Rational design of imidazolium based salts as

Journal of Molecular Liquids 255 (2018) 578–588

Contents lists available at ScienceDirect

Journal of Molecular Liquids journal homepage: www.elsevier.com/locate/molliq

Rational design of imidazolium based salts as anthelmintic agents Prabodh Ranjan a, Mohd Athar a, H. Rather c, Kari Vijayakrishna d, R. Vasita c, Prakash C. Jha b,⁎ a

CCG@CUG, School of Chemical Sciences, Central University of Gujarat, Sector-30, Gandhinagar 382030, Gujarat, India CCG@CUG, Centre for Applied Chemistry, Central University of Gujarat, Sector-30, Gandhinagar 382030, Gujarat, India c School of Life Sciences, Central University of Gujarat, Sector-30, Gandhinagar 382030, Gujarat, India d Department of Chemistry, School of Advanced Sciences, VIT University, Vellore 632014, Tamil Nadu, India b

a r t i c l e

i n f o

Article history: Received 17 November 2017 Received in revised form 21 January 2018 Accepted 1 February 2018 Available online 02 February 2018 Keywords: Ionic liquids (ILs) 1-Butylimidazole Vermicidal Pheretima posthuma Albendazole sulfoxide

a b s t r a c t In present study, we have applied a combined experimental and computational approach to explicate several bred-in-the-bone notions for the rational designing of imidazolium derive salts as anthelmintic agents. Thirty derivatives of 1-methyl-3-alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium differing in the length of N-alkyl chains and various counter anions (Br−, OH− and BF− 4 ) were synthesized and characterized. These ILs were assessed for their vermicidal activity (VA) and cell viability against the Pheretima posthuma and 3T3-40 L1 cell respectively. Subsequently, effects of different N-alkyl side-chain and methyl group at C-2 position of 1,2-dimethyl3-alkylimidazolium derivatives on VA was also reported. The current findings suggest that VA depends on dose, nature of N-alkyl side-chain length, anionic moieties and hydrophobic factors. In most cases, ILs bearing hydroxide counter anion showed better VA than their Br− and BF− 4 counterparts. The experimental findings were complemented by QSAR modeling and molecular docking study to elucidate the contributing factors responsible for their activities and binding patterns in ligand-receptor complexes. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Helminthic infections are neglected tropical diseases (NTDs) caused by helmintic worms (pin worms, hook worms, round worms and tape worms) that lead to pneumonia, eosinophilia, anemia, ascariasis, cysticercosis, filaria, malnutrition etc. [1]. These infections have plagued humans from past few decades and affect wide continuum of population in developing countries due to poor unhygienic conditions [1]. In attempt to combat such infections, different medicinal preparations such as plant extracts and synthetic heterocyclic compounds have been tested [2–4]. As an outcome of such appraisals, benzimidazole carbamates (BZCs), levomesol, antimonisol and paraziquintel etc. were clinically developed for their medicinal use. However, their excessive use has led to resistance problems and thereby posing a severe challenge for the chemists. Therefore, it is need of the hour to develop novel drug with improved efficacy, pharmacokinetics and pharmacodynamics factors than existing anthelmintic drugs [5–27]. Among the clinically tested therapeutics, benzimidazole derivatives showed wide-array of activities against helminths. Efforts have been Abbreviations: ILs, ionic liquids; BZCs, benzimidazole carbamates; APIs, Active Pharmaceutical Ingredients; SIFt, Structural Interaction Fingerprint; GFA, Genetic Function Approximation; QSAR, quantitative structure activity relationship; VAs/VA, vermicidal activities/vermicidal activity. ⁎ Corresponding author at: Centre for Applied Chemistry, Central University of Gujarat, Sector-30, Gandhinagar 382030, India. E-mail addresses: [email protected], [email protected] (P.C. Jha).

https://doi.org/10.1016/j.molliq.2018.02.001 0167-7322/© 2018 Elsevier B.V. All rights reserved.

made to explore the role of different functionalities (1-alkyl, aryl, arylalkyl and acyl benzimidazoles) and hydrogen atom at N-1/N-3 and C-2 position of the benzimidazole ring (Fig. S2) [2]. Structure-activityrelationship (SAR) studies of such compounds indicates that there are certain types of functional group which improves anthelmintic activities (i.e., methoxy carbonyl amino and heteroaryl group attached at C-2 position of benzimidazole derivatives through a CO, CHON, CONH, S, SO) [11]. Notably, the introduction of alkyl or aryl group at C-2 position of the benzimidazole shows weak anthelmintic activity [10–27]. On the contrary, there are few exceptions to this generalizations, such as thiabendazole, cambendazole, triclabendazole and few benzothiazoles demonstrates the high order of anthelmintic activity, even in the absence of a C-2 position hydrogen atom (Fig. S3) [27,28]. Owing to the fact that major class of drug administrated in the form of salts [28,29]; low toxic ionic liquids (ILs) can be potential anthelmintic agents due to their ease of adjustable physicochemical and biological profile (Fig. S4). For this reason, ILs have received attention for their use as Active Pharmaceutical Ingredients (IL-API) [30–52]. In addition, efforts are being made to fine tune their toxicity (by varying their counter anions, alkyl chain length and functional group) in order to afford them as APIs for biomedical applications [33,34]. The maxim “the poison is in the dose”, has encouraged the chemists to develop ILs as a useful drug for biomedical applications [31,32]. Previously, we have assessed the role of increasing alkyl chain length (hydrophobic factors) and different counter anions (Br− and OH−) of 1-methyl-3-alkylimidazolium derive ionic liquids for vermicidal activity (VA) and supported by the in-silico

P. Ranjan et al. / Journal of Molecular Liquids 255 (2018) 578–588

4

3

5

N

6

N 2 H 1

H 7

Benzimidazole

R2 N R1 N R3

X-

Imidazolium Salts

Scheme 1. Structural comparison of benzimidazole derived Antivermicidal drugs with imidazolium derived slats screened for their vermicidal activities to assess the role of hydrogen atom.

approach. In the present study, we have screened the 1-methy-3alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium derived ionic liquids for their VA at different doses in a time dependent manner. The compounds were also evaluated for their cellular toxicity to rationalize their prospective use as anthelmintic. Furthermore, this study has been supported by the molecular docking and quantitative-structure-activity-relationship (QSAR) studies to illustrate the binding mechanism, role of hydrogen atom present at C-2 position and contributing factors in 1-methylimidaolium ring by substituting with a methyl group (Scheme 1).

