Selective oxidation of aromatic sulfide catalyzed by an artificial metalloenzyme: New activity of hemozymes

Article abstract of DOI:10.1039/b907534h

Two new artificial hemoproteins or "hemozymes", obtained by non covalent insertion of Fe(III)-meso-tetra-p-carboxy- and -p-sulfonato- phenylporphyrin into xylanase A from Streptomyces lividans, were characterized by UV-visible spectroscopy and molecular modeling studies, and were found to catalyze the chemo- and stereoselective oxidation of thioanisole into the S sulfoxide, the best yield (85 ± 4%) and enantiomeric excess (40% ± 3%) being obtained with Fe(III)-meso-tetra-p-carboxyphenylporphyrin-Xln10A as catalyst in the presence of imidazole as co-catalyst. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b907534h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Selective oxidation of aromatic sulfide catalyzed by an artificial
metalloenzyme: new activity of hemozymes†
a
a
b
b
R e´ my Ricoux, Mathieu Allard, Roger Dubuc, Claude Dupont,
c
a
Jean-Didier Mar e´ chal and Jean-Pierre Mahy*
Received 14th April 2009, Accepted 16th June 2009
First published as an Advance Article on the web 23rd June 2009
DOI: 10.1039/b907534h
Two new artificial hemoproteins or “hemozymes”, obtained
by non covalent insertion of Fe(III)-meso-tetra-p-carboxy-
and -p-sulfonato-phenylporphyrin into xylanase A from
Streptomyces lividans, were characterized by UV-visible spec-
troscopy and molecular modeling studies, and were found
to catalyze the chemo- and stereoselective oxidation of
thioanisole into the S sulfoxide, the best yield (85 ± 4%) and
enantiomeric excess (40% ± 3%) being obtained with Fe(III)-
meso-tetra-p-carboxyphenylporphyrin-Xln10A as catalyst in
the presence of imidazole as co-catalyst.
was brought by the protein around the iron atom of the porphyrin.
In addition, a very interesting protecting effect against oxidative
13
degradation of the porphyrin was provided by the protein. These
two properties made Fe(TpCPP)-Xln10A an excellent candidate
as catalyst for the enantioselective oxidation of substrates. Sulfur
containing compounds were chosen as substrates as first, they
14
play an important role in medicine and agriculture, second,
their enzymatic oxidation produces sulfoxide metabolites that are
15
responsible for their direct toxicity and, third thioanisole was well
known to be oxidized into an asymmetric sulfoxide by naturally
22
occurring peroxidases such as horseradish peroxidase (HRP).
In the present study, the oxidation of thioanisole by H was
2
O
2
In the last 10 years, several strategies have been reported
for the design of artificial metalloenzymes, including the
introduction of metal binding sites in the active site of a protein
performed in the presence of either 1-Fe or meso-tetra(4-sulfonato-
phenyl)porphyrinatoiron(III) (FeTpSPP) 2-Fe alone or of the cor-
responding Fe-porphyrin-Xln10A complexes acting as a catalyst.
The tetra-arylporphyrins were synthesized according to liter-
ature procedures as described in the ESI. 1 was obtained by
condensation of p-carboxybenzaldehyde with pyrrole in propionic
1
by site-directed mutagenesis, the design of substrate binding
2
3
cavities, chemical modification of prosthetic groups, covalent
4
attachment of metal cofactors and by simple incorporation of
oxometallate ions into the proteins. Artificial hemoproteins
with an interesting peroxidase activity have also been obtained
5,6
16
acid according to the procedure of Adler et al. 2 was prepared by
sulfonation of tetraphenyl porphyrin by concentrated sulfuric acid
by the non-covalent association of a synthetic Fe(III)-a
o-carboxyphenylporphyrin or microperoxidase 8 (MP8) with
3
b-tetra-
17
as described by Fleischer et al. 1-Fe was prepared by reaction of
7
8
1
8
1
with excess FeCl
2
according to Harada et al. 2-Fe was obtained
·4H O in 1 M acetate buffer
pH 4. Xylanase A was purified from the supernatant of S. lividans
monoclonal antibodies raised against these cofactors. These
by treatment of 2 with 1.1 eq. FeCl
2
2
MP8-antibody complexes were additionally shown to catalyze
19
9
the stereoselective oxidation of sulfides by H
2
O
2
. New artificial
13
culture as reported. Hemozymes were obtained by the addition
of 1 eq. 1-Fe or 2-Fe to Xln10A (1.5 eq., 2 mM) in 50 mM citrate
phosphate buffer pH 3. This pH was chosen since it is well known
that at pH values above 5 both iron porphyrins led to the formation
hemoproteins, displaying an interesting peroxidase activity were
also obtained by insertion of modified hemes into mutated sperm
10
whale myoglobin. Heat resistant hemoproteins that are able to
bind and release O in aqueous medium were also obtained by
incorporating Fe(II)-a -tetra-o-pivalamidophenylporphyrin into
a thermo-resistant xylanase or into human serum albumin.
