ARTICLE

Microalgae—Novel Highly Efficient Starch

Producers
Irena Brányiková,1,2 Barbora Maršálková,2 Jiřı́ Doucha,1 Tomáš Brányik,2
Kateřina Bišová,1 Vilém Zachleder,1 Milada Vı́tová1
1
Institute of Microbiology, AS CR, Department of Autotrophic Microorganisms,
Laboratory of Cell Cycle of Algae, Opatovický mlýn, 379 81 Třeboň, Czech Republic;
telephone: þ420-384-340-480; fax: þ420-384-340-415; e-mail: zachleder@gmail.com
2
Department of Fermentation Chemistry and Bioengineering, Institute of Chemical
Technology in Prague, Prague, Czech Republic
Received 30 December 2009; revision received 24 October 2010; accepted 1 November 2010
Published online 1 December 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bit.23016

around 18% of DW, sulfur limitation increased the starch

ABSTRACT: The freshwater alga Chlorella, a highly produc- content to 50% of DW.
tive source of starch, might substitute for starch-rich ter- Biotechnol. Bioeng. 2011;108: 766–776.
restrial plants in bioethanol production. The cultivation ß 2010 Wiley Periodicals, Inc.
conditions necessary for maximizing starch content in KEYWORDS: bioethanol; Chlorella; cycloheximide; light
Chlorella biomass, generated in outdoor scale-up solar intensity; limitation by elements; starch
photobioreactors, are described. The most important factor
that can affect the rate of starch synthesis, and its accumula-
tion, is mean illumination resulting from a combination of
biomass concentration and incident light intensity. While
8.5% DW of starch was attained at a mean light intensity of
215 mmol/(m2 s1), 40% of DW was synthesized at a mean
light intensity 330 mmol/(m2 s1). Another important factor
Introduction
is the phase of the cell cycle. The content of starch was Microalgae for biofuels production are the subject of many
highest (45% of DW) prior to cell division, but during the
course of division, its cellular level rapidly decreased to
current research projects. High growth rate, high photo-
about 13% of DW in cells grown in light, or to about 4% synthetic efficiency, the relatively high content of energy-
in those kept in the dark during the division phase. To rich chemicals (whose level can, to a certain extent, be
produce biomass with high starch content, it is necessary to controlled), and experience with large-scale cultivation and
suppress cell division events, but not to disturb synthesis of down-stream processing technologies has, over the past
starch in the chloroplast. The addition of cycloheximide
(1 mg/L), a specific inhibitor of cytoplasmic protein synth-
decade, focused increased attention on microalgae (Chisti,
esis, and the effect of element limitation (nitrogen, sulfur, 2007; Klass, 2004). Cheap enhanced starch biomass can be
phosphorus) were tested. The majority of the experiments produced from highly productive Chlorella cultures grown
were carried out in laboratory-scale photobioreactors, where in suitable outdoor photobioreactors in which the photo-
culture treatments increased starch content to up to about synthetic carbon dioxide source is derived from combustion
60% of DW in the case of cycloheximide inhibition or sulfur
limitation. When the cells were limited by phosphorus or
of organic waste, fermentation processes or other sources
nitrogen supply, the cellular starch content increased to 55% (Doucha et al., 2005; Douskova et al., 2009; Mann et al.,
or 38% of DW, respectively, however, after about 20 h, 2009).
growth of the cultures stopped producing starch, and the The microorganism used for these experiments was the
content of starch again decreased. Sulfur limited and cyclo- freshwater chlorococcal alga, Chlorella sp. selected in our
heximide-treated cells maintained a high content of starch
(60% of DW) for up to 2 days. Sulfur limitation, the most
laboratory. The strain is characterized by a very high growth
appropriate treatment for scaled-up culture of starch- rate (mmax ¼ 0.20/h) and tolerance to a high culture
enriched biomass, was carried out in an outdoor pilot-scale temperature (408C). Its desirable technological properties
experiment. After 120 h of growth in complete mineral (resistance to shear stress, low adhesion to a surface of the
medium, during which time the starch content reached bioreactor, low tendency to form aggregates) are expected to
offer significant advantages for its use in large-scale
production bioreactors (Doucha and Lı́vanský, 2009). Its
Correspondence to: Vilém Zachleder
Contract grant sponsor: Ministry of Education, Youth and Sports of the Czech Republic ability to grow under high CO2 permits a direct supply of
Contract grant number: OE221; OE09025 flue gas containing up to 13% (v/v) of CO2 (Douskova et al.,

