TAbstract
An unpre
Samples we
mcyB gene t
low in dense
physical inju
in the produ
death (PCD)
Our results
media. Rem
levels of mic
? 2006 Else
Keywords: C
1. Introdu
The pre
of Florida
years since
Istokpoga
studies in t
River Lago
ter ecosyst
Griffin, the
vided infor
light availa
ents of plan
2002). Asid
rial bloom
abiotic nat
? Correspon
E-mail ad
0166-445X/$
doi:10.1016/jAquatic Toxicology 78 (2006) 66?73
oxin release in response to oxidative stress and programmed cell
death in the cyanobacterium Microcystis aeruginosa
Cliff Ross a,?, Lory Santiago-Va?zquez b, Valerie Paul a
a Smithsonian Marine Station at Fort Pierce, 701 Seaway Drive, Ft. Pierce, FL 34949, United States
b Department of Chemistry and Biochemistry, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431, United States
Received 7 December 2005; received in revised form 14 February 2006; accepted 15 February 2006
cedented bloom of the cyanobacterium Microcystis aeruginosa Ku?tz. occurred in the St. Lucie Estuary, FL in the summer of 2005.
re analyzed for toxicity by ELISA and by use of the polymerase chain reaction (PCR) with specific oligonucleotide primers for the
hat has previously been correlated with the biosynthesis of toxic microcystins. Despite the fact that secreted toxin levels were relatively
natural assemblages (3.5
g l?1), detectable toxin levels increased by 90% when M. aeruginosa was stressed by an increase in salinity,
ry, application of the chemical herbicide paraquat, or UV irradiation. The application of the same stressors caused a three-fold increase
ction of H2O2 when compared to non-stressed cells. The application of micromolar concentrations of H2O2 induced programmed cell
as measured by a caspase protease assay. Catalase was capable of inhibiting PCD, implicating H2O2 as the inducing oxidative species.
indicate that physical stressors induce oxidative stress, which results in PCD and a concomitant release of toxin into the surrounding
ediation strategies that induce cellular stress should be approached with caution since these protocols are capable of releasing elevated
rocystins into the environment.
vier B.V. All rights reserved.
aspase; Cellular stress; Hydrogen peroxide; Microcystis aeruginosa; Microcystins; P
ction
valence of toxic cyanobacterial blooms in the state
has received considerable attention in the past 20
first being recorded in Lake Okeechobee and Lake
(Carmichael, 1992; Burns et al., 2002). Long-term
hree major marine ecosystems (Florida Bay, Indian
on, and the Suwannee Estuary) and five freshwa-
ems (Lake Okeechobee, the St. Johns River, Lake
Rainbow River, and the Suwannee River) have pro-
mative data on trophic states, water exchange rates,
bility, and measurements of growth-limiting nutri-
ktonic assemblages (Phlips et al., 1993, 2002; Phlips,
e from increasing anthropogenic input, cyanobacte-
s can form in eutrophic water masses simply from
ural sources such as surface or ground water input
ding author. Tel.: +1 772 465 6630; fax: +1 772 461 8154.
dress: Ross@sms.si.edu (C. Ross).
from natur
urbanizatio
nutrient effl
been quite
(Canfield e
Lake Ok
terial bloo
Florida hur
disturbance
of inorgani
the release
a washout
(SLR) Estu
water syste
square mile
biologicall
Sporadi
tified in th
aeruginosa
St. Lucie w
Florida (Fi
? see front matter ? 2006 Elsevier B.V. All rights reserved.
.aquatox.2006.02.007rogrammed cell death
ally nutrient-rich sediments (Phlips et al., 2002). As
n and agricultural expansion have led to increases in
uxes into Florida water systems, cyanobacteria have
opportunistic in exploiting these available nutrients
t al., 1989).
eechobee is one of best known sources of cyanobac-
ms in the United States (Phlips et al., 2002). The
ricane season of 2004 resulted in a major sediment
to the lake resulting in the release of high levels
c phosphorus. Periods of high rainfall followed by
of water from district canals most likely resulted in
of freshwater cyanobacteria into the St. Lucie River
ary. The SLR Estuary is one of the largest brackish
ms on the east coast of Florida. Encompassing 780
s, the estuary represents an indispensable asset both
y and economically.
c colonies of Microcystis aeruginosa were first iden-
e SLR Estuary in June 2005. By July 2005, M.
abruptly emerged as a dense bloom covering the
aterway within both St. Lucie and Martin counties,
g. 1A and B).
8 (200
M. aeru
have detrim
variants of
fish, domes
et al., 1984
et al., 1998
Microcystis
cases, one
yet may va
cases, som
their toxici
and Neilan
in toxicity
toxin-assoc
have led to
(Bittencour
Vaitomaa e
inosa is fo
of cell-bou
conditions
There a
stress of M
ronment. I
photochem
tion to the
2005). How
an imbalan
serve as a p
et al., 2002
quantify H
has subseq
ety of phot
al., 2005).
The pur
Estuary as
tain environ
could indu
ble toxins
was of inte
ated with o
(PCD).C. Ross et al. / Aquatic Toxicology 7Fig. 1. (A) Bloom of M. aeruginosa in the St. Lucie River, August 2005. (B)
ginosa contains a suite of toxic heptapeptides that
ental impacts on environmental health. Hepatotoxin
microcystin are directly associated with the deaths of
tic livestock, and even human mortalities (Skulberg
; Gunn et al., 1992; Rodger et al., 1994; Jochimsen
; Codd et al., 1999). Interestingly enough, not all
strains produce toxins. In some bloom-specific
species can be morphologically identical to the next
ry in toxicogenicity (Baker et al., 2001). In other
e species are known to upregulate or downregulate
ty under varying laboratory conditions (Kaebernick
, 2001). It is not known why such natural variations
exist. The use of molecular probes that target
iated genes, in conjunction with imunoassays,
advancements in the identification of toxic strains
t-Oliveira, 2003; Kaebernick and Neilan, 2001;
t al., 2003). However, even if a strain of M. aerug-
und to contain toxin-associated genes or low levels
nd toxins, it is not clear exactly what environmental
may induce toxin release.
re no previous reports directly relating the cellular
. aeruginosa with microcystin release into the envi-
n cyanobacteria, H2O2 is commonly produced via
ical reactions where concentrations vary in propor-
amount of sunlight (Palenik et al., 1987; Xue et al.,
ever, the biological production of H2O2 may reflect
ced state of redox within the chloroplast and thus may
roxy for cellular stress (Twiner and Trick, 2000; He
; Choo et al., 2004). Among the approaches used to
2O2, the fluorogenic method is quite successful and
uently been used to monitor oxidative stress in a vari-
osynthetic organisms (He and Hader, 2002; Ross et
pose of this study was to verify toxicity in the SLR
semblage of M. aeruginosa and determine if cer-
mental conditions or potential remediation strategies
ce stress and lead to a significant release of solu-
into the surrounding water column. In addition, it
rest to evaluate whether toxin release was associ-
xidative stress and subsequent programmed cell death
2. Materia
2.1. Micro
Specime
face water
to the Ri
80?19.747?
