Hybridization of long pyridine-dicarboxamide oligomers into multi-turn double helices: Slow strand association and dissociation, solvent dependence, and solid state structures

Article abstract of DOI:10.1002/asia.200900713

Oligoamides of 2,6-diamino- pyridine and 2,6-pyridinediearboxylie aeid eomprised of 5, 7, 9, 11, or 13 units and bearing 4-isobutoxyehains on all pyridine rings and tert-butyl-earba- mate terminal groups have been synthesized stepwise, along with an 11 mer having benzyl-earbamate terminal groups. The erystal strueture of all five Boe-terminated eompounds has been obtained and shows a highly regular and eonserved double helieal hybridization motif of up to 3 eomplete turns for the 13 mer. Four pyridine units span one helieal turn and define a helix piteh of ca 7 A. Solution studies in CDCl₃ demonstrated that the Boe-terminated oligomers strongly hybridize in this solvent, and that K_dim values incerease with oligomer length. The K_dim values are 31000 and 7x 10⁵ Lmol^-1 for the 7 mer and the 9 mer, respeetively, and are too high to be measured by NMR for the 11 mer and the 13 mer. Hybridization and dissoeiation kineties at 2 mM proeeed at deereasing rates upon inereasing oligomer length. The rate was faster than minutes for the 7 mer, of the order of hours for the 9 mer, and days for the 11 mer and 13 mer. The same trend was observed in [D₅]pyridine but with eonsiderably lower K_dim values and faster kineties. The benzylearbamate 11 mer was also found to hybridize into a double helix but with redueed K_dim values and faster kineties eompared to its Boe-terminated analogue. Combined with previous studies, the results presented here frame a global understanding of the hybridization of these pyridineearboxamide oligomers and provide useful guidelines for the design of other artifi- eial double heliees.

Full text of DOI:10.1002/asia.200900713

FULL PAPERS
DOI: 10.1002/asia.200900713
Hybridization of Long Pyridine-Dicarboxamide Oligomers into Multi-Turn
Double Helices: Slow Strand Association and Dissociation, Solvent
Dependence, and Solid State Structures
Benoit Baptiste,[b] Jiang Zhu,[a] Debasish Haldar,[a] Brice Kauffmann,[a]
Jean-Michel Lꢀger,[b] and Ivan Huc*[a]
Abstract: Oligoamides of 2,6-diamino-
pyridine and 2,6-pyridinedicarboxylic
acid comprised of 5, 7, 9, 11, or 13
units and bearing 4-isobutoxychains on
all pyridine rings and tert-butyl-carba-
mate terminal groups have been syn-
thesized stepwise, along with an 11 mer
minated oligomers strongly hybridize
in this solvent, and that Kdim values in-
crease with oligomer length. The Kdim
values are 31000 and 7ꢁ105 LmolÀ1 for
the 7 mer and the 9 mer, respectively,
and are too high to be measured by
NMR for the 11 mer and the 13 mer.
Hybridization and dissociation kinetics
at 2 mm proceed at decreasing rates
upon increasing oligomer length. The
rate was faster than minutes for the
7 mer, of the order of hours for the
9 mer, and days for the 11 mer and
13 mer. The same trend was observed
in [D5]pyridine but with considerably
lower Kdim values and faster kinetics.
The benzylcarbamate 11 mer was also
found to hybridize into a double helix
but with reduced Kdim values and faster
kinetics compared to its Boc-terminat-
ed analogue. Combined with previous
studies, the results presented here
frame a global understanding of the hy-
bridization of these pyridinecarboxa-
mide oligomers and provide useful
guidelines for the design of other artifi-
cial double helices.
having
benzyl-carbamate
terminal
groups. The crystal structure of all five
Boc-terminated compounds has been
obtained and shows a highly regular
and conserved double helical hybridi-
zation motif of up to 3 complete turns
for the 13 mer. Four pyridine units
span one helical turn and define a helix
pitch of ca 7 ꢀ. Solution studies in
CDCl3 demonstrated that the Boc-ter-
Keywords: chirality · helical struc-
tures · structure elucidation · supra-
molecular chemistry · X-ray diffrac-
tion
Introduction
[a] Dr. J. Zhu,++ + Dr. D. Haldar,# + Dr. B. Kauffmann, Dr. I. Huc
Universitꢂ de Bordeaux—CNRS UMR 5248 and UMS 3033
Institut Europꢂen de Chimie et Biologie
Nature has found very diverse applications of the hybridiza-
tion of organic molecular strands into double or triple heli-
cal architectures. In nucleic acids, sequence selective hybrid-
ization is key to genetic information storage,[1] duplication,
and transcription. In the triple helix of collagen,[2] some po-
2 rue Robert Escarpit, 33067 Pessac (France)
Fax : (+33)540-002-215
E-mail: i.huc@iecb.u-bordeaux.fr
[b] B. Baptiste,+ J.-M. Lꢂger
AHCTUNGTRENNUNG
lysachharides[3] or in coiled coils (e.g., keratin[2a,4]) hybridiza-
Universitꢂ Victor Sꢂgalen Bordeaux 2
EA 4138 - Pharmacochimie
146 rue Lꢂo Saignat, 33076 Bordeaux Cedex (France)
tion contributes to the stiffness and mechanical properties
of the polymer fibers. In the bacterial peptide gramicidin D,
hybridization gives rise to a double stranded b-helix archi-
tecture that can bind metal ions in its cavity.[5] Except in the
case of short oligomers such as Gramicidin D, hybridization
in biopolymers occurs over extended lengths which confers
the hybrids with considerable thermodynamic and kinetic
stability. Nevertheless, nucleic acid duplexes or collagen gels
melt at high temperatures as a result of strand dissociation,
which reveals a high negative entropy of hybridization.[1b]
[
++] Present address:
Division of Chemistry, North Sichuan Medical College
234 Fujiang Lu, Nanchong 637007 (China)
[#] Present address:
Department of Chemical Sciences
Indian Institute of Science Education and Research, Kolkata.
Haringhata, Mohanpur, West Bengal 741252 (India)
[+] These authors contributed equally to this work
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.200900713.
1364
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Strand association in those cases is thus a highly cooperative
process.
during hybridization is compensated by the large enthalpy
gain that results from interstrand p–p stacking in the du-
plex.[9i] The study reported here aimed at a thorough struc-
tural investigation of the double helices of long pyridine oli-
goamides sequences. It was carried out in order to explore
whether structural variability as observed for shorter se-
quences persist with more than nine units, and also with the
hope that structural insights would help elucidate the trend
of affinity versus strand length. As shown in the following,
we found compelling crystallographic evidence that hybridi-
zation of pyridinecarboxamide oligomers may occur over ex-
tended lengths. Additionally, solution-phase investigations
of the newly synthesized oligomers allowed us to character-
ize the high solvent dependence of the hybridization process
and the dramatic effect of strand length on the association
and dissociation kinetics which span several orders of mag-
nitude upon adding a few units. In turn, the observation of
slow hybridization kinetics for longer oligomers led us to re-
investigate the association constants and to revise our initial
proposal that hybridization drops upon strand length in-
crease. Indeed, we now unambiguously show that duplex
stability considerably increases upon strand length increase.
Finally, we show that bulky terminal Boc groups favor hy-
bridization thermodynamics and slow down hybridization ki-
netics. Both these findings have implications regarding the
hybridization mechanism.
Inspired by the remarkable efficiency of biopolymer hy-
bridization, chemists have synthesized a number of non-nat-
ural molecular strands that also possess the ability to inter-
twine into multiple helices.[6] Prominent examples include
metal coordination-based helicates,[7] a plethora of nucleic
acid analogues,[8] aromatic oligoamides,[9] d,l-peptides in-
spired from gramicidin D,[10] oligo-resorcinols,[11] ethynylheli-
cene oligomers,[12] m-terphenyl backbone oligomers exploit-
ing amidinium-carboxylate salt bridges,[13] alternate sequen-
ces of aromatic hydrogen-bond donors and acceptors,[14] or
some mannose-derived d-peptides.[15] Mention should also
be made of sheet-like structures in which hybridization
occurs without helical winding.[16,17] These synthetic model
systems allow the exploration of hybridization patterns
which differ from those in Nature and thus shed light on the
essential features of biopolymers that govern their hybridi-
zation behavior. Indeed, the thermodynamic and kinetic be-
haviors of several synthetic duplex families apparently differ
significantly from those of their natural counterparts. The
thermodynamic stability of non natural hybrids was shown
in several cases to level off or even drop upon strand length
increase.[9c,11d,17] Kinetic stability may reach very high levels,
which allows the chromatographic separation of artificial du-
plexes from single strands or other duplexes despite the
small size of these non-covalent assemblies.[9j,l,12c] Addition-
ally, hybridization mechanisms other than a simple zipper-
like process have been proposed. For example, molecular
modeling studies hinted at a slippage mechanism that in-
volves a series of roller-coaster-like discrete steps[18] in
which the intramolecular interactions within a single helix
are replaced by intermolecular interactions in the duplex
when the tail of one of the strands proceeds inside the other
single helix in an eddy-like process. Such mechanisms may
eventually allow direct strand exchange between two du-
plexes without dissociation[11d] and the involvement of pre-
equilibration with structural intermediates prior to reaching
thermodynamic equilibrium.[12c]
These intriguing results concern different artificial duplex
families and obviously call for further physical and structur-
al investigations of the hybridization processes. In particular,
because of the lack of accessibility of well defined oligomers
longer than just a few units, previous studies have generally
focused on helices spanning only one or two turns and data
concerning longer sequences are scarce (for notable excep-
tions, see references [9c, 11d, and 12c]). In the case of pyri-
dine oligoamides, we have initially reported the unexpected
trend that double stranded hybrid stability tends to drop
when oligomer length increases.[9c] We have also shown that
short sequences (5 to 9 monomers) are able to form several
types of duplex architectures in which the two strands may
be offset by zero,[9d] one,[9b,k] or two[9a,f] monomers, consis-
tent with the dynamics observed in solution[9a] and with cal-
culations.[18] Recently, we demonstrated that hybridization is
a mostly enthalpically driven process in which the high cost
of spring-like extension of single helical monomers observed
Results and Discussion
Molecular Design
In previous studies, the long pyridinecarboxamide oligomers
investigated were equipped with n-decyloxy side chains
which provide excellent solubility in organic solvents but
prevent the growth of high quality single crystals for solid-
state structural investigations. In the meantime, studies of
related single helical quinoline carboxamide oligomers
showed that isobutoxy side chains provide both high solubil-
ity in chlorinated and aromatic solvents, and high crystallini-
ty,[19] even for very large oligomers.[20] We thus decided to
undertake the synthesis of pyridinecarboxamide oligomers
5a–13a, possessing from 5 to 13 units and having isobutoxy
side chains in position 4 of each pyridine ring. These sub-
stituents diverge from the helices and are exposed to the
solvent. They also strongly influence the hybridization pro-
cess: oligomers that carry alkoxy groups on every residue
hybridize into double helices possess Kdim values that are
several orders of magnitude larger than oligomers with no
such substituents.[9a,b] However, it was also shown that the
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1365

I. Huc et al.
FULL PAPERS
effect of alkoxy residues on hybridization does not depend
on the nature of the alkyl group, and so n-decyloxychains
have the same effect as methoxy groups.[9k] It was thus as-
sumed that isobutoxy groups would result in a similar pro-
pensity towards hybridization.
