Abstract
Bienertia sinuspersici is a single-cell C4 plant species of which chlorenchyma cells have two distinct groups of chloroplasts spatially segregated in the cytoplasm. The central vacuole encloses most chloroplasts at the cell center and confines the rest of the chloroplasts near the plasma membrane. Young chlorenchyma cells, however, do not have large vacuoles and their chloroplasts are homogenous. Therefore, maturing Bienertia chlorenchyma cells provide a unique opportunity to investigate chloroplast proliferation in the central cluster and the remodeling of chloroplasts that have been displaced by the vacuole to the cell periphery. Chloroplast numbers and sizes increased, more notably, during later stages of maturation than the early stages. Electron tomography analyses indicated that chloroplast enlargement is sustained by thylakoid growth and that invaginations from the inner envelope membrane contributed to thylakoid assembly. Grana stacks acquired more layers, differentiating them from stroma thylakoids as central chloroplasts matured. In peripheral chloroplasts, however, grana stacks stretched out to a degree that the distinction between grana stacks and stroma thylakoids was obscured. In central chloroplasts undergoing division, thylakoids inside the cleavage furrow were kinked and severed. Grana stacks in the division zone were disrupted, and large complexes in their membranes were dislocated, suggesting the existence of a thylakoid fission machinery.
Similar content being viewed by others
Introduction
C4 photosynthesis is an adaptation in plants that suppresses photorespiration and increases efficiency of water and nitrogen use1. A key feature of C4 photosynthesis is the dimorphism of chloroplasts in photosynthetic tissues. One type of chloroplast concentrates atmospheric CO2 into a C4 acid intermediate. CO2 is regenerated in or near the other type of chloroplast where Rubisco brings the CO2 into the Calvin cycle. These differential functions lead to dissimilar energy requirements and ultrastructural features between the two classes of chloroplasts2.
In Kranz anatomy C4 plants, such as maize, the two classes of chloroplasts are segregated in mesophyll cells and bundle sheath cells3. In maize, an NADP-ME type C4 species, chloroplasts in mesophyll cells have higher concentrations of grana stacks because they contain photosystem II (PSII) associated with thylakoid stacking. Linear electron flow operates in the chloroplasts in mesophyll cells to produce both NADPH and ATP. Chloroplasts in bundle sheath cells have fewer stacked thylakoids because they contain less PSII and produce ATP through cyclic electron flow4,5.
It was assumed that C4 photosynthesis was synonymous with Kranz anatomy, but in the early 2000s, several species of single-cell C4 (SCC4) plants were discovered3,4. Unlike plants with Kranz anatomy, the two classes of chloroplasts in SCC4 plants are spatially segregated within a single cell and exhibit dimorphism related to varying abundances of PSII and photosystem I (PSI) in their thylakoids. At least four species, Bienertia cycloptera, Bienertia sinuspersici, Bienertia kavirense, and Suaeda aralocaspica, have been found to exhibit single-cell C4 photosynthesis5. Bienertia sinuspersici (Bienertia), a member of the family Chenopodiaceae, has become a model species in the characterization of SCC4 photosynthesis.
In Bienertia-type SCC4 plants, chlorenchyma cells have central chloroplasts (CCs) and peripheral chloroplasts (PCs). Bienertia CCs are located in a tightly packed central cytosolic compartment (CCC). PCs are dispersed around the outer edges of the cell near the plasma membrane. PCs generate a C4 organic acid from atmospheric CO2, and the organic acid is decarboxylated in mitochondria in the CCC to supply CO2 to CCs. In contrast to the NADP-ME type in maize, Bienertia operates in NAD-ME type C4 photosynthesis. The CO2-capturing PCs have fewer grana stacks than the Rubisco-containing CCs because of CCs’ requirements for reducing power as well as ATP2,3.
Biogenesis of SCC4 systems is of interest to plant biologists due to their unique cytoplasmic organization and regulation of photosynthetic genes6. Young Bienertia chlorenchyma cells have uniform chloroplasts operating in C3 photosynthesis and their cytoplasm is not partitioned7. As the chlorenchyma cells mature, vacuoles enlarge to enclose chloroplasts and mitochondria in the cell center and chloroplasts excluded from CCC become PCs8. In mature Bienertia SCC4 cells, CCs and PCs contain distinct sets of proteins for C4 photosynthesis, and they exchange small molecules through channels composed of cytoplasmic strands. Rubisco large subunit gene in the plastid genome are transcribed specifically in CCs, indicating that transcription in CCs is regulated differently from that in PCs7. A mechanism for selective targeting of chloroplast proteins encoded in the nuclear genome to one of the two chloroplast types has been characterized recently. The N-terminal signal peptides of proteins destined for PCs have two components, one that facilitates general entry into the chloroplast and a second one that inhibits import into CCs9,10. In agreement with their macromolecular compositions, CCs and PCs exhibit differential ultrastructural characteristics, but there have been no detailed morphometric analyses to show how CCs and PCs diverge from a homogenous pool of chloroplasts in young in Bienertia chlorenchyma cells.
Plastids divide via binary fission, and hourglass-shaped dividing proplastids have been observed in electron microscopy (EM) imaging of shoot apical meristem cells11. Constriction of the plastid envelope membranes is mediated by two ring complexes, one in the stroma, termed Z-ring, and the other in the cytosol consisting of a dynamin-related protein12. The thylakoid membrane is densely populated with massive photosynthetic complexes, and some of them constitute close-packed two-dimensional arrays. It seems a daunting task to bisect piles of thylakoid membranes in grana stacks for chloroplast division but almost none is known about thylakoid fission. However, it has been elusive how thylakoids are partitioned into two daughter chloroplasts during division13.