experiment
freshwater
were transp
Fort Pierce
2.2. DNA e
Total ge
Kit (Qiagen
protocol. T
Absorbanc
measured u
Hercules, C
lowing the
(2003) usin
by Neilan e
in 25
l vo
Master Mi
used: 94 ?C
40 ?C for 2
sion at 72 ?
were run on
bromide, a
mance gel
Piscataway
2.3. Seque
McyB g
using the P
technology
an ABI PR
Sequences6) 66?73 67Colony of M. aeruginosa. Scale bar, 45
m.
ls and methods
cystis collections
ns of M. aeruginosa were collected from sur-
s (salinity 0.2?, temperature 32.7 ?C) adjacent
verwatch Marina, Stuart, Florida (27?27.962?N,
?W) on 12 August 2005. For programmed cell death
s, specimens were collected from a nearby unnamed
pond (27?54.288?N, 80?37.234??W). Cyanobacteria
orted to the Smithsonian Marine Station (SMS) at
for immediate analysis after collection.
xtraction and ampli?cation
nomic DNA was prepared using a DNeasy Plant Mini
, Valencia, CA, USA) following the manufacturer?s
ypical DNA yields ranged from 1 to 10
g ml?1.
es (A260/A280) to determine quantity and quality were
sing a SmartSpec Plus Spectrophotometer (Bio-Rad,
A, USA). PCR amplification was conducted by fol-
protocol previously reported by Bittencourt-Oliveira
g the mcyB primers (forward and reverse) designed
t al. (1999). The amplification process was carried out
lumes on a MJ PTC-200 cycler (Bio-Rad) using Taq
x (Qiagen). The following cycling parameters were
for 2 min, followed by 35 cycles at 94 ?C for 10 s;
0 s, and 72 ?C for 1 min, followed by a final exten-
C for 5 min. Aliquots of the PCR reaction product
a 1.2% agarose gel containing 10
g ml?1 ethidium
nd documented with a Typhoon 9410 high perfor-
and blot imager (GE Healthcare Bio-Sciences Corp.,
, NJ, USA).
ncing analysis and accession numbers
ene amplicons were sequenced in both directions
CR primers mcyB and Big Dye Terminator v3.1
(Applied Biosystems, Foster City, CA, USA) in
ISM 3100 Genetic Analyzer (Applied Biosystems).
were viewed and edited using ChromasLite2000
68 8 (200
(www.tech
Biology W
were also c
Alignment
sequences
published m
(Lasergene
sequence w
number DQ
2.4. Detec
Detectio
into the w
Microcysti
according t
assay is ca
ants (Micro
Absorbanc
madzu UV
ments Inc.,
structed us
St. Louis,
from 1.05 ?
0.2?, tem
of 10
E s?
(n = 3). Exp
2.5. Measu
To estab
cells of M.
a direct co
been comm
higher plan
algae (Col
et al., 2005
gen peroxi
diacetate (D
DCFH-DA
this compo
is cleaved o
dihydrofluo
yields the fl
DCFH-DA
stocks (sto
The me
in 50 ml of
Irradiance
(LI-COR,
(upon appl
cyanobacte
cuvette con
The total v
H2O2 was
fluoromete
using the m
trati
s sta
exam
ystin
r an
= 3
ning
tem
ng w
xam
nosa
ter)
ls (n
der
s we
in a
reas
c filt
adia
d -C
artec
, 80

.
aqua
ys pl
hotoC. Ross et al. / Aquatic Toxicology 7
nelysium.com.au/chromas.html) and aligned using
orkbench (http://workbench.sdsc.edu). Sequences
ompared to those in databases using the Basic Local
Search Tool (BLAST) algorithm to identify known
with a high degree of similarity. Homology to other
cyB gene sequences was conducted with MegAlign
, DNASTAR, Madison, WI, USA). The mcyB gene
as submitted to GenBank and assigned the accession
218313.
tion of toxins
n and quantification of microcystin toxins released
ater column were measured using an Envirologix
n Tube Kit (Envirologix Inc., Portland, ME, USA)
o the manufacturer?s instructions. This ELISA based
pable of reacting with four microcystin toxin vari-
cystin LR, LA, RR, and YR) as well as Nodularin.
e measurements (450 nm) were recorded on a Shi-
-265 spectrophotometer (Shimadzu Scientific Instru-
Columbia, MD, USA). A standard curve was con-
ing commercially available Microcystin LR (Sigma,
MO, USA). The measurement of toxins released
108 cells in 50 ml of in situ estuary water (salinity
perature 32.7 ?C, maintained at a constant irradiance
1 m?2) was used as a control for all experiments
concen
used a
To
microc
tomete
cate (n
contai
at room
roundi
To e
aerugi
ary wa
interva
side un
sample
placed
any inc
Acryli
solar r
-B, an
C) (Sp
points
toxins
Par
destro
from perimental stress conditions are described below.
rements of cellular stress
lish a link between cellular stress and toxin release,
aeruginosa were assayed for H2O2 production as
rrelate of stress response. This type of assay has
only employed as a marker for stress response in
ts (Orozco-Cardenas and Ryan, 1999) and marine
len and Pedersen, 1994; Ku?pper et al., 2002; Ross
). The response is based on the reaction of hydro-
de with the fluorogenic probe dichlorofluorescein
CFH-DA, Invitrogen Corp., Carlsbad, CA, USA).
is a non-fluorescing, non-polar compound. When
und reacts with cellular esterases the diacetate group
ff to yield the polar compound DCFH (2?7?-dichloro-
rescein). Oxidation of DCFH by hydrogen peroxide
uorescent product DCF (Collen and Davison, 1999).
was dissolved in dimethylsulfoxide in 10 mM aliquot
red at ?80 ?C).
asurement of H2O2 released from 1.05 ? 108 cells
in situ estuary water was used as a control value.
was measured on a LI-188B Integrating photometer
Lincoln, Nebraska, USA). At selected time points
ication of a stressor), 1 ml of media surrounding the
ria from each replicate was collected and added to a
taining 0.40 U/ml esterase and 25
M of DCFH-DA.
olume was increased to 2000
l in filtered seawater.
quantified fluorometrically on a Biorad VersaFluor
r (Bio-Rad; excitiation: 488 nm, emission: 525 nm)
ethods of Ross et al. (2005). Biologically relevant
directly to m
To examine
1 mM para
1.05 ? 108
of 5 h, 80

toxins.
