lyngby a

7/26/2019 Lyngby A

1/12

The Freshwater Cyanobacterium Lyngbyaaerugineo-coeruleaProduces Compounds Toxicto Mice and to Mammalian and Fish Cells

Ivanka Teneva,1 Dafinka Asparuhova,2 Balik Dzhambazov,3 Rumen Mladenov,4

Kristin Schirmer1

1Junior Research Group, Molecular Animal Cell Toxicology, UFZ Centre for Environmental

Research, Permoserstrasse 15, 04318 Leipzig, Germany

2Cell Biology Laboratory, University of Plovdiv, 24 Tsar Assen Street, 4000 Plovdiv, Bulgaria

3Section for Medical Inflammation Research, Department of Cell and Molecular Biology,

Lund University, I 11, BMC, 22184 Lund, Sweden

4Department of Botany, University of Plovdiv, 24 Tsar Assen Street, 4000 Plovdiv, Bulgaria

Received 26 June 2002; revised 4 September 2002; accepted 26 September 2002

ABSTRACT: Despite a growing awareness of the presence of cyanobacterial toxins, knowledge about the

ability of specific species to produce toxic compounds is still rather limited. It was the overall goal of the current

work to investigate if probes derived from the freshwater species Lyngbya aerugineo-coerulea (Kutz.) Gomont,

a cyanobacterium frequently found in southern Europe and not previously investigated for the presence of

bioactive compounds, were capable of elicitingin vivo andin vitro toxicity. The cyanobacterial extract revealed

signs of neuro- as well as hepatotoxicity in mice, although these signs could not be explained by the

well-known respective cyanobacterial neuro- and hepatotoxins saxitoxin and microcystin. Cytotoxicity was

elicited by the cyanobacterial extract in all mammalian cell lines tested. As well, the rainbow trout liver cell line,

RTL-W1, was found to be susceptible to the cytotoxic effects of the extract, although the cytotoxicity was

dependent on temperature. In contrast, the cyanobacterial growth medium elicited cytotoxicity independent oftemperature, leading to morphological changes indicative of alterations to the cytoskeleton. Overall, the results

suggest that Lyngbya aerugineo-coeruleais an important cyanobacterium to be considered for its potential to

cause health risks on environmental exposure of it to mammals and fish. Applying a combination of mam-

malian and piscine cell line bioassays is a unique approach that, combined with chemical analysis, could

be used in the future to identify the structure and cellular mechanisms of the as-yet-unknown toxic Lyng-

bya aerugineo-coerulea metabolites in particular and to screen cyanobacterial extracts for their toxicity in

general. 2003 Wiley Periodicals, Inc. Environ Toxicol 18: 9 20, 2003.

Keywords: cyanobacteria; Lyngbya aerugineo-coerulea; toxins; mouse bioassay; fish and mammalian cell

cultures; cytotoxicity

INTRODUCTION

Many cyanobacteria (blue-green algae) have been found to

be producers of potent bioactive compounds. The incentive

for identifying these compounds and their biological effects

has been being able to find novel substances applicable to

biomedical research and gaining an understanding of the

exposure risks for humans as well as animals. Toxins pro-

duced by cyanobacteria have been reported in marine- as

Correspondence to: Dr. Kristin Schirmer; e-mail: Kristin.Schirmer@

uoe.ufz.de

Contract grant sponsor: European Commission Marie Curie Training

Site Fellowship (to Ivanka Teneva); contract grant number: HPMT-CT-

2000-00115.

Published online in Wiley InterScience (www.interscience.wiley.com).

DOI 10.1002/tox.10096

2003 Wiley Periodicals, Inc.

9

7/26/2019 Lyngby A

2/12

well as freshwater environments throughout the world (Van

Dolah, 2000). Acute exposure to cyanotoxins has been

shown to result in intoxication, sometimes leading to death

(Carmichael, 1994, 1997; Jochimsen et al., 1998; Falconer,

1999). Recent studies of the presence of cyanobacterial

toxins in drinking water in certain areas of China and the

possible link of these toxins to an increased rate of humanliver cancer led to the revelation that the risk from cya-

nobacterial toxins is not restricted to acute exposure but

may extend to chronic health effects (Ueno et al., 1996;

Carmichael, 1997; Daranas et al., 2001). Despite a growing

general awareness of the presence of cyanobacterial toxins,

knowledge about the ability of specific species to producecompounds with biologically adverse effects is still rather

limited.

Thefilamentous mat-forming cyanobacteria of the genusLyngbya are a potentially rich source of cyanotoxins. The

marine species Lyngbya majuscula Gomont has been the

center of numerous investigations on the discovery of newpharmaceuticals. For example, Koehn et al. (1992) isolated

the immunosuppressive peptides microcolins A and B. Re-

cently, Luesch et al. (2000) discovered structurally similar

apramides. Other intriguing secondary metabolites of

Lyngbya majusculaare the cytotoxins lyngbyabellin A and

B (Luesch et al., 2000; Milligan et al., 2000). In addition,

lyngbyatoxin and aplysiatoxin were identified as the sourceof the dermatitis called swimmers itch, which resultsfrom skin contact with the cyanobacterium (Mynderse et al.,

1977; Cardellina et al., 1979). Although much less is known

about the toxins produced by the freshwater species Lyng-

bya wollei Farlow ex Gomont, one group of toxins recog-

nized are saxitoxin-related neurotoxins, the compounds re-

sponsible for paralytic shellfish poisonings (Carmichael etal., 1997; Yin et al., 1997). No other toxigenic species of the

genusLyngbyahas so far been described, despite nearly 100

species having been classified in the traditional way asbelonging to this genus (Geitler, 1932).

The toxicity of cyanobacterial extracts traditionally has

been investigated by exposing mice and monitoring their

survival, behavior, and histopathology. Aune and Berg

(1986) were among thefirst to suggest the use of isolated rathepatocytes to be able to more quickly and inexpensively

investigate cyanobacterial blooms for their toxicity. Re-

cently, this assay was shown to correlate particularly wellwith the content of microcystin, a common cyanobacterial

hepatotoxin (Heinze et al., 2001). However, inasmuch as the

hepatocyte assay still requires the sacrifice of animals andthe toxicity of many environmental samples cannot be ex-

plained by microcystin alone (Oberemm et al., 2001),

screening cyanobacterial extracts for their impact on cells

might be more easily done using established cell lines. The

use of such cell lines has become particularly attractive in

recent years because of the development of numerous col-

ored and/or fluorescent cell-viability indicator dyes that areapplicable to multiwell plate readers and because of the

increased availability of cultures originating from nonmam-

malian animals, such as fish, which allow cyanobacterialextracts to be evaluated with respect to environmental

health.

