A new route towards fimbrolide analogues: Importance of the exomethylene motif in LuxR dependent quorum sensing inhibition

Article abstract of DOI:10.1039/c2md20298k

Sixteen analogues of fimbrolides (brominated 3-substituted-5-methylene- 2(5H)-furanones), naturally occurring bacterial quorum sensing (QS) inhibitors, were synthesized by a novel sequence. The importance of the methylene group, either brominated or non-brominated, in LuxR-dependent QS inhibition was demonstrated. One of the most active furanone-type QS inhibitors was identified, with an IC₅₀ of 0.6 μM. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2md20298k

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
A new route towards fimbrolide analogues: importance
of the exomethylene motif in LuxR dependent quorum
sensing inhibition†
Cite this: Med. Chem. Commun., 2013,
4, 363
ab
Mohamad Sabbah,ab Maud Bernollin,ab Alain Doutheau,ab Laurent Soulere
and Yves Queneau
*
`
ab
*
Sixteen analogues of fimbrolides (brominated 3-substituted-5-methylene-2(5H)-furanones), naturally
occurring bacterial quorum sensing (QS) inhibitors, were synthesized by a novel sequence. The
importance of the methylene group, either brominated or non-brominated, in LuxR-dependent QS
inhibition was demonstrated. One of the most active furanone-type QS inhibitors was identified, with
an IC50 of 0.6 mM.
Received 3rd October 2012
Accepted 16th November 2012
DOI: 10.1039/c2md20298k
www.rsc.org/medchemcomm
In the communication process known as quorum sensing (QS), vivo,15,16 even aer graing onto biomaterials.17 They have also
bacteria produce and release small diffusible molecules, termed been reported to potentiate the effects of disinfectants and
autoinducers (AIs), which serve as signals to gauge their own antibiotics.18,19
population density and adjust their behavior accordingly.1,2
Many analogues of these compounds have been reported in
When a certain extracellular threshold concentration of the AI is structure–activity relationship studies.11,20–24 Although the role
reached, it binds to a cognate receptor, the transcription factor, of a bromo or a dibromomethylene group, alone or with a
leading to changes in the expression of specic genes. In recent bromine atom in position 4, has been well established, the
years, QS systems have been identied in a great variety of importance of the exomethylene group itself has never been
bacteria and have been shown to control diverse phenotypes, studied in a systematic manner, and neither has the inuence
including bioluminescence, virulence factor production and of the presence/absence of a hydroxyl group on the side chain at
biolm formation.3–5 Consequently, the QS inhibition of path- position 3.
ogenic bacteria would be a promising approach for nding new
anti-virulence strategies.6–8
In keeping with our studies aimed at discovering new
inhibitors of the AHL-regulated QS system,25–28 including some
Higher organisms are able to synthesize compounds furanone-like QS inhibitors,29 we addressed these specic
interfering with QS and, in particular, several natural QS issues by synthesizing a series of sixteen 3-alkyl-a,b-unsaturated
inhibitors have been discovered in marine organisms.9 Among g-lactones, either non-substituted at position 5 (6) or equipped
these, the best known are the brominated 3-substituted-5- at this position with a methylene group (8) or a bromo-
methylene-2(5H)-furanones 1 and 2 (mbrolides), isolated from methylene group (10) (Scheme 2), which were then tested as
the Australian red marine macro-algae Delisea pulchra, which LuxR-dependent QS inhibitors.
are capable of inhibiting surface biofouling by bacteria
(Scheme 1).10
Several synthetic sequences for the preparation of natural
furanones and their synthetic analogues have been repor-
These furanones have been shown to interfere with various ted,30–37 but none of them allowed easy access to our targeted
bacterial QS systems, including important pathogens regulated molecules. We, therefore, developed an original sequence for
by the AI type 1 system (Acylhomoserine lactones, AHLs, 3) in simplied mbrolide analogues, based on the formation of the
Gram-negative bacteria (Scheme 1) and also in QS systems using
the AI type 2 system.11–14 Interestingly they are equally active on
planktonic as well as biolm-embedded cells, in vitro and in
a
ˆ
INSA Lyon, ICBMS, Bat J. Verne, 20 av A. Einstein, 69621 Villeurbanne Cedex, France.
