Identification and quantification of cyanobacterial toxins (microcystins) in two Moroccan drinking-water reservoirs (Mansour Eddahbi, Almassira)

Abstract

Mansour Eddahbi (MED) (30°55′N, 6°53′W) and Almassira (ALM) (31°95′N, 6°72′W) are two Moroccan lake reservoirs located at
an arid and semi-arid hydrographic basin, respectively. Both are used for irrigation, recreational activities and drinking-water
production. This paper deals with the characterization and quantification of microcystins (MC) from two Microcystis aeruginosa blooms occurring in those reservoirs. The toxicity of the blooms was confirmed and evaluated by both mouse and Artemia bioassays. The calculated LD50 values revealed that the MED bloom had a medium toxicity (LD50 = 358mg kg − 1 body weight), whereas the ALM bloom had low toxicity (LD50 = 829mgkg − 1 body weight). The 24-h LC50 values were 1.88 and 4.15mg ml − 1 for the MED and ALM blooms, respectively, using Artemia assay. The identification and quantification of MC variants were carried out by high performance liquid chromatography (HPLC)
equipped with a photodiode array detector, and HPLC coupled to mass spectrometry. The MC content, as Microcystin-LR (MC-LR)
equivalents, was higher in the MED bloom (64.4μg g − 1 dry weight) than in the ALM bloom (9.9μg g − 1 dry weight). Five MC variants were identified in the MED cyanobacteria bloom (MC-RR, MC-YR, MC-LR, MC-FR, and MC-WR) and
only one (MC-LR) in the ALM bloom. The results show that the occurrence of toxic cyanobacteria blooms in the studied reservoirs
may be regarded as a health hazard; therefore, cyanotoxin monitoring in them is highly recommended.

Full text

Page 1

Environ Monit Assess (2010) 160:439–450
DOI 10.1007/s10661-008-0708-5
Identification and quantification of cyanobacterial toxins
(microcystins) in two Moroccan drinking-water reservoirs
(Mansour Eddahbi, Almassira)
M. Douma·Y. Ouahid·F. F. del Campo·
M. Loudiki·Kh. Mouhri·B. Oudra
Received: 24 February 2008 / Accepted: 20 November 2008 / Published online: 8 January 2009
© Springer Science + Business Media B.V. 2008
Abstract Mansour Eddahbi (MED) (30◦55?N,
6◦53?W) and Almassira (ALM) (31◦95?N, 6◦72?W)
are two Moroccan lake reservoirs located at an
arid and semi-arid hydrographic basin, respec-
tively. Both are used for irrigation, recreational
activities and drinking-water production. This
paper deals with the characterization and quan-
tification of microcystins (MC) from two Micro-
cystis aeruginosa blooms occurring in those reser-
voirs. The toxicity of the blooms was confirmed
M. Douma · M. Loudiki · Kh. Mouhri
Department of Biology, Laboratory of Biology
and Biotechnology of Microorganisms,
Fundamental and Applied Phycology Unit,
Faculty of Sciences Semlalia Marrakech,
University Cadi Ayyad, P.O. Box 2390
Marrakech, Morocco
Y. Ouahid · F. F. del Campo
Departamento de Biología,
Laboratorio de Fisiología Vegetal,
Universidad Autónoma de Madrid,
Cantoblanco, 28049 Madrid, Spain
B. Oudra (B )
Department of Biology, Laboratory of Biology
and Biotechnology of Microorganisms,
Microbiology and Ecotoxicology Environmental Unit,
Faculty of Sciences Semlalia Marrakech,
University Cadi Ayyad, P.O. Box 2390
Marrakech, Morocco
e-mail: oudra@ucam.ac.ma, oudas02@yahoo.fr
and evaluated by both mouse and Artemia bioas-
says. The calculated LD50 values revealed that
the MED bloom had a medium toxicity (LD50=
358 mg kg−1body weight), whereas the ALM
bloomhadlowtoxicity(LD50= 829mgkg−1body
weight). The 24-h LC50 values were 1.88 and
4.15 mg ml−1for the MED and ALM blooms,
respectively, using Artemia assay. The identifi-
cation and quantification of MC variants were
carried out by high performance liquid chroma-
tography (HPLC) equipped with a photodiode
array detector, and HPLC coupled to mass spec-
trometry. The MC content, as Microcystin-LR
(MC-LR) equivalents, was higher in the MED
bloom (64.4 μg g−1dry weight) than in the ALM
bloom (9.9 μg g−1dry weight). Five MC variants
were identified in the MED cyanobacteria bloom
(MC-RR, MC-YR, MC-LR, MC-FR, and MC-
WR) and only one (MC-LR) in the ALM bloom.
The results show that the occurrence of toxic
cyanobacteria blooms in the studied reservoirs
may be regarded as a health hazard; there-
fore, cyanotoxin monitoring in them is highly
recommended.
Keywords Cyanobacteria·Drinking water·
Microcystins·Morocco·Health risk ·
Toxicity assessment

