Volume 19, Part 3, June 1994
`
`ISSN 0885-4513
`
`Biotechnology and ..
`
`
`Applied Biochemistry;uN
`2 o 1994
`
`. • 1 7'.-1 S:'..;l�?-:CES LID:lC.rlV
`
`
`Lrnversity of Wicconsln
`
`Editor in Chief
`
`13C5 linden Drive
`
`Maaison, WI 53708
`Assistant Editor
`
`
`P. N. Campbell
`P.M. Brickell
`
`
`Department of Biochemistry and Molecular
`Medical Molecular Biology Unit, Department
`
`
`
`
`
`Biology, University College, Gower Street,
`
`
`
`of Molecular Pathology, UCL Medical School,
`London WCl E 6BT, U.K.
`London Wt P 6DB, U.K.
`
`
`
`Regional Editors
`
`J. R. DeLoach
`K.Soda
`USDA, Agriculture Research Service, Route, 5,
`
`
`
`
`Laboratory of Microbial Biochemistry, Institute
`
`
`
`
`
`Box 810, College Station, TX 77845, U.S.A.
`
`
`for Chemical Research, Kyoto University,
`
`
`Gokasho, Uji, Kyoto-fu 611, Japan
`
`Editors
`
`N. A. Rao
`
`National Institute for Medical Research, Mill
`
`
`
`A. Aitken
`
`
`
`
`Hill, London, U.K.
`
`
`
`Indian Institute of Science, Bangalore, India
`
`C.Bucke
`
`
`
`
`
`University of Westminster, London, U.K.
`
`E. S. Severin
`Research Center or Molecular Diagnostics,
`
`
`
`
`Ministry of Health, Moscow, Russia
`
`A. De Flora
`B. A. Stone
`Universita Degli Studi de Genova, Genova, Italy
`
`
`
`
`
`
`
`La Trobe University, Bundoora, Australia 3083
`
`G.W.Gould
`B.Svensson
`
`Unilever Limited, Shambrook, Bedford, U.K.
`
`
`Carlsberg Laboratory, Valby, Denmark
`
`M.-R. Kula
`Y. H. Tan
`lnstitut fiir Enzymtechnologie der Universitat
`
`
`
`
`
`
`Diisseldorf, Jiilich, Germany
`
`
`
`National University or Singapore, Singapore
`
`B. Mattiasson
`
`University of Lund, Lund, Sweden
`
`M. Uhlen
`
`
`Sweden
`
`Royal Institute of Technology, Stockholm,
`
`J.D. Newman
`S.D. Varfolomeyev
`
`
`
`Cranfield Institute of Technology, Cranfield,
`Bedford, U.K.
`
`
`Russia
`
`M.V. Lomonosov Moscow University, Moscow,
`
`
`
`The Medical School, NewcasUe upon Tyne,
`
`D.O'Shannessy
`R.Virden
`
`
`
`SmithKline Beecham, King of Prussia, U.S.A.
`
`
`T.Oshima
`U.K.
`
`Tokyo Institute of Technology, Yokohama, Japan
`
`
`J. R. Wild
`W. Ostrowski
`
`lnstytut Biochemii Lekarskiej, Krakow, Poland
`
`
`
`
`U.S.A.
`
`
`
`
`
`Texas A&M University, College Station, TX,
`
`
`
`
`
`• J. Lederberg • J. Porath Honorary Editors E. Katchalski-Katzir
`
`
`
`Staff Editor Stuart Hobday
`
`
`
`Copyright© 1994 by Portland Press Ltd
`
`

`

`• Biotechnology and Applied Biochemistry (ISSN 0885-45131 is published bimonthly in February, April, June,
`
`
`
`
`
`
`
`
`August, October and December by Portland Press Ltd, 59 Portland Place, London WlN 3AJ, U.K., on behalf of
`
`
`
`the International Union of Biochemistry and Molecular Biology.
`
`•Subscription orders should be sent to Portland Press Ltd, P.O. Box 32, Commerce Way, Colchester CO28HP,
`
`
`
`
`U.K., or to your usual agent. Volumes 19 and 20 (three issues eachl will appear in 1994. The institutional sub­
`
`
`
`
`scription rate for 1994 is £100, or US$175 for subscribers in North America. Subscription rates lor North
`
`
`
`
`
`Air-mail delivery is Surface Post delivery respectively. America and Japan include airfreight and Accelerated
`
`
`
`
`
`
`
`available at additional cost, details are available on application. Special personal subscription rates for
`
`
`members of the Biochemical Society are also available.
`
`• All rights reserved. Apart from any lair dealing for the purposes of research or private study, or criticism or
`
`
`
`
`
`
`
`
`
`
`review, as permitted under the Copyright, Designs and Patents Act, 1988, this publication may be reproduced,
`
`
`stored or transmitted, in any forms or by any means, only with the prior permission in writing of the pub­
`
`
`
`
`
`lishers, or in the case of reprographk reproduction in accordance with the terms of permissions granted by
`
`
`
`
`
`
`copyright clearing houses such as the Copyright Licensing Agency (U.K.) or the Copyright Clearance Center
`
`
`
`
`
`(U.S.A.). The CCC item-fee code for this publication is 0885-4513/94/$5.00 + 0.00. Inquiries concerning repro­
`
`
`duction outside those terms should be sent to the publishers at the London address.
`
`• Although, at the time of going to press, the information contained in this publication is believed to be correct,
`
`
`
`
`
`
`
`
`
`
`
`the authors, the editors, the publisher and the International Union of Biochemistry and Molecular Biology
`
`
`
`
`assume no responsibility for any errors or omissions herein contained. All material published herein represents
`
`
`
`
`
`the opinions of the author(sl and not necessarily those of the editors, the publishers or the International Union
`
`
`of Biochemistry and Molecular Biology.
`
`
`
`• Typeset by Unicus Graphics Ltd., Horsham, and Printed in Great Britain by Whits table Lit ho Printers Ltd., Kent.
`
`
`
`
`
`
`
`
`
`•Second-class postage paid at Rahway, NJ, U.S.A.
`
`and Applied Biochemistry, • U.S. Postmaster: send address changes to Biotechnology c/o Mercury Airfreight
`
`
`
`
`
`
`lnternational Ltd, 2323 Randolph Avenue, Avenel, NJ 07001.
`
`(ii)
`
`

