Tuning the Stereoselectivity of a DNA-Catalyzed Michael Addition through Covalent Modification

Article abstract of DOI:10.1002/anie.201503838

Complexes of G-quadruplex DNA and Cu^IIions have previously been applied as catalysts in asymmetric reactions, but the largely unspecific and noncovalent nature of the interaction has impeded understanding of the structural basis of catalysis. T

Full text of DOI:10.1002/anie.201503838

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201503838
German Edition: DOI: 10.1002/ange.201503838
G-Quadruplexes
Tuning the Stereoselectivity of a DNA-Catalyzed Michael Addition
through Covalent Modification
Surjendu Dey and Andres Jäschke*
II
II
Abstract: Complexes of G-quadruplex DNA and Cu ions
enantioselectivity. However, in these studies, Cu bound to
have previously been applied as catalysts in asymmetric
reactions, but the largely unspecific and noncovalent nature
of the interaction has impeded understanding of the structural
basis of catalysis. To better control the formation of a catalyti-
cally competent species, DNA quadruplexes were derivatized
the G-quadruplex sequences in an unspecific, most likely
purely electrostatic fashion, thus making it impossible to
localize the catalytic Cu and the substrate binding pocket in
II
the G-quadruplex structure. Therefore, neither a detailed
understanding nor prediction of the factors that determine
activity and selectivity in these reactions is currently possible.
With dsDNA, there are some examples for the covalent
attachment of a transition-metal ligand to DNA, which allows
the placement of the metal center in a much more controlled
way and sets the stage for systematic structure–function
II
with linker-bpy-Cu complexes in a site-specific manner and
applied in asymmetric aqueous Michael additions. These
modified quadruplexes exhibited high rate acceleration and
stereoselectivity. Different factors were found to be important
for the catalytic performance of the modified G-quadruplexes,
among them, the position of modification, the topology of the
quadruplex, the nature of the ligand, and the length of the
linker between the ligand and DNA. Moving the same ligand
by just two nucleotides inverted the stereochemical outcome:
quadruplexes modified at position 10 formed the (À)-enantio-
mer with up to 92% ee, while DNA derivatized at position 12
formed the (+)-enantiomer with up to 75% ee. This stereo-
preference was maintained when applied to structurally differ-
ent Michael acceptors. This work demonstrates a new and
simple way to tune the stereoselectivity in DNA-based
asymmetric catalysis.
[10]
studies.
On the other hand, only one report has been
published for a covalently modified G-quadruplex catalyst
(modified with proline) for an aldol reaction, and no
[
11]
stereoselectivity was observed. To the best of our knowl-
edge, there are no reports on G-quadruplex structures
covalently modified with metal-binding ligands and their
application in asymmetric Michael addition reactions.
We reasoned that a number of different factors govern the
rate acceleration and stereoselectivity of G-quadruplex-
catalyzed reactions, among them the topology of the quad-
ruplex, the attachment site of the metal-chelating ligand, the
nature of the ligand, and the length and structure of the linker
connecting the DNA and ligand. Herein, we report the first
systematic investigation of these parameters in the context of
asymmetric Michael addition reactions, which are important
D
NA-based hybrid catalysis is gaining importance in
[
1]
aqueous-phase homogeneous asymmetric catalysis. In this
catalytic approach, double-stranded (ds) DNA plays a crucial
role in transferring chirality from its characteristic helical
structure to the product. This approach has proven highly
successful for different enantioselective reactions, such as
[12]
CÀC bond forming reactions, catalyzed by c-kit quadru-
plexes.
The 3D structure of the c-kit wild-type (wt) sequence (5’-
AGGGAGGGCGCTGGGAGGAGGG-3’) has been studied
[2]
[3]
Diels–Alder reactions, Friedel–Crafts reactions, Michael
additions, fluorinations, and hydrations. Beside the well-
known double helix structure, DNA can also be present as
triplex, hairpin, G-quadruplex, and i-motif structures. G-
quadruplex structures are topologically more complex and
[
4]
[5]
[6]
[13]
[14]
by both X-ray crystallography and NMR spectroscopy.
According to these studies, c-kit forms a unique parallel G-
quadruplex structure as illustrated in Figure 1a. For our
systematic investigations, we considered attaching the ligand
in two different ways. In one case, ligand attachment should
not affect the correct formation of the G-quadruplex stack,
while it should do so in the other case. For the first case, we
picked position 12, which is located in an apical loop segment,
since crystallographic investigations revealed no influence of
[7]
[
8]
offer more structural variety than dsDNA, thus making
them a promising platform for DNA-based hybrid catalysis.
