One- and two-photon-induced fluorescence from recombinant green fluorescent protein

The fluorescence spectral properties of recombinant green fluorescent protein (rGFP) were examined with one- and two-photon excitations using femtosecond pulses from a Ti:sapphire laser. Intensity-dependent properties of the two-photon-induced fluorescence from rGFP excited by an 800-nm, 100-fs laser beam were reported, and the two-photon excitation cross section of rGFP was measured at 800 nm as about 160 x 10(-50) cm(4)s/photon. The possible excited-state proton transfer between two electronic states at about 400 nm in protonated (RH) species and 478 nm in deprotonated (R(-)) species in rGFP was confirmed by fluorescence and fluorescence excitation anisotropy spectra. A subelectronic state (or vibronic progression) at about 420 nm in RH species was identified, which was relatively stable and not involved in the excited state proton transfer in rGFP upon irradiation.     

Tryptophan fluorescence intensity and anisotropy decays of human serum albumin resulting from one-photon and two-photon excitation.

We measured the emission spectra, intensity decays and anisotropy decays of the single tryptophan residue of human serum albumin (HSA) resulting from one-photon (295-298 nm) and two-photon (590-596) excitation. The emission spectra and intensity decays were independent of the mode of excitation. The anisotropy decays were superficially similar for one- and two-photon excitation. However, upon consideration of the different orientation photoselection for one- and two-photon excitation, the anisotropy data reveal different angles between the absorption and emission oscillators for one-photon and two-photon excitation. This result suggests different relative one-photon and two-photon cross-sections for the 1La and 1Lb transitions of the indole residue. This first report of the time-resolved anisotropy decay of a protein resulting from two-photon excitation suggests that such measurement will yield insights into the complex photophysical properties of tryptophan residues in proteins.     

Analysis of time-resolved fluorescence anisotropy decays.

We discuss the analysis of time-correlated single photon counting measurements of fluorescence anisotropy. Particular attention was paid to the statistical properties of the data. The methods used previously to analyze these experiments were examined and a new method was proposed in which parallel- and perpendicular-polarized fluorescence curves were fit simultaneously. The new method takes full advantage of the statistical properties of the measured curves; and, in some cases, it is shown to be more sensitive than other methods to systematic errors present in the data. Examples were presented using experimental and simulated data. The influence of fitting range on extracted parameters and statistical criteria for evaluating the quality of fits are also discussed.     

Attenuation of photobleaching in two-photon excitation fluorescence from green fluorescent protein with shaped excitation pulses

The two-photon excitation fluorescence (TPEF) process of an enhanced green fluorescent protein (EGFP) for fluorescence signals was adaptively controlled by the phase-modulation of femtosecond pulses. After the iteration of pulse shaping, a twofold increase in the ratio of the fluorescence signal to the laser peak power was achieved. Compared with conventional pulses optimized for peak power, phase-optimized laser pulses reduced the bleaching rate of EGFP by a factor of 4 while maintaining the same intensity of the fluorescence signal. Our method will provide a powerful solution to various problems confronting researchers, such as the photobleaching of dyes in two-photon excitation microscopy.     

Red-edge anisotropy microscopy enables dynamic imaging of homo-FRET between green fluorescent proteins in cells

Steady-state fluorescence anisotropy measurements can be used to detect fluorescence resonance energy transfer (FRET) between identical fluorophores (homo-FRET). However, the contribution of homo-FRET to the steady-state anisotropy must be discerned from those due to the orientational distribution and rotational diffusion, which so far has required photobleaching controls, largely precluding dynamic measurements in live cells. We describe a variation of steady-state anisotropy microscopy in which the contribution of homo-FRET is dynamically isolated from the total anisotropy by exploiting the loss of energy transfer that occurs at red-edge excitation. Excitation of enhanced green fluorescent protein (EGFP) at the red-edge of its absorption band shows the shift in the emission spectrum compared to main-band excitation that is characteristic for photo-selection of static low energy S(0)-S(1) transitions that fail to exhibit FRET. An experimental setup for steady-state fluorescent anisotropy microscopy is described that can be used to acquire anisotropy images in live cells at main-band and red-edge excitation of EGFP. We demonstrate in live cells homo-FRET suppression of protein fusion constructs that consist of two and three EGFP molecules connected by short linkers. This methodology represents a novel approach for the dynamic measurement of homo-FRET in live cells that will be of utility in the biological sciences to detect oligomerization and concentration dependent interactions between identically labeled molecules.     

