Xyloglucan breakdown during cotton fiber development

J. Plant Physiol. 160. 1411 – 1414 (2003)  Urban & Fischer Verlag http://www.urbanfischer.de/journals/jpp

Short Communication Xyloglucan breakdown during cotton fiber development Hayato Tokumoto1, Kazuyuki Wakabayashi1, Seiichiro Kamisaka2, Takayuki Hoson1 * 1

Department of Biological Sciences, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka, 558-8585, Japan

2

Department of Biology, Faculty of Science, Toyama University, Gofuku, Toyama, 930-8555, Japan

Received October 24, 2002 · Accepted January 15, 2003

Summary Cotton (Gossypium herbaceum L.) fibers elongated almost linearly up to about 20 days post anthesis. The molecular mass of xyloglucans in fiber cell walls decreased gradually during the elongation stage. When enzymatically active (native) cell wall preparations of fibers were autolyzed, the molecular mass of xyloglucans decreased. The decrease was most prominent in wall preparations obtained from the rapidly elongating fibers. The xyloglucan-degrading activity was recovered from the fiber cell walls with 3 mol/L NaCl, and the activity was high at the stages in which fibers elongated vigorously. These results suggest the possible involvement of xyloglucan metabolism in the regulation of cotton fiber elongation. Key words: cotton fiber – elongation – Gossypium herbaceum – molecular mass – xyloglucans Abbreviations: DPA = days post anthesis. – HC-II = hemicellulose II. – HPLC = high performance liquid chromatography

Introduction Cotton fibers are differentiated from single epidermal cells of developing seeds without cell division, and thus, are excellent systems for studying the mechanisms controlling cell differentiation and cell elongation (Kim and Triplett 2001). Generally, the cotton fibers are not tip growing but expand via diffuse growth (Tiwari and Wilkins 1995), and the fiber cells elongate up to 30 mm for about 20 days post anthesis (DPA). Therefore, cotton fibers are one of the longest single cells in higher plants. * E-mail corresponding author: [email protected]

The rapid elongation of cotton fibers is believed to be driven by high turgor and a highly extensible primary cell wall (Ruan and Chourey 1998, Ruan et al. 2000, 2001). Critical questions remain regarding the regulation of the capacity of cell walls to extend during cotton fiber elongation, despite recent progress in research, such as the investigation of gene expression of expansin (GhEXP1) in growing cotton fibers (Orford and Timmis 1998). Cell walls of growing cotton fibers contain significant amounts of xyloglucans, and the levels of xyloglucans show considerable changes during fiber development (Huwyler et al. 1979, Shimizu et al. 1997). Recently, we found that the molecular size of xyloglucans also changed prominently dur0176-1617/03/160/11-1411 $ 15.00/0

1412

Hayato Tokumoto et al.

ing cotton fiber development (Tokumoto et al. 2002). In the primary cell wall of dicotyledonous plants, xyloglucans are the major components of the matrix polysaccharides and serve as cross-links between cellulose microfibrils (Hayashi 1989). It has been shown that changes in the molecular size of xyloglucans are involved in the regulation of the capacity of cell walls to extend (Masuda 1990, Sakurai 1991, Hoson 1993, 2002). Furthermore, changes in the activity of xyloglucandegrading enzymes have been demonstrated to contribute to the alteration of the molecular mass of xyloglucans in azuki bean epicotyls (Soga et al. 1999, 2000) and Arabidopsis hypocotyls (Soga et al. 2002). These findings suggest that the activity of xyloglucan-degrading enzymes might be associated with the regulation of the xyloglucan molecular mass in cotton fibers. In the present study, we have measured the activity of xyloglucan-degrading enzymes in vivo and in vitro in elongating cotton fibers to assess this hypothesis.

