Synthesis of new 3- and 4-substituted analogues of acyl homoserine lactone quorum sensing autoinducers

Article abstract of DOI:10.1016/S0960-894X(01)00756-9

The quorum sensing mechanism in Gram-negative bacteria uses small intercellular signal molecules, N-acyl-homoserine lactones (AHLs), to control transcription of specific genes in relation to population density. In this communication, we describe the parallel synthesis of new AHL analogues, in which substituents have been introduced into the 3- and 4-positions of the lactone ring. These analogues have been screened for their ability to activate and inhibit a Vibrio fischeri LuxI/LuxR-derived quorum sensing reporter system.

Full text of DOI:10.1016/S0960-894X(01)00756-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 325–328
Synthesis of New 3- and 4-Substituted Analogues of Acyl
Homoserine Lactone Quorum Sensing Autoinducersy
J. A. Olsen,a R. Severinsen,a T. B. Rasmussen,b M. Hentzer,b M. Givskovb and
J. Nielsena,*
aDepartment of Chemistry, Bld. 207, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark
bMolecular Microbiology, BioCentrum-DTU, Bld. 301, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark
Received 14 June 2001; revised 12 October 2001; accepted 5 November 2001
Abstract—The quorum sensing mechanism in Gram-negative bacteria uses small intercellular signal molecules, N-acyl-homoserine
lactones (AHLs), to control transcription of specific genes in relation to population density. In this communication, we describe the
parallel synthesis of new AHL analogues, in which substituents have been introduced into the 3- and 4-positions of the lactone ring.
These analogues have been screened for their ability to activate and inhibit a Vibrio fischeri LuxI/LuxR-derived quorum sensing
reporter system. # 2002 Elsevier Science Ltd. All rights reserved.
development of anti-infective drugs.4 A well established
example is Pseudomonas aeruginosa, a prevalent oppor-
Introduction
Many host-associated bacteria use pheromone signal
molecules (autoinducers) to monitor population density
and control expression of specific target genes encoding
virulence factors and other functions.1 This phenom-
enon has been termed quorum sensing (QS). The QS-
system in most Gram-negative bacteria is composed
basically of two components: a LuxI homologue which
synthesizes an acyl homoserine lactone (AHL) auto-
inducer, and an autoinducer-dependent transcriptional
activator, a LuxR homologue.
tunistic human pathogen that causes chronic pulmonary
infections in patients with cystic fibrosis. P. aeruginosa
produces a battery of extracellular virulence factors.
Expression of many of these are controlled by QS reg-
ulation.5 Evidence also suggests that QS is involved in
formation of bacterial biofilms, which renders the sessile
population less susceptible to anti-infectives.6
One strategy to modulate QS is by an autoinducer ana-
logue, which inhibits activation by the natural auto-
inducer. Several reports describe the synthesis and
screening of autoinducer analogues to investigate the
specificity of autoinducer interaction with LuxR homo-
logues and to identify small-molecule inhibitors.7 These
reports mainly focus on the effect of varying the fatty
acid tail of the AHL molecules. Using this approach, a
few inhibitors specific to certain bacterial QS-systems
have been identified.7 We envisioned that modification
of the invariable homoserine lactone part of the auto-
inducer molecule could potentially generate a new, more
general class of QS inhibitors.
LuxI/LuxR from the bioluminescent marine bacteria
Vibrio fischeri was the first AHL-based QS-system to be
described.2 LuxI constitutively synthesizes the auto-
inducer N-(3-oxohexanoyl)-l-homoserine lactone 1 (Fig.
1), which binds to the LuxR protein when a threshold
concentration has been reached. AHL binding activates
LuxR, which initiates transcription of the lux operon
and, in turn, confers the bioluminescent phenotype.
Mutants with defects in their QS-system are sig-
nificantly less virulent3 and the QS-systems have there-
fore been envisaged as a potential novel target for
*Corresponding author. Tel.: +45-4525-2131; fax: +45-4593-3968;
e-mail: jni@kemi.dtu.dk
yThis work was presented in part at the 221th National Meeting of the
American Chemical Society, San Diego, 1–5 April 2001, Organic
Section, Abstract no. 86.
