* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Physico-chemical characteristics of cell walls from Arabidopsis
Survey
Document related concepts
Tissue engineering wikipedia , lookup
Cell membrane wikipedia , lookup
Signal transduction wikipedia , lookup
Biochemical switches in the cell cycle wikipedia , lookup
Cell encapsulation wikipedia , lookup
Endomembrane system wikipedia , lookup
Programmed cell death wikipedia , lookup
Cellular differentiation wikipedia , lookup
Extracellular matrix wikipedia , lookup
Cell growth wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
Cytokinesis wikipedia , lookup
Transcript
Journal of Experimental Botany, Vol. 55, No. 405, pp. 2087–2097, September 2004 DOI: 10.1093/jxb/erh225 Advance Access publication 30 July, 2004 RESEARCH PAPER Physico-chemical characteristics of cell walls from Arabidopsis thaliana microcalli showing different adhesion strengths Edouard Leboeuf1,2, Séverine Thoiron2 and Marc Lahaye1,* 1 INRA-Unité de Recherches sur les Polysaccharides, leurs Organisations et Interactions, BP 71627, F-44316 Nantes, France 2 University of Nantes, Groupe de Physiologie et Pathologie Végétales, Faculté des Sciences et Techniques, 2 rue de la Houssinière, BP 92208, F-44322, Nantes cedex 3, France Received 17 September 2003; Accepted 8 June 2004 Abstract Changes in the composition and structure of cell walls and extracellular polysaccharides (ECP) were studied during the growth of suspension-cultured Arabidopsis thaliana microcalli. Three growth phases, namely the cell division phase, the cell expansion phase, and the stationary phase, were distinguished and associated with a decreasing cell cluster adhesion strength. Degradation of the homogalacturonan pectic backbone and of linear pectic side chains (1,4)-b-D-galactan were observed concomitantly with the cell expansion and stationary phases and the decrease in cell adhesion. Also, in the stationary phase, branched (1,5)-a-L-arabinans were linearized. The AGP content of the culture medium increased while it decreased in the cell wall during cell growth and as cell adhesion decreased. These data suggest that, in addition to homogalacturonan, pectic side chains and AGP are involved in plant cell development and particularly in cell–cell attachment. Key words: AGP, Arabidopsis thaliana, (1,5)-a-L-arabinans, cell adhesion, (1,4)-b-D-galactans, homogalacturonan. Introduction Cell adhesion and cell separation are important in the morphogenesis and development of higher plants (Jarvis et al., 2003). The adhered cells of young tissues give rise to more loosely attached cells in mature structures during abscission, dehiscence, and other development stages inducing cell separation processes. A common feature of all cell separation processes is that the cell wall is degraded (Roberts et al., 2002). Plant cell walls are composed primarily of cellulose microfibrils, hemicelluloses, pectic polysaccharides, and small amounts of structural proteins (Carpita and Gibeaut, 1993). Pectins, the major polysaccharides of the middle lamella and the primary cell wall, are thought to be involved in intercellular attachment. Typically, pectins are composed of homogalacturonan (HG) and rhamnogalacturonan I (RGI) (Ridley et al., 2001; Vincken et al., 2003). HG is a polymer of (1,4)-a-linked galacturonic acid residues partially methyl-esterified and/or acetylated. It can be further decorated by b-D-xylosyl residues linked to O-3 of galacturonic acid giving xylogalacturonan (XGA) or by structurally complex sides chains composed of 11 different types of glycosyl residues giving type II rhamnogalacturonan (RGII) (Ridley et al., 2001; Vincken et al., 2003). RGI has a backbone of alternating galacturonic acid and rhamnose residues with 20–80% of the rhamnose residues substituted with side chains of (1,4)-b-D-galactans, (1,5)a-L-arabinans, or arabinogalactans (Ridley et al., 2001; Vincken et al., 2003). It has recently been proposed that HG, XGA, and RGII are other types of side chains of RGI (Vincken et al., 2003). Most studies have focused on the role of HG in cell adhesion, since it is the major component of the middle lamella which constitutes a cementing structure between adjacent cells. Indeed, unsubstituted HG forms in vitro three-dimensional networks through calcium ions-mediated junction zones and it is thought that it forms similar interactions in the middle lamella (Jarvis et al., 2003). However, other unknown covalent linkages are inferred *To whom correspondence should be addressed. Fax: +33 2 40 67 50 66. E-mail: [email protected] Journal of Experimental Botany, Vol. 55, No. 405, ª Society for Experimental Biology 2004; all rights reserved 2088 Leboeuf et al. from studies on the effect of pectin extractions on cell separations (reviewed by Jarvis et al., 2003). Besides HG, other macromolecular components and types of interaction in the primary cell wall are also likely to be involved. These include galactan, arabinan, and/or arabinogalactan side chains of RGI (Kikuchi et al., 1996; Iwai et al., 2001; Wallner and Nevins, 1974), RGII and their borate ester complexes (Iwai et al., 2002) and, probably, proteins (Jarvis et al., 2003). Within the scope of understanding the relationships between cell wall polymer structures and organizations with plant cell adhesion, modifications occurring in the cell wall physico-chemistry and cell cluster fragmentation during growth of A. thaliana microcalli suspension culture are reported here. Materials and methods Arabidopsis thaliana microcalli suspension cultures Chlorophyllian Arabidopsis thaliana (Columbia ecotype) microcalli suspension cultures, cell line T87, were grown according to Axelos et al. (1992). JPL medium was replaced by Gamborg’s B5 medium (Sigma-Aldrich, Saint Quentin Fallavier, France). Determination of cell growth parameters Cell number was counted on Malassey’s cell after a 16 h enzymatic maceration at 25 8C of cell aggregates with 0.25% Macerozyme R-10 (Duchefa Biochemie BV, Kalys, Roubaix, France) and 1% Cellulase Onozuka R-10 (Duchefa Biochemie BV) dissolved in 0.5 M mannitol. The fresh weight was measured after harvesting cells by filtration on a filter paper (Macherey-Nagel, Baeckerroot Labo, Jacou, France) under mild pressure. The dry weight of harvested cells was determined after 24 h at 60 8C. Determination of cell adhesion The cell adhesion