Skip Navigation


JXB Advance Access originally published online on July 26, 2006
Journal of Experimental Botany 2006 57(12):2909-2922; doi:10.1093/jxb/erl064
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
57/12/2909    most recent
erl064v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Van Sandt, V. S. T.
Right arrow Articles by Vissenberg, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Van Sandt, V. S. T.
Right arrow Articles by Vissenberg, K.
Agricola
Right arrow Articles by Van Sandt, V. S. T.
Right arrow Articles by Vissenberg, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

RESEARCH PAPER

Analysis of a xyloglucan endotransglycosylase/hydrolase (XTH) from the lycopodiophyte Selaginella kraussiana suggests that XTH sequence characteristics and function are highly conserved during the evolution of vascular plants

Vicky S. T. Van Sandt1, Yves Guisez2, Jean-Pierre Verbelen1 and Kris Vissenberg1,*

1University of Antwerpen, Biology Department, Plant Physiology and Morphology, Groenenborgerlaan 171, B-2020 Antwerpen, Belgium
2University of Antwerpen, Biology Department, Plant Physiology, Groenenborgerlaan 171, B-2020 Antwerpen, Belgium

*To whom correspondence should be addressed. E-mail: kris.vissenberg{at}ua.ac.be

Received 10 January 2006; Accepted 15 May 2006


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Supplementary data
 References
 
A tissue print followed by a xyloglucan endotransglycosylase assay revealed that XET activity is present at sites of cell elongation in both roots and shoots of the lycopodiophyte Selaginella kraussiana. This paper provides the first report and analysis of a xyloglucan endotransglycosylase/hydrolase (XTH) cDNA sequence, isolated from a club moss. In silico analysis of the deduced amino acid sequence revealed a strong conservation of the XET-domain described in higher plants. The catalytic site (DEIDLEFLG) varies in only one amino acid compared with the consensus sequence and was shown to be functional after recombinant expression of Sk-XTH1 in Pichia pastoris. Sk-XTH1 displays xyloglucan endotransglycosylase activity over a broad pH (4.5–7.5) and temperature range (4–30 °C), but it shows no hydrolase activity. The catalytic site is followed by a consensus sequence for N-linked glycosylation. Four terminal cysteines were shown to stabilize a putative XET-C terminal extension region, which includes conserved amino acids, involved in the recognition and binding of the substrates. The N-linked sugar interactions as well as the disulphide bridges were shown to be necessary to perform XET activity. The presence of a highly conserved XTH sequence and function in a microphyllophyte suggests that XTHs were present before the divergence of lycopodiophytes and euphyllophytes. It also points to a possible key role for XTHs in the production of a cell wall that allowed the further evolution of land plants.

Key words: Evolution, Pichia pastoris, primary cell wall, Selaginella, XET/XEH activity, XTH, vascular plants


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Supplementary data
 References
 
It is generally assumed that ancestors of the present land plants evolved from green algae as they have many basic features in common, like starch as a reserve, a cellulosic cell wall and the presence of chlorophyll a and b, and the accessory pigment ß-carotene (Vettermann, 1973; Apel et al., 1975; Brown, 1990). Early plants colonized land about 425–475 million years ago (Heckman et al., 2001; Wellman et al., 2003) and encountered an entirely different set of environmental conditions compared with an aquatic habitat. The loss of a supporting environment, like water, demanded more structural strength of the plant body. The plants developed an extensive root system for efficient anchorage in the soil and absorption of nutrients and water, and aerial parts bearing leaves as specialized organs for photosynthesis. Directional growth and flexibility became an important aspect to survive in this ‘new’ water-free environment with many new kinds of stresses. Plants generally respond to biotic and abiotic stresses with alterations in growth. During growth and cell elongation the primary cell wall (PCW) is a centre of high activity where the co-ordinated action of a set of enzyme families enables the cell wall to extend without losing its vital strength (Cosgrove, 1999). Cell elongation responds very quickly to stress-related signals (Le et al., 2001) and stresses affect enzymatic activity in the cell wall (Devos et al., 2005).

Research on primary cell wall composition and dynamics was in the past mainly focused on seed plants. Recently, interest arose in the understanding of cell wall evolution, as this can provide new insights into how it arose, evolved, and functions. A comparative study of the cell wall composition including primitive tracheophytes showed that the PCW composition and its associated tannin content have been adapted and modified throughout land plant evolution. By contrast, a frame of cellulose–xyloglucan networks was found in all land plants. Bryophytes were found to contain relatively small amounts of xyloglucan, whereas this hemicellulose became more abundant in the vascular plants, from the lycopodiophytes down to the non-gramineous angiosperms (Popper and Fry, 2003, 2004). These findings underline the importance of the cellulose–xyloglucan network in the PCW's function throughout evolution. Evolutionary studies concerning the cell wall-modifying enzymes, on the other hand, are still rather limited in number. Expansins, suggested to be involved in cell expansion by breaking the hydrogen bonds between the cellulose microfibrils and hemicelluloses, such as xyloglucans (Cosgrove, 2000), were shown to be present in all land plants (Shcherban et al., 1995; Hutchison et al., 1999; Kim et al., 2000; Schipper et al., 2002). They are part of an ancient multi-gene family that was already present before the divergence of plants into mono- and dicotyledons. Together these data suggest the presence of a well-conserved xyloglucan (hemicellulose)–cellulose related cell wall elongation mechanism that is deeply embedded in the evolution of vascular plants.

Besides expansins, xyloglucan endotransglycosylase/hydrolases (XTHs) act upon the cellulose–xyloglucan network. They break donor xyloglucan chains and rejoin them to acceptor xyloglucan chains (XET action, EC 2.4.1.207 [EC] ) or to water (XEH action, EC 3.2.1.151 [EC] ) (Fry et al., 1992; Rose et al., 2002). XET action possibly allows the cellulose microfibrils to move apart and/or past another, driven by turgor pressure (see Fig. 1 in Thompson and Fry, 2001). In contrast to expansin action, XET action itself maintains the integrity of the cellulose–xyloglucan network. This gives XTHs a large cell wall-modifying potential. Evolutionary data concerning the presence and characteristics of XTHs in early vascular plants are lacking. Still these data are crucial in the understanding of the evolution and complexity of the cell wall structure and dynamics and are, therefore, another important step towards the understanding of plant growth in general.


Figure 1
View larger version (65K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Tissue print followed by a XET assay of the Selaginella sporophyte. A Selaginella sporophyte including shoot and root tissue was used to perform a tissue print (sporophyte on the test paper in (A), followed by a XET assay of the print (B)). XET activity is seen as orange fluorescence in the root elongation zone (see arrow) and the microphylls (example is encircled), i.e. the sites of elongation in both tissues.

 
In our previous work, XET action was detected in vivo in the root elongation zone of different land plants from the most complex angiosperms down to lycopodiophytes (Vissenberg et al., 2000, 2003). To compare the characteristics and the functions of XTH proteins from primitive plants with those of seed plants, Selaginella, a representative of the most primitive vascular land plants (lycopodiophytes), was used. The lycopodiophytes diverged from the plant-lineage leading to seed plants, e.g. Arabidopsis (euphyllophytes), about 400 million years ago (Pryer et al., 2001). Such a long period gives evolution ample opportunity to have a drastic impact on DNA sequences and the resulting proteins. In this work, the detection and cloning of a functional XTH from the club moss Selaginella kraussiana, Sk-XTH1, are presented and its DNA and protein sequence characteristics are compared with that of angiosperm XTHs. Sk-XTH1 is heterologously expressed in Pichia pastoris and the enzymatic properties of the purified recombinant protein have been determined.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Supplementary data
 References
 
Plant material
Selaginella kraussiana sporophytes were obtained commercially (Jansen bvba, Edegem, Belgium).

