Enzyme and Microbial Technology 48 (2011) 100–105

Contents lists available at ScienceDirect

Enzyme and Microbial Technology

journal homepage: www.elsevier.com/locate/emt

Enzyme-assisted extraction of flavonoids from Ginkgo biloba leaves:

Improvement effect of flavonol transglycosylation catalyzed by Penicillium
decumbens cellulase
Shuo Chen a,1 , Xin-Hui Xing a , Jian-Jun Huang b , Ming-Shu Xu b,∗
a
Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
b
Marine College, Shandong University at Weihai, Shandong, Weihai 264209, China

a r t i c l e i n f o a b s t r a c t

Article history: We report a novel enzyme-involved approach to improve the extraction of flavonoids from Ginkgo biloba,
Received 31 August 2010 in which the enzyme is employed not only for cell wall degradation, but also for increasing the solubility
Accepted 28 September 2010 of target compounds in the ethanol–water extractant. Penicillium decumbens cellulase, a commercial cell
wall-degrading enzyme with high transglycosylation activity, was found to offer far better performance
Keywords: in the extraction than Trichoderma reesei cellulase and Aspergillus niger pectinase under the presence of
Enzyme-assisted extraction
maltose as the glycosyl donor. TLC, HPLC and MS analysis indicated that P. decumbens cellulase could
Penicillium decumbens cellulase
transglycosylate flavonol aglycones into more polar glucosides, the higher solubility of which led to
Transglycosylation
Flavonoid
improved extraction. The influence of glycosyl donor, pH, solvent and temperature on the enzymatic
Ginkgo biloba transglycosylation was investigated. For three predominant flavonoids in G. biloba, the transglycosyla-
Solubility tion showed similar optimal conditions, which were therefore used for the enzyme-assisted extraction.
The extraction yield turned to be 28.3 mg/g of dw, 31% higher than that under the pre-optimized con-
ditions, and 102% higher than that under the conditions without enzymes. The utilization of enzymatic
bifunctionality described here, naming enzymatic modification of target compounds and facilitation of
cell wall degradation, provides a novel approach for the extraction of natural compounds from plants.
© 2010 Elsevier Inc. All rights reserved.

1. Introduction The extraction of plant flavonoids, which generally have poor sol-
ubility in mild solvents such as ethanol–water solution, is selected
Enzyme-assisted extraction of natural functional compounds as an example to validate our proposal.
from plants is widely investigated in recent years for its advantages Flavonoids are natural compounds showing high physiological
in easy operation, high efficiency, and environment friendship [1]. activities in therapies for inflammations, heart diseases and cancers
Most of the works in this field utilize cellulase, pectinase and ␤- [5,6]. There are three predominant flavonol aglycones, quercetin,
glucosidase to hydrolyze and degrade plant cell wall constituents kaempferol, and isorhamnetin (Fig. 1), in Ginkgo biloba, a Chinese
to improve the release of intracellular contents [2–4]. However, medicinal plant well known for its high content of flavonoids [7].
another important factor in the extraction – the intrinsic prop- Extract from G. biloba is among the most popular phytomedicines
erty especially the solubility of the target compound – has seldom and herbal dietary supplements [8], whose primary active compo-
been concerned according to our knowledge. Low solubility of tar- nents are flavonoids (24%) and small amount of terpene lactones
get compounds in the extractant leads to low extraction yield and (6%) [9].
require large amount of solvents, which largely impedes the eco- Various methods have been used to assist the extraction of
nomic efficiency in industry. Therefore, here we propose a novel flavonoids from plants, such as ultrasonication [10], supercriti-
enzyme-assisted extraction approach, in which the extraction is cal fluids [11], microwave [12], membrane adsorption [13] and
improved not only from enhanced cell wall degradation, but also molecular imprinting [14]. There are also several reports on the
from the increased solubility of target compounds in the extractant. enzyme-assisted extraction of plant favonoids. Pectinase and pro-
tease were employed for the extraction of anthocyanins from black
currant juice press residues [15]. Luteolin and apigenin were enzy-
matically extracted from pigeonpea leaves with pectinase, cellulase
∗ Corresponding author. Tel.: +86 631 5688303; fax: +86 631 5688303.
and ␤-glucosidase [16]. In another study, extraction of antimicro-
E-mail address: [email protected] (M.-S. Xu).
1
Current address: Department of Chemistry and Biotechnology, School of Engi- bial and antioxidant phenolics, mainly anthocyanins, from berries
neering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. was studied, and the enzyme was found to effectively hydrolyze

