Aquatic Ecology 34: 119–126, 2000.

© 2000 Kluwer Academic Publishers. Printed in the Netherlands.

119

Extraction of pigments and fatty acids from the green alga Scenedesmus
obliquus (Chlorophyceae)

Karen H. Wiltshire, Maarten Boersma, Anita Möller and Heinke Buhtz

Max-Planck-Institut für Limnologie, Postfach 165, D-24302 Plön, Germany (Tel: +49 4522 763228; Fax:
+49 4522 763310; E-mail: [email protected])

Accepted 9 May 2000

Key words: carotenoids, chlorophyll, GC, HPLC, lipids

Abstract
In this paper, the efficiency of pigment and fatty acid extraction from resistant algae using Scenedesmus obliquus
as an example was examined. We found that adding quartz sand and solvent to freeze-dried algal material and
subsequent extraction in an ultrasound bath for 90 min at −4 ◦ C resulted in excellent extraction of these compounds.
This extraction method was compared with a method regularly used for extraction of fatty acids and pigments, i.e.
addition of solvents to algal material with subsequent incubation. Our extraction using the ultrasound and sand
method was about twice as efficient as this method for both pigments and fatty acids. The ultrasound method
is simple, extracts over 90% of the different substances in one step and conserves the relationships of pigments
and fatty acids. In addition, no alteration- or breakdown products were observed with the new method. Thus,
this method allows accurate quantitative extraction of both pigments and fatty acids from Scenedesmus obliquus
and other algae. The method was also been found to be as effective for Cryptomonas erosa (Cryptophyceae), Cy-
clotella meneghiniana (Bacillariophyceae), Microcystis aeruginosa (Cyanophyceae), and Staurastrum paradoxum
(Chlorophyceae, Desmidiaceae) and is thus applicable to a wide spectrum of algae.

Introduction and especially so, where the pigment and fatty acid
composition of algae are concerned. Pigments are of
Green algae, particularly members of the genus fundamental importance in algal cell physiology (De-
Scenedesmus, have become the equivalent of labora- venter & Heckman, 1996), and lipids or fatty acids
tory rats in many fields in limnology. They are com- are important components of algal nutritional value
monly used as standard organisms in numerous areas (Drevon et al., 1993).
of aquatic research, technology, and water manage- Substantial problems associated with extraction
ment (e.g., Zachleder et al., 1986). Apart from its om- and analysis complicate the determination of the pig-
nipresent role in research on algal growth, morphology ment composition of algae (Wright et al., 1997;
and life cycles (e.g., Trainor, 1995), Scenedesmus is Wiltshire et al., 1998). The analytical problems are
also one of the most popular food sources in exper- well documented, and it is now widely accepted that
iments with herbivorous zooplankton (e.g., Boersma High Performance Liquid Chromatography (HPLC)
& Vijverberg, 1995). Moreover, it has also become is the only accurate means of quantifying chloro-
the standard alga in the growing research area on in- phyll and other photosynthetic pigments in aquatic
ducible defences, where researchers use this alga to systems (AWWA-APHA, 1985; Wright et al., 1997;
study its reaction to the presence of chemicals excreted Wiltshire et al., 1998). However, exact and uncompli-
by its predator (e.g., Lürling & van Donk, 1997; Wilt- cated extraction of pigments still proves difficult. In
shire & Lampert, 1999). In many of these research an excellent review Wright et al. (1997), have recom-
applications, the quantitative determination of the bio- mended extraction methods for pigments from marine
chemical composition of algal cells is very important algae. However, these methods are still not adequate
120

