Structure-activity relationships of phenyl-furanyl-rhodanines as inhibitors of RNA polymerase with antibacterial activity on biofilms

Article abstract of DOI:10.1021/jm0703183

The dramatic rise of antibiotic-resistant bacteria over the past two decades has stressed the need for completely novel classes of antibacterial agents. Accordingly, recent advances in the study of prokaryotic transcription open new opportunities for such molecules. This paper reports the structure-activity relationships of a series of phenyl-furanyl-rhodanines (PFRs) as antibacterial inhibitors of RNA polymerase (RNAP). The molecules have been evaluated for their ability to inhibit transcription and affect growth of bacteria living in suspension or in a biofilm and for their propensity to interact with serum albumin, a critical parameter for antibacterial drug discovery. The most active of these molecules inhibit Escherichia coli RNAP transcription at concentrations of ≤10 μM and have promising activities against various Gram-positive pathogens including Staphylococcus epidermidis biofilms, a major cause of nosocomial infection.

Full text of DOI:10.1021/jm0703183

J. Med. Chem. 2007, 50, 4195-4204
4195
Structure-Activity Relationships of Phenyl-Furanyl-Rhodanines as Inhibitors of RNA
Polymerase with Antibacterial Activity on Biofilms
Philippe Villain-Guillot,† Maxime Gualtieri,† Lionel Bastide,‡ Franc¸oise Roquet,† Jean Martinez,§ Muriel Amblard,§
Martine Pugniere,† and Jean-Paul Leonetti*,†
CNRS UMR 5236, Centre d‘e´tudes d’agents Pathoge`nes et Biotechnologie pour la Sante´, and CNRS UMR 5247, Institut des Biomole´cules Max
Mousseron, Faculte´ de Pharmacie, 15 aVenue Flahault, 34093 Montpellier cedex 5, France, and Selectbiotics, 69 rue Besse, Parc Scientifique
Besse, 30000 Nˆımes, France
ReceiVed March 20, 2007
The dramatic rise of antibiotic-resistant bacteria over the past two decades has stressed the need for completely
novel classes of antibacterial agents. Accordingly, recent advances in the study of prokaryotic transcription
open new opportunities for such molecules. This paper reports the structure-activity relationships of a
series of phenyl-furanyl-rhodanines (PFRs) as antibacterial inhibitors of RNA polymerase (RNAP). The
molecules have been evaluated for their ability to inhibit transcription and affect growth of bacteria living
in suspension or in a biofilm and for their propensity to interact with serum albumin, a critical parameter
for antibacterial drug discovery. The most active of these molecules inhibit Escherichia coli RNAP
transcription at concentrations of e10 µM and have promising activities against various Gram-positive
pathogens including Staphylococcus epidermidis biofilms, a major cause of nosocomial infection.
Introduction
We have previously described the biological activity of one
of these new synthetic antibacterials: the phenyl-furanyl-
rhodanines (PFRs).8 Its representative molecule 1 (Figure 1)
inhibits the core RNAP-σ70 factor assembly, affects transcription
by the core polymerase and the holoenzyme in vitro, and inhibits
RNA synthesis in Staphylococcus epidermidis. This molecule
is bactericidal against pathogenic Gram-positive bacteria, for
example, Staphylococcus aureus and Bacillus anthracis, but is
not active against Gram-negative pathogens like Escherichia
coli or Pseudomonas aeruginosa. Compound 2 (Figure 1), an
analogue of 1, shows interesting activity against S. epidermidis
biofilms when compared to RIF.10 Furthermore, these molecules
are not toxic against various eukaryotic cells.
These previous results prompted us to undertake a structure-
activity relationship (SAR) study to characterize structural
determinants of this new class of RNAP inhibitors. We
considered how modifications of the three different parts (A,
B, and C) of the molecules affected their biochemical and
antibacterial activities (Figure 1). Keeping in mind that these
molecules may also act as systemic antibacterial drugs, we
evaluated their binding to human serum albumin (HSA) by
surface plasmon resonance (SPR). This parameter is critical
since high free concentrations of antibiotics in plasma are needed
to achieve multiples of the minimum inhibitory concentration
(MIC) at the site of infection over a relatively short time period.
This explains why most of the antibiotics in clinical use are
more hydrophilic and have lower affinities for HSA than most
of the commercial drugs.11 In this paper, we report the synthesis
of relevant analogues of compound 1 and our findings regarding
their biological activity.
Antibiotic-resistant bacteria have become a major concern
for public health professionals since bacterial resistance has
dramatically increased during the last two decades.1 Further-
more, antibacterial chemotherapy alone is ineffective to cure
biofilm related infections, a major cause of nosocomial infec-
tions.1 Indeed, bacteria growing as an adherent biofilm on
catheters, orthopedic implants, and bones are inherently resistant
to antibiotics and are poorly recognized by the host immune
defenses.2 Consequently, there is a growing need for novel
classes of antibacterial agents able to address resistance and
persistent infections because of biofilms.
Within that scope, RNA polymerase (RNAPa), the enzyme
responsible for transcription, remains an attractive target for
antibacterial drug discovery.3 RNAP holoenzyme is a large
protein composed of core subunits (R(2)ââ′) plus a σ factor
that is required for specific promoter site recognition and the
initiation of transcription. Several antibacterial molecules, mainly
from natural origins, inhibit this enzyme with rather different
mechanisms of action; however, some share cross-resistance.4
Among these inhibitors, lipiarmycin, a narrow-spectrum anti-
bacterial, has entered phase III clinical trials under the name of
OPT-80 for the treatment of Clostridium difficile associated
diarrhea.5 Rifampicin (RIF) is still a first-line antibiotic for the
treatment of tuberculosis, and RIF in association with antibiotics
is frequently used to cure implant-associated infections because
of staphylococcal biofilms.6 The potential of RNAP as a target
has also been recently highlighted by the discovery of three
novel synthetic classes of inhibitors with antibacterial activity.7-9
* To whom correspondence should be addressed. Phone: +33 (0) 467
548 607. Fax: +33 (0) 467 548 609. E-mail: jean-paul.leonetti@ univ-
montp1.fr.
Results and Discussion
† CNRS UMR 5236-CPBS.
Synthesis of PFRs. Many compounds reported in this SAR
study are commercially available. When synthesis was required,
a two-reaction sequence of Suzuki-Miyaura coupling12,13 and
Knoevenagel condensation14 was used (Scheme 1). Suzuki-
Miyaura coupling was rather efficient when the reaction mixture
was heated (reaction time < 19 h, 74-100% yields) and was
much slower when performed at room temperature using Pd-
‡ Selectbiotics.
§ CNRS UMR 5247-IBMM.
a Abbreviations: PFR, phenyl-furanyl-rhodanine; RNAP, RNA poly-
merase; RIF, rifampicin; SAR, structure-activity relationships; HSA, human
serum albumin; SPR, surface plasmon resonance; MIC, minimum inhibitory
concentration; CFU, colony-forming unit; WRF, warfarin; MHB, Mueller-
Hinton broth; CHO, Chinese hamster ovary; ATP, adenosine triphosphate;
RU, resonance unit.
10.1021/jm0703183 CCC: $37.00 © 2007 American Chemical Society
Published on Web 08/01/2007

4196 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17
Villain-Guillot et al.
Table 1. SAR of Part A Derivatives of the Molecular Template
Figure 1. Compounds 1 and 2 and molecular template for the SAR
study.
Scheme 1. Synthesis of PFRsa
a (a) Me3SiBr, CHCl3, ∆; (b) cat. (Ph3P)4Pd, K2CO3, toluene/EtOH/H2O,
∆, or cat. Pd-C, Na3PO4,12H2O, EtOH/H2O; (c) 3-allylrhodanine, ethyl-
enediamine diacetate, MeOH, or allylrhodanine, piperidine, DMF.
a S. epidermidis CIP 105777. b E. coli tolC CGSC 5633.
Table 2. SAR of Carboxylic Acid Bioisosters
carbon as catalyst (10 days, 41% yield). Suzuki-Miyaura
coupling did not occur when the rhodanine moiety was present
in the starting material, possibly because of sulfur poisoning of
the catalyst.
