Species specificity and potential roles of Karlodinium micrum toxin

Abstract

Karlodinium micrum is a toxic mixotrophic dinoflagellate that has been responsible for fish kills in coastal environments worldwide. The role that karlotoxins play in the life history of K. micrum is unknown, but may contribute to its bloom-forming ability. We tested the hypothesis that karlotoxins could inhibit the growth of other protists depending on the sterol composition of target cell membranes. We also examined the effect of toxin addition on feeding rates of K. micrum on a flagellated prey, Storeatula major. Dose-dependent effects of isolated karlotoxin (KmTX2) were tested in growth bioassays (24-48h) of K. micrum, three raphidophytes (Heterosigma akashiwo, Fibrocapsa japonica and Chattonella subsalsa), two cryptophytes (S. major andPyrenomonas salina), and the dinoflagellates Amphidinium carterae, Pfiesteria piscicida andP. shumwayae. Growth of K. micrum, P. salina, A. carterae and P. piscicida were not affected by karlotoxin additions up to 1 000ng ml−1. Other organisms showed growth inhibition at concentrations between 500ng ml−1 and 1 000ng ml−1. Predation by K. micrum on S. major was significantly higher in the presence of 25ng ml−1 KmTX2. The results are consistent with a role for karlotoxin in allelopathic inhibition of competitors and/or prey immobilisation depending on sterol composition.

Full text

African Journal of Marine Science 2006, 28(2): 415–419
Printed in South Africa — All rights reserved
Copyright © NISC Pty Ltd
AFRICAN JOURNAL OF
MARINE SCIENCE
EISSN 1814–2338
Species specificity and potential roles of Karlodinium micrum toxin
JE Adolf1*, TR Bachvaroff1, DN Krupatkina1,H Nonogaki1, PJP Brown2, AJ Lewitus2,3, HR Harvey4and AR Place1
1Center of Marine Biotechnology, University of Maryland Biotechnology Institute, Baltimore, MD 21202, USA
2Belle W Baruch Institute for Marine and Coastal Sciences, University of South Carolina, Hollings Marine Laboratory, 331
Fort Johnson Road, Charleston, SC 29412, USA
3Marine Resources Research Institute, South Carolina Department of Natural Resources, Hollings Marine Laboratory, 331
Fort Johnson Road, Charleston, SC 29412, USA
4Chesapeake Biological Laboratory, University of Maryland Center for Environmental Science, Solomons, MD 20688, USA
* Corresponding author, e-mail: adolf@umbi.umd.edu
Introduction
Karlodinium micrum is a cosmopolitan toxic dinoflagellate
that has been associated with fish kills along the eastern
seaboard of the United States and in other temperate
coastal environments (Deeds et al. 2002, Kempton et al.
2002). It produces compounds, putatively named karlo-
toxins, that have been shown to be haemolytic, cytotoxic
and ichthyotoxic in laboratory experiments (Deeds et al.
2002, Kempton et al. 2002).
The ecological function that karlotoxins serve for K. micrum
is unknown. Particularly for dinoflagellates, production of
cytotoxic metabolites may yield a competitive advantage for
the producing organism that enables it to dominate the
phytoplankton and form blooms. Dinoflagellates, including K.
micrum, have relatively slow autotrophic growth rates and
thus may benefit significantly from alternative strategies that
allow them to out-compete other phytoplankton. Allelopathy,
the production of substances that hinder the growth or
physiological function of potentially competitive organisms, is
one such strategy (Legrand et al. 2003).
Mixotrophic nutrition is another means by which K. micrum
may out-compete other phytoplankton: K. micrum can feed on
other protists and achieve mixotrophic growth rates two- to
three-fold higher than its maximum autotrophic growth rate
(Li et al. 1999). Recent work with the prymnesiophyte
Prymnesium parvum (Skovgaard and Hansen 2003) has
suggested that toxin production can immobilise and facili-
tate capture of motile prey. The role of toxin production in
feeding by K. micrum has not been investigated.
