Mice lacking both fat carriers (AM−/−) don't get fat on a high-fat diet.


The two related fatty acid binding proteins (FABPs) called aP2 and mal1 did not seem destined for fame. “These are seemingly very dull molecules,” says Gökhan Hotamisligil (Harvard School of Public Health, Boston, MA). “They are like little clams attaching to fatty acids.” Yet somehow, he says, these “idle chaperone proteins,” which shuttle fatty acids around inside the cell, turn out to “determine lipid metabolism.” They may do so by bringing fatty acid species to regulatory proteins or enzymes.

Mice lacking one or the other FABP had been generated before by Hotamisligil and shown mild phenotypes. Now the Boston group, including Kazuhisa Maeda and Haiming Cao, characterizes mice lacking both proteins. The mice have less fat than normal, and after a high fat diet their body composition, blood glucose, blood insulin, and insulin resistance do not worsen but are all indistinguishable from those of wildtype mice on a normal diet.

The proximal cause of these changes may be an unusual lipid profile. The double mutants have more shorter chain fatty acids in adipocytes and muscle cells, and more longer chain fatty acids in liver. The results, both in the mutants and in vitro after similar manipulations of lipid content, include increased activating phosphorylations of both AMP-activated kinase (AMP-K) and members of the insulin receptor pathway in muscle. AMP-K can induce greater fatty acid oxidation and energy expenditure, as is seen in the mutant mice. And an activated insulin receptor pathway would protect animals from insulin resistance.

FABP acts predominantly in adipose tissue, so the effects on other tissues must be mediated by other messengers. Hotamisligil is pursuing these messengers using cocultures. His “larger aspiration,” meanwhile, is to “tie FABP to a specific biological pathway.”

This may involve tracking down lipid mediators, which is notoriously difficult, even with the protein now in hand. The altered profile of fatty acids is also not yielding answers easily. The pattern seen in mutant adipocytes and muscle “has not been noted as a signature pattern in any other state” such as fasting or feeding, says Hotamisligil, so the significance is not immediately clear. He hopes the answer will come from large-scale “lipomics” efforts to identify lipid signatures at specific sites and under specific dietary conditions.


Maeda, K., et al.
Cell Metab.