APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 2003, p. 7371–7376
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 69, No. 12
Effects of Saxitoxin (STX) and Veratridine on Bacterial Na?-K?
Fluxes: a Prokaryote-Based STX Bioassay
Francesco Pomati,1Carlo Rossetti,2Davide Calamari,2and Brett A. Neilan1*
School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney 2052,
New South Wales, Australia,1and Department of Structural and Functional Biology, University of Insubria, 21100 Varese, Italy2
Received 28 April 2003/Accepted 4 September 2003
Saxitoxin (STX) is a potent natural sodium channel blocker and represents a significant health concern
worldwide. We describe here the antagonistic effects of STX and veratridine (VTD), an Na?channel activator,
on three gram-negative bacteria and their application to an STX bioassay. STX reduced the total cellular levels
of both Na?and K?, as measured by flame photometry, whereas VTD increased the cellular concentrations
relative to control ion fluxes in the cyanobacterium Cylindrospermopsis raciborskii AWT205. Endogenous STX
production in toxic cyanobacterial strains of C. raciborskii and Anabaena circinalis prevented cell lysis induced
by VTD stress. Microscopic cell counts showed that non-STX producing cyanobacteria displayed complete cell
lysis and trichome fragmentation 5 to 8 h after addition of VTD and vanadate (VAN), an inhibitor of sodium
pumps. The addition of STX, or its analogue neoSTX, prior to treatment with VTD plus VAN prevented
complete lysis in non-STX-producing cyanobacteria. VTD also affected cyanobacterial metabolism, and the
presence of exogenous STX in the sample also ameliorated this decrease in metabolic activity, as measured by
the cellular conversion of tetrazolium into formazan. Reduced primary metabolism was also recorded as a
decrease in the light emissions of Vibrio fischeri exposed to VTD. Addition of STX prior to VTD resulted in a
rapid and dose-dependent response to the presence of the channel blocker, with samples exhibiting resistance
to the VTD effect. Our findings demonstrate that STX and VTD influence bacterial Na?and K?fluxes in
opposite ways, and these principles can be applied to the development of a prokaryote-based STX bioassay.
Paralytic shellfish poisoning (PSP) is a deadly affliction that
results from the accidental consumption of some potent natu-
ral neurotoxins, typically via contaminated seafood. These
molecules, of which the most potent representative is saxitoxin
(STX), are highly selective blockers of Na?channels in excit-
able cells, thereby affecting nerve impulse generation in ani-
mals (6). This syndrome can lead, in extreme cases, to death
(31). STX and analogue compounds, known collectively as the
PSP toxins, have been reported to occur in both marine and
dinoflagellates have been described to synthesize PSP toxins
(15, 24, 28), a limited number of bacterial strains (10), and
several species of freshwater filamentous cyanobacteria (1, 5,
26), including strains of Anabaena circinalis and Cylindrosper-
mopsis raciborskii (17, 20). Occurrences of harmful algal
blooms associated with PSP toxins represent a serious health
concern worldwide and have been increasingly reported (18,
25, 31). All nonanalytical detection systems for these neuro-
toxins rely on bioassays based on whole animals or animal cell
lines (14, 21), since much is known regarding the pharmaco-
logical effect of STX on eukaryotes. On the other hand, PSP
toxins have rarely been studied according to their effect on
prokaryotic cells (30). For economic, practical, and ethical
reasons, alternatives to the standard animal biological tests are
desired. Here we describe our findings, which represent the
first investigation of the feasibility of a prokaryote-based bio-
assay for STX and its analogues.
Three generaof marine
In a previous study we reported that, in the STX-producing
cyanobacterium C. raciborskii T3, STX accumulation was di-
rectly correlated to variations in intracellular Na?levels (F.
Pomati, C. Rossetti, G. Manarolla, and B. A. Neilan, unpub-
lished data). These results suggested a possible role of this
toxin in the maintenance of cyanobacterial homeostasis under
Na?stress. Questions regarding whether STX-producing cya-
nobacteria have a potential advantage over other nonneuro-
toxic strains under conditions of critical Na?levels or whether
STX interferes with bacterial Na?fluxes as it does in eukary-
otic cells arose from that study (Pomati et al., unpublished).