To exam
50 ml estu
20 kHz son
son, Danbu
Jacobs (20
subsequent
2.6. Induc
oxidative s
To evalu
grammed c
incubated i
centrations
as a functio
units of ca
specific mo
H2O2, M.
frozen with
20 ml of 1
centrifuged
centrifuge.
tration was
Assay Kit
tions.6) 66?73
ons of commercially obtained H2O2 (Sigma) were
ndards.
ine the effect of increased salinity on the release of
s, M. aeruginosa cells were counted via hemocy-
d subsequently diluted to 1.05 ? 108 cells per repli-
per treatment). Cells were transferred to beakers
50 ml of full strength seawater (32?) and incubated
perature for 5 h. Post-incubation, 80
l of the sur-
ater was assayed for secreted toxins.
ine the oxidative stress initiated by solar radiation M.
samples (1.05 ? 108 cells immersed in 50 ml estu-
were exposed to full sunlight for predetermined time
= 3 per treatment). Experiments were conducted out-
a constant solar irradiance of 1600
E s?1 m?2. All
re placed in 50 ml conical tubes which were in turn
plastic container filled with chilled water to prevent
e in temperature during the course of the experiment.
ers were used to obtain the desired transmittance of
tion reaching the samples: UF 96 (absorbs UV-A,
) and UVT (transmits UV-A and -B but not UV-
h Polyclast, Stamford, CT, USA). At selected time
l of the surrounding water was assayed for secreted
t (Sigma) is a non-selective contact herbicide that
ant tissue by interrupting the steady flow of electrons
system I to NADPH. Instead, electrons are donated
olecular oxygen yielding highly powerful oxidants.
the effect of toxin release upon herbicide treatment,
quat was added to a 50 ml falcon tube containing
cells in estuary water. Following an incubation period
l of the surrounding water was assayed for secreted
ine the effect of sonic injury, cells (1.05 ? 108 cells/
ary water) were subjected to three 10 s pulses of
ic sound on a Branson 1510 ultrasonic cleaner (Bran-
ry, CT, USA) as previously described by Mydlarz and
04). Post-injury cells were gently mixed for 5 h and
ly analyzed for toxin release as described above.
tion of cellular programmed cell death by
tress
ate the relationship between oxidative stress and pro-
ell death, 1.05 ? 108 cells of M. aeruginosa were
n 50 ml of in situ estuary water with selected con-
of H2O2 and assayed for caspase proteolytic activity
n of time. Cells were incubated with H2O2 and 50
talase (Sigma) to verify that H2O2 was indeed the
lecule promoting PCD. Following incubation with
aeruginosa cells were collected via filtration, flash
liquid N2 and soluble proteins were extracted in
00 mM phosphate buffer (pH 7.8). The extract was
at 6600 ? g for 5 min at 4 ?C on a Beckman TJ-6
The supernatant was collected and protein concen-
quantified with a Quick StartTM Bradford Protein
(Bio-Rad) according to the manufacturer?s instruc-
8 (200
The Enz
lized to qu
cells of M.
olytic cleav
(DEVD). A
with 990

Invitrogen)
concentrati
ature for 25
the caspase
110 on a B
N = 3). The
CHO (Invit
samples w
20 min prio
was subtrac
2.7. Statist
To deter
one way A
test. The da
assumption
H2O2 prod
show untra
tion was us
analyses w
Tallahassee
3. Results
3.1. Detec
Samples
0.2?, tem
in Fig. 1A.
the PCR-am
specific pri
was obtain
identity wa
analyzed fo
ments with
identical id
tity (compa
sion numb
AY568035
gene, respo
in the St. L
3.2. Detec
Direct m
of in situ w
When cells
seawater) a
5 h when c
Ethidium bromide-stained 1.0% agarose electrophoresis gel showing
stin synthetase gene amplicons (?750 bp) from toxic Microcystis using
leotide primers FAA and RAA (Neilan et al., 1999). Lane 1, molecular
1 kb DNA plus ladder (Invitrogen, CA); lane 2, negative (no template)
lane 3, St. Lucie River species; lane 4, 1:10 dilution of St. Lucie River
DNA stock.C. Ross et al. / Aquatic Toxicology 7
chek? Caspase-3 Assay Kit #2 (Invitrogen) was uti-
antify programmed cell death activity in stressed
aeruginosa. This assay exploits the specific prote-
age of the amino acid sequence Asp?Glu?Val?Asp
liquots (1 ml) of the supernatant were combined
l of 1? reaction buffer (Caspase-3 Assay Kit #2,
and 10
l Z-DEVD-R110 substrate (final substrate
on, 25
M). Samples were incubated at room temper-
min and subsequently assayed for the appearance of
-catalyzed fluorescent cleavage product Rhodamine-
iorad VersaFluor fluorometer (Ex/Em: 496/520 nm;
reversible aldehyde caspase inhibitor Ac-DEVD-
rogen) was used as a negative control. Cyanobacterial
ere preincubated with 100
M Ac-DEVD-CHO for
r to the addition of H2O2. Background fluorescence
ted for no-enzyme controls.
ics
mine if the treatments had an effect on toxin release a
NOVA was conducted followed by an LSD post-hoc
ta were rank transformed since they did not meet the
s of normality nor equal variances. The analysis for
uction was conducted in the same manner. All figures
nsformed data. A Pearson product moment correla-
Fig. 2.
microcy
oligonuc
marker
control;
speciesed to correlate toxin secretion and H2O2 release. All
ere conducted using Statistix 7 (Analytical Software,
, FL, USA).
tion of mcyB
of M. aeruginosa were collected in situ (salinity
perature 32.7 ?C) from a highly dense area as shown
The presence of the mcyB gene was established by
plification of M. aeruginosa DNA with mcyB gene-
mers. A PCR amplicon of the expected size (759 bp)
ed (Fig. 2). The PCR product was sequenced and its
s confirmed by BLAST analysis. This sequence was
r its homology to other mcyB gene sequences. Align-
other published mcyB sequences resulted in almost
entity matches ranging from 97.6% to 99.7% iden-
red to published mcyB gene sequences with acces-
ers AB092806, AJ224717, AJ224726, AJ492554,
). This data confirms that the presence of the mcyB
nsible for the production of microcystins, was present
ucie River M. aeruginosa assemblage.
tion of extracellular toxins
easurement of toxins from 1.05 ? 108 cells in 50 ml
ater, resulted in associated values around 3.5
g l?1.
were subjected to an osmotic change in media (32?
n 80% increase in toxin level was detected within
ompared to the 3.5
g l?1 toxin value observed in
non-stresse
filtering pa
nificant inc
versely, wh
plates (UV
compared t
cells (Fig.
