Discovery of a potent and selective series of pyrazole bacterial methionyl-tRNA synthetase inhibitors

Article abstract of DOI:10.1016/S0960-894X(03)00298-1

Starting with a micromolar lead identified from high-throughput screening, a series of pyrazoles were discovered with significantly improved potency on bacterial methionyl-tRNA synthetase and selectivity over human methionyl-tRNA synthetase.

Full text of DOI:10.1016/S0960-894X(03)00298-1

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2231–2234
Discovery of a Potent and Selective Series of Pyrazole Bacterial
Methionyl-tRNA Synthetase Inhibitors
John Finn,* Karen Mattia, Mike Morytko, Siya Ram, Yingfei Yang, Ximao Wu,
Elsa Mak, Paul Gallant and Dennis Keith
Cubist Pharmaceutical Inc., 65 Hayden Ave., Lexington, MA 02421, USA
Received 3 October 2002; accepted 26 February 2003
Abstract—Starting with a micromolar lead identified from high-throughput screening, a series of pyrazoles were discovered with
significantly improved potency on bacterial methionyl-tRNA synthetase and selectivity over human methionyl-tRNA synthetase.
# 2003 Elsevier Science Ltd. All rights reserved.
The emergence of bacterial strains with resistance to
currently marketed antibacterial agents has spurred
interest in the discovery of new antibacterial agents with
novel modes of action.1,2 One set of potential novel
targets are the family of bacterial amino acyl-tRNA
synthetases.3 These enzymes are essential for bacterial
growth and have been validated as drug targets by the
discovery and development of pseudomonic acid, whose
mode of action is inhibition of bacterial isoleucinyl-
tRNA synthetase.4 As part of a broad program to dis-
cover bacterial tRNA synthetase inhibitors,5 pyrazole
1a was identified as an inhibitor of methionyl-tRNA
synthetase (MetRS) by high-throughput screening.6
This compound is a modest micromolar inhibitor of the
bacterial MetRS enzyme from two important Gram-
positive pathogens Staphylococcus aureus and Entero-
cocci faecalis (SaMetRS and EfMetRS) but also inhibits
human MetRS (hMetRS) at similar concentrations. In
this paper, we report our efforts that resulted in com-
pounds with improved potency on bacterial MetRS and
selectivity versus hMetRS.
hydrolysis provided pyrazoles 1a–1f. Scheme 2 shows
the synthesis of additional pyrazoles where different
groups are substituted for the dichlorophenyl and/or
carboxylic acid. Addition of a variety of hydrazines to
diketoester 4a followed by ester hydrolysis yielded a set
of 1-substituted pyrazoles 7. Pyrazole acids were con-
verted to amides by coupling chemistry. Primary amides
were converted to tetrazoles 9 by dehydration to the
corresponding nitrile followed by reaction with dibutlytin
oxide and trimethylsilylazide. Treatment of 4a with dif-
ferent pyridylmethyl hydrazines7 afforded mixtures of
ethyl esters of the 1-(pyridylmethyl)-5-biarylpyrazole-3-
carboxylic acid and 1-(pyridylmethyl)-3-biarylpyrazole-
5-carboxylic acid isomers8 that were separated by column
chromatography. Saponification provided the desired
pyrazole acids 10a–10c and 11a–11c.
Pyrazoles with different biphenyl groups at the 5-position
were prepared as shown in Scheme 1. Suzuki coupling
of different boronic acids to 4-bromoacetophenone
provided a set of biphenyl acetophenones. Addition of
an acetophenone anion to diethyl oxalate yielded the
lithium salt of the diketoester. Treatment of these salts
with 2,4-dichlorophenylhydrazine followed by ester
Compounds were evaluated for inhibition of bacterial
and human MetRS.9 Table 1 shows the result of profil-
ing a series of pyrazole analogues where the biphenyl
*Corresponding author. Tel.: +1-781-860-8336; fax: +1-781-861-
1164; e-mail: john.finn@cubist.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00298-1

