Synthesis of Novel Quinazolinone Analogues for Quorum Sensing Inhibition

As bacteria continue to develop resistance mechanisms against antimicrobials, an alternative method to tackle this global concern must be developed. As the pqs system is the most well-known and responsible for biofilm and pyocyanin production, quinazolinone inhibitors of the pqs system in P. aeruginosa were developed. Molecular docking following a rationalised medicinal chemistry approach was adopted to design these analogues. An analysis of docking data suggested that compound 6b could bind with the key residues in the ligand binding domain of PqsR in a similar fashion to the known antagonist M64. The modification of cyclic groups at the 3-position of the quinazolinone core, the introduction of a halogen at the aromatic core and the modification of the terminal group with aromatic and aliphatic chains were investigated to guide the synthesis of a library of 16 quinazolinone analogues. All quinazolinone analogues were tested in vitro for pqs inhibition, with the most active compounds 6b and 6e being tested for biofilm and growth inhibition in P. aeruginosa (PAO1). Compound 6b displayed the highest pqs inhibitory activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively) with no bacterial growth inhibition. However, compounds 6b and 6e only inhibited biofilm formation by 10% and 5%, respectively.


Introduction
In order to minimise the development of antibiotic resistance, research has shifted towards new classes of antimicrobial agents that target quorum sensing (QS), rather than cell viability like conventional antibiotics [1][2][3]. QS is a cell-signalling mechanism that allows bacteria to communicate with each other, causing a population of unicellular bacteria to behave similarly to a multicellular organism [2,3]. As the population density of bacteria grows, small molecules known as autoinducers (AIs) that regulate the expression of virulent genes are produced [2,3]. These AIs are produced intracellularly before being released actively or passively into the extracellular environment [3]. Once the concentration threshold of AIs is reached (known as a 'quorum'), the signalling molecules can be recognised by the cognate receptors [3]. This results in the activation of signal transduction pathways that regulate a wide range of activities such as biofilm maturation, virulence factor production, secondary metabolite production and antibiotic resistance [2,3].
QS has become an emerging drug discovery target as it provides an avenue to prevent virulence and biofilm maturation [2][3][4]. As QS inhibitors (QSIs) disrupt bacterial communication through competitive inhibition instead of targeting cell viability, QSIs impose a less selective pressure on bacteria compared to traditional antibiotics, which reduces the chance of bacteria developing resistance against QSIs [4]. When used in conjunction with growth inhibitory antibiotics, QSIs are capable of combatting against biofilm formation and the virulence of bacteria while simultaneously reducing the associated antibiotic resistance [4].
In 2013, the U.S. Center for Disease Control and Prevention stated that Gram-negative bacteria accounted for a large proportion of resistance threats [5]. In particular, the Gramnegative bacterium Pseudomonas aeruginosa has been identified as a multidrug-resistant

Molecular Docking Using GOLD
In silico computational docking experiments were conducted using GOLD software in conjunction with Discovery Studio (DS) to predict intermolecular interactions between the quinazolinone ligands and PqsR receptor ligand binding domain (LBD). These docking studies facilitated a rational approach to SAR analysis and drug design to develop new scaffolds for the synthesis of pqs inhibitors.

Molecular Docking Using GOLD
In silico computational docking experiments were conducted using GOLD software in conjunction with Discovery Studio (DS) to predict intermolecular interactions between the quinazolinone ligands and PqsR receptor ligand binding domain (LBD). These docking studies facilitated a rational approach to SAR analysis and drug design to develop new scaffolds for the synthesis of pqs inhibitors.

Interactions of M64 and PQS with PqsR
M64 is a known pqs inhibitor that has shown great therapeutic efficiency in the treatment of P. aeruginosa infections [12]. A cocrystal of M64 in complex with the PqsR ligand binding domain obtained from the Protein Data Bank (PDB) of Research Collaboratory for Structural Bioinformatics (RCSB) (PDB-6B8A) was used to observe key ligand-receptor interactions in DS [13]. A site-directed mutagenesis of amino acid residues in the PqsR binding pocket have concluded that hydrogen bonding with GLN194 and aromatic stacking interactions with TYR258 strongly contribute towards its affinity and pqs inhibition ( Figure 2A) [13,14]. Molecular docking of M64 also suggests that the sp 3 nature of the central sulfur atom allows for the correct geometry for the compound to fit into the hydrophobic binding pocket. Additionally, the docking of the natural ligand PQS into the LBD ( Figure 2B) revealed that hydrogen bonding between LEU197 and the carbonyl moiety of the quinolone core could be a crucial interaction for its affinity with the receptor. Therefore, possible quinazolinone inhibitors were designed to incorporate these intermolecular interactions to achieve high affinity to the PqsR receptor.

