Towards the sequence-specific multivalent molecular recognition of cyclodextrin oligomers

Article abstract of DOI:10.3762/bjoc.10.253

Sequence-specific multivalent molecular recognition has been recognized to play a major role in biological processes. Furthermore, sequence-specific recognition motifs have been used in various artificial systems in the last years, e.g., to emulate biological processes or to build up new materials with highly specific recognition domains. In this article, we present the preparation of cyclodextrin (CD)-based strands and complementary and non-complementary strands modified with guest molecules and the investigation of their complexation behavior towards each other by isothermal titration calorimetry (ITC). As complementary binding motifs n-butyl and α-CD and adamantane and β-CD were selected. It was found that it is possible to realize sequencespecific molecular recognition by the use of host-guest chemistry, but the recognition motifs as well as the linkages have to be chosen very carefully. In the case of trivalent systems one adamantane moiety must be included to induce preferred formation of 1:1 adducts. Due to the too weak interaction between n-butyl and α-CD these systems have a negative chelate cooperativity and open adducts are preferentially formed. As soon as two adamantane moieties are present, the complementary systems have a positive chelate cooperativity and double-stranded structures are favored over open adducts. In this system the n-butyl moiety provides insufficient discrimination towards α- and β-CD and no sequence specificity is observed. By the combination of three adamantane moieties sequence specificity can be generated. Exclusively with the complementary CD sequence double-stranded structures are formed, with non-complementary strands aggregates of higher stoichiometry are generated.

Full text of DOI:10.3762/bjoc.10.253

Towards the sequence-specific multivalent molecular
recognition of cyclodextrin oligomers
Michael Kurlemann and Bart Jan Ravoo*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Organic Chemistry Institute, Westfälische Wilhelms-Universität
Münster, Corrensstrasse 40, 48149 Münster, Germany
doi:10.3762/bjoc.10.253
Received: 11 July 2014
Email:
Accepted: 19 September 2014
Published: 20 October 2014
Bart Jan Ravoo* - b.j.ravoo@uni-muenster.de
* Corresponding author
This article is part of the Thematic Series "Superstructures with
cyclodextrins: Chemistry and applications II".
Keywords:
cooperativity; cyclodextrins; molecular recognition; multivalency;
sequence specificity
Guest Editor: G. Wenz
© 2014 Kurlemann and Ravoo; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
Sequence-specific multivalent molecular recognition has been recognized to play a major role in biological processes. Furthermore,
sequence-specific recognition motifs have been used in various artificial systems in the last years, e.g., to emulate biological
processes or to build up new materials with highly specific recognition domains. In this article, we present the preparation of
cyclodextrin (CD)-based strands and complementary and non-complementary strands modified with guest molecules and the
investigation of their complexation behavior towards each other by isothermal titration calorimetry (ITC). As complementary
binding motifs n-butyl and α-CD and adamantane and β-CD were selected. It was found that it is possible to realize sequence-
specific molecular recognition by the use of host–guest chemistry, but the recognition motifs as well as the linkages have to be
chosen very carefully. In the case of trivalent systems one adamantane moiety must be included to induce preferred formation of
1:1 adducts. Due to the too weak interaction between n-butyl and α-CD these systems have a negative chelate cooperativity and
open adducts are preferentially formed. As soon as two adamantane moieties are present, the complementary systems have a posi-
tive chelate cooperativity and double-stranded structures are favored over open adducts. In this system the n-butyl moiety provides
insufficient discrimination towards α- and β-CD and no sequence specificity is observed. By the combination of three adamantane
moieties sequence specificity can be generated. Exclusively with the complementary CD sequence double-stranded structures are
formed, with non-complementary strands aggregates of higher stoichiometry are generated.
Introduction
Multivalency is the interaction of a receptor and a ligand with at role in almost all biological processes, e.g., the recognition of
least two recognition motifs on each binding partner [1]. In cells by other cells, bacteria or viruses, the adhesion of cells or
recent years multivalency has been recognized to play a major signal transduction pathways [2]. By the combination of
2428

