Chemical, biochemical and microbiological properties of a brominated nitrovinylfuran with broad-spectrum antibacterial activity

Article abstract of DOI:10.1016/j.bmc.2012.11.018

A di-bromo substituted nitrovinylfuran with reported broad-spectrum antibacterial activity was found to be a potent inhibitor of MurA, a key enzyme in peptidoglycan biosynthesis. Further characterization of the compound was carried out to assess its reactivity towards thiol nucleophiles, its stability and degradation under aqueous conditions, inhibitory potential at other enzymes, and antibacterial and cytotoxic activity. Our results indicate that the nitrovinylfuran derivative is reactive towards cysteine residues in proteins, as demonstrated by the irreversible inhibition of MurA and bacterial methionine aminopeptidase. Experiments with proteins and model thiols indicate that the compound forms covalent adducts with SH groups and induces intermolecular disulfide bonds, with the intermediate formation of a monobromide derivative. The parent molecule as well as most of its breakdown products are potent antibiotics with MIC values below 4 μg/mL and are active against multiresistant strains such as methicillin-resistant Staphylococcus aureus (MRSA). Further development of the bromonitrovinyl scaffold towards antibiotics with clinical relevance, however, requires optimization of the antibiotic-cytotoxic selectivity profile.

Full text of DOI:10.1016/j.bmc.2012.11.018

Bioorganic & Medicinal Chemistry 21 (2013) 795–804
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Chemical, biochemical and microbiological properties of a brominated
nitrovinylfuran with broad-spectrum antibacterial activity
Therese Scholz a, , Carina L. Heyl a, , Dan Bernardi b, Stefan Zimmermann c, Lars Kattner b,
Christian D. Klein a,
⇑
a Medicinal Chemistry, Institute of Pharmacy and Molecular Biotechnology IPMB, Heidelberg University, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany
b Endotherm Life Science Molecules, Science-Park II, D-66123 Saarbrücken, Germany
c Department of Infectious Diseases, Medical Microbiology and Hygiene, Heidelberg University Hospital, Im Neuenheimer Feld 324, D-69120 Heidelberg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
A di-bromo substituted nitrovinylfuran with reported broad-spectrum antibacterial activity was found to
be a potent inhibitor of MurA, a key enzyme in peptidoglycan biosynthesis. Further characterization of
the compound was carried out to assess its reactivity towards thiol nucleophiles, its stability and degra-
dation under aqueous conditions, inhibitory potential at other enzymes, and antibacterial and cytotoxic
activity. Our results indicate that the nitrovinylfuran derivative is reactive towards cysteine residues in
proteins, as demonstrated by the irreversible inhibition of MurA and bacterial methionine aminopepti-
dase. Experiments with proteins and model thiols indicate that the compound forms covalent adducts
with SH groups and induces intermolecular disulfide bonds, with the intermediate formation of a
monobromide derivative. The parent molecule as well as most of its breakdown products are potent anti-
Received 18 August 2012
Revised 7 November 2012
Accepted 12 November 2012
Available online 24 November 2012
Keywords:
MurA enzyme
Nitrovinylfuran
Antibacterial agent
Inhibition
biotics with MIC values below 4 lg/mL and are active against multiresistant strains such as methicillin-
resistant Staphylococcus aureus (MRSA). Further development of the bromonitrovinyl scaffold towards
Stability
Degradation mechanism
Drug resistance
HPLC
antibiotics with clinical relevance, however, requires optimization of the antibiotic–cytotoxic selectivity
profile.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
sequence of reactions, the monomeric precursor UDP-N-
acetylmuramic acid pentapeptide (Scheme 1).
Resistance of bacteria against antibiotics is an increasingly
problematic clinical issue. The development of novel selective
antibacterial agents as well as the discovery of novel antibacterial
targets is therefore urgently and constantly required.1,2 The cell
wall is an essential and exclusive structural element of prokaryotes
and therefore represents an ideal target for the discovery and
development of novel antibiotics. Peptidoglycan—the macromolec-
ular component of the bacterial cell wall—consists of linear glycan
chains that are cross-linked by short peptides.3 Penicillins and
glycopeptides interfere with the final steps of cell wall synthesis.
The earlier steps of the peptidoglycan biosynthetic pathway,
however, provide further attractive targets for the development
of antibacterial compounds. The cytoplasmic steps in peptidogly-
can synthesis are catalyzed by a series of enzymes that form, in a
Because all enzymes involved in this pathway are essential to
the vital function of the prokaryotic cell, the inhibition of one or
more of the ‘Mur’ enzymes has a lethal effect on bacteria. Ap-
proaches in targeting this multi-step pathway have been compiled
by El Zoeiby et al.4
One inhibitor of the Mur-enzyme biosynthetic pathway has ac-
quired clinical importance: The epoxide fosfomycin is a covalent
and irreversible inhibitor of MurA, the enzyme that catalyzes the
first biosynthetic step. Fosfomycin irreversibly alkylates a cysteine
residue (Cys115 in Escherichia coli MurA) in the active site of
the enzyme. This cysteine residue is of critical relevance for the
catalytic activity of MurA.5–8 Our own efforts in this field led to
the identification of the sesquiterpene lactone cnicin as potent,
non-covalent suicide inhibitor of MurA.9 X-ray structure analysis
revealed the formation of a covalent adduct between the substrate
UNAG and cnicin. A similar binding mode may be relevant for some
natural products from tulips that are structurally related to the cni-
cin side chain.10
Abbreviations: BSA, bovine serum albumine; EP-UNAG, enolpyruvyl-UDP-N-
acetylglucosamine; GSH, glutathione; GSSG, oxidized glutathione; MH medium,
Müller-Hinton medium; PEP, phosphoenolpyruvate; UNAG, UDP-N-acetyl
glucosamine.
Nitrovinylfuran derivatives have been known as antimicrobials
for a long time.11,12 Some of them were evaluated for clinical and
industrial applications.13–18 Nitrovinylfurans with exocyclic nitro
function have been demonstrated to be non-mutagenic in contrast
⇑
Corresponding author. Tel.: +49 6221 544875; fax: +49 6221 546430.
E-mail address: c.klein@uni-heidelberg.de (C.D. Klein).