Development and characterization of green fluorescent protein mutants with altered lifetimes

Multiple-probe fluorescence imaging applications demand an ever-increasing number of resolvable probes, and the use of fluorophores with resolvable fluorescence lifetimes can help meet this demand. Green fluorescent protein (GFP) and its variants have been widely used in spectrally resolved multiprobe imaging, but as yet, there has not been a systematic set of mutants generated with resolvable lifetimes. Therefore, to generate such mutants, we have utilized error-prone PCR and fluorescence lifetime imaging to screen for mutants of UV-excited green fluorescent protein (GFPuv) that exhibit altered fluorescence decay lifetimes. This has resulted in the isolation of GFPuv mutants displaying at least three distinctly different lifetimes in the range of 1.9-2.8 ns. Mutation of Y145 to either histidine or cysteine was found to shift the fluorescence lifetime of GFPuv from 3.03 +/- 0.03 to 2.78 +/- 0.05 ns for the Y145H mutant and to 2.74 +/- 0.05 ns for Y145C. Some of the shorter-lifetime mutants exhibited excitation peaks that were red-shifted relative to their maximal absorption, indicating that the mutations allowed the adoption of additional conformations relative to wtGFPuv. The utility of these mutants for applications in simultaneous imaging and quantification is shown by the ability to quantify the composition of binary mixtures in time-resolved images using a single detector channel. The application of the screening method for generating lifetime mutants of other fluorescent proteins is also discussed.     

Analysis of fluorescence anisotropy decays by a least square method

A method of fluorescence anisotropy decay analysis is described in this work. The transient anisotropy r(ex)(t) measured in a photocounting pulsefluorimeter is fitted by a non linear least square procedure to the ratio of convolutions of the apparatus response function g(t) by sums of appropriate exponential functions. This method takes rigorously into account the apparatus response function and is applicable to any shape of the later as well as to any values of fluorescence decay times and correlation times. The performances of the method have been tested with data simulated from measured response functions corresponding to an air lamp and a high pressure nitrogen lamp. The statistical standard errors of the anisotropy deca parameters have been found to be smaller than the standard errors previously calculated for the moment method. A systematic error delta in the fluorescence decay time entailed an error deltatheta in the correlation time such as Deltatheta/theta < deltatau/tau. By this method, good fitting of experimental data have been achieved very conveniently and accurately.     

Complex homogeneous and heterogeneous fluorescence anisotropy decays: enhancing analysis accuracy

In biological macromolecules, fluorophores often exhibit multiple depolarizing motions that require multiple lifetimes and rotational relaxation times to define fluorescence intensity and anisotropy decays. The related analysis of time-correlated single-photon counting data becomes uncertain due to the multitude of decay parameters and numerical sensitivity to deconvolution of the instrument response function (IRF) via discretization of integrals. By using simulations we show that improved discretizations based on quadratic and cubic local approximations of the IRF yield more accurate estimation of short rotational relaxation times and lifetimes than the commonly used Grinvald-Steinberg discretization, which in turn appears more reliable than two discretizations based on linear local approximations of the IRF. In addition, our simulation suggests that cubic approximation is the most advantageous in discriminating complex heterogeneous and homogeneous anisotropy decay. We show that among three different information criteria, the Akaike information criterion is best suited for detection of heterogeneity in rotational relaxation times. It is capable of detecting heterogeneity even when anisotropy decay appears homogeneous within statistical errors of estimation.     

