Synthesis and metal-binding properties of chelating fluorescein derivatives

Article abstract of DOI:10.1021/ol0344570

(Matrix presented) Two routes to highly functionalized metal-chelating fluorescein derivatives have been pursued. Compound 3 is partially quenched by a variety of first-row transition metal ions in aqueous solution, with EC ₅₀ values ranging fr

Full text of DOI:10.1021/ol0344570

ORGANIC
LETTERS
2003
Vol. 5, No. 12
2051-2054
Synthesis and Metal-Binding Properties
of Chelating Fluorescein Derivatives
Matthew A. Clark, Kathryn Duffy, Jyoti Tibrewala, and Stephen J. Lippard*
Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
lippard@lippard.mit.edu
Received March 15, 2003
ABSTRACT
Two routes to highly functionalized metal-chelating fluorescein derivatives have been pursued. Compound 3 is partially quenched by a variety
of first-row transition metal ions in aqueous solution, with EC values ranging from 0.4 to 60 µM. Compounds of this type may find application
50
in biological sensing.
Fluorescent sensing of ions and small molecules is an active
area of chemical research.1 Sensors for physiologically rele-
vant metal ions such as calcium and zinc have seen extensive
application in biological studies. Developing sensors for other
biologically important molecules such as nitric oxide,2
organic phosphates,3 carbohydrates,4 peptides,5 and others
is the goal of much ongoing research. Practical sensors under-
go spectral shifts or emission intensity changes upon interac-
tion with an analyte. Fluorescence response can be triggered
by a variety of mechanisms,1 some examples of which
include chemical reaction of the fluorophore,6 ligation-
induced abolition of PET,7 and conformational changes that
perturb electronic structure or trigger FRET.8 We have
developed a strategy by which a fluorescence response is
mediated by metal coordination of an analyte, in this case
nitric oxide.9 In this approach, the metal center acts as both
a quenching unit and a structural organizing element.
Coordination of an analyte to the metal displaces the
fluorophore, restoring fluorescence and providing an observ-
able signal (see Figure 1).10
For reasons of synthetic convenience, most work in the
sensor field has focused on relatively short-wavelength
(1) (a) Burdette, S. C.; Lippard, S. J. Coord. Chem. ReV. 2001, 216-
217, 333. (b) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.;
Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem.
ReV. 1997, 97, 1515. (c) Valeur, B.; Leray, I. Coord. Chem ReV. 2000,
205, 3. (d) Czarnik, A. W. Acc. Chem. Res. 1994, 27, 302.
(2) Nagano, T.; Yoshimura, T. Chem. ReV. 2002, 102, 1235.
(3) Best, M. D.; Anslyn, E. V. Chem. Eur. J. 2003, 9, 51.
(4) (a) James, T. D.; Shinmori, H.; Shinkai, S. Chem. Commun. 1997,
71. (b) Adhikiri, D. P.; Heagy, M. D. Tetrahedron Lett. 1999, 40, 7893.
(5) Hioki, H.; Kubo, M.; Yoshida, H.; Bando, M.; Ohnishi, Y.; Kodama,
M. Tetrahedron Lett. 2002, 43, 7949.
(6) (a) Kojima, H.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Hirata, Y.;
Nagano, T. Angew. Chem., Int. Ed. 1999, 38, 3209. (b) Meineke, P.; Rauen,
U.; De Groot, H.; Korth, H.-G.; Sustmann, R. Chem. Eur. J. 1999, 5, 1738.
(7) (a) Walkup, G. K.; Burdette, S. C.; Lippard, S. J.; Tsien, R. Y. J.
Am. Chem. Soc. 2000, 122, 5644. (b) Burdette, S. C.; Walkup, G. K.;
Spingler, B.; Tsien, R. Y.; Lippard, S. J. J. Am. Chem. Soc. 2001, 123,
7831, and references cited therein.
(8) (a) Minta, A.; Kao, J. P. Y.; Tsien, R. Y. J. Biol. Chem. 1989, 264,
8171. (b) Matsuo, T. Biochim. Biophys. Acta 1998, 1379, 178.
(9) (a) Franz, K. J.; Singh, N.; Lippard, S. J. Angew. Chem., Int. Ed.
2000, 39, 2120. (b) Franz, K. J.; Singh, N.; Spingler, B.; Lippard, S. J.
Inorg. Chem. 2000, 39, 4081.
(10) (a) Katayama, Y.; Takahashi, S.; Maeda, M. Anal. Chim. Acta 1998,
365, 159. (b) Mizukami, S.; Nagano, T.; Urano, Y.; Odani, A.; Kikuchi, K.
J. Am. Chem. Soc. 2002, 124, 3920. (c) Soh, N.; Imato, T.; Kawamura, K.;
Maeda, M.; Katayama, Y. Chem. Commun. 2002, 2650.
10.1021/ol0344570 CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/13/2003

