Synthesis of novel rigid triazine-based calix[6]arenes

Article abstract of DOI:10.1016/S0040-4039(02)02884-8

In this report, the stepwise synthesis of a novel rigid and functionalised macrocycle 2 based on triazine and phenylenediamine linkers, is presented. Poor recognition of the macrocycle 2 for its substrates is observed, which shows experimentally that for the meltriazine based-calix[6]arene system, the binding ability of the melamine moiety gets more benefit from the ring flexibility derived from a xylenediamine linker 1 than from a phenylenediamine linker 2.

Full text of DOI:10.1016/S0040-4039(02)02884-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1359–1362
Synthesis of novel rigid triazine-based calix[6]arenes
Xiaoping Yang and Christopher R. Lowe*
Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QT, UK
Received 29 October 2002; revised 17 December 2002; accepted 20 December 2002
Abstract—In this report, the stepwise synthesis of a novel rigid and functionalised macrocycle 2 based on triazine and
phenylenediamine linkers, is presented. Poor recognition of the macrocycle 2 for its substrates is observed, which shows
experimentally that for the meltriazine based-calix[6]arene system, the binding ability of the melamine moiety gets more benefit
from the ring flexibility derived from a xylenediamine linker 1 than from a phenylenediamine linker 2. © 2003 Elsevier Science
Ltd. All rights reserved.
Cavitand bowls are fascinating supramolecular building
blocks.1 Methylene bridged resorcinarene derivatives
form the basis for several classes of host compounds of
significant contemporary interest, including carcerands,
hemicarcerands and calixarenes. The calix[4]arenes in
the cone conformation present an internal cavity which
can potentially host neutral guest molecules of comple-
mentary size.2 The importance of rigidity and, conse-
quently, of preorganisation of the cone conformer of
calix[4]arenes in determining their recognition ability,
recently has been combined.3 Since recent findings show
the possibility of rigidifying calix[4]arene-based hosts by
using different strategies, other approaches to the
recognition of biological molecules by the melamine
moiety have been investigated. The melamine moiety
has been shown to perform well as a recognition ele-
ment providing both hydrogen bond donor and accep-
tor sites for the recognition of biological molecules,
such as uracil,4 thymine4 and carbohydrates.5 More
recently, it has been shown that novel triaminotriazine
scaffold-based receptors inhibit substrate binding in the
presence of Cu(I).6 In this case, the receptors bearing
two bipyridine arms coordinate with the Cu(I) to form
a metal–ligand coordination compound which forces
two of the exocyclic CꢀN bonds of the triazine to rotate
thereby distorting the hydrogen-bonding surface, and
resulting reduction of the affinity of the melamine
receptor for uracil. Recently, some macrocyclic
molecules based on building blocks comprising a tria-
zine ring and a diamine linker were synthesised in our
group.7 These studies have shown that the triazine–
xylenediamine macrocycle 1 (Scheme 1) exhibited
promising binding properties towards cyanuric acid and
glycosides.
The association behaviour is determined predominantly
by the presence of multiple hydrogen bond donor and
acceptor sites. Although there are many examples of
artificial systems that display positive effects, where the
induced conformational change increases the binding
efficacy of the receptor, there are few reports of simple
negative receptors. Obviously, it is very helpful to view
the recognition ability in both ways, in order to under-
stand the intermolecular interactions and allow design
of molecular receptors useful for the preparation of a
new generation of sensing devices where efficiency and
selectivity are governed by molecular recognition pro-
cesses. We report here a stepwise synthetic route to
build up a novel rigid and functionalised macrocycle 2
based on triazine and phenylenediamine linkers, whose
structure is shown in Scheme 1. Investigation of the
binding ability shows limited recognition of molecules,
Keywords: triazine; macrocycle; inhibitor; hydrogen-bonding.
* Corresponding author. Tel.: +44-01223-334160; fax: +44-01223-
334162; e-mail: c.lowe@biotech.cam.ac.uk
Scheme 1.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02884-8