for many of the biochemical and physiological ques- counted lysed) throughout this work. After trying sev-
tions involving Scenedesmus, where exact quantitative eral quite laborious alternatives (e.g., hand grinding
information is required. or French Press, Table 1), we chose the following
The subject of lipid analysis has been extensively method, for its ease of use and its lysis percentage of
reviewed in the literature (Christie, 1982; Parrish, over 90%. We centrifuged 2–10 ml of algal suspen-
1999), and many different preparative and analytical sion, at 4 ◦ C for 10 min at 4000 RPM. After checking
techniques have been published for lipids and fatty for errant cells, the overlying water was carefully
acids (Folch et al., 1956; Bligh & Dyer, 1959; Kattner removed, and the algal pellet frozen at −80 ◦ C. Sub-
& Fricke, 1986). However, the classical work on lipid sequently, the algae were freeze-dried in the dark, and
extraction efficiency was carried out on relatively large the pellet weighed. The algae were freeze-dried, as the
samples of animal tissue. (Folch et al., 1956; Bligh & presence of water can cause the breakdown of chloro-
Dyer, 1959). Given the absence of cell walls in ani- phylls via chlorophyllase (Wright et al., 1997). Di-
mal material, blending or mixing samples during the rectly after drying, the algae were covered with the ap-
extraction sufficed for quantitative lipid extraction. In propriate solvents (pigments: 1 ml of 100% nanograde
contrast, the extraction procedures and efficiencies for acetone for pellets under 0.5 mg, 3 ml for heavier pel-
plant material, especially for algae, are less well es- lets; fatty acids 4 ml of 2:1 dichloromethane/methanol
tablished. Indeed, the literature on fatty acid and lipid mixture), and 0.2 g of analytical grade quartz (particle
content of microalgae contains no standard extraction size 10–30 µm) added. The samples were subse-
method. Published extraction methods include the ad- quently placed in an ultrasound bath (35 kHz; 80 W)
dition of solvents and subsequent incubation (Ackman at −4 ◦ C (using saltwater), and sonicated for 90 min
et al., 1968), agitation (Cartens et al., 1996), stir- (shorter sonication times resulted in lower lysis per-
ring (Whyte, 1988), homogenisation (Ben-Amotz et centages). This ‘maximum extraction’ method was
al., 1985), grinding (Rai et al., 1997), and sonication compared with the method where the appropriate sol-
(Napolitano, 1994). vent was added to freeze-dried algae, shaken manually
In short, no standard technique exists to quantita- for one minute (Ahlgren & Merino, 1991), and left
tively extract pigments and fatty acids from microal- standing. All extractions were followed up with a sec-
gae. Moreover, the choice of Scenedesmus as the stan- ond extraction. The samples were centrifuged, the
dard alga for the applications mentioned above seems original solvent removed, and new solvent added to
unfortunate. Species from this genus have particularly establish the percentage of the pigments and fatty
resistant cell walls (Bisalputra & Weier, 1963; van acids remaining in the algae. The experimental pro-
Donk et al., 1997) and the extraction of pigments cedures of the second extractions were identical to
and fatty acids is notoriously difficult (Wood, 1985; the first extractions. Relevant controls (with only sand
Mouget et al., 1993). Consequently, we set out to and solvent) were also analysed. All extractions were
develop a method for the complete and quantitative done in quintuplet. Once the maximum extraction
extraction of pigments and fatty acids from the green method was established for the chemostat culture of
alga S. obliquus. The criteria we used were maximum Scenedesmus, we applied it to batch cultures of other
extraction efficiency in one step, ease of handling, and algae: Cryptomonas erosa (Cryptophyceae) (CE),
use of solvents of low toxicity. The methods were sub- Cyclotella meneghiniana (Bacillariophyceae) (CM),
sequently tested with different algal species to assess Microcystis aeruginosa (Cyanophyceae) (MA), and
their general applicability. Staurastrum paradoxum (Chlorophyceae, Desmidi-
aceae) (SP). Duplicate samples were taken, and the
extraction efficiencies checked against the control
Materials and methods method.
Many different solvents have been compared for
Scenedesmus obliquus (strain no: SAG 276-3a, Göt- the extraction of pigments in marine algae (Wright
tingen culture collection) was cultured in chemostat et al., 1997) with the general conclusion being that ex-
cultures at 20 ◦ C and at 300 µmol m−2 s−1 of light traction with dimethylformamide (DMF) was the most
in WC (Woods Hole MBL) culture medium. The lysis efficient. However, due to its toxicity this solvent can-
of cell membranes and cell walls is the rate-limiting not be recommended. Although alcohols (methanol
step in the extraction of pigments from Scenedesmus and less frequently ethanol) are usually better sol-
and was determined under the microscope (% of cells vents for extraction than acetone, these are known
 121
Table 1. Initial experiments to establish the percentage lysis of Scenedesmus
obliquus cells. All cells were frozen. For all treatments the algae were frozen
both at −180 ◦ C and at −10 ◦ C. No differences were found between the two
temperatures. The percentage of lysed cells is indicated, with the time of exposure
to the break-up method in minutes in brackets. The percentage of lysed (ruptured)
cells was determined microscopically

