ARTICLE pubs.acs.org/est
Sorption to Dissolved Humic Acid and Its Impacts on the Toxicity of Imidazolium Based Ionic Liquids Zhen Zhang,†,‡ Jing-fu Liu,*,† Xiao-qiang Cai,† Wei-wei Jiang,§ Wen-ru Luo,† and Gui-bin Jiang† †
State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China ‡ School of the Environment, Jiangsu University, Zhenjiang 212013, China § State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China
bS Supporting Information ABSTRACT: Two typical ionic liquids (ILs), 1-butyl-3-methylimidazolium chloride ([C4MIM]Cl) and 1-octyl-3-methylimidazolium chloride ([C8MIM]Cl), are demonstrated to associate strongly with dissolved organic matter (DOM) with distribution coefficients (KDOC) in the range of 104.2 to 104.6 for Aldrich humic acid (used as model DOM). With the increase of humic acid concentration to 11 μg/mL DOC (dissolved organic carbon), the free fraction (ratio of freely dissolved to total concentration) of [C4MIM]Cl and [C8MIM]Cl reduced to about 0.85 and 0.79, respectively. This reduction of freely dissolved concentration gave rise to remarkable reduction of bioavailability and toxicity of the two ILs. MTT assay with HepG2 cell lines showed that the EC50 values were 459 μmol/L for [C4MIM]Cl and 12 μmol/L for [C8MIM]Cl, respectively, and the cell viability increased about 50% in the presence of trace amount of humic acid (1 μg/mL DOC). The SOS/umu test indicated mutagenicity for [C4MIM]Cl at levels above 664 μmol/L, and the genotoxicity was diminished with the addition of trace humic acid (0.00000374-0.374 μg/mL DOC). The studied ILs showed acute toxicity toward model fish medaka with a 96 h median lethal concentration (LC50) of 2254 μmol/L for [C4MIM]Cl and 366 μmol/L for [C8MIM]Cl. The addition of humic acid (5.49 μg/mL DOC for [C8MIM]Cl, 1.37 μg/mL DOC for [C4MIM]Cl) to IL solutions reduced the death rate of medaka to a minimum value of ∼25% of that at zero DOC. Our results suggest that DOM may play an important role in determining the environmental fate and toxicity of imidazolium-based ILs, and its effects should be taken into account in assessing the environmental risk of ILs.
’ INTRODUCTION Room temperature ionic liquids (ILs), resulting from the combination of organic cations and various anions that are liquids at room temperature, are gaining wide recognition as novel solvents for various applications especially as a medium for organic synthesis and catalysis.1 In view of their potential large production quantity and wide applications, ILs will inevitably be released into the environment. Although ILs are considered as green solvents mainly due to their lack of vapor pressure, ILs have relatively low hydrophobicity with KOW < 152 and will likely be introduced into the environment through water. Researchers are more and more concerned the potential environmental impacts of ILs,3-5 as experiments showed that they exhibited toxicities toward different levels of biological complexity as well as several environmental compartments, including reduction of the activity of enzyme3,6 and antibacterial ability,7,8 and inhibition the growth of algae9-11/ duckweed.10,12-14 ILs also exhibited toxicity to cells3,15-19 and invertebrates such as Daphnia magna11,20 and mollusc21,22 as well as vertebrates like zebrafish11,23 and rats/mice.24 Currently most toxicological studies focused on the effects of individual ILs, and a few studies considered the influence of coexisted environmental matrix.13,14,25 Considering the sorption of the incorporated imidazolium cations of ILs to the mineral surfaces and organic matter found in both aquatic sediments and terrestrial soils,13,26-32 it is of great importance to study the r 2011 American Chemical Society
effects of environmental matrix like the organic mater on the toxicity of ILs. Very recently, it was demonstrated that an increase in organic matter and clay contents in soils was observed to decrease the toxicity of imidazolium-based ILs to plants,13,25 and natural dissolved organic matter (DOM) was found to slightly reduce the toxicity of the imidazolium cations to Lemna minor with unclear mechanism.14 Given their similarity to cationic surfactants, it is expected that the cations of imidazolium-based ILs should associate with DOM.33,34 This should reduce the freely dissolved concentration (Cfree) of imidazolium cations and thus the toxicity of ILs, as it is generally believed that only the freely dissolved chemicals are bioavailable.35 Unfortunately, this mechanism has not been verified for ILs due to the lack of approaches for measuring the Cfree of imidazolium cations. The objective of this present work is to study the sorption of ILs to DOM and its effects on the toxicity of ILs. To this end, the Cfree of imidazolium-based ILs was measured through the development of a negligible depletion hollow fiber-protected liquid-phase microextraction (nd-HF-LPME).36 The sorption coefficients of ILs Received: October 18, 2010 Accepted: January 5, 2011 Revised: January 4, 2011 Published: January 14, 2011 1688
dx.doi.org/10.1021/es103514s | Environ. Sci. Technol. 2011, 45, 1688–1694
Environmental Science & Technology to dissolved humic acid were estimated based on the Cfree at various concentrations of humic acid. The influence of humic acid on the toxicity of imidazolium-based ILs was evaluated by testing the cytotoxicity, genotoxicity, and acute toxicity in the presence and absence of humic acid, respectively. While the cytotoxicity was tested based on the viability of HepG2 cell line through MTT assay,37-39 the genotoxicity was evaluated by measuring the induction of umuC gene expression in Salmonella test strains by SOS/umu.40,41 The acute toxicity was tested based on the 96 h median lethal concentration (LC50) to model fish medaka.