Synthesis
Thanks to the experience acquired in the stepwise prepara-
tion of various families of long arylamide oligomers, we
opted for a divergent approach different from that previous-
ly used in oligopyridine-dicarboxamide synthesis, and made
use of the efficient activation of carboxylic acids as acid
chlorides under neutral conditions with 1-chloro-N,N,2-tri-
methylpropenylamine[21] in the presence of Boc amine pro-
tection. This allowed us to avoid benzylcarbamate protec-
tion (Cbz) which, in our experience, sometimes proved to be
difficult to remove from large folded oligomers. The syn-
thetic schemes were designed taking into account that heli-
cal folding creates a severe steric hindrance,[19b,22] and our
observation that this is particularly acute if reagents also un-
dergo hybridization into stable double helices. Thus, our
choice was to attach “small” units that do not hybridize
(dimer acid chlorides) at the ends of longer diamine oligo-
mers. Much higher yields were observed upon coupling a
small acid chloride to a large diamine than coupling a small
amine to a large diacid chloride. Thus, the synthesis pro-
ceeded through the early desymmetrization of diester 1a
into monoacid 1b and of diamine 1d into monoamine 1e
(Scheme 1). Coupling between 1e and 1d afforded the key
intermediate dimer 2a, which was saponified to give 2b.
After activation as an acid chloride, 2b was used repeatedly
to elongate the diamine precursors by four pyridine units
(Scheme 2). Thus, diamine monomer 1b was elongated to
pentamer 5a, which after Boc deprotection, was converted
to nonamer 9a. One more deprotection/coupling round led
to tridecamer 13a. Similarly trimer 3a was converted to
heptamer 7a and then to the undecamer 11a.
Scheme 2. a) 2b or 2d, 1-chloro-N,N,2-trimethylpropenylamine, CH2Cl2,
then DIEA, CH2Cl2; b) TFA, CH2Cl2.
Solid-State Characterization
As anticipated, the isobutoxy group in position 4 of the pyri-
dine rings conferred the oligomers with high solubility in
chlorinated and aromatic solvents and with a remarkable
propensity to crystallize. Single crystals of the five com-
pounds, namely 5a, 7a, 9a, 11a, and 13a, suitable for X-ray
diffraction analysis were all obtained upon diffusion of
MeOH into a CHCl3 solution of the compounds. Alterna-
tively, the same crystals could be obtained upon slow evapo-
ration of a methanol–chloroform solution of the compounds.
In some cases, centimeter long crystals grew in a few days.
All structures were solved using ab initio methods despite
the large asymmetric units of some of them and all revealed
a double helical architecture (Figure 1).
The crystallographic parameters are reported in Table 1.
We have found that the level of crystallographic symmetry
increases with oligomer length. The 5 mer and 7 mer crystal-
¯
lized in the P1 space group with two double helices in the
asymmetric unit whereas the 9 mer and the 11 mer were ob-
¯
tained in the P1 and Pbca space groups, respectively, with
only one double helix in the
asymmetric unit. The 13 mer
was solved in the monoclinic
C2/c group and, in this case, the
two strands of the duplex are
related by a crystallographic C2
axis and the asymmetric unit is
reduced to one strand.
The hybridization motifs of
5a–13a are strictly identical.
Each strand of the duplex spans
one helical turn every four pyri-
dine units for a vertical rise per
turn of 7 ꢀ. Direct contacts be-
tween the two strands consist
essentially of face-to-face aro-
matic stacking. The two strands
of each duplex are helically
Scheme 1. a) 1- NH3 gas, MeOH, 1 h, RT; 2- KOH, Br2, dioxane, 08C for 10 min then 258C for 30 min and
708C for 1.5 h, 78% yield; b) NaOH, MeOH/dioxane, 2 h, 258C, 72% yield; c) NaOH, MeOH/dioxane/water,
2 h, RT, 95% yield; d) lithium bis(trimethylsilyl)amide, ditert-butyldicarbonate, THF, 3.5 h, RT, 52% yield;
e) lithium bis(trimethylsilyl)amide, benzyl chloroformate, THF, 3.5 h, RT, 56% yield; f) 1- 1c, SOCl2, 30 min,
reflux; 2- 1e, DIEA, THF ; 12 h., 258C, 75% yield; g) 1- 1d, SOCl2, 30 min, reflux 2- 1e, DIEA, THF, 12 h.,
258C, 81% yield for 2a; 1- 1d, oxalyl chloride, DMF, toluene, 2 h, RT; 2- 1 f, DIEA, THF, 12 h, RT, 98% for
2c.
1366
www.chemasianj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Hybridization of Oligomers into Double Helices
motions) within the duplex and
that a dominant structural pat-
tern emerges for extended
lengths. For the 13 mer, these
features result in a 2.6 nm long
duplex having
a
molecular
weight above 5.3 kDa in which
each strand spans over three
helical turns. The regularity of
the helical pattern, from the
5 mer to the 13 mer also sug-
gests that it might be propagat-
ed along much longer (e.g.,
polymeric) sequences without
any alteration of the interac-
tions.
The polar cavity of the
double helix channel is filled by
water molecules, one for the
pentamer, and an additional
water molecule for every two
additional pyridine units on
Figure 1. Side view of the solid-state structures in stick (top) and CPK (bottom) representations of: a) pentam-
er 5a; b) heptamer 7a; c) nonamer 9a; d) undecamer 11a; e) tridecamer 13a. The two strands of the double
helices are shown as red and blue and water molecules as yellow sticks. Isobutoxy side chains, hydrogen atoms
and solvent molecules (except water included in the channel) have been omitted for clarity. All helices are dis-
played in the same orientation, as illustrated by the identical appearance of their lower part.
offset with respect to one another. This offset is constant for
all duplexes, which is about one aromatic ring, that is, one
quarter of a turn. Thus, ring 1 of strand A stacks over ring 2
of strand B, and reciprocally. The conservation of the hy-
bridization motif is in contrast with previous observations
that the two strands may adopt different relative positions
within a duplex, which results in offsets of zero, one, or two
pyridine units. These observations were made on shorter
strands (5–9 units) and the highest variability was observed
for pentameric oligomers. Thus, it appears that elongation of
the strands reduces their relative mobility (i.e., screw-like
each strand (Figure 1). Additionally, the extremities of the
channel are closed by the OH group of a bound methanol
molecule that occupies a position similar to those occupied
by water (not shown). We notice here the high conservation
of the position of water molecules throughout the series.
Each water oxygen atom may potentially hydrogen bond to
the four amide protons of the two pyridinecarboxamide
units, one on each strand, oriented perpendicularly to each
other (Figure 2). Other possible interactions involve hydro-
gen-bonding of the water hydrogen atoms to endocyclic ni-
trogens of the neighboring 2,6-diaminopyridine units
Table 1. Crystallographic parameters for 5a–13a.
5a
7a
9a
11a
13a
Solvent of cryst.
Formula
Dimensions [mm]
Color
Crystal system
Space group
Z (Z’)
CHCl3/MeOH
C59H79N11O13
0.2ꢁ0.1ꢁ0.1
colorless
triclinic
¯
P1
CHCl3/MeOH
C79H103N15O17
0.1ꢁ0.05ꢁ0.05
colorless
triclinic
¯
P1
CHCl3/MeOH
C99H127N19O21
0.05ꢁ0.02ꢁ0.02
colorless
triclinic
¯
P1
CHCl3/MeOH
C119H151N23O25
0.1ꢁ0.05ꢁ0.02
colorless
orthorhombic
Pbca
CHCl3/MeOH
C139H175N27O29
0.2ꢁ0.1ꢁ0.1
colorless
monoclinic
C2/c
2 (4)
2 (4)
2 (2)
8 (2)
8 (1)
a [ꢀ]
b [ꢀ]
c [ꢀ]
a [8]
b [8]
g [8]
21.0120 (10)
24.493 (2)
29.916 (2)
111.175 (5)
103.361 (5)
93.590 (5)
193 (2)
13785.8 (16)
1.205
1.54178
20.5203 (8)
32.7070 (12)
34.4280 (12)
66.740 (2)
75.940 (2)
72.402 (2)
133 (2)
20032.7 (13)
1.187
1.54178
20.865 (4)
21.927 (4)
33.746 (7)
73.59 (3)
76.52 (3)
61.90 (3)
100 (2)
12973 (4)
1.184
0.80000
3.46<q<23.58
42570
40.4820 (10)
39.6920 (10)
40.8340 (10)
90
90
90
100 (2)
65613(3)
1.271
0.80000
4.13<q<30.00
458594
38.8452 (10)
22.7274 (3)
46.2275 (18)
90
115.5700
90
113 (2)
36814.8 (18)
1.215
1.54178
6.34<q<50.43
112324
18787
1.020
T [K]
V [ꢀ3]
1 [gcmÀ3
l [ꢀ]
]
q measured
Reflns measured
unique reflections
GOF
6.51<q<54.24
30856
30856
6.50<q<39.97
158533
24106
21957
1.102
32604
1.018
1.013
1.016
R1 (all data)
wR2 (all data)
CCDC#
0.2109
0.5267
757948
0.1720
0.4618
757946
0.2064
0.5208
757944
0.1565
0.4864
757945
0.2009
0.5417
757947
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1367

I. Huc et al.
FULL PAPERS
Figure 2. Top views of the stick representation of fragments of (9a)2 in
the crystalline state and potential interactions (dashed lines) between
each color-coded strand of the duplex and a water molecule: a) potential
hydrogen bonds between the water hydrogens and endocyclic nitrogen
atoms belonging to neighboring 2,6-diaminopyridine rings; b) potential
hydrogen bonds between the water oxygen atom and neighboring carbox-
amide hydrogens. Atomic distances are given in Angstroms. Isobutoxy
side chains and hydrogen atoms have been omitted for clarity.
