rodent obesity models marked the initiation of the molecular genetics of obesity.
Since then autosomal recessive mutations in the genes for leptin [32–34], leptin
receptor [35], prohormone convertase 1 (PC1; 36) and pro-opiomelanocortin
(POMC) [37] have been shown to lead to early-onset obesity in humans. Not
surprisingly, inbreeding was documented in the affected family members with
leptin and leptin receptor gene mutations [32, 33, 35]. Compound heterozygosity
has been detected for PC1 and POMC mutations [36, 37]. All of the
mutations apparently lead to additional phenotypical manifestations including
adrenal insufficiency (POMC), red hair (POMC), reduced or impaired fertility
(PC1, leptin and leptin receptor) and impaired immunity (leptin gene). Whereas
they do not affect intelligence, the pleiotropic effects warrant the consideration
that these recessive disorders be classified as syndromal forms of obesity. All
of the mutations lead to early-onset extreme obesity mostly or totally induced
by an increased energy intake; a reduced energy expenditure as a contributing
pathogenetic factor has not been documented. Finally, all mutations are
evidently exceedingly rare; mutation screenings in the coding and promotor
regions of the respective genes have not revealed other mutations which can
functionally readily be linked to obesity [4]. However, linkage and/or association
have been reported to markers surrounding or within the leptin gene locus
on 7q31 and the leptin receptor locus on 1p32 [4]. In addition, linkage of serum
leptin levels and fat mass has been described upon use of markers that localize
to the POMC region on chromosome 2q and variation of leptin levels has been
associated with POMC polymorphisms [4].
The first autosomal-dominant form of human obesity due to a missense
mutation in the gene coding for the peroxisome-proliferator-activated receptor
gamma2 (PPAR 2) was discovered in 4 of 121 obese unrelated adult Germans,
but not in any of the 237 normal weight controls [38]. Analysis of the mutation
by retroviral transfection and overexpression in murine fibroblasts revealed
functional deficiencies. The same mutation has not been detected in any further
obese individuals including obese German children, adolescents and adults.
The most recent advance has been the identification of functionally relevant
mutations in the melanocortin-4 receptor gene [5, 6, 39–43] which result
in a co-dominantly inherited form of obesity. In Germany, 2.5% out of over 800
more or less extremely obese children and adolescents have recently been
shown to harbor such mutations [6], which encompass frameshift, nonsense and
missense mutations. Worldwide over 30 different mutations have been identified
[6], only single of these have been shown to occur more than once; rare
compound heterozygous or homozygous cases are even more obese than heterozygous
carriers [5, 6]. Pharmacological assays have revealed that these
mutations either lead to a total or partial loss of function [5, 6, 43].
Based on family based BMI comparisons between mutation carriers and
wildtype carriers we estimate that adult male mutation carriers are 15 to 20kg
heavier than their male wild-type relatives [Dempfle et al., unpubl. data]; estimates
for children are not available. The effect size of these mutations is lower
than that of the leptin or leptin receptor gene mutations. In accordance with this
observation, single non-obese MC4R mutation carriers have been identified
[42, 43]. Potentially, such non-obese carriers might have additionally inherited
an allele(s) protecting them from developing obesity. The weight curves of
heterozygous MC4R knockout mice also clearly extend into the range formed
by wild-type animals [44]. Similar to results obtained in Mc4r knockout mice
[44] female carriers in our families are more obese than male carriers [Dempfle
et al., unpubl. data].
Phenotypical effects of MC4R mutation carriers other than obesity have
been shown to encompass an elevated growth rate and a higher bone density
[5]. An interesting question is whether or not MC4R mutations are associated
with aberrant eating behavior and in particular binge eating episodes [45].
Recently, 100% (n 20) of obese carriers of MC4R variants were shown to
have binge eating disorder [46]. However, this study mostly included individuals
with the Val-103-Ile polymorphism, which in all other studies has not been
found to be associated with obesity; carrier frequencies in the range of 2–4%
have repeatedly been demonstrated in both cases and lean controls.
Furthermore, we did not detect any MC4R mutation among extremely obese
adults with binge eating disorder [47]. Finally, in our own families we found
no evidence for elevated rates of binge eating behavior in both carriers of
mutations and the Val-103-Ile polymorphism [Hebebrand et al., unpubl. data].
In conclusion, we do not agree with Branson et al. [46] that binge eating
disorder is strongly associated with MC4R mutations or polymorphisms.
Based on observations in Mc4r / mice an elevated food intake seems to
underlie the development of obesity [44]. These mice are particularly susceptible
to dietary fat [48]. It seems unlikely that a reduced energy expenditure also
contributes to obesity. Children with MC4R mutations have been shown to eat
more at a test meal than controls [5].
A large number of association studies have been performed in ‘normal’
obesity [2]. Whereas many associations have been reported, it is largely unclear,
which of these represent true-positive findings. Presumably, the power of many
of these studies is too low to detect minor genes; false-positive findings can
result from not correcting for multiple testing [49].
Over 20 genome linkage scans pertaining to obesity and related phenotypes
have been performed; specific chromosomal peak regions have repeatedly
been identified in different scans [2, 49]. For example, linkage to
chromosome 10p originally detected in French families with two or more adult
obese offspring [50] was confirmed in a German study [51]. In the first genome
linkage scan for childhood and adolescent obesity we found suggestive linkage
to some of the regions which had previously been identified in adult scans [52].
It appears a matter of time before the genes within these linkage regions are
identified.
The identification of genetic variation underlying obesity should lead
clinicians and the general public to more readily consider genetic factors in the
multifactorial etiology of obesity. We hope that this will entail less stigmatization
of obese children and their families. The fact that 2–3% of extremely obese
children harbor MC4R mutations currently raises the question as to the inclusion
of a systematic mutation screen via sequencing in the diagnostic work-up
of obese children. This and future developments require a consensus as to who
should be screened, for what purpose and who should finance these potentially
expensive diagnostic molecular procedures. In addition, we need to study if and
how knowledge of a genetic cause of obesity influences the course of obesity
particularly in children.
The Deutsche Forschungsgemeinschaft, the Bundesministerium für Bildung und
Forschung and the European Union, Framework V (‘Diet and obesity’; QLKT-CT-2000–00515)
generously supported our research efforts. We thank the families that have contributed to our
genetic studies.
0 komentar:
Posting Komentar