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
multiple, rather weak non-covalent interactions stable yet re- amine respectively a carboxylic acid function at the primary
versible systems are generated, which are responsive to external side, followed by complete deprotection under Zemplén condi-
stimuli. These advantages have made synthetic multivalent tions (Figure 2). The trivalent CD sequences 4–7 (Figure 1B)
systems interesting for a broad field of applications. In the case were prepared by amide coupling of peracetylated 6A,D-di-
of medicinal applications multivalent molecules have been used amine functionalized α- and β-CD with monocarboxylic acid
as inhibitors of toxins or viruses and for imaging and targeted functionalized α- and β-CD, again followed by complete depro-
drug delivery [3]. Hydrogels which are built up by multivalent tection under Zemplén conditions (Figure 2). Based on MALDI
host–guest interactions and vesicles of amphiphilic host mole- mass spectra of the protected and unprotected cyclodextrin
cules have been intensively studied for their ability to function strands impurities by monomeric building blocks respectively
as drug delivery systems as well [4-9]. Additionally, such vesi- dimeric species in the case of trivalent strands can be excluded
cles can be modified with bio-active ligands and serve as model (see Supporting Information File 1). The di- and trivalent guest
systems to mimic biological processes on cell membranes strands 8–14 (Figure 1C, D) were synthesized by solid phase
[10,11]. In the field of materials science multivalency has been peptide synthesis using a standard Fmoc-protocol (Figure 3).
used to create functional polymers [12-14] and self-assembled Therefore the serine derivatives 15 and 16 (Figure 1E) and a
electronic [15-20] or biofunctional materials [21-27]. Even the water-soluble linker molecule were used. The purity of the
molecular recognition of macroscopic gel blocks by multiva- guest strands is estimated to be >95% based on 1H NMR
lent host–guest interactions has been realized [28-32].
spectra (see Supporting Information File 1). The syntheses of
the multivalent host and guest strands are described in detail in
Besides the number of receptor–ligand interactions their spatial Supporting Information File 1.
distribution is crucial for the highly selective molecular recogni-
tion as well. The most important natural example of sequence First of all, the selectivity of the complexation of the unpro-
specific, multivalent molecular recognition is the hybridization tected serine derivatives 17 and 18 towards α- and β-CD was
of complementary DNA strands via the base pairing of adeno- investigated by ITC experiments. The structures of the
sine and thymine respectively guanine and cytosine. Within the host–guest complexes were elucidated by NMR spectroscopy.
last years these binding motifs have been transferred to artifi- The 1-adamantane-functionalized serine 17 shows complexa-
cial systems like peptide nucleic acids (PNA) [33] and exten- tion of α- and β-CD, forming 1:1 complexes. In both cases the
sively used to mimic biological processes [34,35] or to generate adamantane moiety binds into the CD cavity, which is
functional materials [36]. Host–guest chemistry has been confirmed by NMR measurements of 17 and 1:10 mixtures of
studied in the field of sequence-specific molecular recognition 17 with α- and β-CD (Figure 4). After addition of 10 equiva-
as well. The selective recognition of short peptides made of lents of α- or β-CD the signals of the adamantane’s protons are
natural amino acids with aromatic side chains by different host significantly shifted to higher ppm values compared to the
moieties like coordination cages [37] and cucurbiturils [38,39] signals of pure 17. The other signals show almost no variation.
has been demonstrated. For cyclodextrins (CD) a similar ap- The interaction of 17 with β-CD is enthalpically as well as
proach is reported, but by using CD strands and different model entropically favored, while the interaction of 17 with α-CD is
peptides of natural and artificial amino acids no significant exclusively driven by the complexation enthalpy (Table 1). All
selectivity was observed [40]. In this work we present an alter- thermodynamic data are in agreement with literature-known
native approach to realize the hybridization of complementary data of comparable systems [41]. Because the interaction
strands mediated by multivalent host–guest interaction. We between 17 and β-CD has a ca. 400-fold higher binding
used α- and β-CD because of their well-known and regiospe- constant than the interaction of 17 and α-CD the preferred com-
cific modifiability for the preparation of di- and trivalent host plexation behavior towards β-CD can be expected (Figure 5A).
sequences and investigated their binding behavior towards
complementary and non-complementary di- and trivalent guest The n-butyl derivate 18 also interacts with both α- and β-CD. In
sequences which were modified with n-butyl and 1-adamantyl the case of α-CD the formation of a 1:1 adduct with a binding
moieties. Such structures can be used for the self-assembly of constant of ca. 102 M−1 is observed. The complexation is driven
complicated molecular architectures. Furthermore, the results by a negative complexation enthalpy (Table 1) and leads to the
foster the understanding of the basic principles of sequence- inclusion of the n-butyl moiety of 18. This is confirmed by
specific molecular recognition, which is ubiquitous in nature. NMR spectra, where the protons of the n-butyl unit show strong
shifting to higher ppm values after addition of 10 equivalents of
Results and Discussion
α-CD (Figure 6B and C). The interaction of 18 with β-CD
The divalent CD sequences 1–3 (Figure 1A) were synthesized cannot be quantified based on the ITC measurement with 1 mM
by the amide coupling of peracetylated α- and β-CD, bearing an of 18 and 10 mM of β-CD. Nevertheless, the NMR spectra of
2429

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 1: Chemical structures of the di- (A) and trivalent (B) CD sequences, the di- (C) and trivalent (D) guest sequences and the monomeric serine
derivatives (E).
the 1:10 mixture of 18 and β-CD show small variations in the All in all, the monovalent guest molecules 17 and 18 show
chemical shifts of the protons compared to the spectra of pure discrimination towards α- and β-CD in their complexation
18, giving a strong hint that complexation takes place behavior: 17 prefers to complex β-CD, 18 prefers to complex
(Figure 6A and B). Taking the results of both experiments into α-CD.
account, a binding constant lower than 102 M−1 can be assumed
for the interaction between 18 and β-CD. Additional measure- In the next step the divalent guest strands 8, 9 and 10 were
ments with 10-fold increased concentrations of both 18 and investigated regarding their complexation behavior towards the
β-CD, which is assumed to be necessary to determine binding CD dimers 1, 2 and 3. Analysis of the ITC data was done using
constants lower than 102 M−1, were not possible because of too different binding models, based on the host–guest stoichio-
low solubility of the components. Thereby, the preferred metry of each system (Figure 7 and Supporting Information
binding of 18 towards α-CD is observed (Figure 5B).
File 2) [42,43].
2430

