in cells expressing mouse MC3R or MC4R. Hela cells (7x105 cells/well) were transfected with the pET1/MycHis Aexpression vector (4 μg/well) constructed by ligation of genes encoding mouse MC3R or MC4R cDNA, and cells were incubated in serum-free medium. After administration of several doses of MT-II, a MC3/4R agonist and mouse nesfatin-1, cAMP formation for 15 min (a) was measured by cAMP-screen system (Applied Biosystems, T1500)and calcium influx (b) was sequentially monitored by FLIPR calcium assay kit (Molecular Devices, R8041). Data are representative results of duplicate experiments.

1