Harmful Algae 40 (2014) 51–62

A bloom of Karlodinium australe (Gymnodiniales, Dinophyceae)associated with mass mortality of cage-cultured fishes in West JohorStrait, Malaysia

Hong Chang Lim a, Chui Pin Leaw a, Toh Hii Tan b, Nyuk Fong Kon b, Leh Hie Yek b,Kieng Soon Hii b, Sing Tung Teng c, Roziawati Mohd Razali d, Gires Usup e,Mitsunori Iwataki f, Po Teen Lim a,*a Bachok Marine Research Station, Institute of Ocean and Earth Sciences, University of Malaya, 16310 Bachok, Kelantan, Malaysiab Institute of Biodiversity and Environmental Conservation, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysiac Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysiad Fisheries Research Institute Batu Maung, Department of Fisheries Malaysia, 11960 Batu Maung, Pulau Pinang, Malaysiae Faculty of Science and Technology, National University of Malaysia, 43600 Bangi, Selangor, Malaysiaf Asian Natural Environmental Science Center, The University of Tokyo, Tokyo 113-8657, Japan

A R T I C L E I N F O

Article history:

Received 11 June 2014

Received in revised form 2 October 2014

Accepted 15 October 2014

Keywords:

Dinoflagellate

Fish mortality

Harmful algal bloom

Karlodinium australe

Malaysia

West Johor Strait

A B S T R A C T

A recent (February 2014) mass mortality of fishes was observed in the cage-farming region of the West

Johor Strait of Malaysia, involving over four different species of cultured fishes, numbering �50,000 fish.

A field investigation at six stations along the West Johor Strait collected water samples and examined for

the presence of harmful species. Dead fishes were collected for necropsy. The phytoplankton

composition was dominated by a species of Karlodinium, at a considerably high cell concentration (0.31–

2.34 � 106 cells l�1), and constituting 68.8–98.6% of the phytoplankton relative abundance at all

stations. Detailed morphological assessment by light and scanning electron microscopy revealed that the

species was Karlodinium australe de Salas, Bolch and Hallegraeff. This was supported by molecular

evidence of the nuclear encoded large subunit ribosomal gene (LSU rDNA) and the second internal

transcribed spacer (ITS2) via single-cell PCR. The sequences of LSU rDNA yielded 3.6–4.0% divergence

when compared to the sister taxon, K. armiger; and >6.5% when compared to other Karlodinium species.

Fish necropsy showed symptoms similar to those affected by karlotoxin ichthyotoxins. This is the first

report of a mass mortality of cage-cultured and wild fishes attributed to the unarmored dinoflagellate K.

australe.

� 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Harmful Algae

jo u rn al h om epag e: ww w.els evier .c o m/lo cat e/ha l

1. Introduction

Many coastline countries in Southeast Asia, including Malaysia,are experiencing a rapid growth in aquaculture/maricultureindustries, owing to high demands from domestic and interna-tional markets, and the involvement of the government and profit-driven companies (Hishamunda et al., 2009). In Malaysia, theindustry has grown by more than 10% annually between 1990 and2002, with total production of nearly 176,000 tonnes in 2005 (FAO,2007). While the industry has expanded, it has also suffered huge

* Corresponding author. Tel.: +60 9 7785003; fax: +60 9 7785006.

E-mail addresses: ptlim@um.edu.my, poteenlim@gmail.com (P.T. Lim).

http://dx.doi.org/10.1016/j.hal.2014.10.005

1568-9883/� 2014 Elsevier B.V. All rights reserved.

production losses due to frequent fish kills caused by water qualitydegradation or the proliferation of harmful marine diatoms andichthyotoxic dinoflagellates at aquacultures sites.

Several marine microalgal bloom events, some resulting inmassive fish kills, have been documented in Malaysia (reviewed inLim et al., 2012). In 2002, a bloom of Prorocentrum minimum

(Pavillard) Schiller (at >2 � 105 cells l�1) was observed along theJohor Strait, causing massive water discoloration but no fish kills,even near finfish cages (Usup et al., 2003). Subsequent fish-killevents were recorded in the Straits of Malacca, the most importantfinfish-farming area in the country. In 2005–2006, the mostnotable protracted algal bloom associated with the mass mortalityof caged finfish, with an estimated loss of at least 6 million USD,was encountered in Penang, northern Peninsular Malaysia (Sin

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6252

Chew Daily, 2005; reviewed in Lim et al., 2012). The suspectedcausative species was the unarmored dinoflagellate, Cochlodinium

polykrikoides Margalef. Concurrently, fish kills in cage farms werealso reported in Malaysia Borneo (Sepanggar Bay, off KotaKinabalu, Sabah), where blooms and water discoloration wereobserved. It was later found that C. polykrikoides was responsiblefor the event, with a maximum cell density of 6 � 106 cells l�1

(Anton et al., 2008). In a recent field investigation along the JohorStrait, potentially harmful unarmored dinoflagellates, Karlodinium

veneficum (Ballantine) Larsen and Karenia mikimotoi (Miyake &Kominami ex Oda) Hansen and Moestrup, were documented (Tanet al., 2013).

The genus Karlodinium gained significant scientific scrutiny afterthe first described bloom of Karlodinium veneficum (syn. K. micrum;Bergholtz et al., 2005) in Walvis Bay (Namibia), in 1950, harmedfishes and marine invertebrates (Ballantine, 1956; Abbott andBallantine, 1957; Braarud, 1957). In the 2000s, periodic fish-killincidents associated with K. veneficum blooms were reported in theUnited States (Deeds et al., 2002; Kempton et al., 2002; Fensin, 2004;Goshorn et al., 2004). Fish assays using cultured K. veneficum cellsshowed toxic effects on juvenile fish and molluscan shellfish(Nielsen and Strømgren, 1991; Nielsen, 1993; Galimany et al., 2008).

After decades of fish-kill events associated with Karlodinium

veneficum, the responsible toxic compounds were discovered in2002, and named karlotoxins (KmTxs) (Deeds et al., 2002, 2004a;Kempton et al., 2002). These potent amphipathic ichthyotoxins arehemolytic, cytotoxic, algicidal and ichthyotoxic (Mooney et al.,2010). Karlotoxins cause pore formation in biological membranes,increasing their ionic permeability (Deeds, 2003). Their biologicalactivity is dependent on the sterol composition (Adolf et al.,2006a,b; Deeds and Place, 2006). Organisms with a predominanceof 4-desmethyl sterols, such as fishes, are susceptible to the toxins,which damage gill epithelial tissues (Deeds et al., 2006; Place et al.,2012). Toxin production varies geographically by strain and eco-physiological condition (Deeds et al., 2004a; Adolf et al., 2009).

On February 11, 2014, wild as well as caged fishes were founddead in the West Johor Strait of Malaysia. A field investigation wasundertaken to investigate the cause of this mass mortality. Here,we reveal the causative bloom species, based on morphologicaland molecular evidence.

