Diversity of the PSI–PSII Megacomplexes That Conduct Energy Spillover in Green Plants

In oxygenic photosynthesis, two photosystems—photosystem I (PSI) and photosystem II (PSII)—are connected via the Cyt b₆f complex to comprise the photosynthetic electron transport chain, where water is oxidized by PSII, and the resulting electrons are used to reduce ferredoxin via PSI and eventually generate NADPH. In green plants, PSI and PSII bind to their respective light-harvesting complexes (LHCI and LHCII) to form the PSI–LHCI supercomplex and PSII–LHCII supercomplex, respectively. Many reports have indicated that PSII–LHCII localizes in the stacked grana core, while PSI–LHCI in stroma-exposed thylakoid membranes in chloroplasts of land plants (Dekker and Boekema 2005). As such, the classic view is that PSI and PSII function separately in thylakoid membranes. However, this view is not consistent with other previous reports suggesting an energetic interaction between the two types of photosystems (Murata 1969, Satoh et al. 1976), a phenomenon known in later years as ‘spillover’.

Spillover is an excitation energy transfer from PSII to PSI and is considered a mechanism to balance the excitation between the two types of photosystems. In spinach chloroplasts, spillover is enhanced when the PSII reaction centers are closed (Satoh et al. 1976). Spillover is different from state transitions, a relevant mechanism modulating the light-harvesting antenna size of PSI and PSII by relocating LHCII (Rochaix 2007, Minagawa 2011) (Fig. 1A). Theoretically, spillover requires a close interaction between PSI and PSII because excitation energy transfer cannot occur between distantly located pigments (Fig. 1B). It was reported in wheat leaves that PSI migrates toward the central region of the grana stack and associates with PSII during state transitions, which would enhance spillover from PSII to PSI (Tan et al. 1998). However, the structural basis for the PSI–PSII interaction in spillover had not been elucidated at that time.

Fig. 1

Schematic representations of state transitions and spillover. (A) In state transitions (state 1 to state 2 transitions), LHC II protein is phosphorylated ([P], denoted with a circle) by a specific kinase and associated with PSI, while the functional antenna size of PSII is reduced, as indicated by the dashed circular rectangle. This prevents the overexcitation of PSII. (B) The core-mediated type and (C) the antenna-mediated type of the PSI–PSII megacomplex are involved in spillover in Arabidopsis and rice, respectively. The transfer of light energy from PSII to PSI in the PSI–PSII megacomplexes quenches excess energy and prevents photodamage. The dashed arrows indicate the transfer of light energy.

Open in new tabDownload slide

In later years, it was found that PSI–LHCI and PSII–LHCII form a megacomplex (PSI–PSII megacomplex) that can be isolated from Arabidopsis thaliana leaves by large-pore blue-native polyacrylamide gel electrophoresis (lpBN-PAGE, Järvi et al. 2011). Yokono et al. (2015) also isolated PSI–PSII megacomplexes from digitonin-solubilized thylakoid membranes of Arabidopsis by large-pore clear-native polyacrylamide gel electrophoresis (lpCN-PAGE). Furthermore, the PSI–PSII megacomplexes isolated by lpCN-PAGE were analyzed with time-resolved fluorescence analysis, in which the chlorophyll fluorescence of a sample is monitored as a function of time and wavelength after excitation by a flash of light. They demonstrated that excitation energy produced by charge recombination in PSII is transferred to PSI and emitted as a delayed fluorescence around ∼735 nm at 77K. This provided the first evidence that spillover indeed occurs in the PSI–PSII megacomplex. Cyanobacteria also have megacomplexes composed of PSI, PSII and phycobilisomes; however, there is disagreement as to whether energy transfer occurs from PSII to PSI (Liu et al. 2013, Ueno et al. 2016).

This delayed fluorescence due to spillover can also be observed in intact Arabidopsis leaves, suggesting that about half of the excitation energy generated by charge recombination at PSII can be transferred to PSI (Yokono et al. 2015). Delayed fluorescence is also observed as a large PSI peak in the leaves of a chlorophyll b–less mutant (ch1-1), in which the LHCII level is drastically decreased, and in five lhc mutants (lhca1–4 and lhcb5), indicating that LHCs are not required for PSI–PSII complex formation in Arabidopsis (Fig. 1B). The direct interaction between the PSI and PSII core antennae supports the fast energy transfer from PSII to PSI (∼20ps) observed by fluorescence decay–associated spectra (FDAS) via global analysis of time-resolved fluorescence spectra at 77K (Yokono et al. 2015). In addition, electron microscopy (EM) analysis suggests that averaged images of the PSI–PSII megacomplex of A. thaliana contain one PSII core dimer sandwiched by two PSIs with 2-fold rotational symmetry (Yokono et al. 2019).

