- Email: [email protected]
Ecotoxicity of neutral red (dye) and its environmental applications
Ecotoxicology and Environmental Safety 122 (2015) 186–192
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Ecotoxicity of neutral red (dye) and its environmental applications Farzana Kastury a,b,n, Albert Juhasz b, Sabrina Beckmann a, Mike Manefield a a b
School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, NSW 2052, Australia Centre for Environmental Risk Assessment and Remediation, University of South Australia, Mawson Lakes Campus, Adelaide, SA 5095, Australia
art ic l e i nf o
a b s t r a c t
Article history: Received 7 June 2015 Received in revised form 23 July 2015 Accepted 24 July 2015
Neutral red (NR) is a synthetic phenazine with promising prospect in environmental biotechnology as an electron shuttle. Recently, NR injections into coal seam associated groundwater in Australia (final dissolved NR concentration: 8 mM 70.2) were shown to increase methanogenesis up to ten-fold. However, information about NR toxicity to ecological receptors is sorely lacking. The main aim of this study was to investigate the concentration dependent toxicity of NR in microorganisms and plants. Acute toxicity of NR was determined by the modified Microtox™ assay. Microbial viability was determined using Escherichia coli and Bacillus subtilis. Germination and early growth of plants was studied using Lactuca sativa, Daucus carota, Allium cepa and an Australian native Themeda triandra. Lastly, mutagenicity of the coal seam associated groundwater was assessed using the Ames test. The EC50 of acute NR toxicity was determined to be 0.11 mM. The EC50 of microbial viability was between 1 and 7.1 mM NR. Among the concentrations tested, only 0.01, 0.10 and 100 mM of NR significantly affected (po0.001) germination of L. sativa. The EC50 for root elongation in seeds was between 1.2 and 35.5 mM NR. Interestingly, root elongation in seeds was significantly stimulated (p o0.001) between 0.25 and 10 mM NR, showing a hormetic effect. A significant increase in mutagenicity was only observed in one of the three wells tested. The results suggest that the average dissolved NR concentration (8 mM 70.2) deployed in the field trial at Lithgow State Coal Mine, Australia, appears not to negatively impact the ecological receptors tested in this study. & 2015 Elsevier Inc. All rights reserved.
Abbreviations: NR: Neutral red EC10: The concentration where ten percent of the maximum effect is observed EC50: The concentration where fifty percent of the maximum effect is observed LOAEC: Lowest observed adverse effect concentration CFU: Colony forming units SEM: Standard error of the mean PAH: Polycyclic aromatic hydrocarbon Keywords: Coal seam gas Hormesis Phenazine Polyaromatic hydrocarbon Toxicity Dye
1. Introduction There is an increasing awareness that certain chemicals released into the environment in the course of biotechnological application can become persistent hazards, threatening not only human health but that of an entire ecosystem. Neutral red (NR; 3-amino-7-dimethylamino-2-methyl phenazine) is a synthetic phenazine that has found a wide range of applications in multiple disciplines for over a century. It is a tricyclic aromatic amine containing two nitrogen atoms in the aromatic ring structure. Fused aromatic rings generally result in low water solubility, high solid-water distribution ratio and resistance to nucleophilic degradation (Johnsen et al., 2005). However, the presence of a C–N substitution in aromatic compounds, including that found in NR, n Corresponding author at: Centre for Environmental Risk Assessment and Remediation, University of South Australia, Building X, Mawson Lakes, Campus, Adelaide, SA, 5095, Australia. E-mail addresses: [email protected], [email protected] (F. Kastury).
http://dx.doi.org/10.1016/j.ecoenv.2015.07.028 0147-6513/& 2015 Elsevier Inc. All rights reserved.
permits NR to be water soluble, therefore increasing its bioavailability (Kobetičová et al., 2011). This may result in increasing mobility of NR, potentially increasing its toxicity and mutagenicity. Since its earliest known use as a vital stain in 1894 (Koehring, 1930), NR has been broadly used as an intracellular pH indicator (Lamanna and McCracken, 1984), ecological marker (New, 1958), textile dye (Sharma et al., 2009; Zhou, 2001; Sawran et al., 2012), histological stain in cytotoxicity assays (Guerard et al., 2012) and recently as electron shuttles in microbial fuel cells for electricity generation (Park and Zeikus, 2000) and during reductive halogenation (Watanabe et al., 2009; Yee et al., 2010). Despite traditionally being considered relatively non-toxic, Ames tests using Salmonella typhimurium strains TA1535, TA97, TA98, TA100, and TA102 demonstrated that NR may be toxic through metabolic and photo activation (Guerard et al., 2012; Longnecker et al., 1977). Discharge of NR from the textile dye industry is recognised as a serious environmental issue (Sharma et al., 2009; Zhou, 2001). In India, wastewater from the dye industry is often discharged without adequate treatment (Sharma
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
et al., 2009). In China, a total organic dye concentration between 83 and 706 mg/l was recorded in wetlands near dye industries and between 12.3 and 456.2 mg/kg dry weight in surface soil (Zhou, 2001). As detailed by Zhou (2001), dyes may accumulate in crops and plants with the potential consequence of food chain effects. With approximately 1–2 million kg of textile dyes being released into the global aquatic environment each year (Sawran et al., 2012), investigation into the ecotoxicological effect of organic dyes with mutagenic potential such as NR is becoming important. Recently, as a result of the development of coal seam gas industry in Australia, a method has been developed where methane production from a reduced carbon source can be enhanced using NR as an electron mediator (Beckmann et al., 2014). This method can be used in the unconventional gas industry to increase methane yield with promising potential to reduce greenhouse gas emissions as well as dependency on coal fired electricity generation plants (EPA, 2011). This is particularly relevant in the US and China where the recoverable shale gas deposits have been estimated to be 18.8 and 31.6 trillion cubic metres respectively (Wang, 2014). During this process of enhancing methanogenesis from coal seam or oil shale associated groundwater, NR may be introduced in the environment. Because of NR's solubility in water at low concentrations (up to 10 mg/ml), it has the potential of coming in direct contact with microbes, plants and animals. However, other than the Ames test using, information regarding NR's toxicity towards ecological receptors is lacking. This study aimed at determining NR toxicity as part of a larger project investigating its use as a means of accelerating coal digestion to produce biogas. Based on preliminary bench scale data, an optimum NR concentration to enhance methane yield was 0.25 mM. The ecotoxicity study presented here provides data assessing the impact of NR towards ecological receptors for multiple trophic levels at concentrations several orders of magnitude above and below the recommended ‘field’ concentration. It was hypothesised that NR at or below a concentration of 0.25 mM would have minimal impact on ecological receptors.
2. Materials and methods 2.1. NR stock solution A stock solution of 250 mM NR (Sigma-Aldrich) was prepared in filter sterilised MilliQ water and stored at 4 °C in the dark. The solution was stirred thoroughly before each use to dissolve any precipitate that may have formed in storage. All working solutions were made by diluting the stock solution with sterile microbial media (LB10 and LB20) or MilliQ water in the absence of direct light. The presence of precipitates was assessed by visualising a drop (10 ml) of working solution on a microscope slide using an epifluorescence microscope (Olympus BX51WI). 2.2. Microtox™ assay Acute toxicity of NR (0.05–0.25 mM) was assessed using Vibrio fischeri MJ1 in a modified Microtox™ assay. A culture of V. fischeri was grown overnight at 30 °C and 160 rpm shaking in LB20 consisting of tryptone (10 g), NaCl (20 g) and yeast extract (5 g) per litre of MilliQ water. The assay was performed in 96-well microtiter plates using a final volume of 200 ml per well. Untreated media (LB20) was used as a negative control. LB20 amended with 10 ml phenol and 25 mM CuSO4 5H2O were used as positive controls. Following the addition of NR, CuSO4 5H2O or phenol amended LB20 to the wells, plates were inoculated with 20 ml of the overnight V. fischeri culture. Bioluminescence was recorded
187
immediately using a Wallac Victor2 plate reader. Plates were then incubated for 15 min at room temperature (20 °C) and bioluminescence was recorded again. Gamma function: ''the ratio of light lost to light remaining'' was calculated and EC50 (the concentration at which a 50% decrease in bioluminescence occurs) was estimated where Gamma ¼1 (Ribo and Kaiser, 1987). 2.3. Microbial viability assay Concentration dependent microbial viability tests were performed using Escherichia coli to represent enteric Proteobacteria pathogens and Bacillus subtilis to represent Fermicutes soil dwelling microorganisms in the environment. Overnight cultures of both microbes were grown in LB10 (tryptone 10 g, NaCl 10 g and yeast extract 5 g per litre of MilliQ water) at 37 °C and 160 rpm shaking. An aliquot (10 ml) of this overnight culture was used to inoculate sterile media amended with 0–25 mM NR (final volume of 50 ml). Samples (1 ml) were collected after 0, 2, 4, 6, 8 and 24 h and stored in the dark at 4 °C. Colony Forming Units/ml (CFU/ml) were determined using the drop plate method. The generation time during exponential growth was then calculated. 2.4. Seed germination, shoot and root elongation assays Both dicot and monocot seeds were used in the seedling emergence and early growth assay according to OECD Guidelines for Test 208 (OECD, 2006). Lactuca sativa (Lettuce) and Daucus carota (Carrot) were chosen to represent dicot crops, while Allium cepa (Onion) was chosen as a monocot crop. These three seeds were purchased from a local supplier (Randwick, Australia). A second monocot, Themeda triandra (Kangaroo grass) was chosen as an Australian native, purchased from Nindethana Seed Company, South Australia. Sterile MilliQ water was used as a negative control while 0.03 mM Cu2 þ was used as a positive control because this concentration is known to inhibit germination in lettuce (Sharma et al., 2009). Seed Germination Filter Paper (diameter 90 mm, grade 181, Whatman™) was fully soaked in each treatment solution and placed in a petri dish. Fifteen seeds per plate were placed on the filter papers and five plates per treatment were made. To avoid water loss during incubation, the petri dishes were sealed with parafilm and kept in the dark for eight days. Crop seeds were incubated at 20 °C (Sharma et al., 2009) while native seeds were incubated at 30 °C (Mott, 1978). To compensate for water loss from evaporation due to higher temperature in the native species incubation, 1 ml of MilliQ water was added every two days. After eight days, the number of seeds germinated was scored and the length of seedling root and shoot was measured. 2.5. Ames test of coal seam associated groundwater sample Three groundwater samples were collected from the Lithgow State Coal Mine, where NR had been in situ for 6, 10 and 15 months. The samples were filter sterilised and tested for mutagenicity using S. typhimurium TA100 (to test point mutation) and TA98 (to test frameshift mutation) both in the presence and absence of S9 enzymes extracted from rat liver (to test metabolic activation). All strains, reagents and enzymes were purchased from Environmental Bio-detection Products Inc. The Ames test was conducted following methods outlined in the Muta-ChromoPlate™ supplied by Environmental Bio-detection Products Inc. 2.6. Statistical analysis The statistical significance of differences in the microbial viability assays between treatment and untreated media control was
188
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
determined (treating each time point independently) by using ANOVA (Dunnett). The lowest observable adverse effect concentration (where p o0.05) was noted. Logistic dose response curves were used to estimate EC10, EC50 and 95% confidence intervals (Barnes et al., 2003). The significant differences in the seed germination assays between seeds grown in various treatments and MilliQ water was conducted using ANOVA (Dunnett) (treating each treatment independently). The lowest concentration which had a significant adverse effect (where p o0.05) was noted. Logistic dose response curves were generated for the root elongation dataset to estimate EC10, EC50 and 95% confidence intervals (Barnes et al., 2003). Significant increase in mutagenicity in the Ames test was determined according to Gilbert (1980).
