Nature Biotechnology 24, 326 - 328 (2006)
doi:10.1038/nbt0306-326
Self-illuminating quantum dots light the wayJohn V FrangioniJohn V. Frangioni is at the Beth Israel Deaconess Medical Center, Room SL-B05, 330 Brookline Avenue, Boston, Massachusetts 02215, USA. jfrangio@bidmc.harvard.edu A quantum dot decorated with luciferase molecules fluoresces without external illumination.In this issue, So et al.1 describe a clever quantum dot technology that permits improved imaging in vivo compared with existing quantum dots. The new probes rely on bioluminescence resonance energy transfer2, which converts chemical energy into photon energy, resulting in dramatic increases in fluorophore excitation and dramatic reductions in the effects of tissue autofluorescence. The technology has several exciting features. It eliminates the need for fluorescent excitation light and takes advantage of quantum dots' blue-increasing extinction coefficient. It greatly reduces the high background caused by surface illumination and tissue autofluorescence and exploits tissue-penetrating near-infrared wavelengths for fluorescence emission. It efficiently couples chemical energy with light energy. And, finally, because energy transfer is resonant, and nonradiative, absorption of excitation photons by hemoglobin and other tissue pigments is virtually eliminated.
Finding fluorescent objects in scattering tissue using light can be quite difficult. The use of near-infrared wavelengths (typically 700–900 nm) for excitation and emission helps to reduce absorption and scatter, and frequency- or time-domain techniques can be used to improve depth sensitivity even further (reviewed in refs. 3,
4,
5). However, even in the near-infrared range6, tissue autofluorescence always remains a major and often overlooked problem that reduces the signal-to-background ratio and, therefore, object detectability (Fig. 1).
 | | |
To improve fluorophore excitation and reduce the effects of tissue autofluorescence, So et al. covalently attached multiple molecules of Renilla reniformis luciferase to a single fluorescent semiconductor nanocrystal (that is, a quantum dot), forming a large conjugate >22 nm in hydrodynamic diameter. When luciferase binds its substrate coelenterazine, it emits broad-spectrum blue light peaking at 480 nm. So et al. show that the blue light emitted by the luciferase molecules in the conjugate excites the bound quantum dot. As the absorption spectrum of quantum dots increases exponentially towards the blue, the complete overlap of the luciferase emission and quantum dot absorption spectra provides extremely efficient fluorophore excitation without the addition of exogenous light.
The authors characterized this system in both in vitro and in vivo experiments, providing convincing evidence that bioluminescent quantum dots can be spectrally multiplexed, and that they outperform conventional quantum dot imaging. It should be emphasized that the magnitude of improvement of this technology over external excitation is dramatic. As detailed in the paper's online supplementary information, to excite quantum dots embedded 1 cm below scattering tissue equivalently, one would need 9W of 480-nm external excitation light, compared to a single injection of coelenterazine. Tissue autofluorescence from such high levels of excitation light would also be considerable, as would the inevitable excitation-light leakage through emission filters. This helps explain the authors' result that bioluminescent quantum dots embedded in 3 mm of scattering tissue were visible in vivo using bioluminescence resonance energy transfer, but not with external excitation.
Conventional quantum dots have the potential for significantly improved performance over organic fluorophores for in vivo imaging (reviewed in ref. 7). Their advantages include an exponentially increasing extinction coefficient towards the blue, potentially high quantum yield, precise tunability of emission wavelength, photostability, chemical stability and disease-specific targeting. At present, however, the widespread use of quantum dots in animal experiments and their eventual translation to the clinic for patient care are hindered by several fundamental problems.
First, all existing formulations of near-infrared quantum dots contain semiconductor or heavy metals with potential toxicity. Second, an organic coating is needed to render quantum dots soluble in aqueous environments. When the organic coating and its water of hydration are considered, the final hydrodynamic diameter of every quantum dot described to date is much larger than the renal filtration threshold. Hence, near-infrared quantum dots injected intravenously for cancer or other disease targeting cannot be cleared from the body. Indeed, secondary coatings, such as pegylation, often added to large quantum dots, merely delay the inevitable uptake by, and concentration in, the liver. Third, the potential advantage of increasing excitation towards the blue is actually lost in Rayleigh scattering tissue8. That is, the increasing scatter of tissue at bluer wavelengths offsets the increasing absorption of the quantum dot, resulting in net absorption similar to the single absorption peak of a conventional organic fluorophore.
Although the bioluminescent quantum dot described by So et al. solves the third fundamental problem, it still suffers from the first two, as well as several others. Imaging requires systemic administration of a foreign protein (luciferase) and a foreign enzyme substrate (coelenterazine), both of which have immunogenic potential and can alter the biological system under study. Light generation is completely dependent on the biodistribution of the enzyme substrate, and thus on the relative perfusion of the particular organ, tissue or tumor to be imaged. In addition, the large hydrodynamic diameter of the bioluminescent quantum dot will undoubtedly limit its ability to extravasate from the vasculature and thus target normal tissues or organs, and many tumors.
Nevertheless, bioluminescent quantum dot technology has the potential to greatly improve near-infrared fluorescence detection in living tissue. And it stimulates one to think about how the interconversion of energy from one form to another might solve major problems in the field of in vivo imaging.
REFERENCES
-
So, M.-K.
,
Xu, C.
,
Loening, A.M.
,
Gambhir, S.S.
&
Rao, J.
Nat. Biotechnol. 24, 339–343 (2006). | Article |
-
Xu, Y.
,
Piston, D.W.
&
Johnson, C.H.
Proc. Natl. Acad. Sci. USA 96, 151–156 (1999). | Article | PubMed | ChemPort |
-
Sevick-Muraca, E.M.
,
Houston, J.P.
&
Gurfinkel, M.
Curr. Opin. Chem. Biol. 6, 642–650 (2002). | Article | PubMed | ISI | ChemPort |
-
Ntziachristos, V.
,
Bremer, C.
&
Weissleder, R.
Eur. Radiol. 13, 195–208 (2003). | PubMed | ISI |
-
Frangioni, J.V.
Curr. Opin. Chem. Biol. 7, 626–634 (2003). | Article | PubMed | ISI | ChemPort |
-
De Grand, A.M.
et al. J. Biomed. Optics in press (2006).
-
Michalet, X.
et al. Science 307, 538–544 (2005). | Article | PubMed | ISI | ChemPort |
-
Lim, Y.T.
et al. Mol. Imaging 2, 50–64 (2003). | Article | PubMed | ChemPort |
|
