A kinetic and structural investigation of DNA-Based asymmetric catalysis using first-generation ligands

Article abstract of DOI:10.1002/chem.200900456

The recently developed concept of DNA-based asymmetric catalysis involves the transfer of chirality from the DNA double helix in reactions using a noncovalently bound catalyst. To date, two generations of DNAbased catalysts have been reported that differ

Full text of DOI:10.1002/chem.200900456

DOI: 10.1002/chem.200900456
A Kinetic and Structural Investigation of DNA-Based Asymmetric Catalysis
Using First-Generation Ligands
Fiora Rosati, Arnold J. Boersma, Jaap E. Klijn, Auke Meetsma, Ben L. Feringa,* and
Gerard Roelfes*[a]
Abstract: The recently developed con-
cept of DNA-based asymmetric cataly-
sis involves the transfer of chirality
from the DNA double helix in reac-
tions using a noncovalently bound cata-
lyst. To date, two generations of DNA-
based catalysts have been reported that
differ in the design of the ligand for
the metal. Herein we present a study
of the first generation of DNA-based
catalysts, which contain ligands com-
prising a metal-binding domain linked
through a spacer to a 9-aminoacridine
moiety. Particular emphasis has been
placed on determining the effect of
DNA on the structure of the CuII com-
plex and the catalyzed Diels–Alder re-
action. The most important findings
are that the role of DNA is limited to
being a chiral scaffold; no rate acceler-
ation was observed in the presence of
DNA. Furthermore, the optimal DNA
sequence for obtaining high enantiose-
lectivities proved to contain alternating
GC nucleotides. Finally, DNA has been
shown to interact with the CuII com-
plex to give a chiral structure. Compar-
ison with the second generation of
DNA-based catalysts, which bear bipyr-
idine-type ligands, revealed marked dif-
ferences, which are believed to be re-
lated to the DNA microenvironment in
which the catalyst resides and where
the reaction takes place.
Keywords: asymmetric catalysis
Diels–Alder reactions DNA
enantioselectivity · ligand effects
·
·
·
Introduction
transfer hydrogenation,[6] hydrolysis,[7] hydrogenation,[8–10]
epoxidation,[11,12] sulfoxidation,[13–16] Diels–Alder reac-
tions,[17,18] transamination,[19] and allylic alkylation.[20]
The high activities and selectivities achieved by enzymes
under mild conditions continue to be a source of inspiration
for the design of new catalysts. A recent approach that has
proven to be quite versatile is the creation of hybrid cata-
lysts in which the catalytic power of transition-metal com-
plexes is combined with the chiral architecture of biopoly-
mers.[1–5] This concept involves placing a metal complex in
the chiral microenvironment provided by the biomolecular
scaffold using covalent, supramolecular, or dative anchoring
strategies. By using a protein scaffold, this method has re-
sulted in a range of enantioselective artificial metalloen-
zymes that are capable of catalyzing transformations such as
Polynucleotides such as RNA and DNA are arguably
some of the most elegant chiral structures in nature and,
hence, they are attractive scaffolds for the assembly of enan-
tioselective hybrid catalysts. Prior to our studies it was dem-
onstrated that the chirality of DNA can be transferred in a
stoichiometric DNA-templated reaction.[21–23] Furthermore,
a DNAzyme capable of enantioselection of an RNA sub-
strate[24] and an enantioselective Diels–Alderase RNAzyme
have been reported.[25]
In our approach to DNA-based enantioselective cataly-
sis,[26–28] we have demonstrated that the chirality of DNA
can be transferred directly to a metal-catalyzed reaction.
The design involves the modular assembly of a DNA-based
catalyst from natural duplex DNA, usually salmon testes
DNA (st-DNA), and a copper complex of an achiral ligand
that can bind DNA in a noncovalent fashion.[26] As a result,
the active CuII center is brought into the proximity of the
chiral environment of the DNA double helix, allowing the
transfer of chirality and the generation of reaction products
with an excess in one of their enantiomers (Scheme 1).
To date, two generations of DNA-based catalysts have
been developed that differ in the type of ligand used. In our
[a] F. Rosati, A. J. Boersma, Dr. J. E. Klijn, A. Meetsma,
Prof. B. L. Feringa, Dr. G. Roelfes
Stratingh Institute for Chemistry, University of Groningen
Nijenborgh 4, 9747 AG Groningen (The Netherlands)
Fax : (+31)50-363-4296
E-mail: j.g.roelfes@rug.nl
b.l.feringa@rug.nl
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900456.
9596
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9596 – 9605

FULL PAPER
that DNA is more than a chiral scaffold; rate accelerations
of up to two orders of magnitude were observed in the pres-
ence of DNA. Moreover, both the rate acceleration and the
enantioselectivity were found to be DNA-sequence depen-
dent.[29]
The scope of DNA-based catalysis has been extended
beyond the Diels–Alder reaction: good-to-excellent enantio-
selectivities have been reported in the Michael addition re-
actions of malonates and nitromethane to yield a,b-unsatu-
rated 2-acylimidazoles,[30] Friedel–Crafts alkylations,[31] fluo-
rination reactions,[32a] and epoxide hydrolysis.[32b] Compared
with the first generation of catalysts, however, the second
generation of catalysts are less flexible with regard to which
enantiomer of the product is obtained; the product resulting
from attack of the diene or nucleophile on the Si face of the
enone moiety is always obtained.
Recently, a variety of other approaches towards DNA-
based catalysis have been reported, which feature a cova-
lently attached catalytic moiety.[33]
Herein we report the results of a study aimed at obtaining
a better understanding of the mechanistic and structural as-
pects of the first generation of DNA-based catalysts, which
comprise an acridine-based ligand. In particular the effects
DNA has on the observed catalysis and on the structure of
the catalyst will be discussed.
Results and Discussion
Kinetics: The effect of DNA on the reaction rate was inves-
tigated by studying the reaction kinetics. The mechanism by
which a Lewis acid, that is, CuII, can be expected to affect
the rate of the Diels–Alder reaction of azachalcone (1) with
cyclopentadiene (2) in water is depicted in Scheme 2.[34a]
The first step in the cycle comprises a rapid and reversible
coordination of the dienophile to the CuII ion, which results
in the activation of the dienophile. This is followed by the ir-
reversible Diels–Alder reaction. Finally, the product dissoci-
ates from the Lewis acid, regenerating the catalyst for an-
other cycle of the reaction.
