The Journal of Nutrition Vol. 127 No. 9 September 1997, pp. 1909S-1916S
Copyright ©1997 by the American Society for Nutritional Sciences

Mapping of Mouse Obesity Genes: A Generic Approach to a Complex Trait¹

Janis S. Fisler^{*, 2} and Craig H. Warden

^* Department of Medicine, Division of Cardiology, University of California, Los Angeles, CA 90095 and  Department of Pediatrics and Rowe Program in Genetics, University of California, Davis, CA 95616

ABSTRACT

CLASSICAL STRATEGIES FOR IDENTIFICATION OF GENES

APPLICATIONS OF QTL MAPPING TO OBESITY

FOOTNOTES

LITERATURE CITED

ABSTRACT

Identification of genes underlying any complex trait such as obesity is an important and difficult problem in genetics. Traditional candidate gene approaches cannot be relied on to identify all of the genes influencing a complex trait, and positional cloning is very laborious. With the advent of new tools and methods, however, comprehensive approaches to the identification of any genes underlying complex traits are now available. Quantitative trait locus (QTL) mapping is a general technique to map Mendelian factors influencing complex traits. The QTL approach involves the crossing of two strains that differ in the trait of interest to produce F2 or back-cross progeny, individually phenotyping and genotyping each progeny, and statistically associating the typed markers and the phenotype. QTL mapping has been used in the last 4 years to map genes for a wide variety of traits, including body weight and growth, obesity, atherosclerosis and susceptibility to cancer in the mouse, and hypertension, hyperactivity and arthritis in the rat. QTL mapping has also been used to map genes in pigs, poultry, cows, fish and plants. Once a trait has been located in a chromosomal subregion, identifying the underlying gene remains a significant problem. A monogenic model must be developed, isolating one gene influencing a trait from other genes affecting the same phenotype. Then the positional candidate strategy, which relies on a combination of mapping to a chromosomal subregion followed by a survey of the interval to see if attractive candidates reside there, becomes practical.

CLASSICAL STRATEGIES FOR IDENTIFICATION OF GENES

Alternatively, a gene causing obesity can be isolated without using any functional knowledge by positional cloning based on linkage information from human families or animal models, as has been accomplished for many genes, including the Lep (Zhang et al. 1994), tub (Kleyn et al. 1996, Noben-Trauth et al. 1996) and agouti (Bultman et al. 1992, Miller et al. 1993) genes causing monogenic obesity in mice. The remaining two genes causing obesity in mice were also recently identified. The cloning of Lep allowed the expression cloning of its receptor, Lepr (Tartaglia et al. 1995), whereas Cpe^fat was identified by a positional candidate approach. As with the over- and underexpression studies mentioned above, the cloning of these mouse obesity genes identified pathways that were previously unknown. Positional cloning may be possible for genes producing rare monogenic syndromes in humans that include obesity as a phenotype, but little progress has been made.

Difficulties with direct positional cloning of obesity genes may be overcome by cloning of genes in animal models, followed by studies of that gene in humans (Friedman et al. 1991). Positional cloning, however, requires that the model be monogenic and that the chromosomal location of the gene be precisely known. The identification of multiple genes that contribute to a complex phenotype requires strategies differing from classical single-gene/single-protein studies (Frankel 1995). Whole-genome searches in multigenic animal models can be conducted without any previous knowledge of the underlying genes and can identify chromosomal regions controlling a complex trait. Whole-genome searches thus focus attention on specific chromosomal regions that can be isolated and narrowed in congenic strains, then searched for candidate genes or subjected to positional cloning strategies.

Compared with studies of humans, genetic studies in animals are faster and often allow a more detailed and accurate description of phenotype (Lander and Schork 1994, Warden and Fisler 1994). Mice will often be the animal model of choice for genetic studies because of the availability of hundreds of inbred and congenic strains as well as a dense genetic map and physical mapping tools. Moreover, the homologous regions of mouse and human chromosomes are so well defined that it is frequently possible to identify the chromosomal location of a gene in humans by mapping it in mice, often with more precision than is provided by human mapping studies (Copeland et al. 1993). Because of their larger size, rats are used more extensively than mice for physiologic and biochemical studies of energy homeostasis. A number of well-characterized strains are available (Schemmel et al. 1970, Schork et al. 1995); as a result of the rapidly expanding rat genetic map, these will be very useful in identifying genes regulating energy balance. Other critical advantages of studies of mice and rats include the ability to manipulate diet and to obtain any tissue at various times during development. Thus, many studies are more practical in animals than in humans, including studies of diet-responsiveness, longitudinal studies and studies of critical periods in the development of obesity.

The four steps of QTL mapping, which are illustrated in Figure 1, are as follows: 1 ) two different inbred strains are first crossed to produce F2 or back-cross progeny; 2 ) phenotypes are assayed for all of the progeny; 3 ) all of the progeny are individually genotyped for markers that are spread throughout the genome; and 4 ) statistical associations of markers and phenotypes are performed to identify loci underlying the traits.

