Synthesis of Arabino glycosyl triazoles as potential inhibitors of mycobacterial cell wall biosynthesis

Article abstract of DOI:10.1016/j.bmcl.2008.09.082

A series of arabino glycosyl triazoles with varying hydrophobic groups were synthesised as putative mimics of decaprenolphosphoarabinose (DPA) as potential inhibitors of mycobacterial cell wall biosynthesis. Biological testing against Mycobacterium bovis BCG revealed low to moderate anti-mycobacterial activity, with strong dependence on the identity of the hydrophobic side chain.

Full text of DOI:10.1016/j.bmcl.2008.09.082

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6265–6267
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of Arabino glycosyl triazoles as potential inhibitors
of mycobacterial cell wall biosynthesis
Brendan L. Wilkinson a, Hilary Long b, Edith Sim b, Antony J. Fairbanks a,
*
a Department of Chemistry Research Laboratory, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK
b Department of Pharmacology, Oxford University, Mansfield Road, Oxford OX1 3QT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of arabino glycosyl triazoles with varying hydrophobic groups were synthesised as putative mim-
ics of decaprenolphosphoarabinose (DPA) as potential inhibitors of mycobacterial cell wall biosynthesis.
Biological testing against Mycobacterium bovis BCG revealed low to moderate anti-mycobacterial activity,
with strong dependence on the identity of the hydrophobic side chain.
Received 3 September 2008
Revised 22 September 2008
Accepted 23 September 2008
Available online 26 September 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Carbohydrates
Glycosyl triazoles
Click chemistry
Arabinose
Mycobacteria
Tuberculosis
Inhibition of mycobacterial cell wall biosynthesis represents a
significant therapeutic opportunity for the development of new
anti-tubercular agents.1 The cell walls of mycobacteria are com-
plex, and there are therefore many biosynthetic steps that could
in principle be targeted in order to compromise mycobacterial via-
bility. Two cell wall polysaccharides that are unique to mycobacte-
ria and which are crucial to mycobacterial survival and growth are
lipoarabinomannan (LAM) and arabinogalactan (AG). Inhibition of
the biosynthesis of either LAM or AG represents a potentially
highly selective opportunity for therapeutic intervention, and the
design and synthesis of such potential inhibitors has been a field
of intense interest over recent years. Approaches have included2
attempted inhibition of various glycosyl transferases3 used for
polysaccharide assembly, and also of other vital biosynthetic en-
zymes such as the Galp/Galf mutase which catalyses a crucial pyra-
nose/furanose isomerisation during assembly of the galactan cell
wall component.4
O
O P
O
O
O
HO
HO
OH
9
1
Figure 1. Decaprenolphosphoarabinose 1 (DPA).
der to design mimics of DPA that may be expected to display useful
in vivo activity, two immediate considerations present themselves.
Firstly replacement of the labile glycosyl phosphate with a stable
isostere would be highly desirable. Secondly it may be expected
that an alkyl chain of a certain minimum length would be required
to mimic the large hydrophobic decaprenyl moiety of DPA. Various
isosteric replacements for the glycosyl phosphate have been inves-
tigated6 including b-C-glycosyl sulfones and b-C-phosphonates.
Amongst several approaches7 to inhibition of mycobacterial arab-
inan biosynthesis under investigation in this laboratory the poten-
tial use of a b-glycosyl triazole as the replacement for the glycosyl
phosphate moiety of DPA appeared to be an attractive line of inves-
tigation.8 The discovery that Cu(I) salts can efficiently catalyse the
Huisgen 1,3-dipolar cycloaddition of alkynes and azides, reported
simultaneously9 by Meldal and Sharpless, has led to a resurgence
in interest in the triazole functional group, and has popularised
its use as an isosteric replacement in lead generation programs. In-
deed this particular application of ‘Click chemistry’10 has become
one of the mainstays of currently favoured approaches for the
Particular interest has focused on the attempted discovery5 of
potent inhibitors of arabinosyl transferases, which are responsible
for the stepwise assembly of the arabinan portions of both LAM
and AG, and use decaprenol phosphoarabinose 1 (DPA, Fig. 1) as
the glycosyl donor. Metabolically stable analogues of DPA may in-
hibit these arabinosyl transferases, and so in turn represent lead
compounds for the discovery of new anti-tubercular agents. In or-
* Corresponding author. Tel.: +44 1865 275647; fax: +44 1865 275674.