It was the overall goal of the current work to investigate

whether the freshwater speciesLyngbya aerugineo-coerulea

(Kutz.) Gomont, a cyanobacterium frequently found inSouthern Europe (Vodenicharov, 2000), is capable of elic-

iting toxicity. To study this, the traditional in vivo mouse

bioassay as well as in vitro assays using mammalian and a

fish liver cell line were applied to extra- as well as intra-cellular samples obtained from L. aerugineo-coerulea

grown in culture. The results suggest that Lyngbya aeru-

gineo-coerulea is an important cyanobacterium to be con-

sidered as a risk to mammals and fish in the freshwaterenvironment.

MATERIALS AND METHODS

Culture of Lyngbya aerugineo-coerulea andPreparation of Extracts

Cultivation

The cyanobacterium Lyngbya aerugineo-coerulea (Kutz.)

Gomont, PACC (Plovdiv Algal Culture Collection) No.

8601, was grown intensively under sterile conditions, as

described by Dilov et al. (1972.) A Z-nutrient medium was

used for the culture (Staub, 1961). Cultures were synchro-

nized by alternating light and dark periods of 16:8 h. The

temperature was 33C and 22C during the light and darkperiods, respectively. This culture regime was established in

order to closely mimic the conditions for optimal growth of

Lyngbya in its natural habitat in the summer months. The

intensity of light during the light period was 12 000 lux. The

culture medium was aerated with 100 L of air per hour per

liter of medium, adding 1% CO2

during the light cycle. The

cultivation period lasted 14 days.

Extraction

Extracts of the cyanobacterium were obtained using a

slightly modified version of the method of Krishnamurthy etal. (1986). Briefly, cyanobacteria were removed from the Zmedium and weighed, then frozen and thawed, and ex-

tracted twice (3 h and overnight) with a watermethanolbutanol solution (15:4:1 v/v/v, analytical grade) at 22C

while stirring. The extracts were centrifuged at 10 000 rpm

(9500g) for 30 min. The supernatants of the two extracts

were pooled and organic solvents removed via speed-vac

centrifugation (SAVANT Instruments Inc., Farmingdale,

NY, USA) at 37C for 2 h. The resulting extract wassterilized by filtration through a 0.22-m Millipore filter(GS filter typemodified acrylic polypropylene) and pre-

10 TENEVA ET AL.

7/26/2019 Lyngby A

3/12

pared to give equivalentfinal concentrations of 150 mg/mL(wet weight/volume) of suspended cyanobacterial matter.

To investigate whether Lyngbya aerugineo-coerulea re-

leased toxic products into their culture environment, the

nutrient solution in which the cyanobacteria were cultivated

for 14 days was filtered through a 0.22-m Milliporefilter.

The final equivalent concentration of suspended cyanobac-terial matter per milliliter of culture medium was 20 mg/mL

(wet weight/volume). This cyanobacterial medium was

tested for cytotoxicity in the fish cell bioassays.

Toxicity of Lyngbya Extract in vivo

Mouse Bioassay

Six male DBA/1 mice (19 22 g) were used for the exper-iment (three mice per group). All mice were kept in a

climate-controlled environment with 12-h light:dark cycles

in polystyrene cages containing wood shavings. Mice werefed standard rodent chow and water ad libitum in a specific

pathogen-free environment (as defined in http://net.inflam.lu.se). Mice were injected ip with 0.5 mL of the test solution

containing equivalent final concentrations per mouse of 15mg of suspended cyanobacterial matter (682790 mg/kgmouse). To obtain this test solution, the algal extract was

diluted 1:4 with phosphate-buffered saline (PBS). Control

mice were injected with 0.5 mL of PBS. The animals were

observed for 24 h after treatment. Behavioral symptoms,

weight, and survival times were recorded.

Liver HistologyAll animals were subjected to histological examination of

the liver for pathology. After termination of the experiment,

the liver slices were processed for light microscopy accord-

ing to standard procedures. Briefly, the tissue samples werefixed in 4% buffered formalin for 24 h, dehydrated in agraded series of alcohols, cleared in xylene, and embedded

in paraffin wax. Multiple sections from each block wereprepared 5 m thick and stained with hematoxylin and

eosin (McManus and Mowry, 1965).

Toxicity of Lyngbya Extracts in vitroAnimal Cell Culture

Six cell cultures were used: one freshly established cell

culture of mouse thymus fibroblasts; four commercially

available mammalian cell lines, FL (normal amniotic cells,

human, ATCC CCL 62), A2058 (human metastatic mela-

noma, ECACC 91100402), RD (rhabdomyosarcoma, em-

bryonic, human, ATCC CCL 136), and 3T3 (fibroblasts,embryonic, mouse, ATCC CCL 92); and one cell line de-

rived from a normal rainbow trout (Oncorhynchus mykiss)

liver (RTL-W1; Lee et al., 1993).

Continuous Mammalian Cell Lines. FL, A2058, RD,

and 3T3 cells were cultured in 75-cm2 flasks in Dulbeccos

Modified Eagles Medium (DMEM; Gibco, Paisley,

Scotland, UK), supplemented with 10% (v/v) heat-inacti-

vated fetal calf serum (FCS; PAA Laboratories GmbH,

Linz, Austria), 100 U/mL of penicillin, and 100 g/mL of

streptomycin (Sigma, Steinheim, Germany), at 37C with5% CO

2 in air and high humidity. Trypsin treatment and

subculturing were done according to the Invittox protocols

(The Ergatt/Frame, 1990, 1992). Cell viability was mea-

sured with the trypan blue exclusion test (Berg et al., 1972)

prior to seeding.

Primary Mouse Thymus Fibroblast Cell Culture. This

cell culture was prepared from three B10.Q mice. The

thymus of each mouse was removed and stored immediately

in sterile PBS containing 100 U/mL penicillin, 100 g/mL

streptomycin, and 2.5 g/mL amphotericin B (Serva, Hei-

delberg, Germany). All samples were finely minced withscissors, digested for 45 min at 37C in 5 mL of PBS

containing 0.1% trypsin (Gibco BRL, Paisley, Scotland,

UK), and then digested in 0.1% collagenase D (Roche

Diagnostics GmbH, Mannheim, Germany) in DMEM con-

taining 10% fetal calf serum (FCS, heat inactivated) for 90

min at 37C in an atmosphere with 5% CO2

. After passage

of the cells through a 40-m cell strainer (FALCON,

Becton Dickinson, Le Pont De Claix, France), cells were

collected by centrifugation, washed twice with serum-

free DMEM, and cultured in complete DMEM, as de-

scribed above for the continuous mammalian cell lines.

After three passages the culture consisted of fibroblast-like cells only, as judged by cell morphology and recog-

nition by the cells of the (surface) antiglycoproteins

VCAM-1, CD68, and CD11b, determined using a fluo-

rescence-activated cell sorter (FACS; Becton Dickinson,

San Jose, CA, USA).

