Characterization and cytotoxicity evaluation of biocompatible amino acid esters used to convert salicylic acid into ionic liquids

Article abstract of DOI:10.1016/j.ijpharm.2018.05.021

The technological utility of active pharmaceutical ingredients (APIs) is greatly enhanced when they are transformed into ionic liquids (ILs). API-ILs have better solubility, thermal stability, and the efficacy in topical delivery than solid or crystalline drugs. However, toxicological issue of API-ILs is the main challenge for their application in drug delivery. To address this issue, 11 amino acid esters (AAEs) were synthesized and investigated as biocompatible counter cations for the poorly water-soluble drug salicylic acid (Sal) to form Sal-ILs. The AAEs were characterized using ¹H and ¹³C NMR, FTIR, elemental, and thermogravimetric analyses. The cytotoxicities of the AAE cations, Sal-ILs, and free Sal were investigated using mammalian cell lines (L929 and HeLa). The toxicities of the AAE cations greatly increased with inclusion of long alkyl chains, sulfur, and aromatic rings in the side groups of the cations. Ethyl esters of alanine, aspartic acid, and proline were selected as a low cytotoxic AAE. The cytotoxicities of the Sal-ILs drastically increased compared with the AAEs on incorporation of Sal into the cations, and were comparable to that of free Sal. Interestingly, the water miscibilities of the Sal-ILs were higher than that of free Sal, and the Sal-ILs were miscible with water at any ratio. A skin permeation study showed that the Sal-ILs penetrated through skin faster than the Sal sodium salt. These results suggest that AAEs could be used in biomedical applications to eliminate the use of traditional toxic solvents for transdermal delivery of poorly water-soluble drugs.

Full text of DOI:10.1016/j.ijpharm.2018.05.021

Accepted Manuscript
Characterization and cytotoxicity evaluation of biocompatible amino acid esters
used to convert salicylic acid into ionic liquids
Rahman Md. Moshikur, Chowdhury Md Raihan, Rie Wakabayashi, Yoshiro
Tahara, Muhammad Moniruzzaman, Masahiro Goto
PII:
S0378-5173(18)30319-3
https://doi.org/10.1016/j.ijpharm.2018.05.021
IJP 17492
DOI:
Reference:
International Journal of Pharmaceutics
To appear in:
Received Date:
Revised Date:
Accepted Date:
1 March 2018
10 April 2018
7 May 2018
Please cite this article as: R.M. Moshikur, C.M. Raihan, R. Wakabayashi, Y. Tahara, M. Moniruzzaman, M. Goto,
Characterization and cytotoxicity evaluation of biocompatible amino acid esters used to convert salicylic acid into
ionic liquids, International Journal of Pharmaceutics (2018), doi: https://doi.org/10.1016/j.ijpharm.2018.05.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Characterization and cytotoxicity evaluation of
biocompatible amino acid esters used to convert
salicylic acid into ionic liquids
Rahman Md. Moshikur1, Chowdhury Md Raihan1, Rie Wakabayashi1,2, Yoshiro Tahara1,
Muhammad Moniruzzaman3, and Masahiro Goto1,2*
1Department of Applied Chemistry, Graduate School of Engineering,
2Advanced Transdermal Drug Delivery System Center, Kyushu University, 744 Motooka,
Nishi-ku, Fukuoka 819-0395, Japan.
3Chemical Engineering Department, Universiti Teknologi PETRONAS, 32610 Seri Iskandar,
Perak, Malaysia.
*Corresponding author:
Prof. Masahiro Goto,
Graduate School of Engineering, Kyushu University, 744 Motooka, 819-0395 Fukuoka,
Japan. Fax: +81 92 802 2810; Tel: +81 92 802 2806.
e-mail: m-goto@mail.cstm.kyushu-u.ac.jp.
1

Abbreviations
AAE
Amino acid ester
API
Active pharmaceutical ingredient
L-Alanine ethyl ester
AlaEt
AspEt
IL
L-Aspartic acid diethyl ester
Ionic liquid
IC50
Half maximal inhibitory concentration
Minimum essential media
L-Proline ethyl ester
MEM
ProEt
ProBu
Sal
L-Proline isobutyl ester
Salicylic acid
TGA
Thermogravimetric analysis
ABSTRACT
The technological utility of active pharmaceutical ingredients (APIs) is greatly enhanced
when they are transformed into ionic liquids (ILs). API-ILs have better solubility, thermal
stability, and the efficacy in topical delivery than solid or crystalline drugs. However,
toxicological issue of API-ILs is the main challenge for their application in drug delivery. To
address this issue, 11 amino acid esters (AAEs) were synthesized and investigated as
biocompatible counter cations for the poorly water-soluble drug salicylic acid (Sal) to form
1
Sal-ILs. The AAEs were characterized using H and 13C NMR, FTIR, elemental, and
thermogravimetric analyses. The cytotoxicities of the AAE cations, Sal-ILs, and free Sal
were investigated using mammalian cell lines (L929 and HeLa). The toxicities of the AAE
cations greatly increased with inclusion of long alkyl chains, sulfur, and aromatic rings in the
side groups of the cations. Ethyl esters of alanine, aspartic acid, and proline were selected as
a low cytotoxic AAE. The cytotoxicities of the Sal-ILs drastically increased compared with
the AAEs on incorporation of Sal into the cations, and were comparable to that of free Sal.
Interestingly, the water miscibilities of the Sal-ILs were higher than that of free Sal, and the
Sal-ILs were miscible with water at any ratio. A skin permeation study showed that the Sal-
ILs penetrated through skin faster than the Sal sodium salt. These results suggest that AAEs
could be used in biomedical applications to eliminate the use of traditional toxic solvents for
transdermal delivery of poorly water-soluble drugs.
Keywords: Drug Development, Active Pharmaceutical Ingredient, Ionic Liquid, Amino acid,
Salicylic acid, Cytotoxicity.
2

