J. Biomedical Science and Engineering, 2009, 2, 526-531
doi: 10.4236/jbise.2009.27076 Published Online November 2009 (http://www.SciRP.org/journal/jbise/
Published Online November 2009 in SciRes. http://www.scirp.org/journal/jbise
Enzymatic pretreatment and microwave extraction of
asiaticoside from Centella asiatica
Cheng-Xiao Wang, Wei Han*, Liang Fan, Chun-Li Wang
Engineering Center for Traditional Chinese Medicine Modernization, East China University of
Science and Technology, Shanghai, China.
E-mail: *whan@ecust.edu.cn
Received 3 July 2009; revised 13 July 2009; accepted 19 July 2009.
The extraction of asiaticoside from Centella asiatica
by enzymatic pretreatment and microwave extraction
(EPME) was studied in this article. The effects of
several important factors such as temperature of en-
zymatic pretreatment, liquid to solid ratio and mi-
crowave radiation time were investigated by quadric
regression orthogonal design experiment and were
analyzed by response surface. An extraction model
with well forecast performance was then established.
The results indicate that the optimum extraction
condition was as follows: liquid to solid ratio was
36mL/g, temperature of enzymatic pretreatment was
45, enzymatic time was 30min, and microwave ra-
diation time was 110s. On such conditions the yield
was 27.10%.
Keywords: Asiaticoside; Enzymolysis; Microwave Ex-
traction; Centella Asiatica
Centella asiatica (L.) Urban, a perennial herb belonging
to the umbelliferae family, is well known as a traditional
Chinese herbal medicine. Its conventional efficacy is
clearing away heat and toxic, inducing diuresis and re-
ducing edema. The major active constituents of Centella
asiatica are asiaticoside and madecassoside. It has been
used for the treatment of hot and humid jaundice, trau-
matic injuries, infectious hepatitis and dermatosis [1].
Currently, some conventional extraction methods are
mostly adopted for the extraction of asiaticoside [2,3,4],
such as aqueous extraction and ethanol extraction.
In recent years, new kinds of extraction techniques
appeared, including enzymolysis and microwave extrac-
tion. The former has impressive effects with characteris-
tics of high catalytic efficiency, high specificity, mild
reactive conditions and preserving the original efficacy
of active compounds to the maximum [5]. The later
method has many advantages, such as shorter time, less
solvent, higher extraction rate and better products with
lower cost [6,7]. However, the application of the combi-
nation of these two methods on plant materials was rare-
ly reported.
For the purpose of improving the efficient of asiatico-
side extraction, reducing the cost, the EPME method is
employed in this article, quadric regression orthogonal
design is adopted to investigate the effects of three main
extraction parameters including temperature of enzyma-
tic pretreatment, liquid to solid ratio and microwave ra-
diation time on the yield of asiaticoside, and optimum
extraction process is worked out.
2.1. Equipments and Reagents
An ER-692 microwave oven (as shown in Figure 1)
with a power output of 650W, operating at 2450MHz,
was mechanically modified to fit a reflux system that
enables extraction performed under atmospheric settings
and permits solvent containment. The extraction vessel
was a 250-mL three-necked round-bottomed flask con-
nected to a water condenser.
Thermostat with magnetic stirrer (Model DF-101S,
Yuhua Experimental Apparatus Co., Shanghai) was used
as enzymolysis device. UV spectrophotometer (Model
1900PC, Yayan Electronic Co., Shanghai) was used for
Dry Centella asiatica (fitted for the Chinese Pharma-
copoeia); Asiaticoside standard (Provided by National
Institute for the Control of Pharmaceutical and Biologi-
cal Products, Beijing); Cellulase (400U/mg); Anhy-
drous ethanol and concentrated sulfuric acid used in the
experiment were all of analytical grade.
2.2. Analytic Method of Asiaticoside
The concentration of the asiaticoside in this process was
determined by ultraviolet spectrophotometry [8] and the
result was expressed as extraction yield, i.e. unit extrac-
C. X. Wang et al. / J. Biomedical Science and Engineering 2 (2009) 526-531 527
tion quantity (g asiaticoside/g Centella asiatica).