2232
J. Finn et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2231–2234
Scheme 1. Synthesis of 1-(2,4-dichlorophenyl)5-biarylpyrazoles-3-carboxylic acids: (a) Pd(PPh3)4, ArB(OH)2, K2CO3; (b) LiN(TMS)2, diethyl
oxalate; (c) 2,4-dichlorophenylhydrazine, EtOH, H+; (d) NaOH.
Scheme 2. Synthesis of pyrazole analogues: (a) R1NHNH2, EtOH, H+; (b) NaOH; (c) EDC, NH2R2; (d) POCl3; (e) Bu2SnO, TMSN3;
(f) R1CH2NHNH2, EtOH, H+; separate isomers.
group is varied. Table 2 shows the results from testing
analogues where the dichlorobiphenyl group was kept
constant and the other two positions varied. None of
the alternative biphenyl groups examined in Table 1
gave an improvement in either potency or selectivity.
Significant improvements in potency and selectivity
were found by variation of the 1-dichlorophenyl group.
Pyrazoles with small substituents in this position have
potency equal or better than the initial lead 1a on
SaMetRS (7a IC50 3.3 mM, 7h IC50 1.0 mM, 7i IC50 4.5
mM). These results demonstrated that the hydrophobic
dichlorophenyl group in the 1-position was not needed
for enzymatic potency. Improvements in SaMetRS
potency were found in several of the 1-aryl substituted
pyrazoles as demonstrated by 1-(3-benzoic acid)-pyrazole
7d (IC50 1.2 mM) and 1-(2-methoxyphenyl)-pyrazole 7e
(IC50 1.5 mM). The position of phenyl substitution is
important for SaMetRS potency as shown by 1-(4-meth-
oxyphenyl)-pyrazole 7f (IC50 10.5 mM). These results
indicate that a hydrogen bond acceptor group may be
beneficial for enzymatic inhibition.
Introduction of pyridyl groups in the 1-position pro-
vided the most potent and selective compounds in the
series. With a carboxylic acid in the 3-position, both the
1-(2-pyridyl) 7k and the 1-(3-pyridyl) 7l pyrazoles have
submicromolar activity on the SaMetRS enzyme (7k
IC50 0.64 mM; 7l IC50 0.30 mM). In addition, some indi-
cations of Gram-positive enzymatic spectrum were
observed for 1-(pyridyl)-pyrazoles, as shown by their
ability to inhibit EfMetRS (e.g., 7k EfMetRS IC50 1.93
mM.) Multiple compounds also have good selectivity
versus the human MetRS enzyme (>150ꢀ for 7k and
150ꢀ for 7l). As in the 1-phenyl series, substitution of a
tetrazole in the 3-position of the pyridyl seies provided
compounds with improved SaMetRS potency, for
example: 1-(2-pyridyl)-pyrazole 9b (SaMetRS IC50 0.15
mM) and 1-(3-pyridyl)-pyrazole 9c (SaMetRS IC50 0.13
mM). Extension of the acidic functionality by a glycine
linking group yielded pyrazoles 8b and 8c, which were
as potent on SaMetRS as the corresponding 3-car-
boxylic acids 7l and 9c indicating a flexible nature to
this enzymatic interaction.

