Molecular Characterization of PoGT8D and ... - Oxford Academic

Plant Cell Physiol. 48(5): 689–699 (2007) doi:10.1093/pcp/pcm037, available online at www.pcp.oxfordjournals.org ß The Author 2007. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: [email protected]

Molecular Characterization of PoGT8D and PoGT43B, Two Secondary Wall-Associated Glycosyltransferases in Poplar Gong-Ke Zhou 1, Ruiqin Zhong 1, David S. Himmelsbach 2, Brooks T. McPhail

2

and Zheng-Hua Ye

1,

*

1

Department of Plant Biology, University of Georgia, Athens, GA 30602, USA 2 Richard B. Russell Agriculture Research Center, US Department of Agriculture, Agriculture Research Service, Athens, GA 30604, USA

composed of a linear b-(1,4)-linked D-xylosyl backbone with side chains of O-2-linked a-D-glucuronic acid (GlcA) and/or 4-O-methyl-a-D-glucuronic acid (MeGlcA). The xylosyl backbone can be further modified by addition of arabinosyl and acetyl residues (Ebringerova´ and Heinze 2000). In addition, the reducing end of GXs from wood of spruce, birch and poplar and stems of Arabidopsis was found to contain a glycosyl sequence, b-D-Xylp-(1!3)-a-LRhap-(1!2)-a-D-GalpA-(1!4)-D-Xylp, which is distinct from the backbone sequence (Shimizu et al. 1976, Johansson and Samuelson 1977, Andersson et al. 1983, Pena et al. 2007). Conceivably, a number of glycosyltransferases (GTs) and other modifying enzymes are required for the synthesis of the xylosyl backbone and the reducing end glycosyl sequence, and for addition and modification of the side chains. Although the xylan synthase activity has been detected in woody and herbaceous plants (Dalessandro and Northcote 1981, Baydoun et al. 1983, Suzuki et al. 1991, Porchia and Scheller 2000, Kuroyama and Tsumuraya 2001, Gregory et al. 2002), little is known about the genes involved in GX biosynthesis in wood. By using microtome sectioning of developing poplar wood tissues and subsequent microarray analysis, Aspeborg et al. (2005) have found that the expression of 25 GT genes is correlated with secondary wall thickening during wood formation. Except for those that are known to be involved in the biosynthesis of cellulose (Djerbi et al. 2004, Joshi et al. 2004) and glucomannan (Suzuki et al. 2006), the functions of these secondary wall-associated GTs are not known. Because GX constitutes 18–28% of total dry weight of poplar wood (Mellerowicz et al. 2001), it is expected that some of these secondary wall-associated GT genes are involved in GX biosynthesis (Ye et al. 2006). Indeed, three of them, PttGT47C, PttGT8D and PttGT43A/B, are close homologs of the Arabidopsis thaliana FRAGILE FIBER8 (FRA8), IRREGULAR XYLEM8 (IRX8) and IRX9 genes, respectively, which have been shown to be required for normal GX biosynthesis (Zhong et al. 2005, Bauer et al. 2006, Pena et al. 2007). FRA8, IRX8 and IRX9 are putative GTs belonging to family GT47, GT8 and GT43, respectively. The FRA8, IRX8 and IRX9 genes have been shown to be expressed

Dicot wood is mainly composed of cellulose, lignin and glucuronoxylan (GX). Although the biosynthetic genes for cellulose and lignin have been studied intensively, little is known about the genes involved in the biosynthesis of GX during wood formation. Here, we report the molecular characterization of two genes, PoGT8D and PoGT43B, which encode putative glycosyltransferases, in the hybrid poplar Populus alba tremula. The predicted amino acid sequences of PoGT8D and PoGT43B exhibit 89 and 75% similarity to the Arabidopsis thaliana IRREGULAR XYLEM8 (IRX8) and IRX9, respectively, both of which have been shown to be required for GX biosynthesis. The PoGT8D and PoGT43B genes were found to be expressed in cells undergoing secondary wall thickening, including the primary xylem, secondary xylem and phloem fibers in stems, and the secondary xylem in roots. Both PoGT8D and PoGT43B are predicted to be type II membrane proteins and shown to be targeted to Golgi. Overexpression of PoGT43B in the irx9 mutant was able to rescue the defects in plant size and secondary wall thickness and partially restore the xylose content. Taken together, our results demonstrate that PoGT8D and PoGT43B are Golgi-localized, secondary wallassociated proteins, and PoGT43B is a functional ortholog of IRX9 involved in GX biosynthesis during wood formation. Keywords: Glucuronoxylan — Glycosyltransferase — Poplar — Secondary wall synthesis. Abbreviations: CaMV, cauliflower mosaic virus; CFP, cyan fluorescent protein; fra8, fragile fiber8; GFP, green fluorescent protein; GlcA, glucuronic acid; GT, glycosyltransferase; GX, glucuronoxylan; irx8, irregular xylem8; irx9, irregular xylem9; MeGlcA, 4-O-methyl-glucuronic acid; RT–PCR, reverse transcription–PCR; YFP, yellow fluorescent protein. The nucleotide sequences described in Fig. 1 can be found in GenBank with the following accession numbers: PoGT8D (EF501824), PoGT43B (EF501825), IRX8 (NM_124850) and IRX9 (NM_129265).