1. Introduction
The pharmaceutical industry is facing unprecedented challenges to overcome
physicochemical obstacles to achieve optimum pharmacological responses for many drugs
because of their low solubility, polymorphism, and low bioavailability (Rodriguez-Alleret al.
2015; Egorova et al. 2017). Currently, poorly water-soluble drugs represent about 40% of
available oral drugs, 40%–50% of innovative active pharmaceutical ingredient (API) drugs,
and 90% of new chemical entities (Rodriguez-Aller et al. 2015; Martínez Rivas et al. 2017).
These drugs have poor and variable absorption rates in their therapeutic applications. High
doses of these drugs need to be administrated to reach the target therapeutic responses in oral
administration, which can cause inadequate bioavailability and long term gastrointestinal
mucosal toxicity in patients (Peterson et al. 2015). To improve drug solubilities and
therapeutic effects, researchers are investigating changing the dosage forms, physical forms,
and routes of administration to develop efficacious drug delivery systems. Recently, the
solubility characteristics of these drugs have been evaluated by applying several techniques,
either alone or in combination, such as lyophilization, solid dispersion, nanosuspensions,
nanoprecipitation, microencapsulation, soluble salt formation, inclusion complexes with β-
cyclodextrins, cocrystal formation, and ionic liquefaction (Rodriguez-Aller et al. 2015;
Martínez Rivas et al. 2017; Egorova et al. 2017; Lotfi et al. 2017).
Scientists have investigated ionic liquids (ILs) as potential candidates for effective
biocompatible drug delivery systems of poorly water-soluble drugs (Araki et al. 2015). ILs
have gained considerable attention as green designer solvents with unique physiochemical
properties, such as superior solvation properties, wide liquid ranges, negligible vapor
pressures, non-volatility, and non-flammability (Moniruzzaman et al. 2010; Egorova et al.
2015a; Lotfi et al. 2017). Recently, the cytotoxic effects of ILs have been evaluated in
different human and animal cell lines, bacteria, plants, and fungi (Adawiyah et al. 2016;
Egorova et al. 2017). However, ILs do not satisfy all green chemistry criteria, and in some
cases, ILs show higher toxicity than conventional organic solvents in various
biotechnological processes (Hou et al. 2013). Most commonly used imidazolium- and
pyridinium-based ILs are not considered ecofriendly (Hou et al. 2013). Therefore, non-toxic
and ecofriendly ILs need to be developed for sustainable pharmaceutical applications
(Adawiyah et al. 2016). Natural amino acids are potential candidates for synthesis of green
ILs because of their abundance, nontoxicity, biodegradability, biocompatibility, relatively
low price, and environmentally friendly behavior (Hou et al. 2013; Yazdani et al. 2016).
Amino acids have prolific structural diversity and can act as cations or anions in IL synthesis.
3

Recently, amino acid-based ILs, with an amino acid as the anion and phosphonium, choline,
alkyl ammonium, or dialkylimidazolium as the cation, have been developed (Egorova et al.
2017). These amino acid-based ILs possess interesting physicochemical and biological
activities. For example, addition of an amino acid decreases the cytotoxicity of 1-
(2(glycyloxy)ethyl)-3-methylimidazolium tetrafluoroborate compared with that of 1-ethyl-3-
methylimidazolium tetrafluoroborate (Egorova et al. 2015a). Amino acid derivatization at the
carboxyl group was used to reduce the possibility of formation of strong hydrogen bonds
involving carboxylic acid, which led to production of room temperature ILs in many cases.
As biocompatible counter cations, amino acid esters (AAEs) could be used to formulate
poorly water-soluble anionic drugs with attractive physicochemical properties, such as high
water solubility, thermal stability, and skin permeability (Furukawa et al. 2016). However,
the cytotoxicities of ILs synthesized with AAE cations have not been investigated.
Derivatization of poorly water-soluble APIs to their ionized forms in API-ILs is an excellent
approach to increase water solubility. This approach was first reported for pairing of salicylic
acid (Sal) with alkyl amines in propylene ethanol glycol solutions (Megwa et al. 2000). Sal is
commonly used as anti-inflammatory, analgesic, and antipyretic because of its excellent
bacteriostatic, fungicidal, keratolytic, and photoprotective properties (Bashir et al. 2005;
Magaña et. al 2016; Zavgorodnya et al. 2017). Recently, many researchers have formulated
API-ILs using both single active drugs (e.g., lidocaine, ibuprofen, procaine, and docusate)
and multiple active drugs (e.g., lidocaine–ibuprofen, salicylate–ibuprofen, and ephedrine–
ibuprofen) (Egorova et al. 2017). The API-IL formulation improves the physicochemical or
biological properties and is an efficient drug carrier in drug delivery systems (Furukawa, S.,
et al. 2016; Egorova et al. 2017; Moniruzzaman et al. 2017).
The aim of this study was to develop biocompatible amino acid-based API-ILs and to
evaluate their toxicities and skin permeation behavior using Sal as a model drug. We
synthesized and characterized various AAEs as cations, and optimized the alkyl chain length
based on the cytotoxicities with mammalian cell lines. Then, we synthesized Salicylate ionic
liquids (Sal-ILs) by incorporating the model drug into those AAEs with low cytotoxicities,
and investigated the toxicological effects of the Sal-ILs. Finally, skin permeation was
evaluated for selected Sal-ILs.
4

