Improvement of New Dianionic Ionic Liquids vs Monoanionic in Solubility of Poorly Water-Soluble Drugs

Article abstract of DOI:10.1016/j.xphs.2021.01.014

New ionic liquids (ILs) based on dianionic phosphonate anions and ammonium cations were prepared and characterized. They were used as excipients to increase the water solubility of two oral drugs, piroxicam and ibuprofen, that are slightly soluble in water. An increment in solubility of 300-fold was achieved for ibuprofen when compared with pure water, with only 0.25 mol% of IL in water. Interestingly, this was achieved with the less toxic dianionic ionic liquid [N_{4 1 2OH 2OH}]₂ [C₂H₅PO₃], which presents an IC₅₀ of 120 mM (≈0.25 mol%). On the other hand, piroxicam showed an increase of 480-fold for the same dianionic ionic liquid, with the same ionic liquid percentage. In contrast, for monoanionic ionic liquids, the effect was not so pronounced, and only a 10-fold was obtained, in the presence of 0.3 mol% of IL. The lipophilicity (logP) of drugs decreased in the presence of these ILs. Cytotoxicity profile of these ILs was determined and they did not show a significant impact towards healthy fibroblasts. The cytotoxicity of ibuprofen and piroxicam was also determined, and cellular viability almost did not change when ionic liquid was in the presence of 1 mM of oral drug.

Full text of DOI:10.1016/j.xphs.2021.01.014

Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
Contents lists available at ScienceDirect
Journal of Pharmaceutical Sciences
journal homepage: www.jpharmsci.org
Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Improvement of New Dianionic Ionic Liquids vs Monoanionic in
Solubility of Poorly Water-Soluble Drugs
ꢀ
Daniela A.S. Agostinho, Ana R. Jesus, Ana B.P. Silva, Jose M.S.S. Esperança,
Alexandre Paiva, Ana R.C. Duarte**, Patrícia M. Reis*
^
LAQV, Requimte, Departamento de Química, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, Caparica, 2829-516, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
New ionic liquids (ILs) based on dianionic phosphonate anions and ammonium cations were prepared
and characterized. They were used as excipients to increase the water solubility of two oral drugs, pir-
oxicam and ibuprofen, that are slightly soluble in water. An increment in solubility of 300-fold was
achieved for ibuprofen when compared with pure water, with only 0.25 mol% of IL in water. Interestingly,
Received 3 December 2020
Accepted 12 January 2021
Available online 21 January 2021
this was achieved with the less toxic dianionic ionic liquid [N4 1 2OH 2OH]2 [C2H5PO3], which presents an
Keywords:
Ionic liquids
Dianionic ionic liquids
Cytotoxicity
Drug delivery
Bioavailability
IC50 of 120 mM (z0.25 mol%). On the other hand, piroxicam showed an increase of 480-fold for the same
dianionic ionic liquid, with the same ionic liquid percentage. In contrast, for monoanionic ionic liquids,
the effect was not so pronounced, and only a 10-fold was obtained, in the presence of 0.3 mol% of IL. The
lipophilicity (logP) of drugs decreased in the presence of these ILs. Cytotoxicity profile of these ILs was
determined and they did not show a significant impact towards healthy fibroblasts. The cytotoxicity of
ibuprofen and piroxicam was also determined, and cellular viability almost did not change when ionic
liquid was in the presence of 1 mM of oral drug.
© 2021 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Introduction
pharmacokinetic and pharmacodynamics properties of drugs. A
number of strategies to improve the oral delivery of many drugs
Ease of administration, cost-effectiveness, and flexibility in
design of the dosage form are some reasons why oral drug delivery
is the most advantageous and used form for drug delivery.1 The oral
drug bioavailability depends on several aspects, such as aqueous
solubility and drug permeability. Oral drugs that are poorly water
soluble (less than 100 mg/L2), show a limited dissolution rate and
often low bioavailability, thus high doses are necessary to achieve
therapeutic plasma concentrations after oral administration, but on
its turn, it can lead to toxicity in the gastrointestinal mucosal.3
The low bioavailability and high lipophilicity of oral drugs in
combination with high-dose requirements is a problem that phar-
macy industry still has to solve.4
include prodrugs, salt formation, crystal engineering, solid disper-
sions, micellar systems and the use of excipients.3 The common
organic solvent excipients that are used to enhance the poor sol-
ubility of drugs are pyridine, N,N-dimethylformamide (DMF) and
dimethyl sulfoxide (DMSO). These solvents are volatile, flammable
and toxic, so ionic liquids are potential green alternatives for this
purpose.6 ILs are organic salts that are typically liquid below 100 ꢀC.
Some ILs are classified as room temperature ionic liquids (RTILs)
that exist in liquid state at ambient temperature.7 ILs possess
numerous unique and useful physico-chemical properties
including low vapor pressure, good dissolution capabilities, wide
liquid range and good thermal stability.8 As designer solvents, ILs
properties could be manageable, simply by selecting the appro-
priate cation/anion combinations, by changing the side chain
length of ions or by incorporating distinct functionalities in the
cation.9 The assessment of cytotoxicity is a mandatory preliminary
test before application of ILs in either medical or pharmaceutical
applications.5,10 Cholinium-based ILs are some of the most biode-
gradable, less toxic, easily accessible and cost-effective ILs, when
compared with other ILs such as pyridinium and imidazolium-
based ILs.11e13
In the last few years, ionic liquids (ILs) have increasingly being
exploited as solvents, co-solvents and active pharmaceutical in-
gredients because of their unique and tunable physicochemical
properties.5 The use of ILs can markedly improve the
* Corresponding author.
** Corresponding author.
E-mail addresses: ard08968@fct.unl.pt (A.R.C. Duarte), preis@fct.unl.pt
(P.M. Reis).
https://doi.org/10.1016/j.xphs.2021.01.014
0022-3549/© 2021 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.

