Double versus single helical structures of oligopyridine-dicarboxamide strands. Part 2: The role of side chains

Article abstract of DOI:10.1016/j.tet.2007.02.114

A series of heptameric oligoamides comprising 4-alkoxy-substituted 2,6-diaminopyridine and 2,6-pyridine-dicarbonyl units have been synthesized using convergent methods. The hybridization of these compounds into double helical dimers was studied in solutio

Full text of DOI:10.1016/j.tet.2007.02.114

Tetrahedron 63 (2007) 6322–6330
Double versus single helical structures of oligopyridine-
dicarboxamide strands. Part 2: The role of side chains*
b
a,
Debasish Haldar,a Hua Jiang,a,y Jean-Michel Leger and Ivan Huc
*
ꢀ
a
´
´
Institut Europeen de Chimie et Biologie, Universite Bordeaux 1—CNRS UMR5248, 2 rue Robert Escarpit, 33607 Pessac, France
b
´
Laboratoire de Pharmacochimie, Universite Victor Segalen Bordeaux 2, 146 rue Leo Saignat, 33076 Bordeaux, France
´
Received 3 January 2007; revised 26 February 2007; accepted 27 February 2007
Available online 2 March 2007
Abstract—A series of heptameric oligoamides comprising 4-alkoxy-substituted 2,6-diaminopyridine and 2,6-pyridine-dicarbonyl units have
been synthesized using convergent methods. The hybridization of these compounds into double helical dimers was studied in solution by 1H
NMR spectroscopy in CDCl3 or DMSO-d6 at various concentrations, and in the solid state using X-ray crystallographic analysis. Both solid
state and solution data suggest that these compounds follow identical hybridization schemes. In CDCl3, the oligomers possess dimerization
constants considerably (up to 2000-fold) higher than related compounds having no alkoxy substituents on their 2,6-diaminopyridine units.
The origin of this effect can be in part interpreted as a result of interactions between the 4-alkoxy side chains when they are present on all pyr-
idine rings. For example, 4-benzyloxy-substituted oligomer 2 has a higher dimerization constant than 4-decyloxy and 4-methoxy-substituted
analogues 1 and 3. The crystal structure of 2 reveals multiple aromatic–aromatic interactions between the benzyl side chains, both face-to-face
and edge-to-face at various angles surrounding the duplex. In the solid state, these double helices are stacked on top of each other to form long
channels filled with water molecules. The 4-methoxy and 4-decyloxy-substituted analogues 1 and 3 have similar dimerization constants, show-
ing that interactions between side chains are not significant between purely aliphatic residues. Consequently, the high stability of the double
helices formed by 1 and 3 compared to related compounds having alkoxy substituents on their 2,6-pyridine-dicarbonyl units only does not find
its origin in interactions between side chains but in the direct effect of the alkoxy substituents upon main chain aryl–aryl interactions.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
oligo-m-terphenyl derivatives,4 and oligoamides of 2,6-dia-
minopyridine and 2,6-pyridine-dicarboxylic acid.5–12 The
latter oligomers adopt single helical conformations in solu-
tionat low concentration13–16 butalso possess the remarkable
ability to extend like springs and entwine to form double he-
lical dimers (Fig. 1).
Nature has found very diverse applications of the hybridiza-
tion of organic molecular strands into double or triple helical
architectures. In nucleic acids, sequence selective hybridiza-
tion is key to genetic information storage, duplication, and
transcription. In the triple helix of collagen or in coiled coils
(e.g., keratin) hybridization 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-barrel architecture that can bind metal ions in its
hollow.1 It is thus no surprise that chemists have made exten-
sive efforts to design and synthesize artificial oligomers that
can also wind around one another and form multiple helices.
Numerous families of synthetic oligomers have been re-
ported to wind into multiple stranded helices upon coordinat-
ing to transition metal ions.2 Helical hybrids held solely by
direct non-covalent interactions between organic strands
are much less common. They include oligoresorcinols,3
The peculiar hybridization of pyridine dicarboxamide oligo-
mers into double helices has been characterized in solution
in chlorinated solvents and in the solid state.5–11
Additionally, a computational approach has led to the hy-
pothesis that a slippage mechanism involving a series of
roller-coaster-like discrete steps operates during hybridiza-
tion.12 Intramolecular p–p stacking interactions within the
single helical monomers are replaced by intermolecular aro-
matic stacking when the tail of one of the strands proceeds
inside the other single helical strand in an eddy-like process.
From these studies, it can be concluded qualitatively that
hybridization is driven by intermolecular aromatic stacking,
which compensates the energy cost of extending and dou-
bling the helical pitch of the single helical monomers. The
competition between these two factors results in an unex-
pected trend of hybridization strength as a function of olig-
omers’ length as reported in Part 1 of this study.10 Indeed, the
*
See Ref. 10 for Part 1.
* Corresponding author. Fax: +33 (0)54000 2215; e-mail: i.huc@iecb.
u-bordeaux.fr
Present address: CAS Key Laboratory of Photochemistry, Institute of
y
Chemistry, Chinese Academy of Sciences, Beijing 100080, China.
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.02.114

D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
6323
only does not find its origin in interactions between side
chains but in the direct effect of the alkoxy substituents
upon main chain aryl–aryl interactions.
2. Results and discussion
2.1. Synthesis
Heptamer 1 having two terminal amine residues was pre-
pared by hydrogenolysis of the corresponding previously
described bis-benzylcarbamate precursor.10 The benzy-
loxy-substituted heptamer 2 was synthesized from 4-benzy-
loxy-2,6-diaminopyridine17 and 4-benzyloxy-2,6-pyridine-
dicarboxylic acid17 following
a convergent strategy
developed for heptamers having no substituents in position
4 of the pyridine rings (Scheme 1).13 This approach involves
two desymmetrization steps: one at the monomer level to
convert diamine monomer 4 into monoamine 5; and the
other at the trimer level to convert bis-Boc protected trimer
6 into monoamine 7.
Figure 1. Structure of oligomers 1–3. Intramolecular hydrogen bonds in-
volved in helical conformations of pyridine oligoamides and schematic rep-
resentation of single helix/double helix equilibrium implying a spring-like
extension/compression of the strands. The double headed arrows indicate
electrostatic repulsions.
observed dimerization constant was found to increase with
strand length to reach a maximum and then to decrease
down to undetectable levels upon increasing strand length
further.