Figure 3. Part of the 400 MHz 1H NMR spectra at 208C of 2 mm CDCl3
solutions showing the amide resonances of: a) 7a; b) 9a; c) 11a and
d) 13a shortly after dissolving the crystals. Signals belonging to the single
helix are marked with circles.
(Figure 2). These latter interactions are not firmly estab-
lished because the positions of water hydrogen atoms in the
structures have not been determined experimentally. Each
water molecule potentially creates bridging interactions be-
tween the two strands. However, it has not been established
whether these interactions ultimately stabilize double helical
architectures. Indeed, water is also bound in the channel of
single helical conformers,[23] and water has been shown to
promote duplex dissociation in solution.[9b]
helices is in agreement with the solid-state structures, but it
suggests a very high Kdim for these compounds, which is in
apparent contradiction with our previous finding.[9c] This dis-
crepancy cannot be explained by the structural differences
between this series of compounds and the previous one.
Indeed, the replacement of n-decyloxy side chains by isobu-
toxy side chains was not expected to have any significant
effect, and the replacement of the terminal benzylcarba-
mates by tert-butylcarbamates, though it may change duplex
stability, can hardly give rise to an apparent enhancement of
Kdim of over 4 orders of magnitude in the case of 13a.
Rather, we suspected that the equilibrium between single
and double helices may be very slow in CDCl3 for long olig-
omers and that either the spectra in Figure 3 or those on
which we based our previous calculations,[9c] or both, may
not reflect equilibrium and the actual Kdim value. This possi-
bility was supported by a study of the hybridization of an
11 mer into a double helix comprised of pyridine, quinoline,
and diaza-anthracene units, for which single-double helix
equilibrium is so slow that the two species can be separated
on silica gel chromatography and equilibrium can be
reached in solution over several days at room temperatur-
e.[9j] To clarify this, we systematically investigated the associ-
ation and dissociation kinetics of the double helices of 7a,
9a, 11a, and 13a.
As shown in Figure 4, the proportion of (7a)2 drops signif-
icantly with a decrease in concentration and also an increase
in temperature. However, heating and cooling cycles are fol-
lowed by a rapid return to the original NMR spectra, which
indicates that the equilibrium between single and double
helices is fast with respect to the time required to fully ac-
quire the NMR spectra. Thus, Kdim =31000 LmolÀ1 can be
reliably calculated at 208C for 7a. This value is of the order
of those previously reported for other pyridinecarboxamide
heptamers bearing alkoxy groups (R) as side chains and ter-
minal amine residues (82000 LmolÀ1 with R=methoxy)[9k]
or decanoyl residues (65000 LmolÀ1 with R=decyloxy).[9a]
Lower values were observed when the terminal group was
Solution Studies in CDCl3
The association of 5a–13a into double helical architectures
in the solid state underlines the high propensity of the pyri-
dinecarboxamide oligomers to hybridize. However, the crys-
tallization of 11a and 13a as double helices came as a sur-
prise because of our earlier finding that Kdim values quickly
drop upon increasing strand length, leading to a value of the
order of 65 LmolÀ1 for the 13 mer having decyloxy side
chains and Cbz terminal groups.[9c] NMR studies were thus
undertaken to quantify the double helix formation in solu-
tion. NMR is convenient because single helices and double
helices are in slow exchange on the NMR time scale for
oligomers of seven pyridine units or more and give rise to
two sets of signals that can be integrated to calculate Kdim
.
Several spectroscopic features specific to either single or
double helices help in the assignment of the signals. Below a
certain temperature, the lower symmetry of the double helix
gives rise to twice as many signals as the single helix.[9a,c]
The proportion of single helix increases upon heating and
dilution while intermolecular aromatic stacking within
double helices results in stronger ring current effects, which
causes upfield shifts of the double helix signals.
1
The H NMR spectrum of heptamer 7a in CDCl3 shows
that this compound is largely hybridized at 2 mm (broad
signal) in this solvent but signals of the single helix are also
1
clearly visible (Figure 3a). On the other hand, the H NMR
spectra of 9a, 11a, and 13a all show a single set of sharp sig-
nals which can unambiguously be assigned to double helices
based on their multiplicity. The exclusive presence of double
1368
www.chemasianj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Hybridization of Oligomers into Double Helices
nals intensity then decreases over a couple of days but do
not completely disappear. The initial spectrum at 208C thus
appears not to reflect thermodynamic equilibrium but
simply the crystal content. Furthermore, upon incubating
freshly dissolved 9a at 208C, the single helix signals also
emerge after a couple of days without heating (not shown).
The final spectra of these two experiments are identical and
allow to unambiguously calculate Kdim to be 7ꢁ105 LmolÀ1
for 9a. The nonamer thus has a propensity to form double
helices about 20 times larger than the heptamer, as observed
in other series, but does so at much slower association and
dissociation rates. To investigate these kinetics, we prepared
a sample of 9a strongly enriched in single helix by first heat-
ing a solution in pyridine (vide infra), evaporating it, and re-
dissolving the sample in chloroform. Equilibrium was
reached within 1–2 days (Figure 6). Such slow kinetics were
Figure 4. Part of 400 MHz 1H NMR spectra in CDCl3 showing the amide
resonances of 7a at different concentrations and temperatures:
a) 12.8 mm at 208C; b) 2 mm at 208C; c) 2 mm at 408C; d) return to 2 mm
at 208C. Spectra b), c) and d) were recorded from the same 2 mm solu-
tion at regular time intervals (30 min). Signals assigned to the single helix
are marked with circles.
Cbz (1500 LmolÀ1 with R=decyloxy, 1000 LmolÀ1 with R=
methoxy).[9c]
We have found that 9a behaves differently, which is evi-
dent upon heating a solution of pure double helix, wherein
signals of the single helix appear above 508C and remain
upon cooling back to 208C (Figure 5). The single helix sig-
Figure 6. Time trace of duplex formation of 2 mm CDCl3 solutions of 9a
(white circles) and 11a (black circles) and of 11b (black rhombus) after
1
initial enrichment in single helix, as monitored by H NMR at 208C.
not observed before for other 9 mers bearing terminal
amino or Cbz groups. A possible explanation is that, in the
present case, the terminal Boc groups sterically hinder asso-
ciation and dissociation, which is in agreement with the slip-
page mechanism proposed for these processes.[18]
Similar investigations were attempted with 11a and 13a.
However, for these two longer oligomers, heating CDCl3
solutions to 508C or incubating over 6 months in a sealed
NMR tube at room temperature did not lead to any trace of
single helix. This may arise either from extremely slow ki-
netics of duplex dissociation or from a very high Kdim value.
To elucidate this, 11a was treated in the same way as 9a. A
sample of 11a was enriched in single helix by first heating a
solution in pyridine (vide infra), evaporating it, and redis-
solving the sample in chloroform. The proportion of single
helix as a function of time showed a sharp decrease
(Figure 6). Equilibrium was obviously not reached after a
week of incubation, and yet this shows that strand associa-
tion is not kinetically inert, which suggests that the absence
of change in the spectrum of freshly dissolved 11a and 13a
reflects high thermodynamic stability (the proportion of
single helices is negligible at mm concentration). Thus, Kdim
Figure 5. Part of the 400 MHz 1H NMR spectra in 2 mm CDCl3 solution
showing amide resonances of 9a upon changing temperature: a) 208C;
b) 408C; c) 508C; d) 408C; e) 208C; f) 208C after 7 days. Spectra were re-
corded from the same solution at regular time intervals of 15 min per
108C.
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1369

I. Huc et al.
FULL PAPERS
values for 11a and 13a appear to be too high to be mea-
sured by NMR in CDCl3. A few attempts to measure them
by UV spectroscopy also proved unsuccessful. We thus
wished to investigate hybridization in a solvent less favora-
ble to duplex formation than CDCl3. DMSO was often used
by us in the past for this purpose, but both 11a and 13a are
insoluble in this solvent. In the following, we show that
[D5]pyridine proved to be a good medium to promote
duplex dissociation.
Solution Studies in Pyridine
Upon dissolving 7a and 9a crystals (i.e., pure double heli-
ces) in [D5]pyridine, single helices were formed at rates too
fast to monitor by NMR. The final proportions allowed the
calculation of Kdim to be 10 and 19 LmolÀ1 for 7a and 9a,
respectively (Figure 7). Thus this solvent causes a dramatic
Figure 8. Time trace of duplex dissociation of 2 mm solutions of 11a
(white circles) and 13a (black circles) in [D5]pyridine monitored by
1H NMR at 208C. The initial state is a freshly dissolved crystalline
double helix.
firmed in [D5]pyridine, albeit at much lower levels. It is also
confirmed that rates of strand association and dissociation
drop by several orders of magnitude upon increasing strand
length, though at overall faster rates in [D5]pyridine than
CDCl3.