2
20
of m-oxo dimers.
4
Both 1- and 2-Fe-Xln10A complexes were characterized by
UV-visible spectroscopy in 50 mM citrate buffer pH 3. The
spectrum of the 1-Fe-Xln10A complex was very similar to that
of 1-Fe alone (Fig. 1A), with maxima at 411, 565 and 609 nm
11
12
Thus, we have recently generated, by non-covalent incorporation
of Fe(III)-meso-tetra-p-carboxyphenylporphyrin (Fe(TpCPP))
1
-Fe into b-1,4-endoxylanase or xylanase A (Xln10A) from
(
at pH 3), suggesting that no amino acid side-chain of Xln10A
13
Streptomyces lividans , a new artificial hemoprotein, Fe(TpCPP)-
Xln10A, that was found to have peroxidase activity. Enzymatic
kinetic studies, as well as coordination chemistry studies
suggested that steric hindrance and an increased hydrophobicity
was coordinating to the iron. On the contrary, whereas 2-Fe
exhibited a UV-visible spectrum characteristic of an high-spin
pentacoordinate porphyrin-iron(III) species, with maxima at 394,
5
30 and 666 nm, 2-Fe-Xln10A was characterized by a very different
UV-visible spectrum with maxima at 416, 552 and 619 nm,
characteristic of a low-spin hexacoordinate porphyrin-iron(III)
a
Institut de Chimie Mol e´ culaire et des Mat e´ riaux d’Orsay, UMR 8182
CNRS, Laboratoire de Chimie Bioorganique et Bioinorganique, B aˆ t. 420,
Universit e´ Paris 11, 91405, Orsay Cedex, France. E-mail: jpmahy@icmo.u-
psud.fr; Fax: +33(1)69157281; Tel: +33(1)69157421
(Fig. 1B). This clearly indicated that the environment of the iron
atom changed upon insertion of 2-Fe into Xln10A, which could
be due to the binding of one or two amino acid side chains of
Xln10A to the iron atom.
To explain those results, a new analysis by molecular modeling
was done. Both 1-Fe and 2-Fe have been docked into the Xln10A
binding site with the program Gold (version 3.2), following the
b
Institut National de la Recherche Scientifique, Institut Armand-Frappier,
5
31 Boulevard des Prairies, Laval, Qu e´ becH7V 1B7, Canada
c
Unitat de Qu ´ı mica F ´ı sica, Dep. de Qu ´ı mica, Universitat Aut o` noma de
Barcelona, Edifici C.n., 08193, Cerdonyola (Barcelona), Spain
†
Electronic supplementary information (ESI) available: Experimental and
molecular modelling details. See DOI: 10.1039/b907534h
3
208 | Org. Biomol. Chem., 2009, 7, 3208–3211
This journal is © The Royal Society of Chemistry 2009

View Article Online
Fig. 1 Comparison of the UV-visible spectra of 1-Fe (—) and
1
1
-Fe-Xln10A ( ◊ ◊ ◊ ◊ ◊ ◊ ) (A), and 2-Fe (—) and 2-Fe-Xln10A ( ◊ ◊ ◊ ◊ ◊ ◊ ) (B).
.5 mM porphyrin and 1.5 eq of Xln10A in 50 mM citrate phosphate buffer
Fig. 2 Individual lowest energy complexes of the two main orientations
of 2-Fe docked in Xln 10A binding site. Residues at the proximity of the
porphyrin are represented in particular the two putative residues involved
in iron chelation. Panel A Corresponds to the most stable orientation and
in panel B to the lowest energy structure obtained for the second cluster
of solutions.
pH 3.
