Antimicrobial toxicity studies of ionic liquids leading to a 'hit' MRSA selective antibacterial imidazolium salt

Article abstract of DOI:10.1039/c2gc16090k

Imidazolium salts can be classed as surfactants, detergents, ionic liquids, reagents, catalysts or solvents. A study of the toxicity and ecotoxicity of these salts yields valuable information for their use as pharmaceuticals as well as impact on the environment. Our approach to screen a series of chiral imidazolium salts for toxicity to bacteria and fungi, including clinical pathogen strains, has led to the identification of a 'hit' MRSA selective antimicrobial compound. Preliminary structure-activity-relationship (SAR) information (required position of l-phenylalanine and l-valine group) is also elucidated within this first generation of compounds. Conversely, most of the imidazolium salts were nontoxic (IC₉₅ > 2 mM) to the 12 fungi strains and 8 bacteria strains screened, and we propose that they are suitable candidates for 'green chemistry' applications. Ecotoxicity studies (Biodegradation ISO 14593 'CO₂ Headspace Test') of two bromide ionic liquids containing l-phenylalanine residues indicate that these ionic liquids passed the test (>60% in 28 days) and classed as readily biodegradable.

Full text of DOI:10.1039/c2gc16090k

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2012, 14, 1350
www.rsc.org/greenchem
PAPER
Antimicrobial toxicity studies of ionic liquids leading to a ‘hit’ MRSA
selective antibacterial imidazolium salt†
Deborah Coleman,a Marcel Špulák,b M. Teresa Garciac and Nicholas Gathergood*a
Received 1st September 2011, Accepted 25th January 2012
DOI: 10.1039/c2gc16090k
Imidazolium salts can be classed as surfactants, detergents, ionic liquids, reagents, catalysts or solvents.
A study of the toxicity and ecotoxicity of these salts yields valuable information for their use as
pharmaceuticals as well as impact on the environment. Our approach to screen a series of chiral
imidazolium salts for toxicity to bacteria and fungi, including clinical pathogen strains, has led to the
identification of a ‘hit’ MRSA selective antimicrobial compound. Preliminary structure–activity-
relationship (SAR) information (required position of L-phenylalanine and L-valine group) is also
elucidated within this first generation of compounds. Conversely, most of the imidazolium salts were
nontoxic (IC95 > 2 mM) to the 12 fungi strains and 8 bacteria strains screened, and we propose that they
are suitable candidates for ‘green chemistry’ applications. Ecotoxicity studies (Biodegradation ISO 14593
‘CO2 Headspace Test’) of two bromide ionic liquids containing L-phenylalanine residues indicate that
these ionic liquids passed the test (>60% in 28 days) and classed as readily biodegradable.
chain, and demonstrated antimicrobial activity similar to that of
BAC. In a later study, the same research group reported the anti-
Introduction
Ionic liquids (ILs) have been described as molten salts that are
entirely ionic in nature, comprising both a cationic and anionic
species and by definition having a melting point below 100 °C.1
ILs are also known as ‘designer solvents’. The physical, chemi-
cal and biological properties of these salts can be ‘tuned’ for a
particular function. These reagents have been designed as repla-
cements for conventionally used organic solvents, and much
research has been reported on their applications in the area of
‘green’ chemistry.1–3
This has subsequently led ionic liquid researchers to assess
the various biological properties, namely toxicity,4–6 biostability
and biodegradability,7–11 of these ‘green’ solvents. The antimi-
crobial activities of a series of 3-alkoxymethyl-1-methylimidazo-
lium ionic liquids were evaluated by Pernak et al.12 Minimum
inhibitory concentrations (MIC) and minimum bactericidal con-
centrations (MBC) were reported for the ILs and also for ben-
zalkonium chloride (BAC), a commonly used ingredient in
biocides. The most potent IL contained a dodecyl alkoxy side
microbial activities of 1-alkylimidazolium and 1-alkoxymethyl
imidazolium lactate ILs against a range of antibiotic resistant
strains, including Methicillin-resistant Staphylococcus aureus
(MRSA).13 It was again noted that lipophilicity was a principal
factor in the antimicrobial activity of the ILs. ILs with undecyl
and dodecyl alkyl side chains proved most toxic towards the
tested microorganisms.
