A complete cDNA from Pinus pinaster Aiton, potentially coding for an α-xylosidase able to remove the xylose residue from xyloglucan oligosaccharides, has been cloned. Its sequence was homologous to previously published α-xylosidase genes from Arabidopsis and nasturtium. The protein also showed the two signature regions of family 31 of glycosyl hydrolases. The gene expression level was quantified by competitive RT-PCR, under different growth conditions, throughout seedling development, in different regions along the hypocotyls and in auxin-treated hypocotyl segments, and related with growth capacity and α-xylosidase activity. A role of α-xylosidase in regulating the level of xyloglucan oligosaccharides within the apoplast is proposed. The action of an α-xylosidase removing the xylose residue, would make possible the action of a β-glucosidase deblocking the xyloglucan oligosaccharide degradation and it could serve as a control point for the regulation of the apoplastic levels of xyloglucan oligosaccharides.
(Received October 5, 2002; Accepted November 18, 2002)
Plant cell walls are built by two polysaccharide networks, cellulose microfibers cross-linked by xyloglucan chains and pectins, both networks contributing to their functional properties (Roberts 2001). The xyloglucan–cellulose complex has been considered to be responsible for controlling the extension capacity of primary cell walls, xyloglucan being the load-bearing component because of its proposed cross-linking of the cellulose microfibers (Fry 1989, Cosgrove 2001). It consists of a β(1-4)-d-glucan axis that mainly carries α-d-xylosyl and α-l-fucosyl-(1-2)-β-d-galactosyl-(1-2)-α-d-xylosyl as side chains attached at to the HO-6 of the β-glucosyl residues (Acebes et al. 1993b). In addition to its structural role, oligosaccharides derived from xyloglucan have been found to be formed in vivo (Fry 1986) and they have been shown to regulate auxin- (McDougall and Fry 1988, McDougall and Fry 1990) and acid pH-induced (Lorences et al. 1990a) growth.
Enzymes able to modify xyloglucan oligosaccharide side chains have been found in plant cell walls (Fry 1995). An α-xylosidase cleaves the α-xylosyl residue attached to the glucose residue farthest from the reducing end of the xyloglucan oligosaccharide (Fanutti et al. 1991). Plant α-xylosidases have been purified from pea epicotyls (O’Neill et al. 1989), nasturtium (Fanutti et al. 1991) and cabbage (Sampedro et al. 2001). They have an apparent molecular mass on SDS-PAGE of 85 kDa for the two first and of 87 kDa for the latter one. Recently, an α-xylosidase (AtXYL1) from Arabidopsis thaliana (Sampedro et al. 2001) and another one from Tropaeolum majus (Crombie et al. 2002) have been cloned. Both α-xylosidases (GenBank accession nos. AF144078 and AJ131520) showed a very high homology with 75% identity over an 899-amino acid consensus length. Both α-xylosidase amino acid sequences showed the two signature regions of family 31 of glycosyl hydrolases (Henrissat 1991, Frandsen and Svensson 1998) and thus they were members of that family. Furthermore, AtXYL1 gene expression as well as α-xylosidase activity in Arabidopsis thaliana leaves have been shown to be developmentally regulated, their levels being higher in the younger leaves which are in the fast growth phase (Sampedro et al. 2001).
Since the removal of the xylosyl residues from the xyloglucan oligosaccharides by the α-xylosidase seems to be a necessary prerequisite for the oligosaccharide degradation to prevent its accumulation into the apoplast, the apoplastic α-xylosidase might be an important enzyme in relation to cell wall loosening. Thus we studied that enzyme in relation to plant growth. First, we report the search in Pinus pinaster Aiton of sequences homologous to known plant α-xylosidases and the cloning of a putative pine α-xylosidase gene. Changes in the gene expression as well as α-xylosidase activity during intact and auxin-induced growth are also reported.
