New synthetic analogues of N-acyl homoserine lactones as agonists or antagonists of transcriptional regulators involved in bacterial quorum sensing

Article abstract of DOI:10.1016/S0960-894X(02)00124-5

A series of 22 novel synthetic N-acyl-homoserine lactone analogues has been evaluated for both their inducing activity and their ability to competitively inhibit the action of 3-oxo-hexanoyl-L-homoserine lactone, the natural inducer of bioluminescence in the bacterium Vibrio fischeri. In the newly synthesized analogues, the extremity of the acyl chain was modified by introducing ramified alkyl, cycloalkyl or aryl substituents at the C-4 position. Most of the analogues bearing either acyclic or cyclic alkyl substituents showed inducing activity. In contrast, the phenyl substituted analogues displayed significant antagonist activity. We hypothesized that the antagonist activity of the phenyl compounds may result from the interaction between the aryl group and aromatic amino acids of the LuxR receptor, preventing it from adopting the active dimeric form.

Full text of DOI:10.1016/S0960-894X(02)00124-5

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1153–1157
New Synthetic Analogues of N-Acyl Homoserine Lactones as
Agonists or Antagonists of Transcriptional Regulators Involved
in Bacterial Quorum Sensing
a,
b
b
b
Sylvie Reverchon, * Bernard Chantegrel, Christian Deshayes, Alain Doutheau
a
and Nicole Cotte-Pattat
aUnite´ de Microbiologie et Ge´n e´ tique CNRS-INSA-UCB UMR 5122, INSA, Batiment Louis Pasteur, 11 Avenue Jean Capelle,
69621 Villeurbanne, France
Laboratoire de Chimie Organique, INSA, Batiment Jules Verne, 17 Avenue Jean Capelle, 69621 Villeurbanne, France
b
Received 17 October 2001; revised 24 January 2002; accepted 18 February 2002
Abstract—A series of 22 novel synthetic N-acyl-homoserine lactone analogues has been evaluated for both their inducing activity
and their ability to competitively inhibit the action of 3-oxo-hexanoyl-l-homoserine lactone, the natural inducer of bioluminescence
in the bacterium Vibrio fischeri. In the newly synthesized analogues, the extremity of the acyl chain was modified by introducing
ramified alkyl, cycloalkyl or aryl substituents at the C-4 position. Most of the analogues bearing either acyclic or cyclic alkyl sub-
stituents showed inducing activity. In contrast, the phenyl substituted analogues displayed significant antagonist activity. We
hypothesized that the antagonist activity of the phenyl compounds may result from the interaction between the aryl group and
aromatic amino acids of the LuxR receptor, preventing it from adopting the active dimeric form. # 2002 Elsevier Science Ltd. All
rights reserved.
Introduction
these different studies, the following general features
emerge: the homoserine lactone ring is very important
for biological activity while the nature of the acyl chain
is not as critical. Previous attempts to obtain acyl-HSL
antagonists by modifying the acyl chain have focused
mainly on its length, on the modification of the 3-oxo
functionality and on the introduction of unsaturations.
In contrast, compounds with ramified substituents have
received less attention. As part of a program aimed at
obtaining compounds with antagonist activity, we syn-
thesized a range of analogues of 3-oxo-C6-HSL and C6-
HSL bearing either ramified, cycloalkyl or aryl sub-
stituents at the C-4 position of the acyl chain. We then
studied their ability to competitively inhibit the activity
of natural inducers in the V. fischeri LuxR model system.
N-Acyl-l-homoserine lactones (acyl-HSLs) are pro-
duced by a variety of Gram-negative bacteria which use
them as extracellular quorum sensing signals. Depend-
ing on the bacterial species, these molecules differ in the
length of their acyl chain (4–14 carbons) and in the
eventual presence of double bonds or functional groups
1
(
keto, hydroxyl). By activating proteins belonging to the
LuxR family of transcriptional regulators, these signals
allow for population density-dependent expression of
multiple genes. Many pathogens, such as Pseudomonas
2
3
aeruginosa, Erwinia carotovora or Erwinia chry-
4
santhemi, use acyl-HSLs to regulate the production of
virulence factors. Thus, an attractive way of controlling
virulence could be to interfere with quorum sensing sig-
nals by designing acyl-HSL analogues that bind to, but
fail to activate, the LuxR-type transcriptional reg-
ulators. The design of N-acyl-HSL analogues has been
previously performed in four different systems: the 3-
Experimental Procedures
Synthesis of acyl-HSL analogues10
5
,6
oxo-C6-HSL and LuxR of Vibrio fischeri, the 3-oxo-
7
C6-HSL and CarR of E. carotovora, the 3-oxo-C8-
8
Racemic mixtures of 3-oxo-C6-HSL analogues 1–5, 7–
11
HSL and TraR of Agrobacterium tumefaciens and the
9
19 were obtained through a two step synthesis: the
3
-oxo-C12-HSL and LasR of P. aeruginosa. From
reaction of the requisite acid chloride with Meldrum’s
acid, in the presence of pyridine in dichloromethane, at
room temperature and the subsequent work up yielded
crude intermediate acyl Meldrum’s acid, which was then
*
8
Corresponding author. Tel.: +33-4-7243-8088; fax: +33-4-7243-
714; e-mail: revercho@insa-lyon.fr
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00124-5