Gilmore and co-workers were the first researchers to study the
in vitro antibiofilm properties of ionic liquids.14 In this work, a
range of 1-alkyl-3-methylimidazolium chloride ionic liquids
were assessed for antibiofilm activity against clinically relevant
pathogens. It was found that for each strain the minimum biofilm
eradication concentration (MBEC)14 value decreased (increase in
antibiofilm activity) with an increase in the alkyl chain length of
the IL side chain. The decyl derivative displayed antibiofilm
activity against all Gram positive and most Gram negative bac-
teria tested, with tolerance observed against K. aerogenes and
B. cenopacia. The tetradecyl substituted IL exhibited a broad
spectrum of antibacterial activity, giving the greatest activity
against all biofilm strains. Gilmore further investigated the anti-
biofilm activities of 1-alkylquinolium bromide ILs against
pathogen biofilm microbes.15 The authors reported a similar
trend of higher antibiofilm potency with increased alkyl chain
length of the IL side chains.
aSchool of Chemical Sciences, Dublin City University, Glasnevin,
Dublin 9, Ireland. E-mail: Nick.Gathergood@dcu.ie; Fax: +353 1
7005503; Tel: +353 1 7007860
bDepartment of Inorganic and Organic Chemistry, Faculty of Pharmacy,
Charles University, Heyrovského 1203, CZ-500 03, Hradec Králové,
Czech Republic
cDepartment of Surfactant Technology, IQAC-CSIC, Jordi Girona 18-26,
08034 Barcelona, Spain
This toxic nature of some ionic liquids may be undesirable in
‘green’ chemistry applications, but this tunable property may be
exploited and used beneficially in the development of antimicro-
bials and pharmaceuticals. A number of groups have recently
demonstrated the potential biological applications of ionic
†Electronic supplementary information (ESI) available: Experi-
mental procedures (Synthesis, toxicity and biodegradation methods) and
characterisation of ionic liquids is available in the ESI. See
DOI: 10.1039/c2gc16090k
1350 | Green Chem., 2012, 14, 1350–1356
This journal is © The Royal Society of Chemistry 2012

View Online
liquids in drug delivery and as active pharmaceutical ingredients
(APIs). Rogers and co-workers prepared a room temperature
ionic liquid based on the common local anaesthetic Lidocaine.16
The IL was prepared by exchanging the hydrochloride anion
with a docusate ion. The ionic liquid form of the drug delivered
longer pain relief and a slow-release mechanism of drug deliv-
ery. A similar approach in achieving ionic liquid based APIs was
reported by MacFarlane’s group.17 An ionic liquid form of the
API propantheline bromide was prepared by changing the
bromide to an acesulfamate anion. The resulting IL, propanthe-
line acesulfamate, had no observable melting point with a glass
transition state at −20 °C. Antimicrobial ionic liquids have also
been applied as additives for medical devices. Dual functional
ionic liquids have been reported18 with plasticising effects on
poly(vinyl chloride) (widely used in the production of medical
devices) along with antibacterial activities. Contamination of
medical devices with resistant bacteria biofilms, could be treated
with these antibacterial ionic liquids.
The potential of ionic liquids as alternatives to ‘hard’ anti-
microbials may also be exploited. ‘Soft’ reagents are described
as biologically active compounds that are readily degraded into
nontoxic and benign chemicals in vivo, and in the environ-
ment.19,20 A series of quaternary ammonium compounds con-
taining labile ester and amide spacer groups, and long lipophilic
alkyl chains have been reported as soft antimicrobials.20 Ionic
liquids bearing similar structural groups (i.e. amide and ester
moieties, long alkyl chains) have been widely investigated in the
literature, for both toxic and biodegradable properties. The pres-
ence of ester8,9 functionalities and long alkyl chains21 in IL
cations has led to biodegradable examples.
Our research strategy involves a combination of environmental
toxicity, biodegradation and ecotoxicity studies of novel ionic
liquids, with potential medicinal and pharmaceutical appli-
cations. Resistant strains of bacteria (e.g. MRSA), are thus
included in the toxicity screening of novel ionic liquids. A series
of novel imidazolium based ionic liquids with chiral amino acid
ester and dipeptidyl moieties in the cation side chain was
achieved. The presence of an amide and ester moiety in the CIL
(chiral ionic liquid) side chain provides two possible sites for
enzymatic cleavage, we postulate leading to improved IL bio-
degradability. However, before comprehensive biodegradation
and ecotoxicity studies of these ionic liquids are undertaken, our
strategy is to determine their antimicrobial toxicity.
Fig. 1 General structure of amino acid based chiral ionic liquids.