`

`
`Biotechnol.
`
`Appl. Biochem. 19, 341-354 [1994]
`
`341
`(Printed in Great Britain)
`
`
`
`and properties of a /J-o-galactosidase
`Purification
`
`
`from Bifidobacterium bifidum exhibiting a
`1
`
`
`transgalactosylation reaction
`
`
`
`Vincent Dumortier, * Colette Brassart,t Stephanie
`
`
`
`
`Bouquelett
`'Departement Agro-Alimentaire et Biotechnologies, lnstitut Superieur d'Agriculture de Lille,
`
`
`
`
`
`
`Lille, France, and tLaboratoire de Chimie Biologique, Unite mixte du CNRS n° 111
`
`
`
`
`
`
`
`
`Universite des Sciences et Technologies de Lille, 59655 Villeneuve d'Ascq Cedex, France
`
`2
`
`P-o-Galactosidase was extracted by ultrasonic treatment in the presence of
`
`
`
`
`
`
`
`
`Triton X-100. The transgalactosylase fTGase) was separated from other P-o­
`
`
`
`galactostdases and purified 1 720-fold, as measured by the TGase assay. The
`
`
`enzyme was shown to be homogeneous on native PAGE. The complex nature of
`
`
`
`the enzyme was demonstrated by SDS/P AGE, with four major protein bands of
`
`
`163, 170, 178 and 190 kDa. The molecular mass was estimated to be 362 kDa by
`
`
`gel chromatography, and the pl was 5.25. The purified enzyme was stable at
`
`
`
`temperatures below 45 °C and over the pH range from 6.5-8. Lactose hydrolysis
`
`
`by the purified enzyme and by Bifidobacterlum blfidum cells took place at the
`
`
`same pH and temperature, at pH 6.5 and 37 °C respectively. For the TGase,
`
`
`
`
`
`however, B. bifidum cells exhibited specific galactosyltransferase activity at pH
`
`
`4.5 and at 40 °C, even though the transgalactosylation fTG) reaction was prefer­
`
`
`entially established by the purified enzyme at pH 4.8 and 45 °C. The amount of
`
`galacto-ollgosaccharldes (GOS) could be as high as 29% of the total sugar, and
`
`
`
`
`maximum production was achieved when 60% of the tnitial lactose was trans­
`
`
`wereThe Km values of the purilied P-o-galactosidase
`formed by B. bifidum cells.
`
`
`2 mM for p-nitrophenyl galactosidase and 13 mM for lactose. In the TG reaction,
`
`
`the Michaelis constant was 800 mM. Calcium and EDT A affected the TGase/
`
`
`
`
`
`hydrolase ratio. The purified enzyme exhibited a specificity towards glycosidic
`
`
`
`acceptors, and all tested P-o-galacto-oligosides can be used as galactosyl residue
`donor to produce GOS.
`
`in /3-0-Galactosidase is an enzyme which is widespread (EC 3.2.1.23)
`
`
`
`
`
`Nature [ 1] and is able to catalyse not only the hydrolysis of /3-0-galacto­
`
`
`
`sidic linkages, but also the transgalactosylation (TG) reaction to produce
`oligosaccharides
`I 2-14].
`
`
`
`Transfer reaction products do not seem to have a predominant
`physiological role, even though allolactose formed by Escherichia coli
`
`
`
`
`
`
`galactosidase is a natural inducer of the Jae operon [ 15]. Galacto-oligo-
`
`1 This work was supported in part by the Centre National de la Recherche Scientifique
`
`
`
`
`
`
`
`
`Director: membranaires; des Constituants UMR n° 111 (Relations structure-fonction
`
`
`
`
`
`Professeur Andre Verbert) and by the Universite des Sciences et Technologies de Lille.
`
`2To whom correspondence should be sent.
`3 Abbreviations used: pNP-Gal, p-nitrophenyl /3-D-galactopyranoside; GOS, galacto­
`
`
`
`
`
`
`oligosaccharide; TG, transgalactosylation; TGase, transgalactosylase.
`
`

