Starch determination in Chlorella vulgaris—a comparison
between acid and enzymatic methods
Bruno Fernandes & Giuliano Dragone & Ana P. Abreu &
Pedro Geada & José Teixeira & António Vicente
Received: 12 September 2011 /Revised and accepted: 12 September 2011 /Published online: 4 December 2011
#
Springer Science+Business Media B.V. 2011
Abstract Different methods for estimating starch in Chlo-
rella vulgaris were compared with the view of establishing
a procedure suitable for rapid and accurate determination of
starch content in this microalgal species. A close agreement
was observed between methods that use perchloric acid and
enzymatic methods that use α-amylase and amyloglucosi-
dase to hydrolyze the starch of microalgae grow n under
different nitrogen culture conditions. Starch values obtained
by these methods were significantly higher than those
estimated by using hydrochloric acid as solubilizing and
hydrolyzing agent. The enzymatic method (EM1) proved to
be the most rapid and precise method for microalgal starch
quantification. Furthermore, the evaluation of resistant starch
by enzymatic methods assayed in nitrogen-sufficient and
nitrogen-starved cells showed that no formation of this type of
starch occurred in microalgae, meaning that this should not
interfere with starch content determinations.
Keywords Amylolytic enzymes
.
Biofuels
.
Chlorella
vulgaris
.
Microalgae
.
Perchloric acid
.
Starch hydrolysis
Introduction
The ongoing depletion of oil reserv es co upled with
economic growth and stability, and more significantly, the
emerging concern abou t gl obal wa rming arising from
burning fossil fuels have become major drivers for the
development of renewable and cost-effective energy sour-
ces (Stephens et al. 2010; Dragone et al. 2010). Among the
different potential sources of renewable energy, liquid
biofuels such as bioethanol (a petrol additive/substitute)
and biodiesel (a diesel alternative) are of most interest as
they are part of the few options to replace fossil fuels and
have the potential to limit greenhouse gas emissions (Chen
et al. 20 11; Nigam and Singh 2011). However, the
production of biofuels from terrestrial plants is controver-
sial mainly due to the impact on global food markets, on
food security (Brennan and Owende 2010), on arable land
usage, on potable water utilizat ion and on deforestation. In
contrast, microalgae have been considered as a promising
feedstock for biofuel production since they are able to
convert solar energy to chemical energy via CO
2
fixation
and do not compete for land with crops used for food
production (Ahmad et al. 2011). These photosynthe tic
microorganisms accumulate significant qu antities of lipids
and carbohydrates over short periods of time that can be
subsequently proces sed into biofuels (Brennan and Owende
2010; Spolaore et al. 2006; Chen et al. 2009).
Certain species of microalgae such as Chlorella vulgaris
have the ability to produce higher levels of starch than
lipids as reserve polymer (Dragone et al. 2011). Beyond
that, some strains were also found to accumulate large
amounts of starch under nitrogen starvation. These species
are suitable candidates for bioethanol production as starch
from microalgae can be extracted to produce fermentable
sugars (Mussatto et al. 2010).
Therefore, an accura te and rapid method for the
determination of starch is key to the commercial success
of bioethanol production from microalgae. On the other
hand, some difficulties are usually encountered in choosing
and implementing an appropriate methodology for micro-
algal starch quantification. The great number of different
B. Fernandes
:
G. Dragone (*)
:
A. P. Abreu
:
P. Geada
:
J. Teixeira
:
A. Vicente
IBB—Institute for Biotechnology and Bioengineering,
Centre of Biological Engineering, University of Minho,
Campus de Gualtar,
4710-057 Braga, Portugal
e-mail: gdragone@deb.uminho.pt
G. Dragone
e-mail: giulianodragone@hotmail.com
J Appl Phycol (2012) 24:1203–1208
DOI 10.1007/s10811-011-9761-5
starch methods reported in the literature complicates the
task of method evaluation and selection . Furthermore,
variations in the accuracy of starch determination methods
may confound the interpretation and compa rison of results
among different studies.
Methods for starch determination in microalgae can be
broadly grouped into acid hydrolysis or enzymatic proce-
dures. Exampl es of the former include hydrolysis with
perchloric acid (Chader et al. 2009) while the latter comprises
digestion with amylase and amyloglucosidase (Zemke-White
and Clements 1999). Both procedures hydrolyze the starch to
glucose, which is subsequently quantified colorimetrically.
