Identification of a foldaxane kinetic byproduct during guest-induced single to double helix conversion

Article abstract of DOI:10.1021/ja3064364

An aromatic oligoamide sequence was designed and synthesized to fold in a single helix having a large cavity and to behave as a host for a dumbbell-shaped guest derived from tartaric acid. NMR, molecular modeling, and circular dichroism (CD) evidence demo

Full text of DOI:10.1021/ja3064364

Communication
pubs.acs.org/JACS
Identification of a Foldaxane Kinetic Byproduct during Guest-
Induced Single to Double Helix Conversion
Quan Gan,†,‡ Yann Ferrand,†,‡ Nagula Chandramouli,†,‡ Brice Kauffmann,§,∥,⊥ Christophe Aube,#
Didier Dubreuil,# and Ivan Huc*,†,‡
†Universite
‡CNRS, CBMN (UMR 5248), Institut Europee
§Universite
de Bordeaux, Institut Europeen de Chimie Biologie (UMS 3033/US 001), F-33600 Pessac, France
∥CNRS, Institut Europee
n de Chimie Biologie (UMS 3033), F-33600 Pessac, France
⊥INSERM, Institut Europee
n de Chimie Biologie (US 001), F-33600 Pessac, France
#Universite
de Nantes, CNRS, Chimie et Interdisciplinarite: Synthese, Analyse, Model
Sciences et des Techniques, 2 rue de la Houssiniere, BP 92208, 44322 Nantes Cedex 3, France
́
́
de Bordeaux, CBMN (UMR 5248), Institut Europeen de Chimie Biologie, 2 rue Escarpit, 33607 Pessac, France
́
n de Chimie Biologie, 2 rue Escarpit, 33607 Pessac, France
́
́
́
́
́
́
̀
́ ́
isation (CEISAM), UMR 6230, Faculte des
̀
S
* Supporting Information
We recently introduced the helical aromatic oligoamide
foldamer molecular capsule 1, which can bind tartaric acid
diastereoselectively (Chart 1).4 Its design is based on amino
ABSTRACT: An aromatic oligoamide sequence was
designed and synthesized to fold in a single helix having
a large cavity and to behave as a host for a dumbbell-
shaped guest derived from tartaric acid. NMR, molecular
modeling, and circular dichroism (CD) evidence demon-
strated the rapid formation of this 1:1 host−guest complex
and induction of the helix handedness of the host by the
guest. This complex was found to be a long-lived kinetic
supramolecular byproduct, as it slowly transformed into a
2:2 host−guest complex with two guest molecules bound
at the extremities of a double helix formed by the host, as
shown by NMR and CD spectroscopy and a solid-state
structure. The guest also induced the handedness of the
double helical host, but with an opposite bias. The
chiroptical properties of the system were thus found to
revert with time as the 1:1 complex formed first, followed
by the 2:2 complex.
Chart 1. Formulas of Oligomers 1 and 2 and Dumbbell
a
Guest 3
he comprehension of kinetically controlled reaction steps
a
A letter code is used for the abbreviations of the oligomer subunits.
T
in supramolecular events is crucial for the construction of
out-of-equilibrium artificial ordered aggregates,1 complex self-
assemblies,2 and sophisticated molecular machines3 that would
resemble their biological counterparts. In the present study, we
show that mixing a helical host and a chiral guest results in the
formation of a transient supramolecular species that gives rise
to a temporary chiroptical signature opposite to that observed
at equilibrium. Specifically, a tartaric acid-derived guest rapidly
interacts with a single helical aromatic oligoamide foldamer that
can wrap around it to form a 1:1 host−guest complex with a
preferred handedness. This complex then progressively trans-
forms into a thermodynamically more favorable 2:2 architecture
composed of two guests and a double helix and having a
handedness opposite to that of the 1:1 complex. The same
guest is thus able to induce P or M helicity depending on
whether it interacts with a single or a double helix. These
results represent a striking illustration of the multiple processes
that compete and take place on different time scales in
supramolecular reaction sequences.
acid units coding for a large helix cavity in the center of the
sequence and a narrow helix diameter at the ends, thus creating
a binding site completely surrounded by the helix backbone. In
a subsequent work, we showed that multiturn single- or double-
helical oligoamide foldamers possessing open cavities can wind
around urethane dumbbell guests to form thermodynamically
stable complexes in which the two bulky stoppers of the guest
protrude from the two ends of the helix cavity in a
pseudorotaxane-like geometry (Figure 1, top right).3b,5,6
These complexes were termed foldaxanes.5 We envisioned
that a combination of these two concepts may allow the
creation of foldaxanes using dumbbell guests such as 3 derived
from tartaric acid.
