Background
Currently, childhood obesity is considered one of the most serious public health challenges of the twenty-first century. The prevalence of childhood obesity is increasing at an alarming rate, affecting high-income as well as low- and middle-income countries, and the number of overweight and obese children below the age of five is estimated to be 41 million [
http://www.who.int/topics/obesity/en.].
A strong genetic factor is evident in the etiology of obesity, with heritability estimates ranging from 40 to 70% [
1‐
3]. Genetic defects disrupting the leptin-melanocortin signaling pathway very often result in severe early-onset obesity and hyperphagia [
4‐
6], and the genes most commonly involved in monogenic forms of obesity are part of this pathway, including
leptin (
LEP), the
leptin receptor (
LEPR), and the
melanocortin 4 receptor (MC4R) [
7‐
9]. Leptin is a 16-KD hormone secreted by white adipocytes which binds to LEPR and regulates energy expenditure through hypothalamic neurons [
10]. LEPR is a member of the cytokine receptor family with six isoforms (LEPRa-f), yet, leptin signaling is primarily mediated by the long LEPRb expressed in the hypothalamus [
11‐
13]. The short form is expressed in a number of tissues, including the adrenal gland, kidney, lung, and choroid plexus [
14,
15]. The mutations identified to date in
LEP and
LEPR are ethnic-specific, and the prevalence of monogenic obesity caused by mutations within these two genes is as high as > 20% in Pakistani study populations with obesity [
16‐
18].
MC4R encodes a 332- amino acid membrane-bound receptor protein [
19]. It is expressed in the brain and it influences appetite regulation through interaction with adrenocorticotropic and melanocyte stimulating hormones (MSH) through G proteins [
20]. In studies of European populations with severe obesity, the prevalence of heterozygous damaging
MC4R mutations is ~ 6% and ~ 2% in children and adults, respectively [
21‐
23], signifying the different etiologies of childhood and adult obesity.
The disruption of
LEPR and leptin deficiency results in severe early-onset hyperphagic obesity with rapid weight increase during the first few months of life [
17,
24,
25]. Similarly, during the first year of life, MC4R deficiency has been linked with hyperphagia and increased fat and lean mass, increased linear height, increased bone mineral density and severe early hyperinsulinemia [
22,
26], although some of these associations remain controversial [
27,
28]. Obesity caused by damaging mutations in
LEP and
LEPR display an autosomal recessive mode of inheritance while obesity caused by
MC4R mutations exhibit variable penetrance (recessive or co-dominant) [
23].
Genetic sequencing of consanguineous families is an important tool in the identification of deleterious mutations in genes implicated in monogenic forms of obesity [
18], as consanguineous families share homozygous regions in their genomes, enabling the identification of deleterious recessive mutations [
29]. Due to the high degree of inbreeding in the Pakistani population, homozygous deleterious mutations in
LEP, LEPR, and
MC4R have been identified in as many as ~ 30% of severely obese individuals from consanguineous families [
16‐
18]. In comparison, the prevalence of heterozygous
MC4R mutations is 3–5% in Caucasian populations with early-onset obesity [
30,
31]. Both
LEP and
LEPR mutations are rare in Caucasian populations.
The application of next generation sequencing (NGS) provides a powerful method to discover rare disease-causing genetic variants [
32]. Therefore, targeted resequencing of
LEP, LEPR and
MC4R was performed to assess the prevalence of damaging mutations within these genes. In total, genetic screening was performed in 25 severely obese probands from Pakistani families of which 14 probands were from families with known consanguineous marriages.
Methods
Study design
This family-based study applied a phenotype-driven approach to investigate monogenic causes of obesity in selected Pakistani families with obesity segregating in an autosomal recessive pattern.
Participants
Twenty-five families, originating from different regions of Pakistan, were recruited and examined at Children Hospital, Pakistan Institute of Medical Sciences (PIMS), Islamabad. Consanguineous marriages were known in 14 of the included families. The selection of the families was based on three criteria: 1) Body mass index (BMI) of probands > 30 kg/m2; 2) Probands displaying obesity onset before five years of age; and 3) Parents of the probands with BMI < 25 kg/m2, consistent with an autosomal recessive mode of inheritance.
