Skin Irritation-Associated Dinoflagellate Vulcanodinium rugosum Isolated from Cienfuegos Bay, Cuba: Toxin Profile and Cell Growth Characterization Under Laboratory Conditions

TL;DR

This study examines the growth and toxin production of a dinoflagellate linked to skin irritation in Cuba, providing insights for monitoring and forecasting harmful algal blooms.

Contribution

The study provides the first data on toxin profiles and growth characteristics of Vulcanodinium rugosum under controlled laboratory conditions.

Findings

V. rugosum cells can grow under agitated conditions, similar to those with low hydrodynamic energy.

Higher toxin levels were observed during exponential growth, with Portimine-A and pinnatoxin-F as dominant toxins.

Abstract

Blooms of the marine dinoflagellate Vulcanodinium rugosum have been associated with skin lesion outbreaks in Cuba and elsewhere. In this study, cell growth and toxin production were investigated under laboratory-controlled conditions in two strains isolated from Cienfuegos Bay, Cuba. Strains were cultured with and without a mechanical agitation and toxins were analyzed at two stages of the culture growth (exponential and stationary). Although blooms in Cienfuegos Bay occur in a semi-enclosed system characterized by calm waters with no agitation, the results of this study suggest that V. rugosum cells may also exhibit growth capacity under agitated conditions, or in open waters, comparable to that observed in systems with low hydrodynamic energy. Higher toxin levels, as determined by liquid chromatography with mass spectrometry (LC-MS/MS), were detected after exponential growth. Portimine-A and pinnatoxin-F (PnTX-F) were the dominant toxins (up to 1.75 and 1.0 pg·cell−1, respectively). PnTX-E, -D and Portimine-B were also detected at minor concentrations. This study contributes the first data necessary for a proper interpretation of monitoring programs aiming to assess the impact of V. rugosum blooms, particularly when used alongside forecasting models.

Figures

Full text

1. Introduction

The dinoflagellate Vulcanodinium rugosum has been associated with cases of dermatitis and skin lesions in Cienfuegos Bay, Cuba, since 2015 [1]. In the first episode, more than 60 individuals, predominantly children under 14, developed acute skin irritations with severe itching and eczema. Lesions were primarily located in the body’s most moisture-prone areas, including the inguinal and genital regions, pelvis, and gluteal area, often corresponding to the outline of the affected person’s swimsuit. Erythematous blisters and intense itching appeared within 3–4 h of contact with the bloom, progressing to ulcerative lesions that frequently became secondarily infected, taking on a reddish, pustular, and crusted appearance [1]. This incident caused social alarm, and the Cuban authorities temporarily closed the beach. These incidents highlighted the need to better understand the dynamics of harmful algal blooms regarding this species, particularly its cell growth rates and its toxin production.

Vulcanodinium rugosum is a toxic dinoflagellate associated with the production of pinnatoxins and portimines, which are potent neurotoxins acting as selective antagonists of nicotinic acetylcholine receptors [2]. These compounds belong to the so-called fast-acting marine cyclic imines. Due to its potent neurotoxic effects, V. rugosum has also gained significant attention as an agent responsible for seafood contamination, particularly contamination of mussels and other filter-feeding shellfish organisms. In fact, it was due to toxic shellfish extracts detected by a French monitoring program that V. rugosum was first described by Nézan & Chomérat (2011) in a Mediterranean coastal lagoon [3]. To date, it is the only species of this genus, but since its first description, it has been observed globally, including in Italy, Japan, China, New Zealand, Mexico, the USA and Cuba [1,4,5].