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 2: Synthesis of the di- and trivalent CD sequences 1–7 (for detailed reaction conditions see Supporting Information File 1).
The doubly n-butyl substituted strand 8 forms with both the the guest strands supramolecular polymerization can occur in
complementary α–α dimer 1 and the non-complementary β–β these systems (Figure 8A). Additionally, the ITC data of the
dimer 3 1:1 aggregates. Analysis of the ITC data with a multi- systems 1/8 and 3/8 were analyzed by a 1:1 overall binding
valent binding model gives effective molarities (EM) of model. Here all complexation steps are combined in one set of
0.33 mM for the system 1/8 and 0.22 mM for the system 3/8 thermodynamic parameters. With this method similar binding
(Table 2). In combination with the intrinsic binding constants Ki constants and thermodynamic parameters are calculated for
of both systems the specific chelate cooperativities can be both interactions, so that no sequence-specificity in the com-
calculated by multiplication of EM and Ki. With these values a plexation behavior of the divalent guest strand 8 is observed
decision about the structures of the 1:1 aggregates, which can (Table 2).
exist as open aggregates or closed cyclic systems, can be made.
For the system 1/8 the intrinsic binding constant is taken from The heterodivalent guest strand 9 forms a 1:1 aggregate with the
the interaction between 18 and α-CD and set to 102 M−1. In complementary CD strand 2. Due to the combination of two
case of the system 3/8 the monovalent interaction between 18 different recognition motifs in this system the multivalent
and β-CD cannot be quantified by ITC measurements. There- analysis of the ITC data is more complicated than for the previ-
fore, the binding constant was overestimated to be 102 M−1 as ously discussed homodivalent systems. Instead of calculating
well. Based on these assumptions the chelate cooperativity is the EM it was estimated to be 0.25 mM because of the struc-
0.03 for the system 1/8 and 0.02 for the system 3/8. Both values tural similarities of the system 2/9 with the systems 1/8 and 3/8.
are significantly lower than 1. This indicates negative chelate With a view to the much higher binding constant of 17 towards
cooperativity and preferred formation of open 1:1 adducts for β-CD compared to all other possible host–guest interactions, it
both systems. Depending on the concentration of the host and is obvious that this inclusion complex is formed first. In the
2431

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 3: Solid phase peptide synthesis of the di- and trivalent guest strands 8 – 14 (for detailed reaction conditions see Supporting Information
File 1).
second step the intramolecular complexation between the The homodivalent guest strand 8 with two adamantane moieties
n-butyl moiety of 9 and the α-CD of 2, (Ki ≈ 102 M−1) has to shows an alternating complexation behavior towards the
appear to build a double stranded structure. Here again a nega- complementary CD dimer 3. The ITC data suggest that instead
tive chelate cooperativity is present (Ki·EM ≈ 0.03), preventing of 1:1 adducts 2:1 host–guest systems are formed. Therefore,
the intramolecular complexation. The formation of open 1:1 the analysis was done using a 2:1 binding model, assuming two
aggregates is further confirmed by the analysis of the ITC data non-cooperative and independent complexations. This method
by a 1:1 model of overall complexation. The overall binding gives intrinsic thermodynamic parameters (Table 2). The
constant as well as the overall thermodynamic parameters of the intrinsic binding constant is ca. 2 × 105 M−1. This value is
1:1 analysis are very similar to the values observed for the com- 4-fold higher than the binding constant between 17 and β-CD,
plexation of 17 with β-CD (Table 2). Therefore, the interaction caused by the symmetry effect of the interaction between two
between 2 and 9 is only based on the interaction between the homodivalent systems [1]. The intrinsic complexation enthalpy
adamantane moiety of 9 and the β-CD of 2. The interaction and entropy show that both complexations are enthalpically as
between the n-butyl moiety and the α-CD is negligible. As well as entropically driven. The interaction of 8 with the non-
already mentioned for the systems 1/8 and 3/8 in the system 2/9 complementary CD dimer 1 results in the formation of open 1:1
supramolecular polymerization is possible, depending on the adducts. Analysis of the corresponding ITC data with a multiva-
concentrations of the host and the guest strand (Figure 8B).
lent binding model yields an EM of 0, indicating that no
2432

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 4: 1H NMR spectra of 17 (B) and its inclusion complexes with β- (A) and α-CD (C), measured in D2O (600 MHz, 25 °C). A) [17] = 1.0 mM,
[β-CD] = 10 mM. B) [17] = 1.0 mM. C) [17] = 1.0 mM, [α-CD] = 10 mM.
Table 1: Thermodynamic parameters of the host–guest interactions of the serine derivatives 17 and 18 with α- and β-CD.
Serine derivative
CD
17
18
α-CD
β-CD
α-CD
β-CD
n
1.00
1.06 × 102
−11.6
0.94
3.97 × 104
−26.2
1.00
9.71 × 101
−11.3
–
–
–
–
–
K [M−1]
ΔG [kJ/mol]
ΔH [kJ/mol]
ΔS [J/molK]
−19.7
−24.3
−14.9
−27.5
6.5
−11.9
Figure 5: Schematic drawing of the complexation of the serine derivatives 17 (A) and 18 (B) with α- and β-CD. The shown binding constants were
determined by ITC, the structures by NMR. n.d. = not detectable by ITC.
2433

