Original Article

International Journal of Obesity (2010) 34, 1706–1714; doi:10.1038/ijo.2010.97; published online 25 May 2010

A unique genetic defect on chromosome 3 is responsible for juvenile obesity in the Berlin Fat Mouse

C Neuschl1, C Hantschel1, A Wagener1, A O Schmitt1, T Illig2 and G A Brockmann1

  1. 1Department for Crop and Animal Sciences, Humboldt-Universität zu Berlin, Berlin, Germany
  2. 2GSF National Research Center for Environment and Health, Institute of Epidemiology, Neuherberg, Germany

Correspondence: Dr GA Brockmann, Department for Crop and Animal Sciences, Breeding Biology and Molecular Genetics, Humboldt-Universität zu Berlin, Invalidenstrasse 42, D-10115 Berlin, Germany. E-mail: gudrun.brockmann@agrar.hu-berlin.de

Received 31 January 2010; Revised 1 April 2010; Accepted 4 April 2010; Published online 25 May 2010.





This study aimed at the mapping and estimation of genetic and sex effects contributing to the obese phenotype of the Berlin Fat Mouse Inbred line 860 (BFMI860). This mouse line is predisposed for juvenile obesity. BFMI860 mice accumulate 24% total fat mass at 10 weeks of age under a standard maintenance diet.



A total of 471 mice of a (BFMI860 × C57BL/6NCrl) F2 intercross population were fed a standard maintenance diet and were analysed for body composition at 10 weeks when they finished their rapid growth phase.



The most striking result was the identification of a novel obesity locus on chromosome 3 (Chr 3) at 40Mb, explaining 39% of the variance of total fat mass in the F2 population under a standard diet. This locus was named jObes1 (juvenile obesity 1). The BFMI860 allele effect was recessive. Males and females homozygous at jObes1 had on average 3.0 and 3.3g more total fat mass at 10 weeks than the other two genotype classes, respectively. The effect was evident in all white adipose tissues, brown adipose tissue and also in liver. The position of the Chr 3 effect is syntenic to an obesity locus in humans. Additional loci for total fat mass and different white adipose tissue weights with minor effects were detected on mouse Chr 5 and 6. Another locus on Chr 4 had influence especially on liver weight. Many loci including jObes1 affected males and females to a different extent.



The major locus on Chr 3 for juvenile obesity and its interaction with sex is unique and makes the BFMI860 mice an interesting resource for the discovery of novel genetic factors predisposing obesity, which might also contribute to obesity in humans. The results suggested that metabolic and regulatory pathways differed between the sexes.


mouse; quantitative trait loci; QTL mapping; sex effects



The degree of obesity differs between individuals because of their genetic predisposition and lifestyle, for example, nutrition and physical activity. Diverse genetic studies in humans have shown that besides few cases of monogenic obesity, most cases of obesity are caused by multiple genes with each having only a small effect.1 In humans, mutations in the melanocortin-4 receptor (MC4R) gene2, 3, 4 and the fat mass and obesity associated (FTO) gene5, 6 have been repeatedly associated with body mass index. Comprehensive genome-wide association studies in human populations have identified additional single DNA variants and their linked chromosomal regions with influence on obesity measures.7, 8, 9 However, the final identification of the affected genes remains difficult in humans.

During the past years, mutant mouse models have repeatedly contributed to the discovery of key factors in different pathways controlling body weight. In addition to monogenic models of obesity such as the knockout mice (including ob and db mice), polygenic models have been studied.10 The analysis of polygenic obesity is facilitated by the initial study of crosses between diverse inbred mouse lines. Such crosses led to the mapping of genomic regions with linkage to obesity genes on all chromosomes with particularly high density on mouse chromosomes (Chr) 1, 2, 7, 11, 15 and 17.11 So far, only few genes with causal mutations underlying quantitative trait locus (QTL) effects have been identified. These genes further help us to understand the mode of inheritance and determination of obesity in general. For example, recently, a mutation in the Tbc1d1 gene has been identified as a suppressor for high-fat diet-induced obesity in the lean SJL mouse line.12 The SJL allele acts in particular in muscle in which it increases fatty acid uptake and oxidation.

Additional experiments including more genetically distinct mouse lines are desirable because they hold the chance of contributing additional genetic variation for our understanding of the genetic architecture of obesity-related traits. In this respect, obese mice harbouring natural mutations are of particular interest. One of these models is the Berlin Fat Mouse Inbred line 860 (BFMI860), which had been selected for high fat mass over many generations. Therefore, a hitherto unique cross between BFMI860 and the lean line C57BL/6NCrl (B6) was generated for further studies on the genetic basis of body weight control. A previous study showed that BFMI860 mice accumulated on average 24% of body mass as fat, whereas B6 disposed of only 3% fat at 10 weeks of age under a standard diet.13 BFMI860 offspring had higher body weights already at birth. The highest weight gain occurred between 6 and 10 weeks of age. Metabolic examinations revealed that the excessive accumulation of body fat in BFMI860 mice was associated with an altered lipid metabolism and a high energy intake and digestion.14

To assess the genetic determination of the obese phenotype in BFMI860 mice, in the present study, chromosomal regions for obesity traits were mapped in a (BFMI860 × B6) F2 intercross population. This crossbred population facilitated the identification and analysis of potentially new genetic determinants of body weight and fat distribution patterns and enlightened differences between the sexes.