766 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011 ß 2010 Wiley Periodicals, Inc.
2009). This characteristic enhances the ecological and 2.7 ZnSO47H2O, 0.6 CoSO47H2O, and 0.014 NH4VO3 in
economic impact of the proposed technology, because of its distilled water. For preparation of medium, 100 con-
potential to bioremediate carbon dioxide emissions from centrated stock solutions of macroelements and microele-
different CO2 sources including waste incinerators, power ments were used. All components excluding urea were
stations, limekilns, cogeneration units, etc. in situ. Thus, one diluted in distilled water and autoclaved for 20 min. After
of the running costs, the supply of carbon dioxide, is cooling, urea was added and the pH was adjusted to 7 with
eliminated, and it could even have a positive value in carbon 1 M NaOH. The medium for the outdoor culture unit
credits associated with reduction of greenhouse gas (250 L) was not sterilized, and tap water instead of distilled
emissions. water was used for dilution of stock solutions.
We have tested the effect of light intensity, macroelement
(nitrogen, phosphorus, sulfur) limitation and the use of
cycloheximide, a specific inhibitor of cytoplasmic protein Cycloheximide Treatment
synthesis, to induce Chlorella cells to accumulate high levels
of starch. In addition to laboratory experiments, the Cycloheximide (Actidion, Serva Feinbiochemica, Heidelberg,
potential to increase starch content in a pilot outdoor Germany) was added at the beginning of the experiments.
culture unit was also tested. The final concentration in the mineral medium was 1 mg/L.

Materials and Methods Nitrogen, Phosphorus, or Sulfur Limitation

In laboratory experiments where limitation by macroele-
Microorganism ments was tested, the original mineral medium was removed
Chlorella vulgaris Beijerinck, strain CCALA 924, was isolated from the cell suspension by centrifuging, and the cells were
in Southern Greece, further selected in our laboratory, and resuspended in fresh mineral medium as control and in one
deposited in the Culture Collection of Autotrophic of the following modified mineral media:
Organisms (CCALA) in Třeboň, Czech Republic. In the Phosphorus limiting medium: KH2PO4 from the original
collection, the strain has been maintained on agar slants medium was replaced by 130 mg/L KCl, to keep the
under irradiance of about 23 mmol/(m2 s1), 12/12 h (light/ concentration of Kþ ions the same as in the original mineral
dark) regime and at temperature of 12–158C (http:// medium.
www.butbn.cas.cz/ccala/index.php). The strain is also Nitrogen limiting medium: urea was omitted from the
deposited as strain P12 in the Laboratory of Cell Cycles original mineral medium.
of Algae, Institute of Microbiology of ASCR, Opatovický Sulfur limiting medium: MgSO47H2O from the original
mlýn, 379 81 Třeboň. mineral medium was replaced by 168 mg/L MgCl26H2O to
keep the concentration of Mg2þ ions the same as in the
original mineral medium.
Measurement of Light Intensity
A quantum/radiometer-photometer (LI-COR, Inc., Lincoln, Starch Analyses
NE) was used. In the culture unit, dimmable fluorescent
tubes were used for adjustment of irradiance. To obtain a Samples of 2–5 mL of the algal suspension were harvested by
measure of light energy absorbed by a layer of cell centrifuging at 3,000g for 5 min, and the pellets were frozen
suspension grown at different incident light intensities at
208C. The following modification of the method of
and different optical densities (concentrations of cells), the McCready et al. (1950) was used. A paste of algal cells was
mean light intensity (I) was calculated according to the disintegrated by vortexing (Vortex Genie 2, Scientific
Lambert–Beer formula: I ¼ (Ii
It)/ln(Ii/It), where Ii is the Industries, Inc., Bohemia, NY) with 0.5 mL of glass beads
incident light intensity measured at the surface of the culture (0.14–0.39 mm) for 4 min (2,700 rpm) in 0.25 mL of
vessel, and It is the transmitted light intensity measured at distilled water. The pigments were extracted three times
the rear side of the culture vessel. using 4 mL of 80% ethanol for 15 min at 688C. For total
hydrolysis of starch, 3.3 mL of 30% perchloric acid were
added to the sediment, stirred for 15 min at 258C and
centrifuged. This procedure was repeated three times. The
Mineral Medium
extracts were combined and made up to 10 mL. Thereafter,
The mineral medium was based on the mean content of P, 0.5 mL of the extract was cooled to 08C; 2.5 mL of anthrone
N, K, Mg, and S in algal biomass (Doucha and Lı́vanský, solution [2 g of anthrone in 1 L of 72% (v/v) H2SO4] were
2006) and had the following initial composition (mg/L): added and stirred. The mixture was kept in a water bath at
1,100 (NH2)2CO, 237 KH2PO4, 204 MgSO47H2O, 1008C for 8 min. It was then cooled to 208C, and the
40 C10H12O8N2NaFe, 88 CaCl2, 0.83 H3BO3, 0.95 absorbance was measured at 625 nm. Calibration was carried
CuSO45H2O, 3.3 MnCl24H2O, 0.17 (NH4)6Mo7O244H2O, out simultaneously using glucose as the standard. The values

