Design and Synthesis of Lactams Derived from Mucochloric and Mucobromic Acids as Pseudomonas aeruginosa Quorum Sensing Inhibitors

Bacterial infections, particularly hospital-acquired infections caused by Pseudomonas aeruginosa, have become a global threat with a high mortality rate. Gram-negative bacteria including P. aeruginosa employ N-acyl homoserine lactones (AHLs) as chemical signals to regulate the expression of pathogenic phenotypes through a mechanism called quorum sensing (QS). Recently, strategies targeting bacterial behaviour or QS have received great attention due to their ability to disarm rather than kill pathogenic bacteria, which lowers the evolutionary burden on bacteria and the risk of resistance development. In the present study, we report the design and synthesis of N-alkyl- and N-aryl 3,4 dichloro- and 3,4-dibromopyrrole-2-one derivatives through the reductive amination of mucochloric and mucobromic acid with aliphatic and aromatic amines. The quorum sensing inhibition (QSI) activity of the synthesized compounds was determined against a P. aeruginosa MH602 reporter strain. The phenolic compounds exhibited the best activity with 80% and 75% QSI at 250 µM and were comparable in activity to the positive control compound Fu-30. Computational docking studies performed using the LasR receptor protein of P. aeruginosa suggested the importance of hydrogen bonding and hydrophobic interactions for QSI.


Introduction
Exploring new directions to combat bacterial infections has become critically important with the rising incidence of hospital-acquired bacterial infections and the global prevalence of bacterial resistance. Traditional antibiotics are either bactericidal (kill bacteria) or bacteriostatic (inhibit the growth of bacteria) [1]. Therefore, the selective evolutionary pressures exerted by these antibiotics on microorganisms have resulted in the rise and spread of antibiotic resistance [2]. Other factors that have contributed to increased drug resistance include the expanded use of medical devices, treatments for infections in immune-compromised patients and the overuse or mishandling of antibiotics either intentionally or inadvertently [2]. Therefore, novel therapeutic approaches to combat bacterial infection and resistance are required [3].
Bacteria possess an adaptive intracellular mechanism that aids in their communications in a cell density-dependent manner, allowing them to synchronize gene expression as a group using a process termed quorum sensing (QS) [4]. Bacteria sense changes in their population density via the production of diffusible small molecules known as autoinducers, such as the N-acylated homoserine lactones (AHLs) N-(3-oxohexanoyl)-L-homoserine lactone (OHHL) 1 in Vibrio fischeri, and N-butyryl-L-homoserine lactone 2 (C4-HSL, also known as BHL) and N-(3-oxododecanoyl)-L-homoserine lactone 3 (OdDHL, also known as 3-oxo-C12-HSL) in P. aeruginosa ( Figure 1) [5]. P. aeruginosa is an opportunistic and ubiquitous human pathogen and is amongst the most common causative agents of nosocomial and life-threatening infections [6]. In P. aeruginosa, QS is coordinated via a triumvirate of LuxR homologues, namely the LasR, RhlR and QscR systems [7]. These overlapping receptors have significant roles in the regulation of gene expression and QS signals. In P. aeruginosa, QS mediates and controls the gene expressions and phenotypes responsible for its pathogenicity and resistance against the host immune system. It does so by utilizing autoinducers, which trigger the production of virulence factors (e.g., elastase, protease, pyocyanin) and biofilm formation. These phenotypes, however, are not vital to the growth of this pathogen, and thus, their inhibition does not have bacteriostatic or bactericidal effects. Hence, the interference with and antagonism of QS comprise an attractive strategy to overcome and prevent virulence and pathogenicity with minimal likelihood of resistance. Antagonists possessing the lactone head of natural AHLs, but with non-native acyl chains, represent the most extensively-studied class of synthetic QS modulators. For instance, synthetic analogues of the natural autoinducer OdDHL 3 were developed as AHL-based LasR antagonists ( Figure 1) [8]. However, QS antagonists derived directly from AHLs are sensitive to enzymatic and chemical hydrolysis of the lactone ring at physiological pH, giving ring-opened products that lack QS activity [9]. Hence, several research groups have investigated replacement of the lactone group with saturated or unsaturated cyclic and heterocyclic structures [9,10].
