Self-duplicating amplification in a dynamic combinatorial library

Article abstract of DOI:10.1021/ja710248q

A dynamic combinatorial library has been designed to produce a set of constituents among which one is able to self-complementarily direct its duplication. The overall molecular distribution in the library evolves along both kinetic and thermodynamic biases, leading to the amplification of the species that reproduces most efficiently and to the depletion of the other competitors. Copyright

Full text of DOI:10.1021/ja710248q

Published on Web 01/23/2008
Self-Duplicating Amplification in a Dynamic Combinatorial Library
Shengguang Xu‡ and Nicolas Giuseppone*
Institut de Science et d’Inge´nierie Supramole´culaires, UniVersite´ Louis Pasteur, 8 Alle´e Gaspard-Monge, BP 70028,
67083 Strasbourg, and Institut Charles Sadron CNRS - UPR 22, Campus CNRS Cronenbourg, 23 rue du Loess,
67037 Strasbourg Cedex, France
Received November 12, 2007; E-mail: giuseppone@isis.u-strasbg.fr
Scheme 1. Representation of the 11 Library Members Obtained
by Mixing the Three Aldehydes (Al1-Al3) and the Two Amines
(Am1, Am2) in Deuterated Chloroforma
Dynamic combinatorial chemistry (DCC) rests on the design and
the study of libraries of species connected by reversible (supra)-
molecular bonds.1,2 At equilibrium, the expression of the products
in a dynamic combinatorial library (DCL) is governed by thermo-
dynamics and as a consequence, the additional presence of
molecular targets can lead to the in situ screening of the “best-
fitted” constituents. Investigations on DCC by several research
groups have highlighted the wide potentialities of the main concept
in various fields of chemical research such as drug discovery,3 self-
assembly of inorganic architectures, receptor generation, substrate
binding, and catalyst screening.2c,4 DCC has also been shown to
respond to external chemical or physical stimuli such as protons,
phase transition, temperature, or electric field modulations, thus
opening interesting potentialities for material science.5,6 Recently,
we have envisaged the possibility to couple DCC with self-
replicating systems.7 Indeed, the overexpression of a compound
by making copies of itself from a pool of reshuffling constituents
in a series of competing equilibria would be of significance for
biologists and chemists who are interested in various minimal “self-
amplifying” systems in their quest to bridge the gap with a possible
origin of life. Moving in this direction, we describe here a
preliminary investigation of a DCL which amplifies a product that
constitutes its own target, that is, the one capable of self-duplication.
To design such a system, we envisioned several building blocks
capable of (i) reversible covalent associations and (ii) displaying,
or not displaying, complementary supramolecular units in order to
produce, or not produce, a template with self-recognition properties
(Scheme 1). The key molecule was chosen as imine Al1Am1,
synthesized by the condensation of aldehyde Al1 (derived from a
Kemp’s imide) and adenosine amine Am1. This self-complementary
dynamic compound, inspired by a related “non-dynamic” isosteric
system described by Julius Rebek,8,9 is able to strongly associate
with itself into complex [Al1Am1]2. Its dimerization in a chloroform
a In this combinatorial set, one imine, namely Al1Am1, can self-assemble
through hydrogen bonds and produce homodimer [Al1Am1]2. In the frame
(middle right), the connectivity map describes the antagonistic and the
agonistic constitutional relationships between the six imines of the library.11
We then turned to the thermodynamic study of the DCL
described in Scheme 1. We set up a first experiment (DCL1) by
mixing Al2, Al3, Am1, and Am2 (15 mM each at 22 °C in CDCl3),
to determine the distribution of the eight library members at
equilibrium (Figure 1, black bars). The isoenergetic nature2a of
DCL1 is here clearly demonstrated by the statistical distribution
of the products (equal expression of the two amines and two
aldehydes (3.4 mM each) and equal expression of the four imines
Al(2,3)Am(1,2) (5.8 mM each)). This reflects, as expected, the absence
of specific supramolecular interactions between the library members.
In a second experiment (DCL2), we mixed together Al1, Al2, Al3,
Am1, and Am2 (15 mM each at 22 °C in CDCl3). In this library,
the production of homodimer [Al1Am1]2 leads to a strong bias in
the expression of the constituents at equilibrium, far from the
statistical distribution (Figure 1, gray bars). In a third experiment
(DCL3), we divided the process in two steps: (i) set up of a pre-
equilibrated eight member library without Al1 (DCL1, Figure 1,
black bars), and then (ii) addition of Al1 (15 mM) to this library.
In DCL3, it takes five times longer to reach equilibrium (22 days)
compared to DCL2 (t ) 109 h), but the competition produces an
identical distribution of constituents in both DCL2 and DCL3
(Figure 1, gray bars for both conditions). These last two experiments
confirmed the good thermodynamic control of the overall system
in these conditions, that is, without local pseudominima on the
hypersurface of energy/constitution. By comparing the “non-self-
duplicating” DCL1 (black bars) and the “self-duplicating” DCL3
(gray bars), we can conclude on the three following features. First,
the expression of the self-duplicator Al1Am1 (9.03 mM) is increased
by +83% compared to the theoretical statistical distribution (4.94
1
solution through hydrogen bonding was confirmed by H NMR
(see Supporting Information (SI)). It is observed that the N-H imide
resonance signal of compound Al1 is shifted from 7.6 to 8.8 ppm
when Al1 is mixed with Am1 (c ) 15 mM, equimolar ratio), and
to 11.2 ppm for the condensed imine Al1Am1, which is in agreement
with the formation of complex [Al1Am1]2.8b,10 The presence of
homodimer [Al1Am1]2 in solution was also confirmed by exact mass
spectrometry (ESI-TOF). While Al1 contains a free imide bond on
the Kemp recognition group, Al2 is protected by a methyl group
on the nitrogen which prevents the formation of hydrogen bonds
with the adenine moiety. Al3 is also an analogue of Al1 but without
a Kemp’s recognition group and with an acetyl group instead, in
order to display similar activation energy as Al1 for the condensation
reaction with amines. Am1 and Am2 also present close structures,
but in the latter, the hydrogen bonds with the Kemp’s imide are
sterically restricted by the protection of the adenine with a benzyl
group.
‡ Current address: School of Chemistry and Chemical Engineering, Shandong
University, 250100, Jinan, China.
9
1826
J. AM. CHEM. SOC. 2008, 130, 1826-1827
10.1021/ja710248q CCC: $40.75 © 2008 American Chemical Society

