Journal of Biotechnology 81 (2000) 73 – 84

www.elsevier.com/locate/jbiotec

Purification and partial characterization of

fructosyltransferase and invertase from Aspergillus niger
AS0023
Lamia L’Hocine *, Zhang Wang, Bo Jiang, Shiying Xu
School of Food Science, Wuxi Uni6ersity of Light Industry, 170 Huihe Road, 214036 Wuxi, People’s Republic of China

Received 20 April 1999; received in revised form 25 April 1999; accepted 3 May 2000

Abstract

Fructosyltransferase (EC.2.4.1.9) and invertase (EC.3.2.1.26) have been purified from the crude extract of
Aspergillus niger AS0023 by successive chromatographies on DEAE-sephadex A-25, sepharose 6B, sephacryl S-200,
and concanavalin A–Sepharose 4B columns. On acrylamide electophoresis the two enzymes, in native and denatured
forms, gave diffused glycoprotein bands with different electrophoretic mobility. On native-PAGE and SDS-PAGE,
both enzymes migrated as polydisperse aggregates yielding broad and diffused bands. This result is typical of
heterogeneous glycoproteins and the two enzymes have proved their glycoprotein nature by their adsorption on
concanavalin A lectin. Fructosyltransferase (FTS) on native PAGE migrated as two enzymatically active bands with
different electrophoretic mobility, one around 600 kDa and the other from 193 to 425 kDa. On SDS-PAGE, these
two fractions yielded one band corresponding to a molecular weight range from 81 to 168 kDa. FTS seems to
undergo association–dissociation of its glycoprotein subunits to form oligomers with different degrees of polymeriza-
tion. Invertase (INV) showed higher mobility corresponding to a molecular range from 82 to 251 kDa, on native
PAGE, and from 71 to 111 kDa on SDS-PAGE. The two enzymes exhibited distinctly different pH and temperature
profiles. The optimum pH and temperature for FTS were found to be 5.8 and 50°C, respectively, while INV showed
optimum activity at pH 4.4 and 55°C. Metal ions and other inhibitors had different effects on the two enzyme
activities. FTS was completely abolished with 1 mM Hg2 + and Ag2 + , while INV maintained 72 and 66% of its
original activity, respectively. Furthermore, the two enzymes exhibited distinctly different kinetic constants confirming
their different nature. The Km and Vm values for each enzyme were calculated to be 44.38 mM and 1030 mmol ml − 1
min − 1 for FTS and 35.67 mM and 398 mmol ml − 1 min − 1 for INV, respectively. FTS and INV catalytic activity was
dependent on sucrose concentration. FTS activity increased with increasing sucrose concentrations, while INV activity
decreased markedly with increasing sucrose concentration. Furthermore, INV exhibited only hydrolytic

Abbre6iations: FTS, fructosyltransferase; INV, invertase; FOS, fructooligosaccharides; Con A, concanavalin A; Ut, fructosyltrans-
ferase activity; Ui, invertase (hydrolyzing) activity; KU, katal; HPLC, high performance liquid chromatography; SDS, sodium
dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis.
* Corresponding author. Present address: Department of Food Science and Agricultural Chemistry, McGill University, 21,111
Lakeshore, Ste. Anne de Bellevue, Quebec, Canada H9X 3V9. Tel.: + 1-514-3988614; fax: +1-514-3988132.
E-mail address: lamia@macdonald.mcgill.ca (L. L’Hocine).

0168-1656/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 8 - 1 6 5 6 ( 0 0 ) 0 0 2 7 7 - 7
74 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84

activity producing exclusively fructose and glucose from sucrose, while FTS catalyzed exclusively fructosyltransfer
reaction producing glucose, 1-kestose, nystose and fructofuranosyl nystose. In addition, at 50% sucrose concentration
FTS produced fructooligosaccharides at the yield of 62% against 54% with the crude extract. © 281 Elsevier Science
B.V. All rights reserved.

Keywords: Fructosyltransferase; Invertase; Enzyme purification; Aspergillus niger

