Formation of a chiral center and pyrimidal inversion at the single-molecule level

Article abstract of DOI:10.1002/anie.200700736

(Chemical Equation Presented) One-at-a-time stereochemistry: As-S bond-making and bond-breaking reactions are important in pharmacology, toxicology, and cell biology. The aqueous covalent chemistry of individual As^III molecules can be monitored

Full text of DOI:10.1002/anie.200700736

Communications
DOI: 10.1002/anie.200700736
Chiral Organoarsenic Compounds
Formation of a Chiral Center and Pyrimidal Inversion at the
Single-Molecule Level**
Seong-Ho Shin, Mackay B. Steffensen, Tim D. W. Claridge, and Hagan Bayley*
The recent development of single-molecule techniques has
been largely targeted at solving biological problems;[1–5] for
example, protein folding has been examined by means of
force spectroscopy, and enzyme kinetics have been inves-
tigated by single-molecule fluorescence. These approaches
provide insight into individual trajectories that might be
obscured in ensemble measurements. In contrast, bond-
forming and bond-breaking reactions of small molecules in
solution have rarely been observed at the single-molecule
level. We have developed an approach through which the
covalent chemistry of individual molecules can be monitored
in an aqueous environment inside a “nanoreactor”, that is, the
transmembrane protein pore formed by a-hemolysin.[6–9] By
monitoring an ionic current driven through the pore by a
transmembrane potential, subtle changes in the structures of
individual reactants tethered within the lumen can be
detected. We now use this approach to follow both the
formation of a chiral center at AsIII (through the making of an
(residue 137) and observed two distinct conductance levels,
À
which corresponded to the formation of two As S adducts at
the wall of the protein (see Figure 1b). Re-examination of the
earlier data from the reaction at position 117 also revealed
two current levels, but they were very closely spaced. The
unitary conductance of P137-SH—a heteromeric pore with just
one of the seven subunits containing a cysteine residue at
position 137—in 2m KCl at À50 mV is (1.71 Æ 0.04) nS (this is
À
As S bond) and the inversion taking place at that center. The
nanoreactor also serves to isolate the relatively short-lived
AsIII adduct and thereby prevent additional reaction steps
À
that would complicate the chemistry. As S bond making and
breaking are important reactions in pharmacology, toxicol-
ogy,[10–13] and experimental cell biology.[14,15]
À
We earlier reported reversible covalent As S bond
formation between the side chain of a cysteine residue at
position 117 in the a-hemolysin pore and 4-sulfophenylar-
sonous acid (Figure 1a).[9] Herein, we examine the same
reaction at a different position within the nanoreactor
[*] Dr. S.-H. Shin,[+] Dr. M. B. Steffensen,[#] Dr. T. D. W. Claridge,
Dr. H. Bayley
Department of Chemistry
University of Oxford
Oxford, OX1 3TA (UK)
Fax: (+44)1865-275-708
E-mail: hagan.bayley@chem.ox.ac.uk
Homepage: http://www.chem.ox.ac.uk/bayleygroup/
Figure 1. Two adducts are formed when 4-sulfophenylarsonous acid
reacts with the side chain of Cys-137 in the P137-SH pore. a) Structure of
4-sulfophenylarsonous acid and schematic of the P137-SH pore. In
aqueous solution, the dehydrated form, that is, 4-sulfophenylarsane
oxide, may exist in equilibrium with 4-sulfophenylarsonous acid. The
P137-SH pore consists of six cysteine-free wild-type subunits (green) and
one mutated subunit (blue) in which Gly137 is substituted by Cys. The
sulfur atom of the Cys residue is shown in yellow, the carboxy oxygen
of residue 137 in red. b) Single-channel recording at À50 mV and 238C
with a buffer containing 2m KCl, 80mm 3-(4-morpholinyl)-1-propane-
sulfonic acid (MOPS), 100mm ethylenediaminetetraacetate (EDTA)
(pH 8.4) in both chambers. 4-Sulfophenylarsonous acid (100mm) was
in the trans chamber. The current levels corresponding to the free
pore, P137-SH, and the two adducts, A and B, are shown for the depicted
experiment. The assignment of the diastereomeric adducts as A or B
is arbitrary. The asterisks represent inversion events; these events are
over-represented in the trace shown here. c) Kinetic scheme describing
the observed chemistry (R=4-sulfophenyl).