Continuous Fish Cell Line. RTL-W1 cells were cultured

as originally described by Lee et al. (1993) in an atmosphere

of air in 75cm2 Nunc culture flasks at 19C in Leibovitzs

L-15 medium without phenol red (Invitrogen, Karlsruhe,

Germany), supplemented with 5% fetal bovine serum, FBS

(Biochrom, Berlin, Germany), and penicillinstreptomycin

(20 U/mL and 20 g/mL, respectively; Biochrom, Berlin,

Germany), using previously described subcultivation proce-

dures (Bols and Lee, 1994; Schirmer et al., 1994).

Exposure Conditions

Prior to exposure, cells were plated in 96-well tissue culture

plates at a density of 1.5 104 per 200 L of DMEM

medium with 10% FCS for mammalian cells and 5 104

cells per 200 L of L-15 medium with 5% FBS for the

piscine cell line. After 24 h of attachment, the medium was

TOXICITY OF L. AERUGINEO-COERULEA 11

7/26/2019 Lyngby A

4/12

removed and replaced by the exposure medium as described

below.

Exposure to the Cyanobacterial Extract. Mammalian

cells were exposed to one concentration of the cyanobacte-

rial extract. This concentration was equivalent to 15 mg/mL

(w/v) suspended cyanobacterial matter, the same concentra-tion used in the in vivo experiments in mice, and was

obtained by adding 10 L of the extract to 190 L of

DMEM culture medium with 10% FBS. Control wells were

prepared by adding 10 L of Millipore water to 190 L of

culture medium. The cells were exposed to the cell extract

for 4 or 24 h prior to analysis of cytotoxicity by the MTT

assay or of effects on proliferation by the [3H]-thymidine

incorporation assay as described below.

The piscine cell line was exposed to varying concentra-

tions of the cyanobacterial extract, with the highest concen-

tration equivalent to an extract of 15 mg/mL of suspended

cyanobacterial matter. Exposure was done in L-15 mediumin the absence of serum or, in some cases, in the presence of

a 5% FBS supplement. In most cases the exposure temper-

ature was 19C, as for routine maintenance. However, insome experimentsfish cells were exposed at 4C or 25C inorder to investigate if varying temperatures had an effect on

the toxicity of the extract. Cytotoxicity of the cyanobacterial

extract was assessed after 24 h of exposure by a combina-

tion of the alamarBlue/CFDA-AM cytotoxicity assays, asdescribed below.

Exposure to the Cyanobacterial Medium. In addition to

the cyanobacterial extract, the fish cell line was also ex-

posed to varying concentrations of the medium in whichLyngbya had been grown for 14 days. The highest concen-

tration was 50% of the cyanobacterial medium, which was

obtained by adding 100L of the cyanobacterial Z medium

to 100 L of L-15 medium without serum. An appropriate

control was prepared by adding L-15 to the fish cells withup to 50% Z medium. The cytotoxicity of the cyanobacterial

growth medium was assessed after 24 h of exposure by a

combination of the alamarBlue/CFDA-AM cytotoxicityassays as described below.

Cytotoxicity AssaysMTT

The MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltet-

razolium bromide; Sigma, St. Louis, MO, USA] assay was

carried out in accordance with Edmondson et al. (1988).

This assay is based on the capacity of mitochondrial succi-

nyl dehydrogenase to convert the soluble yellow tetrazolium

salt into an insoluble purple-blue formazan product. After

the desired time of contact with the cyanobacterial extract (4

or 24 h), 20 L of a 0.5% (w/v) solution of MTT in PBS

was added directly to each well and incubated at 37C for

4 h in the dark. After incubation, the medium with the dye

was aspirated and plates inverted to drain unreduced MTT,

and 0.1 mL of 0.04 mol/L HCL in isopropanol was added to

each well in order to facilitate solubilization of the formazan

product. The plates were shaken, and absorbance was read

at 570 nm.

alamarBlue and CFDA-AM

Twofluorescent indicator dyesalamarBlue(BioSource,Solingen, Germany) and 5-carboxyfluorescein diacetateacetoxymethyl ester (CFDA-AM, Molecular Probes, Eu-

gene, OR, USA)were used in combination, as previouslydescribed, using L-15/ex as a simplified culture medium

(Schirmer et al., 1997). The alamarBlue dye measures,similar to MTT, the redox potential of a cell, and 5-car-

boxyfluorescein diacetate acetoxymethyl ester measures cellmembrane integrity. After exposure of the cells to the cya-

nobacterial extract or medium, the wells were emptied andfilled with 100 L of a mixuture of 5% (v/v) alamarBlueand 4 M CFDA-AM in L-15/ex and incubated in the dark

for 30 min prior to fluorescent measurement. Fluorescencewas analyzed using a SPECTRAmax Gemini spectroflu-

orometer (Molecular Devices, Munich, Germany) at opti-

mized excitation/emission wavelengths for alamarBlueand CFDA-AM of 530/595 nm and 493/541 nm, respec-

tively.

Proliferation Assay

The cells were plated and exposed to a final equivalentconcentration of cyanobacterial extract of 15 mg/mL as

described above. During the last 18 h of exposure, cells

were pulsed with 1 Ci [3H]-thymidine per well (Amer-

sham Labs, Buckinghamshire, England). After the comple-

tion of 24 h of exposure, the cultures were harvested in a

Filtermate cell harvester (Packard Instrument, Meriden,

CT, USA). Incorporation of [3H]-thymidine was measured

in a Matrix 96 Direct beta counter (Packard). The mean cpm

values of triplicates were determined.

Analysis of Toxins Based on ELISA

Saxitoxins

The samples were analyzed with a Ridascreen saxitoxin

ELISA kit (r-biopharm, Darmstadt, Germany). This is a

competitive enzyme-linked immunosorbent assay (ELISA)

for the quantitative analysis of saxitoxin and related toxins

based on the competition between the free toxins from

samples or standards and an enzyme-conjugated saxitoxin

for the same antibody. The mean lower detection limit of the

Ridascreen saxitoxin assay is about 0.010 ppb.

12 TENEVA ET AL.

7/26/2019 Lyngby A

5/12

Microcystins

Analysis of samples was performed using a Microcystin

Plate kit (EnviroLogix Inc., Portland, OR, USA). As for the

saxitoxin ELISA, this is a quantitative competitive immu-

nosorbent assay. The limit of detection of the EnviroLogix

Microcystin Plate kit is 0.05 ppb.

Statistical Analysis

The results are reported as means SDs from individual

determinations. Statistical differences were analyzed with a

t test using the Sigma Plot (Jandel Scientific) program.Values ofp 0.05 were regarded as significant.