Mild conditions were used for the Knoevenagel condensation.
Ethylenediamine diacetate was used as the required base with
methanol as solvent. These conditions gave the desired com-
pounds in 35-75% yields. When solubility of the starting
materials was poor in methanol, dimethylformamide was used
with piperidine as the required base, but these conditions gave
poor yields (8-22%).
Synthesis of Carboxylic Acid Bioisosteres. The carboxylic
acid moiety of 2 was replaced with various moieties that have
been considered as bioisosteres of this function (Table 2).15
Compounds 5b and 5d were obtained following the procedure
described above. Analogues 5c and 5e were obtained starting
from 4f and 2, respectively. Synthesis of compound 5c is
outlined in Scheme 2. Starting material 4f was reacted with
phosphorus oxychloride and then was hydrolyzed in water.16
Compound 5c was produced without further purification in poor
yield (21%). Synthesis of compound 5e is outlined in Scheme
3. Starting material 2 was reacted under mild conditions with
a S. epidermidis CIP 105777. b E. coli tolC CGSC 5633.
cyanuric chloride, hydroxylamine, N-methylmorpholine, and
dimethylaminopyridine in catalytical amounts.17 Compound 5e

SAR of Antibacterial Inhibitors of RNA Polymerase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17 4197
Table 3. SAR of Part B Derivatives of the Molecular Template
Scheme 2. Synthesis of Carboxylic Acid Bioisoster 5c
(Phosphoric Acid)a
a S. epidermidis CIP 105777. b E. coli tolC CGSC 5633.
a (a) POCl3, pyridine, 0 °C, 50 min, then H2O, 0 °C to rt, 18 h.
Table 4. SAR of Part C Derivatives of the Molecular Template
Scheme 3. Synthesis of Carboxylic Acid Bioisoster 5e
(Hydroxamic Acid)a
a (a) NH2OH‚HCl, cyanuric chloride, NMM, cat. DMAP, THF, 0 °C to
rt, 18 h.
was produced after purification by reverse-phase high-perfor-
mance liquid chromatography (HPLC) in very poor yields (1%).
SAR Study. The compounds were tested for their ability to
inhibit in vitro E. coli RNAP transcription and for their ability
to inhibit bacterial growth of Gram-positive (S. epidermidis)
and Gram-negative (E. coli tolC) species. Indeed, none of the
molecules was able to inhibit the growth of the E. coli wild
strain at concentrations up to 100 µg/mL (Supporting Informa-
tion, Table S1). The E. coli tolC strain is a mutant deficient in
multidrug efflux systems. This property allows for detection of
the antibacterial activity of molecules that are able to penetrate
E. coli but would be readily ejected from the bacteria. In
addition, it allows for comparison of their MIC with data
obtained in the in vitro transcription assay on the E. coli RNA
polymerase. In parallel, the toxicity of the compounds was tested
on eukaryotic cells, and none of these compounds affected their
growth (Supporting Information, Table S1).
Modifications of part A of the molecular template stress the
importance of the carboxylic acid moiety for inhibition of in
vitro transcription (Table 1). In the absence of any functionality
on the phenyl ring (4b), transcription inhibition was decreased
2-fold, but antibacterial activity remained good on E. coli tolC.
Comparison of compounds 1, 2, and 4a affords the conclusion
that the position of the acid on the phenyl ring does not really
matter with respect to the enzymatic activity, but the para- and
meta-positions are required for antibacterial activity. Analogue
4k illustrates that the carboxylic acid moiety can be distanced
from the phenyl ring, only slightly affecting the in vitro
transcription and antibacterial activities. Replacement of the
carboxylic acid moiety by a weakly basic moiety (4e), or by a
stronger electron-withdrawing group (4c), negatively affects the
activity of these compounds in both the transcription and the
antibacterial assays. However, replacement by a weaker electron-
withdrawing group (4d) leads to a slight decrease of transcription
a S. epidermidis CIP 105777. b E. coli tolC CGSC 5633.
inhibition, while antibacterial activity is conserved on E. coli
tolC. Comparison of molecules 4h and 4j with the previous
molecules 4d and 4e reveals that the presence of the carboxylic
acid moiety enhanced the activity on the RNAP. Furthermore,
chlorine does not enhance or disfavor antibacterial activity, while
the amine leads to loss of antibacterial activity. Biological
activity is also lost when the acid is masked by a methyl ester
function (4g) or by a lactone function (4i).
To this point of the discussion, the carboxylic acid was
revealed to be the most efficient moiety for the inhibition of
transcription. Further usual bioisosteres of this function were
then tested to see if they may improve the inhibitory activities
of lead compound 2 on RNAP and bacterial growth but also

4198 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17
Table 5. Biological Activity of Compounds 11a and 11b
Villain-Guillot et al.
transcription
S. epidermidisa
E. coli tolCb
compound
structure
inhibition at 10 µM (%)
MIC (µg/mL)
MIC (µg/mL)
11a
63 ( 5
g100
g100
11b
95 (9
25.0
g100
a S. epidermidis CIP 105777. b E. coli tolC CGSC 5633.
Table 6. Antibacterial Spectrum of PFRs
MIC (µg/mL)
S. aureus
S. epidermidis
S. pneumoniae
C. difficile
B. anthracis
No
CIP 76.25a
49589b
RP62A
40004
48155
CIP 103566
DSM 1296
9131
1
2
4b
4f
4h
4k
5d
5e
6a
6b
11b
12.5
g100
g100
6.25
25.0
g100
g100
25.0
50.0
25.0
g100
g100
25.0
12.5
3.13
g100
12.5
6.25
25.0
g100
g100
25.0
25.0
6.25
ndc
12.5
6.25
25.0
nd
25.0
3.13
nd
12.5
12.5
12.5
nd
50.0
g100
g100
25.0
12.5
g100
g100
50.0
12.5
50.0
g100
g100
6.25
6.25
3.13
0.78
6.25
3.13
12.5
12.5
50.0
6.25
12.5
12.5
50.0
25.0
12.5
g100
g100
25.0
25.0
g100
50.0
g100
g100
50.0
g100
50.0
nd
nd
50.0
50.0
25.0
25.0
50.0
25.0
25.0
50.0
25.0
25.0
12.5
25.0
a Reference strain. b RIF-resistant multiresistant strain. c Not determined.
if they may decrease HSA binding (Table 2). Nitrobenzene (5a),
sulfonic acid (5b), and phosphoric acid (5c) derivatives are
neither biochemically active nor antibacterial. Benzamide (5d)
and hydroxamic acid (5e) inhibit in vitro transcription as
efficiently as lead compounds 1 and 2. None of these bioisosters
are antibacterial against S. epidermidis, but 5e and, more
interestingly, 5d show activity against E. coli tolC. Furthermore,
these compounds present clear advantages regarding HSA
binding (see below).
Modifications of part B of the molecular template show that
biological activity is enhanced by a furan (Table 3). Comparison
of molecules 1, 6a, and 6b afforded the conclusion that the
molecules require a five-membered planar heterocycle to be
optimally biochemically active. However, 6b proves to be
antibacterial against S. epidermidis and not against the permeable
E. coli tolC strain. Antibacterial activity is also observed on B.
anthracis and S. aureus (Table 6). A possible explanation may
be that this compound does not affect transcription of E. coli
but does affect transcription of other microorganisms. This
property remains unclear since, in this study, the transcription
assay was performed with purified RNAP from E. coli. Furan
and pyrrole display comparable activities regarding holoenzyme
dissociation and transcription inhibition, but furan seems to
slightly favor antibacterial activity (Table 6). Nevertheless,
pyrrole seems advantageous when bacteria are grown as a
biofilm (see below).
moiety, prove to be antibacterial agents. Comparison of
compounds 2, 7a, 7b, and 7d highlights the importance of the
thioxo-function and the N-allyl chain in this heterocycle for
antibacterial activity. We felt concerned that the antibacterial
activity of our series of molecules was only due to the
allylrhodanine and not to its ability to inhibit bacterial transcrip-
tion. Indeed, some rhodanine derivatives related to our template
have been previously reported as antibacterial agents.18,19
However, scrutiny of other molecules related to our template
indicates that its biological activity is not limited to only the
allylrhodanine or the presence of specific fragments connected
randomly (Supporting Information, Table S2).
