Mutation Screen of the SIM1 Gene in Pediatric Patients with Early-Onset Obesity

ORIGINAL ARTICLE Mutation screen of the SIM1 gene in pediatric patients with early-onset obesity

D Zegers1, S Beckers1, R Hendrickx1, JK Van Camp1, V de Craemer1, A Verrijken2, K Van Hoorenbeeck3, SL Verhulst3, RP Rooman3, KN Desager3, G Massa4, LF Van Gaal2 and W Van Hul1

BACKGROUND: The transcription factor SIM1 (Single-minded 1) is involved in the control of food intake and in the pathogenesis of obesity. In mice, Sim1 is involved in the development of the paraventricular nucleus, and Sim1 deficiency leads to severe obesity and hyperphagia. In humans, chromosomal abnormalities in the SIM1 gene region have been reported in obese individuals. Furthermore, recent data also suggest that loss-of-function point mutations in SIM1 are responsible for SIM1 haplo-insufficiency that is involved in causing human obesity. In this study, we therefore wanted to expand the evidence regarding the involvement of SIM1 mutations in the pathogenesis of severe early-onset obesity. METHODS: We screened 561 severely overweight and obese children and adolescents and 453 lean adults for mutations in the coding region of the SIM1 gene. Mutation screening in all patients and lean individuals was performed by high-resolution melting curve analysis combined with direct sequencing. To evaluate the effect of the mutations on SIM1 transcriptional activity, luciferase reporter assays were performed. RESULTS: Mutation analysis identified four novel nonsynonymous coding variants in SIM1 in four unrelated obese individuals: p.L242V, p.T481K, p.A517V and p.D590E. Five synonymous variants, p.P57P, p.F93F, p.I183I, p.V208V and p.T653T, were also identified. Screening of the lean control population revealed the occurrence of four other rare SIM1 variants: p.G408R, p.R471P, p.S492P and p.S622F. For variants p.T481K and p.A517V, which were found in obese individuals, a decrease in SIM1 transcriptional activity was observed, whereas the transcriptional activity of all variants found in lean individuals resembled wild type. CONCLUSIONS: In this study, we have demonstrated the presence of rare SIM1 variants in both an obese pediatric population and a population of lean adult controls. Further, we have shown that functional in vitro analysis of SIM1 variants may help in distinguishing benign variants of no pathogenic significance from variants which contribute to the obesity phenotype.

International Journal of Obesity (2014) 38, 1000–1004; doi:10.1038/ijo.2013.188 Keywords: mutation analysis; SIM1; melanocortin signaling

INTRODUCTION hypothalamic oxytocin and PVN melanocortin 4 receptor (Mc4r) SIM1 (Single-minded 1) is a member of the bHLH-PAS (basic helix- mRNA, suggesting that the hyperphagic obesity in Sim1-deficient loop-helix Per-Arnt-Sim) family of transcription factors. In mice, mice may be attributable to changes in the leptin-melanocortin- 6 Sim1 is essential for the formation of the paraventricular nucleus oxytocin pathway. The hypothesis that SIM1 might be a factor (PVN) in the hypothalamus,1 a region that is known to be within the famous leptin-melanocortin signaling pathway is implicated in the regulation of body weight, as neurons in the PVN further supported by the investigation of transgenic mice express murine Mc4r and are a target of a-MSH.2,3 Homozygous overexpressing human SIM1. Overexpression of SIM1 completely Sim1-knockout mice (Sim1 À / À ) lack a PVN and die perinatally. rescued the hyperphagia and partially rescued the obesity In contrast, heterozygous Sim1 þ / À mice are viable, but they show exhibited by agouti yellow mice, in which melanocortin early-onset obesity, increased linear growth and are hyperphagic, signaling is interrupted, suggesting that Sim1 acts downstream making their phenotype resemble mc4r-mutant mice. Partial Sim1 of Mc4r to control food intake.7 deficiency results in hypodevelopment of the PVN that contains SIM1 deficiency in humans has also been reported: a girl with on average 24% fewer cells.4 As lesions in the PVN also induce early-onset severe obesity and a de novo translocation between increased appetite,5 it was suggested that this hypocellularity chromosomes 1p22.1 and 6q16.2 has been described by Holder causes the hyperphagia and obesity