However, numerous aspects of this hybridization phenome-
non remain unclear. For example, we have very early-on ob-
served that the presence of 4-decyloxy groups on all pyridine
rings lead to a dramatic enhancement of the dimerization
constant (Kdim). In CDCl3, a heptameric strand comprising
three 2,6-pyridine-dicarbonyl units and four 2,6-diaminopyr-
idine units show Kdim¼30 L molꢀ1 when it has 4-decyloxy
chains on the three 2,6-pyridine-dicarbonyl rings only, and
a Kdim¼6.5ꢁ104 L molꢀ1—a 2000-fold increase—when it
has 4-decyloxy chains on all rings!5,6 To explain this effect,
we proposed that multiple long side chains could undergo
extensive interstrand van der Waals interactions that would
stabilize the duplex.5,6 In a recent communication, we
showed that such side chain–side chain interactions do take
place between the aryl groups of 4-benzyloxy-substituted
oligomers and result in an enhanced hybridization.11 How-
ever, it remains unclear whether such interactions are also
taking place in the decyloxy-substituted oligomers. In partic-
ular, the poor crystallinity associated with long alkyl chains
has made it impossible to obtain crystals suitable for a solid
state analysis.
Scheme 1. Reagents and conditions: (a) THF, LiHMDS, (BOC)2O
(1 equiv), rt, 3 h, 65% yield; (b) CH2Cl2, 4-benzyloxypyridine-2,6-dicarbo-
nylchloride (0.45 equiv), Et3N, 0 ꢂC, 85% yield; (c) CH2Cl2, TMSI
(1 equiv), 30 min, then MeOH, reflux, 45 min, 62% yield; (d) CH2Cl2,
4-benzyloxypyridine-2,6-dicarbonylchloride (0.45 equiv), Et3N, 0 ꢂC,
71% yield; (e) CH2Cl2, TFA, 2 h, 99% yield.
Methoxy-substituted heptamer 3 was synthesized using a
strategy (Scheme 2) similar to that employed in the 4-decy-
loxy series using benzylcarbamate as amine protective
group.10 Monomers 4-methoxy-2,6-pyridine-dicarboxylic
acid dimethyl ester 9, 4-methoxy-2,6-pyridine-dicarboxylic
acid monomethyl ester 11, 4-methoxy-2,6-pyridine-dicar-
boxylic acid 10, 4-methoxy-2,6-diaminopyridine 12, and
2-amino-6-benzyloxycarbonylamino-4-methoxy-pyridine 13
were all obtained from chelidamic acid using standard pro-
cedures.10,17–20 Monoprotected amine 13 was coupled to
the acid chloride of 10 to yield trimer 16, which was hydro-
genated to diamine 17. Monoamine 13 was also coupled to
the acid chloride of 11 to yield dimer 14. After saponifica-
tion of ester 14 and activation on SOCl2, the acid chloride
of dimer 15 was coupled to diamine 17 to produce dipro-
tected heptamer 18. Hydrogenolysis of the benzylcarba-
mates of 18 gave the desired heptamer 3.
To further investigate the role of the side chains in the hybrid-
ization of pyridine carboxamide oligomers into double heli-
ces, we set to prepare and study oligomers 1–3 (Fig. 1). These
oligomers all possess an identical amine-terminated back-
bone bearing 4-alkoxy substituents on all pyridine rings:
4-decyloxy, 4-benzyloxy, and 4-methoxy for 1, 2, and 3, re-
spectively. Solid state and solution studies of their hybridiza-
tion behavior show that side chain–side chain interactions are
significant only the case of the benzyloxy-substituted oligo-
mer. Consequently, the high stability of the double helices
formed by 1 and 3 compared to related compounds having
alkoxy substituents on their 2,6-pyridine-dicarbonyl units

6324
D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
2.2. Solid state studies
Single crystal X-ray diffraction provided details about the
double helical structures of heptamers 2 and 3 (Table 1 and
Fig. 2). Colorless orthorhombic crystals of heptamer 2
suitable for crystallographic analysis were obtained upon
diffusion of MeOH into a DMSO–CHCl3 solution. The
asymmetric unit contains two molecules of 2, each belonging
to a distinct double helix possessing a C2-symmetry axis. The
two double helices closely resemble but are not related by
symmetry elements; one of them is shown in Figure 2a.
The double helical structure of 2, belongs to one of the two
main types of duplexes previously observed in oligoamides
of diaminopyridine and pyridine-dicarboxylic acid. These
two types of duplexes share a similar helical pitch (about
˚
7 A) and number of units per turn (around 4), but differ by
the relative position of the two stands with respect to one an-
other.7 In the case of 2, the first pyridine ring of a strand
stacks over the second pyridine ring of the other strand.
The screw offset between the two strands is thus approxi-
mately a quarter of a turn. Consequently, the helices consist
of stacks of pyridine rings that are alternatively diaminopyr-
idine and pyridine-dicarboxylic acid units. The duplex pos-
sesses a C2-symmetry axis perpendicular to the helical axis.
In contrast, an analogue of 2 having an additional two units
but the same side chains and terminal residues was shown to
belong to the second type of duplex in the solid state,11 with
the two strands helically offset by two aromatic groups (half
a turn), stacks of identical pyridine rings (either diamino-
pyridine or pyridine-dicarboxylic acid units) and a (pseudo)
C2-symmetry axis parallel to the helical axis.
Scheme 2. Reagents and conditions: (a) MeOH–dioxane, rt, NaOH
(5 equiv), 4 h, 99% yield; (b) MeOH–dioxane, 0 ꢂC, NaOH (1 equiv), 2 h,
85% yield; (c) THF, ꢀ78 ꢂC, n-BuLi (1 equiv), ClCO2Bn (1 equiv), 51%
yield; (d) toluene, 4-methoxypyridine-2,6-dicarbonylchloride (0.45 equiv),
i-Pr2EtN, 0 ꢂC, 78% yield; (e) DMF, MeOH, rt, H2/Pd–C, quant. yield; (f)
SOCl2, reflux, then 13, toluene, i-Pr2EtN, 0 ꢂC, 75% yield; (g) H2O–dioxane,
rt, NaOH (5 equiv), 2 h, 99% yield; (h) SOCl2, reflux, then 17 (0.45 equiv),
toluene, i-Pr2EtN, 0 ꢂC, 29% yield; (i) DMF, EtOAc, AcOH, 60 ꢂC, H2/
Pd–C, 74% yield.