The overall rate of the reaction is determined by the equi-
librium constant for the binding of the azachalcone (1) to
CuII (Ka), the rate of the reaction of cyclopentadiene (2)
with the CuII-bound azachalcone (kcat), and the dissociation
constant for the release of the product from the copper
complex (Kd). In the kinetic measurements Kd did not influ-
ence the overall rate because a large excess of catalyst was
used. Consequently, product inhibition can be excluded.
The apparent second-order rate constant (kapp) was deter-
mined from the initial decrease in the absorption of azachal-
cone (1) at various copper concentrations. From this, Ka and
kcat were determined by using the pseudophase model, as re-
ported by Engberts and co-workers,[34] using an iterative fit-
ting approach. For practical reasons, the experiments were
performed at a pH of 5.5 (MES buffer), which is different
from the pH used in the catalytic experiments, that is,
pH 6.5 (MOPS buffer). The use of a lower pH value result-
Scheme 1. a) Schematic representation of the concept of DNA-based
asymmetric Diels–Alder reaction; b) first (L1–L6) and second (dmbpy)
generation of ligands used in DNA-based asymmetric catalysis.
first design, a metal-binding domain, that is, aminomethyl-
pyridine, was linked through a short spacer to a DNA inter-
calator, that is, 9-aminoacridine. With this design enantio-
meric excess (ee) values of up to 50% were obtained in
CuII-catalyzed Diels–Alder reactions in water.[26] The design
of the ligand proved to have an important influence on the
results of the catalysis. It was found that the ligand should
preferably contain an arylmethyl group, that is, 3,5-dime-
thoxybenzyl or 1-naphthylmethyl, which suggests that p–p
stacking interactions with the substrate are important for
the observed enantioselective catalysis. Moreover, both en-
antiomers of the product were accessible by appropriate
design of the ligand; the enantiomeric preference of the re-
action was demonstrated to depend both on the nature of
the arylmethyl substituent and the length of the spacer.
In an alternative design, the DNA-binding moiety was in-
tegrated into the metal-binding domain; hence, a spacer was
no longer required. With this second generation of ligands, a
dramatic increase in enantioselectivity was observed; up to
99% ee was obtained with 4,4’-dimethyl-2,2’-bipyridine
(dmbpy) as the ligand.[27,28] In these reactions it was found
Chem. Eur. J. 2009, 15, 9596 – 9605
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9597

B. L. Feringa, G. Roelfes et al.
cant. Notably, in the case of
ligand L1 no significant de-
crease in kcat was observed
(Table 1, entries 1 and 2).
Attempts were made to de-
termine the Ka values inde-
pendently by a UV/Vis titration
experiment[29] to validate the
values derived from the model
used. However, these measure-
ments were complicated by the
strong absorption of the acri-
dine chromophore in the wave-
length region of interest. There-
fore, Ka was determined for the
CuII complex of a similar ligand
(ligand L5, Scheme 1), which
contains the same metal-bind-
ing domain but lacks the acri-
dine moiety. By using this
copper complex a Ka value of
(2.97ꢀ0.90)ꢁ104 mꢁ1 was found
for the binding of azachalcone
(1). This is similar to the values
observed with the copper com-
plexes of ligands L1–L4, which
supports the accuracy of the
values derived from the itera-
tive fitting approach.
Scheme 2. Proposed catalytic cycle for the copper-catalyzed Diels–Alder reaction of 1 with 2.
ed in only a slight decrease in the enantioselectivity of the
catalytic experiments.[35]
A comparison of the values of kapp obtained with ligands
L1–L4 showed that the reactions were roughly equally fast,
The results of the kinetic studies are in sharp contrast to
those obtained with the second generation of catalysts: with
Cu–dmpby, the presence of DNA resulted in a rate accelera-
tion with an acceleration factor of 58.[29] This suggests that
the DNA-based catalysts, depending on the type of ligand,
interact differently with the DNA, such that DNA has a dis-
tinct effect on the reaction rate. Most likely the reaction
takes place in a different structural environment: In the case
of the second-generation catalysts the reaction has been pro-
posed to take place in the groove, which, for example, can
provide favorable interactions with the activated complex.[29]
In the case of the first-generation catalysts, as a result of the
design, the reaction is proposed to occur further away from
the DNA groove. This is consistent with the observed de-
crease in ee when using ligands with a longer spacer.[26]
Therefore, the reaction is more comparable to that taking
place in bulk solution, that is, in the absence of DNA.
Hence, only minor effects on the reactivity are observed.
Thus, in the present case, DNA can be considered to act ex-
clusively as the chiral scaffold.
Table 1. Dependence of kinetic data on DNA in the Cu–L catalyzed
Diels–Alder reaction of 1 with 2 in water.[a]
[b]
Entry
Ligand
kapp [10ꢁ2 mꢁ1 sꢁ1
]
Ka [mꢁ1
]
kcat [mꢁ1 sꢁ1
]
1
2
3
4
5
6
7
8
L1
3.1ꢀ0.92
2.6ꢀ0.49
3.50ꢀ0.008
1.5ꢀ0.49
4.0ꢀ0.36
1.50ꢀ0.009
3.7ꢀ0.41
3.4ꢀ0.20
1.8ꢀ0.74ꢁ104
8.2ꢀ0.97ꢁ103
1.0ꢀ0.40ꢁ104
1.2ꢀ0.45ꢁ104
2.3ꢀ1.40ꢁ104
2.5ꢀ0.97ꢁ104
1.9ꢀ0.92ꢁ104
1.2ꢀ0.58ꢁ104
0.111ꢀ0.027
0.100ꢀ0.008
0.218ꢀ0.066
0.057ꢀ0.016
0.113ꢀ0.034
0.046ꢀ0.008
0.124ꢀ0.032
0.064ꢀ0.023
L1[c]
L2
L2[c]
L3
L3[c]
L4
L4[c]
[a] [Cu2+–L]=0.10–0.30 mm, 25.08C, MES buffer (20 mm, pH 5.5); [1]=
6.0ꢁ10ꢁ3 mm, [2]=0.5–2.0 mm. [b] [Cu2+]=0.058 mm. [c] In the presence
of st-DNA.
regardless of the ligand (Table 1). In the presence of DNA a
slight decrease in the reaction rate was observed in all cases.
A more detailed analysis revealed that the Ka values did
not change significantly with the different ligands used.
