Skip to main content
NCBI home page
Toxicology Research logoLink to Toxicology Research
. 2019 Feb 28;8(3):447–458. doi: 10.1039/c8tx00338f

Cytotoxic effect of protic ionic liquids in HepG2 and HaCat human cells: in vitro and in silico studies†

Bruna Varela Zanoni a, Gabriela Brasil Romão b, Rebecca S Andrade c,✉, Regina Maria Barretto Cicarelli d, Eliane Trovatti a, Bruna Galdorfini Chiari-Andrèo a,d, Miguel Iglesias b
PMCID: PMC6505392  PMID: 31160977

graphic file with name c8tx00338f-ga.jpgThe lower toxicological profile of the studied protic ionic liquids should guide their use as solvents for safe human use.

Abstract

Protic ionic liquids (PILs) are innovative chemical compounds, which due to their peculiar nature and amazing physico-chemical properties, have been studied as potential sustainable solvents in many areas of modern science, such as in the industrial fields of textile dyeing, pharmaceuticals, biotechnology, energy and many others. Due to their more than probable large-scale use in a short space of time, a wider analysis in terms of ecotoxicity and biological safety to humans has been attracting significant attention, once many ionic liquids were found to be “a little less than green compounds” towards cells and living organisms. The aim of this study is to investigate the cytotoxicity of 13 recently synthesized PILs, as well as to reinforce knowledge in terms of key thermodynamic magnitudes. All the studied compounds were tested for their in vitro toxic activities on two human cell lines (normal keratinocytes HaCaT and hepatocytes HepG2). In addition, due to the enormous number of possible combinations of anions and cations that can form ionic liquids, a group contribution QSAR model has been tested in order to predict their cytotoxicity. The estimated and experimental values were adequately correlated (correlation coefficient R2 = 0.9260). The experimental obtained results showed their remarkable low toxicity for the studied in vitro systems.

A. Introduction

Ionic liquids (ILs) are chemicals composed of ionic species that are poorly structured and coordinated, which are most often liquid below 100 °C.1 They are synthetic organic salts that contain one or more cations and one or more anions, being an electrically neutral species, where at least one of them includes organic structures. Typically, anions are of a simple nucleophilic inorganic nature and the organic cations usually have molecular functional groups containing phosphorus or nitrogen atoms (imidazolium, pyridinium, phosphonium and ammonium groups), these structures being hardly coordinated with the negative ions. Due to the existence of delocalized pairs of electrons in such structures, stabilization of crystal cells is avoided by steric impedance, electrostatic interactions being the main force holding the ions together as a whole. ILs are viscous liquids mainly prized for their ability as solvents of any kind of substance. In terms of their miscible character, they can form homogeneous, heterogeneous or partially homogeneous mixtures in polar environments (aqueous or alcoholic solutions, for example), depending on the nature of the ions, so these compounds should be modified or tailored by appropriate variations of the ionic structure. Other key benefits of ionic liquids are their large liquid window, low melting point, thermal and chemical stability and high ionic conductivity, opening unimaginable possibilities in terms of industrial uses. Since the first report about stable ionic liquids a few years ago, they have been the object of intense investigation with promising potential applications, usually with the aim of replacing the volatile, organic and toxic solvents traditionally used in laboratories and industries until now,2 as well as in innumerable chemical and biotechnological processes.3 Regardless of their technological potential, modern environmental policies require a wider and deeper knowledge of any concerning hazardous effects on human health and the environment. Their practical application will certainly lead to human exposure and interaction with air, water and soil, with possible dispersion in nature. Due to these facts, these ionic compounds are finding rising relevance in the last few years, not only being studied from a technical point of view, but also from a toxicological and environmental perspective. If a simple search is made in open literature, it can be observed that the mainstream scientific effort related to ILs is mostly confined to salts of organic amines such as substituted imidazolines, alkyl pyridiniums and trialkylamines with nucleophilic inorganic anions as counter ions, commonly defined as aprotic ionic liquids (AILs). Their environmental safety has been widely discussed and the AILs have been reported, usually, as “green compounds” due to their non-volatility and high thermal stability.1 Despite these potential benefits, they also exhibit many disturbing disadvantages, such as: (i) their resistance to photodegradation and low biodegradability,4 (ii) their potential toxicity5 and (iii) their mutagenic character.6 These results have clearly shown so far that common AIL chemical structures are highly toxic and/or do not pass standardised ISO and OECD toxicological tests. Avoiding discussions on the technical advantages of AILs, these environmental issues make it necessary to find better options with a more environmentally friendly character. In the last few years, a family of ionic compounds known as protic ionic liquids (PILs) are being increasingly studied, which are characterized by being able to destabilize the solid crystalline lattice using a smaller number of atoms and produce intense steric hindrance in both ions. These compounds are known for their low cost of synthesis and more highly biodegradable character, mainly because their structures avoid complex functional groups which are barely biodegradable, or potentially hazardous molecular groups (halogens, heterocycles, etc.). Unlike the AILs, which are predominantly of a hydrophobic nature, the PILs are usually soluble in aqueous media, and recent scientific research suggests their strongly environmentally friendly profile.7 Current environmental policies including REACH (European policy concerning registration, evaluation, authorization and restriction of chemicals) in the European Union, and EPA (Environmental Protection Agency) in the United States, make insistent demands for, of course, the physico-chemical information on and safety data for chemicals and, among other things, information about their cytotoxic profile. Attending to this request, in this work is studied and described the synthesis of 13 protic ionic liquids, their thermodynamic trend in terms of density and isentropic compressibility as a function of temperature and their in vitro toxic effects on two human cell lines (normal keratinocytes HaCaT and hepatocytes HepG2). In addition, due to the enormous number of possible combinations of anions and cations that can form ionic liquids, an attempt was made using a group contribution QSAR model to predict their cytotoxicity. The obtained results quantitatively reinforce the previous observed results in open literature regarding the biodegradable profile and remarkably low toxicity of these compounds.

B. Experimental

PIL synthesis

Amine (mono or diethanolamine) chemical compound of better than 98.0% was placed in a three-necked flask all-made-in-glass mounted in an cool ice bath and equipped with a reflux condenser, a PT-100 temperature sensor for controlling temperature and a dropping funnel. The corresponding acid (formic, lactic, acetic, propionic, adipic, citric, benzoic, maleic or salicylic acid) of better than 98.0% was added dropwise to the flask under vigorous stirring with a magnetic bar. Agitation and reaction was continued for 24 h at laboratory temperature, in order to obtain a final viscous yellow liquid. A purification process was carried out additionally for 48 h, under slight heating, low pressure and agitation in order to ensure elimination of residual reagents and potential atmospheric water pollution. The complete synthesis protocol and spectrometric analysis of the protic ionic liquids used in our experiments have been previously reported.8 Chemical structures and full and abbreviated names are listed in Table 1.

Table 1. PIL molecular structures.