Fig. 1. Map of the western Johor S

2. Materials and methods

2.1. Study site and sampling

The study area is located in the West Johor Strait, TanjungKupang, Malaysia (1.21518 N, 103.37588 E) (Fig. 1). The WestJohor Strait receives freshwater inflows from the Pendas Riverand is subjected to tidal influence by the Straits of Malacca.Sampling was undertaken a week after the fish-kill event wasreported in the press on February 20–21, 2014 (Sin Chew, TheStar, 2014). Six sampling stations (Stn 1–6) along the West JohorStrait were selected to include cage-farming areas (Stn 3), as wellas areas influenced by anthropogenic activities such as landreclamation (Stn 2), a jetty (Stn 4), a river mouth (Stn 5), and amarina (Puteri Harbour) (Stn 6). Station 1, a fishing kelong (aplatform built predominantly with wood) served as a control site(Fig. 1).

Quantitative water samples were collected at each station usinga Van Dorn water sampler. Plankton samples were collected usinga 20-mm mesh-size plankton net. Samples containing livedinoflagellates were brought back to the laboratory for cultureestablishment and molecular characterization.

2.2. Determination of phytoplankton composition and water

parameters

Duplicate water samples were collected for analyses. Aliquots(1 l) were concentrated by filtration and preserved with 1% acidicLugol’s solution for phytoplankton enumeration and composition.Cell counts were performed using a Sedgwick-Rafter countingchamber under a Leica CME light microscope (Leica MicrosystemsGmBH, Wetzlar, Germany). Cell abundances of each station wereplotted with Surfer 9 (Golden Software Inc., CO, USA). Phytoplank-ton composition was plotted using GraphPad Prism 5 (GraphPadSoftware Inc., CA, USA).

Temperature, salinity and pH of seawater were measured insitu. The data were plotted with Surfer 9, and Ocean Data View(http://odv.awi.de/) was used to contour the distribution ofdepths, sea-surface temperature, salinity and pH in the studyarea.

trait, showing sampling sites.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–62 53

2.3. Detailed morphological characterizations

Lugol’s preserved cells were examined with an Olympus BX51microscope (Olympus, Tokyo, Japan), using differential interfer-ence contrast (DIC) optics. For SEM examination, Lugol’s preservedfield samples were fixed with 2% osmium tetroxide overnight. Thesamples were then washed with a decreasing salinity of filteredseawater and finally rinsed with distilled water before dehydrationby a graded series of ethyl alcohol. Samples then underwent criticalpoint drying, after intermediate substitution by amyl acetate. Thesamples were coated with gold-palladium using a JEOL JFC-1600magnetron sputter coater (JEOL Ltd., Tokyo, Japan), and observedunder a JEOL JSM-6390LA analytical scanning electron microscope(JEOL).

2.4. DNA isolation, single-cell PCR and sequencing

Single cells collected from the field were isolated by micropi-pette under an Olympus IX51 inverted light microscope (Olym-pus). Each single cell was rinsed in droplets of, sterile-filteredseawater prior to transferring into a 0.2-ml PCR tube containing1 ml of TE buffer (Tris 1 M, EDTA 0.5 M, pH 8). Genomic DNAs(gDNAs) of the single cells were extracted by Proteinase K(20 mg ml�1) at 55 8C for 30 min, followed by 10 min at 95 8C.The gDNAs was then used as template for gene amplification.

Gene amplification was performed using the universal primerpair of D1R (50-ACC CGC TGA ATT TAA GCA TA-30) and D3Ca (50-ACG AAC GAT TGC ACG TCA G-30), targeting the large subunitribosomal RNA gene (LSU rDNA, domain 1–3) (Scholin et al., 1993),and the primer pair ITS1F (50-TCG TAA CAA GGT TTC CGT AGG TG-30) and ITS1R (50–-TA TGC TTA AGT TCA GCG GG-30), targeting theinternal transcribed spacer region (ITS) of rDNA (Leaw et al., 2001).Amplification was performed using an ArktikTM Thermal Cycler(Thermo Scientific, Vantaa, Finland), with the following cyclingconditions: pre-denaturation at 94 8C for 4 min, 35 cycles of 94 8Cfor 35 s, 55 8C for 50 s and 72 8C for 35 s. A final extension wascarried out at 72 8C for 7 min.

The amplicons were then separated by 1% agarose gelelectrophoresis, stained with SYBR Safe DNA Stain (Invitrogen,Life Technologies, Carlsbad, CA, USA), and visualized under a bluetransilluminator (Invitrogen). The amplicons were further purifiedusing the Wizard1 SV Gel PCR Clean-Up System (Promega,Madison, WI, USA) prior to sequencing. The purified ampliconswere directly sequenced for both strands by First Base Laboratories(Selangor, Malaysia). The nucleotide sequences obtained weredeposited in NCBI GenBank (Table 1).

2.5. Sequence alignment and phylogenetic analyses

Nucleotide sequences of LSU rDNA obtained in this study, andsequences of other Karlodinium retrieved from the NCBI nucleotidedatabase, were aligned with MUSCLE (Edgar, 2004) (Table 1). Theresulting alignment obtained from MUSCLE was further editedusing BioEdit Sequence Alignment Editor v7.2.5 (Hall, 1999).

The alignment comprised 23 operational taxonomic units(OTUs), with two outgroup taxa: Takayama tuberculata de Salasand T. acrotrocha (Larsen) de Salas, Bolch & Hallegraeff. Outgroupselection was based on de Salas et al. (2008). Phylogeneticanalyses of maximum parsimony (MP) and maximum likelihood(ML) were performed using PAUP* 4b10 (Swofford, 2011).Bayesian inference (BI) was performed using MrBayes v 3.2.1(Huelsenback and Ronquist, 2001), as described in Lim et al.(2013). The estimates of evolutionary divergence between LSUrDNA sequences of Karlodinium spp. were conducted in MEGA6.06, through pairwise distance analysis using the p-distancemodel (Tamura et al., 2013).

2.6. ITS2 transcript modelling

The proximal stems of 5.8S–28S rRNA interaction weredetermined to annotate the termini of ITS2 in Karlodinium species(Gottschling and Plotner, 2004; Keller et al., 2009). The ITS2secondary structure of K. australe (KJ670418) was predicted usingthe free energy minimization in RNAStructure v. 5.02 (Reuter andMathews, 2010). The ITS2 secondary structure terminology wasfurther confirmed based on Muller et al. (2007). The sequence-structure modelled was used as a template for homologymodelling (Wolf et al., 2005), using the ITS2 Database (Koetschanet al., 2012; Merget et al., 2012). We modelled four sequences of K.

australe isolated from West Johor Strait, one from Aman Island,three of K. armiger from Spain, and 35 of K. veneficum from bothPacific and Atlantic regions (Table 1).

The multiple sequence-structure alignment of Karlodinium ITS2was generated using an ITS sequence structure-specific scoringmatrix (Seibel et al., 2006) in 4SALE v. 1.7 (Seibel et al., 2006, 2008).The compensatory base changes (CBCs) and hemi-compensatorybase changes (HCBCs) were identified using the CBCAnalyzeroption implemented in 4SALE (Wolf et al., 2005; Seibel et al., 2006,2008). The ITS2 secondary structure was illustrated using VARNA(Darty et al., 2009).