Docking Quinazolinone Analogues with PqsR Receptor
Synthetically possible molecules in a library were docked into the PqsR receptor to compare the ligand-receptor interactions with M64 and PQS. The incorporation of cyclic groups at the R 2 position and amide groups have been shown to increase pqs inhibition in known inhibitors and were thus considered when designing possible analogues [15]. Since the natural ligand PQS has a long alkyl chain extending into the hydrophobic pocket of PqsR, the effect of substituting different-sized alkyl and aromatic amines on predicted interactions were also examined at the R 3 and R 4 positions in the respective scaffolds.
Based on docking results, the quinazolinone amide 5a containing a cyclopropyl group at the R 2 position and a butyl chain attached to the amide was designed from Scaffold 1. This analogue showed potential interactions with key residues GLN194 and TYR258 through hydrogen bonding and hydrophobic interactions, respectively. However, aromatic stacking with TYR258 was not observed between 5a and PqsR. To increase the likelihood of the quinazolinone analogue participating in the key aromatic stacking interaction, the butyl chain was replaced with a 4-methoxyphenyl functional group (5e). The docking results indicated that aromatic stacking interactions with TYR258 were predicted to occur, increasing the compound's ability to bind to the PqsR receptor ( Figure  3A). Further synthesis thus involved a focus on incorporating a 4-methoxyphenyl group

Docking Quinazolinone Analogues with PqsR Receptor
Synthetically possible molecules in a library were docked into the PqsR receptor to compare the ligand-receptor interactions with M64 and PQS. The incorporation of cyclic groups at the R 2 position and amide groups have been shown to increase pqs inhibition in known inhibitors and were thus considered when designing possible analogues [15]. Since the natural ligand PQS has a long alkyl chain extending into the hydrophobic pocket of PqsR, the effect of substituting different-sized alkyl and aromatic amines on predicted interactions were also examined at the R 3 and R 4 positions in the respective scaffolds.
Based on docking results, the quinazolinone amide 5a containing a cyclopropyl group at the R 2 position and a butyl chain attached to the amide was designed from Scaffold 1. This analogue showed potential interactions with key residues GLN194 and TYR258 through hydrogen bonding and hydrophobic interactions, respectively. However, aromatic stacking with TYR258 was not observed between 5a and PqsR. To increase the likelihood of the quinazolinone analogue participating in the key aromatic stacking interaction, the butyl chain was replaced with a 4-methoxyphenyl functional group (5e). The docking results indicated that aromatic stacking interactions with TYR258 were predicted to occur, increasing the compound's ability to bind to the PqsR receptor ( Figure 3A). Further synthesis thus involved a focus on incorporating a 4-methoxyphenyl group at the R 3 position.
Antibiotics 2023, 12, x FOR PEER REVIEW 5 of 22 additional hydrogen bond between the triazole ring and SER196 was predicted. While no interactions with TYR258 were predicted to occur and the position of 6b and M64 do not overlap completely in the receptor ( Figure 3C), they still interact with key residues, implying that 6b should still be antagonistic to pqs signalling.

Synthesis of Analogues
The synthesis of 2-mercaptoquinazolin-4(3H)-one analogues was essential for the preparation of possible pqs inhibitors containing a quinazolinone core. Anthranilic acid 7a, methyl anthranilate 7b and 5-chloro-substituted methyl anthranilate 7c were chosen as the precursors for this reaction series. Anthranilates 7a-c were reacted with an appropriate isothiocyanate and triethylamine in ethanol under reflux conditions to afford the 2mercaptoquinazolin-4(3H)-one intermediates 8a-c in 19-95% yield (Scheme 1). Quinazolinone thiol 8c was obtained in low yields due to the electron-withdrawing nature of chlorine in 7c, which withdraws electrons from the amino group, decreasing its nucleophilicity. The quinazolinone thiols were converted to quinazolinone carboxylic acids using a modified version of the procedure described by Savino et al. (2018) [17]. In these reactions, the thiols 8a-8c were stirred with 2.0 equivalents of bromoacetic acid and potassium To increase the QSI's ability to interact with key polar residues in the PqsR receptor, a secondary scaffold was developed, which includes a 1,2,3-triazole ring in between the quinazolinone core and the amide moiety. Triazole rings were chosen because they are known amide bioisosteres and have been proven to be effective in biofilm and virulence inhibition, as they have been used in previous pqs inhibitors such as 4 [11,16]. The amide was incorporated to increase the compound's likelihood of hydrogen bonding with LEU197, a key amino acid residue with which PQS is able to interact.
Molecular docking of quinazolinone-1,2,3-triazole-phenylacetamide 6b supported the above hypotheses, as the desired interactions were predicted. The amide group was predicted to form hydrogen bonds with GLN194 and LEU207. The carbonyl group on the quinazolinone core was predicted to form a hydrogen bond with THR265, which is an interaction that was predicted to occur with effective pqs inhibitors developed by Grossman  [10]. Additionally, a pi-sulfur bond was predicted between the sulfur atom and PHE221 ( Figure 3B). A previous site-directed mutagenesis study showed that PHE221 played a significant role in pqs inhibition, suggesting that the interaction between 6b and the receptor could be key for pqs inhibition [14]. The importance of this sulfur group for orienting the molecule into a hydrophobic binding pocket is reflected in Figure 3C as it mimics the binding position of the known PQS inhibitor M64. Furthermore, an additional hydrogen bond between the triazole ring and SER196 was predicted. While no interactions with TYR258 were predicted to occur and the position of 6b and M64 do not overlap completely in the receptor ( Figure 3C), they still interact with key residues, implying that 6b should still be antagonistic to pqs signalling.

