Over the last decade, several important developments have occurred in our understanding of plastids. In addition to major developments in the understanding of molecular processes which occur during plastid development, a subject considered in several other chapters in this book, a renewed consideration of plastid morphology and the dynamic nature of changes in plastid morphology has also taken place. Central to this latter consideration has been the exploitation of green fluorescent protein targeted to the plastid compartment, which has revealed dramatic new aspects of plastid morphology called stromules (Fig. 1). These long thin membranous tubules containing stroma but not thylakoid membrane or chlorophyll were rediscovered in the late 1990s (Köhler et al. 1997) by imaging GFP fluorescence in plastids of tobacco and petunia containing GFP. These stromules were between 350 and 850 nm in diameter and were highly dynamic in nature extending from and retracting into the plastid body and occasionally interacting with a stromule from a neighbouring plastid. In this case, the movement of GFP from one plastid to another by stromule transfer was shown using photobleaching (Köhler et al. 2000). Ironically, the modern day observation of stromules emanating from plastids was a reconfirmation of many observations made through the last century in which microscopists have observed various protrusions and dynamic extensions of plastids in many different types of tissue (Gray et al. 2001; Kwok and Hanson 2004). Wildmann's laboratory at the University of California was famous in the 1960s for images and movies of highly dynamic plastids producing long thin extensions in the cytoplasm which can fragment, leading to the improbable suggestion that these smaller structures become mitochondria (Wildmann et al. 1962; Wildmann 1967). What we now call stromules are clearly seen in his pictures. Perhaps not surprisingly, stromule-like structures were not considered seriously within the plastid community until their rediscovery 30 years later (Tobin 1997). So how do stromules form and what do they do?
Stromules form by dynamic out growth of the plastid envelope membranes and their movement within the cytosol is controlled in part by the actin microfilament cytoskeletal system in which myosin motors link stromules and plastid bodies to the actin microfilaments (Kwok and Hanson 2003, 2004a). Careful observation of stromules with DIC optics (Gunning 2004, 2005) has revealed a great deal about the precise dynamics of stromule interaction with the microfilament tracks and clearly shows how stromules are pulled out from plastid bodies by attachment to microfilament tracks at points of attachment, not only at stromule tips but also at points along the stromule length. Sudden loss of attachment causes rapid recoil of the stromule. In addition, stromules can also branch and rejoin forming closed loops as well as forming distinct bead-like structures along their length. Beads are particularly clear in stromules on chromoplasts in tomato fruit (Pyke and Howells 2002) although there is little evidence that such structures actually move along the stromule length. Whether the extension of stromules is entire due to pulling by the microfilament strands rather than a pushing out by a stromal pressure is unclear as is the exact source of the new membrane needed to produce a new stromule.
So what do stromules do to aid plastid function? At present the precise role of stromules is unclear but several considerations have been made. It is obvious that production of a stromule by a plastid will increase its surface area significantly and thereby increase the surface of interaction with the cytosol. Since plastids are highly active in cellular biochemistry and are sites of synthesis of many molecules important in cellular function, an increased surface area should potentially improve this interaction. This suggestion makes the assumption that the envelope membranes in the stromule have similar import capacity to that of the plastid body, a fact that has yet to be clearly addressed. The potential for movement of molecules between plastids has been demonstrated but how relevant this process might be to what actually occurs within the cell is difficult to determine. Certainly observation of plastids and stromules in the majority of cell types suggests that such joining is relatively rare and probably transitory in nature. A key point in trying to understand what stromules do is a clear distinction between their propensities in different types of cells and in particular their relative rarity in cells containing mature green chloroplasts. Thus, in mesophyll cells, which are packed with chloroplasts, stromules are rarely seen whereas in other cells containing non-green plastids such as in root cells, petal cells, epidermal cells and cultured suspension cells, stromules are much more abundant. Waters et al. (2004) showed that a decline in plastid density in the epidermal cells of expanding tobacco hypocotyls is correlated with a significant increase in stromule length raising the possibility that stromules act as a density sensing mechanism for plastids which are far apart. This could also tie in with mechanosensing proteins in the plastid envelope which sense when plastids are squashed together (Haswell and Meyerowitz 2006). In many cells containing non-green plastids, stromule networks are extensive and appear to link plastids, which are closely associated with the nucleus and surround it, to the peripheral cell membrane (Kwok and Hanson 2004b). Maybe stromules are involved with intracellular communication in some way. Fragmentation of stromules into distinct vesicles has also been suggested as a method of plastid replication since pieces of broken stromule in ripening tomato fruit cells appear to differentiate as chromoplasts. More work on stromules will be required to understand more fully these enigmatic interesting structures associated with plastids.
Was this article helpful?