  Cations
Anions graphic file with name c8tx00338f-u1.jpg graphic file with name c8tx00338f-u2.jpg
graphic file with name c8tx00338f-u3.jpg 2-Hydroxy ethylammonium formate (2-HEAF) —
graphic file with name c8tx00338f-u4.jpg 2-Hydroxy ethylammonium acetate (2-HEAA) 2-Hydroxy diethylammonium acetate (2-HDEAA)
graphic file with name c8tx00338f-u5.jpg 2-Hydroxy ethylammonium propionate (2-HEAPr) —
graphic file with name c8tx00338f-u6.jpg 2-Hydroxy ethylammonium lactate (2-HEAL) 2-Hydroxy diethylammonium lactate (2-HDEAL)
graphic file with name c8tx00338f-u7.jpg — 2-Hydroxy diethylammonium benzoate (2-HDEABe)
graphic file with name c8tx00338f-u8.jpg — 2-Hydroxy diethylammonium salicylate (2-HDEASa)
graphic file with name c8tx00338f-u9.jpg — 2-Hydroxy diethylammonium maleate (2-HDEAMa)
graphic file with name c8tx00338f-u10.jpg 2-Hydroxy ethylammonium adipate (2-HEAAd) 2-Hydroxy diethylammonium adipate (2-HDEAAd)
graphic file with name c8tx00338f-u11.jpg 2-Hydroxy ethylammonium citrate (2-HDEACi) 2-Hydroxy diethylammonium citrate (2-HDEACi)
Open in a new tab

PIL thermodynamic data

The densities and ultrasonic velocities of pure ionic components were measured with an Anton Paar DSA-5000 M vibrational tube densimeter and sound analyzer in a temperature range of 293.15–323.15 K, and are presented in the ESI.† The uncertainties (combined expanded uncertainty at the 95% confidence level with a coverage factor of k = 2) of the density and ultrasonic velocities measurements are 0.01 and 0.10%, with a repeatability of 0.000001 g cm–3 and 0.10 m s–1, respectively. Apparatus calibration was performed periodically in accordance with vendor instructions using a double reference with Millipore quality water and ambient air in the range of studied temperatures. Accuracy in the temperature of measurement was better than ±10–2 K by means of a temperature control device that applies the Peltier principle to maintain isothermal conditions during the measurements in the density and ultrasonic velocity cells. The molar mass (MM) and experimental and open literature physicochemical data9 under standard conditions for the studied protic ionic liquids (PIL) are shown in Table 2.

Table 2. Comparison of experimental and literature data for the 13 PILs at 298.15 K, and other relevant information. Standard uncertainties, unc, are unc (ρ) = 0.005% and unc (u) = 0.05% (level of confidence = 95%).

Protic ionic liquid (PIL) Molar mass (g mol–1) Exp. Density (g cm–3) Lit. density (g cm–3) Exp. ultrasonic velocity (ms–1) Lit. ultrasonic velocity (ms–1)
2-HEAF 107.110 1.17817 1.17719a 1840.94 1719.599a
1.21059b 1782.599c
1.17629c 1782.879d
1.19299d
1.03359e
1.15109f
 
2-HEAA 121.137 1.14836 1.14909g 1790.94 1790.739g
1.01779h
1.15359i*
 
2-HDEAA 165.190 1.16748 1.16759d 1863.35 1863.359d
1.17559j a
2-HEAPr 135.163 1.09259 1.12119k a 1636.90 1570.019k a
2-HEAL 151.163 1.21361 1.20169l 1877.02 1865.539l
2-HDEAL 195.216 1.21361 1.21879m 1877.02 1877.649m
2-HDEABe 227.260 1.19931 na 1878.58 na
2-HDEASa 243.260 1.26421 na 2017.06 na
2-HDEAMa 326.347 1.28084 na 2101.14 na
2-HEAAd 268.311 1.19411 na 2013.74 na
2-HDEAAd 356.417 1.22170 na 2010.54 na
2-HEACi 375.377 1.32449 na 2200.72 na
2-HDEACi 507.536 1.29154 na 2113.36 na
Open in a new tab

aFitted data.

PIL cytotoxicity evaluation

The cytotoxic effect of the 13 PILs was assessed by MTT assay using human hepatoma cells (HepG2) and human keratinocytes cells (HaCaT), kindly provided by Professor Vera Isaac (Faculdade de Ciências Farmacêuticas, Universidade Estadual Paulista Júlio de Mesquita Filho – UNESP). MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) is a water soluble tetrazolium salt, which is converted to an insoluble purple formazan by cleavage of the tetrazolium ring by succinate dehydrogenase within the mitochondria. The formazan product is impermeable to the cell membranes and therefore it accumulates in healthy cells. The MTT assay was tested for its validity in various cell lines.10

HepG2 and HaCaT cells were grown in DMEM (Dulbecco's modified Eagle's medium) supplemented with 10% fetal calf serum and antibiotics (penicillin, 100 U mL–1, and streptomycin, 0.1 mg mL–1). The cultures were maintained at 37 ± 2 °C in a 5% CO2 controlled atmosphere. When cells reached 80–90% confluence, they were trypsinized.

DMEM with fetal calf serum was used to neutralize trypsin. The cell suspension was centrifuged for 3 min at 1200 rpm, and transferred to a 96-well plate (TPP, Techno Plastic Product), with a cell density of 4 × 105 cells per mL and 2 × 105 cells per mL, respectively, for the HepG2 and HaCaT cell lines.

The 96-well plates were incubated for 24 h for complete cell adhesion. The cells were treated with 100 μL of positive control (10% dimethyl sulfoxide), negative control (DMEM) and different concentrations of each studied PIL. After 24 h, the treatment was removed and the plates were washed with phosphate buffer saline (PBS) solution. 100 μL of MTT (1 mg mL–1 in PBS) was added to each well. The microplates were incubated at 37 ± 2 °C for 3 h, protected from light, to allow the formation of formazan violet crystals. The solubilization of formazan crystals was carried out by adding 100 μL of isopropyl alcohol to each well. The absorbance was read at 595 nm in a Bio-rad Model 550 spectrophotometer, with an accuracy of ±1.5%, and a reproducibility of ±0.005 O.D. The cytotoxicity assays were performed with at least three independent assays, and, for each one, the treatment was done in triplicate.

The percentage of viable cells represents the cytotoxicity of each treatment, and was calculated relative to the negative control, as proposed by Zhang et al. (2004),11 as follows:

graphic file with name c8tx00338f-t1.jpg 1

where %CV is the percentage of cell viability and ABSnc and ABSPIL are the read absorbances for the negative control and the PIL treated sample, respectively. Knowing that cell apoptosis is induced sometimes by environment hypertonicity,12 and not strictly due to the ionic solvent’s toxic character, morphologic studies were carried out to visualize the HaCat cell behaviour in contact with each PIL. HaCat cells were seeded in 96-well microplates and were then incubated at 37 ± 2 °C in a 5% CO2 controlled atmosphere. After cell adhesion, the culture was exposed to 300 μL of protic ionic liquids of 50 wt% for 8 h to observe cellular morphological changes in both control and treated groups, using an inverted phase contrast microscope (Binocular Primostar Carl Zeiss®, with color-corrected infinity optics, a parfocal distance of 45 mm and a 180 mm tube length) at 40× magnification. Photomicrographs were taken at the beginning of the exposure, and after 1, 3, 5 and 8 hours.

