INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant embryogenesis represents the most definitive stages of the plant life cycle, with the overall architectural pattern of the mature organism established during a relatively short interval (Thomas, 1993). Embryogenesis is typically thought of as a morphogenic phase characterized by simple cell division and a subsequent phase in which new cells expand and differentiate. Arrays of genes are expressed in a highly coordinated manner to execute embryo development (Goldberg et al., 1994; Meinke, 1995). The expression and function of genes have been analyzed intensively to understand how changes in metabolism govern structure of cells during differentiation and embryogenesis.

Steward et al. (1958) first demonstrated that the undifferentiated carrot (Daucus carota) callus possesses the ability to form a somatic embryo and that a single cell is totipotent. Since then, somatic embryogenesis has been an attractive model for studying gene expression in plant development and differentiation. Although it has been shown that cultured cells from many other plants can be regenerated into whole plants, the carrot system is still the favored model, and it is used frequently. Consequently, a large number of genes associated with carrot somatic embryogenesis have been identified. Some of these genes are responsible for encoding glycosylated extracellular proteins (Kreuger and van Holst, 1993; Thomas, 1993; Zimmerman, 1993).

Developmental events and exposure to exogenous stresses alters structural, compositional, and physical properties of plant cell walls. Many proteins secreted into the apoplast appear to play roles in environmental sensing, signal transduction, defense reaction, and determination of cell shape (Showalter, 1993). Like carrot, proteins secreted specifically into apoplast have been found during the development of other plant somatic embryos. However, little is known about the genes responsible for secreted proteins with essential functions in embryogenesis.

Cordewener et al. (1991) claimed that a cationic peroxidase (POX) was required for carrot somatic embryogenesis. They found that tunicamycin, an inhibitor of glycosylation of secreted glycoproteins, blocked the somatic embryogenesis before the globular stage. This inhibition can be overcome by the administration of functional cationic POXs of the type secreted during normal somatic embryogenesis. However, the functional characteristics or substrate specificities of the POXs were never clarified. POX catalyzes many oxidation-reduction reactions in the presence of H₂O₂ as an oxidizing agent. Numerous POX isozymes exist in plants, and they are mainly localized in cell walls and vacuoles (Carpin et al., 1999). POXs are induced by wounding (Bernards et al., 1999), pathogens (Lamb and Dixon, 1997), protoplast regeneration (de Marco and Roubelakis-Angelakis, 1996), and plant development (Brownleader et al., 2000). POXs may act on phenylpropanoids in the synthesis of lignin or they may mediate cross-linking of polysaccharides or extensions of cell walls, and hence serve an important role in cell wall construction and in the regulation of cell wall plasticity (Fry, 1982a, 1982b, 1986; Whetten et al., 1998; Brownleader et al., 2000).

We used suspension-cultured asparagus (Asparagus officinalis L. cv Y6) cells to identify proteins secreted specifically into apoplast during embryogenesis. Asparagus is readily regenerated from cultured cells into whole plants (Kohmura et al., 1994). Because asparagus is a monocot, somatic embryogenesis is somewhat different from that of carrot. In carrot embryogenesis, a globular cell is formed first, and then the pattern is shifted to the heart and torpedo stage, whereas in asparagus, such clear transitions are not observed. Globular somatic embryos of asparagus extend their axis, and then root and leaf tissues differentiate.

In this work, the gene for cell wall-bound POX (AoPOX1) expressed in the asparagus somatic embryo was cloned. Using recombinant enzyme, the in vitro substrate specificity was determined. This enzyme may contribute to neolignan synthesis. We propose the function of the gene product in embryogenesis.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Growth of Cultured Cells and Formation of Somatic Embryos

Explants of asparagus embryogenic callus (EC) subcultured on an agar plate were transferred to suspension culture in Murashige and Skoog medium with or without 2,4-dichlorophenoxyacetic acid (2,4-D). In the absence of 2,4-D, somatic embryos were formed after about 14 d (Fig. 1). Cells cultured in the presence of 2,4-D developed into amorphous clusters but failed to initiate somatic embryos. The increase in fresh weight of cells grown with and without 2,4-D commenced after d 7 and continued through d 14 (Fig. 2). Cells without 2,4-D grew more rapidly than those with 2,4-D; after 14 d, the fresh weight of cells in the absence of 2,4-D was about four times greater than cells treated with 2,4-D. The total number of the cells in each culture flask was estimated from cell fresh weight and mean cell size. After 14 d, the cell number in the absence of 2,4-D (6.75 × 10⁷ cells flask¹) was three times more than when 2,4-D was present (2.22 × 10⁷ cells flask¹).