Synthesis of Analogues
The synthesis of 2-mercaptoquinazolin-4(3H)-one analogues was essential for the preparation of possible pqs inhibitors containing a quinazolinone core. Anthranilic acid 7a, methyl anthranilate 7b and 5-chloro-substituted methyl anthranilate 7c were chosen as the precursors for this reaction series. Anthranilates 7a-c were reacted with an appropriate isothiocyanate and triethylamine in ethanol under reflux conditions to afford the 2-mercaptoquinazolin-4(3H)-one intermediates 8a-c in 19-95% yield (Scheme 1). Quinazolinone thiol 8c was obtained in low yields due to the electron-withdrawing nature of chlorine in 7c, which withdraws electrons from the amino group, decreasing its nucleophilicity. The quinazolinone thiols were converted to quinazolinone carboxylic acids using a modified version of the procedure described by Savino et al. (2018) [17]. In these reactions, the thiols 8a-8c were stirred with 2.0 equivalents of bromoacetic acid and potassium carbonate in ethanol at room temperature for 12 h to generate 9a-d in a 72-99% yield (Scheme 1). Amide functional groups have proven to be critical for pqs, biofilm and pyocyanin inhibition in a variety of known inhibitors [11,18]. These amide bonds are generally synthesised by coupling an amine with a carboxylic acid using stoichiometric amounts of a coupling reagent to ensure that the reaction occurs at ambient temperatures [19]. The synthesis of quinazolinone-based amides was attempted under different reaction conditions, and it was identified that the use of hexafluorophosphate azabenzotriazole tetramethyl uronium (HATU) as the coupling reagent provided the highest yield for the reaction (Scheme S1, Table S1). In such an attempt, 9a was dissolved in anhydrous dimethylformamide (DMF) in the presence of triethylamine and 4-methoxyaniline before HATU was added and the reaction mixture was stirred at room temperature overnight, following a modified procedure reported by Guardia et al. (2016) [20]. After the completion of reaction, water was added to the reaction mixture to precipitate the product and afford pure quinazolinone-based amide 5e in 64% yield. With HATU identified as the optimal acidamine coupling regent, compounds 9a-d were reacted with appropriate amines in analogous reactions to afford quinazolinone-based amides 5a-i (Scheme 2, Table 1). Scheme 1. Synthesis of quinazolinone carboxylic acids 9a-d.
Amide functional groups have proven to be critical for pqs, biofilm and pyocyanin inhibition in a variety of known inhibitors [11,18]. These amide bonds are generally synthesised by coupling an amine with a carboxylic acid using stoichiometric amounts of a coupling reagent to ensure that the reaction occurs at ambient temperatures [19]. The synthesis of quinazolinone-based amides was attempted under different reaction conditions, and it was identified that the use of hexafluorophosphate azabenzotriazole tetramethyl uronium (HATU) as the coupling reagent provided the highest yield for the reaction (Scheme S1, Table S1). In such an attempt, 9a was dissolved in anhydrous dimethylformamide (DMF) in the presence of triethylamine and 4-methoxyaniline before HATU was added and the reaction mixture was stirred at room temperature overnight, following a modified procedure reported by Guardia et al. (2016) [20]. After the completion of reaction, water was added to the reaction mixture to precipitate the product and afford pure quinazolinone-based amide 5e in 64% yield. With HATU identified as the optimal acid-amine coupling regent, compounds 9a-d were reacted with appropriate amines in analogous reactions to afford quinazolinone-based amides 5a-i (Scheme 2, Table 1). mide (DMF) in the presence of triethylamine and 4-methoxyaniline before HATU was added and the reaction mixture was stirred at room temperature overnight, following a modified procedure reported by Guardia et al. (2016) [20]. After the completion of reaction, water was added to the reaction mixture to precipitate the product and afford pure quinazolinone-based amide 5e in 64% yield. With HATU identified as the optimal acidamine coupling regent, compounds 9a-d were reacted with appropriate amines in analogous reactions to afford quinazolinone-based amides 5a-i (Scheme 2, Table 1). Scheme 2. Synthesis of quinazolinone-based amides 5a-i. Given that not all quinazolinone-based amides were able to be successfully synthesised by reacting carboxylic acid 9 with a substituted amine and HATU, an alternate strategy was proposed: this involved reacting 2-chloroacetamide 12 with 8b and 8c to form the corresponding quinazolinone-based amides. 4-Methoxyaniline 10a was first converted to 2-chloroacetamide 12 by nucleophilic addition with chloroacetyl chloride 11 at room temperature for 18 h. 2-Chloroacetamide 12 was then reacted with 8b and 8c in DMF at 50 • C to produce their corresponding quinazolinone-based amides 5j and 5k in 37% and 56% yield, respectively (Scheme 3). Given that not all quinazolinone-based amides were able to be successfully synthesised by reacting carboxylic acid 9 with a substituted amine and HATU, an alternate strategy was proposed: this involved reacting 2-chloroacetamide 12 with 8b and 8c to form the corresponding quinazolinone-based amides. 4-Methoxyaniline 10a was first converted to 2-chloroacetamide 12 by nucleophilic addition with chloroacetyl chloride 11 at room temperature for 18 h. 2-Chloroacetamide 12 was then reacted with 8b and 8c in DMF at 50 °C to produce their corresponding quinazolinone-based amides 5j and 5k in 37% and 56% yield, respectively (Scheme 3). Scheme 3. Synthesis of quinazolinone-based amides 5j and 5k from thiol analogues 8b and 8c, respectively.
To synthesise target Scaffold 2 in the effort to develop possible pqs inhibitors, alkynyl quinazolinones 13a-c were reacted with substituted 2-bromoacetamides 15a-c to synthesise a series of quinazolinone-1,2,3-triazole-phenylacetamide derivatives 6a-e. These alkynyl quinazolinones 13a-c were synthesised by reacting thiol analogues 8a-c with propargyl bromide in the presence of potassium carbonate to initiate a nucleophilic substitution reaction (Scheme 4).  To synthesise target Scaffold 2 in the effort to develop possible pqs inhibitors, alkynyl quinazolinones 13a-c were reacted with substituted 2-bromoacetamides 15a-c to synthesise a series of quinazolinone-1,2,3-triazole-phenylacetamide derivatives 6a-e. These alkynyl quinazolinones 13a-c were synthesised by reacting thiol analogues 8a-c with propargyl bromide in the presence of potassium carbonate to initiate a nucleophilic substitution reaction (Scheme 4).
To synthesise target Scaffold 2 in the effort to develop possible pqs inhibitors, alkynyl quinazolinones 13a-c were reacted with substituted 2-bromoacetamides 15a-c to synthesise a series of quinazolinone-1,2,3-triazole-phenylacetamide derivatives 6a-e. These alkynyl quinazolinones 13a-c were synthesised by reacting thiol analogues 8a-c with propargyl bromide in the presence of potassium carbonate to initiate a nucleophilic substitution reaction (Scheme 4).