Our research group has led the development of both fimbrolide-based analogues [11] and their lactam analogues [12,13]. The lactam fimbrolide analogue 6 exhibited good QS activity and was the most active lactam-based fimbrolide derivative tested against AHL-mediated signaling in Escherichia coli [13]. In line with our continuing efforts to develop new QS inhibitors, we explored the potential use of mucochloric acid 7 and mucobromic acid 8 as precursors that could provide access to functionalized lactams [14]. These compounds are inexpensive, commercially available, highly functionalized and possess multiple sites for reactivity, particularly the two halogen atoms situated across one double bond adjacent to a pseudo acid functionality. Mucohalic acids have been used for the synthesis of furanones with antibacterial and antibiofilm activities [15,16], anticancer activity [17] and anti-inflammatory activity [18]. Mucohalic acids have also been used for the synthesis of furanone-based natural products such as rubrolide [19,20], as well as their lactam analogues, showing herbicidal [20] and antibiofilm activities [15]. They have also been used as precursors of the antiseizure agent levetiracetam [21]. In this work, a library of 34 lactam compounds was prepared using the reductive amination of mucochloric and mucobromic acids with selected aliphatic and Molecules 2018, 23,1106 3 of 29 aromatic amines to furnish N-alkyland N-aryl 3,4 dichloro-and 3,4-dibromopyrrole-2-one derivatives. The QS inhibitory activities of these lactam derivatives were determined.

Synthesis of N-alkyl and N-aryl Lactams
In order to generate a diverse array of lactam analogues, various amines were selected including aliphatic, arylated and heterocyclic amines. Lactams 9-31 were prepared following a literature method using sodium triacetoxyborohydride as a reducing agent in acetic acid and mucohalic acids (mucochloric acid 7 and mucobromic acid 8) (Scheme 1) [14]. Most of the products were purified easily by either trituration from methanol or flash column chromatography (ethyl acetate/hexane) (if required). Compounds with a relatively acidic functional group, such as those derived from carboxyaniline (15)(16) or aminophenol (12)(13)(14), precipitated out from the reaction mixture as pure solids. Products were obtained in reasonable yields. The yields obtained are shown in Scheme 1 and were dependent on both the identity of the starting mucohalic acid and the type of functional group introduced. In general, reactions with mucochloric acid gave higher yields compared to those with mucobromic acid. The proposed mechanism for this reaction [14] depends on nucleophilic attack of the amine onto a protonated carbonyl group. The higher electrophilicity of chlorine compared to bromine facilitates this. When aliphatic groups were introduced, butylamine produced a higher percentage yield (79%) for 9 compared to hexylamine (39%) for 10. In the phenolic Compounds 12-14, the yield was lower when the hydroxyl group was installed on the ortho position compared to the meta and para analogues. The yields for these phenols follow the relative nucleophilicity of the nitrogen. In the ortho derivative, the internal hydrogen bond resulted in the lower nucleophilicity of 12, which required longer reaction times. The introduction of para-carboxyphenyl (16) resulted in a higher yield compared to the meta analogue 15, suggesting the impact of the position of the electron-withdrawing group on the nucleophilicity of the aniline group. However, in the reaction of The QS inhibition assay indicated that the synthesized lactams displayed promising QSI activity. Compounds 9, 12-13, 19, 22-24 and 30 exhibited the highest QSI of 71. .0% at 250 µM. The most active compound, the ortho-hydroxyphenyl mucochloric analogue 12, reduced GFP fluorescence by 83% (±2.8) at 250 µM and was comparable in potency to the positive control 5 and better in activity than the triphenyl antagonist (TP-5). The QSI activity of 12 was well maintained at lower concentrations, with 82.0% (±2.6) and 69.4% (±0.3) inhibitions at 125 µM and 62.5 µM, respectively. There was a slight reduction in activity although not significant (78 ± 2.3% at 250 µM, p > 0.05) as the phenol group moved to the meta position as in derivative 13. The para-phenolic lactam 14 was the least active (58.0 ± 3.3% at 250 µM, p < 0.001) when compared to its ortho (12) and meta (13) counterparts, suggesting the importance of the position of the phenolic hydroxyl group on activity. Interestingly, no significant difference (p > 0.05) in activity was observed when comparing Compound 12 with other potent N-aryl analogues including Compound 19 (80.4 ± 1.1%) and Compound 23 (81.7 ± 0.1%) containing the N-3-aminophenyl and N-phenyl groups, respectively, which both exhibited high inhibition at 250 µM. Furthermore, the activity of Compounds 19 and 23 was also retained at lower concentrations, giving a high percentage of inhibition of 78.1% (±1.6) and 76.5% (±0.4) at 62.5 µM, respectively.
