| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Physiology and Biochemistry |
2Unilever R&D Colworth, Sharnbrook, Beds. MK44 1LQ, UK; 3School of Biological and Environmental Sciences, University of Stirling, FK9 4LA, UK; 4Cooperative Research Centre for Bioproducts, School of Botany, University of Melbourne, Victoria 3010, Australia; and 5Centre for Nutrition and Food Sciences, University of Queensland, St. Lucia, Queensland 4072, Australia
Received for publication April 7, 2006. Accepted for publication August 23, 2006.
ABSTRACT
A cellulose/xyloglucan framework is considered to form the basis for the mechanical properties of primary plant cell walls and hence to have a major influence on the biomechanical properties of growing, fleshy plant tissues. In this study, structural variants of xyloglucan have been investigated as components of composites with bacterial cellulose as a simplified model for the cellulose/xyloglucan framework of primary plant cell walls. Evidence for molecular binding to cellulose with perturbation of cellulose crystallinity was found for all xyloglucan types. High molecular mass samples gave homogeneous centimeter-scale composites with extensive cross-linking of cellulose with xyloglucan. Lower molecular mass xyloglucans gave heterogeneous composites having a range of microscopic structures with little, if any, cross-linking. Xyloglucans with reduced levels of galactose substitution had evidence of self-association, competitive with cellulose binding. At comparable molecular mass, fucose substitution resulted in a modest promotion of microscopic features characteristic of primary cell walls. Taken together, the data are evidence that galactose substitution of the xyloglucan core structure is a major determinant of cellulose composite formation and properties, with additional fucose substitution acting as a secondary modulator. These conclusions are consistent with reported structural and mechanical properties of Arabidopsis mutants lacking specific fucose and/or galactose residues.
Key Words: cellulose • Gluconacetobacter • primary cell walls • xyloglucan
The biomechanical performance of plants is influenced by a range of structural factors spanning molecular to tissue scales. A conserved feature of all common plant types, though, is the use of cellulose as a key building block for the structures that underlie the diverse demands of biomechanical performance, accommodation of growth processes, protection of cell contents, and active apoplastic biological processes. Cellulose has the valuable property of being a very effective use of metabolic carbon for load-bearing purposes, but is limited in the variety of structures that it can form on its own. The combination of cellulose with other cell wall polymers is what gives rise to the diversity of biological properties associated with the wide range of plant tissue types. A number of schematic models for how cellulose interacts with other polymers to form functional wall architectures have been proposed (reviewed in Somerville et al., 2004
). This information is derived from a combination of "top down" analytical techniques such as a range of microscopic and spectroscopic techniques (e.g., Jarvis and McCann, 2000
), increasingly coupled with analysis of the consequences of "bottom up" defined genetic modifications, either transgenic (e.g., Sorensen et al., 2000
) or mutational (e.g., Turner et al., 2001
; Somerville et al., 2004
). The rapidly developing knowledge base of functional genomics is undoubtedly going to lead to the generation of a large number of plants with defined genetic variation for phenotypic characterization. The opportunity exists therefore to develop molecular design principles for the biomechanical performance of plants that can be exploited through selection or modification of genotypes.
Two continuing challenges in relating molecular events to biomechanics are the hierarchical structure of plant tissues and the (usually) highly heterogeneous nature of both the cellular architecture and the molecular composition of walls surrounding a single cell, that have only recently been revealed via antibody probes to wall polysaccharides (Willats et al., 2001
). To build a knowledge base about the consequences of specific molecular modification on plant biomechanics, it would be useful to understand how simplified cell wall systems behave as materials. This data can then be used in cell- and tissue-level models for whole-plant mechanical behavior.
We have taken a complementary approach of constructing composites of cellulose with potential non-cellulose cross-linking polymers such as xyloglucan (Whitney et al., 1995
), mannans (Whitney et al., 1998
), and pectins (Chanliaud and Gidley, 1999
). This approach has now been extended to lignin-like polymers (Touzel et al., 2003
). The bacterium Gluconacetobacter xylinus (formerly known as Acetobacter aceti subsp. xylinum) produces cellulose polysaccharides in a way that parallels closely the deposition of cellulose into the extracellular environment of plant cells. After 2–3 d of fermentation, a continuous mat or pellicle of cellulose with centimeter lateral dimensions and several millimeters in depth is formed. If grown under nonagitated conditions, this pellicle is sufficiently homogeneous to be used for mechanical testing. The G. xylinus system allows laboratory study of associations between cellulose and other polymers through incorporation of test polymers into the fermentation medium. This mimics the assembly of plant cell walls to the extent that cellulose is deposited from the cell surface into a medium containing other polymers. In the case of plant cells, these other polymers are present in the extracellular environment after having been synthesized in the Golgi apparatus and secreted in vesicular form across the plasma membrane. Composites formed by the self-assembly of G. xylinus cellulose with representative cell wall polymers can be studied at the molecular level through compositional analysis or NMR spectroscopy, at the microstructural level with microscopy, and at the macroscopic level with materials testing. The molecular and microscopic features of composites may then be compared with those of plant cell walls.
The formation, structure, and properties of composites between G. xylinus cellulose and xyloglucan have been studied extensively. Using tamarind seed xyloglucan as a convenient source of pure, high molecular mass polymer, it has been shown (Whitney et al., 1995
) that composites form spontaneously by deposition of bacterial cellulose into xyloglucan solutions. A cross-linked microstructure is formed that shares microstructural and molecular similarity with related plant cell walls (Whitney et al., 1995
). Cross-linking with xyloglucan has no effect on the small deformation (oscillatory) mechanical properties of cellulose (Whitney at al., 1999), but results in a dramatic weakening under uniaxial tension (Whitney et al., 1999
) and greater extensibility and nonlinear-elastic behavior under biaxial tension (Chanliaud et al., 2002
). The study of xyloglucan–cellulose composites has provided much insight into the molecular mode of action of putative catalysts of plant growth (expansins; Whitney et al., 2000
) and the mechanical consequence of the action of xyloglucan degrading and transglycosylating enzymes (Chanliaud et al., 2004
).
