f o o d a n d b i o p r o d u c t s p r o c e s s i n g 8 6 ( 2 0 0 8 ) 109–115
avai lab le at www.sc iencedi rec t .com
journa l homepage: www.e lsev ier .com/ locate / fbp
ructooligosaccharides production by Aspergillusp. N74 in a mechanically agitatedirlift reactor
scar Sancheza,∗, Felipe Guioa, Diana Garciaa,delberto Silvab, Luis Caicedoa
Universidad Nacional de Colombia, Chemical Engineering Department, Cra 30 Cll 45, Bogota, ColombiaUniversidad Nacional de Colombia, Pharmacy Department, Cra 30 Cll 45, Bogota, Colombia
r t i c l e i n f o
rticle history:
eceived 10 October 2007
ccepted 12 February 2008
eywords:
ructosyltransferase
ructooligosaccharides
spergillus sp.
a b s t r a c t
Batch fructooligosaccharides (FOSs) production by fructosyltransferase from Aspergillus sp.
N74 in a mechanically agitated airlift reactor was investigated. Two biomass concentrations
(6 and 9.5 g L−1 in biomass dry weight) of Aspergillus sp. N74 were used for the evaluation
of fructosyltransferase activity. The enzymatic reactions were carried out at the following
contain glucose, fructose and sucrose in more than 500 g/kg
. Introduction
ructooligosaccharides (FOSs) are oligosaccharides of fruc-ose containing a single glucose moiety, they are producedy the action of fructosyltransferase (FTase, E.C. 2.4.1.9) fromany plants and microorganisms (Yun, 1996; Hidaka et al.,
988). FOS are mainly composed of 1-kestose (GF2), nystoseGF3), and 1-�-fructofuranosyl nystose (GF4), in which fruc-osyl units (F) are bound at the �(2 → 1) position of sucrose
olecule (GF) (Sangeetha et al., 2005b; Kaplan and Hutkins,000; Yun, 1996; Hidaka et al., 1988). FOS with low polymericrade has better therapeutic properties than those with aigh polymeric degree. FOS are about 0.4 and 0.6 times asweet as sucrose and have been used in the pharmaceuticalndustry as a functional sweetener (Sangeetha et al., 2005b;
iedrzycka and Bielecka, 2004; Heyer and Wendenduerg, 2001;un, 1996; Kuhbauch, 1972). FOS present properties such as
low caloric values, non-cariogenic properties, decrease lev-els of phospholipids, triglycerides and cholesterol, help gutabsorption of calcium and magnesium, are useful for diabeticproducts and are used as prebiotics to stimulate the bifidobac-teria growth in the human colon (Sangeetha et al., 2005b;Biedrzycka and Bielecka, 2004; Roberfroid and Delzenne, 1998;Yun, 1996; Crittenden and Playne, 1996; Yamashita et al.,1984).
FOS are industrially produced from sucrose by microbialenzymes with transfructosylating activity, mainly found infungi such as Aspergillus, Aureobasidum, Arthrobacter and Fusar-ium (Sangeetha et al., 2005a,b; Yun, 1996). The theoreticalyield of FOS from sucrose is 75% if 1-kestose is the onlyFOS produced (Yoshikawa et al., 2008). Commercial FOS may
of total FOS dry weight. Thus, the search of new potenttransfructosylating-enzyme producers with their best reac-
neers. Published by Elsevier B.V. All rights reserved.
[F] molar concentrations of fructose[GF2] molar concentrations of 1-kestose[GF3] molar concentrations of nystose[GF4] molar concentrations of 1-�-fructofuranosyl
nystose[S0] initial molar concentrations of sucroseti final timet0 initial timeUh hydrolytic activityUhE hydrolytic enzymatic productivity by reaction
volumeUhS hydrolytic specific activity by dry weight
biomassUtE transfructosylating productivity by reaction
volumeUt transfructosylating activityUtS transfructosylating specific activity by dry
weight biomassYFOS fructooligosaccharides total yieldYGF2 1-kestose yieldYGF3 nystose yield
YGF4 1-�-fructofuranosyl nystose yield
tion conditions is desirable in order to scale-up the process. Inthis study, the batch-FOS production from sucrose by wholecells of Aspergillus sp. N74 in a mechanically agitated airliftreactor was evaluated at bench scale.
2. Materials and methods
2.1. Chemicals
1-Kestose, nystose, and 1-�-fructofuranosyl nystose used forstandard in product analysis were purchased from Wako PureChemical Industries (Osaka, Japan). The sucrose was food-grade, while other chemicals were analytical grade.
