Rewiring state transitions mediated by light-harvesting complex I

Photosynthesis, which takes place in the chloroplasts of cyanobacteria, algae, and plants, converts light into electrochemical energy to fuel the production of oxygen and carbohydrates. This photochemistry relies on the actions of two modular pigment–protein complexes, photosystem I (PSI) and photosystem II (PSII) (Croce and van Amerongen, 2020). To best perceive and capture light, photosystems have evolved peripheral light-harvesting antenna systems that shuttle excitation energy toward the photosystem reaction centers. Balanced allocation of excitation energy between the two photosystems is a prerequisite for optimal photosynthetic efficiency. Over-excitation of one photosystem over another readily leads to the accumulation of deleterious reactive oxygen species and dramatically compromised plant fitness and yield. Because the two photosystems possess distinct pigment compositions and exhibit largely different absorption spectra, the imbalance of energy distribution between them is a common occurrence. Thus, photosynthetic organisms need to constantly re-allocate harnessed energy between the two photosystems in response to rapidly changing environmental factors, such as light, water, and temperature. Understanding the relevant molecular mechanisms is fundamental for the improvement of plant photosynthesis.

The so-called “state transition” is an evolutionarily conserved acclimation mechanism that adjusts relative excitation pressure between the two photosystems by modulating their surrounding antenna size through a mobile pool of the light-harvesting complex II (LHCII) trimers that can differentially associate with the two photosystems under certain light conditions (Minagawa, 2011; Figure 1). When PSI is preferentially activated, such as under far-red light or canopy shade conditions, LHCII is exclusively associated with PSII (State 1). When PSII is over-excited under blue or red light conditions, the reduced plastoquinone pool activates the serine/threonine-protein kinase STATE TRANSITION 7 (STN7), which phosphorylates the mobile LHCII trimers. Phosphorylated LHCII dissociates from PSII and binds to the PSI core complex to form LHCII–PSI–LHCI supercomplexes (State 2). During the State 2 to State 1 transition, the oxidized plastoquinone pool inactivates STN7 and the constitutively active phosphatase PROTEIN PHOSPHATASE 1/THYLAKOID-ASSOCIATED PHOSPHATASE 38 dephosphorylates LHCII, leading to re-association of LHCII with PSII. This reversible mechanism can occur in minutes and is thus essential for plants to protect photosystems from photodamage under fluctuating light conditions.

Figure 1.

A refined model of state transitions in Arabidopsis. In State 1, when PSI is preferentially excited, the plastoquinone pool is oxidized and the thylakoid protein kinase STATE TRANSITION 7 (STN7) is inactive. Non-phosphorylated LHCII are mostly associated with PSII in the grana. A minor portion of LHCII localized in the stroma lamellae transfers excitation energy to the PSI reaction centers via LHCI. In State 2, when PSII is preferentially activated, the reduced plastoquinone pool stimulates STN7, which phosphorylate LHCII trimers. While a fraction of phosphorylated LHCII migrates from grana to stromal lamellae and associates with PSI to form the LHCII–PSI–LHCI supercomplexes, the other phosphorylated or monophosphorylated LHCII enhances the cross section of PSI by trafficking excitation energy to PSI, ideally via LHCI. During the State 2 to State 1 transition, PROTEIN PHOSPHATASE 1/THYLAKOID-ASSOCIATED PHOSPHATASE 38 (PPH1/TAP38) dephosphorylates LHCII. Red dotted lines represent electron transfer flow. The width of red curved arrows indicates the energy transfer rate. Adapted from Schiphorst et al. (2021).

Although light-harvesting complex I (LHCI) is not incorporated into the current canonical state transitions model, characterization of Arabidopsis (Arabidopsis thaliana) mutants with depletion of one LHCI subunit, LHCA4, suggests a potential role of LHCI in mediating energetic interaction between PSI and LHCII during state transitions (Benson et al., 2015). In this issue of Plant Physiology, Schiphorst et al. (2021) addressed this hypothesis by comparing PSI antenna size and excitation energy transfer and trapping in the Arabidopsis wild-type and mutant plants lacking all four LHCI subunits (Δlhca) (Bressan et al., 2016).

The authors quantified PSI antenna size in the thylakoid membranes of wild-type and Δlhca plants using P700 oxidation kinetics (Schiphorst et al., 2021). Surprisingly, while Δlhca lost all LHCI antenna, PSI antenna size was only reduced by approximately 20% under State 1 conditions when the naked PSI core complexes were present on the native PAGE gels. This suggests that “additional” LHCII can serve as the antenna of PSI to compensate for the deficiency of LHCI. However, the association of “additional” LHCII with PSI is difficult to detect through standard methods because the mild non-ionic detergent digitonin, which is applied in native gel electrophoresis to solubilize photosynthetic complexes, including the LHCII–PSI–LHCI supercomplexes accumulated in the non-appressed regions of thylakoid membranes, is capable of removing this part of LHCII from PSI (Grieco et al., 2015; Bos et al., 2017, 2019). Therefore, the “additional” LHCII is referred to as the digitonin-sensitive LHCII.

The two photosystems display distinct localizations in the thylakoid membranes: PSI–LHCI is enriched in the stroma lamellae, whereas PSII–LHCII supercomplexes are predominantly found in the stacked grana (Andersson and Anderson, 1980). In theory, the energy harvested by the antenna localized in the stroma lamellae is finally transferred to PSI reaction centers (Bressan et al., 2018). The authors investigated excitation energy transfer between the digitonin-sensitive LHCII and PSI in the stroma lamellae (Schiphorst et al., 2021). They first quantified the amounts of LHCII relative to PSI in the stroma lamellae during state transitions. The densitometric analysis of the SDS–PAGE revealed that approximately 0.6 digitonin-sensitive LHCII per PSI were present in the stromal lamellae of the wild-type and Δlhca plants in State 1, whereas this number increased to 0.8–1.0 in State 2 (Schiphorst et al., 2021). These data suggest the digitonin-sensitive LHCII can enlarge PSI antenna size during state transitions. The authors went on to quantify and model excitation energy transfer in the stroma lamellae using streak-camera fluorescence decay measurements in combination with decay-associated spectra. A fast decay of fluorescence (also called a short fluorescence lifetime) generally indicates that the excitation energy is efficiently transferred to the reaction centers of photosystems. Notably, the authors observed that the sum of the increased trapping time and the transfer time from the digitonin-sensitive LHCII to the PSI core complexes increased from 74 ± 21 ps in the wild type to 122 ± 14 ps in Δlhca plants during state transitions, suggesting that LHCI can optimize energy transfer from the digitonin-sensitive LHCII to PSI (Schiphorst et al., 2021).

In sum, the work of Schiphorst et al. (2021) provides insight into LHCI-mediated excitation energy transfer from LHCII to PSI. This work emphasizes the significance of the digitonin-sensitive LHCII in adjusting antenna size of PSI during state transitions. The authors therefore propose a refined model of state transitions (Figure 1). While a minor portion of nonphosphorylated LHCII is likely to traffic excitation energy to PSI, ideally via LHCI in State 1, phosphorylated LHCII greatly increases PSI antenna size by the formation of LHCII–PSI–LHCI supercomplexes and the energetical interaction of the “additional” LHCII with PSI facilitated by LHCI. Nevertheless, experiments are still needed to define the binding sites of the additional LHCII at LHCI or PSI.

Funding

This work was supported by the grant from the Deutsche Forschungsgemeinschaft (ref WA 4599/2-1).

Conflict of interest statement. None declared.