A physiological function of spillover via the PSI–PSII megacomplex has been proposed to quench excess light energy by PSI. In fact, the relative amount of PSI–PSII megacomplex in Arabidopsis thylakoid membranes increases with increasing light intensity and decreases quickly as light intensity decreases (Yokono et al. 2015). The presence of the PSI–PSII megacomplex has also been investigated by time-resolved fluorescence analysis among green plants grown under different light environments (Yokono et al. 2019). These studies suggest that sun plants adapted to high-intensity light accumulate the PSI–PSII megacomplex more than shade plants grown under low-intensity light. Some green algae, such as Chlamydomonas reinhardii, accumulate lower amounts of the PSI–PSII megacomplex. Interestingly, the FDAS data suggest that PSI in sun plants has a deeper trap to receive excitation energy, indicated by fluorescence maxima longer than 730 nm (Yokono et al. 2019). This deep trap corresponds to low-energy chlorophylls in PSI and would facilitate excitation energy dissipation. These facts suggest that there seems to be a species diversity in the formation and nature of photosystem megacomplexes as well as in the dynamic response of the formation of the PSI–PSII megacomplex to light intensity.

In this issue, Kim et al. (2023) suggest that modes of interaction between photosystems in the PSI–PSII megacomplex are divergent among green plants. They investigated the formation of the PSI–PSII megacomplex biochemically in C. reinhardii, spinach, Arabidopsis and rice. Thylakoid membranes isolated from each species were solubilized by n-dodecyl-α-D-maltoside (α-DDM) with the ionic amphipol A8-35 to stabilize large mega- and array protein complexes, which were subsequently separated by sucrose density gradient (SDG) centrifugation. Spinach, Arabidopsis and rice accumulate relatively stable photosystem megacomplexes, while the green alga C. reinhardtii does not accumulate them. This observation is consistent with the previous FDAS data indicating the much lower frequency of spillover in C. reinhardtii than in vascular plants (Yokono et al. 2019). Furthermore, higher-molecular-weight fractions containing the PSI–PSII megacomplexes, which appear as green bands in the SDG separation, are more dispersed in rice than in Arabidopsis, indicating a variety of conformations in the different plant species.

Kim et al. (2023) then characterized the rice PSI–PSII megacomplex spectroscopically to examine the energy transfer capabilities from PSII to PSI in the rice PSI–PSII megacomplex. They observed the delayed fluorescence from PSI with a lifetime of ∼25 ns, suggesting an energy spillover from PSII to PSI. The FDAS analysis suggests that the slow PSII to PSI energy transfer component is relatively dominant in rice PSI–PSII supercomplexes than in Arabidopsis ones, indicating the possibility that PSI and PSII may interact through LHCII molecule(s) in the rice PSI–PSII megacomplex (Fig. 1C). This possibility was further examined by negatively stained EM analysis. The EM images of the selected particles suggest that, in most cases, PSI–LHCI binds to LHCII antennae rather than the PSII core in the rice PSI–PSII megacomplex. In addition, there is a remarkable variation in the conformation of PSI–PSII megacomplexes in rice. This suggests that the PSI–PSII supercomplexes do not have a fixed conformation in rice; instead, PSI–LHCI and PSII–LHCII might interact flexibly to cope with the excess light energy.

In summary, spillover—an excitation energy transfer from PSII to PSI—occurs in the PSI–PSII megacomplex and seems to be widely used to dissipate excess light energy in green plants, especially in vascular plants. The findings introduced in this commentary provide a structural basis for a long-known phenomenon and lead to a new understanding of the light-environment response mechanisms of plants. However, it appears that a diversity in the structure and nature of PSI–PSII megacomplexes exists among plant species as well as within individuals of the same species (see Fig. 4E in Kim et al. 2023, in this issue). This opens the door to further research into elucidating the formation of photosystem megacomplexes in response to environmental changes—perhaps through analyses of their detailed 3D structures by cryo-EM.

Data Availability

No new datasets were generated or analyzed in this study.

Disclosures

The author has no conflicts of interest to declare.

References