Xyloglucans are ubiquitous components of primary cell walls in nongraminaceous terrestrial plants, in which they contribute greatly to the ability of plant cell walls to expand and extend during growth (Takeda et al., 2002
). There is also evidence that they play an important role in the formation of secondary cell walls of vascular tissues (Bourquin et al., 2002
). All xyloglucans are based on a cellulosic (1
4)-ß-d-glucose backbone with frequent substitution with 
-d-xylosyl residues at C(O)6 of glucose residues. There are however, multiple variations in both the number and pattern of xylosyl substituents along the glucan backbone and in the presence of a range of additional glycosyl and acetyl substituents (Sims et al., 1996
; Vincken et al., 1997
). Given the structural diversity of xyloglucan polymers, it is of interest to determine whether this translates into a diversity of interactions and composite formation with bacterial cellulose. This information builds on other data on the affects of structural variation in the binding of xyloglucan added to preformed cellulose. These studies led to conclusions that (1) binding increases as xylose content decreases (Chambat et al., 2005
) and (2) that fucosyl-galactose substitution of the xylose residues are key contributors to cellulose binding (Levy et al., 1997
). This last proposal is not supported by subsequent analyses of mutants of Arabidopsis thaliana that suggest that galactosyl substitution of xylose residues is critical for mechanical properties, whereas additional substitution by fucose is not (Ryden et al., 2003
; Pena et al., 2004
).
To assess the role of a range of xyloglucan structural variants on composite formation with cellulose, we now report a comparison of tamarind xyloglucan with xyloglucans obtained from Nicotiana plumbaginifolia (Sims et al., 1996
) and Pyrus communis (pear) suspension culture cells. In addition, tamarind xyloglucan has been treated with a polymer-acting ß-d-galactosidase (Edwards et al., 1988
) to reduce the level of galactose substituents on the xyloglucan core, prior to introduction to the bacterial fermentation system. The effects of acetyl substitution in Nicotiana plumbaginifolia xyloglucan have been evaluated using chemical deacetylation. The core molecular structures of the xyloglucans used in this study are shown schematically in Fig. 1. Effects of molecular mass are also assessed through comparing two tamarind seed xyloglucan samples. The results obtained have been compared with previous data on the interaction of xyloglucans with cellulose, and implications for plant biomechanics and growth processes are discussed.
|
Xyloglucans and modifications
Tamarind seed xyloglucan (Glyloid 3S) with mean molecular mass of 880 kDa was obtained from Dainippon Pharmaceutical, Osaka, Japan, and purified (Gidley et al., 1991
) before use. For comparison, xyloglucan was also purified from a lower molecular mass (ca. 200 kDa) tamarind seed source (Das Industries, Mumbai, India). Xyloglucan from Nicotiana plumbaginifolia and Pyrus communis suspension culture cells was purified as described by Sims and Bacic (1995)
and characterized using methods described in Sims et al. (1996)
. Structural features of N. plumbaginifolia xyloglucan are described in Sims et al. (1996)
. For P. communis xyloglucan, hydrolysis with endo (14)-ß-d-glucanase released a range of oligosaccharides, most (>90%) based on a Glc4Xyl3 structure as shown in Fig. 1. The two most abundant oligosaccharides were Glc4Xyl3 (27%) and Glc4Xyl3Gal1Fuc1 (36%) with smaller amounts of Glc4Xyl3Gal2Fuc1 (11%), Glc4Xyl3Gal1 (12%), and Glc4Xyl3Gal2 (6%) (see Fig. 1), characteristic components of many primary plant cell wall xyloglucans (Vincken et al., 1997
). Characterization of oligosaccharide fragments showed that acetyl substitution was limited to galactose residues but could be on any hydroxyl group (J. Webster and A. Bacic, unpublished manuscript). Deacetylation was achieved by treating xyloglucans with 1 mol/L NaOH at room temperature for 1 h followed by neutralization with glacial acetic acid. Samples were recovered by lyophilization following extensive dialysis against deionized water at 4°C.
For the preparation of galactose-depleted tamarind seed xyloglucan, ß-d-galactosidase was purified from nasturtium seeds as described by Edwards et al. (1988)
except that, after cation exchange on a CM-cellulose column (Whatman CM-52, Brentford, UK) at pH 5.0, fractions containing ß-d-galactosidase activity were pooled and applied to a gel filtration column (Biogel P60, Bio-Rad, Hercules, CA). Fractions containing activity were collected and stored at –18°C. To use galactose-depleted xyloglucans directly in the bacterial cellulose fermentation system, the ß-d-galactosidase preparation (6 mL) was added to a solution (60 mL 0.5% mass/volume) of purified tamarind seed xyloglucan in pre-sterilized Hestrin-Schramm medium (Hestrin and Schramm, 1954
), and the solution was stirred at room temperature. Samples (20 mL) were removed after 16, 24, and 48 h, placed in a boiling water bath for 10 min to inactivate enzymes, and a sample was removed for compositional analysis. The remainder was used directly in G. xylinus fermentations. A control solution containing denatured enzyme preparation was used as a source of native tamarind seed xyloglucan in fermentation experiments. The extent of galactose removal was assessed by monosaccharide analysis (Blakeney et al., 1983
). In enzyme reactions terminated after 16, 24, and 48 h, galactose content in the xyloglucan was reduced from 16.5% at 0 h to 12.4%, 11.3% and 6.6%, respectively (relative reductions in galactose of ca. 25%, 35%, and 60%, respectively).
Preparation and analysis of composites
Gluconacetobacter xylinus (ATCC 53524) was incubated in Hestrin-Schramm medium (Hestrin and Schramm, 1954
) containing 2% glucose for 48 h at 30°C under either static or agitated (50 rpm orbital shaker) conditions (Whitney et al., 1995
). Individual xyloglucans were included in the fermentation medium as 0.5% solutions. Cellulosic pellicles formed after fermentation were washed extensively with fermentation medium without xyloglucan to remove any nonspecifically adsorbed xyloglucan.
