Fluorescent, siderophore-based chelators. design and synthesis of a trispyrenyl trishydroxamate ligand, an intramolecular excimer-forming sensing molecule which responds to iron(III) and gallium(III) metal cations

Article abstract of DOI:10.1021/jo960191h

Two pyrene-labeled hydroxylamines, 5-Bn and 5-Bz, O-protected with the benzyl and the benzoyl group, respectively, have been prepared for the generation of siderophore-based new chelators incorporating both the pyrene chromophore and the hydroxamic acid functionality. 5-Bz formed the starting point toward the synthesis of the tripod-shaped trishydroxamate, 1. That trichromophoric ligand displays remarkable fluorescence emission properties (dual emission: "monomer" and excimer type) which are markedly and selectively modified by binding Fe(NO₃)₃ and Ga(NO₃)₃. Ferric ions induce a quasi total quenching of the pyrene fluorescence, whereas the nonquenching Ga(III) cations are observed to affect the value of the excimer-to-monomer fluorescence intensity ratio. Ethylenediaminetetraacetic acid (EDTA) competition reactions yielded an estimated value of 3.8 for log K* of the complex LFe in methanol/water (80/20 v/v), where K* = ([LFe][H⁺]³)/([LH₃] [Fe³⁺]) and L is the ligand in its totally deprotonated form. Compound 1 is the prototype of a new class of photoresponsive molecular systems which could act as sensitive probes for metal cation detection and recognition.

Full text of DOI:10.1021/jo960191h

3956
J . Org. Chem. 1996, 61, 3956-3961
F lu or escen t, Sid er op h or e-Ba sed Ch ela tor s. Design a n d Syn th esis
of a Tr isp yr en yl Tr ish yd r oxa m a te Liga n d , a n In tr a m olecu la r
Excim er -F or m in g Sen sin g Molecu le Wh ich Resp on d s to Ir on (III)
a n d Ga lliu m (III) Meta l Ca tion s
Fre´de´ric Fages,* Bruno Bodenant, and Tanja Weil
Laboratoire de Photophysique et Photochimie Mole´culaire, CNRS URA 348, 351 Cours de La Libe´ration,
33405 Talence Cedex, France
Received J anuary 30, 1996X
Two pyrene-labeled hydroxylamines, 5-Bn and 5-Bz, O-protected with the benzyl and the benzoyl
group, respectively, have been prepared for the generation of siderophore-based new chelators
incorporating both the pyrene chromophore and the hydroxamic acid functionality. 5-Bz formed
the starting point toward the synthesis of the tripod-shaped trishydroxamate, 1. That trichro-
mophoric ligand displays remarkable fluorescence emission properties (dual emission: “monomer”
and excimer type) which are markedly and selectively modified by binding Fe(NO3)3 and Ga(NO3)3.
Ferric ions induce a quasi total quenching of the pyrene fluorescence, whereas the nonquenching
Ga(III) cations are observed to affect the value of the excimer-to-monomer fluorescence intensity
ratio. Ethylenediaminetetraacetic acid (EDTA) competition reactions yielded an estimated value
of 3.8 for log K* of the complex LFe in methanol/water (80/20 v/v), where K* ) ([LFe][H+]3)/([LH3]
[Fe3+]) and L is the ligand in its totally deprotonated form. Compound 1 is the prototype of a new
class of photoresponsive molecular systems which could act as sensitive probes for metal cation
detection and recognition.
There is an increasing interest in the synthesis of
Whereas a broad effort has been devoted over the last
decade to the synthesis and the coordination studies of
photosensitive ligands that respond to alkali- and alkaline-
earth-metal ions,1,2,7 it is only recently that a few
examples of fluorescent chemosensors for transition
metal ions and heavy metal ions2c,4b,8 have been reported
in the literature. Actually, the design of photoresponsive
chelators enabling the detection, but also the speciation,9
of these metals is a task of prime importance with respect
to the essential role of these species in biological and/or
environmental media.
fluorescent receptors that are endowed with both metal
cation recognition and optical signaling properties.1
These are photoresponsive molecular or supramolecular
systems that contain a chromophore center in close
vicinity to a cation binding site, and that are designed2
to spectrofluorimetrically detect the complexation of a
guest cation. Given the inherent sensitivity of fluores-
cence, the use of such molecular probes is particularly
attractive for the development of a new generation of
optical methods that could enable the real-time, continu-
ous, in situ monitoring of trace metals in various media.3
One of the most successful practical applications of
chemosensors for metal ions has been in biological
research,1,4 and for instance the decisive contribution of
Tsien to intracellular calcium determination5 is to be
particularly emphasized. Besides, the growing need for
such methods in environmental chemistry and oceanog-
raphy represents an appealing direction for the organic
chemist.6
In that connection, we undertook the design of new
fluorescent receptors for the optical recognition of highly
charged metal cations, e.g. transition metal such as Fe3+
,
actinide elements, and group 13 metal ions such as Ga3+
and Al3+ 9,6b Recently, we reported10 the synthesis of the
.
iron(III) and gallium(III) complexes of ligand 3, and their
fluorescence emission features in acetonitrile were shown
to depend remarkably on the nature of the chelated
metal. It thus occurred to us that compound 1 (Scheme
1), a trichromophoric tripodal hydroxamate ligand, could
X Abstract published in Advance ACS Abstracts, May 15, 1996.
(1) Fluorescent Chemosensors for Ion and Molecule Recognition;
Czarnik, A. W., Ed.; ACS Symposium Series No. 538; American
Chemical Society: Washington, DC, 1992.