`

`
`
`342 Dumortier, Brassart and Bouquelet
`
`saccharides (GOS) might be useful as a bifidobacterial growth
`
`
`
`
`
`
`
`factor (16) or for attenuating the intestinal disorders of elderly patients
`
`
`
`(17). Nevertheless, production of GOS by the TG reaction has recently
`
`
`
`
`
`become of interest. Glycoside moieties of glycoconjugates participate in
`
`
`
`
`
`recognition phenomena, and the synthesis of related structures can be
`
`
`
`
`performed using ,8-galactosidase [ 18). The structure elucidation of TG
`
`
`
`
`products showed generally the prevailing formation of /3(1,6) linkages.
`
`
`
`
`Nevertheless, the synthesis reaction also can form other linkages, such as
`
`
`
`,8(1,4) galactosidic linkage [19).
`In previous papers it was reported that B. bifidum, a bacterium from
`
`
`
`
`
`
`
`human colon, is able to exert a beneficial influence on the health of
`
`
`
`
`breast-fed infants (20-21). We showed that a ,8-o-galactosidase from B.
`
`
`
`
`bifidum exhibited a transgalactosylase (TGase) activity (22). The structure
`
`
`
`determination of ten GOS synthesized from lactose showed the predomi­
`
`
`
`nance of /3(1,3) linkages; such GOS had never previously been shown to
`
`
`be products of TG. Hence, it was of interest to know whether
`B. bifidum
`
`
`was able to synthesize in vivo such GOS structures. The GOS production
`
`
`
`was optimized, and the properties of the enzyme are described and
`
`
`
`
`
`discussed. Previous experiments indicated the presence of different
`
`
`of these ,8-o-galactosidases in B. bifidum cells (23). The purification
`
`enzymes allowed us to determine whether all or only one ,8-o­
`
`
`
`galactosidase was responsible for the specific TG reaction.
`
`Experimental
`
`Materials
`BSA, p-nitrophenyl glycosides and chemicals were purchased from
`
`
`
`
`
`
`
`
`Sigma. Lactose, glucose and galactose were obtained from Merck.
`
`
`
`Hydroxyapatite Ultrogel, and Q Fast Flow, Mono P, Superose 6 and
`
`
`
`
`Polybuffer 74 were products of Sepracor and Pharmacia respectively.
`
`
`
`Glucose 6-phosphate dehydrogenase, hexokinase, NADP, ATP, and
`
`
`
`were purchased from Boehringer Mannheim. Arninex HPX 42C
`MgSO
`
`
`
`
`columns (300 mm x 7.8 mm), Bio-Rad protein assay reagent and mole­
`
`
`
`
`cular-mass standards for SDS/PAGE were obtained from Bio-Rad, and
`
`
`
`molecular-mass standards for gel chromatography were from Pharmacia.
`
`4
`
`Organisms
`Cultivation conditions for B. bifidum (DSM 20082) were optimized by
`
`
`
`
`
`Boutry (24). Cultures were grown in anaerobic medium in a Setric
`
`
`
`
`fermenter with a final working volume of 200 litres at constant pH 6 and
`
`
`at 37 °C. Cells were harvested by centrifugation at 7 000 g, 10 h after the
`
`end of the exponential phase of growth.
`
`Protein was measured by the method of Bradford [25], with BSA as
`
`Protein assay
`
`
`
`
`standard.
`
`

`

`bifidum ,8-o-galactosidase
`343
`Bifidobacterium
`
`Enzyme assays
`with 1.25 mM p-nitrophenyl galacto­
`was determined
`activity
`Hydrolytic
`side (pNP-Gal)
`pH 6.5.
`buffer,
`in 0.1 M sodium phosphate
`as substrate
`by the
`was assayed
`was used as the substrate, the
`glucose
`When lactose
`method of Finch [26). One unit of ,8-o-galactosidase
`as the
`was defined
`or 1 µmol of
`1 µmol of p-nitrophenol
`amount of enzyme producing
`at 37 °C and at pH 6.5.
`glucose/min
`The incubation
`method.
`by an h.p.l.c.
`was measured
`TGase activity
`300 µl of 0.05 M
`enzyme,
`of 100 µl of diluted
`was composed
`mixture
`pH 4.25, and 100 µl of
`buffer,
`phosphate
`disodium
`sodium citrate/0.1 M
`was stopped
`by
`reaction
`After 4 h at 45 °C, the enzymic
`0.5 M lactose.
`mixture
`bath, then the incubation
`for 2 min on a boiling-water
`heating
`four
`was diluted
`for 30 min at 35 000 g. The supernatant
`was centrifuged
`(10 µl) on to an h.p.l.c.
`injection
`column (Aminex
`times with water before
`at 85 °C in a SP 8100 apparatus
`controlled
`HPX 42C) thermostatically
`were eluted by
`Oligosaccharides
`Products).
`(Thermo Separation
`by refracto­
`of 0.75 ml/min and detected
`water at a flow rate
`de-aerated
`by using 10 µl of
`were calculated
`concentrations
`Oligosaccharide
`metry.
`were
`The results
`as reference.
`(20 mM lactose)
`external standard
`A unit of TGase was defined
`as
`equivalents.
`as trisaccharide
`expressed
`of 1 µmol of tri­
`the formation
`the amount of enzyme catalysing
`at 45 °C and at pH 4.25.
`saccharide/min
`
`of optimal conditions
`Determination
`to the standard
`according
`was determined
`temperature
`The optimal
`from 4 to 60 °C. For the TG reaction,
`ranging
`at temperatures
`procedure
`in 0.05 M sodium citrate/
`with 0.24 M lactose
`the enzyme was incubated
`pH 4.8, for 20 h in the same range of tempera­
`phosphate,
`0.1 M disodium
`in the
`60 (Merck)
`on Kieselgel
`by t.l.c.
`were separated
`ture. Saccharides
`was done
`acid (2: 1 : 1, v /v). Staining
`system butanol/water
`/ acetic
`solvent
`(0.2%) in 20% (v/v) H2SO4 and heating
`orcinol
`by spraying
`for 5 min at
`(Bio-Rad
`620
`by densitometry
`were estimated
`120 °C. Saccharides
`densitometer).
`using pNP-Gal
`was determined
`activity
`pH for hydrolytic
`The optimal
`in 0.1 M sodium citrate/phosphate/borate
`which had
`buffer,
`as substrate
`a wide
`it covers
`was chosen because
`a pH range from 4 to 10. This buffer
`For the TG reaction,
`the nature of the salt.
`changing
`range of pH without
`at different
`with 0.24 M lactose
`were incubated
`50 µl of ,8-o-galactosidase
`by heating
`and
`was stopped
`After 20 h at 45 °C, the incubation
`pH values.
`above.
`with TLC, as described
`were estimated
`products
`the reaction
`
`on GOS production
`concentration
`of lactose
`Influence
`concentrations
`lactose
`with various
`studies
`Cells were used for kinetic
`phosphate
`M disodium
`(100, 250 or 400 mM) in 0.05 M sodium citrate/0.1
`(0.4 ml) were withdrawn
`times from 30
`at various
`pH 4.25. Samples
`buffer,
`After centrifugation
`min up to 48 h and the reactions stopped
`by heating.
`by h.p.l.c.
`were analysed
`products
`with water, the reaction
`and dilution
`
`