Page 2

440Environ Monit Assess (2010) 160:439–450
Introduction
Cyanobacteria blooms reflect increasing eutroph-
ication in many lakes and reservoirs in different
world regions (Paerl et al. 2001). The presence
of toxic cyanobacteria blooms in water bodies
used for drinking water, irrigation, or recreational
activities poses a hazard to human health and
agriculture (De Figueiredo et al. 2004; Codd et al.
2005).
Cyanotoxins are generally grouped into he-
patotoxins, neurotoxins, cytotoxins, irritants, and
gastrointestinal toxins (Wiegand and Pflugmacher
2005). Among hepatotoxins, microcystins (MCs)
are the best documented, with over 80 natural
structural variants being described (Codd et al.
2005). They are non-ribosomally synthesized hep-
tapeptides, produced by a number of unicellu-
lar and filamentous genera, including Microcystis,
Anabaena, Nostoc, Plankthotrix, and Oscillato-
ria (Sivonen and Jones 1999). Microcystis is the
most frequent bloom-forming genus (Carmichael
1996).
Waters containing cyanotoxins affect in differ-
ent ways diverse organisms such as plants and
animals. Plant growth can be impaired directly
in aquatic plants (Mitrovic et al. 2005) or in-
directly through irrigation in terrestrial plants
(Pflugmacher et al. 2006; Saqrane et al. 2008).
The accumulation of cyanotoxins in crop plants
irrigated with water containing toxic cyanobac-
teria has also been reported (Codd et al. 1999).
This fact poses a technical challenge for water
use managers (Chorus and Bartram 1999) since
many cyanotoxins are not efficiently destroyed by
conventional water treatment methods (Hitzfeld
et al. 2000) such as chlorination.
In order to avoid any sanitary risk, the World
Health Organization has suggested a guideline
value of 1 μg l−1of MC-LR for drinking water
(Chorus and Bartram 1999).
In Morocco, a country with an arid/semi-arid
Mediterranean climate, cyanobacteria blooms are
common in water bodies used for recreational
and/or drinking water. The poisoning of fish,
aquatic birds, and livestock has also been ob-
served during late summer in some reservoirs and
natural ponds, but the poisoning origin was not
investigated in spite of the presence of cyanobac-
teria in those water bodies (Loudiki et al. 2002).
Recently, several reports on toxic cyanobacterial
blooms in Moroccan water reservoirs were pro-
duced, contributing to the idea that cyanotoxi-
city is a world-wide phenomenon (Sbiyaa et al.
1998; Oudra et al. 2001a, b, 2002; Loudiki et al.
2002; Sabour et al. 2002; Douma et al. 2005). In
Morocco, cyanobacteria blooms have been ob-
served in a recurrent way during summer and
autumn in more than 18 out of 26 lake reservoirs
used for human water supply. Several cyanobacte-
ria species occurred in the blooms, but Microcystis
was the dominant genus in all cases (Loudiki et al.
2002).
The present work was carried out within the
framework of a research program on the bio-
diversity and biogeography of toxic cyanobacte-
ria in Morocco. It deals with the toxicity of M.
aeruginosa blooms from two Moroccan drinking-
water reservoirs, Mansour Eddahbi (MED) and
Almassira(ALM),aswellastheidentificationand
quantification of MCs from those blooms.
Study area and general characteristics
The geographical situation of MED and ALM
reservoirs is presented in Fig. 1. The main char-
acteristics and use of both reservoirs are shown
in Table 1. MED reservoir is located 25 km
south of Ouarzazate city (arid climate). It annu-
ally produces about 2.7 Mm3water, mainly used
for human consumption and irrigation of palm
plantations along the course of the Drâa River
(1,800,000 palm trees, covering 40% of the na-
tional oasis surface). This reservoir is classified
as a site of Biological and Ecological Interest
(Ramsar site), being an important refuge for mi-
gratory species.
The ALM reservoir, with an annual water pro-
duction of 500 Mm3, provides drinking and in-
dustrial water for coastal cities between Rabat
and Safi, in particular Casablanca and El Jadida,
the most populated and industrialized area in the
country. It plays a main role in the economic
development of this area, and is also a site of

Page 3

Environ Monit Assess (2010) 160:439–450441
Fig. 1 Geographic
localization of Al Massira
(ALM) and El Manssour
Eddahbi (MED)
drinking-water reservoirs
Rabat
Atlantic ocean
Mediterranean sea
Tanger
Agadir
100 km
Ouazazatate
Marrakech
25 KM
Atlantic ocean
Al Massira Reservoir
Oum Er-RabiaRiver
Oum Er - Rabia Basin
Dr‚a Basin
Mansour Eddahbi Reservoir
Dr‚a River
25 KM
: Reservoir
: City
Biological and Ecological Interest (Ramsar site).
It is considered the most important inland fishing
site in Morocco, with seven commonly caught
species, and catches amounting to 90 tons per
year. It regularly shelters a large number of win-
tering waders and waterfowl; 55,000 birds were
recorded in January 1995, belonging to more than
30 different species. In relation to its trophic sta-
tus, the ALM reservoir is regarded as mesotrophic
(Malki 1994; Oudra et al. 2001b) and MED reser-
voir as hyper-eutrophic (Saadani et al. 2004).
Materials and methods
Sampling and phytoplankton analysis
Cyanobacteria bloom biomass was collected with
a 27-μm mesh phytoplankton net in October 2004
at MED reservoir and in July 2004 at ALM
reservoir. For qualitative studies of the bloom
cyanobacteria, the samples were observed under
light microscope and the cyanobacteria identi-
fied morphologically according to Komarek and
Table 1 Main characteristics and uses of the (MED) and (ALM) reservoirs
Reservoirs
Starting date
Hydrographic basin
Geographical situation
River
Total capacity (Mm3)
Drinking water production (Mm3)
Trophic state
Bioclimate
Other uses
Almassira
1979
Oum Er-Rabia
31◦95?N, 6◦72?W
Oum Er-Rabia
2,760
500
Mesotrophic
Semi-arid
Irrigation, recreation, sport fishing,
electricity
National biological and ecological
interest (Ramsar site no. 1471)
Mansour Eddahbi
1971
Ouarzazate
30◦29?N, 6◦21?W
Drâa
560
2.7
Hyper-eutrophic
Arid
Irrigation, recreation, sport fishing,
electricity
National biological and ecological
interest (Ramsar site no. 1482)
Ecological interest