C O M M U N I C A T I O N S
complementarily direct its own formation. The expression of the
components in the library evolves along both kinetic and thermo-
dynamic biases that both lead to the amplification of the best
duplicator. Because of the double reversibility of the system
(supramolecular H-bonds and molecular imine condensation), the
competition is not only ruled by the differential rates of formation
of the components, but also by the possible takeover of the building
blocks of the antagonistic competitors,11 thus leading to the decrease
of their absolute concentration. From a “Darwinian” point of view,
such a system illustrates the selection of the most efficient self-
duplicator by the destruction of the entities which are not (or less,
such as Al1Am2) able to duplicate themselves. We are currently
studying the possibility to couple a dynamic combinatorial library
with a “self-replicating loop”,7d,12 thus leading to an auto-catalytic
behavior, that is, displaying a sigmoid concentration-time profile
for the self-replicated member of the library.
Figure 1. Representation of the products’ concentration in various dynamic
combinatorial libraries (DCL1-3), determined by 1H NMR at 22 °C in
CDCl3 (c ) 15 mM for each of the starting materials.
Acknowledgment. This work was supported by a doctoral
fellowship from the China Scholarship Council within the frame
of the Colle`ge Doctoral Franco-Chinois between University Louis
Pasteur (Strasbourg, France) and Shandong University (Jinan,
China) (S.X.). We wish to express thanks to Pr. Jean-Marie Lehn
for his help at various stages in the past year.
Supporting Information Available: Characteristic NMR spectra
as well as a discussion of kinetics and thermodynamics data. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4763-4768.
(2) For reviews on dynamic combinatorial chemistry, see for instance: (a)
Lehn, J.-M. Chem. Eur. J. 1999, 5, 2455-2463. (b) Rowan, S. J.; Cantrill,
S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew. Chem.,
Int. Ed. 2002, 41, 898-952. (c) Corbett, P. T.; Leclaire, J.; Vial, J.; West,
K. R.; Wietor, J.-L.; Sanders, J. K. M.; Otto, S. Chem. ReV. 2006, 106,
3652-3711.
(3) (a) Ramstro¨m, O.; Lehn, J.-M. Nature ReV. Drug DiscoVery 2001, 1, 26-
36. (b) Cheeseman, J. D.; Corbett, A. D.; Gleason, J. L.; Kazlauskas, R.
J. Chem. Eur. J. 2005, 11, 1708-1716.
Figure 2. Evolution as a function of time of the concentration of the 6
products Al(1-3)Am(1,2) in the 11-member dynamic combinatorial library
(DCL2) described in Scheme 1. Concentrations were determined by 1H
NMR and products are represented as follows: (9) Al1Am1, (b) Al1Am2,
(O) Al2Am1, (+) Al2Am2, (right solid triangle) Al3Am1, and (∆) Al3Am2.
(4) For examples of recent advances, see for instance: (a) Severin, K. Chem.
Eur. J. 2004, 10, 2565-2580. (b) Saur, I.; Scopelliti, R.; Severin, K. Chem.
Eur. J. 2006, 12, 1058-1066. (c) Giuseppone, N.; Schmitt, J.-L.; Lehn,
J.-M. J. Am. Chem. Soc. 2006, 128, 16748-16763. (d) Nitschke, J. R.
Acc. Chem. Res. 2007, 40, 103-112. (e) Vongvilai, P.; Angelin, M.;
Larsson, R.; Ramstro¨m, O. Angew. Chem., Int. Ed. 2007, 46, 948-950.
(f) Kerckhoffs, J. M. C. A.; Mateos-Timoneda, M. A.; Reinhoudt, D. N.;
Crego-Calama, M. Chem. Eur. J. 2007, 2377-2485. (g) Ludlow, R. F.;
Liu, J.; Li, H.; Roberts, S. L.; Sanders, J. K. M.; Otto, S. Angew. Chem.,
Int. Ed. 2007, 46, 5762-5764. (h) Legrand, Y.-M.; van der Lee, A.;
Barboiu, M. Inorg. Chem. 2007, 46, 9540-9547.
mM). Second, this amplification of the self-duplicator is realized
by the takeover of the resources of its direct competitors, that is,
the imines having antagonistic connectivity11 with Al1Am1 (namely