In the present study we first investigated the effect of STX
and veratridine (VTD), a sodium channel activator that in-
creases Na?permeability in eukaryotic cells (8), on the cya-
nobacterial total Na?and K?cellular contents measured by
flame photometry. C. raciborskii AWT205, a cyanobacterium
known not to produce any neurotoxin, including STX (16), was
exposed to STX and VTD. Further, by using VTD to induce
intracellular Na?stress, we assayed PSP toxins producing and
nontoxic strains of C. raciborskii and A. circinalis in a “cyano-
lytic” test similar to those used with eukaryotic cells (21, 29). In
the assay developed here cyanobacteria were stressed with
VTD and o-vanadate (VAN), an inhibitor of bacterial ion
pumps (ion translocating P-type ATPases) (9, 13). The detec-
tion of cell lysis was used as the endpoint. Two nonneurotoxic
strains (C. raciborskii AWT205 and A. circinalis 271C) were
then used in the same test to assay the presence of channel
blockers such as lidocaine, amiloride, STX, and neoSTX. In
order to investigate whether the demonstrated sensitivity of C.
raciborskii AWT205 to VTD and STX was also related to
changes in the cell’s metabolic activity, the test was applied to
a cell titer cytotoxicity assay. Subsequently, the method based
* Corresponding author. Mailing address: School of Biotechnology
and Biomolecular Sciences, The University of New South Wales, Syd-
ney 2052, NSW, Australia. Phone: 61-2-9385-3235. Fax: 61-2-9385-
1591. E-mail: firstname.lastname@example.org.
on VTD-induced Na?stress was used to assay the presence of
STX with the commercially available and standardized toxicity
test LUMIStox by using the bioluminescent bacterium Vibrio
MATERIALS AND METHODS
Cyanobacterial strains and culture conditions. C. raciborskii AWT205, a non-
PSP-toxin producer (16), was obtained from Peter R. Hawkins (Australia Water
Technologies, EnSight, West Ryde, New South Wales, Australia). C. raciborskii
T3, an STX producer (20), was obtained from Sandra Azevedo (Federal Uni-
versity of Rio de Janeiro, Rio de Janeiro, Brazil). Strains AWT205 and T3 were
maintained in ASM-1 medium (12). PSP toxin-producing strains of A. circinalis,
344B, 134C, 150A, and 131C and the nontoxic strains 271C, 306A, and 332H
were obtained from the Australian Water Quality Centre (Adelaide, South
Australia) and maintained in Jaworski’s medium (17). All cyanobacterial cultures
were grown in 250-ml glass flasks at a constant temperature of 26°C under
continuous irradiance of cool white light at an intensity of 15 ?mol of photons
m?2s?1. Cyanobacterial growth was monitored by recording the optical density
at 750 nm with a Lambda 10 UV/VS spectrometer (Perkin-Elmer, Inc., Shalton,
Conn.). Mid-exponential-phase cultures were chosen for the tests described
Chemicals. All reagents and chemicals were obtained from Sigma-Aldrich
(Dorset, United Kingdom). Lidocaine hydrochloride, amiloride, and VAN solu-
tions (100 ?M, 100 mM, and 10 mM, respectively) were prepared in Milli-Q
water, stored at 4°C (protected from light), and diluted into the culture medium
to obtain the desired concentration. VTD was dissolved to a final concentration
10 mM in acidic Milli-Q water (pH 2) and stored at ?20°C. Certified standard
solutions of PSP toxins (PSP-1C and STXdiHCl-C) were obtained from the
Institute of Marine Bioscience, National Research Council of Canada, Halifax,
Nova Scotia. PSP toxin standards were stored at ?20°C with the stock solutions
diluted in culture medium to obtain the final test concentrations.
Total cellular Na?and K?content and flame photometry. To evaluate the
effect of 1 ?M STX and 100 ?M VTD on total Na?and K?cellular levels,
aliquots of the same culture (20 ml) of C. raciborskii AWT205 were adjusted to
pH 8.1 by adding HEPES buffer to a final concentration of 10 mM. Samples (2
ml) were harvested before exposure (?5 min), immediately after exposure (0
min), and at 10, 30, and 60 min postexposure. Experimental replicates included
negative controls (unexposed culture sample) and positive controls (10 mM
NaCl). Aliquots of challenged AWT205 cultures were collected by centrifugation
in 2-ml plastic tubes at 11,000 ? g for 15 min. All sampled pellets were resus-
pended in 0.5 ml of diluent flame solution (3 mM lithium in MilliQ water) and
analyzed for total Na?and K?cellular content by using a FLM3 flame photom-
eter (Radiometer, Copenhagen, Denmark). All experiments were performed in
quadruplicate. Control traces were subtracted from the tested samples for
threshold correction, and the data were normalized by expressing the values as
the percentile variation over samples at ?5 min.
Cyanobacterial cell lysis test. The cyanobacterial cell lysis assay was based on
the same principles as the animal neuroblastoma and red blood cell culture
assays described elsewhere (21, 29), except for the use of VAN to inhibit Na?/
K?ATPase activity instead of ouabain since ouabain is known to be ineffective
against algal Na?pumps (11). Briefly, 96-well microtiter plates were used for the
cyanolytic assay, in which 100 ?l of the cyanobacterial cultures was inoculated
and exposed to the agents. Controls consisted of untreated aliquots of cyanobac-
terial cultures and samples with 4 ?l of either 10 mM VTD or 10 mM VAN.