of microcalli was determined by the ‘vortexinduced cell separation’ method adapted from Parker and Waldron (1995). Cell cluster size was measured with a particle size analyser (Micromeritics, Saturn DigiSizer 5200, Creil, France) before and after dissociation of cell clusters with a vortex shaker (Top-Mix II, Fisher Bioblock Scientific, Illkirch, France) for 2, 5, and 8 min at 40 Hz. Absorbance of the cell suspension supernatant between 400 and 800 nm was measured by spectrophotometry after centrifugation at 1000 g for 10 min. Preparation of alcohol-insoluble material (AIM), extracellular medium, and extracellular polymer (ECP) Cells were harvested by filtration through a 10 lm stainless-steel mesh. AIM was prepared from about 10 g of cells according to Bouton et al. (2002). The cell-free filtrates were freeze-dried and corresponded to the extracellular medium. The extracellular medium was redissolved in deionized water (1 g l1) and extracellular polymers were precipitated by adding ethanol to reach a final concentration of 80%. The precipitate was collected by centrifugation (10 000 g, 10 min), dissolved in deionized water, dialysed against deionized water, and freeze-dried. Chemical analyses Identification and quantification of neutral sugars by gas–liquid chromotography (GC) was performed after sulphuric acid degradation (Hoebler et al., 1989). Insoluble samples were dispersed in 72% sulphuric acid for 30 min at 25 8C prior to dilution to 1 M and hydrolysis (2 h, 100 8C). The soluble samples were directly dispersed in 1 M sulphuric acid solution and hydrolysed (2 h, 100 8C). For calibration, standard monosaccharide solutions and inositol as the internal standard were used. Starch content was determined by GC according to Englyst and Cummings (1988). Uronic acids were quantified by colorimetry using the automated m-phenylphenol-sulphuric acid method (Thibault, 1979). The nitrogen content of AIM was performed using the Kjeldahl method (Miller and Miller, 1948). The amount of protein was deduced using a conversion factor of 6.25. The degree of methyl- and acetyl-esterification of pectins was determined by HPLC following saponification according to Levigne et al. (2002). Detection of arabinogalactan-proteins in AIM and ECP AIM and ECP were suspended at room temperature in 0.15 M NaCl solution (40 mg ml1 and 10 mg ml1, respectively) for 1 h with shaking. The preparations were assayed for arabinogalactan-proteins by the b-glycosyl-Yariv reagent (Biosupplies, Parkville Victoria, Australia) using the single radial gel diffusion method (van Holst and Clarke, 1985). Enzymatic degradations of AIM and ECP AIM or ECP were degraded by endo-(1,5)-a-L-arabinanase, a-Larabinofuranosidase, endo-(1,4)-b-D-galactanase, endo-polygalacturonase (Megazyme, Bray, Ireland), and b-D-galactosidase (Sigma-Aldrich). AIM and ECP were suspended (10 mg dry weight in 1.5 ml) in 100 mM sodium acetate (pH 4) containing 0.02% thimerosal. Enzyme (0.2 U) was added or not (blank) and the suspensions were incubated for 40 h for AIM or 16 h for ECP at 30 8C with shaking. Enzyme (0.2 U) was added again after 24 h of digestion for AIM. After incubation, the suspensions were centrifuged (5000 g, 10 min) and supernatants were filtered through 0.45 lm filters (Millex-HV; Millipore, St Quentinen-Yvelines, France) and boiled for 5 min. For ECP, ethanol was added to the boiled supernatant to reach a final concentration of 80% ethanol and centrifuged (5000 g, 10 min) to recover enzymatically produced oligomers. Uronic acids and neutral sugars contents of the hydrolysates were measured by colorimetry and GC, respectively. Nuclear magnetic resonance analysis 13 C-NMR spectra of ECP soluble fraction were recorded on a Bruker ARX 400 (Bruker, Wissembourg, France) at 100.57 MHz. ECP were dissolved in water (10 mg ml1) for 16 h at 4 8C. The solution was centrifuged (5000 g, 10 min) and the supernatant freezedried prior to solubilization in D2O (Sigma-Aldrich). Chemical shifts were measured from internal acetone assigned to 31.4 ppm. AIM was extracted by 1% potassium oxalate (1 g in 150 ml) at 25 8C for 1 h with agitation. Residues were recovered by centrifugation (9000 g, 10 min) and supernatants were passed through a sintered glass filter G4 (5–15 lm porosity). Residues were extracted twice more as above and the final residues were washed with water. Starch from the residues was removed by treatment with amyloglucosidase (A7420, Sigma-Aldrich) and residues were dehydrated by 958 ethanol and acetone washes prior to drying at 40 8C. Solid-state magic angle spinning 13C-NMR spectra of the rehydrated residues with D2O were recorded using liquid-like conditions on a Bruker DMX 400 spectrometer operating at carbon frequency of 100.62 MHz (Lahaye et al., 2003). Water-holding capacity of cell walls The water-holding capacity of AIM was determined by the centrifugation method adapted from McConnell et al. (1974). About 150 mg Adhesion strength of Arabidopsis microcalli cells 2089 of AIM were suspended in deionized water (10 ml, 1 h, room temperature) with shaking before centrifugation at 14 000 g for 20 min. The supernatant was discarded and the pellet drained on a sintered glass filter G3 (15–40 lm porosity) before weighing. The weight difference between the wet and dry AIM is ascribed to the water-holding capacity and expressed as g of water retained g1 of dry AIM. Enzyme extraction and assay Culture medium and cell-surface enzymes were extracted according to a method adapted from Wallner and Nevins (1974). Cell-surface enzymes were extracted by shaking cells in cold 1 M sodium citrate buffer pH 5.4 (1 ml g1) for 1 h. Enzyme activity was detected by measuring the release of pnitrophenol from b-D-galactoside/a-L-arabinofuranoside substrates (Sigma–Aldrich) according to Rouau and Odier (1986) or reducing sugars from polysaccharides (laboratory-prepared sugar beet (1,5)-aL-arabinans, lupin (1,4)-b-D-galactans, and commercial polygalacturonic acid and gum arabic (Sigma–Aldrich)) according to Nelson (1944). Caseinolytic activity was assayed as described by Tsumaraya et al. (1990). HPAEC of extracellular medium Freeze-dried extracellular media were solubilized in water (10 mg in 1.5 ml) for 16 h at 4 8C. The solutions (20 ll) were injected on a Dionex analytical Carbopac PA-1 column (Dionex, Sunnyvale, CA ; 43250 mm) equipped with a Carbopac