However, acid-based procedures might be subject to error in
starch estimation due to the extraction of interfering
carbohydrates from other polymers. Furthermore, enzyme
digestion has been the preferred method of determining
starch because, in theory, active, purified starch-degrading
enzymes are specific for the hydrolysis of starch and yield
highly accurate values (Rose et al. 1991).
The objective of this study was to compare different
methodologies for the determination of starch in C. vulgaris,
and to establish a procedure suitable for rapid and accurate
routine measurement of starch content in microalgae.
Materials and methods
The freshwater microalga Chlor ella vulgaris P12 kindly
provided by the Algal Laboratory (CCALA), Institute of
Botany, Academy of Sciences of the Czech Republic was
precultivated in standard (nitrogen-sufficient) medium con-
taining (mg L
−1
): 1100 (NH
2
)
2
CO, 237 KH
2
PO
4
, 204
MgSO
4
·7H
2
O, 40 FeNa–C
10
H
12
O
8
N
2
,88CaCl
2
,0.83
H
3
BO
3
,0.95CuSO
4
·5H
2
O, 3.3 MnCl
2
·4H
2
O, 0.17
(NH
4
)
6
Mo
7
O
24
·4H
2
O, 2.7 ZnSO
4
·7H
2
O, 0.6 CoSO
4
·7H
2
O,
0.014 (NH
4
)VO
3
in distilled water (Fernandes et al. 2010).
Cells in the late exponential growth phase were centrifuged
at 8,750×g for 15 min, washed in distilled water and
resuspended in nitrogen-sufficient and nitrogen-starved (stan-
dard medium without urea) culture media. Photoautotrophic
cultivation of C. vulgaris was performed at 30°C in 1 L
photobioreactors containing 400 mL of medium with a surface
irradiance of 70 μmol photons m
−2
s
−1
provided by four
fluorescent lamps (Sylvania Standard F18W). All cultures
were agitated using air enriched with 2% (v/v) CO
2
at an
aeration rate of 0.833 vvm (volume of gas per volume of
culture suspension per minute).
Methods for starch determination
Microalgal cells at the beginning of the stationary growth
phase were harvested by centrifugation at 8,750×g for
15 min, washed in distilled water, lyophilized and disinte-
grated (10 mg) with a mortar and pestle prior to starch
analysis. The disintegration of the cells was performed for
5 min, monitored through microscopic observation. Cells
were remo ved from the mortar and pestle using solvents
(ethanol solution or acetone, depending on the method).
Starch content was expressed as % w/w (dry weight basis).
The different methods for determination of microalgal starch
compared in this study are summarized in Table 1.
Perchloric acid method (AM1)
In this method (Rose et al. 1991), the cells were first
extracted with acetone and subsequently extra cted with
boiling 80% (v/v) ethanol until the extract remained
colorless, in order to remove interfering substances (e.g.,
pigments, soluble sugars and lipids) and to gelatinize starch
granules. The removal of interfering substances is extreme-
1204 J Appl Phycol (2012) 24:1203–1208
Table 1 Comparison of reagents and estimated time employed in the determination of microalgal starch concentration by acid and enzymatic
methods
Step Method
AM1 AM2 AM3 EM1/EM2
Removal of interfering substances Acetone Acetone
80% ethanol 80% ethanol 80% ethanol 80% ethanol
8h 1h 2h 0.5h
Starch extraction and solubilization 35% HClO
4
30% HClO
4
1.1% HCl α-Amylase
Amyloglucosidase
0.5 h 1 h 0.5 h 1 h
Colorimetric determinationof glucose Anthrone Anthrone Anthrone Glucose oxidase + peroxidase
0.5 h 0.5 h 0.5 h 0.75
Total time 9.0 h 2.5 h 3.0 h 2.25 h
AM1 perchloric acid method (Rose et al. 1991), AM2 modified perchloric acid method (Brányiková et al. 2011), AM3 hydrochloric acid method
(Oren et al. 1988), EM1 enzymatic method (Megazyme 2009), EM2 enzymatic method for resistant starch (Megazyme 2009)
ly important since they are able to react colorimetrically,
thus leading to the overestimation of starch values.
Subsequently, the microalgal starch was extracted and
solubilized with 35% (v/v) perchloric acid. The solubilized
starch solution was then reacted with a mixture of concentrated
sulfuric acid and anthrone (2 g anthrone in 1 L of 72% (v/v)
H
2
SO
4
) to quantify glucose spectrophotometrically at 625 nm.