Scheme 4. Synthesis of alkynyl quinazolinones 13a-c.
In order to incorporate the triazole moiety, a one-pot synthesis of quinazolinone-1,2,3-triazole-phenylacetamide was carried out by performing click chemistry reactions via copper-catalysed azide-alkyne cycloaddition. The required 2-bromoacetamides 15a-c for the click chemistry reactions were first prepared by reacting aniline derivatives 10a-c with bromoacetic acid 14 (Scheme 5). These 2-bromoacetamides 15a-c were then reacted with 13a-c in the presence of sodium azide, sodium ascorbate and a catalytic amount of CuI in DMF/H2O at 90 °C to afford 6a-e in 10-55% yield (Scheme 5, Table 2). 2-Bromoacetamides 15a-c were used in this reaction instead of the corresponding 2-chloro-substituted acetamides because bromide ion is a better leaving group than chloride ion due to its size and ability to stabilise a negative charge, hence resulting in a more efficient substitution reaction. In order to incorporate the triazole moiety, a one-pot synthesis of quinazolinone-1,2,3-triazole-phenylacetamide was carried out by performing click chemistry reactions via copper-catalysed azide-alkyne cycloaddition.
The required 2-bromoacetamides 15a-c for the click chemistry reactions were first prepared by reacting aniline derivatives 10a-c with bromoacetic acid 14 (Scheme 5). These 2-bromoacetamides 15a-c were then reacted with 13a-c in the presence of sodium azide, sodium ascorbate and a catalytic amount of CuI in DMF/H 2 O at 90 • C to afford 6a-e in 10-55% yield (Scheme 5, Table 2). 2-Bromoacetamides 15a-c were used in this reaction instead of the corresponding 2-chlorosubstituted acetamides because bromide ion is a better leaving group than chloride ion due to its size and ability to stabilise a negative charge, hence resulting in a more efficient substitution reaction.  The regioselectivity of the copper-catalysed azide-alkyne 1,3-dipolar cycloaddition reactions in the formation of products 6a-e was confirmed using 2D 1 H: 13 C HMBC NMR spectroscopy. As a representative example, in the 1 H: 13 C HMBC NMR of compound 6b, the protons at both CH2 groups on C6 (5.26 ppm) and C9 (4.57 ppm) showed a correlation with the aromatic CH carbon on the triazole ring at C3, indicating the successful formation of the 1,3-disubstituted triazole ring (Figures 4 and S1).  The regioselectivity of the copper-catalysed azide-alkyne 1,3-dipolar cycloaddition reactions in the formation of products 6a-e was confirmed using 2D 1 H: 13 C HMBC NMR spectroscopy. As a representative example, in the 1 H: 13 C HMBC NMR of compound 6b, the protons at both CH 2 groups on C6 (5.26 ppm) and C9 (4.57 ppm) showed a correlation with the aromatic CH carbon on the triazole ring at C3, indicating the successful formation of the 1,3-disubstituted triazole ring (Figures 4 and S1).
The regioselectivity of the copper-catalysed azide-alkyne 1,3-dipolar cycloaddition reactions in the formation of products 6a-e was confirmed using 2D 1 H: 13 C HMBC NMR spectroscopy. As a representative example, in the 1 H: 13 C HMBC NMR of compound 6b, the protons at both CH2 groups on C6 (5.26 ppm) and C9 (4.57 ppm) showed a correlation with the aromatic CH carbon on the triazole ring at C3, indicating the successful formation of the 1,3-disubstituted triazole ring (Figures 4 and S1).

