Fluorescence time-domain data

There are two ways to collect FLIM data freqnency-domain or time-domain data acqnisition (Alcala et al. 1985 Jameson et al. 1984). Briefly, in freqnency domain FLIM, the fluorescence lifetime is determined by its different phase relative to a freqnency modulated excitation signal nsing a fast Fourier transform algorithm. This method requires a frequency synthesizer phase-locked to the repetition freqnency of the laser to drive an RF power amplifier that modulates the amplification of the detector photomultiplier at the master frequency plus an additional cross-correlation freqnency. In contrast, time-domain FLIM directly measures t using a photon connting PMT and card. 

Prior to describing the possible applications of laser-diode fluorometry, it is important to understand the two methods now used to measure fluorescence lifetimes these being the time-domain (Tl)/4 5 24 and frequency-domain (FD) or phase-modulation methods.(25) In TD fluorometry, the sample is excited by a pulse of light followed by measurement of the time-dependent intensity. In FD fluorometry, the sample is excited with amplitude-modulated light. The lifetime can be found from the phase angle delay and demodulation of the emission relative to the modulated incident light. We do not wish to fuel the debate of TD versus FD methods, but it is clear that phase and modulation measurements can be performed with simple and low cost instrumentation, and can provide excellent accuracy with short data acquisition times. 

The transient response of luminescent substances to pulsed excitation can be captured in the time domain by sampling enough data points within the time span of the decay. For example, fast digitizers are commonly employed to store phosphorescence decays. If fast digitizers are unavailable, time-correlated single-photon counting can be used to monitor fluorescence decays. 

Fig. 2 Data acquisition for time-domain FLIM. FI fluorescence intensity, h gated image no 1,12 gated image no 2. Left Excitation pulse of the light source and synchronized timegated detection with a CCD camera. Right Lifetime determination by two subsequent time-gates according to Eq. 2...

Our objective was to probe fluorescence over a time domain as large as possible. To this end we combined two different detection techniques, FU and TCSPC, allowing us to perform measurements from 100 fs to hundreds of nanoseconds. Notably, we use the same laser excitation source the third harmonic of a titane sapphire laser (267 nm, 100 fs). This is important because the excited state population is created under identical conditions in the two types of experiments. The time-resolution obtained after deconvolution of the recorded signals is 100 fs and 10 ps for FU and TCSPC, respectively. For reasons explained below, FU only detects emission corresponding to highly allowed transitions. TCSPC, on the other hand, is capable to monitor not only allowed but also very weak or forbidden transition. Therefore, particular care must be taken when merging data obtained by these two techniques as described in Ref. 10. 

In spite of its prevalence in the fluorescence decay literature, we were not universally successful with this fitting method. Most reports of hi- or multiexponential decay analysis that use a time-domain technique (as opposed to a frequency-domain technique) use time-correlated photon counting, not the impulse-response method described in Section 2.1. In time-correlated photon-counting, noise in the data is assumed to have a normal distribution. Noise in data collected with our instrument is probably dominated by the pulse-to-pulse variation of the laser used for excitation this variation can be as large as 10-20%. Perhaps the distribution or the level of noise or the combination of the two accounts for our inconsistent results with Marquardt fitting. 

All these studies with femtosecond pulses on the primary photochemical processes of rhodopsin were done by means of transient absorption (pump probe) spectroscopy [10]. However, absorption spectroscopy may not be the best way to probe the excited-state dynamics of rhodopsin, because other spectral features, such as ground-state depletion and product absorption, are possibly superimposed on the excited-state spectral features (absorption and stimulated emission) in the obtained data. Each spectral feature may even vary in the femtosecond time domain, which provides further difficulty in analyzing the data. In contrast, fluorescence spectroscopy focuses only on the excited-state processes, so that the excited-state dynamics can be observed more directly. 

Tune-resolved measurements are widely used in fluorescence spectroscopy, particularly for studies of biological macromolecules. This is because time-iesolved data fie-quently contain more information than is available frcnn the steady-state data. For instance, consider a protein which contains two tryptophan residues, each with a distinct 11 fetime. Because of spectral overlap of the absorption and emission, it is not usually possible to resolve the onission from the two residues. However, the tune-resolved data may reveal two decay times, which can beused to resolve the emission spectra and relative intensities of the two tryptophan residues. Then one can question how each of the tryptophan residues is affected by the interactions of the protein with its substrate or other macromolecules. Is one of the tryptophan residues close to the binding site Is a tryptophan residue in a distal domain affected by substrate binding to another domain Such questions can be answered if one measures the decay times associated with each tryptophan residue. 

The use of nonlinear least-squares analysis is ubiquitous in the analySK of fluorescence data, pairicululy time-domain and fiequency-domain data. A usefiil introduction to the principles of least-squares analysis is found in the compact but informative book by Bevington (1969). The applications of these concepts to diverse types of fluorescence data can be found in edited volumes (Brand and Johnson, 1992 Johnson and Brand, 1994). For more basic information about statistics and spectroscopy, one can examine several introductoiy texts (Ihylor, 1982 Mark and Workman, 1991). 