Synthesis of the amino acid ester CILs was carried out in
three steps (see Scheme 1 and ESI†). The first step involved the
esterification of the C-terminus of the amino acids. The amino
acids were allowed to react with thionyl chloride in the presence
of alkyl alcohols at ambient temperatures. These amino acid
esters were reacted with bromoacetyl bromide to form the corre-
sponding α-bromoamide intermediates in good yields (Table 1,
40–82%). The amino acid ester was dissolved in DCM and
stirred at −78 °C in the presence of Et3N under inert atmospheric
conditions. Bromoacetyl bromide was then added and stirring
was continued for 4–5 h at −78 °C. Purification of the alkylating
intermediates was achieved via column chromatography (eluant
system 50 : 50 ethyl acetate : hexane). Subsequent alkylation of
1-methylimidazole by the α-bromoamides at −15 °C for 3 h,
then at 20 °C for 24 h, resulted in precipitation of the desired
bromide salt in good yields (Table 1, 84–99%).
A series of CILs with dipeptidyl side chains were also pre-
pared. The synthesis of these CILs involves a four step exper-
imental procedure (Scheme 2) where the dipeptide fragments
were prepared using conventional peptide coupling chemistry.
The synthesis involved treating the Boc protected amino acids
(i.e. N-tert-butyloxycarbonyl-L-alanine, L-valine and L-phenyl-
alanine) with 1-hydroxybenzotriazole (HOBt), N-(3-dimethyl-
aminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), and
triethylamine (Et3N) in dichloromethane at 0 °C in the presence
of various amino acid esters. Purification of these dipeptides was
achieved by washing with a dilute acid to remove the urea by-
product from the EDC coupling reagent, 1-{3-(dimethylamino)
propyl}-3 ethyl urea. Recrystallisation using ethyl acetate/pet-
roleum ether 40–60 °C was then employed. The next step of the
synthesis involved the removal of the Boc protecting group from
the dipeptide. This involved treating the Boc-dipeptide in 4 M
HCl in dioxane solution and allowing to stir for 4–5 h at room
temperature. Following removal of the Boc protecting group
from the dipeptidyl moiety, the free amine terminus can react
with bromoacetyl bromide to form the bromo-alkylating reagents
as described previously for the amino acid ester derivatives.
Column chromatography was employed to purify the dipeptidyl
bromoamide intermediates (13b–20b) (the mobile phase used
was a gradient system of 100% hexane to 50 : 50 hexane : ethyl
acetate), in good yields (Table 2, 67–84%). Once the dipeptidyl
alkylating intermediate was purified, N-alkylation of 1-methyl
imidazole was carried out. In this reaction THF was used as the
reaction solvent. To a stirring solution of 1-methyl imidazole in
THF, the bromo-alkylating reagents were added dropwise (at
−15 °C). Stirring at room temperature overnight furnished the
final product (i.e. the bromide salt) in good yields (Table 2,
75–98%).
Results
Synthesis
A range of novel imidazolium-based chiral ionic liquids (CILs)
were prepared in our laboratories using L-, D- and DL-amino acid
esters as chiral building blocks. Incorporation of these deriva-
tives provides the IL side chain with several possible biodegrada-
tion sites. Short synthetic routes, which are modular, are also an
aim for this project to facilitate library expansion. This, com-
bined with utilising sustainable chiral building blocks (amino
acids), further reinforces the role of green chemistry in the devel-
opment of these novel ionic liquids. Fig. 1 demonstrates the
general structure of these CILs.
Two reference ionic liquid compounds were also selected for
the toxicity screening, 3-methyl-1-(pentylcarbonylmethyl)imida-
zolium bromide (21) and 3-methyl-1-(dodecoxycarbonylmethyl)
imidazolium bromide (22).22
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1350–1356 | 1351

View Online
Scheme 1 Three step preparation of amino acid ester CILs (1c–12c)
Table 1 Amino acid esters (1a–12a) and corresponding yields for
α-bromoamide intermediates (1b–12b) and methyl imidazolium bromide
CILs (1c–12c).
yeasts (Candida krusei E28, Candida tropicalis 156, Candida
glabrata 20/I, Candida lusitaniae 2446/I, Trichosporon asahii
1188) and filamentous fungi (Aspergillus fumigatus 231, Absidia
corymbifera 272, Trichophyton mentagrophytes 445) The results
obtained for those dipeptidyl and amino acid ester CILs which
indicated inhibition (defined as 80% inhibition of the growth of
control for yeasts and as 50% inhibition of the growth of control
for filamentous fungi) are summarized in Tables 5 and 6 respect-
ively. Chiral ionic liquids (2c, 9c, 11c, and 18c) and achiral ionic
liquid (21) have MIC values >2 mM for all yeasts and fungi in
this study.