Stepwise two-photon excited fluorescence from higher excited states of chlorophylls in photosynthetic antenna complexes

Stepwise two-photon excited fluorescence (TPEF) spectra of the photosynthetic antenna complexes PCP, CP47, CP29, and light-harvesting complex II (LHC II) were measured. TPEF emitted from higher excited states of chlorophyll (Chl) a and b was elicited via consecutive absorption of two photons in the Chl a/b Qy range induced by tunable 100-fs laser pulses. Global analyses of the TPEF line shapes with a model function for monomeric Chl a in a proteinaceous environment allow distinction between contributions from monomeric Chls a and b, strongly excitonically coupled Chls a, and Chl a/b heterodimers/-oligomers. The analyses indicate that the longest wavelength-absorbing Chl species in the Qy region of LHC II is a Chl a homodimer with additional contributions from adjacent Chl b. Likewise, in CP47 a spectral form at approximately 680 nm (that is, however, not the red-most species) is also due to strongly coupled Chls a. In contrast to LHC II, the red-most Chl subband of CP29 is due to a monomeric Chl a. The two Chls b in CP29 exhibit marked differences: a Chl b absorbing at approximately 650 nm is not excitonically coupled to other Chls. Based on this finding, the refractive index of its microenvironment can be determined to be 1.48. The second Chl b in CP29 (absorbing at approximately 640 nm) is strongly coupled to Chl a. Implications of the findings with respect to excitation energy transfer pathways and rates are discussed. Moreover, the results will be related to most recent structural analyses.     

Measurement of subnanosecond anisotropy decays of protein fluorescence using frequency-domain fluorometry

We report the first anisotropy decays of protein fluorescence obtained using a frequency-domain fluorometer. The ultraviolet light source (300 nm) was a ring dye laser equipped with an intracavity frequency doubler, pumped by an argon ion laser. The data, measured at modulation frequencies from 2 to 200 MHz, reveal the presence of subnanosecond motions (0.1-0.2 ns) of the single tryptophan residues in melittin and monellin. For melittin the data also indicate the presence of slower motions near 1 ns, which may be the result of concerted motions of several peptide units. Smaller amplitude motions, on a similar timescale, were observed for the single tryptophan residue in staphylococcal nuclease. We demonstrate using N-acetyl-L-tryptophanamide in water that the method of frequency-domain fluorometry is capable of measuring correlation times as short as 50 ps. This method can provide data for the direct comparison of measured anisotropy decays with those predicted from molecular dynamics calculations.     

Imaging fluorescence resonance energy transfer between two green fluorescent proteins in living yeast

We show that fluorescence resonance energy transfer between two mutants of the green fluorescent protein (GFP) can be monitored by imaging microscopy in living yeast. This work is based on the constitutive expression of a GFP-containing fusion protein and the inducible expression of the tobacco etch virus (TEV) protease. In the fusion protein, the P4.3 GFP mutant is linked to the YS65T GFP mutant by a spacer bearing the TEV protease-specific cleavage site.     

Nanosecond fluorescence anisotropy decays of 1,6-diphenyl-1,3,5-hexatriene in membranes

Nanosecond decays of the fluorescence anisotropy, $\text{r}$, were studied for the emission of 1,6-diphenyl-l,3,5-hexatriene (DPH) embedded in a series of mixed multilamellar liposomes containing egg yolk phosphatidylcholine, phosphatidylethanolamine and cholesterol in varying molar ratios, as well as in membranes of intact cells and in virus envelopes.The relative contributions of the fast and the infinitely slow decaying component to the steady-state value, $\text{r}$, of the fluorescence anisotropy were very similar for artifical and biological membranes.Angles, θ, of the cone, by which the motion of the fluorescent molecule is limited, were calculated from the intensity of the infinitely slow decaying anisotropy component and compared with steady-state fluorescence anisotropies and with ‘microviscosities’, 〈η〉. An increase in 〈η〉 from 1.5 to 5.2 P in our systems was accompanied by a decrease in θ from 49° to 30° while the decrease in the mean motional relaxation times, $\text{φ}{}_{\mathrm{<mtext>f</mtext>}}$, of the label molecule was not more than 1 ns and due mainly to changes in the potential, by which the diffusion of DPH in the membrane is restricted. From these observations we conclude that differences in the steady-state fluorescence anisotropy and in ‘microviscosities’ of cholesterol-containing membranes ($\text{r}\text{> 0.15}$) represent changes in the degree of static orientational constraint rather than changes in diffusion rates of the label.     

