AOBPreview originally published online on July 25, 2005
Annals of Botany 2005 96(5):769-778; doi:10.1093/aob/mci234
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Perisperm-endosperm Envelope in Cucumis: Structure, Proton Diffusion and Cell Wall Hydrolysing Activity
Institute of Environment Management and Plant Sciences, Vikram University, Ujjain (M.P.) 456 010, India
* For correspondence. E-mail dilipamr{at}sancharnet.in
Received: 17 February 2005 Returned for revision: 15 April 2005 Accepted: 10 June 2005 Published electronically: 25 July 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
• Background and Aims The envelope surrounding the embryo in cucurbit seed, which consists of a single layer of live endosperm cells covered by lipid- and callose-rich layers, is reported to show semi-permeability and also to act as the primary barrier to radicle emergence. Structure, development and permeability of the envelope and activity of cell wall hydrolases during germination of cucumber and muskmelon seeds were investigated.
• Methods Sections of seeds were stained with aniline blue and Sudan III. Proton diffusion and endo-ß-mannanase activity were detected by tissue printing. A gel-diffusion assay was performed to quantify endo-ß-mannanase activity, while the activity of ß-glucanase was determined with laminarin as the substrate and glucose formation measured using the GOD-POD method.
• Key Results The lipid layer differentiated during seed development in cucumber in the epidermis of a multilayered nucellus, whereas the callose layer appeared to develop outside the endosperm cell layer. Accordingly, the envelope has been called the perisperm-endosperm (PE) envelope. Chloroform treatment of seeds, which resulted in a substantial reduction in Sudan staining of the lipid layer, also enhanced the permeability of the PE envelope to 2,3,5-triphenyltetrazolium chloride. Proton diffusion occurred when the PE envelopes from seeds had their inner surface in contact with bromocresol purple-containing agarose gels, but not when their outer surface was in contact. Substantial endo-ß-mannanase activity was present in the caps of the PE envelopes, whereas a marked increase in ß-glucanase activity was observed in radicles prior to germination.
• Conclusions The lipid layer seems to contribute to the semi-permeability of the PE envelope. The diffusion of protons might create an acidic environment conducive to the activity of cell wall hydrolases, namely endo-ß-mannanase (EC 3.2.1.78
[EC]
) and ß-glucanase [ß(1
3)glucanohydrolase; EC 3.2.1.6
[EC]
], which, in turn, may play a role in the weakening of the PE envelope necessary for the protrusion of the radicle in cucumber and muskmelon seeds.
Key words: ß-Glucanase, callose, Cucumis, endo-ß-mannanase, lipid layer, perisperm-endosperm envelope, proton diffusion, seed development, seed germination, tissue printing
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
In many plant species, the seed coverings impose a physical barrier, which must be overcome by the embryo if the seed is to complete its germination. These covering layers, which may include testa, perisperm and endosperm, are known to display unique permeability properties in many seeds on account of the presence of semi-permeable layers, which allow water uptake and gas exchange while restricting or preventing solute transport. The tetrazolium ion, which is commonly used in viability tests, could not penetrate the inner coat of watermelon (Thornton, 1968
), tomato or pepper seeds (Beresniewicz et al., 1995b). Similarly, lanthanum ions were found to accumulate at the inner seed coat adjacent to the endosperm in tomato and pepper (Beresniewicz et al., 1995b; Taylor et al., 1997). The chemical nature of the semi-permeable layer in tomato and pepper was attributed to suberin, while cutin was found in leek and onion (Beresniewicz et al., 1995a). Although massive callose deposition is not a general feature of semi-permeable layers in seed-covering structures, a thick aniline blue-staining (indicative of callose) layer inside the seed coat has been reported in a number of species, e.g. cucumber, muskmelon, watermelon and barley (Yim and Bradford, 1998).
In muskmelon and several other cucurbit seeds, a thin envelope completely encloses the embryo (Singh, 1953). This has been shown to consist of a single layer of endosperm cells, covered by a thick, noncellular layer of callose-rich material and a thin waxy or lipid- and suberin-containing outer layer in muskmelon seeds (Welbaum et al., 1998; Yim and Bradford, 1998). Welbaum and Bradford (1990) suggested that the lipid-containing layer is involved in the semi-permeability of the muskmelon endosperm envelope. In later work, however, Yim and Bradford (1998) demonstrated that the apoplastic semi-permeability of the muskmelon endosperm envelope was dependent on the callose-containing layer, whereas the lipid-containing outer layer, though slowing down water uptake, was not responsible for semi-permeability.