J. Finn et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2231–2234
Table 1. Inhibition of MetRS by 5-biphenyl pyrazole analogues
2233
Compd
R
SaMetRS
IC50 (mM)
hMetRS/SaMetRS ratio selectivity
EfMetRS
8.99
hMetRS
1a
1b
1c
1d
1e
1f
2,4-Cl2C6H3
3-Cl,4-FC6H3
3-(CO2H)C6H4
4-CF3C6H4
4-FC6H4
4.88
17.2
41.3
33.9
43.2
40.9
11.9
20.9
40.9
33.5
41.3
44.3
2
1
1
1
1
1
C6H5
Table 2. Inhibition of MetRS by 5-(20,40-dichlorobiphenyl)pyrazole-3carboxylic acid (i) and 3-(20,40-dichlorobiphenyl)pyrazole-5-carboxylic acid
(ii) analogues
Compd
R1
R2
IC50 (mM)
EfMetRS
8.99
hMetRS/SaMetRS ratio selectivity
SaMetRS
hMetRS
11.9
1a
7a
7b
7c
7d
7e
2,4-Cl2C6H4
H
C6H5
4-CF3OC6H4
3-(CO2H)C6H4
2-CH3O–C6H5
4-CH3O–C6H5
C6H5CH2
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
4.88
3.28
5.47
47.4
1.19
1.51
10.5
3.94
1.02
4.46
3.42
0.63
0.30
2.95
0.30
0.13
0.53
0.15
0.13
1.34
0.66
0.50
2.68
1.09
0.91
2
>100
6.34
17.8
>100
>100
>2
>84
7f
>100
>10
7g
7h
7i
7j
7k
7l
8a
8b
8c
9a
9b
9c
10a
10b
10c
11a
11b
11c
29.9
11.4
1.93
Et
46.4
>100
100
>100
47.0
>100
20.0
13.8
45
>22
29
>159
157
>34
67
106
38
167
82
HOCH2CH2
CH2CO2H
2-Pyridyl
3-Pyridyl
2-Pyridyl
3-Pyridyl
3-Pyridyl
2-CH3O–C6H5
2-Pyridyl
3-Pyridyl
2-Pyridyl-CH2
3-Pyridyl-CH2
4-Pyridyl-CH2
2-Pyridyl-CH2
3-Pyridyl-CH2
4-Pyridyl-CH2
CONH2
CONHCH2CO2H
CONHCH2CN4H
Tetrazole
Tetrazole
Tetrazole
CO2H
3.7
20.0
25.0
10.6
42.5
i
i
i
i
1.79
7.0
32
>152
52
CO2H
CO2H
CO2H
CO2H
CO2H
>100
26.2
22.3
i
ii
ii
ii
8
18
19
19.9
17.3
The 3-position of the pyrazole requires an acidic func-
tionality for enzymatic activity as all of the ester deri-
vatives 6a–6l are inactive in SaMetRS assays at 100 mM.
Replacement of the 3-carboxylic acid with a tetrazole
for the 1-(2-methoxyphenyl)-pyrazole provided a com-
pound, pyrazole 9a (SaMetRS IC50 0.53 mM), with
improved potency relative to the carboxylic acid but
with a reduction in selectivity (38ꢀ) over hMetRS. The
only non-acidic functionality that displayed activity in
the 3-position was the primary amide 8a (SaMetRS IC50
2.95 mM) but this potency was significantly less (5ꢀ)
than the corresponding acid 7k.

