2-Aminotetralones: Novel inhibitors of MurA and MurZ

Article abstract of DOI:10.1016/j.bmcl.2008.01.089

Several 2-aminotetralones were identified as novel inhibitors of the bacterial enzymes MurA and MurZ. A number of these inhibitors demonstrated antibacterial activity against Staphylococcus aureus and Escherichia coli with MICs in the range 8-128 μg/ml. B

Full text of DOI:10.1016/j.bmcl.2008.01.089

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1730–1734
2-Aminotetralones: Novel inhibitors of MurA and MurZ
Colin J. Dunsmore,a Keith Miller,b Katy L. Blake,b Simon G. Patching,c
Peter J. F. Henderson,c James A. Garnett,b William J. Stubbings,b
Simon E. V. Phillips,b Deborah J. Palestrant,d Joseph De Los Angeles,e
Jennifer A. Leeds,f Ian Choprab and Colin W. G. Fishwicka,*
aSchool of Chemistry and Antimicrobial Research Centre, University of Leeds, Leeds LS2 9JT, UK
bInstitute of Molecular and Cellular Biology and Antimicrobial Research Centre, University of Leeds, Leeds LS2 9JT, UK
cInstitute of Membrane and Systems Biology and Antimicrobial Research Centre, University of Leeds, Leeds LS2 9JT, UK
dProtein Structure Group, Novartis Institutes for Biomedical Research, Cambridge, MA 02139, USA
eGlobal Discovery Chemistry, Novartis Institutes for Biomedical Research, Cambridge, MA 02139, USA
fInfectious Diseases Area, Novartis Institutes for Biomedical Research, Cambridge, MA 02139, USA
Received 7 November 2007; revised 11 January 2008; accepted 12 January 2008
Available online 30 January 2008
Abstract—Several 2-aminotetralones were identified as novel inhibitors of the bacterial enzymes MurA and MurZ. A number of
these inhibitors demonstrated antibacterial activity against Staphylococcus aureus and Escherichia coli with MICs in the range
8–128 lg/ml. Based on structure–activity relationships we propose that the a-aminoketone functionality is responsible for the inhib-
itory activity and evidence is provided to support a covalent mode of action involving the C115 thiol group of MurA/MurZ.
ꢀ 2008 Elsevier Ltd. All rights reserved.
Peptidoglycan is an essential component of the cell walls
of Gram-positive and Gram-negative bacteria, provid-
ing protection of the bacterial cell from destruction by
osmotic pressure and conferring cellular shape.1 Inter-
ference with bacterial cell wall biosynthesis is firmly
established as an excellent basis for the development
of potent antibacterials; for example the penicillins
and cephalosporins specifically inhibit enzymes involved
at the crosslinking stages of cell wall synthesis.2
As MurA is conserved across both Gram-positive and
Gram-negative bacterial species and is an essential en-
zyme with no mammalian counterpart, it is an attractive
target for the development of new antibacterials, which
are urgently required as resistance to established antibi-
otics continues to escalate.5,6
Inhibition of MurA by the naturally occurring epoxide-
based antibiotic fosfomycin has been extensively de-
scribed.7–9 This drug acts as a PEP surrogate and forms
a covalent adduct with C115 (Escherichia coli number-
ing) via epoxide ring-opening. Fosfomycin also inhibits
MurZ from Staphylococcus aureus by the same
mechanism.
The bacterial enzyme MurA (UDP-GlcNAc enolpyruvyl
transferase), in the first committed step of peptidoglycan
biosynthesis, catalyses the transfer of enolpyruvate from
phosphoenolpyruvate (PEP) to the 30-hydroxyl group of
UDP-N-acetylglucosamine (UDP-GlcNAc) yielding
enolpyruvyl UDP-N-acetylglucosamine (EP-UDP-Glc-
NAc) and inorganic phosphate (Fig. 1).3 Low-GC
Gram-positive bacteria contain two copies of the murA
gene (murA and murZ) which encode, respectively, the
enolpyruvyl transferase enzymes, MurA and MurZ.4
Here, we report the identification of 2-aminotetralones
as a new class of MurA inhibitors and propose a mode
of action involving formation of a covalent adduct.
