A series of spirocyclic analogues as potent inhibitors of bacterial phenylalanyl-tRNA synthetases

Article abstract of DOI:10.1016/j.bmcl.2003.11.081

We have identified a series of spirocyclic furan and pyrrolidine inhibitors of Enterococeus faecalis and Staphylococcus aureus phenylalanyl-tRNA synthetases. The most potent analogue 1b showed IC₅₀=5 nM (E. faecalis PheRS) and IC₅₀=2

Full text of DOI:10.1016/j.bmcl.2003.11.081

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1339–1342
A series of spirocyclic analogues as potent inhibitors
of bacterial phenylalanyl-tRNA synthetases
Xiang Y. Yu,a,* John Finn,a Jason M. Hill,a Zhong G. Wang,a
Dennis Keith,a Jared Silvermanb and N. Oliverb
aDepartment of Medicinal Chemistry, Cubist Pharmaceuticals, Inc., 65 Hayden Ave, Lexington, MA 02421, USA
bDepartment of Biology, Cubist Pharmaceuticals, Inc., 65 Hayden Ave, Lexington, MA 02421, USA
Received 27 March 2003; accepted 26 November 2003
Abstract—We have identified a series of spirocyclic furan and pyrrolidine inhibitors of Enterococeus faecalis and Staphylococcus
aureus phenylalanyl-tRNA synthetases. The most potent analogue 1b showed IC50=5 nM (E. faecalis PheRS) and IC50=2 nM (S.
aureus PheRS) with high selectivity over the human enzyme. The crystal X-ray structure of analogue 1b was determined.
# 2004 Elsevier Ltd. All rights reserved.
Emergence of resistant bacteria has given new urgency
to the search for new antibacterial agents that act via
novel mechanisms of action.1ꢀ3 Aminoacyl-tRNA
synthetases are essential enzymes for biological cell
growth.4ꢀ8 We have previously described several series
of aminoacyl-tRNA synthetase inhibitors.9ꢀ11 As part
of our high throughput screening efforts identifying
t-RNA synthetase inhibitors, we discovered that a series
of spirocyclic compounds had good inhibition of Enter-
ococcus faecalis and Staphylococcus aureus Phenylala-
nyl-tRNA synthetases (EfPheRS and SaPheRS). The
lead compound 1a inhibited EfPheRS (IC50=0.82 mM)
and SaPheRS (IC50=0.38 mM) with high selectivity
over the human enzyme (IC50 >100 mM). In this
communication, we report on the structure–activity
relationships of the series of spirocyclic analogues based
on compound 1a.
In the case of the 3,4-dichlorophenyl analogue, the two
isomers, cis (1a) 15 and trans (1b), were formed in a ratio
of 2:1 and separated by silica gel chromatography (20%
hexane in dichloromethane).
The spirocyclic furan analogues were prepared by
synthetic route outlined in Scheme 1. Reaction
of 1,3-indandione 2 with aldehydes in the presence of
piperidine produced 2-benzylidene-1,3-indandiones 3.
Treatment with 30% hydrogen peroxide in methanol
formed 3-phenylspirocyclic [oxirane-2,20-indan]-10,30-
dione 4.12 Heating the epoxide 4 gave an intermediate
carbonyl ylide which underwent 1,3-dipolar cycloaddi-
tion with maleimides 513 to give spirocyclic isomers 1
and 6–14.14
* Corresponding author. Tel.: +1-781-372-4840; fax: +1-781-274-
9129; e-mail: xyu@activbiotics.com
Scheme 1.
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.11.081