(2) (a) Valeur, B. In Topics in Fluorescence Spectroscopy; Lakowicz,
J . R., Ed.; Plenum Press: New York, 1994; Vol. 4, pp 21-48. (b)
Fabbrizzi, L.; Poggi, A. Chem. Soc. Rev. 1995, 197. (c) Czarnik, A. W.
Acc. Chem. Res. 1994, 27, 302. (d) Bissel, R. A.; de Silva, A. P.;
Gunaratne, H. Q. N.; Lynch, P. L. M.; Maguire, G. E. M.; Sandanayake,
K. R. A. S. Chem. Soc. Rev. 1992, 187. (e) Desvergne, J .-P.; Fages, F.;
Bouas-Laurent, H.; Marsau, P. Pure Appl. Chem. 1992, 64, 1231. (f)
Lo¨hr, H.-G.; Vo¨gtle, F. Acc. Chem. Res. 1985, 18, 65.
(3) Fiber Optic Chemical Sensors and Biosensors; Wolfbeis, O. S.,
Ed.; CRC Press: Boca Raton, FL, 1991; Vols. I and II.
(4) (a) Haugland, R. P. Handbook of Fluorescent Probes and Research
Chemicals, 5th ed.; Molecular Probes: Eugene, OR, 1992. (b) Loyevsky,
M.; Lytton, S. D.; Mester, B.; Libman, J .; Shanzer, A.; Cabantchik, Z.
I. J . Clin. Invest. 1993, 91, 218.
(7) Fages, F.; Desvergne, J .-P.; Bouas-Laurent, H.; Lehn, J .-M.;
Barrans, Y.; Marsau, P.; Meyer, M.; Albrecht-Gary, A.-M. J . Org. Chem.
1994, 59, 5264.
(8) (a) Parker, D.; Williams, J . A. G. J . Chem. Soc., Perkin Trans. 2
1995, 1305. (b) Vaidya, B.; Zak, J .; Bastiaans, G. J .; Porter, M. D. Anal.
Chem. 1995, 67, 4101. (c) Fabbrizzi, L.; Licchelli, M.; Pallavicini, P.;
Perotti, A.; Sacchi, D. Angew. Chem., Int. Ed. Engl. 1994, 33, 1975.
(d) Ko¨nig, K.-H.; Bosslet, J .; Holzner, C. Chem. Ber. 1989, 122, 59. (e)
Sasaki, D. Y.; Shnek, D. R.; Pack, D. W.; Arnold, F. H. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 905. (f) Schuster, M.; Unterreitmaier, E.
Fresenius J . Anal. Chem. 1993, 346, 630. (g) Barrero, J . M.; Moreno-
Bondi, M. C.; Pe´rez-Conde, M. C.; Camara, C. Talanta 1993, 40, 1619.
(h) Fages, F.; Desvergne, J .-P.; Bouas-Laurent, H.; Marsau, P.; Lehn,
J .-M.; Kotzyba-Hibert, F.; Albrecht-Gary, A.-M.; Al-J oubbeh, M. J . Am.
Chem. Soc. 1989, 111, 8672. (i) Andrianatoandro, H.; Barrans, Y.;
Marsau, P.; Desvergne, J .-P.; Fages, F.; Bouas-Laurent, H. Acta
Crystallogr. 1995, 51(B), 293.
(5) (a) Tsien, R. Y. in ref 1 pp 130-146. (b) Tsien, R. Y. Chem. Eng.
News 1994, 72, 34.
(6) (a) Thompson, R. B.; J ones, E. R. Anal. Chem. 1993, 65, 730. (b)
J ohnson, K. S.; Coale, K. H.; J annasch, H. W. Anal. Chem. 1992, 64,
1065A. (c) Saari, L. A.; Seitz, W. R. Anal. Chem. 1983, 55, 667.
(9) (a) Wells, M. L.; Price, N. M.; Bruland, K. W. Mar. Chem. 1995,
48, 157. (b) Motekaitis, R. J .; Martell, A. E. Mar. Chem. 1987, 21, 101.
(c) Florence, T. M. Talanta 1982, 29, 345.
(10) Bodenant, B.; Fages, F. Tetrahedron Lett. 1995, 1451.
S0022-3263(96)00191-0 CCC: $12.00 © 1996 American Chemical Society

Fluorescent, Siderophore-Based Chelators
J . Org. Chem., Vol. 61, No. 12, 1996 3957
Sch em e 1. Tr isp yr en yl Tr ish yd r oxa m a te
Ch ela tor 1a
response of ligand 1 to iron complexation was expected
to be a significant quenching of the fluorescence inten-
sity.10
On the other hand, attaching a pyrenyl group to each
of the tripodal chain extremities would lead to the
generation of a conformationally-flexible trichromophoric
molecular system whose fluorescence emission spectrum
should display in solution, even at very low concentra-
tions, both “monomer” and intramolecular excimer con-
tributions.15 Thus, we envisioned that compound 1 might
behave as previously reported bispyrenyl sensors16 that
were shown to give rise to guest-influenced conforma-
tional rearrangements leading to excimer formation or
disparition. If such were the case, these features could
be particularly attractive in view of using 1 for detecting
also nonquenching metal ions such as Ga3+ and Al3+
.
