- Email: [email protected]
Ecotoxicology of bromoacetic acid on estuarine phytoplankton
Environmental Pollution 206 (2015) 369e375
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Ecotoxicology of bromoacetic acid on estuarine phytoplankton Ana R. Gordon a, *, Tammi L. Richardson a, b, James L. Pinckney a, b a b
Marine Science Program, University of South Carolina, Columbia, SC 29208, USA Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 February 2015 Received in revised form 7 July 2015 Accepted 14 July 2015 Available online xxx
Bromoacetic acid is formed when effluent containing chlorine residuals react with humics in natural waters containing bromide. The objective of this research was to quantify the effects of bromoacetic acid on estuarine phytoplankton as a proxy for ecosystem productivity. Bioassays were used to measure the EC50 for growth in cultured species and natural marine communities. Growth inhibition was estimated by changes in chlorophyll a concentrations measured by fluorometry and HPLC. The EC50s for cultured Thalassiosira pseudonana were 194 mg L1, 240 mg L1 for Dunaliella tertiolecta and 209 mg L1 for Rhodomonas salina. Natural phytoplankton communities were more sensitive to contamination with an EC50 of 80 mg L1. Discriminant analysis suggested that bromoacetic acid additions cause an alteration of phytoplankton community structure with implications for higher trophic levels. A two-fold EC50 decrease in mixed natural phytoplankton populations affirms the importance of field confirmation for establishing water quality criteria. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Environmental toxicology Estuarine phytoplankton Disinfection by-product Bromoacetic acid
1. Introduction When a disinfectant such as sodium hypochlorite (NaClO) is added to wastewater, it hydrolyses to form hypochlorous acid (HClO). Hypochlorous acid is a weak acid and will undergo partial dissociation between pH 6.5 and 8.5 (Albert and Serjeant, 1984). Aqueous bromine and aqueous chlorine species compete to react with the humic matter. Aqueous chlorine will oxidize bromide to form hypobromous acid (HBrO), which will also undergo a partial dissociation (Abarnou and Miossec, 1992; Allonier et al., 1999; Agus et al., 2009). Hypochlorous acid and hypobromous acid can then react with organic compounds in aqueous solution to produce nonvolatile halogenated aliphatic acids including haloacetic acids (HAAs) which may then be discharged into marine waters (Masters and Ela, 2008; Plewa et al., 2010; Ding et al., 2013). Current literature-reported environmental concentrations of bromoacetic acid in surface waters are predominately <0.1 mg L1 (LeBel et al., 1997; Hashimoto et al., 1998; Williams et al., 1997; Dojlido et al., 1999; Scott et al., 2000, 2002). HAAs readily partition in water because of high Henry's Law constants (1.08 * 105 to 2.26 * 105), low pKas (~1.39) and high water solubility, and rarely volatize back into the atmosphere because
* Corresponding author. E-mail address: [email protected] (A.R. Gordon). http://dx.doi.org/10.1016/j.envpol.2015.07.014 0269-7491/© 2015 Elsevier Ltd. All rights reserved.
complete ionization occurs (Dean and Lange, 1992; Bowden et al., 1998; Hanson and Solomon, 2004; Taylor, 2006). Aquatic environments are particularly vulnerable to HAA contamination because HAAs are unlikely to adsorb onto suspended solids or sediments, but instead will remain dissolved in the water column for exposure to aquatic plants and phytoplankton (Hanson and Solomon, 2004). Free residual chlorine in seawater oxidizes bromide into bromine, making the bromine species of HAAs likely compounds in coastal cities and estuarine ecosystems where salt-water intrusion into the surface waters may cause high bromide levels (Abarnou and Miossec, 1992; Agus et al., 2009; Richardson and Ternes, 2011). Brominated disinfection by-products are typically less volatile and more toxic than their chlorinated analogs (Yang and Zhang, 2013; Liu and Zhang, 2014). Bromoacetic acid (BAA) is of particular interest because mono-HAAs have been shown to be more cytotoxic than their di- and tri-analogs (Giller et al., 1997; Plewa et al., 2010). Phytoplankton can be used as a proxy for estuarine health and productivity to understand the lethal effects and cytotoxicity of BAA. Algae are a particularly good early indicator of pollutant toxicity in an ecosystem because of their high capacity for chemical uptake (due to a high surface area to volume ratio), widespread prevalence, and high turnover rate (DeLorenzo, 2009). Phytoplankton are also a key functional group in estuaries and changes in community abundance or composition may affect ecosystem structure and function (Bougis, 1976; Sournia, 1978; Nyholm and €llqvist, 1989). Ka
370
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 1. Concentration-response curves of cultured phytoplankton to bromoacetic acid, shown by growth measured as fluorometric extracted chlorophyll a inhibition in treated incubations relative to controls. Circles indicate experimental data points used to interpolate curves.
Fig. 2. DCMU ratios of cultured phytoplankton exposed to bromoacetic acid. Experimental treatments were applied at 72 h to ensure that the phytoplankton were in exponential growth. Error bars are standard deviations of treatment replications. A: Thalassiosira pseudonana; B: Dunaliella tertiolecta; C: Rhodomonas salina.
The objective of this research was to characterize the ecotoxicology of BAA in the coastal zone, using marine phytoplankton as an indicator of estuarine health. Phytoplankton have been shown to be particularly sensitive to HAA contamination (Hanson and Solomon, 2004; Agus et al., 2009). To investigate this relationship, both cultured single-algal species and natural communities were exposed to varying concentrations of BAA. Single-species toxicity tests are often conservative estimates of true community response because environmental stressors, inter-species competition, and nutrient availability may make natural phytoplankton communities more sensitive to toxicant exposure (Cairns, 1992; De Laender et al., 2009).
2. Materials and methods 2.1. Lab experiments Due to the difficultly in predicting the variability of toxicity thresholds between species, multiple phytoplankton groups were represented in the experimental design. Unialgal cultures of the diatom Thalassiosira pseudonana (NCMA 1335), the chlorophyte Dunaliella tertiolecta (NCMA 1320) and the cryptophyte Rhodomonas salina (NCMA 1319) were obtained from the ProvasolieGuillard National Center for Marine Algae and Microbiota and used for static case-control toxicity assessments. These phytoplankton were
Table 1 Concentrationeresponse equations for cultured phytoplankton growth inhibition in response to bromoacetic acid (BAA) exposure measured as the change in concentration of extracted chlorophyll a in treated incubations relative to controls. Phytoplankton species Thalassiosira pseudonana Dunaliella tertiolecta Rhodomonas salina
Concentrationeresponse curve equation 15.310
%Inhibition ¼ 0 þ (100-0)/(1 þ (193.769/[BAA]) %Inhibition ¼ 0 þ (100-0)/(1 þ (209.011/[BAA])21.078 %Inihibition ¼ 0 þ (100-0)/(1 þ (239.569/[BAA])9.799
R2
EC50
0.990 0.979 0.978
194 mg L1 209 mg L1 240 mg L1
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 3. Concentration-response curve of natural phytoplankton communities to bromoacetic acid, shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls.
371
chosen based on relatively short doubling times, ease of culturing and environmental prevalence. Algal cultures were maintained in sterile nutrient-enriched seawater media at room temperature (20 ± 3 C) in a 12:12 photoperiod with an irradiance of 70 mmol photons m2 s1. Inoculum cultures were started five days prior to the experimental incubations to allow for an accumulation of phytoplankton cells (Roberts et al., 2010). Crystalline reagent grade BAA (SigmaeAldrich, CAS no. 79-08-3) was dissolved in Milli-Q water to create a stock solution of 100 g L1. The stock solution was added in triplicate to cultures to achieve the concentrations overlapping by 55% (50, 90, 162, 291 and 524 mg BAA L1) omitting addition to control flasks (Rand and Petrocelli, 1985). The high concentrations were selected because they provided a measurable effect over the relatively short incubation times of the bioassays. Therefore, these experiments simulate short-term acute effects rather than longterm chronic responses. The cultures were spiked with BAA 48 h after inoculation while the phytoplankton were in their exponential growth phase as determined by fluorometric chlorophyll a (chl a) measurements. Growth inhibition was then measured daily for an additional 96 h based on extracted chl a concentrations. All chl a measurements (Turner Designs 10-AU Fluorometer)
Fig. 4. Natural phytoplankton community compositions quantified as absolute chlorophyll a in ChemTax at the end of the exposure (36 h). A: Experiment 1 (May 2014); B: Experiment 2 (June 2014); C: Experiment 3 (July 2014).
