Supramolecular catalysis by recognition-encoded oligomers: Discovery of a synthetic imine polymerase

Article abstract of DOI:10.1039/d0sc02234a

All key chemical transformations in biology are catalysed by linear oligomers. Catalytic properties could be programmed into synthetic oligomers in the same way as they are programmed into proteins, and an example of the discovery of emergent catalytic properties in a synthetic oligomer is reported. Dynamic combinatorial chemistry experiments designed to study the templating of a recognition-encoded oligomer by the complementary sequence have uncovered an unexpected imine polymerase activity. Libraries of equilibrating imines were formed by coupling diamine linkers with monomer building blocks composed of dialdehydes functionalised with either a trifluoromethyl phenol (D) or phosphine oxide (A) H-bond recognition unit. However, addition of the AAA trimer to a mixture of the phenol dialdehyde and the diamine linker did not template the formation of the DDD oligo-imine. Instead, AAA was found to be a catalyst, leading to rapid formation of long oligomers of D. AAA catalysed a number of different imine formation reactions, but a complementary phenol recognition group on the aldehyde reaction partner is an essential requirement. Competitive inhibition by an unreactive phenol confirmed the role of H-bonding in substrate recognition. AAA accelerates the rate of imine formation in toluene by a factor of 20. The kinetic parameters for this enzyme-like catalysis are estimated as 1 × 10-3 s-1 for kcat and the dissociation constant for substrate binding is 300 μM. The corresponding DDD trimer was found to catalyse oligomerisation the phosphine oxide dialdehyde with the diamine linker, suggesting an important role for the backbone in catalysis. This unexpected imine polymerase activity in a duplex-forming synthetic oligomer suggests that there are many interesting processes to be discovered in the chemistry of synthetic recognition-encoded oligomers that will parallel those found in natural biopolymers. This journal is

Full text of DOI:10.1039/d0sc02234a

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Supramolecular catalysis by recognition-encoded
oligomers: discovery of a synthetic imine
polymerase†
Cite this: DOI: 10.1039/d0sc02234a
All publication charges for this article
have been paid for by the Royal Society
of Chemistry
ab
a
*
Luca Gabrielli
and Christopher A. Hunter
All key chemical transformations in biology are catalysed by linear oligomers. Catalytic properties could be
programmed into synthetic oligomers in the same way as they are programmed into proteins, and an
example of the discovery of emergent catalytic properties in a synthetic oligomer is reported. Dynamic
combinatorial chemistry experiments designed to study the templating of a recognition-encoded
oligomer by the complementary sequence have uncovered an unexpected imine polymerase activity.
Libraries of equilibrating imines were formed by coupling diamine linkers with monomer building blocks
composed of dialdehydes functionalised with either a trifluoromethyl phenol (D) or phosphine oxide (A)
H-bond recognition unit. However, addition of the AAA trimer to a mixture of the phenol dialdehyde and
the diamine linker did not template the formation of the DDD oligo-imine. Instead, AAA was found to be
a catalyst, leading to rapid formation of long oligomers of D. AAA catalysed a number of different imine
formation reactions, but a complementary phenol recognition group on the aldehyde reaction partner is
an essential requirement. Competitive inhibition by an unreactive phenol confirmed the role of H-
bonding in substrate recognition. AAA accelerates the rate of imine formation in toluene by a factor of
20. The kinetic parameters for this enzyme-like catalysis are estimated as 1 ꢀ 10ꢁ3 sꢁ1 for kcat and the
dissociation constant for substrate binding is 300 mM. The corresponding DDD trimer was found to
catalyse oligomerisation the phosphine oxide dialdehyde with the diamine linker, suggesting an
important role for the backbone in catalysis. This unexpected imine polymerase activity in a duplex-
forming synthetic oligomer suggests that there are many interesting processes to be discovered in the
chemistry of synthetic recognition-encoded oligomers that will parallel those found in natural biopolymers.
