Purification and characterization of a
b
-glucosidase from rye
(
Secale cereale
L.) seedlings
Masayuki Sue *, Atsushi Ishihara, Hajime Iwamura
Di6ision of Applied Life Sciences,Graduate School of Agriculture,Kyoto Uni6ersity,Kyoto606-8502, Japan
Received 29 November 1999; received in revised form 14 January 2000; accepted 14 January 2000
Abstract
Cyclic hydroxamic acids and a glucosidase that occur in rye seedlings were investigated. The concentration of the glucoside of 2,4-dihydroxy-1,4-benzoxazin-3-one (DIBOA-Glc) in shoots increased soon after germination and decreased to a lower, constant level as the plants started autotrophic growth. Cyclic hydroxamic acid glucosideb-glucosidase activity also occurred transiently, and the timing of the increase and decrease was concurrent with that of cyclic hydroxamic acid glucosides. The glucosidase was isolated from 48-h-old rye shoots and purified to apparent homogeneity by using isoelectric precipitation, anion exchange chromatography, and gel filtration. The isoelectric point and the optimum reaction temperature were 4.9 – 5.1 and 25 – 30°C, respectively. The N-terminal amino acid sequence was almost identical to that of the wheat glucosidases, but did not show any similarity to the sequences of other glucosidases of plant origin. SDS – and native – PAGE analyses showed that rye had several isozymes of glucosidase, and each isozyme was an oligomer of 60-kDa monomers with a molecular mass of 300 kDa. The
enzyme was highly active not only for DIMBOA-Glc but also for its 7-demethoxy analogue, DIBOA-Glc, which was different from the specificities of maize and wheat glucosidases. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Cyclic hydroxamic acid;b-Glucosidase; Rye (Secale)
www.elsevier.com/locate/plantsci
1. Introduction
Cyclic hydroxamic acids (Hxs), the most abun-dant secondary metabolites in the Gramineae fam-ily including wheat, maize, and rye, are considered to play an important role in the defense of plants against pests and herbivores [1]. The biosynthetic pathway of Hx branches from that of tryptophan at the point of indole [2 – 4]. In intact plants, Hxs are stored as 2-O-b-D-glucopyranosides, and on
the disruption of tissue by microbial and insect attacks, b-glucosidase (EC 3.2.1.21), which exists in a different cellular compartment from that of the glucosides [5], comes into contact with the glucosides and hydrolyzes them to release toxic
aglycones [1]. Hx glucosides (HxGlcs) and HxGlc glucosidases have been shown to occur in maize and wheat at high levels soon after germination [6 – 8]. Furthermore, some enzymes involved in the biosynthetic pathways of Hx in maize and wheat have been found to be highly expressed in this growth stage [4,9 – 11].
In maize and wheat, 2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one (DIMBOA) is the predomi-nant Hx species whereas the major one in rye is 2,4-dihydroxy-1,4-benzoxazin-3-one (DIBOA) [1]. The characterization of the enzymes involved in Hx biosynthesis has been carried out mainly on maize and wheat. It is thus important to elucidate whether or not the enzymes in rye have similar features to those in maize and wheat. In this study, we purified a b-glucosidase that appears with the DIBOA-Glc in rye seedlings to compare its properties with those of maize and wheat.
* Corresponding author: Tel.: +81-75-7536396; fax: + 81-75-7536408.
E-mail address:[email protected] (M. Sue)
2. Material and methods
2.1. Preparation of hydroxamic acids
DIBOA-Glc and DIMBOA-Glc were extracted from shoots of 3-day-old rye (Secale cerealeL.) and shoots of 3-day-old maize (Zea mays L.), respec-tively, and their aglycones, DIBOA and DIMBOA, were prepared by hydrolyzing the glucosides as described previously [8].
Hydroxy-1,4-benzoxazin-3-one (HBOA) and 2-hydroxy-7-methoxy-1,4-benzoxazin-3-one (HM-BOA) were synthesized chemically, and their glucosides, HBOA-Glc and HMBOA-Glc, were prepared by reducing DIBOA-Glc and DIMBOA-Glc, respectively, according to the method of Honkanen and Virtanen [12].