Introduction Glucuronoxylan (GX) is the second most abundant polysaccharide after cellulose in dicot wood. GX is

*Corresponding author: E-mail, [email protected]; Fax, þ1-706-542-1805. 689

690

Secondary wall-associated glycosyltransferases

specifically in cells undergoing secondary wall thickening, and their encoded proteins are targeted to Golgi (Zhong et al. 2005, Pena et al. 2007), the site for GX biosynthesis (Gregory et al. 2002). Mutations of the FRA8, IRX8 and IRX9 genes cause a drastic reduction in GX content, indicating that they are essential for GX biosynthesis (Brown et al. 2005, Persson et al. 2005, Zhong et al. 2005, Bauer et al. 2006, Pena et al. 2007). Recent structural analysis of GXs from the wild-type and mutant plants further revealed that the irx9 mutation caused a dramatic reduction in GX chain length but an increase in its chain number, whereas the fra8 and irx8 mutations led to a drastic reduction in GX chain number but an increase in GX size distribution (Pena et al. 2007). In addition, while the irx9 mutation caused an increase in the abundance of the reducing end glycosyl sequence of GX compared with the wild type, the fra8 and irx8 mutations resulted in a loss of this sequence. These findings suggest that FRA8 and IRX8 are required for the synthesis of the glycosyl sequence at the reducing end of GX and that IRX9 is involved in the normal elongation of GX chains (Pena et al. 2007). We have recently shown that overexpression of the secondary wall-associated PoGT47C from the hybrid poplar Populus alba tremula is able to rescue the secondary wall and xylan defects caused by the fra8 mutation (Zhou et al. 2006). This finding indicates that PoGT47C is a functional ortholog of FRA8 and it most probably participates in GX synthesis during wood formation. It also suggests the feasibility of using Arabidopsis GT mutants as a complementary approach to study the functions of poplar GTs in wood formation. In the present study, we carried out molecular characterization of two poplar GTs, PoGT8D and PoGT43B, and tested their ability to complement the secondary wall defects caused by mutations of their corresponding Arabidopsis homologous genes.

Results and Discussion The PoGT8D and PoGT43B genes are specifically expressed in cells undergoing secondary wall thickening As a first step toward the molecular characterization of poplar GT8D and GT43B, we isolated their full-length cDNAs by screening a stem cDNA library of the hybrid poplar P. alba tremula using the Arabidopsis IRX8 and IRX9 cDNAs as probes. The longest open reading frame in the PoGT8D cDNA is 1,602 bp long, and it encodes a putative protein of 533 amino acid residues with a predicted molecular mass of 60.6 kDa and a predicted pI of 8.9. The deduced amino acid sequence of PoGT8D exhibits 83% identity and 89% similarity to the Arabidopsis IRX8 protein (Fig. 1A). The longest open reading frame in the PoGT43B

cDNA is 1,071 bp long, and it encodes a putative protein of 356 amino acid residues with a predicted molecular mass of 40.1 kDa and a predicted pI of 8.6. The deduced amino acid sequence of PoGT43B exhibits 64% identity and 75% similarity to the Arabidopsis IRX9 protein (Fig. 1B). Phylogenetic analysis of family GT8 and GT43 genes from the Populus trichocarpa genome showed that there exist two close poplar homologs for Arabidopsis IRX8 (Fig. 2A), and two for IRX9 (Fig. 2B). However, using the IRX8 and IRX9 cDNAs as probes, we were only able to isolate one of the two close poplar homologs, PoGT8D and PoGT43B, respectively. Next we examined the expression patterns of the PoGT8D and PoGT43B genes in different organs and tissues of poplar. Reverse transcription–PCR (RT–PCR) analysis showed that the PoGT8D and PoGT43B genes are highly expressed in leaves and stems (Fig. 1C). In situ hybridization revealed that the PoGT8D and PoGT43B transcripts were present in developing primary xylem and phloem fibers in elongating stems (Fig. 3A, B). In nonelongating stems in which secondary growth was initiated, strong signals were seen in developing secondary xylem (Fig. 3D, E), which is consistent with the previous microarray data (Aspeborg et al. 2005). The PoGT8D and PoGT43B transcripts were also found to be present in developing secondary xylem of roots (Fig. 3G, H). The controls hybridized with the sense PoGT8D probe (Fig. 3C, F, I) or sense PoGT43B probe (data not shown) did not show any signals in developing xylem or phloem fibers. These results demonstrate that the expression of PoGT8D and PoGT43B genes is associated with secondary wall thickening, which is similar to the fact that their Arabidopsis homologs, IRX8 and IRX9, are also specifically expressed in cells undergoing secondary wall thickening (Pena et al. 2007). Although the RT–PCR analysis specifically detects the expression levels of PoGT8D and PoGT43B as judged by sequencing of the PCR products, the PoGT8D and PoGT43B probes used for in situ hybridization experiments might cross-hybridize with the mRNAs of their corresponding closest homologs. This is because the DNA fragments of PoGT8D and PoGT43B used as templates for probe synthesis exhibit 86 and 88% sequence identity to their closest homologs, respectively. Fluorescent protein-tagged PoGT8D and PoGT43B are targeted to Golgi The IRX8 and IRX9 proteins were previously shown to be localized in Golgi (Pena et al. 2007), which is in agreement with their proposed role in xylan biosynthesis. To investigate whether PoGT8D and PoGT43B play roles similar to their Arabidopsis homologs, we first studied their subcellular locations. Both PoGT8D and PoGT43B are predicted to be type II membrane proteins that contain