The life cycle of V. rugosum is particularly interesting and may favor its spread via ballast waters, as it includes the vegetative cell stage but also non-motile cells [6]. Non-motile cells, also designated as temporary cysts, have been found to survive and germinate after fish gut passage, which highlights their strong potential for environment dissemination [7]. In Cuba, the species seems to have been introduced via ship ballast waters, as suggested by the comparative phylogenetic studies carried out [1]. The profile of toxins is another parameter that helps to better understand regional differences among Vulcanodinium populations. Currently, eight pinnatoxins (PnTX-A to –H) and two portimines (A and B) have been described from V. rugosum cells or accumulated in shellfish. Toxin analysis of a 2015 Cienfuegos bloom carried out by liquid chromatography with mass spectrometry detection (LC-MSMS) indicated a dominance of portimine-A and PnTX-F, as well as minor amounts of PnTX-E and D [1]. This profile contrasts with the profile determined for V. rugosum from Florida, which contained only portimines [8], and from France, which was dominated by PnTX-G [5,9], or from China, which only contained PnTX-H [10]; it also differs from New Zealand strains, which contained PnTX-E and –F [11].

The diversity of toxins observed in V. rugosum-enriched phytoplankton samples collected in Cienfuegos Bay in Cuba [1] has not yet been confirmed in cultured strains under laboratory-controlled conditions. This study aims to characterize the growth of V. rugosum isolated from this region and determine its toxin production in the two main culture stages (exponential and stationary phase). It also aims to assess the effect of mechanical agitation, mimicking bay hydrodynamics, on growth and toxin production. Through these objectives, the study seeks to improve our understanding of how rapidly V. rugosum populations can establish and sustain blooms in the environment, potentially reaching hazardous levels in recreational waters.

2. Results

2.1. Growth Curves

For both Vulcanodinium rugosum strains and treatments, the exponential phase occurred between the 4th and the 10th day, as evidenced by the moving average line trends for growth (Figure 1). For strain V5, the growth was slower and more gradual, peaking at around 4000 to 4500 cells·mL^−1^ on day 10 and showing a more sustained plateau. For strain V12, the growth was faster and steeper, also peaking at around 3500 to 4000 cells·mL^−1^ on day 10, and after peaking, it rapidly declined (Figure 1). Regarding the agitation treatment, strain V5 showed variable growth, with no significant differences when subjected to agitation, except for the 4th day of exposure, where growth with no agitation was significantly higher (Figure 1A). Strain V12 showed constant growth for both treatments, with significant differences from the 7th day until the end of the experiment at the 18th day, where growth with agitation was greater than growth with no agitation (Figure 1B).

Growth rates were similar in both treatments for the V5 strain, with an average µ = 0.150 ± 0.031 for agitation and an average µ = 0.144 ± 0.016 for no agitation, with a doubling time of about 5 days, corresponding to an increment of 16% per day (Table 1). Conversely, the V12 strain showed a higher growth rate for cultivation with agitation and greater growth than the V5 strain for both treatments, with an average µ = 0.196 ± 0.030 and a shorter doubling time of approximately 4 days, corresponding to a 21% increase each day. For the treatment without agitation, the V12 strain showed lower growth than the treatment with agitation, with an average µ = 0.102 ± 0.031 and a longer doubling time of approximately 7 days, corresponding to a 10% increase each day (Table 1). The maximum growth recorded was 4063 cells·mL^−1^ at 240 h without agitation for the V5 strain and 3507 cells·mL^−1^ for the V12 strain, also at 240 h, but with agitation (Figure 1).

2.2. Toxin Profiles

The two strains cultivated in the laboratory produced four pinnatoxins, namely PnTX-F, -E, -D and -B, as well as portimine-A and -B (Figure 2). Portimine-A and PnTX-F were the most prevalent toxins determined throughout the culture study. Mean concentrations of 0.9 ± 0.3 pg*·cell^−1^ were determined for Portimine-A, ranging from 0.40 up to 1.75 pg·cell^−1^. PnTX-F varied from 0.2 up to 1.0 pg·cell^−1^. PnTX-E was also observed in each sampling point, varying from 0.03 to 0.48 pg·cell^−1^. Lower levels were determined for PnTX-D and portimine-B, which, when detected, mostly in the stationary phase of the culture (day 18), reached concentrations up to 0.08 and 0.06 pg·*cell^−1^, respectively.