Inbred strains, produced by 20 or more generations of brother × sister matings, provide the basic resource for analysis of both monogenic and multigenic traits. The genes passed on by any one animal of an inbred strain will be identical to those of any other animal of that strain, thus providing an invariant gene pool that can be used for all subsequent phenotyping and genotyping experiments. Ideally, the mean values of the trait under consideration of the parental strains should differ by at least two standard deviations. However, different strains may coincidentally have similar phenotypes, while exhibiting substantial genetic differences. In crosses involving such parental strains, the phenotypes of the progeny are expected to span a much wider range of values than do the phenotypes of the parents. One such instance of this phenomenon, the obesity of BSB mice, is discussed below. Once the model is chosen, a cross is set up between the two genetically divergent strains. The resulting F1 mice are then either crossed with each other to produce F2 progeny or mated to one of the parental strains to produce back-cross progeny, and the resulting progeny are used for the phenotyping and genotyping studies.

The methods used for phenotyping should be the most specific and precise measurements available. For example, the percentage of body fat is a better measure of obesity than body weight or body mass index. Particular care should be taken to minimize the error of the assay and environmental differences, because nongenetic variation will reduce the power of the method (Paterson et al. 1991). QTL mapping can be applied to any continuously varying trait, such as the percentage of body fat, oxygen consumption and weight gain induced by early overfeeding, or to yes/no traits by using a ² analysis.

Genotyping identifies which of the two parental strains (1 or 2 in Fig. 1) has contributed alleles at a specific locus in each F2 or back-cross animal. The two most common methods for genotyping are polymerase chain reaction (PCR) assay of simple sequence repeats (SSR) and restriction fragment length polymorphisms (RFLP) of cDNA or genomic clones. The most common SSR consists of a variable number of cytosine-adenine repeats (CA)n. It is present at ~100,000 copies per genome and is so polymorphic that any two inbred mouse strains will, on average, be polymorphic at 50 % of existing SSR loci (Dietrich et al. 1992). Genotyping with SSR primers is faster and cheaper than RFLP and is generally the method of choice. Genotyping by RFLP remains useful when testing a specific candidate gene for association with a trait.

A linear regression of phenotype on genotype, such as an ANOVA, is the classical method for determining if the mean phenotype values of progeny with different genotypes at a specific marker are significantly different. If Strains 1 and 2 denote inbred strains (Fig. 1) and we assume a back cross with Strain 2 as the recurrent parent, then there would be two possible genotypes at each marker in the back-cross progeny, aa' and a'a'. In F2 mice, there would be three possible genotypes, aa, aa' and a'a'. Finding a significant phenotypic difference between animals with different genotypes suggests that a gene to which the marker is tightly linked controls a measurably detectable difference in the phenotype. This approach is limited by both the number and spacing of the individual markers genotyped, because linear regression can be used only to detect association at each genotyped marker.

QTL analysis requires the construction of genetic linkage maps of ordered markers that can be constructed from primary data using the MAPMAKER/QTL (Lander and Botstein 1989, Lander and Schork 1994, Lander et al. 1987, Paterson et al. 1988) or MapManager/QT software. These genetic linkage maps are used with quantitative phenotype data to locate QTL by an efficient interval mapping method that calculates log-of-the-odds (LOD) scores for each quantitative trait at 1- to 2-cM intervals between each marker. LOD scores provide a measure of the significance of association of a trait and genotype as the log ₁₀ of the likelihood of the odds ratio (LOD). LOD scores of 3.3 and 4.3 or greater are considered statistically significant evidence of association in back-cross and intercross mice, respectively (Lander and Kruglyak 1995). Statistical models for the analysis of multiple QTL (Jansen 1994, Jansen and Stam 1994) and multistep analyses (Schork et al. 1995) have been developed.

APPLICATIONS OF QTL MAPPING TO OBESITY

It is difficult to determine the number of new obesity genes identified by these few studies mapping obesity loci. Several of these QTL are in the same chromosomal regions and may be the same gene. It is likely that the chromosome 15 locus found in the SWR/J × AKR/J, CAST/Ei × C57BL/6J, and C57BL/6J × Mus spretus (BSB) crosses are identical. The BSB mice were maintained on a standard laboratory diet, whereas the SWR/J × AKR/J and CAST/Ei × C57BL/6J crosses were maintained on a high fat, high sucrose "obesifying" diet. It is also likely that the loci identified on distal chromosome 2 in the standard diet-fed M16i × CAST cross and the high fat, high cholesterol diet fed NZB/BINJ × SM/J cross are identical. Thus, the chromosome 15 and chromosome 2 obesity loci do not appear to be influenced by diet composition. Loci on chromosomes 4 and 9 were identified only in mice fed a diet defined as "obesifying" and could reflect diet-responsive obesity genes.

Because variations in environment, including variations in diet, across the study subjects will increase the variance due to nongenetic factors, it is important to reduce such variation (Paterson et al. 1991). Certainly, standard laboratory diets from different suppliers and even from different lots from one supplier vary in composition, particularly in type of fat used in diet formulation (Lardinois et al. 1989). It is, therefore, preferable that defined diets be used in QTL studies.

