Large, rare chromosomal deletions associated with severe early-onset obesity

Abstract

Obesity is a highly heritable and genetically heterogeneous disorder. Here we investigated the contribution of copy number variation to obesity in 300 Caucasian patients with severe early-onset obesity, 143 of whom also had developmental delay. Large (>500 kilobases), rare (<1%) deletions were significantly enriched in patients compared to 7,366 controls (P < 0.001). We identified several rare copy number variants that were recurrent in patients but absent or at much lower prevalence in controls. We identified five patients with overlapping deletions on chromosome 16p11.2 that were found in 2 out of 7,366 controls (P < 5 x 10(-5)). In three patients the deletion co-segregated with severe obesity. Two patients harboured a larger de novo 16p11.2 deletion, extending through a 593-kilobase region previously associated with autism and mental retardation; both of these patients had mild developmental delay in addition to severe obesity. In an independent sample of 1,062 patients with severe obesity alone, the smaller 16p11.2 deletion was found in an additional two patients. All 16p11.2 deletions encompass several genes but include SH2B1, which is known to be involved in leptin and insulin signalling. Deletion carriers exhibited hyperphagia and severe insulin resistance disproportionate for the degree of obesity. We show that copy number variation contributes significantly to the genetic architecture of human obesity.

Figure 1. Discovery of 16p11.2 CNV associated with severe early-onset obesity

a, Affymetrix 6.0 array data for five patients with deletions at 16p11.2 is shown. Log₂ ratios of the five samples are highlighted in dark red, with other samples in the same genotyping plate shown in grey. The structure of extensive segmental duplication that extends to the flanking regions is shown. Annotation of the segmental duplications was taken from UCSC Genome Browser and the darkness of colour coding represents sequence similarity between the duplicated pairs. Protein-coding genes are represented by dark-blue lines; SH2B1 is highlighted in red and by blue vertical shading. The light-pink vertical shading indicates the range of a previous BMI-association signal found in two genome-wide association studies,; the light-grey vertical shading indicates the reported autism-associated CNV region-,. b, c, Pedigrees are shown for the five patients with 16p11.2 deletions. Families are numbered as in Table 2. Filled symbols represent patients with severe early-onset obesity; arrows indicate probands. Age, BMI (body mass index) and BMI s.d.s. (standard deviation score) for children are included where available. Presence (del) or absence (no del) of the 16p11.2 SH2B1-containing CNV is shown where known. Representative MLPA data are shown. MLPA probes for genes in the region of interest are shown. The MLPA target regions labelled as C are control probes located either on chromosome 16 but outside the deleted region or on other chromosomes (Supplementary Information). Patient MLPA traces are in red, overlaid upon the normal control MLPA traces in black. Arrows point to the deleted probes. b, Three probands in whom the 16p11.2 SH2B1-containing deletion co-segregates with severe early-onset obesity alone. c, Two probands harbouring larger de novo 16p11.2 deletions that also encompass a known autism-associated locus and are associated with developmental delay and severe early-onset obesity.

Figure 2. Metabolic phenotype of carriers of the 16p11.2 deletion

a, Weights for three boys with the 16p11.2 deletion are plotted on growth charts based on UK reference data. b, c, Percentage body fat measured by dual energy X-ray absorptiometry (DEXA) (b) and ad libitum energy intake at a test meal (adjusted for fat free mass (FFM) in kilograms) (c) is shown for 16p11.2 deletion carriers (n = 5), leptin-receptor-deficient subjects (n = 10) and normal weight controls (n = 35). Means ± s.e.m. (error bars) are indicated. d, Fasting plasma insulin levels for 16p11.2 deletion carriers (n = 6), leptin-receptor-deficient subjects (n = 10), MC4R-deficient subjects (n = 35), and age and BMI s.d.s. matched controls from the GOOS cohort (n = 535) are shown. Plasma insulin values were analysed after log₁₀-transformation and are presented as geometric mean (1 s.e. range) after back-transformation. e, Plasma insulin levels in response to a 75 g oral glucose load are shown in adult 16p11.2 deletion carriers (n = 3) and severely obese adult controls matched for fasting plasma insulin levels (n = 10) (Supplementary Information). Plasma insulin values were analysed after log₁₀-transformation and are presented as geometric mean (1 s.e. range) after back-transformation. Results were compared using unpaired t-tests. All P values were two-sided. P < 0.05 was considered statistically significant. *P < 0.01; ***P < 0.001.