No significant differences in production of any of the toxins were observed between strains for day 10, which corresponds to the exponential phase of the culture growth. On day 18, corresponding to the stationary phase, both strains produced higher toxin levels compared to the initial sampling day 10. This marked increase was significantly higher on strain V12 compared to V5 regarding the least abundant compounds, namely PnTX-D, -E and portimine-B (Figure 3A).

Comparing cultures with and without agitation, the different treatments did not introduce any notable effect on the production of toxins. There was a tendency for strain V12 under agitation conditions to produce higher levels of Portimine-A and PnTX-F, which are the most prevalent toxins in this study, but this tendency was not confirmed by strain V5 (Figure 3B).

In the overall study, some of the toxins are well correlated, meaning that strains and conditions that produce one toxin tend to produce others at proportional levels. The highest correlation coefficient (>0.94) was observed for PnTX-E and D, which are structural isomers (epimers) sharing the same molecular formula with nearly identical mass spectra. Also, significant correlations were observed for other pairs of toxins, such as Portimine-A and PnTX-F, which are the most abundant in the studied strains, as well as for PnTX-E and -D with Portimine-B, which are the least abundant ones. On the other hand, Portimine-A showed a lower correlation with portimine-B and PnTX-E and -D, suggesting they may be regulated differently (see Supplementary Material, Table S1).

3. Discussion

The presence of the toxic dinoflagellate V. rugosum in Cuba has been pointed out to be the result of a recent introduction of the species, most likely from Japan via ship ballast waters [1]. This dinoflagellate has found favorable conditions to grow and bloom at high cell densities in Cienfuegos Bay. Blooms of this species have been associated with skin lesions and dermatitis in the more confined and sheltered beaches of the Bay.

Few studies have examined the growth dynamics of V. rugosum in batch culture. Abadie et al. (2015, 2016) investigated the effects of different nitrogen sources, as well as temperature and salinity, on this species’ growth [5,9]. In Abadie et al. (2015), no statistically significant differences were observed among nitrate, ammonium, and urea treatments, with average specific growth rates ranging from 0.24 to 0.28 day^−1^ (ammonium: 0.28 ± 0.11; urea: 0.26 ± 0.08; nitrate: 0.24 ± 0.01) [9]. Although the growth rates obtained in the present study were lower (0.144 ± 0.016 to 0.196 ± 0.030 day^−1^), our cultures reached higher cell densities by day 18 (3500–4500 cells·mL^−1^) compared with nitrate-grown cultures in [9], which remained below 3000 cells·mL^−1^. Similarly, Abadie et al. (2016) reported optimal growth between 25 and 30 °C, with maximum cell densities of 3585 and 4252 cells·mL^−1^ and maximum growth rates of 0.39 and 0.20 day^−1^, respectively, which are values more comparable to those observed in our study [5]. Nevertheless, the growth curves described in both studies [5,9] exhibited a longer exponential phase and a more prolonged growth period than those obtained here. These differences may reflect variations in experimental design. In particular, the culture media differed: ENSW is a nutrient-poor medium compared with f/2, and the higher nutrient availability in f/2 can support faster growth and higher biomass. Other culture conditions, such as light intensity, also vary among studies and can influence the carrying capacity of the culture system. For these reasons, comparisons across experiments should be interpreted with caution. Given the limited information available on the growth of V. rugosum in culture, further work is required to clarify its physiological preferences.

Furthermore, the strains from Cienfuegos Bay used in our study exhibited relatively low growth rate values (0.09–0.193 d^−1^), similar to some strains of benthic dinoflagellates species, such as Prorocentrum hoffmannianum (0.18 d^−1^, Cuba) and Prorocentrum sp. (0.19 d^−1^, Brazil) [12]. However, these values were lower than those of other benthic dinoflagellates, such as Amphidinium massartii, A. operculatum, Coolia malayensis, C. monotis, Ostreopsis sp., Prorocentrum lima and P. mexicanum from Brazil, Cuba and the Mediterranean [12,13,14]. The two V. rugosum strains showed different growth rate patterns related to the two treatments. The V5 strain exhibited a similar growth rate for both steady-state and mechanically agitated conditions; however, the V12 strain showed a higher growth rate for the mechanically agitated condition. Therefore, it seems that V. rugosum can proliferate in both agitated and non-agitated conditions in natural environments, although additional environmental factors likely modulate its success [5,9,15]. Evidence of its broad distribution and abundance across diverse habitats worldwide further supports this ecological versatility [16].

