|Home | About | Journals | Submit | Contact Us | Français|
Optimization of osmotic dehydration in different plant cells has been investigated through the variation of parameters such as the nature of the sugar used, the concentration of osmotic solutions and the processing time. In micro-organisms such as the yeast, Saccharomyces cerevisiae, the exposure of a cell to a slow increase in osmotic pressure preserves cell viability after rehydration, while sudden dehydration involves a lower rate of cell viability, which could be due to membrane vesiculation. The aim of this work is to study cytoplasmic vesicle formation in onion epidermal cells (Allium cepa) as a function of the kinetics of osmotic pressure variation in the external medium.
Onion epidermal cells were submitted either to an osmotic shock or to a progressive osmotic shift from an osmotic pressure of 2 to 24 MPa to induce plasmolysis. After 30 min in the treatment solution, deplasmolysis was carried out. Cells were observed by microscopy during the whole cycle of dehydration–rehydration.
The application of an osmotic shock to onion cells, from an initial osmotic pressure of 2 MPa to a final one of 24 MPa for <1 s, led to the formation of numerous exocytotic and osmocytic vesicles visualized through light and confocal microscopy. In contrast, after application of a progressive osmotic shift, from an initial osmotic pressure of 2 MPa to a final one of 24 MPa for 30 min, no vesicles were observed. Additionally, the absence of Hechtian strand connections led to the bursting of vesicles in the case of the osmotic shock.
It is concluded that the kinetics of osmotic dehydration strongly influence vesicle formation in onion cells, and that Hechtian strand connections between protoplasts and exocytotic vesicles are a prerequisite for successful deplasmolysis. These results suggest that a decrease in the area-to-volume ratio of a cell could cause cell death following an osmotic shock.
When a cell is subjected to a hyperosmotic shock, two response mechanisms are observed. First, the cell responds passively, like an osmometer (Alemohammad and Knowles, 1974). As long as the plasma membrane remains intact, water efflux from the cell and a weaker influx of membrane-permeable external solutes into the cell occur simultaneously until a new osmotic equilibrium is reached. The osmotic flow that occurs across the cell membrane is proportional to the osmotic pressure gradient (ΔP) between the cell and the external medium (Kedem and Katchalsky, 1958). This passive response is characterized by a short time-constant (a few seconds). As a consequence of water efflux, a rapid shrinking process occurs, which leads to a reduction in cell volume (Koch, 1984; Meury, 1994; Pohle et al., 1999). The cell volume decreases as soon as the osmotic pressure increases and is clearly associated with the intensity of the osmotic stress (Houssin et al., 1991). The final volume of the cell depends on the intensity of the stress (Munns et al., 1983; Morris et al., 1986). This response can subsequently lead to cell death (Wood, 1999).
Simultaneously with this osmotic stage, and only when the decrease in water potential is not too high, an active, energy-dependent, biological response occurs through the active osmoregulatory system of the cell (Yancey et al., 1982). Osmoregulation occurs more slowly (over minutes or hours) than the passive response (Niedermeyer et al., 1977; Zimmermann, 1978). This metabolic process consists of the biosynthesis of solutes and a rapid modification of membrane permeability that allows cells to import external water. This active response enables the cell to restore its internal volume when the membrane has not been irreversibly damaged by the passive phase of the osmotic response.
Many studies have specifically investigated the effects of an increase in osmotic pressure of the external medium on plant cells (Parida and Das, 2005). Such a stress applied to plant cells first leads to a decrease in cell volume and then to membrane detachment from the cell wall (plasmolysis) (Lang-Pauluzzi and Gunning, 2000). In onion (Allium cepa) epidermal cells, protoplast detachment from the cell wall is incomplete; the living protoplast forms thin, cytoplasmic strands that link the plasma membrane to the cell wall. These strands were first described by Hecht in 1912 and have been confirmed by Sitte (1963), Smith (1972), Schnepf et al. (1986), Oparka et al. (1990) and Lang-Pauluzzi (2000).