[+] Present address:
Lawrence Berkeley National Laboratory
Materials Sciences Division, 1 Cyclotron Road
Berkeley, CA 94720 (USA)
[#] Present address:
Southern Utah University, Physical Science Department
351 West Center, Cedar City, UT 84720 (USA)
[**] Supported by the MRC and the ONR. H.B. is the holder of a Royal
Society–Wolfson Research Merit Award. M.S. was the holder of a
Ruth L. Kirschstein NIH Postdoctoral Fellowship(F32L078236). We
thank Dr. Stephen Cheley for the plasmids pT7-RL3 and pT7-RL3-
D8.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7412
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7412 –7416

Angewandte
Chemie
the mean value obtained from n = 10 separate experiments).
The two adducts formed by 4-sulfophenylarsonous acid
reduce the single-pore current (mean value 85.5 pA) by
(3.7 Æ 0.2) pA (n = 10, adduct A) and (2.5 Æ 0.2) pA (n = 10,
adduct B). The two adducts, A and B, interconvert, although
these interconversion events are rare relative to the rates of
association and dissociation of 4-sulfophenylarsonous acid.
The interconversion events represent (7.8 Æ 1.7)% of all
other current steps, that is, 100 ” (nAB + nBA)/(nPA+nAP+nPB
+
nBP; see the Supporting Information), and are over-repre-
sented in the selected trace (Figure 1b). Based on known[16–21]
and supporting ensemble chemistry, and on the kinetic
À
analysis presented below, we believe that the As S adducts A
and B are the diastereomers[22–24] formed at the surface of the
chiral protein by substitution of one or the other of the
enantiotopic hydroxy groups of 4-sulfophenylarsonous acid.
Because our measurements were carried out under conditions
of dynamic equilibrium, the situation can be represented with
a simple kinetic scheme (Figure 1c). The kinetic analysis
outlined below and the fact that A and B interconvert
eliminate the possibility that A and B are two distinct reaction
products, one formed, for example, by an impurity in the
arsonous acid. Additional experiments showed that buffer
components did not take part in the reaction and that the
Figure 2. Reciprocals of the mean interevent intervals and the dwell
times (t values) versus the concentration of 4-sulfophenylarsonous
À1
acid. a) Reciprocals of the mean interevent intervals (! toAn
;
~ tBonÀ1). toAn is the total time in the unmodified P137-SH state divided by
the number of exits from P137-SH to adduct A (nPA) and tBon is the total
time in the unmodified P137-SH state divided by the number of exits
from P137-SH to adduct B (nPB). b) Reciprocals of the mean dwell times
in states A or B before dissociation (! tAoffÀ1; ~ toBffÀ1). toAff is the total
time in state A divided by the number of exits from A to P137-SH (nAP
)
À À À À
formation of an O As S adduct did not occur, involving
and tBoff is the total time in state B divided by the number of exits from
B to P137-SH (nBP). c) Reciprocals of the mean dwell times in states A or
B before inversion (! tABÀ1; ~ tBAÀ1). tAB is the total time in state A
divided by the number of exits from A to B (nAB) and tBA is the total
time in state B divided by the number of exits from B to A (nBA). For
further details see the Supporting Information.
the hydroxy group of the proximal Thr-125 on the neighboring
subunit of the a-hemolysin heptamer (see the Supporting
Information).
The six rate constants in the proposed scheme were
determined by the analysis of mean inter-event intervals (tON
)
and mean adduct lifetimes (tOFF) (Table 1; for details, see also
Table 1: Rate constants[a] for the formation (kON) of As S adducts within
À
the P137-SH pore and their interconversion (kINV) and dissociation (kOFF).[b]
A
B
kON [mÀ1 sÀ1
kOFF [sÀ1
Kf [mÀ1
kINV [sÀ1
]
(14Æ1)”103
1.5Æ0.1
(5.9Æ0.8)”103
0.40Æ0.02
]
]
(9.3Æ0.9)”103
(15Æ2)”103
]
k
AB =0.071Æ0.007
kBA =0.049Æ0.007
[a] For conditions see Figure 1 (legend). [b] Kf (=kON/kOFF) is the
formation constant for each adduct.