Figure 1. Restoration of fluorescence of a linked bis(dansyl-
aminotroponiminate) Co(II) complex by reaction with nitric oxide.
fluorophores such as dansyl,11 coumarin,10c anthracene,1d or
pyrene.5 Although such fluorophores are well suited for
abiotic analysis, they do not have the optimal spectral
properties for in vivo analysis. The most common fluoro-
phores in biological sensors are those in the fluorescein and
rhodamine families. Their relatively long excitation wave-
lengths (g500 nm) and high quantum yields and extinction
coefficients make them attractive components of biological
sensors. The synthesis of highly functionalized fluoresceins
and rhodamines is nontrivial, however, and has received little
attention.
To design new sensors based on the metal coordination
strategy, we have explored the synthesis and properties of
metal chelating fluorescein derivatives. These derivatives are
hybrids of fluoresceins and rhodamines, which we refer to
as rhodafluors.12,13 Our synthetic strategy is to link a metal-
chelating group to a rhodafluor containing a metal donor site.
As shown in Figure 2, we chose an aminodiacetate group as
the chelator14 and a piperazine-functionalized rhodafluor
(“rhodapip”) 1 as the fluorophore.15 The relatively long
linkers in 2 and 3 were chosen to allow the fluorophore
efficiently to remain away from the influence of the metal
center upon deligation.
Figure 2. Structures of rhodapip 1 and its chelator-linked deriva-
tives 2 and 3 and a hypothetical structure of a complex of 3.
This substrate proved to be amenable to Pd-catalyzed
amination; exposure of 6 to Boc-protected piperazine under
Buchwald conditions resulted in isolation of the expected
arylamine 7 in good yield.18 With compound 7 in hand,
construction of a chelator could commence. The nitrile first
was hydrogenated using 5% rhodium on alumina as a
catalyst.19 Without isolation, the resulting primary amine 8
was alkylated with 2 equiv of tert-butyl bromoacetate.20 The
resulting protected aminodiacetic acid derivative 9 was
isolated in good yield after chromatography.
Compound 9 was properly functionalized for participation
in a condensation reaction with the benzophenone derivative
Scheme 1a
The synthesis of target 2 began with known 2-bromo-p-
anisaldehyde 4, obtained by metalation/bromination of p-
anisaldehyde.16 Exposure of 4 to the Wittig reagent derived
from 6-bromohexanenitrile gave the expected bromostyrene
derivative 5 as a mixture of stereoisomers in good yield
(Scheme 1).17 Reduction of 5 using hydrogen and platinum
catalyst furnished arylalkanenitrile 6 in high yield.
(11) Sulowska, H.; Wiczk, W.; Mlodzianowski, J.; Przyborowska, M.;
Ossowski, T. J. Photochem. Photobiol. A 2002, 150, 249.
(12) Burdette, S. C.; Frederickson, C. J.; Bu, W.; Lippard, S. J. J. Am.
Chem. Soc. 2003, 125, 1778.
(13) Also known as (a) rhodols: Whitaker, J. E.; Haugland, R. P.; Ryan,
D.; Hewitt, P. C.; Haugland, R. P.; Prendergast, F. G. Anal. Biochem. 1992,
207, 267, or (b) fluorhods: Smith, G. A.; Metcalfe, J. C.; Clarke, S. D. J.
Chem. Soc. Perkin Trans. 2 1993, 1195.
(14) Schneppensieper, T.; Finkler, S.; Czap, A.; van Eldik, R.; Heus,
M.; Nieuwenhuizen, P.; Wreesmann, C.; Abma, W. Eur. J. Inorg. Chem.
2001, 491.
(15) Design and synthesis of rhodapip: Hilderbrand, S. A.; Lippard, S.
J. Unpublished results
(16) Lear, Y.; Durst, T. Can. J. Chem. 1997, 75, 817.
(17) Paquette, L. A.; Mendez-Andino, J. L. J. Org. Chem. 1998, 63,
9061.
a Conditions: (a) [Ph3P(CH2)5CN]Br, KN(TMS)2, THF. (b) 1
atm H2, 5% Pt/C, EtOH. (c) Boc-Piperazine, NaOtBu, Pd2(dba)3,
2-dicyclohexylphosphino-2′-(N,N-dimethylamino)-biphenyl, tolu-
ene. (d) (i) H2, Rh/Al2O3; (ii) tert-butyl bromoacetate, K2CO3,
CH3CN.
2052
Org. Lett., Vol. 5, No. 12, 2003

