A xyloglucan-oligosaccharide-specific ?-d-xylosidase or exo-oligoxyloglucan-?-xylohydrolase from germinated nasturtium (Tropaeolum majus L.) seeds

Planta (1991) 184:137-147

P l ~ ' l t ~ 9 Springer-Verlag1991

A xyloglucan-oligosaccharide-specific a-o-xylosidase or exo-oligoxyloglucan-a-xylohydrolase from germinated nasturtium ( Tropaeolum majus L.) seeds Purification, properties and its interaction with a xyloglucan-specific endo-(1--+4)-fl-D-glucanase and other hydrolases during storage-xyloglucan mobilisation Crisfina Fanutti z, Michael J. Gidley 2, and J.S. Grant Reid 1. x Department of Biological and Molecular Sciences, Stirling FK9 4LA, and 2 Unilever Research Laboratories, Colworth House, Sharnbrook, Bedford MK44 1LQ, UK Received 30 August; accepted 11 October 1990

Abstract. The c(-xylosidase which is involved in the postgerminative mobilisation of xyloglucan in nasturtium seed cotyledons has now been purified to apparent homogeneity by a facile procedure involving lectin affinity chromatography. The purified enzyme, a glycoprotein, moved as a single band (apparent molecular weight 85000) on sodium dodecyi sulphate-gel electrophoresis, whilst isoelectric focusing gave a number of enzymatically active protein bands spanning the range pI = 5.0 to 7.1 (maximum activity at pI = 6.1). The enzyme did not hydrolyse the simple a-xylosides p-nitrophenyl-a-Dxylopyranoside and isoprimeverose (~-D-Xyl(1--+6)D-Glc), or polymeric tamarind-seed xyloglucan. It released xylose from a complex mixture of oligosaccharides produced by exhaustive hydrolysis of tamarind seed xyloglucan using the xyloglucan-specific endo-(1--*4)-fl-Dglucanase from germinated nasturtium seeds (M. Edwards et al. 1986, J. Biol. Chem., 261. 9489-9494). The three xyloglucan oligosaccharides o f lowest molecular size were purified from this mixture and were shown by 1H-nuclear magnetic resonance ( t H - N M R ) and enzymatic analysis to have the structures: Gal Xyl

Xyl

Xyl

(A)

Glc--,GlcoGlc~Glc,

Xyl

Xyl

Xyl

(B)

GlcoGlc~GlcoGlc,

Gal Gal ~ $ Xyl Xyl Xyl (C) ~ ~ Glc~Glc~GlcoGlc, respectively, with linkages as in xyloglucans. The enzyme catalysed the release o f xylose from A, B, C and the other higher-molecular-weight saccharides generated from tamarind xyloglucan by the endo-fl-D-glucanase. It also released xylose from cellulase hydrolysates of tamarind xyloglucan. The products of ~-xsylosidase hydrolysis o f oligosaccharides A and C were characterised by 1 H - N M R as:

Xyl Xyl ~ ~, Glc~Glc~Glc~Glc,

Gal

Gal

Xyl

Xyl

Glc~Glc~Glc~Glc,

respectively. With oligosaccharide C as substrate the enzyme had a pH optimum at pH 5.0. The K,, values for oligosaccharides A and C were 0.62 m M and 0.32 mM, respectively, and the corresponding Vmaxvalues were in the ratio A : C = 1.00: 0.54. A model for the enzymatic depolymerisation of nasturtium-seed xyloglucan in vivo is proposed.

* To whom correspondence should be addressed

Abbreviations: Con A = Concanavalin A; DEAE = diethylaminoethyl; Gal = galactose; Glc = glucose; HPLC = high-performance liquid chromatography; Mr = apparent molecular mass; NMR = nuclear magnetic resonance; pI = isoelectric point; SDS-PAGE = sodium dodecyl sulphate-polyacrylamide gel electrophoresis; Xyl = xylose

Key words: Cell wall - Endo-(1--~4)-fl-D-glucanase (xyloglucan-specific) - Storage polysaccharide (seed) - Tropaeolum (xyloglucan mobilisation) - Xyloglucan - ~-Xylosidase

138 Introduction

I n m a n y d i c o t y l e d o n o u s seeds the m a j o r c a r b o h y d r a t e reserve is a cell-wall s t o r a g e p o l y s a c c h a r i d e o f the xyl o g l u c a n type. Seed s t o r a g e x y l o g l u c a n s consist o f a line a r (1----4) fl-linked o - g l u c a n (cellulosic) b a c k b o n e w h i c h carries a - D - x y l o p y r a n o s y l side residues a n d t w o - u n i t fl-Dg a l a c t o p y r a n o s y l - ( 1 - - , 2 ) - o - x y l o p y r a n o s y l side chains. B o t h types o f s u b s t i t u e n t are l i n k e d (1--~6)-a to glucose residues in the cellulosic b a c k b o n e ( R e i d 1985). T h e x y l o g l u c a n s w h i c h a r e p r e s e n t in the n o n - c e l l u l o s i c m a trix o f m a n y h i g h e r - p l a n t p r i m a r y cell walls are similar in s t r u c t u r e , b u t c o n t a i n in a d d i t i o n a p r o p o r t i o n o f L-fucose residues l i n k e d to s i d e - c h a i n g a l a c t o s e ( H a y a s h i 1989). The developmental process of xyloglucan mobilisat i o n after g e r m i n a t i o n h a s been m a p p e d at the u l t r a s t r u c t u r a l level in t a m a r i n d (Tamarindus indica L.) seeds (Reis et al. 1987), a n d s t u d i e d b i o c h e m i c a l l y in n a s t u r t i u m ( E d w a r d s et al. 1985). T h e activities o f f o u r h y d r o l y t i c enzymes, endo-(1---*4)-fl-D-glucanase, fl-D-galactosidase, a-D-xylosidase a n d fl-D-glucosidase, were l i n k e d with xyl o g l u c a n m o b i l i s a t i o n in n a s t u r t i u m ( E d w a r d s et al. 1985). T h e endo-(1--*4)-fl-glucanase has been purified to h o m o g e n e i t y a n d s h o w n to h a v e t o t a l specificity t o w a r d s x y l o g l u c a n s ( E d w a r d s et al. 1986). I t is t h e r e f o r e a novel t y p e o f e n z y m e , d i s t i n c t f r o m o t h e r endo-(1--,4)-fl-Dg l u c a n a s e s (cellulases) in its highly restricted s u b s t r a t e specificity. T h e f l - g a l a c t o s i d a s e has also been purified to homogeneity. Unlike fl-galactosidases from other sources this e n z y m e is c a p a b l e o f r e m o v i n g s i d e - c h a i n g a l a c t o s e residues f r o m x y l o g l u c a n m o l e c u l e s w i t h o u t a n y p r i o r d e p o l y m e r i s a t i o n o f the cellulosic b a c k b o n e ( E d w a r d s et al. 1988). T h e xyloglucan-specific endo-flg l u c a n a s e a n d the f l - g a l a c t o s i d a s e f r o m n a s t u r t i u m seeds h a v e been u s e d as specific c y t o c h e m i c a l p r o b e s for the ultrastructural localisation of xyloglucans by enzymeg o l d c y t o c h e m i s t r y ( V i a n et al. 1991). T h e y h a v e also been u s e d to i n v e s t i g a t e the effect o f c o n t r o l l e d s t r u c t u r a l m o d i f i c a t i o n o n the r h e o l o g i c a l p r o p e r t i e s o f t a m a r i n d x y l o g l u c a n ( R e i d et al. 1988), a n d in a s t u d y o f the s t r u c t u r e a n d s o l u t i o n p r o p e r t i e s o f the s a m e p o l y s a c c h a r i d e ( G i d l e y et al. 1991). I n this p a p e r we r e p o r t the p u r i f i c a t i o n to h o m o g e neity o f the a - D - x y l o s i d a s e f r o m g e r m i n a t e d n a s t u r t i u m seeds, d e s c r i b e s o m e o f its m o l e c u l a r p r o p e r t i e s a n d dem o n s t r a t e t h a t its c a t a l y t i c a c t i v i t y is a p p a r e n t l y restricted to x y l o g l u c a n o l i g o s a c c h a r i d e s f r o m w h i c h it r e m o v e s a single, s p e c i f i c a l l y - l o c a t e d xylose residue. W e f u r t h e r p r o p o s e a m o d e l for the e n z y m a t i c d e p o l y m e r i s a t i o n in vivo o f n a s t u r t i u m x y l o g l u c a n b a s e d o n the restricted specificities o f this e n z y m e a n d o f the endo-fl-D-glucanase ( E d w a r d s et al. 1986).