An original feature of the structure of 2 resides in the way the
duplexes are arranged in the crystal lattice. Double helices
having the same handedness form columnar stacks, which
Table 1. Crystallographic data for 2 and 3
Compound
2
3
Formula
C90H73N15O13–(CH3OH)1.88–(H2O)3.56
1681.65
C48H45N15O13–(C2H5SO)2.25–(H2O)1.38
1237.90
FW (g molꢀ1
)
Cryst. dimensions (mm)
Solvent/precipitant
Crystal color
Crystal system
Space group
0.15ꢁ0.10ꢁ0.10
DMSO–CHCl3/MeOH
Colorless
Orthorhombic
Pnna
0.15ꢁ0.10ꢁ0.10
DMSO/CH3CN
Colorless
Monoclinic
C2/c
Z
a (A)
16
8
˚
˚
19.564(7)
49.684(19)
35.633(15)
90
34,636(2)
1.290
15.311(5)
30.149(6)
26.602(7)
100.3652(11)
12,079.8(6)
1.361
b (A)
˚
c (A)
b (ꢂ)
3
n (A )
˚
r (g cmꢀ3
)
T
l (A)
ꢀ153 ꢂC
1.54178 (Cu Ka)
6.5–50.4
17,860/2321
2106
0.2232
0.6277
1.014
6,31,915
ꢀ153 ꢂC
1.54178 (Cu Ka)
6.4–39.9
24,842/3522
815
0.1758
0.5084
1.06
6,31,916
˚
q measured (ꢂ)
Refl. measured/unique
No. of param.
R1 (I>2s(I))
WR2 (all data)
GOF
CCDC #

D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
6325
Figure 3. Views down the C-axis (top) and down the B-axis (bottom) of the
crystal structure of 2. The view down the C-axis shows a columnar stack of
double helices having the same handedness, which define a channel contain-
ing water molecules (red balls). The view down the B-axis illustrates the
hexagonal packing of the channels and the interdigitation of the benzyl
groups belonging to different channels.
interstrand face-to-face and edge-to-face interactions at var-
ious angles (Fig. 2a). No well-defined motif emerges in the
arrangement of these chains and not every benzyl group
lies in contact with a benzyl group of the same duplex. In
fact, the side chains are also interdigitated (and involved in
interduplex interactions) with neighboring molecules, as il-
lustrated in Figure 3. The structure of 2 may thus be consid-
ered as a snapshot of one of the many possible conformations
of the side chains. This structure is consistent with that of
a longer oligomer11 and demonstrates that multiple benzyl-
oxy residues at the periphery of a duplex do interact. As de-
tailed in Section 2.3, this results in an enhanced dimerization
constant in solution.
The structure of the double helix of methoxy-substituted
heptamer 3 could be solved from small single crystals
obtained upon diffusing CH3CN in a DMSO solution
(Fig. 2b). As expected, the methoxy groups are too small
to bring about significant interstrand side chain–side chain
interactions. Nevertheless, the structure of 3 is strikingly
similar to that of 2 (Fig. 2c). Even the sites where water
molecules are bound in the hollows of the helices are al-
most identical. This shows that the interactions between
the two main chains within a duplex define the relative
positioning of the two strands and overrule interactions
between side chains and interduplex interactions within
the crystal lattice. Importantly, these crystal structures do
not fundamentally differ from those of oligomers that do
not possess any alkoxy subsituent in position 4 of the
diaminopyridine units and which thus have considerably
lower dimerization constants: whatever is responsible for
a strongly enhanced hybridization when 4-alkoxy residues
are found on all rings, it does not lead to an observable
change in the solid state structures.
Figure 2. Crystal structures of the double helical dimers of 2 (a) and 3 (b). In
both cases, the main chains of the two strands of a duplex are color coded in
blue and red while the side chains are shown in gray and white. Water mol-
ecules included in the helix hollow are shown as gray spheres. Other in-
cluded solvent molecules are omitted for clarity. The superposition of the
main chains of duplexes of 2 and 3 is shown in (c).
define long channels along the B-axis—the hollows of the
helices—that are filled with water molecules (Fig. 3). The
water molecules reside at well-defined sites and form hydro-
gen bonds with the amide protons of the strands within the
helices. It is not clear whether the water molecules could
hop from site to site and whether the columnar stacks could
actually work as water channels.
At its periphery, the duplex is surrounded by 14 benzyloxy
chains. These are engaged in multiple intraduplex

6326
D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
Table 2. Dimerization constants of 1–3 at 25 ꢂC calculated from 400 MHz
2.3. Solution studies
1H NMR measurements
As has been extensively discussed previously,5–11 the charac-
terization of the double helices in solution and the calculation
of the dimerization constants Kdim are easily performed using
1H NMR. For heptameric strands like 1–3 or longer, the sin-
gle helical monomer and the double helical dimer are in slow
exchange on the NMR time scale and give rise to distinct sig-
nals, which can conveniently be integrated. The proportions
between monomer and dimer deduced from integral values
give the dimerization constants as previously described.5–11
A qualitative indication of the formation of the hybrids is
also provided by electrospray ionization mass spectrometry
(ESIMS). The double helical dimers often appear as low in-
tensity peaks corresponding to a singly charged double helix
[2M+H]+ (not shown) and much more intense peaks of the
doubly charged double helix [2M+2H]2+, which can be rec-
ognized by the fact that the peaks corresponding to the vari-
ous isotopes are separated by half a mass unit (Fig. 4). Every
other peak of the isotopic mass distribution of the doubly
charged double helix [2M+2H]2+ overlaps with a peak be-
longing to the isotopic mass distribution of the singly charged
single helix [M+H]+, which are separated by a full mass unit.
Thus the overall aspect of the isotopic distribution provides
a qualitative measure of the proportions between the single
helix [M+H]+ and double helix [2M+2H]2+. For example,
upon changing solvent or concentration, we observe that
the proportions between the single helix and double helix
signals on the mass spectra correlate to those observed by
NMR. However, no quantitative analysis could be derived
from the mass spectra.