Hence, it can be concluded that the binding of the Lewis
acid to the dienophile occurs with equal efficiency in the ab-
sence and presence of DNA. The lower overall reaction
rates proved to be due to a lowering of the kcat with ligands
L2–L4. The observed differences are not large, but signifi-
DNA-sequence dependence: The structure of the second co-
ordination sphere provided by DNA has been demonstrated
to have a strong effect on the enantioselectivity of second-
generation DNA-based catalytic Diels–Alder[29] and Frie-
del–Crafts reactions.[31] st-DNA is a heterogeneous polymer
and its sequence can be considered to be random. Loadings
of one complex per six base pairs were used in the catalytic
9598
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9596 – 9605

DNA-Based Asymmetric Catalysis
FULL PAPER
experiments. Therefore, the DNA-based catalyst is a mix-
ture of copper–ligand complexes located in different chiral
microenvironments. Consequently, the observed enantiose-
lectivity most likely is the (weighted) average of all the con-
tributing DNA sequences.
further truncation to an octamer led to a significantly lower
ee (Table 2, entry 7). This is possibly the result of the in-
creased chance that the copper complex is located near one
of the termini, where the influence of the chirality of the
DNA is smaller.
A series of different DNAs were tested to establish the
effect of the chiral microenvironment provided by the DNA
on the Cu–L1-catalyzed Diels–Alder reaction. DNA from
an alternative natural source, that is, calf thymus DNA (ct-
DNA), resulted in an ee comparable to that found with st-
DNA (Table 2, entry 2). Next, a series of synthetic self-com-
It was interesting to note that the results obtained with
the d(GC)6 sequence are dependent on the presence of salt;
in the absence of NaCl a significantly lower ee was found
(Table 2, entries 7 and 8). Surprisingly, the enantioselectivity
could be restored by interchanging the order of the G and C
nucleobases; with d(CG)6 an ee of 60% was obtained. The
origins of these effects are, at present, not well understood,
but are most likely related to small differences in the struc-
ture of the DNA.
Table 2. Dependence of enantioselectivity on the DNA sequence using
ligand L1.[a]
The high enantioselectivities obtained with alternating
GC sequences can tentatively be explained by the acridine
moieties having an increased affinity for GC-rich re-
gions.[36–38] Alternatively, the coordination sphere provided
by GC-rich regions may provide the most favorable environ-
ment for the catalyzed reaction.
Entry
DNA sequence
Conversion [%]
ee[b] [%]
1
2
3
4
5
6
7
8
st-DNA
ct-DNA
d(ATATATATATAT)2
100
100
100
100
100
100
100
100
100
90
100
100
97
97
100
84
37
35
9
6
6
d
A
Poly
Poly
G
N
G
62
27
54
35
60
8
35
26
25
10
10
16
19
8
[c]
dACHTUNGTRENNUNG
[c]
Substrates: An investigation of the substrate scope previous-
ly revealed that dienophiles able to chelate copper in a bi-
dentate fashion is a general requirement for both catalysis
and enantioselectivity.[29] The general design is amenable to
variations in the metal-binding domain, however, an N,O
donor set is required.
For example, the pyridine moiety can be replaced by an
N-methylimidazole moiety.[29] The role of the pyridine in the
interaction with DNA was probed by introducing a methyl
substituent onto the pyridyl ring and systematically varying
its position. It was found that, depending on the ligand used,
this affected the conversion significantly (Table 3). However,
no clear pattern could be discerned. The endo isomer of the
Diels–Alder product was always favored, albeit with ligand
L1 a significantly lower endo/exo ratio was always observed
(Table 3, entries 9–12). The magnitude of the ee did not
change to a major extent for 4–6 compared with the unsub-
stituted azachalcone (Table 3).
9
d(GCGCGCGCGCGC)2
d(CGCGCGCGCGCG)2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
d
N
d(CGCGGCGCCGCG)2
d(CGCGCCGGCGCC)2
d(CGCGGGCCCGCG)2
d(TCAGGGCCCTGA)2
d(TCGCGATCGCGA)2
d(TCGGGTACCCGA)2
d(TCGCGTACGCGA)2
d(CGCGATATCGCG)2
d(CGCGTATACGCG)2
d(GGGCAATTGCCC)2
d(CGCGAATTCGCG)2
d(CGCAAATTTGCG)2
d(GCAAAATTTTGC)2
d(GCATGCATCGTACGTA)2
d(GACTGACTAGTCAGTC)
84
100
100
100
100
92
100
100
100
100
28
3
15
16
4
9
34
[a] Conditions: see the Experimental Section. [b] ee values refer to the
(+) enantiomer in all cases and are the average of two experiments. They
are reproducible to within 2%. [c] 100 mm NaCl.
Which enantiomer is formed in excess depends on the
ligand used. Interestingly, ligand L4 always gives the oppo-
site enantiomer of the Diels–Alder product compared with
ligands L1–L3, except with the substrate 6, which has the
methyl substituent at the 3-position of the pyridyl ring. This
particular behavior can be explained tentatively on the basis
that the azachalcone substrate can adopt either a cisoid or
transoid conformation. In the case of 6, the dienophile is
forced to react in the cisoid conformation because the trans-
oid form is unfavored for steric reasons (Scheme 3).[39] Most
likely this is the reason why with 6 the same enantiomer is
obtained in excess with all ligands.
However, with substrates 4 and 5 this limitation does not
exist and we propose that with the ligand L4 a different sub-
strate-bound complex is formed preferentially,[39] that is,
with the dienophile in the transoid conformation
(Scheme 3). Assuming that the conformation of the Cu-
bound dienophile is an important determinant for the enan-
plementary oligonucleotides of defined sequence were
tested. The sequences were selected to cover a broad range
of the available sequence space. From the data acquired
(Table 2), a general pattern can be discerned. First, sequen-
ces that are rich in AT base pairs generally give rise to very
poor enantioselectivities (Table 2, entries 3–5). Furthermore,
in sharp contrast to the results obtained with Cu–dmbpy,[29]
sequences containing stretches of guanines give ees that are
significantly lower than those obtained with st-DNA
(Table 2, entries 14–15, 17, and 21). The best results were
obtained with sequences rich in GC base pairs, in particular
with alternating GC sequences (Table 2, entries 6–10); up to
62% ee was obtained with poly(GC) (Table 2, entry 6),
which is significantly higher than the value obtained with st-
DNA. A GC dodecamer gave similar ee values, however,
Chem. Eur. J. 2009, 15, 9596 – 9605
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9599

B. L. Feringa, G. Roelfes et al.
Table 3. Substrate variations in the Diels–Alder reaction with cyclopenta-
diene.[a]
Entry Ligand
Dienophile
Conversion[b] endo/ endo
[%]
exo
ee
[%][c]
1
L1
55
88:12 ꢁ46
4
4
4
4
2
3
4
L2
L3
L4
75
100
>80
99:1 ꢁ40
97:3 ꢁ59
96:4
45
5
L1
85
90:10 ꢁ34
5
5
5
5
6
7
8
L2
L3
L4
72
46
50
98:2 ꢁ40
99:1 ꢁ46
99:1
45
9
L1
31
88:12 ꢁ33
Figure 1. ORTEP diagram of the Cu–L1 complexes 1 and 2 in the unit
ꢁ
ꢁ
cell. Bond lengths: Cu Neq: 1.96–2.06 ꢂ; Cu Nax: 2.24 ꢂ.