Indeed, it has been reported that monomer vs excimer
fluorescence intensity-ratio measurements were particu-
larly well-suited for metal sensing applications.2a,7,8e,16,17
In this paper, we report the synthesis of the tripodlike
receptor, 1, and the synthesis of a monochromophoric
bidentate reference compound, 2, which both involved the
preparation of two pyrene-labeled O-protected hydroxy-
lamine synthons (Scheme 1).18 Furthermore, we present
a preliminary account of the fluorescence emission prop-
erties of these two ligands in solution, in the absence and
in the presence of two metal cations, Fe3+ and Ga3+
,
which were selected as representative examples of quench-
ing and nonquenching metals, respectively.19
a
2 and 3 are the monochromophoric bidentate ligands, and 4
is the nonfluorescent analogue14 of 1.
Resu lts
possess a promising prototype structure according to the
two following main converging design objectives.
Syn th esis. The general stepwise procedure is outlined
in Scheme 2. It involves the condensation of a pyrene-
labeled hydroxylamine, 5-Bn or 5-Bz, with a selected
carboxylic acid chloride derivative. This step is then
followed by the removal of the hydroxamate O-substitu-
ent. By varying the nature of the carboxylic acid precur-
sor in the first step, one can expect to generate so a family
of siderophore-like receptors containing both the masked
N-hydroxy-amide function and the pyrene chromophore.
The synthetic route that was followed for the prepara-
tion of 5-Bn and 5-Bz is given in Scheme 3. Carbamate
alkylation were performed by treatment of 7-Bn 20 and
7-Bz21 with the halide compound22 8 in the presence of
NaH (60% in oil) in anhydrous DMF, leading to 6-Bn
(71% yield) and 6-Bz (79% yield), respectively. Treat-
ment of 6-Bn with trifluoroacetic acid (in CH2Cl2), and
of 6-Bz with anhydrous HCl (in dioxane) afforded the
corresponding N-deprotected derivatives in almost quan-
titative yields. Due to their inherent reactivity and
On one hand, trishydroxamate siderophores are natu-
rally occurring and synthetic ion chelators11 that exhibit
a high affinity toward acidic metal cations,12 with a
marked selectivity for Fe3+. This property was observed
for tripod-shaped abiotic siderophores,11,13 and particu-
larly for compound 4, the nonfluorescent analogue of 1,
synthesized by Martell et al.,14 in which three terminal
methyl groups replace the pyrenylmethyl moieties (Scheme
1). Therefore, a synthetic strategy based on a biomimetic
approach, e.g. consisting of the incorporation of sidero-
phore-like complexing structures in fluoroionophore mol-
ecules, was expected to afford systems displaying the
remarkable features needed for iron(III) sensing applica-
tions such as high sensitivity and selectivity. It is worth
to mention that during our investigations promising
results were reported in the recent studies of two series
of monochromophoric linear trishydroxamate sidero-
phores, e.g. naturally-fluorescent pyoverdins8g and fluo-
rescent derivatives4b of desferrioxamine. As iron(III) is
an inherently-quenching metal (vide infra), the optical
(15) (a) Winnik, F. M. Chem. Rev. 1993, 93, 587. (b) De Schryver,
F. C.; Collart, P.; Vandendriessche, J .; Goedeweeck, R.; Swinnen, A.;
Vander Auweraer, M. Acc. Chem. Res. 1987, 20, 159. (c) Reynders, P.;
Ku¨hnle, W.; Zachariasse, K. A. J . Am. Chem. Soc. 1990, 112, 3929.
(16) (a) Aoki, I.; Harada, T.; Sasaki, T.; Kawahara, Y.; Shinkai, S.
J . Chem. Soc., Chem. Commun. 1992, 1341. (b) Takeshita, M.; Shinkai,
S. Chem. Lett. 1994, 125. (c) Aoki, I.; Kawabata, H.; Nakashima, K.;
Shinkai, S. J . Chem. Soc., Chem. Commun. 1991, 1771.
(17) Marquis, D.; Desvergne, J .-P. Chem. Phys. Lett. 1994, 230, 131.
(18) For a preliminary account on the utilization of that synthetic
strategy for the synthesis of 3, see ref 10.
(19) Complexation studies of siderophores with these two metal
cations are known to provide complementary structural informations.
See ref 13i.
(20) Lee, B. H.; Miller, M. J . J . Org. Chem. 1983, 48, 24.
(21) (a) Carpino, L. A.; Giza, C. A.; Carpino, B. A. J . Am. Chem.
Soc. 1959, 81, 955. (b) Wathen, S. P.; Czarnik, A. W. J . Org. Chem.
1992, 57, 6129.
(11) (a) Raymond, K. N. Coord. Chem. Rev. 1990, 105, 135. (b)
Raymond, K. N.; Mu¨ller, G.; Matzanke, B. F. Top. Current Chem. 1984,
123, 49.
(12) Hancock, R. D.; Martell, A. E. Chem. Rev. 1989, 89, 1875.
(13) (a) Dayan, I; Libman, J .; Agi, Y.; Shanzer, A. Inorg. Chem. 1993,
32 1467. (b) Stack, T. D. P.; Karpishin, T. B.; Raymond, K. N. J . Am.
Chem. Soc. 1992, 114, 1512. (c) Baret, P.; Be´guin, C.; Gaude, D.; Gellon,
G.; Mourral, C.; Pierre, J .-L.; Serratrice, G.; Favier, A. Tetrahedron
1994, 50, 2077. (d) Akiyama, M.; Katoh, A.; Ogawa, T. J . Chem. Soc.,
Perkin Trans. 2 1989, 1213. (e) Esteves, M. A.; Vaz, M. C. T.; Gonc¸alves,
M. L. S. S.; Farkas, E.; Santos, M. A. J . Chem. Soc., Dalton Trans.
1995, 2565. (f) Miller; M. J . Chem. Rev. 1989, 89, 1563. (g) Okhanda,
J .; Katoh, A. Tetrahedron 1995, 51, 12995. (h) Baret, P.; Be´guin, C.