1. Introduction enced the results obtained and thus the conclu-

The production of fructooligosaccharides
(FOS) has received particular attention in recent
years because of their excellent biological and
functional properties and necessitates the devel-
opment of efficient enzymatic systems. FOS are
(1F(1-b-D-fructofuranosidase)n − 1 sucrose; GFn,
n= 2–4); they are mainly composed of 1-kestose
(GF2), nystose (GF3), and 1F-fructofuranosyl
nystose (GF4) in which fructosyl units are
bound at the b-2,1 position of sucrose. The en-
zyme source of FOS synthesis can be divided
into two classes; one is plants such as asparagus
roots (Shiomi, 1982), and onion bulbs (Henry
and Darbyshire, 1980); the other consists of mi-
croorganisms such as Aspergillus sp. (Hidaka et
al., 1988; Chang et al., 1994) and Arthrobacter
sp. (Fujita et al., 1990). The nomenclature of
FOS-producing enzymes remains in dispute.
Some researchers use the term fructosyltrans-
ferase (EC.2.4.1.9) (Shiomi and Izawa, 1980; Sh-
iomi, 1982; Praznik et al., 1990) whereas others
designate it as b-D-fructofuranosidase (EC.
3.2.1.26) (Hidaka et al., 1988; Chang et al.,
1994). The latter denomination is probably due
to the fact that fructosyltransferase activity was
originally found from the side action in the
course of invertase preparation when acting on
high concentration of sucrose (Straathof et al.,
1986; Fujita et al., 1990). Purification and pre-
liminary characterization of this enzyme from
various sources have been reported (Nishizawa
et al., 1980; Hidaka et al., 1988; Chang et al.,
1994; Chen et al., 1996). However, until
now the accumulated information on fructosyl-
transferase or fructofuranosidase is rather con-
fusing differing from one source to another,
from one microorganism to another, even from
one strain to another. The source of enzyme
and the difficulty of purification no doubt influ-
sion drawn.
The present paper deals with purification and
partial characterization of FOS-producing en-
zyme from Aspergillus niger AS0023. This strain
showed a very high enzyme productivity and
high fructosyltransferase activity; it produces
FOS at the yield of 54% (L’Hocine et al., 1997).
However, the enzyme used in the crude state
also show hydrolytic activity which is responsi-
ble of the appearance of fructose and the de-
crease of total FOS yield in the final reaction
products. The purification and the characteriza-
tion of this enzyme are necessary steps to im-
prove our understanding of its mode of action,
the nature of the hydrolytic activity and to de-
cide the type of enzyme in which it should be
classified.
2. Materials and methods
2.1. Materials
DEAE-sephadex A-25, sepharose 6B, sep-
hacryl S-200, concanavalin A–sepharose 4B
(ConA-sepharose 4B), and molecular weight
markers for gel electrophoresis were purchased
from Pharmacia. Sodium dodecyl sulfate (SDS),
polyacrylamide gel, and Tris (hydrox-
ymethyl)aminomethane (TRIS) were from
Sigma. Methyl-a-D-mannoside, N,N%-methylene-
bisacrylamide and 2-mercaptoethanol were ob-
tained from Fluka. N,N,N%,N%-Tetramethylene
diamine (TEMED), sodium borohydride, trifl-
uoroacetic acid were from Merck. Kestose, nys-
tose and fructosylnystose were gifts from Meiji
Saika (Japan). All others chemicals were of ana-
lytical grade and obtained from readily available
commercial sources.
 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84
2.2. Culti6ation conditions
A. niger AS0023 was isolated from soil and
characterized by the Microoganism Preservation
Center, Chinese Academy of Science (Jiang and
Wang, 1995). This strain was grown on potato –
agar slant (20% (w v − 1), 2% (w v − 1) sucrose and
2% (w v − 1) agarose); a solution of spores was then
obtained by washing the mycelium with 100 ml
sterilized water and was used as the inoculum. A.
niger AS0023 was cultivated for 24 h at 30°C in an
aerated and stirred 25-l pilot scale reactor. The
culture medium (17 l) consisted of 17.5% sucrose,
1% peptone, 0.5% yeast extract, 0.1% MgSO4, 0.1%
KH2PO4 adjusted to pH 6. The air was supplied at
0.5 v v − 1min − 1 (volume of air per volume of
medium per minute) for the initial 12 h, then
increased to 0.8 v v − 1min − 1 for the remaining 12
h. The reactor pressure was maintained at 0.1 mPa
with a stirring rate of 250 rpm. The mycelium was
then collected and washed several times with
deionic water and 0.1 M McIlvain buffer (pH 5).
It was then freeze dried and stored at −18°C until
use.
2.3. Enzyme assay
The enzyme activity was determined according
to the Hidaka et al. (1988) procedure. The reaction
mixture consisted of 25% sucrose (1 ml) as the
substrate, 0.1 M McIlvain buffer at the required
pH (1 ml) and enzyme solution (0.5 ml). The
enzyme reaction was carried out at 50°C for 1 h
with moderate shaking and terminated by heating
at 100°C for 10 min. The fructosyltransferase
activity (Ut) and the invertase (hydrolyzing) activ-
ity (Ui) were determined from the amount of
1-kestose and fructose produced, respectively. The
enzyme activity was expressed as katal (KU) and
was defined as the amount of enzyme required to
produce 1 mole of the corresponding saccharide
per second under the above conditions. The specific
activity was expressed in KU per kg of protein.
2.4. Analytical method
Enzyme reaction products were analyzed by high
performance liquid chromatography (HPLC, Wa-
75
ters Associates model 209), equipped with a differ-
ential refractive index RI401 detector; using Merck
Lichrosorb–NH2 column (0.4× 15 cm), at a tem-
perature of 30°C. A mixture of acetonitrile–water
(80:20, v/v) was used as the mobile phase at a flow
rate of 1.2 ml min − 1.
2.5. Estimation of protein concentration
Proteins were measured by the method of Lowry
et al. (1951) using bovine serum albumin as stan-
dard. Absorbance at 220 nm was used for monitor-
ing protein in column eluates.
2.6. Purification procedure
A summary of steps involved in the purification
scheme is presented in Table 1. Unless otherwise
specified all steps were conducted at room
temperature.
2.6.1. Enzyme extraction
In a typical experiment 50 g of dried mycelium
were suspended in 0.1 M McIlvain buffer (pH 5)
at a ratio of 1:50 (w/v) and extracted with a
semi-homogenizer (Janke & Kunkel Ultra-Turrax
T25) at 24 000 rpm for 3 min at 4°C. The homog-
enized suspension was then centrifuged at
10 000× g for 20 min and filtrated on cheese-cloth
to get a clear extract. This was considered as the
crude enzyme preparation and its level of activity
was taken to be 100% for calculation of recovery.
2.6.2. Ammonium sulfate fractionation
Solid ammonium sulfate was added, over ice,
into the crude extract to 30% saturation; after
centrifugation (10 000× g, 20 min), ammonium
sulfate was added to bring the supernatant to 100%
saturation. The latter was stored overnight at 4°C
and then centrifuged. The precipitate was redis-
solved and dialyzed against several changes of 0.01
M phosphate buffer (pH 6.5).
2.6.3. Ion exchange chromatography on
DEAE-sephadex A-25
The dialyzed enzyme solution was concentrated
by ultrafiltration, using a membrane with nominal
 76