PQS Inhibition Assays and Structure-Activity Relationships
The ability of the synthesised quinazolinone analogues in inhibiting the pqs system was evaluated using a pqsA:gfp reporter assay, which measures the PqsR regulated expression of the pqsABCDE operon [21]. Ideal lead compounds as QS inhibitors would effectively inhibit cell communication via pqs with a minimal inhibition of bacterial growth to reduce the exerted selective pressure and decrease the likelihood resistance development.

PQS Inhibition Assays and Structure-Activity Relationships
The ability of the synthesised quinazolinone analogues in inhibiting the pqs system was evaluated using a pqsA:gfp reporter assay, which measures the PqsR regulated expression of the pqsABCDE operon [21]. Ideal lead compounds as QS inhibitors would effectively inhibit cell communication via pqs with a minimal inhibition of bacterial growth to reduce the exerted selective pressure and decrease the likelihood resistance development. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially. Table 3. Percentage PQS inhibition on PqsR system of P. aeruginosa (PAO1-pqsA-gfp) using percentage of green fluorescent protein (GFP) fluorescence at 485 nm. All measurements were performed in triplicates with ±standard deviation from mean. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially. The results of the PQS inhibition assay and growth inhibition assay are shown in Table 3 and Table S2, respectively. By developing 16 analogues across two scaffolds (quinazolinone amides 5a-5k and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e), it was discovered that 6b exhibited the most potent pqs inhibition (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). These results were consistent with the numerous interactions with key amino acid residues (i.e., GLN194, LEU207, THR265, SER196 and PHE221) observed in the molecular docking of 6b ( Figure 3C). Moreover, all the analogues did not inhibit the growth of bacteria substantially.  While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). However, it should be noted that 6e possessed the second highest QSI activity of 63.3% at 100 µM and a 58.3 ± 5.9 a 33.0 ± 5.6 a 10.9 ± 0.3 While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). However, it should be noted that 6e possessed the second highest QSI activity of 63.3% at 100 µM and While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively) While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively) While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels c 63.3 ± 2.0 a 37.6 ± 5.5 a 7.3 ± 1.5 a No bacterial growth inhibition; b growth inhibition between 0 and 10%; c growth inhibition between 10 and 20%; NA = Not active; NT = Not tested. While the difference in activity between quinazolinone-based amides and quinazolinone-1,2,3-triazole-phenylacetamides 6a-e is not substantial, the general trend was that compounds containing the electron-donating 4-methoxyphenyl substituent at the terminal position possessed a slightly higher QSI activity, with 6b displaying the highest levels of QSI activity (73.4%, 72.1% and 53.7% at 100, 50 and 25 µM, respectively). However, it should be noted that 6e possessed the second highest QSI activity of 63.3% at 100 µM and contained an electron-withdrawing nitro group. With the exception of 6b, the minimal difference in activity across the two scaffolds was unexpected as numerous existing compounds containing triazole rings exhibited a greater pqs inhibition compared to quinazolinone-based amides. Nevertheless, these analogues displayed slightly higher levels of pqs inhibition compared to the quinazolinone-based amides 5a-k.
The introduction of chlorine at the 6-position of the quinazolinone core in 5d and 5i had no significant effect on their QSI activity at 100 µM despite the previous literature studies showing a significant increase in pqs inhibition [7,15]. A previous study by Sabir et al. found that quinazolinone analogues containing a cyclopropyl group on the 3-position of the quinazolinone core tended to have higher levels of pqs inhibitory activity than compounds containing a phenyl group [14]. The results for the quinazolinone-based amides 5 a-k were not consistent with these findings for the synthesised quinazolinone-based amides. Compounds 5c, 5h and 5j containing a phenyl group had a slightly higher or similar pqs inhibitory activity at all tested concentrations compared to analogous compounds containing a cyclopropyl group. In contrast, quinazolinone-1,2,3-triazole-phenylacetamides 6d with a phenyl group at the 3-position of the quinazolinone core exhibited significantly lower pqs inhibition compared to the corresponding compound 6b bearing a cyclopropyl group, resulting in a similar trend to that observed for quinazolinone-1,2,3-triazole-phenylacetamides in the study by Sabir et al. Since the cyclopropyl-containing compound 6b possessed the highest activity of all compounds, it was concluded that the installation of the cyclopropyl group significantly contributes to pqs inhibitory activity.
The difference in QSI activity between 5e and 5j was also unexpected as molecular docking shows that 5j was unable to participate in pi-sulfur bonds with PqsR, yet it still possesses a higher QSI activity than 5e at 25 and 50 µM. This could be due to the increased number of hydrophobic interactions between the phenyl substituent and the receptor, which could accumulate to provide a stronger interaction than a pi-sulfur bond. Moreover, it can be seen that 5j with an ethyl linker had an inhibitory activity of 43.0% at 50 µM while 5f with a propyl linker had an inhibitory activity of 11.7% at 50 µM, suggesting that increasing the carbon chain on the amide group decreases pqs inhibition. Initial docking studies showed that quinazolinone analogues containing primary amides were predicted to make a higher number of hydrophobic interactions than secondary amides, potentially increasing their affinity to PqsR and thus pqs inhibition. Indeed, compounds containing the secondary amide from morpholine (5g-i) generally showed a lower pqs inhibitory activity at 100 and 25 µM. However, this trend was less clear at 50 µM, with 5h and 5i bearing a secondary amide moiety having a similar QSI activity as their corresponding compounds 5c and 5d, which bear a primary amide moiety.
Docking studies of quinazolinone-1,2,3-triazole-phenylacetamides 6a,c-e showed that these compounds were not predicted to participate in the key ligand-receptor interactions observed for 6b, especially the hydrogen bonds with SER196 and LEU207, and the pi-sulfur bond with PHE221. As 6b was the most active compound at all tested concentrations, the pqs inhibition assay suggests that interactions with these residues of the receptor may be significant for QSI activity. Additionally, compound 6e with a 4-nitro group at the terminal phenyl ring had a QSI activity of 63.3% at 100 µM and was the second most active compound at that concentration. However, at a lower concentration of 25 µM, its corresponding unsubstituted compounds 6a had a higher QSI activity of 21.6% compared to compound 6e with a QSI activity of 7.3%. These suggested that the introduction of the electron-donating 4-methoxy group at the terminal phenyl ring of quinazolinone-1,2,3-triazole-phenylacetamides had a larger effect on QSI activity compared with compounds containing the electron-withdrawing nitro group.

