Fig 2: AHCY overexpression in different cell types.Gene expression levels of (A) adenosylhomocysteinase (AHCY) and (B) itchy E3 ubiquitin protein ligase (ITCH) in stromal vascular fraction (SVF) cells before and after adipocyte differentiation and compared to that of (C) agouti signaling protein (ASIP) in differentiated SVF cells are shown from the patient and controls (Ctr) patients 1-4 (n=4; 3 females, 1 male) matched for age (mean age 13.8 years) and body weight state (all presenting with overweight or obesity). Data points of n=3 independent experiments are indicated for each subject. In peripheral blood leukocytes (PBL) AHCY (D) and ITCH (E) expression are shown from the patient and controls. Data points are indicated for each subject. (F) Relative AHCY expression in Simpson–Golabi-Behmel syndrome (SGBS) cells after transfection with pCMV-Entry control vector (Ctr) or pCMV-AHCY (AHCY) in n=3 independent experiments. Transfected SGBS cells were differentiated into adipocytes and the percentage of differentiated cells was determined by cell counting after Hoechst and Nile red staining. Using the same cells adipogenic capacity was compared by Oil red O staining (representative microscopic pictures from one of three independent experiments; scale bar = 500µM) and gene expression of the adipogenic marker gene peroxisome proliferator-activated receptor gamma (PPARG) after differentiation in n=2 independent experiments was determined. Bar plots show the mean ± SEM. (G) AHCY and ITCH expression levels are shown for SVF cell-derived induced pluripotent stem cells (iPSCs) from three clones of the patient and two clones from each of the control (Ctr) children, 1 and 10. Data points for each clone from each subject are indicated. The bar plots A-E and G present the mean values ± SEM as a fold change of the controls. Gene expression data were normalized to the expression of ß-actin and TATA-box-binding protein. All bars from controls are shown in blue and from the patient in red. P values derived using two-sided Student’s t-test are given. Source data

Fig 3: Clinical characteristics of the index patient and her family.a,b, Course of BMI (a) and height (b) of the patient (red dots) from birth to adolescence in relation to 3rd (P3), 50th (P50) and 97th (P97) sex-specific reference percentiles according to Kromeyer-Hauschild et al.40. Data from the patient were corrected for gestational age until the age of 2 years. c–g, Parameters of glucose metabolism with the respective reference ranges of either the local hospital laboratory depicted as gray areas or with prediabetes cut-offs according to the American Diabetes Association65 indicated by a dotted line. h, Serum levels of liver enzyme alanine aminotransferase (ALAT) in comparison to the reference values according to Bussler et al. 66 (shaded in gray). Blue arrows mark the time point of bariatric surgery at the age of 12.4 years. i, Pedigree of the family; females are indicated by circles, the father by a square. Individuals carrying the ASIP tandem duplication are indicated by black symbols, the open symbol indicates the mother not carrying the mutation. The index patient is indicated by a red arrow. Photographs present patient at age of 3.5 years and father at age 52 years. HbA1c, glycated hemoglobin; HOMA-IR, homeostatic model assessment for insulin resistance; 2 h glucose, glucose levels after 2 h of oral glucose tolerance testing.Source data

Fig 4: Graphical summary.Starting out from a case of a patient with severe early-onset obesity, accelerated linear growth, red hair and hyperinsulinemia (gray shaded box), we found enhanced adipogenesis of her SVF adipose-derived cells. Transcriptomic profiling identified aberrant ASIP expression, which is caused by chromosomal rearrangement with tandem duplication placing ASIP under the control of the ITCH promoter found in genetic profiling. Functionally verified aberrant promoter usage subsequently leads to ectopic ASIP expression as verified in native and patient-derived iPSCs of the index patient. ASIP antagonizes MC4R signaling and thereby ubiquitous ectopic ASIP expression can induce a phenotype phenocopying the agouti mouse models. Cohort screening identified four additional patients carrying the same ASIP mutation, whereas none was identified in a populational cohort. Carriers of the ASIP mutation are shown as red-shaded symbols, controls with obesity as white-shaded symbols and healthy children as gray symbols.

Fig 5: Identification of ectopic ASIP expression in the patient.a, Differentially expressed genes between undifferentiated SVF cells from the patient and control children (Ctr; n = 4) presented as the log2 fold change of gene expression across -log10 false discovery rate (FDR). Similar results have been shown in a second independent experiment. b–d, ASIP gene expression levels are shown for adipocytes (b), SVF cells (c) and peripheral blood leukocytes (PBLs) (d). As PBL RNA samples from control children 1–4 were not available, measurements were performed in control children 5–8 (Supplementary Appendix and Supplementary Table 1). e, ASIP protein was detected by immunoblotting in cell lysates and conditioned medium from untreated and brefeldin A and monensin (B/M)-treated SVF cells from the patient and two control children, 1 and 3. Similar results were obtained by a second independent experiment. Gene expression data shown in b–d were normalized to the expression of ß-actin and TATA-box-binding protein. Bar plots represent mean ± s.e.m. For b–d the fold changes in relation to mean of controls ± s.e.m. are indicated. All bar plots from controls are shown in blue and from the patient in red.Source data