Norbadione A: Synthetic approach to the bis(pulvinic acid) moiety and cesium-complexation studies

Article abstract of DOI:10.1002/anie.200390332

A permethylated analogue 1 of the mushroom pigment norbadione A was synthesized through a double Suzuki-Miyaura coupling of 1,7-diboronated naphthalene 2 and triflate 3. ESI-MS studies showed similar constants for the formation of the complexes of the nor

Full text of DOI:10.1002/anie.200390332

Angewandte
Chemie
Radioactive Isotope Complexation
Norbadione A: Synthetic Approach to the
Bis(pulvinic acid) Moiety and Cesium-
Complexation Studies
Marine Desage-El Murr, StØphanie Nowaczyk,
Thierry Le Gall,* Charles Mioskowski,*
Badia Amekraz,* and Christophe Moulin
Following the reactor accident in Chernobyl in 1986, it was
noted that several species of fungi contained high levels of
[*] Dr. T. Le Gall, Dr. C. Mioskowski,+ M. Desage-El Murr,
Dr. S. Nowaczyk
CEA-Saclay
Service de Marquage MolØculaire et de Chimie Bioorganique
Bât. 547, 91191 Gif-sur-Yvette (France)
Fax : (+33)1-6908-7991
E-mail: legall@dsvidf.cea.fr
mioskowski@aspirine.u-strasbg.fr
Dr. B. Amekraz, Dr. C. Moulin
CEA-Saclay, DEN/DPC/SECR/LSRM
Bât. 391, 91191 Gif-sur-Yvette (France)
Fax : (+33)1-6908-5411
E-mail: amekraz@cea.fr
[þ] Other address: Laboratoire de Synth›se Bioorganique
UMR CNRS 7514, FacultØ de Pharmacie, UniversitØ Louis Pasteur
74 route du Rhin, B.P. 24, 67401 Illkirch (France)
Fax : (+33)3-9024-4306
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2003, 42, No. 11
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4211-1289 $ 20.00+.50/0
1289

Communications
radioactivity, mainly as a result of the radioisotope 137Cs.[1]
OMe
O
Among these species was the edible mushroom bay boletus
(Xerocomus badius (Fr.) Kühn. ex Gilb.). A study by Steglich
and co-workers, who had previously isolated badione A and
norbadione A (two pigments of the cap cuticle of bay boletus)
as their dipotassium salts, showed that these compounds
formed complexes with cesium 137.[2,3] A key observation was
CO2Me
MeO
O
O
O
MeO2C
OMe
3
OMe
O
OH
OMe
O
O
CO2Me
TfO
HO
HO
O
7
B
OH
1
2 x
CO2H
B
O
O
O
O
HO
CO2H
HO
O
O
O
OMe
4
5
O
O
HO2C
OH
OH
Scheme 1. Retrosynthetic analysis of permethylatedanalogue 3. Tf=tri-
O
fluoromethanesulfonyl.
1
2
OH
O
CO2Me
MeO
CO2Me
that norbadione A (1) is a much better ligand than the simple
pulvinic acid derivative atromentic acid (2), a characteristic
that could be attributed to the presence in its structure of two
pulvinic acid chains that are able to participate jointly in the
binding of the cesium cation.[4]
b, c
a
6
OMe
OMe
7
Selective complexation of cesium could be of great value
for such purposes as the separation of the radionuclide 137Cs
from nuclear waste[5] or its removal from contaminated
persons.[6] The evaluation of norbadione A and structurally
related analogues in these applications was envisaged. Herein
we first report a synthetic approach to the bis(pulvinic acid)
backbone characteristic of the pigments of the badione group,
as illustrated by the synthesis of permethylated analogue 3
(Scheme 1). Second, the complexation properties of the
dipotassium salt of norbadione A and of compound 3 were
investigated by electrospray tandem MS (ESI MS/MS).[7] The
utilization of compound 3 was of particular interest for the
evaluation of the degree of cesium complexation specifically
by the pulvinic acid moieties, as complexation by the
carboxylate moieties could not interfere in this case. Full-
scan simple ESI MS spectra were used to investigate the
stoichiometries and stability constants of the cesium com-
plexes. Structural information on the cesium complexes was
then assessed from neutral loss experiments. Such data are
reported for the first time.