Biofilm Inhibition
Selected potent compounds 6b and 6e identified in the QS inhibition assay were tested for their ability to inhibit the formation of P. aeruginosa biofilm. In this assay, compounds 6b and 6e at 25, 50 and 100 µM were incubated with P. aeruginosa at 37 • C under static conditions overnight. After incubation, the supernatant was removed, and the loosely bound and planktonic bacterial cells were washed away. P. aeruginosa biofilms, which adhered to the plate substratum, were then quantified by crystal violet staining [22,23].
Both compounds 6b and 6e were poor P. aeruginosa biofilm inhibitors as they only managed to inhibit 10% and 5%, respectively, of P. aeruginosa biofilm formation at 100 µM ( Figure 5). At lower concentrations of 25 and 50 µM, no inhibition of P. aeruginosa biofilm formation was observed. The results were in contradiction with the QS inhibition assay, as it was expected that these compounds would be able to significantly inhibit the formation of P. aeruginosa biofilm via the pqs-based QS system due to their high QSI activity. This discrepancy could be because of the formation of biofilm via other QS systems, and this will be explored in future studies [24].

In Silico Studies
Ligands were initially sketched and protons were added before performing full energy minimisation using Discovery Studio Client 2018 (Accelrys Inc., San Diego, CA, USA) with CHARM forcefield and default setting with max steps set to 10,000. All minimised ligands were docked using GOLD (Cambridge Crystallography Date Centre, Cambridge, UK) through Discovery Studio onto orthorhombic space group c2221, which is the M64 ligand binding domain of PqsR and OdDHL ligand binding domain of LasR. The protocol was run with default settings except for: poses = 100; detect cavity = false, early termination = false, flexibility-intramolecular hydrogen bond = true. The docking pose in the largest, highest scoring cluster between 2-3 Å RMSD of ligand heavy atoms was used for all analyses. Images of docked compounds were generated in Discovery Studio 18.1 and ChemDraw 19.0.