Volume reduction in plant cells can be controlled by active mechanisms. Thus, cell suspensions of mangrove (Bruguiera sexangula) and root meristematic cells of barley (Hordeum vulgare) demonstrate a rapid increase in their vacuolar volume under salt stress (150 mM NaCl). Such an increase could protect the cytoplasm by decreasing the cytoplasmic volume (Mimura et al., 2003). The volume increase of the vacuole is an active process, which involves the activation of tonoplast H+-ATPase and vacuolar acid phosphatase. During osmotic treatment, a redistribution of the cell membrane that causes a reduction of the surface membrane available for swelling has been observed (Ferrando and Spiess, 2001). The plasma membrane may then be damaged, which reduces cellular viability (Ferrando and Spiess, 2003).
A few authors in the plant field have investigated the correlation between the kinetics of dehydration by air drying and plant viability. Farrant et al. (1999) showed that whole plants of Myrothamnus flabellifolius (homoiochlorophyllous) and Xerophyta humilis (poikilochlorophyllous) survive only if drying is slow. In X. humilis, they observed that there was incomplete replacement of water in vacuoles and that plasma membrane disruption occurred during rehydration. Furthermore, the sudden removal of water has been observed to decrease the nutritional and sensorial value of fruits and vegetable (Lenart, 1996). Many investigations have looked at the optimization of osmotic dehydration in different plant cells. The nature of the sugar used (Ferrando and Spiess, 2001), the concentration of osmotic solutions and the processing time (Seguí et al., 2006; Rizzolo et al., 2007) have been studied; however, to our knowledge, no work has analysed the influence of the kinetics of osmotic treatment on plant cells.
In micro-organisms such as the yeast, Saccharomyces cerevisiae and the bacterium, Escherichia coli, a high cell viability rate is obtained when the osmotic pressure is slowly increased (Marechal et al., 1999; Beney et al., 2001; Mille et al., 2002). This phenomenon has been attributed to preservation of the plasma membrane structure (Laroche et al., 2001). In contrast, the lower cell viability rate obtained after an osmotic shock has been attributed to a rapid and passive water efflux from the cell; such a sudden dehydration would apply too many mechanical constraints on the membrane. In this case, cell mortality could be related to a sudden decrease in cell volume and then a folding of the plasma membrane, which can lead to the formation of cytoplasmic vesicles. Membrane vesiculation has been reported in plant cells and protoplasts of cells, such as Vicia faba [Homann and Thiel, 1999 (guard cell protoplasts)], Secale cereale [Singh, 1979 (epicotyl tissues); Gordon-Kamm and Steponkus, 1984a (leaves); Dowgert et al., 1987 (leaves)], Zea mays [Hübner et al., 1985 (root cap cells)], A. cepa [Oparka et al., 1990 (epidermal cells); Lang-Pauluzzi and Gunning, 2000 (inner epidermal cells)], Chlorophyton comosum [Komis et al., 2002 (leaf cells)], Chenopodium album [Wartenberg et al., 1992 (protoplasts from cell suspension culture)] and Brocchinia reducta (Owen et al., 1991 (basal leaf segments)].
It is well known that vesicles form from exocytotic or endocytotic events in the protoplast plasma membrane (Robinson and Hillmer, 1990; Battey et al., 1999; Verma and Hong, 2005). Nevertheless Meckel et al. (2005) demonstrated that vesicles could also form after hyperosmotic treatment of plant cell protoplasts of different species and that their size increases with an increase in the applied osmotic potential difference. Furthermore, vesicles internalized upon hyperosmotic treatment did not discharge their content into the central vacuole (Oparka et al., 1990; Hawes et al., 1995) and they are thus referred to as osmocytic vesicles in order to distinguish this phenomenon from endocytosis. Indeed endocytotic vesicle internalization is always followed by co-ordinated fusion with other subcellular compartments (Robinson et al., 1991; Diekmann et al., 1993; Hawes et al., 1995).
In summary, previous studies have shown that the kinetics of osmotic pressure variation have a great influence on the viability rate of different types of cells and it is well known that cells generate vesicles during the application of osmotic shock. Thus, the aim of this work is to study vesicle formation in onion cells as a function of the kinetics of osmotic pressure variation in the external medium.