Figure 1. Structures, names, and abbreviations for (A) amino acid ester cations and (B) Sal-
ILs used in this study.
2. Material and methods
2.1. Materials
Salicylic acid (anhydrous, > 99% purity) was purchased from Sigma-Aldrich (Munich,
Germany). L-Proline, L-alanine, L-leucine, L-aspartic acid, sodium salicylate, and thionyl
chloride with high purity (>98.0 %) were obtained from Wako Pure Chemical Industries Ltd.
(Osaka, Japan). L-Glutamic acid, L-methionine, L-cysteine, L-tyrosine, and L-phenylalanine
with high purity (>98.0 %) were purchased from Kishida Chemical Co. Ltd (Osaka, Japan).
Methanol, ethanol, isopropanol, isobutanol, ammonia solution, diethyl ether, sodium chloride,
and octanol were purchased from Wako Pure Chemicals Industries. All reagents, solvents,
and other materials were of analytical grade and were used without any further purification.
Gibco minimum essential media (MEM), Opti-MEM, fetal bovine serum, and antibiotic-
antimycotic were purchased from Thermo Fisher Scientific (Waltham, MA, USA).
Dulbecco's phosphate buffered saline and trypsin/ethylenediaminetetraacetic acid (0.25%
5

trypsin/1 mM ethylenediaminetetraacetic acid) were purchased from Nacalai Tesque (Kyoto,
Japan). A WST-8 cell counting kit containing 2-(2-methoxy-4-nitrophenyl)-3-(4-
nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt was purchased from
Dojindo Molecular Technologies, Inc. (Kumamoto, Japan). Mammalian cell lines L9292 and
HeLa were provided by the RIKEN cell bank (Tsukuba, Japan).
2.2. General synthetic procedure for AAE
AAEs were synthesized using a slight modification of an established method (Furukawa et al.
2016). Briefly, a suspension of 2.0 g of an amino acid in 20 mL of the required alcohol (e.g.,
methanol for methyl ester, ethanol for ethyl ester) was stirred vigorously in an ice bath. Then,
thionyl chloride (mole ratio of thionyl chloride: amino acid = 1.5:1) was added dropwise to
the slurry, and the mixture was kept on ice for 1 h. The resultant solution was stirred
thoroughly for 24 h at room temperature and then concentrated under reduced pressure after
completion of the reaction. The oily residue was dissolved in alcohol (10 mL) and the solvent
was evaporated under reduced pressure to yield the AAE hydrochloride salt. Then, 10 mL of
distilled water was added to the solution, followed by neutralization with ammonia solution
(mole ratio of ammonia: AAE salt = 2:1). Diethyl ether (50 mL) was added, and the solution
was stirred for 2 h at room temperature. The organic layer was extracted and concentrated
under reduced pressure to yield the AAE. The yields of all the synthesized compounds were 
1
13
72.0% (Table S1). All the synthesized compounds were characterized by H and C nuclear
magnetic resonance spectroscopy (NMR) and elemental analysis to check their structures and
purities.
Scheme 1. General synthetic procedure for amino acid alkyl ester. (a) Thionyl chloride-
mediated esterification in alcohol on ice for 1 h, and then at room temperature for 24 h; (b)
ammonia solution and diethyl ether at room temperature for 2 h; and (c) in the dark at 40 °C
for 2 h.
2.3. Synthesis of API-ILs
6

An equimolar mixture of the synthesized AAE (Section 2.2) and Sal was stirred thoroughly in
the dark for 2 h at 40 °C (Scheme 1) (Furukawa et al. 2016). The derived Sal-ILs were clear
and colorless liquids, and their ionization and purities were evaluated using 1H and 13C NMR
spectroscopy.
2.4. NMR Measurements
All NMR measurements were recorded on a JEOL ECZ400S 400 MHz spectrometer (Tokyo,
Japan) operating at 25 °C. Deuterated chloroform (2.0 ppm) or methanol (7.3 ppm) was used
as the solvent, and tetramethylsilane as the internal standard. The NMR solvents were
purchased from Wako Pure Chemical Industries Ltd. The coupling constants (J) are reported
in Hertz (Hz). The same instrument was used to record 13C NMR spectra with complete
proton decoupling (deuterated chloroform, 76.8–77.5 ppm). The spectra were processed using
the DeltaV software package (version 5.0.5.1, JEOL).
The purities of the synthesized compounds were calculated using the following equation:
Purity (%) = [Σ I (product)/Σ I (total)] × 100………………………………………………. (1)
where, I represents the relative area of each signal (Zavgorodnya et al. 2017).
2.5. Elemental analysis
Elemental analysis was carried out for quantitative analysis of carbon, hydrogen, and nitrogen
in organic matter using a Yanaco CHNC MT-5 analyzer (Yanagimoto Seisakusho, Kyoto,
Japan).
2.6. Thermogravimetric analysis
Thermogravimetric analysis (TGA) was carried out on Hitachi High-Technologies TG/DTA
7300 (Tokyo, Japan). Samples weighing about 10 to 20 mg were analyzed in an alumina
crucible using nitrogen as a purge gas. The sample analysis was conducted from 30 to 500 °C
using a constant heating rate of 5 °C min−1. An isothermal hold at 70 °C for 30 min was
performed to remove excess volatile solvents.
2.7. Fourier transform infrared analysis
The Fourier transform infrared (FT-IR) analyses of the Sal-ILs were carried out on a Perkin
Elmer (Frontier FT/IR, Waltham, MA), in attenuated total reflectance mode. All spectra were
obtained with accumulation of 20 scans in the wavenumber range 400–4000 cm−1.
7