In addition to serving as a semi-permeable layer, the envelope in cucumber, muskmelon and other cucurbit seeds is known to act as a barrier to radicle emergence. Since 1963, when Ikuma and Thimann first suggested that endosperm weakening is the consequence of enzymatic action (Ikuma and Thimann, 1963), a number of studies have provided evidence for the collaborative/successive action of several proteins in cell wall modification. Recently, many cell wall-modifying proteins or their expressed mRNAs have been identified in the endosperm tissue from tomato, tobacco and a few other species (Wu et al., 2001; Chen et al., 2002; Leubner-Metzger, 2003; Petruzzelli et al., 2003). Germinative mannanase (LeMAN2), expansin (LeEXP4) and xyloglucan endotransglycosylase (LeXET4) genes showed co-ordinate expression in the endosperm cap in tomato, thus allowing co-operation between these cell wall-modifying factors (Chen and Bradford, 2000; Nonogaki et al., 2000; Chen et al., 2002). Wu et al. (2001) reported expression of chitinase (Chi9) and ß-1,3-glucanase (GluB) in the micropylar endosperm cap of tomato seeds just before radicle emergence, but found no evidence that they were involved in endosperm cap cell wall hydrolysis and weakening. On the other hand, sense-transformation with a chimeric ABA-inducible ß-1,3-glucanase I transgene provided evidence of a causal role for ß-1,3-glucanase I during endosperm rupture in tobacco seeds (Leubner-Metzger and Meins, 2000). Notably, a ß-1,3-glucan substrate has not been detected in tomato and tobacco seeds (Beresniewicz et al., 1995a; Leubner-Metzger, 2003). In muskmelon seed germination, the endosperm envelope acts as the primary barrier to radicle emergence (Welbaum and Bradford, 1990). In addition to having cell walls rich in mannan and galactomannan polymers (Welbaum et al., 1998), the endosperm envelope in muskmelon contains, as mentioned above, a thick callose (ß-1,3-glucan)-rich layer (Yim and Bradford, 1998). Interestingly, while endo-ß-mannanase activity (Welbaum and Wang, 1997; Welbaum, 1999) and chitinase mRNAs and activity have been reported in germinating muskmelon seeds (Zou et al., 2002; Witmer et al., 2003), evidence of ß-1,3-glucan-hydrolysing activity is lacking so far. Indeed, Welbaum and Wang (1997) reported ß-1,3;1,4-glucanase activity in muskmelon endosperm tissue prior to radicle emergence. However, ß-1,3;1,4-glucanases do not hydrolyse ß-1,3-glucans (Leubner-Metzger, 2003).
Thus, one major objective of the present work was to study the structure and development of the envelope surrounding the embryo in Cucumis sativus and Cucumis melo and to determine whether the lipid layer of the envelope contributed in someway to its semi-permeability. In addition, the possible activity of the cell wall hydrolases, namely endo-ß-mannanase (EC 3.2.1.78
[EC]
) and ß-glucanase [ß(13)glucanohydrolase, EC 3.2.1.6
[EC]
], was examined during seed germination in these species.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Plant material
Cucumber (Cucumis sativus L. ‘Green Long’) seeds were purchased from Mahyco Ltd, Jalna, India, whereas muskmelon (Cucumis melo L. ‘Top Mark’) seeds were kindly provided by Professor Kent Bradford, University of California, Davis, USA. After testing the seeds for germination in the laboratory (85–90 % germination at 25 °C in the darkness both in cucumber and muskmelon), they were stored in airtight plastic containers at –10 °C until they were used. Seeds were decoated (testa removed) by carefully removing the seed coat with forceps. Prior to decoating, whole seeds were kept at 100 % relative humidity/25 °C for about 6 h so as to prevent any possible injury to the perisperm-endosperm (PE) envelope and/or embryo during the act of decoating. Efficacy of the treatment was ascertained by autoclaving samples of decoated seeds for 90–95 % osmotic distention. In decoated cucumber seeds soaking in water at 25 °C, the initial phase of water uptake lasted up to 9 h, initiation of radicle emergence occurred at 16 h, and germination was complete by 22 h. The timings for the completion of the above events in decoated muskmelon seeds were 12 h, 18 h and 24 h, respectively. In the case of the cucumber, plants were also grown in the experimental field of the Institute following the cultural practices suggested by Mahyco Ltd. Female flowers were tagged at anthesis and fruits were harvested for seeds at 10-d intervals from 15 to 45 d post-anthesis (DPA). Efforts to obtain fruits of muskmelon ‘Top Mark’ of the desired quality and quantity at various stages of development by growing plants in the experimental field were not successful. Therefore, development of the PE envelope was studied only in cucumber seeds.
Chemicals
Bromocresol purple, 2,3,5-triphenyltetrazolium chloride (TTC) and other chemicals used in this study, unless otherwise stated, were from HiMedia Laboratories, Mumbai and Sisco Research Laboratories, Mumbai, India. TTC was quantified as its reduction product, i.e. formazan (Dawson et al., 1986). Locust bean gum and Congo Red (Fluka, Switzerland), commercially prepared endo-ß-mannanase from Aspergillus niger (Megazyme International, Wicklow, Ireland) and laminarin (Sigma, St Louis, MO, USA) were used.
Seed anatomy
Sections (20 µm thick) of dry, mature decoated cucumber or muskmelon seeds were cut with a plant microtome model MTH-1 (Nippon Medical & Chemical Instruments Co., Ltd, Osaka, Japan). Developing or germinated cucumber seeds were fixed in a formaldehyde/glacial acetic acid/50 % ethanol (FAA) solution (1 : 1 : 18) ready for cutting sections. The sections were stained with aniline blue and Sudan III as described by Yim and Bradford (1998) except that toluidine blue O and Sudan IV were not used.
Chloroform treatment and autoclaving
Chloroform treatment was administered by dipping decoated cucumber or muskmelon seeds in the solvent for 10 min at 25 °C and air-drying for 24 h. For autoclaving, decoated cucumber or muskmelon seeds were soaked in distilled water for 3 h at 25 °C and then autoclaved at 120 °C for 20 min.