0141-0229/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.enzmictec.2010.09.017
 S. Chen et al. / Enzyme and Microbial Technology 48 (2011) 100–105 101

is also the first report on enzyme-assisted extraction of flavonoids

from ginkgo leaves.

2. Materials and methods

2.1. Plant material

Fig. 1. Chemical structures of three predominant flavonol aglycones in G. biloba.

Dried leaves of G. biloba were bought from Henan Medical Lim-
ited (Zhengzhou, China) and ground into powder. The particle size
glycosides to their corresponding aglycones [17]. Although the was controlled within 280 and 600 ␮m using sifters.
enzyme-assisted approach has largely improved the extraction
rate, due to the intrinsic low polarity of flavonol aglycones, the 2.2. Chemicals and enzymes
extraction after enzymatic treatment still has to be performed with
solvents like hexane, acetone and butanone in order to achieve Quercetin, kaempferol and isorhamnetin were purchased from
high productivity. Nevertheless, regarding toxicological limita- Sigma–Aldrich (Steinheim, Germany). Cellulase from Trichoderma
tions, there is a clear trend in the industry to substitute these reesei ATCC 26921 (Celluclast 1.5 L, ≥700 U/g) was bought from
organic solvents for alternative nontoxic solvents, particularly in Novozymes A/S (Bagsvaerd, Denmark). Pectinase from Aspergillus
products for human consumption, with the first option being water niger (>1 U/mg) was bought from Sigma–Aldrich (Steinheim, Ger-
or alcohols [18]. Therefore, large amount of high concentration many). P. decumbens cellulase was provided by Ningxia Cellulase
ethanol–water solution is often used. However, this unfavorably Preparation Plant, China. All other reagents were of analytical grade
leads to a high production cost, making it desirable to increase and commercial available.
the solubility of flavonoids without affecting their physiological
activities. 2.3. Enzyme-assisted extraction of total flavonoids
We come up with an idea to utilize the transglycosylation
activity of glycosidases to introduce hydrophilic groups, e.g. gly- Distilled water was added to enzymes to obtain stock solutions
cosides, into flavonoids to improve their polarity in the extraction at the concentration of 2 mg/mL. For each batch of enzyme-assisted
process, while utilizing the enzyme’s activity in cell wall hydrol- extraction, 50 mL enzyme stock solution and certain amount of
ysis at the same time. Although commercial cellulases are often glycosyl donor were diluted with ethanol–water and adjusted
selected and modified to minimize their transglycosylation activ- with acetate buffer to obtain 500 mL extractant with the desired
ity for more complete degradation of cellulose chains in biomass pH, ethanol–water ratio and glycosyl donor concentration. 30 g
transformation [19], many cellulase systems in microorganisms ginkgo leaf powder was added to the extractant in an Erlenmeyer
show both hydrolysis and the reverse transglycosylation activ- flask covered with aluminum foil. The mixture was incubated
ities [20]. Enzymatic transglycosylation by various glycosidases, under 200 rpm stirring for 30 h on a multichannel magnetic stirrer
including glucosidases [20–22], galactosidases [20], xylosidases with temperature controller (Guohua Electronics Co., Changzhou,
[23], rhamnosidases [24], etc., has already gained importance to China). After incubation, the mixture was filtered and the fil-