Table 1
Purification of fructosyltransferase and invertase from A. niger AS0023

Step of Volume (ml) Total activity Specific activity (KU kg−1) Protein content (mg) Recovery (%) Purification fold Ut/UI
purification (KU 10−3)

Crude extract
Ut 2485 144 37 3819.8 100 0.51
Ui 276 72 100
(NH4)2SO4 fractionation
Ut 210 116 196 592.07 80.55 5.30
Ui 22 371 79.71 5.15 0.53
DEAE-sephadex A25
Ut 432 79 1001 78.88 54.86 27.05
Ui 152 1927 55.07 26.76 0.52
Sepharose 6B FTS fraction
Ut 27 53 2212 23.94 36.8 59.78
Ui 3 113 1.08 1.56 19.57
Sepharose 6B INV fraction
Ut NDa
Ui 19 111 5299 20.95 40.21 73.60 0
Sephacryl S200 FTS fraction
Ut 32 51 2811 18.11 35.42 75.97 22.30
Ui 2 126 0.72
Sephacryl S200 INV fraction
Ut ND
L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84

Ui 40 89 5210 17.16 32.25 72.36 0

ConA-sepharose 4B FTS fraction
Ut 110 23 2905 7.91 15.97 78.51
Ui ND
ConA-sepharose 4B INV fraction
Ut ND
Ui 97 35 7047 4.98 12.68 97.87