1340
X. Y. Yu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1339–1342
The structure of these compounds was determined by
1H NMR experiments and mass spectra. X-ray crystal-
lography of analogue 1b (Fig. 1) further confirmed the
stereochemistry.
more potent than the cis spirocyclic analogues. The
trans 3,4-dichlorophenyl analogue 1b is the most potent
compound in this series. It inhibits EfPheRS (IC50=5
nM) and SaPheRS (IC50=2 nM) with high selectivity
over the human enzyme (IC50 >100 mM). This trend is
also evident in comparing the cis 3-chloro-4-methylphe-
nyl analogue 6a and the trans chloro-4-methylphenyl
analogue 6b. The substitution pattern on the R2 posi-
tion also affects the potency. 3,4-Methylenedioxy ben-
zene analogue 7 has moderate activity whereas 3-
Scheme 2 shows the preparation of spirocyclic pyrroli-
dine analogues. Reaction of ninhydrin 15a, phenyl gly-
cine 16 and substituted maleimide 5a provided two
isomers (17a and b),16 which were separated by Flor-
isil1 chromatography. Analogue 22 was synthesized by
coupling with amine 20, aldehyde 21 and substituted
maleimide 5a. Because formation of carbomethoxy
analogues in one step was unsuccessful, analogue 25 was
prepared stepwise by generating ylide first followed by
treatment with 5a.17,18
(trifluoromethyl)-4-chlorophenyl analogue
9 is less
active. Analogue 13 with an electron-donating methoxy
group and analogue 14 with an electron-withdrawing
nitro group show a significant loss in potency. Interest-
ingly, only limited variations appear to be tolerated at
the R1 position. The 4-methyl and 4-chlorophenyl ana-
logues (8 and 10) showed a significant loss in potency.
The 2-methyl-phenyl analogue 12 is ten times less active
than the phenyl analogue 11.
Spirocyclic analogues were evaluated for inhibition of
the aminoacylation activity19 of EfPheRS and SaPheRS
and the results of spirocyclic furan analogues are shown
in Table 1. The stereochemistry dramatically affects the
activity. The trans spirocyclic analogues are significantly
The results of spirocyclic pyrrolidine analogues are
shown in Table 2. It is worthy to note that the trans
spirocyclic pyrrolidine analogues were also more potent
than the cis spirocyclic analogues. 3,4-Dichlorophenyl
cis analogue 17a is 100 times less active than trans ana-
logue 17b. Analogues without carbonyl groups on the
spirocyclic ring (19a and b) were inactive. Benzoyl ana-
logues 22a and 22b were only moderately active whereas
the carbomethoxy analogue 25 was inactive.
Figure 1. The crystal X-ray structure of analogue 1b.
Figure 2. WT S. aureus macromolecular labeling with PMA.
Scheme 2.

X. Y. Yu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1339–1342
Table 1. Inhibition of EfPheRS and SaPheRS by furan analogues
1341
Compd
R1
R2
IC50 (mM)
(EfPheRS)
IC50 (mM)
(SaPheRS)
IC50 (mM)
(Human PheRS)
1a (cis)
1b (trans)
6a (cis)
6b (trans)
7 (cis:trans=25:1)
8 (cis)
9 (cis)
10 (cis)
11 (cis:trans=18:1)
12 (cis)
13 (cis)
Ph
Ph
Ph
Ph
3,4-Cl2-Ph
3,4-Cl2-Ph
0.85
0.005
8.3
0.22
2.7
0.56
0.002
13.7
0.07
0.97
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
3-Cl-4-Me-Ph
3-Cl-4-Me-Ph
3,4-(OCH2O)-Ph
3,4-(OCH2O)-Ph
3-CF3-4-Cl-Ph
3-CF3-4-Cl-Ph
2-OMe-Ph
Ph
4-Me-Ph
Ph
4-Cl-Ph
Ph
2-Me-Ph
Ph
Ph
>100
32
>100
6.1
57
>100
45
>100
1.8
13
2-OMe-Ph
4-OMe-Ph
4-NO2-Ph
>100
>100
>100
>100
14 (cis)
These spirocyclic furan and pyrrolidine compounds also
exhibit Escherichia coli PheRS inhibitory activity. For
example, the most potent analogue 1b possesses good
activity against E. coli PheRS (IC50=30 nM).
whole cell activity (MIC=50 mg/mL, S. aureus
ATCC6538P). Inhibitors of tRNA synthetase activity
produce a unique profile in assays monitoring macro-
molecular synthesis in S. aureus. As shown in Figure 2,
pseudomonic acid (PMA),20ꢀ24 a known inhibitor of
isoleucine-tRNA synthetase, inhibits synthesis of both
protein and RNA. RNA synthesis is inhibited due to
The analogues were tested for antimicrobial activity
against a panel of organisms. Analogue 1b has weak
Table 2. Inhibition of EfPheRS and SaPheRS by pyrrolidine analogues
Compd
R1
R2/R3
IC50 (mM)
(EfPheRS)
IC50 (mM)
(SaPheRS)
IC50 (mM)
(Human PheRS)
17a (cis)
3,4-Cl2Ph
3,4-Cl2Ph
4.0
3.0
>100
>100
17b (trans)
0.04
0.04
18a (cis)
3-Cl,4-Me-Ph
3-Cl,4-Me-Ph
3,4-Cl2Ph
8.7
0.99
6.1
0.73
>100
>100
>100
>100
18b (trans)
19a (cis)
>100
>100
>100
>100
19b (trans)
3,4-Cl2Ph
22a (cis)a
22b (trans)a
25 (cis)a
3,4-Cl2Ph
3,4-Cl2Ph
3,4-Cl2Ph
PhCO/H
PhCO/H
CO2Me/H
15.7
6.7
>100
59
30
>100
>100
>100
>100
a The stereochemical center adjacent to the benzoyl or methoxy carbonyl centers was not assigned.