Toxic and nontoxic cyanobacteria were also assayed after the addition of a
combination of 4 ?l of 10 mM VTD and 4 ?l of 10 mM VAN. The final
concentration of VTD and VAN, 400 ?M, was chosen based on previous studies
(13, 29). For the treatment of C. raciborskii AWT205 and A. circinalis 271C with
STX and neoSTX at 1 ?M, the PSP toxins were added to the tested wells 30 min
prior to the addition of VTD and VAN to simulate natural the conditions of toxic
cultures. In these experiments, the positive controls consisted of samples exposed
to VTD-VAN added with lidocaine hydrochloride at 1 ?M. Replicates dosed
with amiloride at 10 mM were used as negative controls. In a previous study,
lidocaine hydrochloride was demonstrated to induce an increase in total cellular
Na?, whereas amiloride reduced the cellular ion levels in cyanobacteria (Pomati
et al., unpublished).
In both assays, the microtiter plates were incubated at room temperature
(25°C) for 5 to 8 h (minimum time for complete cell lysis observed in the
VTD-VAN samples). Determination of cell lysis was performed by light micro-
scopic inspection every 30 to 60 min from the onset of the test, and cells were
counted by using a Neubauer improved counting chamber (0.1 mm deep). Com-
plete cyanobacterial lysis in the VTD-VAN samples was utilized as the endpoint
of the assay. If no cyanolysis was observed in the inoculated wells, the presence
of a channel-blocking agent, including PSP toxins, was indicated. Three to five
trials were performed for each strain or treatment to test the reproducibility of
Cell titer assay for metabolic activity. Microtiter plate cytotoxicity assays on
cyanobacterial cells were performed by using the CellTiter 96 nonradioactive cell
proliferation assay kit (Promega Corp., Madison, Wis.). This method utilizes, as
an indicator of a cell’s metabolic activity, the cellular conversion of a tetrazolium
salt into a formazan product that can be quantified with a spectrophotometer
plate reader. Assays were performed essentially, as suggested in the standard
protocol provided with the kit. C. raciborskii cells in mid-exponential growth were
centrifuged (15 min at 4,000 ? g) and concentrated to reach approximately an
optical density at 750 nm of 1. Subsequently, 100-?l aliquots of concentrated
cyanobacterial suspension were inoculated in 96-well microtiter plates and then
exposed to the test agents. For comparison between the metabolic responses of
C. raciborskii strains AWT205 and T3, the culture samples were tested with a
combination of 4 ?l of 10 mM VTD and 4 ?l of 10 mM VAN, yielding a final
concentration of 400 ?M for each compound. Controls consisted of untreated
cyanobacteria. In the evaluation of the effect of combined VTD and STX on C.
raciborskii AWT205, cyanobacterial cells were exposed to 1 ?M STX, incubated
at room temperature for 30 min, and then added to VTD at 100 ?M. Controls
consisted, for each sample, of untreated cells and cyanobacteria exposed to VTD
at 100 ?M. The dye solution was added at different times, and the stop solution
after 4 h of incubation at room temperature (25°C). Plates were allowed to rest
overnight, and then the absorbance at 600 nm was determined with a Metertech
?960 microplate reader (Metertech, Inc., Taipei, Taiwan). All experiments were
performed in quadruplicate, and the data were expressed as an average percent
variation of sample values versus levels in untreated control.
Luminescent bacteria test. Inhibition of bioluminescence in cultures of V.
fischeri NRRL-B-11177 was performed by using a commercially available stan-
dard luminescent bacterium LUMIStox test kit (Dr Bruno Lange GmbH & Co.,
Dusseldorf, Germany) and specific analytical equipment, including the
LUMIStox 300 measuring station and a Lumistherm thermostat. Reactivation of
freeze-dried bacteria and preparation of samples was done according to the
instructions provided. Light emission of reactivated bacteria was adjusted to a
relative intensity of ?1,000 by dilution with sterile 2% NaCl. For the test, 0.5 ml
of luminescent bacterial suspension was combined with 0.5 ml of the test solu-
tions. Test solutions were prepared in sterile saline medium (2% NaCl) and
adjusted to pH 7 with 10 mM phosphate buffer. When needed, STX supple-
mented the solutions to the desired final concentration. Bacterial suspensions
and samples were maintained at 15°C, combined, and monitored with the lumi-
nometer, while allowing bacteria to adapt for 5 to 10 min. Subsequently, 100 ?M
VTD was added, and the light emission was measured over time. Positive and
negative controls were included for each test and consisted of bacteria exposed
to only VTD and unexposed samples, respectively. The effect on biolumines-
cence was monitored and was expressed as the percent inhibition relative to the
untreated controls. Values were calculated by using, as a correction factor, the
changes in intensity of the controls. This was achieved by subtracting the trace
negative control reading from the test samples over the duration of the experi-
Statistical analyzes. All graphical and statistical analyses were performed by
using PC Origin 5.0 software (Microcal Software, Inc., Northampton, Mass.).