A distinctive feature of the Bienertia chloroplast is that large numbers of CCs are packed in the central cytosol; these chloroplasts multiply as chlorenchyma cells develop7,8. The Bienertia CC cluster provides a unique opportunity to examine thylakoids in dividing chloroplast with transmission electron microscopy (TEM)/electron tomography (ET). Most plastid division in plants happens in meristematic cells in which proplastids have primitive thylakoids. It is challenging to find dividing chloroplasts by TEM because chloroplasts in the leaf divide less frequently than proplastids in the meristematic zone, and leaf cells are large, highly vacuolated11,14. It is critical for TEM analysis to acquire unambiguous micrographs of dividing chloroplasts repeatedly in leaf samples of which subcellular structures are preserved close to their native states. The CC division occurs in a restricted volume in the cell center, making it easy to identify many dividing chloroplasts under TEM and capture images of their thylakoids. Furthermore, well-developed thylakoids appear in chlorenchyma cells of young Bienertia leaves15, facilitating high-pressure freezing fixation of the leaf samples.
ET imaging of cryofixed chloroplast has been an essential approach to determine 3D architectures of the photosynthetic organelle as well as macromolecular structures of its constituents that are close to their native states16,17,18,19. Taking advantage of these advanced microscopy techniques, we characterized the gradual modifications of thylakoid architectures in CCs and PCs as SCC4 is established in Bienertia chlorenchyma cells. We were able to delineate thylakoid growth from inner membrane invaginations that expand over planar thylakoid surfaces to add to a layer in the grana stack. Additionally, the concentrated chloroplasts at the central cytoplasm facilitated EM analyses of chloroplast division. Thylakoid division involved distortion of grana stacks at the constricted site before severing, and displacement of PSII components from the deformed stacks.
Results
Morphometric analysis of chloroplasts in maturing Bienertia chlorenchyma cells
In order to monitor the maturation process of dimorphic chloroplasts in Bienertia sinuspersici, we examined chloroplasts in chlorenchyma cells at 1st stage, 2nd stage, 3rd stage, and the mature stage as defined in Park et al.8 using confocal laser scanning microscopy and TEM. The 1st stage and 2nd stage cells were sampled from leaves less than 0.3-cm long, 3rd stage cells from leaves 0.4-cm to 0.5-cm long, and mature stage cells from the tip region of leaves 1.5-cm to 2.0-cm long (Fig. 1A). In isolated mature chlorenchyma cells, two groups of chloroplasts were observed: densely packed chloroplasts in the central cytosol and chloroplasts scattered in the cell periphery (Fig. 1B).
The 1st stage cells were spherical, and their cytoplasms were occupied mostly by chloroplasts (Fig. 1C). Vacuoles divided the cytosol into central and peripheral regions in older cells, but no large vacuoles were observed in 1st stage cells (Fig. 1D–F). Vacuole development was discerned from 2nd stage, separating some chloroplasts to the cell periphery, and by the 3rd stage, chlorenchyma cells had elongated to take on elliptical shapes and vacuoles segregated PCs from CCs. Cells at the mature stage grew into a pill shape with a corresponding increase in the size and number of chloroplasts (Fig. 1H–M). In TEM micrographs, PCs appear oppressed in the thin cytosol between the vacuole and the plasma membrane (Fig. 1M). Numerous mitochondria were intermixed with chloroplasts in the CCC (Fig. 1K–M).
We were able to estimate cell sizes, cell numbers, and sizes of chloroplasts in the cell at the four stages from confocal laser scanning microscopy data (Fig. 1G). Sizes of mature chlorenchyma cells reached 1,330,971 µm3 (s.d. = 487,825 µm3, n = 14) with the largest increase between 3rd stage (103,256 µm3 (s.d. = 71395 µm3, n = 13)) and mature stage (Fig. 1G, left panel). Both CCs and PCs proliferated gradually until the mature stage (Fig. 1G, middle panel). The average chloroplast numbers increased in 1st stage cells to mature cells from 81 (s.d. = 24.64, n = 13) to 137 (s.d. = 44.68, n = 14) for CCs and from 24 (s.d. = 9.82, n = 13) to 56 (s.d. = 16.59, n = 14) for PCs. The volumes of both types of chloroplasts remained similar through stages 1, 2, and 3 (Fig. 1G, right panel). Chloroplasts in mature stage cells were significantly larger than those of 3rd stage cells, however. The volume of CCs increased 7 fold from 76 m3 (s.d. = 45.75 m3, n = 262) to 541 m3 (s.d. = 299.28 µm3, n = 303), and the volume of PCs increased 4 fold from 80 m3 (s.d. = 41.41 m3, n = 218) to 387 m3 (s.d. = 222.10 µm3, n = 262) during the period.