3. Results and discussion 3.1. Test for crystal formation NR forms needle-like crystals in solution, lowering the dissolved NR concentration and affecting its toxicity (Repetto et al., 2008). The bioavailability, exposure and subsequent toxicity of a compound are highly dependent on its aqueous concentration (Kobetičová et al., 2011). Therefore, it was important to establish if crystals formed in the range of NR concentrations used in this study. Crystal formation was not observed in the solutions used in this study for the duration of the assays (data not shown), confirming that the concentration and exposure conditions of NR remained unchanged during the assays. 3.2. Microtox™ assay Since its first application in measuring acute toxicity of air pollutants in 1965 (Serat et al., 1965; Ribo and Kaiser, 1987), the Microtox™ test has been widely used as a fast, sensitive and reproducible measure of aquatic (Ribo and Kaiser, 1987) and soil toxicity (Juhasz et al., 2010). In this study, bioluminescence in V. fischeri decreased in proportion to the increase in NR concentration during the 15 min incubation (Fig. 1) and ceased completely in both positive controls (10 ml phenol and 25 mM CuSO4 5H2O; not shown in Fig. 1). A 50% decrease in bioluminescence or the EC50 (where Gamma ¼1) occurred at 0.11 mM NR. Microbial bioluminescence occurs via a series of metabolic reactions and therefore can be linked to the metabolic activity of a bacterium (Ribo and Kaiser, 1987). Previous studies have reported that aromatic compounds may cause toxic effects such as damage to membranes and proteins, affecting function and fluidity (Tam
Fig. 1. Effect of NR (0–25 mM) on V. fischeri MJ1 using a modified Microtox™ assay. Acute toxicity of NR was determined by calculating Gamma (a ratio of light lost to light remaining) following a 15 min exposure to 0.01–25 mM NR using V. fischeri at 20 °C. A Gamma value of 1 denotes the EC50 or acute toxicity of NR (0.11 mM). Values represent the average of 10 replicates.
et al., 2006; Sverdrup et al., 2003). Redox cycling of NR is known to produce reactive oxygen species capable of damaging cell components (Guerard et al., 2012). Therefore, the decrease in light output during the short exposure to NR in this assay was possibly caused by a diversion of nutrients and energy towards V. fischeri cell repair due to damage caused by NR. However, it is noteworthy to mention that NR toxicity may vary in an ecosystem where multiple organic compounds are present, which are sometimes known to alter toxicity synergistically or additively (Breitholtz et al., 2008). Nonetheless, the modified Microtox™ test may be used as an ideal tool for monitoring the health of the ecosystem where NR and other textile dyes are found as contaminants. 3.3. Microbial viability assays Results of concentration dependent toxicity assays with two microorganisms during aerobic growth are given in Fig. 2, representing an enteric pathogen of warm blooded organisms (E. coli) and a ubiquitous soil inhabitant (B. subtilis). The assay revealed a few noteworthy differences in the tolerance of NR between the two microbes. Firstly, a total loss of viability between 2 and 4 h was observed in B. subtilis at NR concentrations Z 5 mM. In contrast, loss of viability in E. coli was observed at higher concentrations (Z10 mM). Secondly, the mean CFU/ml after 24 h in B. subtilis was not significantly different in 0.5 mM NR compared to the untreated control (p 40.05), while a significant difference was observed in E. coli in 0.5 mM NR (p o0.05). Lastly, during the exponential phase in media containing 0.5 mM NR, the generation time in both E. coli and B. subtilis was not significantly different from the untreated control (p 40.05). However, the generation time was two-fold longer in E. coli at 5 mM NR (po 0.01). Because of a sharp decline in the viability of B. subtilis in 5 mM NR, this assay was unable to determine if NR had an effect on generation times at NR concentrations higher than 0.5 mM. However, from the longer lag phase at 0.5 mM NR in B. subtilis, it can be argued that the Lowest Observable Adverse Effect Concentration (LOAEC) in B. subtilis was 0.5 mM compared to 5 mM in E. coli. Natural phenazines, such as pyocyanin produced by Pseudomonas aeruginosa, are known for being toxic to non-phenazine producing microbes by diverting electrons from cytochromes to reactive oxygen species such as O−2 and H2O2, resulting in cell death (Hassan and Fridovich, 1980). It is possible that in addition to affecting function and fluidity by damaging membranes and proteins (Tam et al., 2006; Sverdrup et al., 2003), exposure to NR also produced reactive oxygen species in this study, contributing to cell death at higher NR concentrations than the LOAEC. Interestingly, pyocyanin has been linked with up-regulation of the SoxR regulon that activates the transcription factor SoxS in P. aeruginosa (Dietrich et al., 2008). One of the roles of SoxS is detoxifying reactive oxygen species such as superoxide and nitric oxide (Dietrich et al., 2008). It has also been reported that the upregulation of SoxS by pyocyanin is restricted to bacteria from the family Enterobacteriaceae (Dietrich et al., 2008), which includes E. coli but not B. subtilis. Similar to pyocyanin, NR may have upregulated SoxS production in E. coli, negating the effects of reactive oxygen species at 0.5 and 5 mM NR. This also explains why B. subtilis, which does not produce SoxS, showed a lower toxicity tolerance in this assay than E. coli. However, the decline in population above the LOAEC in both microbes can be attributed to the saturation of the detoxifying enzymes, which has been reported to cause a rapid death rate, where the toxicity appears to suddenly worsen (Turesky, 2002). 3.4. Plant assays Being essential primary producers in most ecosystems, plants
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
189
Fig. 2. Effect of NR on the growth and viability of E. coli and B. subtilis. CFU/ml were determined using the drop plate method at 0–24 h for the NR treatments (0.5–25 mM), the untreated media control (0 mM NR) and the positive control (25 mM CuSO4). Data represent the mean CFU/ml 7 SEM (n¼ 3). * ¼ p o0.05 and **¼ p o0.01. Viability declined sharply at Z 10 mM NR in E. coli and Z 5 mM in B. subtilis within 2 h. Compared to the untreated control, B. subtilis showed a longer lag phase at 0.5 mM NR.
Fig. 3. The effect of NR (0–100 mM) on seedling germination, root and shoot elongation. The crop seeds were incubated for 8 days at 20 °C (Lactuca sativa, Daucus carota and Allium cepa) and non-crop native seed at 30 °C (Themeda triandra). Data represent % seed germination, mean root/shoot length 7 SEM (n¼ 75). *¼ p o 0.05, **¼ p o 0.01 and ***¼ 0.001. NR negatively affected germination in L. sativa only at 0.01, 0.1 and 100 mM and root/shoot growth in all seeds at 1 mM. A significant stimulation in root and shoot elongation occurred between 0.25 and 10 mM in all crop species.