2.8. Partitioning coefficient determination
The synthesized Sal-ILs were characterized by measuring the log P in octanol and water.
Free Sal (2–3 mg) containing Sal-ILs was mixed with 5 mL of water and 5 mL of octanol.
The mixtures were left overnight at room temperature with constant shaking in the dark. Then,
each solution was centrifuged at 9100g for 30 min to separate the water and octanol layers.
The concentration of Sal in each layer was quantified by ultraviolet–visible spectroscopy (V-
750, JASCO, Tokyo, Japan) with known concentrations of standards at 297 nm. The partition
coefficient, log Po/w, was determined using the following equation:
log Po/w = log (Soluteoctanol/Solutewater)................................................................................... (2)
where, Soluteoctanol and Solutewater represent the concentration of free Sal dissolved in octanol
and water, respectively.
2.9. Water miscibility determination
A viscous sample of Sal-IL (0.7215 g of [Sal][ProEt] containing 0.3565 g of free Sal) was
mixed with 0.3 mL of water in a transparent glass tube. A clear solution without a precipitate
was obtained after shaking for a few minutes. Then, the liquid phase was filtered using 0.2
µm syringe filter to remove the remaining any particles. The concentration of Sal was
quantified by ultraviolet–visible spectroscopy (V-750, JASCO, Tokyo, Japan) with known
concentrations of standards at 297 nm.
2.10. Cytotoxicity evaluation
The WST cell viability assay was conducted using mammalian cell lines L9292 and HeLa as
described in the literature, with some modifications (Furukawa et al. 2016). The cells were
cultivated until reaching approximately 70% monolayer and trypsinized to collect the cells
from the cell culture dish. Then, the cells were seeded into 96-well flat-bottomed plates at
5000 cells/well and cultured in MEM (containing 10 % fetal bovine serum and 1% antibiotic-
antimycotic) for 24 h at 37 °C in a CO2 incubator. AAEs were dissolved in Opti-MEM at
concentrations in the range 1–200 mM. For Sal-ILs and free Sal, the concentration range was
0.001–40 mM because of the limited water solubility of free Sal. One hundred microliters of
each sample solution was replaced with the medium in each well, and the plates were
incubated for 24 h. After incubation, the solution was removed from each well, and the wells
were rinsed twice with Dulbecco's phosphate-buffered saline. Then, 100 µL of Opti-MEM
containing 10 µL of WST assay reagent from the cell counting kit was added to each well to
8

measure the mitochondrial activity of the cells. After incubation for 3 h at 37 °C in a CO2
incubator, the absorbance of the supernatant at 450 nm (Atreated) was measured using a
microplate spectrophotometer (Bio-Rad, Tokyo, Japan). The absorbance of cells soaked in
Opti-MEM medium without AAEs or Sal-ILs was used as a control (Acontrol). The cytotoxicity
is expressed as the relative cell viability, and was calculated using the following equation:
Cell viability (%) = (Atreated/Acontrol)  100……………………………………………….. (3)
All tests were repeated three times and the average values are reported. Statistical analysis of
the data was performed using Microsoft Excel 2016 (Microsoft, Washington, USA). The half
maximal inhibitory concentration (IC50) was calculated for each cation and Sal-IL. Data are
present as the mean with the standard deviation.
2.11. Skin permeation study
Skin permeation tests were carried out using a temperature controlled Franz diffusion cell
(Furukawa et al. 2016). Phosphate buffer solution was used to dilute the samples. The buffer
solution was prepared by mixing 8 g/L of NaCl, 0.2 g/L of KCl, 2.9 g/L of Na2HPO4 and 0.2
g/L of KH2PO4, respectively.Then, the pH of buffer solution was adjusted to 7.4 with NaOH
or HCl solution.
Shoulder skin of a female Yucatan micro pig (Charles River Japan Inc., Tokyo, Japan) was
used for this study. The fat layer was carefully removed from the skin, and then the skin was
cut into small pieces and placed in the Franz diffusion cell with the inside of the skin facing
the receptor phase. The receptor phase was 5 mL of phosphate solution with constant stirring
at 32.5 °C. [Sal][Na] and Sal-IL solutions were prepared using the diluent. Then, a 0.3 mL
aliquot of each sample solution, with an approximate concentration of 3.2 µmol/cm2, was
applied on the outside of skin (donor phase). The donor chamber was covered with parafilm
to minimize adsorption of water from the atmosphere. Aliquots (0.1 mL) of the receiving
phase were withdrawn at fixed intervals for 60 h. After each sampling, 0.1 mL of fresh
diluent was added to maintain a constant volume in the receiving chamber.
The withdrawn samples were diluted and the concentration of Sal was quantified using high
performance liquid chromatography (HPLC) after calibration with known concentrations of
free Sal standard at 245 nm.
9

2.12. Statistical Analysis
Statistical analysis was performed using GraphPad Prism version 6 software (GraphPad
Software, Inc., La Jolla, CA). Statistical significance was evaluated by one-way analysis of
variance (ANOVA) followed by Tukey’s post hoc test for multiple comparisons.
3. Results and discussion
3.1. Synthesis, characterization, and cytotoxicities of the AAEs
3.1.1. Synthesis and characterization of the AAEs
We selected several natural, aromatic, aliphatic, sulfur-containing, and acidic amino acids
(proline, alanine, aspartic acid, glutamic acid, leucine, methionine, phenylalanine, and
tyrosine) that reportedly have low toxicities (Hou et al. 2013; Yazdani et al. 2016), and used
them to prepare AAEs (Figure 1A, Scheme 1a and 1b). To investigate the role of cation alkyl
chain length in toxicity, we synthesized proline-based AAEs with various alkyl chain lengths
(C1–C4) (Figure 1A). The synthesized AAEs were clear and colorless liquids at room
temperature, and were all water miscible except for L-tyrosine ethyl ester (Table S1 and
1
Figure S1). The purities of all the isolated compounds determined using H NMR were 
97.0% (Table S1) (Zavgorodnya et al. 2017). Elemental analysis of the synthesized
compounds confirmed the structures of the AAEs (Table S1).
3.1.2. Cytotoxicities of the AAEs
The mammalian cell lines L929 and HeLa were used to investigate the cytotoxicities of the
synthesized AAE-based cations (Figure 1A). Before comparing the cytotoxicities of the
AAEs, we adjusted the pH of the AAEs containing medium to 7.4 using HCl because the pH
increased on addition of AAEs, in which the –COOH groups were esterified to –COOR
groups and the amino acids were transformed to basic AAE compounds. In all cases, the IC50
values of the AAEs solutions (without pH control) were higher than those with pH control
(Figure S2). The lipophilicity of each AAE increased in the solutions with pH control
because a higher degree of protonation enhanced the penetration rate into the cell membrane,
which disrupted the cell structure (Kobayashi et al. 2007).
First, we evaluated the effect of the alkyl chain length on cytotoxicity using proline-based
cations at pH 7.4. Both cell lines were treated with different doses of proline-based ester
cations with methyl (C1), ethyl (C2), isopropyl (C3), and isobutyl (C4) substituents (Figure S3).
The IC50 values were calculated by a mathematical logarithm function (Figure 2A). The IC50
10