Received: July 2, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja3064364 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Oligomer 2 was designed for this purpose. Its sequence is a
shortened version of 1 in which the terminal quinoline units
that would close the cavity have been replaced by pivaloyl (piv)
groups. The helix of 2 was thus expected to possess an open
cavity. Naphthyridine (N) units were kept in the design
because they are directly involved in intermolecular hydrogen
bonding to tartaric acid.4 The synthesis of 2 involves the piv-
PN2-Boc building block, which after Boc cleavage was coupled
twice to the acids of the central pyridine−pyridazine−pyridine
(pyr−pyz−pyr) unit [see the Supporting Information (SI)].
The two optically pure enantiomers 3-(D) and 3-(L) were
prepared by coupling of 3,5-di-tert-butylbenzoyl chloride to D-
and L-dibenzyl tartrate, respectively, followed by Pd/C-
catalyzed hydrogenolysis of the benzyl groups.7
SI). These results further expand the already wide range of
aromatic amino acids that are compatible with double helix
formation to include the central pyr−pyz−pyr segment.
Double helices do not form in the case of the capsule sequence
1 from which 2 is derived.4 This is explained by the poor ability
of the terminal quinoline units to undergo the springlike
extension required for double helix formation.9 Suppressing (or
reducing the number of) quinoline units that code for a narrow
helix diameter has been observed to promote double helix
formation in other systems.10
Host−guest interactions were then investigated using both
1H NMR and circular dichroism (CD) spectroscopy to estimate
the ability of 2 to fold around tartaric acid derivative 3. Upon
the addition of 3 to a 0.1 mM solution of 2 in CDCl3, a
concentration at which the single-helical conformation of 2
predominates (>99%) (Figure 2A, open circles), the NMR
A preliminary investigation of the behavior of 2 in solution
revealed that like many other aromatic amide sequences,8 its
single-helical conformation is in equilibrium with a double-
helical species (Figure 1 left). This is reflected by, for example,
Figure 1. (left) Equilibrium between oligomer 2 as P (blue) and M
(red) single helices and P and M double helices. (top) Equilibrium
between oligomer 2 as P and M single helices and (M)-2⊃3-(L).
(bottom) Equilibrium between oligomer 2 as P and M double helices
and a 2:2 complex between [(P)-2]2 as a double helix and two 3-(L)
molecules. It should be noted that the P and M single helices are at
equilibrium, whereas the P and M double helices are not unless they
first dissociate into single helices.
Figure 2. (A) Parts of the 300 MHz 1H NMR spectra of 2 (0.1 mM)
in CDCl3 at 298 K (top to bottom) before the addition of guest and 5
min, 8 h, 4 days, 10 days, and 18 days after the addition of 6 equiv of 3.
Open circles denote signals of the empty capsule. The asterisks denote
the hydrogen-bonded acid protons of the guest in 2⊃3. The NH
amide signals of the single-helical host−guest complex are marked with
open squares, whereas those of the double-helical host−guest complex
are marked with a solid square. (B) CD spectra of 2 (0.1 mM in
CDCl3) at 298 K 10 min after the addition of 6 equiv of 3-(D) (blue
spectrum) and after 5 h, 15 h, 30 h, 70 h, and 18 days (increasing red
intensity). (C) CD spectra of 2 (0.1 mM in CDCl3) 10 min after the
addition of 6 equiv of 3-(L) (red spectrum) and after 5 h, 15 h, 30 h,
70 h, and 18 days (increasing blue intensity).