Electron tomography analysis of thylakoid assembly
It has been shown by TEM that thylakoid organizations of the CCs and PCs are dissimilar in mature Bienertia chlorenchyma cells7. We investigated the biogenesis of dimorphic chloroplasts using electron tomography (ET), which allowed three dimensional morphometric analysis of membranous compartments20. No ultrastructural differences were apparent in the thylakoid architectures of the two types of chloroplasts in 2nd stage cells, but in 3rd stage chloroplasts, we were able to observe differential architectures in the thylakoids that became even more apparent in mature cells (Fig. 2A–L). The 3rd stage CCs became enriched with grana stacks (2M, green brackets) and short stroma thylakoids connecting the stacks were seen. Grana stacks widened in PCs without obtaining additional layers, making it difficult to identify discrete stacks (Fig. 2N). By the mature stage, the two types of chloroplasts exhibited the structural dimorphism reported previously15. Mature CCs were ovoid and grana stacks and unstacked thylakoids were well differentiated in their stroma. Their grana stacks had 2–5 layers and some had as many as 10–11 layers (Fig. 2C,F,O,Q). Mature PCs were flattened into a discus-like shape which appeared as a rod in 2D images (Fig. 2I,L). The thylakoid composition PCs was remarkably skewed toward unstacked stroma thylakoids (Fig. 2P,Q), although there were grana stacks with 2–3 layers (Fig. 2P, green brackets).
We generated 3D models of thylakoid membranes and envelope membranes in the tomograms (Fig. 2D–F,J–L) and calculated volume ratios of thylakoids to their chloroplasts. Relative volumes of thylakoids increased in both types of chloroplasts as they matured (Fig. 2R), indicating that thylakoids proliferate more rapidly than chloroplasts enlarge. Given that chloroplast sizes increased most between the 3rd stage and the mature stage (Fig. 1G), we inferred that thylakoid biogenesis is most active in 3rd stage chloroplasts. From the 3rd stage, volumes of thylakoids in CCs were larger than those in PCs (Fig. 1G).
Invaginations from the inner envelope membrane and thylakoid assembly
To investigate how thylakoids assemble in the Bienertia chloroplasts, we examined electron tomograms from CCs of 3rd stage cells, the stage of fastest growth. Round membrane outlines were frequently discerned in the stroma close to the envelope and some of them were connected to the envelope membranes (Fig. 3A,B, arrowheads). Similar invaginations were observed in proplastids of cotyledon cells of germinating Arabidopsis thaliana seedlings (Liang et al.21).
We generated 3D models of membrane elements in the vicinity of the ingrowths including the existing thylakoid network and IEM to better understand their relationship (Fig. 3A–D). It is evident that the volume of thylakoid membrane increases in growing chloroplasts. When we arranged IEM invaginations according to their surface areas, we discerned vesicle-shaped ingrowths (Fig. 3E, Movie S1) and more elaborate ingrowths that seemed to have formed by lateral fusion of isolated ingrowths (Fig. 3G, Movie S3). These thylakoid precursors appeared to expand over a grana stack when it encounters its top surface (Fig. 3H, Movie S4). Ingrowths in close proximity with stroma thylakoid (Fig. 3F, arrowhead, Movie S2) were also observed, suggesting that this may constitute a membrane contact site between the IEM and peripheral membranes of the granum. We were able to identify discrete membrane compartments overlaying a thylakoid that seem to have separated from IEM (Fig. 3I, Movie S5), suggesting that the IEM ingrowths can be detached before they merge with the thylakoid system. However, no free-floating vesicles in the stroma were detected.
Division of chloroplasts within an existing thylakoid network
The number of chloroplasts increased during the Bienertia chlorenchyma cell development (Fig. 1G, left panel), and chloroplasts of chlorenchyma cells as early as 2nd stage accumulate grana stacks (Figs. 1 and 2). Grana stacks are packed with large protein complexes of the electron transport chain embedded in the thylakoid membrane, and these stacks could impede separation of daughter chloroplasts. To understand how grana stacks are partitioned in dividing chloroplasts, we examined 2nd stage and 3rd stage Bienertia cells with electron microscopy/tomography.
Dividing CCs were frequently observed in the central cytoplasmic regions of 2nd stage and 3rd stage cells where dozens of chloroplasts are clustered (Fig. 4A). Very few dividing chloroplasts were seen in the cell periphery under TEM because PCs are rare and scattered over large areas. Dividing chloroplasts were elongated and peanut-shaped due to constriction near the center (arrowheads in Fig. 4A). Interestingly, thylakoids appeared sharply kinked at the division plane at the constricted site (Fig. 4B,C). Electron tomography imaging confirmed that the stack architecture was disrupted. Furthermore, some thylakoids were disjoined as if they had been severed (Fig. 4C,D, arrowheads).
We examined dividing chloroplasts in cryofixed Arabidopsis cotyledons to test whether the thylakoid remodeling in dividing Bienertia chloroplasts is conserved. Thylakoids began to fold into 2 or 3 layers at 60 hours after imbibition (HAI), and grana stacks constituted the majority of thylakoids in chloroplasts by 120 HAI as previously reported (Liang et al.21). It was more challenging to detect dividing chloroplasts in 120 HAI Arabidopsis cotyledon cells by TEM than it was in Bienertia cells.
As in dividing chloroplasts of Bienertia, thylakoid membranes in the division zone were highly twisted in 120 HAI Arabidopsis chloroplasts (Fig. 4E–H). Electron-dense particles of approximately 10 nm constituting a 2D crystalline array were discerned in appressed membranes of grana stacks (Fig. 4I,J), but these particles were not observed in the kinked thylakoid membranes (Fig. 4K,L). We hypothesize that these particles that have dislocated from the fission sites correspond to PSII. To test this, we localized a subunit of the oxygen-evolving complex of PSII, PsbP, in 120 HAI Arabidopsis cotyledon cells using immunogold particles. In dividing chloroplasts, PsbP-specific gold particles were absent from the disrupted thylakoids, indicating that the thylakoid remodeling in dividing chloroplasts involves PSII removal (Fig. 4M,N).