be an efficient and low cost way to synthesize oligosaccharides, trate was forwarded to analysis. For the comparison of extractions
alkyl glucosidases, and glycoconjugates. In a study by Gao et al. assisted by different enzymes, incubation was done at 40 ◦ C in the
[25], catechin, a flavonoid, was successfully glycosylated into its extractant with an ethanol–water ratio of 3:7 (v/v) and at pH 6.0.
glycoside form by several glycosidases. In our case, the trans-
glycosylated product of ginkgo flavonoids, naming the flavonol 2.4. Determination of total flavonoids
glycosides, present physiological activities on almost the same level
with flavonol aglycones, and can be much more readily absorbed by The aluminumchloride colorimetric method described by Chang
the human body [6] directly in the small intestine or after hydrol- et al. [29] was used to determine the total content of flavonoids.
ysis by bacterial enzymes in the intestine [5]. A commercial cell 0.5 mL extract or standard solution was mixed with 1.5 mL
wall-degrading enzyme, cellulase from Penicillium decumbens, was methanol, 0.1 mL of 10% aluminum chloride (substituted with dis-
used in our research for flavonol glycosylation. It is composed tilled water in blank probe), 0.1 mL of 1 M potassium acetate, and
of endoglucanase, cellobiohydrolase and ␤-glucosidase [26,27]. 2.8 mL of distilled water. After 30 min incubation, absorbance at
Besides high activity for cellulose degradation, P. decumbens cel- 415 nm was determined against a distilled water blank on a UV-
lulase shows high transglycosylation activity probably originating 1206 spectrophotometer (Shimadzu, Kyoto, Japan). All samples
from its ␤-glucosidase activity as indicated in our previous research were made in triplicate, and mean values of total flavonoid con-
[28]. tent are expressed as milligrams of quercetin equivalents per gram
In this work, we aim to make it clear whether the enzymatic of dry weight (dw) calculated according to the standard calibration
transglycosylation of flavonoids could improve their extraction curve.
from ginkgo leaves. For this purpose, the effect of P. decumbens
cellulase on the extraction yield was compared with several other 2.5. TLC analysis
cell wall degrading enzymes with little transglycosylation activi-
ties under the addition of glycosyl donors. After validated for its Extract from ginkgo leaves or enzymatic reaction mixture was
improving effect on the extraction, flavonol transglycosylation by loaded to a silica gel plate (10 cm × 20 cm, GF254, Qingdao Haiyang
P. decumbens cellulase was confirmed by analyzing the catalytic Chemical Co. Ltd., China) and developed with the mobile phase of
products of the three predominant flavonoids in G. biloba. Then benzene–ethyl acetate–acetone–acetic acid (10:8:2:2, v/v/v/v).
we optimized the enzymatic conditions, studied the transglyco-
sylation kinetics, and examined whether the extraction could be 2.6. Enzymatic transglycosylation reaction
improved under the optimal transglycosylation conditions. To the
best of our knowledge, this is the first report to utilize enzymatic The reaction system for the enzymatic transglycosylation consi-
bifunctionality – for cell degradation and target product transfor- sted of 2 mL stock solution of P. decumbens cellulase, 25 mg flav-
mation – in the extraction of natural compounds from plants. This onoid substrate and 25 mg glycosyl donor in 48 mL ethanol–water
102 S. Chen et al. / Enzyme and Microbial Technology 48 (2011) 100–105

