Anthracene phytotoxicity in the freshwater flagellate alga Euglena agilis Carter

The freshwater flagellate alga Euglena agilis Carter was exposed to the polycyclic aromatic hydrocarbon (PAH) anthracene for 96 h under optimal photosynthetically active radiation (PAR), and responses of growth, photosynthetic pigment production, and photosynthetic efficiency were assessed. Anthracene reduced the growth rate (μ) and levels of chlorophyll a (Chl a), chlorophyll b (Chl b), and total carotenoids. The growth rate was more sensitive than photosynthetic parameters, with a median effective concentration (EC50) of 4.28 mg L−1. Between 5 and 15 mg L−1, anthracene inhibited the maximum quantum yield (Fv/Fm) of photosystem II (PSII) and the maximum photosynthetic electron transport rate through PSII (rETRmax) with EC50 values of 14.88 and 11.8 mg L−1, respectively. At all anthracene concentrations, intracellular reactive oxygen species (ROS) were elevated, indicating increased oxidative stress. Anthracene presumably reduced the PSII efficiency of photochemical energy regulation and altered the photochemistry through intracellular ROS formation. Acute exposure to PAHs may induce severe physiological changes in phytoplankton cells, which may influence vital ecological processes within the aquatic environments. Additionally, growth and Chl a content may serve as sensitive risk assessment parameters of anthracene toxicity in water management since EC50 values for both overlap with anthracene levels (8.3 mg L−1) permitted by the US Environmental Protection Agency (USEPA).


Materials and Methods
Algal test species and culture conditions. Euglena agilis Carter was cultured in mineral medium (pH 5) 26  Test chemicals and exposure. Anthracene (99% purity, CAS No. 120-12-7) was purchased from Sigma Aldrich (Saint Louis, MO, USA) and test solutions at the desired concentrations were prepared by serial dilution from stocks in high-performance liquid chromatography (HPLC)-grade dimethyl sulphoxide (DMSO; Sigma Aldrich). Microplate toxicity tests of 96 h in duration were conducted in 24-well cell culture plates (well diameter = 15.6 mm, growth area = 1.9 cm 2 ; SPL Life Sciences, Gyeonggi-Do, Korea) with a test volume of 2 mL per well. Equal volumes of cell suspension and anthracene stock solutions were mixed to obtain final concentrations of 0.625, 1.25, 2.5, 5, 10 and 15 mg L −1 , along with untreated controls. The concentration of the carrier solvent did not exceed 0.2% v/v of the test culture volume. The initial cell density was 10 ± 0.5 × 10 4 cells mL −1 of suspension. An additional solvent toxicity test (96 h) was conducted with a maximum DMSO concentration of 0.2% v/v. Organisms were exposed to nominal concentrations of anthracene and all treatments were performed in triplicate. The well plates were covered with parafilm to avoid evaporation and mixing of the volatile toxicant. www.nature.com/scientificreports www.nature.com/scientificreports/ Measurement of growth rate. Growth rates were determined by measuring the number of cells in each well on the first and final days using a hemocytometer (Marienfeld, Germany). The specific growth rate (μ) was calculated using the following formula: where N 1 and N 2 are the number of cells at time t 1 (initial) and t 2 (final), respectively.
Estimation of photosynthetic pigments. Photosynthetic pigment content was estimated using standard protocols 27 . Briefly, 1 mL cell suspension was collected from each replicate culture and centrifuged before extraction, and 1 mL of 90% v/v acetone was added followed by vigorous vortexing and centrifugation at 10,000 × g for 5 min at 4 °C. Supernatants were withdrawn and their optical density was measured spectrophotometrically at 470, 664 and 647 nm using an S-3100 UV/Vis spectrophotometer (Scinco, Seoul, Korea). Pigment concentrations are expressed as μg mL −1 of suspension. is the maximum light-acclimated fluorescence yield and F is the light-acclimated fluorescence yield) by photon flux density (PFD) and plotting against PFD. The ETR is relative because the absorbance of light by cells was not measured. Maximum electron transport rate (ETR max ) was derived from the hyperbolic tangent formula rETR = ETR max * tanh (α/I/ ETR max ), adapted from Jassby and Platt 28  Statistical analyses. Data are presented as means ± 95% confidence intervals (CI). All parameters were compared across treatments with one-way analysis of variance (ANOVA, n = 3, p < 0.05) using the JMP software (JMP ® Pro version 13.1, SAS Institute, USA). Multiple comparison tests based on the least significant difference (LSD) were then carried out to find significant differences (p < 0.05) from controls and between treatments. The effective concentration at which 50% inhibition occurs (EC 50 ) was estimated by the linear interpolation method using ToxCalc 5.0 (Tidepool Science, USA). The coefficient of variation (CV), the standard deviation expressed as a percentage of the mean, was calculated to estimate the precision of test values.

