Expression of an endo-(1,3;1,4)-β-glucanase in response to wounding, methyl jasmonate, abscisic acid and ethephon in rice seedlings

ARTICLE IN PRESS Journal of Plant Physiology 166 (2009) 1814—1825

www.elsevier.de/jplph

Expression of an endo-(1,3;1,4)-b-glucanase in response to wounding, methyl jasmonate, abscisic acid and ethephon in rice seedlings Takashi Akiyamaa,, Shigeki Jinb, Midori Yoshidaa, Tamotsu Hoshinoc, Rodjana Opassirid, James R. Ketudat Cairnsd a

National Agricultural Research Center for Hokkaido Region, 1 Hitsujigaoka, Toyohira-ku, Sapporo 062-8555, Japan Graduate School of Health Sciences, Hokkaido University, N12W5, Kita-ku, Sapporo 060-0812, Japan c National Research Institute of Advanced Industrial Science and Technology, 2-17-2-1 Tsukisamu-higashi, Toyohira-ku, Sapporo 062-8517, Japan d Schools of Biochemistry and Chemistry, Institute of Science, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand b

Received 8 April 2009; received in revised form 4 June 2009; accepted 9 June 2009

KEYWORDS Endo-(1,3,1,4)-bglucanase; Leaf elongation; Phytohormone; Rice seedling; Wounding treatment

Summary We isolated two rice endo-(1,3;1,4)-b-glucanase genes, denoted OsEGL1 and OsEGL2, which encoded proteins that shared 64% amino acid sequence identity. Both the OsEGL1 and OsEGL2 genes were successfully expressed in Escherichia coli to produce functional proteins. Purified OsEGL1 and OsEGL2 proteins hydrolyzed (1,3;1,4)-b-glucans, but not (1,3;1,6)-b-linked or (1,3)-b-linked glucopolysaccharides nor carboxymethyl cellulose, similar to previously characterized grass endo-(1,3;1,4)-b-glucanases. RNA blot analysis revealed that the OsEGL1 gene is expressed constitutively not only in young roots of rice seedlings, but also in mature roots of adult rice plants. Little or no expression of the OsEGL2 gene was observed in all tissues or treatments tested, but database and RT-PCR analysis indicated it is expressed in ripening panicle. In rice seedling leaves, OsEGL1 gene expression significantly increased in response to methyl jasmonate, abscisic acid, ethephon and mechanical wounding. Mechanical wounding also increased the leaf elongation rate in rice seedlings by 16% relative to that of control seedlings at day 4 after treatment. The increase in the leaf elongation rate of rice seedlings treated under mechanical wounding was concomitant with an increase in OsEGL1 expression levels in seedling leaves. & 2009 Elsevier GmbH. All rights reserved.

Abbreviations: ABA, abscisic acid; GA, gibberellin A3; MeJA, methyl jasmonate; Salicylate, sodium salicylate; 2,4-D, 2,4-dichlorophenoxyacetic acid. Corresponding author. Tel.: +81 11 857 9271; fax: +81 11 859 2178. E-mail address: [email protected] (T. Akiyama). 0176-1617/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.jplph.2009.06.002

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice

Introduction Members of glycosyl hydrolase family 17, especially endo-(1,3;1,4)-b-glucanases (EC 3.2.1.73) and endo-(1,3)-b-glucanases (EC 3.2.1.39) from higher plants, are very similar in size and amino acid sequence (Henrissat, 1991). Therefore, it was proposed that the genes encoding the two hydrolytic enzymes have evolved from a common ancestral gene (Høy and Fincher, 1995). Despite this structural similarity, endo-(1,3;1,4)- and endo(1,3)-b-glucanase clearly recognize different substrates and hydrolyze different b-glycosidic linkages (Hrmova et al., 2002). Other than this difference in substrate specificity, there is a difference in distribution between endo-(1,3;1,4)and endo-(1,3)-b-glucanases in higher plants. The (1,3;1,4)-b-glucan substrate for endo-(1,3;1,4)-bglucanase has been found only in the Poaceae family (Stone and Clarke, 1992; Buckeridge et al., 2004). The distribution of endo-(1,3;1,4)-b-glucanase is consistent with that of the substrate, as it is found only in the Poaceae family in monocots, while endo-(1,3)-b-glucanases are known to occur throughout the plant kingdom (Simmons, 1994). (1,3;1,4)-b-Glucan is a major component of grass cell walls (Carpita, 1996) and endo-(1,3;1,4)-bglucanase appears to be involved in cell wall metabolism in certain tissues (Fincher, 1989). A possible physiological role of endo-(1,3;1,4)-bglucanase in endosperm has been studied extensively at the molecular level in germinating barley seeds (Hrmova and Fincher, 2001). Two barley endo-(1,3;1,4)-b-glucanase genes, EI and EII, are expressed in barley endosperm and their expression is regulated quite differently by phytohormones, depending on the tissue and its stage of development (Slakeski et al., 1990; Slakeski and Fincher, 1992a). Although both the EI and EII genes are expressed in the barley aleurone layer and scutellar epithelium, the EI gene is also expressed in vegetative tissues of barley seedlings, such as leaves and roots (Slakeski and Fincher, 1992b). A physiological role of the EI gene expressed in young vegetative tissues of barley seedlings may be difficult to explain. Although expression of the EI gene increased with a concomitant decrease in (1,3;1,4)-b-glucan content when barley seedlings were transferred from a light and dark cycle to continuous darkness, no measurable elongation of leaves was detected, suggesting that cell wall loosening did not occur (Roulin et al., 2002). However, there is other evidence to support an idea that hydrolysis of cell wall (1,3;1,4)-b-glucan by endo-(1,3;1,4)-b-glucanase in barley leaves may be associated with cell wall loosening and expan-