Methyl
Yield imidazolium Yield
Amino acid ester
α-Bromoamide (%)
IL
(%)
L-Phe-OEt·HCl 1a
L-Phe-OBu·HCl 2a
L-Val-OMe·HCl. 3a
L-Val-OEt·HCl 4a
L-Val-OBu·HCl 5a
L-Ala-OEt·HCl 6a
L-Ile-OBu·HCl 7a
DL-Val-OMe·HCl 8a 8b
D-Val-OMe·HCl 9a
D-Val-OEt·HCl 10a
1b
2b
3b
4b
5b
6b
7b
64
79
60
40
58
40
68
82
45
59
65
70
1c
2c
3c
4c
5c
6c
7c
8c
98
99
85
98
96
90
98
87
89
79
87
84
Dipeptidyl CILs 17c and 20c displayed the highest toxicity
towards the bacterial strains, inhibiting the growth of four iso-
lates, namely, Staphylococcus aureus, MRSA, Staphylococcus
epidermidis, and Enterococcus sp. The lowest MIC value
(highest potency) was obtained for 17c against the MRSA strain
(125 μM after 24 h incubation and 500 μM after 48 h). Some
notable trends can be observed from the results (Table 3). Struc-
turally, all four CILs (17c–20c) possess a phenylalanine residue
adjacent to the cation core. ILs bearing this structural motif may
have increased lipophilicities (cf. 13c and 15c), and thus
increased bioactivities with 20c, the IL expected to be most lipo-
philic, (cf. 17c–19c) the most toxic to Gram positive bacterial
isolates. However, greater inhibition (lowest MIC) of the resistant
MRSA strain was noted for CIL 17c. Therefore while lipophili-
city is an important parameter (as mentioned previously) in IL
toxicity, for our IL compounds more work is required to refine
the structure–activity-relationship (SAR). ILs 21 and 22 toxicity
is in agreement with previous antibacterial activity data.22 The
pentyl ester 21 was non-toxic to all bacteria (and fungi) strains
screened (upto 2 mM), while the dodecyl ester 22 is a potent
antimicrobial (Table 3). However, the results clearly show that
no selectivity for MRSA is found, and the broad spectrum
activity of 22 is in agreement with other ionic liquids containing
a long alkyl chain.14–16,19,20 This relationship between CIL struc-
ture and activity was not noted for the amino acid ester deriva-
tives (1c–12c). In the antibacterial studies of the CILs containing
a single peptide moiety in the cation side chain (1c–12c), a rela-
tively low bacterial potency was observed (≥2 mM). Therefore,
the addition of the second amino acid residue, specifically
9b
10b
9c
10c
11c
12c
D-Phe-OMe·HCl 11a 11b
D-Phe-OEt·HCl 12a 12b
Antimicrobial and antifungal toxicity studies
In vitro antibacterial activities23 of the novel chiral ionic liquids
(1c–20c), and achiral ionic liquids 21 and 22 were investigated
against a range of bacteria, including clinically resistant microbes
(ESI †). The microorganisms employed in these studies were
ATCC strains (Staphylococcus aureus ATCC 6538, Escherichia
coli ATCC 8739, Pseudomonas aeruginosa ATCC 9027) and
clinical isolates (Staphylococcus aureus MRSA HK5996/08,
Staphylococcus epidermidis HK6966/08, Enterococcus sp.
HK14365/08, Klebsiella pneumoniae HK11750/08, Klebsiella
pneumoniae ESBL HK14368/08). The results in Tables 3 and 4
represent the MIC values (IC95) obtained for the amino acid
ester and dipeptidyl CILs which displayed antibacterial inhi-
bition. Chiral ionic liquids (1c, 3c, 5c, 6c, 8c, 10c, 11c, 12c, 13c,
14c, 15c and 16c) and achiral ionic liquid (21) have MIC values
(IC95) > 2 mM for all 8 bacteria strains in this study.
In vitro antifungal activities24,25 of the chiral ionic liquids
(1c–20c), and achiral ionic liquids 21 and 22 were evaluated on
a panel of four ATCC strains (Candida albicans ATCC 44859,
Candida albicans ATCC 90028, Candida parapsilosis ATC C
22019, Candida krusei ATCC 6258) and eight clinical isolates of
1352 | Green Chem., 2012, 14, 1350–1356
This journal is © The Royal Society of Chemistry 2012

View Online
Scheme 2 Preparation of dipeptidyl CILs (13c–20c).