6
6
6
6
10
11
12
L2
L3
L4
68
29
100
78:22 ꢁ54
73:27 ꢁ53
95:5 ꢁ54
plexes, six perchlorate anions, and three acetonitrile solvent
molecules.
[a] Conditions: all experiments were carried out with the following re-
agents: [Cu(L)
ACHTUNGTRENNUNG
The asymmetric unit contains two chemically distinct
mononuclear copper complexes. The structure of cation
complex 1 (Figure 1, top) shows a mononuclear CuII center
in a square pyramidal geometry. The ligand is bound in the
equatorial plane and coordinates to the cation through the
two nitrogen atoms of the aminomethylpyridine moiety. The
[2]=15 mm in MOPS buffer (20 mm, pH 6.5) for 3 days at 58C. [b] Deter-
1
mined by H NMR spectroscopy. [c] +/ꢁ refers to the order of elution of
the two enantiomers, first and second, respectively (see the Experimental
Section). The endo/exo ratio, ee, and conversion data are reproducible to
within 3%.
ꢁ
Cu N bond lengths range from 1.96 to 2.06 ꢂ, which is char-
acteristic for CuII complexes.[40,41] The remaining three coor-
dination sites are occupied by acetonitrile molecules, which
was the solvent for crystallization. The structure of the
cation complex 2 (Figure 1, bottom) is very similar, except
for the fact that the three remaining coordination sites are
occupied by two acetonitrile molecules and one water mole-
cule. The latter is bound in the equatorial plane trans to the
tertiary amine of the ligand. It is at these equatorial sites
that the substrate is proposed to bind through the pyridyl ni-
trogen and carbonyl oxygen, thereby displacing the solvent
molecules. An interesting aspect of this structure is that the
acridine moiety of complex 1 lies in a parallel orientation
above the dimethoxybenzyl moiety of complex 2, which in
turn lies above the pyridyl moiety of complex 2. The dis-
tance between the planes of the aromatic rings is approxi-
mately 3.5 ꢂ, which is characteristic of p–p stacking interac-
tions.[42,43]
Scheme 3. Possible reactive conformations of the unsubstituted dieno-
phile: a) cisoid, b) transoid.
tiofacial selectivity of the reaction, this results in the forma-
tion of the opposite enantiomer of the product formed with
ligands L1–L3 in which the substrate is most likely bound to
the CuII center in the cisoid conformation.
Crystal structure: The complexation of L1 with [Cu-
ACHTUNGTRENNUNG(ClO4)2·6H2O] in CH3CN in the presence of HClO4
(1 equiv) resulted in the formation of dark-green crystals, of
which the crystal structure was determined. The identifica-
tion of the atoms and the configuration of the cation moiet-
ies as well as the packing of the molecules in the cell are
shown in Figure 1. Each asymmetric unit contains 11 moiet-
ies consisting of two cations of mononuclear copper com-
The p–p stacking interactions are considered to be impor-
tant for achieving enantioselectivity in the aqueous Diels–
Alder reaction of azachalcone with cyclopentadiene.[26] The
present structure most likely does not reflect the structure
of the complex in aqueous solution, but it does demonstrate
the propensity of the ligand for p–p stacking interactions.
9600
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9596 – 9605

DNA-Based Asymmetric Catalysis
FULL PAPER
Circular dichroism (CD) spectroscopy: Because of the dif-
ferent behavior of the copper complexes of ligands L1 and
L4 in catalysis, a CD study of the DNA–copper–ligand com-
plexes was performed to determine whether this difference
could be assigned to a different interaction with the DNA.
9-Aminoacridine is achiral, but upon interaction with
DNA, which provides a chiral environment, an induced CD
is observed.[44] Signals from DNA are found in the region
below 300 nm. However, in the presence of 9-aminoacridine,
the CD spectrum shows characteristic absorption bands
around 340 and 400 nm (Figure 2a).[44,45]
By using a ligand with a shorter spacer, that is, ligands L1
and L3, a different pattern was observed. An increase in the
intensity of the absorption at around 400 nm was observed
as well as an exciton coupling at 330 nm, which suggests a
chiral interaction between the acridine moiety and one of
the two possible chromophores on the ligand: The pyridyl
ring or the aromatic substituent of the ligand, that is, the 1-
naphthylmethyl or 3,5-dimethoxybenzyl moiety. For the li-
gands with a longer spacer, such as ligands L2 and L4, these
chromophores are apparently too far from the DNA cavity
to have a defined orientation with respect to the acridine.
Then, to identify which chromophore is involved in this
interaction, a new design of ligand was employed in which
the arylmethyl moiety on the nitrogen atom was replaced
with an ethyl group (L6).[35] By using the copper complex of
this ligand with DNA, the exciton coupling was still ob-
served (Figure 2b). From this it can be concluded that the
aromatic side-chain in ligands L2 and L4 are not involved in
the interaction that gives rise to the exciton coupling. Most
likely the pyridyl chromophore interacts with the acridine,
possibly as a result of partial intercalation, which suggests
that it is located closer to the DNA groove. Importantly, the
behavior observed by CD spectroscopy with ligands L1–L4
does not correlate with the enantiomeric outcome of the re-
action; apparently the chiral structure of the complex is only
present in the resting state of the catalyst.
Comparison with the second generation of DNA-based cata-
lysts: Comparison between the two generations of DNA-
based catalysts reveals marked differences, both with regard
to enantioselectivity and the role DNA plays in the catalytic
reactions.
1) Compared with the second-generation catalysts, which
give rise to up to 99% ee in the Diels–Alder reaction,
the ee values obtained with the first generation of DNA-
based catalysts are modest.
