Discovery of powerful uranyl ligands from efficient synthesis and screening

Article abstract of DOI:10.1002/chem.200401056

New tripodal gem-(bis-phosphonates) uranophiles were discovered by a screening method that allowed for the selection of ligands with strong uranyl-binding properties in a convenient microtiter-plate format. The method is based on competitive uranium bindi

Full text of DOI:10.1002/chem.200401056

FULL PAPER
Discovery of Powerful Uranyl Ligands from Efficient Synthesis and Screening
Marcin Sawicki,[a] Jean-Michel Siaugue,[a] Christophe Jacopin,[b] Christophe Moulin,[b]
Thꢀodorine Bailly,[c] Ramon Burgada,[c] Stꢀphane Meunier,[a] Paul Baret,[d]
Jean-Louis Pierre,[d] and Frꢀdꢀric Taran*[a]
Abstract: New tripodal gem-(bis-phos-
phonates) uranophiles were discovered
by a screening method that allowed for
the selection of ligands with strong
uranyl-binding properties in a conven-
ient microtiter-plate format. The
method is based on competitive urani-
um binding by using Sulfochlorophenol
S as chromogenic chelate. This dye
compound was found to present high
uranyl complexation properties and al-
lowed to highlight ligands presenting
association constants for UO22þ up to
1018 at pH 7.4 and 1020 at pH 9. A col-
lection of 40 known ligands including
polycarboxylate, hydroxamate, catecho-
late, hydroxypyridonate and hydroxy-
quinoline derivatives was tested. Also
screened was a combinatorial library
prepared from seven amine scaffolds
and eight acrylates bearing diverse che-
lating moieties. Among these 96 tested
candidates, a tripod derivative bearing
gem-bis-phosphonates moieties was
found to present the highest complexa-
tion properties over a wide range of
pH and was further studied.
Keywords: actinides
· chelates ·
Michael addition · tripodal ligands ·
uranium
Introduction
means to chelate it because handling of uranium in the nu-
clear industry as well as for military purposes is increasing.[2]
The uranyl ion UO22þ, the most stable uranium form in bio-
logical media, is a hard Lewis acid adopting unusual pseudo-
planar penta- or hexa-coordinated complexes structures.
This particular coordination mode stimulated the design and
synthesis of macrocyclic ligands hopefully useful for the se-
lective detection[3] or extraction of uranium from sea water
or nuclear wastes.[4] Stimulated by the lack of effectiveness
of existing treatments,[5] several specific ligands for UO22þ
(i.e., uranophiles) were also prepared for in vivo uranium
removal. After contamination, UO22þ is rapidly transferred
from the blood stream to its target organs, essentially
kidney and bone. Uranyl develops a chemical toxicity in the
kidney[6] whereas long-term accumulation in the bone might
induce cancer.[7] Much of the current research concerns the
synthesis of well designed uranyl ligands which are then di-
rectly evaluated in vivo on mice or rats. The most significant
advances in this area were obtained by K. N. Raymond,
P. W. Durbin and co-workers. Several multidentate cathe-
cholate and hydroxy-pyridonate ligands were synthetized
and their in vivo efficiencies evaluated by this group.[8] This
research highlighted several chelates displaying significant
uranyl removal efficiencies.[9] However, these efficiencies
were often observed when the ligands are administrated im-
mediately (5 to 30 min) after contamination, which shows
that the development of new ligands is still of interest.
The development of powerful chelating agents for actinides
represents a challenging area of research which found multi-
ple applications such as selective separation, nuclear waste
management and in vivo chelation therapy.[1] Among the ac-
tinides, a renewed interest is found in uranium and in the
[a] M. Sawicki, Dr. J.-M. Siaugue, Dr. S. Meunier, Dr. F. Taran
Service de Marquage Molꢀculaire et de Chimie Bio-organique
DBJC/DSV CEA-Saclay
91191 Gif sur Yvette Cedex (France)
Fax : (+33)169-087-991
E-mail: frederic.taran@cea.fr
[b] Dr. C. Jacopin, Dr. C. Moulin
Laboratoire de Spꢀciation des Radionuclꢀides et des Molꢀcules
DEN/DPC/SECR CEA-Saclay
91191 Gif-sur-Yvette Cedex (France)
[c] T. Bailly, Prof. R. Burgada
Laboratoire de Chimie structurale Biomolꢀculaire
UMR 7033 CNRS, 74 rue Marcel-Cachin
93017 Bobigny Cedex (France)
[d] Dr. P. Baret, Prof. J.-L. Pierre
Laboratoire de Chimie Biomimꢀtique, LEDSS, UMR CNRS 5616
Universitꢀ Joseph Fourier, BP 53
38041 Grenoble Cedex 9 (France)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2005, 11, 3689 – 3697
DOI: 10.1002/chem.200401056
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3689

Current combinatorial or parallel synthetic methodologies
allow for the preparation of high number of new molecules.