Received 20th April 2020
Accepted 25th June 2020
DOI: 10.1039/d0sc02234a
rsc.li/chemical-science
Introduction
Recognition-encoded oligomers are the basis for the key
chemical processes that lead to living systems: molecular
replication, self-assembly, molecular recognition and supra-
molecular catalysis.1 The ability to process molecular informa-
tion encoded as a sequence of monomer building blocks
through replication, transcription and translation is currently
unique to nucleic acids. Iterative cycles of selection and repli-
cation have been used in molecular evolution experiments to
select functional nucleic acid sequences with recognition or
catalytic properties from large pools of unbiased sequences.2,3
Synthetic recognition-encoded oligomers have the potential to
display similar properties, which would open the way for
directed evolution of function in synthetic polymers. The rst
Fig. 1 Chemical structure of the AAA$DDD duplex. The comple-
mentary recognition modules are phosphine oxide H-bond acceptors
(red) and 2-trifluoromethyl phenol H-bond donors. Synthesis is based
on reductive amination of bisaniline linkers (green) with dialdehydes
monomer building blocks. The backbone modules (black) provide
a geometry compatible with duplex formation. The 2-ethylhexyl
groups ensure solubility in non-polar solvents.
aDepartment of Chemistry, University of Cambridge, Lenseld Road, Cambridge CB2
1EW, UK. E-mail: herchelsmith.orgchem@ch.cam.ac.uk
bDepartment of Chemistry, University of Padova, via F. Marzolo 1, Padova 35131, Italy
† Electronic supplementary information (ESI) available: Experimental procedures,
HPLC-MS experiments and NMR spectroscopy studies, including 1H-NMR kinetic
experiments and 2D-NMR spectra. See DOI: 10.1039/d0sc02234a
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
steps in this direction have been taken with the development of oligomers is that the coupling chemistry can be carried out
oligomers that form duplexes in a sequence-selective manner, sequentially in two steps: imine formation followed by reduc-
and a range of such systems have been reported.4–9
tion. Imine formation from an aniline and an aldehyde can be
We have recently reported a modular approach for the design carried out under reversible conditions. Thus mixing the bisa-
of duplex forming molecules.6 The basic architecture is illus- niline linker with the two dialdehyde building blocks used to
trated in Fig. 1.
assemble the oligomers in Fig. 1 will lead to a dynamic
There are four key elements to the design: the recognition combinatorial library (DCL) of equilibrating recognition-
modules that lead to duplex assembly are H-bond acceptors encoded oligo-imines.14 Addition of a recognition-encoded
(red) and H-bond donors (blue) displayed as side chains; the amine template, such as AAA shown in Fig. 1, should cause
synthesis of oligomers is based on reductive amination to couple a shi in the DCL equilibrium in favour of the complementary
bisaniline linkers (green) with dialdehyde monomer building DDD oligomer, which would form the most stable duplex with
blocks; the backbone module (black) was chosen to ensure that the template (Fig. 2).15 The DCL could then be trapped by
the oligomer has geometrical properties that are compatible reduction to obtain the complementary copy of the template as
with duplex formation; and the solubilising groups ensure solu- an oligo-amine. Here we describe experiments designed to
bility in the non-polar solvents required for high affinity H- explore non-covalent template directed synthesis of
bonding interactions.
recognition-encoded oligomers based on the duplex shown in
The homo-oligomers illustrated in Fig. 1 form duplexes in Fig. 1. However, rather than acting as a template for selective
toluene and in chloroform solution.6j The association constant formation of DDD, it turns out that AAA is a catalyst for selective
for the intermolecular H-bond between the phenol and phos- elongation of homo-oligomers of H-bond donor monomers.16
phine oxide units used for base-pairing is 3 ꢀ 103 Mꢁ1 in
toluene, which leads to high stability, high delity duplex
formation between complementary oligomers. There is
Results and discussion
a uniform increase in duplex stability with oligomer length
indicating that this supramolecular framework is compatible
with cooperative formation of H-bonds along an extended
duplex. The duplex architecture is the basis for biological
copying of nucleic acid sequence information through
template-directed synthesis. Non-enzymatic DNA template-
directed synthesis of non-natural sequence-dened oligomers
has been performed using amide coupling,10 and reductive
amination,11,12 with further extension to the assembly of
sequence dened polymers.13
The building blocks used for oligomer synthesis are illustrated
in Fig. 3. D and D0 are dialdehyde and mono-aldehyde recog-
nition units equipped with 2-triuoromethylphenol H-bond
donor groups. A and A0 are the complementary dialdehyde
and mono-aldehyde H-bond acceptor recognition units equip-
ped with phosphine oxide groups. N is the bisaniline linker
used for the dynamic covalent assembly of oligomers, as illus-
trated in Fig. 4, and N0 is a monofunctional aniline chain end
capping group. Mixing the linker N with either or both of the
dialdehydes (D, A) should lead to formation of a library of
interconverting oligomers (Fig. 4).