Analogues 11a and 11b reveal how parts A and B of the
molecular template can be connected (Table 5). The merge of
the two fragments (11b) does not disfavor biochemical activity,
while distancing (11a) disfavors it. Antibacterial activity is lost
completely with compound 11a.
Spectrum of Antibacterial Activity. The most interesting
PFRs of this series have a broad spectrum of activity on relevant
Gram-positive pathogens (Table 6). A multiresistant S. aureus
clinical isolate also resistant to RIF is still affected by
compounds active against a S. aureus reference strain. The
molecules are weakly active against Streptococcus pneumoniae
and C. difficile. Nevertheless, the MIC of 6a against C. difficile
is not much higher than that of vancomycin, a second-line
antibiotic for the treatment of intestinal infections caused by
this pathogen.5 Very encouraging activities are observed on B.
anthracis; the best activity is observed with 4b, with an MIC
of 0.78 µg/mL. With respect to this pathogen, a previous report
stated that 4h was a good inhibitor of anthrax lethal factor, and
some analogues were active in an animal infection model.20
Modifications of part C of the molecular template underlined
the importance of the allylrhodanine for antibacterial activity
(Table 4). With the exception of compounds 7c and 7i, all the
compounds are able to inhibit transcription in vitro. Strikingly,
only the lead compounds 1 and 2, bearing the allylrhodanine

SAR of Antibacterial Inhibitors of RNA Polymerase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17 4199
encountered in the literature when RIF and other antibiotics are
used against staphylococcal biofilms.24,25 Furthermore, we have
previously observed that very high concentrations of RIF do
not significantly improve its antibacterial activity, and a limit
is almost reached at a concentration of 8 times the MIC.10
Among the investigated compounds, only 4f and 4h do not
affect S. epidermidis biofilms at all. The other compounds 1,
4k, 6a, 6b, and 11b display good activities on biofilms at a
concentration of 8 times the MIC and are able to decrease the
amount of bacteria by 2-3 Log10 CFU/well. Interestingly,
compound 4k displays similar activities on the three strains at
4 times and 8 times the MIC. These results stress the potential
role of distancing the carboxylic acid moiety from the phenyl
ring with respect to biofilms. The best activity against the S.
epidermidis RP62A, 40004, and 48155 strains is observed with
6a at a concentration of 8 times MIC, with a decrease in levels
of 2.8 Log10 CFU/well, 2.3 Log10 CFU/well, and 1.8 Log10 CFU/
well, respectively. This latter compound is not the most efficient
under planktonic conditions in terms of its MIC.
HSA Binding and Antibacterial Activity. Some of the
molecules described in this paper possess interesting antibacterial
activities. However, like most of the antibacterial molecules
identified after the screening of large synthetic libraries,9,26,27
they are very likely to interact with serum proteins because of
their physicochemical properties.28 The binding of the molecules
to serum proteins, essentially to HSA, reduces their free
concentrations in plasma and thus decreases their antibacterial
efficiency. The common but rather indirect approach to evaluate
the propensity of a molecule to bind to HSA is the determination
of its MIC in the presence of this protein. SPR technology is a
complementary approach, since it measures accurately the
binding of a molecule to HSA. In a previous report, Frostell-
Karlsson et al. used this technique in an attempt to predict HSA
binding levels of drug compounds.29 Their results demonstrate
that it is possible to classify compounds as low (binding < 80%),
medium (binding of 80-90%), and high binders (binding >
90%) of HSA, according to their SPR biosensor response. To
have a further selection criterion for lead optimization, we have
used the same method to truly evaluate and rank the binding of
PFRs to HSA (Figure 3). As expected, HSA bindings, measured
by SPR, of the lead compounds 1 and 2 are very high, with
biosensor responses 10-fold higher than that of warfarin (WRF),
a high binder of HSA.29 All analogues also show high response
values, with the exception of two compounds (4f, 5d) that
display biosensor responses comparable to WRF. Replacement
of the carboxylic acid moiety (2) by a hydroxyl (4f) or an amide
moiety (5d) decreases HSA binding by 10-fold. Comparatively,
substitution by a hydroxamic acid moiety (5e) decreases its HSA
binding by almost 5-fold but decreases its activity on the
enzyme. Despite having no antibacterial activity, the distancing
of the benzoic acid from the furan (11a) decreases HSA binding
by 6-fold. Encouragingly, the lowest binding values were not
far from those of the clinically used antibiotic RIF.
Figure 2. Amount of bacteria measured by ATP-bioluminescence
within 24 h biofilms of S. epidermidis RP62A, 40004, and 48155 after
24 h exposure to RIF (10 µg/mL) and PFRs. The mean amounts of
bacteria (dotted line) within 24 h biofilms were 6.32 ( 0.05, 5.62 (
0.08, and 4.80 ( 0.09 Log10 CFU/well for strains RP62A, 40004, and
48155, respectively. White bars: antibacterial concentration at 4 times
the MIC; black bars: antibacterial concentration at 8 times the MIC.
However, we show that this compound does not only target the
toxin but also affects the growth of B. anthracis with a good
efficacy (MIC of 3.13 µg/mL). This dual effect on the growth
of B. anthracis and its virulence factor could explain the results
achieved in vivo.20
Regarding SAR, a striking aspect of this series of antibacteri-
als is highlighted by 1 and 2. These molecules present rather
different patterns of antibacterial activity. However, this is
despite their very close chemical structure and very close
behavior on RNAP in vitro. Such discrepancies between the
inhibitory effect on the purified target and on bacteria are
frequently reported.21-23 They are likely to reflect the inability
of many molecules to reach their target and to reflect how bac-
teria can select against small molecules present outside the cell.
Activity on Biofilms. Since some compounds display anti-
bacterial activity against the S. epidermidis RP62A strain under
planktonic conditions (i.e., suspended bacteria), they were
challenged against 24 h old biofilms of the same strain. Two
additional S. epidermidis clinical isolates (strains 40004 and
48155) obtained from patients with device-related staphylococcal
infections were also investigated. The MICs of the investigated
compounds do not vary significantly between the different S.
epidermidis strains (Table 6). Biofilms were grown for 24 h
and were then challenged with the antibacteria for 24 h at
concentrations of 4 times and 8 times the MIC (Figure 2). RIF
was tested at 10 µg/mL, a concentration much higher than its
MIC (0.003-0.006 µg/mL), but high concentrations are often
Antibacterial compounds were then evaluated against E. coli
tolC in the presence of HSA to assess the relevance of the SPR
method (Figure 3). All the very high binders of HSA (1, 2, 4h,
and 4k) lose their antibacterial activity in the presence of a
concentration as low as 1% HSA (w/v). MICs of compounds
4b (not evaluated in SPR because of poor solubility) and 5d
increase by more than 10-fold to reach 25 µg/mL. The
hydroxamic acid is an interesting moiety since the MIC of
compound 5e only increases by 2-fold. Surprisingly, 4f, which
ranks as one of the lowest HSA binders, loses its antibacterial
activity in presence of the serum protein. A possible explanation

4200 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17
Villain-Guillot et al.
Experimental Section
Chemistry. All reagents were purchased from commercial
suppliers and were used as provided. Purification by reverse-phase
HPLC was performed on a Waters SymmetryPrep C18 column (19
× 150 mm, 7 µM). Analytical LC-MS was performed on a Waters
Alliance 2690 HPLC coupled to a Micromass Platform II spec-
trometer (electrospray ionization mode, ESI+ or ESI-). All
analyses were carried out using a RP-18, 3.5 µm, 2.1 × 30 mm
reverse-phase column. A flow rate of 600 µL/min and a gradient
of 0-100% B over 5 min was used; eluent-A: water/0.1% TFA,
eluent-B: acetonitrile/0.1% TFA. Proton nuclear magnetic reso-
nance (1H NMR) analyses were performed by Laboratoire des
Mesures Physiques of University of Montpellier-2 (Montpellier,
France). 1H NMR spectra recorded on a Bruker DRX spectrometer
(400 MHz) or on a Bruker AC spectrometer (250 MHz) are given
in parts per million (ppm) and are referenced to an internal standard
of tetramethylsilane (TMS, δ 0.00 ppm). Peak multiplicity is
reported as s (singlet), d (doublet), dd (double doublet), t (triplet),
m (multiplet), and br s (broad singlet). Elemental analyses were
performed by Spectropole of Marseille (Marseille, France).