1815

sion (Sakurai and Masuda, 1978; Hoson and Nevins, 1989; Inouhe and Nevins, 1991; Carpita and Gibeaut, 1993). Rice seeds were proposed to contain an immunologically different endo-(1,3;1,4)-b-glucanase, because an antibody raised against the barley seed enzyme did not detect any strong bands of proteins in rice seed extracts in immunoblot analysis (Stuart et al., 1987). A rice b-glucanase gene, gns1, that shows high amino acid sequence identity to the proteins derived from the barley EI and EII genes has been isolated and shown to be expressed at significant levels in seedling shoots and roots, as well as in callus and immature seeds (Simmons et al., 1992). Another group reported that the rice gns1 gene is expressed only at very low levels in rice seedling caryopsis and coleoptiles (Romero et al., 1998). The two groups used different hybridization probes for RNA blot analysis, which may explain the difference in the observed levels of the rice gns1 mRNA in similar rice seedling tissues. The gns1 gene is expressed preferentially in rice seedling vegetative tissues, suggesting that the gene may be associated with cell wall metabolism in these tissues. To further investigate a physiological role of endo-(1,3;1,4)-b-glucanase in rice, we aimed to isolate previously uncharacterized endo-(1,3;1,4)b-glucanase genes. We cloned two rice endo(1,3;1,4)-b-glucanase genes, denoted OsEGL1 and OsEGL2, and their expression patterns were compared with that of the previously characterized rice gns1 b-glucanase gene (Simmons et al., 1992). In addition, both the OsEGL1 and OsEGL2 genes were successfully expressed in Escherichia coli to produce functional proteins, the enzymatic properties of which were determined.

Materials and methods Plant materials and treatments Rice (Oryza sativa L. cv Yukihikari) seeds were surface sterilized as described previously (Akiyama and Pillai, 2001). Seeds were germinated and grown under complete darkness for 10 d at 25 1C in sterile distilled water. For treatments, 10-d-old rice seedlings were transferred to solutions of phytohormones or chemical reagents. The rice seedlings were also treated under environmental stresses, such as cold, drought, salt, submergence and osmotic stresses. For mechanical wounding treatment, leaf portions of 10-d-old rice seedlings were gently crushed with an edge of the 2-mm thickness plastic ruler at intervals of approximately 1 cm

ARTICLE IN PRESS 1816 from the top to the bottom and then the seedlings were transferred to distilled water. Mature tissues were obtained from 7-week-old rice plants grown in a greenhouse. Seedling leaves, roots and endosperm were obtained from 10-d-old rice seedlings treated with or without phytohormones.

Isolation of OsGEL1 A cDNA library from rice seedling roots was constructed according to the method described previously (Bachem et al., 1996; Akiyama et al., 2004). A full-length barley endo-(1,3;1,4)b-glucanase EII cDNA (Fincher et al., 1986; Wolf 1991; GenBank accession nos. P12257; 1803523A) was labeled by random priming with 32P-dCTP (GE Healthcare) and hybridized to the membranes at 65 1C for 16 h in Rapid-hyb buffer (GE Healthcare). The blots were washed once in 0.1% SDS, 2  SSC for 20 min and once in 0.1% SDS, 1  SSC for 20 min at 65 1C, and exposed to x-Omat film (Eastman Kodak Co.). Positive clones were converted to pBluescript (SK) vector by in vivo excision according to manufacturer’s instruction (Stratagene).

Isolation of OsEGL2 The National Center for Biotechnology Information (NCBI) databases were searched by BLASTp with the OsEGL1 (GenBank accession no. AAV37460)deduced protein sequence as a query. The BLAST search detected another rice putative b-glucanase gene, denoted OsEGL2, in a rice genomic clone (GenBank BAB85436), which had 64% positional identity at the amino acid level with the OsEGL1 gene. The OsEGL2 gene was amplified by PCR with a pair of PCR primers designed based on a nucleotide sequence from the rice genomic clone found in databases (GenBank BAB85436): 50 -ACAAAGGTTGAAA TGGAATCT-30 (sense); 50 -TAATTGCTACTGTACCAACAG-30 (antisense). A PCR product of the expected size was purified, cloned into pGEM-T vector (Promega) and sequenced.

DNA sequence analysis The genes subcloned into the vectors were sequenced with the Dye Terminator II cycle sequencing kit (GE Healthcare) on an ABI-373 DNA sequencer (Perkin-Elmer Applied Biosystems Co.). Sequence data were compared with sequences in the NCBI databases by using the BLAST search program (Altschul et al., 1997). The multiple sequence alignment of the deduced protein sequences and calculation of the neighbor-joining

T. Akiyama et al. tree were performed with ClustalW (Thompson et al., 1994) and the phylogenetic tree was drawn with TreeView (Page, 1996).

RNA blot analysis Total RNA (20 mg per lane) was subjected to electrophoresis on a formaldehyde agarose gel and transferred to a Highbond-N+ nylon membrane (GE Healthcare) with 20  SSC (Sambrook et al., 1989). A gene-specific probe for OsEGL1 was amplified by PCR with a pair of primers designed based on the nucleotide sequence of the 30 -untranslated region (30 -UTR) of the OsEGL1 cDNA (GenBank accession no. AAV37460): 50 -TGCATTCCGTACACATATACG-30 (sense) and 50 -CTACGTGAGGCAAAATACCTC-30 (antisense). A full-length probe for the OsEGL2 gene was obtained by PCR, as described for its cloning. The probes were labeled by Rediprime II random priming with 32P-dCTP according to manufacturer’s instruction (GE Healthcare) and used for RNA blot hybridization as described previously. RT-PCR was done to confirm the expression of OsEGL2 seen in the databases, by reverse transcription of 50 ng of total RNA with SuperScript III reverse transcriptase (Invitrogen), followed by amplification of the cDNA with the primers: 50 -ATCAAGGTGACGACGTCGATC-30 (sense) and 50 -AATACGGGCATGGAAGCTAATG-30 (antisense) to generate a 573 bp band encoding the C-terminal part of the protein, which was qualitatively compared by ethidium bromide staining of the agarose gel.

