Nucleotide variation of sFRP5 gene is not associated with obesity in children and adolescents.

Because sFRP5 was shown to be an important extracellular modulator of the Wnt pathway, regulating adipogenesis, we wanted to investigate the role of sFRP5 variants in human, monogenic obesity by performing mutation analysis. We screened the complete sFRP5 coding region in 622 obese children and adolescents and 503 lean control individuals by high-resolution melting curve analysis and direct sequencing. We found a total of 15 sequence variants in sFRP5, 10 of which resulted in a non-synonymous amino acid change. Five of these variants were, to our knowledge, not previously reported. For one of the variants (c.-3G>A), we identified a trend towards association between the variant frequency and the obese phenotype. We argue that, when looking at conservation and location inside known protein domains, several of the identified variants (D103N, A113V, K212N and H317L), may affect sFRP5 protein function. In addition, we found c.-3G>A, residing in the Kozak sequence, with a lower frequency in cases compared to controls. However, functional studies investigating the effect of sFRP5 variants on protein function are necessary to determine the true role of sFRP5 genetic variation in human, monogenic obesity.

Genetic and structural variation in the SH2B1 gene in the Belgian population.

OBJECTIVE: Animal studies, genome-wide association and genomic structural variation studies have identified the SH2B1 gene as a candidate gene for obesity. Therefore, we have designed an extensive mutation and copy number variation (CNV) analysis investigating the prevalence of genetic and structural variations in SH2B1 in the Belgian population. DESIGN AND METHODS: In the first part of this study, we performed a mutation screen for variants in the SH2B1 coding region in 581 obese children and adolescents and 433 healthy, lean individuals with high-resolution melting curve analysis followed by direct sequencing. In the second part of this study, Multiplex Amplicon Quantification (MAQ) analysis was used to identify CNVs in the distal SH2B1-containing chr.16p11.2 region in 421 obese children and adolescents with no developmental delay or behavioral phenotype. RESULTS: Mutation analysis resulted in the identification of fifteen rare non-synonymous heterozygous variants. Several of these were found both in lean and obese subjects, suggesting that these are neutral polymorphisms. However, six private, heterozygous, non-synonymous variations were present in obese children only. Furthermore, we also identified six missense variants solely in lean individuals. CNV analysis could not identify carriers of the distal 16p11.2 deletion in our population. CONCLUSION: Our mutation analysis has demonstrated that variation in the SH2B1 gene is frequent in both lean and obese groups, with distinctive variations being present on either side of the weight spectrum. Although the equal variation frequency does not immediately support disease causality, it cannot be excluded that some variations are weight-increasing or -decreasing. Further functional testing of the variants will be necessary to fully understand the impact of these variants on SH2B1. We were not able to detect carriers of the distal 16p11.2 deletion in our study population. As we excluded patients with developmental or behavioral problems, we suggest that in addition to obesity, the distal deletion might predispose for these traits. Further characterization of the phenotype is therefore necessary to clearly identify the phenotype of the distal 16p11.2 microdeletion syndrome.

Resistin polymorphisms show associations with obesity, but not with bone parameters in men: results from the Odense Androgen Study.

Resistin is an obesity-related adipokine which has also been implicated in bone metabolism. Therefore, we designed a study to investigate the possible role of resistin gene variation in both obesity and bone mineral density. We included 1,155 individuals from the Odense Androgen Study (663 young subjects and 492 older subjects), a population-based, prospective, observational study on the inter-relationship between endocrine status, body composition, muscle function, and bone metabolism in men, in an association study with resistin (RETN) polymorphisms. Three RETN variants (rs1862513, rs3745367 and rs3745369) were genotyped with TaqMan Pre-Designed Genotyping assays. Linear regression was performed to investigate the possible association of these variants with several obesity- and bone-related parameters. After genotyping 1,155 Danish men, 663 young subjects and 492 older subjects, we found that rs3745367 was associated with several obesity-related measures in both the young and elderly cohort. Rs3745369 was only associated with obesity-phenotypes in the elderly cohort. When studying the combined cohorts, we could confirm the associations of rs3745367 with several obesity-related parameters. We were unable to identify any association between RETN polymorphisms and bone-related measurements. Together, these results illustrate resistin's role in the development of obesity. Rs3745367 gives the most consistent results in the current study and these should be confirmed in other populations. Research into its possible functional effect might also be required. A role for RETN variants in determining bone mineral density seems unlikely from our results.

No important role for genetic variation in the Chibby gene in monogenic and complex obesity.