This approach combined with a screening method has pro-
vided a powerful tool for the discovery of new metal com-
plexes.[10]
The aim of this study was to develop a simple and fast in
vitro test in order to screen the ligands with strong uranyl
binding properties in a library of compounds. With the help
of the assay presented herein, screening of candidate ligands
in a parallel manner offers a rapid readout and allows the
focus on interesting candidates which can then be analyzed
more carefully in view of physico-chemical characterizations
and/or in vivo experiments. Herein are described new pow-
erful uranophiles discovered by this approach.
interest for medical applications.[15] We focused our interest
on biological relevant pH values of 5.5, 7.4 and 9 using
MES, HEPES and CHES buffers, respectively. Control ex-
periments proved that these buffers do not interfere with
the SCP/UO22þ complex.[16] At these pH values, a 1:1 com-
plex is formed instantaneously, which was confirmed by
mass spectroscopy[16] and the Job method of continuous var-
iation (Figure 1).[17] The complex presents strong colorimet-
ric features (Figure 1; e690 =21580, 23390 and 20490mꢀ1 cmꢀ1
for pH 5.5, 7.4 and 9, respectively) and good stability prop-
erties (absorbance remains stable after several weeks).
Further spectrophotometric experiments carried out at
various pH values confirmed the high complexation proper-
ties of SCP for UO22þ. The thermodynamic complexation
constants (Figure 2) were found particularly high therefore
allowing the screening of powerful uranyl ligands. For exam-
ple, the method should revealed ligands with conditional
constants Kcond > 1013 at pH 5.5, Kcond > 1018 at pH 7.4 and
Kcond > 1020 at pH 9 (if they formed a 1:1 complex). The
effect of the addition of a competitive ligand is easily visual-
ized by eye or quantified by UV/Vis spectroscopy by stain-
ing at 690 nm using a 96-well absorbance reader; this allows
for a fast and inexpensive screening without interference
due to the absorbance of most of organic uranophiles.
Results and Discussion
Screening method development: One of the most typical
methods to determine stability constants of uranophiles is
displacement of carbonate from the [UO2(CO3)3]4ꢀ complex
after addition of the ligand investigated.[11] The apparition of
the new complex is usually followed spectrophotometrically.
Unfortunately, such a method usually requires basic pH, suf-
fers of a large range of already described [UO2(CO3)3]4ꢀ as-
sociation constants (from 1018.3 to 1023)[12] and is not easily
applicable to a screening strategy since lmax absorption
might vary from a complex to another.
However, the displacement method might be used for fast
ligands screening by following the disappearance of a pre-
formed chromogenic uranyl complex.[13] Addition of another
complexing agent should induce a significant decrease in ab-
sorbance related to its chelating properties. This implies that
a well-characterized reference chelate can by used, which
displays a good affinity to uranium and generates a stable
complex with strong UV/Vis
Synthesis and screening of the ligand library: This screening
assay was then applied to a chelate library obtained by col-
lecting commercially available compounds and published li-
gands,[20,21] (Figure 3) or by parallel synthesis (Figure 4).
Recently, we described a modular approach for the paral-
lel synthesis of multidentate ligands libraries.[22] The synthe-
sis uses Michael addition of acrylates, which bear chelating
moieties on the amine scaffolds as a key step. This work was
extended by using four new acrylates (B5, B6, B7 and B8).
Combination of seven amines (six dipodals and one tripo-
absorption (lmax
> 500 nm).
We found that sulfochlorophe-
nol S (SCP, Figure 1), a well-
known indicator grade for rare
earth elements,[14] matches with
these criteria.
In aqueous solution, this
compound displays
a
violet
colour whereas the solution of
uranyl complex is blue. The col-
orimetric properties of the
SCP/UO22þ complex allowed to
perform the screening assay at
acidic, neutral or basic pH. This
point is of importance because
pH varies between the different
biological compartments, thus
the development of metal che-
lators displaying high complex-
Figure 1. a) Sulfochlorophenol S (SCP) structure; b) UV/Vis spectra and c) Job plot of SCP and SCP/UO22þ
ation properties in
a large
range of pH could be of great mixtures at pH 7.4 (absorbance measured at 640 nm).
3690
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 3689 – 3697

Uranyl Ligands
FULL PAPER
The A7B6/UO2 complex presents remarkable fluores-
cence properties which were investigated by using time-re-
solved laser-induced fluorescence (TRLIF, Figure 6). The
fluorescence spectrum of the aqueous uranyl ion reflects the
ꢀ
symmetrical vibration of the U O bond. The observed emis-
sion bands correspond to the electronic transition S11!S00
(473 nm) and S10!S0n with n=0–4 (488, 510, 535, 560, and
587 nm).[25] In the absence of the ligand, two species are
present at pH 5.5: UO22þ (t=2 ms) and UO2OH+
(t=80 ms).[26] Progressive addition of ligand A7B6 on a so-
lution of UO22þ induced a hyperchromic and bathochromic
shift (lem: 499, 520, 546 and 572 nm) of the uranyl emission.
Increase of fluorescence reached until saturation corre-
sponding to 1 equiv of A7B6 therefore confirming the 1:1
stoichiometry of the complex. Nonlinear regression fit
showed a tight slope demonstrating the high complexation
properties of A7B6 and making calculation of Kcond not pos-
sible in this experiment. The strong enhancement of the
fluorescence intensity and lifetime (t=176 ꢁ 8 ms) allowed
us to detect less than 10ꢀ11 m of uranyl ion envisioning ana-
lytical applications of this compound.