These two authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.11.018

796
T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
Scheme 1. The cytoplasmic steps in peptidoglycan biosynthesis are catalyzed by the enzymes MurA–F (Scheme for gram-negative bacteria).
to classical 5-nitrofurans.11,19–29 Our interest in this class of
compounds (Scheme 2) was spawned by the published reports
on their broad-spectrum antibacterial and antifungal proper-
ties.11,13,14,16,30–33 Related b-nitrostyrene derivatives are active
against Gram-positive and Gram-negative bacteria.34 The activity
of b-nitrostyrenes against bacteria and fungi, including methicillin-
resistant Staphylococcus aureus (MRSA) and vancomycin-resistant
enterococci (VRE), is currently still under investigation.33,35,36
Among the nitrovinylfurans, 2-bromo-5-(2-bromo-2-nitrovi-
nyl)furan (compound 1) has gained special attention. The biologi-
cal properties of 1 have been described before.13,16,30,31 The
patent specification reports biological activity in vitro and
in vivo, including broad-spectrum antibacterial and antifungal
properties, and a phase 3 study of a topical preparation. Studies
on the stability have also been documented.37 Compound 1 has
been described to be less genotoxic than 5-nitrofuran deriva-
tives21,25,26,38, though Ramos et al. report mutagenic effects.39
The indication spectrum of nitrovinylfuran derivatives was
extended to include leishmaniasis and trypanosomiasis.40
The antibacterial mechanism of action of 1 remains unclear. The
inhibition of bioenergetic processes, such as glycolysis, due to
chemical modification of functional thiol groups in enzymes has
been proposed as the basis of the antimicrobial and cytotoxic activ-
ity of vinylfurans.11,41–45 Inhibition of glyceraldehyde-3-phosphate
dehydrogenase, glucose-6-phosphate dehydrogenase, malate
dehydrogenase and glutathione reductase has been shown
in vitro and the modification of enzyme thiol groups was shown
with purified mammalian glyceraldehyde-3-phosphate dehydro-
genase.44,45 The addition of thiols to the exocyclic double bond of
vinylfurans was demonstrated with model thiols.41,45–48
Given the sensitivity of MurA to alkylation of the catalytically
relevant Cys115 residue, along with the antibacterial activity of
the nitrovinylfurans and their propensity to bind to thiol groups,
we decided to determine whether MurA could be a target of this
compound class. In the course of these investigations, we also
studied the stability of 1 in aqueous media, its reactivity towards
model nucleophiles such as glutathione, and the corresponding
degradation and reaction products.
2. Results and discussion
2.1. Bioactivity
2.1.1. Enzyme inhibition
Inhibitory effects of 1 and its degradation products were
evaluated for several MurA species, including E. coli, Pseudomonas
aeruginosa and Staphylococcus aureus (isoform 2). Table 1 com-
prises enzyme inhibition data, as well as the in vitro antibacterial
data, expressed as minimum inhibitory concentrations (MIC). IC50
values of cytotoxicity against various cell lines complement the
profiling of the compounds. Data related to the known E. coli MurA
inhibitory agents fosfomycin and cnicin are given as references.
Compound 1 is a potent inhibitor of the MurA enzymes, with
an IC50 of 2.8 ( 0.2)
lM for the E. coli MurA enzyme (Fig. 1).
Compound 1 is also a potent inhibitor of E. coli MetAP (methionine
aminopeptidase), but not of the human MetAPs and the two serine
proteases (thrombin, Dengue NS2B-NS3). The nitrovinyl moiety
in 1 and several congeners could be responsible for a Michael-
addition to nucleophilic structures in the target proteins. In
E. coli MurA, the Cys115 residue is solvent-exposed in the
unliganded state and is attacked by the epoxide electrophile of
the antibiotic fosfomycin.49 In E. coli MetAP, the active site also
contains exposed cysteine residues.50,51
To study the reactivity of the compounds towards Cys115 in
MurA, we compared the inhibitory effects on the C115D MurA mu-
tant. This mutant is catalytically active but resistant to the antibac-
terial agent fosfomycin as it lacks the Cys115 nucleophile and is
therefore not reactive towards electrophilic attack at this site.7
The results (cf. Fig. 1) indicate a somewhat reduced but still signif-
icant inhibition of the mutant enzyme with an IC50 of 18.7 ( 4.8)
lM for 1. Cys115 is therefore not the only target site. The different
steepness of the dose–response curves of 1 for native and C115D
MurA indicates a particular involvement of the geminal bromoni-
trovinyl moiety and Cys115 in the binding mode (Fig. 1). This effect
is under further investigation.
The de-bromo derivative 2 was significantly less potent against
all tested enzymes. Interestingly, the IC50 values of 2 at the various
MurA enzymes, including the C115D mutant, are nearly identical.
In contrast, inhibition by the parent compound 1 was less pro-
nounced for the mutant lacking the active site thiolate.
The aldehyde 3 is inactive against the enzymes and has neither
antibacterial nor cytotoxic activity, whereas bromonitromethane 4
shows significant inhibition of the MurA enzymes and antibacterial
activity. The antibiotic effect of 4 is known34,53–55 and its use as
Scheme 2. The nitrovinylfuran 1, its degradation pathways and breakdown
products.

T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
797
Table 1
Enzymatic, antibacterial and cytotoxicity profile of compound 1 and its breakdown products
Compound
1
2
3
4
5
Fosfomycin
Cnicin
IC50 (lM) on MurA species
E. coli MurA
E. coli MurA C115D
Ps. aerug. MurA
S. aureus MurA2
2.8 ( 0.2)
18.7 ( 4.8)
2.8 ( 0.3)
3.9 ( 0.3)
27.3 ( 4.4)
23.8 ( 7.2)
20.3 ( 1.9)
20.7 ( 1.6)
n.i.a
n.i.
n.i.
n.i.
4.6 ( 2.8)
7.8 ( 3.3)
4.0 ( 0.5)
6.2 ( 0.1)
n.i.
n.i.
n.i.
n.i.
0.1e ( 0.01)
n.i.e
16.7e ( 0.7)
e
n.i.
10.5e ( 0.5)
1.1 ( 0.1)
10.3e ( 1.5)
20.4 ( 1.1)
Other enzymes (% inh)c
E. coli MetAP
Hs MetAP 1
Hs MetAP 2
Thrombin
85 ( 3)
n.i.
13 ( 4)
8 ( 3)
12 ( 2)
30 ( 9)
n.i.
7 ( 2)
19 ( 5)
26 ( 3)
n.i.
n.i.
n.i.
n.i.
n.i.
85 ( 2)
n.i.
22 ( 1)
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
9 ( 1)
n.i.
n.i.
Dengue NS2B-NS3
n.i.
MIC (lg/mL)
Gram-positive
MSSA (ATCC 25923)
MRSA (NCTC 10442)
4
4
2
2
>32
>32
2
2
8
>32
4
4
>32
>32
Gram-negative
E. coli (ATCC 25922)
Ps. aerug. (ATCC 27853)
4
2
2
4
>32
>32
1
0.5
32
8
2
1
>32
>32
Cytotoxicity IC50 (lM)
Non cancer cell lines
HaCaT
Flow
33
9
15
6
P100
P100
25
14
P100
P100
P100
P100
80
82
Cancer cell lines
A-431
MeWo
Calu-6
CaSki
24
43
10
50
36
47
12
21
5
18
19
22
P100
P100
P100
P100
P100
P100
38
40
23
99
69
73
P100
P100
P100
P100
P100
P100
P100
P100
P100
P100
P100
P100
27
P100
44
P100
93
P100
Capan-1
SW-707
t1/2 (h)d
Tris buffer, pH 7.8
Citrate buffer, pH 6
0.5
6
58
116
n.d.b
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
GSH-reactivity
GSH-adduct formation
(addition)
Oxidation to GSSG
+
+
+
ꢀ
ꢀ
ꢀ
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
ꢀ
+
Enzyme concentrations were 12 nM for the native MurA and C115D mutant and S. aureus MurA2. Ps. aeruginosa MurA was assessed at 150 nM. Other enzyme concentrations
were: MetAP enzymes 500 nM, Thrombin 10 nM, Dengue NS2B-NS3 100 nM. Compound concentration for determining percentage inhibition of investigated enzymes were
as follows: MetAP assay 10
l
M, Thrombin assay 25
lM and DV NS2B-NS3 assay 50
lM.
a
No inhibition
Not determined
Percentage inhibition
Half-life
b
c
d
e
Bachelier et al. (2006).52
biocide has been patented,13,56 but a mechanism of action has not
been proposed yet. Considering the structure–activity relation-
ships presented here, it appears that the bromonitromethyl
2.1.1.1. Time-dependent enzyme inhibition.
Reversible and
irreversible binding modes can be distinguished by varying
the pre-incubation period of inhibitor and enzyme. For a
reversible inhibitor, the inhibitory potency is independent of the
pre-incubation time, whereas irreversible inhibitors appear as
increasingly potent if the incubation time is prolonged.
moiety, either isolated or in conjunction with
a vinylfuran
structure, is essential for the antibacterial and MurA inhibition
properties of this compound class.