a
ND, not detected.
 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84
cutoff of 10 000 Da, and applied on DEAE-sep-
hadex A-25 column (2.6×50 cm) equilibrated
with 0.01 M phosphate buffer (pH 6.5). The non-
adsorbed protein fraction was eluted from the
column with starting buffer (500 ml) and the
adsorbed enzyme was eluted with a linear gradient
of sodium chloride (0 – 0.35 M) in the same buffer
(1500 ml) at a flow rate of 40 ml h − 1.
2.6.4. Gel filtration on Sepharose 6B
The active fractions eluted from DEAE-sep-
hadex A-25 were pooled and concentrated by
ultrafiltration (nominal cutoff of 10 000 Da) then
loaded on Sepharose 6B column (1.6 ×120 cm)
equilibrated with 0.1 M phosphate buffer (pH
6.5). The proteins were eluted with the same
buffer at the flow rate of 8 ml h − 1. Fructosyl-
transferase activity and invertase activity ap-
peared as two separable activities, designated as
fructosyltransferase (FTS) and invertase (INV)
fractions, respectively.
2.6.5. Gel filtration on sephacryl S-200
Each active fraction was concentrated by ul-
trafiltration (nominal cut off of 10 000) and chro-
matographed separately on a sephacryl S-200
column (1.6×150 cm) equilibrated with 0.1 M
phosphate buffer (pH 6.5). The proteins were
eluted with the same buffer at a flow rate of 12 ml
h − 1. Protein fractions corresponding to the same
activity were pooled and concentrated by ultrafil-
tration (nominal cutoff of 10 000).
2.6.6. Affinity chromatography on
ConA-sepharose 4B
Fructosyltransferase and invertase fractions
were applied separately for further purification on
a column of ConA-sepharose 4B (1× 31 cm) equi-
librated with 0.05 M McIlvain buffer (pH 6.5)
containing 0.5 M NaCl. The non-adsorbed
proteins were eluted from the column with the
same buffer and the adsorbed proteins with a
continuous gradient of methyl a-D-mannoside in
0.1 M McIlvain buffer (pH 6.5, 0.5 M NaCl) at a
flow rate of 30 ml h − 1. The eluting gradient for
fructosyltransferase was from 0 to 0.6 M methyl-
a-D-mannoside (160 ml). Invertase was eluted
with a stepwise gradient of methyl-a-D-man-
77
noside, first from 0 to 0.05 M (150 ml), then from
0.05 to 0.5 M (100 ml).
ConA-sepharose 4B was regenerated by wash-
ing the gel with 2–3 bed volumes of 0.1 M
Tris–HCl buffer (pH 8.5) containing 0.5 M NaCl,
then with 2–3 bed volumes of 0.1 M acetate
buffer (pH 4.5) containing 0.5 M NaCl. This cycle
was repeated three times. Re-equilibration was
done with 0.05 M NaCl, 0.001 M CaCl2 and 0.001
M MnCl2 in 0.05 M McIlvain buffer. The sepa-
rated and purified active fractions were concen-
trated and dialyzed against water, then
freeze-dried and stored at − 18°C.
2.7. Electrophoresis
Polyacrylamide gel slab electrophoresis (PAGE)
under native and denaturing conditions was per-
formed according to Laemmli’s discontinuous
Tris–glycine buffer system (Laemmli, 1970) at 10
mA for 2–3 h. For native-PAGE, 6.5% acrylam-
ide separating gel was used; the molecular weight
markers applied were Pharmacia high molecular
weight calibration kit. For SDS-PAGE, 7.5% ac-
rylamide and 0.1% SDS were used; the molecular
weight markers were Pharmacia high molecular
weight-SDS calibration kit.
Proteins in the gel were stained with Coomassie
blue G-250 and destained with 0.5M NaCl
aqueous solution. Fructosyltransferase and inver-
tase activities were detected in situ after native-
PAGE by the staining procedure of Gabriel and
Wang (1969).
2.8. Determinations of kinetics parameters
The initial reaction rates were determined for
the two enzymes, at their optimum conditions, for
various sucrose concentrations (FTS, 0.05–0.25
M and INV, 0.0125–0.05 M).
2.9. Effect of sucrose concentration on FTS and
INV acti6ity
The effect of sucrose concentration on the en-
zyme reaction rates was studied by monitoring the
enzyme activity and analyzing the carbohydrate
composition in the reaction mixtures at different
78 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84
sucrose concentrations (0.5 – 80%; w/v). Enzyme
dosage of 8.33× 10 − 6 KU of enzyme per g of
sucrose was used.
3. Results
3.1. Purification of fructosyltransferase and
in6ertase
The results of the purification procedure are
reported in Table 1. With ammonium sulfate frac-
tionation as an initial step of purification, about
80% of total fructosyltransferase and invertase
Fig. 1. DEAE-sephadex A-25 column chromatogram of the
‘ammonium sulfate fraction’. —, Protein (A220);
, fructosyl-
transferase and invertase activities; --, NaCl gradient.
Fig. 2. Sepharose 6B column chromatogram of the ‘DEAE-
sephadex A-25 fraction’. —, Protein (A220);
, fructosyltrans- publications (Smith and Ballou, 1974; Chu et al.,
ferase activity; , invertase activity.
activities could be recovered in the fraction of
30–100% ammonium sulfate saturation with a
purification of 5.3- and 5.1-fold, respectively.
Maximal saturation was needed to precipitate the
fructofuranosidase active fraction showing their
high hydrophilic nature. The second step of purifi-
cation was ion exchange chromatography on
DEAE-sephadex A-25 (Fig. 1) using a linear
sodium chloride gradient. In this step a large part
of the contaminating proteins was removed as
non-adsorbed proteins; fructosyltransferase and
invertase activities were eluted from the ion ex-
changer at the same NaCl concentration (0.05 M),
as one broad peak.
The elution profile of sepharose 6B gel filtration
chromatography (Fig. 2) revealed the presence of
two types of enzymes, one having mainly trans-
fructosylating activity and the other exclusively
hydrolyzing activity. However, the separation be-
tween the two proteins was not complete. Chro-
matography of both fractions on sephacryl S-200
(Fig. 3) improved considerably the purification for
fructosyltransferase (FTS) and invertase (INV) by
75.9- and 72.3-fold, respectively. Nevertheless, af-
ter analytical polyacrylamide gel electrophoresis
there still appeared to be contaminating proteins
in both fractions. Further purification was, there-
fore, pursued by affinity chromatography on
ConA-sepharose 4B column (Fig. 4). FTS and
INV were both retained on ConA-affinity gel
showing their glycoprotein nature and the pres-
ence of a-D-mannopyranosyl residues in their gly-
cosidic moiety. Besides, the different affinity of
the two enzymes for ConA-sepharose 4B sug-
gested a great difference in the composition and/
or the structure of their carbohydrate part. In fact
the invertase was eluted at concentrations of
methyl-a-D-mannoside as low as 0.01 M, while
fructosyltransferase was initiated at 0.25 M. The
latter enzyme yielded two active fractions; the first
was eluted from 0.25 to 0.4 M and the second
part from 0.4 to 0.6 M.
A noteworthy phenomenon is that in all chro-
matographic steps the two enzymes were eluted as
broad and asymmetric peaks. This behavior is
quite typical of polydisperse glycoproteins
(Beeley, 1985) and it has been reported in several
1983; Zarate and Belda, 1996).
 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84 79