The Effect of End Groups
The dimerization constants of 7a and 9a in CDCl3 have
been shown to be 3.1ꢁ104 LmolÀ1 and 7ꢁ105 LmolÀ1, re-
spectively, and those of 11a and 13a are too high to mea-
sure, which means well above 106 LmolÀ1. These values are
to be compared to those we previously reported for ana-
logues of 7a–13a bearing decyloxy side chains and Cbz end
groups instead of isobutoxy and Boc, which are 1500, 5200,
650, and 65 LmolÀ1 from 7 mer to 13 mer, respectively.[9c]
The discrepancy at the 7 mer and 9 mer level may be as-
cribed to an effect of the Boc versus Cbz end-groups, which
results in Kdim values 20 fold (7 mer) and 130 fold (9 mer)
larger in favor of Boc end groups. A possible origin of this
effect is that bulky Boc end groups may have a destabilizing
effect on the single helices because of steric hindrance,
which facilitates their spring-like extension and thus de-
creases the associated enthalpy cost during hybridization.[9i]
However, the discrepancy at the 11 mer and 13 mer level is
far too large to simply reflect an effect of end groups which
would be expected to result in a DDG independent of strand
length (i.e., a constant ratio of Kdim values). Instead, we sus-
pected that our initially reported values were erroneous be-
cause they were measured away from equilibrium. The ki-
netics of strand dissociation for 11 mers and 13 mers are
slow, and the proportions used to calculate Kdim may instead
have reflected snapshots on the way to equilibrium.[24,25] In
order to lift any doubt on this issue, we prepared 11b, an an-
alogue of 11a bearing terminal Cbz groups (Scheme 1 and
2).
Figure 7. Part of the 300 MHz 1H NMR spectra at 208C of 2 mm
[D5]pyridine solutions showing the amide resonances of: a) 7a; b) 9a;
c) 11a and d) 13a at equilibrium. Signals belonging to the single and
double helix are respectively marked with white and black circles.
decrease in double helix stability (or an increase in single
helix stability), and also allows much faster strand dissocia-
tion. This effect was confirmed in the case of 11a and 13a.
Duplexes dissociated to reach equilibrium in only one day
for 11a and 3 days for 13a (Figure 8). That equilibrium is
indeed reached could be validated through a heating and
cooling cycle which led to a further increase of the propor-
tion of single helix (heating) and a return to the equilibrium
state (cooling). The time scale of strand association moni-
tored after heat-induced dissociation was found to be the
same as the time scale of strand dissociation upon dissolving
crystals of double helices (not shown). From the NMR spec-
tra Kdim was found to be 43 and 307 LmolÀ1 for 11a and
13a, respectively. However, the kinetics could not be fitted
to a simple model because the reaction is reversible. Thus,
the overall trend observed in chloroform, that double helix
formation is enhanced upon increasing strand length, is con-
Crystals of 11b were grown from a liquid–liquid diffusion
of methanol into a chloroform solution. Upon dissolving this
material in CDCl3, an NMR spectrum very similar to that of
1370
www.chemasianj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Hybridization of Oligomers into Double Helices
11a (Figure S6 in the Supporting Information) was recorded.
Only one species was observed and the chemical shift values
and multiplicity of the signals were clearly indicative of a
double helical architecture. After several days, no single
helix was detected and the spectrum remained unchanged.
However, upon dissolving 11b in [D5]pyridine, rapid dissoci-
ation occurred, which led to an almost quantitative conver-
sion to single helix that indicated a very low Kdim in this sol-
vent. After evaporation of pyridine and dissolving 11b in
CDCl3, the single helix was progressively transformed
almost completely into a double helix over a period of 1 day
(Figure 9). This hybridization process was thus considerably
more rapid than for 11a (Figure 6).
Conclusions
In summary, we have validated our assumption that isobu-
toxy side chains provide both high solubility and high crys-
tallinity and have shown compelling crystallographic evi-
dence that hybridization of pyridinecarboxamide oligomers
may occur over extended strand lengths and follow a highly
conserved structural motif. Solution studies have revealed
that increasing strand length results in a strong increase of
the propensity to hybridize in double helical structures. In-
creasing strand length also causes a dramatic decrease in the
strand association and dissociation kinetics, where time-
scales range from minutes to weeks upon elongation of a
heptamer to a tridecamer. These effects have been observed
in CDCl3 in which hybridization is favored (Kdim ranging
from 104 to above 106 LmolÀ1). They also persist in
[D5]pyridine where hybridization is much weaker (Kdim rang-
ing from 10 to 300 LmolÀ1) and strand association and disso-
ciation proceeds at faster rates. This study underlines the po-
tentially critical importance of kinetic aspects when assess-
ing the stability of non-covalent assemblies even though of
relatively small size. Indeed, for having overlooked the slow
kinetics of hybridization of the longest strands, two Kdim
values published by us were found to be erroneous.[9c]
We also observed that bulky Boc terminal groups result in
a one to two orders of magnitude enhancement of Kdim
when compared to terminal flat Cbz groups. This effect was
tentatively assigned to a facilitated spring-like extension of
single helices. Bulky Boc terminal groups also result in
slower kinetics of strand association and dissociation than
terminal Cbz groups, which is in agreement with the current
proposed slippage mechanism of hybridization, in which one
single helical strand proceeds within another one in a screw-
like motion.
1
Figure 9. Part of the 300 MHz H NMR spectra showing the formation of
The hybridization of pyridinecarboxamide oligomers into
double helices results from a subtle combination of various
non independent parameters, namely, ability to extend like
springs, oligomer length, and nature of side chains and of
terminal groups. All of them critically influence both the
thermodynamic and kinetic aspects. Combined with previous
investigations, the results presented here have framed a
global understanding of this complex and intriguing phe-
nomenon and provide useful guidelines for the design of
other artificial double helices.
double helix from a 2 mm solution of Z-11 mer-Z. The selected window
presents the amide resonances as a function of time. The solution was
first heated in [D5]pyridine to obtain a totally dissociated state and
cooled to room temperature. Then, the spectra were recorded at 208C in
CDCl3.
Altogether, these result indicate that: (i) oligomers with Z
end-groups have a propensity to hybridize one to two orders
of magnitude lower that those with Boc end groups both in
CDCl3 and [D5]pyridine; (ii) the Kdim of 11b in CDCl3 is
nevertheless above 106 LmolÀ1, in disagreement with that re-
ported for an analogue of 11b having decyloxy chains;[9c]
(iii) the kinetics of association and dissociation are faster
with Z-terminated oligomers, but may nevertheless take
over a day for an 11 mer. It thus clearly appears that large
end groups slow down kinetics and also shift equilibria in
favor of double helices. Also, we conclude that our low Kdim
value reported for a 13 mer and an 11 mer were in fact mea-
sured out of equilibrium because of the slow kinetics in-
volved. This may have been enhanced by the decyloxy
chains of the oligomers investigated at the time which might
have expectedly resulted in slower kinetics than with isobu-
toxy chains, though without alteration in thermodynamics.
Experimental Section
General procedure and materials. Unless specified, materials were ob-
tained from commercial suppliers and used without further purification.
NMR spectra were recorded on a Bruker-Avance 400 NB US NMR spec-
trometer by means of a 5 mm direct QNP 1H/X probe with gradient ca-
pabilities. Samples were not degassed. Chemical shift are reported in
ppm and are calibrated against residual solvent signals of CDCl3 (d=
7.26, 77.2 ppm), [D6]DMSO (d=2.50, 39.4 ppm), [d5]pyridine (d=8.74,
7.58, 7.22 ppm). All coupling constants are reported in Hz. Silica gel
chromatography was performed using Merck Kieselgel Si 60. Electro-
nspray ionization (ESI) mass spectra were obtained in the positive ion
mode.
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1371

I. Huc et al.
FULL PAPERS
Dimethyl 4-isobutyloxy-2,6-pyridine dicarboxylate 1a. Anhydrous di-
methyl chelidamic ester[26] (8.3 g, 34 mmol, 1equiv) was dissolved in
60 mL of freshly distilled DMF and heated to 1208C under a N2 atmos-
phere. K2CO3 (14.0 g, 102 mmol, 3 equiv) was added and heating at
1208C was continued for 1.5 h. After cooling down to 708C, isobutyl
iodide (6.2 mL, 51 mmol, 1.5 equiv) was added and the mixture was
stirred at 708C for 3 h. The solution was cooled to room temperature.
Solid residues were filtered off. Toluene (80 mL) and water (50 mL) were
added. The mixture was extracted twice with toluene (80 mL). The com-
bined organic layers were dried with Na2SO4 and evaporated under
vacuum to obtain the product as a white solid. Yield 8.7 g (90%); m.p.:
15 min at 258C. After 10 min,
a
solution of di-tert-butyldicarbonate
(1.72 g, 1 equiv) in dry THF (10 mL) was added slowly over 20 min. The
reaction was allowed to proceed at 258C for 3 h. The solvent was re-
moved and the residue was dissolved in CH2Cl2 (50 mL) and extracted
with water (2ꢁ50 mL). The combined organic layers were evaporated
and the residue was purified by column chromatography (SiO2) with cy-
clohexane/ethylacetate/triethyl amine (70:25:5, vol/vol/vol) as eluent to
afford 1e as a light yellow wax (1.15 g, 52%). IR (liquid layer): n˜ =3484,
3379, 3205, 2969, 2931, 2876, 1782, 1720, 1614, 1571, 1447, 1391, 1372,
1285, 1235, 1161, 1037 cmÀ1 1H NMR (400 MHz, CDCl3): d=6.95 (1H,
;
s), 6.92 (1H, s), (5.7, 1H, s), 4.16 (2H, s), 3.74 (2H, d, J=6.4 Hz), 2.04
(1H, s), 0.99 ppm (6H, d, J=6.8 Hz); 13C NMR (CDCl3): d=169.24,
168.64, 159.43, 158.8, 152.88, 152.29, 11.42, 92.05, 90.00, 89.27, 83.15,
81.09, 74.54, 28.71, 28.36, 19.59 ppm; ESI-MS: m/z (%) cacld for
C14H23N3O3: 281.2; found: 281.9 [M+H]+.
