Dihydropyrrolones as bacterial quorum sensing inhibitors

Article abstract of DOI:10.1016/j.bmcl.2019.03.004

Bacteria regulate their pathogenicity and biofilm formation through quorum sensing (QS), which is an intercellular communication system mediated by the binding of signaling molecules to QS receptors such as LasR. In this study, a range of dihydropyrrolone (DHP) analogues were synthesized via the lactone-lactam conversion of lactone intermediates. The synthesized compounds were tested for their ability to inhibit QS, biofilm formation and bacterial growth of Pseudomonas aeruginosa. The compounds were also docked into a LasR crystal structure to rationalize the observed structure-activity relationships. The most active compound identified in this study was compound 9i, which showed 63.1% QS inhibition of at 31.25 μM and 60% biofilm reduction at 250 μM with only moderate toxicity towards bacterial cell growth.

Full text of DOI:10.1016/j.bmcl.2019.03.004

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Dihydropyrrolones as bacterial quorum sensing inhibitors
Basmah Almohaywia,b, Tsz Tin Yua, George Iskandera, Daniel S.H. Chana, Kitty K.K. Hoa,
⁎
Scott Ricec, David StC. Blacka, Renate Griffithd, Naresh Kumara,
b School of Pharmacy, King Khalid University Abha 62529, Saudi Arabia
a School of Chemistry, UNSW Sydney, Sydney, NSW 2052, Australia
c The Singapore Centre of Environmental Life Sciences Engineering, Nanyang Technological University, Singapore 639798, Singapore
d Department of Pharmacology, School of Medical Sciences, UNSW Sydney, Sydney, NSW 2052, Australia
A R T I C L E I N F O
A B S T R A C T
Keywords:
Bacteria regulate their pathogenicity and biofilm formation through quorum sensing (QS), which is an inter-
cellular communication system mediated by the binding of signaling molecules to QS receptors such as LasR. In
this study, a range of dihydropyrrolone (DHP) analogues were synthesized via the lactone-lactam conversion of
lactone intermediates. The synthesized compounds were tested for their ability to inhibit QS, biofilm formation
and bacterial growth of Pseudomonas aeruginosa. The compounds were also docked into a LasR crystal structure
to rationalize the observed structure-activity relationships. The most active compound identified in this study
was compound 9i, which showed 63.1% QS inhibition of at 31.25 µM and 60% biofilm reduction at 250 µM with
only moderate toxicity towards bacterial cell growth.
Quorum sensing
Biofilm inhibition
Dihydropyrrolone
P. aeruginosa
Bacteria regulate their population density using quorum sensing
(QS). QS is a communication process that involves the production of
diffusible chemical signaling molecules, known as autoinducers.1 At
low bacterial density, the autoinducers are produced in small amounts
that are not sufficient to provoke effects. However, as the bacterial
population density and number of signaling molecules increase, a
‘quorum’ is reached. When the binding of signaling molecules to their
cognate receptors exceeds a certain threshold, this triggers the expres-
sion of different bacterial phenotypes including virulence factor pro-
duction and biofilm formation.2 Thus, QS antagonists have potential to
be developed as novel alternative antibacterial agents or as com-
plementary agents to conventional antibiotic drugs. Importantly, the
inhibition of QS does not lead to bactericidal mechanisms, and is
therefore unlikely to exert selective pressure on bacteria that could lead
to resistance.3
bear a nitrogen atom in place of the oxygen in the heterocyclic ring, are
more hydrolytically stable under physiological conditions compared to
the corresponding lactones.5 DHP is a common moiety in several classes
of biologically active molecules such as pulchellalactam (2a, Fig. 1),
jatropham and rolipram.5–7 In our earlier investigations, DHPs were
synthesized as a head group replacement for the lactone motif of fim-
brolides.8 A number of DHP compounds possessed QS inhibitory (QSI)
activities against Pseudomonas aeruginosa with little or no bactericidal
activity.8 The attachment of several DHPs on surfaces as a means of
reducing biofilm formation on biomedical devices was also ex-
plored.9,10 Furthermore, our group has previously reported the synth-
esis of different derivatives of N-substituted DHPs (e.g. 2b, Fig. 1)11
which have shown good QSI activities against Escherichia coli11 and P.