Effect of Na?stress, VTD, and STX on the total cellular
Na?and K?content. In C. raciborskii AWT205, the stress
induced by 10 mM NaCl increased the total cellular Na?levels
compared to the untreated controls (Fig. 1A). Na?uptake by
the cells was shown to be very rapid, and the total cyanobac-
terial sodium content remained stable over the 60-min course
of the experiment. The total K?content of cells was only
slightly affected by 10 mM NaCl, indicating that the homeosta-
sis of K?is of marginal consequence in the overall cyanobac-
terial Na?stress response. On the other hand, both Na?and
K?cellular levels in C. raciborskii AWT205 were altered due to
the effects of STX at 1 ?M and VTD at 100 ?M (Fig. 1B). The
addition of VTD dramatically stimulated cyanobacterial Na?
7372 POMATI ET AL.APPL. ENVIRON. MICROBIOL.
and K?accumulation, whereas STX markedly inhibited the
cellular uptake and thus the intracellular levels of both ions.
These results suggest that Na?flux is not the only cellular
response elicited by these two compounds in cyanobacteria.
Lysis test with toxic and nontoxic cyanobacteria. The effects
of VTD and VAN each at 400 ?M and in combination were
monitored by microscopically counting the cells of the two
strains of C. raciborskii: AWT205 and T3. Cell densities were
recorded over a 180-min time period (Fig. 2). VAN had no
significant effect on both strains, whereas culture samples of
AWT205 were manifestly more sensitive to VTD and the com-
bination of both VTD and VAN compared to the STX-pro-
ducing strain T3. For AWT205, the first indications of cya-
nobacterial lysis occurred after 60 min of exposure to VTD or
VTD-VAN, whereas cell lysis was evident for both T3 and
AWT205 at 120 min after the onset of the experiment. The
combination of VTD and VAN was more effective than VTD
alone in both strains as indicated by the decrease in the num-
ber of intact cells (Fig. 2). In AWT205 cultures, 180-min ex-
posure to VTD-VAN resulted in almost complete lysis of cells.
By microscopic examination, cyanobacterial lysis occurred af-
ter evident enlargement of the cells, which subsequently burst.
This indicated that lysis could have been caused by a dramatic
increase in the intracellular osmotic pressure due to excessive
PSP toxin-producing and nonneurotoxic strains of the cya-
nobacterial species C. raciborskii and A. circinalis were assayed for
resistance to VTD-VAN by the method described above, and the
results are summarized in Table 1. All of the non-PSP-toxin-
producing strains tested showed complete cell lysis after 3 to 8 h,
with no filaments recorded, whereas the neurotoxic strains were
characterized by intact filaments and a lack of complete cell lysis
even after overnight exposure to these agents.
Two non-PSP toxins producing cyanobacterial strains, C.
raciborskii AWT205 and A. circinalis 271C, were also chosen to
evaluate the effect of STX and neoSTX at 1 ?M on the stress
induced by exposure to VTD-VAN. The results, summarized
in Table 2, demonstrated that the production of PSP toxins
prevented complete cell lysis caused by VTD-VAN in culture
samples, as seen for the positive controls and in the cyanobac-
teria added with lidocaine hydrochloride at 1 ?M. In cyanobac-
teria exposed to lidocaine, cell lysis was observed at the same
time as the untreated controls. On the other hand, the effects
of amiloride at 1 mM, STX at 1 ?M, and neoSTX at 1 ?M were
comparable with regard to the level of inhibition of cyanobac-
Effect of VTD and STX on cyanobacterial metabolic activity.
In order to investigate whether the increased Na?uptake in
cyanobacterial cells was coupled to a consequent measurable
FIG. 1. (A) Time course of total cellular Na?and K?levels in C.
raciborskii AWT205 cultures exposed to 10 mM NaCl (Na?I and K?
?) compared to untreated control samples (Na?F and K?E). (B) Ef-
fects of STX at 1 ?M (Na?? and K?ƒ) and VTD at 100 ?M (Na?
Œ and K?‚) on total cellular Na?and K?concentrations in C.
raciborskii AWT205. All values are the mean of four experimental
replicates and are expressed as the percent variation over time.
FIG. 2. Effects of VTD and VAN at 400 ?M and their combination
on cell numbers in samples of C. raciborskii AWT205 and T3. All
values represent the average of five experimental replicates and are
expressed as the percent variation over time. Symbols for C. raciborskii
AWT205: I, VAN; F, VTD; and ?, VAN-VTD. Symbols for C.
raciborskii T3: ?, VAN; E, VTD; and ƒ, VAN-VTD.