Meristem cells have proplastids, and they do not have extensive grana stacks22. We examined dividing proplastids to determine whether their prothylakoids are severed when the ring complexes at the envelope squeeze the envelope membranes. In consecutive sections of an Arabidopsis meristem cells where two daughter proplastids are about to separate, we observed a connecting neck and a tubular prothylakoid (Fig. 4O). This observation indicates that prothylakoids are not disconnected in preparation for the shrinking envelope membranes probably because they do not interfere with plastid constriction.
Discussion
We analyzed development of dimorphic chloroplasts in chlorenchyma cells of Bienertia sinuspersici with structural biology approaches. Our results indicate that thylakoids assemble through growth and fusion of inner membrane invaginations and that disjoining of thylakoids precedes constriction of the envelope membranes. The cytoplasmic partitioning through the vacuole development in the early stages separated the chloroplasts into central and peripheral groups. The increases in the cell sizes, numbers of chloroplasts, and chloroplast sizes accelerated in the 3rd stage, culminating in the pill-shape of mature cells with fully diversified chloroplasts. These characteristics of maturation are different from what Koteyeva et al. (2016) reported because the stage classification scheme in their research was based on the developmental gradient in a single leaf from the base to the tip7.
Our TEM and ET analyses indicated that the 3rd stage is critical in the establishment of the Bienertia dimorphic chloroplasts. In 1st stage and 2nd stage cells, no significant ultrastructural variations were detected among chloroplasts. Thylakoids with 2 or 3 layers were abundant in both types of chloroplasts in these early stages. In CCs, beginning in the 3rd stage, the number of layers per thylakoid stack increased and discrete stacks appeared. By contrast, thylakoids spread laterally in PCs, and they lost the layered architecture. That protoplasts from “young cells” are capable of selective protein targeting to the PC was demonstrated by Wimmer et al. (2017)9; these authors took advantage of this targeting to characterize mechanisms of PC-specific protein accumulation. Considering cell structure and size, 3rd stage cells correspond to “young cells”. It is likely that the morphological variations in the thylakoids in 3rd stage cells result from targeting of proteins involved in thylakoid stacking to CCs and their exclusion from PCs; these proteins include those that stabilize the stack margins such as LHCB subunits and CURT1 family proteins18,23,24. Lateral stretching of thylakoids (Fig. 2I–L) and lack of grana-forming proteins in 3rd stage PCs could lead to their distinct thylakoid architecture in the mature stage. Offermann et al. (2015) isolated central and peripheral chloroplasts to measure relative abundances of PSII and PSI subunits in the chloroplasts using mass spectrometry analysis6. In agreement with the preferential grana stacking in CCs, LHCB proteins mediating membrane stacking are more abundant (up to two-fold) in CCs than PCs.
Thylakoid assembly is the hallmark of chloroplast biogenesis25,26. Conversions of proplastids and etioplasts into chloroplasts have been characterized by ET, and these studies led to our current understanding of construction of the photosynthetic apparatus21,22,27. We provide evidence for a mechanism of thylakoid assembly involving IEM invaginations that enlarge and become integrated with existing thylakoids. They also formed contact sites with membranes of planar thylakoids (Fig. 3). In Arabidopsis proplastids, thylakoid assembly takes place throughout the stroma in a linear order21,27. By contrast, the Bienertia chloroplast has an elaborate thylakoid network in the stroma and the construction process is confined to the vicinity of the envelope membrane (within 100 nm from IEM). Given that we did not identify vesicle-like carriers in the stroma, roles of vesicle carriers reported earlier from studies in higher plants28 seems limited in the Bienertia thylakoid assembly.
The intermediate structures derived from IEM resemble what was proposed in Bastien et al.29. We have detected IEM ingrowth, the formation of flattened invaginations, and the development of contact sites between IEM and planar thylakoids (Fig. 3). These membrane junctions are reminiscent of the endoplasmic reticulum (ER) - chloroplast contact sites. Galactolipids monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG) comprise >75% of the thylakoid membrane and IEM. The rest is composed of two anionic lipids, sulfoquinovosyldiacylglycerol (SQDG) and phosphatidylglycerol (PG)30. These four lipids are partly or entirely transferred from ER. The lipid compositions of IEM and thylakoid membrane are broadly similar. Expansion of thylakoid membranes that cannot directly interact with IEM may acquire lipid molecules through the IEM-thylakoid contact sites.
FZL is a dynamin-related protein targeted to the chloroplast. Arabidopsis mutant lines in which FZL is inactivated have fragmented thylakoids, suggesting that FZL is involved in thylakoid membrane fusion21,31. Recently, Findinier et al. demonstrated that FZL in Chlamydomonas mediates thylakoid membrane fusion using an ingenious biochemical assay32. FZL localizes to thylakoid margins suggesting that grana stacks and stroma lamellae are interlinked through their periphery. FZL orthologs of Bienertia could contribute to combining the thylakoid precursors derived from IEM to margins of grana stacks or thylakoid lamellae.