aeruginosa.12
To elaborate on our earlier findings,8–10 we report in this work
further modifications to DHPs focusing on five features around the DHP
motif: a) varying the aromatic substituents at C4; b) replacement of the
C3 hydrogen by a methyl group (R1 = CH3); c) modification of the C5-
bromomethylene group (R2 = Br); d) installing aryl substituents on the
nitrogen atom (R3) and e) generating DHP analogues without the exo-
cyclic alkene (Fig. 2). The synthesized DHP analogues, as a class of non-
lactone AHL mimics, were subsequently tested for QS and biofilm in-
hibitory activities against P. aeruginosa, an opportunistic human pa-
thogen with multidrug resistant strains.
Different species of bacteria use different classes of autoinducers to
regulate QS. Gram-negative bacteria typically produce N-acyl-L-homo-
serine lactones (AHLs).3 Antagonists that mimic the structure of natural
AHLs can disrupt the interaction between AHLs and their cognate re-
ceptor, ultimately inhibiting QS-mediated phenotypes such as virulence
factor production and biofilm formation. Halogenated lactones and
fimbrolides (e.g. 1a–b, Fig. 1) exhibit antimicrobial activities against
several bacterial species by reducing the expression of pathogenic
phenotypes.4 Related dihydropyrrol-2-one (DHP) analogues, which
⁎ Corresponding author.
E-mail address: n.kumar@unsw.edu.au (N. Kumar).
https://doi.org/10.1016/j.bmcl.2019.03.004
Received 15 January 2019; Received in revised form 28 February 2019; Accepted 4 March 2019
0960-894X/©2019ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:BasmahAlmohaywi,etal.,Bioorganic&MedicinalChemistryLetters,https://doi.org/10.1016/j.bmcl.2019.03.004

B. Almohaywi, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 1. Biologically active lactones and pyrrole-2-ones.
Scheme 2. Synthesis of 5-bromomethylene NH-DHPs. Reagents and conditions:
i) NBS, benzoyl peroxide, hν, CCl4; ii) TFA, TFAA, CH2Cl2.
Fig. 2. Modifications to DHP compounds.
The synthesis of the target 5-methylene-4-phenyl-1,5-dihydro-2H-
pyrrol-2-ones employed the lactone-lactam conversion method pre-
viously developed within our group (Scheme 1).8 The synthesis began
with the acid-catalyzed condensation of phenylacetones
3 with
glyoxylic acid or pyruvic acid 4, producing 5-hydroxyfuranones 5a–n
and 6a–e respectively. In the key lactone-lactam conversion step, fur-
anones 5a–n and 6a–e were treated with aqueous ammonia to provide
the intermediate 5-hydroxylactams 7a–n and 8a–e, which were sub-
sequently dehydrated with P2O5 or BF3·Et2O to give the product DHPs
9a–n and 10a–e respectively. The 5-bromomethylene DHPs (12a, 12d
and 12i) were synthesized by radical bromination of hydroxy lactams
7a, 7d and 7i with N-bromosuccinimide (NBS) to give 5-bromo-
methylene-5-hydroxylactams 11a, 11d and 11i, followed by dehydra-
tion to the target compounds (Scheme 2).
The N-aryl DHPs 13–15 (Fig. 3) were synthesized using the Chan-
Evans-Lam CeN arylation reaction using the DHPs 9d and 9i as pre-
cursors, and their synthesis and spectral data have been previously
described.13 However, their biological activities have not been pre-
viously reported. The N-alkyl DHP derivatives 16 and 17 were prepared
following the lactone-lactam conversion method using the lactone
precursor 5d in the presence of excess butylamine and benzylamine
respectively.