Al1Am2, Al2Am1, and Al3Am1, (3.0 mM each), see middle right
frame in Scheme 1). Thus, the amplification of the self-duplicator
Al1Am1 compared to its direct competitors reaches a value of
+200%. And last, the agonistic connectivity11 between the self-
duplicator Al1Am1 together with Al2Am2 and Al3Am2 leads to a
small increase of the last two products compared to the statistical
distribution (5.7 mM; +14%).
We also studied the kinetic behavior of the 11-member library
DCL2 by plotting the concentration of the constituents as a function
of time (Figure 2). This kinetic evolution is biased compared to
the “non-self-duplicating” DCL1 involving Al2, Al3, Am1, and Am2,
in which the 4 imines display very similar initial rates in the
competition experiment (≈4.5 × 10-1 mM‚h-1, see SI). In DCL2,
the self-duplicator Al1Am1 is produced with a V0 of 59 × 10-1
mM‚h-1, which is about 60 times faster than the condensation of
Al2Am1, and Al3Am1 (0.96 × 10-1 mM‚h-1); 13 times faster than
Al2Am2 and Al3Am2 (4.6 × 10-1 mM‚h-1); and 6 times faster than
Al1Am2 (9.6 × 10-1 mM‚h-1). These differential rates lead to a
maximum (+200%) of amplification (kinetic amplification) of the
self-duplicator (10.98 mM) compared to its immediate competitor
Al1Am2sand of +160% compared to the average concentration
of the six iminessat t ) 16 h (Figure 1, striped bars). We assume
that the kinetic amplification is mainly the result of a pre-association
complex between Al1 and Am1 as indicated by the kinetic studies
of the individual reactions (SI).
(5) (a) Giuseppone, N.; Lehn, J.-M. Chem.-Eur. J. 2006, 12, 1715-1722.
(b) Giuseppone, N.; Fuks, G.; Lehn, J.-M. Chem. Eur. J. 2006, 12, 1723-
1735. (c) Giuseppone, N.; Lehn, J.-M. Angew. Chem., Int. Ed. 2006, 45,
4619-4624.
(6) Lehn, J.-M. Chem. Soc. ReV. 2007, 36, 151-160.
(7) For reviews on self-replication, see: (a) Morgan Conn, M.; Rebek, J. Curr.
Opin. Struct. Biol. 1994, 4, 629-635. (b) Robertson, A.; Sinclair, A. J.;
Philp, D. Chem. Soc. ReV. 2000, 29, 141-152. (c) Paul, N.; Joyce, G. F.
Curr. Opin. Chem. Biol. 2004, 8, 634-639. (d) Patzke, V.; von
Kiedrowski, G. ARKIVOC 2007, 293-310. For a study of the kinetic
acceleration of a single imine condensation upon the initial addition of
the final product, see: (e) Terfort, A.; von Kiedrowski, G. Angew. Chem.,
Int. Ed. 1992, 31, 654-656.
(8) (a) Tjivikua, T.; Baluster, R.; Rebek, J. J. Am. Chem. Soc. 1990, 112,
1249-1250. (b) Nowick, J. S.; Feng, Q.; Tjivikua, T.; Ballester, P.; Rebek,
J. J. Am. Chem. Soc. 1991, 113, 8831-8839.
(9) The corresponding Rebek “self-replicator” is made by the association of
adenosine amine 1 and a Kemp’s imide activated ester, producing an
irreVersible amide bond instead of a reVersible imine bond in our case.
(10) Askew, B.; Ballester, P.; Buhr, C.; Jeong, K. S.; Jones, S.; Parris, K.;
Williams, K.; Rebek, J. J. Am. Chem. Soc. 1989, 111, 1082-1090.
(11) Constituents directly related through one of their components may be
considered as antagonists (increasing one, decreases the others), whereas
constituents with no common component behave as agonists (increasing
one, increases the other), as defined in reference 5a; see also 4a.
(12) We use the term “self-replication” for auto-catalytic systems displaying
a sigmoid concentration-time profile and the term “self-duplication” for
a system displaying the general property to thermodynamically or
kinetically (or both) favor its own formation (see also SI).
In conclusion, we have demonstrated that it is possible to self-
amplify one product in a DCL, namely the one that can self-
JA710248Q
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 1827