Compound 20 containing a 4-aminophenyl group was less active compared to the meta-amino phenyl analogue 19. These observations from both the N-phenolic and the N-aminophenyl compounds indicate the importance of the position of the electron donating groups installed on the N-phenyl ring of the lactam, with both para-substituted compounds showing reduced activity. Activity was also retained with the mucobromic analogue 24 (80.7 ± 3.0% at 250 µM), suggesting that both of the dichloroand dibromo-pyrrolone were acceptable for activity for this type of scaffold. The N-carboxyphenyl lactams including Compounds 15-16 and Compound 21 (N-butanoic acid group), both derived from mucochloric acid, were less active compared to other N-aryl lactams. The position of the carboxyl group did not improve the activity, and the same was observed with their mucobromic analogues 26 and 27, which were even less active.

QSI of C-Linked and N-Linked Amide Analogues
In general, amides 32-42 displayed only low to moderate activity, as their QSI activity reduced by 30-54% at 250 µM, with low to no activity at 62.5 µM. The C-linked amides 32-38 were generally more active than the N-linked amides 39-42. This indicates the possible importance of the position or the orientation of the amide group. The C-linked amides had higher activity, which ranges from 30-53.5% inhibition at 250 µM when compared to their parent acid lactam 16 (QSI = 19.8% ± 3.3 at 250 µM). Compounds 32 and 34 had a moderate QSI of 52.5% (±6.2) and 54.2% (±5.9), respectively. A comparison of the N-linked amides to their parent 4-aminophenyl lactam 20 (QSI = 58.8% ± 5.8 at 250 µM) showed reduced QSI activity of these amides, which ranges from 53.6-34.5%. The octyl group produced inhibition of 53.6% (±3.1), which was relatively higher than the butyl and hexyl groups.

Evaluation of Growth Inhibition
To ensure that the decrease in GFP fluorescence was related to QS inhibition and not a result of toxicity or a decline in the population of the bacteria, the OD of the cultures were also measured and the degree of growth inhibition noted along with the QS inhibition data in Table 1 (full results are  presented in the Supplementary Information). Overall, our synthesized library of lactams displayed low growth inhibition against P. aeruginosa. At 62.5 µM, all synthesized compounds had little to no effect on the growth of bacteria. Although the synthesized compounds have minimal effect on bacterial growth, it should be noted that there may be other factors that can affect the QS activity [23]. The best compounds in this study and TP-5 reduced bacterial growth by less than 30%, whereas the positive control compound Fu-30 (5) inhibited P. aeruginosa growth by 84% at 250 µM. Therefore, this study showed that the tested compounds inhibit QS with minimal effect on the growth of P. aeruginosa.

Pyocyanin Inhibition
The virulence factor pyocyanin is produced when cell density is high in response to the AHL molecule interacting with LasR. Since LasR is one of the determinant factors for the production of pyocyanin in P. aeruginosa [24], the ability of our compounds to inhibit pyocyanin production was investigated. Wild-type P. aeruginosa (PAO1) were grown in the presence of Compounds 12, 13, 19, 23 and 24, and the amount of pyocyanin in the culture supernatants was quantified based on its absorbance at 695 nm following a reported protocol [25]. The ability of the compounds to reduce pyocyanin levels was determined with respect to the levels of pyocyanin in the DMSO-treated positive control ( Figure 2). Growth inhibition at OD 600 was monitored, and the compounds showed moderate reduction in bacterial growth (22-30%). The compounds were effective in inhibiting pyocyanin in the range of 90-94% at 250 µM. The compounds also showed potent activity at a lower concentration of 32 µM (80-89% inhibition). Our QS-based inhibitors showed comparable pyocyanin inhibition activity to those reported by others [24]. The outcome of this assay correlates well with the potent QS activity of these compounds.

