Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
•The involvement of callose in the mechanism of stomatal pore opening and closing in the fern Asplenium nidus was investigated by examination of the pattern of callose deposition in open and closed stomata, and by examination of the effects of callose degradation and inhibition or induction of callose synthesis in stomatal movement.
•Callose was identified with aniline blue staining and a callose antibody and degraded via β-1,3-d-glucanase. Callose synthesis was inhibited with 2-deoxy-d-glucose and induced by coumarin or dichlobenil. Stomatal pore opening and closing were assessed by estimation of the stomatal pore width.
•The open stomata entirely lacked callose, while the closed ones displayed distinct radial fibrillar callose arrays in the external periclinal walls. The latter displayed local bending at the region of callose deposition, a deformation that was absent in the open stomata. Both callose degradation and inhibition of callose synthesis reduced the stomatal ability to open in white light and close in darkness. By contrast, callose synthesis induction considerably improved stomatal pore opening and reduced stomatal closure in same conditions.
•The present data revealed that: during stomatal closure the external periclinal guard cell walls experience a strong mechanical stress, probably triggering callose synthesis; and that callose participates in stomatal movement.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Stomata are developmentally and functionally complicated epidermal cell complexes. The constituent guard cells (GCs) have the ability to undergo reversible changes in shape, leading to opening and closing of the stomatal pore. They are the outcome of a mechanical interplay between the cell walls and the protoplast of the GCs triggered by the vacuoles. Vacuoles undergo reversible volume changes, in response to environmental factors, through fairly complicated biochemical pathways (Sack, 1987; Wilmer & Fricker, 1996; Hetherington, 2001; Galatis & Apostolakos, 2004).
The GC walls display unique mechanical properties as a result of their particular form, fine structure and composition. Stomatal movement is largely based on the radial arrays of cellulose microfibrils in the periclinal GC walls around the stomatal pore (Galatis, 1980; Galatis & Mitrakos, 1980). During stomatal movement, the elastic properties of the GC walls seem to change, a phenomenon probably related to compositional changes of the matrix cell wall polysaccharides (Majewska-Sawka et al., 2002; Jones et al., 2003, 2005).
Considering the available information, the behavior and the probable role of callose in functioning stomata of the fern A. nidus was investigated by examination of the pattern of callose deposition in closed and open stomata and by examination of the effects of enzymatic degradation of callose, as well as the effects of the experimental inhibition or induction of callose synthesis in the opening and closing of the stomatal pore. Stomatal movement (i.e. stomatal opening and closure) was assessed by estimation of the stomatal pore width in stomata opening, after illumination in white light and closing in darkness.
Materials and Methods
Plant materials used were leaves of the fern A. nidus L. Plants grown in a glasshouse were further developed in plant growth chambers (16 h period of light (5.3 W m−2), at 25 ± 1°C; control conditions). Particular attention was given to: leaf regions in which the cuticle had not been completely developed (developing leaf regions), which displayed stomata at an advanced stage of differentiation, able to open and close; and to mature leaf regions covered by a thick cuticle layer, which exhibited fully differentiated stomata. The study was carried out on untreated and treated leaves.
Opening of stomata was achieved by exposure of whole plants or removed apical leaf regions to white light of 40 μmol m−2 s−1 at 30 ± 2°C and closing by transferring them to darkness, at 25 ± 1°C. The width of the stomatal pore was estimated in a light microscope equipped with a micrometric scale. The data were statistically analysed by the Student’s t-test. P-values < 0.05 were deemed significant.
In hand-made sections of living material, callose was stained with 0.05% (w : v) aniline blue (C.I. 42725; Sigma, St Louis, MA, USA) in 0.07 M K2HPO4 buffer, pH 8.5 (O’Brien & McCully, 1981; Apostolakos & Galatis, 1999). In fixed freehand and semithin leaf sections callose was identified with a monoclonal anti-callose antibody (Meikle et al., 1991; Ferguson et al., 1998). Following immunogold labeling, thin stomatal sections were examined for greater detail of the position of callose polymers in GC walls.