Here, we tested the species specificity and potential roles
of karlotoxins produced by K. micrum. We hypothesised
that the lysis of different target species by karlotoxin would
depend on sterol composition of the target cell’s membrane
(see Deeds and Place 2006). Because laboratory cultures
and natural populations of K. micrum have been observed
feeding on flagellated cryptophyte prey (Li et al. 1999,
2000), we also hypothesised that pre-treatment of prey with
karlotoxin would result in higher feeding rates as a result of
prey immobilisation.
Material and Methods
Cultures
The cultures used in these experiments are listed in Table 1.
Karlodinium micrum is a toxic mixotrophic dinoflagel-
late that has been responsible for fish kills in coastal
environments worldwide. The role that karlotoxins play
in the life history of K. micrum is unknown, but may
contribute to its bloom-forming ability. We tested the
hypothesis that karlotoxins could inhibit the growth of
other protists depending on the sterol composition of
target cell membranes. We also examined the effect of
toxin addition on feeding rates of K. micrum on a flagel-
lated prey, Storeatula major. Dose-dependent effects of
isolated karlotoxin (KmTX2) were tested in growth
bioassays (24–48h) of K. micrum, three raphidophytes
(Heterosigma akashiwo, Fibrocapsa japonica and
Chattonella subsalsa), two cryptophytes (S. major and
Pyrenomonas salina), and the dinoflagellates Amphi-
dinium carterae, Pfiesteria piscicida and P. shumwayae.
Growth of K. micrum, P. salina, A. carterae and P. pisci-
cida were not affected by karlotoxin additions up to
1 000ng ml–1. Other organisms showed growth inhibi-
tion at concentrations between 500ng ml–1 and 1 000ng
ml–1. Predation by K. micrum on S. major was signifi-
cantly higher in the presence of 25ng ml–1 KmTX2. The
results are consistent with a role for karlotoxin in
allelopathic inhibition of competitors and/or prey immo-
bilisation depending on sterol composition.
Keywords: allelopathy, Karlodinium, karlotoxin, mixotrophy

Adolf, Bachvaroff, Krupatkina, Nonogaki, Brown, Lewitus, Harvey and Place416
Growth rate bioassays
Bioassays were conducted in three separate sets of
experiments. Purified karlotoxin, KmTX2, was added to
growth medium at twice the target treatment concentration
in sterile glass tubes. An equivalent volume of carrier
methanol was added to controls that did not receive toxin.
Equal volumes of algal culture and karlotoxin-containing
media were then pipetted, using glass pipettes, into wells
of a 24-well polystyrene well plate. In Experiment 1, run in
triplicate, cells were enumerated under a microscope after
48h exposure. In Experiment 2, a series of target cell
concentrations (3 250 cells ml–1, 7 250 cells ml–1, 75 000
cells ml–1 and 150 000 cells ml–1) was used and cells were
enumerated by a Coulter®Counter after 20h exposure. In
Experiment 3, cells were enumerated under the micro-
scope after 24h exposure to toxin. The cell count data for
each karlotoxin concentration tested in each experiment
were expressed as a growth rate,
where N0and N1are the initial and final target cell concen-
trations respectively (cells ml–1), and t0and t1are the start
and end times respectively (days). Growth rate was plotted
against karlotoxin concentration and subjected to a linear
regression analysis (S-PLUS Pro 6.1) to determine whether
karlotoxin concentration had a significant effect on growth
rate. In Experiments 1 and 3, where each karlotoxin
concentration was replicated, all data were used in fitting
the linear regressions.
Feeding experiments
Cultures of K. micrum strains CCMP 1974 (34 110 cells
ml–1) and CCMP 2064 (87 865 cells ml–1) grown in f/2-Si
medium were diluted to equal cell concentrations (34 000
cells ml–1) with growth medium. S. major prey (334 300
cells ml–1) were diluted three-fold with f/2-Si medium
containing a 10% P concentration. This diluted prey
suspension was used to prepare a double karlotoxin treat-
ment concentration of 50ng ml–1 KmTX2 in a sterile glass
tube. Treatments with no toxin but an equivalent volume
(0.03% v/v) of carrier methanol were prepared as controls.