E-mail address: antony.fairbanks@chem.ox.ac.uk (A.J. Fairbanks).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.09.082

6266
B. L. Wilkinson et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6265–6267
introduction of structural diversity in medicinal chemistry pro-
grams. The synthesis of glycosyl triazoles derived from anomeric
azides has been extensively investigated, and glycosyl triazoles11
have been demonstrated to posses a wide range of interesting bio-
logical activity. Pertinently to the anti-mycobacterial field of inves-
tigation other carbohydrate-derived triazoles in which the triazole
moiety is not linked to the anomeric position have also generated
interest, and a series of arabino and ribo-configured pentose tria-
zoles in which azide was introduced at the 5-position before cyclo-
series of spot tests was performed on E. coli (strain JM109) in which
none of the compounds 6a–f displayed any anti-bacterial effects at
the highest concentrations (1 mg/mL), indicating a selective anti-
mycobacterial mode of action. The minimum inhibitory concentra-
tions (MIC) of the anti-mycobacterial activity of compounds 6a–e
were then measured in a second series of tests on M. bovis BCG
using the Alamar Blue microplate assay, in which the Alamar Blue
dye changes from the oxidised indigo blue (and non-fluorescent)
form to the reduced pink (and fluorescent) form in the presence
of growing bacteria (Fig. 2).20 The results are shown in Table 1
and Figure 2.
addition with
a series of alkynes have very recently been
synthesised as potential anti-tubercular agents.12
The synthesis of a series b-arabinofuranose glycosyl triazoles
was therefore undertaken as putative mimics of DPA; herein a vari-
ety of hydrophobic groups of varying lengths were installed as
mimics of the extended polyprenyl side chain in order to delineate
structure activity relationships. Anti-bacterial activity of the ensu-
ing compounds was tested in a variety of cell growth assays against
both Escherichia coli and Mycobacterium bovis BCG.
Several points are worthy of note, though general conclusions
are somewhat harder to arrive at. First of all the complete lack of
activity of the phenyl triazole 6f, and the low activity of the hexyl
triazole 6a indicate that hydrophobicity of the side chain is impor-
The known benzoylated arabino a
-glycosyl bromide 213 was re-
acted with sodium azide under phase transfer conditions to give
the corresponding glycosyl azide (64% overall yield) as an anomeric
mixture in which the desired b-anomer 314 predominated (
a:b ra-
tio ꢀ1:2), and from which pure b-anomer was easily obtained after
purification by flash column chromatography (isolated yield of 3,
43%). Modified Cu (I) catalysed Huigsen cycloaddition15 of azide
3 with alkyne 4a in which Cu(I) was generated in situ by the addi-
16
tion of sodium ascorbate to a solution of CuSO4 proved sluggish,
and therefore an alternative procedure was investigated. Reaction
of azide 3 with alkyne 4a and copper (I) iodide in toluene, in the
presence of diisopropylethylamine (DIPEA) smoothly gave the de-
sired triazole 5a. Similar reaction of the benzoate protected azide 3
with the remainder of the alkynes 4b–f17 gave the corresponding
b-glycosyl triazoles 5a–f18 in excellent yield. Finally de-protection
with sodium methoxide in methanol yielded the triol triazoles 6a–
f for biological screening (Scheme 1).
Figure 2. Alamar Blue assay of deprotected glycosyl triazoles 6a–e and control with
isoniazid (INH).
Table 1
Inhibitory effects of deprotected glycosyl triazoles 6a–f against M. bovis BCG
Compound
R0
MICa
(lg/mL)
Anti-mycobacterial testing of compounds 6a–f was initially per-
formed by a spot-test method using M. bovis BCG as a model for M.
tuberculosis. M. bovis BCG cultures were spotted onto 6 well plates
containing solid media and the test compounds at various concen-
trations. The cultures were then grown in an incubator at 37 °C for
7–14 days. The effect of test compounds on cell growth was then
assessed by measuring the area of the circular mycobacterial spot
cultures after photography using a Bio-Rad Gel-Doc 2000 system, a
reduction in this value indicating anti-mycobacterial effects.19
Whilst compound 6f did not display any anti-mycobacterial effects
at the highest concentration investigated (1 mg/mL) all of the other
triazoles showed effects on mycobacterial cell growth. A second
INH
6a
—
0.1
500
62
–(CH2)5CH3
–(CH2)7CH3
–(CH2)9CH3
–CH2O(CH2)13CH3
6b
6c
250
31
6d
6e
62
O
6f
–Ph
No activity
a
MIC, minimum inhibitory concentration; the lowest concentration of the
compound which inhibited the growth of M. bovis BCG >90% from the Alamar Blue
assay. Isoniazid (INH) was used as a control (MIC 0.1 lg/mL).