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 6: 1H NMR spectra of 18 (B) and its inclusion complexes with β- (A) and α-CD (C), measured in D2O (600 MHz, 25 °C). A) [18] = 1.0 mM,
[β-CD] = 10 mM. B) [18] = 1.0 mM. C) [18] = 1.0 mM, [α-CD] = 10 mM.
intramolecular complexation takes place. Analysis of the data In the last step the interactions of the trivalent guest strands 11,
with a 1:1 overall binding model gives a binding constant of 12, 13 and 14 with complementary and non-complementary
ca. 3 × 102 M−1, which is higher than the binding constant trivalent CD strands were investigated. The three times n-butyl-
between 17 and α-CD due to the symmetry effect. The com- substituted guest strand 11 forms a 2:1 host–guest adduct with
plexation enthalpy and entropy differ from the monovalent the complementary CD strand 4. The intrinsic binding constant
interaction α-CD/17 because of the bridging of the host and the and the intrinsic thermodynamic parameters (Table 3), taken
guest molecules (Figure 8C).
from a 2:1 binding model, are in agreement with the values of
the monomeric interaction between 18 and α-CD. This indi-
In summary, for none of the divalent strands sequence-specific cates that the 2:1 host–guest adduct is formed by two inde-
binding can be observed. This is due to negative chelate cooper- pendent, non-cooperative interactions between two n-butyl
ativities of the systems which are caused by too low intrinsic moieties of 11 and α-CD each of one CD trimer 4 (Figure 9A).
binding constants on the one hand and too low EM on the other The interaction of 11 with the non-complementary CD strand 7
hand. While the intrinsic binding constant can easily be leads to a host–guest aggregate which cannot be characterized
increased by the variation of the guest moieties, especially for by the data of the corresponding ITC measurement.
the interaction with α-CD, the EM cannot be increased that
easy. Here the linkers of the guest and the host moieties have a For the trivalent guest strand 12 bearing one adamantane moiety
crucial influence on the structure of the divalent molecules and the selective formation of a 1:1 host–guest adduct with the
the host–guest complexes before the intramolecular complexa- complementary CD strand 5 is observed. Because of the
tion [1]. Changes in the linkers’ structures can result in different complexity of a multivalent binding model for a heterotrivalent
steric environments during the interactions, but the exact effects system the EM of the interaction between 5 and 12 was not
cannot be predicted and have to be solved by theoretical calcu- calculated but estimated from the structural similar divalent
lations and simulations.
systems 1/8 and 3/8 and set to be 0.25 mM. Taking into account
2434

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 7: Selected binding models for the analysis of ITC data. (A) Monovalent receptor (R)-ligand (L) interaction. (B) Multivalent interaction of a diva-
lent receptor (RR) and a divalent ligand (LL). (C) 1:1 overall interaction of a divalent receptor (RR) and a divalent ligand (LL). (D) 2:1 interaction of a
divalent receptor (RR) and a divalent ligand (LL). EM = effective molarity. See Supporting Information File 2 for details.
Table 2: Thermodynamic parameters of the interactions of the divalent guest strands 8, 9 and 10 with complementary and non-complementary diva-
lent CD strands.
guest strand
CD strand
8
9
2
10
1
3
1
3
n
1.00
3.05 × 102 a
−14.2 a
−13.3 a
3.0 a
1.00
3.69 × 102 a
−14.7 a
−4.2 a
1.06
3.11 × 104 a
−25.6 a
−25.5 a
0.3 a
1.00
3.12 × 102 a
−14.2 a
−12.0 a
7.5 a
2.37
1.69 × 105 b
−29.8 b
−16.5 b
44.7 b
K [M−1]
ΔG [kJ/mol]
ΔH [kJ/mol]
ΔS [J/molK]
EM [mM]
35.0 a
0.33 c
0.22 c
0.25 d
0 c
–
a1:1 overall binding model, b2:1 binding model, intrinsic value, cmultivalent binding model, destimated value.
the complexation behavior of the monovalent systems and the The following intramolecular complexation of one of the
results from the heterodivalent system 2/9 it is obvious that n-butyl moieties by one of the α-CD (Ki ≈ 102 M−1) does not
during the interaction between 5 and 12 first the complexation occur due to a negative chelate cooperativity (Ki·EM ≈ 0.03).
between the adamantane of 12 and the β-CD of 5 takes place. Therefore only one inclusion complex is formed, leading to an
2435

