Medicine has a long tradition of pragmatic empiricism, in which one does what works, regardless of understanding why it works. In the era of rational drug design, however, it is a surprise when new therapeutics make it through clinical trials without knowledge of their molecular targets. But such was the case for the cholesterol-lowering agent ezetimibe (Zetia; Merck/Schering-Plough).

Ezetimibe was approved by the US FDA in November 2002 for the treatment of hypercholesterolaemia (alone and in combination with statins, which inhibit cholesterol synthesis). At the time, its mechanism of action was unknown. A recent paper in Science by Scott Altmann and colleagues from the Schering-Plough Research Institute now sheds light on this enigma by identifying a molecular pathway that ezetimibe affects.

Serum cholesterol levels are principally controlled by two organs: the liver, which synthesizes cholesterol, and the intestine, which absorbs dietary cholesterol. Acyl-coenzyme A:cholesterol acyltransferase (ACAT) is an enzyme involved in cholesterol absorption in the intestine, a process that is poorly understood. Schering-Plough, targeting ACAT for the development of a cholesterol-lowering drug, identified a class of compounds that inhibited cholesterol absorption in a cholesterol-fed hamster model. Experimental data, however, indicated that these agents acted upstream of ACAT. Nonetheless, ezetimibe emerged as a potent cholesterol-lowering agent.

So, if ACAT is not the target of ezetimibe, what is? To address this question, Altmann and colleagues carried out a genomic search to identify, from intestinal cells, rat transcripts that possessed features of a cholesterol transporter. Only one credible candidate emerged from this search: the rat homologue of the gene encoding human Niemann–Pick C1-like 1 (NPC1L1) protein. Analysis of the tissue distribution of NPC1L1 showed that expression is enriched in the small intestine.

The authors then investigated the in vivo role of NPC1L1 using mice engineered to lack the protein. Although otherwise phenotypically normal, knockout mice absorbed just 31% of the amount of orally administered cholesterol that mice with functional NPC1L1 absorbed, indicating that the protein has a key role in cholesterol absorption from the intestine. And when the authors tested the effects of ezetimibe in normal and NPC1L1-knockout mice, they found that it reduced cholesterol absorption in mice with functional NPC1L1 to a level similar to that in untreated mice that lacked NPC1L1. NPC1L1-deficient mice, however, were completely insensitive to the drug. These data indicate that ezetimibe acts on an NPC1L1-containing pathway, and although the group was unsuccessful in detecting binding of ezetimibe to NPC1L1, it seems that the secrets of this drug's action are at last being revealed.