In order to obtain information on the target-selectivity of these
compounds, we screened their activity against other enzymes. No
compound was active against the serine proteases thrombin and
Dengue virus NS2B-NS3 and the two human methionine amino-
peptidases MetAP I and II. In contrast, the methionine aminopepti-
dase from E. coli is highly sensitive towards 1 and 4. Bacterial
MetAPs are potential targets for antibacterial drug discovery. The
bromonitromethyl moiety may therefore be considered as a
fragment in the design of inhibitors of bacterial MetAPs.
For compound 1 the enzyme inhibition increased with preincu-
bation time, in analogy to the irreversible MurA inhibitor fosfomy-
cin (Fig. 2), indicating an irreversible mode of action.
2.1.1.2. UNAG dependency.
Inhibitor binding to MurA often
depends on the presence of the substrate UNAG.52,57 UNAG triggers
a
conformational change towards a ‘closed’ conformation of
MurA.58 Thus, augmented inhibitory potency in the presence
of UNAG indicates that the inhibitor binds to the ‘closed’

798
T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
Figure 3. The inhibitory effect of fosfomycin, but not of compound 1 depends on
the presence of UNAG.
Figure 1. Dose–response curves for native E. coli MurA (circle) and C115D mutant
(triangle) upon incubation with 1.
sensitive towards these agents, indicating that the compounds
were not inactivated by the resistance mechanisms of that strain.
Antibacterial efficacy of 1 against resistant strains was already
noted and hence an orthogonal mechanism of action was pre-
sumed, although no detailed analysis of the phenomenon was
made.31 Antibacterial potency was also striking for 2, although it
was considerably less potent than 1 as MurA inhibitor. This clearly
indicates that MurA is not the only target of the nitrovinylfuran
derivatives. Results of our examination are in good accordance
with the antibacterial susceptibilities already reported else-
where.16,31 The antibacterial activity of the decomposition
products 3 and 5 was not significant. Thus, the bromonitromethyl
element is essential for antibacterial activity and enzyme
inhibition.
2.1.3. Cytotoxicity
Profiling of the compounds under study was completed by
cytotoxicity data on several cancer and non-cancer cell lines. All
compounds which were active on the enzymatic and bacterial le-
vel also displayed cytotoxic effects with no significant selectivity
towards tumorigenic cell lines (Table 1). The cytotoxic effects of 1
and 2 was more pronounced than that of the halo-nitro com-
pound 4. For 3 and 5 no cytotoxic properties were observed, in
analogy to the lack of activity in enzymatic or bacterial tests.
Interestingly, the cytotoxic action of the de-bromo compound 2
was greater than for the parent molecule 1 on all investigated cell
lines. Cytotoxic effects of 2 on cultured human lymphocytes have
Figure 2. Time-dependent inhibition of E. coli MurA by
reference inhibitor.
1 and fosfomycin as
conformation. Additionally, MurA isolated from E. coli contains
PEP bound covalently to Cys115,59 which can be liberated from
the enzyme by addition of UNAG, leading to the formation of
the reaction product EP-UNAG. Thus, if the binding site of an inhib-
itor overlaps with the PEP binding site, the effect of UNAG would
be to liberate PEP, thereby opening access for the inhibitory
compound.
been reported at 15 lg/mL and the reduction of cytotoxic events
The binding of fosfomycin is clearly, and absolutely, dependent
on the presence of UNAG. This effect is not present for 1. Figure 3
shows the results of an experiment where fosfomycin and 1 were
pre-incubated with MurA in the absence and presence of UNAG.
Whereas fosfomycin remains inactive in absence of substrate
UNAG, compound 1 binds in a way that is independent of UNAG.
These results show that 1 does not require the UNAG-triggered
closure of the MurA active site for binding.
due to exogenous metabolic inactivation by S9 fraction proteins
was demonstrated.23 In a study using a human lymphoblastoid
cell line viability was >70% at the highest tested dose which
was 1
investigations demonstrate a low cytotoxicity for 1 in mouse
spleen macrophages (cytotoxicity 27% at 10 g/mL resp.
34
M).60 In a comparative study, the nitrostyrene scaffold was
lg/mL for 1 (3.4 lM) and 10 lg/mL for 2 (46 l
M).27 Recent
l
l
demonstrated to be twofold more cytotoxic than unsubstituted
nitrovinylfuran.61 In contrast, the reference compounds fosfomy-
cin and cnicin did not display cytotoxic effects at the highest con-
centration tested. Nitrofurantoin, as classical nitrofuran derivative
with the nitro group attached to the furan ring, did not exhibit
cytotoxic effects on the investigated cell lines at the highest dose
(data not shown) in accordance with the documented need for
metabolic activation.62–64 The therapeutic window between anti-
biotic and cytotoxic effects is limited for all tested compounds.
There is no selectivity between cancerous and non-cancerous cell
lines.
2.1.2. Antimicrobial activity
Encouraged by the enzyme inhibition data, the antibacterial
properties of the compounds against selected strains of Gram-
positive and Gram-negative bacteria were determined (Table 1).
Compounds that exerted inhibitory activity on the isolated MurA
enzymes, for example, 1, 2 and 4 had antibacterial MIC (minimal
inhibitory concentration) values of 4
clinical isolate with highly problematic resistance profile
(methicillin-resistant Staphylococcus aureus, MRSA) was equally
lg/mL or less. Interestingly,
a

T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
799
2.2. Compound stability and reactivity
2.2.2. Reactivity of compound 1 and derivatives towards
glutathione
2.2.1. Nitrovinylfuran derivatives decompose to aldehydes in
aqueous media
To investigate the reactivity of nitrovinylfuran 1 and its break-
down products towards thiols, we used glutathione (GSH) as a
model sulphur nucleophile to simulate the SH groups of target pro-
teins. The reactivity of 1 towards thiols has not been studied be-
fore. We studied the reaction between 1 and GSH in equal
concentrations at pH 6. The samples were analyzed by HPLC-UV
and HPLC-ESI-MS. Reactivity towards thiol-nucleophiles has been
analyzed for b-nitrostyrenes and several vinylfurans and the thio-
ether conjugate has been the reported product.41,45–47,65
Surprisingly, 1 turned out to react with GSH in a different man-
ner. The major reaction product is the de-bromo derivative 2, and
not the expected GSH-adduct. The product was verified by isola-
tion and 1H NMR and APCI-MS-analysis of 2 from the reaction of
1 and GSH in citrate buffer (data not shown). In contrast to the deg-
radation of 1 in citrate buffer, the incubation with GSH in the same
buffer does not lead directly to the formation of 3. HPLC-UV-ESI-
MS analysis shows, besides 2, the formation of the GSH-adducts
of 1 and the released 2 (Fig. 4).
