High-throughput screening of metal-N-heterocyclic carbene complexes against biofilm formation by pathogenic bacteria

Article abstract of DOI:10.1002/cmdc.201402012

A set of molecules including a majority of metal-N-heterocyclic carbene (NHC) complexes (metal=Ag, Cu, and Au) and azolium salts were evaluated by high-throughput screening of their activity against biofilm formation associated with pathogenic bacteria. The anti-planktonic effects were compared in parallel. Representative biofilm-forming strains of various genera were selected (Listeria, Pseudomonas, Staphylococcus, and Escherichia). All the compounds were tested at 1 mg-L^-1 by using the BioFilm Ring Test. An information score (IS, sum of the activities) and an activity score (AS, difference between anti-biofilm and anti-planktonic activity) were determined from normalized experimental values to classify the most active molecules against the panel of bacterial strains. With this method we identified lipophilic Ag^I and Cu^I complexes possessing aromatic groups on the NHC ligand as the most efficient at inhibiting biofilm formation. Film screening: Metal-N-heterocyclic carbene (NHC) complexes of group 11 metals (Ag, Au, Cu) strongly inhibit biofilm formation by representative pathogenic bacteria at low concentrations. The highest and broadest activities are obtained with silver as the metal complexed with lipophilic NHCs. This opens new perspectives to develop alternative drugs to treat biofilm-associated pathogens.

Full text of DOI:10.1002/cmdc.201402012

CHEMMEDCHEM
COMMUNICATIONS
DOI: 10.1002/cmdc.201402012
High-Throughput Screening of Metal-N-Heterocyclic
Carbene Complexes against Biofilm Formation by
Pathogenic Bacteria
Thierry Bernardi,*[a] Stꢀphanie Badel,[a] Pascal Mayer,[a] Jꢀrome Groelly,[a] Pierre de Frꢀmont,[b]
Bꢀatrice Jacques,[b] Pierre Braunstein,[b] Marie-Laure Teyssot,[c] Christelle Gaulier,[c]
Federico Cisnetti,[c] Arnaud Gautier,*[c] and Sylvain Roland*[d]
A set of molecules including a majority of metal-N-heterocyclic
carbene (NHC) complexes (metal=Ag, Cu, and Au) and azoli-
um salts were evaluated by high-throughput screening of their
activity against biofilm formation associated with pathogenic
bacteria. The anti-planktonic effects were compared in parallel.
Representative biofilm-forming strains of various genera were
selected (Listeria, Pseudomonas, Staphylococcus, and Escheri-
chia). All the compounds were tested at 1 mgLꢀ1 by using the
BioFilm Ring Test. An information score (IS, sum of the activi-
ties) and an activity score (AS, difference between anti-biofilm
and anti-planktonic activity) were determined from normalized
experimental values to classify the most active molecules
against the panel of bacterial strains. With this method we
identified lipophilic AgI and CuI complexes possessing aromatic
groups on the NHC ligand as the most efficient at inhibiting
biofilm formation.