2.2. Plant materials
Rye (S. cereale L. cv. haru-ichiban) seeds were germinated on a sheet of wet paper and incubated at 25°C with a 12-h period of illumination (60 W/m2).
2.3. Purification of b-glucosidase
All operations were carried out at 4°C. About 5 g of shoots from rye seedlings (48 h after seeding) were frozen in liquid nitrogen and ground to a powder, followed by homogenization in 25 ml of 50 mM sodium acetate, pH 6.0. After the extract was placed in a centrifuge at 15 000×gfor 20 min, the supernatant was ultrafiltered to obtain a crude enzyme solution. The pH of the enzyme solution was adjusted to 5.0 with 50% (v/v) acetic acid, and the solution was centrifuged at 20 000×g for 30 min. The precipitation was resuspended in a small vol. of 50 mM Bis-Tris – HCl buffer (pH 6.8) and loaded onto an DEAE – Sepharose (Pharmacia) column (3 ml) that had been equilibrated with a 50 mM Bis-Tris – HCl buffer (pH 6.8). The proteins were eluted stepwise with 0, 70, 350 and 500 mM NaCl. The 350 mM NaCl fraction was placed on a Mono Q HR 5/5 column equilibrated with a 50 mM Bis-Tris – HCl buffer (pH 6.8). The proteins were eluted with a linear gradient of NaCl (150 – 400 mM) at a flow rate of 1 ml/min. The fractions with
b-glucosidase activity were collected and further purified by gel filtration on a Superdex 200 HR 10/30 that was equilibrated with a 50 mM Bis-Tris –
HCl buffer (pH 6.8) containing 150 mM NaCl. To estimate molecular mass on the gel filtration column, we used the following proteins as stan-dards: ferritin (440 kDa), human IgG (160 kDa), transferrin (81 kDa), ovalbumin (43 kDa), and myoglobin (17.6 kDa). b-Glucosidase activity was measured using 1.5 mM of DIBOA-Glc and DIM-BOA-Glc as substrates in each purification step. We estimated protein contents by following the method of Bradford [13] using bovine serum albumin as a standard.
2.4. Enzyme assay
The activity ofb-glucosidase was measured in 100 mM citrate – 200 mM phosphate buffer (McIlvaine buffer) (pH 5.5) as described previously [8].
2.5. Electrophoresis and analysis of N-terminal amino acid sequence
For activity staining, the crude enzyme solution prepared from shoots of 48-h-old rye was elec-trophoresed (8% resolving gel) under a non-dena-turing condition (native – PAGE) at 4°C [14]. After electrophoresis, the gel was equilibrated in McIl-vaine buffer (pH 5.5) for 30 min at 4°C. Then, the gel was incubated in the same buffer containing both 0.5 mM 6-bromo-2-naphthyl-b-D
-glucopyra-noside, which was previously dissolved in small volume of hot acetone, and 10 mM Fast Blue BB salt in darkness for 2 h at 30°C with shaking. After staining, the gel was rinsed twice in distilled water and fixed in 10% (v/v) methanol containing 10% (v/v) acetic acid. We analyzed the polypeptides constituting b-glucosidase by SDS – PAGE (8% re-solving gels) after staining them with silver nitrate in the procedure described by Laemmli [15].
The proteins separated on the SDS – PAGE gel were blotted onto a polyvinylidenedifluoride mem-brane, and stained with Coomassie Brilliant Blue. The 60-kDa bands were cut off and the N-terminal sequence was analyzed (HP G1005A protein se-quencing system).
3. Results
3.1. Occurrence of Hx glucosides (HxGlcs) and HxGlc glucosidase
after seeding, and started to green 30 h after
seeding. DIBOA-Glc was the only detectable Hx species in the shoots of the rye cultivar, and it started to increase steeply after germination, reaching a maxima concentration 36 h after seeding (21 nmol/mg FW) (Fig. 1). As the plants started autotrophic growth, DIBOA-Glc decreased gradually to the lower constant level, which was about a half of the maximum (11 nmol/mg FW at 96 h after seeding). In roots, DIMBOA-Glc was detected along with DIBOA-Glc. DIBOA-Glc appeared with germination and attained its maximum concentration 24 h after seeding (7.1 nmol/mg FW). The concentration of DIMBOA-Glc increased continuously during the experimental period to reach an amount of 3.3 nmol/mg FW 60 h after seeding.