Page 4

442Environ Monit Assess (2010) 160:439–450
Anagnostidis (1999, 2005). The concentrated bio-
mass was lyophilized and stored at −20◦C until
further use.
For quantitative analysis, 100-ml water samples
were fixed with neutralized formaldehyde solu-
tion. After disintegrating the colonies by ultra-
sonication (50 kHz, 60 s), cyanobacteria cells were
counted using a Mallassez hematocytometer.
Mouse bioassay
Toxicity was measured using a mouse bioassay
with 18–22 g male Swiss mice. Suspensions of
lyophilized bloom material in 0.9% NaCl were
injected intraperitoneally (i.p.) into pairs of mice
per dose. All experimental animals used were
animals from a laboratory (breeding animals),
treated in accordance with institutional guidelines
for the protection of human and animal wel-
fare. Toxicity symptoms were registered and dead
animals were observed for signs of hepatotoxi-
city. The toxicity was determined as LD50 and
expressed on the basis of dry cyanobacteria bio-
mass (μg) that caused death of 50% of the tested
mice (μg kg−1dry weight).
Artemia salina bioassay
The brine shrimp Artemia salina is used exten-
sively for cyanotoxicity testing, since MCs have
been shown to be highly toxic to this organism
(Metcalf et al. 2002; Sabour et al. 2005). Toxins
were extracted from lyophilized cells as follows:
first,theyweresuspendedinultrapurewaterMilli-
Qplus to get the desired biomass concentration
(mg dry weight ml−1) and sonicated 5 min with
a cell disruptor (Sonifer®B-12, Branson Sonic
Power Company), set at 50 kHz and 100 W/cm2;
then, MC extraction was carried out three times
with 75% methanol. Extracts were centrifuged
(5,000 g, 10 min, 4◦C). The supernatants (crude
extracts), after being diluted with ultrapure
water (Milli-Qplus) to reach a methanol concen-
tration of 20%, were loaded onto a Bond Elut
C18 cartridge (Sep-Pack®Vac C18 6 cc, Waters
Co., Milford, USA) previously activated with 20%
aqueousmethanol.Methanol20%wasfirstpassed
through the column, and MC-enriched fractions
were obtained by eluting with 75% methanol.
The fractions were dried at 40◦C and the residues
suspended in methanol and passed through a
0.2-μm Nylon Acrodisc (Gelman Sciences Inc.)
filter. The filtrates were dried and finally dissolved
in 2 ml of sterilized sea water (35?). MC concen-
trationisexpressedonadriedbiomass(DW)basis
(μg g DW−1).
The A. salina larvae utilized were obtained 24 h
after egg hatching. The toxicity bioassay was per-
formed as described by Sabour et al. (2005). The
larvae, in natural seawater, were exposed to MC-
enriched fractions (final volume, 1 ml; 8–20 larvae
in each well) in loosely covered 96-well microtiter
plates at 25◦C. The test results were expressed
as percentage of dead individuals (Vezie et al.
1996, 1998a) after 24 and 40 h of exposure to MC-
enriched extracts. The LC50 is expressed as μg
DW mL−1(assay volume) to attain 50% lethality.
It was estimated by EPA Probit Analysis Program
(version 1.5).
Detection and quantification of microcystins
Microcystin extraction
Lyophilized biomass was extracted with 70%
aqueous methanol (2 mg DW ml−1) and dried
by rotary evaporation at 45◦C. Two extract types
were obtained: concentrated extract, by extracting
the biomass with 70% methanol twice, and a di-
lute extract, by extracting the residue of the bio-
mass with 70% methanol. This procedure ensured
total recovery of MCs. All extracts were filtered
through a GF/C glass filter before being subjected
to HPLC.
Microcystin analysis by HPLC-PDA
Chromatographic analysis of MCs was performed
byHPLC(Waters, model2695)with a photodiode
array detector (model 996). The column used was
Kromasil C18 (ID 5 μm, 250 × 0.46 mm). The
mobile phase consisted of a discontinuous gra-
dient of two eluents: A, water with 0.05% (v/v)
trifluoroacetic acid (TFA); and B, acetonitrile
with 0.05% (v/v) TFA. The gradient profile of the
mobile phase is shown in Table 2. The flow rate
was 1.0 ml min−1. After each chromatography, the
column was cleaned with 100% acetonitrile with