Two plant α-xylosidase genes, one from Arabidopsis thaliana (Sampedro et al. 2001) and the other from nasturtium (Crombie et al. 2002) have been cloned previously. Based on both sequences (Fig. 1), a search on public databases was performed, and a Pinus radiata EST (AA220895) with a high similarity was found. The sequence of the P. radiata EST was completed, and degenerated oligonucleotide primers designed from conserved regions were used in RT-PCR with RNA isolated from 7-day-old Pinus pinaster seedlings. The PCR product was extended towards the 5′ and the 3′ ends by RACE and the complete sequence obtained. That sequence from Pinus pinaster was considered to code for an α-xylosidase and named as PpXYL1 (GenBank accession no. AF44821). The translated sequence was 910 amino acids long and it showed a secretion signal peptide, 21 amino acids long. It has also eight potential N-glycosylation sites as predicted by PROSITE (Falquet et al. 2002). The amino acid sequence identity between the protein coded by PpXYL1 and the two signature regions of family 31 glycosyl hydrolases (Henrissat 1991, Frandsen and Svensson 1998) showed that the PpXYL1 was a member of that family as the α-xylosidases from Arabidopsis (Sampedro et al. 2001) and nasturtium (Crombie et al. 2002) were.
An α-xylosidase activity able to remove xylosyl residue from xyloglucan oligosaccharides was present in hypocotyls of etiolated seedling of Pinus pinaster. That activity has been described previously for pea (O’Neill et al. 1989), Arabidopsis (Sampedro et al. 2001) and nasturtium (Crombie et al. 2002). The α-xylosidase activity and the expression levels of PpXYL1 were measured using whole hypocotyls or hypocotyl regions with different growth capacity. Fig. 2 shows the changes in the α-xylosidase activity and also in the PpXYL1 transcript levels with age. The α-xylosidase activity decreased with the hypocotyl age (Fig. 2A), that decrease being higher from the 7th to the 10th day after sowing. The changes in PpXYL1 expression with the development of pine seedlings were studied using competitive RT-PCR. The expression levels of PpXYL1 was referred to the expression of an ubiquitin gene (PpUBIQ1), considered as a control gene, and noted as the ratio between the transcripts from PpXYL1 and PpUBIQ. The changes in the expression of PpXYL1 during seedling growth are shown in Fig. 2B. Its expression dramatically decreased from the 7th to the 10th day, the decrease continuing more slowly up to the 16th day. As it has been shown previously (Lorences and Zarra 1986), pine hypocotyls grew very fast up to the 7th day and after that their growth rate continuously decreased until the 13th day with growth ceasing completely at the 16th day.
Although the hypocotyls actively grew up to the 10th day, they showed a growth gradient, the apical region showing an active growth while the basal one almost ceased to grow (Lorences et al. 1990b). The measurements using the whole hypocotyl of 10-day-old seedlings masked the differences among the different hypocotyl regions. Thus, it was necessary to study the α-xylosidase activity and the PpXYL1 expression levels in the different regions along the hypocotyl (Fig. 3). The α-xylosidase activity was higher at the apical region and decreased towards the basal region. The expression of PpXYL1 was also studied along the hypocotyl axis of 10-day-old seedlings (Fig. 3). The expression was very high at the apical region and it was low in the other regions.
The effect of indole-3-acetic acid on the α-xylosidase activity and on the PpXYL1 expression of pine hypocotyl segments has been also studied (Fig. 4). The pine hypocotyl segments were incubated with 10–4 M, the optimum concentration for growth induction in this material (Lorences and Zarra 1986). It also induced an increase in the α-xylosidase activity from the same material. The effect was noted after 3 h of incubation and later it decreased. Furthermore, PpXYL1 transcript levels in hypocotyl segments treated with auxin, under conditions that induced growth, were higher than in controls segments. The effect was noted at 90 min and after it decreased. Both α-xylosidase activity and PpXYL1 transcript levels of non-treated material (controls) rapidly decreased in the first h, suggesting an active turnover.