Brányiková et al.: Microalgae for Starch 767

Biotechnology and Bioengineering
measured for glucose were multiplied by 0.9 to obtain a Germany) was used. In graphs where straight lines between
calibration curve for starch determination. experimental points were not used, regression curves of
various orders were applied to fit the data. With the
exception of the outdoor culture unit, experiments were
Biomass Determination
carried out in triplicate, and the mean values were used to
For dry weight determination the biomass was separated construct the graphs.
from the medium by centrifuging the cell suspension at
3,000g for 5 min washing with distilled water and drying at
1058C for 12 h.
Cell volume and number were measured in glutaralde- Synchronization Procedure
hyde (0.2% (v/v) final concentration) fixed samples using a
The synchronization was carried out by alternating light/
Beckman Coulter Multisizer III (Coulter Corporation,
dark periods. To obtain synchronous populations of
Miami, FL) by diluting 10–50 ml of fixed cell suspension
daughter cells, optimal growth conditions were used
into 10 mL of 0.9% (w/w) NaCl as an electrolyte solution.
(incident light intensity 780 mmol/(m2 s1), temperature
308C, 2% CO2 in aeration mixture, 1  106 cells/mL,
Laboratory Photobioreactor light/dark 12/6 h). Under such conditions cell division
started after about 12 h of illumination, and the cells divided
A set of glass cylinders (inner diameter 36 mm, height
mostly into eight daughter cells. These cultures were
500 mm) were placed in a thermostatic bath (308C) and
observed using light microscopy for two cell cycles to set
continuously illuminated by a panel dimmable fluorescent
the correct length of both the light and dark periods. The
lamps (OSRAM DULUX L55 W/950 Daylight, Italy)
proper moment for darkening the cells was when about 10%
allowing adjustment of the incident light intensity from
of cells started their first protoplast fission. The length of the
16 to 780 mmol/(m2 s1). The incident and transmitted light
dark period was chosen to allow all cells of the population to
intensities were measured using a quantum radiometer–
release their daughter cells. Once the illumination regime
photometer (LI-COR, Inc.). The cylinders were ‘‘aerated’’
was adapted to the length of the cell cycle under a given light
using a mixture of air and CO2 (2%, v/v). The volume of the
intensity, the duration of the light and dark periods was kept
algal suspension in each cylinder was 300 mL, and each
constant. The synchronized daughter cells were then used as
cylinder was supplied with gas at a flow rate of 15 L/h.
an inoculum for experimental cultures.
The pH of cultures was maintained in the range 6.5–7.5 by
the addition of 1 M NaOH. The experiments were carried
out in a batch culture regime.
Results
Pilot Scale Outdoor Open Solar Photobioreactor
Effect of Mean Light Intensity
Algae were grown in an outdoor open thin-layer solar
photobioreactor with 24 m2 of culture area (length 24 m,
Synchronized Cultures
slope 1.7%), and a 6- to 7-mm thick layer of algal culture.
The suspension volume in the bioreactor was 250 L. The Accumulation of starch during the cell cycle and its phases
cultivation was performed in batch mode; the mineral was studied in synchronized batch cultures. The culture
medium at the beginning of experiment was of the same proliferation and changes in starch content were monitored
composition as in the laboratory experiments. On the 4th during two cell cycles (Fig. 1), while continuously
day of cultivation, mineral salts were added in an amount illuminated with 780 mmol/(m2 s1). As the biomass density
that corresponded to the ‘‘sulfur limiting medium’’ used for increased, the mean light intensity decreased from
laboratory cultivation. During the 5th day, the sulfates from 410 mmol/(m2 s1) to 215 mmol/(m2 s1) between the begin-
the mineral medium were consumed by the algae and on the ning and the end of the first cell cycle and fell further to
6th day of cultivation sulfur limitation occurred. The algal 170 mmol/(m2 s1) at the end of the second cycle (Fig. 1A).
suspension was exposed to sunlight, which resulted in an The decline in the mean light intensities was best fitted to an
illumination duration of approximately 11 h a day; during exponential decay curve (Fig. 1A, dotted curve).
the night, the suspension was kept in a retention tank and The relative starch content reached a maximum of 46% of
aerated. The experimental culture unit was located at 498N, DW (Fig. 1B) just prior to the first cell division, which
Czech Republic. The principles and details of the cultivation commenced at about the 10th hour of growth in light
system are described in Doucha and Lı́vanský (2006). (Figs. 1C and 2B). During cell division, the relative starch
content decreased, even in cells grown in light, to 13% of
DW (Fig. 1B, solid line). If the cells were transferred to the
Evaluation of Results
dark (Fig. 1, vertical arrows), they exhausted the starch
For creation of graphs and regression curve calculation, storage nearly completely (4% of DW) (Figs. 1B, dashed
Sigmaplot version 8 (Jandel Scientific Software, Erkrath, curve, and 2A). The decrease in starch content was

768 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011

 accompanied by a decrease in concentration of algal biomass
(Fig. 1A, dashed curve).
In the second cell cycle, the relative content of starch was
substantially lower (maximum 26% of DW) due to an
increased biomass concentration and correspondingly
decreased mean light intensity. During the second division
period the store of starch was decreased during the phase of
cell division in light to 7% of DW (Fig. 1B).