The effective molarity for the intramolecular interactions
leading to duplex formation was previously measured as 30 mM
in toluene.6j Working at a monomer concentration of 10 mM
should therefore favour oligomerisation in the presence of
a template. At this concentration, more than 80% of the
monomers will be bound to the template, and the AAA$DDD
duplex will be more than 99% bound.17 Even though the aniline
The oligomers shown in Fig. 1 should be able to template
their own synthesis in the same way. An attractive feature of the
reductive amination chemistry used to synthesise these
Fig.
2 Non-covalent template-directed synthesis of recognition-
encoded oligomers. The template is an amine (green) oligomer
bearing recognition units (blue and red). Dynamic imine libraries can
be built from dialdehydes and diamines equipped with recognition Fig. 3 Structures of the acceptor and donor dialdehyde (D, A) and
units. The template shifts the equilibrium, favouring the formation of dianiline (N) monomers used for building libraries of imine oligomers.
the complementary sequence, which can be trapped as an amine Structures of monoaldehyde (D0, A0) and monoaniline (N0) monomers
oligomer.
used as chain end capping units (R ¼ 2-ethylhexyl, R0 ¼ pentyl).
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
Fig. 5 HPLC traces of a toluene-d8 solution of donor D (10 mM) 5
minutes after the addition of dianiline N (16 mM) in the presence of
phosphine oxides. (a) 10 mM nBu3PO. The peak corresponding to the
monomeric mono-imine mono-aldehyde is labelled D0. (b) 3.3 mM
AAA. The imine oligomers are labelled D, DD, DDD, D4, D5 etc. For
each oligomer, three peaks were observed, corresponding to dia-
ldehyde (traces), mono-imine mono-aldehyde and diimine terminal
groups. HPLC method: coagent phenyl hydride 2.o 3 cm ꢀ 3 mm
column, flow rate 0.4 ml minꢁ1, injection volume 0.1 mL, 45 ꢂC; eluent
A 60% NH4OAc 10 mM, pH 5.70, 40% THF; eluent B 85% THF, 10% IPA,
5% A; 45% B for 1 min, then increased to 75% over 3.5 min, and then
increased to 80% over 2 min. The UV/vis absorbance was recorded at
292–308 nm.
Fig. 4 Equilibria involved in the formation of a library of imine oligo-
mers. If no cyclic oligomers are formed, the degree of oligomerization
can be controlled with the amount of dianiline. X is a recognition
module, and R is a solubilizing group.
oligomers with aldehyde end groups were observed, conrming
that there is no imine hydrolysis on the HPLC column (see Fig.
S79†). The HPLC peaks were assigned using the corresponding
mass spectra. The HPLC results in Fig. 5 show that there is very
little imine formation aer 5 minutes in the absence of
template. Addition of the AAA leads to a much faster reaction
and the formation a complex mixture of different length linear
oligomers. However, there is no obvious template effect, and the
yield of DDD does not appear to be enhanced relative to the
other oligomers. When the control reaction without template
was le to equilibrate for 13 hours, the resulting HPLC trace was
similar to that obtained in the presence of AAA (see Fig. S80†).
In neither case were any cyclic oligomers detected. Thus AAA
appears to be a catalyst for imine formation rather than
a template.
linkers were equipped with solubilising groups, mixing D and N
in toluene in the absence of a template lead to precipitation.
The presence of one equivalent of tri-n-butyl phosphine oxide
was sufficient to keep everything in solution, and so the
untemplated control experiment was carried out under these
conditions. However, mixing D and N in toluene in the presence
of AAA lead to rapid precipitation. These initial observations
already indicate that AAA has a signicant effect on the reaction.
Based on the assumption that the precipitate was due to
formation of polymers, the reaction was tested using excess N in
order to limit the chain length. More than 1.6 equivalents of N
was sufficient to keep everything in solution in the presence of
AAA. The results are shown in Fig. 5.
Imine formation was monitored using HPLC-MS. Imine-
based DCLs are typically xed for HPLC analysis by reduction
to the corresponding amines.18 However, the reagents required
could interfere with H-bonding interactions with the AAA
template and alter the oligomer population in this experiment,
so direct chromatographic analysis of the imine library is
preferable. It was possible to minimise imine hydrolysis on the
HPLC column by using a mixture of THF and ammonium
acetate buffer (pH 5.7) as the eluent, allowing direct analysis of
the imine DCL. The HPLC results shown in Fig. 5 show peaks
due to oligomers with both aldehyde and imine end groups.