A Pinus pinaster cDNA sequence encoding a putative α-xylosidase (PpXYL1) as deduced by comparison with homologous sequences from Arabidopsis thaliana (Sampedro et al. 2001) and nasturtium (Crombie et al. 2002) has been reported (Fig. 1). The homology between the putative α-xylosidase from pine and those of Arabidopsis and nasturtium was 72% for both amino acidic sequences. All three sequences are members of the family 31 of glycosyl hydrolases (Fig. 5) (Henrissat 1991). Most of the characterized enzymes in family 31 are α-glucosidases, although two α-xylosidases from prokaryotes (accession nos. AAC62251 and AJ251975) have been discovered recently (Chaillou et al. 1998, Moracci et al. 2000). These α-xylosidases are not closely related to their plant counterparts and seem to have evolved independently. On the other hand, all known plant α-xylosidases appear in the same branch of family 31 cladogram which include the putative pine α-xylosidase as well as several unidentified proteins. Moreover, Crombie et al. (2002) found that in nasturtium both activities were coded by the same cDNA. In fact, both substrates, α-xylose and α-glucose, are structurally very similar. Taylor et al. (2001) proposed that the α-glucosidases located in the apoplast function primarily as α-xylosidases and have a role in xyloglucan oligosaccharide metabolism. In addition, the translated protein showed a signal peptide, in agreement with an apoplastic localization as expected for an enzyme able to modify the xyloglucan oligosaccharides present in the apoplastic fluid (Fig. 1). Thus, it seems reasonable to assume that the protein encoded by PpXYL1 may act in vivo on xyloglucan oligosaccharides generated within the cell wall, although other possibilities cannot be discarded. Furthermore, the changes in the expression of PpXYL1 resemble the changes in the α-xylosidase activity under the different growth conditions (Fig. 2, 3, 4) supporting the idea that PpXYL1 codes for a α-xylosidase. However, the existence of more than one gene coding for α-xylosidase cannot be excluded.
Our previous work on Pinus pinaster hypocotyls has shown that xyloglucan is the major hemicellulosic polysaccharide in the primary walls (Acebes et al. 1993b), where it forms a macromolecular complex with cellulose microfibers (Acebes et al. 1993a) and constitutes the load-bearing network as it has been demonstrated for dicot plants (Fry 1989, Cosgrove 2001). Furthermore, changes in xyloglucan during auxin-induced growth (Lorences and Zarra 1987), acid pH-induced growth (Lorences et al. 1989) and intact growth (Lorences et al. 1990b) have been related to the biochemical basis of the cell wall loosening process that leads to cell wall extension. Consequently, enzymes that can act on xyloglucan may play an important role in the loosening process. A glycanase system able to degrade xyloglucan associated with pine cell walls has been described already (Acebes and Zarra 1992). Furthermore, Barrachina and Lorences (1998) have demonstrated also that a XET bound to pine cell walls was able to act on wall-bound xyloglucan as well as on soluble xyloglucan using them as substrates for the endotransglycosylation reaction. A crucial step to degrade in vivo the xyloglucan oligosaccharides is the removal of the xylosyl residue from their non-reducing end, opening it up to further degradation by a β-glucosidase. In this study we report the presence of an α-xylosidase activity in pine hypocotyls able to remove xylosyl residues from the xyloglucan oligosaccharides used as substrates (Fig. 2, 3, 4). Its activity was related to the growth capacity of the plant material. So, the activity was high in the hypocotyls from young pine seedlings (Fig. 2) which have been shown to be in the fast growth phase (Lorences and Zarra 1986). The α-xylosidase activity was also higher in the hypocotyl apical region (Fig. 3) which is the most active growing region (Lorences et al. 1990b). Furthermore, auxin treatment of the hypocotyl segments also increased the α-xylosidase activity (Fig. 4) under the same conditions that were also able to induce elongation growth (Lorences and Zarra 1987). A positive relationship between growth rate and α-xylosidase activity has already been found in Arabidopsis thaliana (Sampedro et al. 2001).
The expression level of PpXYL1 was high in the hypocotyls of young pine seedlings and decreased with age (Fig. 2), as well as along the hypocotyl axis (Fig. 3). Furthermore, auxin also increased the expression level of PpXYL1 (Fig. 4). Thus, PpXYL1 transcript levels and their changes with age, as well as along the hypocotyl, and during auxin-induced growth were in agreement with the changes found for α-xylosidase activity.