values of L-proline methyl ester, L-proline ethyl ester (ProEt), L-proline isopropyl ester, and
L-proline isobutyl ester (ProBu) were 17.6, 18.3, 6.5 and 2.9, respectively, in L929 cells. The
corresponding values in HeLa cells were 38.9, 44.8, 33.4 and 15.2 mM. The toxicities of the
cations depended on the alkyl chain length in the ester group. The cation toxicity increased
with increasing alkyl chain length from C2 to C4. Among all the cations in both cell lines, the
lowest IC50 (highest toxicity) was obtained for ProBu because of the strong interaction
between the lipophilic long alkyl chain of ProBu and the plasma cell membrane (Jing et al.
2016). This interaction has been proven by an integrated experimental and computational
approach (Yoo et al. 2016). Generally, the hydrophobic or lipophilic nature of an IL increases
with increasing alkyl chain length in the cation. High hydrophobicity leads to interaction with
the lipid bilayer of biological membranes and enhances the penetration of cations into the cell
membrane. Consequently, the lipid bilayer becomes swollen and the cell membrane
physiological functions are disrupted, ultimately leading to cell death (Jing et al. 2016; Yoo et
al. 2016; Egorova et al. 2017). By contrast, ProEt showed low cytotoxicity because of its
short alkyl chain, which resulted in weak interactions with the cell membrane and increased
mobility of the cation in the medium (Kumar et al. 2009). L-Proline methyl ester showed
relatively high toxicity compared with ProEt. A similar result was found in toxicity studies of
L-leucine methyl ester and L-leucine ethyl ester (LeuEt) in Natural killer (NK) cells
(Rabinovitch el at 1986) and sophorolipid derivatives in human pancreatic cancer (HPAC)
cells (Fu et al. 2008). Further studies are needed to define the precise mechanisms involved.
However, our results showed that ethyl ester-based amino acid cations had lower toxicities
than cations with other alkyl chain lengths.
HeLa
L929
B
A
HeLa
L929
****
***
**
80
60
40
20
0
60
40
20
0
****
**
ns
****
ProMe ProEt ProPr ProBu
AlaEt AspEt ProEt GluEtMetEtLeuEt PheEt
11

Figure 2. Cytotoxicities of AAEs towards L929 and HeLa cell lines at pH 7.4. Effect of (A)
alkyl chain length derived from alcohols, and (B) side chains derived from amino acids. n = 3,
mean ± SD, ns for non-significant, ** for P < 0.01, *** for P < 0.001, **** for P < 0.0001.
Next, we evaluated the influence of the amino acid side chain on cytotoxicity using ethyl
ester-based AAEs (Figure 2B and Figure S4). High toxicity was observed when the cation
contained sulfur, a long alkyl chain, and an aromatic ring. L-Alanine ethyl ester (AlaEt)
showed lower toxicity than the other AAEs in both cell lines because of the short carbon
chain in its structure. If the number of carbons increased in the cation structure, the
cytotoxicity gradually increased. For example, LeuEt and L-glutamic acid diethyl ester
showed higher toxicities than AlaEt and L-aspartic acid diethyl ester (AspEt), respectively.
These results are similar to those reported by Hou et al. (2013) and Yazdani et al. (2016). The
hydrophobic/lipophilic nature of ILs in solutions reportedly increases with increasing carbon
chain length or the presence of functional groups in the cation core (Yazdani et al. 2016;
Egorova et al. 2017), and the cytotoxicities of ILs depend on the cation molecular weight
(Monnery et al. 2017; Egorova et al. 2017). ProEt also showed lower toxicity than the other
cations. The presence of a cyclic secondary amine in the proline structure could reduce the
toxicity because of the higher basicity of the secondary amine compared with that of the
primary amine. Among all the AAE cations, L-phenylalanine ethyl ester showed the lowest
IC50 in both cell lines because of the presence of an aromatic ring in the cation. This high
cytotoxicity of PheEt could be attributed to a strong π-π interaction between the cation and
the cell membrane, leading to rapid disruption of the cell structure (Hou et al. 2013).
3.2. Synthesis, characterization, and cytotoxicities of the Sal-ILs
3.2.1. Synthesis and characterization of the Sal-ILs
AAE-based API-ILs were synthesized by ionization of a poorly water-soluble drug (Sal) and
AAEs with relatively low cytotoxicities (AlaEt, ProEt, and AspEt). All API-ILs were clear
and colorless liquids. The results of 1H NMR analysis showed the stoichiometry between the
two constituents for the [Sal][ProEt] compounds was 1:1 (Figure 3). The acidic proton of Sal
(-COOH peak at approximately 10.37 ppm in Figure 3B) was absent in the Sal-IL spectra
(Figure 3C), and a new broad peak was detected at 8.96 ppm, which corresponded to the
protonated amine group. These results indicated that all of the Sal and ProEt were converted
into [Sal][ProEt].
12