However, these HPLC traces were recorded aer 5 minutes, and
when the HPLC experiment was repeated aer an hour, no
In order to assess whether H-bonding interactions between
the phenol groups on the monomer units and the phosphine
oxide recognition groups on AAA play an important role the
catalytic activity, we investigated the reaction between an
aniline and an aldehyde that have no H-bonding sites (N0 and P
1
in Fig. 6a). H-NMR spectroscopy was used to monitor the rate
of appearance of the imine signal at 9 ppm due to the product in
the presence of various additives: nBu3PO, AA or AAA. In
comparison with the control experiment, none of the phos-
phine oxide additives have any signicant effect on the reaction
rate. This observation suggests that H-bonding interactions
between the D monomer units and AAA play an important role
in the catalysis of imine formation.
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 6 Effects of phosphine oxide oligomers on the rate of imine
formation for substrates with no recognition sites. (a) Reaction of 4-
pentylbenzaldehyde with 4-pentylaniline (R0 ¼ n-pentyl). (b) Half-life
of the aldehyde starting material measured by integration of 500 MHz
1H NMR signals as a function of time. Reaction conditions: 20 mM
aldehyde, 40 mM aniline and 10 mM phosphine oxide in toluene-d8
(i.e. 10 mM nBu3PO, 5 mM AA, or 3.3 mM AAA). (c) Chemical structure
of AA.
Fig. 8 (a) Kinetics of imine formation measured by integration of all
imine signals in the 500 MHz 1H NMR spectra of a mixture of N (20 mM)
and D (10 mM) in toluene-d8 in the presence of different amounts of
AAA: 0 (red), 0.83 (yellow), 1.65 (blue) or 3.3 (green) mM. The dotted
lines show the linear region where the initial rate (Vinit) was measured.
(b) Relationship between the initial rate (Vinit) and the concentration of
AAA. The straight line of best fit is shown.
Next we investigated whether a trimer is required for catal-
ysis. Fig. 7 compares the rate of reaction of D with N in the
presence of a monomeric phosphine oxide (nBu3PO), AA or AAA.
Although this reaction generates many different oligomers (see
Fig. 5), it was possible to measure the total amount of imine
formed by integrating all of the signals between 8.9 and 9.1 ppm
rst order in AAA, conrming that this oligomer is the active
species. These data also show that there is efficient turnover of
the catalyst. In all of the reactions in Fig. 8b, AAA is present in
substoichiometric amounts (8–33 mol%), but it converts 100%
of the substrate into product.
1
in the H NMR spectrum. In the presence of AA, the reaction
The oligomerisation of N and D is a multi-step process, and
the solubility of the oligomeric products limits the substrate
concentration range that can be studied, so we have not been
able to demonstrate saturation kinetics with this system.
Therefore in order to conrm the role played by substrate
recognition in the catalytic activity of AAA, 2-triuoromethyl
phenol was added as a competitive inhibitor. The results of 1H-
NMR kinetic experiments are shown in Fig. 9. The rate of
reaction between D and N is signicantly slower in the presence
of the inhibitor: Vinit drops from 5.2 to 3.6 mM sꢁ1 in the pres-
ence of one equivalent of 2-triuoromethyl phenol. Note that
proceeds at a similar rate to the control experiment with
nBu3PO. In the presence of AAA, the rate of reaction is about one
order of magnitude higher, consistent with the HPLC results in
Fig. 5. This observation indicates that the trimer is the
minimum length oligomer required for catalysis.
The order of reaction with respect to AAA was determined by
measuring the rate as a function of catalyst concentration using
1H-NMR spectroscopy. Fig. 8a shows that the rate of imine
formation increases with increasing amounts of AAA. The initial
rates Vinit are plotted as a function of AAA concentration in
Fig. 8b, and the linear relationship shows that the reaction is
Fig. 9 Kinetics of imine formation measured by integration of all imine
Fig. 7 Kinetics of imine formation measured by integration of all imine signals in the 500 MHz 1H NMR spectra of a mixture of N (20 mM) and
signals in the 500 MHz 1H NMR spectra of a mixture of N (20 mM) and D (10 mM) in toluene-d8 in the presence of 3.3 mM AAA in the absence
D (10 mM) in toluene-d8 in the presence of 10 mM nBu3PO (red data), (filled circles) or presence of 10 mM 2-trifluoromethyl phenol (open
5 mM AA (yellow data) or 3.3 mM AAA (green data).
circles).
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
the addition of 2-triuoromethyl phenol was observed to cause
a small increase in the rate of imine formation in the reaction of
P with N0 (see Fig. S52†), which suggests that the observed
inhibition is an underestimate of the true effect of competitive
substrate binding. These results conrm that H-bond interac-
tions between AAA and D play an important role in the catalytic
process.