372
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
triplicated. All bottles were incubated for 36 h (with a minimum of 24 h of daylight) in situ in screened (fiberglass mesh) water tables (reducing the ambient irradiance by ca. 30%) to simulate environmental conditions of light and temperature. Both immediately after toxicant spike (time ¼ 0 h) and at the end of the experiment (time ¼ 36 h), separate aliquots were taken for measurements of BAA concentration and high performance liquid chromatographic (HPLC) analysis of photosynthetic pigments. Samples for HPLC were filtered onto Whatman GF/F filters and immediately stored at 80 C.
2.3. Analysis
Fig. 5. Canonical discriminant function plots for phytoplankton group data classified by bromoacetic acid treatments. Centroids are labeled in open squares with respective bromoacetic acid concentrations (mg L1) for Experiment 1 (May 2014) and Experiment 3 (July 2014). Functions for individual data points are indicated by solid circles. There was no significant difference between treatment responses for Experiment 2 (June 2014).
were determined by the emission of chl a florescence at the red wavelength (~680 nm) when samples were excited with blue light (430 nm). DCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea, SigmaeAldrich, CAS no. 330-54-1, 0.1 M) was then added to these samples to estimate photosynthetic capacity by inducing an increase of in vivo fluorescence relative to uninhibited bulk fluorescence (Cullen and Renger, 1979). Additional water samples from the cultures were filtered onto Whatman GF/F filters and extracted in 90% acetone at 20 C for 24 h (Arar and Collins, 1997). In vitro fluorescence (extracted chl a) was then measured and adjusted according to the volume of acetone extract and the volume of the sample filtered to calculate chl a concentrations (Welschmeyer, 1994). Chl a concentration was used as a proxy for phytoplankton biomass. 2.2. Field experiments Three separate short-term bioassays were conducted in situ in North Inlet Estuary at the Belle W. Baruch Marine Field Lab in Georgetown, South Carolina during the summer of 2014 to investigate multi-species acute algal toxicity. Water samples (1.75 L) were collected during the incoming high tide from Clambank Landing (33 200 0200 N, 791103400 W) in acid-washed 2-L clear polycarbonate Nalgene bottles. Bottles were then spiked with crystalline BAA dissolved in Milli-Q water (to a stock concentration of 100 g L1) in triplicate at concentrations ranging from 5 to 524 mg L1. Control bottles containing only seawater were also
Phytoplankton community composition was measured by HPLC analysis of photosynthetic pigments (Roy et al., 2011). Filters for HPLC analysis were lyophilized for 24 h at 50 C and 0.57 mBar to remove remaining moisture. Photopigments were extracted by adding 750 mL of 90% aqueous acetone solvent followed by storage for 12e20 h at 20 C. Filtered extracts (250 mL) were injected into a Shimadzu HPLC with a single monomeric column (Rainin Microsorb, 0.46 1.5 cm, 3 mm packing) and a polymeric (Vydac 201TP54, 0.46 25 cm, 5 mm packing) reverse-phase C18 column in series. Ammonium acetate (1.0 M in volume to volume ratio of 1.25) was added to extracts for ion pairing prior to injection. A non-linear binary gradient consisting of solvent A (80% methanol: 20% 0.5 M ammonium acetate) and solvent B (80% methanol: 20% acetone) was used for the mobile phase (Pinckney et al., 2001). Absorption spectra and chromatograms (440 ± 4 nm) were obtained using a Shimadzu SPD-M10av photodiode array detector and pigment peaks were identified by comparing retention times and absorption spectra with pure standards (DHI, Denmark). The synthetic carotenoid b-apo-80 -carotenal (SigmaeAldrich) was used as an internal standard. The software ChemTax (v. 1.95; Mackey et al., 1996; Wright et al., 1996) was used to determine the relative abundance of major phytoplankton groups (Pinckney et al., 2001; Lewitus et al., 2005; Higgins et al., 2011). The initial pigment ratio matrix used for this analysis was derived from Mackey et al. (1996). The convergence procedure outlined by Latasa (2007) was used to minimize errors in algal group biomass due to inaccurate pigment ratio seed values. The chl a contribution of each algal class was partitioned into the total chl a of major phytoplankton groups providing an estimate of the community composition. BAA concentrations in all lab and field incubations were quantified through derivative analysis on a HewlettePackard 5890 gas chromatograph equipped with an electron-capture detection system following the methods described by Ford et al. (2007). Derivitization was done on a Varian 3800 gas chromatograph with a Saturn 2000 ion trap mass spectrometry system (Ford et al., 2007). Samples were preserved at 6 C in amber glass vials, and run within 48 h of sampling as specified by EPA Method 552.3. Half-maximum effective concentrations (EC50) were estimated using the extracted chl a data for each experiment. Inhibition of chl a for each treatment relative to controls were calculated using a non-linear regression method (IBM SPSS v. 21). Sigmoidal concentrationeresponse curves were fitted to the following equation (Motulsky and Christopoulos, 2004).
% Inhibition ¼ y axisminimum þ ðy axismaximum y axisminimum Þ=ð1 þ ðEC50 =½BAAÞHillslope Differences in community composition between BAA treatment levels were verified using a randomized complete block design multivariate analysis of variance (MANOVA) procedure with
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
373
Fig. 6. Sigmoidal concentrationeresponse curves of natural phytoplankton groups to bromoacetic acid, shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls. Inhibition was calculated based on relative change in algal group abundance in chlorophyll a units of each treatment compared to the controls at the end of the exposure (36 h) for each experiment before being pooled for response estimations. Circles indicate experimental chlorophyll a inhibition and the curve is the derived sigmoid function.
experiment number as the blocking factor and BAA concentration as the main factor (Pinckney and Lee, 2008). Discriminant analyses were performed following the MANOVA to build a predictive model of community composition shifts based on the observed characteristics of BAA treatments. These results were used to separate similar community compositions among BAA treatment levels (Pinckney and Lee, 2008). 3. Results Indicators of growth inhibition in all three cultured phytoplankton species suggest similar responses for extracted chl a (Fig. 1). The two lowest doses of BAA (50 mg L1 and 90 mg L1) had no measured effect on any of the test species (Fig. 1). No effect was
also seen at 162 mg L1 in T. pseudonana and D. tertiolecta, however partial inhibition was observed in R. salina. The two highest doses (291 mg L1 and 524 mg L1) resulted in complete inhibition among all species. Similarly, DCMU results suggested that photosynthetic capacity was minimal at the both 291 and 524 mg L1 levels in all three species (after the toxicant spike at 72 h) and reduced, but not minimal, at 162 mg L1 in R. salina only (Fig. 2). The half-maximum effective concentration (EC50) of BAA was 194 mg L1 for T. pseudonana, 209 mg L1 for D. tertiolecta and 240 mg L1 for R. salina (Table 1). In contrast, the overall EC50 in natural phytoplankton communities relative to uncontaminated controls was 80 mg L1 (Fig. 3). To investigate shifts in community composition, data for all measured phytoplankton groups were pooled and analyzed using a two-
Table 2 Concentrationeresponse curves of natural phytoplankton communities to bromoacetic acid (BAA), as shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls. Inhibition was calculated based on relative change in algal group abundance in chlorophyll a units of each treatment compared to the controls at the end of the exposure (36 h) for each experiment before being pooled for response estimations. Phytoplankton group
Concentrationeresponse curve equation
Cryptophytes Diatoms Dinoflagellates Prasinophytes Haptophytes
%Inhibition %Inhibition %Inhibition %Inhibition %Inhibition
¼ ¼ ¼ ¼ ¼
0 0 0 0 0
þ þ þ þ þ
(100-0)/(1 (100-0)/(1 (100-0)/(1 (100-0)/(1 (100-0)/(1
þ þ þ þ þ
(33.139/[BAA])1.424 (79.595/[BAA])2.386 (21.912/[BAA])1.882 (31.138/[BAA])1.910 (156.045/[BAA])5.467
R2
EC50
0.691 0.759 0.606 0.562 0.784
33 mg L1 80 mg L1 22 mg L1 31 mg L1 156 mg L1
374
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 7. Chlorophyte response to bromoacetic acid as measured by high performance liquid chromatographic chlorophyll a concentration.