2.2. Microorganism and spore production
The fungus Aspergillus sp. N74 was isolated from a sugar canecrop in La Pena (Colombia). In a previous study (Sanchez,2006), this strain showed a high transfructosylating activity
Fig. 1 – Upper (a) and lower (b) crossview and ax
s s i n g 8 6 ( 2 0 0 8 ) 109–115
and the best sugar-bioconversion was at pH 5.5, 60 ◦C and ini-tial sugar concentrations higher than 55% (w/v). The strainwas cultivated on malt extract agar (MEA) plates at 30 ± 1 ◦Cfor 7 days. To prepare spore suspensions, spores were scrapeddown from the MEA plates with a sterilized tensoactive solu-tion (15%, w/v glycerol, 0.1%, w/v Tween 80 and acetate buffer0.1 M (pH 6.0) q.s.f. 100 mL) and diluted to a concentrationof about 1 × 107 spores mL−1 with sterilized water. The sporesuspensions were kept at −20 ± 1 ◦C and subcultured once amonth.
2.3. Filter module
The filter module used in this study is shown in Fig. 1. Thefilter was made of stainless steel with a mean pore size of20 �m. It has an irregular geometry in order to increase thefiltration area per working volume (4.5 L). The total filtrationsurface was 1620 cm2 for an area–operation volume ratio of360 cm2 L−1. The airlift draft–tube was made up the filtrationmodule.
The filter module was submerged and linked both forextraction of culture broth and for gas sparging. The outletfiltrate was connected to a peristaltic pump (Cole-Parmer-77200) and a three-way valve that allow the gas sparger tobe connected to the air supply. The alternating linkage wasperiodically switched by manual change of the valve line, inorder to increase the oxygen transfer and decrease the mem-brane fouling. Transmembrane pressure (TMP) was monitoredthrough a U-type mercury manometer as an indicator of mem-brane fouling.
2.4. Reactor description
The mechanically agitated membrane draft–tube airlift reac-tor scheme is shown in Fig. 2. Agitation and aeration systemswere located at the centerline of the membrane–draft–tube.Agitation was made with two Rushton turbines and aerationwas made through a perforated pipe sparger, the vertical dis-tance between the lower turbine and the sparger was 3.5 cm(Cts). The 6-bladed Rushton turbines were 7.5 cm in diameter(di) and every blade was 1.5 cm high (w), for a w/di rela-tion of 1/5. The vertical distance between the turbines was
5.0 cm (sc) and the lower turbine was located 8.5 cm from thevessel bottom (C), for a sc/di relation of 2/3. The bioreactorvessel was 16 cm in internal diameter (D) and 32 cm in over-
ial view (c) of the design filtration module.
f o o d a n d b i o p r o d u c t s p r o c e s s
Fig. 2 – Reactor dimensions (in cm). Nomenclature: impellerblade height (w), impeller diameter (di), bioreactor vesseldiameter (D), bioreactor vessel height (H), impellerclearance (C), initial static liquid height (hL), distancebetween turbines (Sc), clearance between the uppermembrane–draft–tube shape and the liquid surface (Cu),membrane–draft–tube length (hm) andmembrane–draft–tube clearance (Cl).
a1v0vhm(
2
P50Npu3ct3tfipu5cdb
biomass dry weight (Fernandez et al., 2004; Hidaka et al., 1988;
ll height (H) with round bottom. The membrane–draft–tube,5 cm length (hm), was located 5.0 cm above the bottom of theessel (Cl) and the raiser and downcomer relation area was.83 (Ar/Ad). The working volume and the overall bioreactorolume were 4.5 and 7 L, respectively. The initial static liquideight was 22.0 cm (hL) and the clearance between the upperembrane–draft–tube shape and the liquid surface was 2.2 cm
Cu).
