Morphological and molecular characterization of Kareniaceae (Dinophyceae, Gymnodiniales) in Kuwait’s waters

2.1 Study area

Kuwait’s marine area occupies the northwestern corner of the Gulf, stretching for about 180 km from Iraq in the north to Saudi Arabia in the south (Figure 1A). Kuwait’s waters are hypersaline, with an average salinity of 41.6, shallow, with a maximum depth of about 30 m, and are generally well mixed year round (Al-Yamani 2021; Al-Yamani et al. 2004, 2017, 2021). Kuwait’s subtropical location and arid climate results in marked amplitudes of summer/winter temperature differences (Sheppard et al. 2010). The average annual seawater temperature in Kuwait is 23.8 °C, with seasonal variations ranging from an average of 14 °C in winter to 30.5 °C in the summer months (Al-Yamani et al. 2004). One of the most prominent features of Kuwait’s waters is the semi-enclosed Kuwait Bay, which extends approximately 46 km westward, accounting for nearly half of the country’s coastline and covering an area of approximately 720 km² (Figure 1B). The bay is a shallow, meso-tidal hypersaline inverse estuary, with an average depth of 5.2 m and a maximum depth at its mouth of 23 m and experiences a high tidal range of 4 m (Al-Yamani et al. 2004).

Figure 1:

Area of investigation: (A) general location of the sampling area; (B) map of Kuwait’s marine area; (C) satellite image of Kuwait Bay and adjacent waters showing location of sampling sites.

2.2 Sampling

During 2014, 2016, and 2021, morphological and molecular data were obtained from field seawater samples and from cultures of Karenia and Karlodinium species collected or isolated at several sampling sites during routine phytoplankton monitoring or in response to phytoplankton bloom events (Figure 1C; Table 1).

One-liter seawater samples were collected from the surface layer (1 m depth) using a 5-liter Niskin bottle sampler and were carefully transferred to sampling containers through a flexible hose. Samples were preserved with 4 % acidified Lugol’s solution. A replicate sampling was applied to obtain both preserved and living material simultaneously. Live samples were screened and utilized for taxonomic purposes and the establishment of cultures. Associated water temperature and salinity were measured at the sampling sites using a CTD profiler JFE AAQ1183 (JFE Advantech Co., Japan).

2.3 Species identification

Species identification was primarily based on live material obtained from unpreserved environmental samples or clonal cultures. For a detailed examination, as soon as possible after collection, when cells were in a live state, they were isolated by micropipetting and transferred to a glass slide in preparation for high-magnification photomicroscopy. Cells were examined and photographed using a Leica DM LM compound microscope (Leica Microsystems GmbH, Wetzlar, Germany) equipped with DFC425 color digital camera at 630–1,000 × magnification. A range of microscopic tools were used to examine the cells, including brightfield (BF) illumination to show the natural color, and Nomarski differential interference contrast (DIC) to visualize a three-dimensional appearance of cellular content and cell surface structure in unstained, transparent specimens. Autofluorescence of chloroplasts and nuclei stained by 4ʹ,6-diamidino-2-phenylindole (DAPI, Sigma Chemical Co.) were observed under fluorescence illumination (100 W short-arc mercury lamp, ultraviolet excitation light). Cell dimensions were measured from the images using Leica Application Suite v. 3.7 software (Leica Microsystems (Switzerland) Ltd).

2.4 Cell isolation and maintenance of cultures

Ten monoclonal cultures were established from samples collected in the central part of Kuwait Bay and artificial small semi-enclosed marinas in October 2014, June 2016, and March 2021 (Figure 1C; Table 1). Single cells were isolated using sterilized drawn out capillary pipettes under an inverted microscope, washed at least three times in a drop of sterilized seawater, and were then transferred to a 96-well plate filled with 350 μl of sterile K medium without silicate (Keller et al. 1987) prepared with filtered (0.22 μm, Millipore) and autoclaved natural seawater with a salinity of approximately 40. The isolated cells were incubated at 23.5 °C under 80 μmol photons·m⁻²·s⁻¹ of light and a 12:12 h light:dark photoperiod and checked regularly. When the cell concentration became sufficiently high, the contaminant-free isolates with dividing and motile cells were transferred to 25-cm² tissue culture flasks containing 40 ml sterile K medium as stock cultures. The established cultures are stored and maintained at the Laboratories of Phytoplankton Taxonomy of the Kuwait Institute for Scientific Research (KISR, Kuwait) and the IOC Science and Communication Centre on Harmful Algae, University of Copenhagen, Denmark. They are checked biweekly to maintain healthy growing cultures, and 1 ml of each strain is transferred to fresh medium monthly.

