A new receptor with a FRET based fluorescence response for selective recognition of fumaric and maleic acids in aqueous medium

Article abstract of DOI:10.1039/c3cc45051a

Preferential binding of a new reagent to fumaric acid could be utilized for its estimation in aqueous medium and in commercial fruit juice.

Full text of DOI:10.1039/c3cc45051a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A new receptor with a FRET based fluorescence
response for selective recognition of fumaric and
maleic acids in aqueous medium†
Cite this: Chem. Commun., 2013,
49, 9818
Received 5th July 2013,
Upendar Reddy G.,a Rabindranath Lo,b Sovan Roy,a Tanmay Banerjee,a
Bishwajit Ganguly*b and Amitava Das*a
Accepted 27th August 2013
DOI: 10.1039/c3cc45051a
www.rsc.org/chemcomm
Preferential binding of a new reagent to fumaric acid could be to ortho- and meta-dibenzoic acids in mixed aq. buffer medium
utilized for its estimation in aqueous medium and in commercial at nearly neutral pH. A synthetic molecular probe was reported for
fruit juice.
estimation of citrate or tartrate–malate in mixed aq. buffer
medium.4b Thus, the issue of developing a reversible sensor
Molecular recognition studies of dicarboxylic acids have gained specific to maleate and fumarate ions in an ensemble of common
importance because of their presence as key structural moieties in mono- and di-carboxylate ions in aqueous medium has eluded
many bioactive molecules, their role in various metabolic processes researchers to date.3–8
and involvement in the biosynthesis of some important inter-
A new thiourea based receptor A was synthesized following a
mediates.1 Fumaric (Fum) and maleic (Mal) acids find a broad multistep procedure with reasonable yield and purity of this
range of uses in medicine, food and polymer industries.1 More reagent was ensured.† Model reagents (B and R) were synthe-
recently, fumaric acid derivatives have been tested for treatment of sized for unambiguous assignment of the spectral responses of
multiple sclerosis and patients with psoriasis.2 Maleic acid is A upon binding to fumarate or maleate ions (Scheme 1).
known for its role as an inhibitor of the Krebs cycle and in
The electronic and emission spectra of compounds A and B were
different kidney diseases.2 Due to the widespread use of these recorded in aq. HEPES buffer–CH3CN (1 : 1, v/v; pH 7.4) medium at
two acids as ingredients in food as well as beverages and their room temperature. The electronic spectrum of compound A showed
possible adverse influences on human health upon prolonged three shoulders at B353 nm, B290 nm and B244 nm, whereas in
exposure, it is important to develop an efficient reagent for their the case of compound B one distinct absorption band at 290 nm was
recognition and quantitative estimation in aqueous medium. In observed.† The common shoulder observed at 290 nm could be
the recent past, considerable progress has been made in the ascribed to a charge transfer (CT) transition that is typical for the
recognition of various dicarboxylic acids through hydrogen naphthalene moiety. A shoulder at B353 nm for A was ascribed to a
bonded adduct formation using hydrogen bond donor units like dansyl-based CT transition. UV-vis spectra of A and B remained
urea/thiourea/amide, guanidium groups, and allosteric phosphate- unchanged in the presence of excess of all mono and di-carboxylate
based receptors.3–6 Among these, the only report that reveals the ions under identical experimental conditions.†
recognition of certain dicarboxylate ions in pure aqueous tris
A solution of A in aq. HEPES buffer–CH3CN (1 : 1, v/v; pH 6.0)
buffer medium of pH 7.5 shows the lack of the desired specificity medium showed a strong emission band at 542 nm (lExt
=
towards maleate or fumarate.4a Alternative approaches to investi- 290 nm).† Emission spectra recorded for compound B, having
gate the coordinative interactions of these dicarboxylate ions with only a comparable naphthalene moiety as the fluorescence
certain metal ions or Lewis acid based receptors for recognition active unit (lExt of 290 nm), showed three emission bands at
studies are truly limited and in these examples the key issue of 313, 390 (broad) and 464 nm.† Bands at 313 and 390 nm were
specificity remains unaddressed.4,7 A chemodosimetric reagent
was reported to be able to discriminate between maleate and
fumarate ions.8 However, this reagent was also found to bind
a Organic Chemistry Division, CSIR-National Chemical Laboratory, Pune-411008,
India. E-mail: a.das@ncl.res.in; Fax: +91 2025902629; Tel: +91 2025902385
b CSIR-Central Salt & Marine Chemicals Research Institute, Bhavnagar – 364002,
India. E-mail: ganguly@csmcri.org; Fax: +91 278 2567562
† Electronic supplementary information (ESI) available: Synthesis and character-
ization details of A, B and R. Energy optimized structures and analysis of fumaric
Scheme 1 Molecular structures of A, B and R.