RESULTS

Toxicity of Lyngbya Extract in vivo

Injection of Lyngbya aerugineo-coerulea extract into

DBA/1 mice caused symptoms typical of neurotoxins. Im-

mediately after injection of the Lyngbya extract, mice

showed reduced activity and had convulsions and spasms as

well as respiratory difficulties. These symptoms lasted for

about 2 h. When survival was recorded 24 h after injection,

no lethality was observed. However, mice treated with the

Lyngbya extract showed an average weight loss of 1.5

0.3 g (n 3), whereas control mice gained an average of

0.18 0.16 g (n 3).

Histological examination of liver slices of treated mice

revealed signs of hepatotoxicity. Most notable was a granu-lovacuolar degeneration with perinuclear clumping of cyto-

plasm and an increased number of cells undergoing mitosis

(Fig. 1). Other signs of hepatotoxicity were minor inflam-mations and sinusoid congestions.

Toxicity of Lyngbya Extract in vitro

Toxicity to Mammalian Cells

Exposure of the extract led to distinct responses depending

on the cell line and the exposure period. Both stimulatory

and cytotoxic effects were observed after 4 h of exposure,

whereas after 24 h cytotoxicity was revealed in all cells(Fig. 2). The most pronounced response was seen in FL

cells. After 4 h of exposure, cells showed a stimulation of

MTT reduction above control levels of more than 40%,

whereas after 24 h of exposure, a decrease in cell viability

to 32% of control levels was observed. In contrast, 3T3 cells

seemed least sensitive, with 60% cell viability remaining

after 24 h of exposure. It was found that after a 24 h

exposure, most cells began to round up and detach from the

plates, floating into the medium.The apparent cytotoxicity elicited by theLyngbyaextract

was confirmed by a significant inhibition of proliferation in

the primary mouse thymus fibroblast cell culture and in the3T3 cell line. Incorporation of [3H]-thymidine was inhibited

by 80% and 92% in the thymus- and 3T3-cell cultures,

respectively (Fig. 3).

Toxicity to the Fish Cell Line, RTL-W1

In contrast to the mammalian cell cultures,fish cells respondedmuch less clearly to exposure to the Lyngbya extract at 19Cunder routinefish-cell-culture conditions. In both the presenceand absence of serum, cell viability was little affected by

exposure to the cyanobacterial extract for 24 h, as measured

with the alamarBlue/CFDA-AM assay (Fig. 4). In somecases a slight increase in fluorescent unit readings was ob-

Fig. 1. Appearance of liver cells obtained from mice treatedin vivo with (A) the appropriate vehicle control or (B) theLyngbya aerugineo-coerulea extract. Mice were injected ipwith an equivalentfinal concentration of 15 mg suspendedcyanobacterial matter for 24 h prior to preparation of the liverslices. Arrows indicate typical signs of hepatotoxicitycaused by the extract. These signs are (a) an increasednumber of mitotic cells, (b) minor inflammations and sinu-soid congestions, and (c) a granulovacuolar degenerationwith perinuclear clumping of cytoplasm. This is hematoxylin-eosin staining. The original magnification was 500.


7/26/2019 Lyngby A

6/12

served, but this was generally no more than 25% and was not

dose dependent. Likewise,fluorescent unit readings occasion-ally dropped below 100%, but, again, these readings either

were not significantly different from the control or did notoccur in a dose-dependent manner. Comparable results were

obtained when cells were exposed to the extract for 4 h at 19C(data not shown). However, when cells were exposed to the

extract for 48 h, fluorescence unit readings for both alamar-

Blueand CFDA-AM increased significantly. For example,at an equivalent concentration of 15 mg/mL suspended cyanobac-

terial matter, which was also the concentration applied to mam-

malian cells (Fig. 2), alamarBlue readings increased to 127%

11% and 136% 12% (average of three culture wells) in me-dium with and without serum, respectively. For CFDA-AM, these

readings were 120% 24% in the medium with serum and 178%

9% for the medium without serum. The dose dependence of

thisincrease in fluorescence unit readings was seen most clearly in

the medium without serum (Fig. 5).

Inasmuch as temperature could be a determinant of the

cytotoxic effects of the Lyngbya extract, RTL-W1 cells

were exposed for 24 h at 4C and 25C. At the lowertemperature cell viability was not affected (data not shown)

and was comparable to the cell viability data obtained at

19C (Fig. 4). However, exposure at 25C led to a signifi-

cant loss of cell viability (Fig. 6). Cell response differed

depending on the presence or absence of serum. Without

serum in the culture medium, fluorescence unit readings

dropped to 20%30% of control levels, with no striking

differences in the readings for the different extract concen-

trations [Fig. 6(B)]. In contrast, a dose-dependent decrease

of cell viability was observed in the presence of serum [Fig.

Fig. 3. [3H]-thymidine incorporation into 3T3 cells (top panel)and mouse thymusfibroblasts (bottom panel) on exposure tocell culture medium containing 10%L. aerugineo-coeruleaex-tract or 10% tissue culture water as a control. Exposure wasdone for 24 h at 37C in the presence of 10% serum. Oneexperiment is shown, in which each bar represents the averageof three culture wells, with vertical lines indicating the standarddeviation. Asterisks denote a significant response from thecyanobacterial extract compared with the extract-free control,as determined by attest ( 0.05).

Fig. 2. Viability of four mammalian cell lines on exposure toa cell culture medium containing 10% L. aerugineo-coeruleaextract for 4 h (black bars) and 24 h (white bars) at 37C in

the presence of 10% serum. Cell viability was measuredwith the MTT assay and absorbance readings expressed asa percentage of the control, which received no extract but anappropriate amount of tissue culture water. A volume of10% Lyngbya extract is equivalent to a suspended cya-nobacterial matter concentration of 15 mg/mL. One experi-ment is shown, for which each bar represents the average ofthree culture wells, with vertical lines indicating the standarddeviation. Asterisks denote a significant response from thecyanobacterial extract compared with the extract-free con-trol, as determined by a ttest ( 0.05).

14 TENEVA ET AL.

7/26/2019 Lyngby A

7/12

6(A)]. At the highest concentration applied in this study, the

average cell viability of three culture wells was 42% 4%

for alamarBlue and 35% 7% for CFDA-AM.

Toxicity of the Lyngbya aerugineo-coeruleaGrowth Medium to RTL-W1 Cells

The medium in whichLyngbyahad been grown for 14 days

prior to extraction was applied to the fish liver cells inconcentrations of up to 50%. A similar concentration of Z

medium in L-15 had no detrimental effect on cell viability

compared with 100% L-15 culture medium (data not

shown). In contrast to the cyanobacterial extract, significant

cytotoxicity was observed with the cyanobacterial mediumat 19C, the temperature used for routine maintenance of the

fish cell line. This cytotoxicity was far greater in the absencethan in the presence of serum (Fig. 7). A change in exposure

temperature to 4C or 25C did not significantly alter theseresults (data not shown).