Le

af

C

tio le St em I St em II

B

Pe

A

691

PoGT8D PoGT43B Actin

Fig. 1 PoGT8D and PoGT43B are close homologs of Arabidopsis IRX8 and IRX9, respectively. (A) Amino acid sequence alignment of PoGT8D and IRX8. The numbers shown at the left of each sequence are the positions of amino acid residues in the corresponding proteins. Gaps (marked with dashes) were introduced to maximize the sequence alignment. Identical and similar amino acid residues are shaded with black and gray, respectively. (B) Amino acid sequence alignment of PoGT43B and IRX9. (C) RT–PCR analysis of the expression of PoGT8D and PoGT43B genes in poplar organs. The expression level of an actin gene was used as an internal control. The leaves and petioles used were at the expanding stage, and stem I and stem II were from elongating and non-elongating parts of the stems, respectively.

a short cytoplasmic N-terminus followed by a single transmembrane helix and a long non-cytoplasmic C-terminus (Fig. 4A, B). To examine their subcellular locations, PoGT8D and PoGT43B were tagged with green fluorescent protein (GFP) and expressed in transgenic Arabidopsis plants. Confocal imaging of the fluorescence signals in root epidermal cells showed that both PoGT8D– GFP (Fig. 4E, F) and PoGT43B–GFP (Fig. 4G, H) exhibited a punctate pattern in the cytoplasm, suggesting that they are targeted to certain subcellular organelles. The control GFP alone showed signals throughout the cytoplasm (Fig. 4C, D). To determine in which subcellular organelles PoGT8D and PoGT43B are located, yellow fluorescent protein

(YFP)-tagged PoGT8D and PoGT43B were co-transfected into carrot protoplasts with cyan fluorescent protein (CFP)tagged MUR4, which was previously shown to be located in Golgi (Burget et al. 2003). It was demonstrated that both PoGT8D–YFP and PoGT43B–YFP signals were co-localized with the MUR4–CFP signals (Fig. 4K–R). The control YFP alone had signals distributed in the cytoplasm and the nucleus (Fig. 4I, J). Because the Golgi targeting of a protein is determined by its intrinsic amino acid sequences (Machamer 1993), the fact that the fluorescence signals of PoGT8D–YFP and PoGT43B– YFP but not YFP alone were located specifically in Golgi indicates that the protein sequences of PoGT8D and PoGT43B are responsible for their Golgi targeting.

692


A

556346 577718 IRX8

774919

410293 831613

834622/PoGT8D/PttGT8D 409728

569069 282518 254623

244180 293912 560055 828111 661826 581394

418911 418609 664389

663529 783905

552490 266020

805206

590338/PttGT8G

766024

799474

228663 420166

420023

796839 286701/PttGT8F

834449

219005/PttGT8E

744421

412772

570686 830464 572572

676387

803462 805284 657524 658324

646104

558398 562455/PttGT8C

573752

651523 699864

662758 571125

204782/PttGT8B 782892 703114/PttGT8A

578089

B

428200 At1g27600

IRX9

816398

735460/PoGT43B /PttGT43B 826399/PttGT43A 218550 716281 At5g27630 At4g36890

Fig. 2 Phylogenetic trees of poplar glycosyltransferase families GT8 (A) and GT43 (B). The phylogenetic relationship of Populus trichocarpa GTs was analyzed with ClustalW and TREEVIEW. Populus trichocarpa GT sequences were retrieved from the DOE Joint Genome Institute website, and the numbers shown are the IDs of the proteins. Arabidopsis GT43 family members were included in the phylogenetic analysis of poplar GT43 for comparison.


693

PoGT43B

PoGT8D

A

pf

Control

B

pf

C

px px

px

D

E

SX

F

SX

G

CO

SX

SX

H

CO

SX

I

CO

SX

Fig. 3 In situ hybridization detection of the expression of PoGT8D and PoGT43B genes in poplar. Cross-sections of stems and roots were hybridized with digoxigenin-labeled antisense (A, B, D, E, G and H) or sense (C, F and I) RNA probes, and the hybridization signals were detected with alkaline phosphatase-conjugated antibodies and are shown as purple color. (A) and (B) Cross-sections of the elongating region of stems showing the expression of PoGT8D (A) and PoGT43B (B) in developing primary xylem and phloem fibers. (C) A control section of the elongating region hybridized with the sense PoGT8D probe showing the absence of hybridization signals. (D) and (E) Crosssections of the non-elongating region of stems showing the expression of PoGT8D (D) and PoGT43B (E) in developing secondary xylem. (F) A control section of the non-elongating region hybridized with the sense PoGT8D probe showing the absence of hybridization signals. (G) and (H) Cross-sections of roots showing the expression of PoGT8D (G) and PoGT43B (H) in developing secondary xylem. Note that the secondary xylem was detached from cortex during sectioning (arrows). Due to the distortion of sections, the vascular cambium is not discernible. (I) A control root section hybridized with the sense PoGT8D probe showing the absence of hybridization signals. co, cortex; pf, phloem fiber; px, primary xylem; sx, secondary xylem. Bars ¼ 69 mm in (A)–(F), and 66 mm in (G)–(I).