3-(5-Formyl-furan-2-yl)-benzoic Acid (3a). A solution of K2-
CO3 (476 mg, 3.44 mmol) in water (3 mL) was added to a mixture
of 3-carboxyphenylboronic acid (316 mg, 1.90 mmol) and 5-bromo-
2-furaldehyde (301 mg, 1.72 mmol) in toluene/ethanol (20 mL).
(Ph3P)4Pd (20 mg, 0.02 mmol) was added, and the reaction mixture
was stirred at 90 °C for 7 h. The reaction mixture was cooled to
room temperature and was diluted with water (15 mL); the pH was
then adjusted to 1 by addition of a solution of HCl (6 N). The
diluted reaction mixture was extracted with CH2Cl2 (3 × 100 mL);
the combined organic fractions were washed with brine (1 × 10
mL), were dried over MgSO4, and were concentrated under reduced
pressure to afford 367 mg of 3a in quantitative yield. MS: 217
([M + H]+).
3-(5-Formyl-furan-2-yl)-aniline (3b). A solution of K2CO3 (486
mg, 3.52 mmol) in water (3 mL) was added to a mixture of
3-aminobenzeneboronic acid monohydrate (300 mg, 1.94 mmol)
and 5-bromo-2-furaldehyde (308 mg, 1.76 mmol) in toluene/ethanol
(20 mL). (Ph3P)4Pd (20 mg, 0.02 mmol) was added, and the reaction
mixture was stirred at 90 °C for 7 h. The reaction mixture was
cooled to room temperature and was diluted with water (15 mL).
The diluted reaction mixture was extracted with CH2Cl2 (3 × 100
mL), and the combined organic fractions were washed with brine
(1 × 10 mL), were dried over MgSO4, and were concentrated under
reduced pressure to afford 215 mg of 3b. The product was used
without further purification.
4-(5-Formyl-furan-2-yl)-phenol (3c). A solution of K2CO3 (301
mg, 2.18 mmol) in water (3 mL) was added to a mixture of
4-hydroxyphenylboronic acid (200 mg, 1.45 mmol) and 5-iodo-2-
furaldehyde (386 mg, 1.74 mmol) in toluene/ethanol (20 mL).
(Ph3P)4Pd (15 mg, 0.01 mmol) was added, and the reaction mixture
was stirred at 90 °C for 19 h. The reaction mixture was carried out
using the method described for the synthesis of 3a to afford 214
mg of 3c with a 78% yield. MS: 189 ([M + H]+).
Figure 3. Ranking of HSA binding levels of PFRs according to
biosensor response and influence of HSA on antibacterial activity.
Rifampicin (RIF), a low binder of HSA (73% binding), and warfarin
(WRF), a high binder of HSA (98% binding), were used as reference
molecules to assess protein binding.29 Compound 4b was not tested
because of its poor solubility in the SPR buffer. White bars: biosensor
response ( standard error; black square: E. coli tolC MIC in the
absence of HSA; black triangle: E. coli tolC MIC in the presence of
1% (w/v) HSA.
would be that some PFRs bind to some sites of HSA that may
be unaccessible subsequent to NH2-tethering of the protein to
the biosensor chip. These putative sites would be different for
the major drug binding sites according to Frostell-Karlsson et
al.’s conclusions.29
Despite this limitation, SPR results are rather informative,
and it appears that amide and hydroxamic acid moieties should
be valuable for development of this series with respect to
attenuation of binding to serum proteins.
Conclusion
The tested compounds inhibited bacterial RNAP in the
micromolar range with an interesting spectrum of activity against
clinically relevant Gram-positive pathogens and biofilms. Map-
ping of the binding site of the PFRs on the core RNAP would
be a milestone that could lead to understanding their mechanism
of action and a more rational design of improved ligands. Not
surprisingly, from an antibacterial provided by the screening
of a large synthetic library, the original molecule interacts too
strongly with HSA. However, it is possible to significantly
decrease this binding by slight modifications of the benzoic acid
moiety.
SAR studies on antibiotics are often difficult to achieve. Many
molecules with good activity on the purified target and bacteria
turn out to be inactive after modification either because these
modifications alter their interaction with the target or because
permeability limitations are difficult to assess. We conclude from
this study that the PFR skeleton can be modified without losing
its antibacterial properties and that it is a promising template
for development of new drugs to fight infection.
3-Amino-5-(5-formyl-furan-2-yl)-benzoic Acid (3d). A solution
of K2CO3 (86 mg, 0.62 mmol) in water (1 mL) was added to a
mixture of 3-amino-5-carboxyphenylboronic acid (75 mg, 0.41
mmol) and 5-iodo-2-furaldehyde (110 mg, 0.50 mmol) in toluene/
ethanol (10 mL). (Ph3P)4Pd (4 mg, 3 nmol) was added, and the
reaction mixture was stirred at 90 °C for 19 h. The reaction mixture
was cooled to room temperature and was diluted with water (10
mL), and the pH was adjusted to 1 by addition of a solution of
HCl (6 N). The product was precipitated after addition of CH2Cl2
(40 mL). The solid was collected by filtration and was washed with
cold methanol (3 × 5 mL) and cold ether (3 × 5 mL) to afford 70
mg of 3d in a 74% yield. MS: 232 ([M + H]+).
3-[4-(5-Formyl-furan-2-yl)-phenyl]-propionic Acid (3e). 10%
Pd-carbon (1 mg) was added to a solution of Na3PO4,12H2O (261
mg, 0.69 mmol), 4-(2-carboxyethyl)benzeneboronic acid (67 mg,
0.34 mmol), and 5-bromo-2-furaldehyde (60 mg, 0.34 mmol) in
water/ethanol (4:1, 7 mL) under a nitrogen atmosphere. The reaction

SAR of Antibacterial Inhibitors of RNA Polymerase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17 4201
mixture was stirred at room temperature for 10 days. The reaction
mixture was filtered through a Celite bed, and pH was adjusted to
1 by addition of a solution of HCl (6 N). The mixture was extracted
with CH2Cl2 (3 × 25 mL), and the combined organic fractions were
washed with brine (1 × 5 mL), were dried over MgSO4, and were
concentrated under reduced pressure to afford 34 mg of 3e in a
41% yield. MS: 487 ([2M - H]-).
4-(5-Formyl-furan-2-yl)-benzamide (3f). A solution of K2CO3
(129 mg, 0.93 mmol) in water (1 mL) was added to a mixture of
4-aminocarbonylphenylboronic acid (100 mg, 0.62 mmol) and
5-iodo-2-furaldehyde (165 mg, 0.75 mmol) in toluene/ethanol (10
mL). (Ph3P)4Pd (4 mg, 3 nmol) was added, and the reaction mixture
was stirred at 90 °C for 17 h. The reaction mixture was carried out
using the method described for the synthesis of 3b to afford 114
mg of 3f in an 86% yield. MS: 216 ([M + H]+).
(2H, t, J ) 7.8 Hz), δ 2.87 (2H, t, J ) 7.8 Hz), 4.65 (2H, d, J )
4.8 Hz), 5.13 (1H, d, J ) 17.5 Hz), 5.19 (1H, d, J ) 10.6 Hz),
5.82 (1H, m), 7.30 (1H, d, J ) 3.7 Hz), 7.38 (1H, d, J ) 3.8 Hz),
7.41 (2H, d, J ) 8.4 Hz), 7.70 (1H, s), 7.78 (1H, d, J ) 8.2 Hz);
MS: 398 ([M + H]+); anal. (C20H17NO4S2): C, H, N, S.