Acetone added (ml)

1.5 3 5

Hand grinding – 40–60 (10) 20–40 (10)

Mechanical grinding – 40–60 (10) 20–40 (10)
Ultrasound microtip 70 (8) 50–70 (8) 50 (8)
Ultrasound microtip & sand – 70–80 (5) 70–80 (5)
Ultrasound bath −4 ◦ C <20 (90) <20 (90) <40 (90)
Ultrasound bath & sand −4 ◦ C >90 (90) >90 (90) >90 (90)

to promote the formation of allomers of chlorophyll were checked against the spectra from the diode ar-
(Strain & Svec, 1966; Bowles et al., 1985). A mix- ray detector. The instrument was calibrated for the
ture of 90% acetone and 10% water has been the relevant xanthophylls and chlorophyll pigments us-
solvent of choice in many studies, but chlorophyl- ing a five-point calibration every 100 samples with
lase activity is still substantial in this mixture, hence commercial standards in 100% acetone (chlorophyll
100% acetone is the preferred solvent. The HPLC a and b; Sigma) with a regression coefficient of be-
method used was optimised for the separation of the tween 0.98 and 0.99 in the range of 0.01 mg l−1
xanthophylls, chlorophylls and carotenes. Extracts and 5 mg l−1 . A two-point calibration was carried
were filtered through a 0.45 µm pore-size cellulose out every 15 samples. The accuracy and purity of the
filter and then 60 µl of sample was packed in be- commercial standards were checked using spectropho-
tween two 20 µl water ‘plugs’ (Villerius et al., 1996) tometric measurements. The absolute detection limit
in the injection loop and injected in duplicate via a of the system was 0.08 µg l−1 of chlorophyll in ace-
cooled autosampler straight into an HPLC system. tone extract. The instrument standard error is no more
This system consisted of a low-pressure pump and than 1% for five replicate measurements.
autosampler (Waters Alliance), a column oven and Most current fatty acid extraction methods are
a diode array detector (Waters 996). The flow rate based on the extraction methods of Bligh & Dyer
used was 1 ml min−1 , the column used was a re- (1959), where mixtures of chloroform and methanol
versed phase 5C18 , (Vertex, Knauer) column, 25 cm have been used to extract lipids. As there is grow-
long. This column was kept thermostated at 15 ◦ C in ing concern about the potential health hazards as-
a column oven. The gradient required three solvent sociated with the use of chloroform, we substituted
mixtures: A (80:10:10 methanol:water:ammonium ac- the chloroform with the less toxic dichloromethane
etate), B (90:10 methanol:acetone), and C (10:7.7 (methylene chloride) (Chen et al., 1981; Parrish &
methanol:propanol). The initial solvent was A, after Wangersky, 1987). Dichloromethane is also more
5 min, this was changed to a 1:1 A:B mixture, held volatile than chloroform and evaporation of the sol-
there for 5 min, thereafter changed linearly 100% B at vent is more easily achieved. To avoid auto-oxidation
15 min. Then the solvent was changed back to 100% of the unsaturated fatty acids, 200 mg l−1 of buty-
A at 27 min and then this converted linearly to 100% lated hydroxytoluene (BHT; Sigma) was added to the
C. After 29 min C was reduced back to A. Finally, CH2 Cl2 /MeOH mixture (Christie, 1982). Kattner &
after 35 min the system was set back to the initial Fricke (1986) showed that for fatty acids a washing
solvent conditions (A). All solvents were degassed step with a NaCl solution (Folch et al., 1956) is not es-
nanograde HPLC solvents (Baker). The identification sential. Therefore, the extraction mixture with the dis-
of the pigments was carried out using retention times solved lipids was evaporated to dryness under N2 , and
(Francis et al., 1973; Fawley, 1991) in combination trans-esterified with 2 ml of 3% H2 SO4 in methanol
with commercial standards (VKI & Sigma), which (Kattner & Fricke, 1986) for four hours at 70 ◦ C. The
122