(bloodstream infections) associated with S. aureus (one of the
most frequently occurring bacteria found in clinical isolates)
and E. coli lead to significant mortality and morbidity among
patients, due to methicillin-resistant S. aureus (MRSA) and
cephalosporin-resistant E. coli strains.[2] Resistant P. aeruginosa
or MRSA are also responsible for respiratory tract infections,
which are difficult to control, resulting in high mortality world-
wide.[3] Bacterial resistance is induced by various cellular mech-
anisms involving efflux pumps, target modification, or drug in-
activation by enzymes.[4] In addition, drug resistance in refrac-
tory infections is closely connected with the presence of bacte-
rial biofilms,[5–7] which are a major cause of failure in the thera-
py of infections. Antibiotics have been shown to be 1000- to
1500-fold less active against bacteria that are confined within
a biofilm than against individual free-form planktonic bacte-
ria.[8] Strategies alternative to those involving conventional an-
tibiotic treatments are needed to control bacterial resistance
and tackle refractory infections. Among these, the develop-
ment of new drugs able to prevent pathogenic bacteria from
forming biofilms is essential.[9]
The treatment of certain bacterial infections has become
a major public health concern due to the increasing emer-
gence of pathogenic bacterial strains that are resistant to most
available antibiotics. This phenomenon involves all major mi-
crobial pathogens.[1] The increasing inefficiency of conventional
antibiotic drugs used in clinical protocols renders antibacterial
therapy more and more problematic. For instance, bacteremia
The combination of metal ions with organic molecules offers
the opportunity to develop alternative drugs with new modes
of action.[10] Although organometallic compounds are com-
monly used for the chemotherapy of some diseases such as
cancer, the development of new metal-based drugs is of par-
ticular interest in the context of increasing chemoresistance to
conventional therapeutics. This could open new possibilities
for the treatment of pathogenic bacteria.[11] In the last decade,
metal-N-heterocyclic carbene (NHC) complexes with various
metal ions have been investigated for their anticancer and an-
tibacterial properties.[12,13] Metal–NHCs are well adapted for
drug design, owing to their synthetic flexibility and to the
strength of the M–NHC bond that makes them good candi-
dates as carriers for a wide range of metal ions in biological
media.[14,15] In the antibacterial field, the most significant results
have been obtained with Ag–NHCs,[16] which are capable of in-
hibiting the growth of various bacterial strains, including
highly pathogenic strains, at very low silver concentration-
s.[12a,e,17] Metals are known as anti-biofilm agents. Metal ions
can be incorporated into the matrix of biofilms and interfere
with their formation.[18] To date, few metal–NHCs have been
evaluated for their potential against bacterial biofilms.[19] A sil-
ver(I) imidazole cyclophane gem-diol and two xanthine-derived
silver(I) complexes were shown to be effective against biofilms
of B. anthracis and MRSA at low silver concentrations. Two pyr-
[a] Dr. T. Bernardi, S. Badel, Dr. P. Mayer, J. Groelly
Biofilm Control SAS, Biopꢀle Clermont-Limagne
63360 Saint-Beauzire (France)
E-mail: thierry.bernardi@biofilmcontrol.com
[b] Dr. P. de Frꢁmont, Dr. B. Jacques, Dr. P. Braunstein
Laboratoire de Chimie de Coordination, UMR CNRS 8232
Universitꢁ de Strasbourg
4, rue Blaise Pascal, CS 90032, 67081 Strasbourg (France)
[c] Dr. M.-L. Teyssot, C. Gaulier, Dr. F. Cisnetti, Dr. A. Gautier
Clermont-Universitꢁ – Universite Blaise Pascal and ENSCCF
Institut de Chimie de Clermont-Ferrand, 63177 Aubiꢂre (France)
E-mail: arnaud.gautier@univ-bpclermont.fr
[d] Dr. S. Roland
Sorbonne Universitꢁs, UPMC Univ Paris 6
Institut Parisien de Chimie Molꢁculaire, UMR CNRS 8232
4, place Jussieu, 75005 Paris (France)
Fax: (+33)1-44273056
E-mail: sylvain.roland@upmc.fr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402012: details of biological assays,
along with references and synthetic protocols for preparation of the
compounds.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1140 – 1144 1140

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
azine-functionalized AgI and AuI complexes were also found to
affect S. mutans and E. coli biofilms.
Recently, we gathered a significant library of metal–NHCs
and azolium salts for biological application. This prompted us
to evaluate their anti-biofilm activity against representative
pathogenic bacteria. The automatized BioFilm Ring Test,
a high-throughput screening tool with high reproducibility
was used to detect molecules with high test scores. The most
relevant results for 89 compounds are presented herein.