The monosaccharide composition of composites was determined after hydrolysis in 12 mol/L H2SO4 at room temperature for 1 h, followed by 3 h in 1 mol/L H2SO4 at room temperature, then neutralized with 0.1 mol/L NaOH. The resulting monosaccharides were separated using high-performance anion exchange chromatography (Dionex, Sunnyvale, CA). Samples were eluted from a PA-10 column with deionized water then treated with 0.3 mol/L NaOH and analyzed with a PAD 2 detector (Dionex, Sunnyvale, CA). Knowledge of the monosaccharide composition of the xyloglucans allows the data to be expressed as cellulose/xyloglucan ratios within composites.
The molecular-level organization of polymers within composites was assessed with solid state 13C NMR as described previously (Whitney et al., 1995
). Chemical shifts for the diagnostic anomeric carbon signals of xyloglucans were obtained either from literature values (Gidley et al., 1991
; Watt et al., 1999
) for tamarind and N. plumbaginifolia xyloglucan, or by correlation of assigned 1H anomeric peaks (York et al., 1996
) with 13C signals using two-dimensional NMR.
Composites were microstructurally characterized using deep-etch freeze-fracture transmission electron microscopy (Whitney et al., 1995
).
Mechanical properties of composites were assessed by uniaxial tensile testing, because tamarind xyloglucan composites have major differences to pure bacterial cellulose (Whitney et al., 1999
; Chanliaud et al., 2004
). Rectangular strips of pellicle of typical geometry 30 x 3 x 1.5 mm were cut using a razor blade. The two ends were placed directly between vise grips in a Minimat materials tester (Polymer Laboratories, Loughborough, UK), and the grips moved apart at a constant speed of 10 mm/min. A 20-N load beam was used to record the force required for extension as a function of time. From geometrical measurements, force–deformation data were converted to apparent stress–strain profiles. At least six independent measurements were made for each composite type to assess consistency of results.
RESULTS
Xyloglucan content in composites
From monosaccharide analysis of washed composites, the ratio of cellulose to xyloglucan in composites was determined (Table 1). A combination of static and agitated culture conditions was used. Static incubations result in more xyloglucan incorporation into composites. Less of the lower molecular mass tamarind xyloglucan was incorporated than the higher molecular weight mass material, even when the concentration in the fermentation media was quadrupled. This effect may explain the relatively low incorporation levels for N. plumbaginifolia xyloglucan, which had lower viscosity in solution than the 230-kDa tamarind xyloglucan. The P. communis xyloglucan had similar solution viscosity to the 230 kDa in agreement with previous data (Sims et al., 1998
) and a somewhat higher incorporation level. Galactose-depleted tamarind xyloglucan gave a wide range of incorporation rates up to nearly 1 : 1 for the lowest galactose content sample.
|
and Iß, have characteristic resonances that can be used to quantify their relative amounts based on published spectra with defined ratios (Debzi et al., 1991
; Yamamoto and Horii, 1993
). Results are shown in Table 1. Apart from the 60% galactose-depleted sample, all composites has a significant reduction in I to Iß ratio from 70 : 30 to between 40 : 60 and 50 : 50, respectively. This reduction provides direct evidence for the molecular perturbation of cellulose assembly by the presence of all but one of the xyloglucans tested. The differences in allomorph ratio in cellulose : xyloglucan composites are modest, so to a first approximation, all the xyloglucans tested have a similar ability to perturb cellulose crystallinity, presumably by direct association with cellulose chains close to the point of deposition and before mature fibrillar ribbons are formed (Whitney et al., 1995
). The one xyloglucan that has a negligible effect on cellulose crystalline form is the most galactose-depleted tamarind sample. The I : Iß differences are exemplified in Figs. 2–5, which show cross polarization and magic angle spinning (CPMAS) spectra in original (c) and resolution-enhanced (b) forms. The signals ranging from 88 to 92 ppm are from C-4 sites within crystalline regions of cellulose. The ratio of intensities of the two peaks at ca. 89 and 91 ppm either side of the major peak at 90 ppm are diagnostic of the I to Iß ratio. Figs. 4b and 5b have a slightly higher peak at 89 compared to 91 ppm corresponding to I : Iß of 45 : 55 (Table 1), whereas Fig. 3b have a higher peak at 91 compared to 89 ppm corresponding to a ratio of 70 : 30 (Table 1) (Debzi et al., 1991
; Yamamoto and Horii, 1993
).
|
|
|
|
) and, as found previously for composites of bacterial cellulose with tamarind xyloglucan (Whitney et al., 1995
), gives spectra with chemical shifts characteristic of solubilized polymer, with underrepresentation of less-mobile backbone glucose residues. For example, Figs. 2a and 3a have C-1 signals at ca. 100, 103.5, and 105 ppm assigned to xylose, glucose, and galactose residues, respectively (Gidley et al., 1991
). For P. communis xyloglucan (Fig. 4a), the spectrum in this region is more complex, reflecting the greater structural diversity of this polymer (Fig. 1). The fucose residue is expected to give a C-1 resonance at 100.7 ppm, just downfield of the xylose resonances at 99.5–100.2 (Watt et al., 1999
). It is possible that the double-intensity peak seen in Fig. 4b also reflects a combination of xylose and fucose intensity, because the xylose signal from tamarind xyloglucan composites gives an apparent single resonance (Figs. 2 and 3). For N. plumbaginifolia xyloglucan, the intensity of xyloglucan signals is low (Fig. 5), reflecting the low level of incorporation in the composite (Table 1). Two features of the CPMAS spectra of composites in Figs. 2, 4, and 5 are (1) the presence of signal intensity at the same chemical shift as that of xylose resonances in the SPMAS spectra (ca. 100 ppm) and (2) the apparent absence of intensity at the chemical shift of glucose resonances in the SPMAS spectrum (103.5 ppm). Figure 3c shows a broader cellulose signal covering the range 103.5 ppm, so this latter conclusion cannot be extended to highly galactose-reduced samples.