Material and methods

Materials. Nasturtium (Tropaeolum majus L.) seeds were purchased from Royal Sluis, Leyland, Preston, UK. Soluble tamarind-seed xyloglucan (Glyloid 3S) was obtained from the Dainippon Pharmaceutical Corporation, Osaka, Japan. The commercial prepara-

C. Fanutti et al.: ~-Xylosidase from nasturtium tion contained about 5 % glucose which was removed by dialysis and freeze-drying. Isoprimeverose (C~-D-Xylp-(1--*6)-D-Glc)was prepared from tamarind-seed xyloglucan by hydrolysis with the commercial enzyme preparation Driselase (Sigma, Poole, Dorset, UK) which contains enzyme activities capable of hydrolysing all the glycosyl linkages in the polysaccharide except the e-o-xylosyl linkage. A limit Driselase hydrolysate of a xyloglucan therefore contains monosaccharides plus the disaccharide isoprimeverose (Fry 1987). Tamarind xyloglucan (288 mg, final concentration 10 mg 9m l - 1) in 50 mM ammonium-acetate buffer (pH 5.0) was mixed with Driselase (600 mg) and the solution was incubated for 24 h at 30~ C. The progress of the reaction was monitored by TLC. When the reaction was complete (after 24 h) the mixture was heated to 100~ C for 15 min, and centrifuged (26000 99 for 30 min). The supernatant was freeze-dried, dissolved in the minimum volume of water and fractionated on a column (100 cm long, 2.2 cm i.d.) of Bio-Gel P2. Fractions were analysed for carbohydrate content by the phenolsulphuric acid method (Dubois et al. 1956) and examined by TLC. Those fractions which contained only disaccharide were pooled and freeze-dried to give a white powder (96 mg). The identity of the material was checked by acid hydrolysis (Saeman et al. 1945) and quantitative determination of the sugars released by gas-liquid chromatography (GLC) of their alditol acetates (Edwards et al. 1986). Only xylose and glucose were detected in equimolar amounts. The mixture of cello-oligosaccharides used as TLC reference standards was a gift from Dr. R. Sturgeon, Heriot Watt University, Edinburgh, UK. Enzymes for analytical purposes were purchased from Sigma, except for hexokinase which was from Boehringer, Mannheim FRG. General chemicals and buffer salts were of analytical grade or better.

General assays. The activity of c~-xylosidase was routinely assayed at pH 5.0 and 30~ C using either 50 mM ammonium-acetate buffer or McIlvaine phosphate-citrate (Dawson et al. 1982). The volatile ammonium-acetate system was always used in assays for TLC analysis. The activity of the enzyme was determined by assaying the free pentose released, as described below. Endo-fl-glucanase activity was determined using tamarind xyloglucan as substrate as before (Edwards et al. 1986). The activities of fl-galactosidase and fl-glucosidase were determined using the appropriate p-nitrophenyl glycoside as substrate (Edwards et al. 1985). Galactose was determined using D-galactose dehydrogenase (Edwards et al. 1985), glucose using hexokinase and glucose-6-phosphate dehydrogenase (Bergmeyer et al. 1974), and pentose by a modification of the p-bromoaniline procedure of Roe and Rice (1948) (Edwards et al. 1985). Protein was determined quantitatively by the dye-binding procedure of Sedmak and Grossberg (1977) using bovine serum albumin as standard. Total carbohydrate was determined by the phenol-sulphuric method (Dubois et al. 1956) or by the anthrone method (Dische 1962) using an appropriate monosaccharide or mixture of monosaccharides as standard. The reducing power of saccharide-containing solutions was determined by the direct ferricyanide method (Halliwell and Riaz 1970). Wherever possible, published assays were scaled down to allow them to be carried out in 1.5 ml-capacity microcentrifuge tubes. General separation methods. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out essentially according to Laemmli (1970), as described elsewhere (Edwards et al. 1986). Gels were stained for protein using Coomassie Blue, and for carbohydrate using periodic acid and Schiff's reagent (Fairbanks et al. 1971). Electroblots on nitrocellulose (Schleicher and Schuell, Dassel, FRG) were prepared from gels using the buffer system of Towbin et al. (1979) and the Hoefer (Newcastle-underLyme, UK) mini transfer apparatus. Proteins were detected on blots using Amido Black. Glycoproteins which were capable of binding Concanavalin A (Con A) were localised on electroblots by treating them first with Con A (Sigma), and then with horseradish peroxidase (a glycoprotein). Peroxidase activity was then made

C. Fanutti et al. : a-Xylosidase from nasturtium visible on the blots using diaminobenzidine. The procedure used was that of Clegg (1982). Analytical isoelectric focusing was carried out using Ampholine PAGplates (LKB, Bromma, Sweden). They were run and stained for proteins with Coomassie Blue exactly according to the manufacturers' instructions. Thin-layer chromatography was carried out on aluminium foil-backed silica-gel layers 0.2 mm in thickness (Merck DC-Alufolien, Kieselgel 60). For oligosaccharide separation, triple development in the solvent propan-1-ol 5 :nitromethane 2 :water 3 (by vol.) was used. Monosaccharides were separated using propan-l-ol 7 :ethanol 1 :water 2 (by vol. ; double development). Carbohydrates were detected by spraying lightly with a 5% (v/v) solution of concentrated sulphuric acid in methanol and heating in an oven at 120 ~ C for 5 min. This procedure gave sensitive detection of oligosaccharides. Monosaccharides were detected more sensitively by spraying plates (or the appropriate areas of plates) with a solution (3% w/v) ofp-anisidine phthalate (Kirchner 1967) in methanol and heating at 120 ~ C for 5 min. Separations of monosaccharide alditol acetates by GLC were carried out using a Pye Series 104 chromatograph (Pye-Unicam, Cambridge, UK) with twin columns and flame-ionisation detectors. The columns (3% polyester E C N S S - M on Gas Chrom Q, dimensions 200 cm long, 0.4 cm i.d.) were operated isothermally at 180 to 190 ~ C, with a nitrogen carrier-gas flow of 50 ml 9rain-1 The high-performance liquid chromatography (HPLC) analysis of oligosaccharides was carried out using a Dionex BioLC chromatograph equipped with CarboPac PAl ion-exchange column, pulsed amperometric detector, and microinjector system. The column was eluted with a gradient of sodium acetate (50 mM to 100 mM) in sodium hydroxide (100 mM). Gel filtration was carried out using Bio-Gel P (BioRad, Watford, Herts, UK), hydrophobic interaction chromatography using phenylSepharose (Pharmacia, Uppsala, Sweden) and chromatofocusing using Polybuffer Exchanger PBE94 and Polybuffer 96 (Pharmacia).