2.5 DNA amplification and sequencing

DNA was extracted from eight monoclonal strains of Karenia isolated from Kuwait’s waters in 2014 and 2021 and from two monoclonal strains of Karlodinium isolated in 2014 and 2016. Cells were harvested as pellets from 15-ml aliquots of late exponential phase cultures via gentle centrifugation at 4,000g for 5 min at room temperature. Total genomic DNA was extracted from the cell pellets using the DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. DNA quality was assessed by agarose gel electrophoresis and quantified using a Nanodrop™ 1,000 spectrophotometer (Thermo Scientific, Labtech International Ltd, UK). Aliquots of DNA were kept at 4 °C for immediate PCR analysis, and replicate stocks were held at −80 °C for long-term storage.

Genomic DNA resulting from the 10 microalgal isolates was examined by PCR amplification using the D1/D2 region of the large nuclear subunit (LSU) of the ribosomal operon with the forward primer D1R-F and the reverse primer D2C-R (Scholin et al. 1994) for the eight isolates and the D1/D3 region of the LSU rDNA with the forward primer D1C-F (Scholin et al. 1994) and the reverse primer, either 28–1483R (Daugbjerg et al. 2000) or D3B-R (Nunn et al. 1996), for the five isolates (Tables 1 and 2).

PCRs were carried out using a Bio-Rad C1000 thermal cycler (Bio-Rad Laboratories Inc., Hercules, CA, USA). PCR amplification conditions were as follows: initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 1 min and extension at 72 °C for 1 min, and a final extension step at 72 °C for 5 min. Reactions were performed in 50 µl of the AccuPower PCR Premix (Bioneer, Daejeon, Korea) and 5X HOT FIREPol Blend Master Mix, containing 1 µl template DNA (3.3–47.5 ng), 20 pmol forward primer, 20 pmol reverse primer. All the PCR products were separated through 1.6 % (w/v) agarose gels in 1 × TAE (Tris-acetate-EDTA) buffer, stained with GelRed (Biotium Inc., Hayward, CA, USA), and visualized using a UV fluorescent gel documentation system (Bio-Rad, UK).

Amplicons were sequenced after two-step qualitative PCR with the KAPA2G Robust HotStart Kit using Buffer B and Enhancer A (Roche, Kit Code KK5525, Cat. No 07961120001), then amplicons showing a clear single band were sequenced using the ABI 3730XL DNA Analyzer (PE Biosystems, Vernon Hills, IL, USA). The identity of PCR products was confirmed by using BLASTN analysis (www.ncbi.nlm.nih.gov/BLAST). The consensus sequences were assembled and edited by pair-wise alignment using BioEdit 7.7.1 (Hall 1999). New sequence data were deposited in the GenBank (National Center for Biotechnology Information, NCBI) under the following accession numbers: PP951870-PP951879.

2.6 Phylogenetic analyses

The obtained sequences of partial LSU rDNA were aligned with other related Kareniaceae species retrieved from the NCBI nucleotide database (http://www.ncbi.nlm.nih.gov/) using the ClustalW algorithm (Thompson et al. 1994) using MEGA 10.2.6 (Kumar et al. 2018) and were subsequently verified manually. The LSU rDNA dataset of this study contained 1,002 characters and comprised 10 nucleotide sequences of Kareniaceae from Kuwait’s waters and 47 sequences of other closely related taxa, which were used for phylogenetic analyses. Phylogenetic relationships were analyzed by the maximum likelihood (ML) method using MEGA 10.2.6 and by Bayesian inference (BI) using MrBayes v. 3.1.2 (Ronquist et al. 2012). The Tamura-Nei model of nucleotide substitution (Tamura and Nei 1993) with gamma-distributed rates among sites (TN93 + G) was considered best to describe the substitution pattern for ML based on the lowest Bayesian Information Criterion (BIC) scores and corrected Akaike Information Criterion (AICc) values. The initial tree for the heuristic search was obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances, and then selecting the topology with the superior log likelihood value. A discrete gamma distribution was used to model evolutionary rate differences among sites (5 categories (+G, parameter = 0.4071)). Bootstrap analysis was carried out with 1,000 replicates to assess statistical reliability. For Bayesian inference, the GTR substitution model with gamma-distributed rate variation across sites and a proportion of invariable sites (GTR + I + Г) was used. Four independent Markov chain Monte Carlo simulations were run simultaneously for 1,000,000 generations, and trees were sampled at every 1,000 generations. The first 25 % of the trees were discarded as burn-in. Convergence was judged by the average standard deviation of split frequencies (less than 0.01). The remaining trees were used to generate a consensus tree and calculate the posterior probabilities of all branches using a majority-rule consensus approach. FigTree 1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/) was used to view and edit the trees. The ML topology was used to visualize the phylogenetic tree. The tree with the highest log likelihood (−3,690.53) is shown. GenBank accession numbers and collection geographic region for Kareniaceae sequences used in this study are provided in the phylogenetic tree. Genetic distances (p-distances) between sequences were calculated using an uncorrected genetic distance (UGD) model with a pairwise deletion option (MEGA 10.2.6).

