Evolution of a constitutional dynamic library driven by self-organisation of a helically folded molecular strand

Article abstract of DOI:10.1002/chem.200903437

Conversion of macrocyclic imine entities into helical strands was achieved through three- and four-component exchange reactions within constitutionally dynamic libraries. The generation of sequences of the intrinsic helicity codon, based on the hydrazone-pyrimidine fragment obtained by condensation of pyrimidine dialdehyde A with pyrimidine bis-hydrazine B, shifted the equilibrium between all the possible macrocycles and strands to-wards the full expression (>98%) of helical product [A/B]. Furthermore, it was shown that chain folding accelerated the dynamic exchange reactions among the library members. Lastly, in four-component experiments (involving A, B, E and either C or D), even though the macrocyclic entities ([A/C], [B/E]; [A/D], [B/E]) were the kinetically preferred products, over time dialdehyde A relinquished its initial diamine partners C or D to opt for bishydrazine B, which allowed the preferential formation of the helically folded strand. The present results indicate that self-organisation pressure was able to drive the dynamic system towards the selective generation of the strand undergoing helical folding.

Full text of DOI:10.1002/chem.200903437

FULL PAPER
DOI: 10.1002/chem.200903437
Evolution of a Constitutional Dynamic Library Driven by Self-Organisation
of a Helically Folded Molecular Strand
Luciana Lisa Lao, Jean-Louis Schmitt, and Jean-Marie Lehn*[a]
Abstract: Conversion of macrocyclic
imine entities into helical strands was
achieved through three- and four-com-
ponent exchange reactions within con-
stitutionally dynamic libraries. The gen-
eration of sequences of the intrinsic
helicity codon, based on the hydra-
zone–pyrimidine fragment obtained by
condensation of pyrimidine dialdehyde
A with pyrimidine bis-hydrazine B,
shifted the equilibrium between all the
possible macrocycles and strands to-
wards the full expression (>98%) of
helical product [A/B]. Furthermore, it
was shown that chain folding accelerat-
ed the dynamic exchange reactions
among the library members. Lastly, in
four-component experiments (involving
A, B, E and either C or D), even
though the macrocyclic entities ([A/C],
[B/E]; [A/D], [B/E]) were the kineti-
cally preferred products, over time di-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
hydrazine B, which allowed the prefer-
ential formation of the helically folded
strand. The present results indicate
that self-organisation pressure was able
to drive the dynamic system towards
the selective generation of the strand
undergoing helical folding.
Keywords: constitutionally dynam-
ic chemistry · helical structures ·
imines · macrocycles · self-assembly
Introduction
analyses on the design and control of the folding of molecu-
lar strands have been well documented.[3] Diverse building
blocks as well as synthetic avenues for the formation of heli-
cal polymers or foldamers have been reported.[4] Of particu-
lar interest is a special class of dynamic foldamers, that is,
helical molecular strands that posses reversible covalent
connections.[5–8] In the context of dynamic systems, it has
been shown that 1) helical folding can serve as the driving
force of imine metathesis polymerisation, which elongates
the starting oligomers,[5] 2) dynamic foldamers can respond
to external effectors such as ligand templation by amplifying
the most suitable foldamer,[6a] and participate in cooperative
self-assembly of supramolecular coordination polymers[6b]
and 3) reversible formation of helical foldamers can be con-
trolled by external chemical effectors such as Lewis acids
and metal ions.[7] Such studies have shown the response and
adaptability of dynamic foldamers towards external factors.
The present work is devoted to the investigation of the in-
trinsic ability of a dynamic system capable of generating a
helical molecular strand to select those components in a dy-
namic library that favour the formation of the helical prod-
uct. Such a process represents a component selection driven
by helical folding propensity.
Constitutional dynamic chemistry (CDC)[1] aims at studying
chemical systems composed of entities that are able to un-
dergo continuous variations in their constitution both on the
molecular as well as supramolecular levels through dissocia-
tion into, exchange and recombination of their components
through reversible breaking and formation of covalent or
non-covalent connections, thus generating a whole set of in-
terconverting entities. Such constitutional dynamic libraries
(CDLs) are able to respond to physical stimuli or chemical
effectors by constitutional variations. On the molecular
level, a particularly rich set of reversible covalent connec-
tions is the family of C=N double bonds, which result from
the condensation of an amine with a carbonyl group and in-
clude imine, hydrazone and oxime functions.[2]
On the other hand, the characteristic geometrical features
of helices have been the subject of research investigations in
both biology and chemistry. Numerous reviews and in-depth
[a] Dr. L. L. Lao, Dr. J.-L. Schmitt, Prof. Dr. J.-M. Lehn
ISIS, Universitꢀ de Strasbourg
8 allꢀe Gaspard Monge
67083 Strasbourg (France)
In our earlier work, we have demonstrated that it is possi-
ble to generate well-defined molecular architectures, such as
helically folded strands, by self-assembly from suitable het-
erocyclic components connected through reversible hydra-
Fax : (+33)3-6885-5140
E-mail: lehn@isis.u-strasbg.fr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903437.