1
3
same protocol as described earlier (see ESI). To better explore
the [Porph-Xln10A)] structural variability, the number of docking
runs has been extended to 100. From the resulting solutions, it
appears that, when inserted inside the cleft of Xln10A, both -SO
or -CO
-
3
-
2
porphyrin derivatives can display two different main
moment, flexible protein dockings are in progress to confirm this
hypothesis.
The influence of the Xln10A protein on the yield and selectivity
of the oxidation of thioanisole catalyzed by the 1-Fe and 2-Fe was
orientations (Fig. 2).
In the most stable orientation, which is very similar to that
13
reported in our previous study, the main polar contacts occur
between the negatively charged substituents of the phenyl groups
of the ligands are with the patches {His80, Lys48, Trp266} and
then studied. Thioanisole (8.5 mM) was thus oxidized by H
2
O
2
(final concentration, 3.5 mM) catalyzed either by 20 mM 1-Fe or
2-Fe alone, or by 1-Fe or 2-Fe Xln10A prepared by preincubation
{
Ser212, Asn208, Asn173} of the receptor (Fig. 2A). In terms
◦
of binding energies, sulfonic species are predicted to be about
of 20 mM Fe-porphyrin with 25 mM XlnA for 2 h at 4 C. The
-
1
6
-
kJ mol less stable than carboxy ones (-24.55 for 2-Fe versus
reactions with 2-Fe and 2-Fe-Xln10A were performed at pH 3,
30.27 kJ mol- for 1-Fe). In the less stable orientation, only the
1
since at this pH, 2-Fe had been shown to be able to catalyze the
21
polar contacts with the first patch of residues remain, whereas
weaker interactions exist, mainly in the form of hydrogen bonds
with residues located on the top of the external part of the
cavity highly solvent exposed, particularly Ser275 and Arg138 or
Gln93. Those interactions are less stabilizing than for the previous
cluster as reflected by the values of the predicted binding energies
oxidation of aromatic sulfides by H
2
O . Three kinds of buffers
2
have been tested, 50 mM glycine pH 3, 50 mM formate buffer
pH 3 and 50 mM phosphate/citrate buffer pH 3. In the three
buffers, no reaction occurred in the presence of only Xln10A, but
in the presence of 2-Fe and 2-Fe-Xln10A, a chemoselective reaction
was observed, with the formation of only sulfoxide as oxidation
product (Table 1). When the reactions were performed under
anaerobic conditions, no change in the yields and enantiomeric
excesses was observed, which clearly indicated that the oxygen
-
1
-1
(
-22.63 kJ mol and -14.10 kJ mol resp. for 1-Fe and 2-Fe).
Considering the most stable orientation, owing to the larger size
of the sulfonato substituents with respect to the carboxylato ones,
a slightly less deepening of 2-Fe into the Xyln10A binding site with
respect to 1-Fe was observed. In addition, in the case of 2-Fe, a
detailed inspection of close contact residues present the Serine 86
and the Histidine 85 though lying about 6 A from the iron ion (6.19
and 6.35 A respectively) that could coordinate it after a structural
rearrangement of the protein, which could be at the origin of
the spectral change observed when the insertion of 2-Fe into
Xln10A is monitored by UV-visible spectroscopy (Fig. 2B). At the
atom of the sulfoxide was not coming from O . The use of 2-Fe
2
as catalyst led to rather similar yields (42 to 50%) and turnover
numbers after 2.3 h (69 to 82 mol. sulfoxide/mol. catalyst) but no
enantiomeric excess was observed. In the three buffers, insertion
of 2-Fe into Xln10A led to a decrease in the yields (18 to 27%)
and turnover numbers after 2.3 h. (29 to 45 mol. sulfoxide/mol.