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 8: Schematic drawing of the interactions of the divalent guest molecules 8 (A), 9 (B) and 10 (C) with complementary (right) and non-comple-
mentary (left) divalent CD strands.
open 1:1 adduct of 5/12. This assumption is confirmed by the adamantane respectively n-butyl moieties of 12 is no limitation
modeling of the ITC data with a 1:1 overall binding model. and the simplified 3:1 model is a valid approach to analyse the
Here a binding constant of ca. 3 × 104 M−1 is calculated. This ITC data. The intrinsic binding constant of ca. 2 × 103 M−1 is
value as well as the overall thermodynamic parameters one order of magnitude higher than the comparable monovalent
(Table 3) are in good agreement with the values of the inter- binding constants of α-CD/17 and α-CD/18. This is caused by
action between 17 and β-CD. In contrast to the 1:1 adduct of the the symmetry effect, which is even higher due to the negli-
complementary strands 12 forms a 3:1 host–guest system with gence of different molecular recognition motifs for the simpli-
the non-complementary CD strand 4 and a 1:3 host–guest fied modelling. The thermodynamic parameters surprisingly
system with the non-complementary CD strand 7. The ITC data indicate that every inclusion process is mainly driven by the
of the interaction between 4 and 12 can be modeled with a 3:1 complexation entropy although four molecules are combined in
model assuming that three independent, non-cooperative one aggregate. The analysis of the ITC data of the system 7/12
complexations with the same intrinsic binding constant occur was done with a 1:3 binding model where only the interaction
(Table 3). With a view to the almost identical binding constants between the adamantane moiety of 12 and the β-CD of 7 was
between α-CD and 17 respectively 18 the assumption of same taken into account. This approach yields an intrinsic binding
binding constants for the interactions between α-CD and the constant and intrinsic thermodynamic parameters (Table 3)
Table 3: Thermodynamic parameters of the interactions of the trivalent guest strands 11, 12, 13 and 14 with complementary and non-complementary
divalent CD strands.
guest strand
CD strand
11
4
12
4
13
14
5
7
6
7
7
6
n
2.38
0.93
2.70
0.31
1.22
0.77
1.21
2.10
K [M−1]
1.56 × 102 a 2.53 × 104 b 2.23 × 103 c 1.74 × 104 d 3.91 × 105 b 5.47 × 105 b 1.39 × 106 b 5.26 × 104 a
ΔG [kJ/mol]
ΔH [kJ/mol]
ΔS [J/molK]
EM [mM]
−12.5 a
−15.0 a
−8.5 a
–
−25.1 b
−27.7 b
−8.6 b
−19.1 c
−2.6 c
55.4 c
–
−24.2 d
−24.3 d
−0.5 d
–
−31.9 b
−35.3 b
−11.4 b
0.07 f
−32.7 b
−52.8 b
−67.5 b
0.12 f
−35.1 b
−57.4 b
−75.1 b
0.25 e
−26.9 a
−36.0 a
−30.4 a
–
0.25 e
a2:1 binding model, intrinsic value, b1:1 overall binding model, c3:1 binding model, intrinsic value, d1:3 binding model, intrinsic value, eestimated value,
fsimplified, multivalent binding model.
2436

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
Figure 9: Schematic drawing of the interactions of the trivalent guest molecules 11 (A), 12 (B), 13 (C) and 14 (D) with complementary (right) and non-
complementary (left) trivalent CD strands.
which are consistent with the values of the interaction between binding constant between β-CD and 17 as intrinsic value
β-CD and 17. Therefore, three independent, non-cooperative (Ki ≈ 4 × 104 M−1), the system 6/13 has a chelate cooperativity
complexations of the adamantane moieties of three equivalents around 3. Because this value is greater than 1 a positive chelate
of 12 by the β-CD of 7 can be assumed (Figure 9B).
cooperativity is observed for the interaction of 6 and 13,
favoring the intramolecular complexation and the formation of a
The heterotrivalent guest strand 13 with two adamantane double-stranded structure. For the interaction of 13 and the non-
moieties is the first strand which shows formation of double- complementary strand 7 the same approach of analyzing the
stranded aggregates. The ITC data of the interaction between 13 data can be done. The 1:1 overall binding model gives a binding
and the complementary CD strand 6 can be analyzed with a 1:1 constant of ca. 5 × 105 M−1, which is similar to the value of the
overall binding model resulting in stoichiometry of 1.22 and an complementary system 6/13. Neglecting the interaction between
overall binding constant of ca. 4 × 105 M−1 (Table 3). This is the n-butyl moiety of 13 and the β-CD of 6 the simplified multi-
one order of magnitude higher than the binding constant valent binding model can be used for modelling the ITC data,
between 17 and β-CD, which can be attributed to the combina- resulting in an EM of 0.12 mM. Thus, the system 7/13 shows
tion of two adamantane-β-CD interactions in one value. Taking positive chelate cooperativity as well (Ki·EM ≈ 5) and double-
into account that in the prior discussed heteromultivalent stranded structures are preferentially formed. All in all, the
systems the interactions between n-butyl moieties and α-CD can guest strand 13 shows preferred formation of double-stranded
be neglected, this is presumably valid for the system 6/13 as 1:1 adducts, but not in a sequence-specific way. This is caused
well. Therefore, the complicated heteromultivalent system can by the too weak discrimination of the n-butyl moiety of 13
be simplified to be the interaction of a homodivalent guest- towards the α- and β-CD units of 6 respectively 7 (Figure 9C).
strand with two adamantane moieties with a homodivalent CD
strand of two β-CD and the ITC data can be analyzed with a Finally, the homotrivalent guest strand 14 shows sequence-
multivalent binding model with two non-cooperative complexa- specific molecular recognition. In combination with the comple-
tions. This analysis gives an EM of ca. 0.07 mM. Taking the mentary CD strand 7 the ITC data show the formation of 1:1
2437