2490
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
Fig. 1. Chemical structure of ILs used: a e monoanionic and b e dianionic ionic liquids.
We have previously shown that N-acetyl amino acid N-alkyl
prepared 10 new biocompatible ionic liquids, 5 monoanionic and 5
dianionic ionic liquids. The cation is based in the well-known ammo-
nium cation, functionalized with different number of hydroxyethyl
groups, and ether moietie and/or different alkyl chains, and the anion is
based on phosphate and phosphonate groups, as presented in Fig. 1.
These new ILs were tested as excipients for two oral drugs,
ibuprofen and piroxicam, which structure is presented in Fig. 2.
The application of this new ammonium ionic liquids as green
excipients was evaluated, and their influence on the solubility of
two poorly water soluble nonsteroidal anti-inflammatory drugs
(NSAID), ibuprofen and piroxicam is reported herein. These drugs
have analgesics and antipyretics properties, due to the inhibition of
the cyclooxygenase enzyme, and are usually used in the treatment
of headache, dysmenorrhea, dental pain, post-operative, rheuma-
toid arthritis, spondylitis, gouty arthritis and osteoarthritis.18e22
cholinium-based ILs were able to significantly increase the water
solubility of two commercial drugs, paracetamol and sodium
diclofenac, with no significant effect on normal human cells
viability.14 Our group also synthesized new biocompatible N-alkyl
cholinium N-alkyl sulfonate ionic liquids, that were used to
enhance the aqueous solubility of the same drugs.15 In addition,
four new dianionic ionic liquids, containing a succinyl-DL-alaninate
anion and N-alkyl cholinium cation, were also prepared.
The aim of this work is to test the use of dianionic ILs as excipients,
for the dissolution of low water soluble drugs, such as piroxicam and
ibuprofen. Excipients are very important because of their ability to
transport the active drug to the site in the body where the drug is
intended to exert its action.16,17 They also keep the drug from being
released too early in the assimilation process. Therefore, we have

D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
2491
Fig. 2. Chemical structure of Ibuprofen (left) and Piroxicam (right).
Materials and Methods
dried in the vacuum for 1e2 days, and their purity confirmed by 1H
NMR. These ILs were obtained in 95e99% yield.
Materials
General Procedure for the Metathesis Reaction
N-Methyldiethanolamine (ꢁ98.0%), 1-butylphosphonic acid
(98.0%), ethylphosphonic acid (ꢁ98.0%), methylphosphonic acid
An aqueous solution of N-alkyl derivative cholinium bromide or
chloride (20 mmol) was slowly passed through an anion exchange
column Amberlite™ IRN-78. Then, the corresponding hydroxide
solution was slowly added to a solution of dibutyl phosphate (1.0
equiv.) diluted in acetone or the corresponding aqueous alkyl-
phosphonate acid (0.5 equiv.). The reaction mixture was stirred at
room temperature for 2e3 h prior to water removal by evaporation.
Ionic liquids were washed with diethyl ether, obtained in quanti-
tative yield, and dried in high vacuum for 2 days at 50e60 ꢀC to
guarantee minimum water content. Coulometric Karl-Fischer ti-
trations yielded final water contents below 1000e1500 ppm
depending on the IL. Moreover, AgNO3 test was used to confirm the
absence of halogen presence in the final IL.
(98.0%), 1-octanol (99.0%), Amberlite™ IRN-78, 4-isobutyl-a-
methylphenylacetic acid (Ibuprofen) (99.0%), piroxicam (97.0%
min.) and -lactose (98.0% min.) were purchased from Alfa Aesar. 1-
D
chlorobutane (ꢁ99.0%), 2-chloroethanol (99.0%), N-butyldietha-
nolamine (ꢁ98.6%), dibutyl phosphate (ꢁ97.0%), 1-bromobutane
(ꢁ98.0%), 2-dimethylaminoethanol (ꢁ99.5%), 2-bromoethyl ethyl
ether (90.0%) and sodium bicarbonate (ꢁ99.7%) were acquired from
Sigma-Aldrich. Diethyl ether (ꢁ99.8%) and acetone (ꢁ99.5%) were
€
obtained from Honeywell e Riedel-de-Haen. Ethanol absolute
anhydrous, deuterium oxide (99.9%) and hexane were purchased
from Carlo Erba Reagents, Eurisotop and Valente e Ribeiro Lda.,
respectively.
Solubility Assays
General Procedure for the Synthesis of [N4 1 2OH 2OH]Cl and [N4 2OH
2OH 2OH]Cl
A calibration curve was prepared for each drug, in water, at their
maximum absorbance wavelength, by UVeVis VWR® spectro-
photometer model UV-6300PC, 222 nm for ibuprofen and 353 nm
for piroxicam, at 25 ꢀC, see Figure S1. Absorbance values were kept
below 1. The statistical analysis was performed by Graphpad
Prism 7.
Methyldiethanolamine or butyldiethanolamine (100 mmol) was
mixed with the alkylating agent, chlorobutane or 2-chloroethanol
respectively (4.0 equiv.), and n-hexane in a 100 mL pressure reac-
tion vessel. The mixture was kept at 80e90 ꢀC for 4 days. The
chloride ionic liquids were thoroughly washed with diethyl ether to
remove the unreacted alkylating agent. Ionic liquids were then
dried in the vacuum for 1e2 days, and their purity confirmed by 1H
NMR. These ILs were obtained in 95e99% yield.
General Procedure for the Determination of Solubility Limit of
Ibuprofen and Piroxicam
Aqueous solutions (2 mL) of ionic liquids were prepared.
Ibuprofen or piroxicam was added to each sample guaranteeing
that we are above the solubility limit. In the case of ibuprofen, we
used an aqueous solution of IL at the same concentration as our
blank sample, because these ILs absorb in the region 190e230 nm.
All solutions were stirred for 24 h at 37 ꢀC until equilibrium was
reached. Keeping the solutions in a 37 ꢀC thermostatized water
bath, the non-dissolved drugs were filtered off through a 0.22 mm
syringe filter. Samples of each filtered solution were diluted in
MilliQ® water, until the absorbance was smaller than 1. The drug
General Procedure for the Synthesis of [N4 1 1 2OH]Br, [N2O2 1 1 2OH]Br
and [N1 1 2OH 2OH]Cl
2-Dimethylaminoethanol (100 mmol) was mixed with the
alkylating agent, 1-bromobutane, 2-bromo ethyl ether or 2-
chloroethanol respectively (1.1 equiv.), and n-hexane in a 100 mL
pressure reaction vessel. The mixture was kept at 80e90 ꢀC for 24 h.
The ionic liquids were thoroughly washed with diethyl ether to
remove the unreacted alkylating agent. Ionic liquids were then