The instability of 1 in aqueous media has been reported before.
The decomposition was studied by TLC and UV spectroscopy and
the monobromide derivative 2-bromo-5-(2-nitrovinyl)furan 2 and
5-bromofurfural 3 were the reported degradation products.30,37
We further investigated the stability of parent molecule 1 in
aqueous solution by HPLC and calculated half-lives of the com-
pounds (Table 1). This confirmed the formation of 3 out of 1 with
concomitant release of bromonitromethane 4, but the formation
of 2 could not be verified. Reference chromatograms of commer-
cially available furan-aldehydes showed that all studied nitrovinyl
derivatives decompose to the corresponding aldehyde under aque-
ous conditions (Scheme 3).
In all cases, the decrease of the starting material follows a first
order reaction rate law with logarithm of concentration plotted
versus time resulting in a linear slope. The half-lives can be calcu-
lated from this slope. The geminal bromonitro derivative show
much shorter half-lives than the corresponding nitrovinyl com-
pounds, for example, 29 min for 1, but 58 h for 2.
2.2.3. Compound 1 promotes glutathione oxidation
For compounds 1 and 2, the degradation was also studied at pH
6 and in Müller-Hinton (MH) medium, pH 7.2 (data not shown). At
pH 6, all compounds show longer half-lives compared to pH 7.8
(e.g., 6 h for 1). In MH medium, we observed a time-dependent dis-
appearance of the starting materials, but were unable to detect de-
fined degradation products. This is probably due to a reaction of
either the starting materials or the released derivatives with mac-
romolecular constituents of the MH medium. Compared to Tris
buffer at pH 7.8, the disappearance of the starting material in
MH medium is much faster (e.g., 8 min for 1).
The half-lives for 1 lie within the incubation periods of the per-
formed bioassays. This means—apart from any possible influence
of the enzyme—that a large quantity of the molecules decomposes
within the assay time span to give the respective furaldehyde.
However, this decomposition product has been shown to be inac-
tive against the isolated enzyme and to have insignificant antibi-
otic efficacy. The presumed co-product of this degradation route,
however, would be bromonitromethane 4, for which potent inhib-
itory efficacy was demonstrated (Table 1). In any case, the release
of 4 as active component can not be the only mechanism of en-
zyme inhibition, because the nitrovinyl derivative 2—which re-
leases inactive nitromethane 5—also shows significant MurA
inhibition.
Because a shift of pH to slightly acidic conditions provokes a
clear increase in compound stability, the relevance of these find-
ings was monitored for the determination of bacterial susceptibil-
ity. MIC determination was repeated using MH medium which had
been previously adjusted to pH 5.8. At lower pH (change from 7.2
to 5.8) the half-life of 1 increases from 8 min to 60 min. However,
the MICs were not affected by the pH-change of the medium.
While the compound is more stable under acidic conditions, cellu-
lar uptake or binding to the target site might also be hindered at
lower pH.
As the conversion of 1–2 represents a reduction reaction, we
examined whether oxidized glutathione (GSSG) is formed during
the incubation of 1 and GSH. As shown in Figure 5, formation of
GSSG occurs.
Scheme 4 shows the proposed mechanism for the reaction be-
tween 1 and GSH. Oxidation of protein thiol groups could be an-
other possible mechanism of action for 1 at the studied enzymes.
Formation of intramolecular, non-native disulfide bonds of ac-
tive site cysteines, resulting in conformational changes as enzyme
inactivating mechanism has already been suggested for the ageing
of E. coli MetAP and could account for the observed inhibitory ef-
fects of 1 and 4 on this enzyme.66 However, other enzymes that
were investigated in the course of this study were resistant to 1
and 4 because they do not harbour sensitive cysteine residues.
Table 1 includes an overview of the observed reaction products
of investigated substances with GSH. Oxidation of GSH was also
demonstrated for the decomposition product 4. In contrast, 2
forms the GSH-conjugate without concomitant oxidation product.
These results indicate that the geminal bromonitro feature is
responsible for the observed oxidative processes. In accordance
to the results in the enzymatic assays, compounds 3 and 5 were
inactive towards GSH.
The interference of compounds 1, 2 and 4 with thiol nucleo-
philes in proteins can have severe consequences for the vitality
of prokaryotic organisms. Similar observations have been made
previously for antibiotic compounds that contain related structural
elements: Oxidative modification of cysteine residues has been de-
scribed as the mechanism of action of the broad-spectrum antibi-
otic nitrofurantoin.67,68 Depletion of GSH pool levels as
consequence of the oxidative stress has been linked to eukaryotic
toxicity of nitrofurantoin.69–71 Thiol protein oxidation has also
been observed for the preservative 2-bromo-2-nitropropane-1,3-
diol, which is a close structural analog of the compounds described
here.72 We therefore conclude that the thiol-reactivity of com-
pounds 1 and 4 could provoke uncontrolled oxidative processes
comprising formation of disulfide bonds within and between vital
enzymes, thereby leading to cell damage.
2.2.4. Compound 1 interacts with MurA cysteines
To verify the proposed interaction between 1 and cysteine res-
idues of MurA, 2,20-dithiodipyridine (2-DPS) was used to quantify
free thiol groups of the enzyme. 2-DPS reacts with thiols in a
thiol-disulfide exchange and releases a stoichiometric amount
of the chromogenic 2-thiopyridone (2-TP), which can easily be
Scheme 3. Degradation of nitrovinylfuran derivatives to aldehydes. This decom-
position route was demonstrated for aqueous conditions and for the following
buffer systems: Tris pH 7.8; citrate pH 6.

800
T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
Figure 4. Chromatogram of 1 after 60 s incubation with GSH. Peak labelling: (A) GSH-adduct of 2, (B) GSH-adduct of 1, (C) compd 2, (D) compd 1. Corresponding ESI mass
spectra of peaks A and B. (A) GSH-adduct of 2 showing the characteristic Br1-substitution pattern and fragment molecules through HNO2-loss (m/z 478.0096, m/z 47.0022)
D
and sodium adduct (m/z 546.9902). Mass accuracy ꢀ0.4 ppm with respect to the calculated mass m/z 525.0120. (B) GSH-adduct of 1 showing the characteristic Br2-
substitution pattern and fragment molecule through HBr-loss (m/z 522.9960,
D
m/z 79.9254). Mass accuracy ꢀ1.3 ppm with respect to the calculated mass m/z 602.9225.
and mutant enzyme. This experiment provides another indication
that 1 interacts with cysteine residues in MurA besides the cata-
lytic C115. On the other hand, this experiment clearly points out
that enzyme inhibition is not promoted through 1:1 stoichiometry,
but involves several accessible cysteine residues.