Materials and methods

Mouse lines

The mouse lines used in this study were phenotypically extremely different. The BFMI860 was generated from an outbred population. Founder animals of the BFM population were originally purchased in several pet shops in Berlin, Germany. This led to a genetically highly heterogeneous base population for the subsequent selection experiment. The selection process comprised several distinct phases that have been described recently.13 In brief, mice were repeatedly selected first for low protein content and afterward for high body weight. All selection decisions were performed on mice on a standard breeding diet.

Owing to the fact that the genetic constitution of the base population for the selection of BFM was unknown (mice of unknown origin were crossed) and an internal unselected control line of the selection experiment became extinct during the 40 years of breeding history, we used B6 (National Institutes of Health, Charles River Laboratories, Sulzfeld, Germany) as a lean contrast inbred line. In a pre-test, growth and body composition of C57BL/6J (The Jackson Laboratory) and C57BL/6NCrl mice were compared under our conditions. C57BL/6NCrl accumulated less white adipose tissue than C57BL/6J in our mouse facility, which was an advantage for the experimental design to map obesity QTLs. Furthermore, higher reproduction rates were found in C57BL/6NCrl animals, which had on average 1.4 more born offspring per litter and fewer losses before weaning (18% in C57BL/6NCrl and 32% in C57BL/6J; Neuschl C, unpublished data). Such a higher reproduction performance was required for this intercross experiment and for intended follow-up studies, such as repeated backcrosses to B6.

Animal husbandry and feeding conditions

The animals were treated in accordance to and all experimental protocols were approved by the German Animal Welfare Authorities (approval no. G0152/04, T0149/04, O0145/04).

Mice were maintained under conventional conditions and controlled lighting with a 12:12h light/dark cycle at a temperature of 22±2°C and a relative humidity of 65%. They were reared in groups of three to four mice of the same sex in macrolon cages with a 350cm2 floor space (E. Becker & Co (Ebeco) GmbH, Castrop-Rauxel, Germany) and with bedding type S 80/150, dust-free (Rettenmeier Holding AG, Wilburgstetten, Germany). All mice had ad libitum access to food and water.

After weaning at 21 days, F2 mice were fed a standard maintenance diet. This diet (Ssniff diet V1534-0, Castrop-Rauxel, Germany) contained 19.0% crude protein, 3.3% crude fat, 4.9% crude fibre, 6.4% crude ash, 54.1% nitrogen-free extract (thereof 36.5% starch and 4.7% sugar), vitamins, trace elements, amino acids and minerals (12.8MJkg–1 metabolizable energy; thereof 9% energy from fat, 58% from carbohydrates and 33% from proteins). The fat in the standard maintenance diet was derived from soy oil (50–60%), wheat and barley (40–50%).

Pedigree structure

An F2 intercross population was generated by crossing one male of the obese line BFMI860 (ninth generation of inbreeding) to eight females of the control line B6 with a total of 471 F2 offspring (222 males and 249 females). The population size was realized by repeated matings (one to four times) of the parents and subsequently repeated matings (one to five times) of 28 pairs of F1 animals, building sub-families. Litter size ranged from 7 to 14 offspring, and smaller litters were immediately excluded from the experiment. For the statistical analyses, all 471 F2 animals were considered. For the QTL analysis, only 365 F2 mice of 22 sub-families were used.


In all, 5 males and 5 females of each of the two parental lines BFMI860 and B6, 11 male and 13 female F1 and all F2 animals were phenotyped at 10 weeks of age. In addition, body weight, total fat mass and total lean mass were measured weekly on the basis of the birthdays of animals from weeks 4 to 9.

Body weight was recorded using a digital balance with a computer interface accurate to 0.01g. Total fat mass and total lean mass were determined in non-anaesthetized mice by quantitative magnetic resonance interference analysis using the EchoMRI-100 whole body composition analyser (Echo Medical Systems, Houston, TX, USA).15, 16 This instrument creates a contrast between adipose tissue, muscle and free body fluids by taking advantage of the differences in relaxation times of the hydrogen spins and hydrogen density in these tissues. Each magnetic resonance measurement per animal was repeated two to three times and the median was used for further analyses. Total fat mass represented the sum of all fat in the body. Total lean mass included mainly muscle and inner organs. Skeletal muscle mass accounted for the largest portion of total lean mass.