High throughput screening of the Novartis compound
collection identified compounds 3 and 4 as good inhibi-
tors of E. coli10 MurA with IC50 values of 3.1 and
8.5 lM, respectively (Table 1).11 Compounds 3 and 4
were also identified as inhibitors of MurA and MurZ
from S. aureus12,13 with IC50’s in the range 12–23 lM
Keywords: 2-Aminotetralone; MurA inhibitor; MurZ inhibitor; Anti-
bacterial activity.
*
Corresponding author. Tel.: +44 113 3436510; fax: +44 113
3436565; e-mail: colinf@chem.leeds.ac.uk
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.089

C. J. Dunsmore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1730–1734
1731
-
3-
2-O3PO
PEP
CO2
PO4
CH
2OH
CH2OH
OH
O
O
OH
OH
O
HN
HN
-O2C
O
UDP O
UDP O
O
UDP-GlcNAc
EP-UDP-GlcNAc
Figure 1. MurA catalyses the formation of enolpyruvyl-UDP-N-acetylglucosamine (EP-UDP-GlcNAc) from phosphoenolpyruvate (PEP) and UDP-
N-acetylglucosamine (UDP-GlcNAc).
Table 1. Some 2-aminotetralone-derivatives that inhibit isolated MurA and MurZ enzymes and have antibacterial activities against E. coli and
S. aureus
Compound
Structurea
IC50
(E. coli MurA)b
(lM)
IC50
(S. aureus MurA)b
(lM)
IC50
(S. aureus MurZ)c
(lM)
MIC
(E. coli)d
(lg/ml)
MIC
(S. aureus)e
(lg/ml)
OH
O
N
MeO
N
3
3.13 ( 1.51)
8.53 ( 2.39)
22.56 ( 4.48)
22.49 ( 0.74)
>120
17.35 ( 3.30)
12.04 ( 0.40)
29.00 ( 0.16)
12.52 ( 1.82)
>120
18.83 ( 2.37)
22.91 ( 0.80)
28.93 ( 1.02)
16.09 ( 3.15)
>120
256
128
64
MeO
O
N
N
4
128
O
N
Ph
MeO
N
5
>256
128
64
MeO
O
OH
N
6
8
N
OH
OH
OH
N
N
7f
>256
>256
>256
8
>256
>256
>256
16
N
N
10
>120
>120
>120
O
12
>120
>120
>120
O
H
2-O3P
H
Fosfomycin
0.46 ( 0.06)
1.36 ( 0.06)
1.37 ( 0.02)
CH3
a Compounds 3, 4, 6, 7, and 10 were assayed as their dihydrochloride salts.
b Values are means of four experiments, standard deviation is given in parentheses.
c Values are means of three experiments, standard deviation is given in parentheses.
d Strain JM109.10
e Strain SH1000.12
f Single diastereomer isolated after purification. Confirmed as trans (J C(1)H-C(2)H coupling constant = 9.5 Hz).
(Table 1). These are the first reported synthetic inhibi-
tors of both forms of UDP-GlcNAc enolpyruvyl trans-
ferase from S. aureus.
was maintained following removal of the aryl methoxy
groups (compound 6) or incorporation of 4-benzylpiper-
azine (compound 5) (Table 1). These findings led us to
investigate the importance of the ketone moiety to the
mode of action of these inhibitors. Hence, we synthesized
the des-carbonyl compound 10, from b-tetralone 8 and 1-
(2-hydroxylethyl)piperazine 9 as shown in Scheme 2.
Compound 10 failed to inhibit MurA and MurZ (Table
1) suggesting the ketone is crucial for inhibitory activity.