Two well-characterized G-quadruplex-forming sequences,
namely the human telomeric G-quadruplex (h-Tel) and
a region of the c-kit promoter (c-kit), have already been
utilized for asymmetric Diels–Alder and Friedel–Crafts
[13]
a bulky substitution at this position on the folding topology.
II
[9]
reactions in the presence of Cu ions. Of these two, h-Tel
showed greater versatility in terms of conversion and
Position 10, on the other hand, was identified as important for
the unique all-parallel folding topology of the c-kit quad-
ruplex, and NMR spectra suggested a G10T mutant to favor
[
14]
an alternative topology. However, the exact structure(s) of
this mutant quadruplex c-kit-T10 are, however, unknown. A
hypothetical structure is shown in Figure 1c.
[
*] S. Dey, Prof. Dr. A. Jäschke
Institut für Pharmazie und Molekulare Biotechnologie
Universität Heidelberg, 69120 Heidelberg (Germany)
E-mail: jaeschke@uni-hd.de
We thus substituted the natural nucleotides at positions 10
or 12 with a deoxyuridine derivative covalently modified at
the C5-position with a bipyridine moiety attached via linkers
of different lengths (Figure 1b and d). Modified nucleosides
Homepage: http://www.jaeschke.uni-hd.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201503838.
Angew. Chem. Int. Ed. 2015, 54, 11279 –11282
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11279

.
Angewandte
Communications
[a]
Table 1: Michael addition catalyzed by dU12-modified c-kit DNA.
[b]
[b,c]
Entry DNA
Cu(NO3)2
mol%)
n
conv. [%]
ee [%]
(
1
2
3
4
5
6
7
8
9
–
–
–
10
–
–
–
–
–
1
2
3
4
4
<1
8
<5
6
94
92
94
95
22
0
0
c-kit(wt)
c-kit(wt)
Propargyl-bpy-dU12 10
Butynyl-bpy-dU12
Pentynyl-bpy-dU12
Hexynyl-bpy-dU12
Hexynyl-bpy-dUrd
<À5
À9
+31
+40
+43
+52
0
10
10
10
10
10
[
a] See the Experimental Section for detailed reaction conditions. All
experiments were performed in triplicate. [b] Both conversion and ee
values were calculated by using chiral-phase HPLC; results are
reproducible within Æ5%. [c]+ and À symbols refer to isomers with low
Figure 1. G-quadruplex folding of different DNA sequences in the
and high retention time, respectively, from a chiral-phase HPLC column.
II
[13,14]
presence of Cu . a) Observed folding of c-kit(wt) DNA.
b) Hypo-
thetical folding and metal binding of dU12-modified DNA. c) Hypo-
thetical folding of c-kit-T10 DNA. d) Hypothetical folding and metal
binding of dU10-modified DNA.
Table 2: Michael addition of 1a with 2 catalyzed by dU10-modified c-kit
DNA.
[a]
[b]
[b,c]
Entry DNA
Cu(NO3)2
mol%)
n
conv. [%]
ee [%]
(
1
2
3
4
5
6
c-kit-T10
c-kit-T10
Propargyl-bpy-dU10 10
Butynyl-bpy-dU10
Pentynyl-bpy-dU10
Hexynyl-bpy-dU10
–
10
–
–
1
2
3
4
<5
8
92
92
91
95
<À5
À11
À32
À57
À69
À92
were prepared by Sonogashira coupling of 5’-DMT-5-iodo-2’-
deoxyuridine and bipyridine alkynes with different linker
lengths (see the Supporting Information for details). By using
standard methodology, these four modified nucleosides were
protected and converted into phosphoramidites. Solid-phase
synthesis was applied to synthesize dU12- and dU10-modified
DNA sequences (see the Supporting Information). After
purification and analytical characterization, these modified
10
10
10
[
a], [b], [c] see footnotes to Table 1.
