Exciton annihilation as a probe of the light-harvesting antenna transition into the photoprotective mode
Graphical abstract
Highlights
► We study non-photochemical quenching in non-aggregated LHCIIs. ► Excitation annihilation is a sensitive conformational probe. ► LHCII conformation is found to change with environmental conditions.
Introduction
Photosynthesis in plants occurs through the absorption of light by a number of light-harvesting pigment–protein complexes embedded in the thylakoid membrane and subsequent excitation energy transfer to the reaction center (RC), where initial charge separation takes place to initiate the light phase of photosynthesis [1]. En route to the RC the singlet chlorophyll (Chl) excitation is apt to undergo intersystem crossing to a triplet state. In the triplet state the Chl molecule possesses sufficient energy to excite molecular oxygen to the singlet state, known to be destructive for biological molecules. Formation of singlet oxygen is inhibited by carotenoid (Car) molecules as they accept Chl triplets through intersystem crossing, and the Car triplet energy is too low to excite singlet oxygen [2].
At high light, however, the amount of absorbed excitation energy exceeds the turn-over rate of the RC, and singlet excitations lingering on the antenna pigments in the presence of oxygen result in highly reactive and potentially hazardous molecular species. To cope with this, plants have evolved a defensive mechanism by which absorbed excitation excess is harmlessly dissipated to heat. This process is commonly referred to as non-photochemical quenching (NPQ) [3]. Its major part, the energy-dependent quenching (qE) [4], depends upon the thylakoid trans-membrane proton gradient [5], the presence of the PsbS protein [6] and the xanthophyll cycle of zeaxanthin (Zea), a specific Car molecule [7]. Several mechanisms have been proposed to underlie qE (for the recent review see [3]). Various experiments of in vitro aggregation of isolated LHCII antenna complexes revealed a drastic decrease of the fluorescence yield and lifetime which is the main spectroscopic signature of qE. This has led to the suggestion that oligomerized LHCII provides an excellent in vitro system for studies of the qE mechanism [8]. Consequently, from the spectroscopic measurements on LHCII aggregates a number of molecular sites and mechanisms have been proposed to be responsible for excitation energy dissipation: (i) a conformational change bringing about the enhancement of excitation transfer from Chl to the rapidly and non-radiatively decaying S1 state of lutein (Lut) 1 [9]; (ii) charge transfer [10] or excimer formation [11] in a strongly-coupled Chl dimer; (iii) low energy, short-lived and non-radiative exciton state formation through an increase of Car S1 state and Chl interaction [12].
Each LHCII monomer contains 14 Chls (8 Chl a and 6 Chl b molecules) and 4 Cars [13], [14], [15]. Excitation energy flow between these pigments with different spectroscopic signature in LHCII has been studied extensively using transient absorption, time-resolved fluorescence and photon echo techniques [16], [17], [18], [19], [20], [21], [22] revealing energy transfer steps occurring on different time scales ranging from hundreds of femtoseconds to many picoseconds. Detailed modeling based on the modified Redfield theory and high resolution crystal structure allowed association of the process of the energy transfer with specific pigments [23], [24], [25], [26], [27]. According to this model, 7 Chls a of each LHCII monomer form three excitonically coupled clusters where relaxation among the excitonic levels and the energy transfer among the clusters occurs on a sub-ps time scale. Similarly, Chl b cluster is also characterized by sub-ps relaxation and transfer to Chl a clusters. Mixed Chl b and remaining Chl a cluster equilibrates on a sub-ps time scale resulting in predominant population of so-called “bottleneck” Chl a and Chl b with relatively slow (several ps) energy transfer between them and even slower (tens of ps) transfer to the other Chl a sites. In equilibrium most of the energy is localized on the Chl a610–a611–a612 trimer, with Chl a610 featuring the highest population probability.