Next, we investigated whether an amine alpha to the ke-
To probe the key structural features responsible for inhi-
bition, analogues were synthesized by reaction of 2-bro-
motetralones 2a/b (obtained from bromination of
commercially available 6,7-dimethoxy-1-tetralone 1a
and a-tetralone 1b, respectively) with the appropriate
substituted piperazine (Scheme 1).14 Inhibitory activity

1732
C. J. Dunsmore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1730–1734
R3
O
O
O
N
R1
R2
R1
R2
Br
R1
R2
N
a
b
1a R1 = R2 = OMe
1b R1 = R2 = H
2a/b
3 R1 = R2 = OMe
R3 = CH2CH2OH
4 R1 = R2 = H
R3 = Me
5 R1 = R2 = OMe
R3 = CH2Ph
6 R1 = R2 = H
R3 = CH2CH2OH
c
7
Scheme 1. Synthesis of 2-aminotetralone derivatives 3–7. Reagents and conditions: (a) N-bromosuccinimide, amberlyst-15, EtOAc, rt, 24 h; (b) 4-R3-
piperazine, K2CO3, DMF, rt, 24 h; (c) NaBH4, THF, rt, 24 h.
OH
N
O
N
OH
a
HN
N
8
9
10
Scheme 2. Synthesis of 10. Reagents and conditions: (a) NaBH(OAc)3, AcOH, DCM, rt, 24 h.
0.22
tone was necessary for activity by synthesizing the cyclo-
hexyl derivative 12 by a previously published route
(Scheme 3).15 When tested in the MurA and MurZ assays,
12 did not inhibit the enzymes (Table 1) suggesting that
0.20
0.18
0.16
0.14
0.12
the a-aminoketone functionality is responsible for the
activity of this class of inhibitors.
0.10
0.08
0.06
To determine whether the binding mode of the amino-
tetralone inhibitors was reversible, a dilution assay was
0.04
0.02
0.0
performed. E. coli MurA was treated with 4, UDP-Glc-
NAc and PEP followed by a 50-fold dilution of the
-0.02
0
200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600
enzyme–inhibitor complex. Assuming a reversible mode
of action, there should be recovery of enzyme activity
upon dilution of the reaction mix. However, as can be
seen from Figure 2, the level of inhibition was
maintained over the course of an hour in a manner
similar to that of the covalent inhibitor, fosfomycin.
Hence, these results indicate a mode of action of these
inhibitors consistent with either covalent or very tight
non-covalent binding.
Time (s)
(A) MurA + no inhibitor (C) MurA + 4
(B) No enzyme + 4
(D) MurA + fosfomycin
Figure 2. Test for reversible inhibition of E. coli MurA by 4. No
recovery of enzyme activity over time after diluting out inhibitor
supports a tight-binding mode of inhibition possibly involving a
covalent adduct.
Since the mode of action of fosfomycin involves attack
by C115, we investigated if this residue was required
for inhibition by the 2-aminotetralone inhibitors. We
generated and purified the C115D MurA mutant pro-
tein16 of E. coli and treated it separately with 3, 4, and
fosfomycin. Although this mutant protein was found
to be functional and, as expected, resistant to fosfomy-
cin, no inhibition was observed with 3 and 4 at the high-
est concentration tested (120 lM) (data not shown).
These data support the hypothesis that C115 is involved
in the mechanism of these inhibitors, presumably
through attack by the thiol group on the ketone moiety
to generate a thiohemiketal adduct 13 (Fig. 3), which
have been shown to be formed by some cysteine prote-
ase inhibitors.17 The limited stability associated with this
type of thiohemiketal may account for our failure to
However, attempts to detect a covalent adduct between
MurA and 4 using electrospray mass spectrometry were
unsuccessful, under conditions that did detect an adduct
between MurA and fosfomycin (data not shown). If the
covalent adduct did form in the case of 4, it was not sta-
ble to the conditions of mass spectrometry.
O
O
O
Cl
a
b
1b
11
12
Scheme 3. Synthesis of 12.15 Reagents and conditions: (a) N-chloro-
succinimide, amberlyst-15, EtOAc, rt, 24 h; (b) c-C6H11MgCl + ZnCl2,
Cu(acac)2, Et2O, rt, 24 h.