solvent. Prior to the addition of the enzyme, pH of the solution was

adjusted by acetate buffer. The enzymatic catalysis was carried out
for 30 h under 200 rpm stirring.
To determine the transglycosylation rate, 50 ␮L enzymatic reac-
tion mixture was loaded on a silica gel plate and developed with
the same mobile phase as described above. Under the exposure in
the ultraviolet light from a portable UV lamp, flavonoids and the
reaction products at the corresponding Rf positions were scraped
and extracted in 10 mL ethanol through 0.5 h ultrasonication. The
extract was filtrated and the absorbance of the filtrate at 370 nm
was measured on the UV-1206 spectrophotometer (Shimadzu,
Kyoto, Japan). According to the standard curve prepared by the
same method, the residual flavonoid concentration (C) after trans-
glycosylation was obtained. Transglycosylation rate was calculated
by the following equation:
C0 − C
transglycosylation rate (%) = × 100,
C0
Fig. 2. Effect of different enzyme and maltose additions on the extraction yields of
where C0 and C refer to the initial and residual flavonoid concen- total flavonoids.
tration (mg/mL), respectively.
To study the kinetics of the transglycosylation, a series of exper-
iments is carried out with varying substrate concentrations under due to its better performance in cell wall degradation with its com-
the conditions of pH 6, 50% ethanol, 60 ◦ C and 0.5 g/L maltose addi- prised activities from pectinesterase, polygalacturonase and pectin
tion. The substrate concentrations at 0 h and 0.5 h in the reaction lyase [16].
system were determined from TLC in the same way as described In order to study the effect of transglycosylation on the extrac-
above. The reaction rate calculated from the concentration change tion, maltose was added as a glycosyl donor to each of the extraction
of the substrate in the first 0.5 h was used as the initial velocity. systems containing different enzymes. Adding glycosyl donors is
The measurements were then plotted in a Lineweaver–Burk plot, generally considered to be an efficient way to enhance transglyco-
in which the inverse of substrate concentration was plotted against sylation while inhibiting the reverse hydrolysis. It is interesting to
the inverse of the initial velocity. The values of the desired constants find that the yield of the extraction assisted by P. decumbens cellu-
Km and Vmax were read directly off the plot. lase, in contrast to the other two enzymes, was largely increased
with the addition of maltose. When there was 2 g/L maltose in the
2.7. HPLC analysis extractant, the extraction yield with treatment by P. decumbens cel-
lulase achieved 21.6 mg/g of dw, much higher than those by the
HPLC analysis was carried out using a Symmetry C18 column other two enzymes. This indicated that, glycosyl donor addition has
(5 ␮m, 250 mm × 4.6 mm, Waters Corp., Milford, USA) attached to notable improving effect on the extraction involving P. decumbens
a Shimadzu HPLC LC-10A system (Shimadzu, Kyoto, Japan) with a cellulase but almost no effect on that involving A. niger pectinase or
photodiode array detector monitored at 370 nm. The mobile phase T. reesei cellulase. It has been reported that P. decumbens cellulase
was methanol–water (1:1, v/v), the flow rate was set at 0.5 mL/min has high transglycosylation activity [28], while A. niger pectinase
and the column temperature was maintained at 40 ◦ C. has no transglycosylation activity, and T. reesei cellulase has little
transglycosylation activity due to loss of its ␤-glucosidase con-
2.8. MS analysis stituent in the production process [19]. Therefore, it is reasonable
to consider that, it is the maltose-induced transglycosylation that
The enzymatic reaction mixtures were fractionated using a improves the extraction assisted by P. decumbens cellulase.
semi-preparative scale HPLC system consisting of the same type of
column (5 ␮m, 250 mm × 4.6 mm, Waters Corp., Milford, USA) at a 3.2. Confirmation of transglycosylation by P. decumbens cellulase
flow rate of 5.0 mL/min. The fractionated reaction product was iso-
lated and further analyzed by mass spectrometry. The mass spectral In order to validate the transglycosylation of flavonoids in
analyses were performed on a PE Sciex API 3000 mass spectrome- the extraction process with P. decumbens cellulase, we analyzed
ter with ESI ion source (Applied Biosystems, USA) (m/z 100–1000; the profile of the extraction products. TLC analysis of the total
0.3 amu; dwell: 1.0 ms; pause: 2.0 ms). flavonoid extracts with P. decumbens cellulase treatment showed
size decrease of the spots at high Rf and size increase of the spots
3. Results and discussion at low Rf compared with that without enzyme treatment, revealing
the generation of compounds with higher polarity in the extraction
3.1. Effect of enzymes with different transglycosylation activities process. In order to identify the generated products, three pre-
on extraction yield dominant flavonoids in ginkgo leaves, quercetin, kaempferol and
isorhamnetin, were used as model substrates to study the catalysis
Extraction of total flavonoids from ginkgo leaves was conducted by P. decumbens cellulase.
with three enzymes, Trichoderma reesei cellulase, A. niger pectinase TLC results for the three substrates and their catalytic prod-
and P. decumbens cellulase. As shown in Fig. 2, the extraction yields ucts were shown in Table 1. Generation of compounds with higher
of total flavonoids increased with treatment by the three enzyme polarity was confirmed after enzyme treatment. It is important to
preparations. The amounts of flavonoids extracted with A. niger note that, the spots for quercetin, kaempferol and isorhamnetin at
pectinase were slightly higher than those with T. reesei cellulase Rf 0.68, 0.87 and 0.70, together with the spots for their catalytic
and P. decumbens cellulase. Fu et al. also reported that pectinase had products at Rf 0.10, 0.10 and 0.14 after enzymatic reaction, could
higher improving effect on the extraction of luteolin and apigenin all be clearly identified on the TLC for the extracts after enzyme-
from pigeonpea leaves than cellulase or ␤-glucosidase, probably assisted extraction.
 S. Chen et al. / Enzyme and Microbial Technology 48 (2011) 100–105 103