Page 5

Environ Monit Assess (2010) 160:439–450443
Table 2 Eluent gradient in the HPLC analysis
Time (min)
015
70 60
30 40
21
60
40
35
30
70
40 45 46
70
30
50
70
30
Solvent A (%)
Solvent B (%)
00
100 100
Solvent A = water–0.05% trifluoroacetic acid (TFA) (v/v);
B = acetonitrile–0.05% TFA
0.05% TFA, then conditioned with the appropri-
ate eluent mixture for the next run.
MCs were identified on the basis of their UV
spectra and retention time. Standard MC-LR,
MC-YR,MC-RR,MC-LF,andMC-LWwerepur-
chased from Calbiochem (Germany). Other MCs
were quantified using MC-LR as a standard.
Microcystin identification by LC-mass
spectrometry analysis
The LC-mass spectrometry (LC-MS) experiments
were carried out on an Agilent 1100 series
HPLC system (Agilent Technologies, CA, USA)
consisting of a vacuum degasser, binary pump,
auto sampler, and a DAD detector coupled to
a hybrid quadrupole time-of-flight instrument
(QStar/Pulsar i; Applied Biosystems, CA, USA)
equipped with a turbo spray ion source interface.
Chromatographic separation was carried out as
described above. Full scan spectra were acquired
in the positive ion mode, using a source potential
of 5,000 V, over the mass range of 100–1,500 at 1 s.
ESIMS/CID mass spectra were measured using
nitrogen as a collision gas (collision energy, 3 kV)
in the pressure range of 65 bar. The nitrogen gas
drying temperature was set at 300◦C, and the cone
voltage was fixed at 70 V.
Results
Bloom-forming cyanobacteria in MED
and ALM reservoirs
The dominant bloom-forming species in the sam-
ples from both MED and ALM reservoirs was
Microcystis aeruginosa Kütz. The relative abun-
dance was higher than 95%, with a cell density
of 3 × 104cells ml−1(ALM bloom) and 5.6 ×
104cells ml−1(MED bloom).
The species was identified according to diverse
morphological features as described by Watanabe
(1996) and Komarek and Anagnostidis (1999).
The Microcystis colonies were mucilaginous, and
usually formed a clathrate-like net with distinct
holes. Cells were generally spheric, with a diame-
ter of 3.5–6.5 μm, and numerous aerotopes. Other
minor cyanobacteria species in the blooms were
Pseudanabaena mucicola Nau & Hub., Oscillato-
ria sancta Kütz., Oscillatoria ornata (Kütz.) Gom.,
Synechocystis pevalkii Erc., Spirulina major Kütz.
in ALM bloom, and Pseudanabaena papillatermi-
nata Kuk., Oscillatoria sp. in MED bloom.
Toxicity evaluation by bioassays
Toxicity of both MED and ALM blooms, as
measured by LD50mouse bioassay, was 358 and
829 mg kg−1DW, respectively. The animal sur-
vival time at a lethal dose was about 2–4 h with
the MED bloom extract and 3–6 h with the ALM
extract; therefore, the MED bloom appeared to
be more toxic than the ALM bloom. The most
conspicuous poisoning manifestations observed
during the bioassay were ataxia, paleness, and
severe diarrhea, which appeared 30–60 min af-
ter injection. After death, the autopsied animals
showed grossly enlarged livers. These symptoms
are similar to those usually observed in hepato-
toxicosis cases.
Toxicity was also assayed with Artemia. In this
case, MC extracts corresponding to two different
bloom biomass concentrations were used, 10 and
5 mg ml−1, in order to ascribe toxicity according
to Kiviranta et al. (1991). These authors reported
that the Artemia test allows to classify toxic sam-
ples that cause mortality higher than 50% at 10 mg
biomass ml−1assay volume after 24 h exposure
and higher than 20% at 5 mg ml−1. These con-
centrations have a good correspondence with the
toxicity in the mouse bioassay (Vezie et al. 1996).
The mortality rate after 24 h caused by partially
purified MC extracts from the two reservoir sam-
ples, corresponding to a dried biomass of 10 and
5 mg ml−1, were 79.41% and 72.66% for MED
bloom and 65.86% and 51.43% for the ALM
bloom, respectively (Table 3). The effect after

Page 6

444Environ Monit Assess (2010) 160:439–450
Table 3 Toxicity of cyanobacteria bloom extract collected from MED and ALM drinking-water reservoirs
Extract biomass concentration (mg ml−1)
10
MED bloom 24 h 79.41 (>50)
40 h 87.00
ALM bloom24 h65.86 (>50)
40 h73.13
SampleExposure time Result of bioassay
5
72.66 (>20)
78.47
51.43 (>20)
58.47
2.5
63.07
69.29
40.66
48.66
1.25
42.80
47.78
31.61
43.00
0.625
21.77
29.62
16.31
29.17
Toxic
Toxic
Toxicity evaluated by A. salina bioassay. The numbers represent the percentage of mortality
24- and 40-h exposure to the same MC-containing
extracts is compared (Fig. 2). The LC50after 24
and 40 h of the MED bloom were 1.71 and 1.29 mg
DW ml−1, respectively, and 4.34 and 2.49 mg ml−1
for the ALM bloom.
Identification and quantification of microcystins
(MC) in the bloom samples
Quantitative and qualitative analyses of MCs in
the two reservoirs were carried out to see if there
was a correlation between the MC content of the
bloom samples assayed and their toxicity. HPLC
analysis revealed a very different MC content. In
the MED bloom sample, it was 64.4 μg (equiva-
lent MC-LR) g−1, much higher than that in the
0
20
40
60
80
100
Extract biomass concentration (mg/ml)
Mortality (%)
24h
40h
(B)
0
20
40
60
80
100
024681012
024681012
Extract biomass concentration (mg/ml)
Mortality (%)
24h
40h
(A)
Fig. 2 Effect of MC-containing extracts from MED bloom
(a) and ALM bloom (b) on A. salina after 24 h of exposure
ALM sample (9.9 μg g−1). Moreover, the two
blooms differed in the MC profile (Fig. 3). In the
ALM bloom, the only MC present was MC-LR,
while in the MED bloom, five MC variants were
detected (Fig. 3a). Using commercial standards,
three of them were identified as MC-LR, MC-RR,
and MC-YR; the other two were suspected to be
MC-FR and MC-WR, based on retention times
and UV spectra. To confirm the identity of these
suspected components, a HPLC-MS analysis was
carried out. The full scan spectrogram of LC-MS
presented five peaks. Three of them corresponded
to the three MCs identified by HPLC-PDA. The
other two peaks of m/z ions at 1,029.5 and 1,068.5
(Fig. 4) corresponded to MC-FR and MC-WR,
respectively.
Discussion
The toxicity and the microcystin content of two
cyanobacteria blooms which appeared in the
Moroccan water reservoirs Mansour Eddahbi and
Almassira was analyzed. In the two blooms, the
predominant species was Microcystis aeruginosa
Kütz. Amongst the seven species of Microcys-
tis cited in Morocco, M. aeruginosa is the most
widespread, proliferating in the majority of high
trophic water bodies (Loudiki et al. 2002). Dur-
ing a eutrophication survey program in the two
reservoirs (Malki 1994; Oudra et al. 2002; Saadani
et al. 2004), Microcystis blooms occurred season-
ally every year during summer and autumn (from
July to December). During blooms, thick scums
of M. aeruginosa tend to accumulate near the
shoreline of the dams, and its persistence in the
reservoirs seems to be determined by high water
column stability.