The distinct cytotoxic responses observed with the cya-

nobacterial medium were also confirmed by characteristicmorphological changes, which were not observed with the

cyanobacterial extract. At concentrations of 20% cyanobac-

terial medium and above, cells acquired a more fibroblast-like appearance and formed circles around empty culture

surface areas [Fig. 8 B(a)]. In some areas cells were reducedto long, thin extensions and started to detach [Fig. 8B(b)]. In

contrast to the measurements of cytotoxicity, these morpho-

logical changes were marginally more pronounced at the

higher exposure temperature, 25C.

Analysis of Toxins Based on ELISA

Although in the assay of saxitoxins slightly increased back-

ground levels were found in both the extract and the growth

medium, neither saxitoxin nor microcystin could be de-

tected above the mean lower limits of detection of the

Fig. 4. Viability of RTL-W1 cells on exposure to an extract ofL. aerugineo-coeruleafor 24 h at 19C in the (A) presence or(B) absence of serum. Cell viability was measured withalamarBlue (black bars) and CFDA-AM (white bars), andfluorescent units readings are expressed as a percentage ofthe control, which received no extract but an appropriateamount of tissue culture water. A volume of 10% Lyngbyaextract is equivalent to a suspended cyanobacterial matter

concentration of 15 mg/mL. One representative experimentis shown. Each bar represents the average of three culturewells, with vertical lines indicating the standard deviation.

Asterisks denote a significant response from the cyanobac-terial extract compared with the extract-free control, as de-termined by a ttest ( 0.05).

Fig. 5. Viability of RTL-W1 cells upon exposure to an extractofL. aerugineo-coeruleafor 48 h at 19C in the presence (A)or absence (B) of serum. Cell viability was measured withalamarBlue (black bars) and CFDA-AM (white bars) andfluorescent unit readings were expressed as described inFigure 4. One representative experiment is shown. Each barrepresents the average of three culture wells with verticallines indicating the standard deviation. Asterisks denote asignificant response from the cyanobacterial extract ascompared with the extract-free control, as determined by attest ( 0.05). In one instance, a significant difference wasfound between the reading obtained with alamarBlueandthat with CFDA-AM, ( 0.05; indicated by $).

Fig. 6. Viability of RTL-W1 cells on exposure to an extract ofL. aerugineo-coeruleafor 24 h at 25C in the (A) presence or(B) absence of serum. Cell viability was measured withalamarBlue (black bars) and CFDA-AM (white bars), andfluorescent unit readings were expressed as described inFigure 4. One representative experiment is shown. Each barrepresents the average of three culture wells, with verticallines indicating the standard deviation. Asterisks denote a

significant response from the cyanobacterial extract com-pared with the extract-free control, as determined by attest( 0.05).


7/26/2019 Lyngby A

8/12

ELISA assays. This result indicates that toxins capable of

cross-reacting in the ELISA tests were not present at sig-

nificant amounts. These toxins include decarbamoyl saxi-toxin and gonyautoxins II, III, B1, C1, and C2 for the assay

of saxitoxins. For the assay of microcystin, toxins capable

of cross-reacting but not apparently present in the Lyngbya

samples were the microcystin variants LR, LA, RR, and YR

as well as nodularin.

DISCUSSION

Several cyanobacterial species that belong to the order

Oscillatorialeshave previously been reported to be capable

of producing potent toxins. Most notable are the investiga-

tions on species of the prominent genus Oscillatoria, in-

cluding a number of hepatotoxin- as well as neurotoxin-

producing strains (Mynderse et al., 1977; stensik et al.,1981; Skulberg et al., 1992). In contrast to Oscillatoria,

only two toxigenic species have been reported for Lyngbya

so far. These are the marine species L. majuscula(Milligan

et al., 2000a, 2000b; Luesch et al., 2000a, 2000b, 2001a,

2001b; Nogle et al., 2001; Osborne et al., 2001) and the

freshwater species L. wollei (Carmichael et al., 1997; On-

odera et al., 1997; Yin et al., 1997). The results of our study

have revealed for the first time the ability of another fresh-water species of Lyngbya, namely, L. aerugineo-coerulea,

to be toxigenic. Both an intracellular extract and an extra-

cellular sample ofL. aerugineo-coeruleaelicited significant

adverse effects when applied to the mammalian andfish-cellbioassays used in this report.

Injection of L. aerugineo-coerulea extracts into mice

yielded signs of neuro- as well as hepatotoxicity, but these

signs could not be attributed to either the neurotoxin saxi-

toxin or the hepatotoxin microcystin and their analogues

cross-reacting in the ELISA assays. The lack of micro-cystins and nodularin is in line with other reports on toxins

produced by theLyngbyaspecies, for which microcystins or

nodularin have not yet been noted. Analogues of saxitoxin,

which have previously been identified in L. wollei (Car-michael et al., 1997; Onodera et al., 1997; Yin et al., 1997)

andL. majuscula(Fujiki et al., 1981; Nakayasu et al., 1981),

are decarbamoyl gonyautoxin II and III (paralytic shellfishpoison toxins). Although their presence in the extract pre-

pared from L. aerugineo-coerulea cannot entirely be ruled

Fig. 8. Phase-contrast appearance of RTL-W1 cells on ex-posure to cell culture medium containing 30% of the cya-nobacterial medium in which Lyngbya had been grown for14 days (B) compared to the appropriate control (A). Arrowsindicate the formation (a) of circles around empty culturesurface areas because of the cyanobacterial medium and (B)of long, thin extensions originating from the normally epithe-lial-like cells. The photographs were taken at 100.

Fig. 7. Viability of RTL-W1 cells on exposure to the mediumin whichL. aerugineo-coeruleahad been grown for 14 days.Cells were exposed to the medium for 24 h at 19 C in the (A)presence or (B) absence of serum. Cell viability was mea-sured with alamarBlue (black bars) and CFDA-AM (whitebars), and fluorescent unit readings are expressed as apercentage of the control, which received an appropriateamount of tissue culture water with Z medium. A volume of

50% cyanobacterial medium is equivalent to the volume inwhich 10 mg of suspended cyanobacterial matter has beengrown. One representative experiment is shown. Each barrepresents the average of three culture wells, with verticallines indicating the standard deviation. Asterisks denote asignificant response from the growth medium comparedwith the control, as determined by a ttest ( 0.05).