G.; Boukhalfa, H.; Caris, C.; Laulhe`re, J .-P.; Pierre, J .-L.; Serratrice,
G. J . Am. Chem. Soc. 1995, 117, 9760. (i) Tor, Y.; Libman, J .; Shanzer,
A.; Felder, C. E.; Lifson, S. J . Am. Chem. Soc. 1992, 114, 6661.
(14) Motekaitis, R. J .; Sun, Y.; Martell, A. E. Inorg. Chem. 1991,
30, 1554.
(22) Georgescaud, D.; Desmare`z, J . P.; Lapouyade, R.; Babeau, A.;
Richard, H.; Winnik, M. A. Photochem. Photobiol. 1980, 31, 539.

3958 J . Org. Chem., Vol. 61, No. 12, 1996
Fages et al.
Sch em e 2. Syn th esis of Liga n d s 1 a n d 2a
F igu r e 1. Corrected fluorescence emission spectra (λexc ) 340
nm) of the free ligands 1 and 2 and of 1 in the presence of 1
mol equiv of Fe(NO3)3 in nondegassed solution (MeOH/H2O
80/20 (v/v), concn ca. 5 × 10-7 M, -log[H+] ca. 7.4 (HEPES,
10 mM), µ ) 0.1 M (NaNO3), 25 °C).
reagent24 were also unsuccessful. From these experi-
ments, it occurred to us that the benzyl substituent as
O-protecting group was of limited utility in the case of
trishydroxamates bearing bulky aromatic moieties. There-
fore, the benzoyl group was chosen because of its stability
towards planned reactions and yet its ease of removal
by methanolic ammonia at room temperature.25 Indeed,
reaction of 1-Bz with ammonia in methanol allowed to
obtain 1 in relatively good yield and with the high purity
required in order to carry out fluorimetric measurements.
All new compounds described here gave analytical and
spectral data in agreement with their structure. In
particular, 250-MHz proton NMR spectroscopy fully
confirmed the structure of the free ligands, 1 and 2, in
solution at room temperature. Furthermore, the 1H
NMR spectrum of 1 in DMSO-d6 showed that the
structure of that ligand is symmetrical on the NMR time
scale, suggesting a random orientation of each of the
tripod chains.
a
(a) CH3COCl, Et3N, toluene; (b) H2, 10% Pd/C, THF/MeOH;
(c) 9, Et3N, benzene; (d) 9, pyridine, THF; (e) H2, 10% Pd/C, THF/
MeOH, or BF3-Et2O, EtSH, CH2Cl2; (f) NH3 gas, MeOH/CH2Cl2.
Sch em e 3. Syn th esis of th e Syn th on s 5-Bn a n d
5-Bza
F lu or escen ce Em ission . Compound 1 exhibited in
solution (MeOH or MeOH/H2O 80/20 (v/v), concn < 10-6
M) a fluorescence spectrum composed of the monomerlike
emission band, and of a broad intense red-shifted band
with an emission maximum at 485 nm (Figure 1). The
latter was attributed to the emission of the pyrene
excimer arising from the intramolecular interaction of a
pyrene chromophore in its singlet excited state with a
second one in the ground state.26 In contrast, the
monochromophoric reference compound, 2, displayed only
the monomer structured fluorescence spectrum charac-
teristic of a 1-substituted derivative of pyrene (Figure 1).
The addition of 1 equiv of iron(III) nitrate to the
solution of 1 was observed to cause the quantitative
quenching of the fluorescence emission of both monomer
and excimer species (Figure 1). The fluorescence emis-
sion properties of a solution of 1-ethylpyrene were not
modified by the presence of 1 molar equiv of iron(III)
salts, which indicated that the deactivation of the ligand
excited state by diffusion-controlled intermolecular reac-
tion with Fe3+ was not operative under these conditions.
a
(a) 8, NaH, DMF; (b) TFA, CH2Cl2; (c) HCl gas, dioxane.
photosensitivity, the N-pyrenylmethyl-O-protected-hy-
droxylamines, 5-Bn and 5-Bz, were used directly without
further purification.
The condensation of 5-Bn with acetyl chloride in the
presence of triethylamine afforded 2-Bn (56% yield).
Similarly, 5-Bn , and 5-Bz were condensed with the
triacid trichloride derivative,23 9, and the corresponding
O-protected tripodal ligands, 1-Bn (17% yield) and 1-Bz
(23% yield), were obtained, respectively (Scheme 2).
The standard treatment of 2-Bn (H2, Pd/C, in metha-
nol), in order to effect the reductive debenzylation,
generated the desired ligand, 2, cleanly. On the other
hand, the same method totally failed to afford the
O-deprotected trishydroxamate, 1, from 1-Bn . The cata-
lytic hydrogenolysis of 1-Bn was observed to generate a
complex mixture containing partially deprotected com-
pounds and also products resulting from the hydrogena-
tion of the aromatic moieties. Furthermore, attempts to
cleave selectively the benzyl ether groups by treatment
of 1-Bn with a boron trifluoride etherate-ethanethiol
(24) Mulqueen, G. C.; Pattenden, G.; Whiting, D. A. Tetrahedron
1993, 49, 9137.