Our approach to the synthesis of 3 was based on the
palladium-catalyzed double Suzuki–Miyaura cross coupling
of triflate 5 and diboronate 4 (Scheme 1).[8] Such a strategy
has the advantage that it should be amenable to the
preparation of other analogues in which the distance between
the two pulvinic acid chains could be modified simply by
changing the diboronate.
The preparation of triflate 5, described in Scheme 2, relied
mainly on the Lewis acid catalyzed Langer cyclization of 1,3-
bis(trimethylsilyloxy)-1,3-butadiene derivative 8 with oxalyl
chloride.[9,10] a-Aryl-b-ketoester 7 was obtained in good yield
from the reaction of 2 equivalents of the lithium enolate
derived from methyl para-anisylacetate (6) and methyl
Me3SiO
MeO
OMe
OMe
OSiMe3
CO2Me
HO
e
d
5
O
O
8
OMe
9
OMe
Scheme 2. Synthesis of triflate 5. a) 6 (2 equiv), iPr2NLi (2 equiv), THF,
À788C, then MeOCH2CO2Me (1 equiv), À788C!RT, 77%; b) Et3N,
Me3SiCl, THF, room temperature; c) iPr2NLi, THF, À788C, then
Me3SiCl, À788C!RT, 95%; d) oxalyl chloride (1.3 equiv), Me3SiOTf
(0.3 equiv), CH2Cl2, À788C!RT, 54%; e) Tf2O, pyridine, CH2Cl2,
À788C!RT, 68%.
methoxyacetate. Ketoester 7 was converted into the bis(sily-
lated) diene 8 in two steps, which proceeded in very good
yield on a multigram scale. However, compound 8 did not
withstand distillation and was therefore used in the next step
without purification. The cyclization of 8 with oxalyl chloride
was then carried out under a variety of conditions. The best
yield of lactone 9 (54%) was obtained with 1.3 equivalents of
oxalyl chloride in the presence of 0.3 equivalents of trime-
thylsilyl triflate, upon allowing the reaction mixture to warm
from À788C to room temperature overnight. Only one
stereoisomer was obtained. The attribution of the indicated
E configuration is based on general observations by Langer
et al.[9] on the stereochemical outcome of related cyclization
reactions and on the preparation of the known compound 10
(see below). Alcohol 9 was readily converted into the
corresponding triflate 5 by treatment with triflic anhydride
in the presence of pyridine.
A Suzuki–Miyaura reaction of triflate 5 was then carried
out using phenylboronic acid in the presence of the catalyst
1290
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4211-1290 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 11

Angewandte
Chemie
binding constants for such noncovalent complexes.[18] These
studies have shown good agreement between relative solu-
tion-phase and gas-phase ionic abundances of metal com-
plexes present in thermodynamic equilibrium in solution.
Recently, ESI MS has allowed the acquisition of solution data
such as stoichiometries and stability constants of LnIII
complexes of the whole lanthanide series with various
di(dialkyltriazinyl)pyridine ligands.[19] ESI MS thus seemed
appropriate for the study of the complexation of cesium by
the potassium salt of norbadione A or by methylated
compound 3. There have been no reports published that
deal with the stability constants of Cs+ complexes of the
potassium salt of norbadione A, which is the naturally
occurring form of this compound.
TfO
OMe
CO2Me
OMe
CO2Me
PhB(OH)2
[Pd(PPh3)4] (3 mol%)
K3PO4 (1.5 equiv)
O
O
O
O
dioxane, reflux, 4 h
86%
OMe
OMe
5
10
Scheme 3. Preparation of pulvinic acidderivative 10 by the Suzuki–
Miyaura reaction of phenylboronic acidwith triflate 5.