Docking Studies
Lactams 9-42 were docked into the binding site of an X-ray crystal structure of the LasR receptor protein (supplementary materials Table S2) in complex with OdDHL (PDB Code 2UV0) using the Genetic Optimisation for Ligand Docking (GOLD) algorithm through the Accelrys Discovery Studio software package. Before docking, the compounds were minimized using a CHARMM force field. The crystal structure consisted of four subunits (two sets of dimers); however, compounds were only docked into the binding site of subunit E following the results of previous control dockings.
The co-crystallised OdDHL ligand (3) was docked back into the protein giving an acceptable root mean square of deviation (RMSD) of 0.94 Å (heavy atoms) with the X-ray crystal structure.
The ligand-LasR interactions of Compounds 9-42 were analysed based on the highest scoring docked pose of the largest cluster. The predicted binding poses of Compounds 10, 11, 13, 15, 22, 26, and 31-42 were observed to be very similar to OdDHL, with the lactam carbonyl group of most compounds forming a hydrogen bond with Trp60 from the same position as the lactone ring of the natural ligand ( Figure 3A,B). Interestingly, the remaining compounds were predicted to bind in a different pose, with the molecule flipped ( Figure 3C). This alternate pose still allowed the carbonyl group of the lactam ring to form a hydrogen bond with Trp60, as was predicted for Compounds 9, 12, 14 and 25; however, it also allowed the lactam CO to form a hydrogen bond with Arg61, which was observed for the remaining compounds. It was further observed that most of the compounds with high QSI were predicted to bind in a pose different from OdDHL. One explanation for this may be similar to that proposed by Bottomley et al., in that this pose severely limits the interaction with the hydrophobic pocket of the receptor and thus inhibits the formation of a stable protein conformation [26]. Previous studies by Gerdt et al. and Bottomley et al. have identified that hydrogen bonds to residues Ser129 and Thr115 are important for agonistic activity of ligands [26,27]. Notably, none of the QS inhibitors in this study were observed to form these hydrogen bonds, which is consistent with the antagonistic activity seen for these compounds. The predicted binding poses for Compounds 12 and 14 both shared the lactam ring in the same position, resulting in the main difference between the two being the interactions of the hydroxyl group of each compound. In Compound 12, the ortho hydroxyl group was predicted to form a hydrogen bond to Tyr56, whilst in the less active Compound 14, the para hydroxyl group formed a hydrogen bond with Tyr93. Hydrogen bonding to Tyr56 has been identified to be a key interaction in determining the agonistic or antagonistic behaviour of a compound, which may explain the much greater QSI activity observed for 12 [27].
The potent QS inhibitors 19, 23 and 24 form the same hydrogen bond with Arg61 and have a flipped orientation relative to the lactone head, but they form different interactions between the phenyl and the LasR binding pocket. The 4-aminophenyl group in Compound 19 forms an electrostatic π-anion with Asp73. This suggested that NH 2 is not required for activity since Compounds 23 and 24 lack the amino group and yet have potent QS inhibition. Although there are no significant differences in QS activity between 23 and 24, the phenyl groups differ in the type of interactions they produce. Compound 23 makes hydrophobic interactions as shown in Figure 3C,D, while Compound 24 only forms an electrostatic π-anion interaction with Asp73. For the majority of docked compounds, the halogen at the 3-position of the lactam ring was not predicted to make any interactions with the LasR pocket except halogen interaction with Leu110 for Compound 13 and the least active compounds including 15, 33, 35-38 and 40-41. In contrast, the halogen at the 4-position was predicted to form hydrophobic interactions with the pocket for every docked compound including the most active ones. To further investigate the importance of these observations, a new series of compounds similar to those detailed in this study should be synthesized that lack a halogen at the 3-position of the lactam ring and have various hydrophobic alkyl or aromatic groups at the 4-position. Detailed docking results are provided in the Supplementary Information.