`

`
`
`
`
`344 Dumortier, Brassart and Bouquelet
`
`Enzyme extraction
`B.bifidum cells were washed twice with 0.03 M sodium phosphate buffer,
`
`
`
`pH 6.8. The cells were suspended (0.4 g/ml) in sodium phosphate buffer,
`
`
`pH 6.8, containing 0.2% Triton X-100. They were subjected for 15 min at
`
`
`
`
`a temperature of 4 °C to a pulsed (0.3 s/s) ultrasonic treatment (Branson
`
`
`
`sonifier B 30) with a frequency of 20 kHz and an amplitude of 30 µm. The
`
`
`
`extract was centrifuged at 4 °C at 35 000 g to obtain a clear crude extract.
`
`
`
`
`The pellet was subjected to three other ultrasonic treatments under the
`same conditions.
`
`3
`
`Enzyme purification
`All procedures were conducted at room temperature in sodium phosphate
`
`
`
`
`
`
`(0.02%) as preservative. The following steps were
`
`buffer, with NaN
`
`carried out.
`The crude extract
`
`Step 1: Hydroxyapatite-Ultrogel chromatography
`
`
`was chromatographed on a column (5 cm x 20 cm) of HA Ultrogel equili­
`
`
`brated with 30 mM sodium phosphate buffer, pH 6.8, at a flow rate of 350
`
`
`
`ml/h. Elution was performed with a stepwise change of sodium
`
`
`
`
`phosphate buffer (0.03, 0.1 and 0.3 M), pH 6.8. Absorbance at 280 nm was
`
`
`
`
`measured. The enzyme fractions were pooled, concentrated and dialysed
`
`
`
`against 0.01 M sodium phosphate buffer, pH 6.8.
`The enzyme solution from Step 2: Q Fast-Flow chromatography
`
`
`
`
`
`previous step was layered on to a Q Fast Flow column (1.6 cm x 15 cm)
`
`
`
`
`equilibrated with 0.01 M sodium phosphate buffer, pH 6.8. Proteins were
`
`
`
`
`
`eluted by a stepwise increase of NaCl concentration (0.1, 0.25, 0.5, 1 M) in
`
`
`equilibration buffer at a flow rate of 250 ml/h.
`Q Fast Step 3: Chromatofocusing method The enzyme solution from
`
`
`
`
`
`
`Flow chromatography was dialysed against a 25 mM BisTris/HCl buffer,
`
`pH 6.5, and injected into a Mono P column (0.5 cm x 20 cm) at a flow rate
`
`
`
`
`of 30 ml/h. The enzyme was eluted by feeding Polybuffer 74/HCl, pH 4.5,
`
`
`
`which formed a pH gradient ranging from 6.5 to 4.5 (Pharmacia Technical
`Bulletin).
`Purification was achieved by gel
`
`
`
`Step 4: Gel-filtration chromatography
`
`
`chromatography on a Superose 6 HR 10/30 column (1 cm x 30 cm)
`
`
`
`
`equilibrated with 0.25 M NaCl/0.05 M sodium phosphate buffer, pH 6.5,
`at a flow rate of 24 ml/h.
`
`Electrophoretic analysis
`
`Non-denaturating PAGE was carried out in a 7.5% polyacrylamide gel at
`
`
`
`
`pH 8.9. After the electrophoretic run at 4 °C, the gel was cut into two. The
`
`
`
`
`
`first part was stained for protein with Coomassie Brilliant Blue G-250. The
`
`second part was cut into 5 mm bands to test for ,8-o-galactosidase activity.
`
`
`
`SDS/PAGE according to Weber and Osborne [27) was carried out with
`
`
`a 5-15% polyacrylamide gel. The samples were heated at 100°C for 5 min
`
`
`with SDS (2%) with or without ,8-mercaptoethanol (4.5%) and Bromo­
`
`
`
`
`phenol Blue as migration marker. The proteins were stained as previously
`described.
`
`

