Microalgae are an important component of the aquatic environment (Elser et al., 2007), and are key players in primary productivity and biogeochemical cycles (Sarthou et al., 2005). They are rich in nutrients and organic compounds, and hence are used as aquaculture feeds, health supplements and alternative energy sources (Becker, 2007). Moreover, microalgae are a diverse assemblage of both autotrophs and heterotrophs, have substantial biomass, and are abundant in the marine ecosystem (Shi et al., 2011). They are especially useful as bioindicators of environmental changes, for both short and long-term environmental monitoring as well as ecotoxicology assessments (Franklin et al., 2002). Owing to the involvement of microalgae in the global cycling of toxic chemicals in the aquatic environment, monitoring the effects of chemicals, such as metals and endocrine disruptors, on microalgae is of considerable importance (Torres et al., 2008).

Algae-based bioassays are commonly employed in environmental risk assessments to assess the toxicity of metals, novel chemicals and other emerging contaminants and to establish regulatory guidelines (Stauber and Davies, 2000). Algal toxicity tests routinely use freshwater green algae (e.g., Chlamydomonas sp., Chlorella vulgaris, Pseudokirchneriella subcapitata, and Scenedesmus subspicatus) and the diatom Navicula pelliculosa (Franklin et al., 2007). However, in aquatic ecosystems, algae comprise diverse taxa, including green algae, cyanobacteria, diatoms, and dinoflagellates (Shi et al., 2011). Each of these taxa responds differently to chemical toxicants; therefore, it is mandatory to conduct tests on a wide range of species representing different classes to determine safe discharge guidelines. Moreover, the available algae-based toxicity data were determined using freshwater algae (Nalewajko and Olaveson, 1998; Sverdrup et al., 2001); relatively little emphasis has been placed on marine algae, hence the need to assess the responses of marine species to various toxicants.

In the present study, we quantified the sub-lethal effects of metals and endocrine-disrupting chemicals (EDCs) on the marine green alga Tetraselmis suecica. The test toxicants included Cu, Ni, Pb, bisphenol A (BPA), endosulfan (ES), and polychlorinated biphenyl (PCB). Finally, we calculated the median effective concentrations (EC₅₀) of the chemicals tested and discussed their significance in terms of an environmental risk assessment. T. suecica is a marine green alga belonging to Chlorodendrophyceae, Chlorophyta. It has a high growth rate, and it has been used commercially as an aquaculture feed and as a nutritional supplement (Brown, 2002). In addition, T. suecica is considered a potential candidate for biodiesel production (Montero et al., 2011). Thus determining its potential as a model for toxicity assays could be advantageous.

T. suecica (P009) was obtained from the Korea Marine Microalgae Culture Center (Pukyung National University, Busan, Korea). The cells were cultured in f/2 medium, using filtered seawater with additional macronutrients, vitamins and trace metals (CuSO₄, ZnSO₄, CoCl₂, MnCl₂, and NaMoO₄) according to Guillard and Ryther (1962). The cells were maintained at 20℃, under a 12:12-h light: dark cycle with a photon flux density of ca. 65 μmol photons/m²/s.

In this study, three metals (Cu, Ni, and Pb) and three EDCs (BPA, ES, and PCB) were selected as test toxicants. The concentrations of each were chosen based on EC₅₀ values reported for other aquatic organisms (Millan de Kuhn et al., 2006; Soto-Jimenez et al., 2011). Accordingly, a range of concentrations of each chemical was prepared, as described below.

For Cu (as CuSO₄; Cat. No. C1297, Sigma, St. Louis, MO, USA), the concentrations used were 0.5, 1.5, 3, 6, 12, 25, 75, 150, 500, and 750 mg/L. For Pb (as PbCl₂; Cat. No. 268690, Sigma), the chosen concentrations were 0.5, 1.5, 5, 10, 25, 50, 100, 150, 250, 500 and 750 mg/L. For Ni (as NiCl₂; Cat. No. 339350, Sigma), the concentrations used were 0.1, 0.5, 1, 5, 10, 50, 100, 250, and 500 mg/L. All test concentrations were higher than that added to the f/2 medium. For example, amongst the four test metals, only 0.0068 mg/L CuSO₄ was added to the f/2 medium, and thus the medium-containing metals or EDCs contained negligible endpoint concentrations.