Fig. 3. N-aryl DHPs 13a–h, 14a–j and 15a–b, and N-alkyl compounds 16 and
17.
prepared via the Reformatsky reaction of ethyl bromoacetate with the
appropriate acetophenones and ethyl bromoacetate in the presence of
zinc, which were subjected to dehydration followed by NBS bromina-
tion to give the key intermediate 18 as previously described.14–16
Compound 18 was treated with ammonia gas in ether to yield the DHPs
19a–h in 16–60% yields (Scheme 3). The synthesized compounds were
characterized spectroscopically and the spectra are provided in the
In order to examine the importance of the exocyclic double bond on
QS activity, a novel series of DHPs without the double bond at C5
(19a–h) were synthesized. Hydroxybutanoic acid ethyl esters were
Scheme 1. Lactone-lactam conversion procedure.8 Reagents and conditions: i) H3PO4, 75 °C, 5 h, then rt, 10 h; ii) SOCl2, reflux, 2 h; iii) ammonia (aq), CH2Cl2, rt,
24 h; iv) P2O5, CH2Cl2, rt, 30 min or BF3.Et2O, rt, 18 h.
2

B. Almohaywi, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
31.25 µM), the resulting N-aryl DHPs 13a–h showed similar QSI values
(25.3–46.4%) to the parent compound. Furthermore, attaching an alkyl
chain (compound 16) or a phenylalkyl group (compound 17) at the N1-
position of 9d also gave similar QSI values (38.2% and 27.0% respec-
tively) to the parent compound 9d.
The effect of a 3-methyl substitution (R1 in Fig. 2) at the DHP
scaffold was then studied. Compounds 10a–e bearing the 3-methyl
substituent possessed lower QSI activity (23.7–33.7% at 31.25 µM)
compared to their corresponding parent compounds (9b, 9d, 9e, 9i, 9h
respectively) which had QSI values between 33.3% and 63.1%. This
suggested that 3-methyl substitution on the DHP scaffold was un-
favorable for QSI activity.
Scheme 3. Synthesis of compounds 19a–h. Reagents and conditions: i) gaseous
ammonia, ether.
Supplementary Material, together with detailed synthesis procedures
and compound characterization.
To investigate whether the exocyclic alkene of the DHP scaffold is
important for QSI activity, compounds 19a–h without the exocyclic
alkene were synthesized. Removing the exocyclic alkene reduced the
QSI values of compounds 19b–h to 21.9–27.1% compared to their
corresponding parent compounds 9b–o which had QSI values between
33.3% and 63.1%. The only exception was the 4-phenyl-substituted
compound 19a, which showed higher QSI (31.7%) compared to the
parent compound 9a (19.6%). These results suggested that the exo-
cyclic alkene of the DHP scaffold is generally important to the QSI
activity of the compound. With this finding, a terminal bromine group
was added to the exocyclic alkene (R2 in Fig. 2) for compounds 12a,d,i
and its effect on QSI activity was explored. It was found that the effect
of a terminal bromine group substitution at the exocyclic alkene also
depends on the substituent on the C4-phenyl ring of the DHP com-
pound. For compound 12a with an unsubstituted phenyl ring and
compound 12d with a 2-fluoro-substituted phenyl ring, the addition of
a terminal bromine group significantly increased QSI activities (56.8%
and 61.5%, respectively) compared to their parent compounds 9a and
9d, which had QSI values of 19.6% and 37.8% respectively. In contrast,
compound 12i with a 4-bromo-substituted phenyl ring has significantly
lower QSI activity (38.2%) compared to its parent compound 9i, which
had a QSI value of 63.1%.
The library of 53 synthesized compounds was evaluated for QSI
activity using a QS reporter screening system17 following an established
literature protocol.18,19 The reporter strain P. aeruginosa PlasB::gfp
(PAMH602), which produces its own AHL, has a lasB promoter fused to
the green fluorescent protein (GFP-ASV) reporter gene. Therefore, the
level of GFP fluorescence is a measure of AHL-mediated QS in this
strain, and inhibitors of QS will result in a reduction in the expression of
GFP that is correlated with the efficacy of the inhibitor. The bacteria
cultures were incubated with different concentrations of the synthe-
sized compound ranging from 250 to 31.25 µM at 37 °C for 15 h and the
fluorescence of GFP at λ = 535 nm was determined. QSI activity was
expressed as a percentage compared to the negative control (PAMH602
in the absence of test compound). The bacterial optical density (OD) at
600 nm was also measured at the same time to evaluate the potential
growth inhibition of the compounds. Full QSI and growth inhibition
data are shown in Table S1. Furanone 30 (1a)18,20 was used as a po-
sitive control in the assays.