Figure 1. LuxR natural autoinducer 1 and furanone QS inhibitor 2.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00756-9

326
J. A. Olsen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 325–328
Halogenated furanone type QS inhibitors 2 (Fig. 1),8
isolated from the red marine alga Delisea pulchra, have
been shown to inhibit several different QS-systems uti-
lizing different autoinducer molecules. The furanone
type inhibitors have substituents in the 3- and 4-positions
of the furanone ring 2. The structural resemblance
between the furanone inhibitors and AHL suggests that
it would be interesting to explore AHL substituted in
the 3- and 4-positions (Fig. 1).
by substitution with NaI afforded the iodide 3. Similarly
to methods developed for the equivalent Boc derivative,
treatment with Zn afforded the insertion product 3a.10
Palladium catalyzed Negishi cross coupling reaction
with vinyl bromide in presence of 5% Pd2(dba)3 and
P(o-Tolyl)3 afforded protected l-allylglycine 4.
Dihydroxylation by the Upjohn procedure11 was fol-
lowed by spontaneous ring closure to afford a diastero-
meric mixture of the most stable g-lactones 5 and 6. The
mixture was separated by flash chromatography to give
enantiopure lactones 5 and 6.12
Only a few examples exist where the homoserine lactone
part of the AHL molecules has been substituted. To the
best of our knowledge there are no reports on analogues
with substituents in the 3- and 4-positions of the homo-
serine lactone ring. Introduction of substituents should
also provide interesting information of the interaction
between the LuxR homologue and the autoinducer
homoserine lactone moiety.
The cis-(9) and trans-(10) 3-hydroxy substituted homo-
serine lactones13 were synthesized (Scheme 2) from l-
methionine in only six steps by taking advantage of a
modified literature procedure for protected vinylglycine
8.14 Protection of l-methionine followed by oxidation
with metaperiodate gave the sulfoxide 7. Thermal elim-
ination generated protected vinylglycine 8. Dihydroxy-
lation under the conditions as described for the
allylglycine derivative afforded a mixture of 8a, 9 and
10. Full conversion to the g-lactone was accomplished
by heating in the presence of 5% PTSA in toluene. The
diastereomeric mixture was purified by column chro-
matography to afford enantiopure 9 and 10.12
Herein, we describe a new strategy which enables the
introduction of a variety of different substituents into
the 3- and 4-positions of the homoserine lactone ring
in a stereoselective way. These analogues have been
screened for their ability to activate and inhibit a LuxR-
based quorum sensing system.
A new simple one-pot procedure for substituting the
Cbz group with an appropriate fatty acid chain, similar
to the natural autoinducers, was developed.15 Hydro-
genolysis of the Cbz group over Pd/C in MeOH was
accomplished in the presence of hexanoic acid hydroxy-
succinimide ester, which acylates the liberated amino
group (Scheme 3). All four scaffolds (5, 6, 9 and 10)
were subjected to this mild one-pot procedure to give
the corresponding hexanoylated derivatives (5a, 6a, 9a and
10a). Analogues with more bulky substituents derived
from ethylisocyanate were synthesized on a small scale.
The hydroxy group of the hexanoylated lactones (5a, 6a
and 9a) underwent reaction with ethylisocyanate in the
presence of CuCl in THF (Scheme 3) to give the corre-
sponding carbamate lactones (5a, 6a and 9a).16 Lactone
(10a), however, decomposed under these reaction con-
ditions as well as the Cbz-protected lactone 10.17
Chemistry
We envisioned that a range of 3- and 4-substituted
autoinducers could be synthesized in a parallel manner
by using enantiopure cis- and trans-3-hydroxy and
4-hydroxymethyl substituted l-homoserine lactones as
central scaffolds. Acylation, carbamoylation, and alkyl-
ation would then allow the introduction of sustituents
with different size and character.
The cis-(5) and trans-(6) 4-hydroxymethyl substituted
homoserine lactones9 were synthesized (Scheme 1) in
only six linear steps starting from l-serine via a new
synthesis of protected l-allylglycine. First, l-serine was
protected as the appropriate Cbz amino protected
methyl ester. Tosylation of the hydroxy group followed
Scheme 1. Reagents and conditions: (a) AcCl, MeOH, reflux; (b) CbzCl, K2CO3, EtOAc/H2O; (c) TsCl, pyridine; (d) NaI, acetone, reflux; (e) Zn,
.
DMA; (f) 1.0 M vinylbromide in THF, Pd2(dba)3 (5%), P(o-tolyl)3; (g) K2OsO4 2H2O (1%), NMO, acetone; H2O; (h) separation of diastereomers
by flash chromatography.

J. A. Olsen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 325–328
327
Scheme 2. Reagents and conditions: (a) AcCl, MeOH, reflux; (b) CbzCl,ꢀK2CO3, EtOAc/H2O; (c) NaIO4, MeOH/H2O; (d) xylene, reflux;
.
(e) K2OsO4 2H2O (1%), NMO, AcOH, acetone, H2O; (f) PTSA, toluene, 100 C; (h) separation of diastereomers by column chromatography.