2) Both enantiomers of the Diels–Alder product are acces-
sible by using the first generation of catalysts, whereas
with the complexes formed between DNA and the bipyr-
idine (bpy)-based ligands, the enantiomer resulting from
the approach of the cyclopentadiene at the Si face of the
enone moiety is always favored.
3) DNA plays a different role in the catalysis depending on
the ligand used. In combination with the copper complex
of the 4,4’-dimethylbipyridine ligand, the presence of
DNA results in a significant rate acceleration of up to
two orders of magnitude. In contrast, with the first gener-
ation of ligands, there is a slight decrease in the reaction
rate in the presence of DNA.
Figure 2. a) CD spectra of copper–DNA–Lx: L1 (g), L2 (d), L3
(c), and L4 (a) complexes; b) comparison of the complexes of li-
gands L1 (c), L3 (a), and L6 (g).
4) The first and second generation of catalysts also have dif-
ferent requirements regarding the DNA sequence and,
hence, structure, with the acridine-based catalysts the
best results are obtained with alternating GC sequences,
whereas with the bpy-based catalysts the presence of
guanine tracts is especially beneficial.
For the ligands with a propyl spacer, that is, L2 and L4
(Scheme 1), a spectrum similar to 9-aminoacridine was ob-
served with characteristic absorption bands around 330 and
400 nm (Figure 2a), which correspond to two (p–p*) elec-
tronic transitions polarized along the short and long axes of
the acridine molecule, respectively.[45]
Chem. Eur. J. 2009, 15, 9596 – 9605
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9601

B. L. Feringa, G. Roelfes et al.
These differences can be rationalized by considering that
the catalyzed reaction can take place in different microen-
vironments depending on the type of ligand used. The gen-
eral design of the catalysts comprises a metal-binding
domain and a DNA-binding domain to anchor the catalyst
to the DNA. In the second generation of catalysts, which
comprises planar, symmetric polyaromatic molecules, the
DNA binding moiety is integrated into the metal binding
site. Therefore, the catalytic metal center is most likely very
close to the DNA helix and the distance from the DNA de-
creases on going from larger, for example, dipyrido[3,2-
a:2’,3’-c]phenazine (dppz), to smaller polyaromatic moieties
such as bpy and dmbpy. It has been proposed that with the
catalysts based on the bpy-type ligands, the Diels–Alder re-
action takes place in the DNA groove, which provides a
chiral environment that is potentially structurally compatible
with the transition state of the reaction.[29] The resulting sta-
bilization of the transition state would explain the observed
58-fold rate enhancement and the high enantioselectivity
that was found with Cu–dmpby. This requires a specific
structure of the DNA groove, as demonstrated by the strong
DNA-sequence dependence of the enantioselectivity of the
reaction, which was shown to be positively affected by se-
quences containing guanine tracts, which give rise to B-
DNA, but with A-DNA characteristics.
bination of the structure of the ligand, which is supported
by the fact that both enantiomers of the product can be ac-
cessed depending on the design of the ligand, and the inter-
action of the DNA with the substrate-bound complex.
Conclusions
A detailed study of the first generation of DNA-based cata-
lysts in copper-catalyzed Diels–Alder reactions in water was
performed with a particular focus on establishing the role of
DNA in catalyzed enantioselective Diels–Alder reactions. A
comparison with the second generation of DNA-based cata-
lysts, which involves bpy-type ligands, revealed marked dif-
ferences, which are most likely the result of different micro-
environments for the catalyzed reactions. These findings un-
derline the fact that the design of the ligand is an important
aspect in the development of hybrid catalysts: Depending
on the ligand used, the catalyzed reactions can take place in
very different microenvironments, which results in different
reaction rates and enantioselectivities.
Experimental Section
In the first generation of ligands, the DNA binding
moiety, that is, 9-aminoacridine, is linked to a metal-binding
domain through a spacer. Hence, the catalyzed reaction is in
this case expected to take place at the edge of the DNA
groove and, therefore, will more resemble the reaction as it
occurs in solution in the absence of DNA. As a result only
small differences in reaction rate are observed between the
reactions with and without DNA. Furthermore, the enantio-
selectivity does not originate from the structural compatibil-
ity of the DNA groove with one enantiomer of the transi-
tion state, but most likely by the shielding of one face of the
enone by the arylmethyl side-chain of the ligand, as has also
been suggested for the enantioselective Diels–Alder reac-
tion catalyzed by copper–amino acid complexes.[34] The in-
teraction of the ligand with the substrate has been proposed
to occur by p–p stacking interactions. This hypothesis im-
plies that, in contrast to the bpy-based catalysts, with the
first generation of DNA-based catalysts the chirality is not
transferred directly from the DNA to the catalyzed reaction.
Instead, the chirality is transferred in two steps; the DNA
forces a chiral conformation on the bound copper complex,
which is in turn translated into enantioselective bond forma-
tion in the catalyzed Diels–Alder reaction. Indeed, the in-
duced CD effects, in particular the observed exciton cou-
pling, demonstrate that the DNA can enforce a chiral struc-
ture on the copper complex, even though the observed in-
duced CD effect proved to be unrelated to the enantiomeric
preference of the reaction. Based on the results obtained by
varying the substrates, we hypothesize that an important
aspect in determining which enantiomer is formed in excess
is whether the substrate binds in the cisoid or transoid con-
formation. This is in turn most likely determined by a com-
General: st-DNA was obtained from Sigma, pAT and pGC were obtained
from Amersham. Synthetic oligonucleotides were obtained from BioTez
(Germany). Cyclopentadiene was freshly prepared from its dimer prior
to use. Azachalcone dienophiles and acridine-based ligands (L1–L4)
were prepared by following published procedures.[35] For kinetic studies
st-DNA was dialyzed extensively against MES buffer (20 mm, pH 5.5)
prior to use. 1H and 13C NMR spectra were recorded on a Varian 400
(400 MHz) spectrometer in CDCl3. Chemical shifts (d) are given in ppm
using residual solvent as the internal standard (dC =77.0 ppm and dH =
7.26 ppm for CDCl3).
Physical methods—General: Circular Dichroism spectra were measured
on a JASCO J-715 spectropolarimeter equipped with a temperature con-
trol attachment. The UV/Vis spectra were measured on JASCO V-560
and JASCO V-570 spectrophotometers. The ees were determined by
HPLC analysis performed on a Shimadzu 10AD-VP system equipped
with a Daicel ODH column (n-heptane/iPrOH, 98:2, 0.5 mLminꢁ1) or a
Daicel OD column (n-heptane/iPrOH, 98:2, 1 mLminꢁ1).