QS inhibition was concentration-dependent for all compounds
(Table S1). As all compounds showed some QSI effect even at the lowest
tested concentration (31.25 µM), the structure-activity relationships
(SARs) of the tested compounds were studied using QSI data at
31.25 µM. At this concentration, the positive control compound 1a
showed 65.9% QSI activity.
In summary, SAR studies of the DHP compounds suggests that a
bromine substituent at the 4-position of the C4-phenyl ring, a hydrogen
at the C3 position of the DHP scaffold, the presence of an exocyclic
alkene group and the absence of a substitution at the N1-position of the
DHP contributed to high QSI activity of the molecule (Fig. 4). Com-
pound 9i containing all of these qualities exhibited the highest QSI
activity of 63.1% at the lowest tested concentration, 31.25 µM.
While the QSI activity of the DHP compounds was being tested,
their bacterial growth inhibition property was also examined. It was
found that at the highest tested concentration (250 µM), only com-
pounds 1a, 12a, 12d, 12i and 15a inhibited bacterial growth by more
than 30% (Table S1). Compounds 9b, 9i, 9j, 9l, 14a, 14g, 14j and 15b
inhibited bacterial growth by between 15 and 30% and the rest of the
tested compounds had small or no effect on growth (≤15%). At
125 µM, only compounds 1a, 12a, 12d and 12i inhibited the growth by
more than 30%. At 62.5 µM, only 1a still inhibited the growth by more
than 30%. The toxicity of the positive control compound, furanone 30
(1a), has been reported previously.20 The most bacteriostatic com-
pounds synthesized in this study (12a, 12d and 12i) are similar to 1a in
that they all contain a bromine substituent on the exocyclic alkene. No
other compounds in this study had this structural feature, suggesting
The parent DHP compound, 5-methylene-4-phenyl-1,5-dihydro-2H-
pyrrol-2-one (9a), showed 19.6% QSI activity at 31.25 µM. The effect of
attaching a substituent at the C4 phenyl ring (R in Fig. 2) of the DHP
scaffold was first compared. DHPs 9b–j, 9l–m bearing halogen, nitro,
methoxy or alkyl groups at different positions of the phenyl ring pos-
sessed higher QSI activities (with inhibition values between 26.3% and
63.1%) than the parent compound 9a. However, adding a carboxylic
substituent at the 4-position of the phenyl ring resulted in DHP 9k,
which had the same inhibition value of 19.6% as the parent compound
9a. Overall, compound 9i bearing a bromine substituent at the 4-po-
sition of the phenyl ring showed the highest QSI value of 63.1% at
31.25 µM.
Moreover, it was found that the position of the substituent on the
phenyl ring only had small effects on QSI activity for this series of
compounds. For example, as compounds 9b–d bearing a fluoro sub-
stituent at the 4, 3 or 2- positions of the phenyl ring respectively
showed similar QSI values ranging between 33.3% and 39.5%.
Similarly, the chloro-substituted compounds 9e-g showed similar in-
hibition values of 26.3 to 37.8%.
Given that the 4-bromophenyl-substituted DHP 9i showed the
highest QSI activity (63.1% at 31.25 µM), the effect of adding a sub-
stituent at the N1-position (R3 in Fig. 2) of the DHP scaffold was then
studied. Attaching a heterocyclic group such as thiophene (compound
15a) or benzothiophene (compound 15b) reduced the QSI value to
49.9% and 34.1% respectively. When the heterocyclic group was re-
placed by various phenyl groups (compounds 14a–j), the QSI value of
the molecules was further diminished to less than 22.0%. This suggests
that attaching an aromatic group at the N1-position is unfavored for QSI
activity for the lead compound 9i. On the other hand, when aryl groups
were added to the 2-fluorophenyl-substituted DHP 9d (37.8% QSI at
Fig. 4. Summary of SARs for QSI activity.