2492
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
concentration was analysed in a UVeVis VWR® spectrophotom-
eter, model UV-6300PC. Triplicates were prepared for each sample
and maximum absorbance was measured at 222 nm and 353 nm to
ibuprofen and piroxicam, respectively. Solubility of drugs in the
presence of each ionic liquid was calculated based on the calibra-
tion curve. The pH was checked during assay and no variation was
observed.
Kow and log P of the three samples were determined using
Equations (1) and (2), and the mean value was taken.
.
Kow ¼ ½drugꢂoct ½drugꢂaq
log P ¼ log Kow
(1)
(2)
General Procedure for the Determination of Partition Coefficient and
Permeability
Log P
Permeability measurements were conducted using a glass
Franz-type diffusion cell (PermeGear) with a 10 mL reactor
compartment with an effective mass transfer area of 1 cm2.24,25
The membrane used was a polyethersulphone (PES-U), with
The Kow and log P values of ibuprofen and piroxicam in the
presence of ionic liquids were determined using the shake-flask
method reported in literature23 and also in our previous work.14
Briefly, an octanol-saturated aqueous solution was used to pre-
pare the ionic liquid solutions at 0.05 mol%. Known amounts of
drug were dissolved in 1 g of IL solution. Afterwards, we mixed it
with 1 g of a water-saturated octanol solution. The solutions were
stirred vigorously for 18e24 h and then sat for some hours until full
separation of organic and aqueous layers. Three independent ex-
periments were performed. The drug concentration in the water-
rich phase was analysed in a UVeVis VWR® spectrophotometer,
model UV-6300PC, using a previously prepared calibration curve,
see Figure S2.
150 mm thickness and 0.45 mm pore size (Sartorius Stedim Biotech,
Germany), which was placed between the two compartments and
held with a stainless-steel clamp. The receptor compartment (8 mL)
was filled with MilliQ® water and donor compartment (2 mL)
loaded with a IL solution (0.05 mol%) containing the oral drug
(40 mg), which was filtered after being stirred 24 h at 37 ꢀC. Ali-
quots of 200 mL were collected from the receptor compartment at
predetermined time periods (0, 5, 15, 30 min and 1, 2, 3, 4, 6, 8 h)
and fresh MilliQ® water was added to complete the volume. The
concentration of the drug in the receptor compartment was ana-
lysed in a microplate reader (VICTOR Nivo™ PerkinElmer, USA) at
222 nm for ibuprofen and 353 nm for piroxicam. The experiments
were performed at 37 ꢀC and triplicates were performed.
The concentration of drug in the octanol-rich phase was directly
calculated by subtracting the amount in the water-rich phase to the
initial amount dissolved.
Fig. 3. Solubility of Ibuprofen in aqueous solutions with different mol percentages of monoanionic ionic liquids. The dashed line represents the solubility limit of ibuprofen in water.
The results were expressed as the mean SD of the three independent experiments. Significantly statistical differences are represented in asterisks: ****p < 0.0001 and ns - non-
significant (as compared to values of the same IL with different percentages).

D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
2493
Fig. 4. Solubility of Ibuprofen in aqueous solutions with different mol percentages of dianionic ionic liquids. The dashed line indicates ibuprofen solubility in water and is use for
comparison purposes. The results were expressed as the mean SD of the three independent experiments. Statistical significant differences are represented in asterisks:
****p < 0.0001, ***p ¼ 0.0001 or p ¼ 0.0002 ([N4 1 2OH [C2H5PO3] and [N4 1 2OH [CH3PO3], respectively).
]
]
2
2OH
2
2OH
The permeability, P (cm sꢃ1) was calculated according to
Equation (3):
Cytotoxicity Evaluation
Cell Culture
Fibroblasts L929 cells (L929 (DSMZ - German Collection of Mi-
croorganisms and cell culture GmbH) were cultivated in MEM
(Eagle's Minimum Essential Medium) supplemented with 10% of
FBS and 1% penicillin-streptomycin at 37 ꢀC and 5% CO2. The sup-
plements and MEM were obtained from Corning, USA.
Ct
ꢃlnð1 ꢃ 2 Þ ¼
C0
2A
V
ꢄ P ꢄ t
(3)
where Ct is the concentration in the receptor compartment at time
t, C0 is the initial concentration in the donor compartment, A is the
effective mass transfer area, V is the total volume of solution in both
compartments.
Viability Assays
Fibroblasts were incubated for 24 h at 37 ꢀC and 5% CO2 in a 96-
well plate at a density of 1.0 ꢄ 104 cells/well with different concen-
trations of IL, ranging from 25 to 250 mM. Control cells were incu-
bated with complete media only. After 24 h, cell monolayers were
washed with PBS and cell viability was evaluated using the CellTiter
96® AQueous One Solution Cell Proliferation Assay (Promega), which
is based on tetrazolium active component ((3-(4,5-dimethylthiazol-
2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazo-
lium, MTS). The amount of formazan product formed was measured
by spectroscopy, in a microplate reader (VICTOR Nivo™ PerkinElmer,
USA) at 490 nm, as absorbance is directly proportional to the number
of viable cells in culture. Cell viability was expressed as percentage of
cells exposed to extracts vs control cells. Graphpad Prism 7 was used
for statistical analysis. Data is expressed as average standard devi-
ation from at least three independent experiments.
According to a derived solution of Fick's law of diffusion it is
possible to determine the diffusion coefficient, D (cm2 ꢃ1) of the
s
ibuprofen and piroxicam in the presence of ILs, through the
membrane, as can be seen in Equation (4):
V1 ꢄ V2
V1 þ V2
h
A
1
t
Cf ꢃ Ci
Cf ꢃ Ct
D ¼
ꢄ
ꢄ
ln ð
Þ
(4)
where Cf and Ci are the final and initial concentration in the re-
ceptor compartment and Ct is the concentration in the receptor
compartment at time t. V1 and V2 are the volume in the donor and
in the receptor compartment respectively and h is the thickness of
the membrane.
The partition coefficient, Kd is defined as measure of the solu-
bility of the solute in the membrane, and is calculated as follow in
Equation (5):
Results and Discussion
Solubility Assays
P ꢄ h
Kd ¼
(5)
D
In this work, the solubility of ibuprofen and piroxicam, two
nonsteroidal anti-inflammatory drugs (NSAID) typically orally
administered, and which present a low solubility in aqueous media,
where P is permeability, h is thickness of the membrane and D is
the diffusion coefficient.