For this study we selected various NHC complexes with
group 11 metals (CuI, AgI, and AuI), with a majority being Ag–
NHCs (Figures S1 and S2 in the Supporting Information). Most
of the complexes are heteroleptic neutral complexes of the
general formula [(NHC)MX] (M=Ag, Au, Cu) and cationic
[(NHC)2M+, Xꢀ] complexes (M=Ag, Cu). To evaluate the influ-
ence of the NHC ligand on anti-biofilm activity, various NHC
ligand types derived from imidazolium or imidazolinium salts
were tested, in which the nitrogen atoms are substituted with
aromatic, heterocyclic (triazolyl, pyridyl), aliphatic, or function-
alized groups. The activities of representative metal-complexed
N,N’-diaryl-NHCs substituted with additional polar side chains
(ammonium groups, alcohols) were also assessed. A set of imi-
dazol(in)ium salts previously prepared by our research groups
was also tested in parallel (Supporting Information Figure S3).
All compounds were tested at a final concentration of
Figure 1. a) Principle of the BioFilm Ring Test: images of the wells containing
super-paramagnetic beads after magnetization in the absence and presence
of biofilm. b) Example of a 96-well plate (S. aureus aureus CIP7625) after
magnetization, in which active compounds are visualized by the presence of
the brown spots.
The activity of metal–NHCs on planktonic growth and biofilm
formation was quantified with the BioFilm Control Elements
software tools (BioFilm Control, France). The turbidity of the
solution, estimated from an image acquired before magnetiza-
tion and contrasting liquid addition, gives an indicator of
planktonic growth, coined TURB. In parallel, the variation of
the information content between pre- and post-magnetization
images, which is a result of sub-pixel-aligned image analysis al-
gorithms,[23] gives an indicator of biofilm formation coined BFI
(BioFilm Index). These indices were normalized respectively as
TURBnorm and BFInorm, by applying the normalization transform
[ValsampleꢀValcontrol]/[Valsample +Valcontrol]). Both normalized values
range from +1 to ꢀ1, i.e., from activation to inhibition, respec-
tively, of biofilm formation and planktonic growth.
1 mgmLꢀ1
.
The anti-biofilm activity was evaluated on the following bac-
terial strains: Gram+ Listeria monocytogenes EGDe (food patho-
gen),[20] Staphylococcus aureus (ATCC25923, CIP7625), Staphylo-
coccus epidermis (CIP105777), and Gramꢀ Pseudomonas aerugi-
nosa (CIP104116, CIP76110) and Escherichia coli DH5a. The re-
sults were collected after incubation times of 4, 6, and 8 h. This
corresponds to 21 experiments per molecule.
To get a general view of the ability of the compounds to in-
hibit biofilm formation, we determined an information score
(IS), which indicates the general level of activity of a given
compound against the panel of bacterial strains at different in-
cubation times, regardless of its activity type (anti-biofilm
or bactericidal), and an activity score (AS), which indicates
the specificity of the molecule as an anti-biofilm agent.
The IS and AS values were derived from a global anti-
planktonic score (TURBscore) and a global anti-biofilm score
(BFIscore). For each compound, these values, calculated as
TURBscore =[{Count(TURBnorm >0.4)}ꢀ{Count(TURBnorm<ꢀ0.4)}]/21
and BFIscore =[{Count(BFInorm>0.4)}ꢀ{Count(BFInorm<ꢀ0.4)}]/21,
are the average differences across all 21 experiments between
the count of the number of normalized values >0.4 and that
of the number of normalized values less than ꢀ0.4. The arbi-
trary limit at 0.4 introduced in this calculation enabled the de-
tection of compounds with the most significant activities. An
information score was then expressed as IS= jBFIscore j + j
TURBscore j, and an activity score as AS=BFIscore +TURBscore. Be-
cause TURBscore is negative for compounds with significant anti-
bacterial activity, a negative AS indicates a more pronounced
antibacterial activity, whereas a positive AS demonstrates
a general anti-biofilm specificity for the compound in question.