3

B. Almohaywi, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
that the presence of a bromine substituent on the exocyclic alkene in-
creases growth inhibition against bacteria. Therefore, despite com-
pound 12d’s high QSI activity (61.5% at 31.25 µM), it cannot be ad-
ministrated at high quantity without causing bacterial cell death. In
contrast, the most potent compound 9i shows both high QSI activity
(63.1% at 31.25 µM) as well as low bacterial toxicity, suggesting that it
is a better candidate when the drug has to be administrated at high
dosage.
10d-e) docked in a different orientation with the molecule being in-
serted deeply into the hydrophobic pocket occupying the same space as
the end of the tail of OdDHL. While the lactam group was located far-
ther away from the protein residues, the carbonyl group of the lactam
still H-bond with Arg61 (Fig. 5B).
The docking scores, indicative of the relative affinity of the com-
pounds for the LasR binding pocket, were in the range of 45–76 for all
test compounds. However, the docking scores failed to correlate with
the QSI activities of the compound. For example, compounds 14a–j
with an aromatic group at the N1-position possessed lower QSI activity
than their parent compound 9i, but they showed higher docking scores
of between 64 and 76 compared to 9i, which had a score of 55.27.
In terms of predicted interactions between the analogues and the
receptor, other than the formation of a hydrogen bond between their
carbonyl group and the Arg61 side chain of LasR (Fig. 5), nearly all NH-
DHPs 9, apart from 9g, 9h and 9n, additionally formed an electrostatic
π-anion interaction22 between the lactam ring of the small molecule
with Asp73. For compound 9h, the π-anion interaction was instead
formed between the C4-phenyl ring and Asp73. The most active com-
pound (9i) in this study was predicted to form a number of hydrophobic
and π-π interactions with LasR, including three hydrophobic interac-
tions between the Br group of 9i with the amino acid residues Trp88,
Ala105 and Tyr93 of LasR. All 3-methyl DHPs (10a–e) except 10c
formed a hydrogen bond with Arg61. Additionally, all compounds in
this series apart from 10b formed an electrostatic interaction with
Asp73.
In order to understand how the DHP compounds interact with the
LasR QS receptor, docking studies were performed. All of the test
compounds were docked into the crystal structure of LasR with the
autoinducer N-(3-oxododecanoyl)-L-homoserine (OdDHL) (PDB code:
2uv0)21 using the GOLD algorithm implemented in Discovery Studio.
Complete docking methods and results are presented in Table S2.
From the docking analysis, the lactone ring of the autoinducer
OdDHL interacts with the lactone pocket and the aliphatic tail buries in
the hydrophobic pocket. For the synthesized compounds, it was found
that all compounds were predicted to form hydrogen bond interactions
between their carbonyl group with the Arg61 side chain of LasR, while
some of the compounds formed additional hydrogen bond interaction
with Arg61 and Trp60. Some of the DHP compounds (9a–f, 9h–j, 9l–n,
10c, 12a, 12d, 15b, 19a–h) docked in a ‘flipped orientation’ relative to
OdDHL. While the lactone ring of OdDHL interacts with the lactone
pocket and the aliphatic tail buries in the hydrophobic pocket, some of
the DHPs docked in a reverse orientation with the hydrophobic phenyl
group interacting with the protein residues and the lactone group sit-
ting in the hydrophobic region of the LasR protein (Fig. 5A). This
‘flipped orientation’ has also been observed previously in the docked
structures of similar DHP derivatives.12 While this flipped orientation
was observed for some of the DHPs, others (9g, 9k, 9n, 12i, 10a-b,
While all three 5-bromomethylene compounds formed a hydrogen
bond with Arg61, only 12d was able to form an additional hydrogen
bond interaction with Trp60 via its 2-fluorine group. Meanwhile, the
bromine in the 5-bromomethylene atom in 12i formed a halogen in-
teraction with Leu125 and hydrophobic interactions with Ala50 and
Leu40.
For the N-substituted compounds 13–15 (except for 15b), the C4
phenyl group no longer overlapped with the lactone head in OdDHL.
