Planta (2002) 214: 468±475 DOI 10.1007/s004250100641

O R I GI N A L A R T IC L E

Geraldine A. Toole á Andrew C. Smith Keith W. Waldron

The effect of physical and chemical treatment on the mechanical properties of the cell wall of the alga Chara corallina Received: 22 December 2000 / Accepted: 20 April 2001 / Published online: 11 October 2001 Ó Springer-Verlag 2001

Abstract Single large internode cells of the charophyte (giant alga) Chara corallina were dissected to give sheets of cell wall, which were then notched and their mechanical properties in tension determined. The cells were subjected to a thermal treatment in excess water (cf. cooking), which had little e€ect on strength but increased the sti€ness, contrasting with the behaviour of higher-plant tissues. Extraction in CDTA (cyclohexanetrans-1,2-diamine-N,N,N¢,N¢-tetraacetate) or 4 M KOH reduced the strength from 17 MPa to 10 MPa, although sequential extraction in CDTA and 4 M KOH reduced the strength further to 4 MPa. The sti€ness decreased from 500 MPa to 300 MPa on extraction in CDTA or 4 M KOH, while falling to 70 MPa after extraction in CDTA followed by 4 M KOH. Conventional sequential extraction in CDTA, Na2CO3 at 1 °C and 20 °C, and KOH at 0.5 M, 1 M, 2 M and 4 M caused a gradual decrease in sti€ness and strength after the CDTA treatment to the same lower values. This result is in keeping with mechanical properties for plant tissues, but in contrast to the removal of pectic polysaccharides from model cell wall systems, which does not reduce the sti€ness. Keywords Cell wall á Chara (cell wall) á Mechanical properties (cell wall) á Sti€ness (cell wall) á Strength (cell wall) Abbreviation CDTA: cyclohexane-trans-1,2-diamine-N, N,N¢,N¢-tetraacetate

Introduction Plant cells are surrounded by a tough, yet ¯exible, polymeric wall that determines the shape of the cell, G.A. Toole á A.C. Smith (&) á K.W. Waldron Institute of Food Research, Norwich Research Park, Colney, Norwich, NR4 7UA, UK E-mail: [email protected] Fax: +44-1603-507723

while being strong enough to enable turgor pressures of up to 108 N m±2 to develop (Cosgrove 1993); hence the mechanical properties of the cell wall are of great importance. By imposing a series of physical and chemical treatments on plant cell wall samples, then determining the changes in tensile strength and modulus of elasticity, it is possible to show the relative impact of the various components of the cell wall on its overall mechanical properties. The extremely large size of the individual cells of charophytes, such as Chara corallina, makes it possible to obtain material for a useful model system for studying the mechanical properties of plant cell walls. These giant algae consist of a linear series of nodes and internodes. The nodes are multicellular with lateral cells of limited and unlimited growth. The internodes are single cylindrical multinucleate cells which have a central vacuole surrounded by a thin layer of cytoplasm. These cells are large enough to be opened up to give intact sheets of primary cell wall suitable for mechanical testing. This is not easily achieved in higher plants, which consist of heterogeneous small cells held together by components of the middle lamella. Nevertheless, previous studies on multicellular tissue strips have revealed that primary cell walls from the tissues of higher plants show stress relaxation and creep properties that are similar to those determined for characean internode walls (Cleland 1971). The chemical composition of the cell walls of charophytes is similar to that of dicotyledonous plants (Gillet and Liners 1996). The plant cell wall may be modelled as three interwoven polymeric networks: a micro®brillar framework made up of long, crystalline cellulose micro®brils interlaced with xyloglucan polymers; an amorphous matrix of pectins, consisting of pectic polysaccharides, polygalacturonic acid and rhamnogalacturonan (substituted with small polymeric side groups of arabinan, galactan and arabinogalactan), linked ionically by calcium bridges; and structural proteins covalently cross-linked to elements in the wall matrix. The pectin within the cell wall may be highly branched, in

469

areas often described as the `hairy regions', which alternate with areas where the pectin is unbranched, or the `smooth regions'. These networks interact with one another in many ways (Carpita and Gibeaut 1993) and at least one bears the mechanical stresses in the wall, yet very little is known about the contributions of stresses among these cell wall components. The aim of this research was to determine the sti€ness and fracture strength, in tension, of intact cell-wall material derived from single cells of Chara corallina, following physical and chemical extraction treatments designed to remove classes of cell wall components by sequentially deconstructing the chemical structure of the cell wall. Earlier reports by MeÂtraux (1982) and Morrison et al. (1993) used the cell wall of Nitella to investigate the changes in carbohydrate composition of the cell wall during maturation. Morrison et al. (1993) also determined which carbohydrates were extracted from the cell wall during a sequential extraction. Here, the carbohydrate composition of the cell walls was determined following each treatment. The tensile strength, modulus and notch sensitivity of the Chara cell wall were determined previously (Toole et al. 2001) and this approach was used here to determine the changes in the mechanical properties of the cell wall following selected physical and chemical treatments. Tensile tests, using prepared test pieces of extracted Chara cell wall with a single-edge notch, were carried out on a universal testing machine. The tensile strength and the Young's modulus were calculated from the data obtained, and compared with the cell-wall carbohydrate composition.