Clinical examination
Through interview sessions, information on age of obesity onset (years), other major chronic disease(s) (if any), metabolic disorder(s) running in the family, eating habits, physical activity, along with obesity-related co-morbidities was recorded. Waist circumference (cm) and height (cm) were measured with a non-elastic plastic tape with the participant standing in an upright position without shoes. Weight (kg), without shoes and in light clothes, was measured to the nearest 0.1 kg using a digital scale. From these measures, BMI was calculated as the weight in kilograms divided by the square of the height in meters (kg/m
2), and using the LMS method [
33], a BMI standard deviation score (SDS) was calculated based on a World Health Organization (WHO) reference population [
34]. Approximately 3–5 ml of venous non-fasting blood from affected and un-affected family members were collected in 8.5 ml vacutainer tubes (BD Vacutainer® ACD, Franklin Lakes NJ, USA). Clinical characteristics of the families are presented in Additional file
1.
DNA was extracted from blood samples from 36 affected and 88 unaffected family members. Genomic DNA was primarily extracted using the standard phenol-chloroform method [
35], however, in some families, the QIAamp DNA Mini Kit (Qiagen, Germany) was used.
Targeted resequencing
The probands (
n = 25) from each family as well as four additional affected individuals (OB2–6, OB4–8, OB4–9, and OB4–10) underwent targeted resequencing. Using a chip-based customized nucleotide probe, targeted resequencing was performed to examine the coding regions of
LEP, LEPR, and
MC4R. Methods for target region capture and NGS have previously been extensively described [
36]. According to the manufacturer’s standard cluster generation and sequencing protocols, the final captured DNA libraries were sequenced using the Illumina HiSeq2000 Analyzers as PE 90 bp reads. Only variants having a minimum mean depth of 20x were included. Identified variants were annotated according to the transcripts 1)
LEP: NM_000230; 2)
LEPR: NM_002303.5 and 3)
MC4R: NM_005912.
Chip genotyping
Illumina Infinium Human CoreExome Bead Chip (CoreExomeChip) genotyping was performed in 124 individuals from 25 families using Illumina’s HiScan system at the laboratory facilities of the Novo Nordisk Foundation Center for Basic Metabolic Research at Symbion, Copenhagen, Denmark. The standard pipeline in Illumina Genome Studio software was used for the genotype calling. The pipeline yielded 551,839 genetic variations, which entered our quality control (QC) pipeline. The Illumina final report was converted to plink format using custom scripts and aligned with the positive strand of the GRCh37 reference.
QC included removal of variants with missing genotypes above 5% as well as individuals with more than 5% missing genotype calls, individuals with negative inbreeding, indication of duplicated samples, and discrepancy between genetic identified pedigree and the pedigree obtained from the family. A total of 12 individuals were removed in the QC. We did not remove SNPs deviating from Hardy-Weinberg equilibrium as is otherwise usual, since these variants may be of interest given the specific mode of data sampling.
Homozygosity mapping
Based on genotyping, runs of homozygosity were determined in each family using the “homozyg” command in PLINK [
37].
Discussion
In the present study, we employed targeted resequencing as a means to identify the genetic etiology in cases with severe early-onset obesity. The sequencing data was analyzed with respect to the three main genes known to be involved in monogenic forms of obesity, i.e.
LEP, LEPR and
MC4R. Given the high prevalence of
LEP,
LEPR and
MC4R mutations previously reported in Pakistani populations [
18], a similarly high prevalence of causal variants within these genes was expected. However, only two of the 25 examined probands carried homozygous recessive mutations, and both mutations were positioned in
LEPR.