The extraction and the assay of the xyloglucan-degrading enzymes were carried out by the methods of Hoson et al. (1995) and Soga et al. (1999). The native cell wall materials were suspended in 20 mmol/L sodium phosphate buffer (pH 6.0) containing 3 mol/L NaCl, and kept for 24 h at 4 ˚C and then filtered through polypropylene mesh. The filtrate was dialyzed against 20 mmol/L sodium phosphate buffer (pH 6.0) at 4 ˚C. The dialyzed solution was used for the assay of the xyloglucan-degrading activity. The protein content in the extract was determined using the Bio-Rad Protein Assay Kit (Bio-Rad Lab., Hercules, CA, USA). The reaction mixtures (200 µL of 20 mmol/L sodium phosphate buffer, pH 6.0, containing 0.02 % NaN3) contained 200 µg of purified azuki bean xyloglucans (400 – 600 kDa, Tabuchi et al. 1997) and ca. 20 µg cell wall proteins, and they were incubated for 72 h at 37 ˚C. As a control, the enzymes denatured by heating were used. The activity of xyloglucan-degrading enzymes was determined by the iodine method and expressed in terms of the decrease in absorbance at 640 nm of the xyloglucan-iodine complex.

Results and Discussion Materials and Methods Seeds of cotton (Gossypium herbaceum L. cv. Shisomen) were sown in the middle of May 1999 at the experimental field adjacent to the university campus. On the day of flowering, each individual flower was tagged and healthy bolls were harvested at different developmental stages (9, 12, 14, 17, 21 and 28 DPA). The pericarp was removed immediately after harvest, and the ovules were frozen in liquid nitrogen and stored at – 30 ˚C. The measurements of length and dry mass of cotton fibers were performed as described previously (Tokumoto et al. 2002). The matrix polysaccharides of cotton fiber cell walls were fractionated into the pectin, the hemicellulose-I, and the hemicellulose-II (HCII) fractions by sequential extraction with 50 mmol/L ethylenediaminetetraacetic acid, 4 % KOH and 24 % KOH, respectively (Tokumoto et al. 2002). Total sugar and xyloglucan content in the fractions was determined by using the phenol-sulfuric acid method (Dubois et al. 1956) and the iodine staining method (Kooiman 1960), respectively. The molecular mass of xyloglucans in the HC-II fraction was analyzed by high performance liquid chromatography (HPLC) using a gel-permeation column (TSKgel G5000PW, Tosoh Co. Ltd., Tokyo) (Tokumoto et al. 2002). Xyloglucan content in each fraction was determined by the iodine method. The weight-average molecular mass of xyloglucans was calculated from the equation proposed by Nishitani and Masuda (1981). Dextrans (Sigma) of 10, 40, 70, 120 and 500 kDa were used as molecular mass markers. Cotton fibers, which had been detached from the thawed ovules, were homogenized in ice-cold 20 mmol/L sodium phosphate buffer (pH 7.0), and washed on a polypropylene mesh (32 µm) with the same buffer. The cell walls thus prepared were enzymatically active and designated as native cell walls. The cell wall samples were suspended in 20 mmol/L sodium phosphate buffer (pH 6.0) containing 0.02 % NaN3 and incubated for 72 h at 37˚C. Part of the samples were boiled for 10 min in 80 % ethanol before autolysis experiments as denatured controls. After the incubation, the suspension was centrifuged (1000 g, 10 min), and the residual cell wall materials were washed with cold water and boiled in 80 % ethanol for 10 min. The HC-II fractions were extracted from the autolyzed cell walls and the molecular masses of xyloglucans were analyzed by HPLC, as described above.

The length of cotton fibers increased almost linearly up to 20 DPA and reached a final length of ca. 30 mm, while the fiber dry mass increased slowly for the first 20 days and then rapidly after the cessation of fiber elongation (Table 1). Therefore, the cotton fiber development consisted of the elongation phase up to 20 DPA and the following cell wall thickening phase. We had similar results in repeated experiments in 1997 and 1998 (Tokumoto et al. 2002). Most xyloglucans were extracted with strong alkali (24 % KOH). Thus, xyloglucans were the major components of the HC-II fraction (Tokumoto et al. 2002). The weight-average molecular mass of xyloglucans in the HC-II fraction was high when the fibers were young (9–12 DPA) and decreased grad-

Table 1. Changes in length, dry mass and molecular mass of xyloglucans during cotton fiber development. DPA

Length (mm)

Dry mass (mg seed – 1)

Molecular mass of XGs (kDa)