Tissue print
A xyloglucan test paper, prepared as described by Fry (1997), was soaked in 25 mM MES buffer. Part of a Selaginella kraussiana plant was pressed onto the moist paper for 16 h while being sealed in an acetate envelope and incubated at 30 °C. The test paper was washed and visualized as described before (Fry, 1997).

RNA isolation
Roots and shoots (100 µg) of the club moss Selaginella kraussiana were homogenized in liquid nitrogen in the presence of 150 µl homogenization buffer (100 mM TRIS–HCl pH 7.5, 4 M guanidine isothiocyanate, 1% N-lauryl sarcosine, 1% polypyrrolidone; 1% ß-mercaptoethanol was added prior to use and an equal volume of extraction buffer (100 mM TRIS–HCl pH 7, 2 M NaCl, 25 mM EDTA, 2% cetyl triethyl ammonium bromide preheated at 70 °C) was added. The sample was mixed by inversion and kept at 70 °C in a warm water bath for 5 min. A phenol:chloroform:isoamylalcohol (25:24:1 by vol.) separation was followed by one chloroform:isoamylalcohol (24:1, v:v) separation step. The RNA in the supernatant was precipitated overnight at 4 °C in the presence of 1/4 volume 10 M LiCl. The pellet was washed with 70% ethanol, air-dried, and resuspended in highly purified water.

Cloning of a Selaginella-XTH gene
Five µg of total RNA was reverse transcribed using Superscript II RNase H-Reverse Transcriptase (Invitrogen) according to the manufacturer's instructions. The resulting cDNA served as template in a PCR reaction using degenerate primers (designed on conserved sequences from aligned Arabidopsis thaliana, Hordeum vulgare, and Festuca pratensis XTH sequences). Primers 5'-ATCGACWTYGAGTTCMTGGG-3' and 5'-GCCGYSCTGCGTKGCCCAGT-3' were used to amplify a potential XTH cDNA fragment in a touch-down PCR [(56 °C)x10; (46 °C)x15]. A 300 bp cDNA sequence was amplified and sequenced to confirm the presence of the proposed catalytic site of angiosperm XTHs, DEIDFEFLG or a variant. The 3'-end of the cDNA was obtained by 3'-RACE using a gene-specific primer (RACE1 5'-ATCGACTTTGAGTTCCTGGGC-3') and an oligodT26. The 5'-end was identified using the 5'-RACE System for Rapid Amplification of cDNA Ends (Invitrogen, ver. 2.0) with 5 µg total RNA as template and the following gene-specific primers: 5RACE1 (5'-CTGCGTGGCCCAGTCGTCG-3') and 5RACE2 (nested; 5'-CTCGAGTAGACGCCCATGGG-3'). The amplified products were cloned into the pGEM-T Easy vector (Promega) and sequenced. From the different partial sequences, the full-length cDNA was deduced and designated Sk-XTH1 (accession no. AY580314).

Specific primers, based on the known cDNA sequence, were used to amplify the full gDNA-sequence of Sk-XTH1.

In silico sequence analysis
The physico-chemical parameters of the full-length deduced amino acid sequence of Sk-XTH1 were determined (http://au.expasy.org/tools/protparam.html) as well as functional sites using the Eukaryotic Linear Motif search engine (http://elm.eu.org).

The amino acid sequence was aligned with those of other XTHs and a tree was generated using the Neighbor–Joining method in ClustalW (http://clustalw.genome.jp). The GenBank accession numbers of the dicots XTHs are: Arabidopsis AtXTH1, At4g13080; AtXTH2, At4g13090; AtXTH3, At3g25050; AtXTH4, At2g06850; AtXTH5, At5g13870; AtXTH6, At5g65730; AtXTH7, At4g37800; AtXTH8, At1g11545; AtXTH9, At4g03210; AtXTH10, At2g14620; AtXTH11, At3g48580; AtXTH12, At5g57530; AtXTH13, At5g57540; AtXTH14, At4g25820; AtXTH15, At4g14130; AtXTH16, At3g23730; AtXTH17, At1g65310; AtXTH18, At4g30280; AtXTH19, At4g30290; AtXTH20, At5g48070; AtXTH21, At2g18800; AtXTH22, At5g57560; AtXTH23, At4g25810; AtXTH24, At4g30270; AtXTH25, At5g57750; AtXTH26, At4g28850; AtXTH27, At2g01850; AtXTH28, At1g14720; AtXTH29, At4g18890; AtXTH30, At1g32170; AtXTH31, At3g44990; AtXTH32, At2g36870; AtXTH33, At1g10550; Nasturtium TmXTHa, X68254; TmXTHb, L43094; tobacco NtXTHa, AB017025; NtXTHb, D86730; Kiwifruit AdXTHa, L46792; Vigna radiata VrXTHa, AY841854; Vigna angularis VaXTHa, D16458; bean PsXTHa, AB015428; Cicer arietinum CaXTHa, AJ004917; rapeseed BrXTHa, AY834281; tomato (from http://labs.plantbio.cornell.edu/XTH/toma.html) LeXTH1 D16456; LeXTH2, AF176776; LeXTH3, LeXTH4, AF186777; LeXTH5, LeXTH6, LeXTH7, LeXTH8, AB036338; LeXTH9, LeXTH10, X82684; LeXTH11, X82685; LeXTH12, AF205069; soybean GmXTHa, L22162; beech FsXTHa, AJ130885; poplar PttXTH34, AAN87142; cauliflower BobXET16A, Q6YDN9; and for the monocots: Rice OsXTH1, AP003899; OsXTH2, AC136481; OsXTH3, AL606650; OsXTH4, AP003907; OsXTH5, AP004657; OsXTH6, AL606638; OsXTH7, AL662996; OsXTH8, AP004705; OsXTH9, AL606638; OsXTH10, AP005445; OsXTH11, AP005445; OsXTH12, AP005445; OsXTH13, AP005398; OsXTH14, AP005398; OsXTH15, AP004784; OsXTH16, AL662996; OsXTH17, AP004705; OsXTH18, AP005445; OsXTH19, AC113930; OsXTH20, AC025783; OsXTH21, AP003839; OsXTH22, AP004026; OsXTH23, AP005859; OsXTH24, AC120506; OsXTH25, AC018929; OsXTH26, AP004886; OsXTH27, AP079037; OsXTH28, AC118981; OsXTH29, AP005430; wheat TaXTHa, D16457; barley HvXTHa, X91659; HvXTHb, X93175; HvXTHc, X93174; HvXTHd, X91660; HvXTHe, X93173; maize ZmXTHa, U15781; ZmXTHb, AJ875021; Festuca pratensis FpXTHa, AJ295943; FpXTHb, AJ295944; FpXTHc, AJ295945. The names of the genes were changed according to the systematic nomenclature of the XTH genes (Rose et al., 2002), for example, TRUXET1G, L43094 was changed into TmXTHb in this case.