Ultrastructures within composites
Figures 6–9 show representative ultrastructures observed for cellulose/xyloglucan composites. The characteristic cross-linked structure observed for high molecular mass tamarind xyloglucan (Whitney et al., 1995
; Chanliaud et al., 2004
) is seen in 25% (Fig. 6) and 35% galactose-depleted xyloglucan, to a lesser extent for P. communis xyloglucan grown under agitated conditions, and hardly at all for other composites. The fact that low molecular mass tamarind xyloglucan does not exhibit cross-links shows that polymer size is a major factor. Estimates of the molecular mass for P. communis xyloglucan by light scattering after size exclusion chromatography suggests a molecular mass of between 200 and 250 kDa, with that for N. plumbaginifolia xyloglucan being considerably less (data not shown). A summary of structural features seen in micrographs is given in Table 2.
|
|
|
). In this study, local alignment was seen for many samples, with high molecular mass tamarind and P. communis xyloglucans having the strongest effect. In general, agitation seemed to favor local alignment over static conditions presumably due to local shear flow effects. Apart from P. communis xyloglucan, lower molecular mass also lowered the extent of fiber alignment.
Composites prepared during this study had a greater diversity of ultrastructures than the cross-linked and aligned structure characterized for high molecular mass tamarind xyloglucan (Whitney et al., 1995
). This diversity is reflected both in additional structural types (discussed later) as well as in greater heterogeneity within single samples, i.e., different fields of view within the same composite often had different microstructures. This finding is in contrast to high molecular mass tamarind xyloglucan composites with cellulose for which structural features are conserved both within one sample and across sample replicates. The three additional structure types observed are exemplified in Figs. 7–9. Structures shown are representative of those found in other composites as identified in Table 2. Figure 7 shows a region that appears to be purely cellulosic, although it is possible that xyloglucan polymers are aligned with fibers without giving rise to any cross-links. This feature was seen in many composites (Table 2), notably for the most galactose-depleted tamarind composite and P. communis xyloglucan composite grown under static conditions. Figure 8 shows a region with a high density of thin strands assigned to xyloglucan chains and interspersed with thicker and partially aligned fibers assigned to cellulose. This "intimate mixing" is not seen for composites containing high molecular mass tamarind or 60% galactose-depleted tamarind but is a feature of all other composites (Table 2). Figure 9 shows a feature that is only observed for galactose-depleted xyloglucans. The "rounded" local network structure is characteristic of weak gels such as those formed by certain galactomannans either on their own (locust bean gum; Brigham et al., 1994
) or in the presence of cellulose (guar gum; Whitney et al., 1998
). This structure is taken as evidence for extensive self-association of xyloglucan chains within the composite structure. For the most galactose-depleted xyloglucan, composites contained a mixture of this xyloglucan-rich feature with the cellulosic network of the type shown in Fig. 7. This finding demonstrates a high degree of segregation (phase separation) of the two polymer types within the composite.
|
|
; Chanliaud et al., 2002
, 2004
). Uniaxial mechanical properties of the composites prepared in this study were investigated in an extension test. For many samples (low molecular mass tamarind, P. communis, and N. plumbaginifolia), replicate mechanical tests gave highly variable results (data not shown). This observation is probably related to the presence of a large number of ultrastructural features within a single piece of composite material (Table 2) giving rise to variable levels of heterogeneity within composite samples. The only samples that gave reproducible mechanical data (i.e., between replicate fermentation products) were the galactose-depleted tamarind family of composites. Representative traces are shown in Fig. 10 compared with cellulose alone and high molecular mass tamarind composites (Whitney et al., 1999
). In all cases, strain and stress to break were considerably reduced compared with either cellulose or the native xyloglucan composite. Again, this finding is probably due to greater ultrastructural heterogeneity, resulting in more facile failure. However, significant differences in stiffness (slope) were seen prior to failure (Fig. 10), with the most galactose-depleted sample having a stiffness value very similar to cellulose. Mechanical failure of composites was essentially elastic i.e., fractured material recovered original dimensions. Ultrastructural observation with TEM of fractured material showed no obvious difference in microstructure compared with pre-stretched composites.
|
A range of xyloglucans are incorporated into composites and bound to cellulose
For the wide range of xyloglucan structural types examined (Fig. 1), significant amounts are incorporated in composites formed by deposition of bacterial cellulose into a xyloglucan solution. Extensive washing of composites prior to analysis is expected to remove any loosely associated polymers, as has been previously observed, e.g., in pectin in the absence of calcium (Chanliaud and Gidley, 1999
) and in fenugreek galactomannan (Whitney et al., 1998
). Although there are variations in the amounts of xyloglucan incorporated, it is clear that all structural types have a sufficiently strong molecular affinity for cellulose to be incorporated within composites, irrespective of whether cross-links are observed (Table 2).