Modified perchloric acid method (AM2)
In this method (Brányiková et al. 2011), the removal of
interfering substances using acetone was avoided, and the
duration of this and subsequent (starch extraction and
solubilization) steps was shortened, when compared with
AM1 (Table 1). Pigments were extracted three times using
80% ethanol for 15 min at 68°C. For total hydrolysis of
starch, 30% perchloric acid was added to the sediment,
stirred for 15 min at 25°C and centrifuged. This procedure
was repeated three times. The solubilized starch solution was
then reacted with a mixture of concentrated sulfuric acid and
anthrone as described for AM1.
Hydrochloric acid method (AM3)
The microalgal biomass was extracted with acetone and
boiling 80% (v/v) ethanol as described in the AM1 method
aiming at the removal of interfering substances. Subse-
quently, starch granules were hydrol yzed with 1.1% hydro-
chloric acid at 100°C for 30 min (Oren et al. 1988).
Glucose was determined colorimetrically by the anthrone
reaction after starch hydrolysis.
Enzymatic method (EM1)
The starch content of C. vulgaris was assayed by enzymatic
degradation of the starch to glucose with α-amylase and
amyloglucosidase, using the total starch assay procedure from
Megazyme (Megazyme 2009) accepted by AOAC—Associa-
tion of Analytical Communities (Official Method 996.11) and
AACC—American Association of Cereal Chemists (Method
76.13). Lyophilized microalgal biomass, previously disinte-
grated, was resuspended in 80% (v/v) ethanol and incubated in
a water bath at 80–85°C for 5 min, in order to extract interfering
compounds. Thermostable α-amylase (3,000 U mL
−1
)in
MOPS buffer (50 mM, pH 7.0) containing 5 mM CaCl
2
,
was added to each sample. The samples were incubated in a
boiling water bath for 6 min, with mixing at 2 min intervals,
andthenplacedinablockheaterat50°Candallowedto
equilibrate for 5 min. Amyloglucosidase (3,300 U mL
−1
)in
sodium acetate buffer (200 mM, pH 4.5) plus sodium azide
(0.02% w/v) was subsequently added to each sample. After
that, samples were incubated at 50°C for 30 min, and
centrifuged for 10 min at 4,500 × g to separate any remaining
insoluble material. Aliquots of the supernatant were assayed
for glucose. Each aliquot was added to 3.0 mL of GOPOD
reagent in distilled water. This reagent contained, according
to the manufacturer’s spec ifications: glucose oxidase
(>12,000 U); peroxidase (>650 U) and 4-aminoantipyrine
(80 mg). Samples were incubated at 50°C for 20 min and
then cooled to room temperature. The absorbance of samples
and the
D-glucose control were measured at 510 nm in a
spectrophotometer against a reagent blank solution consist-
ing of 0.1 mL of water and 3.0 mL of GOPOD reagent.
Enzymatic method for resistant starch (EM2)
The enzymatic method for resistant starch proposed by
Megazyme (Megazyme 2009) was based on the EM1 method
detailed above, except that after extraction of interfering
compounds with hot ethanol, the microalgal biomass was
predissolved with 2 M KOH in an ice/water bath, followed by
neutralization with sodium acetate buffer and further hydro-
lysis with α-amylase and amyloglucos idas e.
Statistical design and analysis
Each of the five methods was tested in triplicate on the
microalgal biomass. The replicates were performed using cells
from two cultivations, one under nitroge n sufficient and another
under nitrogen starved conditions. The accuracy of each
method was evaluated as the percentage relative error (%Er)
between their mean results and those obtained by the official
method EM1 according to the equation given below (Eq. 1).
%Er ¼ 100
»
C
M
C
OF
ðÞ=C
OF
ð1Þ
where C
M
= mean starch concentration measured by a specific
method and C
OF
= mean starch content obtained by the
official enzymatic method EM1. The precision of each
method was obtained by determining the percentage relative
standard deviation (%RSD) according to Eq. 2.
%RSD ¼ 100
»
s=C
M
ð2Þ
where s = standard deviation and C
M
= mean starch
concentration measured by a specific method. Results were
analyzed by the Experimental Design Module of the Statistica
8.0 software (Statsoft, USA).
Results and discussion
Comparison of starch content in C. vulgaris determined by
acid and enzymatic methods
Figure 1 shows the comparison of different methods for
determining the starch content in microalgae cultivated
J Appl Phycol (2012) 24:1203–1208 1205
under nitrogen sufficient conditions. It can be observed that
there was no statistically significant difference (p<0.05)
between the starch content of C. vulgaris determined either
by perchloric acid methods (AM1 and AM2) and enzymatic
methods (EM1 and EM2).