Chem. Eur. J. 2010, 16, 4903 – 4910
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4903

zone functions.[7,8] In analogy to 2,2’-bipyridine, the hydra-
zone–pyrimidine (hyz-pym) unit shows marked conforma-
tional preference for a transoid orientation around the con-
necting bond. Consequently, a sequence of such units under-
goes helical self-wrapping and enforces helicity of the mo-
lecular strands.[8a] It forms the basic helicity codon and three
such units constitute one helical turn. Molecular strands
built from hyz-pym sequences may be generated by polycon-
densation of difunctional monomers, such as A and B, yield-
ing helically folded polymers [A/B], as shown in
Scheme 1.[8c]
the dialdehydes A and E with bis-hydrazine and diamines B,
C and D in chloroform at 258C for 24 h. Table 1 summarises
the products obtained. The composition of the CDLs
formed was analysed by high-resolution mass spectrometry
(see the Supporting Information). Only macrocyclic struc-
tures of various sizes were identified in the mixtures formed
from the [A/C], [A/D] and [B/E] combinations, but isolation
was unsuccessful. Linear polymers were obtained for the [C/
E] and [D/E] combinations (see Scheme 2).
The non-covalent interactions of p–p stacking of the aro-
matic rings stabilise the helical conformation of polymer [A/
B]. Similar aromatic stacking, coupled with solvophobic and
hydrogen-bonding interactions, operates in the formation of
other foldamers described in the literature.[3b]
The CDC concept presents a unique avenue for generat-
ing dynamic molecular systems and adaptive materials that
are able to respond to external stimuli/driving forces. In our
laboratory, several examples developed include tempera-
ture,[9] pH,[9] gelation,[10] metal ions[7,11] and electric field[12]
responsive materials.
Herein, we aim to investigate if helical folding through
the formation of a helicity codon can act as a driving force
for component exchange and selection between different
imine entities and hence favour the formation and amplifi-
cation of a helical polymer [A/B] in a library of dynamic
constituents.
Folding-driven formation of a helical polymer through
three-component exchange reactions: Three-component ex-
change reactions were set up according to the equilibria
shown in Scheme 3 to study the evolution of macrocycles
[A/C] and [A/D] towards the formation of the helical poly-
mer [A/B].
Results and Discussion
Table 1. Imine and hydrazone products from condensation of carbonyl (A,E) and amino-containing (B,C,D)
components.[a]
Dynamic libraries of imine and
hydrazone compounds formed
from components A–E: Sets of
dynamic compounds were gen-
erated through reversible con-
densation reactions between
Pyrimidine dialdehyde A
helical polymer [A/B]
Glutaraldehyde E
B
macrocycles [B/E] (2B+2E, 3B+3E,
4B+4E 5B+5E, and 6B+6E)
linear polymer [C/E]
C
D
macrocycles [A/C] (2A+2C and 4A+4C)
macrocycles [A/D] (2A+2D, 3A+3D and 4A+4D)
linear polymer [D/E]
[a] Condensation reactions performed in chloroform at 258C for 24 h.
Scheme 1. Helical polymer [A/B] (right) generated by polycondensation of pyrimidine dialdehyde A with pyrimidine bis-hydrazine B; linear representa-
tion (left). Lateral groups (R=Me, A=Ph, B=3,4,5-(MeO)3C6H2) are omitted for clearer helical representation (right).
4904
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4903 – 4910

Evolution of a Constitutional Dynamic Library
FULL PAPER
demonstrate this exchange re-
action. Initially macrocycles [A/
C] showed sharp, well-defined
signals. However, after the ex-
change reaction occurred, these
distinct signals disappeared and
were replaced by a set of small,
more or less broad signals char-
acteristic of the helical polymer
[A/B]. At the same time, sharp
signals
for
diamine
C
emerged—a sign of release
(>98%, Figure 1) of free di-
AHCTUNGTRENNUNG
1H NMR spectra, see Figure S7
in the Supporting Information).
