Ultrafast infrared spectroscopy reveals a key step for successful entry into the photocycle for photoactive yellow protein

Abstract

Photoactive proteins such as PYP (photoactive yellow protein) are generally accepted as model systems for studying protein signal state formation. PYP is a blue-light sensor from the bacterium Halorhodospira halophila. The formation of PYP's signaling state is initiated by trans-cis isomerization of the p-coumaric acid chromophore upon the absorption of light. The quantum yield of signaling state formation is approximately 0.3. Using femtosecond visible pump/mid-IR probe spectroscopy, we investigated the structure of the very short-lived ground state intermediate (GSI) that results from an unsuccessful attempt to enter the photocycle. This intermediate and the first stable GSI on pathway into the photocycle, I0, both have a mid-IR difference spectrum that is characteristic of a cis isomer, but only the I0 intermediate has a chromophore with a broken hydrogen bond with the backbone N atom of Cys-69. We suggest, therefore, that breaking this hydrogen bond is decisive for a successful entry into the photocycle. The chromophore also engages in a hydrogen-bonding network by means of its phenolate group with residues Tyr-42 and Glu-46. We have investigated the role of this hydrogen bond by exchanging the H bond-donating residue Glu-46 with the weaker H bond-donating glutamine (i.e., Gln-46). We have observed that this mutant exhibits virtually identical kinetics and product yields as WT PYP, even though during the I0-to-I1 transition, on the 800-ps time scale, the hydrogen bond of the chromophore with Gln-46 is broken, whereas this hydrogen bond remains intact with Glu-46.

Fig. 1.

Schematic drawing of the active site of WT PYP. The p-coumaric acid chromophore is covalently bound to the protein backbone by means of Cys-69; in addition, it takes part in a distal hydrogen-bonding network. In E46Q, glutamic acid (Glu) is changed to glutamine by using site-directed mutagenesis.

Fig. 2.

Selection of time traces measured on E46Q PYP. Absorption difference (in mOD) is plotted as a function of time (in picoseconds). The time axis is linear up to 3 ps and logarithmic until 3 ns. The black lines are fits to the data based on a target analysis.

Fig. 3.

Decay-associated difference spectra (including error bars) of the E46Q data. The data are fitted with four time constants: 1.4 ps (black), 6 ps (green), 800 ps (red), and long-lived (blue).

Fig. 4.

Model used for the target analysis. GS is the ground state; I₀ and I₁ are the first two transient, on-pathway intermediates.

Fig. 5.

Species-associated difference spectra of WT PYP and E46Q as a result of the target analysis. A comparison of the ES spectra of WT (gray) and E46Q (black) is shown in a, and a comparison of GSI^WT (gray) and GSI^E46Q (green) is shown in b. The I₀ (red) and I₁ (blue) spectra for E46Q and WT PYP are shown in c and d, respectively. Note that c and d share the same legend. Negative features in these spectra originate from the ground state bleach, and positive ones originate from ES or product state absorption.

Fig. 6.

Comparison of the species-associated difference spectra of E46Q and WT in the Glu/Gln CO (blue box) and chromophore CO (red box) regions.