2234
J. Finn et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2231–2234
Preparation of 1-(pyridylmethyl)-pyrazole analogues
provided an opportunity to compare the 1,5-di-
substituted-pyrazole-3-carboxylic acid analogues with
1,3-disubstituted-pyrazole-5-carboxylic acid analogues
Using a series of pyridylmethylhydazines, six different
1-pyridylmethyl-pyrazoles were prepared. In every case,
the 1,5 disubstituted-pyrazole-3-carboxylic acid ana-
logue was preferred for SaMetRS potency. These
1-(pyridylmethyl)-pyrazoles have similar potency to the
1-(pyridyl)-pyrazoles.
Ther. Targets 2000, 4, 1. (b) Yu, X.; Hill, J.; Yu, G.; Yang, Y.;
Kluge, A.; Keith, D.; Finn, J.; Gallant, P.; Silverman, J.
Bioorg. Med. Chem. Lett. 2001, 11, 541.
6. (a) For references for other methionyl tRNA synthetase
inhibitors, see: Blanquet, S.; Fayat, G.; Poire, M.; Waller, J.
Eur. J. Biochem. 1975, 51, 567. (b) Lee, J.; Kang, M.; Kyoung,
C.; Moon, W.; Jo, Y.; Kwak, J.; Kim, S. Bioorg. Med. Chem.
Lett. 1998, 8, 3511. (c) Yeong, J.; Lee, S.; Jo. Myung, K.; Lee,
J.; Kang, M.; Yoon, J.; Kim, S. J. Biochem. Mol. Biol. 1999,
32, 547. (d) Lee, J.; Kang, S.; Kang, M.; Chun, M.; Jo, Y.;
Kwak, J.; Kim, S. Bioorg. Med. Chem. Lett. 1999, 9, 1365. (e)
Jarvest, R.; Berge, J.; Berry, V.; Boyd, H.; Brown, M.; Elder,
J.; Forrest, A.; Fosberry, A.; Gentry, D.; Hibbs, M.; Jaworski,
D.; O’Hanlon, P.; Pope, A.; Rittenhouse, S.; Sheppard, R.;
Slater-Radosti, C.; Worby, A. J. Med. Chem. 2002, 45, 1959.
7. Ghali, N. I.; Venton, D.; Hung, S.; LeBreton, G. J. Org.
Chem. 1981, 46, 5413.
8. Treatment of lithium aroylpyruvates with aryl hydrazines is
known to provide the 1,5-diarylpyrazole-3-carboxylate isomer
with high regioselectivity. See: Murray, W.; Wachter, M. J.
Hetero. Chem. 1989, 26, 1389. When different pyridyl-
methylhydrazines were used the 1,3-disubstituted-pyrazole-5-
carboxylate isomers were also isolated as the minor products
with the 1,5-disubstituted-pyrazole-3-carboxylate isomer with
ratios ranging between 3:2 and 3:1. The isomers are readily
differentiated by the NMR signal of the methylene group with
the methylene signal for the 5-carboxylate isomer is 0.5 ppm
downfield of that for the 3-carboxylate isomer (e.g., 5.60 ppm
for 10c and 6.10 ppm for 11c).
In summary, optimization of a micromolar pyrazole
lead that lacked selectivity provided a set of sub-
micromolar 1-pyridyl-pyrazoles. These compounds have
significant increased selectivity for the bacterial MetRS
enzyme over the human MetRS enzyme. Although
multiple pyrazole analogues displayed antibacterial
activity versus S. aureus, the mechanism of action could
not be tied soley to inhibition of MetRS. The advances
in potency and selectivity for the pyrazole series sug-
gests that MetRS may be a useful target for discovering
other series of inhibitors with potency and selectivity.
References and Notes
1. Gold, H.; Moellering, R., Jr. New Engl. J. Med. 1996, 335,
1445.
2. Payne, D.; Wallis, N.; Gentry, D.; Rosenberg, M. Curr.
Opin. Drug Discov. Develop. 2000, 3, 177.
3. (a) Schimmel, P.; Tao, J.; Hill, J. FASEB J. 1998, 12, 1599.
(b) Tao, J.; Schimmel, P. Exp. Invest. Drugs 2000, 9, 1767.
4. (a) Hughes, J.; Mellows, G. J. Antibiot. 1978, 31, 330. (b)
Hughes, J.; Mellows, G. Biochem. J. 1980, 191, 209.
5. (a) Gallant, P.; Finn, J.; Keith, D.; Wendler, P. Emerg.
9. Enzymatic inhibition was determined by measuring the
amount of radiolabeled methionine incorporated into the
product, the charged methionyl–tRNA complex. The enzy-
matic reaction was run at Km concentrations for ATP and
methionine and a saturating level of 90 mM for tRNA. A
compound’s IC50 was determined by fitting the results from a
10-point dose/response curve. All results are the average of at
least two measurements.