PA-1 guard column (Dionex; 4350 mm) using alkaline conditions according to Bouton et al. (2002). Results Cell growth and cell adhesion Although not synchronized, the growth of suspensioncultured Arabidopsis cells (Fig. 1) showed three reproducible phases, namely a cell division phase (days 0–6), a cell expansion phase (days 6–12) and a stationary phase (after day 12). During the cell division phase, the cell number increased about 8-fold with only a 3-fold increase in fresh weight. The cell expansion phase was characterized by a 2-fold increase in fresh weight corresponding with water uptake while the cell number did not change. After 12 d of culture, cell growth stopped as already described by Axelos et al. (1992) and probably corresponded with cell senescence (Callard et al., 1996). The change in cell adhesion at the different growth stages was observed by cell cluster size measurement before and after a mild mechanical treatment (vortex shaking) for 2, 5, and 8 min. There was no change in turbidity or in absorption maxima at 430, 490, and 680 nm for chlorophyll b, carotenoid, and chlorophyll a absorption, respectively, demonstrating that the cell integrity was conserved after vortexing (data not shown). The size distribution of cell clusters was essentially monomodal and gaussian. The cell aggregates were grouped into small (<250 lm), middle (250–500 lm), and big cell clusters (>500 lm). Before vortexing, the proportion of cell clusters in the three groups was similar for cells in the division and expansion stages (Fig. 2). By contrast, cell clusters in the Fig. 1. Growth curves of Arabidopsis thaliana microcalli suspension culture and definition of the division, expansion, and stationary growth phases. Lines are drawn to help reader’s eyes; bars indicate SD (n=5). stationary phase were more fragmented (Fig. 2). This fragmentation possibly resulted from the shaking conditions of the culture and indicated that cells in the stationary phase were less adherent than cells in division or expansion. After 2, 5 and 8 min of vortexing, cell aggregates in the expansion phase were richer in small clusters and poorer in large clusters, than cell aggregates in the division phase (Fig. 2). Thus, expanding cells are less adherent than dividing cells. Cell aggregates in the stationary phase were richer in small clusters and poorer in big clusters than those of dividing and expanding cells (Fig. 2). These results suggest that cell adhesion decreases from the cell division phase to the stationary phase in this Arabidopsis cellculture system. Cell wall polysaccharides composition and structure The alcohol-insoluble material (AIM) composition of the cells at the different growth phases indicated the presence of uronic acids, glucose, galactose, arabinose, xylose, fucose, mannose attributed to cell wall polysaccharides, and to starch (glucose; Fig. 3). It also contained proteins and uncharacterized compounds (Fig. 3A). The proportion of cell wall sugars increased during the cell expansion phase and accounted for about 30–50% of the AIM while that of starch decreased. Proteins and uncharacterized material remained constant at about 30% and 15% of dry weight, respectively. The relative amount of cell wall sugars in the AIM of cells in the division, expansion and stationary phases is shown in Fig. 3B, C, and D. Glucose and uronic acid were the major sugars (Fig. 3B) while arabinose, galactose, and xylose represented about 20–30 mol% of all sugars (Fig. 3C, D). Rhamnose, fucose, and mannose were minor sugars amounting to about 5 mol% of the total sugars. There were no significant changes in the proportions of uronic acid, (non-starch) glucose, rhamnose, xylose, fucose, and mannose along the cell culture. By contrast, the proportion of 2090 Leboeuf et al. Fig. 2. Proportion of dividing, expanding and stationary cell aggregates of size <250 lm, between 250 and 500 lm and >500 lm before (0 min) and after (2, 5, 8 min) vortex shaking. Results are expressed as a percentage of the total cell cluster population. Bars indicate SD (n=6). Fig. 3. Chemical composition of the alcohol-insoluble materials from cell aggregates in the dividing, expanding, and stationary phases. (A) Overall dry weight composition; (B, C, D) molar proportion of wall sugars. (B) Uronic acid and glucose, (C) rhamnose, galactose, and arabinose; (D) fucose, xylose, and mannose. Bars indicate SD (n=5). galactose increased about 1.5-fold during the cell division phase and then gradually decreased to reach a minimum in the stationary phase. The proportion of arabinose increased 2-fold during the cell division phase, remained stable during the cell expansion phase, and then decreased 2-fold in the stationary phase. Expressed on the uronic acid content of the AIM, the pectin degree of methyl-esterification varied from 30% to 40% and tended to be higher in the 4d 2269 1467 35612 4367 2266 1362 30622 37611 1264 964 1663 3265 74622 666 – – 7769 1869 – – 7362 2667 – – – – – – – – – – – – – – – 663 – – – 1765 – – – 2464 – – 5066 – – – 52620 – – – 53611 – – – 2566 – – – 1263 – – – 964 – – – Ara Gal Rha UA 14 d 14 d 8d 4d 14 d 8d 4d 14 d 8d 14 d 8d 4d 4d 8d 14 d 4d 8d Polygalacturonase Arabinanase+ arabinofuranosidase+ galactanase Galactosidase Galactanase Arabinofuranosidase Arabinanase Sugars Results were expressed as weight percentages of invididual sugars content in AIM;6 indicate SD (n=3). Table 1. Release of arabinose (Ara), galactose (Gal), rhamnose (Rha), and uronic acid (UA) from the alcohol-insoluble material of dividing, expanding, and stationary cells collected after 4, 8, and 14 d of culture, respectively, by endo-(1,5)-a-L-arabinanase,a -L-arabinofuranosidase, endo-(1,4)-b-D-galactanase, b-galactosidase, and endopolygalacturonase Adhesion strength of Arabidopsis microcalli cells 2091 cell division phase, while its degree of acetylation increased from 20% to 30% during the cell division phase and then gradually decreased during the cell expansion phase to reach a constant value of 20% in the stationary phase (data not shown). To explore the structure of arabinose- and galactosecontaining polysaccharides, alcohol-insoluble material of cells grown for 4, 8, and 14 d were treated with endo(1,5)-a-L-arabinanase, a-L-arabinofuranosidase, endo-(1,4)b-D-galactanase, and b-D-galactosidase (Table 1). The endo-a-L-arabinanase released about 