Scheme 3. Reagents and conditions: (a) H2 (1atm), 5% Pd/C ( 2.7 mol%), MeOH; (b) CuCl, THF. *not observed; decomposition of starting
material 10a.
Biological Results and Discusion
is harbored by a pUC18 vector maintained in Escheri-
chia coli JM105. The green fluorescent QS reporter
system is activated by exogenous autoinducers. Quan-
tification of the Gfp generated fluorescence gives a
direct measure of the degree of activation. N-Hexanoyl-
l-homoserine lactone 11 was able to activate the QS
reporter system and used as positive control. In corre-
spondence with previous studies,7 this indicates that the
fatty acids’ 3-oxo group is not an absolute requirement
for activity.
The Cbz protected compounds (5, 6, 9 and 10) and the
hexanoylated compounds (5a, 5b, 6a, 6b, 9a, 9b and
10a) were screened for their ability to activate and inhi-
bit a LuxR-based QS reporter system (Fig. 2). The QS
reporter system contains elements of V. fischeri quorum
sensing circuit encoded by luxR-PluxI fused to a reporter
gene encoding an unstable form of the green fluorescent
protein (Gfp).18 The genetic construct, termed pJBA89,
Figure 2. Ability of 3- and 4-substituted homoserine lactones to activate and inhibit a LuxR-based QS screening system. N-Hexanoyl-l-homoserine
lactone 11 was included as positive control in screening for activation. Inhibition was measured by adding the natural autoinducer 1 (concentration 2
nM) together with the compound. A furanone-based inhibitor 2 was included in the experiment as a positive control. After 3 h, green fluorescence
was measured at 535 nm and bacterial density (OD) at 450 nm.

328
J. A. Olsen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 325–328
The Cbz protected compounds (5, 6, 9 and 10) were
neither able to activate nor inhibit the QS-system. This
underlines the importance of the fatty acid chain moiety
of these autoinducer molecules, as also demonstrated in
other studies.7 The 4-substituted analogues (5a, 5b, 6a
and 6b) were only very weak activators, implying that
this part of the AHL molecule is crucial for recognition
of the autoinducers by LuxR. One can speculate that
this part of the molecules is embedded in a narrow
binding pocket of the LuxR protein. In contrast, the
3-substituted analogues (9a, 9b and 10a) were sig-
nificantly more potent as activators. Interestingly, 9a was
able to activate LuxR as efficiently as 11, while inverting
the hydroxy group as in 10a reduced the potency by an
order of magnitude. Derivatization of 9a hydroxy group
as in the case of 9b reduced the potency dramatically.
5. Delden, C. V.; Iglewski, B. H. Emerg. Infect. Dis. 1998, 4, 551 .
6. (a) Davies, D. G.; Parsek, M. R.; Pearson, J. P.; Iglewski,
B. H.; Costerton, J. W.; Greenberg, E. P. Science 1998, 280,
295. (b) Costerton, J. W.; Stewart, P. S.; Greenberg, E. P.
Science 1999, 284, 1318. (c) Lewis, K. Antimicrob. Agents
Chemother. 2001, 45, 999.
7. (a) Eberhard, A.; Widrig, C. A.; McBath, P.; Schineller,
J. B. Arch. Microbiol. 1986, 146, 35. (b) Chhabra, S. R.; Stead,
P.; Bainton, N. J.; Salmond, G. P. C.; Stewart, G. S. A. B.;
Williams, P.; Bycroft, B. W. J. Antibiot. 1993, 46, 441. (c)
Schaefer, A. L.; Hanzelka, B. L.; Eberhard, A.; Greenberg,
E. P. J. Bacteriol. 1996, 178, 2897. (d) Passador, L.; Tucker,
K. D.; Guertin, K. R.; Journet, M. P.; Kende, A. S.; Iglewski,
B. H. J. Bacteriol. 1996, 178, 5995. (e) Zhu, J.; Beaber, J. W. R.;
More, M. I.; Fuqua, C.; Eberhard, A.; Winans, S. C. J. Bac-
teriol. 1998, 180, 5398. (f) Kline, T.; Bowman, J.; Iglewski,
B. H.; Kievit, T. D.; Kakai, Y.; Passador, L. Bioorg. Med.
Chem. Lett. 1999, 9, 3447.
8. (a) Givskov, M.; Nys, R. d.; Manefield, M.; Gram, L.;
Maximilien, R.; Eberl, L.; Molin, S.; Steinberg, P. D.; Kjelle-
berg, S. J. Bacteriol. 1996, 178, 6618. (b) Manefield, M.; Nys,
R. D.; Kumar, N.; Read, R.; Givskov, M.; Steinberg, P.;
Kjelleberg, S. Microbiology 1999, 145, 283.