TABLE 1. Cyanobacterial strains used in this study and their
relative resistance to treatment with VTD (400 ?M) plus
VAN (400) ?M
C. raciborskii AWT205
C. raciborskii T3
A. circinalis 344B
A. circinalis 134C
A. circinalis 150A
A. circinalis 131C
A. circinalis 271C
A. circinalis 306A
A. circinalis 332H
a?, Complete cell lysis and filaments absent; ?, partial cell damage with intact
bP. Baker, unpublished data.
VOL. 69, 2003 STX BLOCKS BACTERIAL Na?AND K?UPTAKE 7373
decrease in general metabolic activity, cultures of C. raciborskii
AWT205 were treated with 400 ?M VTD–400 ?M VAN, 100
?M VTD, or 100 ?M VTD–1 ?M STX and then monitored by
using a CellTiter 96 cell proliferation assay. As a control,
cultures of the STX-producing C. raciborskii T3 were also
exposed to 400 ?M VTD–400 ?M VAN, and the metabolic
activity was measured over time. The treatments with VTD
and VTD-VAN both resulted in a reduced metabolic activity
of C. raciborskii strains over a 90-min exposure; however, the
toxic strain T3 was affected less than AWT205 (Fig. 3). Treat-
ment with STX at 1 ?M alone did not result in any adverse
effect on cyanobacterial metabolism (data not shown). The
addition of 1 ?M STX to cultures of AWT205, followed by 100
?M VTD, induced a less dramatic decrease in metabolic ac-
tivity compared to exposure to 100 ?M VTD alone (Fig. 3).
Effect on bioluminescence by V. fischeri. In the standard
toxicity test LUMIStox, no significant effect of varied concen-
trations of STX (from 100 nM to 1 ?M) on the luminescence
of V. fischeri was observed. On the other hand, treatment with
VTD at 100 ?M reduced bacterial light emission over time, as
shown in Fig. 4A. The addition of STX (1 ?M) to the Vibrio
cultures after exposure to 100 mM VTD resulted in a slight
recovery of bioluminescence (maximum of 4.4% after 70 min)
compared to samples with no STX added (Fig. 4A). Exposure
of the bioluminescent bacteria to STX prior to the addition of
100 ?M VTD, however, led to different responses over time for
the test samples compared to the controls with no STX added.
Figure 4B shows the time course of the percent light emission
for the various treatments, with values corrected by subtracting
the trace levels of unexposed control samples to eliminate
threshold variations in bioluminescence. In the first 2 min after
the addition of 100 ?M VTD, samples not exposed to STX
drastically decreased their light emission compared to bacterial
solutions treated with the channel-blocking toxin. STX at 600
nM was the most effective treatment for preventing the VTD
effect (21% compared to VTD controls), followed by STX at
300 nM (19.5%) and at 1.2 ?M (18.3%). By 5 min after the
addition of VTD, samples with no STX added reached a level
of bioluminescence that were, on average, 6% higher than the
untreated controls. Bacterial solutions with STX at 300 and
600 nM showed no significant variation over control levels.
Samples with STX added to 1.2 ?M did not completely recover
control levels of light emission, attaining an average of ?3.8%
of the luminescence in unexposed bacteria.
During the present study, we observed opposing effects of
STX and VTD on cyanobacterial Na?and K?ion fluxes, as
FIG. 3. Time course of metabolic activity in culture samples of C.
raciborskii AWT205 treated with 100 ?M VTD (Œ), 400 ?M VTD–400
?M VAN (?), and 100 ?M VTD–1 ?M STX (E). Control culture
samples of C. raciborskii T3 were also tested with 400 ?M VTD–400
?M VAN (?). All values are the average of 4 experimental replicates
and are expressed as the percent variation over time.
FIG. 4. (A) Time course of bioluminescence in V. fischeri. VTD
(100 ?M) was added 52 min after the onset of the experiment. Sub-
sequently, half of the experimental replicates were supplemented with
1 ?M STX (F), whereas the other half were supplemented with phys-
iological saline solution (?). (B) Effects of 100 ?M VTD on light
emission by samples of V. fischeri exposed to STX at 0 (I), 300 (E),
600 (shaded triangles), and 1,200 (?) nM, expressed as the percent
variation over time and corrected for the control levels. All values are
the average of four experimental replicates.
TABLE 2. Results of the application of channel-blocking agents to
two non-PSP-toxin-producing cyanobacterium strains over a 5-h
exposure time to VTD (400 ?M) plus VAN (400 ?M)
C. raciborskii AWT205
A. circinalis 271C
aControl, samples with no channel blocker added. ?, Complete cell lysis and
filaments absent; ?, partial cell damage with intact filaments present.
7374 POMATI ET AL.APPL. ENVIRON. MICROBIOL.
measured by flame photometry analysis. The effects detected
were rapid but not Na?specific, as predicted by the interaction
of these compounds with eukaryotic cells. Our results sug-
gested either that STX and VTD have less specific effects on
prokaryotic cells than those reported in the literature for eu-
karyotic sodium fluxes or that the target of these two agents on
cyanobacterial cells is a binding protein involved in both Na?
and K?homeostasis. The latter hypothesis is consistent with
several reports in the literature demonstrating the presence, in
cyanobacterial cells, of channel proteins that are permeable to
both sodium and potassium ions (19, 22, 23).