acid in fruit juice by HPLC.See DOI: 10.1039/c3cc45051a
c
9818 Chem. Commun., 2013, 49, 9818--9820
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
assigned as the vibronic bands of the naphthyl moiety, whereas a (lExt = 290 nm) were also recorded in the presence of 200 mole
relatively less intense emission band at 464 nm was assigned to an equivalent excess of Fum and Mal acids, while a significant quench-
emission process from an excited naphthyl dimer.3h Emission ing of the naphthalene-based emission was observed.† This clearly
spectra of R showed a band at 528 nm for a lExt of 355 nm, while a suggested that the binding of the maleate/fumarate ion to either of
very weak emission band was observed when excited at 290 nm.† two receptors (A and B) could cause an efficient quenching of the
Naphthalene moiety is expected to absorb predominantly for lExt naphthalene-based luminescence. Such an observation is anticipated
of 290 nm. A comparison of the emission spectra of A, B and R for the binding of the anionic analytes to a fluorophore without
clearly revealed the complete absence of the naphthalene-based much change in the molecular rigidity.9 Once the naphthalene
emission and presence of strong dansyl-based emission in A.† All based emission is effectively quenched, no FRET based ET and thus
¨
these tend to suggest that a Forster Resonance Energy Transfer dansyl-based emission at 542 nm are anticipated for A upon
(FRET) process was being operational for the probe molecule A at excitation at 290 nm. Thus, the observed fluorescence quenching
a lExt of 290 nm with the naphthalene moiety as the donor and of receptor A at 542 nm (lExt = 290 nm) could be attributed to the
the dansyl moiety as the acceptor fragments. Such a possibility is quenching of the luminescence of the naphthyl fragment and the
further supported by the fact that two model compounds R (with subsequent interruption of the FRET process. The residual lumines-
predominant dansyl-based absorption) and B (with naphthalene- cence of receptor A in the presence of excess of maleate/fumarate
based emission) showed a significant spectral overlap.† Energy ions could be accounted for by the weak dansyl-based emission at a
optimized structures of receptor A revealed that the distance
lExt of 290 nm. Low emission quantum yield measured for the model
between the dansyl moiety and two naphthalene units was 8 Å compound R at 528 nm (lExt of 290 nm) further confirmed this.