3.1 Species morphology

Morphological descriptions were based on the results of observations using light and epifluorescence microscopy of cultured material gathered during this study and field samples collected in the coastal waters of Kuwait. Conservative diagnostic characters used to distinguish between different kareniacean morphotypes included cell shape (length/width ratio), apical groove length, degree of extension of the sulcal groove onto the epicone, presence of a ventral pore, and shape and position of the nucleus (Table 3). Based on morphological characteristics, three distinct morphotypes from Kuwait’s waters (Figures 2–4; Table 3) were identified, matching well the descriptions of K. papilionacea, K. selliformis, and Karlodinium ballantinum (de Salas et al. 2008; Haywood et al. 2004; Steidinger et al. 2008).

Figure 2:

Light micrographs of Karenia papilionacea cells (Figure 2A–C: strain KW-F3-015; Figure 2D–G: strain KW-E1-014; Figure 2N: strain KW-G4-062; Figure 2H–M, O-R: isolates from field samples; arrows point to the apical groove; arrowheads point to the open sulcal extension onto the epicone; n marks the nucleus): (A, B) ventral view showing apical groove, open-ended sulcus onto the epicone, and nucleus in the left lobe of hypocone; (C) epifluorescence microscopy under blue-light illumination showing numerous kidney-shaped chloroplasts; (D, E) ventral and dorsal views of the same cell showing apical groove, sulcus, and nucleus; (F) apical view in deep focus showing dorso-ventral compression; (G) epifluorescence microscopy showing several elongated chloroplasts; (H) two cells in ventral and dorsal view; (I–K) cells in left lateral, apical, and antapical views showing dorso-ventral compression; (L) ventral view showing apical groove and open-ended sulcus onto the epicone; (M) dorsal view showing apical groove over epicone; (N) Lugol’s-fixed cell in ventral view; (O–R) contracted cells during butterfly-like movement. Scale bars: 10 µm.

Figure 3:

Light micrographs of Karenia selliformis cells (Figure 3A–D, L: strain KW-D10-060; Figure 3C–K: isolates from field samples; arrows point to the apical groove; arrowheads point to the sulcal intrusion onto the epicone; n marks the nucleus): (A, B) ventral view showing apical groove and sulcal intrusion onto the epicone; (C) squashed cell showing numerous elongated chloroplasts and transversely elongated nucleus in the hypocone; (D) epifluorescence microscopy under blue-light illumination showing numerous peripherally located, elongated chloroplasts; (E, F) ventral view showing apical groove and sulcal intrusion onto the epicone; (G) lateral view showing strong dorso-ventral compression; (H, I) dorsal view of different cells showing apical groove over epicone and nucleus position; (J) epifluorescence microscopy of DAPI-stained cell showing transversely elongated nucleus in hypocone; (K) epifluorescence microscopy showing numerous peripherally located, elongated chloroplasts; (L) Lugol’s-fixed cell in ventral view. Scale bars: 10 µm.

