Affinity ionic liquid

Article abstract of DOI:10.1039/b916411a

An affinity ionic liquid, based on biomolecular recognition, was developed and found to be capable of quantitative partitioning of biomacromolecules from aqueous buffer to ionic liquid.

Full text of DOI:10.1039/b916411a

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Affinity ionic liquidw
Ming-Chung Tseng,a Min-Jen Tsengb and Yen-Ho Chu*a
Received (in Cambridge, UK) 10th August 2009, Accepted 20th October 2009
First published as an Advance Article on the web 9th November 2009
DOI: 10.1039/b916411a
An affinity ionic liquid, based on biomolecular recognition, was
developed and found to be capable of quantitative partitioning of
biomacromolecules from aqueous buffer to ionic liquid.
demonstrate the technology. As a popular reporter of gene
expression, GFP is a well known protein routinely used in
molecular biology as the proof-of-concept that a gene can be
expressed throughout a given organism.5 The AIL 2 was
readily synthesized by standard organic chemistry protocols.
The synthesis of AIL 2 is outlined in Scheme 1. Based on
our previous report on the synthesis of a series of new bicyclic
imidazolium ionic liquids that are far more chemically
stable than the common [bmim][PF6] and [bdmim][PF6],6
it prompted us to incorporate this structural core for the
development of AILs. In short, the synthesis of AIL 2 began
with commercially available 4-chlorobutyronitrile 3 and,
having the imidate 4 as the intermediate compound, the key
This paper reports the use of an affinity ionic liquid (AIL) for
the study of biomolecular recognition in pure ionic liquid.
Ionic liquids have attracted much attention in recent years and
carry many desirable properties, such as wide liquid range,
thermal stability, high ionic conductivity, enhanced solubility
with many organic compounds, attractive recyclability, very
low flammability, and negligible vapor pressure, that are well
suited for numerous innovative applications, including high
temperature lubricants, re-writeable image surfaces, and reaction
media for synthesis and catalysis, and novel electrolytes for
solar and fuel cells.1 Ionic liquids are also considered as
environmentally friendly substitutes to conventional organic
solvents and often outperform molecular solvents in chemical
reactions and applications.1 Recently, significant progress has
been made to explore functionalized ionic liquids through
incorporation of functional groups as a part of developing
ionic liquids possessing tailored properties.2 Here, we are
reporting our initial development of AILs for use to investi-
gate biomolecular interactions in ionic liquids, with a direct
and immediate aim to facilitate the one-step affinity isolation
and purification of recombinant fusion proteins in bacterial
cells such as E. coli.
bicyclic imidazole
5 was prepared using the protocol
previously developed by us (59% isolated yield in three steps)
(Scheme 1).6 A direct alkylation of the resulting imidazole 5
with benzyl 6-bromohexanoate under heated conditions
followed by deprotection using catalytic hydrogenation and
subsequently an ion exchange in water readily afford the
precursor ionic liquid acid 6 with a high 93% isolated yield
in three steps. Finally, the EDC coupling of 6 with a
protected NTA amine 7 in dichloromethane followed by the
deprotection using catalytic hydrogenation optimally achieved
the preparation of AIL 2 as a viscous liquid at room temperature
(76% yield in 2 steps). In our hands, the overall isolated yield
of AIL 2 synthesis was high: 42% with a total of eight steps
starting from 3 (Scheme 1). Details of the synthetic procedures
are summarized in the ESI.w
Ionic liquids possess remarkable solubility with many small
molecules and, in the literature, have shown their superior
applicability toward extraction of metal ions and amino
acids.3 Moreover, ionic liquids are polar ionic solvents and
have been demonstrated as excellent media for peptide synthesis.4
In this paper, we illustrate our AIL methodology using a
fluorescein-labeled hexahistidine peptide FITC-(His)6-NH2 (1),
a His-tag green fluorescent protein (GFP) and an AIL 2 as the
model system. This system of biomolecular recognition has
Results in Fig. 1 clearly show that, upon association with
Ni(II), the AIL 2 dissolved in [bdmim][NTf2] ionic liquid (8)
been extensively studied using other techniques and
a
substantial amount of literature concerning its interaction is
available (footnote 1, ESIw). Peptide 1 could be prepared in a
straightforward manner by solid-phase peptide synthesis using
the standard Fmoc chemistry. In this work, the recombinant
His-tag GFP was used as a model protein with which to
a Department of Chemistry and Biochemistry, National Chung Cheng