Two lines of evidence support a model of direct molecular binding between cellulose and the range of xyloglucans tested. One is the finding of altered relative amounts of cellulose crystalline forms (I : Iß; Table 1) for cellulose deposited into xyloglucan solutions. This effect on cellulose crystallinity is interpreted as being due to the molecular association of xyloglucan with nascent cellulose fibrils as secreted from bacteria, prior to formation of the final ribbon-shaped fiber characteristic of Gluconacetobacter cellulose (Haigler et al., 1980
). As noted previously (Whitney et al., 1995
), the I : Iß ratio found for xyloglucan-containing composites is typical of that described for plant cell walls (ca. 50 : 50; Foster et al., 1996
) compared with the 70 : 30 value for bacterial cellulose (Table 1). The second line of evidence for direct molecular binding of all xyloglucan types to cellulose is the differential xyloglucan backbone glucose C-1 but not side chain xylose C-1 NMR chemical shifts (Figs. 2–4) for "rigid" segments (detected by the CPMAS experiment) and "mobile" segments (detected by the SPMAS experiment; Foster et al., 1996
). Within washed composites, examined under never-dried conditions, all cellulose and some xyloglucan signal intensity are found in the "rigid-segment" CPMAS spectrum, with some xyloglucan intensity also in the "mobile-segment" SPMAS spectrum. This last is assigned to thin-stranded cross-links (e.g., Fig. 6) or surface "loops" of xyloglucan that are tethered by molecular binding to cellulose. The chemical shift values are essentially identical to those found in aqueous solution (Gidley et al., 1991
). Where significant cross-links are not observed (e.g., for N. plumbaginifolia), there is low intensity in the SPMAS spectrum (Fig. 5). Xyloglucan intensity in CPMAS spectra shows a signal corresponding to xylose C-1 at ca. 100 ppm, but a lack of intensity at 103.5 ppm, the chemical shift of "mobile" glucose C-1. We interpret this as due to a conformation-induced chemical shift change, most probably to chemical shift values very close to cellulose C-1 (105–106 ppm). Taken together with the observation of altered cellulose crystallinity, a likely explanation is that xyloglucan segments bind to nascent cellulose secreted from bacterial surfaces by adopting the two-fold conformation characteristic of crystalline cellulose. This mode of binding leads to both a change in chemical shift of bound glucose residues and the perturbation of chain organization such that the cellulose I : Iß ratio is altered.
The fact that qualitatively similar behavior is found for all of the xyloglucan types examined suggests that the common xyloglucan backbone is the principal determinant of the ability to bind to cellulose and that the various substitution patterns studied do not prevent this binding. Although N. plumbaginifolia and P. communis xyloglucans are able to bind to cellulose, they were isolated from a cell suspension culture in a noncellulose-bound state. This lack of apparent interaction with cellulose is probably because the relatively low level of cellulose typical of plant cell walls in culture (ca. 20%; Iraki et al., 1980
) only binds a fraction of the xyloglucan secreted by cells, with the remainder forming part of the readily separated extracellular polysaccharide fraction (Sims and Bacic, 1995
).
Molecular mass has a major influence on composite formation and properties
This study has shown large effects of xyloglucan molecular mass on the composition, ultrastructure, and heterogeneity of cellulose-based composites. Table 1 shows that the lower the molecular mass of the tamarind seed xyloglucan, the lower the level of incorporation within composites. This molecular weight effect may well be linked to the absence of cross-linking domains in micrographs for composites containing the lower molecular mass xyloglucan. The extensive cross-linking observed for higher molecular mass xyloglucan (Fig. 6) would lead to a higher level of incorporation for the same amount of direct cellulose-xyloglucan binding. The lower level of incorporation for P. communis and particularly N. plumbaginifolia xyloglucans is probably related to their relative molecular masses. Although no cross-linking was observed for composites of N. plumbaginifolia xyloglucan, some cross-linking was observed for P. communis xyloglucan under agitated conditions (Table 2). Previous studies utilizing high molecular mass tamarind xyloglucan have generated composites with reproducible mechanical behavior in both uniaxial and biaxial tension (Whitney et al., 1999
, 2000
; Chanliaud et al., 2002
, 2004
). For the uniaxial tests in this study, reproducible results were also found (Fig. 7). However, for lower molecular mass tamarind xyloglucan and both N. plumbaginifolia and P. communis xyloglucans, composites gave irreproducible uniaxial tensile data, showing variation in both apparent stiffness and, particularly, a high level of variation in values for the stress and strain to break. This variation is indicative of significant microstructural heterogeneity within individual composites, such that mechanical properties reflect both variable microstructures and failure in heterogeneous zones. The observation of a wide range of ultrastructure features in micrographs (Table 2) supports this proposal.
The molecular mass effects reported here probably represent a feature of the bacterial cellulose composite system that does not model the plant cell wall effectively. The molecular masses of xyloglucans isolated from either plant tissue or cell suspension cultures are typically in the range 100–200 kDa (Hayashi, 1989
; McCann et al., 1990
; Sims et al., 1996
) and are clearly associated with extensive cross-linking of cellulose within the plant cell wall (McCann et al., 1990
). In this study, xyloglucans of comparable molecular mass do not lead to extensive cross-linking. The lengths of cellulose cross-links formed by high molecular mass xyloglucans in composites are similar to those observed in depectinated onion cell walls (Whitney et al., 1995
; McCann et al., 1990
), suggesting that cross-link length is not the varying factor. One feature of the bacterial fermentation system that differs from the plant cell wall is the local concentration of polymers. In the fermentation system, it is likely that the architecture of the composite is defined primarily by the architecture of cellulose secretion sites (so-called terminal complexes) in conjunction with previously deposited cellulose, because there are no space restrictions in the liquid fermentation medium. In contrast, cell walls are likely to present a more concentrated polymer environment at least in part due to the presence of turgor pressure from adjacent cells. The molecular binding of xyloglucans to cellulose is essentially irreversible under both fermentation and cell wall conditions (Hayashi, 1989
; Whitney et al., 1995
), with solubilization of bound xyloglucan requiring alkaline treatment. We propose that initial binding of xyloglucan with cellulose occurs over relatively short segments (Vincken et al., 1995) with the possibility of a single xyloglucan chain binding with more than one cellulose fibril being increased by both xyloglucan molecular mass and fibril proximities. Once one xyloglucan chain is bound to two or more fibrils, we envisage that these fibrils are then tethered (i.e., kinetically trapped). We further propose that annealing of such cross-linked structures can occur to the extent that the lengths of cross-linking and bound segments of xyloglucan will equilibrate to the lowest free energy by balancing the enthalpy of binding with the entropy of flexible cross-links (Whitney et al., 1995
). An analogous process is envisaged for noncross-linked xyloglucan with an equilibration in terms of the lengths of loops and chains (Vincken et al., 1995). This proposed mechanism does not involve any extracellular biological effectors. Whilst this mechanism is likely to be valid for the bacterial fermentation, the cell wall contains a range of re-modeling agents such as xyloglucan endotransglycosylase and expansin, which may act on bound segments and cross-links, respectively, in xyloglucan–cellulose assemblies (Whitney et al., 2000
; Chanliaud et al., 2004
).