Conclusions
A small library of 34 lactam compounds was synthesized and evaluated for QS inhibition against P. aeruginosa. Compounds 9-42 were prepared in moderate to high yields via the reductive amination of mucochloric and mucobromic acid with a wide range of amines, including aliphatic, aromatic and heteroaromatic amines. In biological testing, several compounds possessed promising activities, with 12, 13, 23 and 24 being the most active and showing comparable or even superior activity to the positive controls TP-5 and Fu-30 (5). The tested compounds showed low bacterial growth inhibition in contrast to Fu-30. Amides were also introduced to Compounds 16 and 20 to give the C-linked amides 32-38 and the N-linked amides 39-42, respectively. Generally, the amides 32-42 were less active compared to lactams 9-31, although the C-linked amides 32-38 were more potent than the N-linked amides 39-42. Several compounds showed high efficacy in pyocyanin inhibition, and the results are consistent with their potent QS activity. Docking of the synthesized compounds to the LasR receptor protein predicted favourable intermolecular interactions similar to OdDHL, including a hydrogen bond with the conserved polar Arg61 and Trp60 residues. The most active compounds were docked in a different orientation compared to OdDHL and did not form hydrogen bonds implicated in the stabilization of LasR by agonists. Overall, the results obtained from this study suggest that lactams derived from mucochloric and mucobromic acid could serve as new lead compounds for the development of potent QSI compounds that are unlikely to exert selective pressure on bacteria.

Chemistry
Commercially-available reagents were purchased from Aldrich, Acros Organics, Alfa Aesar. The synthetic procedures have been reported for all compounds as general methods and appropriate references have been given for known compounds. Melting points were measured using a Mel-Temp melting point apparatus and were used uncorrected. High-resolution [+ESI] mass spectra were recorded by the Bioanalytical Mass Spectrometry Facility, UNSW, on an Orbitrap LTQ XL ion trap mass spectrometer using a nanospray (nano-electrospray) ionization source. 1 H and 13 C NMR spectra were determined in the designated solvent on a Bruker DPX 300 spectrometer or a Bruker Avance 400 spectrometer unless otherwise stated. Chemical shifts (δ) are quoted in parts per million (ppm), to the nearest 0.01 ppm and internally referenced relative to the solvent nuclei. 1 H NMR spectral data are reported with their chemical shift in parts per million (ppm). The multiplicity in 1 H NMR is abbreviated as follows: brs, broad; s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; sext, sextet; m, multiplet; or as a combination (e.g., dd, dt, etc.). The coupling constant (J) in hertz, integration and proton count were determined.

General Procedure
Procedure A: Reductive amination Mucohalic acid (1 eq) was added to a solution of 5:3 v/v dichloromethane/glacial acetic acid. Then, an amine (1 eq) was added, and the mixture was stirred for 10 min. To that mixture, sodium triacetoxyborohydride (3 eq) solution in 5:3 v/v dichloromethane and glacial acetic acid was added. The mixture was left to stir at room temperature for 24 h unless otherwise stated.
Procedure B: Amide coupling 1 Acid 16 (1 eq) was treated with thionyl chloride (3 mL) and refluxed for 3 h at 70 • C. The mixture was then cooled and left to stir overnight at room temperature. Thionyl chloride was removed under high vacuum, and the resultant acid chloride was used without further purification. The acid chloride was dissolved in 10 mL of dry tetrahydrofuran, and triethylamine (1 eq) and amine (1 eq) were then added. The reaction mixture was left to stir at room temperature for 24 h.
Procedure C: Amide coupling 2 The amine was dissolved in 10 mL of dry tetrahydrofuran, and then, triethylamine (1 eq) and acid chloride (1.0 eq) were added. The reaction mixture was left to stir at room temperature for 24 h.

Chemistry
Commercially-available reagents were purchased from Aldrich, Acros Organics, Alfa Aesar. The synthetic procedures have been reported for all compounds as general methods and appropriate references have been given for known compounds. Melting points were measured using a Mel-Temp melting point apparatus and were used uncorrected. High-resolution [+ESI] mass spectra were recorded by the Bioanalytical Mass Spectrometry Facility, UNSW, on an Orbitrap LTQ XL ion trap mass spectrometer using a nanospray (nano-electrospray) ionization source. 1 H and 13 C NMR spectra were determined in the designated solvent on a Bruker DPX 300 spectrometer or a Bruker Avance 400 spectrometer unless otherwise stated. Chemical shifts (δ) are quoted in parts per million (ppm), to the nearest 0.01 ppm and internally referenced relative to the solvent nuclei. 1 H NMR spectral data are reported with their chemical shift in parts per million (ppm). The multiplicity in 1 H NMR is abbreviated as follows: brs, broad; s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; sext, sextet; m, multiplet; or as a combination (e.g., dd, dt, etc.). The coupling constant (J) in hertz, integration and proton count were determined.