View larger version (88K): [in this window] [in a new window]   Figure 1.   Suspension-cultured asparagus cells after 14 d. A, Cells of somatic embryo in the absence of 2,4-D. B, Cells in the presence of 2,4-D. Bar = 250 µm.

View larger version (14K): [in this window] [in a new window]   Figure 2.   Changes in the fresh weight of suspension-cultured asparagus cells in the absence or presence of 105 M 2,4-D. Mean values and SEs (n = 3) are shown.

Differential Expression of Cell Wall Proteins

Wall proteins were extracted from 14-d-old cells cultured in the presence or absence of 2,4-D and fractionated by chromatography on a cation-exchange column using stepwise elution. The individual protein fractions were further resolved by SDS-PAGE. The SDS-PAGE of the 200 mM NaCl eluted fraction is shown in Figure 3. Removal of 2,4-D specifically expressed some proteins (e.g. 62, 34, and 31 kD) and suppressed the others (32, 20, and 17.5 kD). The expression of the bottom band of the doublet (38 kD) was particularly enhanced in the absence of 2,4-D (Fig. 3, arrowhead). The 15 amino acids at the N terminus (DGSLTPQFYDHSCPR) of the protein band were identified.

View larger version (23K): [in this window] [in a new window]   Figure 3.   Differential expression of asparagus cell wall proteins in the presence or absence of 2,4-D. Crude cell wall proteins were fractionated by a cation exchange column and NaCl stepwise elution (100, 200, 300, and 400 mM) and then 0.5 µg of proteins of each fraction was subjected to SDS-PAGE. Proteins were visualized by silver-staining. The 200 mM NaCl fraction was shown. Arrow indicates AoPOX1 protein.

Cloning of AoPOX1

A sense primer was designed based on the amino acid sequence and was used for reverse transcriptase (RT)-PCR, and the full-length cDNA sequence was determined (Fig. 4A; accession no. AB042103 designated as AoPOX1). The AoPOX1 clone predicted a 325-amino acid sequence containing a signal peptide at the N terminus (29 amino acids) with a predicted pI value of 8.72. AoPOX1 showed moderate similarity with many homologs of plant POX (Fig. 4B). The derived amino acid sequence for AoPOX1 showed similarities to the Arabidopsis expressed sequence tag (EST) X98320 (74.1%) and X98804 (61.0%). Homologs with other monocots and dicots showed fewer similarities (35%-47%).

View larger version (31K): [in this window] [in a new window]   Figure 4.   cDNA sequence of AoPOX1. A, Nucleotide sequence and deduced amino acid sequence of AoPOX1 cDNA clone. A bold line indicates a position of signal peptide. A double broken line indicates an N-terminal amino acid sequence derived from secreted AoPOX1 protein. A sense primer designed according to N-terminal amino acid sequence of mature AoPOX1 protein and used for RT-PCR is hybridized with the position of S1. An antisense primer was designed according to A1 sequence for 5'-RACE-PCR. The fragment between S2 and A2 was amplified for expression of recombinant AoPOX1 protein. Two arrowheads indicate SacI restriction site, and the fragment between them was used as template for probe of northern analysis. Broken lines show the possible N-glycosylation sites. B, Phylogenetic tree of plant POXs. Phylogenetic tree was obtained from the alignment of the putative amino acid sequences by a Clustal method. Gene accession number is shown in parentheses of each gene.

Expression of AoPOX1 during Embryogenesis

Northern analysis was performed with an 800-bp cDNA probe obtained by RT-PCR (Fig. 5A). AoPOX1 transcripts were scarcely expressed in undifferentiated callus. In the absence of 2,4-D, transcripts increased 8 h after incubation and continued up to d 7. The signal intensity on d 7 was 13 times higher than callus (Fig. 5B). After d 14 the signal diminished. In the presence of 2,4-D a transient peak in expression of AoPOX1 was observed at 24 h. After d 1 the expression in the presence of 2,4-D exponentially decreased through d 14. 

View larger version (39K): [in this window] [in a new window]   Figure 5.   Expression of AoPOX1 during asparagus somatic embryogenesis. A, Northern-blot analysis. Twenty micrograms of total RNA was separated and hybridized with AoPOX1 cDNA probe. 28S rRNA was stained by methylene blue as an index of amount of applied RNA. B, Signal intensity of each bands is shown by the ratio to that of callus.