Figure 3. 1H NMR spectra of ProEt (A), free Sal (B), and [Sal][ProEt] (C).
To further characterize the Sal-ILs, we obtained a FT-IR spectrum of [Sal][AspEt] and
compared it to FT-IR spectra of AspEt, free Sal, and [Sal][Na] (Figure 4). In the FT-IR
spectrum of [Sal][AspEt], characteristic peaks were observed at 1668 cm−1 and 1609 cm−1 for
C=O stretching of –COOH groups and C=C stretching of the Sal benzene ring, respectively.
In the [Sal][Na] spectrum, these peaks shifted slightly to 1650 cm−1 and 1615 cm−1. Another
characteristic peak was observed at 1731 cm−1 for C=O stretching of –COOR groups of
AspEt, and this shifted to 1727 cm−1 in [AspEt]. An ethyl group C-H stretching peak was
observed at 2982 cm−1 in the [Sal][AspEt] spectrum and attributed to the ethyl group of
AspEt. A broad peak at 3241 cm−1 for the –OH group of Sal was not observed in the
spectrum of [Sal][AspEt] or [Sal][Na]. Another characteristic peak at 3384 cm-1 for N-H
bending of the primary amine in AspEt was absent in the [Sal][AspEt] spectrum. However, a
weak shoulder peak appeared at 1592 cm−1, and this could be assigned to C=O and –NH2
stretching. A similar weak shoulder peak at 1582 cm−1 was observed in the spectrum of
[Sal][Na].
13

Figure 4. FT-IR spectra of free Sal, AspEt, [Sal][AspEt], and [Sal][Na].
Thermal stability is important for ILs and mainly depends on the chemical structure. The
thermal stabilities of the synthesized Sal-ILs were investigated by TGA, and we measured the
onset (T5%onset) temperatures (Table 1 and Figure 5). When Sal was used as the anion, the
thermal stabilities of the Sal-ILs depended on the cation structures. Thermal decomposition
(T5% onset) of [Sal][ProEt], [Sal][AlaEt], and [Sal][AspEt] occurred at 125.7 °C, 126.1 °C, and
136.4 °C, respectively, whereas the value previously reported for Sal was 162 °C (T5% onset
)
(Zavgorodnya et al. 2017). Therefore, the synthesized Sal-ILs showed lower thermal stability
than free Sal, but all of the compounds decomposed at temperatures above 100 °C.
[Sal][AspEt] had higher thermal stability than [Sal][ProEt] and [Sal][AlaEt] because of the
high van der Waals forces between the alkyl side chains of [Sal][AspEt] (Fukumoto et al.
2005).
14

Figure 5. Thermogravimetric analysis of [Sal][AspEt], [Sal][ProEt], and [Sal][AlaEt].
However, the solubility of Sal greatly increased in the IL form. The Sal-ILs were miscible
with water at any ratio (Table 1 and Figure 6), whereas free Sal is only slightly water-soluble
(2.2 g/L) (Egorova et al. 2015b).
Table 1. Physical properties and water miscibility of Sal-ILs at room temperature.
Compound
Appearance
T5% onset (°C)
Estimated solubility in water
g Sal-IL/ 0.3 mL
> 0.722[a]
> 0.623[a]
> 0.708[a]
-
g active/ mL
[Sal][ProEt]
[Sal][AlaEt]
[Sal][AspEt]
Sal
Clear colorless
liquid
125.7
126.1
136.4
162[b]
> 1.130[a]
>0.967[a]
> 0.958[a]
0.0022[c]
Crystalline solid
[a] Estimated lower limit of solubility in water; [b] Data was obtained from Zavgorodnya et
al. 2017; [c] Data was obtained from Egorova et al. 2015b.
15

Figure 6. Water miscibility of Sal-ILs. 1: Sal-IL viscous sample in transparent glass tube. 2:
Added water in Sal-IL sample, 3: Clear solution formed after shaking.
3.2.2. Cytotoxicities of the Sal-ILs
We investigated the cytotoxicities of the Sal-ILs synthesized from basic AAE cations and
acidic Sal. The Sal in free acid form showed higher toxicity than the Sal-ILs, and the IC50
values were similar in both HeLa and L929 cell lines (Table 2). The cytotoxicities of the Sal-
ILs were much higher than those of the AAEs, because Sal is a cytotoxin with a strong
anticancer effect (Egorova et al. 2015b). When cells were treated with Sal-ILs, hydrophilic
cations reduced the lipophilicities of Sal-ILs and slightly reduced the toxicities of the Sal-ILs.
In both cell lines, [Sal][AlaEt] is low cytotoxicity because of short alkyl chain in the side
group on the AlaEt cation. Moreover, The hydrophilicity of [Sal][AlaEt] is higher than that of
[Sal][AspEt] or [Sal][ProEt], which could reduce the penetration of Sal through cell
membrane by decreasing the lipophilicity of the IL form (log Po/w in Table 2). Consequently,
the cytotoxicities of the Sal-ILs and Sal were similar (Table 2).
Table 2. Cytotoxicities of the Sal-ILs towards L929 and HeLa cell lines with partitioning
coefficient.
Compound
[Sal][ProEt]
[Sal][AlaEt]
[Sal][AspEt]
Sal
Log Po/w HeLa (IC50, mM) L929 (IC50, mM)
-0.40
-0.60
-0.19
-
4.2
4.9
4.6
3.2
4.1
6.2
3.7
3.0
16