1154
S. Reverchon et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1153–1157
dissolved in 1,2-dichloroethane and refluxed in the
presence of d,l-homoserine lactone hydrobromide and
pyridine. Racemic mixtures of C6-HSL analogues 6, 20–
HSL, the most efficient inducer and C6-HSL, a secondary
inducer. We synthesized a range of derivatives bearing a
ramified alkyl (1, 2), cycloalkyl (3–6), aryl (7–17) or
heteroaromatic (18, 19) substituent at the C-4 position
of the 3-oxo-acyl chain. Three analogues of the C6-
2
2 were prepared by reacting, at room temperature,
d,l-homoserine lactone hydrobromide, pyridine and the
corresponding acid chlorides in dichloromethane. After
the usual workup, crude products were purified by silica
gel chromatography and obtained in yields ranging from
6
HSL, bearing either a phenyl group at C-4 (20 ) or a
phenyl group at C-4 together with either an oxygen or a
sulphur atom instead of the C-3 methylene (21, 22),
were also prepared.
2
0 to 70%.
Bacterial strain and culture conditions
Effect of acyl-HSL analogues on the induction of
bioluminescence
We used the recombinant Escherichia coli strain NM522
containing the sensor plasmid pSB401 to measure the
induction of luminescence by various acyl-HSL analo-
gues. In pSB401, the LuxR and the LuxI promoter from
V. fischeri have been coupled to the entire Lux struc-
tural operon (LuxCDABE) from Photorhabdus lumines-
cens. This construct, present in the biosensor strain
The ability of each analogue to activate the LuxR pro-
tein was determined by following the luminescence of
the biosensor strain. The natural inducers 3-oxo-C6-
HSL and C6-HSL were used as controls (Fig. 1). Four
compounds (1, 3, 4, 6) with alkyl or cycloalkyl sub-
stitution activated luminescence whereas the other
compounds showed no inducing activity. Compound 3
(cyclopentyl) is as active as 3-oxo-C6-HSL whereas 1
(isopropyl) and 4 (cyclohexyl) displayed a lower activ-
ity. Compound 6 (cyclohexyl with a C-3 methylene)
appears to be as active as the corresponding C6-HSL
natural inducer. These results indicate that the inducing
activity is more or less retained if one branching is
introduced at the C-5 position of the acyl chain, either
in acyclic or cyclic form. The introduction of two
branchings at C-5 resulted in the absence of any inducing
activity, as shown with compounds 2 (tert-butyl) and 5
(adamentyl). Interestingly, all of the aryl compounds
(7–17) were also deprived of inducing activity, indicat-
ing that the flattening of the chain extremity due to the
(
NM522/pSB401), responds to a wide range of acyl-
1
2
HSLs by producing bioluminescence. Bacterial cul-
tures were grown in Luria broth, in the presence of tet-
racycline (20 mg mL ), at 30 C.
À1
ꢀ
Detection of biological activity using the biosensor strain
The inducing activity of the various acyl-HSL analogues
was monitored using the E. coli biosensor strain. Acyl-
HSL activity was measured in a microtitre plate format,
with bioluminescence quantified using a Luminoskan
luminometer. Concentrations of analogues, ranging
from 20 to 40 mM, were made up to 0.1 mL volumes
with growth medium. The same volume (0.1 mL) of a
3
1
strain was then added and the plate was incubated at
:10 dilution of an overnight culture of the biosensor
hybridization modification of the C-5 atom, from sp to
2
sp , resulted in a dramatic effect.
ꢀ
3
0 C. The amount of light produced by the bacteria was
detected after 4–5 h, when the ratio of induced to back-
ground light was at its maximum. The amount of light
measured was expressed in relative light units (RLU).
Inhibition of 3-oxo-C6-HSL activity by analogues
The analogues devoid of inducing activity were tested
for their ability to interfere with the induction of lumi-
nescence by 3-oxo-C6-HSL. The alkyl derivative 2,
bearing a tert-butyl group at C-4, proved to be less
active that all the 4-aryl-3-oxo-derivatives (7–15) (Figs.
2and 3). Among the latter compounds, the antagonist
activity appeared to be of the same significant magnitude
with the phenyl compound 7 and the p-substituted
phenyl compounds 8, 9 and 10 bearing, respectively, a
chloro, methoxy or bromo group. The activity was slightly
lower with the p-fluorophenyl and tolyl derivatives 11 and
12 and significantly decreased in the case of the p-tri-
fluoromethylphenyl derivative 13 (Fig. 2). The influence of
the position of the substituent on the aromatic ring was
examined with chloro derivatives (Fig. 3). A decrease in
inhibitory activity was observed following the sequence
para (8), meta (15) and ortho (14). Thus the best antago-
nist activity was obtained with a phenyl group or a phenyl
group bearing a heteroatom in position para.
Competition assays using the biosensor strain
The influence of acyl-HSL analogues on the induction
of bioluminescence by 3-oxo-C6-HSL was determined
as described above, except that 3-oxo-C6-HSL was
included at a final concentration of 200 nM together
with the analogue. This concentration of 3-oxo-C6-HSL
was required for 1/2maximal induction of luminescence
under our conditions. This value is 2- to 4-fold higher
5
,6
than that previously reported. This is probably due to
the fact that we used a racemic mixture of 3-oxo-C6-
1
3
HSL. In addition, bioassays were performed under
different cultural conditions.
Results
Structures of synthetic analogues of the natural LuxR
inducers
Naphthyl (16) and biphenyl (17) compounds (Fig. 4)
showed no inhibitory activity, presumably because of
steric hindrance limitation; this kind of limitation was
also observed with the alkyl derivative 5 bearing a bulky
adamentyl group at C-4 which displayed no antagonist
We prepared a series of analogues of the two natural
inducers of the LuxR regulator which are produced by
the acyl-HSL synthase LuxI of V. fischeri: 3-oxo-C6-