Purification of the enzyme. During the purification procedure the "activity" of the c~-xylosidase in column fractions was estimated by assaying the pentose released after incubating an appropriate volume of the fraction for up to 16 h with tamarind xyloglucan (final concentration 10 mg-m1-1) and an aliquot of pure xyloglucanspecific endo-fl-o-glucanase (Edwards et al. 1986) (usually 10 lag enzyme protein.) Seed germination and seedling growth conditions were as before (Edwards et al. 1986). The cotyledons were removed manually from 13-d seedlings (typically about 800 g fresh weight, approx. 3000 cotyledon pairs) and a clarified homogenate was prepared exactly as described previously (Edwards et al. 1988). The homogenate was treated with ammonium sulphate, and the precipitated proteins were collected by centrifugation, dissolved in Tris-HC1 buffer (20 mM, pH 7.8), dialysed against the same buffer and subjected to anion-exchange chromatography on diethylaminoethyl (DEAE)-cellulose again exactly as before (Edwards et al. 1988). The ct-xylosidase activity was eluted from the column in the sodiumchloride gradient. Those fractions which contained ct-xylosidase activity were pooled and brought to 90% saturation with ammonium sulphate by addition of the crystalline salt. The precipitate was collected by centrifugation (26000 - 9 for 30 min), dissolved in the minimum volume of a buffer solution (pH 5.0) containing sodium acetate (5 mM), calcium chloride (1 raM), manganous chloride (1 raM) and sodium chloride (200 mM), and dialysed against the same solution. The dialysed solution was applied to a column of Con A-Sepharose (Sigma) (250 mm long, 10 mm i.d.) which had been pre-equilibrated with the same buffer. The column was eluted stepwise with methyl-a-D-glucopyranoside (80 ml each of 0, 10, 20 and 100 mM) in the same buffer. Fractions were analysed by S D S - P A G E and assayed for a-xylosidase. Those fractions from the 10- and 20-mM elution steps which contained a-xylosidase activity and only a single protein component (Mr 85000) on S D S - P A G E were pooled. The enzyme was concentrated by adsorption (after

139 ammonium-sulphate precipitation and dialysis) onto a second Con A-Sepharose column from which it was eluted in a narrow band by 50 mM methyl-a-D-glucopyranoside. In some preparations (see main text) an additional purification step was necessary to remove a minor contamination by a fl-galactosidase. The enzyme was dissolved in ammonium acetate (50 mM, pH 5.0) and applied to a column (100 mm long, 5 mm i.d.) of Whatman (Maidstone, Kent, UK) carboxymethylcellulose CM 52. The column was eluted with o-galactose (0.5 M) in the same buffer. The galactosidase was eluted from the column in the first galactose-containing fractions whereas the ct-xylosidase was eluted slowly from the column over several bed volumes. The xylosidase-containing fractions were dialysed free of galactose and concentrated by lectin-affinity chromatography as above.

Preparation of a limit digest of tamarind-seed xyloglucan usino the xyloolucan-specific endo-fl-olueanase (Edwards et al. 1986), and fraetionation of the dioest. Tamarind-seed xyloglucan (600 mg, final concentration 10 mg - ml-1) was treated with the endo-fl-glucanase (50 lag enzyme protein) at 30 ~ C until there was no further change in the reducing power of the system and no apparent change in the pattern of oligosaccharide products on TLC. The enzyme was heat-denatured (100 ~ C for 10 min) and any insoluble material was removed by centrifugation (26 000"9 for 10min). Ethanol was added to the supernatant to give a final concentration of 65% ethanol by volume. The white precipitate which formed immediately was collected by centrifugation (26 000 99; 20 min), washed with 65% ethanol twice with intermediate centrifugation, collected by centrifugation, dissolved in water and freeze-dried. The supernatant from the ethanol precipitation was taken to dryness by vacuum rotary evaporation at 45 ~ C, dissolved in 50 m M ammonium acetate pH 5.0 (25 ml) and fractionated on a Bio-Gel P4 column (100 cm long, 2.2 cm i.d.). The column was eluted with the same ammonium-acetate solution. Fractions (2.5 ml) were collected and analysed for total carbohydrate (glucose as standard). Individual carbohydrate-containing fractions were further analysed by TLC. The three saccharide components of lowest molecular size in the enzyme digest were partially separated one from the other on the Bio-Gel P4 column. Chromatographically (TLC) pure samples of each were obtained by preparative TLC. Plates (20 920 cm z) were strip-loaded each with about 10-15 mg carbohydrate. After development the plates were dried in a stream of warm air and marker strips cut from the edges and from the middle of each plate were treated to reveal the separated zones. The silica gel was removed from the appropriate areas of the remainder of the plate and suspended in water (10 ml per zone per plate) for 60 min at room temperature. To remove inorganic contaminants from the carbohydrate fractions the solutions were centrifuged (26000-9; 15 rain), taken to dryness by rotary evaporation at 45 ~ C, dissolved in ammonium acetate (50 mM, pH 5.0) and fractionated on Bio-Gel P4 columns (40 cm long, 2.2 cm i.d.). Carbohydrate-containing column fractions were pooled and freeze-dried. The amounts of the three saccharides were determined at first by weighing, and checked later by total-carbohydrate determination using as standards solutions containing mixtures of monosaccharides corresponding to their compositions. The purity of the three saccharide fractions (saccharides A, B and C, respectively) was checked by TLC in both solvents and by HPLC.

Substrate specificity of the ~-xylosidase. Individual fractions from the Bio-Gel P4 column were examined by TLC before and after incubation with the purified ct-xylosidase. The incubation mixtures contained enzyme (50 lag protein) and column fraction (concentrated if necessary) in a total volume of 100 lal. Incubations were terminated by heating to 100 ~ C for 5 min. Aliquots (up to 8 lal) of incubation mixtures and of appropriately concentrated column fractions (without enzyme digestion) were spotted directly onto TLC plates.

Isolation of the oligosaccharide products of digestion of saccharides A and C with the purified c~-xylosidase. Saccharides A and C (0.625

140 and 0.64 mM, respectively) were incubated with the enzyme (187 lag enzyme protein 9ml 1) at 30 ~ C in 50 m M ammonium-acetate buffer (pH 5.0). The progress of the reactions was monitored by estimation of the pentose released. When no further pentose was released the enzyme was heat-denatured (100 ~ C, 5 min). The digest was centrifuged (26 000 - g for 5 min) and the supernatant was applied to a column (40 cm long, 2.2 cm i.d.) of Bio-Gel P4. The column was eluted with 5 0 m M ammonium-acetate (pH 5.0) and fractions were analysed for total carbohydrate. Two clearly resolved peaks were obtained, one of which was shown by TLC analysis to contain only xylose. The fractions comprising the other peak (the first to emerge from the column) were pooled and freeze-dried.