Based on these observations, PSP toxin-producing and non-
neurotoxic cyanobacterial strains were assayed to investigate
whether, in vivo, the production of STX would have prevented
the excessive ion uptake and subsequent cell lysis induced by
VTD stress. As noted from microscopic observations, non-
PSP-toxin-producing cyanobacteria under VTD-VAN stress
were swollen, probably due to an increase in the internal os-
motic pressure, and subsequently collapsed within 5 to 8 h.
Similar findings were reported in previous studies with animal
cells subjected to the activity of the sodium channel activator
and the ion pump inhibitor ouabain (29). In contrast, PSP
toxin-producing cultures exhibited lower rates of cell lysis. This
differential sensitivity to VTD and VAN exposure is proposed
to be due to the presence of channel-blocking compounds in
the cultures, which interfere with the action of the channel-
activating agents. To verify this hypothesis and to exclude the
possibility of a variable intrinsic sensitivity of the toxic strains
to the chemicals used, the two nonneurotoxic cyanobacteria C.
raciborskii AWT205 and A. circinalis 271C were exposed to
VTD-VAN in the presence of STX and neoSTX at 1 ?M. This
experiment clearly revealed the acquired resistance to lysis of
nonneurotoxic strains after the addition of PSP toxins to the
cultures. Therefore, in vivo, a direct antagonism of STX and
VTD was demonstrated in a prokaryotic microorganism, a
result similar to what has been noted in the eukaryotic cell-
based assays for channel-blocking toxins (21).
Consistent with our hypothesis that the main effect of VTD
on cyanobacterial cells was due to the increased uptake of Na?
and K?ions, we investigated whether such stress was corre-
lated with a decrease in the metabolic activity of the cyanobac-
teria. Sodium, above a certain critical concentration, repre-
sents a threat to normal cellular functions. This ion, if in excess
compared to normal physiological levels, can disrupt several
crucial biological functions, such as photosynthetic and elec-
tron transport activities in cyanobacterial cells (2). The cell
titer toxicity assays, applied in the present study with the ad-
dition of VTD-VAN, demonstrated that antagonistic effects of
STX and VTD can also be detected via the monitoring of
bacterial metabolic activity.
Since measurements of metabolic activity, compared to cell
lysis, represent a more precise, easy-to-quantify, and standard-
ized means of investigation, we applied the principle of induc-
ing ion uptake by using VTD in the bioluminescent bacterium
V. fischeri. Such stress, resulting in decreased primary metab-
olism, can be measured in this microorganism as variations in
light emission by a commercially standardized method. In sus-
pensions of V. fischeri, VTD exposure reduced biolumines-
cence. Light emission also followed a similar pattern after the
post-VTD addition of STX (Fig. 4A). This effect could be
explained by the occurrence of nonreversible cell damage
caused by either the initial VTD stress or a critical increase in
cytoplasmic Na?levels (2). Alternatively, these data could
indicate a difference in the affinity of VTD and STX for a
putative binding site on the bacterial cells. STX may have less
specificity than VTD for the receptor molecule in V. fischeri, or
the binding protein(s) on the bacterial cell membranes could
have a completely different structure compared to the defined
targets of these two agents, i.e., the eukaryotic voltage-gated
On the other hand, exposure to STX prior to the addition of
VTD to bioluminescent bacteria resulted in the observed dif-
ference in the time course of cellular metabolic activity. As
expected for changes in membrane ion fluxes, the effect dis-
played by V. fischeri was rapid and dose dependent. A 1.2 ?M
concentration of STX was shown to have a minor level of
toxicity to bioluminescent bacteria. This effect could be a result
of the prolonged inhibition of basal bacterial Na?and K?
activity by such high concentrations of the channel-blocking
toxin. In general, however, during the course of the present
study, STX alone did not have any particular effect on bacterial
(i.e., Vibrio sp. or cyanobacteria) growth or metabolism. Ac-
cordingly, exposure to STX could not be used directly in a
bacterial bioassay. These data were consistent with reports that
show low to no toxicity of STX on other microorganisms (14,
30). In contrast, no investigations of the effects of STX on
prokaryotic ion fluxes have been reported. Bacterial ion chan-
nels are single-domain proteins (for reviews, see references 3
and 7) that have been reported to be insensitive to both STX
and TTX and confirmed by recent findings regarding the volt-
age-gated prokaryotic Na?channel in the halophilic bacterium
Bacillus halodurans (27). Further comprehensive investigations
of the effect of STX on microorganisms may lead to important
evolutionary findings regarding an ancestral ion channel sen-
sitive to neurotoxins.