Southern analysis
Genomic DNA of Selaginella kraussiana was prepared as described in Lodhi et al. (1994). The resulting gDNA (20 µg) was digested with the indicated restriction enzymes (ClaI, NcoI, and XbaI; NheI, HindIII, and XbaI), fractionated by electrophoresis on 1% (w/v) agarose gels, and transferred to a Hybond-N membrane (Amersham, Arlington Heights, IL). Blots were hybridized with a 297 bp Sk-XTH1 DIG-labelled cDNA PCR product [5'-CTGCGTGGCCCAGTCG-3'; 5'-ATCGACCTCGACTTCCTGGGC-3'; (30x56 °C)] at 44 °C in DIG easyHyb hybridization buffer (Roche). The blot was washed twice in 2x SSC (0.3 M sodium chloride, 30 mM sodium citrate, pH 7.0) with 0.1% (w/v) SDS at room temperature for 5 min, followed by three washes in 0.1x SSC at 65 °C for 15 min.

For detection, the membrane was washed briefly in washing buffer, blocked for 1 h in blocking solution (DIG Wash and Block Buffer Set, Roche) and subsequently incubated with alkaline phosphatase-conjugated sheep anti-DIG antibody (1:10 000 in blocking solution, Roche) for 1 h. After two 15 min washes in washing buffer, the membrane was soaked in detection buffer (DIG Wash and Block Buffer Set, Roche) for 3 min. The chemiluminescent substrate, CSPD (Roche), was added drop-wise onto the surface of the membrane, which was then placed in a tightly sealed acetate envelope for 10 min at 37 °C. Signals could be detected after exposure on X-ray film ranging from 10 min to 1 h.

Construction of the expression vector
The Pichia pastoris expression system (Invitrogen) was used for the production of recombinant Sk-XTH1 protein. The cDNA of Sk-XTH1 was amplified without its signal peptide, but in-frame with the alpha factor secretion signal peptide of the pPICZ{alpha}A vector (Invitrogen), including the stop codon using the primers: Sk-XTH1NF 5'-GAATTCACGCCTTGCATTGCCGCG-3' and Sk-XTH1NR 5'-TCTAGAGCTCACGGAGCGCGCGCGCATTC-3' (10x53 °C) (20x58 °C), to introduce EcoRI and XbaI restriction sites at the 5' and 3' ends, respectively. The PCR product was TA-subcloned in pGEM-T, digested with EcoRI and XbaI and cloned into the pPICZ{alpha} vector (reading frame A, Invitrogen). A clone with an open reading frame for the {alpha}-factor sequence and the Sk-XTH1-coding region, including the stop codon, was sequenced to verify that no sequence errors were introduced.

Recombinant protein production in P. pastoris
The Sk-XTH1 expression vector and the empty pPICZ{alpha}A vector, used as a negative control for the expression analysis, were linearized with PmeI and used to transform Pichia pastoris (strain GS115) by electroporation (1800 V) according to the manufacturer's manual (EasySelectTM Pichia Expression Kit, Invitrogen). Multiple copy integrants were selected on YPDS (regeneration yeast peptone dextrose sorbitol) plates, containing elevated zeocin concentrations (<2000 µg ml–1). Colony-PCR confirmed the presence of the expression construct in the yeast genome. Recombinant yeast colonies were inoculated in 5 ml buffered glycerol-complex medium (BMGY), with 100 µg ml–1 zeocin (Invitrogen) and grown overnight (29 °C, 250 rpm). 100 µl of this culture was subsequently transferred to 100 ml BMGY and grown overnight under the same conditions, without selective antibiotic. When the culture reached OD600 {cong} 4, cells were harvested by centrifugation (3000 g, 5 min) and resuspended in BMMY (Buffered Methanol-complex medium) to OD600 {cong} 1, and divided in 4 1.0 l flasks covered with two layers of cheesecloth (22 °C, 200 rpm). The methanol concentration was kept at 1% by adding 100% methanol every 24 h. After induction for 3 d the yeast cells and culture media were separated by centrifugation (6000 g, 5 min), the supernatant was further analysed for the expression of active enzyme.

Protein purification
Sk-XTH1 was purified from the supernatant that was recovered from 400 ml of culture medium of methanol-induced P. pastoris, expressing Sk-XTH1. Proteins were precipitated from the supernatant at 4 °C, for 4 h with increasing amounts (25–55% w/v) of (NH4)2SO4. The precipitates were collected by centrifugation at 10 000 g for 30 min at 4 °C and dissolved in 50 mM sodium citrate buffer at pH 4.5. The different fractions were analysed on SDS–PAGE (using silver stain) and with a XET activity assay. 35–40% of (NH4)2SO4 was experimentally found to induce precipitation of most of the recombinant Sk-XTH1. This fractional precipitation was used as the first purification step.

After precipitation, the proteins were desalted and equilibrated in 50 mM sodium acetate buffer pH 4.5, 50 mM NaCl, 10% glycerol (=buffer A), using PD10 columns (GE Health care), according to the manufacturer's manual. A 5 ml Hi-trap SP-Sepharose column was equilibrated with 10 vols of buffer A. Prior to application, the desalted sample was diluted five times in buffer A, to lower the viscosity. The sample was loaded onto the column using a syringe and washed with five column volumes of buffer A. The bound proteins were subsequently eluted with five column volumes 50 mM sodium acetate buffer pH 4.5, 300 mM NaCl, and 10% glycerol. Fractions containing Sk-XTH1 were pooled and concentrated using centricon plus-20 with an ultra cell-30 membrane (Millipore). Protein expression and purification steps were analysed on SDS–PAGE using an XcellTM Mini Cell (Novex, San Diego, CA, USA) followed by staining with SimplyBlue or SilverQuest (Invitrogen) and the XET dot spot activity assay. The purified protein was sequenced in the CEPROMA centre at the University of Antwerp (Belgium). Protein concentrations were determined spectrophotometrically according to the Bradford method (Bradford, 1976), using bovine serum albumin as a standard.

XET dot blot activity assay
The recombinant Sk-XTH1 was assayed for xyloglucan endotransglycosylase (XET) activity using the dot-blot assay as described by Fry (1997). In this assay XET activity is defined as the ability to transfer a sulphorhodamine-labelled xyloglucan oligosaccharide (XGO-SR, = acceptor substrate) to the reducing end of a non-fluorescently labelled high molecular weight xyloglucan (tamarind xyloglucan = donor substrate). Five µl of the purified Sk-XTH1 was spotted on xyloglucan-containing test paper together with XGO-SRs. As a control for the specificity of the assay, a parallel experiment was carried out using a trisaccharide-SR (=non-XET substrate). The test paper was spotted in a cold room and firmly sealed in an acetate envelope to prevent drying, and was then incubated at 28 °C overnight. XET activity catalyses the transfer of the fluorescent acceptor-substrates to the non-labelled xyloglucans attached to the filter paper. Non-reacted fluorescent xyloglucan oligosaccharides were washed away in (1:1:1 by vol. 90% formic acid:ethanol:water) for 4 h followed by a 5 min wash in water. The remaining fluorescence on the test paper upon UV illumination is hence indicative of XET activity. Pictures were taken using an Olympus C-5050 ZOOM digital camera with identical settings (1/10 shutter time, F 2.0 diaphragm) to allow comparison of the resulting fluorescence.