The asiaticoside standard sample solution (was diluted
by anhydrous ethanol) and extraction solution were scan
ed at 200-400nm respectively, there were the same ab-
sorption peak at 277nm (Figure 2), which was close to
the literature values [9]. So 277nm was chosen for the
UV wavelengths.
The asiaticoside standard sample solution (concentra-
tion: 92µg/mL) was precisely measured at 0.0, 0.5, 1.0,
1.5, 2.0, 2.5mL, and put into a 10mL volumetric flask.
Firstly, volatilized out the solvent ethanol in the boiling
water bath, and then added the 2mL concentrated sulfu-
ric acid (H2SO4) after the flask cooling to the ambient
temperature, heated for 30min in 80 water bath. Fi-
nally, added anhydrous ethanol to the scale after the
flask getting to the ambient temperature. According to
the ultraviolet spectrophotometry, the prepared solution
was measured at 277nm. Regression equation and corre-
lation coefficient were y=43.40x-2.49 and r=0.9992 (n=7)
respectively. The linear range was 4.6~23.0µg/mL.
The 0.5mL test sample solution was accurately meas-
ured and placed into a 10mL volumetric flask, following
the preparation method of regression equation. The ab-
sorbance of test sample was determined, and the yield of
asiaticoside in the test sample was calculated in accor-
dance with the following equation:
Yield of asiaticoside (%, w/w)=
where, A, the absorbance of asiaticoside in test sample;
V, volume of solvent, mL; n, diluted times.
Figure 1. Microwave equipment diagram; 1-Water condenser;
2-Air condenser; 3-Copper tube; 4-Tailored tube; 5-Air agitator;
6-Status display; 7-Microwave oven timer; 8-Microwave oven;
9-Flask; 10-Base of flask.
Figur e 2 . UV spectra of reference solution and extraction solution.
2.3. Enzymatic Pretreatment and Microwave
The dry Centella as iatica (sieved through 10 screen me-
sh) 3.0g was accurately weighed and placed into a three-
neck flask with 3% cellulase solution (dissolved by de-
ionized water), then the deionized water as solvent was
added in according to a certain ratio (mL/g) of material
to solvent volume, and the mass of system was weighed.
Three-neck flask was put into the thermostat with
magnetic stirrer, setting enzymolysis time at 30min ac-
cording to the pre-experiment, while the temperature of
enzymatic reaction and stirring speed were set at certain
values. Then the flask was taken out and placed into the
microwave oven immediately. The radiation exposure
was 30s for preventing the serious evaporation of solvent.
At the end of each exposure, the system was brought
back to ambient temperature during 2~3 min interval by
cooling it with a water bath. An extraction cycle was
defined as the combination of phases of radiation and
phases without radiation in which the solvent cooled.
The sum of radiation exposure of processing extraction
cycles served as the overall intensity of microwave ra-
diation. The extraction solution was agitated with an air
pump to promote heat uniformity while exposing to mi-
crowave radiation. After the radiation, the flask was tak-
en out to weigh the total mass again, and the lost weight
was supplemented. The extract was filtered through
0.45µm millipore filter, and then abandoned the initial
filtrate, added 0.5mL subsequent filtrate to a 10mL volu-
metric flask with anhydrous ethanol as the test sample.
3.1. Effect of Temperature of Enzymatic
Pretreatment on EPME
As shown in Figure 3, the results indicate that the yield
of asiaticoside was increased with the increase of tem-
perature of enzymatic pretreatment, reached a high point
Copyright © 2009 JBiSE
C. X. Wang et al. / J. Biomedical Science and Engineering 2 (2009) 526-531
at 45. Because the temperature of enzymatic pretreat-
ment was a significant factor in the process of enzymo-
lysis, it affected the enzyme activity as well as the rates
of enzyme-catalyzed reactions. When the temperature
was lower than 45, the increase of temperature can
improve the cellulase activity, accelerate the degradation
of cytoderm. While the temperature of enzymatic pre-
treatment was higher than 45, the cellulase activity
was decreased, leading to the reduction of the yield.
Therefore the temperature of enzymatic pretreatment for
45 was used.