Asynchronous Cultures
In outdoor scale-up solar photobioreactors, it is not possible
to keep algal cultures synchronous and to harvest them at
the moment of maximum starch content before the cell
population begins to divide. The loss of synchrony is caused
by high culture density, where the length of the cell cycle due
to limited light is greater than 24 h. Thus only a part of the
population can divide during the dark period, and this
population decreases with increasing biomass concentra-
tion. Growth experiments with asynchronous cultures at
different mean light intensities were therefore carried out to
monitor the impact of different starting biomass concen-
trations (Fig. 3A) as well as the effect of continuously
decreasing mean light intensities with increasing biomass
concentration (Fig. 3D) on the content of starch (Fig. 3B).
The relative starch content (% of DW) increased in all
experiments until the mean light intensity failed to reach the
limit value of about 265 mmol/(m2 s1) (Fig. 3D, horizontal
dashed line), which corresponded to a biomass concentra-
tion about 1.4 g of DW/L (Fig. 3A, horizontal dashed line).
The higher the mean light intensity, the higher was the
relative content of starch (Fig. 3B). A decrease in the mean
light intensity below 265 mmol/(m2 s1) caused a gradual
decline in the rate of the starch synthesis leading finally to no
net increase of starch content (variant 0.3) or even to its
decrease in the case of more concentrated variants (0.5 and
0.9) (Fig. 3B), and a decrease in relative starch content (% of
DW) in all experimental variants (Fig. 3C). The decrease
occurred even in the variant, which did not attain the
limiting mean light intensity, but the increase in starch
content was linear (Fig. 3B, variant 0.2) while the increase in
biomass concentration was exponential.

Effect of the Treatment With Cycloheximide

The final content of starch in cells is determined by two main
factors: (1) the mean light intensity, which determines the
rate of the starch synthesis and its final content, and (2)
processes (mainly cell division) consuming the starch for
Figure 1. The effect of mean light intensity on changes in biomass concentra- their energy demands. One way to produce biomass with
tion (DW in g/L) (A), relative starch content (% of DW) (B) and cell number (mil/mL) (C)
in synchronized cultures of Chlorella. Triangular symbols represent the change in increased starch content would be to inhibit cell division
mean light intensity. The cultures were either grown for two cell cycles in continuous and/or other processes consuming starch but to allow light-
light (open symbols) or were transferred into dark during the division period of the first driven starch synthesis to continue in the chloroplast.
cell cycle (solid symbols). A regression curve of the 2nd order (exponential decay) was
fitted to the experimental data. To verify this hypothesis, we added cycloheximide, an
antibiotic that inhibits eukaryotic cytoplasmic protein
synthesis, without significantly reducing protein synthesis
in the chloroplast. The presence of cycloheximide (1 mg/L)

Brányiková et al.: Microalgae for Starch 769

Biotechnology and Bioengineering
Figure 2. Electron microscopic photographs of daughter (A) and mother (B) cells of Chlorella grown in complete mineral medium, in the presence of cycloheximide (1 mg/L)
(C), and in sulfur limiting medium (D). N, nucleus; S, starch granules. Bars: panels A–C ¼ 2 mm; bar panel D ¼ 5 mm.

prevented cells from undergoing nuclear division and was investigated as a viable and environmental-friendly
consequently no cell division occurred (Fig. 2). Starch, option to the control of cell cycle. During essential
however, was intensively synthesized (within 4–5 h) up to a macroelement limitation, biomass synthesis was reduced
level at least 60% of DW (Figs. 2 and 4), while the starch compared to control cultures (Fig. 5A), while starch content
content of the untreated cultures grown at the highest mean increased significantly during the first 12 h of growth.
light intensity was about 40% of DW (Fig. 3; for Limitation by individual elements had, however, different
comparison, see also Fig. 2A–C). Moreover, the content effects on (i) the growth rate of the culture (Fig. 5A), (ii) the
of starch in g/L was much lower than in cycloheximide rate of starch synthesis (Fig. 5B), (iii) the final relative
treated culture. After 5 h the rate of starch synthesis starch content (Fig. 5C), and (iv) the interval after which
equilibrated with biomass growth, and the high relative starch content began to decline (Fig. 5). In the case of the
content of starch remained constant in cycloheximide untreated culture (Fig. 5, dotted curves), a gradual decrease
treated cultures (Fig. 4C). in the rate of starch synthesis started after about 18 h of the
Another difference from untreated cultures (Fig. 3) was growth and continued up until the 24th hour. As was shown
that while the growth rate decreased with increasing biomass in Figure 2, this was caused by decreasing mean light
concentration, the final relative content of starch did not intensity with increasing culture density (above 1.4 g/L,
differ significantly. In the presence of cycloheximide, the horizontal dashed line in Fig. 5A). No decrease in starch
starch content increased continuously albeit at a slightly content occurred until hour 32 of the experiment. Despite a
decreased rate (compare Figs. 3C and 4C). continued increase in total biomass the relative starch
Cycloheximide is effective, reproducible and its mechan- content (in % of DW) decreased (compare dotted curves in
ism of action is well known. However, for environmental Fig. 5A–C).
security reasons the use of an antibiotic substance would not In the case of element limitation, starch synthesis was not
be appropriate for large-scale cultivation in open outdoor inhibited by a decrease in mean light intensity (Fig. 5A)
bioreactors. because the biomass concentration did not attain a value of
1.4 g/L at which, the reduction in mean light intensity would
start to limit the cell growth (Fig. 5A). Thus, reductions in
Limitation by Macroelements—Nitrogen, Phosphorus,
total biomass synthesis, as well as starch content, were
Sulfur
caused by shortage of the given element. In all variants, cell
The possibility of increasing starch content by limitation division was blocked, as observed by microscopic examina-
of a macroelement (either nitrogen, phosphorus or sulfur) tion. In the case of nitrogen and phosphorus limitation, the