Fig. 2. Changes in HxGlc glucosidase activity in shoots (A) and roots (B)., DIBOA-Glc glucosidase; , DIMBOA-Glc
glucosidase.
Fig. 1. Changes in contents of cyclic hydroxamic acids and their glucosides in shoots (A) and roots (B) of rye seedlings.
, DIBOA; , DIBOA-Glc; , DIMBOA; ,
DIMBOA-Glc.
DIBOA-Glc and DIMBOA-Glc glucosidase activities started to appear with germination. These activities reached their maxima 36 h after seeding in both shoots and roots (in shoot, 410 and 320 pkat/mg FW for DIBOA-Glc and DIMBOA-Glc, respectively; in roots, 150 and 105 pkat/mg FW for DIBOA-Glc and DIMBOA-Glc, respectively) and eventually decreased to approximately a half of their maxima (Fig. 2).
3.2. Purification of b-glucosidase from rye seedlings
Fig. 3. Elution profiles ofb-glucosidase activities on Mono Q. Fraction volumes were 1 ml. Shadowing shows fractions subjected to further purification studies., DIBOA-Glc glu-cosidase;, DIMBOA-Glc glucosidase.
Fig. 4. SDS – PAGE analysis of proteins at various purifica-tion steps. Fracpurifica-tions exhibiting b-glucosidase activity were applied to SDS – PAGE and proteins were visualized by silver staining. Arrow at right indicates 60-kDa polypeptide posi-tion. M, marker protein; lane 1, crude enzyme solution; lane 2, isoelectric precipitation; lane 3, active fractions from DEAE – Sepharose; lane 4, active fractions from Mono Q; lane 5, active fractions from Superdex 200 (fractions 24 and 25 of Mono Q); lane 6, active fractions from Superdex 200 (fraction 15 of Mono Q).
The plant material was extracted with 50 mM sodium acetate (pH 6.0). Most of the glucosidase activity was precipitated when the pH of the crude enzyme solution was adjusted to 5.0 by adding 50% (v/v) acetic acid. After resuspending the pre-cipitates in 50 mM Bis-Tris – HCl buffer (pH 6.8), we performed anion exchange chromatography on the solution on a DEAE – Sepharose column. The protein was eluted stepwise with 70, 350 and 500 mM NaCl, and the active fraction (350 mM NaCl) was further purified on a Mono Q column. As shown in Fig. 3, DIBOA-Glc and DIMBOA-Glc glucosidase activities were eluted in several peaks, and the elution profiles of both activities were identical. The fractions corresponding to the ma-jor peak (fractions 24 and 25 in Fig. 3) were collected and gel filtrated on a Superdex 200 column. The b-glucosidase activity was eluted at 10.3 ml, which corresponded to a molecular mass of 300 kDa. The HxGlc glucosidase was
purified 45-fold for both substrates,
DIBOA-Glc and DIMBOA-DIBOA-Glc (Table 1), and the SDS – PAGE analysis of the most active fraction gave a single 60-kDa band on the gel stained with silver nitrate (Fig. 4). The 20 residues in N-terminal region of the rye glucosidase were Gly – Thr – Pro – Ser – Lys – Pro – Ser – Glu – Pro – Ile – Gly – Pro – Val – Phe – Thr – Lys – Leu – Lys – Pro – Trp. The se-quence of the first 12 residues was identical to those of wheat glucosidase monomers with the exception of the seventh residue, Ser, which is Ala in wheat [8]. However, the sequence of the rye glucosidase did not show any similarity to the sequences of glucosidases from other plants than wheat. The glucosidase in the second major peak
Table 1
Purification of a b-glucosidase from rye shoots
Specific activity (nkat/mg protein) Recovery (%) Purification (fold)
DIBOA-Glc DIMBOA-Glc DIBOA-Glc DIMBOA-Glc DIBOA-Glc DIMBOA-Glc
42.0 53.0
Crude 100 100 1.0 1.0
77.1 94.6 2.6
Isoelectric precipitation 110.4 172.0 3.2
223
165.3 73.5 78.1 3.9 4.2
DEAE–Sepharose
19.2 22.6
Mono Q 1033 1538 24.6 28.9
43.6 45.8
9.2 9.6
1922 2322
on the Mono Q chromatogram (fraction 15 in Fig. 3) was purified with the Superdex 200 to give the same results as above: 300 kDa on gel filtration
and a single band on SDS – PAGE gel with a molecular mass of 60 kDa (Fig. 4).