one was equivalent to a molecular weight range

over 600 kDa and the second one from 193 to 425
kDa (average of 309 kDa). When the denatured
enzyme was examined on 7.5% SDS-PAGE (Fig.
6), both fractions yielded one band with elec-
trophoretic mobility corresponding to a molecular
weight range from 81 to 168 kDa (average of 125
kDa) confirming the polymeric structure of FTS.
The fructosyltransferase fraction with a molecular
weight in the range 600 kDa could be a series of
higher polymer of the enzyme (tetramer, hex-
amer). Native INV on 6.5% polyacrylamide gel
showed an electrophoretic mobility corresponding
to a molecular range from 82 to 251 kDa (average

Fig. 3. Sephacryl-S200 column chromatogram of the ‘sep-

harose-6B active fractions’ (a) fructosyltransferase fraction; (b)
invertase fraction. — , Protein (A220);
, fructosyltransferase
activity; , invertase activity.

When electrophoresed on SDS – acrylamide gels

(Figs. 5 and 6), fructosyltransferase and invertase
migrated as oligomeric aggregates yielding diffuse
and broad bands. The fructosyltransferase on na-
tive PAGE yielded two bands, in which both were
revealed to have enzymatic activity. This result
was in agreement with the elution profile obtained
from ConA-affinity chromatography.

3.2. General properties of fructosyltransferase and

in6ertase from A. niger
3.2.1. Molecular size
, fructosyltransferase activity;
Electrophoresis of native FTS on 6.5% poly-
acrylamide gel (Fig. 5) indicated that the two
bands of proteins had different mobility; the first
Fig. 4. Con A-sepharose 4B chromatogram of ‘sephacryl ac-
tive fractions’. (a) Fructosyltransferase fraction; (b) invertase
fraction. — , Protein (A220);
, invertase activity, -- -, methyl-a-D-mannoside gradient. The
arrow indicates the end of the elution and starting of the
regeneration process.
80 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84

Fig. 5. Native polyacrylamide gel electrophoresis (6.5%) of fructosylransferase and invertase fractions. (a) Gel was stained with
Coomassie blue G-250; (b) Gel was stained for enzyme activity. Lane 1, INV; lane 2, FTS; lane 3, proteins standards. The molecular
weight markers used were: thyroglobulin, 669 kDa; ferritin, 440 kDa; catalase, 232 kDa; lactate dehydrogenase, 140 kDa; albumin,
67 kDa.

of 166 kDa); when the denatured enzyme was

examined on 7.5% SDS-PAGE (Fig. 6) its mobil-
ity corresponded to a molecular weight range
from 71 to 111 kDa (average of 91 kDa).

3.2.2. Effect of pH on fructosyltransferase and

in6ertase acti6ity and stability
The activity of the two enzymes was measured
at various pH from 2 to 11 (Using 0.1 M McIl-
vain buffer for pH 2 – 8 and 0.2 M glycine buffer
from pH 8.2 to 11). As shown in Fig. 7a, FTS was
most active at pH 5.8. INV showed maximal
activity at pH 4.4, which rapidly decreased above
pH 5 and the enzyme had almost no activity
above pH 7.
To determine the effect of pH on stability the
enzymes solutions were incubated, for 2 h at
25°C, in various pH buffers (from pH 2 to11). Fig. 6. SDS-polyacrylamide electrophoresis (7.5%) of fructo-
The remaining enzyme activity was then assayed syltransferase and invertase fractions. Lane 1, INV; lane 2,
by the standard method. The two enzymes were FTS; lane 3, proteins standards. The gel was stained with
Coomassie blue G-250. The molecular weight markers used
stables from pH 4.5 to 11 (Fig. 7b). In this range were: myosin, 212 kDa; macroglobulin, 170 kDa; galactosi-
of pH the fructosyltransferase showed a slight dase, 116 kDa; transferrin, 76 kDa; glutamic dehydrogenase,
higher stability than invertase. 53 kDa.
 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84 81

3.2.3. Effect of temperature on FTS and INV

acti6ity and stability
The two enzymes activity was measured by the
standard method, at their optimum pH, at various
temperatures from 30 to 70°C. The optimum tem-
perature for FTS was 50°C, while INV showed an
optimum at 55°C (Fig. 7c). The effect of tempera-
ture on stability was studied by keeping the en-
zymes, for 2 h, at various temperatures
(30–70°C). The remaining activities were then
assayed by the standard method. As shown in
Fig. 7d. FTS was stable below 50°C, while INV
retained over 80% of its original activity at tem-
perature as high as 65°C.