and 10 Å, respectively.† This further supported the feasibility of a
Systematic luminescence titrations were carried out for A
FRET process. Efficiency of the FRET process for the energy (2.0 Â 10À5 M) with varying [Fum] (0 to 3.2 Â 10À3 M) and [Mal]
transfer (ET) between naphthyl and dansyl moieties in A was (0 to 2.8 Â 10À3 M) in aq. HEPES buffer–CH3CN (1:1, v/v; pH 6.0)
evaluated to be 75% under the experimental conditions.†
medium at a lExt of 290 nm and a lMon of 542 nm (Fig. 1). A gradual
Emission quantum yield (FF of 0.0137) for the model com- decrease in emission intensity at 542 nm was observed upon addition
pound B was evaluated in aq. HEPES buffer–CH3CN (1 : 1, v/v; of either of these two carboxylic acids (Fig. 1). Stoichiometry for the
pH 6.0) medium (lExt = 285 nm, lEms = 394 nm) using naphtha- adduct formation between receptor A and the fumarate or maleate
lene as the reference compound. It is known that the thiourea ion was evaluated to be 1 : 1 from the Benesi–Hildebrand (B–H) plot
moiety could even act as a hydrogen bond donor for polar solvent of the data obtained from the systematic fluorescence titration.†
molecules like DMSO.9 Thus, more polar water molecules are This was also confirmed from the data obtained from the ESI-MS
expected to interact with B thereby favouring the non-radiative analysis. Signals at m/z of 843.28 and m/z 843.66 were attributed to
deactivation of the naphthalene based excited state. This could [A + fumarate + Na+] and [A + maleate + Na+], respectively. The
account for the low quantum yield value of FRET based ET (FET
75%) despite a strong spectral overlap between B and R.†
=
respective binding affinity of Fum and Mal acids towards A was
evaluated to be (8.6 Æ 0.02) Â 102 MÀ1 and (2.1 Æ 0.04) Â 102 MÀ1 (an
In order to examine the response of the probe molecule A average of four independent experimental data) from the subsequent
towards different carboxylic acids, we recorded the luminescence B–H plot in aq. HEPES buffer-acetonitrile (1:1, v/v; pH 6.0) medium.†
spectra of A in the absence and presence of various mono, di- and
The interactions between the TBA salt of fumaric acid (TBAF) and
tri-carboxylic acids (Fig. 1). Luminescence spectra of A remained maleic acid (TBAM) with A were also investigated by 1H NMR
virtually unchanged when recorded in the presence of all other titration with varying [TBAF] (Fig. 2) or [TBAM].† The initial chemical
carboxylic acids, except Fum and Mal acids. For these two acids, a shifts for signals for the four protons of two dissymmetric thiourea
significant quenching of the FRET based luminescence at 542 nm functionalities in A appeared as two sharp singlets at d = 8.28 ppm
was observed (Fig. 1), which was presumably due to the formation (belonging to the N1-atom) and d = 6.52 ppm (belonging to the
of a hydrogen bonded adduct between reagent A and the fumarate N2-atom), respectively. Distinct downfield shifts of four thiourea-
or maleate ion. The extent of changes was more pronounced for the protons in A were observed upon subsequent increase in the [TBAF]
fumarate ion. Luminescence responses of the model receptor B or [TBAM]. These shifts were smaller for studies with TBAM, which
Fig. 1 Luminescence spectra A (2.0 Â 10À5 M) in the presence of varying
(a) [Fum] (0 to 3.2 Â 10À3 M) and (b) [Mal] (0 to 2.8 Â 10À3 M); inset: the
emission response of A in the absence and presence of 200 mole equiv. of Fum &
Mal and various carboxylic acids (X = acetic, oxalic, malonic, citric, succinic, adipic,
glutaric, suberic, malic). All studies were performed in 10 mM aq. HEPES buffer–
Fig. 2 Partial 1H NMR spectra of A(2.83 mM) in the absence and in the presence
CH3CN (1 : 1, v/v; pH 6.0) medium; lExt = 290 nm, lMon = 542 nm.