A number of these QTL for body fat are near the mapped positions for the single-gene mutations causing obesity. The chromosome 6 locus that was identified in BSB mice is very near the Lep gene (Warden et al. 1993 and 1995) at which the Lep^ob mutation causes massive obesity in mice (Zhang et al. 1994); there is preliminary evidence that Lep alleles influence fat pad weight in this cross (Warden et al. 1996). Similarly, the chromosome 7 locus in BSB mice includes the tub gene, which causes maturity-onset obesity in mice (Kleyn et al. 1996, Noben-Trauth et al. 1996). The chromosome 4 locus identified in the SWR/J × AKR/J cross includes the Lepr gene (West et al. 1994b), which codes for the leptin receptor (Tartaglia et al. 1995); our laboratory has shown that AKR/J mice fed a high fat diet are resistant to exogenous leptin (Lembertas et al. 1996a and 1996b). Finally, the locus we identified on distal chromosome 2 in the NZB/BINJ × Sm/J cross (Lembertas et al. 1997) contains the agouti gene whose mutations (A^y, A^vy) result in the obesity of the yellow (agouti) mouse (Bultman et al. 1992, Miller et al. 1993). It thus seems likely that normal genetic variation in these genes contributes to multigenic obesity.

QTL mapping has also been applied to crosses involving lines of mice that have been divergently selected for body weight or fat content or body weight gain. F2 populations of crosses selected for low × high body weight lines and low × high body fat lines revealed a QTL on the proximal X chromosome for body weight but not for body fat (Hastings and Veerkamp 1993, Rance et al. 1994). Still other studies have identified QTL for growth or body weight in mice (Cheverud et al. 1996, Collins et al. 1993, Dragani et al. 1995, Keightley et al. 1996) and pigs (Andersson et al. 1994, Clamp et al. 1972).

BSB mice were examined for measures of obesity including percentage of body fat, body mass index, weight of fat pads and for biochemical parameters related to obesity, including glucose, insulin, corticosterone, glycerol, plasma triglycerides, nonesterified fatty acids and total and HDL cholesterol. Unlike the single-gene recessive obesities, there was an increase in fat-free tissue with increasing carcass fat (Fisler et al. 1993). The most notable biochemical variable was the lack of response of corticosterone to ACTH stimulation in obese BSB mice (Fisler et al. 1993, Pace et al. 1993).

BSB mice (n = 412) were typed for 148 markers covering all chromosomes except the Y chromosome (all Y chromosomes in BSB mice are from the C57BL/6J parent). Figure 2 shows the markers typed on chromsomes 6, 7, 12 and 15 in the BSB mice (Warden et al. 1995). We searched for genetic loci underlying the obesity phenotypes that were measured in BSB mice by using the MAPMAKER/QTL program (Paterson et al. 1988), which calculates the strength of associations between genotypes and phenotypes as the log₁₀ of the likelihood of the odds ratio (LOD) score (Fig. 3). It has been calculated that, in a mouse back cross, a LOD score of 3.3 is the threshold for statistically significant linkage (Lander and Kruglyak 1995).

To date, we have identified four chromosomal loci that contribute to obesity in BSB mice: obesity results from the heterozygous SB genotype at three loci and from the homozygous BB genotype at one locus. These four loci explain <50% of the variation in percentage of body fat in BSB mice, suggesting that additional genes, perhaps with effects too small to detect in this experiment, contribute to obesity in this model. It is also possible that interactions among the QTL contributed to phenotypic variance (Paterson et al. 1991). We observed that the markers D7Mit8 and D7Ucla1 on distal mouse chromosome 7 were significantly linked with percentage of body fat (peak LOD score 4.2), total plasma cholesterol (LOD score 4.4) and hepatic lipase (HL) activity (LOD score 5.1) (Fig. 3, 4) (Warden et al. 1993 and 1995). The chromosome 7 locus explained 6.5% of the variance in the percentage of body fat. Mice that were heterozygous for Mus spretus and C57BL/6J alleles at the D7Mit8 marker had 28% more body fat than mice that were homozygous for C57BL/6J alleles at that locus (P < 0.0003). A locus on mouse chromosome 6 near the marker D6Mit1, which lies very near the Lep gene, exhibited a LOD score of 4.8, explaining 7.1% of the variance in femoral fat pad weight, and a LOD score of 3.4 for total plasma cholesterol (Fig. 4) (Warden et al. 1993 and 1995). Mice that were heterozygous at the D6Mit1 locus had ~twofold heavier femoral fat pads than mice that were homozygous C57BL/6J at that locus (P < 0.0001). A third locus was found on mouse chromosome 12 near the marker D12Mit27. It exhibited a LOD score of 4.8 and explained 7% of the variance in percentage of fat (Fig. 4). Mice that were heterozygous at the D12Mit27 locus had 44% greater body fat than mice homozygous for C57BL/6J alleles at that locus (P < 0.0002). A fourth locus was identified in BSB mice on mouse chromosome 15 near D15Mit13 (LOD score 3.4) explaining 5.9% of the variance in mesenteric fat. Mice heterozygous at this locus had 32% smaller mesenteric fat pads than mice that were homozygous for C57BL/6J alleles (P < 0.0001). Each of the four loci identified in BSB mice does not appear to act in an additive manner with the other three loci. Mus spretus alleles promote obesity at the chromosome 6, 7, and 12 loci, whereas C57BL/6J alleles promote obesity at the chromosome 15 locus. Furthermore, the chromosome 15 locus appears to promote obesity only if the homozygous C57BL/6J genotype occurs on a background where the chromosome 6, 7 and 12 loci are all heterozygotes (Fig. 5).

