Synthetic and structural studies on macrocyclic amino cyclitols - Conformational chameleons

Article abstract of DOI:10.1039/b716347a

Starting from quinic acid 7 the synthesis of 1,4-butanediol-linked macrocyclic aminocyclitols 30, 32, 34, 36 and 38 is described. Assembly was achieved by olefin cross-metathesis of appropriate cyclohexyl allyl ethers followed by ring-closing metathesis of bis-O-allyl homodimers. In all five cases studied, the only products that were formed were those resulting from direct ring-closing metathesis; the formation of larger rings was not detected. These macrocycles exhibited diverse conformational behaviour which included formation of stable separable conformers 31a and 31b as well as conformationally dynamic macrocycles 35 in which a ring flip in one cyclohexane chair conformer induces a ring flip of the other cyclohexane ring through the linking chains of the macrocycles. The activation energy for the inversion of the chair conformation in this process was determined to be about 38 kJ mol^-1, which is about 7 kJ mol^-1 lower than the activation energy for the ring flip of the unsubstituted cyclohexane ring. In all cases, the conformational studies strongly suggest that intramolecular H-bonding between 1,3-diaxially oriented amido and alcohol or ether groups exerts a decisive contribution to the overall stabilisation of the preferred cyclohexane chair conformation.

Full text of DOI:10.1039/b716347a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthetic and structural studies on macrocyclic amino cyclitols –
conformational chameleons†
Benjamin Oelze, Dieter Albert and Andreas Kirschning*
Received 23rd October 2007, Accepted 7th April 2008
First published as an Advance Article on the web 7th May 2008
DOI: 10.1039/b716347a
Starting from quinic acid 7 the synthesis of 1,4-butanediol-linked macrocyclic aminocyclitols 30, 32, 34,
36 and 38 is described. Assembly was achieved by olefin cross-metathesis of appropriate cyclohexyl allyl
ethers followed by ring-closing metathesis of bis-O-allyl homodimers. In all five cases studied, the only
products that were formed were those resulting from direct ring-closing metathesis; the formation of
larger rings was not detected. These macrocycles exhibited diverse conformational behaviour which
included formation of stable separable conformers 31a and 31b as well as conformationally dynamic
macrocycles 35 in which a ring flip in one cyclohexane chair conformer induces a ring flip of the other
cyclohexane ring through the linking chains of the macrocycles. The activation energy for the inversion
of the chair conformation in this process was determined to be about 38 kJ mol−1, which is about
7 kJ mol−1 lower than the activation energy for the ring flip of the unsubstituted cyclohexane ring. In all
cases, the conformational studies strongly suggest that intramolecular H-bonding between 1,3-diaxially
oriented amido and alcohol or ether groups exerts a decisive contribution to the overall stabilisation of
the preferred cyclohexane chair conformation.
Introduction
Many natural products of pharmaceutical relevance are composed
of macrocycles with embedded smaller rings such as pyrans or
furans (often present as lactols or lactones). Typical examples
are the polyketides bryostatin and sorangicin, and the diterpene
tonantzitlolone. Their ability to target their receptors is very likely
closely related to their conformational flexibility. On one hand
the macrocycle adopts a preferred global conformation when
being bound to the biomacromolecule.1 But besides the influence
imposed by the receptor it can be assumed that the conformational
characteristics of the embedded five- or six-membered rings will
also influence the macrocycle conformation in a relay type mode
of action. Relay of conformational information has emerged as a
new important research field for so-called small molecules.2–4
Recently, we and other research groups have been involved in
synthesising aminoglycoside antibiotics and derivatives as well as
mimetics that are able to selectively bind to biomacromolecules.5–8
In this context, we described the first preparation of novel
macrocyclic 1,4-butanediol-linked aminodeoxyglycosides 3 and 4
(Fig. 1),8 which can be regarded as glycomimetics derived from
Fig. 1 Structures of neomycin (1) and streptomycin (2), and macrocyclic
neoaminoglycosides 3 and 4.
aminoglycoside antibiotics. Like the natural products described
above, these glycosides are composed of a macrocycle with
embedded pyranose rings. In detailed NMR studies,9 we showed
that the conformational flexibility of the macrocycle, as well as
conformational changes in the pyranose units, were responsible
for 3 forming a complex with TAR RNA.10
Aminoglycoside antibiotics like neomycin (1) and streptomycin
(2) are commonly not only composed of hexoses or pentoses but
also contain cylictols such as the 2-deoxystreptamine ring system
(3). In continuation of our recent work, we now disclose detailed
synthetic and conformational studies on aminocyclitols embedded
in a macrocyclic environment, as depicted in the general structures
5 and 6 (Fig. 2). Compared to pyranoses the cyclohexane ring
shows a larger conformational flexibility,10 which in the present
case should have an effect on the outcome of the RCM and
the conformational dynamics of the macrocycles formed. In this
Institut fu¨r Organische Chemie, Leibniz Universita¨t Hannover, Schneiderberg
1B, D-30167, Hannover, Germany. E-mail: andreas.kirschning@oci.uni-
hannover.de; Fax: int. +49-(0)511-762 3011
† Electronic supplementary information (ESI) available: Additional
experimental information; determination of E/Z ratios; discussion
of compounds 29a,b, 33 and 37; crystal structure data. See DOI:
10.1039/b716347a
2412 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
report, we provide detailed synthetic studies and conformational
analyses on two out of four macrocycles of type 5 and 6 that can
also be of relevance for macrocyclic natural products containing
six-membered rings within a macrocycle.
Fig. 2 General formulae of macrocyclic aminocyclitols 5 and 6.
Results and discussion
Synthesis of macrocycles
The syntheses of macrocyclic cyclitols utilised D-quinic acid 7 as
the starting building block (Scheme 1).11 It readily provides four
stereogenic centres of which three can be conserved for the target
products. Quinic acid 7 was converted via known sequences into
ketone 8,12–14 which was then transformed into the corresponding
oxime 9. Reduction to the corresponding amine was achieved
with sodium borohydride in the presence of a catalytic amount of
nickel(II) acetate, a rather uncommon reagent system, but in our
hands the best we tested for this transformation.15 After protection
as trifluoroacetamide and formation of diastereoisomers 10a and
10b followed by O-allylation at C-5, two cyclitol derivatives
11a and 11b (2 : 1) were obtained and separated. However, we
encountered distortion of the cyclohexane ring by the annelated
acetonide, which hampered determination of the configuration
at the amido-substituted stereogenic center at C-1. Removal of
the this protection yielded diols 12a and 12b, at which point
structure elucidation of both diastereoisomers became possible.
Diastereoisomer 10a and all products derived therefrom have
the L-configuration, while 10b was determined as having the D-
configuration.16
Scheme 1 Preparation of homodimers 16 and 17. Reagents and conditions:
a) HONH3Cl, NaOAc, MeOH, rt, 24 h (95%); b) NaBH4, cat. Ni(OAc)2,
MeOH, rt, 3 h; c) CF3CO2Et, Et3N, MeOH, rt, 24 h (93% over two steps);
d) allyl iodide, Ag2O, MeCN, 45 ◦C, 24 h (75–85%); e) conc. HCl(aq)
,
In the following transformations, the allyl moiety served as a
functional group for metathesis olefinations. Thus, independent
treatment of allyl ethers 11a and 11b with the Grubbs I catalyst 1317
yielded homodimers (E/Z = 10 : 1, configurational assignment
was conducted with compound 18 which is discussed in detail in
the ESI†).18 However, the transformation was incomplete, and so
the starting material was re-isolated and re-employed in order to
achieve almost quantitative yield.
The double bonds were then hydrogenated (to furnish homo-
dimers 14 and 15) and the acetal groups were hydrolysed to
yield tetrols 16 and 17 (Scheme 1). At this stage, we chemically
distinguished both hydroxy groups on each cyclohexane ring
by stannylidene formation and ring opening using allyl iodide
as an electrophile, which yielded allyl ethers 19–21, 25 and 26
(Scheme 2).19 Usually, this method is very selective in distinguish-
ing syn alcohol groups.20 Nevertheless, we encountered a total
absence of selectivity in the case of the L-diastereoisomer 16 (1 : 1
allylation at O-3 and O-4) and only a slight preference at O-4 for the
D-diastereoisomer 17 (1.7 : 1 in favour of O-4 allylation). As shown
in the second part of this report, most of these oligosubstituted
amino cyclohexanes are conformationally flexible, which affects
MeOH–H2O (10 : 1), 1 h, quantitative yield for 12a and 12b; f) 13, CH2Cl2,
37 ◦C, 52 h, 77% from 11a (23% recovered 11a) and 46% from 11b (50%
recovered 11b); g) PtO2, H2, EtOAc–CH2Cl2–MeOH (16 : 8 : 1), 16 h; h)
conc. HCl(aq), MeOH–H2O (10 : 1), 1 h (86% of 16 for two steps and 69%
of 17 for two steps); Tfa = trifluoroacetyl, Cy = cyclohexyl.
the outcome of reactions like the present protection protocol.
Separation of positional isomers was achieved after formation
of the corresponding acetates 22–24, 27 and 28.
With these dimeric bis-O-allylated 1,4-butanediol-linked cy-
clitols in hand, we conducted macrocyclisations under RCM
conditions, again employing the Grubbs I catalyst 13. In our
earlier work with O-allylated glycosides we found that the mode
of macrocyclisation depends on several factors, one of them being
the relative configuration of all substituents at the pyran ring.
Thus, starting from a diene precursor, direct macrocyclisation
(e.g. formation of macrocycle 4 with an arabino configuration
around the pyranose ring) or dimerisation and trimerisation (e.g.
formation of macrocycle 3 with a lyxo configuration around the
pyranose ring)8c can occur, making the outcome of the RCM
difficult to predict.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2413
©

View Article Online
Scheme 2 Reagents and conditions: a) Bu2SnO, toluene, D, 2 h; b) allyl bromide, CsF, DMF, 60 ◦C to rt, 18 h (from 16: 67% for two steps; from 17: 86%
for two steps); c) Ac2O, DMAP, Et3N, 30 min; quantitative yields.
In the present cases (acetates 22–24, 27 and 28) RCM yielded
only direct ring-closing products irrespective of the stereochem-
istry associated with the amido group, as was confirmed by mass
spectrometric analysis. Formation of larger macrocycles could not
be detected by TLC and MS after sampling the crude RCM
products. The free alcohols 19–21, 25 and 26 could also be
employed but RCM afforded lower yields. The RCM products are
again E/Z mixtures of C2-symmetric alkenes that can be analysed
by the NMR method described in the ESI†.18c After hydrogenation
of the newly formed olefinic double bond we obtained macrocycles
29, 31, 33, 35 and 37 (Scheme 3).
To our surprise each metathesis reaction yielded two or three
products with distinctly different TLC Rf values. However, in most
cases the compounds could not be separated by flash column
chromatography. The mixtures clearly revealed only one mass in
the mass spectra, while the NMR spectrum distinctly showed the
presence of different compounds or very broad signals, which
hampered detailed assignment. Careful investigations of these
products indicate the presence of conformational isomers, which
will be discussed in detail below. After having worked out this
behaviour, we terminated the sequence by deprotection under
standard conditions to yield the target macrocycles 30, 32, 34,
36 and 38. As for many other aminoalcohols of rather complex
structure, the NMR spectra of these products were difficult to fully
solve and interpret. Therefore, we first focused on the precursors
and protected aminoalcohols 29, 31, 33, 35 and 37.
based on the original labelling of the starting D-quinic acid 7. In
order to differentiate the cyclohexane rings with acetyl groups at
O-3 from those acetylated at O-4 we shall label the cyclohexanes
to be endo when the acetate group is at C-4 while exo-acetate
groups are at C-3 irrespective of whether acyclic RCM precursors
or macrocycles are being discussed. In fact, in macrocycles the C-3
position is located outside the macrocycle while the C-4 position
lies inside.