The data in Fig. 8 can therefore be used to estimate the
kinetic parameters that describe the enzyme-like properties of
AAA. The dissociation constant Kd for the H-bonding interac-
tion between the dialdehyde substrate D and the phosphine
oxide groups on AAA was previously measured as 300 mM in
toluene. The reactions in Fig. 8b were all carried out at
a substrate concentration of 10 mM and with excess N, so if we
assume that the value of KM is similar to the value of Kd reported
above, the catalyst AAA must be saturated under the reaction
conditions. Thus the slope of Fig. 8b (1.3 ꢀ 10ꢁ3 sꢁ1) is equiv-
alent to the value of kcat, and the value of the corresponding
pseudo rst order rate constant for the uncatalyzed reaction
kuncat is the ratio of the intercept of Fig. 8b and the substrate
concentration (6.2 ꢀ 10ꢁ5 sꢁ1). Thus we can estimate the rate
acceleration for the catalysed reaction kcat/kuncat as approxi-
mately 20.
Fig. 11 a) Half-life for reactions in the presence of DD (5 mM, shaded
data) or DDD (3.3 mM, black data) relative to the corresponding half-
life in the presence of 2-trifluoromethyl phenol (10 mM) (t1/2 (rel)).
Half-lives were measured by integration of the 500 MHz 1H NMR signal
due to the aldehyde as a function of time for mixtures of 20 mM
aldehyde (P or A0) and 40 mM total aniline (40 mM N0, or 20 mM N) in
toluene-d8. (b) Chemical structure of DD.
1
in toluene to measure reaction rates using H-NMR spectros-
Next we investigated the effect of AAA on systems that do not
form polymers by using the mono-aldehyde D0 in place of D and
the mono-aniline N0 in place of N (see Fig. 3). The results are
shown in Fig. 10. For all four combinations of substrate, the
catalytic effect of AAA on imine formation is similar. This result
shows that formation of a H-bond with one of the coupling
partners is sufficient for catalysis. The aldehyde substrate must
have a H-bonding recognition site that allows binding to the
catalyst, but reaction with any aniline will then be accelerated.
This observation explains why no length selectivity is observed
in the product distribution in Fig. 5: AAA simply acts as an
elongation catalyst, an imine polymerase.
copy (see Fig. S56†). It was possible to measure the rate of imine
formation with the corresponding mono-aldehyde A0, and the
results are shown in Fig. 11. In this case, both the dimer DD and
the trimer DDD are effective catalysts for imine formation with
either the monofunctional aniline N0 or with the bisaniline N.
Both DD and DDD increase the rate by an order of magnitude
To evaluate the possible role of the phosphine oxide groups
in catalysis, this experiment was repeated but with the recog-
nition units swapped between the substrate and catalyst.
Addition of DDD increased the rate of reaction of the dia-
ldehyde phosphine oxide monomer A with the bisaniline linker
N, but the imine oligomers formed were not sufficiently soluble
1
Fig. 10 Half-life for reactions in the presence of AAA (3.3 mM) relative Fig. 12 Partial 500 MHz H-NMR spectra of a mixture of D (10 mM)
to the corresponding half-life in the presence of nBu3PO (10 mM) (t1/2 and B (10 mM) in toluene-d8 solution 5 minutes after the addition of N
(rel)). Half-lives were measured by integration of the 500 MHz 1H NMR (20 mM) in the presence of (a) AAA (3.3 mM) or (b) nBu3PO (10 mM).
signal due to the aldehyde as a function of time for mixtures of 20 mM The signals at 10.4–10.5 ppm are due to the aldehyde protons at the
total aldehyde (20 mM D0, 10 mM D, or 20 mM P) and 40 mM total end of D oligomers; the signal at 10.1–10.2 ppm is due to the aldehyde
aniline (40 mM N0, or 20 mM N) in toluene-d8.
proton of the mono-imine of B.
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
and the backbone is an oligo-amine made from dialdehyde and
bisaniline building blocks. Mixing the dialdehyde and bisani-
line components in toluene leads to a dynamic combinatorial
library (DCL) of different imine oligomers. The effects of adding
recognition-encoded amine oligomers to this reaction mixture
was investigated using NMR spectroscopy and HPLC to char-
acterise the rate of reaction and the product distribution.
Specically, addition of the AAA amine oligomer to a mixture of
the D dialdehyde and bisaniline linker did not template
formation of the complementary DDD imine oligomer, and no
change in product distribution was observed.