2.2.5. MurA induces the debromination of compound 1
The reaction of 1 with GSH leads to formation of the de-bromo
derivative 2 and oxidized GSH. We studied the conversion of 1 in
the presence of MurA to examine whether cysteine oxidation
might play a role in enzyme inhibition. In order to rule out
non-specific transformation of 1 by the assay additive BSA (bo-
vine serum albumine) we conducted appropriate controls of com-
pound incubation in presence and absence of 0.1% BSA. Reaction
progress was followed by HPLC-UV. The presence of BSA has only
minor effects on the conversion of 1. We also showed that the en-
zyme substrates UNAG or PEP have no significant effects on the
conversion of 1 (data not shown). The main product of the degra-
dation in absence of MurA is 3, while the presence of 2 is near
detection limit (Fig. 7). In contrast, in the presence of MurA the
formation of the de-bromo derivative 2, which must occur via re-
dox reaction, can be detected, while only a small amount of 3 is
formed (Fig. 7). Additionally, the decomposition of 1 in solution
proceeds much faster in presence of the enzyme. The concentra-
tion of 1 is greatly diminished even at time point zero. The rapid
disappearance of 1 in solution indicates either a covalent binding
to the enzyme or an enzyme-catalysed transformation. The de-
crease of 2 can be attributed to a binding process to the enzyme,
because 2 is decomposing very slowly under assay conditions
(half-life 58 h) and there is no further rise in concentration of fur-
aldehyde 3.
Figure 5. Compound 1 promotes GSSG formation.
Scheme 4. Proposed mechanism for the redox reaction between 1 and GSH.
detected by HPLC. MurA was incubated with an equimolar
concentration of 1 for 30 min before adding 2,20-dithiodipyridine.
The amount of 2-TP released from an excess of 2-DPS was
compared to a control containing only MurA.
2.2.6. Compound 1 induces disulfide bond formation between
MurA and a fluorescent thiol
Figure 6 shows the experiment of SH group detection for 1 and
the reference compound fosfomycin upon incubation of equimolar
and excess substance/inhibitor ratios. The results are compared to
an experiment with the C115D mutant enzyme. For the reference
substance fosfomycin, the amount of detected thiol groups is not
further decreased under compound excess, correlating with the
C115-selectivity of this compound in a 1:1 stoichiometry. Upon
incubation with the C115D mutant, the amount of detected thiol
groups remains unchanged for fosfomycin compared to the control
without inhibitor, again correlating with the reported specificity of
fosfomycin for the active site nucleophile C115 and the resistance
of the C115D mutant towards fosfomycin. For 1 the amount of free
SH groups is reduced to a much larger extent when incubated in
equimolar ratio and decreases even more upon compound excess.
This decrease in detected SH groups is obvious for both wild-type
To determine whether the formation of disulfide bonds medi-
ated through 1 also occurs with the MurA enzyme (analogous to
the reaction with GSH), MurA was incubated with the fluorescent
coumarin derivative 6 in the absence and presence of 1. The sam-
ples were analyzed by SDS–PAGE and the fluorescent dye was visu-
alized under UV.
Figure 8 shows the formation of a fluorescent adduct with the
molecular weight of MurA in the sample containing 1 and 6 (lanes
2 and 8) according to the proposed reaction scheme (Scheme 5). A
minor band of this size can also be seen in lane 3, where 1 was
omitted. Addition of the thiol nucleophile DTT in reduced Laemmli
buffer (lane 4) reverses disulfide bond formation, coinciding with a
disappearance of this distinct band. This confirms that a disulfide
bond was initially formed between enzyme and dye.

T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
801
Figure 6. Quantification of free cysteines after treatment of MurA with fosfomycin and 1. HPLC-analysis of 2-TP released from native and C115D MurA.
Figure 7. Time course of compound 1 conversion into its degradation products 2 and 3 in presence and absence of E. coli MurA (100 lM).
3. Conclusion
Nitrovinylfuran and bromonitromethane derivatives possess a
significant antibacterial activity. We started to explore the
efficacy of these compounds as inhibitors of enzymes within
the early peptidoglycan biosynthetic pathway. Our data identify
the nitrovinylfuran and bromonitromethane scaffolds as inhibi-
tors of MurA from various species and E. coli MetAP. In particu-
lar, compounds 1 and 4 specifically target bacterial and not
human MetAPs. The observed activity on the enzymatic and
cellular level is induced by a complex interplay of 1 and its
breakdown products. Depending on environmental conditions,
the transformation of 1 to the aldehyde proceeds either directly
in aqueous media or through intermediate formation of the
de-bromo derivative
2
in presence of thiol nucleophiles
(Scheme 2). Reactivity of 1 and its descendants involves thiol
functionalities through addition reactions or oxidative processes,
including the formation of non-native disulfide bonds, thereby
impairing stability and catalytic properties of bacterial proteins.
Enzymatic, antibacterial, cytotoxic and GSH-reactivity data pre-
sented here indicate that neither MurA, nor the other targets
which were previously identified, are exclusively responsible
for the antibacterial effect of 1, but that this molecule exerts
pleiotropic effects on numerous proteins by interference with
cysteine residues.
Figure 8. SDS-PAGE of fluorescently labelled MurA. Excitation at 365 nm, emission
at 475 nm. Lane 1: Molecular weight marker (at visible light); the arrow indicates
the 47 kDa band corresponding to E. coli MurA (47.4 kDa). Intermolecular disulfide
bond formation of E. coli MurA and 6 is markedly increased in presence of 1.
The cytotoxic activity of the compounds against human cell
lines is significant and does not encourage the further development
of the compounds as antibacterial drugs. However, the bromoni-
trovinyl scaffold may be used for the design of inhibitors that spe-
cifically interact with cysteine residues in target proteins. This will
require a significant selectivity, which must be obtained by other,
non-covalent recognition elements.
Scheme 5. Fluorescent labelling of E. coli MurA with the coumarine derivative 6,
induced by redox reaction with 1.

802
T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
Müller-Hinton broth. Briefly, the samples were pipetted to 96-well
4. Experimental
4.1. Materials
microtiter plates in an appropriate medium followed by a twofold
serial dilution. Hundred microliters of microbial suspension was
added to give a final concentration of 5 ꢁ 105 cfu/mL. After incuba-
tion at 36 °C for 20–24 h, minimal inhibitory concentration (MIC)
was determined as the lowest concentration at which no growth
occurred.
Chemicals and solvents for HPLC were from Sigma–Aldrich
(Schnelldorf, Germany). 96-well plates, U-bottom (polystyrene),
and 96-well cell culture plates (CellstarÒ, sterile, F-bottom) were
purchased from Greiner Bio-One (Frickenhausen, Germany).
4.4. Cytotoxicity studies
Compounds
1 and 2 were synthezised according to the
procedures specified in the supporting information.
Compound stock solutions were 10 mM in DMSO. Fosfomycin
stock solution was prepared freshly in water. All investigated
enzymes were expressed in E. coli BL21 (k DE3) and purified by
chromatographic techniques according to procedures described
previously,52,73,74 except for thrombin which was obtained from
Sigma–Aldrich.