Inhibition or induction of callose synthesis To investigate the probable role of callose in stomatal opening and closure, the effects of 2-deoxy-d-glucose (2-DDG), coumarin and 2,6-dichlorobenzonitrile (dichlobenil) in stomatal movement were examined. 2-DDG inhibits callose synthesis (Jaffe & Leopold, 1984; Schreiner et al., 1994), while coumarin and dichlobenil inhibit cellulose synthesis (Montezinos & Delmer, 1980; Anderson et al., 2002), inducing simultaneously callose synthesis (Vaughn et al., 1996; Sabba et al., 1999; De Bolt et al., 2007a). In these experiments, only stomata at an advanced stage of differentiation were examined. These stomata exhibited radial cellulose microfibril arrays in the periclinal GC walls, which were deposited before the onset of the treatment.
The study was carried out in developing leaf regions, unless stated otherwise, treated with one of the following water solutions: (1) 500 μM or 1 mM 2-DDG for 48–72 h; (2) 500 μM coumarin, for 48–72 h; (3) 100 μM dichlobenil for 48–72 h; and (4) 100 μM cytochalasin B (CB) for 48–72 h. Each experiment was repeated three to five times. Water solutions of coumarin and dichlobenil were prepared from stock solutions in dimethyl sulfoxide (DMSO). The treatments were carried out in plant growth chambers (initially 16 h period of light, 25 ± 1°C) and afterwards in conditions inducing stomatal opening or closing. In all cases, apical leaf regions were removed and placed vertically with their cut edge immersed in the drug solutions. In each experiment, apical leaf regions placed in distilled water were used as controls. In the drug solutions used, the maximum DMSO concentration was 0.1% (v : v). At this concentration, DMSO did not induce detectable side-effects (Panteris et al., 2006).
Enzymatic callose degradation Callose degradation was carried out in paradermal sections of living leaves, by the enzyme β-1,3-d-glucanase (0.65 U mg−1, C.I. 49103; Sigma). A slightly modified protocol of Jones et al. (2003) was used to investigate the implication of callose in stomatal movement. Using this protocol the above authors examined the effects of arabinanase and other enzymes in stomatal opening and closure of Commelina communis. Thus, to study the effects of β-1,3-d-glucanase in the stomatal opening of A. nidus, paradermal sections of developing leaf regions were initially placed in 10 mM KCl containing 0.1 M CaCl2 and 5 mg ml−1β-1,3-d-glucanase for 2 h, at 25 ± 1°C in darkness. The sections were then transferred to 75 mM KCl containing 5 mg ml−1β-1,3-d-glucanase and exposed to white light at 30 ± 2°C for 2–5 h. Controls consisted of sections from the same leaves processed as above without glucanase.
To study the effects of β-1,3-d-glucanase in stomatal closure, paradermal sections of developing leaf regions were initially placed in 75 mM KCl containing 5 mg ml−1β-1,3-d-glucanase for 4 h in white light, at 30 ± 2°C. Afterwards, they were transferred to 10 mM KCl containing 0.1 mM CaCl2 and 5 mg ml−1β-1,3-d-glucanase in darkness at 25 ± 1°C for 2–4 h. Controls consisted of sections from the same leaves processed as described earlier but without glucanase. Each experiment was repeated three times. In all cases, the width of the stomatal pore was calculated.
Light and transmission electron microscopy (TEM) – immunogold staining
Small pieces of control and treated leaves were prepared for light microscope and TEM examination according to procedures recently described (Apostolakos et al., 2009b). Briefly, the material was fixed with glutaraldehyde and osmium tetroxide, dehydrated in acetone or alcohol series and embedded in Spurr’s resin Serva, Heidelberg, Germany. The semithin sections were stained with toluidine blue and the thin ones with uranyl acetate and lead citrate.
For immunogold staining, small leaf pieces were fixed in 2% (w : v) paraformaldehyde and 0.1% (v : v) glutaraldehyde in PEM (50 mM piperazine-1,4-bis(2-ethanesulfonic acid) (PIPES), 5 mM ethylene glycol-bis(beta-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 10 mM MgSO4), pH 6.8, at 4°C for 1.5 h. After fixation, the specimens were washed in the same buffer and dehydrated in a graded ethanol series (10–90%) diluted in distilled water, and in absolute ethanol three times, for 30 min (each step) at 0°C. The material was post-fixed with 0.25% (w : v) osmium tetroxide added to the 30% ethanol step for 2 h. Then, the material was infiltrated with LR White (LRW) (Sigma) acrylic resin diluted in ethanol, in 10% steps to 100% (1 h in each), at 4°C and finally with pure resin overnight. The samples were embedded in gelatin capsules filled with LRW resin and polymerized at 60°C, for 48 h.