3. Results

3.1. Fish kills in Tanjung Kupang cage cultures, West Johor Strait

The cage-fish cultures are located along the coasts of TanjungKupang, Johor, the West Johor Strait, Malaysia. Nine cages wereoperated at the area. Most of the affected cages were situated atclose quarters between Stn 3 and Stn 4 (Fig. 1). The affectedcultured fish included estuary cod (Epinephelus coioides Hamilton),barramundi (Lates calcarifer Bloch), paddletail snapper (Lutjanus

gibbus Forsskal) and fourfinger threadfin (Eleutheronema tetra-

dactylum Shaw). Fish kills started on February 11, 2014, becomingmore serious during the subsequent three days, and continued forabout a week (Fig. 2A–D). The net cages were unable to be movedduring the incident and no aeration system was available. Aboutfour tonnes of fish died during the incident. Other than the cage-cultured fishes, some small wild fishes in the surrounding areawere also affected; their carcasses were found floating andwashing away from the coastal area.

The dead fishes observed during the incident showed somesymptoms similar to those of other fishes exposed to high levels ofkarlotoxins (see Mooney et al., 2010). These included protrudingeyes and reddening of the iris, discoloured and sloughing skin,reddening at the base of fins, and damaged gills showing abrownish cast (Fig. 2E–I). Human skin irritation, swelling, rashesand itchiness were also observed (Fig. 2J).

3.2. Phytoplankton composition and its related water parameters

A total of 33 taxa were identified from the area: 26 taxaof Bacillariophyceae (19 centric and 7 pennate diatoms) and sevenof Dinophyceae. Karlodinium sp., a minute unarmored dinoflagel-late, dominated at most of the sampling stations. The averagecell density of Karlodinium sp. was 1.25 � 106 cells l�1, withrelative abundances of 68.8–98.6% (Fig. 3A). Cell densitiesincreased inwardly along the West Johor Strait; the lowest celldensity was observed at Stn 1 (the outermost station;0.31 � 106 cells l�1), and the highest at Stn 6 (the innermoststation; 2.34 � 106 cells l�1) (Fig. 3B). Other dinoflagellates (Cer-

atium, Prorocentrum, Protoperidinium, Gyrodinium, Ostreopsis) wereobserved in the samples, but comprised only a minority of theoverall phytoplankton composition (<0.2%). Cell densities of these

Table 1LSU rDNA sequences of Karlodinium species used in this study. Isolates in bold face indicate single cells isolated from natural field samples. Asterisks indicate type specimens.

Species Strain/isolate Locality Date of collection Accession

LSU ITS

Karlodinium antarcticum KDANSO10* Southern Ocean 508330 S, 1458350 E Mar, 2006 EF469234

Karlodinium armiger K-0668* Alfacs Bay, Spain 2000 DQ114467

GC-3 IRTA Ebro River Delta, Alfacs Bay, Spain 2000 AM184205

GC-2 IRTA Ebro River Delta, Alfacs Bay, Spain 2000 AM184204

GC-7 IRTA Alfacs Bay, Ebro Delta, Spain 2000 AJ557024

Karlodinium australe F2-05LSU Tanjung Kupang, West Johor Strait, Malaysia Feb 23, 2014 KJ670422F2-06LSU Tanjung Kupang, West Johor Strait, Malaysia Feb 23, 2014 KJ670423F2-07LSU Tanjung Kupang, West Johor Strait, Malaysia Feb 23, 2014 KJ670424F2-08LSU Tanjung Kupang, West Johor Strait, Malaysia Feb 23, 2014 KJ670425F2-09LSU Tanjung Kupang, West Johor Strait, Malaysia Feb 23, 2014 KJ670426F201ITS Tanjung Kupang, West Johor Strait, Malaysia 2014 KJ670418F202ITS Tanjung Kupang, West Johor Strait, Malaysia 2014 KJ670419F203ITS Tanjung Kupang, West Johor Strait, Malaysia 2014 KJ670420F204ITS Tanjung Kupang, West Johor Strait, Malaysia 2014 KJ670421KaPA Aman Island, Malacca Strait, Malaysia 2013 KM975627

GT5 Singapore – DQ156228

KDAGT03* Grants Lagoon, Tasmania, Australia Apr 15, 2002 DQ151559

KDATL11 Tuggerah Lakes, New South Wales, Australia Mar, 2006 DQ151560

Karlodinium ballantinum KDBMP01* Mercury Passage, Tasmania, Australia Mar, 2006 EF469232

MC701-B1 Gulf of Naples, Italy Feb 14, 2006 FJ024699

Karlodinium conicum KDCSO15* Southern Ocean 448410 S, 1478070 E Mar, 2006 EF469231

Karlodinium corrugatum KDGSO08* Southern Ocean 538220 S, 1448530 E Mar, 2006 EF469233

Karlodinium decipiens Bilbao Abra de Bilbao, Spain – EF469237

KDDSB01* Spring Bay, Tasmania, Australia Mar, 2006 EF469236

KDDSO10 Southern Ocean 508330 S, 1458350 E Mar, 2006 EF469235

Karlodinium veneficum CCMP415 – – AJ557026

CG-1 CSIC-1 Alfacs Bay, Ebro Delta, Spain 1998 AJ534656

GgaITSC36 North Carolina, Neuse River, USA 1997 AF352367

GgaITSC35 North Carolina, Neuse River, USA 1997 AF352366

GgaITSC34 North Carolina, Neuse River, USA 1997 AF352366

GC-5 IRTA Alfacs Bay, Ebro Delta, Spain 2000 AJ557028

GC-8 IRTA Alfacs Bay, Ebro Delta, Spain 2000 AJ557027

ICMB283 Ebro River Delta, Spain – JF906097

ICMB282 Ebro River Delta, Spain – JF906096

ICMB280 Ebro River Delta, Spain – JF906095

ICMB279 Ebro River Delta, Spain – JF906094

ICMB278 Ebro River Delta, Spain – JF906093

ICMB277 Ebro River Delta, Spain – JF906092

ICMB276 Ebro River Delta, Spain – JF906091

ICMB275 Ebro River Delta, Spain – JF906090

ICMB274 Ebro River Delta, Spain – JF906089

ICMB273 Ebro River Delta, Spain – JF906088

ICMB271 Ebro River Delta, Spain – JF906087

ICMB269 Ebro River Delta, Spain – JF906085

ICMB268 Ebro River Delta, Spain – JF906084

ICMB267 Ebro River Delta, Spain – JF906083

ICMB266 Ebro River Delta, Spain – JF906082

ICMB265 Ebro River Delta, Spain – JF906081

ICMB264 Ebro River Delta, Spain – JF906080

ICMB263 Ebro River Delta, Spain – JF906079

ICMB261 Ebro River Delta, Spain – JF906077

ICMB260 Ebro River Delta, Spain – JF906076

ICMB259 Ebro River Delta, Spain – JF906075

ICMB258 Ebro River Delta, Spain – JF906074

K-0522 Oslofjord, Norway 1977 AF200675

KDMPT01 The Swan River, Perth, Western Australia Mar 11, 2001 AY263964

KM1 CSIC-1 – – AJ557025

LAMB010601 German coastal water, North Sea – HQ434333

LAMB090611 East China Sea – HQ434334

LJ-2011-1 Sanggou Bay, China – JN986578

NMBjah047 – – DQ459434

Pim05JulC4 Florida, USA – AY245692

Plymouth 103 Devonport, UK 1950 DQ114466

Takayama tuberculata TTBSO11.1* The Southern Ocean, Australia Mar, 2006 EF469230

Takayama acrotrocha GT7 Singapore – DQ656115

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6254

other dinoflagellates were relatively low (<4 � 103 cells l�1).Ostreopsis sp. was found only at Stn 2 (a land reclamation areawith dredging activity) and Stn 6 (the marina, Puteri Harbour).