190
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
are particularly relevant for ecotoxicity studies (Sverdrup et al., 2003). The results identify concentration dependent toxic effects of NR on seedling germination, root and shoot elongation for three crops and a fourth non-crop native Australian plant (Fig. 3). 3.4.1. Germination assay According to the Terrestrial plant test (OECD, 2006), the seedling emergence must be at least 70% in the negative control for the test to be considered valid. Germination in all replicates of the untreated crop seed negative control (L. sativa, D. carota and A. cepa), was over 90%, confirming the validity of these assays (OECD, 2006). In contrast, an average seed germination of only 10.7% was achieved in the native Australian variety (T. triandra), which is a noteworthy limitation for this dataset. Germination was completely inhibited in the positive control at a Cu2 þ concentration of 0.03 mM. Low germination of native Australian species is well supported by existing literature (Mitchell et al., 1988; Mott, 1978). Although toxicity studies using native plants are rare, less than 50% germination was recorded for the native Australian Banksia ericifolia, Casuarina distyla and Eucalyptus eximia (Mitchell et al., 1988). It is recommended that the validity requirement is relaxed for native plants so that more non-crop species can be included in ecotoxicity studies in the future. Overall, L. sativa was found to be the most sensitive to NR treatment. Compared to the untreated negative control (0 mM NR), a significant reduction (po 0.001) in germination occurred at low concentrations (0.01 and 0.1 mM), as well as the highest concentration tested (100 mM NR), although there was no significant effect at 0.05 mM (p 40.05). In contrast, germination of D. carota, A. cepa and T. triandra was not affected by NR during the exposure (p 40.05). Previous studies have also reported that seedling germination was not significantly affected by similarly structured aromatic compounds, such as PAHs (Juhasz et al., 2010; Khan et al., 2013). Furthermore, it has been suggested that seedling germination alone may not be predictive of PAH toxicity (Sverdrup et al., 2003), which is also in accord with results observed in this test. Moreover, the results indicate that L. sativa is very sensitive to NR and is ideal for future ecotoxicity studies. Unexpectedly, stimulation in germination and reduction of fungal contamination was observed at NR concentrations between 0.25 and 10 mM for all species. Within this concentration range, germination mostly occurred within 24–48 h, in contrast to 48– 72 h for the remaining NR plates. The relationship where a toxin causes low dose stimulation and high dose inhibition is known as a hormetic dose response (Calabrese and Baldwin, 2000, 2003). The hormetic effect of NR in seed germination assays agrees with recent advances in the knowledge of the hormesis literature. For example, a low dose of 10% postharvest residue increased sugarcane bud germination by 45% compared to controls, while higher doses (25% or 100%) showed toxic effects (Viator et al., 2006). 3.4.2. Root and shoot elongation assay Compared to the untreated control (0 mM NR), T. triandra and L. sativa appeared to be the most sensitive to NR treatment in the root elongation assay (significantly inhibited at 10 mM, p o0.001), followed by D. carota and A. cepa (significantly inhibited Z50 mM, p o0.001). This result agrees with a prior root elongation study using another dye (methyl red), where L. sativa was also found to be more sensitive than D. carota (Lamanna and McCracken, 1984). In contrast, shoot elongation was significantly inhibited only in D. carota at 0.1 mM and Z1 mM (po 0.01), A. cepa and T. triandra at Z10 mM (p o0.05). Interestingly, an unexpected hormesis was also observed in the root and shoot elongation assays with 0.25–10 mM NR, which
paralleled hormesis during the germination assay. Compared to the untreated negative control, maximum root stimulation occurred in L. sativa (1.42 fold), followed by D. carota (1.25 fold) and A. cepa (1.14 fold). Similar to the germination test, root elongation in L. sativa also appeared to be significantly inhibited at 0.1 mM NR (p o0.01). Hormesis in shoot elongation assays was limited to L. sativa and A. cepa. Compared to the negative control, stimulation in L. sativa shoot elongation occurred over a wide NR concentration range: 0.01–1 mM (po0.001), the maximum of 1.52 fold being at 0.25 mM. On the other hand, significant stimulation of 1.22 fold shoot elongation occurred at a single NR concentration (0.25 mM) in A. cepa (p o0.01). The evidence of hormesis found in this study agrees with recent findings. More than 350 instances of hormesis have been found in the literature (Calabrese and Baldwin, 1997) across all taxonomic levels (Stebbing, 1982). For example, 10 mg/L Cr6 þ , Cu2 þ and Ni2 þ was demonstrated to increase Medicago sativa L. (alfalfa) root length by 37%, 54% and 37% respectively, while 20 and 40 mg/L caused significant root inhibition (Peralta et al., 2001). Several oxido-reductases are present on plant plasma membranes, which are capable of producing reactive oxygen species, such as NADH oxidases, peroxidases and NADPH oxidases (Crane and Barr, 1989). Although reactive oxygen species are best known for their toxicity and destructive effects on biological systems, it has been proposed that they are involved in cell signalling pathways and induced during seed germination, early seedling development, aging, as well as during a pathogenic attack (Schopfer et al., 2001; Patnaik et al., 2013). A burst of reactive oxygen species preceding germination was recently discovered by the seed coat and embryo of Raphanus sativus (radish), suggesting that reactive oxygen species may play a role in inducing germination and act as an early growth defence mechanism (Schopfer et al., 2001). In this study, NR may have induced reactive oxygen species production in the seeds, which in turn caused oxidative damage to plant embryos, such as lipid peroxidation, DNA damage or cell death (Patnaik et al., 2013). However, at low NR concentrations, the damage may have disrupted cell homoeostasis and stimulated growth to overcompensate the adverse effect (Stebbing, 1982). It has been suggested that biological systems achieve this overcompensation either by removing end product inhibition in the biosynthetic pathways or by increasing metabolic rates (Stebbing, 1982). This conclusion is in accordance with a recent study where reactive oxygen species generation by Cr6 þ was linked with induction of hormetic root stimulation in A. cepa (Patnaik et al., 2013). Generation of reactive oxygen species induced by NR exposure may also explain the decrease in contamination in the petri dishes where root stimulation was also observed. 3.5. Summary of microbial and plant assays The effective concentration values derived from logistic models of concentration response curves of microbial viability and root elongation assay are given in Table 1. The EC50 of all assays fell within the 1–35.5 mM NR range, suggesting that 50% lethality of NR lies in the mM concentration range. On the other hand, the EC10 fell between 0.1 and 12.9 mM NR, suggesting that 90% of the species in these two tropic level will be unaffected by NR below 0.1 mM. In addition, the lowest EC10 occurred for the native Australian plant (0.1 mM NR), suggesting that native non-crop species are likely to be the most threatened by NR exposure. An important deviation from this trend was the significant inhibition (p o0.001) of L. sativa at concentrations of 0.01 and 0.1 mM NR. The reason for this is currently unknown.
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
Table 1 Summary of logistic models of concentration dependent response curves in microorganisms and plants as the lowest observed adverse effect concentration (LOAEC) and 10% and 50% lethal concentrations (EC10 and EC50). 95% confidence interval is shown in parentheses. LOAEC (mM)
EC10 (mM)
EC50 (mM)
Microorganisms Escherichia coli Bacillus subtilis
0.50 0.50
5.16 (0.88–30.3) 0.37 (0.01–11.3)
7.08 (2.17–23.1) 1.00 (0.07–15.2)
Seed Lactuca sativa Daucus carota Allium cepa Themeda triandra
0.01 0.10 50.0 0.01
1.85 (0.12–28.4) 5.60 (1.70–18.6) 12.9 (2.20–75.5) 0.10 (0.01–0.90)
10.0 (4.5–22.0) 19.9 (10.2–39.1) 35.5 (20.7–60.8) 1.20 (0.70–2.20)
3.6. Ames test of coal seam associated groundwater sample Aromatic amines such as NR are known to become metabolically and photochemically activated into strong mutagens (Guerard et al., 2012; Longnecker et al., 1977). An Ames test was used in this study to assess the mutagenicity of NR impacted groundwater samples from the Lithgow State Coal Mine, NSW, Australia. Strain TA 100 assessed the potential to cause point mutation, whereas strain TA98 assessed the potential to cause frame-shift mutations. An enzyme S9 demonstrated the potential to cause exogenous metabolic activation. An average of 8 mM and a maximum of 24.3 þ/ 0.2 mM aqueous NR concentration was recorded in the groundwater where NR had been in situ for 6, 10 or 15 months (unpublished data). The results (Fig. 4) showed that only the groundwater sample containing NR in situ for 10 months demonstrated a significant increase in point mutations in the absence of metabolic activation. In addition, significant increases were observed in both point and in frame-shift mutations once bio-transformed by the S9 enzymes. However, it has been suggested that chemicals may exhibit synergistic or additive toxicity when they co-occur with one or more chemicals in the environment (Breitholtz et al., 2008). Characterisation of these coal seam associated groundwater samples revealed a mixture of total organic compounds incumbent on the coal in the range of 1–10 mg/L, including aromatic compounds such as anthracene and methylated naphthalene, which may have affected mutagenicity of the samples containing NR. It is not currently understood why only one of the three
191
samples was found to be significantly mutagenic and emphasised the need to study toxicity of coal seam groundwater samples containing a multitude of elements.
4. Conclusion There is an increasing awareness that contaminants from anthropogenic sources concern not only humans but can also threaten entire ecosystems. This study was conducted to generate ecotoxicity data of NR to support a recently discovered environemntal application of NR to enhance methanogenesis from coal seam associated groundwater. During a bench trial, the average aqueous concentration of NR detected in NR treated gas wells was 8 þ/ 0.2 mM and the highest concentration was 24.3þ/ 0.2 mM. The purpose of this study was to investigate if NR will exert any adverse effect on the ecosystem at this concentration. The findings of this study can be summarised as:
The acute toxicity of NR, as determined by the EC50 in the Microtox™ assay, was 0.11 mM.
The lowest EC50 value for microbial viability assays was 1 mM NR.