General Information
Bruker Avance III 300 and Bruker Avance III HD 400 spectrometers (Bruker Pty Ltd., Preston, Victoria, Australia) were used to obtain all 1 H and 13 C NMR spectra with the respective solvents using chemical shifts (δ) in parts per million (ppm). Multiplicities for NMR spectra have been assigned using singlet (s), doublet (d), doublet of doublet (dd), doublet of doublet of doublet (ddd), doublet of triplet (dt), triplet (t), quartet (q), doublet

In Silico Studies
Ligands were initially sketched and protons were added before performing full energy minimisation using Discovery Studio Client 2018 (Accelrys Inc., San Diego, CA, USA) with CHARM forcefield and default setting with max steps set to 10,000. All minimised ligands were docked using GOLD (Cambridge Crystallography Date Centre, Cambridge, UK) through Discovery Studio onto orthorhombic space group c222 1 , which is the M64 ligand binding domain of PqsR and OdDHL ligand binding domain of LasR. The protocol was run with default settings except for: poses = 100; detect cavity = false, early termination = false, flexibility-intramolecular hydrogen bond = true. The docking pose in the largest, highest scoring cluster between 2-3 Å RMSD of ligand heavy atoms was used for all analyses. Images of docked compounds were generated in Discovery Studio 18.1 and ChemDraw 19.0.

General Information
Bruker Avance III 300 and Bruker Avance III HD 400 spectrometers (Bruker Pty Ltd., Preston, Victoria, Australia) were used to obtain all 1 H and 13 C NMR spectra with the respective solvents using chemical shifts (δ) in parts per million (ppm). Multiplicities for NMR spectra have been assigned using singlet (s), doublet (d), doublet of doublet (dd), doublet of doublet of doublet (ddd), doublet of triplet (dt), triplet (t), quartet (q), doublet of quartet (dq), pentet (p), hextet (h), septet (sept), multiplet (m) and broad singlet (br) as necessary and coupling constants (J) in Hertz (Hz). Optimelt melting point apparatus (SRS, Sunnyvale, CA, USA) was used for all measurements of all melting points, uncorrected. High-resolution mass spectra (HRMS) were conducted using Thermo LTQ Orbitrap XL instrument (Thermo Scientific, Waltham, MA, USA) under positive-mode electrospray ionisation. Infrared (IR) spectra were recorded using a Cary 630 FTIR spectrometer (Agilent, Mulgrave, Victoria, Australia) fitted with a diamond attenuated total reflectance (ATR) sample interface. Flash column chromatography was performed using Grace Davisil LC60A silica. All reagents were bought commercially from Sigma Aldrich (Castle Hill, NSW, Australia), Alfa Aesar (Haverhill, MA, USA) and ChemImpex (Wood Dale, IL, USA) and used without extra purification. Anhydrous solvents were acquired from PureSolv MD Solvent Purification System (Inert, Amesbury, MA, USA).

Synthetic Procedures and Experimental Characterisation Data
General synthetic procedure A for 2-mercaptoquinazoline 8a-c A mixture of corresponding methyl anthranilate 7 (1.0 equivalent), substituted isothiocyanate (1.2 equivalent) and triethylamine (1.2 equivalent) in ethanol (30 mL) was stirred and heated at 100 • C under reflux for 8 h, using thin layer chromatography (TLC) to monitor the reaction. The reaction mixture was then cooled to room temperature. The resulting white solid precipitate was filtered and washed with diethyl ether. The solid was then dried to afford the corresponding 2-mercaptoquinazoline-4(3H)-one 8.

3-Cyclopropyl-2-mercaptoquinazolin-4(3H)-one (8a)
The title compound 8a was synthesised from methyl anthranilate (1.00 g, 6.61 mmol), 2-cyclopropyl isothiocyanate (0.673 mL, 7.27 mmol) and triethylamine (1.10 mL, 7.90 mmol) following general synthetic procedure A. The product was obtained as a white fluffy solid General synthetic procedure B for quinazolinone carboxylic acid derivatives 9 Appropriate carboxylic acid (1.7 equivalent) in EtOH (5 mL) was added to a solution of compound 8 (1.0 equivalent) in EtOH (5 mL). Potassium hydroxide (10 mL) was then added. The reaction mixture was stirred at room temperature overnight. After completion, water (10 mL) was added to the reaction mixture, which was acidified with HCl to pH 2 to produce a cloudy white suspension. The crude product was then filtered and washed with water before being purified with flash column chromatography on silica gel using n-hexane/EtOAc as eluent to afford a pure white solid product 9.
2-((3-Cyclopropyl-4-oxo-3,4-dihydroquinazolin-2-yl)thio)acetic acid (9a) The title compound 9a was synthesised from thiol 8a (0.500 g, 2.29 mmol) and bromoacetic acid (0.541 g, 3.89 mmol) following general synthetic procedure B. The product was obtained as a fluffy white solid (0.458 g, 64%); mp 165. General synthetic procedure C for quinazolinone-based amides 5a-i A solution of the appropriate acetic acid 9 (1.0 equivalent), substituted amine (1.1 equivalent) and triethylamine (2.0 equivalent) in anhydrous DMF (2 mL) was prepared. HATU (1.1 equivalent) was then added to the solution and the reaction mixture was stirred at room temperature overnight. After completion of the reaction, water was added and the mixture extracted with EtOAc (3 × 15 mL). The combined organic extracts were washed with brine (30 mL), dried over anhydrous magnesium sulfate, and concentrated in vacuo. The crude mixture was purified with flash column chromatography on silica gel using n-hexane/EtOAc as eluent to afford the solid product 5a-i.

