The use of deep cavity tetraformyl calix[4]arenes in the synthesis of static and dynamic macrocyclic libraries

Article abstract of DOI:10.1016/j.tetlet.2005.01.168

This communication reports the first synthesis of a number of deep cavity tetraformyl calix[4]arene macrocycles and their use as key building blocks in the synthesis of a dynamic and the first static non-peptide macrocyclic library.

Full text of DOI:10.1016/j.tetlet.2005.01.168

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 2059–2062
The use of deep cavity tetraformyl calix[4]arenes in the synthesis
of static and dynamic macrocyclic libraries
Nikolai Kuhnert* and Adam Le-Gresley
Supramolecular and Biological Chemistry Laboratory, Chemistry, School of Biomedical and Molecular Sciences,
The University of Surrey, Guildford GU2 7XH, UK
Received 23 September 2004; revised 17 January 2005; accepted 26 January 2005
Abstract—This communication reports the first synthesis of a number of deep cavity tetraformyl calix[4]arene macrocycles and their
use as key building blocks in the synthesis of a dynamic and the first static non-peptide macrocyclic library.
ꢀ 2005 Published by Elsevier Ltd.
We have recently argued that in order to be successful in
host–guest chemistry using synthetic macrocyclic recep-
tors the principles of combinatorial chemistry should be
applied. Since host–guest chemistry is guest-centred, a
synthetic receptor binding to a given guest of choice
with a high binding constant and high selectivity needs
to be identified efficiently, necessitating the development
of novel synthetic methodology for the synthesis of
libraries of macrocycles.1 These methodologies can either
aim at static libraries2 or at dynamic combinatorial
libraries (DCL)3 as proposed by Lehn4,5 and Sanders
and co-workers.6–8 In contrast to Sandersꢀ approach
where a series of difunctional building blocks were
allowed to form a DCL comprising macrocyclic receptors
of various ring sizes along with open-chain receptors in
the presence or absence of a guest, we present in this
work a novel approach. In this approach we used a pre-
organised macrocycle, in our case a calix[4]arene, as an
ideal platform for a macrocyclic library. The lipophilic
cavity of the calix[4]arene provides a mildly attractive
recess for similarly lipophilic components of a guest,
without impinging on any self-assembly process of
appropriately functionalised building blocks. The upper
rim of the calix[4]arene should possess several function-
alities suitable for further elaboration with a series of
structurally diverse reagents to obtain static and dy-
namic libraries. Dynamic imine libraries of carcerands9
and
a
series of static cyclopeptide macrocyclic
libraries10–12 have been reported.
To achieve the goal of developing a novel synthetic
methodology for macrocyclic library synthesis we
O
O
O
O
O
O
O
O
I
I
I
I
O
O
5 mol% Pd(OAc)2
dppp/NEt3/DMF
O
O
O
O
5 mol% Pd(OAc)2
dppp/NEt3/DMF
O
O
O
O
2
O
O
O
O
O
O
3
OR
OR
RO
RO
O
O
OHC
O
O
OR
OR
OHC
OR
OR
1
RO
RO
RO
RO
R = n-Bu
*
Corresponding author. E-mail: n.kuhnert@surrey.ac.uk
0040-4039/$ - see front matter ꢀ 2005 Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2005.01.168

2060
N. Kuhnert, A. Le-Gresley / Tetrahedron Letters 46 (2005) 2059–2062
extended our recently developed Heck coupling method-
ology13,14 to obtain deep cavity tetraformyl calix[4]arene
acrylates 2, 3 from tetraiodo compound 1 and the form-
ylacrylate using palladium acetate and 1,3-bis-(diph-
enylphosphino)propane as a coligand. The coupling
proceeded as expected to give the deep cavity calix[4]-
arenes 2 and 3 in good yields as the all-trans isomers.
All spectroscopic data were in agreement with the struc-
ture. The chemical shifts of the calix[4]arene methylene
protons provided evidence for the deep cavity nature
of the compounds as discussed previously.13,15
was carried out with a 2-fold excess of the amine in chlo-
roform in the presence of molecular sieves. Within the
static library 26 compounds were targeted and 21 could
be isolated in analytically pure form and were fully char-
acterised (see Table 1, entries 1–30). All the tetra-imines
displayed the expected C4V symmetry along with the ex-
pected spectroscopical data, supporting their structures.
