Available online at www.sciencedirect.

com

Bioresource Technology 99 (2008) 24662470

Microwave-assisted hydrolysis of Zymomonas mobilis levan envisaging

oligofructan production
Valdemir Cordeiro de Paula a, Irapuan Oliveira Pinheiro b, Carlos Edison Lopes a,
Glcia Maria Torres Calazans a,*
a

Department of Antibiotics, Federal University of Pernambuco, 50.670-901 Recife PE, Brazil

b
Institute of Biological Sciences, University of Pernambuco, 50.100-130 Recife PE, Brazil
Received 4 November 2005; received in revised form 30 April 2007; accepted 30 April 2007
Available online 2 July 2007

Abstract
Levan, a polysaccharide from Zymomonas mobilis, was hydrolyzed to obtain oligofructans or fructooligosaccharides with a degree of
polymerization varying from 4 to 14. Fructooligosaccharides (FOS) are short chain fructans that benecially aects the host by selective
stimulation of growth and activity of one or a number of bacteria including probiotic bacteria in the colon. The hydrolysis was performed in a microwave oven to shorten the reaction time. The experiments showed that it is possible to maximize selected oligomers
by interrupting the hydrolysis at the due time. The results allow one to infer that the procedure may also be useful for production of
oligomers from other polysaccharides.
 2007 Elsevier Ltd. All rights reserved.
Keywords: Fructooligosaccharides; Hydrolysis; Levan; Microwave; Prebiotics

1. Introduction
Polysaccharides are macromolecules composed of hundreds of monosaccharide units. If the molecular chain contains less than twenty monosaccharides it may be
considered an oligosaccharide (Kennedy et al., 1989). Both
polysaccharides and oligosaccharides are readily hydrolyzed in acid solutions yielding monosaccharides. However,
oligosaccharides may be obtained from polysaccharides by
stopping the hydrolysis so as to leave the desired oligomer
maximized for subsequent isolation (Li, 1998). This isolation may be performed at an industrial level by continuous
chromatography.
Fructooligosaccharides (FOS) are short chain fructans
having prebiotic potential. Fructans are oligomers or polymers of fructose. A prebiotic is a nondigestible food ingre-

Corresponding author. Tel.: +55 81 2126 8347; fax: +55 81 2126 8346.
E-mail address: calazans@ufpe.br (G.M.T. Calazans).

0960-8524/$ - see front matter  2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2007.04.062

dient that benecially aects the host by selective

stimulation of growth and activity of one or a number of
bacteria including probiotic bacteria in the colon. Probiotic
is a food or supplement, containing viable microorganisms
that are present in sucient numbers to actively enhance
consumers health by improving the balance of microora
in the gastrointestinal tract. Probiotic bacteria are mostly
represented by genera Lactobacillus and Bidobacterium
and specic prebiotic substances, such as FOS, that stimulate their development. Lactobacilli and bidobacteria do
not need specic prebiotic substances for growth. However,
in the complex ecosystem of the intestinal tract, prebiotics
tend to selectively enrich these groups. Many fruit and vegetables contain prebiotic oligosaccharides such as FOS but
the levels are too low to have any signicant eects. FOS
can be commercially produced by hydrolysis of polysaccharides or by enzymatic generation. A number of benets
are attributed to prebiotic intake: protection against colon
cancer, resistance to pathogens by increasing bidobacteria
and lactobacilli, better calcium absorption, reduction of