We then determined the stability constant of A7B6 at
pH 9 using first the competitive displacement method of the
SCP/UO2 complex.[27] When A7B6 was added to a solution
of SCP/UO2 complex (80 mm in CHES buffer), competitive
displacement takes place and reaches equilibrium after 24 h
according to the spectral changes. We varied the concentra-
tion of A7B6 from 0 to 10 equiv[16] and estimated the Kcond
values at each concentration. On the basis of these mea-
surements accurate association constant was found to be
logKcond = 19.2 ꢁ 0.4. In a parallel manner, we took ad-
vantage of the fluorescence properties of the A7B6/UO2
complex to carry out a competitive displacement of the
nonfluorescent uranyl carbonate complex [UO2(CO3)3]4ꢀ.
For this purpose, uranyl (40 mm) was dissolved in 30 mm car-
bonate buffer (pH 9) and the compound A7B6 was added
(4 to 400 mm). Displacement of carbonate by A7B6 takes
place and the formation of A7B6/UO2 complex was fol-
lowed by fluorescence staining.[16] Association constants, de-
termined at each A7B6 concentration, were found to be
logKcond =19.1 ꢁ 0.1. Both relative constants independently
determined are in good agreement, affording a reliable
value.
Figure 2. Thermodynamic constants and speciation diagram of SCP/UO22þ
complexes calculated by using pHab[18] and HYSS[19] programs, respec-
tively (see Experimental Section). Values in parentheses give the estimat-
ed in the least significant figure based on the variation between replicate
experiments. LH2U: logb112
logb110 = 24.9(3).
= 37.8(1), LHU: logb111 = 33.0(1), LU:
dal) and eight acrylates conducted to the formation of 56
new chelates with high structural diversity (Figure 4).
The 96 ligands were then screened in a parallel manner
on microtiter plates for their uranium binding properties at
pH 5.5, 7.4 and 9 by using SCP as a reference chelate. The
results are summarized in Figure 5.
Remarkably, the compounds bearing bis-phosphonate
chelating moieties were found to be the most potent among
the library containing polycarboxylate, hydroxamate, cate-
cholate, hydroxypyridonate and hydroxyquinoline deriva-
tives. This is highlighted by the comparison of the results ob-
tained at pH 5.5 and 7.4 with a family of tripods 3A–H and
4A constructed with the same tris(2-aminoethyl)amine
group anchoring arms. At pH 7.4, 3,4,3-LIHOPO 2G was
found the only compound without a bis-phosphonate chelat-
ing function, which was able to displace around 60% of the
SCP/UO2 complex. This ligand has already been tested for
in vivo uranyl removal with some success.[23] Comparison of
the results obtained with dipodal bis-phosphonates A1–5B6
(Figure 5, pH 7.4) show the influence of the spacer separat-
ing the two chelating moieties, the maximum of uranyl-bind-
ing was observed with ligand A4B6 (12 atoms separating the
chelating moieties). The uranyl binding properties of bis-
phosphonates overall remains superior regardless of the pH
value. More interesting is the fact that, among the bis-phos-
phonate family, tripods 4B and A7B6 presented the highest
Kcond values at pH 9 comparable to catechol based-ligand
TRENCAMS (3B) which exhibit high Kcond enhancement in
basic conditions in accordance with previous findings.[24]
Next we focused our interest on ligand A7B6 which in-
duced the highest displacement value of the SCP/UO2 com-
plex at all the tested pH.
The selectivity of the complex was then investigated using
TRLIF experiments which consist in adding a competing
metal cation (Mn+) in the presence of the A7B6/UO2 com-
plex. A decrease of the fluorescence signal will be observed
if those metals form a new complex with A7B6. Competing
metal cations, such as K+, Ca2+, Mg2+, Zn2+ and Fe3+ have
been added to the solution of the 1:1 A7B6/UO2 complex at
pH 5.5. The A7B6/UO2 complex lifetime has been examined
in order to verify that no interference due to the addition of
complexing cations occurred (variation of the lifetime would
indicate dynamic quenching and not complexation). Repre-
sentative results are displayed on Figure 7.
Studies on compound A7B6: Electrospray mass spectrome-
try (ES-MS) was first used to confirm the 1:1 stoichiometry
of the A7B6/UO22þ complex. In the negative mode, ES-MS
showed one major peak at m/z 367 corresponding to [(UO2)-
(H7L)]3ꢀ (A7B6=H12L).[16] No polymetallic complex were
detected even in the presence of an excess of UO22þ.
The data revealed that the A7B6/UO2 complex presents a
good stability even in the presence of excess of other metals
Chem. Eur. J. 2005, 11, 3689 – 3697
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3691

F. Taran et al.
Figure 3. Ligand library.
(100 equiv) except for Fe3+ which dramatically affect A7B6/
UO2 complex.
phosphonates) ligands, especially compounds 4B and A7B6,
were found the most powerful uranyl ligands among our li-
brary. These ligands display large association constants at
acidic, neutral and basic pH envisioning analytical and medi-
cal possible applications. The complexing properties of phos-
phonates are well known,[28] but to the best of our knowl-
edge, this is the first time that such a comparative study of
highly diverse ligands was accomplished with uranium.
