Fluorescence and phosphorescence resemble each other and in many ways can give the same type of information. Both originate from a dipolar interaction between light and the molecule. In this regard, both are polarized and subject to the same type of quenching phenomena. In other respects the information which they divulge are complementary. The fluorescence quantum yield is higher for exposed tryptophans and this is expressed in longer lifetime (Grinvald and Steinberg, 1976); in contrast long lifetime of phosphorescence appears to correlate with burial. Phosphorescence, spin-disallowed, is much longer lived than fluorescence. This allows the structural/dynamic characterization of proteins to be studied on a new time regime. A really remarkable finding of studies of protein phosphorescence is that there is such variability both in phosphorescence lifetime and quenchability. We would interpret this to indicate that the tryptophan environment can range from essentially a crystal, almost comparable in rigidity as found at 77 K, to tryptophans in a flexible environment, almost as flexible as free in solution. An interesting task will be to examine the relationship between the yield and lifetime of phosphorescence and details of the tryptophan environment in terms of rigidity and adjacent amino acids among the proteins with known three dimensional structure.

Much evidence, on both theoretical and experimental sides, indicates the importance of local fluctuations (in energy levels, conformational substates, etc.) of the macromolecular matrix in the biological activity of proteins. We describe here a novel application of the Förster-type energy-transfer process capable of monitoring changes both in local fluctuations and in conformational states of macromolecules. A new energy-transfer parameter, f, is defined as an average transfer efficiency, [E], normalized by the actual average quantum efficiency of the donor fluorescence, [phi D]. A simple oscillator model (for a one donor-one acceptor system) is presented to show the sensitivity of this parameter to changes in amplitudes of local fluctuations. The different modes of averaging (static, dynamic, and intermediate cases) occurring for a given value of the average transfer rate, [kt], and the experimental requirements as well as limitations of the method are also discussed. The experimental tests were performed on the ribonuclease T1-pyridoxamine 5'-phosphate conjugate (a one donor-one acceptor system) by studying the change of the f parameter with temperature, an environmental parameter expectedly perturbing local fluctuations of proteins. The parameter f increased with increasing temperature as expected on the basis of the oscillator model, suggesting that it really reflects changes of fluctuation amplitudes (significant changes in the orientation factor, k2, as well as in the spectral properties of the fluorophores can be excluded by anisotropy measurements and spectral investigations). Possibilities of the general applicability of the method are also discussed.

The triplet state absorption and phosphorescence of Zn and Pd derivatives of myoglobin were compared. Both metal derivatives exhibit long triplet state lifetimes at room temperature, but whereas the Pd derivative showed exponential decay and an isosbestic point in the transient absorption spectra, the decay of the Zn derivative was nonsingle exponential and the transient absorption spectra showed evidence of more than one excited state species. No difference was seen in triplet quenching by oxygen for either derivative, indicating that differences in the polypeptide chain between the two derivatives are not large enough to affect oxygen penetrability. Quenching was also observed by anthraquinone sulfonate. In this case, the possibility of long-range transfer by an exchange mechanism is considered.

Ten In earlier work, we showed that the phenomenon of roomtemperature protein phosphorescence, though previously seen only rarely (1, 2), in fact can be found in the great majority of proteins (3). Of 40 proteins surveyed, 29 were found to exhibit phosphorescence in aqueous solution at room temperature with a wide range of lifetimes-between about 0.5 msec and 2 sec. The central requirement for the observation of protein phosphorescence in solution is to reduce dissolved oxygen to a sufficiently low level, since oxygen can efficiently quench the excited tryptophan triplet state, even when the tryptophan is buried in the protein matrix (3-6).A subsequent study (7) revealed that a variety of smallmolecule agents in addition to dioxygen can quench the phosphorescence of protein tryptophans. Surprisingly, even though the phosphorescent tryptophans are well buried within the protein, the quenching efficiency of most of the agents tested (those larger than three atoms in size) was found to be independent of the size and polarity of the quenching agent. This indicates that the quenching process does not involve the penetration of these quenchers through the protein matrix to the position of the buried tryptophan. Although the sensitivity of the different proteins to quenching was found to be spread over a wide range, the various agents tested all quenched any given protein with similar efficiency, indicating that the quenching reaction is determined by some property of the individual tryptophan or the protein itself rather than by the particular quenching agent used. Finally, the quenching rate was essentially independent of solution viscosity. This rules out the possibility of a protein-opening reaction that might transiently bring the buried tryptophan into contact with solvent and the added quenchers. All these properties might be explained, it was noted, if the quenching process involves long-range electron transfer occurring on a long time scale, so that the quenchers in solution are effectively in the rapid diffusion limit.The present work represents an attempt to identify the protein parameters that determine the sensitivity of its tryptophans to the quenching process. We compared various structural parameters of 10 different proteins with their ability to be quenched by four different small-molecule agents. The results show that quenching rate constants decrease exponentially with the distance of the tryptophan from the protein surface, consistent with an electronexchange reaction in the rapid diffusion limit. Analysis of the data then provides an estimate of the dependence of electrontransfer rate on distance when the intervening space is filled with averaged protein matrix. MATERIALS AND METHODSGlyceraldehyde-3-phosphate dehydrogenase (GAPDH) from porcine muscle, Pronase type XIV from Streptomyces griseus, and protease type X from Bacillus thermoproteolyticus rokko were obtained from Sigma. Nuclease from Staphylococcus aureus was a gift of E. E. Lattman (Baltimore). The sources of other proteins and su...

Ca2+-ATPase molecules were labeled in intact sarcoplasmic reticulum (SR) vesicles, sequentially with a donor fluorophore, fluorescein-5'-isothiocyanate (FITC), and with an acceptor fluorophore, eosin-5'-isothiocyanate (EITC), each at a mole ratio of 0.25-0.5 mol/mol of ATPase. The resonance energy transfer was determined from the effect of acceptor on the intensity and lifetime of donor fluorescence. Due to structural similarities, the two dyes compete for the same site(s) on the Ca2+-ATPase, and under optimal conditions each ATPase molecule is labeled either with donor or acceptor fluorophore, but not with both. There is only slight labeling of phospholipids and other proteins in SR, even at concentrations of FITC or EITC higher than those used in the reported experiments. Efficient energy transfer was observed from the covalently bound FITC to EITC that is assumed to reflect interaction between ATPase molecules. Protein denaturing agents (8 M urea and 4 M guanidine) or nonsolubilizing concentrations of detergents (C12E8 or lysolecithin) abolish the energy transfer. These results are consistent with earlier observations that a large portion of the Ca2+-ATPase is present in oligomeric form in the native membrane. The technique is suitable for kinetic analysis of the effect of various treatments on the monomer-oligomer equilibrium of Ca2+-ATPase. A drawback of the method is that the labeled ATPase, although it retains conformational responses, is enzymatically inactive.