Preparation of oligosaccharides and their corresponding alditols for analysis by nuclear-magnetic resonance (NMR). Alditols were prepared from saccharides A to C, and from the oligosaccharide products of their digestion with c~-xylosidase,by borohydride reduction. Saccharide (2 mg) was dissolved directly in 1 ml of a solution of sodium borohydride (0.5 M) in 1 M ammonia and left at room temperature for 2 h. The solution was acidified with glacial acetic acid to decompose excess borohydride and the saccharide was freed of salts and boric acid. This was done by pasing the solution through a column (4ml bed volume) of Amberlite IR 120 (H +) eluted with deionised water, concentrating pooled carbohydratecontaining fractions to dryness under vacuum, dissolving the residue in methanol and removing boric acid as the volatile methyl ester by vacuum rotary evaporation; the methanol-evaporation step was repeated five times and the final residue was dissolved in water and freeze dried. The 1H-NMR spectra of D/O solutions (0.5 to 5 mg in 0.5 ml) of saccharides or of their corresponding alditols were recorded at 85-90~ C on a model AM 200 spectrometer (Bruker Analytische Messtechnik, Rheinstetten, FRG) operating at 200.13 MHz. To ensure full signal responses for accurate integration, a 10-s recycle time was employed between successive 90~ pulses. Samples were lyophilised once or twice from D20 solution prior to analysis to reduce interference from residual protiated solvent. Chemical shifts are referenced to external tetramethylsilane. Catalytic properties of the purified c~-xylosidase. These were determined using the pure oligosaccharides A and C as substrates. Initial velocities were determined by measuring pentose levels in the incubation mixtures at intervals after mixing enzyme and substrates. No pentose was ever detected in enzyme- or substrate-free control incubations. The pH optimum was determined using oligosaccharide C and McIlvaine phosphate-citrate buffers (Dawson et al. 1982) spanning the range pH 2.6 to pH 7.6. Kinetic parameters ( K and Vax values were obtained by statistical analysis of initial velocity/substrate concentration data using the computer programme of Cleland (1979). Results

Purification and molecular properties of the enzyme. The enzyme was purified f r o m extracts prepared f r o m the cotyledons of nasturtium seedlings 13 d after sowing, when ~-xylosidase activity in cotyledonary extracts is at its m a x i m u m (Edwards et al. 1985). It had been observed earlier that crude extracts of nasturtium cotyledons did not catalyse the hydrolysis of p-nitrophenyl-e-D-xylopyranoside, yet did catalyse the release of xylose from tamarind xyloglucan (Edwards et al. 1985). Tamarind xyloglucan was therefore chosen as substrate to monitor e-xylosidase activity during purification. In our first purification experiments it became obvious that only crude seed extracts and those partially-purified preparations which still contained xyloglucan-specific endo-glucanase

C. Fanutti et al. : a-Xylosidase from nasturtium 2.0

1.6

13-

1.2

<

I

-1.0 0.8-

~;

E

0.8

0.6 0.4

0.4-

8 x

0.2 ! P 10

20 30 40 50 Fraction number 0 9 ~ 10 20 Methyl-o~- D- Glucopyranoside(mM)

Fig. I. Affinity chromatography on Con A-Sepharose of the nasturtium-seed ~-xylosidase after DEAE-cellulose chromatography (Fig. 2, lane 3).

activity would catalyse the release of pentose from xyloglucan. This indicated that the native xyloglucan was not a substrate, and that substrates were probably generated from it by endo-glucanase action. In all subsequent purification procedures c~-xylosidase levels were monitored semi-quantitatively by determining the pentose released on incubating column fractions with tamarind xyloglucan plus an aliquot of the pure, xyloglucan-specific endo(1--,4)-fl-D-glucanase from nasturtium (Edwards et al. 1986). A partial purification of the e-xylosidase was obtained by ion-exchange c h r o m a t o g r a p h y on DEAE-cellulose. Attempts at further purification by gel filtration, hydrophobic interaction c h r o m a t o g r a p h y and chromatofocusing had only limited success because the enzyme behaved as if it had a very high degree of molecular dispersity. This indicated that it might be a glycoprotein and therefore susceptible to purification by lectin-affinity chromatography. The activity bound quantitatively to a column of Con A-Sepharose and most of it was eluted as a symmetrical peak by 10 m M methyl-aD-glucopyranoside: the remainder could be eluted by higher concentrations of the glucoside (Fig. 1). A twostep purification schedule, comprising DEAE-cellulose c h r o m a t o g r a p h y followed by Con-A affinity chromatography with graded elution, gave an c~-xylosidase preparation which was homogeneous on S D S - P A G E (Fig. 2A). The purified enzyme corresponded to about 40% of the activity applied to the affinity column, the remainder being contaminated with other proteins. Occasionally this procedure gave an ~-xylosidase preparation which was contaminated by very small amounts of a protein band of higher electrophoretic mobility (Mr 63000); such preparations contained low levels of fl-galactosidase activity. The contaminant was

C. Fanutti et al. : a-Xylosidase from nasturtium 1

2

3

4

141

5

IEF

A

0.6

B

Fig. 2A, B. Separation by S D S - P A G E of the purified nasturtium enzyme and of partially purified seed extracts. A. Gel stained with Coomassie Blue. B Electroblot from an identical gel stained to reveal glycoproteins. Lanes 1 , 4 - purified enzyme. L a n e 2 - seed extract after ammoniumsulphate precipitation. L a n e 3 - enzyme after DEAE-cellulose chromatography. L a n e 5 - marker proteins: p-galactosidase (116000), phosphorylase B (97400), bovine albumin (66000), ovalbumin (45000, a glycoprotein), carbonic anhydrase (29000)

removed successfully by a (presumed) affinity elution procedure. The mixture was applied to a carboxymethylcellulose column and the contaminant was selectively eluted with D-galactose. The ~-xylosidase was recovered in 70% yield. The enzyme preparation described in this paper was purified using the simpler two-step procedure. It was not possible to construct a conventional purification table for the enzyme, since no suitable substrate was available to assay the activity of the enzyme in crude and partially purified extracts. The degree of purification achieved can be appreciated qualitatively by comparing the electrophoretograms of crude extract and purified enzyme (Fig. 2A). No endo-fl-glucanase, fl-galactosidase or ~-glucosidase activity (Edwards et al. 1985) could be detected in the purified preparation. The glycoprotein nature of the enzyme was confirmed by staining SDS gels directly using periodic acid and Schiff's reagent, and by staining electroblots of gels using the Con A-horseradish peroxidase procedure of Clegg (1982) (Fig. 2B). On isoelectric focusing, several protein bands were resolved, spanning the range of isoelectric points from pI = 5.0 to pI = 7.1. The ~-xylosidase activity was distributed over the same pI range, with a maximum at pI = 6.1 (Fig. 3).

Substrate requirement of the enzyme. As indicated by earlier work (Edwards et al. 1985) the purified enzyme did not catalyse the hydrolysis of p-nitrophenyl-~-Dxylopyranoside, and it did not release pentose from solutions of tamarind-seed xyloglucan. Pentose was released, however, from solutions of tamarind xyloglucan which had been pre-digested with the xyloglucan-specific endo-

0.5

~" E

"r

7

r

Q.

