Grant A. Mitchell

Macrophage infiltration of white adipose tissue (WAT) is implicated in the metabolic complications of obesity. The precipitating event(s) and function(s) of macrophage infiltration into WAT are unknown. We demonstrate that >90% of all macrophages in WAT of obese mice and humans are localized to dead adipocytes, where they fuse to form syncytia that sequester and scavenge the residual "free" adipocyte lipid droplet and ultimately form multinucleate giant cells, a hallmark of chronic inflammation. Adipocyte death increases in obese (db/db) mice (30-fold) and humans and exhibits ultrastructural features of necrosis (but not apoptosis). These observations identify necrotic-like adipocyte death as a pathologic hallmark of obesity and suggest that scavenging of adipocyte debris is an important function of WAT macrophages in obese individuals. The frequency of adipocyte death is positively correlated with increased adipocyte size in obese mice and humans and in hormone-sensitive lipase-deficient (HSL-/-) mice, a model of adipocyte hypertrophy without increased adipose mass. WAT of HSL-/- mice exhibited a 15-fold increase in necrotic-like adipocyte death and formation of macrophage syncytia, coincident with increased tumor necrosis factor-alpha gene expression. These results provide a novel framework for understanding macrophage recruitment, function, and persistence in WAT of obese individuals.

Identifying the genes responsible for human diseases requires combining information about gene position with clues about biological function. The recent availability of whole-genome data sets of RNA and protein expression provides powerful new sources of functional insight. Here we illustrate how such data sets can expedite disease-gene discovery, by using them to identify the gene causing Leigh syndrome, French-Canadian type (LSFC, Online Mendelian Inheritance in Man no. 220111), a human cytochrome c oxidase deficiency that maps to chromosome 2p16-21. Using four public RNA expression data sets, we assigned to all human genes a "score" reflecting their similarity in RNA-expression profiles to known mitochondrial genes. Using a large survey of organellar proteomics, we similarly classified human genes according to the likelihood of their protein product being associated with the mitochondrion. By intersecting this information with the relevant genomic region, we identified a single clear candidate gene, LRPPRC. Resequencing identified two mutations on two independent haplotypes, providing definitive genetic proof that LRPPRC indeed causes LSFC. LRPPRC encodes an mRNA-binding protein likely involved with mtDNA transcript processing, suggesting an additional mechanism of mitochondrial pathophysiology. Similar strategies to integrate diverse genomic information can be applied likewise to other disease pathways and will become increasingly powerful with the growing wealth of diverse, functional genomics data.