of the Sim þ / À mice.4 More et al.8 In this case, the translocation separates the 50 promoter recently, Tolson et al.6 demonstrated that the role of Sim1 in region and bHLH domain from the rest of the gene. It was feeding regulation is not limited to formation of the PVN. They hypothesized that haplo-insufficiency of SIM1 was responsible for conditionally deleted Sim1 in mice postnatally, thereby generating the severe obesity in the subject.8 Further, in multiple obese conditional Sim1 heterozygous mice and conditional Sim1 patients with a Prader-Willi-like phenotype, deletions of the homozygote mice and revealing a dosage-dependent effect of 6q16.2-6q16.3 region have been reported. These deletions on 6q Sim1 on obesity. These mice exhibited a remarkable decrease in disrupt or delete the SIM1 gene, possibly giving rise to obesity.9–11

1Department of Medical Genetics, University of Antwerp and Antwerp University Hospital, Antwerp, Belgium; 2Department of Endocrinology, Diabetology and Metabolic Diseases, Antwerp University Hospital, Antwerp, Belgium; 3Department of Paediatrics, Antwerp University Hospital, Antwerp, Belgium and 4Department of Paediatrics, Jessa Hospital, Hasselt, Belgium. Correspondence: Professor Dr W Van Hul, Department of Medical Genetics, University of Antwerp, Universiteitsplein 1, Antwerp B-2610, Belgium. E-mail: [email protected] Received 11 April 2013; revised 12 September 2013; accepted 24 September 2013; accepted article preview online 7 October 2013; advance online publication, 29 October 2013 Involvement of SIM1 mutations in childhood obesity D Zegers et al 1001 More recent data also suggest that loss-of-function point standard PCR protocol. Real-time PCR was performed as previously mutations in SIM1 are responsible for SIM1 haplo-insufficiency described.19 After amplification, samples were first denaturated at 95 1C that is involved in causing human obesity12–14 and Prader- and subsequently renaturated at 40 1C and then heated during Willi-like syndrome.12,13 Because of the alleged role of SIM1 fluorescence acquisition between 70 and 95 1C. When melting curves downstream of the leptin-melanocortin pathway and the reports deviated from WT, direct sequencing using an ABI Prism 3130xl Genetic Analyzer with ABI BigDye Terminator v1.1 Cycle Sequencing kits (Applied on SIM1 haplo-insufficiency in obese humans, we hypothesized Biosystems, Foster City, CA, USA) was performed in order to characterize that heterozygous point mutations in the SIM1 gene might be the putative sequence variations. Primer sequences for high-resolution melting cause of severe early-onset obesity, and we set out to screen the curve analysis and direct sequencing are available on request. SIM1 gene region in an extensive cohort of obese children. Constructs MATERIALS AND METHODS Wild-type (WT) human SIM1 subcloned into a pDR2 vector was kindly provided by Professor Chrast (University of Lausanne, Lausanne, Switzer- Study population land). The entire coding region of the hSIM1 gene was amplified using Mutation screening of the SIM1 gene was performed in a cohort of 561 specific primers introducing a HindIII and an EcoRI restriction site 50 and 30 severely overweight and obese children and adolescents and 453 healthy, of the SIM1 sequence, respectively. The PCR product was subsequently lean adults. Children and adolescents (243 boys, 318 girls) were recruited cloned into a pcDNA3 expression vector (Invitrogen, Carlsbad, CA, USA). from the pediatrics departments of the Antwerp University Hospital and Mutations were introduced into WT SIM1, subcloned into the pcDNA3 the Jessa Hospital, Hasselt, both in Belgium. This group (age range 0–12) expression vector, using the QuikChange II Site-Directed Mutagenesis Kit included 257 children aged 8.7±2.5 years (mean±s.d.) and 304 (Stratagene, La Jolla, CA, USA). The WT sequence and the presence of the adolescents aged 15.2±2.3 years, (age range 12–21; summary of all mutations were checked and confirmed by direct sequencing using an ABI characteristics: Table 1). The cohort of healthy lean adults (18.5 kg m À 2 3130xl Genetic Analyzer (Applied Biosystems) according to the protocol Xbody mass index (BMI)X24.9 kg m À 2) was recruited among employees described above to ensure that no errors had been