Carbohydrate-based switch-on molecular sensor for Cu(II) in buffer: Absorption and fluorescence study of the selective recognition of Cu(II) ions by galactosyl derivatives in HEPES buffer

Article abstract of DOI:10.1021/ol061274f

1-(β-D-Galactopyranosyl-1′-deoxy-1′-iminomethyl) -2-hydroxynaphthalene (L₁), possessing an ONO binding core, was found to be selective for Cu²⁺ ions in N-[2-hydroxyethyl]piperazine- N′-[2-ethanesulfonic acid] buffer, at concentrations ≤580 ppb, at physiological pH by eliciting switch-on behavior, whereas the other ions, viz., Mg²⁺, Ca²⁺, Mn²⁺, Fe²⁺, Co ²⁺, Ni²⁺, Zn²⁺, and Cd²⁺, caused no significant change in the fluorescence. Whereas the binding characteristics were ascertained by absorption spectroscopy, the species formed were shown by Q-TOF ES MS.

Full text of DOI:10.1021/ol061274f

ORGANIC
LETTERS
2006
Vol. 8, No. 16
3525-3528
Carbohydrate-Based Switch-On
Molecular Sensor for Cu(II) in Buffer:
Absorption and Fluorescence Study of
the Selective Recognition of Cu(II) Ions
by Galactosyl Derivatives in HEPES
Buffer
Nitin Kumar Singhal,† Balaji Ramanujam,† Vairamani Mariappanadar,‡ and
Chebrolu Pulla Rao*,†
Bioinorganic Laboratory, Department of Chemistry, Indian Institute of Technology
Bombay, Mumbai 400 076, India, and Analytical Chemistry DiVision, Indian Institute
of Chemical Technology, Hyderabad 500 007, India
cprao@iitb.ac.in
Received May 25, 2006
ABSTRACT
1-(
â
-
D
-Galactopyranosyl-1
′
-deoxy-1
′-iminomethyl)-2-hydroxynaphthalene (L1), possessing an ONO binding core, was found to be selective for
+
Cu2 ions in N-[2-hydroxyethyl]piperazine-N
′
-[2-ethanesulfonic acid] buffer, at concentrations e580 ppb, at physiological pH by eliciting switch-
on behavior, whereas the other ions, viz., Mg2 , Ca2 , Mn2 , Fe2 , Co2 , Ni2 , Zn2 , and Cd2 , caused no significant change in the fluorescence.
+
+
+
+
+
+
+
+
Whereas the binding characteristics were ascertained by absorption spectroscopy, the species formed were shown by Q-TOF ES MS.
Copper is an essential element of life by being involved in
the functional part of a number of enzymes and also being
involved in transcriptional events.1 This necessitates sensing
as well as quantifying copper in biological systems, and
hence, there is a need to develop biologically compatible
sensors. Because carbohydrates are biologically compatible
as well as water soluble, it is desirable to have sensors based
on such molecular systems. Synthetic carbohydrate-based
sensors capable of reporting metal ions, including Cu2+, are
scarce in the literature. However, non-carbohydrate-based
sensors suitable for biological applications are known in the
literature for both copper2 and zinc.3 Therefore, the present
communication deals with the absorption and fluorescence
studies of a simple galactosyl-based naphthyl derivative,
1-(â-D-galactopyranosyl-1′-deoxy-1′-iminomethyl)-2-hydroxy-
naphthalene (L1, Scheme 1) (Supporting Information, S1),
in the recognition of Cu2+ ions in HEPES buffer at pH 7.2-
7.4.
† Indian Institute of Technology Bombay.
‡ Indian Institute of Chemical Technology.
(1) (a) Messerschmidt, A.; Huber, R.; Poulos, T.; Weighardt, K.; Eds.
Hand. Metalloproteins 2001, 2, 1149. (b) Hu, S.; Furst, P.; Hamer, D. New
Biol. 1990, 2, 544.
10.1021/ol061274f CCC: $33.50
© 2006 American Chemical Society
Published on Web 07/12/2006

5.3 ( 0.3 and a Ka yields of 50 500 ( 700 M-1. Similar
titrations performed between L1 and other divalent ions of
Mg, Ca, Mn, Fe, Co, Ni, Zn, and Cd exhibited no significant
change in the emission intensities (Figure 1a). At lower and
higher pH, viz., 6.0 and 8.0, the fluorescence is further
quenched. A change in the anion, viz., Cu(NO3)2 and CuCl2,
indicated no influence of the counteranion on the fluores-
cence results. Binding of Cu2+ with L1 in HEPES buffer was
further confirmed by absorption spectral studies (Figure 1b)
that provide a Ka of 42 500 ( 1000 M-1. However, Zn2+
does not show any significant binding in buffer (Figure 1c)
(Supporting Information, S4).