General Procedure
Procedure A: Reductive amination Mucohalic acid (1 eq) was added to a solution of 5:3 v/v dichloromethane/glacial acetic acid. Then, an amine (1 eq) was added, and the mixture was stirred for 10 min. To that mixture, sodium triacetoxyborohydride (3 eq) solution in 5:3 v/v dichloromethane and glacial acetic acid was added. The mixture was left to stir at room temperature for 24 h unless otherwise stated.
Procedure B: Amide coupling 1 Acid 16 (1 eq) was treated with thionyl chloride (3 mL) and refluxed for 3 h at 70 °C. The mixture was then cooled and left to stir overnight at room temperature. Thionyl chloride was removed under high vacuum, and the resultant acid chloride was used without further purification. The acid chloride was dissolved in 10 mL of dry tetrahydrofuran, and triethylamine (1 eq) and amine (1 eq) were then added. The reaction mixture was left to stir at room temperature for 24 h.
Procedure C: Amide coupling 2 The amine was dissolved in 10 mL of dry tetrahydrofuran, and then, triethylamine (1 eq) and acid chloride (1.0 eq) were added. The reaction mixture was left to stir at room temperature for 24 h.
(1) 1-butyl-3,4-dichloro-1,5-dihydro-2H-pyrrol-2-one (9) The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), butyl amine (170 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated. A residue of oil was obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (390 mg; 79%). 1   The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), butyl amine (170 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated. A residue of oil was obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (390 mg; 79%). 1   (2) 3,4-dichloro-1-hexyl-1,5-dihydro-2H-pyrrol-2-one (10) The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), hexyl amine (300 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated and a residue oil obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (220 mg; 39%); 1 The title compound was made according to Procedure A following the reported method [14].  (12) The title compound was prepared according to Procedure A from mucochloric acid (600 mg, 3.55 mmol), ortho-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 48 h. The precipitated solid was filtered by vacuum filtration, and a white solid was obtained (280 mg; 32%). m.p. 171.6 °C; 1   The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), hexyl amine (300 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated and a residue oil obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (220 mg; 39%); 1 H NMR (CDCl 3 , 400 MHz): δ 0.90 (t, J = 6.9, 13.5 Hz, 3H, CH 3 ), δ 1.24 (m, 6H, CH 2 ), δ 1.60 (m, 2H, CH 2 ), δ 3.49 (t, J = 7.4, 14.6 Hz, CH 2 ), δ 4.04 (s, 2H, C5-CH 2 ). 13 Molecules 2018, 23, x FOR PEER REVIEW 12 of 28 The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), hexyl amine (300 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated and a residue oil obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (220 mg; 39%); 1 The title compound was made according to Procedure A following the reported method [14].  (12) The title compound was prepared according to Procedure A from mucochloric acid (600 mg, 3.55 mmol), ortho-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 48 h. The precipitated solid was filtered by vacuum filtration, and a white solid was obtained (280 mg; 32%). m.p. 171.6 °C; 1 Molecules 2018, 23, x FOR PEER REVIEW 12 of 28 The title compound was prepared according to Procedure A from mucochloric acid (400 mg, 2.3 mmol), hexyl amine (300 mg, 270 mL, 2.3 mmol) and sodium triacetoxyborohydride (1.46 g, 6.9 mmol). The reaction mixture was left to react for 24 h. The crude mixture was extracted with dichloromethane (30 mL) and washed with water (15 mL) followed by brine (15 mL), dried over sodium sulphate, and the solvent was evaporated and a residue oil obtained, which was purified by flash column chromatography using a gradient eluent of 25-50% ethyl acetate in hexane. A yellow oil was obtained (220 mg; 39%); 1 The title compound was made according to Procedure A following the reported method [14].