factor MANOVA (Fig. 4). Significant differences were confirmed (p < 0.0001) between concentrations of BAA added among all treatments. Discriminant analysis, using all phytoplankton groups as variables, was used to further explore the MANOVA results by classifying community composition using discriminant functions. For Experiment 1 (May 2014), the first two discriminant functions explained 92 and 6.9% of the variance, respectively (cumulative total ¼ 98.9%). The group centroids, based on canonical discriminant functions, suggest similar community compositions among BAA treatments above 50 mg L1 (90, 162, 294 and 524 mg L1), with the 50 mg L1 and control incubations each showing distinct differences (Fig. 5). Discriminant analysis on Experiment 2 (June 2014) indicated no significant differences among any BAA treatments. For Experiment 3 (July 2014), the first two discriminant functions explained 79.5 and 20.5% of the variance, respectively (cumulative total ¼ 100%). The group centroids suggest similar community compositions among the four lowest treatments (5, 9, 16 mg L1 and the control incubations), the middle three treatments (29, 40 and 52 mg L1) and those above 52 mg L1 (60, 85, 94, 170 mg L1) (Fig. 5). To investigate any toxicity threshold differences among individual algal classes, the individual concentrationeresponse curves were plotted separately (Fig. 6) and indicate that toxicity thresholds ranged from 31 to 156 mg L1 (Table 2). Although chlorophytes were present, their response did not fit the sigmoidal concentrationeresponse curve. Only at low (<20 mg L1) and high (>90 mg L1) concentrations of BAA, were the chlorophytes completely inhibited (Fig. 7). There even appears to be a slight increase in chlorophyte concentration at mid-range levels of BAA dose, which was not seen in the representative cultured chlorophyte, D. tertiolecta. 4. Discussion The complete inhibition of all phytoplankton cultures treated with the two highest doses (291 mg L1 and 524 mg L1) suggests that BAA is algicidal past the toxicity threshold (Fig. 1). Published literature has suggested that R. salina may be more sensitive to some contaminants than other phytoplankton species, which is evident at the middle dose, 162 mg L1, where BAA acid is algistatic in R. salina (cryptophyte) but has no effect on either T. pseudonana
(diatom) or D. tertiolecta (chlorophyte) (Hampel et al., 2001). The DCMU results (Fig. 2) further suggest that BAA is lethal at the two highest doses. DCMU blocks the electron flow in Photosystem II, and the ratio between bulk fluorescence and the DCMUinhibited fluorescence is an indicator of photosynthetic capacity. When this ratio declines, it suggests that the phytoplankton cells are unable to photosynthesize properly. The two highest doses of BAA (291 and 524 mg L1) appear to have greatly impacted photosynthetic capacity, suggesting that the mode of action for BAA may be related to photosynthesis (Fig. 2). However, it is evident that the photosynthetic capacity of T. pseudonana declined even with the lowest doses. One explanation for the decline of photosynthetic capacity during the T. pseudonana exposure is that, unlike the other two species exposed, T. pseudonana was spiked during the senescence phase. This is the most sensitive phase of the phytoplankton life cycle, and suggests that BAA may be lethal at lower concentrations if exposed as the cells reach stationary growth and decline. Curve shapes and EC50 values (194e240 mg L1) for all species suggest minimal toxicity differences among these cultured phytoplankton, despite phylogenetic diversity (Fig. 1). The steep concentrationeresponse curves suggest a toxicity threshold, where there is no or minimal toxic effect below and complete cell death above. Single-species toxicity tests are often conservative estimates of true community response because environmental stressors, interspecies competition, and nutrient availability may make natural phytoplankton communities more sensitive to toxicant exposure (Cairns, 1992; De Laender et al., 2009). This study was no exception, where the EC50 of BAA on the bulk phytoplankton community chl a inhibition (80 mg L1) was lower than all of the single algal species EC50 values (194e240 mg L1) (Fig. 3). Although the coefficient of determination (R2) for this concentrationeresponse curve (based on relative change in chl a) is high (0.873), there is a certain level of response variability below doses of 100 mg L1. These inconsistencies may be due to slightly differing phytoplankton communities (confirmed by the MANOVA test) across the three different incubations skewing the experimental data points (Figs. 4 and 6). The discriminant analysis at each treatment level suggests differences in community composition (Fig. 5) within this range, with phytoplankton population shifts at ~30 mg L1 and again at ~50 mg L1. Concentration-response curves for individual natural phytoplankton groups (Fig. 6) indicate some variability in EC50 values (Table 2). As different phytoplankton have different responses to BAA, it is apparent that the concentrationeresponse curve for the entire population will have some variability. Although the chlorophytes could not be fitted to a concentrationeresponse curve, and therefore an EC50 value could not be calculated, their response is noteworthy. Chlorophytes may be opportunistic at the medium doses when competition with other algal groups is reduced due to the lower lethal toxicities for other algal groups (such as the dinoflagellates). A reduction of these competing groups at lower concentrations of BAA may open up a niche for chlorophytes at medium dose levels (50 mg L1). This response was not seen in the cultured D. tertiolecta, suggesting more work should be done to qualify this specific difference in response between natural and cultured phytoplankton. 5. Conclusions The results of these experiments suggest that the halfmaximum effective concentration of BAA to natural phytoplankton communities is below that of single-species algal cultures. Individual phytoplankton groups respond with widely varying sensitivity, making community composition an important factor in
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