Expression and Purification of Recombinant AoPOX1 Protein

To analyze its function, the AoPOX1 protein was extracted from cell walls of asparagus somatic embryos, and purified with column chromatography (Fig. 6A). However, The amount and quality of concentrated enzyme were not adequate for further biochemical experiments. Therefore, the AoPOX1 protein was expressed in Escherichia coli as a 6× His-tagged recombinant protein. Expression of recombinant protein was induced by the addition of isopropyl--D-thiogalactoside. Calcium was added to facilitate accurate folding. After sonication and centrifugation, POX activity was detected in the supernatant with guaiacol as substrate. Recombinant AoPOX1 protein was purified by cation-exchange and Ni-affinity column chromatography (Fig. 6B). Because recombinant AoPOX1 protein was not glycosylated, its M_r was slightly smaller than the native enzyme. Recombinant enzyme was used to examine substrate specificity.

View larger version (44K): [in this window] [in a new window]   Figure 6.   SDS-PAGE of native and recombinant AoPOX1 protein. A, AoPOX1 protein was extracted from somatic embryo cells and purified by three steps of column chromatography. The eluants of each step were applied to SDS-PAGE. The gel was silver-stained. AoPOX1 protein was indicated by black arrowhead. B, His-tagged recombinant AoPOX1 protein was expressed in E. coli and purified using a cation exchange column and Ni-affinity column. Purified AoPOX1 protein was applied to SDS-PAGE. The gel was stained with Coomassie Brilliant Blue (CBB) reagents. Recombinant AoPOX1 protein was indicated by white arrowhead.

Substrate Specificity and Product Analysis

Selected monolignols (coniferyl alcohol and sinapyl alcohol) and phenolic acids (ferulic acid, sinapic acid, p-coumaric acid, and caffeic acid) were tested as substrates for recombinant AoPOX1 protein by stopped-flow spectrometry (Table I). Of these substrates, coniferyl alcohol and ferulic acid were oxidized more effectively than others. Product separation with a C₁₈ reversed-phase HPLC system showed coniferyl alcohol was converted by AoPOX1 to a product with a distinctly different elution time (Fig. 7, A and B). The native AoPOX1 protein also converted coniferyl alcohol to the same product (Fig. 7C). The reaction product with ferulic acid was also analyzed by the HPLC system, but no reaction product was detected, and the peak height of ferulic acid detected at 280 and 320 nm was simply reduced, suggesting that AoPOX1 converted ferulic acid to a compound without absorption at 280 and 320 nm. The product of coniferyl alcohol eluted at 16.9 min was collected, silylated, and applied to gas chromatography-mass spectrometry (GC-MS) to determine its structure (Fig. 8A). The m/z 574 ion was the molecular ion. The mass spectrum showed that coniferyl alcohol had been converted to a dimer by AoPOX1 protein, and the dimer had three hydroxyl groups silylated. The structure was also confirmed by ¹H-NMR (Fig. 8B). ¹H-NMR spectrum was consistent with the spectrum of dehydrodiconiferyl alcohol (DDCA; Hirai et al., 1994).

View this table: [in this window] [in a new window]   Table I.   Substrate specificity of AoPOX1 Maximum catalytic rates were determined by measuring the initial rate of substrate disappearance with a range of substrate concentrations (0-400 mM) and H2O2 (4 mM). The recombinant AoPOX1 protein was used at a 0.1 µg mL1 final concentration. Figures in parentheses are relative Vmaxapp to coniferyl alcohol.

View larger version (6K): [in this window] [in a new window]   Figure 7.   Enzyme product analysis. A, Coniferyl alcohol (indicated by white arrowhead). B, The reaction mixture of coniferyl alcohol converted by recombinant AoPOX1 protein. C, The reaction mixture of coniferyl alcohol reacted with purified AoPOX1 protein extracted from somatic embryo. DDCA was indicated by black arrowhead.

View larger version (16K): [in this window] [in a new window]   Figure 8.   Characterization of recombinant enzyme product. A, Mass spectrum of the product. B, 1H-NMR spectrum. The peaks of solvent (CDCl3, S) were indicated.

Amount of Lignin

The lignin content of the cell walls was measured using the Klason (KL) method and the acetyl bromide (AB) method (Table II). The amount of lignin in EC measured by the AB method accounts for 6% of cell wall dry matter. Growth of cells in the presence of 2,4-D did not alter the amount of lignin on d 7 and only slightly increased the amount on d 14. In the absence of 2,4-D in the medium, the lignin content decreased to 25% by d 7. On d 14, the lignin content increased somewhat but was still much lower than that of walls of cells grown in the presence of 2,4-D. KL also showed similar results. On d 7, KL in the absence of 2,4-D was 46% of the contents in the presence of 2,4-D. On d 14, almost the same amount of KL was present in the presence or absence of 2,4-D.