In dividing chloroplasts of Bienertia chlorenchyma cells as well as Arabidopsis cotyledon cells, kinked thylakoid membranes were seen along the midline connecting the division furrows (Fig. 4). Peanut-shaped chloroplasts that appear to go through fission have been documented before and thylakoids constricted in the neck region have been observed in several land plant species33,34. But the unusual thylakoid architecture that we have observed in this study has not been reported previously. We infer that chemical fixation may fail to preserve the intriguing U-shaped thylakoids. It has been demonstrated that high-pressure freezing and subsequent freeze substitution can capture transient and fragile subcellular structures in plant cells that conventional TEM protocols cannot35,36. The thylakoid remodeling and splitting visualized in our study benefited from the congregated chloroplast arrangement in growing Bienertia chlorenchyma cells and the improved sample preservation methods employed for electron microscopy37. It was possible to discern dividing chloroplasts in 120 HAI Arabidopsis cotyledon cell samples, although these were extremely rare, and we examined the displacement of thylakoid membrane proteins in the samples. However, we failed to identify dividing chloroplasts in true leaf cells of Arabidopsis seedlings.
Components of the plastid division machinery have been identified and molecular mechanisms of their function have been characterized38. Two ring complexes, the DRP5B ring on OEM and the Z-ring inside IEM, assemble at the midzone of a dividing chloroplast to constrict the envelope membrane39,40,41. Linker complexes connecting the two rings on the opposite sides of the envelope regulate formation and contraction of the rings42. However, almost no information about thylakoid fission is available13. Based on our observations, we speculate that the thylakoid severing machinery bends thylakoid membranes and disrupts the array of PSII-LHCB complexes initiating their relocation, probably by exerting forces (Fig. 4P). Given that thylakoid cutting occurs precisely at the cleavage plane, it is conceivable that assembly and operation of the thylakoid division machinery is coordinated with those of the ring complexes. Because of the distance between the two apparatuses, it is puzzling how the contractile ring complexes at the envelope membranes communicate with the thylakoid division machinery in the stroma. The chloroplast Min system controls position of Z-ring formation and the inner membrane ring seems to trigger subsequent assembly of the linker and outer membrane ring complexes43. Establishment of the thylakoid fission machinery could be regulated by the Z-ring. Alternatively, the Min system might control assembly and activation of the thylakoid disjoining directly.
Conclusions
Bienertia sinuspersici, a single-cell C4 plant species, is a useful model organism for studying the development of dimorphic chloroplasts and the selective targeting of nucleus-encoded plastid proteins to central and peripheral chloroplasts. Through light and electron microscopy analyses of Bienertia chloroplasts, we delineated the assembly process of thylakoid membranes in a growing chloroplast, observed the gradual gain of grana stacks in the central chloroplasts and their loss in peripheral chloroplasts, and the fission of grana thylakoids in dividing chloroplasts. The observations of thylakoid membranes and the division of chloroplasts presented here diverge from what has been seen in those of proplastids, expanding our understanding of chloroplast biogenesis in higher plants. These observations were made possible by state-of-the-art cryo-preservation and three-dimensional electron microscopy techniques, and investigations of plastids employing these tools in other species are well warranted. Mitochondria are vital in the NAD-ME type C4 photosynthesis occurring in Bienertia leaf cells. Further work is needed to characterize the subgroups of mitochondria that associate with the tightly-packed CCs.
Methods
Plant growth conditions
Bienertia sinuspersici samples were obtained from Dr. Joonho Park (Seoul National University of Science and Technology). The plants were propagated asexually and grown in magenta boxes with full strength Murashige-Skoog (MS) media (1x MS media, 10 mM MES, 10 mM NaCl, 2% sucrose, 0.8% phytoagar, pH 5.8, DW) in a growth chamber (Cat No. MLR-352H-PB, Panasonic, Japan) at a photon flux intensity of 50–100 mol quanta m−2 s−1 of light under a 16 h/ 8 h light/dark cycle, with temperature of 20–22 °C. Afterwards, Bienertia samples are transferred onto soil and grown in a greenhouse with 100–200 µmol quanta m−2 s−1 of light where they are watered three times a week with 50 mM NaCl and given a nutritional supplement (1 g/L, NPK 5.1–10–5, BIO-NEX, South Korea) once a week. 2–4 month old plants were used for the experiments. The leaves were then cut and sized to different stages as shown in Fig. 1A. Arabidopsis seedlings were grown as described in Lee et al.44. Briefly, the seeds were germinated and grown on 0.75% phytoagar plates supplemented with ½ MS salt (0.5 g/l, pH 5.7) under continuous light at a photon flux density of 120 mol quanta m−2 s−1 in a growth chamber at 22 °C before their cotyledons were dissected for high-pressure freezing and electron microscopy experiments.
Confocal laser scanning microscopy
Isolated chlorenchyma cells were examined using confocal laser scanning microscopes FluoView FV1000 (Olympus, Japan) & TCS SP5 II MP with SMD (Leica Microsystems, Germany) to assess their chlorophyll autofluorescence. Stacked images of the chloroplast in Fig. 1C–F were prepared using FV10-ASW Ver. 3.1 Viewer (Olympus, Japan) and other images were processed using ImageJ (LOCI, University of Wisconsin, USA).