of cell wall. Then, leaving a section (at least 5 mm long) at the end intact, the needle was withdrawn slowly using the sharp tip as a knife to cut along the length of the cell. The tube was opened up to provide a rectangular strip of cell wall with a section of cell wall tube at one end. The cell wall was rinsed thoroughly with distilled water to remove all cell contents. The cell wall strips were then treated by boiling and by extraction in cyclohexane-trans-1,2-diamine-N,N,N¢,N¢-tetraacetate (CDTA) and 4 M KOH separately and sequentially. A full sequential extraction with CDTA, Na2CO3 at 1°C and 20°C, and 0.5 M, 1 M, 2 M and 4 M KOH was also carried out. The strips were then assessed for mechanical properties, wall thickness and carbohydrate content. Some cells were classi®ed according to maturity for carbohydrate analysis to complement mechanical-property results from the earlier study (Toole et al. 2001).

Materials and methods

4 M KOH

Plant materials

Opened internode cells were extracted in 4 M KOH solution for 2 h. They were then neutralised in 0.1 M acetic acid and stored in distilled water. Tensile tests were conducted using 30 extracted samples, and approximately 20 sample strips were retained for carbohydrate analysis.

Chara corallina was grown in 20-l glass tanks according to the method described by Toole et al. (2001). Branches of algae were obtained from the tanks on the day of use and stored in distilled water. Instrumentation A binocular light microscope was used to aid the preparation of test samples. Tensile tests were conducted and recorded using a TA XT2 Texture Analyser with Texture Expert software (Stable Micro Systems, Godalming, Surrey, UK). A Leitz Ortholux (Wetzlar, Germany) microscope equipped with di€erential interference contrast (DIC) optics and a calibrated graticule were used to measure the thickness of the cell walls. Gas chromatography was conducted on a Perkin-Elmer (Beacons®eld, Bucks, UK) AutoSystem XL gas chromatograph using a Restek Rtx-225 WCOT column (15 m long, 0.32 mm i.d.; 0.25 lm ®lm; Thames Chromatography, Windsor, Berks., UK). During uronic acid analysis, absorbances were determined using a Perkin Elmer 550S UV/Vis spectrophotometer. Sample preparation Single internode cells were excised from the axis using a razor blade then placed in a petri dish containing distilled water. Using a dissecting microscope, a hypodermic needle was inserted into the tube

Extraction of cell walls Thermal treatment Opened internode cells were extracted in a 3-l vessel, containing 1 l of boiling distilled water. After time intervals of 0 h, 0.5 h, 1 h, 2 h and 3 h, 10 cell-wall strips were removed and stored in distilled water at room temperature for tensile testing. Approximately 10 additional sample strips were removed from the boiling water after 2 h for the determination of carbohydrate content. CDTA Opened internode cells were extracted in 50 mM CDTA (Na+ salt, adjusted to pH 6.5 using 1 M NaOH). Cells were removed after 1 h, 3 h, and 24 h, then rinsed thoroughly and stored in distilled water. Tensile tests were conducted using 20 samples extracted for 1 h, 40 samples extracted for 3 h, and 30 samples extracted for 24 h. Additionally, approximately 20 sample strips were extracted in CDTA for 3 h then retained for carbohydrate determination.

CDTA followed by 4 M KOH Opened internode cells were extracted in 50 mM CDTA solution (adjusted to pH 6.5 using 1 M NaOH) for 1 h then rinsed and extracted in 4 M KOH solution for 1 h. They were then neutralised in 0.1 M acetic acid and stored in distilled water. Tensile tests were conducted using 30 extracted samples, and approximately 20 sample strips were retained for carbohydrate analysis. Full sequential extraction Opened internode cells were extracted sequentially using a method adapted from that used to extract cell walls of onion (Redgwell and Selvendran 1986), asparagus tissues (Waldron and Selvendran 1992) and Chinese water chestnut (Parker and Waldron 1995). Cell wall strips were extracted with 50 mM CDTA (Na+ salt, pH 6.5) at 20 °C for 6 h; 50 mM Na2CO3 and 20 mM NaBH4 at 1 °C overnight (Na2CO3-I), followed by 20 °C for 2 h (Na2CO3-II); 0.5 M KOH and 20 mM NaBH4 at 20 °C for 2 h; 1 M KOH and 20 mM NaBH4 at 20 °C for 2 h; 2 M KOH and 20 mM NaBH4 at 20 °C for 2 h; and ®nally 4 M KOH and 20 mM NaBH4 at 20 °C for 2 h. The alkali extractions were carried out with O2-free