We analyzed the protein profile of each active fraction on Mono Q by SDS – PAGE and native – PAGE. The SDS – PAGE analysis showed that all of the fractions had the 60-kDa polypeptide (Fig. 5A). On the native – PAGE gel, the mobilities of
the proteins corresponded to molecular masses of
300 kDa, but were slightly different between the
fractions; the enzymes in latter fractions had larger mobilities (Fig. 5B). The results confirmed the electrical heterogeneity of the rye glucosidase. When freshly prepared crude extract was elec-trophoresed under a non-denaturing condition and stained for glucosidase activity, a few bands (zones) were observed on the gel with the same mobility as the glucosidases in the Mono Q frac-tions (Fig. 5B)
3.3. Characterization of purified b-glucosidase
The b-glucosidase was characterized using the purified enzyme from the major peak (fractions 24 and 25) on the Mono Q chromatography.
The optimum pH, the isoelectric point, and the optimum temperature for the DIBOA-Glc and DIMBOA-Glc glucosidase were 5.5, 4.9 – 5.1 and 25 – 30°C, respectively. The effects of several metal cations, EDTA, and castanospermine were exam-ined (Table 2). A sodium acetate buffer (pH 5.5) was used as the reaction medium in place of a citrate/phosphate buffer, since some cations formed insoluble salts in a buffer containing phos-phate. The activity was completely inhibited by 0.2 mM castanospermine, a potent inhibitor of glu-cosidase, but was not affected by EDTA. A mono-valent cation, Ag+, resulted in a complete
inhibition at a concentration of 200 mM, and Cu2+ inhibited strongly the activity of the rye
glucosidase (98% inhibition at 0.5 mM). Among other divalent cations, Zn2+ decreased the activity
to 66% of its maximum.
We used the compounds in Table 3 to investi-gate the substrate specificity of the glucosidase. DIMBOA-Glc was shown to be the best substrate among the compounds tested (Vmax=4952 nkat/
mg protein, Km=0.617 mM). Its 7-demethoxy
analogue, DIBOA-Glc, was also hydrolyzed effi-ciently. Although the Vmax value for DIBOA-Glc
was highest, the Km value was twice as large as
that for DIMBOA-Glc (Vmax=5870 nkat/mg
protein, Km=1.19 mM). The Vmax values for
lac-tam glucosides, 2-hydroxy-1,4-benzoxazin-3-one (HBOA-Glc) and 2-hydroxy-7-methoxy-1,4-ben-zoxazin-3-one (HMBOA-Glc), were 1420 and 1005 nkat/mg protein, respectively, and the Km values
were 2.0 and 0.89 mM, respectively. Esculin and artificial substrates, pNP-b-glucose and pNP-b
-fu-Fig. 5. SDS – PAGE (A) and native – PAGE (B) of Mono Q fractions. Numbers on lanes in both figures correspond to fraction number. In (A) arrow at right indicates 60-kDa polypeptide position. In (B) C is the crude enzyme solution stained forb-glucosidase activity.