1360
X. Yang, C. Lowe / Tetrahedron Letters 44 (2003) 1359–1362
Scheme 2. Reaction conditions and yields:8 (i) (3-amino-phenyl) carbamic acid tert-butyl ester, NaHCO3, in acetone–H2O, at 0°C,
85%; (ii) (3-amino-phenyl) acetic acid benzyl ester, NaHCO3, in acetone–H2O, at 50°C, 91%; (iii) CH3(CH2)4NH2, THF, at 60°C,
95%; (iv) TFA, CH2Cl2; (v) 3, NaHCO3, in acetone–H2O, at 50°C, 84%; (vi) CH3(CH2)4NH2, THF, at 60°C, 91%; (vii) H2/Pd/C,
MeOH; (viii) cyanuric chloride, NaHCO3, in acetone–THF–H2O, at 0°C, 77%; (ix) 2 M HCl in dioxane; (x) DIPEA, DMF, 62%;
(xi) CH3(CH2)4NH2, DMF, at 60°C, 73%.
such as cyanuric acid, Me-a-O-D-glucopyranoside,
biotin and thymine, and suggests that for the meltri-
azine based-calix[6]arene system, the binding ability of
the melamine moiety gets more benefit from the ring
flexibility derived from a xylenediamine linker 1 than
from a phenylenediamine linker 2, where the methylene
bridge makes the melamine moiety form a well-ordered
hydrogen bonding surface which offers good recogni-
tion to the substrates.
The receptor 2 was prepared from cyanuric chloride
and mono-protected 1,3-phenylenediamine, as outlined
in Scheme 2. At each step of the synthesis, the chain
can either be elongated or cyclised to a macrocycle of
interest. The use of two different orthogonal protective
groups on either side of the oligomers allows control of
the length of the triazine–phenylenediamine chain. With
the aim of creating
a rigid macrocycle bearing
Figure 1. The plot of change of chemical shift (Dl) versus
equivalents of titrant (x). Titration of triazine–xylenediamine
melamine moieties, phenylenediamine was chosen as a
linker, instead of the xylenediamine exploited in macro-
cycle 1.7 The cyclic chloride 9 was synthesised from the
triazine–pheneylenediamine oligomers 8 by subsequent
treatment with acid and base, and was substituted with
macrocycle 1 with n-octyl-a-
b-
-glucopyranoside (") and cyanuric acid (ꢁ) in CDCl3.7b
Titration of triazine–phenylenediamine macrocycle 2 with Me-
a- -glucopyranoside (×) and cyanuric acid (ꢂ) in d6-DMSO.
D-glucopyranoside (ꢀ), n-octyl-
D
D

X. Yang, C. Lowe / Tetrahedron Letters 44 (2003) 1359–1362
1361
Figure 2. (a) Modelling structure for the binding of macromolecule 1 (R1=R2=R3=NH2) with cyanuric acid; (b) the lowest
energy structure for macrocycle 2 bearing a rigid cavity.
an excess of a third amine to give the macrocycle 2. No
significant dimer as identified by mass spectrometry was
found under the conditions employed. All new species
1994; pp. 85–130; (b) Haino, T.; Rudkevich, D. M.; Shiv-
anyuk, A.; Rissanen, K.; Rebek, J., Jr. Chem. Eur. J. 2000,
6, 3797–3805.
1
were characterised by H and 13C NMR spectroscopy
2. (a) Bohmer, V. Angew. Chem., Int. Ed. Engl. 1995, 34, 713;
(b) Cho, Y. L.; Rudkevich, D. M.; Shivanyuk, A.; Ris-
sanen, K.; Rebek, J., Jr. Chem. Eur. J. 2000, 6, 3788–3796;
(c) Arduini, A.; Secchi, A.; Pochini, A. J. Org. Chem.
2000, 24, 8127–8406.
3. Arduini, A.; McGregor, W. M.; Paganuzzi, D.; Pochini,
A.; Secchi, A.; Ugozzoli, F.; Ungaro, R. J. Chem. Soc.,
Perkin Trans. 2 1996, 839–846.
4. Beijer, F. H.; Sijbesma, R. P.; Vekemans, J. A. J. M.;
Meijer, E. W.; Kooijman, H.; Spek, A. L. J. Org. Chem.
1996, 61, 6371.
5. Vyas, N. K. Curr. Opin. Struct. Biol. 1991, 1, 732.
6. Ai-Sayah, M. H.; Branda, N. R. Angew. Chem., Int. Ed.
2000, 39, 945–947.
and mass spectrometry.8
The binding efficacy of the receptor 2 to a substrate was
compared with macrocycle 1, and evaluated by
analysing the change that occurred in the 1H NMR
spectra upon titrating solutions of cyanuric acid, Me-a-
O-
cycle
D
-glucopyranoside, biotin and thymine with macro-
2
in d6-DMSO. 1H NMR data of both
macrocycles 2 and 1 with their substrates are shown in
Figure 1, although no significant downfield shift for the
macrocycle 2 was observed for any substrates. In our
previous results,7b the macrocycle 1 exhibited promising
binding properties towards cyanuric acid, n-octyl-a-
D-
7. (a) Lowik, D. W. P. M.; Lowe, C. R. Tetrahedron Lett.
2000, 41, 1837–1840; (b) Lowik, D. W. P. M.; Lowe, C. R.
Eur. J. Org. Chem. 2001, 2825–2839.
glucopyranoside and n-octyl-b- -glucopyranoside.
D
Computer modelling showed that the successful binding
with cyanuric acid benefits from the relative flexibility
of macrocycle 1, where the methylene bridge is an
automatic adjuster, and changes its conformation to
create well-ordered hydrogen-bonding surfaces for the
melamine moiety, depending on the different substrates.
This self-adjusting is limited, nevertheless. Unlike
macrocycle 1, there is no methylene bridge adjuster in
macrocycle 2 derived from the xylenediamine linker,
resulting in distortion of the hydrogen-bonding surfaces
in the rigid macrocycle. The modelled structure is
shown in Figure 2(b), which suggests why the rigid
macrocycle 2 presents a poor binding ability.
8. Oligomers 8: Pd/C (10%, 350 mg) was added to a solution
of 7 (0.27 mmol) in MeOH (14 mL). After stirring the
solution under a hydrogen atmosphere for 3 h, the catalyst
was filtered off and the filtrate was evaporated. The
residue was dissolved in acetone–THF (1:1, 2 mL) and the
resulting solution was added to a fresh suspension of
cyanuric chloride (0.26 mmol) in acetone–H2O (1:2, 6 mL)
followed by the addition of NaHCO3 (0.27 mmol). After
stirring for 2 h at 0°C, water was added, the aqueous
suspension was extracted with DCM and the combined
organic layers were dried with MgSO4 and evaporated.
Flash column chromatography (eluent: 10% MeOH in
DCM) yielded 77% of product 8. 1H NMR (CDCl3; 400
MHz): 0.82 (m, 6H), 1.23 (m, 8H), 1.47 (m, 13H), 3.15 (m,
4H), 5.52 (b, 2H), 6.28 (m, 2H), 6.88 (m, 10H), 7.66 (m,
6H); 13C NMR (CDCl3; 100.5 MHz): 171.06, 170.44,
166.88, 165.39, 164.99, 153.67, 141.78, 141.45, 141.07,
137.79, 129.60, 129.38, 129.12, 118.00, 115.13, 113.21,
111.17, 79.92, 41.54, 29.96, 29.95, 28.62, 23.17, 14.37 MS-
ESI: m/z 896.5 [M+].
Acknowledgements
Authors would like to acknowledge Prometic Bio-
sciences Inc. and the BBSRC for financial support.
References
Macrocycle 9: A solution of 8 (57 mmol) in 2 M HCl in
dioxane (5 mL) was stirred for 3 h. The volatiles were
removed and the residue was coevaporated with THF
twice. The intermediate was dried overnight, then dis-
1. (a) Cram, D. J.; Cram, J. M. Container Molecules and
Their Guests; Royal Society of Chemistry: Cambridge,