579

with hydroxide anions. A demonstrative example for the synthesis of [RMIM]OH (i.e., [EMIM]OH, [BMIM]OH, [HMIM]OH, [OMIM]OH and [DMIM]OH) and [RMMIM]OH (i.e., [EMMIM]OH, [BMMIM]OH, [HMMIM]OH, [OMMIMOH and [DMMIM]OH) derivatives are as follows: 5 g of [RMIM]Br or [RMMIM]Br IL was dissolved in methanol and passed slowly through a column packed with Amberlite® IRA-400 hydroxide resin. The solvents were rotary evaporated from the collected solution and then vacuum dried which afforded the 1-methyl-3alkylimidazolium hydroxide and 1,2-dimethyl-3-alkylimidazolium hydroxide as a viscous liquid in quantitative yields (Scheme 2, Step-2). Metathesis reaction was performed as per a literature report with minor modifications (Scheme 2, Step-3) to achieve ILs with boron tetrafluoride as their counter anion (IL- BF4). An illustrative example for the synthesis of [RMIM]BF4 (i.e., [EMIM]BF4, [BMIM]BF4, [HMIM]BF4, [OMIM]BF4, [DMIM]BF4) and [RMMIM]BF4 (i.e.,[EMMIM]BF4, [BMMIM] BF4, [HMMIM]BF4, [OMMIM]BF4 and [DMMIM]BF4) derivatives is as follows: 1 equivalent of 1-methyl-3-alkylimidazolium bromide ([RMIM] Br) or 1,2-methyl-3-alkylimidazolium bromide ([RMMIM]Br) and 2 equivalent of NaBF4 was added in to the round bottom flask containing 25 mL of methanol and stirred at room temperature till completion of the reaction. The reaction progress was monitored by TLC. After reaction completion, the reaction mixture was filtered off using Whatman filter paper to remove the salt. Further, the salt was isolated by centrifuging the filtrate for 10 min at 3000 rpm. The solvent from filtrate was evaporated under reduced pressure and characterized using 1H and 13C NMR and mass spectroscopy (Scheme 2, Step-3). The spectral details of the synthesized ILs have been mentioned in the Supplementary information Figs. S5 to S34.

2. Materials and methods 1-Methylimidazole, 1,2-dimethylimidazole, ethyl bromide, butyl bromide, hexyl bromide, octyl bromide, and decyl bromide, NaBF4, NaOH and organic solvents were purchased from Avara Synthesis Pvt. Ltd. and R & D Chemicals, India, and used as received. Albendazole sulfoxide was purchased from Hi-Media, India and used as received. 1H and 13 C NMR spectra were recorded at room temperature (rt) in 5 mm tube using Bruker 400 MHz and 500 MHz spectrometer. The mass spectral analysis was accomplished with Agilent Technologies G6520B LC–MS (QTOF) mass spectrometry with +ESI ionization method. 2.1. General synthetic procedure for the ionic liquids (ILs) ILs with bromide counter anion (IL-Br) and varying alkyl chains, i.e., n-C2H5, n-C4H9, n-C6H13, n-C8H17 and n-C10H21 were synthesized according to the reported procedure [53]. The quaternization was carried out separately in a sealed flask after adding 12 mmol of 1methylimidazole or 1,2-dimethylimidzole (Scheme 1) and 14 mmol of N-alkyl bromide, heated in an oil bath maintained at 140 °C for 20 min and cooled to room temperature. Further, the reaction mixture was heated at the same temperature for 10 min and then cooled down to room temperature. The obtained IL was washed with ethyl acetate (3 × 25 mL) to remove the unreactant (unreacted starting materials), vacuum dried and characterized using NMR (1H and 13C) and mass spectroscopy. Similarly, 1-methyl-3-alkylimidazolium bromide ([RMIM]Br): [EMIM]Br, [BMIM]Br, [HMIM]Br, [OMIM]Br and [DMIM] Br) and 1,2-dimethyl-3-alkylimidazolium bromide ([RMMIM]Br): [EMMIM]Br, [BMMIM]Br, [HMMIM]Br, [OMMIM]Br and [DMMIM]Br) were synthesized (Scheme 2, Step-1) and characterized. Further, ILs with hydroxide counter anion (IL-OH) and varying alkyl chains (i.e., nC2H5, n-C4H9, n-C6H13, n-C8H17 and n-C10H21) were synthesized according to the reported procedure [54,55]. Primarily, 3 g of the Amberlite® IRA-400 chloride resin was put in a round bottomed flask containing 100 mL of 1 M NaOH and stirred for 12 h for anion exchange. Subsequently, the resin was neutralized by washing several times with double distilled water. Later, it was filtered and dried, which resulted in a resin

2.2. Evaluation of vermicidal activity (VA) ILs viz. 1-methyl-3-alkylimidazolium bromide/or hydroxide/or boron tetrafluoride ([RMIM] Br/or OH/or BF4) and 1,2-dimethyl-3alkylimidazolium bromide/or hydroxide/or boron tetrafluoride ([RMMIM] Br/or OH/or BF4) were screened for their vermicidal activity. Indian earthworms (P. posthuma) were considered for this assay due to their physiological and anatomical resemblance with intestinal parasitic worms and round worms. They were collected from wastewater treatment, management Plant of Golden Temple, Vellore, India. Albendazole sulfoxide and sterilized distilled water were used as a positive and negative control, respectively, against the test organisms. The assay was performed as per reported procedure [55,56] utilizing four different concentrations, i.e., 2 mM, 4 mM, 8 mM and 16 mM of each IL in time dependent manner (15, 30, 45, 60, 75, 90, 120 min) (refer Figs. S35 to S49). Their activities were recorded in triplicates as function of percentage paralysis and percentage mortality.

2.3. Cell culture study of ionic liquids (ILs) The alamar blue assay was performed to evaluate the 3T3-L1 (fibroblast) cells test viability against the ILs i.e., [DMIM]Br, [DMIM]OH, [DMIM]BF4, [DDMIM]Br, [DDMIM]OH and [DDMIM]BF4. The 3T3-L1 cells were grown in Roswell Park Memorial Institute (RPMI) media supplemented with 10% Fetal Bovine Serum (FBS). At around 70–80% confluency, cells were trypsinized and seeded in 96 well plates as 10,000 cells per well. After 24 h, 4 mM dose of each IL in the media was treated for another 24 h. As the stock of compounds were prepared in water. Hence, an equal volume of water was added to media and incubated in control wells. Once more, after 24 h, the media were replaced with fresh media containing 10% v/v of the alamar blue solution and incubated for 1 h in the dark. After that, absorbance was recorded at 570 nm to evaluate the cell viability using multi-plate reader (Synergy™ H1, BioTek®) [57]. All the experiments were performed in triplicates.

580


R2

Amberlite Resin

N

Y R1

RT Step-2

N CH3

R2

Step-1 N R1

X

N

Quaternization

Y = OH-

Metathesis R1

N

R-X Heat, 140°C

CH3

R2

N CH3 NaBF4

X = Br-

[EMIM] Br [BMIM] Br [HMIM] Br [OMIM] Br [DMIM] Br

N CH3

Z = BF4-

n-C6H13 ; n-C8H17 ; n-C10H21

[EMMIM] Br [BMMIM] Br [HMMIM] Br [OMMIM] Br [DMMIM] Br

X = Br-

Z R1

Methanol, RT

MIM Where, R1 = H R1 = CH3 MMIM R2 = n-C2H5; n-C4H9 ;

N

Step-3

[EMIM] OH [BMIM] OH [HMIM] OH [OMIM] OH [DMIM] OH

[EMMIM] OH [BMMIM] OH [HMMIM] OH [OMMIM] OH [DMMIM] OH

Y = OH-

[EMIM] BF4 [BMIM] BF4 [HMIM] BF4 [OMIM] BF4 [DMIM] BF4

[EMMIM] BF4 [BMMIM] BF4 [HMMIM] BF4 [OMMIM] BF4 [DMMIM] BF4

Z = BF4-

Scheme 2. General synthetic procedure for ionic liquids (ILs).

2.4. Computational details Geometry optimization of 1-methyl-3-alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium bromide/or hydroxide/or boron tetrafluoride derivatives were performed at B3LYP/6-311G(d) level of approximation with Conductor-like Polarizable Continuum Solvation Model (CPCM) using Gaussian 09 program [58]. Altered guess geometry of ILs was given by manual positioning of anions with respect to the cationic core to approximate the real geometry with minimal energy. Frequency analysis was also performed at the same level of theory for each IL in order to assure the structure is at global minima. The details of the optimized guess geometries have been given in electronic supplementary information Fig. S50.