the Supporting Information). Plots of 1/tON versus the
concentration of 4-sulfophenylarsonous acid (Figure 2a)
were of the form 1/t = k[A], which shows that the formation
of the proposed adducts involves bimolecular reactions. As
expected, analogous plots for the reversal of the adduct
formation and the interconversion of the adducts were of the
form 1/t = k, thus confirming that these are unimolecular
reactions (Figure 2b,c). As a further test of the kinetic
scheme, we note that at equilibrium, the product of the rate
constants for a clockwise movement around the triangle
(Figure 1c) must equal the product for the anticlockwise
Figure 3. Structure of various S-adducts of 4-sulfophenylarsonous acid.
a) 1:1 adduct with N-acetyl-l-cysteine methyl ester; b) 2:1 adduct with
N-acetyl-l-cysteine methyl ester; c) 1:1 adduct with glutathione (GSH);
d) 2:1 adduct with GSH. All molecules are tetrahedral at the As center,
but the configuration is not defined at the stereogenic As in com-
pounds (a) and (c).
movement.[25] In keeping with this, we found koAn kAB kBoff
=
(0.39 Æ 0.05) ” 103 mÀ1 sÀ3 and koBn kBA kAoff = (0.43 Æ 0.08) ”
103 mÀ1 sÀ3. We also found that the formation constants of A
and
B
(KAf = koAn/koAff = (9.3 Æ 0.9) ” 103 mÀ1
, =
KBf = kBon/kBoff
(15 Æ 2) ” 103 mÀ1) are similar to that of the adduct between
Angew. Chem. Int. Ed. 2007, 46, 7412 –7416
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7413

Communications
4-sulfophenylarsonous acid and Cys117, namely, Kf = (14 Æ
(GSH, which is the tripeptide g-Glu–Cys–Gly). We titrated
GSH with 4-sulfophenylarsonous acid (and vice versa) in a
100mm sodium phosphate solution (pD 7.8). With the aid of
3) ” 103 mÀ1 [9]
.
Therefore, the assignment of the current levels
and transitions makes sense in terms of the kinetic scheme.
Furthermore, if it is assumed that the
“concentration” of 4-sulfophenylarsonous
acid in the lumen of the pore (that is, the
probability of finding the reactant in a
unit volume of solution) is equal to that in
free solution—an assumption that has
been borne out by experiment[8]—the
bimolecular rate constants found herein
should approximate the solution rate
constants for the reaction of thiols and
arsonous acids.
Electrospray ionization time-of-flight
mass spectrometry (ESI-TOF-MS) and
1HNMR experiments provided informa-
tion about the formation of covalent thiol
adducts by 4-sulfophenylarsonous acid
and their chirality. The monothiol N-
acetyl-l-cysteine methyl ester was mixed
with 4-sulfophenylarsonous acid in water.
The solution was allowed to stand (at
238C) for 5 min before performing the
ESI-TOF-MS experiments in the nega-
tive-ion mode. At a thiol/AsIII ratio of 1:2,
a peak at m/z 423.83 appeared in the
spectrum, which was assigned to the 1:1
adduct (Figures 3a and 4a). At a higher
thiol/AsIII ratio (of 4:1), the major peak
was observed at m/z 582.82 and assigned
to the 2:1 adduct (Figure 3b, theoretical
582.99). The two remaining prominent
peaks at this ratio were that of the 1:1
adduct (m/z 423.84, theoretical 423.95)
and that of the dehydrated monoanionic
form of the arsonous acid (namely, 4-
sulfophenylarsane oxide m/z 246.86, the-
oretical 246.91).
The peaks at m/z 583 and 424 were
examined by tandem mass spectrometry
(MS–MS, Figure 4b,c). In both cases,
elimination to form N-acetyl-l-dehydro-
cysteine[26] yielded peaks at m/z 262.83
(corresponding to the mono-anion of 4-
sulfophenylarsane sulfide, theoretical
262.88). In the case of the peak at m/z
424, breakdown to form a species with
m/z 246.87 (that is, the mono-anion of 4-
sulfophenylarsane oxide) also occurred
(Figure 4b).
These
observations
strengthen the assignments of the peaks
at m/z 583 and 424 as being derived from
the 2:1 and 1:1 adducts of N-acetyl-l-
cysteine methyl ester with 4-sulfopheny-
larsonous acid.
