Excess accumulation of incompletely oxidized non-esterified FAs (FAs) in muscle cells is increasingly appreciated to be a critical causal event in the development of insulin resistance (IR). In parallel, branched chain amino acids (BCAAs) have recently moved front and center in the field of diabetes, as unbiased metabolomic profiling in large prospective epidemiological studies have shown that serum elevations in BCAAs predict insulin resistance and diabetes as much as 20 years prior to clinical presentation. We have uncovered a novel molecular pathway that links these two observations. We find that active catabolism of the BCAA valine in skeletal muscle causes the paracrine secretion of metabolites that promote entry of FAs into skeletal muscle, and subsequent lipotoxicity. The identification of this pathway makes numerous predictions, three of which have especially strong translational potential and thus form the object of this proposal: 1. BCAA breakdown, rather than accumulation, and specifically in muscle, causes insulin resistance. 2. Specifically Valine breakdown is causal, whereas for example leucine may be protective. 3. Therapeutic shunting of valine catabolism away from muscle should improve insulin resistance. 4. Treating human subjects with valine, in the presence of excess fat, should trigger insulin resistance. We propose experiments to test these predictions in detail. Lipid trafficking in and out of tissues is at the heart of diabetes, and yet its relationship to trafficking amino acids, and especially the abundant BCAAs, is little understood. The proposed work thus stands to identify a novel targets for the treatment of diabetes.