II
sequences were investigated along with Cu as catalysts in the
Next, we investigated the dU10-modified G-quadru-
asymmetric Michael addition of standard acceptor 1a and the
nucleophile dimethyl malonate (DMM, 2).
plexes, which likely fold differently. Again, near-quantitative
conversions were achieved (91–95%), and an increase in the
stereoselectivity with the linker length was observed (Table 2,
entries 3–6), reaching the practically useful 92% ee for
hexynyl-bpy-dU10 (Table 2, entry 6). As with c-kit(wt)
(Table 1, entries 3 and 4), negligible activity was measured
for the c-kit-T10 sequence without the ligand (Table 2,
entries 1 and 2). Surprisingly, for all dU10-modified quad-
ruplexes, the (À)-enantiomer was obtained in excess, which is
the opposite result to that obtained with the dU12-modified
sequences (Table 2 vs. Table 1).
While the buffer controls (Table 1, entries 1 and 2), and
the unmodified c-kit(wt) sequence (Table 1, entries 3 and 4)
showed negligible activity and stereoinduction in the absence
II
or presence of Cu , the four different dU12-modified G-
quadruplexes yielded near-quantitative conversion (92–95%)
with 10 mol% of catalyst (Table 1, entries 5–8). These results
II
indicate that bpy-coordinated Cu covalently attached to
quadruplex DNA is a better catalyst than copper ions
unspecifically bound to quadruplex DNA.
The stereoselectivity increased almost linearly with the
linker length from 31% for propargyl to 52% ee for the
hexynyl linker. In all cases, the (+)-enantiomer was obtained
in excess. To ensure that stereoinduction is caused by the
topology of the DNA scaffold and not by the chirality of the
modified nucleoside, the reaction was investigated in the
presence of the different linker-bpy-modified deoxyuridine
nucleosides, with or without added unmodified quadruplex
DNA (Table 1, entry 9, and Table S1 in the Supporting
Information), and much lower ee (always in favour of the
The c-kit quadruplex derivatized with the hexynyl-bpy
ligand in position 12 yielded 95% conversion and 52% excess
of the (+)-enantiomer, while the catalyst with the same ligand
attached to position 10 gave also 95% conversion but with
95% excess of the (À)-enantiomer, which was unpredicted
and not easy to rationalize (see Figure 2a,b). To get a first
structural insight into the molecular basis of this phenomenon,
CD spectra were recorded. Figure 2c presents the CD spectra
of c-kit(wt) and hexynyl-bpy-dU12 sequences in the absence
II
and presence of Cu . These spectra are virtually identical and
(
+)-enantiomer) and conversion values were observed. These
show the typical features of a parallel quadruplex, namely
a minimum around 240 nm and a maximum around 260 nm,
thus suggesting that neither the attachment of the linker-
findings demonstrate that the ligand-modified quadruplexes
provide both rate acceleration and stereoselectivity.
1
1280 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 11279 –11282

Angewandte
Chemie
Inspired by the primary results, we chose
hexynyl-bpy-dU12 and hexynyl-bpy-dU10 com-
II
bined with Cu as catalysts to explore the Michael
addition reaction in more depth with different
acceptor substrates 1a–g (Table 4). In all cases, we
observed preference for the (+)-enantiomers of
the products (3a–g) when using hexynyl-bpy-dU12
as the catalyst and a preference for the (À)-
enantiomers when using hexynyl-bpy-dU10 (see
Figures S2–S8). The two catalysts thus show
defined preferences for stereoselective bond for-
mation, which they keep when they are confronted
with varying substrates. Except for substrate 1f
(which is a poor acceptor owing to the strongly
electron-withdrawing CF3 group), all of the other
substrates showed good to excellent conversion in
the presence of both hexynyl-bpy-dU12 and
hexynyl-bpy-dU10. Substrate 1g turned out to be
best for hexynyl-bpy-dU12; while substrate 1a was
most appropriate for hexynyl-bpy-dU10 in terms
of enantioselectivity.
In conclusion, we found that G-quadruplex
DNA sequences covalently modified with linker-
bpy-Cu complexes are efficient catalysts for
asymmetric Michael addition reactions in water.