In conclusion, SCP was found highly suitable for the rapid
screening of putative uranium ligand library. The described
procedure allowed us to compare the binding properties of
96 highly diverse ligands. The main chemical information is
that ligands bearing bis-phosphonate moieties display by far
the strongest binding features for the uranyl ion. Tris(bis-
3692
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 3689 – 3697

Uranyl Ligands
FULL PAPER
Figure 4. Parallel synthesis of ligands A1B1 to A7B8. Michael additions were run with an excess of acrylates B in THF or DMF and then quenched by
an excess of Ac2O. Crude mixtures were then concentrated and purified through filtration on silica gel. Removal of the protective groups (PG) of the
chelating moieties (L1 and L2) were carried out by Pd hydrogenation or by TMSBr to afford A1–7B1–5 and A1–7B6–8, respectively (see Experimental
Section). LC-MS controls showed more than 85% purities for each products. All compounds have been fully characterized by using 1 H, 31P NMR and
mass spectroscopy.
Chem. Eur. J. 2005, 11, 3689 – 3697
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3693

F. Taran et al.
Figure 5. Screening results: Displacements (%) of the SCP/UO22þ complex are presented in a color-coded format for clarity reason. logKcond (estimates)
were calculated by using the HYSS program supposing that the complex competitive ligand/UO22þ is 1:1. For details on the procedure, see the Experi-
mental Section.
layer chromatography (TLC) was per-
formed by using 0.25 mm silica gel
coated MERCK 60 F254 plates. Visuali-
zation of the chromatogram was car-
ried out by UV absorbance, ethanolic
solution of phosphomolybdic acid and
iodine. Flash chromatography was per-
formed using compressed air with the
indicated solvent system and silica gel
60 (Merck, 230–400 mesh). Purifica-
tions through silica cartridges were
performed by using Bond Elut Jr. 18C
(1 g, Varian) as reverse phase.
UV measurements were carried out
on a VARIAN CARRY 50 Scan UV/
Vis spectrophotometer in
a
1 cm
Figure 6. Structure of ligand A7B6. Spectrofluorescence titration of a solution of UO22þ (4.2 mm) by A7B6 in
0.1m NaClO4, pH 5.5. The inset shows the experimental and theoretical uranyl fluorescence intensity as a func-
tion of A7B6/UO22þ ratio at lexc 266 nm, lem 519 nm.
quartz Suprasil cells purchased from
HELLMA operating with Cay Win
UV software. The pH of the solutions
was measured with an automatic titra-
tor system DMS 716 titrino with
a
Experimental Section
combined glass electrode (Methrom filled with saturated KCL). NMR
spectra were recorded on Bruker instrument (AC300). MS chromato-
grams were carried out on ESI/TOF Mariner spectrometer. Analytical
LC-MS separations were carried out on Waters HPLC 2525 with a C18
column (Xterra, 5 mm, 4.6ꢂ50 mm), elution was done with H2O/CH3CN/
General methods: Unless otherwise stated, starting materials were ob-
tained from commercial suppliers and used without purification. THF
was distilled from sodium/benzophenone before use. Analytical thin
3694
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 3689 – 3697

Uranyl Ligands
FULL PAPER
B5, then CH2Cl2/MeOH 8:2) to yield protected chelate. After removal of
the solvents, the residues were then dissolved in a solution of 0.37m am-
monium formate in 9:1 DMF/H2O (10 equiv per benzyl group) and stir-
red overnight in the presence of Pd/C (10% by weight). The solutions
were filtered, acidified with 10% formic acid to pH 4, then evaporated.
Sublimation at 508C under vacuum eliminated excess of formate salts.
After filtration through a reverse phase silica cartridge (water/methanol
3:7) and evaporation, the target chelates A1–7B5 were dissolved in
MeOH and precipitated by addition of Et2O (32–75%).
Analytical data for selected compounds (data for compounds A1–7B5
are available in Supporting Information):
Figure 7. Selectivity of the A7B6/UO2 complex in the presence of metal
cations. [A7B6/UO2]=4.2 mm; [NaClO4]=0.1m; [competitive metal]=0
to 20 mm; pH 5.5
For A6B5: Yield 49%; 1H NMR (300 MHz, DMSO): d=1.83, 1.93 (2s,
6H), 2.16 (s, 6H), 3.65–4.00 (m, 4H), 4.05–4.70 (m, 4H), 4.82 (m, 2H),
6.13 (m, 2H), 6.87 (m, 1H), 7.00 (m, 2H), 7.28 (m, 1H), 7.51 (m, 2H);
13C NMR (75.47 Hz, DMSO): d=13.14, 13.50, 22.11, 22.88, 35.43, 48.41,
53.11, 111.84, 126.30, 126.56, 127.63, 130.35, 131.05, 139.28, 146.29, 170.18,
170.89, 171.54; IR (NaCl, cmꢀ1): 3392, 3078, 1905, 1727, 1633, 1504, 1421,
1248, 1174, 1026, 991, 829; HRMS: calc. for [M+H]+: 611.2353; found:
611.2339.
For A7B5: Yield 75%; 1H NMR (300 MHz, DMSO): d=1.80, 1.84, 1.95
(3s, 9H), 2.32 (s, 9H), 2.95–3.80 (m, 12H), 3.85–4.30 (m, 6H), 5.58 (m,
3H), 6.77 (m, 3H), 7.19 (m, 3H); 13C NMR (75.47 Hz, DMSO): d=14.18,
22.62, 22.72, 35.40, 47.50, 51.30, 112.10, 127.61, 134.16, 145.36, 169.78,
172.79, 173.58; IR (NaCl): n˜ = 3147, 3045, 2808, 2007, 1739, 1634, 1403,
1260, 1025, 995, 828, 659 cmꢀ1; HRMS: m/z: calcd for: 858.3521; found:
858.3549 [M+H]+.