2.5-fold more of the total wall arabinose at day 14 than at days 4 and 8 while the percentage of total wall galactose released by the endo(1,4)-b-D-galactanase gradually decreased during the cell culture. The a-L-arabinofuranosidase released about 50% of total wall arabinose at the three growth phases, but bD-galactosidase released no galactose. In order to test whether these enzymes could act synergistically on the degradation of the arabinose- and galactose-containing polysaccharides, endo-(1,5)-a-L-arabinanase, a-L-arabinofuranosidase, and endo-(1,4)-b-D-galactanase were also used together (Table 1). The combined enzymes failed to release more galactose than that released by the endo-(1,4)-b-D-galactanase alone. By contrast, they increased the amount of arabinose released from cells grown for 4 d and 8 d by about 10%, while no increase was obtained from cells grown for 14 d. Altogether, these results indicate that (1,4)-b-D-galactans are present in the cell wall material and are unlikely to be ramified. Furthermore, the gradual decrease in galactose content and the decrease in galactose released by enzymes as the cell culture progressed to the stationary phase indicate that pectic galactans side chains are metabolized and/or diluted by the synthesis of RGI poor in galactan side chains. Similarly, these results indicate that (1,5)-linked arabinans are present in these cell walls, but are more branched in dividing and expanding cells than in nongrowing ones. The decrease in the amount of arabinose in non-growing cells, the increase in arabinose released by the endo-(1,5)-a-L-arabinanase alone from these cells, and the inefficiency of the arabinofuranosidase to release more arabinose when used in combination with the endoarabinanase suggest that arabinose branches on (1,5)-a-Larabinan are removed as the cell growth stops and/or are diluted by the synthesis of RGI substituted by few linear arabinan side chains. The proportion of pectic arabinans (linear (1,5)-linked a-L-arabinan) and galactans to the total arabinose and galactose in the wall is low (about 25%, see Table 1) and suggests the presence of other arabinose/ galactose containing polymers in these cell walls. The inefficiency of the combined enzymes endo-(1,5)-a-Larabinanase, a-L-arabinofuranosidase, and endo-(1,4)b-D-galactanase in releasing more galactose than endo-(1,4)-b-D-galactanase alone and the positive reaction of aqueous suspensions of these different AIM with bglycosyl-Yariv reagent (data not shown) demonstrated the 2092 Leboeuf et al. presence of arabinogalactan proteins (AGP) in these materials. These cell wall preparations were also treated with an endo-polygalacturonase to follow modifications of the galacturonan pectic backbone during the different cellgrowth phases. The amount of uronic acid released by the endo-polygalacturonase was higher from cells grown for 8 d and 14 d than for 4 d (Table 1). As this endopolygalacturonase cleaves blocks of unsubstituted galacturonic acids, the difference in degradation could reflect the lower methyl- and acetyl-ester substitution of pectins in expanding and non-growing cells than in dividing cells. The enzyme released similar amounts of RGI from the cells at the three growth phases as shown by the close molar proportions of rhamnose and uronic acids recovered in the hydrolysate (Table 2). However, the neutral side chains differ, since the arabinose proportion was higher in the expanding cells and markedly lower in non-growing cells. These results agree with wall sugar composition. To complement the structural analysis of cell wall polysaccharides, the solid-state 13C NMR spectrum was determined with liquid-like magic angle spinning acquisition conditions (HR-MAS 13C NMR) of the alcohol-insoluble material from dividing, expanding and non-growing cells (Fig. 4A). This technique allows the observation of highly mobile polysaccharides or polysaccharides segments such as RGI side chains within the cell wall matrix (Foster et al., 1996; Ha et al., 1996; Rondeau-Mouro et al., 2003). Prior to observation, the different AIMs were enriched for tightly linked cell wall materials by oxalate extraction and were then Table 2. Mole percentage of rhamnose (Rha), arabinose (Ara), galactose (Gal), and uronic acids (UA) released by endopolygalacturonase from the alcohol-insoluble material of dividing, expanding, and stationary cells collected after 4, 8, and 14 d of culture, respectively;6 indicate SD (n=3) Sugars Rha Ara Gal UA Culture age (d) 4 8 14 4.460.5 12.561.6 8.361.6 74.863.5 5.362.5 18.563.8 9.561.8 66.867.9 6.261.3 9.461.2 7.861.1 76.663.4 Fig. 4. 13C NMR spectra of cell walls (A) and extracellular polysaccharides in the culture medium (B). (A) Solid-state HR-MAS 13C NMR spectra of oxalate-treated AIM of cells in division, expansion, and stationary phases and spectrum of the empty rotor, and (B) 13C NMR spectra of solutions of native and a-L-arabinofuranosidase-treated extracellular polysaccharides. Letters correspond to signal in terminal arabinofuranoside (At), (1,4)-linked bD-galactan (G), (1,4)-linked a-D-galacturonan (P), and methyl ester (Me); C-1 to 6 correspond to carbon 1 to 6 in terminally-linked galactose or arabinose (t-Gal, t-Ara, respectively), 3-, 6-, and 3,6-linked galactose (3-Gal, 6-Gal, 3,6-Gal, respectively); signals referred to as Gal have chemical shifts similar to those of galactose in Latrix sibirica arabinogalactan (Karacsonyi et al., 1984). Adhesion strength of Arabidopsis microcalli cells enzymatically de-starched. The oxalate extract was essentially composed of uronic acids, galactose, arabinose, and rhamnose that represented 10–30% of the corresponding total wall sugar (data not shown) which probably corresponded to loosely interacting or ionically linked rhamnogalacturonans and homogalacturonans. The complex solid-state 13C NMR spectra obtained from the residues (Fig. 4A) showed major signals attributed to AGP (see below and Fig. 4B) and pectic polysaccharides (Ryden et al., 1989; Catoire et al., 1998). Signals for (1,4)-linked b-D-galactan were observed, but those for (1,5)-linked a-L-arabinan were not seen, presumably because of overlaps with those of AGP or losses during preparation of the residues (Fig. 4A). Other signals at about 20–40 ppm were attributed to proteins, at 171–174 ppm to carboxylic C of pectins (not shown), at 17.0 ppm