Recombinant expression of the OsEGL1 and OsEGL2 proteins A region coding for the putative mature OsEGL1 protein was amplified by PCR with the OsEGL1 cDNA (GenBank accession no. AAV37460) as template and a pair of primers designed to introduce restriction sites at the ends of the predicted mature proteincoding region: 50 -GCCTCCGGATCCCAAAAGGCG GAG-30 (sense; the BamHI site is in bold print) and 50 -ATGCGTGAATTCGTACGGAATGCA-30 (antisense; the EcoRI site is in bold). A region encoding the putative OsEGL2 protein was amplified by PCR with a rice genomic DNA as template and a pair of primers designed to introduce restriction sites at the ends of the OsEGL2 protein-coding region: 50 -ATTTGAATTC GGTTGAAATGGAAT-30 (sense; EcoRI site in bold) and 50 -GCTAGTCGACCAACAGCACTGCAG-30 (antisense; SalI site in bold). The PCR-amplified OsEGL1 and OsEGL2 fragments were digested with the restriction enzymes and ligated into the restriction sites of the pGEX-4T-3 vector (GE Healthcare). The

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice pGEX-4T-3 vectors containing the OsEGL1 and OsEGL2 genes were transformed into DH5a E. coli cells. The transformed cells were grown in LB media containing 60 mg/mL of ampicillin, and protein expression was induced with 100 mM isopropyl-b-D-thiogalactosylpyranoside (IPTG) for 16 h at 30 1C. The cells were harvested by centrifugation at 10,000g for 5 min at 5 1C and the precipitated cells were kept at 80 1C until use.

Protein purification The bacterial cells were resuspended in phosphate-buffered saline (PBS), pH 7.2, and sonicated on ice for 10 min. The cell lysate was centrifuged at 10,000g for 20 min at 5 1C. The supernatant was applied to a glutathione-Sepharose 4B column (1.5  3 cm; GE Healthcare) pre-equilibrated with PBS. The column was washed with 10 volumes of PBS and bound protein was eluted with 50 mM Tris–HCl, pH 8.0, containing 10 mM reduced glutathione. Enzyme fractions were pooled, dialyzed extensively against PBS at 5 1C, concentrated in a Centricon YM-10 filter (Millipore, Bedford) and digested with 100 units of thrombin (GE Healthcare) in 1 mL of PBS for 16 h at 20 1C. The digested sample was passed through a glutathione-Sepharose 4B column and a benzamidine-Sepharose 6B column (1.5  3 cm; GE Healthcare) to remove glutathioneS transferase (GST) and thrombin, respectively. SDS-PAGE of proteins at each purification step was conducted on 12% polyacrylamide gels, as described by Laemmli (1970).

Enzyme assay In the reductometric assay, a standard reaction mixture comprising 0.25% (w/v) barley (1,3;1,4)-bglucan (Megazyme), 50 mM sodium acetate, pH 5.2, and enzyme in a total volume of 50 mL was incubated for 5 min at 37 1C. Enzyme activity was measured by monitoring an increase in reducing sugars by the method of Lever (1972). One unit of enzyme activity was defined as an amount of enzyme required to release 1 mmol glucose equivalents per min at 37 1C. In the viscometric assay, a standard reaction mixture consisting of 1.25% (w/v) barley (1,3;1,4)-b-glucan, 0.01% (w/v) sodium azide and 50 mM sodium acetate, pH 5.2, in a total volume of 10 mL was equilibrated for 1 h at 30 1C before the reaction was started by addition of 10 mL of enzyme solution. Aliquots were taken at indicated periods of time and the relative viscosity was determined as described by Huber and Nevins (1977). Viscosity units are indicated as time in s

1817

required for 1 mL of reaction mixture to evacuate a 1-mL measuring glass pipette at 30 1C. Protein concentrations were determined by the method of Bradford (1976) with bovine serum albumin as a standard (Pierce, Rockford).

Substrate specificity and mode of action To determine substrate specificity, reaction mixture containing 0.25% (w/v) b-glucans from different sources, including barley and oat (1,3;1,4)-b-glucan (Megazyme), Cetralia islandica lichenan (Sigma), Laminaria digitata laminarin (Sigma), Eisenia bicyclis laminarin (Tokyokasei, Tokyo), Alcaligenes faecalis curdlan (swollen beads; Wako, Osaka), and carboxymethyl (CM)cellulose (Sigma), 50 mM sodium acetate, pH 5.2, and enzyme in a total volume of 50 ml was incubated for 5 min at 37 1C. The increase in reducing sugars was measured as described above (Lever, 1972). To determine the action pattern, aliquots were withdrawn at the indicated periods of time and reaction products were analyzed by thin layer chromatography (TLC) as described previously (Akiyama et al., 1997).

Results and discussion We have previously purified and characterized two rice endo-(1,3;1,4)-b-glucanases from rice bran that differed in molecular mass and isoelectric point, suggesting that at least two genes encoding endo-(1,3;1,4)-b-glucanase may be present in rice (Akiyama et al., 1996, 1998). However, until now, only a single (1,3;1,4)-b-glucanase gene has been isolated from rice and studied (Simmons et al., 1992; Romero et al., 1998; Thomas et al., 2000a, 2000b). In this study, we have isolated two rice (1,3;1,4)-b-glucanase genes, denoted OsEGL1 and OsEGL2, and their expression patterns and biochemical properties were determined for comparison with those of previously reported rice and barley endo-(1,3;1,4)-b-glucanase genes (Simmons et al., 1992; Slakeski and Fincher, 1992a, 1992b).