Table 1
TLC analysis for enzymatic catalysis and extraction of flavonol aglycones.

TLC sample Rf value

Quercetin standard 0.68

Quercetin after enzymatic catalysis 0.68, 0.10
Kaempferol standard 0.87
Kaempferol after enzymatic catalysis 0.87, 0.10
Isorhamnetin standard 0.70
Isorhamnetin after enzymatic catalysis 0.70, 0.14
Total flavonoid extract without 0.68, 0.87, 0.70
enzyme treatment
Total flavonoid extract with enzyme 0.68, 0.87, 0.70, 0.10, 0.10, 0.14
treatment

In order to identify the polar products generated in the

enzymatic catalysis, HPLC and MS analyses were performed. Chro-
matograms for the catalytic products of quercetin, kaempferol and
isorhamnetin all clearly indicated the generation of a new com-
pound with a shorter retention time after catalysis (Fig. 3). The m/z
ratios of the new compounds were determined to be 464 [M+H]+
(calcd for C21 H20 O12 , 464), 448 [M+H]+ (calcd for C21 H20 O11 , 448)
and 478 [M+H]+ (calcd for C22 H22 O12 , 478) on MS, correspond-
ing to the glucosides of quercetin, kaempferol and isorhamnetin,
respectively. The results proved the transglycosylation of flavonol
aglycones into flavonol glucosides by P. decumbens cellulase dur-
ing the extraction process. Flavonol glucosides, which have higher
polarity, are much more soluble in the ethanol–water extractant
than flavonol aglycones. This could explain why after P. decum-
bens cellulase treatment, the extraction efficiency was remarkably
improved.

3.3. Optimal conditions for enzymatic transglycosylation

As suggested by the maltose addition study, enhanced flavonoid

extraction could be achieved through facilitating the transgly-
cosylation by P. decumbens cellulase. So the optimization of P.
decumbens cellulase catalyzed transglycosylation was studied.
Quercetin, kaempferol and isorhamnetin were used as the model
substrates.

3.3.1. Glycosyl donor

Soluble starch, maltose and dextrin were selected as candidate
glycosyl donors. The concentration of all the glycosyl donors was
controlled at 0.5 g/L. As shown in Fig. 4A, for all three substrates,
highest transglycosylation rate was obtained when maltose was
used as the glycosyl donor.

