Discovery of the first antibacterial small molecule inhibitors of MurB

Article abstract of DOI:10.1016/S0960-894X(02)01076-4

A series of imidazolinone analogues was synthesized and shown to possess potent MurB inhibitory as well as good antibacterial activity.

Full text of DOI:10.1016/S0960-894X(02)01076-4

Bioorganic & Medicinal Chemistry Letters 13 (2003) 873–875
Discovery of the First Antibacterial Small Molecule
Inhibitors of MurB
Joanne J. Bronson, Kenneth L. DenBleyker, Paul J. Falk, Robert A. Mate,
Hsu-Tso Ho, Michael J. Pucci and Lawrence B. Snyder*
Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway, Wallingford, CT 06492, USA
Received 22 November 2002; revised 12 December 2002; accepted 12 December 2002
Abstract—A series of imidazolinone analogues was synthesized and shown to possess potent MurB inhibitory as well as good
antibacterial activity.
# 2003 Elsevier Science Ltd. All rights reserved.
Resistance of pathogenic bacteria to available anti-
biotics is quickly becoming a major problem in the
community and hospital based healthcare settings. For
example, the incidence of penicillin resistant Strepto-
coccus pneumoniae (PRSP) in the United States has
increased from less than 5% in the late 1980s to over
40% today.1 This dramatic rise in resistance rate is also
reflected in other parts of the world and by other strains
of bacteria. Consequently, the search for novel agents to
combat resistant bacteria has become one of the most
important areas of antibacterial research today. The
Mur enzymes2 are integral components in bacterial
peptidoglycan biosynthesis and therefore represent
attractive newtargets for antibacterial design.
step of bacterial peptidoglycan biosynthesis. Subsequent
enzymes in the Mur pathway catalyze cytosolic forma-
tion of the UDP-MurNAc pentapeptide which is ulti-
mately incorporated into the nascent bacterial
peptidoglycan. We felt that a small molecule inhibitor
of MurB would have the potential to be a superior
antibacterial agent for a number of reasons. MurB has
been shown to be essential for bacterial cell growth,4
which means that an inhibitor could be expected to be
bactericidal. As MurB has no known counterparts in
eukaryotes,5 selectivity and toxicity should not be a
problem. Furthermore, as MurB is found in both gram
positive and gram negative organisms it would be
expected that an inhibitor might possess broad spec-
trum antibacterial activity. Lastly, since previous studies
have elucidated the biochemical,6 mechanistic,7 and
structural8 features of the MurB enzyme, we had the
potential to use rational drug design techniques in our
quest.
It was recently reported9 that tri-substituted thiazolidi-
nones, exemplified by 1, inhibit the MurB enzyme at the
lowmicromolar level. These compounds were prepared
using a stereorandom parallel synthesis approach,
which provided each thiazolidinone analogue as a mix-
ture of all four possible diastereoisomers. We report
herein the discovery of a heterocyclic bioisosteric repla-
cement for the thiazolidinone nucleus which eliminates
the need to deal with multiple diastereoisomers. This
newheterocycle not only retains potent in vitro MurB
inhibitory activity, but it demonstrates whole cell anti-
bacterial activity—an attribute not demonstrated by the
aforementioned thiazolidinones.
The MurB enzyme,3 an NADPH dependant enolpyr-
uvyl reductase, is responsible for the second committed
*Corresponding author. Tel.: +1-203-677-6190; fax: +1-203-677-
7702; e-mail: lawrence.snyder@bms.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(02)01076-4

