Both authors contributed equally to this work. SUMMARYCyclic electron transport (CET) around photosystem I (PSI) plays an important role in balancing the ATP/ NADPH ratio and the photoprotection of plants. The NAD(P)H dehydrogenase complex (NDH) has a key function in one of the CET pathways. Current knowledge indicates that, in order to fulfill its role in CET, the NDH complex needs to be associated with PSI; however, until now there has been no direct structural information about such a supercomplex. Here we present structural data obtained for a plant PSI-NDH supercomplex. Electron microscopy analysis revealed that in this supercomplex two copies of PSI are attached to one NDH complex. A constructed pseudo-atomic model indicates asymmetric binding of two PSI complexes to NDH and suggests that the low-abundant Lhca5 and Lhca6 subunits mediate the binding of one of the PSI complexes to NDH. On the basis of our structural data, we propose a model of electron transport in the PSI-NDH supercomplex in which the association of PSI to NDH seems to be important for efficient trapping of reduced ferredoxin by NDH.

Photosystem I (PSI) is a pigment-protein complex required for the light-dependent reactions of photosynthesis and participates in light-harvesting and redox-driven chloroplast metabolism. Assembly of PSI into supercomplexes with light harvesting complex (LHC) II, cytochrome bf (Cytbf) or NAD(P)H dehydrogenase complex (NDH) has been proposed as a means for regulating photosynthesis. However, structural details about the binding positions in plant PSI are lacking. We analyzed large data sets of electron microscopy single particle projections of supercomplexes obtained from the stroma membrane of Arabidopsis thaliana. By single particle analysis, we established the binding position of Cytbf at the antenna side of PSI. The rectangular-shaped Cytbf dimer binds at the side where Lhca1 is located. The complex binds with its short side rather than its long side to PSI, which may explain why these supercomplexes are difficult to purify and easily disrupted. Refined analysis of the interaction between PSI and the NDH complex indicates that in total up to 6 copies of PSI can arrange with one NDH complex. Most PSI-NDH supercomplexes appeared to have 1-3 PSI copies associated. Finally, the PSI-LHCII supercomplex was found to bind an additional LHCII trimer at two positions on the LHCI side in Arabidopsis. The organization of PSI, either in a complex with NDH or with Cytbf, may improve regulation of electron transport by the control of binding partners and distances in small domains.

SummaryPhotosynthesis in plants and algae relies on the coordinated function of photosystems (PS) I and II. Their efficiency is augmented by finely-tuned light-harvesting proteins (Lhcs) connected to them. The most recent Lhcs (in evolutionary terms), Lhcb6 and Lhcb3, evolved during the transition of plants from water to land and have so far been considered to be an essential characteristic of land plants.We used single particle electron microscopy and sequence analysis to study architecture and composition of PSII supercomplex from Norway spruce and related species.We have found that there are major land plant families that lack functional lhcb6 and lhcb3 genes, which notably changes the organization of PSII supercomplexes. The Lhcb6 and Lhcb3 proteins have been lost in the gymnosperm genera Picea and Pinus (family Pinaceae) and Gnetum (Gnetales). We also revealed that the absence of these proteins in Norway spruce modifies the PSII supercomplex in such a way that it resembles its counterpart in the alga Chlamydomonas reinhardtii, an evolutionarily older organism.Our results break a deep-rooted concept of Lhcb6 and Lhcb3 proteins being the essential characteristic of land plants, and beg the question of what the evolutionary benefit of their loss could be.

The monomeric photosystem I-light-harvesting antenna complex I (PSI-LHCI) supercomplex from the extremophilic red alga represents an intermediate evolutionary link between the cyanobacterial PSI reaction center and its green algal/higher plant counterpart. We show that the PSI-LHCI supercomplex is characterized by robustness in various extreme conditions. By a combination of biochemical, spectroscopic, mass spectrometry, and electron microscopy/single particle analyses, we dissected three molecular mechanisms underlying the inherent robustness of the PSI-LHCI supercomplex: (1) the accumulation of photoprotective zeaxanthin in the LHCI antenna and the PSI reaction center; (2) structural remodeling of the LHCI antenna and adjustment of the effective absorption cross section; and (3) dynamic readjustment of the stoichiometry of the two PSI-LHCI isomers and changes in the oligomeric state of the PSI-LHCI supercomplex, accompanied by dissociation of the PsaK core subunit. We show that the largest low light-treated PSI-LHCI supercomplex can bind up to eight Lhcr antenna subunits, which are organized as two rows on the PsaF/PsaJ side of the core complex. Under our experimental conditions, we found no evidence of functional coupling of the phycobilisomes with the PSI-LHCI supercomplex purified from various light conditions, suggesting that the putative association of this antenna with the PSI supercomplex is absent or may be lost during the purification procedure.