QSAR modelling

In order to perform the QSAR modelling, a database of the IC50 values of protic and aprotic ionic liquids for the HepG2 cytotoxicity test was built, composed using the experimentally obtained IC50 values for our 13 PILs, and the remaining 16 IC50 values taken from the available open literature13 for ILs derived from disubstituted imidazolium, pyridazinium phosponium and choline, which were selected in order to build a larger database of IC50 values. Due to the scarcity of information about the cytotoxicity of ionic liquids to HaCat cells, a database could not be built for this cell line.

The QSAR model was based on the group contribution method, proposed by Luis et al. (2010)14 where molecular properties can be computed as a sum of the contributions of its atoms and/or molecular fragments. Every IL has been divided into basic structural groups, which can contribute to the toxicity profile. The groups were encoded in a Boolean distribution, meaning that they are equal to 1 if the group is present in the IL molecule and to 0 if not. The IC50 values were converted into a logarithmic scale, and the cytotoxicity was expressed as the calculated dimensionless cytotoxicity Inline graphic, according to the following equation:

graphic file with name c8tx00338f-t3.jpg 2

where log IC50max and log IC50min are the highest and the lowest IC50 value in the IC50 database, respectively. The log IC50 is the logarithm of the experimentally obtained IC50 value. The predicted dimensionless toxicity Inline graphic used in this model gathers the sum of the all contributions of the present molecular groups enclosed in the IL, as following:

graphic file with name c8tx00338f-t5.jpg 3

where A, C and S represent the main groups in the ionic liquid: anions, cations and substitutions of the cations, respectively. ai, cj and sk are the fitting coefficients representing the contribution of each molecular group to the total IL cytotoxicity.

C. Results and discussion

Physico-chemical properties

Physicochemical data of protic ionic liquids are important for both the design of future cleaner technological processes and understanding the interactions in such compounds. As with other thermodynamic properties, an enormous gap in information is present in open literature in terms of density, ultrasonic velocity, viscosity or any other parameters of ionic liquids. Only in the last few years has a relative increase in the number of papers devoted to these measurements been observed. A simple search in any popular scientific finder shows that the situation is much worse for protic ionic liquids.8,9

Densities and ultrasonic velocities of these 13 studied protic ionic liquids are given in the ESI.† In Fig. 1–3, the temperature trend of these magnitudes, as well as of the isentropic compressibility (κs), derived from the Laplace–Newton equation, is gathered. The observed inverse dependence on temperature of density and ultrasonic velocity for these protic ionic liquids indicates the special trend of packing and the strong dependence on ion kinetics and polar/dispersive interaction. In general terms, the protic ionic liquids with a higher molecular mass have higher values of density and ultrasonic velocity at any temperature. The observed values for acetates and propionate salts are surprisingly lower than expected9a due to the stronger influence of the steric hindrance of aliphatic residues in the cation and anions. These residual structures produce a huge disturbance in the packing order of ions. If different functional series of salts were compared (for example 2-HEAF, 2-HEAA, 2-HEAPr in terms of anion aliphatic chain length, or 2-HEAA versus 2-HDEAA in terms of functional substitution length), it is clear that the bulk cation is under less influence of the steric hindrance than the linear anion. This fact may be observed by the higher values of density and ultrasonic velocity for those salts with lighter anions (at each case) and heavier cations (for the same anion) (Fig. 2 and 3, for example). As earlier observed, all of them show higher values for both properties if compared with common covalent compounds in the same range of temperatures.

Fig. 1. Curves of density of the 13 studied protic ionic liquids in the temperature range 293.15–323.15 K.

Fig. 1

Open in a new tab

Fig. 2. Curves of ultrasonic velocity of the 13 studied protic ionic liquids in the temperature range 293.15–323.15 K.

Fig. 2

Open in a new tab

Fig. 3. Curves of isentropic compressibility of the 13 studied protic ionic liquids in the temperature range 293.15–323.15 K.

Fig. 3

Open in a new tab

The increasing values of isentropic compressibility with rising temperature suggest that ion kinetics increasingly disturb the flexible relatively ordered bulk lattice under higher thermal conditions to generate a more and more relaxed structure. As observed in Fig. 3, flag structured (2-HDEABe) or short aliphatic (2-HEAA, 2-HEAPr, 2-HDEAL) anions in PILs show the strongest variation of compressibility as a function of temperature.

A frequently applied derived magnitude for industrial compounds is the temperature dependence of volumetry which is expressed as isobaric expansibility or a thermal expansion coefficient (α). The data reported in literature normally only give values of thermal expansion coefficients both of pure compounds and their mixtures, showing the relative changes in density, calculated as a function of temperature and assuming that –Δρ/ρ remains a constant term in any thermal range. As in the case of pure covalent chemicals, it can be computed using the expression:

graphic file with name c8tx00338f-t6.jpg 4

taking into account the temperature dependence of density. These values are included in Fig. 4. All PILs show negative values in the studied range of temperatures. Two tendencies are observed: firstly, a group of PILs show rising values for higher temperatures, mainly PILs of a polyelectrolyte nature (PILs with adipate, citrate, or maleate anions); secondly, a different group show constant values or relatively decreasing values versus temperature (PILs with salicylate, lactate, acetate, lactate, propionate, or benzoate anions). Only 2-HDEABe shows a clearly decreasing tendency with rising temperatures. If each pair of PILs (cations with mono or disubstitutions of the ethyl functional group) was compared (2-HEAAd and 2-HDEAAd, 2-HEACi and 2-HDEACi, 2-HEAA and 2-HDEAA, 2-HEAL and 2-HDEAL), it should be pointed out that the polyelectrolytic PILs show higher values and opposite trends in the following order: 2-HDEAAd > 2-HEAAd > 2-HDEACi > 2-HEACi. As discussed, the monoelectrolyte group shows an inverse trend and almost constant value versus temperature: 2-HEAL > 2-HEAA > 2-HDEAA > 2-HDEAL.

Fig. 4. Curves of isobaric expansibility of the 13 studied protic ionic liquids in the temperature range 293.15–323.15 K.

Fig. 4

Open in a new tab

Cytotoxic effect on HepG2 and HaCat cell lines

Cytotoxicity assays permit the study of cell behaviour in a controlled micro-environment using suitable cell lines.10 The in vitro cell culture has been successfully used in pre-clinical tests in order to predict the toxicity of chemicals15 and could allow the extrapolation of assay data with regard to possible effects on humans.

HEpG2 primary hepatocyte cultures appear to be a very significant in vitro system, as the liver's specific functions and responses to toxic insults are retained for several days up to several weeks.16 This cell line is well characterized for its relevance to toxicity models, and has been used,13,17 as well as HeLa,18 CaCo-2,19 MCF7,20 and PC12,21 to evaluate the cytotoxic effect of aprotic ionic liquids.

Environmental agencies also mandate testing of skin irritation for chemicals that are used at over 1 ton per annum before they are released into the market. Skin is one of the tissues that is directly exposed to toxic substances in the working environment and everyday life; therefore, the outcome of chemical exposure to the skin should be fully studied for industrial chemicals. The human keratinocyte HaCaT, the most abundant cell line of the epidermis, has been widely used in studying cytotoxicity caused by direct contact with hazardous compounds,22 including ionic liquids.