9. (a) Bernardo, S. D.; Tengi, J. P.; Sasso, G. J.; Weigele, M.
J. Org. Chem. 1985, 50, 3457. (b) Girard, A.; Greck, C.;
Genet, J. P. Tetrahedon Lett. 1998, 39, 4259. (c) Liao, L.-X.;
Zhou, W.-S. Tetrahedron 1998, 54, 12571.
10. Jackson, R. F. W.; Moore, R. J.; Dexter, C. S.; Elliott, J.;
Mowbray, C. E. J. Org. Chem. 1998, 63, 7875.
11. VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron
Lett. 1976, 23, 1973.
Compounds 5a and 6a were able to inhibit QS at max-
imum concentration tested, though less efficiently than
the furanone inhibitor 2. Even though the other syn-
thesized analogues did not show significant antagonistic
activity, our results suggests that further exploration of
3- and 4-substituted analogues could generate a potent
QS inhibitor. This is being pursued at the moment,
together with 3- and 4-substituted analogues having a
3-oxo-dodecanoyl fatty acid side chain for targeting the
therapeutically important QS-system of P. aeruginosa.
22
12. The optical purity corresponded to the literature: 5 ½ꢀꢁD
22
ꢂ2.0 ꢀ (c 0.57, MeOH) [lit.10b ½ꢀꢁD ꢂ1.0 ꢀ (c 0.6, MeOH]. 6
Acknowledgements
22
22
½ꢀꢁD ꢂ48.2 ꢀ (c 0.48, MeOH) [lit.10c ½ꢀꢁD ꢂ46 ꢀ (c 1.3, MeOH)].
Financial support was provided in part by the Technical
University of Denmark, The Danish Technical
Research Council (Grant no. 56-00-0134) and the Vil-
lum Kann Rasmussen Foundation.
22
22
9 ½ꢀꢁD +29.0 ꢀ (c 0.48, EtOAc) [lit.14a ½ꢀꢁD +36.9 ꢀ (c 0.5,
22
22
ꢀ
EtOAc)]. 10 ½ꢀꢁD ꢂ54.2 ꢀ (c 0.98, MeOH) [lit.14b ½ꢀꢁD ꢂ59.2
(c 1.01, MeOH)].
13. (a) Sendai, M.; Hashiguchi, S.; Tomimoto, M.; Kishi-
moto, S.; Matsuo, T.; Ochiai, M. Chem. Pharm. Bull. 1985, 33,
3798. (b) Herranz, R. J. Chem. Soc., Perkin Trans. 1, 649.
14. Afzali-Ardakani, A.; Rapoport, H. J. Org. Chem. 1980,
45, 4817.
References and Notes
15. Dzubeck, V.; Schneider, J. P. Tetrahedron Lett. 2000, 41,
9953.
16. Holten, R. A.; Zhang, Z.; Clarke, P. A.; Nadizadeh, H.;
Procter, D. J. Tetrahedron Lett. 1997, 39, 2883.
17. The ethylisocyanate benzylalcohol adduct (O-benzyl-N-
ethyl-carbamate) was isolated as major by-product for com-
pound 10. The reason has not yet been elucidated.
18. (a) Andersen, J. B.; Sternberg, C. B.; Poulsen, L. K.;
Bjørn, S. P.; Givskov, M.; Molin, S. Appl. Environ. Microbiol.
1998, 64, 2240. (b) Andersen, J. B.; Heydorn, A.; Hentzer, M.;
Eberl, L.; Geisenberger, O.; Christiansen, B. B.; Molin, S.;
Givskov, M. Appl. Environ. Microbiol. 2001, 67, 575.
1. (a) Dunny, G. M.; Winans, S. C. Cell–Cell Signaling in
Bacteria; ASM: Washington, DC, 2000. (b) Parsek, M. R.;
Greenberg, E. P. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8789.
2. (a) Nealson, K. H.; Platt, T.; Hastings, J. W. J. Bacteriol.
1970, 104, 313. (b) Eberhard, A. J. Bacteriol. 1972, 109, 1101.
3. Tang, H. B.; DiMango, E.; Bryan, R.; Gambello, M.;
Iglewski, B. H.; Goldberg, J. B.; Prince, A. Infect. Immun.
1996, 64, 37.
4. (a) Hartman, G.; Wise, R. The Lancet 1998, 351, 848. (b)
Finch, R. G.; Pritchard, D. I.; Bycroft, B. W.; Williams, P.;
Stewart, G. S. A. B. J. Antimicrob. Chemother. 1998, 42, 569.