S. Reverchon et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1153–1157
1155
Figure 1. Agonist effects of N-acyl-HSL analogues 1, 2, 3, 4, 5, 6 and 7. The concentrations (mM) required for half-maximal activation are given in
parentheses.
Figure 2. Antagonist effects of N-acyl-HSL analogues 2, 7, 8, 9, 10, 11, 12 and 13. The concentrations (mM) required for 50% inhibition are given in
parentheses.
activity. Surprisingly, when the phenyl group in compound
1
(5) or aryl (16, 17) moieties are deprived of any activity,
indicating that the presence of an excessively bulky
group prevents interaction with LuxR.
4
7 was replaced by an isosteric thiophenyl group (18
and 19, Fig. 4), no antagonist activity was observed
either.
To our knowledge, such clear-cut differences of behaviour
among HSL analogues bearing alkyl and aryl sub-
stituents have not previously been reported. It is very
To evaluate the influence of the 3-oxo moiety for
antagonist activity, we tested the C6-HSL analogues 20,
2
1 and 22 bearing a phenyl substituent at C-4. These
compounds were inhibitors but they appeared to be
slightly less efficient than the corresponding 3-oxo-C6-
HSL analogue 7 (Fig. 5). Thus the 3-oxo moiety, which is
important for the inducing capacity of natural inducers,
also favours the antagonist activity of phenyl derivatives.
This statement is in agreement with previous results
6
from the literature.
Discussion
Considering their biological activity, secondary alkyl or
aryl acyl-HSLs clearly delimit two classes of com-
pounds: secondary alkyl derivatives (1, 3, 4, 6) show
agonist activity while the aryl derivatives (7–15, 20–22)
and, to a lesser extent, the tertiary alkyl derivative 2
show antagonist activity. Compounds with larger alkyl
Figure 3. Antagonist effects of N-acyl-HSL analogues 7, 8, 14 and 15.
The concentrations (mM) required for 50% inhibition are given in
parentheses.