Sequence analysis of the OsEGL1 and OsEGL2 genes The deduced amino acid sequences of the OsEGL1 and OsEGL2 precursor proteins were similar to many grass endo-(1,3;1,4)- and endo-(1,3)-b-glucanase genes. In phylogenetic analysis of their deduced amino acid sequences, OsEGL1, OsEGL2 and other grass endo(1,3;1,4)-b-glucanase proteins are classified into a

ARTICLE IN PRESS T. Akiyama et al.

common branch (I), while grass endo-(1,3)-bglucanases show relatively divergent protein sequences (Figure 1). The OsEGL1 precursor protein (100%; GenBank accession no. AAV37460) showed high amino acid sequence identities with the rice gns1 (97%; GenBank X58877; Simmons et al., 1992), barley EI (83%; GenBank CAB41401; Slakeski et al., 1990), barley EII (79%; GenBank P12257; Fincher et al., 1986; Wolf, 1991), wheat (84%; GenBank CAA80493; Lai et al., 1993) and oat (83%; GenBank CAA78834; Yun et al., 1993) endo-(1,3;1,4)-bglucanase precursor proteins. In contrast, the OsEGL2 protein (GenBank BAB85436) exhibited only 64% amino acid sequence identity with the OsEGL1 protein, although it retained many strikingly conserved sequences (Figure 2).

OsEGL1 (AAV37460) 100 gns1 (X58877) 98

100 63

OsEGL2 (BAB85436) gns9 (U72255) gns7 (U72253)

99 100

gns8 (U72254)

Osg1 (AB070742) 100

gns5 (U72251)

83

Expression of the OsEGL1 and OsEGL2 genes in germinating seeds and tissues

barley GIII (X67099) 100 96

barley GLU2 (S35156) gns6 (U72252)

OsGLN1 (AF337174) 95 barley GV (M96939) 50 gns2 (U72248)

( II )endo-(1,3) - -glucanase

In barley, two genes encoding the EI and EII endo(1,3;1,4)-b-glucanases have been isolated and characterized extensively (Fincher et al., 1986; Litts et al., 1990; Wolf, 1991; Loi et al., 1998). The barley EII gene is expressed in the aleurone layer and scutellum in germinated barley seeds, while the EI gene is expressed not only in germinated seeds, but also in developing vegetative tissues, such as young seedling roots (Slakeski and Fincher, 1992a). The association of the barley EI and EII gene expression patterns with seed germination prompted us to examine whether the OsEGL1 and OsEGL2 genes are expressed in rice seeds during germination. The OsEGL1 expression levels were increased after 6-d imbibition and remained high at least until day 14 (Figure 3a). In 10-d-old rice seedlings, OsEGL1 expression was observed to be highly detected in seedling roots and moderate in seedling leaves and endosperm (Figure 3b). In 7-week-old adult rice plants, a high level of OsEGL1 expression was observed in mature roots, whereas a low level was detected in flowers. The root-preferential expression pattern of the OsEGL1 gene in rice seedlings is, at least in part, consistent with the earlier observation that the rice gns1 b-glucanase gene is expressed constitutively in young roots of germinated rice seedlings (Simmons et al., 1992). Expression of the barley EI and rice OsEGL1 genes in young vegetative tissues of barley seedlings suggests that their gene products may act in cell wall metabolism in young growing tissues, apart from roles in endosperm mobilization in germinating seeds (Slakeski and Fincher, 1992a). It was speculated that the EI gene is likely to be associated with

barley EI (CAB41401) 99 wheat (CAA80493) 81 oat (CAA78834) 99 barley EII (P12257)

(I) endo-(1,3;1,4) - -glucanase

1818

OsGLN2 (AF443600) 100 gns4 (U72250) 67

barley GI (M96938) 100

barley GII (M23548) barley GIV (M96940)

0.1

gns3 (U72249)

Figure 1. Phylogenetic analysis of grass endo-(1,3;1,4)and endo-(1,3)-b-glucanase proteins. The deduced OsEGL1, OsEGL2 (shown in boxes), other grass endo(1,3;1,4)- and endo-(1,3)-b-glucanase protein sequences were aligned to calculate this neighbor-joining tree. As indicated by the scale bar (which shows the length for a substitution/site rate of 0.1%), the lengths of the horizontal branches are proportional to the number of base substitutions. Bootstrap values over 50% are shown at the branch points. Cluster I containing (1,3;1,4)-bglucanases and cluster II containing (1,3)-b-glucanases are indicated by vertical brackets and GenBank accession numbers are shown in parentheses.

formation of air spaces (aerenchyma) in the vegetative tissues of barley seedlings (Slakeski et al., 1990). The constitutive expression of OsEGL1 mRNA in young roots of 10-d-old rice seedlings and adult roots from 7-week-old mature rice plants

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice

1819

Figure 2. Amino acid sequence alignment of OsEGL1, OsEGL2 and other grass endo-(1,3;1,4)-b-glucanase proteins. The deduced protein sequences of rice OsEGL1 (GenBank accession no. AAV37460), OsEGL2 (GenBank BAB85436), rice gns1 (GenBank X58877), barley EI (GenBank CAB41401), barley EII (GenBank P12257), wheat (GenBank CAA80493) and oat (GenBank CAA78834) endo-(1,3;1,4)-b-glucanases were aligned with the ClustalW program package. Identical residues are indicated with an asterisk and conservative substitutions are indicated with a period or a colon.

suggests that the OsEGL1 gene may be involved in aerenchyma tissue formation in roots of paddy rice, where aerenchyma tissues may play an important role to carry oxygen from the plant parts above the water surface (Setter et al., 1997).

Little or no expression of the OsEGL2 gene was detected in germinated seeds and tissues from rice seedlings and mature rice plants (Figures 3a and b). However, inspection of the expressed sequence tags (ESTs) for this loci (Os01g0942300, Unigene ID

ARTICLE IN PRESS 1820

T. Akiyama et al. Germination 0

2

4

6

8

10

12

14 (d)

seen to increase in panicle from flowering up to 30 d after flowering (data not shown).