For BPA (Cat. No. A133027, Sigma), concentrations of 0.5, 1, 2.5, 7.5, 15, 25, 50, 75, and 100 mg/L were prepared from a stock solution. BPA was dissolved in 10% dimethyl sulfoxide. For ES (Cat. No. 36676, Sigma), the concentrations used were 0.001, 0.01, 0.1, 0.5, 1 and 10 mg/L. PCBs were prepared from Aroclor 1016 (Cat. No. 48701, Sigma) at 0.001, 0.01, 0.05, 0.1, 0.5, 1, 10, 20, and 50 mg/L. All dilutions were from standard stock solutions.

Fifty-milliliter aliquots of the algal culture, comprising cells in the exponential phase of growth, were transferred into sterile tubes. The chemicals, at the concentrations mentioned above, were transferred into test tubes in duplicate. The initial cell concentration was 5.5 ± 0.1 × 10⁵/mL. and samples were withdrawn for cell counts and chlorophyll a (Chl a) estimation at 0, 12, 24, 48, and 72 h.

Cells were enumerated using a hemocytometer (Marienfeld GmbH, Lauda, Germany). Cell counts were plotted against exposed time as log₁₀ values. In addition, Chl a was measured by concentrating a 10 mL sample of the culture at various time points. The pigments were extracted with 90% acetone after incubation overnight in the dark. The supernatants extracted were measured using a DU730 Life Science UV/Vis spectrophotometer (Beckman Coulter, Fullerton, CA, USA). The Chl a concentrations were estimated following Parsons et al. (1984).

The median effective concentration (EC₅₀) and the percentile growth inhibition were calculated as recommended in the Organisation for Economic Cooperation and Development (OECD) test guidelines (OECD, 2011). The 72-h EC₅₀ values were estimated using a sigmoidal dose-response curve plotted in Origin 8.5 (MicroCal Software Inc., Northampton, MA, USA) based on the sigmoidal four parameter equation (Teisseyre and Mozrzymas, 2006): Log EC₅₀ = a + (b ？ a)/[1 + 10(x ？ c)^d], where a is the response value at zero or minimum asymptote, b is the response value for an infinite concentration or maximum asymptote c is the mid-range point, d is the steepness of the curve or the Hill slope and x is the dilution coefficient.

All data presented are the mean values of duplicate determinations. A one-way analysis of variance (ANOVA) with post hoc Student’s Newmann Keul’s test in Graphpad InStat (Graphpad Software, Inc., La Jolla, CA, USA) was used for comparisons between non-treated and treated cultures. P < 0.05 was accepted as significant. In addition, correlations between cell counts and Chl a using Pearson’s correlation coefficient (r²) were calculated using an Excel spreadsheet (Microsoft Corporation, Redmond, WA, USA).

[Fig. 1.] Variation in cell count of Tetraselmis suecica following exposure to metals. (A, C, E) Variation in cell numbers following exposure to metals. (B, D, F) Variation in cell count and chlorophyll a following 72-h exposure to metals.

Exposure of T. suecica to metals (Cu, Ni, and Pb) induced a wide range of responses, depending on the toxicant concentrations (Fig. 1). In all cases, the initial experimental concentrations used (i.e., 0.1-10 mg/L) of the metals had very little or no effect on the cell counts or Chl a levels. However, when cells were exposed to higher concentrations of metals (50-750 mg/L), significant reductions (P < 0.001) were recorded. We observed no significant changes in T. suecica cells exposed to levels up to 150 mg/L Cu and up to 10 mg/L Pb and Ni (Fig. 1A, 1C, and 1E). However, above these concentrations, significant reductions (P < 0.001) in cell count and Chl a level were observed for all of the tested metals (Fig. 1B, 1D, and 1E).

[Fig. 2.] Variation in cell count of Tetraselmis suecica following exposure to endocrine-disrupting chemicals (EDCs). (A, C, E) Variation in cell numbers following exposure to EDCs. (B, D, F) Variation in cell count and chlorophyll a following 72-h exposure to EDCs. PCB, polychlorinated biphenyl.