3.2.3. Skin permeation studies of Sal-ILs
To evaluate possible medical applications in a transdermal drug delivery, we investigated
skin permeation of the Sal-ILs at 32.5 °C using a Franz diffusion cell system with the sodium
salt of Sal in a PBS solution as a control. Transport of [Sal][Na] and Sal-ILs in the receiver
phase was detected by HPLC analysis at regular intervals. The Sal-ILs easily permeated
through the skin, at a speed approximately nine times that of [Sal][Na] (Figure 7). After 60 h,
the cumulative transport of Sal-IL for [Sal][AspEt] was 0.18 mg/cm2, whereas the transport
of [Sal][Na] was 0.03 mg/cm2. The differences between the neutral and ionized forms of Sal
were similar to data from previous studies, which demonstrated that ionic species cannot
easily permeate the skin barrier (Zavgorodnya et al. 2017). This difference in permeation also
arises because [Sal][Na] is more hydrophilic (log Po/w of −1.5) than the Sal-ILs (log Po/w in
Table 2). Therefore, it is difficult for [Sal][Na] to permeate through the skin compared with
the Sal-ILs.
0.3
****
***
**
0.2
0.1
0
Figure 7. Skin permeation of Sal-ILs and [Sal][Na] after 60 h. n = 3, mean ± SD, ** for
P < 0.01, *** for P < 0.001, **** for P < 0.0001.
Variation in skin permeation because of strong ion-pairing between a cation and an anionic
drug was reported previously for ibuprofen and ProEt cations (Furukawa et al. 2016). In that
case, the speed of permeation of [ibuprofen][ProEt] through pig skin was 10 times that of free
ibuprofen. However, among the synthesized Sal-ILs, the AspEt cation-based Sal-IL
permeated the skin faster than either the ProEt or AlaEt cation-based Sal-IL. The cumulative
amount of drug in the receiver phase for [Sal][AspEt] was approximately three times that of
17

[Sal][AlaEt] and 1.5 times that of [Sal][ProEt] after 60 h. The presence of a long alkyl chain
in the side group on the cation could enhance skin transport of Sal by increasing the
lipophilicity of the IL form (log Po/w in Table 2).
4. Conclusions
We have demonstrated a new approach to develop API–ILs that are biocompatible and
effectively penetration the skin. All synthesized compounds are clear and colorless liquids at
room temperature, except for L-tyrosine ethyl ester, with high purities (> 97.0 %). Among the
synthesized AAEs, AlaEt, AspEt, and ProEt had low cytotoxicities and were investigated as
potential components for synthesis of Sal-ILs with the anionic drug Sal. All synthesized Sal-
ILs were fully ionized and had high thermal stabilities. In contrast to Sal, the Sal-ILs show
good solubility in aqueous media. When toxic Sal was incorporated into amino acid cations,
the cytotoxicity decreased slightly. However, a skin permeation study demonstrated that skin
penetration of the Sal-ILs was approximately nine times faster than that of the sodium salt of
Sal. The AspEt formulation showed better transport behavior than ProEt, AlaEt, and
[Sal][Na]. We did not use any environmentally unfriendly reagents to increase the solubilities
or enhance skin permeation of the Sal-ILs. Therefore, this API-IL technique can be used for
effective formulation of poorly water-soluble drugs in an IL form, and it eliminates the use of
traditional solvent vehicles for transdermal drug delivery.
Acknowledgments
This work was supported by a Grant-in-Aid for Scientific Research (S) JP16H06369 from the
Ministry of Education, Culture, Sports, Science, and Technology of Japan. M.M.R thanks the
Government of Japan (MEXT) for providing a scholarship. The authors also thank Mr. Y.
Hirano for help with elemental analysis and Dr. M. Watanabe for facility support for NMR
and TGA.
Declaration of interests
None.
Appendix A.
Supplementary material
18

References
Araki, S., Wakabayashi, R., Moniruzzaman, M., Kamiya, N., Goto, M., 2015. Ionic
liquid-mediated transcutaneous protein delivery with solid-in-oil nanodispersions.
Med Chem Commun 6, 2124–2128. https://doi.org/10.1039/C5MD00378D
Adawiyah, N., Moniruzzaman, M., Hawatulaila, S., Goto, M., 2016. Ionic liquids as a
potential tool for drug delivery systems. Med Chem Commun 7, 1881–1897.
https://doi.org/10.1039/C6MD00358C.
Bashir, S.J., Dreher, F., Chew, A.L., Zhai, H., Levin, C., Stern, R., Maibach, H.I., 2005.
Cutaneous bioassay of salicylic acid as a keratolytic. Int J Pharm 292, 187–194.
https://doi.org/10.1016/j.ijpharm.2004.11.032
Egorova, K.S., Seitkalieva, M.M., Posvyatenko, A. V., Ananikov, V.P., 2015a. An
unexpected increase of toxicity of amino acid-containing ionic liquids. Toxicol Res
(Camb) 4, 152–159. https://doi.org/10.1039/C4TX00079J
Egorova, K.S., Seitkalieva, M.M., Posvyatenko, A. V., Khrustalev, V.N., Ananikov, V.P.,
2015b. Cytotoxic Activity of Salicylic Acid-Containing Drug Models with Ionic and
Covalent Binding. ACS Med Chem Lett 6, 1099–1104.
https://doi.org/10.1021/acsmedchemlett.5b00258
Egorova, K.S., Gordeev, E.G., Ananikov, V.P., 2017. Biological Activity of Ionic Liquids
and Their Application in Pharmaceutics and Medicine. Chem Rev 117, 7132–7189.
https://doi.org/10.1021/acs.chemrev.6b00562
Fukumoto, K., Yoshizawa, M., Ohno, H., 2005. Room temperature ionic liquids from 20
natural amino acids. J Am ChemSoc 127, 2398–2399.
https://doi.org/10.1021/ja043451i
Furukawa, S., Hattori, G., Sakai, S., Kamiya, N., 2016. Highly efficient and low toxic
skin penetrants composed of amino acid ionic liquids. RSC Adv 6, 87753–87755.
https://doi.org/10.1039/C6RA16926K
Fu, S.L., Wallner, S.R., Bowne, W.B., Hagler, M.D., Zenilman, M.E., Gross, R., Bluth,
M.H., 2008. Sophorolipids and Their Derivatives Are Lethal Against Human
Pancreatic Cancer Cells. J Surg Res 148, 77–82.
https://doi.org/10.1016/j.jss.2008.03.005
Hou, X.D., Liu, Q.P., Smith, T.J., Li, N., Zong, M.H., 2013. Evaluation of Toxicity and
Biodegradability of Cholinium Amino Acids Ionic Liquids. PLoS One 8.
https://doi.org/10.1371/journal.pone.0059145
Jing, B., Lan, N., Qiu, J., Zhu, Y., 2016. Interaction of Ionic Liquids with a Lipid
Bilayer: A Biophysical Study of Ionic Liquid Cytotoxicity. J PhysChem B 120,
2781–2789. https://doi.org/10.1021/acs.jpcb.6b00362
19