[Pd(PPh3)4] and the base K3PO4 at reflux in dioxane,[11] to
afford compound 10 in 86% yield (Scheme 3). The H NMR
1
spectrum of compound 10 was identical to that reported for
the E isomer of O-methylisopinastric acid, which is quite
different from that of the Z isomer.[12] This allowed us to
determine the E configuration of the central double bond in
alcohol 9 unambiguously.
The preparation of diboronate 4 was carried out in two
steps from 1,7-dihydroxynaphthalene (11; Scheme 4). Treat-
ment with triflic anhydride in pyridine converted 11 into the
corresponding known ditriflate.[13] The procedure described
by Masuda and co-workers.[14] for the palladium-catalyzed
borylation of aryl triflates with pinacolborane enabled the
preparation of diboronate 4 in 45% yield.
We first studied the complexation behavior of the
dipotassium salt of norbadione A (hereafter denoted
NboK2) by means of a speciation study. A 10À4 m solution of
NboK2 in methanol was treated with increasing quantities of
cesium chloride.[20] Three different species were identified:
free NboK2, observed as the potassium complex [Nbo + 2K +
K]+ (and as the sodium complex [Nbo + 2K + Na]+ of weaker
relative intensity),[21] and cesium complexes [Nbo + 2K +
Cs]+ and [Nbo + K + 2Cs]+. The species [Nbo + 3Cs]+ was
not observed. Its formation might be prevented for steric
reasons. Based on the assumption that the ion current in the
gas phase quantitatively represents the solution equilibrium,
the species distribution was obtained from the total signal
response of these singly charged ions in spectra taken at each
cesium/norbadione A ratio (see Supporting Information). By
doing so, it was possible to plot the speciation curves shown in
Figure 1.
c
a, b
O
B
3
OH
B
O
OH
O
O
As CsCl was added, the quantity of complex [Nbo + 2K +
Cs]+ present reached a maximum at a 1.2 molar ratio of
cesium to ligand, while the quantity of complex [Nbo + K +
2Cs]+ increased regularly. The formation of the latter
complex is observed even when only small amounts of
cesium are present, thus showing that this species is favored.
Neutral loss experiments were also performed to provide
structural information about the cesium complexes. The
neutral loss of the component NboK2 was only seen from
[Nbo + 2K + Cs]+ and the neutral loss of the component
NboKCs was only seen from [Nbo + K + 2Cs]+. This suggests
11
4
Scheme 4. Preparation of permethylatedanalogue 3 by double Suzuki–
Miyaura reaction of diboronate 4 andtriflate 5. a) Tf2O, pyridine, 95%;
b) pinacolborane (6 equiv), [PdCl2(dppf)] (0.06 equiv), Et3N (12 equiv),
dioxane, 808C, 13 h, 45%; c) 5 (2 equiv), [PdCl2(PPh3)2] (cat.), aqueous
Na2CO3, THF, reflux, 3 h, 63%. dppf=1,1’-bis(diphenylphosphanyl)fer-
rocene.
Our initial idea for the preparation of compound 3
through a double Suzuki–Miyaura reaction was to make use
of the corresponding diboronic acid as the starting material.
However, several attempts to hydrolyze the boronate groups
of 4 were unsuccessful. Fortunately, conditions recently
described for the palladium-catalyzed cross coupling of
vinyl triflates with boronic esters, in the presence of
[PdCl2(PPh3)2] as the catalyst and aqueous Na2CO3 as the
base,[15] were found to be successful for the conversion of
diboronate 4 and triflate 5 (2 equiv) into the desired
norbadione A analogue 3, obtained in 63% yield.
ESI MS has been applied previously to the study of metal-
ion complexation by a variety of ligands to provide the
structures, stoichiometries, and oxidation states of ligand–
metal complexes.[16] Several reports have shown the potential
of ESI MS for studying the interactions between alkaline
cations and ligands.[17] It also appears that ESI MS can be used
as a reliable technique for the direct determination of solution
Figure 1. ESI–MS speciation curves of norbadione A dipotassium salt
^
in the presence of cesium chloride in methanol. : free ligand,
+
+
~
&
: [Nbo+2K+Cs] complex, : [Nbo+K+2Cs] complex. Relative
abundances are reproducible within 5% from run to run.