The most important tool used to determine the cyclohexane ring
3
conformation is the analysis of the J scalar coupling constants
according to the Karplus equation which can straightforwardly
can be adopted to cyclohexanes in ideal chair conformations.22
In the present cases, however, we often encountered broadened
signals in the 1H as well as 13C NMR spectra at room temperature,
which can be taken as an indication of a fast equilibrium relative
to the NMR timescale. If necessary, we also used NOE contacts to
determine the arrangement of atoms in space and thus elucidate
the cyclohexane chair conformation.
Acyclic precursors
We first conducted conformational studies on the cyclohexane
rings in precursor building blocks as well as on acyclic homod-
imers. These analyses turned out to be important for understand-
ing conformational peculiarities found in the macrocycles, which
are strongly influenced by the conformations of the individual
cyclohexane moieties present (vide supra) (Table 1).
1
In chloroform the cyclohexane ring in 12a adopts a C4 chair
Analytical and conformational considerations
conformation and the amido group is oriented in an axial position.
An intramolecular H-bond is possible and would stabilise the 1C4
conformation. The chemical shift of the NH proton in chloroform
at 295 K can be regarded as an indicator of intramolecular H-
bonding. NH chemical shifts higher than 7.4 ppm are always a
diagnostic tool for an intramolecular H-bond to an oxygen atom
Prior to discussing conformational aspects in detail, we wish to
1
4
define conformations C4 and C1 for clarity, in analogy to the
IUPAC rules for carbohydrates (Fig. 3).21 In this context we regard
the carbon C-6 as the pseudo ring oxygen atom. The numbering is
consistent with the labelling in all schemes in this paper, which is
2414 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Scheme 3 RCM of dienes 22, 23, 24, 27 and 28. Reagents and conditions: a) 13 (10 mol%), CH2Cl2, 37 ◦C; b) PtO2, H2, CH2Cl2–ethyl acetate–MeOH
(29: 43%; 31: 66%; 33: 53%; 35: 53%; 37: 39% for two steps); c) NaOH, H2O–MeOH (30: 67%; 32: 81%; 34: 83%; 36: 61%; 38: 86%).
4
adopts a C1 chair conformation with the amido group oriented
in an equatorial position for which no intramolecular H-bond is
possible.
NMR analyses of the more polar compounds 12b, 16 and 17 had
to be conducted in CD3OD instead of CDCl3. Consequently the
amido group occupies an equatorial position and no intramolec-
ular H-bond is formed.
present context.
Fig. 3 Definition of the prefixes “endo” and “exo” and 1C4 and 4C1 in the
RCM products
in a 1,3-diaxial relationship. The NH signal of the NHTfa group
of 12a in chloroform is located at 8.11 ppm and indicates an
intramolecular H-bond. On the other hand, in methanol, which
is a polar H-bond competing solvent, the cyclohexane ring of 12a
Conformational analysis was carried out with N-acylated macro-
cycles 29, 31, 33, 35 and 37 and with the fully deprotected
derivatives 30, 32, 34, 36 and 38. The NMR spectra of the latter
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2415
©

View Article Online
macrocycles commonly revealed poor resolution. In part, the
results collected for the protected macrocycles could be used for
the analysis of the latter ones. Among the five acylated macrocycles
only the conformations of the two macrocycles 31 and 35 could
be analysed in great detail. They showed remarkably different
behaviour (vide supra). Analysis of the other three macrocycles 29,
33 and 37 can be found in the ESI†. In essence, they show similar
conformational effects.
of 31a and 31b cannot unequivocally be attributed to a certain
chair conformation. The cyclohexane ring that holds the endo-
acetate group gives sharp NMR signals in both conformers of
31. From the coupling constants it can be concluded that the
4
1
endo-acetate ring adopts a C1 conformation in 31a and a C4
conformation in 31b. The chair conformations of the endo-acetate
1
cyclohexane rings (4C1 in 31a and C4 in 31b) cannot invert into
each other due to steric hindrance between the acetyl group
and the backbone of the macrocycle. The two conformers 31a
and 31b are stable for months in solution at room temperature.
Even in refluxing toluene (1 h) inversion of conformation was
not observed. Further structural proof was collected from X-ray
analysis of crystalline 31b (Fig. 4).23 In the crystal the cyclohexane
ring with the endo-acetyl group clearly adopts a 1C4-conformation,
while the cyclohexane ring with the exo-acetyl group shows a 4C1
conformation.
Free aminoalcohol 32 was independently prepared by deprotec-
tion of 31a or 31b, respectively. As is evident from the coupling
constants in Table 2, the whole ring systems appears to be highly
flexible, as only the amino group adopts an equatorial position,
whereas for the other ring protons no defined coupling constants
could be determined. This observation clearly reveals a change of
conformation with respect to the NMR timescale.
RCM products 31a,b
The diene precursor 23 is composed of an endo- and an exo-
acetate cyclohexane ring. For better recognition, the labels of
the exo-acetate cyclohexane rings are marked with a prime (ꢀ)
and are always placed on the right-hand side. In CDCl3 both
1
cyclohexane rings adopt a C4 conformation. The amido groups
at C-1 and at C-1ꢀ are placed in an axial position and can form a
stabilizing hydrogen bond with O-3 and O-3ꢀ. NMR investigations
in CDCl3 of the macrocycles 31a and 31b prepared from 23
(coupling constants J, selective TOCSY and selective NOESY
NMR experiments) showed that both compounds are identical
with respect to their stereo- and regiochemistry. Further proof
for these considerations was collected when both conformers 31a
and 31b separately as well as a mixture of both conformers were
independently deacetylated to yield macrocycle 32 as a single
product.
RCM products 35a,b
It is important to note that in the macrocycles 31a and 31b
the exo-acetate rings did not give sharp NMR signals even
at low temperatures (210 K), which created difficulties in fully
interpreting all data. The coupling constants listed in Table 2 of
the remaining sharp signals of the exo-acetate cyclohexane ring
Diene 27 is the C-1 epimer of diene 24 with D-configured cyclohex-
ane rings and two exo-acetate groups. As for 28, the cyclohexane
rings of diene 27 adopt an axial-rich C1 conformation that is
stabilised by an intramolecular H-bond from NH to O-5 (see NH
chemical shift in Table 3). Macrocycle 35 has two exo-acetate
4
Table 1 Conformations of cyclohexanes 12a and 12b as well as dimers 16 and 17 obtained from NMR analysis
3J/Hz a
d/ppm b
HNTfa
Compound
Solvent
1-H
2ax-H
2eq-H
3-H
4-H
5-H
6ax-H
6eq-H
12a
12a
12b
16
CDCl3
4.1, 3.6, 3.6, 3.5
m
11.5, 11.2, 3.9, 3.9 11.5, 3.5
10.9, 10.9, 4.6, 4.6
10.9, 10.8, 4.2, 4.1 10.8, 2.6
4.1, 2.9
m
3.6, 3.0, 2.9
m
4.5, 4.0
m
3.0, 3.0, 2.9
10.7, 4.5, 3.2
m
10.5, 4.8, 3.0
5.1, 2.8, 2.6
9.1, 3.0
3.5, 3.2
m
3.8, 3.0
8.3, 2.8
11.2, 9.1, 3.9
3.7, 3.7, 3.5
10.5, 9.1, 4.9
3.8, 3.6, 3.6
m
11.2, 3.5
m
11.2, 10.5 4.9, 3.9
3.9, 3.6
m
8.11
n.d.
n.d.
n.d.
n.d.
CD3OD
CD3OD
CD3OD
CD3OD
m
m
m
17
5.1, 4.2
10.9, 10.9 6.3, 4.1
2
a Coupling constants between 1-H and NH (around 8.0 Hz) not listed; large geminal J couplings at positions 2 and 6 (around 12 Hz) not listed; m =
multiplet, caused by overlapping of signals, higher order signals or broad signals. b Determined at 295 K; n.d. = not detectable because of use of CD3OD
as solvent.
2416 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 2 Conformations of homodimer 23 as well as cyclohexanes 31a,b and 32 obtained by NMR analysis in CDCl3
3J/Hz a
d/ppm b
Compound
Solvent
1-H
2ax-H
2eq-H
3-H
4-H
5-H
6ax-H
6eq-H
HNTfa
23 (endo)
23 (exo)
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CD3OD
CD3OD
5.4, 5.3, 5.2, 5.0
5.0, 4.8, 4.6, 4.5
12.4, 12.4, 4.2, 3.9
4.6, 4.6, 4.5, 4.3
m
m
m
4.2, 2.4
m
m
m
3.9, 3.8
m
4.1, 4.1, 3.0
m
11.9, 3.6, 2.9
br s
br dd 5.6, 2.7
br s
m
7.8, 3.0
m
8.0, 7.8, 4.0
m
m
br dd 9.0, 5.4
11.2, 10.3, 4.2
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
br d
m
m
7.55
6.58
6.50
7.19
8.24
6.90
n.d.
n.d.
31a (endo)
31a (exo)
31b (endo)
31b (exo)
32 (endo)
32 (exo)
2.9, 2.7
7.3, 2.5
10.3, 2.7
6.8, 3.0
br d 2.7
m
4.4, 2.4
5.6, 2.6
m
m
m
m
m
m
m
11.2, 11.2, 3.9, 3.9
9.5, 9.5, 3.9, 3.9
10.0, 3.5, 2.7
m
2
a Coupling constants between 1-H and NH (around 8.0 Hz) not listed; large geminal J couplings at positions 2 and 6 (around 12 Hz) not listed; m =
multiplet, caused by overlapping of signals, higher order signals or broad signals. b Determined at 295 K; n.d. = not detectable because of use of CD3OD
as solvent.
positions in the cyclohexane ring, the conformation is rather more
defined than in 32, where only two substituents can adopt the same
orientation. Nevertheless, this configuration still enables the ring
system to be very flexible, as can be judged from the behaviour of
35 in CDCl3 (Fig. 5).
These sets of signals belong to two interconverting confor-
mations (see equilibrium in Fig. 6). ROESY spectra recorded
at 220 K in CDCl3 clearly reveal exchange cross-peaks between
the two sets of signals. One set of NMR signals belongs to an
unsymmetrical conformation (94% for 35a) whereas the other
set belongs to a C2-symmetrical conformation (6% of 35b)
(Fig. 6).
In macrocycle 35a the two cyclohexane rings adopt different
conformations: an equatorial-rich 1C4 conformation (denoted B)
and an axial-rich 4C1 conformation (denoted A).25 Consequently,
the 1H-NMR chemical shifts of the two cyclohexanes differ. The
Fig. 4 X-Ray analysis of macrocycle 31b.24
coupling constants at 3-H in conformation A (singlet, line width =
10 Hz) and at 3-H in the conformation B (dd, J = 11.6, 5.5 Hz)
indicate correspondingly an equatorial-rich 1C4 and an axial-rich
groups and derives from diene 27. In CD3OD, only one con-
4C1 cyclohexane ring conformation.
formation is present with a single set of sharp signals for both
The axial-rich conformer is presumably stabilised by an in-
tramolecular hydrogen bond between NH and O-5. Additional
support for this conformation was collected from the ROESY
spectra recorded at 220 K. Characteristic ROE connectivi-
ties found for the two chair conformations are depicted in
Fig. 7.
cyclohexane rings. This indicates that the molecule has a C2-
symmetry in methanol. The analysis of the coupling constants
indicates that both cyclohexane rings adopt an equatorial-rich
1C4 conformation.
Deprotected aminoalcohol 36 adopts also a 1C4 conformation,
as does 35 in methanol. With three substituents in equatorial
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2417
©

View Article Online
Table 3 Preferred conformations of epimeric diene 27 as well as corresponding macrocycles 34 and 35
3J/Hz a
d/ppm b
Compound Solvent
1-H
2ax-H
2eq-H
3-H
4-H
5-H
6ax-H
6eq-H
HNTfa
27
CDCl3
CDCl3
CDCl3
CDCl3
3.9, 3.9, 3.9, 3.9
12.5, 3.9 br d 3.9 12.5, 4.0, 3.0 4.0, 3.8 4.1, 3.8, 3.7
br dd 4.1, 3.9
m
m
m
12.0, 11.4
11.8, 11.6
br d 3.9 7.83
35a (A)
35a (B)
35b (C)
35
m
m
m
m
m
m
m
m
m
m
3.9, 3.5
11.6, 5.5
br s
br s
3.5, 3.2, 3.1
3.5, 3.1, 2.6
m
m
m
m
m
m
m
m
m
8.28c
7.00–7.08c
8.02c
CD3OD 12.1, 12.0, 4.1, 3.9
CD3OD 11.6, 11.5, 4.0, 3.9 11.5, 2.6 6.5, 3.5
9.4, 3.2 11.4, 9.4, 4.4
9.2, 3.0
4.5, 4.4
7.1, 4.0
n.d.
n.d.