2.4.1. QSAR model development Thirty ionic liquids with various alkyl chain n-C2H5, n-C4H9, n-C6H13, − − n-C8H17, and n-C10H21 and varying counter anions (BF− 4 , Br , and OH ) were considered for the QSAR model development. The critical electronic and solvation effects were deliberated during the generation of independent variables for mapping the anthelmintic activity as the dependent variable. Indeed, computationally; it is tedious to mimic the structure of IL due to their characteristic interaction between the [C]+X− (C+ = cations, and X− = anions) [59,60]. Therefore, as proposed in literatures, cations ([EMIM]+, [BMIM]+, [HMIM]+, [OMIM]+, and [DMIM]+, [EMMIM]+, [BMMIM]+, [HMMIM]+, [OMMIM]+, and − − [DMMIM]+) and anions (BF− 4 , Br , and OH ) were optimized separately in solvation phase using CPCM model without any symmetry constraint [60]. The average percentage mortality at 4 mM conc. were converted to −log10[%mortality] and used as the dependent variable. Further, calculated DFT based descriptors of cationic head and anionic counterparts were rendered as an independent variable (Table S2) [58,61]. Thereafter, the data set of thirty ILs was divided into training set and test set of 23 and 7 compounds respectively. Genetic functional approximation (GFA) was used for the QSAR modeling [62,63] that

involves the multivariate adaptive regression algorithm combining with the genetic algorithm (GA) to evolve the population of equations (each containing only a subset of variables) that best fit the training set data. With this method, a series of potential solutions to a problem (the population of organisms) are derived and tested repeatedly until an approximate optimal solution is found. The ability to prevent overfitting of the GFA model is critical to the successful construction of a statistically significant QSAR model. In our experiments, Friedman lack-of-fit (LOF) was used for the selection of GFA-derived equations, while, the correlation coefficient r2, r2adjusted, and leave one-out cross-validation (r2cv) were taken as objective function to select the model predictability. The LOF [62,64] examines over-fitting, which is a problem often encountered in constructing statistical models. LOF ¼

SSE i2 M 1−λ c þ dp M h

ð1Þ

whereas, SSE is the sum of squares of errors, c is the number of basic functions in the model, the constant term d is the smoothing parameter, p is the total number of descriptors contained in all model ignoring the constant term, M is the number of data points in the training set, and λ is a safety factor, with the value of 0.99, to ensure that the denominator of the expression never become zero. 2.4.2. Molecular docking of cationic head group of ILs The anthelmintic potentials of the ionic liquids were mechanistically studied using the molecular docking which enable to comprehend the receptor-binding interactions at the bio-molecular level. For this, the DFT optimized ligand structures of the cationic head group [RMIM]+ and [RMMIM]+ derivatives and 3D homology receptor model of H. contortus β-tubulin [65] (PDB entry: 1OJ0) [66] were used. Flexible docking protocol of CDOCKER (Discovery Studio v4.0) followed by CHARMm force field was considered to recognize Structural Interaction


Fingerprint (SIFt) patterns [67,68] of β-tubulin inhibitors as inferred by Massarotti et al., [69]. Prior docking, protein was prepared and the binding site was defined by selecting the grid box of size 8 Å. CDOCKER algorithm was used for docking the low energy conformations in flexible binding pockets [70,71]. Conformation generation and flexibility of the binding pocket was modeled by the ChiFlex and ChiRotor algorithm respectively [70]. Thereafter, Simulated Annealing (SA) method was applied to minimize the poses by using the following elements i.e., heating temperature (500 K), cooling temperature (300 K), grid extension (8.0 Å), conformer generation method (BEST) and maximum conformation (10). Afterwards, SIFt binding pattern of β-tubulin inhibitors and distribution of colchicine binding domain for IL was also recognized [67,68,72]. The interactions present in the dock poses was represented in 2D format using ligand-receptor interaction map. Further, we attempted to identify all three different zones present in colchicine binding domain of β-tubulin inhibitors [67,68]. The current in-silico method is the advanced technique for computational illustrations of ILs as it answers the factors accountable for the VA [66]. 3. Results and discussion Benzimidazole derivatives are the most widely administered anthelmintic drugs contains an imidazole ring substituted with different functionalities [5–8]. It has been found that hydrogen atom present at C-2 position of BZCs derivatives plays an important role in anthelmintic activities with few exceptions [10,23,27,73]. Also, it is evident that majority of the drugs are being administrated in the form of salts. Therefore, ILs can be worth to explore as promising anthelmintic candidate due to its tunable physicochemical and biological properties. Recent advancement has been made to develop them as Active Pharmaceutical Ingredients (APIs) [35]. Therefore, in this study 1-methyl-3-alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium derived ionic liquids have been employed (Scheme 2) for VA as per reported method to check the role of C-2 positioned hydrogen atom by substituting with methyl group and the role of different anions by fixing the cationic head group. The VAs of each derivatives have been discussed below (for spectral details of ILs refer electronic Supplementary information Figs. S5 to S34). 3.1. Assessment of vermicidal activity of ILs Albendazole sulfoxide (ABZSO) was used as reference drugs during the screening of vermicidal activities (VAs) against P. posthuma of

581

different ionic liquids at four different concentrations (2, 4, 8 and 16 mM) in a time dependent manner. The details of their VA have been given in the electronic Supporting information (ESI Figs. S35 to S49). Vermicidal activities of 1-methyl-3-alkylimidazolium bromide ILs (i.e., [EMIM]Br, [BMIM]Br, [HMIM]Br, [OMIM]Br, and [DMIM]Br) (Figs. S36 and S37), 1-methyl-3-alkylimidazolium hydroxide (i.e., [EMIM] OH, [BMIM]OH, [HMIM]OH, [OMIM]OH, and [DMIM]OH) (Figs. S38 and S39) and 1-methyl-3-alkylimidazolium borontetrafluoride derived ILs (i.e., [EMIM]BF4, [BMIM]BF4, [HMIM]BF4, [OMIM]BF4, and [DMIM] BF4) (Figs. S40 and S41) at four different concentrations (i.e., 2, 4, 8 and 16 mM) were recorded as function of percentages paralysis and percentage mortality in a time dependent manner (15, 30, 45, 60, 75, 90 and 120 min) against the P. posthuma. The percentage paralysis and mortality of the earthworms varied drastically with respect to concentrations and types of ILs employed (Figs. S36 to S41). Fig. S48, clearly illustrates that at 4 mM dose these ILs can be arranged in descending order of their VA which are as follows: 1-methyl-3-alkylimidazolium bromide: [DMIM]Br N [OMIM]Br N [HMIM]Br N [BMIM]Br N [EMIM]Br (Fig. S48 a and b), 1-methyl-3-alkylimidazolium hydroxide: [DMIM]OH N [OMIM] OH N [HMIM]OH N [BMIM]OH N [EMIM]OH (Fig. S48 c and d), and 1methyl-3-alkylimidazolium boron tetrafluoride: [DMIM]BF4 N [OMIM] BF4 N [HMIM]BF4 N [BMIM]BF4 N [EMIM]BF4 (Fig. S48 c and d). The similar trends can be observed from the rest of the concentrations (2, 8 and 16 mM) too (Fig. S48 d and e) for the same ILs. The details discussions have been given in supplementary information page number 81 to 86. Similarly, ILs 1,2-dimethyl-3-alkylimidazolium bromide (i.e., [EMMIM]Br, [BMMIM]Br, [HMMIM]Br, [OMMIM]Br, [DMMIM]Br) (Figs. S42 and S43), 1,2-dimethyl-3-alkylimidazolium hydroxide (i.e., [EMMIM]OH, [BMMIM]OH, [HMMIM]OH [OMMIM]OH and [DMMIM] OH) (Figs. S44 and S45) and 1,2-dimethyl-3-alkylimidazolium boron tetrafluoride (i.e., [EMMIM]BF4, [BMMIM]BF4 [HMMIM]BF4 [OMMIM] BF4 and [DMMIM]BF4) (Figs. S46 & S47) derivatives were screened for their vermicidal activities. The vermicidal activity varied considerably with reverence to concentrations and kinds of ILs employed (Figs. S42 to S47). To make it more significant, a graph plotted at 4 mM for all five ILs with respect to time, expresses the dose dependent VA (Fig. S49). From supplementary Fig. S49, the following trends of the VA can be perceived: 1,2-dimethyl-3-alkylimidazolium bromide; [DMMIM]Br N [OMMIM]Br N [HMMIM]Br N [BMMIM]Br N [EMMIM]Br (Fig. S49 a & b), 1,2-dimethyl-3-alkylimidazoliumhydroxide [DMMIM]OH N [OMMIM] OH N [HMMIM]OH N [BMMIM]OH N [EMMIM]OH (Fig. S49 c and d), and 1,2-dimethyl-3-alkylimidazolium boron tetrafluoride; [DMMIM]BF4