In order to study the pH-profile of the Sk-XTH1 XET-activity, the purified enzyme was buffer-exchanged to a sodium citrate buffer set ranging from pH 3 to 5.5 and a phosphate buffer system from pH 6 to 8. The samples were spotted onto XGO-SR test paper as described above to test for the presence of XET activity.

For the determination of the temperature optimum of recombinant Sk-XTH1, dot-blots were prepared as described above, spotted at 4 °C, and immediately transferred to different temperatures (4 °C to 60 °C).

Deglycosylation
A range from 0–1000 units of Endo glycosidase H (New England Biolabs) was added to 50 µl of the purified Sk-XTH1 (1µg µl–1) in 50 mM sodium acetate buffer pH 4.5, 50 mM NaCl, 10% glycerol (final volume of 55 µl). The samples were incubated at 30 °C for 3 h to allow optimal deglycosylation conditions. The stability of the recombinant protein was monitored, comparing two samples without Endo H at 4 °C and at 30 °C. All samples were assayed for XET activity on a test paper overnight at 30 °C. The progress of deglycosylation of Sk-XTH1 was studied on SDS–PAGE.

Reduction
In a 50 µl Sk-XTH1 sample (1µg µl–1) DTT was added to a final concentration of 500 mM and kept at 4 °C for 2h. XET activity was assayed using the dot-spot assay, as described above.

Colorimetric XEH assay
XEH activity was studied in detail by the iodine staining method (Kooiman, 1960; Sulova et al., 1995). 200 µl of the recombinant Sk-XTH1 (250 ng µl–1), dissolved in 50 mM sodium acetate buffer, pH 6 was added to 200 µl 0.2% high molecular weight xyloglucan, 0.04% xyloglucan heptasaccharide (Megazyme) in 50 mM sodium acetate buffer at pH 6 together with 6 mM sodium azide. To measure the contribution of substrate hydrolysis, control samples without the xyloglucan heptasaccharides were measured in parallel. All samples were incubated at 30 °C. Samples were withdrawn at appropriate time intervals; the reaction was terminated by the addition of 1/2 volume 1 M HCl. All reactions were repeated in buffered conditions at pH 4.5 and pH 7.5.

Using this method XEH activity is expressed in terms of the decrease in absorbance at OD630nm (Elx808i, Ultra Microplate reader, Bio-tek instruments) of the xyloglucan–iodine complex. To 20 µl of each sample, 55 µl of 20% Na2SO4 and 14 µl potassium tri-iodide (KI3) reagent were added. The samples were pipetted into a 96-well plate (polystyrene, clear wells in white matrix, Wallac, Finland) and incubated in the dark at room temperature for 30 min, to allow colour development. Each sample was compared with one in which buffer replaced the xyloglucan solution. XEH-activity was expressed as the difference in A630 at the different time points between samples with and without xyloglucan heptasaccharides. Control reaction mixtures in which an equal concentration of Aspergillus cellulase (Sigma) replaced the recombinant Sk-XTH1, were performed in parallel. All assays were carried out in duplicate.

Reducing sugar assay
Using the BCA (disodium 2,2-bicinchoninate) reducing sugar assay (Henriksson et al., 2003) xyloglucan hydrolysis was measured as the accumulation of reducing ends in 0.2% high molecular weight xyloglucan in 50 mM sodium acetate buffer, pH 6, 6 mM sodium azide with and without Sk-XTH1. Samples were withdrawn at appropriate time intervals; the reaction was terminated by the addition of 1/2 volume 1 M HCl. The same assay was repeated in buffered conditions at pH 4.5 and pH 7.5.

To 5 µL of each sample 45 µl of highly distilled water and 50 µl of the BCA reagent was added. The samples were pipetted into a 96-well plate (polystyrene, clear wells in white matrix, Wallac, Finland) and incubated at 80 °C for 30 min. XEH activity was measured as the increase of A550 (Elx808i, Ultra Microplate reader, Bio-tek instruments). Reaction mixtures, in which an equal concentration of cellulase replaced the recombinant Sk-XTH1, were performed in parallel. Xyloglucan heptasaccharides were used to generate a standard curve over the concentration range 0–200 µM. All measurements were performed in duplicate.

Thin layer chromatography
The breakdown of xyloglucan by Sk-XTH1 and cellulase at different pH was visualized on TLC. TLC was performed using (5:2:3 by vol.) propan-1-ol:nitromethane:water as the solvent system. The saccharides were visualized by briefly dipping the TLC plate in MeOH, 3% H2SO4, 0.3% naphthyl ethylene diamine dihydrochloride and heating at 120 °C. A mixture of xyloglucan heptasaccharides and high molecular weight xyloglucans were spotted in parallel.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Supplementary data
 References
 
XET activity in Selaginella
So far, XET action of XTHs has been documented for the root elongation zone of Selaginella (Vissenberg et al., 2003). A tissue print of the root and shoot system of Selaginella (Fig. 1A) followed by a XET assay demonstrated the presence of XET activity in both parts of the sporophyte (Fig. 1B). XET activity was visible as a small fluorescent spot corresponding to the elongation zone of the root (see arrows). Where the microphylls were pressed onto the blot paper bright XET signals were observed (one example is encircled).

Cloning and in silico analysis of Sk-XTH1
In former work XET activity was shown to be ubiquitous in roots of vascular land plants, including the primitive Selaginellales (Vissenberg et al., 2003). In this study a XTH cDNA-fragment was amplified by RT-PCR from the Selaginella kraussiana-sporophyte using different sets of degenerate primers. These were based on nucleotide sequences conserved among known XTHs. The corresponding full-length cDNA (849 bp, accession no. AY580314) was obtained by 3'- and 5'-RACE PCRs (Fig. 2A). The amino acid sequence was deduced and analysed in silico. The Selaginella sequence aligned with a number of putative XTHs from the GH16 group and was therefore tentatively named Sk-XTH1 according to the proposed unified nomenclature system for XTH-genes (Yokoyama and Nishitani, 2001; Rose et al., 2002).


Figure 2
View larger version (84K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 cDNA, gDNA and gene structure of Sk-XTH1. (A) The cDNA- (849 bp) and gDNA-sequence (1024 bp) of Sk-XTH1. (B) In a schematic diagram of the Sk-XTH1 gene structure, the white boxes represent exons, lines introns, the black box the putative signal peptide sequence and the grey box the catalytic site in the translated protein.

 
The full gDNA sequence of Sk-XTH1, from start to stop codon (3' and 5' UTRs were not studied) was identified (Fig. 2A). Figure 2B gives a schematic diagram of the gDNA structure. Three small introns, with a length of 58, 61, and 68 bp (bold in Fig. 2A), separate four exons (boxes with lengths of 166, 92, 206, and 385 bp, respectively). The gene structure contains a secretion signal (in black) at the start of the first exon and the catalytic site (in grey) at the beginning of the third exon. This gene structure is characteristic for XTHs belonging to the phylogenetic Group I (Yokoyama and Nishitani, 2001).