R2
O
O
O
Br
(i)
N3
(ii)
N
N
R1O
BzO
BzO
N
BzO
BzO
OBz
R1O
OR1
OBz
5a-f (R1 = Bz)
6a-f (R1 = H)
2
3
(iii)
4a
4c
4b
O
O
4e
4d
4f
Scheme 1. Reagents and conditions: (i) NaN3 (5.0 equiv), Bu4NHSO4 (1.0 equiv), 1:1 DCM-NaHCO3 (aq), 43%; (ii) 4a–f (5–10 equiv), CuI (1.0 equiv), DIPEA (1.0 equiv), 110 °C,
72–92%; (iii) NaOCH3 (0.4 equiv), CH3OH-THF (2:1 v/v), 92–100%.

B. L. Wilkinson et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6265–6267
6267
tant for biological activity.21 These results agree with the lack of
activity previously reported for arabino-C-glycosyl phenyl sulfones,
3a but the fact that the octyl triazole 6b is more active that the cor-
responding decyl triazole 6c indicates that there is not a simple
correlation between biological activity and the length of an alkyl
side chain; an observation in agreement with a previous report
by Lowary6b on a study of arabino-C-glycosyl sulfones. It is also
notable in the current study that the most potent compound, ether
6d which contains a tetradecyl side chain, is approximately twice
as active as the corresponding farnesyl ether 6e, though the latter
more closely resembles the natural substrate DPA. The significance
of these results remains unclear, particularly in view of the modest
anti-mycobacterial activity observed herein; even the most potent
compound 6d is at least two orders of magnitude less active than
isoniazid (INH). The synthesis of more derivatives will be necessary
in order to access more potent compounds and to further delineate
precise structure–activity relationships.
6. (a) Centrone, C. A.; Lowary, T. L. J. Org. Chem. 2002, 67, 8862; (b) Centrone, C. A.;
Lowary, T. L. Bioorg. Med. Chem. 2004, 12, 5495.
7. Smellie, I. A.; Bhakta, S.; Sim, E.; Fairbanks, A. J. Org. Biomol. Chem. 2007, 5,
2257.
8. A referee queried the choice of triazole as replacement for phosphate, and we
agree that triazole is undoubtedly a better isostere of an amide. However
besides simply acting as a spacer of relevant atomic proportions, triazole,
although not negatively charged like phosphate, can either act as a H-bond
acceptor, or bind to a metal centre. For a previous report on the attempted use
of triazole as a bioisosteric replacement for phosphate for the development of
anti-tubercular compounds see: Somu, R. V.; Boshoff, H.; Qiao, C.; Bennett, E.
M.; Barrry, C. E.; Aldrich, C. C. J. Med. Chem. 2006, 49, 31.
9. (a) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057; (b)
Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int. Ed.
2002, 41, 2596.
10. Bock, V. D.; Hiemstra, H.; van Maarseveen, J. H. Eur. J. Org. Chem. 2006, 51.
11. (a) El Akri, K.; Bougrin, K.; Balzarini, J.; Faraj, A.; Benhida, R. Bioorg. Med. Chem.
Lett. 2007, 17, 6656; (b) Rossi, L. L.; Basu, A. Bioorg. Med. Chem. Lett. 2005, 15,
3596; (c) Wilkinson, B. L.; Innocenti, A.; Vullo, D.; Supuran, C. T.; Poulsen, S.-A.
J. Med. Chem. 2008, 51, 1945; (d) Wilkinson, B. L.; Bornaghi, L. F.; Houston, T. A.;
Innocenti, A.; Supuran, C. T.; Poulsen, S.-A. J. Med. Chem. 2006, 49, 6539; (e)
Wilkinson, B. L.; Bornaghi, L. F.; Houston, T. A.; Innocenti, A.; Vullo, D.; Supuran,
C. T.; Poulsen, S.-A. Bioorg. Med. Chem. Lett. 2007, 17, 987; De las (f) Heras, F. G.;
Alonso, R.; Alonso, G. J. Med. Chem. 1979, 22, 496; (g) De las Heras, F. G.;
Camarasa, M. J. Nucleosides Nucleotides 1982, 1, 45; (h) De las Heras, F. G.;
Sanchez-Perez, R. M.; Aguado, M. L. Eur. J. Med. Chem. 1981, 16, 339.