Permeation of TTC into seeds and embryos
Developing cucumber seeds or embryos isolated from them were treated with a 2 mM TTC solution in 0·05 M phosphate buffer (pH 6·8) at 25 °C in the dark. After 12 h, the seeds were washed with distilled water and the embryos were isolated from the treated seeds. These embryos, as well as those isolated and then treated with TTC were washed with distilled water, wiped dry with tissue paper, and then dried over anhydrous silica gel. Three replicates of 25 mg of the tissue were extracted in acetone to quantify formazan. After centrifugation at 10 000 g for 10 min and desired dilution of the supernatant with acetone, the absorbance was recorded at 485 nm. In a further set of seeds, mature, decoated chloroform-treated or -untreated cucumber or muskmelon seeds were soaked in distilled water. After 3 h, the seeds or the embryos isolated from them were treated with a 2 mM TTC solution in buffer at 25 °C. After 12 h, the seeds were washed with distilled water and the embryos were isolated from them. These embryos as well as those isolated then treated with TTC were washed with distilled water, dried over anhydrous silica gel, and extracted in acetone in triplicate to quantify formazan as described above.
Proton diffusion
Tissue printing
A gel solution containing agarose (1·2 % w/v) and 0·1 mM bromocresol purple (water soluble) was prepared in sterilized distilled water, maintaining the pH at 6·5 with 0·01 N NaOH. The gel solution (0·25 mL) was added to a 1·5-cm-diameter well in a ten-welled plastic strip. Decoated muskmelon seeds were soaked in distilled water for 3 h at 25 °C in the dark. The PE envelopes were isolated by opening up with fine-tip forceps along the distal cotyledonary end and then taking them out with blunt-tip forceps. One of the two halves of each PE envelope was placed with its inner surface on the agarose gel in a well, whereas the other one was placed with its outer surface on the gel in a well opposite to the former. After incubation at 25 °C in the dark in water vapour-saturated, self-sealing polythene bags for 12 h, the halves of the PE envelopes were removed and the gels were immediately photographed.
Measurement of proton diffusion
PE envelopes (n = 10) were isolated from decoated cucumber or muskmelon seeds that had imbibed water for 3 h. The two halves of each envelope were floated with their inner surface down on 0·5 ml of carbon dioxide-free distilled water (pH 6·5) in 1-cm quartz cuvettes. The PE envelopes were also isolated from viable or autoclaved seeds. Caps (n = 10) excised from the radicle end of the PE envelopes, or from the distal cotyledonary end, were placed as bulk samples in triplicate in 0·5 mL distilled water, as above, in 1-cm quartz cuvettes. After replacing the Teflon lids and securing them tightly with parafilm, the cuvettes were shaken (100 rpm) at 25 °C in the dark. After incubation for 12 h, the tissue was removed and 1·5 mL of 0·025 mM bromocresol purple solution (pH 6·5) prepared in carbon dioxide-free distilled water was added to each cuvette. Absorbance at 436 nm was recorded against identically treated controls without the tissue. To quantify proton diffusion in the incubation medium, a calibration curve was prepared by adding measured quantities of 0·01 M HCl to a 0·025 mM aqueous solution of bromocresol purple (pH 6·5) and recording the corresponding increase in absorbance at 436 nm.
Activity of endo-ß-mannanase (EC 3.2.1.78)
Tissue printing
Gels (5 x 3 x 0·5 cm) containing 7·5 % (w/v) polyacrylamide and 0·1 % (w/v) locust bean gum (LBG) were prepared in 0·1 M citric acid/0·2 M disodium phosphate buffer (pH 6·2). PE envelopes were isolated from decoated muskmelon seeds soaked in distilled water for 3 h at 25 °C in the darkness. One of the two halves of each PE envelope was placed with its inner surface on the gel, whereas the other one was placed with its outer surface on the gel opposite to the former. After incubation at 25 °C in the dark for 12 h in water vapour-saturated, self-sealing polythene bags, the halves of the PE envelopes were removed and the gels were immediately stained as described below.
Diffusible/extractable activity
Hollow plastic cylinders (diameter 3 mm; height 2 mm) were gently pressed on a polyacrylamide gel (5 x 3 x 0·5 cm) prepared as described above. The hollow plastic cylinders were employed because, in preliminary experiments, satisfactory results were not obtained with the small sample volume that a 2- or 3 mm-diameter well punched in a 0·5-mm thick polyacrylamide gel would otherwise hold. PE envelopes or embryos were isolated from decoated cucumber or muskmelon seeds which had imbibed water for 3 or 15 h. Caps (n = 25) were excised from the radicle end of the PE envelopes and incubated at 25 °C for 12 h in triplicate in 1·25-cm-diameter wells in plastic strips, each containing 0·1 ml of 0·1 M citric acid/0·2 M disodium phosphate buffer (pH 6·2) after being covered with parafilm. Radicles (n = 25) were excised from the embryos and similarly incubated. After a 12-h incubation period, a 10-µL aliquot of the incubation medium from a well was transferred (diffusible endo-ß-mannanase activity) to a plastic cylinder placed on the gel. The remaining buffer was removed and 0·2 ml of fresh buffer was added. The tissues were ground with a mortar, centrifuged at 10 000 g for 10 min at 4 °C, and 10 µL supernatant was assayed for extractable endo-ß-mannanase activity. To determine the total endo-ß-mannanase activity in the embryonic axis, cotyledons and the PE envelopes, 50 mg of tissue from seeds soaked for various time intervals were extracted in 0·5 ml buffer in triplicate. After centrifugation, a 10-µL aliquot of the supernatant was assayed for total endo-ß-mannanase activity. The enzyme was allowed to act for 24 h at 25 °C in the darkness, after which the cylinders were removed and the gels, immediately stained with Congo Red, following Toorop et al. (1996) and Amritphale et al. (2005), and photographed. The diameters of the hydrolysed areas were measured in two directions to the nearest 0·1 mm with calipers and averaged. The enzyme activity in nkatals was calculated according to a standard curve for Aspergillus mannanase.