16 TENEVA ET AL.

7/26/2019 Lyngby A

9/12

out, they were below the level of detection in the saxitoxin

ELISA and thus unlikely to be the cause of the observed

neurological effects. Rather, recovery from signs of neuro-

toxicity after 2 h of injection into mice, as observed in our

study, has recently been described for another toxin identi-

fied in L. majuscula: lyngbyatoxin A (Ito et al., 2002). In

addition to causing transient signs of neuropathology, lyng-byatoxin A has been shown by the same authors to cause

increased mitosis and small granulomas (inflammatorycells) in livers of mice upon oral administration (Ito et al.,

2002). Similar signs of hepatotoxicity were observed in the

current study. Thus, lyngbyatoxin A or a structurally similar

compound is a potential candidate for eliciting at least some

of the neuro- as well as hepatotoxicity by L. aerugineo-

coerulea.

The ability of the L. aerugineo-coeruleaextract to cause

adverse effects in vivo was confirmed in vitro using mam-malian cell lines. The extract was a strong inhibitor of

[

3

H]-thymidine incorporation, indicating a significant re-duction in DNA synthesis. Inhibition of DNA synthesiscould have been the cause of the increase in mitosis found

in the livers of treated mice. A similar correlation between

the rate of DNA synthesis and mitosis has previously been

observed, for a cyanobacterial indolinone, welwistatin,

which caused a depletion of microtubules in human ovarian

carcinoma and vascular smooth-muscle cells (Zhang and

Smith, 1996). Yet, from the current in vitro data, a specificmode of toxicity from the L. aerugineo-coerulea extract

could not be deduced. One reason for this is the cytotoxicity

that developed in a time-dependent manner to varying de-

grees in the different mammalian cell lines. This cytotoxic-

ity, as measured with the MTT assay, could be both a cause

or a result of the inhibition of DNA synthesis. To elucidate

the mechanisms by which the L. aerugineo-coeruleaextract

can inhibit DNA synthesis and cause cytotoxicity will be an

important future task.

One intriguing aspect of exposing a fish liver cellculture along with those of mammalian cell lines was that

it revealed unique reactions of the two animal cell sys-

tems to the cyanobacterial extract. The rainbow trout

liver cell line, RTL-W1, was derived from a normal male

liver (Lee et al., 1993) and is the first fish cell line to beinvestigated for toxicity caused by cyanobacteria. Nor-

mally,fish cells are grown at 19C, which is close to theorganisms temperature. At this temperature no cytotox-icity was observed within 24 h of exposure to the L.

aerugineo-coerulea extract. However, an increase in ex-

posure temperature to 25C, generally the highest tem-

perature at which salmonid cells can be maintained (Bols

et al., 1992), led to severe toxicity by the algal extract,

particularly when serum was absent. A greater sensitivity

offish cells to toxic substances in the absence of serumhas been observed repeatedly in the past (Kocan et al.,

1983; Lyons-Alcantara et al., 1996; Schirmer et al.,1997,

2000). In these previous studies the protective effect of

serum was attributed to an affinity of the toxic substancesto serum components or to the provision by the serum of

protective molecules, such as antioxidants (Schirmer et

al., 2000). In light of the unknown identity of the cyto-

toxic compounds produced by L. aerugineo-coerulea, the

exact mechanism of protection by serum cannot be deci-

phered at this point. However, knowledge about the in-creased sensitivity of thefish cells and their ability to besustained in a serum-free medium should give rise to an

experimental system in which protective agents can be

studied without interference by serum components.

As it did for mammalian cells, time appeared to be one

determinant of whether the Lyngbya extract yielded mea-

surable effects in the fish cells. This was seen whenRTL-W1 cells were exposed at 19C for 48 h instead of 4or 24 h. Surprisingly, this exposure regime resulted in a

dose-dependent increase in both alamarBlue andCFDA-AM readings. Inasmuch as this phenomenon oc-

curred both in the presence and in the absence of serum, itis unlikely to be a result of cell proliferation because fishcells show very little growth without serum and within 48 h

(Lee et al., 1993). Thus, this increase could be from a

modulation of cellular membrane function. In support of

this hypothesis, pendolmycin, an indole alkaloid which is

structurally similar to lyngbyatoxin A, has previously been

shown to modify membrane function by inhibiting phospha-

tidylinositol turnover in A431 cells and by activating ara-

chidonic acid release and hexose transport in C3H10T1/2

cells (Umezawa et al., 1989). The alamarBlueis acted onby oxidoreductases (Goegan et al., 1995; OBrien et al.,

2000) and the CFDA-AM by cellular esterases, both of

which can be membrane-bound, and alteration of membrane

composition or function potentially could increase the avail-

ability of these enzymes for the fluorescent dyes. A similarmechanism might have led to the increase in the conversion

of MTT in the FL cell line after 4 h of exposure to the

cyanobacterial extract. Exposure to the extract of the same

cell line led to a significant reduction in cell viability 20 hlater. Thus, the stimulation of MTT, alamarBlue, and

CFDA-AM conversion may reflect subtle changes to cellu-lar membranes that lead after prolonged exposure to cyto-

toxicity.

In contrast to the L. aerugineo-coerulea extract, the

medium in whichLyngbyahad been grown for 14 days priorto extraction showed significant cytotoxic effects inRTL-W1 cells at 19C within 24 h of exposure. Severallines of evidence suggest that different cytotoxic com-

pounds with different modes of action must have been

involved in the effects elicited by the medium and the

extract, respectively. First, susceptibility of the fish cells tothe cytotoxicity of the L. aerugineo-coerulea extract was

dependent on temperature, whereas sensitivity to the cya-

nobacterial medium was not. Second, the L. aerugineo-

coerulea medium led to morphological changes in the fishliver cells that were not observed with the cyanobacterial


7/26/2019 Lyngby A

10/12

extract. The morphological changes elicited by the cya-

nobacterial medium suggest alterations to the cytoskeleton.

One toxin of the marine speciesLyngbya majusculathat has

been described as having disrupted a cellular microfilamentnetwork is Lyngbyabellin A (Luesch et al., 2000). This

toxin caused the normallyfibroblastic A-10 cells to exhibit

a neuronlike morphology (Luesch et al., 2000). A similarmorphology was seen in RTL-W1 cells, which normally are

epithelial-like, on exposure to the L. aerugineo-coerulea

medium.