12. Singh, B. K.; Yadav, A. K.; Kumar, B.; Gaikwad, A.; Sinha, S. K.; Chaturvedi, V.;
Tripathia, R. P. Carbohydr. Res. 2008, 343, 1153.
In conclusion the synthesis of a variety of b-arabino glycosyl tri-
azoles possessing various hydrophobic chains was achieved using
modified Click chemistry. These glycosyl triazoles displayed weak
to moderate yet selective anti-mycobacterial activity in assays
against M. bovis BCG; activity was strongly dependent on the nat-
ure of the hydrophobic group.
13. Lee, H. C.; Kumar, P.; Wiebe, L. I.; McDonald, R.; Mercer, J. R.; Ohkura, K.; Seki,
K.-I. Nucleosides Nucleotides Nucleic Acids 1999, 18,
14. Stimac, A.; Kobe, J. Carbohydr. Res. 2000, 329, 317.
1995.
15. Typical experimental procedure: to a solution of the glycosyl azide 3 (150 mg,
0.31 mmol) in dry toluene (3 mL) was successively added
tetradecanolpropargyl ether 4d (390 mg, 1.51 mmol, 5.0 equiv), CuI (59 mg,
0.31 mmol, 1.0 equiv) and N,N-diisopropylethylamine (51 L, 0.31 mmol,
Acknowledgement
l
We gratefully acknowledge financial support from the EPSRC
(Project Grant D051495/1).
1.0 equiv) under argon. The heterogeneous mixture was heated to gentle
reflux (ca. 110 °C) for 12 h at which time t.l.c. analysis (petrol/ethyl acetate,
7:3) indicated the consumption of starting material (Rf 0.5) and formation of
product (Rf 0.4). The mixture filtered through CeliteÒ and eluted with ethyl
acetate. The filtrate was concentrated in vacuo and the residue purified by
gradient flash column chromatography (100% DCM ? DCM/ethyl acetate, 9:1)
to afford 5d as a pale yellow oil, which slowly crystallized at rt to form a pale
yellow solid (178 mg, 84%).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.09.082.
16. Wilkinson, B. L.; Bornaghi, L. F.; Poulsen, S.-A.; Houston, T. A. Tetrahedron 2006,
62, 8115.
References and notes
17. The farnesyl propargyl ether 4e was made by modification of the
previously reported procedure by alkylation of farnesol with propargyl
bromide and sodium hydride in DMF. For the original procedure see:
Schmialek, P.; Geyer, A.; Miosga, V.; Nuendel, M.; Zapf, B. Insect Biochem.
1976, 6, 19.
1. Lowary, T. L. Mini Rev. Med. Chem. 2003, 3, 689.
2. Owen, D. J.; Davis, C. B.; Hartnell, R. D.; Madge, P. D.; Thomson, R. J.; Chong, A. K.
J.; Coppel, R. L.; von Itzstein, M. Bioorg. Med. Chem. Lett. 2007, 17, 2274.
3. For some leading references see: (a) Maddy, J. A.; Bansal, N.; Bermudez, L. E.;
Comber, R. N.; Orme, I. M.; Suling, W. J.; Wilson, L. N.; Reynolds, R. C. Bioorg.
Med. Chem. Lett. 1998, 8, 237; (b) Pathak, A. K.; Pathak, V.; Madry, J. A.; Suling,
W. J.; Gurcha, S. S.; Besra, G. S.; Reynolds, R. C. Bioorg. Med. Chem. Lett. 2002, 10,
923; (c) Pathak, A. K.; Pathak, V.; Madry, J. A.; Suling, W. J.; Gurcha, S. S.; Besra,
G. S.; Reynolds, R. C. Bioorg. Med. Chem. Lett. 2002, 10, 3145; (d) Chaumontet,
M.; Pons, V.; Marotte, K.; Prandi, J. Tetrahedron Lett. 2006, 47, 1113.