The lowest EC50 value for plant assays was 10 mM. However, it is
important to note that the lowest observable adverse effect concentration was 10 mM NR in L. sativa, the mechanism of which is not understood. One of the most interesting findings of this study was the evidence of hormesis, observed by significantly longer root and shoot growth, faster germination in the plant assays and concomitant reduction in fungal contamination between the NR concentrations of 0.25–10 mM. Significant increase in mutagenicity was observed in the sample containing NR in situ for 10 months in the absence of S9 enzymes using TA100 only and both in the presence and absence of S9 enzymes using TA98.
Overall, the results of ecotoxicity studies of NR suggest that the average concentration of NR recommended by the bench trial of enhanced methanogenesis, has no adverse ecotoxicological effects. These findings constitute essential background data to enable the development of real time methane production from coal seams, thus extending the lifetime of existing gas well infrastructure and limiting extensive deployment of new wells.
Fig. 4. Mutagenicity of NR amended ground water samples from Lithgow State Coal Mines where NR has been in situ for 6, 10 and 15 months. Samples were incubated at 37 °C for five days after being inoculated with Salmonella typhimurium strains TA100 and TA98. Data represent mean positive wells7SEM (n¼ 3). * ¼ p o0.05, ** ¼p o 0.01 and *** ¼0.001.
192
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
Acknowledgements MM was supported on an Australian Research Council Future Fellowship (FT100100078). This research was funded by ARC Linkage Project LP100100128. We thank University of New South Wales for supporting FK with an Honours Year Scholarship (2014UNSW Honours Year Scholarship (UGCA1120) SCI(SP)).
References Barnes, M., Correll, R., Stevens, D., 2003. A simple spreadsheet for estimating low effect concentrations and associated confidence intervals with logistic dose response curves. In: Proceedings of the Society of Environmental Toxicology and Chemistry (SETAC) Asia/Pacific–Australasian Society for Ecotoxicology (ASE) Meeting. Christchurch, New Zealand. Beckmann, S., Lee, M.J., Manefield, M., Oo, Y.M., 2014. Enhanced Biogas Production. Australia Patent Application # AU2014202663. Breitholtz, M., Nyholm, J.R., Karlsson, J., Andersson, P.L., 2008. Are individual NOEC levels safe for mixtures? A study on mixture toxicity of brominated flame–retardants in the copepod Nitocra spinipes. Chemosphere 72, 1242–1249. Calabrese, E., Baldwin, L., 2000. Chemical hormesis: its historical foundations as a biological hypothesis. Hum. Exp. Toxicol. 19, 2–31. Calabrese, E.J., Baldwin, L.A., 1997. A quantitatively‐based methodology for the evaluation of chemical hormesis. Hum. Ecol. Risk Assess. 3, 545–554. Calabrese, E.J., Baldwin, L.A., 2003. Toxicology rethinks its central belief. Nature 421, 691–692. Crane, F.L., Barr, R., 1989. Plasma membrane oxidoreductases. Crit. Rev. Plant Sci. 8, 273–307. Dietrich, L.E., Teal, T.K., Price-Whelan, A., Newman, D.K., 2008. Redox-active antibiotics control gene expression and community behavior in divergent bacteria. Science 321, 1203–1206. US EPA, 2011. National Greenhouse Gas Emissions Data. Gilbert, R.I., 1980. The analysis of fluctuation tests. Mutat. Res.—Environ. Mutatagen. Relat. Subj. 74, 283–289. Guerard, M., Zeller, A., Singer, T., Gocke, E., 2012. In vitro genotoxicity of neutral red after photo-activation and metabolic activation in the Ames test, the micronucleus test and the comet assay. Mutat. Res. 746, 15–20. Hassan, H.M., Fridovich, I., 1980. Mechanism of the antibiotic action pyocyanine. J. Bacteriol. 141, 156–163. Johnsen, A.R., Wick, L.Y., Harms, H., 2005. Principles of microbial PAH-degradation in soil. Environ. Pollut. 133, 71–84. Juhasz, A.L., Smith, E., Waller, N., Stewart, R., Weber, J., 2010. Bioavailability of residual polycyclic aromatic hydrocarbons following enhanced natural attenuation of creosote-contaminated soil. Environ. Pollut. 158, 585–591. Khan, M.I., Cheema, S.A., Tang, X., Hashmi, M.Z., Shen, C., Park, J., et al., 2013. A battery of bioassays for the evaluation of phenanthrene biotoxicity in soil. Arch. Environ. Contam. Toxicol. 65, 47–55. Kobetičová, K., Šimek, Z., Brezovský, J., Hoffman, J., 2011. Toxic effects of nine polycyclic aromatic compounds on Enchytraeus crypticus in artificial soil in relation to their properties. Ecotoxicol. Environ. Saf. 74, 1727–1733. Koehring, V., 1930. The neutral-red reaction. J. Morphol. 49, 45–137. Lamanna, J.C., McCracken, K.A., 1984. The use of neutral red as an intracellular pH indicator in rat brain cortex in vivo. Anal. Biochem. 142, 117–125. Longnecker, D.S., Curphrey, T.J., Daniel, D.S., 1977. Mutagenicity of neutral red.
Mutat. Res./Fundam. Mol. Mech. Mutagen. 48, 109–111. Mitchell, R., Burchett, M., Pulkownik, A., McCluskey, L., 1988. Effects of environmentally hazardous chemicals on the emergence and early growth of selected Australian plants. Plant Soil. 112, 195–199. Mott, J., 1978. Dormancy and germination in five native grass species from savannah woodland communities of the northern territory. Aust. J. Bot. 26, 621–631. New, J.G., 1958. Dyes for studying the movements of small mammals. J. Mammal. 39, 416–429. OECD, 2006. Test No. 208: Terrestrial Plant Test: Seedling Emergence And Seedling Growth Test, OECD Guidelines for the Testing of Chemicals. OECD Publishing, Paris, Section 2. Park, D.H., Zeikus, J.G., 2000. Electricity generation in microbial fuel cells using neutral red as an electronophore. Appl. Environ. Microbiol. 66, 1292–1297. Patnaik, A.R., Achary, V.M.M., Panda, B.B., 2013. Chromium (VI)-induced hormesis and genotoxicity are mediated through oxidative stress in root cells of Allium cepa L . Plant Growth Regul. 71, 157–170. Peralta, J.R., Gardea-Torresdey, J.L., Tiemann, K.J., Gomez, E., Arteaga, S., Rascon, E., et al., 2001. Uptake and effects of five heavy metals on seed germination and plant growth in Alfalfa (Medicago sativa L.). Bull. Environ. Contam. Toxicol. 66, 727–734. Repetto, G., Del Peso, A., Zurita, J.L., 2008. Neutral red uptake assay for the estimation of cell viability/cytotoxicity. Nat. Protoc. 3, 1125–1131. Ribo, J.M., Kaiser, K.L.E., 1987. Photobacterium phosphoreum toxicity bioassay. I. Test procedures and applications. Toxic. Assess. 2, 305–323. Sawran, B., Pare, B., Acharya, A.D., Jonnalagadda, S.B., 2012. Mineralization and toxicity reduction of textile dye neutral red in aqueous phase using BiOCl photocatalysis. J. Photochem. Photobiol. B Biol. 116, 48–55. Schopfer, P., Plachy, C., Frahry, G., 2001. Release of reactive oxygen intermediates (superoxide radicals, hydrogen peroxide, and hydroxyl radicals) and peroxidase in germinating radish seeds controlled by light, gibberellin, and abscisic acid. Plant Physiol. 125, 1591–1602. Serat, W.F., Budinger Jr ., F.E., Mueller, P.K., 1965. Evaluation of biological effects of air pollutants by use of luminescent bacteria. J. Bacteriol. 90, 832. Sharma, S., Sharma, S., Upreti, N., Sharma, K.P., 2009. Monitoring toxicity of an azo dye methyl red and a heavy metal Cu, using plant and animal bioassays. Toxicol. Environ. Chem. 91, 109–120. Stebbing, A.R.D., 1982. Hormesis—the stimulation of growth by low levels of inhibitors. Sci. Total Environ. 22, 213–234. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Kjær, C., Stenersen, J., 2003. Toxicity of eight polycyclic aromatic compounds to red clover (Trifolium pratense), ryegrass (Lolium perenne), and mustard (Sinapsis alba). Chemosphere 53, 993–1003. Tam, L.T., Eymann, C., Albrecht, D., Sietmann, R., Schauer, F., Hecker, M., et al., 2006. Differential gene expression in response to phenol and catechol reveals different metabolic activities for the degradation of aromatic compounds in Bacillus subtilis. Environ. Microbiol. 8, 1408–1427. Turesky, R.J., 2002. Heterocyclic aromatic amine metabolism, DNA adduct formation, mutagenesis, and carcinogenesis. Drug Metab. Rev. 34, 625–650. Viator, R.P., Johnson, R.M., Grimm, C.C., Richard, E.P., 2006. Allelopathic, autotoxic, and hormetic effects of postharvest sugarcane residue. Agron. J. 98, 1526–1531. Wang, Q., 2014. China should aim for a total cap on emissions. Nature 512, 115. Watanabe, K., Manefield, M., Lee, M., Kouzuma, A., 2009. Electron shuttles in biotechnology. Curr. Opin. Biotechnol. 20, 633–641. Yee, L.H., Aagaard, V., Johnstone, A., Lee, M., Kjelleberg, S.J., Manefield, M., 2010. Development of a treatment solution for reductive dechlorination of hexachloro-1, 3-butadiene in vadose zone soil. Biodegradation 21, 947–956. Zhou, Q., 2001. Chemical pollution and transport of organic dyes in water–soil–crop systems of the Chinese coast. Bull. Environ. Contam. Toxicol. 66, 784–793.
