Correlated behavior implicates stromules in increasing the interactive surface between plastids and ER tubules

Stromules are extended by plastids but the underlying basis for their extension and retraction had not been understood until recently. Our live-imaging aided observations on coincident plastid stromule branching and ER tubule dynamics open out new areas of investigation relating to these rapid subcellular interactions. This addendum provides a testable hypothesis on the formation of stromules, which argues against the need for new membrane incorporation and suggests that stromal extensions might result from a remodeling of the plastid envelope membrane in an ER aided manner.Key words: stromules, plastids, endoplasmic reticulum, fluorescent proteins, subcellular interactions, FNR-EGFP, RFP-ERThe extension and retraction of stromules (stroma-filled tubules) from both chlorophyll containing and achlorophyllous plastids is well established for diverse plant species.¹ Many different conditions such as increased subcellular redox stress,² symbiotic interactions,^3–5 elevated temperatures,⁶ viral infection⁷ and alterations in plastid size and density^8,9 have been associated with stromule formation. Stromules extended from different plastids have been observed as forming connecting bridges^10–13 through which an exchange of proteins has been demonstrated.^{10,11,14–16} The latter observations strongly suggested stromules as transient communication channels between plastids for exchanging metabolites.^10,11 The connectivity of plastids via stromules also led to the suggestion that plant cells might contain a plastidom, defined as a well-connected plastid-compartment.¹² However, as pointed out by Netasan and co-workers¹⁷ most stromules do not appear to interconnect plastids and thus the movement of macromolecules between plastids might not be their sole function. In concurrence with this viewpoint the observations of Gunning¹⁸ and Lütz and Engel¹⁹ suggest that stromules have a role in increasing plastid interactions with mitochondria and peroxisomes. Thus a more generalized and acceptable statement is that stromules serve to increase the plastid stromal surface area that is exposed to the subcellular environment.Notably, the largest membrane surface area within a cell is provided by the endoplasmic reticulum (ER). Transmission electron micrographs often suggest a close proximity between plastids and the ER^20,21 and the presence of a chloroplast envelope associated-ER has been demonstrated.²² However, studies aimed at uncovering possible dynamic interactions between stromules and the ER in living plant cells had not been carried out. Our recent work²³ investigated this possible relationship by simultaneously visualizing stromules (labeled using a ferredoxin NADP(H) oxidoreductase (FNR) transit peptide fused to enhanced GFP; hereafter referred to as FNR-EGFP) and the ER (highlighted using a chimeric red fluorescent protein (RFP) carrying a basic chitinase signal sequence and an ER retention sequence; referred to as RFP-ER). Our observations clearly identified the ER as providing the nearest set of membranes with which the plastid envelope can interact. The observations are most pertinent for the narrow cortical sleeve where the relatively large sized plastids (including chloroplasts) are pressed against the cell boundary (plasma membrane and comparatively rigid cell wall) on one side and a turgid vacuole on the inner side. This space is also traversed by the cortical ER mesh created by constantly reorganizing ER-tubules. As shown in Figure 1A the side of a plastid pressed against the plasma membrane is surrounded by a loose mesh of ER tubules while the side pressed close to the vacuolar membrane (Fig. 1B) is free of the ER cradle. The cortically located plastids display strong behavioral correlations between their stromules and the neighboring ER tubules (Fig. 1C–E). Notably stromules form triangular junctions that appear very similar to the 3-way junctions displayed by the cortical ER. The three-way junctions of the ER are cisternal locations from which tubules are extended to create ER polygons. Stromule branches are extended from the stromal triangles in a very similar manner along comparable angles. Interestingly the extension and retraction of stromules and their branches occurs in tandem with contiguous ER tubules. These observations on the coincident behavior of the two organelles indicate that stromule branching and dynamic behavior might rely upon the creation of multiple membrane contact sites (MCSs) with the ER. The MCSs could aid in stretching stromules along ER tubules while loss of contact between the two membranes could cause rapid stromule retraction. While the possibility of identifying MCSs on the stromule envelope is being explored further by us our initial observations evoke a long-standing but very relevant question. Where does the membrane for forming stromules come from?Open in a separate windowFigure 1Confocal laser scanning microscopy based imaging of living Nicotiana benthamiana cells co-expressing FNR-EGFP (labels stromules) and RFP-ER (labels ER), 3D isosurface rendering and a model based on the observations. (A) 3D iso-surface rendering of a chloroplast facing the plasma membrane side shows the plastid situated in a loose cradle of ER tubules. (B) The 3D rendered chloroplast ER reconstruction seen in (A) observed from the vacuolar side after rotation by 180° shows the chloroplast appears smooth and free of the ER mesh. (C–E) Sequential frames