The Golgi localization of PoGT8D and PoGT43B indicates that they are most probably involved in the biosynthesis of non-cellulosic polysaccharides during secondary wall synthesis in wood. PoGT43B is a functional ortholog of Arabidopsis IRX9 that is known to be involved in GX biosynthesis To investigate whether PoGT8D and PoGT43B are functional orthologs of Arabidopsis IRX8 and IRX9, respectively, we overexpressed the full-length PoGT8D and PoGT43B cDNAs in the corresponding Arabidopsis mutants and examined their ability to rescue the mutant

phenotypes. The irx8 and irx9 mutations were shown to cause a reduction in plant size and stem strength, a decrease in secondary wall thickness and a collapsed vessel phenotype (Brown et al. 2005, Pena et al. 2007). More than 100 transgenic plants were generated and their phenotypes were examined. It was found that whereas overexpression of PoGT8D did not complement the irx8 mutant (data not shown), overexpression of PoGT43B in irx9 was able to rescue the mutant phenotypes (Figs. 5, 6). The results from eight representative lines were presented. Both the rosette size and the inflorescence height were restored to the wild-type level (Fig. 5B, C), and the stem

694

Secondary wall-associated glycosyltransferases 1.2

A

1.2

B

1.0

0.8

Probability

Probability

1.0 PoGT8D

0.6 0.4

0.8

PoGT43B

0.6 0.4 0.2

0.2

0

0 100

200

300

400

500

50

Amino acid residue Transmembrane

Inside

Outside

100 150 200 250 Amino acid residue

Transmembrane

Inside

300

350

Outside

C

D

GFP

E

F

PoGT8D-GFP

G

H

PoGT43B-GFP

I

J

YFP

K

L

PoGT8D-YFP

M

MUR4-CFP

N

Merged

O

P

PoGT43B-YFP

Q

MUR4-CFP

R

Merged


strength was partially restored (Fig. 5D). The rescued stem strength in the PoGT43B-complemented irx9 plants was accompanied by the restoration of secondary wall thickness in interfascicular fibers, xylem vessels and xylary fibers (Fig. 6, Table 1) and the rescue of the collapsed vessel phenotype (Fig. 6D–F). It has been shown that the cell walls from irx9 mutant stems had a reduction in xylose and an elevation in several other monosaccharides, including mannose, galactose and arabinose (Brown et al. 2005, Bauer et al. 2006). Overexpression of PoGT43B was able partially to rescue these cell wall sugar defects (Table 2). It is interesting to note that although the cell wall sugar defects were only partially rescued, the thickness of secondary walls in fibers and vessels could be restored to a level comparable with that of the wild type. In addition, despite the fact that the breaking strength of stems showed a variable range, the rosette size and stem height of the transgenic plants were equivalent to those of the wild type. In conclusion, we have shown that the PoGT8D and PoGT43B genes are expressed in cells undergoing secondary wall thickening in poplar stems and roots, and their encoded proteins are targeted to Golgi, the site for GX biosynthesis. We further demonstrate that overexpression of PoGT43B is able to rescue the deficiency in secondary wall synthesis caused by the irx9 mutation. Our results suggest that PoGT43B most probably performs a biochemical function identical to that of IRX9, thus leading to the rescue of the irx9 mutant phenotypes. Because IRX9 has been shown to be required for GX biosynthesis in Arabidopsis, it is reasonable to propose that PoGT43B is also involved in GX biosynthesis during wood formation in poplar. The present study together with another report by Zhou et al. (2006) suggests that genes involved in GX biosynthesis are conserved in herbaceous Arabidopsis and woody trees. Because GX is the second most abundant polysaccharide in dicot wood, it is expected that understanding GX biosynthesis would provide novel strategies for improvement of wood quantity and quality in tree species.

695

Materials and Methods Isolation of poplar GT cDNAs and sequence analysis The full-length PoGT8D and PoGT43B cDNAs were isolated by screening a poplar (P. tremula P. alba) stem cDNA library (Zhong et al. 2000) with IRX8 and IRX9 cDNA probes, and sequenced using a dye-based cycle sequencing kit (Applied Biosystems, Foster City, CA, USA). The protein sequences of family GT8 and GT43 in P. trichocarpa were retrieved under the GO numbers GO:0016051 and GO:0015018, respectively, from the DOE Joint Genome Institute website (http://genome.jgi-psf.org/ Poptr1/Poptr1.home.html). Individual GT sequences can be accessed using the protein ID number through the website (http://genome.jgi-psf.org/cgi-bin/dispGeneModel?db¼Poptr1&id¼ protein ID number). The GT sequences were aligned using the ClustalW 1.8 program (http://www.ebi.ac.uk/clustalw/), and the resulting alignment parameters were used to generate the phylogenetic tree using TREEVIEW (Page, 1996). Expression analysis Poplar (P. tremula P. alba) plants were propagated on modified Murashige and Skoog medium as described by Leple et al. (1992). Total RNA was isolated from leaves and stems using a Qiagen RNA isolation kit (Qiagen, Valencia, CA, USA), and treated with DNase I to remove any potential genomic DNA contamination. The total RNA was then used for first-strand cDNA synthesis and subsequent semi-quantitative PCR analysis of the expression of PoGT8D and PoGT43B genes with gene-specific primers (50 -atgcagcttcatatatcgccgagc-30 and 50 -tctaatgtttgaggaatatcactc-30 for PoGT8D; and 50 -taaatcgcaatccttgccgcaacc-30 and 50 acgattttgtcttcttgattttcc-30 for PoGT43B). The RT–PCR products were sequenced and it was found that they corresponded to PoGT8D and PoGT43B, respectively. The sequences of their closest homologs were absent in the RT–PCR products, indicating that the RT–PCR analysis specifically detected the expression of PoGT8D and PoGT43B. The RT–PCRs were repeated three times, and identical results were obtained. The expression of an actin gene was used as an internal control to determine the RT–PCR amplification efficiency among different samples. For the synthesis of antisense and sense probes used for in situ hybridization, a 530 bp fragment of PoGT8D cDNA and the full-length cDNA of PoGT43B were PCR amplified with their corresponding primers (50 -atgcagcttcatatatcgccgagc-30 and 50 -ctcagggcaaggcaatgaagctgc-30 for PoGT8D; 50 -atgggctctgtggaaagatcaaag-30 and 50 - acgattttgtcttcttgattttcc-30 for PoGT43B), and the amplified DNA fragments were ligated into the pGEM vector