determining overall community responses to dissolved toxicants. The reduction in photosynthetic capacity of the cultured phytoplankton at high doses of BAA suggests that the mode of action for BAA may be related to photosynthesis, although further work should be done for positive confirmation. The difference in the toxicity threshold between cultured and natural phytoplankton suggest that standard acute laboratory toxicology tests may not provide an accurate estimate of environmental health, and emphasize the importance of confirming lab experiments with field work whenever possible. Acknowledgments The authors would like to acknowledge Benson Solomon and the Ferry Lab at the University of South Carolina for assistance with the chemical analysis of BAA in solution. References Abarnou, A., Miossec, L., 1992. Chlorinated waters discharged to the marine environment chemistry and environmental impact. An overview. Sci. Total Environ. 126 (1), 173e197. Agus, E., Voutchkov, N., Sedlak, D.L., 2009. Disinfection by-products and their potential impact on the quality of water produced by desalination systems: a literature review. Desalination 237 (1), 214e237. Albert, A., Serjeant, E.P., 1984. The Determination of Ionization Constants, vol. 2. Chapman and Hall, New York. Allonier, A.S., Khalanski, M., Camel, V., Bermond, A., 1999. Characterization of chlorination by-products in cooling effluents of coastal nuclear power stations. Mar. Pollut. Bull. 38 (12), 1232e1241. Arar, E.J., Collins, G.B., 1997. Method 445.0: In Vitro Determination of Chlorophyll a and Pheophytin a in Marine and Freshwater Algae by Fluorescence. United States Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory. Bougis, P., 1976. Marine Plankton Ecology. Bowden, D.J., Clegg, S.L., Brimblecombe, P., 1998. The Henry's law constants of the haloacetic acids. J. Atmos. Chem. 29 (1), 85e107. Cairns Jr., J., 1992. The threshold problem in ecotoxicology. Ecotoxicology 1 (1), 3e16. Cullen, J.J., Renger, E.H., 1979. Continuous measurement of the DCMU-induced fluorescence response of natural phytoplankton populations. Mar. Biol. 53 (1), 13e20. Dean, J.A., Lange, N.A., 1992. Lange's Handbook of Chemistry, 14th ed. McGraw-Hill, New York. Print. De Laender, F., De Schamphelaere, K.A., Vanrolleghem, P.A., Janssen, C.R., 2009. Comparing ecotoxicological effect concentrations of chemicals established in multi-species vs. single-species toxicity test systems. Ecotoxicol. Environ. Saf. 72 (2), 310e315. DeLorenzo, M.E., 2009. Utility of Dunaliella in ecotoxicity testing. In: BenAmotz, Ami, Polle, Jurgen, Subba Rao, D.V. (Eds.), The Alga Dunaliella: Biodiversity, Physiology, Genomics & Biotechnology. Science Publishers, New Hampshire, USA. Ding, G., Zhang, X., Yang, M., Pan, Y., 2013. Formation of new brominated disinfection byproducts during chlorination of saline sewage effluents. Water Res. 47 (8), 2710e2718. ^ Dojlido, J., Zbie c, E., Swietlik, R., 1999. Formation of the haloacetic acids during ozonation and chlorination of water in Warsaw waterworks (Poland). Water Res. 33 (14), 3111e3118. Ford, Q.L., Burns, J.M., Ferry, J.L., 2007. Aqueous in situ derivatization of carboxylic acids by an ionic carbodiimide and 2, 2, 2-trifluoroethylamine for electroncapture detection. J. Chromatogr. A 1145 (1), 241e245. Giller, S., Le Curieux, F., Erb, F., Marzin, D., 1997. Comparative genotoxicity of halogenated acetic acids found in drinking water. Mutagenesis 12 (5), 321e328. n, L.M., Blasco, J., 2001. Acute Hampel, M., Moreno-Garrido, I., Sobrino, C., Lubia toxicity of LAS homologues in marine microalgae: esterase activity and
375
inhibition growth as endpoints of toxicity. Ecotoxicol. Environ. Saf. 48 (3), 287e292. Hanson, M.L., Solomon, K.R., 2004. Haloacetic acids in the aquatic environment. Part II: ecological risk assessment. Environ. Pollut. 130 (3), 385e401. Hashimoto, S., Azuma, T., Otsuki, A., 1998. Distribution, sources, and stability of haloacetic acids in Tokyo Bay, Japan. Environ. Toxicol. Chem. 17 (5), 798e805. Higgins, H.W., Wright, S.W., Schlüter, L., 2011. Quantitative interpretation of chemotaxonomic pigment data. In: Roy, S., Llewellyn, C.A., Egeland, E.S., Johnsen, G. (Eds.), Phytoplankton Pigments. Cambridge University Press, NY, pp. 257e313. Latasa, M., 2007. Improving estimations of phytoplankton class abundances using CHEMTAX. Mar. Ecol. Prog. Ser. 329, 13e21. LeBel, G.L., Benoit, F.M., Williams, D.T., 1997. A one-year survey of halogenated disinfection by-products in the distribution system of treatment plants using three different disinfection processes. Chemosphere 34 (11), 2301e2317. Lewitus, A.J., White, D.L., Tymowski, R.G., Geesey, M.E., Hymel, S.N., Noble, P.A., 2005. Adapting the CHEMTAX method for assessing phytoplankton taxonomic composition in southeastern US estuaries. Estuaries 28 (1), 160e172. Liu, J., Zhang, X., 2014. Comparative toxicity of new halophenolic DBPs in chlorinated saline wastewater effluents against a marine alga: halophenolic DBPs are generally more toxic than haloaliphatic ones. Water Res. 65, 64e72. Mackey, M.D., Mackey, D.J., Higgins, H.W., Wright, S.W., 1996. CHEMTAX-a program for estimating class abundances from chemical markers: application to HPLC measurements of phytoplankton. Mar. Ecol. Prog. Ser. 144, 265e283. Masters, G.M., Ela, W.P., 2008. Introduction to Environmental Engineering and Science, third ed. Pearson Education, Inc, Upper Saddle River, NJ. Print. Motulsky, H., Christopoulos, A., 2004. Fitting Models to Biological Data Using Linear and Nonlinear Regression: a Practical Guide to Curve Fitting. Oxford University Press. €llqvist, T., 1989. Methods for growth inhibition toxicity tests with Nyholm, N., Ka freshwater algae. Environ. Toxicol. Chem. 8 (8), 689e703. Pinckney, J.L., Richardson, T.L., Millie, D.F., Paerl, H.W., 2001. Application of photopigment biomarkers for quantifying microalgal community composition and in situ growth rates. Org. Geochem. 32 (4), 585e595. Pinckney, J.L., Lee, A.R., 2008. Spatiotemporal patterns of subtidal benthic microalgal biomass and community composition in Galveston Bay, Texas, USA. Estuaries Coasts 31 (2), 444e454. Plewa, M.J., Simmons, J.E., Richardson, S.D., Wagner, E.D., 2010. Mammalian cell cytotoxicity and genotoxicity of the haloacetic acids, a major class of drinking water disinfection by-products. Environ. Mol. Mutagen. 51 (8e9), 871e878. Rand, G.M., Petrocelli, S.R., 1985. Fundamentals of Aquatic Toxicology: Methods and Applications. FMC Corporation, Princeton, NJ. Richardson, S.D., Ternes, T.A., 2011. Water analysis: emerging contaminants and current issues. Anal. Chem. 83 (12), 4614e4648. Roberts, J.F., van Egmond, R., Price, O.R., 2010. Toxicity of haloacetic acids to freshwater algae. Ecotoxicol. Environ. Saf. 73 (1), 56e61. Roy, S., Llewellyn, C.A., Egeland, E.S., Johnsen, G. (Eds.), 2011. Phytoplankton Pigments: Characterization, Chemotaxonomy and Applications in Oceanography. Cambridge University Press. Scott, B.F., MacTavish, D., Spencer, C., Strachan, W.M., Muir, D.C., 2000. Haloacetic acids in Canadian lake waters and precipitation. Environ. Sci. Technol. 34 (20), 4266e4272. Scott, B.F., Spencer, C., Marvin, C.H., MacTavish, D.C., Muir, D.C., 2002. Distribution of haloacetic acids in the water columns of the Laurentian Great Lakes and Lake Malawi. Environ. Sci. Technol. 36 (9), 1893e1898. Sournia, A., 1978. Phytoplankton Manual. Unesco, Paris. Print. Taylor, C.J., 2006. The effects of biological fouling control at coastal and estuarine power stations. Mar. Pollut. Bull. 53 (1), 30e48. Welschmeyer, N.A., 1994. Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol. Oceanogr. 39 (8), 1985e1992. Williams, D.T., LeBel, G.L., Benoit, F.M., 1997. Disinfection by-products in Canadian drinking water. Chemosphere 34 (2), 299e316. Wright, S.W., Thomas, D.P., Marchant, H.J., Higgins, H.W., Mackey, M.D., Mackey, D.J., 1996. Analysis of phytoplankton of the Australian sector of the Southern Ocean: comparisons of microscopy and size frequency data with interpretations of pigment HPLC data using the CHEMTAX matrix factorisation program. Mar. Ecol. Prog. Ser. Oldend. 14 (1), 285e298. Yang, M., Zhang, X., 2013. Comparative developmental toxicity of new aromatic halogenated DBPs in a chlorinated saline sewage effluent to the marine polychaete Platynereis dumerilii. Environ. Sci. Technol. 47 (19), 10868e10876.