In both treatments, the majority of V. rugosum cells in culture appeared as rounded vegetative cysts, commonly referred to as temporary cysts. These structures are typically produced during the stationary phase or under stress conditions, such as exposure to very high or low light intensities, suboptimal temperatures, or nutrient limitation. This life cycle strategy may favor both persistence and dissemination of the species in the environment. Cells can remain dormant in sediments until favorable conditions for germination are reached while simultaneously promoting dispersal by surviving passage through fish digestive tracts and subsequently germinating [7]. In V. rugosum cultures, the non-motile temporary cyst is frequently described as the dominant life form [6,17], a pattern also observed in other dinoflagellates, including Scrippsiella hangoei, Gloeodinium montanum, and several species of Bysmatrum and Fragilidium [17].

The profile of toxins has been determined from natural samples collected during blooms, highlighting the presence of portimines, PnTX-F, -E, -D and traces of -G [1]. The previously established toxin profile of V. rugosum was confirmed and further refined in the present study using two strains cultured under laboratory conditions, isolated from Cienfuegos Bay in 2023 and 2024. Two portimines were identified: portimine-A, which, together with PnTX-F, was found to be the dominant compound, and portimine-B, which was detected at minor levels. In addition, PnTX-E and -D were also detected. Compared with the results of [1], slightly higher cell concentrations were observed in the present study, likely reflecting differences between natural samples and laboratory-cultured strains.

Portimine-A, found here as the most abundant toxin compound, has been shown to exert much lower acute toxicity to mice than pinnatoxins, despite inducing apoptosis in certain cell lines [18]. Nevertheless, portimine-A was recently found to be associated with acute dermatitis in over a thousand of Senegalese fishermen exposed to V. rugosum that attached to drift nets [19]. The same study indicated possible mechanisms for skin inflammation caused by portimine-A acting on ribosome export and inducing inflammasome activation. It also suggested that people with a genetic mutation in the NLRP1 gene are protected against the effects of portimine-A [19]. In addition to portimine-A, portimine-B was also detected in the strains from Cienfuegos Bay tested in this study. However, portimine-B occurred at much lower levels, mostly associated with the stationary phase of the cell culture, when higher levels of secondary metabolites are commonly produced by microalgae. In terms of toxicity, portimine-B seems less potent in inducing cell apoptosis than portimine-A [20].

Pinnatoxin-F was detected in both strains at levels comparable to those of portimine-A. PnTX-F is the most potent pinnatoxin according to studies carried out in mice, either orally or by the intraperitoneal route [21,22,23]. PnTX-F is the precursor compound of PnTX-E, as well as of PnTX-D [21], which explains the good correlations observed between the concentrations of these toxins determined in the V. rugosum strains analyzed.

In Cienfuegos Bay, blooms of V. rugosum have mostly occurred in sheltered and more confined areas of the bay, suggesting that a higher hydrodynamic environment may not favor bloom development. However, in the present study, cultured strains subjected to continuous mechanical agitation did not show significant differences in cell growth or toxin production, suggesting that the intensity of V. rugosum blooms in certain areas of Cienfuegos Bay is unlikely to be associated with calm water conditions. These results indicate that V. rugosum is a highly tolerant species that grows and produces toxins under contrasting conditions, supporting the view that it has a strong potential to disseminate in the environment. Previous studies, carried out on the V. rugosum population from the Ingril Lagoon, France, suggest that temperature and organic nutrients are the main drivers that facilitate their growth [15]. These parameters also characterize the environmental conditions of Cienfuegos Bay [24].

