PERSPECTIVEhomology in the active site. These data would point a medicinal chemist to expect that it should be relatively easy to gain selectivity between CPT2 and CPT1 inhibitors, while it might prove to be a daunting task for a competitive inhibitor to differentiate CPT1A from CPT1B. An alternative option is to achieve tissue selective distribution of an unspecific inhibitor via permeability and pharmacodynamic properties.The homology between human, rat, and mouse enzymes, on the other hand, is rather high (see Table 3b), and species differences at the in vitro level are not likely to be significant (the rate of FAO in different species and the level of CPT activity control can be considerably different). 22 CPT1 is subject to multiple controlling factors. Expression level and basal activity of CPT1 depend on fasting state, 23 exercise, 24 type of diet, 25 exposure to cold, infections, metabolic disease, and enzyme inhibition. 9 Diabetes does not affect CPT1B activity 26 but increases CPT1A activity. 27 Malonyl-CoA, the first committed step of fatty acid synthesis, whose concentration is in turn highly regulated by multiple mechanisms, is a potent inhibitor of CPT1B (∼0.03 μM IC 50 for the rat enzyme) and a less potent inhibitor (by about 2 orders of magnitude) of CPT1A. 28,29 The sensitivity of CPT1A to allosteric inhibition by malonyl-CoA varies under different physiological conditions, amplifying the effect of the cellular malonyl-CoA concentration. In particular, declining malonyl-CoA concentrations reduce the sensitivity of the enzyme to allosteric inhibition (and vice versa). Mitochondrial outer membrane composition also affects sensitivity of CPT1A to malonyl-CoA. Studies regarding the binding of malonyl-CoA and other active-site directed inhibitors have shown that there are two separate binding sites for malonyl-CoA: a high-affinity binding site located on the cytoplasmic side of the protein and a second low-affinity site that corresponds to the catalytic site and where also CoA exerts a product-inhibition action. 30 It is postulated that one of these binding sites is a contact interaction between the N-and C-terminal segments of the protein, which explains the sensitivity to membrane fluidity as well as the loss of sensitivity to malonyl-CoA if the N-terminal portion of the protein is deleted (which also leads to loss of function) or by specific single point mutations in this region. The N-terminal domain is also responsible for the targeting of the enzyme to the outer mitochondrial membrane. 31 There is evidence that CPT activity is controlled by phosphorylation 32 and nitration, 33 although the hypothesis has been advanced that CPT is constitutively phosphorylated. 28