From IS and AS values, the set of all molecules could be rap-
idly classified according to their general activity (IS) and anti-
biofilm specificity (AS) (Supporting Information Figures S4 and
S5). An initial observation of the results showed that a large
majority of the compounds have visible anti-biofilm or anti-
The standard crystal violet method has been widely used to
evaluate the propensity of a given compound to inhibit biofilm
formation. However, this method is not well adapted for rapid
screening of large libraries, mainly due to accumulation of
non-standardized manipulations (washing, staining, de-stain-
ing, and drying steps), which can lead to high result variations
for the same compound.[21] In addition, the test requires at
least 24–48 h for completion. In the present study, we pre-
ferred the automatized BioFilm Ring Test (BRT), which is much
less time-consuming owing to fewer manipulations after the
initial bacterial inoculation (no washing and staining steps).
Moreover, this method was shown to be highly reproducible.
In short, the BRT is based on the immobilization of well-dis-
persed super-paramagnetic beads by forming bacterial aggre-
gates (biofilm) with sufficient strength to overcome displace-
ment when a magnetic field is applied.[22] In the absence of
a biofilm, aggregation of the magnetic beads into a single
point is easily detectable after magnet contact at the center
bottom of the wells (Figure 1). In contrast, in the presence of
an established biofilm, the well-dispersed beads remain in
place, and no aggregation is observed upon magnetization.
Bead mobility is indicative of the status of biofilm formation,
which can be determined by comparison of the images ob-
tained in the presence or absence of a given test compound.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1140 – 1144 1141

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Figure 2. Information score (IS) and activity scores (AS) of the 13 most active
molecules selected for IS>0.4 (see text). a) IS: sum of the absolute values of
the global anti-biofilm score (BFIscore, black bars) and the global anti-plank-
tonic score (TURBscore, grey bars). b) AS=BFIscore +TURBscore: positive values
indicate more pronounced anti-biofilm activity.
planktonic activities. In a more precise analysis, we observed
that 13 metal–NHCs exhibited IS>0.4, suggesting high anti-
biofilm and anti-planktonic activities against several of the bac-
terial strains (Figure 2a). As depicted in Figure 3, this family of
active complexes is largely characterized by the presence of
lipophilic NHC ligands with bulky aromatic groups on the ni-
trogen atoms. It is exclusively composed of neutral heterolep-
tic complexes with predominantly silver as the metal. Silver is
also present in the three most active complexes (P1, C31, and
P17) for which IS>0.8, and a BFIscore value of ~0.5 was deter-
mined. Among these, C31 exhibits the highest positive AS
value, which corresponds to the highest anti-biofilm specificity
(Figure 2b). In parallel, two Cu–NHCs (C50 and C2) were also
found to display high general antibacterial activity, with
a BFIscore close to 0.3.
Figure 3. Structures of the 13 most active complexes selected for IS>0.4.
order (Figure S5). Consequently, it was possible to compare the
compounds according to their BFInorm value at 8 h, which is
representative of their general potential as anti-biofilm agents
(Table 1).
A general analysis of the data shows first that most of the
13 complexes display high activity against S. aureus strains re-
gardless of the metal (Ag or Cu). Secondly, none of the 13
complexes is active against the whole set of bacterial strains.
Besides, none of the complexes is significantly active against
S. epidermis or P. aeruginosa CIP76110. The superior scores ob-
served with P1 and C31 are clearly associated with high BFInorm
values (>0.4) for four of the strains. Both complexes strongly
inhibit biofilm formation of S. aureus and display significant ac-
tivity against E. coli and P. aeruginosa CIP104116. The highest
BFInorm value with L. monocytogenes was also observed for P1,
although the inhibition is moderate (~0.4). Both complexes
(P1 and C31) are silver chloride complexes with bulky aromatic
substituents on the NHC ligand. The other complexes can be
classified in four different categories. The first one includes
complexes C2, P29, P24 (NHC–AgCl) and P13 (NHC–CuCl),
which exclusively display anti-biofilm activity against S. aureus.