Measurement of chlorophyll a (Chl
Capsule. Anthracene significantly reduces growth and photosynthesis in the freshwater flagellate Euglena agilis via intracellular ROS generation.

Results and Discussion
Effect of anthracene on cell growth. The carrier solvent used in this study (DMSO) had no significant inhibitory effects on cell growth (ANOVA, df = 6, F = 1.3, P > 0.05) or photosynthetic efficiency (ANOVA, df = 5, F = 0.56, P > 0.05) of E. agilis, even at the maximum concentration of 0.2% (v/v) in the growth medium ( Fig. 1). Okumura et al. 31 previously demonstrated the suitability of DMSO as a carrier solvent in Euglenoid tests.
Most of the toxicity data available for the effects of anthracene on freshwater microalgae are based on growth inhibition (Table 1). Growth is an important endpoint parameter that reflects the overall vitality of a population under the tested conditions. Addition of anthracene to the culture medium resulted in a concentration-dependent inhibition of the specific growth rate of E. agilis (Fig. 2). Compared with controls, the final day cell densities were significantly lower at all tested concentrations, and μ was significantly reduced from 0.53 for control cells to 0.12 at the highest anthracene dose (ANOVA, df = 6, F = 198.82, P < 0.001). The EC 50 value for growth was 4.28 mg L −1 ( Table 2), which is greater than the values previously reported for several freshwater microalgae (Table 1). At nominal concentrations exceeding 0.05 mg L −1 , anthracene significantly inhibits the growth of freshwater phytoplankton 32 . For example, the growth of Selenastrum capricornutum was extremely sensitive to anthracene, aggravated by UV radiation, with a 22 h EC 50 value of 3.9-37.4 μg L −1 33 . Our results suggest that anthracene itself is (2019) 9:15323 | https://doi.org/10.1038/s41598-019-51451-y www.nature.com/scientificreports www.nature.com/scientificreports/ potentially toxic to freshwater primary producers, and could add synergistic effects with other stressors such as UV radiation 32 .
Growth inhibition due to PAH exposure in microalgae and higher aquatic plants has been previously reported 34 , and the extent of growth inhibition depends on the species studied, the chemicals tested and the duration of exposure. Reduction in growth can result from an accumulation of anthracene within the lipid fraction of cells and subsequent changes in membrane properties 35 . PAH accumulation in membranes can cause an expansion of the membrane surface area, inhibition of primary ion pumps, and an increase in proton permeability, leading to dissipation of the electrical potential and pH gradient, which ultimately results in inhibition of cellular growth 36 . Additionally, a reduction in photosynthesis can lead to impaired growth, since these are highly interrelated phenomena, each being a function of the utilization of energy from light and nutrients. Even moderate changes in the function of the photosynthetic apparatus can lead to a marked reduction in energy production within chloroplasts 34 . Effect of anthracene on pigment content. Euglena contains both Chl a and b as light-harvesting pigments, along with the carotenoids, diadinoxanthin, and diatoxanthin 37 . Despite studies on the effect of anthracene on growth and photosynthesis in algae, limited information is available on their interference with photosynthetic pigment production. Anthracene did not affect chlorophyll biosynthesis in Chlamydomonas reinhardtii strain cw92 at concentrations up to 1 mg L −1 38 , or in three Desmodesmus spp. up to 0.25 mg L −1 37 . However, in the present study, anthracene (>0.625 mg L −1 ) had a pronounced effect on photosynthetic pigments content in E. agilis. The most abundant pigment in E. agilis was Chl a (7.14 μg mL −1 ) followed by carotenoids (1.72 μg mL −1 ) and Chl b (1.25 μg mL −1 ). At the lowest test concentration (0.625 mg L −1 ), there were significant reductions in Chl a, Chl b and total carotenoids of up to 20%, 16%, and 17%, respectively, while at the highest concentration, reductions of 58%, 64%, and 49% were observed (Fig. 3). The adverse effect on pigment content was concentration-dependent, with 96 h EC 50 values of 5.59 mg L −1 , 8.14 mg L −1 and >15 mg L −1 for Chl a (ANOVA, df = 6, F = 334.54, P < 0.05), Chl b (ANOVA, df = 6, F = 40.05, P < 0.05) and total carotenoids (ANOVA, df = 6, F = 130.11, P < 0.05), respectively ( Table 2).
