Biochemists have uncovered how certain small molecules, including nonsteroidal anti-inflammatory drugs such as the ibuprofen analog tarenflurbil, reduce production of a form of amyloid-β peptide, the plaque-forming protein associated with Alzheimer's disease (Nature 2008, 453, 925). In addition to potentially leading to a new family of drugs to combat Alzheimer's and related disorders, the finding suggests a new tactic for general drug-screening efforts. A form of amyloid-β peptide known as Aβ42 is created when the protease enzyme γ-secretase cuts up β-amyloid precursor protein (APP). Small-molecule γ-secretase modulators (GSMs) such as tarenflurbil are known to influence the enzyme's activity, but little was known about how they work until now. A study led by Thomas L. Kukar and Todd E. Golde at the Mayo Clinic, Jacksonville, Fla., shows that some GSMs directly target the APP substrate to limit Aβ42 production and stop proteins from aggregating into plaques. That mechanism is at odds with the usual process of targeting the enzyme. "This is an astonishing result," given that there are "precious few examples of substrate-targeted enzyme inhibitors," says Thomas J. Kodadek of the University of Texas Southwestern Medical Center in an accompanying Nature commentary. If the mechanism proves to be relevant to other proteases, then "drug-screening efforts should not just seek compounds that bind to proteases themselves, but also those that bind to their substrates," Kodadek adds.