Thin sections mounted on gold grids were treated with phosphate buffered saline (PBS) for 30 min, blocked with 5% (w : v) BSA in PBS for 3–5 h, then incubated with mouse anti-(1→3)-β-d-glucan monoclonal antibody (Biosupplies Australia Ltd, Parkville, Australia) for 2.5 h at 37°C and finally incubated with 10 nM monodisperse colloidal gold-conjugated anti-mouse IgG (Sigma) overnight. Anti-(1→3)-β-d-glucan antibody and gold-conjugated IgG were diluted 1 : 40 and 1 : 10 in blocking buffer respectively. The sections were counterstained with 2% (w : v) aqueous uranyl acetate for 20 min. Both TEM and immuno-TEM specimens were examined with a Philips 300 transmission electron microscope. Semithin sections were examined with a Zeiss Axioplan microscope equipped with a Zeiss Axiocam MRc5 digital camera.
Semithin sections of control material embedded in LRW resin were transferred to glass slides and blocked with 5% (w : v) BSA in PBS for 5 h. After washing with PBS, anti-(1→3)-β-d-glucan antibody diluted 1 : 40 in PBS containing 2% (w : v) BSA was applied overnight, at room temperature. Following rinsing with PBS and blocking again with 2% (w : v) BSA in PBS, the sections were incubated for 1 h, at 37°C in fluorescein isothiocyanate (FITC) anti-mouse IgG (Sigma) diluted 1 : 40 in PBS containing 2% (w : v) BSA. After rinsing with PBS, the sections were mounted on glass slides using an antifade mounting medium containing p-phenylenediamine.
Hand-made leaf sections of control and treated material were fixed in 8% (w : v) paraformaldehyde in PEM, pH 6.8, for 45 min, at room temperature, washed three times with PEM for 15 min and treated with 1% (w : v) cellulase (Onozuka, Yakult Honsha, Tokyo, Japan) in PEM for 30 min. After washing with PEM, the sections were extracted with 3% (v : v) Triton X-100 and 5% (v : v) DMSO in PBS for 1 h and then transferred to PBS containing 3% (w : v) BSA, for 1 h. Sections were incubated overnight with the same callose antibody and rinsed with PBS three times, for 15 min. They, were then transferred to PBS containing 2% (w : v) BSA and incubated in FITC-IgG as described earlier, washed with PBS and finally covered with antifade solution.
Semithin and hand-made sections were examined with a Zeiss Axioplan microscope equipped with an UV source, proper filters and a Zeiss Axiocam MRc5 digital camera. The aniline blue-stained sections were examined using a filter set provided with exciter G 365 and barrier LP 420, while the immunolabeled ones were examined with a filter set provided with exciter BP 450–490 and barrier BP 515–565. All samples were checked for UV autofluorescence in an epifluorescence microscope using the above filters.
The A. nidus stomata exhibit kidney-shaped GCs (Fig. 1a). The anticlinal wall separating the GCs, at the middle of which the stomatal pore forms, represents the ventral wall. The anticlinal walls of the GCs shared with the surrounding epidermal cells are defined as dorsal walls (Fig. 1a), while those parallel to the leaf surface are defined as periclinal walls (external and internal periclinal walls; Figs 2c, 3b).
Both aniline blue staining and immunolabeling with callose antibody applied in this study to localize callose gave reliable, almost identical results (Fig. 1d; cf. Fig. 1f; see also Apostolakos et al., 2009a,b). However, callose localization in intact stomata was possible only in stomata residing in developing leaf regions, where the cuticle was not fully developed. Although the GC wall thickening was still underway in them, these stomata were functional (i.e. they opened in the presence of white light and closed when they were transferred to darkness). Of 1616 stomata counted in illuminated leaves, 938 (58%) were open and 678 (42%) closed, while of 1718 stomata measured in leaves placed in darkness, 533 (31%) were open and 1185 (69%) closed. Closed stomata were considered as those displaying a completely closed stomatal pore.
In the mature leaf regions, where the cuticle was completely developed, callose localization in intact stomata was impossible with both the procedures used. The presence of a thick cuticle layer in GCs did not permit the entrance of either aniline blue or callose antibody. To overcome this problem, the surface of mature leaves was covered with a warm mixture of wax with resins that formed a relatively thick layer on the leaf. By removal of this material the cuticle was detached from the leaf surface. After cuticle removal, callose staining in mature stomata became efficient.