770 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011

Figure 3. The effect of mean light intensity (D) on changes in biomass concentration (DW in g /L) (A), starch content in g/L (B) and relative starch content (% of DW) (C) in
asynchronous cultures of Chlorella. The cultures were grown in a laboratory photobioreactor continuously illuminated by a constant incident light intensity of 780 mmol/(m2 s1) and
at various biomass concentrations (numerals on curves indicate initial dry weight in g/L). Dashed horizontal lines indicate the biomass concentration (panel A) and the mean light
irradiance (panel D) at which the starch content begins to decrease.

treatment was severe enough to reduce biomass synthesis, Sulfur Limitation and Light–Dark Alteration
from the very beginning of the experiment and to stop it
The results of the preceding section show that to apply
completely after 12 and 20 h, respectively. In the case of
element limitation with large scale-up cultures for starch
nitrogen limitation, a slow decrease in biomass concentra-
overproduction in commercially attractive amounts, it
tion occurred due to cell death (Fig. 5A). In both
would be necessary to induce starch synthesis in high-
experimental variants, starch began to break down, and
density cultures. This is because with element limitation, cell
its concentration (in g/L) decreased as well as its relative
growth and division are slowed, and finally stop completely.
content (% of DW) (Fig. 5B and C).
Large-scale cultivation usually takes place in outdoor solar
In the case of sulfur limitation, the rate of the starch
photobioreactors under an illumination regime of alternat-
synthesis slowed after 12 h of growth (Fig. 5B) in parallel
ing light (day) and dark (night) periods. Photosynthetically
with the slowing of total biomass synthesis (Fig. 5A). The
active radiation is the sole source of energy for microalgae
maximum relative starch content was higher than in
grown under autotrophic conditions, so cellular main-
untreated cells, but in contrast to them, remained constant
tenance energy during the dark period must be obtained
until the end of the experiment (compare Fig. 5A–C, dashed
from intracellular storage compounds. We have tested
curves). From the point of possible application in large-scale
untreated (Fig. 6A–C) and sulfur limited (Fig. 6D–F)
culture units, the most promising results were those
asynchronous cultures at three different starting biomass
obtained under sulfur limitation. Relative starch content
concentrations grown at the same incident light intensity for
reached the highest level of all three experimental variants
two light/dark cycles (12 h/12 h) (Fig. 6).
(60% of DW) (Figs. 5C and 3D) and was stable for at least
the next 15 h (Fig. 5B) over which time period, growth of the
Untreated Cultures
culture stopped. Even after this long sulfur limitation, the
cells retained viability and were able to fully recover after The cultures grown under alternating light/dark intervals
sulfate readdition (data not shown). characteristically decrease biomass concentration during the

Brányiková et al.: Microalgae for Starch 771

Biotechnology and Bioengineering
Figure 4. Effect of cycloheximide (1 mg/L) treatment on changes in biomass
concentration (DW in g /L) (A), starch content in g/L (B), and relative starch content
(% of DW) (C) in asynchronous cultures of Chlorella. The cultures were grown in a
laboratory photobioreactor at a constant incident light intensity of 780 mmol/(m2 s1)
and various biomass concentrations (numerals on curves indicate initial dry weight
in g/L).
Figure 5. Effect of nitrogen, phosphorus, or sulfur limitation on changes in
biomass concentration (DW in g/L) (A), starch content (g/L) (B), and relative starch
content (% of DW) (C) in asynchronous cultures of Chlorella. Dashed horizontal line
(panel A) indicates the biomass concentration at which the gradual decrease in the
dark period. This decrease is mainly due to extensive rate of starch synthesis starts. The cultures were grown in a laboratory photobior-
degradation of starch (Fig. 6B) and consequently, the eactor with continuous illumination at an incident light intensity of 780 mmol/(m2 s1) in a
complete mineral medium (untreated) or in nitrogen (-N), phosphorus (-P), or sulfur (-S)
relative content of starch falls during the dark interval to limited media. The cultures started growth at the same initial biomass concentrations
very low levels (Fig. 6C). The decrease in starch synthesis (0.1 g/L).
during the light period, as a consequence of decreasing mean