in detectibl
of soluble
than the co
Fig. 3. Measu
stressed cells
Cells of M. ae
seawater for 5
tion (UVT) an
were incubate
disrupted by
are represente
cated by the le
followed by a6) 66?73 69d cells (Fig. 3). The absorbing ability of ultraviolet
nels (UF 96UV) was capable of preventing any sig-
rease in soluble toxin over the same time course. Con-
en cells were incubated under ultraviolet transmitting
T) a 55% increase in soluble toxin was detected when
o the 3.5
g l?1 toxin value detected in non-stressed
3). The use of paraquat resulted in a 90% increase
e toxins. Physical injury yielded the highest release
toxins with spectrophotometric values 95% greater
ntrol samples. All stressors (aside from the use of UF
rement of toxins (as microcystin LR equivalents) secreted from
compared to undisturbed cells maintained in natural estuary water.
ruginosa were osmotically stressed by being transferred to 32?
h (Osmotic). Cells were stressed by full exposure to solar radia-
d compared to Ultraviolet filtered solar irradiation (UF 96). Cells
d with a lethal dose of the herbicide (Paraquat), and physically
ultrasonication (Sonicate) (n = 3 for each experiment). Data bars
d as the mean with error bars of ?1. Significant groupings are indi-
tters above bars, analyses were conducted by a one-way ANOVA
n LSD post-hoc test.
70 8 (200
Fig. 4. Measu
bars are repre
are indicated
ANOVA follo
96UV) resu
(p < 0.001)
3.3. Hydro
stress
H2O2 w
of salinity,
irradiation
?0.25 pmo
would be e
such as H2
thesis. Wh
elevated to
ing materia
when com
(UVT) to
0.60 pmol H
lular homo
oxygen spe
the cyanob
under 2.0 p
UF 96UV)
to control c
relation be
r2 = 0.59).
3.4. Progr
Exogen
vitro (Fig.
post-introd
10
M H2O
9
mol rho
H2O2 elici
tial detecti
24 h-post in
rhodamine
with exoge
level of cas
100
M H2
suppressedC. Ross et al. / Aquatic Toxicology 7rement of H2O2 released as a function of stress treatment. Data
sented as the mean with error bars of ?1. Significant groupings
by the letters above bars, analyses were conducted by a one-way
wed by an LSD post-hoc test.
lted in a significant increase of toxin concentration
.
gen peroxide production in response to cellular
as released by M. aeruginosa following an increase
physical injury, application of paraquat, or UV
(Fig. 4). Control cells were shown to release
l H2O2 cell?1 after sitting undisturbed for 5 h. This
xpected as basal levels of reactive oxygen species,
O2, are exuded as simple byproducts of photosyn-
en cells were placed in 32? seawater, H2O2 levels
0.75 pmole H2O2 cell?1. The use of the UV absorb-
l UF 96UV prevented any change in H2O2 release
pared to the control. The ability for UV radiation
reach the cells resulted in H2O2 levels reaching
2O2 cell?1. The use of paraquat to compromise cel-
estasis resulted in a considerable release of reactive
cies (3.25 pmol H2O2 cell?1). Physical disruption of
acterial cells resulted in H2O2 concentrations just
mol H2O2 cell?1. All stressors (aside from the use of
resulted in a significant production of H2O2 relative
ells (p < 0.001). There was a significant positive cor-
tween toxin release and H2O2 production (p < 0.001,
ammed cell death by oxidative stress
ous addition of H2O2 elicited caspase activity in
5). Caspase activity was initially detected by 6 h
uction when M. aeruginosa cells were incubated in
2. By 18 h post-incubation, caspase activity reached
damine 110 mg?1 protein. The addition of 100
M
ted a more rapid response in caspase activity (ini-
on by 1 h post introduction of 100
M H2O2). By
troduction caspase activity was detected at 50
mol
110 mg?1 protein. Control cells that were not treated
nous amounts of H2O2 did not show any significant
pase activity (Fig. 5). When cells were incubated with
O2 in addition to 50 U catalase, caspase activity was
for up to 10 h. After this time point caspase activ-
Fig. 5. Time
caspase activi
selected conc
was quantified
the fluorescen
and the revers
used as negat
protein.
ity slowly i
24 h. When
inhibitor A
capase acti
4. Discuss
M. aeru
been linked
1984; Jew
logical and
great succe
M. aerugin
balance an
Cyanobact
to their hig
M. aerugin
organism to
tage over o
Whitford a
High co
late with hi
cyanobacte
upon cell l
(White et a
acteristical
bloom ages
lular toxins
collection o
in Fig. 1A)
Upon expo
capable of
over 90% a
blages of M6) 66?73course of the onset of programmed cell death as measured by
ty (DEVD cleavage) in M. aeruginosa. Cells were incubated with
entrations of H2O2 (n = 3 for each experiment). Caspase activity
by the cleavage of Z-DEVD-R110 resulting in the production of
t product rhodamine 110 (Ex/Em: 496/520 nm). Catalase (1 U/ml)
ible aldehyde caspase inhibitor Ac-DEVD-CHO (100
M) were
ive controls. Units are expressed in 
mol rhodamine 110 mg?1
ncreased to 9
mol rhodamine 110 mg?1 protein by
cells were preincubated with the reversible caspase
c-DEVD-CHO prior to the addition of 100
M H2O2,
vity was drastically reduced (Fig. 5).
ion
ginosa is a ubiquitous cyanobacterium that has often
to toxic blooms world-wide (Watanabe and Oishi,
el et al., 2003; Silva, 2003). A series of morpho-
physiological characteristics may account for its
ss amongst the phytoplankton community. Primarily,
osa requires little energetic input to sustain cellular
d is capable of persisting in nutrient deplete areas.
eria can outcompete other planktonic organisms due
h affinity for phosphorus and nitrogen. In addition,
osa contains numerous gas vacuoles allowing this
modify its buoyancy, again giving it a distinct advan-
ther phytoplankton in the community (Smith, 1950;
nd Schumacher, 1984).
ncentrations of Microcystis do not necessarily corre-
gh levels of microcystins in the water column. Many
ria retain cyanotoxins within their cell structure, and
ysis release these toxins into the surrounding water
l., 2005). The early stages of a toxic bloom are char-
ly associated mostly with intracellular toxins. As the
, cell death ensues and the concentration of extracel-
increases (Lahti et al., 1997, White et al., 2005). Our
f water from a dense bloom (surface scum as shown
displayed surprisingly low toxin values (3.5
g l?1).
sure to selected stressors, cell bound toxins were
being released into the immediate vicinity at levels
bove what was normally secreted by dense assem-
. aeruginosa.