Measurement of DDCA in the Culture Media

The 7-d culture medium in the absence of 2,4-D was collected and extracted with chloroform liquid-liquid extraction. DDCA was detected in this extract by GC-MS (Fig. 9A). To measure the level of DDCA in the culture medium, chloroform fraction was applied to GC-SIM analysis (Fig. 9B). The relative level of DDCA extracted from medium was estimated from the ion intensity at m/z 484 observed by GC-SIM comparing with DDCA produced by recombinant AoPOX1 protein. The estimated content of DDCA in the 7-d culture medium in the absence of 2,4-D was 1.33 × 10⁸ M. However, DDCA could not be detected in the 7-d culture medium in the presence of 2,4-D (Fig. 9C).

View larger version (17K): [in this window] [in a new window]   Figure 9.   Determination of DDCA in the culture medium. A, GC-MS total-ion current trace of the cultured medium in the absence of 2,4-D on d 7. B, GC-selected ion monitoring (SIM) mass trace of m/z 484 in A. C, GC-SIM mass trace of m/z 484 of the culture medium in the presence of 2,4-D on d 7.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Somatic embryos are readily formed from asparagus callus when cells are cultured in hormone-free suspension medium. As was found with carrot (Zimmerman 1993), asparagus somatic embryogenesis is regulated by a relatively simple manipulation. Hence, this regeneration system has the potential to serve as a suitable model for studying development and differentiation in monocots. The cells grew faster, based on both fresh weight accumulation and cell number increase, in the absence of 2,4-D than in its presence. The results show prolific cell division in the absence of 2,4-D.

As cells shift to embryogenic development, a number of proteins are newly detected (Fig. 3). One of these secreted proteins is AoPOX1. A signal peptide sequence deduced from the cDNA sequence supports the conclusion that the protein is destined for localization in the cell wall. The molecular mass deduced from the amino acid sequence was 33 kD, which is somewhat less than the mass of the isolated of native AoPOX1 protein resolved by SDS-PAGE. Like other secreted proteins, this POX is likely to be glycosylated. The fact that the calculated pI value was 8.72 would account for the binding of this protein to a cation exchange column. AoPOX1 was classified as a plant POX by homology analysis. Two Arabidopsis EST clones exhibited the highest homology with AoPOX1, but the functions of these EST clones are not known (Tognolli et al., 2002). Other plant POXs had relatively low similarity with AoPOX1. Therefore, AoPOX1 lacks homology with POXs that have been characterized.

Northern analysis revealed that AoPOX1 transcripts increased after subculture both in the presence and absence of 2,4-D. However, in the presence of 2,4-D, AoPOX1 transcripts decreased exponentially after d 1. One explanation is that exposure of cells to fresh medium for subculture would provide oxidative stress. There are a number of reports that expression of some POXs is activated by oxidative stress (Lamb and Dixon, 1997; Morita et al., 1999). Hence, the short term stimulation of AoPOX1 is likely the result of transferring cells to fresh medium. Nevertheless, in the absence of 2,4-D, elevation of AoPOX1 gene transcripts continued from 8 h to d 7. Somatic embryogenesis promoted further increases in AoPOX1 transcript accumulation. Because the transcript level decreased by d 14, AoPOX1 expression seems to be restricted during the early stage of somatic embryogenesis. Using a cDNA encoding nearly full-length AoPOX1 protein, the recombinant AoPOX1 protein was expressed in E. coli. The extract of E. coli strain XL1-Blue showed no POX activity toward guaiacol as substrate. After transformation and induction of recombinant AoPOX1 protein, strong POX activity was observed. Fortunately, active recombinant AoPOX1 protein was soluble in the cytoplasm, so a refolding step was not required. AoPOX1 protein was purified and used for analysis of its function.

Somatic embryogenesis in carrot is inhibited by the glycosylation inhibitor, tunicamycin (Cordewener et al., 1991), suggesting that protein glycosylation is necessary for embryogenesis. Tunicamycin inhibition is reversed by supplying the cells with exogenous cationic POXs of the type secreted into medium during normal somatic embryogenesis. On the basis of the assumption that glycosylation is a prerequisite for the secretion, one can suggest that expression and secretion of functional cationic POXs into the apoplast is essential for the early stage of carrot somatic embryogenesis. AoPOX1 protein is also a putative cationic POX secreted into apoplast during early stages of asparagus somatic embryogenesis, and it might also play a role during the early stages of embryogenesis.