University, Chia-Yi 621, Taiwan, ROC. E-mail: cheyhc@ccu.edu.tw;
Fax: +886 5 2721040; Tel: +886 5 2428148
b Institute of Molecular Biology, National Chung Cheng University,
Chia-Yi 621, Taiwan, ROC
w Electronic supplementary information (ESI) available: Synthesis,
NMR spectra and spectral data of AIL 2; experimental procedures
and characterization of affinity extractions of AIL 2 with the peptide 1
and His-tag GFP (14 pages). See DOI: 10.1039/b916411a
Scheme 1 Synthesis of AIL 2. Experimental conditions: (a) HCl,
MeOH, ether, À20 1C, 24 h; (b) H2NCH2CH(OMe)2, Et3N, CH2Cl2,
reflux, 2 h; (c) HCO2H, reflux, 20 h; (d) Br(CH2)5CO2Bn, 80 1C, 2 h;
(e) H2, Pd(OH)2/C, MeOH, rt, 1 h; (f) LiNTf2, H2O, rt, 12 h; (g) EDC,
DIEA, CH2Cl2, rt, 5 h; (h) H2, Pd(OH)2/C, MeOH, rt, 1 h.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7503–7505 | 7503

View Online
specifically recognized the fluorescent peptide 1 in PBS buffer
(upper layer) and, as a result, effectively extracted it into the
ionic liquid phase (bottom layer). In these affinity extraction
experiments, the ionic liquid phase was prepared by dissolving
AIL 2 in ionic liquid 8 and the aqueous solution was
separately prepared by dissolving peptide 1 in PBS buffer.
Equal volumes of the ionic liquid 8 and buffer solution were
mixed and the resulting mixture was then gently shaken at
ambient temperature to reach the completion of extraction. As
shown in Fig. 1 (tube a), without AIL 2 the bright green
fluorescence remained in the upper layer, indicating that the
peptide 1 was retained in the PBS buffer. Also, in the absence
of Ni(II) cation, the partitioning of the peptide 1 into ionic
liquid phase containing AIL 2 was negligible (tube b). Further-
more, the affinity extraction experiment performed in tube c
(Fig. 1) unambiguously proved that the transfer of the peptide
1 from the aqueous phase into the ionic liquid 8 was solely due
to affinity recognition by the Ni(II)-chelated AIL 2. This
quantitative extraction of peptide 1 was also apparent to the
naked eye, since the bright green fluorescence of the peptide
visually indicated its distribution between two layers (Fig. 1B).
It is worth mentioning that no precipitation of peptide at the
ionic liquid–water interface was observed; that is, 1 was
completely soluble in the ionic liquid 8. Fig. 1 also demon-
strated that the total recovery of the peptide 1 from ionic
liquid phase into the buffer layer could be readily achieved by
competitive binding with AIL 2 using a high concentration of
imidazole (tube d).
measurement further verified that the steady-state fluorescence
of the fluorophore FITC itself in ionic liquid 8 was essentially
quenched (Fig. S1B, ESIw). Therefore, it appears likely that,
once in the ionic liquid phase, the close contact between the
fluorophore and the ionic liquid molecules result in an efficient
static fluorescence quenching (tube c in Fig. 1B).
For a receptor–ligand interaction, the stoichiometry of the
binding is valuable to measure the concentration of active
receptor and to probe ligand specificity. To determine the
stoichiometry of our binding system, samples having a fixed
concentration of the AIL 2 in ionic liquid 8 and various
concentrations of the Fmoc-(His)6-NH2 peptide (9) in buffer
were prepared to perform affinity extractions. Results in Fig. 2
showed that, as the ratio of [9]total/[Ni(II)-chelated AIL 2]
increased, the concentration of 9 extracted into the ionic liquid
phase increased (i.e., peptide was transferred to the ionic liquid
and its concentration in buffer phase was minimal) until the
ratio reached its binding stoichiometry; beyond this value no
more peptide 9 could be extracted because all Ni(II)-chelated
AIL 2 in the ionic liquid phase has been fully titrated and
associated with 9, and therefore excessive 9 remained unbound
in the buffer layer. An abrupt change in the slope, in a plot of
the concentration of free peptide 9 in buffer phase vs. the ratio
of [9]total/[Ni(II)-chelated AIL 2] in samples, corresponded to
the binding stoichiometry of the system studied; that is, an
expected 1 : 1 stoichiometry of this model binding system was
obtained (Fig. 2). Furthermore, a similar experiment of the
[9]buffer vs. the ratio of [9]total/[Ni(II)] clearly indicated that,
with Ni(II) alone but without AIL 2, the peptide 9 was mostly
retained in aqueous phase, suggesting that both AIL 2 and
Ni(II) are required for efficient extraction of peptide 9 into the
ionic liquid phase (Fig. S2, ESIw). Overall, we were notably
pleased that the Ni(II)-chelated AIL 2 not only dissolved and
delivered the peptide 9 into ionic liquid 8, but also maintained
its binding integrity (to form a stable 1 : 1 complex) and high
specificity to peptide in pure ionic liquid.