The functional stomata of A. nidus exhibited a definite pattern of autofluorescence emitted by the portion of the ventral wall between the stomatal pore and the dorsal walls (Fig. 1b,c). Autofluorescence was easily distinguishable from callose fluorescence (Fig. 1b,c; cf. Fig. 1d,f).
Callose deposition in functioning stomata
All the closed stomata that were in advanced stages of differentiation exhibited extensive callose accumulations along the midregion of the periclinal walls, at the margins of the stomatal pore wall thickenings (Fig. 1d). These callose deposits were consistently found in intact stomata as well as in their transverse semithin sections, following immunolabeling (Fig. 2b,d). The external periclinal walls contained much more callose than the internal ones (Fig. 2b,d). In paradermal view, the callose deposits appeared as fine fibrillar structures, which, similarly to the cellulose microfibrils, were radially arranged around the stomatal pore region (Fig. 1d; see also Apostolakos et al., 2009a). Callose was also present in the ventral wall thickenings delimiting the stomatal pore (Fig. 2b,d). The presence of the above callose deposits was also confirmed in closed mature stomata, after the mechanical removal of the cuticular layer (Fig. 1f). The internal periclinal walls of mature stomata almost completely lacked callose.
By contrast, callose was definitely absent from all the open stomata. This was observed in functioning, not-fully differentiated stomata (Figs 1e, 2g) and in mature stomata after cuticle detachment (Fig. 1g). The periodic stomatal callose synthesis and degradation was confirmed in > 15 different experiments and permits the suggestion that this glucan participates in stomatal movement of A. nidus. The arrows in Fig. 1(f,g) show, at the stomatal pore region, remnants of the wax–resin mixture applied to detach the cuticle from the leaf surface (Fig. 1h; cf. Fig. 1a), which emits autofluorescence (Fig. 1i).
Continuous microscope examination of living, opened stomata stained with aniline blue enabled the observation of callose appearance in the periclinal GC walls, parallel to the stomatal pore closure. Callose was initially detected in semiclosed stomata (Fig. 2h; cf. Fig. 2g) and as the stomatal pore closure progressed, the callose deposits increased considerably (Fig. 2i). The whole phenomenon was completed in c. 60 min. Callose deposition in the semiclosed stomata was also confirmed in transverse sections, after callose immunolabeling. The callose deposits in the external periclinal wall of the semiclosed stoma shown in Fig. 2(c,d) were not as intense as those of the closed ones in Fig. 2(a,b). Callose was also present at the junctions of the ventral and dorsal walls (Fig. 2e,f).
Examination in serial transverse semithin or thin sections of > 20 closed stomata, which were at an advanced stage of differentiation, revealed that the external periclinal GC walls were bent at definite regions (i.e. at the margins of the growing wall thickenings of the stomatal pore; Figs 2a, 3a,e inset). At these regions the external periclinal GC walls were relatively thin (Fig. 3a,e inset). Callose was mainly deposited in the highly stressed periclinal wall regions adjacent to this bending (Fig. 2a,b).
Immunogold labeling showed that callose resides close to the plasmalemma lining the GC wall thickenings but mainly in their inner regions (Fig. 3e). Callose exhibited the same pattern of deposition in ventral wall thickenings lining the stomatal pore (Fig. 3f). This distribution has been found in other plant cell types (Gregory et al., 2002; Salnikov et al., 2003) and shows that the callose synthesized on the plasmalemma is rapidly forwarded to inner GC wall layers. The gold grains were usually accumulated in wall regions displaying a moderate electron density (Fig. 3e,f), an observation also made in the secondary wall thickenings of cotton fibers (Salnikov et al., 2003). In contrast to the external periclinal GC walls of the closed stomata, those of the opened or semi-closed ones that were at an advanced stage of differentiation did not show any sign of bending (Figs 2c, 3b). This was confirmed in serial semithin sections of > 20 stomata examined.