We reasoned that the genetic variations influencing obesity in BSB mice may be relatively common because many inbred strains of laboratory mice vary in body fat content (West et al. 1992). Chromosome 7 is the site of several histocompatibility genes that have been isolated in several congenic strains with two different donor strains, 129/SnJ and BALB/cByJ, on two different background strains, C57BL/6ByJ and C57BL/10SnJ. Two of these congenic strains, B10.129(5M)/nSn and B10.C(41N)/Sn, contain a region of chromosome 7 that includes the QTL identified in BSB mice. The two congenic strains differ from the background C57BL/10SnJ strain in several parameters related to obesity (Fig. 7A). Compared with strain C57BL/10SnJ mice, the two congenic strains exhibited only ~25% of the retroperitoneal fat pad weight and ~60 % of the body fat. Total body weight and plasma HL activity were decreased, and total cholesterol was increased in the B10.129(5M)/nSn congenic strain. Thus, the characteristics of the locus isolated in these congenic strains are strikingly similar to those of the chromosome 7 locus identified in BSB mice. We mapped the breakpoints between the background strain-derived and donor strain-derived chromosomes. Both congenic strains contain between 27 and 29 cM of donor strain DNA on chromosome 7 (Fig. 7B).

We generated a small cross of the B10.129(5M)/nSn congenic with the C57BL/10SnJ background strain to test its suitability for fine mapping and to test the observed difference in obesity between the congenic and background strains. Sixty F2 mice have been partially phenotyped and genotyped. We observed ~1.6-fold variation in BMI and 76-fold variation in retroperitoneal fat pad weight among the F2 progeny. Examination of the retroperitoneal fat pad/BMI ratio in 32 male F2 mice suggests that segregation of the phenotype may be sufficient to allow fine mapping that will further narrow the QTL region. In addition, subcongenic strains that break the congenic region into smaller segments can be generated from this cross. A subcongenic that differs from the background strain will contain the gene of interest, whereas a subcongenic that does not differ from background will be eliminated from further consideration. This procedure, although laborious, proved successful in fine mapping in several studies (Frankel et al. 1995a, Yui et al. 1996).

Once the QTL has been narrowed to just a few centimorgans, an examination of candidate genes within that region is the next step in identifying the gene. An example of this positional candidate approach applied to a monogenic mutation is the description of Cpe^fat causing the obesity of the fat mouse. Genetic mapping showed that the obese phenotype had a chromosomal position co-incident with the gene Cpe, coding for carboxypeptidase E, which is involved in prohormone (particularly proinsulin) processing (Naggert et al. 1995). Cpe of Cpe^fat mice is also defective in pituitary in brain, suggesting that obesity may develop as a result of defective prohormone processing for neuroendocrine hormones (Naggert et al. 1995).

Several candidate genes have been identified within the confidence limits of the QTL on chromosome 7 including Ucp2 (uncoupling protein-2) (Fleury et al. 1997), Snrpn (Leff et al. 1992), Igf1r (Hochberg et al. 1992), tub (Coleman and Eicher 1990, Kleyn et al. 1996, Noben-Trauth et al. 1996), Stp (phenol preferring sulfotransferase), Ins2 (the second mouse insulin locus) (Jones et al. 1992) and Igf2 (the locus for insulin-like growth factor 2). Mapping of the donor strain chromosomal region of the B10.129(5M)/nSn congenic strain showed that it contains several of these candidate genes, including Ucp2, Snrpn, Igf1r and tub (Fig. 7B) but excludes Stp, Ins2 and Igf2. A summary of obesity QTL and candidate genes is provided in Figure 7 to simplify identification of positional candidates that are co-incident with existing and future QTL.

