| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Departments of Pediatrics (B.M.K.) and Internal Medicine (B.M.K., A.R.Z.) and McDermott Center for Human Growth and Development (A.R.Z.), University of Texas Southwestern Medical Center, Dallas, Texas 75390-8591
Address all correspondence and requests for reprints to: Andrew R. Zinn, Department of Internal Medicine, McDermott Center for Human Growth and Development, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-8591. E-mail: andrew.zinn{at}utsouthwestern.edu.
Although the heritability of human body mass index (BMI) is quite high, mutations causing severe obesity have been found with appreciable frequency in only one gene: MC4R, encoding the melanocortin-4 receptor (1). This receptor integrates orexigenic and anorexigenic signals in the hypothalamus and elsewhere in the central nervous system to regulate food intake and energy expenditure. The MC4R gene has only a single exon and thus is amenable to mutation analysis, which is available clinically. In 1998 (2, 3), two groups first reported heterozygous MC4R frameshift mutations associated with autosomal dominant early onset severe obesity. Since then, more than 70 mutations have been reported in various study populations, most with severe obesity (4). Associated findings include hyperphagia, accelerated linear growth in children (although final height appears to be normal), and marked increase in bone mineral density above and beyond that normally seen with obesity (1). Probands with homozygous MC4R mutations show more severe obesity than their heterozygous relatives; thus, the mode of inheritance is codominant (1). So far 57 nonsynonymous missense, five nonsense, and 10 frameshift mutations have been reported with a combined frequency of 2.28% in obese cohorts (5). The prevalence of pathogenic MC4R mutations in these obese populations varied widely, ranging from 0.5 to 5.8% (6, 7, 8, 9).
Proposed explanations for the 10-fold range in frequency of pathogenic MC4R mutations among various cohorts include differences in ethnicity, age of onset, and severity of obesity. The highest prevalence of pathogenic MC4R mutations, 5.8%, was found in a cohort of 500 subjects from the United Kingdom with early onset (before 10 yr of age), severe obesity [mean BMI SD score (SDS), +4.2] (9). A cohort of 209 French adults with even more severe obesity (mean BMI SDS, +7.8), only a third of whom had childhood onset, yielded a prevalence of 4% (8). The prevalence of pathogenic MC4R mutations in a German cohort of 808 children and adolescents with moderate obesity (mean BMI, 32.5 kg/m2) was 1.9% (10). A Danish cohort of 750 adults (mean BMI, 33.3 kg/m2) with childhood onset of obesity showed a prevalence of pathogenic MC4R mutations of 2.5%, with no pathogenic mutations found in a similar number of lean controls (11). An Italian study of 208 children and adolescents with severe early onset obesity (mean age of onset, 4.5 yr; mean BMI SDS, +3.6) found only one patient with a pathogenic MC4R mutation, yielding a prevalence of less than 0.5% (7). A small Japanese study revealed no pathogenic mutations among 50 obese patients with a mean BMI of 38.7 kg/m2 (12).
The aforementioned studies highlight the effects of inclusion criteria on the prevalence of pathogenic MC4R mutations in the cohorts studied. The highest prevalence was seen in obese adults selected for severe childhood-onset obesity, suggesting that severe obesity and especially early age of onset may be characteristic of MC4R mutation carriers. They also suggest that pathogenic MC4R mutations may be more prevalent in northern European populations than in Asian or Mediterranean populations. It is important to note that pathogenic MC4R mutations by and large are "private," with individual mutations having a very low frequency in any population. Collectively, the French, Danish, and UK studies cited included more than 2000 patients selected for obesity, and these studies reported a total of 68 patients with pathogenic MC4R mutations (8, 10, 11). Only one mutation (Y35stop) was found in all three studies. Thus, any statistical comparison of MC4R mutations in human populations requires pooling of individual mutations. Fortunately, in vitro assays have been developed that allow MC4R mutations to be classified on the basis of their effects on receptor signaling (4).