of varying [TBAF] in a CD3CN–DMSO(d6) (99 : 1, v/v) medium.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9818--9820 9819

View Article Online
ChemComm
Communication
These data support the suitability of the proposed method for
its application to real samples. For analysis of the commercial
sample of apple juice, 0.2 ml of this was used, which was
eventually adjusted to the final volume of 5 ml after adding an
appropriate amount of reagent A (1.0 Â 10À4 M) and aq. HEPES
buffer–acetonitrile (1 : 1, v/v; pH 7.4) as solvent. This solution
along with solutions spiked with known [Fum] (20 mM, 40 mM
and 60 mM) as an internal standard were used for emission
measurements without further treatment. The fumaric acid
concentration in the commercial apple juice was determined
to be 3.8 Â 10À5 M, which is within the allowed limit for fumaric
acid content in good apple juice.10 These results were compared
and validated with the results of the HPLC studies.†
2À
Fig. 3 B3LYP/6-31G*//RHF/PM3 calculated binding energies for A with F
À2H+
À
and A with M
and M
2À. Distances are given in Å (atom colour code:
ÀH+
À2H+
red = O, blue = N, white = H, yellow = S, magenta = C).
further corroborate its weaker interaction with A.† These shifts were
attributed to the net deshielding effect induced by the hydrogen-
bonding interaction between the thiourea protons and the anion.
Higher Dd for the N1-protons signified a stronger interaction with
the fumarate or maleate ion than Dd for N2-protons. To understand
this and to examine the relative binding affinities of A towards
fumarate and maleate ions, detailed computational studies
were performed.
In brief, the new thiourea-based receptor (A) could be used
for selective recognition of fumarate and maleate ions in an
ensemble of several other mono- and di-carboxylic acids in an
aqueous environment based on a binding induced modulated
FRET response. This reagent could further be used for quantita-
tive estimation of the fumarate ion in commercial fruit juice.
The authors thank the CSIR-Network project (M2D) for
financial support. URG and RL thank UGC (India) and TB
thanks CSIR (India) for their research fellowships.
The binding of Fum and Mal acids with A could occur in
the mono- or bis-deprotonated states of these diacids. The
fractional distribution curve for respective acids reveals that at
Notes and references
pH 6.0, mono (M
À) and bis-deprotonated (M
2À) forms of
´
´
1 (a) I. M. Perez-Dıaz and R. F. McFeeters, J. Food Sci., 2010, 75, M204;
(b) S. Rouse and D. J. van Sinderen, J. Food Prot., 2008, 71, 1724.
2 (a) D. M. Khiabani, A. Blank, T. Skripuletz, E. Miller, A. Kotsiari,
V. Gudi and M. Stangel, PLoS One, 2010, 5, e11769; (b) B. D. Trapp
and K. A. Nave, Annu. Rev. Neurosci., 2008, 31, 247; (c) J. Wernerman
and F. Hammarkvist, Lancet, 1990, 335, 701; (d) R. H. Grubbs and
W. Tumas, Science, 1989, 243, 907.
ÀH+
À2H+
the Mal acid exist in comparable concentrations, while the bis-
deprotonated form (F
2À) exists almost exclusively for Fum
À2H+
acid.† Thus, binding energies were calculated for binding of A to
M
F
À, M
2À and F
2À. The higher binding energy of the
ÀH+
À2H+
À2H+
2À with A was accounted for by the interaction of two –N1H
À2H+
3 (a) K. Ghosh and T. Sen, J. Phys. Chem. B, 2011, 115, 8597; (b) E. Fan,
S. A. V. Arman, S. Kincaid and A. D. Hamilton, J. Am. Chem. Soc.,
1993, 115, 369; (c) T. R. Kelly and M. H. Kim, J. Am. Chem. Soc., 1994,
groups adjacent to the naphthyl ring via strong H-bonding with
2À
the two OCOOÀ of F
was found to be similar to that of F
binding energy of A with M
2À (Fig. 3). The binding mode of M
À2H+
À2H+
´
116, 7072; (d) J. M. Benito, M. G. Garcia, J. L. J. Blanco, C. O. Mellet
2À. The computed
was reasonably lower when
À2H+
´
and J. M. G. Fernandez, J. Org. Chem., 2001, 66, 1366; (e) S.-Y. Liu,
À
L. Fang, Y.-B. He, W.-H. Chan, K.-T. Yeung, Y.-K. Cheng and
R.-H. Yang, Org. Lett., 2005, 7, 5825; ( f ) A. M. Costero, M. Colera,
ÀH+
À
compared with M
2À. However, the presence of M
in
À2H+
ÀH+
˜
P. Gavina and S. Gil, Chem. Commun., 2006, 761; (g) T. Gunnlaugsson,
comparable concentration seems to mask the effective binding
of the bis-deprotonated maleate ion at this pH.