However, addition of AAA did increase the rate of imine
oligomerisation by an order of magnitude (Fig. 13). AAA was
found to catalyse imine formation reactions between different
substrates, but the minimal requirement for activity is that the
aldehyde must have a phenol recognition group to allow the
substrate to bind to the catalyst. Addition of a phenol as
a competitive inhibitor was found to reduce the rate of reaction,
conrming the role of substrate recognition in the observed
activity. AAA accelerates the rate of imine formation in toluene
by a factor of 20. The kinetic parameters for this enzyme-like
catalysis are estimated as 1 ꢀ 10ꢁ3 sꢁ1 for kcat and the dissoci-
ation constant for substrate binding is 300 mM. Neither a simple
phosphine oxide nor the shorter AA oligomer showed any
catalytic activity. The corresponding phenol oligomers DD and
DDD were found to catalyse imine formation with an aldehyde
bearing a complementary phosphine oxide recognition group,
which suggests a key role for the backbone in catalysis. The
secondary amines in the backbone are the most likely func-
tional groups to play a role in catalysis. One possible mecha-
nism is intramolecular reaction of a backbone amine of AAA
with an aldehyde on the end of a growing D oligomer, which is
bound to AAA. The resulting iminium ion intermediate would
provide a pathway for nucleophilic covalent catalysis of imine
formation. Oligomerisation of the phenol monomer D is rst
order in AAA, which is also active in catalytic amounts. This
unexpected imine polymerase activity in a synthetic oligomer
suggests that there are many interesting processes to be
discovered in the chemistry of synthetic recognition-encoded
oligomers that will parallel those found in natural
biopolymers.19
Fig. 13 Elongation of an imine oligomer of D catalysed by AAA. H-
bonding interactions between phenol and phosphine oxide recogni-
tion units are required for catalysis.
compared to the experiment with the corresponding mono-
meric phenol, 2-triuoromethyl phenol. No catalysis was
observed for reactions with an aldehyde lacking a recognition
site, P, which implies that phenol–phosphine oxide H-bonding
is important for catalysis by the donor oligomers. It is possible
that the phenol oligomers could catalyse imine formation by
a different mechanism from the phosphine oxide oligomers,
but the similarities between Fig. 10 and 11 suggests that it is the
backbone rather than the recognition units that are responsible
for the catalysis of imine formation by AAA. The phenol recog-
nition module is signicantly longer than the phosphine oxide
recognition module, and the difference in the geometric rela-
tionship between the substrate binding site, the position of the
reactive aldehyde, and the backbone might explain why DD is
a catalyst and AA is not.
Finally, the ability of AAA to selectively catalyse oligomer-
isation of D in the presence of a dialdehyde lacking any H-bond
1
recognition sites (B) was investigated. Fig. 12 shows H NMR
spectra of mixtures of D, B and N 5 minutes aer addition of
either nBu3PO or AAA. In the presence of the monomeric
phosphine oxide, the major species observed were the two
dialdehydes D and B, and only traces of imine were formed. In
the presence of AAA, B did not react, but D was rapidly con-
verted into imine oligomers. When the reaction was monitored
for longer time periods in the presence of the monomeric
phosphine oxide, the amount of imine formed was similar for
the two dialdehydes D and B (see Fig. S69†), so the differences
observed in the presence of AAA are not due a difference in the
inherent reactivity of the two substrates. Thus H-bonding
interactions between the D monomers and AAA leads to selec-
tive oligomerisation in the presence of other possible
substrates.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
LG thanks Marie Skłodowska-Curie Actions Individual Fellow-
ship (H2020-MSCA-IF-2016, DyNAmics – number 745730),
University of Padova (P-DiSC#05BIRD2019-UNIPD project) for
funding, and Dr Ana Belenguer for advice on the HPLC method.
Conclusions
Oligomers equipped with complementary recognition groups as
side chains form stable duplexes. In the system described here,
the recognition groups are 2-triuoromethyl phenol and phos-
phine oxide, which form strong H-bonds in toluene solution,
Notes and references
1 P. Chandra, A. M. Jonas and A. E. Fernandes, J. Am. Chem.
Soc., 2018, 140, 5179.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
2 (a) C. Tuerk and L. Gold, Science, 1990, 249(4968), 505; (b)
A. D. Ellington and J. W. Szostak, Nature, 1990, 346, 818;
(c) D. L. Robertson and G. F. Joyce, Nature, 1990, 344, 467.