E-mail: yves.queneau@insa-lyon.fr; laurent.soulere@insa-lyon.fr; Tel: +33 4 72 43 61
69; +33 4 72 43 83 42
b
´
ˆ
CNRS, UMR 5246, ICBMS, Universite Lyon 1, INSA-Lyon, CPE-Lyon, Bat CPE, 43 bd
du 11 novembre 1918, 69622 Villeurbanne Cedex, France
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2md20298k
Scheme 1
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun., 2013, 4, 363–366 | 363

View Article Online
MedChemComm
Concise Article
Scheme 2 Preparation of furanones 6, 8 and 10.
Table 1 Inhibition of bioluminescence for compounds 6 to 10 and natural
dibromo-furanone 2
exomethylene appendage, via the introduction, at position 5, of
a dimethylamino group38 which is subsequently eliminated,
thus allowing straightforward access to all three groups of
compounds, 6, 8 and 10, successively (Scheme 2). Lactone 4
was either alkylated or subjected to an aldolisation reaction
leading to compounds 5a,b and 5c–e in 60% to 80% isolated
yields, respectively. Compounds 5 were then subjected to an
oxidation–elimination reaction to give the corresponding
unsaturated lactone, 6a–e, without purication of the oxidized
intermediate. Introduction of the methylene group was per-
formed by rst introducing a dimethylamino group to produce
compound 7, subsequently eliminated to give compound 8. We
observed that the usual procedure, which consists of a qua-
ternization of the dimethylamino group with methyl iodide
IC50 (mM)
6
8
10
a
b
c
d
e
>200
>200
Inactive
Inactive
Inactive
>200
>200
52 (ꢀ2.3)
8 (ꢀ1.2)
15 (ꢀ1.5)
0.6a (ꢀ0.05)b
3 (ꢀ0.1)
(Z) 14 (ꢀ1.1) (E) 18 (ꢀ1.7)
6 (ꢀ0.8)
25 (ꢀ2.1)
a
Concentration (mM) required to reduce to 50% of the bioluminescence
b
intensity (IC50). Values are the means of three experiments; standard
deviation is given in brackets.
The furanones 6a and 6b displayed very low activity (IC50
>
before undergoing a Hoffmann elimination reaction,38 was not 200 mM) and the furanones 6c–e proved to be inactive (Table 1).
necessary since purication on silica gel of compound 7 Surprisingly, the compounds equipped with a methylene group
resulted directly in furanone 8. The brominated furanones 10 without a hydroxyl function, 8a and 8b, showed only low
were obtained from compounds 8 by a dibromination–elimi- inhibitory activity, unlike the other furanones which displayed a
nation sequence.36 The stereochemistry of the exocyclic potent inhibitory activity capable of totally inhibiting the
carbon–carbon double bond in bromo-lactones 10 was bioluminescence at higher concentrations. Except for 8c (IC50
¼
assigned according to their 1H NMR spectra. Indeed, it is 52 mM), they proved to be signicantly more active (IC50 ¼ 0.6 to
known that the chemical shi value for the exocyclic vinylic 25 mM) than the natural furanone 2 (IC50 ¼ 45 mM).
proton in 5-bromomethylene-2(5H)-furanones is characteristic
From these biological results, we can conclude that a
for either the (Z)-isomer (d z 6 ppm) or the (E)-isomer (d z 6.5 methylene group at position 5 is required to confer inhibitory
ppm). In our case, we observed, for the exocyclic vinylic proton, activity to furanones. However, the benecial effect on the
a singlet at either 5.93 ppm (10a-Z), (10b-Z), or 6.06 ppm (10c- activity of this group is much more pronounced with lactones
Z), (10d-Z), (10e-Z), or at 6.46 ppm (10c-E). In 10c-Z, this 8c, 8d and 8e, bearing a hydroxyl-alkyl substituent at the 3-
assignment was conrmed by an observed NOE effect between position, than for the 3-alkyl substituted lactones 8a and 8b.
the two vinylic protons.