At 71 days and after a fasting period of 2h, mice were anaesthetized under isofluorane and decapitated using surgical scissors. After exsanguination, white adipose tissues were dissected and weighed. These tissues comprised the reproductive adipose tissue, which was the epididymal adipose tissue in males and the periuterine and the periovarian adipose tissues in females, the renal adipose tissue (peritoneal and retroperitoneal adipose tissues) and the subcutaneous adipose tissue. The subcutaneous adipose tissue was all white adipose tissue underneath the skin, including the depot near the hind limb (inguinal), the depot between the shoulder blades, near the brown adipose tissue (subscapular) and the depot around the tailhead. The brown adipose tissue and the liver were also weighed.


DNA was extracted from tail tips using the Invisorb Spin Tissue Mini Kit (Invitek, Berlin, Germany). Out of 471 phenotyped F2 animals, 365 were genotyped at 132 informative markers, 69 single-nucleotide polymorphisms and 63 microsatellites, covering all chromosomes except Y with an average distance of 21.08Mb (Figure 1). Microsatellites were genotyped as described previously.17 Single-nucleotide polymorphisms were analysed using allele-specific primers and a MALDI TOF MS (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry) system (Sequenom, Inc., San Diego, CA, USA). Among the single-nucleotide polymorphisms was a polymorphism in the leptin gene (rs13478682). One of the markers (D11Mit227) was heterozygous in the parental BFMI860 male.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Map of 69 reference single-nucleotide polymorphisms and 63 microsatellite markers used in this study (drawn using MapChart; http://www.biometris.wur.nl/uk/Software/MapChart). Positions are given in Mb (Ensembl release 47). Bars indicate the confidence intervals of identified highly significant QTLs.

Full figure and legend (193K)

The marker order was checked and a pedigree-specific marker map was built using the program JoinMap4 (http://www.kyazma.nl/index.php/mc.JoinMap).18 As the pedigree-specific order was consistent with the marker order in the genomic mouse sequence (Ensembl database release 47 at http://www.ensembl.org/Mus_musculus), we used marker positions of the genomic sequence in Mb divided by two (Mb/2) for QTL mapping in the F2 population. In mice, 1 Mb is equivalent to approximately 0.5cM.19, 20 Using Mb/2 instead of the pedigree-specific map might have slightly changed test statistics and estimates but it would have not changed the results in principle. The advantage of using the Mb/2 map is that QTL locations given in Mb would allow the direct incorporation of detected QTLs into the physical reference map.

Statistical analyses

Calculations of basic statistics were performed using SAS/STAT statistical software package version 9.1.3 (SAS Institute Inc., Cary, NC, USA) and the MEANS procedure. The data were tested for normality using the UNIVARIATE procedure. The generalized linear model procedure was applied for analyses of variance.

QTL mapping

In this study, log10-transfomed trait values were used in the QTL analyses because the raw data were not normally distributed for the recorded phenotypes.

QTLs were mapped with GridQTL (http://www.gridqtl.org.uk),21 in which the multiple regression least squares approach is implemented.22 This program enables QTL mapping in outbred populations and thus the inclusion of heterozygous parental marker genotypes into linkage analyses.

A general linear model was fitted to the trait values including sex, sub-family (22 levels) and season (three levels) as fixed effects. Litter size (three levels: 7–9, 10–12 and 13 to 14 offspring, respectively) and the litter number of the F1 dam (six levels) were included as covariates. For each significant QTL identified with this standard model, sex was fitted as additional interaction term into the genetic model. The QTL was assumed to be affected by sex if the difference between the F-value of the standard and the interaction model ΔF was >4.6. An F-value of approximately 4.6 corresponded to a LOD (logarithm (base 10) of odds) score of 2.0 in this F2 population. A LOD difference of 2.0 was significant (Pless than or equal to0.05) in a simulation study with a sample size of 200 intercross mice.23

For the detection of QTLs, genome-wide scans were performed using the forward selection interval mapping approach.24 Interactions between QTLs were tested by a two-dimensional grid search as described previously.25 Pless than or equal to0.001 was used as stringent and Pless than or equal to0.01 as relaxed threshold for the acceptance of QTL interactions.

Chr X was analysed as a pseudo-autosome in all analyses as all markers were located in that region. The direction of the genetic effects was given as BFMI860 allele effect compared with B6.