induced. Luciferase from the university and the university hospital and among couples seeking reporter plasmid pGL3-HRE-Luc was kindly provided by professor Whitelaw prenatal counseling at the Department of Medical Genetics. This control (University of Adelaide, Adelaide, Australia). Wild-type human ARNT2 cDNA group included 160 men and 293 women, aged 35.2±7.2 years, with a cloned in a pCMV6-XL4 expression vector was obtained from Origene mean BMI of 22.1±1.7 kg m À 2. The study was approved by the local ethics Technologies (Rockville, MD, USA). committee, and all participants gave their written informed consent once the aim and design of the study had been explained. Reporter assay SIM1 For monitoring SIM1 transcriptional activity in cells transiently expressing Anthropometry WT or mutant SIM1, a luciferase assay was performed. HEK293 cells were Height was measured to the nearest 0.5 cm, and body weight was maintained in DMEM supplemented with 10% fetal bovine serum, measured on a digital scale to the nearest 0.1 kg. BMI was calculated for all 100 units ml À 1 penicillin and 100 mgmlÀ 1 streptomycin (Invitrogen). Cells patients and BMI-for-age percentile (490th percentile for all patients were incubated at 37 1C in humidified air containing 5% CO2. Cells were included) and Z-scores were based on the ‘Flemish Growth Charts 2004’.15 plated in 24-well plates at a density of 50 000 cells per well. Eight hours Using the ‘Flemish Growth Charts 2004’, the cutoff values for overweight later, cells were transiently transfected with 40 ng ARNT2 in pCMV6-XL4 and obesity are defined as the percentile line on the chart that crosses BMI expression vector, 120 ng pGL3-HRE-Luc, 10 ng pRL-TK and 40 ng WT or 25 kg m À 2 and BMI 30 kg m À 2, respectively, at 18 years of age. mutant SIM1 in pcDNA3 expression vector per well. Fourty-eight hours post-transfection, cells were then assayed for luciferase activity, using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) on a Mutation analysis Glomax-multi Detection System (Promega). Firefly luciferase readings, Blood samples from all patients and control subjects were obtained for measured as a reference for SIM1 transcriptional activity, were normalized extraction of genomic DNA by standard procedures.16 Mutation screening to the readings of the pRL-SV40 Renilla luciferase control reporter. Relative of the entire coding region of the SIM1 gene (GenBank accession number ratios of Firefly luciferase readings over Renilla luciferase readings were NM_005068.2), including intron-exon boundaries was performed by high- normalized to the maximal relative ratio calculated for WT SIM1 (%). Each resolution melting curve analysis in combination with direct sequencing. transfection was performed in duplicate, and the experiment was repeated High-resolution melting curve analysis (HRM), which is performed using three times with similar results. Data are expressed as mean values±s.d. P- the LightCycler 480 Real-Time PCR System (Roche Applied Science, values were obtained by performing Wilcoxon signed-rank test (IBM SPSS Mannheim, Germany), is a highly sensitive technique to detect novel Statistics, version 20). mutations in a large sample set.17,18 For high-resolution melting curve analysis, the coding region of SIM1 (exon 1–exon 11, Figure 1) was divided into 14 amplicons, which were amplified using either a touch-down or a RESULTS Mutation analysis Screening of the coding region of the SIM1 gene in our population Table 1. Characteristics of the study populations of obese children and adolescents led to the identification of multiple sequence variants. We detected five different synon- Parameter Obese Obese Lean adults ymous variants, c.171A4T (P57P), c.279C4T (F93F; rs145361258), children adolescents c.549C4T (I183I), c.624G4A (V208V) and c.1959C4T (T653T; N 257 304 453 rs41318041), that do not alter the protein sequence and two Age (years) 8.7±2.5 15.2±2.3 35.2±7.2 common nonsynonymous SNPs P352T (rs3734354) and A371V Male/female 120/137 123/181 160/293 (rs3734355). Both variants, which are in complete linkage Weight (kg) 52.5±17.6 95.1±22.5 65.2±8.7 disequilibrium, were observed in the population of obese children Height (m) 1.37±0.21 1.65±0.20 1.71±0.10 and the population of lean controls with a frequency of 23% and À 2 BMI (kg m ) 27.5±4.3 34.3±5.1 22.1±1.7 25.4% respectively (Pearson Chi-Square; P ¼ 0.376). Further, we BMI Z-score 2.64±0.64 2.51±0.41 NA also found four nonsynonymous coding variants in SIM1 BMI-for-age percentile 490th 490th NA (c.724C4G, c.1442C4A, c.1550C4T and c.1770C4G) leading to percentile percentile the amino-acid substitutions p.L242V, p.T481K, p.A517V and Abbreviation: BMI, body mass index; NA, not applicable. Mean value±s.d. p.D590E, respectively. When we subsequently screened our lean is given, except for the BMI-for-age percentile, which exceeds the 90th control population, we found four of the synonymous variants also percentile for all children and adolescents included. BMI Z-scores were 15 observed in the obese population. In addition, we found four calculated using the ‘Flemish Growth Charts 2004’. nonsynonymous coding variants in the SIM1 gene (c.1222G4C,

& 2014 Macmillan Publishers Limited International Journal of Obesity (2014) 1000 – 1004 Involvement of SIM1 mutations in childhood obesity D Zegers et al 1002

Figure 1. Graphical representation of SIM1. The SIM1 protein is a transcription factor containing a basic helix-loop-helix motif (bHLH), two PAS (per-arnt-sim) domains, a PAC domain (PAS associated C-terminal domain) and a C-terminal domain (C-term), which includes the nuclear localization signal (NLS). The relative position of the variants found in this study is indicated on the protein structure of SIM1: variants found in obese patients are indicated with gray arrowheads (m), variants found in lean controls are indicated with ﬁne black arrows (k). Amino acid numbering (1–766) is indicated below the domain structure for easy reference.

c.1412G4C, c.1474T4C and c.1865C4T) leading to the amino- Table 2. Summary of mutations identified in the coding region of the acid substitutions p.G408R, p.R471P, p.S492P and p.S622F. An SIM1 gene overview of the different coding variants found in our cohorts of obese children and lean control adults is given in Table 2. Nucleotide Amino-acid Domain Obese cases Lean controls We performed in silico analysis using a set of five different substitution substitution (N ¼ 561) (N ¼ 453) prediction programs (PolyPhen,20 PolyPhen-2,21 SIFT,22 SNPs&Go23 and MutPred24) for all SIM1 variants identified in the present study. c.171A4T P57P(*) /1 1 For most of the eight reported nonsynonymous variants, there is c.279C4T F93F PAS1 2 1 c.549C4T I183I(*) /1 0 large variation in prediction between the different programs used. (*) However, the R471P variant, which was found in a lean control c.624G4A V208V /1 1 c.724C4G L242V PAS2 1 0 individual, was predicted to be benign by all programs. As this (*) 25 c.1442C4A T481K C-term 1 0 variant has also been identified in the 1000 genomes project (*) 26 c.1550C4T A517V C-term 1 0 and the NHLBI GO Exome Sequencing Project (ESP), it is c.1770C4G D590E C-term 1 0 expected to be a rare variant, which is not associated with obesity. c.1222G4C G408R(*) C-term 0 2 Whereas none of the other variants have been reported in the c.1412G4C R471P C-term 0 1 1000 genomes project, both D590E and S622F have been reported c.1474T4C S492P(*) C-term 0 1 once in the current ESP6500 release of the NHLBI GO Exome c.1865C4T S622F C-term 0 1 Sequencing Project. However, as the ESP project cannot release c.1959C4T T653T C-term 11 8 phenotype information about any particular individual, no Abbreviations: C-term, Carboxy-terminal domain; PAS, Per-Arnt-Sim conclusions regarding causality of the variants can be drawn. domain. All subjects referred to as ‘obese cases’ in this table are obese as defined by the ‘Flemish Growth Charts 2004’.15 Variants indicated with Reporter assay an (*) are novel variants, as they have not been previously reported. To evaluate the effect of the mutations on SIM1 trancriptional activity, we performed an in vitro luciferase reporter assay. For T481K and A517V-mutant SIM1, a decrease in transcriptional activity is observed, when co-transfected with the obligate dimerization partner ARNT2. Relative to wild-type SIM1 (100%), their transcriptional activities were reduced to 81.1±7.4% and 55.1±7.1%, respectively (Po0.05). However, transcriptional activity for all other variants (including the variants found in lean controls) resembled wild type (Figure 2).