4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-benzenesulfonic Acid (5b). 3-Allylrhodanine (17 mg, 0.10
mmol) was added to a solution of 4-(5-formyl-2-furyl)-benzene-
sulfonic acid (25 mg, 0.10 mmol) and piperidine (10 µL, 0.10 mmol)
in dimethylformamide (3 mL); the reaction was continued as
described for the synthesis of 4j to afford 9 mg of 5b in a 22%
1
yield. H NMR (400 MHz, DMSO): δ 4.66 (2H, d, J ) 5.2 Hz),
5.14 (1H, d, J ) 17.2 Hz), 5.19 (1H, d, J ) 10.4 Hz), 5.85 (1H,
m), 7.36 (1H, d, J ) 3.7 Hz), 7.40 (1H, d, J ) 3.7 Hz), 7.48 (1H,
s), 7.71 (1H, s), 7.75 (2H, d, J ) 8.4 Hz), 7.83 (2H, d, J ) 8.31
Hz); MS: 408 ([M + H]+).
3-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-benzoic Acid (1). 3-Allylrhodanine (24 mg, 0.14 mmol) was
added to a solution of 3a (30 mg, 0.14 mmol) and ethylenediamine
diacetate (75 mg, 0.42 mmol) in methanol (3 mL). The reaction
mixture was stirred at room temperature for 18 h, and the product
was precipitated. The solid was collected by filtration and was
washed with water (3 × 5 mL) to afford 33 mg of 1 in a 64%
yield as a red powder. 1H NMR (400 MHz, DMSO): δ 4.65 (2H,
d, J ) 3.7 Hz), 5.14 (1H, d, J ) 17.2 Hz), 5.19 (1H, d, J ) 10.4
Hz), 5.86 (1H, m), 7.40 (2H, dd, J ) 5.8, 6.0 Hz), 7.60 (1H, dd, J1
) J2 ) 7.7 Hz), 7.71 (1H, s), 7.97 (2H, m), 8.39 (1H, d, J ) 1.2
Hz); MS: 370 ([M - H]-); Anal. (C18H13NO4S2): C, H, N, S.
3-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-aniline (4e). 3-Allylrhodanine (23 mg, 0.13 mmol) was added
to a solution of 3b (25 mg, 0.13 mmol) along with ethylenediamine
diacetate (72 mg, 0.40 mmol) in methanol (3 mL); the reaction
was then continued as described for the synthesis of 1 to afford 30
Phosphoric Acid Mono-(4-[5-(3-allyl-4-oxo-2-thioxo-thiazo-
lidinylidenemethyl)-furan-2-yl]-phenyl) Ester (5c). Phosphorus
oxychloride (135 µL, 0.15 mmol) was added dropwise to a solution
of 4f in pyridine (5 mL) at 0 °C. The reaction mixture was stirred
for 50 min, and water (30 mL) was added. The reaction mixture
was allowed to warm to room temperature and was stirred for 18
h. The reaction mixture was then extracted with CH2Cl2 (3 × 100
mL), and the organic fractions were combined and concentrated
under reduced pressure to afford 13.1 mg of 5c in a 21% yield. 1H
NMR (400 MHz, DMSO): δ 4.65 (2H, d, J ) 5.2 Hz), 5.14 (1H,
d, J ) 17.2 Hz), 5.19 (1H, d, J ) 10.4 Hz), 5.85 (1H, m), 7.28
(1H, d, J ) 3.7 Hz), 7.39 (1H, d, J ) 3.4 Hz), 7.59 (2H, dd, J )
7.6, 5.9 Hz), 7.70 (1H, s), 7.86 (2H, d, J ) 8.6 Hz); MS: 424 ([M
+ H]+); anal. (C17H14NO6PS2): C, H, N, S.
4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-benzamide (5d). 3-Allylrhodanine (24 mg, 0.14 mmol) was
added to a solution of 3f (30 mg, 0.14 mmol) and piperidine (15
µL, 0.14 mmol) in dimethylformamide (3 mL); the reaction was
continued as described for the synthesis of 4j to afford 6 mg of 5d
in a 12% yield. 1H NMR (250 MHz, DMSO): δ 4.64 (2H, d, J )
5.4 Hz), 5.12 (1H, d, J ) 17.2 Hz), 5.18 (1H, d, J ) 10.8 Hz),
5.83 (1H, m), 7.40 (1H, d, J ) 4.4 Hz), 7.46 (1H, d, J ) 3.1 Hz),
7.71 (1H, s), 7.95 (2H, d, J ) 7.5 Hz), 8.03 (2H, d, J ) 9.3 Hz);
MS: 371 ([M + H]+).
1
mg of 4e in a 68% yield. H NMR (400 MHz, DMSO): δ 4.65
(2H, d, J ) 5.2 Hz), 5.13 (1H, d, J ) 17.2 Hz), 5.19 (1H, d, J )
10.4 Hz), 5.40 (2H, s), 5.86 (1H, m), 6.63 (1H, d, J ) 7.7 Hz),
7.01 (3H, m), 7.13 (1H, d, J ) 3.7 Hz), 7.19 (1H, dd, J1 ) J2 )
8.0 Hz), 7.36 (1H, d, J ) 3.7 Hz), 7.68 (1H, s); MS: 343 ([M +
H]+); Anal. (C17H14N2O2S2): C, H, N, S.
4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-phenol (4f). 3-Allylrhodanine (46 mg, 0.27 mmol) was added
to a solution of 3c (50 mg, 0.27 mmol) along with ethylenediamine
diacetate (48 mg, 0.27 mmol) in methanol (3 mL). The reaction
mixture was stirred at room temperature for 17 h. The pH of the
reaction mixture was adjusted to 1 by addition of a solution of
HCl (6 N). The reaction mixture was extracted with CH2Cl2 (3 ×
25 mL), and the combined organic fractions were washed with brine
(1 × 5 mL), were dried over MgSO4, and were concentrated under
4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-yl)-N-hydroxybenzamide (5e). A solution of cyanuric chloride
(149 mg, 0.81 mmol) in tetrahydrofurane (3 mL) was added
dropwise to a solution of 2 (200 mg, 0.54 mmol), N-methylmor-
pholine (89 µL, 0.81 mmol), dimethylaminopyridine (0.7 mg, 5
nmol), and hydroxylamine hydrochloride salt (112 mg, 1.62 mmol)
in tetrahydrofurane (12 mL) at 0 °C. The reaction mixture was
allowed to warm to room temperature and was stirred for 18 h.
The reaction mixture was diluted with water (15 mL) and was
extracted with CH2Cl2 (3 × 100 mL). The organic fractions were
combined, were washed with brine (1 × 15 mL), were dried over
MgSO4, and were concentrated under reduced pressure. The product
was diluted with water/acetonitrile (7:3, 0.1% TFA, 30 mL) and
was purified by reverse-phase HPLC with a linear gradient of 30-
90% acetonitrile (0.1% TFA) over 55 min at 10 mL/min. Relevant
fractions were combined, and the solvent was removed by freeze-
1
reduced pressure to afford 69 mg of 4f in a 75% yield. H NMR
(400 MHz, DMSO): δ 4.64 (2H, d, J ) 3.7 Hz), 5.13 (1H, d, J )
17.2 Hz), 5.19 (1H, d, J ) 10.4 Hz), 5.84 (1H, m), 6.95 (2H, d, J
) 6.7 Hz), 7.12 (1H, d, J ) 3.7 Hz), 7.35 (1H, d, J ) 3.7 Hz),
7.65 (1H, s), 7.71 (2H, d, J ) 6.7 Hz), 10.08 (1H, s); MS: 344
([M + H]+); Anal. (C17H13NO3S2): C, H, N, S.
3-Amino-5-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-
furan-2-yl)-benzoic Acid (4j). 3-Allylrhodanine (19 mg, 0.11
mmol) was added to a solution of 3d (25 mg, 0.11 mmol) and
piperidine (11 µL, 0.11 mmol) in dimethylformamide (3 mL). The
reaction mixture was stirred at room temperature for 16 h. The
reaction mixture was diluted with water/acetonitrile (3:2, 0.1% TFA,
30 mL) and was purified by reverse-phase HPLC with a linear
gradient of 40-90% acetonitrile (0.1% TFA) over 45 min at 10
mL/min. Relevant fractions were combined, and the solvent was
removed by freeze-drying to afford 3 mg of 4j in an 8% yield.