In order to find the copy number of Sk-XTH1-coding genes, a Southern blot was performed on genomic DNA that was extracted from Selaginella sporophytes (Fig. 3). Digestion was done with ClaI, NcoI, and XbaI (lane a) and with NheI, HindIII, and XbaI (lane b), respectively. A Southern blot was performed on the digested genomic DNA using a Sk-XTH1-specific probe, yielding a single restriction band, indicating that Sk-XTH1 is probably encoded by a single gene. In a control experiment, the gDNA was digested with EcoRI, XbaI, and XhoI, where XhoI cuts into the sequence recognized by the Sk-XTH1-specific probe. No band was detected after Southern analysis (results not shown).


Figure 3
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Genomic Southern blot analysis of Sk-XTH1. Genomic DNA of Selaginella sporophytes was digested completely with ClaI, NcoI, and XbaI (lane a) and with NheI, HindIII, and XbaI (lane b), respectively. The digested DNA was separated using agarose gel electrophoresis, blotted onto a Hybond-N membrane and hybridized with a 297 bp Sk-XTH1-probe. In each lane, one fragment with sizes of ±7.4 and 5.5 kb were present suggesting Sk-XTH1 is encoded by a single gene. Molecular size markers are on the right.

 
The presence of conserved amino acids and motifs was studied in an alignment with PttXTH34 (Geisler-Lee et al., 2006), a well-characterized XTH from Populus tremulaxtremuloides that was formerly known as PttXET16A (Bourquin et al., 2002), BobXET16A, EXGT-A1, XTR9, TmNXG1, TCH4, Meri5, and LeXET2 (Fig. 4). The Eukaryotic Linear Motif Resource for Functional Sites in Proteins (http://elm.eu.org/) predicted a putative secretion signal peptide, rich in hydrophobic amino acids, in the N-terminal region. Cleavage of this predicted signal would probably occur after the 21st amino acid (Fig. 4, small letters; von Heijne, 1986). A variant of the functional site DEIDFEFLG, conserved among most XTHs studied in seed plants (Nishitani, 1997), but also present in other family 16 glycoside hydrolases (Michel et al., 2001; Strohmeier et al., 2004) is present between positions 89 and 97 (Fig. 4; bold letters). The predicted catalytic site differs in only one amino acid (DEIDLEFLG) compared with the consensus sequence; a phenylalanine (position 105) is replaced by a leucine. This site is, as in most other XTHs studied, immediately followed by a potential site for N-linked glycosylation (N-X-T/S; Shakin-Eshleman et al., 1996) at positions 98–100 (Fig. 4, underlined). Certain arginine, aspartic, and glutamic acid residues, conserved at well-defined positions in some angiosperm XTHs, are thought to play an important role in the three-dimensional structure of the active cleft and the positioning of the donor and acceptor substrate, as found in poplar PttXTH34 (E208-R116, E208-R204, D88-R116; Fig 4, orange letters) (Johansson et al., 2003, 2004; Kallas et al., 2005). Most of these amino acids, however, are, as in TCH4 (AtXTH22) and AtMeri5 (AtXTH24) of Arabidopsis, not conserved in Sk-XTH1 (Fig. 4, encircled). D88 is the only amino acid involved in additional electrostatic interactions in the poplar XTH that is also present in Sk-XTH1 (D80). At position 201 the hydrophobic valine residue in Sk-XTH1, replaces the negatively charged glutamic acid residue found at the corresponding position 208 in PttXTH34. In addition, both interacting arginines at positions R116 and R204 in PttXTH34 are replaced by a threonine at position 109 and a glutamine residue at position 197 in Sk-XTH1. The amino acids involved in N-glycan interaction are conserved or substituted to homologue amino acids in Sk-XTH1 (Fig 4, red letters). By contrast, all the amino acids involved in the recognition and binding of the xyloglucan substrates, which are strongly conserved amongst angiosperm XTHs, are present in Sk-XTH1 (Fig. 4, black and grey boxes).


Figure 4
View larger version (69K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Alignment of the Sk-XTH1 and all characterized angiosperm XTH amino acid sequences. TmNXG1, displaying mixed (XET/XEH) activity, is marked with an asterisk, PttXTH34, BobXET16A, AtXTH4, –14, –22, –24, and LeXET2 display exclusively XET activity. The proteins whose functionality is dependent on the presence of the N-glycan are shown in blue, the deglycosylation-independent ones in green. The signal peptide sequences of the proteins are presented in small letters, the conserved catalytic site in bold and the N-glycosylation site is underlined. The amino acid interacting with the catalytic site on the opposing beta-strand is marked in yellow. The acceptor binding loop of all proteins is emboxed, the amino acids possibly involved in the formation of additional salt-bridges are shown in orange, those interacting with the N-glycan are marked in red. Conserved cysteines are represented in purple. Amino acids involved in the recognition and binding of the xyloglucan substrates are shaded respectively in grey and in black.

 
Together these in silico data thus suggest that the Sk-XTH1-cDNA encodes a novel putative XTH from Selaginella.

Recombinant expression and functional analysis of Sk-XTH1
In order to verify whether Sk-XTH1 encodes a fully functional XTH protein exhibiting XET (EC 2.4.207) and/or XEH (EC 3.2.1.151 [EC] ) activity, recombinant Sk-XTH1 was produced using the Pichia pastoris heterologous expression system (Invitrogen). The cDNA, encoding the mature protein, was cloned into the pPICZ{alpha}A expression vector in-frame with the N-terminal {alpha}-factor secretion signal. The construct was then transformed into the Pichia pastoris GS115 strain by electroporation. SDS-PAGE showed the presence of a 35 kDa protein in the ammonium sulphate-precipitated culture medium of yeast transformed with the Sk-XTH1 expression vector (Fig. 5, lane A), but not in the medium from control yeast transformed with an empty vector.


Figure 5
View larger version (69K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Purification of recombinant Sk-XTH1 from methanol-induced P. pastoris culture medium visualized using SDS–PAGE and silver stain. (A) Total protein fraction present in the supernatant after ammonium sulphate precipitation (55%). (B–H) Fractional ammonium sulphate precipitation (25–55% w/v in 5% steps). 35–40% of (NH4)2SO4 (lanes D, E) was experimentally determined to induce precipitation of most of the recombinant Sk-XTH1. This fractional precipitation was used as the first purification step. Both fractions were pooled, desalted (I) and further purified on a SP-Sepharose column (K, flow through; L, wash; M, elution). Fractions containing Sk-XTH1 were pooled and concentrated using centricon plus-20 with ultra cell-30 membrane (Millipore). This purification protocol yielded two glycoforms of Sk-XTH1 (29 and 35 kDa).

 
A recombinant colony producing high levels of Sk-XTH1 was selected and used to scale up protein production. Sk-XTH1 was subsequently purified from the supernatant (Fig. 5, lanes B–M), resulting in two protein bands on SDS PAGE (Sk-XTH1A, 35 kDa; and Sk-XTH1B, 29 kDa). Both proteins were sequenced and shown to be identical in their amino acid sequence. The difference in length is due to alternative glycosylation as was reported earlier by Kallas et al. (2005).