3.3.2. pH
Effects of pH in the range of 2–9 on the transglycosylation was
Fig. 3. Chromatograms of flavonoid samples before and after treatment with P.
examined by preparing the reaction solutions with different initial decumbens cellulase: (A) quercetin; (B) kaempferol; (C) isorhamnetin.
pHs. As shown in Fig. 4B, P. decumbens cellulase showed transg-
lycosylation activities within a wide pH range. Around pH 6 the
catalytic rate was the highest for all three substrates.
the substrates. When ethanol concentration was higher than 60%,
3.3.3. Solvent ethanol–water ratio the transglycosylation rate decreased rapidly with the increase of
Substrate solubility generally has large effects on the catalytic ethanol, which was presumably due to the enzyme inactivation
reaction. Flavonol aglycones have very low solubility in water, and caused by excess ethanol in the solution.
the solubility rises as the ethanol concentration increases in water
solution. So the effects of solvent ethanol concentrations from 0 to
90% (v/v) on the enzymatic transglycosylation were examined. As 3.3.4. Temperature
shown in Fig. 4C, the transglycosylation rate reached the highest As shown in Fig. 4D, the transglycosylation rate reached the peak
at the ethanol concentration of 50% for kaempferol and isorham- at 60 ◦ C for all of quercetin, kaempferol and isorhamnetin. When
netin, and 60% for quercetin. When ethanol concentration was the temperature was higher than 60 ◦ C, the catalysis was notably
between 20 and 50%, the catalysis was facilitated with the increase inhibited, probably due to the inactivation of P. decumbens cellulase
of ethanol, probably resulted from the increase of the solubility of at a high temperature.
104 S. Chen et al. / Enzyme and Microbial Technology 48 (2011) 100–105

Fig. 4. Effect of the glycosyl donor (A), pH (B), ethanol concentration (C), and temperature (D) on the enzymatic transglycosylation of quercetin, kaempferol, and isorhamnetin.

3.3.5. Transglycosylation kinetics by P. decumbens cellulase thermore, the kinetics of the enzymatic transglycosylation was
According to the above results, quercetin, kaempferol and studied. The initial rate of transglycosylation showed a good rela-
isorhamnetin have very similar optimal conditions for their tionship with the initial substrate concentration, which fits to the
transglycosylation by P. decumbens cellulase. Time dependence Michaelis–Menten equation. Table 2 gives the kinetic parameters
experiment showed that the transglycosylation reached the for the three flavonoid substrates under the optimal catalytic con-
plateau after 30 h of enzymatic treatment under the optimal con- ditions.
ditions of pH 6, 50% ethanol, 60 ◦ C and 0.5 g/L maltose (Fig. 5).
The transglycosylation rates of quercetin, kaempferol and isorham- 3.4. Extraction under optimal transglycosylation conditions
netin reached 45%, 50% and 41%, respectively (Table 2), compared
with 25%, 27% and 27% under the pre-optimized conditions. Fur- We further examined whether the extraction of flavonoids is
improved when the transglycosylation reaches its highest under
the optimal catalytic conditions of pH 6, 50% ethanol and 60 ◦ C.
As shown in Table 3, total flavonoid content in the extracts
was improved under the optimal transglycosylation conditions

Table 2
Transglycosylation rates and kinetic parameters of flavonol aglycones catalyzed by
P. cellulase under optimal conditions (pH 6, 50% ethanol, 0.5 g/L maltose, 60 ◦ C, 30 h).

Flavonol aglycone Transglycosylation Km (mM) Vmax (␮M/h)

rate (%)

Quercetin 45 2.40 3.57

Kaempferol 50 3.70 20.0
Isorhamnetin 41 2.43 0.55

Table 3
Extraction yield of total flavonoids from G. biloba.

Extraction condition Yield (mg/g dw)

Without enzymes 14.0 ± 0.3

Fig. 5. Time dependence of the transglycosylation rate of flavonol aglycones (25 mg
Under pre-optimized enzymatic conditions 21.6 ± 0.3
in 50 mL solvent) under optimal enzymatic conditions (pH 6, 50% ethanol, 60 ◦ C and
Under optimized enzymatic conditions 28.3 ± 0.5
0.5 g/L maltose).
 S. Chen et al. / Enzyme and Microbial Technology 48 (2011) 100–105 105