Galactose content has a greater effect on composites than other xyloglucan variants
The xyloglucan with the greatest difference in behavior in cellulose composites is a high molecular mass tamarind xyloglucan in which ca. 60% of starting galactose content was removed enzymatically. This modification results in the highest apparent level of incorporation of xyloglucan in composites (ca. 1 : 1 with cellulose; Table 1) yet cellulose I : Iß ratio is unchanged from pure cellulose, implying a lack of direct binding. Similarly, Fig. 3c shows evidence for signal intensity at 103.5 ppm corresponding to noncellulose-bound xyloglucan. Unlike other high molecular mass samples, there is no microscopic evidence for cross-linking of cellulose by xyloglucan. Instead, the composite microstructure is dominated by features that seem to arise from either cellulose (Fig. 7) or xyloglucan (Fig. 9) alone. Mechanical data (Fig. 10) show a low stress and strain to break for this composite with stiffness similar to cellulose alone. Taken together, data for this galactose-depleted xyloglucan is consistent with self-association of xyloglucan, leading to phase separation into xyloglucan-rich and cellulose-rich phases in the composite with limited direct binding between the two polymer types. The mechanical consequence is a composite that is probably cellulose continuous (from the stiffness) but fails readily on extension probably because of microscopic heterogeneity. Extrapolating to the cell wall, this combination of properties is unlikely to be beneficial to the plant. Indeed, a mutant of Arabidopsis thaliana (mur 3), that lacks the presence of a specific XG-specific galactosyl transferase, has markedly different mechanical and morphological properties (Pena et al., 2004
), whereas loss of fucose substituents through mutation of a fucosyl transferase (mur 2) does not have a mechanical phenotype (Ryden et al., 2003
; Pena et al., 2004
). Pena et al. (2004)
proposed that the mechanical phenotype of mur 3 is due to altered susceptibility to xyloglucan endotransglycosylase caused by either specificity of enzyme binding and activity or self-association of the xyloglucan, thereby inhibiting enzyme access. This work has suggested the alternative possibility that the cellulose–xyloglucan framework is compromised at low galactose levels because self-association of xyloglucan occurs in preference to binding to cellulose. The reported mechanical phenotype (Ryden et al., 2003
) would be consistent with our observations of weakened composite properties and lack of integration of galactose-depleted xyloglucan with cellulose in composites.
The self-association of galactose-depleted xyloglucan has been reported previously (Reid et al., 1988; Shirakawa et al., 1998
; Sims et al., 1998
), is often thermally reversible, and is considered to be analogous to the gelation of native tamarind xyloglucan in the presence of alcohol by random aggregation of "insolubilized" chain segments (Yamanaka et al., 2000
). This description is consistent with the observed "rounded" appearance of network pores in aggregated galactose-depleted xyloglucan (Fig. 9). The previous discussion concerns xyloglucan with 60% of galactose residues removed. For samples with intermediate galactose levels (25% and 35% removal), intermediate properties are observed. Thus both xyloglucan networking and cellulose cross-linking are observed at 35% depletion (Table 2). Mechanical properties at 25% and 35% depletion are similar to the 60%-depleted sample in that failure occurs at relatively low stress and strain values, again symptomatic of a heterogeneous microstructure. However, the stiffness of these composites follows that of composites of unmodified tamarind xyloglucan, suggesting that a cellulose-xyloglucan network forms the continuous phase in these two composites. The subtlety and diversity of properties as a function of galactose content are consistent with a potential role of this feature in modulating cellulose-xyloglucan networks in vivo. Previous studies of the effect of galactose content on binding of xyloglucans to isolated cellulose (Lima and Buckeridge, 2001
; Lima et al., 2004
) have also emphasized a role in controlling direct binding to cellulose, but have suggested that lower molecular masses lead to greater binding, the opposite of what has been found here. This different molecular mass effect most likely reflects a difference in the experimental systems. In vitro binding studies will favor binding at low molecular mass because there is little entropic penalty. However, in self-assembling networks, we propose that the initial binding topology is controlled by kinetic rather than thermodynamic factors.
Fucose substituents on xyloglucan may modulate aspects of composite formation
Earlier molecular modelers had suggested a key role for fucose substitution in controlling binding of xyloglucans to cellulose through a preference for formation of two-fold backbone conformations that are likely to be involved in steric binding to cellulose (Levy et al., 1991
). Subsequently, it was shown that a fucosylated xyloglucan bound to microcrystalline cellulose at twice the rate of a nonfucosylated xyloglucan (Levy et al., 1997
). However, earlier work with tamarind xyloglucan composites with cellulose (Whitney et al., 1995
) showed that fucosylation was not required for formation of cross-linked networks involving direct binding of xyloglucan to cellulose. It seems likely that the template effect of a two-fold conformation of cellulose is sufficient to overcome any inherent solution conformational preferences of xyloglucans as influenced by their substitution pattern. The consistent evidence for molecular interaction between xyloglucans and cellulose in this study provides further evidence for this proposal.
Recent analyses of Arabidopsis thaliana mutants deficient in fucosylated xyloglucan have shown a normal mechanical phenotype, again suggesting that fucosylation is not required for either mechanical performance or plant development (Ryden et al., 2003
; Pena et al., 2004
).