Water depths at the stations ranged from 6–30 m (Fig. 4A). Thedistribution of two basic physical properties (sea-surface temper-ature and pH) in the vicinity of the sampling stations was rather

homogenous, with temperatures ranging from 27.2–27.5 8C(Fig. 4B) and pH of �8 (Fig. 4C). However, salinity was lower inthe inner channel of the West Johor Strait (<30, Stn 4–6) comparedto the outer stations (Fig. 4D). A low water visibility (<0.5 m) wasobserved, compared to a normal visibility of �1 m. The waterappeared to be stagnant.

Fig. 2. (A–D) Mass mortality of cage-cultured fishes. (E) Skin sloughing. (F) Fin reddening at the base of the fin (arrow). (G–H) Brownish gills (arrows). (I) Eye with reddening of

the iris. (J) Rashes and swelling on human skin.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–62 55

3.3. Species description

Samples collected in the area were comprised mainly of a singlespecies of an unarmored dinoflagellate. Detailed morphologicalobservations and comparisons showed that the species resembledKarlodinium australe, as described in de Salas et al. (2005).

3.3.1. Morphology

Cells are ovoid (Fig. 5), with a mean length of 22.53 � 2.15 mmand a mean width of 16.16 � 1.86 mm (n = 30); the meanlength:width ratio is 1.40 � 0.11 mm. Cells are lightly pigmented,with 7–19 (n = 10) ribbon-shape chloroplasts observed along the cellperiphery, distributed in both epi- and hypocone (Fig. 5A–B). The cellnucleus is at the anterior of the cell (Fig. 5A and C). The epicone isslightly smaller and conical compared to the round hypocone (Fig. 5C).

Under SEM, cells are ovoid without a dorsal-ventral compres-sion (Fig. 5D–G). A ventral pore is present to the left of the apicalgroove, and slightly above the sulcal intrusion. A tube-shapedpeduncle-like structure is present, lying along the sulcus (Fig. 5D–E). An apical groove descends 1/3 the length, to the dorsal side ofthe epicone (Fig. 5F). The deeply excavated premedian cinguluminvades into the epicone and descends to the right, and is abouttwo cingulum widths (1/3 the total cell length). A smooth, curlytransverse flagellum is observed in the cingulum (Fig. 5G). Theamphiesma (alveola) is polygonal (mostly pentagonal-hexagonal),with some rounded trychocysts observed (Fig. 5H).

3.3.2. Diagnosis

Cells are larger compared to Karlodinium antarcticum, K.

armiger, K. ballantinum, K. corrugatum and K. vitiligo (Bergholtz

et al., 2005; de Salas et al., 2008), but are in the range of thoseobserved in the type species, K. australe (de Salas et al., 2005). Thecingulum displacement was 1/3 the total cell length, whichresembles that of K. corsicum and K. decipiens (de Salas et al., 2008;Siano et al., 2009), and is slightly larger than the type specimens(25% of total cell length; de Salas et al., 2005). Generally, thenumber of chloroplasts is larger (7–19) than the type species (6–10; de Salas et al., 2005).

3.3.3. Distribution

Karlodinium australe is documented for the first time inMalaysian coastal waters, in samples collected along the WestJohor Strait. Concurrently, K. australe was also found in AmanIsland (Penang), the Straits of Malacca. The species was previouslyreported from Australia (New South Wales [Tuggerah Lake];Victoria [Port Phillip Bay]; South Australia [Port Lincoln]; Tasmania[Grants Lagoon, Moulting Bay]) (de Salas et al., 2005) andSingapore (de Salas et al., 2008).

3.4. Molecular dataset and phylogenetic inferences

The LSU rDNA D1–D3 nucleotide sequences of Karlodinium

species were aligned and yielded 900 characters (includinggaps), of which 729 were constant and 129 were parsimonyinformative. The best substitution and rate heterogeneitymodels of TIM1 + I + G were selected for ML (parameter valuesset: A = 0.2423, C = 0.1957, G = 0.2851, T = 0.2769; rate matrixof A–C = 1.0000, A–G = 0.8872, A–T = 0.6048, C–G = 0.6048,C–T = 5.3201, G–T = 1.0000, with the proportion of invariable sitesestimated at 0.5110 and a gamma shape parameter of 0.6290). For

Fig. 3. (A) Phytoplankton relative abundances at the six stations. (B) Cell densities of Karlodinium along the sampling stations.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6256

BI, the best model selected with Bayesian Information Criterion(BIC) was TrN + I + G (parameter values set: A = 0.2408, C = 0.1948,G = 0.2865, T = 0.2779; rate matrix of A–C = 1.0000, A–G = 1.1010,A–T = 1.0000, C–G = 1.0000, C–T = 6.6521, G–T = 1.0000; propor-tion of invariable sites estimated at 0.5070 and a gamma shapeparameter of 0.6110).

The LSU rDNA phylogenetic inferences based on MP, ML and BIyielded identical tree topologies. Only the ML tree is shown here,with MP, ML bootstraps supports and BI clade credibility valuespresented (Fig. 6). The phylogenetic trees consistently revealed ahighly support group consisting of the K. australe isolates from theWest Johor Strait, one strain of K. australe from Aman Island(Malacca Strait), one strain from Singapore, and two strains fromAustralia (including the type specimens, KDAGT03) (MP/ML/BI = 100%). Its sister taxon, K. armiger, formed the basal of theclade of K. australe, with moderately strong bootstrap supports(MP/ML/BI: 75/76/99).

The LSU rDNA sequence divergence of Karlodinium australe andK. armiger strains ranged from 3.6–4.0%; and were �6.5% whencompared to other species. The lowest sequence divergenceobserved was between K. antarcticum de Salas and K. decipiens

de Salas and Laza-Martınez, by a sequence divergence of 0.8–0.9%.The intraspecific LSU rDNA sequence divergence of the eightK. australe strains ranged from 0–0.3%, comparable to that inK. decipiens (0.1–0.2%), while lower than those in K. ballantinum

de Salas (0.6%) and K. veneficum (0.3–1.1%) (SupplementaryMaterial S1).

3.5. ITS2 transcript

The secondary structure of the ITS2 transcript of Karlodinium

species was determined based on the 5.8S–28S LSU interactions(Keller et al., 2009) (Fig. 7). The structure showed typically fourconserved helices (I–IV). The average length of ITS2 in K. australe

Fig. 4. Basic physical water parameters in the West Johor Strait. Contour plots of depth (A), temperature (B), pH (C), and salinity (D) at the six stations.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–62 57

was 211 bp. Secondary structure comparisons of ITS2 from threeKarlodinium species (K. australe, K. armiger Bergholtz, Daugbjergand Moestrup and K. veneficum) revealed high structural conser-vation. The universal motif of pyrimidine–pyrimidine (U–U)mismatch was found in helix II (Fig. 7).