8 (200
In photo
a variety of
xenobiotics
cyanobacte
coordinate
of program
Levine et a
let radiatio
effect on th
teria (Tyag
from the d
the leakage
port chain t
species (RO
tive lipid p
oxidation o
phycobilin
M. aerugin
specimens
2001). Asi
was no men
indicate tha
full sunligh
elevation i
under UV t
tion of H2
ultraviolet
inosa. Mor
panied with
media, whe
rial showed
In an att
strategy, sp
at salt conc
ity values s
reduction in
lysis (Atkin
an 80% in
to water w
tant with an
correlates w
Since cy
of algicide
Studies hav
blooms of M
per sulfate o
and Orr, 1
was detecte
studies. Th
of algicide
(Jones and
viologen) i
in a catalyt
gen and ch
are produc
the organis
in toxin rel
osur
rs ha
ariet
(Lam
orop
dino
in an
2004
atin
(Da
erich
irec
n as
(Tho
ppea
y co
tion
ive d
mo
et a
amp
hode
dativ
yano
pase
sinceC. Ross et al. / Aquatic Toxicology 7
synthetic organisms H2O2 is produced in response to
exogenous factors including irradiation, pesticides,
, or pathogens (Xue et al., 2005). In several species of
ria and unicellular algae, the production of H2O2 can
a series of cellular responses including the induction
med cell death or apoptosis (Korsmeyer et al., 1995;
l., 1996; Vardi et al., 1999). For example, ultravio-
n (particularly UV-B: 280?320 nm) has a detrimental
e development and general metabolism of cyanobac-
i et al., 1992; Sinha and Hader, 1998, 2002). Aside
irect damage to key proteins, enzymes and DNA,
of electrons from the photosynthetic electron trans-
o oxygen enhances the formation of reactive oxygen
S). The indirect damage by ROS includes oxida-
eroxidation, inhibition of photosynthesis, and the
f photosynthetic pigments such as chlorophylls and
s (He et al., 2002). A recent study demonstrated that
osa released H2O2 into the surrounding media when
were irradiated with an ultraviolet source (Alam et al.,
de from the observation that cell death ensued there
tion of toxin release into the environment. Our results
t when cultures of M. aeruginosa were irradiated by
t, the UV absorbing material UF 96UV prevented any
n H2O2 release. Conversely, M. aeruginosa placed
ransmitting material significantly increased produc-
O2. This observation supports the hypothesis that
Exp
stresso
in a v
plants
lar chl
2003),
Frankl
et al.,
chrom
plasm
(Hoeb
to be d
functio
mals)
It a
directl
produc
oxidat
ulatory
(Slater
(Van C
in Tric
of oxi
other c
Cas
zoansradiation triggers the formation of ROS in M. aerug-
e importantly, elevated levels of H2O2 were accom-
a 40% increase in toxin release into the surrounding
reas M. aeruginosa protected by UV-blocking mate-
no sign of H2O2 increase or toxin elevation.
empt to use salinity alterations as a bloom elimination
ecimens of M. aeruginosa were found to be viable
entrations up to 9.8? (Atkins et al., 2000). Salin-
urpassing this critical mark were accompanied with a
total viable cell concentration and an increase in cell
s et al., 2000; Orr et al., 2003). Our data show nearly
crease in toxin release when cells were transferred
ith a salinity of 32?. This increase was concomi-
increase in secreted H2O2, showing oxidative stress
ith toxin release.
anotoxins are stored intracellularly, the application
s should be used with caution to avoid toxin release.
e demonstrated that microcystins are released when
. aeruginosa are treated with algicides such as cop-
r sodium carbonate peroxyhydrate (Pak-27TM; Jones
994; Touchette et al., 2005). Post lysis, free toxin
d at concentrations reaching 1.8
g l?1 in mesocosm
is release can be quite rapid and occurs within 3?24 h
application depending on the administered dose
Orr, 1994; Kenefick et al., 1993). Paraquat (methyl
s a well known herbicide/algicide that elicits toxicity
ic manner requiring the presence of molecular oxy-
loroplast photo-activation. Ultimately free radicals
ed that compromise the photosynthetic integrity of
m. Our studies with paraquat showed a 90% increase
ease compared to control specimens.
in photosy
related pro
iterative ho
pase or ?ca
fungi, mult
chlorophyt
and plants
Lam, 1998
caspase ac
spp. (Berm
to utilize c
nosa (Hug
caspase act
sible for the
oxidative s
that when c
DEVD-CH
findings are
caspase act
systems bu
et al., 2000
in conjunct
chodesmiu
PCD has an
than previo
In conc
of M. aeru
demonstrat
lar methodo
to biosynth
tions rangi6) 66?73 71
e of cells to low levels of physical and chemical
ve been known to trigger apoptosis-like conditions
y of photosynthetic organisms including vascular
and del Pozo, 2000; Lam et al., 2001), unicellu-
hytes (Berges and Falkowski, 1998; Segovia et al.,
flagellates (Vardi et al., 1999; Franklin et al., 2004;
d Berges, 2004), and cyanobacteria (Berman-Frank
). Characteristics of cells undergoing PCD include
condensation, shrinkage of the nucleus and cyto-
non et al., 2000), and ordered DNA fragmentation
ts and Woitering, 2003). These changes are believed
ted by caspases (a family of cysteine proteases that
a regulatory switch for many forms of PCD in ani-
rnberry and Lazebnik, 1998).
rs that toxin release in stressed M. aeruginosa was
rrelated with an H2O2-elicited signal for PCD. The
of H2O2 can trigger the initiation of PCD by direct
amage to DNA, or by the indirect oxidation of reg-
lecules that ultimately provide an entrance into PCD
l., 1995). This process is termed oxidative cell stress
et al., 1998). Aside from what has been documented
smium (Berman-Frank et al., 2004), the involvement
e stress on PCD has not been previously shown in
bacterium.
activity was initially thought to be limited to meta-
direct sequence homologies have not been identifiednthetic organisms. However, a family of caspase-
teases has now been recognized in higher plants via
mology matches (Del Pozo and Lam, 1998). Cas-
spase-like? proteolytic activity is currently known in
iple bacterial species (Uren et al., 2000), unicellular
es (Segovia et al., 2003; Segovia and Berges, 2005)
during the hypersensitive response (Del Pozo and
, Chichkova et al., 2004). The most recent findings of
tivity in the marine cyanobacterium Trichodesmium
an-Frank et al., 2004) prompted our investigation
aspase activity as a proxy for PCD in M. aerugi-
et al., 1999, Liu et al., 1999). We demonstrated that
ivity in stressed M. aeruginosa cells might be respon-
onset of PCD when induced with H2O2 as a result of
tress (Fig. 4). This conclusion is supported by the fact
ells were preincubated with the caspase inhibitor Ac-
O, caspase activity was essentially eliminated. These
in agreement with recent reports demonstrating that
ivity is a common constituent not only in mammalian
t in higher plants and lower algae as well (Korthout
, Elbaz et al., 2002, Segovia et al., 2003). Our results,
ion with the recent caspase-like activity found in Tri-
m (Berman-Frank et al., 2004), support the notion that
earlier evolutionary origin and broader significance
usly thought.
lusion, this is the first incidence of a large bloom
ginosa in the St. Lucie River, Estuary. We have
ed, through the use of immunodetection and molecu-
logies, that this assemblage has the genetic potential
esize microcystins and release them at concentra-
ng from 3.5 to 6.8
g l?1. Considering the World
72 8 (20
Health Org
microcystin
10
g l?1 f
remediatio
that are ba
risks are n
removal of
ing the sal
bloom rem
cyanobacte
into the su
released to
assemblage
weeks (Jon
1999). The
rial blooms
inputs and
been identi
ities (Chor
Remediatio
ment the pr
Acknowled
We than
Russell G.
ular biolog
Raphael R
sis and mi
from the Fl
Biotechnol
NSF Mino
upon work
a grant aw
#0310283)
mendations
and do not
Foundation
Pierce con
Biomedica
Reference
Alam, M.Z.B
inactivatio
(4), 1008?