Activity of ß-glucanase [ß(13)glucanohydrolase, EC 3.2.1.6]
PE envelopes or embryos were isolated from decoated cucumber or muskmelon seeds which had imbibed water for 15 h. Caps (n = 75) were excised from the radicle end of the PE envelopes and incubated at 25 °C for 12 h in triplicate in 1·75-cm-diameter wells in plastic strips each containing 0·3 ml of 15 mM Na-acetate buffer (pH 5·5) after being covered with parafilm. Radicles or plumules (n = 75) were excised from the embryos and similarly incubated. After a 12-h incubation period, the diffusate was pipetted into a Millipore Ultrafree-MC centrifugal filter unit PL-5 and centrifuged at 5000 g for 45 min at 4 °C. The retentate was diluted to 0·1 mL with buffer and used for diffusible ß-glucanase assay. To determine remaining (extractable) ß-glucanase activity, the tissue from each well was extracted in 0·6 ml of buffer with a mortar and was centrifuged at 10 000 g for 10 min at 4 °C. The supernatant was pipetted into Ultrafree-MC filter units PL-5 and the retentate, after dilution to 0·1 mL with buffer, was assayed for extractable ß-glucanase activity. To determine total ß-glucanase (diffusible plus extractable) activity in embryonic axis, cotyledons and PE envelopes, 50 mg tissue from seeds soaked for various time intervals was extracted in 0·5 mL of 15 mM Na-acetate buffer (pH 5·5) in triplicate. The rententate, obtained by centrifugation and subsequent ultrafiltration as above, was diluted to 0·1 mL with buffer and assayed for ß-glucanase activity using Laminaria digitata laminarin as the substrate, following Salyers et al. (1977) and Morohashi and Matsushima (2000). The enzyme activity was evaluated by measuring glucose formation with the GOD-POD method using the kit and the procedure given by E. Merck (India) Ltd, Mumbai, India.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Anatomy and development of the PE envelope
The PE envelope in cucumber and muskmelon seeds contained a single layer of live endosperm cells covered by a thick aniline blue-staining layer, which in turn was covered by a relatively thin Sudan-staining layer. Yim and Bradford (1998)
demonstrated that in muskmelon seeds the former layer was primarily composed of callose, whereas the latter one was rich in lipids. Hence, in the present work, they have been referred to as the callose layer and the lipid layer, respectively (Fig. 1A and B). The lipid layer showed considerably reduced staining coincident with seed germination in both cucumber and muskmelon (Fig. 1C and D). However, the callose layer apparently remained more or less unaffected. Treating decoated cucumber (Fig. 1E and F) or muskmelon (data not shown) seeds with chloroform at 25 °C for 10 min also resulted in considerably reduced staining of the lipid layer, whereas there was not any apparent effect on the callose layer.
|
During early development (15 DPA) of cucumber seeds, a lipid layer was seen differentiating in the epidermis of the four- to five-layered nucellus. An incipient callose layer, which did not stain with aniline blue, was also seen developing above the layer of endosperm cells (Fig. 2A). By 25 DPA, the incipient callose layer appeared a little more developed, but it still did not stain with aniline blue (Fig. 2B). The lipid layer also appeared to develop further within the inner epidermal cell walls of the nucellus (Fig. 2C). At 35 DPA, not only the lipid layer was quite distinct, but a thick deposit of callose, as indicated by aniline blue staining, was also observed outside the walls of endosperm cells (Fig. 2D). In addition, a little deposition of callose was observed below the lipid layer. Notably, callose was not detected on the inner side of the primary wall of endosperm cells (Fig. 2E). By 45 DPA, the nucellar cells, which had already been reduced to one or two layers at 35 DPA (Fig. 2D), degraded completely leaving behind the apparently fully developed lipid and callose layers in close proximity (Fig. 2F), more or less similar to that found in mature cucumber seeds.
|
Developing cucumber seeds at 25, 35 and 45 DPA were treated with TTC. After 12 h, embryos were isolated, dried and extracted in acetone for formazan. There was little difference in the formazan content of embryos in TTC-treated seeds between 25 and 35 DPA (Table 1). On the other hand, a reduction of about two-thirds was recorded in the formazan content of embryos in TTC-treated seeds from 35 to 45 DPA. To check whether the lower formazan content of embryos from TTC-treated seeds at 45 DPA was due to decreased permeability of the seed coverings to TTC or was on account of diminished capacity of embryos to reduce TTC, the embryos were isolated from seeds at 25, 35 and 45 DPA and treated with TTC. It is evident from the data (Table 1) that reduction in formazan content of the embryos in TTC-treated seeds at 45 DPA was attributable to a decrease in the permeability of the seed coverings. The marked decrease in the permeability of seed coverings, which occurred at 45 DPA, coincided with the complete development of lipid and callose layers (Fig. 2F).