(25) (a) Mitchell, M. S.; Walker, D.-L.; Whelan, J .; Bonisch, B. Inorg.
Chem. 1987, 26, 396. (b) Rodgers, S. J . Raymond, K. N. J . Med. Chem.
1983, 26, 439.
(26) Birks, J . B. Photophysics of Aromatic Molecules; Wiley-
Interscience: London, 1970.
(23) Sun, Y.; Martell, A. E. Tetrahedron 1990, 46, 2725.

Fluorescent, Siderophore-Based Chelators
J . Org. Chem., Vol. 61, No. 12, 1996 3959
Addition of a great excess of desferrioxamine B methyl-
sulfonate (Desferal) to the solution of 1 containing 2 equiv
of iron(III) cations induced the reversed effect, restoring
the original emission of the free ligand. The observed
fluorescence quenching was ascribed to the formation a
complex in the ground state between 1 and ferric cations.
Indeed, the UV-visible absorption spectrum of 1 in the
presence of an equimolar amount of iron in methanol
(conc ca. 5 × 10-5 M) showed an absorption (λmax ) 470
nm, ꢀmax ) 2700 M-1cm-1) which could be indicative of
the formation of a monoprotonated 1‚FeH+ complex as
observed14 by Martell for the nonfluorescent receptor 4.
Addition of a base in excess (pyridine or sodium acetate)
to that solution induced a blue shift of that absorption
band which pointed then at 435 nm. Furthermore, UV-
visible absorption titration of 1 with Fe(NO3)3 in metha-
nol established 1:1 binding stoichiometry.
As expected from the results reported for previous
examples of fluorescent siderophores,4b,8g,27 the fluores-
cence emission of receptor 1 is markedly quenched upon
complexation of Fe3+ cations. Actually, quenching of
electronically excited state of aromatic hydrocarbons by
ferric ions or iron(III) chelates is a known phenomenom
that has been the subject of extensive investigations. It
has been suggested that two main pathways could
account for the efficient radiationless deactivation of the
singlet excited state, i.e. electron transfer from the excited
aromatic chromophore to the metal and/or to energy
transfer from the excited pyrene to low-lying metal
centered energy states.28 Such processes could be par-
ticularly effective in the complex of 1 with Fe3+, due
probably to the fact that the chelated metal cation is held
very close to the excited chromophore.
F igu r e 2. Corrected fluorescence emission spectra (λexc ) 340
nm) of chelator 1 in the presence of gradual amounts of Ga-
(NO3)3 in methanol (non-degassed, 25 °C). [1]0 ) 1.3 × 10-5
M. [Ga3+
]
x 105 M: 1, 0; 2, 2.6; 3, 13; 4, 26; 5, 52. Inset:
0
spectrofluorimetric titration curve.
chains would experience a poorer overlap and the mono-
mer emission would be favored at the expense of that of
the excimer. This interpretation lies on the assumption
that conformational changes of the receptor molecule are
associated with the complexation processes.7,15,16 It is to
be mentioned that, consistently with published observa-
tions for ferrichrome tripodal models,13a the 250-MHz 1H
NMR spectrum of 1 was strongly altered by the addition
of slightly over 1 equiv of Ga(NO3)3. In methanol/water
(80/20 (v/v)), the value of the IE/IM ratio was found to be
equal to 17 and 11 for the free ligand and the ligand in
the presence of Ga3+, respectively. This would indicate
a greater trend for the pyrene chromophores to interact
intramolecularly in the excited state in both the free and
the complex ligand in aqueous solution relative to neat
methanol.15a
The thermodynamic stability constant of the ferric
complex of ligand 1 was estimated by spectrofluorimetric
competitions against EDTA in methanol/water (80/20 v/v)
at -log[H+] ) 4.29 Since the solubility of 1 was too low
in aqueous methanol to allow the determination of its
protonation constants, these experiments yielded the
proton-dependent formation constant K*:30
In contrast to the case of Fe3+, the addition of Ga3+
cations to solutions of receptor 1 was not followed by the
quenching of the total fluorescence emission. Instead, a
marked modification of the shape of the spectrum was
observed. In methanol, the value of the ratio of the
fluorescence intensities of excimer and monomer (IE/IM)
,
that equals 14 for the free ligand, dropped off to about 4
upon addition of two molar equivalents of Ga(NO3)3. The
fluorimetric titration of ligand 1 in methanol with Ga3+
(Figure 2) illustrated well the pronounced effect of that
trivalent metal ion on the IE/IM ratio. With increased
Ga3+ concentration, the excimer emission intensity de-
creased while the monomer emission increased, and the
presence of an isoemissive point was detected at ca. 430
nm. Addition of an excess of Desferal to the solution of
1 containing 2 equiv of gallium(III) cations restored
completely the initial spectrum corresponding to that of
the free ligand. From these results and by analogy with
those obtained by Martell with siderophore 4,14 the effect
of gallium might be ascribed to the formation of a 1:1
Ga(III)/ligand complex. Since Ga3+ is not redox active
and does not possess any empty low-energy orbitals, no
quenching interactions are likely to develop between that
metal and the pyrene chromophore in the excited state.
Rather, the chelation of the gallium cation might alter
the conformational properties of the ligand, which could
induce a rigidification of the tripodal structure and
impede the intramolecular excimer formation. In the
chelate, the pyrene nuclei at the extremities of the tripod
LH3
K* ) â110FeL/â013
where âmlh ) [FemLlHn]/([Fe]m[L]l[H]n), Fe ) ferric ion, L
) ligand 1 in its deprotonated form, H ) hydrogen ion.