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
adducts. These exist preferentially as closed double-stranded while the n-butyl modified serine 18 prefers the complexation
structures because of a positive chelate cooperativity. Assuming of α-CD. Going to divalent systems neither the formation of
that the EM of the system 7/14 is similar to the EM of the struc- closed 1:1 adducts nor is the desired sequence specific molec-
tural related divalent systems (EM ≈ 0.25 mM) and taking the ular recognition observed due to negative chelate cooperativi-
binding constant between 17 and β-CD as intrinsic value ties of all systems. Finally, the trivalent guest strands show a
(Ki ≈ 4 × 104 M−1) the chelate cooperativity is around 10. This specific 1:1 interaction with the complementary CD strand as
means that the intramolecular complexation, which leads to soon as one adamantyl moiety is included. The monoadamantyl
double-stranded structures, is favored over the formation of substituted guest strand 12 forms an open 1:1 adduct with the
open 1:1 adducts. These assumptions are confirmed by the complementary CD trimer 5. Because of negative chelate coop-
analysis of the ITC data with a 1:1 overall binding model. Here erativity only the inclusion complex between the adamantyl
a binding constant of ca. 106 M−1 is obtained (Table 3). This is moiety of 12 and the β-CD of 5 is formed. If two adamantyl
one order of magnitude higher than the overall binding constant substituents are present in the guest strands, positive chelate
of the systems 6/13 and 7/13 and indicates that three complexa- cooperativity is generated. As long as the concentrations of the
tions between β-CD and adamantane are combined in one value. single strands are lower than the system specific effective
In contrast to that both the non-complementary systems 5/14 molarity 1:1 double-strands are preferentially built up. The
and 6/14 show the formation of 2:1 host–guest adducts. The diadamantyl-substituted guest strand 13 shows no sequence
corresponding ITC data can be analyzed using 2:1 models specific molecular recognition, although double stranded struc-
where the complexation of the adamantane moieties by α-CD, tures are formed. With both the complementary cyclodextrin
which has a much lower binding constant than the interaction trimer 6 and the non-complementary cyclodextrin trimer 7
between adamantane and β-CD, is neglected. The intrinsic cyclic 1:1-adducts are preferred, because the n-butyl substituent
binding constants, which are obtained by this method, are ca. of 13 provides no sufficient discrimination in the complexation
5 × 104 M−1 for the system 6/14 and ca. 2 × 103 M−1 for the of α- and β-CD. The trivalent homoadamantyl substituted guest
system 5/14 (Table 3). The clear difference is caused by statis- strand 14 shows sequence specific molecular recognition.
tical as well as steric reasons. In the case of the interaction Cyclic 1:1 adducts are exclusively generated with the comple-
between 6, which has two β-CD, and 14 more complexations mentary cyclodextrin trimer 7. With non-complementary
between one adamantane moiety and one β-CD are possible in cyclodextrin trimers host–guest systems of higher stoichio-
comparison to the interaction between 5, which has only one metry are formed. Based on these results further development of
β-CD, and 14. This difference influences the symmetry effect the guest strands can be done to create even heteromultivalent
and leads to different intrinsic binding constants. Additionally, systems which show highly selective molecular recognition and
the steric circumstances of the 2:1 host−guest adducts 5/14 and can be used for the defined self-assembly of molecular struc-
6/14 differ from each other. The CD strand 6 has terminally tures.
located β-CD units while the β-CD of 5 is located in the centre
of the sequence. Thereby the 2:1 adduct of 6 and 14 can avoid Experimental
sterical hindrance of the uncomplexed cyclodextrins more Materials: Throughout this work, chemicals were used as
easily than the 2:1 adduct of 5 and 14, resulting in a higher received from Acros Organics (Thermo Fischer Scientific Inc.,
intrinsic binding constant (Figure 9D). These results demon- Waltham, Massachusetts, USA), Aldrich (Sigma-Aldrich Corp.,
strate, that the homotrivalent guest strand 14 shows sequence- St. Louis, Missouri, USA), Alfa Aesar (Alfa Aesar, Ward Hill,
specificity in its molecular recognition. With the complemen- Massachusetts, USA), Carbolution Chemicals (Carbolution
tary CD strand a closed 1:1 adduct is preferentially formed, Chemicals GmbH, Saarbrücken, Germany), Fluka (Sigma-
with non-complementary CD strands host–guest adducts of Aldrich Corp., St. Louis, Missouri, USA), Iris Biotech GmbH
higher stoichiometry are favored.
(Iris Biotech GmbH, Marktredwitz, Germany), Merck (Merck
KGaA, Darmstadt, Germany), Novabiochem (Merck KGaA,
Darmstadt, Germany) or Wacker (Wacker Chemie AG,
Conclusion
In this work we present our first successful attempt to realize München, Germany). The synthesis and analysis of the mole-
the sequence-specific multivalent molecular recognition of cules 1–18 is reported in Supporting Information File 1.
cyclodextrin sequences and complementary strands modified
with guest moieties. To this end we selected the interactions Methods: ITC measurements were performed with a TA Instru-
between adamantane and β-CD respectively n-butyl and α-CD ments Nano ITC Low Volume (Waters Corp., Milford, Massa-
as molecular recognition motifs. In the case of monovalent chusetts, USA) with a cell volume of 170 µL using ITCRun
interactions the expected discrimination was observed. The Version 2.1.7.0 Firmware version 1.31 (TA Instruments, Waters
adamantine-substituted serine 17 prefers to complex β-CD, Corp., Milford, Massachusetts, USA) as software. All titrations
2438

