Synthesis of the penta-oxazole core of telomestatin in a convergent approach to poly-oxazole macrocycles

Article abstract of DOI:10.1039/b613791c

The synthesis of the penta-oxazole core of telomestatin whose methyl derivative could be an advanced intermediate in its total synthesis by subsequent condensation with a suitably substituted oxazole was investigated. The biosynthesis of telomestatin was interpreted as a formal assembly of one cysteine, five serine, and two threonine sub-units. The successive use of amines bearing unprotected hydroxymethyl groups permitted iterative assembly of polyoxazoles without the need for repeated protection and deprotection. It also permitted regioselective introduction of substituents at the 5-position of one or more oxazole rings and a variety of analogues to probe telomerase function. The synthetic route also provided several linked polyoxazole systems of increasing complexity and derived from serine as the only amino acid.

Full text of DOI:10.1039/b613791c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis of the penta-oxazole core of telomestatin in a convergent approach
to poly-oxazole macrocycles
Charles M. Marson* and Mona Saadi
Received 21st September 2006, Accepted 21st September 2006
First published as an Advance Article on the web 2nd October 2006
DOI: 10.1039/b613791c
A protocol for the construction of poly-oxazoles with consec-
utive 2,4ꢀ-linkages is described, and has afforded an efficient
route to a penta-oxazole which demarcates a route to telomes-
tatin and related macrocyclic poly-oxazole systems.
appeared to offer a practical and flexible strategy. Accordingly, an
efficient route to di-serine building blocks (Scheme 1) was devel-
oped based upon the Williams–Wipf cyclisation–dehydrogenation
protocol using (dimethylamino)sulfur trifluoride (DAST) followed
by treatment with BrCCl3 and DBU.6 N-Cbz-L-Ser (prepared from
L-serine and benzyl chloroformate in aqueous 2 M NaOH) was
protected using dimethoxypropane in the presence of p-TsOH as
catalyst to give acid 4 (57% over two steps) which, as its mixed
anhydride, was reacted with the hydrochloride salt of L-Ser-OMe
to give the di-serine derivative 5. This was cyclised using DAST at
−78 ◦C to give the oxazoline 6, which with BrCCl3–DBU afforded
the oxazole 7.
Telomestatin (1), the most potent inhibitor of telomerase function
known (IC50 = 5 nM), was recently isolated from Streptomyces
anulatus 3533-SV4.1 Telomestatin is also a specific inhibitor of
telomerase function, and does not inhibit DNA polymerases
or reverse transcriptases such as Taq polymerase or HIV-RT.1
The mode of inhibition is considered to involve stabilisation of
human DNA G-quadruplex structures or facilitation of their
formation.2 The synthesis of telomestatin requires the assembly of
seven oxazole rings, each with a consecutive 2,4ꢀ-linkage. Whereas
both linear3 and convergent4 routes to such tris-oxazoles have
been reported, the synthesis of poly-oxazoles, especially with
terminal substituents appropriate to natural product synthesis,
remains a challenge.5 Here, we report the synthesis of the
penta-oxazole core 2a of telomestatin whose methyl derivative
2b could be an advanced intermediate in its total synthesis,
by subsequent condensation with a suitably substituted oxazole
such as 3.
Scheme 1 Reagents and conditions: i, t-BuOCOCl, Et3N, CH2Cl2; ii,
L-serine methyl ester hydrochloride, −30 ◦C, 2.5 h, 95% over two steps; iii,
DAST, CH2Cl2, −78 ◦C, 2.5 h, 85%; iv, BrCCl3, DBU, CH2Cl2, −10 ◦C to
20 ◦C, 17 h, 92%.
Oxazole 7 was the common precursor of the di-serine building
blocks 8 and 9 (Scheme 2), which in a convergent strategy
afforded the advanced intermediates 11, 12 and 13 derived from
four serine sub-units. Acid-catalysed hydrolysis of the N,O-acetal
of 7 followed by hydrogenolysis of the Cbz group afforded
the amino alcohol 8. Alkaline hydrolysis of ester 7 afforded
the acid 9, which acylated 8 giving 10 without difficulty when
performed at −62 ◦C in DMF using benzyltriazol-1-yloxy-
tris(dimethylamino)phosphonium hexafluorophosphate (BOP) in
the presence of (i-Pr)2NEt and 1-hydroxybenzotriazole (HOBt).
Cyclisation of amide 10 with DAST at −78 ◦C afforded oxazoline
11, which underwent dehydrogenation with BrCCl3–DBU to give
the oxazole 12. Alkaline hydrolysis of oxazole 12 afforded the
acid 13, which could also be used to acylate the amino group