4. For some leading references see: (a) Cren, S.; Gurcha, S. S.; Blake, A. J.; Besra, G.
S.; Thomas, N. R. Org. Biomol. Chem. 2004, 2, 2418; (b) Pathak, A. K.; Pathak, V.;
Kulshrestha, M.; Kinnaird, D.; Suling, W. J.; Gurcha, S. S.; Besra, G. S.; Reynolds,
R. C. Tetrahedron 2003, 59, 10239; (c) Wen, X.; Crick, D. C.; Brennan, P. J.; Hultin,
P. G. Bioorg. Med. Chem. Lett. 2003, 11, 3579; (d) Burton, A.; Wyatt, P.; Boons, G.-
J. J. Chem. Soc., Perkin Trans. 1 1997, 2375.
5. See for example: (a) Marotte, K.; Ayad, T.; Genisson, Y.; Besra, G. S.; Baltas, M.;
Prandi, J. Eur. J. Org. Chem. 2003, 2557; (b) Pathak, A. K.; Pathak, V.; Maddry, J.
A.; Suling, W. J.; Gurcha, S. S.; Besra, G. S.; Reynolds, R. Bioorg. Med. Chem. 2001,
9, 3145; (c) Joe, M.; Lowary, T. L. Carbohydr. Res. 2006, 341, 2723; (d) Cociorva,
O. M.; Gurcha, S. S.; Besra, G. S.; Lowary, T. L. Bioorg. Med. Chem. 2005, 13, 1369;
(e) Cociorva, O. M.; Lowary, T. L. Carbohydr. Res. 2004, 339, 853; (f) Bosco, M.;
Bisseret, P.; Constant, P.; Eustache, J. Tetrahedron Lett. 2007, 48, 153.
18. See Supporting Information for full experimental details and complete
characterisation of all new compounds. Selected data: 6d,
solid; mp 67–68 °C; [a] D –19.6 (c 0.65, CH3OH); dH (400 MHz, CDCl3), 0.9 (t, J
a
pale yellow
25
6.8 Hz, CH3), 1.24–1.36 (m, 24H, 12ꢁ CH2), 1.54–1.61 (2H, m, CH2), 3.50 (t, J
6.5 Hz, OCH2), 3.81 (1H, dd, J4-5 5.1 Hz, J5-5 12.1 Hz, H-50), 3.86 (1H, dd, J4-5
0
0
3.4 Hz, H-500), 3.94–3.98 (1H, m, H-40), 4.28 (1H, at, J 5.5 Hz, H-20), 4.34 (1H, at, J
5.5 Hz, H-30), 4.58 (s, 2H, C@CCH2), 6.32 (1H, d, J1-2 5.5 Hz, H-10), 8.15 (1H, s,
C@CH); dC (125 MHz, CD3OD), 14.5 (q, CH3), 23.8 (t, CH2), 27.3 (t, CH2), 30.5 (t,
CH2), 30.6 (t, CH2), 30.7 (t, CH2), 30.8 (2ꢁ t, 2ꢁ CH2), 30.8 (2ꢁ t, 2ꢁ CH2), 30.8
(2ꢁ t, 2ꢁ CH2), 30.9 (t, CH2), 33.1 (t, CH2), 62.6 (t, C-50), 64.7 (t, OCH2), 71.7
(C@CCH2), 75.9 (d, C-20), 78.8 (d, C-30), 85.9 (d, C-10), 124.9 (C@CH), 145.6
(C@CH); HRMS (ES+) Calcd. for C22H41N3O5 (MNa+) 450.2938. Found 450.2938.
19. (a) Anderton, M. C.; Bhakta, S.; Besra, G. S.; Jeavons, P.; Eltis, L. D.; Sim, E. Mol.
Microbiol. 2006, 59, 181; (b) Madikane, V. E.; Bhakta, S.; Russell, A. J.; Campbell,
W. E.; Claridge, T. D.; Elisha, B. G.; Davies, S. G.; Smith, P.; Sim, E. Bioorg. Med.
Chem. 2007, 15, 3579.
20. Collins, L. A.; Franzblau, S. G. Antimicrob. Agents Chemother. 1997, 41, 1004.
21. The importance of lipophilic chain length for recognition of donor
analogues of DPA by arabinosyltransferases has been reported previously:
see Lee, R. E.; Brennan, P. J.; Besra, G. S. Bioorg. Med. Chem. Lett. 1998, 8,
951.