9 12 14 17 21 28

5.0 ± 0.2 11.5 ± 0.3 14.7 ± 0.3 23.6 ± 0.4 29.6 ± 0.7 29.9 ± 0.8

0.5 ± 0.0 2.0 ± 0.1 2.7 ± 0.2 3.8 ± 0.1 11.9 ± 0.6 34.6 ± 2.7

500 ± 51 459 ± 56 349 ± 13* 310 ± 10* 296 ± 14* 411 ± 58

Length and dry mass of cotton fibers at different growth stages were measured. Data are means ± SE (n = 30 and 3 for the determination of fiber length and dry mass, respectively). The molecular mass distribution of xyloglucans (XGs) in the HC-II fraction extracted from cell walls of cotton fibers at different growth stages was determined by HPLCgel permeation chromatography. Weight-average molecular masses were calculated based on the elution profiles of xyloglucans. Data are means ± SE from three independent samples. * Significant decrease from the value at 9 DPA at 5 % level.

Xyloglucan breakdown in cotton fibers

1413

The autolysis products released from native cell walls were almost completely composed of glucose, but did not contain xyloglucan fragments (data not shown). Bucheli et al. (1987) reported that the glucose residues in the autolysis products of cotton fiber cell walls were mostly derived from (1– 3)-β-glucans. Molecular mass downshift of xyloglucans during autolysis treatment without the release of xyloglucan fragments has been observed in cell wall preparations of azuki bean epicotyls (Soga et al. 2000). Therefore, xyloglucans were extracted from the fiber cell walls remaining after autolysis with 24 % KOH, and their molecular masses were determined (Fig. 1). The molecular mass distribution of xyloglucans obtained from autolyzed cell walls shifted to lower values compared to the initial cell walls. The calculated average molecular mass of xyloglucans decreased during autolysis. The decrease in the molecular mass was significant (P < 0.05) in cell wall preparations from rapidly elongating fibers (14 DPA), but not significant at 17 and 21 DPA. The molecular mass of xyloglucans obtained from heat-inactivated cell wall materials did not change during the incubation. These results indicate that cotton fiber cell walls have a capacity to degrade xyloglucans autolytically and that this capacity is high when the breakdown of xyloglucans is active in the cell walls (Table 1). The xyloglucan depolymerization during autolytic incubation of the native cell wall preparations (Fig. 1) suggests that the xyloglucan breakdown in fiber cell walls was mediated by the action of the cell wall-bound enzymes. The xyloglucandegrading activity present in a protein fraction extracted with 3 mol/L NaCl from the fiber walls was determined by the iodine method using purified azuki bean xyloglucans (Table 2). The activity was high when fibers elongated vigorously (12 and 14 DPA), but decreased substantially and was kept low thereafter (after 17 DPA). The activity per seed also tended to decrease with aging (data not shown). The xyloglucandegrading activity is in good accordance with the changes in the molecular mass of xyloglucans (Table 1). Therefore, there was a parallel relation among fiber elongation, the decrease Figure 1. Elution profiles of xyloglucans in the HC-II fraction from autolyzed cell walls of cotton fibers at different growth stages. The HC-II fraction extracted from the initial (0 h) and autolyzed cell walls (72 h) were subjected to HPLC-gel permeation chromatography. The amount of xyloglucans in each fraction was determined by the iodine method. Each elution profile is the mean of three independent samples. Numbers in parentheses are the weight-average molecular mass (in kDa) calculated from the elution profiles of xyloglucans. Data are means ± SE from three independent samples. The elution positions of molecular mass standards (dextrans of 500, 70 and 10 kDa, and glucose: Glc) and the void volume (V0) are shown at the top.

ually during fiber elongation (from 9 to 21 DPA) (Table 1). Although the molecular mass of xyloglucans slightly increased from 21 to 28 DPA, the increase was not significant (at 5 % level). These results suggest that xyloglucans are actively degraded in rapidly elongating fibers and that the degradation ceases when fiber elongation is completed.

