Excitation quenching in chlorophyll\u2013carotenoid antenna systems: \u2018coherent\u2019 or \u2018incoherent\u2019

Abstract

Plants possess an essential ability to rapidly down-regulate light-harvesting in response to high light. This photoprotective process involves the formation of energy-quenching interactions between the chlorophyll and carotenoid pigments within the antenna of Photosystem II (PSII). The nature of these interactions is currently debated, with, among others, ‘incoherent’ or ‘coherent’ quenching models (or a combination of the two) suggested by a range of time-resolved spectroscopic measurements. In ‘incoherent quenching’, energy is transferred from a chlorophyll to a carotenoid and is dissipated due to the intrinsically short excitation lifetime of the latter. ‘Coherent quenching’ would arise from the quantum mechanical mixing of chlorophyll and carotenoid excited state properties, leading to a reduction in chlorophyll excitation lifetime. The key parameters are the energy gap, Updelta varepsilon =varepsilon_{rm Car}-varepsilon_{rm Chl}, and the resonance coupling, J, between the two excited states. Coherent quenching will be the dominant process when -J<Updelta varepsilon <J, i.e., when the two molecules are resonant, while the quenching will be largely incoherent when varepsilon_{rm Chl}>(varepsilon_{rm Car}+J). One would expect quenching to be energetically unfavorable for varepsilon_{rm Chl}<(varepsilon_{rm Car}-J). The actual dynamics of quenching lie somewhere between these limiting regimes and have non-trivial dependencies of both J and Updelta varepsilon. Using the Hierarchical Equation of Motion (HEOM) formalism we present a detailed theoretical examination of these excitation dynamics and their dependence on slow variations in J and Updelta varepsilon. We first consider an isolated chlorophyll–carotenoid dimer before embedding it within a PSII antenna sub-unit (LHCII). We show that neither energy transfer, nor the mixing of excited state lifetimes represent unique or necessary pathways for quenching and in fact discussing them as distinct quenching mechanisms is misleading. However, we do show that quenching cannot be switched ‘on’ and ‘off’ by fine tuning of Updelta varepsilon around the resonance point, Updelta varepsilon =0. Due to the large reorganization energy of the carotenoid excited state, we find that the presence (or absence) of coherent interactions have almost no impact of the dynamics of quenching. Counter-intuitively significant quenching is present even when the carotenoid excited state lies above that of the chlorophyll. We also show that, above a rather small threshold value of J>10,mathrm{cm}^{-1}quenching becomes less and less sensitive to J (since in the window -J<Updelta varepsilon <J the overall lifetime is independent of it). The requirement for quenching appear to be only that J>0. Although the coherent/incoherent character of the quenching can vary, the overall kinetics are likely robust with respect to fluctuations in J and Updelta varepsilon. This may be the basis for previous observations of NPQ with both coherent and incoherent features.