Phenotype of T481K and A517V SIM1 variant carriers T481K. The patient, in whom we discovered the T481K mutation, is a 20-year-old female presenting with obesity. Her body weight is 99 kg for 1.70 m, with a BMI of 34.3 kg m À 2, the waist-to-hip ratio (WHR) is 0.98 and the fat mass percentage (measured by bioelectrical impedance analysis) is 48.1%. The patients father is also obese as well as the maternal grandparents. However, as no Figure 2. Reporter assay for SIM1 transcriptional activity. HEK293 further family data were available, it was not possible to determine cells were transiently transfected with pGL3-HRE-Luc, pRL-TK, whether the mutation segregates with obesity in this family. pCMV6-XL4_ARNT2 and WT or mutant SIM1. The influence of the mutations on SIM1 transcriptional activity was evaluated using A517V. The patient harboring the A517V mutation is a 13-year- luciferase reporter assay. Data were generated in three independent old boy, suffering from obesity with an early onset. Body weight experiments, each performed in duplicate. Bars represent mean reaches 87.5 kilograms with a BMI of 30.5 kg m À 2. BMI Z-score is values±s.d. *P o0.05 cells transfected with mutant SIM1 compared 2.23. The patient already suffered from several complications with cells transfected with WT SIM1 (100%). All data were corrected because of the extreme obesity and showed hypertriglyceridemia for differences in transfection efficiency by co-transfection with (248 mg dl À 1), hyperinsulinism (36 mUmlÀ 1) and hepatic steatosis. constitutively active Renilla luciferase expression plasmids (pRL-TK). Both parents were overweight as well and the mother underwent gastric banding surgery. Unfortunately, both were not available for DISCUSSION genetic analysis. As no family members were willing to cooperate The involvement of SIM1 in the leptin-melanocortin signaling for further research, it could not be determined whether the pathway that regulates energy homeostasis and food intake is mutation was inherited from one of the overweight parents or had firmly established. Mice heterozygous for Sim1 develop severe, arisen de novo. early-onset obesity and are reported to be hyperphagic.4

International Journal of Obesity (2014) 1000 – 1004 & 2014 Macmillan Publishers Limited Involvement of SIM1 mutations in childhood obesity D Zegers et al 1003 In addition, SIM1 deficiency in humans is associated with a severe, performed in vitro functional experiments to evaluate the early-onset obesity phenotype.8–11,27 Different research groups set transcriptional activity of the (mutated) SIM1 protein. We were out to determine the exact role of the SIM1 gene in human obesity able to demonstrate that both the T481K and the A517V mutation by performing both mutation screens and association studies.28–30 affect SIM1 transcriptional activity, whereas transcriptional activity Until recently, reports on SIM1 mutation screening in cohorts of of all other identified variants resembled wild type. obese individuals have been scarce, as only one missense While the ‘common disease, common variant’ hypothesis has mutation (I128T) had been reported in a 5-year-old Caucasian led to inconsistent results regarding the influence of common girl with severe early-onset obesity, hyperphagia and SIM1 variants on obesity susceptibility, one might suggest that developmental delay for which disease causality could not be there could be a possibility for the ‘common disease, rare variant’ proven as the variant did not cosegregate with obesity in the hypothesis to hold true for the SIM1 gene. This hypothesis family.29 Only recently, two novel studies on SIM1 rare mutations suggests an important role of multiple rare gene variants of major in obesity emerged.13,14 In the paper by Bonnefond et al.,13 the effect, each of which is found in only a few people, to underlie authors report on four variants identified in children with Prader- susceptibility to common disease. However, our data emphasize Willi-like syndrome features and four other variants in morbidly that the firm identification of these rare, causative variants for obese adults. After performing luciferase gene reporter assays, obesity is not an easy task because of the background noise three of the variants showed strong loss-of-function effects. created by multiple rare variants also found in lean control Corroborated by functional experiments, they established a firm individuals. The most decisive evidence to distinguish rare causal link between SIM1 loss-of-function mutations and severe obesity. variants from benign, neutral variants can only be provided by The authors also reported that mutation of SIM1 is not always extensive co-segregation analysis combined with comprehensive responsible for a fully penetrant form of obesity. Contradictory to functional in vitro and even in vivo experiments. our results, they did not identify any novel rare mutations in their As most of the variants identified in the present study were population of 383 normal weight controls.13 In the paper by found scattered