Compound
1
2
3
Kdim (CDCl3)
Kdim (DMSO-d6)
6.9ꢁ104
52.8ꢁ104
8.2ꢁ104
a
—
120
320
a
Insoluble.
between benzyl groups do stabilize the double helix. This is
in agreement both with the solid state structure of 2 (Fig. 2)
and with previous results obtained for a longer analogue of 2
for which the effect is even more pronounced.11 Duplex sta-
bilization due to interstrand interactions between benzylic
side chains apparently does not operate in more polar sol-
vents like DMSO, in which methoxy-substituted oligomer
3 is actually found to hybridize slightly more strongly than 2.
The most striking result is the similar hybridization of decy-
loxy-substituted oligomer 1 and methoxy-substituted oligo-
mer 3. This shows that interactions between numerous long
aliphatic chains at the periphery of a double helix have no
effect on its stability and thus do not account, as was initially
hypothesized,5,6 for the high Kdim values of those oligomers,
which bear 4-alkoxy residues on all pyridine rings with re-
spect to those which do not. The very large effect of the alk-
oxy residues thus finds its origin in other factors. Given that
the crystal structures of 2 and 3 (Fig. 2) do not show any re-
markable difference from the solid state structures of less
stable double helices,5–9 the effect appears to belong to the
strength of p–p interactions between the two strands within
the duplex with respect to those of the single helical mono-
mers. This is surprising in a way because electron donor
groups such as alkoxy residues are a priori expected to en-
hance electron density of the pyridine p-clouds and thus to
weaken p–p interactions.
The dimerization constants of 1–3 measured in CDCl3 and
DMSO-d6 are shown in Table 2. All of them are characteris-
tic of oligomers bearing alkoxy residues in position 4 of all
pyridine rings. They are considerably higher than those of
comparable oligomers having no 4-alkoxy substituents on
the diaminopyridine rings, which are typically found in the
range of 100 L molꢀ1 in chlorinated solvents and below de-
tection levels in DMSO. The dimerization constant of the
benzyloxy substituted heptamer 2 in CDCl3 is almost 10-
fold higher than those of 1 and 3, suggesting that interactions
We also envisioned that 4-alkoxy substituents could enhance
the hydrogen bonding acceptor strength of the correspond-
ing endocyclic pyridine nitrogen.21 Some bifurcated inter-
strand hydrogen bonds which have occasionally been
Figure 4. Experimental high-resolution ESI mass spectrum of 2 showing the isotopic mass distribution of the single helical monomer [M+H]+, which adds to
that of the diprotonated double helical dimer [2M+2H]2+. The inset shows the theoretical isotopic distributions of [M+H]+ (top) and [2M+2H]2+ (bottom).

D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
6327
observed in the double helices of aromatic amide oligomers4
(though not in 2 and 3) could then be reinforced. However,
this hypothesis was ruled out after measuring that 2,6-diace-
tylamino-pyridine has the same hydrogen-bonding directed
dimerization constant whether it bears a 4-alkoxy substituent
or not (not shown). Finally, the most plausible, though un-
substantiated hypothesis that we could formulate concerned
dipolar effects. Considering that all the components that
constitute the oligomeric backbone—amide groups and pyr-
idine rings—possess relatively large dipole moments, it is
possible that some unfavorable dipolar interactions are re-
duced in those oligomers bearing 4-alkoxy residues on the
diaminopyridine rings, which are expected to possess
a much smaller dipole moment than pyridine itself.
400 MHz): d 0.90 (br, 30H), 1.33 (m, 140H), 2.39 (s, 68H),
3.64 (m, 36H), 4.19 (m, 28H), 4.90 (s, 4H), 5.33 (s, 2H),
5.48 (br, 10H), 6.71 (s, 4H), 6.87 (s, 1H), 6.99 (s, 2H), 7.52
(s, 2H), 7.59 (br, 8H), 7.66 (s, 4H), 7.77 (s, 2H), 7.83 (s, 2H),
7.98 (s, 2H), 8.21 (s, 1H), 8.24 (s, 1H), 10.9 (br, 12H), 10.32
(s, 2H), 10.60 (s, 2H), 10.73 (s, 2H). 13C NMR (CDCl3,
100 MHz): 14.1, 22.7, 25.9, 26.0, 29.1, 29.4, 29.6, 31.9,
66.5, 67.6, 68.1, 69.1, 69.3, 89.4, 91.7, 96.2, 110.4, 110.6,
110.7, 110.8, 111.1, 111.3, 149.1, 149.4, 149.6, 149.7,
149.9, 150.1, 150.2, 150.3, 150.5, 151.7, 157.4, 160.5,
160.8, 161.0, 161.1, 161.4, 161.5, 161.7, 161.8, 166.9,
167.0, 167.1, 167.2, 167.4, 167.5, 167.6, 167.7. IR (liquid
layer) nmax: l3319, 2924, 2854, 1701, 1610, 1578, 1528,
1439, 1343, 1174, 1047. HRMS calcd for C111H172N15O13:
1923.3259; found (ESI), 1923.3250 [M+H]+.
3. Conclusion
4.1.2. 2-Amino-4-benzyloxy-6-tertiobutoxycarbonyl-
amino-pyridine 5. A LiHMDS solution (21 mL, 1 M,
21 mmol, 2 equiv) was added to a solution of compound 4
(2.25 g, 10.46 mmol)17 in 40 mL THF at room temperature
over 15 min. After an additional 10 min, a solution of
di-tert-butyl dicarbonate (2.28 g, 1 equiv, 10.46 mmol) in
20 mLTHF was added dropwise for 20 min. And the reaction
was allowed to proceed at room temperature for 3 h. The sol-
vent was evaporated and the residue was purified by silica gel
chromatography using CH2Cl2/EtOAc 80:20 vol/vol to af-
ford compound 5 in 65% yield (2.14 g, 6.8 mmol) as a white
solid. Mp 105–107 ꢂC. 1H NMR (CDCl3, 400 MHz): d 1.50
(s, 5H), 1.54 (s, 4H), 4.31 (s, 2H), 5.07 (s, 2H), 5.77 (s, 1H),
6.88 (s, 1H), 7.06 (s, 1H), 7.39 (m, 5H). 13C NMR (CDCl3,
100 MHz): 28.2, 69.6, 80.7, 89.2, 89.7, 127.5, 128.0, 128.5,
136.2, 152.1, 152.5, 158.4, 168.2. MS calcd (C17H21N3O3),
315.16; found (TOFMS ES+), 315.9 [M+H]+.