However, in the case of more electron-deficient amines
such as h and i the reaction proceeded slowly even in
the presence of small quantities of mineral or Lewis
acids at elevated temperatures. Condensation reactions
producing mixtures of mono-, di-, tri- and tetra-imines
1
With the tetraformyl calix[4]arene acrylates16 in hand we
proceeded to synthesise a small library of tetra-imines
using 2 and 3 along with a selection of 16 structurally di-
verse aromatic, aliphatic and functionalised amines a–p.
We attempted the condensation of 10 selected amines a–
j with 4-formyl derivative 2 to give tetra-imines 4aaaa–
4jjjj (each lower case letter signifies one imine linkage
with a particular amine derived substituent) and the
condensation of all 16 amines with 3-formyl derivative
3 to give tetra-imines 5aaaa–5pppp. Imine condensation
were characterised as mixtures by H NMR and FAB
mass spectrometry (see Table 1, entries 4, 10, 15, 24
and 30).
Reaction of the bases cytosine j and adenine k occurred
to a limited extent of 2–4% and even this could only be
achieved in a DMSO-d6/CDCl3 mixture (1:1) owing to
their low solubility in CDCl3. Similarly, the free amino
acids L-alanine, L-serine, L-cysteine, L-tryptophan, L-his-
tidine, L-valine and L-methionine failed to react despite
NH2
NH2
NH2
H2N
NH2
R'
N
R'
O
O
O
R'
N
O
R'
N
N
O
a
c
d
e
O
b
NH2
OMe
OH
CO2Me
OMe
OtBu
H2N
NH2
Amines a-p
O
H2N
O
O
O
N
H
O
O
g
O
O
O
f
h
i
O
O
O
O
2
O
4aaaa-
O
O
O
NH2
NH2
NH2
NH2
4pppp
OH
N
HN
N
NH2
N
H
O
N
N
n
j
k
l
m
OR
OR
RO
RO
OR
OR
RO
RO
(EtO)3Si
NH2
NH2
o
p
Table 1. Yields of tetra-imines 4aaaa–4jjjj and 5aaaa–5pppp
Entry
Major producta
Yieldb (%)
Entry
Amines
Major producta
Yieldb (%)
1
a
4aaaa
4bbbb
4cccc
4dddd
4eeee
4ffff
80
78
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
e
5eeee
5ffff
83
77
2
b
f
3
c
88
Mixtured
g
5gggg
5hhhh
5iiii
45
4
d
h
49%
78
Mixtured
59
83
5
e
54
86
75
73
74
i
6
f
j
5jjjj
5kkkk
5llll
7
g
4gggg
4hhhh
4iiii
k
8
h
l
9
i
m
5mmmm
5nnnn
5oooo
5pppp
4accg
4aacf
4ccg
39
67
10
11
12
13
14
15
j
4jjjj
Mixtured
c
n
Amino acidsc
o
84
Mixtured
a
b
c
5aaaa
5bbbb
5cccc
5dddd
75
72
p
e
acgf
acgf
acgf
g
f
68
Mixtured
d
a As identified by 1H NMR and FAB-MS.
b Isolated yield.
c No reaction occurred with L-alanine, L-serine, L-cysteine, L-tryptophan, L-histidine, L-valine and L-methionine.
d Mixture of isomers not separated.
e In the absence of external template.
f In the presence of biotin.
g In the presence of barbituric acid.

N. Kuhnert, A. Le-Gresley / Tetrahedron Letters 46 (2005) 2059–2062
2061
OMe
OMe
N
N
O
N
N
O
N
N
N
N
O
amines a, c, f, g
amines a, c, f, g
O
HN
NH
OR
O
O
O
O
O
O
O
O
2
4aacf
O
O
O
O
O
O
O
O
HN
NH
Barbituric acid
H
H
OH
4accg
S
OR
OR
O
OR
OR
RO
RO
Biotin
RO
RO
long reaction times according to the literature condi-
tions.17,18 As observed in non-macrocyclic cases19,20 es-
ter protected amino acids h and i showed an increased
degree of reactivity and successfully yielded the required
tetra-imines.21
As a control experiment we added two non-amine tem-
plates (barbituric acid and biotin) to the dynamic library
described above. Again LSIMS-MS allowed us to iden-
tify the major components of the library after 48 h of
equilibration. The major product when barbituric acid
was present was 4aacf, which was not present in the
above mixture. In the case of biotin the major reaction
Generally the reaction times of amines a–p loosely re-
flect (Table 1) their relative nucleophilicity. Within the
static library, 26 compounds were targeted and 21 could
be isolated in analytically pure form.