These two cell lines were used with the aim of detecting any damage caused by the studied PIL or its metabolites, and to determine the IC50, the PIL concentration required for achieving 50% inhibition of the cell culture.

This inhibition is characterised by changes in morphology, which include cell membrane disruption, chromosome condensation and apoptotic body formation.23 Apoptosis is a physiological cell suicide program that helps to maintain homeostasis, in which cell death naturally occurs during tissue turnover.24 The cellular changes due to the apoptosis mechanism include the formation of plasma membrane blebs, reduction in cell volume, chromatin condensation and DNA fragmentation, retaining only (to some extent) the organelle integrity.25 The studied PILs showed a time-dependent increase in the number of apparent apoptotic fragments at a specific concentration indicating their capability to induce apoptosis.

The previous visual results for the highest tested concentration (50 wt%) indicated an intrinsic feature of molten salts, i.e., their dissociation capability in aqueous medium, increasing the environmental osmotic forces and provoking cell apoptosis. Fig. 5 shows the morphology of HaCat cells exposed to 50 wt% of each PIL. At the beginning of the exposure time (0 h), it was possible to observe the cell morphology integrity of all tested samples, including the cytoplasmic membrane. The control sample (cell culture not exposed to the PILs) kept its integrity.

Fig. 5. Images of HaCat cells exposed to PILs after 0 h (at the beginning of the exposure), 1 h, 3 h, 5 h and 8 h.

Fig. 5

Open in a new tab

Morphological changes like apoptosis, found in most cell types, can be induced by any highly concentrated material through an osmosis mechanism, provoking cell volume reduction and nucleus condensation, then increasing the cell membrane permeability to water, resulting in the swelling and death of cells. This change is followed by plasma membrane blebbing and nuclear fragmentation to form apoptotic bodies, which could be observed in most of the cells exposed to PILs. In general, the reduction of cell volume was observed in the first hour of exposure (Fig. 5b2–n2). After 3 hours of exposure (Fig. 5b3–n3), cell membranes became permeable to water and swelled. In Fig. 5 b4–n4, after 5 hours of exposure, it is possible to observe the membrane blebbing and its rupture and, in the last hour (Fig. 5 b5–n5), almost all cells suffered damage, probably apoptosis. In some cases, as occurred with the cells exposed to 2-HEAA (Fig. 5c), the cells did not suffer reduction, but on the other hand, they swelled and in the last hour it is possible to observe the rupture of their cell membranes.

These results are insufficient to elucidate PIL cytotoxicity, because the highly hypertonic medium, and not their toxic potential, probably caused the osmotic transport of solvents out from inside the cell, through a well-known inhibition mechanism that can be compared to that caused by any other ionic substance.26 The classical example is NaCl solution that at 0.9 wt% is comfortable to the skin and safe to cells. However, in higher concentrations, it totally inhibits cell growth and promotes plasmolysis.26a,b The effect of growth inhibition was detected for all the PILs at 50 wt%.

The results of morphological assay using 50 wt% PIL were useful to elucidate the mechanism of HaCat cell viability reduction promoted by the tested PILs at high concentrations. Thus, the cytotoxicity experiments were performed with lower PIL concentrations and an additional cell line – HepG2. The IC50 (μM) for HepG2 and HaCat is shown in Table 3.

Table 3. Calculated IC50 values (95% confidence interval) for the tested PILs (values expressed in μM).

PILs IC50 for HepG2 IC50 for HaCat
2-HEAF 91.97 ± 4.01 195.60 ± 1.68
2-HEAA 142.30 ± 0.16 193.80 ± 27.52
2-HDEAA 25.98 ± 3.48 178.40 ± 53.29
2-HEAPr 58.79 ± 2.26 66.78 ± 5.21
2-HEAL 66.13 ± 8.03 372.10 ± 33.96
2-HDEAL 41.24 ± 1.65 88.54 ± 10.56
2-HDEABe 29.11 ± 2.20 54.48 ± 4.42
2-HDEASa 8.91 ± 1.14 48.23 ± 2.34
2-HDEAMa 34.76 ± 0.15 143.80 ± 37.45
2-HEAAd 20.54 ± 4.76 91.52 ± 3.93
2-HDEAAd 28.77 ± 1.23 75.54 ± 34.15
2-HEACi 15.12 ± 0.16 35.67 ± 0.82
2-HDEACi 19.07 ± 1.43 36.92 ± 7.17
Open in a new tab

A chemical’s toxicity depends on its permeation through the cells and tissues, and also on the mechanism of action (interaction with cell organelles, receptors and pathways triggered), which are strictly related to the chemical structure.

Some important details should be considered to correlate the PILs’ chemical structure with their activity, such as the toxicity of the cations/anions that compose each PIL and the hydrophilic/hydrophobic features (partition coefficient) of the molecule.

In general, the obtained results showed a greater cytotoxic effect on HepG2 than on HaCat cells. The concentration of the PILs 2-HEAF, 2-HEACi, 2-HDEAL, 2-HDEABe and 2-HDEACi needed to reduce by 50% the viability of the HaCat cells was approximately twice the concentration necessary to produce the same effect in HepG2 cells. For the samples 2-HEAL, 2-HEAAd, 2-HDEAA, 2-HDEAAd, 2-HDEASa and 2-HDEAMa this effect was increased, with higher IC50 values.

The increased effect of PILs on HepG2 cells can be possibly attributed to the metabolic activity of these cells, which could generate toxic metabolites from the PIL. However, it is reasonable to suppose that simple human skin contact with protic ionic liquids, without permeation, would not allow their metabolism. According to the IC50 values, the most cytotoxic PILs were 2-HDEASa, 2-HEACi and 2-HDEACi. Citric acid is a three-carboxylic acid with three negative charges, which requires three counter ions. Supposing they do not penetrate the cells,27 their dissociation generates six free charges at the external medium, which could disrupt the cell membranes, leading to growth inhibition. This effect can also be correlated to the higher polarity of the samples – polar ionic liquids composed of ether, hydroxyl and nitrile functional groups showed a toxic character when tested in rat cells.28,29

The effect of the anionic chain length was observed, comparing the cytotoxicity results for 2-HEAAd and 2-HDEAAd (six methylene groups at the alkyl chain) with those of 2-HEAA and 2-HEAL (relatively short alkyl chain). The long alkyl chains with a polar head could alter the cell integrity, quickly disrupting the lipid bilayer, resulting in membrane rupture and then final cell death. This highly cytotoxic effect of long chain ionic liquids was discussed in previous work.30

2-HEAF, 2-HEAA, 2-HEAL and 2-HDEAA are short aliphatic PILs, composed of counter anions of small carboxylic acids. The cytotoxicity tests revealed they had the lowest toxic effects, as expected.