The second category involves P20, P22, P23, C34 (NHC–AgCl)
and C50 (NHC–CuCl), and is characterized by activity against
Previous results showed that P1 and P17 display different
antibacterial activities against S. aureus and E. coli, suggesting
that the NHC ligand and the halide play an important role in
the selectivity of the complexes for the bacterial strain. Similar
factors could affect the selectivity of metal–NHCs as anti-bio-
film agents. Therefore, their activity against the different bacte-
rial strains was examined separately.
To select the best method to analyze BFInorm values, we first
compared the BFInorm values at t=8 h with the average of the
BFInorm values recorded at t=4, 6, and 8 h. Classification of the
molecules according to these values gave similar results for
some bacterial strains such as P. aeruginosa CIP104116 or
S. aureus ATCC25923 (Supporting Information Figure S5). With
other strains such as E. coli, the active compounds are also de-
tected with both values, although they appear in a different
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1140 – 1144 1142

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
Table 1. BFInorm values at t=8 h for molecules with IS>0.4.[a]
Compd
Bacterial strain
E. coli
L. monocytogenes
P. aeruginosa
S. epidermis
S. aureus aureus
P. aeruginosa
S. aureus
EGDe
CIP104116
CIP105777
DH5a
CIP7625
CIP76110
ATCC25923
P1
0.398
0.290
0.287
ꢀ0.01
ꢀ0.084
0.038
0.054
ꢀ0.007
0.278
0.016
ꢀ0.264
0.193
0.407
0.543
0.651
0.045
0.137
0.229
ꢀ0.067
ꢀ0.084
ꢀ0.067
0.026
0.033
0.006
0.045
0.038
0.713
0.476
0.522
0.472
0.661
0.144
0.458
ꢀ0.031
0.432
0.236
0.063
0.343
0.703
0.913
0.919
0.229
0.920
0.916
0.887
0.914
ꢀ0.040
0.916
0.915
0.789
0.736
0.916
ꢀ0.047
ꢀ0.012
0.101
ꢀ0.093
ꢀ0.070
ꢀ0.055
0.045
0.001
0.001
ꢀ0.033
ꢀ0.026
ꢀ0.070
ꢀ0.077
0.897
0.902
0.231
0.904
0.899
0.887
0.892
ꢀ0.054
0.900
0.898
0.907
0.820
0.899
C31
P17
C34
C50
C2
P22
P15
P20
P24
P29
P13
P23
ꢀ0.110
ꢀ0.321
ꢀ0.025
ꢀ0.321
0.641
ꢀ0.325
ꢀ0.181
ꢀ0.311
0.003
0.006
ꢀ0.008
ꢀ0.142
ꢀ0.476
[a] Values are the result of one experiment. The screening strategy was oriented toward the detection of anti-biofilm molecules with high reliability against
a range of representative bacterial strains (multiple targets) rather than on obtaining highly precise results on a single target (a given bacterial strain). Low
standard deviations have been routinely obtained with the BioFilm Ring Test in repeated experiments (see ref. [22], savg ~6%).
Table 2. BFInorm values at t=8 h for selected molecules with IS<0.4.
Compd[a]
Bacterial strain
E. coli
L. monocytogenes
P. aeruginosa
S. epidermis
S. aureus aureus
P. aeruginosa
S. aureus
EGDe
CIP104116
CIP105777
DH5a
CIP7625
CIP76110
ATCC25923
LR36
P33
LR40
P4
P7
P8
0.410
0.714
0.270
0.144
0.085
ꢀ0.127
0.181
ꢀ0.077
ꢀ0.042
ꢀ0.150
0.256
ꢀ0.224
0.463
0.626
0.562
0.527
0.469
0.594
0.056
0.062
ꢀ0.057
ꢀ0.092
ꢀ0.037
0.082
0.00
ꢀ0.017
ꢀ0.063
0.050
0.019
0.150
0.093
ꢀ0.040
0.092
ꢀ0.053
ꢀ0.079
ꢀ0.062
ꢀ0.049
ꢀ0.022
0.073
0.066
0.051
0.366
0.509
ꢀ0.041
ꢀ0.018
0.031
0.009
ꢀ0.033
ꢀ0.025
ꢀ0.094
ꢀ0.094
ꢀ0.086
ꢀ0.017
ꢀ0.070
0.455
0.020
ꢀ0.101
ꢀ0.077
0.045
ꢀ0.012
0.441
P9
0.019
P14
P38
P44
ꢀ0.092
ꢀ0.014
ꢀ0.007
ꢀ0.280
ꢀ0.227
0.041
0.508
0.536
[a] Compound structures are provided in Supporting Information Figures S1–S3.