In this issue, two articles (5, 13) address important albeit different questions relating to MC4R mutations and obesity. First, does age of onset of obesity predict which obese adults will be found to carry MC4R mutations? Second, what is the prevalence of MC4R mutations in an adult population not selected for obesity? To answer the first question, Lubrano-Berthelier et al. (13) examined a group of very obese French adults for whom childhood weight and height data were available. They found a similar prevalence of MC4R mutations in subjects with childhood- vs. adult-onset obesity, implying that early age of onset is not characteristic of MC4R mutations. They also found no evidence that MC4R mutations are associated with binge eating, a controversial previous finding (14, 15).
Despite the lack of overall association of MC4R mutations with age of onset of obesity, Lubrano-Berthelier et al. (13) did find that mutations causing intracellular retention of the receptor in vitro were associated with earlier age of onset and greater severity of obesity than other mutations, including those causing complete loss of function. How might this be the case? Mutant receptor molecules retained intracellularly could have dominant negative effects on signaling by the wild-type receptor in heterozygotes. The in vitro assays used to characterize mutations did not test this possibility. Alternatively, the authors have previously suggested that intracellular retention of receptors could cause toxic degeneration of MC4R neurons (16), a hypothesis that is readily testable in transgenic mice. If confirmed, their findings could have important treatment implications for MC4R mutation carriers, e.g. predicting responsiveness to receptor agonists.
The study by Hinney et al. (5) addressed the prevalence of MC4R mutations in the general population. They screened an unselected population of 4068 German adults (mean BMI, 26–27 kg/m2) as well as a second cohort of 1003 obese German adults (mean BMI, 35–36 kg/m2) for MC4R mutations. Several important findings emerged. Like Lubrano-Berthelier et al. (13), they performed in vitro assays of the identified MC4R mutations. Based upon the results of these assays, mutations were grouped for statistical analyses into those that impaired MC4R function, those that did not, and those with an uncertain effect. Four mutations that impaired receptor function were identified in the unselected population, but none of the carriers were obese, raising important questions about the penetrance of MC4R mutations and our ability to predict in vivo effects from in vitro data. Familial studies of MC4R mutations have also shown incomplete penetrance (17), and it is likely that penetrance estimates based on high risk families (9) are exaggerated.
The prevalence of pathogenic MC4R mutations in the cohort of moderately obese German adults was only 0.2% (5), much lower than the prevalence of 1.86% found by the same workers in extremely obese German children and adolescents (10). This discrepancy could be due to the difference in severity of obesity in the populations studied. Alternatively, the contribution of pathogenic MC4R mutations to obesity may be diluted by additional genetic and environmental factors leading to obesity in adults.
These studies demonstrate several difficulties in this field. Mutations that clearly impair MC4R function are found with significant frequency in only the most severely obese cohorts. Although age of onset is an important factor in selection of patients to be tested for MC4R mutations, it is not the only factor. Other clinical clues to the diagnosis of MC4R mutations include accelerated linear growth and a positive family history (18). These studies also point to our limited ability to predict phenotypes from in vitro assays. The extremely low prevalence of pathogenic MC4R mutations in the general population underscores the fact that the obesity epidemic is not an epidemic of new mutations, although it is conceivable that epigenetic modifications may play a role (19). The fact that four large studies with more than 3000 obese patients shared only one pathogenic MC4R mutation underscores the rarity of individual mutations and the need for very large multicenter studies of highly selected populations and extremely large prospective, population-based cohort studies (on the order of 200,000 subjects; Ref.20) to understand the spectrum of human MC4R mutations and associated phenotypes. The present studies also suggest that the yield of MC4R testing in clinical diagnosis of obesity is likely to be low except in highly selected patients.
Footnotes
This work was supported by the American Diabetes Association and National Institutes of Health Grant P20 RR020691.
Abbreviations: BMI, Body mass index; SDS, SD score.
Received March 10, 2006.
Accepted March 14, 2006.
References
-melanocyte-stimulating hormone genes in severely obese children. J Pediatr 139:204–209[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