Pairwise comparison of the consensus ITS2 transcript ofKarlodinium australe to its sister taxa, K. armiger, showed thepresence of four HCBCs: two in helix I (both U–G $ C–G), one ofeach in helix III (G–C $ G–U) and helix IV (G–U $ G–C), and16 SNPs were found. No CBCs were observed between the twospecies. When compared to the type species, K. veneficum, threeCBCs were identified: one situated in helix III (U–A $ C–G) and two

in helix IV (both G–C $ U–A); seven HCBCs, one in helix I (U–G $ C–G), one in helix II (G–C $ G–U), three in helix III (two G–C $ G–U and one C–G $ U–G), two in helix IV (A–U $ G–U and G–U $ A–U), and 23 SNPs were found.

4. Discussion

4.1. Karlodinium bloom in the West Johor Strait

On February 11, 2014, a mass mortality of �50,000 caged fishes,involving over four species, occurred along the West Johor Strait,Malaysia. A phytoplankton bloom, with patches of water

Fig. 5. Karlodinium australe from the West Johor Strait. (A–C) LM. (A) Cell showing numerous peripheral chloroplasts and dorsal position of nucleus (n). (B) Auto-fluorescent

peripheral chloroplasts. (C) Cells showing cell outline and nuclei in the epicone (n). (D–H) SEM. (D) Ventral view of cell showing ventral pore (arrow) and a tube-shaped

peduncle-like structure in the sulcus (arrowhead). (E) Ventral view of cell; note the straight apical groove (arrow). (F) Dorsal view of cell showing the extent of straight apical

groove (arrow). (G) Cell with flagella. (H) Close-up of alveolae and trichocysts (arrow). Scale bars: 10 mm (A–C); 2 mm (D–G); 1 mm (H).

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6258

discoloration, had been witnessed by the farm operators a weekprior to the incident (February 6). It is postulated that theKarlodinium australe cell densities capable of causing the fish killshad been much higher than our observations, because our sampleswere collected two weeks after the incident (February 20–21),partly due to poor information dissemination. In spite of that, ourresults showed that the phytoplankton composition was dominat-ed by K. australe (70–100%), with a mean cell density of1.25 � 106 cells l�1. The bloom concentration of K. veneficum

associated with previous fish kills was �6 � 107 cells l�1 (Deedset al., 2002), and it is capable of forming intense blooms of up to108 cells l�1 (Place et al., 2012).

Several studies have recognized that, in lieu of regular seasonalblooms, the timing of some dinoflagellates blooms is unpredictable(Trainer et al., 2010; Smayda and Trainer, 2010). We theorize thatthere is a potential link between the outbreak of Karlodinium

australe blooms and the influx of nutrients into the WestJohor Strait. This phenomenal, unprecedented bloom was likely

Kar

lodi

nium

aus

trale

GT5 (Singapore)

F2-09LSU (WJS, Malaysia)

F2-08LSU (WJS, Malaysia)

F2-07LSU (WJS, Malaysia)

F2-06LSU (WJS, Malaysia)

F2-05LSU (WJS, Malaysia)

KDAGT03* (Australia)

KDATL11 (Australia)

Karlodinium armiger K-0668*

Plymouth103

K-0522

Karlodinium veneficum KDMPT01

Karlodinium conicum KDCSO15

KDBMP01*

MC701-B1

Karlodinium corrugatum KDGSO08*

KDDSB01*

KDDSO10

Bilbao

Karlodinium antarcticum KDANSO10*

Takayama tuberculatum TTBSO1 1.1*

Takayama acrotrocha GT7

Karlodinium veneficum

Karlodinium decipiens

Karlodinium ballantinum

64/62/85

97/96/100

65/63/60

74/77/100

75/74/95

0.1

KaPa (Malacca Strait, Malaysia)

75/76/99

90/88/100

Fig. 6. Phylogenetic inference of Karlodinium species based on the large subunit rDNA sequences. The tree is rooted using Takayama tuberculata and T. acrotrocha. Bootstraps

value �50% are shown in the order: maximum parsimony (MP)/maximum likelihood (ML)/Bayesian (BI). Branches marked in thick line indicated bootstraps values of 100%

for MP/ML/BI. Taxa in boldface indicate isolates from this study.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–62 59

triggered by anthropogenic nutrients from nearby areas, includingland-based and riverine inputs and disturbances caused by ongoingcoastal reclamation and maintenance dredging activities (near Stn2, Fig. 1), which might affect the nutrient dynamics in the strait.These anthropogenic activities may contribute to excess nutrientenrichment that favours the phytoplankton directly, as well asstimulating the growth of its algal prey (cf. Adolf et al., 2008, 2009).The species is known as a mixotroph, consuming cyptophyte prey(de Salas et al., 2005). It is not surprising that the K. veneficum bloommay be promoted by the abundance of the cryptophyte prey, asdemonstrated by several studies on a closely related species, K.veneficum (summarized in Burkholder et al., 2008).

During the outbreak, the area experienced a three-day neaptide, when the water remained stagnant, with very little waterexchange. This condition was envisaged to initiate the bloom. Thehighest concentration of Karlodinium australe cells was observed inthe innermost station (Stn 6), where a marina is situated. This is inagreement with Garces et al. (2006), who also showed that cellstend to concentrate in harbours, i.e. areas where waters are lessturbulence. At the same time, water currents in the strait moveinwards (Goyal and Rathod, 2011), which might also carry the cellpopulation towards Stn 6.

The basic necropsy findings presented here link the toxicKarlodinium australe cells to the acute fish kills observed in the WestJohor Strait. It is well known that some Karlodinium species kill fish,but not all species are capable of producing karlotoxins. Among the

ten known species, only K. armiger, K. conicum de Salas and K.

veneficum were found to be ichthyotoxic (Garces et al., 2006;Bachvaroff et al., 2008, 2009; Mooney et al., 2009). The ability of K.australe to produce toxins has not yet been elucidated, but thenecropsy evidence of dead fish examined in this study showsimilarities to those reported in naturally occurring or experimentalfish kills due to karlotoxins (Deeds et al., 2002, 2006; Kempton et al.,2002). Karlotoxins primarily damage the fish gill epithelia (Deedset al., 2006; Place et al., 2012), and this was observed in our study.

Other than causing fish mortality, one of the congeners, KmTx2,was found to be toxic to mammalian epithelial cells, neurons,fibroblasts, cardiac myocytes and lymphocytes (Deeds et al.,2004b). The toxins have never been associated with human skinirritation; however, we eye-witnessed this during the samplingsurvey (Fig. 2J). Dermatological problems such as swollen skin andrashes were developed after handling the dead fishes and planktonnet samples. Analyses of the ichthyotoxicity and dermal toxicity ofKarlodinium australe cells are thus pivotal and urgently needed.