The cytotoxicity of the investigated compounds was deter-
mined using the following cell lines: HaCaT (human skin keratino-
cyte) and Flow 2002 (diploid human embryonic lung fibroblast) as
nontumorigenic cell lines; A-431 (human skin epidermoid carci-
noma), MeWo (human skin melanoma), Calu-6 (human lung ade-
nocarcinoma), CaSki (human cervical carcinoma), Capan-1
(human pancreas adenocarcinoma) and SW-707 (human colon car-
cinoma) as cancerous representatives. The assay was performed in
96-well plates with an inoculum of 3 ꢁ 104 cells/well, which were
seeded the day before treatment. Compound test solutions were
prepared freshly by diluting the stock solution 1:10 with the
4.2. Enzyme inhibition studies
Experiments were performed in triplicate with standard
deviations of average values generally 610%.
respective strain medium. Total assay volume was 100 lL with a
final DMSO concentration of 1% (v/v). Plates were incubated for
72 h at 37 °C, then deprived of medium and fixed with formalin
solution 3% (v/v). For quantification of cytotoxicity cells were
4.2.1. IC50 determination
Dose–response curves were established by measuring enzyme
activity in triplicate determinations of at least eight different
compound concentrations. Assays were performed as described
above. The resulting data were plotted using GraFit version
5.0.12 (Erithacus Software Ltd, Horley, Surrey, United Kingdom)
and IC50 values, representing the concentration of compound at
which the residual activities were 50%, were determined from a
four-parameter fit.
stained by adding 80 lL of crystal violet solution 0.5% (w/v) per
well. After removal of excessive dye and desiccation of the plates,
the optical density was measured at 590 nm using a BMG Labtech
Fluostar Galaxy multiplate reader. IC50 values were calculated from
eight different compound concentrations using GraFit version
5.0.12 (Erithacus Software Ltd, United Kingdom).
4.5. Chromatographic analysis (compound stability and
reactivity)
4.2.2. MurA assay
In a standard assay, MurA enzyme was pre-incubated with the
substrate UNAG and inhibitor for 10 min at 37 °C.52 To determine
the influence of the substrate UNAG on the binding process, exper-
iments were performed in absence of UNAG during pre-incubation.
The reaction was started by the addition of the second substrate
PEP, resulting in a total volume of 100 lL with the following con-
centrations: E. coli wild-type, C115D mutant MurA or S. aureus
MurA2 12 nM, Ps. aeruginosa MurA 150 nM, BSA 0.1% (v/v), UNAG
4.5.1. Instrumentation and chromatographic conditions
Samples of degradation studies and for reactivity analysis with
GSH or enzyme were analyzed by HPLC. The HPLC system con-
sisted of an Agilent 1200 series device (consisting of G1322A deg-
asser, G1311A quaternary pump, G1329A autosampler, G1316A
column oven and G1315A diode array detector controlled by
Chemstation Software, Agilent Technologies, Germany) equipped
with a Waters XTerraÒ C18 column (2.1 ꢁ 50 mm, particle size
250
lM, PEP 125
l
M, Tris 25 mM pH 7.8, DMSO 1% (v/v). The reac-
L of Lanzet-
tion was stopped after 60 min at 37 °C by adding 100
l
2.5 lm) unless otherwise noted. A linear gradient mobile phase
ta reagent, composed of malachite green solution (0.045% (w/v))
and ammonium heptamolybdate (8.4% (w/v) in 8 N HCl) in a ratio
of 4:1, and 0.03% (w/v) Tergitol NP-40 as dye-stabilizing detergent.
After 10 min the absorbance at 620 nm was measured using a BMG
Labtech Fluostar Galaxy (Ortenberg, Germany) multiplate reader to
quantify the released inorganic phosphate.
was adequate for all analytical separations and accomplished using
0.1% TFA in water (solvent A) and 0.1% TFA in acetonitrile (solvent
B) with a flow rate of 0.3 mL/min and a column temperature of
37 °C. A typical gradient elution profile started with 10% solvent
B for 1 min, was ramped linearly to 95% over 8 min and maintained
at 95% for 2 min. Injection volume was 10 lL unless indicated
otherwise. The column eluate was monitored by measuring the
absorbance as indicated within the respective experimental setup.
RP-LC/MS-analysis was performed using the same chromato-
graphic setup coupled to a micrOTOF-Q II (Bruker Daltonik,
Bremen, Germany) operating in electrospray negative ionization
mode. Chromatographic conditions were employed as mentioned,
except that the solvent additive TFA was replaced by formic acid
0.1% to allow for MS-compatibility. Reference masses of sodium
formate clusters were used for external calibration of each chro-
matographic run.
4.2.3. Other enzymes
Inhibitor screening on the Dengue virus NS2B-NS3 protease and
thrombin were conducted as reported previously.74,75 Activity on
the MetAP enzymes was assessed as described elsewhere.75
4.3. Microbiological studies—determination of MIC in
pathogenic bacteria
Antimicrobial activities were determined against S. aureus
(MSSA) ATCC 25923, methicillin-resistant S. aureus (MRSA) NCTC
10442, E. coli ATCC 25922 and Ps. aeruginosa ATCC 27853 strains.
The strains were sub-cultured on Columbia agar containing 5%
sheep blood (Becton Dickinson, Germany) 24 h prior to any antimi-
crobial test. Broth microdilution assays were performed according
to CLSI (Clinical and Laboratory Standards Institute, 2011) in
4.5.2. Half-life of nitrovinylfuran derivatives
The decomposition of nitrovinylfuran derivatives into corre-
sponding aldehydes was studied using HPLC-UV. Solutions of the
compounds (100
lM) in water as well as different buffers (sodium
citrate 50 mM pH 6, Tris 50 mM pH 7.8) and in Müller-Hinton

T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
803
medium (pH 7.2) were prepared and aliquots of 10
l
L of these
additive. After 1 h, a 10
lL aliquot was quenched with 10
l
L of
samples were injected into the HPLC system at different time
points. Disappearance of the nitrovinylfuran derivatives was
monitored at 390 nm. Formation of corresponding aldehydes was
observed at 290 nm. MicrosoftÒ Office Excel 2003 was used to
construct first order plots from the logarithm of peak area versus
time and the half-lives were calculated using the equation
non-reducing Laemmli buffer. Additionally, one sample containing
MurA and compounds 1 and 6 was quenched with reducing Lae-
mmli buffer. After heating to 95 °C for 3 min, 10 lL were analyzed
by SDS–PAGE (12%). Protein bands were analyzed using an Alpha-
Imager (Alpha Innotec, Kasendorf, Germany) gel documentation
system with an excitation wavelength of 365 nm to visualize pro-
tein-bound fluorescent 6. Compound 6 was synthesized from 6-
methoxy-coumarin-3-carboxylic acid and 4,40-dimethoxytrityl
protected cysteine methylester. Details are described in the Sup-
plementary data.
t1/2 = ln2/kobs
.
4.5.3. Reactivity of nitrovinylfuran derivatives towards
glutathione
In order to study the reactivity towards glutathione, 10 lL
inhibitor stock solution (10 mM in DMSO) and 10
solution (10 mM in H2O) were dissolved in 980 L sodium citrate
buffer (50 mM pH 6.0), resulting in final concentrations of
100 M. Aliquots (100 L) of the reaction mixture were withdrawn
at regular intervals (every 30 s over a period of 3 min, and at 4 and
5 min) and the reaction in these aliquots was stopped by quench-
l
L GSH stock
Acknowledgments
l
The authors thank Christoph Nitsche and Michael Wacker for
performing the screening assays on the serine proteases and
MetAP enzymes. This study was supported by a grant from the
German Bundesministerium für Wirtschaft und Technologie,
Pro-INNO-2, KF0647401UL8.
l
l
ing with
1 lL of orthophosphoric acid (85%). Samples were
analyzed by HPLC.