from a time-lapse movie depict a FNR-EGFP labeled stromule undergoing a branching event. Note that in (C) the stromule extending along an ER channel with possible contacts at several points along the stromule length. (D) depicts a stromal triangle (arrowhead) that forms a branch initial. (E) depicts a branched stromule with the branches aligned with the ER (size bar = 5 µm). (F) A plastid exhibiting a short, wide stroma-filled area after stromule retraction has occured. (G) The same plastid shown in (F) without the FNR-EGFP labeled stroma (arrowhead) provides an appreciation of the “mobile jacket” created by the stromal contents around the chlorophyll containing thylakoids (size bar = 5 µm). (H and I) A diagrammatic depiction based on our observations and pertinent literature. (H) The narrow cortical region between the vacuolar membrane and the cell boundary is shared by large plastids and a dynamic ER. A depiction of the loose stroma filled jacket (arrowhead) of a plastid along with contiguous ER. Note the relative positions of the vacuolar membrane and the plasma membrane. The plastid size cannot increase isotropically within the narrow confines of the cortex and thus stromule formation along ER tubules is favored (Arrow pointing direction of stromule extension along the ER). (I) Multiple contact sites might be created between the extending stromule and the neighboring ER tubules. The diagrammatic depiction emphasizes the remodeling of the loose plastid jacket for stromule extension along ER. The schematic does not depict the strong possibility that both organelles might share an F-actin based mechanism for their extension.Stromule extension visibly enlarges the area occupied by the stromal contents of a plastid. The general conclusion of stromules increasing the plastid surface is clearly validated. Indeed Hanson and coworkers¹ calculated that for a model plastid of 3 µm diameter extending a 0.2 µm diameter and 10 µm long stromule the stromule represents about 20% of the overall plastid envelope surface area. A point that remains unclear is whether stromule extension involves a net increase in plastidic membrane or merely involves a remodeling of the existing membrane into an elongated shape. In the first scenario stromule formation would involve the creation of fresh membrane vesicles and their incorporation into an existing bi-layered plastid coat. This would be followed by their dispersal into inner and outer plastid envelopes along with complete complements of protein import and export machinery. Currently there is no experimental evidence to support this scenario. Moreover, this viewpoint does not consider observations of stromule retraction at all. If new membranes have been added during stromule extension then what happens to them during the retraction stage? Further, live imaging clearly shows that the time involved in sporadic stromule extension, branching and retraction is in the order of seconds to minutes. Such short periods do not favor the viewpoint involving creation of new membrane and its incorporation into stromules.Interestingly, observations of plastids in living cells often suggest that they have a rather wobbly form than a tight compact shape. The unstable form suggests the presence of relaxed membranes enveloping the main plastid body (Fig. 1F and G). These loose fitting membranes have been described as a “mobile jacket”^1,24 whose presence and irregular protrusions can be clearly distinguished from long stromules using a shape index.⁶ Hanson and co-workers¹ speculated that the “mobile jacket” might be stretched out and could thus provide the extra membrane needed for stromule formation. The notion is supported by the work of Gunning, which shows stromules retracting into short amorphous protrusions.¹⁸ Our observations of short protrusions or “beaks”²³ being created sporadically on a relatively regular-appearing plastid surface support these ideas. According to our observations one or two of the beaks might develop into long stromules that would be extended along neighboring ER tubules (Fig. 1H and I).Moreover, in vitro experiments on sucrose filled liposomes and giant vesicles^25,26 are particularly interesting in this context. Plasmolytic shrinkage of the sucrose filled vesicles creates states in which excess membrane surrounds a small volume. Over time the membrane exhibits small protrusions that are very similar in shape to the protrusions formed by plastid. Taken together the above-mentioned observations support behavior that might be expected upon a reshaping of already existing membranes. Notably the membrane-remodeling scenario for the formation of a stromule does not involve an increase in the net membrane of a plastid but it does achieve an increase in the surface area over which plastid stromal contents can interact with their surroundings. The viewpoint is also able to explain stromule elasticity that is suggested by their rapid extension and retraction.The new observations and ensuing discussion suggesting stromules being formed by stretching of the plastid envelope and identifying the ER as an interactive membrane partner in the process lay down the foundation for a number of other questions. These include investigations relating to the role of stromules in metabolite import and export between the two organelles, understanding the relationship between stromule and the ER behavior in relation to the underlying actin cytoskeleton as well as assessing the specificity of myosin motors that might be involved in these dynamic processes.