Representative procedure for the [Cu(L)ACTHNUTRGNEUNG(NO3)2]/DNA-catalyzed Diels–
Alder cycloaddition reaction:
A solution of st-DNA (10 mL of a
2 mgmLꢁ1 solution of st-DNA dissolved in 30 mm MOPS buffer, pH 6.5,
and prepared 24 h in advance) was added to a catalyst solution of
[Cu(L)ACHTNUGTRNEUGN(NO3)2] (0.9 mm of Cu and 1.17 mm of ligand in a total volume of
5 mL) to a final concentration of 1.3 mgmLꢁ1. An aliquot of a stock solu-
tion of dienophile azachalcone in CH3CN (0.5m, 30 mL) was added to
obtain a final concentration of 1 mm and the mixture was subsequently
cooled to 58C. The reaction was started by the addition of cyclopenta-
diene (20 mL, 0.016m final concentration) and allowed to continue for 3 d
at 58C, while mixing continuously. The product was isolated by extraction
with Et2O. After drying (Na2SO4) and removal of the solvent, the crude
1
product was analyzed by H NMR spectroscopy and HPLC.
Dissolution of synthetic oligonucleotides: The synthetic oligonucleotides
were obtained from BioTez. The lyophilized powders were dissolved in
MOPS buffer (20 mm, pH 6.5) and the solution was heated to 948C,
slowly cooled to 58C, and left for 2 h at 58C prior to use. The concentra-
tion was determined by UV/Vis spectrophotometry at 258C.
Reactions in the presence of synthetic oligonucleotides: A 2 mL Eppen-
dorf container was loaded with oligomer solution (400 mL, 2 mgmLꢁ1) in
buffer (20 mm MOPS, pH 6.5), [CuL1ACHTNUGTRNEUNG(NO3)2] complex (200 mL, 0.3 mm),
9602
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9596 – 9605

DNA-Based Asymmetric Catalysis
FULL PAPER
kobs
½Cpꢂtot
and a stock solution of azachalcone (1; 2.4 mL, 3.1 mg/60 mL CH3CN,
0.024 mm) at 58C. Freshly distilled cyclopentadiene (2; 0.8 mL, 0.016m
final concentration) was added to this mixture through a microsyringe.
The reaction mixture was mixed by continuous inversion for 3 d at 58C.
After extraction with Et2O, drying with Na2SO4, and concentration under
reduced pressure, the products were analyzed by HPLC.
kobs;app
¼
ð6Þ
The apparent rate constant (kapp) can then be calculated after correcting
for incomplete binding of the dienophile to the catalyst [Eq. (7)].
Determination of kapp: The procedure to determine kapp, Ka, and kcat were
adapted from the procedure of Engberts and co-workers.[34] A fresh solu-
tion of azachalcone (1; 2.0 mL, 0.95 mm in CH3CN) was added to the ap-
propriate catalyst in buffer (20 mm MES, pH 5.5) in a quartz cuvette.
After the absorption had stabilized, a freshly prepared cyclopentadiene
solution in CH3CN (1–10 mL, 0.5m) was added, with a final concentration
0.5–2.0 mm in the reaction mixture. The cuvette was closed immediately
to prevent evaporation of cyclopentadiene and sealed. The reaction was
monitored at the appropriate temperature at 326 nm on a JASCO V-560
or JASCO V-570 spectrophotometer. The decrease in absorption of 1 was
followed for the first 15% of the reaction, and Equation (1) was used to
calculate kapp, in which e1 and e3 are the extinction coefficients of dieno-
phile 1 and product 3, respectively, and d is the path length of the cuv-
ette. The observed rate constants were determined at different concentra-
tions of cyclopentadiene 2 and the value of kapp was extracted from the
slope of the resulting plot.
kobs;app
Xb
ð7Þ
kapp
¼
Because the decrease in the dienophile concentration as a function of
time is given by Equation (8) and because the conditions are such that
the uncatalyzed reaction is negligible, Equations (5) and (8) can be com-
bined to give Equation (9).
d½azaꢂtot
ð8Þ
ð9Þ
¼ kapp½azaꢂtot½Cpꢂtot
dt
nþ
Ka½M
ꢂ
tot
nþ
kapp ¼ kcat
1 þ Ka½M
ꢂ
tot
Equation (9) can be rewritten as Equation (10) so that a straight line is
obtained by plotting 1/kapp versus 1/
[Mn+]. From the values of the slope
and the intercept, values of Ka and kcat can be obtained. Because kapp can
only be calculated if Ka is known the data has to be fitted iteratively until
the value of Ka inserted into Equation (4) converges to the value ob-
tained from the ratio of the intercept to the slope.
dA1
dt
1
kapp
¼
ð1Þ
AHCTUNGTRENNUNG
dðe1 ꢁ e3Þ½1ꢂ0½2ꢂ0
Determination of Ka and kcat: The rate of the reaction of the dienophile
with the diene can be described by the sum of the rate of the catalyzed
and uncatalyzed reactions [Eq. (2)].
1
1
1
nþ
1
kcat
¼
þ
ð10Þ
kapp kcatKa
½M
ꢂ
tot
d½azaꢂ
ð2Þ
¼ kuncat½Cpꢂtot½azaꢂf þ kcat½Cpꢂtot½azaꢂb
dt
Determination Ka:[34] The Ka values were determined also independently
by titration experiments using ligand L5 which contains the same binding
site as the other ligands tested, but depleted of the acridine moiety. The
catalyst solution without DNA was prepared as previously described in
the literature.