Table 2
Effects of EDTA, castanospermine, and metal cations on
b-glucosidase activity
Relative activity (%)
Control 100
EDTA (0.5 mM) 101 Castanospermine (0.2 mM) 0 Ag+(0.2 mM) 0
1.6 Cu2+(0.5 mM)
Mn2+(0.5 mM) 89.7
92.7 Fe2+(0.5 mM)
Zn2+(0.5 mM) 66.6
Ca2+(0.5 mM) 101
Table 3
Kinetic parameters of ryeb-glucosidase
Substratea K pNP-b-xyloside 78.2
0.616 1671
pNP-b-fucoside
pNP-a-glucoside 11d
32.4e
apNP,p-nitrophenol. bNot detected. cActivity at 4.7 mM. dActivity at 2 mM. eActivity at 0.05 mM.
tants [8,16]. This should not only be due to the difference in the isoelectric point between the en-zymes, because the isoelectric points of maize, wheat, and rye glucosidases were not so different (the pIs of maize, wheat, and rye glucosidases were 5.2, 5.1 – 5.6 and 4.9 – 5.1, respectively). The rye glucosidase may be more hydrophobic than the glucosidases of maize and wheat. On the Mono Q chromatogram, several peaks of DIBOA-Glc and DIMBOA-DIBOA-Glc glucosidases were detected (Fig. 3), and the elution profile of both activities were almost identical. These results indicate that rye has several isozymes ofb-glucosidase and there is little difference in their activities for DIBOA-Glc and DIMBOA-Glc. This property is reflected in the results shown in Fig. 1, where the changes in DIBOA-Glc glucosidase activities were parallel to changes in DIMBOA-Glc glucosidase activities in both shoots and roots, with similar extents of activity.
The results of the Superdex 200 column chro-matography and SDS – PAGE analysis indicated that the glucosidases were oligomers with molecu-lar masses of 300 kDa consisting of 60-kDa
monomers. Because an estimation of the molecu-lar mass according to gel filtration is not accurate, we could not define the number of the monomers comprising the native glucosidase oligomers. All of the glucosidase isozymes separated on the Mono Q chromatogram were also shown to be oligomers of 60-kDa polypeptides with native molecular mass of 300 kDa (Fig. 5). However,
the slightly different mobilities on a native – PAGE gel (Fig. 5B) indicated that they are electrically heterogenous. Esen and Cokums [17] have shown that glucosidase in maize crude extracts show elec-trophoretic multiplicity when incubated at 25 and 37°C for several hours. They concluded that the several distinct bands generated at lower pH (pH 4 – 6) were due to the action of SH-proteases, and that more anodic diffuse bands (zone) generated at a higher pH (\pH 6) were due to the action of
another class of proteases which were not inhib-ited by any of the protease inhibitors tested. In our experiment, freshly prepared extracts dis-played activity bands (zone) with the same mobili-ties as the glucosidases in the Mono Q fractions (Fig. 5B). Therefore, the charge heterogeneity of the rye glucosidase is not likely to be a proteolytic artifact. The maize glucosidase is a homodimer of 60-kDa polypeptides, and is encoded by a highly cose, were also hydrolyzed efficiently, but pNP-a
-glucose and salicin were hardly or not at all hy-drolyzed. Among the flavonoid and isoflavonoid glucosides, genistein glucoside was hydrolyzed with good efficiency (837 nkat/mg protein).
4. Discussion
The predominant Hx species in rye is known to be DIBOA-Glc, and it was the only detectable Hx species in shoots of the rye cultivar we used. DIBOA-Glc and DIMBOA-Glc was in roots. The concentration of DIBOA-Glc started to increase after germination, and decreased gradually to a lower constant level as the plant started au-totrophic growth. The HxGlc glucosidase activities were concurrent with HxGlcs, especially in the shoots. The transient occurrence of Hxs and HxGlc glucosidase activities were similar to results reported in maize and wheat [6 – 8], suggesting that the three Gramineae plants have a common mech-anism in producing Hxs and glucosidase during this stage of growth.
superna-polymorphic locus. This causes maize hybrid lines to show a few activity bands on native – PAGE gels [18]. The locus encoding the glucosidase and the polymorphism of the glucosidase has not been reported for rye. However, considering the fact that the rye glucosidase is an oligomer and con-sists of only 60-kDa polypeptides, the several bands on a native – PAGE gel may be caused by a similar mechanism. The glucosidases from wheat have also been shown to have several isozymes that can be clearly separated on native – PAGE [8]. Although the wheat glucosidases are comprised of two different subunits, 60- and 58-kDa polypep-tides, these electrophoretic multiplicities and the similar native molecular masses may indicate that the glucosidases from both plants resemble each other.