HCOOH (gradient from 95:05:0.1 to 0:100:0.1) with a 1.0 mLminꢀ1 flow
rate; detection were carried out by UV detector (254 nm) and DEDL
PL-ELS 1000 (Polymer laboratories); mass spectra were taken on Waters
Micromass ZQ by positive electrospray method (ES+). Screening was
performed in the 96 well microtiter polystyrene plates purchased from
NUNC. The plates were read using an absorbance plate reader SPEC-
TRA max PLUS (Molecular Devices).
Time resolved laser-induced fluorescence (TRLIF) measurements were
carried out by using a Nd/YAG laser (model Minilite, Continuum, Santa
Clara, USA) operating at 266 nm (quadrupled) or 355 nm (tripled) and
delivering an energy of 2 mJ in a 4 ns pulse duration at a repetition rate
of 15 Hz was used as the excitation source. The laser output energy was
monitored by a laser powermeter (Scientech, Boulder, USA). The excita-
tion laser beam was focused on the cell of the spectrofluorometer “Fluo
2001” (Dilor, France). The light emitted from the cell was focused onto
the entrance slit of the polychromator. Taking into account dispersion of
the holographic grating used in the polychromator, measurement range
extended to approximately 200 nm into the visible spectrum with a reso-
lution of 1 nm. The detection was performed by an intensified array of
photodiodes (1024 diodes) cooled by Peltier effect (ꢀ358C) and posi-
tioned at the polychromator exit. Recording of spectra was performed by
integration of the pulsed light signal given by the intensifier. The integra-
tion time was adjustable from 1 to 99 s and allowed for variation in de-
tection sensitivity. Logic circuits, synchronized with the laser shot, al-
lowed the intensifier to be active with a defined time delay and during a
selected aperture time. The whole system was controlled by a microcom-
puter. TRLIF experiments were performed by using a gate delay of
300 ms, a gate length of 100 ms and integration time of 0.5 s. The emission
wavelength was 520 nm (excitation at 266 nm) and the temperature was
208C.
General procedure for the synthesis of ligands A1–7B6–8: Acrylate B6
(60 mg, 0.2 mmol, 8 equiv for amines A1–6 or 135 mg, 0.45 mmol,
18 equiv for amine A7) or acrylate B7 (22 mg, 0.1 mmol, 4 equiv for
amines A1–6 or 44 mg, 0.2 mmol, 8 equiv for amine A7) or acrylate B8
(27 mg, 0.1 mmol, 4 equiv for amines A1–6 or 54 mg, 0.2 mmol, 8 equiv
for amine A7) was added to a solution of amine A1–7 (0.025 mmol,
1 equiv) in dry DMF (2 mL). After reaction at room temperature (48 h
for acrylate B6, 24 h for acrylate B7 and 4 h for acrylate B8), acetic anhy-
dride (40 mL, 0.43 mmol, 17 equiv for amines A1–6 or 60 mL, 0.63 mmol,
25 equiv for amine A7) was added. The solution was stirred for 15 min at
room temperature, evaporated and then directly filtered through a silica
gel (5 g, hexane/acetone 1:1 to remove the excess of acrylate, then ace-
tone/MeOH 7:3) to yield protected chelate. After removal of the sol-
vents, protected chelates were then dissolved in dry CH3CN and TMSBr
(3 equiv per functional group) was added under argon. The mixture was
stirred under reflux for 3 h and then quenched with water (3 equiv per
functional group). After refluxing for additional 15 min, the solvents
were evaporated. The residue co-evaporated with MeOH and subse-
quently dissolved in a minimum of MeOH and precipitated with diethyl
ether. The resulting slurry was decanted, washed with diethyl ether and
dried under high vacuum in the presence of P2O5 to afford the target che-
lates A1–7B6–8 (48–99%).
Electrospray ionization mass spectrometric detection of negative ions
was performed by using a Quattro (Micromass, Manchester, UK). Sam-
ples were introduced into the source with a syringe pump (Harvard Ap-
paratus, Cambridge, MA, USA). Nitrogen was employed as both the
drying and spraying gas with a source temperature of 808C. The cone
voltage was set to 22 and 30 V, the voltage applied on the capillary is
3500 kV, and the sample solution flow rate was 10 mLminꢀ1. Spectra were
recorded by averaging 40 scans from 100 to 1000 m/z at a scan rate of 6 s
per scan.
Analytical data for selected compound (data for compounds A1–7B6–8
are available in Supporting Information):
For A6B6: Yield 90%; 1H NMR (300 MHz, D2O): d=2.18, 2.23, 2.32,
2.52 (4s, 6H), 2.83–3.32 (m, 2H), 4.06 (m, 4H), 4.84, 4.95 (2m, 4H), 7.25
(m, 1H), 7.40(m, 2H), 7.60 (m, 1H); 31P NMR (121.5 MHz, D2O): d=
18.73, 19.34; 13C NMR (75.47 Hz, D2O): d=21.63; 37.12 (t, 1J(C,P)=
124.5 Hz), 44.37 (t, 1J(C,P)=73.1 Hz), 53.59, 126.54, 126.96, 129.84,
137.54, 175.19, 175.68; IR (NaCl): n˜ = 3112, 2150, 1610, 1442, 1204, 1176,
990, 915, 821, 760, 699 cmꢀ1; HRMS: m/z: calcd for: 595.0413; found:
595.0408 [MꢀH]+.