to the internal hexamethyl benzene reference and C6 of rhamnose in RGI. The broad signal at 111 ppm was a contribution from the rotor and/or the probe. Many small signals could not be assigned, probably resulting in part from the structural complexity of AGPs and branched pectic side chains. Modifications in the signals intensity were observed between the spectra of dividing, expanding, and nongrowing cells. Compared with the signal intensity of the internal reference, there was an increased contribution of galacturonan signals in the expanding cell. There was a decrease in the signals for (1,4)-linked b-D-galactan and a marked decrease in the signals for terminal arabinofuranosyl residues in the non-growing cell. Because changes were observed in sugar composition, these modifications probably resulted from the decrease in the proportion of some cell wall components rather than a reduction in their mobility. Signals within the 20–40 ppm range attributed to proteins were also affected by the growth phases, as their intensity decreased from dividing to non-growing cells. Water retention capacity of AIMs The composition and structural analyses of AIM demonstrated modifications in pectic polysaccharides and AGP during the cell culture. Pectins are thought to be involved in the control of cell wall porosity (Willats et al., 2001) while 2093 the function of AGP are still unclear (Showalter, 2001) but can also play a role through their highly ramified carbohydrate structure on wall porosity and cell wall polymer interactions. To evaluate the consequence of the modifications of these polysaccharides contents and structure on the cell wall properties, the water-holding capacity of the different AIM was measured. The water-holding capacity increased during cell culture from 5.661.2 at 4 d to 10.261.5 at 8 d, and 13.660.1 at 14 d and suggested an increase in cell wall porosity. Extracellular polysaccharides Given the elimination of arabinose and galactose among the cell wall polysaccharides during cell culture, the presence of the corresponding degradation products in the culture medium was investigated. At day 4, the culture medium was mainly composed of glucose and mannose, that had probably originated from sucrose isomerization, while at days 8 and 14 these sugars were minor compared with the arabinose, galactose, and uronic acid contents (Table 3). No monomers and oligomers of arabinose and galactose were detected in the culture medium by HPAEC. Only glucose, fructose, and sucrose (peaks 1, 2, and 3, respectively, Fig. 5) were seen and their proportion decreased during the cell culture in agreement with the GC data. Peak 4 detected at day 8 and, more importantly, at day 14, was assigned to galacturonic acid. Extracellular polymers (ECP) were precipitated from the culture medium by ethanol and their yield on the basis of cell number (ng cell1) increased during the culture: 0.1760.07 at day 4, 0.3560.02 at day 8, and 0.4060.03 at day 14. Sugars accounted for 3166, 5162, and 4564% of ECP dry weight at days 4, 8, and 14, respectively, and were mainly composed of arabinose, galactose and uronic acids (Table 3). These sugars amounted to nearly all the arabinose, galactose, and uronic acids measured in the culture medium (data not shown) and thus belonged to high molecular weight compounds. ECP was degraded by a-L-arabinofuranosidase (6762% of total arabinose at day 8 and 6464% at day 14), but the addition of endo-(1,5)-a-L-arabinanase, Table 3. Sugar composition (mol%) of the extracellular medium and extracellular polymers of dividing, expanding, and stationary microcalli recovered after 4, 8, and 14 d of culture, respectively;6 indicate SD (n=3) Sugarsa Rha Fuc Ara Xyl Man Gal Glc UA a Extracellular medium Extracellular polymers 4d 8d 14 d 4d 8d 14 d 0.060.0 0.060.0 0.660.0 0.360.3 10.161.1 0.960.2 85.261.2 2.960.3 1.760.2 1.360.4 15.361.7 3.460.6 5.160.6 15.161.1 12.163.6 45.862.8 1.860.2 1.160.1 17.360.4 3.260.5 4.760.1 16.661.0 7.660.6 47.761.3 1.360.1 1.060.2 15.261.2 6.660.9 4.060.8 19.061.4 15.661.3 34.463.0 1.360.2 0.660.1 14.362.0 2.260.2 2.560.3 14.562.3 5.860.7 55.566.3 1.660.0 0.760.0 17.660.5 1.460.1 2.160.4 17.460.8 4.160.4 52.264.3 Rha, rhamnose; Fuc, fucose; Ara, arabinose; Xyl, xylose; Man, mannose; Gal, galactose; Glc, glucose; UA, uronic acids. 2094 Leboeuf et al. endo-(1,4)-b-D-galactanase, and b-D-galactosidase to the arabinofuranosidase had no effect. On solubilization of ECP for NMR analysis, a uronic acid-rich insoluble material precipitated leaving nearly all the neutral sugars in solution. The combined solutions of ECP recovered from the 8 d and 14 d cell cultures yielded a 13C NMR spectrum (Fig. 4B) with major signals attributed to type II arabinogalactans (Gane et al., 1995; Karacsonyi et al., 1984; Saulnier et al., 1992). These soluble fractions also yielded a reddish precipitate with the b-glycosyl-Yariv reagent. All these results indicated that ECP were made of type II arabinogalactans belonging to AGPs and partly of homogalacturonan. Activities of cell wall-degrading enzymes during cell culture Together with the structural changes of AIM polysaccharides during the cell growth phases, cell wall and culture medium enzymatic activities were measured (Table 4). (1,5)-a-L-Arabinanase, ‘AGPase’ (tested against gum arabic substrate) and protease activities (tested against azocasein substate) were not detected. Cell surface bgalactosidase activity was the only enzyme detected at day 4 and the level of this activity slightly increased during the cell culture. The culture medium also contained bgalactosidase activity at day 4, but this activity was higher at days 8 and 14 than at day 4. Cell surface and culture medium a-arabinofuranosidase activities were very low at days 4 and 8, but increased at day 14. There was no cell surface and culture medium (1,4)-b-D-galactanase activity at day 4. This activity appeared at day 8 and increased at day 14. Similarly, a low polygalacturonase activity at the cell surface and in the culture medium appeared at day 8 and markedly increased at day 14 (Table 4). Discussion Fig. 5. HPAEC chromatograms of the extracellular culture medium of Arabidopsis microcalli in division, expansion, and stationary phases. Peaks labelled Glc, Fru, Suc, and GalA were assigned on the basis of their retention time to glucose, fructose, sucrose, and