`

`bifidum fJ-o-galactosidase
`345
`Bifidobacterium
`
`Immunological studies
`
`An antiserum against the B. bifidum crude extract was prepared by
`
`
`
`
`
`
`
`
`subcutaneous inoculation of rabbit with enzyme crude extract suspended
`
`
`
`
`
`in complete Freund's adjuvant as described by VaitukaHis [28). A booster
`
`
`
`injection was made a month later. Electrophoresis on 0.9% Indubiose gel
`
`
`
`
`(Sepracor) in veronal buffer was carried out with 4 V / cm for 2 h. Precipi­
`
`
`tated lines appeared after 24 h of antiserum diffusion.
`
`Determination of M,
`
`The M, of native TG was determined by gel chromatography on a TSK
`
`
`
`HW 55 S (Fractogel; Merck) column (1.6 cm x 90 cm) equilibrated with
`
`
`0.05 M sodium phosphate buffer, pH 6.5, containing 0.5 M NaCl, at a flow
`
`
`
`(232 (440 kDa); catalase rate of 21 ml/h. Thyroglobulin (669 kDa), ferritin
`kDa) and BSA (67 and 134 kDa) were used as M, markers.
`
`Determination of pl
`
`The pl was determined by the chromatofocusing method previously
`
`
`
`
`described (step 3 of the purification procedure).
`
`Enzyme stability
`The kinetics for the purified enzymes were determined at different
`
`
`
`
`
`
`buffer, pH 6.5. temperatures (18-60°C) in 0.1 M sodium phosphate
`
`
`(2 mM) were checked for protective effects.
`
`Lactose (0.2 M) and CaCl2
`
`
`The enzyme was preincubated for 4 h at 37 °C at different pH values in
`
`
`
`0.1 M sodium citrate/phosphate/borate buffer. The influences of lactose
`
`
`
`
`and calcium on pH-stability were performed with 24 mM lactose and 0.4
`mM CaC12•
`
`Kinetic constants
`
`For hydrolytic activity, kinetic constants were determined with lactose
`
`
`
`
`
`
`
`and pNP-Gal as subtrates in 0.1 M sodium phosphate buffer, pH 6.5, with
`
`
`concentrations ranging from 0.005 to 0.3 M and from 0.5 to 10 mM for
`
`
`
`lactose and pNP-Gal respectively. For TGase, determination of kinetic
`
`
`
`parameters was made in 0.05 M sodium citrate/0.1 M disodium phosphate
`
`
`
`buffer, pH 4.8, at 45 °C with lactose concentration ranging from 0.05 to
`
`
`
`
`0.4 M. Hydrolytic and TG kinetic constants were determined by using
`
`Hanes [29) and Lineweaver-Burk [30] plots respectively.
`
`Effectors
`The effects of different cations (2.5 mM) and EDTA (5 mM) were tested
`
`
`
`
`
`
`with 1.25 mM pNP-Gal for hydrolytic activity and with 0.25 M lactose for
`
`
`
`
`TGase. In the latter case, relative TG activity was estimated by densito­
`metry after t.l.c.
`
`Specificity of the TG reaction
`
`
`
`The specificity of the donor was studied by using the following reaction
`
`
`
`
`mixture: 25 µl of the purified galactosidase (50 munits) was added to 125
`
`

`

`
`
`
`
`346 Dumortier, Brassart and Bouquelet
`
`µl of 0.1 M sodium citrate/0.2 M phosphate buffer, pH 4.8, and 100 µl of
`
`
`
`
`of the acceptor was
`
`
`0.5 M lactose or 0.05 M pNP-Gal. The specificity
`
`
`studied with different sugars and sugar derivatives (concentrations
`
`
`ranging from 0.75 M to 1.5 M). In that case, 50 munits of purified enzyme
`
`
`(25 µl) were incubated with 25 µl of 0.15 M sodium citrate/0.3 M
`
`
`
`
`phosphate buffer, pH 4.8, 0.75 or 1.5 M acceptor molecule (50 µl) and
`
`
`
`0.5 M lactose or 0.05 M pNP-gal (25 µl) as galactosyl donor. After 1-5 h of
`
`
`
`incubation the reaction was stopped by heating and the incubation
`
`
`mixtures were analysed by t.l.c.
`
`Results
`
`The choice of different conditions for assaying TG reaction and hydrolysis
`
`
`
`
`cells catalysed respectively can be explained by the fact that B. bifidum
`
`
`
`
`
`
`
`the transfer reaction optimally at pH 4.5, in contrast with lactose
`
`
`hydrolysis, which took place pH 6.5. In order to minimize the hydrolysis
`
`
`
`
`of GOS by ,8-o-galactosidase activity during lactose incubation, the TG
`
`
`
`reaction was assayed at pH 4.25 and at 45 °C. Moreover, the TG reaction
`
`
`was more favoured by sodium citrate/phosphate buffer than by sodium
`
`acetate buffer (results not shown).
`Different lactose concentrations were used for kinetic assays (0.1-0.4 M
`
`
`
`
`
`
`
`
`
`
`lactose). The typical course for 0.4 M lactose transformation by B. bifidum
`
`
`cells is illustrated in Figure 1. For 0.4 M lactose, the maximal yield of
`
`GOS reached 15% of the total sugar (w/w) after 5 h incubation. The
`
`
`amount of GOS can be related to the conversion degree (CD), which is
`
`
`defined according to the following relationship:
`
`CD= _10_0_-_re_s_id_u_al_d_is_a_cc_h_an_· d_e_(:....%�)
`100
`
`100
`
`Figure 1
`
`Course of lactose hydrolysis by B. bifidum
`
`
`
`
`
`A 0.4 M lactose solution was incubated at
`pH 4.25 and 45 °c with Bifidobacterium
`
`
`
`cells. Aliquots were taken and analysed by
`
`
`h.p.l.c. as described in the texl Amounts of
`( x ), di­tri-, tetra-and penta-saccharides
`
`( + ), glucose (I:!.) and galactosesaccharides
`
`(□) were expressed
`
`as relative percentages
`(w/w).
`
`
`of initial substrate concentration
`
`� 80
`
`(/)
`
`(/)
`
`60
`
`0
`C 40
`0
`'.;J
`
`0
`
`0
`
`c.. 20
`
`5 10 15 20 25 30 35 40 45 50
`Times (h)
`
`