Figure 4:

Light micrographs of Karlodinium ballantinum cells (Figure 4A–C, F, G: strain KW-JL-07; Figure 4D, E, H–L: strain KW-E9-046; n marks the nucleus): (A–C) ventral view showing apical groove (arrow) and short finger-like sulcal intrusion onto the epicone (arrowhead); (D) dorsal view showing apical groove (arrow) over epicone; (E) apical view showing apical groove (arrow) over epicone and no dorso-ventral compression; (F) dividing motile cell; (G) squashed cell in dorsal view showing the knob-like structures lining the lower margin of the cingulum (arrows); (H) squashed cell showing large central nucleus surrounded by several chloroplasts; (I) epifluorescence microscopy of DAPI-stained cell in dorsal view showing centrally located nucleus; (J) epifluorescence microscopy of DAPI-stained cell in lateral view showing nucleus close to the dorsal side; (K) epifluorescence microscopy showing numerous peripherally located, elongated chloroplasts with internal pyrenoids (white arrows); (L) Lugol’s-fixed cell in ventral view. Scale bars: 10 µm.

3.2 Phylogenetic relationships

A total of 10 kareniacean strains isolated from Kuwait’s waters were sequenced during this study and included in the phylogenetic analyses. The construction of phylogenetic trees using Maximum Likelihood (ML) and Bayesian Inference (BI) methods based on the D1/D2 and D1/D3 domains of the LSU rDNA resulted in a consistent topology, with high bootstrap support and posterior probability for most clades (Figure 5).

Figure 5:

Maximum-likelihood (ML) phylogenetic tree of the D1-D2/D1-D3 LSU rDNA sequences of the Kareniaceae from Kuwait’s waters and closely related species from other locations around the world. The numbers on the branches are statistical support values (ML bootstrap support (%), n = 1,000/Bayesian posterior probability, n = 1,000,000). GenBank accession numbers and collection geographic regions follow the taxon names. New sequences are shown in bold. Scale bar = number of nucleotide substitutions per site.

The newly sequenced Karenia strains were placed in well-supported monophyletic clades of K. selliformis or K. papilionacea (Figure 5) and distinct from related Karenia species. Within the highly supported (ML bootstrap value of 99 and BI posterior probability of 0.97) clade of K. selliformis strains with a worldwide distribution, Kuwait’s strain KW-D10-060 was closest to those originating from New Zealand, including the holotype sequence U92250, forming a subclade with moderate bootstrap (62) and high posterior probability (0.99) values. Other K. selliformis strains formed three distinct, geographically consistent subclades, including either 1) Asian strains originating from Japan and China (66/0.94), 2) Chilean strains (87/0.99), or 3) Russian strains isolated from Kamchatka, Bering Sea (86/0.65).

The highly supported clade of K. papilionacea (100/1.00) comprised three subclades. The seven strains from Kuwait were found in a large subclade among strains originating from New Zealand (the type locality), Australia, Japan, China, Hong Kong, and Korea (89/0.99) (Figure 5). The remaining K. papilionacea sequences split into two well-supported subclades consisting of either European (86/0.99) or Japanese (85/0.99) strains. The strains isolated in Kuwait Bay in 2014 (KW-E1-14 and KW-F3-15) and those obtained during the Karenia bloom in 2021 (KW-G4-062, KW-G5-063, Kw-G8-064, and KW-G9-065) were identical (the pairwise genetic distance = 0).

Sequences obtained for Karlodinium from Kuwait were found in a clade of other K. ballantinum strains with strong support (99/1.00). The two Kuwait strains appeared in a subclade with Asian strains from Japan and the Philippines and the Australian strain, representing the holotype. A minor genetic divergence (p-distance = 0.0044, 3 bp difference) was detected between K. ballantinum strains isolated in Kuwait Bay in 2014 (KW-E9-46) and in the artificial KISR Salmiya marina to the south in 2016 (KW-JL-07), but in the phylogenetic analyses they clustered together (Figure 5).