.5. Biomass production
re-inoculum was made in a 1 L Erlenmeyer flask with00 mL of culture medium (11% sucrose, 0.84% K2HPO4,.102% MgSO4·7H2O, 0.088% KCl, 0.007% FeSO4·7H2O, 0.085%aNO3·4H2O, 2.0% yeast extract, 0.136% CaCO3, adjusted toH 5.5 ± 0.1 with HNO3 (Hernandez and Mendoza, 2004)) inoc-lated with 600 �L of 1 × 107 spores mL−1 shaken for 12 h at0 ± 1 ◦C and 250 rpm (New Brunswick C76). The inoculatedulture medium (pH 5.5 and total volume 4.5 L) was cul-ured for 24 or 48 h, under the following conditions: 30 ◦C,00 rpm and superficial aeration rate (UGr), 0.008 m s−1. Athe end of the growing time (24 or 48 h), the membrane-ltered culture medium was extracted with a peristalticump (Cole-Parmer-77200) connected to the membrane mod-le and the held biomass was washed several times with0 mM phosphate buffer, pH 5.5, in order to remove theulture medium remnant. The biomass concentration was
etermined at the end of the enzymatic reaction, aftereing washed with 50 mM phosphate buffer, pH 5.5 and
i n g 8 6 ( 2 0 0 8 ) 109–115 111
dried for 48 h at 105 ◦C (Dorta et al., 2006; Cruz et al.,1998).
To study the effect of the membrane filtration module overthe biomass production, fermentations without the mem-brane module were carried out under culture conditions,medium and reactor, described above.
2.5.1. Morphological characterizationDuring the culture, samples were taken for the fungal pelletmorphology. It was characterized using image analysis (Casaset al., 2005; Paul and Thomas, 1998). Prior to imaging, eachsample was filtered and washed twice with 10 mL of distilledwater. For each sample, 50 pellets were analyzed and the totalpellet core diameter, as the one-dimensional projected area,was measured. The image was captured in an inverted micro-scope (Leica DMIL) with a CMOS camera (Evolution LC Color;Media Cybernetics Inc., Silver Spring, MD, USA). All sampleswere analyzed at 40× magnification.
2.6. FTase assay and FOS production
The biomass used in the enzymatic reactions was obtainedseparately from two selected cultured times (24 or 48 h). Thereaction volume and conditions were 4.5 L, initial sucrose con-centration 70% (w/v), pH 5.5, 60 ◦C, 350 rpm, and superficialaeration rate 0.012 m s−1. The reactor was operated in batchfor 26 h; samples of the culture were taken for the analysis ofenzymatic activity and carbohydrates.
The total yield of fructooligosaccharides (YFOS) was calcu-lated from the yield of 1-kestose (YGF2), nystose (YGF3) and1-�-fructofuranosyl nystose (YGF4) (Eqs. (1)–(4)). The selectiv-ity of conversion from sucrose to FOS was calculated by Eq. (5)(Madlova et al., 1999):
where [GF2], [GF3], [GF4], [S0] and [F] are the molar concentra-tions of 1-kestose, nystose, 1-�-fructofuranosyl nystose, initialsucrose and fructose, respectively.
The transfructosylating activity and the hydrolytic activ-ity were determined by measuring both the glucose (G) andfructose (F) present in the reaction mixture. One unit of trans-fructosylating activity was defined as the amount of enzymerequired to transfer 1 �mol of fructose min−1. One unit ofhydrolytic activity was defined as the amount of enzymerequired to release 1 �mol of free fructose min−1.
The enzymatic productivity was calculated as the trans-fructosylating (UtE) or hydrolytic (UhE) activity per reactionvolume, while the specific activity was calculated as the trans-fructosylating (UtS) or hydrolytic (UhS) activity per unit of
Nguyen et al., 1999). Transfructosylating (Ut) and hydrolytic(Uh) activities were calculated at a time interval by Eqs. (6) and
r o c e
�mo
1-�-fructofuranosyl nystose production was observed (Fig. 6a).The highest sucrose bioconversion was observed at 24 h of
Specific activity and volumetric productivity of the enzymewere evaluated through Eqs. (8) and (9) and Eqs. (10) and (11),respectively
UtS = Ut
mg dried biomass(8)
UhS = Uh
mg dried biomass(9)
UtE = Ut
reaction volume(10)
UhE = Uh
reaction volume(11)
2.7. Analysis of sugars
The analysis of sugars was performed by high-performanceliquid chromatography (HPLC). The HPLC equipment con-sisted of a pump Waters 515 with an on line degasser, arefractive index (RI) detector Waters 410 and an injection valvewith a 20 �L loop.
A Sugar-PakTM (Waters) column was used for sucrose,glucose and fructose identification and quantification. Chro-matographic conditions were column temperature, 84 ◦C;mobile phase, water at a flow rate of 0.4 cm3 min−1 and RIdetector temperature, 40 ◦C (Sanchez, 2006).