catalyst), but an enantiomeric excess in favor of the S sulfoxide
was observed. This enantiomeric excess was dependant upon the
˚
˚
This journal is © The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3208–3211 | 3209

View Article Online
Table 1 Yield, turnover number and enantiomeric excesses obtained for the oxidation of thioanisole into sulfoxide by H
their complexes with Xln10A in various buffers at pH 3 or 50 mM phosphate buffer pH 7
2
O
2
catalyzed by 1-Fe, 2-Fe and
a
b
Buffer
Catalyst
Yield (%)
Turnover number
ee (%)
5
5
5
5
0 mM phosphate/citrate pH 3
0 mM glycine pH 3
2-Fe
50 ± 2
27 ± 3
45 ± 2
24 ± 3
42 ± 2
18 ± 3
33 ± 2
93 ± 3
24 ± 3
85 ± 4
86 ± 2
47 ± 3
77 ± 2
42 ± 3
72 ± 2
31 ± 3
57 ± 2
162 ± 3
42 ± 3
145 ± 3
—
10 ± 1
—
24 ± 2
—
8 ± 1
—
—
36 ± 3
40 ± 3
2
-Fe-Xln10A
2-Fe
2
-Fe-Xln10A
0 mM formate pH 3
2-Fe
2
1-Fe
1
1
1
-Fe-Xln10A
0 mM phosphate pH 7.4
-Fe + ImH
-Fe-Xln10A
-Fe-Xln10A + ImH
a
Reaction conditions: 1-Fe or 2-Fe 20 mM, Xln10 A 25 mM (1.25 eq), thioanisole 8.5 mM (425 eq), ImH ± 2mM (100 eq) and H
2
O
2
3.5 mM final (175 eq).
b
2 2
The yield was calculated as a function of the final H O concentration. Turnover number: mol. sulfoxide/mol. catalyst in 2.3 h.
nature of the buffer, and, whereas it was weak in 50 mM formate
buffer pH 3 and 50 mM phosphate/citrate buffer pH 3 (Resp. 8 and
responsible for the oxidation. Accordingly, as can be seen from
Table 1, addition of imidazole to 1-Fe and 1-Fe-Xln10A led in
both cases to a tremendous increase in the yield of sulfoxide
formed (resp. 93 and 85%) and in the turnover number (resp.
162 and 145 mol. sulfoxide/mol. catalyst in 2.3 h). However,
no enantiomeric excess was observed with 1-Fe and imidazole,
whereas a slight increase in the enantiomeric excess in favor of
the S sulfoxide, that reached 40%, was observed in the presence of
Xln10A, indicating that the role of the xylanase protein is mainly
to provide a chiral environment to the iron-porphyrin complex
and to induce a stereoselectivity in the reaction. Under those best
conditions, the rate of the reaction was determined as a function
1
0%) it could reach 24% in 50 mM glycine pH 3. Both the lower
yields and turnover numbers, and the induction of an enantiomeric
excess observed in all the cases with 2-Fe as catalyst were in
agreement with the insertion of 2-Fe into the cleft of Xln10A.
The reactions with 1-Fe and 1-Fe-Xln10A were performed at
pH 7.4, since at this pH the para-substituents of the meso-phenyl
groups of 1 were predominantly in the carboxylate form, and also
since it was previously shown that an amino acid with a pK of
A
13
about 7.5 could be participating in the catalysis. Once again,
under those conditions better yield (33%) and turnover number
(
1
4
54 mol. sulfoxide/mol. catalyst in 2.3 h.) were obtained with
of the H
the kinetic parameters to be determined: kcat = 8.0 ± 0.1 min and
= 5.4 ± 0.2 mM for H
In conclusion, the above mentioned results show that the as-
sociation of iron(III)-meso-tetra-p-carboxylatophenyl-porphyrin
1-Fe with xylanase A leads, in the presence of 100 eq. imidazole,
to a new "hemozyme" that is able to catalyze the chemo- and
2
O concentration. Lineweaver Burk plots then allowed
2
-
1
-Fe than with 1-Fe-Xln10A (yield = 24%, turnover number =
0 mol. sulfoxide/mol. catalyst in 2.3 h). The most remarkable
K
M
2
O .
2
effect was an increase in the enantiomeric excess in favor of the
S sulfoxide that reached 36% with 1-Fe-Xln10A as catalyst, that
was even larger than that observed with 2-Fe-Xln10A in 50 mM
glycine pH 3. In agreement with the molecular modeling studies,
this could be due to a deeper insertion of 1-Fe than 2-Fe into the
cleft of Xln10A.