both E. coli and S. aureus. Interestingly, except P23, all these
complexes possess saturated NHC ligands derived from imida-
zolinium salts. In contrast, P15 displays a unique and interest-
ing profile owing to selective inhibition of biofilm formation
by P. aeruginosa CIP104116. Finally, P17 exhibits activity similar
to that of P15 against P. aeruginosa CIP104116 (~0.65), but was
also found to inhibit E. coli biofilm formation. Both P15 and
P17 are silver complexes with iodido ligands. This suggests
that iodide could favor selectivity against Gramꢀ bacteria. Im-
portantly, P23 is as active as P1 against E. coli and S. aureus.
However, it also promotes biofilm formation in the case of
P. aeruginosa CIP104116, with a negative BFInorm value of
ꢀ0.476.
and selectivity against S. aureus biofilm formation (Supporting
Information Table S1). Analysis of the results revealed that the
lower scores of these compounds are due to their ability to ac-
tivate biofilm formation in the case of P. aeruginosa CIP104116.
This property was observed with a dozen molecules that dis-
play negative BFInorm values for this strain, ranging from ꢀ0.3
to ꢀ0.5 (Table S1). Note that C47 and C18 are azolium salts
(NHC·HCl), precursors of P24 and C19/C21, respectively. In the
latter case, it is not clear whether the activity is metal- or
ligand-based. From the data, we also observed that the cation-
ic homoleptic Ag complex LR36 and the CuI complex P33 ex-
hibited high selectivity for L. monocytogenes biofilms (Table 2).
This particular selectivity is rarely observed across the library of
compounds. Another set of molecules (LR40, P4, P7, P8, P9,
and P14) was identified for near exclusive selectivity against
P. aeruginosa CIP104116 (Table 2). LR40 and P9 are both cation-
ic complexes with AuI or AgI, respectively, as the metal. More,
interestingly, P4, P7, P8, and P14 are Ag–NHC complexes char-
acterized by the presence of iodido ligands. Their profile is sim-
ilar to that of the AgI complex P15. Finally, the study highlight-
Next, we carefully analyzed the anti-biofilm activities of all
compounds that were not identified for IS>0.4. This was done
to check the validity of the initial selection method and to
detect other molecules with significant activities against a par-
ticular bacteria. This showed that an significant set of com-
pounds (C12, C17, C18, C19, C21, C47, and P5) exhibited a pro-
file similar to that of C2, P29, P24, and P13 with high activity
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1140 – 1144 1143

CHEMMEDCHEM
COMMUNICATIONS
www.chemmedchem.org
[11] M. Patra, G. Gasser, N. Metzler-Nolte, Dalton Trans. 2012, 41, 6350–
6358.
ed two compounds (P38 and P44), which are azolium salts, as
having notable activity against P. aeruginosa CIP76110.
[12] a) K. M. Hindi, M. J. Panzner, C. A. Tessier, C. L. Cannon, W. J. Youngs,
Chem. Rev. 2009, 109, 3859–3884; b) M. L. Teyssot, A. S. Jarrousse, M.
Manin, A. Chevry, S. Roche, F. Norre, C. Beaudoin, L. Morel, D. Boyer, R.
Mahiou, A. Gautier, Dalton Trans. 2009, 6894–6902; c) L. Mercs, M. Al-
brecht, Chem. Soc. Rev. 2010, 39, 1903–1912; d) A. Gautier, F. Cisnetti,
Metallomics 2012, 4, 23–32; e) L. Oehninger, R. Rubbiani, I. Ott, Dalton
Trans. 2013, 42, 3269–3284; f) W. Liu, R. Gust, Chem. Soc. Rev. 2013, 42,
755–773.