V.C. de Paula et al. / Bioresource Technology 99 (2008) 24662470

cholesterol and triglycerides by lactic acid bacteria and

immunological eect (Bekers et al., 2002; Manning and
Gibson, 2004).
The main interest of food industries in FOS is due to
their functional properties. FOS have a number of interesting properties. FOS are free of calories, scarcely hydrolyzed
by digestive enzymes and they are not utilized as an energy
source in the body, thus they are safe for diabetics; they are
noncariogenic; they encourage the growth of bidobacteria, they discourage the growth of potentially putrefactive
microorganisms and they decrease the level of serum cholesterol, phospholipids, and triglyceride (Yun, 1996).
As is often the case with carbohydrate mixtures, it is also
very dicult to separate the FOS components from each
other (Yun, 1996). Therefore, it is important to develop
an economical method for the production of high-content
FOS.
Our interest in fructooligosaccharides mainly arises
from their possible use in pharmaceutical applications as
an antitumourous means since experiments have pointed
towards low molecular weight levans as having higher
antitumourous activities (Calazans et al., 1997, 2000).
The levan molecular weight distribution curve shows that
fractions with low molecular weights are produced in small
quantities during the fermentation process. Therefore, it is
necessary to develop an ecient way to increase the
amount of FOS produced.
The FOS utilized in this paper was produced by Z. mobilis, a Gram-negative, facultative anaerobic bacterium that
ferments glucose, fructose, or sucrose as carbon sources
(Viikari, 1988). These carbohydrates are metabolized via
the same biochemical route, the Entner-Doudoro pathway. Z. mobilis are rods 26 lm in length and 11.5 lm
in width, agellated but lack spores or capsules.
Recently the microwave has become a useful tool for use
in organic and inorganic synthesis, in analytical chemical
laboratories and in many chemical manipulations during
the process of standardizing procedures (Corsaro et al.,
2004; Das, 2004; Lindstrom et al., 2001; Joergensen and
Thestrup, 1995; Morales-Rubio et al., 1993; Singh et al.,
2003). The main advantage of applying microwave
approaches is the time saving it achieves. The heating eect
is due mainly to dielectric polarization.
In this paper a microwave oven was used to accelerate
the levan acid hydrolysis envisaging the production of oligofructans for antitumour uses. The experiments were
directed to measuring the inuence of hydrolysis time on
levan hydrolysis at specic conditions of pH and microwave oven potency.
2. Methods
2.1. Levan
Levan (Mw = 769,500 and Mn = 50,100) produced by
Z. mobilis ZAG-12 in sucrose medium was used in all
experiments. The purity of levan was approximately 100%.

2467

2.2. Standards
As qualitative standard oligomer a light corn syrup
ALO 3038 and the standard fructose n 374 both
acquired from Phenomenex (Torrance, CA/USA) were
used.
2.3. Hydrolysis
A 20 g/L levan solution was prepared and the pH was
adjusted to 2.5 using 0.1 M hydrochloric acid. The hydrolysis was performed in a microwave oven working at
650 Watts that correspond to 60% of the nominal potency.
The hydrolysis was performed in an open 250 mL beaker at
atmospheric pressure. The working volume was 30 mL of
levan solution. The initial temperature was around 30 C.
The beaker was placed into another 600 mL beaker with
200 mL of water to avoid temperatures above 100 C for
safety reasons. The maximum temperature in the hydrolysis was 87 C. The evolution of temperature during the
hydrolysis was not measured. The samples were cooled
and neutralized using diluted NaOH (2.5 M, approximately 50 lL) solution to interrupt the reaction at the
due time. To eliminate chloride content of the solution
the levan was resuspended in distilled water and precipitated twice by ethanol addition. The samples were obtained
by evaporation of 200 mL of hydroalcoholic solution followed by recuperation in 4 mL of water. Due to these procedures the concentrations were multiplied by 2.5.
2.4. Oligomer analysis
The separation technique for oligomer analysis was size
exclusion (or gel fltration) chromatography (SEC). The
hydrolized samples were analyzed by HPLC (Agilent
1100 series) using refractive index detector, the mobile
phase was deionised water and the column used was Phenomenex Rezex-RS0 Oligosaccharide (12 lm 10 mm
200 mm, USA).
2.5. Fructose analysis
The fructose was analyzed using deionised water as the
mobile phase and in the same equipment the column used
was Beckman U-Spherogel Carbohydrate (10 lm
6.5 mm 300 mm, USA). The oven temperature was
80 C in both cases.
3. Results and discussion
The chromatogram of the standard mixture revealed various peaks associated to dierent degrees of polymerization
(DP). These peaks correspond to the monosaccharide and
oligosaccharides present in the standard mixture with DP
varying from 4 to 14. The chromatographic retention times
for each oligomer varied from 17 to 45 min. In Fig. 1
the chromatograms of the standard oligomer (light corn

2468

V.C. de Paula et al. / Bioresource Technology 99 (2008) 24662470

Fig. 1. Chromatograms of the oligomers standard (light corn syrup ALO 3038) and of a sample of hydrolyzed levan solution.