2494
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
Fig. 5. Solubility of Piroxicam in aqueous solutions with different mol percentages of dianionic ionic liquids. The dashed line indicates buprofen solubility in water and is use for
comparison purposes. The results were expressed as the mean SD of the three independent experiments. Statistical significant differences are represented in asterisks:
****p < 0.0001, ***p ¼ 0.0003 ([N4 2OH 2OH 2OH 2 [C4H9PO3]), p ¼ 0.0008 or p ¼ 0.0009 ([N4 1 2OH 2OH 2 [C2H5PO3] 0.05%e0.1% and 0.2%e0.25%, respectively) and p ¼ 0.0010 [N4 1 2OH
[CH3PO3]).
]
]
]
2
2OH
was determined. The assays were performed at 37 ꢀC to mimic body
temperature. At this temperature, their solubility was 60 mg/L and
32 mg/L, respectively, which is consistent with literature data.26
The effect of a small quantity of IL, as excipient, to promote the
solubility of both drugs was evaluated. New monoanionic ionic
liquids were tested as excipients, for ibuprofen, with promising
results as presented in Fig. 3. Even at 0.3 mol% of [N4
]
1 1 2OH
[C8H18PO4] and [N4 1 2OH 2OH][C8H18PO4], it was possible to achieve
an increment of 10-fold in the solubility of ibuprofen when
compared to its value in pure water.
In contrast, dianionic ionic liquids induced a greater increment
in solubility compared to monoanionic ionic liquids. The solubility
Figure-6. Solubility of Ibuprofen and Piroxicam with dianionic ILs and two commonly used excipients (0.2 mol%) compared to solubility of two drugs in water. The results were
expressed as the mean SD of the three independent experiments. Statistical significant differences are represented in asterisks: ****p < 0.0001, Ibuprofen: **p ¼ 0.0013 or
p ¼ 0.0085 ([N4 1 2OH 2OH
]
2 [C4H9PO3] - [N4 2OH 2OH 2OH
]
2 [C4H9PO3] or [N4 1 2OH 2OH
[C4H9PO3] - [N4
]
2 [C2H5PO3] - [N4 1 2OH 2OH
]
2 [CH3PO3], respectively) and *p ¼ 0.0119 ([N4 1 1 2OH 2 [C4H9PO3] - [N4
[C4H9PO3] - [N4 ] [C4H9PO3]) and
2OH 2OH 2OH 2
]
]
[CH3PO3]); Piroxicam: ** *p ¼ 0.0005 ([N4 1 1
]
]
[C4H9PO3]), **p ¼ 0.0032 ([N4 1 1
]
1
2OH 2OH
2
2OH
2
1
2OH 2OH
2
2OH
2
*p ¼ 0.0102 ([N4 1 2OH
]
[C4H9PO3] - [N4 1 2OH
] [C2H5PO3], and ns - non-significant (as compared to values of the same IL with different percentages).
2OH 2
2OH
2

D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
2495
Fig. 7. Solubility of ibuprofen with ionic liquid [N4 1 2OH
]
[C2H5PO3] at 0.01 and 0.028 mol% at 37 ꢀC during 5 min.
2
2OH
results obtained for ibuprofen in aqueous solutions of dianionic ILs
are presented in Fig. 4. Using ionic liquids with the same anion,
[C4H9PO3]2-, it is possible to see that the insertion of additional
hydroxyethyl groups (-CH2CH2OH) in the cation moiety, does not
increase the solubility of ibuprofen. The same behaviour was
observed when the alkyl chain of the anion was changed, from C4 to
C1, using ILs with the same cation [N4 1 2OH 2OH]þ.
more than 3 times when the amount of IL rise from 0.05 mol% to
0.2 mol%. The solubility enhancement in IL aqueous solutions
(0.2 mol%) is over 250-fold when comparing with that in pure water.
With the monoanionic ionic liquids, the effect in the aqueous
solubility of piroxicam is very small, see Figure S3. In contrast,
major changes were observed, when using dianionic ionic liquids
and the results are presented in Fig. 5. For this drug, the changes in
the cation and/or anion had a distinct effect on the solubility. In the
cation moiety, the insertion of a second hydroxyethyl group, lead to
The solubility of ibuprofen increases with increasing concen-
tration of IL. For all ILs tested, there is an increase in solubility of
Figs. 8. 2D 1He1H NOESY spectrum of ibuprofen with 0.2 mol% of [N4 1 1 2OH
] [C4H9PO3] in DMSO‑d6 and representation of the interactions between ibuprofen and IL.
2