QSAR method

In silico methods such as QSAR (quantitative structure–activity relationship) can be used for predicting toxicity and, thus, reduce costly and time-consuming toxicity testing, either in vitro or in vivo.14 The use of QSAR modeling for the prediction of toxicity is recognized by REACH regulation,31 stating that properties determined with the use of in silico methods are equivalent to laboratory testing. The QSAR modelling, as its name indicates, provides toxicity data based on the quantitative relationship between chemical structure and biological and/or toxicological activity, using chemical descriptors generated from the molecular structure. These descriptors are statistically analyzed in order to develop a model that would describe the desired activity, e.g. toxicity.32 Until now, QSAR studies on the toxicity of ILs have been far less frequent compared to those on other chemicals.33

In this work, the experimental cytotoxic data of the studied PILs were compared with previously reported experimental cytotoxic values of other ionic liquids, and this database used to build a predictive model based upon the QSAR modelling method, aiming to elucidate the chemical and structural factors that govern the toxicity of these compounds. A detailed overview of the chemical descriptors used can be found in Table 4, as well as the respective IC50 values converted into a logarithmic scale and the calculated dimensionless cytotoxicity Inline graphic parameters. The dataset of 29 IC50 values was fitted to the QSAR model by multiple linear regression using Sigmaplot software. The calculated dimensionless toxicity obtained from experimental IC50 data and the predicted dimensionless toxicity obtained after QSAR modelling were plotted and the data plot graph is shown in Fig. 6.

Table 4. Ionic liquid cytotoxicity in μmol L–1 for the HepG2 test, calculated dimensionless toxicity Inline graphic and group contribution descriptors.

Ref. No. Compound IC50 (μM) log IC50 graphic file with name c8tx00338f-t9.jpg A1 A2 A3 A4 A5 C1 C2 C3 C4 C5 S1 S2 S3 S4 S5
Liu et al. (2017)13 1 [C2Mim][OAc] 40.02 1.60228 0.09752 0 1 0 0 0 1 0 0 0 0 2 1 0 0 0
Liu et al. (2017)13 2 [C2Mim][Cl] 35.66 1.55218 0.10639 1 0 0 0 0 1 0 0 0 0 2 1 0 0 0
Liu et al. (2017)13 3 [Ch][OAc] 135.53 2.13204 0.00375 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0
Liu et al. (2017)13 4 [Ch][Cl] 105.97 2.02518 0.02266 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0
Liu et al. (2017)13 5 [C2OHMim][Cl] 66.57 1.82328 0.05840 1 0 0 0 0 1 0 0 0 0 2 1 1 0 0
Ferraz et al. (2015)13 6 [P6,6,6,14] 3.22 × 10–4 –3.49214 0.99928 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
Ferraz et al. (2015)13 7 [Ch] 1.62 × 10–3 –2.79075 0.87513 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
Ferraz et al. (2015)13 8 [C2mim] 2.43 × 10–2 –1.61493 0.66700 0 0 0 0 0 1 0 0 0 0 2 1 0 0 0
Ferraz et al. (2015)13 9 [C2OHMim] 3.19 × 10–4 –3.49621 1.00000 0 0 0 0 0 1 0 0 0 0 2 1 1 0 0
Messali et al. (2015)13  a 10 EtO2C(CH2)Cl 17.20 1.23553 0.16244 1 0 0 0 0 0 0 1 0 0 1 0 0 1 0
Messali et al. (2015)13  a 11 MeO2C(CH2)Br 32.20 1.50786 0.11423 1 0 0 0 0 0 0 1 0 0 5 0 0 1 0
Messali et al. (2015)13  a 12 4-NO2C6H4(CH2)Br 15.60 1.19312 0.16994 1 0 0 0 0 0 0 1 0 0 1 0 0 0 1
Zhang et al. (2011)13 13 [C4Mim][Cl] 4.59 × 10–1 –0.33819 0.44100 1 0 0 0 0 1 0 0 0 0 4 1 0 0 0
Zhang et al. (2011)13 14 [C8Mim][Cl] 1.20 × 10–2 –1.92082 0.72114 1 0 0 0 0 1 0 0 0 0 8 1 0 0 0
Jodynis-Liebert et al. (2010)13 15 [DDA][Sac] 4.80 0.68124 0.26055 0 0 4 1 0 0 0 0 0 1 0 0 0 0 0
This Work 16 2-HEAF 91.97 1.96365 0.03355 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0
This Work 17 2-HEAA 142.30 2.15320 0.00000 0 0 1 0 1 0 0 0 0 1 0 0 0 0 0
This Work 18 2-HDEAA 25.98 1.41464 0.13073 0 0 1 0 1 0 0 0 0 2 0 0 0 0 0
This Work 19 2-HEAPr 58.79 1.76930 0.06795 0 0 1 0 2 0 0 0 0 1 0 0 0 0 0
This Work 20 2-HEAL 66.13 1.82040 0.05891 0 0 1 0 2 0 0 0 0 1 0 0 0 0 0
This Work 21 2-HDEAL 41.24 1.61532 0.09521 0 0 1 0 2 0 0 0 0 2 0 0 0 0 0
This Work 22 2-HDEABe 29.11 1.46404 0.12199 0 0 1 1 0 0 0 0 0 2 0 0 0 0 0
This Work 23 2-HDEASa 8.91 0.94988 0.21300 0 0 1 1 0 0 0 0 0 2 0 0 0 0 0
This Work 24 2-HDEAMa 34.76 1.54108 0.10835 0 0 2 0 2 0 0 0 0 2 0 0 0 0 0
This Work 25 2-HEAAd 20.54 1.31260 0.14879 0 0 2 0 4 0 0 0 0 1 0 0 0 0 0
This Work 26 2-HDEAAd 28.77 1.45894 0.12289 0 0 2 0 4 0 0 0 0 2 0 0 0 0 0
This Work 27 2-HEACi 15.12 1.17955 0.17235 0 0 3 0 3 0 0 0 0 1 0 0 0 0 0
This Work 28 2-HDEACi 19.07 1.28035 0.15450 0 0 3 0 3 0 0 0 0 2 0 0 0 0 0
Open in a new tab

aThe given IC50 values in mg mL–1 have been transformed to μM using the respective compound densities.

Fig. 6. Calculated and predicted dimensionless cytotoxicity dataplot for HepG2 assay. The dotted lines delimit the region of residual values <0.10.

Fig. 6

Open in a new tab

Two additional zoomed-in figures are shown in Fig. 6, the upper one indicating the good agreement of the protic ionic liquid values with the QSAR predictions. The lower one shows the clear tendency of absolute deviations of dimensionless toxicity from computed IC50 values, showing a strong decrease for those PILs of lower potential toxicity.

A reasonably good fitting was achieved in the QSAR proposed model, with n = 27, R2 = 0.9260 and 15 descriptors. Residuals were calculated as the differences between the calculated dimensionless toxicity and predicted by QSAR method (taking into account the absolute values). The distribution of the residuals in Table 5 and the main plot graph in Fig. 6 indicate that there are practically no outliers and 85% of the residuals are lower than 0.10. The most and the least toxic ILs were well predicted by the model. The group contribution, as well as a detailed comment on each descriptor, is shown in Table 6.

Table 5. Distribution of residuals between the calculated and predicted dimensionless toxicity and the mean residual for the HepG2 cell assay.