Triplicate treatments were prepared in a polystyrene 24-well
plate by pipetting prey suspension into wells containing an
equal volume of either growth medium (f/2-Si), K. micrum
CCMP 1974 or K. micrum CCMP 2064. The well plate was
covered, wrapped with polyvinyl chloride film and returned
to the growth chamber. After 5h, the treatments were fixed
by adding gluteraldehyde (1% final concentration) and
allowed to settle for >24h at 4°C. At least 100 random cells
were counted at 400Xmagnification, under an Olympus
inverted microscope equipped with phase contrast, and
scored for the presence or absence of an orange inclusion
indicative of an ingested cryptophyte.
Sterol analyses
Sterol analyses were performed as described by Deeds
and Place (2006) and Place et al. (2006).
Results and Discussion
In Experiment 1, raphidophytes (with exception of H. aka-
shiwo 2) and the cryptophyte S. major were affected by
karlotoxin addition, resulting in a reduction of growth rate that
was dependent on toxin concentration (Figure 1, Table 2). K.
micrum was not affected by toxin addition (Figure 1, Table 2).
In Experiment 2, the dinoflagellate A. carterae and the
cryptophyte P. salina were not affected by karlotoxin
although the cryptophyte S. major was affected, except in
one of the cell density treatments (Table 2). P. shumwayae
was affected by karlotoxin addition whereas P. piscicida was
not (Table 2). P. shumwayae responded to toxin addition by
forming cysts, resulting in an apparent reduction of cell
numbers that was expressed as a reduction of growth rate
in the experiments. The sterol composition of these two
species of Pfiesteria differed in that, when compared with P.
piscicida, P. shumwayae showed elevated levels of brassi-
casterol, the presence of [[(3β, 22E)-ergosta-7,22-dien-3-
yl]oxy]trimethyl-sterols, and the absence of two 4α-methyl
sterols that were abundant in P. piscicida (Table 3).
Feeding experiments
Ingestion rates were 2–3 times higher on prey in the
presence of 25ng ml–1 karlotoxin (Figure 2a, b). Treatment
of prey with the double toxin concentration of 50ng ml–1 for
approximately 10 minutes before dilution at the beginning
of the feeding assay had no effect on prey cell numbers.
The production of bioactive compounds by phytoplankton
presumably yields an ecological advantage to the producer,
although this aspect of toxin production is far less under-
stood than the effects of toxins on human health or econo-
mic interests such as fisheries. Our interest in the biological
role(s) of karlotoxin stems from the desire to understand the
evolutionary origin of toxin production, and from a desire to
understand what factors help K. micrum form blooms in
nature. The present results show that high doses of
karlotoxin (>500ng m–1) inhibited the growth of raphido-
phytes, and one of two cryptophyte and Pfiesteria species
Table 1: Cultures used in the experiments
Maintenance
Culture Strain conditions
Chattonella subsalsa CAAE 1662X 20ppt f/2-Si
Fibrocapsa japonica C-101 20ppt f/2-Si
Heterosigma akashiwo CAAE 1663X 20ppt f/2-Si
Karlodinium micrum 010410-1C6 30ppt f/2-Si
K. micrum CCMP 1974 15ppt f/2-Si
K. micrum CCMP 2064 15ppt f/2-Si
Pyrenomonas salina 15ppt f/2-Si
Storeatula major Strain-g 15ppt f/2-Si
Pfiesteria piscicida CCMP 1829 15ppt
P. shumwayae CCMP 2089 15ppt
Growth rate day
N
N
tt
−=−
⎛
⎝
⎜⎞
⎠
⎟
10
10
1
ln

African Journal of Marine Science 2006, 28(2): 415–419 417
tested. A. carterae and K. micrum were not affected by the
toxin. Furthermore, K. micrum ingested more cryptophytes
in the presence of a low dose (25ng ml–1) of karlotoxin in a
5h feeding assay.
Allelopathic chemicals are most effective when the
producing organism is not affected by the allelochemical.