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
3. Joshi, A.; Vance, D.; Rai, P.; Thiyagarajan, A.; Kane, R. S.
were done using a 50 µL syringe and 20 injections of 2.5 µL at
a temperature of 25 °C and a stirring rate of 350 rpm. All
samples were prepared in 100 mM phosphate buffer pH 7.4 and
degassed for 10 minutes before use. The data were analysed
using NanoAnalyse Data Analysis version 2.36 (TA Instru-
ments, Waters Corp., Milford, Massachusetts, USA),
Microsoft® Excel version 14.07113.5005 as part of Microsoft®
Office Professional Plus 2010 (Microsoft Corp., Redmond,
Washington, USA) and OriginPro 9.1.0G (OriginLab Corp.,
Northampton, Massachusetts, USA). Before analysis all data
were corrected by substraction of a dilution measurement of the
titrated component into pure solvent. Concentrations of the
components and all ITC data are shown in Supporting Informa-
tion File 2.
Chem. – Eur. J. 2008, 14, 7738–7747. doi:10.1002/chem.200800278
4. Peng, K.; Tomatsu, I.; Kros, A. Chem. Commun. 2010, 46, 4094–4096.
doi:10.1039/c002565h
5. Appel, E. A.; Loh, X. J.; Jones, S. T.; Dreiss, C. A.; Scherman, O. A.
Biomaterials 2012, 33, 4646–4652.
doi:10.1016/j.biomaterials.2012.02.030
6. Jung, H.; Park, K. M.; Yang, J.-A.; Oh, E. J.; Lee, D.-W.; Park, K.;
Ryu, S. H.; Hahn, S. K.; Kim, K. Biomaterials 2011, 32, 7687–7694.
doi:10.1016/j.biomaterials.2011.06.060
7. Park, K. M.; Yang, J.-A.; Jung, H.; Yeom, J.; Park, J. S.; Park, K.-H.;
Hoffman, A. S.; Hahn, S. K.; Kim, K. ACS Nano 2012, 6, 2960–2968.
doi:10.1021/nn204123p
8. Park, K. M.; Lee, D.-W.; Sarkar, B.; Jung, H.; Kim, J.; Ko, Y. H.;
Lee, K. E.; Jeon, H.; Kim, K. Small 2010, 6, 1430–1441.
doi:10.1002/smll.201000293
9. Nalluri, S. K. M.; Voskuhl, J.; Bultema, J. B.; Boekema, E. J.;
Ravoo, B. J. Angew. Chem., Int. Ed. 2011, 50, 9747–9751.
doi:10.1002/anie.201103707
NMR measurements for elucidation of the structures of the
host–guest complexes were recorded on an Agilent DD2 600
(Agilent Technologies, Santa Clara, California, USA). Samples
were prepared by dissolving 10 equivalents of CD in the corres-
ponding volume of a 1 mM stock solution of the guest mole-
cules in D2O. Analysis of the data was done using MNova 9.0.0
(Mestrelab Research S. L., Santiago de Compostela, Spain).
The spectra were referenced to the residual solvent signal.
10.Voskuhl, J.; Ravoo, B. J. Chem. Soc. Rev. 2009, 38, 495–505.
doi:10.1039/b803782p
11.Kauscher, U.; Ravoo, B. J. Beilstein J. Org. Chem. 2012, 8,
1543–1551. doi:10.3762/bjoc.8.175
12.Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J. B.;
Hirschberg, J. H. K. K.; Lange, R. F. M.; Lowe, J. K. L.; Meijer, E. W.
Science 1997, 278, 1601–1604. doi:10.1126/science.278.5343.1601
13.Kautz, H.; van Beek, D. J. M.; Sijbesma, R. P.; Meijer, E. W.
Macromolecules 2006, 39, 4265–4267. doi:10.1021/ma060706z
14.Abbel, R.; Grenier, C.; Pouderoijen, M. J.; Stouwdam, J. W.;
Leclère, P. E. L. G.; Sijbesma, R. P.; Meijer, E. W.;
Schenning, A. P. H. J. J. Am. Chem. Soc. 2009, 131, 833–843.
doi:10.1021/ja807996y
Supporting Information
Supporting Information File 1
15.El-ghayoury, A.; Peeters, E.; Schenning, A. P. H. J.; Meijer, E. W.
Chem. Commun. 2000, 1969–1970. doi:10.1039/b005903j
16.Ajayaghosh, A.; George, S. J. J. Am. Chem. Soc. 2001, 123,
5148–5149. doi:10.1021/ja005933+
Synthetic procedures and analytical data of 1–18.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-253-S1.pdf]
17.Schoonbeek, F. S.; van Esch, J. H.; Wegewijs, B.; Rep, D. B. A.;
de Haas, M. P.; Klapwijk, T. M.; Kellogg, R. M.; Feringa, B. L.
Angew. Chem., Int. Ed. 1999, 38, 1393–1397.
Supporting Information File 2
ITC Measurements.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-253-S2.pdf]
doi:10.1002/(SICI)1521-3773(19990517)38:10<1393::AID-ANIE1393>3
.0.CO;2-H
18.Messmore, B. W.; Hulvat, J. F.; Sone, E. D.; Stupp, S. I.
J. Am. Chem. Soc. 2004, 126, 14452–14458. doi:10.1021/ja049325w
19.Zubarev, E. R.; Pralle, M. U.; Sone, E. D.; Stupp, S. I.
J. Am. Chem. Soc. 2001, 123, 4105–4106. doi:10.1021/ja015653+
20.Hill, J. P.; Jin, W.; Kosaka, A.; Fukushima, T.; Ichihara, H.;
Shimomura, T.; Ito, K.; Hashizume, T.; Ishii, N.; Aida, T. Science 2004,
304, 1481–1483. doi:10.1126/science.1097789
Acknowledgements
We acknowledge the financial support of the Deutsche For-
schungsgemeinschaft (DFG) and the International Research and
Training Group Münster–Nagoya (IRTG 1143). β-CD was
kindly donated by Wacker Chemie AG.
21.Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294,
1684–1688. doi:10.1126/science.1063187
22.Hartgerink, J. D.; Beniash, E.; Stupp, S. I.
References
Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 5133–5138.
doi:10.1073/pnas.072699999
1. Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.;
Koksch, B.; Dernedde, J.; Graf, C.; Knapp, E.-W.; Haag, R.
Angew. Chem., Int. Ed. 2012, 51, 10472–10498.
doi:10.1002/anie.201201114
23.Rajangam, K.; Behanna, H. A.; Hui, M. J.; Han, X.; Hulvat, J. F.;
Lomasney, J. W.; Stupp, S. I. Nano Lett. 2006, 6, 2086–2090.
doi:10.1021/nl0613555
24.Tysseling-Mattiace, V. M.; Sahni, V.; Niece, K. L.; Birch, D.;
Czeisler, C.; Fehlings, M. G.; Stupp, S. I.; Kessler, J. A. J. Neurosci.
2008, 28, 3814–3823. doi:10.1523/JNEUROSCI.0143-08.2008
2. Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int. Ed.
1998, 37, 2754–2794.
doi:10.1002/(SICI)1521-3773(19981102)37:20<2754::AID-ANIE2754>3
.0.CO;2-3
2439

