The dominant paradigm of protein engineering is structurebased site-directed mutagenesis. This rational approach is generally more effective for the engineering of local properties, such as substrate specificity, than global ones such as allostery. Previous workers have modified normally unregulated reporter enzymes, including -galactosidase, alkaline phosphatase, and -lactamase, so that the engineered versions are activated (up to 4-fold) by monoclonal antibodies. A reporter that could easily be "reprogrammed" for the facile detection of novel effectors (binding or modifying activities) would be useful in high throughput screens for directed evolution or drug discovery. Here we describe a straightforward and general solution to this potentially difficult design problem. The transcription factor p53 is normally regulated by a variety of post-translational modifications. The insertion of peptides into intrinsically unstructured domains of p53 generated variants that were activated up to 100-fold by novel effectors (proteases or antibodies). An engineered p53 was incorporated into an existing high throughput screen for the detection of human immunodeficiency virus protease, an arbitrarily chosen novel effector. These results suggest that the molecular recognition properties of intrinsically unstructured proteins are relatively easy to engineer and that the absence of crystal structures should not deter the rational engineering of this class of proteins.Cells generally employ sensor proteins (also called "biosensors" or "switches") to detect chemical stimuli and activate downstream components of signal transduction systems. We sought to fabricate artificial molecular sensors by engineering proteins that are specifically activated when bound or modified by novel effectors. Such sensors have practical utility in high throughput screens for drug discovery or directed protein evolution. They have also proved to be useful as research reagents. For example, two-hybrid systems (1, 2) and protein fragment complementation assays (3) couple the interactions of fusion proteins within transgenic cells to the production of signals. Cleverly designed sensors based upon fluorescent resonance energy transfer between green fluorescent protein analogues have also enabled the observation of intracellular protein modification events (4), including protein phosphorylation (5) and proteolysis (6). We expect that the utility of engineered protein sensors will continue to increase as they are deployed as diagnostic reagents (7) and pathogen-activated biotherapeutics (8, 9).Rational protein design is generally synonymous with structure based site-directed mutagenesis (10). Reporter proteins are usually selected as starting points for sensor design because their structurehave been solved and because their activities are amenable to high throughput screening. Previous workers have inserted peptide epitopes into -galactosidase (11), alkaline phosphatase (12), or -lactamase (13). This approach has generally produced catalytically compromis...