The presence of glutathione-conjugates was shown by RP-LC/
MS. To verify the formation of oxidized glutathione (GSSG), a
Merck ChromolithÒ Performance RP-18e column (4.6 ꢁ 100 mm)
was used. Elution started with 100% aqueous conditions for
3 min, followed by a linear gradient to 95% solvent B over 5 min
at a flow rate of 1 mL/min. Formation of GSSG could be detected
at 200 nm at a retention time of 3.0 min.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.11.018.
References and notes
1. Rice, L. B. Clin. Inf. Dis. 2008, 46, 491.
2. Silver, L. L. Clin. Microbiol. Rev. 2011, 24, 71.
3. van Heijenoort, J. Glycobiology 2001, 11, 25R.
4.5.4. Detection of protein thiols using 2,20-dithiodipyridine
Ten millilitres of E. coli MurA stock solution (500
of compound 1 stock solution (1 mM for equimolar ratio and
10 mM for 10-fold excess experiments) were dissolved in 35
Tris buffer 50 mM pH 7.8 and incubated at room temperature.
After 30 min, 50
L of a 2,20-dithiodipyridine solution (2 mM in so-
lM) and 5 lL
4. El Zoeiby, A.; Sanschagrin, F.; Levesque, R. C. Mol. Microbiol. 2003, 47, 1.
5. Eschenburg, S.; Priestman, M.; Schonbrunn, E. J. Biol. Chem. 2005, 280, 3757.
6. Jiang, S.; Gilpin, M. E.; Attia, M.; Ting, Y.-L.; Berti, P. J. Biochemistry 2011, 50,
2205.
7. Kim, D. H.; Lees, W. J.; Kempsell, K. E.; Lane, W. S.; Duncan, K.; Walsh, C. T.
Biochemistry 1996, 35, 4923.
lL
l
8. Wanke, C.; Amrhein, N. Eur. J. Biochem. 1993, 218, 861.
9. Steinbach, A.; Scheidig, A. J.; Klein, C. D. J. Med. Chem. 2008, 51, 5143.
10. Mendgen, T.; Scholz, T.; Klein, C. D. Bioorg. Med. Chem. Lett. 2010, 20, 5757.
11. Drobnica, L.; Šturdík, E. Folia Microbiol. 1980, 25, 467.
12. Latif, N.; Mishriky, N.; Guirguis, N. S.; Hussein, A. J. Prakt. Chem. 1973, 315, 419.
13. Castanedo Cancio, N. R.; Diaz Martinez, G.; Ramirez Dieguez, A.; Martin Triana,
E. L.; Salazar Yera, E.; Gonzalez Bedia, M. M.; Machado Rodriguez, R. M.;
Gonzalez Lorenzo, C. M.; Cueto Sanchez, M. D. L. C. WO Patent 9749283 (A2),
2001.
dium citrate buffer 0.5 M at pH 4, containing 6 M urea) were
added. After subsequent 30 min incubation at room temperature,
the reaction was quenched by the addition of 10
reaction product 2-thiopyridone was detected by HPLC at
342 nm. Injection volume was 20 L. The resulting peak area was
lL TFA (4%). The
l
compared to a control in which 1 was omitted.
14. McCoy, W. F.; Thornburgh, S. WO Patent 8911793 (A1), 1989.
15. McCoy, W. F. U.S. Patent 5,045,104, 1991.
4.5.5. Reactivity of compound 1 towards MurA
To determine the effects of MurA on the degradation route of 1,
conditions were identical to those described for the standard
enzyme activity assay, except for enzyme and compound concen-
16. Castanedo Cancio, N. R.; Goizueta Dominguez, R. D.; Gonzalez Garcia, O.; Perez
Donato, J. A.; Gonzalez Feitot, J.; Silveira Prado, E. A.; Cuesta Mazorra, M.;
Martinez del Pino, A. R.; Lugo Farinas, E. Eur. Pat. Appl. 0678516A1, 1995.
17. Gerns, F. R.; Timberlake, L. D. U.S. Patent 5,138,076, 1992.
18. Whitekettle, W. K.; Donofrio, D. K. U.S. Patent 5,158,972, 1992.
19. McCalla, D. R. Environ. Mutagen. 1983, 5, 745.
20. Šturdík, E.; Rosenberg, M.; Štibrányi, L. Chem. Biol. Interact. 1985, 53, 145.
21. Yahagi, T.; Nagao, M.; Hara, K.; Matsushima, T.; Sugimura, T.; Bryan, G. T.
Cancer Res. 1974, 34, 2266.
trations which were 100 lM. For reactions in the absence of
enzyme, identical reaction buffers were used without addition of
the enzyme. For measurements in absence of the assay additive
BSA, the respective volume was replaced with water. Reactions
were stopped at different time points by addition of 1
l
L ortho-
22. Estrada, E.; Gómez, M.; Castañedo, N.; Pérez, C. J. Mol. Struct. 1999, 468, 193.
23. González Borroto, J. I.; Creus, A.; Marcos, R. Mutat. Res. 2002, 519, 179.
24. González Borroto, J. I.; Creus, A.; Marcos, R.; Zapatero, J. Environ. Toxicol.
Pharmacol. 2005, 20, 241.
25. González Borroto, J. I.; Pérez, G.; Creus, A.; Marcos, R. Food Chem. Toxicol. 2004,
42, 187.
26. González Borroto, J. I.; Pérez Machado, G.; Creus, A.; Marcos, R. Mutagenesis
2005, 20, 193.
27. González Borroto, J. I. G.; Creus, A.; Marcos, R. Mutat. Res. 2001, 497, 177–184.
28. González Borroto, J. I. G.; Creus, A.; Marcos, R.; Molla, R.; Zapatero, J. Toxicol. Sci.
2003, 72, 359.
phosphoric acid (85%). For determinations at time point zero
the stopping reagent was added before all other components.
Standards for the retention time determination of 1, 2 and 3 were
prepared at 100 lM in 0.1% formic acid in order to prevent unspe-
cific degradation steps. Conversion of 1 in presence of enzyme was
tracked at 290 and 390 nm according to the expected decomposi-
tion products.
ˇ
29. Šturdík, E.; Benová, M.; Miertuš, S. Chem. Biol. Interact. 1986, 58, 69.
4.6. Detection of disulfide bond formation using the fluorescent
thiol 6
30. Castro-Hermida, J. A.; Gómez-Couso, H.; Ares-Mazás, M. E.; González-Bedia, M.
M.; Castañeda-Cancio, N.; Otero-Espinar, F. J.; Blanco-Mendez, J. J. Pharm. Sci.
2004, 93, 197.
31. Blondeau, J. M.; Castanedo, N.; Gonzalez, O.; Mendina, R.; Silveira, E. Int. J.
Antimicrob. Agents 1999, 11, 163.
32. Kellova, G.; Sturdik, E.; Stibranyi, L.; Drobnica, L.; Augustin, J. Folia Microbiol.
(Praha) 1984, 29, 23.
