Quantitative Comparison of the Sensitivity of Detection of Fluorescent and Bioluminescent Reporters in Animal Models

Bioluminescent and fluorescent reporters are finding increased use in optical molecular imaging in small animals. In the work presented here, issues related to the sensitivity of in vivo detection are examined for standard reporters. A high-sensitivity imaging system that can detect steady-state emission from both bioluminescent and fluorescent reporters is described. The instrument is absolutely calibrated so that animal images can be analyzed in physical units of radiance allowing more quantitative comparisons to be performed. Background emission from mouse tissue, called autoluminescence and autofluorescence, is measured and found to be an important limitation to detection sensitivity of reporters. Measurements of dual-labeled (bioluminescent/fluorescent) reporter systems, including PC-3M-luc/DsRed2-1 and HeLa-luc/PKH26, are shown. The results indicate that although fluorescent signals are generally brighter than bioluminescent signals, the very low autoluminescent levels usually results in superior signal to background ratios for bioluminescent imaging, particularly compared with fluorescent imaging in the green to red part of the spectrum. Fluorescence detection sensitivity improves in the far-red to near-infrared, provided the animals are fed a low-chlorophyll diet to reduce autofluorescence in the intestinal region. The use of blue-shifted excitation filters is explored as a method to subtract out tissue autofluorescence and improve the sensitivity of fluorescent imaging.