1
two sets of signals in the H NMR spectrum that are assigned
to the monomer and the dimer, the proportions of which vary
with concentration. A 1H diffusion-ordered spectroscopy
(DOSY) measurement on a 2 mM solution of 2 at 293 K
(Figure S2 in the SI) showed two distinct diffusion coefficients
for 2 and its double helix (2)2, which were calculated to be 7.38
× 10−10 and 6.44 × 10−10 m2 s−1, respectively. However, unlike
for other aromatic amide double helices,8 we found that the
NMR signals of the double helix also varied with concentration
(Δδ up to 0.2 ppm), indicating another aggregation
phenomenon that takes place rapidly on the NMR time scale.
Self-aggregation of the double helices themselves due to the
aromatic and hydrogen bonding groups present at their termini
could explain these chemical shift variations. This resulted in a
bias of the single helix ⇄ double helix equilibrium, which
shifted in favor of the double helix at high concentration. For
example, the apparent dimerization constants were measured to
be 420 M−1 at 0.5 mM and 4900 M−1 at 8 mM (Table S1 in the
SI). Exchange spectroscopy (EXSY) NMR experiments on a
2.4 mM solution at 293 K allowed the rate constants for double
helix formation and dissociation to be calculated as 107 s−1 M−1
and 0.061 s−1, respectively. The double-helical dimer was also
characterized in the solid state by X-ray crystallography (see the
signals of the latter disappeared, and signals due to a new
complex in which 3 is bound within the cavity of 2 emerged
(Figure 2A, open squares). The binding constant was measured
accurately through the integral ratios of the free and bound
receptor amide resonances to be Ka = 2.7 × 104 L mol−1 at 298
K in CDCl3. On the basis of previous investigations of the
binding of tartaric acid to 1,4 the sharp signal at 15.7 ppm in the
spectrum of 2⊃3 was assigned to the acid protons of the guest,
which are hydrogen-bonded to the central N units of 2,
suggesting similar modes of binding of tartaric acid by 1 and 2.
This similarity was further expressed in CD titrations, for which
strong responses with opposite signs appeared upon the
addition of 3-(D) or 3-(L).
B
dx.doi.org/10.1021/ja3064364 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
The host−guest interaction is fully diastereoselective and
results in an induced handedness in 2. The strong negative
Cotton effect at 360 nm for 2⊃3-(L) (Figure 2C, red spectrum)
suggests that the same handedness preference occurs in 2⊃3 as
in 1⊃tartaric acid: the L enantiomer favors M helicity. CD was
also employed as an additional technique to support the NMR
titrations. The progressive addition of 3-(D) into a 10 μM
solution of 2 resulted in the appearance a positive signal that
gave an excellent fit to a 1:1 binding model.11 The binding
constant measured by CD at 273 K (4.4 × 104 L mol−1) was
reasonably close to that obtained by NMR analysis at 298 K.
These results, along with the detailed structural data available
for the 1⊃tartaric acid complex,4a allowed the proposal of an
energy-minimized model of the structure of (M)-2⊃3-(L)
having the expected foldaxane architecture with the helix
wrapped around the dumbbell guest (Figure 3A,B) and the
carboxylic acid groups of the guest hydrogen-bonded to the N
units of the host.
monitored by both NMR spectroscopy, where all of the initial
signals, including the guest resonance at 15.7 ppm, disappeared
(Figure 2A), and by CD spectroscopy, where a remarkable
inversion of sign at 360 nm was observed. This modification
reflects an inversion of the handedness13 of 2 mediated by the
guest that changes the helicity from P to M in the case of 3-(D)
and from M to P in the case of 3-(L) (Figure 2B,C).
X-ray-quality single crystals were obtained by slow diffusion
of hexane into a chloroform solution of 2 mixed with racemic 3
at thermodynamic equilibrium, and the solid-state structure was
solved (Figure 3C,D). It revealed an original architecture
completely different from that of the foldaxane 2⊃3 composed
of a 2:2 complex consisting of (2)2 as a double helix with a
bound molecule of 3 capping the helix cavity at each extremity.