The XET activity of both purified recombinant glycoforms was tested using the fluorescence dot-blot assay (Fry, 1997; Fig. 6). Spotting the purified Sk-XTH1 onto the XET test paper resulted in the incorporation of fluorescence (Fig. 6A). As a control for the specificity of the assay, fluorescently labelled non-XET acceptor substrate was used, yielding no fluorescence (Fig. 6B).


Figure 6
View larger version (7K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 XET assay of recombinant Sk-XTH1. XET activity is defined as the ability to transfer sulphorhodamine-labelled xyloglucan oligosaccharides (XGO-SR=acceptor substrate) to the reducing end of a non-fluorescent labelled high molecular weight xyloglucan (=donor substrate). Spotting the purified recombinant Sk-XTH1 on XET test paper resulted in the incorporation of a fluorescent spot (A). As a control for specificity of the assay, fluorescent-labelled non-XET substrate was used, yielding no fluorescence when Sk-XTH1 was spotted (B).

 
The effect of pH and temperature on XET activity of recombinant Sk-XTH1 were studied using the dot blot assay. Sk-XTH1 acts within a broad pH range between 4.5 and 7.5 (Fig. 7A) and a temperature range between 4 °C and 37 °C. At temperatures of 50 °C and higher, activity clearly decreases (Fig. 7B).


Figure 7
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 pH and temperature profile of XET activity of Sk-XTH1. Sk-XTH1 was assayed for XET activity under different pH (A) and temperature conditions (B). The enzyme is active in a broad pH range from 4.5 to 7.5 and in the temperature range from 4 °C to 37 °C.

 
Using the iodine staining method (Kooiman, 1960; Sulovà et al., 1995; Suda et al., 2003) the presence of XEH activity was studied in more detail at different pH points (4.5, 6, and 7.5). A small decrease in A630 was observed, incubating high molecular weight xyloglucan with Sk-XTH1. Including oligo xyloglucans in the reaction mixture resulted in fast drop in A630, as a result of oligo xyloglucan transglycosylation (see supplementary data at JXB online). This small decrease seen in the reactions without added oligos is therefore most likely to be the result of the formation of enzyme–substrate complexes, rather than XEH activity. The BCA reducing assay and TLC analysis confirmed the absence of detectable XEH activity at different pH suggesting that Sk-XTH1 preferably accepts another xyloglucan and not water, as an acceptor substrate (see supplementary data at JXB online).

Effect of deglycosylation of Sk-XTH1 on XET activity
The presence, importance, and necessity of the N-linked sugar modification near the active site of Sk-XTH1 was tested by deglycosylation. Endoglycosidase H cleaves the chitobiose core of high-mannose-type N-glycans to leave a single asparagine-linked N-acetylglucosamine residue. Sk-XTH1 was incubated with Endoglycosidase H for 3 h at 30 °C. The progress of deglycosylation was analysed on SDS-PAGE, followed by silver staining (Fig. 8A) and the effect on the XET activity was assayed (Fig. 8B).


Figure 8
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 The effect of deglycosylation by endoglycosidase H on the molecular weight (A) and XET activity (B) of recombinant Sk-XTH1 and the effect of reduction on XET activity (C). (A) Both glycoforms of Sk-XTH1 (lane 1) were separated by SDS–PAGE and silver stained. Deglycosylation of this sample by increasing amounts of Endoglycosidase H (1–1000 units) caused a shift downwards of the two protein bands. (B) In parallel, the samples were assayed for XET activity. In the last lane deglycosylation is near to complete, resulting in a drastic loss of function, indicating that N-glycosylation is needed for the stability and/or functionality of the protein. In control experiments at 4 °C and 30 °C, no Endo H was included. (C) Sk-XTH1 protein was reduced in the presence of 500 mM DTT and assayed for XET activity (2). The control sample was kept under the same conditions without DTT (1).

 
SDS-PAGE shows the effect of deglycosylation by Endo H on the mobility of the proteins. The proteins shift down on the gel after increasing deglycosylation in the following lanes (1–1000 units). Accordingly XET-activity (seen as fluorescence intensity) decreases with increasing Endo H units added.

Effect of reduction of Sk-XTH1 on XET activity
The in silico analysis of Sk-XTH1 revealed four C-terminal cysteines (Fig. 4A, boxed amino acids) that form inter- or intra-molecular disulphide bonds, thereby stabilizing the three-dimensional structure of the C-terminal region of the enzyme. To monitor the importance of the disulphide bonds in Sk-XTH1, the protein was reduced and a XET-assay was done. Incubation of Sk-XTH1 with DTT for 2 h at 4 °C resulted in a clear decrease in XET-activity (Fig. 8C).

Phylogenetic tree
The evolutionary relationship of Sk-XTH1 with other known XTHs was studied, generating a phylogenetic tree using full-length protein sequences of XTHs from different plant species. To increase the readability of the figures, two separate trees were made, one including Sk-XTH1 and XTHs from dicots (Fig. 9A) and one including Sk-XTH1 and XTHs from monocots (Fig. 9B). Arabidopsis XTH members were classified into three distinct phylogenetic groups (Group I, red, Group II, green, and Group III, blue; Yokoyama and Nishitani, 2001; Rose et al., 2002). Also for monocots it is proposed that three different groups should be considered, although Yokoyama and Nishitani (2004) claimed that Group I and II from rice were difficult to separate. However, the division into three groups becomes evident when other monocots are included in the alignment. In both cases Sk-XTH1 was, accordingly with its gene structure classified in Group I, a group known to share a high level of sequence identity among different species.


Figure 9
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 9 Representation of the evolutionary relationship of Sk-XTH1 to XTHs from modern plant species using a phylogenetic tree. The deduced amino acid sequence of Sk-XTH1 was aligned with 97 XTH sequences from other plant species in ClustalW and visualized with TreeView. Details and GenBank accession numbers can be found in the Materials and methods; the boxed proteins are biochemically characterized. (A) Evolutionary relationship of Sk-XTH1 to known dicot XTHs. The XTH members are classified into three groups based on the classification for Arabidopsis. Sk-XTH1 is located in Group I. (B) Evolutionary relationship of Sk-XTH1 to known monocot XTHs. The XTH members are classified into three groups based on the classification when Arabidopsis was included. Sk-XTH1 is located in Group I.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Supplementary data
 References
 
In previous work, xyloglucan endotransglycosylase (XET) action was localized in the elongation zone of roots of different vascular land plants, including the most primitive ones that persist until today (Vissenberg et al., 2003). In this work, Sellaginella was revisited and XET activity was also localized in the shoot of this spore plant, more specifically at the sites of microphyll development. The data suggest a correlation between XET activity and cell elongation in lycopodiophytes. This group of organisms diverged from the plant-lineage that gave rise to the mono- and dicotyledons more than 400 million years ago (Pryer et al., 2001). The Selaginellales lack true leaves, but have microphylls and are therefore considered as intermediates between non-vascular plants (algae and bryophytes) and all other vascular plants. The presence of XET activity in these plants and its correlation with elongating cells thus underlines the intrinsic role for XTHs in the evolutional development of an efficient cell wall elongation and restructuring mechanism in vascular plants.

Therefore the isolation of Sk-XTH1, a XTH from the lycopodiophyte Selaginella, and the functional analysis of the recombinantly expressed Sk-XTH1 (in the yeast Pichia pastoris) was undertaken. This approach allowed the identification of the main or basic features of XTHs that were maintained during evolution and for them to be distinguished from those that were changed without affecting the proteins functionality.

