The Antimicrobial Natural Product Chuangxinmycin and Some
Synthetic Analogues are Potent and Selective Inhibitors of
Bacterial Tryptophanyl tRNA Synthetase
Murray J. Brown, Paul S. Carter, Ashley E. Fenwick, Andrew P. Fosberry,
Dieter W. Hamprecht, Martin J. Hibbs, Richard L. Jarvest,* Lucy Mensah,
Peter H. Milner, Peter J. O’Hanlon, Andrew J. Pope, Christine M. Richardson,
Andrew West and David R. Witty
GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
Received 29 May 2002; accepted 23 July 2002
Abstract—The antimicrobial natural product chuangxinmycin has been found to be a potent and selective inhibitor of bacterial
tryptophanyl tRNA synthetase (WRS). A number of analogues have been synthesised. The interaction with WRS appears to
be highly constrained, as only sterically smaller analogues afforded significant inhibition. The only analogue to show inhibition
comparable to chuangxinmycin also had antibacterial activity. WRS inhibition may contribute to the antibacterial action of
# 2002 Elsevier Science Ltd. All rights reserved.
Chuangxinmycin (1) is a natural product first isolated
from Actinoplanes tsinanensis.1,2The compound was
reported to have in vitro antibacterial activity against a
number of Gram-positive and Gram-negative bacteria
and to show in vivo efficacy in mouse infection models
against Escherichia coli and Shigella dysenteriae.1Pre-
liminary clinical results indicated that chuangxinmycin
was effective in the treatment of septicaemia, urinary
and biliary infections caused by E. coli.1Chuangxin-
mycin is reported to interact with the tryptophan bio-
synthetic pathway, but the molecular target for its
antibacterial activity has not been described.3
The aminoacyl tRNA synthetases are a family of
enzymes that catalyse attachment of amino acids to
their cognate tRNA in protein biosynthesis. The activity
and fidelity of these enzymes are essential to living cells.
Inhibition of bacterial isoleucyl tRNA synthetase is the
mode of action of the antibacterial agent mupirocin
(marketed as Bactroban1) and selective inhibitors of
other tRNA synthetases are of interest as they could
afford new antibacterials with a novel mode of action.
natural product inhibitors of tRNA synthetase frequently
contain a structural fragment that is readily recognisable
Chuangxinmycin has some structural similarity to tryp-
tophan (2), the substrate of tryptophanyl tRNA syn-
thetase (WRS). It also
resemblance to the known WRS inhibitor indolmycin
(3). We hypothesised that chuangxinmycin might be
exerting its antibacterial activity via inhibition of WRS.
In this report, we describe studies of the inhibition of
WRS by chuangxinmycin, as well as the synthesis and
testing of some analogues designed to elucidate the role
of the dihydrothiopyran carboxylic acid in binding.
Chuangxinmycin was found to potently inhibit the
aminoacylation of tRNA by purified Staphylococcus
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
Bioorganic & Medicinal Chemistry Letters 12 (2002) 3171–3174
*Corresponding author. Fax: +44-1279-622260; e-mail: richard_l_
aureus WRS,7,8and in a pyrophosphate exchange assay9
at varying concentrations of the amino acid substrate,
chuangxinmycin was found to be competitive with
respect to tryptophan with a Kiof 20 nM. Chuangxin-
mycin was highly selective for the bacterial enzyme,
with no inhibition of ovine WRS observed at con-
centrations up to 30 mM.
To further investigate the interactions of chuangxin-
mycin with bacterial WRS, we decided to synthesise
some key analogues of the inhibitor. We concentrated
on modification of the substituents on the characteristic
dihydrothiopyran ring of chuangxinmycin. A number of
derivatives of chuangxinmycin, some of which are pro-
ducts of modification of the carboxylic acid group, have
been reported in the Chinese literature.10
We decided to use synthetically derived (?)-chuangxin-
mycin methyl ester (7) as the starting point for the pre-
paration of analogues. This material was prepared using
a modification of the route reported by Matsumoto and
Watanabe,11starting from 3-acetylindole (Scheme 1).
