| Bioluminescence imaging is utilized to study a wide variety of biochemical processes and signal transduction pathways, generally by determination of total photon flux from a luciferase to characterize pathways of interest in real time. The purpose of this dissertation is to exploit not only the raw photon output from each enzyme, but also bioluminescent spectra for analysis of biochemical pathways in live cells.; To this end, a Java program was written for use with Image J to deconvolute the bioluminescent signals from multiple luciferases simultaneously and in real time. The program was validated by both simulated bioluminescence experiments, and actual bioluminescence experiments. While the program can in principle deconvolute unlimited number of luciferases, in practice, up to three commonly used luciferases could be readily deconvoluted. The utility of this methodology was demonstrated by measuring a ∼5-fold increase in precision of measurements of I kappa B kinase (IKK) activity in living cells and by demonstrating the feasibility of high throughput screening (HTS). Control reporters generated Z' values acceptable for HTS and signal to noise ratios ranging from 68 to 320.; Furthermore, detecting spectral shifts of luciferases through bioluminescence resonance energy transfer (BRET) were demonstrated to be both predictable and useful. The Forster resonance energy transfer equation was modified from first principles and adapted for BRET experiments. The predictive power of the equation was tested in bacteria by constructing five lux A-fluorescent protein fusion pairs: luxA-eGFP, luxA-mRFP, luxA-mCherry, luxA-tdTomato, and luxA-mPlum. The generalizability of this approach was further validated in mammalian cells by constructing and testing three click beetle green (CBG) luciferase/fluorophore fusion constructs: CBG-mRFP, CBG-tdTomato, and CBG-mPlum. The modified Forster equation fit the data in both bacterial and mammalian cells (R2 > 0.95; p < 0.05). CBG-DEVD-tdTomato, a high precision DEVDase (caspase) reporter based upon the most efficient luciferase/fluorophore pair, was validated as a real time protease biosensor. At peak reduction of BRET (12 hours post treatment with 10 muM doxorubicin) the signal to noise ratio was ∼33. Thus, exploiting the spectral properties of luciferase provides multiple novel opportunities for investigation in cellulo of signal transduction pathways. |