Table 2 Boc-dipeptide-OR (R = Me or Et) (13a–20a) and corresponding yields for α-bromoamides intermediates (13b–20b) and dipeptidyl methyl
imidazolium bromide CILs (13c–20c)
Boc-dipeptide
α-Bromoamide
Yield (%)
Methyl imidazolium IL
Yield (%)
Boc-L-Ala-L-Val-OMe 13a
Boc-L-Ala-L-Phe-OEt 14a
Boc-L-Val-L-Ala-OEt 15a
Boc-L-Val-L-Phe-OEt 16a
Boc-L-Phe-L-Val-OEt 17a
Boc-L-Phe-L-Leu-OMe 18a
Boc-L-Phe-L-Ala-OMe 19a
Boc-L-Phe-D-Phe-OEt 20a
13b
14b
15b
16b
17b
18b
19b
20b
67
76
84
77
72
74
73
69
13c
14c
15c
16c
17c
18c
19c
20c
97
84
95
91
96
75
97
98
Table 3 MIC (μM, IC95) values obtained for dipeptidyl CILs (17c, 18c, 19c and 20c) and 22
IL
Organism
Time (h)
17c
500, 1000
125, 500
500, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
18c
19c
20c
1000, 1000
2000, 2000
2000, >2000
2000, 2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
22
S. aureus ATCC 6538)
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
1000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
15.62, 15.62
15.62, 15.62
15.62, 31.25
31.25, 62.50
31.25, 62.50
31.25, 125
Methicillin-res.S.a. (HK5996/08)
S. epidermidis (HK6966/08)
Enterococcus sp. (HK14365/08)
E. coli (ATCC 8739)
K. pneumoniae (HK11750/08)
K. pneumoniae-ESBL (HK14368/08)
P. aeruginosa (ATCC 9027)
31.25, 125
31.25, 125
D-valine (within the scope of this study), is required for MRSA
selective antimicrobial toxicity to be realised. 13c, the ionic
liquid with a D-Val–D-Val sequence, exhibited no antibacterial
toxicity (up to the maximum concentration screened, 2 mM) for
all 8 bacteria in the study. This further supports the requirement
of the location of the D-phenylalanine group in 17c. While
detailed discussion of the pharmacophore would be premature,
consideration of host–guest interactions based on hydrogen
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1350–1356 | 1353

View Online
Table 4 MIC (μM, IC95) values obtained for amino acid ester CILs (2c, 4c, 7c, 9c and 11c)
IL
Organism
Time (h)
2c
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
4c
7c
9c
11c
S. aureus (ATCC 6538)
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
Methicillin-res.S.a. (HK5996/08)
S. epidermidis (HK6966/08)
Enterococcus sp. (HK14365/08)
E. coli (ATCC 8739)
K. pneumoniae (HK11750/08)
K. pneumoniae-ESBL (HK14368/08)
P. aeruginosa (ATCC 9027)
bonding, van der Waals forces, ionic bonding and π–π stacking
are fundamental for drug development.
dioxide) was determined by measuring the net increase in total
inorganic carbon (TIC) levels over time compared to unamended
blanks. Sodium n-dodecyl sulfate (SDS) was used as a reference
substance. The test ran for 28 days.
A trend can also be noted between CIL activity and the bac-
terial strains tested. The CILs displayed increased toxicity
towards the Gram positive type bacteria (namely Staphylococcus
aureus, Staphylococcus aureus-MRSA, Staphylococcus epider-
midis, and Enterococcus sp.), whilst the Gram negative species
appeared the most tolerant towards the test compounds. This
result may be understood by considering the different structural
features present in both types of bacteria. Gram positive bacterial
cell walls lack the outer membrane found in Gram negative
species. Certain components of the Gram negative bacterial cell
well (e.g. lipopolysaccharide layer) can protect the bacteria from
various antibiotics and chemical entities.
Moderate antifungal activity was noted for the dipeptidyl
CILs, with inhibition observed at high MIC values (low toxi-
city). Furthermore, in terms of SAR studies, no strong trends can
be deduced from the obtained data. Amino acid ester CILs
appeared more toxic towards the tested fungal isolates. CIL 7c
proved to be the most active of the ILs tested, inhibiting six
fungal strains (Table 6), Candida albicans (ATCC 44859),
Candida parapsilosis (ATCC 22019), Candida tropicalis (156),
Trichosporon asahii (1188), Absidia corymbifera (272), Tricho-
pyton mentagrophytes (445). The lowest MIC value obtained for
7c was at 500 μM against Candida albicans (ATCC44859) after
24 h incubation, and further inhibited growth after 48 h (MIC
2000 μM). However 5c gave the lowest measured MIC value
(highest toxicity) of 62.50 μM against Trichopyton mentagro-
phytes (445) even after 48 h incubation.