Molecular heterogeneity of O-acetylserine sulfhydrylase by two-photon excited fluorescence fluctuation spectroscopy

O-acetylserine sulfhydrylase, a homo-dimeric enzyme from Salmonella typhimurium, covalently binds one pyridoxal 5'-phosphate molecule per subunit as a fluorescent coenzyme. Different tautomers of the Schiff base between the coenzyme and lysine 41 generate structured absorption and fluorescence spectra upon one-photon excitation. We investigated the protein population heterogeneity by fluorescence correlation spectroscopy and lifetime techniques upon two-photon excitation. We sampled the fluorescence intensity from a small number of molecules (approximately 10) and analyzed the distribution of photon counts to separately determine the number and the fluorescence brightness of excited protein molecules. The changes in the average number of molecules and in the fluorescence brightness with the excitation wavelength indicate the presence of at least two fluorescent species, with two-photon excitation maxima at 660 and 800 nm. These species have been identified as the enolimine and ketoenamine tautomers of the protein-coenzyme internal aldimine. Their relative abundance is estimated to be 4:1, whereas the ratio of their two-photon cross sections is reversed with respect to the single-photon excitation case. Consistent results are obtained from the measurement of the lifetime decays, which are sensitive to the excited-state heterogeneity. At least two components were detected, with lifetimes of approximately 2.5 and 0.5 ns. The lifetimes are very close to the values measured in bulk solutions upon one-photon excitation and attributed to the ketoenamine tautomer and to a dipolar species formed upon proton dissociation in the excited state.     

Synthesis and characterization of water-soluble two-photon excited blue fluorescent chromophores for bioimaging.

We report here the synthesis and characterization of a new type of non-ionic blue fluorescent water-soluble chromophores specifically designed for two-photon absorption microscopy. The water solubility is induced by introduction of short oligo(ethylene glycol) monomethyl ether moieties. This work has led to low molecular weight dyes with efficient two-photon absorption cross sections and high fluorescence quantum yield in organic solvents as well as in aqueous solutions.     

Picosecond resolution of tyrosine fluorescence and anisotropy decays by 2-GHz frequency-domain fluorometry

We extended the technique of frequency-domain fluorometry to an upper frequency limit of 2000 MHz. This was accomplished by using the harmonic content of a laser pulse train (3.76 MHz, 5 ps) from a synchronously pumped and cavity-dumped dye laser. We used a microchannel plate photomultiplier as the detector to obtain the 2-GHz bandwidth. This new instrument was used to examine tyrosine intensity and anisotropy decays from peptides and proteins. These initial data sets demonstrate that triply exponential tyrosine intensity decays are easily recoverable, even if the mean decay time is less than 1 ns. Importantly, the extended frequency range provides good resolution of rapid and/or multiexponential tyrosine anisotropy decays. Correlation times as short as 15 ps have been recovered for indole, with an uncertainty of +/- 3 ps. We recovered a doubly exponential anisotropy decay of oxytoxin (29 and 454 ps), which probably reflects torsional motions of the phenol ring and overall rotational diffusion, respectively. Also, a 40-ps component was found in the anisotropy decay of bovine pancreatic trypsin inhibitor, which may be due to rapid torsional motions of the tyrosine residues and/or energy transfer among these residues. The rapid component has an amplitude of 0.05, which is about 16% of the total anisotropy. The availability of 2-GHz frequency-domain data extends the measurable time scale for fluorescence to overlap with that of molecular dynamics calculations.     

Resolution of initially excited and relaxed states of tryptophan fluorescence by differential-wavelength deconvolution of time-resolved fluorescence decays