After spectral equilibration in LHCII the electronic excitations are subject to decay through internal conversion, intersystem crossing and fluorescence, whereas in a system of interconnected pigments (the case of both trimers and aggregates) multiple excitations are also prone to mutual annihilation. Moreover, in the case of aggregates an additional non-radiative decay channel responsible for NPQ has to be considered. The compartmental model taking into account all these effects was used to interpret the rich transient absorption data with the resulting assignment of the NPQ site to Lut 1 [9]. Subsequently the model was refined by reducing the number of compartments and at the same time taking into account singlet–singlet annihilation explicitly [28]. Remarkably, assuming the absence of excitation annihilation and a different spectral signature of the quencher, interpretation of a similar dataset resulted in entirely different conclusions as to the identity of the NPQ site [10]. In the present work we wish to take advantage of the recent discovery that excitation energy quenching in LHCII aggregates occurs not due to the aggregation as such but because of the removal of detergent leading to protein aggregation as a byprocess [29]. Since the detergent micelle is thought to mimic the native membrane environment, its removal has a profound effect on the condition of LHCII resulting in its quenching. Importantly, quenching could be achieved in isolated LHCII trimers without their aggregation if they were embedded in the polymer gel. This presents us with unprecedented opportunity to monitor excitation energy flow in quenched LHCII trimers free from factors complicating quantitative analysis of LHCII aggregates, such as the uncertainty in the oligomerization extent, the geometry of the LHCII trimer connectivity and the spatial distribution of non-photochemical quenchers. Then comparison of the annihilation parameters from the unquenched trimers in solution and quenched trimers in the gel allows us to draw conclusions about the nature of changes within LHCII leading to its quenching.
Section snippets
LHCII preparation
LHCII trimers were prepared from BBY particles using IEF procedure as described in [30]. Further purification of LHCII was carried out by sucrose density gradient centrifugation. Sucrose gradients were seven step exponential gradients from 0.15 to 1.0 M sucrose dissolved in 15 mM HEPES buffer containing 0.03% (w/v) n-dodecyl β-D-maltoside (β-DM. The run time was 18 h at 200,000 g in a SW41 rotor at 4 °C. Sucrose was removed from LHCII using a PD10 column (GE Healthcare) in a buffer containing 15 mM
Results
Transient absorption (TA) kinetics of the LHCII trimers and aggregates in solution and (un)quenched LHCII trimers in the acrylamide gel together with corresponding model fits (discussed below) are presented in Fig. 1. Pump at 650 nm predominantly excites Chls b which transfer excitation energy to Chls a on a subpicosecond to picosecond time scale. Subsequent equilibration of excitations within the LCHII monomer takes place within a few tens of ps. After that time the ground state bleaching of
Discussion
The annihilation rate is defined by the exciton migration rate within the LHCII complex [1] and therefore must be sensitive to the possible conformational changes of the protein scaffold, which determines the geometry and electronic properties of the chromophores. With relatively fast excitation equilibration within the LHCII monomer, the annihilation rate should be mainly determined by the inter-monomer excitation energy transfer. Thus in terms of the coarse-grained model [34], [35] the
Acknowledgements
The authors thank Giedrius Sasnauskas for indispensable help with acrylamide gel preparation and immobilization and Cristian Ilioaia for advising on LHCII gels. This research was supported by the Research Council of Lithuania through European Union Structural Funds project “Postdoctoral Fellowship Implementation in Lithuania” and grant No. MIP-110/2010, the European Social Fund under the Global Grant measure and the international collaboration grant from the Royal Society and UK EPSRC grant
References (39)
- et al.
Biophys. J.
(1995) - et al.
Biochim. Biophys. Acta, Bioenerg.
(2012) - et al.
Biochim. Biophys. Acta Bioenerg.
(1979) - et al.
Biochim. Biophys. Acta
(1992) - et al.
Chem. Phys.
(1995) - et al.
Chem. Phys.
(1996) - et al.
Biophys. J.
(2003) - et al.
Biophys. J.
(2001) - et al.
Chem. Phys.
(2009) - et al.
J. Biol. Chem.
(2008)