The reaction catalyzed by POX is both complex and fast and does not follow simple Michaelis-Menten kinetics. Therefore, we measured the initial rate of substrate depletion with a stopped-flow spectrophotometer under conditions of saturating H₂O₂ according to Bernards et al. (1999). Recombinant AoPOX1 protein oxidized feruloyl substrates (Table I). This result followed other results of cationic POX substrate specificity (Takahama, 1995; Melo et al., 1997; Bernards et al., 1999). Because coniferyl alcohol is a lignin precursor (Whetten et al., 1998), walls of cells cultured in the absence of 2,4-D, which had high AoPOX expression, were expected to exhibit an increase in lignin content. However, the amount of lignin was the same or even lower in the absence of 2,4-D than in its presence (Table II). Lignin is usually a more abundant component of secondary cell walls and would be expected to be synthesized in later stage of plant cell development (Fry, 1982a). Because cells of the somatic embryo are typically juvenile, it is unlikely that the lignin synthesis is activated during embryogenesis. Nevertheless, as cell fresh weight increases the cell wall and cell wall components were generally produced actively. Cells during somatic embryogenesis might not increase lignin content. Therefore the secreted AoPOX1 might reduce or keep the low level of lignin content of cell walls by depleting lignin precursors.

POX, monolignols, and H₂O₂ are necessary not only for lignin synthesis, but they also contribute to the synthesis of neolignan, a dimer of monolignol. GC-MS analysis showed that the product formed from coniferyl alcohol by recombinant AoPOX1 protein was a specific dimer of coniferyl alcohol, DDCA. Another possible form of dimer, pinoresinol, was not detected in the reaction medium. DDCA was also produced by AoPOX1 protein extracted from cell walls of asparagus somatic embryos, and it existed in the culture medium at sub-micromolar level concentration. DDCA is a neolignan that is thought to be a precursor of lignin. However, other physiological roles for DDCA have been suggested (Hirai et al., 1994). Dehydrodiconiferyl glucoside (DCG), one of the neolignans, was found to have a cytokinin-like activity in promoting cell division in tobacco (Nicotiana tabacum) callus at micromolar concentration (Binns et al., 1987; Teutonico et al., 1991). DDCA, the aglycone of DCG, was assumed the direct precursor of DCG (Orr and Lynn, 1992). We suggest that AoPOX1 protein be secreted into cell walls where it catalyzes dimerization of coniferyl alcohol in the synthesis of DCG. DCG produced by this mechanism may play a role in the activation of cell division and differentiation in asparagus cultures.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Materials and Culture Conditions

Asparagus (Asparagus officinalis L. cv Y6) EC cultures were maintained as previously described (Kohmura et al., 1994). The stable EC line was subcultured every 4 weeks on Murashige and Skoog (1962) medium containing 10⁵ M 2,4-D and 0.8% (w/v) agar at pH 5.8. At 3 weeks, an explant of the EC was suspension-cultured in an 100-mL flask containing 30 mL of Murashige and Skoog medium with or without 10⁵ M 2,4-D on a shaker (120 rpm) at 25°C in the light. After 14 d in suspension culture, the cells grown in the absence of 2,4-D developed into somatic embryos, but cells in the presence of 2,4-D did not.

Counting of Cell Number

Asparagus-cultured cells were harvested by filtration through 25-µm nylon mesh. Fresh weight of the sampled cells was measured, and the cells were fixed by 4% (w/v) formaldehyde in 20 mM cacodylic acid buffer (pH 7.2) at 4°C for 12 h. The fixed samples were dehydrated in a graded ethanol series (from 30%-100%, v/v), followed by the gradual replacement of absolute ethanol by Technovit 7100 (Kulzer, Wehrheim, Germany). The samples embedded in Technovit were sectioned with a microtome (MT-3, Nippon Medical & Chemical Instruments, Osaka). The magnified (×400) cell clusters were photographed with a digital camera (C-4040, Olympus, Tokyo) attached to an Olympus BX-41 microscope. The mean cell number (n) observed in the microscope field (0.08 mm²) was determined in 30 images. The total cell number in a flask (N) was calculated assuming that the density of cells is 1, as follows,
where FW is fresh weight of cells in a flask.