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Ecotoxicology of bromoacetic acid on estuarine phytoplankton Ana R. Gordon a, *, Tammi L. Richardson a, b, James L. Pinckney a, b a b
Marine Science Program, University of South Carolina, Columbia, SC 29208, USA Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 February 2015 Received in revised form 7 July 2015 Accepted 14 July 2015 Available online xxx
Bromoacetic acid is formed when effluent containing chlorine residuals react with humics in natural waters containing bromide. The objective of this research was to quantify the effects of bromoacetic acid on estuarine phytoplankton as a proxy for ecosystem productivity. Bioassays were used to measure the EC50 for growth in cultured species and natural marine communities. Growth inhibition was estimated by changes in chlorophyll a concentrations measured by fluorometry and HPLC. The EC50s for cultured Thalassiosira pseudonana were 194 mg L1, 240 mg L1 for Dunaliella tertiolecta and 209 mg L1 for Rhodomonas salina. Natural phytoplankton communities were more sensitive to contamination with an EC50 of 80 mg L1. Discriminant analysis suggested that bromoacetic acid additions cause an alteration of phytoplankton community structure with implications for higher trophic levels. A two-fold EC50 decrease in mixed natural phytoplankton populations affirms the importance of field confirmation for establishing water quality criteria. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Environmental toxicology Estuarine phytoplankton Disinfection by-product Bromoacetic acid
1. Introduction When a disinfectant such as sodium hypochlorite (NaClO) is added to wastewater, it hydrolyses to form hypochlorous acid (HClO). Hypochlorous acid is a weak acid and will undergo partial dissociation between pH 6.5 and 8.5 (Albert and Serjeant, 1984). Aqueous bromine and aqueous chlorine species compete to react with the humic matter. Aqueous chlorine will oxidize bromide to form hypobromous acid (HBrO), which will also undergo a partial dissociation (Abarnou and Miossec, 1992; Allonier et al., 1999; Agus et al., 2009). Hypochlorous acid and hypobromous acid can then react with organic compounds in aqueous solution to produce nonvolatile halogenated aliphatic acids including haloacetic acids (HAAs) which may then be discharged into marine waters (Masters and Ela, 2008; Plewa et al., 2010; Ding et al., 2013). Current literature-reported environmental concentrations of bromoacetic acid in surface waters are predominately <0.1 mg L1 (LeBel et al., 1997; Hashimoto et al., 1998; Williams et al., 1997; Dojlido et al., 1999; Scott et al., 2000, 2002). HAAs readily partition in water because of high Henry's Law constants (1.08 * 105 to 2.26 * 105), low pKas (~1.39) and high water solubility, and rarely volatize back into the atmosphere because
* Corresponding author. E-mail address: [email protected] (A.R. Gordon). http://dx.doi.org/10.1016/j.envpol.2015.07.014 0269-7491/© 2015 Elsevier Ltd. All rights reserved.
complete ionization occurs (Dean and Lange, 1992; Bowden et al., 1998; Hanson and Solomon, 2004; Taylor, 2006). Aquatic environments are particularly vulnerable to HAA contamination because HAAs are unlikely to adsorb onto suspended solids or sediments, but instead will remain dissolved in the water column for exposure to aquatic plants and phytoplankton (Hanson and Solomon, 2004). Free residual chlorine in seawater oxidizes bromide into bromine, making the bromine species of HAAs likely compounds in coastal cities and estuarine ecosystems where salt-water intrusion into the surface waters may cause high bromide levels (Abarnou and Miossec, 1992; Agus et al., 2009; Richardson and Ternes, 2011). Brominated disinfection by-products are typically less volatile and more toxic than their chlorinated analogs (Yang and Zhang, 2013; Liu and Zhang, 2014). Bromoacetic acid (BAA) is of particular interest because mono-HAAs have been shown to be more cytotoxic than their di- and tri-analogs (Giller et al., 1997; Plewa et al., 2010). Phytoplankton can be used as a proxy for estuarine health and productivity to understand the lethal effects and cytotoxicity of BAA. Algae are a particularly good early indicator of pollutant toxicity in an ecosystem because of their high capacity for chemical uptake (due to a high surface area to volume ratio), widespread prevalence, and high turnover rate (DeLorenzo, 2009). Phytoplankton are also a key functional group in estuaries and changes in community abundance or composition may affect ecosystem structure and function (Bougis, 1976; Sournia, 1978; Nyholm and €llqvist, 1989). Ka
370
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 1. Concentration-response curves of cultured phytoplankton to bromoacetic acid, shown by growth measured as fluorometric extracted chlorophyll a inhibition in treated incubations relative to controls. Circles indicate experimental data points used to interpolate curves.
Fig. 2. DCMU ratios of cultured phytoplankton exposed to bromoacetic acid. Experimental treatments were applied at 72 h to ensure that the phytoplankton were in exponential growth. Error bars are standard deviations of treatment replications. A: Thalassiosira pseudonana; B: Dunaliella tertiolecta; C: Rhodomonas salina.
The objective of this research was to characterize the ecotoxicology of BAA in the coastal zone, using marine phytoplankton as an indicator of estuarine health. Phytoplankton have been shown to be particularly sensitive to HAA contamination (Hanson and Solomon, 2004; Agus et al., 2009). To investigate this relationship, both cultured single-algal species and natural communities were exposed to varying concentrations of BAA. Single-species toxicity tests are often conservative estimates of true community response because environmental stressors, inter-species competition, and nutrient availability may make natural phytoplankton communities more sensitive to toxicant exposure (Cairns, 1992; De Laender et al., 2009).
2. Materials and methods 2.1. Lab experiments Due to the difficultly in predicting the variability of toxicity thresholds between species, multiple phytoplankton groups were represented in the experimental design. Unialgal cultures of the diatom Thalassiosira pseudonana (NCMA 1335), the chlorophyte Dunaliella tertiolecta (NCMA 1320) and the cryptophyte Rhodomonas salina (NCMA 1319) were obtained from the ProvasolieGuillard National Center for Marine Algae and Microbiota and used for static case-control toxicity assessments. These phytoplankton were
Table 1 Concentrationeresponse equations for cultured phytoplankton growth inhibition in response to bromoacetic acid (BAA) exposure measured as the change in concentration of extracted chlorophyll a in treated incubations relative to controls. Phytoplankton species Thalassiosira pseudonana Dunaliella tertiolecta Rhodomonas salina
Concentrationeresponse curve equation 15.310
%Inhibition ¼ 0 þ (100-0)/(1 þ (193.769/[BAA]) %Inhibition ¼ 0 þ (100-0)/(1 þ (209.011/[BAA])21.078 %Inihibition ¼ 0 þ (100-0)/(1 þ (239.569/[BAA])9.799
R2
EC50
0.990 0.979 0.978
194 mg L1 209 mg L1 240 mg L1
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 3. Concentration-response curve of natural phytoplankton communities to bromoacetic acid, shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls.
371
chosen based on relatively short doubling times, ease of culturing and environmental prevalence. Algal cultures were maintained in sterile nutrient-enriched seawater media at room temperature (20 ± 3 C) in a 12:12 photoperiod with an irradiance of 70 mmol photons m2 s1. Inoculum cultures were started five days prior to the experimental incubations to allow for an accumulation of phytoplankton cells (Roberts et al., 2010). Crystalline reagent grade BAA (SigmaeAldrich, CAS no. 79-08-3) was dissolved in Milli-Q water to create a stock solution of 100 g L1. The stock solution was added in triplicate to cultures to achieve the concentrations overlapping by 55% (50, 90, 162, 291 and 524 mg BAA L1) omitting addition to control flasks (Rand and Petrocelli, 1985). The high concentrations were selected because they provided a measurable effect over the relatively short incubation times of the bioassays. Therefore, these experiments simulate short-term acute effects rather than longterm chronic responses. The cultures were spiked with BAA 48 h after inoculation while the phytoplankton were in their exponential growth phase as determined by fluorometric chlorophyll a (chl a) measurements. Growth inhibition was then measured daily for an additional 96 h based on extracted chl a concentrations. All chl a measurements (Turner Designs 10-AU Fluorometer)
Fig. 4. Natural phytoplankton community compositions quantified as absolute chlorophyll a in ChemTax at the end of the exposure (36 h). A: Experiment 1 (May 2014); B: Experiment 2 (June 2014); C: Experiment 3 (July 2014).