K. micrum is immune to its own toxin (Deeds and Place
2006), and we suspect that specificity of karlotoxin
depends on the sterol composition of the target cell’s
membrane. Sterol composition is a critical factor determin-
ing toxicity in several pore-forming toxins (see Deeds and
Place 2006 for references). The sterol dominant in K.
micrum (gymnodinosterol [(24S)-4α-methyl-5α-ergosta-
8(14),22-dien-3β-ol]; Leblond and Chapman 2002) did not
interact strongly with karlotoxin (Deeds and Place 2006).
Ergosterol and cholesterol did interact strongly with karlo-
toxin, making membrane pore formation and cytotoxicity
more likely in organisms containing these membrane
sterols (Deeds and Place 2006). These observations point
to possible roles of the 4α-methyl group or the 8(14)
double bond (both present in gymnodinosterol, both absent
in cholesterol and ergosterol) in determining whether or not
a sterol interacts with karlotoxin. Our growth response
results were consistent with this association in eight of nine
species (P. salina was the exception, see below). The
sterol composition of A. carterae (immune to KmTX2) is
dominated by amphisterol that contains both the 4α-methyl
and 8(14) double bond (Withers et al. 1979, the present
data). P. piscicida (immune to KmTX2) sterols are domi-
nated by dinosterol that contains a 4α-methyl but no 8(14)
double bond. Raphidophytes, cryptophytes and P. shum-
wayae have desmethyl sterols as major membrane
components (Volkman 2003). In the analyses, desmethyl
Figure 1: Representative plot showing growth rate plotted as a
function of KmTX2 (karlotoxin) concentration for K. micrum, F.
japonica and S. major. Similar plots were used to derive the
statistics in Table 2
Table 2: Linear regression results for growth rate vs KmTX2 concentration in bioassay experiments. KmTX2 concentration ranged from 0ng
ml–1 to 1 000ng ml–1 in each experiment. t0indicates cell abundance at the start of the experiment. The y-intercept (intercept) corresponds to
the growth rate of the target species in the absence of toxin. A negative slope indicates a suppression of growth rate by toxin
Cell abundance
Target species Class (t0, cells ml–1) Intercept Slope r2p-value
Experiment 1
H. akashiwo 1 Raphidophyte 28 634 0.74 – 0.001 0.35 0.020*
H. akashiwo 2 Raphidophyte 26 132 –0.80 –0.0009 0.26 0.062*
F. japonica Raphidophyte 36 153 0.34 – 0.0027 0.71 <0.001*
C. subsalsa Raphidophyte 5 850 – 0.17 – 0.0011 0.43 0.015*
S. major Cryptophyte 42 024 0.73 –0.014 0.68 < 0.001*
K. micrum Dinoflagellate 28 634 0.10 – 0.0001 0.08 0.310*
Experiment 2
A. carterae Dinoflagellate 3 250 0.85 0.0003 0.18 0.475*
7 500 0.67 – 0.0002 0.61 0.117*
75 000 0.42 < 0.0000 0.01 0.887*
150 000 0.44 – 0.0002 0.32 0.435*
P. salina Cryptophyte 3 250 1.41 –0.0006 0.68 0.088*
7 500 0.20 0.0001 0.03 0.782*
75 000 – 0.88 0.0002 0.02 0.856*
150 000 – 0.95 0.0008 0.42 0.349*
S. major Cryptophyte 3 250 1.13 – 0.0011 0.92 0.009*
7 500 0.34 – 0.0008 0.50 0.182*
75 000 – 0.77 – 0.0034 0.90 0.049*
150 000 –1.03 – 0.0038 0.94 0.033*
Experiment 3
P. piscicida Dinoflagellate 39 375 –0.74 – 0.0002 0.03 0.362*
P. shumwayae Dinoflagellate 9 813 –0.28 –0.0007 0.28 0.004*
* Significant regression (p < 0.05)
1
2
3
0
1
200 400 600 800
KARLOTOXIN (ng ml1)
GROWTH RATE (day1)
K. micrum
(not affected)
F. japonica
(affected)
S. major
(affected)
0

Adolf, Bachvaroff, Krupatkina, Nonogaki, Brown, Lewitus, Harvey and Place418
brassicasterol was the major component of both S. major
and P. salina (>90%). The analyses also showed a higher
abundance of desmethyl sterols in the karlotoxin-sensitive
species of P. shumwayae (Table 3). The lack of a negative
effect of karlotoxin on P. salina may indicate that secondary
mechanisms protect this organism from lysis. It is note-
worthy that the growth rate bioassays used in this study
would not detect sublethal effects of karlotoxin.