Several observations from this study do, however, point to a modulating effect of fucose substitution on properties of cellulose–xyloglucan composites. The fucosylated xyloglucan studied (P. communis; see Fig. 1) is similar in size to the lower molecular mass sample of nonfucosylated tamarind xyloglucan (Fig. 1), so comparisons between these two materials provides an index of potential roles for fucosylated residues. Compared to low molecular mass tamarind xyloglucan, P. communis xyloglucan had a higher level of incorporation within composites (0.32/0.36 vs. 0.20/0.24), a greater level of alignment of cellulose strands in the composite, and the presence of xyloglucan cross-links (after agitated fermentation). In these features, the behavior of P. communis xyloglucan is closer to that of high molecular mass tamarind xyloglucan and plant cell wall properties than is low molecular mass tamarind xyloglucan. Although these are not major effects (compared with galactosylation), this study does provide some evidence for a secondary role of fucosylation in promoting the formation of a cross-linked cellulose–xyloglucan network.
FOOTNOTES
1 The authors thank A. Darke for NMR spectra and J. Brigham (Unilever R&D Colworth) for electron micrographs. 
7 Author for correspondence (mike.gidley{at}uq.edu.au
)
101 6 Present address: The Babraham Institute, Babraham Research Campus, Cambridge, CB2 4AT, UK
LITERATURE CITED
Blakeney A. B. Harris P. J. Henry R. J. Stone B. A.. 1983. A simple and rapid preparation of alditol acetates for monosaccharide analysis. Carbohydrate Research 113: 291-299.[CrossRef][Web of Science]
Bourquin V. Nishikobo N. Abe H. Brumer H. Denman S. Eklund M. Christiernin M. Teeri T. T. Sundberg B. Mellerowicz E. J.. 2002. Xyloglucan endotransglycosylases have a function during the formation of secondary cell walls of vascular tissues. Plant Cell 14: 3073-3088.
Brigham J. E. Gidley M. J. Hoffmann R. A. Smith C. G.. 1994. Microscopic imaging of network strands in agar, carrageenan, locust bean gum and kappa carrageenan/locust bean gum gels. Food Hydrocolloids 8: 331-344.[Web of Science]
Chambat G. Karmous M. Costes M. Picard M. Joseleau J.-P. 2005. Variation of xyloglucan substitution pattern affects the sorption on celluloses with different degrees of crystallinity. Cellulose 12: 117-125.
Chanliaud E. Burrows K. M. Jeronimidis G. Gidley M. J.. 2002. Mechanical properties of primary plant cell wall analogues. Planta 215: 989-996.[CrossRef][Web of Science][Medline]
Chanliaud E. DeSilva J. Strongitharm B. Jeronimidis G. Gidley M. J.. 2004. Mechanical effects of plant cell wall enzymes on cellulose / xyloglucan composites. Plant Journal 38: 27-37.[CrossRef][Web of Science][Medline]
Chanliaud E. Gidley M. J.. 1999. In vitro synthesis and properties of pectin/Acetobacter xylinus cellulose composites. Plant Journal 20: 25-35.[CrossRef][Web of Science][Medline]
Debzi E. M. Chanzy H. Sugiyama J. Tekely P. Excoffier G.. 1991. The IIß transformation of highly crystalline cellulose by annealing in various mediums. Macromolecules 24: 6816-6822.[CrossRef][Web of Science]
Edwards M. Bowman Y. J. L. Dea I. C. M. Reid J. S. G.. 1988. A ß-galactosidase from nasturtium (Tropaeolum majus L) cotyledons. Purification, properties and demonstration that xyloglucan is the natural substrate. Journal of Biological Chemistry 263: 4333-4337.
Foster T. J. Ablett S. McCann M. C. Gidley M. J.. 1996. Mobility-resolved 13C NMR spectroscopy of primary plant cell walls. Biopolymers 39: 51-66.
Gidley M. J. Lillford P. J. Rowlands D. W. Lang P. Dentini M. Crescenzi V. Edwards M. Fanutti C. Reid J. S. G.. 1991. Structure and solution properties of tamarind-seed polysaccharide. Carbohydrate Research 214: 299-314.[CrossRef][Web of Science][Medline]
Haigler C. H. Brown Jr. M. Benziman M.. 1980. Calcofluor white alters the in vivo assembly of cellulose microfibrils. Science 210: 903-906.
Hayashi T.. 1989. Xyloglucans in the primary cell wall. Annual Review of Plant Physiology 40: 136-168.
Hestrin S. Schramm M.. 1954. Synthesis of cellulose by Acetobacter xylinum. Biochemical Journal 58: 345-352.[Medline]
Iraki N. M. Bressan R. A. Hasegawa P. M. Carpita N. C.. 1989. Alteration of the physical and chemical structure of the primary cell wall of growth-limited plant cells adapted to osmotic stress. Plant Physiology 91: 39-47.
Jarvis M. C. McCann M. C.. 2000. Macromolecular biophysics of the plant cell wall: concepts and methodology. Plant Physiology and Biochemistry 38: 1-13.
Levy S. Maclachlan G. Staehelin L. A.. 1997. Xyloglucan sidechains modulate binding to cellulose during in vitro binding assays as predicted by conformational dynamics simulations. Plant Journal 11: 373-386.[CrossRef][Web of Science][Medline]
Levy S. York W. S. Stuike-Prill R. Meyer B. Staehelin L. A.. 1991. Simulations of the static and dynamic molecular conformations of xyloglucan: the role of the fucosylated side-chain in surface-specific side-chain folding. Plant Journal 1: 195-215.[Web of Science][Medline]
Lima D. U. Buckeridge M. S.. 2001. Interaction between cellulose and storage xyloglucans: the influence of the degree of galactosylation. Carbohydrate Polymers 46: 157-163.[CrossRef][Web of Science]
Lima D. U. Loh W. Buckeridge M. S.. 2004. Xyloglucan-cellulose interaction depends on the sidechains and molecular weight of xyloglucan. Plant Physiology and Biochemistry 42: 389-394.[CrossRef][Web of Science][Medline]
McCann M. C. Wells B. Roberts K.. 1990. Direct visualisation of cross-links in the primary plant cell wall. Journal of Cell Science 96: 323-334.