|
Amritphale et al. (1993)
and Singh et al. (1993) demonstrated that acetone increases the permeability of seed coverings to TTC in cucumber, muskmelon and several other cucurbit seeds. However, the effect of acetone on the lipid or callose layer of the PE envelope was not investigated. As evident from Fig. 1E and F, the chloroform treatment resulted in reduced staining of the lipid layer, but not of the callose layer. To examine whether there was any change in the permeability of the PE envelope due to chloroform-induced weakening of the lipid layer, chloroform-treated cucumber or muskmelon seeds were tested for permeation of TTC. The formazan content in the embryos in chloroform-treated seeds was approx. 4-fold greater compared with that in the embryos in untreated seeds (Table 2). In contrast, no significant difference in the formazan content was observed when TTC treatment was given to the embryos isolated from chloroform-treated or -untreated seeds. The increase in formazan content in the embryos in chloroform-treated seeds could thus be attributed to enhanced permeability of the PE envelope rather than to increased metabolic activity in embryos.
|
Proton diffusion from the endosperm cell layer
In a related study, aimed at observing the permeation of 2,6-dichlorophenolindophenol, sodium salt (DCPIP) into seeds, the PE envelopes from cucumber or muskmelon seeds often showed a pink coloration. The oxidized form of DCPIP is known to be blue at neutral pH and pink at acidic pH. To check any possible proton diffusion, tissue printing was performed by placing halves of the PE envelopes isolated from seeds on agarose gels containing bromocresol purple, an acid–base indicator (pH range 5·2–6·8). Tissue prints having the characteristic yellow colour on account of proton diffusion became visible when the halves of the PE envelopes isolated from muskmelon (Fig. 3) or cucumber (data not shown) seeds were placed with their inner surface on the gel. On the other hand, no or little proton diffusion was detected when the halves of the PE envelopes had their outer surface on the gel, indicating that proton diffusion was either polarized in the endosperm cells or the lipid/callose layer prevented proton diffusion. Proton diffusion from the PE envelopes was also measured by recording the increase in the absorbance of bromocresol purple solution at 436 nm. A perceptible variation in proton diffusion did exist among individual PE envelopes in cucumber as well as muskmelon seeds, but the range was relatively narrow (Fig. 4A). Further, there was no significant difference in proton diffusion between the bulk samples of caps from the radicle end of the PE envelopes from viable seeds and those from the distal cotyledonary end (Fig. 4B). Autoclaving seeds resulted in complete loss of proton diffusion both in cucumber and muskmelon.
|
|
Activities of endo-ß-mannanase and ß-glucanase
A tissue printing method was employed to locate endo-ß-mannanase activity in the PE envelopes of cucumber and muskmelon seeds. On the basis of preliminary experiments, 7·5 % polyacrylamide gels having an initial pH of 6·2 and containing 0·1 % LBG were used. Perceptible clearing of the LBG-containing gel, indicative of endo-ß-mannanase activity, was observed in muskmelon seeds when the inner surface of halves of the PE envelopes was in contact with the gel (Fig. 5). The enzyme activity observed in the outer surface of the PE envelope was lower than that in the inner surface. Furthermore, instead of being localized to the radicle end or any other specific region, the hydrolysing activity was more or less uniformly distributed throughout the surface of the PE envelope. In cucumber, the enzymatic activity was relatively weak as compared with that in muskmelon, but the trend was more or less similar (data not shown).
|
The gel diffusion assay for quantifying endo-ß-mannanase activity showed it to be present in the PE envelopes in cucumber as well as muskmelon seeds, but not in the embryonic axis or cotyledons. Figure 6A shows changes in endo-ß-mannanase activity in the PE envelopes of cucumber and muskmelon seeds while imbibing water. A gradual increase in enzyme activity was observed from the time imbibition started to when the radicle emerged. Enzyme activity was relatively greater in muskmelon PE envelopes than that in cucumber PE envelopes. The activity of ß-glucanase was determined photometrically by measuring glucose formed from Laminaria digitata laminarin. Unlike endo-ß-mannanase activity, the activity of ß-glucanase was present in the embryonic axis, but not in the PE envelopes or cotyledons. Also, the activity of ß-glucanase was detected much later during the course of imbibition than was endo-ß-mannanase activity (Fig. 6B). A 3-fold increase in the enzyme activity observed between 12 h and 15 h of imbibition was nearly coincident with the initiation of seed germination in cucumber and muskmelon.