772 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011

Figure 6. Effect of alternating light and dark periods on changes in biomass concentration (DW in g/L) (A and D), starch content (g/L) (B and E) and relative starch content (%
of DW) (C and F) in untreated (A–C) and in sulfur limited (D–F) cultures of Chlorella. The cultures were grown in a laboratory photobioreactor at an incident light intensity of 780 mmol/
(m2 s1) and various biomass concentrations (numerals at curves indicate initial dry weight in g/L). Dark periods are marked by black stripes and separated by vertical solid lines.

light intensity, contributes to the very low total and relative Sulfur limited cultures
starch contents (Fig. 6B and C).
In the most diluted untreated cultures, the starch content Using the same biomass concentrations, three experimental
reached only 36% of DW during the first 12 h and decreased variants were grown under the same growth conditions as
to almost the initial starting value during the first dark untreated cultures, but in sulfur limited mineral medium.
period (Fig. 6C). The second cell cycle in untreated cultures The cultures grew slower than untreated cultures, but no
was already affected by a low mean irradiance caused by significant decrease in biomass concentration occurred
increasing cell biomass (Fig. 6A), and the maximum content during the dark interval (compare Fig. 6A and D). Sulfur
of starch during the light period decreased substantially limitation stopped cell division, causing the starch content
(18% of DW) (Fig. 6C). to increase in the light more than in untreated cultures, but

Brányiková et al.: Microalgae for Starch 773

Biotechnology and Bioengineering
 varied from 188C at night to 308C at noon. It was rainy and
cloudy during the 3rd day of culture, causing a very low
change in biomass concentration and starch content
(Fig. 7A). On day 1 of cultivation, the starch content
increased to 30% of the DW. This was caused by the high
mean illumination and relatively low culture density (1.3 g/
L) of the culture. Most of this starch, however, was degraded
during the following night, when respiration and cell
division took place. Over the following 3 days of cultivation,
the average starch content varied linearly, with a slight
increase to a value of about 1g/L (Fig. 7A, dotted line). In
addition, the relative starch content oscillated between 18%
in the evening and 10% in the morning (Fig. 7B).
To prevent exhaustion of mineral medium, fresh medium
was added on the 4th day in an amount that corresponded to
the ‘‘sulfur limiting medium’’ used for laboratory cultiva-
tion. The starch content started to increase significantly once
sulfur limitation occurred on the 6th day (Fig. 7, see the
linear regression lines for the untreated and sulfur limited
cultures). The increase during the day was higher and the
decrease during the night lower than during the preceding
growth period in mineral replete medium (Fig. 7A and B,
dotted curves). In spite of the fact that the outdoor
experiment had to be terminated at the 156th hour because
of poor weather conditions, the findings provide evidence
that a relative starch content of at least 50% of the DW of the
algal biomass (12 g/L) can be attained.
Figure 7. Changes in biomass concentration (DW in g /L), starch content (g/L)
(A), and relative starch content (% of DW) (B) in cultures of Chlorella. The cultures
were grown in an outdoor scale-up thin layer photobioreactor in complete mineral
medium for 96 h and then in sulfur limiting medium (-S). Dark periods (nights) are Discussion
marked by black stripes and separated by vertical solid lines. Linear regression curves
of the 1st order (solid lines) were fitted to the experimental data separately for the To produce starch economically, conditions for culturing
periods of untreated and sulfur limited cultivation. starch-enriched algae in dense cultures must be attained.
Using a thin layer algal suspension in outdoor cultures,
linear growth continues up to very high biomass concen-
it did not decrease markedly during the dark interval. trations (about 40 g/L) enabling easy and cheap harvesting
Consequently, at the end of the second dark interval, the and processing (Doucha and Lı́vanský, 2006, 2009).
starch content was about 2 g/L, while in untreated cultures, However, the content of starch in the biomass is low
the value was about 0.4 g/L (compare Fig. 6B and E). (15% of DW or less).
Relative starch content during the light period was In the present experiments, we identify conditions under
significantly higher in sulfur limited compared to untreated which starch content increases in commercially produced
cultures (compare Fig. 6C and F). From an initial relative algal biomass to a level that would be viable for bioethanol
starch content of 8% of DW, it increased up to 60% of DW production. To achieve this, the processes and events during
during the first 12 h, but the loss of starch during the which starch is extensively degraded should be slowed down,
following dark period decreased by only 8–52% of total DW. or stopped completely, while the factors supporting starch
Moreover, the relative starch content remained high even synthesis, namely light intensity, should be sustainable.
during the second day (Fig. 6F). Using cycloheximide to inhibit cytoplasmic protein
synthesis, nuclear and cellular division, and related
processes, we have shown that the relative content of starch
Pilot-Scale Experiment
(% of DW) increased markedly (Fig. 4). However, the
To confirm the potential application of laboratory findings practical large-scale use of cycloheximide, particularly in
to large-scale commercial devices, the algae were grown outdoor bioreactors, is unrealistic both environmentally and
using a pilot scale-up outdoor open solar photobioreactor. economically.
The culture was grown for 4 days in complete mineral In contrast, essential macroelement limitation is a more
medium. Over the entire experiment, the solar light promising and practical approach. The results for both
intensity varied in the range from 2,000 to 2,600 mmol/ synchronous and asynchronous cultures of Chlorella
(m2 s1) during the day, and the average temperature of 228C confirmed that as in the case of cycloheximide application,