The biosynthesis of telomestatin (1), although not yet elu-
cidated, can be interpreted as a formal assembly of one cys-
teine, five serine and two threonine sub-units. From a synthetic
viewpoint, condensation of pairs of amino acid residues to give
2,4-disubstituted oxazoles such as 7, followed by convergent
condensation to give tetra-serine building blocks such as 12,
Department of Chemistry, University College London, Christopher Ingold
Laboratories, 20 Gordon Street, London, WC1H 0AJ, UK. E-mail:
c.m.marson@ucl.ac.uk; Fax: +44 (0)207 679 7463; Tel: +44 (0)207 679
4712
◦
of 8 using the BOP procedure, at −62 C for this condensation.
3892 | Org. Biomol. Chem., 2006, 4, 3892–3893
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Scheme 2 Reagents and conditions: i, MeOH, p-TsOH, reflux 2.5 h; ii, H2, 10% Pd/C, 92% over two steps; iii, LiOH, THF–H2O (8 : 1), 50 ◦C, 17 h, 83%;
iv, BOP, (i-Pr)2NEt, HOBt, DMF, −30 ◦C, 77%; v, DAST, CH2Cl2, −78 ◦C, 2.5 h, 79%; vi, BrCCl3 (2 equiv.), DBU (4 equiv.), CH2Cl2, −10 ◦C to 20 ◦C,
17 h, 72%; vii, LiOH, THF–H2O (8 : 1), 60 ◦C, 17 h, 95%; viii, BOP, (i-Pr)2NEt, HOBt, DMF, −62 ◦C, 40%.
Scheme 3 Reagents and conditions: i, DAST, CH2Cl2, −78 ◦C, 2.5 h, 80%; ii, BrCCl3 (22 equiv.), DBU (40 equiv.), CH2Cl2, −10 ◦C to 20 ◦C, 17 h, 68%.
The successive use of amines bearing unprotected hydroxymethyl
groups is a notable feature of the strategy, and permits iterative
assembly of polyoxazoles without the need for repeated protection
and deprotection.
Notes and references
1 K. Shin-ya, K. Wierzba, K. Matsuo, T. Ohtani, Y. Yamada, K. Furihata,
Y. Hayakawa and H. Seto, J. Am. Chem. Soc., 2001, 123, 1262–1263.
2 M.-Y. Kim, H. Vankayalapati, K. Shin-ya, K. Wierzba and L. H. Hurley,
J. Am. Chem. Soc., 2002, 124, 2098–2099.
3 (a) P. Liu, C. A. Celatka and J. S. Panek, Tetrahedron Lett., 1997, 38,
5445–5448; (b) C. A. Celatka and J. S. Panek, Tetrahedron Lett., 2002,
43, 7043–7046.
A further round of acylation–cyclisation–dehydrogenation was
successful: coupling acid 13 with amine 8 using the BOP procedure
◦
at −62 C gave the amide 14, which cyclised with DAST to give
the oxazoline 15 (Scheme 3). Again, the internal oxazoline ring
could be dehydrogenated, and thus gave the penta-oxazole 2a.7
In addition to demarcating an approach to telomestatin (1), the
penta-oxazole 2a could also be reacted with oxazole 8, using
the protocol outlined, to furnish the C8-symmetric octa-oxazole
analogue 16, which would provide a valuable comparison with the
natural product 1, especially in terms of their relative biological
effects on telomerase. The strategy permits regioselective intro-
duction of substituents at the 5-position of one or more oxazole
rings (as required), and thus a variety of analogues with which
to probe telomerase function. The synthetic route also provides
several linked polyoxazole systems of increasing complexity, but
derived from serine as the only amino acid.
4 (a) J. Deeley and G. Pattenden, Chem. Commun., 2005, 797–799; (b) S. K.
Chattopadhay and G. Pattenden, Synlett, 1997, 1342–1344.
5 For a synthesis of a 5-phenyltetra-oxazole via an isonitrile anion addition
to glyoxylic acid, see:J. M. Atkins and E. Vedejs, Org. Lett., 2005, 7,
3351–3354.
6 (a) D. R. Williams, D. A. Brooks and M. A. Berliner, J. Am. Chem. Soc.,
1999, 121, 4924–4925; (b) A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno and
D. R. Williams, Org. Lett., 2000, 2, 1165–1168.
7 2a: dH (DMSO-d6, 313 K, 400 MHz) 8.99 (1H, s), 8.97 (1H, s), 8.90 (1H,
s), 8.85 (1H, s), 8.78 (1H, s), 7.24 (5H, m), 5.30 (1H, dd, J = 6.5 and
2.5 Hz, CHCH2), 5.14 (1H, d, J = 12.7 Hz, CHHPh), 5.02 (1H, d, J =
12.7 Hz, CHHPh), 4.34 (1H, dd, J = 9.3 and 6.5 Hz, CHHCH), 4.16
(1H, dd, J = 9.3 and 2.7 Hz, CHHCH), 3.87 (3H, s, OCH3), 1.70 (3H, s,
CH3), 1.57 (3H, s, CH3) ppm; dC (DMSO-d6, 353 K, 100 MHz) 163.4
(s), 160.1 (s), 155.1 (s), 155.0 (s), 154.9 (s), 154.3 (s), 151.0 (s), 144.3
(d), 140.1 (d), 140.05 (×2, d), 140.0 (d), 135.7 (s), 133.1 (s), 129.7 (s),
129.6 (s), 129.5 (s), 129.7 (s), 127.5 (d), 127.0 (d), 126.7 (d), 94.1 (NCO),
66.4 (CH2OCO), 65.8 (OCH2CH), 54.0 (OCH2CH), 50.9 (OCH3), 24.9
(CH3), 23.6 (CH3) ppm; m/z found: 651.1462; C30H24N6O10 (M + Na)+
requires 651.1452.
Support from the EPSRC for a studentship (to MS) under the
DTA initiative is gratefully acknowledged. We thank Dr Abil Aliev
for assistance with NMR spectroscopy.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3892–3893 | 3893
©