Range Residuals %
<0.10 22 75.58
0.10–0.15 3 10.71
0.15–0.20 3 10.71
Open in a new tab

Table 6. Details and contribution of the descriptors used in the QSAR modelling.

Group Molecular descriptor Comments Contribution
Anion a 1 Influence of anions: chloride or bromide. –0.7137
a 2 Influence of the acetate anion –0.7774
a 3 Influence of the number of carbons in the PIL anion 0.0340
a 4 Influence of the aromatic ring in the PIL anion 0.1070
a 5 Influence of the number of carboxylic groups in the PIL anion 0.0116
 
Cation c 1 Influence of the imidazolium cation 0.3469
c 2 Influence of the phosphonium cation 0.9993
c 3 Influence of the piridazinium cation 0.3352
c 4 Influence of the choline cation 0.7976
c 5 Influence of the amine cation 0.0141
 
Cation Subst. s 1 Number of carbons in the long chains of the AIL molecule 0.06824
s 2 Number of substitutions in the imidazolium cation of the AIL molecule 0.3469
s 3 Number of hydroxyls in the long chains of the AIL molecule 0.0275
s 4 Number of ethyl or methyl groups in the long chains of the AIL molecule 0.2697
s 5 Number of ethyl or NO2 groups in the long chains of the AIL molecule 0.4661
Open in a new tab

D. Conclusions

As previously discussed, the thermodynamic characterization in this paper, in terms of the density and ultrasonic velocity of 13 recently synthesized protic ionic liquids as a function of temperature, provides useful information for industrial process design and development of new prediction models.

The studied PILs’ cytotoxicity evaluated using MTT assay revealed higher IC50 values (lower toxicological profile) when compared to imidazolium derived ionic liquids assessed for similar cell types.30 These results should guide their use as solvents or additives for products and processes safe for human use, in agreement with the predictions based on their designed chemical structure.

An important fact that deserves be highlighted is that all the studied PILs showed IC50 values ranging between 1.25 and 5.0 wt% in solution. In 2008, Jaitely et al.34 assessed ionic liquids (hexafluorophosphate salts) for pharmaceutical use, and found almost no effect in terms of toxicity to Caco-2 cells at the maximum concentration of 1%, and they were considered largely non-toxic. Depending on concentration and structural features, the toxicity profile of ILs may vary, but by rational design, the toxicity can be mitigated knowing a structure–activity relationship. The model based on group contribution (QSAR) used in this work allowed the prediction of protic ionic liquids’ cytotoxicity, with relative accuracy of experimental and then predicted toxicity data. In view of the need to perform a toxicity assessment to confirm fully green PIL behaviour, the QSAR method could be the key towards providing the predictive ability which could guide the design of novel greener ILs for industrial application.

Conflicts of interest

There are no conflicts to declare.

Supplementary Material

Click here for additional data file. (908KB, pdf)

Acknowledgments

The authors would like to acknowledge FAPESB - Fundação de Amparo à Pesquisa do Estado da Bahia (scientific grant of G. B. Romão and Programa Ação Referência – PET0071/2013) and the National Council for Scientific and Technological Development – CNPq (Chamada Universal MCTI/CNPq 28/2018) for its support in developing this research.

Footnotes

†Electronic supplementary information (ESI) available: Densities (ρ) and ultrasonic velocities (u) for the studied protic ionic liquids at 293.15–323.15 K. See DOI: 10.1039/c8tx00338f