samples were then cooled, and shaken twice with 2 ml

of iso-hexane. The solvent was evaporated, and 50 µl
of iso-hexane were added.
The Fatty Acid Methyl Esters (FAMEs) were
analysed on a Hewlett Packard 5890 gas chromato-
graph with a HP-225 silica-fused column (30 m;
0.25 mm I.D.). The sample volume was 1 µl; the
temperature of the injector was 175 ◦ C. We used a
splitless injection technique in a split/splitless liner,
with a purge after 40 s. The temperature program
was one minute at 45 ◦ C; then a temperature increase
to 180 ◦ C at a rate of 25 ◦ C min−1 , followed by a Figure 1. Comparison of the extraction efficiency of the ultrasound
3 ◦ C min−1 increase to 220 ◦ C and finally a hold at and passive (control) extraction for chlorophyll a., and total fatty
220 ◦ C for 18 min. Helium was used as the carrier acids from Scenedesmus obliquus. Error bars indicate standard er-
gas (1 ml min−1 ). The temperature of the flame- rors (n = 5). The amounts extracted in the second extractions (open
bars) are stacked on the amounts extracted in the first extraction
ionisation detector was 250 ◦ C. FAMEs were identi- (filled bars).
fied by the comparison of their retention times with
retention times from single standard FAMEs. As in-
ternal standards we used a series of odd-chained fatty
acids (C13:0–C21:0; Restek). Differences in response
factors of the detector for different FAMEs were es-
tablished with the help of quantitative mixtures of
different FAMEs (Supelco). HP-chemstation software
was used to identify and quantify different FAMEs.
We analysed our data in two way ANOVAs with
treatment (ultrasound-control), and extraction (First-
Second) as fixed factors, and the different pigments
and fatty acids as the dependent variables. Because
within one sample the different compounds are not in-
dependent, we used a repeated measure design, where
all compounds were analysed in one ANOVA. Signif-
icant interactions of treatment with compounds would Figure 2. Amounts of different pigments extracted using the two
indicate that the different compounds are extracted extraction techniques from Scenedesmus obliquus. All differences
between first and second extraction are highly significant (Table 2).
with different efficiencies. To corroborate this, we sub- Error bars indicate standard errors (n = 5). The amounts extracted
sequently carried out the same analysis with ratios to in the second extractions (open bars) are stacked on the amounts
chlorophyll a (pigments), or total fatty acids. extracted in the first extraction (filled bars).

culture of S. obliquus contained β-carotene. We ob-

Results
served a highly significant effect of all of the effects
in the ANOVA (Table 2a). The extraction efficiency
The main carotenoid pigments present in S. obliquus
of the ultrasound method was significantly higher
were neoxanthin, loroxanthin, violaxanthin, and
than that of the control treatment. For example, for
lutein. The main porphyrin pigments present were
chlorophyll a, the first extraction using the ultrasound
chlorophyll a, b and their allomers. No chlorophyll
method yielded more than double the amount of the
breakdown products (chlorophyllides and phaeopig-
control treatment (Figure 1). The other pigments were
ments) were found. In the continuous cultures of
also extracted to a significantly higher degree using the
Scenedesmus obliquus we observed no carotenes, al-
ultrasound treatment (Figure 2). In the second extrac-
though these are usually found in green algae. The
tion, a significant difference between the two methods
lack of carotenes is not an artefact of extraction but
was found for one substance only, caused by the fact
was specific to the culturing conditions. In fact, in the
that most of the pigments were extracted in the first
experiment with the different algal species, the batch
 123
Table 2. Summary ANOVA tables of two way ANOVAs with treatment and extraction as fixed factors, and
the different substances (pigments, fatty acids) as repeated measures of one sample. Absolute amounts as well
as the ratio to chlorophyll a (pigments) or total fatty acid amount (fatty acids) were analysed