Empirically derived significance thresholds for the one-QTL versus no-QTL test statistics were estimated with a permutation test.26 In all, 1000 permutations of the data were performed. Owing to the fact that the estimated thresholds (chromosome- and genome-wide, 1 and 5%) were very similar for all chromosomes and all traits, the means over all appropriate thresholds for total fat mass, total lean mass and subcutaneous adipose tissue weight were used as representative thresholds for all chromosomes, all traits and all models. The following thresholds were taken from the resulting test statistic distribution: ‘genome-wide highly significant’ (P=0.01) corresponded to an F-value of 10.2; ‘genome-wide significant’ (P=0.05) corresponded to F=8.2; and ‘genome-wide suggestive’ (P=0.63) corresponded to ‘chromosome-wide significant’ (P=0.05) and ranged between 4.2less than or equal toFless than or equal to5.4 for the different chromosomes. A parametric bootstrap with 1000 iterations27 was performed to estimate the 95% confidence interval of a single QTL location.

Symbols were assigned to QTLs at the genome-wide highly significant level (P=0.01). If QTLs for several fat deposition traits were colocalized at the same chromosomal region, they were assigned to the same QTL symbol (juvenile obesity (jObes)).



Phenotypes of parental lines, F1 and F2 animals

Irrespective of the sex, animals of the BFMI860 line accumulated on average 12.1 times more total fat mass, showed a 1.3 times higher total lean mass and were overall 1.6 times heavier than B6 animals (Table 1). In contrast to most other mouse lines, in which males express a higher phenotype than females,28, 29 BFMI860 females accumulated the same high amount of total fat mass and single adipose tissues as the males. Taking body weight into account, females had a higher total fat mass percentage than males. The means of all recorded traits of the F1 and also the F2 mice shifted toward the respective weights found in B6 animals. The F2 population did not exceed the parental limits on average, although single animals expressed a more extreme phenotype. In the comparison of the two sexes, F2 males and females were similar with respect to total fat mass and subcutaneous adipose tissue weight, but differed significantly in all other traits. Males had significantly more reproductive, renal (peritoneal and retroperitoneal) and brown adipose tissue weights and increased liver weights than females.

QTL effects

The most striking result of this study was the mapping of a novel major QTL for total fat mass on Chr 3 at 40Mb (closest marker D3Mit21 at 37Mb) with a narrow confidence interval of 10Mb. The locus was genome-wide highly significant with an extremely high F-value of 87.7 (Table 2 and Figure 2). This locus was named jObes1 (juvenile obesity 1) QTL. It explained 39% of the variance of total fat mass in 10-week-old mice of the (BFMI860 × B6) F2 population. The genetic effect of the obesity-inducing BFMI860 allele at jObes1 was recessive. The QTL accounted for 3.2g difference in total fat mass between homozygous BFMI/BFMI and B6/B6 mice. Male and female BFMI860/BFMI860 homozygous mice accumulated on average 3.0 and 3.3g more fat mass than B6/B6 and BFMI860/B6 animals, respectively (Figure 3). The jObes1 effect on fat deposition was evident in all measured white adipose tissues, brown adipose tissue and liver. Furthermore, significant differences between the genotype classes were already apparent for total fat mass percentage from weeks 5 to 6 in males and females, respectively (Figure 4, Supplementary Table 1).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Genome scans ((a) Chr 3 and (b) all chromosomes except Chr 3) for total fat mass (FAT, solid line) and total lean mass (LEAN, dotted line) at the age of 10 weeks in the (BFMI860 × B6) F2 population. Note the different scales for the F-value in (a) and (b). The two horizontal lines represent F-value thresholds at the genome-wide 5% level (dashed line) and the suggestive level (dotted line), respectively.

Full figure and legend (82K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Effect plots showing means and s.d. of various traits at 10 weeks of the three genotype classes at QTL peak marker D3Mit21 on Chr 3 at 37Mb in males (filled squares) and females (open squares) of the F2 population (note that the overall phenotypic performance is not only due to jObes1 but also to the genetic background). In both sexes, differences in body weight, total fat mass, adipose tissue weights and liver weight between B6/B6 homozygous and BFMI860/B6 heterozygous animals were not significant, but these two genotype classes differed significantly from BFMI860/BFMI860 homozygous mice with P<0.01. For total lean mass, all genotype classes were not significantly different from each other. For significant differences between the two sexes, see Supplementary Table 1.

Full figure and legend (122K)

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Age-dependent effect of the Chr 3 jObes1 QTL on body fat percentage, conditional for the three genotype classes at QTL peak marker D3Mit21 in F2 males (left) and females (right). Body fat percentage was calculated as the proportion of total fat mass to body weight. In both sexes, differences in all seven traits between B6/B6 homozygous and BFMI860/B6 heterozygous animals were not significant. These two genotype classes differed significantly from BFMI860/BFMI860 homozygous mice with P<0.02 in all cases except: B6/B6 vs BFMI/BFMI for body fat percentage at 28 days in males, B6/B6 vs BFMI/BFMI and BFMI/B6 vs BFMI/BFMI for body fat percentage at 28 days in females and B6/B6 vs BFMI/BFMI for body fat percentage at 35 days in females (in all these cases P>0.05). For significant differences between the two sexes, see Supplementary Table 1.