A value of 3.8 was obtained for log K*. It is interesting
to note that this value is higher than that of the related
ligand 414 and in agreement with those observed for tris-
(hydroxamate) ligands30,31 exhibiting effective iron bind-
ing abilities. However, it should be mentioned that the
use of different experimental conditions in our study (e.g.
solvent systems of lower polarity and of decreased
solvation ability) renders the comparison with literature
data difficult. Preliminary experiments showed that the
proton-dependent stability constant of the gallium com-
plex was 2 to 3 orders of magnitude smaller than that of
the ferric complex.32
(29) L'Eplattenier, F.; Murase, I.; Martell, A. E. J . Am. Chem. Soc.
1967, 89, 837.
(30) Wong, G. B.; Kappel, M. J .; Raymond, K. N.; Matzanke, B.;
Winkelmann, G. J . Am. Chem. Soc. 1983, 105, 810.
(31) Ng, C. Y.; Rodgers, S. J .; Raymond, K. N. Inorg. Chem. 1989,
28, 2062.
(27) Chae, M.-Y.; Czarnik, A. W. J . Fluoresc. 1992, 2, 225.
(28) (a) Hug, G. L.; Marciniak; B. J . Phys. Chem. 1994, 98, 7523.
(b) Foss, R. P.; Cowan, D. O.; Hammond, G. S. J . Phys. Chem. 1964,
68, 3747. (c) Wilkinson, F.; Farmillo, A. J . Chem. Soc., Faraday Trans.
1976, 72, 604.

3960 J . Org. Chem., Vol. 61, No. 12, 1996
Fages et al.
protonation constants of EDTA in aqueous methanol (80/20
v/v) were taken from literature data.36
In contrast to the behavior of receptor 1, the fluores-
cence emission properties of a solution of the reference
ligand 2 in methanol/water (80/20 (v/v); -log[H+] ) 7.4)
were not affected by the presence of gradual amounts of
iron salts (up to 3 mol equiv). Still spectrophotometric
titrations32,33 confirmed ground-state metal complexation
and indicated the formation of mixtures of iron(III)
chelates with different ligand/metal stoichiometries.
Similarly, the fluorescence of 2 was found to be insensi-
Physical and spectroscopic methods were described else-
where,7 except that FT-IR spectra and fluorescence emission
spectra were recorded on a Perkin Elmer Paragon 1000PC
instrument and on a Hitachi F4500 spectrofluorimeter, re-
spectively.
O-Ben zyl-N-(ter t-bu toxycar bon yl)-N-(1-pyr en yl)h ydr ox-
yla m in e (6-Bn ). To a solution of tert-butyl-N-(benzyloxy)-
carbamate,20 7-Bn , (10.1 g; 45 mmol) in DMF (200 mL) was
added portionwise NaH (60% in oil, 2.2 g; 50 mmol NaH), and
the mixture was stirred under N2 atmosphere over 15 min.
Then a solution of 1-(chloromethyl)pyrene22 8 (11.3 g; 45 mmol)
in DMF (50 mL) was added dropwise at room temperature.
The reaction mixture was heated at 80-90 °C for 3 h and
stirred one additional night at room temperature. The solution
was poured into 500 mL of ice-cooled water. After extraction
with 4 × 250 mL of ethyl acetate, the usual workup was
followed by a chromatography on a SiO2 column, eluting with
CH2Cl2/petroleum ether 70/30 (v:v). 6-Bn was obtained (14
g; 71%) as an orange waxy product. Rf ) 0.64 (CH2Cl2). 1H
NMR (250 MHz, CDCl3): 1.45, 4.38, 5.41, 7.36, 8-8.6 ppm.
MS (EI) m/ z 437 (calcd 437). Anal. Calcd for C29H27NO3: C,
79.61; H, 6.22; N, 3.20; O, 10.97. Found: C, 80.78; H, 6.16;
N, 2.90; O, 10.15.
O-Ben zoyl-N-(ter t-b u t oxyca r b on yl)-N-(1-p yr en yl)h y-
d r oxyla m in e (6-Bz). By the same procedure as above, using
3 g (13 mmol) of tert-butyl-N-(benzoyloxy)carmate,21 7-Bz, and
0.52 g of NaH (60% in oil, 13 mmol of NaH) in DMF (80 mL),
and 3.5 g (14 mmol) of 1-(chloromethyl)pyrene,22 8, in DMF
(50 mL), 6-Bz was obtained as a green-yellow solid (4.5 g,
79%): mp 118-120 °C; Rf ) 0.72 (CH2Cl2). 1H NMR (60 MHz,
CDCl3): 1.5, 5.45, 7.1-8.2 ppm. MS (EI) m/ z 451 (calcd 451).
Anal. Calcd for C29H25NO4: C, 77.14; H, 5.58; N, 3.10.
Found: C, 76.96; H, 5.66; N, 3.12.
tive to the addition of up to 5 mol equiv of Ga3+ 33
These
.
results underlined the necessity of designing molecular
systems that combine both the chelating sites and the
chromophoric subunits in a highly preorganized structure
in order to ensure effective binding properties and
sensitive optical responses to complexation.34
Con clu sion
The main purpose of this paper is to provide a general
synthetic approach toward the synthesis of fluorescent
hydroxamic siderophores incorporating the pyrene chro-
mophore. To this end, the synthesis of two pyrene-
labeled O-protected hydroxylamines, 5-Bn and 5-Bz, is
outlined, which should be versatile building-blocks for
the elaboration of a wide array of fluorescent chelators.