The toxin concentrations that were cytotoxic in growth
rate bioassays (>500ng ml–1) would require production by
K. micrum abundances >50 000–500 000 cells ml–1 assu-
ming a toxin quota between 1pg cell–1 and 10pg cell–1
(Deeds et al. 2004) and the release of all cellular toxin into
the surrounding medium. However, observations suggest
that >90% of karlotoxin is cell-associated, albeit loosely,
under normal culture conditions (Deeds et al. 2002). K.
micrum sometimes blooms to concentrations exceeding
50 000 cells ml–1, but generally exists at numbers well
below 10 000 cells ml–1 (Li et al. 2000, Goshorn et al.
2004). These observations suggest that cell-to-cell contact,
enabling delivery of toxin directly to the target cell mem-
brane, may be an important aspect of the allelopathic/prey
immobilisation mechanism in K. micrum. Further experi-
ments are needed to address this conjecture.
We suggest that the stimulation of prey ingestion in the
presence of a low dose of karlotoxin results from sub-lethal
effects of the toxin, resulting in prey immobilisation. When K.
micrum is feeding on a cryptophyte prey, S. major, the pro-
cess of phagocytosis begins with a prey capture sequence
that lasts approximately 30 seconds. During that time, K.
micrum appears to capture the cryptophyte with a narrow
appendage resembling a line (i.e. tow line) that extends
2–5µm between predator and prey (Li et al. 1999). S. major is
flagellated and approximately half the mass of K. micrum, and
during this initial phase of prey capture the two cells are
observed to struggle against each other, often ending in prey
escape (videos available at adolf@umbi.umd.edu). Our
tentative conclusion is that low doses of karlotoxin immobilise
S. major, reducing prey escape success resulting in higher
ingestion rates by K. micrum. Based on these findings, it
appears that karlotoxins may provide both a potential allelo-
pathic and prey capture advantage to K. micrum.
Acknowledgements — This work was funded in part by grants from
Ecology and Oceanography of Harmful Algal Blooms (ECOHAB) and
National Institute of Environmental Health Sciences (to ARP), and
National Oceanic and Atmospheric Administration and Centers for
Disease Control and Prevention (to AJL). This is contribution no. 05-136
from the UMBI Center of Marine Biotechnology, no. 3993 from the
University of Maryland Center for Environmental Science, and no.136
from the ECOHAB programme.
