As a single polypeptide, cytochrome P450 BM3 fuses oxidase and reductase domains and couples each domain's function to perform catalysis with exceptional activity upon binding of substrate for hydroxylation. Mutations introduced into the enzyme to change its substrate specificity often decrease coupling efficiency between the two domains, resulting in unproductive consumption of cofactors and formation of water and/or reactive species. This phenomenon can correlate with leakage, in which P450 BM3 uses electrons from NADPH to reduce oxygen to water and/or reactive species even without bound substrate. The physical basis for leakage is not yet well understood in this particular member of the cytochrome P450 family. To clarify the relationship between leakage and coupling, we used simulations to illustrate how different combinations of kinetic parameters related to substrate-free consumption of NADPH and substrate hydroxylation can lead to either minimal effects on coupling or a dramatic decrease in coupling as a result of leakage. We explored leakage in P450 BM3 by introducing leakage-enhancing mutations and combining these mutations to assess whether doing so increases leakage further. The variants in this study provide evidence that while a transition to high spin may be vital for coupled hydroxylation, it is not required for enhanced leakage; substrate binding and the consequent shift in spin state are not necessary as a redox switch for catalytic oxidation of NADPH. Additionally, the variants in this study suggest a tradeoff between leakage and stability and thus evolvability, as the mutations we Abbreviations: P450 BM3, cytochrome P450 monooxygenase from B. megaterium; P450 cam, cytochrome P450 monooxygenase from P. putida; WT, wild type; k cat , R-H, turnover rate constant with respect to hydroxylation; K m , R-H, Michaelis constant with respect to hydroxylation; k cat , leakage, turnover rate constant with respect to leakage; K m , leakage, Michaelis constant with respect to leakage; K d , dissociation equilibrium constant; k on , association rate constant; k off , dissociation rate constant; T 50 , incubation temperature at which half of protein becomes denatured after 10 min; NADPH, nicotinamide adenine dinucleotide phosphate; ROS, reactive oxygen species; H 2 O 2 , hydrogen peroxide; O 2 2 , superoxide; IPTG, isopropyl b-d-1-thiogalactopyranoside; KPi, potassium phosphate; HRP, horseradish peroxidase; ABTS, 2,2 0 -azino-bis(3-ethylbenzothiazoline-6-sulphonic acid); CO, carbon monoxide; Tris, tris(hydroxyl)aminomethane; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis Additional Supporting Information may be found in the online version of this article. Published by Wiley-Blackwell. V C 2015 The Protein Society investigated were far more deleterious than other mutations that have been used to change substrate specificity.

Highlights• We characterized the signal of HyPer, a fluorescent probe for peroxide, in E. coli.• Each strain requires its own specific signal characterization.• HyPer's signal is reversible rather than real-time.• Expression of HyPer reduced the rate of peroxide scavenging by the expression host.• Careful, controlled use of HyPer facilitates quantitative comparisons across studies. AbstractGenetically encoded, fluorescent biosensors have been developed to probe the activities of various signaling molecules inside cells ranging from changes in intracellular ion concentrations to dynamics of lipid second messengers. HyPer is a member of this class of biosensors and is the first to dynamically respond to hydrogen peroxide (H 2 O 2 ), a reactive oxygen species that functions as a signaling molecule. However, detailed characterization of HyPer's signal is not currently available within the context of bacteria exposed to external oxidative stress, which occurs in the immunological response of higher organisms against invasive pathogenic bacteria. Here, we performed this characterization, specifically in Escherichia coli exposed to external H 2 O 2 . We found that the temporal behavior of the signal does not correspond exactly to peroxide concentration in the system as a function of time and expression of the sensor decreases the peroxide scavenging activity of the cell. We also determined the effects of cell number, both before and after normalization of externally added H 2 O 2 to the number of cells. Finally, we report quantitative characteristics of HyPer's signal in this context, including the dynamic range of the signal, the signal-to-noise ratio, and the half saturation constant. These parameters show statistically meaningful differences in signal between two commonly used strains of E. coli, demonstrating how signal can vary with strain. Taken together, our results establish a systematic, quantitative framework for researchers seeking to better understand the role of H 2 O 2 in the immunological response against bacteria, and for understanding potential differences in the details of HyPer's quantitative performance across studies.