874
J. J. Bronson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 873–875
We chose as our initial target the imidazolinone analo-
gue of 1. Although there are a number of very efficient
methods known to produce substituted imidazoli-
nones,10 there are fewwhich allowfor the incorporation
of optically active amino acids. We therefore chose to
employ a method11 wherein an a-amino ketone is con-
densed with an appropriately substituted isocyanate.
The requisite aminoketone was prepared in a straight-
forward manner starting from the commercially avail-
able biphenyl ether 2 (Scheme 1). Henry reaction
followed by oxidation afforded the a-nitroketone 3
which was subsequently reduced to provide 4. Con-
densation with the isocyanate derived from norleucine12
provided urea 5, which was in turn cyclodehydrated to
yield imidazolinone 6 along with minor amounts of the
corresponding hydantoin 13. Alkylation of the remain-
ing unsubstituted nitrogen atom was accomplished by
allowing 6 to react with potassium carbonate in the
presence of t-butyl bromoacetate and tetra-
butylammonium iodide at 100 ꢀC followed by liberation
of the free carboxylic acid 9 by the action of
trifluoroacetic acid.
are typically generated with cyanuric fluoride or DAST
and isolated prior to amine addition. A recent report14
has
delineated
the
usage
of
tetramethyl-
fluoroformamidinium hexafluorophosphate (TFFH) to
convert carboxylic acids to their respective acyl fluorides
which are then reacted with amines without prior iso-
lation and/or purification of the intermediate acyl fluo-
ride. We have found that commercially available DAST
can also function in this regard. In the event, treatment
of 9 with DAST in dichloromethane followed by addi-
tion of amine provided good yields of amides 14 and 15
which were subsequently saponified to provide 16 and
17, respectively.15 In order to investigate the role of the
amide side chain in conferring MurB activity, we pre-
pared the NH imidazolinone 6 as well as the alkyl sub-
stituted analogues 10–12. It is worth mentioning that
this synthetic sequence should in principal be amenable
to a solid supported parallel synthesis approach by
attaching the starting amino acid to a solid support.
This would result in the production of a library of imi-
dazolinones with three points of diversity.
Although amidation could in principal be accomplished
with any number of traditional reagents, we have found
that the acyl fluoride, generated in situ, functions
remarkably well as an activated carboxylic acid syn-
thon. Acyl fluorides have long been known to be com-
petent partners in amidation reactions,13 however, they
Biological Results
Both 16 and 17 exhibited good potency against the iso-
lated MurB enzyme6,9,16 (Table 1). Furthermore, as
anticipated, they were equipotent to the previously
reported stereochemical mixture of thiazolidinones9
Scheme 1. Reagents and conditions: (a) CH3NO2, EtOH, then PCC, CH2Cl2; (b) SnCl2, EtOH then HCl/Et2O; (c) Norleucine methyl ester, DMAP,
(BOC)2O, CH2Cl2 (72%); (d) H2SO4, MeOH, reflux (50%, along with ꢁ10% of 13); (e) NaOH, MeOH; (f) K2CO3, Bu4N+IÀ, t-butyl bromoace-
tate, DMF, 100 ꢀC (84%); (g) TFA, CH2Cl2 (87%); (h) K2CO3, MeI then NaOH, MeOH (43%); (i) K2CO3, allyl bromide then NaOH, MeOH
(56%); (j) K2CO3, benzyl bromide then NaOH, MeOH (38%); (k) (i) DAST, CH2Cl2; (ii) For compound 14: 3,4-dichlorobenzyl amine; For
compound 15: 2-chlorophenethyl amine.