0,4 O o~ O

0.3 x

0.2

0.1

1

2

3

I

I

I

4

5

6

7

8

9

Gel length (cm)

Fig. 3. Isoelectric focusing ~-xylosidase

(IEF)

of the purified nasturtium-seed

fl-o-glucanase from nasturtium seeds. Analysis of these incubation mixtures by TLC confirmed that the only monosaccharide released by the enzyme was xylose. The products of hydrolysis of tamarind xyloglucan with crude, commercial cellulase preparations from Trichoderma viride and Penicillium funiculosum also contained a substrate or substrates for the enzyme. The simple xyloglucan disaccharide isoprimeverose (~-D-xylopyranosyl-(1-*6)-o-glucose) was however not hydrolysed even on prolonged incubation. To identify the e-xylosidase substrate(s) generated on digestion of tamarind xyloglucan with the xyloglucanspecific glucanase a solution of the polysaccharide was digested with the pure endo-glucanase until the reducing power of the incubation mixture was constant. Analysis of this mixture by TLC showed that it contained oligosaccharides varying widely in molecular size and no monosaccharides (Fig. 4). These were fractionated according to molecular size. A relatively high-molecularweight fraction was obtained as a precipitate from 65% ethanol. The supernatant was then further fractionated on a column of Bio-Gel P4. All the carbohydratecontaining fractions from this column were analysed by TLC before and after digestion with the purified e-xylosidase. After the ~-xylosidase digestion all the fractions contained xylose (Fig. 5). They therefore all contained substrates for the ~-xylosidase. In a separate experiment it was shown that the enzyme released xylose also from the higher-molecular-size oligosaccharide fraction which had been obtained by 65% ethanol precipitation of the

142

C. Fanutti et al.: a-Xylosidase from nasturtium

Fig. 4. Separation by TLC (after removal of a high-molecular-weight fraction by 65% ethanol precipitation) of the saccharides produced on complete digestion of tamarind xyloglucan by the pure xyloglucan-specific endo-( 1--~4)-fl-D-glucanase (Edwards et al. 1986) from nasturtium seeds. Lt~[? lane - enzyme digest. Ri,qht lane - reference standards. A, B, C - oligosaccharides later purified to homogeneity (see text)

s~3 s:2 s3 s~0 4~ 4:8 4:7 ~

~

PPM

Fig. 6. The anomeric region of the ~H-NMR spectra of native

tamarind xyloglucan (A), tamarind xyloglucan after complete digestion with the pure endo-glucanase from nasturtium seeds (B), oligosaccharide C (C), oligosaccharide A (D)

Fig. 5. Separation by TLC, after ~-xylosidase treatment, of individual fractions from a BiD-Gel P4 column separation of the oligosaccharide mixture (after 65 % ethanol precipitation of a highmolecular-weight fraction) produced on complete digestion of tamarind xyloglucan with the pure endo-glucanase from nasturtium seeds (see Fig. 4). Lanes 1 to 9 - column fractions individually digested with the purified ~t-xylosidase.Lane 10 - reference standards of o-xylose (uppermost spot), D-glucoseand o-galactose. The lower part of the plate was sprayed with H2SO4 in ethanol, and the upper part with p-anisidine phthalate for the sensitive detection of penroses endo-glucanase incubation mixture. Surprisingly, the enzyme-catalysed release of xylose from the endo-glucanase-generated oligosaccharides did not cause any observable change in their T L C mobilities. This indicated that the structural changes brought about by the action of the enzyme on any individual saccharide might be limited. M o d e o f action o f the enzyme on xylogluean oligosaccharides. The xyloglucan oligosaccharides generated by the

action of the xyloglucan-specific endo-fl-glucanase on tamarind xyloglucan were substrates for the e-xylosidase whereas the polymeric xyloglucan was not. This indicated the presence in the oligosaccharides of a molecular environment for some ~-o-xylosyl residues which was not present in the original polymer. This was investigated by comparing the 1 H - N M R spectra of polymeric and endo-glucanase-digested tamarind xyloglucan. The anomeric proton regions of these spectra are compared in Fig. 6. The spectra (Fig. 6, traces A, B) showed good resolution of the anomeric proton signals arising from unsubstituted side-chain xylose (4.96 ppm), galactosesubstituted sidechain xylose (5.14 ppm) and reducing glucose (Fig. 6, trace B; 4.68 and 5.24 ppm). The signals arising from galactose and non-reducing glucose were incompletely resolved (4.52 to 4.62 ppm). Nevertheless sufficient information was present to allow the calculation (Gidley et al. 1991) of the ratio galactose: xylose: glucose in the polysaccharide by integration of the signals. (Since all galactose in xyloglucans is linked to xylose, galactose-substituted xylose must equal galactose, allowing the calculation of glucose from the total integral of galactose+glucose.) The calculated ratio (Gal:Xyl:Glc = 17:37:46) was in complete agreement with the value obtained experimentally by hydrolysis of the polysaccharide and quantitative analysis of the sugars released. By comparing spectra A and B in Fig. 6 an increased proportion of reducing glucose residues generated by endo-glucanase digestion is observed. In addition the spectrum of the enzyme-depolymerised xyloglucan (Fig. 6, trace B) showed a splitting (arrow in Fig. 6, trace B) of the signal arising from unsubstituted,

C. Fanutti et al.: a-Xylosidase from nasturtium

Gal

A B

C

1

2

3

4

5

Fig. 7. Analysis by TLC of a nasturtium-seed fi-galactosidase (Edwards et aL 1988) digest of oligosaccharide C. Lane 1- oligosaccharide C. Lane 2- digest immediately after adding the enzyme. Lane 3- digest after 13 min. Lane 4- digest after 37 min. Lane 5- digest after 2 h. Lane 6- digest after 4.5 h. A, B, C, Gal = mobilities on the same chromatogram of reference standards of oligosaccharides A, B, C and D-galactose

6

side-chain xylose which was not present in that of the polymer (Fig. 6, trace A). This was clear confirmation that endo-glucanase cleavage had created an environment for some of the side-chain xylose residues which did not exist in the intact polymer. Since the action of the endo-glucanase is to create new chain-ends it seemed logical that the "new" environment for unsubstituted side-chain xylose would be at a newly created glucan chain terminal (reducing or non-reducing). This interpretation was confirmed by the N M R analysis o f a series of fractions obtained by partial separation of an endoglucanase digestion on Sephadex G-50. The fractions all showed the same splitting of the signal arising from non-substituted side-chain xylose, with the relative intensity of the newly created signal increasing with decreasing molecular size. In the following text those unsubstituted side-chain xylose residues giving chemical shift values corresponding to those in the polymeric xyloglucan are called "internal", whereas those giving rise to the "new" signal associated with the glucanase-digested xyloglucan are called "terminal". To investigate further the environments of "terminal" and "internal" side-chain unsubstituted xylose, the three xyloglucan oligosaccharides with the highest T L C mobilities were purified from the mixture o f saccharides generated on endo-glucanase digestion o f tamarind xyloglucan and subjected to enzymatic analysis and 1 H - N M R . The oligosaccharides, labelled A, B and C in Fig. 4, were purified by gel filtration on Bio-Gel P4 followed by preparative TLC. Saccharides A and C gave single spots on TLC and single peaks on H P L C analysis. Saccharide B gave a single spot on T L C but was clearly resolved into two unequal peaks by HPLC. When oligosaccharide A was treated with the pure fl-D-galactosidase from nasturtium seeds (Edwards et al. 1988), no galactose was released and there was no change in its T L C