On the other hand, the hydrochloric acid method (AM3)
provided nearly 20% lower estimat es of starch content than
those obtained by perchloric acid methods (AM1 and
AM2). These results may be attributed to the higher
efficiency of starch extraction and hydrolysis with HClO
4
in comparison with HCl. According to Raessler et al.
(2010), the accurate determin ation of starch is dependent on
both its complete ext raction from the sample and its
complete hydrolysis into glucose. In green algae, starch is
synthesized and stored within the chloroplast, which
considerably limits the accessibility of the solvent during
extraction. Consequently, extraction of starch generally
needs rather harsh conditions to allow thorough access of
the solvent. A previous study (Fontana et al. 2001) reported
that the higher the acid strength, the higher the yield of
glucose released from starch by acid hydrolysis. As a result,
the higher starch values obtained by the perchloric acid
method can be related to the greater acid strength of HClO
4
compared to that of HCl. In this sense, perchloric acid has
been considered to be the most efficient solvent for starch
extraction from plant tissues (Ghiena et al. 1993).
Starch content of C. vulgaris estimated by the hydro-
chloric acid method was also significantly lower than those
obtained by both perchloric acid methods at the end of
microalgae cultivation under nitrogen starvation conditions
(Fig. 2). As can be seen in Fig. 2, the specific starch
concentration in nitrogen-depleted microalgae reached 30%
as determined by the hydrochloric acid method, while the
estimation of the microalgal starch content attained 33.5%
and 35% according to AM1 and AM2 methods, respec -
tively, which use HClO
4
as a solubilizing and hydrolyzing
agent. The significant differences among the starch values
obtained by both acid methods may be related to the
difference in acid strengths between HClO
4
and HCl, as
explained above.
Figure 2 also shows that enzymatic methods yielded
significantly higher microalgal starch concentrations than
that obtained by the hydrochloric acid method. On the other
hand, the starch values estimated by using amylolytic
enzymes did not differ significantly (at a 95% confidence
level) from those resul ting from perchloric acid methods.
Table 2 summarizes the precision and accuracy of the
tested methods for the determination of starch in microalgae
cultivated under nitrogen sufficient conditions. According
to this table, the accuracy of AM2 method was similar to
that of the perchloric acid method AM1 for the estimation
of the microalgal starch content. Moreover, it can be
observed in Table 2 that precisions (10.8%) achieved by
both perchloric acid methods were comparable to that
obtained by the enzymatic method EM1 (9.2%).
Although the perchloric acid method AM2 provided a
higher relative error (lower accuracy) than method AM1
AM1 AM2 AM3 EM1 EM2
0
10
20
30
40
b
a
a
a
Starch (%)
Anal
y
tical Method
a
Fig. 2 Starch concentration (mean of three replicates±standard
deviation) in C. vulgaris grown under nitrogen starvation conditions.
Columns with the same letters are not significantly different (p<0.05)
according to the Tukey's test for mean comparisons. (AM1 perchloric
acid method (Rose et al. 1991), AM2 modified perchloric acid method
(Brányiková et al. 2011), AM3 hydrochloric acid method (Oren et al.
1988), EM1 enzymatic method (Megazyme 2009), EM2 enzymatic
method for resistant starch (Megazyme 2009))
AM1 AM2 AM3 EM1 EM2
0
1
2
3
4
ab
b
ab
a
Starch (%)
Anal
y
tical Method
a
Fig. 1 Starch concentration (mean of three replicates±standard
deviation) in C. vulgaris grown under nitrogen sufficient conditions.
Columns with the same letters are not significantly different (p<0.05)
according to the Tukey’s test for mean comparisons. (AM1 perchloric
acid method (Rose et al. 1991), AM2 modified perchloric acid method
(Brányiková et al. 2011), AM3 hydrochloric acid method (Oren et al.
1988), EM1 enzymatic method (Megazyme 2009), EM2 enzymatic
method for resistant starch (Megazyme 2009))
1206 J Appl Phycol (2012) 24:1203–1208
(3.4 and 0.9, respectively) for determining starch concen-
tration in nitrogen-starved cultures of C. vulgaris (Table 3),
starch values estimated by both perchloric acid methods did
not differ significantly from that obtained by the enzymatic
method EM1 (Fig. 2). Additionally, both perchloric acid
methods yielded sim ilar precisions to that reported for the
determination of starch in microalgae cultivated under
nitrogen sufficient conditions (Table 2).