FOOTNOTES

LITERATURE CITED

• Andersson, L., Haley, C. S., Ellegren, H., Knott, S. A., Johansson, M., Andersson, K., Andersson-Eklund, L., Edfors-Lilja, I., Fredholm, M., Hansson, I., Håkansson, J. & Lundström, K. (1994) Genetic mapping of quantitative trait loci for growth and fatness in pigs. Science (Washington, DC) 263: 1771-1774.
• Bailey D. W. Genetics of histocompatibility in mice: I. New loci and congenic lines. Immunogenetics 1975; 2:249-256
• Bultman S. J., Michaud E. J., Woychik R. P. Molecular characterization of the mouse agouti locus. Cell 1992; 71:1195-1204 [Medline][Medline]
• Cheverud J. M., Routman E. J., Duarte F. A., van Swinderen B., Cothran K., Perel C. Quantitative trait loci for murine growth. Genetics 1996; 142:1305-1319 [Medline][Abstract]
• Clamp P. A., Beever J. E., Fernando R. L., McLaren D. G., Schook L. B. Detection of linkage between genetic markers and genes that affect growth and carcass traits in pigs. J. Anim. Sci. 1972; 70:2695-2706
• Clark J. S., Jeffs B., Davidson A. O., Lee W. K., Anderson N. H., Bihoreau M. T., Brosnan M. J., Devlin A. M., Kelman A. W., Lindpainter K., Dominiczak A. F. Quantitative trait loci in genetically hypertensive rats. Possible sex specificity. Hypertension 1996; 28:898-906 [Medline][Abstract/Free Full Text]
• Clement K., Vaisse C., Manning B.S.J., Basdevant A., Guy-Grand B., Ruiz J., Silver K. D., Shuldiner A. R., Froguel P., Strosberg A. D. Genetic variation in the beta3-adrenergic receptor and an increased capacity to gain weight in patients with morbid obesity. N. Engl. J. Med. 1995; 333:352-354 [Medline][Abstract/Free Full Text]
• Coleman D. L., Eicher E. M. Fat (fat) and tubby (tub): two autosomal recessive mutations causing obesity syndromes in the mouse. J. Hered. 1990; 81:424-427 [Medline][Abstract/Free Full Text]
• Collins A. C., Martin I. C., Kirkpatrick B. W. Growth quantitative trait loci (QTL) on mouse chromosome 10 in a Quackenbush-Swiss x C57BL/6J backcross. Mammalian Genome 1993; 4:454-458 [Medline]
• Copeland, N. G., Jenkins, N. A., Gilbert, D. J., Eppig, J. T., Maltais, L. J., Miller, J. C., Dietrich, W. F., Weaver, A., Lincoln, S. E., Steen, R. G., Stein, L. D., Nadeau, J. H. & Lander, E. S. (1993) A genetic linkage map of the mouse: current applications and future prospects. Science (Washington, DC) 262: 57-66.
• Deng A. Y., Dene H., Rapp J. P. Mapping of a quantitative trait locus for blood pressure on rat chromosome 2. J. Clin. Invest. 1994; 94:431-436 [Medline]
• Dietrich W., Katz H., Lincoln S. E., Shin H.-S., Friedman J., Dracopoli N. C., Lander E. S. A genetic map of the mouse suitable for typing intraspecific crosses. Genetics 1992; 131:423-447 [Medline][Abstract]
• Dragani T. A., Zeng Z. B., Canzian F., Gariboldi M., Ghilarducci M. T., Manenti G., Pierotti M. A. Mapping of body weight loci on mouse chromosome X. Mammalian Genome 1995; 6:778-781 [Medline]
• Fijneman F. J., de Vries S. S., Jansen R. C., Demant P. Complex interactions of new quantitative trait loci, Sluc1, Sluc2, Sluc3, and Sluc4, that influence the susceptibility to lung cancer in the mouse. Nat. Genet. 1996; 14:465-467[Medline]
• Fisler J. S., Warden C. H., Pace M. J., Lusis A. J. BSB: a new mouse model of multigenic obesity. Obes. Res. 1993; 1:271-280[Medline]
• Fleury C., Neverova M., Collins S., Raimbault S., Champigny O., Levi-Meyrueis C., Bouillaud F., Seldin M. F., Surwit R. S., Ricquier D., Warden C. H. Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat. Genet. 1997; 15:269-272 [Medline][Medline]
• Frankel W. N. Taking stock of complex trait genetics in mice. Trends Genet. 1995; 11:471-477 [Medline][Medline]
• Frankel W. N., Johnson E. W., Lutz C. M. Congenic strains reveal effects of the epilepsy quantitative trait locus, El2, separate from other El loci. Mammalian Genome 1995a; 6:839-843 [Medline]
• Frankel W. N., Valenzuela A., Lutz C. M., Johnson E. W., Dietrich W. F., Coffin J. M. New seizure frequency QTL and the complex genetics of epilepsy in EL mice. Mammalian Genome 1995b; 6:830-838 [Medline]
• Friedman J. M., Leibel R. L., Bahary N., Siegel D. A., Truett G. Genetic analysis of complex disorders: molecular mapping of obesity genes in mice and humans. Ann. N.Y. Acad. Sci. 1991; 630:100-115 [Medline][Medline]