In summary, this study has confirmed our preliminary re-
port11 that interstrand interactions between side chains
may stabilize the double helices of aromatic oligoamides
of 2,6-diaminopyridine and 2,6-pyridine-dicarboxylic acid
when the side chains consist of benzyl residues. Larger
aromatic side chains or side chains endowed with specific
recognition properties may thus be envisaged to amplify
this phenomenon. However, no significant side chain–side
chain interactions occur between exclusively aliphatic side
chains. Thus, the large dimerization constants of oligomers
such as 1 and 3 result from the direct effect of the alkoxy
substituents on interactions at the level of the aromatic back-
bone. It is fair to say that this effect, despite its large ampli-
tude, remains largely unexplained. Detailed investigations
involving the replacement of the alkoxy groups by substitu-
ents having various electron donor and acceptor strengths
may provide better insights into these complex phenomena
and are now underway.
4.1.3. Trimer 6 and general procedure for coupling reac-
tions. 4-Benzyloxy pyridine-dicarboxylic acid17 (0.79 g,
2.88 mmol, 1 equiv) and SOCl2 (15 mL) were heated to re-
flux. Excess SOCl2 was removed by distillation after 1.5 h.
Anhydrous toluene (10 mL) was added to the residue, and
then evaporated to azeotrope remaining SOCl2 to give the
corresponding diacid chloride. To a solution of compound
5 (2 g, 6.35 mmol, 2.2 equiv) and triethylamine (2.3 mL,
15.84 mmol, 5.5 equiv) in CH2Cl2, a solution of the above-
mentioned diacid chloride in anhydrous CH2Cl2 (10 mL)
was added over a period of 10 min at 0 ꢂC. The reaction mix-
ture was then stirred at room temperature for 2 h. The solvent
was removed under reduced pressure and the residue was
purified by silica gel chromatography using CH2Cl2/EtOAc
98:2 vol/vol to afford compound 6 in 85% yield (2.12 g,
4. Experimental section
4.1. General
Solvents (THF, toluene, CH2Cl2) were dried by filtration
over activated alumina on a commercially available setup.
FTIR spectra were recorded on a Brucker IFS 55 FTIR spec-
trometer. H NMR (400 MHz) and 13C NMR (100 MHz)
1
spectra were recorded on a Bruker 400 Ultrashield spectrom-
eter. The chemical shifts are expressed in parts per million
(ppm) using the residual solvent peak as an internal standard.
The following notations are used for the H NMR spectral
splitting patterns: singlet (s), doublet (d), triplet (t), multiplet
(m), broad (br). Melting points are uncorrected.
1
1
2.45 mmol) as a white solid. Mp 129–131 ꢂC. H NMR
(CDCl3, 400 MHz): d 1.52 (s, 10H), 1.60 (s, 8H), 5.18 (s,
4H), 5.29 (s, 2H), 7.18 (s, 2H), 7.42 (m, 18H), 7.83 (s, 2H),
8.04 (s, 2H), 10.05 (s, 2H). 13C NMR (CDCl3, 100 MHz):
28.2, 70.1, 70.8, 81.1, 95.1, 96.3, 112.3, 127.6, 128.2,
128.5, 134.7, 135.8, 150.2, 150.6, 151.6, 152.1, 161.5,
167.8, 168.5. IR (liquid layer): nmax: 3301, 2979, 2932,
1731, 1693, 1612, 1582, 1509, 1474, 1392, 1367, 1350,
1289, 1230, 1169, 1082, 1047, 1028. HRMS calcd for
C48H49N7NaO9: 890.3489; found (ESI), 890.3508 [M+Na]+.
4.1.1. Heptamer 1. A mixture of the bis-benzylcarbamate
protected precursor of 1 (7.5 mg, 0.0034 mmol)10 dissolved
in DMF/EtOAc 1:1 vol/vol (6 mL), 10% Pd/C (75 mg), and
acetic acid (1 mL) was stirred in an autoclave under 4 bar
of hydrogen at 55 ꢂC for 3 days. After cooling, the mixture
was filtered through Celite. The solvents were removed under
reduced pressure. The residue was purified by silica gel chro-
matography using CH2Cl2/EtOAc 96:4 vol/vol to afford
compound 1 in 76% yield (5 mg, 0.0026 mmol) as a light yel-
4.1.4. Trimer 7. TMSI (0.5 g, 0.38 mL, 2.5 mmol,
1.1 equiv) was added through a dry syringe to a solution of
6 (2 g, 2.3 mmol, 1 equiv) dissolved in dichloromethane
1
low waxy solid. H NMR (CDCl3, 1 mM, 25 ꢂC, dimer,

6328
D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
(50 mL). The mixture was stirred at room temperature for
30 min. The solvent was then evaporated; the residue was
dissolved in methanol (30 mL) and heated to reflux for
45 min. Excess triethylamine was added. After evaporation
of the solvent, the solid residue was purified by silica gel
chromatography using CH2Cl2/EtOAc 98:2 vol/vol to afford
compound 7 in 62% yield (1.1 g, 1.43 mmol) as a white
solid. Mp 107–109 ꢂC. 1H NMR (CDCl3, 400 MHz):
d 1.53 (s, 9H), 4.49 (br, 2H), 5.14 (s, 2H), 5.18 (s, 2H), 5.28
(s, 2H), 5.91 (s, 1H), 7.42 (m, 22H), 7.63 (s, 1H), 7.83 (s,
1H), 8.03 (s, 2H), 10.22 (s, 1H), 10.25 (s, 1H). 13C NMR
(CDCl3, 100 MHz): 28.2, 69.9, 70.7, 81.1, 90.5, 92.1, 95.1,
196.2, 112.1, 128.2, 127.6, 128.6, 134.7, 135.9, 150.3,
151.6, 152.2, 158.2, 161.5, 167.7, 168.4. IR (liquid layer):
nmax 3366, 3315, 3034, 2979, 2932, 1729, 1694, 1613,
1582, 1532, 1447, 1392, 1368, 1348, 1335, 1291, 1233,
1170, 1058, 1047, 1028. MS calcd (C43H41N7O7), 767.31;
found (TOFMS ES+), 768.1 [M+H]+ 790.1 [M+Na]+.
phase was evaporated and dried under vacuum. The product
contaminated with some starting diester 9 was used without
further purification. Mp 159–161 ꢂC. 1H NMR (CDCl3,
400 MHz): d 4.00 (s, 3H), 4.01 (s, 3H), 7.83 (s, 1H), 7.87
(s, 1H). 13C NMR (CDCl3, 100 MHz): 53.1, 56.3, 115.6,
148.2, 164.2, 168.6. MS calcd (C9H9NO5), 211.05; found
(TOFMS ES+), 212.1 [M+H]+.