470 solutions under nitrogen. Following each extraction stage, cell-wall strips were removed for mechanical testing and carbohydrate analysis. Five replicate extractions were conducted. Tensile testing was conducted using 30 samples each, for the 7 extraction stages and the control (non-extracted). Sucient amounts of extracted (and non-extracted) samples at each stage were also retained for carbohydrate analysis. Cell-wall thickness measurement Following treatment, the section remaining in tube form at the end of each cell wall strip was removed using a razor blade; it was retained in distilled water and used to determine the thickness of the individual cell wall which was measured to an accuracy of ‹1 micro;m as described previously (Toole et al. 2001). Tensile testing Each opened cell-wall strip, typically 30 mm long and 2±3 mm wide, was placed onto a slide (with 0.1-mm graduations) within a petri dish of distilled water, and its width was determined. With the aid of a low-power microscope, a notch, 0.3 mm in length, was cut at the midpoint of a long edge using a razor blade. Cyanoacrylate adhesive (Cyanolit AdheÂsif 223-F; Eurobond, Sittingbourne, UK) was used to glue the short edges of the notched specimen between two sets of custom-made metal plates held 15 mm apart by a transfer clamp (Toole et al. 2001). A single-edge notch was thereby located midway along and perpendicular to the sample length. The sample was immersed in a custom-made plastic tank containing distilled water. Tensile tests were then carried out at a test speed of 0.1 mm s±1. The force versus distance pro®les were recorded and used to determine the maximum force and the initial slope of the force-displacement curve. Calculations Young's modulus, E, was calculated from: Eˆ

L dF t:w dx

…1†

where L is original length, t is the thickness and w is the width of the sample, and dF/dx is the initial slope of the force-displacement curve. The strength, r, of each sample was calculated from: rˆ

Fmax t:w

…2†

where Fmax is the maximum force. Carbohydrate analysis Following each treatment or extraction the cell-wall strips were airdried then ground to a powder and cell-wall neutral sugar and uronic acid contents determined. Several replicate analyses were conducted, each requiring 5 mg each of dried sample. Sugars were released from the cell wall by dispersion in 72% (v/v) H2SO4 for 3 h followed by dilution to 1 M and hydrolysis for 2.5 h at 100 °C (Saeman et al. 1954). Neutral sugars were reduced with NaBH4 and acetylated by the method of Blakeney et al. (1983) using 2-deoxyglucose as an internal standard. The resulting alditol acetates were quanti®ed by gas chromatography on a Perkin-Elmer AutoSystem XL gas chromatograph after automatic injection by autosampler. Alditol acetates were separated with baseline resolution on a Restek Rtx-225 WCOT column (15 m long, 0.32 mm i.d.; 0.25 lm ®lm) using an oven temperature programme of 140 °C for 5 min, 2.5 °C per min to 210 °C and 210 °C for 2 min. Helium was used as the carrier gas at a column head pressure of 5 bar, detection

was by ¯ame ionisation and the data were collected and integrated using the software PE Turbochrom. Total uronic acid content of the cell-wall samples was determined colorimetrically by a modi®cation of the method of Blumenkrantz and Asboe-Hansen (1973) in which the cell-wall strips were dispersed in 72% (v/v) H2SO4 for 3 h at room temperature, diluted to 1 M H2SO4, and hydrolysed for 1 h at 100 °C. The cell wall hydrolysates were dispersed in dilute acid then reacted quantitatively with m-phenylphenol in 0.5% NaOH to give a pink derivative with an absorption at 524 nm. The mole percent of each carbohydrate present in the cell wall samples was calculated. Cell-wall maturity Control wall strips were also produced from individual internode cells, which were classi®ed by maturity following Proseus et al. (1999): young cells from the region of new growth at the top of the plant (3±12 days old), medium from the centre of the plant and old from the mature base of the plant (over 30 days old). This complements the earlier mechanical-property studies of walls from cells at these stages of maturity (Toole et al. 2001). In all other cases in this paper, all stages of maturity were combined randomly.