References

  1. Welton T. Chem. Rev. 1999;99:2017. doi: 10.1021/cr980032t. [DOI] [PubMed] [Google Scholar]
  2. Plechkova N. V., Seddon K. R., Feng R., Zhao D., Guo Y., Andreani L., Rocha J. D., Hajipour A. R., Rafiee F., Jenkin G. R. T., Al-Bassam A. Z. M., Harris R. C., Abbott A. P., Smith D. J., Holwell D. A., Chapman R. J., Stanley C. J., Zhu Y., Hosmane N. S., Sasikumar B., Arthanareeswaran G., Ismail A. F., Ullah Z., Khan A. S., Muhammad N., Ullah R., Alqahtani A. S., Shah S. N., Ghanem O. B., Bustam M. A., Man Z. Chem. Soc. Rev. J. Environ. Prot. Braz. J. Chem. Eng. Org. Prep. Proced. Int. Miner. Eng. Eur. J. Inorg. Chem. J. Mol. Liq. J. Mol. Liq. 2008;2010;2012;2015;2016;2017;2018;2018;37129478738266266:123. 95, 1, 249, 18, 4369, 330, 673. [Google Scholar]
  3. Greaves T. L., Drummond C. J., Hallett J. P., Welton T., Khashavar G., Greaves T. L., Drummond C. J., Egorova K. S., Ananikov V. P. Chem. Rev. Chem. Rev. Green Sustainable Chem. Chem. Rev. J. Mol. Liq. 2008;2011;2014;2015;2018;1081114115272:206. 3508, 44, 11379, 271. [Google Scholar]
  4. Gathergood N. N., Scammells P. J., Garcia M. T., Zhao D., Liao Y., Zhang Z. D., Zhu S., Chen R., Wu Y., Chen Q., Zhang X., Peric B., Martí E., Sierra J., Iglesias M., Garau M. A., Jordan A., Gathergood N., Deng Y., Beadham I., Ghavre M., Costa Gomes M. F., Gathergood N., Husson P., Légeret B., Quilty B., Sancelme M., Besse-Hoggan P., Romero A., Santos A., Tojo J., Rodríguez A., Jordan A., Haiß A., Spulak M., Karpichev Y., Kümmerer K., Gathergood N. Green Chem. Clean: Soil, Air, Water. Chem. Biochem. Eng. Q. Environ. Toxicol. Chem. Chem. Soc. Rev. Green Chem. J. Hazard. Mater. Green Chem. 2006;2007;2009;2011;2015;2015;2008;2016;8352330441715118:156. 42, 207, 2802, 8200, 1479, 268, 4374. [Google Scholar]
  5. Petkovic M., Seddon K. R., Rebelo L. P. N., Pereira C. S., Egorova K. S., Ananikov V. P., Amde M., Liu J.-F., Pang L. Chem. Soc. Rev. ChemSusChem. Environ. Sci. Technol. 2011;2014;2015;40749:1383. 336, 12611. doi: 10.1039/c004968a. [DOI] [PubMed] [Google Scholar]
  6. Docherty K. M., Hebbeler S. Z., Kulpa Jr. C. F., Kumar M., Trivedi N., Reddy C. R. K., Jha B., Guo Y., Liu T., Zhang J., Wang J., Wang J., Zhu L., Yang J., Thamke V. R., Tapase S. R., Kodam K. M. Green Chem. Chem. Res. Toxicol. Environ. Toxicol. Chem. Water Res. 2006;2011;2013;2017;82435125:560. 1882, 411, 237. [Google Scholar]
  7. Peric B., Martí E., Sierra J., Iglesias M., Garau M. A., Peric B., Sierra J., Martí E., Bottin-Weber U., Stolte S., Peric B., Siera J., Martí E., Cruañas R., Garau M. A., Peric B., Sierra J., Martí E., Cruañas R., Garau M. A., Tzani A., Elmaloglou M., Kyriazis C., Aravopoulou D., Kleidas I., Papadopoulos A., Ioannou E., Kyritsis A., Voutsas E., Detsi A., Sun J., Konda N. V. S. N. M., Parthasarathi R., Dutta T., Valiev M., Xu F., Simmons B. A., Singh S., Khan A. S., Nasrullah A., Ullah Z., Bhat A. H., Ghanem O. B., Muhammad N., Rashid M. U., Man Z., Reid J. E. S. J., Prydderch H., Spulak M., Shimizu S., Walker A. J., Gathergood N. Environ. Toxicol. Chem. J. Hazard. Mater. Chemosphere. Ecotoxicol. Environ. Saf. J. Mol. Liq. Green Chem. J. Mol. Liq. Sustainable Chem. Pharm. 2011;2013;2014;2015;2016;2017;2018;2018;30261108115224192497:2802. 99, 418, 257, 366, 3152, 583, 17. doi: 10.1002/etc.683. [DOI] [PubMed] [Google Scholar]
  8. Andrade R. S., Torres D., Ribeiro F. R., Chiari-Andréo B. G., Oshiro Junior J. A., Iglesias M. ACS Sustainable Chem. Eng. 2017;5:8756. [Google Scholar]
  9. (a) Iglesias M., Torres A., Gonzalez-Olmos R., Salvatierra D. J. Chem. Thermodyn. 2008;40:119. [Google Scholar]; (b) Ghatee M. H., Bahrami M., Khanjari N., Firouzabadi H., Ahmadi Y. J. Chem. Eng. Data. 2012;57:2095. [Google Scholar]; (c) Bharmoria P., Trivedi T. J., Malek N. I., Kumar A. Indian J. Chem. 2016;55A:544. [Google Scholar]; (d) Andrade R., Carreras A., Iglesias M. J. Serb. Chem. Soc. 2017;82:1155. [Google Scholar]; (e) Nazet A., Buchner R. J. Chem. Phys. 2018;148:193836. doi: 10.1063/1.5013097. [DOI] [PubMed] [Google Scholar]; (f) Hosseini S. M., Aparicio S., Alavianmehr M. M., Khalifer R. J. Mol. Liq. 2018;266:751. [Google Scholar]; (g) Alvarez V., Mattedi S., Martin-Pastor M., Aznar M., Iglesias M. J. Chem. Thermodyn. 2011;43:997. [Google Scholar]; (h) Hou M., Xu Y., Han Y., Chen B., Zhang W., Ye Q., Sun J. J. Mol. Liq. 2013;178:149. [Google Scholar]; (i) Pentilla A., Uusi-Kyyny P., Salminen A., Seppala J., Alopaeus V. Int. J. Greenhouse Gas Control. 2014;22:313. [Google Scholar]; (j) Santos D., Lourenço A., Santos M. F. C., Franceschi E., Dariva C., Barison A., Zuber A., Castier M., Mattedi S. J. Mol. Liq. 2016;222:867. [Google Scholar]; (k) Sarabando J., Magano P. J. M., Ferreira A. G. M., Santos J. B., Carvalho P. J., Mattedi S., Fonseca I. M. A. J. Chem. Thermodyn. 2018;122:183. [Google Scholar]; (l) Barros S., Andrade R. S., Iglesias M. Int. J. Thermodyn. 2018;21:70. [Google Scholar]; (m) Kulhavy J., Andrade R. S., Barros S., Serra J., Iglesias M. J. Mol. Liq. 2016;213:92. [Google Scholar]
  10. Babich H., Borenfreund E., Castaño A., Cantarino M. J., Castillo P., Tarazona J. V., Lachance B., Robidoux P. Y., Hawari J., Ampleman G., Thiboutot S., Sunahara G. I., Murtoniemi T., Penttinen P., Nevalainen A., Hirvonen M.-R., Hamers T., Kamstra J. H., Sonneveld E., Murk A. J., Kester M. H. A., Andersson P. L., Legler J., Brouwer A., Bell S. M., Chang X., Wambaugh J. F., Allen D. G., Bartels M., Brouwer K. L. R., Casey W. M., Choksi N., Ferguson S. S., Fraczkiewicz G., Jarabek A. M., Ke A., Lumen A., Lynn S. G., Paini A., Price P. S., Ring C., Simon T. W., Sipes N. S., Sprankle C. S., Strickland J., Troutman J., Wetmore B. A., Kleinstreuer N. C. Toxicol. in Vitro. Chemosphere. Mutat. Res., Genet. Toxicol. Environ. Mutagen. Inhalation Toxicol. Toxicol. Sci. Toxicol. in Vitro. 1991;1996;1999;2005;2006;2018;532444179247:91. 2141, 25, 681, 157, 213. [Google Scholar]
  11. Zhang Y., Wu L., Tashiro S., Onodera S., Ikejima T. Acta Pharm. 2004;25:83. [PubMed] [Google Scholar]
  12. Horio M., Ito A., Matsuoka Y., Moriyama T., Orita Y., Takenaka M., Imai E. Nephrol., Dial., Transplant. 2001;16:483. doi: 10.1093/ndt/16.3.483. [DOI] [PubMed] [Google Scholar]