The sequence of the first 12 residues in N-termi-nal region of the rye glucosidase was identical to the sequences of wheat glucosidase monomers with the one exception [8], suggesting a close relationship between the rye and wheat dases. However, the sequence of the rye glucosi-dase did not show any similarity to the sequences of glucosidases from other plants than wheat. Manyb-glucosidases, including maize glucosidase, are assigned to the family 1 glucosidases according to the similarity of amino acid sequences around the active sites [19]. The N-terminal region is one of the least conserved regions in plant glucosidase sequences, and the three crystal structures of the family 1 glucosidases have shown that the N-ter-minus is located on the surface of the enzymes and does not have defined secondary structures [20 – 22]. Therefore, the difference in the N-terminal amino acid sequence does not necessarily mean a great difference in the total amino acid sequence between rye glucosidase and other family 1 glucosidases.
The optimum pH and the isoelectric point for the rye glucosidase were comparable to results reported for maize and wheat glucosidases [8,16], but the optimum temperature was lower than that for the maize enzyme, 50°C [16]. Among the metal ions, Ag+ and Cu2+ were strong inhibitors
al-though Cu2+ reportedly has no effect on the
maize glucosidase [16]. The Vmax values of the rye
glucosidase for DIBOA-Glc and DIMBOA-Glc were four to five times higher than those for lactam glucosides, HBOA-Glc and HMBOA-Glc, while the Km values were 1.4 to 1.7 times smaller.
These results indicated that the N-4-hydroxy group is important for the exhibition of higher activity in the enzymes and it has a larger influence on the Vmax value than Km value, as has been
shown for maize and wheat glucosidases [8]. On the other hand, the influence of 7-methoxy group on the rye glucosidase was different from that on the other two glucosidases. The maize and wheat enzymes showed much higher substrate specific-ities for DIMBOA-Glc and HMBOA-Glc than for their respective 7-demethoxy analogues, DIBOA-Glc and HBOA-DIBOA-Glc [6,8], whereas the rye glucosi-dase did not display large differences between 7-methoxy and 7-demethoxy compounds. Further-more, the rye glucosidase showed relatively higher activities for such compounds as pNP-b-glucose, pNP-b-fucose, esculin, and genistein-Glc than maize and wheat glucosidases. Therefore, the rye enzyme is considered to have broader substrate specificity than maize and wheat glucosidases. This is interesting with respect to the fact that the predominant Hx species in maize and wheat is DIMBOA-Glc while the major one in rye is DIBOA-Glc.
Analyses of the total amino acid sequence and of the genomic DNA encoding the rye glucosidase would provide us with a clue to what causes the difference in the substrate specificities and the multiplicity on a native – PAGE gel, and would provide us with an insight into evolutional differ-ences between the three plants.
References
[1] H. Niemeyer, Hydroxamic acids (4-hydroxy-1,4-benzox-azin-3-ones), defence chemicals in the Gramineae, Phyto-chemistry 11 (1988) 3349 – 3358.
[2] S.R. Desai, P. Kumar, W.S. Chilton, Indole is an inter-mediate in the biosynthesis of cyclic hydroxamic acids in maize, Chem. Commun. (1996) 1321.
[3] M. Frey, P. Chomet, E. Glawischnig, C. Stettner, S. Gru¨n, A. Winklmair, W. Eisenreich, A. Bacher, R.B. Meeley, S.P. Briggs, K. Simcox, A. Gierl, Analysis of a chemical plant defense mechanism in grasses, Science 277 (1997) 696 – 699.
[4] D. Melanson, M.-D. Chilton, D. Masters-Moore, W.S. Chilton, A deletion in an indole synthase gene is respon-sible for the DIMBOA-deficient phenotype of bxbx maize, Proc. Natl. Acad. Sci. USA 94 (1997) 13345 – 13350.