throughout the carboxy-terminal domain of the Ramachandrappa et al.,14 2100 individuals with early-onset obesity protein with no clear clustering pattern for the variants found in have been screened for mutations in the SIM1 gene. Nine out of obese cases compared with the variants found in lean controls thirteen different heterozygous SIM1 variants significantly reduced (Figure 1), there was no a priori indication for disease causality of SIM1 transcriptional activity; however, they were reported to any of the variants. Nevertheless, by performing in vitro experi- segregate with obesity with variable penetrance. Further, variant ments we demonstrated that two of the identified variants (T481K carriers presented with an increased ad libitum food intake at a and A517V) decreased SIM1 transcriptional activity and that final test meal, normal basal metabolic rate and evidence of autonomic conclusions regarding pathogenicity of the variants can only be dysfunction.14 based on functional studies. There are also several reports available in which the contribu- Building on the recently generated and growing body of data tion of common variants in the SIM1 gene to human complex regarding SIM1 mutations involved in the pathogenesis of obesity, obesity is analyzed.28–31 Results of these association studies are we evaluated the prevalence of rare SIM1 mutations in human rather conflicting, as an initial association of the coding individuals with severe early-onset obesity as well as in lean polymorphisms rs3734354 (P352T) and rs3734355 (A371V) with controls in the current study. However, as rare variants were BMI in obese men29 was not replicated in a large scale association identified in both obese cases and lean controls and not all study performed in a French cohort comprising 1275 obese and variants seem to affect transcriptional activity, this raises questions 1395 lean control individuals.30 Further, the association of a regarding the pathogenic impact of SIM1 variants on obesity. linkage disequilibrium block spanning from 50-UTR to intron 8 with Therefore, future perspectives include more extensive and obesity risk in a population of 6194 Pima Indians31 could also not detailed studies to clarify the role of rare SIM1 mutations and to be replicated in a cohort of French individuals.30 In the most evaluate complementary mechanisms such as gene–gene and recent study, homozygosity for the 352T/371V allele was gene–environment interactions, as these may add to disease risk associated with higher BMI in a white male population and and contribute to the complex relationship between genotype three other SNPs (rs9390322, rs7746743 and rs3734353) were and phenotype. significantly associated with various adiposity measures,28 indicating that the role for common genetic variants in SIM1 is still not clearly defined. In this study, we have demonstrated the presence of rare SIM1 CONFLICT OF INTEREST variants in both an obese pediatric population as well as a The authors declare no conflict of interest. population of lean adult controls. The majority of these variants is situated in the C-terminal domain of the transcription factor, whereas the L242V variant is located in the PAS2 domain ACKNOWLEDGEMENTS (Figure 1). Observation from deletion studies showed that the Research was funded by a PhD grant of the Institute for the Promotion of Innovation HLH-PAS domain in SIM1 is a bipartite dimerization domain with through Science and Technology in Flanders (IWT-Vlaanderen) to DZ. SB is a the general dimerization partner ARNT2 and that the C-terminus postdoctoral fellow of the Research Foundation—Flanders (FWO Vlaanderen). This 32 contains the transcriptional regulatory domains. Further, there is work was supported by a grant (G0028.05) from the Research Foundation—Flanders a high level of homology between the human SIM1 and SIM2, (FWO Vlaanderen) to LVG and WVH and by a TOP-research grant from the University mouse Sim1 and Sim2 and even drosophila sim genes in the of Antwerp to WVH. This study was supported by an Interuniversity Attraction Pole amino-terminal part of the protein in which the conserved bHLH Project (Phase VII project 43, BELSPO). (basic helix-loop-helix), PAS (Per-Arnt-Sim) and PAC (PAS- associated C-terminal) domains are located. The carboxy- terminal part is only homologous between human SIM1 and REFERENCES mouse Sim1.33 When comparing peptide sequences, amino acids L242 and T481 show the strongest conservation compared with 1 Michaud JL, Rosenquist T, May NR, Fan CM. Development of neuroendocrine lineages requires the bHLH-PAS transcription factor SIM1. Genes Dev 1998; 12: the other sim genes. However, as all variants identified in this 3264–3275. study are situated in crucial protein domains, it is impossible to 2 Palkovits M, Mezey E, Eskay RL. Pro-opiomelanocortin-derived peptides differentiate between benign and functionally relevant variants (ACTH/beta-endorphin/alpha-MSH) in brainstem baroreceptor areas of the rat. based on their location within the protein. Therefore, we Brain Res 1987; 436: 323–338.