Preparation of Cell Wall-Bound Proteins

Proteins were extracted from cell walls by the method of Inouhe and Nevins (1991). The cultured cells were homogenized in 5 mM NaCl with 100 µM Pefabloc SC (Merck, Darmstadt, Germany) with a mortar and pestle. The homogenate was centrifuged at 1,000g for 10 min. The pellet was washed three times with deionized water and then finally suspended in 3 M LiCl with 10 µM Pefabloc SC at 4°C overnight. The LiCl extract was obtained by centrifugation and dialyzed against 20 mM sodium-acetate buffer (pH 5.4) for 72 h at 4°C and was used as a crude enzyme fraction.

Differential Display of Cell Wall Proteins and N-Terminal Amino Acid Sequence of AoPOX1 Protein

The LiCl protein fraction was subjected to cation exchange chromatography (SP-Toyopearl, Tosoh, Tokyo). Protein fractions were eluted with 20 mM sodium-acetate buffer (pH 5.4) in a stepwise gradient from 0 to 400 mM NaCl at increasing concentrations by 100 mM at each step. The protein of each fraction was applied to SDS-PAGE. The gel applied 200 mM NaCl fractions in the presence and absence of 2,4-D was electroblotted on a polyvinylidene difluoride membrane (Hybond-P, Amersham Biosciences UK, Little Chalfont, Buckinghamshire, UK) using a blotting apparatus (Horize Blot AE-6675, Atto, Tokyo). Membrane was stained with CBB reagent. The band of 38-kD protein expressed in the absence of 2,4-D (Fig. 3, arrowhead) was cut out and subjected to an automated Edman degradation sequencer (G1005A, Hewlett-Packard, Palo Alto, CA).

SDS-PAGE

All of SDS-PAGE was carried out using 12.5% (w/v) acrylamide gel as described by Laemmli (1970). Molecular mass markers ("DAIICHI".II, Daiichi Pure Chemicals, Tokyo) were used as standards. Gels were silver-stained or stained with CBB reagent.

Cloning of AoPOX1 cDNA

Total RNA was prepared from 14-d-old suspension-cultured (±2,4-D) asparagus cells using the ISOGEN (Nippongene, Tokyo), from which the poly(A⁺) fraction was isolated with the Oligotex-dT30 Super (Roche Diagnostics, Mannheim, Germany). Reverse transcriptase PCR was performed by RNA LA PCR Kit (Takara, Tokyo). The first-strand cDNA was synthesized with an oligo(dT)-adapter primer containing an M13 primer M4 sequence (contained in the RNA LA PCR Kit). For 3'-RACE-PCR, M13 primer M4 was used as antisense primer (5'-GTTTTCCCAGTCACACGAC-3', contained in the RNA LA PCR Kit). The sense primer was designed according to the N-terminal amino acid sequence of AoPOX1 (5'-ACNCCNCARTTYTAYGAYCA-3'). The nucleotide sequence of the 5' region was obtained by a 5'-RACE-PCR procedure. A homopolymeric tail was added using terminal deoxynucleotidyl transferase (Takara) and dGTP. PCR amplification was accomplished with an oligo(dC) primer (degree of polymerization = 18) used as adapter primer and a specific antisense primer (5'-GCAACCGCCTT CTCAACTAC-3').

The PCR products were subcloned into pGEM-T vector with a TA-cloning kit (Promega, Madison, WI), and the recombinant plasmid was transformed into Escherichia coli strain XL1-Blue. The clones were sequenced by a dideoxy method (Thermo Sequence II dye terminator cycle sequencing premix kit, Amersham Biosciences UK) with a sequencing system (373A DNA sequencing system, Applied Biosystems, Foster City, CA).

Alignment of Protein Sequences and Prediction of pI and M_r

The determined AoPOX1 cDNA sequence was subjected to homology analysis with the FASTA program (Lipman and Pearson, 1985). The region representing the signal peptide of POX homolog was predicted with a primary structure analysis program, SignalP (http://www.cbs.dtu.dk/ services/SignalP/), and then removed for the alignment. The amino acid sequences of AoPOX1 and its homologs were aligned by a Clustal method, and the similarities were calculated with a help of Megalign (DNASTAR, Madison, WI). The amino acid sequence of AoPOX1 without the signal peptide was subjected to pI and M_r analysis with Compute pI/M_r tool program (http://www.expacy.ch/tools/pi_tools.html/).