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Ecotoxicity of neutral red (dye) and its environmental applications Farzana Kastury a,b,n, Albert Juhasz b, Sabrina Beckmann a, Mike Manefield a a b
School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, NSW 2052, Australia Centre for Environmental Risk Assessment and Remediation, University of South Australia, Mawson Lakes Campus, Adelaide, SA 5095, Australia
art ic l e i nf o
a b s t r a c t
Article history: Received 7 June 2015 Received in revised form 23 July 2015 Accepted 24 July 2015
Neutral red (NR) is a synthetic phenazine with promising prospect in environmental biotechnology as an electron shuttle. Recently, NR injections into coal seam associated groundwater in Australia (final dissolved NR concentration: 8 mM 70.2) were shown to increase methanogenesis up to ten-fold. However, information about NR toxicity to ecological receptors is sorely lacking. The main aim of this study was to investigate the concentration dependent toxicity of NR in microorganisms and plants. Acute toxicity of NR was determined by the modified Microtox™ assay. Microbial viability was determined using Escherichia coli and Bacillus subtilis. Germination and early growth of plants was studied using Lactuca sativa, Daucus carota, Allium cepa and an Australian native Themeda triandra. Lastly, mutagenicity of the coal seam associated groundwater was assessed using the Ames test. The EC50 of acute NR toxicity was determined to be 0.11 mM. The EC50 of microbial viability was between 1 and 7.1 mM NR. Among the concentrations tested, only 0.01, 0.10 and 100 mM of NR significantly affected (po0.001) germination of L. sativa. The EC50 for root elongation in seeds was between 1.2 and 35.5 mM NR. Interestingly, root elongation in seeds was significantly stimulated (p o0.001) between 0.25 and 10 mM NR, showing a hormetic effect. A significant increase in mutagenicity was only observed in one of the three wells tested. The results suggest that the average dissolved NR concentration (8 mM 70.2) deployed in the field trial at Lithgow State Coal Mine, Australia, appears not to negatively impact the ecological receptors tested in this study. & 2015 Elsevier Inc. All rights reserved.
Abbreviations: NR: Neutral red EC10: The concentration where ten percent of the maximum effect is observed EC50: The concentration where fifty percent of the maximum effect is observed LOAEC: Lowest observed adverse effect concentration CFU: Colony forming units SEM: Standard error of the mean PAH: Polycyclic aromatic hydrocarbon Keywords: Coal seam gas Hormesis Phenazine Polyaromatic hydrocarbon Toxicity Dye
1. Introduction There is an increasing awareness that certain chemicals released into the environment in the course of biotechnological application can become persistent hazards, threatening not only human health but that of an entire ecosystem. Neutral red (NR; 3-amino-7-dimethylamino-2-methyl phenazine) is a synthetic phenazine that has found a wide range of applications in multiple disciplines for over a century. It is a tricyclic aromatic amine containing two nitrogen atoms in the aromatic ring structure. Fused aromatic rings generally result in low water solubility, high solid-water distribution ratio and resistance to nucleophilic degradation (Johnsen et al., 2005). However, the presence of a C–N substitution in aromatic compounds, including that found in NR, n Corresponding author at: Centre for Environmental Risk Assessment and Remediation, University of South Australia, Building X, Mawson Lakes, Campus, Adelaide, SA, 5095, Australia. E-mail addresses: [email protected], [email protected] (F. Kastury).
http://dx.doi.org/10.1016/j.ecoenv.2015.07.028 0147-6513/& 2015 Elsevier Inc. All rights reserved.
permits NR to be water soluble, therefore increasing its bioavailability (Kobetičová et al., 2011). This may result in increasing mobility of NR, potentially increasing its toxicity and mutagenicity. Since its earliest known use as a vital stain in 1894 (Koehring, 1930), NR has been broadly used as an intracellular pH indicator (Lamanna and McCracken, 1984), ecological marker (New, 1958), textile dye (Sharma et al., 2009; Zhou, 2001; Sawran et al., 2012), histological stain in cytotoxicity assays (Guerard et al., 2012) and recently as electron shuttles in microbial fuel cells for electricity generation (Park and Zeikus, 2000) and during reductive halogenation (Watanabe et al., 2009; Yee et al., 2010). Despite traditionally being considered relatively non-toxic, Ames tests using Salmonella typhimurium strains TA1535, TA97, TA98, TA100, and TA102 demonstrated that NR may be toxic through metabolic and photo activation (Guerard et al., 2012; Longnecker et al., 1977). Discharge of NR from the textile dye industry is recognised as a serious environmental issue (Sharma et al., 2009; Zhou, 2001). In India, wastewater from the dye industry is often discharged without adequate treatment (Sharma
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
et al., 2009). In China, a total organic dye concentration between 83 and 706 mg/l was recorded in wetlands near dye industries and between 12.3 and 456.2 mg/kg dry weight in surface soil (Zhou, 2001). As detailed by Zhou (2001), dyes may accumulate in crops and plants with the potential consequence of food chain effects. With approximately 1–2 million kg of textile dyes being released into the global aquatic environment each year (Sawran et al., 2012), investigation into the ecotoxicological effect of organic dyes with mutagenic potential such as NR is becoming important. Recently, as a result of the development of coal seam gas industry in Australia, a method has been developed where methane production from a reduced carbon source can be enhanced using NR as an electron mediator (Beckmann et al., 2014). This method can be used in the unconventional gas industry to increase methane yield with promising potential to reduce greenhouse gas emissions as well as dependency on coal fired electricity generation plants (EPA, 2011). This is particularly relevant in the US and China where the recoverable shale gas deposits have been estimated to be 18.8 and 31.6 trillion cubic metres respectively (Wang, 2014). During this process of enhancing methanogenesis from coal seam or oil shale associated groundwater, NR may be introduced in the environment. Because of NR's solubility in water at low concentrations (up to 10 mg/ml), it has the potential of coming in direct contact with microbes, plants and animals. However, other than the Ames test using, information regarding NR's toxicity towards ecological receptors is lacking. This study aimed at determining NR toxicity as part of a larger project investigating its use as a means of accelerating coal digestion to produce biogas. Based on preliminary bench scale data, an optimum NR concentration to enhance methane yield was 0.25 mM. The ecotoxicity study presented here provides data assessing the impact of NR towards ecological receptors for multiple trophic levels at concentrations several orders of magnitude above and below the recommended ‘field’ concentration. It was hypothesised that NR at or below a concentration of 0.25 mM would have minimal impact on ecological receptors.
2. Materials and methods 2.1. NR stock solution A stock solution of 250 mM NR (Sigma-Aldrich) was prepared in filter sterilised MilliQ water and stored at 4 °C in the dark. The solution was stirred thoroughly before each use to dissolve any precipitate that may have formed in storage. All working solutions were made by diluting the stock solution with sterile microbial media (LB10 and LB20) or MilliQ water in the absence of direct light. The presence of precipitates was assessed by visualising a drop (10 ml) of working solution on a microscope slide using an epifluorescence microscope (Olympus BX51WI). 2.2. Microtox™ assay Acute toxicity of NR (0.05–0.25 mM) was assessed using Vibrio fischeri MJ1 in a modified Microtox™ assay. A culture of V. fischeri was grown overnight at 30 °C and 160 rpm shaking in LB20 consisting of tryptone (10 g), NaCl (20 g) and yeast extract (5 g) per litre of MilliQ water. The assay was performed in 96-well microtiter plates using a final volume of 200 ml per well. Untreated media (LB20) was used as a negative control. LB20 amended with 10 ml phenol and 25 mM CuSO4 5H2O were used as positive controls. Following the addition of NR, CuSO4 5H2O or phenol amended LB20 to the wells, plates were inoculated with 20 ml of the overnight V. fischeri culture. Bioluminescence was recorded
187
immediately using a Wallac Victor2 plate reader. Plates were then incubated for 15 min at room temperature (20 °C) and bioluminescence was recorded again. Gamma function: ''the ratio of light lost to light remaining'' was calculated and EC50 (the concentration at which a 50% decrease in bioluminescence occurs) was estimated where Gamma ¼1 (Ribo and Kaiser, 1987). 2.3. Microbial viability assay Concentration dependent microbial viability tests were performed using Escherichia coli to represent enteric Proteobacteria pathogens and Bacillus subtilis to represent Fermicutes soil dwelling microorganisms in the environment. Overnight cultures of both microbes were grown in LB10 (tryptone 10 g, NaCl 10 g and yeast extract 5 g per litre of MilliQ water) at 37 °C and 160 rpm shaking. An aliquot (10 ml) of this overnight culture was used to inoculate sterile media amended with 0–25 mM NR (final volume of 50 ml). Samples (1 ml) were collected after 0, 2, 4, 6, 8 and 24 h and stored in the dark at 4 °C. Colony Forming Units/ml (CFU/ml) were determined using the drop plate method. The generation time during exponential growth was then calculated. 2.4. Seed germination, shoot and root elongation assays Both dicot and monocot seeds were used in the seedling emergence and early growth assay according to OECD Guidelines for Test 208 (OECD, 2006). Lactuca sativa (Lettuce) and Daucus carota (Carrot) were chosen to represent dicot crops, while Allium cepa (Onion) was chosen as a monocot crop. These three seeds were purchased from a local supplier (Randwick, Australia). A second monocot, Themeda triandra (Kangaroo grass) was chosen as an Australian native, purchased from Nindethana Seed Company, South Australia. Sterile MilliQ water was used as a negative control while 0.03 mM Cu2 þ was used as a positive control because this concentration is known to inhibit germination in lettuce (Sharma et al., 2009). Seed Germination Filter Paper (diameter 90 mm, grade 181, Whatman™) was fully soaked in each treatment solution and placed in a petri dish. Fifteen seeds per plate were placed on the filter papers and five plates per treatment were made. To avoid water loss during incubation, the petri dishes were sealed with parafilm and kept in the dark for eight days. Crop seeds were incubated at 20 °C (Sharma et al., 2009) while native seeds were incubated at 30 °C (Mott, 1978). To compensate for water loss from evaporation due to higher temperature in the native species incubation, 1 ml of MilliQ water was added every two days. After eight days, the number of seeds germinated was scored and the length of seedling root and shoot was measured. 2.5. Ames test of coal seam associated groundwater sample Three groundwater samples were collected from the Lithgow State Coal Mine, where NR had been in situ for 6, 10 and 15 months. The samples were filter sterilised and tested for mutagenicity using S. typhimurium TA100 (to test point mutation) and TA98 (to test frameshift mutation) both in the presence and absence of S9 enzymes extracted from rat liver (to test metabolic activation). All strains, reagents and enzymes were purchased from Environmental Bio-detection Products Inc. The Ames test was conducted following methods outlined in the Muta-ChromoPlate™ supplied by Environmental Bio-detection Products Inc. 2.6. Statistical analysis The statistical significance of differences in the microbial viability assays between treatment and untreated media control was
188
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
determined (treating each time point independently) by using ANOVA (Dunnett). The lowest observable adverse effect concentration (where p o0.05) was noted. Logistic dose response curves were used to estimate EC10, EC50 and 95% confidence intervals (Barnes et al., 2003). The significant differences in the seed germination assays between seeds grown in various treatments and MilliQ water was conducted using ANOVA (Dunnett) (treating each treatment independently). The lowest concentration which had a significant adverse effect (where p o0.05) was noted. Logistic dose response curves were generated for the root elongation dataset to estimate EC10, EC50 and 95% confidence intervals (Barnes et al., 2003). Significant increase in mutagenicity in the Ames test was determined according to Gilbert (1980).