Kobayashi, M., Kagawa, T., Takano, R., Itagaki, S., Hirano, T., Iseki, K., 2007. Effect of
medium pH on the cytotoxicity of hydrophilic statins. J Pharm PharmSci 10, 332–
339.
Kumar, R. A., Papaïconomou, N., Lee, J. M., Salminen, J., Clark, D. S., &Prausnitz, J.
M. 2009. In vitro cytotoxicities of ionic liquids: Effect of cation rings, functional
groups, and anions. Environmental Toxicology, 24(4), 388–395.
https://doi.org/10.1002/tox.20443
Lotfi, M., Moniruzzaman, M., Sivapragasam, M., Kandasamy, S., Abdul Mutalib, M.I.,
Alitheen, N.B., Goto, M., 2017. Solubility of acyclovir in nontoxic and
biodegradable ionic liquids: COSMO-RS prediction and experimental verification.
J Mol Liq 243, 124–131. https://doi.org/10.1016/j.molliq.2017.08.020
Magaña, H., Palomino, K., Cornejo-Bravo, J.M., Díaz-Gómez, L., Concheiro, A.,
Zavala-Lagunes, E., Alvarez-Lorenzo, C., Bucio, E., 2016. Polymeric prodrug–
functionalized polypropylene films for sustained release of salicylic acid. Int J
Pharm 511, 579–585. https://doi.org/10.1016/j.ijpharm.2016.07.044
Martínez Rivas, C.J., Tarhini, M., Badri, W., Miladi, K., Greige-Gerges, H., Nazari,
Q.A., Galindo Rodríguez, S.A., Román, R.Á., Fessi, H., Elaissari, A., 2017.
Nanoprecipitation process: From encapsulation to drug delivery. Int J Pharm 532,
66–81. https://doi.org/10.1016/j.ijpharm.2017.08.064
Megwa, S. a, Cross, S.E., Benson, H. a, Roberts, M.S., 2000. Ion-pair formation as a
strategy to enhance topical delivery of salicylic acid. J Pharm Pharmacol 52, 919–
928. https://doi.org/10.1211/0022357001774804
Moniruzzaman, M., Tamura, M., Tahara, Y., Kamiya, N., Goto, M., 2010. Ionic liquid-
in-oil microemulsion as a potential carrier of sparingly soluble drug:
Characterization and cytotoxicity evaluation. Int J Pharm 400, 243–250.
https://doi.org/10.1016/j.ijpharm.2010.08.034
Moniruzzaman, M., Mahmood, H., Goto, M., 2017. Ionic Liquid Based Nanocarriers for
Topical and Transdermal Drug Delivery. Eftekhari, A. (Eds.), Ionic Liquid Devices.
RSC, Smart Materials Series No.28, pp. 390-403.
Monnery, B.D., Wright, M., Cavill, R., Hoogenboom, R., Shaunak, S., Steinke, J.H.G.,
Thanou, M., 2017. Cytotoxicity of polycations: Relationship of molecular weight
and the hydrolytic theory of the mechanism of toxicity. Int J Pharm 521, 249–258.
https://doi.org/10.1016/j.ijpharm.2017.02.048
Peterson, D.E., Boers-Doets, C.B., Bensadoun, R.J., Herrstedt, J., 2015. Management of
oral and gastrointestinal mucosal injury: ESMO Clinical Practice Guidelines for
diagnosis, treatment, and follow-up. Ann Oncol 26, v139–v151.
https://doi.org/10.1093/annonc/mdv202
Rabinovitch, M., Zilberfarb, V., Ramazeilles, C., 1986. Destruction of
Leishmaniamexicanaamazonensis amastigotes within macrophages by
20

lysosomotropic amino acid esters. J Exp Med 163, 520–35.
https://doi.org/10.1084/jem.163.3.520
Rodriguez-Aller, M., Guillarme, D., Veuthey, J.L., Gurny, R., 2015. Strategies for
formulating and delivering poorly water-soluble drugs. J Drug DelivSciTechnol 30,
342–351. https://doi.org/10.1016/j.jddst.2015.05.009
Yazdani, A., Sivapragasam, M., Leveque, J.M., Moniruzzaman, M., 2016. Microbial
Biocompatibility and Biodegradability of Choline-Amino Acid Based Ionic Liquids.
J MicrobBiochemTechnol 8, 415–421. https://doi.org/10.4172/1948-5948.1000318
Yoo, B., Jing, B., Jones, S.E., Lamberti, G.A., Zhu, Y., Shah, J.K., Maginn, E.J., 2016.
Molecular mechanisms of ionic liquid cytotoxicity probed by an integrated
experimental and computational approach. Sci Rep 6, 2–8.
https://doi.org/10.1038/srep19889
Zavgorodnya, O., Shamshina, J.L., Mittenthal, M., McCrary, P.D., Rachiero, G.P., Titi,
H.M., Rogers, R.D., 2017. Polyethylene glycol derivatization of the non-active ion in
active pharmaceutical ingredient ionic liquids enhances transdermal delivery. New
J Chem 41, 1499–1508. https://doi.org/10.1039/C6NJ03709G
21

Graphical Abstract
22