PLoS ONE (Jan 2011)

A stratified transcriptomics analysis of polygenic fat and lean mouse adipose tissues identifies novel candidate obesity genes.

  • Nicholas M Morton,
  • Yvonne B Nelson,
  • Zoi Michailidou,
  • Emma M Di Rollo,
  • Lynne Ramage,
  • Patrick W F Hadoke,
  • Jonathan R Seckl,
  • Lutz Bunger,
  • Simon Horvat,
  • Christopher J Kenyon,
  • Donald R Dunbar

DOI
https://doi.org/10.1371/journal.pone.0023944
Journal volume & issue
Vol. 6, no. 9
p. e23944

Abstract

Read online

BACKGROUND:Obesity and metabolic syndrome results from a complex interaction between genetic and environmental factors. In addition to brain-regulated processes, recent genome wide association studies have indicated that genes highly expressed in adipose tissue affect the distribution and function of fat and thus contribute to obesity. Using a stratified transcriptome gene enrichment approach we attempted to identify adipose tissue-specific obesity genes in the unique polygenic Fat (F) mouse strain generated by selective breeding over 60 generations for divergent adiposity from a comparator Lean (L) strain. RESULTS:To enrich for adipose tissue obesity genes a 'snap-shot' pooled-sample transcriptome comparison of key fat depots and non adipose tissues (muscle, liver, kidney) was performed. Known obesity quantitative trait loci (QTL) information for the model allowed us to further filter genes for increased likelihood of being causal or secondary for obesity. This successfully identified several genes previously linked to obesity (C1qr1, and Np3r) as positional QTL candidate genes elevated specifically in F line adipose tissue. A number of novel obesity candidate genes were also identified (Thbs1, Ppp1r3d, Tmepai, Trp53inp2, Ttc7b, Tuba1a, Fgf13, Fmr) that have inferred roles in fat cell function. Quantitative microarray analysis was then applied to the most phenotypically divergent adipose depot after exaggerating F and L strain differences with chronic high fat feeding which revealed a distinct gene expression profile of line, fat depot and diet-responsive inflammatory, angiogenic and metabolic pathways. Selected candidate genes Npr3 and Thbs1, as well as Gys2, a non-QTL gene that otherwise passed our enrichment criteria were characterised, revealing novel functional effects consistent with a contribution to obesity. CONCLUSIONS:A focussed candidate gene enrichment strategy in the unique F and L model has identified novel adipose tissue-enriched genes contributing to obesity.