No abstract

SUMMARYPlant photosystem II (PSII) is organized into large supercomplexes with variable levels of membrane-bound light-harvesting proteins (LHCIIs). The largest stable form of the PSII supercomplex involves four LHCII trimers, which are specifically connected to the PSII core dimer via monomeric antenna proteins. The PSII supercomplexes can further interact in the thylakoid membrane, forming PSII megacomplexes. So far, only megacomplexes consisting of two PSII supercomplexes associated in parallel have been observed. Here we show that the forms of PSII megacomplexes can be much more variable. We performed single particle electron microscopy (EM) analysis of PSII megacomplexes isolated from Arabidopsis thaliana using clear-native polyacrylamide gel electrophoresis. Extensive image analysis of a large data set revealed that besides the known PSII megacomplexes, there are distinct groups of megacomplexes with non-parallel association of supercomplexes. In some of them, we have found additional LHCII trimers, which appear to stabilize the non-parallel assemblies. We also performed EM analysis of the PSII supercomplexes on the level of whole grana membranes and successfully identified several types of megacomplexes, including those with nonparallel supercomplexes, which strongly supports their natural origin. Our data demonstrate a remarkable ability of plant PSII to form various larger assemblies, which may control photochemical usage of absorbed light energy in plants in a changing environment.

In nature, plants are continuously exposed to varying environmental conditions. They have developed a wide range of adaptive mechanisms, which ensure their survival and maintenance of stable photosynthetic performance. Photosynthesis is delicately regulated at the level of the thylakoid membrane of chloroplasts and the regulatory mechanisms include a reversible formation of a large variety of specific protein-protein complexes, supercomplexes or even larger assemblies known as megacomplexes. Revealing their structures is crucial for better understanding of their function and relevance in photosynthesis. Here we focus our attention on the isolation and a structural characterization of various large protein supercomplexes and megacomplexes, which involve Photosystem II and Photosystem I, the key constituents of photosynthetic apparatus. The photosystems are often attached to other protein complexes in thylakoid membranes such as light harvesting complexes, cytochrome b f complex, and NAD(P)H dehydrogenase. Structural models of individual supercomplexes and megacomplexes provide essential details of their architecture, which allow us to discuss their function as well as physiological significance.

Photosystem II (PSII) complexes are organized into large supercomplexes with variable amounts of lightharvesting proteins (Lhcb). A typical PSII supercomplex in plants is formed by four trimers of Lhcb proteins (LHCII trimers), which are bound to the PSII core dimer via monomeric antenna proteins. However, the architecture of PSII supercomplexes in Norway spruce[Picea abies (L.) Karst.] is different, most likely due to a lack of two Lhcb proteins, Lhcb6 and Lhcb3. Interestingly, the spruce PSII supercomplex shares similar structural features with its counterpart in the green alga Chlamydomonas reinhardtii [Kou ril et al. (2016) New Phytol. 210, 808-814]. Here we present a single-particle electron microscopy study of isolated PSII supercomplexes from Norway spruce that revealed binding of a variable amount of LHCII trimers to the PSII core dimer at positions that have never been observed in any other plant species so far. The largest spruce PSII supercomplex, which was found to bind eight LHCII trimers, is even larger than the current largest known PSII supercomplex from C. reinhardtii. We have also shown that the spruce PSII supercomplexes can form various types of PSII megacomplexes, which were also identified in intact grana membranes. Some of these large PSII supercomplexes and megacomplexes were identified also in Pinus sylvestris, another representative of the Pinaceae family. The structural variability and complexity of LHCII organization in Pinaceae seems to be related to the absence of Lhcb6 and Lhcb3 in this family, and may be beneficial for the optimization of light-harvesting under varying environmental conditions.