Beilstein J. Org. Chem. 2014, 10, 2428–2440.
25.Spoerke, E. D.; Anthony, S. G.; Stupp, S. I. Adv. Mater. 2009, 21,
425–430. doi:10.1002/adma.200802242
License and Terms
26.Mata, A.; Geng, Y.; Henrikson, K. J.; Aparicio, C.; Stock, S. R.;
Satcher, R. L.; Stupp, S. I. Biomaterials 2010, 31, 6004–6012.
doi:10.1016/j.biomaterials.2010.04.013
This is an Open Access article under the terms of the
Creative Commons Attribution License
27.Shah, R. N.; Shah, N. A.; Del Rosario Lim, M. M.; Hsieh, C.; Nuber, G.;
Stupp, S. I. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 3293–3298.
doi:10.1073/pnas.0906501107
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
28.Harada, A.; Kobayashi, R.; Takashima, Y.; Hashidzume, A.;
Yamaguchi, H. Nat. Chem. 2011, 3, 34–37. doi:10.1038/nchem.893
29.Nakahata, M.; Takashima, Y.; Yamaguchi, H.; Harada, A.
Nat. Commun. 2011, 2, 511. doi:10.1038/ncomms1521
30.Yamaguchi, H.; Kobayashi, Y.; Kobayashi, R.; Takashima, Y.;
Hashidzume, A.; Harada, A. Nat. Commun. 2012, 3, 603.
doi:10.1038/ncomms1617
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
31.Takashima, Y.; Hatanaka, S.; Otsubo, M.; Nakahata, M.; Kakuta, T.;
Hashidzume, A.; Yamaguchi, H.; Harada, A. Nat. Commun. 2012, 3,
1270. doi:10.1038/ncomms2280
doi:10.3762/bjoc.10.253
32.Kakuta, T.; Takashima, Y.; Nakahata, M.; Otsubo, M.; Yamaguchi, H.;
Harada, A. Adv. Mater. 2013, 25, 2849–2853.
doi:10.1002/adma.201205321
33.Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991,
254, 1497–1500. doi:10.1126/science.1962210
34.Ura, Y.; Beierle, J. M.; Leman, L. J.; Orgel, L. E.; Ghadiri, M. R.
Science 2009, 325, 73–77. doi:10.1126/science.1174577
35.Lygina, A. S.; Meyenberg, K.; Jahn, R.; Diederichsen, U.
Angew. Chem., Int. Ed. 2011, 50, 8597–8601.
doi:10.1002/anie.201101951
36.Keren, K.; Krueger, M.; Gilad, R.; Ben-Yoseph, G.; Sivan, U.; Braun, E.
Science 2002, 297, 72–75. doi:10.1126/science.1071247
37.Tashiro, S.; Tominaga, M.; Kawano, M.; Therrien, B.; Ozeki, T.;
Fujita, M. J. Am. Chem. Soc. 2005, 127, 4546–4547.
doi:10.1021/ja044782y
38.Bush, M. E.; Bouley, N. D.; Urbach, A. R. J. Am. Chem. Soc. 2005,
127, 14511–14517. doi:10.1021/ja0548440
39.Heitmann, L. M.; Taylor, A. B.; Hart, P. J.; Urbach, A. R.
J. Am. Chem. Soc. 2006, 128, 12574–12581. doi:10.1021/ja064323s
40.Staunstrup Christensen, H.; Sigurskjold, B. W.; Frihed, T. G.;
Marinescu, L. G.; Pedersen, C. M.; Bols, M. Eur. J. Org. Chem. 2011,
5279–5290. doi:10.1002/ejoc.201100671
41.Rekharsky, M. V.; Inoue, Y. Chem. Rev. 1998, 98, 1875–1918.
doi:10.1021/cr970015o
42.Hunter, C. A.; Anderson, H. L. Angew. Chem., Int. Ed. 2009, 48,
7488–7499. doi:10.1002/anie.200902490
43.Ercolani, G.; Schiaffino, L. Angew. Chem., Int. Ed. 2011, 50,
1762–1768. doi:10.1002/anie.201004201
2440