Fig. 4 Fluorescent protein-tagged PoGT8D and PoGT43B are targeted to Golgi. Fluorescent protein-tagged proteins were expressed in Arabidopsis plants and carrot protoplasts, and their fluorescence signals were visualized using a laser confocal microscope. PoGT8D (A) and PoGT43B (B) are type II membrane proteins predicted by the TMHMM2.0 program. They are predicted to contain a short N-terminal region located on the cytoplasmic side of the membrane (inside) followed by a transmembrane helix and a long stretch of C-terminal region located on the non-cytoplasmic side of the membrane (outside). Differential interference contrast (DIC) image (C) and the corresponding fluorescence signal (D) of Arabidopsis root epidermal cells expressing GFP alone. Note the presence of the GFP signal throughout the cytoplasm and the nucleus. DIC image (E) and the corresponding fluorescence signal (F) of Arabidopsis root epidermal cells expressing PoGT8D–GFP. Note the punctate pattern of the PoGT8D–GFP signal. DIC image (G) and the corresponding fluorescence signal (H) of Arabidopsis root epidermal cells expressing PoGT43B–GFP. Note the punctate pattern of the PoGT43B–GFP signal. DIC image (I) and the corresponding fluorescence signal (J) of a carrot protoplast expressing YFP alone. DIC image (K) and the corresponding PoGT8D–YFP signal (L), MUR4–CFP signal (M) and a merged image (N) of a carrot cell expressing PoGT8D–YFP and MUR4–CFP. Note the superimposition of PoGT8D–YFP and MUR4–CFP signals. DIC image (O) and the corresponding PoGT43B–YFP signal (P), MUR4–CFP signal (Q) and a merged image (R) of a carrot cell expressing PoGT43B–YFP and MUR4–CFP. Note the superimposition of PoGT43B–YFP and MUR4–CFP signals. Bars ¼ 20 mm in (C)–(H) and 31 mm in (I)–(R).

696


A

irx9 plants transformed with poplar PoGT43B WT

1

2

3

4

5

6

7

8

PoGT43B transgene

PoGT43B transcript

EF1α

B

irx9

irx9+PoGT43B

wild type

C

irx9

irx9+PoGT43B

Wild type

D

Breaking force (g)

4000

3000

2000

1000

0 irx9

irx9+PoGT43B

Wild type

Fig. 5 Restoration of plant size and stem strength in the Arabidopsis irx9 plants by overexpression of the poplar PoGT43B gene. The results shown are data from eight representative transgenic Arabidopsis lines. (A) PCR detection of the PoGT43B transgene (upper panel) and its transcript (middle panel) in the transgenic irx9 plants. The expression of the EF1a gene was used as an internal control (lower panel). WT, wild type. (B) The irx9