1
76–788C; H NMR (400 MHz, [D6]DMSO): d=7.73 (2H, s), 4.01 (2H, d,
J=6.0), 3.91 (6H, s), 2.07 (1H, m), 1.00 ppm (6H, d, J=6.4); 13C NMR
(CDCl3): d=166.40, 164.50, 149.25, 113.96, 74.50, 52.56, 27.37, 18.65 ppm;
ESI-MS: m/z (%) calcd for C13H17NO5: 267.1; found: 268.0 [M+H]+,
290.1 [M+Na]+, 556.8 [2M+Na]+.
Cbz-momomer amine 1 f. Lithium bis(trimethylsilyl)amide (LiHMDS)
(1.0mÀ1, 11.0 mL, 11.0 mmol, 2 equiv) was added dropwise to a solution
of 4-isobutyloxy-2,6-diamino-pyridine (1b, 1.0 g, 5.5 mmol, 1 equiv) in
dry THF (10 mL) at 08C over a period of 15 min. After 10 min, a solu-
tion of benzyl chloroformate (0.80 mL, 7.9 mmol, 1 equiv) in dry THF
(10 mL) was added over 20 min. The reaction was then allowed to pro-
ceed at room temperature for 3 h. The solvent was removed and the resi-
due was dissolved in CH2Cl2 (50 mL) and extracted with water (2ꢁ
50 mL). The organic layer was evaporated and the residue was purified
by column chromatography using cyclohexane/ethyl acetate/triethyl
amine (70:25:5 vol/vol/vol) as eluent to afford a light yellow wax (0.97,
56%). 1H NMR (400 MHz, CDCl3): d=7.61 (1H, brs), 7.38–7.35 (4H,
m), 6.99 (1H, s), 5.71 (s, 1H), 5.21 (2H, s), 4.24 (2H, brs), 3.74 (2H, d),
2.06 (1H, m), 1.01 ppm (6H, d); ESI-MS: m/z (%) calcd for C17H21N3O3:
315.2; found: 316.1 [M+H]+).
4-isobutyloxy-2,6-pyridinedicarboxylic acid 1c. Diester 1a (1.34 g,
5 mmol) was dissolved in a mixture of methanol (10 mL), 1,4-dioxane
(10 mL) and water (5 mL). NaOH (0.6 g, 15 mmol) was added and the
solution was stirred at 258C. The progress of saponification was moni-
tored by TLC. Acetic acid (1.3 mL) was added and the resulting white
precipitate was filtered off and washed thoroughly with ethyl acetate and
diethyl ether. Yield 1.13 g (95%); m.p.: 171–1738C; 1H NMR (400 MHz,
[D6]DMSO): d=7.51 (2 h, s), 4.03 (2H, d, J=6.4 Hz), 2.08 (1H, m),
1.01 ppm (6H, d, J=6.8); ESI-MS: m/z (%) cacld for C11H13NO5: 239.1;
found: 240.2 [M+H]+.
4-isobutyloxy-2,6-pyridinedicarboxylic acid monomethyl ester 1d. Diester
1a (1.34 g, 5 mmol) was dissolved in a mixture of 1,4-dioxane (24 mL)
and methanol (7 mL) and the solution was cooled to 08C. NaOH (0.2 g,
5 mmol) was added and the solution was stirred for 2 h at 258C. The
progress of saponification was monitored by TLC. The solution was acidi-
fied with acetic acid and poured into 100 mL water, and then extracted
with CH2Cl2 (2ꢁ50 mL). The solvent was evaporated and the residue was
dried under vacuum and used directly without further purification. Yield
0.91 g (72%); m.p.: 93–958C; 1H NMR (400 MHz, CDCl3): d=7.85 (1 h,
s), 7.82 (1H, s), 4.01 (3H, s), 3.92 (2H, d, J=6.4), 2.16 (1H, m), 1.05 ppm
(6H, d, J=6.4); 13C NMR (CDCl3): d=176.78, 164.03, 163.75, 147.99,
115.53, 114.20, 111.90, 75.19, 52.75, 27.6, 18.61 ppm; ESI-MS: m/z (%)
Trimer di-Boc 3a. Diacid 1c (434 mg, 1.82 mmol, 0.7 equiv) and thionyl
chloride (10 mL) were heated to reflux for 0.5 h, under a nitrogen atmos-
phere. Excess thionyl chloride was removed first under reduced pressure,
and then with a toluene azeotrope. The residue was dissolved in dry THF
(10 mL). This solution was added dropwise to a previously prepared solu-
tion of 1e (850 mg, 3.02 mmol, 1 equiv) in dry THF (10 mL) and diiso-
propylethylamine (1.08 mL, 2.5 equiv) at 258C. The reaction mixture was
stirred at 258C for 12 h. The solvent was removed and the residue was
purified by flash chromatography (SiO2) with ethyl acetate/CH2Cl2 (5:95
vol/vol) as the eluent to yield the product (860 mg, 75%). M.p.: 140–
1428C; IR (liquid layer): n˜ =3644, 3375, 3300, 3138, 3056, 2963, 2931,
2875, 1731, 1694, 1613, 1581, 1513, 1444, 1394, 1363, 1225, 1163, 1081,
cacld for C12H15NO5: 253.1; found: 254.1 [M+H]+, 528.8 [2M+Na]+
2,6-diamino-4-isobutyloxy-pyridine 1b. Dry NH3 gas was gently bubbled
into solution of 4-isobutyloxy-2,6-pyridinedicarboxylate 1a (3.5 g,
.
a
13.1 mmol) in methanol (45 mL) for 1 h. The resulting white precipitate
of 4-isobutyloxy-2,6-pyridinedicarboxamide was filtered off, dried under
vacuum and used without further purification (2.8 g, 90%). M.p.: 190–
1928C; 1H NMR (400 MHz, [D6]DMSO): d=8.83 (2H, s), 7.69 (2H, s),
7.63 (2H, s), 3.97 (2H, d, J=6.0), 2.05 (1H, m), 0.99 ppm (6H, d, J=
5.6); ESI-MS: m/z (%) calcd for C11H15N3O3: 237.1; found: 238.1
[M+H]+, 260.1 [M+Na]+. KOH (10 g, 178 mmol) was dissolved in water
(20 mL) at 08C. Crushed ice (25 g) was added, followed by the slow addi-
tion of Br2 (1.3 mL, 25.4 mmol) with continuous stirring. After 10 min, 4-
isobutyloxy-2,6-pyridinedicarboxamide (2.37 g, 10 mmol) was added, fol-
lowed by 1,4-dioxane (45 mL). The mixture was stirred in an ice bath for
10 min, at 258C for 1 h and then heated to 708C for 1.5 h. Acetic acid
(6 mL) was added and, afer 30 min, the reaction mixture was cooled to
258C, and KOH (4 g) was added. The solution was extracted with di-
chloromethane. The organic phase was evaporated and the residue was
purified by flash chromatography (SiO2) with methanol/ethyl acetate
(5:95 vol/vol) as eluent to yield 1.42 g of 1b as a light yellow solid
(78%); m.p.: 104–1068C; IR (Liquid layer): n˜ =3435, 3373, 3192, 2963,
1038 cmÀ1 1H NMR (400 MHz, CDCl3): d=10.31 (2H, s), 7.95 (2H, s),
;
7.76 (2H, s), 7.31 (2H, s), 7.14 (2H, S), 3.95 (2H, d, J=5.6), 3.86 (4H, d,
J=4.8), 2.13 (3H, m), 1.52 (18H, s), 1.05 ppm (18H, d, J=6.4); 13C NMR
(CDCl3): d=168.96, 168.20, 161.56, 152.13, 151.45, 150.45, 150.13, 111.80,
95.98, 95.00, 80.88, 75.30, 74.48, 30.26, 29.65, 28.14, 27.92, 19.15 ppm;
ESI-MS: m/z (%) calcd for C39H55N7O9: 765.4; found: 766.2 [M+H]+.
Trimer diamine 3b. Trimer di-Boc 3a (850 mg, 1.11 mmol) was dissolved
in CH2Cl2 (5 mL) and TFA (2 mL) was added to the solution. After stir-
ring for 2 h, the reaction mixture was washed with aqueous NaHCO3 so-
lution and extracted with chloroform (2ꢁ50 mL). The solvent was evapo-
rated to yield the product as a white solid (621.5 mg, 99%). M.p.: 114–
1168C; IR (liquid layer): n˜ =3370, 3352, 2963, 2932, 2876, 1684, 1616,
1350, 1450, 1338, 1245, 1233, 1172, 1036 cmÀ1 1H NMR (400 MHz,
;
CDCl3): d=10.23 (2H, s), 7.94 (2H, s), 7.52 (2H, s), 5.84 (2H, s), 4.43
(4H, S), 3.95 (2H, d, J=4.4), 3.81 (4H, d, J=5.6), 2.13 (3H, m),
1.03 ppm (18H, d, J=4.8); 13C NMR (CDCl3): d=168.88, 168.45, 158.53,
150.38, 111.67, 91.85, 90.23, 75.30, 74.31, 28.09, 19.16 ppm; ESI-MS: m/z
(%) calcd for C29H39N7O5: 565.3; found: 566.36 [M+H]+.
1
2919, 2876, 1639, 1447, 1198, 1030 cmÀ1; H NMR (400 MHz, CDCl3): d=
5.48 (2H, s), 4.12 (4H, s), 3.65 (2H, d, J=6), 2.03 (1H, m), 0.99 ppm
(6H, d, J=6.8); 13C NMR (CDCl3): d=168.76, 158.91, 84.67, 73.88, 28.04,
19.14 ppm; ESI-MS: m/z (%) calcd for C9H15N3O: 181.1; found: 182.1
[M+H]+.
Dimer Boc-ester 2a.Monoacid 1d (404 mg, 1.6 mmol, 1.2 equiv) and thio-
nyl chloride (10 mL) were heated to reflux for 0.5 h under a nitrogen at-
mosphere. The excess thionyl chloride was removed first under reduced
pressure and then with a toluene azeotrope. The residue was dissolved in
dry THF (10 mL). This solution was added dropwise to a previously pre-
pared solution of monoamine 1e (374 mg, 1.33 mmol, 1 equiv) in dry
THF (5 mL) and diisopropylethylamine (1.3 mL) at room temperature.