• Gnudi L., Tozzo E., Shepherd P. R., Bliss J. L., Khan B. B. High level overexpression of glucose transporter-4 driven by an adipose-specific promoter is maintained in transgenic mice on a high fat diet, but does not prevent impaired glucose tolerance. Endocrinology 1995; 136:995-1002 [Medline][Abstract]
• Graff R. J., Snell G. D. Histocompatibility genes of mice: VIII. The alleles of the H-1 locus. Transplantation 1968; 6:598-617 [Medline][Medline]
• Hastings I. M., Veerkamp R. F. The genetic basis of response in mouse lines divergently selected for body weight or fat content. I. The relative contributions of autosomal and sex-linked genes. Genet. Res. 1993; 62:169-175 [Medline][Medline]
• Hilbert P., Lindpaintner K., Beckmann J. S., Serikawa T., Soubrier F., Dubay C., Cartwright P., De Gouyon B., Julier C., Takahasi S., Vincent M., Ganten D., Georges M., Lathrop G. M. Chromosomal mapping of two genetic loci associated with blood-pressure regulation in hereditary hypertensive rats. Nature (Lond.) 1991; 353:521-529 [Medline][Medline]
• Hochberg Z., Hertz P., Varda C., Ish-Shalom S., Yeshuran D., Youdim M.B.H., Amit T. The distal axis of growth hormone (GH) in nutritional disorders: GH binding protein, insulin-like growth factor-1 (IGF-1) and IGF-1 receptors in obesity and anorexia nervosa. Metabolism 1992; 41:106-112 [Medline][Medline]
• Hofstetter J. R., Mayeda A. R., Possidente B., Nurnberger J. I. Jr. Quantitative trait loci (QTL) for circadian rhythms of locomotor activity in mice. Behav. Genet. 1995; 25:545-556 [Medline][Medline]
• Hyman R. W., Frank S., Warden C. H., Daluiski A., Heller R., Lusis A. J. Quantitative trait locus analysis of susceptibility to diet-induced atherosclerosis in recombinant inbred mice. Biochem. Genet. 1994; 32:397-407 [Medline][Medline]
• Jacob H. J., Lindpaintner K., Lincoln S. E., Kusumi K., Bunker R. K., Mao Y. P., Ganten D., Dzau V. J., Lander E. S. Genetic mapping of a gene causing hypertension in the stroke-prone spontaneously hypertensive rat. Cell 1991; 67:213-224 [Medline][Medline]
• Jansen R. C. Controlling the type I and type II errors in mapping quantitative trait loci. Genetics 1994; 138:871-881 [Medline][Abstract]
• Jansen R. C., Stam P. High resolution of quantitative traits into multiple loci via interval mapping. Genetics 1994; 136:1447-1455 [Medline][Abstract]
• Jones J. M., Meisler M. H., Seldin M. F., Lee B. K., Eicher E. M. Localization of insulin-2 (Ins-2) and the obesity mutant tubby (tub) to distinct regions of mouse chromosome 7. Genomics 1992; 14:197-199 [Medline][Medline]
• Katz E. B., Stenbit A. E., Hatton K., DePinho R., Charron M. J. Cardiac and adipose tissue abnormalities but not diabetes in mice deficient in GLUT4. Nature (Lond.) 1995; 377:151-155 [Medline][Medline]
• Keightley P. D., Hardge T., May L., Bulfield G. A genetic map of quantitative trait loci for body weight in the mouse. Genetics 1996; 142:227-235 [Medline][Abstract]
• Kleyn P. W., Fan W., Kovats S. G., Lee J. J., Pulido J. C., Wu Y., Berkemeier L. R., Misumi D. J., Holmgren L., Charlat O., Woolf E. A., Tayber O., Brody T., Shu P., Hawkins F., Kennedy B., Baldini L., Ebeling C., Alperin G. D., Deeds J., Lakey N. D., Culpepper J., Chen H., Glucksmann-Kuis M. A., Moore K. J. Identification and characterization of the mouse obesity gene tubby: a member of a novel gene family. Cell 1996; 85:281-290 [Medline][Medline]
• Kopecky J., Clarke G., Enerback S., Spiegelman B., Kozak L. P. Expression of the mitochondrial uncoupling protein gene from the aP2 gene promoter prevents genetic obesity. J. Clin. Invest. 1995; 96:2914-2923 [Medline]
• Kozak L. P., Kozak U. C., Clarke G. T. Abnormal brown and white fat development in transgenic mice overexpressing glycerol 3-phosphate dehydrogenase. Genes Dev. 1991; 5:2256-2264 [Medline][Abstract/Free Full Text]
• Lander E. S., Botstein D. Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 1989; 121:185-199 [Medline][Abstract/Free Full Text]
• Lander E. S., Green P., Abrahamson J., Barlow A., Daly M. J., Lincoln S. E., Newburg L. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1987; 1:174-181 [Medline][Medline]
• Lander E. S., Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat. Genet. 1995; 11:241-247[Medline]
• Lander, E. S. & Schork, N. J. (1994) Genetic dissection of complex traits. Science (Washington, DC) 265: 2037-2048.