galacturonic acid, respectively. In the present work, A. thaliana microcalli suspension culture was assessed as a model to approach the macromolecular basis of cell adhesion. Even though this cell culture was asynchronous, it presented three reproducible growth phases enriched in dividing, expanding, and nongrowing cells. These phases were associated with a decrease in cell cluster adhesion strength, modifications in the cell wall polysaccharides composition and structure, as well as in cell wall-degrading enzyme activities. The overall cell wall sugar composition of these microcalli is similar to that of A. thaliana seedlings primary wall (Zablackis et al., 1995; Bouton et al., 2002; Peng et al., 2000). However, major variations were seen in the galactose and arabinose contents which peak during the cell division phase and then decrease at different rates during the cell expansion and Table 4. Cell wall degrading enzyme activities at the cell surface and in the culture medium of dividing, expanding and stationary microcalli (4, 8, and 14 d, respectively);6 indicate SD (n=3) Enzymes a-L-arabinofuranosidase b-galactosidase (1,4)-b-D-galactanase Polygalacturonase Cell surface (nkat g1) Culture medium (nkat ml1) 4d 8d 14 d 4d 8d 14 d 0.0160.01 0.5160.04 0.0060.00 0.0060.00 0.0160.01 0.6060.11 1.3160.52 0.0360.04 0.2560.06 0.6760.07 3.4161.04 2.3160.05 0.0060.00 0.0260.00 0.0060.00 0.0060.00 0.0160.00 0.1060.01 0.1560.12 0.0260.02 0.0560.00 0.1260.03 0.9960.67 0.5760.38 Adhesion strength of Arabidopsis microcalli cells stationary phases. Part of these sugars was clearly shown to originate from linear (1,4)-b-D-galactans and branched (1,5)-a-L-arabinans belonging to RGI side chains. The loss of these galactans and arabinans was well supported by the appearance of extracellular (1,4)-b-D-galactanase and arabinofuranosidase activities and led to an increase in waterholding capacity of wall material taken as reflecting an increase in wall porosity. The decrease in b-D-galactan during the cell expansion phase agrees with results obtained on cell suspension cultures of Populus alba (Kakegawa et al., 2000). The b-D-galactosidase activity observed throughout the culture cycle and, particularly, peaking when the cells became less adherent and lost galactose, agrees with results reported from cell walls of Paul’s Scarlet Rose cell suspension cultures (Wallner and Nevins, 1974). All the enzymes reaction products were likely to be rapidly metabolized because only some galacturonic acid accumulated in the extracellular medium. AGP and HG were the main accumulating extracellular polymers in agreement with literature reports (Darjania et al., 2002; Takeuchi and Komamine, 1978). The decrease of pectic arabinan and galactan side chains found in the microcalli culture system used here can be related to both cell growth and/or adhesion. Willats et al. (1999) showed that the onset of pectic galactan side chains in the cell wall of proliferating cells in the carrot root apex or in carrot calli appeared after the cells were first conditioned for expansion. Similarly, galactan RGI side chains were observed at the transition zone near or at the onset of cell elongation in different tissues of Arabidopsis seedlings (McCartney et al., 2003). Willats et al. (1999) also showed that arabinan side chains appeared restricted to proliferating cells. The different rates of galactose and arabinose disappearance observed here in Arabidopsis microcalli cell walls, notably during the elongation phase, support modulations of RGI neutral side chains composition during cell development. However, the overwhelming contribution of AGP to the arabinose and galactose contents and cell asynchrony probably masked fine modifications of the pectic side chains. In particular, the de-branching of RGI arabinan observed during the stationary phase of the microcalli culture would be interesting to follow in planta with regard to development and cell differentiation. RGI and their neutral side chains have also been associated with adhesion and rigidity of primary walls. The molar ratio of arabinose to galactose in the pectic fraction of carrot callus cell walls has been correlated with the average size of cell clusters (Kikuchi et al., 1996) and probably reflected different cell adhesion. The presence of galactan side chains have often been related with cell wall rigidity and with firmer textures of tissues and organs (McCartney et al., 2000; McCartney and Knox, 2002; Smith et al., 2002). With regard to arabinan, low levels and abnormal deposition in cell walls have been found with large intercellular spaces in the pericarp of the Cnr tomato 2095 mutant (Orfila et al., 2001, 2002) and with loose cells in Nicotiana plumbaginifolia callus (Iwai et al., 2001). However, these observations are difficult to interpret because not only the RGI neutral side chains were affected but also the galacturonan backbone. In fact, recent mutants studies suggest that the polyuronan backbone of pectin may be more important in intercellular attachment than RGI side chains (Bouton et al., 2002; Iwai et al., 2002; Sorensen et al., 2000; Skjot et al., 2002; Oomen et al., 2002). The release of homogalacturonan in the microcalli culture medium as they became more fragmented supports their role in adhesion. Since RGI is located in the primary wall (Jarvis et al., 2003), its role in cell adhesion is likely to be indirect. It is expected to involve molecular entanglement of its neutral side chains and ionic interactions of the uronan backbone with other cell wall and middle lamella polymers. It is also possible that it be covalently linked, for example, with xyloglucans (Thompson and Fry, 2000). These linkages and interactions would rigidify the cell wall, anchor the middle lamella polymers to the primary wall, and control wall porosity, thus affecting the diffusion of enzymes. The proposition that the middle lamella HG represent side chains of peculiar RGI (Vincken et al., 2003) remains to be established, but would fit with the above conception. However, the bulk of galactose and arabinose affected by the different growth phase of these A. thaliana microcalli suspension culture was attributable to AGP. These polymers were found both in the culture medium as previously reported (Darjania et al., 2002), and tightly bonded in the cell wall where they are structurally