1
product was tris-imine 4ccg.23 The H NMR spectra of
the dynamic library support the formation of a major
tetra-imine 4aacf and tris-imine 4ccg, respectively.21
Additional evidence for binding of the guest template
to a component of the dynamic library comes from dif-
fusion NMR experiments,14,24 indicating a reduced dif-
fusion coefficient of barbituric acid and biotin benzyl
ammonium salt in the presence of the dynamic library
as opposed to the templates in the absence of the dy-
namic library. These experiments clearly indicated that
when alternative guest templates were introduced to
the dynamic library the equilibrium mixture shifted con-
siderably towards a thermodynamically more stable
host–guest system and result in molecular amplification
of individual DCL members. Current work is focussing
on the isolation and full characterisation of the novel
macrocyclic receptors.
After the successful synthesis of a static library we
turned our attention to a dynamic library based on the
imine condensation reaction of tetraformyl compounds
2 and 3. Imine condensation is a reversible reaction
therefore allowing the exploitation of thermodynamic
control in a highly modular approach. In a dynamic
combinatorial library the potential number of unique
calix[4]arene compounds possible when 2 or 3 are re-
acted with four different amines could comprise 146 dif-
ferent imines in the cone conformation (including mono-,
di-, tri- and tetra-imines, regioisomers and possible ste-
reoisomers). The inclusion of an additional amine adds
another 105 possible permutations of imine reaction
products to the dynamic library, clearly illustrating the
immense diversity that can be achieved with our ap-
proach. The possibility of exerting thermodynamic con-
trol via a guest and the subsequent predominance of one
(or significantly fewer than 146) compounds would cer-
tainly illustrate the potential value of the method
described.
In conclusion, we have demonstrated that deep cavity
tetraformyl calix[4]arenes are ideal scaffolds for the syn-
thesis of the first static macrocyclic library and can be
used successfully to create a dynamic combinatorial li-
brary. The synthetic methodology developed will allow
us to exploit the libraries in supramolecular chemistry.
A representative example of this thermodynamic prefer-
ence was given by LSIMS–MS analysis of the products,
when 2 was mixed with stoichiometric amounts of
amines a, c, f and g. After 12 h reaction time a statistical
mixture of at least 30 imines (as judged by the intensity
of the molecular ions in the LSIMS MS) was observed.22
After a prolonged reaction time of 48 h there was strong
evidence for equilibration under thermodynamic control
as illustrated by the predominance of a heterocondensa-
tion product 4accg (m/z 1684) with smaller amounts of
4aacc, 4cccf and 4aaaf as judged by the relative intensi-
ties of the molecular ions in the LSIMS-MS. Longer
reaction times did not change the composition of the
DCL significantly. It can be assumed that one of the
amines acts as a templating guest to induce the forma-
tion of the four most stable tetra-imines.
Acknowledgements
We acknowledge EPSRC funding for a studentship to
ALG and further financial support by the European
Union (project pesticide No. ICA2-CT-2000-10002).
References and notes
1. Kuhnert, N.; Rossignolo, G.; Lopez-Periago, A. Org.
Biomol. Chem. 2003, 1, 1157–1170.
2. Terret, N. Combinatorial Chemistry, 1st ed.; OUP:
Oxford, 1998.
3. Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J.
K. M.; Stoddard, J. F. Angew. Chem., Int. Ed. 2002, 41,
898–952.

2062
N. Kuhnert, A. Le-Gresley / Tetrahedron Letters 46 (2005) 2059–2062
4. Huc, C.; Lehn, J. M. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 2106–2110.
5. (a) Lehn, J. M. Supramolecular Chemistry: Concepts and
Perspectives; VCH: Weinheim, 1995; (b) Lehn, J. M.
Chem. Eur. J. 1999, 5, 2455–2463.
heated to 100 ꢁC and stirred for 36 h. After addition of
(50 ml) diethylether the mixture was washed with HCl
(3 · 20 ml 3 M HCl). The organic extract was dried
(Na2SO4), filtered and evaporated under reduced pressure.