HRMS were performed in TOF MS ES+ mode at the “Institut de
Chimie des Substances Naturelles” (Gif sur Yvette, France).
Synthesis: The preparation of compounds A1–7B1–4 has been published
elsewhere.[22] The synthesis of acrylates B5–8 is described in the Support-
ing Information.
1
For A7B6: Yield 48%; H NMR (300 MHz, D2O): d=2.23, 2.41, 2.47 (4s,
9H), 2.77–3.02 (m, 3H), 3.70 (m, 6H), 3.98 (m, 6H), 4.17 (m, 6H);
31P NMR (121.5 MHz, D2O): d=18.43; 13C NMR (75.47 Hz, D2O): d=
21.46, 38.09 (t, 1J(C,P)=123.1 Hz), 42.13 (t, 1J(C,P)=72.8 Hz), 47.12,
52.78, 176.61; IR (NaCl): n˜ = 3331, 2920, 2360, 1615, 1428, 1148, 998,
914, 702 cmꢀ1; HRMS: calcd for: 837.0809; found: 837.0846 [M+H]+.
General procedure for the synthesis of ligands A1–7B5: Acrylate B5
(94 mg, 0.25 mmol, 4 equiv for amines A1–6 or 188 mg, 0.5 mmol, 8 equiv
for amine A7) was added to a solution of amine A1–7 (0.063 mmol,
1 equiv) in dry THF (2 mL). After 2 h of reaction at room temperature,
acetic anhydride (100 mL, 1.07 mmol, 17 equiv for amines A1–6 or
150 mL, 1.6 mmol, 25 equiv for amine A7) was added. The solution was
stirred for 15 min at room temperature, evaporated and then directly fil-
tered through a silica gel (5 g, Et2O/MeOH 9:1 to remove the excess of
Screening: Uranyl stock solution (20 mm) was prepared by dissolving
UO2(OAc)2·2H2O in 0.1m perchloric acid. Sulfochlorophenol S (SCP,
purchased from Fluka) stock solution (4 mm) was prepared by dissolving
Chem. Eur. J. 2005, 11, 3689 – 3697
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3695

F. Taran et al.
SCP in Millipore quality water. All experiments were carried out at 258C
in buffers with a fixed ionic strength of 0.1m prepared as follows: MES
(for pH 5.5), HEPES (for pH 7.4) or CHES (for pH 9) (acid forms)
(12.5 mmol) and nBu4N+Clꢀ (112.5 mmol) were dissolved in Millipore
quality water (1 L). Adjustment of the pH was done using nBu4N+
OH·30H2O. A 100 mm SCP/UO22þ complex solution was prepared by
mixing: SCP stock solution (6.25 mL) diluted in the appropriate buffer
(12.5 mm, 242.5 mL) with uranyl stock solution (1.25 mL) for 1 h at room
temperature. Candidates ligands were dissolved in water to get 400 mm
solutions.
[2] a) C. Madic, Radiat. Prot. Dosim. 1989, 26, 15–22; b) L. A. Dietz,
Bull. Soc. Pharm. Lille Bull. Atomic Scientists 1991, 47, 47; c) A.
Max, T. Mason, ATW-Int. Z. Kernenergie 1996, 41, 79–84.
[3] Q. Lu, J. H. Callahan, G. E. Collins, Chem. Commun. 2000, 1913–
1914.
[4] a) S. Shinkai, H. Koreishi, K. Ueda, T. Arimura, O. Manabe, J. Am.
Chem. Soc. 1987, 109, 6371–6376; b) J. Vincent, J. F. Desreux, Inorg.
Chem. 1996, 35, 7205–7210; c) J. L. Sessler, D. Seidel, A. E. Vivian,
V. Lynch, B. L. Scott, D. W. Keogh, Angew. Chem. 2001, 113, 611–
614; Angew. Chem. Int. Ed. 2001, 40, 591–594; d) I. Tabushi, Y.
Kobuke, K. Ando, M. Kishimoto, E. Ohara, J. Am. Chem. Soc. 1980,
102, 5948–5949; e) T. N. Lambert, L. Dasaradhi, V. J. Huber, A. S.
Gopalan, J. Org. Chem. 1999, 64, 6097–6101.
[5] a) M. H. Battacharyya, B. Breitenstein, H. Mꢀtivier, B. A. Muggen-
burg, G. N. Stradling, V. Volf, Radiat. Prot. Dosim. 1992, 41, 27–36;
b) G. N. Stradling, M. H. Henge-Napoli, F. Paquet, J. L. Poncy, P.
Fritsch, D. M. Taylor, Radiat. Prot. Dosim. 2000, 87, 29–40; c) F.
Paquet, B. Ramounet, M. H. Henge-Napoli, H. Mꢀtivier, P. Gour-
melon, Radiat. Prot. Dosim. 1998, 79, 467–470; d) M. Archimbaud,
M. H. Henge-Napoli, D. Lilienbaum, M. Desloges, C. Montagne,
Radiat. Prot. Dosim. 1994, 53, 327–330; e) J. L. Domingo, A.
Ortega, J. M. Llobet, J. L. Paternain, J. Corbela, Res. Commun.