10. This condensation, similar to a Friedel-Crafts acylation,
is the common transformation in the synthesis of fluoresceins
and their derivatives.7b Little is known, however, about the
amenability of this reaction to highly functionalized arenes
such as 9.
Scheme 3a
In practice, condensation of 9 with 10 was carried out in
TFA at 120-130° in a sealed vessel (Scheme 2). The
Scheme 2a
a Conditions: (a) Br(CH2)5CN, (Bu4N)OH, 50% aqueous NaOH,
benzene. (b) TFA, CH2Cl2. (c) 3-Bromoanisole, NaOtBu, Pd2(dba)3,
2-dicyclohexylphosphino-2′-(N,N-dimethylamino)-biphenyl, tolu-
ene. (d) (i) NaBH4, CoCl2, MeOH; (ii) tert-butyl bromoacetate,
K2CO3, CH3CN. (e) 10, TFA, 130 °C.
a Conditions: (a) TFA, 130°, 5 days.
reaction proved to be quite sluggish, and complete disap-
pearance of starting material could only be achieved after at
least 5 days of heating. Analysis of the products by mass
spectrometry showed that the desired fluorescein 2 was
indeed formed, as well as an unwanted byproduct, the
structure of which is 11. This byproduct arises from
dealkylation of the aminodiacetate grouping.
The difficulty encountered in the condensation of 9 may
be due to steric interference of the piperazine ring with the
o-alkyl substituent. To alleviate A1,3-strain between the two
aryl substituents, rotation around the aryl-N bond would
orient the piperazine ring orthogonal to the phenyl ring plane
(Scheme 2). In such a conformation, the nitrogen lone pair
would be unable to overlap with the benzene π-system.
Without delocalization of the nitrogen lone pair, the nucleo-
philicity of the benzene ring is reduced. Presumably, a
substrate that lacked this crowding would be better suited
for condensation.21
CH2Cl2, affording the desired product 14 in good yield after
chromatography. Coupling of 14 with 3-bromoanisole pro-
ceeded smoothly under the standard conditions, furnishing
the desired arylamine 15 in 81% yield. As before, we sought
to construct the chelating moiety by reduction of the nitrile
and subsequent amine alkylation. Exposure of 15 to CoCl2
and excess NaBH4 in MeOH, followed by alkylation with
tert-butyl bromoacetate, led to the isolation of 16 in 36%
yield. With compound 16 in hand, we investigated the
previously discussed condensation reaction. As anticipated
for the less congested substrate, condensation of 16 with the
benzophenone 10 in TFA proceeded more quickly than the
analogous reaction of 9. Complete disappearance of starting
material was typically observed after 18-24 h reaction time,
and only traces of the dealkylation product could be
observed.
Because of the difficulty in removing the dealkylation
product 11 from 2, we concentrated on studying the metal-
binding properties of 3. Titration of 3 with metal halides or
sulfates was investigated by fluorescence spectroscopy. All
titrations were conducted in pH 7.5 10 mM PIPES containing
100 mM KCl. Representative fluorescence spectra for
titration with Co(II) are shown in Figure 3. The results of
these titrations are summarized in Table 1. To confirm that
the chelator was responsible for the quenching effect,
titrations were performed with rhodapip 1, the nonchelating
As illustrated in Scheme 3, our approach to 3 began with
differentially protected piperazine-2-methanol 12.22 We
sought to attach a linker to the hydroxyl group of 12 through
an ether functionality. Accordingly, substrate 12 was alky-
lated with bromohexanenitrile under biphasic conditions.23
Deprotection of product 13 was achieved by use of TFA in
(18) Zhang, X.-X.; Buchwald, S. L. J. Org. Chem. 2000, 65, 8027.
(19) Freifelder, M. J. Am. Chem. Soc. 1960, 82, 2386.
(20) Achilefu, S.; Wilhelm, R. R.; Jimenez, H. N.; Schmidt, M. A.;
Srinivasan, A. J. Org. Chem. 2000, 65, 1562.
(21) A related compound, N-(6-acetamido-3-hydroxyphenyl)piperazine
was completely inert to condensation. Clark, M. A.; Lippard, S. J.
Unpublished results.
(22) Take, K.; Kasahara, C.; Shigenaga, S.; Azami, H.; Eikyu, Y.; Nakai,
K.; Morita, M. PCT Int. Appl. WO 0255518, 2002.
(23) Chiou, A.; Markidis, T.; Constantinou-Kokotou, V.; Verger, R.;
Kokotos, G. Org. Lett. 2000, 2, 347.
Org. Lett., Vol. 5, No. 12, 2003
2053