The external periclinal GC wall deformation has been also found in the closed mature stomata examined. A median transverse hand-made section of such a living stoma is illustrated in Fig. 3c. The external periclinal walls are bent at regions close to the dorsal walls, where they were relatively thin compared with the rest of periclinal walls (Fig. 3c, arrowed). This phenomenon was not confirmed in semithin or thin sections of mature stomata, because during embedding of specimens in resins the stomatal pore opened. Resin polymerization probably increases the GC volume, resulting in the opening of the stomatal pore. All modifications of the embedding procedure failed to keep the pore of mature stomata closed. The external periclinal walls of the open mature stomata did not display any local deformation (Fig. 3d; cf. Fig. 3c). The above findings are in accord with the hypothesis that during stomatal closure the surrounding epidermal cells exert strong mechanical stresses on the external periclinal GC walls, leading to their deformation. These stresses are relieved in opened stomata, the external periclinal walls of which recover and become straight or slightly curved (Figs 2c, 3b,d).
The inhibition of callose synthesis affects stomatal opening and closure
Callose deposition in the differentiating and functioning 2-DDG-affected stomata was blocked (Fig. 4a; cf. Fig. 1d; see also Apostolakos et al., 2009a,b). The absence of callose seems to affect the stomatal pore opening. In 2-DDG-treated developing leaf regions exposed to white light for 3 h, 42.5% of the stomata were open (666 out of 1572 counted stomata), while the respective percentage of the control leaves exposed to white light for the same time period was 58% (938 out of 1616 counted stomata). In addition, the stomatal pore width in the 2-DDG-affected open stomata was 19% smaller than that of control open stomata (Fig. 5). These differences are statistically significant (P <0.001). In leaves treated with 1 mM 2-DDG for 24–48 h in darkness at 25 ± 1°C, 88% of the stomata were open or semi-open (2349 out of 2669 counted stomata) and only 12% were closed (320 out of 2669 counted stomata). By contrast, in control leaves placed in darkness, 69% of the stomata were closed. Therefore, inhibition of callose synthesis seems to decrease the ability of the A. nidus stomata to open in white light and close in darkness.
The radial cellulose microfibril arrays in the periclinal GC walls constitute basic elements of the mechanism of stomatal pore opening and closure (Galatis & Apostolakos, 2004). In untreated stomata, they create a characteristic pattern of birefringence under a polarized microscope (Fig. 6a,b; see also Apostolakos & Galatis, 1999). It is important that under the polarized microscope the 2-DDG-affected functional stomata exhibited a birefringence pattern that did not differ from that of the untreated stomata (Fig. 6c,d; cf. Fig. 6a,b). Therefore, during treatment the radial cellulose microfibrillar arrays in the periclinal GC walls were not detectably disturbed.
The induction of callose synthesis affects the stomatal opening and closure
The coumarin or dichlobenil-affected stomata displayed more extensive callose depositions than the control ones (Fig. 4b,d; cf. Fig. 1d,f; see also Apostolakos et al., 2009a,b). The affected closed or semi-closed stomata showed prominent radial fiber-like callose depositions in the periclinal walls (Fig. 4b,d). In contrast to the control stomata, they were also present in the open coumarin- or dichlobenil-affected stomata (Fig. 4c,e; cf. Figs 1e,g, 2g). Furthermore, examination of coumarin or dichlobenil-affected functional stomata under the polarized microscope revealed a birefringence pattern similar to that of the untreated stomata (Fig. 6e,f; cf. Fig. 6a,b). The periclinal GC walls of the affected stomata displayed the radial cellulose microfibrils deposited before treatment with the cellulose synthesis inhibitors. Radial cellulose microfibril arrays were not found in stomata affected at an early stage of differentiation with these inhibitors (Apostolakos et al., 2009a). These stomata were not yet able to open and close and were not considered in the present study.
In the coumarin- or dichlobenil-treated developing leaf regions exposed to white light for 3 h, the percentage of the open stomata reached to 93% (1279 out of 1369 counted stomata) and 96% (2774 out of 2883 counted stomata), respectively, while the respective percentages in control leaves exposed to the same conditions were 58%. In addition, the width of the stomatal pore in coumarin- or dichlobenil-affected stomata was, respectively, 28% and 54% greater than that of the control open stomata (Fig. 5; also Fig. 4g,h; cf. Fig. 1a). Therefore, the induction of callose synthesis is related to the excessive opening of the stomatal pore or coumarin and dichlobenil alone may induce this phenomenon.