C. J. Dunsmore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1730–1734
1733
supporting the interpretation that the 2-aminotetralones
are not ‘promiscuous’ inhibitors. These findings suggest
that inhibition of MurA/MurZ is an important factor in
the antimicrobial action of these inhibitors although fur-
ther studies are required in order to establish whether
this inhibition is the primary mode of antibacterial activ-
ity associated with these inhibitors.
HS
Cys
Cys
N
O
N
N
HO
S
N
4
13
Figure 3. Proposed mechanism of inhibition for 2-aminotetralone
inhibitors.
In summary, we have discovered a new series of inhibi-
tors of MurA based around a 2-aminotetralone motif.
In addition, these inhibitors also displayed activity
against MurA and MurZ from S. aureus, the first time
these enzymes have been targeted by synthetic inhibi-
tors. These results continue to provide insight into the
design of new antibacterial agents.
detect an adduct using MS. Indeed, rearrangement of
such adducts has been previously proposed for certain
cysteine protease inhibitors derived from halomethyl
ketones.18
Using the docking program eHiTS19 together with the
X-ray crystal structure of E. coli MurA9 (pdb code:
1UAE), 4 was docked into the enzyme active site
(Fig. 4). As illustrated in Figure 4, 4 is predicted to bind
close to the key C115 residue. Since the loop that con-
tains C115 is known to exhibit a high degree of flexibil-
ity,20 the cysteine residue should be capable of
approaching the ketone prior to attack. Furthermore,
the ketone is predicted to make a hydrogen-bond to
the conserved active site residue R120, which would be
expected to facilitate thiohemiketal formation.
Acknowledgments
The authors gratefully acknowledge the scientific contri-
butions of Peter Kotsonis, Carl Rollins, Greg Chenail,
Chad Vickers, Gejing Deng, Lac Lee, Meena Sachdeva,
Daniel Wall, and Brian Granda at Novartis.
References and notes
1. Russell, A. D.; Chopra, I. Understanding Antibacterial
Action and Resistance, 2nd ed.; Ellis Horwood: London,
1996.
2. Green, D. W. Expert Opin. Ther. Targets 2002, 6, 1.
3. Brown, E. D.; Vivas, E. I.; Walsh, C. T.; Kolter, R.
J. Bacteriol. 1995, 177, 4194.
Antimicrobial activities were tested according to the
Clinical Laboratory Standards Institute (CLSI)
protocol.21
Of the compounds that showed in vitro activity against
MurA and MurZ, 3, 4, 5, and 6 had MICs of 128, 64, 64,
and 8 lg/ml, respectively, against S. aureus (Table 1). In
addition, 4 and 6 had an MIC of 128 lg/ml against
E. coli.
4. Du, W.; Brown, J. R.; Sylvester, D. R.; Huang, J.;
Chalker, A. F.; So, C. Y.; Holmes, D. J.; Payne, D. J.;
Wallis, N. G. J. Bacteriol. 2000, 182, 4146.
5. Chopra, I. Expert Rev. Anti-Infect. Ther. 2003, 1, 45.
6. Zoeiby, A. E.; Sanschagrin, F.; Levesque, R. C. Mol.
Microbiol. 2003, 47, 1.
7. Kahan, F. M.; Kahan, J. S.; Cassidy, P. J.; Kroop, H.
Ann. N.Y. Acad. Sci. 1974, 235, 364.
8. Marquardt, J. L.; Brown, E. D.; Lane, W. S.; Haley, T.
M.; Ichikawa, Y.; Wong, C. H.; Walsh, C. T. Biochemistry
1994, 33, 10646.
Selectivity assays were performed using malate dehydro-
genase (MDH) and chymotrypsin.22 Those compounds
that inhibited E. coli MurA (3–6) showed no activity
against either MDH or chymotrypsin (data not shown),
9. Skarzynski, T.; Mistry, A.; Wonacott, A.; Hutchinson, S.
E.; Kelly, V. A.; Duncan, K. Structure 1996, 4, 1465.