modified according to the growth phase of the cells. These modifications and their accumulation in the extracellular medium suggest specific biological roles (Showalter, 2001), such as on cell growth (Kreuger and Van Holst, 1993; Toonen et al., 1997). Nevertheless, like RGI neutral side chains, physicochemical contributions can be expected of the arabinogalactan ramified structure of AGP to the cell wall mechanical properties and in the control of wall porosity. Furthermore, being associated or anchored to the plasma membrane (Showalter, 2001), with b-glycan binding properties (Baldwin et al., 1993; Pennell et al., 1989; Yariv et al., 1967), and with the ability to form cross-links (Kjellbom et al., 1997), AGP are likely to play a role in linking cell wall macromolecules to the cell membrane, and thus to be involved in cellular cohesion. Altogether, these results stress on the complex multimolecular aspect of cell adhesion involving polyuronans, RGI side chains, and AGP in the whole wall, from the plasma membrane to the middle lamella. In particular, it is shown that linearization of RGI arabinan side chains and modulation of AGP structure are also related to cell growth and separation processes. However, it is likely that other cell wall polymers are also involved in cell adhesion, such as structural wall proteins known to rigidify tissues and to be able to form complexes with pectins (Cassab, 1998; 2096 Leboeuf et al. MacDougall et al., 2001). Identification of Arabidopsis mutants specifically affected in the synthesis, structure, and location of these different wall polymers will help in elucidating their function on cell growth and adhesion. References Axelos M, Curie C, Mazzolini L, Bardet C, Lescure B. 1992. A protocol for transient gene expression in Arabidopsis thaliana protoplasts isolated from cell suspension cultures. Plant Physiology and Biochemistry 30, 123–128. Baldwin TC, McCann MC, Roberts K. 1993. A novel hydroxyproline-deficient arabinogalactan protein secreted by suspensioncultured cells of Daucus carota. Purification and partial characterization. Plant Physiology 103, 115–123. Bouton S, Leboeuf E, Mouille G, Leydecker MT, Talbotec J, Granier F, Lahaye M, Höfte H, Truong HN. 2002. Quasimodo1 encodes a putative membrane-bound glycosyltransferase required for normal pectin synthesis and cell adhesion in Arabidopsis. The Plant Cell 14, 2577–2590. Callard D, Axelos M, Mazzolini L. 1996. Novel molecular markers for late phases of the growth cycle of Arabidopsis thaliana cellsuspension cultures are expressed during organ senescence. Plant Physiology 112, 705–715. Carpita NC, Gibeaut DM. 1993. Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. The Plant Journal 3, 1–30. Cassab G. 1998. Plant cell wall proteins. Annual Review of Plant Physiology and Plant Molecular Biology 49, 281–309. Catoire L, Goldberg R, Pierron M, Morvan C, Herve du Penhoat C. 1998. An efficient procedure for studying pectin structure which combines limited depolymerization and 13C NMR. European Biophysics Journal 27, 127–136. Darjania L, Ichise N, Ichikawa S, Okamoto T, Okuyama H, Thompson Jr GA. 2002. Dynamic turnover of arabinogalactan proteins in cultured Arabidopsis cells. Plant Physiology and Biochemistry 40, 69–79. Englyst HN, Cummings JH. 1988. Improved method for measurement of dietary fiber as non-starch polysaccharides in plant foods. Journal of the Association of Official Analytical Chemists 71, 808–814. Foster TJ, Ablett S, McCann MC, Gidley MJ. 1996. Mobilityresolved 13C-NMR spectroscopy of primary plant cell wall. Biopolymer 39, 51–66. Gane AM, Craik D, Munro SLA, Howlett GJ, Clarke AE, Bacic A. 1995. Structural analysis of the carbohydrate moiety of arabinogalactan-proteins from stigmas and styles of Nicotiana alata. Carbohydrate Research 277, 67–85. Ha MA, Evans BW, Jarvis MC, Apperley DC, Kenwright AM. 1996. CP-MAS of highly mobile hydrated biopolymers: polysaccharides of Allium cell walls. Carbohydrate Research 288, 15–23. Hoebler C, Barry JL, David A, Delort-Laval J. 1989. Rapid acid hydrolysis of plant cell wall polysaccharides and simplified quantitative determination of their neutral monosaccharides by gas–liquid chromatography. Journal of Agricultural and Food Chemistry 37, 360–367. Iwai H, Ishii T, Satoh S. 2001. Absence of arabinan in the side chains of pectic polysaccharides strongly associated with cell walls of Nicotiana plumbaginifolia non-organogenic callus with loosely attached constituent cells. Planta 213, 907–915. Iwai H, Masaoka N, Ishii T, Satoh S. 2002. A pectin glucuronyltransferase gene is essential for intercellular attachment in the plant meristem. Proceedings of the National Academy of Sciences, USA 99, 16319–16324. Jarvis MC, Briggs SPH, Knox JP. 2003. Intercellular adhesion and cell separation in plants. Plant, Cell and Environment 26, 977–989. Kakegawa K, Edashige Y, Ishii T. 2000. Metabolism of cell wall polysaccharides in cell suspension cultures of Populus alba in relation to cell growth. Physiologia Plantarum 105, 420–425. Karacsonyi S, Kovacik V, Alfoldi J, Kubackova M. 1984. Chemical and 13C-NMR studies of an arabinogalactan from Larix sibirica. Carbohydrate Research 134, 265–274. Kikuchi A, Edashige Y, Ishii T, Fujii T, Satoh S. 1996. Variations in the structure of neutral sugar chains in the pectic polysaccharides of morphologically different carrot calli and correlations with the size of cell clusters. Planta 198, 634–639. Kjellbom P, Snogerup L, Stöhr C, Reuzeau C, McCabe PF, Pennel RI. 1997. Oxidative cross-linking of plasma membrane arabinogalactan proteins. The Plant Journal 12, 1189–1196. Kreuger M, Van Holst GJ. 1993. Arabinogalactan proteins are essential in somatic embryogenesis of Daucus carota L. Planta 189, 243–248. Lahaye M, Rondeau-Mouro C, Deniaud E, Buleon A. 2003. Solid-state 13C NMR spectroscopy studies of xylans in the cell wall of Palmaria palmata. Carbohydrate Research 338, 1559–1569. Levigne S, Thomas M, Ralet MC, Quemener B, Thibault JF. 2002. Determination of the degrees of methylation and acetylation of pectins using a C18 column and internal standards. Food Hydrocolloids 16, 547–550. MacDougall AJ, Brett GM, Morris VJ, Rigby NM, Ridout MJ, Ring SR. 2001. The effect of peptide–pectin interactions in the gelation behaviour of a plant cell wall pectin. Carbohydrate Research 335, 115–126. McCartney L, Ormerod AP, Gidley