Titration of a mixture of {L1 and Cu2+} with M2+ (Figure
1d) in HEPES buffer resulted in a decrease of fluorescence
intensity by ∼10% in the case of Mg2+, Ca2+, Mn2+, and
Zn2+, by ∼30% in the case of Fe2+, Co2+, and Ni2+, and by
∼45% in the case of Cd2+, indicating that there is a net
fluorescence increase of L1 by 290-470% when Cu2+ is
bound in the presence of other biologically relevant metal
ions reported here. This can also be gauged by the fluores-
cence decay constants obtained during the titrations of {L1
and Cu2+} with M2+, viz., 37 100, 38 000, 40 000, 23 100,
and 29 400 M-1, respectively, for Fe2+, Co2+, Ni2+, Cu2+,
and Cd2+. Fluorescence studies performed in HEPES buffer
by varying the concentration of Cu2+ but keeping the [Cu2+]/
[L1] mole ratio at 1:2 clearly indicated that the detection of
580 ppb can be established with an increase in fluorescence
by at least 150% (Supporting Information, S4). Thus, Cu2+
can be selectively recognized and quantified in HEPES buffer
at a physiological pH using L1 to quite low concentrations
even in the presence of other biologically relevant metal ions,
Scheme 1
Titration of L1 with Cu2+ in MeOH resulted in a substantial
increase in the fluorescence intensity of a 354 nm band until
the [Cu2+]/[L1] mole ratio is 0.5 (I/I0 ratio is 18 ( 1) and
then a decrease thereafter. Similar titration with Zn2+ in
MeOH shows a gradual increase in the fluorescence intensity
of a 450 nm band until the [Zn2+]/[L1] mole ratio of 1.0
(I/I0 ratio of 18 ( 2) and saturation thereafter. The titration
of L1 in MeOH with other biologically relevant divalent ions
of Mg, Ca, Mn, Fe, Co, Ni, and Cd exhibited no significant
change in the fluorescence intensity of either the 350 or the
450 nm bands. On the basis of the Benesi-Hildebrand
equation, Ka’s of 50 000 ( 1000 and 50 500 ( 1000 M-1
were derived, respectively, for Zn2+ and Cu2+ (Supporting
Information, S2).
The binding of Cu2+ and Zn2+ to L1 in MeOH was further
ascertained from the absorption spectral changes noticed in
the titrations. A plot of absorbance vs the [M2+]/[L1] mole
ratio for various bands clearly indicated the formation of a
1:1 complex in the case of Zn2+ and a 1:2 complex in the
case of Cu2+, the same as that obtained from the fluorescence
studies, and the corresponding logarithmic plots yielded Ka’s
of 44 500 ( 1500 and 50 500 ( 3500 M-1 in the cases of
Cu2+ and Zn2+, respectively. Similar titrations performed in
MeOH with other metal ions did not show any interpretable
changes in the spectra (Supporting Information, S3).
However, the fluorescence titrations performed between
M2+ and L1 in HEPES buffer, at pH 7.2-7.4, yielded
altogether different results. In buffer, only the titration of
Cu2+ with L1 showed a gradual enhancement in the fluo-
rescence intensity of the 350 nm band until the [Cu2+]/[L1]
mole ratio was 0.5 and a decrease thereafter with I/I0 being
by eliciting a switch-on fluorescence behavior (ΦL1
0.0005, ΦL1ꢀL1 ) 4.7 M-1 cm-1; and ΦL1+Cu ) 0.0035,
L1+CuꢀL1+Cu ) 34.4 M-1 cm-1) with a 7.0-fold increase in
)
Φ
quantum yield and a ∼7.3-fold increase in brightness
(ΦL1ꢀL1) upon Cu2+ addition.
These studies were appropriately compared with those of
the synthetic control molecular systems, L2, L3, and L4
(Figure 2, Supporting Information, S1) to establish the role
of imine and carbohydrate units in the sensor property of
L1.
(2) (a) Zeng, Li; Miller, E. W.; Pralle, A.; Isacoff, E. Y.; Chang, C. J. J.