Full figure and legend (59K)

Additional genome-wide significant QTLs influencing body weight were found on Chr 1, 4, 5, 6 and 9 (Table 2). The region on Chr 5 with the most likely QTL positions between 98 and 106Mb (jObes2) had also influence on the mass of reproductive, renal and subcutaneous adipose tissues. Brown adipose tissue weight was also affected, but the most likely position was found at 70Mb (jObes3). In the Chr 6 region between 26 and 28Mb (jObes4), QTLs on total fat mass, reproductive and subcutaneous adipose tissues coincided. The loci affecting body weight on Chr 4, 6 and 9 had influence on total lean mass, but not on fat deposition. The effects of Chr 4 on body weight (jBw) and liver weight (jLiv) were highly significant. Further loci linked to increased liver weight were found on Chr 3, 4 and 6. In addition, on Chr 6, loci for total fat and total lean mass mapped to the same position at 26cM, suggesting pleiotropic effects of one gene or the combined effect of several genes in the detected QTL region. The effects of the BFMI860 alleles of the identified loci were additive increasing, except for the QTL on Chr 9, in which the B6 allele was the high performance allele.

Interactions between QTLs were tested but no significant result was found.

Sex effects

In the F2 population, 13 out of 24 (54%) QTLs reported in Table 2 were affected by sex (sex interaction model). Most of these QTLs affected females more than males. The only QTL with higher additive genetic effects in males than in females was detected on Chr 6 for subcutaneous adipose tissue weight. The sex differences at the jObes1 QTL on Chr 3 are depicted in Figure 3.



This study was directed toward the decomposition of genetic effects leading to obesity in BFMI860 mice. A linkage analysis in a classical F2 intercross between the extremely different mouse lines, BFMI860 and B6, provided evidence for a novel major QTL for total fat mass in juvenile mice on Chr 3 at 40Mb (jObes1) with a very narrow confidence interval. This QTL controlled the fat deposition in all analysed adipose tissues and liver weight. The BFMI860 allele was recessive and acted differently in males and females. Despite the huge recessive jObes1 effect, this study confirmed, with the identification of additional QTLs on five chromosomes, that most genetic variance contributing to body weight and obesity as complex traits is additive.30, 31

Although the mouse genome is multi-saturated with many QTLs for body weight- and obesity-related traits,11 loci on Chr 3 were only identified in two crossbred experiments (Mouse Genome Informatics database version 4.32; http://www.informatics.jax.org). In a F2 cross between the high-body-weight-selected line DU6i and DBA/2J, three QTLs were linked to body weight and white adipose tissue between 10 and 30cM (approximately 20 to 60Mb): Abfp4 (26.0cM, approximately 52Mb),24 Afpq1 (26.0cM, approximately 52Mb) and Afw1 (30.0cM, approximately 60Mb).32 Abfp4 was only significant when interacting with a locus on Chr 5, but it was another hint for a QTL for abdominal white adipose tissue percentage in this population. The QTL allele originating from DU6i had an additive genetic effect and decreased fat deposition in all cases.

In a cross between C57BL/6J and PWK/PhJ mice, a locus at 25cM (approximately 50Mb) had significant effects on total fat and total lean mass at 18 weeks of age (Bwtq13).33 The plus allele originated from C57BL/6J (which is a different sub-line compared with C57BL/6NCrl in this study), and interestingly, the mode of inheritance of this B6 allele was also dominant as in this study. However, the effect size of Bwtq13 was much smaller. Overall, the jObes1 locus with major effects on total fat mass and different adipose tissue weights in BFMI860 is unique.

The syntenic region of the confidence interval of jObes1 in the human genome is located on Chr 3q26.33, Chr 4q27–4q28.2 and Chr Xq28 (National Center for Biotechnology Information (NCBI) build 37.1). Within these human regions, loci associated with body mass index or waist circumference were only found on Chr 3q26.33 in five different populations covering four ethnic groups (Caucasians, African Americans, Mexican Americans and Asians).34, 35, 36, 37, 38, 39 In non-Hispanic whites and African Americans, sex-specific effects on body mass index and body fat percentage of this chromosomal region were identified.34 The effect on body mass index was evident in men but not in women.

The genomic region comprising the confidence interval of jObes1 between 34 and 44Mb in the mouse genome harbours 47 known protein coding genes (Ensembl release 56). These genes are positional candidates most likely underlying the effect of this QTL. This chromosomal region is particularly rich in genes acting in energy partitioning. For example, Mccc1 and Acad9, genes encoding a coenzyme-A-carboxylase and a coenzyme-A-dehydrogenase, respectively, are key enzymes in the mitochondrial fatty acid oxidation. Another gene is Slc25a31, a mitochondrial carrier and thus involved in the energy transfer. But further fine mapping of the target jObes1 region is necessary to suggest most potential candidate genes.