Pena M. Ryden P. Madson M. Smith A. C. Carpita N. C.. 2004. The galactose residues of xyloglucan are essential to maintain mechanical strength of the primary cell walls in Arabidopsis during growth. Plant Physiology 134: 443-451.
Ryden P. Sugimoto-Shirasu K. Smith A. C. Findlay K. Reiter W.-D. McCann M. C.. 2003. Tensile properties of Arabidopsis cell walls depend on both a xyloglucan cross-linked microfibrillar network and rhamnogalacturonan II-borate complexes. Plant Physiology 132: 1033-1040.
Shirakawa M. Yamatoya K. Nishinari K.. 1998. Tailoring of xyloglucan properties using an enzyme. Food Hydrocolloids 12: 25-28.
Sims I. M. Bacic A.. 1995. Extracellular polysaccharides from suspension cultures of Nicotiana plumbaginifolia. Phytochemistry 38: 1397-1405.[CrossRef][Web of Science]
Sims I. M. Gane A. M. Dunstan D. Allan G. C. Boger D. V. Melton L. D. Bacic A.. 1998. Rheological properties of xyloglucans from different plant species. Carbohydrate Polymers 37: 61-69.[CrossRef][Web of Science]
Sims I. M. Munro S. L. A. Currie G. Craik D. Bacic A.. 1996. Structural characterisation of xyloglucan secreted by suspension-cultured cells of Nicotiana plumbaginifolia. Carbohydrate Research 293: 147-172.[CrossRef][Web of Science][Medline]
Somerville C. Bauer S. Brininstool G. Facette M. Hamann T. Milne J. Osborne E. Paredez A. Persson S. Raab T. Vorwek S. Youngs H.. 2004. Towards a systems approach to understanding plant cell walls. Science 306: 2206-2211.
Sorensen S. O. Pauly M. Bush M. Skjot M. McCann M. C. Borkhardt B. Ulskov P.. 2000. Pectin engineering: modification of potato pectin by in vivo expression of an endo-1,4-beta-d-galactanase. Proceedings of the National Academy of Sciences, USA 97: 7639-7644.
Takeda T. Furuta Y. Awano T. Mizuno Y. Mitsuishi Y. Hayashi T.. 2002. Suppression and acceleration of cell elongation by integration of xyloglucans in pea stem segments. Proceedings of the National Academy of Sciences, USA 99: 9055-9060.
Touzel J.-P. Chabbert B. Monties B. Debeire P. Cathala B.. 2003. Synthesis and characterisation of dehydrogenation polymers in Gluconacetobacter xylinus cellulose and cellulose/pectin composite. Journal of Agricultural and Food Chemistry 51: 981-986.[CrossRef][Web of Science][Medline]
Turner S. Taylor N. Jones L.. 2001. Mutations of the secondary cell wall. Plant Molecular Biology 47: 209-219.[CrossRef][Web of Science][Medline]
Vincken J.-P. York W. S. Beldman G. Voragen A. G. J.. 1997. Two general branching patterns of xyloglucan, XXXG and XXGG. Plant Physiology 114: 9-13.[CrossRef][Web of Science][Medline]
Watt D. K. Brasch D. J. Larsen D. S. Melton L. D.. 1999. Isolation, characterisation, and NMR study of xyloglucan from enzymatically depectinised and non-depectinised apple pomace. Carbohydrate Polymers 39: 165-180.
Whitney S. E. C. Brigham J. E. Darke A. Reid J. S. G. Gidley M. J.. 1995. In vitro assembly of cellulose/xyloglucan networks: ultrastructural and molecular aspects. Plant Journal 8: 491-504.
Whitney S. E. C. Brigham J. E. Darke A. Reid J. S. G. Gidley M. J.. 1998. Structural aspects of the interaction of mannan-based polysaccharides with bacterial cellulose. Carbohydrate Research 307: 299-309.[CrossRef][Web of Science]
Whitney S. E. C. Gidley M. J. McQueen-Mason S. J.. 2000. Probing expansin action using cellulose/hemicellulose composites. Plant Journal 22: 327-334.[CrossRef][Web of Science][Medline]
Whitney S. E. C. Gothard M. G. E. Mitchell J. T. Gidley M. J.. 1999. Roles of cellulose and xyloglucan in determining the mechanical properties of plant cell walls. Plant Physiology 121: 657-663.
Willats W. G. T. Orfilla C. Limberg G. Bucholt H. Van Alebeek G. J. W. M. Voragen A. G. J. Marcus S. E. Christensen T. M. I. E. Mikkelsen J. D. Murray B. S. Knox J. P.. 2001. Modulation of the degree and pattern of methyl-esterification of pectic homogalacturonan in plant cell walls: implications for pectin methyl esterase action, matrix properties and cell adhesion. Journal of Biological Chemistry 276: 19404-19413.
Yamanaka S. Yuguchi Y. Urakawa H. Kajiwara K. Shirakawa M. Yamatoya K.. 2000. Gelation of tamarind seed xyloglucan in the presence of ethanol. Food Hydrocolloids 14: 125-128.[CrossRef][Web of Science]
Yamomoto H. Horii F.. 1993. CP/MAS 13C NMR analysis of the crystal transformation induced for Valonia cellulose by annealing at high temperatures. Macromolecules 26: 131-1317.
York W. S. Kolli V. S. K. Orlando R. Albersheim P. Darvill A. G.. 1996. The structures of arabinoxyloglucans produced by solanaceous plants. Carbohydrate Research 285: 99-128.[Web of Science][Medline]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Schroder, R. G. Atkinson, and R. J. Redgwell Re-interpreting the role of endo-{beta}-mannanases as mannan endotransglycosylase/hydrolases in the plant cell wall Ann. Bot., August 1, 2009; 104(2): 197 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Niklas, H.-C. Spatz, and J. Vincent Plant biomechanics: an overview and prospectus Am. J. Botany, October 1, 2006; 93(10): 1369 - 1378. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