The title compound was prepared according to Procedure A from mucochloric acid (600 mg, 3.55 mmol), ortho-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 48 h. The precipitated solid was filtered by vacuum filtration, and a white solid was obtained (280 mg; 32%). m.p. 171.6 °C; 1 (13) The title compound was prepared according to Procedure A following the reported method [14] from mucochloric acid (600 mg, 3.55 mmol), meta-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 24 h. As the reaction progresses, a yellow precipitated solid was evident. The solid was filtered by vacuum filtration and a yellow solid was obtained (578 mg; 66%); m.p. 163.6 °C; 1 (14) The title compound was prepared according to Procedure A from mucochloric acid (600 mg, 3.55 mmol), 4-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 24 h. As the reaction progressed, a yellow precipitated solid was evident. The solid was filtered by vacuum filtration, and a yellow solid was obtained (500 mg; 58%); m.p. 126.0 °C; 1 The title compound was synthesized according to Procedure A from mucochloric acid (1 g, 5.91 mmol), 3-aminobenzoic acid (0.81 g, 5.91 mmol) and sodium triacetoxyborohydride (3.76 g, 17.75 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The reaction mixture was left to stir at room temperature for 18 h, during which time, a yellow precipitate was evident. The mixture The title compound was prepared according to Procedure A following the reported method [14] from mucochloric acid (600 mg, 3.55 mmol), meta-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 24 h. As the reaction progresses, a yellow precipitated solid was evident. The solid was filtered by vacuum filtration and a yellow solid was obtained (578 mg; 66%); m.p. 163.6 • C; 1 (14) Molecules 2018, 23, x FOR PEER REVIEW 13 of 28 (13) The title compound was prepared according to Procedure A following the reported method [14] from mucochloric acid (600 mg, 3.55 mmol), meta-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 24 h. As the reaction progresses, a yellow precipitated solid was evident. The solid was filtered by vacuum filtration and a yellow solid was obtained (578 mg; 66%); m.p. 163.6 °C; 1 (14) The title compound was prepared according to Procedure A from mucochloric acid (600 mg, 3.55 mmol), 4-aminophenol (388 mg, 3.55 mmol) and sodium triacetoxyborohydride (2.26 g, 10.65 mmol). The reaction mixture was left to react for 24 h. As the reaction progressed, a yellow precipitated solid was evident. The solid was filtered by vacuum filtration, and a yellow solid was obtained (500 mg; 58%); m.p. 126.0 °C; 1 The title compound was synthesized according to Procedure A from mucochloric acid (1 g, 5.91 mmol), 3-aminobenzoic acid (0.81 g, 5.91 mmol) and sodium triacetoxyborohydride (3.76 g, 17.75 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The reaction mixture was left to The title compound was synthesized according to Procedure A from mucochloric acid (1 g, 5.91 mmol), 3-aminobenzoic acid (0.81 g, 5.91 mmol) and sodium triacetoxyborohydride (3.76 g, 17.75 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The reaction mixture was left to stir at room temperature for 18 h, during which time, a yellow precipitate was evident. The mixture was filtered under vacuum, and the filtered solid was purified by flash chromatography. The solid The title compound was synthesized according to Procedure A from mucochloric acid (1 g, 5.91 mmol), 3-aminobenzoic acid (0.81 g, 5.91 mmol) and sodium triacetoxyborohydride (3.76 g, 17.75 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The reaction mixture was left to stir at room temperature for 18 h, during which time, a yellow precipitate was evident. The mixture was filtered under vacuum, and the filtered solid was purified by flash chromatography. The solid was then recrystallized in methanol after chromatography to yield the pure title product as a white solid The title compound was synthesized according to Procedure A, by first dissolving mucochloric acid (1 g, 5.91 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). To this mixture, a solution of p-aminobenzoic acid (0.81 g, 5.91 mmol) in dichloromethane (8 mL) was added followed by sodium triacetoxyborohydride (3.76 g, 17.75 mmol). The reaction mixture was stirred at room temperature for 18 h, during which time a yellow precipitate was evident. The mixture was filtered under vacuum and washed with dichloromethane and distilled water to yield a yellow solid (0.6 g; 37%). The title compound was prepared according to Procedure A from mucochloric acid (0.9 g, 5.36 mmol), N-Boc-p-phenylenediamine (1.1 g, 5.36 mmol) and sodium triacetoxyborohydride (3.38 g, 15.98 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The mixture was stirred at room temperature for 3 h. The reaction mixture was