Effect of phytohormones and chemical reagents on the OsEGL1 and OsEGL2 gene expression in rice seedlings

OsEGL1

rRNA

Treatment of young barley leaves with indole acetic acid or gibberellic acid (GA) results in an increase in the levels of endo-(1,3;1,4)-b-glucanase EI mRNA (Litts et al., 1990; Slakeski et al., 1990; Slakeski and Fincher, 1992b). Addition of GA and ABA to barley young seedling leaves, roots and isolated aleurone layer also resulted in an increased level of EI mRNA in these tissues (Slakeski and Fincher, 1992a). This hormonal regulation of the barley EI and EII genes led us to investigate the effects of phytohormones and chemical reagents on

OsEGL2

rRNA

Ethephon

ABA

MeJA

Salicylate

2,4-D

GA

OsEGL1

Kinetin

H2O

Root

Seedlings

Leaf

Endosperm

Stem

Seedlings

Node

Root

Leaf sheath

Leaf blade

Flower

Mature plants

OsEGL1

rRNA

rRNA

OsEGL2

rRNA OsEGL2

Figure 3. Expression of the OsEGL1 and OsEGL2 genes in germinating rice seeds and tissues of 7-week-old rice plants. (a) Total RNA was extracted from whole seeds or seedlings germinated at 25 1C for the indicated period of time and subjected to RNA blot analysis. (b) Blot of RNA extracted from various tissues of 10-d-old rice seedlings or 7-week-old mature plants. The ethidium bromidestained gel shows the approximately equal RNA loading (20 mg per lane) before blotting.

Os.7551) in the NCBI database showed 24 ESTs from panicle and 9 from flower, corresponding to approximately 173 and 65 transcripts per million (tpm), respectively. ESTs were found at lower frequencies in seed, stem and leaf. This expression pattern was confirmed by RT-PCR, in which low signals were seen in most tissues, but the signal was

rRNA

Figure 4. Effects of phytohormones and chemical reagents on OsEGL1 and OsEGL2 gene expression in rice seedlings. RNA gel blots of 10-d-old rice seedlings treated for an additional 2 d with the indicated phytohormones or chemical reagents: H2O (control); kinetin (100 mM); GA, 100 mM gibberellin A3; 2,4-D, 100 mM 2,4-dichlorophenoxy acetic acid; salicylate, 100 mM sodium salicylate; MeJA, 100 mM methyl jasmonate; ABA, 100 mM abscisic acid and ethephon (10 mM). The radiolabeled gene-specific OsEGL1 30 -UTR and full-length OsEGL2 gene probes were used to detect their respective mRNA. The ethidium bromide-stained gels below the blots show the nearly equal RNA loading before blotting.

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice

Leaf

Root

Endosperm

Ethephon

Root

Leaf

Endosperm

H2O

Root

Leaf

Endosperm

ABA

Root

Leaf

Endosperm

H2O

Root

Leaf

Endosperm

MeJA

Root

Leaf

Endosperm

H2O

1821

72

96 (h)

OsEGL1

rRNA MeJA Leaf

0

12

24

48

ABA 72

96

0

12

24

Ethephon 48

72

96

0

12

24

48

OsEGL1

rRNA

Figure 5. Tissue specificity and time course of OsEGL1 gene expression in rice seedlings in response to MeJA, ABA and ethephon. (a) RNA gel blots of various tissues from 10-d-old rice seedlings treated with MeJA, ABA and ethephon for an additional 2 d, as in Figure 4. (b) RNA gel blots of leaves from 10-d-old rice seedlings treated with MeJA, ABA or ethephon for the indicated periods of time. The radiolabeled gene-specific OsEGL1 30 -UTR probe was used to detect its mRNA. The ethidium bromide-stained gel below the blots shows the nearly equal RNA loading before blotting.

the OsEGL1 and OsEGL2 gene expression in rice seedlings (Figure 4). Treatments with MeJA, ABA and ethephon significantly increased the OsEGL1 expression in rice seedlings, whereas the GA and 2,4-D treatments slightly increased the level of the OsEGL1 expression (Figures 4). Little or no expression of the OsEGL2 gene was observed in rice seedlings treated with any phytohormones or chemical reagents (Figure 4). Although treatments with MeJA, ABA and ethephon significantly increased the levels of OsEGL1 mRNA in rice seedling leaves, this regulation was tissue specific and the relatively high constitutive expression of OsEGL1 observed in rice seedling roots (Figure 3b) did not increase in response to the phytohormones (Figure 5a). The OsEGL1 transcript levels in rice seedling leaves increased as early as 12 h after treatments with MeJA, ABA and ethephon, reached the maximum level by 48–72 h and then decreased (Figure 5b). It was previously reported that gns1 b-glucanase mRNA expression in rice seedling shoots significantly increased in response to ethylene, salicylic acid and cytokinin, but not by auxin, GA and ABA (Simmons et al., 1992), while our results showed that level of OsEGL1 mRNA expression significantly increased in response to MeJA, ABA and ethephon, but not by kinetin and salicylic acid. Nonetheless,

inspection of the rice genome suggests that gns1 and OsEGL1 likely correspond to the same gene in the two cultivars. Since it has been reported that rice cultivars may have differences in responses to phytohormones (Das and Jat, 1977; Justin and Armstrong, 1991), the use of different cultivars may explain this discrepancy.