The residue was further dried in a desiccator (SiO2) under
reduced pressure for 12 h. Purification via column chro-
6. Brady, P. A.; Sanders, J. K. M. J. Chem. Soc., Perkin
Trans. 1 2001, 3237–3239.
matography (SiO2, DCM) gave the calix[4]arene
3
7. Furlan, R. L. E.; Otto, S.; Sanders, J. K. M. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 4801–4804.
8. Otto, S.; Furlan, R. L. E.; Sanders, J. K. M. Drug Discov.
Today 2002, 7, 117–125.
9. Ro, S.; Rowan, S. J.; Pease, A. R.; Cram, D. J.; Stoddard,
J. F. Org. Lett. 2000, 2, 2411–2414.
10. Sutton, P. W.; Bradley, A.; Farras, J.; Romea, P.; Urpi,
F.; Vilarrasa, J. Tetrahedron 2000, 56, 7947–7958.
11. Quin, C. G.; Bu, X. Z.; Zhong, X. F.; Ng, N. L. G.; Guo,
Z. H. J. Comb. Chem. 2004, 6, 398–406.
12. Takahashi, T.; Nagamiya, H.; Doi, T.; Griffith, P. G.;
Bray, A. M. J. Comb. Chem. 2003, 5, 414–428.
13. Kuhnert, N.; Le-Gresley, A. J. Chem. Soc., Perkin Trans.
1 2001, 3393–3398.
14. Kuhnert, N.; Le-Gresley, A. Chem. Commun. 2003, 2426–
2427.
(0.429 g,68%) as cream shards; mp 98–104 ꢁC; mmax
(Nujol)/cmÀ1 1730 (C@O), 1720 (C@O), 1630 (C@C); dH
(300 MHz, CDCl3) 9.81 (4H, s, CHO), 7.76 (4H, d, J 6.9,
Ar-H), 7.61 (4H, s, Ar-H), 7.78–7.75 (4H, m, Ar-CH@),
7.41–7.35 (8H, m, Ar-H), 6.93 (8H, s, Ar-H), 6.34 (4H, d, J
15.2, CH–C@O), 4.51 (4H, d, J 14.0, CHAHBAr), 3.99
(8H, q, J 6.9, OCH2CH2), 3.26 (4H, d, J 14.0, CHAHBAr),
1.91–1.90 (8H, m, CH2CH2O), 1.48–1.41 (8H, m,
OCH2CH2CH2), 1.00 (12H, t, J 8.3, CH3CH2); dc (CDCl3)
191.2, 165.4, 151.4, 147.3, 137.7, 135.6, 129.9, 129.9, 128.7,
128.5, 128.1, 126.6, 123.3, 114.6, 75.6, 32.5, 31.3, 19.5,
14.2; m/z (LSIMS) 1368 [M+Na]; Found: C, 69.29; H,
5.92. C84H80O166H2O requires C, 69.41; H, 6.38.
17. Felder, E. Chem. Ber. 1967, 100, 555–560.
18. Hope, A.; Halpern, B. Tetrahedron Lett. 1972, 13, 2261–
2264.
15. Gutsche, C. D. Aldrichim. Acta 1995, 28, 7.
16. (a) Synthesis of 2: Triethylamine (0.335 ml, 0.157 g,
19. Han, O.; Frey, P. J. Am. Chem. Soc. 1990, 112, 8982–8983.
20. Bey, P.; Vevert, J. P. Tetrahedron Lett. 1997, 17, 1455–
1458.
1.56 mmol)
1.56 mmol) were added to
and
4-acryloylbenzaldehyde
a
(0.276 g,
stirred solution of
21. Analytical data data of representative imine 4eeee: pale
brown powder; mp 74–79 ꢁC; mmax (Nujol)/cmÀ1 1732
(C@O), 1634 (C@N), 1602 (C@C); dH (300 MHz; CDCl3)
8.28 (4H, s, CH@N), 7.69 (8H, d, J 8.5, HCCCH@N), 7.51
(4H, d, J 15.5, Ar–CH@), 7.34 (4H, s, Ar-H (furfuryl), 6.90
(8H, s, Ar-H), 6.78 (8H, d, J 8.5, HCCO), 6.34–6.30 (8H, m,
CH-C@O, Ar-H (furfurylamine)), 6.12 (4H, s, Ar-H (furfur-
ylamine)), 4.76 (8H, s, CH2N), 4.50 (4H, d, J 13.9,
CHAHBAr), 4.19 (8H, q, J 6.3, OCH2CH2), 3.23 (4H, d, J
13.85, CHAHBAr), 1.91–1.88 (8H, m, CH2CH2O), 1.50–1.35
(8H, m, OCH2CH2CH2), 1.00 (12H, t, J 8.3, CH3CH2); dC
(CDCl3) 165.1, 160.3, 160.1, 152.1, 146.3, 135.8, 134.9, 133.1,
133.0, 128.9, 128.1, 127.0, 122.0, 121.5, 121.2, 118.9, 118.6,
110.7, 74.7, 61.2, 31.7, 30.4, 18.7, 13.9; m/z (LSIMS) 1662
[M+]; Found C, 68.23; H, 6.11; N, 3.10; C104H100O16N4 1.5
CHCl3 requires C, 68.83; H, 5.56; N 3.04.