J. J. Bronson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 873–875
875
Table 1. Antibacterial and MurB inhibitory activity of imidazoli-
nones
6. Dhalla, A. M.; Yanchunas, J., Jr.; Ho, H.-T.; Falk, P. J.;
Villafranca, J. J.; Robertson, J. G. Biochemistry 1995, 34,
5390.
Compd
In vitro inhibition of MurB
IC50 (mM)
Antibacterial activitya
7. (a) Benson, T. E.; Marquardt, J. L.; Marquardt, A. C.;
Etzkorn, F. A.; Walsh, C. T. Biochemistry 1993, 32, 2024. (b)
Lees, W. J.; Benson, T. E.; Hogle, J. M.; Walsh, C. W. Bio-
chemistry 1996, 35, 1342. (c) Benson, T. E.; Walsh, C. T.;
Massey, V. Biochemistry 1997, 36, 796. (d) Constantine, K. L.;
Mueller, L.; Goldfarb, V.; Wittekind, M.; Metzler, W. J.;
Yanchunas, J., Jr.; Robertson, J. G.; Malley, M. F.; Frie-
drichs, M. S.; Farmer, B. T., II J. Mol. Biol. 1997, 267, 1223.
8. (a) Benson, T. E.; Walsh, C. T.; Hogle, J. M. Structure
1996, 4, 47. (b) ref 5. (c) Benson, T. E.; Walsh, C. T.; Hogle,
J. M. Biochemistry 1997, 36, 806. (d) Axley, M. J.; Fairman,
R.; Yanchunas, J., Jr.; Villafranca, J. J.; Robertson, J. G.
Biochemistry 1997, 36, 812.
MIC (mg/mL)
7
9
>118
>101
>115
40
—
—
—
4
2
4
10
11
12
16
17
1
16
15
25
12
4
—
aStaphylococcus aureus A9537.
9. Andres, C. J.; Bronson, J. J.; D’Andrea, S. V.; Deshpande,
M. S.; Falk, P. J.; Grant-Young, K. A.; Harte, W. E.; Ho, H.-
T.; Misco, P. F.; Robertson, J. G.; Stock, D.; Sun, Y.; Walsh,
A. W. Bioorg. Med. Chem. Lett. 2000, 10, 715.
10. (a) Divanfard, H. R.; Ibrahim, Y. A.; Joullie, M. M. J.
Heterocycl. Chem. 1978, 15, 691. (b) De Kimpe, N.; De Cock,
W.; Keppens, M.; De Smaele, D.; Meszaros, J. Heterocycl.
Chem. 1996, 33, 1179. (c) Meanwell, N. A.; Sit, S. Y.; Gao, J.;
Wong, H. S.; Gao, Q.; St Laurent, D. R.; Balasubramanian,
N. J. Org. Chem. 1995, 60, 1565.
From the data shown in Table 1 it is apparent that a
lipophilic substituent on the nitrogen distal to the
biphenyl ether moiety (N-1) is necessary for MurB
inhibitory activity. Compounds 16 and 17 both meet
this requirement and effectively inhibit the isolated
enzyme. Additionally, it is not necessary to have an
amide linkage as exemplified by 11 and 12. It is not
sufficient, however, to merely ‘cap’ the NH with a
methyl group as this leads to complete loss of activity
(10). The requirement for lipophilicity of the N-1 sub-
stituent is again demonstrated by the inactivity of 7 and
9. Of equal importance, these novel imidazolinone ana-
logues possess whole cell antibacterial activity which
tracks with MurB inhibitory activity. It remains, how-
ever, to be shown that the antibacterial activity
observed in S. aureus is due to inhibition of the MurB
enzyme.
11. Holzmann, G.; Krieg, B.; Lautenschlager, H.; Konieczny,
¨
P. J. Heterocycl. Chem. 1979, 16, 983.
12. Knolker, H. J.; Braxmeier, T. Synlett. 1997, 925.
¨
13. (a) Carpino, L. A.; Beyermann, M.; Wenschuh, H.; Bien-
ert, M. Acc. Chem. Res. 1996, 29, 268. (b) Kaduk, C.; Wen-
schuh, H.; Beyermann, M.; Forner, K.; Carpino, L.; Bienert,
M. Lett. Pep. Sci. 1995, 2, 285.
14. Carpino, L. A.; El-Faham, A. J. Am. Chem. Soc. 1995,
117, 5401.
15. Experimental procedure for the conversion of 9 to 15: to
50 mg (0.10 mmol) carboxylic acid 9 in 5 mL dichloromethane
under nitrogen was added 0.12 mL (1 mM in dichloro-
methane) of DAST. After 10 min, 0.15 mL (1 mM in di-
chloromethane) of 3, 4-dichlorobenzylamine was added. After
an additional 10 min, the reaction was filtered through a pre-
conditioned SAX cartridge followed by filtration through a
preconditioned SCX cartridge. Concentration provided 36 mg
(55%) of analytically pure ester 14. 1H NMR (300 MHz;
CDCl3) d 7.40–7.30 (m, 5H), 7.18–6.91 (m, 6H), 6.31 (s, 1H),
4.52–4.38 (m, 5H), 3.62 (s, 3H), 2.33–2.19 (m, 1H), 2.09–1.95
(m, 1H), 1.32 (s, 9H), 1.20–1.08 (m, 4H), 0.79 (app t, 6.9 Hz,
3H). To 12.5 mg (0.019 mmol) ester 14 in 0.5 mL methanol
was added 57 mL (0.057 mmol) 1 N NaOH. After 24 h the
reaction mixture was acidified with 1 N HCl, extracted with
Et2O, washed with brine and dried over MgSO4 to provide
In conclusion, the imidazolinone analogues described in
this communication are the first reported stereo-
chemically discrete MurB inhibitors possessing anti-
bacterial activity, and as such, represent a promising
chemotype in the search for novel antibacterial agents.
References and Notes
1. Thornsberry, C.; Ogilvie, P. T.; Holley, H. P., Jr.; Sahm,
D. F. Antimicrob. Agents Chemother. 1999, 43, 2612.
2. Bugg, T. D.; Walsh, C. T. Nat. Prod. Rep. 1992, 9, 199.
3. (a) Sarver, R. W.; Rogers, J. M.; Epps, D. E. J. Biomol.
Screening 2002, 7, 21. (b) Harris, M. S.; Hergerg, J. T.; Cial-
della, J. I.; Martin, J. P., Jr.; Benson, T. E.; Choi, G. H.;
Baldwin, E. T. Acta Cryst., Section D: Biol. Chryst. 2001, D57,
1032.
4. (a) Miyakawa, T.; Matsuzawa, H.; Matsuhashi, M.;
Sugino, Y. J. Bacteriol. 1972, 112, 950. (b) Rowland, S. L.;
Errington, J.; Wake, R. G. Gene 1995, 164, 113.
5. Benson, T. E.; Filman, D. J.; Walsh, C. T.; Hogle, J. M.
Nat. Struct. Biol. 1995, 2, 644.
1
10mg (82%) of 16. H NMR (300 MHz, CD3OD) d 7.49–7.39
(m, 5H), 7.27–7.23 (m, 1H), 7.12–7.03 (m, 2H), 6.98–6.93 (m,
3H), 6.60 (s, 1H), 4.54 (dd, 4.79 and 10.8 Hz, 1H), 4.43–4.40
(m, 4H), 2.17–2.10 (m, 1H), 2.03–1.99 (m, 1H), 1.34 (s, 9H),
1.20–0.90 (m, 4H), 0.797 (t, 6.9 Hz, 3H).
16. (a) Sylvester, D. R.; Alvarez, E.; Patel, A.; Ratnam, K.;
Kallender, H.; Wallis, N. G. Biochem. J. 2001, 355, 431. (b)
Pucci, M. P.; Discotto, L. F.; Dougherty, T. J. J. Bacteriol.
1992, 174, 1690.