|
It was mentioned earlier that (a) tissue weakening in the region of micropylar endosperm to allow radicle emergence is known to require the action of several wall hydrolases in tomato, tobacco and other seeds, and (b) in addition to the presence of a callose-rich layer, the cell walls of the envelope in muskmelon, a species closely related to cucumber, are also rich in mannan and galactomannan polymers (Welbaum et al., 1998
). Thus, the activities of endo-ß-mannanase and ß-glucanase may be necessary for the weakening in the micropylar region of the PE envelope in cucumber and muskmelon. Endo-ß-mannanase activity was determined in the caps of the PE envelopes and in the radicles. Marked endo-ß-mannanase activity was present in the caps of the PE envelopes (Table 3). In contrast, radicles did not show any measurable enzyme activity in the gel diffusion assay. Diffusible, as well as extractable, activities showed only about a 1·5-fold increase from the start of imbibition (3 h) to the initiation of radicle emergence (15 h), both in cucumber and muskmelon. About 40–45 % of the total endo-ß-mannanase (diffusible plus extractable) activity diffused into the incubation medium during the 12-h incubation period. Enzyme activity was 1·5–2·0 times greater in the caps of muskmelon PE envelopes as compared with that in cucumber. To examine which part of the embryonic axis had greater ß-glucanase activity, plumule and radicles were excised and the activity in each was determined separately. In addition, the enzyme activity was determined in the caps of the PE envelopes. Diffusible and extractable activities were both about 2·5- to 3·5-fold greater in radicles relative to plumules in cucumber as well as muskmelon seeds (Table 4). In contrast, the caps of the PE envelopes showed neither diffusible nor extractable ß-glucanase activity.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
In seeds of cucumber, muskmelon and several other members of the Cucurbitaceae the embryo is surrounded by a membranous envelope. Yim and Bradford (1998)
showed that the envelope is composed entirely of a single layer of endosperm cells covered by a thin lipid-rich layer and a thick callose-containing layer and referred to it as the ‘endosperm envelope’. A relatively thick deposit of callose at 35 DPA was observed, as indicated by aniline blue staining, outside the walls of the endosperm cells, but not on their inner side (Fig. 2D and E). Like other cellulosic polysaccharides, callose is synthesized at the plasma membrane (Minorsky, 2002). Hence, a mechanism(s) must exist to deposit callose or its precursors outside the endosperm cell walls. No vesicles/globules were observed in any of the preparations in the present study, but, earlier, Yim and Bradford (1998) had reported discrete vesicles or globules stained with aniline blue when callose accumulation occurred at about 35 DPA in the muskmelon envelope. Interestingly, at 35 DPA, a little deposition of callose about one or two cells apart from the endosperm cell layer was also observed in the present work immediately below the lipid layer (Fig. 2D). Apparently, an apoplastic route might also exist to deposit callose or its precursors a few cells away from the endosperm cell layer. In contrast to the callose layer, which could have been derived from the endosperm, the lipid layer of the envelope in cucumber seed appeared to differentiate in the epidermis of the multilayered nucellus, a maternal tissue (Fig. 2A–C). In other words, while the callose layer of the envelope could be endospermic in origin, the lipid layer appeared to be of nucellar origin. Therefore, it would perhaps be appropriate to call the envelope surrounding the embryo in cucumber, and maybe also in other members of Cucurbitaceae, the perisperm-endosperm (PE) envelope.
The PE envelope in muskmelon was found to give a positive reaction for cutin and suberin when stained with Sudan IV (Welbaum and Bradford, 1990). Besides occurring as a major constituent of the cuticle, cutin is also found within the cell wall, i.e. in the interfibrillar and intermicellar spaces of cellulose (Fahn, 1982). Notably, the Sudan-staining layer was seen differentiating within the inner epidermal cell walls of the nucellus in developing cucumber seeds (Fig. 2C). Yim and Bradford (1998) found chloroform treatment removed the Sudan-staining layer and thus suggested that it was not composed of suberin. In the present study, substantially reduced staining, though not dissolution, of the Sudan-staining layer was also observed when decoated cucumber (compare Fig. 1E with 1F) or muskmelon (data not shown) seeds were treated with chloroform. The difference between the present observations and those of Yim and Bradford (1998) might perhaps be because their sections were obtained by sectioning by hand instead of with a plant microtome.
Yim and Bradford (1998) suggested that the callose layer caused the semi-permeability of the muskmelon envelope, whereas the lipid layer was of little consequence. Their suggestion was based on the observations that (a) removal of the lipid layer by chloroform treatment did not prevent autoclaved seeds from attaining osmotic distention, but degradation of the callose layer by ß-1,3-glucanase did, and (b) the lipid layer developed much before the acquisition of semi-permeability. While ß-1,3-glucanase action might be specific to callose, it is not unlikely that the degradation of the callose layer might also have affected the organization, if not the chemistry, of the lipid layer lying in close proximity. Chloroform treatment was found to enhance the permeation of TTC into cucumber seeds (Table 2) and to affect the lipid layer, but not the callose layer (compare Fig. 1E with 1F). It is thus possible that greater permeation of TTC into seeds on account of chloroform treatment could be related to its effect on the lipid layer. The present data for the initiation of differentiation of the lipid layer in developing cucumber seeds (Fig. 2A and B) agree with those of Yim and Bradford (1998) who showed that muskmelon envelope could be stained with Sudan III and IV at 25 DPA. It was observed, however, that, although the lipid layer started developing quite early, it only developed fully by 45 DPA, coincident with the complete development of callose layer (compare Fig. 2A with 2F). Permeation of TTC into developing cucumber seeds was also reduced by two-thirds from 35 to 45 DPA (Table 1). It appears in the light of the above findings that the lipid layer may have a modifying effect on the semi-permeability of the callose layer in cucumber, and perhaps also in muskmelon.
Edelstein and Kigel (1990) reported inhibition of seed germination in muskmelon ‘Noy Yizre’el at sub-optimal temperatures. Since decoating allowed the seeds to germinate at a lower temperature compared with that for intact seeds, the seed coat has been suggested to play a significant role in the inhibition of muskmelon seed germination (Edelstein and Kigel, 1990; Edelstein et al., 1995). Furthermore, Edelstein et al. (1995) also observed that the pressure necessary to split the seed coat at the radicle end decreased during seed imbibition. On the other hand, Welbaum and Bradford (1990) found that the perisperm envelope, and not the seed coat, in muskmelon ‘Top Mark’ acted as the primary barrier to radicle emergence. In later work, Welbaum et al. (1995) reported that the force and energy required to penetrate the envelope declined steadily during imbibition. Sreenivasulu and Amritphale (1999) also observed that a marked decrease in the penetration force preceded the splitting of the PE envelope, the first visible sign of germination in cucumber. The cell walls of the PE envelope in muskmelon contain >60 % mannose (Welbaum and Wang, 1997). Substantial LBG-hydrolysing activity was observed in the PE envelopes in cucumber and muskmelon seeds (Figs 5 and 6A). Welbaum (1999) observed endo-ß-mannanase activity to increase gradually in the endosperm tissue of developing muskmelon seeds from 25 DPA onwards and to peak between 45 and 50 DPA as seeds reached maximum germinability. Pronounced diffusible and extractable endo-ß-mannanase activity was also found in the caps of the PE envelopes, but not in the radicles, much before the emergence of the radicle in cucumber and muskmelon (Table 3).
