UC Berkeley

Biological photoreceptors are molecules that harvest light and convert it into chemical energy, ultimately driving a physiological response. They typically consist of a protein moiety and a strongly light-absorbing chromophore, and respond to the absorption of a photon with an elementary chemical reaction such as electron/proton transfer or cis-trans isomerization. The process of light-information conversion is of fundamental importance in biology, but it is not well understood at the molecular level. In this dissertation I investigate the primary photochemistry of two important photoreceptors, phytochrome Cph1 and rhodopsin. Although these proteins vary greatly in composition, architecture and function – phytochrome Cph1 drives a phototactic response in the cyanobacterium Synechocystis and rhodopsin initiates the visual process in vertebrates – they both respond to light stimuli through the isomerization of a carbon-carbon double bond in a covalently bound, extended conjugated chromophore. The primary purpose of my study is to understand the mechanism of the initial photoisomerization event that harnesses light energy for the purpose of driving biological function.

In my primary study, resonance Raman intensity analysis is used to investigate the structure and excited-state dynamics of the P_r and P_fr forms of phytochrome Cph1. I have accurately modeled the experimental absorption spectra of each conformer with a unique set of vibronic parameters, showing that both P_r and P_fr are structurally homogeneous in the ground state. The excited-state dynamics of the bilin chromophore are interesting because, although the low fluorescence quantum yield (φ_f < 10^–3 – 10^–4) suggests fast excited-state dynamics, the isomerization is relatively slow (~1−3 ps) and inefficient (φ ∼ 0.15). I investigate this discrepancy and find that, although torsional Franck-Condon analysis predicts a <200 fs Z/E isomerization for both the P_r → P_fr and P_fr → P_r reactions, the reaction does not proceed as rapidly as predicted due to steric interaction between the D-ring of the chromophore and the protein.

In collaboration with researchers at Politecnico di Milano, Università di Bologna, Universität Duisburg-Essen, and The University of Oxford, we investigated the 11-cis → all-trans isomerization of rhodopsin using ultrafast pump-probe spectroscopy with sub-20-fs time resolution and a spectral coverage from the visible to the near-infrared. We tracked the coherent wavepacket motion from the initial photoexcited state to the photoproduct and observed a pattern consistent with the presence of a conical intersection. During the first ∼80 fs, we observed the loss of stimulated emission from the reactant and a decrease in the energy gap as the ground- and excited-state potential energy surfaces approached each other. Once the molecules pass through the conical intersection, we saw the rise of photoproduct absorption and an increase in the probed energy gap. These results provide the first experimental observation of a conical intersection in the unique reactivity of rhodopsin.

These studies demonstrate that although phytochrome Cph1 and rhodopsin are very different proteins, their primary photochemistry is surprisingly similar. By responding to light stimuli through the isomerization of a carbon-carbon double bond, they are able to very efficiently mediate light-information conversion for the purpose of driving biological function.