Figure 4. MS data supporting the interpretation of the single-molecule experiments. a) Neg-
ative-ion ESI-TOF-MS of mixtures (in various ratios) of 4-sulfophenylarsonous acid and the
monothiol N-acetyl-l-cysteine methyl ester in water. b) MS–MS of the ion at m/z 424. c) MS–
1HNMR spectroscopy was used to
investigate the interaction of 4-sulfophe-
nylarsonous acid with l-glutathione MS of the ion at m/z 583.
7414
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7412 –7416

Angewandte
Chemie
2D correlation spectroscopy (COSY), we then assigned all the
1H resonances to individual molecular species (see the
Supporting Information). The titrations suggested that the
GSH–4-sulfophenylarsonous acid 2:1 adduct (Figure 3d) was
more stable than the 1:1 adduct (Figure 3c)—as found
elsewhere for related compounds.[19] Peaks that were assigned
to the 1:1 adduct were apparent in the presence of an excess of
arsonous acid (Figure 5a), but were of low intensity when
a 100mm sodium phosphate solution (pD 8.8), two of the
À
À
SCH2 resonances of the 2:1 complex, which arise from
protons on different cysteine Cb atoms (according to the 2D
COSYexperiment), coalesced into a single broad peak, whilst
À
À
the two remaining SCH2 resonances of near-identical
chemical shifts fully coalesced and sharpened to a single
double doublet (see Figure 5c and the Supporting Informa-
tion). Simulation of the spectra from a temperature series
yielded a rate constant for the exchange process of approx-
imately 8 sÀ1 at 308C (see the Supporting Information). The
À
À
exchange process for the same two SCH2 resonances was
also examined in greater detail in a 1D magnetization-transfer
experiment (in 100mm sodium phosphate, pD 8.8), which
yielded a rate constant of 4.7 sÀ1 at 308C (see the Supporting
Information). These experiments demonstrate that sulfur
ligands at the As center of the 2:1 adduct with GSH exchange
À
at approximately the same rate as that observed for As S
bond cleavage in the 1:1 adduct seen in the single-molecule
experiments.
Pyrimidal inversion at an arsenic center with three carbon
substituents is presumed to occur without bond breaking and
is extremely slow;[27–29] for example, ethylmethylphenylarsine
has a half-life for racemization in decalin of about 6 days at
2188C.[27] In contrast, AsIII compounds with Si substituents
isomerize faster; for example, C6H5As[SiH(CH3)2]2 has a
solvent-independent barrier to inversion of 74 kJ molÀ1
,
1
À
À
Figure 5. HNMR spectra illustrating the cysteine methylene ( SCH2
)
which corresponds to a half-life for racemization of about
1 s at 258C.[30] Although detailed kinetic studies are lacking,
AsIII compounds with S substituents also isomerize faster; for
example, stereoisomers of the cyclic 1:1 adducts of phenyl-
dichloroarsine with 1,3-dimercapto-2-propanol and 1,2-
dimercaptopropane interconvert with rate constants in the
range of 0.42 to 3.1 sÀ1 at about 258C in acidic CD3OD,[20] in
contrast with earlier measurements in neutral apolar sol-
vents.[31] The cyclic diastereomeric adducts between phenyl-
arsonous acid and a dicysteine peptide isomerized in aceto-
nitrile–water (pH ꢀ 2) with t1/2 ꢀ 40 h.[32] The diastereomeric
adducts between methylphenylarsinous acid and glutathione
could be separated by means of high-performance liquid
chromatography (HPLC), but were interconverted upon
removal of the solvent.[22]
While these reactions may proceed without bond break-
ing, as suggested for ethylmethylphenylarsine,[27] an alterna-
tive is a dissociative mechanism, in which one of the bonds to
As is broken and reformed.[20,24,32] The case for such a
mechanism is bolstered by the existence of arsenium cat-
ions,[33,34] which would be intermediates in the process.