The structure of (2)2 complexed with 3 much resembles its
structure without bound 3. As a slight difference, the duplex
diameter is slightly decreased in the complex with 3, as reflected
by a helix span longer by about half a pyridine ring. In the
crystal structure, 3-(L) is bound to [(P)-2]2 and 3-(D) is bound
to [(M)-2]2, consistent with the inversion of preferred
handedness observed over time when 2 was titrated with
either 3-(L) or 3-(D) (Figure 2). Each tartaric acid derivative
adopts a typical conformation with trans acid groups and
gauche hydroxy groups. The carboxylic acid groups are oriented
perpendicular to the edge of the helical strand and doubly
hydrogen-bonded to the distal pyridines (dOH···N = 2.63 Å,
dNH···OC = 2.99 Å), giving a total of eight intermolecular
hydrogen bonds. The interactions between 3 and (2)2 differ
from those between 3 and single-helical 2 in that the hydrogen
bonds occur with the N units of 2 and the pyridine units of
(2)2. Also, van der Waals contacts between the 3,5-di-tert-
butylphenyl rings of 3 and 2 appear to be more extensive in the
crystal structure of (2)2⊃(3)2 than in the calculated structure of
2⊃3. However, it is unclear why these differences result in the
much higher stability of (2)2⊃(3)2.
Consistent with the solid-state structure, evidence of the
double helical nature of the host in solution was found. In
particular, the 1H signals of 2 in (2)2⊃(3)2 are strongly upfield-
shifted from those of 2 in 2⊃3 as a result of enhanced ring
current effects in the double-helical dimer.8
A
1H DOSY
experiment confirmed the larger size of (2)2⊃(3)2 relative to
2⊃3 (Figure S4). Adding racemic 3 to a racemic (P/M)
solution of (2)2 (at 30 mM, the double helix prevails) readily
caused shifts of both the aromatic and amide signals, reflecting
the fact that [(P/M)-2]2 and its complexes with 3-(L) and 3-(D)
exchange rapidly on the NMR time scale (Figure S12). The
association constant of (2)2 and 3 was too large to be calculated
accurately from curve fitting of the titration data. Indeed, this
binding constant is expected to be significantly larger than that
for binding of 3 to single-helical 2 (i.e., larger than 2.7 × 104 L
mol−1), as the apparent dimerization of 2 to form (2)2 is
strongly enhanced in the presence of 3 (Figure 2). Also, this
titration experiment did not allow any cooperativity between
the binding of the first and second guests to be characterized.
In an other titration, a single enantiomer of 3 was added to a
racemic (P/M) solution of (2)2 (Figure S14). This resulted in
Figure 3. (A) CPK (guest) and tube (host) and (B) CPK (host and
guest) representations of the structure of the foldaxane (M)-2⊃3-(L)
as obtained by molecular modeling using Maestro version 6.5 with the
Merck molecular force field. (C) CPK (guest) and tube (host) and
(D) CPK (host and guest) representations of the solid-state structure
of [(P)-2]2⊃[3-(L)]2. The isobutyl side chains and solvent molecules
have been omitted for clarity, except for the water cluster included in
the cavity of the double helix (purple balls).
The kinetics of the quantitative formation of 2⊃3 from the
single helix of 2 and the kinetics of guest-induced single-helix
handedness inversion are fast: a steady state was reached within
seconds (before CD or NMR measurements can be carried
out).6a,b,12 When stored at −18 °C, a sample remained
unchanged for weeks. However, upon standing at 25 °C, the
complex 2⊃3 slowly disappeared over the course of days and
was quantitatively replaced by another species, showing that
2⊃3 is not the thermodynamic product. This conversion was
1
the induction of CD bands and splitting of H NMR signals
into two sets, one having chemical shift values identical to those
observed in the titration with racemic 3 and one having
chemical shift values almost identical to those of free (2)2 (see
the SI). This reflects the fact that the pure enantiomer of 3
forms a stable complex with the (2)2 duplex that has the
matching stereochemistry and a weak complex or no complex
C
dx.doi.org/10.1021/ja3064364 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(d) Cangelosi, V. M.; Carter, T. G.; Zakharov, L. N.; Johnson, D. W.
Chem. Commun. 2009, 5606.