A Shodex® column was used for 1-kestose, nystose and1-�-fructofuranosyl nystose identification and quantification.Chromatographic conditions were column temperature, 65 ◦C;mobile phase, water–acetonitrile (72:28) at a flow rate of1.0 cm3 min−1 and RI detector temperature, 45 ◦C (Sanchez,2006).
3. Results and discussion
3.1. Biomass production and enzyme activity
Dissolved oxygen (DO), pH, dried biomass concentration andapparent viscosity during the 48 h microorganism culture inthe reactor, with and without the filter module, are shownin Fig. 3. DO and biomass profiles were comparable withreported profiles for aerobic fermentations (Casas et al., 2005),with an oxygen tension drop while the biomass concentrationincreases. The biomass produced with the membrane filtra-tion module after 24 and 48 h of culture was 6 and 9.5 g L−1 (indry weight), respectively, which is 45% higher than obtainedwithout the membrane filtration module. This increase in thebiomass weight is possibly due to the enhancement of theoxygen transfer by a better air distribution obtained with thesubmerged filter module (Chang et al., 1994; Suzuki et al.,1994). Likewise, apparent viscosity was correlated with thebiomass concentration. Similar results were reported for a
submerged culture of filamentous fungi (Aspergillus terreus) byRodrıguez-Porcel et al. (2005) and Casas et al. (2005). The pHprofile was in the range of 5.50 ± 0.06 during the culture.
s s i n g 8 6 ( 2 0 0 8 ) 109–115
l Fructose|t0 )(6)
It was noticed that for the culture conditions (stirring rate300 rpm and superficial gas velocity 0.008 m s−1), the pellet sizeafter the first six growing hours had a mean core diameter of2200 ± 100 �m (Fig. 4). Similar pellet sizes were obtained byCasas et al. (2005) for A. terreus cultures at the same stirringrate.
As shown in Fig. 5, the specific transfructosylating activityreached a plateau after a 24 h of culture, although the trans-fructosylating volumetric productivity did not show a peak.Also, a maximum in the hydrolytic activity was detected after12 h of culture. This phenomenon may be associated withthe fact that before 12 h there is not a high FTase produc-tion, since there is a low biomass concentration and invertaseenzyme production is enhanced under lower biomass andFOS concentration conditions (Fernandez et al., 2004; Yun,1996).
3.2. FOS production
The bioconversion of sucrose and FOS production showed adependence on the reaction time and biomass concentration(Fig. 6). In fact, 1-kestose and nystose were the FOS obtainedfor the 6 g L−1 biomass concentration, and no evidence of the
Fig. 3 – pH (�), DO (�), apparent viscosity (�) and biomassdry weight (�) profiles during the 48 h fungi culture. Withthe filter module (a) and without it (b).
f o o d a n d b i o p r o d u c t s p r o c e s s i n g 8 6 ( 2 0 0 8 ) 109–115 113
Fig. 4 – Pellet formation sequence at the first 6 h of culture (a.1–a.6). Pellet core size follow-up during the culture time: 6 h(a-6), 18 h (b), 30 h (c) and 48 h (d) (40× magnification).
rona3tfAa
FUU
eaction (86.5%, w/w) with a sucrose and fructose remnantf 13.5 and 5.5% (w/w), respectively. After 18 h of reaction,o significant variations were observed in FOS profiles, withproduction of 1-kestose and nystose on a sucrose basis of
6 ± 2 and 14 ± 3% (w/w), respectively, for a total FOS produc-ion of 50 ± 5%. Transformation from sucrose to 1-kestose androm 1-kestose to nystose showed expected profiles (Fig. 6a).
fter 26 h of culture FOS yield (YFOS) was 69% (43% 1-kestosend 26% nystose).
ig. 5 – Enzymatic activity profiles. Specific activity (UtS (�),
hS (�)) and volumetric productivity of the enzyme (UtE (�),
hE (�)) obtain during the fungi culture.