For the result shown in tube c of Fig. 1, we noted that the
fluorescence of peptide 1 in ionic liquid phase was diminished
completely. This experimental observation is in agreement
with previous literature reports that ionic liquids could act
as quenchers for fluorescence probes.7 We have also experi-
mentally confirmed by UV-vis spectroscopy that, upon affinity
extraction by the Ni(II)-chelated AIL 2, the peptide 1 parti-
tioned entirely in the bottom ionic liquid layer and underwent
a major blue shift with significant decrease in absorption
intensity in its spectrum: lmax = 495 nm in buffer vs. 445 nm
in ionic liquid 8 (Fig. S1A, ESIw). Moreover, the fluorescence
Encouraged as we were by the peptide result, we then
turned our attention to the green fluorescent protein (GFP),
a powerful imaging tool, of the jellyfish Aequorea aequorea5
and investigated it as a model protein with which to develop
an effective platform for the one-step affinity isolation and
Fig. 1 Extraction of the fluorescent peptide 1 in PBS buffer (upper
layer) by AIL 2 solubilized in ionic liquid 8 (bottom layer). (A) and (B)
are optical and fluorescence photographic images, respectively, of
peptide extraction into ionic liquid: (a) extraction of peptide 1
(100 mM) in PBS buffer into ionic liquid 8 containing no AIL 2;
(b) extraction of peptide 1 (100 mM) in PBS buffer into ionic liquid 8
having AIL 2 (20 mM) but without NiCl2; (c) extraction of peptide 1
(100 mM) in PBS buffer layer into ionic liquid layer containing
Ni(II)-chelated 2 (20 mM); (d) competitive extraction of peptide 1
(100 mM) in ionic liquid layer into the PBS buffer layer using imidazole
(500 mM). Note that fluorescence of the peptide 1 was, in (c),
quenched in ionic liquid (bottom layer) and, in (d), re-established in
aqueous buffer (upper layer).
Fig. 2 Determination of the binding stoichiometry of the Ni(II)-
chelated AIL 2 (1.00 mM) dissolved in ionic liquid 8 to peptide 9
(0.067–1.33 mM) in PBS buffer. The Fmoc-(Gly)6-NH2 peptide 10 was
used as the internal standard in buffer. After each extraction, the
molar ratios of [9]/[10] in the aqueous layer were quantitatively
determined at 280 nm by C18-HPLC. Peptides 9 and 10 were used in
this experiment, primarily for the reason that they could be readily
separated and quantified by HPLC.
ꢀc
This journal is The Royal Society of Chemistry 2009
7504 | Chem. Commun., 2009, 7503–7505

View Online
Fig. 4 Direct affinity extraction of the His-tag GFP from cell lysate:
(A) before extraction and (B) after extraction. The upper layer was
aqueous phase containing cell lysate (5 mL) in PBS (25 mL) and the
bottom layer was Ni(II)-chelated AIL 2 (89 mM) solubilized in ionic
liquid 8 (30 mL). Note that, after extraction, GFP remains fluorescent
in the suspended ionic liquid layer.
Fig. 3 SDS-polyacrylamide gel electrophoresis of direct extraction of
the His-tag GFP in crude E. coli cell lysate by the Ni(II)-chelated AIL
2. Lane 1, protein molecular weight markers (170, 130, 100, 72, 55, 40,
33, 24, 17, and 11 kDa); lane 2, total cell lysate; lane 3, the cell lysate
after affinity extraction by Ni(II)-chelated AIL 2; lane 4, competitive
extraction of His-tag GFP in ionic liquid into aqueous buffer by
imidazole; lane 5, purified GFP (2 g). Electrophoresis was performed
under reduced conditions on 12% polyacrylamide gel.
hydrogen bonds and other interactions within the protein are
not significantly perturbed by suspending in ionic liquid,
leading to the conservation of the overall stability in protein.