4.1 Karenia selliformis

K. selliformis was the first species of the family Kareniaceae identified in Kuwait’s waters. Its appearance was linked to red tide and large fish kills in Kuwait Bay in September-October 1999 (Heil et al. 2001; Husain and Faraj 2000). It was subsequently recorded again in 2001 but did not form a bloom and was reported at a concentration of 400 cells l⁻¹ (Glibert et al. 2002). This species was initially reported as Gymnodinium cf. mikimotoi (Husain and Faraj 2000) and then as Gymnodinium sp. or G. selliforme (Heil et al. 2001), and Karenia sp. (Glibert et al. 2002) (Table 4). In the course of a taxonomic examination of bloom material by A. Haywood and K. Steidinger, the Kuwaiti taxon was believed to be conspecific with yet undescribed Gymnodinium sp. from New Zealand coastal waters, which possessed the laterally elongated nucleus in the hypocone (Chang 1995; MacKenzie et al. 1995, 1996). Upon the description of the genus Karenia (Daugbjerg et al. 2000), this taxon was recognized as fitting the newly erected genus and redescribed as K. selliformis (Haywood et al. 2004). The authors noted the conspecificity of material from Kuwait due to similar morphology (Haywood et al. 2004; Steidinger et al. 2008). K. selliformis was recurrently found in Kuwait’s waters during the phytoplankton taxonomic survey in 2003–2007, when its morphology was illustrated by LM and SEM (Al-Kandari et al. 2009, pl. 7) (Table 4).

The observed morphological characters of K. selliformis cells from field samples and monoclonal culture (strain KW-D10-60) isolated from Kuwait’s waters are consistent with the original description of this species (Haywood et al. 2004) and previous records from Kuwait (Al-Yamani and Saburova 2019; Al-Kandari et al. 2009; Heil et al. 2001) in terms of their shape, apical groove length and path, and position of the nucleus (Figure 3; Table 3). The cells observed in this study were slightly larger than those described initially. The cells were 30–33 µm long, as compared to 26–32 µm (MacKenzie et al. 1996) and 20–27 μm (Haywood et al. 2004), but our measurements fall within the range previously reported from Kuwait (17–37 μm, Heil et al. 2001; 30–34 μm, Al-Yamani and Saburova 2019). The most notable characteristics distinguishing this species from the other Karenia are the large, transversely elongated nucleus located posteriorly and the higher number of chloroplasts (Table 3).

Molecular phylogenetic analysis using partial LSU rDNA showed that the strain from Kuwait clearly belonged to the K. selliformis clade and was genetically distinct from other Karenia species (Figure 5). Kuwait’s strain was most closely related to isolates from New Zealand and was 99.86 % identical (1 bp difference) to the holotype sequence of K. selliformis (U92250). This study provides the first molecular data on K. selliformis isolated from Kuwait’s waters and confirms the taxonomic identity of this dinoflagellate that has caused harmful blooms.

Recently, the genetic variability among the LSU rDNA and ITS sequences of K. selliformis strains originating from distant geographical areas worldwide was found to be high enough to form distinct nested subclades, suggesting the existence of two or three distinct phylotypes, depending on the molecular marker used (Iwataki et al. 2022; Mardones et al. 2020; Orlova et al. 2022). Moreover, the delineated phylotypes were distinguished phenotypically and by their temperature tolerance (Iwataki et al. 2022; Mardones et al. 2020; Orlova et al. 2022), pointing to the potential K. selliformis “species complex” (Mardones et al. 2020).

Our morphological and molecular data support the hypothesis of phylotype separation within K. selliformis, differentiating phenotypically and by temperature preference. In the phylogeny inferred from partial LSU rDNA, the topology of the K. selliformis clade is consistent with that previously described (Mardones et al. 2020). The K. selliformis strains were grouped into distinct nested subclades representing the different phylotypes (Figure 5). The basal subclade (phylotype I) comprised strains isolated from K. selliforms bloom along the coast of the Kamchatka Peninsula in 2020. The strain from subtropical Kuwait clustered with other warm-water strains originating from New Zealand, which were closely related to the Asian strains, belonging to phylotype II, and separated from another subclade that included the Chilean strains (phylotype III). The warm-water strain from Kuwait was represented by strongly dorso-ventrally flattened cells with a distinct antapical notch consistent with the described initially K. selliformis from New Zealand waters. In contrast, strains blooming in Chile (1999), Kamchatka, Russia (2020), and Hokkaido, Japan (2021), which belong to phylotype I, exhibited cells with abundant granular to strap-shaped chloroplasts and bloomed at low temperatures <20 °C (Iwataki et al. 2022; Orlova et al. 2022; Uribe and Ruíz 2001). The strains isolated during the K. selliformis bloom in Chile in 2018 (MN203220 and MN203221) were genetically distinct enough to form phylotype III. Morphologically, cells of this phylotype differed from the holotype strain by weak dorsal-ventral compression, shallower antapical excavation, and a smaller number of lateral pores in the hyposome (Mardones et al. 2020).