Additional toxicity tests for the three EDCs (BPA, ES, and PCB) were performed over a wide range of concentrations (Fig. 2). BPA was administered at concentrations of 0.5-100 mg/L. T. suecica cells were not significantly affected (P > 0.05) at lower BPA concentrations (~25 mg/L). However, higher concentrations of BPA (>50 mg/L) caused a significant decrease in the cell count (79-100%). Exposure of T. suecica to the lower concentrations of ES (0.001-0.05 mg/L) led to a 14-30% reduction in cell count, but there was a highly significant (P < 0.0001) reduction at higher concentrations (0.1-10 mg/L). PCB exposure followed a similar trend, with concentrations greater than 1 mg/L significantly decreasing both cell numbers and Chl a levels.

[Fig. 3.] Dose response curve of Tetraselmis suecica following 72-h exposure to metals and endocrine-disrupting chemicals. (A) Copper, (B) bisphenol A (BPA), (C) lead, (D) endosulfan (ES), (E) nickel, and (F) polychlorinated biphenyl (PCB).

Table 1 shows the Pearson’s correlations between cell count and Chl a level; the data for both endpoints were valid and correlated in most experiments. In this study, we calculated 72-h EC₅₀ values (Table 2) using the cell count data and plotted sigmoidal dose-response curves for T. suecica exposed to metals and EDCs (Fig. 3). The EC₅₀ values for the metals were as follows: EC₅₀ (Cu) = 43.03 ± 0.154 mg/L, EC₅₀ (Pb) = 9.62 ± 0.135 mg/L, and EC₅₀ (Ni) = 16.11 ± 0.187 mg/L. The EC₅₀ values of the EDCs were as follows: EC₅₀ (BPA) = 15.55 ± 0.274 mg/L, EC₅₀ (ES) = 0.045 ± 0.032 mg/L, and EC₅₀ (PCB) = 3.96 ± 0.180 mg/L.

Toxicity assessments using marine species can be challenging compared with assessments using freshwater species. This is because the marine environment can have a more profound influence on toxicity evaluation than freshwater ecosystems. In addition, the higher ionic strength and buffering capacity of seawater can alter the bioavailability of discharged chemicals due to complex chemical reactions and the subsequent formation of by-products (Moffett and Zika, 1987). As noted previously, relatively few species can be described as “standard” for marine algae, although several guidelines have been published and some marine species have been recommended as model species (Nalewajko and Olaveson, 1998; Sverdrup et al., 2001). In this study, we present additional toxicity data for various metals and EDCs to the marine green alga T. suecica.

Discharge guidelines for metals in the marine environment are as follows: Cu, 0.5 mg/L; Pb and Ni, 0.1 mg/L (United States Environmental Protection Agency, 1996). Individual discharges of Cu, Pb, or Ni confined to the stipulated levels should have very little, if any, effects on cell counts or Chl a levels in T. suecica (see Figs. 1 and 3). In addition, compared to most of the frequently used freshwater green algae, such as Chlorella vulgaris, Selenastrum capricornutum, Pseudokirchnereiella subcapitata, etc., T. suecica is highly resistant to these commonly used metal pollutants. For example,

[Table 1.] Pearson’s correlation between cell count and Chl a level in Tetraselmis suecica cells following exposure to toxic chemicals

[Table 2.] 72-h EC50 values of chemicals exposed to Tetraselmis suecica

the 72-h EC₅₀ value for Cu was 0.016 mg/L in S. capricornutum at a similar cell density (Franklin et al., 2002) to that used in this study. Vasseur et al. (1988) reported that the EC₅₀ values for Cu and Zn in S. capricornutum were 10 μg/L and 90 μg/L, respectively. However, Millan de Kuhn et al. (2006) reported markedly higher EC₅₀ values for Cu in various marine algae than freshwater algae. The authors recorded EC₅₀ values of Cu of 220 mg/L for the green alga Dunaliella salina, 34.0 mg/L for the flagellate Euglena gracilis, 13.5 mg/L for the dinoflagellate Heterocapsa triquetra, and 7.0 mg/L for the dinoflagellate Prorocentrum minimum, respectively. Our previous studies also showed high EC₅₀ values for Cu (12.74 mg/L) and Pb (46.70 mg/L) in the marine dinoflagellate Cochlodinium polykrikoides (Ebenezer and Ki, 2012). These data suggest that marine algae may be more tolerant to metal exposure than freshwater species, particularly at the level of the safe discharge standards recommended by the United States Environmental Protection Agency (US EPA) (US EPA, 1996).