2496
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
Table 1
values are presented for both drugs in Fig. 6. The lactose does not
- Log P values for Ibuprofen and Piroxicam.
have any effect in solubility of both drugs. With sodium bicarbon-
ate, the solubility of piroxicam has a 7-fold increment when
comparing with that in pure water and with ibuprofen the incre-
ment in solubility was very high, but lower than the new dianionic
ionic liquids, which demonstrated the advantage in the use of these
new ILs.
Both ibuprofen and piroxicam are taken at 6e8 h intervals,
hence we have studied their solubility in selected ionic liquids
aqueous solutions after 6 h. After that period, it was observed that
the solubility was practically the same than after 24 h, see Figure S4.
The mean onset of action per 400 mg dose of ibuprofen is 45 min,
which can be critical in an acute pain situation where a fast relief is
critical.28 So, we decide to decrease the time in assays to 5 min and
the quantity of ionic liquid used, until 0.01 mol%.
System/Media
Log P
Ibuprofen/water
Piroxicam/water
3.9732
3.0633
Ibuprofen/water þ0.05 mol% [N4 1 2OH 2OH
]
[C2H5PO3]
0.60 0.05
0.57 0.01
0.33 0.06
0.29 0.02
2
Ibuprofen/water þ 0.05 mol% [N4 1 1 2OH
] [C4H9PO3]
2
Piroxicam/water þ 0.05 mol% [N4 1 2OH 2OH
] [C2H5PO3]
2
Piroxicam/water þ 0.05 mol% [N4 1 1 2OH
]2 [C4H9PO3]
an increase in the solubility, but the addition of a third hydrox-
yethyl group lead to a decrease. These effect could be explained by
our previous work,27 which shows that addition of hydroxyl groups
in the ammonium cation, enhances the hydrogen bonding ability of
ILs with water. When the third hydroxyethyl group is added, the
quantity of hydrogen bonds formed between IL and water is
harmful, for interaction of piroxicam with ionic liquid. Cþomparing
the results for ILs with the same cation ([N 4 1 2OH 2OH] ) we can
explore the effect of changes in the anion. In this case, the effect of
increasing the alkyl chain from methyl to ethyl had a positive effect
in solubility of piroxicam. However, increasing further the alkyl
chain length in the anion, from an ethyl group to a butyl one, did
not show the trend previously described and the solubility of pir-
oxicam presents similar values in both IL/aqueous solutions. For the
As can be seen in Fig. 7, with only 0.01 mol% and 0.028 mol% of
ionic liquid [N4 1 2OH 2OH]2 [C2H5PO3] at 5 min is possible to achieve
a solubility of z600 mg/L and z2400 mg/L corresponding to one
dose and maximum take daily, that an adult can take per day,
however the normal dose should be 1200 mg/L.
Two Dimensional 1He1H NOESY
A sample of ibuprofen with 0.2 mol% of [N4 1 1 2OH]2 [C4H9PO3]
dianionic IL [N4
possible to have a 480-fold improvement on the drug solubility in
comparison to pure water, and with [N4
0.3 mol%, an enhance of 530-fold.
] [C2H5PO3], with 0.25 mol% it was
1
2OH 2OH 2
was prepared in DMSO‑d6 to observe the interaction between the
two using a 2D 1He1H NOESY experiment. The molar concentra-
tions used mimic the conditions in the solubility assay. The 2D
1He1H NOESY spectrum obtained is showed in Fig. 8.
] [C4H9PO3]
1
2OH 2OH 2
The behaviour of some selected excipients that are used in the
market to enhance the solubility of ibuprofen and piroxicam (so-
dium bicarbonate and lactose) were herein studied and the results
are compared with the values obtained with the new dianionic
ionic liquids herein designed, with an amount of 0.2 mol%. The
The OH at
d 12.24 ppm from ibuprofen disappeared, which
clearly demonstrate the intermolecular interaction between the
oral drug and the ionic liquid, see Figure S15. It is also clear that
exists a molecular interaction between the proton of ibuprofen at
d
1.80 ppm and methyl group of phosphonate dianion at
Fig. 9. Concentration of Ibuprofen and Piroxicam in solution, in the presence or absence of IL along time, through PES-U membrane.

D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
2497
Fig. 10. Linearization of the permeation curves of Ibuprofen and Piroxicam with 0.05 mol% ILs.
d
0.87 ppm. These interactions may justify the improvement on the
again. According to literature these optimum log P values may
vary between 0 and 3.31 According to literature, ibuprofen has a
log P of 3.9732 and piroxicam has a log P of 3.0633 meaning that
are distributed to hydrophobic areas such as lipidic bilayers of
cells.
solubility of this drug in the aqueous solution of this IL.
Analogously, a NOESY experiment of a sample of piroxicam with
0.2 mol% of the dianionic IL [N4 1 1 2OH]2 [C4H9PO3] was performed
in CD3CN. The OH at
d 13.52 ppm from piroxicam disappeared,
suggesting the intermolecular interaction between the oral drug
and the ionic liquid, see Figure S16. There is also a molecular
We have determined the log P value of these drugs using ionic
liquids as co-solvents using the shake flask method. For each drug
we have selected the ionic liquids that promoted the highest sol-
ubility. Namely, we have selected [N4 1 2OH 2OH]2 [C2H5PO3] and [N4
interaction between the NH of piroxicam at
methylene group connected to atom from the cation
(NCH2CH2OH) at 3.34 ppm, which may justify the improvement
d 11.40 ppm and the
N
d
] [C4H9PO3] dianionic ionic liquids to calculate log P.
1 1 2OH 2
on the solubility of this drug in the aqueous solution of this IL, see
Figure S17.
Log P of ibuprofen and piroxicam in water were not determined
by this method, due to the fact that the separation of octanol-water
phases was not possible, therefore the values reported in the
literature were considered for comparison purposes. It is important
to highlight that in ionic liquid solutions we did not encounter the
same problems described for the determination of log P of
ibuprofen and piroxicam in water. The ability of both drugs to move
to the aqueous phase seems to be therefore, highly affected by the
presence of dianionic ionic liquids in solution. Log P for both drugs
with only 0.05 mol% was very similar for both dianionic ionic liq-
uids, as can be seen in Table 1.
In this study, the dianionic ionic liquids not only have an
incredible increment in the solubility of both drugs but also
meliorate its log P. In the presence of 0.05 mol% of ionic liquid both
log P of ibuprofen and piroxicam have values more hydrophilic. As
mentioned previously there must be a hydrophilicity/lipophilicity
balance to achieve an ideal drug to be orally administered.
Octanol-Water Partition Coefficient and Log P
In the context of pharmacokinetics, the partition coefficient
(log P) can be a useful parameter to estimate the distribution
profile of a drug within the body, which has a great influence in
its ADME (absorption, distribution, metabolism and excretion)
profile. It also helps scientists to determine the most likely way
of administration. It is relevant to notice that both drugs studied
in this work belong to class II according to the Biopharmaceutics
Classification System (BCS),29,30 low solubility and high perme-
ability. In the case of an orally absorbed drug, it has to permeate
the lipid bilayers in the intestinal epithelium. Therefore, it is
necessary a certain lipophilicity to partition into that bilayer but
also require some hydrophilicity so that will not partition out
Table 2
Permeability, Diffusion and Partition Coefficients Obtained for Oral Drugs With 0.05% ILs.
Permeability, P (10ꢃ5 cm/s)
Diffusion coefficient, D (10ꢃ6 cm2/s)
Partition coefficient, Kd (10ꢃ1
)
Ibuprofen
e
e
e
Ibuprofen þ [N4 1 1 2OH
]
[C4H9PO3]
] [C2H5PO3]
2
3.70 0.99
2.12 0.59
e
3.04 1.05
3.08 0.20
e
1.83 0.23
1.03 0.27
e
2
Ibuprofen þ [N4 1 2OH 2OH
Piroxicam
Piroxicam þ [N4 1 1 2OH
]
2 [C4H9PO3]
1.71 0.68
1.01 0.22
2.12 0.98
2.60 0.70
1.21 0.22
0.58 0.17
Piroxicam þ [N4 1 2OH 2OH
]
2
[C2H5PO3]