as expected. The extraction yield turned to be 28.3 mg/g of dw, [8] Tang J, Sun J, Zhang Y, Li L, Cui F, He Z. Herb–drug interactions: effect of Ginkgo
31% higher than that under the pre-optimized conditions, and biloba extract on the pharmacokinetics of theophylline in rats. Food Chem
Toxicol 2007;45:2441–5.
102% higher than (or 2 folds of) that under the conditions without [9] Ahlemeyer B, Krieglstein J. Neuroprotective effects of Ginkgo biloba extract. Cell
enzyme involvement. Mol Life Sci 2003;60:1779–92.
As shown in Fig. 2, the extraction yield was increased to 1.3 folds [10] Smelcerovic A, Spiteller M, Zuehlke S. Comparison of methods for the exhaus-
tive extraction of hypericins, flavonoids, and hyperforin from Hypericum
by treatment with pectinase resulting from its facilitating effect on perforatum L. J Agric Food Chem 2006;54:2750–3.
cell wall degradation. In the case of P. decumbens cellulase, even [11] Wang L, Yang B, Du X, Yi C. Optimisation of supercritical fluid extraction of
the cell wall degradation activity could be seen at the same level flavonoids from Pueraria lobata. Food Chem 2008;108:737–41.
[12] Chen L, Jin H, Ding L, Zhang H, Li J, Qu C, Zhang H. Dynamic microwave-assisted
as pectinase, there was another 0.7 fold increase in the extraction extraction of flavonoids from Herba Epimedii. Sep Purif Technol 2008;59:50–7.
yield, which should be accounted for flavonol transglycosylation. [13] Yu FC, Lai SM, Suen SY. Extraction of flavonoid glycosides from Ginkgo
As mentioned above, the flavonol transglycosylation rate was about biloba leaves and their adsorption separations using hydrophobic and anion-
exchange membranes. Sep Sci Technol 2003;38:1033–50.
50% under the optimal conditions, revealing that when half of the
[14] Xie J, Zhu L, Luo H, Zhou L, Li C, Xu X. Direct extraction of specific pharma-
flavonol aglycones were enzymatically glycosylated into more sol- cophoric flavonoids from gingko leaves using a molecularly imprinted polymer
uble glycosides, the extraction yield was increased by at least 70%. for quercetin. J Chromatogr A 2001;934:1–11.
The present study demonstrates P. decumbens cellulase to be [15] Landbo AK, Meyer AS. Enzyme-assisted extraction of antioxidative phenols
from black current juice press residues (Ribes nigrum). J Agric Food Chem
a high efficient enzyme to assist the extraction of flavonoid com- 2001;49:3169–77.
pounds in mild solvents like ethanol–water. The high performance [16] Fu YJ, Liu W, Zu YG, Tong MH, Li SM, Yan MM, Efferth T, Luo H. Enzyme assisted
of P. decumbens cellulase originates from its high activity not only extraction of luteolin and apigenin from pigeonpea [Cajanus cajan (L.) Millsp.]
leaves. Food Chem 2008;111:508–12.
in facilitating cell wall degradation, but also in transglycosylat- [17] Puupponen-Pimiä R, Nohynek L, Ammann S, Oksman-Caldentey KM, Buchert
ing flavonol aglycones into more polar glycosides which have J. Enzyme-assisted processing increases antimicrobial and antioxidant activity
higher solubility in the extractant. In general, this study provides of bilberry. J Agric Food Chem 2008;56:681–8.
[18] Santamaría RI, Reyes-Duarte MD, Bárzana E, Fernando D, Gama FM, Mota M,
a novel view for enzyme-assisted extraction of plant secondary López-Munguía A. Selective enzyme-mediated extraction of capsaicinoids and
metabolites that, besides focusing on enzyme-facilitated cell wall caratenoids from chili guajillo puya (Capsicum annum L.) using ethanol as sol-
degradation, targeting on enzymatic modification of the proper- vent. J Agric Food Chem 2000;48:3063–7.
[19] Andrić P, Meyer AS, Jensen PA, Dam-Johansen K. Reactor design for minimizing
ties of target compounds could be an alternative approach. In this product inhibition during enzymatic lignocellulose hydrolysis II. Quantifi-