Despite the cyanobacterial extracts investigated in this

study having originated from L. aerugineo-coerulea grown

in culture, several environmentally relevant conclusions can

be drawn about the Lyngbya species in specific and cya-nobacterial toxins in general. First,L. aerugineo-coeruleais

the second freshwater species of Lyngbya shown to be

capable of producing toxic substances. The culture condi-

tions used were adopted to reflect temperatures and light

exposures typical of the summer months in southern Eu-rope, where L. aerugineo-coerulea can be found and where

it is involved in the formation of cyanobacterial blooms

(Vodenicharov, 2000). Thus, the data obtained from cul-

tured strains can serve as a guide for the toxic potential to be

expected in nature. Second, L. aerugineo-coerulea appears

to release substantial amounts of cytotoxic compounds into

the surrounding medium. This release is unlikely to be a

result of cell lysis alone because different toxic effects were

seen in the cells and the medium. Extracellular concentra-

tions of cyanotoxins in water have only recently been in-

cluded in some surveys and have been found to be rather

low (Chorus, 2001). However, given the toxic potency of

the cyanobacterial medium in the current study and consid-

ering the general lack of knowledge about toxin release,

fate, and transport, the possibility of human and wildlife

exposure to extracellular toxic cyanobacterial metabolites in

water should not be ignored. Finally, as previously observed

with other cyanobacterial samples, the toxicity elicited byL.

aerugineo-coerulea could not be explained by the promi-

nent cyanobacterial toxin microcystin or by saxitoxin. This

highlights the importance of an overall toxicological anal-

ysis in the environmental assessment of risk of exposure to

cyanobacteria. Applying a combination of mammalian and

piscine in vitro bioassays is a promising approach that,

combined with an effect-directed chemical analysis, couldbe used in the future to identify the structure and cellular

mechanisms of the as-yet-unknown toxic Lyngbya aeru-

gineo-coeruleametabolites, in particular, and to screen cya-

nobacterial extracts for their toxicity, in general.

REFERENCES

Aune T, Berg K. 1986. Use of freshly prepared rat hepatocytes to

study toxicity of blooms of the blue-green algae Microcystis

aeruginosaand Oscillatoria agardhii. J Toxicol Environ Health

19:325336.

Berg T, Boman D, Seglen PO. 1972. Induction of tryptophan

oxygenase in primary rat liver cell suspensions by glucocorti-

coid hormone. Exp Cell Res 72:571574.

Bols NC, Lee LEJ. 1994. Cell lines: availability, propagation and

isolation. In: Hochachka, PW, Mommsen TP, editors. Biochem-istry and molecular biology of fishes, Volume 3. Amsterdam,

The Netherlands: Elsevier Science. p 145159.

Bols NC, Mosser DD, Steels GB. 1992. Temperature studies and

recent advances withfish cellsin vitro. Comp Biochem Physiol

103A:114.

Cardellina JH, Marner FJ, Moore RE. 1979. Seaweed dermatitis:

structure of lyngbyatoxin A. Science 204:193195.

Carmichael WW, Evans WR, Yin QQ, Bell P, Moczydlowski E.

1997. Evidence for paralytic shellfish poisons in the freshwater

cyanobacterium Lyngbya wollei (Farlow ex Gomont) comb.

nov. Appl Environ Microbiol 63(8):3104 3110.

Carmichael WW. 1994. The toxins of cyanobacteria. Sci Am

270(1):78 86.

Carmichael WW. 1997. The Cyanotoxins. Adv Botanical Res

27:211256.

Chorus I, editor. 2001. Cyanotoxin occurrence in freshwaters. In:

Cyanotoxins occurrence, causes, consequences. Springer,

Berlin, Heidelberg. p 5101.

Daranas AH, Norte M, Fernandez JJ. 2001. Toxic marine microal-

gae. Toxicon 39: 11011132.

Dilov Chr, Bozkova M, Staev St. 1972. Prakticeskij blok dlja

kultivirovanija mikrovodoroslej v laboratornych uslovijach [A

practical block for cultivation of microalgae in laboratory con-

ditions]. Dokl Akad selskoch Nauk Bolg 4:169 175.

Edmondson JM, Armstrong LS, Martinez AO. 1988. A rapid andsimple MTT-based spectrophotometric assay for determining

drug sensitivity in monolayer cultures. J Tissue Cult Meth

11:1517.

Falconer IR. 1999. An overview of problems caused by toxic

blue-green algae (cyanobacteria) in drinking and recreational

water. Environ Toxicol 14:512.

Fujiki H, Mori M, Nakayasu M, Terada M, Sugimura T, Moore

RE. 1981. Indole alkaloids: dihydroteleocidin B, teleocidin, and

lyngbyatoxin A as members of a new class of tumor promoters.

Proc Natl Acad Sci USA 78(6):38723876.

Goegan P, Johonson G, Vincent R. 1995. Effects of serum protein

and colloid on the alamar blue assay in cell culture. Toxicol In

Vitro 9:257266.Geitler L. 1932. Cyanophycea. In: Rabenhorst L, editor. Krypto-

gamen Flora. Leipzig, Germany: Akademische Verlagsgesell-

schaft.

Heinze R, Fastner J, Neumann U, Chorus I. 2001. In: Chorus I,

editor. Testing cyanobacterial toxicity with primary rat hepato-

cyte and cell-line assays. Cyanotoxins occurrence, causes,

consequences. Springer, Berlin, Heidelberg. p 317324.

Invittox: The Ergatt/Frame. 1990. The frame cytotoxicity test.

Protocol 3b.

Invittox: The Ergatt/Frame. 1992. DataBank ofin vitro techniques

in toxicology. Protocol 46.

18 TENEVA ET AL.

7/26/2019 Lyngby A

11/12

Ito E, Satake M, Yasumoto T. 2002. Pathological effects of lyng-

byatoxin A upon mice. Toxicon 40:551556.

Jochimsen EM, Carmichael WW, An JS, Cardo DM, Cookson ST,

Holmes CE, Antunes MB, de Melo Filho DA, Lyra TM, Barreto

VS, Azevedo SM, Jarvis WR. 1998. Liver failure and death after

exposure to microcystins at a hemodialysis center in Brazil.

N Engl J Med 338:873 878.Kocan RM, Chi EY, Eriksen N, Benditt EP, Landolt ML. 1983.

Sequestration and release of polycyclic aromatic hydrocarbons

by vertebrate cells in vitro. Environ Mutagen 5:643 656.

Koehn FE, Longley RE, Reed JK. 1992. Microcolins A and B, new

immunosuppressive peptides from the blue-green algaLyngbya

majuscula. J Nat Prod 55:613 619.

Krishnamurthy T, Carmichael WW, Sarver EW. 1986. Toxic pep-

tides from freshwater cyanobacteria (blue-green algae). Isola-

tion, purification and characterization of peptides from Micro-

cystis aeruginosa and Anabaena flos-aquae. Toxicon 24:865

873.

Lee LEJ, Clemons JH, Bechtel DG, Caldwell SJ, Han KB, Pa-

sitschniak-Arts M, Mosser DD, Bols NC. 1993. Development

and characterization of a rainbow trout liver cell line expressing

cytochrome P450-dependent monooxygenase activity. Cell Biol

Toxicol 9:279294.