Am. Chem. Soc. 2006, 128, 10. (b) Banthia, S.; Samanta, A. New J. Chem.
2005, 29, 1007. (c) Singh, A.; Yao, Q.; Tong, L.; Clark Still, W.; Sames,
D. Tetrahedron Lett. 2000, 41, 9601. (d) Cano-Raya, C.; Fern’andez-Ramos,
M. D.; Capit’an-Vallvey L. F. Anal. Chim. Acta 2006, 555, 299. (e) Brunner,
J.; Kraemer, R. J. Am. Chem. Soc. 2004, 126, 13626. (f) Royzen, M.; Dai,
Z.; Canary, J. W. J. Am. Chem. Soc. 2005, 127, 1612. (g) Torrado, A.;
Walkup, G. K.; Imperiali, B. J. Am. Chem. Soc. 1998, 120, 609. (h) Xu,
Z.; Qian, X.; Cui, J. Org. Lett. 2005, 7, 3029. (i) Xu, Z.; Xiao, Y.; Qian,
X.; Cui, J.; Cui, D. Org. Lett. 2005, 7, 889. (j) Gunnlaugsson, T.; Leonard,
J. P.; Murray, N. S. Org. Lett. 2004, 6, 1557.
(3) (a) Woodroofe, C. C.; Lippard, S. J. J. Am. Chem. Soc. 2003, 125,
11458. (b) Shults, M. D.; Pearce, D. A.; Imperiali, B. J. Am. Chem. Soc.
2003, 125, 10591. (c) Taki, M.; Wolford, J. L.; O’Halloran, T. V. J. Am.
Chem. Soc. 2004, 126, 712. (d) Walkup, G. K.; Imperiali, B. J. Am. Chem.
Soc. 1996, 118, 3053. (e) Nolan, E. M.; Burdette, S. C.; Harvey, J. H.;
Hilderbrand, S. A.; Lippard, S. J. Inorg. Chem. 2004, 43, 2624. (f) Chang,
C. J.; Nolan, E. M.; Jaworski, J.; Burdette, S. C.; Sheng, M.; Lippard, S. J.
Chem. Biol. 2004, 11, 203. (g) Chang, C. J.; Jaworski, J.; Nolan, E. M.;
Sheng, M.; Lippard, S. J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1129.
(h) Chang, C. J.; Nolan, E. M.; Jaworski, J.; Okamoto, K.; Hayashi, Y.;
Sheng, M.; Lippard, S. J. Inorg. Chem. 2004, 43, 6774. (i) Royzen, M.;
Durandin, A.; Young, V. G., Jr.; Geacintov, N. E.; Canary, J. W. J. Am.
Chem. Soc. 2006, 128, 3854. (j) Ajayaghosh, A.; Carol, P.; Sreejith, S. J.
Am. Chem. Soc. 2005, 127, 14962.
Titration of Cu2+ with L2 in HEPES buffer showed no
significant change in the fluorescence intensity, suggesting
that the -HCdN- moiety in L1 favors Cu2+ binding over
that in L2 owing to the formation of a chelate in the case of
L1. The effect of such a chelation on the recognition of one
metal ion over the other has been recently demonstrated in
the case of calix[4]arene derivatives.4 However, the titration
with Cu2+ resulted in an increase of I/I0 to only 2.7 in the
case of L3 and to 2.3 in the case of L4 (Figure 3a). These
data suggest that the marginal increase observed in I/I0 in
the case of L3 and L4 can be explained on the basis of the
presence of imine and naphthylic -OH moieties suitable for
forming six-membered chelate and precludes the involvement
of the carbohydrate moiety in L4 in binding at least at the
(4) (a) Dessingou, J.; Joseph, R.; Rao, C. P. Tetrahedron Lett. 2005, 46,
7967. (b) Kumar, A.; Ali, A.; Rao, C. P. J. Photochem. Photobiol., A Chem.
2006, 117, 164.
3526
Org. Lett., Vol. 8, No. 16, 2006

Figure 1. (a) Plot of relative fluorescence intensity (I/I0) vs [M2+]/[L1] mole ratio in HEPES buffer. The symbols refer to: 9 (Mg2+), b
(Ca2+), 2 (Mn2+), 1 (Fe2+), triangle pointing left (Co2+), triangle pointing right (Ni2+), [ (Cu2+), > (Zn2+), ` (Cd2+). UV-vis absorption
spectra measured in the titration of M2+ with L1 in HEPES buffer at different [M2+]/[L1] mole ratios ranging from 0.0 to 2.0: (b) Cu2+; (c)
Zn2+. (d) Histogram showing the relative fluorescence intensity (I/I0) in the titration of {L1 + 0.5 equiv of Cu2+} with M2+. The concentration
of the competitor metal ions used was 0-2.0 equiv, and at the highest ratio, this comes out to be 4 mol equiv with respect to the copper.
