Mutational Screening of the CART Gene in Obese Children: Identifying a Mutation (Leu34Phe) Associated With Reduced Resting Energy Expenditure and Cosegregating With Obesity Phenotype in a Large Family

Food intake is regulated via neural circuits located in the hypothalamus. Over the last few years we have gained a good understanding of the specific mediators and neuronal networks that regulate food intake and body weight (1,2). Cocaine- and amphetamine-regulated transcript (CART) was originally described as an mRNA induced in the brain after acute administration of cocaine to rats; it has recently been shown that intracerebroventricular administration of this neuropeptide inhibits feeding and induces the expression of c-Fos in several nuclei involved in the regulation of food intake (3). Furthermore, an antiserum against CART increases feeding in normal rats, suggesting that CART may be an endogenous inhibitor of food intake in normal animals (4). This action is modulated by leptin (4,5). In obese animal models with disrupted leptin signaling, CART mRNA is almost absent from the arcuate nucleus (4). Peripheral administration of leptin to obese mice stimulates CART mRNA expression (4). Moreover, in the lateral arcuate nucleus, there is a population of leptin-activated cells containing CART and directly innervating sympathetic preganglionic neurons in the thoracic spinal cord (5). These neurons regulate interscapular brown adipose tissue, which plays an integral role in regulating body temperature, energy expenditure, and diet-induced thermogenesis (5). Because leptin mediates many of its physiological effects by increasing the activity of the sympathetic nervous system, it is conceivable that the engagement of this novel pathway may contribute to the increased thermogenesis and energy expenditure characteristic of leptin action. The detection of CART-like immunoreactivity in sympathetic ganglia and adrenal glands of the rat has further supported the idea that the CART peptide(s) functions as a signaling molecule in the sympathetic nervous system (6,7,8). The aim of this study was to investigate whether variation in the CART gene may contribute to the variation of the clinical phenotype in Italian children and adolescents with early onset obesity.

We examined 130 (72 girls) unrelated obese Italian children and adoloscents aged (mean ± SD [range]) 10.7 ± 3.1 (2–16) years. The mean of their Z-score relative to BMI percentiles (9) was 3.9 ± 1.8 (2.1–8.8), and the average age at obesity onset was 4.7 ± 2.6 (1–11) years.

The CART gene was screened by single-strand conformation polymorphism (SSCP). Automatic sequencing of PCR products showing an aberrant pattern of migration was performed. The amino acid numbering is in agreement with Kuhar and Dall Vechia (10), and nucleotide positions are numbered according to Douglas and Daoud (11). Two previously described silent polymorphisms were found in the 3′ untranslated region (UTR): a deletion of A at position 1457 in 9 patients (allele frequency 0.035) and the change of A with G at position 1475 in 11 patients (allele frequency 0.042). A similar allelic frequency (0.041 for the mutation at position 1457 and 0.039 for the mutation at position 1475) was found by screening 100 unrelated lean control subjects.

A missense mutation of G729C, TTG̅→TTC̅, resulting in the substitution of Leu with Phe at the codon 34 (Fig. 1), within the NH₂-terminal CART region has been detected in the heterozygous state in a 10-year-old obese boy who has been obese since the age of 2 years (Fig. 2). His birth weight was normal. His BMI at age 10 years was 32 kg/m² (weight 60 kg, height 142 cm, Z-score 8.3). Serum leptin concentration (23 ng/ml) was adequate for BMI, sex, and pubertal development (12) (leptin Z-score 0.5). The patient belongs to a large family of obese subjects (Fig. 2). His mother and many components of the maternal kindred had a very similar history of early onset obesity. His father, on the contrary, was not obese. To test for the inheritance of this missense mutation in the maternal proband’s family, exon 2 of the CART gene was sequenced. The mutation cosegregated with the severe obesity phenotype over three generations. Subjects II-2, -4, and -6 have type 2 diabetes. The mutation was not found in the control population. Comparing resting metabolic rates with those expected after adjusting energy expenditure for age, sex, weight, height, and an age-adjusted value for urinary nitrogen, we found a difference of −230 Kcal/24 h (−14%) and −240 Kcal/24 h (−16%) for the patients III-2 and IV-1. We also found differences of +50 Kcal/24 h (+4%), −40 Kcal/24 h (−4%), and +50 Kcal/24 h (+6%) for the normal subjects IV-2, IV-3, and IV-4, respectively.

Although the CART gene has recently been localized in the 5q13–14 locus (13), a known human obesity locus (14), three mutational screenings performed in Danes with early onset obesity (13), in a group of obese Pima Indians (15), and in 91 severely obese Caucasian children (16) failed to detect genetic variations clearly contributing to the obesity phenotype. The only mutation found in the coding region, Ser39Thr, did not cosegregate with obesity in family studies (16). Three polymorphisms in the 3′ UTR region have been identified (A1475G, 1457delA, and, among Pima Indians, C1442G). The prevalence of these variants was determined among obese and control subjects, but no associations between these alleles and variations of BMI were demonstrated. However, it is noteworthy that in one study, the A1475G variant has been associated with a lower waist-to-hip ratio in male A/G heterozygotes (16). When comparing the age of obesity onset, BMI Z-scores, and leptin levels adjusted for BMI, sex, and pubertal stage (leptin Z-scores), we found no difference between the obese patients heterozygous for one of these polymorphisms and those homozygous for the wild allele (Table 1), further eliminating the possibility that these mutations are a significant source of obesity variation.

CART is a neuropeptide that is secreted and processed. This idea is based on the identification of a hydrophobic leader sequence at the NH₂-terminus, which suggests involvement in a secretion pathway, and on the presence of several pairs of basic amino acids commonly found in propeptides that are processed before their subsequent use (10,11). In the rat, the CART precursor has been found to be processed differently in central and peripheral sites. The CART long-form peptides (1–89 and 10–89) were isolated from the adrenal gland, and they most likely represent the CART peptides present in the adrenal medulla, in contrast to the hypothalamus, from which the shorter-form peptides (42–89 and 49–89) were purified (17,18). Although the involvement of central CART in the regulation of energy expenditure is possible (CART is also found in the hypothalamic neurons innervating the sympathetic preganglionic neurons), this tissue-specific processing suggests that different CART peptides may have different biological functions in the periphery (energy expenditure increase) and in the central nervous system (food intake inhibition). However, because of the limitations of the techniques used in these studies (i.e., antibody against the COOH-terminal part of the CART precursor), it has not been possible to determine whether further processing of the NH₂-terminal part of CART (1–39) takes place in the periphery. Leucine at codon 34, conserved in this position in the human and rat sequences, immediately precedes a couple of lysine residues that may well represent a dibasic processing site. The substitution of a phenylalanine for the wild-type leucine in this position might alter the susceptibility to proteolysis of a potential processing site. Alternatively, this mutation might produce a perturbation of the protein’s three-dimensional structure large enough to increase the sensitivity to proteolysis; this increase in sensitivity might even occur at sites distant from the mutation itself, also affecting, in this case, the activity of central CART peptides.

In agreement with results from resting metabolic rates in the proband and his mother, we suggest that the Leu34Phe mutation may play a role in contributing to the obese phenotype by altering the CART mediated leptin effect on thermogenesis and energy expenditure.