α-Xylosidase has been proposed as a key enzyme in regulating the level of xyloglucan oligosaccharides within the apoplast (Sampedro et al. 2001). There are several oligosaccharides in the apoplast and some of them show a powerful effect on the growth rate (McDougall and Fry 1988, McDougall and Fry 1990, Lorences et al. 1990a), and in some cases, that effect requires a xylose residue in the non-reducing-end of the oligosaccharide molecule (Lorences and Fry 1993). For these oligosaccharides, the level of α-xylosidase would be critical. Furthermore, the action of an α-xylosidase removing the xylose residue, would make possible the action of a β-glucosidase deblocking the xyloglucan oligosaccharide degradation and it could serve as a control point for the regulation of their apoplastic levels (Crombie et al. 2002). If xyloglucan oligosaccharides accumulate in the apoplast, they would compete with de novo synthesized xyloglucan chains, to act as XET acceptors. The incorporation of small xyloglucan oligosaccharides would shorten the xyloglucan linkages between the cellulose microfibrils, leading to a weakening of the cell walls and the cessation of growth. Further studies on the β-glucosidase that acts on xyloglucan oligosaccharides in combination with α-xylosidase will be necessary to fully understand that regulation.
Material and methods
Seedlings of Pinus pinaster Aiton grown at 25°C under darkness were used as experimental material (Lorences and Zarra 1986). Whole hypocotyls were harvested at 7, 10, 13 and 16 d after being soaked. Ten-day-old hypocotyls were divided into four 5-mm sections from the cotyledonary node towards the base and the sections were named I, II, III and IV, as the distance from the cotyledonary node increased.
RNA was extracted from 100 mg of plant material with RNeasy Plant Mini Kit (Qiagen, Valencia, CA, U.S.A.) and was quantified through A260.
Similarity searches were done with the National Center of Biological Information BLAST 2.0 (Altschul et al. 1997). Multiple alignments were done with MultAlin (Corpet 1988), and phylogenetic and molecular evolutionary analyses were conducted using MEGA version 2.1 (http://www.megasoftware.net).
Pinus radiata EST (AA220895) was a kind gift of Sharon L. Bishop-Hurley from the Forest Research Institute (Rotorua, New Zealand). The clone was fully sequenced by CIB (CSIC, Madrid, Spain).
Enhanced Avian RT-PCR Kit (Sigma España, Madrid, Spain) was employed as recommended by the manufacturer to perform two-step RT-PCR. Total RNA from Pinus pinaster (1 µg) was used as the template for reverse transcription with an oligo d(T) primer to produce single-strand cDNA.
Based on the conserved amino acid domains of known plant α-xylosidase genes (Fig. 1), two degenerated oligonucleotide primers (5′-GTGAGCAAACAGTTYCTICTNGG-3′ and 5′-ATCATCAATTTCTCNACNACCCA-3′) were designed and employed for PCR to produce double-stranded cDNA. The product obtained (489 bp long) was then fully sequenced.
Generation of full-length cDNA
The 5′ and 3′ ends of the cDNA fragment from Pinus pinaster were obtained by RACE performed with a Gene RACER Kit (Invitrogen, Groningen, The Netherlands), following the manufacturer instructions.
Amplification of the 3′ end was performed using a forward gene-specific primer 3GSP (5′-TCTGCCCATGCAAAGAGGGGAATGA-3′) and the GeneRacer 3′ primer (Invitrogen), and the 5′ end was PCR amplified using a reverse gene specific primer 5GSNestP (5′-GTACCGGCACTTCAGGTCACCCGT-3′) and the GeneRacer 5′Nested primer (Invitrogen). ThermoZyme DNA polymerase (Invitrogen) was used according to the manufacturer recommendations. PCR products of the expected size were observed on agarose electrophoresis, purified from the gel with QIAEX II gel extraction kit (QIAGEN GmbH, Hilden, Germany), cloned into pCR 4-TOPO (TOPO TA Cloning Kit for Sequencing, Invitrogen) and sequenced.