The primary amide (8) was formed by treatment of the
methyl ester with methanolic ammonia, although the
reaction was very slow. In general, mixtures containing
similar amounts of the respective cis and trans deriva-
tives were obtained.12
The ester 7 and the primary amide 8 were tested for
inhibition against a crude preparation of wild type WRS
from S. aureus Oxford. Neither compound showed sig-
nificant inhibition up to 20 mM.
The ester was hydrolysed to the free acid 10 with
sodium hydroxide (Scheme 2). The hydroxamic acid 11
was formed from the acid by carbodiimide-HOAt
mediated coupling with hydroxylamine. The aldehyde 9
was prepared from the ester 7 by first protecting the
indole nitrogen with a triisopropylsilyl group, followed
by DIBAL-H reduction and TBAF deprotection.
Compounds 9 and 11 were tested in a standard amino-
acylation assay against purified recombinant S. aureus
WRS in which chuangxinmycin gave an IC50value of
about 30 nM. The hydroxamic acid gave almost no
inhibition up to 10 mM but the aldehyde was a potent
inhibitor with an IC50of 90 nM. It was shown that the
aldehyde was stable in the DMSO stock solution used
for the assay and was not oxidised to chuangxinmycin
itself. The possibility exists that the aldehyde could form
a covalent adduct, for example by Schiff base formation
with an enzymatic lysine residue. However, following
incubation with sodium borohydride13no evidence for a
covalent adduct was obtained by LC–MS.
Dehydration of the primary amide to the nitrile resulted
in a 1:7 ratio of cis-12: trans-12 (Scheme 3). The ratio
could be changed to a 2.1:1 ratio in favour of the
desired cis isomer by epimerisation with catalytic DBU.
Cycloaddition of azide to the nitrile afforded the tetra-
zole analogue 13.
In order to investigate functionalisation of the methyl
group, we adapted a diastereo- and enantioselective
synthesis of chuangxinmycin described by Akita and co-
workers14in order to include a silyl protected hydroxy-
methyl in place of the methyl group (Scheme 4).
Chiral epoxide 15, available from 14 in five steps,15,16
alkylates 4-iodoindole in the presence of Lewis acid to
give 16 in virtually identical yield to that reported for
the parent compound.14Conversion of the a-hydroxy
group in 16 to thiol 17 was assumed to proceed with
retention of configuration by analogy to the literature
precedent.14A significant amount of disulfide 18 was
formed as a by-product of the reaction, however this
was smoothly converted to 17 by heating with dithio-
threitol in triethylamine. The synthesis was completed
by palladium-mediated cyclisation followed by desilyla-
tion with concomitant lactonisation, to afford lactone
20.17This was converted to the hydroxymethyl analo-
gue 21 by treatment with lithium hydroxide. Compound
Scheme 1. Reagents and conditions: (i) Tl(CF3CO2)3/CuI/I2, quant;
(ii) Me3SnSCH2CO2Me, quant; (iii) TsNHNH2, 97%; (iv) NaH then
?, 44%; (v) NH3/MeOH, 21 d, 50%.
Scheme 2. Reagents and conditions: (i) LiHMDS/THF/?78?C then
TIPS-Cl, 82%; (ii) DIBAL-H/toluene/DCM(4:1), 84%; (iii) TBAF/
THF, 73%; (iv) NaOH/H2O/EtOH, quant; (v) NH2OH.HCl/EDCI/
Scheme 3. Reagents and conditions: (i) POCl3/pyridine/0–10?C, 61%,
then DBU (0.2 equiv)/DMSO, quant; (ii) NaN3/Et3N.HCl/DMF/
3172M. J. Brown et al./Bioorg. Med. Chem. Lett. 12 (2002) 3171–3174
21 was isolated as virtually exclusively the cis isomer,
the unusually high stereoselectivity probably being due
to intramolecular hydrogen bonding effects.