Results and discussion Cell-wall maturity The strength and Young's modulus of the Chara cell wall at di€erent stages of maturity were determined earlier (Toole et al. 2001). The strength of the cell wall remained unchanged throughout growth; however, the sti€ness increased gradually with maturity. Carbohydrate analysis of the cell wall (Table 1) showed that there was a decrease in galacturonic acid content as the cell wall matured from young (rapidly growing), to medium (slow growing), to old (not growing or fully grown), the di€erence between medium and old cells being statistically signi®cant (P<0.05). These results are consistent with the work of MeÂtraux (1982) on Nitella cell walls where the relative proportions of uronides, expressed as a percentage of the dry weight of the wall, decreased while the proportion of cellulose increased with maturity. Morrison et al. (1993) also found that the uronic acid content of the Nitella cell wall dropped with increasing maturity, from 33.7% for rapidly growing cells, to 31.4% for slowly growing cells, to 26.3% for recently mature cells. The fall in uronic acid content in Chara cell walls coincided with increased proportions of neutral sugars and cellulose within the cell wall, as it matured, which is consistent with the results for Nitella cell walls (MeÂtraux 1982; Morrison et al. 1993). Galacturonic acid confers the ability to stretch to the cell wall (Van Buren 1979) and therefore its reduction during increasing maturity of the Chara cell wall is possibly responsible for the increase in sti€ness. Such changes occur in the inner, most recently deposited layer of the algal wall as the cell matures, consistent with the ®ndings of Richmond (1983) who highlighted the stressbearing role of the inner part of the Nitella wall. The present results also suggest that the synthesis of these cell wall components occurred at di€ering rates throughout

471 Table 1 Carbohydrate compositions of the Chara corallina cell wall before and after individual treatments. For control and treated cells all stages of maturity were combined randomly. Data Extracted wall

n

Cell wall carbohydrates (mol%)

Untreated Young

3

Medium

3

Old

3

Control Treated Water 100 °C

are means and SD (in parentheses) for n replicates. Rha Rhamnose, Fuc fucose, Ara arabinose, Xyl xylose, Man mannose, Gal galactose, Glc glucose, Urn uronic acid

12 3

CDTA

3

4 M KOH

3

CDTA/4 M KOH

4

Rha

Fuc

Ara

Xyl

Man

Gal

Glc

Urn

1.40 (0.27) 1.73 (0.01) 2.15 (0.06) 1.64 (0.08)

2.67 (0.25) 2.80 (0.04) 3.50 (0.10) 2.49 (1.16)

1.89 (0.18) 1.86 (0.13) 2.11 (0.07) 1.88 (0.07)

4.97 (0.46) 4.94 (0.18) 5.42 (0.02) 4.61 (0.26)

5.88 (0.10) 6.63 (0.75) 5.06 (0.66) 5.22 (0.26)

1.29 (0.11) 1.73 (0.09) 2.08 (0.05) 1.84 (0.26)

54.74 (4.91) 57.42 (0.58) 60.01 (0.30) 53.81 (2.98)

27.18 (6.30) 22.88 (1.47) 19.20 (0.09) 28.51 (4.05)

1.47 (0.11) 1.33 (0.09) 1.08 (0.03) 0.77 (0.04)

2.68 (0.18) 3.71 (0.16) 2.57 (0.04) 3.80 (0.13)

1.83 (0.08) 1.98 (0.07) 1.71 (0.09) 2.04 (0.05)

4.68 (0.26) 5.58 (0.20) 4.72 (0.13) 5.05 (0.05)

5.29 (0.32) 6.91 (0.25) 5.02 (0.22) 6.22 (0.18)

1.57 (0.15) 1.14 (0.06) 0.91 (0.22) 0.59 (0.03)

53.69 (3.76) 71.61 (2.95) 57.44 (2.34) 74.01 (0.60)

28.78 (4.79) 7.74 (3.63) 26.55 (2.82) 7.53 (0.78)

maturity, which could be involved in the regulation of growth rate and cell elongation (MeÂtraux 1982). Thermal treatment The strength of the Chara cell wall remained constant at 17±19 MPa over the 3-h cooking period (Fig. 1), whereas the elastic modulus or sti€ness of the wall increased from 484 MPa to 880 MPa (Fig. 2), the increase becoming statistically signi®cant (P<0.01) after 3 h cooking. This coincided with only a very slight reduction in rhamnose and galactose in the wall, determined

Fig. 1 Strength of the Chara corallina cell wall after various treatments. The treatments and number of replicates (in parentheses) were: control (n=10); thermal treatment for 0.5 h (n=10), 1 h (n=10), 2 h (n=10) and 3 h (n=10); CDTA extraction for 1 h (n=20), 3 h (n=40) and 24 h (n=30); extraction in 4 M KOH for 2 h (n=30); extraction in CDTA for 2 h followed by 4 M KOH for 2 h (n=30). Data are means ‹ SE

during carbohydrate analysis (Table 1), which suggested a very slight solubilisation of cell wall pectic polysaccharides. The e€ects of thermal treatment on the mechanical properties of cell walls are dicult to determine using multicellular plant tissues, due to cell separation in conjunction with numerous other changes. During thermal treatment, pectic substances involved in cell adhesion become soluble, and the cells separate easily, thereby softening the tissue. A loss of turgor experienced during heat treatment also adds to the softening of the plant tissue. The denaturing of proteins in the cell membrane allows water to move out of the cell, thereby increasing its softness (Ainsworth 1994). Starch granules within the cells of a tissue absorb cellular water and swell