Page 7

Environ Monit Assess (2010) 160:439–450445
Fig. 3 HPLC
chromatograms of
Microcystis blooms of the
MED (a) and ALM (b)
drinking-water reservoirs.
HPLC conditions are
described in “Materials
and methods” section
Absorbance (AU)
Retention time (min)
Absorbance (AU)
A
B
In spite of M. aeruginosa being the predom-
inant species in the two studied water bodies,
their toxicity was different, as determined by the
mouse and Artemia bioassays. The LD50 in the
mouse bioassay for the MED bloom, 358 mg kg−1
(Table 4), allowed us to classify the bloom as
toxic, according to Lawton et al. (1994). The
toxicity of the MED bloom could be regarded
as medium toxic when compared with other
Microcystis blooms reported from Moroccan wa-
ter reservoirs studied so far (Table 4). Thus, it is
higher than that of a bloom from the Oued Mellah
lake (LD50= 502 mg kg−1), where the dominant
species was M. ichtyoblabe (Sabour et al. 2002),
but lower than that of a M. aeruginosa bloom
from the Lalla Takerkoust reservoir (Oudra et al.
2002).
Using the mouse bioassay, the ALM bloom was
only slightly toxic (829 mg kg−1). The toxicity was
six times lower than that observed in a Microcystis
bloom from the same reservoir in 1999 (142 mg
kg−1, Oudra et al. 2001a). The difference in the
toxicity between the 1999 and 2004 blooms may
be due to differences in the genotype composi-
tion and/or the growth phase of the cyanobac-
teria (Sabour et al. 2002), while following up
a Microcystis bloom in Oued Mellah reservoir,
observed a change in the mouse LD50, from 518 to