372
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
triplicated. All bottles were incubated for 36 h (with a minimum of 24 h of daylight) in situ in screened (fiberglass mesh) water tables (reducing the ambient irradiance by ca. 30%) to simulate environmental conditions of light and temperature. Both immediately after toxicant spike (time ¼ 0 h) and at the end of the experiment (time ¼ 36 h), separate aliquots were taken for measurements of BAA concentration and high performance liquid chromatographic (HPLC) analysis of photosynthetic pigments. Samples for HPLC were filtered onto Whatman GF/F filters and immediately stored at 80 C.
2.3. Analysis
Fig. 5. Canonical discriminant function plots for phytoplankton group data classified by bromoacetic acid treatments. Centroids are labeled in open squares with respective bromoacetic acid concentrations (mg L1) for Experiment 1 (May 2014) and Experiment 3 (July 2014). Functions for individual data points are indicated by solid circles. There was no significant difference between treatment responses for Experiment 2 (June 2014).
were determined by the emission of chl a florescence at the red wavelength (~680 nm) when samples were excited with blue light (430 nm). DCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea, SigmaeAldrich, CAS no. 330-54-1, 0.1 M) was then added to these samples to estimate photosynthetic capacity by inducing an increase of in vivo fluorescence relative to uninhibited bulk fluorescence (Cullen and Renger, 1979). Additional water samples from the cultures were filtered onto Whatman GF/F filters and extracted in 90% acetone at 20 C for 24 h (Arar and Collins, 1997). In vitro fluorescence (extracted chl a) was then measured and adjusted according to the volume of acetone extract and the volume of the sample filtered to calculate chl a concentrations (Welschmeyer, 1994). Chl a concentration was used as a proxy for phytoplankton biomass. 2.2. Field experiments Three separate short-term bioassays were conducted in situ in North Inlet Estuary at the Belle W. Baruch Marine Field Lab in Georgetown, South Carolina during the summer of 2014 to investigate multi-species acute algal toxicity. Water samples (1.75 L) were collected during the incoming high tide from Clambank Landing (33 200 0200 N, 791103400 W) in acid-washed 2-L clear polycarbonate Nalgene bottles. Bottles were then spiked with crystalline BAA dissolved in Milli-Q water (to a stock concentration of 100 g L1) in triplicate at concentrations ranging from 5 to 524 mg L1. Control bottles containing only seawater were also
Phytoplankton community composition was measured by HPLC analysis of photosynthetic pigments (Roy et al., 2011). Filters for HPLC analysis were lyophilized for 24 h at 50 C and 0.57 mBar to remove remaining moisture. Photopigments were extracted by adding 750 mL of 90% aqueous acetone solvent followed by storage for 12e20 h at 20 C. Filtered extracts (250 mL) were injected into a Shimadzu HPLC with a single monomeric column (Rainin Microsorb, 0.46 1.5 cm, 3 mm packing) and a polymeric (Vydac 201TP54, 0.46 25 cm, 5 mm packing) reverse-phase C18 column in series. Ammonium acetate (1.0 M in volume to volume ratio of 1.25) was added to extracts for ion pairing prior to injection. A non-linear binary gradient consisting of solvent A (80% methanol: 20% 0.5 M ammonium acetate) and solvent B (80% methanol: 20% acetone) was used for the mobile phase (Pinckney et al., 2001). Absorption spectra and chromatograms (440 ± 4 nm) were obtained using a Shimadzu SPD-M10av photodiode array detector and pigment peaks were identified by comparing retention times and absorption spectra with pure standards (DHI, Denmark). The synthetic carotenoid b-apo-80 -carotenal (SigmaeAldrich) was used as an internal standard. The software ChemTax (v. 1.95; Mackey et al., 1996; Wright et al., 1996) was used to determine the relative abundance of major phytoplankton groups (Pinckney et al., 2001; Lewitus et al., 2005; Higgins et al., 2011). The initial pigment ratio matrix used for this analysis was derived from Mackey et al. (1996). The convergence procedure outlined by Latasa (2007) was used to minimize errors in algal group biomass due to inaccurate pigment ratio seed values. The chl a contribution of each algal class was partitioned into the total chl a of major phytoplankton groups providing an estimate of the community composition. BAA concentrations in all lab and field incubations were quantified through derivative analysis on a HewlettePackard 5890 gas chromatograph equipped with an electron-capture detection system following the methods described by Ford et al. (2007). Derivitization was done on a Varian 3800 gas chromatograph with a Saturn 2000 ion trap mass spectrometry system (Ford et al., 2007). Samples were preserved at 6 C in amber glass vials, and run within 48 h of sampling as specified by EPA Method 552.3. Half-maximum effective concentrations (EC50) were estimated using the extracted chl a data for each experiment. Inhibition of chl a for each treatment relative to controls were calculated using a non-linear regression method (IBM SPSS v. 21). Sigmoidal concentrationeresponse curves were fitted to the following equation (Motulsky and Christopoulos, 2004).
% Inhibition ¼ y axisminimum þ ðy axismaximum y axisminimum Þ=ð1 þ ðEC50 =½BAAÞHillslope Differences in community composition between BAA treatment levels were verified using a randomized complete block design multivariate analysis of variance (MANOVA) procedure with
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
373
Fig. 6. Sigmoidal concentrationeresponse curves of natural phytoplankton groups to bromoacetic acid, shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls. Inhibition was calculated based on relative change in algal group abundance in chlorophyll a units of each treatment compared to the controls at the end of the exposure (36 h) for each experiment before being pooled for response estimations. Circles indicate experimental chlorophyll a inhibition and the curve is the derived sigmoid function.
experiment number as the blocking factor and BAA concentration as the main factor (Pinckney and Lee, 2008). Discriminant analyses were performed following the MANOVA to build a predictive model of community composition shifts based on the observed characteristics of BAA treatments. These results were used to separate similar community compositions among BAA treatment levels (Pinckney and Lee, 2008). 3. Results Indicators of growth inhibition in all three cultured phytoplankton species suggest similar responses for extracted chl a (Fig. 1). The two lowest doses of BAA (50 mg L1 and 90 mg L1) had no measured effect on any of the test species (Fig. 1). No effect was
also seen at 162 mg L1 in T. pseudonana and D. tertiolecta, however partial inhibition was observed in R. salina. The two highest doses (291 mg L1 and 524 mg L1) resulted in complete inhibition among all species. Similarly, DCMU results suggested that photosynthetic capacity was minimal at the both 291 and 524 mg L1 levels in all three species (after the toxicant spike at 72 h) and reduced, but not minimal, at 162 mg L1 in R. salina only (Fig. 2). The half-maximum effective concentration (EC50) of BAA was 194 mg L1 for T. pseudonana, 209 mg L1 for D. tertiolecta and 240 mg L1 for R. salina (Table 1). In contrast, the overall EC50 in natural phytoplankton communities relative to uncontaminated controls was 80 mg L1 (Fig. 3). To investigate shifts in community composition, data for all measured phytoplankton groups were pooled and analyzed using a two-
Table 2 Concentrationeresponse curves of natural phytoplankton communities to bromoacetic acid (BAA), as shown by growth measured as high performance liquid chromatographic chlorophyll a inhibition in treated incubations relative to controls. Inhibition was calculated based on relative change in algal group abundance in chlorophyll a units of each treatment compared to the controls at the end of the exposure (36 h) for each experiment before being pooled for response estimations. Phytoplankton group
Concentrationeresponse curve equation
Cryptophytes Diatoms Dinoflagellates Prasinophytes Haptophytes
%Inhibition %Inhibition %Inhibition %Inhibition %Inhibition
¼ ¼ ¼ ¼ ¼
0 0 0 0 0
þ þ þ þ þ
(100-0)/(1 (100-0)/(1 (100-0)/(1 (100-0)/(1 (100-0)/(1
þ þ þ þ þ
(33.139/[BAA])1.424 (79.595/[BAA])2.386 (21.912/[BAA])1.882 (31.138/[BAA])1.910 (156.045/[BAA])5.467
R2
EC50
0.691 0.759 0.606 0.562 0.784
33 mg L1 80 mg L1 22 mg L1 31 mg L1 156 mg L1
374
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
Fig. 7. Chlorophyte response to bromoacetic acid as measured by high performance liquid chromatographic chlorophyll a concentration.