774 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011

energy requirements from starch for events like nuclear and out. The present findings on the effect of sulfur limitation
cellular division are blocked, while starch synthesis is under field conditions using an outdoor scale up, thin-layer
allowed to continue (Fig. 5). solar photobioreactor demonstrated the overproduction of
This provides a tool for the production of starch enriched starch in microalgal biomass.
commercially usable algal biomass. It is in line with findings An increase in the production of starch in sulfur-limited
with the chlorococcal alga Scenedesmus quadricauda that culture up to a maximum of 50% starch content of algal
macroelement limitation inhibited DNA replication, biomass (DW) was demonstrated under field conditions
nuclear and cellular division, and protein synthesis within using the outdoor scale up, thin-layer solar photobioreactor.
one cell cycle while starch synthesis was not substantially Despite the relatively unfavorable climatic conditions of
affected in cultures limited for nitrogen (Ballin et al., 1988; Trebon (Czech Republic), a total of 7 tones of starch per
Šetlı́k et al., 1988; Zachleder et al., 1988). hectare was produced from a photoautotrophic alga culture,
Nitrogen limitation is often used for increasing produc- over a season of 150 days (Fig. 7). In optimum locations for
tion of starch (Eriksen et al., 2007; Maeda et al., 2006; phototrophic production of algae like Greece, with a season
Rodolfi et al., 2009). A disadvantage of nitrogen or lasting approximately 250 days, the overall harvest might be
phosphorus limitation is that while the cells succeeded in increased by a factor of 10 (Doucha and Lı́vanský, 2006).
increasing starch content before dying, their viability This can be compared to corn, which usually only produces
decreased within a relatively short time interval due to about 4 tones of starch per hectare, in a similar or even more
effects on cell metabolism (Fig. 5). Cell lysis due to nitrogen favorable climate (Chisti, 2007).
limitation is in line with findings by other authors for The cultures must be supplied with carbon dioxide, and it
Chlorella and Chlamydomonas (Eriksen et al., 2007). Thus, has been shown that the direct use of flue gas is possible
from a practical standpoint, in order to generate starch (Douskova et al., 2009). Carbon dioxide sequestration can
enriched biomass in large-scale production, it will be therefore be regarded as an added benefit (Benemann, 1997;
necessary to achieve the longest interval between attainment Brown, 1996; Doucha et al., 2005).
of maximum starch content and cell death due to the lack of
a macroelement. The present findings suggest that sulfur This work was supported by the projects EUREKA of the Ministry of
limitation offers just such a process. Education, Youth and Sports of the Czech Republic (no. OE221 and
no. OE09025).
While use of algae with enriched starch content is
conventional for bioethanol production, another attractive
exploitation of starch from algae might be the production of References
hydrogen, which may be realized in the near future Ballin G, Doucha J, Zachleder V, Šetlı́k I. 1988. Macromolecular syntheses
(Chochois et al., 2009; Melis et al., 2004; Miura et al., and the course of cell cycle events in the chlorococcal alga Scenedesmus
1982; Tsygankov et al., 2002). Sulfur limitation could be one quadricauda under nutrient starvation: Effect of nitrogen starvation.
of the ways to support hydrogen production (Melis et al., Biol Plant 30:81–91.
2000; Zhang et al., 2002). Sulfur deprivation is considered as Benemann JR. 1997. CO2 mitigation with microalgae systems. Energy
Convers Mgmt 38:S475–S479.
an efficient way to trigger long-term hydrogen photo- Brown LM. 1996. Uptake of carbon dioxide from flue gas by microalgae.
production in unicellular Chlamydomonas by a decrease in Energy Convers Mgmt 37:1363–1367.
PSII, which allows anaerobiosis to be reached and starch Chader S, Hacene H, Agathos SN. 2009. Study of hydrogen production by
storage (Chochois et al., 2009). It has been shown recently three strains of Chlorella isolated from the soil in the Algerian Sahara.
Int J Hydrogen Energy 34:4941–4946.
that some strains of Chlorella can produce and accumulate
Chisti Y. 2007. Biodiesel from microalgae. Biotechnol Adv 25:294–306.
significant volume of hydrogen gas under anaerobic Chochois V, Dauvillée D, Beyly A, Tolleter D, Cuiné S, Timpano H, Ball S,
conditions and sulfur deprivation such as it is reported in Cournac L, Peltier G. 2009. Hydrogen production in Chlamydomonas:
literature using C. reinhardtii (Chader et al., 2009). Photosystem II-dependent and independent pathways differ in their
On a large scale, at first sight, the potentially outstanding requirement for starch metabolism. Plant Physiol 151:631–640.
algal biomass per hectare could be a convincing argument for Doucha J, Lı́vanský K. 2006. Productivity, CO2/O2 exchange and hydraulics
in outdoor open high density microalgal (Chlorella sp.) photobior-
the production of starch by macroelement limitation. For eactors operated in a Middle and Southern European climate. J Appl
example, in pilot tests, microalgal content per hectare compare Phycol 18:811–826.
well with traditional agricultural crops (Rodolfi et al., 2009), Doucha J, Lı́vanský K. 2009. Outdoor open thin-layer microalgal photo-
and much has been written about their potential—so far bioreactor: Potential 5 productivity. J Appl Phycol 21:111–117.
theoretical—to produce 200–300 tones of algal biomass per Doucha J, Straka F, Lı́vanský K. 2005. Utilization of flue gas for cultivation
of microalgae (Chlorella sp.) in an outdoor open thin-layer photo-
hectare (Chisti, 2007). Such projections are, however, often bioreactor. J Appl Phycol 17:403–412.
extrapolated from laboratory experiments and are therefore Douskova I, Doucha J, Livansky K, Machat J, Novak P, Umysova D,
unrealistic, particularly at the level of environmentally Zachleder V, Vitova M. 2009. Simultaneous flue gas bioremediation
sustainable light intensities. Experimentation using large-scale and reduction of microalgal biomass production costs. Appl Microbiol
production units to verify laboratory findings is the only way to Biotechnol 82:179–185.
Eriksen NT, Riisgård FK, Gunther WS, Lønsmann Iversen JJ. 2007. On-line
demonstrate real outcomes of any proposed approach, but estimation of O(2) production, CO(2) uptake, and growth kinetics of
studies on large-scale production of starch enriched algal microalgal cultures in a gas-tight photobioreactor. J Appl Phycol 19:
biomass as a feedstock for bioethanol have not been carried 161–174.