Table 1. Metal Ion Concentration at 50% Emission and
Quenching Ability of Metal Complexes of 3 (1 µM)
metal ion
EC50 (µM)
% quenching
Mn2+
Co2+
Ni2+
Cu2+
Zn2+
60
6
2
0.4
2
60
60
50
90
35
with various transition metals, notably Cu(II).11,26,27 Both
electron and energy transfer mechanisms have been proposed.
Further structural and photophysical studies may clarify the
mechanism of quenching.
In conclusion, the present work demonstrates that rhodapip-
based fluorionophores are synthetically accessible and have
favorable photophysical and metal-binding properties. Since
excess metal ion is required to achieve full fluorescence
quenching, future work will focus on chelating groups with
higher denticities such as azamacrocycles. We anticipate that
such systems will be well suited for a variety of sensing
applications in which displacement of the fluorophore by
an analyte restores fluorescence according to the strategy of
Figure 1.
Figure 3. Fluorescence spectra during titration of 3 (1 µM) with
CoCl2 in pH 7.5 10 mM PIPES, 100 mM KCl. Excitation at 513
nm.
parent of 2 and 3. No quenching was observed under these
conditions. Additionally, we observed complete restoration
of fluorescence when EDTA was added at the end of a metal
titration. This result confirms that only metal ions available
for coordination are able to quench the fluorescence.
We observed no change in the absorbance spectrum of
the ligand upon metal addition, nor was there any change in
the shape of the emission band. That the quenching is
dependent on availability of half-filled d-orbitals is supported
by the fact that all the metal ions studied except Zn(II) were
efficiently able to quench the fluorescence.24 This result
would seem to be consistent with a ligand-to-metal energy
transfer quenching mechanism. The efficiency of energy
transfer, however, would be a function of the overlap of the
fluorophore emission band with the metal excitation band.25
The fact that there is little difference in quenching efficiency
between the d5-d8 metals, despite appreciable differences
in the energy and extinction coefficient of their d-d
absorptions, seems to indicate that energy transfer may not
be important. Instead, metal-to-ligand electron transfer may
be responsible for the quenching. In this case, the weak
quenching ability of Zn(II) would be explained by its stability
toward oxidation. Similarly constituted conjugated fluori-
onophores have displayed comparable quenching efficiencies
Acknowledgment. This work was supported by the
National Institute of General Medical Sciences (GM-65519)
and the National Science Foundation (CHE-0234951). M.A.C.
is an NRSA Fellow (GM66501-01). K.D. received funding
from the MIT Undergraduate Research Opportunities Pro-
gram John Reed Fund and from an MIT Undergraduate
Research Fellowship in Macromolecular Interactions. The
MIT DCIF NMR spectrometer was funded through NSF
Grant CHE-9808061. We thank Carolyn Woodroofe for
supplying compound 10.
Supporting Information Available: All experimental
1
procedures and analytical data, H NMR of all synthetic
intermediates, and procedures for fluorescence titrations. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0344570
(24) Similar effects in the absorption and emission spectra and indiffer-
ence to Zn(II) have been previously observed: Klein, G.; Kaufmann, D.;
Schu¨rch, S.; Reymond, J.-L. Chem. Commun. 2001, 561.
(25) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum
Press: New York, 1983.
(26) (a) Fabrizzi, L.; Licchelli, M.; Pallavicini, P.; Perotti, A.; Taglietti,
A.; Sacchi, D. Chem. Eur. J. 1996, 2, 75. (b) Fabrizzi, L.; Licchelli, M.;
Pallavicini, P.; Perodi, L. Angew. Chem., Int. Ed. 1998, 37, 800.
(27) Prodi, L.; Bolletta, F.; Montalti, M.; Zaccheroni, N. Eur. J. Inorg.
Chem. 1999, 455.
2054
Org. Lett., Vol. 5, No. 12, 2003