3. Results and discussion 3.1. Test for crystal formation NR forms needle-like crystals in solution, lowering the dissolved NR concentration and affecting its toxicity (Repetto et al., 2008). The bioavailability, exposure and subsequent toxicity of a compound are highly dependent on its aqueous concentration (Kobetičová et al., 2011). Therefore, it was important to establish if crystals formed in the range of NR concentrations used in this study. Crystal formation was not observed in the solutions used in this study for the duration of the assays (data not shown), confirming that the concentration and exposure conditions of NR remained unchanged during the assays. 3.2. Microtox™ assay Since its first application in measuring acute toxicity of air pollutants in 1965 (Serat et al., 1965; Ribo and Kaiser, 1987), the Microtox™ test has been widely used as a fast, sensitive and reproducible measure of aquatic (Ribo and Kaiser, 1987) and soil toxicity (Juhasz et al., 2010). In this study, bioluminescence in V. fischeri decreased in proportion to the increase in NR concentration during the 15 min incubation (Fig. 1) and ceased completely in both positive controls (10 ml phenol and 25 mM CuSO4 5H2O; not shown in Fig. 1). A 50% decrease in bioluminescence or the EC50 (where Gamma ¼1) occurred at 0.11 mM NR. Microbial bioluminescence occurs via a series of metabolic reactions and therefore can be linked to the metabolic activity of a bacterium (Ribo and Kaiser, 1987). Previous studies have reported that aromatic compounds may cause toxic effects such as damage to membranes and proteins, affecting function and fluidity (Tam
Fig. 1. Effect of NR (0–25 mM) on V. fischeri MJ1 using a modified Microtox™ assay. Acute toxicity of NR was determined by calculating Gamma (a ratio of light lost to light remaining) following a 15 min exposure to 0.01–25 mM NR using V. fischeri at 20 °C. A Gamma value of 1 denotes the EC50 or acute toxicity of NR (0.11 mM). Values represent the average of 10 replicates.
et al., 2006; Sverdrup et al., 2003). Redox cycling of NR is known to produce reactive oxygen species capable of damaging cell components (Guerard et al., 2012). Therefore, the decrease in light output during the short exposure to NR in this assay was possibly caused by a diversion of nutrients and energy towards V. fischeri cell repair due to damage caused by NR. However, it is noteworthy to mention that NR toxicity may vary in an ecosystem where multiple organic compounds are present, which are sometimes known to alter toxicity synergistically or additively (Breitholtz et al., 2008). Nonetheless, the modified Microtox™ test may be used as an ideal tool for monitoring the health of the ecosystem where NR and other textile dyes are found as contaminants. 3.3. Microbial viability assays Results of concentration dependent toxicity assays with two microorganisms during aerobic growth are given in Fig. 2, representing an enteric pathogen of warm blooded organisms (E. coli) and a ubiquitous soil inhabitant (B. subtilis). The assay revealed a few noteworthy differences in the tolerance of NR between the two microbes. Firstly, a total loss of viability between 2 and 4 h was observed in B. subtilis at NR concentrations Z 5 mM. In contrast, loss of viability in E. coli was observed at higher concentrations (Z10 mM). Secondly, the mean CFU/ml after 24 h in B. subtilis was not significantly different in 0.5 mM NR compared to the untreated control (p 40.05), while a significant difference was observed in E. coli in 0.5 mM NR (p o0.05). Lastly, during the exponential phase in media containing 0.5 mM NR, the generation time in both E. coli and B. subtilis was not significantly different from the untreated control (p 40.05). However, the generation time was two-fold longer in E. coli at 5 mM NR (po 0.01). Because of a sharp decline in the viability of B. subtilis in 5 mM NR, this assay was unable to determine if NR had an effect on generation times at NR concentrations higher than 0.5 mM. However, from the longer lag phase at 0.5 mM NR in B. subtilis, it can be argued that the Lowest Observable Adverse Effect Concentration (LOAEC) in B. subtilis was 0.5 mM compared to 5 mM in E. coli. Natural phenazines, such as pyocyanin produced by Pseudomonas aeruginosa, are known for being toxic to non-phenazine producing microbes by diverting electrons from cytochromes to reactive oxygen species such as O−2 and H2O2, resulting in cell death (Hassan and Fridovich, 1980). It is possible that in addition to affecting function and fluidity by damaging membranes and proteins (Tam et al., 2006; Sverdrup et al., 2003), exposure to NR also produced reactive oxygen species in this study, contributing to cell death at higher NR concentrations than the LOAEC. Interestingly, pyocyanin has been linked with up-regulation of the SoxR regulon that activates the transcription factor SoxS in P. aeruginosa (Dietrich et al., 2008). One of the roles of SoxS is detoxifying reactive oxygen species such as superoxide and nitric oxide (Dietrich et al., 2008). It has also been reported that the upregulation of SoxS by pyocyanin is restricted to bacteria from the family Enterobacteriaceae (Dietrich et al., 2008), which includes E. coli but not B. subtilis. Similar to pyocyanin, NR may have upregulated SoxS production in E. coli, negating the effects of reactive oxygen species at 0.5 and 5 mM NR. This also explains why B. subtilis, which does not produce SoxS, showed a lower toxicity tolerance in this assay than E. coli. However, the decline in population above the LOAEC in both microbes can be attributed to the saturation of the detoxifying enzymes, which has been reported to cause a rapid death rate, where the toxicity appears to suddenly worsen (Turesky, 2002). 3.4. Plant assays Being essential primary producers in most ecosystems, plants
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
189
Fig. 2. Effect of NR on the growth and viability of E. coli and B. subtilis. CFU/ml were determined using the drop plate method at 0–24 h for the NR treatments (0.5–25 mM), the untreated media control (0 mM NR) and the positive control (25 mM CuSO4). Data represent the mean CFU/ml 7 SEM (n¼ 3). * ¼ p o0.05 and **¼ p o0.01. Viability declined sharply at Z 10 mM NR in E. coli and Z 5 mM in B. subtilis within 2 h. Compared to the untreated control, B. subtilis showed a longer lag phase at 0.5 mM NR.
Fig. 3. The effect of NR (0–100 mM) on seedling germination, root and shoot elongation. The crop seeds were incubated for 8 days at 20 °C (Lactuca sativa, Daucus carota and Allium cepa) and non-crop native seed at 30 °C (Themeda triandra). Data represent % seed germination, mean root/shoot length 7 SEM (n¼ 75). *¼ p o 0.05, **¼ p o 0.01 and ***¼ 0.001. NR negatively affected germination in L. sativa only at 0.01, 0.1 and 100 mM and root/shoot growth in all seeds at 1 mM. A significant stimulation in root and shoot elongation occurred between 0.25 and 10 mM in all crop species.