syrup ALO 3038) and a sample of hydrolyzed levan

solution just for comparison are presented. The number
of dierent oligomers detected by HPLC varied with the
hydrolysis time (Table 1). The peak with the retention time
of 15 min, the chromatogram shown below in Fig. 1, is a
recorded chromatographic signal related to the undegraded

Table 1
Relation between hydrolysis time and number of dierent oligomers
obtained and optimum times for oligomers production
Hydrolysis time
(min)

Number of dierent
oligomers

Degree of polymerization
maximized

0.5
1
1.5
2.0
2.5

0
11
10
8
8

2.8
3.0
3.5
4
5

Not determined
8
2
3
3

None
DP = 14
DP = 13 and DP = 12
None
DP = 11; DP = 10 and
DP = 9
DP = 8; DP = 7 and DP = 6
DP = 5 and DP = 4
None
None
DP = 1 (fructose)

and non-precipitated levan. The residual levan concentration in the samples has always been below 1 g/L.
The complete levan hydrolysis in a water bath at 100 C
takes approximately 30 min. In this condition the temperature of the levan solution is around 100 C too. If performed in a microwave oven the hydrolysis may be done
in approximately 5 min with the levan solution temperature
below 80 C. After that, the reacting mixture turns yellowish indicating molecular degradation, therefore 5 min was
considered as the upper limit for all experiments.
The fructose evolution as function of the hydrolysis time
is shown in Fig. 2. As expected the amount of fructose
increased continuously up to 4.5 min when the net fructose
production zeroed which was probably due to fructose degradation. The total amount of fructose liberated after a
hydrolysis time of 4.5 min (Fig. 2) was 73% (0.44 g) in comparison with the added amount of levan, taking into
account the increase in the molecular mass due to the
hydrolysis reaction.
The relative detection sensitivities achieved by the RI
detector for the various oligofructans were estimated by
ChemWindow software.
The kinetic behaviour of the oligomer containing 14
fructose units which is the highest oligomer size considered

V.C. de Paula et al. / Bioresource Technology 99 (2008) 24662470

2469

Fig. 2. Relation between areas of fructose versus the hydrolysis time.

Fig. 5. Relation between area of the oligomer versus hydrolysis time
(DP = 9, 10, 11).

here, is shown in Fig. 3. Its initial concentration was

assumed to be zero, so the concentration increases until
reaching a maximum at around 1 min. It is possible that
the real maximum had occurred before that time but to verify this the hydrolysis sampling interval should be shortened. The graph also reveals the complete disappearance
of this oligomer at around 1.5 min reaction time.
The both oligomer with DP = 13 and DP = 12 are
shown in Fig. 4. They reached the highest value at around
1.5 min. The graph shows that these oligomers disappeared
when the hydrolysis time reached 2 min.
The oligomers having DP 9, 10 and 11 reached a maximum at around 2.5 min decreasing to zero at 3.5 min as
shown in Fig. 5.

Fig. 3. Relation between area of the oligomer versus hydrolysis time

(DP = 14).

Fig. 6. Relation between area of the oligomer versus hydrolysis time

(DP = 6, 7, 8).

Fig. 7. Relation between area of the oligomer versus hydrolysis time

(DP = 4, 5).

The peak for the oligomers having DP = 6, 7 and 8

occurred when the hydrolysis time was 2.8 min as shown
in Fig. 6.
In Fig. 7 the peaks of the oligomers with DP = 5 and 4
are shown that occurred when the hydrolysis time was
3 min. Table 1 presents the hydrolysis time in which the
maximization of the selected oligomer occurs.
4. Conclusions
Fig. 4. Relation between area of the oligomer versus hydrolysis time
(DP = 12, 13).

