Daniel Hoersch

Strong Hydrogen Bond between Glutamic Acid 46 and Chromophore Leads to the Intermediate Spectral Form and Excited State Proton Transfer in the Y42F Mutant of the Photoreceptor Photoactive Yellow Protein

In the Y42F mutant of photoactive yellow protein (PYP) the photoreceptor is in an equilibrium between two dark states, the yellow and intermediate spectral forms, absorbing at 457 and 390 nm, respectively. The nature of this equilibrium and the light-induced protonation and structural changes in the two spectral forms were characterized by transient absorption, fluorescence, FTIR, and pH indicator dye experiments. In the yellow form, the oxygen of the deprotonated p-hydroxycinnamoyl chromophore is linked by a strong low-barrier hydrogen bond to the protonated carboxyl group of Glu46 and by a weaker one to Thr50. Using FTIR, we find that the band due to the carbonyl of the protonated side chain of Glu46 is shifted from 1736 cm(-1) in wild type to 1724 cm(-1) in the yellow form of Y42F, implying a stronger hydrogen bond with the deprotonated chromophore in Y42F. The FTIR data suggest moreover that in the intermediate spectral form the chromophore is protonated and Glu46 deprotonated. Flash spectroscopy (50 ns-10 s) shows that the photocycles of the two forms are essentially the same except for a transition around 5 mus that has opposite signs in the two forms and is due to the chemical relaxation between the two dark states. The two cycles are coupled, likely by excited state proton transfer. The Y42F cycle differs from wild type by the occurrence of a new intermediate with protonated chromophore between the usual I(1) and I(2) intermediates which we call I(1)H (370 nm). Transient fluorescence measurements indicate that in I(1)H the chromophore retains the orientation it had in I(1). Transient proton uptake occurs with a time constant of 230 mus and a stoichiometry of 1. No proton uptake was associated however with the formation of the I(1)H intermediate and the relaxation of the yellow/intermediate equilibrium. These protonation changes of the chromophore thus occur intramolecularly. The chromophore-Glu46 hydrogen bond in Y42F is shorter than in wild type, since the adjacent chromophore-Y42 hydrogen bond is replaced by a longer one with Thr50. This facilitates proton transfer from Glu46 to the chromophore in the dark by lowering the barrier, leading to the protonation equilibrium and causing the rapid light-induced proton transfer which couples the cycles.

Monitoring the conformational changes of photoactivated rhodopsin from microseconds to seconds by transient fluorescence spectroscopy.

The transient changes of the tryptophan fluorescence of bovine rhodopsin in ROS membranes were followed in time from 1 micros to 10 s after flash excitation of the photoreceptor. Up to about 100 micros the fluorescence did not change, suggesting that the tryptophan lifetimes in rhodopsin and the M(I) intermediate are similar. The fluorescence then decreases on the millisecond time scale with kinetics that match the rise of the M(II) state as measured on the same sample by the transient absorption increase at 360 nm. Both the sign and kinetics of the fluorescence change strongly suggest that it is due to an increase in energy transfer to the retinylidene chromophore caused by the increased spectral overlap in M(II). Calculation of the Forster radius of each tryptophan from the high-resolution crystal structure suggests that W265 and W126 are already completely quenched in the dark, whereas W161, W175, and W35 are located at distances from the retinal chromophore that are comparable to their Forster radii. The fluorescence from these residues is thus sensitive to an increase in energy transfer in M(II). Similar results were obtained at other temperatures and with monomeric rhodopsin in dodecyl maltoside micelles. A large light-induced transient fluorescence increase was observed with ROS membranes that were selectively labeled with Alexa594 at cysteine 316 in helix 8. Using transient absorption spectroscopy the kinetics of this structural change at the cytoplasmic surface was compared to the formation of the signaling state M(II) (360 nm) and to the kinetics of proton uptake as measured with the pH indicator dye bromocresol purple (605 nm). The fluorescence kinetics lags behind the deprotonation of the Schiff base. The proton uptake is even further delayed. These observations show that in ROS membranes (at pH 6) the sequence of events is Schiff base deprotonation, structural change, and proton uptake. From the temperature dependence of the kinetics we conclude that the Schiff base deprotonation and the transient fluorescence have comparable activation energies, whereas that of proton uptake is much smaller.

Dynamics of light-induced activation in the PAS domain proteins LOV2 and PYP probed by time-resolved tryptophan fluorescence.

Light-induced activation of the LOV2-Jα domain of the photoreceptor phototropin from oat is believed to involve the detachment of the Jα helix from the central β-sheet and its subsequent unfolding. The dynamics of these conformational changes were monitored by time-resolved emission spectroscopy with 100 ns time resolution. Three transitions were detected during the LOV2-Jα photocycle with time constants of 3.4 μs, 500 μs, and 4.3 ms. The fastest transition is due to the decay of the flavin phosphorescence in the transition of the triplet LOV(L)(660) state to the singlet LOV(S)(390) signaling state. The 500 μs and 4.3 ms transitions are due to changes in tryptophan fluorescence and may be associated with the dissociation and unfolding of the Jα helix, respectively. They are absent in the transient absorption signal of the flavin chromophore. The tryptophan fluorescence signal monitors structural changes outside the chromophore binding pocket and indicates that there are at least three LOV(S)(390) intermediates. Since the 500 μs and 4.3 ms components are absent in a construct without the Jα helix and in the mutant W557S, the fluorescence signal is mainly due to tryptophan 557. The kinetics of the main 500 μs component is strongly temperature dependent with activation energy of 18.2 kcal/mol suggesting its association with a major structural change. In the structurally related PAS domain protein PYP the N-terminal cap dissociates from the central β-sheet and unfolds upon signaling state formation with a similar time constant of ∼1 ms. Using transient fluorescence we obtained a nearly identical activation energy of 18.5 kcal/mol for this transition.

Let there be light: how to use photoswitchable cross-linker to reprogram proteins

Azobenzene is a photo-isomerizing molecule whose end-to-end distance changes upon external illumination. When combined with site-specific reactive groups, it can be used as molecular tweezers to remote-control the structure and function of protein targets. The present study gives a brief overview over the rational design strategies that use an azobenzene-based photoswitchable cross-linker to engineer ON/OFF switches into functional proteins or to reprogram proteins for novel functions. The re-engineered proteins may be used as remote controls for cellular pathways, as light-gated drug delivery platforms or as light-powered machinery of synthetic cells and micro-scaled factories.

Engineering a light-controlled F1 ATPase using structure-based protein design

The F1 sub-complex of ATP synthase is a biological nanomotor that converts the free energy of ATP hydrolysis into mechanical work with an astonishing efficiency of up to 100% (Kinosita et al., 2000). To probe the principal mechanics of the machine, I re-engineered the active site of E.coli F1 ATPase with a structure-based protein design approach: by incorporation of a site-specific, photoswitchable crosslinker, whose end-to-end distance can be modulated by illumination with light of two different wavelengths, a dynamic constraint was imposed on the inter-atomic distances of the α and β subunits. Crosslinking reduced the ATP hydrolysis activity of four designs tested in vitro and in one case created a synthetic ATPase whose activity can be reversibly modulated by subsequent illumination with near UV and blue light. The work is a first step into the direction of the long-term goal to design nanoscaled machines based on biological parts that can be precisely controlled by light.