OsEGL1 gene expression and leaf elongation in rice seedling leaves in response to wounding Since ethylene, MeJA and ABA have been implicated in wounding and other environmental stress signaling in higher plants (Leo ´n et al., 2001), we also examined the effects of environmental stresses and wounding on expression of the OsEGL1 gene in 10-d-old rice seedling leaves. The OsEGL1 expression was most highly induced in response to mechanical wounding (Figure 6a). Salinity and submergence stresses also slightly increased the OsEGL1 expression levels in rice seedling leaves. OsEGL1 transcript levels increased by 12 h after wounding treatment, reached maximum levels by 72 h and then decreased (Figure 6b). These results are, at least in part, consistent with the previously

ARTICLE IN PRESS 1822

T. Akiyama et al.

Wounding

Mannitol

Submergence

Salt

Drought

Cold

Leaf

H2O

reported results on expression of the gns1 gene (Simmons et al., 1992). Mechanical wounding treatment of the leaf portions of 10-d-old rice seedlings also promoted the leaf elongation rate by 16% in comparison with that of control seedlings, when compared 4 d after the treatment (Figure 6c). It was proposed that partial hydrolysis of cell wall (1,3;1,4)-b-glucan by endo-(1,3;1,4)-b-glucanase may be involved in loosening the cell wall structure sufficiently to allow elongation, which is induced by an increased turgor pressure in growing tissues (Cosgrove, 1999).

OsEGL1

rRNA

In a dual assay, both changes in substrate viscosity and appearance of oligosaccharide products were determined with purified OsEGL1 and OsEGL2 proteins using barley (1,3;1,4)-b-glucan as substrate. The relative viscosity of barley (1,3;1,4)b-glucan in a reaction mixture dropped rapidly by 15% and 26% of their original viscosity within 10 min after the addition of purified OsEGL1 and OsEGL2, respectively (Figure 8a). In contrast,

0

12

24

48

72

96 (h)

OsEGL1

rRNA

25 Control 20 Leaf length (cm)

To determine whether the OsEGL1 and OsEGL2 genes encode active enzymes capable of hydrolyzing grass (1,3;1,4)-b-glucans, regions coding for putative mature OsEGL1 and OsEGL2 proteins were expressed as glutathione-S transferase fusion proteins in E. coli. The accumulations of the GSTOsEGL1 (62 kDa) and GST-OsEGL2 (64 kDa) fusion proteins were observed when E. coli cells containing pGEX4T-3-OsEGL1 and pGEX-4T-3-OsEGL2 vectors were induced with IPTG (Figures 7a and b, lanes 2). GST-OsEGL1 and GST-OsEGL2 fusion proteins were purified with a glutathione-Sepharose 4B column and eluted as major bands of 62 and 64 kDa, respectively (Figures 7a and b, lanes 3). The two fusion proteins were digested with thrombin and passed through glutathione-Sepharose 4B and benzamidine-Sepharose 6B columns to eliminate the GST tag and thrombin, respectively. The purified OsEGL1 and OsEGL2 proteins migrated as predominant bands of 33 and 35 kDa, respectively (Figures 7a and b, lanes 5).

Mode of action and substrate specificity of the OsEGL1 and OsEGL2 proteins Wounding

Leaf

Characterization of the OsEGL1 and OsEGL2 proteins expressed in E. coli

Wounding

15 10

5

0 0

24

48 Time (h)

72

96

Figure 6. Effect of abiotic stresses on OsEGL1 gene expression and promotion of the leaf elongation rate of rice seedlings in response to mechanical wounding. (a) RNA gel blot from leaf portions of 10-d-old rice seedlings treated with various abiotic stresses and wounding for an additional 2 d: H2O (control); cold (5 1C); drought (about 50% decrease in fresh weight); salt (0.3 M NaCl); submergence (full submergence in distilled water); mannitol (0.5 M); wounding. (b) Gel blot of RNA from the leaf portion of 10-d-old rice seedlings grown for the indicated periods of time after wounding treatment. The ethidium bromide-stained gel below the blots shows the nearly equal RNA loading before blotting. (c) Leaf portions of 10-d-old rice seedlings were crushed gently with a plastic ruler and the leaf elongation rate was compared with that of control rice seedlings at indicated periods of time after the wounding treatment. The error bars denote 7SE of three experiments, each with n ¼ 40–50.

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice

1

2

3

4

5

6

1823

14

kDa

OsEGL1

- 94 - 67 - 43

- 30

Relative viscosity (sec)

12

OsEGL2

10 8 6 4 2 0 0

10

20

- 20.1

30 40 Time (min)

50

60

- 14.4

1

2

3

4

5

6

G

C2

kDa C3

- 94 C4

- 67 - 43

OsEGL2

OsEGL1

- 30

0

15

30

45

60

0

15

30

45

60

SD

Time (min)

- 20.1

- 14.4

Figure 7. SDS-PAGE of OsEGL1 (a) and OsEGL2 (b) proteins expressed in E. coli. Proteins at each purification step were separated by SDS-PAGE on a 12% polyacrylamide gel. Lanes 1, proteins from E. coli cells containing pGEX-4T-3 vector with OsEGL insert without IPTG induction; lanes 2, proteins of E. coli cells containing pGEX-4T3 with OsEGL insert induced by IPTG; lanes 3, GST-fusion proteins purified with glutathione-Sepharose 4B; lanes 4, fusion proteins digested with thrombin; lanes 5, proteins after passing through the glutathione-Sepharose 4B and benzamidine-Sepharose 6B columns; lanes 6, molecular mass markers (kDa).

oligosaccharide products released from barley (1,3;1,4)-b-glucan treated with OsEGL1 and OsEGL2 proteins were not clearly detectable until 45 min after the initiation of the reaction (Figure 8b). The two oligosaccharides released have similar relative