Table 2. Changes in the xyloglucan-degrading activity during cotton fiber development. DPA

Xyloglucan-degrading activity (% µg protein – 1)

12 14 17 21 28

4.0 ± 0.3 2.4 ± 0.1 0.4 ± 0.1 0.2 ± 0.1 0.02 ± 0.00

Proteins, extracted from fiber cell walls with 3 mol/L NaCl, were incubated with azuki bean xyloglucans in 20 mmol/L sodium phosphate buffer (pH 6.0) at 37 ˚C for 72 h. The activities were assayed by the iodine method and expressed in terms of the decrease in absorbance at 640 nm of the xyloglucan-iodine complex per µg proteins. Data are means ± SE from three independent samples.

1414

Hayato Tokumoto et al.

in molecular mass of xyloglucans, and the activity of xyloglucan-degrading enzymes in vivo and in vitro. Since such a decrease was clearly observed in the specific activity basis (Table 2), it may be due to a decrease in the net amount of xyloglucan-degrading enzymes but not a reduction of the efficiency of protein extraction with aging. In the present study, we detected xyloglucan-degrading activity in the cell wall of cotton fibers. However, whether such an activity is attributed to xyloglucan endo-(1– 4)-β-glucanases (Shimizu et al. 1997) or to xyloglucan endotransglucosylases/hydrolases (Nishitani 1997, Tabuchi et al. 1997, 2001), is an unsolved problem. In xyloglucan degradation in vivo, the molecular mass downshift of xyloglucans attained was 250 – 260 kDa and no xyloglucan fragments were released from cell walls (Fig. 1). Such a pattern of degradation could be explained by the action of a member of the xyloglucan endotransglucosylases/hydrolases with hydrolytic activity (Tabuchi et al. 1997, 2001). The enzymes preferentially degraded azuki bean xyloglucans rather than those from tamarind seeds, which lack terminal fucose residues (data not shown). The nature of xyloglucan-degrading enzymes in cotton fiber cell walls should be clarified by further studies.

References Bucheli P, Buchala AJ, Meier H (1987) Physiol Plant 70: 633 – 638 Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956) Anal Chem 28: 350 – 356

Hayashi T (1989) Annu Rev Plant Physiol Plant Mol Biol 40: 139–168 Hoson T (1993) J Plant Res 106: 369 – 381 Hoson T (2002) J Plant Res 115: 277– 282 Hoson T, Tabuchi A, Masuda Y (1995) J Plant Physiol 147: 219 – 224 Huwyler HR, Franz G, Meier H (1979) Planta 146: 635 – 642 Kim HJ, Triplett BA (2001) Plant Physiol 127: 1361–1366 Kooiman P (1960) Recl Trav Chim Pays-Bas 79: 675 – 678 Masuda Y (1990) Bot Mag 103: 345 – 370 Nishitani K (1997) Int Rev Cytol 173: 157– 206 Nishitani K, Masuda Y (1981) Physiol Plant 52: 482 – 494 Orford SJ, Timmis JN (1998) Biochim Biophys Acta 1398: 342 – 346 Ruan YL, Chourey PS (1998) Plant Physiol 118: 399 – 406 Ruan YL, Llewellyn DJ, Furbank RT (2000) Aust J Plant Physiol 27: 795 – 800 Ruan YL, Llewellyn DJ, Furbank RT (2001) Plant Cell 13: 47– 60 Sakurai N (1991) Bot Mag 104: 235 – 251 Shimizu Y, Aotsuka S, Hasegawa O, Kawada T, Sakuno T, Sakai F, Hayashi T (1997) Plant Cell Physiol 38: 375 – 378 Soga K, Wakabayashi K, Hoson T, Kamisaka S (1999) Plant Cell Physiol 40: 581– 585 Soga K, Wakabayashi K, Hoson T, Kamisaka S (2000) Plant Cell Physiol 41: 509 – 514 Soga K, Wakabayashi K, Kamisaka S, Hoson T (2002) Planta 215: 1040–1046 Tabuchi A, Kamisaka S, Hoson T (1997) Plant Cell Physiol 38: 653 – 658 Tabuchi A, Mori H, Kamisaka S, Hoson T (2001) Plant Cell Physiol 42: 154–161 Tiwari SC, Wilkins TA (1995) Can J Bot 73: 746–757 Tokumoto H, Wakabayashi K, Kamisaka S, Hoson T (2002) Plant Cell Physiol 43: 411– 418