containing both T7 and SP6 promoters (Promega, Madison, WI, USA). The antisense and sense RNA probes were synthesized with the DIG RNA labeling mix (Roche, Mannheim, Germany) using T7 or SP6 RNA polymerase, respectively. Segments of poplar stems and roots were fixed in 2.5% formaldehyde and 0.5% glutaraldehyde, and embedded in paraffin. Tissues sections (12 mm) were cut, mounted onto glass slides and hybridized with digoxygenin-labeled PoGT8D and PoGT43B antisense or sense RNA probes. The hybridization signals were detected by incubating with alkaline phosphatase-conjugated antibodies against digoxigenin and subsequent color development with alkaline phosphatase substrates. The images were collected under a bright field using a digital camera and processed with Adobe Photoshop Version 7.0 (Adobe Systems, San Jose, CA, USA). Subcellular localization of fluorescent protein-tagged PoGT8D and PoGT43B For subcellular localization experiments, GFP-tagged PoGT8D and PoGT43B were used. The GFP cDNA (ABRC, Columbus, OH, USA) was ligated between the cauliflower mosaic virus (CaMV) 35S promoter and the nopaline synthase terminator in the binary vector pBI121 (Clontech, Moutain View, CA, USA) to create the pBI121-GFP construct. The full-length PoGT8D and PoGT43B cDNAs were PCR-amplified with a high fidelity DNA polymerase, and ligated into pBI121-GFP at the BamHI site to create an in-frame fusion with the GFP cDNA. The resulting constructs encode fusion proteins with PoGT8D or PoGT43B located at the N-terminus and GFP at the C-terminus, and seven additional amino acid residues (RIQGDIT) at the junction region. The constructs were stably transformed into Arabidopsis plants (Bechtold and Bouchez 1994), and transgenic plants were selected on kanamycin. The GFP signals in roots of 3-day-old transgenic seedlings were observed using a Leica TCs SP2 spectral confocal laser scanning microscope (Leica Microsystems, Heidelberg, Germany). Images were saved and processed with Adobe Photoshop. The co-localization of fluorescent protein-tagged GTs with a Golgi marker was carried out in carrot protoplasts according to Liu et al. (1994). The YFP cDNA (Clontech) was ligated between the CaMV 35S promoter and the nopaline synthase terminator in the pBI221 vector (Clontech) to create the pBI221–YFP construct. The full-length PoGT8D and PoGT43B cDNAs were PCR-amplified and ligated into PBI221–YFP at the engineered MluI site to create an in-frame fusion with the YFP cDNA. The resulting constructs encode fusion proteins with PoGT8D or PoGT43B located at the N-terminus and YFP at the C-terminus, and two additional amino acid residues (TR) at the junction region. The YFP fusion construct together with the MUR4–CFP construct was co-transfected into carrot (Daucus carota) protoplasts. After 20 h incubation, the transfected protoplasts were examined for fluorescence signals using a confocal laser scanning mutant (left) has a small rosette size, and overexpression of PoGT43B in irx9 (middle) restored the rosette size to that of the wild type (right). The rosette size of all eight representative lines was comparable with that of the wild type. (C) The irx9 mutant (left) has a short inflorescence stem, and overexpression of PoGT43B in irx9 (middle) restored the stem height to that of the wild type (right). The height of stems in all eight lines was equivalent to that of the wild type. (D) Breaking force measurement showing that overexpression of the PoGT43B gene in irx9 partially restored the stem strength. Each bar represents the stem breaking force of individual plants.


CO

A

D

G

ve

697

J

xf

irx9

ve ph if ve

B

H

E

CO

K

irx9+PoGT43B

xf

ph ve

if

ve ve

C

I

F

CO

L

Wild type

xf if ph ve ve

Fig. 6 Restoration of secondary wall thickness of fibers and vessels in the transgenic Arabidopsis irx9 plants overexpressing the poplar PoGT43B gene. The bottom internodes of 10-week-old plants were sectioned for examination of fibers and vessels. The PoGT43Bcomplemented irx9 line used was the transgenic line with the highest breaking strength as shown in Fig. 5D. (A)–(C) Cross-sections of interfascicular regions of stems of irx9 (A), PoGT43B-complemented irx9 (B) and wild type (C). (D)–(F) Cross-sections of xylem regions of stems of irx9 (D), PoGT43B-complemented irx9 (E) and wild type (F). Note that the irx9 mutant exhibited a collapsed vessel phenotype (arrow in D), which was rescued by expression of the PoGT43B gene (E). (G)–(I) Transmission electron micrographs of interfascicular fiber cells of irx9 (G), PoGT43B-complemented irx9 (H) and wild type (I). (J)–(L) Transmission electron micrographs of walls of vessels and xylary fibers in irx9 (J), PoGT43B-complemented irx9 (K) and wild type (L). co, cortex; if, interfascicular fiber; ph, phloem; ve, vessel; xf, xylary fiber. Bars ¼ 70 mm in (A)–(F), 9 mm in (G)–(I) and 3.6 mm in (J)–(L).

Table 1 Wall thickness of interfascicular fibers, vessels and xylary fibers in the stems of wild type, irx9, and irx9 complemented with PoGT43B Sample

Interfascicular fibers

Vessels

Xylary fibers

irx9 irx9 þ PoGT43B Wild type

0.84 0.11 2.44 0.28

0.44 0.09 1.41 0.16

0.21 0.04 0.71 0.16

2.39 0.31

1.34 0.20

0.67 0.06

Wall thickness was measured from transmission electron micrographs of interfascicular fibers, vessels and xylary fibers of stems. The irx9 þ PoGT43B line used was the transgenic line with the highest breaking strength as shown in Fig. 5D. Data are means (mm) SE from 10 cells.

microscope. Images from single optical sections were collected and processed with Adobe Photoshop. Mutant complementation analysis For complementation analysis, the full-length PoGT8D and PoGT43B cDNAs were PCR-amplified and ligated downstream of the CaMV 35S promoter in the binary vector pBI121. The constructs were introduced into the Arabidopsis irx8 or irx9 mutant plants by Agrobacterium-mediated transformation. Transgenic plants were selected on kanamycin, and the first generation of transgenic plants were used for analysis. Basal parts of the main inflorescence of 10-week-old wild-type Arabidopsis plants, irx9, and irx9 complemented with PoGT43B were measured for breaking force using a digital force/length tester (Model DHT4-50; Larson System, Minneapolis, MN, USA). The breaking force (g) was calculated as the force needed to break apart a stem segment (Zhong et al. 1997).