factor MANOVA (Fig. 4). Significant differences were confirmed (p < 0.0001) between concentrations of BAA added among all treatments. Discriminant analysis, using all phytoplankton groups as variables, was used to further explore the MANOVA results by classifying community composition using discriminant functions. For Experiment 1 (May 2014), the first two discriminant functions explained 92 and 6.9% of the variance, respectively (cumulative total ¼ 98.9%). The group centroids, based on canonical discriminant functions, suggest similar community compositions among BAA treatments above 50 mg L1 (90, 162, 294 and 524 mg L1), with the 50 mg L1 and control incubations each showing distinct differences (Fig. 5). Discriminant analysis on Experiment 2 (June 2014) indicated no significant differences among any BAA treatments. For Experiment 3 (July 2014), the first two discriminant functions explained 79.5 and 20.5% of the variance, respectively (cumulative total ¼ 100%). The group centroids suggest similar community compositions among the four lowest treatments (5, 9, 16 mg L1 and the control incubations), the middle three treatments (29, 40 and 52 mg L1) and those above 52 mg L1 (60, 85, 94, 170 mg L1) (Fig. 5). To investigate any toxicity threshold differences among individual algal classes, the individual concentrationeresponse curves were plotted separately (Fig. 6) and indicate that toxicity thresholds ranged from 31 to 156 mg L1 (Table 2). Although chlorophytes were present, their response did not fit the sigmoidal concentrationeresponse curve. Only at low (<20 mg L1) and high (>90 mg L1) concentrations of BAA, were the chlorophytes completely inhibited (Fig. 7). There even appears to be a slight increase in chlorophyte concentration at mid-range levels of BAA dose, which was not seen in the representative cultured chlorophyte, D. tertiolecta. 4. Discussion The complete inhibition of all phytoplankton cultures treated with the two highest doses (291 mg L1 and 524 mg L1) suggests that BAA is algicidal past the toxicity threshold (Fig. 1). Published literature has suggested that R. salina may be more sensitive to some contaminants than other phytoplankton species, which is evident at the middle dose, 162 mg L1, where BAA acid is algistatic in R. salina (cryptophyte) but has no effect on either T. pseudonana
(diatom) or D. tertiolecta (chlorophyte) (Hampel et al., 2001). The DCMU results (Fig. 2) further suggest that BAA is lethal at the two highest doses. DCMU blocks the electron flow in Photosystem II, and the ratio between bulk fluorescence and the DCMUinhibited fluorescence is an indicator of photosynthetic capacity. When this ratio declines, it suggests that the phytoplankton cells are unable to photosynthesize properly. The two highest doses of BAA (291 and 524 mg L1) appear to have greatly impacted photosynthetic capacity, suggesting that the mode of action for BAA may be related to photosynthesis (Fig. 2). However, it is evident that the photosynthetic capacity of T. pseudonana declined even with the lowest doses. One explanation for the decline of photosynthetic capacity during the T. pseudonana exposure is that, unlike the other two species exposed, T. pseudonana was spiked during the senescence phase. This is the most sensitive phase of the phytoplankton life cycle, and suggests that BAA may be lethal at lower concentrations if exposed as the cells reach stationary growth and decline. Curve shapes and EC50 values (194e240 mg L1) for all species suggest minimal toxicity differences among these cultured phytoplankton, despite phylogenetic diversity (Fig. 1). The steep concentrationeresponse curves suggest a toxicity threshold, where there is no or minimal toxic effect below and complete cell death above. Single-species toxicity tests are often conservative estimates of true community response because environmental stressors, interspecies competition, and nutrient availability may make natural phytoplankton communities more sensitive to toxicant exposure (Cairns, 1992; De Laender et al., 2009). This study was no exception, where the EC50 of BAA on the bulk phytoplankton community chl a inhibition (80 mg L1) was lower than all of the single algal species EC50 values (194e240 mg L1) (Fig. 3). Although the coefficient of determination (R2) for this concentrationeresponse curve (based on relative change in chl a) is high (0.873), there is a certain level of response variability below doses of 100 mg L1. These inconsistencies may be due to slightly differing phytoplankton communities (confirmed by the MANOVA test) across the three different incubations skewing the experimental data points (Figs. 4 and 6). The discriminant analysis at each treatment level suggests differences in community composition (Fig. 5) within this range, with phytoplankton population shifts at ~30 mg L1 and again at ~50 mg L1. Concentration-response curves for individual natural phytoplankton groups (Fig. 6) indicate some variability in EC50 values (Table 2). As different phytoplankton have different responses to BAA, it is apparent that the concentrationeresponse curve for the entire population will have some variability. Although the chlorophytes could not be fitted to a concentrationeresponse curve, and therefore an EC50 value could not be calculated, their response is noteworthy. Chlorophytes may be opportunistic at the medium doses when competition with other algal groups is reduced due to the lower lethal toxicities for other algal groups (such as the dinoflagellates). A reduction of these competing groups at lower concentrations of BAA may open up a niche for chlorophytes at medium dose levels (50 mg L1). This response was not seen in the cultured D. tertiolecta, suggesting more work should be done to qualify this specific difference in response between natural and cultured phytoplankton. 5. Conclusions The results of these experiments suggest that the halfmaximum effective concentration of BAA to natural phytoplankton communities is below that of single-species algal cultures. Individual phytoplankton groups respond with widely varying sensitivity, making community composition an important factor in
A.R. Gordon et al. / Environmental Pollution 206 (2015) 369e375