3.2. Effect of Liqu id to Solid Rati o on EPME
As it is known, the liquid to solid ratio is very important
in the extraction. From the perspective of mass transfer,
it mainly affects the concentration gradient between liq-
uid phase and solid phase. Figure 4 shows that the yield
of asiaticoside was increased with the increase of liquid
to solid ratio. After the peak, the yield of asiaticoside
was decreased with the increase of liquid to solid ratio.
The higher liquid to solid ratio, the longer time for the
solution elevated to the same temperature was required.
At the same microwave radiation, the temperature of the
system with larger amount of solvent was relatively
lower; the solute diffusion would be affected. So the
ratio for 30mL/g was chosen.
3.3. Effect of Microwave Radiation Time on
Figure 5 described that the extraction yield sharply in-
creased before 60s and was asymptotic to a steady value
during 60~120s, then falls down after 120s. At the pre-
liminary stage of extraction, the velocity of molecular
thermal motion quickened and the asiaticoside was qui-
ckly separated from the cell into solution. The extraction
process had tended towards equilibrium since 60s. When
Figure 3. Effect of temperature of enzymatic pretreatment on EPME.
Figure 4. Effect of liquid to solid ratio on EPME.
Figure 5. Effect of microwave radiation time on EPME.
the microwave radiation time exceeded 120s, the system
was hyperthermal, the vaporization reflux of solution
increased. All of above brought about the reduction of
effective contact interval of solvent and plants, and
caused the decrease of the thermal effects. In addition,
the extension of microwave radiation time would in-
crease the consumption of energy. Thus the microwave
radiation time for 60s was used in the experiment.
3.4. Quadratic Regression Orthogonal
Design Results
The quadratic regression orthogonal design was employ-
ed to evaluate the relevance of the three main extraction
factors including enzymatic temperature, liquid to solid
ratio and the microwave radiation time, while other fac-
tors including grinding degree of plant material, concen-
tration of enzyme, enzymatic pretreatment time and pH
of the solvent were constant according to the pre-expe-
riments. The multivariate study allows the identification
of interactions between variables and provides a com-
plex exploration of the experimental domain to be stud-
ied with a number of experiments optimized.
The three key variables studied were pointed at five
separate coded levels [11], –1, 0, +1, +γ (=1.682) and
their values were selected on the basis of previous ex-
periments. The natural values and coded levels used in
this multivariate study are presented in Table 1.
Copyright © 2009 JBiSE
C. X. Wang et al. / J. Biomedical Science and Engineering 2 (2009) 526-531 529
Table 1. Factors of orthogonal experiment.
X1-Microwave X2-Liquid to X3-Enzymatic
-γ 9 13 28
-1 30 20 35
0 60 30 45
1 90 40 55
γ 111 47 62
Δ 30 10 10
The statistical analysis software was applied in this
experiment to establish a regression Eq.2, in this equa-
tion, the terms of Y, X1, X2, and X3 respectively repre-
sent the yield of asiaticoside, microwave radiation time,
liquid to solid ratio and enzymatic temperature:
Y=25.8963+1.8150X1-0.7646X2+0.346 4X3
+0.9363X1X2+0.4 688X 1X3+0.1363X2X3 (2)
Analysis of variance was carried out in order to test
the signification of the regression model. Thus, various
statistical data such as sum of squares (SS), mean sq-
uares (MS), F-ratio were given in Table 2.
The F-ratio in Table 2 was the ratio of the meansq-
uared error to the pure error obtained from the replicates
at the design center. The significance of the F-ratio de-
pends on the number of degrees of freedom (d. f.) in the
model. Thus, the effects lower than 0.05 in this column
were significant.
As shown in Table 2, FRegression=10.866>F0.05 (9,
8)=3.388, the regression of (2) was significant, while
FLack of fit=3.478<F0.05 (5, 8) =3.687, the lack of fit was
non-significant. Therefore, (2) had well predictivity un-
der the experimental condition. In the test of regression
coefficient, F-ratio for terms X3, X1X3, X2X3 (1.253,
1.344, 0.114) were lower than F0.05 (1, 8) =5.318, so
these terms were not significant. On the contrary, the
F-ratio for other terms (X1, X2, X1X2, X1
2, X2
2 and X3
which were higher than F0.05, indicated the significance
of these terms. As the orthogonality of this experiment,
the insignificant terms were cut out to simplify the (3):
Y=25.8963+1.8150X1-0.7646X2+0.936 3X1X2
2 (3)
Table 2. Variance analysis of test results.