698


Table 2 Monosaccharide composition of cell walls from the stems of wild type, irx9, and irx9 complemented with PoGT43B Sample

Glucose

Xylose

Mannose

Galactose

Arabinose

Rhamnose

Fucose

irx9 irx9 þ PoGT43B Wild type

312 5 296 28 310 14

78.4 3.0 94.6 4.6 140 0.2

40.9 0.4 26.1 1.8 17.1 0.5

24.7 0.01 17.9 0.9 12.3 0.7

21.1 0.4 17.8 0.5 10.8 0.3

9.9 0.04 9.9 0.5 10.1 1.9

2.9 0.3 2.5 0.7 2.4 1.2

Cell walls were prepared from stems of 10-week-old plants. Data are means (mg g1 dry cell wall) SE of two independent assays.

Histology Basal internodes of the main inflorescence stems of 10-weekold plants were fixed in 2% formaldehyde in PEMT buffer (50 mM PIPES, 2 mM EGTA, 2 mM MgSO4 and 0.05% Triton X-100, pH 7.2) at 48C overnight. After washing in PEMT buffer, samples were post-fixed in 1% OsO4 for 2 h and then dehydrated in a gradient of ethanol and embedded in Spurr’s resin (Electron Microscopy Sciences, Fort Washington, PA, USA) for histological analysis (Burk et al. 2006). Sections 1 mm thick were cut with a microtome and stained with toluidine blue for light microscopy. For transmission electron microscopy, 85 nm thick sections were cut, post-stained with uranyl acetate and lead citrate, and visualized using a Zeiss EM 902A transmission electron microscope (Carl Zeiss, Jena, Germany).

Cell wall composition analysis Inflorescence stems of 10-week-old plants were used for cell wall monosaccharide determination. Stems were ground into a fine powder in liquid nitrogen with a mortar and pestle, homogenized with a polytron, and extracted in 70% ethanol at 708C. The resulting cell wall residues were dried at 658C in a vacuum oven. Cell wall sugars (as alditol acetate) were determined following the procedure described by Hoebler et al. (1989). Briefly, cell walls were incubated with 70% sulfuric acid at 378C for 1 h followed by the addition of inositol as the internal standard and dilution with water to 2 N sulfuric acid. After heating for 2 h at 1008C, the solution was cooled and treated with 25% ammonium solution. After reduction with sodium borohydride in dimethylsulfoxide, the solution was heated for 1.5 h at 408C, followed by sequential treatment with glacial acetic acid, acetic anhydride, 1-methylimidazole, dichloromethane and water. The organic layer containing the alditol acetates of the hydrolyzed cell wall sugars was washed three times with water, and the sugars were analyzed on an Agilent 6890N gas–liquid chromatograph (Wilmington, DE, USA) equipped with a 30 m 0.25 mm (i.d.) silica capillary column DB 225 (Alltech Assoc., Deerfield, IL, USA). All samples were run in duplicate.

Acknowledgments We thank E. A. Richardson for her help with transmission electron microscopy, and the editor and reviewers for their suggestions. This work was supported by a grant from the US Department of Energy-Bioscience Division (DE-FG0203ER15415).

References Andersson, S.-I., Samuelson, O., Ishihara, M. and Shimizu, K. (1983) Structure of the reducing end-groups in spruce xylan. Carbohydr. Res. 111: 283–288. Aspeborg, H., Schrader, J., Coutinho, P.M., Stam, M., Kallas, A., et al. (2005) Carbohydrate-active enzymes involved in the secondary cell wall biogenesis in hybrid aspen. Plant Physiol. 137: 983–997. Bauer, S., Vasu, P., Persson, S., Mort, A.J. and Somerville, C.R. (2006) Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proc. Natl Acad. Sci. USA 103: 11417–11422. Baydoun, E.A.-H., Waldron, K.W. and Brett, C.T. (1983) The interaction of xylosyltransferase and glucuronyltransferase involved in glucuronoxylan synthesis in pea (Pisum sativum) epicotyls. Biochem. J. 257: 853–858. Bechtold, N. and Bouchez, D. (1994) In planta Agrobacterium-mediated transformation of adult Arabidopsis thaliana plants by vacuum infiltration. In Gene Transfer to Plants. Edited by Potrykus, I. and Spangenberg, G. pp. 19–23. Springer-Verlag, Berlin. Brown, D.M., Zeef, L.A.H., Ellis, J., Goodacreb, R. and Turner, S.R. (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell 17: 2281–2295. Burget, E.G., Verma, R., Molhoj, M. and Reiter, W.-D. (2003) The biosynthesis of L-arabinose in plants: molecular cloning and characterization of a Golgi-localized UDP-D-xylose 4-epimerase encoded by the MUR4 gene of Arabidopsis. Plant Cell 15: 523–531. Burk, D.H., Zhong, R., Morrison, W.H. 3rd and Ye, Z.-H. (2006) Disruption of cortical microtubules by overexpression of green fluorescent protein-tagged a-tubulin 6 causes a marked reduction in cell wall synthesis. J. Integr. Plant Biol. 48: 85–98. Dalessandro, G. and Northcote, D.H. (1981) Increase of xylan synthetase activity during xylem differentiation of the vascular cambium of sycamore and poplar trees. Planta 151: 61–67. Djerbi, S., Aspeborg, H., Nilsson, P., Sundberg, B., Mellerowicz, E., Blomqvist, K. and Teeri, T.T. (2004) Identification and expression analysis of genes encoding putative cellulose synthase (CesA) in the hybrid aspen, Populus tremula (L.) P. tremuloides (Michx.). Cellulose 11: 301–312. Ebringerova´, A. and Heinze, T. (2000) Xylan and xylan derivatives— biopolymers with valuable properties, 1. Naturally occurring xylans: structures, isolation procedures and properties. Macromol. Rapid Commun. 21: 542–556. Gregory, A.C.E., Smith, C., Kerry, M.E., Wheatley, E.R. and Bolwell, G.P. (2002) Comparative subcellular immunolocation of polypeptides associated with xylan and callose synthases in French bean (Phaseolus vulgaris) during secondary wall formation. Phytochemistry 59: 249–259. Hoebler, C., Barry, J.L., David, A. and Delort-Laval, J. (1989) Rapid acidhydrolysis of plant cell wall polysaccharides and simplified quantitative determination of their neutral monosaccharides by gas–liquid chromatography. J. Agric. Food Chem. 37: 360–367. Johansson, M.H. and Samuelson, O. (1977) Reducing end groups in birch xylan and their alkaline degradation. Wood Sci. Technol. 11: 251–263.