Interestingly, a “thiaarsahydroxy” species was observed in
the active site of an arsenate reductase,[35] and [RAsOH]+
(where R = 3-nitro-4-hydroxyphenyl) was found in the ESI-
MS of the oxidized form of Ehrlichꢀs organoarsenic thera-
peutic salvarsan.[11] Herein, we did not observe any inter-
mediate that might be ascribed to an arsenium cation at the
time resolution of the experiments (that is, 50 ms). Finally, in
the presence of excess ligand (in this case water), an
associative mechanism for ligand exchange and racemization
is a possibility.
resonances in 1:1 and 2:1 adducts of glutathione (GSH) and 4-
sulfophenylarsonous acid. a) A mixture containing 1:1 and 2:1 adducts
(2.3mm GSH: 30mm 4-sulfophenylarsonous acid, 258C). Labeled are
the eight double-doublet resonances observed for the 2:1 complex
[upper labels A–D; see also trace (b)] and the two diastereomeric 1:1
complexes (lower labels E–H). b) A mixture in which the 2:1 adduct is
dominant (28.2mm GSH: 30mm 4-sulfophenylarsonous acid, 258C).
c) Spectrum of the 2:1 adduct at 808C (18mm GSH: 21mm 4-
sulfophenylarsonous acid). The NMR spectra were recorded in a
100mm sodium phosphate solution [pD 7.8 in (a) and (b), and pD 8.8
in (c)].
both reagents were present at a concentration of about 30mm
À
À
(Figure 5b). The cysteine methylene ( SCH2 ) resonances
(Figure 5b) in the 2:1 adduct can be ascribed to four
chemically non-equivalent protons, which is consistent with
the fact that this adduct is tetrahedral at the arsenic center
and contains two identical ligands of the same chirality[22,23]
(see the Supporting Information). In contrast, the 1:1 adducts
comprise a pair of diastereomers, each of which is stereogenic
at As and exhibits diastereotopic methylene protons. The
1
À
À
H NMR resonances for the SCH2 group of each diaste-
reomer appear as a pair of double doublets, thus giving a total
of sixteen lines. All of these peaks could be identified in the
1HNMR spectrum of a mixture of the 2:1 and 1:1 adducts by
analyzing the titration data (see Figure 5a and the Supporting
Information). It is worth emphasizing that a 1:1 species is
isolated within the nanoreactor in our single-molecule
approach.
À
To shed light on the kinetics of As S bond formation and
dissociation, we examined the temperature dependence of the
1HNMR spectrum and performed a magnetization-transfer
experiment (see the Supporting Information). At 808C in
À
The inversion is slow relative to the breakdown of the As
S adducts, and it is instructive to note that this infrequent
Angew. Chem. Int. Ed. 2007, 46, 7412 –7416
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7415

Communications
[10] A. J. Nok, Parasitol. Res. 2003, 90, 71.
pathway would not be observed in most ensemble measure-
[11] N. C. Lloyd, H. W. Morgan, B. K. Nicholson, R. S. Ronimus,
Angew. Chem. 2005, 117, 963; Angew. Chem. Int. Ed. 2005, 44,
941.
[12] E. Lengfelder, S. Saussele, A. Weisser, T. Buchner, R. Hehl-
mann, Crit. Rev. Oncol. Hematol. 2005, 56, 261.
[13] S. Kapaj, H. Peterson, K. Liber, P. Bhattacharya, J. Environ. Sci.
Health Part A 2006, 41, 2399.
[14] B. R. Martin, B. N. Giepmans, S. R. Adams, R. Y. Tsien, Nat.
Biotechnol. 2005, 23, 1308.
[15] B. A. Griffin, S. R. Adams, R. Y. Tsien, Science 1998, 281, 269.
[16] R. A. Rey, O. W. Howarth, E. C. Pereira-Maia, J. Inorg.
Biochem. 2004, 98, 1151.
ments, such as a dynamic NMR spectroscopy experiments
where dissociation and re-association of the sulfur ligand
would dominate the signal. Furthermore, in bulk solution, the
initial adduct would react with a second thiol, thereby
complicating the chemistry.[16–19] In the a-hemolysin nano-
reactor, the initial adduct is isolated at the reaction site.
Finally, the surface of the protein on which inversion occurs
has a weak preference for one of the two diastereomers (kAB
/
k
BA = 1.4). It would be interesting to attempt to strengthen
that preference by the appropriate positioning of neighboring
groups.
[17] N. Scott, K. M. Hatlelid, N. E. MacKenzie, D. E. Carter, Chem.
Res. Toxicol. 1993, 6, 102.
[18] K. Dill, E. R. Adams, R. J. OꢀConnor, S. Chong, E. L. McGown,
Arch. Biochem. Biophys. 1987, 257, 293.
[19] A. M. Spuches, H. G. Kruszyna, A. M. Rich, D. E. Wilcox, Inorg.