Using the RACE-technique based on conserved sequences among known XTHs, a cDNA encoding a putative XTH was amplified, named Sk-XTH1 (according to an agreement made at the 9th Cell Wall Meeting held in Toulouse, France, 2001), and the amino acid composition of the resulting protein was deduced. The comparative in silico analysis of the DNA- and protein sequence of Sk-XTH1 and that of present angiosperm XTHs revealed that different aspects, described in modern XTHs, originated before the diversification of vascular plants.

A hydrophobic amino terminus was predicted to function as a secretion signal peptide that allows the newly formed enzyme to be deposited into the cell wall. Sk-XTH1 contains a variation of the DEIDFEFLG motif, conserved among the GH16 family, which comprises the xyloglucan endotransglycosylases, 1,3 ß-glucanases, and 1,3-1,4-ß-glucan-hydrolases in group B and ß-agarases, {kappa}-carrageenases, and 1,4 ß-glucanases in group A. Chrystallographic analysis of both groups (Keitel et al., 1993; Hahn et al., 1995a; Michel et al., 2001; Strohmeier et al., 2004) as well as site-directed mutagenesis (Juncosa et al., 1994; Hahn et al., 1995b) have shown that both glutamic acid residues in the conserved motif act as the nucleophile and general base involved in the cleavage of the ß-glycosyl linkages. Recent work of Campbell and Braam (1998) and Johansson et al. (2003, 2004) confirmed the presence and functional necessity of these residues in XTHs, as cleavage of a ß-glycosyl linkage is the first step of both XET- and XEH actions of the group B XTHs. In agreement with these data both glutamic acid residues are present in Sk-XTH1, which was shown to display XET-activity. Variation of the other amino acids within the catalytic site appears to be limited in Sk-XTH1, as seen in other XTHs which were shown to be functional (Campbell and Braam, 1998; Catalá et al., 2001; de Silva et al., 1993; Henrikkson et al., 2003; Kallas et al., 2005) (Fig. 4, bold letters). This strong conservation might be crucial for the formation and stability of the enzyme itself and/or of the enzyme–substrate complex. Although the catalytic sites of group A and B GH16 glycoside hydrolases differ in the number of amino acids between both the glutamic acid residues (Group A: ExxxxE; Group B: ExxxE), their catalytic residues can be exactly superimposed (Michel et al., 2001). The extra amino acid present in GH16A enzymes forms a ß-bulge, which is compensated by the presence of a smaller interacting amino acid in the opposing ß-strand, compared with group GH16B enzymes. These data are indicative for the strong conservation of the positioning of the catalytic residues within the active cleft. In most of the characterized XTHs (PttXTH34, AtXTH22 (=TCH4), AtXTH24 (=Meri5), AtXTH14 (=XTR9) and LeXET2) the composition of the catalytic cleft as well as the interacting amino acid on the opposing ß-strand is conserved. The first phenylalanine residue of the catalic site is here thought to interact with a phenylalanine residue in the opposing strand (Michel et al., 2001) (Fig. 4, Catalytic site in bold letters, interacting F is marked in yellow). In Sk-XTH1 the first phenylalanine residue of the catalytic site is replaced by a leucine (DEIDFEFLG->DEIDLEFLG). This substitution to a smaller amino acid might be in accordance with the presence of a slightly bigger interacting amino acid (Y142, Fig. 4, marked in yellow), present in the opposing ß-strand. An analogous amino acid substitution in the active cleft (DEIDFEFLG->DEIDIEFLG) and opposing ß-strand (F->Y) is seen in NXGI from Tropaeolum majus, a XTH that displays both XET and XEH activity (de Silva et al., 1993; Fanutti et al., 1993). This aspect, however, needs be studied in more detail since both EXGT and BobXET16A have a (F->Y) substitutions without according changes in their catalytic site. More XTHs therefore need to be analysed in silico and checked for functionality, for example, by recombinant expression followed by XET- and XEH assay as presented in this study.

A potential N-linked glycosylation site follows the catalytic site of Sk-XTH1, as is the case in most angiosperm XTHs. This modification, shown to be crucial for the XET function of XTH22 and XTH24 in Arabidopsis (Campbell and Braam, 1998) was also crucial for the XET function of Sk-XTH1. By contrast, nearly complete deglycosylation of poplar PttXTH34 and cauliflower BobXET16A maintained most of its XET activity (Henriksson et al., 2003; Kallas et al., 2005). Recently Johansson et al. (2004) proposed that certain amino acids interact in PttXTH34, thereby stabilizing the three-dimensional structure of the active enzyme. These salt bridges, that stabilize the positioning of the acceptor loop towards the catalytic site in the poplar XTH (D88-R116, E208-R116, and E208-R204; Johansson et al., 2004), are missing in the club moss XTH, since the aspartic acid (D80) residue is the only one conserved. Analogue substitutions are seen in AtXTH14 and TmNXG1, both enzyme functions, however, survive deglycosylation, in contrast to Sk-XTH1. Kallas et al. (2005) proposed the presence of alternative electrostatic interactions between E197 and K202 (Fig. 4, according to Sk-XTH1 numbering) in the acceptor binding loop. Since these interactions are also absent in Sk-XTH1, the stabilizing amino acid interactions (N111, Y140, and Q118 of Sk-XTH1) with the N-glycan could hence be more important in the maintenance of the three-dimensional structure of the active cleft in Sk-XTH1. Disruption of the stabilizing sugar interactions therefore causes, in contrast to the poplar and cauliflower XTH, loss of XET activity, as the 3-D structure of the catalytic site is probably changed or lost. Although two of the three sugar-interacting amino acids are substituted in Sk-XTH1, compared with PttXTH34, the replacing amino acids (Q119->N112; R204->Q197) most likely allow homologue interactions with the N-glycan. Possibly the formation of additional salt bridges arose later in plant evolution and became an advantageous feature to maintain functional, stable XTHs.

Although the structural features of GH16 hydrolases and XTHs are very similar, a notable difference is present in the C-terminal region. As in angiosperm XTHs, four cysteines, near the C-terminus of Sk-XTH1, have the potential to form two disulphide bridges. These are thought to stabilize a (XET)-C terminal region (pfam06955), unique to XTHs and possibly involved in the recognition and guidance of xyloglucans to the active site (Johansson et al., 2004). Reduction of these disulphide bonds within Sk-XTH1 indeed lowered the XET activity of the protein. In sugar-binding enzymes, aromatic residues establish van der Waals interactions with oligo and/or polysaccharides, thereby keeping the substrates near the catalytic site of the enzyme. Several of these amino acids are shared with endoglucanases and are also present in the putative (XET)-C extension region of Sk-XTH1 (Fig. 4, shaded black). Some, on the other hand, are unique to XTHs and are thought to be involved in binding of the acceptor substrate (e.g. Y185 and Y269 in Sk-XTH1 (Fig. 4, shaded grey), Y192 and Y272 in PttXTH34). In addition to the other highly conserved XTH motifs, all characteristics of a C-terminal extension that distinguish XTHs from other sugar-modifying enzymes were thus present before the divergence of vascular plants.