Brányiková et al.: Microalgae for Starch 775

Biotechnology and Bioengineering
Klass DL. 2004. Biomass for renewable energy and fuels. In: Clevel CJ, Rodolfi L, Zittelli GC, Bassi N, Padovani G, Biondi N, Bonini G, Tredici
editor. Encyclopedia of energy, Vol. 1 Amsterdam: Elsevier, Inc. p 193– MR. 2009. Microalgae for oil: Strain selection, induction of lipid
212. synthesis and outdoor mass cultivation in a low-cost photobioreactor.
Maeda I, Seto Y, Ueda S, Cheng Y, Hari J, Kawase M, Miyasaka H, Yagi Biotechnol Bioeng 102:100–112.
K. 2006. Simultaneous control of turbidity and dilution rate through Šetlı́k I, Ballin G, Doucha J, Zachleder V. 1988. Macromolecular syntheses
adjustment of medium. Biotechnol Bioeng 94:722–729. and the course of cell cycle events in the chlorococcal alga Scenedesmus
Mann G, Schlegel M, Schumann R, Sakalauskas A. 2009. Biogas-condition- quadricauda under nutrient starvation: Effect of sulphur starvation.
ing with microalgae. Agronomy Res 7:33–38. Biol Plant 30:161–169.
McCready RM, Guggolz J, Silviera V, Owens HS. 1950. Determination of Tsygankov A, Kosourov S, Seibert M, Ghirardi ML. 2002. Hydrogen
starch and amylose in vegetables. Anal Chem 22:1156–1158. photoproduction under continuous illumination by sulfur-deprived,
Melis A, Zhang L, Forestier M, Ghirardi ML, Seibert M. 2000. Sustained synchronous Chlamydomonas reinhardtii cultures. Intl J Hydrogen
photobiological hydrogen gas production upon reversible inactivation Energy 27:1239–1244.
of oxygen evolution in the green alga Chlamydomonas reinhardtii. Plant Zachleder V, Ballin G, Doucha J, Šetlı́k I. 1988. Macromolecular syntheses
Physiol 122:127–135. and the course of cell cycle events in the chlorococcal alga Scenedesmus
Melis A, Seibert M, Happe T. 2004. Genomics of green algal hydrogen quadricauda under nutrient starvation: Effect of phosphorus starvation.
research. Photosynth Res 82:277–288. Biol Plant 30:92–99.
Miura Y, Yagi K, Shoga M, Miyamoto K. 1982. Hydrogen production by a Zhang L, Happe T, Melis A. 2002. Biochemical and morphological char-
green alga, Chlamydomonas reinhardtii, in an alternating light/dark acterization of sulfur-deprived and H2-producing Chlamydomonas
cycle. Biotechnol Bioeng 24:1555–1563. reinhardtii (green alga). Planta 214:552–561.

776 Biotechnology and Bioengineering, Vol. 108, No. 4, April, 2011