In mature leaves treated in darkness at 25 ± 1°C with 500 μM coumarin for 24–48 h or 100 μM dichlobenil for 48–72 h, 2647 and 1644 stomata were measured, respectively. From the coumarin-affected stomata counted, 2134 (80.5%) were semi-closed and 513 (19.5%) closed. From the dichlobenil-affected stomata counted, 1378 (84%) were semiclosed and 266 (16%) closed. By contrast, in control leaves placed in darkness 69% of the stomata were closed. Therefore, either coumarin or dichlobenil alone or the excessive callose synthesis prevents stomatal pore closing in darkness.
Excessive stomatal pore opening affects the pattern of callose deposition
The above data show that the excessive opening of the stomatal pore is accompanied by the presence of abundant callose in the periclinal GC walls. To investigate further this phenomenon, the pattern of callose deposition in CB-affected stomata was investigated. In the elliptical stomata, actin filament disorganization by cytochalasins induces an exceptional increase of the stomatal pore width (Hwang et al., 2000). This was also confirmed in A. nidus stomata (Fig. 5; also Fig. 4i; cf. Fig. 1a). Treatment with CB disintegrated actin filaments in the A. nidus stomata but did not affect the pattern of callose deposition at any stage of stomatal development (Apostolakos et al., 2009a). However, the CB-affected opened stomata, in contrast to the control ones, displayed radial callose fiber arrays in the periclinal walls (Fig. 4f; cf. Figs 1e,g, 2g).
The enzymatic degradation of callose affects stomatal opening and closure
Preliminary experiments revealed that a 2–5 h β-1,3-d-glucanase treatment completely or almost completely degrades callose in all the stomata examined (Fig. 7a,c,e; cf. Fig. 7b,d). Examination of the affected stomata under the polarized microscope showed that their radial cellulose microfibril arrays remain unaffected (Fig. 7f; cf. Fig. 6a).
The data presented in Fig. 8 show that β-1,3-d-glucanase reduces the ability of stomata to open in white light. After 2 h in white light the width of the stomatal pore of the β-1,3-d-glucanase-affected stomata was 40% lower than that of the control stomata. During the further illumination of the leaf sections, the width of the stomatal pore increased but did not reach that of the control stomata. After 5 h of illumination, it appeared 30% smaller than that of the control stomata (Fig. 8). In Fig. 8, the width of the stomatal pore of the control stomata is greater than that of the control stomata shown in Fig. 5, because the measured stomata presented in these figures were under different conditions. Fig. 5 shows stomata taken from intact leaves, while Fig. 8 shows stomata taken from paradermal hand-made leaf sections floating on KCl solutions.
The data presented in Fig. 9 show that in untreated hand-made leaf sections, floating on KCl solutions, the stomatal pore width decreases c. 30% in darkness. Interestingly, in the β-1,3-d-glucanase-affected hand-made leaf sections the stomata did not close in the darkness. In those stomata, the width of the stomatal pore after 4 h in darkness appears increased by c. 30% (Fig. 9). Therefore, β-1,3-d-glucanase interferes with the ability of A. nidus stomata to close in darkness.
Pattern of callose deposition in open and closed stomata
This work showed for the first time a periodic callose synthesis/degradation during stomatal pore opening and closure in A. nidus. Callose is definitely absent from the open stomata, but appears rapidly in the external periclinal walls of the closing ones in the form of radial fibrillar arrays. Recent experimental work showed that, similarly to cellulose microfibrils, the alignment of the callose fibrils is determined by the underlying radial cortical microtubule arrays (Apostolakos et al., 2009a).
Our findings are not in accord with the view of Waterkeyn & Bienfait (1966, 1979) that callose degrades in the mature fern stomata. A closer examination of Figs 6, 7 and 22–30 presented in the paper of Waterkeyn & Bienfait (1979) revealed intense callose deposits in the closed stomata but less intense ones in the semiclosed stomata (i.e. in the fern species examined by the above authors, callose seems to degrade in the open stomata but is present in the closed stomata). Therefore, the periodic callose synthesis/degradation in the closed/open stomata may be a rather general feature among fern stomata.
The data of this work showed also for the first time that in A. nidus GC wall composition changes during stomatal movement. It should be noted here that in stomata of some angiosperm species examined, compositional changes of homogalacturonans, arabinans and phenolic esters of pectins have been described, which were correlated with the stomatal movement (Majewska-Sawka et al., 2002; Jones et al., 2003, 2005). Experimental degradation of the GC wall arabinans prevents both stomatal opening and closing (Jones et al., 2003, 2005).