2498
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
Fig. 11. Cytotoxicity of the ionic liquids in fibroblasts L929 cells to different concentrations of the respective ILs. The results are expressed as the average SD from three inde-
pendent biological assays. The symbols *, **, *** and **** indicates that the viabilities are statistically significant with p ¼ 0.0152, p ¼ 0.0018, p ¼ 0.0002 and p ¼ 0.0001,
respectively, when compared to the control.
Permeability
The permeability of ibuprofen and piroxicam was studied with
0.05 mol% in aqueous solution. In absence of ILs it was not possible
to determine the permeability, diffusion coefficient and partition
the presence and in absence of IL along time. The concentration of
the drug that permeate the membrane (mg/L) vs time is repre-
sented in Fig. 9. It is evident the effect that these dianionic ionic
liquids have in the diffusion of ibuprofen and piroxicam over time
through the membrane polyethersulphone (PES-U). Different pa-
rameters influence the diffusion. In particular, the amount of the
solute permeated through a membrane depends on solute size,
membrane pores size, pH, temperature and the affinity between
membrane and the solute.34 A temperature of 37 ꢀC and a pH value
of 6 were maintained constant during the whole assay.
Table 3
- IC50 Determination of Mono and Dianionic Ionic Liquids and Drugs (Ibuprofen and
Piroxicam).
Ionic Liquid
IC50
[N4 1 2OH 2OH][C8H18PO4]
[N2O2 1 1 2OH][C8H18PO4]
[N4 2OH 2OH 2OH
[N4 1 2OH 2OH 2 [C4H9PO3]
[N4 1 1 2OH 2 [C4H9PO3]
[N4 1 2OH 2OH
[N4 1 2OH 2OH
Ibuprofen
70.2 5.6
111.6 3.7
72.6 3.4
54.5 2.3
93.6 5.5
119.9 5.2
86.3 5.2
2.5 0.3
]2 [C4H9PO3]
]
]
]
]
2 [C2H5PO3]
2 [CH3PO3]
The permeability of the oral drugs ibuprofen and piroxicam was
determined with the presence of dianionic ionic liquids [N4 1 2OH
Piroxicam
2.3 0.3
]
[C2H5PO3] and [N4
] [C4H9PO3], in a concentration
1 1 2OH 2
2OH 2