  13. Jodynis-Liebert J., Nowicki M., Murias M., Adamska T., Ewertowska M., Kujawska M., Piotrowska H., Konwerska A., Ostalska-Nowicka D., Pernak J., Zhang Z., Liu J. F., Cai X. Q., Jiang W. W., Luo W. R., Jiang G. B. X., Li X., Ma J., Wang J. M., Messali M., Almtiri M. N., Abderrahman B., Salghi R., Aouad M. R., Alshahateet S. F., Ali A. A.-S., Ferraz R., Costa-Rodrigues J., Fernandes M., Santos M. M., Marrucho I. M., Rebelo L. P., Prudêncio C., Noronha J. P., Petrovski Z., Branco L. C., Liu X., Ji C., Yang Q., Bao Z., Fan X., Yang Y., Xing H. Regul. Toxicol. Pharmacol. Environ. Sci. Technol. Ecotoxicol. Environ. Saf. S. Afr. J. Chem. ChemMedChem. ACS Sustainable Chem. Eng. 2010;2011;2015;2015;2015;2017;574512068105:266. 1688, 342, 219, 1480, 1974. doi: 10.1016/j.yrtph.2010.03.006. [DOI] [PubMed] [Google Scholar]
  14. Luis P., Garea A., Irabien A. J. Mol. Liq. 2010;152:28. [Google Scholar]
  15. Botham P. A., Earl L. K., Fentem J. H., Roguet R., Van De Sandt J. J. M., Livny O., Kaplan I., Reifen R., Polak-Charcon S., Madar Z., Schwartz B., Abu-Yousif A. O., Smith K. A., Getsios S., Green K. J., Van Dross R. T., Pelling J. C., Rehberger K., Kropf C., Segner H. ATLA, Altern. Lab. Anim. Nutr. Cancer. Cancer Res. Environ. Sci. Eur. 1998;2003;2008;2018;26476830:195. 195, 3057, 23. [PubMed] [Google Scholar]
  16. Wilkening S., Stahl F., Bader A., Gerets H. H. J., Tilmant K., Gerin B., Chanteux H., Depelchin B. O., Dhalluin S., Atienzar F. A. Drug Metab. Dispos. Cell Biol. Toxicol. 2003;2012;3128:1035. 69. [Google Scholar]
  17. Dias A. M. A., Marceneiro S., Braga M. E. M., Coelho J. F. J., Ferreira A. G. M., Simões P. N., Veiga H. I. M., Tomé L. C., Marrucho I. M., Esperança J. M. S. S., Matias A. A., Duarte C. M. M., Rebelo L. P. N., De Sousa H. C., Xue J.-J., Chen Q.-Y., Nagaraj K., Arunachalam S., Li T. H., Jing C. Q., Gao K. L., Yue W. Y., Lee S. F., Wan R., Xia X., Wang P., Huo W., Dong H., Chang Z. Acta Biomater. Spectrochim. Acta, Part A. New J. Chem. GMR, Genet. Mol. Res. Toxicol. in Vitro. 2012;2014;2014;2015;2018;8120381452:1366. 161, 366, 13342, 1. doi: 10.1016/j.actbio.2011.10.034. [DOI] [PubMed] [Google Scholar]
  18. Stepnowski P., Składanowski A. C., Ludwiczak A., Łaczyńska E. Hum. Exp. Toxicol. 2004;23:513. doi: 10.1191/0960327104ht480oa. [DOI] [PubMed] [Google Scholar]
  19. Frade R. F. M., Simeonov S., Rosatella A. A., Siopa F., Afonso C. A. M. Chemosphere. 2013;92:100. doi: 10.1016/j.chemosphere.2013.02.047. [DOI] [PubMed] [Google Scholar]
  20. Kumar R. A., Papaïconomou N., Lee J. M., Salminen J., Clark D. S., Prausnitz J. M. Environ. Toxicol. 2009;24:388. doi: 10.1002/tox.20443. [DOI] [PubMed] [Google Scholar]
  21. Samori C., Malferrari D., Valbonesi P., Montecavalli A., Moretti F., Galleti P., Sartor G., Tagliavini E., Fabbri E., Pasteris A. Ecotoxicol. Environ. Saf. 2010;73:1456. doi: 10.1016/j.ecoenv.2010.07.020. [DOI] [PubMed] [Google Scholar]
  22. Abu-Yousif A. O., Smith K. A., Getsios S., Green K. J., Van Dross R. T., Pelling J. C., Heise R., Mey J., Neis M. M., Marquardt Y., Joussen S., Ott H., Wiederholt T., Kurschat P., Megahed M., Bickers D. R., Merk H. F., Baron J. M., Bedoni M., Sforza C., Dolci C., Donetti E., Leme D. M., Sehr A., Grummt T., Gonçalves J. P., Jacomasso T., Winnischofer S. M. B., Potrich F. B., Oliveira C. C. D., Trindade E. D. S., de Oliveira D. P., Hwang J.-H., Park H., Choi D. W., Nam K. T., Lim K.-M., Santos de Almeida T., Júlio A., Saraiva N., Fernandes A. S., Araújo M. E. M., Baby A. R., Rosado C., Mota J. P. Cancer Res. J. Invest. Dermatol. J. Dermatol. Sci. J. Toxicol. Environ. Health, Part A. Toxicol. in Vitro. Drug Dev. Ind. Pharm. 2008;2006;2007;2018;2018;2017;6812646814643:3057. 2473, 139, 589, 194, 1858. doi: 10.1158/0008-5472.CAN-07-2763. [DOI] [PubMed] [Google Scholar]
  23. Elmore S. Toxicol. Pathol. 2007;35:495. doi: 10.1080/01926230701320337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kataoka S., Tsuruo T., Antonio J. R., Juan F. A., Ewan T. M., Oliver K., Lorenzo G., Marta L., Bin Z., Lu L., Kai C., George V. C., Kumar D. N., Suresh P., Kumar R. A. Oncologist. Parasites Vectors. Nat. Rev. Drug Discovery. Asian Pac. J. Cancer Prev. Asian Pac. J. Cancer Prev. 1996;2010;2011;2011;2012;13101213:399. 1, 221, 2675, 2015. [Google Scholar]
  25. Kuang Y. F., Chen Y.-H., Oliver L., Hue E., Rossignol J., Bougras G., Hulin P., Naveilhan P., Heymann D., Lescaudron L., Vallette F. M. Food Chem. Toxicol. PLoS One. 2004;2011;4212:1711. 19820. [Google Scholar]
  26. (a) Kultz D., Chakravarty D. Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol. 2001;130:421. doi: 10.1016/s1095-6433(01)00440-8. [DOI] [PubMed] [Google Scholar]; (b) Huang Z., Tunnacliffe A. J. Physiol. 2004;558:181. doi: 10.1113/jphysiol.2004.065540. [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Harms C. T., Oertli J. J. J. Plant Physiol. 1985;120:29. [Google Scholar]; (d) Munns R. Plant, Cell Environ. 2002;25:239. doi: 10.1046/j.0016-8025.2001.00808.x. [DOI] [PubMed] [Google Scholar]; (e) Schwarz H., Koch A. L. Microbiology. 1995;141:161. doi: 10.1099/13500872-141-12-3161. [DOI] [PubMed] [Google Scholar]; (f) Friis M. B., Friborg C. R., Schneider L., Nielsen N. M-B., Lambert I. H., Christensen S. T., Hoffmann E. K. J. Physiol. 2005;567:427. doi: 10.1113/jphysiol.2005.087130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Akeson M. A., Munns D. N. Biochim. Biophys. Acta, Biomembr. 1989;984:200. doi: 10.1016/0005-2736(89)90217-4. [DOI] [PubMed] [Google Scholar]
  28. Palmer M., Chan A., Dieckmann T. and Honek J., Notes to biochemical pharmacology, John Wiley & Sons, 2013. [Google Scholar]
  29. Stolte S., Arning J., Bottin-Weber U., Muller A., Pitner W. R., Welz-Biermann U., Jastorff B., Ranke J. Green Chem. 2007;9:760. [Google Scholar]
  30. Wang X., Ohlin C. A., Lu Q., Fei Z., Hub J., Dysonc P. J. Green Chem. 2007;9:1191. [Google Scholar]
  31. European Union. 2006. Regulation (EC) No. 1907/2006 of the European Parliament and of the Council concerning the Registration, Evaluation, Authorization and Restriction of Chemicals (REACH). Off J Eur Union 396:1–856
  32. Amberg A., Drug Discovery and Evaluation: Safety and Pharmacokinetic Assays, Springer, Berlin Heidelberg, 2013, p. 1273. [Google Scholar]
  33. Das R. N., Roy K. Mol. Diversity. 2013;17:151. doi: 10.1007/s11030-012-9413-y. [DOI] [PubMed] [Google Scholar]
  34. Jaitely V., Karatas A., Florence A. T. Int. J. Pharm. 2008;54:168. doi: 10.1016/j.ijpharm.2008.01.034. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file. (908KB, pdf)

Articles from Toxicology Research are provided here courtesy of Oxford University Press

RESOURCES