• Lardinois C. K., Caudill T., Starich G. H. Dissimilar fatty acid composition of standard rat chow. Am. J. Med. Sci. 1989; 298:305-308 [Medline][Medline]
• Leff, S. E., Brannan, C. I., Reed, M. L., Ozcelik, T., Francke, U., Copeland, N. G. & Jenkins, N. A. (1992) Maternal imprinting of the mouse Snrpn gene and conserved linkage homology with the human Prader-Willi syndrome region. Nat. Genet.2: 259-264.
• Lembertas, A. V., Lusis, A. J. & Fisler, J. S. (1996a) Differential effects of leptin on growth and body weight in five mouse strains. Obes. Res. 4: 37S (abs.).
• Lembertas, A. V., Lusis, A. J. & Fisler, J. S. (1996b) Differential effects of leptin on mass of four fat pads in five mouse strains. Obes.Res. 4: 37S (abs.).
• Lembertas, A. V., Pérusse, L., Chagnon, Y. C., Fisler, J. S., Warden, C. H., Purcell-Huynh, D. A., Dionne, F. T., Gagnon, J., Nadeau, A., Lusis, A. J. & Bouchard, ÉC. (1997) Identification of an obesity quantitative trait lows on mouse chromosome 2 and evidence of linkage to body fat and insulin on the human homologous region 20q. J. Clin. Invest. 100 (in press).
• Luetteke N. C., Lee D. C., Palmiter R. D., Brinster R. L., Sandgren E. P. Regulation of fat and muscle development by transforming growth factor alpha in transgenic mice and in cultured cells. Cell Growth Dev. 1993; 4:203-213
• Miller M. W., Duhl D. M. J., Vrieling H., Cordes S. P., Ollmann M. M., Winkes B. M., Barsh G. S. Cloning of the mouse agouti gene predicts a secreted protein ubiquitously expressed in mice carrying the lethal yellow mutation. Genes Dev. 1993; 7:454-467 [Medline][Abstract/Free Full Text]
• Moisan M. P., Courvoisier H., Bihoreau M. T., Gauguier D., Hendley E. D., Lathrop M., James M. R., Mormede P. A major quantitative trait locus influences hyperactivity in the WKHA rat. Nat. Genet. 1996; 14:471-473 [Medline][Medline]
• Morel L., Yu Y., Blenman K. R., Caldwell R. A., Wakeland E. K. Production of congenic mouse strains carrying genomic intervals containing SLE-susceptibility genes derived from the SLE-prone NZM2410 strain. Mammalian Genome 1996; 7:335-339 [Medline]
• Nagase H., Bryson S., Fee F., Balmain A. Multigenic control of skin tumour development in mice. Ciba Found. Symp. 1996; 197:156-168 [Medline][Medline]
• Naggert J. K., Fricker L. D., Varlamov O., Nishina P. M., Rouille Y., Steiner D. F., Carroll R. J., Paigen B. J., Leiter E. H. Hyperproinsulinaemia in obese fat/fat mice associated with a carboxypeptidase E mutation which reduces enzyme activity. Nat. Genet. 1995; 10:135-141 [Medline][Medline]
• Noben-Trauth K., Naggert J. K., North M. A., Nishina P. M. A candidate gene for the mouse mutation tubby. Nature (Lond.) 1996; 380:534-538 [Medline][Medline]
• Pace, M. J., Tieu. L., Warden, C. H. & Fisler, J. S. (1993) Differential response to ACTH stimulation in lean and obese mice in a model of polygenic obesity. Obes. Res. (Suppl. 1): 39S.
• Paterson, A. H., Damon, S., Hewitt, J. D., Zamir, D., Rabinowitch, H. D., Lincoln, S. E., Lander, E. S. & Tanksley, S. D. (1991) Mendelian factors underlying quantitative traits in tomato: comparison across species, generations, and environments. Genetics 127:181-197.
• Paterson A. H., Lander E. S., Hewitt J. D., Peterson S., Lincoln S. E., Tanks- ley, S. D. Resolution of quantitative traits into Mendelian factors by using a complete linkage map of restriction fragment length polymorphisms. Nature (Lond.) 1988; 335:721-726 [Medline][Medline]
• Pepin M.-C., Pothier F., Barden N. Impaired type II glucocorticoid-receptor function in mice bearing antisense RNA transgene. Nature (Lond.) 1992; 355:725-728 [Medline][Medline]
• Pomp, D., Cushman, M. A., Foster, S. C., Drudik, D. K., Fortman, M. & Eisen, E. J. (1994) Identification of quantitative trait loci for body weight and body fat in mice. In: 5th World Congress on Genetics Applied to Livestock Production, pp. 209-212. University of Guelph, Guelph, Canada.
• Purcell-Huynh D. A., Weinreb A., Castellani L. W., Mehrabian M., Doolittle M. K., Lusis A. J. Genetic factors in lipoprotein metabolism: analysis of a genetic cross between inbred mouse strains NZB/B1NJ and SM/J using a complete linkage map approach. J. Clin. Invest. 1995; 96:1845-1858 [Medline]
• Rance, K. A., Hastings, I. M., Hill, W. G. & Keightley, P. D. (1994) Mapping of putative QTL influencing body weight on the X chromosome in mice. In: 5th World Congress on Genetics Applied to Livestock Production, pp. 268-269. University of Guelph, Guelph, Canada.