The molecular mechanism of the reduction in pigment levels may involve the accumulation of lipophilic anthracene in thylakoid membranes 39 , resulting in conformational changes in their structure and composition. In general, reduced pigmentation under chemical stress results from inhibition of enzymes related to chlorophyll synthesis, degradation of chlorophyll and DNA damage 40 , or accelerated degradation of pigments due to increased ROS formation at various positions in the photosynthetic electron transport chain. Moreover, carotenoids prevent photo-oxidative destruction of chlorophylls 41 and, therefore, a reduction in carotenoids could have  www.nature.com/scientificreports www.nature.com/scientificreports/ additional serious consequences on chlorophyll molecules. The simultaneous reduction in all three photosynthetic pigments suggests that the major target of anthracene toxicity is the thylakoid compartment of chloroplasts. These results also indicate that in addition to causing a severe reduction in growth, anthracene exposure may reduce photosynthetic performance via the destruction of pigments responsible for harvesting available photons.

Inhibition of photosynthesis. Anthracene is a strong inhibitor of phytoplankton photosynthesis
in vivo 32,35,38,42 . We conducted in vivo Chl a fluorescence measurements as an intriguing tool to reveal the toxic effects of anthracene on the photosynthetic machinery of E. agilis (Fig. 4). The quantum yield and quantum efficiency parameters are indicators of the efficiency of solar energy absorption, which decreases under chemical stress, implying that stress negatively impacts photon absorption and conversion of solar energy during photosynthesis 43 . We found that at higher anthracene concentrations (>5 mg L −1 ), there were significant reductions in dark fluorescence (F 0 ; ANOVA, df = 6, F = 40.90, P < 0.05), which reflects emission by excited Chl a molecules in the antennae structure of PSII, and in maximal fluorescence (F m ; ANOVA, df = 6, F = 33.54, P < 0.05), which can be attributed to severe loss of pigments and/or inactivation of PSII reaction centres. It is evident that at higher anthracene doses, reduction in the number of cells and pigment levels resulted in an overall decline in the light-harvesting by E. agilis. Moreover, no significant variation (ANOVA, df = 3, F = 2.01, P > 0.05) in F 0 was observed between 0-2.5 mg L -1 anthracene, despite significant reductions in the concentration of Chl a (Fig. 3a), suggesting that pigment molecules associated with PSII reaction centres are less affected. Instead, anthracene may pose a more serious threat to the pigment pool of PSI. This interpretation is supported by the findings of Huang et al. 44 , who suggested that PSI is the primary site of action of anthracene. However, Chl a fluorescence measurements in plants and algae have suggested inhibition of the cytochrome-b6/f complex and/or photo-oxidative damage to PSII as additional modes of anthracene toxicity 6,38 .
F v /F m , an estimate of the photochemical conversion efficiency of PSII in the dark, has been widely used to assess the acute toxicity of aromatic hydrocarbons in freshwater plants and phytoplankton 6,35 . An F v /F m value of ~0.55 relative units (RU) was recorded in our control E. agilis population, comparable to the value reported previously for Euglena gracilis 21 . F v /F m did not significantly vary up to 1.25 mg L −1 anthracene (Fig. 4c). However, at higher concentrations, F v /F m was declined (ANOVA, df = 6, F = 63.18, P < 0.05) by 17% (5 mg L −1 ), 36% (10 mg L −1 ), and 55% (15 mg L −1 ) with an EC 50 of 13.74 mg L −1 ( Table 2). Toxicity of anthracene on F v /F m in microalgae taxa has not been reported previously, so direct comparison of the sensitivity of E. agilis with other species is not possible. Nevertheless, in the macrophyte Lemna gibba, F v /F m appeared to be a more sensitive biomarker of anthracene toxicity, with a 4 h EC 50 value of 2 mg L −1 44 .
We noted that the reduction in F v /F m at >2.5 mg L −1 anthracene was accompanied by a significant loss of NPQ (Fig. 4d). Although values were not statistically significant, NPQ tended to increase up to 1.25 mg L −1 and then decreased significantly thereafter (ANOVA, df = 3, F = 44.74, P < 0.05). This decline in Chl a fluorescence quenching can be attributed to impairment of electron transport downstream from PSII and an elevated reduction of the PQ pool 45 . NPQ is produced through the generation of an H + electrochemical gradient across the thylakoid membranes 46 and is an indicator of absorbed energy that is dissipated through heat loss and other non-photochemical mechanisms. We assume that a severe reduction in the photosynthetic process at high anthracene levels likely reduces the magnitude of the pH gradient, thereby affecting NPQ values, consistent with the view of González-Moreno et al. 47 who reported similar observations on fluorescence quenching in E. gracilis under salt stress. Severe reduction in F v /F m of E. gracilis exposed to NaCl had resulted in a diminution of the pH gradient across the thylakoid membranes and subsequently, up to 95% reduction in NPQ 47 .