3 (a) Z. Chen, P. A. Lichtor, A. P. Berliner, J. C. Chen and
D. R. Liu, Nat. Chem., 2018, 10, 420; (b) L. D. Usanov,
A. I. Chan, J. P. Maianti and D. R. Liu, Nat. Chem., 2018,
10, 704.
4 (a) A. Eschenmoser, Origins Life Evol. Biospheres, 1997, 27,
535; (b) S. Obika, D. Nanbu, Y. Hari, K. Morio, Y. In,
T. Ishida and T. Imanishi, Tetrahedron Lett., 1997, 38, 8735;
(c) A. A. Koshkin, S. K. Singh, P. Nielsen, V. K. Rajwanshi,
R. Kumar, M. Meldgaard, C. E. Olsen and J. Wengel,
Tetrahedron, 1998, 54, 3607; (d) A. Aerschot Van,
C. A. Hunter, Chem. Sci., 2019, 10, 2444; (i)
´
F. T. Szczypinski, L. Gabrielli and C. A. Hunter, Chem. Sci.,
˜
´
2019, 10, 5397; (j) L. Gabrielli, D. Nunez-Villanueva and
C. A. Hunter, Chem. Sci., 2020, 11, 561.
7 R. Amemiya, N. Saito and M. Yamaguchi, J. Org. Chem., 2008,
73, 7137.
8 (a) Y. Tanaka, H. Katagiri, Y. Furusho and E. A. Yashima,
Angew. Chem., Int. Ed., 2005, 44, 3867; (b) H. Ito,
Y. Furusho, T. Hasegawa and E. A. Yashima, J. Am. Chem.
Soc., 2008, 130, 14008.
9 (a) R. Kramer, J. M. Lehn and A. Marquis-Rigault, Proc.
Natl.Acad. Sci. U. S. A., 1993, 90, 5394; (b) P. N. Taylor and
H. L. Anderson, J. Am. Chem. Soc., 1999, 121, 11538.
¨
I. Verheggen, C. Hendrix and P. Herdewijn, Angew. Chem., 10 C. Bohler, P. E. Nielsen and L. E. Orgel, Nature, 1995, 376,
Int. Ed., 1995, 34, 1338; (e) K. U. Schoning, P. Scholz, 578.
S. Guntha, X. Wu, R. Krishnamurthy and A. Eschenmoser, 11 X. Li, Z. Y. J. Zhan, R. Knipe and D. G. Lynn, J. Am. Chem.
Science, 2000, 290, 1347; (f) L. Zhang, A. Peritz and Soc., 2002, 124, 746.
E. Meggers, J. Am. Chem. Soc., 2005, 127, 4174; (g) 12 J. T. Goodwin and D. G. Lynn, J. Am. Chem. Soc., 1992, 114,
A. B. Gerber and C. J. Leumann, Chem.–Eur. J., 2013, 19, 9197.
6990; (h) L. Zhang, Z. Yang, K. Sefah, K. M. Bradley, 13 D. M. Rosenbaum and D. R. Liu, J. Am. Chem. Soc., 2003, 125,
S. Hoshika, M. J. Kim, H. J. Kim, G. Zhu, E. Jimenez, 13924.
S. Cansiz, I. T. Teng, C. Champanhac, C. McLendon, 14 M. Ciaccia, R. Cacciapaglia, P. Mencarelli, L. Mandolini and
C. Liu, W. Zhang, D. L. Gerloff, Z. Huang, W. Tan and S. Di Stefano, Chem. Sci., 2013, 4, 2253.
S. A. Benner, J. Am. Chem. Soc., 2015, 137, 6734; (i) 15 (a) J. M. Lehn, Chem. –Eur. J., 1999, 5, 2455; (b) S. J. Rowan,
J. A. Piccirilli, T. Krauch, S. E. Moroney and S. A. Benner,
Nature, 1990, 343, 33; (j) B. A. Schweitzer and E. T. Kool, J.
Org. Chem., 1994, 59, 7238; (k) H. Liu, J. Gao, S. R. Lynch,
Y. D. Saito, L. Maynard and E. T. Kool, Science, 2003, 302,
S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders and
J. F. Stoddart, Angew. Chem., Int. Ed., 2002, 41, 898; (c)
P. T. Corbett, V. Leclaire, L. Vial, K. R. West, J. L. Wietor,
J. K. M. Sanders and S. Otto, Chem. Rev., 2006, 106, 3652.
868; (l) J. C. Delaney, J. Gao, H. Liu, N. Shrivastav, 16 For examples of supramolecular catalysis of polymerization
J. M. Essigmann and E. T. Kool, Angew. Chem., Int. Ed.,
2009, 48, 4524; (m) D. H. Appella, Curr. Opin. Chem. Biol.,
2009, 13, 687; (n) H. Isobe, T. Fujino, N. Yamazaki,
M. Guillot-Nieckowski and E. Nakamura, Org. Lett., 2008,
10, 3729; (o) K. Burgess, R. A. Gibbs, M. L. Metzker and
R. Raghavachari, J. Chem. Soc., Chem. Commun., 1994, 915.
5 (a) A. P. Bisson, F. J. Carver, C. A. Hunter and J. P. Waltho, J.
