Acknowledgements

References

Copyright

Obesity is rapidly reaching epidemic proportions in many populations, including most Western societies. Through related disorders, which include cardiovascular disease, non-insulin-dependent diabetes mellitus, hypertension, stroke and certain forms of cancer, obesity can increase the mortality rates by 35–1000% for weights of between 130% and greater than 200% of expected values. Obesity clearly has a strong genetic basis, with estimates of heritability ranging from 30% to 70%.

Certain rare mutations have been identified that account for a small minority of extreme obese phenotypes ¹. However, multiple genes with modest effect on components of energy balance, in combination with environmental factors, probably account for the normal distribution of body-fat proportion in nearly all human populations. Although tremendous progress has recently been made in understanding the biochemical, physiological and endocrine bases of obesity, the underlying genetic nature of obesity in humans is still unknown.

There are literally hundreds of animal models for obesity, including dozens of vertebrate species. Those with the greatest impact on our understanding of the genetics and physiology of human obesity are the prolific and variable rodent models. Despite a general lack of similarity to humans in many aspects of nutritional physiology, rodents are, in general, extremely valuable experimental models that have human-like primary metabolic pathways and can be genetically manipulated in many ways. In particular, two important categories of mouse models will be described: polygenic (quantitative) variation ( Fig. 1a) and monogenic (spontaneous and targeted mutations) variation ( Fig. 1b).

Figure 1. Two Mouse Models of Obesity: (a) Polygenic model, showing two long-term selection lines. These lines differ significantly in body-fat proportion, owing to the combined action of a large number of quantitative trait loci (QTLs). Bottom: L6, selected for low 6-week body weight. Top: M16, selected for rapid 3–6-week body weight gain. Lines were selected and are maintained by Dr Gene Eisen, North Carolina State University, Raleigh, NC, USA. Reproduced with permission from Gene Eisen. (b) Monogenic model. These mice differ dramatically in obesity owing to the direct action of a single gene. Left: mutant mouse (Leprdb-3J/Leprdb-3J) lacking functional receptors for the hormone leptin, caused by a spontaneous mutation in the gene coding for leptin receptor. Right: wild-type control mouse. Reproduced with permission from The Jackson Laboratory, Bar Harbor, ME, USA.

The vast spectrum of existing genetic variation, coupled with a short generation time and low husbandry costs, make mice particularly well suited to studying the quantitative genetics of obesity. Rodent strains undergoing long-term directed selection for obesity-related traits have been particularly useful as a model to characterize the complex genetic basis of obesity in humans, establishing a strong heritable component for body-fat proportion and quantifying the genetic and phenotypic correlations between obesity and traits involved in energy balance ².

The genomes of many animal models are now routinely screened for the individual genes or quantitative trait loci (QTLs) that influence body composition at the population level ³. The QTL approach has been very useful for understanding the quantitative genetics of obesity, including interactions with diet, gender, age and genetic background, and has yielded over 50 chromosomal regions harboring loci with a potential influence on obesity ^1,3 . Such approaches are also beginning to be fruitful in human populations but are limited by high expense and inherently low statistical power. Although the rodent models promise to provide insight into the polygenic predisposition to obesity, no obesity QTLs have yet been cloned and so the physiological relevance of these loci has not been established. In addition, it remains to be seen whether or not rodent and human obesity QTLs are homologous in identity and function. The burgeoning tools of genomics and comparative mapping should assist these endeavors in the future.

By contrast, the characterization of spontaneous or targeted gene mutations in mice has led to a great advance in our understanding of the molecular physiology of weight regulation in humans. The cloning of the genes responsible for the mouse mutant models ob and db has led to the discovery of the hormone leptin and its receptor, arguably the most important components in the complex physiological system regulating energy homeostasis and obesity in mammals ^4,5 . Additional successes include the cloning and characterization of other known mouse mutations that lead to obesity, such as the agouti signaling protein (Asp), carboxypeptidase E (fat), and tubby (tub) ( Table 1). Most recently, the obesity-suppressing gene mahogany (mg) has been identified.

Table 1. Monogenic mouse models with major effects on obesity

Mutation     Gene     Product     Function     Human cases a     Obese (ob)     Lep     Leptin     Regulates energy intake and expenditure     5                                   Diabetes (db)     Lepr     Leptin receptor     Receptor for leptin, mediating effects     3                                   Fat (fat)     Cpe     Carboxypeptidase E     Prohormone processing     0 b                                   Tubby (tub)     Tub     Unknown     Unknown     0                                   Agouti (Ay)     Asip     Agouti signaling protein     Antagonism of Mc4r?     0                                   Mahogany (Mg)     Mg     Mahogany protein     Suppresses diet-induced obesity     0     Mc4R (knockout)     Mc4r     Melanocortin 4 receptor     Receptor for melanocortin     9

[a]Data from Ref. 1[b]One case of obesity in humans has been reported with a mutation in the PCSK1 gene, which is also involved in prohormone processing.

The ability to cause targeted genetic defects in mice using homologous recombination in embryonic stem cells has expanded the use of mutational analysis to almost any gene of interest. This has revealed the roles of numerous genes in the etiology of obesity, including the melanocortin-4 receptor and protein-tyrosine-phosphatase 1B, among many others. The possibility of treating obesity by pharmaceutical intervention will benefit from these efforts, as gene products could be tested as therapeutic agents in the treatment of excessive fat deposition.

An interesting paradox emerges when trying to correlate results from QTLs and mutational approaches. Even though proteins such as leptin, leptin receptor and melanocortin-4 receptor clearly play a major role in the regulation of body composition, mutations within these genes have only rarely been identified in a small number of obese humans ¹ ( Table 1). Furthermore, extensive analysis in mice has not resulted in repeated identification of obesity QTLs in chromosomal regions harboring these (and many other) major obesity genes. Thus, it is unlikely that allelic variants within these loci contribute greatly to quantitative variations in body composition within or between populations. This is where the QTL and mutational approaches will synergistically intersect in the future. It can be hypothesized that QTLs, forming the heritable component, and thus the genes leading to genetic predisposition to obesity, might represent regulatory loci that lead to subtle changes in the expression of the primary genes whose protein products directly determine obesity phenotypes. Animal models will be critical in evaluating this hypothesis and continuing to expand our knowledge of the genetics and molecular physiology of obesity in humans.

This article is published as paper number 12640 of the Journal Series, Nebraska Agricultural Research Division.

[1] Perusse L., Chagnon Y.C., Weisnagel J. and Bouchard C. (1999) The human obesity gene map: the 1998 update.
Obesity Res., 7:111-129.

[2] Pomp D. and Nielsen M.K. (1999) Quantitative genetics of energy balance: lessons from animal models.
Obesity Res., 7:106-110.

[3] Pomp D. (1997) Genetic dissection of obesity in polygenic animal models.
Behav. Genet., 27:285-306. MEDLINE Cited by

[4] Friedman J.M. and Halaas J.L. (1998) Leptin and the regulation of body weight in mammals.
Nature, 395:763-770. Full text MEDLINE Cited by

[5] Leibel R.L., Chung W.K. and Chua S.C. Jr. (1997) The molecular genetics of rodent single gene obesities.
J. Biol. Chem., 272:31937-31940. Full text MEDLINE Cited by

© 1999 Elsevier Science Ltd. All rights reserved.