The jObes1 QTL explained approximately 39% of the variance of total fat mass at 10 weeks in the F2 population; jObes2, 3 and 4 accounted for 6.76 to 8.08% of the variance of different adipose tissue weights.

The gene underlying jObes4 on Chr 6 at 29Mb was very likely the gene encoding the appetite-suppressing hormone leptin. The genotyped single-nucleotide polymorphism marker rs13478682 is located in the intronic region of the leptin gene.

The QTL for liver weight (jLiv) on Chr 4 at 72Mb coincided with Lvrq9 and Lwq, QTLs that were detected in the crosses M16i × L640 and in DU6 × DUKs and DU6i × DBA/2,32, 41 respectively. In all cases, including this study, the allele of the high-body-weight-selected line (BFMI860, M16i, DU6 and DU6i) had an increasing effect on the trait. Therefore, the underlying gene/s for high liver weight might be the same.

In this study, QTLs on Chr 3 and 6 influencing obesity were affected by sex. Such interactions between QTLs and sex were also observed in other obesity-related studies in mice,42, 43, 44, 45 suggesting that underlying genes were differentially expressed in males and females. For example, estrogens that differ between the two sexes can act as transcription factors and might regulate the QTL alleles in a sex-specific manner. This led to differences in metabolic and regulatory pathways between males and females.

This study was the initial but important step in the way of identifying the genetic causes of juvenile obesity in BFMI860 mice. As a result of the detected major QTL jObes1 on Chr 3 and its narrow confidence interval of 10Mb, the identification of the underlying gene or genes with their causal mutation(s) will be feasible in the near future and will further contribute significantly in our understanding of body weight control. The Chr 3 mutation in the BFMI860 mouse line could provide new insights into the molecular mechanisms responsible for the effects in syntenic human regions.


Conflict of interest

The authors declare no conflict of interest.