The mixture was stirred at ambient temperature for 12 h. The solvent
(4-isobutyloxy-6-amino-pyridin-2-yl)-carbamic acid tert-butyl ester 1e.
Lithium bis(trimethylsilyl)amide (1.0m in THF, 15.8 mL, 2 equiv) was
added dropwise to a solution of 4-isobutyloxy-2,6-diaminopyridine 1b
(1.43 g, 7.9 mmol, 1 equiv) in dry THF (10 mL) through a period of
1372
www.chemasianj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Hybridization of Oligomers into Double Helices
was removed and the residue was purified by flash chromatography
(SiO2) with cyclohexane/ethyl acetate (2:1 vol/vol) as the eluent to yield
the product (520 mg 81%). M.p.: 134–1368C; IR (liquid layer): n˜ =3375,
2963, 2938, 2875, 1725, 1713, 1700, 1600, 1538, 1506, 1444, 1394, 1388,
the residue was purified by silica gel chromatography with CH2Cl2/
EtOAc (1:5 vol/vol) containing 1% AcOH as eluent, to yield the product
as a white solid (190 mg, yield: 88%). 1H NMR (400 MHz, CDCl3): d=
10.55 (1H, s), 7.99 (1H, s), 7.90 (2H, d, J=2.8 Hz), 7.76 (1H, s), 7.45–
1363, 1331, 1225, 1156, 1113, 1088, 1038 cmÀ1 1H NMR (400 MHz,
;
7.37 (5H, m), 5.27ACTHNUTRGNEUNG(2H, s), 3.94 (2H, d, J=8.8 Hz), 3.88 (2H, d, J=
8.8 Hz), 2.16 (2H, m), 1.06 ppm (12H, m); ESI-MS: m/z (%) calcd for
CDCl3): d=10.16 (1H, s), 7.92 (1H, s), 7.76 (1H, s), 7.66 (1H, s), 7.3
(1H, s), 7.03 (1H, s), 4.02 (3H, s), 3.92 (2H, d, J=6.4), 3.85 (2H, d, J=
6.8), 2.13 (2H, m), 1.52 (9H, s), 1.04 ppm (12H, d, J=4.0); 13C NMR
(CDCl3): d=169.31, 168.02, 165.33, 162.06, 152.51, 150.40, 148.59, 115.22,
111.32, 95.83, 95.02, 81.27, 75.57, 74.84, 53.27, 28.55, 1048 ppm; ESI-MS:
m/z (%) calcd for C26H36N4O7: 516.3; found: 517.0 [M+H]+.
C28H32N4O7: 536.2; found: 537.3 [M+H]+.
Pentamer diamine 5b. TFA (2 mL) was added to pentamer di-Boc 5a
(180 mg, 0.16 mmol) dissolved in CH2Cl2 (2 mL). After stirring for 2 h,
the reaction mixture was washed with saturated aqueous NaHCO3 solu-
tion and extracted with chloroform (2ꢁ50 mL). The solvent was evapo-
rated to yield the product as a white solid (146 mg, 98%). M.p.: 172–
Dimer Boc-acid 2b. Sodium hydroxide (80 mg, 2 equiv) was added to a
solution of dimer ester 2a (516 mg, 1 mmol) in 1,4-dioxane (25 mL) and
water (2 mL) at room temperature, and the resulting solution was stirred
for 2 h, then acidified with acetic acid. CH2Cl2 (40 mL) was then added
to the reaction mixture. The organic phase was washed with water, then
dried with MgSO4 and filtered. The solvent was removed and the residue
was purified by flash chromatography (SiO2) with cyclohexane/ethyl ace-
tate (from 2:1 to 1:1 vol/vol) as the eluent to give the product (452 mg,
90%). M.p.: 160–1628C; IR (liquid layer): n˜ =3250, 3133, 2967, 2930,
2874, 1724, 1706, 1693, 1583, 1447, 1392, 1361, 1331, 1238, 1158, 1085,
1
1748C; H NMR (400 MHz, CDCl3): d=10.61 (2H, s), 10.42 (2H, s), 7.88
(2H, s), 7.85 (2H, s), 7.58 (2H, s), 7.08 (2H, s), 5.31 (2H, s), 4.28 (4H,
br), 3.99 (6H, d, J=5.6), 3.82 (2H br), 3.6 (2H, br), 2.17 (5H, m),
1.04 ppm (30H, d, J=6.0); 13C NMR (CDCl3): d=168.34, 167.98, 162.06,
161.80, 158.08, 150.84, 150.37, 111.52, 111.22, 98.02, 90.24, 89.76, 89.72,
75.55, 74.28, 28.39, 28.35, 19.47, 19.40 ppm; ESI-MS: m/z (%) calcd for
C49H63N11O9: 949.5; found: 950 [M+H]+.
Heptamer di-Boc 7a was prepared from dimer acid 2b (425 mg,
0.85 mmol) and trimer diamine 3a (192 mg, 0.34 mmol) using the same
procedure as for 5a. Yield 506 mg (98%); m.p.: 124–1268C; IR (liquid
layer): n˜ =3391, 3336, 3317, 2967, 2936, 2911, 2887, 1687, 1724, 1737,
1035 cmÀ1 1H NMR (400 MHz, CDCl3): d=10.56 (1H, s), 8.11 (1H, s),
;
7.98 (1H, s), 7.93 (1H, s), 7.73 (1H, s), 7.66 (1H, s), 3.96 (2H, d, J=4.4),
3.87 (2H, d, J=6.8), 2.16 (2H, m), 1.48 (9H, s), 1.05 ppm (12H, d, J=
4.0); 13C NMR (CDCl3): d=170.35, 168.23, 162.87, 152.56, 151.30, 149.87,
114.80, 112.13, 96.25, 95.20, 81.69, 75.57, 75.25, 28.65, 19.56 ppm; ESI-
MS: m/z (%) calcd for C25H34N4O7: 502.2; found: 503.19 [M+H]+.
1613, 1583, 1527, 1509, 1490, 1441, 1386, 1324, 1171, 1152, 1041 cmÀ1
;
1H NMR (400 MHz, [D6]DMSO): d=10.92 (2H, s), 10.49 (2H, s), 10.19
(2H, s), 9.31 (2H, s), 7.96 (2H, s), 7.77 (2H, s), 7.73 (2H, s), 7.50 (2H,
s), 7.37 (2H, s), 7.19 (2H, s), 6.96 (2H, s), 4.11 (6H, d, J=4.4), 3.88 (4H,
d, J=5.6), 3.84 (4H, d, J=5.2), 2.15 (7H. m), 1.32 (18H, s), 1.11 ppm
(42H, br); 13C NMR (CDCl3): d=168.17, 167.70, 167.57, 161.02, 150.98,
150.39, 150.09, 149.62, 149.41, 111.56, 111.03, 96.28, 95.04, 93.94, 80.41,
75.15, 73.96, 28.14, 27.97, 27.68, 19.15 ppm; ESI-MS: m/z (%) calcd for
C79H103N15O17: 1533.8; found: 1534.2 [M+H]+.
Pentamer
di-Boc
5a.
1-Chloro-N,N,2-trimethylpropenylamine[20]
(1.2 mmol, 0.1 mL, 6 equiv) was added to dimer acid 2b (251 mg,
0.5 mmol, 2.5 equiv) dissolved in dry CH2Cl2 (2 mL) under a nitrogen at-
mosphere. The mixture was stirred for 12 h at 258C. Excess chloroena-
mine was removed under reduced pressure. The residue obtained was dis-
solved in dry CH2Cl2 (2 mL) and added dropwise to a previously pre-
pared solution of monomer di-amine 1b (36 mg, 0.2 mmol, 1 equiv) in
dry CH2Cl2 (2 mL) and diisopropylethylamine (1.3 mL) at 258C. The
mixture was stirred at ambient temperature for 12 h. The solvent was re-
moved and the residue was purified by flash chromatography (SiO2) with
ethyl acetate/cyclohexane (5:95 vol/vol) as the eluent to yield the product
(218 mg, 95%). M.p.: 116–1188C; 1H NMR (400 MHz, CDCl3): d=10.35
(2H, s), 10.26 (2H, s), 7.76 (2H, s), 7.68 (2H, s), 7.40 (2H, s), 7.07 (2H,
s), 6.66 (4H, s), 3.95 (6H, d, J=6.8), 3.70 (2H br), 3.53 (2H, br), 2.17
(5H, m), 1.15 (18H, s), 1.11 ppm (30H, d, J=6.4); 13C NMR (CDCl3):
d=168.26, 167.75, 161.30, 161.06, 151.20, 150.80, 150.41, 149.99, 111.44,
110.97, 97.43, 95.63, 93.95, 80.45, 75.3, 74.08, 48.53, 39.11, 28.23, 27.76,
Heptamer diamine 7b was prepared from heptamer di-Boc 7a (450 mg,
0.29 mmol) using the same procedure as for 5b. Yield 382 mg (99%);
m.p.: 180–1828C; IR (liquid layer): n˜ =3308, 3284, 2969, 2932, 2876, 1690,
1610, 1585, 1573, 1530, 1473, 1425, 1447, 1425, 1431, 1345, 1332, 1289,
1215, 1172, 1166, 1036 cmÀ1 1H NMR (400 MHz, [D6]DMSO): d=11.38
;
(2H, s) 10.74 (2H, s), 10.30 (2H, s), 7.88 (4H, s), 7.71 (2H, s), 7.64 (2H,
s), 7.56 (2H, s), 6.76 (2H, s), 5.61 (2H, s), 5.40 (4H, s), 4.08 (2H, d, J=
6.4), 4.04 (4H, d, J=6.4), 3.91 (4H, d, J=6.4), 3.70 (4H, d, J=6.4), 2.14
(7H, m), 1.08 ppm (42H, br); 13C NMR (CDCl3): d=168.13, 167.69,
161.65, 157.72, 150.67, 150.06, 111.34, 96.78, 91.88, 89. 55, 75.38, 74.07,
29.97, 28.46, 19.38 ppm; ESI-MS: m/z (%) calcd for C69H87N15O13: 1333.7;
found: 1334.32 [M+H]+.