Source d.f. SS MS F F0.05
X1 1 44.995 44.995 34.409 5.318
X2 1 7.985 7.985 6.106
X3 1 1.639 1.639 1.253
X1X2 1 7.012 7.012 5.363
X1X3 1 1.758 1.758 1.344
X2X3 1 0.148 0.148 0.114
X12 1 7.935 7.935 6.068
X22 1 7.463 7.463 5.707
X32 1 48.949 48.949 37.432
Regression 9 127.885 14.209 10.866 3.388
Lack of fit 5 22.740 4.548 3.478 3.687
Pure error 8 10.461 1.308
Total 22 161.087 7.322
Eq.3 indicated that the microwave radiation time and
liquid to solid ratio were the main factors that influence
the yield because of the significance of the terms X1 and
X2 .The significance of X1X2 suggested the obvious in-
teraction between microwave radiation time and liquid
to solid ratio. And the significance of all the quadratic
terms demonstrated the nonlinear relationship between
the three factors and the yield of extraction.
3.5. Analysis of Response Surface
The 3D surface curves were drawn to illustrate the main
and interactive effects of the three factors on the yield.
The response surfaces are shown in Figures 6,7,8 with
one variable kept at optimum level and the other two
varied within the experimental range.
Figure 6 shows the effect of liquid to solid ratio(X2)
and extraction time(X1) on the yield. A quadratic effect
for both factors on the response can be observed. At a
low level of X2 (-2), the system was readily hyperther-
mal and vaporization of the solvent could reduced the
yield with the increasing of microwave time. And at a
high level of X2 (2), the yield displayed an increasing
curve in the experimental range of X1. It is due to the
distinctly interaction between X1 and X2, which was
implied in the Eq.3. The maximum yield was predicted
when X1 was in the range of 1.5 to 1.7 and X2 varied
from 0.4 to 0.6.
Figure 7 depicts the effect of enzymatic tempera-
ture(X3) and extraction time(X1)as both them exerting
a quadratic effect. As shown in the Figure 7, an increase
in yield resulted when X3 was increased in the level
range from –2 to 0, then the curve started to go down,
which may indicate that a level of X1 at approximate 0
is required to achieve maximum yield. Likewise, an in-
crease in yield resulted when X1 was increased in the
code range from –2 to 1.5, and then the yield was
slightly reduced. In the response surface, X1 exerted a
more significant effect on yield than X3, and no obvious
interaction between X1 and X3 was observed, which
was well in agreement with Eq. 3.
In Figure 8, yield showed quadratic curve depending
upon the liquid to solid ratio(X2), whereas no significant
effect was observed in the enzymatic temperature(X3).
Because the ratio was a key factor which influences the
impetus in mass transfer of both enzymatic pretreatment
and microwave extraction processes, it exerted a more
significant effect on yield than the factor of enzymatic
temperature. According to the response surface, there
was no obviously interaction between X2 and X3, and it
was also supported by the Eq.3.
3.6. Optimization of EPME Condition
Yield of extraction was employed as the evaluation ob-
jective in the optimization of the parameter to ensure it
reaches the peak under constraint conditions. According
Copyright © 2009 JBiSE
C. X. Wang et al. / J. Biomedical Science and Engineering 2 (2009) 526-531
Figure 6. Response surface graph of microwave radiation time
and liquid/solid ratio.
Figure 7. Response surface graph of microwave radiation time
and enzymatic temperature.
Figure 8. Response surface graph of liquid/solid ratio and
enzymatic temperature.