Page 8

446Environ Monit Assess (2010) 160:439–450

Page 9

Environ Monit Assess (2010) 160:439–450447
? Fig. 4 Full scan electrospray mass spectra of MC-FR and
MC-WR with the base peak at 1,029.5 m/z and 1,068.5 m/z,
respectively
1,924 mg kg−1, corresponding to the exponential
and decline growth phase of the bloom, respec-
tively. During the decline growth phase of Micro-
cystis, most of the soluble MC should be detected
in the water, since these toxins are released into
the medium after cell disruption.
Considering the toxicity using the mouse bioas-
say of the two blooms, along with that of previ-
ous reports (Table 4), it seems that cyanobacteria
toxicity in Moroccan water supplies is rather vari-
able, both among different supplies and even in
one given supply, as in the ALM reservoir. The
highest toxicity reported in Moroccan reservoirs
corresponds to a M. aeruginosa bloom from the
Lalla Takerkoust reservoir (Oudra et al. 2002). A
highlyvariabletoxicityinMicrocystisbloomsfrom
a specific reservoir along to Mediterranean basin
has previously been reported, as in Kastoria lake
(Greece), with a LD50ranging from 40 to 1,500 mg
kg−1(Cook et al. 2004).
According to the Artemia bioassay, both MED
and ALM blooms are toxic. But again, a signif-
icant difference between them is observed: 24-h
LC50= 1.71 mg DW ml−1for MED bloom and
4.34 mg DW ml−1for ALM bloom (Fig. 2). As
with the mouse bioassay, these toxicity values are
higher than those previously obtained with the
Microcystis bloom from Oued Mellah reservoir,
24 h LC50 of 6–46 mg DW ml−1(Sabour et al.
2002).
Taking together, our results with the Artemia
bioassay clearly confirm that it can be used as an
alternative test to evaluate cyanobacteria toxicity.
A direct relationship between toxicity and MC
content in the two blooms was found, since in the
MED bloom the MC content was about seven
times bigger than that of the ALM one. At any
rate, the MC concentration in both blooms was
higher than that observed before in the ALM
and Oued Mellah reservoirs (Oudra et al. 2001b;
Sabour et al. 2002). However, the content is very
low compared with that reported for Microcystis
blooms of diverse origin from the Mediterranean
region, as: theMoroccan
(8.8 mg g−1; Oudra et al. 2001a), various
Portuguese reservoirs (1–7.1 mg g−1; Vasconcelos
2001), the Spanish Santillana reservoir (13.5 mg
g−1; Padilla et al. 2006), and some French
reservoirs (0.07–3.97 mg g−1; Vezie et al. 1998b).
The MED and ALM blooms were also qualita-
tively different with respect to the MCs variants.
In the ALM bloom, only MC-LR was clearly de-
tected (Fig. 3a). In a Microcystis bloom occurring
in this reservoir in 1999, two microcystin variants,
MC-LR and MC-RR, were detected (Oudra et al.
2001b). The detection of one MC variant does
not necessarily mean the presence of just one
cyanobacteria strain in the bloom. The coexis-
tence of MC-producing and non-producing strains
is a normal event (Vezie et al. 1998a; Rohrlack
et al. 2001; Kurmayer et al. 2002), and it may occur
that the MC-producing strains are not dominant.
In the MED bloom, five MC variants were
present, the three major ones being MC-LR, MC-
RR, and MC-YR. Predominance of these three
MCs was also reported in blooms of other coun-
tries from the Mediterranean region: Algeria
(Nasri et al. 2004), Egypt (Abdel-Rahman et al.
1993), Morocco (Oudra et al. 2001a), Portugal
(Vasconcelos 2001), and Spain (Padilla et al.
2006). The two minor, relatively more hydropho-
Takerkoustlake
Table 4 Toxicity assessment of Microcystis blooms occurring in some Moroccan lake reservoirs
Reservoirs
Lalla Takerkoust
Date of sampling
1999
2003
1999
2005
1999
2005
Dominant strain forming-bloom
Microcystis aeruginosa
Microcystis aeruginosa
Microcystis aeruginosa
Microcystis aeruginosa
Microcystis ichtyoblabe
Microcystis aeruginosa
LD50mg kg-1
34
177
142
829
502
358
Reference
Oudra et al. (2002)
Douma et al. (2005)
Oudra et al. (2002)
This work
Sabour et al. (2002)
This work
Almassira
Oued Mellah
Mansour Eddahbi
Toxicity evaluated by mouse bioassay

Page 10

448 Environ Monit Assess (2010) 160:439–450
bic MC variants from the MED bloom, MC-FR
and MC-WR, are reported for the first time in
Moroccan waters. It would be interesting to know
if the different MC variants arise from the same or