190
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
are particularly relevant for ecotoxicity studies (Sverdrup et al., 2003). The results identify concentration dependent toxic effects of NR on seedling germination, root and shoot elongation for three crops and a fourth non-crop native Australian plant (Fig. 3). 3.4.1. Germination assay According to the Terrestrial plant test (OECD, 2006), the seedling emergence must be at least 70% in the negative control for the test to be considered valid. Germination in all replicates of the untreated crop seed negative control (L. sativa, D. carota and A. cepa), was over 90%, confirming the validity of these assays (OECD, 2006). In contrast, an average seed germination of only 10.7% was achieved in the native Australian variety (T. triandra), which is a noteworthy limitation for this dataset. Germination was completely inhibited in the positive control at a Cu2 þ concentration of 0.03 mM. Low germination of native Australian species is well supported by existing literature (Mitchell et al., 1988; Mott, 1978). Although toxicity studies using native plants are rare, less than 50% germination was recorded for the native Australian Banksia ericifolia, Casuarina distyla and Eucalyptus eximia (Mitchell et al., 1988). It is recommended that the validity requirement is relaxed for native plants so that more non-crop species can be included in ecotoxicity studies in the future. Overall, L. sativa was found to be the most sensitive to NR treatment. Compared to the untreated negative control (0 mM NR), a significant reduction (po 0.001) in germination occurred at low concentrations (0.01 and 0.1 mM), as well as the highest concentration tested (100 mM NR), although there was no significant effect at 0.05 mM (p 40.05). In contrast, germination of D. carota, A. cepa and T. triandra was not affected by NR during the exposure (p 40.05). Previous studies have also reported that seedling germination was not significantly affected by similarly structured aromatic compounds, such as PAHs (Juhasz et al., 2010; Khan et al., 2013). Furthermore, it has been suggested that seedling germination alone may not be predictive of PAH toxicity (Sverdrup et al., 2003), which is also in accord with results observed in this test. Moreover, the results indicate that L. sativa is very sensitive to NR and is ideal for future ecotoxicity studies. Unexpectedly, stimulation in germination and reduction of fungal contamination was observed at NR concentrations between 0.25 and 10 mM for all species. Within this concentration range, germination mostly occurred within 24–48 h, in contrast to 48– 72 h for the remaining NR plates. The relationship where a toxin causes low dose stimulation and high dose inhibition is known as a hormetic dose response (Calabrese and Baldwin, 2000, 2003). The hormetic effect of NR in seed germination assays agrees with recent advances in the knowledge of the hormesis literature. For example, a low dose of 10% postharvest residue increased sugarcane bud germination by 45% compared to controls, while higher doses (25% or 100%) showed toxic effects (Viator et al., 2006). 3.4.2. Root and shoot elongation assay Compared to the untreated control (0 mM NR), T. triandra and L. sativa appeared to be the most sensitive to NR treatment in the root elongation assay (significantly inhibited at 10 mM, p o0.001), followed by D. carota and A. cepa (significantly inhibited Z50 mM, p o0.001). This result agrees with a prior root elongation study using another dye (methyl red), where L. sativa was also found to be more sensitive than D. carota (Lamanna and McCracken, 1984). In contrast, shoot elongation was significantly inhibited only in D. carota at 0.1 mM and Z1 mM (po 0.01), A. cepa and T. triandra at Z10 mM (p o0.05). Interestingly, an unexpected hormesis was also observed in the root and shoot elongation assays with 0.25–10 mM NR, which
paralleled hormesis during the germination assay. Compared to the untreated negative control, maximum root stimulation occurred in L. sativa (1.42 fold), followed by D. carota (1.25 fold) and A. cepa (1.14 fold). Similar to the germination test, root elongation in L. sativa also appeared to be significantly inhibited at 0.1 mM NR (p o0.01). Hormesis in shoot elongation assays was limited to L. sativa and A. cepa. Compared to the negative control, stimulation in L. sativa shoot elongation occurred over a wide NR concentration range: 0.01–1 mM (po0.001), the maximum of 1.52 fold being at 0.25 mM. On the other hand, significant stimulation of 1.22 fold shoot elongation occurred at a single NR concentration (0.25 mM) in A. cepa (p o0.01). The evidence of hormesis found in this study agrees with recent findings. More than 350 instances of hormesis have been found in the literature (Calabrese and Baldwin, 1997) across all taxonomic levels (Stebbing, 1982). For example, 10 mg/L Cr6 þ , Cu2 þ and Ni2 þ was demonstrated to increase Medicago sativa L. (alfalfa) root length by 37%, 54% and 37% respectively, while 20 and 40 mg/L caused significant root inhibition (Peralta et al., 2001). Several oxido-reductases are present on plant plasma membranes, which are capable of producing reactive oxygen species, such as NADH oxidases, peroxidases and NADPH oxidases (Crane and Barr, 1989). Although reactive oxygen species are best known for their toxicity and destructive effects on biological systems, it has been proposed that they are involved in cell signalling pathways and induced during seed germination, early seedling development, aging, as well as during a pathogenic attack (Schopfer et al., 2001; Patnaik et al., 2013). A burst of reactive oxygen species preceding germination was recently discovered by the seed coat and embryo of Raphanus sativus (radish), suggesting that reactive oxygen species may play a role in inducing germination and act as an early growth defence mechanism (Schopfer et al., 2001). In this study, NR may have induced reactive oxygen species production in the seeds, which in turn caused oxidative damage to plant embryos, such as lipid peroxidation, DNA damage or cell death (Patnaik et al., 2013). However, at low NR concentrations, the damage may have disrupted cell homoeostasis and stimulated growth to overcompensate the adverse effect (Stebbing, 1982). It has been suggested that biological systems achieve this overcompensation either by removing end product inhibition in the biosynthetic pathways or by increasing metabolic rates (Stebbing, 1982). This conclusion is in accordance with a recent study where reactive oxygen species generation by Cr6 þ was linked with induction of hormetic root stimulation in A. cepa (Patnaik et al., 2013). Generation of reactive oxygen species induced by NR exposure may also explain the decrease in contamination in the petri dishes where root stimulation was also observed. 3.5. Summary of microbial and plant assays The effective concentration values derived from logistic models of concentration response curves of microbial viability and root elongation assay are given in Table 1. The EC50 of all assays fell within the 1–35.5 mM NR range, suggesting that 50% lethality of NR lies in the mM concentration range. On the other hand, the EC10 fell between 0.1 and 12.9 mM NR, suggesting that 90% of the species in these two tropic level will be unaffected by NR below 0.1 mM. In addition, the lowest EC10 occurred for the native Australian plant (0.1 mM NR), suggesting that native non-crop species are likely to be the most threatened by NR exposure. An important deviation from this trend was the significant inhibition (p o0.001) of L. sativa at concentrations of 0.01 and 0.1 mM NR. The reason for this is currently unknown.
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
Table 1 Summary of logistic models of concentration dependent response curves in microorganisms and plants as the lowest observed adverse effect concentration (LOAEC) and 10% and 50% lethal concentrations (EC10 and EC50). 95% confidence interval is shown in parentheses. LOAEC (mM)
EC10 (mM)
EC50 (mM)
Microorganisms Escherichia coli Bacillus subtilis
0.50 0.50
5.16 (0.88–30.3) 0.37 (0.01–11.3)
7.08 (2.17–23.1) 1.00 (0.07–15.2)
Seed Lactuca sativa Daucus carota Allium cepa Themeda triandra
0.01 0.10 50.0 0.01
1.85 (0.12–28.4) 5.60 (1.70–18.6) 12.9 (2.20–75.5) 0.10 (0.01–0.90)
10.0 (4.5–22.0) 19.9 (10.2–39.1) 35.5 (20.7–60.8) 1.20 (0.70–2.20)
3.6. Ames test of coal seam associated groundwater sample Aromatic amines such as NR are known to become metabolically and photochemically activated into strong mutagens (Guerard et al., 2012; Longnecker et al., 1977). An Ames test was used in this study to assess the mutagenicity of NR impacted groundwater samples from the Lithgow State Coal Mine, NSW, Australia. Strain TA 100 assessed the potential to cause point mutation, whereas strain TA98 assessed the potential to cause frame-shift mutations. An enzyme S9 demonstrated the potential to cause exogenous metabolic activation. An average of 8 mM and a maximum of 24.3 þ/ 0.2 mM aqueous NR concentration was recorded in the groundwater where NR had been in situ for 6, 10 or 15 months (unpublished data). The results (Fig. 4) showed that only the groundwater sample containing NR in situ for 10 months demonstrated a significant increase in point mutations in the absence of metabolic activation. In addition, significant increases were observed in both point and in frame-shift mutations once bio-transformed by the S9 enzymes. However, it has been suggested that chemicals may exhibit synergistic or additive toxicity when they co-occur with one or more chemicals in the environment (Breitholtz et al., 2008). Characterisation of these coal seam associated groundwater samples revealed a mixture of total organic compounds incumbent on the coal in the range of 1–10 mg/L, including aromatic compounds such as anthracene and methylated naphthalene, which may have affected mutagenicity of the samples containing NR. It is not currently understood why only one of the three
191
samples was found to be significantly mutagenic and emphasised the need to study toxicity of coal seam groundwater samples containing a multitude of elements.
4. Conclusion There is an increasing awareness that contaminants from anthropogenic sources concern not only humans but can also threaten entire ecosystems. This study was conducted to generate ecotoxicity data of NR to support a recently discovered environemntal application of NR to enhance methanogenesis from coal seam associated groundwater. During a bench trial, the average aqueous concentration of NR detected in NR treated gas wells was 8 þ/ 0.2 mM and the highest concentration was 24.3þ/ 0.2 mM. The purpose of this study was to investigate if NR will exert any adverse effect on the ecosystem at this concentration. The findings of this study can be summarised as:
The acute toxicity of NR, as determined by the EC50 in the Microtox™ assay, was 0.11 mM.
The lowest EC50 value for microbial viability assays was 1 mM NR.
The lowest EC50 value for plant assays was 10 mM. However, it is
important to note that the lowest observable adverse effect concentration was 10 mM NR in L. sativa, the mechanism of which is not understood. One of the most interesting findings of this study was the evidence of hormesis, observed by significantly longer root and shoot growth, faster germination in the plant assays and concomitant reduction in fungal contamination between the NR concentrations of 0.25–10 mM. Significant increase in mutagenicity was observed in the sample containing NR in situ for 10 months in the absence of S9 enzymes using TA100 only and both in the presence and absence of S9 enzymes using TA98.