Fresh bulbs of onion (Allium cepa L.) obtained at a local market of Dijon (France), 6 months after harvesting, were used in the investigations. The inner epidermal peels derived from the bulb were prepared in accordance with the method described by Lichtscheidl and Url (1990). Epidermal strips (between 50 and 100 µm thick) were peeled away using fine forceps and cut into small squares (5 × 5 mm) using a fine scalpel. For complete re-expansion of the cells, the epidermal pieces were briefly immersed in distilled water (<1 s) before the osmotic treatment. The strips were then placed between a slide and a coverslip to prevent air dehydration. Onion epidermis comprises a single layer of thin-walled epidermal cells, which allows accurate microscopic observations of cells.
Cells were first submitted either to an osmotic shock or to a progressive osmotic shift to induce plasmolysis. All osmotic perturbations were performed at 25 °C by putting 200 µL of the osmotic solution in contact with the onion strips on one edge of the coverslip.
An osmotic shock corresponds to a drastic and immediate change in external osmotic pressure. Such a change is clearly dependent on the mixing time of the cell suspension with the osmotic solution (Martinez de Maranon et al., 1997). Times were selected after preliminary experiments for osmotic shocks and osmotic progressive shifts in order to destroy or protect, respectively, a large part of the cell population.
Osmotic shock was achieved with a 24 MPa shock solution of 300 g L−1 KCl and 80 g L−1 CaCl2, which act as water activity depressors, in distilled water. The shock consisted of a rapid addition of 200 µL of this solution on one edge of the coverslip. The osmotic solution moved from this edge of the coverslip to the opposite one, which could be described as a front of solute at 24 MPa. So for all the cells of the sample, the external osmotic pressure changed from an initial value of 2 MPa to a final value of 24 MPa in <1 s, which was confirmed by the microscopic examination of the beginning of the plasmolysis. Cells were kept in the treatment solution for 30 min before measurements were taken.
Progressive osmotic shift was achieved with ten KCl/CaCl2 osmotic solutions of increasing osmotic pressures (2, 3, 7, 9, 11,13, 15, 18, 21 and 24 MPa) prepared with the following amounts of KCl/CaCl2 added to 1000 g of distilled water, respectively: 30/5, 55/12, 105/26, 130/33, 150/41, 179/48, 200/56, 240/65, 272/73 and 300/80 g. The weight of solute required to reach the different osmotic pressure levels quoted above was determined using the Norrish equation (Norrish, 1966), and confirmed with an Aqualab CX2 osmometer (Decagon Devices, Pullman, WA, USA). A progressive osmotic shift was achieved in 30 min by ten successive additions of 200 µL of each KCl/CaCl2 solution to the epidermal cells in increasing order of osmotic pressure. Each osmotic addition on the edge of the coverslip was followed by a 3 min time period to permit the solute to diffuse towards the cells and to achieve a corresponding balance in the cell volume. Just before each successive addition of KCl/CaCl2 solution, any excess of the preceding osmotic solution was rapidly withdrawn by capillarity.
After 30 min in the treatment solution for the osmotic shock as well as for the progressive osmotic shift, deplasmolysis was carried out. An osmotic solution of 30 g/5 g KCl/CaCl2 in 1000 g of distilled water (2 MPa) at 25 °C was added to one edge of the coverslip in 200 µL additions and allowed to diffuse. The excess liquid was removed from the opposite side of the slide by capillarity. Rehydration was then completed by adding distilled water instead of the 2 MPa solution to the edge of the coverslip.
The highly fluorescent dye Lucifer Yellow CH (LYCH) is commonly used to visualize endocytotic vesicles (Wright and Oparka, 1989; Oparka et al., 1990; Owen et al., 1991; Wartenberg et al., 1992). Addition of this fluorescent probe to cells allows exocytotic and endocytotic vesicles to be distinguished by microscopic observations. Indeed, because LYCH does not enter the cytoplasm, endocytotic vesicles appear as green spots in a black-coloured cytoplasm, whereas exocytotic vesicles appear as black spots in a green-coloured medium.
A 10-mg aliquot of LYCH (Sigma, St Louis, MO, USA) was added to 1 mL of distilled water and stored at +4 °C. About 2 s before osmotic treatments, 200 µL of osmotic solution was mixed with 10 µL of LYCH solution and immediately transferred onto each piece of epidermis. For each treatment, and before visualization, LYCH was washed out using the corresponding osmotic solution.
