Whole animal bioluminescence imaging (BLI) provides a powerful technological means to observe physiological events inside living systems with minimal perturbation [1–5]. It therefore has utility in studies on, for example, tumor development [3, 6], immunity [4], infection [7], gene expression [8], and inflammation [9]. However, a critical challenge for in vivo BLI comes from an abundance of endogenous optical absorbers, such as hemoglobin [2, 10, 11]. Hemoglobin readily absorbs blue-yellow light (400–650 nm); strongly overlapping the most commonly used bioluminescent sources (i.e., Photorhabdus luminescens, Gaussia, and even firefly luciferase to some extent). Consequently, the light output of luminescent probes can be quenched, and/or scattered by tissues, eventually resulting in a reduced amount of detectable signal. Likewise, these same constraints are encountered under in vitro conditions using animal extracts such as blood or homogenized tissue. Because biological tissue is more permissive to red light, much effort has been directed towards improving BLI sensitivity by red-shifting the emission wavelengths to the desired optical window of 650–900 nm, a range in which optical absorption is minimized significantly [12]. To achieve this, researchers have tried either to create mutations of the common luciferases (for example, the work by Branchini et al.) [12], or by proxy using bioluminescence resonance energy transfer (BRET) [13–15].

Unlike classical fluorescence excitation, where excitation light radiatively excites a fluorophore, BRET is a process by which the photon energy of a luminescent “donor” molecule can pass directly (via a non-radiative resonance energy transfer, RET) to a fluorescent “acceptor” molecule within 10 nm proximity, resulting in fluorescence. The Stokes shift of the fluorophore yields fluorescence emission at a red-shifted, longer wavelength than the original luminescence. Recently we reported an alternative method to achieve a similar red shifted emission by using Fluorescence by Unbound Excitation from Luminescence (FUEL) [2, 10]. The FUEL method differs from BRET because it arises from an irradiative epifluorescent excitation/emission interaction between the bioluminescent molecule and fluorophore. The experimental conditions favoring FUEL are very similar to those for BRET: a strong spectral overlap between the bioluminescent source and the fluorophore must exist. The major difference is the distance tolerance over which the phenomena may occur. The working distance of FUEL is much greater than 10 nm. Under in vitro conditions and in the absence of an optical absorber, the FUEL effect has been observed with up to 3 cm between the luminescent source and the fluorophore moieties, without covalent coupling chemistries or surface modifications of either of the FUEL components.

The differences between FUEL and BRET underline their distinct utility as experimental tools. While the use of BRET is well established in the literature, we propose that FUEL brings a different utility to the experimental toolbox for both in vitro and in vivo applications. For example, the possibility to measure entity co-proximity at a macroscopic level would be invaluable. Towards these ends, we present in detail simple and robust methods to reconstitute FUEL using bioluminescent Escherichia coli and quantum dots, under controlled in vitro and in vivo demonstrations. With the principles demonstrated here it should be possible to extrapolate the appropriate techniques for more complicated in vitro or in vivo applications thereof.