Fig. 1. Comparison of % paralysis (a) and mortality (b) of 1-methyl-3-alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium derivatives (IL-Br, IL-BF4 and IL-OH) containing three different anions and ABZSO as standard drug.

582


Fig. 2. Cytotoxicity assay of [DMIM]Br, [DMIM]OH, [DMIM]BF4, [DMMIM]Br, [DMMIM]OH, and [DMMIM]BF4 on 3T3-L1 cells after 24 h. The data shows decrease in the % cell growth as a function of cytotoxic level of the compounds.

N [OMMIM]BF4 N [HMMIM]BF4 N [BMMIM]BF4 N [EMMIM] BF4 (Fig. S49 e and f). From outcome of our investigations, we have comprehended that VA depends on dose, hydrophobicity and nature of the N-alkyl chains, an− − ions (BF− 4 , Br , and OH ). Morover, IL-OH showed significant activities followed by IL-BF4, IL-Br and Albendazole sulfoxide (ABZSO) due to gradual decrease in positive charge on N-3 nitrogen atom [74], hydrophobicity and lipophilicity. Besides, there are few exceptions among the tested ILs (i.e., [EMIM]Br, [BMIM]Br and [EMIM]BF4) that showed slight poor activities and ILs (i.e., [BMIM]BF4, [EMIM]OH and [BMIM] OH) competitive activities with ABZSO. The possible explanations can be given from the literature [17,75,76]. One of the possible reason is attributed to their poor lipophilic and hydrophobic (N-alkyl side chains) ratio compared to the standard drug (ABZSO). 3.1.1. Role of C-2 hydrogen substituted with methyl group in imidazolium ring Here, our motive was to explain the role of hydrogen atom present at C-2 position of the imidazolium ring of ILs in VAs. Because, it has been claimed that presence of hydrogen atom at C-2 position of benzimidazole carbamates derivatives plays very significant role in anthelmintic activities of the clinically available anthelmintic drugs, but there are few exception of this concept (i.e., thiabendazole, cambendazole, triclabendazole etc.). Moreover, majority of the drugs exist and administrated in salt form [18–29]. Therefore, to develop imidazolium derive ILs as a potential anthelmintic candidate one need to understand the role of hydrogen atom presents at C-2 position of the imidazolium ring. In order to comprehend it, Fig. 1 depicts the average values of percentage mortality at 4 mM conc. [RMMIM]Br/or BF4 derivatives shows very competitive and slightly better VAs than [RMIM]Br/BF4 derivatives. Conversely, [RMIM]OH compounds show very competitive activities with [RMMIM]OH, except [EMIM]OH and [BMIM]OH (Fig. 1). It signifies that the substitution of

methyl group hamper the anthelmintic activity, due to increase in hydrophobic factors (substitution of methyl group at C-2 position of imidazolium ring) and lipophilicity change. Molecular modeling was also subsequently performed to emphasize the role hydrogen atom and other contributing factors in VAs. 3.2. Cell culture study Cytotoxicity of [DMIM]Br, [DMIM]OH, [DMIM]BF4, [DMMIM]Br, [DMMIM]OH and [DMMIM]BF4 were evaluated by the alamar blue assay. The cells were grown for 24 h and treated with 4 mM dose of each compound for the next 24 h. The results are shown in Fig. 2, which depicts the different cytotoxic levels of the test compounds. [DMIM]OH and [DMIM]BF4 showed around 36% and 32% cell survival. The trend was further followed by [DMIM]Br and [DMMIM]OH with around 15%, and 3% cell survival. However, [DMMIM]Br and [DMMIM] BF4 exhibit maximum level of cytotoxicity, with no cell survival. Among the test compounds [DMMIM] series of IL showed the highest level of cytotoxicity. It may be due the presence of methyl group at C2 position of imidazolium ring, which makes the [DMMIM] series of ILs more hydrophobic and more cytotoxic. Whereas, [DMIM] series of compounds showed intermediate levels of cytotoxic. 3.3. Density functional theory (DFT) based studies The geometries of thirty ILs were optimized at B3LYP/6-311G(d) level of theory by manual positioning [77,78] of anions. Out of converged structures, we selected the geometry corresponding to the lowest energy (Fig. S50). It was found that distance between cationic head group and anionic head group increases with the bulkiness of ions (Table S1). Table 2 Details of statistics of the developed quantum mechanical based GFA.

Table 1 List of quantum mechanical descriptors used in GFA. a

Descriptors

X

a_LUMO c_FPSA c_HOMO Intercept

−1.9063 −21.9340 9.3945 5.6136

DX

b

0.3971 5.2506 7.5093 2.0111

t-Test value −4.8003 −4.1774 1.2511 2.7913

c

p-Value

d

0.0001 0.0005 0.2261 0.0116

Xa, regression coefficient of descriptor; DXb, Standard error of descriptor regression coefficient; t-test valuec, t-statistics of the regression coefficient; p-valued, statistics of the calculated probability. Fractional Polar Surface Area of cation (c_FPSA), and lowest unoccupied molecular orbital energy of anion (a_LUMO) GFA Pred. activity = 5.6136 − 1.9063 ∗ a_LUMO − 21.934 ∗ c_FPSA + 9.3945 ∗ c_HOMO.

SL no.

Statistic

GFA

1 2 3

N r2 r2adjusted r2predicted RMSE residual error RMS residual error (cross-validation) Friedman lack-of-fit (LOF) S.O.R. p-value

30 0.8174 0.7880 0.7886 0.1626 0.1910 0.0352 3.1820

4 6 7 8

N = no. of components; r2 = regression coefficient; r2adjusted = adjusted correlation coefficient; RMSE = Root Mean Square Error; q2 = cross-validation, S.O.R. p-value = p-value for significance of regression.