washed with water and brine and then extracted into ethyl acetate. The organic layer was dried over sodium sulphate and evaporated in vacuo to yield the title compound as a dark red solid (1. The title compound was prepared according to Procedure A using Compound 17. The Boc group was cleaved by treating 17 (1.36 g) with trifluoroacetic acid (6 mL) at room temperature for 1 h followed by evaporation of TFA under high vacuum. The residue was neutralized and washed with saturated solution of sodium bicarbonate, and the solid obtained was filtered under vacuum to yield a red solid (0.75 g, 78%). m.p. 109 °C; 1 H NMR (300 MHz, DMSO-d6) δ 4.75 (s, 2H, CH2), 5.21 (brs, 2H, NH2), δ 6.43 (dd, J = 6.42, 7.7 Hz, 1H, ArH), 6.78 (dd, J = 6.4, 7.7 Hz, 1H, ArH), 7.04-7.12 (m, 2H, ArH); The title compound was prepared according to Procedure A from mucochloric acid (0.9 g, 5.36 mmol), N-Boc-p-phenylenediamine (1.1 g, 5.36 mmol) and sodium triacetoxyborohydride (3.38 g, 15.98 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The mixture was stirred at room temperature for 3 h. The reaction mixture was washed with water and brine and then extracted into ethyl acetate. The organic layer was dried over sodium sulphate and evaporated in vacuo to yield the title compound as a dark red solid (1. The title compound was prepared according to Procedure A from mucochloric acid (0.9 g, 5.36 mmol), N-Boc-p-phenylenediamine (1.1 g, 5.36 mmol) and sodium triacetoxyborohydride (3.38 g, 15.98 mmol) in 5:3 v/v dichloromethane/glacial acetic acid (12 mL). The mixture was stirred at room temperature for 3 h. The reaction mixture was washed with water and brine and then extracted into ethyl acetate. The organic layer was dried over sodium sulphate and evaporated in vacuo to yield the title compound as a dark red solid (1.

Pyocyanin Assay
An overnight culture of P. aeruginosa PAO1 was diluted 1 in 100 with LB10 medium. To a 5-mL test tube were added the tested compounds (compounds prepared from a DMSO stock of 20 mM), and 2.5 mL of the prepared culture was added, giving final concentrations of 250 µM and 32 µM. An equivalent amount of DMSO was prepared for the positive control containing the bacterial culture without the tested compounds. The same medium without bacterial culture in an equivalent amount of DMSO was used as the background reading. The cultures were grown with shaking at 37 • C for 17 h. The final cell density was measured by reading the absorbance at 600 nm (OD 600 ), and the solutions were then centrifuged for 5000 rpm. The clear supernatant was then transferred by pipetting into a plastic 96-well plate, and the absorbance was measured at 695 nm.

Docking
The crystal structure of LasR complexed with OdDHL (PDB: 2UV0) was used. The protein was prepared before docking by removing the other subunits, and only subunit E was used for docking based on a previous control docking. The binding site and cavity were prepared as follows. Water molecules and ligands including the LasR bound ligand in the binding pocket were removed, and the crystal structure was protonated. The binding pocket was chosen from the 'receptor cavities' tool, and this defined the binding site sphere in the proposed pocket. Alternate conformers were identified (Ser20 and Ser131), but were outside the binding pocket sphere. The geometry of the protonated OdDHL ligand was optimized with the CHARMM force field [28], and it was docked back to the prepared protein using the Genetic Optimization for Ligand Docking (GOLD) algorithm, Versions 5.2.1, 5.2.2, and 5.4.0 (Cambridge Crystallographic Data Centre, U.K.) [29]. The number of the docking runs was set to 100, while the 'detect cavity' and 'early determination' were set to 'false', but all other parameters were set to their defaults. The GoldScore was specified as the fitness score function. After the docking run, ligand poses were analysed based on their clusters using the RMSD of heavy atoms and within each cluster; poses were ranked in order of decreasing GoldScore value. The largest clusters at RMSD of around 2 Å were usually considered. The best pose of the largest cluster for OdDHL when docked back to the prepared protein was selected. The ligand from the crystal structure was superimposed with the selected pose, and an acceptable RMSD was obtained of 0.94 Å (heavy atoms).
The tested compounds were sketched, protonated and their energy minimized as above with the CHARMm force field [28]. Docking was then performed and analysed as described above. Receptor ligand interactions of selected poses were analysed using the 'view interactions' tool. Different types of interactions including hydrogen bonds, hydrophobic and π-interactions were examined and compared to the crystal structure. Any other attractive forces, repulsive or unfavourable interactions were noted.