Visualization of endocytotic, or osmocytic, vesicles is more complex than visualization of exocytotic vesicles. Indeed, even after fluorescent staining, osmocytic vesicles cannot be distinctly observed using light microscopy. Confocal microscopy, which allows reconstitution of an image from numerous optical sections, combined with the use of the fluorescent probe, LYCH, enables osmocytic and exocytotic vesicles to be clearly visualized.
A confocal laser scanning system, the Leica TCS 4D microscope (Leica, Wetzlar, Germany), equipped with an argon and krypton laser and epifluorescence attachments (excitation 488 nm, emission PB-FITC) was used. Oil immersion objectives, ×40 (numerical aperture 1·00) and ×63 (numerical aperture 1·40) were used for most observations. Epidermal cells were observed under bright-field, fluorescence and differential interference contrast optics. To visualize the inner part of cells, optical sectioning at 1·3 µm was performed. The optical section number of each projection was between 5 and 20.
The vesicle formation was recorded with a video camera (×100; CCD 6710 model; Cohu, San Diego, CA, USA) connected to the camera port of the microscope (×1·5; Leica DMLB; Leica, Wetzlar, Germany). Selected images were then transferred to a computer.
To ensure that the plasma membrane was intact before the osmotic perturbation, cells were put in a KCl/CaCl2 solution (2 MPa) containing LYCH (Fig. 1A, B). As shown in Fig. 1B, under normal conditions LYCH was unable to cross the plasma membrane and enter the cytoplasm, which attested to the integrity of the plasma membrane.
The application of KCl/CaCl2 solution to generate an osmotic pressure shock of 24 MPa with LYCH led to plasmolysis (≤30 s) of the epidermal cells and the formation of numerous osmocytic vesicles (Fig. 1C, D). These results provided evidence that the extracellular medium entered the protoplasts. About 90 % of the cells contained fluorescent vesicles, and counts, which were repeated three times, revealed a mean number of 15 (s.d. = 4·58) osmocytic vesicles per cell. The fluorescent vesicles were not uniform in size and their diameters ranged from 1 to 10 µm. During this step, the protoplasts were not completely detached from the cell wall and Hechtian strands maintained a connection between protoplasts and cell walls. Such a phenomenon, not clearly visible in Fig. 1C, can be observed in Fig. 3A for an osmotic pressure of 24 MPa before the rehydration steps.
The first step of the rehydration process was achieved with a solution at an osmotic pressure of 2 MPa for 2–3 min. During deplasmolysis, protoplasts gradually re-expanded and osmocytic vesicles observed during plasmolysis still appeared clearly, especially under fluorescence optics (Fig. 2A, B, thick arrows). Exocytotic vesicles, which probably formed during plasmolysis but were not visible at this stage, gradually appeared and became progressively easier to observe because of the increase in cell volume (Fig. 2A, B, thin arrows).
Counting exocytotic vesicles made on three deplasmolysis cycles (about ten cells per observation field) revealed a mean number of 25 (s.d. = 4·36) exocytotic vesicles per cell.
Part of these exocytotic vesicles, also called sub-protoplasts (Komis et al., 2002), were clearly interconnected and linked to the protoplasts by a network of Hechtian strands (Fig. 2A, medium-sized arrows).
During microscopic observations of this step, Hechtian strands were moved by the flow of water from the external medium into the protoplast and were then clearly observed because of this motion. However, they did not appear distinctly in all the images: in Fig. 3B, for example, they were not visible. At the end of this first step of rehydration, the protoplasts and the connected vesicles fused (Fig. 3C).
Other exocytotic vesicles observed during deplasmolysis formed autonomously. The absence of any common movement between the cell and these vesicles confirmed that they were formed autonomously by the mother cell. These vesicles could be generated by the rupture of Hechtian strands or from the original plasma membrane. During the rehydration phase, these vesicles grew in volume after the addition of 2 MPa solution (Fig. 4B, thin arrows).
