Characterization of a novel obesity phenotype caused by interspecific hybridization

Abstract

Interspecific hybridization in mammals causes hybrid dysgenesis effects, such as sterility and abnormal placentation. Here, we describe a novel obesity syndrome caused by interspecific hybridization in the genus Mus and show that this obesity, appearing sporadically in F1 littermates derived from inbred strains, has an epigenetic basis. Mus hybrids from various strains of M. musculus and M. spretus were generated and the sporadic obese phenotype was confirmed through assessment of physiological and biochemical parameters in littermates. To understand the underlying mechanisms, large-scale and candidate gene expression assays, global DNA methylation assays and allelic expression analysis were performed. Studies showed that obese hybrids are similar to other known models of obesity. While increased axial growth indicated a defect in POMC pathway, comparison of global gene expression patterns in brain of obese F1 and obese Pomc mutant mice showed little similarity. In F1 obese mice many genes involved in the maintenance of epigenetic states, as well as several imprinted genes, were differentially expressed. Global DNA methylation analysis in brain showed that increased methylation levels were associated with obesity. The imprinted gene Gnasxl, known to be important in lipid homeostasis, was found over expressed in the obese hybrids. Allelic expression and methylation analysis of Gnasxl showed that alterations of epigenetic marks underlying F1 obesity are probably many and multi-factorial. Conclusions: This model of obesity, which is both spontaneous and epigenetic, may be a useful tool to address the epigenetic aspects of clinical obesity.

Key words:

• Lipid homeostasis
• epigenetic
• imprinting
• DNA methylation

Acknowledgements

We are grateful to Dr. Ute Hochgeschwender and Peggy Hunnewell for giving us POMC mouse tissues. This work was supported by grants from the Swedish Research Council (RHF), the Wallenberg Consortium North (RF, TJE), the Swedish Cancer Foundation (TJE), the Swedish AFA foundation (TJE), the UK Biotechnology and Biological Sciences Research Council and the Medical Research Council (GK, AP). MK is supported by a scholarship from the Ministry of Health of the I.R. Iran. This paper is ISE-M contribution N° 2008-065.

Appendix

Supplementary Figure 1.  Volcano plot for all the reporters in the six micro-array hybridizations performed on pairs (1), (2) and (3). The X axis represents M value and the Y axis represents B value for each reporter.

Supplementary Table I.  List of genes deregulated in F1 obese brain compared to F1 lean brain. Commonality refers to occurrence of same EST in Supplementary Table III. M value is a very commonly used standard parameter in micro-array statistics, calculated as logarithm to the base 2 of the ratio of expression in test (obese) upon control (lean litter mates). Calculating 2 raised to the power of the M value will give the folds change. All positive M values are increase in expression and all negative values are decrease in expression in obese brains compared to lean controls. B values are suited for probability calculation in micro-array results with small sample size as p values with such sample sizes will be misleading. Higher B values represent higher probability of being differentially expressed.

Supplementary Table II.  Functional classes, their Gene Ontology definitions and distribution of ESTs deregulated in F1 obese brain (refer to Supplementary Table I).

Supplementary Table III.  List of genes deregulated in Pomc^−/− brain compared to wild-type controls. Commonality refers to occurrence of same EST in Supplementary Table I. For details of M and B values, please see caption to Supplementary Table I.

Supplementary Table IV.  CpG islands exhibiting altered methylation in obese F1 brain as compared to those of lean controls. For details of M and B values, please see caption to supplementary Table I. Distance between CpG islands and genes is calculated as the minimum number of bases between the end of CpG islands and the transcription start sites of the genes.

Supplementary Table V.  Primers and reaction conditions for real-time expression analysis of different genes.