In this equation kuncat and kcat are the uncatalyzed and catalyzed rate con-
stants, respectively. The dienophile is either bound to the catalyst (Lewis
acid) or free in solution, whereas the diene is distributed homogeneously
through the solution. The concentrations of diene and dienophile are
given by [Cp]tot, [aza]tot, [aza]f, and [aza]b, which represent the total con-
centration of the diene, the total concentration of the dienophile, the
concentration of the free dienophile, and the concentration of the bound
dienophile, respectively. The total concentration of the dienophile is
given by the sum of the free and bound concentrations. Considering that
at most 10% of the catalyst Mn+ is bound to the dienophile (see below)
instead of using the concentration of the free catalyst ([Mn+]f), the total
Crystal structure determination: Data were collected on
a Bruker
SMART APEX CCD diffractometer (Platform with full three-circle goni-
ometer). The diffractometer was equipped with a 4K CCD detector set
60.0 mm from the crystal. The crystal was cooled to 100 K using the
Bruker KRYOFLEX low-temperature device. Intensity measurements
were performed using graphite monochromated MoKa radiation from a
sealed ceramic diffraction tube (Siemens). Generator settings were
50 kV/40 mA. SMART was used for preliminary determination of the
unit cell constants and data collection control. Data integration and
global cell refinement was performed with the program SAINT. Intensity
data were corrected for Lorentz and polarization effects, scale variation,
for decay and absorption: a multi-scan absorption correction was applied,
based on the intensities of symmetry-related reflections measured at dif-
concentration of the catalyst ([Mn+
]tot) can be used to calculate the bind-
ing constant (Ka) of the dienophile to the catalyst [Eq. (3)]. From this,
the mole fraction of bound dienophile (Xb) can be calculated [Eq. (4)].
½azaꢂb
2
Ka
Xb
¼
¼
ð3Þ
ð4Þ
ferent angular settings (SADABS), and reduced to Fo . The program
nþ
½azaꢂf½M
ꢂ
tot
suite SHELXTL was used for space group determination (XPREP). The
structure was solved by Patterson methods and extension of the model
was accomplished by direct methods applied to difference structure fac-
tors using the program DIRDIF. Details of the data collection and refine-
ment are given in Table 4.
nþ
Ka½M
1 þ Ka½M
ꢂ
tot
nþ
ꢂ
tot
CCDC-719143 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Substitution of Equation (3) into Equation (2) leads to Equation (5).
nþ
d½azaꢂtot
dt
Ka½M
ꢂ
tot
nþ
ð5Þ
¼ kuncat½Cpꢂtot½azaꢂf þ kcat½azaꢂf þ kcat½Cpꢂtot
1 þ Ka½M
ꢂ
tot
Under pseudo first-order conditions (excess diene) the observed apparent
rate constant (kobs,app) can be obtained from the observed rate constant
(kobs) and the concentration of the diene [Eq. (6)].
Chem. Eur. J. 2009, 15, 9596 – 9605
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9603

B. L. Feringa, G. Roelfes et al.
Table 4. Crystal data and details of the structure determination.
[9] R. R. Davies, M. D. Distefano, J. Am. Chem. Soc. 1997, 119, 11643–
11652.
[10] a) J. Collot, J. Gradinaru, N. Humbert, M. Skander, A. Zocchi, T. R.
Ward, J. Am. Chem. Soc. 2003, 125, 9030–9031.
[11] M. E. Wilson G. M. Whitesides, J. Am. Chem. Soc. 1978, 100, 306–
307.
[12] K. Okrasa R. J. Kazlauskas, Chem. Eur. J. 2006, 12, 1587–1596.
[13] A. Fernandez-Gacio, A. Codina, J. Fastrez, O. Riant, P. Soumillion,
ChemBioChem 2006, 7, 1013–1016.
[14] T. Sugimoto, T. Kokubo, J. Miyazaki, S. Tanimoto, M. Okano, J.
Chem. Soc. Chem. Commun. 1979, 402–404.
[15] A. Mahammed, Z. Gross, J. Am. Chem. Soc. 2005, 127, 2883–2887.
[16] J. R. Carey, S. K. Ma, T. D. Pfister, D. K. Garner, H. K. Kim, J. A.
Abramite, Z. L. Wang, Z. J. Guo, Y. Lu, J. Am. Chem. Soc. 2004,
126, 10812–10813.
[17] F. van de Velde, L. Kçnemann, F. van Rantwijk, R. A. Sheldon,
Chem. Commun. 1998, 1891–1892.
[18] M. T. Reetz, N. Jiao, Angew. Chem. 2006, 118, 2476–2479; Angew.
Chem. Int. Ed. 2006, 45, 2416–2419.
[19] S. Colonna, A. Manfredi, R. Annunziata, Tetrahedron Lett. 1998, 39,
3347–3350.
[20] R. S. Roy, B. Imperiali, Protein Eng. 1997, 10, 691–698.
[21] J. Pierron, C. Malan, M. Creus, J. Gradinaru, I. Hafner, A. Ivanova,
A. Sardo, T. R. Ward, Angew. Chem. 2008, 120, 713–717; Angew.
Chem. Int. Ed. 2008, 47, 701–705.
[22] X. Y. Li, D. R. Liu, J. Am. Chem. Soc. 2003, 125, 10188–10189.
[23] a) I. A. Kozlov, L. E. Orgel, P. E. Nielsen, Angew. Chem. 2000, 112,
4462–4465; Angew. Chem. Int. Ed. 2000, 39, 4292–4295; b) M.
Bolli, R. Micura, A. Eschenmoser, Chem. Biol. 1997, 4, 309–320.
[24] P. Ordoukhanian, G. F. Joyce, J. Am. Chem. Soc. 2002, 124, 12499–
12506.
[25] B. Seelig, S. Keiper, F. Stuhlmann, A. Jꢃschke, Angew. Chem. 2000,
112, 4764–4768; Angew. Chem. Int. Ed. 2000, 39, 4576–4579.
[26] G. Roelfes, B. L. Feringa, Angew. Chem. 2005, 117, 3294–3296;
Angew. Chem. Int. Ed. 2005, 44, 3230–3232.
formula
C36H40CuN7O2]3+·[C34H39CuN6O3]3+
·6 (C2H3N)
[ClO4]ꢁ·3
N
ACHTUNGTRENNUNG
formula weight [gmolꢁ1]
crystal system
space group (no.)
2029.43
triclinic
P1 (2)
¯
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
13.806(1)
16.682(2)
19.765(2)
100.922(2)
90.722(2)
99.422(2)
4404.9(8)
b [8]
g [8]
V [ꢂ3]
q range unit cell: min.–max. [8],
2.52–27.38, 7360
reflections
Z
2
1calcd [gcmꢁ3
]
1.530
F
N
2096
m
G
]
7.54
green, platelet
0.33ꢁ0.20ꢁ0.08
0.71073
100(1)
2.23, 25.03
31406
lACHTUNGTRENNUNG
T [K]
q range: min., max. [8]
total data
unique data
15289
8551
0.0685
0.1354
15289
data with criterion: Fo ꢃ4.0s(F2o)
R
R
int =S[jF2oꢁFo2(mean)j]/S[F2o]
sig =Ss(F2o)/s[Fo2]
number of reflections
number of refined parameters
final agreement factors:
1153
wR(F2)={S[w(Fo2ꢁF2c)2]/
0.1735
0.0690
1
S[w(F2o)2]} =
2
R(F)=S(jjFo jꢁjFc jj)/SjFo j
[27] G. Roelfes, A. J. Boersma, B. L. Feringa, Chem. Commun. 2006,
635–637.