Predicted Activity (GFA)

2.5

3.4. Genetic Function Approximation (GFA) based QSAR modeling

y = 0.8174x + 0.305 r² = 0.8174

2

1.5

1 Training Set Test Set

0.5 0.5

1

583

1.5 % Mortality

2

2.5

Fig. 3. Correlation plot of observed % mortality (on X-axis) versus the predicted activities (on Y-axis) by GFA model.

Quantitative structure activity relationship (QSAR) based study was performed via utilizing quantum mechanical based descriptors to identify the role of cation and anion vermicidal activity. As stated, the Genetic Function Approximation (GFA) technique was applied for the selection of optimal combination of descriptors out of the calculated descriptors (Table S2). The data set of 30 ligands was divided in a training set and test set of 23 and 7 respectively. Thereafter, the GFA based modeling was carried out by considering the parameters like 100 population size, 5000 iterations, 10% mutation probability, Friedman lack-offit (LOF) smoothness parameter 0.5, Max. Correlation 0.99, model development finger print FCFP_2 and total number of variables was thirty. Based on the RMSE, p-value and regression statistics, we selected the model equation (Table 1) with a_LUMO (LUMO energy of anion), c_FPSA (Fractional Polar Surface Area of cation) and c_HOMO (HOMO energy of cation) descriptors. From Table 2, anionic electron affinity (a_LUMO) and cation ionization energy (c_HOMO) and cationic surface area were the contributing factors that encode the observed trend of % mortality. This indicates that these descriptors contributed in activity of the tested ILs and thus supports our experimental findings that also

(b)

(a) THR136

ILE4 ARG241

PHE200

ILE47

GLN134 ARG46

LEU250

ILE24

HIS6

PHE167

THR237

VAL236 THR238 LEU42

THR237 PHE242

MET233

GLU27

(c)

(d)

HIS137

VAL169

GLN134

LEU250 PHE167

PHE200

CYS12

GLN134 THR136 PHE200

VAL236 THR237 HIS6

SER165 LEU250

GLY13 THR136

PHE167

TYR50 VAL229

MET230 VAL229 SER230

(e)

PHE20

THR238

ILE16

MET233

(f)

VAL169

THR136

SER165 LEU250

ILE202

GLN134

PHE167 VAL236

HIS6

THR238 ILE16

VAL229

GLY17

MET233 TYR50 ASN226

PHE20

Fig. 4. 3D view of (a) [EMIM]+ (b) [BMIM]+ (c) [HMIM]+ (d) [OMIM]+ (e) [DMIM]+ and (f) [RMIM]+ docked poses of β-tubulin inhibitors on the H. contortus β-tubulin model.

584


SER166

(a)

(b) ALA254

SER165

GLU198

GLN134

SER165

HSI6 PHE200 MET237

PHE167

PHE167 HIS6

LEU250

PHE200

THR237 PHE200

VAL236

TYR50

LEU253

PHE20

MET233 VAL236 SER234

ILE24

MET233

LEU240

(c)

(d)

HIS137

SER165 PHE200

THR136 GLN134

VAL169

THR136

HIS6

CYS12 GLN134

SER165

VAL236 THR237

HIS6

PHE200

ILE16

ILE16

LEU250 VAL229 PHE20

TYR50

VAL236

MET233

PHE167

VAL229

SER230

LYS19

TRY50

MET233

(e)

PHE20

(f) VAL169

ILE202

SER165 LEU250 HIS6 ILE16 PHE167

GLY17

VAL236 VAL229

TYR50

ASN226 PHE20 THR238

Fig. 5. 3D view of (a) [EMMIM]+ (b) [BMMIM]+ (c) [HMMIM]+ (d) [OMMIM]+ (e) [DMMIM]+ and (f) [RMMIM]+ docked poses of β-tubulin inhibitors on the H. contortus β-tubulin model.

account the effects of anion exchange. The list of descriptors is given in the above equation and Table 1 given below. t-Test values for the corresponding descriptors viz. a_LUMO, c_HOMO and c_FPSA (independent variables) were found to be significantly imperative. It reveals the statistical significance of the regression coefficients. The statistics of the developed quantum mechanical based GFA model have been given in Table 2, as the correlation coefficient (r2), predicted correlation coefficient (r2adjusted) and LOF. The obtained GFA statistics with significant regression coefficient (r2) 0.8174, p-value b 0.5 and RMSE was 0.01626 respectively, that signify the fitness ability of the QSAR model. In addition, the r2adjusted is a better measure of the ratio than r2 that avoids the over-parameter value for the GFA (0.7880). It indicates that the variation and correlation in the whole data were in good agreement. Leave one-out cross-validation r2cv value was 0.191 for the GFA model, indicates the predictability power of the generated model. Besides, lesser value (0.03526) of the Friedman lackof-fit (LOF), indicates that the developed GFA model resist the over fitting of the model and allows control over the smoothness of fit.

The correlation graph plotted between predicted activities (on the Y-axis) Vs observed % mortality (on the X-axis) in Fig. 3 shows good correlation with r2 = 81.74%. Thus, the predicted QSAR model can be used to foresee a new lead molecule against helminths. These predicted values are in good agreement with the experimental values. The observed % mortality and predicted values for the whole data set of the GFA models are listed in Tables S3 and S4 for the train set and tests set correspondingly (Supplementary data). 3.5. Molecular docking In continuation of our earlier published work and recent reports, colchicine binding domain of β-tubulin mainly involved in the anthelmintic activities through the alternate three binding consisting of zone-1 to 3 [55,67–69]. Therefore, we present here the molecular docking study of ten different cationic part of 1-methyl-3-alkylimidazolium derivatives (i.e., [EMIM]+, [BMIM]+, [HMIM]+, [OMIM]+ and [DMIM]+) and 1,2-dimethyl-3-alkylimidazolium derivatives (i.e., [EMMIM]+, [BMMIM]+, [HMMIM]+, [OMMIM]+ and [DMMIM]+) of ionic liquids with five


585

Table 3 Docking score of 1-methyl-3-alkylimidazolium and 1,2-dimethyl-3-alkylimidazolium derive ILs and interacting amino acids.

ILs [EMIM]+

Docking score (Kcal/mol) −21.6182

[BMIM]+

−27.8344

[HMIM]+

−30.6652

[OMIM]+

−38.678

[DMIM]+

−27.9547

[EMMIM]+

−23.8875

[BMMIM]+

−30.6931

[HMMIM]+

−33.5881

[OMMIM]+

−36.6832

[DMMIM]+

−27.9547

Interacting amino acids LEU250, PHE200, MET233, PHE20, SER165, THR237, VAL236, GLN134, TYR50, HIS6, THR136 TYR50, ILE4, THR237, VAL49, ILE24, ARG241, THR238, ILE47, GLU27, HIS28, ARG46, LEU42, GLN134, PHE242 SER230, VAL229, ILE16, THR136, TYR50, MET233, PHE20, HIS6, PHE167, PHE200, SER165, LEU250, GLN134, THR238, THR237, VAL236 HIS137, CYS12, VAL169, GLY13, VAL229, GLY17, LEU250, MET233, GLN134, ILE16, TYR50, THR136, HIS6, PHE20, PHE167, VAL236, PHE200, SER165, THR237 VAL236, TYR50, THR238, GLN134, LEU250, SER165, PHE200, THR237, PHE167, GLY17, ILE16, VAL229, ASN226, MET233, THR136, HIS6, PHE20, ILE202, VAL169, LEU250, SER165 THR238, ILE24, HIS6, THR237, SER165, PHE200, PHE20, VAL236, PHE167, SER166, GLN134, TYR230, MET233 HIS6, PHE20, MET233, SER165, PHE200, GLU198, LEU250, PHE167, THR237, VAL236, LEU253, MET257, ALA254, GLN134 PHE200, SER165, TYR50, HIS6, PHE167, THR136, LYS19, VAL229, LEU250, ILE16, VAL236, THR237, SER230, MET233, PHE20, GLN134 THR237, TYR50, VAL236, LEU250, PHE200, PHE167, ILE16, VAL169, PHE20, CYS12, SER165, HIS137, THR136, HIS6, MET233, GLY13, VAL229, GLN134 LEU250, SER165, PHE200, VAL236, THR238, HIS6, MET233, THR136, PHE20, ILE202, THR237, PHE167, GLY17, ILE16, VAL229, ASN226, VAL169, TYR50, GLN134