143 mobility. When oligosaccharides B and C were treated in the same way, galactose was released and they were each converted quantitatively into a compound with the same T L C mobility as oligosaccharide A. In the case of oligosaccharide C the transitory appearance of a compound with the same T L C mobility as oligosaccharide B was observed in the early stages o f the enzyme digestion (Fig. 7). These data indicated that oligosaccharides A, B and C had a c o m m o n xyloglucan core substituted by zero, one and two terminal non-reducing fl-galactosyl residues, respectively. Well-resolved N M R spectra (e.g. Fig. 6, traces C, D) were obtained for all three saccharides, and were integrated to give separate values for glucose (reducing and non-reducing) residues, galactose residues, galactose-substituted xylose residues and unsubstituted xylose residues ("terminal" an d "internal"). The results are summarised in Table 1. Saccharides A, B and C were clearly xyloglucan oligosaccharides with the compositions Glc4Xyl3 (a heptasaccharide), Glc4Xyl3Gal (an octasaccharide) and Glc4Xy13Ga12 (a nonasaccharide) respectively. Saccharides A, B and C each contained only a single unsubstituted xylose residue of the "terminal" type and 2, 1 and 0 respectively of the "internal" type. To determine whether the "terminal" xylose residue in xyloglucan 01igosaccharides A, B and C was associated with the reducing end or with the non-reducing end of the molecule, samples of A, B and C were reduced with sodium borohydride to convert the reducing residue to an alditol, re-purified and re-examined by N M R . The spectra were well resolved. They contained no signals arising from reducing glucose residues, confirming complete reduction. On comparing the spectra of A, B and C before and after reduction, no change in the signal associated with "terminal" unsubstituted xylose residues was observed (e.g. Fig. 8, traces A, B), indicating very strongly that it was remote from the reducing end of the

Table 1. Analysis by 1H-NMR of anomeric protons in saccharides A, B and C Signal"

Assignment

Relative integral A B C

5.143 (3.5) 4.961 (3.5)

1,2-1inkedct-xylose "Internal" 1-1inked ~-xylose "Terminal" 1-1inked ~-xylose Reducing 4-1inked ") a-glucose j~ Reducing 4-1inked [3-glucose 1,4 linked -~ [3-glucose j~ 1-1inked 13-galactose

0 2

1 1

2 0

1

1

1

1

1

1

3

4

5

4.944 (3.6) 5.245 (3.8) 4.676 (7.8) 4.54-4.60 (7.7-7.9) 4.53~.60 b

a Chemical shifts in ppm downfield from standard, with H-l, H-2 coupling constants (Hz) in parentheses b Unresolved coupling

144

C. Fanutti et al. : a-Xylosidasefrom nasturtium

molecule. In contrast, significant chemical-shift changes were observed for non-reducing glucose residues in saccharide A following borohydride reduction (Fig. 8, traces A, B). In particular one signal (Fig. 8, trace B) is shifted at least 0.07 ppm downfield (to 4.655 ppm), indicating that this signal is probably due to H-1 of glucose attached to the reducing-reduced terminal glucose. If xylose had been attached to reducing terminal glucose, a similar shift would have been expected in comparison with the undetectable (< 0.002 ppm) shift in the signal for "terminal" xylose following borohydride reduction. Small changes were observed also in the signals associated with galactose-substituted xylose in oligosaccharides B and C, indicating that this structural feature was closer to the reducing end of the molecule than the "terminal" unsubstituted xylose. Chemical shifts and integrals for reduced and non-reduced saccharide A are given in Table 2. The highest-field glucose signal (4.56 ppm) was assigned to the non-reducing terminal residue by comparison of relative shifts with those for cellotetraose and by the fact that this signal was the least-shifted glucose signal following borohydride reduction. The chemical-shift values shown in Table 2 for saccharide A (Glc4Xyl3) are esssentially identical to those for structure I as determined by York et al. (1988). Oligosaccharides A, B and C were therefore assigned the structure I, II +III, and IV, respectively. Xyll

Xyll

~a

~a

Xyll

~a

6,8 6,8 6,8 G l c l -~4G 1c 1 ---}4G1c 1--}4G 1c

Structure I (oligosaccharide A)

Gall

Gall 2 Xyll J,a

Xyll ~a

6

,8

PPM

Fig. 8. The anomeric region of the 1H-NMR spectra of oligosaccharide A (A), oligosaccharideA after borohydridereduction (B), the borohydride-reducedoligosaccharideproduct of ~-xylosidase digestion of oligosaccharideA (C)

Xyll

,La

6

,8 6 , 8 G l c l -*4G lcl ---,4G1cl ~ 4 G l c Structure H

Xyll

J,a

Xyll

J,,~

2 Xyll

J,a

6,8 6,8 6,8 G l c l - , 4 G l c l -,4G 1cl ~ 4 G l c Structure 111

(oligosaccharide B is a mixture of structures II and III) Gall

~P

Gall

~B

2 2 Xyll Xyll Xyll +a +a Sa 6,8 6,8 6,8 Glcl ~4Glcl ~4Glcl ~4Glc

Structure I V (oligosaccharide C)

The two pure oligosaccharides A and C were separately incubated with the pure ~-xylosidase. The progress of the reaction was followed by quantitative analysis of the pentose liberated and by HPLC of the reaction mixture. It was clear from the pentose analyses that a single xylose residue was removed by the action of the enzyme in each case. The HPLC analyses showed that the enzymatic removal of xylose was accompanied in each case by the appearance of a single oligosaccharide

product of lower HPLC mobility than the oligosaccharide substrate. The data obtained for oligosaccharide A are shown in Fig. 9. In the case of nonasaccharide C (structure IV) only a single, unsubstituted xylose residue was available for enzymatic cleavage. This residue was of the "terminal" type. By contrast, heptasaccharide A (structure I) had three unsubstituted side-chain xylose

C. Fanutti et al. : a-Xylosidase from nasturtium

145

Table 2. Anomeric signals" in oligoxyloglucan 1H-NMR spectra Assignment

Saccharide A

Saccharide A reduced

Saccharide A + c~-xylb

Saccharide A + c~-xylb reduced

Cellotetraose

"Terminal" 1-1inked ct-xylose "Internal" l-linked e-xylose Reducing 4-1inked c~-glucose Reducing 4-1inked 13-glucose Non-reducing l-linked [3-glucose 1,4 linked [3-glucose

4.945 (1 H)

4.945 (1 H)

absent

absent

absent

4.961 (2 H)

4.961 (2 H)

4.961 (2 H)

4.961 (2 H)

absent

5.245 I (1 H) 4.677

absent

absent

absent

5.246 I (1 H) I 4.677

absent

5.246 I (1 H) 4.676

4.561 (1 H)

4.559 (1 H)

4.537 (1 H)

4.537 (1 H)

4.536 (1 H)

4.586 (2 H)

4.598 (1 H) 4.655 (1 H)

4.580 (2 H)

4.594 (1 H) 4.655 (1 H)

4.564 (2 H)

Chemical shifts downfield from standard in ppm, relative integrals shown in parentheses. H - l , H-2 coupling constants for a and 13residues were all in the ranges 3.4-3.8 Hz and 7.6-8.0 Hz, respectively b Oligosaccharide product of digestion of oligosaccharide A with the purified ct-xylosidase "

1

2

a

3

X

X

spectra both before and after (Fig. 8C) borohydride reduction, which were integrated as before (Table 2). It is clear from Table 2 that the oligosaccharide product of digestion of heptasaccharide A (structure I) with the e-xylosidase was a hexasaccharide (Glc4Xyl 2). On comparing Table 2 with Table 1 and Fig. 8, traces B and C, it is clear that the single xylose residue cleaved from heptasaccharide A by the c~-xylosidase was the one in the "terminal" environment, i.e. attached to the backbone glucose residue remote from the reducing end of the molecule. The hexasaccharide product may therefore be assigned the structure V.