When the enzyme reaction was carried out with the9.5 g L−1 biomass concentration, 94% of the initial sucroseconcentration was biotransformed in the first 4 h of reac-tion (Fig. 6b), with a sucrose and fructose remnant of 6%and 10%, respectively. However, after 26 h of reaction anincrement in the remnant of sucrose (8.2%) and fructose(10.1%) was observed. This behaviour could be associated withthe Ping–Pong mechanism that has been reported for fruc-tosyltransferase, when sucrose can be produced from FOS(Crittenden and Playne, 1996; Yun, 1996). Biotransformation ofnystose from 1-kestose was higher in the first 2 h of reaction.Additional to the FOS produced in the 6 g L−1 biomass con-centration, 1-�-fructofuranosyl nystose was observed when9.5 g L−1 biomass concentration was used, but it showed alow and constant biotransformation rate from nystose. After4 h of reaction a peak in the YFOS of 70% was observed (43%1-kestose, 25% nystose and 2% 1-�-fructorianosyl nystose),which was reduced to the 57% (18% 1-kestose, 33% nystoseand 6% 1-�-fructofuranosyl nystose) at the end of the reactiontime. This FOS yield reduction could be due to the presenceof a high concentration of glucose in the medium that inac-tivated the FTase and increased the FOS hydrolysis, leadingto an increase in the free fructose and sucrose concentra-tion in the medium (Hirayama et al., 1989; Song and Jacques,1999). It is noticed, that at 9.5 g L−1 biomass concentration is
important to control the reaction time, in order to get a con-siderable sucrose bioconversion to FOS and to handle theircomposition.
Few studies at bench scale have been reported. Yun et al.(1990) evaluated the semibatch production of FOS in a 1 Lreactor using Aureobasidium pullulans. Results showed a FOSconversion of 57% and 55% for the free and immobilized cells,respectively, in a reaction solution of 77% (w/w) sucrose, pH 5.5and at 55 ◦C. Sangeetha et al. (2005b) scaled up the FOS pro-duction using a 10 L reactor and Aspergillus oryzae CFR 202 ina reaction solution of sucrose 60% (w/w), pH 5.5 and at 55 ◦C,reaching a FOS yield of 52% after 18 h of culture.
The results suggest that the used of fructosyltransferasefrom Aspergillus sp. N74 in the designed reactor is an alterna-tive to study the FOS production in a large scale or even in anindustrial scale.
4. Conclusions
The used of a submerged membrane airlift reactor agitatedmechanically, allowed high cell density cultures of the nativestrain Aspergillus sp. N74, which were 45% higher than thebiomass reached without the filtration module.
The FTase specific transfructosylating activity showed adependence on the biomass concentration and the reactiontime, with a plateau after 24 h of culture. Although the volu-metric productivity also showed a dependence on the biomassconcentration, a peak was not observed.
The composition of the produced FOS was related with thebiomass concentration and the reaction time. Therefore, for
s s i n g 8 6 ( 2 0 0 8 ) 109–115
a biomass concentration of 9.5 g L−1 1-�-fructofuranosyl nys-tose was synthesized, while for the biomass concentration of6.0 g L−1 it was not.
The best reaction time for the batch FOS production witha biomass concentration of 6.0 and 9.5 g L−1 was 24 and 4 h,respectively. For each case the FOS yield was 69% and 70%,respectively.
The designed reactor and fructosyltransferase from thenative strain Aspergillus sp. N74 can be considered as an indus-trial alternative for the fructooligosaccharides production.
e f e r e n c e s
Biedrzycka, E. and Bielecka, M., 2004, Prebiotic effectiveness offructans of different degrees of polymerization. Trends FoodSci Technol, 15: 170–175.
Casas, J.L., Sanchez, J.A., Fernandez, J.M., Rodrıguez, E.M. andChisti, Y., 2005, Pellet morphology, culture rheology andlovastatin production in cultures of Aspergillus terreus. JBiotechnol, 116: 61–77.
Chang, H., Yoo, I. and Kim, B., 1994, High density cell culture bymembrane-based cell recycle. Biotechnol Adv, 12: 467–487.
Crittenden, R.G. and Playne, M.J., 1996, Production, properties andapplications of food-grade oligosaccharides. Trends Food SciTechnol, 7: 353–361.
Cruz, R., Cruz, V.D., Belini, M.Z., Belote, J.G. and Vieira, C.R., 1998,Production of fructooligosaccharides by the mycelia ofAspergillus japonicus immobilized in calcium alginate.Bioresour Technol, 65: 139–143.
Dorta, C., Cruz, R., Oliva-Neto, P. and Camargo, D.J., 2006,Sugarcane molasses and yeast powder used in thefructooligosaccharides production by Aspergillus japonicus-FCL119T and Aspergillus niger ATCC 20611. J Ind MicrobiolBiotechnol, 33: 1003–1009.
Fernandez, R.C., Guilarte, B., Juarez, A. and Martinez, J., 2004,Production of fructooligosaccharides by �-fructofuranosidasefrom Aspergillus sp. 27H. J Chem Technol Biotechnol, 79:268–272.