MS: 387 ([M + H]+).
3-[4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-fu-
ran-2-yl)-pheny l]-propionic Acid (4k). 3-Allylrhodanine (14 mg,
0.08 mmol) was added to a solution of 3e (20 mg, 0.08 mmol) and
ethylenediamine diacetate (30 mg, 0.16 mmol) in methanol (3 mL);
the reaction was continued as described for the synthesis of 4f.
The product was further washed with methanol (2 × 5 mL) to afford
11 mg of 4k in a 35% yield. 1H NMR (250 MHz, DMSO): δ 2.58
1
drying to afford 5 mg of 5e in a 1% yield. H NMR (400 MHz,
DMSO): δ 4.66 (2H, d, J ) 5.1 Hz), 5.14 (1H, d, J ) 17.2 Hz),
5.20 (1H, d, J ) 10.5 Hz), 5.85 (1H, m), 7.41 (1H, d, J ) 3.7 Hz),
7.46 (1H, d, J ) 3.7 Hz), 7.73 (1H, s), 7.92 (4H, s), 9.14 (1H, br
s), 11.32 (1H, s); MS: 387 ([M + H]+).
3′-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-biphenyl-
3-carboxylic Acid (6b). 3-Allylrhodanine (31 mg, 0.18 mmol) was
added to a solution of 3′-formyl-(1,1′-biphenyl)-3-carboxylic acid
(40 mg, 0.18 mmol) and ethylenediamine diacetate (64 mg, 0.35
mmol) in methanol (3 mL); the reaction was continued as described
1
for the synthesis of 1 to afford 37 mg of 6b in a 54% yield. H
NMR (400 MHz, DMSO): δ 4.67 (2H, d, J ) 5.1 Hz), 5.16 (1H,
d, J ) 18.7 Hz), 5.20 (1H, d, J ) 11.7 Hz), 5.85 (1H, m), 7.58
(2H, dd, J1 ) J2 ) 7.7 Hz), 7.65 (2H, m), 7.85 (1H, d, J ) 7.3

4202 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17
Villain-Guillot et al.
Hz), 7.89 (1H, d, J ) 7.8 Hz), 7.97 (1H, s), 7.99 (1H, s), 8.30 (1H,
s); MS: 380 ([M - H]-); anal. (C20H15NO3S2): C, H, N, S.
(GENbag anaer, bioMe´rieux, Marcy l’Etoile, France). The test
medium for S. pneumoniae was Mueller-Hinton broth (MHB)
supplemented with 0.5% (v/v) of sheep blood. The inoculum was
5 × 105 CFU/mL for every microorganism. The inoculated
microplates were incubated at 37 °C for 18 h before being read.
Cytotoxicity. Chinese hamster ovary (CHO) cells were grown
in 96-well plates in RPMI medium supplemented with 5% fetal
calf serum. The cells were incubated for 24 h at 37 °C in the absence
of serum and in the presence of the compounds, and the cytotoxicity
was measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltet-
razolium bromide at a final concentration of 300 µg/mL. After 3 h
incubation at 37 °C in the dark, acidified isopropanol (10 µL
isopropanol, 0.4 M HCl) was added and mixed in the well. The
plates were incubated at room temperature for 1 h in the dark. The
solution optical density in the 96-well plate was measured at 595
nm using a spectrophotometer.
S. epidermidis Biofilms. The wells of a black 96-well polystyrene
microtiter plate (Greiner bio-one, Frickenhausen, Germany) were
filled with 100 µL aliquots of S. epidermidis inoculum (107 CFU/
mL) in MHB; the plate was then incubated for 24 h at 37 °C. Each
well was rinsed three times with 200 µL sterile water to remove
nonadherent bacteria. Subsequently, 100 µL MHB containing the
desired antibiotic concentration was added to each well, and the
plate was incubated at 37 °C for 24 h without shaking. After the
challenge, the plate was washed three times with 200 µL sterile
water, and 10 µL of DMSO was added to each well to lyse the
bacteria. The plate was shaken for 3 min, and 90 µL of ATP-
counting bioluminescence buffer (25 mM Hepes, 24 mM MgCl2,
2 mM ethylenediamine tetra-acetate, 2 mM dithiothreitol, 60 µg/
mL bovine serum albumine, 20 µg/mL luciferase, 0.6 mM D-
luciferin, pH 7.25) was added to each well before being read by a
luminometer (Chameleon, Hidex, Turku, Finland).32 All of the
experiments were done in triplicate.
5-Bromomethyl-2-furaldehyde (9). A solution of 5-hydroxym-
ethyl-2-furaldehyde (150 mg, 1.19 mmol) in CHCl3 (2 mL) was
added dropwise to a solution of Me3SiBr (462 µL, 3.57 mmol) in
CHCl3 (8 mL) under an argon atmosphere.30 The reaction mixture
was stirred at 45 °C for 6 h. The reaction mixture was then cooled
to room temperature and was diluted with saturated NaHCO3
solution (6 mL). The reaction mixture was extracted with CH2Cl2
(3 × 50 mL), and the combined organic fractions were washed
with brine (1 × 10 mL), were dried over MgSO4, were concentrated
under reduced pressure, and were purified by flash column
chromatography to provide 198 mg of 9 in an 88% yield as a brown
oil. Thin layer chromatography: retention factor 0.6 (CH2Cl2).
4-(5-Formyl-furan-2-ylmethyl)-benzoic Acid (10). A solution
of K2CO3 (92 mg, 0.66 mmol) in water (1 mL) was added to a
mixture of 4-carboxyphenylboronic acid (50 mg, 0.37 mmol) and
9 (63 mg, 0.33 mmol) in toluene/ethanol (10 mL). (Ph3P)4Pd (4
mg, 3 nmol) was added, and the reaction mixture was stirred at
90 °C for 7 h. The reaction mixture was cooled to room temperature
and was diluted with water (5 mL), and the pH was adjusted to 1
by addition of a solution of HCl (6 N). The diluted reaction mixture
was extracted with CH2Cl2 (3 × 100 mL), and the combined organic
fractions were washed with brine (1 × 10 mL), were dried over
MgSO4, and were concentrated under reduced pressure to afford
45 mg of 10 in a 59% yield. TLC: Rf 0.7 (ethyl acetate).
4-(5-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-furan-
2-ylmethyl)-benzoic Acid (11a). 3-Allylrhodanine (17 mg, 0.10
mmol) was added to a solution of 8 (22 mg, 0.10 mmol) and
ethylenediamine diacetate (52 mg, 0.29 mmol) in methanol (3 mL).
The reaction mixture was stirred at room temperature for 16 h.
The reaction mixture was then diluted with water/acetonitrile (3:1,
0.1% TFA, 40 mL) and was purified by reverse-phase HPLC with
a linear gradient of 15-100% acetonitrile (0.1% TFA) over 60 min
at 10 mL/min. Relevant fractions were combined, and the solvent
was removed by freeze-drying to afford 1 mg of 11a in a 4% yield.
MS: 386 ([M + H]+).
To establish the correlation between luminescence values and
the amount of biofilm-embedded cells, bacteria of a 24 h biofilm
grown in a well of the microtiter plate were resuspended in sterile
water by sonication using a Branson 450 Sonifier with a microtip
(4 × 2 s, 10% of the maximal amplitude). The bacterial suspension
was serially diluted in sterile water; 10 µL of this serial dilution
was drawn from the wells to be measured by bioluminescence while
the number of bacteria in the wells was estimated by classical plate
counting. Measures by bioluminescence and by plate counting
afforded establishment of the following correlation equations:
2-[3-Allyl-4-oxo-2-thioxo-thiazolidinylidenemethyl]-1H-indole-
7-carboxylic Acid (11b). 3-Allylrhodanine (23 mg, 0.13 mmol)
was added to a solution of 3-formyl-1H-indole-7-carboxylic acid
(25 mg, 0.13 mmol) and ethylenediamine diacetate (71 mg, 0.40
mmol) in methanol (3 mL); the reaction was continued as described
for the synthesis of 1 to afford 25 mg of a mixture of 11b in a
1
56% yield. H NMR (400 MHz, DMSO): δ 4.65 (2H, d, J ) 5.2
S. epidermidis RP62A
Hz), 5.13 (1H, d, J ) 17.2 Hz), 5.19 (1H, d, J ) 10.4 Hz), 5.86
(1H, m), 7.27 (1H, dd, J1 ) J2 ) 7.6 Hz), 7.73 (1H, s), 7.81 (1H,
d, J ) 7.4 Hz), 8.12 (1H, d, J ) 7.9 Hz), 8.14 (1H, s); MS: 345
([M + H]+); anal. (C16H12N2O3S2): C, H, N, S.