Fig. 2 Modulus of elasticity of the Chara cell wall following various treatments. Treatments and numbers of replicates were as for Fig. 1. Data are means ‹ SE

472

to form a gel (Andersson et al. 1994), which also has an e€ect on the mechanical properties of the tissue. Studies on the e€ects of heat treatment on most vegetable tissues show a drop in both strength and Young's modulus following treatment. Verlinden et al. (1996) showed that a rupture stress of 0.54 MPa for raw carrot was reduced to 0.09 MPa following cooking, and a 20% reduction in the modulus of carrot tissue was also reported. The apparent modulus of elasticity for potato tissue fell from 3.34 MPa to 0.53 MPa following heat treatment (Alvarez and Canet 1998). Waldron et al. (1997) reported a fall in tensile strength of raw potato from 0.34 MPa to 0.04 MPa on steaming for 16 min. These cases showed the mode of fracture moving from cell wall fracture to cell separation, suggesting that the loss in strength and modulus was due to a change in mechanical properties of the whole tissue rather than the cell wall. The tensile strength of Chinese water chestnut tissue did not change following 20 min of steaming because, as a result of the substantial presence of phenolics within the middle lamella, cell wall breakage rather than cell separation occurred (Waldron et al. 1997). By using a single strip of cell wall, the complexity of multicellular systems described earlier was eliminated, and therefore, in this study the e€ect of heat treatment on the cell wall alone was determined. Extraction in CDTA CDTA is a chelating agent; therefore, it solubilises the pectic polysaccharides which are held adjacent to each other by Ca2+ bridges (Selvendran 1985). Extraction in CDTA reduced the strength of the cell wall dramatically to almost half its original value, falling from 17.0 MPa to 9.7 MPa after 1 h (P<0.01), thereafter remaining almost constant after 24 h extraction (Fig. 1). The modulus of elasticity also decreased from 484 MPa to 322 MPa after 24 h, (signi®cant at P<0.05) (Fig. 2). Carbohydrate analysis of the cell wall (Table 1) showed a decrease in rhamnose and galactose and a considerable decrease in galacturonic acid, which was consistent with the removal of calcium-bound pectic polysaccharides. CDTA solubilisation of galacturonic acid-rich pectic polysaccharides is consistent with results on the extraction of dried and powdered Nitella cell walls by Morrison et al. (1993). They observed that 67% of the polysaccharide extracted from the wall using CDTA was galacturonic acid. The e€ect of CDTA extraction on the mechanical properties of the Chara cell wall can be compared with the work of Chanliaud and Gidley (1999). They studied cell wall mechanical properties, using a model cell-wall system based on cellulose produced by the bacterium Acetobacter xylinus, and found that the sti€ness in tension was decreased and the toughness, obtained from the area under force-displacement curves, was increased by the inclusion of pectin to form a cellulose/pectin composite. More importantly, they observed that the subse-

quent removal of pectin from the composite, using CDTA, had no e€ect on the mechanical properties, concluding that within the cellulose/pectin composites the load-bearing component was the cellulose network alone. This contrasts with the study reported here in that the removal of pectin from the cell wall of Chara caused a signi®cant drop in its strength and modulus of elasticity, indicating that in this system the pectin network does contribute to the mechanical properties of the cell wall. A slight de-lamination of the cell wall strips was also observed following extraction with CDTA, and was presumably due to the removal of the Ca2+-bound pectins which are concentrated between the micro®brillar layers, or lamellae, of the cell wall (Kamiya et al. 1963). Slippage between the micro®brillar layers, due to the removal of pectin, may have reduced the cell wall sti€ness. Extraction in 4 M KOH Extraction of the Chara cell wall with 4 M KOH was an extreme treatment, which reduced the strength of the cell wall considerably from 17 MPa to 8.5 MPa (P<0.01) (Fig. 1), and the modulus of elasticity from 484 MPa to 337 MPa (P<0.05) (Fig. 2). The carbohydrate analysis of the cell wall (Table 1) showed a reduction in rhamnose, arabinose, and galactose, with no signi®cant change in galacturonic acid content. This was consistent with the removal of highly branched pectic polysaccharides only. The pectin chains are likely to have been subjected to b-elimination. Due to the absence of NaBH4, a stepwise peeling of the polysaccharide chains may have a€ected the structural integrity of the wall leading to a signi®cant decrease in its mechanical properties. It is, therefore, surprising that so little galacturonic acid was solubilised. Extraction in CDTA followed by extraction in 4 M KOH Extraction in CDTA prior to extraction in 4 M KOH produced a synergistic e€ect compared with the separate treatments, and an extreme loss of strength and reduction in Young's modulus was observed. Compared with the values for cells extracted in CDTA or 4 M KOH, the strength and modulus decreased signi®cantly (P<0.01) to 4.4 MPa (Fig. 1) and to 72 MPa (Fig. 2), respectively. The cell wall strips remaining after extraction were deformed, and had greatly reduced width and length but increased thickness, re¯ecting delamination of the micro®brillar layers. They were easily stretched and ¯exible in nature, consistent with the considerable reduction in modulus. Carbohydrate analysis (Table 1) revealed a large decrease in rhamnose, arabinose, galactose and galacturonic acid, indicating that virtually all of the pectic polysaccharides had been removed. The reductions in carbohydrates observed during the individual extraction treatments, when added together, were