’ EXPERIMENTAL SECTION Chemicals and Materials. 1-Butyl-3-methylimidazolium chloride ([C4MIM]Cl) and 1-octyl-3-methylimidazolium chloride ([C8MIM]Cl) with purity of 99% were bought from Henan Lihua Pharmaceutical Co., Ltd. (Henna, China). At room temperature these two chemicals are solid and are miscible with water. Aldrich humic acid (tech., lot. S15539-384, SigmaAldrich) with 35.1% DOC (dissolved organic carbon) was used as obtained, and the stock solution was prepared by dissolution in water filtering through a 0.45 μm membrane. Dimethyl sulfoxide (DMSO) was purchased from Ameresco (ACS grade, purity >99%, USA). Dubelcco’s modified Eagle medium (DMEM) was obtained from Hyclone (Logan, UT, USA). Antibiotic solution (10000 units 1tg/mL penicillin and 10000 1tg/mL streptomycin) and heat inactivated fetal bovine serum was purchased from Gibco BRL Life Technologies (Grand Island, NY, USA). The 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) was obtained from Amresco (Solon, OH, USA). HPLCgrade acetonitrile was provided by Scharlau Chemie (Barcelona, Spain). Bis(2-ethylhexyl) phosphate (DEPHA) and undecane were purchased from Sigma-Aldrich (Steinheim, Germany). Other reagents and chemicals were of analytical grade from Beijing Chemicals (Beijing, China), and ultra pure water (18.3 MΩ) produced with a Milli-Q Gradient system (Millipore, Bedford, USA) was used throughout. The Accurel Q3/2 polypropylene hollow fiber membrane (200 μm wall thickness, 600 μm inner diameter, 0.2 μm pore size) was obtained from Membrana GmbH (Wuppertal, Germany). HepG2 cell lines were provided by Chinese Peking Union Medical College. The test strain Salmonella typhimurium TA1535/ PSK1002 was provided by the Osaka Prefectural Institute of Public Health, Japan. Multimode microplate sepectrophotometer was purchased from Thermo Scientific (FL, USA). Determination of Freely Dissolved Concentration. The Cfree of [C4MIM] and [C8MIM] was determined, respectively, by nd-HF-LPME coupled with high-performance liquid chromatography (HPLC) modified from our previous study.36 Procedures for measurement of the Cfree of the two ILs were developed by optimizing the nd-HF-LPME conditions as described in the Supporting Information (SI). In brief, the disposable nd-HF-LPME device was prepared by cutting the hollow fiber membrane into pieces of ∼5.2 cm and sealing the two ends with heated tweezers and followed by sonicating in the sampling phase (6% and 30% (v/v) DEPHA in undecane for [C8MIM] and [C4MIM], respectively) for 2 h to impregnate the lumen with sampling phase. Before use, the nd-HF-LPME device was taken out and immersed into water with shaking for approximately 1 min to wash out surplus organic solvent. For extraction, one sealed end of the nd-HF-LPME device was clamped with a
ARTICLE
stainless steel clamp to sink the entire sampling device into the 200-mL water sample placed in a capped vial. After shaking at 140 rpm for 2 h, the nd-HF-LPME device was removed from the sample solution and subsequently the two sealed ends were cut open to flush the acceptor (∼11 μL) in the lumen of the hollow fiber into a 200-μL glass vial with a microsyringe. Then, 10 μL of the collected acceptor was manually aspirated into a HPLC microsyringe for injected into the HPLC system for the analysis of [C4MIM] or [C8MIM]. HPLC analysis was performed with an Agilent 1100 series HPLC system (Wilmington, DE, USA), consisting of a binary pump, a variable wavelength detector set at 220 nm, a personal computer equipped with an Agilent ChemStation program for LC for processing chromatographic data, and an Agilent Zorbax Eclipse XDB-C18 column (150 mm 4.6 mm I.D, 5 μm