1156
S. Reverchon et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1153–1157
Figure 4. Antagonist effects of N-acyl-HSL analogues 16, 17, 18 and 19.
of LuxR dimerization in the presence of 4-aryl-3-oxo-
butanoyl-HSLs.
Acknowledgements
We thank G. Condemine for his critical opinions and
V. James for reading the manuscript. We are much
indebted to Professor Loic Blum for generously making
his Luminometer available to us and to Agnes Degiuli
for guiding us in its use. This work was supported by
grants from the CNRS, from the Ministere de l’Education
`
Nationale et de la Recherche and by specific grants
from INSA (Bonus Qualite Recherche) and from the
Fonds National pour la Science 2000 (Programme
Microbiologie).
Figure 5. Antagonist effects of N-acyl-HSL analogues 7, 20, 21 and
2. The concentrations (mM) required for 50% inhibition are given in
parentheses.
2
References and Notes
unlikely that these differences can be explained on the
basis of steric effects. More probably, the presence of an
aryl group allows for a very specific interaction to take
place with the LuxR protein. The specific character of
this interaction is also illustrated by the surprising
absence of inhibitory activity observed with the isosteric
thiophenyl derivatives 18 and 19.
1. Fuqua, C.; Winans, S. C.; Greenberg, E. P. Annu. Rev.
Microbiol. 1996, 50, 727.
2. Latifi, A.; Winson, M. K.; Foglino, M.; Bycroft, B. W.;
Stewart, G. S.; Lazdunski, A.; Williams, P. Mol. Microbiol.
1995, 17, 333.
3
. Jones, S.; Yu, B.; Bainton, N. J.; Birdsall, M.; Bycroft,
B. W.; Chhabra, S. R.; Cox, A. J. R.; Golby, P.; Reeves, P. J.;
Stephens, S.; Winson, M. K.; Salmond, G. P. C.; Stewart,
G. S. A. B.; Williams, P. EMBO J. 1993, 12, 2477.
. Reverchon, S.; Bouillant, M.; Salmond, G.; Nasser, W.
Mol. Microbiol. 1998, 29, 1407.
. Eberhard, A.; Widrig, C. A.; McBath, P.; Schineller, J. B.
Arch. Microbiol. 1986, 146, 35.
6. Schaefer, A. L.; Hanzelka, B. L.; Eberhard, A.; Greenberg,
E. P. J. Bacteriol. 1996, 178, 2897.
7. Chhabra, S. R.; Stead, P.; Bainton, N. J.; Salmond, G. P.;
Stewart, G. S.; Williams, P.; Bycroft, B. W. J. Antibiot.
(Tokyo) 1993, 46, 441.
It is well recognized that LuxR-type transcriptional
regulators activate gene transcription only after dimeri-
zation. For the LuxR protein, the accepted model is
that dimerization requires the preliminary binding of
4
5
1
5
acyl-HSL. One can speculate that the antagonist
activity of 4-aryl acyl-HSLs may result from their
interaction with one or several aromatic amino acid(s)
of the LuxR protein, in the region containing the HSL-
binding site or in the dimerization domain or both.
1
6,17
Previous work indicates
that the HSL-binding site
8
. Zhu, J.; Beaber, J. W.; More, M. I.; Fuqua, C.; Eberhard,
A.; Winans, S. C. J. Bacteriol. 1998, 180, 5398.
. Passador, L.; Tucker, K. D.; Guertin, K. R.; Journet,
M. P.; Kende, A. S.; Iglewski, B. H. J. Bacteriol. 1996, 178,
of LuxR contains three aromatic residues and is fol-
lowed by the hydrophobic dimerization domain which
contains six aromatic residues.
9
5
1
995.
0. For a recent review on chemical synthesis of acyl-HSLs
We plan to use a genetic system, based on the cI
1
8
repressor of phage l, to determine whether the
mechanism of inhibition is really due to the prevention
and analogues, see: Eberhard, A.; Schineller, J. B. Methods
Enzymol. 2000, 305, 301.

S. Reverchon et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1153–1157
1157
1
1. Moya, P.; Cantin, A.; Castillo, M. A.; Primo, J.; Miranda,
M. A.; Primo-Yufera, E. J. Org. Chem. 1998, 63, 8530.
2. Winson, M. K.; Swift, S.; Fish, L.; Throup, J. P.; Jorgen-
15. Fuqua, C.; Greenberg, E. P. Curr. Opin. Microbiol. 1998,
1, 183.
16. Shadel, G. S.; Young, R.; Baldwin, T. O. J. Bacteriol.
1990, 172, 3980.
17. Slock, J.; VanRiet, D.; Kolibachuk, D.; Greenberg, E. P.
J. Bacteriol. 1990, 172, 3974.
18. Hu, J. C. Structure 1995, 3, 431.
1
sen, F.; Chhabra, S. R.; Bycroft, B. W.; Williams, P.; Stewart,
G. S. A. B. FEMS Microbiol. Lett. 1998, 163, 185.
1
3. Ikeda, T.; Kajiyama, K.; Kita, T.; Takiguchi, N.; Kuroda,
A.; Kato, J.; Ohtake, H. Chem. Lett. 2001, 314.
4. Patani, G. A.; LaVoie, E. J. Chem. Rev. 1996, 96, 3147.
1