[13] a) D. Bourissou, O. Guerret, F. P. Gabbai, G. Bertrand, Chem. Rev. 2000,
100, 39–91; b) W. A. Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1290–
1309; Angew. Chem. 2002, 114, 1342–1363; c) F. E. Hahn, M. C. Jahnke,
Angew. Chem. Int. Ed. 2008, 47, 3122–3172; Angew. Chem. 2008, 120,
3166–3216.
[14] L. Benhamou, E. Chardon, G. Lavigne, S. Bellemin-Laponnaz, V. Cꢀsar,
Chem. Rev. 2011, 111, 2705–2733.
[15] H. Jacobsen, A. Correa, A. Poater, C. Costabile, L. Cavallo, Coord. Chem.
Rev. 2009, 253, 687–703.
[16] a) A. Kascatan-Nebioglu, M. J. Panzner, C. A. Tessier, C. L. Cannon, W. J.
Youngs, Coord. Chem. Rev. 2007, 251, 884–895; b) J. C. Garrison, W. J.
Youngs, Chem. Rev. 2005, 105, 3978–4008.
In conclusion, we have shown that various metal–NHC com-
plexes with group 11 metals, as well as some azolium salts, are
capable of inhibiting biofilm formation of representative
pathogenic bacteria at low concentrations. We have demon-
strated the pertinence of the BioFilm Ring Test as a high-
throughput screening method for the discovery of antibacterial
metallodrug candidates. The results highlight that Ag com-
plexes possessing lipophilic NHC ligands have the broadest
anti-biofilm activity, although some Cu complexes also dis-
played high scores. In addition, we demonstrated that some
metal–NHCs and azolium salts, associated with lower scores,
exhibited selective but significant activity against one or two
pathogenic bacterial strains. This opens new perspectives for
the development of new metal-based drugs for the treatment
of refractory bacterial infections associated with biofilms.
[17] S. Roland, C. Jolivalt, T. Cresteil, L. Eloy, P. Bouhours, A. Hequet, V.
Mansuy, C. Vanucci, J.-M. Paris, Chem. Eur. J. 2011, 17, 1442–1446.
[18] a) V. Hancock, M. Dahl, P. Klemm, Appl. Environ. Microbiol. 2010, 76,
3836–3841; b) F. Rivardo, M. G. Martinotti, R. J. Turner, H. Ceri, Can. J.
Microbiol. 2010, 56, 272–278; c) J. J. Harrison, R. J. Turner, D. A. Joo,
M. A. Stan, C. S. Chan, N. D. Allan, H. A. Vrionis, M. E. Olson, H. Ceri, Anti-
microb. Agents Chemother. 2008, 52, 2870–2881; d) M. L. Workentime,
J. J. Harrison, P. U. Stenroos, H. Ceri, J. Raymond, Environ. Microbiol.
2008, 10, 238–250.
[19] a) M. J. Panzner, A. Deeraksa, A. Smith, B. Wright, K. M. Hindi, A. Kasca-
tan-Nebioglu, A. Torres, B. Judy, C. Hovis, J. Hilliard, R. Mallett, E. Cope,
M. Estes, C. L. Cannon, J. Leid, W. J. Youngs, Eur. J. Inorg. Chem. 2009,
1739–1745; b) G. Roymahapatra, S. M. Mandal, W. F. Porto, T. Samanta,
S. Giri, J. Dinda, O. L. Franco, P. K. Chattaraj, Curr. Med. Chem. 2012, 19,
4184–4193.
[20] a) J. M. Farber, P. I. Peterkin, Microbiol. Rev. 1991, 55, 476–511; b) J. A.
Vazquez-Boland, M. Kuhn, P. Berche, T. Chakraborty, G. Dominguez-
Bernal, W. Goebel, B. Gonzalez-Zorn, J. Wehland, J. Kreft, Clin. Microbiol.