Finally to improve the catalytic efficiency of our system, 100 eq.
of imidazole (with respect to catalyst) were added. Indeed, it is
well known that, in the oxidation reaction by H
stereoselective oxidation of sulfides into sulfoxide by H
2
2
O . On
the one hand, the metalloporphyrin cofactor induces a remarkable
-
1
catalytic efficiency (kcat = 8.0 min ), and on the other hand,
xylanase A not only protects the cofactor against self-oxidation,
but also induces an enantiomeric excess in favor of the S sulfoxide
(40%).
2
2
O catalyzed
by metalloporphyrins, it can act both as an axial ligand of the
metal and as a general base to help in the heterolytic cleavage
23
When compared to other artificial metalloenzymes (Table 2),
of the O-O bond of H
2
O
2
to form the metal-oxo species that is
1-Fe-Xln10A appears to be best hybrid catalyst for the S-oxidation
2 2
Table 2 Enantiomeric excesses, yields or turnovers obtained for the sulfoxidation of sulfides by H O catalyzed by artificial metalloenzymes
Protein
Metal complex
Anchoring
ee (%)
Yield (%)
85
Conf.
Ref.
c
Xylanase A
Antibody 7A3
Apomyoglobin and mutants
1-Fe
SM
40
8
S
S
S
S
S
R
S
S
S
S
This work
(25)
(27)
(28)
(26)
(9)
(24)
(24)
(5)
c
a
Estradiol-Fe-Porph
Cr-Salen
SM
—
c
SM
5–13
51
60
45
38
74
66
93
54
d
b
Mn-Salen
Mn-Salen
MP8
C
—
c
SM
SM
SM
SM
49.4
49
87
16
100
53
c
Antibody 3A3
BSA
c
Fe-corrole
Mn-corrole
c
3-
Phytase
Streptavidin
VO
VOSO
4
—
—
4
(6)
a
Yield not mentioned, turnover : 0.3 min-1. b Yield not mentioned, turnover : 0.39 min-1. c SM: supramolecular. C: covalent.
d
3
210 | Org. Biomol. Chem., 2009, 7, 3208–3211
This journal is © The Royal Society of Chemistry 2009

View Article Online
of sulfides obtained by insertion of an iron-complex into a protein.
Indeed, with the exception of abzyme 3A3-MP8, which leads
to a 45% enantiomeric excess in favor of the (R)-sulfoxide, but
4 R. R. Davies and M. D. Distefano, J. Am. Chem. Soc., 1997, 119,
1
1643.
5
F. van de Velde, L. K o¨ nemann, F. van Rantwijk and R. Sheldon, Chem.
Commun., 1998, 1891.
9
with a lower yield (45%), all other artificial Fe-proteins induce
6 A. Pordea, M. Creus, J. Panek, C. Duboc, D. Mathis, M. Novic and
T. R. Ward, J. Am. Chem. Soc., 2008, 130, 8085.
a lower enantiomeric excess in favor of the S sulfoxide. Fe-
24
7 S. de Lauzon, R. Quilez, L. Lion, B. Desfosses, B. Desfosses, I. Lee,
M. A. Sari, S. J. Benkovic, D. Mansuy and J. P. Mahy, Eur. J. Biochem.,
tetrapyrrolic compounds into resp. apomyoglobin and BSA lead
to enantiomeric excesses ranging from 5 to 38%, whereas the
insertion of an iron-porphyrin into an anti-steroid antibody, using
1
998, 257, 121.
8 R. Ricoux, E. Girgenti, H. Sauriat-Dorizon, D. Blanchard and J. P.
25
Mahy, J. Protein Chem., 2002, 21, 473.
the “Trojan horse” strategy, only led to a 8% enantiomeric excess.
9
R. Ricoux, E. Lukowska, F. Pezzotti and J. P. Mahy, Eur. J. Biochem.,
It is noteworthy that the enantiomeric excess can be increased
by covalent anchoring of metal complexes into apomyoglobin
2
004, 271, 1277.
1
0 H. Sato, T. Hayashi, T. Ando, Y. Hisaeda, T. Ueno and Y. Watanabe,
2
6–28
and mutants
or by changing the metal, such as for example
J. Am. Chem. Soc., 2004, 126, 436.
11 T. Komatsu, S. Ishihara, E. Tsuchida, H. Nishide, C. Morokuma and
24,26–28
replacing iron by manganese.