Luesch H, Pangilinan R, Yoshida WY, Moore RE, Paul VJ. 2001a.

Pitipeptolides A and B, new cyclodepsipeptides from the marine

cyanobacterium Lyngbya majuscula. J Nat Prod 64:304 307.

Luesch H, Yoshida WY, Moore RE, Paul VJ, Corbett TH. 2001b.

Total structure determination of apratoxin A, a potent novel

cytotoxin from the marine cyanobacterium Lyngbya majuscula.

J Am Chem Soc 123:5418 5423.

Luesch H, Yoshida WY, Moore RE, Paul VJ, Mooberry SL.

2000a. Isolation, structure determination, and biological activityof Lyngbyabellin A from the marine cyanobacterium Lyngbya

majuscula. J Nat Prod 63:611 615.

Luesch H, Yoshida WY, Moore RE, Paul VJ. 2000b. Isolation and

structure of the cytotoxin lyngbyabellin B and absolute config-

uration of lyngbyapeptin A from the marine cyanobacterium

Lyngbya majuscula. J Nat Prod 63:14371439.

Lyons-Alcantara M, Tarazona JV, Mothershill C. 1996. The dif-

ferential effects of cadmium exposure on the growth and sur-

vival of primary and established cells from fish and mammals.

Cell Biol Toxicol 12:29 38.

McManus JFA, Mowry RW. 1965. General methods for study of

the cell and its structure. In: McManus JFA, Mowry, RW,

editors. Staining methods: histologic and histochemical. Harper,NY. p 7390.

Milligan KE, Marquez B, Williamson RT, Davies-Coleman M,

Gerwick WH. 2000a. Two new malyngamides from a Mada-

gascan Lyngbya majuscula. J Nat Prod 63:65968.

Milligan KE, Marquez BL, Williamson RT, Gerwick WH. 2000b.

Lyngbyabellin B, a toxic and antifungal secondary metabolite

from the marine cyanobacteriumLyngbya majuscula. J Nat Prod

63:1440 1443.

Mynderse JS, Moore RE, Kashiwagi M, Norton TR. 1977. Anti-

leukemia activity in the Osillatoriaceae: isolation of Debro-

moaplysiatoxin from Lyngbya. Science 196:538 540.

Nakayasu M, Fujiki H, Mori M, Sugimura T, Moore RE. 1981.

Teleocidin, lyngbyatoxin A and their hydrogenated derivatives,

possible tumor promoters, induce terminal differentiation in

HL-60 cells. Cancer Lett 12(4):271277.

Nogle LM, Okino T, Gerwick WH. 2001. Antillatoxin B, a neu-

rotoxic lipopeptide from the marine cyanobacterium Lyngbya

majuscula. J Nat Prod 64:983985.Oberemm A, Heinze R, Papendorf O, Fastner J. 2001. Significance

of unidentified toxic compounds and approaches to their iden-

tification. In: Chorus I, Editor. Cyanotoxins occurrence,

causes, consequences. Springer, Berlin, Heidelberg. p 286 295.

Onodera H, Satake M, Oshima Y, Yasimoto T, Carmichael WW.

1997. New saxitoxin analogues from the freshwaterfilamentous

cyanobacterium Lyngbya wollei. Nat Toxins 5(4):146 151.

O Brien J, Wilson I, Orton T, Pognan F. 2000. Investigation of

the Alamar Blue (resazurin) fluorescent dye for the assess-

ment of mammalian cell cytotoxicity. Eur J Biochem 267:

54215426.

Osborne NJ, Webb PM, Shaw GR. 2001. The toxins of Lyngbyamajuscula and their human and ecological health effects. Envi-

ron Int 27(5):381392.

stensvik O, Skulberg OM, Soli N. 1981. Toxicity studies with

blue-green algae from Norwegian inland waters. In: Carmichael

WW, editor. The water environment: Algal Toxins and Health.

New York: Plenum Press. p 315324.

Schirmer K, Chan AGJ, Bols NC. 2000. Transitory metabolic

disruption and cytotoxicity elicited by benzol(a)pyrene in two

cell lines from rainbow trout liver. J Biochem Mol Toxicol

14(5):262276.

Schirmer K, Chan AGJ, Greenberg BM, Dixon DG, Bols NC.

1997. Methodology for demonstrating and measuring the pho-

tocytotoxicity offluoranthene tofish cells in culture. Toxicol InVitro. 11:107119.

Schirmer K, Ganassin RC, Brubacher JL, Bols NC. 1994. A DNA

fluorometric assay for measuring fish cell proliferation in mi-

croplates with different well sizes. J Tissue Cult Meth 16:133

142.

Skulberg OM, Carmichael WW, Andersen RA, Matsunaga S,

Moore RE, Skulberg R. 1992. Investigations of a neurotoxic

oscillatorialean strain (Cyanohyceae) and its toxin. Isolation and

characterization of homoanatoxin-a. Environ Toxicol Chem 11:

321329.

Staub R. 1961. Ernahrungsphysiologisch- antokologische Unter-

suchungen an der planktonischen Blaualga Oscillatoria rubes-cens DC. Schweiz Z Hydrol 23:82198.

Ueno Y, Nagata S, Tsutsumi T, Hasegawa A, Watanabe MF, Park

HD, Chen GC, Chen G, Yu SZ. 1996. Detection of micro-

cystins, a blue-green algal hepatotoxin, in drinking water sam-

pled in Haimen and Fusui, endemic areas of primary liver cancer

in China, by highly sensitive immunoassay. Carcinogenesis

17:13171321.

Umezawa K, Imoto M, Yamashita T, Sawa T, Takeuchi T. 1989.

Modification of cellular membrane function by pendolmycin.

Jpn J Cancer Res 80(1):1518.

Van Dolah FM. 2000. Diversity of marine and freshwater algal


7/26/2019 Lyngby A

12/12

toxins. In: Botana, LM, editor. Seafood and freshwater toxins:

pharmacology, physiology, and detection. New York: Marcel

Dekker. p 19 43.

Vodenicharov D. 2000. Sistematika na rasteniata. Chast 1. Pro-

kariotni vodorasli i talusni rastenia, [Systematics of the plants.

Part 1. Prokaryotic algae and thalloid plants]. Sofia, Bulgaria:

Pensoft. p 31 40.

Zhang X, Smith CD. 1996. Microtubule effects of welwistatin, a

cyanobacterial indolinone that circumvents multiple drug resis-

tance. Mol Pharmacol 49(2):288 294.

Yin QQ, Carmichael WW, Evans WR. 1997. Factors influencing

growth and toxin production by cultures of the freshwater cya-

nobacterium Lyngbya wollei Farlow ex Gomont. J Appl Phycol

9:55 63.

20 TENEVA ET AL.

lyngby a

Documents