Chibby (CBY) has been identified as a potent proadipogenic factor required for adipocyte differentiation. It has been shown that CBY inhibits the canonical Wnt pathway, and therefore promotes the development of new fat cells. Our objective therefore is to investigate the contribution of rare and common genetic variation in CBY to the development of human obesity. A mutation analysis was performed on a total of 566 obese patients and 432 lean individuals. To investigate the involvement of CBY in complex obesity, we performed a genetic association analysis of the entire CBY gene region on 1,011 obese individuals and 523 control samples. Four rare, novel variants were identified in either obese patients or lean control subjects, among which two non-synonymous variations and one frameshift mutation. In addition, four previously reported CBY variants were found. In the association analysis, logistic and linear regression showed no association between common genetic variation in CBY and obesity parameters. Several novel variations were found, but no definite role in the pathogenesis of obesity could be confirmed. Results from the association analysis suggest that common variation in CBY is not a cause for obesity in the Belgian population.

Identification of mutations in the NUCB2/nesfatin gene in children with severe obesity.

Nesfatin-1 is the N-terminal fragment of nucleobindin-2 (NUCB2) that was identified as a novel satiety molecule in rodents. The protein is reported to exert anorexigenic effects and appears to play an important role in hypothalamic pathways regulating energy homeostasis and food intake. In this study, we hypothesized that mutations in the nesfatin encoding gene NUCB2 might cause obesity in humans. Therefore, we screened the entire coding region of the NUCB2 gene for mutations in a population of 471 obese children and adolescents. Mutation analysis of NUCB2 identified a total of seven sequence variants of which four were previously reported as polymorphisms. The remaining three variants included ex9+6G>C, L125H and K178X and were found in 3 unrelated individuals in the obese population only (0.6%). Biochemical experiments including ELISA and western blot were performed on plasma samples of the obese patient carrying the nonsense mutation K178X. However, neither NUCB2/nesfatin-1 immunoreactive plasma levels of the patient, nor expression of full length NUCB2 differed significantly from matched obese control individuals. In conclusion, we have identified the first genetic variants in the NUCB2 gene in obese individuals, although further functional characterization will be essential to verify disease causality of the mutations.

Association study of PNPLA2 gene with histological parameters of NAFLD in an obese population.

INTRODUCTION: The prevalence of non-alcoholic fatty liver disease (NAFLD) and the closely associated metabolic syndrome is high and is related to risk factors such as obesity and type 2 diabetes. A genetic basis for NAFLD has been suggested, but only few causal genes have been identified. The most significant association reported to date is the robust association of the PNPLA3 I148M variant with susceptibility to NAFLD. We therefore hypothesized that the PNPLA2 gene might also be involved in NAFLD pathogenesis, because of its close sequence similarity with PNPLA3 and its possible involvement in ectopic fat accumulation. METHODS: In this study, we investigated the association of PNPLA2 polymorphisms with the development of non-alcoholic fatty liver disease in a prospectively recruited Belgian obese population comprising 633 individuals with varying degrees of fatty liver disease. We selected 3 PNPLA2 SNPs for genotyping, including 2 tagSNPs that cover most information on common genetic variation in the selected region. RESULTS: After performing linear regression analysis, we found that 2 of the analyzed PNPLA2 SNPs were associated with anthropometric and metabolic parameters. In our subcohort of patients that underwent liver biopsy (n=372/633 or 58.7%), we assessed the influence of the PNPLA2 variants on the severity of histologically determined liver damage, but we did not find convincing evidence for association. CONCLUSION: Although we found evidence for moderate association between PNPLA2 tagSNPs and anthropometric and metabolic parameters in our cohort, no evidence for association between polymorphisms in the PNPLA2 gene and the presence and severity of NAFLD was identified.

Investigation of common and rare genetic variation in the BAMBI genomic region in light of human obesity.

The aim of this study was to confirm the previously identified link between BAMBI and human obesity by means of a genetic and functional analysis. We performed both a mutation analysis, using high-resolution melting curve analysis, and a genetic association study, including 8 common tagSNPs in the BAMBI gene region. Three of the identified genetic variants (R151W, H201R, and C229R) were evaluated for their Wnt signaling enhancing capacity in a Wnt luciferase reporter assay. Mutation screening of the BAMBI coding region and exon-intron boundaries on our population of 677 obese children and adolescents and 529 lean control subjects resulted in the identification of 18 variants, 10 of which were not previously reported and 12 of which were exclusively found in obese individuals. The difference in variant frequency, not taking into account common polymorphisms, between obese (3.1 %) and lean (0.9 %) subjects was statistically significant (p = 0.004). Our Wnt luciferase assay, using WT and mutant BAMBI constructs, showed a significantly reduced activity for all of the investigated variants. Logistic and linear regression analysis on our Caucasian population of 1022 obese individuals and 606 lean controls, did not identify associations with obesity parameters (p values >0.05). We found several rare genetic variations, which represent the first naturally occurring missense variants of BAMBI in obese patients. Three variants (R151W, H201R, and C229R) were shown to reduce Wnt signaling enhancing capacity of BAMBI and we believe this result should encourage further study of this gene in other obese populations. In addition, we did not find evidence for the involvement of BAMBI common variation in human obesity in our population.