Chem. Pathol. Pharmacol. 1989, 64, 161–164; f) G. N. Strandling,
S. A. Gray, J. C. Moody, M. Ellender, Hum. Exp. Toxicol. 1991, 10,
195–198.
General procedure for the screening of the ligands library: In each well
of a microtiter plate 100 mm SCP/UO22þ solution (200 mL) and 400 mm
ligand solution (50 mL) were mixed at 258C for 36 h. Controls experi-
ments (without ligand and without UO22þ) were made in each plates. All
experiments were done in duplicate. Control experiments (SCP alone,
SCP/UO2 and SCP/competitive ligand mixtures) were carried out on
each plate and used to calculate the percentage of SCP/UO2 displace-
ment or to control that the competitive ligand did not interfere with the
UV/Vis properties of SCP. Absorbance measurements were carried out
on an absorbance plate reader by staining at 690 nm.
Using the speciation program Hyss, it was possible to determine the
value of the conditional constants, defined in Equation (1), of the com-
petitive ligands L’ in case of the formation of UL’ 1:1 complex.
Kcond ¼ ½UL0ꢂ=ð½Uꢂð½L0ꢂ þ ½HL0ꢂ þ ½H2L0ꢂ þ . . . þ ½HnL0ꢂÞÞ
ð1Þ
[6] R. W. Leggett, Health Phys. 1989, 57, 365–383.
[7] M. P. Finkel, Proc. Soc. Exp. Biol. Med. 1953, 83, 494–498.
[8] a) J. Xu, K. N. Raymond, Inorg. Chem. 1999, 38, 308–315; b) P. W.
Durbin, B. Kullgren, J. Xu, K. N. Raymond, Radiat. Prot. Dosim.
1994, 53, 305–309; c) P. W. Durbin, B. Kullgren, J. Xu, K. N. Ray-
mond, Radiat. Prot. Dosim. 1998, 79, 433–443.
[9] a) P. W. Durbin, B. Kullgren, J. Xu, K. N. Raymond, Health Phys.
1997, 72, 865–879; b) P. W. Durbin, B. Kullgren, S. N. Ebbe, J. Xu,
K. N. Raymond, Health Phys. 2000, 78, 511–521.
[10] a) M. C. Pirrung, K. Park, L. N. Tumey, J. Comb. Chem. 2002, 4,
329–344; b) T. Opatz, R. M. Liskamp, J. Comb. Chem. 2002, 4, 275–
284; c) A. J. Brouwer, H. J. van der Linden, R. M. J. Liskamp, J. Org.
Chem. 2000, 65, 1750–1757; d) C. Gennari, H. P. Nestler, U. Piarulli,
B. Salom, Liebigs Ann. 1997, 637–647; e) I. Chataigner, C. Gennari,
S. Ongeri, U. Piarulli, S. Ceccarelli, Chem. Eur. J. 2001, 7, 2628–
2634; f) C. Gennari, U. Piarulli, Chem. Rev. 2003, 103, 3071–3100.
[11] a) I . Tabushi, Y. Kobuke, K. Ando, M. Kishimoto, E. Ohara, J. Am.
Chem. Soc. 1980, 102, 5948–5949; S. Shinkai, H. Koreishi, K. Ueda,
T. Arimura, O. Manube, J. Am. Chem. Soc. 1987, 109, 6371–6376.
[12] I . Tabushi, Y. Kobuke, A. Yoshizawa, J. Am. Chem. Soc. 1984, 106,
2481–2482 and references therein.
Determination of the SCP/UO2 complexation constants: The complexa-
tion constants were determined by measuring the absorbance spectra of
an equimolar solution of SCP/UO2 at pH ranging from 1 to 10. An aque-
ous stock solution (250 mL) containing nBu4N+Clꢀ (0.1m), SCP (20 mm)
and UO2(OAc)2·2H2O (20 mm) was prepared and the pH adjusted to 1
with an aqueous solution of HCl. A series of 44 samples were prepared
in separate sealed containers at various pH using nBu4N+OHꢀ·30H2O.
After the samples had been equilibrated in the dark at 258C for at least
48 h, a potentiometric measurement of pH and an absorbance spectrum
were recorded for each solution. Data analysis was performed with the
pHab spectral componentization and least-square program. All equilibri-
um constants were defined as cumulative formation constants according
to Equation (2); SCP is designated as L.
m
n
z
mU þ nL þ zH Ð ½UmLnHzꢂ; bmlh ¼ ½UmLnHzꢂ=ð½Uꢂ ½Lꢂ ½Hꢂ Þ ð2Þ
A typical analysis of an experiment included approximately 15 equilibri-
um constants (Ka and UO22þ constants for SCP, UO22þ hydrolysis constants
and Kw). Despite this complexity the refinements were stable since only
three constants were refined (formation constants b110, b111, and b112 are
given in Figure 2). The theoretical UV spectra are in good agreement
with the experimental UV spectra (see Supporting Information).
[13] O. Braun, C. Contino, M. H. Hengꢀ-Napoli, E. Ansoborlo, B. Pucci,
Analusis 1999, 27, 65–68.
[14] a) Z. Cizek, J. Dolezal, Anal. Chim. Acta 1979, 109, 381; b) V. M.
Ivanov, N. V. Ermakova, J. Anal. Chem. 2002, 57, 110–117; c) Z.
Cizek, Anal. Chim. Acta 1979, 109, 381–391.