Biodegradation data. Results for biodegradation of ionic
liquids containing a single phenylalanine residue (2c) and two
phenylalanine residues (18c) were obtained at weekly intervals
over 28 days and are shown in Tables 7 and 8 respectively. 2c
gave 61% biodegradation over 28 days (95% confidence limit
55–66%, based on 4 replicates) while 18c gave 64% biodegrada-
tion over 28 days (95% confidence limit 59–68%, based on 4
replicates). These bromide salts pass the “CO2 Headspace” test
(ISO 14593) (>60% in 28 days) and classed as readily bio-
degradable28 under aerobic conditions. Both 2c and 18c were
non-toxic in the biodegradation assay. When compared to bio-
degradation data for other imidazolium bromide ionic liquids
containing an amide8 (<5% in the Closed Bottle Test
(OECD301D)28), significant progress to identify amide contain-
ing ionic liquids which biodegrade has been demonstrated.
Biodegradation studies are on-going with the aim of linking
antimicrobial toxicity to ecotoxicity studies. The goal is to
develop an MRSA selective antibacterial drug, which readily
biodegrades.
Conclusions
A series of novel imidazolium chiral ionic liquids, containing
amino acid ester (1c–12c) and dipeptidyl (13c–20c) functional-
ities, have been reported. Antimicrobial toxicity studies incorpor-
ating drug resistant bacterial and fungal strains allowed for
simultaneous screening of toxicity in the environment and medi-
cinal properties of the synthesised CILs. A medicinal ‘hit’ was
observed against an antibiotic resistant strain of Methicillin-
resistant Staphylococcus aureus (MRSA). CIL 17c displayed
an inhibitory effect on this isolate at low MIC values (corre-
sponding to a high level of toxicity). After 48 h the MIC value
had increased by a factor of 4, which we propose is due to the
breakdown of 17c into less toxic metabolites. This CIL also has
low antifungal toxicities (MIC values of >2000 μM against
Candida albicans ATCC 44859, Candida albicans ATCC 90028
and Candida lusitaniae 2446/I), low antibacterial toxicities
(MIC values of ≥2000 μM against Enterococcus sp. HK14365/
08, Klebsiella pneumoniae HK11750/08, Klebsiella pneumoniae
ESBL HK14368/08, Escherichia coli ATCC 8739 and
Biodegradation studies
CO2 Headspace test. To evaluate the biodegradability of the
ionic liquids, the “CO2 Headspace” test (ISO 14593)26 was
implemented.27 This method allows the evaluation of the ulti-
mate aerobic biodegradability of an organic compound in an
aqueous medium at a given concentration of microorganisms by
analysis of inorganic carbon. The test ionic liquid, as the only
source of carbon and energy, was added at a concentration of
20 mg C L−1 to a buffer–mineral salts medium which had been
inoculated with a mixed population of microorganisms derived
from activated sludge collected from a sewage treatment plant.
These solutions were incubated in sealed vessels with a head-
space of air, which provided a reservoir of oxygen for aerobic
biodegradation. Biodegradation (mineralization to carbon
1354 | Green Chem., 2012, 14, 1350–1356
This journal is © The Royal Society of Chemistry 2012

Table 5 Antifungal activitiesa of dipeptidyl CILs (13c, 14c, 15c, 16c, 17c, 19c, 20c) and 22
IL
Organism
Time (h)
13c
14c
15c
16c
17c
2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
19c
20c
22
7.81, 31.25
7.81, 31.25
31.25, 62.50
1.95, 3.90
1.95, 3.90
3.90, 7.81
1.95, 3.90
C. albicans (ATCC44859)b
C. albicans (ATCC 90028)b
C. parapsilosis (ATCC 22019)b
C. krusei (ATCC 6258)b
C. krusei (E28)b
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
72, 120
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
2000, >2000
>2000, >2000
C. tropicalis (156)b
C. glabrata (20/I)b
C. lustaniae (2446/I)b
T. asahii (1188)b
31.25, 62.50
15.62, 31.25
15.62, 31.25
A. fumigatus (231)c
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
>2000, >2000
A. corymbifera (272)c
T. mentagrophytes (445)c
>2000, >2000 >2000, >2000
>2000, >2000
>2000, >2000
15.62, 31.25
a Reported as MIC (μM, bIC80 or cIC50) values.