In this report, the O-benzoyl derivative has been used
in order to obtain the new fluorescent tripod-shaped
ligand, 1. That compound is the first example of a
siderophore-based chelator that displays a dual fluores-
cence emission (monomer- and excimerlike) which is
sensitive to the presence of Fe3+ and Ga3+ metal cations.
The optical response is dependent on the nature of the
complexed metallic species. This work is a first step
toward the development of a new class of fluorescent
chemosensors that could be of potential interest for
environmental applications.
O-Ben zyl-N-(1-p yr en yl)h yd r oxyla m in e (5-Bn ). To a
solution of 6-Bn (8 g, 18 mmol) in 15 mL of CH2Cl2 was added
15 mL of trifluoroacetic acid. The reaction mixture was
allowed to stir at room temperature for 4 h and then evapo-
rated to dryness. The residue was dissolved in CH2Cl2 (200
mL), the organic layer was washed with 2 x 200 mL of 10%
aqueous NaHCO3 and then dried over K2CO3. The evaporation
of CH2Cl2 yielded to a pasty orange-yellow product (5.5 g, 89%)
which was used directly without further purification. 1H NMR
(60 MHz, CDCl3): 4.9, 5.5, 7.6, 8-8.7 ppm.
Exp er im en ta l Section
All reagents and solvents (spectrometric grade) were pur-
chased commercially and used without further purification,
except as below. Benzene, DMF, and dichloromethane were
distilled from calcium hydride. THF was refluxed with and
distilled from sodium benzophenone ketyl. Triethylamine and
pyridine were distilled from and stored over KOH pellets. The
purity of all compounds was examined by thin-layer chroma-
tography, using a two-wavelength detection system (365 and
254 nm) and, in the case of the hydroxamate ligands, a
visualization method (FeCl3). 1,1,1-Tris((2-(chlorocarbonyl)-
ethoxy)methyl)ethane,23 9, 1-(chloromethyl)pyrene,22 8, tert-
butyl N-(benzyloxy)carbamate,20 7-Bn , and tert-butyl-N-
(benzoyloxy)carbamate,21 7-Bz, were prepared according to
published procedures. The following abbreviations have been
used: Bn: benzyl; Bz: benzoyl. Fe(NO3)3‚9H2O (98%) and
gallium nitrate (99.999%) were purchased from Aldrich. Stock
solutions of the metals (in ca. 0.1 M HCl) were standardized
with EDTA by using Pyrocatechol violet as an indicator.35
Desferrioxamine B methylsulfonate salt was used as a gift
from Ciba-Geigy and used without further purification. The
O-Ben zoyl-N-(1-p yr en yl)h yd r oxyla m in e (5-Bz).
A
stream of HCl gas was passed through a solution of 6-Bz (4 g,
9 mmol) in dioxane (150 mL), and the course of the reaction
was followed by TLC. After the disparition of the starting
product, the dioxane was evaporated and the residue was
taken up into CH2Cl2. The organic layer was washed with 10%
aqueous Na2CO3 and dried over MgSO4. The evaporation of
the solvent left a green solid (2.8 g, 89%) which was used
without further purification. 1H NMR (60 MHz, CDCl3): 4.7,
7.1-8.1 ppm.
N-(Ben zyloxy)-N-(1-p yr en yl)a ceta m id e (2-Bn ). To a
stirred solution of 5-Bn (5.5 g, 16 mmol) and triethylamine
(10 mL) in dried toluene (100 mL) was added dropwise acetyl
chloride (2 mL) in toluene (50 mL). After the usual workup
and chromatography on silica gel, eluting with CH2Cl2, 2-Bn
was obtained as a white solid (3.4 g, 56%): mp 140 °C, Rf )
0.86 (CH2Cl2/MeOH 98/2 (v/v)). 1H NMR (250 MHz, CDCl3):
2.25, 4.28, 5.49, 7.13, 7.6-8.6 ppm. Anal. Calcd for C26H21
-
NO2: C, 82.30; H, 5.58; N, 3.69; O, 8.43. Found: C, 82.10; H,
5.61; N, 3.59; O, 8.02.
(32) The detailed description of the complexation properties of
ligands 1 and 2 will be reported in a forthcoming paper, together with
photophysical studies.
(33) Consistently, we observed that the pyrene fluorescence emission
of the neutral iron(III) complex of ligand 3, which is very weak in
acetonitrile (see ref 10), was totally restored in methanol. The
photophysical properties of (2)3M(III) and (3)3M(III) (M ) Fe and Ga),
and their dependence on the nature of the solvent, have been
investigated (see ref 34). To be published.
1,1,1-Tr is[[[[N-(ben zyloxy)-N-(1-pyr en ylm eth yl)am in o]-
ca r bon yl]eth oxy]m eth yl] eth a n e (1-Bn ). To a solution of
5-Bn (3 g, 9 mmol) and triethylamine (10 mL) in benzene (200
mL) at room temperature was added a solution of triacid
chloride23 9 (1.1 g, 3 mmol) in benzene (50 mL) over a period
of 1 h, under a nitrogen atmosphere and with vigorous stirring.
The mixture was allowed to stir an additional 48 h at room
temperature. After washing with 3 x 200 mL water, the
(34) Bodenant, B. The`se No. 1325, Universite´ de Bordeaux I,
Talence, France, 1995.
(35) Schwarzenbach, G.; Flaschka, H. Die Komplexometrische Ti-
tration; Verlag Enke: Stuttgart, 1965.