In the present study, anthracene reduced the ETR across PSII at higher concentrations (Fig. 4e). rETR is an empirical estimate of the rate of the flow of electrons through the electron transport chain. rETR max was not significantly affected up to 1.25 mg L −1 anthracene, whereas 5, 10 and 15 mg L −1 doses resulted in 18%, 46%, and 56% reductions, respectively, with an EC 50 value of 11.8 mg L −1 (Table 2). Thus, the threshold value of anthracene for a significant reduction in ETR was higher than that for growth inhibition. In aquatic plants, PAHs inhibit photosynthetic electron transport at concentrations below which growth and CO 2 fixation are inhibited 44 . For PAHs in general, the target of their toxicity to photosynthesis is the electron transport downstream from PSII, specifically www.nature.com/scientificreports www.nature.com/scientificreports/ at cytochrome-b6f. Inhibition of electron transport blocks reoxidation of the reduced plastoquinone pool (PQH 2 ) and the absorbed energy cannot be used in photochemistry 6 . A probable consequence of inhibition of the electron transport chain at PSII is the transfer of energy from triplet chlorophylls to oxygen, forming singlet oxygen species, which induces oxidative damage of cells 48 . The generation of free radicals and subsequent intracellular oxidative stress is a prominent mechanism of anthracene toxicity in freshwater phytoplankton 38 .
When DCFH-DA fluorescence emission in anthracene exposed E. agilis cells was measured, a significant increase was observed at all dosages (ANOVA, df = 6, F = 81.11, P < 0.05), indicating a rise in intracellular ROS levels (Fig. 5). ROS level at 2.5 mg L −1 anthracene was almost double than that in the controls. The subsequent reduction in fluorescence at high anthracene (>5 mg L −1 ) can be attributed to reduced cell growth and diminished www.nature.com/scientificreports www.nature.com/scientificreports/ enzyme activities, although values were still significantly higher (p < 0.5) than in controls. The major site of ROS production in photosynthetic organisms is the disrupted electron transport chain across PSII 49 . We report here, for the first time in freshwater microalga taxa, the significant elevation of intracellular ROS levels under anthracene stress. In Euglena spp., ROS play a significant role in metal toxicity 45 , UV damage and defense mechanisms 50 . However, ROS, generated by chemical stressors, trigger adverse effects through multifaceted actions inside the cell. They attack thylakoid lipids and initiate peroxyl radical chain reactions, eventually destroying membranes and pigment-protein complexes 45 . Moreover, in chloroplasts, ROS cause lipid peroxidation, which results in the disruption of photosynthetic pigments, and the inactivation and degradation of RuBisCo and other components of the Calvin cycle 51 . Our results correspond to Babu et al. 52 , who found that 1,2-dihydroanthraquinone, a photoproduct of anthracene, inhibited photosynthetic electron transport, leading to the overproduction of O 2 − and subsequent oxidation of proteins, membranes, and pigments in Lemna gibba.
DCFH-DA is more suitable to estimate total ROS production rather than as a probe for a particular type of ROS 53 . The superoxide anion radicals (O 2 − ) produced in the electron transport chain is a precursor for many other ROS species. They are rapidly converted to hydrogen peroxide (H 2 O 2 ) and subsequently to hydroxy radicals (OH • ) by enzymatic reactions 54 . DCFH-DA does not directly react with O 2 − but can be oxidized to highly fluorescent DCF by H 2 O 2 and OH • radicals 55 . Thus, DCFH-DA probing of ROS revealed the overall cellular redox status www.nature.com/scientificreports www.nature.com/scientificreports/ of E. agilis under anthracene stress. Higher ROS also reflect the inefficiency of both photochemical pathways and protective regulatory mechanisms to process the excitation energy at PSII. Our data suggest that ROS generation and consequential oxidative stress play a pivotal role in acute anthracene toxicity in the model organism, E. agilis. We detected significant ROS levels under optimal PAR irradiation, where photo-modification of the parent compound is less likely. Under high oxidative damage, Euglena relies on the activation of antioxidant enzymes such as ascorbate peroxidase (APX) and glutathione peroxidase (GPX) 56 , and biosynthesis of antioxidant metabolites such as reduced glutathione (GSH) and its derivatives 57 . Furthermore, some canonical metabolites act as indicators of oxidative damage, such as malondialdehyde (MDA) 58 . Thus, antioxidant/oxidant responses upon anthracene exposure may represent a promising area for further investigation.