Table 3. Optimum values and verification results
Value Calculated
Value Esti-
X1(s) 1.672 110
X2(mL/g)0.584 36
X3() 0 45
27.19 27.10
to (3), model was then built up as described hereinafter
Objective function: Y(X1,X2,X3);
Constraint conditions: –1.682Xi1.682; (i=1,2,3)
The optimum parameters and maximal yield were
worked out by Newton’s iteration method. The results
were shown in Ta bl e 3 . To compare the predicted result
(82.10%) with the practical value, the rechecking ex-
periment was performed using this deduced optimal
condition. The mean value of 27.10% (n=3), obtained
from real experiments, demonstrated the validity of the
model, since there was no significant differences be-
tween 27.19% and 27.10%. The strong correlation be-
tween the real and the predicted results confirmed that
the extraction model was adequate to reflect the ex-
pected optimization.
The quadratic regression orthogonal design was used in
this research, an extraction model which can accurately
predict the yield of asiaticoside extraction under the ex-
perimental condition was established.
Through the analysis of experiment data, it can be
found that the microwave radiation time and liquid to
solid ratio significantly influence the yield of the extrac-
tion. Especially, there was an obvious interaction be-
tween the microwave radiation time and liquid to solid
The optimum combination of the parameters for the
extraction of asiaticoside was obtained by the mathema-
tical methology; it was microwave radiation time for
110s; liquid to solid ratio for 36mL/g, and enzymatic
pretreatment temperature for 45. On this condition,
the maximum yield of extraction was 27.10%, closed to
the estimated value 27.19%.
The EPME procedure had the advantages of less time,
high efficiency of extraction, and environmentally frie-
ndly. It can be applied to other extraction of plant mate-
rials. But the expansive cost of the enzyme and difficulty
in industrialization of microwave extraction would limit
the further application of EPME.
The work was supported by National Natural Science Foundation of
China (No. 20006003).
Copyright © 2009 JBiSE
C. X. Wang et al. / J. Biomedical Science and Engineering 2 (2009) 526-531
SciRes Copyright © 2009
REFERENCES [6] Bayramoglu, B., Sahin, S., and Sumnu, G., (2008) Sol-
ventfree microwave extraction of essential oil from
oregano, J. Food Eng., 88, 535540.
[1] Randriamampionona, D., Diallo, B., and Rakotoniriana,
F., (2007) Comparative analysis of active constituents in
Centella asiatica samples from Madagascar: Application
for ex situ conservation and clonal propagation, Fitotera-
pia, 78, 482–489.
[7] Proestos, C. and Komaitis, M., (2008) Application of
microwave-assisted extraction to the fast extraction of
plant phenolic compounds, Food Sci. Technol-LEB, 41,
[2] Verma, K., Bhartariya, G., and Gupta, M., (1999) Rever-
sephase high performance liquid chromatography of asi-
aticoside in Centella asiatica, Phytochem. Anal., 10,
[8] Pan, X. J., Niu, G. G., and Liu, H. Z., (2003) Micro-
wave-assisted extraction of tea polyphenols and tea caf-
feine from green tea leaves, Chem. Eng. Process, 42, 129
[3] Sarma, K., Khosa, L., and Chansauria, N., (1995) An-
tistress activity of Tinospora cordfolia and Centella asi-
atica extracts, Phytother. Res., 10, 181183.
[9] Lu, X. Y., “Determination of asiaticosides by UV- spec-
trophotography. (2005) J Modern Food Drug, 16, 1516.
[10] Ma, S. F., Wang, L. Q., and Hu, Z. C., (2005) Enzymatic
extraction of the submerged mycelium polysaccharide
from Pleurotus nebrodensis, T CSAE, 22, 198201.
[4] Flora, S. and Gupta, R., (2007) Beneficial effects of Cen-
tella asiatica aqueous extract against arsenic-induced
oxidative stress and essential metal status in rats, Phyto-
ther. Res., 21, 980988. [11] Virot, M., Tomao, V., and Colnagui, G., (2007) New mi-
crowave-integrated Soxhlet extraction an advantageous
tool for the extraction of lipids from food products, J.
Chromatogr. A, 1174, 138144.
[5] Li, B. B., Smith, B., and Hossain, M., (2006) Extraction
of phenolics from citrus peels II, Enzyme-assisted ex-
traction method, Sep. Purif. Technol., 48, 189196.