(36) Rorabacher, D. B.; MacKellar, W. J .; Shu, F. R.; Bonavita, S.
M. Anal. Chem. 1971, 43, 561.

Fluorescent, Siderophore-Based Chelators
J . Org. Chem., Vol. 61, No. 12, 1996 3961
reaction mixture was dried over Na2SO4 and evaporated to
dryness. Chromatography of the crude product on a silica gel
column, eluting with CH2Cl2/MeOH 98/2 (v/v), followed by a
crystallization in CH2Cl2/petroleum ether mixtures yielded to
the tris-O-protected receptor 1-Bn (0.64 g, 17%) as a pale
yellow solid: mp 120 °C, Rf ) 0.31 (CH2Cl2/MeOH 98/2 (v/v)).
1H NMR (250 MHz, CDCl3): 1.00, 2.82, 3.39, 3.82, 4.45, 5.53,
7.0-8.4 ppm. FT-IR (KBr pellet): 1650 cm-1. MS (FAB) m/ z
1294.5 (calcd for M + H+ 1294.5); 1316.1 (calcd for M + Na+
1316.5). Anal. Calcd for C86H75N3O9‚H2O: C, 78.70; H, 5.91;
N, 3.20. Found: C, 78.79; H, 6.10; N, 3.39.
1,1,1-t r is[[[[N-(Ben zoyloxy)-N-(1-p yr en ylm et h yl)a m i-
n o]ca r bon yl]eth oxy]m eth yl]eth a n e (1-Bz). Similarly, us-
ing 5-Bz (4 g, 11.4 mmol) and pyridine (3 mL) in THF (50 mL),
and 923 (1.1 g, 3 mmol) in THF (20 mL), 1-Bz was obtained as
a pasty yellow solid (0.9 g, 23%): Rf ) 0.71 (CH2Cl2/MeOH
98/2 (v/v)). 1H NMR (250 MHz, CDCl3): 1.28, 2.56, 3.17, 3.66,
5.71, 7.2-8.4 ppm. FT-IR (KBr pellet): 1750, 1660 cm-1. MS
(FAB) m/ z 1336.5 (calcd for M + H+ 1336.5), 1358.5 (calcd for
M + Na+ 1358.5). Anal. Calcd for C86H69N3O12‚H2O: C, 76.25;
H, 5.28; N, 3.10. Found: C, 76.02; H, 5.27; N, 3.01.
N-Hyd r oxy-N-(1-p yr en yl)a ceta m id e (2). 2-Bn (0.7 g, 1.8
mmol) was dissolved in 80 mL of the THF/MeOH (50/50 v/v)
solvent system, and 10% Pd/C (0.5 g) was added. The
hydrogenation was carried out for 4 h at room temperature
and atmospheric pressure. The filtration of the reaction
mixture afforded a white solid which was crystallized in hot
CH2Cl2 to give the free monohydroxamate 2 (0.45 g, 84%): mp
215-220 °C; Rf ) 0.48 (CH2Cl2/ethyl acetate 50/50 (v/v)). 1H
NMR (250 MHz, DMSO-d6): 2.10, 5.42, 8.0-8.5, 9.89 ppm. FT-
IR (KBr pellet): 1605, 1590 cm-1. MS (FAB) m/ z 290.1 (calcd
for M + H+ 290.1), 312.0 (calcd for M + Na+ 312.1). Anal.
Calcd for C19H15NO2: C, 78.87; H, 5.22; N, 4.84; O, 11.06.
Found: C, 78.05; H, 5.51; N, 4.97; O, 11.00.
1,1,1-tr is[[[[N-Hydr oxy-N-(1-pyr en ylm eth yl)am in o]car -
bon yl]eth oxy]m eth yl] eth a n e (1). The tris-O-protected
ligand 1-Bz (0.7 g, 5.2 mmol) was dissolved in 100 mL of the
CH2Cl2/MeOH (20/80 v/v) solvent system at room temperature
and the solution was saturated with NH3 gas for 30 mn. The
slightly cloudy solution was concentrated and left to stand at
0-5°C for 2 h. A beige solid was deposited, which was filtrated
and dried under vacuum over P2O5, yielding 1 (0.2 g, 37%):
mp 145-150 °C; Rf ) 0.58 (CH2Cl2/MeOH 90/10 (v/v)). 1H
NMR (250 MHz, DMSO-d6): 0.83, 2.69, 3.22, 3.64, 5.40, 7.9-
8.5, 9.75 ppm. FT-IR (KBr pellet): 1640 cm-1. MS (FAB) m/ z
1024.5 (calcd for M + H+ 1024.4), 1046.6 (calcd for M + Na+
1046.4). Anal. Calcd for C65H57N3O9‚H2O: C, 74.91; H, 5.71;
N, 4.03. Found: C, 74.96; H, 5.66; N, 3.92.
Ack n ow led gm en t. The authors thank Dr. A.-M
Albrecht-Gary, Prof. J .-L. Pierre, and Prof. A. W.
Czarnik for stimulating discussions. The CNRS, the
University of Bordeaux I, and La Re´gion Aquitaine are
gratefully acknowledged for financial assistance. B.B.
is grateful to the CNRS and La Re´gion Aquitaine for
granting a research fellowship. La Re´gion Aquitaine
is thanked for partial support for the equipment of the
CESAMO (Centre d’Etude et d’Analyse des Mole´cules
Organiques). We thank Ciba Geigy for a generous gift
of Desferal.
J O960191H