The auxin treatment was performed as described previously (Lorences and Zarra 1987). Hypocotyl apical segments (5 mm long) were pre-incubated for 2 h with 50 mM sodium citrate-phosphate buffer pH 6.5 with orbital shaking under dark and at 25°C. Later, the sections were transferred to the same medium with or without 10–4 M indolacetic acid, incubated under the same conditions for different periods, and the length of the sections measured with a binocular microscope to the nearest 0.1 mm. All growth experiments were performed under dim green light. After completion of the growth experiments, the plant material was frozen in liquid nitrogen and stored at –80°C.
Protein extract preparation
The protein extract was prepared as previously described (Sampedro et al. 2001). Plant material frozen with liquid nitrogen was homogenized with a mortar and pestle to a fine powder. The powder was suspended in 1 M sodium acetate buffer pH 4.5 containing a protease inhibitor cocktail (P9599, Sigma España). After 1 h at 4°C and magnetic stirring, the suspension was centrifuged at 16,000×g for 15 min. The supernatant was dialyzed against 100 mM sodium acetate pH 4.5 overnight, and considered as the protein extract.
α-Xylosidase activity was measured as described previously (Sampedro et al. 2001). Aliquots of the protein extract were incubated with 200 µg of xyloglucan oligosaccharides in 0.1 M sodium acetate pH 4.5 at 37°C for a variable period according to the extract activity. The activity was measured as pentose release, xylose being the only pentose in xyloglucan, according to the method of Roe and Rice (1948) as modified by Edwards et al. (1985). The xyloglucan oligosaccharides used as substrate for α-xylosidase were an oligosaccharide mixture prepared by Trichoderma viride endoglucanase (Sigma España) digestion of tamarind (Tamarindus indica) xyloglucan.
Protein content was measured using Coomassie protein assay reagent (Pierce, Rockford, IL, U.S.A.).
The patterns of expression of the gene PpXYL1 were studied by competitive PCR. The competitors were prepared with a Competitive DNA Construction Kit (Takara Biomedicals) to amplify a sequence of 450 bp to compete with the amplification of a fragment of the gene PpXYL1 (550 bp). A Pinus pinaster ubiquitin (PpUBIQ1) was used as the control gene. As there were no sequences in public databases of any Pinus pinaster ubiquitin, it was necessary to obtain one. Degenerate oligonucleotide primers MIUBIQ1 (5′-GGAGCTTCAGGATYTGCAGARGGATCC-3′) and MIUBIQ2 (5′-TCAGTTCATGGCATAYTTYTGRGT-3′) designed from conserved regions of other plant ubiquitin sequences (Pinus resinosa, AF001948; Arabidopsis, L00639; tomato, L23762) were used for RT-PCR with RNA isolated from Pinus pinaster. The PCR product was cloned, sequenced and sent to GenBank (accession number AF461687).
Once both competitors for α-xylosidase and ubiquitin were prepared as above, they were purified on an agarose gel and quantified through A260. RT-PCRs were performed using RNA obtained from pine hypocotyls of different ages (7, 10, 13 and 16 days), from different regions along the 10-day-old hypocotyls and from apical region I treated or untreated with exogenous auxins. First strand cDNA was prepared using a First Strand RT-PCR Kit (Roche). RNA was treated previously with DNase (Promega) following the manufacturer instructions. PCRs were performed with the different cDNA templates and by adding variable amounts of the competitor. At the same time as the fragments of PpXYL1 or PpUBIQ1 cDNA were amplified, their respective competitors were produced. PCR products were loaded on an agarose gel and the differing density of the bands was quantified from a digital picture with Scion Image (Scion Corporation, U.S.A.).
We are grateful to Dr. Ester P. Lorences for her helpful comments and to Duncan A. Lindsay for correcting the English version. We are also grateful to S.L. Bishop-Hurley from the Forest Research Institute, Rotorua, New Zealand for his kind gift of Pinus radiata EST (accession no. AA220895). This work was supported by a Marie Curie Fellowship to M. Sánchez (HPMF-CT-1999-0330) and a DGIC (PB98-0640, MCyT, Spain) and a XUNTA de Galicia (PGIDT00PXI20002PN) grants.
Corresponding author: E-mail: email@example.com; Fax: 34-981-596904.