Angew. Chem. Int. Ed. 2003, 42, No. 11
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4211-1291 $ 20.00+.50/0
1291

Communications
that the norbadione A potassium salt binds to Cs+ through the
pulvinic acid moieties, and that the second Cs+ binding
involves one of the carboxylate groups.
The complexation constant determined by ESI MS for the
formation of the complex that contains one cesium cation, as
represented in Scheme 5, is lgb = 4.9 ꢀ 0.4.
Scheme 5. Formation of the NboK2·Cs+ complex from norbadione A
dipotassium salt.
We also undertook a speciation study on the permethyl-
ated analogue 3 by using the same experimental procedure as
described above. For reasons of coherence with the first
speciation study, 1.2 equivalents of potassium were added to
each sample to express the competitive influence of potas-
sium on Cs+–3 complex formation (Figure 2). Two different
Figure 3. Structural hypothesis for the cesium complex of NboK2
(Chem3D model). One potassium ion is located above the plane of
the naphthalene moiety, the other one is locatedunderneath that
plane.
in a pseudocavity formed by a special arrangement of the two
pulvinic acid moieties. Further work will include a decorpo-
ration study of radioactive cesium on animals, by treatment
with norbadione A, the development of our synthetic
approach towards a total synthesis of norbadione A, and the
preparation of analogues designed for the selective complex-
ation of different metal cations.
Received: October 10, 2002 [Z50332]
À
Keywords: C C coupling · cesium · mass spectrometry · metal
complexation
.
Figure 2. ESI–MS speciation curves of the permethylatedanalogue 3 in
&
^
the presence of cesium chloride in methanol. : free ligand, : com-
plex [3+Cs]+. Relative abundances are reproducible within 5% from
run to run.
ˇ
[1] For a review, see: P. Kalac, Food Chem. 2001, 75, 29 – 35.
[2] a) B. Steffan, W. Steglich, Angew. Chem. 1984, 96, 435; Angew.
Chem. Int. Ed. Engl. 1984, 23, 445 – 447; b) D. C. Aumann, G.
Clooth, B. Steffan, W. Steglich, Angew. Chem. 1989, 101, 495;
Angew. Chem. Int. Ed. Engl. 1989, 28, 453 – 454.
[3] Norbadione A was also isolated from the mushroom Pisolithus
tinctorius (Pers.) Coker et Couch; see: a) M. Gill, D. A. Lally,
Phytochemistry 1985, 24, 1351 – 1354; b) M. Gill, M. J. Kiefel,
Aust. J. Chem. 1994, 47, 1967 – 1977; the dipotassium salt of
norbadione A employed in this study was provided by Dr.
Melvyn Gill, University of Melbourne, Australia.
[4] For a recently reported physicochemical study of the diacid
corresponding to norbadioneA, see: S. GaraudØe, M. Elhabiri,
D. Kalny, C. Robiolle, J.-M. Trendel, R. Hueber, A. Van Dors-
selaer, P. Albrecht, A.-M. Albrecht-Gary, Chem. Commun. 2002,
944 – 945.
[5] J. F. Dozol, M. Dozol, R. M. Macias, J. Inclusion Phenom.
Macrocyclic Chem. 2000, 38, 1 – 22.
[6] For articles on the decorporation of radionuclides from the
human body, see: M. H. Henge-Napoli, G. N. Stradling, D. M.
Taylor, Radiat. Prot. Dosim. 2000, 87, 3 – 57.
[7] P. Kebarle, J. Mass Spectrom. 2000, 35, 804 – 817.
[8] For reviews, see: a) A. Suzuki, J. Organomet. Chem. 1999, 576,
147 – 168; b) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457 –
2483; c) S. R. Chemler, D. Trauner, S. J. Danishefsky, Angew.