A. P. Davis, J. E. O’Brien and M. Glynn, Org. Biomol. Chem., 2005,
3, 48; (h) Z. H. Lin, L. X. Xie, Y. G. Zhao, C. Y. Duan and J. P. Qu, Org.
Biomol. Chem., 2007, 5, 3535; (i) S. Goswami, K. Ghosh and
S. Dasgupta, J. Org. Chem., 2000, 65, 1907; ( j) S. K. Kim, B.-G.
Kang, H. S. Koh, Y. J. Yoon, S. J. Jung, B. Jeong, K.-D. Lee and
To explore the possibility of evaluating unknown [Fum] in
commercial fruit juice, a standard curve was generated from the
plot of the D[I0 À I] vs. known [Fum] (0 to 1.0 Â 10À4 M) (Fig. 4).†
The linearity of the calibration curve for known [Fum] was
ensured (Fig. 4) with the lowest detection limit of 10 ppm.
˜
J. Yoon, Org. Lett., 2004, 6, 4655; (k) P. Mateus, R. Delgado, P. Brandao
´
and V. Felix, J. Org. Chem., 2012, 77, 4611.
4 (a) J. Raker and T. E. Glass, J. Org. Chem., 2002, 67, 6113; (b) A. Metzger
and E. V. Anslyn, Angew. Chem., 1998, 37, 649; (c) J. J. Lavigne and
E. V. Anslyn, Angew. Chem., Int. Ed., 1999, 38, 3666.
5 (a) D. A. Jose, I. Mon, H. F. Perez, E. C. E. Adan, J. B. Buchholz and
A. V. Ferran, Org. Lett., 2011, 13, 3632; (b) M. Takeuchi, T. Imada and
S. Shinkai, Angew. Chem., 1998, 37, 2096; (c) T. Ikeda, O. Hirata,
M. Takeuchi and S. Shinkai, J. Am. Chem. Soc., 2006, 128, 16008.
6 Y. Wu, H. Guo, X. Zhang, T. D. James and J. Zhao, Chem.–Eur. J.,
2011, 17, 7632.
7 (a) M. Hu and G. Feng, Chem. Commun., 2012, 48, 6951; (b) L. Tang,
J. Park, H.-J. Kim, Y. Kim, S. J. Kim, J. Chin and K. M. Kim, J. Am.
Chem. Soc., 2008, 130, 12606; (c) F. Li, S. Carvalho, R. Delgado,
´
M. G. B. Drew and V. Felix, Dalton Trans., 2010, 39, 9579.
´
´
´˜
´
8 F. Sancenon, R. Martınez-Manez, M. A. Miranda, M.-J. Seguı and
J. Soto, Angew. Chem., 2003, 115, 671.
9 (a) A. Ghosh, B. Ganguly and A. Das, Inorg. Chem., 2007, 46, 9912;
(b) A. Ghosh, S. Verma, B. Ganguly, H. N. Ghosh and A. Das, Eur. J.
Inorg. Chem., 2009, 2496.
Fig. 4 Plot of DI = (I0 À I) vs. [Fum], where I0 and I are emission intensities of
receptor A in the absence and presence of known [Fum] as well as apple juice
and apple juice spiked with a known amount of fumaric acid.
10 V. Gokmen and J. Acar, Food Addit. Contam., 2004, 21, 626.
c
9820 Chem. Commun., 2013, 49, 9818--9820
This journal is The Royal Society of Chemistry 2013