Because Cu1+ is possible to be present in association with
Cu2+ in biological systems, the interaction of Cu1+ with L1
was studied. Titration of L1 with Cu1+ in HEPES buffer
shows an increase in the fluorescence of the 356 nm band
to an I/I0 value of 3.0 ( 0.5, which is comparable with that
observed with Cu2+. The results of the Cu1+ titration were
confirmed by repeating the experiment four times. Further,
in the titration of Cu1+, the 440 nm band also exhibited
Figure 2. Schematic structures of the control molecules L2, L3,
and L4.
similar behavior, but the peak I/I0 value was only 1.5 ( 0.1,
indicating a rather less significant change in this band. Thus,
the fluorescence behavior of Cu1+ with L1 is almost the same
as that of the Cu2+ titration. To further confirm this, two
additional sets of experiments were carried out. This includes
titration of [L1 + 0.5 equiv of Cu2+] with Cu1+ and a reverse
experiment, viz., titration of [L1 + 1.0 equiv of Cu1+] with
Cu2+, where the results indicate that both of the ions were
behaving similarly. This means that L1 can be used to detect
and/or measure the copper ions either as 1+ or 2+ or as a
mixture. By combining these results, it is possible to conclude
that the carbohydrate-based ligand (L1) reported here can
detect both Cu2+ and Cu1+ and hence can be very well suited
for application in biological systems.
Figure 3. (a) Plot of I/I0 vs [Cu2+]/[L] mole ratio in HEPES buffer
(pH ) 7.2). (b) Histogram indicating I/I0 at the [M2+]/[L] mole
Titration of L3 and L1 with Cu2+ was also carried out using
Q-TOF ES MS. In the titration of L3 with Cu2+, besides the
ligand molecular ion peak (m/z ) 228), only a 2:1 (ligand/
Cu2+ ratio, m/z ) 516) species was present at the 0.2 mole
ratio of Cu2+. At 0.5 mol equiv of Cu2+, the 2:1 species
further increased with the appearance of a small portion of
1:1 (m/z ) 289). However, further addition of Cu2+ in the
titration exhibits a large increase in the 1:1 species whereas
the 2:1 species does not increase any further as shown
(Supporting Information, S6); the corresponding species
formed are shown in Scheme 2a. Comparison of this with
the fluorescence behavior indicates clearly that the fluoresc-
ing behavior of these species follows a trend, 1:1 > 2:1. In
the titration of L3 with Cu2+, no traces of 2:2 species were
formed.
ratio of 0.5 in the titration of L1, L2, L3, and L4 with different M2+
.
[Cu2+]/[L4] mole ratio of 0.5. The sharp contrast observed
in I/I0 between L1 and L4 is attributable to the axial
orientation of C2-OH in the case of ribosyl-based L4.
However, at the mole ratio of [Cu2+]/[L] beyond 0.5, a
dinuclear copper species is favored in which the carbohydrate-
C2-OH is involved in bridging. Thus, all these results
obtained in the titrations of aqueous HEPES buffer support
the involvement of the imine and the carbohydrate moieties
present in L1 in Cu2+ recognition. Further, the titration of
Mg2+, Ca2+, Mn2+, Fe2+, Co2+, Ni2+, Zn2+, and Cd2+ with
the sensor molecule, L1, and the control molecules, L2, L3,
and L4, under similar conditions in HEPES buffer did not
alter the emission intensities to any significant extent (Figure
3b).