10. JM109 strain: Yanisch-Perron, C.; Vieira, J.; Messing, J.
Gene 1985, 33, 103.
11. MurA activity was monitored by the detection of inor-
ganic phosphate generated during the reaction based on
the colourimetric malachite green method described by
Marquadt, J.L.; Siegele, D.A.; Walsh, C.T. J. Bacteriol.
1992, 174, 5748. Inhibition of MurA was measured
according to the method of Baum, E.Z.; Montenegro,
D.A.; Licata, L.; Turchi, I.; Webb, G.C.; Foleno, B.D.;
Bush, K. Antimicrob. Agents Chemother. 2001, 45, 3182.
All assays were carried out at 25 ꢁC. To measure MurA
inhibition, a 2-fold dilution series of the inhibitor (15 lL)
was added to 33.3 lg/mL MurA preincubated for 10 min.
in 50 mM Tris–HCl buffer (pH 7.8) containing 187.5 lM
UDP-N-acetylglucosamine (20 lL). The reaction was
started with the addition of 15 lL of 66.7 lM phospho-
enolpyruvate and the reaction mixture was incubated for
30 min. 50 lL malachite green-molybdate (0.045% w/v
malachite green hydrochloride and 4.2% w/v ammonium
molybdate in 4 N HCl mixed in a 3:1 ratio for 30 min and
R91
R120
C115
Figure 4. Inhibitor 4 docked into the active site of E. coli MurA. The
keto-carbonyl group shown to be essential for inhibition lies in close
proximity to the key C115 residue of the enzyme.

1734
C. J. Dunsmore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1730–1734
then passed through a Millipore Steritop 0.22 mm bottle
reduced in vacuo to give the crude product which was
purified by flash chromatography on silica gel using
petroleum ether (40–60 ꢁC)/EtOAc (4:1) as eluent. 2-
Bromo-6,7-dimethoxy-1-tetralone (2.10 mmol) 2a (Scheme
1) was dissolved in anhydrous DMF (20 ml) and anhy-
drous potassium carbonate (2.10 mmol) added. A solution
of 1-(2-hydroxyethyl)piperazine (2.10 mmol) 9 in anhy-
drous DMF (10 ml) was then added dropwise with
stirring. The reaction was stirred at room temperature
under dry N2 for 24 h. The mixture was poured into ice
water (100 ml) and extracted with DCM (3 · 75 ml). The
organic phase was exhaustively washed with water, dried
(MgSO4), and reduced in vacuo to give the crude product
which was purified using flash chromatography on silica
gel using DCM/MeOH/NH3 (90:9:1) as eluent. The
purified product was dissolved in DCM (10 ml), cooled
to 0 ꢁC, and treated dropwise with ethereal HCl (1 M,
1.5 ml). The solvent was removed in vacuo and the salt
recrystallised from ethyl acetate/MeOH to afford pure 3.
mp = 172–173 ꢁC; 1H NMR (300 MHz, D2O) d 7.11 (s,
1H), 6.59 (s, 1H), 4.22 (m, 1H), 3.66 (m, 3H), 3.58 (m, 3H),
3.53 (m, 6H), 3.42 (m, 4H), 3.17 (m, 2H), 2.84 (m, 2H),
2.29 (m, 1H), 2.01 (m, 1H); 13C NMR (75 MHz, D2O) d
191.1, 155.0, 147.8, 140.8, 123.6, 111.1, 108.9, 69.2, 58.5,
56.4, 56.1, 55.2, 49.7, 46.5, 27.5, 23.7; HRMS (ES+) m/z
335.1959 (MH+); C18H26N2O4H+ requires 335.1965; LC-
MS (20–95% MeCN) tR 2.88 min (m/z 335.1 (MH+),
100%).
top filter unit) was added and incubated for 10 min.
Absorbance at 660 nm was then measured with a Molec-
ular Devices SpectraMax 250 plate reader. An accompa-
nying series of inorganic phosphate standards was run for
each plate. It was established that under these conditions,
the production of phosphate versus time was linear.
12. SH1000 strain: Horsburgh, M. J.; Aish, J. L.; White, I. J.;
Shaw, L.; Lithgow, J. K.; Foster, S. J. J. Bacteriol. 2002,
184, 5457.
13. The murA and murZ genes from Staphylococcus aureus
SH1000 were cloned into pET-26b for expression with a
C-terminal His Tag. Constructs were used to transform
E. coli BL21 Star (DE3) cells for protein expression.
Expression of murA was induced with 0.1 mM isopropyl-
b-D-thiogalactopyranoside (IPTG) over 24 h with shaking
at 18 ꢁC. The expression of murZ was achieved using the
autoinduction media described by Studier, F.W. Protein
Expression & Purification 2005 41, 207. The poly(His)-
tagged MurA and MurZ were purified from the soluble
protein extract by affinity chromatography to Co2+ resin
(Talon resin, Clontech). The Co2+ resin was equilibrated
with buffer containing 50 mM sodium phosphate, 300 mM
NaCl and 0.5 mM THP. The resin and soluble protein
were incubated on ice with gentle shaking for 3 h, then
applied to 25 ml Cell Thru Disposable Columns (BD
Biosciences). The resin was washed with 200 ml of 50 mM
sodium phosphate, 300 mM NaCl, 20 mM imidazole.
Protein was eluted using 300 mM imidazole. Eluted
protein was then incubated with 1 mM UDP-N-acetylglu-
cosamine (UDP-GlcNAc) for 20 min at 4 ꢁC with shaking.
Imidazole and UDP-GlcNAc were removed by dialysis
into size exclusion buffer containing 50 mM HEPES (pH
7.5), 500 mM KCl, 3 mM dithiothreitol (DTT). MurA and
MurZ were further purified using a HiLoad 16/60 Super-
dex 200 preparation grade gel filtration column. Pooled
enzymes were dialysed into storage buffer containing
50 mM HEPES (pH 7.5), 500 mM KCl, 3 mM DTT and
50% glycerol and stored at ꢀ20 ꢁC.
15. Malosh, C. F.; Ready, J. M. J. Am. Chem. Soc. 2004, 126,
10240.
16. Kim, D. H.; Lees, W. J.; Kempsell, K. E.; Lane, W. S.;
Duncan, K.; Walsh, C. T. Biochemistry 1996, 35, 4923.
17. Powers, J. C.; Asgian, J. L.; Ekici, O. D.; James, K. E.
Chem. Rev. 2002, 102, 4639.
18. Drenth, J.; Kalk, K. H.; Swen, H. M. Biochemistry 1976,
15, 3731.
19. Zsoldos, Z.; Reid, D.; Simon, A.; Sadjad, B. S.; Johnson,
A. P. Curr. Protein Pept. Sci. 2006, 7, 421.
20. Klein, C. D.; Bachelier, A. J. Comput. Aided Mol. Des.
2006, 20, 621.
14. Typical procedure: To a solution of 6,7-dimethoxy-1-
tetralone (2.4 mmol) 1a (Scheme 1) in ethyl acetate (25 ml)
was added N-bromosuccinimide (2.5 mmol) followed by
Amberlyst-15ꢂ (0.75 g) (Ref.: Hershram, H.M.; Reddy,
P.N.; Sadashiv, K.; Yadav, J.S. Tetrahedron Lett. 2005,
46, 623). The reaction was stirred at room temperature for
24 h, then filtered and the catalyst washed with ethyl
acetate. The filtrate and washings were combined and
21. National Committee for Clinical Laboratory Standards.
Methods for dilution antimicrobial susceptibility tests for
bacteria that grow aerobically, 5th ed.; Approved standard
M7-A6. National Committee for Clinical Laboratory
Standards: Wayne, PA, 2003.
22. Seidler, J.; McGovern, S. L.; Doman, T. N.; Shoichet, B.
K. J. Med. Chem. 2003, 46, 4477.