The application of the bioluminescent bacterium method
described here for toxicity assays showed a nanomolar order of
magnitude detection range for STX and a limit of ?300 nM.
Preliminary data also confirmed (29) that varying the concen-
trations of VTD or VAN or the number of cells in the assay
can affect the sensitivity of the test. In the present study, the
parameters were selected to provide the best results in a short
time scale, as required for a rapid bioassay. We predict that
additional development and standardization of this test would
afford a novel and accurate method for the detection and
quantification of PSP toxins. All three gram-negative bacteria
tested with VTD showed a lower affinity to STX compared to
their potency against eukaryotic cells. However, the use of
microorganisms represents an easy, economic, and ethical al-
ternative to animal tests for screening environmental, clinical,
and industrial samples for neurotoxins, including PSP toxins,
tetrodotoxin, ciguatoxins, and brevetoxins.
In conclusion, the present study presents the first evidence
of the effect of the Na?channel blocker STX, as well as the
sodium channel activator VTD, on the Na?and K?ion fluxes
and metabolism of bacterial cells. Previously, these two agents
were thought to act almost exclusively on eukaryotic mem-
brane channels. In addition, we demonstrated the applicability
of VTD and STX antagonism in prokaryotic cells for the de-
velopment of a novel PSP toxin bioassay.
VOL. 69, 2003STX BLOCKS BACTERIAL Na?AND K?UPTAKE7375
ACKNOWLEDGMENTS Download full-text
We thank G. Manarolla for helpful experimental assistance and L.
Llwellyn for advice.
F.P. is the recipient of research scholarships from the University of
New South Wales and the School of Biotechnology and Biomolecular
Sciences, together with an A. Lee Travel Scholarship kindly granted by
the School of Microbiology and Immunology.
1. Alam, M., M. Ikawa, J. J. Sasner, Jr., and P. J. Sawyer. 1973. Purification of
Aphanizomenon flos-aquae toxin and its chemical and physiological proper-
ties. Toxicon 11:65–72.
2. Allakhverdiev, S. I., A. Sakamoto, Y. Nishiyama, M. Inaba, and N. Murata.
2000. Ionic and osmotic effects of NaCl-induced inactivation of photosystems
I and II in Synechococcus sp. Plant Physiol. 123:1047–1056.
3. Anderson, P. A., and R. M. Greenberg. 2001. Phylogeny of ion channels:
clues to structure and function. Comp. Biochem. Physiol. B Biochem. Mol.
4. Baker, P. D., A. R. Humpage, and D. A. Steffensen. 1993. Cyanobacterial
blooms in the Murray-Darling basin: their taxonomy and toxicity. Report
8/93. Australian Centre for Water Quality Research, Adelaide, Australia.
5. Carmichael, W. W., W. R. Evans, Q. Q. Yin, P. Bell, and E. Moczydlowsky.
1997. Evidence of paralytic shellfish poisons in the freshwater cyanobacte-
rium Lyngbya wollei (Farlow ex Gomont) comb. nov. Appl. Environ. Micro-
6. Catterall, W. A. 1980. Neurotoxins that act on voltage-sensitive sodium
channels in excitable membranes. Annu. Rev. Pharmacol. Toxicol. 20:15–43.
7. Catteral, W. A. 2001. A one-domain voltage gated sodium channel in bac-
teria. Science 294:2306–2308.
8. Catteral, W. A., and M. Nirenberg. 1973. Sodium uptake associated with
activation of action potential ionophores of cultured neuroblastoma and
muscle cells. Proc. Natl. Acad. Sci. USA 70:3759–3763.
9. Eddy, M. L., and P. E. Jablonski. 2000. Purification and characterization of
a membrane-associated ATPase from Natronococcus occultus, a haloalkali-
philic archaeon. FEMS Microbiol. Lett. 189:211–214.
10. Gallacher, S., and E. A. Smith. 1999. Bacteria and paralytic shellfish toxins.
11. Glimmer, H. 2000. Primary sodium plasma membrane ATPases in salt-
tolerant algae: facts and fictions. J. Exp. Botany 51:1171–1178.
12. Gorham, P. R., J. McLachlan, U. T. Hammer, and W. K. Kim. 1964. Isolation
and culture of toxic strains of Anabaena flos-aquae (Lyngb.) de Breb. Verh.
Int. Verein. Theor. Angew. Limnol. 15:796–804.
13. Hafer, J., A. Siebers, and E. P. Bakker. 1989. The high-affinity K?-translo-
cating ATPase complex from Bacillus acidocaldarius consists of three sub-
units. Mol. Microbiol. 3:487–495.
14. Harada, K., F. Kondo, and L. Lawton. 1999. Laboratory analysis of cyano-
toxins, p. 369–389. In I. Chorus and J. Bartram (ed.), Toxic cyanobacteria in
water. World Health Organization, London, United Kingdom.