from different cyanobacteria. To establish these
possibilities, it would be necessary to obtain dif-
ferent isolates and compare their MC profile.
Conclusion
The occurrence of M. aeruginosa blooms in two
Moroccan drinking-water reservoirs, ALM and
MED, is described. The toxicological data, both
by mouse and Artemia bioassays, showed the
toxic nature of both blooms. There is a correla-
tion between toxicity and MC content. This work
constitutes the first report on MC quantification
in M. aeruginosa blooms from MED reservoir.
Five MC variants were identified by HPLC-MS
in that bloom; two of them, MC-FR and MC-
WR, are reported for the first time in Morocco.
The occurrence of toxic cyanobacteria blooms
in the two reservoirs may eventually contami-
nate water in an important manner, contributing
to a human and animal health risk. To prevent
intoxications, continuous monitoring of the two
reservoirs is strongly advised. Such an action is
urgently needed in the MED reservoir, among
other reasons due to its biological and ecological
significance.
Acknowledgements
Moroccan government and the Spanish Agency of In-
ternational Cooperation AECI (bilateral Spain–Morocco
Project A/011422/07). Mountasser DOUMA was recipi-
ent of an excellence grant from the National Center of
Scientific and Technical Research—Ministry of National
Education, Higher Education, Staff Training and Scien-
tific Research. The authors would like to thank Dr. Lee
Robertson for his help in English language and anonymous
reviewers for useful comments on the manuscript.
This work was funded by the
References
Abdel-Rahman, S., El-Ayouty, Y. M., & Kamael, H. A.
(1993). Characterization ofheptapeptide toxins
extracted from Microcystis aeruginosa (Egyptian
isolate). Comparison with some synthesized analogs.
International Journal of Peptide and Protein Research,
41, 1–7.
Carmichael, W. W. (1996). Toxic Microcystis and the
environment. In F. Watanabe, K. H. Harada, W. W.
Carmichael, & H. Fujiki (Eds.), Toxic microcystis
(pp. 1–11). Boca Raton: CRC.
Chorus, I., & Bartram, J. (1999). Toxic cyanobacteria in
water. A guide to their public health consequences,
monitoring and managements. WHO, London: E. & F.
N. Spon, Routledge, 416 pp.
Codd, G. A., Metcalf, J. S., & Beattie, K. A. (1999).
Retention of Microcystis aeruginosa and microcystin
by salad lettuce (Lactica sativa) after spray irriga-
tion with water containing cyanobacteria. Toxicon, 37,
1181–1185. doi:10.1016/S0041-0101(98)00244-X.
Codd, G. A., Morrison, L. F., & Metcalf, J. S. (2005).
Cyanobacterial toxins: Risk management for health
protection. Toxicology and Applied Pharmacology,
203, 264–272. doi:10.1016/j.taap.2004.02.016.
Cook, C. M., Vardaka, E., & Lanaras, T. (2004).
Toxic cyanobacteria in Greek freshwaters, 1987–2000:
Occurrence, toxicity, and impacts in the Mediter-
ranean region. Acta Hydrochimica et Hydrobiologica,
32, 107–124. doi:10.1002/aheh.200300523.
De Figueiredo, D. R., Azeiteiro, U. M., Esteves, S.
M., Goncalves, F. J. M., & Pereira, M. J. (2004).
Microcystin-producing blooms—a serious global pub-
lic health issue. Ecotoxicology and Environmen-
tal Safety, 59, 151–163. doi:10.1016/j.ecoenv.2004.04.
006.
Douma, M., Loudiki, M., Sabour, B., Oudra, B., Mouhri,
K., & Vasconcelos, V. (2005). Cyanobactéries des
zones humides du Maroc: Inventaire & Distribution
géographique. Acte des résumés du deuxième congrès
Méditerranéen “Ressources en Eau dans le Bassin
Méditerranéen” (WATMED 2), Marrakech, Maroc,
p. 345
Hitzfeld, B. C., Hoger, S. J., & Dietrich, D. R. (2000).
Cyanobacterial toxins: Removal during drinking
water treatment, and human risk assessment. Environ-
mental Health Perspectives, 108, 113–122. doi:10.2307/
3454636.
Kiviranta, J., Sivonen, K., Niemela, S. I., & Huovinen, K.
(1991). Detection of toxicity of cyanobacteria by
Artemia salina bioassay. Environmental Toxicol-
ogy and Water Quality, 6, 423–436. doi:10.1002/tox.
2530060407.
Komarek, J., & Anagnostidis, K. (1999). Cyanoprokary-
ota. 1. Teil: Chroococcales. Süßwasserflora von Mit-
teleuropa; Band 19/1. Spectrum Akademischer verlag,
Gustav Fischer, Berlin; 643 figures, 548 pp.
Komarek, J., & Anagnostidis, K. (2005). Cyanoprokary-
ota. 2. Teil: Oscillatoriales. Süßwasserflora von Mit-
teleuropa; Band 19/2. Spectrum Akademischer verlag,
Elsevier GmbH, München; 110 figures, 759 pp.