36
m
2
a Coupling constants between 1-H and NH (around 8.0 Hz) not listed; large geminal J couplings at positions 2 and 6 (around 12 Hz) not listed; m =
multiplet, caused by overlapping of signals, higher order signals or broad signals. b Determined at 295 K; n.d. = not detectable because of use of CD3OD
as solvent. c Determined at 220 K.
For the minor conformer only one set of signals could be
detected. This is a clear indication that 35b has a C2-symmetry with
two cyclohexane rings populating the same chair conformation.
The 1H NMR chemical shifts of the cyclohexane protons in
It should be noted that in CD3OD only one conformation can be
detected in which both cyclohexane rings adopt the equatorial-rich
1C4 chair conformations.
1
35b resemble the chemical shifts of the equatorial-rich C4 (B)
Conclusions
conformation in 35a. This indicates that in 35b both cyclohexane
rings presumably adopt an equatorial-rich 1C4 conformation.
In conclusion, we were able to synthesise several cyclohexane-
based macrocyclic aminoalcohols 30, 32, 34, 36 and 38 with poten-
tial nucleic acid binding properties.8,9,28 Each of these macrocycles
contains two tetrasubstituted cyclohexane rings linked though two
butanediol units. Unlike sugar-derived pyrans, these individual
tetrasubstituted cyclohexane moieties show high conformational
flexibility. On being incorporated into a macrocyclic system as
in 29, 31, 33, 35 and 37, diverse static as well as dynamic
conformational phenomena occurred. The relative configuration
and the nature of substituents around the cyclohexane moieties
influence not only the overall conformation of the macrocycle, but
1
Line-shape analysis of the H NMR spectra of 35 provided
cyclohexane chair flipping rates (k).26 Computer-generated line
shapes for the 3-H resonances at temperatures between 300 K and
220 K and the determined rate constants are listed alongside the
experimental spectra in Fig. 8.
By plotting ln(k/T) vs. 1/T and fitting these results to the Eyring
equation, one is able to determine the enthalpy of activation
for the cyclohexane ring flip. The activation energy amounts to
38 kJ mol−1, which is about 7 kJ mol−1 lower than the activation
energy for the ring flip of the unsubstituted cyclohexane ring.27
2418 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 8 Experimental vs. calculated line shapes for the 1H NMR reso-
nances of 3-H in 35 at different temperatures.
studies strongly suggest that the intramolecular H-bonding from
an amido H-atom to an alcohol or ether functionality exerts a
decisive contribution to the overall stabilisation of the cyclohexane
chair conformation, thus overcoming increased steric energy when
switching from an equatorial-rich to an axial-rich arrangement of
substituents. Thus, oligosubstituted cyclohexane moieties embed-
ded in macrocycles behave very differently to analogous pyran
rings.8 It needs to be noted that the conformational switch
occurs in non-polar solvents like chloroform and not in a protic
medium like methanol, the latter more closely resembling a
biological system. However, biomacromolecules induce active
conformations upon small ligands. Therefore, the present study
is of importance as it clearly reveals that cyclohexane-based
ligands are conformationally more flexible than the corresponding
pyranyl analogues. The enhanced flexibility of both acyclic as
well as embedded cyclohexanes also has an impact on the
chemical behaviour; for example, protection protocols based on
stannylidene intermediates derived from 16 and 17 proceeded with
poor regioselectivity compared to pyran-based carbohydrates.
RCM always gave the small macrocycles, an observation which
is in contrast to the corresponding sugar-based precursors.8,29
Finally, we are certain that our conformational studies should
be of relevance to related macrocyclic natural products in general
because such conformational flexibility may enable molecules of
this kind to bind to different substrate structures. Investigations
on that subject are underway in our laboratories.
1
Fig. 5 Temperature dependence of the H NMR spectra of macrocycle
35. The signal splitting at 220 K is exemplified for 3-H.
Fig. 6 Exchange cross-peaks in the ROESY spectra of 35 at 220 K; the
chemical exchange between 3-H in 35a (conformation A and B) and 35b
(conformation C) is shown (dotted line).
Experimental
General remarks and starting materials
NMR spectra were recorded on Bruker Avance DPX spectrome-
ters at 200 and 400 MHz and on a Bruker Avance DRX spectrom-
eter at 500 MHz, using tetramethylsilane as the internal standard,
1
if not otherwise mentioned. H multiplicities are described using
the following abbreviations: s = singlet, d = doublet, t = triplet,
q = quartet, m = multiplet, br = broad. Chemical shift values of
13C NMR spectra are reported as values in ppm relative to residual
CDCl3 (77 ppm) or CD3OD (49 ppm) as internal standards.
The multiplicities refer to the resonances in the off-resonance
spectra and were elucidated using the distortionless enhancement
by polarisation transfer (DEPT) spectral editing technique, with
secondary pulses at 90◦ and 135◦. 13C multiplicities are reported
using the following abbreviations: s = singlet (due to quaternary
carbon), d = doublet (methine), q = quartet (methyl), t = triplet
(methylene). The atom labelling is consistent with the numbering
used in the respective schemes. Mass spectra were recorded on a
Fig. 7 Significant NOE connectivities found for 35a which support the
two chair conformations A and B.
also the interaction between the two cyclohexane rings through
the linking chains. In macrocycle 31 we observed the existence
of two distinct stable conformations that do not interconvert and
thus can be separated. On the other hand, compound 34 contains
cyclohexane rings that rapidly switch from one conformation to
the other at room temperature. The energy barrier for the ring flip
was calculated and determined to be significantly lower than in
unsubstituted cyclohexane rings. In all cases, the conformational
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2419
©

View Article Online
LCT spectrometer (Micromass) with a lock-spray unit for ESI or
on a VG Autospec spectrometer (Micromass). Ion mass (m/z)
signals are reported as values in atomic mass units, followed
in parentheses by the peak intensities relative to the base peak
(100%). Optical rotations [a] were collected on a Polarimeter 341
(Perkin Elmer) at a wavelength of 589 nm and are given in 10−1
deg cm2 g−1. All solvents used were of reagent grade and were
further dried. Reactions were monitored by thin layer chromatog-
raphy (TLC) on silica gel 60 F254 (E. Merck, Darmstadt), and
spots were detected either by UV-absorption or by charring with
KMnO4/NaOH in water. Amines were detected using a ninhydrin
solution in propanol. Preparative column chromatography was
performed on silica gel 60 (E. Merck, Darmstadt). Quinic acid
7 was purchased from Fluka in >98% purity. All compounds
starting from quinic acid 7 to yield ketone 8 were prepared
according to the literature.12–14 The preparation of compounds
9, 10a,b, 11a,b, 12a,b, 14, 15 and macrocycles 29, 33, 37 as well as
aminoalcohols 30, 34 and 38 are described in the ESI†. Compound
purities were assessed by NMR analysis.
(d, 5-C), 117.5 (s, CF3, 1JC,F = 286.8 Hz), 158.0 (s, COCF3, 2JC,F
=
36.9 Hz) ppm; HRMS (ESI): m/z for negative ions; calculated:
539.1834 (M − H+); found: 539.1846.
1ꢀ,4ꢀ-O-(5)-Di-3-allyl-(3(R),4(S),5(R)-trihydroxycyclohexane)-1-
(R)-(trifluoroacetamido)-1,4-butane (19), 1ꢀ,4ꢀ-O-(5)-(3-Allyl)-(4ꢀ-
allyl)-di-(3(R),4(S),5(R)-trihydroxycyclohexane)-1(R)-(trifluoroac-
etamido)-1,4-butane (20) and 1ꢀ,4ꢀ-O-(5)-di-4-allyl-(3(R),4(S),5-
(R)-trihydroxy-cyclohexane)-1(R)-(trifluoroacetamido)-1,4-butane
(21). Tetrol 16 (674 mg, 1.248 mmol) and Bu2SnO (776 mg,
3.12 mmol) were dissolved in 20 ml toluene and heated under
reflux for 2 h until the solution has become homogeneous. The
solvent was removed in vacuo and the residue was dissolved in
20 ml anhydrous DMF under an argon atmosphere and cooled to
◦
−15 C. Allyl bromide (1.43 ml, 6.24 mmol) was added and the
solution stirred for 10 min. Afterwards CsF (947 mg, 6.24 mmol)
was added to the mixture and stirring was continued for 1 h.
After removal of the icebath the reaction mixture was stirred
for additional 20 h at RT. After dilution with 70 ml CH2Cl2, the
organic layer was washed three times with water and brine and
dried (Na2SO4). The solvent was removed under reduced pressure
and the crude oil was purified by column chromatography over
silica gel (petroleum ether–ethyl acetate = 3 : 1) to afford a
colourless oil (517 mg, 0.834 mmol; 67% yield) consisting of
three regioisomers which were separated in the following step.
For analytical reasons only a small amount was separated and
characterised at this stage.
1ꢀ,4ꢀ-O-(5)-Di-(3(R),4(S),5(R)-trihydroxycyclohexane)-1(R)-(tri-
fluoroacetamido)-1,4-butane (16). The butanediol-linked dimer
14 (416 mg, 0.670 mmol) was dissolved in 25 ml of MeOH–H2O
(5 : 1) and treated with Amberlyst 15 (230 mg). The solution was
shaken for 18 h. After filtration and evaporation of the solvent,
product 16 (342 mg, 0.633 mmol; 94% yield) was isolated as
a colourless solid. Likewise, the HCl procedure described for
compound 17 works equally well.
19: Rf = 0.40 (CH2Cl2–MeOH = 9 : 1); [a]20 = −76.3 (c = 1,
CHCl3); 1H-NMR (400 MHz, CDCl3, TMS =D0 ppm): d = 1.51–
1.58 (m, 2 H, 6ax-H), 1.65 (m, 4 H, 11-H), 1.79 (ddd, J = 14.0, 4.5,
3.9 Hz, 2 H, 2ax-H), 1.94 (ddd, J = 14.0, 5.4, 4.7 Hz, 2 H, 2eq-H),
2.18 (br d, J = 13.6 Hz, 2 H, 6eq-H), 3.26 (br s, 2 H, OH), 3.45
(ddd, J = 12.0, 5.8, 3.0 Hz, 2 H, 10b-H), 3.61 (ddd, J = 12.0, 8.0,
4.0 Hz, 2 H, 10a-H), 3.58–3.67 (m, 4 H, 5-H, 4-H), 3.93 (ddd, J =
5.4, 3.9, 3.0 Hz, 2 H, 3-H), 4.12 (ddt, J = 12.5, 6.0, 1.4 Hz, 1 H,
Rf = 0.39 (CH2Cl2–MeOH = 9 : 1); [a]2D0 = −3.5 (c = 1, MeOH);
1H NMR (400 MHz, CD3OD, TMS = 0 ppm): d = 1.64 (m, 4 H,
8-H), 1.70–1.88 (m, 8 H, 2-H, 6-H), 3.48 (ddd, J = 11.5, 5.6,
3.4 Hz, 2 H, 7b-H), 3.59 (ddd, J = 11.5, 5.6, 3.4 Hz, 2 H, 7a-H),
3.63 (ddd, J = 3.8, 3.6, 3.6 Hz, 2 H, 5-H), 3.81 (dd, J = 3.8,
3.0 Hz, 2 H, 4-H), 3.88 (ddd, J = 10.5, 4.8, 3.0 Hz, 2 H, 3-H),
4.31 (dddd, J = 10.9, 10.9, 4.6, 4.6 Hz, 2 H, 1-H) ppm; 13C NMR
(100 MHz, CD3OD, CD3OD = 49 ppm): d = 27.9 (t, C-8), 30.8
(t, C-6), 34.3 (t, C-2), 45.2 (d, C-1), 68.7 (d, C-3), 70.1 (t, C-7),
71.0 (d, C-4), 78.5 (d, C-5), 117.5 (s, CF3, 1JC,F = 286.4 Hz), 158.0
(s, COCF3, 2JC,F = 36.8 Hz) ppm; HRMS (ESI): m/z for positive
ions; calculated: 563.1804 (M + Na+); found: 563.1804.