References
Deeds JR, Kibler SR, Tester PA, Place AR (2004) Geographic strain
variation in toxin production in Karlodinium micrum (Dinophyceae)
from Southeastern estuaries of the United States. In: Steidinger
KA, Landsberg JH, Tomas CR, Vargo GA (eds) Harmful Algae
Table 3: Major sterols detected in Pfiesteria species used in this study. Values are the percentage abundance of each sterol, determined
from the flame ionisation detector (FID) trace obtained by gas chromatography
Abundance (%)
Sterol P. shumwayae P. piscicida
(dinosterol) — 4α,23,24-trimethyl-5α-cholest-22-en-3β-ol 20.4 28.5
(brassicasterol) — 24-methylcholesta-5,22-dien-3β-ol 33.1 07.0
4α,23,24-trimethylcholest-17(20)-en-3β-ol 23.4 23.5
4α,24-dimethylcholestan-3β-ol nd 12.2
4α,23,24-trimethylcholest-8(14)-en-3β-ol nd 28.9
[[(3β, 22E)-ergosta-7,22-dien-3-yl]oxy]trimethyl- 11.6 nd
nd = Not detected
CONTROL 25ng ml1
KmTX2
INGESTED PREY PER K. micrum
K. micrum CCMP 2064
Prey: S. major (cryptophyte)
5h assay
p = 0.006
K. micrum CCMP 1974
Prey: S. major (cryptophyte)
5h assay
p = 0.006
0.20
0.15
0.10
0.05
0.20
0.15
0.10
0.05
(a)
(b)
Figure 2: Ingested prey per K. micrum without (control) or with
(25ng ml–1 KmTX2) karlotoxin treatment in K. micrum strains
CCMP for (a) 2064 and (b) 1974

African Journal of Marine Science 2006, 28(2): 415–419 419
2002. Florida Fish and Wildlife Conservation Commission, Florida
Institute of Oceanography, and Intergovernmental Oceanographic
Commission of UNESCO, St Petersburg, Florida, pp 145–147
Deeds JR, Place AR (2006) Sterol-specific membrane interactions
with the toxins from Karlodinium micrum (Dinophyceae) — a
strategy for self-protection? In: Pitcher GC, Probyn TA, Verheye
HM (eds) Harmful Algae 2004. African Journal of Marine Science
28(2): 421–425
Deeds JR, Terlizzi DE, Adolf JE, Stoecker DK, Place AR (2002) Toxic
activity from cultures of Gyrodinium galatheanum (Dinophyceae) –
a dinoflagellate associated with fish mortalities in an estuarine
aquaculture facility. Harmful Algae 1: 169–189
Goshorn D, Deeds J, Tango P, Poukish C, Place AR, McGinty M,
Butler W, Luckett C, Magnien R (2004) Occurrence of Karlodinium
micrum and its association with fish kills in Maryland estuaries. In:
Steidinger KA, Landsberg JH, Tomas CR, Vargo GA (eds) Harmful
Algae 2002. Florida Fish and Wildlife Conservation Commission,
Florida Institute of Oceanography, and Intergovernmental
Oceanographic Commission of UNESCO, St Petersburg, Forida, pp
361– 363
Kempton JW, Lewitus AJ, Deeds JR, Law J McH, Wilde SB, Place
AR (2002) Toxicity of Karlodinium micrum (Dinophyceae) associ-
ated with a fish kill in a South Carolina brackish retention pond.
Harmful Algae 1: 233–241
Leblond JD, Chapman PJ (2002) A survey of the sterol composition
of the marine dinoflagellates Karenia brevis, Karenia mikimotoi,
and Karlodinium micrum: distributions of sterols within other
members of the class Dinophyceae. Journal of Phycology 38:
670–682
Legrand C, Rengefors K, Fistarol GO, Granéli E (2003) Allelopathy
in phytoplankton – biochemical, ecological and evolutionary
aspects. Phycologia 42: 406–419
Li A, Stoecker DK, Adolf JE (1999) Feeding, pigmentation, photosyn-
thesis and growth of the mixotrophic dinoflagellate Gyrodinium
galatheanum. Aquatic Microbial Ecology 19: 163–176
Li A, Stoecker DK, Coats DW (2000) Spatial and temporal aspects
of Gyrodinium galatheanum in Chesapeake Bay: distribution and
mixotrophy. Journal of Plankton Research 22: 2105–2124
Place AR, Harvey HR, Bai X, Coats DW (2006) Sneaking under the
toxin surveillance radar: parasitism and sterol content. In: Pitcher
GC, Probyn TA, Verheye HM (eds) Harmful Algae 2004. African
Journal of Marine Science 28(2): 347–351
Skovgaard A, Hansen PJ (2003) Food uptake in the harmful alga
Prymnesium parvum mediated by excreted toxins. Limnology
and Oceanography 48: 1161–1166
Volkman JK (2003) Sterols in microorganisms. Applied Microbiology
and Biotechnology 60: 495–506
Withers NW, Goad LJ, Goodwin TW (1979) A new sterol, 4α-methyl-
5α-ergosta-8(14),24(28)-dien-3β-ol, from the marine dinoflagellate
Amphidinium carterae. Phytochemistry 18: 899–901
Manuscript received January 2005; accepted November 2005