  1. Barsh GS, Farooqi IS, O’Rahilly S. Genetics of body-weight regulation. Nature 2000; 404: 644–651. | Article | PubMed | ISI | ChemPort |
  2. Roth CL, Ludwig M, Woelfle J, Fan ZC, Brumm H, Biebermann H et al. A novel melanocortin-4 receptor gene mutation in a female patient with severe childhood obesity. Endocrine 2009; 36: 52–59. | Article | PubMed | ChemPort |
  3. Ochoa MC, Azcona C, Biebermann H, Brumm H, Razquin C, Wermter AK et al. A novel mutation Thr162Arg of the melanocortin 4 receptor gene in a Spanish children and adolescent population. Clin Endocrinol (Oxf) 2007; 66: 652–658. | Article | PubMed | ChemPort |
  4. Hinney A, Bettecken T, Tarnow P, Brumm H, Reichwald K, Lichtner P et al. Prevalence, spectrum, and functional characterization of melanocortin-4 receptor gene mutations in a representative population-based sample and obese adults from Germany. J Clin Endocrinol Metab 2006; 91: 1761–1769. | Article | PubMed | ChemPort |
  5. Hinney A, Nguyen TT, Scherag A, Friedel S, Bronner G, Muller TD et al. Genome wide association (GWA) study for early onset extreme obesity supports the role of fat mass and obesity associated gene (FTO) variants. PLoS ONE 2007; 2: e1361. | Article | PubMed | ChemPort |
  6. Rampersaud E, Mitchell BD, Pollin TI, Fu M, Shen H, O’Connell JR et al. Physical activity and the association of common FTO gene variants with body mass index and obesity. Arch Intern Med 2008; 168: 1791–1797. | Article | PubMed
  7. Lindgren CM, Heid IM, Randall JC, Lamina C, Steinthorsdottir V, Qi L et al. Genome-wide association scan meta-analysis identifies three loci influencing adiposity and fat distribution. PLoS Genet 2009; 5: e1000508. | Article | PubMed | ChemPort |
  8. Thorleifsson G, Walters GB, Gudbjartsson DF, Steinthorsdottir V, Sulem P, Helgadottir A et al. Genome-wide association yields new sequence variants at seven loci that associate with measures of obesity. Nat Genet 2009; 41: 18–24. | Article | PubMed | ChemPort |
  9. Willer CJ, Speliotes EK, Loos RJ, Li S, Lindgren CM, Heid IM et al. Six new loci associated with body mass index highlight a neuronal influence on body weight regulation. Nat Genet 2009; 41: 25–34. | Article | PubMed | ChemPort |
  10. Speakman J, Hambly C, Mitchell S, Krol E. The contribution of animal models to the study of obesity. Lab Anim 2008; 42: 413–432. | Article | PubMed | ChemPort |
  11. Wuschke S, Dahm S, Schmidt C, Joost HG, Al-Hasani H. A meta-analysis of quantitative trait loci associated with body weight and adiposity in mice. Int J Obes (Lond) 2007; 31: 829–841. | PubMed | ChemPort |
  12. Chadt A, Leicht K, Deshmukh A, Jiang LQ, Scherneck S, Bernhardt U et al. Tbc1d1 mutation in lean mouse strain confers leanness and protects from diet-induced obesity. Nat Genet 2008; 40: 1354–1359. | Article | PubMed | ChemPort |
  13. Wagener A, Schmitt AO, Aksu S, Schlote W, Neuschl C, Brockmann GA. Genetic, sex, and diet effects on body weight and obesity in the Berlin Fat Mouse Inbred lines. Physiol Genomics 2006; 27: 264–270. | Article | PubMed | ChemPort |
  14. Meyer CW, Wagener A, Rink N, Hantschel C, Heldmaier G, Klingenspor M et al. High energy digestion efficiency and altered lipid metabolism contribute to obesity in BFMI mice. Obesity (Silver Spring) 2009; 17: 1988–1993. | Article | PubMed | ChemPort |
  15. Taicher GZ, Tinsley FC, Reiderman A, Heiman ML. Quantitative magnetic resonance (QMR) method for bone and whole-body-composition analysis. Anal Bioanal Chem 2003; 377: 990–1002. | Article | PubMed | ChemPort |
  16. Tinsley FC, Taicher GZ, Heiman ML. Evaluation of a quantitative magnetic resonance method for mouse whole body composition analysis. Obes Res 2004; 12: 150–160. | Article | PubMed
  17. Bevova MR, Aulchenko YS, Aksu S, Renne U, Brockmann GA. Chromosome-wise dissection of the genome of the extremely big mouse line DU6i. Genetics 2006; 172: 401–410. | Article | PubMed | ISI | ChemPort |
  18. Stam P. Construction of integrated genetic linkage maps by means of a new computer package: Join Map. Plant J 1993; 3: 739–744. | Article | ChemPort |
  19. Shifman S, Bell JT, Copley RR, Taylor MS, Williams RW, Mott R et al. A high-resolution single nucleotide polymorphism genetic map of the mouse genome. PLoS Biol 2006; 4: e395. | Article | PubMed | ChemPort |
  20. Cox A, Ackert-Bicknell CL, Dumont BL, Ding Y, Bell JT, Brockmann GA et al. A new standard genetic map for the laboratory mouse. Genetics 2009; 182: 1335–1344. | Article | PubMed | ChemPort |
  21. Seaton G, Hernandez J, Grunchec JA, White I, Allen J, de Koning DJ et al. GridQTL: A grid portal for QTL mapping of compute intensive datasets. World Congress on Genetics Applied to Livestock Production. Proceedings of the 8th World Congress on Genetics Applied to Livestock Production, Abstract No. 27_633-916. 2006. Belo Horizonte, Brazil.
  22. Haley CS, Knott SA. A simple regression method for mapping quantitative trait loci in line crosses using flanking markers. Heredity 1992; 69: 315–324. | PubMed | ISI | ChemPort |
  23. Li R, Lyons MA, Wittenburg H, Paigen B, Churchill GA. Combining data from multiple inbred line crosses improves the power and resolution of quantitative trait loci mapping. Genetics 2005; 169: 1699–1709. | Article | PubMed | ISI | ChemPort |