The formation of the helical
polymer [A/B] was also con-
firmed by ESI-TOF mass spec-
Scheme 2. Schematic representation of imine and hydrazone products listed in Table 1.
trometry, as shown by Figure 2. No macrocycles [A/C] were
detected, while helical strands [A/B] of up to seven repeat-
ing units were formed after 24 h of exchange reaction in
chloroform at 608C.
Scheme 3. Dynamic component exchange reactions leading to the forma-
tion of helical polymer [A/B] from the sets of macrocycles resulting from
the combinations a) [A/C] and b) [A/D] (in CDCl3, at 608C for 24 h).
When the sets of macrocycles [A/C] were subjected to an
equimolar amount of pyrimidine bis-hydrazine B in equili-
brating conditions (608C, 24 h in CDCl3), the macrocycles
[A/C] were destroyed, diamine C was liberated and replaced
by the bis-hydrazine B in favour of the formation of helical
polymer [A/B]. The 1H NMR spectra in Figure 1 clearly
Figure 2. Formation of helical polymer [A/B] from macrocycles [A/C] ac-
cording to Scheme 3a, detected by mass spectrometry ESI-TOF analysis
of the mixture obtained (24 h, 608C, CDCl3).
The reaction mixture was also tested by LC–MS. Howev-
er, analysis of the results was difficult because polymer [A/
B] was destroyed by the acidic aqueous environment of the
LC–MS apparatus. As a result, only smaller fragments of
the polymers could be detected (Figure S4 in the Supporting
Information). Therefore, ESI-TOF mass spectrometry was
chosen for subsequent analyses of the exchange reactions
mixtures.
When macrocycles [A/D] were similarly subjected to an
equimolar amount of pyrimidine bis-hydrazine B, the helical
polymer [A/B] was formed at the expense of diamine D,
which was eventually released into the solution. The reac-
tion was again monitored by 1H NMR spectroscopy and
mass spectrometry (see the Supporting Information).
Figure 1. Dynamic exchange reaction between macrocycles [A/C] and
bis-hydrazine B according to Scheme 3a, monitored by 1H NMR spec-
troscopy (24 h, 608C, CDCl3). The amounts of macrocycles (top) and of
released diamine (second from top) were obtained by integration with re-
spect to the signal of internal reference TMS.
Chem. Eur. J. 2010, 16, 4903 – 4910
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4905

J.-M. Lehn et al.
Table 3. Kinetic evolution of three-component dynamic library towards
Comparison of the equilibrations of foldable and non-folda-
ble molecular strands: All the observations in the previous
section indicate that the library constituents underwent an
exchange of their components, which resulted in the final
generation of the helical polymer [A/B], via the hyz-pym
fragment formed by condensation of the A and B compo-
nents. To further investigate if folding-driven selection of
the “correct” components that allow the formation of a heli-
cal strand is indeed a driving force of this process, two paral-
lel, equivalent studies utilising foldable and non-foldable
compounds were conducted.
the amplification of helical polymer [A/B] constituents.[a]
Reaction
a
t
G
G
ACHTUNGTRENUN[NG A/D], y [%]
(Æ1.5%)
1
19
33
50
237 >98
<2
b
1
4
23
50
77
50
240 >98
<2
Pyridine-2-aldehyde F, the structure of which is similar to
pyrimidine dialdehyde A, but cannot polymerise into a
folded strand, was selected to represent A in the competi-
tive condensation reactions outlined in Table 2.
[a] Equilibration performed in CDCl3, at 258C for up to 250 h. At equi-
librium, helical polymer [A/B] was formed quantitatively (>98%) and
the free diamines C and D were obtained. Due to the complex NMR
spectrum of helical polymer [A/B], its amount was determined by com-
paring the sharp, unique peaks of unreacted B (H5 proton, see Figure S8
in the Supporting Information) and unreacted C (CH2O protons, see
Figure 1) or D (CH2N protons, see Figure S5 in the Supporting Informa-
tion). Over time, the peak of unreacted B decreased while that of un-
reacted C or D increased.
Table 2. Relative amount of hydrazone and imine entities at equilibrium
for the competitive condensation formation of hydrazone [F/B] against
imine [F/C] and imine [F/D] in CDCl3.[a]
The data in Table 3 show that the major kinetic products
obtained at early times were the sets of macrocycles [A/C]
and [A/D], respectively. Over time, these macrocycles
evolved towards formation of the helical polymer [A/B], in-
corporating bis-hydrazine B and liberating diamine C or D
in the process. At equilibrium, helical polymer [A/B]
reached full conversion (>98%) and no macrocycles [A/C]
or [A/D] were present.