4.2. Identity of the bloom species responsible for the fish kills in the

West Johor Strait

The systematics of unarmored dinoflagellates have undergoneseveral taxonomic modifications, and several species weretransferred to the genus Karlodinium Larsen. The erection of thegenus was based primarily on morphological and ultrastructure

U

A

G

C

C

UG

CU

UC

AGU

G

U

C

UAC

UG

AU

UCCCAUGCUCCU

GA

CA

UAUGCCAAGUGUUU

UC G C G C U U G G U G

C C GU

CA

U G G G UG C G U

CUC

UG

UG

UG

UC

AA

GG

AG

CCU C

GGCC

CU

UG

AC

GC

AU U

UA

GU G C A C

AGGUA

GU

GU

CUG

C

A

A

CGAG

CA

ACUU

AAUAAG

CA

CCU

U GUGUGU C U

CUUU

CGUUGC

ACA

G

U

U

G

CU

G

AUCC

AC

GCCUG

CCCCACAC A G C G C G C

AGCGCGUUG

CG

CUUUUCUU

UC

AA

G

A

C

AUGAAGUUA

G

G

UCAG

CAAA

C

1

10

20

30

4050

60

7080

90

100

110

120

130

140

150

160

170

180

190

200

210

220

230

240

250

260

UU

C

G

G

C

U

UG

U

G

U

C U

G

CG

U

UC

C U

UG

GU

GU

U

C

U

G

U

A

U

U

A

C

U

CA

C

C

G

U

A

C

C

U

AC

C

G

C

U

ITS2

5.8S 28S

I

II

III

IV

Karlodinium australe vs K. armiger & K. veneficumKarlodinium australe vs K. armigerKarlodinium australe vs K. veneficum

Fig. 7. Pair-wise consensus comparison of ITS2 transcript based on 42 sequence-structures of Karlodinium. Shaded rectangles indicate compensatory base changes (CBCs) and

unfilled-rectangles indicate hemi-compensatory base changes (HCBCs). Circles represent single nucleotide polymorphisms.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6260

reassessments, particularly on the character of the complex cellcovering, the amphiesma; these are strongly supported by thephylogenetic data and pigment characterization (Daugbjerg et al.,2000). Prior to 2005, there were four known Karlodinium species,viz. K. armiger, K. australe, K. veneficum and K. vitiligo (Ballantine)Larsen (Bergholtz et al., 2005; de Salas et al., 2005). At the same

time, Karlodinium micrum (Leadbeater and Dodge) Larsen wassynonymized as K. veneficum (Bergholtz et al., 2005), based on thelack of morphological difference and the low genetic divergence ofthe LSU rDNA marker. Five more Karlodinium species were laterdescribed, i.e. K. antarcticum, K. ballantinum, K. conicum, K.corrugatum de Salas and K. decipiens (de Salas et al., 2008). The

H.C. Lim et al. / Harmful Algae 40 (2014) 51–62 61

most recent, Gyrodinium corsicum Paulmier (Paulmier et al., 1995),was revised to K. corsicum (Paulmier) Siano and Zingone (Sianoet al., 2009); hence, the genus comprises ten species.

The cells collected during this study were ovoid and small. SEMobservation revealed typical Karlodinium characteristics, such as astraight apical groove and the presence of a ventral pore. Based onthe outer morphology and the chloroplast numbers, we designatedthe species as K. australe. Precise identification of species in thisgenus has been difficult due to their morphological plasticity.Bergholtz et al. (2005) claimed that sample preparation prior toelectron microscopy might produce artefacts; morphologicalcharacteristics such as amphiesma vesicles tend to deform duringthe fixation process. de Salas et al. (2008) also suspected thatculturing artefacts might distort the morphology of these delicatedinoflagellates. Hence, a total-evidence approach sensu McManusand Katz (2009), applying both detailed morphological assessmentand molecular evidence, was used to support the species identityof this tiny dinoflagellate. In this study, we adopted the single-cellPCR technique, applied to field samples, and found efficientamplification of the targeted genetic markers (LSU and ITS rDNA),with nucleotide sequences successfully obtained for subsequentmolecular characterizations. The sensitivity of PCR allows theamplification of minute amounts of nucleic acid from a single cell.The single-cell PCR technique is useful when the sample amount islimited, and is a powerful molecular technique for diagnosis ordetection at the single-cell level (e.g. Bolch, 2001; Ki et al., 2005;Kai et al., 2006; Auinger et al., 2008).

The phylogeny of Karlodinium based on LSU rDNA was wellresolved, and the phylogenetic position of K. australe is inaccordance with that of Bergholtz et al. (2005) and de Salaset al. (2008). The species is genetically distinct from otherKarlodinium species based on its sequence divergence, forming asister group with K. armiger in our phylogenetic analyses.

In addition, we applied another biological species marker, thesecondary structure of ITS2 of the nuclear-encoded ribosomal RNAgene, to confirm the species identity. The CBCs in the secondarystructure of ITS2 have proven useful for delimiting biologicalspecies (e.g. Coleman, 2003; Muller et al., 2007). As demonstratedby Muller et al. (2007), the presence of at least one CBC is a goodindicator that two organisms belong to distinct species, while theabsence of CBCs is not necessarily an indicator that they belong tothe same species. The species in this study is readily distinguishedfrom the toxic Karlodinium veneficum, with a difference of threeCBCs. It is noteworthy that pairwise comparisons of CBCs in theITS2 transcripts of K. australe in this study, and its sister species K.armiger, showed four HCBCs, although no CBCs were found. Itshould be noted that when the structural features are comparedamong the Karlodinium ITS2 transcripts, helix IV showed the mostvariable base-pairing interaction (CBCs and HCBCs) among species.

The identification of Karlodinium australe in this study wassupported not only by morphological evidence, but also bymolecular data. This is the first report of K. australe in Malaysia,and the first to associate it with a fish-kill event in the world. Thisichthyotoxic species warrants a thorough long-term study, as itthreatens not only the economy but also the welfare of localfishermen and breeders.

Acknowledgements

The authors are grateful to Dr. Stephen S. Bates (Fisheries andOceans Canada) for his reviewed comments and English revision.This work was supported by the Malaysian Government throughthe Ministry of Education under KPM–JSPS COMSEA MatchingFund (GA002-2014) and the Higher Institution Centre of ExcellenceResearch Grant; and the University of Malaya Research Grants(RU006-2014, BK013-2014).[TS]

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.hal.2014.10.005.

References

Abbott, B.C., Ballantine, D., 1957. The toxin from Gymnodinium veneficum Ballantine.J. Mar. Biol. Assoc. U.K. 36, 169–189.

Adolf, J.E., Bachvaroff, T.R., Krupatkina, D.N., Nonogaki, H., Brown, P.J.P., Lewitus, A.J.,Harvey, H.R., Place, A.R., 2006a. Species specificity and potential roles ofKarlodinium micrum toxin. Afr. J. Mar. Sci. 28, 415–421.

Adolf, J.E., Stoecker, D.K., Harding, L.W., 2006b. The balance of autotrophy andheterotrophy during mixotrophic growth of Karlodinium micrum (Dinophy-ceae). J. Plank. Res. 28, 737–751.

Adolf, J.E., Bachvaroff, T., Place, A.R., 2008. Does cryptophyte abundance trigger toxicKarlodinium veneficum blooms in eutrophic environments? Harmful Algae 8,119–128.

Adolf, J.E., Bachvaroff, T.R., Place, A.R., 2009. Environmental modulation of karlo-toxin levels in strains of the cosmopolitan dinoflagellate, Karlodinium veneficum(Dinophyceae). J. Phycol. 45, 176–192.

Anton, A., Teoh, P.L., Mohd-Shaleh, S.R., Mohammad-Noor, N., 2008. First occurrenceof Cochlodinium blooms in Sabah, Malaysia. Harmful Algae 7, 331–336.