20.58, 19.16, 19.08 ppm; ESI-MS: m/z (%) calcd for C59H79N11O13
1149.6; found: 1150.4 [M+H]+.
:
Nonamer di-Boc 9a was prepared from dimer acid 2b (140 mg,
0.28 mmol, 2.5 equiv) and pentamer diamine 5b (107 mg, 0.11 mmol,
1 equiv) using the same procedure as for 5a. Yield 152 mg (72%); m.p.:
138–1408C; 1H NMR (400 MHz, [D6]DMSO, at 1018C): d=10.17 (2H,
s), 10.02 (2H, s), 9.94 (2H, s), 9.85 (2H, s), 7.93 (2H, s), 7.81 (2H, s),
7.48 (2H, s), 7.37 (4H, s), 7,24 (2H, s), 6.86 (2H, br), 4.12 (8H, d, J=
5.6), 3.97 (3H, br), 3.81 (3H, br), 2.24 (18H, s), 1.15 ppm (54H, d, J=
5.0 Hz); 13C NMR (CDCl3): d=168.17, 167.78, 167.38, 161.28, 160.87,
150.76, 150.20, 149.30, 117.03, 111.87, 110.37, 97.29, 94.81, 88.06, 84.03,
82.38, 81.96, 75.03, 73.89, 65.98, 29.66, 28.12, 27.6, 19.15, 19.04 ppm; ESI-
Dimer Cbz Ester 2c. Under a nitrogen atmosphere, 4-isobutyloxy-2,6-
pyridinedicarboxylic acid monomethyl ester (1d, 471 mg, 1.8 mmol,
1.2 equiv) was suspended in toluene (10 mL). Oxalyl chloride (0.5 mL,
4.1 mmol) and two drops of DMF were added to the suspension. The re-
action mixture was stirred at room temperature for 2 h, and then excess
oxalyl chloride and toluene were removed under reduced pressure. The
residue was dissolved in dry THF (10 mL) and was added dropwise to a
previously prepared solution of 1 f (315 mg, 1.0 mmol, 1 equiv) in dry
THF (5 mL) and N,N-diisopropylethylamine (0.4 mL) at room tempera-
ture. The mixture was stirred at ambient temperature for 12 h. The sol-
vent was removed and the residue was purified by silica gel chromatogra-
phy, with cyclohexane/ethyl acetate (2:1 vol/vol) as eluent to yield the
product (570 mg, 98%). 1H NMR (400 MHz, CDCl3): d=10.20 (1H, s),
7.92 (1H, d), 7.36 (5H, m), 5.24 (2H, s), 4.04 (2H, s), 3.94 (2H, d), 3.88
(2H, d), 2.16 (2H, m), 1.06 ppm ( 12H, m); ESI-MS: m/z (%) calcd for
C29H34N4O7: 550.2; found: 551.2 [M+H]+.
MS: m/z (%) calcd for C99H127N19O21: 1917.9; found: 1919 [2M++2H]2+
1940 [M+Na]+ and 1956 [M+K]+.
,
Nonamer diamine 9b was prepared from 9a (120 mg, 0.06 mmol) using
the same procedure as for 5b. Yield 99.5 mg (98%); m.p.: 190–1928C;
1H NMR (400 MHz, [D6]DMSO, 628C): d=10.61 (2H, s), 10.39 (2H, s),
10.25 (2H, s), 9.73 (2H, s), 8.25 (2H, s), 7.77 (2H, s), 7.48 (2H, s), 7.42
(4H, s), 7.36 (2H, s), 6.84 (2H, br), 5.43 (2H, s), 5.06 (2H, s), 4.09 (8H,
d, J=5.6), 3.95 (3H br), 3.87 (3H, br), 1.14 ppm (54H, d, J=5.0);
13C NMR (CDCl3): d=167.93, 167.25, 161.49, 160.68, 157.34, 149.78,
149.47, 149.02, 11.93, 111.20, 110.63, 97.39, 96.93, 96.87, 95.99, 95.67,
91.27, 89.05, 75.04, 73.34, 48.94, 28.09, 28.03, 19.12, 19.07 ppm; ESI-MS:
Dimer Cbz acid 2d. To a solution of dimer ester 2c (220 mg, 0.40 mmol)
in 1,4-dioxane (15 mL) and water (2 mL) at room temperature, sodium
hydroxide (40 mg, 1.0 mmol, 2.5 equiv) was added and the resulting solu-
tion was stirred for 2 h, and then acidified with acetic acid, followed by
the addition of CH2Cl2 (40 mL). The organic layer was washed with
water, then dried with MgSO4 and filtered. The solvent was removed and
m/z (%) calcd for C89H111N19O17: 1717.8; found: 1719 [2M+2H]2+
.
Undecamer di-Boc 11a was prepared from dimer acid 2b (329 mg,
0.66 mmol, 2.5 equiv) and heptamer diamine 7b (350 mg, 0.26 mmol,
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1373

I. Huc et al.
FULL PAPERS
1 equiv) using the same procedure as for 5a. Yield 414 mg (70%); m.p.:
146–1488C; 1H NMR (400 MHz, [D6]DMSO, 1018C): d=10.23 (2H, s),
9.97 (2H, s), 9.78 (2H, s), 9.65 (2H, s), 9.35 (2H, s), 8.13 (2H, s), 8.01
(2H, s), 7.83 (2H, s), 7.69 (4H, s), 7.39 (2H, s), 7.30 (2H, br), 7.03 (2H,
br), 6.98 (2H, s), 6.76 (2H, s) 4.08 (12H, d, J=9.6), 3.92 (10H, d, J=
11.6), 3.72 (3H, br), 2.23 (11H, m), 1.24 (18H, s), 1.15 ppm (66H, d, J=
5.0); 13C NMR (CDCl3): 168.12, 167.88, 167.35, 161.21, 160.84, 160.46,
151.01, 150.43, 149.71, 149.09, 148.47, 124.58, 112.50, 111.64, 110.48,
109.98, 102.04, 97.26, 95.78, 95.03, 93.38, 80.56, 89.96, 74.93, 73.75, 28.18,
27.99, 27.53, 27.51, 1938, 18.99 ppm; ESI-MS: m/z (%) calcd for
Ramshow, A. Kirkpatrick, B. Brodsky, Biochemistry 2005, 44, 1414;
e) R. Improta, R. Berisio, L. Vitagliano, Protein Sci. 2008, 17, 955.
[3] S. Janaswamy, R. Chandrasekaran, Carbohydr. Res. 2001, 335, 181.
[4] F. H. C. Crick, Nature 1952, 170, 882; R. D. B. Fraser, T. P. Macrae,
D. A. D. Parryl, E. Suzuki, Proc. Natl. Acad. Sci. USA 1986, 83,
1179.
[5] D. A. Langs, Science 1988, 241, 188; B. M. Burkhart, R. M. Gass-
man, D. A. Langs, W. A. Pangborn, W. L. Duax, V. Pletnev, Biopoly-
mers 1999, 51, 129.
[6] D. Haldar, C. Schmuck, Chem. Soc. Rev. 2009, 38, 363.
[7] Selected examples: B. Hasenknopf, J.-M. Lehn, G. Baum, D. Fenske,
Proc. Natl. Acad. Sci. USA 1996, 93, 1397; C. R. Woods, M. Bena-
glia, F. Cozzi, J. S. Siegel, Angew. Chem. 1996, 108, 1977; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1830; A. Orita, T. Nakano, D. L. An,
K. Tanikawa, K. Wakamatsu, J. Otera, J. Am. Chem. Soc. 2004, 126,
10389; H. Katagiri, T. Miyagawa, Y. Furusho, E. Yashima, Angew.
Chem. 2006, 118, 1773; Angew. Chem. Int. Ed. 2006, 45, 1741; A.
Marquis, V. Smith, J. Harrowfield, J.-M. Lehn, H. Herschbach, R.
Sanvito, E. Leise-Wagner, A. V. Dorsselaer, Chem. Eur. J. 2006, 12,
5632.
C119H151N23O25: 2301; found: 2303 [2M+2H]2+
.
Undecamer bis-Cbz 11b was prepared from dimer acid 2d (182 mg,
0.34 mmol, 2.5 equiv) and heptamer diamine 7b (181 mg, 0.14 mmol,
1 equiv) using the same procedure as for 5a. Yield: 249 mg (75%).
1H NMR (400 MHz, CDCl3): d=10.09 (1H, s), 10.04 (1H, s), 9.91 (1H,
s), 9.77 (3H, m, br), 9.70 (1H, s), 9.68 (1H,s), 9.63 (1H, s), 9.58 (1H, s),
7.62 (1H, d), 7.54 (3H, m), 7.51 (1H, d), 7.46 (1H, d), 7.43 (1H,d), 7.40
(1H, d), 7.33 (3H, m), 7.16–7.11 (5H, m), 6.96–6.89 (5H, m), 6.76 (1H,
s), 6.66 (1H, s), 6.61–6.56 (3H, m), 6.51 (1H, s), 6.30 (3H, s, br.), 6.20
(1H, s), 6.12 (1H, s), 4.98 (2H, t, J=12.6 Hz), 4.48 (1H, d, J=12.3 Hz),
4.24 (1H, d, J=12.3 Hz), 4.07–3.77 (11H, m), 3.50–3.38 (11H, m), 2.27–
1.87 (11H, m), 1.19–0.89 ppm (66H, m); ESI-MS: m/z (%) calcd for
C125H147N23O25: 2370.09; found: 2371.1 [M+H]+.
[8] C. J. Leumann, Bioorg. Med. Chem. 2002, 10, 841; P. E. Nielsen, M.
Egholm, R. H. Berg, O. Buchardt, Science 1991, 254, 1497; P. E.
Nielsen, Acc. Chem. Res. 1999, 32, 624.