A. Pigments Absolute amounts Ratios to chlorophyll a

MS df F P MS df F P

Treatment 212620 1 25.7 <0.001 0.130 1 29.4 <0.001

Extraction 3458335 1 417.3 <0.001 0.003 1 0.7 0.40
Pigment 975727 5 659.8 <0.001 0.387 4 176.5 <0.001
Treat × Extract 259371 1 31.3 <0.001 0.005 1 1.1 0.31
Treat × Pigment 35894 5 24.3 <0.001 0.047 4 21.5 <0.001
Extract × Pigment 718933 5 486.2 <0.001 0.001 4 0.5 0.77
3-way interaction 44935 5 30.4 <0.001 0.002 4 0.8 0.52
Error 1479 80 0.002 64

B. Fatty acids Absolute amounts Ratios to total fatty acids

MS df F P MS df F P

Treatment 16.8 1 29.5 <0.001 0.0001 1 0.3 0.60

Extraction 166.6 1 293.2 <0.001 0.0001 1 5.8 0.03
Fatty acid 42.5 12 167.5 <0.001 0.2094 12 276.5 <0.001
Treat × Extract 31.5 1 55.4 <0.001 0.0001 1 0.3 0.60
Treat × Fatty acid 2.6 12 10.2 <0.001 0.0005 12 0.6 0.83
Extract × Fatty acid 25.0 12 98.5 <0.001 0.0078 12 10.4 <0.001
3-way interaction 5.0 12 19.9 <0.001 0.0012 12 1.6 0.10
Error 0.3 168 0.0007 168

area of chlorophyll a, the ubiquitous pigment. Table 2a

shows a significant treatment, pigment and interaction
(treatment × pigment) effect. This was mainly caused
by the higher extraction of chlorophyll a due to the
ultrasound method (Figure 1).
As was the case with the pigments, we observed
highly significant treatment, extraction, and com-
pound effects for the fatty acids (Table 2b). Higher
extraction efficiencies were found for the fatty acids
with the ultrasound-extraction (Figures 1 and 3). We
assessed changes in the fatty acid signature by com-
puting the ratios of single fatty acids to the total
amount of fatty acids found. Because of the fact that
substances with very low initial concentrations were
Figure 3. Amounts of different fatty acids extracted using the two
extraction techniques from Scenedesmus obliquus. All differences not detected in the second extraction, we observed a
between first and second extraction are highly significant (Table 2). significant extraction effect for the ratios of the fatty
Error bars indicate standard errors (n = 5). The amounts extracted acids to the total of the fatty acid pool. With the excep-
in the second extractions (open bars) are stacked on the amounts tion of one fatty acid (18:0), the amounts relative to the
extracted in the first extraction (filled bars).
total amount of fatty acids extracted showed no differ-
ences between the ultrasound method and the controls,
extraction with the ultrasound treatment and the con- resulting in a non-significant treatment effect for the
trol treatment extracted less. To assess the effect of the ratios. Hence, for fatty acids both methods extract the
different extractions on the total pigment signature, we same spectrum, but are quantitatively different.
computed the ratios of pigment peak areas to the peak
124