`

`
`
`Bifidobacterium bifidum /3-0-galactosidase
`347
`
`Figure 2 shows that the maximum amount of GOS was obtained when
`
`
`
`
`the conversion degree of disaccharides was about 60%. For higher values
`
`
`of CD, GOS were hydrolysed. The optimization of GOS production
`
`
`achieved a maximum production of 29% of the total sugar after 80 h of
`
`
`incubation with 500 mM lactose at pH 4.25.
`
`Extraction by ultrasonic treatment
`
`
`Previous experiments suggested that the various /j-o-galactosidases from
`
`
`
`
`
`
`
`
`were intracellular {soluble form) or membrane-located [23,24].
`B.bifidum
`
`
`in Table 1. The totalThe results of the present extraction are summarized
`
`
`
`
`
`
`recovery (cell remnants and supernatant) after ultrasonic treatment for
`
`
`
`
`activity (149%) of thethe hydrolytic activity is higher than the initial
`
`
`
`
`intact resuspended cells. It suggests that a part of the total P-o-galacto­
`
`
`
`
`
`sidase activity is intracellularly located. About 60% of hydrolytic activity
`
`
`
`
`
`was still retained in cell remains, as was TG activity (55.7%). Only 70% of
`
`
`
`
`the total TGase activity was recovered, suggesting that drastic conditions
`
`Figure 2
`
`GOS formation by B. bifidum as a function
`
`
`
`
`of the degree of conversion and initial lac­
`
`
`tose concentration. GOS quantities were
`
`
`
`expressed as weighable percentages of total
`
`sugar concentration (w/w). The degree of
`
`
`
`conversion was calculated according to the
`
`
`relationship described in the text. Initial
`
`
`
`substrate (lactose) concentrations were 100
`mM (6), 250 mM (+) and 400 mM (□).
`
`20
`
`;a
`15
`£
`0
`�
`
`Cl)
`
`10
`·;:;
`
`"' u
`
`
`u
`
`"' .;,
`
`
`5
`
`+
`·.:
`
`10 20 30 40 50 60 70 80 90 100
`
`
`
`Degree of conversion(%)
`
`
`
`
`
`Table 1 Extraction of /1-galactosidase and /1-transgalactosylase activities from B. bifldum
`
`
`
`cells
`
`Fraction
`
`/J-Galactosidase
`/J-Transgalactosylase
`TGIH ratio•
`
`Activity (units)
`
`2176
`Cells
`Cell remnants
`1 293
`
`Supernatant after disruption
`1943
`Total yield(%) ...
`149
`89
`
`Extract yield(%) ...
`
`11.2
`6.24
`1.59
`70
`14.2
`
`5.15x10 3
`4.83 X 10-3
`0.82 X 10-J
`
`
`
`
`
`• TG/H ratio, TG reaction/hydrolysis ratio.
`
`