Log10 (CFU/well) ) 0.90 × Log10 (luminescence) + 0.85
R2 ) 0.9994
S. epidermidis 40004
Transcription Inhibition Assay. Transcription inhibition was
evaluated using a commercial RNAP inhibitor screening kit (Kool
NC-45, Epicentre Biotechnologies, Madison, WI). This screening
assay is based on the Rolling Circle Transcription technology and
relies on the property that some small circular, single-stranded
DNAs mimic an open promoter and are efficiently transcribed by
RNAP. This assay does not rely on σ factors and therefore was
performed without any.
Log10 (CFU/well) ) 0.93 × Log10 (luminescence) + 0.83
R2 ) 0.9911
S. epidermidis 48155
Log10 (CFU/well) ) 0.94 × Log10 (luminescence) + 0.91
R2 ) 0.9982
The same procedure was applied with a 48 h biofilm, and no
significant differences were observed between the correlation
equations (Supporting Information, Table S4). Thus, 24 h correlation
equations were used in this study to assess the amount of biofilm-
embedded bacteria.
HSA Binding Ranking by SPR. SPR experiments were
performed at 25 °C using a BIACORE 3000 apparatus (Biacore
AB, Uppsala Sweden). HSA (Sigma St. Louis, MO) diluted to 40
µg/mL in 10 mM acetate buffer, pH 5.2, was immobilized on CM5
sensor chips using amine-coupling chemistry. Immobilization levels
were between 9000 and 11 000 RU (resonant unit). The surface
was blocked with 1 M ethanol amine, pH 8.0, and was washed by
three 30 s pulses of 50 mM NaOH to remove noncovalently bound
HSA.
MIC Determinations and Bacterial Strains. The MICs were
determined as recommended by the CLSI.31 Compounds predis-
solved in DMSO were tested at final concentrations (prepared from
serial 2-fold dilutions) ranging from 100 to 0.4 µg/mL. The MIC
was defined as the lowest concentration which yielded no visible
growth. MICs were evaluated for S. aureus CIP 76.25, S. epider-
midis RP62A CIP 105777, E. coli CIP 76.24, E. coli tolC CGSC
5633, C. difficile DSM 1296, and S. pneumoniae CIP 103566. The
multiresistant S. aureus strain 49589 (see Supporting Information,
Table S3 for antibiotics susceptibility) and S. epidermidis strains
40004 and 48155 were clinical isolates obtained from University
Hospital of Nˆımes, France. The nonpathogenic B. anthracis strain
9131 was obtained from Institut Pasteur, France. C. difficile was
grown in Wilkins-Chalgren broth under anaerobic conditions

SAR of Antibacterial Inhibitors of RNA Polymerase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17 4203
(13) Zhang, G. Ligand-free Suzuki-Miyaura reaction catalysed by Pd/C
at room temperature. J. Chem. Res. 2004, 593-595.
(14) Howard, M. H.; Cenizal, T.; Gutteridge, S.; Hanna, W. S.; Tao, Y.;
Totrov, M.; Wittenbach, V. A.; Zheng, Y. J. A novel class of
inhibitors of peptide deformylase discovered through high-throughput
screening and virtual ligand screening. J. Med. Chem. 2004, 47,
6669-6672.
(15) Wermuth, C. G. Molecular variations based on isosteric replacements.
In The Practice of Medicinal Chemistry, 2nd ed.; Wermuth, C. G.,
Ed.; Academic Press: London, UK, 2003; pp 189-214.
(16) Toya, M.; Takagi, M.; Kondo, T.; Nakata, H.; Isobe, M.; Goto, T.
Improved synthetic methods of firefly luciferin derivatives for use
in bioluminescent analysis of hydrolytic enzymes; carboxylic esterase
and alkaline phosphatase. Bull. Chem. Soc. Jpn. 1992, 65, 2604-
2610.
(17) Giacomelli, G.; Porcheddu, A.; Salaris, M. Simple one-flask method
for the preparation of hydroxamic acids. Org. Lett. 2003, 5, 2715-
2717.
(18) Zervosen, A.; Lu, W. P.; Chen, Z.; White, R. E.; Demuth, T. P., Jr.;
Frere, J. M. Interactions between penicillin-binding proteins (PBPs)
and two novel classes of PBP inhibitors, arylalkylidene rhodanines
and arylalkylidene iminothiazolidin-4-ones. Antimicrob. Agents
Chemother. 2004, 48, 961-969.
(19) Grant, E. B.; Guiadeen, D.; Baum, E. Z.; Foleno, B. D.; Jin, H.;
Montenegro, D. A.; Nelson, E. A.; Bush, K.; Hlasta, D. J. The
synthesis and SAR of rhodanines as novel class C beta-lactamase
inhibitors. Bioorg. Med. Chem. Lett. 2000, 10, 2179-2182.
(20) Forino, M.; Johnson, S.; Wong, T. Y.; Rozanov, D. V.; Savinov, A.
Y.; Li, W.; Fattorusso, R.; Becattini, B.; Orry, A. J.; Jung, D.;
Abagyan, R. A.; Smith, J. W.; Alibek, K.; Liddington, R. C.; Strongin,
A. Y.; Pellecchia, M. Efficient synthetic inhibitors of anthrax lethal
factor. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 9499-9504.
(21) Nie, Z.; Perretta, C.; Lu, J.; Su, Y.; Margosiak, S.; Gajiwala, K. S.;
Cortez, J.; Nikulin, V.; Yager, K. M.; Appelt, K.; Chu, S. Structure-
based design, synthesis, and study of potent inhibitors of beta-
ketoacyl-acyl carrier protein synthase III as potential antimicrobial
agents. J. Med. Chem. 2005, 48, 1596-1609.
Drugs were prepared as 10 mM stock solutions in 100% DMSO.
Drugs were diluted in PBS to make a final DMSO concentration
of 5%. Binding studies were measured in phosphate buffer saline
containing 5% DMSO with a running buffer flow rate of 30 µL/
min. For ranking of drug compound binding to HSA, duplicate
injections of drugs at 50 µM were done during 60 s over the HSA
and reference surface. The surfaces were washed with running
buffer until complete regeneration was reached. For cleaning the
flow system, a bypass wash was performed with DMSO 50%
and DMSO 5% between each injection. Buffer blanks were in-
jected for double referencing33 before each drug injection, and
binding responses were corrected for DMSO bulk differences via
calibration curves (eight DMSO solutions between 4.5 and 5.8%
DMSO).29
Acknowledgment. We are grateful to Dr. S. Michaux-
Charachon (University Hospital, Nˆımes, France) for the gift of
clinical isolate bacterial strains and to Dr. M. Mock (Institut
Pasteur, Paris, France) for the gift of the nonvirulent B. anthracis
strain. We thank N. Aslaoui and N. Ricard for technical support.
P. Villain-Guillot was supported by a postgraduate fellowship
from DGA. This work was supported by institutional funds from
the Centre National de la Recherche Scientifique and the Grant
Biose´curite´ 2004.
Supporting Information Available: Antibacterial activity on
E. coli, cytotoxicity on CHO cells; biological activity of some
allylrhodanine derivatives; antibiotic susceptibility of the multire-
sistant S. aureus clinical isolate 49589; correlation equations
between luminescence values and the amount of 48 h old biofilm-
embedded bacteria; elemental analyses; HPLC purity. This material
is available free of charge via the Internet at http://pubs.acs.org.