4.1.8. 2-Amino-6-benzyloxycarbonylamino-4-methoxy-
pyridine 13. To a solution of 4-methoxy-2,6-diaminopyri-
dine 12 (1.5 g, 10.8 mmol)20 in anhydrous THF (60 mL) at
ꢀ78 ꢂC, a 2 M solution of n-butyllithium in hexane (2.7 mL,
1 equiv) was added slowly. After 15 min, benzyl chlorofor-
mate (0.88 mL, 1 equiv) was added at once. The mixture
was stirred at ꢀ78 ꢂC for 5 h, then at room temperature for
12 h. The reaction was quenched with a small amount of
water. The solvent was evaporated, and the residue was dis-
solved in CH2Cl2. This solution was washed with water, dried
over MgSO4, filtered, and evaporated. The residue was puri-
fied by silica gel chromatography using CH2Cl2/EtOAc 96:4
vol/vol to afford compound 13 in 51% yield (1.5 g,
4.1.5. Heptamer 8. Compound 8 was prepared from 4-
benzyloxy pyridine-dicarboxilic acid17 (76 mg, 0.28 mmol,
1 equiv) and trimer 7 (0.47 g, 0.62 mmol, 2.2 equiv) using
the general procedure for coupling described above. The
product was purified by silica gel chromatography using
CH2Cl2/EtOAc 98:2 vol/vol to afford compound 8 in 71%
yield (354 mg, 0.2 mmol) as a white solid. Mp 141–
1
5.5 mmol) as an off-white solid. Mp 91–93 ꢂC. H NMR
(CDCl3, 400 MHz): d 1.57 (s, 3H), 3.79 (s, 3H), 4.21 (s, 2H),
5.19 (s, 2H), 5.71 (s, 1H), 6.99 (s, 1H), 7.37 (s, 3H). 13C NMR
(CDCl3, 100 MHz): 55.2, 66.9, 88.8, 89.3, 128.2, 128.5,
135.8, 151.6, 153.0, 158.5, 169.1. MS calcd (C14H15N3O3),
273.11; found (TOFMS ES+), 274.1 [M+H]+.
1
143 ꢂC. H NMR (DMSO-d6, 1 mM, single helix, 75 ꢂC,
400 MHz): d 1.32 (s, 18H), 5.12 (s, 4H), 5.21 (s, 4H), 5.38
(s, 4H), 5.50 (s, 2H), 7.10 (s 2H), 7.42 (br, 37H), 7.85 (s,
2H), 7.89 (s, 2H), 8.05 (s, 2H), 8.81 (s, 2H), 10.13 (s, 2H),
10.34 (s, 2H), 10.74 (s, 2H). 13C NMR (CDCl3, 100 MHz):
30.1, 67.9, 70.4, 81.3, 94.2, 97.0, 114.1, 128.9, 129.8,
133.5, 135.1, 141.0, 152.1, 152.6, 161.9, 172.3. IR (liquid
layer) nmax: 3312, 3033, 2923, 2852, 1733, 1699, 1609,
1580, 1511, 1435, 1377, 1220, 1169, 1123, 1058, 1046.
MS calcd (C100H89N15O17), 1771.65; found (TOFMS
ES+), 1772.17 [M+H]+.
4.1.9. Dimer ester 14. A solution of acid 11 (1 g, 4.74 mmol)
in thionyl chloride (5 mL) was heated to reflux for 30 min.
Excess thionyl chloride was distilled under reduced pressure,
and azeotroped with dry toluene. The residue was dissolved
in dry toluene (10 mL). To this solution at 0 ꢂC, a solution
of amine 13 (1.16 g, 4.27 mmol, 0.9 equiv) in dry toluene
(10 mL) was added, followed by distilled N,N-diisopropyl-
ethylamine (4 mL, 23.7 mmol, 5 equiv). The mixture was al-
lowed to warm to ambient temperature and stirred overnight.
The solvent was removed and the residue was purified by sil-
ica gel chromatography using CH2Cl2/EtOAc 96:4 vol/vol to
afford dimer 14 in 75% yield (1.65 g, 3.55 mmol) as a white
solid. Mp 171–173 ꢂC. 1H NMR (CDCl3, 400 MHz): d 1.55
(s, 3H), 3.91 (s, 3H), 3.99 (s, 3H), 4.02 (s, 3H), 5.22 (s,
2H), 7.40 (m, 5H), 7.71 (s, 1H), 7.77 (s, 1H), 7.93 (s, 1H),
10.20 (s, 1H). 13C NMR (CDCl3, 100 MHz): 53.0, 55.6,
56.1, 67.1, 94.4, 95.5, 110.6, 114.5, 128.1, 128. 6, 135.7,
150.2, 151.1, 152.8, 161.6, 168.1, 169.5. HRMS calcd for
C23H23N4O7: 467.1567; found (ESI), 467.1548 [M+H]+.
4.1.6. Heptamer 2. Diprotected heptamer 8 (250 mg,
0.14 mmol) was dissolved in CH2Cl2 (5 mL) and TFA
(2 mL) was added. After 2 h, the reaction mixture was
washed with saturated NaHCO3 solution and extracted with
chloroform (2ꢁ50 mL). The solvent was evaporated to
afford compound 2 in 99% yield (204 mg, 0.13 mmol) as
a white solid. Mp 278–280 ꢂC. 1H NMR (DMSO-d6,
2 mM, 65 ꢂC, single helix, 400 MHz): d 5.01 (s, 4H), 5.24
(s, 4H), 5.32 (s, 8H), 5.44 (s, 4H), 5.77 (s, 2H), 6.96 (s,
2H), 7.41 (br, 35H), 7.69 (s, 2H), 7.72 (s, 2H), 7.76 (s, 2H),
7.79 (s, 2H), 7.92 (s, 2H), 10.13 (s, 2H), 10.55 (s, 2H),
11.02 (s, 2H). 13C NMR (CDCl3, 100 MHz): 69.4, 70.4,
89.7, 91.7, 97.1, 111.2, 127.3, 127.5, 128.4, 128.5, 134.7,
135.5, 150.1, 150.4, 157.6, 161.4, 167.3. HRMS calcd for
C90H74N15O13: 1572.5591; found (ESI), 1572.5575 [M+H]+.