4.2 Karenia papilionacea

The taxonomic survey by Al-Kandari et al. (2009) revealed the second morphotype of Karenia in Kuwait’s waters. It was distinguished from K. selliformis by its cell shape, being wider rather than oblong. The small globular nucleus of this morphotype was located in the left lobe of the hypocone versus the larger, transversely elongated, and posteriorly located nucleus in K. selliformis (Table 3). While the morphology of this taxon resembled K. brevis, its unambiguous identification remained uncertain for a long time. In taxonomic and ecological phytoplankton studies in Kuwait’s waters, this taxon has been variously referred to as either Gymnodinium breve (Al-Yamani et al. 2004), Karenia cf. brevis (Al-Kandari et al. 2009, plate 8A, LM), K. brevis (Devlin et al. 2019), or K. papilionacea (Al-Yamani and Saburova 2019; Al-Yamani et al. 2012, plate 60, LM; Polikarpov et al. 2020) (Table 4).

Morphologically, K. papilionacea is similar to K. brevis, and these two species can be distinguished only by minor details. The cells of K. papilionacea are wider (24–36 µm vs 18–48 µm), possess pointed rather than bulbous apical protrusions (carina), and have slightly deeper antapical excavations (Haywood et al. 2004). Given the high morphological variability observed in both species, even within the same strain (Persson et al. 2013; Stuart 2011), it is doubtful whether species discrimination based on morphological criteria alone can be considered reliable. Nevertheless, these two species are genetically distinct (Haywood et al. 2004).

K. papilionacea may have previously been misidentified as K. brevis worldwide. The geographic range of K. brevis appears to be limited to the Gulf of Mexico (Brand et al. 2012; Magaña et al. 2003; Steidinger 2009). Previous reports of fish mortality caused by K. brevis in Asia and the Mediterranean Sea require molecular analysis to verify if they are actually misidentifications of K. papilionacea or other Karenia species (e.g., Tsikoti and Genitsaris 2021; Yamaguchi et al. 2016; Yeung et al. 2005).

The seven strains of K. papilionacea examined in this study displayed similarity to each other, and the morphological characters observed under LM were consistent with previous reports of this species (e.g., Al-Yamani and Saburova 2019; Kim et al. 2023; Yamaguchi et al. 2016), including its original description (Haywood et al. 2004). Despite some morphological variability within and among strains in cell size and shape, the typical cells were wide, dorsoventrally flattened, with short apical grooves, pointed apices, and round nuclei in the left lobe of the hypocone (Figure 2; Table 3).

In the phylogenetic tree based on partial LSU rDNA sequences, the strains from Kuwait were grouped together within the monophyletic clade of K. papilionacae among strains isolated from New Zealand (the type locality) and Japan with strong statistical support (Figure 5). The sequences obtained from Kuwait’s strains were identical, sharing 99.7 % similarity (2 bp differences) to that of the type strain of K. papilionacea (U92252), and were clearly genetically distinct from the K. brevis clade. Therefore, our results provide the first unambiguous molecular identification of K. papilionacea from Kuwait’s coastal waters and resolve the previous regional taxonomic uncertainty for this species. This implies that earlier records of K. brevis-like species in Kuwait’s waters need to be reconsidered and attributed to K. papilionacea. Furthermore, the genetic similarity among the strains isolated seven years apart (first in 2014 and then in 2021) and from different sampling sites (Kuwait Bay and semi-enclosed marina, Figure 1C; Table 1) suggests the persistent presence of the same population of this species throughout the study area.

Similar to K. selliforms, recent phylogenetic analyses based on both LSU and ITS have revealed genetic divergence among the K. papilionacea strains, resulting in the distinction of at least two phylotypes within this species (Kim et al. 2023; Yamaguchi et al. 2016). No morphological differences were observed between phylotypes, while different physiological growth traits were assumed to distinguish them (Yamaguchi et al. 2016). In our LSU-based phylogeny, the topology of the K. papilionacea clade was almost the same as previously reported by Yamaguchi et al. (2016) and Kim et al. (2023) and consisted of three distinct, well-supported subclades (Figure 5). All of Kuwait’s strains were placed in a large, diverse, and strongly supported clade (89/0.99) among strains from Australia, New Zealand, Japan, China, and Korea, corresponding to the original phylotype (as per Yamaguchi et al. 2016). The second distinct subclade, with high nodal support (85/0.99), solely comprised K. papilionacea strains restricted to the western Japanese coast (Yamaguchi et al. 2016), representing phylotype I. Two strains originating from the French Atlantic and north-western Mediterranean Sea, Spain, were genetically divergent from those of the original phylotype and phylotype I, clustering separately into a well-supported subclade (86/0.99).