Figure 8. Relative viscosity (a) and hydrolysis products (b) of barley (1,3;1,4)-b-glucan treated with purified OsEGL1 and OsEGL2 proteins. (a) Relative viscosity of reaction mixture was measured at the indicated times and viscosity units are presented as time in seconds required for the reaction mixture to evacuate a 1-mL glass pipette. (b) Four-microliter aliquots were taken from a reaction mixture after the indicated periods of time and oligosaccharide products were analyzed by silica gel TLC with 75% (v/v) acetonitrile. Standard (SD) includes glucose (G), cellobiose (C2), cellotriose (C3) and cellotetraose (C4).

mobilities on TLC to those reported for the hydrolysis products of barley (1,3;1,4)-b-glucan with barley endo-(1,3;1,4)-b-glucannase (Stone and Clarke, 1992; Lai et al., 1993). Substrate specificity of the purified OsEGL1 and OsEGL2 proteins was evaluated by comparing hydrolysis of several potential b-glucan substrates (Table 1). Both the OsEGL1 and OsEGL2 proteins specifically hydrolyzed (1,3;1,4)-b-glucan, including barley and oat b-glucans and C. islandica

ARTICLE IN PRESS 1824 Table 1.

T. Akiyama et al. Substrate specificity of purified OsEGL1 and OsEGl2 protein against potential b-glucans.

Substrates

Barley b-glucan (Hordeum vulgare) Oat b-glucan (Avena sativa) Lichenan (Cetraria islandica) Laminarin (Laminaria digitata) Laminarin (Eisenia bicyclis) Curdlan (swollen beads) (Alcaligenes faecalis) CM-cellulose a b

Major linkage typesa

1,4;1,3-b-(2.3-2.7:1) 1,4;1,3-b1,4;1,3-b-(2:1) 1,3;1;6-b-(7:1)) 1,3;1,6-b-(3:2) 1,3-b1,4-b-

Relative rate (%)b OsEGL1

OsEGL2

100 95 48 0 0 0 0

100 97 53 0 0 0 0

Ratios of linkage types are from Hrmova and Fincher (1993). Expressed as % of enzyme unit.

lichenan, but did not show activity against other potential b-glucan substrates, such as laminarin, curdlan and CM-cellulose. Therefore, we speculate that a possible role of OsEGL1 protein expressed in rice seedling leaves in response to wounding may be to hydrolyze high molecular weight (1,3;1,4)-bglucan in the cell wall matrix and to loosen the cell wall to allow cell elongation and expansion as a step to regenerate injured cell walls in wounded leaf tissues (Sakurai and Masuda, 1978; Hoson and Nevins, 1989). In summary, concomitant with a significant increase in the OsEGL1 transcript level in rice seedling leaves treated with mechanical wounding, the leaf elongation rate of rice seedlings increased in response to wounding treatment, suggesting that the OsEGL1 gene product may act in cell wall loosening and elongation by hydrolyzing (1,3;1,4)b-glucan in cell wall matrix on mechanical wounding. Expression of the OsEGL2 gene, in contrast, appeared to be the highest in ripening panicle and was nearly negligible in the tissues and conditions tested by RNA blot analysis, suggesting that it is not stress regulated in these tissues and plays a different developmental role than does OsEGL1.

References Akiyama T, Pillai MA. Molecular cloning, characterization and in vitro expression of a novel endo-1,3-bglucanase up-regulated by ABA and drought stress in rice (Oryza sativa L.). Plant Sci 2001;161:1089–98. Akiyama T, Kaku H, Shibuya N. Purification and partial characterization of an endo-(1,3;1,4)-b-glucanase from rice, Oryza sativa L. Biosci Biotechnol Biochem 1996;60:2078–80. Akiyama T, Kaku H, Shibuya N. Purification, characterization and NH2-terminal sequencing of an endo-1,3;1,4b-glucanase from rice (Oryza sativa L.) bran. Plant Sci 1998;134:3–10.

Akiyama T, Pillai MA, Sentoku N. Cloning, characterization and expression of OsGLN2, a rice endo-1,3-bglucanase gene regulated developmentally in flowers and hormonally in germinating seeds. Planta 2004; 220:129–39. Akiyama T, Shibuya N, Hrmova M, Fincher GB. Purification and characterization of a (1-3)-b-D-glucan endohydro]ase from rice (Oryza sativa) bran. Carbohydr Res 1997;29:365–74. Altschul SF, Thomas LM, Alejandro AS, Jinghui Z, Webb M, David JL. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucl Acids Res 1997;25:3389–402. Bachem CW, van der Hoeven RS, de Bruijin SM, Vreugdenhil D, Zabeau M, Visser RG. Visualization of differential gene expression using novel method of RNA fingerprinting based on AFLP: analysis of gene expression during potato tuber development. Plant J 1996;9:745–53. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 1976;72:248–54. Buckeridge MS, Rayon C, Urbanowicz B, Tine MAS, Carpita NC. Mixed linkage (1-3),(1-4)-D-glucans of grasses. Phytochemistry 2004;81:115–27. Carpita NC. Structure and biogenesis of the cell walls of grasses. Annu Rev Plant Physiol Plant Mol Biol 1996;47: 445–76. Carpita NC, Gibeaut DM. Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J 1993;3:1–30. Cosgrove DJ. Enzymes and other agents that enhance cell wall extensibility. Annu Rev Plant Physiol Plant Mol Biol 1999;50:391–417. Das DK, Jat RL. Influence of three soil–water regimes on root porosity and growth of four rice varieties. Agron J 1977;69:197–200. Fincher GB. Molecular and cellular biology associated with endosperm mobilization in germinating cereal grains. Annu Rev Plant Physiol Plant Mol Biol 1989;40: 305–46.