The carbonyl group in the DHP ring of 15a formed hydrogen bonds
with both Arg61 and Trp60, while the thiophene ring formed hydro-
phobic and π-π interactions with Leu110 and Tyr56. Compared with
OdDHL, the thiophene of 15a was in the ‘flipped’ conformation and in
close proximity to the lactone head of OdDHL, while the 4-bromo-
phenyl substituent of 15a occupied the same space as the hydrophobic
tail of OdDHL (Fig. 6).
Our docking results suggested that it is important for the compound
to be able to occupy the same space as the OdDHL autoinducer.
Furthermore, a hydrogen bond with Arg61 and several hydrophobic, π-
π or halogen bond interactions with the hydrophobic part of the re-
ceptor pocket are required. On the other hand, the lactam ring of the
DHP compounds is not required to overlap with the lactone ring of
OdDHL and the head-tail arrangement can be flipped relative to
OdDHL. It was also found that there is no correlation between the
docking scores and the QSI activities of the molecule.
Biofilm formation is one of the key phenotypes that protect bacteria
from stresses such as the host immune system and antibiotics. QS in-
hibitors have been shown to stop the formation of biofilms.19 We tested
the ability of our compounds to prevent biofilm formation against P.
aeruginosa PA01 following an established literature protocol.19 Briefly,
P. aeruginosa cultures were grown in the presence or absence of test
compounds at 250 μM for 6 h, and biofilm biomass was determined by
crystal violet staining (OD550). The detailed method is provided in the
Supplementary Material, and the full results are provided in Table S3.
Selected results are presented in Fig. 7.
The majority of the NH-DHPs 9a–9o had moderate effects on bio-
film formation, with inhibition ranging from 40 to 60% at 250 µM. The
N-substituted thiophene analogue 15a was the most active derivative
showing 87% biofilm inhibition at 250 µM. 15a showed higher activity
than the parent NH molecule 9i (60% biofilm reduction at 250 µM), and
was comparably potency to furanone 30 (1a) at 250 µM. 15a also had
Fig. 5. A) Docked pose of 9i with LasR in comparison with OdDHL. The protein
backbone is shown as a white ribbon. Selected interactions are shown as dashed
lines: π-anion (orange), π-π (pink), hydrogen bonds (green). Atoms are color-
coded: C green for OdDHL, purple for 9i, grey for interacting sidechains; O red;
N blue; H white; Br brown. B) Overlay of OdDHL with docking pose of 9i and 9k
(yellow carbons).
4

B. Almohaywi, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 8. Biofilm inhibition against P. aeruginosa PA01 by compound 15a.
p < 0.05 for concentrations of 125 µM and 62.5 µM as compared to 250 µM.
inhibition activity. Similarly, compound 10a (3-methyl-4-(p-fluor-
ophenyl) DHP) had a very small effect on biofilm formation even
though it inhibited 67% of QS at 250 µM.
Fig. 6. Docked pose of 15a in comparison with OdDHL. Selected interactions
shown as colored dashed lines: π-anion orange, π-π pink. Interacting residue
sidechains shown. Atoms colour-coded: C grey for amino acids, green for
OdDHL, dark blue for 15a; O red; N blue; H white; Br brown; S yellow.
The removal of the exocyclic double bond (compounds 19a–h) had a
deleterious effect on biofilm inhibition. Four of the eight compounds had
no activity at 250 µM, and the remaining compounds inhibited biofilm
formation by between 10 and 21%. This is in contrast to their modest QSI
values which ranged from 58% (19a) to 72% (19h) at 250 µM.