We describe a new procedure for the analysis of time-resolved decays of fluorescence intensity. This procedure was used to resolve the emission spectra of the initially excited and solvent relaxed states of a tryptophan derivative in viscous solution. Specifically, we examined N-acetyl-l-tryptophanamide (AcTrpNH₂) in viscous and nonviscous solutions of propylene glycol. Time-resolved decays of fluorescence intensity were collected at wavelengths across the emission spectra. Instead of the usual procedure of deconvolving these data with the time profile of the exciting pulse, we deconvolved these data using the response observed on the short-wavelength side of the emission. If one assumes that this emission results only from the initially excited state (F), then the nonzero decay time calculated using deconvolution is that of the solvent relaxed state (R). For our specific case of AcTrpNH₂ the emission spectra of the F and R states overlap at most wavelengths longer than the short-wavelength side of the emission (310 nm). As a result, differential-wavelength deconvolution yields two lifetimes and amplitudes, one pair representing the relaxed state and the other the initially excited state. The latter appears as a zero-decay-time component whose amplitude can be readily quantified. The wavelength-dependent amplitude of this zero-lifetime component can be used to calculate the emission spectrum of the F state and. by difference, the emission spectrum of the relaxed state. For AcTrpNH₂ in propylene glycol at ?20°C the emission maxima of the F and R states are near 320 and 350 nm, respectively, and the relative proportion of the emission from each state was near 50%. At lower temperatures the emission from the F state becomes dominant and at high temperatures the emission from the R state dominates. We note that this resolution of states is somewhat arbitrary because we assumed a two-state model and the absence of solvent relaxed emission at 310 nm. Nonetheless, differential-wavelength deconvolution simplifies and facilitates the analysis of time-resolved fluorescence data from samples which undergo excited state reactions. Moreover, this deconvolution procedure considerably simplifies the determination of the kinetic constants for reversible excited state reactions. The application of differential-wavelength deconvolution does not increase the time reqaired for data acquisition. This differential analysis procedure should enhance the usefulness and precision of pulse fluorometric methods in studies of nanosecond time scale processes in proteins and membranes.     

Photostability of green and yellow fluorescent proteins with fluorinated chromophores, investigated by fluorescence correlation spectroscopy

The photostability of the widely used autofluorescent proteins EGFP and EYFP and their fluorinated counterparts were compared by means of fluorescence correlation spectroscopy. We analyzed the reduction of the apparent diffusional time in analogy to fluorescence quenching in which the 'photon concentration' is treated as the quencher concentration. The quantum yields of photobleaching Phi(bl) of EYFP (6.1x10(-)(5)) and EGFP (8.2x10(-)(5)) are in agreement with the previously published values. Among the investigated mutants, EYFP has the highest photostability and there is an enhanced photobleaching in (2-F) Tyr-EYFP. It turns out that the chromophore fluorination has no significant influence on the photostability.     

In vivo imaging of the actin polymerization state with two-photon fluorescence anisotropy

Using two-photon fluorescence anisotropy imaging of actin-GFP, we have developed a method for imaging the actin polymerization state that is applicable to a broad range of experimental systems extending from fixed cells to live animals. The incorporation of expressed actin-GFP monomers into endogenous actin polymers enables energy migration FRET (emFRET, or homoFRET) between neighboring actin-GFPs. This energy migration reduces the normally high polarization of the GFP fluorescence. We derive a simple relationship between the actin-GFP fluorescence polarization anisotropy and the actin polymer fraction, thereby enabling a robust means of imaging the actin polymerization state with high spatiotemporal resolution and providing what to the best of our knowledge are the first direct images of the actin polymerization state in live, adult brain tissue and live, intact Drosophila larvae.     

Nanosecond fluorescence anisotropy decays of 1,6-diphenyl-1,3,5-hexatriene in membranes.

Nanosecond decays of the fluorescence anisotropy, r, were studied for the emission of 1,6-diphenyl-1,3,5-hexatriene (DPH) embedded in a series of mixed multilamellar liposomes containing egg yolk phosphatidylcholine, phosphatidylethanolamine and cholesterol in varying molar ratios, as well as in membranes of intact cells and in virus envelopes. The relative contributions of the fast and the infinitely slow decaying component to the steady-state value r, of the fluorescence anisotropy were very similar for artifical and biological membranes. Angles, theta, of the cone, by which the motion of the fluorescent molecule is limited, were calculated from the intensity of the infinitely slow decaying anisotropy component and compared with steady-state fluorescence anisotropies and with 'microviscosities', (eta). An increase in (eta) from 1.5 to 5.2 P in our systems was accompanied by a decrease in theta from 49 degrees to 30 degrees while the decrease in the mean motional relaxation times, phi f, of the label molecule was not more than 1 ns and due mainly to changes in the potential, by which the diffusion of DPH in the membrane is restricted. From these observations we conclude that differences in the steady-state fluorescence anisotropy and in 'microviscosities' of cholesterol-containing membranes (r greater than 0.15) represent changes in the degree of static orientational constraint rather than changes in diffusion rates of the label.