[28] A. J. Boersma, B. L. Feringa, G. Roelfes, Org. Lett. 2007, 9, 3647–
3650.
(For Fo >4.0s(Fo))
1
GooF=S={S[w(F2oꢁFc2)2]/(nꢁp)} = 0.958
2
[29] A. J. Boersma, J. E. Klijn, B. L. Feringa, G. Roelfes, J. Am. Chem.
Soc. 2008, 130, 11783–11790.
[30] D. Coquiꢄre, B. L. Feringa, G. Roelfes, Angew. Chem. 2007, 119,
9468–9471; Angew. Chem. Int. Ed. 2007, 46, 9308–9311.
[31] A. J. Boersma, B. L. Feringa, G. Roelfes, Angew. Chem. 2009, 121,
3396–3398; Angew. Chem. Int. Ed. 2009, 48, 3346–3348.
[32] a) N. Shibata, H. Yasui, S. Nakamura, T. Toru, Synlett 2007, 1153–
1157; b) E. W. Dijk, B. L. Feringa, G. Roelfes, Tetrahedron: Asym-
metry 2008, 19, 2374–2377.
Acknowledgements
Financial support from ERA-Chemistry, the Netherlands Organisation
for Scientific Research (NWO-CW), and NRSC-Catalysis is gratefully ac-
knowledged.
[33] a) T. N. Grossmann, A. Strohbach, O. Seitz, ChemBioChem 2008, 9,
2185–2192; b) N. Sancho Oltra, G. Roelfes, Chem. Commun. 2008,
6039–6041; c) Z. Tang, A. Marx, Angew. Chem. 2007, 119, 7436–
7439; Angew. Chem. Int. Ed. 2007, 46, 7297–7300; d) L. Ropartz,
N. J. Meeuwenoord, G. A. van der Marel, P. van Leeuwen, A. M. Z.
Slawin, P. C. J. Kamer, Chem. Commun. 2007, 1556–1558; e) Z.
Tang, D. P. N. GonÅalves, M. Wieland, A. Marx, J. S. Hartig, Chem-
BioChem 2008, 9, 1061–1064.
[1] a) J. Steinreiber, T. R. Ward, Coord. Chem. Rev. 2008, 252, 751–766;
b) G. Roelfes, Mol. BioSyst. 2007, 3, 126–135; c) R. Krꢃmer, Angew.
Chem. 2006, 118, 872–874; Angew. Chem. Int. Ed. 2006, 45, 858–
860.
[2] M. D. Mihovilovic, J. Chem. Technol. Biotechnol. 2007, 82, 1067–
1071.
[34] a) S. Otto, J. Engberts, J. Am. Chem. Soc. 1999, 121, 6798–6806;
b) S. Otto, F. Bertoncin, J. B. F. N. J. Engberts, J. Am. Chem. Soc.
1996, 118, 7702–7707; c) F. M. Menger, C. E. Portnoy, J. Am. Chem.
Soc. 1967, 89, 4698–4703.
[35] See the Supporting Information.
[36] F. Bailly, C. Bailly, M. J. Waring, J. P. Hꢅnichart, Biochem. Biophys.
Res. Commun. 1992, 184, 930–937.
[3] Y. Lu, Curr. Opin. Chem. Biol. 2005, 9, 118–126.
[4] D. F. Qi, C. M. Tann, D. Haring, M. D. Distefano, Chem. Rev. 2001,
101, 3081–3111.
[5] M. T. Reetz, Proc. Natl. Acad. Sci. USA 2004, 101, 5716–5722.
[6] a) C. Letondor, A. Pordea, N. Humbert, A. Ivanova, S. Mazurek, M.
Novic, T. R. Ward, J. Am. Chem. Soc. 2006, 128, 8320–8328; b) C.
Letondor, N. Humbert, T. R. Ward, Proc. Natl. Acad. Sci. USA
2005, 102, 4683–4687.
[37] C. Liu, Z. Jiang, Y. Zhang, Z. Wang, X. Zhang, F. Feng, S. Wang,
Langmuir 2007, 23, 9140–9142.
[7] G. Klein, N. Humbert, J. Gradinaru, A. Ivanova, F. Gilardoni, U. E.
Rusbandi, T. R. Ward, Angew. Chem. 2005, 117, 7942–7945; Angew.
Chem. Int. Ed. 2005, 44, 7764–7767.
ˇ
ˇ
[38] J. Sebestꢆk, I. Stibor, J. Hlavꢇcek, J. Pept. Sci. 2006, 12, 472–480.
[39] M. P. Sibi, Y. H. Yang, Synlett 2008, 83–88.
[8] M. T. Reetz, J. J. P. Peyralans, A. Maichele, Y. Fu, M. Maywald,
Chem. Commun. 2006, 4318–4320.
[40] M. E. Quiroz-Castro, G. A. vanAlbada, I. Mutikainen, U. Turpeinen,
J. Reedijk, Inorg. Chim. Acta 2000, 297, 129–133.
9604
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9596 – 9605

DNA-Based Asymmetric Catalysis
FULL PAPER
[41] R. Kannappan, S. Tanase, I. Mutikainen, U. Turpeinen, J. Reedijk,
Inorg. Chim. Acta 2005, 358, 383–388.
[42] H. Kooijman, A. L. Spek, D. Rehorst, W. L. Driessen, J. Reedijk,
Acta Crystallogr. Sect. C 1997, 53, 1596–1598.
[45] M. Ardhammar, T. Kurucsev, B. Nordꢅn in Circular Dichroism:
Principles and Applications, 2nd. ed. (Eds.: N. Berova, K. Nakanishi,
W. R. Woody), Wiley-VCH, Weinheim, 2000,Chapter 26.
[43] R. Paulini, K. Mꢈller, F. Diederich, Angew. Chem. 2005, 117, 1820–
1839; Angew. Chem. Int. Ed. 2005, 44, 1788–1805.
Received: February 19, 2009
Revised: April 22, 2009
[44] D. Fornasiero, T. Kurucsev, Biophys. Chem. 1985, 23, 31–37.
Published online: July 3, 2009
Chem. Eur. J. 2009, 15, 9596 – 9605
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9605