Reaction
Hydrazone, x [%] (Æ1.5%)
80 [F/B]
70 [F/B]
Imine, y [%] (Æ1.5%)
20 [F/C]
30 [F/D]
a
b
[a] Mixture of F/B/C in CDCl3 reached equilibrium after 288 h at 258C.
Mixture of F/B/D in CDCl3 reached equilibrium after 240 h at 258C fol-
lowed by 68 h at 608C. At extended times, the amounts of hydrazone and
imine did not change.
In Figure 3 (top), open triangles and squares represent
the evolution of library constituents of a dynamic system
possessing no folding capability as in reaction a described in
Table 2, whereas filled triangles and squares represent reac-
tion a described in Table 3 in which self-folding capability is
present through the (A+B) combination. In a similar fash-
ion, Figure 3 (bottom) compares reaction b in Tables 2 and
3. Thus, Figure 3 displays parallel comparisons of the evolu-
tion of dynamic libraries in the presence and absence of
hyz-pym helicity codon formation from condensations of A
and B.
The intrinsic folding capability of the [A/B] molecular
strand[8] has at least two profound effects: 1) Thermodynam-
ically, it shifted the equilibrium towards the full expression
(>98%, corresponding to >2.3 kcalmolÀ1 free energy dif-
ference) of helical polymer constituent [A/B]. As described
in Table 2, the amount of non-helical hydrazone [F/B] at
equilibrium is 80 or 70% (corresponding to 0.8 or
0.5 kcalmolÀ1 free energy difference) when the competing
diamine present in the library is C or D, respectively, indi-
cating a moderate bias in favour of hydrazone over imine
formation. However, upon the replacement of F by A,
driven by more than three-fold increase in stability, all non-
helical entities of macrocycles [A/C] and [A/D] were de-
stroyed and fully replaced by the self-folding helical poly-
mer [A/B], as shown in Table 3 and Figure 3. 2) Kinetically,
it accelerated the dynamic exchange reactions among the li-
brary members. In Figure 3 (top), helical polymer [A/B]
Following this reaction, aldehyde F, bis-hydrazine B and
diamine C or D in a 2/1/1 ratio were dissolved in CDCl3 and
the mixture was equilibrated. The ratio of the products, x
and y (x+y=100%), was determined by comparing the
area of the NMR signals belonging to each hydrazone and
imine. Each reaction was followed until equilibrium was
reached and no change in product distribution was observed
afterwards. The results at equilibrium are listed in Table 2.
Since reaction b (Table 2) proceeded very slowly at 258C,
the mixture was heated to 608C to accelerate the equilibra-
tion. To confirm that the equilibrium had been reached, it
was also approached by mixing and equilibrating solutions
in which hydrazone [F/B] and imine [F/C] or [F/D] had
been pre-formed. The same product distributions were ob-
tained at equilibrium: hydrazone [F/B] reached 80 and
70%, respectively. These results point to a moderate bias in
favour of hydrazone over imine formation.
Next, another three-component equilibration study in-
volving components that can lead to a folded strand was set
up according to the equilibria shown in Table 3. It is impor-
tant to note that these reactions were designed to be equiva-
lent, differing only in the absence and presence of chain
self-folding in the former and latter, respectively.
Dynamic libraries were generated by mixing equimolar
amounts of A, B and C or D. The changes in the relative
amounts of each constituent, x and y (x+y=100%), were
monitored over time and the results are given in Table 3.
4906
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4903 – 4910

Evolution of a Constitutional Dynamic Library
FULL PAPER
Scheme 4. Dynamic exchange reaction between mixtures of non-folded
entities (see Table 1) leading to the formation of the helical polymer [A/
B] (in CDCl3, at 608C for 72 h).
amount of bis-hydrazine B (10% of the total imine bonds
present), the two macrocycles exchanged their components,
generating two new polymers, namely, the helical polymer
[A/B] and the linear polymers [C/E] or [D/E], not expected
to adopt a helical form. The exchange reactions were moni-
1
tored by H NMR spectroscopy through the disappearance
of the characteristic signals of the existing macrocycles and
appearance of the signals of the new polymers. Figure 4 il-
lustrates the exchange reaction of Scheme 4a.