Genetic association study of WNT10B polymorphisms with BMD and adiposity parameters in Danish and Belgian males.

Because of the importance of the Wnt pathway in the development and maintenance of both adipose and bone tissue, we wanted to evaluate the involvement of WNT10B, a Wnt pathway activator, in adipogenesis and osteoblastogenesis in humans. Genetic association between WNT10B polymorphisms and adiposity parameters as well as bone mineral density (BMD) measurements was analysed in two independent populations. The first is a population of 1,228 Danish men (702 aged 20-29 years; 532 aged 60-74 years) from the Odense Androgen Study (OAS), which was designed as a cross-sectional, population-based study. The second population, called SIBLOS, includes 922 Belgian men (34 ± 5 years old) and contains siblings selected from over 500 families. Four tagSNPs (rs833840, rs833841, rs10875902 and rs4018511) that capture variation of ten SNPs (MAF > 5 %) in a 15.2 kb region spanning the WNT10B gene and its flanking regions were genotyped. Although no association with body mass index was found, we found all tagSNPs to be associated with BMD parameters (BMD whole body, total hip and femoral neck) and height in the OAS population. The association of rs10875902 was most prominent (nominal p = 0.012) and confirmed a previously shown negative effect on BMD. No significant associations were observed in the SIBLOS population. In the present study, no association between WNT10B polymorphisms and adiposity parameters was found. However, our results clearly illustrate a role for WNT10B variants in determining human BMD. The effect of WNT10B polymorphisms on height should be evaluated in additional populations.

Prevalence of rare MC3R variants in obese cases and lean controls.

The role of mutations in the melanocortin-3 receptor (MC3R) gene, which is implicated in the regulation of energy homeostasis, is still under debate. Animal studies have clearly proven that, together with the melanocortin-4 receptor (MC4R), the MC3R is a critical receptor for melanocortin peptides within the leptin-melanocortin signaling cascade. However, as several mutations have been found in lean individuals and not all mutations seem to cause receptor dysfunction, results from mutation screens in obese humans remain controversial. In the present study, we screened for rare variants in the MC3R gene of obese children and lean controls to assess the prevalence of MC3R mutations in the Belgian population. We screened 249 severely overweight and obese children and adolescents and 239 lean adults for mutations in the coding region of MC3R. Mutation screening was performed by high resolution melting curve analysis and direct sequencing. We identified four non-synonymous coding variations in the obese population, all of which had been reported previously. In addition, we also found four novel rare MC3R variants in the lean control population, suggesting that not all MC3R mutations are disease-causing. Overall, the total prevalence of rare MC3R variants was 1 % in Belgian obese children and adolescents compared to 1.02 % in lean controls. Ultimately, cosegregation studies combined with comprehensive functional analysis is required to determine the potential pathogenic role of rare MC3R variants in causing human obesity.

Mutation analysis of WNT10B in obese children, adolescents and adults.

Wingless-type MMTV integration site family, member 10B (WNT10B) is an activator of the Wnt pathway. The Wnt pathway is known to play an important role in maintenance and differentiation of stem cells and has been implicated in the origination of obesity. To evaluate the role of genetic variation in WNT10B in obesity further, we performed a mutation analysis on Belgian obese patients and control subjects. A mutation analysis of WNT10B by means of high-resolution melting curve analysis and direct sequencing was performed on 546 obese children and adolescents (mean Z-score of 2.6 ± 0.6 and 2.5 ± 0.4 respectively), 86 morbidly obese adults (mean BMI of 48.0 ± 0.4 kg/m(2)) and 447 lean, healthy controls (mean BMI of 22.1 ± 1.7 kg/m(2)). A total of five novel non-synonymous variants were identified. R228Q was the only coding, non-synonymous variant that was exclusively found in patients, but the variant did not co-segregate with obesity in the three investigated siblings. The remaining four variants were either found both in cases and in control samples (G181D) or only in control samples (A108P, S187R and P315S). The frequency of non-synonymous variants in lean individuals (0.9 %) was higher than in obese individuals (0.3 %) and familial co-segregation of the most promising variant in patients could not be demonstrated. Therefore, we conclude that variations in WNT10B do not contribute to human monogenic obesity in our population.