ꢀ
7b-H), 4.14 (ddt, J = 8.8, 5.7, 1.3 Hz, 1 H, 7b -H), 4.25 (ddt, J =
12.5, 5.7, 1.3 Hz, 2 H, 7a-H), 4.30 (ddddd, J = 8.7, 4.7, 4.4, 4.4,
3.9 Hz, 2 H, 1-H), 5.22 (ddd, J = 10.3, 2.7, 1.3 Hz, 2 H, 9b-H), 5.29
(ddd, J = 17.2, 3.1, 1.3 Hz, 2 H, 9a-H), 5.91 (ddt, J = 17.2, 10.3,
5.7 Hz, 2 H, 8-H), 7.70 (br s, 2 H, NH) ppm; 13C-NMR (100 MHz,
CDCl3, CDCl3 = 77 ppm): d = 26.7 (t, C-11), 31.4 (t, C-2), 32.6
(broad signal, t, C-6), 45.0 (d, C-1), 69.3 (t, C-10), 72.1 (t, C-7),
73.3 (broad signal, d, C-4), 75.4 (d, C-5), 77.2 (d, C-3), 115.8 (s,
1ꢀ,4ꢀ-O-(5)-Di-(3(R),4(S),5(R)-trihydroxycyclohexane)-1(S)-(tri-
fluoroacetamido)-1,4-butane (17). The butanediol-linked dimer
15 (959 mg, 1.545 mmol) was dissolved in 33 ml MeOH–H2O
(10 : 1) and 0.5 ml concentrated HCl(aq). This solution was stirred
for 1 h at RT. The solvents were evaporated in vacuum to yield
the product 17 (838 mg, 1.551 mmol) in quantitative yield as
a colourless solid. The Amberlyst 15 procedure described for
compound 16 works equally well. The product is not very soluble
in methanol or chloroform.
1
CF3, JC,F = 287.9 Hz), 117.7 (t, C-9), 134.1 (d, C-8), 156.0 (s,
2
COCF3, JC,F = 36.9 Hz) ppm; HRMS (ESI): m/z for positive
ions; calculated: 643.2430 (M + Na+); found: 643.2430.
20: Rf = 0.67 (CH2Cl2–MeOH = 9 : 1); [a]20 = −77.3 (c = 1,
CHCl3); 1H-NMR (400 MHz, CDCl3, TMS =D0 ppm): dH = 1.54
(m, 2 H, 2ꢀ-H), 1.63 (m, 4 H, 11-H), 1.78 (m, 2 H, 6ꢀ-H), 1.96 (m,
2 H, 6-H), 2.19 (m, 2 H, 2-H), 2.83 (s, 1 H, OH), 3.06 (s, 1 H,
OH), 3.31 (dd, J = 8.4, 2.9 Hz, 1 H, 4-H), 3.40–3.47 (m, 1 H,
10ꢀ-H), 3.49–3.55 (m, 1 H, 10ꢀ-H), 3.55–3.67 (m, 4 H, 4ꢀ-H, 5-H,
10-H), 3.92 (ddd, J = 5.6, 2.7, 2.7 Hz, 1 H, 3ꢀ-H), 4.12 (ddt, J =
◦
1
Rf = 0.15 (CH2Cl2–MeOH = 9 : 1); m.p. = 155–160 C; H
NMR (400 MHz, CDCl3–CD3OD, TMS = 0 ppm): d = 1.35 (ddd,
J = 12.2, 10.9, 10.9 Hz, 2 H, 6ax-H), 1.62 (ddd, J = 13.3, 10.8,
2.6 Hz, 2 H, 2ax-H), 1.66 (m, 4 H, 8-H), 1.97 (ddd, J = 13.3, 5.1,
4.2 Hz, 2 H, 2eq-H), 2.22 (ddd, J = 12.2, 6.3, 4.1 Hz, 2 H, 6eq-H),
3.46 (dd, J = 8.3, 2.8 Hz, 2 H, 4-H), 3.55–3.67 (m, 6 H, 5-H, 7-H),
4.01 (ddd, J = 5.1, 2.8, 2.6 Hz, 2 H, 3-H), 4.22 (dddd, J = 10.9,
10.8, 4.2, 4.1 Hz, 2 H, 1-H) ppm; 13C NMR (100 MHz, CDCl3–
CD3OD, CD3OD = 49 ppm): d = 27.8 (t, 8-C), 35.1 (t, 6-C), 36.7
(t, 2-C), 44.9 (d, 1-C), 69.4 (d, 3-C), 70.9 (t, 7-C), 75.2 (d, 4-C), 78.3
ꢀ
12.7, 6.0, 1.3 Hz, 1 H, 7a -H), 4.18 (ddt, J = 12.7, 6.9, 1.3 Hz,
ꢀ
1 H, 7b -H), 4.21–4.26 (m, 4 H, 3-H, 5ꢀ-H, 7a-H, 7b-H), 4.26–4.34
(m, 2 H, 1-H, 1ꢀ-H), 5.21 (ddd, J = 10.2, 6.3, 2.3 Hz, 2 H, 9ꢀ-H),
5.30 (ddd, J = 17.2, 6.2, 3.2 Hz, 2 H, 9-H), 5.89 (ddd, J = 10.2,
5.8, 4.1 Hz, 1 H, 8-H), 5.95 (ddd, J = 10.2, 6.9, 4.1 Hz, 1 H,
8ꢀ-H), 7.69 (br s, 1 H, NHꢀ), 7.95 (br s, 1 H, NH) ppm; 13C-NMR
2420 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
(100 MHz, CDCl3, CDCl3 = 77 ppm): d = 26.7 (t, C-11ꢀ), 26.8 (t,
C-11), 31.4 (t, C-6ꢀ), 32.6 (t, C-6), 32.6 (t, C-2ꢀ), 33.6 (t, C-2), 45.0
(d, C-1), 45.0 (C-1ꢀ), 69.1 (d, C-5), 69.2 (t, C-10ꢀ), 70.0 (t, C-10),
71.9 (t, C-7), 74.2 (d, C-3ꢀ), 75,3 (broad signal, d, C-4ꢀ), 76,8 (d,
C-3), 81.2 (broad signal, d, C-4), 115.8 (s, CF3, 1JC,F = 287.9 Hz),
117.2 (t, C-9ꢀ),117.6 (t, C-9), 134.0 (d, C-8ꢀ), 134.6 (d, C-8), 156.2
(s, COCF3, 2JC,F = 36.7 Hz) ppm; HRMS (ESI): m/z for positive
ions; calculated: 643.2430 (M + Na+); found: 643.2430.
C-6), 44.7 (d, C-1), 69.8 (t, C-10), 71.9 (t, C-7), 72.8 (d, C-5), 74.6
(broad signal, d, C-4), 75.3 (broad signal, d, C-3), 115.8 (s, CF3,
1JC,F = 287.5 Hz), 118.0 (t, C-9), 133.8 (d, C-8), 156.2 (s, COCF3,
2JC,F = 36.7 Hz), 170.2 (s, COCH3) ppm; HRMS (ESI): m/z for
positive ions; calculated: 727.2641 (M + Na+); found: 727.2626.
23: Rf = 0.44 (petroleum ether–ethyl acetate = 1 : 1); [a]D20
=
−33.8 (c = 1, CHCl3); 1H-NMR (400 MHz, CDCl3, TMS =
0 ppm): d = 1.61 (m, 4 H, 11-H), 1.70–2.05 (m, 8 H, 6-H, 2-
21: Rf = 0.70 (CH2Cl2–MeOH = 9 : 1); [a]2D0 = −84.2 (c = 1,
H), 2.08 (s, 3 H, COCH3 ), 2.12 (s, 3 H, COCH3), 3.43–3.61 (m,
ꢀ
1
CHCl3); H-NMR (400 MHz, CDCl3, TMS = 0 ppm): d = 1.52
4 H, 10-H), 3.64–3.68 (m, 2 H, 4ꢀ-H, 5ꢀ-H), 3.72 (ddd, J = 8.0,
8.0, 4.0 Hz, 1 H, 5-H), 3.94 (ddd, J= 4.1, 4.1, 3.9 Hz, 1 H, 3-H),
4.00–4.10 (m, 2 H, 7-H), 4.11–4.17 (m, 2 H, 7-H), 4.21 (ddddd,
J = 9.3, 5.0, 4.8, 4.6, 4.5 Hz, 1 H, 1ꢀ-H), 4.30 (ddddd, J= 7.2, 5.4,
5.3, 5.2, 5.0 Hz, 1 H, 1-H), 4.97 (dd, J= 7.5, 2.1 Hz, 1 H, 4-H),
5.17 (m, 1 H, 3ꢀ-H), 5.18 (ddd, J= 10.4, 3.0, 1.3 Hz, 1 H, 9-H),
5.22 (ddd, J = 10.4, 2.8, 1.3 Hz, 1 H, 9-H), 5.24–5.31 (m, 2 H,
9-H), 5.87 (dddd, J = 16.0, 8.9, 5.7, 1.3 Hz, 1 H, 8ꢀH), 5.89 (dddd,
J = 16.1, 11.5, 5.7, 1.1 Hz, 1 H, 8-H), 6.58 (br d, J = 7.3 Hz, 1 H,
NHꢀ), 7.55 (br s, 1 H, NH) ppm; 13C-NMR (100 MHz, CDCl3,
CDCl3 = 77 ppm): d = 21.0 (q, COCH3), 26.6 (t, C-11ꢀ), 26.7 (t,
C-11), 31.2 and 31.9 (t, C-2, C-6), 43.8 (d, C-1ꢀ), 44.6 (d, C-1),
69.3 (t, C-10ꢀ), 69.7 (t, C-10), 70.0 (d, C-3ꢀ), 72.0 (broad signal, t,
C-7ꢀ), 72.1 (t, C-7), 72.9 (broad signal, d, C-5), 72.9 (broad signal,
(ddd, J = 13.6, 10.4, 3.7 Hz, 2 H, 6ax-H), 1.61 (m, 4 H, 11-H),
1.75 (ddd, J = 14.4, 3.4, 2.9 Hz, 2 H, 2ax-H), 1.99 (m, 2 H, 2eq-H),
2.21 (br d, J = 12.9 Hz, 2 H, 6eq-H), 3.03 (s, 2 H, OH), 3.30 (dd,
J = 8.2, 3.1 Hz, 2 H, 4-H), 3.51 (ddd, J = 6.0, 5.8, 2.7 Hz, 2 H,
10ꢀ-H), 3.56 (ddd, J = 6.0, 5.8, 2.7 Hz, 2 H, 10-H), 3.62 (ddd,
J = 10.4, 8.2, 4.3 Hz, 2 H, 5-H), 4.16 (ddt, J = 12.7, 5.6, 1.4 Hz,
2 H, 7b-H), 4.24–4.19 (m, 2 H, 3-H), 4.23 (ddt, J = 12.7, 5.6,
1.4 Hz, 2 H, 7a-H), 4.28 (ddddd, J = 7.9, 3.9, 3.9, 3.7, 3.4 Hz,
2 H, 1-H), 5.20 (ddd, J = 10.5, 3.0, 1.4 Hz, 2 H, 9b-H), 5.29 (ddd,
J = 17.2, 3.0, 1.4 Hz, 2 H, 9a-H), 5.92 (dddd, J = 17.2, 10.5, 5.6,
5.6 Hz, 2 H, 8-H), 7.93 (br s, 2 H, NH) ppm; 13C-NMR (100 MHz,
CDCl3, CDCl3 = 77 ppm): d = 26.8 (t, C-11), 32.6 (t, C-6), 33.6
(t, C-2), 45.0 (d, C-1), 69.1 (d, C-5), 70.0 (t, C-10), 71.9 (t, C-7),
74.2 (d, C-3), 81.2 (d, C-4), 115.8 (s, CF3, 1JC,F = 287.5 Hz), 117.2
d, C-4), 74.8 (d, C-5ꢀ), 75.1 (broad signal, d, C-4ꢀ), 75.2 (broad
2
ꢀ
(t, C-9), 134.6 (d, C-8), 156.2 (s, COCF3, JC,F = 36.5 Hz) ppm;
signal, d, C-3), 115.8 (s, CF3 , J = 288.2 Hz), 115.8 (s, CF3, 1JC,F
=
HRMS (ESI): m/z for positive ions; calculated: 643.2430 (M +
Na+); found: 643.2430.