Table 6 Antifungal activitiesa of amino acid ester CILs
IL
Organism
Time (h) 1c
3c
4c
5c
6c
7c
8c
10c
12c
C. albicans (ATCC 44859)b
C. albicans (ATCC 90028)b
C. parapsilosis (ATCC 22019)b
C. krusei (ATCC 6258)b
C. krusei (E28)b
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
24, 48
72, 120
2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
>2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
>2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
500, 2000 >2000, >2000 >2000, >2000 >2000, >2000
C. tropicalis (156)b
>2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
C. glabrata (20/I)b
>2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
>2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
C. lustaniae (2446/I)b
T. asahii (1188)b
>2000, >2000 >2000, >2000
>2000, >2000 >2000, 2000
>2000, >2000 >2000, >2000 >2000, >2000
1000, 2000 1000, 1000 62.5, 62.5
500, 500 >2000, >2000 >2000, >2000
500, 1000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000 >2000, >2000
1000, >2000
2000, >2000 >2000, >2000 >2000, >2000
A. fumigatus (231)c
A. corymbifera (272)c
T. mentagrophytes (445)c
250, 250
>2000, >2000
2000, 2000
500, 500
1000, >2000 >2000, >2000
1000, 2000 250, 250
2000, >2000
1000, >2000
2000, >2000
1000, >2000
a Reported as MIC (μM, bIC80 or cIC50) values.

View Online
Table 7 Biodegradation data obtained for amide CILs 2c vs. SDS
reference
PHD-ET-7). The antibacterial and antifungal screening was sup-
ported by the Czech Science Foundation (project No. P207/10/
2048) (MS). Biodegradation studies supported by financial
support of the Spanish Ministry of Science and Innovation
(project CTQ2010-17990) (MTG).
% Biodegradation
Compound
7 d
14 d
21 d
28 d
95% aCI
SDS
2c
78
16
89
59
91
61
94
61
89–99
55–66
IL initial concentration = 20 mg C L−1
calculated from 4 replicates.
,
a confidence limits were
References
1 R. D. Rogers and K. R. Seddon, Ionic Liquids; Industrial Applications to
Green Chemistry, American Chemical Society, Washington, DC, 2002,
pp. 1–492.
2 Ionic Liquids in Synthesis, ed. P. Wasserschid and T. Welton Wiley-VCH,
Weinheim, 2003, pp. 1–380.
3 P. Wasserscheid, R. Van Hal and A. Bösmann, Green Chem., 2002, 4,
400–404.
4 J. Pernak, I. Goc and I. Mirska, Green Chem., 2004, 6, 323–329.
5 M. Rebros, H. Q. Nimal Gunaratne, J. Ferguson, K. R. Seddon and
G. Stephens, Green Chem., 2009, 11, 402–408.
Table 8 Biodegradation data obtained for amide CILs 18c vs. SDS
reference
% Biodegradation
Compound
6 d
13 d
20 d
28 d
95% aCI
6 D. Zhao, Y. Liao and Z. Zhang, Clean: Soil, Air, Water, 2007, 35, 42–48.
7 N. Gathergood and P. J. Scammells, Aust. J. Chem., 2002, 55, 557–560.
8 N. Gathergood, M. T. Garcia and P. J. Scammells, Green Chem., 2004, 6,
166–175.
9 N. Gathergood, M. T. Garcia and P. J. Scammells, Green Chem., 2006, 8,
156–160.
SDS
18c
67
42
91
65
91
68
81
64
76–88
59–68
IL initial concentration = 20 mg C L−1
calculated from 4 replicates.
,
a confidence limits were
10 S. Stolte, S. Abdulkarim, J. Arning, A. Blomeyer-Nienstedt, U. Bottin-
Weber, M. Matzke, J. Ranke, B. Jastorff and J. Thoeming, Green Chem.,
2008, 10, 214–224.
11 D. Coleman and N. Gathergood, Chem. Soc. Rev., 2010, 39, 600–637.
12 J. Pernak, K. Sobaszkiewicz and I. Mirska, Green Chem., 2003, 5, 52–
56.
13 J. Pernak, I. Goc and I. Mirska, Green Chem., 2004, 6, 323–329.
14 L. Carson, P. K. W. Chau, M. J. Earle, M. A. Gilea, B. F. Gilmore, S.
P. Gorman, M. T. McCann and K. R. Seddon, Green Chem., 2009, 11,
492–497.