22. FAB-MS spectra of representative equimolar mixtures of
tetra-imines revealed differences of relative intensities of
the molecular ions within 20%, indicating that LSIMS-MS
is in this case a valid technique for the analysis of
structurally closely related compounds of comparable
molecular weight. It is furthermore worth noting that the
library components could not be sufficiently resolved by
HPLC. For the validity of FAB-MS quantification of
compounds in mixtures see also: Manzi, A. E.; Dell, A.;
Azadi, P.; Varki, A. J. Biol. Chem. 1990, 265, 8094.
23. The DCL products 4aacf exist as three possible isomers
(regioisomers and enantiomeric stereoisomers) and 4ccg as
two possible regioisomers. From the NMR data obtained
it is currently not possible to distinguish and unambi-
giously assign the structure of the major reaction products.
24. Frish, L.; Mathews, L. E.; Bo¨hmer, V. J. Chem. Soc.,
Perkin Trans. 1999, 2, 669–672.
5,11,17,23-tetraiodocalixarene 25,26,27,28 tetra-n-butyl-
ether (0.3 g, 0.26 mmol) 1 in DMF (20 ml). After 10 min
vigorous stirring at 25 ꢁC palladium acetate (0.006 g,
0.026 mmol) and bis-diphenylphosphinopropane (0.011 g,
0.026 mmol) were added and the mixture heated to 100 ꢁC
and stirred for 48 h. After addition of (50 ml) diethylether
the mixture was washed with HCl (3 · 20 ml 3 M HCl).
The organic extract was dried (Na2SO4), filtered and
evaporated under reduced pressure. The residue was
further dried in a desiccator (SiO2) under reduced pressure
for 12 h. Purification via column chromatography (SiO2,
DCM) gave the calix[4]arene 2 (0.101 g, 32%) as cream
shards mp 103–105 ꢁC; mmax (Nujol)/cmÀ1 1748 (C@O
(ester)), 1710 (C@O), 1615 (C@C); dH (270 MHz; CDCl3)
9.99 (4H, s, CHO), 7.79 (8H, d, J 8.6, Ar-H (benzenal-
dehyde)), 7.62 (4H, d, J 15.9, Ar-CH@), 7.28 (8H, d, J 8.6,
Ar-H (benzenaldehyde)), 6.93 (8H, s, Ar-H), 6.37 (4H, d, J
15.9, Ar-CH@CH–C@O), 4.50 (4H, d, J 13.2, CHAHBAr),
3.97 (8H, t, J 6.4, CH2O), 3.19 (4H, d, J 13.2, CHAHBAr),
1.90 (8H, m, CH2CH2O), 1.50–1.35 (8H, m,
OCHCH2CH2), 1.00 (12H, t, J 8.3, CH3CH2); dc (CDCl3)
191.5, 165.4, 156.1, 147.2, 142.3, 136.0, 131.7, 131.3, 129.7,
128.8, 123.1, 114.8, 75.6, 32.5, 31.2, 19.7, 14.3; m/z (EI)
1345 [M+]. Found: C, 68.97; H, 5.92. C84H80O16Æ6H2O
requires C, 69.41; H, 6.38; (b) Synthesis of 3: Triethyl-
amine (0.650 ml, 0.320 g, 3.12 mmol) and 3-acryloylbenz-
aldehyde (0.540 g, 3.12 mmol) were added to a stirred
solution of 5,11,17,23-tetraiodocalixarene 25,26,27,28
tetra-n-butylether (0.6 g, 0.510 mmol) 1 in DMF (20 ml).
After 10 min vigorous stirring at 25 ꢁC palladium acetate
(0.0120 g, 0.052 mmol) and bis-diphenylphosphinopro-
pane (0.022 g, 0.052 mmol) were added and the mixture