Concerning the consequences of bromination of the
The sixteen synthesized furanones, 6a–e, 8a–e, 10a, b, d, e, methylene group, they largely depend on the nature of the
10c-Z and 10c-E, were tested for their ability to interfere with the substituent in position 3 (Table 1). The activity of the hydroxy-
QS dependent bioluminescence production in an Escherichia alkylated lactones is either almost unchanged (compare 8d and
coli biosensor strain containing a plasmid that couples the luxR 10d), slightly enhanced (compare 8c and 10c), or diminished
and luxICDABE promoter regions of Vibrio scheri to the lux- (compare 8e and 10e). Furthermore, the stereochemistry of the
CDABE operon of Photorhabdus luminescens.39 First, these carbon–carbon double bond is of almost no importance
compounds were tested as potential inducers of biolumines- (compare 10c-Z and 10c-E). By contrast, the activity of the
cence, but none of them exhibited any agonistic activity. Next, alkylated lactones 8a and 8b is dramatically enhanced, the
they were tested for their ability to decrease bioluminescence corresponding brominated lactones 10a and 10c, respectively,
induced by 200 nM of 3-oxo-C6-HSL. For comparison, we also being the most active compounds of the series. Finally, the
tested natural dibromofuranone 2 (Scheme 1) prepared by a length of the substituent in position 3 has only a moderate
recently reported synthetic sequence.40
effect on the activity. In the series of hydroxylated furanones,
364 | Med. Chem. Commun., 2013, 4, 363–366
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
the best inhibitory activity was obtained with a 6 carbon atom exomethylene group in the inhibitory activity of the V. scheri
substituent, brominated or not, 10d (IC50 ¼ 6 mM) and 8d QS system for such furanone compounds. Our study enabled us
(IC50 ¼ 8 mM) respectively. The most active compound, and one to identify one of the most active new mbrolide analogues,
of the best inhibitors reported to date among analogs from the namely 3-butyl-5-bromomethylene-2(5H)-furanone (10a), with
mbrolides family, is from the non-hydroxylated series, the an IC50 of 0.6 mM. We have also shown that strong inhibitory
bromo-methylene furanone 10a (IC50 ¼ 0.6 mM) substituted by a activity can be obtained with the non-brominated furanones
butyl alkyl chain. For the most active compound 10a, we have 8d,e, bearing a methylene group, provided that a hydroxyl group
veried that bioluminescence inhibition did not result from is present in the alkyl substituent at position 3 of the furanone.
antimicrobial activity rather than QS inhibition. Compound 10a
Financial support from MESR, CNRS, the “Cluster de
´
ˆ
was tested using the disk diffusion assay, a common assay for recherche chimie/infectiologie de la Region Rhone-Alpes” and
antibiotic susceptibility.41 A control disk diffusion assay was from ANR (project ECORUM) is gratefully acknowledged.
conducted using the relevant kanamycin antibiotic. 10a was
tested at concentrations ranging from 20 mM to 20 mM. Aer
overnight incubation at 37 ꢁC, there is no inhibition of bacterial
growth even at 20 mM, that is, several orders of magnitude
Notes and references
higher than QS inhibition.
1 C. M. Waters and B. L. Bassler, Annu. Rev. Cell Dev. Biol.,
2005, 21, 319–346.
2 S. Uroz, Y. Dessaux and P. Oger, ChemBioChem, 2009, 10,
205–216.
3 J. S. Dickschat, Nat. Prod. Rep., 2010, 27, 343–369.
4 L. C. M. Antunes, R. B. R. Ferreira, M. M. C. Buckner and
B. B. Finlay, Microbiology, 2010, 156, 2271–2282.
5 J. D. Shrout, T. Tolker-Nielsen, M. Givskov and M. R. Parsek,
MRS Bull., 2011, 36, 367–373.