  24. Carlborg O, Brockmann GA, Haley CS. Simultaneous mapping of epistatic QTL in DU6i x DBA/2 mice. Mamm Genome 2005; 16: 481–494. | Article | PubMed | ChemPort |
  25. Brockmann GA, Karatayli E, Neuschl C, Stylianou IM, Aksu S, Ludwig A et al. Genetic control of lipids in the mouse cross DU6i x DBA/2. Mamm Genome 2007; 18: 757–766. | Article | PubMed | ChemPort |
  26. Churchill GA, Doerge RW. Empirical threshold values for quantitative trait mapping. Genetics 1994; 138: 963–971. | PubMed | ISI | ChemPort |
  27. Visscher PM, Thompson R, Haley CS. Confidence intervals in QTL mapping by bootstrapping. Genetics 1996; 143: 1013–1020. | PubMed | ISI | ChemPort |
  28. Reed DR, Bachmanov AA, Tordoff MG. Forty mouse strain survey of body composition. Physiol Behav 2007; 91: 593–600. | Article | PubMed | ChemPort |
  29. Svenson KL, Von SR, Magnani PA, Suetin HR, Paigen B, Naggert JK et al. Multiple trait measurements in 43 inbred mouse strains capture the phenotypic diversity characteristic of human populations. J Appl Physiol 2007; 102: 2369–2378. | Article | PubMed | ChemPort |
  30. Hill WG, Goddard ME, Visscher PM. Data and theory point to mainly additive genetic variance for complex traits. PLoS Genet 2008; 4: e1000008. | Article | PubMed | ChemPort |
  31. Hager R, Cheverud JM, Wolf JB. Relative contribution of additive, dominance and imprinting effects to phenotypic variation in body size and growth between divergent selection lines of mice. Evolution 2009; 63: 1118–1128. | Article | PubMed
  32. Brockmann GA, Kratzsch J, Haley CS, Renne U, Schwerin M, Karle S. Single QTL effects, epistasis, and pleiotropy account for two-thirds of the phenotypic F2 variance of growth and obesity in DU6i x DBA/2 mice. Genome Res 2000; 10: 1941–1957. | Article | PubMed | ISI | ChemPort |
  33. Shao H, Reed DR, Tordoff MG. Genetic loci affecting body weight and fatness in a C57BL/6J x PWK/PhJ mouse intercross. Mamm Genome 2007; 18: 839–851. | Article | PubMed
  34. Lewis CE, North KE, Arnett D, Borecki IB, Coon H, Ellison RC et al. Sex-specific findings from a genome-wide linkage analysis of human fatness in non-Hispanic whites and African Americans: the HyperGEN study. Int J Obes (Lond) 2005; 29: 639–649. | Article | PubMed | ChemPort |
  35. Luke A, Wu X, Zhu X, Kan D, Su Y, Cooper R. Linkage for BMI at 3q27 region confirmed in an African-American population. Diabetes 2003; 52: 1284–1287. | Article | PubMed | ISI | ChemPort |
  36. Wu X, Cooper RS, Borecki I, Hanis C, Bray M, Lewis CE et al. A combined analysis of genomewide linkage scans for body mass index from the National Heart, Lung, and Blood Institute Family Blood Pressure Program. Am J Hum Genet 2002; 70/5: 1247–1256.
  37. Zhu X, Cooper RS, Luke A, Chen G, Wu X, Kan D et al. A genome-wide scan for obesity in African-Americans. Diabetes 2002; 51: 541–544. | Article | PubMed | ChemPort |
  38. Kissebah AH, Sonnenberg GE, Myklebust J, Goldstein M, Broman K, James RG et al. Quantitative trait loci on chromosomes 3 and 17 influence phenotypes of the metabolic syndrome. Proc Natl Acad Sci USA 2000; 97: 14478–14483. | Article | PubMed | ChemPort |
  39. Rankinen T, Zuberi A, Chagnon YC, Weisnagel SJ, Argyropoulos G, Walts B et al. The human obesity gene map: the 2005 update. Obesity (Silver Spring) 2006; 14: 529–644. | Article | PubMed
  40. Rocha JL, Eisen EJ, Van Vleck LD, Pomp D. A large-sample QTL study in mice: II. Body composition. Mamm Genome 2004; 15: 100–113. | Article | PubMed | ISI | ChemPort |
  41. Brockmann GA, Haley CS, Renne U, Knott SA, Schwerin M. Quantitative trait loci affecting body weight and fatness from a mouse line selected for extreme high growth. Genetics 1998; 150: 369–381. | PubMed | ISI | ChemPort |
  42. Farber CR, Corva PM, Medrano JF. Genome-wide isolation of growth and obesity QTL using mouse speed congenic strains. BMC Genomics 2006; 7: 102. | Article | PubMed | ChemPort |
  43. Farber CR, Medrano JF. Fine mapping reveals sex bias in quantitative trait loci affecting growth, skeletal size and obesity-related traits on mouse chromosomes 2 and 11. Genetics 2007; 175: 349–360. | Article | PubMed
  44. Wang S, Yehya N, Schadt EE, Wang H, Drake TA, Lusis AJ. Genetic and genomic analysis of a fat mass trait with complex inheritance reveals marked sex specificity. PLoS Genet 2006; 2: e15. | Article | PubMed | ChemPort |
  45. Fawcett GL, Roseman CC, Jarvis JP, Wang B, Wolf JB, Cheverud JM. Genetic architecture of adiposity and organ weight using combined generation QTL analysis. Obesity (Silver Spring) 2008; 16: 1861–1868. | Article | PubMed | ChemPort |


We thank Wenhua Wei and Anna Wolc for their efforts and valuable comments. Furthermore, we appreciate the help of Ralf Bortfeldt and Mark Kendell Clement in bioinformatical issues. This research was supported by the German National Genome Research Network (NGFN Plus (01GS0829) and by the German Research Foundation (DFG) Graduate College 1208 ‘Hormonal Regulation of Energy Metabolism, Body Weight and Growth’.

Supplementary Information accompanies the paper on International Journal of Obesity website

Extra navigation