Figure 3. Comparison of dynamic library evolution in the presence and
absence of hyz-pym helicity codon formation in A/B/C versus F/B/C sys-
tems (top) and A/B/D versus F/B/D systems (bottom) at 258C in CDCl3.
The release of free C follows the same curve as that of formation of [A/
B] (top) and similarly the release of free D follows the same curve as
~
~
&
that of the formation of [A/B] (bottom). : [A/C], : [A/B], : [F/C] and
&: [F/B].
reached 50% in 20 h, while its non-helical counterpart [F/B]
required 50 h to reach the same level. A more pronounced
acceleration is demonstrated in Figure 3 (bottom) through
the comparison of F/B/D and A/B/D dynamic library sys-
tems. It is important to note that F+D imine connection is
slightly favoured over the F+C one and thus the transfor-
mation of the [F/D] towards the [F/B] condensation product
proceeded at a slower pace. From Figure 3 (bottom), it is
shown that to reach 50% [F/B] and 50% [F/D], more than
250 h were required. In comparison, on replacement of F by
A, which makes the generation of the hyz-pym helicity
codon possible, 50% helical polymer [A/B] was readily ob-
tained in 50 h.
Figure 4. Dynamic exchange reaction according to Scheme 4a, recorded
by 1H NMR spectroscopy after 5 min and 72 h equilibration in CDCl3 at
608C; bottom trace: helical polymer formed from A+B (see text for de-
scription).
At early times (5 min), the NMR spectrum showed the
combined signals of macrocycles [A/C] and [B/E]. There-
after, component exchange occurred and the initially sharp,
characteristic signals of macrocycles [A/C] and [B/E] disap-
peared, indicating that these entities were broken down. At
the same time, the release of diamine C and dialdehyde E
and their reversible recombination into polymer [C/E] were
observed along with the formation of helical polymer [A/B],
shown by the appearance of many small, broad polymer
peaks. The total amount of diamine C, present in polymer
[C/E] and free in solution, was quantified with respect to an
internal standard (TMS) and was found to be >98%. The
same attempt to quantify dialdehyde E was unsuccessful as
its peak overlaps with that of helical polymer [A/B] (see
also Figure S9 in the Supporting Information). At the end
Folding-driven formation of helical polymer through four-
component exchange reaction: In the next phase, we investi-
gated the behaviour of mixtures of macrocyclic compounds
(see Table 1) not containing the helicity codon and therefore
not expected to be folded. The component exchange and
the resulting products were analysed according to the equili-
bria shown in Scheme 4.
When macrocycles [A/C] and [B/E] were mixed and
heated to 608C for 72 h in the presence of a catalytic
Chem. Eur. J. 2010, 16, 4903 – 4910
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4907

J.-M. Lehn et al.
(about 72 h), the solution mixture contained only the helical
polymer [A/B] and linear polymer [C/E].
This result was further confirmed by mass spectrometry
analysis, as shown in Figure 5. Both macrocycles [A/C] and
[B/E] were no longer present. Instead, helical polymers [A/
The four-component dynamic library formed according to
the equilibration reactions shown in Table 4 gave rise to the
four possible constituents, the relative amounts of which
changed with time. The absence of distinct peaks for both
helical and linear polymers made the quantification of the
product constituents difficult. Therefore, changes in the rela-
tive composition of the products could only be monitored
1
qualitatively by using H NMR spectroscopy and identified
by mass spectrometry. The results are summarised in
Table 4.
At the beginning, when the four components were free to
choose their condensation partners, diamines C and D were
quick to combine with dialdehyde A to form macrocycles
[A/C] and [A/D], respectively, while dialdehyde E combined
with bis-hydrazine
B to form macrocycles [B/E] (see
Table 1). For both reactions (Table 4), the change in the
amounts of the library constituents was slow at 258C. How-
ever, upon heating to 608C for about 70 h, the amounts of
the constituents were reversed, displaying a decrease/disap-
pearance of macrocycles [A/C], [A/D] and [B/E] and a
prominent increase in the formation of helical polymers [A/
B] and linear polymers [C/E] and [D/E].
Figure 5. Formation of helical polymer [A/B] and linear polymer [C/E]
from macrocycles [A/C] and [B/E] according to Scheme 4a, detected by
mass spectrometry ESI-TOF analysis of the mixture obtained after equili-
bration (72 h, 608C, CDCl3). In a related study, NMR analysis shows that
[B/E] combination decreased over time and was eventually replaced by
[A/B] (see Figure S9 in the Supporting Information).