15 A. Busetti, D. E. Crawford, M. E. Earle, M. A. Gilea, B. F. Gilmore, S.
P. Gorman, G. Laverty, A. F. Lowry, M. McLaughlin and K. R. Seddon,
Green Chem., 2010, 12, 420–425.
16 W. L. Hough, M. Smiglak, H. Rodriguez, R. P. Swatloski, S. K. Spear, D.
T. Daly, J. Pernak, J. E. Grisel, R. D. Carliss, M. D. Soutullo, J. H. Davis
and R. D. Rogers, New J. Chem., 2007, 31, 1429.
17 P. M. Dean, J. Turanjanin, M. Yoshizawa-Fujita, D. R. Mac Farlane and J.
L. Scott, Cryst. Growth Des., 2009, 9, 1137–1145.
18 S. Y. Choi, H. Rodriguez, A. Mirjafari, D. F. Gilpin, S. McGrath, K.
R. Malcolm, M. M. M. Tummey, R. D. Rogers and T. Mc Nally, Green
Chem., 2011, 13, 1527–1535.
Pseudomonas aeruginosa ATCC 9027), with selectivity for
MRSA inhibition established. 18c, 19c and 20c also demon-
strated antibacterial behaviour against the test species. A relation-
ship between these CIL structures and their resulting
bioactivities (SAR) has been highlighted. The presence of the
lipophilic, aromatic phenylalanine unit in the ILs (17c–20c)
dipeptidyl side chain results in active antibacterials. However,
20c, with the Phe-Phe sequence was less active, with no MRSA
selectivity (cf. 17c). The relationship between antimicrobial tox-
icity and ecotoxicity (including biodegradation studies) is
complex. Antimicrobial toxicity determines toxicity to a single
strain, while an ecotoxicity test determines toxicity to a commu-
nity of organisms. We postulate that an MRSA selective antibac-
terial ionic liquid (17c) will be less toxic in an ecotoxicity test
than a proven antimicrobial biocide (22). Biodegradation studies
of 2c and 18c show that these two ionic liquids ‘pass’ the ‘CO2
Headspace Test’ and are ‘readily biodegradable’. As low toxicity
in an ecotoxicity (e.g. biodegradation) test, does not ensure bio-
degradation, persistence of recalcitrant parent compounds or
breakdown products must also be considered. Studies are on-
going to establish the biodegradation properties of the class of
ionic liquids disclosed herein. Our overall aim is to develop
MRSA selective drugs which are readily biodegradable and
avoid bioaccumulation in the environment. Our combined med-
icinal and green chemistry research strategy was successful in
discovering an antimicrobial ‘hit’. Further work is required to
determine the ‘lead’ compound and to define the
pharmacophore.
19 N. Nodor, J. Med. Chem., 1980, 23, 566–569.
20 T. Thorsteinsson, M. Másson, K. G. Kristinsson, M. A. Hjálmarsdóttir,
H. Hilmarsson and T. Loftsson, J. Med. Chem., 2003, 46, 4173–4181.
21 K. M. Docherty, J. K. Dixon and C. F. Kulpa, Biodegradation, 2007, 18,
481–493.
22 S. Morrissey, B. Pegot, D. Coleman, M. T. Garcia, D. Ferguson,
B. Quilty and N. Gathergood, Green Chem., 2009, 11, 475–483.
23 Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that
Grow Aerobically. Approved Standard – Seventh Edition. Document
M07-A7. Clinical Laboratory Standard Institute, Wayne, PA, 2006.
24 Reference Method for Broth Dilution Antifungal Susceptibility Testing of
Yeasts. Approved standard. Document M27-A3. Clinical Laboratory
Standard Institute, Wayne, PA, 2008.
25 Reference Method for Broth Dilution Antifungal Susceptibility Testing of
Filamentous Fungi. Approved standard. Document M38-A2. Clinical
Laboratory Standard Institute, Wayne, PA, 2008.
26 ISO 14593: Water quality, Evaluation of ultimate aerobic biodegradability
of organic compounds in aqueous medium. Method by analysis of
inorganic carbon in sealed vessels CO2, headspace test.
27 For previous work by M. Teresa Garcia using the ISO 14593 test, see ref.
8, 9, 11, 22.
Acknowledgements
The authors (NG, DC) would like to thank the Environmental
Protection Agency in Ireland for funding (EPA 2006-
28 OECD, Ready Biodegradability-CO2 in Sealed Vessels (Headspace Test
and Closed Bottle Test), in Guidelines for Testing of Chemicals, 310, 2006.
1356 | Green Chem., 2012, 14, 1350–1356
This journal is © The Royal Society of Chemistry 2012