Search

The role of histidine 148 and glutamate 222 is analyzed by studying the fluorescence dynamics of green fluorescence protein mutant (GFPmut) and its H148G and E222Q mutants. The power dependent but pH-independent relaxation mode is due to an excited-state process that is related to conformational rearrangements of the chromophore after the photoexcitation, more than to the chromophore excited-state proton transfer.

The fluorescence of Green Fluorescent Protein (wtGFP) and variants has been exploited in distinct applications in cellular and analytical biology. GFPs emission depends on the population of the protonated (A-state) and deprotonated (B-state) forms of the chromophore. Whereas wtGFP is pH-independent, mutants in which Ser65 is replaced by either threonine or alanine (as in GFPmut2) are pH-dependent, with a p K a around 6. Given the wtGFP pH-independence, only the structure of the protonated form was determined. The deprotonated form was deduced on the basis of the crystal structure of the Ser65Thr mutant at basic pH, assuming that it corresponds to the conformation populated in solution. Here, we present an investigation where structures of the protonated and deprotonated forms of GFPmut2 were determined from crystals grown in either MPD at pH 6 or PEG at pH 8.5, and moved to either higher or lower pH. Both crystal forms of GFPmut2 were titrated monitoring the process via polarized absorption microspectrophotometry in order to precisely correlate the protonation process with the structures. We found that (i) in solution, chromophore titration is not thermodynamically coupled with any residue and Glu222 is always protonated independent of the protonation state of the chromophore; (ii) the lack of coupling is reflected in the structural behavior of the chromophore and Glu222 environments, with only the former showing variations with pH; (iii) titrations of low-pH and high-pH grown crystals exhibit a Hill coefficient of about 0.75, indicating an anticooperative behavior not observed in solution; (iv) structures where pH was changed in the crystal point to Glu222 as the ionizable group responsible for the outset of the anticooperative behavior; and (v) in GFPmut2 the canonical GFP proton wire involving the chromophore is not interrupted at the level of Ser205 and Glu222 at basic pH as in the Ser65Thr mutant. This allows proposing the structure of the deprotonated state of GFPmut2 as an alternative model for the analogous state of wtGFP.

New probes for kinetic intracellular measurements in the millisecond range are desirable to monitor protein biochemical dynamics essential for catalysis, allosteric regulation, and signaling. Good candidates to this aim are the photoswitchable mutants of the green fluorescent protein, whose anionic fluorescence, primed by blue light, is markedly enhanced under an additional excitation at a shorter wavelength and relaxes within a few milliseconds. The aim of this report is to study how the brightness enhancement kinetics depends on the physical-chemical and spectroscopic parameters and to provide proof-of-concept experiments for the use of the fluorescence enhancement in conditions in which the protein diffusion is hindered and thereby photobleaching can be a limiting critical issue. Future, direct applications of photochromic mutants for modulated excitation imaging would in fact require such a detailed knowledge. We present here an extensive study of the photoswitching mechanism of the E222Q mutant of GFPMut2 (Mut2Q), pumped by visible 488 nm light and probed at 400-420 nm, as a function of pH, viscosity, temperature, and light intensity. In solution, two characteristic photoswitching times are found by means of modulated double beam fluorescence correlation spectroscopy in the 1-30 ms range, depending on the solution pH. The photoswitching kinetics is solved in terms of the eigenvalues and the eigenvectors of a specific energy diagram and used directly to fit the data, suggesting that the observed photoswitching amplitudes and kinetics are related to a single three-level transition loop. Finally, we give in vitro examples of the use of modulated excitation microscopy, based on fluorescence enhancement amplitude and kinetics detection, on Mut2Q protein samples immobilized in acrylamide gels.