Rev. 2001, 14, 584–640.
[21] S. Vesterlund, J. Paltta, M. Karp, A. C. Ouwehand, J. Microbiol. Methods
2005, 60, 225–233.
[22] a) N. Bara, T. Bernardi, Int. PCT Pub. No. WO 2005/090944 A1 (and PCT/
FR05/000427), 2005; b) P. Chavant, B. Gaillard-Martinie, R. Talon, M. Hꢀ-
braud, T. Bernardi, J. Microbiol. Methods 2007, 68, 605–612.
[23] a) “Fourier and SpectralApplications”, W. H. Press, S. A. Teukolsky, W. T.
Vetterling, B. P. Flannery in Numerical Recipes in C: The Art of Scientific
Computing, 2nd Edition, Cambridge University Press, 1992, pp. 545–
547; b) D. Marr, E. Hildreth, Proc. R. Soc. London Ser. B 1980, 207, 187–
217; c) Image Analysis and Mathematical Morphology, Vol. 2: Theoretical
Advances (Ed.: J. Serra), Academic Press, London, 1988; d) B. Jahne, Dig-
ital Image Processing, 4th ed., Springer, Heidelberg, 1997.
Acknowledgements
The Auvergne Region, France, is gratefully acknowledged for
funding (PNC program).
Keywords: antibiotics · biofilms · carbene ligands · group 11
elements · high-throughput screening
[1] a) S. B. Levy, B. Marshall, Nat. Med. 2004, 10, S122–S129; b) R. E. W. Han-
cock, Nat. Rev. Drug Discovery 2007, 6, 28–34.
[2] M. E. A. de Kraker, P. G. Davey, H. Grundmann, PLoS Med. 2011, 8, 1–7.
[3] a) A. J. Morrison, Jr., R. P. Wenzel, Rev. Infect. Dis. 1984, 6, S627–S642;
b) Cystic Fibrosis Foundation 2011 Annual Report, Cystic Fibrosis Foun-
dation, Bethesda, MD (USA).
[4] a) M. C. McManus, Am. J. Health-Syst. Pharm. 1997, 54, 1420–1433;
b) F. C. Tenover, Am. J. Infect. Control 2006, 34, S3–10.
[5] a) R. Patel, Clin. Orthop. Relat. Res. 2005, 437, 41–47; b) G. G. Anderson,
G. A. O’Toole, Curr. Top. Microbiol. Immunol. 2008, 322, 85–105; c) N.
Høiby, T. Bjarnsholt, M. Givskov, S. Molin, O. Ciofu, Int. J. Antimicrob.
Agents 2010, 35, 322–332.
[6] J. W. Costerton, P. S. Stewart, E. P. Greenberg, Science 1999, 284, 1318–
1322.
[7] P. Hunter, EMBO Rep. 2008, 9, 314–317.
[8] a) H. Ceri, M. E. Olson, C. Stremick, R. R. Read, D. Morck, A. Buret, J. Clin.
Microbiol. 1999, 37, 1771–1776; b) M. Simꢂes, M. O. Pereira, M. J. Vieira,
Water Sci. Technol. 2003, 47, 217–223; c) R. Slinger, F. Chan, W. Ferris,
S. W. Yeung, M. St Denis, I. Gaboury, S. D. Aaron, Diagn. Microbiol. Infect.
Dis. 2006, 56, 247–253.
[9] M. Chen, Q. Yu, H. Sun, Int. J. Mol. Sci. 2013, 14, 18488–18501.
[10] a) N. P. E. Barry, P. J. Sadler, Chem. Commun. 2013, 49, 5106–5131;
b) C. G. Hartinger, P. J. Dyson, Chem. Soc. Rev. 2009, 38, 391–401; c) J. C.
Dabrowiak, Metals in Medicine, Wiley, Chichester, 2009.
Received: February 5, 2014
Published online on April 11, 2014
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1140 – 1144 1144