However, the best system
S. Nakamura, Biomacromolecules, 2005, 6, 1489.
is simply obtained by incorporation of the VOSO
4
ion into
1
2 R. M. Wang, T. Komatsu, A. Nakagawa and E. Tsuchida, Bioconjugate
Chem., 2005, 16, 23.
6
steptavidin, which leads to an ee value of 93%. These strategies
will be applied to our new hybrid metalloproteins constructed from
xylanase A and to extend their use to the selective oxidation of
other substrates including phenol derivatives, alkenes and alkanes.
13 R. Ricoux, R. Dubuc, C. Dupont, J. D. Marechal, A. Martin, M. Sellier
and J. P. Mahy, Bioconjugate Chem., 2008, 19, 899.
1
4 L. A. Damani, Sulphur-Containing Drugs and Related Compounds –
Chemistry Biochemistry and Toxicology. Metabolism of Sulphur Func-
tional Groups, 1989, Vol. 1, Ellis Horwood, Chichester.
1
1
5 A. H. Hofstra and J. P. Uetrecht, Toxicology, 1993, 82, 221.
6 A. D. Adler, F. R. Longo, J. D. Finarelli, J. Goldmacher, J. Assour and
L. Korsakoff, J. Org. Chem., 1967, 32, 476.
Acknowledgements
We thank the Minist e` re des Relations Internationales du Qu e´ bec,
the Consulat G e´ n e´ ral de France in Qu e´ bec and the 61 Commis-
sion Permanente de Coop e´ ration Franco-Qu e´ becoise for a two-
years collaboration grant between the team of Prof. J-P. Mahy
17 E. B. Fleischer, J. M. Palmer, T. S. Srivastava and A. Chatterjee, J. Am.
Chem. Soc., 1971, 93, 3162.
1
th
8 A. Harada, H. Fukushima, K. Shiotsuki, H. Yamaguchi, F. Oka and
M. Kamachi, Inorg. Chem., 1997, 36, 6099.
9 Z. Dong and P. J. Scammells, J. Org. Chem., 2007, 72, 9881.
1
(
France) and that of Prof. C. Dupont (Quebec) that helped
20 V. E. Yushmanov, Inorg. Chem., 1999, 38, 1713.
2
1 E. Baciocchi, M. F. Gerini, O. Lanzalunga, A. Lapi and M. G. Lo
Piparo, Org. Biomol. Chem., 2003, 1, 422.
2 L. Casella, M. Gullotti, R. Ghezzi, S. Poli, T. Beringhelli, S. Colonna
and G. Carrea, Biochemistry, 1992, 31, 9451.
to develop this project. Dr J.-D Marechal thanks the financial
support of the Spanish Ministry of Science and Technology
through the Consolider Ingenio 2010 Project No. CSD2007–
2
00006.
23 C. Letondor and T. R. Ward, ChemBioChem, 2006, 7, 1845.
2
2
4 A. Mahammed and Z. Gross, J. Am. Chem. Soc., 2005, 127, 2883.
5 Q. Raffy, R. Ricoux and J. P. Mahy, Tetrahedron Lett., 2008, 49,
1865.
6 J. L. Zhang, D. K. Garner, L. Liang, Q. Chen and Y. Lu, Chem.
Commun., 2008, 1665.
Notes and references
2
1
2
3
J. A. Sigman, B. C. Kwok and Y. Lu, J. Am. Chem. Soc., 2000, 122,
192.
I. Hara, T. Ueno, S. OzakiItoh, K. Lee, N. Ueyama and Y. Watanabe,
J. Biol. Chem., 2001, 276, 36067.
T. Hayashi, Y. Hitomi, T. Ando, T. Mizutani, Y. Hisaeda, S. Kitagawa
and H. Ogoshi, J. Am. Chem. Soc., 1999, 121, 7747.
8
27 M. Ohashi, T. Koshiyama, T. Ueno, M. Yanase, H. Fujii and Y.
Watanabe, Angew. Chem., Int. Ed., 2003, 42, 1005.
28 J. R. Carey, S. K. Ma, T. D. Pfister, D. K. Garner, H. K. Kim, J. A.
Abramite, Z. Wang, Z. Guo and Y. Lu, J. Am. Chem. Soc., 2004, 126,
10812.
This journal is © The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3208–3211 | 3211