473

similar to those achieved by the treatments in sequence, suggesting that the removal of the relatively poorly branched pectins by CDTA extraction was followed by the removal of the branched pectins by extraction in 4 M KOH. In addition, there was possibly a small reduction in xylose, suggesting the removal of xyloglucan hemicelluloses. The remaining cell wall strip had a glucose content of 75 mol% neutral sugars, indicating that it consisted mainly of cellulose micro®brils. These results indicate that the CDTA- and 4 M KOH-induced changes in extensibility and cell wall chemistry are separate and additive.

Full sequential extraction This extraction procedure was designed to sequentially deconstruct the chemical structure of the cell wall, whilst minimising degradation of the component polysaccharides (Redgwell and Selvendran 1986). The addition of NaBH4 to the alkali served to minimise alkaline peeling during the initial stages of extraction and to solubilise the polysaccharides in as close to their native form as possible (Redgwell and Selvendran 1986). Extraction in CDTA had similar chemical (Table 2) and mechanical (Figs. 3, 4) e€ects as before (Table 1; Figs. 1, 2). The decreases in sti€ness and strength were signi®cant at the same level as before (Extraction in CDTA). The initial Na2CO3 (Na2CO3-I) extraction was performed at 1 °C in order to minimise b-elimination of pectins (Redgwell and Selvendran 1986). Some of the remaining CDTA-insoluble, unbranched pectic polysaccharides were subsequently solubilised by dilute Na2CO3 at 1 °C and then 20 °C (Na2CO3-II), by hydrolysis of weak ester cross-links. Carbohydrate analysis showed a slight reduction in pectic polysaccharides with little change in the galacturonic acid, galactose, arabinose and rhamnose contents following each of the two weak-alkali

Fig. 3 Strength of the Chara cell wall following sequential extraction with: CDTA, Na2CO3-I, Na2CO3-II, 0.5 M KOH, 1 M KOH, 2 M KOH, 4 M KOH. Data are means ‹ SE, n=30

Fig. 4 Modulus of elasticity of the Chara cell wall following sequential extraction with: CDTA, Na2CO3-I, Na2CO3-II, 0.5 M KOH, 1 M KOH, 2 M KOH, 4 M KOH. Data are means ‹ SE, n=30

Table 2 Carbohydrate compositions of the Chara cell wall before and after sequential extraction. Data are means and SD (in parentheses) for n replicates. Rha Rhamnose, Fuc fucose, Ara arabinose, Xyl xylose, Man mannose, Gal galactose, Glc glucose, Urn uronic acid Extracted wall

n

Control

10

CDTA

10

Na2CO3-I

10

Na2CO3-II

10

0.5 M KOH

10

1 M KOH

10

2 M KOH

10

4 M KOH

10

Cell wall carbohydrates (mol%) Rha

Fuc

Ara

Xyl

Man

Gal

Glc

Urn

1.89 (0.04) 1.18 (0.12) 1.04 (0.12) 1.08 (0.15) 0.71 (0.11) 0.68 (0.12) 0.53 (0.12) 0.51 (0.06)

2.77 (1.15) 3.87 (0.31) 3.89 (0.23) 3.73 (0.32) 3.84 (0.27) 4.01 (0.30) 3.85 (0.29) 4.11 (0.22)

1.88 (0.09) 2.36 (0.33) 2.34 (0.16) 2.20 (0.18) 2.15 (0.14) 2.09 (0.20) 2.08 (0.19) 1.99 (0.13)