(3) (a) Kinbara, K.; Aida, T. Chem. Rev. 2005, 105, 1377. (b) Gan,
at all with the (2)2 duplex having the mismatching stereo-
chemistry, suggesting that the interaction between (2)2 and 3 is
also strongly diastereoselective. Since the equilibrium between
[(P)-2]2 and [(M)-2]2 is slow on the NMR time scale, 3 acts as
a sort of chiral shift reagent that resolves them into distinct
signals before their interconversion occurs.
Interestingly, the induced CD spectrum of 2 resulting from a
handedness bias by 3 was found to be ∼5 times more intense in
(2)2⊃(3)2 than in 2⊃3 (Figure 2B,C) despite the fact that
handedness induction is (close to) quantitative in both cases.
The Δε values appear to be higher for double-helical (2)2 than
for single-helical 2, whose helix pitch is half as small.
In summary, guest 3 enhances the thermodynamic stability of
the double-helical duplex (2)2 but simultaneously reduces the
kinetics of (2)2 production upon forming a foldaxane with
monomeric 2, which can be long-lived at low temperature.
While each individual equilibrium appears to be relatively fast
when assessed in the millimolar range, the formation of 2⊃3 at
0.1 mM results in a very low concentration of single-helical 2,
which slows the formation of (2)2. Similarly, the formation of
(2)2⊃(3)2 is slower when the concentrations of both (2)2 and
free 3 are low. The sequence of steps for conversion of 2⊃3
into (2)2⊃(3)2 is shown in Figure 1: (i) dissociation into 2 and
3; (ii) inversion of the helix handedness of 2 (see ref 14 for a
possible mechanism); (iii) association of 2 into (2)2; and (iv)
binding of 3 to (2)2. The time scales involved in the equilibria
at each step allow the isolation of both the kinetic and
thermodynamic supramolecular products and monitoring of the
inversion of chiroptical properties with time as the system first
evolves toward a product with one handedness and then reverts
into another product having the opposite handedness.
Q.; Ferrand, Y.; Bao, C.; Kauffmann, B.; Grel
Science 2011, 331, 1172.
(4) (a) Ferrand, Y.; Kendhale, A. M.; Kauffmann, B.; Grel
Marie, C.; Blot, V.; Pipelier, M.; Dubreuil, D.; Huc, I. J. Am. Chem. Soc.
2010, 132, 7858. (b) Ferrand, Y.; Nagula, C.; Kendhale, A. M.; Aube,
́
ard, A.; Jiang, H.; Huc, I.
́
ard, A.;
́
C.; Kauffmann, B.; Grelard, A.; Laguerre, M.; Dubreuil, D.; Huc, I. J.
Am. Chem. Soc. 2012, 134, 11282.
(5) Ferrand, Y.; Gan, Q.; Kauffmann, B.; Jiang, H.; Huc, I. Angew.
Chem., Int. Ed. 2011, 50, 7572.
(6) (a) Nishinaga, T.; Tanatani, A.; Oh, K.; Moore, J. S. J. Am. Chem.
Soc. 2002, 124, 5934. (b) Tanatani, A.; Hughes, T. S.; Moore, J. S.
Angew. Chem., Int. Ed. 2002, 41, 325. (c) Petitjean, A.; Cuccia, L. A.;
Schmutz, M.; Lehn, J. M. J. Org. Chem. 2008, 73, 2481.
(7) Sugiura, M.; Tokudomi, M.; Nakajima, M. Chem. Commun. 2010,
46, 7799.
(8) (a) Berl, V.; Huc, I.; Khoury, R.; Krische, M. J.; Lehn, J.-M.
Nature 2000, 407, 720. (b) Gan, Q.; Bao, C.; Kauffmann, B.; Grel
A.; Xiang, J.; Liu, S.; Huc, I.; Jiang, H. Angew. Chem., Int. Ed. 2008, 47,
1715. (c) Baptiste, B.; Zhu, J.; Haldar, D.; Kauffmann, B.; Leger, J.-M.;
Huc, I. Chem.Asian J. 2010, 5, 1364.
(9) Jiang, H.; Leger, J.-M.; Huc, I. J. Am. Chem. Soc. 2003, 125, 3448.
́
ard,
́
́
(10) (a) Berni, E.; Kauffmann, B.; Bao, C.; Lefeuvre, J.; Bassani, D.