*Green color residues shows van der Wall interactions, light blue color shows π-π interaction, amino acid residues shows in pink color and shows interaction with positively charged nitrogen of imidazolium ring.

varying alkyl chains (n-C2H5, n-C4H9, n-C6H13, n-C8H17 and n-C10H21) to visualize the 2D interaction pattern of ligands given in Figs. S51 and S52 respectively. The 3D Dock pose view of ligand and receptor complex has been given in the Figs. 4 and 5 for [RMIM]+ and [RMMIM]+ respectively. The cationic part of ligands highlighted in red color (in a ball and CPK form) with interacting amino acids (Figs. 4 and 5). The list of interacting amino acids have been given in Table 3. Among them, amino acids viz. HIS6, MET233, GLN134, PHE200, PHE167, PHE20, MET233, THR237, VAL236 and SER165 are universally conserved in every ligand and receptor complex (Table 3). There were three types

of interactions observed i.e., van der Waal, π-π, hydrophobic and positively charged interactions. Residues like PHE20 and PHE200 interact with the positively charged nitrogen of imidazolium ring, π-π interactions by PHE242, HIS28, PHE20, PHE200, PHE167 whereas, GLN134 engaged in forming non-classical H-bond interaction (Table S5). However, remaining amino acids highlighted in green color to represent hydrophobic interactions with alkyl chains. On the other hand, it can be noticed that all five ligands of [RMIM]+ (Fig. 6b) and [RMMIM]+ (Fig. 6c) derivatives are superimposed over reference ligands (ABZSO) colored in blue. It can be comprehend from

Fig. 6. (a). Interacting amino acids from zone vid 1- to -3 (colchicine binding domain) are represented in the CPK view (orange -zone-1, blue -zone-2, light green -zone-3), identified according to SIFt patters, (b). Superimposed structures of [RMIM]+ (c) and [RMMIM]+ with reference ligands Albendazole sulfoxide (blue color). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

586


Fig. 7. The correlation plots of the average percent mortality of earthworms (in different time interval at 4 mM conc. in red color) and the docking score (in dotted black color) of the [RMIM] Br/or BF4/or OH and [RMMIM] Br/or BF4/or OH. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the interaction patterns of ligands that all the docked ligands overlap in zone-1 to 3 as has been highlighted in Fig. 6 (with three different colors: orange -zone-1, blue -zone-2, light green -zone-3). Interestingly, a major portion of the ligands overlapped in zone-2 (highlighted in fluorescence green color) in analogous to our previous findings [68]. It can be stated, that VAs was increases with increase in alkyl chain length and an escalation in magnitude of positive charge on quaternary nitrogen with substitution of different counter anions. Docking score for [RMIM]+ and [RMMIM]+ derivatives increases with increase in the alkyl chain (or hydrophobic factors); complemented the trend of observed experimental VAs (Fig. 7). Notably, docking score of [DMMIM]Br, [DMMIM]OH and [DMMIM]BF4 (circled in black color, Fig. 7) does not follow the generalized trends. This behavior conferre, that the increase in hydrophobicity and charge on the quaternary nitrogen of the imidazolium ring resulted in increase in VAs. 4. Conclusion A new class of 1,2-dimethyl-3-alkylimdiazolium and 1-methyl-3alkylimidazolium derived ionic liquids were synthesized and screened for their antivermicidal activity and cell viability study. We report that the nature of N-alkyl chains (or hydrophobic factors), anions and lipophilicity was accountable for triggering the vermicidal and cell cytotoxicity. IL-OH shows significant activities followed by IL-BF4, IL-Br and ABZSO; in agreement with earlier studies that suggest “decrease in positive charge on the quaternary nitrogen atom reduces the vermicidal activity”. Further, the study reveals that 1,2-dimethyl-3-alkylimidazolium derivatives were more active than 1-methyl-3-alkylimidazolium derivatives due to greater hydrophobic factors. Besides, QSAR study displayed that c_FPSA of cation, a_HOMO and a_LUMO of the anion exhibits a significant relationship (correlation coefficient for GFA (r2), 0.8174; RMSE, 0.16) with the vermicidal activity and strengthen our experimental findings. We conclude that hydroxyl group (OH−) largely increases the polar surface area of the cationic head then BF− 4 and Br− groups. The molecular docking studies of the cationic heads of the 1,2-dimethyl-3-imidazolium and 1-methyl-3-alkylimdiazolium derive ILs suggests that they binds in zone-2 and partially share the zone-1 and zone-3 of the colchicine binding domain. Docking scores and ligand superimposition complements the VAs of ILs. Overall, our study advocates that these ILs can be used as initial hits for the anthelmintics drug development processes. Moreover, efforts are under progress to address the

issues related to toxicity, bioavailability, and pharmacological features. Nonetheless, these findings are encouraging and offer new vistas for ILs advancement as a potential anthelmintic candidate. Acknowledgement Prabodh Ranjan thanks financial support from University Grants Commission (UGC), Govt. of India. Mr. Mohd Athar acknowledges generous support from the Department of Science and Technology (DST), Govt. of India as Doctoral INSPIRE Fellowship. Dr. Prakash Chandra Jha would also like to thanks Science and Engineering Research Board (SERB) for providing grants and Central University of Gujarat for providing basic computational facilities. Author contributions Prabodh Ranjan, Mohd Athar, Kari Vijayakrishna and Prakash C. Jha designed the research problem. Prabodh Ranjan and Mohd Athar performed the calculations and interpreted the results. H. Rather and R. Vasita performed the cytotoxiicty study. Conflict of interest Authors have no personal, financial or non-financial conflicts of interest. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.molliq.2018.02.001. References [1] S. Plotkin, et al., Hookworm vaccines, Clin. Infect. Dis. 46 (2) (2008) 282–288. [2] H. Hoste, et al., The effects of tannin-rich plants on parasitic nematodes in ruminants, Trends Parasitol. 22 (6) (2006) 253–261. [3] T. Eguale, et al., Haemonchus contortus: in vitro and in vivo anthelmintic activity of aqueous and hydro-alcoholic extracts of Hedera helix, Exp. Parasitol. 116 (4) (2007) 340–345. [4] T. Eguale, et al., In vitro and in vivo anthelmintic activity of crude extracts of Coriandrum sativum against Haemonchus contortus, J. Ethnopharmacol. 110 (3) (2007) 428–433. [5] D. Davyt, et al., A new indole derivative from the red alga Chondria atropurpurea. Isolation, structure determination, and anthelmintic activity 1, J. Nat. Prod. 61 (12) (1998) 1560–1563.