Auinger, B.M., Pfandl, K., Boenigk, J., 2008. Improved methodology for identificationof protists and microalgae from plankton samples preserved in Lugol’s iodinesolution: combining microscopic analysis with single-cell PCR. Appl. Environ.Microbiol. 74, 2505–2510.

Bachvaroff, T.R., Adolf, J.E., Squier, A.H., Harvey, H.R., Place, A.R., 2008. Characteri-zation and quantification of karlotoxins by liquid chromatography–mass spec-trometry. Harmful Algae 7, 473–484.

Bachvaroff, T.R., Adolf, J.E., Place, A.R., 2009. Strain variation in Karlodinium vene-ficum (Dinophyceae): toxin profiles, pigments, and growth characteristics. J.Phycol. 45, 137–153.

Ballantine, D., 1956. Two new marine species of Gymnodinium isolated from thePlymouth area. J. Mar. Biol. Assoc. U.K. 35, 467–474.

Bergholtz, T., Daugbjerg, N., Moestrup, Ø., Fernandez-Tejedor, M., 2005. On theidentity of Karlodinium veneficum and description of Karlodinium armiger sp.nov, (Dinophyceae), based on light and electron microscopy, nuclear-encodedLSU rDNA, and pigment composition. J. Phycol. 42, 170–193.

Braarud, T., 1957. A red tide organism from Walvis Bay. Galathea Rep. 1, 137–138.Bolch, C.J.S., 2001. PCR protocols for genetic identification of dinoflagellates directly

from single cysts and plankton cells. Phycologia 40, 162–167.Burkholder, J.M., Glibert, P.M., Skelton, H.M., 2008. Mixotrophy, a major mode

of nutrition for harmful algal species in eutrophic waters. Harmful Algae 8,77–93.

Coleman, A.W., 2003. ITS2 is a double-edged tool for eukaryote evolutionarycomparisons. Trends Genet. 19, 370–375.

Darty, K., Denise, A., Ponty, Y., 2009. VARNA: interactive drawing and editing of theRNA secondary structure. Bioinformatics 25, 1974–1975.

Daugbjerg, N., Hansen, G., Larsen, J., Moestrup, Ø., 2000. Phylogeny of some of themajor genera of dinoflagellates based on ultrastructure and partial LSU rDNAsequence data, including the erection of three new genera of unarmoureddinoflagellates. Phycologia 39, 302–317.

de Salas, M.F., Bolch, C.J.S., Hallegraeff, G.M., 2005. Karlodinium australe sp. nov.(Gymnodiniales, Dinophyceae), a new potentially ichthyotoxic unarmoreddinoflagellate from lagoonal habitats of south-eastern Australia. Phycologia44, 640–650.

de Salas, M.F., Laza-Martınez, A., Hallegraeff, G.M., 2008. Novel unarmored dino-flagellates from the toxigenic family Kareniaceae (Gymnodiniales): five newspecies of Karlodinium and one new Takayama from the Australian sector of theSouthern Ocean. J. Phycol. 44, 241–257.

Deeds, J.R., 2003. Toxins and Toxicity From the Cosmopolitan Bloom-formingDinoflagellate Karlodinium micrum. University of Maryland, College Park, pp.259 (Ph.D. dissertation).

Deeds, J.R., Terlizzi, D.E., Adolf, J.E., Stoecker, D.K., Place, A.R., 2002. Toxic activityfrom cultures of Karlodinium micrum (=Gyrodinium galatheanum) (Dinophyceae)– a dinoflagellate associated with fish mortalities in an estuarine aquaculturefacility. Harmful Algae 1, 169–189.

Deeds, J.R., Kibler, S.R., Tester, P.A., Place, A.R., 2004a. Geographical strain variationin toxin production in Karlodinium micrum (Dinophyceae) from southeasternestuaries of the United States. In: Steidinger, K.A., Landsberg, J.H., Tomas, C.R.,Vargo, G.A. (Eds.), Harmful Algae 2002. Florida and Wildlife ConservationCommission, Florida Institute of Oceanography, and Intergovernmental Ocean-ographic Commission of UNESCO, Paris, pp. 145–147.

Deeds, J.R., Kao, J.P.Y., Hoesch, R.E., Place, A.R., 2004b. Toxic mode of action of KmTx2, a new fish-killing toxin from Karlodinium micrum (Dinophyceae). In: SecondSymposium on Harmful Marine Algae in the United States, Woods Hole, MA, pp.185–268.

Deeds, J.R., Place, A.R., 2006. Sterol specific membrane interactions with the toxinsfrom Karlodinium micrum (Dinophyceae) – a strategy for self-protection? Afr. J.Mar. Sci. 28, 421–427.

Deeds, J.R., Reimschuessel, R., Place, A.R., 2006. Histopathological effects infish exposed to the toxins from Karlodinium micrum. J. Aquat. Anim. Health18, 136–148.

H.C. Lim et al. / Harmful Algae 40 (2014) 51–6262

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy andhigh throughput. Nucleic Acids Res. 32, 1792–1797.

FAO, 2007. FishStat Plus – Universal Software for Fishery Statistical Time Series.Fisheries and Aquaculture Department, Food and Agriculture Organization ofthe United Nations, Rome, Italy.

Fensin, E.E., 2004. Occurrence and ecology of the dinofagellate Karlodinium micrumin estuaries of North Carolina, USA. Harmful Algae 2002 In: Steidinger, K.A.,Landsberg, J.H., Tomas, C.R., Vargo, G.A. (Eds.), Florida Fish and Wildlife Con-servation Commission. Florida Institute of Oceanography, and Intergovernmen-tal Oceanographic Commission of UNESCO, St Petersburg, pp. 62–64.

Galimany, E., Place, A.R., Ramon, M., Jutson, M., Pipe, R.K., 2008. The effects offeeding Karlodinium veneficum (PLY # 103; Gymnodinium veneficum Ballantine)to the blue mussel Mytilus edulis. Harmful Algae 7, 91–98.

Garces, E., Fernandez, M., Penna, A., Van Lenning, K., Gutierrez, A., Camp, J., Zapata,M., 2006. Characterization of NW Mediterranean Karlodinium spp., (Dinophy-ceae) strains using morphological, molecular, chemical, and physiologicalmethodologies. J. Phycol. 42, 1096–1112.

Goshorn, D., Deeds, J., Tango, P., Poukish, C., Place, A.R., McGinty, M., Butler, W.,Luckett, C., Magnien, R., 2004. Occurrence of Karlodinium micrum and itsassociation with fish kills in Maryland estuaries. In: Steidinger, K.A., Landsberg,J.H., Tomas, C.R., Vargo, G.A. (Eds.), Harmful Algae 2002. Florida and WildlifeConservation Commission, Florida Institute of Oceanography, and Intergovern-mental Oceanographic Commission of UNESCO, Paris, pp. 361–363.

Gottschling, M., Plotner, J., 2004. Secondary structure models of the nuclear internaltranscribed spacer regions and 5.8S rRNA in Calciodinelloideae (Peridiniaceae)and other dinoflagellates. Nucleic Acids Res. 32, 307–315.

Goyal, R., Rathod, P., 2011. Hydrodynamic modelling for salinity of SingaporeStrait and Johor Strait using MIKE 3FM. International Proceedings ofChemical, Biological and Environmental Engineering, vol. 4. IACSIT Press,Singapore, pp. 295–300.

Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor andanalysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98.

Hishamunda, N., Bueno, P.B., Ridler, N., Yap, W.G., 2009. Analysis of AquacultureDevelopment in Southeast Asia: A Policy Perspective. FAO Fisheries and Aqua-culture Technical Paper. No. 509. FAO, Rome, pp. 69.

Huelsenback, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetictrees. Bioinformatics 17, 754–755.

Kai, A.K.L., Cheung, Y.K., Yeung, P.K.K., Wong, J.T.Y., 2006. Development of single-cellPCR methods for the Raphidophyceae. Harmful Algae 5, 649–657.

Keller, A., Schleicher, T., Schultz, J., Muller, T., Dandekar, T., Wolf, M., 2009. 5.8S-28SrRNA interaction and HMM-based ITS2 annotation. Gene 430, 50–57.

Kempton, J.W., Lewitus, A.J., Deeds, J.R., Law, J.M., Place, A.R., 2002. Toxicity ofKarlodinium micrum (Dinophyceae) associated with a fish kill in a SouthCarolina brackish retention pond. Harmful Algae 1, 233–241.

Ki, J.-S., Jang, G.Y., Han, M.-S., 2005. Integrated method for single-cell DNA extrac-tion, PCR amplification, and sequencing of ribosomal DNA from harmful dino-flagellates Cochlodinium polykrikoides and Alexandrium catenella. Mar. Biotech.6, 587–593.

Koetschan, C., Hackl, T., Muller, T., Wolf, M., Forster, F., Schultz, J., 2012. ITS2Database IV: Interactive taxon sampling for internal transcribed spacer 2 basedphylogenies. Mol. Phylogenet. Evol. 63, 585–588.

Leaw, C.P., Lim, P.T., Ahmad, A., Usup, G., 2001. Genetic diversity of Ostreopsis ovata(Dinophyceae) from Malaysia. Mar. Biotechnol. 3, 246–255.

Lim, H.C., Teng, S.T., Leaw, C.P., Lim, P.T., 2013. Three novel species in the Pseudo-nitzschia pseudodelicatissima complex: P. batesiana sp. nov., P. lundholmiae sp.nov., and P. fukuyoi sp. nov. (Bacillariophyceae) from the Strait of Malacca,Malaysia. J. Phycol 49, 902–916.

Lim, P.T., Usup, G., Leaw, C.P., 2012. Harmful algal blooms in Malaysian waters. SainsMalaysiana 41, 1509–1515.

McManus, G.B., Katz, L.A., 2009. Molecular and morphological methods for identi-fying plankton: what makes a successful marriage? J. Plank. Res. 31, 1119–1129.

Merget, B., Koetschan, C., Hackl, T., Forster, F., Dandekar, T., Muller, T., Schultz, J.,Wolf, M., 2012. The ITS2 Database. J. Vis. Exp. 61, e3806.

Mooney, B.D., De Salas, M., Hallegraeff, G.M., Place, A.R., 2009. Survey for karlotoxinproduction in 15 species of Gymnodinioid dinoflagellates (Kareniaceae, Dino-phyta). J. Phycol. 45, 164–175.

Mooney, B.D., Hallegraeff, G.M., Place, A.R., 2010. Ichthyotoxicity of four species ofgymnodinioid dinoflagellates (Kareniaceae, Dinophyta) and purified karlotox-ins to larval sheepshead minnow. Harmful Algae 9, 557–562.

Muller, T., Philippi, N., Dandekar, T., Schultz, J., Wolf, M., 2007. Distinguishingspecies. RNA 13, 1469–1472.

Nielsen, M.V., 1993. Toxic effect of the marine dinoflagellate Gymnodinium galathea-num on juvenile cod Gadus morhua. Mar. Ecol. Prog. Ser. 95, 273–277.

Nielsen, M.V., Strømgren, T., 1991. Shell growth response of mussels (Mytilus edulis)exposed to toxic microalgae. Mar. Biol. 108, 263–267.

Paulmier, G., Berland, B., Billard, C., Nezan, E., 1995. Gyrodinium corsicum nov. sp.(Gymnodiniales, Dinophycees), organisme responsable d’une ‘‘au verte’’ dansl’etang marin de Diana (Corse), en avril 1994. Crypt. Algol. 16, 77–94.

Place, A.R., Bowers, H.A., Bachvaroff, T.R., Adolf, J.E., Deeds, J.R., Sheng, J., 2012.Karlodinium veneficum – the little dinoflagellate with a big bite. Harmful Algae14, 179–195.

Reuter, J.S., Mathews, D.H., 2010. RNAstructure: software for RNA secondarystructure prediction and analysis. BMC Bioinform. 11, 129.

Scholin, C.A., Anderson, D.M., Sogin, M.L., 1993. Two distinct small-subunit ribo-somal RNA genes in the North American toxic dinoflagellate Alexandriumfundyense (Dinophyceae). J. Phycol. 29, 209–216.

Seibel, P.N., Muller, T., Dandekar, T., Schultz, J., Wolf, M., 2006. 4SALE – a tool forsynchronous RNA sequence and secondary structure alignment and editing.BMC Bioinform. 7, 498.

Seibel, P.N., Muller, T., Dandekar, T., Wolf, M., 2008. Synchronous visual analysis andediting of RNA sequence and secondary structure alignments using 4SALE. BMCEvol. Biol. 1, 91.

Siano, R., Kooistra, W.H.C.F., Montresor, M., Zingone, A., 2009. Unarmoured andthin-walled dinoflagellates from the Gulf of Naples, with the descriptionof Woloszynskia cincta sp. nov, (Dinophyceae, Suessiales). Phycologia 48,44–65.

Sin Chew Daily, January 23, 2005. http://sinchew-i.com (in Chinese).Smayda, T.J., Trainer, V.L., 2010. Dinoflagellate blooms in upwelling systems:

seeding, variability, and contrasts with diatom bloom behaviour. Prog. Ocea-nogr. 85, 92–107.

Swofford, D.L., 2011. PAUP*. Phylogenetic Analysis Using Parsimony (*and OtherMethods). Version 4. Sinauer Associates, Sunderland, MA.

Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6:Molecular Evolutionary Genetics Analysis version 6.0. Mol. Biol. Evol. 30,2725–2729.

Tan, T.H., Lim, P.T., Mohd Razali, R., Leaw, C.P., 2013. Harmful algal species in theTebrau Strait: an SEM observation of the dinoflagellate assemblage. Ann.Microsc. 13, 4–13.

Trainer, V.L., Pitcher, G.C., Reguera, B., Smayda, T.J., 2010. The distribution andimpacts of harmful algal bloom species in eastern boundary upwelling systems.Prog. Oceanogr. 85, 33–52.

Usup, G., Cheah, M.Y., Roziwani, Ng, B.K., Leaw, C.P., Othman, M., Faazaz, A.L., 2003.Identification of the species responsible for the harmful algal bloom event inTebrau Strait in 2002. Malaysian Appl. Biol. 32, 59–62.

Wolf, M., Achtziger, M., Schultz, J., Dandekar, T., Muller, T., 2005. Homologymodelling revealed more than 20,000 rRNA internal transcribed spacer 2(ITS2) secondary structures. RNA 11, 1616–1623.