Northern-Blot Analysis

The clone from RT-PCR was digested by SacI (Toyobo, Osaka) to obtain 0.7-kb cDNA fragment. This fragment without a poly(A) tail was used as a template for DNA probe for the detection of mRNA of the AoPOX1. The ³²P-labeled probes were prepared with a help of BcaBest labeling kit (Takara). Total RNA was separated on 1.2% (w/v) formaldehyde-agarose gel, blotted onto a nitrocellulose membrane (Nitrobind, Micron Separations, Westborough, MA). The blotted membrane was baked at 80°C for 2 h and preincubated at 42°C for 2 h in 50% (v/v) formamide, 5× SSPE (0.75 M sodium chloride, 50 mM sodium phosphate buffer [pH 7.4], and 5 mM EDTA disodium salt), 5× Denhardt's solution (0.1% [w/v] Ficoll, 0.1% [w/v] polyvinylpyrrolidone, and 0.1% [w/v] bovine serum albumin), 0.1% (w/v) SDS, and 0.1 mg mL¹ salmon sperm DNA. After pre-incubation, ³²P-labeled probe was added and hybridized at 42°C for 16 h. Then the membrane was washed twice with 2× SSC (0.15 M sodium chloride and 0.075 M sodium citrate) and 0.1% (w/v) SDS at 55°C for 10 min and twice with 0.5× SSC and 0.1% (w/v) SDS at 65°C for 15 min. The washed membrane was examined by an imaging analyzer (BAS-2000, Fuji Xerox, Tokyo). rRNA was stained by methylene blue for the estimation of amount of applied RNA.

Purification of AoPOX1 Protein from Somatic Embryo Cells

The crude enzyme fraction obtained from somatic embryo cell walls was purified through three different chromatographic protocols. First, the crude enzyme fraction was applied to an SP-Toyopearl column and eluted with a stepwise gradient of 0 to 400 mM NaCl by every 100 mM. Solid ammonium sulfate was added to the active fraction to 40% saturation. Then the fraction was loaded onto a Butyl-Toyopearl column (Tosoh). The column was eluted with a stepwise gradient from 40% to 0% saturation by every 10%. The active fraction was applied to a G3000SW HPLC column (0.8 × 30 cm, Tosoh) and eluted with 20 mM sodium-acetate buffer (pH 5.4) by HPLC system (LC-6A, Shimadzu, Kyoto). Fractions containing POX activity were pooled and stored at 4°C. The protein concentration was monitored in each purification step at A₂₈₀. POX activity was assayed with guaiacol as substrate. The reaction medium for enzyme assay contained 20 mM sodium-acetate buffer (pH 5.0), 4 mM H₂O₂, and 0.1% (v/v) guaiacol. The reaction was started by adding 100 µL of reaction medium to 10 µL of each enzyme fraction and was then measured at A₄₇₀.

Expression of Recombinant AoPOX1 Protein in E. coli

RT-PCR was performed using primers containing restriction site (sense primer, 5'-ATGCATGCTTTCTTGTTCTCATCATCTCCC-3', containing SphI restriction site; antisense primer, 5'-ATCTGCAGTCACCACCGCAACAACGTAC-3', containing PstI site). Amplified AoPOX1 cDNA was digested with restriction enzyme, and cloned into the expression vector pQE-30 (QIAexpressionist, Qiagen, Valencia, CA). The recombinant plasmid was transformed into E. coli strain XL1-Blue. LB-broth medium (800 mL) containing 0.1 mg L¹ ampicillin was inoculated with 16 mL of an overnight culture of cells. Cultures were grown at 20°C with 120 rpm shaking until the A₆₀₀ reached 0.7. Then 0.1 mM of isopropylthio--galactoside and 40 mM of CaCl₂ were added, and incubation continued for 24 h. Recombinant AoPOX1 protein was extracted with sonication (VP-5S, TAITEC, Saitama, Japan). The crude extract was dialyzed against 20 mM sodium-acetate buffer (pH 5.4) and was applied to an SP-Toyopearl column. Protein fractions were eluted with a stepwise gradient from 0 to 300 mM NaCl at increasing concentrations by 100 mM at each step. AoPOX1 protein was extracted between a 100 to 200 mM NaCl fraction. The active fraction was dialyzed against 50 mM sodium-phosphate buffer (pH 8.0) containing 300 mM NaCl and then applied to nickel-nitrilotriacetic acid agarose resin column (contained in the QIAexpressionist). Pure AoPOX1 protein was eluted by reducing the pH (pH 6.0).

Substrates

Guaiacol was purchased from Kanto Chemical (Tokyo). Coniferyl alcohol was from Sigma-Aldrich (St. Louis). Sinapyl alcohol and sinapic acid were from Aldrich Chemical Co. (Milwaukee). Ferulic acid, caffeic acid, and lignin (de-alkaline) were from Tokyo Chemical Industry (Tokyo). p-Coumaric acid was from Katayama Chemical (Osaka).