Our results expand current knowledge on the physiology and toxin production of V. rugosum confirming its ecological versatility. These findings highlight the need for continued monitoring, as its physiological plasticity and toxin production under diverse conditions underscore a highly adaptable species with the potential to persist and expand, capable of thriving in dynamic coastal environments, with important implications for future bloom development posing recurrent risks to environmental and public health impacts.

4. Materials and Methods

4.1. Sampling and Sampling Site

Cienfuegos Bay is a semi-enclosed embayment area with estuarine characteristics. The bay has a surface area of 90 km^2^ and an average depth of 14 m. It is connected to the Caribbean Sea by a narrow, 3 km long channel. The northeastern basin of the bay receives most of the anthropic impact from the outfall of Cienfuegos City (129,508 inhabitants) and its industrial area [25]. The southeastern basin is subjected to a lower degree of anthropic pollution originating from the Caonao and Arimao rivers [26]. Weather in this area is characterized by two seasons, comprising one dry (November–April) and one warm rainy (May–October) period [27].

Water samples were collected during a V. rugosum bloom in May 2023 from the bathing area “Círculo Juvenil” (22°7′25″ N–80°27′11″ W), outer part of Cienfuegos city, located on the Cienfuegos Bay, central-southern Cuba (Figure 4). Samples were placed in 500 mL plastic bottles and maintained fresh. The bathing area “Círculo Juvenil” is characterized by a sandy–rocky shallow nearshore part and patches of seaweeds (mainly the brown Padina sanctae-crucis and the red Acanthophora spicifera, Hypnea spinella and Palisada perforata) and the seagrass Halodule wrightii in the inner parts.

4.2. Cell Isolation and Culturing

On arrival to the laboratory, single V. rugosum motile cells were isolated from water samples by a capillary pipette under an inverted microscope and individually transferred to a 96-multiwell plate containing autoclaved filtered seawater enriched with f/2 culture medium [28] at a practical salinity of Sp = 33. The culture plate was incubated at 26 °C, 100–140 μmol photons m^−2^ s^−1^, and a light:dark cycle of 12 h:12 h in static condition. As the cell abundance of the cultures increased, monoclonal cultures of V. rugosum were subsequently transferred to 24- and 6-well tissue culture plates and then to 50 mL glass Erlenmeyer flasks containing fresh f/2 seawater medium under the same culture conditions.

4.3. Growth Curves and Experimental Conditions

The two strains of V. rugosum (Strains V5 and V12) were cultured under both steady-state and mechanically agitated conditions. For each condition, 250 mL T-flasks were filled with f/2 medium with 33 of salinity and inoculated with stock cultures to achieve an initial concentration of approximately 500 cells·mL^−1^ at a final volume of 250 mL. Three biological replicates per strain and condition were incubated in a controlled culture chamber (Fitoclima 600PL, Aralab, Sintra, Portugal) at 50 µmol m^−2^ s^−1^ under a 12:12 light–dark cycle at 26 °C. Cultures assigned to the agitation treatment were placed on a plankton wheel rotating at 400 RPM. Growth was monitored by collecting 1 mL aliquots every 3 days until day 10 (240 h) and every 4 days after that for a total of 18 days (432 h). The 1 mL aliquots collected during the growth experiment were fixed with Lugol iodine solution, and cell concentrations were determined using a Sedgwick-Rafter counting chamber under an inverted microscope (Leica Microsystems CMS GmbH, Wetzlar, Germany).