288.2 Hz), 116.9 (t, C-9ꢀ), 117.9 (t, C-9), 133.8 (d, C-8), 134.7 (d,
C-8ꢀ), 156.2 (s, COCF3, 2JC,F = 36.7 Hz), 169.8 (s, COCH3 ), 170.2
ꢀ
Compounds 25 and 26 were not isolated and were immediately
employed in the next step for the preparation of dimers 27 and 28.
(s, COCH3) ppm; HRMS (ESI): m/z for positive ions; calculated:
727.2641 (M + Na+); found: 727.2626.
24: Rf = 0.54 (petroleum ether–ethyl acetate = 1 : 1); [a]2D0 = −3.1
(c = 1, CHCl3); 1H-NMR (400 MHz, CDCl3, TMS = 0 ppm): dH =
1.64 (m, 4 H, 11-H), 1.77 (ddd, J = 13.1, 10.2, 2.4 Hz, 2 H, 6ax-
H), 1.82 (ddd, J = 11.7, 10.7, 10.5 Hz, 2 H, 2ax-H), 1.90–2.08 (m,
4 H, 6eq-H, 2eq-H), 2.10 (s, J = 1.0 Hz, 3 H, COCH3), 2.11 (s, J =
1.0 Hz, 3 H, COCH3), 3.49 (ddd, J = 8.6, 5.8, 3.0 Hz, 2 H, 10b-H),
3.54 (ddd, J = 8.6, 5.6, 3.2 Hz, 2 H, 10a-H), 3.65–3.70 (m, 4 H,
4-H, 5-H), 4.01 (ddt, J = 13.0, 5.6, 1.4 Hz, 2 H, 7b-H), 4.14 (ddt,
J = 13.0, 5.6, 1.5 Hz, 2 H, 7a-H), 4.21 (ddddd, J = 10.5, 10.2, 8.6,
4.7, 4.7 Hz, 2 H, 1-H), 5.15–5.22 (m, 4 H, 9b-H, 3-H), 5.29 (ddd,
J = 17.2, 3.3, 1.7 Hz, 2 H, 9a-H), 5.89 (ddt, J = 17.2, 10.4, 5.6 Hz,
2 H, 8-H), 6.48 (br d, J = 8.6 Hz, 2 H, NH) ppm; 13C-NMR
(100 MHz, CDCl3, CDCl3 = 77 ppm): dC = 21.1 (q, COCH3),
26.6 (t, C-11), 31.2 (t, C-6, C-2), 43.8 (d, C-1), 69.3 (t, C-10), 70.0
(d, C-3), 72.1 (t, C-7), 74.8 (d, C-4), 75.3 (d, C-5), 115.7 (s, CF3,
1JC,F = 288.1 Hz), 117.0 (t, C-9), 134.8 (d, C-8), 156.2 (s, COCF3,
2JC,F = 36.9 Hz), 169.8 (s, COCH3) ppm; HRMS (ESI): m/z for
positive ions; calculated: 727.2641 (M + Na+); found: 727.2647.
1ꢀ,4ꢀ-O-(5)-Di-3-allyl-di-4-aceto-(3(R),4(S),5(R)-trihydroxycycl-
ohexane)-1(R)-(trifluoroacetamido)-1,4-butane (22), 1ꢀ,4ꢀ-O-(5)-
(3-Allyl)-(4-aceto)-(3ꢀ-aceto)-(4ꢀ-allyl)-di-(3(R),4(S),5(R)-trihydro-
xycyclohexane)-1(R)-(trifluoroacetamido)-1,4-butane (23) and
1ꢀ,4ꢀ-O-(5)-Di-4-allyl-di-3-aceto-[3(R),4(S),5(R)-trihydroxycycloh-
exane)-1(R)-trifluoroacetamido]-1,4-butane (24). To the mixture
of regioisomers 19, 20 and 21 (451 mg, 0.727 mmol) in 20 ml
CH2Cl2 were added Et3N (0.30 ml, 2.181 mmol), Ac2O (0.27 ml,
2.908 mmol) and DMAP (40 mg, 0.327 mmol). This mixture
was stirred for 2 h at RT. After removal of the solvent under
reduced pressure the crude material was purified by flash column
chromatography over silica gel (petroleum ether–ethyl acetate =
3 : 1) to yield the products 22 (63 mg, 0.089 mmol; 12%), 23
(253 mg, 0.359 mmol; 49%) and 24 (124 mg, 0.176 mmol; 24%)
all as colourless oils.
22: Rf = 0.38 (petroleum ether–ethyl acetate = 1 : 1); [a]D20
=
−55.3 (c = 1, CHCl3); 1H-NMR (400 MHz, CDCl3, TMS =
0 ppm): d = 1.55 (m, 4 H, 11-H), 1.65–1.74 (m, 2 H, 6b-H), 1.85–
1.91 (m, 4 H, 2-H), 2.04–2.09 (m, 2 H, 6a-H), 2.10 (s, 3 H, COCH3),
2.11 (s, 3 H, COCH3), 3.42 (ddd, J = 11.7, 5.7, 3.1 Hz, 2 H, 10b-H),
3.54 (ddd, J = 11.7, 5.7, 3.2 Hz, 2 H, 10a-H), 3.69 (ddd, J = 8.3,
8.1, 4.0 Hz, 2 H, 5-H), 3.93 (ddd, J = 4.4, 4.3, 3.2 Hz, 2 H, 3-H),
4.01 (ddt, J = 12.5, 5.8, 1.2 Hz, 2 H, 7b-H), 4.12 (ddt, J = 12.4,
5.8, 1.4 Hz, 2 H, 7a-H), 4.24 (ddddd, J = 8.3, 4.8, 4.8, 4.6, 4.5 Hz,
2 H, 1-H), 4.93 (dd, J = 8.1, 2.4 Hz, 2 H, 4-H), 5.21 (ddd, J =
10.4, 2.7, 1.2 Hz, 2 H, 9b-H), 5.26 (ddd, J = 17.2, 3.0, 1.6 Hz, 2 H,
9a-H), 5.86 (ddt, J = 17.2, 10.4, 5.8 Hz, 2 H, 8-H), 7.54 (br s, 2 H,
NH) ppm; 13C-NMR (100 MHz, CDCl3, CDCl3 = 77 ppm): d =
21.0 (q, COCH3), 26.7 (t, C-11), 31.9 (t, C-2), 33.5 (broad signal, t,
1ꢀ,4ꢀ-O-(5)-Di-4-allyl-di-3-aceto-(3(R),4(S),5(R)-trihydroxycyc-
lohexane)-1(S)-(trifluoroacetamido)-1,4-butane (27) and 1ꢀ,4ꢀ-O-
(5)-(3-Allyl)-(4-aceto)-(3ꢀ -aceto)-(4ꢀ-allyl)-di-(3(R),4(S),5(R)-tri-
hydroxycyclohexane)-1(R)-(trifluoroacetamido)-1,4-butane (28).
In analogy to the procedure described for the synthesis of
compounds 19, 20 and 21, tetrol 17 (490 mg, 0.907 mmol) was
dissolved in 25 ml toluene and Bu2SnO (564 mg, 2.268 mmol) was
added. For the second step CsF (689 mg, 4.535 mmol) and allyl
bromide (0.38 ml, 4.535 mmol) in 25 ml DMF were employed.
The resulting products were filtered through a pad of silica gel
(ethyl acetate–petroleum ether = 3 : 1) to yield the mixture
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2421
©

View Article Online
of two regioisomers 25 and 26 (661 mg, 1.066 mmol), which
were employed in the next step without further separation or
purification. According to the procedure for compounds 22–24,
the residue was dissolved in 20 ml CH2Cl2 and treated with
Et3N (0.34 ml, 3.192 mmol), Ac2O (0.40 ml, 4.256 mmol) and
DMAP (12.5 mg, 0.105 mmol). The crude material was purified
by column chromatography over silica gel (petroleum ether–ethyl
acetate = 2.5 : 1) to yield bisallyl ethers 27 (411 mg, 0.584 mmol;
64% for two steps) and 28 (101 mg, 0.143 mmol, 16%).
Macrocycles 31a and 31b. According to the procedure de-
scribed for the preparation of macrocycle 29, bisallylether 23
(328 mg, 466 mmol) and Grubbs catalyst 13 (38 mg, 0.047 mmol)
were dissolved in 50 ml degassed benzene under an argon
atmosphere and heated at 40 ◦C for 36 h. After 6 h an additional
portion of catalyst 13 (19 mg, 0.023 mmol) was added. Then, all
volatile compounds were removed under reduced pressure and
the residue was filtered through silica gel (petroleum ether–ethyl
acetate = 2 : 1) to yield a colourless oil (236 mg, 0.349 mmol; 75%
yield). The product was immediately employed in the next step.
A portion of this crude material (180 mg, 0.266 mmol) was
dissolved in ethyl acetate–CH2Cl2–MeOH (16 : 8 : 1; 12.5 ml),
Pt2O (24.2 mg, 0.106 mmol) was added and the reaction mixture
was kept under hydrogen atmosphere for 14 h. The resulting
mixture of products 31a and 31b were separated by flash column
chromatography over silica gel (petroleum ether–ethyl acetate =
2 : 1) to yield macrocycle 31a (77.5 mg, 0.114 mmol; 42%) as a
colourless oil. Besides a mixed fraction (27.4 mg, 0.040 mmol;
15%), macrocycle 31b was collected in a minor fraction (15.9 mg,
0.023 mmol; 9%).
27: Rf = 0.44 (petroleum ether–ethyl acetate = 1 : 1); [a]2D0 = −36
(c = 1, CHCl3); 1H-NMR (400 MHz, CDCl3, TMS = 0 ppm): d =
1.65–1.71 (m, 4 H, 11-H), 1.74 (br dd, J = 4.1, 3.9 Hz, 2 H, 6b-H),
1.90 (ddd, J = 12.5, 4.0, 3.9 Hz, 2 H, 2b-H), 2.07 (s, 6 H, COCH3),
2.08 (br dd, J = 12.1, 3.9 Hz, 2 H, 6a-H), 2.11 (br dd, J = 7.3,
3.9 Hz, 2 H, 2a-H), 3.52 (ddd, J = 11.1, 5.5, 3.1 Hz, 2 H, 10ꢀ-H),
3.69 (ddd, J = 11.8, 5.6, 2.9 Hz, 2 H, 10-H), 3.79 (ddd, J = 4.1, 3.8,
3.7 Hz, 2 H, 5-H), 3.89 (dd, J = 4.0, 3.8 Hz, 2 H, 4-H), 4.04 (ddt,
J = 12.9, 5.6, 1.4 Hz, 2 H, 7ꢀ-H), 4.14 (ddt, J = 12.9, 5.6, 1.4 Hz,
2 H, 7-H), 4.34 (ddddd, J = 7.6, 3.9, 3.9, 3.9, 3.9 Hz, 2 H, 1-H),
5.15 (ddd, J = 12.5, 4.0, 3.0 Hz, 2 H, 3-H), 5.19 (ddd, J = 10.4,
2.9, 1.3 Hz, 2 H, 9ꢀ-H), 5.27 (ddd, J = 17.2, 3.2, 1.6 Hz, 2 H, 9-H),
5.87 (ddt, J = 17.2, 10.4, 5.6 Hz, 2 H, 8-H), 7.83 (br s, 2 H, NH)
ppm; 13C-NMR (100 MHz, CDCl3, CDCl3 = 77 ppm): d = 21.1
(q, COCH3), 26.7 (t, C-11), 29.0 (broad signal, t, C-6), 29.9 (broad
signal, t, C-2), 45.3 (d, C-1), 67.9 (d, C-3), 69.9 (t, C-10), 72.4
(t, C-7), 73.0 (broad signal, d, C-4), 77.6 (d, C-5), 115.9 (s, CF3,
1JC,F = 288.1 Hz), 117.2 (t, C-9), 134.6 (d, C-8), 156.1 (s, COCF3,
2JC,F = 36.7 Hz), 170.3 (s, COCH3) ppm. HRMS (ESI): m/z for
positive ions; calculated: 727.2641 (M + Na+); found: 727.2650.