Though unlike the peptide case that is completely soluble in
ionic liquid (Fig. 1), our affinity extraction of proteins
evidently shows that the GFP maintains fluorescence (Fig. 4
and S2, ESIw) and can be isolated and purified using the
protocol developed in this work.
purification of proteins by AILs (footnote 2, ESIw). Because
GFP is fluorescent, the recombinant His-tag GFP is
well-suited for our AIL experiments.
Our GFP (26.9 kDa) was expressed in E. coli as His-tag
fusion to the N terminus of GFPmut1 that contains the
double-amino-acid substitution (F64L and S65T), and
reportedly has been optimized for brighter fluorescence and
higher expression in E. coli.8 With this protein in hand, we
embarked upon direct affinity extraction of His-tag GFP in
crude E. coli lysate by Ni(II)-chelated AIL 2 onto ionic liquid.
It was reported in literature that if GFP is denatured its
fluorescence is quenched, and GFP emits fluorescence only
when it has the correct tertiary structure of the native form.9
Results of affinity extraction of Ni(II)-chelated AIL 2 with
His-tag GFP shown in Fig. 3 and 4 and S3 (ESIw) clearly
indicate that, because of the presence of fluorescence in the
ionic liquid layer, this His-tag GFP likely maintains its native
conformation with the ionic liquid (footnote 3, ESIw). No
precipitation of GFP at the ionic liquid–water interface was
observed. Both denaturing and native polyacrylamide gel
electrophoresis experiments illustrated that His-tag GFP could
be effectively extracted from the crude cell lysate onto the ionic
liquid and then competitively back-extracted using imidazole
to a newly prepared buffer solution (lane 4 in Fig. 3; lane 3 in
Fig. S3, ESIw). We noted that, upon affinity extraction, the
protein-bound AIL 2 was not totally miscible with ionic liquid
8 but uniformly dispersed in the ionic liquid phase (but clearly
not at all partitioning in aqueous layer) (Fig. 4B) (footnote 3,
ESIw). This phenomenon may be explained by the fact that
Ni(II)-chelated AIL 2 has a strong affinity for His-tag GFP and
therefore tightly coordinates with it to enable the complex
formation, ultimately leading to the complete departure of the
protein from the aqueous layer and quantitative transfer to
suspension in the ionic liquid phase.
In conclusion, we described our initial development of AILs
and demonstrated that Ni(II)-chelated AIL 2 is eminently
capable of performing binding interactions with His-tag
peptides and proteins in ionic liquid. Based upon biospecific
molecular recognition, this is the sole binding system of
receptor–ligand interactions in pure ionic liquid to date. As
the advance in research of protein compatible ionic liquids are
of great interest, the affinity protocol developed in this work is
practical and, most significantly, the GFP protein studied is
fluorescent and remains active. The results presented in this
report hold compelling possibilities for advancing biosensors
targeting a new range of analytes and applications in bio-
technology. Quantitative binding measurements of Ni(II)-
chelated AIL 2 with His-tag peptides and proteins in pure
ionic liquids are actively being pursued and the result will be
reported in due course.
This work was supported by a multi-year Grant-in-Aid
from the ANT Technology (Taipei, Taiwan) and in part by
the National Science Council (Taiwan, ROC). We thank
reviewers for their constructive comments.
Notes and references
1 N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37,
123.
2 S.-G. Lee, Chem. Commun., 2006, 1049.
3 H. Zhao, S. Xia and P. Ma, J. Chem. Technol. Biotechnol., 2005, 80,
1089.
4 X. He and T. H. Chan, Org. Lett., 2007, 9, 2681.
5 M. Zimmer, Chem. Rev., 2002, 102, 759.
6 H.-C. Kan, M.-C. Tseng and Y.-H. Chu, Tetrahedron, 2007, 63,
1644.
7 M. Blesic, A. Lopes, E. Melo, Z. Petrovski, N. V. Plechkova,
J. N. C. Lopes, K. R. Seddon and L. P. N. Rebelo, J. Phys.
Chem. B, 2008, 112, 8645.
This ionic liquid effect on proteins at the molecular level is
highly complex and understanding is far from complete, what
can be said is that in hydrophobic ionic liquid 8 this His-tag
GFP has shown its high stability (Fig. 4B). Furthermore, the
hydrophobic solvent usually has a lesser tendency to take
away the essential water from the protein surface and the
8 B. P. Cormack, R. H. Valdivia and S. Falkow, Gene, 1996, 173, 33.
9 S. Enoki, K. Saeki, K. Maki and K. Kuwajima, Biochemistry, 2004,
43, 14238.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7503–7505 | 7505