In the titration of L1 with Cu2+, as the mole ratio of Cu2+
increases from 0 to 0.2 to 0.5, the formation of 2:1 (ligand/
Cu2+ ratio, m/z ) 728) species is increased considerably and
Org. Lett., Vol. 8, No. 16, 2006
3527

Scheme 2a
a Species formed based on the Q-TOF ES MS experiment: titration of (a) L3 and (b) L1 with Cu2+
.
decreases thereafter, and the species is completely absent at
higher Cu2+ mole ratios. On the other hand, the formation
of 1:1 (m/z ) 395) species increases as Cu2+ is added and a
steep increase is observed beyond the Cu2+ mole ratio of
2.0. As the concentration of Cu2+ increases, the formation
of 2:2 (m/z ) 789) species also increases, whereas the
molecular ion peak of the ligand (m/z ) 334) decreases
progressively and disappears beyond 2 equiv of Cu2+
addition. The isotopic pattern of the peaks resulting from
the copper-bound species was consistent with the isotopes
of copper and their abundance ratio. The extent of all these
species can be clearly seen from the Supporting Information
(S6), and the species formed are shown in Scheme 2b.
Comparison of this with the fluorescence titration indicates
clearly that the fluorescing behavior of these species follows
a trend, 2:1 . 2:2 > 1:1. The nature of the 1:1 species
formed in the case of L1 and L3 could differ in the fact that
the coordination in L3 is simply bidentate, whereas L1 can
extend a pseudocoordination from the carbohydrate C2-OH
additionally. An optimized structure using hybrid density
functional theory (B3LYP) with a basis set of LANL2DZ
for Cu2+ resulted in a Cu2+‚‚‚Ocarb distance of 3.26 Å. The
Cu2+ is coordinatively unsaturated in both the 1:1 species,
and to fill the coordination either the anion or the solvent,
or both, will be used up. Although L1 can give rise to 2:2
species, L3 does not exhibit any such species. This contrasting
behavior observed between L1 and L3 is explainable based
on the fact that L1 can form dinuclear species using its
carbohydrate C2-OH moiety, whereas L3 cannot, owing to
the presence of a butyl moiety instead of the carbohydrate
moiety.
[M2+]/[L1] mole ratio >1.0 (Supporting Information, S5).
The 2:2 dinuclear complex proposed for 1 was supported
by crystal structure determinations of 4,6-di-O-protected-D-
glucopyranosylamine derivatives where the Cu2O2 core was
stabilized by two intramolecular H-bonds utilizing equatorial
C2-OH of the carbohydrate.5 The presence of equatorial C2-
OH is still retained in L1 (galactosyl based) but not in L4
(ribosyl based), and hence, L1 stands as a better molecule
for sensing Cu2+.
Although the carbohydrate-based metal ion sensor mol-
ecules are scarce in the literature, L1 stands as a unique
molecular system that works well (by exhibiting a 7-fold
increase in the quantum yield) in detecting very low
concentrations of Cu2+ even in the presence of other
biological metal ions under physiological pH. Also demon-
strated were the merits of L1 against three control molecules,
the involvement of the carbohydrate moiety in metal ion
recognition, the binding characteristics and the detection of
copper-bound species, recognition of Cu1+, and the switch-
on behavior in the presence of Cu2+. Thus, the galactosyl-
based L1 may offer its utility for studying Cu2+ in biological
samples even in association with other metal ions as well as
Cu1+, as no such synthetic glyco-conjugate sensor is known
to date for transition metal ions.
Acknowledgment. C.P.R. acknowledges the financial
support from DST, CSIR, and DAE-BRNS.
Supporting Information Available: Synthesis, charac-
terization, and experimental details (S1); fluorescence and
absorption in MeOH (S2 and S3); fluorescence in HEPES
buffer (S4); 1 and 2 (S5); and mass spectral data for Cu2+
titration (S6). This material is available free of charge via
the Internet at http://pubs.acs.org.
Synthetic reactions carried out between M(OAc)2‚nH2O
and L1 in MeOH resulted in the formation of {Cu(L1)(OAc)-
(H2O)2}2, 1, and {Zn(L1)(OAc)(MeOH)2}, 2, and these
exhibited fluorescence characteristics similar to those ob-
tained in the solution titration studies reported here. The
dinuclear Cu(II) complex, 1, exhibits a spectrum that is in
agreement with that obtained from the titration studies at
the [Cu2+]/[L1] mole ratio >1.0. Similarly, the absorption
spectra of 1 and 2 agree well with those obtained at the
OL061274F
(5) (a) Rajsekhar, G.; Sah, A. K.; Rao, C. P.; Guionneau, P.; Bharathy,
M.; GuruRow, G. N. Dalton Trans. 2003, 3126. (b) Sah, A. K.; Rao, C. P.;
Saarenketo, P. K.; Rissannen, K.; van Albada, G. A.; Reedijk, J. Chem.
Lett. 2002, 348.
3528
Org. Lett., Vol. 8, No. 16, 2006