15. Harada, T., Y. Oshima, and T. Yasumoto. 1982. Structure of two paralytic
shellfish toxins, gonyautoxins V and VI, isolated from a tropical dinoflagel-
late, Pyrodinium bahamense var. compressa. Agric. Biol. Chem. 46:1861–
16. Hawkins, P. R., N. R. Chandrasena, G. J. Jones, A. R. Humpage, and I. R.
Falconer. 1997. Isolation and toxicity of Cylindrospermopsis raciborskii from
an ornamental lake. Toxicon 1997 35:341–346.
17. Humpage, A. R., J. Rositano, A. H. Bretag, R. Brown, P. D. Baker, B. C.
Nicholson, and D. A. Steffensen. 1994. Paralytic shellfish poisons from Aus-
tralian cyanobacterial blooms. Aust. J. Mar. Freshwater Res. 45:761–771.
18. Kaas, H., and P. Henriksen. 2000. Saxitoxins (PSP toxins) in Danish lakes.
Water Res. 34:2089–2097.
19. Kaneko, T., S. Sato, H. Kotani, A. Tanaka, E. Asamizu, Y. Nakamura, N.
Miyajima, M. Hirosawa, M. Sugiura, S. Sasamoto, T. Kimura, T. Hosouchi,
A. Matsuno, A. Musaraki, N. Nakazaki, K. Nauro, S. Okumura, S. Shimpo,
C. Takeuchi, T. Wada, A. Watanabe, M. Yamada, M. Yasuda, and S. Tabata.
1996. Sequence analysis of the genome of the unicellular cyanobacterium
Synechocystis sp. PCC6803: II. Sequence determination of the entire genome
and assignment of potential protein coding regions. DNA Res. 3(Suppl.
20. Lagos, N., H. Onodera, P. A. Zagatto, D. Andrinolo, S. M. F. Q. Azevedo, and
Y. Oshima. 1999. The first evidence of paralytic shellfish toxins in the fresh-
water cyanobacterium Cylindrospermopsis raciborskii, isolated from Brazil.
21. Manger, R. L., L. S. Leja, S. Y. Lee, J. M. Hungerford, and M. M. Wekell.
1993. Tetrazolium-based cell bioassay from neurotoxins active on voltage-
sensitive sodium channels: semiautomated assay for saxitoxin, brevetoxin,
and ciguatoxins. Anal. Biochem. 214:190–194.
22. Murata, T., K. Takase, I. Yamato, K. Igarashi, and Y. Kakiuma. 1996. The
ntpJ gene in the Enterococcus hirae ntp operon encodes a component of KtrII
potassium transport system functionally independent of vacuolar Na?-
ATPase. J. Biol. Chem. 271:10042–10047.
23. Nakamura, T., R. Yuda, T. Unemoto, and E. P. Bakker. 1998. KtrAB, a new
type of bacterial K?-uptake system from Vibrio alginolyticus. J. Bacteriol.
24. Oshima, Y., M. Hasegawa, T. Yasumoto, G. Hallegaeff, and S. Blackburn.
1987. Dinoflagellate Gimnodium catenatum as the source of paralytic shell-
fish toxins in Tasmanian shellfish. Toxicon 25:1105–1111.
25. Pereira, P., H. Onodera, D. Andrinolo, S. Franca, F. Araujo, N. Lagos, and
Y. Oshima. 2000. Paralytic shellfish toxins in the freshwater cyanobacterium
Aphanizomenon flos-aquae, isolated from Montargil reservoir, Portugal.
26. Pomati, F., S. Sacchi, C. Rossetti, S. Giovannardi, H. Onodera, Y. Oshima,
and B. A. Neilan. 2000. The freshwater cyanobacterium Planktothrix sp. FP1:
molecular identification and detection of paralytic shellfish poisoning toxins.
J. Phycol. 36:553–562.
27. Ren, D., B. Navarro, H. Xu, L. Yue, Q. Shi, and D. E. Clapham. 2001. A
prokaryotic voltage-gated sodium channel. Science 294:2372–2375.
28. Shimizu, Y. 1977. Chemistry and distribution of deleterious dinoflagellate
toxins, p. 261–269. In D. J. Faulkner and W. H. Fenical (ed.), Marine natural
products chemistry. Plenum Press, Inc., New York, N.Y.
29. Shimojo, R. Y., and W. T. Iwaoka. 2000. A rapid hemolysis assay for the
detection of sodium channel-specific marine toxins. Toxicology 154:1–7.
30. Tisa, L. S., J. J. Sekelsky, and J. Adler. 2000. Effects of organic antagonists
of Ca2?, Na?, and K?on chemotaxis and motility of Escherichia coli. J.
31. Zingone, A., and H. O. Enevoldsen. 2000. The diversity of harmful algal
blooms: a challenge for science and management. Ocean Coastal Manage-
7376 POMATI ET AL.APPL. ENVIRON. MICROBIOL.