Quadruplex topology, the ligand and its attach-
ment site, as well as the length of the linker
Figure 2. Correlation between quadruplex topology and stereoselectivity in the
Michael addition. a) Separation of product 3a enantiomers for the Michael addition
catalyzed by c-kit(wt) and hexynyl-bpy-dU12. b) Separation of product 3a enantio-
mers for the Michael addition catalyzed by c-kit-T10 and hexynyl-bpy-dU10. c) CD
II
II
spectra of c-kit(wt) and hexynyl-bpy-dU12 in the absence and presence of Cu . d) CD
II
spectra of c-kit-T10 and hexynyl-bpy-dU10 in the absence and presence of Cu .
bound bpy ligand nor the binding of copper disturbs the
Table 3: Dependence of the DNA-catalyzed Michael addition of 1a with 2
on catalyst loading.
folding. CD spectra of other dU12-modified sequences
displayed the same features, thus indicating the same type
of structural folding (Figure S9). For the dU10 quadruplexes
[a]
[d]
[b]
[b,c]
Entry
DNA
Cu(NO3)2
mol%)
conv. [%]
ee [%]
(
(
Figure 2d), the spectra display the features of a parallel
II
1
2
3
4
5
6
7
8
Hexynyl-bpy-dU12
Hexynyl-bpy-dU12
Hexynyl-bpy-dU12
Hexynyl-bpy-dU12
Hexynyl-bpy-dU10
Hexynyl-bpy-dU10
Hexynyl-bpy-dU10
Hexynyl-bpy-dU10
10
5
2.5
1
10
5
95
81
72
46
95
95
94
83
+52
+52
+52
+52
À92
À92
À92
À92
quadruplex in the absence of Cu for c-kit-T10. The addition
of copper ions to c-kit-T10 leads to attenuation of both the
2
40 and the 260 nm bands and the appearance of a small
shoulder around 290 nm, features often observed in hybrid
[
15,16]
structures with mixed parallel/anti-parallel strands.
hexynyl-bpy-dU10 sequence shows a small shoulder around
90 nm that becomes more prominent upon the addition of
The
2.5
1
2
II
Cu . Again, other dU10-modified sequences yielded very
similar CD spectra (Figure S10). Thus, while not providing the
same level of information as high-resolution structures, the
CD spectra support our design assumption that dU10- and
dU12-modified c-kit quadruplexes fold into different catalyti-
[
a], [b], [c] see footnotes to Table 1. [d] DNA concentration was varied in
proportion to that of Cu(NO ) .
3
2
[
a]
Table 4: Substrate variation for the Michael addition reaction.
II
cally competent structures in the presence of Cu , which
cause the opposing stereochemical outcomes of the reactions.
To further investigate the G-quadruplex-catalyzed
Michael reactions, we compared the two best cases, namely
the hexynyl-bpy-modified dU10 and dU12 catalysts, with
respect to conversion and stereoselectivity at decreasing
catalyst loadings (Table 3). Surprisingly, the dU10 catalyst, in
which the regular quadruplex folding has been disturbed,
turns out to be the better catalyst: even at 1 mol% catalyst,
Entry Substrate, R Hexynyl-bpy-dU12
Hexynyl-bpy-dU10
[b]
[b,c]
[b]
[b,c]
conv. [%]
ee [%]
conv. [%]
ee [%]
1
2
3
4
5
6
7
H (1a)
p-Me (1b)
p-OMe (1c) 96
p-Br (1d) 65
o-OMe (1e) 96
95
92
+52
+55
+42
+62
+51
+46
+75
95
90
97
75
92
10
98
À92
À85
À71
À83
À61
À45
À66
8
1
3% conversion is measured, giving a turnover frequency of
À1
.2 h . The dU12 catalyst gave 46% conversion under these
À1
o-CF
3
(1 f)
13
99
conditions, which equates to a turnover frequency of 0.6 h .
In both cases the enantioselectivity was not affected by
variation of the catalyst loading.
o-Br (1g)
[a], [b], [c] see footnotes to Table 1.
Angew. Chem. Int. Ed. 2015, 54, 11279 –11282
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11281

.
Angewandte
Communications
connecting the DNA and ligand were all found to have an
effect on the reaction outcome. The finding that longer linkers
yield higher ee values is counterintuitive. This result might be
due to an optimal stacking orientation of the bpy ligand upon
the G-quadruplex for stereoselective catalysis, which is more
easily achieved with longer linkers. Importantly, the stereo-
selectivity of the reaction can be easily tuned by selecting
different attachment positions on the DNA sequence. Moving
the same linker-derivatized bpy ligand by just two nucleotides
completely changed the stereochemical outcome of the
reaction: while quadruplexes modified at position 10 prefer-
entially formed the (À)-enantiomer with high ee, those
derivatized at position 12 preferentially formed the (+)-enan-
tiomer. For each of these catalysts, the stereopreference was
conserved across a panel of differently substituted Michael
acceptors. This first example of covalently modified quad-
ruplex DNA catalysts and the reported dependence of the
catalytic performance on structural variations expands the
scope of DNA-based hybrid catalysis.