" 0.6-

Xyl 1

Xyl 1

~a

0.4-

fl

6

~a

fl

6

fl

Glcl~4Glcl--*4Glcl~4Glc

Structure V (hexasaccharide)

O

%, 0,2, X

i

i

i

i

1

2

3

4

Time (h)

Fig. 9. c~-Xylosidase digestion of otigosaccharide A (0.625 mM). L o w e r p a r t - xylose release. U p p e r p a r t - HPLC analysis of the digest after 30 s (1), 30 min (2), 120 min (3), a = oligosaccharide A; x = xylose

residues two of which were "internal" and one "terminal". A further sample of heptasaccharide A was digested to completion using the ct-xylosidase and the oligosaccharide product was purified free of protein, xylose and buffer salts. This saccharide gave well-resolved N M R

Confirmatory evidence for this structure is obtained by comparing 1H chemical shifts with those for cellotetraose (Table 2). Essentially identical (+ 0.001 ppm) chemical shifts are observed both for reducing and non-reducing terminal glucoses, with internal glucoses showing significant differences (0.016 ppm).

Some catalytic properties of the ~-D-xylosidase. With nonasaccharide C (structure IV) as substrate the enzyme showed a pH optimum at pH 5.0 (half-maximal activity at pH 4 and pH 6.5). The enzyme was inactive below pH 3 and above pH 7.5. No cofactor requirements were demonstrated. The Km and Vmax values for heptasaccharide A and nonasaccharide C are given in Table 3. The enzyme clearly does not differentiate greatly between the two oligosaccharides.

146

C. Fanutti et al.: a-Xylosidase from nasturtium

Table 3. Substrate specificity of the c~-xylosidase from nasturtium

seeds Substrate

Km (mM)

Vmax (nkat 9mg protein- 1)

Oligosaccharide A (structure I) Oligosaccharide C (structure III)

0.62

14

0.32

7.6

Isoprimeverose

Not hydrolysed

[~-D-xyl (1 ~ 6 ) D-glc] p-Nitrophenyl-c~-Dxylopyranoside

Not hydrolysed

Discussion

It may be deduced from the evidence presented here that the nasturtium-seed c~-xylosidase acts only on xyloglucan oligosaccharides and catalyses the selective hydrolytic cleavage of an unsubstituted side-chain xylose residue attached to the backbone glucose residue situated farthest from the reducing end of the molecule, i.e. "terminal" xylose. Our data also allow the deduction that the xyloglucan-specific endo-(1---,4)-fl-D-glucanase from nasturtium seeds (Edwards et al. 1986) acts on tamarindseed xyloglucan to produce only oligosaccharides which are e-xylosidase substrates, i.e. xyloglucan oligosaccharides which have a "terminal" o-xylose residue. The structures of these oligosaccharides will be discussed further in a future publication concerning the molecular mode of action of the endo-glucanase. Only three e-xylosidases have previously been purified to demonstrable homogeneity. Two of them, from Aspergillus niger (Matsushita et al. 1985) and Bacillus species No 693-1 (Zong and Yasui 1989; Zong et al. 1989) respectively, could be termed "typical" e-xylosidases in that they hydrolyse simple cr such as p-nitrophenyl-c~-D-xylopyranoside and isoprimeverose (e-D-Xyl (1--, 6)Glc). Both release "terminal" xylose from simple xyloglucan-derived oligosaccharides, and neither hydrolyses polymeric xyloglucan. The Bacillus and As~ pergillus enzymes differ markedly in their pH optima (7.5 and 2.2-3.0, respectively), their pI values (4.25 and 5.6, respectively) and their molecular weights (subunit Mr = 82 000 and 123 000 respectively). The c~-xylosidases from Aspergillus and Bacillus are not obviously related to each other, or to the nasturtium c~-xylosidase in their molecular and catalytic properties. A third c~-xylosidase has recently been purified in small amounts from auxin-treated pea epicotyls using an elaborate purification scheme (O'Neill et al. 1989). It did not hydrolyse p-nitrophenyl-c~-D-xylopyranoside, isoprimeverose or polymeric xyloglucan, yet specifically cleaved the "terminal" xylosyl residue from xyloglucan oligosaccharides obtained by digestion of sycamore extracellular xyloglucan with Trichoderma viride endo(1----,4)-fl-D-glucanase. The pea enzyme had an M r of 85 000 and comprised several molecular species spanning the range pI = 7.35 to 7.7; its pH optimum was at pH

4.9-5.1. The e-xylosidase from germinated nasturtium seeds clearly has the same highly restricted specificity as the pea enzyme: neither plant enzyme possesses simple c~-xylosidase activity, and their action might be more appropriately termed "exo-oligoxyloglucan-e-xylohydrolase". The nasturtium and pea enzymes have the same Mr and both comprise a range of molecular species differing in pI. The enzyme isolated from pea covers the pI range 7.35 to 7.7 whereas that from nasturtium covers the pI range of 5.0 to 7.1. It should be noted, however, that only a part of the c~-xylosidase activity eluted from our affinity chromatography column was obtained free of contaminating proteins, and it is possible that our preparation does not define the full native pI range. A similar argument could be applied to the purification of the pea enzyme, which also involved the rejection of an appreciable amount of e-xylosidase activity to obtain an electrophoretically homogeneous preparation (O'Neill et al. 1989). Our enzyme is a glycoprotein and the microheterogeneity observed on isoelectric focusing may reflect differences in the nature and-or number of glycosyl substituents. If the pea enzyme is also a glycoprotein the e-xylosidases from nasturtium and pea could be closely similar molecules. In the nasturtium seed the cell-wall storage xyloglucan is completely depolymerised after germination by the concerted action of a xyloglucan-specific endo-(14 4)fl-D-glucanase (Edwards et al. 1986), a fl-D-galactosidase (Edwards et al. 1988), fl-D-glucosidase (Edwards et al. 1985) and this ~-D-xylosidase. Crude enzyme extracts from nasturtium seeds, which contain all of these activities, catalyse the complete conversion of tamarind xyloglucan to glucose, galactose and xylose (data not shown), indicating that the end-products of xyloglucan degradation in vivo are the monosaccharides. It is clear from the data presented here that the action of the endoglucanase alone on seed xyloglucans is to produce a mixture of xyloglucan oligosaccharides with backbone glucan chain lengths of four residues or more. These saccharides all carry a "terminal" ~-xylosyl substituent and would thus not be susceptible to further degradation by a conventional fl-glucosidase, although their terminal galactosyl residues could be removed by the action of the fl-galactosidase. We envisage that the role of this ~-xylosidase in xyloglucan mobilisation in vivo is the "deblocking" of endo-glucanase-generated oligosaccharides (before or after the fl-galactosidase-catalysed removal of their terminal galactose residues) to allow the release of a glucose residue by fl-glucosidase action. Oligosaccharide A (structure I) would in this way be converted quantitatively to glucose and xylose by three rounds each of hydrolysis by the c~-xylosidase and a fl-glucosidase. This is assuming that the c~-xylosidase will remove terminal xylose from xyloglucan oligosaccharides with backbone chain-lengths of two units or more. (Oligosaccharides with backbone chain-lengths of two and three units were not available to us in the present study.) We have not observed any accumulation of xyloglucan oligosaccharides in nasturtium cotyledons during xyloglucan breakdown, and conclude that the activity of the c~-xylosidase is not rate-limiting in vivo.