33. Milhazes, N.; Calheiros, R.; Marques, M. P. M.; Garrido, J.; Cordeiro, M. N. D. S.;
Rodrigues, C. T.; Quinteira, S.; Novais, C.; Peixe, L.; Borges, F. Bioorg. Med. Chem.
2006, 14, 4078.
E. coli MurA (final concn 100
derivative 6 (final concn 200 M) were incubated in Tris buffer
50 mM pH 7.8 at room temperature in the absence or presence
of 1 (final concn 200 M). Appropriate controls were performed
by incubation of enzyme and compound 1 alone and without any
lM) and the fluorescent thiol
l
l

804
T. Scholz et al. / Bioorg. Med. Chem. 21 (2013) 795–804
34. Schales, O.; Graefe, H. A. J. Am. Chem. Soc. 1952, 74, 4486.
35. Denisenko, P. P.; Sapronov, N. S.; Tarasenko, A. A. WO Patent 02102789 (A1),
2002.
36. Lo, K.; Cornell, H.; Nicoletti, G.; Jackson, N.; Hügel, H. Appl. Sci. 2012, 2, 114.
37. González Bedia, M. M. Tesis Doctoral, Universidad de Santiago de Compostela,
2001.
38. McCalla, D. R.; Voutsinos, D.; Olive, P. L. Mutat. Res. 1975, 31, 31.
39. Ramos, A.; Vizoso, A.; Edreira, A.; Betancourt, J.; Décalo, M. Mutat. Res. 1997,
390, 233.
55. Fridman, A. L.; Zalesov, V. S.; Surkov, V. D.; Kratynskaya, L. V.; Plaksina, A. N.
Pharm. Chem. J. 1976, 10, 752.
56. Dallmier, A. W.; Termine, E. J.; Yeoman, A. M. WO Patent 9851149 (A1), 1998.
57. Baum, E. Z.; Montenegro, D. A.; Licata, L.; Turchi, I.; Webb, G. C.; Foleno, B. D.;
Bush, K. Antimicrob. Agents Chemother. 2001, 45, 3182.
58. Schönbrunn, E.; Svergun, D. I.; Amrhein, N.; Koch, M. H. J. Eur. J. Biochem. 1998,
253, 406.
59. Brown, E. D.; Marquardt, J. L.; Lee, J. P.; Walsh, C. T.; Anderson, K. S.
Biochemistry 1994, 33, 10638.
40. Castanedo Cancio, N. R.; Sifontes Rodriguez, S.; Monzote Fidalgo, L.; Lopez
Hernandez, Y.; Montalvo Alvarez, A. M.; Infante Bourzac, J. F.; Olazabal Manso,
E. E. WO Patent 2007033616 (A1), 2007.
41. Rosenberg, M.; Balaz, S.; Sturdik, E.; Kuchar, A. Collect. Czech. Chem. Commun.
1987, 52, 425.
60. Tenorio-Borroto, E.; Penuelas Rivas, C. G.; Vásquez Chagoyán, J. C.; Castanedo,
N.; García-Mera, X.; González-Díaz, H. Bioorg. Med. Chem. 2012, 20, 6181.
61. Kaap, S.; Quentin, I.; Tamiru, D.; Shaheen, M.; Eger, K.; Steinfelder, H. J.
Biochem. Pharmacol. 2003, 65, 603.
62. Boyd, M. R.; Stiko, A. W.; Sasame, H. A. Biochem. Pharmacol. 1979, 28, 601.
63. Spielberg, S. P.; Gordon, G. B. J. Clin. Invest. 1981, 67, 37.
64. Guay, D. R. Drugs 2001, 61, 353.
42. Drobnica, L.; Sturdík, E.; Kovác, J.; Végh, D. Neoplasma 1981, 3, 281.
ˇ
43. Šturdík, E.; Drobnica, L.; Baláz, Š. Collect. Czech. Chem. Commun. 1983, 48, 336.
ˇ
44. Šturdík, E.; Drobnica, L.; Baláz, Š. Collect. Czech. Chem. Commun. 1983, 48, 327.
65. Gullner, G.; Cserhati, T.; Mikite, G. Pest. Biochem. Physiol. 1991, 39, 1.
66. Larrabee, J. A.; Thamrong-nawasawat, T.; Mon, S. Y. Anal. Biochem. 1999, 269,
194.
67. Bandow, J. E.; Brotz, H.; Leichert, L. I. O.; Labischinski, H.; Hecker, M. Antimicrob.
Agents Chemother. 2003, 47, 948.
ˇ
45. Šturdík, E.; Drobnica, L.; Baláz, Š.; Marko, V. Biochem. Pharmacol. 1979, 28,
2525.
ˇ
ˇ
46. Baláz, Š.; Šturdík, E.; Drobnica, L. Collect. Czech. Chem. Commun. 1982, 47, 1659.
47. Šturdík, E.; Liptaj, T.; Baláz, Š.; Drobnica, L. Collect. Czech. Chem. Commun. 1982,
ˇ
47, 1523.
68. Hochgräfe, F.; Mostertz, J.; Albrecht, D.; Hecker, M. Mol. Microbiol. 2005, 58,
409.
69. Hoener, B.; Noach, A.; Andrup, M.; Yen, T. S. B. Pharmacology 1989, 38, 363.
70. Rossi, L.; Silva, J. M.; McGirr, L. G.; Obrien, P. J. Biochem. Pharmacol. 1988, 37,
3109.
ˇ
ˇ
48. Baláz, Š.; Végh, D.; Šturdík, E.; Augustín, J.; Liptaj, T.; Kovác, J. Collect. Czech.
Chem. Commun. 1987, 52.
49. Marquardt, J. L.; Brown, E. D.; Lane, W. S.; Haley, T. M.; Ichikawa, Y.; Wong, C.
H.; Walsh, C. T. Biochemistry 1994, 33, 10646.
50. Lowther, W. T.; Orville, A. M.; Madden, D. T.; Lim, S. J.; Rich, D. H.; Matthews, B.
W. Biochemistry 1999, 38, 7678.
51. Swierczek, K.; Copik, A. J.; Swierczek, S. I.; Holz, R. C. Biochemistry 2005, 44,
12049.
52. Bachelier, A.; Mayer, R.; Klein, C. D. Bioorg. Med. Chem. Lett. 2006, 16, 5605.
53. Bartoshevich, Y. E.; Kopranenkov, V. N.; Burdelev, O. T.; Unkovskii, B. V. Pharm.
Chem. J. 1972, 6, 11.
71. Shen, W.; Hoener, B. A. Hum. Exp. Toxicol. 1996, 15, 428.
72. Shepherd, J. A.; Waigh, R. D.; Gilbert, P. Antimicrob. Agents Chemother. 1988, 32,
1693.
73. Altmeyer, M. A.; Marschner, A.; Schiffmann, R.; Klein, C. D. Bioorg. Med. Chem.
Lett. 2010, 20, 4038.
74. Steuer, C.; Heinonen, K. H.; Kattner, L.; Klein, C. D. J. Biomol. Screen. 2009, 14,
1102.
54. Clark, N. G.; Croshaw, B.; Leggetter, B. E.; Spooner, D. F. J. Med. Chem. 1974, 17,
977.
75. Mendgen, T.; Steuer, C.; Klein, C. D. J. Med. Chem. 2012, 55, 743.