A. J. Boersma, B. L. Feringa, Chem. Commun. 2006, 635 – 637;
c) A. J. Boersma, B. L. Feringa, G. Roelfes, Org. Lett. 2007, 9,
3
647 – 3650; d) J. Wang, E. Benedetti, L. Bethge, S. Vonhoff, S.
Klussmann, J. J. Vasseur, J. Cossy, M. Smietana, S. Arseniyadis,
Angew. Chem. Int. Ed. 2013, 52, 11546 – 11549; Angew. Chem.
2
013, 125, 11760 – 11763.
[
[
3] a) A. J. Boersma, B. L. Feringa, G. Roelfes, Angew. Chem. Int.
Ed. 2009, 48, 3346 – 3348; Angew. Chem. 2009, 121, 3396 – 3398;
b) S. Park, K. Ikehata, R. Watabe, Y. Hidaka, A. Rajendran, H.
Sugiyama, Chem. Commun. 2012, 48, 10398 – 10400.
4] a) D. Coqui›re, B. L. Feringa, G. Roelfes, Angew. Chem. Int. Ed.
2
007, 46, 9308 – 9311; Angew. Chem. 2007, 119, 9468 – 9471;
b) R. P. Megens, G. Roelfes, Chem. Commun. 2012, 48, 6366 –
368; c) Y. H. Li, C. H. Wang, G. Q. Jia, S. M. Lu, C. Li,
Tetrahedron 2013, 69, 6585 – 6590.
[5] N. Shibata, H. Yasui, S. Nakamura, T. Toru, Synlett 2007, 1153 –
157.
6
1
[
6] A. J. Boersma, D. Coqui›re, D. Geerdink, F. Rosati, B. L.
Feringa, G. Roelfes, Nat. Chem. 2010, 2, 991 – 995.
[
7] a) R. D. Wells, J. Biol. Chem. 1988, 263, 1095 – 1098; b) R. D.
Wells, Trends Biochem. Sci. 2007, 32, 271 – 278; c) S. C. Ragha-
van, M. R. Lieber, Front. Biosci. 2007, 12, 4402 – 4408; d) A.
Bacolla, R. D. Wells, J. Biol. Chem. 2004, 279, 47411 – 47414;
e) A. Majumdar, D. J. Patel, Acc. Chem. Res. 2002, 35, 1 – 11.
8] a) Y. Qin, L. H. Hurley, Biochimie 2008, 90, 1149 – 1171; b) S.
Neidle, S. Balasubramanian, Quadruplex Nucleic Acids, Royal
Society of Chemistry, Cambridge, 2006; c) S. Burge, G. N.
Parkinson, P. Hazel, A. K. Todd, S. Neidle, Nucleic Acids Res.
[
Experimental Section
Standard Michael addition procedure: An aqueous stock solution
containing 40 nmol of the oligonucleotide was lyophilized. The
lyophilized DNA was then dissolved in a 297 mL solution of 3-(N-
morpholino)propanesulfonic acid (MOPS) buffer (20 mm, pH 7)
containing KCl (100 mm) then 3 mL Cu(NO ) (10 mm) solution was
2
006, 34, 5402 – 5415; d) A. N. Lane, J. B. Chaires, R. D. Gray,
J. O. Trent, Nucleic Acids Res. 2008, 36, 5482 – 5515.
3
2
II
[9] a) S. Roe, D. J. Ritson, T. Garner, M. Searle, J. E. Moses, Chem.
Commun. 2010, 46, 4309 – 4311; b) C. H. Wang, G. Q. Jia, J.
Zhou, Y. H. Li, Y. Liu, S. M. Lu, C. Li, Angew. Chem. Int. Ed.
added. The final DNA and Cu concentrations were 133 mm and
00 mm, respectively. The solution was heated for 5 min at 908C, and
1
slowly cooled to room temperature. The solution was kept at 58C
overnight before use. To the catalyst solution, 3 mL of a fresh stock
solution of Michael acceptor (100 mm) in acetonitrile was added,
followed by the addition of 3.42 mL dimethyl malonate. The reaction
mixture was stirred for 3 days at 58C. Extraction was performed with
2
012, 51, 9352 – 9355; Angew. Chem. 2012, 124, 9486 – 9489;
c) C. H. Wang, Y. H. Li, G. Q. Jia, Y. Liu, S. M. Lu, C. Li, Chem.