Molecular docking studies, using the LuxR model,42 were
performed with the active compounds 10a, 10d, 8d and with the
inactive furanones 6d and 8a. These investigations suggest that
the interaction between the exomethylene group, either
brominated or non-brominated, and the residues Trp94 and
Leu118 is correlated with the biological activity (see Fig. 1 for
compound 10a). Indeed, the inactive furanone 6d, lacking this
substituent, cannot perform these interactions in the absence
of the methylene group. Docking of the compounds 10d and 8d
with a hydroxyl group indicates other interactions, via hydrogen
bonds, with Tyr70 and Asp79. We suggest that the alcohol
function serves as an anchor for the ligand to interact with
Trp94 and Leu118, whether the exocyclic double bond is
brominated or not. Indeed, the inactive compound 8a, lacking
the alcohol function and the bromine atom, could not interact
with Trp94 and Leu118 via the exocyclic double bond (see ESI†).
6 G. D. Geske, J. C. O'Neill and H. E. Blackwell, Chem. Soc. Rev.,
2008, 37, 1432–1447.
7 C. A. Lowery, N. T. Salzameda, D. Sawada, G. F. Kaufmann
and K. D. Janda, J. Med. Chem., 2010, 53, 7467–7489.
8 G. Brackman, P. Cos, L. Maes, H. J. Nelis and T. Coenye,
Antimicrob. Agents Chemother., 2011, 55, 2655–2661.
9 J. W. Blunt, B. R. Copp, M. H. G. Munro, P. T. Northcote and
M. R. Prinsep, Nat. Prod. Rep., 2011, 28, 196–268.
¨
Finally, the presence of a bromine atom (compound 10a) is 10 R. de Nys, A. D. Wright, G. M. Konig and O. Sticher,
sufficient to obtain interactions with both residues.
Tetrahedron, 1993, 49, 11213–11220.
To summarize, we have developed a new synthetic sequence 11 M. Maneeld, T. B. Rasmussen, M. Henzter, J. B. Andersen,
for the preparation of 3-substituted furanones from the m-
brolides family, either non-substituted at position 5 (6) or
P. Steinberg, S. Kjelleberg and M. Givskov, Microbiology,
2002, 148, 1119–1127.
equipped, at this position, with a methylene group or a bro- 12 C. Kim, J. Kim, H. J. Park, H. J. Park, J. Lee, C. Kim and
momethylene group. Having these compounds available made J. Yoon, Appl. Microbiol. Biotechnol., 2008, 80, 37–47.
it possible to obtain crucial evidence of the importance of the 13 J. C. A. Janssens, H. Steenackers, S. Robijns, E. Gellens,
J. Levin, H. Zhao, K. Hermans, D. De Coster,
T. L. Verhoeven, K. Marchal, J. Vanderleyden, D. E. De Vos
and S. C. J. De Keersmaecker, Appl. Environ. Microbiol.,
2008, 74, 6639–6648.
14 H.-B. Liu, J.-H. Lee, J. S. Kim and S. Park, Biotechnol. Bioeng.,
2010, 106, 119–126.
15 T. Defoirdt, R. Crab, T. K. Wood, P. Sorgeloos, W. Verstraete
and P. Bossier, Appl. Environ. Microbiol., 2006, 72, 6419–
6423.
¨
16 J. Lonn-Stensrud, M. A. Landin, T. Benneche, F. C. Petersen
and A. A. Scheie, J. Antimicrob. Chemother., 2009, 63, 309–
316.
17 K. Glinel, P. Thebault, V. Humblot, C. M. Pradier and
T. Jouenne, Acta Biomater., 2012, 8, 1670–1684.
¨
18 L. K. Vestby, J. Lonn-Stensrud, T. Møretrø, S. Langsrud,
A. Aamdal-Scheie, T. Benneche and L. L. Nesse, J. Appl.
Microbiol., 2010, 108, 771–778.
Fig. 1 Proposed binding mode of furanone 10a (in green) in a model of the
LuxR pocket (in blue).
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun., 2013, 4, 363–366 | 365

View Article Online
MedChemComm
Concise Article
19 L. D. Christensen, M. van Gennip, T. H. Jakobsen, M. Alhede, 30 C. M. Beechan and J. J. Sims, Tetrahedron Lett., 1979, 20,
H. P. Hougen, N. Høiby, T. Bjarnsholt and M. Givskov, J.
Antimicrob. Chemother., 2012, 67, 1198–1206.