& 2014 Macmillan Publishers Limited International Journal of Obesity (2014) 1000 – 1004 Involvement of SIM1 mutations in childhood obesity D Zegers et al 1004 3 Mountjoy KG, Mortrud MT, Low MJ, Simerly RB, Cone RD. Localization of the 18 Zhou L, Wang L, Palais R, Pryor R, Wittwer CT. High-resolution DNA melting melanocortin-4 receptor (MC4-R) in neuroendocrine and autonomic control analysis for simultaneous mutation scanning and genotyping in solution. circuits in the brain. Mol Endocrinol 1994; 8: 1298–1308. Clin Chem 2005; 51: 1770–1777. 4 Michaud JL, Boucher F, Melnyk A, Gauthier F, Goshu E, Levy E et al. Sim1 19 Zegers D, Beckers S, de Freitas F, Peeters AV, Mertens IL, Verhulst SL et al. haploinsufficiency causes hyperphagia, obesity and reduction of the para- Identification of three novel genetic variants in the melanocortin-3 receptor of ventricular nucleus of the hypothalamus. Hum Mol Genet 2001; 10: 1465–1473. obese children. Obesity 2011; 19: 152–159. 5 Sims JS, Lorden JF. Effect of paraventricular nucleus lesions on body weight, food 20 Ramensky V, Bork P, Sunyaev S. Human non-synonymous SNPs: server and survey. intake and insulin levels. Behav Brain Res 1986; 22: 265–281. Nucleic Acids Res 2002; 30: 3894–3900. 6 Tolson KP, Gemelli T, Gautron L, Elmquist JK, Zinn AR, Kublaoui BM. Postnatal 21 Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P et al. Sim1 deficiency causes hyperphagic obesity and reduced Mc4r and oxytocin A method and server for predicting damaging missense mutations. Nat Methods expression. J Neurosci 2010; 30: 3803–3812. 2010; 7: 248–249. 7 Kublaoui BM, Holder Jr JL, Tolson KP, Gemelli T, Zinn AR. SIM1 overexpression 22 Ng PC, Henikoff S. Predicting deleterious amino acid substitutions. Genome Res partially rescues agouti yellow and diet-induced obesity by normalizing food 2001; 11: 863–874. intake. Endocrinology 2006; 147: 4542–4549. 23 Calabrese R, Capriotti E, Fariselli P, Martelli PL, Casadio R. Functional annotations 8 Holder Jr JL, Butte NF, Zinn AR. Profound obesity associated with a balanced improve the predictive score of human disease-related mutations in proteins. translocation that disrupts the SIM1 gene. Hum Mol Genet 2000; 9: 101–108. Hum Mutat 2009; 30: 1237–1244. 9 Faivre L, Cormier-Daire V, Lapierre JM, Colleaux L, Jacquemont S, Genevieve D et al. 24 Li B, Krishnan VG, Mort ME, Xin F, Kamati KK, Cooper DN et al. Automated Deletion of the SIM1 gene (6q16.2) in a patient with a Prader-Willi-like phenotype. inference of molecular mechanisms of disease from amino acid substitutions. J Med Genet 2002; 39: 594–596. Bioinformatics 2009; 25: 2744–2750. 10 Varela MC, Simoes-Sato AY, Kim CA, Bertola DR, De Castro CI, Koiffmann CP. A new 25 1000 Genomes Project Consortium. A map of human genome variation from case of interstitial 6q16.2 deletion in a patient with Prader-Willi-like phenotype population-scale sequencing. Nature 2010; 467: 1061–1073. and investigation of SIM1 gene deletion in 87 patients with syndromic obesity. 