M.; Huc, I. Chem.Eur. J. 2007, 13, 8463. (b) Berni, E.; Garric, J.;
Lamit, C.; Kauffmann, B.; Leger, J.-M.; Huc, I. Chem. Commun. 2008,
́
1968.
(11) Association constants (Ka) were obtained using the
WINEQNMR software. See: Hynes, M. J. J. Chem. Soc., Dalton
Trans. 1993, 311.
(12) (a) Hou, J.-L.; Shao, X.-B.; Chen, G.-J.; Zhou, Y.-X.; Jiang, X.-K.;
Li, Z.-T. J. Am. Chem. Soc. 2004, 126, 12386. (b) Maurizot, V.; Dolain,
C.; Huc, I. Eur. J. Org. Chem. 2005, 1293. (c) Goto, H.; Furusho, Y.;
Yashima, E. J. Am. Chem. Soc. 2007, 129, 9168. (d) Waki, M.; Abe, H.;
Inouye, M. Angew. Chem., Int. Ed. 2007, 46, 3059.
(13) For examples of handedness inversion, see: (a) Fujiki, M. J. Am.
Chem. Soc. 2000, 122, 3336. (b) Sakurai, S.-i.; Okoshi, K.; Kumaki, J.;
Yashima, E. J. Am. Chem. Soc. 2006, 128, 5650. (c) Okoshi, K.; Sakurai,
S.-i.; Ohsawa, S.; Kumaki, J.; Yashima, E. Angew. Chem., Int. Ed. 2006,
45, 8173. (d) Johnson, R. S.; Yamazaki, T.; Kovalenko, A.; Fenniri, H.
J. Am. Chem. Soc. 2007, 129, 5735. (e) Meudtner, R. M.; Hecht, S.
Angew. Chem., Int. Ed. 2008, 47, 4926. (f) Peterca, M.; Imam, M. R.;
Ahn, C.-H.; Balagurusamy, V. S. K.; Wilson, D. A.; Rosen, B. M.;
Percec, V. J. Am. Chem. Soc. 2011, 133, 2311. (g) Akine, S.; Hotate, S.;
Nabeshima, T. J. Am. Chem. Soc. 2011, 133, 13868.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental details for synthetic procedures, spectroscopic
data, and CIFs for (2)2 and (2)2⊃(3)2. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
i.huc@iecb.u-bordeaux.fr
(14) Delsuc, N.; Kawanami, T.; Lefeuvre, J.; Shundo, A.; Ihara, H.;
Takafuji, M.; Huc, I. ChemPhysChem 2008, 9, 1882.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Conseil Reg
(predoctoral fellowship to Q.G.) and by ANR (Grant ANR-09-
́
ional d’Aquitaine
́
BLAN-0082-01). We thank Ms. Axelle Grelard for her
assistance with NMR measurements.
REFERENCES
■
(1) (a) Mathias, J. P.; Simanek, E. E.; Seto, C. T.; Whitesides, G. M.
Angew. Chem., Int. Ed. Engl. 1993, 32, 1766. (b) Paraschiv, V.; Crego-
Calama, V.; Ishi-i, T.; Padberg, C. J.; Timmerman, P.; Reinhoudt, D.
N. J. Am. Chem. Soc. 2002, 124, 7638. (c) Korevaar, P. A.; George, S.
J.; Markvoort, A. J.; Smulders, M. M. J.; Hilbers, P. A. J.; Schenning, A.
P. H. J.; De Greef, T. F. A.; Meijer, E. W. Nature 2012, 481, 492.
(2) (a) Hasenknopf, B.; Lehn, J.-M.; Boumediene, N.; Dupont-
Gervais, A.; Van Dorsselaer, A.; Kneisel, B.; Fenske, D. J. Am. Chem.
Soc. 1997, 119, 10956. (b) Hasenknopf, B.; Lehn, J.-M.; Boumediene,
N.; Leize, E.; Van Dorsselaer, A. Angew. Chem., Int. Ed. 1998, 37, 3265.
(c) Levin, M. D.; Stang, P. J. J. Am. Chem. Soc. 2000, 122, 7428.
D
dx.doi.org/10.1021/ja3064364 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX