Structure-activity relationships of novel potent MurF inhibitors

Article abstract of DOI:10.1016/j.bmcl.2003.09.073

A novel class of MurF inhibitors was discovered and structure-activity relationship studies have led to several potent compounds with IC ₅₀=22～70 nM. Unfortunately, none of these potent MurF inhibitors exhibited significant antibacterial activity even in the presence of bacterial cell permeabilizers.

Full text of DOI:10.1016/j.bmcl.2003.09.073

Bioorganic & Medicinal Chemistry Letters 14 (2004) 267–270
Structure–activity relationships of novel potent MurF inhibitors
Yu Gui Gu,* Alan S. Florjancic, Richard F. Clark, Tianyuan Zhang, Curt S. Cooper,
David D. Anderson, Claude G. Lerner, J. Owen McCall, Yingna Cai,
Candace L. Black-Schaefer, Geoffrey F. Stamper, Philip J. Hajduk and Bruce A. Beutel
Infectious Disease Research, Global Pharmaceutical Research and Development, Abbott Laboratories, 200 Abbott Park Road,
Abbott Park, IL 60064, USA
Received 8 July 2003; accepted 8 September 2003
Abstract—A novel class of MurF inhibitors was discovered and structure–activity relationship studies have led to several potent
compoundswith IC 50=22ꢀ70 nM. Unfortunately, none of these potent MurF inhibitors exhibited significant antibacterial activity
even in the presence of bacterial cell permeabilizers.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
The emergence of drug resistance poses a major chal-
lenge to the antibacterial research community and clin-
iciansworldwide. Among the widely prcersibed
antibiotics, resistance rates for b-lactamsand macro-
lideshave reached ꢀ25%.1,2 Although resistance rates
to fluoroquinolone antibacterial agentshave been low in
the United States, with increased usage and over-pre-
scription the emergence of resistance to this class of
2. Synthesis
agentsisinevitable, and hasbeen reported in esveral
countries.3,4 Consequently, efforts to discover novel
antibacterial agents capable of overcoming drug resis-
tance have been a continuing interest in our laboratories.
2-Aminothiophene 3 wasreadily ysntheizsed from
cyclopentanone and malononitrile according to a litera-
ture procedure.10 The reaction of one equivalent of 3 in
the presence of a base such as triethylamine with m-
chlorocarboxbenzenesulfonyl chloride derivatives 5,
which were prepared from the corresponding benzoic
acid compounds 4 via chlorosulfonation, produced the
sulfonyl chloride derivatives 7 in good yields. Treatment
of 7 with either primary or secondary amines gave the
varioususlfonamides 8 (Scheme 1). The yieldsvaried
from 20 to 80% depending on the additional functional
groupsthat the aminesbear.
Peptidoglycan is an essential and unique building block
of the bacterial cell wall and hasbeen the target of many
drug classes including b-lactams, cephalosporines and
glycopeptides. Murein enzymes are involved in the bio-
synthesis of bacterial peptidoglycan at various stages.5
Gene knockout studies have shown that these Murein
enzymes are essential for the survival of the bacterial
cell and therefore are attractive targets.6ꢁ8 Recently, we
discovered two credible MurF inhibitor leads (1, 2) via
an affinity selection screening technology developed at
Abbott.9 Here we wish to report our structure–activity
relationship study of these novel MurF inhibitors.
The modification of the central portion of the molecule
was accomplished by straightforward functional group
manipulations. The cyano group of compounds 8 was
hydrolyzed to amide 9 or selectively reduced to primary
amine 10 with borane and then further acylated to
amide 11. The ethyl ester analogue 12 wasprepared
from the commercially available ethyl 2-amino-5,6-
dihydro-4H-cyclopenta[b]thiophene-3-carboxylate and 5
by following the reaction sequences for compounds 8 (b
* Corresponding author. Tel.: + 1-847-937-4792; fax: + 1-847-938-
3403; e-mail: yu-gui.y.gu@abbott.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.09.073

268
Y. G. Gu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 267–270
Scheme 1. Reagentsand conditions: (a) ClSO 3H, 95 ꢂC, 70–88%; (b) toluene, reflux, 20 h, 55–83%; (c) 10 or 20 amines, NEt3, THF, rt, 2–24 h, 20–
80%; (d) concd H2SO4, 21 h, 55%; (e) BH3, THF, 65 ꢂC, 2 h, 40%; (f) Ac2O, CH2Cl2, rt, 59%; (g) MeI, K2CO3, DMA, 74%; (h) (1) benzene, cat. p-
TSA, reflux, 30 min; (2) NaCNBH3, 32–40%; (i) amine, NEt3, THF, rt, 16 h, 40–70%; (j) 10 or 20 amines, dioxane, reflux, 1–5 days, 22–79%
(R3=amines), or 10 equiv alkoxide, DMA, 85 ꢂC, 1–24 h, 35–80% (R3=alkoxy).
and c of Scheme 1). Methylation of the amide NH of 8
provided compound 13. The compoundswith an amine
linkage (14 and 15) were prepared via reductive amina-
tion of 3 with an aldehyde or ketone 6. Selective
nucleophilic displacement of 8 (R1=F) with amines
afforded the variousaniline analogues( 8, R1=amino
groups) in good yield under mild conditions. On the
other hand, displacement of monochlorobenzene-
sulfonamide 8 (R1=H) with aminesrequired harhs
conditions(high temperature and long reaction time) to
produce desired compounds 16, while the reaction with
alkoxidesproceeded more smoothly.
radiolabeled inorganic phosphate from ATP. In a
modification of the ATPase end-point assay of Seals et
al.,11 purified recombinant MurF from Streptococcus
pneumoniae (preparation of which will be described
elsewhere9) wascombined with UDP- N-acetylmuramyl-
l-Ala-g-d-Glu-Lys(purified from Staphylococcus aur-
eus12), d-Ala-d-Ala (Sigma, St. Louis, MO, USA) and
adenosine 50-[g-33P]-triphosphate (Amersham, Piscat-
away, NJ, USA). Progress of the reaction was determined
by measuring the number of counts released as free
phosphate. Compound concentrations resulting in 50%
inhibition of enzyme activity (IC50) were graphically
determined from four-point plotsof concentration ranges
spanning this value. Each plotted value was the mean of
at least two determinations, and each IC50 value wasthe
mean derived from at least two independent plots.
Similarly, analogues 19 and 20 with different cycloalkyl
ring sizes were synthesized from compounds 17 and 1810
(Scheme 2).
Variouusslfonamidesare well tolerated exhibiting
MurF inhibitory activity comparable to 1 (8, entries1–
4, Table 1). Interestingly, the compound with a basic
amine group attached to the sulfonamide is poorly tol-
erated (8, entry 5) while the corresponding hydroxyl
analogue (8, entry 4) followsthe general trend. A halo-
gen atom ortho to the sulfonamide improves activity,
especially for the ortho-chloro morpholino sulfonamide
(8, entry 8), which is30 timesmore potent than the
corresponding parent compound (8, entry 1). However,
replacement of the halogen with an amine group sig-
nificantly lowers potency especially in those analogues
containing a large amine (8, entries9–11).
Scheme 2. Reagentsand condition:s (a) 5, toluene, relux, 20 h, 60–
80%; (b) 10 or 20 amines, NEt3, THF, rt, 2–24 h, 40–75%.
3. Results and discussion
The MurF enzyme is responsible for incorporation of a
d-alanyl d-alanyl moiety during peptidoglycan synth-
esis.5 The d-alanyl d-alanine adding activity of MurF
was monitored by measuring the concomitant release of
It isinteresting that the cyano group of compound 1 is
essential for MurF inhibitory potency. Conversion of

Y. G. Gu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 267–270
Table 1. MurF activities from the radiolabeled phosphate release assay
269
Entry
1
Compd
R1
H
R2
R3
MurF IC50 (mM)
8
9
2
3
4
5
8
8
8
8
H
H
H
H
–NHMe
–NHPh
–NH(CH2)3OH
–NH(CH2)3NMe2
15
26
8
>100
6
8
F
6.4
7
8
8
8
Cl
Cl
–NEt2
5.2
0.3
9
8
8
8
–NHMe
62
>100
>100
10
11
–NH(CH2)3OH
–NH(CH2)3NMe2
12
13
14
15
16
17
18
9
–NEt2
–NEt2
–NEt2
–NEt2
–NEt2
–NEt2
–NEt2
66
>100
>100
>100
>100
>100
>100
10
11
12
13
14
15
19
20
21
22
23
24
16
16
16
16
16
16
–NH(CH2)3NMe2
74
>100
>100
>100
>100
4.2
–NMe2
–OEt
–OCH2CHMePh
H
Br
25
26
19
19
H
H
–NEt2
–NEt2
H
H
1.4
1.7
27
28
19
19
Cl
H
Ph
0.07
0.054
29
30
19
19
H
–NEt2
–NEt2
–CO2Et
6
Cl
0.067
31
19
Cl
0.022
32
33
20
20
H
H
–NEt2
H
H
3.4
4.2
the cyano group to amides 9, 11, amine 10, or ethyl ester
12, all resulted in the significant or complete loss of
activity. The amide linker in the center of the molecule
is also essential for potency. A total loss of activity was
observed when the amide of compound 1 wasmethyl-
ated (13) or converted to amines( 14, 15). When the
chlorine atom ortho to the amide linker wasdipslaced
with amino or alkoxy groups, the resulting analogues 16
(entries19–22) were inactive compared with the parent
compound (8, entry 1), regardless of the size of the
substituents (entry 19 vs 20, and entry 21 vs 22). A
cyano group at the C-3 position of thiophene, a NH
amide linker, and a chlorine ortho to the amide linker
are all required for MurF activity. These strict require-
ments suggest that both the cyano and an amide NH
may be involved in hydrogen bonding interactionswith

270
Y. G. Gu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 267–270
the enzyme. The chlorine atom may provide proper
orientation for such interactions by forcing the phenyl
ring into a non-coplanar position with the amide linker.
to MurF not catalyzing a rate limiting step of the bio-
synthetic pathway. More research is needed to under-
stand why the inhibition of MurF does not result in
whole cell antibacterial activity.
Indeed, removal of the chlorine results in loss of activity
13
(entry 23), whereasthe corresponding bromide
analo-
gue (entry 24) slightly increases potency. Itai and co-
workersreported that N-methylbenzanilide exists in a
cis-amide conformation while the unsubstituted NH
benzanilide exists in a trans-amide conformation.14
Therefore, it is also possible that the lack of activity of
Acknowledgements
We would like to thank Drs. Peter Dandliker and
Kenneth Comess for helpful discussions.
15
13 isdue to the change of amide conformation.
Cyclohexyl analogues 19 exhibit better potency than the
corresponding cycloheptyl counterparts 20, which in
turn seem to be more potent than cyclopentyl analo-
gues. This is demonstrated by the decreasing potency
from 19 (entry 25, IC50=1.4 mM) to 20 (entry 32,
IC50=3.4 mM) to 1 (IC50=8 mM) and from 19 (entry 26,
IC50=1.7 mM) to 20 (entry 33, IC50=4.2 mM) to 8 (entry
1, IC50=9 mM). A major potency boost was achieved
when an aryl group wasappended to the cyclohexyl
group. A phenyl group increases potency 14-fold (2,
IC50=1 mM vs 19, entry 27, IC50=70 nM) and 4-
hydroxyphenyl boosts potency by 45-fold (19, entry 31,
IC50=22 nM). The corresponding diethyl sulfonamide
analogues 19 (entries 28 and 30) also show high potency
with IC50=54 nM and IC50=67 nM, respectively.
Introduction of an ester at the same position of the
cyclohexyl group reducespotency more than 4-fold ( 19,
entry 25 vs29), indicating that a p–p interaction or
stacking of the aryl group with the enzyme may be
responsible for the potency boost.
References and notes
1. Doern, G. V.; Heilmann, K. P.; Huynh, H. K.; Rhom-
berg, P. R.; Coffman, S. L.; Brueggemann, A. B. Anti-
microb. Agents Chemother. 2001, 45, 1721.
2. Doern, G. V. Clin. Infect. Dis. 2001, 33, S187.
3. Goldsmith, C. E.; More, J. E.; Murphy, P. G.; Ambler,
J. E. J. Antimicrob. Chemother. 1998, 41, 420.
4. Ho, P. L.; Que, T. L.; Tsang, D. N. C.; Ng, T. K.; Chow,
K. H.; Seto, W. H. Antimicrob. Agents Chemother. 1999,
43, 1310.
5. (a) Murein synthesis: Heijenoort, J. V. In Escherichia
coli and Salmonella Cellular and Molecular Biology, 2nd
ed.; Neidhardt F. C., Ed.; ASM: Washington, DC,
1996; Vol. 1, p 1025. (b) For a review on Mur enzyme
inhibitors, see: Zoeiby, A. E. et al. Mol. Microbiol. 2003,
47, 1.
6. For MurB inhibitors, see: Snyder, L. B. Bioorg. Med.
Chem. Lett. 2003, 13, 873.
7. For weak transition state inhibitors of MurF, see: Miller,
D. J. et al. J. Chem. Soc., Perkin Trans. 1. 1998, 131.
8. (a) For MurI inhibitors, see: de Dios, A., et al. J. Med.
Chem. 2002, 45, 4559. (b) Tanner, M. E. et al. Bioorg.
Med. Chem. Lett. 1997, 7, 2265.
9. Manuscript in preparation for publication.
10. For a review, see: Sabnis, R. W. et al. J. Heterocyclic.
Chem. 1999, 36, 333.
In conclusion, we have discovered a novel class of
potent MurF inhibitors. Unfortunately, even the most
potent compounds do not show significant antibacterial
activity. There could be many reasons for the lack of
antibacterial activity, including poor cellular perme-
ability, efflux, non-selective intracellular binding of our
compoundsto other proteinsor other unknown rea-
sons. To address the permeability and efflux issues, we
measured antibacterial activity of our MurF inhibitors
in the presence of some well-known permeabilizers
(Escherichia coli with 1 mg/mL ethylenediaminete-
traacetic acid, S. aureus with 64 mg/mL nisin16) aswell
as E. coli AcrAB efflux pump mutants.17 Unfortunately,
no antibacterial activitieswere observed. Since the gene
encoding MurF is essential for bacterial survival, the
lack of antibacterial activity of thisseriescould be due
11. Seals, J. R.; McDonald, J. M.; Bruns, D.; Jarett, L. Anal.
Biochem. 1978, 90, 785.
12. Kohlrausch, U.; Holtje, J.-V. FEMS Microbiol. Lett.
1991, 78, 253.
¨
13. Prepared from the commercially available 2-bromo-5-
sulfo-benzoic acid.
14. Itai, A.; Toriumi, Y.; Tomioka, N.; Kagechika, H.; Azu-
maya, I.; Shudo, K. Tetrahedron Lett. 1989, 30, 6177.
15. For an example, see: Hirayama, F. et al. Bioorg. Med.
Chem. 2002, 10, 2597.
16. Vaara, M. Microbiol. Rev. 1992, 56, 395.
17. Ma, D.; Cook, D. N.; Alberti, M.; Pon, N. G.; Nikaido,
H.; Hearst, J. E. Mol. Microbiol. 1995, 16, 45.