In summary, the four-component dynamic libraries above
would first generate the kinetic products comprising macro-
cycles [A/C] or [A/D] and [B/E] as the intermediates. After
some time, the constitutional dynamic library evolved and
rearranged its constituents through component exchange to
yield the sets of final products, the helical [A/B] and the
linear [C/E] or [D/E] polymers.
The relationship among the library members described by
reaction a in Table 4 can be described by the network dia-
gram, shown in Figure 6. A similar diagram can also be con-
structed for the library members of reaction b in Table 4.
The four constituents of the dynamic library occupy the cor-
ners of a square graph. Related constituents, sharing one of
their components, share an edge and are antagonists, that is,
an increase in one leads to a decrease in the other. On the
other hand, unrelated constituents, sharing no common com-
ponents, are located on the diagonals and behave as ago-
nists, that is, they both increase together. In Figure 6, it is an
B] of different numbers of repeating units as well as a linear
chain [C/E] of up to 18 repeating units were detected.
The same study performed following the equilibrium
shown in Scheme 4b gave similar results. Macrocycles [A/D]
and [B/E] exchanged their components to yield helical poly-
mer [A/B] and linear polymer [D/E]. In both scenarios, it is
important to note that the driving force for the helical poly-
mer formation is strong enough to achieve 1) breaking of
macrocycles [A/C] or [A/D] and subsequent exchange of di-
amine C or D for the bis-hydrazine B and 2) breaking of
macrocycles [B/E] to supply bis-hydrazine B to the system
and hence allow for the formation of helical polymer [A/B].
Next, four-component kinetic equilibration was performed
to find out if the system would directly select the suitable
pair of dialdehyde and bis-hydrazine/diamine that gives the
most stable constituent or if some intermediate constituents
were formed prior to the final products. This experiment
was carried out as outlined in Table 4.
Table 4. Kinetic evolution of four-component dynamic library towards
the amplification of helical polymer [A/B] constituent and its agonists,
linear polymers [C/E] and [D/E].[a]
Reaction T [8C] t [h]
[A/C]
E
E
[A/B]
G
ACHTUNGTNER[NUNG D/E]
a
25
25
60
25
25
60
3
70
64
3
77
74
high
high
low
high
high
low
high
high
low
low
low
high
low
low
high
low
low
high
Figure 6. Connectivities between the constituents in the square network
of the four-membered dynamic library described in Table 4, representing
agonistically (diagonals) and antagonistically (edges) related constituents.
The thicker diagonal indicates the preferential expression/upregulation of
the agonistic constituents [A/B] and [C/E].
b
high
high
low
low
low
high
[a] Equilibration performed in CDCl3.
4908
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4903 – 4910

Evolution of a Constitutional Dynamic Library
FULL PAPER
internal stimulus, the stability of the helically folded struc-
ture, that drives the constitutional dynamic library towards
the expression/upregulation of the helical polymer [A/B]
and its agonistic partner [C/E].
7) Most significantly, the present results may in principle be
extendable to biopolymers (biodynamers), whereby for
instance the formation of a particular folded strand
would induce the selection of the appropriate compo-
nents. Such effects may be considered as potentially op-
erative in prebiotic evolution, driving the system by
either internal (present case) or external[10a] self-organi-
sation.
Conclusion
Three- and four-component constitutionally dynamic sys-
tems were generated through reversible imine and hydra-
zone bond formation with the aim of investigating the con-
tribution provided by the helical folding ability of a molecu-
lar strand to the selection of its components. Based on the
results and the discussions above, the following conclusions
can be drawn:
Experimental Section
General: All reagents and solvents were purchased at the highest com-
mercial quality and used without further purification unless stated other-
wise. Pyrimidine dialdehyde A and pyrimidine bis-hydrazine B were syn-
thesised according to a literature procedure.[8b] Diamine (N2O) C was ob-
tained from TCI Europe, whereas diamine (N2C4) D, glutaraldehyde E
(50% in H2O) and pyridine-2-aldehyde F were obtained from Sigma–Al-
drich. Deuterated chloroform used for all reaction schemes was flashed
through basic alumina immediately prior to use to remove any trace of
acid. 1H NMR spectra were recorded on a Bruker Avance 400 spectrome-
ter at 400 MHz. The spectra were internally referenced to the residual
proton solvent signal. Electrospray (ESI-TOF) analyses were performed
by Service de Spetromꢀtrie de Masse de lꢂInstitut de Chimie, Universitꢀ
de Strasbourg, on a Bruker MicroTOF mass spectrometer. LC–MS was
performed by coupling a Thermo Finnigan Accela chromatography appa-
ratus and a Thermo Finnigan Surveyor MSQ+ mass spectrometer.