In addition to the breakdown of mannan-rich endosperm cell walls at the micropylar end, the degradation of callose may be necessary for the weakening of the PE envelope associated with radicle emergence in cucumber or muskmelon seeds. A measurable laminarin-hydrolysing activity was observed in the embryonic axes (Fig. 6B), as well as in the diffusate or extract of radicles (Table 4) excised from cucumber or muskmelon seeds just before the splitting the PE envelope. In contrast to the observations of Yim and Bradford (1997), who reported that the callose layer degrades in germinating muskmelon seeds, a general degradation of the callose layer in cucumber or muskmelon seeds, coincident with the protrusion of radicle, was not observed in the present study (Fig. 1C and D). Presumably, overall degradation of the callose layer could be a post-germinative event in cucumber and muskmelon.
Many of the hydrolases including endo-ß-mannanase and ß-glucanase are known to have acidic pH optima. Izhar and Frenkel (1971) found that in fertile Petunia hybrida anthers, the pH during meiosis was 6·8–7·0 and the activity of callase could not be detected. However, the pH dropped to 5·9–6·2 at the tetrad stage, concomitant to a sharp rise in callase activity resulting in the breakdown of callose around the tetrads. Their results thus indicated that a drop in pH was a prerequisite for callase activity. It is thus possible that the diffusion of protons from the cells of the endosperm layer (Figs 3 and 4A and B) might create an acidic environment conducive to the activation of endo-ß-mannanase and ß-glucanase in cucumber and muskmelon seeds. Further, it has been proposed that endosperm cap weakening in tomato is a biphasic process (Karssen et al., 1989; Toorop et al., 2000). While the first phase in the endosperm weakening in tomato seed was correlated with endo-ß-mannanase, the expression of ß-1,3-glucanase and chitinase occurred in the second phase prior to radicle emergence. The presence of separate mannan-rich and callose-rich layers and a structural obligation of the radicle to pierce them at two different points of time during the course of germination imply that endo-ß-mannanase and ß-glucanase may play a role in the germination of cucumber and muskmelon seeds. However, further work evaluating their direct role in cell wall modification or tissue weakening and involving gene expression is necessary before drawing any conclusion in this regard.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
The authors are grateful to Professor Kent J. Bradford (University of California, Davis, CA, USA) for providing muskmelon seeds and also for his kind suggestions during the initial phase of this work. Thanks are also due to Megazyme International, Wicklow, Ireland for their generous gift of Aspergillus niger endo-ß-mannanase.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Amritphale D, Dixit S, Singh B. 1993. Effect of acetone on the induction and breakage of secondary dormancy in seeds of cucumber. Journal of Experimental Botany 44: 1621–1626.
Amritphale D, Yoneyama K, Takeuchi Y, Ramakrishna P, Kusumoto D. 2005. The modulating effect of perisperm-endosperm envelope on the ABA-inhibition of seed germination in cucumber. Journal of Experimental Botany 56: 2173–2181.
Beresniewicz MM, Taylor AG, Goffinet MC, Koeller WD. 1995a. Chemical nature of a semipermeable layer in seed coats of leek, onion (Liliaceae), tomato and pepper (Solanaceae). Seed Science and Technology 23: 135–145.
Beresniewicz MM, Taylor AG, Goffinet MC, Terhune BT. 1995b. Characterization and location of a semipermeable layer in seed coats of leek and onion (Liliaceae), tomato and pepper (Solanaceae). Seed Science and Technology 23: 123–134.
Chen F, Bradford KJ. 2000. Expression of an expansin is associated with endosperm weakening during tomato seed germination. Plant Physiology 124: 1265–1274.
Chen F, Nonogaki H, Bradford KJ. 2002. A gibberellin-regulated xyloglucan endotransglycosylase gene is expressed in the endosperm cap during tomato seed germination. Journal of Experimental Botany 53: 215–223.
Dawson RMC, Elliott DC, Elliott WH, Jones KM. 1986. Data for biochemical research, 3rd edn. Oxford: Clarendon Press.
Edelstein M, Kigel J. 1990. Seed germination of melon (Cucumis melo) at sub- and supra-optimal temperatures. Scientia Horticulturae 45: 55–63.[CrossRef]
Edelstein M, Corbineau F, Kigel J, Nerson H. 1995. Seed coat structure and oxygen availability control low-temperature germination of melon (Cucumis melo) seeds. Physiologia Plantarum 93: 451–456.[CrossRef]
Fahn A. 1982. Plant anatomy, 3rd edn. Oxford: Pergamon Press.
Ikuma H, Thimann KV. 1963. The role of the seed-coats in germination of photosensitive lettuce seeds. Plant Cell Physiology 4: 169–185.
Izhar S, Frankel R. 1971. Mechanism of male-sterility in Petunia. I. The relationship between pH, callase activity in the anthers and the breakdown of microsporogenesis. Theoretical and Applied Genetics 41: 104–108.