Commun. 2012, 48, 6232 – 6234; d) C. H. Wang, G. Q. Jia, Y. H.
Li, S. F. Zhang, C. Li, Chem. Commun. 2013, 49, 11161 – 11163;
e) M. Wilking, U. Hennecke, Org. Biomol. Chem. 2013, 11,
Et O (3 ” 500 mL). After drying the reaction mixture with anhydrous
2
6
940 – 6945; f) Y. H. Li, G. Q. Jia, C. H. Wang, M. P. Cheng, C.
Na SO , the solvent was removed to obtain the crude product. Chiral-
2
4
Li, ChemBioChem 2015, 16, 618 – 624.
phase HPLC was performed directly with this crude product to
determine the conversion and enantiomeric excess.
[
10] a) P. Fournier, R. Fiammengo, A. Jäschke, Angew. Chem. Int.
Ed. 2009, 48, 4426 – 4429; Angew. Chem. 2009, 121, 4490 – 4493;
b) N. S. Oltra, G. Roelfes, Chem. Commun. 2008, 6039 – 6041;
c) S. Park, L. Zheng, S. Kumakiri, S. Sakashita, H. Otomo, K.
Ikehata, H. Sugiyama, ACS Catal. 2014, 4, 4070 – 4073; d) L.
Gjonaj, G. Roelfes, ChemCatChem 2013, 5, 1718 – 1721; e) U.
Jakobsen, K. Rohr, S. Vogel, Nucleosides Nucleotides Nucleic
Acids 2007, 26, 1419 – 1422.
Acknowledgements
This work was supported by the Deutsche Forschungsge-
meinschaft, SFB 623 “Molecular Catalysis”. The authors
thank Dr. R. Wombacher and Dr. M. Sunbul for their
valuable comments on the manuscript. S. Suhm is gratefully
acknowledged for technical assistance.
[
11] Z. Tang, D. P. GonÅalves, M. Wieland, A. Marx, J. S. Hartig,
ChemBioChem 2008, 9, 1061 – 1064.
12] a) J. S. Johnson, D. A. Evans, Acc. Chem. Res. 2000, 33, 325 –
[
3
9
35; b) D. A. Evans, D. Seidel, J. Am. Chem. Soc. 2005, 127,
958 – 9959.
Keywords: asymmetric catalysis · catalyst tuning · DNA ·
G-quadruplexes · Michael addition
[13] D. G. Wei, G. N. Parkinson, A. P. Reszka, S. Neidle, Nucleic
Acids Res. 2012, 40, 4691 – 4700.
[
14] A. T. Phan, V. Kuryavyi, S. Burge, S. Neidle, D. J. Patel, J. Am.
Chem. Soc. 2007, 129, 4386 – 4392.
How to cite: Angew. Chem. Int. Ed. 2015, 54, 11279–11282
Angew. Chem. 2015, 127, 11432–11436
[15] E. Y. N. Lam, D. Beraldi, D. Tannahill, S. Balasubramanian, Nat.
Commun. 2013, 4, 1796.
[
16] Alternatively, these spectral changes might be explained by
a mixed population of parallel quadruplexes and antiparallel
quadruplexes. This hypothesis, however, is difficult to reconcile
with the high and opposing stereoselectivity of the reaction.
[
1] a) A. J. Boersma, R. P. Megens, B. L. Feringa, G. Roelfes, Chem.
Soc. Rev. 2010, 39, 2083 – 2092; b) S. Park, H. Sugiyama, Angew.
Chem. Int. Ed. 2010, 49, 3870 – 3878; Angew. Chem. 2010, 122,
3
7
960 – 3969; c) S. K. Silverman, Angew. Chem. Int. Ed. 2010, 49,
180 – 7201; Angew. Chem. 2010, 122, 7336 – 7359; d) J. Bos, G.
Roelfes, Curr. Opin. Chem. Biol. 2014, 18, 135 – 143.
2] a) G. Roelfes, B. L. Feringa, Angew. Chem. Int. Ed. 2005, 44,
Received: April 27, 2015
Revised: June 15, 2015
[
3
230 – 3232; Angew. Chem. 2005, 117, 3294 – 3296; b) G. Roelfes,
Published online: July 29, 2015
1
1282 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 11279 –11282