3.89 (0.11) 5.17 (0.18) 5.40 (0.29) 5.18 (0.31) 5.33 (0.35) 5.21 (0.22) 5.03 (0.28) 3.95 (0.15)

4.51 (0.31) 6.16 (0.30) 6.51 (0.52) 6.00 (0.40) 6.44 (0.89) 6.31 (0.31) 6.11 (0.39) 6.48 (0.06)

2.03 (0.28) 1.47 (0.18) 1.34 (0.21) 1.22 (0.25) 0.98 (0.18) 0.88 (0.18) 0.66 (0.22) 0.70 (0.10)

49.61 (1.10) 71.70 (2.26) 72.60 (1.96) 72.97 (2.62) 74.89 (3.14) 74.69 (2.57) 76.68 (1.99) 79.81 (0.09)

33.42 (1.09) 7.72 (1.33) 6.89 (1.31) 7.62 (2.23) 5.67 (2.54) 6.13 (2.75) 5.06 (1.10) 2.46 (0.11)

474

extractions (Table 2). The strength and elastic modulus were not signi®cantly changed (Figs. 3, 4). Extraction in KOH solutions of increasing strength reduced the strength and elastic modulus of the cell wall even further, and there was a rapid decrease in the strength of the cell wall as the alkalinity of the extractant increased. The strength fell from 10.3 MPa following the Na2CO3 extractions to 6.4 MPa after 2 M KOH (P<0.01) and then ®nally to only 5.2 MPa after 4 M KOH extraction (P<0.05) (Fig. 3). The modulus also decreased rapidly with increasing extractant alkalinity, falling from 360 MPa following the Na2CO3 extractions to 170 MPa after 2 M KOH (P<0.01) and then ®nally to only 75 MPa after 4 M KOH (P<0.01) extraction (Fig. 4). Similar reductions in strength and modulus were achieved with extraction in CDTA followed by 4 M KOH without the presence of NaBH4 (Figs. 1, 2). Carbohydrate analysis (Table 2) showed a slight reduction in rhamnose, arabinose and galactose following the 0.5 M and 1 M KOH extractions, and a further reduction in galacturonic acid was observed following the 2 M and 4 M KOH extractions, as the remaining cell wall pectic polysaccharides were solubilised. A signi®cant decrease in xylose (Table 2) following the 4 M KOH extraction suggested the gradual removal of cell wall hemicelluloses. The remaining cell wall strip had a glucose content of 80 mol% neutral sugars, indicating that it consisted mainly of cellulose micro®brils. A greater percentage of non-cellulosic sugars was present than in the cellulosic residue from Nitella (Morrison et al. 1993) where sub-millimeter dried Nitella cell wall particles were sequentially extracted. Less non-cellulosic sugar was present than in sequentially extracted tissues such as carrot (Massiot et al. 1988), potato (Ryden and Selvendran 1990) and onion (Redgwell and Selvendran 1986) where fractions of cell wall and middle lamella (and possibly traces of the cell contents) were extracted. However, the Chara cell wall was both chemically and structurally intact in its native form for the sequential extractions. To date, sequential extractions of this type have usually been conducted using cell wall material (CWM) extracted from plant tissues (Redgwell and Selvendran 1986; Ng and Waldron 1997). Parker (2000) conducted extractions on potato parenchyma, and recorded a drop in tensile strength from 0.3 MPa to 0.15 MPa following a 12-h extraction in CDTA; extraction for 48 h resulted in total cell separation of the tissue. A 3-h extraction in CDTA followed by a 3-h extraction in NaCO3 also reduced the strength of the potato tissue by half; however, any subsequent extractions with KOH resulted in a breakdown of the tissue. Sequential extractions of CWM show a comparable breakdown of the chemical structure of the cell wall, with similar amounts of polysaccharide being removed at each stage of the extraction; however, it is impossible to follow the e€ects of this deconstruction on the mechanical properties of the cell wall as they deal with sub-millimeter particles of cell wall and middle lamella.

Conclusions The strength and therefore the mechanical integrity of the Chara cell wall were una€ected by heat treatment in an excess of distilled water (akin to cooking), and the modulus of elasticity was increased by less than a factor of 2. On extraction with CDTA or KOH, the strength of the cell wall decreased from 17 MPa to 9±10 MPa and the sti€ness decreased from 500 MPa to 320± 340 MPa, whereas sequential extraction in CDTA and KOH reduced the strength to 4 MPa and the elastic modulus decreased to 75 MPa. A full sequential extraction of the cell wall, resulting in the removal of almost all of the wall polysaccharides except cellulose, reduced the cell-wall strength to 5 MPa and the modulus to 75 MPa. The sequential deconstruction and removal of polysaccharides from the cell wall were directly responsible for the loss in tensile strength and reduction in modulus of elasticity. Cellulose content alone does not determine mechanical properties. Acknowledgements This work was funded from the Biotechnology and Biological Sciences Research Council Competitive Strategic Grant and as part of a BBSRC Research Studentship (G.A.T.). We thank Dr. John Thain at the University of East Anglia for the initial cultures of Chara.