High-pressure freezing, sample processing, electron microscopy, and electron tomography
High-pressure freezing, freeze substitution, resin embedding, and ultra-microtomy were performed as described45. Bienertia chlorenchyma tissue and Arabidopsis cotyledon tissue were dissected and rapidly frozen with an HPM100 high-pressure freezer (Leica Microsystems, Germany). Two different freeze-substitution protocols were used and processed in an EM AFS2 freeze-substitution machine (Leica Microsystems, Germany). For ultrastructural analysis, the samples were freeze-substituted at −80 °C for 72 h in 2% osmium tetroxide (OsO4) dissolved in anhydrous acetone, and excess OsO4 was removed by rinsing with precooled acetone. After being slowly warmed up to room temperature over 48 h, tissue samples were separated from planchettes and embedded in Embed-812 resin (Electron Microscopy Sciences, USA). For immunogold labelling samples, anhydrous acetone containing 0.25% glutaraldehyde and 0.1% uranyl acetate was used for freeze-substitution media at −80 °C for 72 h before being slowly warmed up to −45 °C where they were embedded with Lowicryl HM20 resin (Electron Microscopy Sciences, USA) and polymerized under ultraviolet light for 24 h at −45 °C.
The blocks were trimmed and sectioned into 100 nm sections using a diamond knife (Diatome, Switzerland) and an UC7 ultramicrotome (Leica Microsystems, Germany) onto slot grids (Electron Microscopy Sciences, USA) and examined with a Hitachi 7400 TEM (Hitachi-High Technologies, Japan) operated at 80 kV. For ultrastructural analysis, 250 nm thick sections were collected on formvar-coated copper slot grids (Electron Microscopy Sciences, USA) and stained with 2% uranyl acetate in 70% methanol followed by Reynold’s lead citrate (Electron Microscopy Sciences, USA). A 200-kV Tecnai F20 intermediate voltage electron microscopy (Thermo-Fischer, USA) was used to collect tilt series from ±60 at 1.5° intervals in brightfield TEM mode and from ±51° at 1.5° intervals in scanning transmission electron microscopy (STEM) mode. For dual-axis tomography analysis, tilt series around two orthogonal axes were acquired from each section using FEI TEM Tomography and FEI STEM Tomography (Thermo-Fischer, USA)46.
The TEM/ET data sets that we produced for this study include 675 electron micrographs (355 Bienertia, 320 Arabidopsis) of 52 thin sectioned cells, ET reconstruction of 16 chloroplasts (10 Bienertia, 6 Arabidopsis) with a total of ~5200 slice images, and STEM ET reconstruction of 3 Bienertia chloroplasts with a total of 256 slice images. Each micrograph and slice image was analyzed individually. The IMOD (University of Colorado Boulder, USA) software package was used for the reconstruction of tomograms as described in Toyooka and Kang47
Immunogold labelling
The immunogold labelling experiments were performed according to the protocol explained by Kang48. Antibodies for PsbP were donated by Kenneth Cline (University of Florida, USA)49.
Automatic segmentation from tomogram
The 3D models of the more complete chloroplast were generated with a total variation-based convex optimization algorithm developed for automatic segmentation of tomograms with incomplete frequency information50. The computer codes are written in MATLAB (MathWorks, USA). The data term, a z-axis-dependent threshold, and the regularizing terms were specifically fine-tuned for each tomogram to achieve the most visibly sensible results. The segmented images generated above were then meshed with a MATLAB implementation of the marching cubes algorithm51. The meshes were then exported to.stl files with the stlwrite function52, which were opened in MeshLab (ISTI-CNR, Italy) for visualization.
Thylakoid density measurements
Five lines orthogonal to the longest axis of the chloroplast were drawn equidistant from each other. At each line, the thickness of the thylakoid membranes in number of layers was counted using ImageJ (LOCI, University of Wisconsin, USA) along with the thickness of the chloroplast. This data was then quantified using Microsoft Excel (Microsoft, USA).
References
Gowik, U. & Westhoff, P. The path from C3 to C4 photosynthesis. Plant Physiology 155, 56–63 (2011).
Edwards, G. E., Franceschi, V. R. & Voznesenskaya, E. V. Single-cell C(4) photosynthesis versus the dual-cell (Kranz) paradigm. Annual Review of Plant Biology 55, 173–196 (2004).
Voznesenskaya, E. V., Franceschi, V. R., Pyankov, V. I. & Edwards, G. E. Anatomy, chloroplast structure and compartmentation of enzymes relative to photosynthetic mechanisms in leaves and cotyledons of species in the tribe Salsoleae (Chenopodiaceae). J Exp Bot 50, 1779–1795 (1999).
Voznesenskaya, E. V. et al. Proof of C4 photosynthesis without Kranz anatomy in Bienertia cycloptera (Chenopodiaceae). Plant J 31, 649–662 (2002).
Sharpe, R. M. & Offermann, S. One decade after the discovery of single-cell C4 species in terrestrial plants: what did we learn about the minimal requirements of C4 photosynthesis? Photosynthesis Research 119, 169–180 (2013).
Offermann, S. et al. Developmental and Subcellular Organization of Single-Cell C 4Photosynthesis in Bienertia sinuspersiciDetermined by Large-Scale Proteomics and cDNA Assembly from 454 DNA Sequencing. J. Proteome Res. 14, 2090–2108 (2015).
Koteyeva, N. K., Voznesenskaya, E. V., Berry, J. O., Cousins, A. B. & Edwards, G. E. The unique structural and biochemical development of single cell C4 photosynthesis along longitudinal leaf gradients in Bienertia sinuspersici and Suaeda aralocaspica (Chenopodiaceae). J Exp Bot 67, 2587–2601 (2016).