Taken together, the strong conservation of these motifs over at least 400 million years of plant evolution makes them key elements to determine and predict XTH identity.

Purification and functional characterization of different XTHs of mono- and dicotyledons have demonstrated that their enzymatic activity is not limited to transglycosylation. A spectrum of activities, ranging from transglycosylase to exclusive hydrolase has indeed been identified (Okazawa et al., 1993; Xu et al., 1995; Tabuchi et al., 1997, 2001; Catalá et al., 2001). Within these extremes some XTHs were shown to act as transglycosylase or as hydrolase depending on acceptor substrate availability (de Silva et al., 1993; Fanutti et al., 1993). This functional diversity of the XTHs of higher plants is another interesting aspect to study in the evolutionary perspective. Is the capacity to hydrolyse xyloglucans an ancient feature of XTHs, common with the ß-endoglycosidases? Or is it the result of the loss of XET-function of some evolutionary more recent XTHs? Incubation of Sk-XTH1 with high molecular weight xyloglucan (200 kDa) caused only a small decrease in iodine staining, compared with the reaction where oligo xyloglucans were included. Most likely this small decrease was only the result of enzyme–substrate complex formation, since both BCA reducing assay and TLC were unable to detect hydrolysis. These data suggest that Sk-XTH1 does not function as a xyloglucan hydrolase. The main function of this ‘ancient XTH’ is clearly not just cleavage, but probably the reconnection of the xyloglucans and hence the restructuring and maintenance of the xyloglucan/cellulose ‘load-bearing’ network of the PCW.

A phylogenetic tree using full-length amino acid sequences of dicotyledons (tree A) and monocotyledons (tree B) generated three groups of related XTHs (Campbell and Braam, 1999; Nishitani, 1997; Schünmann et al., 1997; Yokoyama and Nishitani, 2001). This phylogenetic divergence possibly reflects the evolution of XTH subgroups with different mechanisms of action. So far, members of group I and II have been demonstrated to mediate exclusively transglycosylation between xyloglucan in vitro. Some members of group III were shown to catalyse xyloglucan endo-hydrolysis. In accordance with its activity and gene structure Sk-XTH1 was phylogenetically classified in group I. XTH-genes of Arabidopsis belonging to this group are expressed in young developing tissues (Catalá et al., 1997; Campbell and Braam, 1999) and are thought to be mainly involved in cell wall expansion. In seed plants the different XTHs have evolved specific functions, for example, in fruit ripening (Arrowsmith and de Silva, 1995) and the mobilization of xyloglucan in seeds (Fanutti et al., 1993). Such functions are irrelevant in spore plants. It is therefore presumed that Sk-XTH1 is mainly involved in the restructuring of the cell wall during growth and development. The broad pH- and temperature optima of Sk-XTH1 support this idea and allow the enzyme to be active in a wide range of physiological conditions.

Club mosses diverged early from the ancestors of the euphyllophytes (i.e. plants with true leaves). Similarities of their primary cell wall composition (Popper and Fry, 2004) as well as of the modifying enzymes with those of higher plants, suggest the presence of a well-conserved cell wall structure and expansion mechanism. The data presented in this paper point to the deeply embedded role of XTH within vascular plant cell wall elongation and demonstrate the strong conservation of the relevant protein motifs and amino acids during plant evolution.

Since the genome sequencing of Physcomitrella patens is in progress, the presence of XTHs in bryophytes is currently of interest in our laboratory. Going one step down on the evolutionary ladder will reveal more of the involvement of XTHs in the cell wall dynamics of the land plants.


   Supplementary data
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary data
References
 
Supplementary data, including the figures of the XEH assays, are available at JXB online.


   Acknowledgements
 
V Van Sandt is funded by a PhD grant of the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT, Vlaanderen). K Vissenberg is a Postdoctoral Fellow of the Research Foundation, Flanders (FWO, Vlaanderen). The research was partially funded by a University of Antwerpen-grant (UA-BOF). The authors would like to thank Professor SC Fry (University of Edinburgh, UK) for the labelled tri- and oligosaccharides (XGO-SRs) and the CEPROMA centre of the University of Antwerp for the protein sequencing.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary data
 References
 
Apel K, Bogorad L, Woodcock CL. (1975) Chloroplast membranes of the green alga Acetabularia mediterranea. I. Isolation of the photosystem II. Biochimica et Biophysica Acta 387:568–579.[Medline]

Arrowsmith DH and de Silva J. (1995) Characterisation of two tomato fruit-expressed cDNAs encoding xyloglucan endo-transglycosylase. Plant Molecular Biology 28:391–403.[CrossRef][ISI][Medline]

Bradford MM. (1976) A rapid and sensitive method for the quantitation of microgram quantities of proteins utilizing the principle of protein dye binding. Analytical Biochemistry 72:248–254.[CrossRef][ISI][Medline]

Brown RM Jr. (1990) Algae as tools in studying the biosynthesis of cellulose, nature's most abundant macromolecule. In Wiessner W, Robinson DG, Starr RC (Eds.). Experimental phycology. Cell walls and surfaces, reproduction, photosynthesis (Springer-Verlag, Berlin) pp. 20–39.

Bourquin V, Nishikubo N, Abe H, Brumer H, Denman S, Eklund M, Christiemin M, Teeri TT, Sundberg B, Mellerowicz EJ. (2002) Xyloglucan endotransglycosylases have a function during the formation of the secondary cell walls of vascular tissues. The Plant Cell 14:3073–3088.[Abstract/Free Full Text]

Campbell P and Braam J. (1998) Co- and/or post-translational modifications are critical for TCH4 XET activity. The Plant Journal 15:553–561.[CrossRef][ISI][Medline]

Campbell P and Braam J. (1999) Xyloglucan endotransglycosylases: diversity of genes, enzymes and potential wall-modifying functions. Trends in Plant Science 4:361–366.[CrossRef][ISI][Medline]

Catalá C, Rose JKC, Bennett AB. (1997) Auxin regulation and spatial localization of an endo-1,4-ß-D-glucanase and a xyloglucan endotransglycosylase in expanding tomato hypocotyls. The Plant Journal 12:417–426.[CrossRef][ISI][Medline]

Catalá C, Rose JKC, York WS, Albersheim P, Darvill AG, Bennett AB. (2001) Characterisation of a tomato xyloglucan endotransglucosylase gene that is down-regulated by auxin in etiolated hypocotyls. Plant Physiology 127:1180–1192.[Abstract/Free Full Text]

Cosgrove DJ. (1999) Enzymes and other agents that enhance cell wall extensibility. Annual Review of Cell and Development Biology 13:171–201.

Cosgrove DJ. (2000) Loosening of plant cell walls by expansins. Nature 407:321–326.[CrossRef][Medline]

De Silva J, Jarman CD, Arrowsmith DA, Stronach MS, Chengappa S, Sidebottom C, Reid JSG. (1993) Molecular characterization of a xyloglucan-specific endo-(1->4)-ß-D-glucanase (xyloglucan endotransglycosylase) from nasturtium seeds. The Plant Journal 3:701–711.[CrossRef][ISI][Medline]

Devos S, Vissenberg K, Verbelen J-P, Prinsen E. (2005) Infection of Chinese cabbage by Plasmodiophora brassicae leads to a stimulation of plant growt