Callose deposition is usually a response of the cells to mechanical or chemical stresses exerted on them (Kauss, 1987; Parre & Geitmann, 2005; Vaughn et al., 2007). In stomata, the plasmalemma–cell wall complex of the turgid, opened stomata should be under mechanical stress compared with those of the relaxed GCs of the closed stomata. Therefore, callose should be expected to form in the opened stomata and not in the closed ones. The consistent formation of callose in the closed stomata of A. nidus can be explained if we accept that an exceptional stress-inducing activity is applied on the external periclinal GC walls. Careful examination of the GC shape changes in moving A. nidus stomata showed that the external GC periclinal walls of the closed stomata bend locally (Figs 2a, 3a,e inset). Obviously, intense mechanical stresses are exerted on them. During stomatal closure, solutes are transferred from the GCs to surrounding cells leading to an increase in their volume, which probably generates mechanical stresses that are exerted on the closed stomata. These stresses, in correlation with the decrease in GC turgor, probably deform the external periclinal walls. Callose is consistently deposited in the deformed external periclinal wall regions (Figs 2b, 3e).
The plasmalemma lining these stressed wall regions probably senses a strong mechanical stress, which may result in the local increased Ca2+ uptake in the GCs, which triggers callose synthesis (Kauss, 1987). Mechanical forces have been shown to alter plasmalemma ion channel permeability, which is associated with Ca2+ and other ion fluxes (Haley et al., 1995; Silver & Siperko, 2003). This hypothesis may also explain the presence of callose at the sites of the junction of the ventral wall with the dorsal walls in the semiclosed and closed A. nidus stomata.
The bending of the external GC periclinal wall is also obvious in the mature stomata but lies closer to the dorsal walls (Fig. 3c). The bending ‘shifts’ towards the dorsal walls because in mature stomata the thinner regions of the external periclinal walls lie closer to the dorsal walls. This is the result of the mode of the local wall deposition in the periclinal GC walls (Apostolakos & Galatis, 1999). By contrast, the periclinal GC walls of the opened stomata do not show any kind of local deformation (Fig. 3b,d).
The different behavior of the internal periclinal walls of the closed stomata can also be explained. The inner portion of the stomata is more or less freely suspended in the substomatal cavity (Figs 2a,c, 3a,c). Therefore, mechanical forces cannot be directly applied on them. The inner periclinal walls of the closed mature stomata do not show any local bending or any callose deposits.
During stomatal opening the deformed GC wall regions probably recover not only structurally but also chemically. New polysaccharidic material, including cellulose microfibrils, may be incorporated in them. Synthesis of wall material other than callose during stomatal movement has been reported by Takeuchi & Kondo (1988). They correlated stomatal opening with the synthesis of pectic substances and cellulose in GCs of Vicia faba. If in A. nidus stomata, a cellulose synthesis/degradation cycle occurs during stomatal pore opening/closure, it should be restricted to a structurally undetectable portion of GCs cellulose, because both open and closed stomata display well-organized radial cellulose microfibrils (Fig. 6a,b). The work of Majewska-Sawka et al. (2002) and Jones et al. (2003, 2005) allows the suggestion that during stomatal movement the composition of GC wall matrix polysaccharides changes. In consideration of this information, it may be suggested that in A. nidus stomata callose, among other probable functions, creates a suitable microenvironment in the periclinal GC walls for the deposition of cell wall polysaccharides and in particular cellulose during stomatal opening, a hypothesis previously made for other cell types (Waterkeyn, 1981; Vaughn et al., 2007).
Callose and stomatal opening and closure
The present work shows for first time the implication of callose in the mechanism of stomatal movement in A. nidus. This conclusion is adequately substantiated by the following structural and experimental data:
1The periodic synthesis and degradation of callose in the closed and open stomata, respectively. This is a consistent phenomenon possibly limited to ferns. To date, no such periodic callose appearance/disappearance in closing/opening stomata has been described in angiosperms.
2The worsening of the ability of stomata to open in white light and close in darkness after the enzymatic degradation of callose by β-1,3-d-glucanase (Figs 8, 9). This experimental approach should not have side-effects, since β-1,3-D-glucanase affects callose only. Using similar experimental procedures, the involvement of matrix cell wall polysaccharides, other than callose, in the mechanism of angiosperm stomatal movement has been shown (Jones et al., 2003, 2005).