`

`and Bouquelet
`Brassart
`348 Dumortier,
`
`of the enzyme. The
`and the solubilization
`during the extraction
`prevailed
`is the same in cells or in cell remnants
`and
`ratio TG reaction/hydrolysis
`in the
`are located
`activity
`the idea that the TGase and hydrolytic
`supports
`in the ·solubilized
`decreased
`This ratio was dramatically
`same structure.
`for TGase activity.
`It
`with only 14.2% of recovery
`enzyme· fraction,
`to the fJ-o­
`relative
`showed that there was 6-fold less TGase activity
`used as crude extract compared
`in the supernatant
`activity
`galactosidase
`cells.
`with the intact
`
`Purification
`fJ-o-galactosidases
`various
`procedure,
`purification
`During a four-step
`but only one was able to form GOS at pH 4.25 with
`were characterized,
`as substrate.
`lactose
`three fJ-o-galactosidases
`were
`chromatography,
`By hydroxyapatite
`Only the
`buffer).
`0.1 and 0.3 M sodium phosphate
`characterized
`(0.03,
`gave the TG reaction.
`This
`enzyme eluted by 0.1 M sodium phosphate
`of the TC/hydrolysis ratio
`first step gave the most significant increase
`of the enzyme by Q Fast Flow chromato­
`(4.6-fold). Further purification
`peak showed
`but only the 0.1 M NaCl-eluted
`graphy gave two entities,
`23-fold.
`was purified
`During this step the TGase activity
`the TG reaction.
`of all traces of fJ-o­
`the elimination
`permitted
`Mono P chromatography
`by Superose
`6
`of TGase was achieved
`The purification
`galactosidase.
`chromatography.
`procedure.
`the purification
`concerning
`the results
`Table 2 summarizes
`in a final yield of 12% from the super-
`TGase was obtained
`The purified
`
`of P-o-gaJactosidase/!ransgaJaclosylase
`from a B. bifidum extract
`Table 2 Purification
`{a) Hydrolysis.
`
`Step
`
`activity Yield Purification
`Activity Specific
`(fold)
`(units) (units/mg) (%)
`
`0.92
`Crude extract 1553
`0.66
`Hydroxyapatite 187
`57.4 5.50
`Q Fast Flow
`18.5 29.3
`Mono P
`12.02 49.8
`Superose
`6
`
`100
`12
`3.7
`1.2
`0.8
`
`1
`0.71
`5.98
`31.85
`54.13
`
`(b)TransgaJactosylation
`reaction
`
`Step
`
`x TG!H
`10 3
`activity Yield Purification
`Activity Specific
`(munits) (mu nits/mg) (%) (fold) ratio•
`
`0.82
`Crude extract 1272
`2.22
`710
`Hydroxyapalite
`50.2
`Q Fast Flow 446
`237.5 831.6
`Mono P
`155 1410
`Superose
`6
`
`0.82
`100 1
`3.80
`55.8 2.7
`35.8 61.2 7.75
`12.82
`18.8 1014
`12:!J0
`12.2 1720
`
`TG reaction/hydrolysis
`ratio.
`• TG/H ratio,
`
`

`

`
`
`Bifidobacterium bifidum
`
`,8-o-galactosidase 349
`
`natant of extraction and a purification factor of 1 720. This enzyme
`
`
`
`
`
`
`
`represented only 0.8% of the initial galactosidase activity present in the
`
`
`
`crude extract. The increase in the TGase/galactosidase ratio during the
`
`
`procedure showed that not all the ,8-o-galactosidases were able to
`produce GOS.
`Immunoelectrophoresis of purified enzyme versus crude extract
`
`
`
`
`
`antiserum showed only one precipitatin line. Electrophoresis under non­
`
`
`
`denaturing conditions in 7.5%-polyacrylamide gel showed that the trans­
`
`
`
`
`
`galactosylase was homogeneous. Coomassie Brilliant Blue staining showed
`
`
`
`one major band. /J-o-Galactosidase activity detected on the polyacryl­
`
`
`
`amide gel was superimposed on the protein band (results not shown).
`
`Physical properties
`
`SDS/PAGE showed the presence of four subunits of molecular masses
`
`
`
`
`
`respectively 163, 170, 178 and 190 kDa as estimated by using molecular­
`
`
`
`
`mass markers (Figure 3). The molecular mass of the native galactosidase
`
`
`was estimated to be about 362 kDa by gel filtration on TSK HW 55 S. In
`
`
`
`
`comparison, the sum of the different subunits gave a molecular mass of
`
`700 kDa. It was twice that found by chromatography. The isoelectric point
`
`
`was estimated as 5.25 by chromatofocusing on Mono P.
`
`Enzyme stability
`The purified transgalactosylase may be stored for several weeks at 4 °C in
`
`
`
`
`
`
`0.1 M sodium phosphate buffer, pH 7, with 0.02% NaN
`as preservative.
`
`
`The enzyme is stable for 4 h below 45 °C, but was completely denaturated
`
`
`
`, and particularly lactose, provided a substantial
`after 1 h at 60 °C. Ca2+
`
`
`
`heat-protective effect. After 3 h of incubation at 50 °C, transgalactosylase
`
`
`
`
`in phosphate buffer lost more than 70% of its initial activity, but in the
`
`
`
`
`presence of 0.2 M lactose the enzyme remained 100% active. Similarly,
`2 mM Ca2+ showed a moderate
`
`
`
`
`protective effect resulting in retention of
`
`
`
`60% of the initial activity. The enzyme was stable at near-neutral pH
`
`and Ca2+ were also able to preserve galactosidase
`
`(6.5-8). Lactose
`
`
`
`
`activity, particularly at acid pH values. Therefore, at pH 5.2, 95% of
`
`3
`
`Figure 3
`A B C D E
`F A Molecular
`mass (Da)
`SOS/PAGE of the different enzymic frac­
`
`
`tions with a 5-15%-polyacrylamide gra­
`
`dienL A, molecular-mass markers (myosin,
`
`200 000 Da; E. coli p-galactosidase, 116 250
`
`Da; phosphorylase b, 97 400 Da; BSA,
`
`66 200 Da; ovalbumin, 42 700 Da); B, crude
`
`
`extract; C "HA 100· fraction; D, "QFF 0.1 M
`
`
`NaCl' fraction; E, 'Mono P' fraction; F,
`
`
`'Superose 6" fraction .
`
`w:i 116250
`..... ◄ 97400
`wJ,. 66200
`
`..i ◄ 200000
`
`..- 42700
`
`