Chem. 2005, 44, 2964.
[20] K. Dill, L. Huang, D. W. Bearden, E. L. McGown, R. J.
OꢀConnor, Chem. Res. Toxicol. 1991, 4, 295.
[21] M. Delnomdedieu, M. M. Basti, J. D. Otvos, D. J. Thomas, Chem.
Res. Toxicol. 1993, 6, 598.
Experimental Section
Procedures for MS and NMR spectroscopy, additional MS and NMR
data, procedures for site-directed mutagenesis and the preparation of
protein pores, the kinetic analysis, and control experiments that
eliminate the participation in the chemistry of buffer components and
neighboring groups on the nanoreactor surface are available in the
Supporting Information.
[22] J. S. Edmonds, T. Nakayama, T. Kondo, M. Morita, Magn. Reson.
Chem. 2006, 44, 151.
Received: February 17, 2007
Revised: May 30, 2007
Published online: August 14, 2007
[23] T. Nakayama, T. Isobe, K. Nakamiya, J. S. Edmonds, Y. Shibata,
M. Morita, Magn. Reson. Chem. 2005, 43, 543.
[24] A. von Döllen, H. Strasdeit, Eur. J. Inorg. Chem. 1998, 61.
[25] E. A. Richard, C. Miller, Science 1990, 247, 1208.
[26] M. Piraud, C. Vianey-Saban, K. Petritis, C. Elfakir, J. P. Steghens,
A. Morla, D. Bouchu, Rapid Commun. Mass Spectrom. 2003, 17,
1297.
[27] G. H. Senkler, K. Mislow, J. Am. Chem. Soc. 1972, 94, 291.
[28] R. D. Baechler, J. D. Andose, J. Stackhouse, K. Mislow, J. Am.
Chem. Soc. 1972, 94, 8060.
[29] J. D. Andose, A. Rauk, K. Mislow, J. Am. Chem. Soc. 1974, 96,
6904.
[30] R. D. Baechler, J. P. Casey, R. J. Cook, G. H. Senkler, K. Mislow,
J. Am. Chem. Soc. 1972, 94, 2859.
[31] J. P. Casey, K. Mislow, J. Chem. Soc. Chem. Commun. 1970, 999.
[32] B. A. Griffin, PhD thesis, University of California, San Diego
1998.
Keywords: arsenic · chirality · nanoreactors ·
.
organoarsenic chemistry · single-molecule studies
[1] G. Bokinsky, X. Zhuang, Acc. Chem. Res. 2005, 38, 566.
[2] X. Zhuang, Annu. Rev. Biophys. Biomol. Struct. 2005, 34, 399.
[3] X. Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay, S. Doose,
J. J. Li, G. Sundaresan, A. M. Wu, S. S. Gambhir, S. Weiss,
Science 2005, 307, 538.
[4] K. Kinosita, Jr., K. Adachi, H. Itoh, Annu. Rev. Biophys.
Biomol. Struct. 2004, 33, 245.
[5] E. J. Peterman, H. Sosa, W. E. Moerner, Annu. Rev. Phys. Chem.
2004, 55, 79.
[6] S.-H. Shin, H. Bayley, J. Am. Chem. Soc. 2005, 127, 10462.
[7] T. Luchian, S.-H. Shin, H. Bayley, Angew. Chem. 2003, 115, 1970;
Angew. Chem. Int. Ed. 2003, 42, 1928.
[8] T. Luchian, S.-H. Shin, H. Bayley, Angew. Chem. 2003, 115, 3896;
Angew. Chem. Int. Ed. 2003, 42, 3766.
[33] K. A. Porter, A. C. Willis, J. Zank, S. B. Wild, Inorg. Chem. 2002,
41, 6380.
[34] N. Burford, T. M. Parks, B. W. Royan, J. F. Richardson, P. F.
White, Can. J. Chem. 1992, 70, 703.
[35] P. Martin, S. DeMel, J. Shi, T. Gladysheva, D. L. Gatti, B. P.
Rosen, B. F. P. Edwards, Structure 2001, 9, 1071.
[9] S.-H. Shin, T. Luchian, S. Cheley, O. Braha, H. Bayley, Angew.
Chem. 2002, 114, 3859; Angew. Chem. Int. Ed. 2002, 41, 3707.
7416
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7412 –7416