26 Exome Variant Server, NHLBI Exome Sequencing Project (ESP), Seattle, Eur J Med Genet 2006; 49: 298–305. WA [database on the Internet] 2012, [cited July 14]. Available from: URL 11 Bonaglia MC, Ciccone R, Gimelli G, Gimelli S, Marelli S, Verheij J et al. Detailed http://evs.gs.washington.edu/EVS/. phenotype-genotype study in five patients with chromosome 6q16 deletion: 27 Wang JC, Turner L, Lomax B, Eydoux P. A 5-Mb microdeletion at 6q16.1-q16.3 with narrowing the critical region for Prader-Willi-like phenotype. Eur J Hum Genet SIM gene deletion and obesity. Am J Med Genet A 2008; 146A: 2975–2978. 2008; 16: 1443–1449. 28 Swarbrick MM, Evans DS, Valle MI, Favre H, Wu SH, Njajou OT et al. Replication and 12 Stutzmann F, Ghoussaini M, Couturier C, Marchand M, Vatin V, Corset L et al. extension of association between common genetic variants in SIM1 and human Loss-of-function mutations in SIM1 cause a specific form of Prader-Willi-like adiposity. Obesity 2011; 19: 2394–2403. syndrome. Diabetologia 2009; 52(Suppl 1): 104. 29 Hung CC, Luan J, Sims M, Keogh JM, Hall C, Wareham NJ et al. Studies of the SIM1 13 Bonnefond A, Raimondo A, Stutzmann F, Ghoussaini M, Ramachandrappa S, gene in relation to human obesity and obesity-related traits. Int J Obes 2007; 31: Bersten DC et al. Loss-of-function mutations in SIM1 contribute to obesity and 429–434. Prader-Willi-like features. J Clin Invest 2013; 123: 3037–3041. 30 Ghoussaini M, Stutzmann F, Couturier C, Vatin V, Durand E, Lecoeur C et al. 14 Ramachandrappa S, Raimondo A, Cali AM, Keogh JM, Henning E, Saeed S et al. Analysis of the SIM1 contribution to polygenic obesity in the French population. Rare variants in single-minded 1 (SIM1) are associated with severe obesity. J Clin Obesity 2010; 18: 1670–1675. Invest 2013; 123: 3042–3050. 31 Traurig M, Mack J, Hanson RL, Ghoussaini M, Meyre D, Knowler WC et al. Common 15 Roelants M, Hauspie R, Hoppenbrouwers K. References for growth and pubertal variation in SIM1 is reproducibly associated with BMI in Pima Indians. Diabetes development from birth to 21 years in Flanders, Belgium. Ann Hum Biol 2009; 36: 2009; 58: 1682–1689. 680–694. 32 Crews ST. PAS Proteins: Regulators and Sensors of Development and Physiology. 16 Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting 1 edn, Kluwer Academic Publishers, 2003. DNA from human nucleated cells. Nucleic Acids Res 1988; 16: 1215. 33 Chrast R, Scott HS, Chen H, Kudoh J, Rossier C, Minoshima S et al. Cloning of two 17 Wittwer CT, Reed GH, Gundry CN, Vandersteen JG, Pryor RJ. High-resolution human homologs of the Drosophila single-minded gene SIM1 on chromosome genotyping by amplicon melting analysis using LCGreen. Clin Chem 2003; 49: 6q and SIM2 on 21q within the Down syndrome chromosomal region. Genome 853–860. Res 1997; 7: 615–624.

International Journal of Obesity (2014) 1000 – 1004 & 2014 Macmillan Publishers Limited