approach, the key point is to find an enzyme which could modify cation of inhibition and suitability of membrane reactors. Biotechnol Adv
the properties, such as solubility, of target compounds for more 2010;28:407–25.
[20] Hansson T, Andersson M, Wehtje E, Adlercreutz P. Influence of water activity
efficient extraction without damaging their bioactivity. Moreover,
on the competition between ␤-glycosidase-catalysed transglycosylation and
by using bifunctional enzymes or mixed enzyme preparations, the hydrolysis in aqueous hexanol. Enzyme Microb Technol 2001;29:527–34.
two approaches might be combined for even higher extraction per- [21] Kato N, Suyama S, Shirokane M, Kato M, Kobayashi T, Tsukagoshi N, Novel. ␣-
formance. Glucosidase from Aspergillus nidulans with strong transglycosylation activity.
Appl Environ Microbiol 2002;68:1250–6.
[22] Yang S, Jiang Z, Yan Q, Zhu H. Characterization of a thermostable extracellu-
lar ␤-glucosidase with activities of exoglucanase and transglycosylation from
References Paecilomyces thermophila. J Agric Food Chem 2008;56:602–8.
[23] Eneyskaya EV, Brumer H, Backinowsky LV, Ivanen DR, Kulminskaya AA,
[1] Sowbhagya HB, Chitra VN. Enzyme-assisted extraction of flavorings and Shabalin KA, Neustroev KN. Enzymatic synthesis of ␤-xylanase substrates:
colorants from plant materials. Crit Rev Food Sci Nutr 2010;50:146– transglycosylation reactions of the ␤-xylosidase from Aspergillus sp. Carbohydr
61. Res 2003;338:313–25.
[2] Barzana E, Rubio D, Santamaria RI, Garcia-Correa O, Garcia F, Sanz VER, [24] Martearena MR, Daz M, Ellenrieder G. Synthesis of rutinosides and rutinose by
López-Munguía A. Enzyme-mediated solvent extraction of carotenoids reverse hydrolysis catalyzed by fungal ␣-l-rhamnosidases. Biocatal Biotrans-
from marigold flower (Tagetes erecta). J Agric Food Chem 2002;50: form 2008;26:177–85.
4491–6. [25] Gao C, Mayon P, MacManus DA, Vulfson EN. Novel enzymatic approach to
[3] Salgado-Roman M, Botello-Alvarez E, Rico-Martínez R, Jiménez-Islas H, the synthesis of flavonoid glycosides and their esters. Biotechnol Bioeng
Cárdenas-Manríquez M, Navarrete-Bolaños JL. Enzymatic treatment to 2001;71:235–43.
improve extraction of capsaicinoids and carotenoids from chili (Capsicum [26] Sun X, Liu Z, Zheng K, Song X, Qu Y. The composition of basal and induced cellu-
annuum) fruits. J Agric Food Chem 2008;56:10012–8. lase systems in Penicillium decumbens under induction or repression conditions.
[4] Lavecchia R, Zuorro A. Improved lycopene extraction from tomato peels using Enzyme Microb Technol 2008;42:560–7.
cell-wall degrading enzymes. Eur Food Res Technol 2008;228:153–8. [27] Chen M, Qin Y, Liu Z, Liu K, Wang F, Qu Y. Isolation and characterization of a
[5] Stalikas CD. Extraction, separation, and detection methods for phenolic acids ␤-glucosidase from Penicillium decumbens and improving hydrolysis of corn-
and flavonoids. J Sep Sci 2007;30:3268–95. cob residue by using it as cellulase supplementation. Enzyme Microb Technol
[6] Middleton E, Kandaswami C, Theoharides TC. The effects of plant flavonoids 2010;46:444–9.
on mammalian cells: implications for inflammation, heart disease, and cancer. [28] Xu MS, Luo MF, Xing XH, Chen HZ. Characteristics of quercetin transglycosi-
Pharmacol Rev 2000;52:673–751. dation catalysed by Penicillium decumbens glycosidase. Food Bioprod Process
[7] Gray D, LeVanseler K, Pan M, Waysek EH. Evaluation of a method to determine 2006;84:237–41.
flavonol aglycones in Ginkgo biloba dietary supplement crude materials and [29] Chang CC, Yang MH, Wen HM, Chern JC. Estimation of total flavonoid content
finished products by high-performance liquid chromatography: collaborative in propolis by two complementary colorimetric methods. J Food Drug Anal
study. J AOAC Int 2007;90:43–53. 2002;10:178–82.