P. Ranjan et al. / Journal of Molecular Liquids 255 (2018) 578–588 [6] D.E. Akiyoshi, et al., Analysis of the β-tubulin genes from Enterocytozoon bieneusi isolates from a human and rhesus macaque, J. Eukaryot. Microbiol. 54 (1) (2007) 38–41. [7] S. Parida, et al., Anthelmintic potential of crude extracts and its various fractions of different parts of Pterospermum acerifolium Linn, Int. J. Pharma. Sci. Rev. Res. 1 (Suppl. 2) (2010) 107–111. [8] R.G. Mali, A.A. Mehta, A Review on Anthelmintic Plants, 2008. [9] G. Lämmler, H. Herzog, H.R. Schütze, Chemotherapeutic studies on Litomosoides carinii infection of Mastomys natalensis: 2. The activity of drugs against microfilariae, Bull. World Health Organ. 44 (6) (1971) 757. [10] N. Anand, S. Sharma, Approaches to Design and Synthesis of Antiparasitic Drugs, vol. 25, Elsevier, 1997. [11] V.E. Relf, et al., Anthelmintic efficacy on UK Thoroughbred stud farms, Int. J. Parasitol. 44 (8) (2014) 507–514. [12] J.C. Van De Steene, W.E. Lambert, Validation of a solid-phase extraction and liquid chromatography–electrospray tandem mass spectrometric method for the determination of nine basic pharmaceuticals in wastewater and surface water samples, J. Chromatogr. A 1182 (2) (2008) 153–160. [13] M. Zrnčić, et al., Analysis of anthelmintics in surface water by ultra high performance liquid chromatography coupled to quadrupole linear ion trap tandem mass spectrometry, Chemosphere 99 (2014) 224–232. [14] R. Prichard, Anthelmintic resistance, Vet. Parasitol. 54 (1–3) (1994) 259–268. [15] F. Nchu, et al., Anthelmintic and cytotoxic activities of extracts of Markhamia obtusifolia Sprague (Bignoniaceae), Vet. Parasitol. 183 (1) (2011) 184–188. [16] R. Dahiya, D. Pathak, Synthetic studies on novel benzimidazolopeptides with antimicrobial, cytotoxic and anthelmintic potential, Eur. J. Med. Chem. 42 (6) (2007) 772–798. [17] P. Köhler, The Biochemical Basis of Anthelmintic Action and Resistance, Elsevier, 2001. [18] S. Sharma, et al., Studies in potential filaricides. 19. Synthesis of 1-methyl-4substitutedcarbonylpiperazines as diethylcarbamazine analogs, Indian J. Chem., Sect. B: Org. Chem. Incl. Med. Chem. 26 (8) (1987) 748–751. [19] A.E. Mourad, D.S. Wise, L.B. Townsend, Synthesis of imidazo [1, 2-b] pyridazines: fenbendazole, oxifenbendazole analogs and related derivatives, J. Heterocycl. Chem. 30 (5) (1993) 1365–1372. [20] R. Chatterjee, et al., Comparative antifilarial efficacy of the N-oxides of diethylcarbamazine and two of its analogues, Trop. Med. Parasitol. 40 (4) (1989) 474–475. [21] E.F. Elslager, S. Perricone, F.H. Tendick, Antifilarial agents. I. Effects of 4-[(7-chloro-4quinolyl) amino]-. alpha.-(mono-and dialkylamino)-o-cresols and related compounds against Litomosoides carinii in gerbils, J. Med. Chem. 12 (6) (1969) 965–969. [22] S. Sharma, Advances in the treatment and control of tissue-dwelling helminth parasites, Progress in Drug Research/Fortschritte der Arzneimittelforschung/Progrès des recherches pharmaceutiques, Springer 1986, pp. 473–547. [23] S. Sharma, Treatment of filariasis—diethylcarbamazine and its congeners, Pharm. Res. 3 (2) (1986) 75–80. [24] M. Go, T. Ngiam, A.S. Wan, Synthesis of some novel amodiaquine analogs as potential antimalarial and antifilarial compounds, J. Med. Chem. 24 (12) (1981) 1471–1475. [25] P. Wadia, N. Anand, Studies in Potential Filaricides: Part III - Synthesis of Homopiperazine Analogues of Antifilarial Piperazines, 1958. [26] J. Reinertson, P. Thompson, Effects of JGS-11o (N, N-diethyl-4-methyUi-diazacycloheptaae-i-carboxamide, hydrochloride) against fiiariasis in cotton rats, Antibiot. Chemother. 1 (5) (1955) 566–570. [27] P.A. Sturm, et al., Antifilarial agents. 3-Aminopyrrolidine and 1, 4-diazabicyclo [3.2. 1] octane derivatives as analogs of diethylcarbamazine, J. Med. Chem. 20 (10) (1977) 1333–1337. [28] A. Bousvaros, B. Mueller, Thalidomide in gastrointestinal disorders, Drugs 61 (6) (2001) 777–787. [29] G.B. Vogelsang, et al., Thalidomide for the treatment of chronic graft-versus-host disease, N. Engl. J. Med. 326 (16) (1992) 1055–1058. [30] V. Kumar, S.V. Malhotra, Study on the potential anti-cancer activity of phosphonium and ammonium-based ionic liquids, Bioorg. Med. Chem. Lett. 19 (16) (2009) 4643–4646. [31] S.V. Malhotra, et al., Imidazolium-derived ionic salts induce inhibition of cancerous cell growth through apoptosis, Med. Chem. Commun. 5 (9) (2014) 1404–1409. [32] S.V. Malhotra, V. Kumar, A profile of the in vitro anti-tumor activity of imidazoliumbased ionic liquids, Bioorg. Med. Chem. Lett. 20 (2) (2010) 581–585. [33] R. Ferraz, et al., Ionic liquids as active pharmaceutical ingredients, ChemMedChem 6 (6) (2011) 975–985. [34] W.L. Hough-Troutman, et al., Ionic liquids with dual biological function: sweet and anti-microbial, hydrophobic quaternary ammonium-based salts, New J. Chem. 33 (1) (2009) 26–33. [35] W.L. Hough, et al., The third evolution of ionic liquids: active pharmaceutical ingredients, New J. Chem. 31 (8) (2007) 1429–1436. [36] M.S. Sitze, et al., Ionic liquids based on FeCl3 and FeCl2. Raman scattering and ab initio calculations, Inorg. Chem. 40 (10) (2001) 2298–2304. [37] N.K. Kaushik, et al., Synthesis and antiproliferative activity of ammonium and imidazolium ionic liquids against T98G brain cancer cells, Molecules 17 (12) (2012) 13727–13739. [38] K. Maddali, et al., Biological evaluation of imidazolium-and ammonium-based salts as HIV-1 integrase inhibitors, Med. Chem. Commun. 2 (2) (2011) 143–150. [39] M.R. Cole, et al., Design, synthesis, and biological evaluation of β-lactam antibioticbased imidazolium-and pyridinium-type ionic liquids, Chem. Biol. Drug Des. 78 (1) (2011) 33–41. [40] J. Pernak, P. Chwała, Synthesis and anti-microbial activities of choline-like quaternary ammonium chlorides, Eur. J. Med. Chem. 38 (11) (2003) 1035–1042.