3.2.4. Effect of metals and other reagents

The effects of different metal ions and other
Fig. 7. Effect of pH and temperature on enzyme activity and reagents on the activity of FTS and INV were
stability. (a) pH – enzyme activity; (b) pH–enzyme stability; (c) examined by incubating the enzymes in the pres-
Temperature-enzyme activity; (d) Temperature–enzyme stabil-
ence of the reagents at 30°C for 1 h. The residual
ity.
, FTS; , INV.
activity was assayed by the standard method. As
shown in Table 2 HgCl2 and AgNO3 completely
abolished FTS while in the presence of the same
concentration of these inhibitors INV retained 72
Table 2
and 66% of its initial activity, respectively. The
Effect of metal ions and other chemicals on FTS and INV
activity invertase was slightly affected by FeSO4, SrCl2
and EDTA. Urea inhibited 30% of its original
Compound mM Relative activity (%) activity. These reagents did not show any in-
hibitory effect towards FTS. Aniline, potent in-
FTS INV
hibitor of yeast and Neurospora invertase (de la
Control (H2O) 100 100 Vega et al. 1991), had only a slight inhibitory
FeSO4 1 96 88 effect on A. niger invertase.
AlCl3 1 100 100
KCl 1 101 95 3.2.5. Determination of the kinetic parameters
MgSO4 1 99 94
The Km and Vmax values for each enzyme were
SrCl2 1 99 84
CuSO4 1 98 99 determined using the Lineweaver–Burk plots (not
SnCl2 1 99 100 shown). They were calculated to be 44.38 mM
BaCl2 1 101 99 and 1030 mmol ml − 1 min − 1 for FTS and 35.67
Pb(NO3)2 1 99 97 mM and 398 mmol ml − 1 min − 1 for INV
AgNO3 1 1 66
respectively.
MnCl2 1 99 93
CaCl2 1 100 92
CoCl2 1 92 95 3.2.6. Effect of sucrose concentration on FTS and
KI 0.4 100 90 INV acti6ity
HgCl2 1 0 72 FTS and INV catalytic activity was very much
SDS 10 42 59
dependent on sucrose concentration. FTS activity
EDTA 25 97 80
Urea 500 99 71 increased with increasing sucrose concentrations
Aniline 1.5 105 92 to reach a maximum activity at a sucrose concen-
tration of 50% (Fig. 8). At higher concentrations
82 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84

the rate of transfer reaction decreased slowly 4. Discussion

probably due to the decrease in water activity.
However FTS maintained 74.8% of its maximal
activity at concentrations as high as 80% (w/v).
On the contrary INV activity decreased greatly at
sucrose concentrations greater than 5% (w/v)
(Fig. 8). Furthermore, FTS exclusively catalyzed
fructosyltransfer reaction producing glucose, 1-
kestose, nystose and fructofuranosyl nystose. In
addition, at 50% sucrose concentration, FTS pro-
duced fructooligosaccharides at a yield of 62%
against 54% with the crude extract (Table 3). On
the other side INV exhibited only hydrolytic ac-
tivity producing equimolar amounts of fructose
and glucose, no tri- or higher oligosaccharides
were detected (Table 3).
The first important result of this investigation is
that two enzymes, fructosyltransferase and inver-
tase, were separated and isolated from A. niger
AS0023. Owing to the similarity of their behavior
it was necessary to use four consecutive chro-
matographic procedures to achieve the separation
of the two proteins; they could be purified to
78.5-and 97.8-fold, respectively. The fructosyl-
transferase catalyzed exclusively fructosyltransfer
reaction, while the invertase did not show any
transfer activity. Previous studies on FOS-produc-
ing enzyme from Aspergillus sp. reported that it
was a b-fructofuranosidase (invertase) with high
fructosyltransferase activity at high sucrose con-

Fig. 8. Effect of sucrose concentration on FTS (

) and INV () enzyme activity.