Tridecamer di-Boc 13a was prepared from dimer acid 2b (60 mg,
0.12 mmol, 2.5 equiv) and nonamer diamine 9b (80 mg, 0.047 mmol,
1 equiv) using the same procedure as for 5a. Yield 102 mg (81%); m.p.:
152–1548C; 1H NMR (400 MHz, [D6]DMSO, 758C): d=10.00 (2H, s),
9.88 (4H, s), 9.76 (2H, s), 9.70 (2H, s), 9.53 (4H, s), 8.14 (2H, s) 7.72
(2H, s), 7.67 (2H, s), 7.31 (12H, br), 7.10 (2H, s), 6.99 (4H, s), 6.72 (2H,
s), 5.68 (2H,s), 4.09 (14H, d, J=5.6), 3.85 (12H, d, J=3.4), 2.24 (13H,
m), 1.24 (18H, s), 1.16 ppm (78H, d, J=5.0); 13C NMR (CDCl3): d=
167.77, 166.83, 162.83, 160.43, 160.05, 154.88, 154.48, 153.79, 149.89,
149.29, 148.71, 145.19, 112.17, 95.35, 94.51, 94.35, 88.42, 74.59, 73.36,
27.83, 27.25, 27.20, 18.68 ppm; ESI-MS: m/z (%) calcd for
C139H175N27O29: 2686.3; found: 2687.0 [2M+2H]2+ (Calcd for)..
[9] a) V. Berl, I. Huc, R. Khoury, M. J. Krische, J.-M. Lehn, Nature
2000, 407, 720; b) V. Berl, I. Huc, R. Khoury, J.-M. Lehn, Chem.
Eur. J. 2001, 7, 2810; c) H. Jiang, V. Maurizot, I. Huc, Tetrahedron
2004, 60, 10029; d) C. Dolain, C. Zhan, J.-M. Lꢂger, I. Huc, J. Am.
Chem. Soc. 2005, 127, 2400; e) C. Zhan, J.-M. Lꢂger, I. Huc, Angew.
Chem. 2006, 118, 4741; Angew. Chem. Int. Ed. 2006, 45, 4625; f) D.
Haldar, H. Jiang, J.-M. Lꢂger, I. Huc, Angew. Chem. 2006, 118,
5609; Angew. Chem. Int. Ed. 2006, 45, 5483; g) Q. Gan, C. Bao, B.
Kauffmann, A. Grꢂlard, J. Xiang, S. Liu, I. Huc, H. Jiang, Angew.
Chem. 2008, 120, 1739; Angew. Chem. Int. Ed. 2008, 47, 1715; h) E.
Berni, B. Kauffmann, C. Bao, J. Lefeuvre, D. Bassani, I. Huc, Chem.
Eur. J. 2007, 13, 8463; i) E. Berni, J. Garric, C. Lamit, B. Kauffmann,
J.-M. Lꢂger, I. Huc, Chem. Commun. 2008, 1968; j) D. Haldar, H.
Jiang, J.-M. Lꢂger, I. Huc, Tetrahedron 2007, 63, 6322; k) Q. Gan, F.
Li, G. Li, B. Kauffmann, J. Xiang, I. Huc, H. Jiang, Chem. Commun.
2010, 46, 297.
[10] a) B. Di Blasio, E. Benedetti, V. Pavone, C. Pedone, Biopolymers
1989, 28, 203; b) G. Bunkꢄczi, L. Vꢂrtesy, G. M. Sheldrick, Angew.
Chem. 2005, 117, 1364; Angew. Chem. Int. Ed. 2005, 44, 1340.
[11] a) H. Goto, H. Katagiri, Y. Furusho, E. Yashima, J. Am. Chem. Soc.
2006, 128, 7176; b) H. Goto, Y. Furusho, E. Yashima, J. Am. Chem.
Soc. 2007, 129, 109; c) H. Goto, Y. Furusho, E. Yashima, J. Am.
Chem. Soc. 2007, 129, 9168; d) H. Goto, Y. Furusho, K. Miwa, E.
Yashima, J. Am. Chem. Soc. 2009, 131, 471.
[12] a) H. Sugiura, Y. Nigorikawa, Y. Saiki, K. Nakamura, M. Yamagu-
chi, J. Am. Chem. Soc. 2004, 126, 14858; b) H. Sugiura, M. Yamagu-
chi, Chem. Lett. 2007, 36, 58; c) H. Sugiura, R. Amemiya, M. Yama-
guchi, Chem. Asian J. 2008, 3, 244; d) R. Amemiya, N. Saito, M. Ya-
maguchi, J. Org. Chem. 2008, 73, 7137.
[13] a) Y. Tanaka, H. Katagiri, Y. Furusho, E. Yashima, Angew. Chem.
2005, 117, 3935; Angew. Chem. Int. Ed. 2005, 44, 3867; b) H. Ito, Y.
Furusho, T. Hasegawa, E. Yashima, J. Am. Chem. Soc. 2008, 130,
14008; c) H. Ito, Y. Furusho, T. Hasegawa, E. Yashima, J. Am.
Chem. Soc. 2008, 130, 14008; d) T. Maeda, Y. Furusho, S.-I. Sakurai,
J. Kumaki, K. Okoshi, E. Yashima, J. Am. Chem. Soc. 2008, 130,
7938.
X-ray crystallography. Single crystals of 5a, 7a, and 13a were mounted
on a Rigaku R-Axis Rapid diffractometer equipped with a MM07 micro-
focus rotating anode generator (monochromatized Cuka radiation,
1.54178 ꢀ). The data collection, unit cell refinement and data reduction
were performed by using the CrystalClear software package. The position
of non-H atoms were determined by the program SHELXD and refined
with SHELXH. The position of the H atoms were deduced from coordi-
nates of the non-H atoms and confirmed by Fourrier synthesis. H atoms
were included for structure factor calculations but not refined. Single
crystals of 9a and 11a were measured on Synchrotron beam line ID 29
(ESRF, Grenoble). Data processing was performed using the same soft-
wares as for other crystals. Details of crystallographic data can be found
in the Supporting Information.
Acknowledgements
This work was supported by the French Ministry of Research (predoctor-
al fellowship to B.B. and postdoctoral fellowship to D.H.) and by an
ANR grant (project no. NT05-3_44880). We warmly acknowledge Chris-
toph Mueller-Dieckmann for beamtime and help during data collection
on ID29 Beamline at ESRF.
[14] J. Li, J. A. Wisner, M. C. Jennings, Org. Lett. 2007, 9, 3267.
[15] T. K. Chakraborty, D. Koley, R. Ravi, A. C. Kunwar, Org. Biomol.
Chem. 2007, 5, 3713.
[16] H. Gong, M. J. Krische, J. Am. Chem. Soc. 2005, 127, 1719; J. Zhu,
J.-B. Lin, Y.-X. Xu, X.-B. Shao, X.-K. Jiang, Z.-T. Li, J. Am. Chem.
Soc. 2006, 128, 12307; M. Li, K. Yamato, J. S. Ferguson, B. Gong, J.
Am. Chem. Soc. 2006, 128, 12628; Y. Yang, J.-F. Xiang, M. Xue, H.-
Y. Hu, C.-F. Chen, Org. Biomol. Chem. 2008, 6, 4198.
[1] a) Principles of Nucleic Acid Structure (Ed.: Stephen Neidle), Elsev-
ier 2008; b) D. H. Turner, Curr. Opin. Struct. Biol. 1996, 6, 299; c) N.
Sugimoto, S. Nakano, M. Yoneyama, K. Honda, Nucl. Acids Res.
1993, 24, 4501.
[2] a) K. Beck, B. Brodsky, J. Struct. Biol. 1998, 122, 17; b) G. Auerbach,
F. Gaill, R. Jaenicke, T. Schulthess, R. Timpl, J. Engel, Matrix Biol.
1995, 14, 589; c) R. Berisio, A. De Simone, A. Ruggiero, R. Improta,
L. Vitagliano, J. Pept. Sci. 2009, 15, 131; d) A. V. Persikov, J. A. M.
[17] E. A. Archer, M. J. Krische, J. Am. Chem. Soc. 2002, 124, 5074.
1374
www.chemasianj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1364 – 1375

Hybridization of Oligomers into Double Helices
[18] A. Acocella, A. Venturini, F. J. Zerbetto, J. Am. Chem. Soc. 2004,
126, 2362.
[24] The compounds described in reference [9c] are not crystalline. Crys-
tallization did not drive the equilibrium towards a fully hybridized
or fully dissociated state. The proportions of single and double heli-
ces were probably those arising from chromatographic separation.
[25] We believe it is reasonable to assume that changing decyloxy side
chains for isobutoxy side chains has no effect as far as Kdim values
are concerned. However, these side chains may result in different
association and dissociation rates, the bulkier decyloxy presumably
resulting in slower kinetics.
[19] H. Jiang, J.-M. Lꢂger, I. Huc, J. Am. Chem. Soc. 2003, 125, 3448.
[20] a) N. Delsuc, J.-M. Lꢂger, S. Massip, I. Huc, Angew. Chem. 2007,
119, 218; Angew. Chem. Int. Ed. 2007, 46, 214; b) D. Sꢅnchez-
Garcꢆa, B. Kauffmann, T. Kawanami, H. Ihara, M. Takafuji, M.-H.
Delville, I. Huc, J. Am. Chem. Soc. 2009, 131, 8642.
[21] L. Ghosez, B. Haveaux, H. G. Viehe, Angew. Chem. 1969, 81, 468;
Angew. Chem. Int. Ed. Engl. 1969, 8, 454.
[22] A. Zhang, J. S. Ferguson, K. Yamato, C. Zheng, B. Gong, Org. Lett.
2006, 8, 5117.
[26] R. ChÞnevert, M. Dickman, J. Org. Chem. 1996, 61, 3332–3341.
[23] a) V. Berl, I. Huc, R. Khoury, J.-M. Lehn, Chem. Eur. J. 2001, 7,
2798; b) I. Huc, V. Maurizot, H. Gornitzka, J.-M. Lꢂger, Chem.
Commun. 2002, 578.
Received: December 14, 2009
Published online: April 7, 2010
Chem. Asian J. 2010, 5, 1364 – 1375
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1375