Mechanistic investigation and engineering of fluorescent proteins

One of our central themes concerns photophysical studies of phototransformable fluorescent proteins (PTFPs). PTFPs are fundamental players in super-resolution microscopy as well as other advanced fluorescence methods such as pulse-chase imaging, modulated-illumination imaging, photochromic Förster Resonance Energy Transfer (FRET) or biological data storage.

These genetically encoded fluorescence markers are fascinating because their fluorescence state can be tuned by illumination with proper laser light.
 For example, the emission color of some PTFPs can be changed from green to red upon illumination with violet light (so-called photoconvertible FPs, PCFPs).
 Other PTFPs can be reversibly switched from on states to off states by alternating cyan and violet light (so-called reversibly photoswitchable FPs, RSFPs).

Phototransformation properties like photoactivation, photoconversion and photoswitching need to be understood and optimized for each application. In addition, photoblinking (the stochastic and transient loss of fluorescence) and photobleaching (the irreversible loss of fluorescence) are crucial photophysical properties that apply to all fluorescent proteins and that we investigate in great details, although they still hide a lot of mysteries.

Many groups worldwide study and develop fluorescent proteins to create ever more performing markers or sensors. We contribute to this field of research by investigating PTFPs using an extensive set of tools : we typically employ a combination of kinetic crystallography (including XFEL in collaboration with the Dynamop group), in crystallo spectroscopy, single molecule imaging in vitro and in cellulo and modelling in silico (collaboration LCP, Paris Saclay), and recently NMR in collaboration with the NMR group).

In the last years, we have largely focused our attention on photoblinking and photobleaching in popular PCFPs such as mEos4b. These photophysical processes seem to exhibit a never ending complexity, but we discover more and more about them. For example, in collaboration with the team of P. Dedecker (KUL, Belgium) we recently deciphered a major mechanism of blinking in red mEos4b, typically causing serious trouble in the quality and interpretation of quantitative PALM (qPALM) or sptPALM data, and showed that this mechanism very closely relates to the reversible photoswitching mechanisms at play in RSFPs. This allowed us to propose a trick to reduce blinking in sptPALM, so as to enable the reconstruction of significantly longer tracks. We also turned our attention to green state photophysics in PCFPs. Indeed, although PCFPs are only detected once photoconverted to the red state, what happens before photoconversion considerably affects their performance as markers for SMLM. Recently, in collaboration with the teams of J.B Sibarita and M. Sainlos (IINS, Bordeaux), we started an ANR-funded project aiming at further improving the photostability of PCFPs, by combining structural studies with high-content-screening single-molecule imaging approaches to achieve efficient semi-rational engineering.
More and more, we realize the important link between PTFPs protein dynamics and PTFPs photophysical behavior, so that NMR, which can address the dynamical behavior of proteins with great detail, is also becoming a central tool for our investigations.