[15] A.-M. Albrecht-Gary, S. Blanc, F. Biaso, F. Thomas, P. Baret, G.
Gellon, J-L. Pierre, G. Serratrice, Eur. J. Inorg. Chem. 2003, 2596–
2605.
Acknowledgements
[16] See Supporting Information for details.
This work was supported by the “Nuclear Toxicology” program of the
CEA for which grateful acknowledgement is made. We also thank E. An-
soborlo for helpful discussions, E. Zekri and D. Buisson for experimental
assistance with MS and LC/MS measurements.
[17] V. M. S. Gil; N. C. Oliveira, J. Chem. Educ. 1990, 67, 473–478.
[18] P. Gans, A. Sabatini, A. Vacca, Anal. Chim. Acta 1999, 89, 45–49.
[19] L. Alderighi, P. Gans, A. Ienco, D. Peters, A. Sabatini, A. Vacca,
Coord. Chem. Rev. 1999, 184, 311–318.
[20] For compounds 1D–G, 2F–H, 3F–G, 4B–F, 4H, 5B and 5E: a) R.
Burgada, T. Bailly, M. Lecouvey, V. Larue, Y. Leroux, Eur. J. Org.
Chem. 2001, 349–352; b) T. Bailly, R. Burgada, T. Prangꢀ, M. Le-
couvey, Tetrahedron Lett. 2003, 44, 189–192; c) R. Burgada, T.
Bailly, M. Lecouvey, C. R. Chim. 2004, 7, 35–39; d) M. Kanaan, R.
Burgada, Phosphorus Sulfur Relat. Elem. 1988, 37, 217–229; e) T.
Bailly, R. Burgada, Phosphorus, Phosphorus Sulfur Silicon Relat.
Elem. 1994, 86, 217–228; f) T. Bailly, R. Burgada, Phosphorus,
Phosphorus Sulfur Silicon Relat. Elem. 1995, 101, 131–140; g) T.
Bailly, R. Burgada, C. R. Acad. Sci. Paris IIc, 1998, 241–245.
[1] a) A. E. V. Gorden, J. Xu, K. N. Raymond, P. Durbin, Chem. Rev.
2003, 103, 4207–4282; b) O. Anderson, Chem. Rev. 1999, 99, 2683–
2710; c) D. M. Taylor, G. N. Stradling, M. H. Henge-Napoli, Radiat.
Prot. Dosim. 2000, 87, 11–17; d) R. Wood, C. Sharp, P. Gourmelon,
B. Le Guen, G. N. Stradling, D. M. Taylor, M. H. Henge-Napoli,
Radiat. Prot. Dosim. 2000, 87, 51–57; e) P. W. Durbin, B. Kullgren,
K. N. Raymond, Radiat. Prot. Dosim. 1998, 79, 433–443; f) G. N.
Strandling, J. Alloys Comp. 1998, 271–273, 72–77.
3696
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 3689 – 3697

Uranyl Ligands
FULL PAPER
[21] For compounds 3A–E, 3H and 4A: a) P. Baret, V. Beaujolais, C.
Bꢀguin, D. Gaude, J-L. Pierre, G. Serratrice, Eur. J. Inorg. Chem.
1998, 613–619; b) G. Serratrice, H. Boukhalfa, C. Bꢀguin, P. Baret,
C. Caris, J-L. Pierre, Inorg. Chem. 1997, 36, 3898–3910; c) M.
Streater, P. D. Taylor, R. C. Hider, J. Porter, J. Med. Chem. 1990, 33,
1749–1755.
[22] S. Meunier, J. M. Siaugue, M. Sawicki, F. Calbour, S. Dezard, F.
Taran, C. Mioskowski, J. Comb. Chem. 2003, 5, 201–204.
[23] a) M. H. Henge-Napoli, M. Archimbaud, E. Ansoborlo, H. Metivier,
P. Gourmelon, Int. J. Radiat. Biol. 1995, 68, 389–393; b) B. Ramou-
net-Le Gall, G. Grillon, G. Rateau, R. Burgada, T. Bailly, P. Fritsch,
Radiat. Prot. Dosim. 2003, 105, 535–538.
[25] a) J. T. Bell, R. E. Biggers, J. Mol. Spectrosc. 1968, 25, 312; b) E. Ra-
binowitsch, R. L. Belford, Spectroscopy and Photophysics of Uranyl
Compounds, Pergamon Press, 1964, Oxford.
[26] C. Moulin, P. Decambox, V. Moulin, J. G. Decaillon, Anal. Chem.
1995, 67, 348.
[27] Attempts to determine the stability constants at lower pH values
were unsuccessful due to the too strong complexation of UO22þ by
A7B6 in these conditions.
[28] a) J. E. Bollinger, Q. M. Roundhill, Inorg. Chem. 1994, 33, 6421–
6424; b) C. Jacopin, M. Sawicki, G. Plancque, D. Doizi, F. Taran, E.
Ansoborlo, B. Amekraz, C. Moulin, Inorg. Chem. 2003, 42, 5015–
5022.
[24] F. Thomas, C. Bꢀguin, J. L. Pierre, G. Serratrice, Inorg. Chim. Acta
1999, 291, 148–157.
Received: October 18, 2004
Published online: April 5, 2005
Chem. Eur. J. 2005, 11, 3689 – 3697
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3697