MJ, Knox JP. 2000. Temporal and spatial regulation of pectic (1,4)-b-D-galactan in cell walls of developing pea cotyledons: implications for mechanical properties. The Plant Journal 22, 105–113. McCartney L, Knox JP. 2002. Regulation of pectic polysaccharide domains in relation to cell development and cell properties in the pea testa. Journal of Experimental Botany 53, 707–713. McCartney L, Steele-King CG, Jordan E, Knox JP. 2003. Cell wall pectic (1!4)-b-D-galactan marks the acceleration of cell elongation in Arabidopsis seedling root meristem. The Plant Journal 33, 447–454. McConnell AA, Eastwood MA, Mitchell WD. 1974. Physical characteristics of vegetable foodstuffs that could influence bowel function. Journal of the Science of Food and Agriculture 25, 1457–1464. Miller GL, Miller E. 1948. Determination of nitrogen in biological materials. Analytical Chemistry 20, 481–488. Nelson N. 1944. A photometric adaptation of the Somogyi method for determination of glucose. Journal of Biological Chemistry 153, 375–380. Oomen RJFJ, Doeswijk-Voragen CHL, Bush MS, et al. 2002. In muro fragmentation of the rhamnogalacturonan I backbone in potato (Solanum tuberosum L.) results in a reduction and altered location of the galactan and arabinan side-chains and abnormal periderm development. The Plant Journal 30, 403–413. Orfila C, Seymour GB, Willats WG, Huxham IM, Jarvis MC, Dover CJ, Thompson AJ, Knox JP. 2001. Altered middle lamella homogalacturonan and disrupted deposition of (1!5)-aL-arabinan in the pericarp of Cnr, a ripening mutant of tomato. Plant Physiology 126, 210–221. Orfila C, Huisman MMH, Willats WGT, van Alebeek GJWM, Schols HA, Seymour GB, Knox JP. 2002. Altered cell wall Adhesion strength of Arabidopsis microcalli cells disassembly during ripening of Cnr tomato fruit: implications for cell adhesion and fruit softening. Planta 215, 440–447. Parker ML, Waldron KW. 1995. Texture of chinese water chesnut: involvement of cell wall phenolics. Journal of the Science of Food and Agriculture 68, 337–346. Peng L, Hocart CH, Redmond JW, Williamson RE. 2000. Fractionation of carbohydrates in Arabidopsis root cell walls shows three radial swelling loci are specifically involved in cellulose production. Planta 211, 406–414. Pennell RI, Knox JP, Scofield GN, Selvendran RR, Roberts K. 1989. A family of abundant plasma membrane-associated glycoproteins related to arabinogalactan proteins is unique to flowering plants. Journal of Cell Biology 108, 1967–1977. Ridley BL, O’Neill MA, Mohnen D. 2001. Pectins: structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry 57, 929–967. Roberts K, Elliott KA, Gonzalez-Carranza ZH. 2002. Abscission, dehiscence, and other cell separation processes. Annual Review of Plant Biology 53, 131–158. Rondeau-Mouro C, Crepeau MJ, Lahaye M. 2003. Application of CP-MAS and liquid-like solid-state NMR experiments for the study of the ripening-associated cell wall changes in tomato. International Journal of Biological Macromolecules 31, 235–244. Rouau X, Odier E. 1986. Purification and properties of two enzymes from Dichomitus squalens which exhibit both cellobiohydrolase and xylanase activity. Carbohydrate Research 145, 279–292. Ryden P, Colquhoun IJ, Selvendran RR. 1989. Investigation of structural features of the pectic polysaccharides of onion by 13C NMR spectroscopy. Carbohydrate Research 185, 233–237. Saulnier L, Brillouet JM, Moutounet M. 1992. New investigations of the structure of grape arabinogalactan-protein. Carbohydrate Research 224, 219–235. Showalter AM. 2001. Arabinogalactan-proteins: structure, expression and function. Cellular and Molecular Life Sciences 58, 1399–1417. Skjot M, Pauly M, Bush MS, Borkhardt B, McCann MC, Ulvskov P. 2002. Direct interference with rhamnogalacturonan I biosynthesis in Golgi vesicles. Plant Physiology 129, 95–102. Smith DL, Abbott J, Gross KC. 2002. Down-regulation of tomato b-galactosidase 4 results in decreased fruit softening. Plant Physiology 129, 1755–1762. Sorensen SO, Pauly M, Bush M, Skjot M, McCann MC, Borkhardt B, Ulvskov P. 2000. Pectin engineering: modification of potato pectin 2097 by in vivo expression of an endo-1,4-b-D-galactanase. Proceedings of the National Academy of Sciences, USA 97, 7639–7644. Takeuchi Y, Komamine A. 1978. Changes in the composition of cell wall polysaccharides of suspension-cultured Vinca rosea cells during culture. Physiologia Plantarum 42, 21–28. Thibault JF. 1979. Automatization du dosage des substances pectiques par la méthode au Méta-hydroxydiphenyl. Lebensmittel Wissenschaft und Technologie 12, 247–251. Thompson JE, Fry SC. 2000. Evidence for covalent linkage between xyloglucan and acidic pectins in suspension-cultured rose cells. Planta 211, 275–286. Toonen MAJ, Schmidt EDL, van Kammen A, de Vries SC. 1997. Promotive and inhibitory effects of diverse arabinogalactan proteins on Daucus carota L. somatic embryogenesis. Planta 203, 188–195. Tsumaraya Y, Mochizuki N, Hashimoto Y, Kovac P. 1990. Purification of an exo-b-(1,3)-D-galactanase of Irpex lacteus (Polyporus tulipiferae) and its action on arabinogalactan-proteins. Journal of Biological Chemistry 265, 7207–7215. van Holst GJ, Clarke AE. 1985. Quantification of arabinogalactanprotein in plant extracts by single radial gel diffusion. Analytical Biochemistry 148, 446–450. Vincken J-P, Schols HA, Oomen RJFJ, McCann MC, Ulvskov P, Voragen AGJ, Visser RGF. 2003. If homogalacturonan were a side chain of rhamnogalacturonan I. Implications for cell wall architecture. Plant Physiology 132, 1781–1789. Wallner SJ, Nevins DJ. 1974. Changes in cell walls associated with cell separation in suspension cultures of Paul’s Scarlet rose. Journal of Experimental Botany 25, 1020–1029. Willats WG, Steele-King C, Marcus SE, Knox JP. 1999. Side chains of pectic polysaccharides are regulated in relation to cell proliferation and cell differentiation. The Plant Journal 20, 619–628. Willats WG, McCartney L, Mackie W, Knox JP. 2001. Pectin: cell biology and prospects for functional analysis. Plant Molecular Biology 47, 9–27. Yariv J, Lis H, Katchalski E. 1967. Precipitation of arabic acid and some seed polysaccharides by glycosylphenylazo dyes. Biochemical Journal 105, 1–2. Zablackis E, Huang J, Müller B, Darvill AG, Albersheim P. 1995. Characterization of the cell wall polysaccharides of Arabidopsis thaliana leaves. Plant Physiology 107, 1129–1138.