determining overall community responses to dissolved toxicants. The reduction in photosynthetic capacity of the cultured phytoplankton at high doses of BAA suggests that the mode of action for BAA may be related to photosynthesis, although further work should be done for positive confirmation. The difference in the toxicity threshold between cultured and natural phytoplankton suggest that standard acute laboratory toxicology tests may not provide an accurate estimate of environmental health, and emphasize the importance of confirming lab experiments with field work whenever possible. Acknowledgments The authors would like to acknowledge Benson Solomon and the Ferry Lab at the University of South Carolina for assistance with the chemical analysis of BAA in solution. References Abarnou, A., Miossec, L., 1992. Chlorinated waters discharged to the marine environment chemistry and environmental impact. An overview. Sci. Total Environ. 126 (1), 173e197. Agus, E., Voutchkov, N., Sedlak, D.L., 2009. Disinfection by-products and their potential impact on the quality of water produced by desalination systems: a literature review. Desalination 237 (1), 214e237. Albert, A., Serjeant, E.P., 1984. The Determination of Ionization Constants, vol. 2. Chapman and Hall, New York. Allonier, A.S., Khalanski, M., Camel, V., Bermond, A., 1999. Characterization of chlorination by-products in cooling effluents of coastal nuclear power stations. Mar. Pollut. Bull. 38 (12), 1232e1241. Arar, E.J., Collins, G.B., 1997. Method 445.0: In Vitro Determination of Chlorophyll a and Pheophytin a in Marine and Freshwater Algae by Fluorescence. United States Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory. Bougis, P., 1976. Marine Plankton Ecology. Bowden, D.J., Clegg, S.L., Brimblecombe, P., 1998. The Henry's law constants of the haloacetic acids. J. Atmos. Chem. 29 (1), 85e107. Cairns Jr., J., 1992. The threshold problem in ecotoxicology. Ecotoxicology 1 (1), 3e16. Cullen, J.J., Renger, E.H., 1979. Continuous measurement of the DCMU-induced fluorescence response of natural phytoplankton populations. Mar. Biol. 53 (1), 13e20. Dean, J.A., Lange, N.A., 1992. Lange's Handbook of Chemistry, 14th ed. McGraw-Hill, New York. Print. De Laender, F., De Schamphelaere, K.A., Vanrolleghem, P.A., Janssen, C.R., 2009. Comparing ecotoxicological effect concentrations of chemicals established in multi-species vs. single-species toxicity test systems. Ecotoxicol. Environ. Saf. 72 (2), 310e315. DeLorenzo, M.E., 2009. Utility of Dunaliella in ecotoxicity testing. In: BenAmotz, Ami, Polle, Jurgen, Subba Rao, D.V. (Eds.), The Alga Dunaliella: Biodiversity, Physiology, Genomics & Biotechnology. Science Publishers, New Hampshire, USA. Ding, G., Zhang, X., Yang, M., Pan, Y., 2013. Formation of new brominated disinfection byproducts during chlorination of saline sewage effluents. Water Res. 47 (8), 2710e2718. ^ Dojlido, J., Zbie c, E., Swietlik, R., 1999. Formation of the haloacetic acids during ozonation and chlorination of water in Warsaw waterworks (Poland). Water Res. 33 (14), 3111e3118. Ford, Q.L., Burns, J.M., Ferry, J.L., 2007. Aqueous in situ derivatization of carboxylic acids by an ionic carbodiimide and 2, 2, 2-trifluoroethylamine for electroncapture detection. J. Chromatogr. A 1145 (1), 241e245. Giller, S., Le Curieux, F., Erb, F., Marzin, D., 1997. Comparative genotoxicity of halogenated acetic acids found in drinking water. Mutagenesis 12 (5), 321e328. n, L.M., Blasco, J., 2001. Acute Hampel, M., Moreno-Garrido, I., Sobrino, C., Lubia toxicity of LAS homologues in marine microalgae: esterase activity and
375
inhibition growth as endpoints of toxicity. Ecotoxicol. Environ. Saf. 48 (3), 287e292. Hanson, M.L., Solomon, K.R., 2004. Haloacetic acids in the aquatic environment. Part II: ecological risk assessment. Environ. Pollut. 130 (3), 385e401. Hashimoto, S., Azuma, T., Otsuki, A., 1998. Distribution, sources, and stability of haloacetic acids in Tokyo Bay, Japan. Environ. Toxicol. Chem. 17 (5), 798e805. Higgins, H.W., Wright, S.W., Schlüter, L., 2011. Quantitative interpretation of chemotaxonomic pigment data. In: Roy, S., Llewellyn, C.A., Egeland, E.S., Johnsen, G. (Eds.), Phytoplankton Pigments. Cambridge University Press, NY, pp. 257e313. Latasa, M., 2007. Improving estimations of phytoplankton class abundances using CHEMTAX. Mar. Ecol. Prog. Ser. 329, 13e21. LeBel, G.L., Benoit, F.M., Williams, D.T., 1997. A one-year survey of halogenated disinfection by-products in the distribution system of treatment plants using three different disinfection processes. Chemosphere 34 (11), 2301e2317. Lewitus, A.J., White, D.L., Tymowski, R.G., Geesey, M.E., Hymel, S.N., Noble, P.A., 2005. Adapting the CHEMTAX method for assessing phytoplankton taxonomic composition in southeastern US estuaries. Estuaries 28 (1), 160e172. Liu, J., Zhang, X., 2014. Comparative toxicity of new halophenolic DBPs in chlorinated saline wastewater effluents against a marine alga: halophenolic DBPs are generally more toxic than haloaliphatic ones. Water Res. 65, 64e72. Mackey, M.D., Mackey, D.J., Higgins, H.W., Wright, S.W., 1996. CHEMTAX-a program for estimating class abundances from chemical markers: application to HPLC measurements of phytoplankton. Mar. Ecol. Prog. Ser. 144, 265e283. Masters, G.M., Ela, W.P., 2008. Introduction to Environmental Engineering and Science, third ed. Pearson Education, Inc, Upper Saddle River, NJ. Print. Motulsky, H., Christopoulos, A., 2004. Fitting Models to Biological Data Using Linear and Nonlinear Regression: a Practical Guide to Curve Fitting. Oxford University Press. €llqvist, T., 1989. Methods for growth inhibition toxicity tests with Nyholm, N., Ka freshwater algae. Environ. Toxicol. Chem. 8 (8), 689e703. Pinckney, J.L., Richardson, T.L., Millie, D.F., Paerl, H.W., 2001. Application of photopigment biomarkers for quantifying microalgal community composition and in situ growth rates. Org. Geochem. 32 (4), 585e595. Pinckney, J.L., Lee, A.R., 2008. Spatiotemporal patterns of subtidal benthic microalgal biomass and community composition in Galveston Bay, Texas, USA. Estuaries Coasts 31 (2), 444e454. Plewa, M.J., Simmons, J.E., Richardson, S.D., Wagner, E.D., 2010. Mammalian cell cytotoxicity and genotoxicity of the haloacetic acids, a major class of drinking water disinfection by-products. Environ. Mol. Mutagen. 51 (8e9), 871e878. Rand, G.M., Petrocelli, S.R., 1985. Fundamentals of Aquatic Toxicology: Methods and Applications. FMC Corporation, Princeton, NJ. Richardson, S.D., Ternes, T.A., 2011. Water analysis: emerging contaminants and current issues. Anal. Chem. 83 (12), 4614e4648. Roberts, J.F., van Egmond, R., Price, O.R., 2010. Toxicity of haloacetic acids to freshwater algae. Ecotoxicol. Environ. Saf. 73 (1), 56e61. Roy, S., Llewellyn, C.A., Egeland, E.S., Johnsen, G. (Eds.), 2011. Phytoplankton Pigments: Characterization, Chemotaxonomy and Applications in Oceanography. Cambridge University Press. Scott, B.F., MacTavish, D., Spencer, C., Strachan, W.M., Muir, D.C., 2000. Haloacetic acids in Canadian lake waters and precipitation. Environ. Sci. Technol. 34 (20), 4266e4272. Scott, B.F., Spencer, C., Marvin, C.H., MacTavish, D.C., Muir, D.C., 2002. Distribution of haloacetic acids in the water columns of the Laurentian Great Lakes and Lake Malawi. Environ. Sci. Technol. 36 (9), 1893e1898. Sournia, A., 1978. Phytoplankton Manual. Unesco, Paris. Print. Taylor, C.J., 2006. The effects of biological fouling control at coastal and estuarine power stations. Mar. Pollut. Bull. 53 (1), 30e48. Welschmeyer, N.A., 1994. Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol. Oceanogr. 39 (8), 1985e1992. Williams, D.T., LeBel, G.L., Benoit, F.M., 1997. Disinfection by-products in Canadian drinking water. Chemosphere 34 (2), 299e316. Wright, S.W., Thomas, D.P., Marchant, H.J., Higgins, H.W., Mackey, M.D., Mackey, D.J., 1996. Analysis of phytoplankton of the Australian sector of the Southern Ocean: comparisons of microscopy and size frequency data with interpretations of pigment HPLC data using the CHEMTAX matrix factorisation program. Mar. Ecol. Prog. Ser. Oldend. 14 (1), 285e298. Yang, M., Zhang, X., 2013. Comparative developmental toxicity of new aromatic halogenated DBPs in a chlorinated saline sewage effluent to the marine polychaete Platynereis dumerilii. Environ. Sci. Technol. 47 (19), 10868e10876.