4.3 Karlodinium ballantinum

Like Karenia, Karlodinium species possess the characteristic straight apical groove but differ in the presence of a ventral pore on the epicone in most species and a unique type of amphiesma with plugs. Typical Karlodinium cells are small, globular rather than flattened like most Karenia, and their straight versus sigmoid apical groove distinguishes them from the morphologically similar Takayama species (Daugbjerg et al. 2000; de Salas et al. 2003). Most Karlodinium species share similar morphology and often overlap in size, shape, and diagnostic morphological characters. Species differentiation within this genus is based on cell shape, the presence/absence of a ventral pore, the position of the nucleus, and the arrangement of chloroplasts. These characters are often difficult to distinguish in preserved samples but may be discerned by observing live cells (e.g., Bergholtz et al. 2006; de Salas et al. 2008).

Until recently, dinoflagellates belonging to the genus Karlodinium have not been recognized from Kuwait’s waters due to research being restricted to Lugol’s preserved samples and possible misidentification with other small gymnodinioid taxa. Five Karlodinium morphotypes have recently been reported from Kuwait based on the observations of live cells in freshly collected water samples (Table 4), including Karl. australe, Karl. decipiens, Karl. digitatum (as Karenia digitata), Karl. gentienii, and Karl. veneficum, as reported by Al-Yamani and Saburova (2019); however, no molecular data were provided for these species. In this study, two Kuwait strains, KW-JL-07 and KW-E9-046, exhibited the typical Karlodinium morphology, with small, nearly spherical to ellipsoidal cells, possessing a short straight apical groove, a large centrally located nucleus, and several elongated peripheral chloroplasts. The lack of a ventral pore in this species (as discerned by LM) sets it apart from most Karlodinium taxa previously reported in Kuwait, suggesting the finding of an unrecorded taxon. Among the currently described Karlodinium species, an inconspicuous or lacking ventral pore has been reported in Karl. antarcticum, Karl. ballantinum, Karl. digitatum, Karl. jejuense, and Karl. zhouanum (de Salas et al. 2008; Li and Shin 2018; Luo et al. 2018; Siano et al. 2009; Yang et al. 2000). The cell morphology of Kuwait’s strains differed from that described in four out of five pore-lacking Karlodinium species in terms of shape (ellipsoidal instead of elongated in Karl. antarcticum), nucleus position (centrally located nucleus versus those in the epicone in Karl. zhouanum or in the hypocone in Karl. antarcticum and Karl. digitatum), the apical groove length that extends for a short distance onto the dorsal epicone compared to its longer path (1/4–1/2 down the dorsal epicone) in Karl. antarcticum, Karl. digitatum, Karl. jejuense, and Karl. zhouanum, and the knob-like structures (microprocesses) lining the lower margin of the cingulum (Figure 4G) have not been observed in other Karlodinium species. The morphology observed in Kuwait’s strains corresponds well with that described for Karl. ballantinum (Table 3). Although de Salas et al. (2008) originally reported this species as lacking a ventral pore, further SEM analysis by Benico et al. (2020) of strains from the Philippines and Japan revealed a pore or pore-like shallow depression in this species. As our morphological observations of this species were limited to light microscopy, further SEM examination is required to determine whether this character varies in Kuwait’s material.

The phylogenetic tree inferred from the LSU rDNA sequences in the present study also showed a high genetic similarity between the sequences obtained from Kuwait and the type material of Karl. ballantinum from Tasmania (Figure 5). Kuwait’s strains shared 98.19–99.27 % similarity (5 bp differences) with the holotype strain EF469232. Through combined evidence from molecular phylogeny and morphology, this study reports the identification of K. ballantinum in Kuwait’s waters, marking the first record of this species in the entire Gulf.

4.4 Risk assessment of kareniacean harmful algal blooms

The family Kareniaceae is of particular global ecological concern because it contains many toxigenic bloom-forming species with a wide range of negative impacts. However, the types of harmful algal blooms (HABs), the toxins they produce, and hence the impacts of the blooms are species-specific and vary significantly from region to region. This makes a robust taxonomic identification of species within the Kareniaceae ecologically and economically important for regional HAB risk assessments.