D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
2499
Fig. 12. Cytotoxicity of ibuprofen and piroxicam at 1 mM with some concentrations of [N4 1 2OH 2OH]2 [C2H5PO3] in fibroblasts L929 cells. The results are expressed as the
average SD from three independent biological assays. The symbols * and **** indicates that the viabilities are statistically significant with p ¼ 0.0124 and p ¼ 0.0001, respectively,
when compared to the control.
coefficient of drugs, since they not diffuse through membrane PES-
U, as can be seen in Fig. 9.
The permeability coefficient was determined in the linear part of
the permeation curve, up to 60% of the total drug permeated and as
can be seen in Fig. 10 for ibuprofen and piroxicam.
2OH]2 [C2H5PO3] and [N4 1 2OH 2OH]2 [CH3PO3] which have the same
cation with different anions we observed a decrease in IC50 with the
decrease of the alkyl side chain of the anion, from ethyl to methyl.
However, comparing ILs [N4 1 2OH
] [C4H9PO3] and [N4 1 2OH
2OH 2
]
[C2H5PO3], there is an increment in IC50 when the anion
2OH 2
A. R. Duarte et al.24 determined the permeability and diffusion
coefficient of ibuprofen in PBS solution, 4.6 0.14 ꢄ 10ꢃ5 cm/s and
2.39 0.43 ꢄ 10ꢃ6 cm2/s, respectively. The permeability obtained
for ibuprofen with both dianionic ILs were lower, as can be seen in
Table 2, but diffusion coefficient was similar. Ibuprofen and pirox-
icam are drugs that belong to BCS class II, being high permeable. It
would then be expected a different result in the permeability of
these drugs. However, permeability depends on factors such as the
medium and membrane, which may lead to completely different
results. The difference in values may be due to small changes in the
experimental procedure, such as the use of PBS in permeability
studies, which is not the case in this study.
changes from butyl to ethyl.
Interestingly, in this study the trend observed in other studies14
regarding the increasing of IL's lipophilicity with the increase of
toxicity, was not observed. In fact, in some cases we could not
observe a direct correlation between the side chain length and the
cytotoxicity. This may somehow suggest that dianionic ILs behave
in different way than the monoanionic ones.
The cytotoxicity of both drugs was also tested, in fibroblasts
L929 and the IC50 of ibuprofen is 2.5 mM and of piroxicam 2.3 mM,
see Table 3 and Figure S18.
It is not straightforward though that if the IL and the API are
non-toxic that the combination of the two renders a non-toxic
formulation. Therefore, the cytotoxicity of ibuprofen and pirox-
icam at 1 mM (a non-toxic concentration) with different concen-
trations (25, 100 and 150 mM) of dianionic ionic liquid [N4 1 2OH
Cytotoxicity Evaluation
]
[C2H5PO3] was tested. The percentage of cell viability with
2OH 2
For the cytotoxicity study, we have selected the five dianionic
25 mM of IL is identical to that obtained for the same concentration
of IL with 1 mM ibuprofen or piroxicam (Fig. 12). At 100 mM of IL
with 1 mM of ibuprofen or piroxicam, the viability of cells is close
50% or below (29%) in the case of piroxicam. At 150 mM of IL þ drug
(1 mM), the cytotoxicity was highly visible.
ILs, [N4 1 1 2OH 2
]
[C4H9PO3], [N4 1 2OH 2OH 2
[C2H5PO3] and [N4 ]
1 2OH 2OH 2
] [C4H9PO3], [N4 2OH 2OH
]
[C4H9PO3], [N4
]
2OH 2
1
2OH 2OH 2
[CH3PO3], which are those that showed the best results in terms of
solubility for both drugs ibuprofen and piroxicam. For comparison
purposes we have also chosen two monoanionic ILs, [N4 1 2OH 2OH
[C8H18PO4] and [N2O2 1 1 2OH][C8H18PO4].
]
According to results showed in Fig. 11 and in Table 3, mono and
dianionic ionic liquids presented low toxicity IC50 values ranging
from 54.5 mM up to 119.9 mM.
Comparing [N4 1 2OH 2OH][C8H18PO4] and [N2O2 1 1 2OH][C8H18PO4]
(both monoanionic with the same anion) it was possible to observe
that the introduction of an ether group at the alkyl side chain is able
to reduce significantly the cytotoxicity of these types of ILs. More-
over, the introduction of an N-hydroxyethyl group can lead to a
decrease of IC50 values, fact that can be observed in ILs [N4 1 2OH
Conclusions
The new dianionic ionic liquids synthesized and characterized
in this work, have an excellent ability to improve the solubility of
two poorly soluble oral drugs, ibuprofen and piroxicam, that belong
to class II of the BCS (poor aqueous solubility and high perme-
ability). Even at very small IL concentration (0.05e0.3 mol%) it was
possible to convert these drugs into a class I (high permeability and
high solubility). Also, at 0.05 mol% of IL it was possible to decrease
the lipophilicity (log P) of both drugs which is important to enhance
]
[C4H9PO3] and [N4 1 1 2OH]2 [C4H9PO3]. For the ILs [N4 1 2OH
2OH 2