particles) for separation of the analytes. Separation was conducted by isocratic elution with a mobile phase at a flow rate of 1.0 mL/min. The mobile phase consists of 13% acetonitrile, 25% ethanol, and 62% CH3COONH4 buffer (20 mM, pH 7.0) for analysis of [C8MIM], while that for analysis of [C4MIM] consists of 4% ethanol and 96% CH3COONH4 buffer (20 mM, pH 7.0). Estimation of Sorption Coefficients. The sorption coefficients of ILs to dissolved humic acid are estimated based on the free fraction (defined as the ratio between the free and total concentrations) of ILs at various concentrations of humic acid (normalized to DOC). The data were fitted to the following model adapted from Kukkonen et al.42 by using Graphpad Prism (ver. 4.1, GraphPad Software, San Diego, CA) f ¼ 1=ð1 þ KDOC 3 CDOC Þ
ð1Þ
where f is the free fraction, KDOC is the distribution coefficient between DOC and water of the studied chemical, and CDOC is the aqueous concentration of DOC. The value of f was calculated based on the ratio of Cfree to the nominal total concentration. MTT Cell Viability Assay. The HepG2 cell lines were cultured in DMEM (in 75 mL flasks) supplemented with the antibiotic solution (1% of penicillin/streptomycin stock), glutamine, NaHCO3, and 10% fetal bovine serum at 37 °C in a humidified atmosphere of 5% CO2. The cells at a logarithmic phase of growth were used for the exposure tests. Cell viability was measured using MTT assay with 96-well culture plates. Cells were exposed to individual ILs or a mixture of ILs and humic acid at varied concentrations for 24 h, and then 150 μL of MTT stock solution (0.5 mg/mL in PBS, phosphate buffered saline, pH 7.4) was added to each well. After incubation at 37 °C for 4 h, the supernatants were removed, and 150 μL of DMSO was added. The optical density (OD) value of each well was read with an automated microplate reader (model 550, Bio-Rad Laboratories, Hercules, CA, USA) at 490 nm wavelength. Cell viability was calculated by ratio of OD value to that of the control. SOS/umu Bioassay. The genotoxicities of chemical carcinogens and mutagens were determined by measuring induction of umuC gene expression in Salmonella test strains as described in the literature.40,41 Briefly the test strain Salmonella typhimurium TA1535/PSK1002 was incubated in Luria-Bertani medium at 37 °C and 175 rpm overnight, and then the medium was half diluted by tryptone glucose ampicillin medium and cultured at the same condition as previous. After 1 h incubation, the optical density at 600 nm (OD600) was adjusted to 0.5 with the fresh tryptone glucose ampicillin medium. The 96-well microplate was incubated for 2 h at 37 °C, and the OD600 was measured to 1689
dx.doi.org/10.1021/es103514s |Environ. Sci. Technol. 2011, 45, 1688–1694
Environmental Science & Technology
ARTICLE
quantify growth inhibition during the exposure. After enzyme reaction, the OD420 and OD550 were measured to determine βgalactosidase activity. DMSO and 4-nitroquinoline-1-oxide were used as the solvent control and positive control, respectively. The units of β-galactosidase activity (A) was calculated by A ¼ ðOD420 - OD550 1:75Þ=ðOD600 t vÞ
ð2Þ
where OD420, OD550, and OD600 are the optical density at 420 nm, 550, and 600 nm, respectively, t is the incubation duration (min) of the enzyme reaction, and v is the dilution rate of the bacterial suspension (mL). To avoid interferences, the genotoxicity was quantitatively expressed as induction ratios (IR = Asample/Asolvent blank) at a given IL concentration. IR >2.0 means the genotoxic potency of sample to induce a 2-fold increase to that of solvent blank and is considered to be positive, whereas IR