Park, J., Knoblauch, M., Okita, T. W. & Edwards, G. E. Structural changes in the vacuole and cytoskeleton are key to development of the two cytoplasmic domains supporting single-cell C4 photosynthesis in Bienertia sinuspersici. Planta 229, 369–382 (2008).
Wimmer, D., Bohnhorst, P., Shekhar, V., Hwang, I. & Offermann, S. Transit peptide elements mediate selective protein targeting to two different types of chloroplasts in the single-cell C4 species Bienertia sinuspersici. Sci. Rep. 7, 41187 (2017).
Erlinghaeuser, M., Hagenau, L., Wimmer, D. & Offermann, S. Development, subcellular positioning and selective protein accumulation in the dimorphic chloroplasts of single-cell C4species. Current Opinion in Plant Biology 31, 76–82 (2016).
Pyke, K. Plastid division and development. The Plant Cell Online 11, 549–556 (1999).
Yoshida, Y., Miyagishima, S.-Y., Kuroiwa, H. & Kuroiwa, T. The plastid-dividing machinery: formation, constriction and fission. Current Opinion in Plant Biology 15, 714–721 (2012).
Chen, C., MacCready, J. S., Ducat, D. C. & Osteryoung, K. W. The Molecular Machinery of Chloroplast Division. Plant Physiology 176, 138–151 (2018).
Pyke, K. A. & LEECH, R. M. Chloroplast Division and Expansion Is Radically Altered by Nuclear Mutations in Arabidopsis thaliana. Plant Physiology 99, 1005–1008 (1992).
Voznesenskaya, E. V. et al. Differentiation of cellular and biochemical features of the single-cell C4 syndrome during leaf development in Bienertia cycloptera (Chenopodiaceae). American Journal of Botany 92, 1784–1795 (2005).
Austin, J. R. & Staehelin, L. A. Three-dimensional architecture of grana and stroma thylakoids of higher plants as determined by electron tomography. Plant Physiology 155, 1601–1611 (2011).
Austin, J. R., Frost, E., Vidi, P.-A., Kessler, F. & Staehelin, L. A. Plastoglobules are lipoprotein subcompartments of the chloroplast that are permanently coupled to thylakoid membranes and contain biosynthetic enzymes. Plant Cell 18, 1693–1703 (2006).
Daum, B., Nicastro, D., Austin, J., McIntosh, J. R. & Kühlbrandt, W. Arrangement of photosystem II and ATP synthase in chloroplast membranes of spinach and pea. The Plant Cell Online 22, 1299–1312 (2010).
Engel, B. D. et al. Native architecture of the Chlamydomonaschloroplast revealed by in situ cryo-electron tomography. eLife 4, 3583–29 (2015).
Staehelin, L. A. & Kang, B.-H. Nanoscale architecture of endoplasmic reticulum export sites and of Golgi membranes as determined by electron tomography. Plant Physiology 147, 1454–1468 (2008).
Liang, Z. et al. Thylakoid-Bound Polysomes and a Dynamin-Related Protein, FZL, Mediate Critical Stages of the Linear Chloroplast Biogenesis Program in Greening Arabidopsis Cotyledons. The Plant Cell Online 30, 1476–1495 (2018).
Charuvi, D. et al. Gain and loss of photosynthetic membranes during plastid differentiation in the shoot apex of Arabidopsis. The Plant Cell Online 24, 1143–1157 (2012).
Armbruster, U. et al. Arabidopsis CURVATURE THYLAKOID1 proteins modify thylakoid architecture by inducing membrane curvature. The Plant Cell Online 25, 2661–2678 (2013).
Pribil, M., Labs, M. & Leister, D. Structure and dynamics of thylakoids in land plants. J Exp Bot 65, 1955–1972 (2014).
Pogson, B. J., Ganguly, D. & Albrecht-Borth, V. Insights into chloroplast biogenesis and development. BBA - Bioenergetics 1847, 1017–1024 (2015).
Bastien, O. et al. New Insights on Thylakoid Biogenesis in Plant Cells. International Review of Cell and Molecular Biology 323, 1–30 (Elsevier Inc., (2016).
Kowalewska, Ł., Mazur, R., Suski, S., Garstka, M. & Mostowska, A. Three-Dimensional Visualization of the Tubular-Lamellar Transformation of the Internal Plastid Membrane Network during Runner Bean Chloroplast Biogenesis. The Plant Cell Online 28, 875–891 (2016).
Lindquist, E. & Aronsson, H. Chloroplast vesicle transport. Photosynthesis Research 138, 361–371 (2018).
Bastien, O. et al. In International Review of Cell and Molecular Biology 323, 1–30 (Elsevier, (2016).
LaBrant, E., Barnes, A. C. & Roston, R. L. Lipid transport required to make lipids of photosynthetic membranes. Photosynthesis Research 138, 345–360 (2018).
Gao, H., Sage, T. L. & Osteryoung, K. W. FZL, an FZO-like protein in plants, is a determinant of thylakoid and chloroplast morphology. Proc Natl Acad Sci USA 103, 6759–6764 (2006).
Findinier, J., Delevoye, C. & Cohen, M. M. The dynamin-like protein Fzl promotes thylakoid fusion and resistance to light stress in Chlamydomonas reinhardtii. PLoS Genetics 15, e1008047–30 (2019).
Leech, R. M., Thomsom, W. W. & Platt Aolia, K. A. Observations on the Mechanism of Chloroplast Division in Higher Plants. New Phytologist 87, 1–9 (1981).