(22) Tedder, M. E.; Nie, Z.; Margosiak, S.; Chu, S.; Feher, V. A.; Almassy,
R.; Appelt, K.; Yager, K. M. Structure-based design, synthesis, and
antimicrobial activity of purine derived SAH/MTA nucleosidase
inhibitors. Bioorg. Med. Chem. Lett. 2004, 14, 3165-3168.
(23) Molteni, V.; He, X.; Nabakka, J.; Yang, K.; Kreusch, A.; Gordon,
P.; Bursulaya, B.; Warner, I.; Shin, T.; Biorac, T.; Ryder, N. S.;
Goldberg, R.; Doughty, J.; He, Y. Identification of novel potent
bicyclic peptide deformylase inhibitors. Bioorg. Med. Chem. Lett.
2004, 14, 1477-1481.
(24) Cerca, N.; Martins, S.; Cerca, F.; Jefferson, K. K.; Pier, G. B.;
Oliveira, R.; Azeredo, J. Comparative assessment of antibiotic
susceptibility of coagulase-negative staphylococci in biofilm versus
planktonic culture as assessed by bacterial enumeration or rapid XTT
colorimetry. J. Antimicrob. Chemother. 2005, 56, 331-336.
(25) Lee, J. Y.; Ko, K. S.; Peck, K. R.; Oh, W. S.; Song, J. H. In vitro
evaluation of the antibiotic lock technique (ALT) for the treatment
of catheter-related infections caused by staphylococci. J. Antimicrob.
Chemother. 2006, 57, 1110-1115.
(26) Li, Z.; Francisco, G. D.; Hu, W.; Labthavikul, P.; Petersen, P. J.;
Severin, A.; Singh, G.; Yang, Y.; Rasmussen, B. A.; Lin, Y. I.;
Skotnicki, J. S.; Mansour, T. S. 2-Phenyl-5,6-dihydro-2H-thieno[3,2-
c]pyrazol-3-ol derivatives as new inhibitors of bacterial cell wall
biosynthesis. Bioorg. Med. Chem. Lett. 2003, 13, 2591-2594.
(27) Payne, D. J.; Miller, W. H.; Berry, V.; Brosky, J.; Burgess, W. J.;
Chen, E.; DeWolf, W. E., Jr.; Fosberry, A. P.; Greenwood, R.; Head,
M. S.; Heerding, D. A.; Janson, C. A.; Jaworski, D. D.; Keller, P.
M.; Manley, P. J.; Moore, T. D.; Newlander, K. A.; Pearson, S.;
Polizzi, B. J.; Qiu, X.; Rittenhouse, S. F.; Slater-Radosti, C.; Salyers,
K. L.; Seefeld, M. A.; Smyth, M. G.; Takata, D. T.; Uzinskas, I. N.;
Vaidya, K.; Wallis, N. G.; Winram, S. B.; Yuan, C. C.; Huffman,
W. F. Discovery of a novel and potent class of FabI-directed
antibacterial agents. Antimicrob. Agents Chemother. 2002, 46, 3118-
3124.
References
(1) Rice, L. B. Unmet medical needs in antibacterial therapy. Biochem.
Pharmacol. 2006, 71, 991-995.
(2) Costerton, J. W.; Stewart, P. S.; Greenberg, E. P. Bacterial biofilms:
a common cause of persistent infections. Science 1999, 284, 1318-
1322.
(3) Villain-Guillot, P.; Bastide, L.; Gualtieri, M.; Leonetti, J. P. Progress
in targeting bacterial transcription. Drug DiscoVery Today 2007, 12,
200-208.
(4) O’Neill, A.; Oliva, B.; Storey, C.; Hoyle, A.; Fishwick, C.; Chopra,
I. RNA polymerase inhibitors with activity against rifampin-resistant
mutants of Staphylococcus aureus. Antimicrob. Agents Chemother.
2000, 44, 3163-3166.
(5) Finegold, S. M.; Molitoris, D.; Vaisanen, M. L.; Song, Y.; Liu, C.;
Bolanos, M. In vitro activities of OPT-80 and comparator drugs
against intestinal bacteria. Antimicrob. Agents Chemother. 2004, 48,
4898-4902.
(6) Trampuz, A.; Zimmerli, W. Antimicrobial agents in orthopaedic
surgery: Prophylaxis and treatment. Drugs 2006, 66, 1089-1105.
(7) Artsimovitch, I.; Chu, C.; Lynch, A. S.; Landick, R. A new class of
bacterial RNA polymerase inhibitor affects nucleotide addition.
Science 2003, 302, 650-654.
(8) Andre´, E.; Bastide, L.; Michaux-Charachon, S.; Gouby, A.; Villain-
Guillot, P.; Latouche, J.; Bouchet, A.; Gualtieri, M.; Leonetti, J. P.
Novel synthetic molecules targeting the bacterial RNA polymerase
assembly. J. Antimicrob. Chemother. 2006, 57, 245-251.
(9) Arhin, F.; Belanger, O.; Ciblat, S.; Dehbi, M.; Delorme, D.; Dietrich,
E.; Dixit, D.; Lafontaine, Y.; Lehoux, D.; Liu, J.; McKay, G. A.;
Moeck, G.; Reddy, R.; Rose, Y.; Srikumar, R.; Tanaka, K. S.;
Williams, D. M.; Gros, P.; Pelletier, J.; Parr, T. R., Jr.; Far, A. R. A
new class of small molecule RNA polymerase inhibitors with activity
against Rifampicin-resistant Staphylococcus aureus. Bioorg. Med.
Chem. 2006, 14, 5812-5832.
(28) Kragh-Hansen, U.; Chuang, V. T.; Otagiri, M. Practical aspects of
the ligand-binding and enzymatic properties of human serum albumin.
Biol. Pharm. Bull. 2002, 25, 695-704.
(29) Frostell-Karlsson, A.; Remaeus, A.; Roos, H.; Andersson, K.; Borg,
P.; Hamalainen, M.; Karlsson, R. Biosensor analysis of the interaction
between immobilized human serum albumin and drug compounds
for prediction of human serum albumin binding levels. J. Med. Chem.
2000, 43, 1986-1992.
(10) Gualtieri, M.; Bastide, L.; Villain-Guillot, P.; Michaux-Charachon,
S.; Latouche, J.; Leonetti, J. P. In vitro activity of a new antibacterial
rhodanine derivative against Staphylococcus epidermidis biofilms.
J. Antimicrob. Chemother. 2006, 58, 778-783.
(11) Kratochwil, N. A.; Huber, W.; Muller, F.; Kansy, M.; Gerber, P. R.
Predicting plasma protein binding of drugs: a new approach.
Biochem. Pharmacol. 2002, 64, 1355-1374.
(12) Wiesner, J.; Mitsch, A.; Jomaa, H.; Schlitzer, M. Structure-activity
relationships of novel anti-malarial agents. Part 7: N-(3-benzoyl-4-
tolylacetylaminophenyl)-3-(5-aryl-2-furyl)acrylic acid amides with
polar moieties. Bioorg. Med. Chem. Lett. 2003, 13, 2159-2161.
(30) Komla, S.; Rigal, L.; Gaset, A. Synthesis of 5-(bromomethyl)- and
of 5-(chloromethyl)-2-furancarboxaldehyde. Carbohydr. Res. 1989,
187, 15-23.

4204 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 17
Villain-Guillot et al.
(31) National Committee for Clinical Laboratory Standards. Methods for
Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow
Aerobically, 6th ed.; Approved Standard M07-A6; National Com-
mittee for Clinical Laboratory Standards: Villanova, PA, 2003.
(32) Gracia, E.; Fernandez, A.; Conchello, P.; Alabart, J. L.; Perez, M.;
Amorena, B. In vitro development of Staphylococcus aureus biofilms
using slime-producing variants and ATP-bioluminescence for auto-
mated bacterial quantification. Luminescence 1999, 14, 23-31.
(33) Myszka, D. G. Improving biosensor analysis. J. Mol. Recognit. 1999,
12, 279-284.
JM0703183