ARTICLE IN PRESS Wounding- and phytohormone-induced endo-(1,3;1,4)-b-glucanase in rice Fincher GB, Lock PA, Morgan MM, Lingelbach K, Wettenhall RE, Mercer JF, et al. Primary structure of the (1-3, 1-4)-b-D-glucan 4-glucohydrolase from barley aleurone. Proc Natl Acad Sci USA 1986;83:2081–5. Henrissat B. A classification of glycosyl hydrolase based on amino acid sequence similarities. Biochem J 1991; 280:309–16. Hoson T, Nevins DJ. b-D-Glucan antibodies inhibit auxininduced cell elongation and changes in the cell wall of Zea coleoptile segments. Plant Physiol 1989;90: 1353–8. Høy PB, Fincher GB. Molecular evolution of plant b-glucan endohydrolases. Plant J 1995;7:367–79. Hrmova M, Fincher GB. Purification and properties of three (1-3)-b-D-glucanase isoenzymes from young leaves of barley (Hordeum vulgare). Biochem J 1993; 289:453–61. Hrmova M, Fincher GB. Structure–function relationships of b-D-glucan endo- and exohydrolases from higher plants. Plant Mol Biol 2001;47:73–91. Hrmova M, de Gori R, Smith BJ, Fairweatherd JK, Driguez H, Varghese JN, et al. Structural basis for broad substrate specificity in higher plant b-D-glucan glucohydrolases. Plant Cell 2002;14:1033–52. Huber DJ, Nevins DJ. Preparation and properties of a b-Dglucanase for the specific hydrolysis of b-D-glucans. Plant Physiol 1977;60:300–4. Inouhe M, Nevins DJ. Inhibition of auxin-induced cell elongation of maize coleoptiles by antibodies specific for cell wall glucanases. Plant Physiol 1991;96:426–31. Justin SHFW, Armstrong W. Evidence for the involvement of ethene in aerenchyma formation in adventitious roots of rice (Oryza sativa L.). New Phytol 1991;118: 49–62. Lai DM, Høj PB, Fincher GB. Purification and characterization of (1-3,1-4)-b-glucan endohydrolases from germinated wheat (Triticum aestivum). Plant Mol Biol 1993;22:847–59. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. Leo ´n J, Rojo E, Sa ´nchez-Serrano JJ. Wound signaling in plants. J Exp Bot 2001;52:1–9. Lever M. A new reaction for colorimetric determination of carbohydrates. Anal Chem 1972;47:273–9. Litts JC, Simmons CR, Kareer EE, Huang N, Rodriguez RL. The isolation and characterization of a barley 1,3-1, 4-b-glucanase gene. Eur J Biochem 1990;194:831–8. Loi L, Ahluwalina B, Fincher GB. Chromosomal location of genes encoding barley (1-3,1-4)-b-glucan 4-glucanohydrolases. Plant Physiol 1998;87:300–2. Page RD. TreeView: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 1996;12:357–8. Romero GO, Simmons C, Yaneshima M, Doan M, Thomas BR, Rodriguez RL. Characterization of rice endo-b-

1825

glucanase genes (Gns2-Gnsl4) defines a new subgroup within the gene family. Gene 1998;223:311–20. Roulin S, Buchala AJ, Fincher GB. Induction of (1-3, 1-4)b-D-glucan hydrolases in leaves of dark-incubated barley seedlings. Planta 2002;215:51–9. Sakurai N, Masuda Y. Auxin-induced changes in barley coleoptile cell wall composition. Plant Cell Physiol 1978;19:1217–23. Sambrook J, Fritsch EF, Maniatis TA. Molecular cloning: a laboratory manual, second ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989. Setter TL, Ellis M, Laureles EV, Ella ES, Senadhira D, Mishra SB, et al. Physiology and genetics of submergence tolerance in rice. Ann Bot Suppl 1997;69:67–77. Simmons CR. The physiology and molecular biology of plant 1,3-b-D-glucanases and 1,3;1,4-b-D-glucanases. Critic Rev Plant Sci 1994;13:325–87. Simmons CR, Litts JC, Huang N, Rodriguez RL. Structure of a rice b-glucanase gene regulated by ethylene, cytokinin, wounding, salicylic acid and fungal elicitors. Plant Mol Biol 1992;18:33–45. Slakeski N, Fincher GB. Developmental regulation of (1-3, 1-4)-b-glucanase gene expression in barley. Plant Physiol 1992a;99:1226–31. Slakeski N, Fincher GB. Barley (1-3,1-4)-b-glucanase isoenzyme EI gene expression is mediated by auxin and gibberellic acid. FEBS Lett 1992b;306:98–102. Slakeski N, Baulcombe DC, Devos KM, Ahluwalia B, Doan DNP, Fincher GB. Structure and tissue-specific regulation of genes encoding barley (1-3,1-4)-b-glucan endohydrolases. Mol Gen Genet 1990;224:437–49. Stone BA, Clarke AE. Chemistry and biology of (1-3)b-glucans. Victoria, Australia: La Trobe University Press; 1992. Stuart IM, Loi L, Fincher GB. Immunological comparison of (1-3,1-4)-b-glucan endohydrolases in germinating cereals. J Cereal Sci 1987;6:45–52. Thomas BR, Romero GO, Nevins DJ, Rodriguez RL. New perspectives on the endo-beta-glucanases of glycosyl hydrolase family 17. Biol Macromol 2000a;27: 139–44. Thomas BR, Inouhe M, Simmons CR, Nevins DJ. Endo1,3;1,4-b-glucanase from coleoptiles of rice and maize: role in the regulation of plant growth. Biol Macromol 2000b;27:145–9. Thompson JD, Higgins DG, Gibson TJ. ClustalW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positionsspecific gap penalties and weight matrix choice. Nucl Acids Res 1994;22:4673–80. Wolf N. Complete nucleotide sequence of a Hordeum vulgare gene encoding (1-3, 1-4)-b-glucanse isoenzyme II. Plant Physiol 1991;96:1382–4. Yun SJ, Martin DJ, Gengenbach BG, Rines HW, Somers DA. Sequence of a (1-3,1-4)-b-glucanase cDNA from oat. Plant Physiol 1993;103:295–6.