Whatley, J. M. Mechanisms and morphology of plastid division. In SA Boffey, D Lloyd, eds, Division and Segregation of Organelles. Cam- bridge University Press, Cambridge, 63–84 (1988).
Kang, B.-H. & Staehelin, L. A. ER-to-Golgi transport by COPII vesicles in Arabidopsis involves a ribosome-excluding scaffold that is transferred with the vesicles to the Golgi matrix. Protoplasma 234, 51–64 (2008).
Donohoe, B. S., Kang, B.-H., Gerl, M. J., McMichael, C. M. & Staehelin, L. A. Cis-Golgi cisternal assembly and biosynthetic activation occur sequentially in plants and algae. Traffic 14, 551–567 (2013).
Karahara, I. & Kang, B.-H. In Plant Cell Morphogenesis (eds. Zarsky, V. & Cvrckova, F.) 1080, 147–157 (Humana Press, (2013).
Miyagishima, S.-Y. Mechanism of plastid division: from a bacterium to an organelle. Plant Physiology 155, 1533–1544 (2011).
TerBush, A. D., Yoshida, Y. & Osteryoung, K. W. FtsZ in chloroplast division: structure, function and evolution. Curr Opin Cell Biol 25, 461–470 (2013).
Gao, H., Kadirjan-Kalbach, D., Froehlich, J. E. & Osteryoung, K. W. ARC5, a cytosolic dynamin-like protein from plants, is part of the chloroplast division machinery. Proc Natl Acad Sci USA 100, 4328–4333 (2003).
Miyagishima, S.-Y. et al. A Plant-Specific Dynamin-Related Protein Forms a Ring at the Chloroplast Division Site. Plant Cell 15, 655–665 (2003).
Wang, W. et al. Structural insights into the coordination of plastid division by the ARC6–PDV2 complex. Nature Plants 3, 17011 (2017).
Nakanishi, H., Suzuki, K., Kabeya, Y. & Miyagishima, S.-Y. Plant-Specific Protein MCD1 Determines the Site of Chloroplast Divisionin Concert with Bacteria-Derived MinD. Current Biology 19, 151–156 (2009).
Lee, K.-H. et al. Defective chloroplast development inhibits maintenance of normal levels of abscisic acid in a mutant of the Arabidopsis RH3 DEAD-box protein during early post-germination growth. Plant J 73, 720–732 (2013).
Wang, P., Chen, X., Goldbeck, C., Chung, E. & Kang, B.-H. A distinct class of vesicles derived from the trans-Golgi mediates secretion of xylogalacturonan in the root border cell. Plant J 92, 596–610 (2017).
Kang, B.-H. STEM Tomography Imaging of Hypertrophied Golgi Stacks in Mucilage-Secreting. Cells. 1496, 55–62 (2016).
Toyooka, K. & Kang, B.-H. In Plant Cell Morphogenesis (eds. Zarsky, V. & Cvrckova, F.) 1080, 159–170 (Humana Press, Totowa, NJ, (2014).
Kang, B. Electron microscopy and high-pressure freezing of Arabidopsis. Methods in Cell Biology 96, 259–283 (2010).
Henry, R., Carrigan, M., McCaffery, M., Ma, X. & Cline, K. Targeting Determinants and Proposed Evolutionary Basis for the Sec and the Delta pH Protein Transport Systems in Chloroplast Thylakoid Membranes. The Journal of Cell Biology 136, 823–832 (1997).
Bauer, B., Cai, X., Peth, S., Schladitz, K. & Steidl, G. Variational-based segmentation of bio-pores in tomographic images. Computers and Geosciences 98, 1–8 (2017).
Hammer, P. Marching Cubes - File Exchange - MATLAB Central. Available at: https://www.mathworks.com/matlabcentral/fileexchange/32506-marching-cubes. (Accessed: 20 August (2019).
stlwrite - write ASCII or Binary STL files - File Exchange - MATLAB Central. Avaiable at: https://www.mathworks.com/matlabcentral/fileexchange/20922-stlwrite-write-ascii-or-binary-stl-files (Accessed: 20 August (2019).
Acknowledgements
We thank Dr. Raymond Honfu Chan (City University of Hong Kong) for his help in automatic segmentation with MatLab. We also thank Emily J. Kang (Phillips Exeter Academy, USA) for generating electron tomography models that were not included in the paper. This work was supported by Cooperative Research Program for Agriculture Science & Technology Development (Project No. 0109532019) Rural Development Administration, Republic of Korea and Hong Kong Research Grant Council (GRF14126116, AoE/M-05/12, C4002-17G).
Author information
Authors and Affiliations
Contributions
K.M., I.H. and B.-H.K. designed the research; K.M. and S.-Y.H. performed the research; W.-T. Y. and X.C. prepared electron tomographic models of thylakoids, K.M., W.-T.Y., S.-Y.H. and B.-H.K. analyzed data and prepared the figures; and K.M. and B.-H.K. wrote the paper.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Mai, K.K.K., Yeung, WT., Han, SY. et al. Electron Tomography Analysis of Thylakoid Assembly and Fission in Chloroplasts of a Single-Cell C4 plant, Bienertia sinuspersici. Sci Rep 9, 19640 (2019). https://doi.org/10.1038/s41598-019-56083-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-019-56083-w
This article is cited by
-
Accelerated remodeling of the mesophyll-bundle sheath interface in the maize C4 cycle mutant leaves
Scientific Reports (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.