Compounds 12, 13, 20 and 21 were tested for inhibition
in a scintillation proximity assay for WRS activity.18In
this assay format chuangxinmycin gave a higher IC50
value of 90–200 nM. The tetrazole analogue 13 and the
hydroxymethyl derivative 21 afforded no inhibition up
to the highest concentration tested, 10 mM. However,
significant inhibition was shown by the nitrile 12 (IC50
1100 nM) and by the lactone 20, which had an IC50
value of 230 nM, comparable to that of chuangxinmy-
It is noteworthy that the only active analogues of
chuangxinmycin, the aldehyde 9, the nitrile 12, and the
lactone 20, all have smaller steric footprints than
chuangxinmycin, indicative of a tight constraint to
binding in this region of WRS. The lack of activity of
the tetrazole, hydroxamic acid, ester and amide, as well
as of the hydroxymethyl analogue 21, confirm this steric
constraint. Rather than an ionisable feature, the pre-
sence of a small dipolar group with Lewis base char-
acter seems to be critical for binding in this region of
In a standard antibacterial assay to determine minimum
inhibitory concentration (MIC), the only analogue to
show activity was the compound that showed compar-
able inhibition to chuangxinmycin, the lactone 20. This
compound was active against the bacterial pathogens S.
aureus Oxford, MIC 4 mg/mL, Haemophilus influenzae
Q1, MIC 16 mg/mL, and Moraxella catarrhalis 1502,
MIC 16 mg/mL. The activity is likely to be due to the
intact lactone as the potential hydrolysis product 21 was
completely inactive at concentrations up to 32 mg/mL.
In summary we have shown that chuangxinmycin is a
potent and selective inhibitor of bacterial WRS. The
interaction is highly constrained with only sterically
smaller analogues showing significant inhibition. The
only analogue to show inhibition comparable to
chuangxinmycin, the lactone 20, also had antibacterial
activity. We suggest that WRS inhibition is likely to
contribute to the mechanism of antibacterial activity of
We thank Stephen Rittenhouse and his team for anti-
References and Notes
1. Chuangxinmycin Research Group, Institute of Materia
Media, Chinese Academy of Medical Sciences Sci. Sin. (Engl.
Ed.) 1977, 20, 106.
2. Liang, H.-T.; Hsu, H.-D.; Chang, C.-P.; Ku, H.-E.; Wang,
W.-S. Hua Hsueh Hsueh Pao 1976, 34, 129. Gu Z.-P.; Liang,
X.-T. Acta Chim. Sin. 1985, 43, 250. Hsu, H.-C.; Shao, M.-C.;
Chang, C.-Y.; Li, K.-P.; Chou, K.-C.; Tang, Y.-C. K’o Hsueh
T’ung Pao 1980, 25, 350.
3. Qi, T. Q.; Liu, X.; Yand, Y. F. Chung-Kuo i Hsueh Ko
Hsueh Yuan Hsueh Pao Acta Academiae Medicinae Sinicae
1980, 2, 32.
4. Tyrosyl tRNA Synthetase Inhibitor: Stefanska, A. L.;
Coates, N. J.; Mensah, L. M.; Pope, A. J.; Ready, S. J.; Warr,
S. R. J. Antibiotics 2000, 53, 345. Houge-Frydrych, C. S. V.;
Readshaw, S. A.; Bell, D. J. J. Antibiotics 2000, 53, 351.
5. Isoleucyl tRNA Synthetase Inhibitor: Stefanska, A. L.;
Cassels, R.; Ready, S. J.; Warr, S. R. J. Antibiotics 2000, 53,
357. Houge-Frydrych, C. S. V.; Gilpin, M. L.; Skett P. W.;
Tyler J. W. J. Antibiotics 2000, 53, 364.