The use of microwave oven in the levan hydrolysis is viable and leads to a considerable reduction in the hydrolysis

2470

V.C. de Paula et al. / Bioresource Technology 99 (2008) 24662470

time. It was observed that the reacting mixture turns

yellowish indicating fructose decomposition when the
hydrolysis time exceeds 5 min. The experiments showed
that it is possible to obtain selected oligomers (DP from
4 to 14) by interrupting the hydrolysis at the correct time.
The optimum times for each oligomer production were
found. The results allow one to infer that the procedure
may be useful for oligosaccharide production from
any polysaccharides. The 30 ml of levan solution (20 g/L)
was partially converted to oligosaccharides and the
yields were 1.8; 2.4; 10.5; 20.5; 25.4; 5.4; 5.8; 2.0% for 1;
1.5; 2.0; 2.5; 3.0; 3.5; 4.0; 5.0 min of hydrolysis time,
respectively.
These preliminary results show that the use of microwaves can be a promising technique for industrial FOS
production leading to high purity product. Using polyfructan as raw material this technique should lead to a nal
product with no glucose and sucrose. Currently these sugars are present in high concentration in the commercial
FOS limiting the use of their favoured properties. Microwave is able to reduce the reaction times from hours to
minutes. Further studies should be performed to exactly
locate the optimum yield times for each oligomer. For
the purpose of better isolation it appears to be worthwhile
to decrease the hydrolysis rate.
It would also be worthwhile to perform experiments to
observe the eects that pH change and microwave potency
have on hydrolysis performance.
It may be possible that the isolation of the oligomers
would be uninteresting for antitumour purposes and that
it would be advantageous to mix them.
Acknowledgements
The authors thank CNPq, CAPES and FACEPE
for nancial assistance and Dr. Alexandre Jose da Silva
Goes for his assistance in the determination of relative

detection sensitivities achieved by the RI detector for

oligofructans.
References
Bekers, M., Laukevics, J., Upite, D., Kaminska, E., Vigiants, A., Viesturs,
U., Pankova, L., Danilevics, A., 2002. Fructooligosaccharide and
levan producing activity of Zymomonas mobilis extracellular levansucrase. Process Biochem. 38 (5), 701706.
Calazans, G.M.T., Lopes, C.E., Lima, R.O.C., Franca, F.P., 1997.
Antitumour activities of levans produced by Zymomonas mobilis
strains. Biotechnol. Lett. 19 (1), 1921.
Calazans, G.M.T., Lima, R.C., de Franca, F.P., Lopes, C.E., 2000.
Molecular weight and antitumour activity of Zymomonas mobilis
levans. Int. J. Biol. Macromol. 27, 245247.
Corsaro, A., Chiacchio, U., Pistara, V., Romeo, G., 2004. Microwaveassisted chemistry of carbohydrates. Curr. Org. Chem. 8 (6), 511538.
Das, S.K., 2004. Application of microwave irradiation in the synthesis of
carbohydrates. Synlett 6, 915932.
Joergensen, N.L., Thestrup, H.N., 1995. Determination of amino acids in
biomass and protein samples by microwave hydrolysis ion-exchange
chromatography. J. Chromatogr. A 706 (12), 421428.
Kennedy, J.F., Stevenson, D.L., White, C.A., Viikari, L., 1989. The
chromatographic behaviour of a series of fructooligosaccharides
derived from levan produced by the fermentation of sucrose by
Zymomonas mobilis. Carbohyd. Polym. 10, 103113.
Li, B.W., 1998. Comparison of microwave oven and convection oven for
acid hydrolysis of dietary ber polysaccharide. J. Assoc. O. Anal.
Chem. Int. 81 (6), 12771280.
Lindstrom, P., Thierney, J., Wathey, B., Westman, J., 2001. Microwave
assisted organic synthesis a review. Tetrahedron 57, 92259283.
Manning, T.S., Gibson, R.G., 2004. Microbial-gut interactions in health
and disease Prebiotics. Best Pract. Res. Clin. Gastroenterol. 18 (2),
287298.
Morales-Rubio, A., Cerezo, J., Salvador, A., De La Gardia, M.J., 1993.
Rapid acid hydrolysis of albumin in a microwave oven. Microchem. J.
47, 270277.
Singh, V., Tiwari, A., Tripathi, D.N., Malviya, T., 2003. Microwave
promoted methylation of plant polysaccharides. Tetrahedron Lett. 44,
72957297.
Viikari, L., 1988. Carbohydrate Metabolism in Zymomonas mobilis. CRC
Cr. Rev. Biotechnol. 7, 237261.
Yun, W.Y., 1996. Frutooligosacharides Occurrence, preparation and
application. Enzyme Microbiol. Technol. 19, 107117.