A long-term phytoplankton monitoring program revealed that several kareniacean species are seasonally part of the phytoplankton community along Kuwait’s coast, often present at background to low concentrations (100–10,000 cells l⁻¹) (Al-Kandari et al. 2009; Al-Yamani et al. 2012; Al-Yamani and Saburova 2019; Devlin et al. 2019; Polikarpov et al. 2020). Three species belonging to the genus Karenia, namely K. selliformis, K. papilionacea, and K. mikimotoi, as well as Karlodinium spp. and Takayama spp., can be considered bloom-forming due to their ability to form sporadic outbreaks (>10⁵ cell l⁻¹) in Kuwait’s waters resulting in surface water discolouration (red tides) (Al-Yamani and Saburova 2019; Al-Yamani et al. 2012; Heil et al. 2001; Polikarpov et al. 2020).

The dinoflagellate species of particular concern for Kuwait is the ichthyotoxic K. selliformis. Its intense bloom in Kuwait Bay in 1999 resulted in massive fish kills and considerable economic losses (Al-Yamani et al. 2004, 2012; Heil et al. 2001). Although no harmful blooms of this species have been recorded since then, Kuwait’s marine environment remains at risk of further K. selliformis blooms. This species routinely co-occurs with K. papilionacea and K. mikimotoi in Kuwait’s coastal waters, forming sporadic multi-species blooms (Polikarpov et al. 2020). The presence of these toxigenic Karenia species in the phytoplankton composition could potentially contribute to new HAB events in response to nutrient surplus and increased seawater temperatures, particularly in areas with low water exchange rates, such as the shallow, semi-enclosed Kuwait Bay and numerous artificial marinas along Kuwait’s coast.

The bloom-forming toxigenic kareniacean species can be considered a latent threat to Kuwait’s marine environment and requires more detailed and comprehensive studies of their taxonomy, dynamics, and toxicity. Accurate identification and quantification of cell abundance in the water column is essential for routine monitoring of coastal waters. Potential misidentification of tiny, fragile, and poorly preservable kareniacean cells can result in significant underestimation of abundance and consequently affect the quality of the risk assessment. The key diagnostic characters that distinguish different kareniacean morphotypes, in particular the apical and sulcal grooves and the ventral pore, can be discerned by high-resolution microscopy of live cells. However, these taxonomically important morphological characters are often indistinct in preserved samples, resulting in the loss of data when Lugol’s-fixed samples are used for routine monitoring programs. To facilitate further phytoplankton monitoring in Kuwait’s waters, representative images of Lugol’s-fixed cells were documented from the cultured strains of K. papilionacea, K. selliformis, and Karl. ballantinum, which were identified through close morphological and molecular analyses (Figures 2N, 3L, and 4L).

Routine monitoring based on microscopy alone is insufficient for unambiguous species identification among the species belonging to the family Kareniaceae; therefore, supporting molecular data is necessary to evaluate potential risks, particularly for species verification of high biomass blooms. Establishing a HAB culture collection at KISR with the morphological and genetic characterization of the cultured species is the first valuable resource for HAB research in Kuwait and the entire Gulf region. Testing the ichthyotoxicity of these local strains to fish and invertebrates could be a priority topic for further research.

The long-term uncertainty surrounding the identity of K. brevis/K. papilionacea has affected the recent regional checklist of phytoplankton (Attaran-Fariman and Asefi 2022), in which both species were listed. Further molecular studies are required to ascertain whether previous records of the K. brevis-like morphotype should be attributed to K. papilionacea or if both species co-occur across the Gulf, as has been observed in the Gulf of Mexico (Steidinger et al. 2008) and along Florida’s east coast (Wolny et al. 2015).

Further investigations should focus on the development of an efficient molecular marker to discriminate among the isolates of Kareniaceae species from Kuwait, the region, and worldwide in order to track their origin and possible spreading pathways. In this regard, the plastid gene ribulose-bisphosphate carboxylase (rbcL) sequences have been found to be more discriminating than the LSU and ITS nuclear markers (e.g., Daugbjerg et al. 1995) and have been successfully used to distinguish Karenia mikimotoi isolates originating from different geographical regions (Al-Kandari et al. 2011).