In addition, comparisons of available EC₅₀ data revealed that our test species was generally more tolerant than other freshwater algae, including Desmodesmus subspicatus and Chlorella kessleri (Pavli？ et al., 2006). The high tolerance of T. suecica is generally in accordance with previous reports stating that T. suecica was more tolerant and bioaccumulative upon exposure to metals (Perez-Rama et al., 2002) than other algae. Millan de Kuhn et al. (2006) reported an EC₅₀ (Cu) value of 40 mg/L in T. suecica. Moreover, Debelius et al. (2009) found that Tetraselmis chuii [EC₅₀ (Cu) = 0.33 mg/L and EC₅₀ (Pb) = 2.6 mg/L] was more tolerant in toxicity assessments than the other marine microalgae tested, including the diatom Chaetoceros sp. and green algae Rhodomonas salina, Isochrysis galbana, and Nannochloropsis gaditana. Overall, previous reports and our findings suggest that the marine algae genus Tetraselmis is more tolerant to metals than other algae.

To date, the available data regarding the toxicity of EDCs to algae is limited compared to metals. One reason for this is that algae do not have an endocrine system, and thus may show limited effects of exposure to EDCs. However, recent studies have shown that most EDCs, such as BPA, ES, PCB or metolachlor, do have toxic effects on algae (Liu et al., 2010; Ebenezer and Ki, 2012), in particular by damaging photo system II energy fluxes in chloroplasts (Perron and Juneau, 2011). In contrast, the data from this study indicated that T. suecica was relatively tolerant to BPA and PCB (Table 2), although not to ES, when compared with other algae. For example, the EC₅₀ values of BPA were recorded as 3.73 mg/L and 7.96 mg/L for the diatoms Navicula incerta and Cyclotella caspia, respectively (Li et al., 2009; Liu et al., 2010). The 96-h EC₅₀ value of PCBs for the dinoflagellate Lingulodinium polyedrum was 0.122 mg/L (Leitao et al., 2003). In the marine dinoflagellate Cochlodinium polykrikoides, the 72-h EC₅₀ of PCB was 1.07 mg/L (Ebenezer and Ki, 2012). Alternatively, the EC₅₀ value of ES for the green alga Pseudokirchneriella subcapitata was 0.427 mg/L (De Lorenzo et al., 2002), thus T. suecica was more sensitive [EC₅₀ (ES) = 0.045 mg/L]. EC₅₀ values for the EDCs tested were mostly higher than the environmental discharge limits set by the US EPA for coastal waters, which would have had no effect on the marine algae. The US EPA guidelines set safe discharge levels of 0.09 μg/L for BPA (Kolpin et al., 2002), 0.034 μg/L for ES (Agency for Toxic Substances and Disease Registry, 2000) and 0.02 μg/L for PCBs (Nagpal et al., 2006). Although these compounds do not pose a risk as acute toxicants, their chronic effects can be deleterious. The high tolerance of T. suecica to the EDCs is in accordance with previous metal toxicity data.

In conclusion, T. suecica exhibited a dose-dependent response upon exposure to selected metals and EDCs. This species was generally tolerant to most of the chemicals at their permissible concentrations. According to the EC₅₀ values obtained, in terms of the metals tested, T. suecica was most sensitive to Pb (9.62 mg/L) and most tolerant to Cu (43.03 mg/L). Of the three tested EDCs, T. suecica was most sensitive to ES (0.045 mg/L) and most tolerant to BPA (15.55 mg/L). These data indicate that T. suecica may be relatively tolerant of toxic chemicals compared with other marine algae, and has a markedly higher tolerance than freshwater algal species.