Overall, the results of ecotoxicity studies of NR suggest that the average concentration of NR recommended by the bench trial of enhanced methanogenesis, has no adverse ecotoxicological effects. These findings constitute essential background data to enable the development of real time methane production from coal seams, thus extending the lifetime of existing gas well infrastructure and limiting extensive deployment of new wells.
Fig. 4. Mutagenicity of NR amended ground water samples from Lithgow State Coal Mines where NR has been in situ for 6, 10 and 15 months. Samples were incubated at 37 °C for five days after being inoculated with Salmonella typhimurium strains TA100 and TA98. Data represent mean positive wells7SEM (n¼ 3). * ¼ p o0.05, ** ¼p o 0.01 and *** ¼0.001.
192
F. Kastury et al. / Ecotoxicology and Environmental Safety 122 (2015) 186–192
Acknowledgements MM was supported on an Australian Research Council Future Fellowship (FT100100078). This research was funded by ARC Linkage Project LP100100128. We thank University of New South Wales for supporting FK with an Honours Year Scholarship (2014UNSW Honours Year Scholarship (UGCA1120) SCI(SP)).
References Barnes, M., Correll, R., Stevens, D., 2003. A simple spreadsheet for estimating low effect concentrations and associated confidence intervals with logistic dose response curves. In: Proceedings of the Society of Environmental Toxicology and Chemistry (SETAC) Asia/Pacific–Australasian Society for Ecotoxicology (ASE) Meeting. Christchurch, New Zealand. Beckmann, S., Lee, M.J., Manefield, M., Oo, Y.M., 2014. Enhanced Biogas Production. Australia Patent Application # AU2014202663. Breitholtz, M., Nyholm, J.R., Karlsson, J., Andersson, P.L., 2008. Are individual NOEC levels safe for mixtures? A study on mixture toxicity of brominated flame–retardants in the copepod Nitocra spinipes. Chemosphere 72, 1242–1249. Calabrese, E., Baldwin, L., 2000. Chemical hormesis: its historical foundations as a biological hypothesis. Hum. Exp. Toxicol. 19, 2–31. Calabrese, E.J., Baldwin, L.A., 1997. A quantitatively‐based methodology for the evaluation of chemical hormesis. Hum. Ecol. Risk Assess. 3, 545–554. Calabrese, E.J., Baldwin, L.A., 2003. Toxicology rethinks its central belief. Nature 421, 691–692. Crane, F.L., Barr, R., 1989. Plasma membrane oxidoreductases. Crit. Rev. Plant Sci. 8, 273–307. Dietrich, L.E., Teal, T.K., Price-Whelan, A., Newman, D.K., 2008. Redox-active antibiotics control gene expression and community behavior in divergent bacteria. Science 321, 1203–1206. US EPA, 2011. National Greenhouse Gas Emissions Data. Gilbert, R.I., 1980. The analysis of fluctuation tests. Mutat. Res.—Environ. Mutatagen. Relat. Subj. 74, 283–289. Guerard, M., Zeller, A., Singer, T., Gocke, E., 2012. In vitro genotoxicity of neutral red after photo-activation and metabolic activation in the Ames test, the micronucleus test and the comet assay. Mutat. Res. 746, 15–20. Hassan, H.M., Fridovich, I., 1980. Mechanism of the antibiotic action pyocyanine. J. Bacteriol. 141, 156–163. Johnsen, A.R., Wick, L.Y., Harms, H., 2005. Principles of microbial PAH-degradation in soil. Environ. Pollut. 133, 71–84. Juhasz, A.L., Smith, E., Waller, N., Stewart, R., Weber, J., 2010. Bioavailability of residual polycyclic aromatic hydrocarbons following enhanced natural attenuation of creosote-contaminated soil. Environ. Pollut. 158, 585–591. Khan, M.I., Cheema, S.A., Tang, X., Hashmi, M.Z., Shen, C., Park, J., et al., 2013. A battery of bioassays for the evaluation of phenanthrene biotoxicity in soil. Arch. Environ. Contam. Toxicol. 65, 47–55. Kobetičová, K., Šimek, Z., Brezovský, J., Hoffman, J., 2011. Toxic effects of nine polycyclic aromatic compounds on Enchytraeus crypticus in artificial soil in relation to their properties. Ecotoxicol. Environ. Saf. 74, 1727–1733. Koehring, V., 1930. The neutral-red reaction. J. Morphol. 49, 45–137. Lamanna, J.C., McCracken, K.A., 1984. The use of neutral red as an intracellular pH indicator in rat brain cortex in vivo. Anal. Biochem. 142, 117–125. Longnecker, D.S., Curphrey, T.J., Daniel, D.S., 1977. Mutagenicity of neutral red.
Mutat. Res./Fundam. Mol. Mech. Mutagen. 48, 109–111. Mitchell, R., Burchett, M., Pulkownik, A., McCluskey, L., 1988. Effects of environmentally hazardous chemicals on the emergence and early growth of selected Australian plants. Plant Soil. 112, 195–199. Mott, J., 1978. Dormancy and germination in five native grass species from savannah woodland communities of the northern territory. Aust. J. Bot. 26, 621–631. New, J.G., 1958. Dyes for studying the movements of small mammals. J. Mammal. 39, 416–429. OECD, 2006. Test No. 208: Terrestrial Plant Test: Seedling Emergence And Seedling Growth Test, OECD Guidelines for the Testing of Chemicals. OECD Publishing, Paris, Section 2. Park, D.H., Zeikus, J.G., 2000. Electricity generation in microbial fuel cells using neutral red as an electronophore. Appl. Environ. Microbiol. 66, 1292–1297. Patnaik, A.R., Achary, V.M.M., Panda, B.B., 2013. Chromium (VI)-induced hormesis and genotoxicity are mediated through oxidative stress in root cells of Allium cepa L . Plant Growth Regul. 71, 157–170. Peralta, J.R., Gardea-Torresdey, J.L., Tiemann, K.J., Gomez, E., Arteaga, S., Rascon, E., et al., 2001. Uptake and effects of five heavy metals on seed germination and plant growth in Alfalfa (Medicago sativa L.). Bull. Environ. Contam. Toxicol. 66, 727–734. Repetto, G., Del Peso, A., Zurita, J.L., 2008. Neutral red uptake assay for the estimation of cell viability/cytotoxicity. Nat. Protoc. 3, 1125–1131. Ribo, J.M., Kaiser, K.L.E., 1987. Photobacterium phosphoreum toxicity bioassay. I. Test procedures and applications. Toxic. Assess. 2, 305–323. Sawran, B., Pare, B., Acharya, A.D., Jonnalagadda, S.B., 2012. Mineralization and toxicity reduction of textile dye neutral red in aqueous phase using BiOCl photocatalysis. J. Photochem. Photobiol. B Biol. 116, 48–55. Schopfer, P., Plachy, C., Frahry, G., 2001. Release of reactive oxygen intermediates (superoxide radicals, hydrogen peroxide, and hydroxyl radicals) and peroxidase in germinating radish seeds controlled by light, gibberellin, and abscisic acid. Plant Physiol. 125, 1591–1602. Serat, W.F., Budinger Jr ., F.E., Mueller, P.K., 1965. Evaluation of biological effects of air pollutants by use of luminescent bacteria. J. Bacteriol. 90, 832. Sharma, S., Sharma, S., Upreti, N., Sharma, K.P., 2009. Monitoring toxicity of an azo dye methyl red and a heavy metal Cu, using plant and animal bioassays. Toxicol. Environ. Chem. 91, 109–120. Stebbing, A.R.D., 1982. Hormesis—the stimulation of growth by low levels of inhibitors. Sci. Total Environ. 22, 213–234. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Kjær, C., Stenersen, J., 2003. Toxicity of eight polycyclic aromatic compounds to red clover (Trifolium pratense), ryegrass (Lolium perenne), and mustard (Sinapsis alba). Chemosphere 53, 993–1003. Tam, L.T., Eymann, C., Albrecht, D., Sietmann, R., Schauer, F., Hecker, M., et al., 2006. Differential gene expression in response to phenol and catechol reveals different metabolic activities for the degradation of aromatic compounds in Bacillus subtilis. Environ. Microbiol. 8, 1408–1427. Turesky, R.J., 2002. Heterocyclic aromatic amine metabolism, DNA adduct formation, mutagenesis, and carcinogenesis. Drug Metab. Rev. 34, 625–650. Viator, R.P., Johnson, R.M., Grimm, C.C., Richard, E.P., 2006. Allelopathic, autotoxic, and hormetic effects of postharvest sugarcane residue. Agron. J. 98, 1526–1531. Wang, Q., 2014. China should aim for a total cap on emissions. Nature 512, 115. Watanabe, K., Manefield, M., Lee, M., Kouzuma, A., 2009. Electron shuttles in biotechnology. Curr. Opin. Biotechnol. 20, 633–641. Yee, L.H., Aagaard, V., Johnstone, A., Lee, M., Kjelleberg, S.J., Manefield, M., 2010. Development of a treatment solution for reductive dechlorination of hexachloro-1, 3-butadiene in vadose zone soil. Biodegradation 21, 947–956. Zhou, Q., 2001. Chemical pollution and transport of organic dyes in water–soil–crop systems of the Chinese coast. Bull. Environ. Contam. Toxicol. 66, 784–793.