28: Not every signal could be unequivocally assigned to each
of the two cyclohexane rings. Rf = 0.39 (petroleum ether–ethyl
acetate = 1 : 1); [a]2D0 = −46.2 (c = 1, CHCl3); 1H-NMR (400 MHz,
CDCl3, TMS = 0 ppm): d = 1.60–1.79 (m, 7 H, 11-H, CH2), 1.86–
31a: m.p.: 230–232 ◦C; Rf = 0.60 (CH2Cl2–MeOH = 9 : 1);
[a]2D0 = −42.4 (c = 1, CHCl3); 1H-NMR (500 MHz, CDCl3, TMS =
0 ppm): d = 1.43–1.54 (m, 4 H, 6b-H, CH2), 1.55–1.76 (m, 12 H,
ꢀ
6b -H, CH2), 1.79 (ddd, J = 14.0, 4.2, 2.4 Hz, 1 H, 2b-H), 1.87–1.97
ꢀ
ꢀ
(m, 2 H, 2a -H, 2b -H), 2.02 (ddd, J = 14.0, 3.9, 3.8 Hz, 1 H, 2a-H),
ꢀ
2.09–2.12 (m, 1 H, 6a -H), 2.11 (s, 3 H, COCH3), 2.13 (s, 3 H,
COCH3), 2.14–2.21 (m, 1 H, 6a-H), 3.38 (ddd, J = 9.5, 8.6, 3.6 Hz,
1 H, 7b -H), 3.41 (dd, J = 7.3, 2.5 Hz, 1 H, 4ꢀ-H), 3.50 (ddd, J =
ꢀ
8.6. 8.6, 3.7 Hz, 1 H, 10b-H), 3.54–3.64 (m, 5 H, 5-H, CH2), 3.65
(ddm, J = 9.0, 5.4 Hz, 1 H, 5ꢀ-H), 3.68 (ddd, J = 9.5, 4.7, 4.6 Hz,
ꢀ
1 H, 10a-H), 3.78 (ddd, J = 10.1, 4.9, 4.9 Hz, 1 H, 7a -H), 3.82
(ddd, J= 11.9, 3.6, 2.9 Hz, 1 H, 3-H), 4.20 (ddddd, J = 12.4, 12.4,
8.0, 4.2, 3.9 Hz, 1 H, 1-H), 4.35 (ddddd, J = 8.1, 4.6, 4.6, 4.5,
4.3 Hz, 1 H, 1ꢀ-H), 5.29 (dd, J = 2.9, 2.7 Hz, 1 H, 4-H), 5.50 (br s,
3ꢀ-H), 6.50 (br d, J = 8.0 Hz, 1 H, NH), 7.19 (br d, J = 8.1 Hz,
1 H, NHꢀ) ppm; 13C-NMR (125 MHz, CDCl3, CDCl3 = 77 ppm):
d = 20.9 (t, COCH3), 21.1 (t, COCH3), 25.2 (t, C-8ꢀ, C-11ꢀ), 26.2 (t,
C-8), 27.0 (t, C-11), 31.3 (t, C-2), 31.5 (t, C-2ꢀ), 32.4 (t, C-6), 33.3
(t, C-6ꢀ), 43.7 (d, C-1ꢀ), 44.6 (d, C-1), 67.8 (d, C-4), 68.1 (t, C-10ꢀ),
69.4 (d, C-3ꢀ), 69.7 (t, C-7ꢀ), 69.9 (t, C-7), 70.3 (t, C-10), 71.0 (d,
C-3), 73.6 (d, C-5ꢀ), 74.2 (d, C-5), 78.7 (broad signal, d, C-4ꢀ), 115.7
(s, CF3, 1JC,F = 288.0 Hz), 115.8 (s, CF3, 1JC,F = 287.9 Hz), 155.2
ꢀ
ꢀ
2.08 (m, 4 H, 2b -H, CH2), 2.08 (s, 3 H, COCH3 ꢀ), 2.11 (s, 3 H,
COCH3), 2.14 (dddm, J = 8.1, 4.1, 0.9 Hz, 1 H, 2a -H), 3.51 (ddd,
J = 8.8, 5.7, 5.7 Hz, 1 H, 10-H), 3.58 (ddd, J = 9.0, 5.8, 5.8 Hz, 1 H,
10ꢀH), 3.68 (ddd, J = 11.8, 6.5, 2.4 Hz, 2 H, 10-H), 3.75–3.82 (m,
3 H, 3-H, 5ꢀ-H, 5-H), 3.88 (br s, 1 H, 4ꢀ-H), 3.99 (ddt, J = 12.6,
5.8, 1.3 Hz, 1 H, 7b-H), 4.03 (ddt, J = 5.8, 4.5, 1.3 Hz, 1 H, 7a-
ꢀ
H), 4.06 (ddt, J = 5.6, 4.4, 1.4 Hz, 1 H, 7b -H), 4.14 (ddt, J =
ꢀ
12.9, 5.6, 1.4 Hz, 1 H, 7a -H), 4.35 (ddddd, J = 8.0, 4.0, 4.0, 4.0,
4.0 Hz, 2 H, 1ꢀ-H, 1-H), 5.14 (ddd, J = 11.6, 4.1, 2.6 Hz, 1 H,
3ꢀ-H), 5.18 (ddd, J = 5.7, 3.0, 1.3 Hz, 1 H, 9b-H), 5.20 (ddd, J =
5.7, 3.0, 1.3 Hz, 1 H, 9a-H), 5.24 (ddd, J = 7.5, 3.2, 1.6 Hz, 1 H,
2
2
(s, COCF3, JC,F = 36.8 Hz), 156.4 (s, COCF3, JC,F = 37.0 Hz),
169.3 (q, COCH3), 170.0 (q, COCH3) ppm; HRMS (ESI): m/z for
positive ions: calculated: 701.2485 (M + Na+), found: 701.2490.
31b: m.p.: 205–208 ◦C; Rf = 0.67 (CH2Cl2–MeOH = 9 : 1);
[a]2D0 = −63.8 (c = 1, CHCl3); 1H-NMR (500 MHz, CDCl3, TMS =
0 ppm): d = 1.43–1.54 (m, 4 H, 6b-H, CH2), 1.55–1.76 (m, 12 H,
ꢀ
ꢀ
9b -H), 5.29 (ddd, J = 7.4, 3.2, 1.6 Hz, 1 H, 9a -H), 5.33 (br s, 1 H,
4-H), 5.85 (dddd, J = 11.3, 10.4, 5.8, 3.3 Hz, 1 H, 8-H), 5.88
(dddd, J = 11.4, 10.4, 5.8, 3.3 Hz, 1 H, 8-H), 7.68 (br s, 1 H,
NH), 7.77 (br s, 1 H, NHꢀ) ppm; 13C-NMR (100 MHz, CDCl3,
CDCl3 = 77 ppm): d = 21.0 (q, COCH3), 21.1 (q, COCH3), 26.6
(t, C-11), 26.7 (t, C-11), 29.2 (broad signal, t), 29.9 (broad signal,
t), 30.0 (broad signal, t), 31.1 (broad signal, t), 45.3 (d, C-1ꢀ), 45.3
(d, C-1), 67.9 (d, C-3ꢀ), 69.9 (broad signal, d, C-4), 70.0 (t, C-10),
70.3 (t, C-7), 72.4 (t, C-7ꢀ), 74.0 (broad signal, d, C-4ꢀ), 76.3 (broad
ꢀ
6b -H, CH2), 1.79 (ddd, J = 15.0, 4.4, 2.4 Hz, 1 H, 2b-H), 1.87–1.97
ꢀ
ꢀ
(m, 2 H, 2a -H, 2b -H), 2.02 (ddd, J = 15.1, 5.7, 2.6 Hz, 1 H, 2a-H),
ꢀ
2.11 (s, 3 H, COCH3), 2.09–2.12 (m, 1 H, 6a -H), 2.13 (s, 3 H,
COCH3), 2.42 (br d, J = 13.7 Hz, 1 H, 6a-H), 3.37 (dd, J = 6.8,
3.0 Hz, 1 H, 4ꢀ-H), 3.43–3.50 (m, 4 H, CH2), 3.50–3.56 (m, 2 H,
CH2), 3.57–3.63 (m, 2 H, CH2), 3.63–3.70 (m, 4 H, CH2, 5ꢀ-H), 3.83
(dd, J = 8.3, 4.0 Hz, 1 H, CH2), 3.89 (ddd, J = 11.2, 10.5, 4.2 Hz,
1 H, 5-H), 4.00 (br dd, J = 5.4, 2.7 Hz, 2 H, 3-H), 4.29–4.34 (m,
1 H, 1ꢀ-H), 4.35–4.40 (m, 1 H, 1-H), 4.77 (dd, J = 10.1, 2.7 Hz, 1 H,
4-H), 5.45 (br s, 3ꢀ-H), 6.90 (br s, 1 H, NHꢀ), 8.24 (br d, J = 7.8 Hz,
1 H, NH) ppm; 13C-NMR (125 MHz, CDCl3, CDCl3 = 77 ppm):
signal, d, C-3), 77.2 (d, C-5), 77.6 (d, C-5), 115.9 (s, CF3, 1JC,F
=
287.5 Hz), 117.2 (t, C-9ꢀ), 117.4 (t, C-9), 134.4 (d, C-8), 134.6 (d,
C-8), 156.0 (s, COCF3, 2JC,F = 36.5 Hz), 156.0 (s, COCF3, 2JC,F
=
36.6 Hz), 170.2 (s, COCH3), 170.3 (s, COCH3) ppm. HRMS (ESI):
m/z for positive ions; calculated: 727.2641 (M + Na+); found:
727.2653.