The interpretation of the data relies on the assumption that the fluorescence yield in the observed time domain is dominated by the redox state of (for a review see ref. 6). From the fluorescence yield changes concentrations were calculated on the basis of the equation described in ref. 7 accounting for excitation energy transfer by a factor of 0.5 and for PS II heterogeneity by a factor of 0.7. The relaxation of Q " induced by saturating laser flashes (t = 10 ns) can be properly described by a biexponential decay ... 

The pH effect on the steady-state fluorescence spectrum, Figure 2, can also be seen in the time domain Between pH=6 and 3 there is no change in the time resolved data, but at lower pH values the long time tail increases (Figure 5). This corresponds to the pH range for which the R OH peak in Figure 2 increases in amplitude. The increase in the tail in strongly acidic... 

A general discussion of the use of least-squares fitting in fluorescence measurements may be found in (28). The global analysis of fluorescence data is discussed in (29). Commercially available time-resolved fluorimeters are typically sold with data analysis software included. Available stand-alone packages include the Globals Unlimited suite, which is capable of analysing both time- and frequency-domain data, stopped-flow kinetics, etc. The Center for Fluorescence Spectroscopy at the University of Maryland (USA) also offers software for frequency- and time-domain fluorescence lifetime analysis. 

There is one domain, however, in which EWIF is probably unsurpassable. Due to the high sensitivity of fluorescence measurements, significant data can be obtained in much less than one minute. Therefore, the kinetics of formation of the adsorbed layer can be monitored almost in real time. Schemes in which the incident beam is kept fixed but where the detection is performed at various angles should further speed up the measurements especially if the single channel photomultiplier is replaced by a linear position sensitive detector. 

Most of the photophysical data for azulene and its simple derivatives have come from measurements of the quantum yields of Sj - Sq fluorescence, ( )s2, and time-domain measurements of the lifetimes of the fluorescent Sj state, Xjj, in fluid solution. The first order rate constants for the parallel radiative and nonradiative processes that depopulate Sj are then determined from k = ( )s2/Xs2 and Ek, = (1 - ( )s2)/ s2- Whereas Sj - Si internal conversion is clearly the major, and perhaps the exclusive, pathway for Sj s nonradiative decay in azulene itself, other nonradiative processes can be expected to occur in some substituted azulenes. 

One limitation, however, is that only a limited number of spots can be measured simultaneously. A compromise between the temporal analysis of FCS and fluorescence fluctuation analysis in the spatial domain [39] can be obtained by exploiting the time structure of sample/laser scanning confocal microscope images [40,41]. Thereby, spatial correlation analysis of the emitted fluorescence is combined with temporal characterization of the fluorescence emission from the serial data stream of subsequently scanned pixels. This... 

The complexity observed in polymer fluorescence decay kinetics is further exacerbated when fluorescent polyelectrolytes are dissolved in aqueous media [29,30,33,35,37,43,120,122,128-132] segregation of the macromolecular structure into hydrophobic and hydrophilic-rich domains results in differing degrees of water penetration which further complicates the time-resolved fluorescence [26]. Within this context, more recent attempts to describe time-resolved polymer photophysical data include use of the blob model [133,134], which accounts for the range of environments encountered in heterogeneous systems by invoking a distribution of rate constants for excimer formation. 

Fluorescence lifetime data of 1, 4, 5, and 6 in presence of 10 M -CD were collected with frequency-domain fluorometry. These probes gave only 1 1 complexes with P-CD [58] and, given the association constant values, the complex molar fraction was >0.95 for 4 and 5 and 0.1 for 6. The fluorescence decay of all the probes was best described by unimodal Lorentzian lifetime distribution [51,59] rather than by a mono- or biexponential function corresponding to the emission of the complexed and the free probe. This distribution was attributed to the coexistence of molecules included in the cavity to different extents. It was proposed that, in the case of 4, the apolar benzene ring enters the cavity first and penetrates until the whole naphthalene is included. This is the most stable and, hence, the most populated conformation of the complex. The distribution of the lifetimes suggests that at any time there is an ensemble of molecules in different stages of complexation which have slightly different lifetimes. 

Fluorescence anisotropy decay of [Leu ] enkephalin tyrosine was measured using the frequency- domain up to 10 GHz. The data indicate a 44 ps cori elation time for local tyrosine motions and a 219 ps correlation time for overall rotational diffusion of the pentapeptide (Lakowicz et al. 1993). Also a rotational correlation time of 26 ps was measured by H NMR for Ha of tyrosine in position 1 of L-dermorphin (Simenel, 1990). These ps values determined by NMR and by fluorescence spectroscopy are the result of possible significant atomic fluctuations that occur in the picosecond time scale (Karplus and Me Gammon, 1981). Since it was difficult in quenching experiments performed on DREK to measure such short correlation times we do not know whether these atomic fluctuations would depend on the conformation of the peptide or not. However, our results clearly put into evidence the presence of a local rotation within DREK. 

---