2500
D.A.S. Agostinho et al. / Journal of Pharmaceutical Sciences 110 (2021) 2489-2500
phosphonium-hydrophobic ionic liquids. Phys Chem Chem Phys. 2015;17:4569-
4577.
10. Gomes JM, Silva SS, Reis RL. Biocompatible ionic liquids: fundamental behav-
iours and applications. Chem Soc Rev. 2019;48:4317-4335.
the pharmacodynamic and toxicological profile of both drugs. In
studies at 37 ꢀC, 5 min dissolution time and 0.01 mol% of [N4 1 2OH
]
[C2H5PO3], it was achieved a solubility of 600 mg/L, which
2OH 2
11. Gouveia W, Jorge TF, Martins S, et al. Toxicity of ionic liquids prepared from
biomaterials. Chemosphere. 2014;104:51-56.
12. Jordan A, Gathergood N. Biodegradation of ionic liquids e a critical review.
Chem Soc Rev. 2015;44:8200-8237.
13. Ventura SPM, e Silva FA, Gonçalves AMM, Pereira JL, Gonçalves F, Coutinho JAP.
Ecotoxicity analysis of cholinium-based ionic liquids to Vibrio fischeri marine
bacteria. Ecotoxicol Environ Saf. 2014;102:48-54.
14. Jesus AR, Soromenho MRC, Raposo LR, et al. Enhancement of water solubility of
poorly water-soluble drugs by new biocompatible N-acetyl amino acid N-alkyl
cholinium-based ionic liquids. Eur J Pharm Biopharm. 2019;137:227-232.
15. Jesus AR, Raposo LR, Soromenho MRC, et al. N-alkyl Cholinium Based Ionic
Liquids as Enhancers of Water Solubility for Poorly Water-Soluble Drugs, Unpubl.
Results. 2020.
corresponds to one dose for an adult. The dianionic ionic liquid [N4 1
[C2H5PO3] is the less toxic IL, having an IC50 of 120 mM
(z0.25 mol%). Furthermore, the new dianionic ILs synthesized are
not toxic to fibroblasts L929 cells. Thus, these new biocompatible
ionic liquids are promising vehicles for aqueous administration of
poorly water-soluble drugs.
]
2OH 2OH 2
Declarations of Interest
None.
16. Caparica R, Júlio A, Baby AR, et al. Choline-amino acid ionic liquids as green
functional excipients to enhance drug solubility. Pharmaceutics. 2018;10.
17. Júlio A, Antunes C, Mineiro R, et al. Influence of two choline-based ionic liquids
on the solubility of caffeine. Biomed Biopharm Res. 2018;15:96-102.
Acknowledgements
ꢀ
18. Garzon LC, F. Martínez Fleming. Temperature dependence of solubility for
~
^
The authors thank Fundaçao para a Ciencia e Tecnologia, FCT/
MCTES (Portugal) for financial support through an Investigator
contract (IF/00621/2015 e P.M. Reis) and (IF/01146/2015 - A. Paiva),
and through projects IF/00621/2015 and CryoDES (PTDC/EQU-EQU/
29851/2017). This project has received funding from the European
Research Council (ERC) under the European Union's Horizon 2020
research and innovation programme, under grant agreement No
ERC-2016-CoG 725034. This work was also partially supported by
the Associate Laboratory for Green Chemistry - LAQV which is
financed by national funds from FCT/MCTES (UIDB/50006/2020
and UIDP/50006/2020) The NMR spectrometers are part of The
National NMR Facility, supported by FCT/MCTES (RECI/BBB-BQB/
0230/2012).
ibuprofen in some organic and aqueous solvents. J Solut Chem. 2004;33:1379.
19. Potthast H, Dressman JB, Junginger HE, et al. Biowaiver monographs for im-
mediate release solid oral dosage forms: Ibuprofen. J Pharm Sci. 2005;94:2121-
2131.
20. Escandar G, Bystol A, Campiglia A. Spectrofluorimetric method for the deter-
mination of piroxicam and pyridoxine. Anal Chim Acta. 2002;466:275-283.
21. Shohin IE, Kulinich JI, Ramenskaya GV, et al. Biowaiver monographs for im-
mediate release solid oral dosage forms: piroxicam. J Pharm Sci. 2014;103:367-
377.
22. Mahamat Nor SB, Woi PM, Ng SH. Characterisation of ionic liquids nano-
emulsion loaded with piroxicam for drug delivery system. J Mol Liq. 2017;234:
30-39.
ꢀ
ꢀ
23. Baka E, Comer JEA, Takacs-Novak K. Study of equilibrium solubility measure-
ment by saturation shake-flask method using hydrochlorothiazide as model
compound. J Pharm Biomed Anal. 2008;46:335-341.
24. Duarte ARC, Ferreira ASD, Barreiros S, Cabrita E, Reis RL, Paiva A. A comparison
between pure active pharmaceutical ingredients and therapeutic deep eutectic
solvents: solubility and permeability studies. Eur J Pharm Biopharm. 2017;114:
296-304.
Appendix A. Supplementary Data
25. Silva JM, Duarte ARC, Caridade SG, Picart C, Reis RL, Mano JF. Tailored free-
standing multilayered membranes based on chitosan and alginate. Bio-
macromolecules. 2014;15:3817-3826.
26. Sintra TE, Shimizu K, Ventura SPM, Shimizu S, Canongia Lopes JN, Coutinho JAP.
Enhanced dissolution of ibuprofen using ionic liquids as catanionic hydro-
tropes. Phys Chem Chem Phys. 2018;20:2094-2103.
27. de Ferro AM, Reis PM, Soromenho MRC, et al. Designing the ammonium cation
to achieve a higher hydrophilicity of bistriflimide-based ionic liquids. Phys
Chem Chem Phys. 2018;20:19307-19313.
28. Santos MM, Raposo LR, Carrera GVSM, et al. Ionic liquids and salts from
ibuprofen as promising innovative formulations of an old drug. Chem Med
Chem. 2019;14:907-911.
29. García-Arieta A, Gordon J, Potthast H. On the Biopharmaceutics classification
system biowaiver of ibuprofen. J Pharm Sci. 2015;104:2429-2432.
30. Mirza S, Miroshnyk I, Habib MJ, Brausch JF, Hussain MD. Enhanced dissolution
and oral bioavailability of piroxicam formulations: modulating effect of phos-
pholipids. Pharm Times. 2010;2.
31. Pajouhesh H, Lenz GR. Medicinal chemical properties of successful central
nervous system drugs. NeuroRx. 2005;2:541-553.
32. Czyrski A. Determination of the lipophilicity of ibuprofen, naproxen, ketopro-
fen, and flurbiprofen with thin-layer chromatography. J Chemother. 2019;2019:
3407091.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.xphs.2021.01.014.
References
1. Krishnaiah YSR. Pharmaceutical technologies for enhancing oral bioavailability
of poorly soluble drugs. J Bioequiv Availab. 2010;2:28-36.
2. Horter D, Dressman JB. Influence of physicochemical properties on dissolution
of drugs in the gastrointestinal tract. Adv Drug Deliv Rev. 2001;46:75-87.
3. Savjani KT, Gajjar AK, Savjani JK. Drug solubility: importance and enhancement
techniques. ISRN Pharm. 2012;2012:195727.
€
4. Boyd BJ, Bergstrom CAS, Vinarov Z, et al. Successful oral delivery of poorly water-
soluble drugs both depends on the intraluminal behavior of drugs and of
appropriate advanced drug delivery systems. Eur J Pharm Sci. 2019;137:104967.
5. Adawiyah N, Moniruzzaman M, Hawatulaila S, Goto M. Ionic liquids as a po-
tential tool for drug delivery systems. Med Chem Commun. 2016;7:1881-1897.
6. Moniruzzaman M, Goto M. Ionic liquids: future solvents and reagents for
pharmaceuticals. J Chem Eng Jpn. 2011;44:370-381.
7. Buzzeo MC, Evans RG, Compton RG. Non-haloaluminate room-temperature
ionic liquids in electrochemistry–a review. Chem Phys Chem. 2004;5:1106-
1120.
8. Chiappe C, Pieraccini D. Ionic liquids: solvent properties and organic reactivity.
J Phys Org Chem. 2005;18:275-297.
9. Kurnia KA, V Quental M, Santos LMNBF, Freire MG, Coutinho JAP. Mutual sol-
ubilities between water and non-aromatic sulfonium-{,} ammonium- and
33. Shokri J, Azarmi S, Fasihi Z, Hallaj-Nezhadi S, Nokhodchi A, Javadzadeh Y. Ef-
fects of various penetration enhancers on percutaneous absorption of pirox-
icam from emulgels. Res Pharm Sci. 2012;7:225-234.
34. He H, Cao X, Lee LJ. Design of a novel hydrogel-based intelligent system for
controlled drug release. J Control Release. 2004;95:391-402.