2-Chloro-N-(4-methoxyphenyl)acetamide (12)
A stirred solution of 4-methoxyaniline 10a (0.500 g, 4.06 mmol, 1.0 equivalent) in dichloromethane (DCM) (15 mL) was cooled to −10 • C under an argon atmosphere. Triethylamine (0.622 mL, 0.470 mmol, 1.1 equivalent) and chloroacetyl chloride 11a (0.550 g, 4.872 mmol, 1.2 equivalent) were added successively and the reaction mixture was stirred at room temperature overnight. The resulting reaction mixture was treated with 10% HCl (10 mL) and then added to ice. The product was extracted into DCM (3 × 30 mL), and the combined organic extracts washed with brine, dried over anhydrous magnesium sulfate and concentrated in vacuo. The resulting crude compound was purified with flash column chromatography on silica gel using DCM/MeOH as eluent to afford the acetamide 12a as an off-white solid powder (0.626 g, 77%); mp 120.9-123.9 • C; 1 H NMR (400 MHz, DMSO-d 6  General synthetic procedure D for quinazolinone-based amides 5j and 5k A solution containing the appropriate thiol compound 8b or 9c (1.0 equivalent), 2-chloroacetamide 12a (1.0 equivalent) and potassium carbonate (2.0 equivalent) in DMF was heated at 50 • C for 12 h. After completion, the reaction mixture was poured into water/ice and the product was extracted with EtOAc (3 × 15 mL). The combined organic extracts were washed with brine, dried over anhydrous magnesium sulfate and concentrated in vacuo to afford the final solid product (5j or 5k).

PQS Inhibition Assay
The assay for PqsR inhibition activity was performed using the PAO1 P. aeruginosa strain carrying the PqsR-regulated pqsA promoter fused to gfp. The compounds were dissolved in 100% DMSO to make 20 mM stock solutions. The test compounds (serially diluted with medium) were then incubated with overnight cultures of PAO1-pasA-gfp using MHB (Mueller-Hinton Broth) in 96-well plates at 37 • C with intermittent shaking. Readings were taken at 30 min intervals for at least 8 h and both GFP fluorescence and OD 600 were recorded. The fluorescence values shown in the graph were normalised with respect to OD 600 . Negative control refers to the medium containing DMSO (0.5%) as the highest concentration of the test compound. The pqs inhibition assay was carried out in triplicate manner [21].

Biofilm Inhibition Assay
A single colony of P. aeruginosa was cultured in Mueller-Hinton Broth (MHB) at 37 • C with shaking at 120 rpm for 24 h. The resulting bacterial culture was washed twice with MHB with centrifugation after each wash. The bacterial solution was then diluted with fresh MHB to a turbidity of OD 660 = 0.1 in MHB (equivalent to 10 8 colony-forming unit (CFU)/mL of bacteria), followed by diluting to 10 6 CFU/mL in MHB. 100 µL of the bacterial solution was added to wells of a flat-bottom 96-well plate (Costar) containing 100 µL serially diluted test compound. After incubation at 37 • C for 18 h, loosely bound cells were washed away with 1× phosphate-buffered saline (PBS, pH 7.4). Biofilms adhered to the plate substratum were quantified, using crystal violet staining as described previously [22,23]. Untreated bacteria were used as a negative control, where the percentage of biofilm mass reflected 100% biofilm growth. The experiment was performed in triplicate.

Conclusions
In conclusion, GOLD docking studies were carried out in order to guide the synthesis of quinazolinone analogues targeting the PqsR receptor in P. aeruginosa. Modifications of two quinazolinone-based scaffolds led to the synthesis of a library of 16 quinazolinone analogues using three different synthetic pathways. Eleven quinazolinone-based amides 5a-i were generated via acid-amine coupling using quinazolinone carboxylic acids 9a-d with substituted amines. Another two quinazolinone-based amides 5j and 5k were synthesised via a nucleophilic substitution of 2-chloroacetamide 12a using thiol compounds 8a-d. Furthermore, quinazolinone-1,2,3-triazole-phenylacetamides 6a-e were synthesised utilising 1,3-dipolar cycloaddition of alkyl quinazolinones 13a-c and 2-bromoacetamides 12b-d.
In vitro pqs inhibition assays of compounds 6a-e identified that introducing the methoxy electron-donating group on the 4-position of the terminal phenyl ring (6b) could play a significant role in pqs inhibition. The pqs inhibition results of 6a and 6e emphasise this as analogues without a substituent or bearing an electron-withdrawing nitro group, respectively, possessed lower levels of pqs inhibition. This suggests that the higher levels of pqs inhibition could be due to the hydrogen bonding interactions with GLN194, LEU207, THR265 and SER196, pi-sulfur bonds with PHE221 and hydrophobic interactions with TYR258 and LEU197, as shown by in silico molecular docking studies.