Table 3
Products of enzymatic reaction of A. niger crude extract and purified FTS and INV on 50% sucrose solution after 5 h reaction time

Carbohydrate (% of total sugars) Crude extract Purified FTS Purified INV

Fructose 4.3 NDa 32.1

Glucose 30.79 27.2 30.3
Sucrose 10.45 10.28 37.5
1-Kestose 30.05 35.54 ND
Nystose 22.89 24.13 ND
1F-Fructofuranosyl nystose 1.52 2.79 ND
Total FOS 54.46 62.46 –

a
ND, not detected.
 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84
centration (Hidaka et al., 1988; Fujita et al., 1990;
Chang et al., 1994; Chen and Liu, 1996); because
of the purification difficulty it was not possible to
separate the two enzymes. Consequently,
the two activities appeared to belong to one single
enzyme. Fructosyltransferase without hydro-
lytic activity has only been isolated from higher
plants (Henry and Darbyshire, 1980; Shiomi and
Izawa, 1980) and not from microbial sources. In
A. niger AS0023 fructosyltransferase and
invertase appear to be two different constitutive
enzymes. Owing to their specificity they can be
classified as fructosyltransferase (EC.2.4.1.9)
and b-fructofuranosidase (EC.3.2.1.26), respec-
tively.
On polyacrylamide gel electrophoresis, fructo-
syltransferase and invertase migrated as polydis-
perse aggregates yielding broad and diffused
bands. This result is typical of glycoproteins
and the two enzymes have proved their glyco-
protein nature by their adsorption on ConA-
affinity gel (Goldstein, 1976). These polydisperse
profiles are probably due to the presence of an
heterogeneous population of oligosaccharides. In
fact even if the subunits of glycoprotein are of the
same average size, they may contain varying
amounts of carbohydrates (Trimble and Maley,
1977). Thus the observed heterogeneity could
have resulted from a difference in the number
and/or length of the associated oligosaccharide
chains.
Although an accurate molecular size cannot be
determined on native gel, information on the mul-
timericity of the two enzymes could be provided.
Fructosyltransferase showed a higher molecular
weight with an approximate average subunit size
of 125 kDa. In native form this enzyme seems to
undergo association – dissociation; its subunits
tend to aggregate to form oligomers (hexamer,
tetramer, and dimer). In glycoproteins with high
carbohydrate content strong intermolecular asso-
ciation may occur as result of protein – carbohy-
drate or carbohydrate – carbohydrate interactions.
Thus, its polydispersity could conceivably arise
either from heterogeneity within its carbohydrate
moiety or from a different degree of aggregation
of its glycoprotein subunits. The electrophoretic
mobility suggested the native invertase to be a
83
dimer with an approximate average subunit size
of 91 kDa.
From the catalytic properties of A. niger
AS0023 purified fructosyltransferase and inver-
tase, we could see that parameters like tempera-
ture, pH, metal ions and other inhibitors had
different effects on their respective activities. Be-
sides, determination of the kinetic constants such
as Km and Vmax gave some light about the essen-
tial features of the affinity of the two enzymes
toward sucrose. Thus it may be postulated that,
although very similar in some properties, the two
enzymes possess different catalytic sites.
With regard to sucrose concentration it was
confirmed that at concentrations exceeding 5%,
INV activity deviated from Michaelis–Menten ki-
netics since the rate of hydrolysis gradually de-
creased. A very similar behavior has been
reported by Straathof et al. (1986) and Combes
and Monsan (1983). This effect has been at-
tributed to the decrease in the concentration of
water (Bowski et al., 1971), substrate inhibition
(Combes and Monsan, 1983) and substrate aggre-
gation (Bock and Lemieux, 1982). The invertase-
catalysed fructosyltransfer in concentrated sucrose
solutions has also been reported to account for
the decrease in hydrolytic activity. However, in
this study the purified invertase did not show any
transfer reaction even at high sucrose concentra-
tion. Similarly the fructosyltransferase did not
show hydrolytic action on sucrose at low concen-
trations. It has been often observed that b-fructo-
furanosidase commonly possess both
fructosyltransferase and hydrolyzing activity and
that the hydrolysis is a particular case of transfer
to water in dilute reaction systems (Straathof et
al., 1986; Hidaka et al., 1988; Fujita et al., 1990).
However, the use of purified enzymes presented
the evidence that the fructosyltransferase activity
(Ut) and the invertase (hydrolyzing) activity (Ui)
belong to two separate enzymes. From the results