Figure 4. Amounts of different pigments extracted using the two

extraction techniques from different algae, exemplified for chloro- Figure 5. Amounts of unsaturated fatty (open bars) stacked on satu-
phyll a (open bars) stacked on β-carotene (hatched bars). Error rated (hatched bars) acids using the two extraction techniques from
bars indicate standard errors (n = 2). Other pigments found in different algae. Error bars indicate standard errors (n = 2).
the different species were: Alloxanthin (CE), Chlorophyll b (SO,
SP) , Chlorophyll c2 (CE), Crocoxanthin (CE), b-cryptoxanthin
(MA, SO), α-carotene (MA), Diadinoxanthin (CM), Diatoxanthin let alone substances in algal cells bound to mem-
(CM), Fucoxanthin (CM), Echinone (MA), Loroxanthin (SO, SP), branes and associated with protein complexes. Thus,
Lutein (SO, SP), Monadoxanthin (CE), Myxoxanthophyll (MA),
Neoxanthin (SO, SP), Violaxanthin (SO, SP), Zeaxanthin (MA).
in work on the pigment and fatty acid composition
of Scenedesmus and other algae under different envi-
ronmental conditions one cannot simply assume total
Figure 4 shows the extraction efficiencies of the extraction. Consequently, as the first rate-limiting step
ultrasound method relative to the control method, for in an extraction is cell lysis and having found a method
pigments in a variety of algae with different struc- that provided maximum cell lysis, we investigated the
tures. Different pigments were found in the differ- efficiency of pigment and fatty acid extraction using
ent algae; for all of them the extraction efficiencies two sequential extractions. Both absolute amounts and
were higher for the ultrasound method. This is ex- relative amounts were considered.
emplified by Figure 4 for β-carotene (filled bars) and The ultrasound and sand method enables us
chlorophyll a (open bars). For most of the identified to quantitatively extract pigments and fatty acids
pigments and fatty acids the extracted amounts were from the difficult-to-extract green alga Scenedesmus
higher with the ultrasound method. After Scenedesmus obliquus, (Chlorophyceae) (cf., Jespersen & Christof-
obliquus, the desmid Staurastrum paradoxum proved fersen, 1987). Even though Scenedesmus, and chloro-
to be most difficult to extract showing the greatest phytes in general, might be especially problematic, we
differences between the two methods for pigments, found, that even in the ‘easier’ algae substantial differ-
whereas for fatty acids the extraction from Microcystis ences existed between the methods used in this study.
proved very difficult (Figure 5). The extraction effi- Especially Staurastrum paradoxum (Chlorophyceae,
ciency of each compound also varied for the different Desmidiaceae) for pigments and Microcystis aerug-
algae. Chlorophyll c2 and oleic acid (18:1ω9) were inosa (Cyanophyceae) for fatty acids showed large
examples of substances, which extracted very badly differences between the methods. Hence, we do not
from Cryptomonas and Cyclotella, using the control agree with Ahlgren & Merino (1991), who stated that
method. freeze-drying breaks the cells, and turns algal mater-
ial into a loose, fine powder, making homogenisation
unnecessary.
Discussion The current literature (see review of Wright et al.,
1997) on the analyses of pigments, and particularly the
Under ideal extraction conditions, solid substances chlorophylls, points out that a primary criterion in the
should dissolve in a solvent rapidly and totally in a extraction of substances is their preservation. The pro-
short time period. However, this cannot be assumed duction of chlorophyllides and phaeo-pigments from
when dealing with complex mixtures of substances,
 125