1) The condensation reactions in Table 2 show that in com-
petition with imines [F/C] and [F/D], the amount of non-
foldable hydrazone [F/B] ranged from 70 to 80% at
equilibrium—thermodynamic distribution in the absence
of chain folding.
2) On replacement of F by A, which allows the generation
of the hyz-pym helicity codon, imine macrocycles formed
by [A+C] and [A+D] converted fully to helical strand
[A/B] (Table 3). In other words, the amount of foldable
hydrazone [A/B] reached completion (>98%) at equilib-
rium. This finding indicates that the generation of a heli-
cally folded strand by helicity codon formation drove the
General procedure for imine/hydrazone condensation reactions: Equimo-
lar amounts of dialdehydes (A or E) and diamines or bis-hydrazine (B, C
or D) were dissolved in fresh CDCl3 at a concentration of 10 mm. The so-
lution was stirred overnight at room temperature prior to use in other ex-
periments.
À
NH2 (amine or hydrazine) selection and finally shifted
the equilibrium towards full expression of helical poly-
mer [A/B]—thermodynamic shift due to chain folding.
3) The capability to form a helical molecular strand [A/B]
accelerated the dynamic exchange reactions among the
library members. Comparing reaction b in Table 3 with
Table 2, expressing 50% helical constituent product [A/
B] took approximately 50 h, whereas expressing 50%
non-helical product [F/B] under the same conditions re-
quired more than 250 h—exchange rate acceleration due
to chain folding.
4) Helical polymer [A/B] is not the preferred kinetic prod-
uct. When A, B and D were freely mixed, pyrimidine di-
aldehyde A first reacted preferentially with diamine D,
forming macrocycles [A/D] as kinetic products—kinetic
selection. Thereafter, the dynamic library evolved to-
wards the expression of helix [A/B]. A similar situation
was encountered with other combinations and also in
four-component systems.
Procedure for three-component exchange reactions: Macrocycles [A/C]
or [A/D] were prepared according to the above procedure in an NMR
tube (0.6 mL, 10 mm). An equimolar amount of pyrimidine bis-hydrazine
B was added to the mixture and the solution was heated to 608C for
24 h. The NMR tubes were topped with Teflon caps to keep a constant
concentration upon heating. Exchange reactions were monitored by
NMR spectroscopy and LC–MS or ESI-TOF.
Procedure for four-component exchange reactions: Macrocycles [A/C],
[A/D] and [B/E] were individually prepared. Equimolar amounts of two
different macrocycles were mixed and a catalytic amount of pyrimidine
bis-hydrazine B (equivalent to 10% of the total imine bonds present)
was added. An aliquot of the mixture was transferred to an NMR tube
(topped with a Teflon cap) and heated to 608C for 72 h. Exchange reac-
tions were monitored by NMR spectroscopy and LC–MS or ESI-TOF.
Procedure for equilibration experiments and determination of kinetic
data: A typical protocol is realised by the preparation of a fresh solution
of the desired compounds in CDCl3, allowed to equilibrate in an NMR
tube (0.6 mL, 10 mm). The first NMR spectrum was usually recorded
3 min after mixing all of the compounds with spinning the tube at room
temperature. Subsequently, NMR spectra were recorded at pre-deter-
mined time points. All NMR tubes were topped with Teflon caps to keep
a constant concentration throughout measurement.
5) In more general terms, it is the intrinsic property of the
system, the formation of a helically folded strand, that
provides a driving force for the adaptation/evolution of
the CDL network, leading to the selection and amplifica-
tion/upregulation of the most stable library member of a
constitutionally dynamic system.
Acknowledgements
6) As an effect of an internal factor, point 5) complements
the numerous cases in which a constitutional dynamic
system is driven by external factors, either physical stim-
uli (such as temperature[9]) or chemical effectors.[1,2,9,11]
We thank Prof. Freddy Boey Yin Chiang, the School of Materials Science
and Engineering, Nanyang Technological University, Singapore for his
support and the Economic Development Board, EDB, Singapore for a
post-doctoral fellowship (L.L.L.).
Chem. Eur. J. 2010, 16, 4903 – 4910
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4909

J.-M. Lehn et al.
[6] a) T. Nishinaga, A. Tanatani, K. Oh, J. S. Moore, J. Am. Chem. Soc.