We further analyzed the three photochemical quantum yields of PSII measured by imaging PAM to describe the response of PSII photochemistry to anthracene (Fig. 6). Y(II) represents the fraction of excitation energy converted photochemically at PSII. The remaining fraction, 1-Y(II), is the sum of the yields of regulated dissipation, referred to as Y(NPQ), and unregulated dissipation, indicated by Y(NO) 59 . As stated before, anthracene reduced the F v /F m in a dose-dependent manner, and the photon energy requirements for a complete reduction of Q A decrease drastically. This explains the gradual reduction in Y(II) with increasing anthracene doses and a corresponding increase in Y(NO) (Fig. 6). Y(NO) denotes the excess energy, the fraction of absorbed energy used for the generation of free radicals (ROS) via an apparent catalytic transfer of electrons occurred from the reduced PQ pool to O 2 . The higher quantum yield of non-regulated non-photochemical energy loss of PSII (Y(NO)) is a significant stress response, suggesting potential damage to the photosynthetic apparatus exerted by anthracene. Moreover, the declining Y(NPQ) is an indicator of the failure in regulated non-photochemical quenching mechanisms to process the excess energy at PSII. These PSII quantum yield parameters collectively indicates the reduced efficiency of photochemical energy regulation imposed by anthracene exposure.
On the downside of our methodology, anthracene has a higher n-octanol/water partition coefficient (K OW ) of 4.54 60 and, therefore, multi-well plate assay is likely to underestimate the toxicity potential, because a loss of exposure concentration due to lipophilicity is expected for hydrophobic compounds with K OW > 4 61 . We controlled the evaporative loss of the toxicant by sealing the well plates, however, loss in nominal concentration due to physicochemical properties of the tested PAH may have adversely affected the toxicity thresholds reported here. Nevertheless, our data confirmed the mode of anthracene toxicity in E. agilis through Chl a fluorescence technique and the role of ROS in the overall toxicity response.

Conclusions
Microalgae play a pivotal role in primary production in aquatic ecosystems, hence microalgal ingestion in polluted water bodies is a major route by which toxic chemicals can enter the food chain. The results of the present study confirm that anthracene exerts phytotoxic effects on E. agilis by disrupting growth, pigmentation and photosynthesis. Any severe reduction in these parameters will be followed by a loss of ecological competence and diminished survival of the entire E. agilis population, which could have a devastating impact on associated food chains.
Five principal conclusions derived from this study are: (1) The addition of anthracene resulted in a concentration-dependent reduction in cellular growth which appears to be highly related to a reduction in photosynthesis. (2) Anthracene had a pronounced negative effect on photosynthetic pigment content and a simultaneous reduction in all three photosynthetic pigments suggests that the major target of anthracene toxicity is the thylakoid compartment of chloroplasts. These results also indicate that in addition to causing a severe reduction in growth, anthracene exposure may reduce photosynthetic performance via the destruction of pigments responsible for harvesting available photons. (3) Toxicity of anthracene on Y(II), rETR max and non-photochemical quenching parameters in microalgae taxa has for the first time been reported in the current study. The PSII quantum yield parameters collectively indicate the reduced efficiency of photochemical energy regulation, impairment of electron transport downstream from PSII, and an elevated reduction of the PQ pool imposed by anthracene exposure. (4) There was a significant increase in DCFH-DA fluorescence emission in anthracene exposed E. agilis cells, indicating a rise in the intracellular ROS levels. A probable source of generation of ROS would be an inhibition of the electron transport chain at PSII which would have transferred energy from triplet chlorophylls to oxygen, forming singlet oxygen species. A corresponding increase in Y(NO) with increasing anthracene also confirmed that the fraction of absorbed energy might have been used for the generation of free radicals (ROS) via an apparent catalytic transfer of electrons that occurred from the reduced PQ pool to O 2 . (5) Growth and Chl a content of E. agilis may serve as sensitive risk assessment parameters of anthracene toxicity in water management since EC 50 values for both overlap with anthracene levels (8.3 mg L −1 ) permitted by the US Environmental Protection Agency (USEPA).

Data availability
Data can be obtained by contacting the corresponding author. Published: xx xx xxxx Figure 6. Effect of exposure to anthracene on overall energy conversion at PSII in terms of three quantum yields; (i) photochemical quantum yield of photosystem II, Y(II); (ii) quantum yield of non-photochemical fluorescence quenching due to downregulation of the light harvesting function, Y(NPQ); (iii) quantum yield of non-photochemical fluorescence quenching other than that caused by down-regulation of the light harvesting function, Y(NO). Values are given as % relative to untreated controls.