Am. Chem. Soc., 1994, 116(22), 10292; (b) A. P. Bisson,
F. J. Carver, D. S. Eggleston, R. C. Haltiwanger,
C. A. Hunter, D. L. Livingstone, J. F. McCabe, C. Rotger
and A. E. Rowan, J. Am. Chem. Soc., 2000, 122, 8856; (c)
H. Gong and M. J. Krische, J. Am. Chem. Soc., 2005, 127,
1719; (d) Y. Yang, Z. Y. Yang, Y. P. Yi, J. F. Xiang,
C. F. Chen, L. J. Wan and Z. G. Shuai, J. Org. Chem., 2007,
72, 4936.
see (a) J. Ferguson and S. A. O. Shah, Eur. Polym. J., 1968,
4, 343; (b) E. Tsuchida and Y. Osada, J. Polym. Sci. Part A:
Polym. Chem., 1975, 13, 559; (c) R. Saito and H. Kobayashi,
Macromolecules, 2002, 35, 7207; (d) R. C. Pratt,
B. G. G. Lohmeijer, D. A. Long, P. N. P. Lundberg,
A. P. Dove, H. Li, C. G. Wade, R. M. Waymouth and
J. L. Hedrick, Macromolecules, 2006, 39, 7863; (e) S. Koeller,
J. Kadota, A. Deffieux, F. Peruch, S. Massip, J.-M. Leger,
J.-P. Desvergne and B. Bibal, J. Am. Chem. Soc., 2009, 131,
15088; (f) Y. Takashima, M. Osaki, Y. Ishimaru,
H. Yamaguchi and A. Harada, Angew. Chem., Int. Ed., 2011,
50, 7524; (g) J. U. Pothupitiya, N. U. Dharmaratne,
T. M. M. Jouaneh, K. V. Fastnacht, D. N. Coderre and
M. K. Kiesewetter, Macromolecules, 2017, 50, 8948; (h)
X. Tang, Z. Huang, H. Chen, Y. Kang, J.-F. Xu and
X. Zhang, Angew. Chem., Int. Ed., 2018, 57, 8545; (i)
F. Eisenreich, M. Kathan, A. Dallmann, S. P. Ihrig,
T. Schwaar, B. M. Schmidt and S. Hecht, Nat. Catal., 2018,
1, 516; (j) J. Y. C. Lim, N. Yuntawattana, P. D. Beer and
C. K. Williams, Angew. Chem., Int. Ed., 2019, 58, 6007.
6 (a) G. Iadevaia, A. E. Stross, A. Neumann and C. A. Hunter,
Chem. Sci., 2016, 7, 1760; (b) A. E. Stross, G. Iadevaia and
C. A. Hunter, Chem. Sci., 2016, 7, 5686; (c) A. E. Stross,
G. Iadevaia and C. A. Hunter, Chem. Sci., 2016, 7, 94; (d)
˜
´
D. Nunez-Villanueva and C. A. Hunter, Chem. Sci., 2017, 8,
˜
´
206; (e) D. Nunez-Villanueva, G. Iadevaia, A. E. Stross, 17 This estimate assumes that the amine template binds the
M. A. Jinks, J. A. Swain and C. A. Hunter, J. Am. Chem. Soc.,
complementary imine with the same association constant
˜
´
2017, 139, 6654; (f) A. E. Stross, G. Iadevaia, D. Nunez-
as measured for the amine duplex AAA$DDD shown in Fig. 1.
Villanueva and C. A. Hunter, J. Am. Chem. Soc., 2017, 139, 18 I. Huc and J. M. Lehn, Proc. Natl. Acad. Sci. U. S. A., 1997, 94,
12655; (g) J. A. Swain, G. Iadevaia and C. A. Hunter, J. Am. 2106.
Chem. Soc., 2018, 140, 11526; (h) F. T. Szczypinski and 19 D. M. J. Lilley, Philos. Trans. R. Soc., B, 2011, 366, 2910.
´
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.