In conclusion, we have designed and synthesized a library of 53
DHP compounds as non-lactone AHL mimics. The compounds were
evaluated as antagonists of LasR signaling in P. aeruginosa. This study
provided preliminary SAR trends around five main features of the DHP
core motif. Attaching a substituent at the C4-phenyl ring can generally
enhance the QSI activity of the DHP, although position of the sub-
stituent on the ring was less important. Introducing a methyl group at
the C3 position was not favorable for QS inhibition. Modification of the
exocyclic alkene by introducing a bromine group improved QSI in two
out of three cases. However, complete removal of the exocyclic double
bond was not favorable for QSI activity. Incorporating substituents at
the nitrogen atom of the DHPs was generally not favorable. The most
potent QS inhibitor, NH-DHP 9i, incorporated all of the favorable SAR
elements. The majority of the DHPs did not inhibit the growth of the
bacteria appreciably. Most of the compounds had moderate effects on
the inhibition of biofilm formation. The most potent biofilm inhibitor
was the N-thiophene compound 15a, which effectively inhibited bio-
film formation at concentrations where bacterial growth was un-
affected. Compound 9i also showed decent biofilm inhibition, and to-
gether with its potent QS inhibitory activity and low bacterial toxicity,
is considered to be the most promising lead scaffold emerging from this
study. In the docking analysis, most molecules bound to LasR in a si-
milar region to the natural autoinducer OdDHL, although the precise
orientation in the binding pocket was dependent on the nature of the
substituents on the DHP scaffold. A hydrogen bond with Arg61 and
several interactions with the hydrophobic part of the OdDHL pocket
appeared to be important for antagonistic activity at LasR. Overall, this
research provides further insights for the design and development of
new generations of QS inhibitors.
Fig. 7. Inhibition of biofilm formation in P. aeruginosa PA01 by selected com-
pounds at 250 µM. Biofilm biomass was quantified using a crystal violet stain
(OD550). % biofilm inhibition was determined with respect to the DMSO
control.
high QSI activity at 250 µM (86%), and was comparably potent to 1a at
this concentration. This suggested that analogue 15a can effectively
inhibit biofilm formation. Compound 15a was tested at two more
concentrations (125 µM and 62.5 µM) and 47% inhibition was observed
at 62.5 µM (Fig. 8). At this concentration, 15a inhibited QS to 60% and
did not inhibit the growth of the bacteria appreciably.
The 2-fluorophenyl N-aryl compounds 13a–h showed moderate
biofilm inhibition at 250 µM, ranging from 38% (13a) to 59% (13b),
compared to 51% inhibition for the parent NH-DHP 9d. These com-
pounds also showed high QSI values of 72% (9d), 64% (13a), and 61%
(13b) at 250 µM, suggesting that their biofilm inhibitory activity could
be related to their ability to disrupt QS in bacteria. The 4-bromophenyl
N-aryl 14a-j series of compounds (derived from NH-DHP 9i) varied in
their ability to inhibit biofilm formation. Compound 14j inhibited
biofilm formation by just 10% at 250 µM, while compound 14h pro-
duced 55% inhibition. Compounds 14j and 14h showed similar QSI
activities (59% and 73% at 250 µM respectively), suggesting that
having decent QSI activity may not always translate to strong biofilm
Acknowledgments
This work was supported by a linkage project grant from the
Australian Research Council (LP150100752). We thank the NMR and
Bioanalytical Mass Spectrometry Facility (BMSF) facilities at UNSW.
The authors would like to acknowledge the Saudi Ministry of Education
and King Khalid University for an Endowment Fund and financial
support given to B. Almohaywi.
5

B. Almohaywi, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Appendix A. Supplementary data
11. Goh W-K, Gardner CR, Sekhar KVC, et al. Synthesis, quorum sensing inhibition and
docking studies of 1, 5-dihydropyrrol-2-ones. Bioorg Med Chem.
2015;23(23):7366–7377.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.03.004.
12. Almohaywi B, Taunk A, Wenholz DS, et al. Design and synthesis of lactams derived
from mucochloric and mucobromic Acids as pseudomonas aeruginosa quorum sensing
inhibitors. Molecules. 2018;23(5):1106.
13. Almohaywi B, Iskander G, Yu TT, Bhadbhade M, Black DS, Kumar N. Copper-
mediated Chan-Evans-Lam N-arylation of 5-methylene-4-aryl-1, 5-dihydro-2H-
pyrrol-2-one derivatives. Tetrahedron Lett. 2018;59(9):811–814.
References
14. Chavan SP, Pathak AB, Pandey A, Kalkote UR. Short and efficient synthesis of ru-
brolide E. Synth Commun. 2007;37(23):4253–4263.