obtained in the present study we have direct proof
for the different nature of A. niger fructosyltrans-
ferase and invertase. The purification of these two
enzymes offers a great potential for understanding
the functional role of each of the enzymes and a
better use of them.
84 L. L’Hocine et al. / Journal of Biotechnology 81 (2000) 73–84
References
Beeley, J.G., 1985. Glycoprotein and proteoglycan techniques.
In: Burdon, R.H., Van Khippenberg, P.H. (Eds.), Labora-
tory Techniques in Biochemistry and Molecular Biology,
vol. 16. Elsevier, Amsterdam, pp. 65–67.
Bock, K., Lemieux, R.U., 1982. The conformational proper-
ties of sucrose in aqueous solution: intramolecular hydro-
gen-bonding. Carbohydr. Res. 100, 63–75.
Bowski, R.S., Ryu, D.Y., Vieth, W.R., 1971. Kinetic modeling
of the hydrolysis of sucrose by invertase. Biotechnol. Bio-
eng. 13, 641 – 656.
Chang, C.T., Lin, Y.Y., Tang, M.S., Lin, C.F., 1994. Purifica-
tion and properties of b-fructofuranosidase from Asper-
gillus oryzae ATCC 76080. Biochem. Mol. Biol. Int. 32,
269 – 277.
Chen, W.C., Liu, C.H., 1996. Production of b-fructofuranosi-
dase by Aspergillus japonicus. Enz. Microbiol. Technol. 18,
153 – 160.
Chen, J.S., Saxton, J, Hemming, F.W., Peberdy, J.F., 1996.
Purification and partial characterization of the high and
low molecular weight (S- and F-form) of invertase secreted
by Aspergillus nidulans. Biochem. Biophys. Acta 1296,
207 – 218.
Chu, F.K., Watorek, W., Maley, F., 1983. Factors affecting
the oligomeric structure of yeast external invertase. Arch.
Biochem. Biophys. 223, 543–555.
Combes, D., Monsan, P., 1983. Sucrose hydrolysis by inver-
tase. Characterisation of products and substrate inhibition.
Carbohydr. Res. 117, 215–228.
Fujita, K., Hara, K., Hashimoto, H., Kitahata, S., 1990.
Transfructosylation catalysed by b-fructofuranosidase I
from Arthrobacter sp. K-1. Agric. Biol. Chem. 54, 2655–
2661.
Gabriel, O., Wang, S.F., 1969. Determination of enzymatic
activity in polyacrylamide gels I. Enzymes catalyzing the
conversion of non-reducing substrates to reducing prod-
ucts. Anal. Biochem. 27, 545–554.
Goldstein, J.I., 1976. Carbohydrate binding specificity of Con-
canavalin A. In: Bittiger, H., Schenebli, H.P. (Eds.). Wiley,
London, pp. 55 – 68.
Henry, R.J., Darbyshire, B., 1980. Sucrose fructosyltransferase
and fructan: fructan fructosyltransferase from Allium cepa.
Phytochemistry. 19, 1017–1020.
Hidaka, H., Hirayama, M., Sumi, N., 1988. A fruc-
tooligosacharides-producing enzyme b-fructofuranosidase
from Aspergillus niger ATCC 20611. Agric. Biol. Chem. 52,
1181 – 1187.
Jiang, B., Wang, Z., 1995. Properties of fructosyltransferase
from Aspergillus niger AS0023. J. Wuxi. Univ. Light. Ind.
14, 290 – 295.
Laemmli, U.K., 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 277,
680 – 685.
L’Hocine, L., Jiang, B., Wang, Z., 1997. Production of high-
content fructooligosaccharides by a multi-enzyme system.
Food. Sci. (China) 18, 24 – 27.
Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J.,
1951. Protein measurement with the Folin phenol reagent.
J. Biol. Chem. 193, 265 – 275.
Nishizawa, M., Maruyama, Y., Nakumara, M., 1980. Purifica-
tion and characterization of invertase isoenzymes from
Fusarium oxysporum. Agric. Biol. Chem. 44, 489 – 498.
Praznik, W., Beck, R.H.F., Spies, T., 1990. Isolation and
characterization of sucrose: sucrose 1F-b-D-fructosyltrans-
ferase from tubers of Helianthus tuberosis L. Agric. Biol.
Chem. 54, 2429 – 2431.
Shiomi, N., Izawa, M., 1980. Purification and characterization
of sucrose: sucrose fructosyltransferase from the roots of
asparagus (Asparagus officinalis L.). Agric. Biol. Chem. 44,
603 – 614.
Shiomi, N., 1982. Purification and characterisation of 1F-fruc-
tosyltransferase from the roots of asparagus (Asparagus
officinalis L.). Carbohydr. Res. 99, 157 – 169.
Smith, W.L., Ballou, C.E., 1974. The effect of dithiothreitol on
external yeast invertase. Biochem. Biophys. Res. Commun.
59, 314 – 321.
Straathof, A.J.J., Kieboom, A.P.G., Bekkum, H., 1986. Inver-
tase-catalysed fructosyl transfer in concentrated solutions
of sucrose. Carbohydr. Res. 146, 154 – 159.
Trimble, R.B., Maley, F., 1977. Subunit structure of external
invertase from Saccharomyces cere6iciae. J. Biol. Chem.
252, 4409 – 4412.
de la Vega, M.G., Cejudo, F.J., Paneque, A., 1991. Purifica-
tion and some properties of extracellular invertase from
Azotobacter chroococcum. Enzym. Microbiol. Technol. 13,
267 – 271.
Zarate, V., Belda, F., 1996. Characterization of the het-
erologous invertase produced by Schizosaccharomyces
pombe from SUC2 gene of Saccharomyces cere6iciae. J.
Appl. Bacteriol. 80, 45 – 52.