The second step of rehydration was achieved through the addition of distilled water. During this step, which could take 3–5 min, the protoplasts continued to increase in size. Fusion between protoplasts and exocytotic vesicles linked by Hechtian strands was completed during this stage. In this case, protoplasts could be restored to their initial volume (Fig. 3D). In contrast, autonomous exocytotic vesicles burst after the addition of water (Fig. 4C). This phenomenon prevented the protoplasts from recovering their size and shape due to lack of membrane area, and led to cell death (Fig. 4D).
These results show that Hechtian strand connections between protoplasts and vesicles are a prerequisite for successful deplasmolysis.
The entire rehydration process took approx. 8 min.
After the application of a progressive osmotic shift between 2 and 24 MPa for 30 min, no exocytotic or osmocytic vesicles were visualized by microscopic analysis, as seen in Fig. 5A (bright-field optic) and B (fluorescence optic).
Recently, vesiculation events after application of stress to cells have been the subject of many studies. For example, in E. coli the presence of vesicles after osmotic shock has been demonstrated (Alemohammad and Knowles, 1974; Schwarz and Koch, 1995; Mille et al., 2002). Additionally, observations of vesiculation in plant cells have been made. Oparka et al. (1990) and Wartenberg et al. (1992), who used the fluorescent marker LYCH, induced osmocytic vesiculation in onion and Chenopodium album cells. These authors found that vesiculation was dependent on the presence of CaCl2 in the external medium and that deplasmolysis resulted in disappearance of the LYCH fluorescence.
To date, no study has attempted to relate vesicle formation to the rate of osmotic stress. In this investigation, we have clearly demonstrated that the formation of cytoplasmic vesicles is dependent on the kinetics of osmotic dehydration. Indeed a sudden change in the osmotic pressure level induced the formation of many plasmatic vesicles, whereas no vesicle appeared when cells were submitted to progressively increasing osmotic pressure (i.e. progressive osmotic shift). It is thus concluded that the rate of dehydration is an important parameter that controls the plasmolysis cycle, as previously proposed by Mille et al. (2002) for bacteria. From our results, vesiculation appears to be the answer to rapid osmotic stress in cells and such a phenomenon has already been observed in cells submitted to rapid freezing stress (Steponkus, 1999). Thus, membrane vesiculation is a result of a drastic modification in cell volume. Indeed, a significant and rapid increase in osmotic pressure led to a high level of cell shrinkage and cell surface folding. This folding, as proposed by Gordon-Kamm and Steponkus (1984b) in a study on cold stress, could generate vesicles. The effects of the kinetics of osmotic treatment on cell viability have been previously shown by Gervais et al. (1992) and Poirier et al. (1998) in the yeast S. cerevisiae and the bacterium E. coli by using glycerol as a water activity depressor. In our case, KCl/CaCl2 solution was used to increase the osmotic pressure level; the influence of the selected osmotic substance has to be investigated further.
Using our technique, some of the vesicles observed were very large (up to 10 µm in diameter) compared with those observed in previous studies. Homann and Thiel (1999) reported the formation of exocytotic vesicles of around 300 nm using guard cell protoplasts. Most of the osmocytic vesicles identified in onion cells by Oparka et al. (1990) were between 0·5 and 3 µm in diameter.
In our study some of the vesicles burst, showing that the integrity of cells involved in an osmotic shock is primarily related to the type of vesicles that form during plasmolysis events. Nevertheless and according to our observations, deplasmolysis could preserve the viability of a cell only when the vesicles formed were connected to each other and to the protoplasts through Hechtian strands.
We propose that the formation of autonomous vesicles involves a decrease in the area-to-volume ratio of the cell due to the folding of plasma membrane during plasmolysis. During deplasmolysis, the cell has to restore its initial volume, which is linked to the initial mass of osmolytes in the cell. However, the cell burst because its area-to-volume ratio had decreased and the reduced membrane area could no longer contain this volume. Therefore, a fast dehydration involves the formation of vesicles, which modify the original area-to-volume ratio of a cell in such a way that during the following rehydration, cells cannot recover their initial internal volume and rupture.
Studying cell reactions to other physical perturbations, such as heat or application of hydrostatic pressure stress, would be interesting in order to verify whether cell membrane vesiculation may be a general answer for any type of rapid-onset stress on cells.
We acknowledge the ‘Région de Bourgogne’ for its financial support.