1. Welsh MA, Blackwell HE. Chemical probes of quorum sensing: from compound de-
velopment to biological discovery. FEMS Microbiol Rev. 2016;40(5):774–794.
2. Papenfort K, Bassler BL. Quorum sensing signal-response systems in Gram-negative
bacteria. Nat Rev Microbiol. 2016;14(9):576–588.
15. Fürstner A. Recent advancements in the Reformatsky reaction. Synthesis.
1989;1989(08):571–590.
16. Deng J, Duan Z-C, Huang J-D, et al. Rh-catalyzed asymmetric hydrogenation of γ-
phthalimido-substituted α, β-unsaturated carboxylic acid esters: an efficient en-
antioselective synthesis of β-aryl-γ-amino acids. Org Lett. 2007;9(23):4825–4828.
17. Hentzer M, Riedel K, Rasmussen TB, et al. Inhibition of quorum sensing in
Pseudomonas aeruginosa biofilm bacteria by a halogenated furanone compound.
Microbiology. 2002;148(1):87–102.
3. LaSarre B, Federle MJ. Exploiting quorum sensing to confuse bacterial pathogens.
Microbiol Mol Biol Rev. 2013;77(1):73–111.
4. Manefield M, Rasmussen TB, Henzter M, et al. Halogenated furanones inhibit quorum
sensing through accelerated LuxR turnover. Microbiology. 2002;148(4):1119–1127.
5. Albrecht D, Bach T. Synthesis of 4-substituted 1, 5-dihydropyrrol-2-ones and 5, 6-
dihydro-1H-pyridin-2-ones by Negishi cross-coupling reactions: Short access to the
antidepressant ( )-rolipram. Synlett. 2007;2007(10):1557–1560.
18. Biswas NN, Kutty SK, Barraud N, et al. Indole-based novel small molecules for the
modulation of bacterial signalling pathways. Org Biomol Chem. 2015;13(3):925–937.
19. Kutty SK, Barraud N, Pham A, et al. Design, synthesis, and evaluation of fim-
brolide–nitric oxide donor hybrids as antimicrobial agents. J Med Chem.
2013;56(23):9517–9529.
6. Li W-R, Lin ST, Hsu N-M, Chern M-S. Efficient total synthesis of pulchellalactam, a
CD45 protein tyrosine phosphatase inhibitor. J Org Chem. 2002;67(14):4702–4706.
7. Mase N, Nishi T, Takamori Y, Yoda H, Takabe K. First synthesis of (R)-(−)-5-hy-
droxy-3-methyl-3-pyrrolin-2-one (jatropham) by lipase-catalyzed kinetic resolution.
Tetrahedron Asymmetry. 1999;10(23):4469–4471.
20. Nizalapur S, Kimyon Ö, Biswas NN, et al. Design, synthesis and evaluation of N-aryl-
glyoxamide derivatives as structurally novel bacterial quorum sensing inhibitors. Org
Biomol Chem. 2016;14(2):680–693.
8. Kumar N, Iskander G. Novel lactams. WO2007085042, 2007.
9. Ho KKK, Cole N, Chen R, Willcox MD, Rice SA, Kumar N. Characterisation and in vitro
activities of surface attached dihydropyrrol-2-ones against Gram-negative and Gram-
positive bacteria. Biofouling. 2010;26(8):913–921.
21. Bottomley MJ, Muraglia E, Bazzo R, Carfì A. Molecular insights into quorum sensing
in the human pathogen Pseudomonas aeruginosa from the structure of the virulence
regulator LasR bound to its autoinducer. J Biol Chem. 2007;282(18):13592–13600.
22. Frontera A, Gamez P, Mascal M, Mooibroek TJ, Reedijk J. Putting anion–π interac-
tions into perspective. Angew Chem Int Ed. 2011;50(41):9564–9583.
10. Ho KKK, Cole N, Chen R, Willcox MD, Rice SA, Kumar N. Immobilization of anti-
bacterial dihydropyrrol-2-ones on functional polymer supports to prevent bacterial
infections in vivo. Antimicrob Agents Chemother. 2012;56(2):1138–1141.
6