[5] A. Esen, D.A. Stetler, Subcellular localization of maize
[6] K. Ebisui, A. Ishihara, N. Hirai, H. Iwamura, Occur-rence of 2,4-dihydroxy-7-methoxy- 1,4-benzoxazin-3-one (DIMBOA) and ab-glucosidase specific for its glucoside in maize seedlings, Z. Naturforsch. 53c (1998) 793 – 798. [7] E. Nakagawa, T. Amano, N. Hirai, H. Iwamura, Non-induced cyclic hydroxamic acids in wheat during juvenile stage of growth, Phytochemistry 38 (1995) 1349 – 1354. [8] M. Sue, A. Ishihara, H. Iwamura, Purification and
char-acterization of a hydroxamic acid glucosideb-glucosidase from wheat (Triticum aesti6umL.) seedlings, Planta 210
(2000) 432 – 438.
[9] B.A. Bailey, R.L. Larson, Maize microsomal benzoxazi-none N-monooxygenase, Plant Physiol. 95 (1991) 792 – 796.
[10] M. Frey, R. Kliem, H. Saedler, A. Gierl, Expression of a cytochrome P450 gene family in maize, Mol. Gen. Genet. 246 (1995) 100 – 109.
[11] J. Tanabe, M. Sue, A. Ishihara, H. Iwamura, Occurrence and characterization of 2-hydroxy-1,4-benzoxazin-3-one and indole hydroxylases in juvenile wheat, Biosci. Bio-technol. Biochem. 63 (1999) 1614 – 1617.
[12] E. Honkanen, A.I. Virtanen, The synthesis of precursor II of benzoxazolinone formed in rye plants, and the enzymic hydrolysis of precursor I, the glucoside, Acta Chem. Scand. 14 (1960) 504 – 507.
[13] M.M. Bradford, A rapid sensitive methods for the quan-titation of microgram quantities of protein utilizing the principle of protein – dye binding, Anal. Biochem. 72 (1976) 248 – 254.
[14] B.J. Davis, Disc electrophoresis. II. Methods and appli-cation to human serum proteins, Ann. N.Y. Acad. Sci. 121 (1964) 404 – 427.
[15] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 277 (1970) 680 – 685.
[16] A. Esen, Purification and partial characterization of maize (Zea mays L.) b-glucosidase, Plant Physiol. 98 (1992) 174 – 182.
[17] A. Esen, C. Cokums, pH- and temperature-dependent
b-glucosidase multiplicity in maize (Zea mays L.) is a proteolysis artifact, Plant Sci. 74 (1991) 17 – 26.
[18] C.W. Stuber, M.M. Goodman, F.M. Johnson, Genetic control and racial variation ofb-glucosidase isozymes in maize (Zea mays L.), Biochem. Genet. 15 (1977) 383 – 394.
[19] B. Henrissat, A classification of glycosyl hydrolases based on amino acid sequences similarities, Biochem. J. 280 (1991) 309 – 316.
[20] T. Barrett, C.G. Suresh, S.P. Tolley, E.J. Dodson, M.A. Hughes, The crystal structure of a cyanogenicb -glucosi-dase from white clover, a family 1 glycosyl hydrolase, Structure 3 (1995) 951 – 960.
[21] W.P. Burmeister, S. Cottaz, H. Driguez, R. Iori, S. Palmieri, B. Henrissat, The crystal structures of Sinaps alba myrosinase and a covalent glycosyl – enzyme inter-mediate provide insights into the substrate recognition and active-site machinery of anS-glycosidase, Structure 5 (1997) 663 – 675.
[22] J. Sanz-Aparicio, J.A. Hermoso, M. Martı´nez-Ripoll, J.L. Lequerica, J. Polaina, Crystal structure ofb -glucosi-dase A fromBacillus polymyxa: insights into the catalytic activity in family 1 glycosyl hydrolases, J. Mol. Biol. 275 (1998) 491 – 502.