`

`
`
`
`
`350 Dumortier, Brassart and Bouquelet
`
`
`
`
`
`activity was preserved as compared with only 65% for lactose-free con­
`
`
`
`trol.
`
`Optimum pH values and temperatures for the purified enzyme
`
`
`
`
`The optimum conditions for the TG reaction were pH 4.8 and 45 °C. In
`
`
`
`
`comparison, hydrolysis using pNP-Gal or lactose as substrates showed its
`
`
`optimal pH and temperature as pH 6.5 and 37-39 °C respectively.
`
`Kinetic constants
`constants
`
`
`
`Under their optimal pH and temperature conditions, Michaelis
`
`
`
`for hyrolase and TGase activities were determined (see the Experimental
`between the Km values for hydrolytic
`
`
`section). The difference obtained
`
`
`
`
`activity towards lactose (13 mM) and the TG reaction (800 mM) is not
`
`
`
`
`solely due to the differences in experimental conditions, but also reflects
`
`
`
`
`the requirement for TGase to assemble at least two substrate molecules in
`
`its catalytic subsite.
`were 2.2 for pNP-Gal and lactose At pH 6.5, Km values of galactosidase
`
`
`
`
`
`and 13 mM and the maximum reaction velocities were 3.04 and 2.70
`µmo!· min -1 ( V maxJ respectively.
`
`Effects of metal ions and EDTA
`The effective action of different components was determined using pNP­
`
`
`
`
`
`
`
`
`Ga1 hydrolysis at pH 6.5, lactose hydrolysis at pH 4.8 and the TG reaction
`
`at pH 4.8. The results are summarized in Table 3. Except for Mn2+ and
`
`
`
`
`Ca2+ most of the cations decreased both activities to the same extent.
`Ca2+ at 2.5 mM increased
`
`
`the hydrolytic activity by 230% under acidic
`
`
`
`
`conditions, whereas at the same time GOS production decreased by more
`
`
`
`
`than 50%. In contrast, EDT A decreased the hydrolytic activity at pH 6.5
`
`
`
`
`and particularly favoured the TG reaction, with an increase of 46%. This
`
`action occurred at pH 4.8, which is an unfavourable pH value for Ca2 +
`
`
`
`Table 3 Effect of effectors on hydrolase and transgalactosylase activities
`
`
`
`
`
`
`
`
`Hydrolysis of pNP-Gal was measured at pH 6.5 by a standard procedure. Hydrolysis of
`
`
`
`
`lactose and transferase activity were determined at pH 4.80 and 45 •c by a method based
`
`
`
`
`upon monosaccharide or trisaccharide formation and detection by t.l.c.
`
`Concn.
`Effector (mM)
`
`Transferase
`pNP-Gal Lactose activity•
`
`
`
`Hydrolysis• of:
`
`CaCl2
`MnCl2
`MgCl2
`C0Cl2
`ZnC12
`SnCl2
`EDTA
`
`2.5
`2.5
`2.5
`2.5
`2.5
`2.5
`5
`
`150
`126
`98
`102.5
`76
`69
`70.3
`
`230
`121
`94
`66
`92
`325
`95
`
`46.2
`87.5
`98
`80
`90
`67
`146
`
`
`
`• Activity compared with effector-free control(• 100).
`
`
`
`

`

`bifidum fi-o-galactosidase
`351
`Bifidobacterium
`
`by EDTA. Ca2+ at 25 µM increased
`5-fold
`the Vmax. of pNP-Gal
`chelation
`the Vmax. (2.6-fold)
`hydrolysis
`at pH 6.5, whereas
`50 µM EDTA decreased
`and the Km (2.0-fold)
`of hydrolysis.
`
`TG reaction specificity
`The specificity
`was determined
`after a 5 h incubation
`at pH 4.8 and at
`45 °C. No reversal
`of the reaction
`was detected
`when galactose
`is used as
`substrate.
`All ,B-o-galactosidase,
`i.e. pNP-Gal, lactose, lactulose,
`lactitol,
`allolactose,
`,8Gal(1-3)Glc,
`and ,8Gal(l-3),8Gal(l-4)Glc
`can act as a
`substrate
`to produce
`GOS with the purified
`enzyme.
`Lactose
`at 0.1 M and pNP-Gal at 0.01 M were used as donors of
`galactosyl
`residues
`for the acceptor-specificity
`study. Different
`com­
`ponents were
`studied:
`saccharides,
`hydroxyamino
`acids,
`alcohols
`and
`polyols.
`Transgalactosylase
`exhibited
`the specificity
`for glyco­
`sidic acceptors
`shown in Table 4. Glucose
`and galactose
`can serve as
`acceptors, whereas
`mannose,
`fructose
`and fucose
`were unable to accept a
`galactosyl
`residue.
`Furthermore,
`when their hemiacetal
`groups were
`blocked
`by a methyl or a p-nitrophenyl
`group, their capacities
`to receive
`a galactosyl
`residue
`increased.
`These structural
`modifications,
`however,
`did not provide
`acceptor
`capacity
`for mannose.
`Interestingly,
`when the
`2-hydroxy
`groups were replaced
`by amino or acetamido
`groups,
`glucose
`and galactose
`derivatives
`lost