LEPR mutations have previously been reported to influence the risk of developing severe early-onset obesity, hypogonadotropic hypogonadism and hypothalamic hypothyroidism [
24,
44], which is similar to the clinical characteristics of leptin deficiency [
24]. Furthermore, in mice,
Lepr mutations have been found to influence the susceptibility of type 2 diabetes [
45]. Hypogonadotropic hypogonadism in LEPR-deficient individuals may be due to a defect both at the hypothalamic and the pituitary level [
46]. However, hypogonadism may change over time, as in the case of spontaneous pubertal development and a natural pregnancy [
24].
Both of the novel
LEPR mutations identified in the current study, are positioned in the intracellular domain of LepRb, which is involved in energy homeostasis, glucose metabolism, fertility, growth and the action of insulin [
47,
48] . Upon binding, leptin activates the LepRb through the mediation of multiple signaling pathways, including phosphorylation of cytoplasmic tyrosine kinases of Janus Kinase 2 (JAK2), conscription of signal transducer and activator of transcription 3 (STAT3) and mitogen-activated protein kinase (MAPK) cascade [
47,
49‐
51]. In both humans and mice, multiple forms of LepRb are known, including short intracellular domain forms ranging from 32 to 40 amino acids and the long form comprising 303 amino acids, which is predominantly expressed in the hypothalamus [
14,
48,
52‐
54]. The missense mutation (p.Pro892Arg), identified in a family OB25, is located in the Box 1 motif which is important both for leptin-dependent JAK2 activation through the mediation of signaling by the intracellular domain and for the physiologic actions of leptin [
52,
55]. The CADD score of the identified mutation (p.Pro892Arg) is 28, which indicates this is a very likely disease-causing variant. The second mutation, i.e. frameshift (Ser1090Trpfs*6) identified in the family OB4, is located in the long intracellular domain of LepRb and has sequence motifs resulting in the truncation of the domain, thereby suggesting a dysfunctional effect on its intracellular signal-transducing capabilities.
The highly deleterious nature of the identified two mutations (p.Pro892Arg and Ser1090Trpfs*6) is consistent with the clinical conditions of hyperphagia, rapid weight gain and extreme obesity observed in proband OB4–7. In both probands, the mutations were found in a homozygous state, and based on genotyping of family members, the homozygous region was found to be shared among affected individuals only, while parents of the probands were heterozygous carriers. Thus, when combined, our results strongly indicate that the identified mutations are causal.
Previously, it has been observed that boys in the Pakistani population are more prone to obesity than girls [
56]. In addition, Iranian consanguineous families have revealed that LEPR deficiency may be more severe in females compared to males [
43]. However, in OB4 where the mutation was found in both affected boys and girls, it showed the same level of severity irrespective of sex.
Increasing knowledge of genetic factors involved in the development of childhood obesity leads towards an improved understanding of the genetic etiology of this disorder. For this purpose the Pakistani population is unique due to its large size, its high number of families with known consanguineous marriages and the high frequency of large pedigrees [
56]. Especially the identification of rare, damaging variants predisposing to obesity holds promise to the future development of novel therapeutic options and personalized medicine based on molecular diagnosis [
24,
57]. In the case of congenital leptin deficiency caused by deleterious
LEP mutations, hormonal leptin therapy has proved to have dramatic treatment effects, successfully decreasing the body weight and hyperphagia of the carriers [
58,
59]. Recently, treatment with mechanism-based therapy using a MC4R agonist (setmelanotide) in two patients with damaging
pro-opiomelanocortin (POMC) mutations completely reversed hyperphagia and induced a remarkable weight loss while normalizing insulin sensitivity [
60]. Albeit no effective drug therapies are currently available for LEPR deficient individuals, treatment of dysfunctional POMC with MC4R-agonist suggests its efficacy in other monogenic defects of the hypothalamic leptin-melanocortin pathway, including
LEPR deficient patients [
60]. Hence, the treatment with setmelanotide might be an effective in treatment of probands with non-functional LEPR, as identified in our study.
Yet, within 23 of the probands included in the present study, no causal variants were identified. For these remaining probands, the application of whole exome sequencing may be an important means to examine whether damaging mutations in other, yet unknown genes, may be the cause of their inherited early-onset obesity.