Chem. 2001, 113, 4676 – 4701; Angew. Chem. Int. Ed. 2001, 40,
4545 – 4568.
species were identified: free 3, observed as the potassium
complex [3 + K]+, and the cesium complex [3 + Cs]+. The
stability constant calculated for the 1:1 complex is lgb’ = 4.6 ꢀ
0.4 (Scheme 6). The stability constants of the 1:1 complexes
that contain ligands 3 and norbadione A are thus quite
similar. These results are in agreement with the structural
hypothesis made from the neutral loss experiments whereby
the neutral component of the [Nbo + 2K + Cs]+ complex is
NboK2. We also briefly evaluated the binding selectivity of 3
for Na+, K+, and Cs+. In the presence of 0.8 equivalents of
each alkali cation (in total, 2.4 equivalents), we observed a
distribution of 60% cesium complex, 24% potassium com-
plex, and 16% sodium complex. This distribution clearly
indicates selectivity for cesium over potassium and sodium.
In summary, these results suggest that the complex formed
by the dipotassium salt of norbadione A and one cesium
cation results from an interaction between this cation and the
pulvinic acid moieties of norbadione A. Thus, the structure
depicted in Figure 3 might be favored: the cesium ion is held
[9] a) P. Langer, T. Schneider, M. Stoll, Chem. Eur. J. 2000, 6, 3204 –
3214; b) P. Langer, T. Eckardt, Synlett 2000, 6, 844 – 846; c) P.
Scheme 6. Formation of the 3·Cs+ complex from compound 3.
1292
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4211-1292 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 11

Angewandte
Chemie
Langer, T. Eckardt, T. Schneider, C. Göbel, R. Herbst-Irmer, J.
Org. Chem. 2001, 66, 2222 – 2226.
[10] For a review, see: P. Langer, Synthesis 2002, 441 – 459.
[11] T. Oh-e, N. Miyaura, A. Suzuki, J. Org. Chem. 1993, 58, 2201 –
2208.
[12] G. Pattenden, M. W. Turvill, A. P. Chorlton, J. Chem. Soc. Perkin
Trans. 1 1991, 2357 – 2361.
[13] M. Takeuchi, T. Tuihiji, J. Nishimura, J. Org. Chem. 1993, 58,
7388 – 7392.
[14] M. Murata, T. Oyama, S. Watanabe, Y. Masuda, J. Org. Chem.
2000, 65, 164 – 168.
[15] E. G. Occhiato, A. Trabocchi, A. Guarna, J. Org. Chem. 2001, 66,
2459 – 2465.
[16] a) C. A. Schalley, Mass Spectrom. Rev. 2001, 20, 253 – 309; b) R.
Colton, A. D'Agostino, J. C. Traeger, Mass Spectrom. Rev. 1995,
14, 79 – 106.
[17] a) W. Z. Shou, R. F. Browner, Anal. Chem. 1999, 71, 3365 – 3373;
b) G. A. Lawrance, M. J. Robertson, Sutrisno, E. I. von Nagy-
Felsobuki, Inorg. Chim. Acta 2002, 328, 159 – 168; c) M. L.
Reyzer, J. S. Brodbelt, A. P. Marchand, Z. Chen, Z. Huang,
I. N. N. Namboothiri, Int. J. Mass Spectrom. 2001, 204, 133 – 142.
[18] a) K. Wang, G. W. Gokel, J. Org. Chem. 1996, 61, 4693 – 4697;
b) P. A. Brady, J. K. M. Sanders, New J. Chem. 1998, 22, 411 –
417; c) E. C. Kempen, J. S. Brodbelt, Anal. Chem. 2000, 72,
5411 – 5416.
[19] S. Colette, B. Amekraz, C. Madic, L. Berthon, G. Cote, C.
Moulin, Inorg. Chem., 2003, in press.
[20] Methanol was preferred to water as the solvent, since decar-
boxylation reactions readily occur in the ESI source with
aqueous solutions.
[21] Sodium adducts detected in the spectra result from the back-
ground concentration of Na+ present in the samples.
Angew. Chem. Int. Ed. 2003, 42, No. 11
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4211-1293 $ 20.00+.50/0
1293