C. Fanutti et al. : a-Xylosidase from nasturtium It has been argued that the specialised role o f cellwall polysaccharides as substrate reserves in seeds has arisen as a result o f the evolutionary a d a p t a t i o n o f m o r e widespread t u r n o v e r processes (Reid 1985). X y l o g l u c a n t u r n o v e r in the p r i m a r y cell walls o f certain dicoty ledonous species has been observed and has been associated with the control o f elongation growth, with the action o f auxin, and with cellulase (endo-(l~4)-fl-Dglucanase) activity (Hayashi 1989). F u r t h e r m o r e , a cellulase-generated fucosylated xyloglucan oligosaccharide with a b a c k b o n e chain length o f four glucose units is k n o w n to inhibit auxin-induced elongation g r o w t h ( Y o r k et al. 1984). This oligosaccharide, like all other cellulase-generated xyloglucan oligosaccharides reported up to now, has a xylosyl residue o f the terminal type. I f xyloglucan is turned over in p r i m a r y cell walls by a c o m p l e m e n t o f enzymes similar to those in the storage cell walls o f the germinated nasturtium seed, then a factor in the regulation o f the c o n c e n t r a t i o n in vivo o f this a n d other xyloglucan oligosaccharides might be the rate o f their exposure to further d e g r a d a t i o n by the ~-xylosidase-catalysed " d e b l o c k i n g " reaction. Much of the work reported in this paper was carried out with the aid of the European Community's "Science Stimulation Action" (Contract No. ST2P4)250-UK), and we wish to record our appreciation of this support.

References Bergmeyer, H.U., Berndt, E., Schmidt, F., Stark, H. (1974) D-Glucose determination with hexokinase and glucose-6-phosphate dehydrogenase. In: Methods of enzymatic analysis, pp. 1196-1201, Bergmeyer, U., ed. Verlag Chemie, Weinheim Clegg, J.C.S. (1982) Gtycoprotein detection in nitrocellulose transfers of electrophoretically separated protein mixtures using Concanavalin A peroxidase: application to Arenavirus and Flavivirus proteins. Anal. Biochem. 127, 389-394 Cleland, W.W. (1979) Statistical analysis of enzyme kinetic data. Methods Enzymol. 63, 103-138 Dawson, R.M.C., Elliott, D.C., Elliott, W.H., Jones, K.M. (1982) Data for biochemical research. Clarendon Press, Oxford Dische, Z. (1962) Color reactions of carbohydrates. In: Methods in carbohydrate chemistry, vol. 1, pp. 475-514, Whistler, R.L., Wolfrom, M.L., eds. Academic Press, New York Dubois, M., Gilles, K.A., Hamilton J.K., Rebers, P.A., Smith F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356 Edwards, M., Dea, I.C.M., Bulpin, P.V., Reid, J.S.G., (1985) Xyloglucan (amyloid) mobilisation in the cotyledons of Tropaeolum majus L. seeds following germination. Planta 163, 133-140 Edwards, M., Dea, I.C.M., Bulpin, P.V., Reid, J.S.G. (1986) Purification and properties of a novel xyloglucan-specific endo(1--~4)-fl-o-glucanase from germinated nasturtium seeds (Tropaeolum majus L.). J. Biol. Chem. 261, 9489-9494 Edwards, M., Bowman, Y.J.L., Dea, I.C.M., Reid, J.S.G. (1988) A fl-D-galactosidase from nasturtium (Tropaeolum majus L.) coty-

147 ledons. Purification, properties and demonstration that xyloglucan is the natural substrate. J. Biol. Chem. 263, 43334337 Fairbanks, G., Steck, T.L., Wallach, D.F.H. (1971) Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10, 2606 2617 Fry, S.C. (1987) Formation of isodityrosine by peroxidase isozymes. J. Exp. Bot. 38, 853-862 Gidley, M.J., Lillford, P.J., Rowlands, D.W., Lang, P., Dentini, M., Crescenzi, V., Edwards, M., Fanutti, C., Reid, J.S.G. (1991) Structure and solution properties of tamarind seed polysaccharide. Carbohydr. Res., in press Halliwell, G., Riaz, M. (1970) The formation of short fibres from native cellulose by components of Trichoderma koningii cellulase. Biochem. J. 116, 3542 Hayashi, T. (1989) Xyloglucans in the primary cell wall. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40, 139-168 Kirchner, J.G. (1967) Thin layer chromatography. In: Technique of organic chemistry, vol. 12, pp. 147-185, Perry, E.S., Weissberger, A., eds. Interscience, New York Matsushita, J., Kato, Y., Matsuda, K. (1985) Purification and properties of an Ct-D-xylosidase from Aspergillus niger. J. BiDchem. 98, 825-832 O'Neill, R.A., Albersheim, P., Darvill, A.G. (1989) Purification and characterization of a xyloglucan oligosaccharide-specific xylosidase from pea seedlings. J. Biol. Chem. 264, 20430-20437 Reid, J.S.G. (1985) Cell wall storage carbohydrates in seeds. Biochemistry of the seed "gums" and "hemicelluloses". Adv. Bot. Res. 11, 125-155 Reid, J.S.G., Edwards, M., Dea, I.C.M. (1988) Enzymatic modification of natural seed gums. In: Gums and stabilisers for the food industry, vol. 4, pp 391-398, Phillips, G.O., Wedlock, D.J., Williams, P.A., eds. IRL Press, Oxford Reis, D., Vian, B., Darzens, D., Roland, J.-C. (1987) Sequential patterns of intramural digestion of galactoxyloglucan in tamarind seedlings. Planta 170, 60-73 Roe, J.H., Rice, E.W. (1948) A photometric method for the determination of free pentoses in animal tissue. J. Biol. Chem. 173, 507-512 Saeman J.F., Buhl, J.L., Harris, E.E. (1945) Quantitative saccharifi"cation of wood and cellulose. Ind. Eng. Chem. Anal. Ed. 17, 35-37 Sedmak, J.J., Grossberg, S.E. (1977) A rapid, sensitive and versatile assay for protein using Coomassie Brilliant Blue G250. Anal. Biochem. 79, 544-552 Towbin, H., Staehlin, T., Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 9350-9359 Vian, B., McGowan, J., Reid, J.S.G. (1991) Enzyme-gold cytochemistry of seed xyloglucans using two xyloglucan-specific hydrolases. Importance of prior heat-deactivation of the enzymes. Histochem. J., in press York W.S., Darvill, A.G., Albersheim, P. 0984) Inhibition of 2,4-dichlorophenoxyacetic acid-stimulated elongation of pea stem segments by a xyloglucan oligosaccharide. Plant Physiol. 75, 295-297 York, W.S., Oates J.E., van Halbeek, H., Darvill, A.G., Albersheim, P. (1988) Location of the O-acetyl substituents on a nonasaccharide repeating unit of sycamore extracellular xyloglucan. Carbohydr. Res. 173, 113-132 Zong, N., Yasui, T. 0989) Purification and some properties of an a-o-xylosidase from Bacillus sp. No. 693-1. Agric. Biol. Chem. 53, 187-195 Zong, N., Kamiyama, Y., Yasui, T. (1989) Substrate specificity of Bacillus a-D-xylosidase. Agric. Biol. Chem. 53, 2129-2139