Atkins, R., Ro
teria bloo
107?114.
Baker, J.A., N
cyanobact
molecular
Berges, J.A.,
marine ph
light limi
Berman-Fran
demise of
catalyzed
urt-O
obact
ful A
J., W
in Flo
eedin
e of t
, D.E
e of
?123
ael, W
n alg
va, N
tsov,
t casp
t Cel
., Sn
entou
a. J.
I., M
ds.),
seque
ll, R.
ht, A
n from
temp
161.
.A.,
. Cy
col. 3
J., PeC. Ross et al. / Aquatic Toxicology 7
anization (WHO) placed a provisional guideline on
concentrations of 1.0
g l?1 for potable water and
or recreational use, it is critical that cyanobacterial
n strategies be approached with caution. Guidelines
sed upon cell concentrations for calculating human
ot adequate. The utilization of algicides, physical
surface scum using oil spill equipment, and increas-
inity of water reservoirs are all currently used as
ediation strategies that are capable of resulting in
rial stress and subsequently the release of toxins
rrounding water column. Residence times for these
xins are subject to water composition and bacterial
s yet could persist in the water column for up to 3
es and Orr, 1994; Lahti et al., 1997; Chriswell et al.,
key management plan for minimizing cyanobacte-
is to intervene at the source of the problem. Nutrient
hydrologic sources are two such examples that have
fied as bloom targets by watershed management facil-
us and Mur, 1999; Paerl et al., 2001; Piehler, 2005).
n strategies further downstream may potentially aug-
oblem rather than improve the quality of the water.
gements
k Sherry Reed for assistance with collections, Dr.
Bittenco
cyan
Harm
Burns,
ins
Proc
Stat
Canfield
danc
1232
Carmich
gree
Chichko
Rub
plan
Plan
Choo, K
filam
rian
Chorus,
J. (E
Con
Chriswe
wrig
toxi
pH,
155?
Codd, G
1999
Phy
Collen,
Kerr for the use of facilities and equipment for molec-
y studies, Dr. Jared Lucas for useful suggestions, and
itson-Williams for assistance with statistical analy-
crophotography. We acknowledge financial support
orida Center of Excellence in Biomedical and Marine
ogy. Lory Z. Santiago-Va?zquez was funded by an
rity postdoctoral fellowship. This material is based
supported by the National Science Foundation under
arded in 2003 to Lory Z. Santiago-Va?zquez (award
. Any opinions, findings, and conclusions or recom-
expressed in this publication are those of the authors
necessarily reflect the views of the National science
. This represents Smithsonian Marine Station at Ft.
tribution #642 and Florida Center of Excellence in
l and Marine Biotechnology #P200606.
s
., Otaki, M., Furumai, H., Ohgaki, S., 2001. Direct and indirect
n of Microcystis aeruginosa by UV-radiation. Water Res. 35
1014.
se, T., Brown, R.S., Robb, M., 2000. The Microcystis cyanobac-
m in the Swan River?February. Water Sci. Technol. 43 (9),
eilan, B.A., Entsch, B., McKay, D.B., 2001. Identification of
eria and their toxigenicity in environmental samples by rapid
analysis. Environ. Toxicol. 16, 472?482.
Falkowski, P.G., 1998. Physiological stress and cell death in
ytoplankton: Induction of proteases in response to nitrogen or
tation. Limnol. Oceanogr. 43, 129?135.
k, I., Bidle, K.D., Haramaty, L., Falkowski, P.G., 2004. The
the marine cyanobacterium, Trichodesmium spp., via an auto-
cell death pathway. Limnol. Oceanogr. 49 (4), 997?1005.
platycladu
Collen, J., D
intertidal
Danon, A., D
cell death
del Pozo, O.
hypersens
1132.
Elbaz, M., Av
cutes prog
Franklin, D.J.
and degen
pistillata
Franklin, D.J
Amphidin
Soc. Lond
Gunn, G.F., R
tie, K.A.,
blue-green
He, Y., Hader
damage to
66, 73?80
He, Y., Klisc
stress cor
Photobiol
Hoeberichts,
grammed
plant-spec
Hug, H., Lo
amino aci
apoptosis
Jewel, M.A.S
terial bloo
6 (12), 10
Jochimsen, E
Holmes, C
V., Azeve06) 66?73
liveira, M., 2003. Detection of potential microcystin-producing
eria in Brazilian reservoirs with a mcyB molecular marker.
lgae 2, 51?60.
illiams, C., Chapman, A., 2002. Cyanobacteria and their tox-
rida Surface Waters. In: Johnson, D., Harbinson, R.D. (Eds.),
gs of the Health Effects of Exposure to Cyanobacteria Toxins:
he Science. August 13?14.
., Phlips, E.J., Duarte, C., 1989. Factors influencing the abun-
blue-green algae in Florida lakes. Can. J. Fish. Aquat. Sci. 46,
7.
., 1992. A Status Report on Planktonic Cyanobacteria (blue-
ae) and their Toxins. EPA/600/R-92/079.
.V., Kim, S.Y., Titova, E.S., Kalkum, M., Morozov, V.S.,
Y.P., Kalinina, N.O., Taliansky, M.E., Vartapetian, A.B., 2004. A
ase-like protease activated during the hypersensitive response.
l 16, 157?171.
oeijs, P., Pedersen, M., 2004. Oxidative stress tolerance in the
s green algae Cladophora glomerata and Enteromorpha ahlne-
Exp. Mar. Biol. Ecol. 298, 111?123.
ur, L., 1999. Preventative measures. In: Chorus, I., Bartram,
Toxic Cyanobacteria in Water: A Guide to their Public Health
nces, Monitoring and Management. F and FN Spon, London.
K., Shaw, G.R., Eaglesham, G., Smith, M.J., Norris, R.L., Sea-
.A., Moore, M.R., 1999. Stability of cylindrospermopsin, the
the cyanobacterium, Cylindrospermopsis raciborskii: effect of
erature and sunlight on decomposition. Environ. Toxicol. 14,
Bell, S.G., Kaya, K., Ward, C.J., Beattie, K.A., Metcalf, J.S.,
anobacterial toxins, exposure routes and human health. Eur. J.