References Ainsworth P (1994) Chemistry in the kitchen ± fruit and vegetables 1. Nutr Food Sci 5:25±28 Alvarez MD, Canet W (1998) Rheological characterization of fresh and cooked potato tissues (cv. Monalisa). Z Lebensm Unters Forsch A 207:55±65 Andersson A, Gekas V, Lind I, Oliveira F, OÈste R (1994) E€ect of preheating on potato texture. Crit Rev Food Sci Nutr 34:229± 251 Blakeney AB, Harris PJ, Henry RJ, Stone BAA (1983) Simple and rapid preparation of alditol acetates for monosaccharide analysis. Carbohydr Res 113:291±299 Blumenkrantz N, Asboe-Hansen G (1973) New method for quantitative determination of uronic acids. Anal Biochem 54:484± 489 Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in ¯owering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J 3:1±30 Chanliaud E, Gidley MJ (1999) In vitro synthesis and properties of pectin/Acetobacter xylinus cellulose composites. Plant J 20:25±35 Cleland R (1971) Cell wall extension. Annu Rev Plant Physiol 22:197±222 Cosgrove DJ (1993) How do plant cell walls extend? Plant Physiol 102:1±6 Gillet G, Liners F (1996) Changes in distribution of short pectic polysaccharides induced by monovalent ions in the Nitella cell wall. Can J Bot 74:26±30 Kamiya N, Tazawa M, Takata T (1963) The relation of turgor pressure to cell volume in Nitella with special reference to mechanical properties of the cell wall. Protoplasma 57:501±521 Massiot P, Rouau X, Thibault JF (1988) Isolation and characterisation of the cell-wall ®bres of carrot. Carbohydr Res 172:217±227 MeÂtraux JP (1982) Changes in cell-wall polysaccharide composition of developing Nitella internodes. Planta 155:459±466

475 Morrison JC, Greve LC, Richmond PA (1993) Cell wall synthesis during growth and maturation of Nitella internodal cells. Planta 189:321±328 Ng A, Waldron KW (1997) E€ect of steaming on cell wall chemistry of potatoes (Solanum tuberosum cv. Bintje) in relation to ®rmness. J Agric Food Chem 45:3411±3418 Parker CC (2000) The (bio)chemistry of cell adhesion in edible plant tissues: its role in texture. PhD thesis, University of East Anglia, UK Parker ML, Waldron KW (1995) Texture of chinese water chestnut: Involvement of cell wall phenolics. J Sci Food Agric 68:337±346 Proseus TE, Ortega JKE, Boyer JS (1999) Separating growth from elastic deformation during cell enlargement. Plant Physiol 119:775±784 Redgwell RJ, Selvendran RR (1986) Structural features of cell wall polysaccharides of onion Allium cepa. Carbohydr Res 157:183± 199 Richmond PA (1983) Patterns of cellulose micro®bril deposition and rearrangement in Nitella: in vivo analysis by a birefringence index. J Appl Polym Sci: Appl Polym Symp 37:107±122 Ryden P, Selvendran RR (1990) Structural features of cell wall polysaccharides of potato (Solanum tuberosum). Carbohydr Res 195:257±272

Saeman JF, Moore WE, Mitchell RL, Millett MA (1954) Techniques for the determination of pulp constituents by quantitative paper chromatography. Tech Assoc Pulp Paper Ind J 37:336±343 Selvendran RR (1985) Developments in the chemistry and biochemistry of pectic and hemicellulosic polymers. J Cell Sci [Suppl] 2:51±88 Toole GA, Gunning PA, Parker ML, Smith AC, Waldron KW (2001) Fracture mechanics of the cell wall of Chara corallina. Planta 212:606±611 Van Buren JP (1979) The chemistry of texture in fruits and vegetables. J Text Stud 10:1±12 Verlinden BE, De Barsy T, De Baerdemaeker J, Deltour R (1996) Modelling the mechanical and histological properties of carrot tissue during cooking in relation to texture and cell wall changes. J Text Stud 27:15±28 Waldron KW, Selvendran RR (1992) Cell wall changes in immature Asparagus stem tissue after excision. Phytochemistry 31:1931±1940 Waldron KW, Smith AC, Parr AJ, Ng A, Parker ML (1997) New approaches to understanding and controlling cell separation in relation to fruit and vegetable texture. Trends Food Sci Tech 8:213±221