chlorophyll (for example) is a common problem in may have been fine for one group of compounds but
extraction of algae. Therefore, no matter which ex- sub-optimal for the other.
traction method is applied to substrates containing In contrast to the pigment literature, multiple ex-
pigments (from algae through to sediments for ex- tractions have been regularly applied in the extraction
ample) the preservation of substances should always of lipids and fatty acids (Folch et al., 1956; Bligh &
be addressed and checked. We observed no alteration Dyer, 1959). However, many of the papers on fatty
products resulting from the ultrasound extraction (e.g., acid content of biological material are not specific as
phaeo-pigments, chlorophyllides and carotenoid alter- to the exact extraction procedures (e.g., omitting ex-
ation) at any wavelength (330–700 nm). Moreover, traction temperatures and duration, not reporting the
we did not observe any alteration products when stan- amount of extracted material). The original papers by
dards of pigments (chlorophyll a, b and lutein) were Folch et al. (1956) and Bligh & Dyer (1959) both
treated identically. This was also corroborated using a emphasise the point of sample homogenisation, and
Kontron SFM25 fluorescence detector. mention grinding for tougher tissues. Christie (1982)
Generally, the question as to the amount remain- already suggested the use of clean sand while ho-
ing in a cell after extraction of pigments seems largely mogenising difficult tissues, which would certainly
to be ignored in the literature. Although authors have include microalgae (Ahlgren & Merino, 1991). Our
been intent on developing effective extraction meth- results show that sonifying samples with sand as an
ods (e.g., Wright et al., 1997), it is often assumed abrasive substantially increases the amount of ex-
that one extraction is enough and rarely is the error tracted fatty acids from algal material, as illustrated by
associated with estimations of pigments remaining in the significant difference in extraction efficiency be-
biomass even considered when discussing quantitative tween the ultrasound and the non-ultrasound method.
results. The fact that not all substances extract at the Using the ultrasound and sand method, we were able
same rate is also generally simply assumed to be of to extract double the amount of fatty acids from algal
negligible importance. Of course, when developing an material and most of that in the first extraction.
extraction method ease of use (i.e., one extraction step) This work represents a major step forward in the
must be balanced against the fact that, for chemical extraction of pigments and fatty acids from an alga
reasons alone, in solute-substrate mixtures no single (Scenedesmus) which is notoriously difficult to ex-
step will ever be 100% efficient. Multiple extractions tract. The method is easy to use, allows the extraction
of pigments from algae are a rarity in the phycological of pigments and fatty acids highly quantitatively (over
literature and, when discussed at all, are inconclusive 90%) in one step, and conserves the relationships of
and confusing (see Wright et al., 1997). In develop- different pigments and fatty acids to one another. The
ing our method, we checked the extraction efficiency substances are conserved and do not breakdown in
of the methods using a double sequential extraction. the course of the extractions. Moreover, the method
Both extraction techniques extracted highly signifi- is effective for a wide range of other algae, and hence
cantly more pigments in the first extraction than the should be generally applicable in research on pigments
second extraction (ten to twenty times more). In terms and fatty acids.
of absolute amounts, the treatment with sand and ultra-
sound was more effective than the treatment without.
Between 92–96% of the total extracted amounts of Acknowledgements
each pigment were extracted using the sand and ul-
trasound treatment and as low as 70% (chlorophyll a) We thank Winfried Lampert for his support, Ivonne
in the treatment without ultrasound and sand. Conse- Harder for her help with the culture of the algae
quently, less was extracted in the second extraction and Martin Losch for help with part of the analy-
with the former technique, and more residual pig- ses. This research was partly supported by European
ments were removed in the second extraction in the Commission contracts ENV4-CT97-0402 and MAS3-
treatment without sand and ultrasound. There was no CT97-0158.
significant difference between the extraction efficiency
of carotenoids and chlorophylls (ratios) using either
method. This is reassuring, as both groups of com-
pounds exhibit different chemical properties and it
could have been possible that the extraction method
126