4, 405?415.
dersen, M., 1994. A stress-induced oxidative burst in Eucheuma
m (Rhodophyta). Physiol. Plantarum. 92, 417?422.
avison, I.R., 1999. Reactive oxygen production and damage in
Fucus spp. (Phaeophycae). J. Phycol. 35, 54?61.
elorme, V., Mailhac, N., Gallois, P., 2000. Plant programmed
: A common way to die. Plant Physiol. Biochem. 38, 647?655.
, Lam, E., 1998. Caspases and programmed cell death in the
itive response of plants to pathogens. Curr. Biol. 8, 1129?
ni, A., Weil, M., 2002. Constitutive caspase-like machinery exe-
rammed cell death in plant cells. Cell Death Diff. 9, 726?733.
, Hoegh-Guldberg, O., Jones, R.J., Berges, J.A., 2004. Cell death
eration in the symbiotic dinoflagellate of the coral Stylophora
during bleaching. Mar. Ecol. Prog. Ser. 272, 117?130.
., Berges, J.A., 2004. Mortality in cultures of the dinoflagellate
ium carterae during culture senescence and darkness. Proc. R.
. B 271, 2099?2107.
afferty, A.G., Rafferty, G.C., Cockburn, N., Edwards, C., Beat-
Codd, G.A., 1992. Fatal canine neurotoxicosis attributed to
algae (Cyanobacteria). Vet. Rec. 130, 301?302.
, D., 2002. Involvement of reactive oxygen species in the UV-B
the cyanobacterium Anabaena sp. J. Photochem. Photobiol. B
.
h, M., Hader, D., 2002. Adaptation of cyanobacteria to UV-B
related with oxidative stress and oxidative damage. Photochem.
. 76 (2), 188?196.
F.A., Woitering, E.J., 2003. Multiple mediators of plant pro-
cell death: interplay of conserved cell death mechanisms and
ific regulators. Bioessays 25, 47?57.
s, M., Werner, H., Debatin, K., 1999. Rhodamine 110-linked
ds and peptides as substrates to measure caspase activity upon
induction in intact cells. Biochemistry 38, 13906?13911.
., Affan, M.A., Khan, S., 2003. Fish mortality due to cyanobac-
m in an aquaculture pond in Bangladesh. Pakistan J. Biol. Sci.
46?1050.
.M., Carmichael, W.W., An, J., Cardo, D., Cookson, S.T.,
.E.M., Antunes, M.B.C., Melo-Filho, D.A., Lyra, T.M., Barreto,
do, S.M.F.O., Jarvis, W.R., 1998. Liver failure and death fol-
8 (20
lowing ex
New Eng
Jones, G., Or
algicide t
lake, as d
Water Re
Kaebernick, M
of cyanot
Kenefick, S.L
from Micr
27 (3?4),
Korsmeyer, S
1995. Re
Bcl-2 gen
Korthout, H.A
The prese
plant cell
Ku?pper, F.C.,
ginate rec
induced r
(10), 2057
Lahti, K., Rap
of cyanob
dissolved
Lam, E., Del
of plant c
Lam, E., Kat
and the p
Levine, A., P
mediated
Curr. Bio
Liu, J., Bhal
1999. Flu
fluoromet
Mydlarz, L.D
burst in fr
pseudopte
Neilan, B.A.,
nen, K., B
ity of cya
Orozco-Carde
temically
pathway.
Orr, P.T., Jon
aeruginos
and conse
Freshwate
Paerl, H.W.,
water alg
76?113.
Palenik, B., Z
tion by a
Phlips, E.J.,
Ritter, P.,
eters in a
Hydrobio
Phlips, E.J., 2
dia of En
Phlips, E.J.,
potentially
Effects of
13?14.
Piehler, M.F.,
rial harmf
Symposiu
Park, NC
H.D.
e-gree
Loc
., Ku?
e of
t unic
541.
, M.,
e uni
ution
05.
, M.,
DNA
activ
olecta
.I.L.,
, Kan
.P., H
cyan
.P., H
ochem
g, O.M
ms in
.F.G
ants i
.M.,
York
rry, N
1312
te, B.
in relC. Ross et al. / Aquatic Toxicology 7
posure to microcystin toxins at a hemodialysis center in Brazil.
l. J. Med. 36, 373?378.
r, P.T., 1994. Release and degradation of microcystin following
reatment of a Microcystis aeruginosa bloom in a recreational
etermined by HPLC and protein phosphatase inhibition assay.
s. 28 (4), 871?876.
., Neilan, B.A., 2001. Ecological and molecular investigations
oxin production. FEMS Microbiol. Ecol. 35, 1?9.
., Hrudey, S.E., Peterson, H.G., Prepas, E.E., 1993. Toxin release
ocystis aeruginosa after chemical treatment. Water Sci. Technol.
433?440.
.J., Yin, X.M., Oltvai, Z.N., Veis-Novack, D.J., Linette, G.P.,
active oxygen species and the regulation of cell death by the
e family. Biochim. Biophys. Acta 1271, 63?66.
.A.J., Berecki, G., Bruin, W., van Duijn, B., Wang, M., 2000.
nce and subcellular localization of caspase 3-like proteinases in
s. FEBS Lett. 475, 139?144.
Muller, D.G., Peters, A.F., Kloared, B., Potin, P., 2002. Oligoal-
ognition and oxidative burst play a key role in natural and
esistance of sporophytes of Laminariales. J. Chem. Ecol. 28
?2081.
ala, J., Fardig, M., Niemela, M., Sivonen, K., 1997. Persistance
acterial hepatotoxin, Microcystin-LR, in particulate material and
in lake water. Water Res. 31, 1005?1012.
Pozo, O., 2000. Caspase-like protease involvement in the control
ell death. Plant Mol. Biol. 44, 417?428.
o, N., Lawton, M., 2001. Programmed cell death, mitochondria
lant hypersensitive response. Nature 411, 848?853.
ennell, R., Alvarez, M., Palmer, R., Lamb, C.J., 1996. Calcium
apoptosis in a plant hypersensitive disease resistance response.
l. 6, 427?437.
Rodger,
(blu
L. in
Ross, C
denc
gian
531?
Segovia
in th
evol
99?1
Segovia
and
like
terti
Silva, E
body
Sinha, R
field
Sinha, R
Phot
Skulber
bloo
Slater, A
oxid
Smith, G
New
Thornbe
281,
Touchet
otoxgat, M., Zhang, C., Diwu, Z., Hoyland, B., Klaubert, D.H.,
orescent molecular probes V: a sensitive caspase-3 substrate for
ric assays. Bioorg. Med. Chem. Lett. 9, 3231?3236.