587

[41] J. Pernak, I. Goc, I. Mirska, Anti-microbial activities of protic ionic liquids with lactate anion, Green Chem. 6 (7) (2004) 323–329. [42] J. Pernak, et al., Synthesis and anti-microbial activities of some pyridinium salts with alkoxymethyl hydrophobic group, Eur. J. Med. Chem. 36 (11) (2001) 899–907. [43] M. Ghavre, et al., Low toxicity functionalised imidazolium salts for task specific ionic liquid electrolytes in dye-sensitised solar cells: a step towards less hazardous energy production, Green Chem. 16 (4) (2014) 2252–2265. [44] G.G. Mandawad, et al., An efficient synthesis of isoxazoline libraries of thiophene analogs and its antimycobacterial investigation, Med. Chem. Res. 23 (10) (2014) 4455–4463. [45] P.N. Dube, et al., Synthesis and biologic evaluation of substituted 5-methyl-2-phenyl-1H-pyrazol-3 (2H)-one derivatives as selective COX-2 inhibitors: molecular docking study, Chem. Biol. Drug Des. 84 (4) (2014) 409–419. [46] M. Zhao, et al., Combined use of hydroxypropyl-β-cyclodextrin and ionic liquids for the simultaneous enantioseparation of four azole antifungals by CE and a study of the synergistic effect, J. Sep. Sci. 37 (1–2) (2014) 151–157. [47] P. Khloya, et al., Synthesis of some novel 4-arylidene pyrazoles as potential antimicrobial agents, Org. Med. Chem. Lett. 3 (1) (2013) 9. [48] J. Feder-Kubis, K. Tomczuk, The effect of the cationic structures of chiral ionic liquids on their antimicrobial activities, Tetrahedron 69 (21) (2013) 4190–4198. [49] M. Messali, Eco-friendly synthesis of a new class of pyridinium-based ionic liquids with attractive antimicrobial activity, Molecules 20 (8) (2015) 14936–14949. [50] N. Ferlin, et al., Biomass derived ionic liquids: synthesis from natural organic acids, characterization, toxicity, biodegradation and use as solvents for catalytic hydrogenation processes, Tetrahedron 69 (30) (2013) 6150–6161. [51] P. Ranjan, B.S. Kitawat, M. Singh, 1-Butylimidazole-derived ionic liquids: synthesis, characterisation and evaluation of their antibacterial, antifungal and anticancer activities, RSC Adv. 4 (96) (2014) 53634–53644. [52] Y. Yu, Y. Nie, Toxicity and antimicrobial activities of ionic liquids with halogen anion, J. Environ. Prot. 2 (03) (2011) 298. [53] S.V. Dzyuba, R.A. Bartsch, Efficient synthesis of 1-alkyl (aralkyl)-3-methyl (ethyl) imidazolium halides: precursors for room-temperature ionic liquids, J. Heterocycl. Chem. 38 (1) (2001) 265–268. [54] I. Dinarès, et al., Imidazolium ionic liquids: a simple anion exchange protocol, Green Chem. 11 (10) (2009) 1507–1510. [55] K.P. Charan, et al., Evaluation of imidazolium-based ionic liquids towards vermicidal activity: in vitro & in silico studies, RSC Adv. 5 (92) (2015) 75415–75424. [56] E. Ajaiyeoba, P. Onocha, O. Olarenwaju, In vitro anthelmintic properties of Buchholzia coriaceae and Gynandropsis gynandra extracts, Pharm. Biol. 39 (3) (2001) 217–220. [57] D. Jhala, H. Rather, R. Vasita, Polycaprolactone–chitosan nanofibers influence cell morphology to induce early osteogenic differentiation, Biomater. Sci. 4 (11) (2016) 1584–1595. [58] M. Frisch, et al., Gaussian 09, Revision a. 02, Gaussian, Inc., Wallingford, CT, 2009 200. [59] C.-W. Cho, S. Stolte, Y.-S. Yun, Comprehensive approach for predicting toxicological effects of ionic liquids on several biological systems using unified descriptors, Sci. Rep. 6 (2016). [60] N.L. Mai, et al., Prediction of cellulose dissolution in ionic liquids using molecular descriptors based QSAR model, J. Mol. Liq. 215 (2016) 541–548. [61] Accelrys Discovery Studio Version 4.0. , Accelrys, San Diego, USA, 2016http:// accelrys.com/products/collaborative-science/biovia-discovery-studio/. [62] S. Ma, et al., Computational study of the effects of cations and anions to the cytotoxicity of diverse ionic liquids by supervised machine learning, Chemom. Intell. Lab. Syst. 144 (2015) 138–147. [63] C. Croux, K. Joossens, Influence of observations on the misclassification probability in quadratic discriminant analysis, J. Multivar. Anal. 96 (2) (2005) 384–403. [64] M. Cruz-Monteagudo, M.N.D.S. Cordeiro, Chemoinformatics profiling of ionic liquids —uncovering structure-cytotoxicity relationships with network-like similarity graphs, Toxicol. Sci. 138 (1) (2013) 191–204. [65] M.W. Robinson, et al., A possible model of benzimidazole binding to β-tubulin disclosed by invoking an inter-domain movement, J. Mol. Graph. Model. 23 (3) (2004) 275–284. [66] F.C. Bernstein, et al., The protein data bank, FEBS J. 80 (2) (1977) 319–324. [67] Z. Deng, C. Chuaqui, J. Singh, Structural interaction fingerprint (SIFt): a novel method for analyzing three-dimensional protein−ligand binding interactions, J. Med. Chem. 47 (2) (2004) 337–344. [68] P. Ranjan, et al., Exploration of interaction zones of β-tubulin colchicine binding domain of helminths and binding mechanism of anthelmintics, Comput. Biol. Chem. 68 (2017) 78–91. [69] A. Massarotti, et al., The tubulin colchicine domain: a molecular modeling perspective, ChemMedChem 7 (1) (2012) 33–42. [70] V.Z. Spassov, L. Yan, P.K. Flook, The dominant role of side-chain backbone interactions in structural realization of amino acid code. ChiRotor: a side-chain prediction algorithm based on side-chain backbone interactions, Protein Sci. 16 (3) (2007) 494–506. [71] G. Wu, et al., Detailed analysis of grid-based molecular docking: a case study of CDOCKER—A CHARMm-based MD docking algorithm, J. Comput. Chem. 24 (13) (2003) 1549–1562. [72] J.R. Koska, et al., Fully automated molecular mechanics based induced fit protein−liigand docking method, J. Chem. Inf. Model. 48 (10) (2008) 1965–1973. [73] H. Singh, S. Sharma, R.N. Iyer, N. Anand, Indian J. Chem. 528 (14B) (1976). [74] P. Rousset, Maladies péritonéales: place et apport de l'imagerie par résonance magnétique, Université Claude Bernard-Lyon I, 2015.

588


[75] R. Aguayo-Ortiz, et al., Towards the identification of the binding site of benzimidazoles to β-tubulin of Trichinella spiralis: insights from computational and experimental data, J. Mol. Graph. Model. 41 (2013) 12–19. [76] L.M. MacDonald, et al., Characterisation of benzimidazole binding with recombinant tubulin from Giardia duodenalis, Encephalitozoon intestinalis, and Cryptosporidium parvum, Mol. Biochem. Parasitol. 138 (1) (2004) 89–96.

[77] C. Lee, W. Yang, R.G. Parr, Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density, Phys. Rev. B 37 (2) (1988) 785. [78] K. Vanasundari, et al., Spectroscopic investigation, vibrational assignments, Fukui functions, HOMO-LUMO, MEP and molecular docking evaluation of 4–[(3, 4– dichlorophenyl) amino] 2–methylidene 4–oxo butanoic acid by DFT method, J. Mol. Struct. 1147 (2017) 136–147.