By comparing Tables 2 and 3, it can also be noticed that
methods AM1 and AM2 provided a more accurate value of
the starch content in nit rogen-starv ed ce lls than that
estimated in cells grown under nitrogen-sufficient condi-
tions. This result may be due to the lower concentration of
interfering compounds (e.g., proteins) present in microalgae
grown under nitrogen starvation conditions. It has been
reported (Esposito et al. 2006) that the protein content in
green microalgae is about threefold higher in nitrogen-
sufficient cells with respect to nitrogen-starved cells.
According to Chow and Landhäusser (2004), water-
soluble compounds such as proteins react with the
concentrated sulfuric acid in the glucose assay and,
therefore, significantly interfere with the absorbance read-
ing. Tables 2 and 3 also show that the enzymatic method
EM1 had the best precision in starch quantification, while
method AM3 yielded the highest percentage relative
standard deviation.
Evaluation of resistant starch in C. vulgaris
Many of the properties of starches including gelatinisation
characteristics, solubility, and the formation of resistant
starch determine their suitability for particular applications
(Maršálková et al. 2010). By defin ition, resistant starch is
the portion of starch that resists hydrolysis by amylolytic
enzymes in the small intestine. It is a linear molecule of α-
1,4-D-glucan, essentially derived fr om the retrog raded
amylose fraction, and has a relatively low molecular weight
(1.2×10
5
Da) (Fuentes-Zaragoza et al. 2010). Resistant
starch formation depends on several factors such as
physical structure of starch, possible cross-linking/structural
modification of starch and protein–starch/lipid–starch inter-
actions (Tharanathan and Mahadevamma 2003).
In our study, formation of resistant starch in C. vulgaris
was evaluated by comparing starch contents determined
either by the enzymatic method EM1 for nonresistant starch
and the enzymatic method for samples containing resistant
starch (EM2). According to Figs. 1 and 2, no statistically
significant differences were found between the starch
values determined by both enzymatic methods at the end
of the cultivation period under nitrogen sufficient and
nitrogen starvation conditions, respectively. These findings
led us to hypothesize that formation of resistant starch did
not occur under either growing condition.
Conclusions
We conclude that perchloric acid methods and enzymatic
methods gave similar values for the concentrations of starch
in C. vulgaris grown under different nitrogen conditions.
On the other hand, hydrochloric acid method resulted in
significantly lower estimates of star ch in microalgae. The
enzymatic method EM1 that uses α-amylase and amylo-
glucosidase to hydrolyze the starch of microalgae proved to
be the m ost rapid and precise procedure for starch
determination in C. vulgaris.
By comparing the enzymatic method for nonresistant
starch and the enzymatic method for samples containing
resistant starch, we concluded that no resistant starch was
formed in microalgae cultivated neither under nitrogen
sufficient nor nitrogen star vation conditions.
Acknowledgement This research work was supported by the grants
SFRH/BD/44724/2008 (Bruno Fernandes) from Fundação para a
CiênciaeaTecnologia(Portugal)andSFRH/BPD/44935/2008
(Giuliano Dragone). The authors also acknowledge the financial
support received through the projects INNOVALGAE (FCT PTDC/
AAC-AMB/108511/2008) and ALGANOL.
Table 2 Comparison of accuracy and precision of methods for
estimating starch content in nitrogen-sufficient microalgae
Method Accuracy (%) Precision (%)
AM1 17.8 10.8
AM2 15.1 10.8
AM3 −11.7 40.3
EM1 – 9.2
EM2 1.1 9.3
AM1 perchloric acid method (Rose et al. 1991), AM2 modified
perchloric acid method (Brányiková et al. 2011), AM3 hydrochloric
acid method (Oren et al. 1988), EM1 enzymatic method (Megazyme
2009), EM2 enzymatic method for resistant starch (Megazyme 2009)
Table 3 Comparison of accuracy and precision of methods for
estimating starch content in nitrogen-starved microalgae
Method Accuracy (%) Precision (%)
AM1 −0.9 11.8
AM2 3.4 11.4
AM3 −11.2 7.7
EM1 – 1.5
EM2 0.1 1.7
AM1 perchloric acid method (Rose et al. 1991), AM2 modified
perchloric acid method (Brányiková et al. 2011), AM3 hydrochloric
acid method (Oren et al. 1988), EM1 enzymatic method (Megazyme
2009), EM2 enzymatic method for resistant starch (Megazyme 2009)
J Appl Phycol (2012) 24:1203–1208 1207