., Jacobs, R.S., 2004. Comparison of an inducible oxidative
ee-living and symbiotic dinoflagellates reveals properties of the
rosins. Phytochemistry 65, 3231?3241.
Dittmann, E., Rouhiainen, L., Bass, R.A., Schaub, V.A., Sivo-
orner, T., 1999. Non-ribosomal peptide synthesis and toxigenic-
nobacteria. J. Bacteriol. 181 (13), 4089?4097.
nas, M., Ryan, C.A., 1999. Hydrogen peroxide is generated sys-
in plant leaves by wounding and systemin via the octadecanoid
Proc. Natl. Acad. Sci. U.S.A. 96 (11), 6553?6557.
es, G.J., Douglas, G.B., 2003. Response of cultured Microcystis
a from the Swan River, Australia, to elevated salt concentration
quences for bloom and toxin management in estuaries. Mar.
r Res. 55 (3), 227?283.
Fulton, R.S., Moisander, P.H., Dyble, J., 2001. Harmful fresh-
al blooms, with an emphasis on cyanobacteria. Sci. World 1,
afiriou, O.C., Morel, F.M.M., 1987. Hydrogen peroxide produc-
marine phytoplankter. Limnol. Oceanogr. 32 (6), 1365?1369.
Aldridge, F.J., Hansen, P., Zimba, P.V., Ihnat, J., Conroy, M.,
1993. Spatial and temporal variability of trophic state param-
shallow subtropical lake (Lake Okeechobee, FL, USA). Arch.
l. 128, 437?458.
002. Eutrophication and algae. In: Bitton, G. (Ed.), Encyclope-
vironmental Microbiology. John Wiley & Sons, New York.
Bledsoe, E., Badylak, S., Frost, J., 2002. The distribution of
toxic cyanobacteria in Florida. In: Proceedings of the Health
Exposure to Cyanobacteria Toxins: State of the Science, August
2005. Watershed management strategies to control cyanobacte-
ul algal blooms. In: Proceedings of the Interagency International
m on Cyanobacterial Harmful Algal Blooms, Research Triangle
, September 6?10.
carbonate
Symposiu
Park, NC
Twiner, M.J.,
duction o
akashiwo.
Tyagi, R., Sr
ential eff
in a chro
407.
Uren, A.G.,
Koonin, E
caspases:
a key role
Vaitomaa, J.,
L., Sivon
Microcyst
in lakes.
Van Camp, W
signals in
Vardi, A., Be
A., 1999
gatunense
9, 1061?1
Watanabe, M
Microcyst
White, S.H.,
framewor
cyanotoxi
10, 25?37
Whitford, L.A
Sparks Pr
Xue, L., Zh
ultraviolet
31, 79?8906) 66?73 73
, Turnbull, T., Edwards, C., Codd, G.A., 1994. Cyanobacterial
n-algal) bloom associated pathology in brown trout Salmo trutta
h Leven. Scotland J. Fish. Dis. 17, 177?181.
pper, F.C., Vreeland, V.J., Waite, J.H., Jacobs, R.S., 2005. Evi-
a latent oxidative burst in relation to wound repair in the
ellular chlorophyte Dasycladus vermicularis. J. Phycol. 41 (3),
Haramaty, L., Berges, J.A., Falkowski, P.G., 2003. Cell death
cellular chlorophyte Dunaliella tertiolecta: an hypothesis on the
of apoptosis in higher plants and metazoans. Plant Physiol. 132,
Berges, J.A., 2005. Effect of inhibitors of protein synthesis
replication on the induction of proteolytic activities, caspase-
ities and cell death in the unicellular chlorophyte Dunaliella
. Eur. J. Phycol. 40, 21?30.
2003. Emergence of a Microcystis bloom in an urban water
dy Lake. Sri Lanka Curr. Sci. 85 (6), 723?725.
ader, D.P., 1998. Effects of ultraviolet-B radiation in three rice
obacteria. J. Plant Physiol. 153, 763?769.
ader, D.P., 2002. UV-induced DNA damage and repair: a review.
. Photobiol. Sci. 1, 225?236.
., Codd, G.A., Carmichael, W.W., 1984. Toxic blue-green-algal
Europe?a growing problem. Ambio 13, 244?247.
., Nobel, C.S.I., Orrenius, S., 1995. The role of intracellular
n apoptosis. Biochim. Biophys. Acta 1271, 59?62.
1950. The Freshwater Algae in the United States. McGraw-Hill,
, NY.
.A., Lazebnik, Y., 1998. Caspases: enemies within. Science
?1316.
W., Edwards, C.T., Alexander, J., 2005. A comparison of cyan-
ease following bloom treatments with copper sulfate or sodium
peroxyhydrate. In: Proceedings of the Interagency International
m on Cyanobacterial Harmful Algal Blooms, Research Triangle
, September 6?10.
Trick, C.G., 2000. Possible physiological mechanisms for pro-
f hydrogen peroxide by the ichthyotoxic flagellate Heterosigma
J. Plankton Res. 22 (10), 1961?1975.
inivas, G., Vyas, D., Kumar, A., Kumar, H.D., 1992. Differ-
ect of ultraviolet-B radiation on certain metabolic processes
matically adapting Nostoc. Photochem. Photobiol. 55, 401?
O?Rourke, K., Aravind, L., Pisabarro, M.T., SEshagiri, S.,
.V., Dixit, V.M., 2000. Identification of paracaspases and meta-
two ancient families of caspase-like proteins, one of which plays
in MALT lymphoma. Mol. Cell 6, 961?967.
Rantala, A., Halinen, K., Rouhiainen, L., Tallberg, P., Mokelke,
en, K., 2003. Quantitative real-time PCR for determination of
in Synthetase E copy numbers for Microcystis and Anabaena
Appl. Environ. Microbiol. 69 (12), 7289?7297.
., Van Montagu, M., Inze, D., 1998. H2O2 and NO: redox
disease resistance. Trends Plant Sci. 3 (9), 330?334.
rman-Frank, I., Rozenberg, T., Hadas, O., Kaplan, A., Levine,
. Programmed cell death of the dinoflagellate Peridinium
is mediated by CO2 limitation and oxidative stress. Curr. Biol.
064.
.F., Oishi, S., 1984. Toxic substances from a natural bloom of
is aeruginosa. Appl. Environ. Microbiol. 43 (4), 819?822.
Duivenvoorden, L.J., Fabbro, L.D., 2005. A decision-making
k for ecological impacts associated with the accumulation of
ns (cylindrospermopsin and microcystin). Lake Reserv. Manage.
.
., Schumacher, G.J., 1984. A Manual of Freshwater Algae.
ess, Raleigh, NC.
ang, Y., Zhang, T., Wang, X., 2005. Effects of enhanced
-B radiation on algae and cyanobacteria. Crit. Rev. Microbiol.
.