4.1.10. Dimer acid 15. Compound 15 was prepared from es-
ter 14 (1.5 g, 3.2 mmol) and NaOH (5 equiv) using a proce-
dure comparable to that of the preparation of 11. The product
was used without further purification (1.4 g, 3.1 mmol,
1
99%). Mp 245–247 ꢂC. H NMR (DMSO-d6, 400 MHz):
d 3.86 (s, 3H), 4.01 (s, 3H), 5.18 (s, 2H), 7.30 (s, 1H), 7.41
(m, 5H), 7.59 (s, 1H), 7.77 (s, 1H), 7.86 (s, 1H), 10.37 (s,
1H), 10.44 (s, 1H). 13C NMR (CDCl3, 100 MHz): 55.4,
56.3, 65.7, 66.9, 94.2, 94.8, 110.7, 113.5, 127.7, 127.3,
150.0, 152.2, 153.3, 165.0, 168.0, 168.4. MS calcd
(C22H20N4O7), 452.13; found (TOFMS ES+), 453.1
[M+H]+, 475.1 [M+Na]+.
4.1.7. 4-Methoxy-2,6-pyridine-dicarboxylic acid mono-
methyl ester 11. Dimethyl 4-methoxypyridine 2,6-dicarb-
oxylate 9 (2 g, 8.89 mmol)20 was dissolved in 1,4-dioxane
(32 mL) and methanol (8 mL) and the solution was cooled
to 0 ꢂC. Sodium hydroxide (355 mg, 8.89 mmol, 1 equiv)
was added and the mixture was stirred at 0 ꢂC for 2 h and an-
other 2 h at ambient temperature. The solution was neutral-
ized with acetic acid and poured into water (50 mL). The
product was extracted with CH2Cl2 (2ꢁ50 mL). The organic
4.1.11. Trimer 16. Compound 16 was prepared from amine
13 (0.4 g, 1.46 mmol, 1 equiv) and 4-methoxypyridine-2,6-

D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
6329
dicarboxylic acid 10 (132 mg, 0.67 mmol, 0.45 equiv)19 (see
4.2. X-ray crystallography
Section 4.1.3). The residue was purified by flash chromato-
graphy on silica gel using CH2Cl2/MeOH 99:1 vol/vol to
afford compound 16 in 78% yield (0.25 g, 0.35 mmol) as a
white solid. Mp 139–141 ꢂC. 1H NMR (CDCl3, 400 MHz):
d 1.70 (m, 10H), 3.90 (s, 6H), 3.99 (s, 3H), 5.09 (s, 4H),
7.27 (m, 2H), 7.71 (s, 2H), 7.74 (s, 2H), 7.88 (s, 2H), 10.31
(s, 2H). 13C NMR (CDCl3, 100 MHz): 46.3, 50.8, 55.6,
67.2, 94.4, 95.3, 95.7, 108.7, 111.5, 120.7, 127.8, 128.5,
133.2, 137.7, 138.5, 143.8, 151.7, 156.5, 157.2, 161.5,
162.1, 166.2, 168.9, 169.4. HRMS calcd for C36H34N7O9:
708.2404; found (ESI), 708.2424 [M+H]+.
Single crystals of heptamers 2 and 3 were mounted on a Ri-
gaku R-Axis Rapid diffractometer equipped with a MM007
micro focus rotating anode generator with monochromatized
˚
Cu Ka radiation (1.54178 A). The data collection, unit cell
refinement, and data reduction were performed using the
CrystalClear software package. The positions of non-H
atoms were determined by the program SHELXS 87, and
the position of the H atoms were deduced from coordinates
of the non-H atoms and confirmed by Fourier synthesis. H
atoms were included for structure factor calculations but
not refined.
4.1.12. Trimer diamine 17. A mixture of trimer 16 (100 mg,
0.14 mmol) dissolved in DMF (5 mL) and methyl alcohol
(5 mL), and of 10% Pd/C (10 wt %) was stirred overnight
under hydrogen at atmospheric pressure. The mixture was
filtered through Celite. The solvents were removed under re-
duced pressure to give the product, which was used without
further purification. Mp 114–116 ꢂC. 1H NMR (CDCl3,
400 MHz): d 3.86 (6H, s), 4.01 (s, 3H), 4.80 (s, 4H), 5.64
(s, 6H), 5.83 (s, 2H), 7.52 (s, 2H), 7.95 (s, 2H), 10.59 (s,
2H). 13C NMR (CDCl3, 100 MHz): 46.3, 50.7, 55.7, 67.2,
94.3, 95.3, 95.7, 108.7, 111.6, 151.6, 156.5, 157.5, 161.6,
162.2, 166.2, 168.8, 169.5. MS calcd (C20H21N7O5),
439.16; found (TOFMS ES+), 440.21 [M+H]+.
Acknowledgements
This work was supported by the Centre National de la
Recherche Scientifique, the Universities of Bordeaux 1
and Victor Segalen Bordeaux 2, by the Ministere de la
Recherche (postdoctoral fellowships to D.H. and H.J.),
and by ANR grant NT05-3_44880. We are also grateful
to Ms. Bathany for assistance with MS measurements
and to S. Massip for helping with the crystallographic
analysis.
ꢁ
4.1.13. Heptamer 18. Compound 18 was prepared from
dimer acid 15 (170 mg, 0.37 mmol, 1 equiv) and trimer
diamine 17 (75 mg, 0.17 mmol, 0.45 equiv). The residue
was purified by silica gel chromatography using CH2Cl2/
MeOH 99:1 vol/vol to afford compound 18 in 29% yield
References and notes
1. Langs, D. A. Science 1988, 241, 188; Burkhart, B. M.;
Gassman, R. M.; Langs, D. A.; Pangborn, W. A.; Duax,
W. L.; Pletnev, V. Biopolymers 1999, 51, 129.
2. For reviews on helicates: Albrecht, M. Chem. Rev. 2001, 101,
3457; Piguet, C.; Bernardinelli, G.; Hopfgartner, G. Chem.