• Remmers E. F., Longman R. E., Du Y., O'Hare A., Cannon G. W., Griffiths M. M., Wilder R. L. A genome scan localizes five non-MHC loci controlling collagen-induced arthritis in rats. Nat. Genet. 1996; 14:82-85 [Medline][Medline]
• Samani N. J., Gauguier D., Vincent M., Kaiser M. A., Bihoreau M. T., Lodwick D., Wallis R., Parent V., Kimber P., Rattray F., Thompson J. R., Sassard J., Lathrop M. Analysis of quantitative trait loci for blood pressure on rat chromosomes 2 and 13. Age-related differences in effect. Hypertension 1996; 28:1118-1122 [Medline][Abstract/Free Full Text]
• Schemmel R., Mickelson O., Gill J. L. Dietary obesity in rats: body weight and body fat accretion in seven strains of rats. J. Nutr. 1970; 100:1041-1048 [Medline]
• Schork N. J., Krieger J. E., Trolliet M. R., Franchini K. G., Koike G., Kreiger E. M., Lander E. S., Dzau V. J., Jacob H. J. A biometrical genome search in rats reveals the multigenic basis of blood pressure variation. Genome Res. 1995; 5:164-172 [Medline][Abstract/Free Full Text]
• Tartaglia L. A., Dembski M., Weng X., Deng N., Culpepper J., Devos R., Richards G. J., Campfield L. A., Clark F. T., Deeds J., Muir C., Sanker S., Moriarty A., Moore K. J., Smutko J. S., Mays G. G., Woolf E. A., Monroe C. A., Tepper R. I. Identification and expression cloning of a leptin receptor, OB-R. Cell 1995; 83:1263-1271 [Medline][Medline]
• Taylor B. A., Phillips S. J. Detection of obesity QTLs on mouse chromosomes 1 and 7 by selective DNA pooling. Genomics 1996; 34:389-398 [Medline][Medline]
• Tecott L. H., Sun L. M., Akana S. F., Strack A. M., Lowenstein D. H., Dallman M. F., Jullus D. Eating disorder and epilepsy in mice lacking 5-HT2c serotonin receptors. Nature (Lond.) 1995; 374:542-546 [Medline][Medline]
• Warden, C. H. & Fisler, J. S. (1994) Identification of genes underlying polygenic obesity in animal models. In: The Genetics of Obesity (Bouchard, C., ed.), pp. 181-197. CRC Press, Boca Raton, FL.
• Warden, C. H. & Fisler, J. S. (1997) Molecular genetics of obesity. In: Handbook of Obesity (Bray, G. A., Bouchard, C. & James, W.P.T., eds.), CRC Press (in press).
• Warden C. H., Fisler J. S., Pace M. J., Svenson K. L., Lusis A. J. Coincidence of genetic loci for plasma cholesterol levels and obesity in a multifactorial mouse model. J. Clin. Invest. 1993; 92:773-779 [Medline]
• Warden C. H., Fisler J. S., Shoemaker S. M., Wen P.-Z., Svenson K. L., Pace M. J., Lusis A. J. Identification of four chromosomal loci determining obesity in a multifactorial mouse model. J. Clin. Invest. 1995; 95:1545-1552 [Medline]
• Warden, C. H., Kirchgessner, T., Lusis, A. J. & Fisler, J. S. (1996) Evidence that alleles of the leptin gene influence femoral fat pad weight in the BSB obesity model. Obes. Res. 4: 9S (abs.).
• Warden C. H., Mehrabian M., He K., Yoon M.-Y., Diep A., Xia Y.-R., Wen P.-Z., Svenson K. L., Sparkes R. S., Lusis A. J. Linkage mapping of 40 randomly isolated liver cDNA clones in the mouse. Genomics 1993; 18:295-307 [Medline][Medline]
• West, D. B., Boozer, C. N., Moody, D. L. & Atkinson, R. L. (1992) Dietary obesity in nine inbred mouse strains. Am. J. Physiol. 262: R1025-R1032.
• West, D. B., Goudey-Lefevre, J., York, B. & Truett, G. E. (1994a) Dietary obesity linked to genetic loci on chromosomes 9 and 15 in a polygenic mouse model. J.Clin.Invest. 94: 1410-1416.
• West D. B., Waguespack J., York B., Goudey-Lefevre J., Price R. A. Genetics of dietary obesity in AKR/J X SWR/J mice: segregation of the trait and identification of a linked locus on Chromosome 4. Mammalian Genome 1994b; 5:546-552 [Medline]
• York B., Lei K., West D. B. Sensitivity to dietary obesity linked to a locus on chromosome 15 in a CAST/Ei x C57BL/6J intercross. Mammalian Genome 1996; 7:677-681 [Medline]
• Yui M. A., Muralidharan K., Moreno-Altamirano B., Perrin G., Chestnut K., Wakeland E. K. Production of congenic mouse strains carrying NOD-derived diabetogenic genetic intervals: an approach for the genetic dissection of complex traits. Mammalian Genome 1996; 7:331-334 [Medline]
• Zhang Y., Proenca R., Maffel M., Barone M., Leopold L., Friedman J. M. Positional cloning of the mouse obese gene and its human homologue. Nature (Lond.) 1994; 372:425-432 [Medline]

0022-3166/97 $3.00 ©1997 American Society for Nutritional Sciences