Secondary wall-associated glycosyltransferases Joshi, C.P., Bhandari, S., Ranjan, P., Kalluri, U.C., Liang, X., Fujino, T. and Samuga, A. (2004) Genomics of cellulose biosynthesis in poplars. New Phytol. 164: 53–61. Kuroyama, H. and Tsumuraya, Y. (2001) A xylosyltransferase that synthesizes b-(1!4)-xylans in wheat (Triticum aestivum L.) seedlings. Planta 213: 231–240. Leple, J.C., Brasileiro, A.C.M., Michel, M.F., Delmotte, F. and Jouanin, L. (1992) Transgenic poplars: expression of chimeric genes using four different constructs. Plant Cell Rep. 11: 137–141. Liu, Z. B., Ulmasov, T., Shi, X., Hagen, G. and Guilfoyle, T.J. (1994) Soybean GH3 promoter contains multiple auxin-inducible elements. Plant Cell 6: 645–657. Machamer, C.E. (1993) Targeting and retention of Golgi membrane proteins. Curr. Biol. 5: 606–612. Mellerowicz, E.J., Baucher, M., Sundberg, B. and Boerjan, W. (2001) Unravelling cell wall formation in the woody dicot stem. Plant Mol. Biol. 47: 239–274. Page, R.D.M. (1996) TREEVIEW: an application to display phylogenetic trees on personal computers. CABIOS 12: 357–358. Pena, M.J., Zhong, R., Zhou, G.-K., Richardson, E.A., O’Neill, M.A., Darvill, A.G., York, W.S. and Ye, Z.-H. (2007) Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis. Plant Cell 19: 549–563. Persson, S., Wei, H., Miline, J., Page, G.P. and Somerville, C.R. (2005) Identification of genes required for cellulose synthesis by repression analysis of public microarray data sets. Proc. Natl Acad. Sci. USA 102: 8633–8638. Porchia, A.C. and Scheller, H.V. (2000) Arabinoxylan biosynthesis: identification and partial characterization of b-1,4-xylosyltransferase from wheat. Physiol. Plant. 110: 350–356.

699

Shimizu, K., Ishihara, M. and Ishihara, T. (1976) Hemicellulases of brown rotting fungus, Tyromyces palustris. II. The oligosaccharides from the hydrolysate of a hardwood xylan by the intracellular xylanase. Mokuzai Gaikkashi 22: 618–625. Suzuki, K., Ingold, E., Sugiyama, M. and Komamine, A. (1991) Xylan synthase activity in isolated mesophyll cells of Zinnia elegans during differentiation to tracheary elements. Plant Cell Physiol. 32: 303–306. Suzuki, S., Li, L., Sun, Y.-H. and Chiang, V.L. (2006) The cellulose synthase gene superfamily and biochemical functions of xylem-specific cellulose synthase-like genes in Populus trichocarpa. Plant Physiol. 142: 1233–1245. Ye, Z.-H., York, W.S. and Darvill, A.G. (2006) Important new players in secondary wall synthesis. Trends Plant Sci. 11: 162–164. Zhong, R., Morrison, W.H., III, Himmelsbach, D.S., Poole, F.L. and Ye, Z.-H. (2000) Essential role of caffeoyl CoA O-methyltransferase in lignin biosynthesis in woody poplar plants. Plant Physiol. 124: 563–577. Zhong, R., Pena, M.J., Zhou, G.-K., Nairn, C.J., Wood-Jones, A., Richardson, E.A., Morrison, W.H., Darvill, A.G., York, W.S. and Ye, Z.-H. (2005) Arabidopsis Fragile Fiber8, which encodes a putative glucuronyltransferase, is essential for normal secondary wall synthesis. Plant Cell 17: 3390–3408. Zhong, R., Taylor, J.J. and Ye, Z.-H. (1997) Disruption of interfascicular fiber differentiation in an Arabidopsis mutant. Plant Cell 9: 2159–2170. Zhou, G.K., Zhong, R., Richardson, E.A., Morrison, W.H., 3rd Nairn, C.J., Wood-Jones, A. and Ye, Z.-H. (2006) The poplar glycosyltransferase GT47C is functionally conserved with Arabidopsis fragile fiber8. Plant Cell Physiol. 47: 1229–1240.

(Received January 19, 2007; Accepted March 13, 2007)