Hernandez, A., Mendoza, C., 2004. Design of a Culture Mediumfor Fructosyltranferase Production by Aspergillus sp.,Pharmacy Thesis, Universidad Nacional de Colombia, SedeBogota (in Spanish).
Heyer, A.G. and Wendenduerg, R., 2001, Gene cloning andfunctional characterization by heterologous expression of thefructosyltranferase of Aspergillus sydowy IAM 2544. ApplEnviron Microbiol, 67: 363–370.
Hidaka, H., Hirayama, M. and Sumi, N., 1988, Afructooligosaccharide-producing enzyme from Aspergillusniger ATCC 20611. Agric Biol Chem, 52: 1181–1187.
Hirayama, M., Sumi, N. and Hidaka, H., 1989, Purification andproperties of fructo-oligosaccharides producingbeta-fructofuranosidase from Aspergillus niger ATCC 20611.Agric Biol Chem, 53: 667–673.
Kaplan, H. and Hutkins, R.W., 2000, Fermentation offructooligosaccharides by lactic acid bacteria andbifidobacteria. Appl Environ Microbiol, 66: 2682–2684.
Kuhbauch, W., 1972, Changes of the contents in glucose, fructose,sucrose and fructosans, as well as in the degree ofpolymerization of fructosan’s molecules, in small leaves andstems of Orchard Grass during their growth. LandwirtschForsch, 26(2): 173–181 (in German)
Madlova, A., Antosova, M., Barathova, M., Polakovie, M., Stefuca,V. and Bales, V., 1999, Screening of microorganisms fortransfructosylating activity and optimization ofbiotransformation of sucrose to fructooligosaccharides. ChemPapers, 53(6): 366–369.
Nguyen, Q.D., Mattes, F., Hoschke, A., Rezessy-Szabo, J. and Bhat,
M.K., 1999, Production, purification and identification offructooligosaccharides produced by �-fructofuranosidasefrom Aspergillus niger IMI 303386. Biotechnol Lett, 21: 183–186.
c e s s
P
R
R
S
S
S Biotechnol, 24/25: 299–308.
f o o d a n d b i o p r o d u c t s p r o
aul, G.C. and Thomas, C.R., 1998, Characterisation of mycelialmorphology using image analysis. Adv Biochem Eng, 60: 1–59.
odrıguez-Porcel, E.M., Casas-Lopez, J.L., Sanchez-Perez, J.A.,Fernandez-Sevilla, J.M. and Chisti, Y., 2005, Effects of pelletmorphology on broth rheology in fermentations of Aspergillusterreus. Biochem Eng J, 26: 139–144.
anchez, O.F., 2006. Study of Fructosyltransferase Production byan Aspergilllus sp. Native Strain from Sucrose in a Bench-scaleMembrane Reactor. M.Sc. Chemical Engineering Thesis,Universidad Nacional de Colombia, Sede Bogota (in Spanish).
angeetha, P.T., Ramesh, M.N. and Prapulla, S.G., 2005,Fructooligosaccharide production using fructosyl transferaseobtained from recycling culture of Aspergillus oryzae CFR 202.Process Biochem, 40: 1085–1088.
angeetha, P.T., Ramesh, M.N. and Prapulla, S.G., 2005,
Maximization of fructooligosaccharide production by twostage continuous process and its scale up. J Food Eng, 68:57–64.
i n g 8 6 ( 2 0 0 8 ) 109–115 115
Song, D.D. and Jacques, N.A., 1999, Purification and enzymicproperties of the fructosyltransferase of Streptococcus salivariusATCC 25975. Biochem J, 341: 285–291.
Suzuki, T., Sato, T. and Kominami, M., 1994, A dense cell retentionculture system using a stirred ceramic membrane reactor.Biotechnol Bioeng, 44: 1186–1192.
Yamashita, K., Kawai, K. and Itakura, M., 1984, Effects offructooligosaccharides on blood glucose and serum lipids indiabetic subjects. Nutr Res, 4: 961–966.
Yoshikawa, J., Amachi, S., Shinoyama, H. and Fujii, T., 2008,Production of fructooligosaccharides by crude enzymepreparations of b-fructofuranosidase from Aureobasidiumpullulans. Biotechnol Lett, 30: 535–539.
Yun, J.W., Jung, K.H., Oh, J.W. and Lee, J.H., 1990, Semibatchproduction of fructo-oligosaccharides from sucrose byimmobilized cells of Aureobasidium pullulans. Appl Biochem