1342
X. Y. Yu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1339–1342
G.; Weinberger, D. A.; Wilcoxen, K. M.; Ghadiri, M. R.
Nature (London, U. K.) 2001, 452.
2. Frosco, M.; Barrett, J. F. Exp. Opin. Invest. Drugs 1998,
7, 175.
3. Sanders, C. C. Clin. Infect. Dis 1992, 14, 1089.
4. Davis, M. W.; Buechter, D. D.; Schimmel, P. Biochem-
istry 1994, 33, 9904.
5. Delarue, M.; Moras, D. BioEssays 1993, 15, 675.
6. Hou, Y. M.; Francklyn, C.; Schimmel, P. Trends Biochem.
Sci. 1989, 14, 233.
7. Schulman, L. H.; Abelson, J. Science 1988, 240, 1591.
8. Warner, J. R.; Soeiro, R. N. Engl. J. Med. 1967, 12,
675.
9. Yu, X. Y.; Hill, J. M.; Yu, G.; Yang, Y.; Kluge, A. F.;
Keith, D.; Gallant, P.; Silverman, J.; Lim, A. Bioorg.
Med. Chem. Lett. 2001, 11, 541.
10. Yu, X. Y.; Hill, J. M.; Yu, G.; Wang, W.; Kluge, A. F.;
Wendler, P.; Gallant, P. Bioorg. Med. Chem. Lett. 1999, 9,
375.
11. Hill, J. M.; Yu, G.; Shue, Y.; Zydowsky, T. M.; Rebek, J.
Jr. U.S. patent, 5,726,195.
12. Weigele, M.; Tengi, J. P.; De Bernardo, S.; Czajkowski,
R.; Leimgruber, W. J. Org. Chem. 1976, 41, 388.
13. Cava, M. P.; Deana, A. A.; Muth, K.; Mitchell, M. J.
Organic Synthesis 1961, 41, 93.
Figure 3. WT S. aureus macromolecular labeling with analogue 1b.
induction of the stringent response.25 Induction of the
stringent response is inhibited by addition of the protein
synthesis inhibitor chloramphenicol (CAM), allowing
RNA synthesis to continue uninhibited until the dose of
PMA is increased by 10–100-fold. As shown in Figure 3,
analogue 1b also triggers the stringent response, as
indicated by the altered profile of RNA synthesis inhi-
bition in the presence of CAM. This result suggests a
synthetase component to the mechanism of action.
In conclusion, a novel structural class of spirocyclic
furan and pyrrolidine analogues with potent activity
against E. faecalis and S. aureus PheRSs with high selec-
tivity over the human enzyme has been discovered. These
compounds also exhibit E. coli PheRS inhibitory activity.
Further study shows that an analogue 1b was unstable in
Mueller Hinten Broth (MHB) bacterial media. The results
of our efforts to improve the stability and the antibacterial
activity of this series are described in the next paper.
14. Krysin, M. Y.; Anokhina, I. K.; Zalukaev, L. P. Khim.
Geterotsikl. Soedin 1987, 11, 1463.
15. Studies on the mechanism of action for cis (1a) shows that
it is a competitive inhibitor with respect to phenylalanine
(Ki=0.39 mM) and ATP (Ki=0.39 mM). Attempts to
separate the chiral racemic mixtures using various chiral
columns were unsuccessful.
16. Vaidyanathan, G.; Wilson, J. W. J. Org. Chem. 1989, 54,
1810.
17. Galley, G.; Liebscher, J.; Patzl, M. J. Org. Chem. 1995,
60, 5005.
18. Hamper, B. C.; Duckesherer, D. R.; South, M. S. Tetra-
hedron Lett. 1996, 37, 3671.
19. The detailed aminoacylation assays were described in the
patent: Finn, J.; Yu, X. Y.; Wang, Z.; Hill, J. M.; Keith,
D.; Gallant, P.; Wendler, P. PCT int. Appl. (2000), WO
0018772
20. Brown, P.; Davies, D. T.; O’Hanlon, P. J.; Wilson, J. M.
J. Med. Chem. 1996, 39, 446.
Acknowledgements
The authors would like to thank the Biological and Ana-
lytical staff at Cubist for enzyme assay data and spectra,
Professor J. A. Golen at University of Masechusetts at
Dartmouth for the crystal X-ray structure, and the staff at
Bristol-Myers for 1H NMR NOE experiments.
21. Class, Y. J.; DeShong, P. Chem. Review 1995, 95, 1843.
22. Hughes, J.; Mellows, G. Biochem. J. 1980, 191, 209.
23. Hughes, J.; Mellows, G. J. Antibiotics 1978, 31, 330.
24. Hughes, J.; Mellows, G. Biochem. J. 1978, 176, 305.
25. Cashel, M. Encycl. Microbiol. (2nd Ed.) 2000, 4, 467.
References and notes
1. Fernandez-Lopez, S.; Kim, H.; Choi, E. C.; Delgado, M.;
Granja, J. R.; Khasanov, A.; Kraehenbuehl, K.; Long,