The fluorescence process offers two parameters to be measured for imaging: the intensity and the fluorescence lifetime. Fluorescence lifetime designates the time, the molecule stays in the excited state. The typical lifetime of a fluorochrome can be measured by observing a sufficiently large ensemble of excitation-emission events. We can measure the typical lifetime for all pixels in our image and write these numbers into the array-elements. That is Fluorescence Lifetime Imaging (FLIM, also known as τ-mapping). Typical fluorescence lifetimes range from 0.2 to 20 nanoseconds.
The fluorescence lifetime is independent of the concentration of fluorochromes. Whether the sample structure is sparsely decorated with fluorochromes or heavily loaded: the lifetime signal is always the same and indicates the presence of the same fluorochrome in the same environment. Fluorescence Lifetime is therefore not affected by bleaching. When recording deep in the sample, images will be substantially darker than the surface images – the lifetime does but not change at all. This is the main benefit for lifetime measurements.
If the molecular environment causes the excited state to decay without emitting a photon, the fluorescence intensity is reduced (quenched). Quenching is a separate path to emission and thus kinetically competes with the fluorescence process. The reservoir of excited states can now decay through more than one process and consequently the fluorescence lifetime shortens. This alteration in lifetime can be utilized to gather information on the molecular environment.
A special type of quenching is nonradiative transfer of the exciting energy to a neighboring, different fluorochrome: “Förster Resonance Energy Transfer”, FRET. Here not only the first fluorochrome (Donor) becomes darker and a shorter lifetime, but also the second fluorochrome (Acceptor) starts emitting upon the “wrong” excitation color. As this effect requires close contact of the two fluorochromes (less than 10 nm), it is used as a “molecular ruler” to investigate molecular interactions. It is also the base of many modern FRET-biosensors, tailored to measure all sorts of intracellular parameters in living samples: Ca2+ or other ion concentration tracking, pH, polarity and potential measurements, protein-protein interactions, and many more.
The gold standard for FLIM is “Time Correlated Single Photon Counting” TCSPC which is compatible with confocal and multiphoton microscopy. The sample is flashed with a light pulse and the time until the emission photon arrives is measured. To reconstruct the typical life time, such measurement is repeated several times at the same sample position. To reach some 10% accuracy requires a few 100 events to be measured. The data are then time-binned in a histogram. The decay of occurrences over time in this histogram is then mathematically fitted and directly provides the desired fluorescence lifetime (Fig 01a-c).
This procedure is the most precise and reproducible, but takes time. After a measurement the system is ready for the next shot by some 100 nanoseconds dead-time. To register 400 events, the time to measure one pixel amounts to 40 microseconds. A 512x512 image will take some 12 seconds - not compatible with fast- changing and moving living systems.
Current solutions require quite some technical dexterity, complex manual data handling and elaborate evaluation procedures.
Leica Microsystems presents a new FLIM microscope that solves all these problems: the SP8 FALCON (for FAst Lifetime CONtrast).
This system is based on Hybrid Detectors (HyDs), the ideal sensors for lifetime imaging and featuring a very short dead-time. Both the laser pulse train and the emission photon pulses are immediately digitized at a high sampling rate and the measured distance is directly fed into a data pool. These arrival times are used to output a “fast-FLIM” image.
The new approach allows to use most laser pulses and a filter to be applied that restricts to only pulse events with one photon emitted. The unavoidable abstraction of photons that arrive very shortly after the first photon, can be corrected by a smart mathematical approach. All these factors sum up to a 10 times faster image recording.
All FALCON FLIM imaging is integrated in a confocal microscope. Recording FLIM is just like activation of an additional channel. Multidimensional acquisition modes for 3D, time and spectral series are immediately available for FLIM imaging. An example is the title-image: a mouse embryo with classical histological staining. Using the NAVIGATOR software, a tile-artwork was created with 722 tiles, 512x512 pixels each. The original image has 190 Megapixels. Lifetime was fitted with four exponentials, the lifetime coded in color.
In the field of functional imaging we wish to monitor dynamic changes of small molecules, ions or electrical potential. This is often done with fluorescent probes. Many of them show both intensity and lifetime changes. It is also possible to decorate proteins with fluorescent proteins and express these constructs in living cells. This technique allows to follow molecular interactions in living cells by exploiting the FRET phenomena. The most modern tool are proteins or peptides, that bind to the desired analyte (e.g. Ca2+) and are decorated with a pair of fluorescent proteins performing FRET. Upon binding the analyte the peptide will undergo a conformational change and FRET will occur or disappear. These probes are called FRET-biosensors. An example for a biosensor is Epac, a sensor for cAMP.
The new technology in the SP8 FALCON is fast enough to allow fluorescence lifetime imaging with chemosensors and FRET-biosensors in living cells.
A no-stain application is analysis of endogenous fluorescence in live material. An example: cancer cells avoid cellular respiration and prefer anaerobic glycolysis (Warburg-effect), which results in accumulating fluorescent NADH in higher concentration. Multiphoton microscopy enables deep imaging in skin tissue, even without biopsy. Here again, lifetime contrast is much more reliable, as microscopy in deep layers will suffer from absorption and shadowing effects that distort the intensity signal.
Classically, different fluorochromes are distinguished by their different color. If excitation and emission are identical, they may still be distinguished by their lifetime. Emission intensity and lifetime can independently be recorded as a function of the color. The number of distinguishable fluorochromes is therefore the product of emission bands and fitted lifetimes.
The SP8 FALCON (FAst Lifetime CONtrast) confocal and multiphoton microscope from Leica Microsystems brings together all of these new technologies and concepts for a FLIM solution suited to multiple areas of application. Image acquisition is 10 times faster than classical TCSPC, the gold standard in lifetime imaging so far. This opens up the possibility of monitoring kinetics and dynamics in living samples using lifetime contrast.
Straightforward implementation and automation relieves the researcher from time-consuming hardware adjustments and data evaluation. Complex data acquisition modes like 3D-stacking, time lapse sequences, mosaic-scanning and excitation or emission wavelength scans are seamlessly combined with FLIM.
- Gerritsen et al: Fluorescence Lifetime Imaging in Scanning Microscopy. Pawley J (Ed) Handbook of Biological Confocal Microscopy, 3rd ed. Springer New York (2006)
- Förster T: Zwischenmolekulare Energiewanderung und Fluoreszenz. Ann. Physik. 437, p 55 (1948)
- Becker W: Fluorescence lifetime imaging – techniques and applications. J. Microscopy 247/2, pp 119-136 (2012)
- Borlinghaus RT, Birk H & Schreiber F: Detectors for Sensitive Detection: HyD. In: Mendez-Vilas A (ed.): Current microscopy contributions to advances in science and technology, Formatex Vol. 5, 818-825 (2012)
- Ponsioen B et al: Detecting cAMP-induced Epac activation by fluorescence resonance energy transfer: Epac as a novel cAMP indicator. EMBO reports 5/12 pp 1176-1180 (2004)
- Georgakoudi I et al: NAD(P)H and Collagen as in Vivo Quantitative Fluorescent Biomarkers of Epithelial Precancerous Changes. Cancer Research 62, 682– 687 (2002)