2422 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
d= 20.9 (t, COCH3), 21.1 (t, COCH3), 25.2 (t), 26.2 (t), 27.4 (t),
32.2 (t, C-2, C-6ꢀ), 34.4 (t, C-6), 45.4 (d, C-1ꢀ), 45.4 (d, C-1), 68.4 (t,
C-10, C-10ꢀ), 68.9 (t), 69.5 (d, C-3ꢀ), 70.0 (d, C-5), 70.7 (t), 72.9 (t,
C-7, C-7ꢀ), 73.1 (d, C-5ꢀ), 77.1 (d, C-4), 77.8 (d, C-3), 78.5 (broad
(s, 3 H, COCH3), 2.53–2.61 (m, 1 H, 6B, eq-H), 3.02–3.10 (m, 1 H,
ꢀ
ꢀ
ꢀ
7B -H), 3.12–3.19 (m, 2 H, 4B-H, 10B -H), 3.20–3.31 (m, 1 H, 7A -
ꢀ
ꢀ
H), 3.33–3.53 (m, 1 H + 2 H, 5C-H, 10A -H, 10C -H), 3.57–3.69 (m,
3 H, 5B-H, 10A-H, 7A-H), 3.70–3.77 (m, 2 H, 4A-H, 5A-H), 3.78–
3.84 (m, 1 H, 10B-H), 3.85–3.92 (m, 1 H, 7B-H), 4.13–4.24 (m,
1 H, 1B-H), 4.24–4.30 (m, 1 H, 1A-H), 4.99 (dd, J = 11.6, 5.5 Hz,
3A-H), 5.44 (br t, 1 H, 3c-H), 5.53–5.59 (br s, 1 H, 3B-H), 7.00–7.08
(br s, 1 H, NHB), 8.28 (br d, J = 7.4 Hz, 1 H, NHA) ppm; at room
temperature with CD3OD as solvent the signals also sharpened:
1H NMR (400 MHz, CD3OD, TMS = 0 ppm): d = 1.37 (ddd, J =
12.1, 12.0, 11.4 Hz, 2 H, 6b-H), 1.50–1.77 (m, 10 H, 2b-H, 8-H,
11-H), 2.00 (ddd, J = 13.8, 3.9, 3.5 Hz, 2 H, 2a-H), 2.09 (s, 6 H,
COCH3), 2.27 (ddd, J = 12.1, 4.5, 4.4 Hz, 2 H, 6a-H), 3.22 (dd, J =
9.4, 3.2 Hz, 2 H, 4-H), 3.44–3.51 (m, 2 H, CH2), 3.52–3.61 (m, 2 H,
CH2), 3.53 (ddd, J = 11.4, 9.4, 4.4 Hz, 2 H, 5-H), 3.65–3.73 (m,
4 H, CH2), 4.10 (dddd, J = 12.1, 12.0, 4.1, 3.9 Hz, 2 H, 1-H), 5.49
(ddd, J = 3.5, 3.2, 3.1 Hz, 2 H, 3-H) ppm; at room temperature
with CDCl3 as solvent the signals showed broadened signals; C-4
not detectable: 13C NMR (125 MHz, CDCl3, CDCl3 = 77 ppm):
d= 21.1 (t, COCH3), 26.0 and 26.3 (t, C-8, C-11), 32.2 (t, broad
signal), 44.4 (d, C-1), 67.2 (d, C-3), 69.2 (t), 70.3 (t, broad signal),
75.6 (d, C-5), no signal for C-4, 115.7 (s, CF3, 1JC,F = 288.1 Hz),
156.4 (s, COCF3, 2JC,F = 36.9 Hz), 170.2 (s, COCH3) ppm; room
temperature in CD3OD: 13C NMR (125 MHz, CD3OD, CD3OD =
49 ppm): d= 21.0 (t, COCH3), 27.8 and 28.2 (t, C-8, C-11), 34.7
1
signal, d, C-4ꢀ), 115.8 (s, CF3, JC,F = 288.1 Hz), 116.0 (s, CF3,
2
1JC,F = 288.6 Hz), 155.7 (s, COCF3, JC,F = 36.5 Hz), 156.0 (s,
2
COCF3, JC,F = 36.7 Hz), 169.2 (q, COCH3), 170.1 (q, COCH3)
ppm. HRMS (ESI): m/z for positive ions: calculated: 701.2485
(M + Na+); found: 701.2452. HRMS (ESI): m/z for negative ions:
calculated: 677.2509 (M − H+); found: 677.2493.
Aminoalcohol 32. Based on the procedure described for the
preparation of aminoalcohol 30, a mixture of the protected
aminoalcohols 31a and 31b (27.4 mg, 0.040 mmol) were dissolved
in a mixture of water and methanol (1 : 1; 4 ml) and were treated
with NaOH (19.4 mg, 0.485 mmol). The solution was stirred for
2 h at RT and was then neutralised with dry ice. After evaporation
of all volatile compounds the residue was purified by reverse-phase
chromatography (RP18; H2O → H2O–MeOH = 4 : 1 → 3 : 2 →
1 : 1) to furnish product 32 (12.9 mg, 0.032 mmol; 81%) as a
colourless sticky solid.
1
[a]2D0 = −11.1 (c = 1, MeOH); H NMR (400 MHz, CD3OD,
CD3OH = 3.31 ppm): d = 1.46–1.80 (m, 14 H), 1.81–1.91 (m, 2 H,
2-H, 6-H), 2.90 (dddd, J = 11.2, 11.2, 3.9, 3.9 Hz, 1 H, 1-H), 2.96
(dddd, J = 9.5, 9.5, 3.9, 3.9 Hz, 1 H, 1ꢀ-H), 3.40–3.47 (m, 2 H,
4ꢀ-H, CH2), 3.52–3.63 (m, 5 H, 5-H, CH2), 3.63–3.70 (m, 3 H, 5-H,
2 × CH2), 3.72–3.79 (m, 2 H, 5ꢀ-H, CH2), 3.88 (ddd, J = 10.0, 3.5,
2.7 Hz, 1 H, 3ꢀ-H), 3.94 (br d, J = 2.7 Hz, 1 H, 4-H) ppm; 13C
NMR (100 MHz, CD3OD, CD3OD = 49 ppm): d 26.7 (t), 27.1 (t),
27.2 (t), 28.3 (t), 34.9 (t, C-2), 35.6 (t, C-6), 36.3 (t), 38.6 (t), 45.7
and 45.9 (d, C-1, C-1ꢀ), 68.6 (t), 69.1 (d, C-4, C-3ꢀ), 70.2 (t), 70.8
(t), 71.4 (t), 76.1 (d, C-5, C-5ꢀ), 79.2 (d, C-3), 80.0 (d, C-4ꢀ) ppm.
HRMS (ESI): m/z for positive ions: calculated: 425.2628 (M +
Na+); found: 425.2636.
(t, C-2), 36.6 (t, C-6), 44.6 (d, C-1), 69.3 (d, C-3), 71.6 and 71.7
1
(t, C-7, C-10), 77.2 (d, C-5), 82.7 (d, C-4), 117.4 (s, CF3, JC,F
=
2
286.7 Hz), 158.2 (s, COCF3, JC,F = 37.0 Hz), 171.9 (s, COCH3)
ppm. HRMS (ESI): m/z for positive ions: calculated: 701.2485
(M + Na+); found: 701.2482.
Aminoalcohol 36. Based on the procedure described for the
preparation of aminoalcohol 30, the protected aminoalcohol 35
(77.6 mg, 0.114 mmol) and NaOH (91.5 mg, 2.288 mmol) were
dissolved in a mixture of water and methanol (1 : 1; 4 ml). The
resulting residue was purified by reverse-phase chromatography
(RP18; H2O → H2O–MeOH = 4 : 1 → 3 : 2 → 1 : 1) to isolate
aminoalcohol 36 (28 mg, 0.070 mmol; 61%) as a colourless syrup.
Macrocycle 35. According to the procedure described for
the preparation of macrocycle 29, bisallylether 27 (361 mg,
0.512 mmol) and Grubbs catalyst 13 (31 mg, 0.010 mmol)
were dissolved in 170 ml degassed CH2Cl2. The resulting crude
product was purified by column chromatography over silica gel
(ethyl acetate–petroleum ether = 3 : 1) to yield the unsaturated
macrocycle (230 mg, 0.340 mmol; 66%) as a colourless oil as
well as the starting material 27 (43.2 mg, 0.061 mmol; 12%).
The macrocycle (230 mg, 0.340 mmol) was dissolved in 5 ml
of ethyl acetate–CH2Cl2–MeOH and treated with Pt2O (31 mg,
0.136 mmol) under a hydrogen atmosphere to furnish macrocycle
35 (187 mg, 0.276 mmol; 81%) as a colourless oil.
1
[a]2D0 = −93.4 (c = 1, MeOH); H NMR (400 MHz, CD3OD,
CD3OH = 3.31 ppm): d = 1.11 (ddd, J = 11.8, 11.8, 11.6 Hz, 2 H,
6b-H), 1.29 (ddd, J = 13.5, 11.5, 2.6, 2 H, 2b-H), 1.68 (m, 4 H,
11-H), 1.75 (m, 4 H, 8-H), 2.02 (ddd, J = 13.5, 6.5, 3.5 Hz, 2 H,
2a-H), 2.21 (ddd, J = 11.8, 7.1, 4.0 Hz, 2 H, 6a-H), 3.08 (dddd,
J = 11.6, 11.5, 4.0, 3.9 Hz, 2 H, 1-H), 3.11 (dd, J = 9.2, 3.0 Hz,
2 H, 4-H), 3.47 (dt, J = 4.5, 5.2 Hz, 2 H, 7b-H), 3.52–3.66 (m, 4 H,
5-H, 10b-H), 3.67 (ddd, J = 8.9, 6.0, 5.6 Hz, 2 H, 10a-H), 3.75 (dt,
J = 8.7, 5.4 Hz, 2 H, 7a-H), 4.14 (ddd, J = 3.1, 3.5, 2.6 Hz, 2 H,
3-H) ppm; 13C NMR (100 MHz, CD3OD, CD3OD = 49 ppm):
d = 28.0 and 28.1 (t, C-8, C-11), 41.0 (t, C-2, C-6), 44.5 (d, C-1),
67.5 (d, C-3), 71.3 (t, C-7, C-10), 77.5 (d, C-5), 85.0 (d, C-4) ppm;
HRMS (ESI): m/z for positive ions: calculated: 403.2808 (M +
H+); found: 403.2819.
Rf = 0.48 (CH2Cl2–MeOH = 9 : 1); [a]2D0 = −48.2 (c = 1, CHCl3);
room temperature NMR spectra with CDCl3 as solvent showed
broadened signals: 1H NMR (400 MHz, CDCl3, TMS = 0 ppm):
d = 1.31–1.45 (m, 2 H, 6b-H), 1.46–1.71 (m, 10 H, 2b-H, 8-H and
11-H), 1.98–2.09 (m, 8 H, Ac, 2a-H), 2.11–2.29 (m, 2 H, 6a-H),
3.21–3.33 (m, 2 H, CH2), 3.35–3.48 (m, 4 H, 4-H, CH2), 3.57–3.74
(m, 6 H, 5-H, CH2), 4.12–4.26 (m, 2 H, 1-H), 5.18–5.35 (br s,
2 H, 3-H), 6.64–7.11 (br m, 2 H, NH) ppm; at 220 K signals
sharpened considerably and three ring conformations (labelled A,
B, C) became visible: 1H NMR (400 MHz, CDCl3, TMS = 0 ppm):
dH = 1.11–1.21 (m, 3 H, 6B, ax-H), 1.21–1.30 (m, 1 H), 1.31–1.43 (m,
3 H), 1.45–1.61 (m, 4 H), 1.62–1.69 (m, 1 H, 6A, eq-H), 1.79–1.94
(m, 3 H, 6A, ax-H), 2.00–2.06 (m, 3 H), 2.08 (s, 3 H, COCH3), 2.12
Acknowledgements
This work was supported by the DFG (SFB 416, project B2) and
the Fonds der Chemischen Industrie. We are indebted to E. Hofer
(Leibniz University of Hannover) and S. Glaser (Technical
University Mu¨nchen) for expert advice and support in NMR
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2423
©

View Article Online
studies. R. Wartchow and Dr Wiebcke (Institute of Inorganic
Chemistry, Leibniz University of Hannover) are thanked for the
X-ray analysis of 31b.
11 A. Barco, S. Benetti, C. De Risi, P. Marchetti, G. P. Pollini and V.
Zanirato, Tetrahedron: Asymmetry, 1997, 8, 3515–3545.
12 B. M. Trost and A. G. Romero, J. Org. Chem., 1986, 51, 2332–2342.
13 J. E. Audia, L. Boisvert, A. D. Patten, A. Villalobos and S. J.
Danishefsky, J. Org. Chem., 1989, 54, 3738–3740.
14 Ketone 8 was synthesised via a NaIO4-supported diol cleavage. The
reported yield for this step was substantially improved (to 96%) by
using silica-supported NaIO4 in CH2Cl2: Y.-L. Zhong and T. K. M.
Shing, J. Org. Chem., 1997, 62, 2622–2624.
15 In contrast to our reported protocol in which catalytic amounts of
nickel acetate were efficient, the author in this example employed a
stoichiometric amount of nickel acetate: J. Ipaktschi, Chem. Ber., 1984,
117, 856–858.