Enzyme Assays

Reaction with purified recombinant AoPOX1 protein was assayed as described by Rasmussen et al. (1995) and Bernards et al. (1999) using a stopped-flow spectrophotometer (RA-401, Union Giken, Osaka). Substrate (0-800 µM) in the reaction buffer (50 mM sodium-acetate buffer, pH 4.5, with 1 mM CaCl₂) was placed in one reservoir (3 mL), while the mixture of 3 mL of H₂O₂ (8 mM) and enzyme (0.2 µg mL¹) in reaction buffer was placed in the other. Solution was mixed by simultaneous injection into the flow cell. Initial rate from 5 to 10 s was measured by linear decrease in absorbance at A₂₆₀ for coniferyl alcohol and sinapyl alcohol, and A₃₁₀ for ferulic acid, sinapic acid, p-coumaric acid, and caffeic acid.

Product Formation and Structural Characterization

The reaction medium contained 1 mM coniferyl alcohol, 20 mM sodium-acetate buffer (pH 5.8), 4 mM H₂O₂, and 0.5 µg mL¹ purified POX. The reaction mixture was purged with N₂ gas and then incubated for 1 h at 25°C in the dark. After incubation, reaction mixture was subjected to HPLC separation with a C₁₈ column (3.9 mm × 15 cm, Waters, Milford, MA) and eluted with a linear gradient of 5% to 50% (v/v) acetonitrile/deionized water containing 20 mM acetic acid. The eluant from 16.5 to 17 min was collected. The collected product was evaporated to dryness by a stream of N₂ gas and stored overnight in desiccators. For GC-MS, an aliquot of residue was redissolved in 500 µL of acetonitrile. Three hundred and seventy microliters of N,O-bis(trimethylsilyl) trifluoroacetamide (GL Sciences, Tokyo) was added and incubated at 100°C for 30 min. A portion was introduced into a GC-MS system (GCMS-QP5000, Shimadzu). A CP-Sil 5CB capillary column (0.25 mm × 30 m, Varian Medical Systems, Palo Alto, CA) was used. The column was maintained at 60°C for 6 min and then the temperature was raised to 320°C at the rate of 10°C min¹. For ¹H-NMR, the reaction medium was spotted on a thin layer chromatography plate (silica gel 60 F₂₅₄, Merck). Developing solvent was chloroform:MeOH (5:1, v/v). The reaction product was eluted with chloroform. The purified product was dissolved into 600 µL of chloroform-d (Aldrich), and applied to ¹H-NMR system (JNM-A400, JEOL, Tokyo). Tetramethyl silane (Aldrich) at 0.05% (v/v) was added as standard. Spectra were recorded at 400 MHz.

Lignin Determination

Cell walls, from which the POX had been extracted, were treated with 100 units of -amylase (Type VII A from porcine pancreas, Sigma-Aldrich) for 2 h at 37°C to remove starch. Hemicellulose and phenolic acids were removed with 1 M NaOH for 2 h. The precipitate was washed twice each with 0.03 M acetic acid, ethanol, ethanol:diethylether (1:1, v/v), and diethylether and then dried completely at 80°C. Lignin was extracted from the precipitate by AB (Tokyo Chemical Industry) and determined at A₂₈₀ (Morrison, 1972). KL lignin was determined as the residue remaining after two-step sulfuric acid hydrolysis of cell wall polysaccharides (Hatfield et al., 1994).

Extraction of DDCA from Culture Medium

The culture media in the presence or absence of 2,4-D on d 7 (2 L) were collected. NaCl was added to the culture medium to 2% (w/v). Then medium was partitioned with 600 mL of chloroform. The organic layer was concentrated in vacuo. Concentrated extract was dried under N₂ gas and then dissolved in 200 µL of acetonitrile. Five hundred microliters of N,O-bis(trimethylsilyl) trifluoroacetamide was added and incubated at 100°C for 30 min. Silylated sample was concentrated to 60 µL under the N₂ gas. One microliter of sample was applied to GC-SIM analysis. GC was maintained at 220°C for 10 min, and then the temperature was raised to 320°C at the rate of 2°C min¹. Three ions, m/z 574, 484, and 454, were monitored by SIM. Ten micrograms of DDCA produced by recombinant AoPOX1 protein in vitro was used for quantification of DDCA in the medium by GC-SIM. The amount of in vitro DDCA was estimated from A₂₈₀ observed by HPLC comparing with the area of known amount of coniferyl alcohol.