Page 11

Environ Monit Assess (2010) 160:439–450449
Kurmayer, R., Dittman, E., Fastner, J., & Chorus, I. (2002).
Diversity of microcystin genes within a population
of the toxic cyanobacterium Microcystis spp. in lake
Wannsee (Berlin, Germany). Microbial Ecology, 43,
107–118. doi:10.1007/s00248-001-0039-3.
Lawton, L. A., Beattie, K. A., Hawser, S. P., Campbell,
D.L.,&Codd,G.A.(1994).Evaluationofassaymeth-
ods for the determination of cyanobacterial hepato-
toxicity. In G. A. Codd, T. M. Jeffries, C. W. Keevil, &
E. Potter (Eds.), Detection methods for cyanobacterial
toxins (pp. 111–116). Cambridge: The Royal Society of
Chemistry.
Loudiki, M., Oudra, B., Sabour, B., Sbiyyaa, B., &
Vasconcelos, V. (2002). Taxonomy and geographic
distribution of potential toxic cyanobacterial strains in
Morocco. Annales de Limnologie, 38, 101–108.
Malki, M. (1994). Etude de la communauté phytoplanc-
tonique et des caractéristiques physico-chimiques des
eaux du lac reservoir Al Massira. Thèse de Doctorat
d’état, Univ. Hassan II, Casablanca, 287 p.
Metcalf, J. S., Lindsay, J., Beattie, K. A., Birmingham,
S., Saker, M. L., Torokne, A. K., et al. (2002).
Toxicity of cylindrospermopsin to the brine shrimp
Artemia salina: Comparisons with protein synthesis
inhibitors and microcystins. Toxicon, 40, 1115–1120.
doi:10.1016/S0041-0101(02)00105-8.
Mitrovic, S. M., Allis, O., Furey, A., & James, K. J. (2005).
Bioaccumulation and harmful effects of microcystin-
LR in the aquatic plants Lemna minor and Wolffia
arrhiza and the filamentous alga Chladophora fracta.
Ecotoxicology and Environmental Safety, 61, 345–352.
doi:10.1016/j.ecoenv.2004.11.003.
Nasri, A. B., Bouaicha, N., & Fastner, J. (2004). First report
of a microcystin containing bloom of the cyanobacte-
ria Microcystis spp. in Lake Oubeira, eastern Algeria.
Archives of Environmental Contamination and Toxi-
cology, 46, 197–202.
Oudra, B., Loudiki, M., Sbiyyaa, B., Martins, R.,
Vasconcelos, V., & Namikoshi, M. (2001a). Isola-
tion, characterization and quantification of micro-
cystins (heptapeptides hepatotoxins) in Microcystis
aeruginosa dominated bloom of Lalla Takerkoust
lake-reservoir (Morocco). Toxicon, 39, 1375–1381.
doi:10.1016/S0041-0101(01)00093-9.
Oudra, B., Loudiki, M., Sabour, B., Sbiyyaa, B., &
Vasconcelos, V. (2001b). Étude des blooms toxiques
à cyanobactéries dans trois lacs réservoirs du Maroc:
Résultats préliminaires. Revue Des Sciences De l’eau,
15, 301–313.
Oudra, B., Loudiki,M.,
Martins, R., Amorim, A., et al. (2002). Detection
and variation of microcystin contents of Microcystis
blooms in eutrophic Lalla Takerkoust Lake, Morocco.
Lakes and Reservoirs: Research and Management, 7,
35–44. doi:10.1046/j.1440-1770.2002.00165.x.
Padilla, C., Soledad, S.-A., & Del Campo, F. F. (2006).
Toxin characterisation and identification of a Micro-
Sbiyyaa, B., Sabour,B.,
cystis flos-aquae strain from a Spanish drinking-water
reservoir. Archiv für Hydrobiologie, 165, 383–399.
doi:10.1127/0003-9136/2006/0165-0383.
Paerl, H. W., Fulton, R. S., Moisander, P. H., & Dyble,
J. (2001). Harmful freshwater algal blooms, with an
emphasis on cyanobacteria. Science World Journal, 1,
76–113.
Pflugmacher, S., Jung, K., Lundvall, L., Neumann, S., &
Peuthert, A. (2006). Effects of cyanobacterial toxins
and cyanobacterial cell-free crude extract on the ger-
mination of alfafa (Medicago sativa) and induction
of oxidation stress. Environmental Toxicology and
Chemistry, 9(25), 2381–2387. doi:10.1897/05-615R.1.
Rohrlack, T., Henning, M., & Kohl, J. G. (2001). Isolation
and characterization of colony-forming Microcystis
aeruginosa strains. In I. Chorus (Ed.), Cyanotoxins—
occurrence, effects, controlling factor (pp. 152–158).
Heidelberg: Springer.
Saadani, M., Ouazzani, N., & Mandi, L. (2004). Impact
de la sécheresse sur l’évolution de la qualité des eaux
du lac Mansour Eddahbi (Ouarzazate, Maroc). Revue
Des Sciences De l’eau, 17, 69–90.
Sabour, B., Loudiki, M., Oudra, B., Vasconcelos, V.,
Martins, R., Oubraim, S., et al. (2002). Toxicology of a
MicrocystisichthyoblabewaterbloomfromLakeOued
Mellah (Morocco). Environmental Toxicology, 17, 24–
31. doi:10.1002/tox.10028.
Sabour, B., Loudiki, M., Oudra, B., Vasconcelos, V.,
Oubraim, S., & Fawzi, B. (2005). Dynamics and toxi-
city of Anabaena aphanizomenoides (Cyanobacteria)
waterblooms in the shallow brackish Oued Mellah
lake(Morocco).AquaticEcosystemHealth&Manage-
ment, 8, 95–104. doi:10.1080/14634980590914944.
Saqrane, S., El ghazali, I., Oudra, B., Bouarab, L., &
Vasconcelos, V. (2008). Effects of cyanobacteria pro-
ducing microcystins on seed germination and seedling
growth of several agricultural plants. Journal of En-
vironmental Science and Health. Part. B, Pesticides,
Food Contaminants, and Agricultural Wastes, 43, 1–9.
doi:10.1080/03601230802062307.
Sbiyaa, B., Oudra, B., Loudiki, M., Bouguerne, A., &
Tifnouti, A. (1998). Acute toxicity of Microcystis
aeruginosa KÜTZ. to three cladoceran species. In
B. Reguera, J. Blonco, M. L. Fernandez, & T. Wyatt
(Eds.), Harmful algae (pp. 29–31). Xunta de Galicia
& Intergovernmental Oceanographic Commission of
UNESCO.
Sivonen, K., & Jones, G. (1999). Cyanobacterial toxins. In
I. Chorus, & J. Bartram (Eds.), Toxic cyanobacteria
in water: A guide to their public health consequences,
monitoring and management (pp. 41–111). London:
E & Spon.
Vasconcelos, V. M. (2001). Toxic freshwater cyanobacte-
ria and their toxins in Portugal. In I. Chorus (Ed.),
Cyanotoxins—occurrence, effects, controlling factors
(pp. 64–69). Heidelberg: Springer.

Page 12

450Environ Monit Assess (2010) 160:439–450
Vezie, C., Brient, L., Sivonen, K., Betru, G., Lefeuvre,
J. C., & Salkinoja-Salonen, M. (1998a). Variation of
microcystin content of cyanobacterial blooms and iso-
lated strains in Lake Grand-Lieu (France). Microbial
Ecology, 35, 126–135. doi:10.1007/s002489900067.
Vezie,C.,Brient,L.,Sivonen,K.,Betru,G.,Lefeuvre,J.C.,
& Salkinoja-Salonen, M. (1998b). Occurrence of mi-
crocystins containing cyanobacterial blooms in fresh-
waters of Brittany (France). Archiv für Hydrobiologie,
139, 401–413.
Vezie, C., Sivonen, K., Brient, L., Bertru, G., & Lefeuvre,
J. C. (1996). Développement de cyanobactéries dans
l’Ouest de la France. Détection de la toxicité par des
tests sur Artemia salina. Annales de Limnologie, 32,
123–128.
Watanabe, M. F. (1996). Toxic Microcystis in eutrophic
lakes. In M. F. Watanabe, K.-I. Harada, & W. W.
Carmichael (Eds.), Toxic Microcystis (pp. 57–77).
Boca Raton: CRC.
Wiegand, C., & Pflugmacher, S. (2005). Ecotoxicological
effects of selected cyanobacterial secondary metabo-
lites: A short review. Toxicology and Applied Phar-
macology, 203, 201–218. doi:10.1016/j.taap.2004.11.
002.