References
1 An illustrative example has been published for epothilone: T.
Carlomagno, M. J. J. Blommers, J. Meiler, W. Jahnke, T. Schupp,
F. Petersen, D. Schinzer, K.-H. Altmann and C. Griesinger, Angew.
Chem., 2003, 115, 2537–2539, (Angew. Chem., Int. Ed., 2003, 42, 2511–
2512).
2 (a) J. Clayden and N. Vassilou, Org. Biomol. Chem., 2006, 4,
2667–2678; (b) Molecular switches, ed. B. L. Feringa, Wiley-VCH,
Weinheim/Chichester, 2001; (c) A. P. de Silva, H. Q. Gunaratne, T.
Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and
T. E. Rice, Chem. Rev., 1997, 97, 1515–1566.
16 P. Fresenius, K. Go¨rlitzer, Organisch-chemische Nomenklatur: 4. Au-
flage, Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart, 1998,
pp. 300–302.
17 The second generation Grubbs II precatalyst, unlike the first generation
analogue, led to almost quantitative double bond isomerisation and
formation of the corresponding enol ethers (see ref. 8b): (a) R. Uma,
C. Crevisy and R. Gree, Chem. Rev., 2003, 103, 27–52 (review); (b) M.
Arisawa, Y. Terada, M. Nakagawa and A. Nishida, Angew. Chem.,
2002, 114, 4926–4928, (Angew. Chem., Int. Ed., 2002, 41, 4732–4734);
(c) M. K. Gurjar and P. Yakambram, Tetrahedron Lett., 2001, 42,
3633–3636; (d) D. C. Braddock and A. J. Wildsmith, Tetrahedron Lett.,
2001, 42, 3239–3242; (e) T. R. Hoye and H. Zhao, Org. Lett., 1999, 1,
1123–1125.
18 (a) Analysis of the E/Z ratio has been achieved: E. Breitmaier and
W. Voelter, in Carbon-13 NMR Spectroscopy. High Resolution Methods
and Applications in Organic Chemistry and Biochemistry, VCH, New
York, 1987, p. 192; (b) S. T. Diver and S. L. Schreiber, J. Am. Chem.
Soc., 1997, 119, 5106–5109; (c) B. Luy, G. Hauser, A. Kirschning and
S. J. Glaser, Angew. Chem., 2003, 115, 1338–1341, (Angew. Chem. Int.
Ed., 2003, 42, 1125–1132), and are described in detail in the ESI†.
19 (a) S. David, in Preparative Carbohydrate Chemistry, Marcel Dekker,
New York, 1997, p. 69–72; (b) S. David and S. Hanessian, Tetrahedron,
1985, 41, 643–663.
3 (a) R. Krauss and U. Koert, Synlett, 2003, 598–608; (b) M. Karle, D.
Bockelmann, D. Schumann, C. Griesinger and U. Koert, Angew. Chem.,
2003, 115, 4684–4687, (Angew. Chem., Int. Ed., 2003, 42, 4546–4549).
4 (a) H. Yuasa and H. Hashimoto, J. Am. Chem. Soc., 1999, 121, 5089–
5090; (b) S. E. O’Connor and B. Imperiali, J. Am. Chem. Soc., 1997, 119,
2296; (c) H. J. Schneider and F. Werner, J. Chem. Soc., Chem. Commun.,
1992, 490–491; (d) V. A. Palyulin, S. V. Emets, V. A. Chertkov, C. Kasper
and H.-J. Schneider, Eur. J. Org. Chem., 1999, 3479–3482; (e) J. N. H.
Reek, H. Engelkamp, A. E. Rowan, J. A. A. W. Elemans and R. J. M.
Nolte, Chem. Eur. J., 1998, 4, 716–722; (f) V. V. Samoshin, B. Brazdova,
V. A. Chertkov, D. E. Gremyachinskiy, A. K. Shestakova, E. K.
Dobretsova, L. P. Vatlina, J. Yuan and H.-J. Schneider, ARKIVOC,
2005, 192–141.
5 A particularly impressive library was recently disclosed: B. Clique, A.
Ironmonger, B. Whittaker, J. Colley, J. Titchmarsh, P. Stockley and A.
Nelson, Org. Biomol. Chem., 2005, 3, 2776–2785.
6 For modified aminoglycosides in which substitution of one or more
carbohydrates was achieved, see: (a) W. C. Park, M. Auer, H. Jaschke
and C.-H. Wong, J. Am. Chem. Soc., 1996, 118, 10150–10155; (b) Y.
Ding, S. A. Hofstadler, E. E. Swayze and R. F. Griffey, Org. Lett.,
2001, 3, 1621–1623; (c) P. B. Alper, M. Hendrix, P. Sears and C.-H.
Wong, J. Am. Chem. Soc., 1998, 120, 1965–1978; (d) H. Wang and
Y. To r , Angew. Chem., 1998, 110, 117–120, (Angew. Chem., Int. Ed.,
1998, 37, 109–111); (e) D. Vourloumis, M. Takahashi, G. C. Winters,
K. B. Simonsen, B. K. Aiyda, S. Barluenga, S. Qamar, S. Shandrick, Q.
Zhao and T. Hermann, Bioorg. Med. Chem. Lett., 2002, 12, 3367–3372;
(f) B. Franc¸ois, J. Szychowski, S. S. Adhikari, K. Pachamuthu, E. E.
Swayze, R. H. Griffey, M. T. Migawa, E. Westhof and S. Hanessian,
Angew. Chem., 2004, 116, 6903–6906, (Angew. Chem., Int. Ed., 2004,
43, 6735–6738); (g) F.-S. Liang, S. K. Wang, T. Nakatani and C.-H.
Wong, Angew. Chem., 2004, 116, 6658–6662, (Angew. Chem., Int. Ed.,
2004, 43, 6496–6500); (h) W. Greenberg, E. S. Priestley, P. S. Sears,
P. B. Alper, C. Rosenbohm, M. Hendrix, S.-C. Hung and C.-H. Wong,
J. Am. Chem. Soc., 1999, 121, 6527–6541; (i) C.-W. T. Chang, Y. Hui,
B. Elchert, J. Wang, J. Li and T. Rai, Org. Lett., 2002, 4, 4603–4606;
(j) J. Wang, J. Li, D. Tuttle, Y. Takemoto Jon and T. Chang Cheng-Wie,
Org. Lett., 2002, 4, 3997–4000.
7 For orthogonally-protected sugar diamino acids as building blocks for
linear and branched amino glycoside derivatives, see: F. Sicherl and V.
Wittmann, Angew. Chem., 2005, 117, 2133–2136, (Angew. Chem., Int.
Ed., 2005, 44, 2096–2099).
8 (a) J. Jaunzems and A. Kirschning, Org. Biomol. Chem., 2004, 2, 3448–
3456; (b) G.-w. Chen, J. Jaunzems, M. Jesberger, M. Kalesse, M. Lindner
and A. Kirschning, Tetrahedron, 2004, 60, 3505–3521; (c) G.-w. Chen
and A. Kirschning, Chem. Eur. J., 2002, 8, 2717–2729; (d) A. Kirschning
and G.-w. Chen, Synthesis, 2000, 1133–1137; (e) A. Kirschning and G.-
w. Chen, Tetrahedron Lett., 1999, 40, 4665–4668.
20 Selected examples:(a) A. Bernardi, D. Arosio, L. Manzoni, F. Micheli,
A. Pasquarello and P. Senesi, J. Org. Chem., 2001, 66, 6219–
6216; (b) D.-S. Wang, A.-L. Hsu, X. Song, C.-M. Chiou and C.-S.
Chen, J. Org. Chem., 1998, 63, 5430–5437; (c) U. Schlueter, J. Lu and
B. Fraser-Reid, Org. Lett., 2003, 5, 255–257.
21 A. D. McNaught, ‘Nomenclature of Carbohydrates’, Pure Appl. Chem.,
1996, 68, 1919–2008.
22 (a) M. Karplus, J. Chem. Phys., 1959, 30, 11–15; (b) M. Karplus, J. Am.
Chem. Soc., 1963, 2870–2871.
23 X-ray crystallographic study of 31b (Fig. 4): A single crystal (0.10 ×
0.44 × 0.22 mm) of 31b was investigated on a Stoe IPDS area detec-
tor diffractometer using graphite-monochromated Mo-Ka radiation
(71.073 pm). The lattice constants were determined by the automatic
Search and Indexing routines of the diffractometer. Indexing and
refinement of 4179 reflections led to a monoclinic unit cell with the
dimensions: a = 1194.9(2) pm, a = 90◦, b = 931.3(1) pm, b = 108.84(2)◦,
c = 1528.9(3) pm, c = 90◦ (V = 1665.2(5) × 106 pm3; Z = 2; Dx =
1.353 g cm−3). The space group P21 (see ref. 30) was determined by
the reflection conditions taken into account. The intensity data of
24154 reflections were measured at room temperature (300(2) K) in
the H-range of 2.42◦ to 26.22◦, yielding 6594 unique reflections. For Lp
correction and data processing the programs of the Stoe IPDS software
package were used. The structure was solved by direct methods and
refined using the programs SHELXS-86 and SHELXL-97 (ref. 31).
The atomic parameters of the molecule (excluding H atoms) were
determined by Fourier difference syntheses and refined (anisotropic
displacement parameters for non-F,O,N-atoms only) by full-matrix
least squares (F2) to the R values R1 = 0.0457 [for 1826 reflections
I > 2r(I)], wR2 = 0.0730. For all reflections, the hydrogen atoms
were included as riding atoms using constraints and restraints. Carbon
atoms were refined isotropically only due to the low ratio of observed
reflections to parameters. The absolute configuration of the compound,
which is known from chemical reasoning, could only be proven in
this X-ray study using Mo-Ka radiation. The Flack X parameter was
−0.1(10) and is meaningless here. CCDC reference number 673913.
For crystallographic data in CIF or other electronic format see DOI:
10.1039/b716347a†.
9 D. Ragunathan, V. M. Sa`nchez-Pedregal, J. Junker, J. Jaunzems, K.
Michaelis, M. Kalesse, A. Kirschning and T. Carlomagno, Nucleic Acid
Res., 2006, 34, 3599–3608.
10 For selected examples for enhanced conformational flexibility of
substituted cyclohexanes (cyclitols) compared to substituted pyranes
(sugars), see: (a) M. Carpintero, A. Bastida, E. Garc´ıa-Junceda, J.
Jime´nez-Barbero and A. Ferna´ndez-Mayorales, Eur. J. Org. Chem.,
2001, 21, 4127–4135; (b) T. Sakakibara, K. Suzuki, A. Sakai, M. Scindo,
C. Nagano, S. Nerumi, Y. Kajihara and K. Mochizuki, Tetrahedron
Lett., 2003, 30, 5711–5712.
24 Although macrocycle 31a is also a crystalline solid, we were unable to
obtain crystals suitable for X-ray analysis.
2424 | Org. Biomol. Chem., 2008, 6, 2412–2425
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
25 K. Okajima, T. Mukae, H. Imagawa, Y. Kwamura, M. Nishizawa and
H. Yamada, Tetrahedron, 2005, 61, 3497–3506.
26 (a) G. Binsch, J. Am. Chem. Soc., 1969, 91, 1304–1309; (b) D. A. Kleier
and G. Binsch, J. Magn. Reson., 1970, 3, 146–160.
27 D. Ho¨fner, S. A. Lesko and G. Binsch, Org. Magn. Reson., 1978, 11,
179.
29 Depending on the relative configuration of all substituents at the pyran
ring, these dienes cyclise directly or after dimerisation/trimerisation,
respectively, to yield larger macrocycles, for which the larger conforma-
tional rigidity of the pyran rings can be held responsible.
30 International Tables for Crystallography, Vol. A, 2nd edn, ed. T. Hahn,
Springer, Dordrecht, 1989.
28 The biological properties of these macrocycles will be reported else-
where.
31 G. M. Sheldrick, SHELXS-86, University of Go¨ttingen, Germany;
G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 467–473.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 2412–2425 | 2425
©