References Mouget JL, Legendre L and De La Noüe J (1993) Initial and total
activity of Rubisco in Scenedesmus ecornis: is the french press a
Ackman RG, Tocher CS and McLachlan J (1968) Marine phyto- suitable tool for enzyme extraction? Plant Cell Physiol 34: 281–
plankter fatty acids. J Fish Res Bd Can 25: 1603–1620 288
Ahlgren G and Merino L (1991) Lipid analysis of freshwater Napolitano GE (1994) The relationship of lipids with light and
microalgae: a method study. Arch Hydrobiol 121: 295–306 chlorophyll measurements in freshwater algae and periphyton.
AWWA-APHA (1985) Standard methods for the examination of J Phycol 30: 943–950
water and waste water. 16th edition. Washington DC. Parrish CC (1999) Determination of lipid, lipid classes, and fatty
Ben-Amotz A, Tornabene TG and Thomas WH (1985) Chemical acids in aquatic samples. In: Arts MT and Wainmain BC (eds),
profile of selected species of microalgae with emphasis on lipids. Lipids in Freshwater Ecosystems. Springer-Verlag, New York
J Phycol 21: 72–81 Parrish CC and Wangersky PJ (1987) Particulate and dissolved lipid
Bisalputra T and Weier TE (1963) The cell wall of Scenedesmus classes in cultures of Phaeodactylum tricornutum grown in cage
quadricauda. Am J Bot 50: 1011–1019 culture turbidostats with a range of nitrogen supply rates. Mar
Bligh EG and Dyer WJ (1959) A rapid method of total lipid Ecol Progr Ser 35: 119–128
extraction and purification. Can J Biochem Physiol 37: 911–917 Rai H, Arts MT, Wainman BC, Dockal N and Krambeck HJ (1997)
Boersma M and Vijverberg J (1995) Synergistic effects of dif- Lipid production in natural phytoplankton communities in a
ferent food species on life-history traits of Daphnia galeata. small freshwater Baltic lake, Lake Schöhsee, Germany. Freshwat
Hydrobiologia 307: 109–115 Biol 38: 581–590
Bowles ND, Paerl HW and Tucker J (1985) Effective solvents and Strain H and Svec WA (1966) Extraction, separation, estimation and
extraction periods employed in phytoplankton carotenoid and isolation of chlorophylls. In: Vernon LP and Seely GR (eds.) The
chlorophyll determinations. Can J Fish Aquat Sci 42: 1127–1131 Chlorophylls.
Cartens M, Grima EM, Medina AR, Gimenez AC and Gonzalez Trainor FR (1995) The sequence of ecomorph formation in a
JI (1996) Eicosapentaenoic acid (20:5n-3) from the marine mi- phenotypicly plastic, multispined Scenedesmus species (Chloro-
croalga Phaeodactylum tricornutum. J Am Oil Chem Soc 73: phyceae). Arch Hydrobiol 133: 161–171
1025–1031 van Donk E, Lürling M, Hessen DO and Lokhorst GM (1997) Al-
Chen IS, Shen CSJ and Sheppard AJ (1981) Comparison of tered cell wall morphology in nutrient-deficient phytoplankton
methylene chloride and chloroform for the extraction of fats from and its impact on grazers. Limnol Oceanogr 42: 357–364
food products. J Am Oil Chem Soc 58: 599–601 Villerius L, Wiltshire KH, Gieskes W and Paterson DM (1996)
Christie WW (1982) Lipid analysis: Pergamon, Oxford The use of a binary reversed phase, non-endcapped C30 method
Deventer B and Heckman CW (1996) Effects of prolonged darkness for the separation of algal pigments. International Carotenoid
on the relative pigment content of cultured diatoms and green Symposium of IUPAC, Leiden, The Netherlands.
algae. Aquatic Sci 58: 241–252 Whyte JNC (1988) Fatty acid profiles from direct methanolysis of
Drevon CA, Baksaas I and Krokan HE (eds.) (1993) Omega-3 Fatty lipids in tissue of cultured species. Aquaculture 75: 193–203
Acids: metabolism and Biological Effects. Birkhäuser, Basel Wiltshire KH, Harsdorf S, Smidt B, Blocker g, Reuter R and
Fawley MW (1991) Disjunct distribution of the xanthophyll lorox- Schroeder F (1998) The determination of algal biomass (as
anthin in the green algae (Chlorophyta). J Phycol 27: 544–548 chlorophyll) in suspended matter from the Elbe estuary and
Folch J, Lees M and Sloane-Stanley GH (1956) A simple method for the German Bight: a comparison of high-performance liquid
the isolation and purification of total lipids from animal tissues. chromatography, delayed fluorescence and prompt fluorescence
J Biol Chem 226: 497–509 methods. J Exp Marine Biol Ecol 222: 113–131
Francis GW, Knutsen G and Lien T (1973) Loroxanthin from Wiltshire KH and Lampert W (1999) Urea excretion by Daphnia:
Chlamydomonas reinhardii. Acta Chemica Scandinavica 27: A colony-inducing factor in Scenedesmus? Limnol Oceanogr 44:
3599–3600 1894–1903
Jespersen AM and Christoffersen K (1987) Measurements of Wood LW (1985) Chloroform-methanol extraction of chlorophyll a.
chlorophyll-a from phytoplankton using ethanol as extraction Can J Fish Aquat Sci 42: 38–43
solvent. Arch Hydrobiol 109: 445–454 Wright SW, Jeffrey SW and Mantoura RFC (1997) Evaluation of
Kattner G and Fricke HSG (1986) Simple gas-liquid chromato- methods and solvents for pigment extraction. In: Jeffrey SW,
graphic method for the simultaneous determination of fatty acids Mantoura RFC, and Wright SW (Eds) Phytoplankton Pigments
and alcohols in wax esters of marine organisms. J Chrom 361: in Oceanography: Guidelines to Modern Methods. UNESCO,
263–268 Paris
Lürling M and van Donk E (1997) Morphological changes in Zachleder V, Wittenburg E and Abarzua S (1986) Factors control-
Scenedesmus induced by infochemicals released in situ from ling the inhibitory effects of 3,4-benzo(a)pyrene on the chloro-
zooplankton grazers. Limnol Oceanogr 42: 783–788 coccal alga Scenedesmus quadricauda. Arch Hydrobiol Suppl
73: 281–296