Rev. 1997, 97, 2005.
3. Goto, H.; Katagiri, H.; Furusho, Y.; Yashima, E. J. Am. Chem.
Soc. 2006, 128, 7176.
4. Tanaka, Y.; Katagiri, H.; Furusho, Y.; Yashima, E. Angew.
Chem., Int. Ed. 2005, 44, 3867; Furusho, Y.; Tanaka, Y.;
Yashima, E. Org. Lett. 2006, 8, 2583.
1
(70 mg, 0.05 mmol) as a white solid. Mp 159–161 ꢂC. H
NMR (DMSO-d6, 75 ꢂC single helix, 400 MHz): d 1.26
(m, 6H), 3.88 (s, 7H), 3.93 (s, 7H), 4.08 (s, 7H), 5.00 (s,
4H), 6.96 (s, 2H), 7.26 (s, 2H), 7.41 (s, 2H), 7.44 (s, 2H),
7.71 (s, 2H), 7.77 (s, 2H), 9.42 (s, 2H), 10.06 (s, 2H),
10.24 (s, 2H), 10.61 (s, 2H). 13C NMR (CDCl3, 100 MHz):
44.9, 45.1, 50.7, 51.8, 56.0, 62.4, 66.5, 67.5, 84.7, 90.2,
94.8, 97.5, 99.8, 100.7, 103.7, 112.7, 110.7, 119.4, 120.6,
121.1, 127.9, 128.3, 136.9, 139.7, 141.1, 142.6, 151.6,
154.3, 156.2, 159.9, 161.9, 165.9, 167.6, 169.9, 172.6,
184.6. MS calcd (C64H57N15O17), 1307; found (TOFMS
ES+), 1308.32 [M+H]+.
5. Berl, V.; Huc, I.; Khoury, R. G.; Krische, M. J.; Lehn, J.-M.
Nature 2000, 407, 720.
6. Berl, V.; Huc, I.; Khoury, R. G.; Lehn, J.-M. Chem.—Eur. J.
2001, 7, 2810.
ꢀ
4.1.14. Heptamer 3. A mixture of heptamer 18 (30 mg,
0.023 mmol) and 10% Pd/C (10 wt %) in DMF/EtOAc/
AcOH (10:10:1, vol/vol/vol, 10.5 mL) was placed in auto-
clave under 4 bar of hydrogen and was heated at 60 ꢂC for
36 h. The solvents were evaporated and the residue was
purified by silica gel chromatography using CH2Cl2/
MeOH 99:1 vol/vol to afford compound 3 in 74% yield
7. Maurizot, V.; Leger, J.-M.; Guionneau, P.; Huc, I. Russ. Chem.
Bull. 2004, 7, 1572.
ꢀ
8. Zhan, C.; Leger, J.-M.; Huc, I. Angew. Chem., Int. Ed. 2006,
45, 4625.
9. Dolain, C.; Zhan, C.; Leger, J.-M.; Huc, I. J. Am. Chem. Soc.
2005, 127, 2400.
10. Jiang, H.; Maurizot, V.; Huc, I. Tetrahedron 2004, 60,
10029.
11. Haldar, D.; Jiang, H.; Leger, J.-M.; Huc, I. Angew. Chem., Int.
Ed. 2006, 45, 5483.
12. Acocella, A.; Venturini, A.; Zerbetto, F. J. J. Am. Chem. Soc.
2004, 126, 2362.
13. Berl, V.; Huc, I.; Khoury, R. G.; Lehn, J.-M. Chem.—Eur. J.
2001, 7, 2798.
ꢀ
1
(18 mg, 0.017 mmol) as a white solid. Mp 164–166 ꢂC. H
ꢀ
NMR (CDCl3, dimer, 1 mM, 400 MHz): d 3.96 (s, 12H),
4.06 (s, 9H), 5.45 (s, 4H), 4.94 (s, 4H), 6.75 (s, 7H),
6.99 (s, 7H), 7.58 (s, 9H), 7.65 (s, 10H), 7.73 (s, 7H),
10.08 (s, 2H), 10.16 (s, 1H), 10.25 (s, 1H), 10.59 (s, 1H),
10.84 (s, 1H). 13C NMR (CDCl3, 100 MHz): 44.8, 45.1,
50.7, 51.8, 55.9, 62.7, 66.4, 67.5, 84.7, 90.2, 94.8, 97.5,
99.8, 100.7, 103.1, 112.7, 110.7, 151.6, 154.3, 156.2,
159.9, 161.8, 165.9, 167.6, 169.8, 172.6, 184.7. HRMS calcd
for C48H46N15O13: 1040,3400; found (ESI), 1040.3388
[M+H]+.
ꢀ
14. Huc, I.; Maurizot, V.; Gornitzka, H.; Leger, J.-M. Chem.
Commun. 2002, 578.
ꢀ
15. Dolain, C.; Grelard, A.; Laguerre, M.; Jiang, H.; Maurizot, V.;
Huc, I. Chem.—Eur. J. 2005, 11, 6135.

6330
D. Haldar et al. / Tetrahedron 63 (2007) 6322–6330
€
19. M€uller, W. M.; M€uller, U.; Mieden-Gundert, G.; Vogtle, F.;
16. Maurizot, V.; Dolain, C.; Huc, I. Eur. J. Org. Chem. 2005, 1293.
17. Braxmeier, T.; Demarcus, M.; Fessmann, T.; McAteer, S.;
Kilburn, J. D. Chem.—Eur. J. 2001, 7, 1889; Tamura, N.;
Mitsui, K.; Nabeshima, T.; Yano, Y. J. Chem. Soc., Perkin
Trans. 2 1994, 2229.
ꢀ
Lescanne, M.; Heuze, K.; D’Aleo, A.; Fages, F. Eur. J. Org.
ꢀ
Chem. 2002, 2891.
20. Markees, D. G.; Kidder, G. W. J. Am. Chem. Soc. 1956,
78, 4130.
21. Chang, S.-Y.; Kim, H. S.; Chang, K.-J.; Jeong, K.-S. Org. Lett.
2004, 6, 181.
18. Sugano, Y.; Kittaka, A.; Otsuka, M.; Ohno, M.; Sugiura, Y.;
Umezawa, H. Tetrahedron Lett. 1986, 27, 3635.