1362
X. Yang, C. Lowe / Tetrahedron Letters 44 (2003) 1359–1362
Macrocycle 2: A solution of 9 (20 mmol) and pentylamine
solved in DMF (25 mL) and a solution of DIPEA (1.2
(0.20 mmol) in DMF (3 mL) was heated at 60°C
overnight. The solvent was removed in vacuo and the
residue was purified by column chromatography (eluent:
10% MeOH in DCM) followed by gel permeation chro-
matography (eluent: DCM/MeOH, 2:1) to yield 73% of
mmol) in DMF (5 mL) was added dropwise at 45°C, to
the resulting solution and stirring continued for 5 h at
45°C. The mixture was further purified by flash column
chromatography (eluent: 1% MeOH in DCM) to afford
62% of macrocycle 9. 1H NMR (d6-DMSO+D2O; 500
MHz): 0.81 (m, 3H), 0.87 (m, 3H), 1.30 (m, 8H), 1.52 (m,
4H), 3.29 (m, 4H), 6.79–7.16 (m, 12H); 13C NMR (d6-
DMSO; 100.5 MHz): 168.42, 165.62, 163.72, 162.46,
150.47, 141.00, 128.37, 113.22, 40.69, 29.08, 29.03, 28.93,
28.89, 22.17, 22.14, 14.21, 14.17; HRMS-ESI: m/z calcd
760.3576, found: 760.3560 [MH+].
1
product 2. H NMR (d6-DMSO+D2O; 500 MHz): 0.84 (m,
9H), 1.30 (m, 12H), 1.52 (m, 6H), 3.29 (m, 6H), 6.79–7.16
(m, 12H); 13C NMR (d6-DMSO; 100.5 MHz): 162.31,
160.87, 151.48, 139.20, 128.04, 124.91, 40.65, 29.07, 28.90,
28.77, 28.69, 28.55, 22.03, 21.99, 21.04, 14.09, 14.02, 13.91;
HRMS-ESI: m/z calcd 811.4857, found: 811.4858 [MH+].