6. Seryl tRNA Synthetase Inhibitor: Stefanska, A. L.; Ful-
ston, M.; Houge-Frydrych, C. S. V.; Jones J. J.; Warr, S. R. J.
Antbiotics, 2000, 53, 1346.
7. Recombinant S. aureus WRS19was overexpressed in E. coli
and purified using standard procedures.
8. The aminoacylation assay was carried out essentially as
described for other aminoacyl tRNA synthetases.9,20SAR
screening was carried at a tryptophan concentration equal to
Km and with ATP in excess (2.5 mM).
9. Pope, A. J.; Moore, K. J.; McVey, M.; Mensah, L.; Benson,
N.; Osborne, N.; Broom, N.; Brown, M. J. B.; O’Hanlon, P. J.
J. Biol. Chem. 1998, 273, 31691.
10. Su, S.; Tu, J.; Zhang, S. Yiyao Gongye 1984, 2, 17.
11. Matsumoto, K.; Watanabe, N. Heterocycles 1987, 26,
12. The ratio of cis:trans isomer in samples that were tested
for biological activity was as follows: 7 1:1, 8 6:1, 9 2:1, 11 1:1,
12 1.3:1, 13 1.1:1, 20 >20:1, 21 >20:1.
13. Laber, B.; Gomis-Ruth, F. X.; Romao, M. J.; Huber, R.
Biochem. J. 1992, 288, 691.
14. Kato, K.; Ono, M.; Akita, H. Tetrahedron Lett. 1997, 38,
1805. Kato, K.; Ono, M.; Akita, H. Tetrahedron: Asymm.
1997, 8, 2295.
15. Mori, K.; Iwasawa, H. Tetrahedron 1980, 36, 87.
Scheme 4. Reagents and conditions: (i) refs 15 and 16, followed by
TBS-Cl; (ii) 4-I-indole/SnCl4, 35%; (iii) MsCl/pyridine/0–25?C then
KSAc/DMF/40?C then K2CO3/MeOH, 35% 17 and 25% 18; (iv)
dithiothreitol/Et3N/60?C, 78%; (v) Pd(PPh3)4/Et3N/THF 86%; (vi)
TBAF/THF, 74%; (vii) LiOH/MeOH then HCl then NaOH 27%.
M. J. Brown et al./Bioorg. Med. Chem. Lett. 12 (2002) 3171–31743173
16. Manchand, P. S.; Luk, K.-C.; Belica, P. S.; Choudhry, Download full-text
S. C.; Wei, C. C. J. Org. Chem. 1988, 53, 5507.
17. The lactone 20, has a coupling constant S-CH-CH of 6.2
Hz, significantly larger than normally observed values (around
3 Hz for cis isomers, around 4–4.5 Hz for trans isomers). Since a
trans fusion of the lactone would geometrically be highly dis-
favoured this suggests a close-to-eclipsed conformation of the
two hydrogen atoms (Karplus curve). Indeed, modelling 20 and
chuangxinmycin shows a reduced dihedral angle (from 52.6 to
34.1?) in the tetracyclic compound. The hydrolysed hydroxy-
carboxylate 21 has a normal cis coupling constant of 3.0 Hz.
18. Macarro ´ n, R.; Mensah, L.; Cid, C.; Carranza, C.; Benson,
N.; Pope, A.; Diez, E. Anal. Biochem. 2000, 284, 183.
19. Hodgson, J. E.; Lawlor, E. J. Eur. Pat. Appl., EP-785263
(to SmithKline Beecham plc), 1997.
20. Brown, P.; O’Hanlon, P. J.; Osborne, N.; Mensah, L.;
Pope, A. J.; Richardson, C. M.; Walker, G. Bioorg. Med.
Chem. 1999, 7, 2473.
3174M. J. Brown et al./Bioorg. Med. Chem. Lett. 12 (2002) 3171–3174