3The decrease in the ability of stomata to open in white light and close in darkness after inhibition of callose synthesis by the 2-DDG treatment. Although 2-DDG might interfere with the synthesis of other cell wall polysaccharides, the similarity of the effects of 2-DDG with those of β-1,3-d-glucanase and the presence of well-organized radial cellulose microfibril arrays in 2-DDG-affected stomata (Fig. 6c,d) support the hypothesis that 2-DDG affects the stomatal movement via inhibition of callose synthesis.
4The improvement of the ability of stomata to open in white light and the prevention of their complete closure in darkness after induction of callose synthesis by coumarin or dichlobenil. In the case of coumarin, the above phenomena might be due to the result of direct interference of this drug with stomatal movement, which apart from cellulose synthesis may affect other cellular processes, such as membrane ion transport (Purohit et al., 1992). In contrast to A. nidus, in some angiosperms coumarin induced stomatal closure (Purohit et al., 1992; Zacchini & Morini, 1998). However, in A. nidus, dichlobenil has the same effects as coumarin. Dichobenil has been considered to have a more specific function than coumarin affecting cellulose synthesis only (Delmer et al., 1987; De Bolt et al., 2007b). There is also the possibility that both coumarin and dichlobenil affect stomatal movement by disturbance of cellulose synthesis. In that case, it is expected that the stomatal pore opening should be worsened and not improved. In addition, the treated stomata possess well-organized radial cellulose microfibrils (Fig. 6e,f), which had been deposited before the onset of the treatment with both drugs used. Therefore, it is tempting to suggest that the above drugs interfere with stomatal opening and closure by induction of callose synthesis
As for the particular role of callose in the mechanism of stomatal movement, only hypotheses can be made. It has already been mentioned that callose accumulations in the closing stomata might create the proper microenvironment for the rapid synthesis of other cell wall polysaccharides including cellulose, which may accompany stomatal opening (Takeuchi & Kondo, 1988; see also Fukuda et al., 1998, 2000).
Callose is also considered as material that makes the cell wall rigid and stiff (Parre & Geitmann, 2005). The deposition of callose in the form of a continuous sheet or as large amorphous masses in the periclinal GC walls could be expected to block or greatly reduce their particular mode of expansion. However, in A. nidus periclinal GC walls, callose is deposited as radially arranged fibrils (Figs 1d, 4d; see also Apostolakos et al., 2009a). This mode of callose deposition does not prevent the ‘tangential’ periclinal GC wall expansion that is critical for the stomatal pore opening. By contrast, the radial callose fibril arrays should reinforce the function of cellulose microfibrils in this process.
Moreover, the depositing fibrillar callose may loosen or disrupt the development of chemical bonds between the matrix cell wall polysaccharides changing the elastic properties of the periclinal GC walls. This might facilitate their tangential expansion, promoting stomatal pore opening. In the case of modification of the elastic properties of the periclinal GC walls by callose in fern stomata, this glucan might play a role analogous to that of the arabinans in the angiosperm stomata (Jones et al., 2003, 2005).
Unlike the untreated opened stomata of A. nidus, those treated with CB or cellulose synthesis inhibitors (coumarin and dichlobenil) form abundant callose in their periclinal walls. In elliptical stomata, actin filament disintegration induces a detectable increase in the width of the stomatal pore (Hwang et al., 2000, and literature therein), a phenomenon also confirmed in A. nidus stomata (Fig. 5). In addition, the width of the stomatal pore in A. nidus stomata treated with coumarin or dichlobenil was obviously larger than that of control stomata (Fig. 5) and possessed abundant callose. It is difficult to conclude whether callose is responsible for the excessive opening of the stomatal pore or vice versa. The induction of callose synthesis, after coumarin or dichlobenil treatment, which results in the excessive opening of the stomatal pore, may explain the presence of this glucan in the open stomata but cannot explain the presence of callose in the open CB-affected stomata because CB is not implicated in callose synthesis. Thus, it seems probable that the excessive opening of the stomatal pore generates excessive mechanical stresses in the GC walls, which might induce the synthesis of callose in the opening stomata.
The present study was financed from the University of Athens (project ‘Kapodistrias’).