1649–1652.
31 H. Kotsuki, M. Monden and M. Ochi, Chem. Lett., 1983, 7,
20 N. Ni, M. Li, J. Wang and B. Wang, Med. Res. Rev., 2009, 29,
65–124.
1007–1008.
32 D. Caine and V. C. Ukachukwu, J. Org. Chem., 1985, 50, 2195–
21 W. R. Galloway, J. T. Hodgkinson, S. D. Bowden, M. Welch
and D. R. Spring, Chem. Rev., 2011, 111, 28–67.
22 M. Maneeld, R. de Nys, N. Kumar, R. Read, M. Givskov,
2198.
33 C. W. Jefford, D. Jaggi and J. Boukouvalas, Tetrahedron Lett.,
1989, 30, 1237–1240.
P. Steinberg and S. Kjelleberg, Microbiology, 1999, 145(Pt 34 P. de March, J. Font, A. Gracia and Z. Qingying, J. Org. Chem.,
2), 283–291. 1995, 60, 1814–1822.
23 Y. Han, S. Hou, K. A. Simon, D. Ren and Y. Y. Luk, Bioorg. 35 A. J. Manny, S. Kjelleberg, N. Kumar, R. de Nys, R. W. Read
Med. Chem. Lett., 2008, 18, 1006–1010. and P. Steinberg, Tetrahedron, 1997, 53, 15813–15826.
24 H. P. Steenackers, J. Levin, J. C. Janssens, A. D. Weerdt, 36 K. P. Haval and N. P. Argade, Synthesis, 2007, 2198–
J. Balzarini, J. Vanderleyden, D. E. De Vos and S. C. De
Keersmaecker, Bioorg. Med. Chem., 2010, 18, 5224–5233.
25 M. Sabbah, F. Fontaine, L. Grand, M. Boukraa, M. L. Efrit,
2202.
37 G. Iskander, R. Zhang, D. S. H. Chan, D. S. Black, M. Alamgir
and N. Kumar, Tetrahedron Lett., 2009, 50, 4613–4615.
`
A. Doutheau, L. Soulere and Y. Queneau, Bioorg. Med. 38 S. V. Ley, M. L. Trudell and D. J. Wadsworth, Tetrahedron,
Chem., 2012, 20, 4727–4736. 1991, 47, 8285–8296.
26 M. Boukraa, M. Sabbah, L. Soulere, M. L. El Efrit, Y. Queneau 39 S. Reverchon, B. Chantegrel, C. Deshayes, A. Doutheau and
`
and A. Doutheau, Bioorg. Med. Chem. Lett., 2011, 21, 6876–6879.
N. Cotte-Pattat, Bioorg. Med. Chem. Lett., 2002, 12, 1153–
1157.
`
27 M. Sabbah, L. Soulere, S. Reverchon, Y. Queneau and
A. Doutheau, Bioorg. Med. Chem., 2011, 19, 4868–4875.
28 A. M. Stevens, Y. Queneau, L. Soulere, S. von Bodman and
A. Doutheau, Chem. Rev., 2011, 111, 4–27.
40 C. A. Lowery, T. Abe, J. Park, L. M. Eubanks, D. Sawada,
G. F. Kaufmann and K. D. Janda, J. Am. Chem. Soc., 2009,
131, 15584–15585.
29 J. Estephane, J. Dauvergne, L. Soulere, S. Reverchon, 41 G. Kronvall and S. Ringertz, APMIS, 1991, 99, 295–306.
`
Y. Queneau and A. Doutheau, Bioorg. Med. Chem. Lett., 42 L. Soulere, M. Frezza, Y. Queneau and A. Doutheau, J. Mol.
2008, 18, 4321–4324.
Graph. Model., 2007, 26, 581–590.
366 | Med. Chem. Commun., 2013, 4, 363–366
This journal is ª The Royal Society of Chemistry 2013