4.4. Toxin Analysis

Toxin extraction and analysis was carried out as described in [29]. V. rugosum cells from cultured samples were harvested by centrifugation at 500 rcf for 15 min at 12 °C using an Eppendorf 5810 centrifuge (Eppendorf, Hamburg, Germany). Following centrifugation, the culture medium (supernatant) was separated from the cell pellet, and both fractions were stored at −20 °C until analysis. All chemicals and reagents used for toxin extraction and analysis were of analytical grade or higher. Methanol (>99.8%, p.a.), formic acid (98–100%) and acetonitrile (analytical grade) were obtained from Sigma-Aldrich (Sintra, Portugal); ammonium formate (>99% purity) was from Fluka (Oeiras, Portugal), and water was purified using a Milli-Q 185 Plus system from Millipore (Oeiras, Portugal). To prepare the toxin extracts, the cell pellet was resuspended in 4 mL of methanol and subjected to sonication for 2 min at 25 W with a 50% pulse duty cycle using a Vibracell sonicator (Sonic & Materials, Newtown, CT, USA) while being kept in an ice bath. The resulting suspension was centrifuged at 2500 rcf for 5 min at room temperature using an Eppendorf 5810 centrifuge (Hamburg, Germany). An aliquot of 1 mL of the supernatant was subsequently filtered through 0.20 μm cellulose filters directly into vials for analysis.

The extracts were analyzed by LC-MS/MS for the detection and quantification of pinnatoxins and portimines. The LC-MS/MS equipment used consisted of an Agilent 1290 Infinity chromatograph coupled to an Agilent 6470 triple quadrupole mass spectrometer (Agilent Technologies, Waldbronn, Germany). Chromatographic separation was achieved on a Zorbax SBC8 RRHT column (2.1 × 50 mm, 1.8 μm), equipped with a guard column (2.1 × 5 mm, 1.8 μm). Mobile phase A consisted of purified water with 2 mM ammonium formate and 50 mM formic acid, while mobile phase B was composed of 95% acetonitrile and 5% MilliQ water supplemented with 2 mM ammonium formate and 50 mM formic acid. Elution was carried out at a flow rate of 0.4 mL min^−1^ using the gradient program 0–3 min, 88% to 50% A; 3–6.5 min, 50% to 10% A; 6.5–8.9 min, held at 10% A; and 8.9–10 min, 10% to 88% A to re-establish initial conditions. Detection was conducted in accordance with the Standardized Operating Procedure of the European Reference Laboratory for Marine Biotoxins for the determination of marine lipophilic biotoxins [30]. Two multiple-reaction-monitoring (MRM) transitions were monitored in positive polarity mode for both pinnatoxins and portimines (all details of the MRM transitions are provided in the Supplementary Material, Table S2). The optimized ion source parameters were set as follows: gas temperature 150 °C, gas flow 13 L min^−1^, nebulizer 50 psi, sheath gas temperature 400 °C, sheath gas flow 12 L min^−1^ and capillary voltage 2500 V. Quantification was performed using a five-point calibration curve prepared with PnTX-G (National Research Council, Ottawa, Canada), over a concentration range of 0.5–12 ng mL^−1^, yielding a correlation coefficient greater than 0.990. The lowest calibration level was defined as the limit of quantification (LOQ). All toxins, pinnatoxins and portimines, were indirectly quantified against the PnTX-G standard, assuming that all compounds have a similar response factor to PnTX-G.

4.5. Statistics

All statistical analyses were done in JASP (v0.95.4, Amsterdam, The Netherlands). For toxin analysis, a significance level of α < 0.01 was considered. Assumptions of normality and equality of variance were verified using the Shapiro–Wilk test and Levene’s test, respectively, for analysis of variance (ANOVA). Comparisons among pairs of means were conducted using Tukey’s post hoc tests. Pearson’s correlation analysis was used to assess the correlations between toxin concentrations throughout the study.

For growth curve analysis, Welch’s t-test was performed for unequal variances. Comparisons were made independently at each time point to evaluate growth treatment differences within each strain; p ≤ 0.05 was used, corresponding to a 95% confidence interval. Growth rate (µ) was calculated according to [31] considering the cell densities between day 1 and day 10 of the growth curve:

$$[eqn]$$

Doubling time (DT) and percentage of increase (% IC) were calculated based on the growth rate as follows:

$$[eqn]$$ $$[eqn]$$