Karssen CM, Haigh A, van der Toorn P, Weges R. 1989. Physiological mechanisms involved in seed priming. In: Taylorson RB, ed. Recent advances in the development and germination of seeds. New York: Plenum Press, 269–280.
Leubner-Metzger G, Meins F. 2000. Sense transformation reveals a novel role for class I ß-1,3-glucanase in tobacco seed germination. The Plant Journal 23: 215–221.[CrossRef][Web of Science][Medline]
Leubner-Metzger G. 2003. Functions and regulation of ß-1, 3-glucanases during seed germination, dormancy release and after-ripening. Seed Science Research 13: 17–34.[CrossRef][Web of Science]
Minorsky PV. 2002. The wall becomes surmountable. Plant Physiology 128: 345–353.
Morohashi Y, Matsushima H. 2000. Development of ß-1,3-glucanase activity in germinated tomato seeds. Journal of Experimental Botany 51: 1381–1387.
Nonogaki H, Gee OH, Bradford KJ. 2000. A germination-specific endo-ß-mannanase gene is expressed in the micropylar endosperm cap of tomato seeds. Plant Physiology 123: 1235–1245.
Petruzzelli L, Muller K, Hermann K, Leubner-Metzger G. 2003. Distinct expression patterns of ß-1,3-glucanases and chitinases during the germination of solanaceous seeds. Seed Science Research 13: 139–153.[CrossRef][Web of Science]
Salyers AA, Palmer JK, Wilkins TD. 1977. Laminarinase (ß-glucanase) activity in Bacteroides from the human colon. Applied and Environmental Microbiology 33: 1118–1124.
Singh B. 1953. Studies on the structure and development of seeds of Cucurbitaceae. Phytomorphology 3: 224–239.
Singh B, Dixit S, Amritphale D. 1993. A modified tetrazolium chloride test for seed viability in some cucurbitaceae. Indian Journal of Experimental Biology 31: 1002–1003.
Sreenivasulu Y, Amritphale D. 1999. Membrane fluidity changes during ethanol-induced transition from dormancy to germination in cucumber seeds. Journal of Plant Physiology 155: 159–164.
Taylor AG, Beresniewicz MM, Goffinet MC. 1997. Semipermeable layer in seeds. In: Black EM, Murdoch AJ, Hong TD, eds. Basic and applied aspects of seed biology. Dordrecht: Kluwer Academic Publishers, 429–436.
Thornton ML. 1968. Seed dormancy in watermelon Citrullus vulgaris Shrad. Proceedings of the Association of Official Seed Analysts 58: 80–84.
Toorop PE, Bewley JD, Hilhorst HWM. 1996. Endo-ß-mannanase isoforms are present in the endosperm and embryo of tomato seeds, but are not essentially linked to the completion of germination. Planta 200: 153–158.[Web of Science]
Toorop PE, van Aelst AC, Hilhorst HWM. 2000. The second step of the biphasic endosperm cap weakening that mediates tomato (Lycopersicon esculentum) seed germination is under control of ABA. Journal of Experimental Botany 51: 1371–1379.
Welbaum GE. 1999. Cucurbit seed development and production. HortTechnology 9: 341–348.
Welbaum GE, Bradford KJ. 1990. Water relations of seed development and germination in muskmelon (Cucumis melo L.). IV. Characteristics of the perisperm during seed development. Plant Physiology 92: 1038–1045.
Welbaum GE, Wang Y. 1997. Enzymatic degradation of perisperm envelope tissue in germinating muskmelon seeds. Plant Physiology 114 (Suppl. 292): abstract no. 1523.
Welbaum GE, Muthui WJ, Wilson JH, Grayson RL, Fell RD. 1995. Weakening of muskmelon perisperm envelope tissue during germination. Journal of Experimental Botany 46: 391–400.
Welbaum GE, Bradford KJ, Yim KO, Booth DT, Oluoch MO. 1998. Biophysical, physiological and biochemical processes regulating seed germination. Seed Science Research 8: 161–172.
Witmer X, Nonogaki H, Beers EP, Bradford KJ, Welbaum GE. 2003. Characterization of chitinase activity and gene expression in muskmelon seeds. Seed Science Research 13: 167–178.[CrossRef]
Wu C-T, Leubner-Metzger G, Meins F Jr, Bradford KJ. 2001. Class I ß-1,3-glucanase and chitinase are expressed in the micropylar endosperm of tomato seeds prior to radicle emergence. Plant Physiology 126: 1299–1313.
Yim KO, Bradford KJ. 1997. Relationship of callose accumulation to semipermeability of the perisperm of muskmelon seeds. Plant Physiology 114 (Suppl. 288): abstract no. 1506.
Yim KO, Bradford KJ. 1998. Callose deposition is responsible for apoplastic semipermeability of the endosperm envelope of muskmelon seeds. Plant Physiology 118: 83–90.
Zou X, Nonogaki H, Welbaum GE. 2002. A gel diffusion assay for visualization and quantification of chitinase activity. Molecular Biotechnology 22: 19–23.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. A. Salanenka, M. C. Goffinet, and A. G. Taylor Structure and Histochemistry of the Micropylar and Chalazal Regions of the Perisperm-endosperm Envelope of Cucumber Seeds Associated with Solute Permeability and Germination J. Amer. Soc. Hort. Sci., July 1, 2009; 134(4): 479 - 487. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||