2002, 124, 5934–5935; b) J. W. Wackerly, J. S. Moore, Macromole-
cules 2006, 39, 7269–7276.
[7] a) N. Giuseppone, J.-L. Schmitt, J.-M. Lehn, J. Am. Chem. Soc. 2006,
128, 16748–16763; b) N. Giuseppone, J.-L. Schmitt, J.-M. Lehn,
Angew. Chem. 2004, 116, 5010–5014; Angew. Chem. Int. Ed. 2004,
43, 4902–4906.
[8] a) K. M. Gardinier, R. G. Khoury, J.-M. Lehn, Chem. Eur. J. 2000, 6,
4124–4131; b) J.-L. Schmitt, A.-M. Stadler, N. Kyritsakas, J.-M.
Lehn, Helv. Chim. Acta 2003, 86, 1598–1624; c) J.-L. Schmitt, J.-M.
Lehn, Helv. Chim. Acta 2003, 86, 3417–3426.
[9] N. Giuseppone, J.-M. Lehn, Chem. Eur. J. 2006, 12, 1715–1722.
[10] a) N. Sreenivasachary, J.-M. Lehn, Proc. Natl. Acad. Sci. USA 2005,
102, 5938–5943; b) N. Sreenivasachary, J.-M. Lehn, Chem. Asian J.
2008, 3, 134–139.
[1] J.-M. Lehn, Chem. Soc. Rev. 2007, 36, 151–160.
[2] a) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor,
J. K. M. Sanders, S. Otto, Chem. Rev. 2006, 106, 3652–3711; b) J.-M.
Lehn, Chem. Eur. J. 1999, 5, 2455–2463.
[3] a) S. Hecht, I. Huc, Foldamers: Structure, Properties, and Applica-
tions, Wiley-VCH, Weinheim, 2007; b) D. J. Hill, M. J. Mio, R. B.
Prince, T. S. Hughes, J. S. Moore, Chem. Rev. 2001, 101, 3893–4011;
c) S. H. Gellman, Acc. Chem. Res. 1998, 31, 173–180; d) A. E.
Rowan, R. J. M. Nolte, Angew. Chem. 1998, 110, 65–71; Angew.
Chem. Int. Ed. 1998, 37, 63–68; e) T. J. Katz, Angew. Chem. 2000,
112, 1997–1999; Angew. Chem. Int. Ed. 2000, 39, 1921–1923; f) K.
Maeda, E. Yashima, Macromolecules 2008, 41, 3–12; g) W. Wang,
A. D. Shaller, A. D. Q. Li, J. Am. Chem. Soc. 2008, 130, 8271–8279.
[4] a) D. Sanchez-Garcia, B. Kauffmann, T. Kawanami, H. Ihara, M.-H.
Delville, I. Huc, J. Am. Chem. Soc. 2009, 131, 8642–8648; b) D.
Zhao, J. S. Moore, Org. Biomol. Chem. 2003, 1, 3471–3491; c) D.
Zhao, J. S. Moore, J. Am. Chem. Soc. 2003, 125, 16294–16299; d) P.
Jonkheijm, P. van der Schoot, A. P. H. J. Schenning, E. W. Meijer,
Science 2006, 313, 80–83; e) J. Q. Nguyen, B. L. Iverson, J. Am.
Chem. Soc. 1999, 121, 2639–2640; f) M. Ohkita, J.-M. Lehn, G.
Baum, D. Fenske, Chem. Eur. J. 1999, 5, 3471–3481; g) V. Berl, I.
Huc, R. G. Khoury, J.-M. Lehn, Chem. Eur. J. 2001, 7, 2798–2809.
[5] a) K. Oh, K.-S. Jeong, J. S. Moore, Nature 2001, 414, 889–893; b) D.
Zhao, J. S. Moore, J. Am. Chem. Soc. 2002, 124, 9996–9997.
[11] S. Fujii, J.-M. Lehn, Angew. Chem. 2009, 121, 7771–7774; Angew.
Chem. Int. Ed. 2009, 48, 7635–7638.
[12] a) N. Giuseppone, J.-M. Lehn, Angew. Chem. 2006, 118, 4735–4740;
Angew. Chem. Int. Ed. 2006, 45, 4619–4624; b) A. Herrmann, N.
Giuseppone, J.-M. Lehn, Chem. Eur. J. 2009, 15, 117–124.
Received: December 15, 2009
Published online: March 22, 2010
4910
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4903 – 4910