Identification of novel phenylalanine derivatives bearing a hydroxamic acid moiety as potent quorum sensing inhibitors

Abstract

Phenylalanine derivatives are a well-known small moiety responsible for controlling the virulence factors of several bacteria. Herein, for the first time, we report novel structures of phenylalanine derivatives bearing a hydroxamic acid moiety which were designed, synthesized, and evaluated for use as quorum sensing inhibitors. Biological results reveal that six compounds showed good quorum sensing inhibitors properties with an IC₅₀ ranging from 7.12 ± 2.11 μM–92.34 ± 2.09 μM (4NPO, a reference compound, IC₅₀ = 29.13 ± 0.88 μM). In addition, three out of the six compounds (4a, 4c, 4h) showed strong anti-biofilm formation and CviR inhibitory activity when compared to that of 4NPO. These biological data were also confirmed by computational studies. In this series of compounds, 4h is the most promising compound for future drug development targeting quorum sensing. Our results concluded that the fragment-based drug design is a good approach for the discovery of novel quorum-sensing inhibitors in the future.

Phenylalanine derivatives bearing a hydroxamic acid moiety as potent quorum sensing inhibitors.

1. Introduction

The abundance of microbial drug-resistance has negatively affected the security of a vast number of countries around the world, especially for underdeveloped and developing countries.^1,2 The unmonitored prescription use and overuse of antimicrobial drugs have accelerated the growth of a new generation of antibiotic-resistant bacteria.^3,4 Therefore, research and development for new antimicrobials with novel mechanisms of action is needed.

Pseudomonas aeruginosa is at the top of the list of WHO “priority pathogens” that might pose a great threat to global health, due to the growing number of resistant antibiotics caused by this bacterium.⁵ Therefore, our group is currently interested in the design, synthesis, and evaluation of novel quorum-sensing inhibitors for use as alternative treatments for infections caused by this bacterium.⁶

Quorum sensing is a communication pathway for bacteria in response to external stress.^6–9 Bacteria use quorum sensing as a biological tool for responding to the presence of nutrition, antimicrobials, and others. The results of the response to the external factors of bacteria include the production of virulence factors, increase of the bacterial mass, forming of biofilms, and host toxicity. As a result, many studies have proved that the inhibition of the quorum sensing pathway may lead to the prevention of bacterial infection and biofilm formation, and reduce the production of toxins by bacteria.^6,9

Design of a compound library

It is well-known that the quorum-sensing protein is a metalloenzyme that contains a zinc ion in the active site.¹⁰ We envision that the compound containing the metal-binding group will potentially act as quorum sensing inhibitors. Carboxylic acid is a weak metal binding group that was also found to be a quorum-sensing inhibitor in our previous work.^6b In addition, hydroxamic acid, a strong metal-binding group, was previously reported to be a quorum-sensing gene inhibitor.^11,12 Therefore, the replacement of carboxylic by the hydroxamic moiety could bring more benefits for the design of a novel quorum-sensing inhibitor. Furthermore, previous studies have revealed that several hydroxamic acids do not disturb the normal growth of bacteria, a key element for quorum-sensing inhibitor agents.¹³ Overall, the hydroxamic moiety could be a great privileged structure for designing novel quorum-sensing inhibitors.

On the other hand, phenylalanine and its derivatives are well-known as small molecules responsible for controlling the virulence factors of several bacteria, in particular, it is one of the key fragments in the structure of quorum sensing molecules cyclo(l-Phe-l-Pro) (Fig. 1).^10,14 In this work, we chose phenylalanine as the main skeleton and designed a novel hybrid structure containing both phenylalanine and hydroxamic acid moieties (Fig. 1).

Fig. 1. Design of library compounds.

2. Materials and methods

2.1. Chemistry

The reagents and solvent were supplied by TCI, and Sigma-Aldrich and were used directly without further purification. The NMR spectra were recorded using a Bruker Avance system at 400/500 MHz at 300 K (for protons), and 151 MHz or 101 MHz for carbon. Chemical shifts (δ) were calibrated to the peak of the solvent in ppm. The J are reported in Hz. Pre-coated silica gel 60 F₂₅₄ (Sigma-Aldrich) was used for the thin layer chromatography (TLC) and visualized under 254 nM (UV). The HRMS was recorded using an electrospray ionization (ESI) mass spectrometer.

2.2. Biology

2.2.1. Quorum sensing assay

In this work, the quorum sensing (QS) assay was adapted from the previously described method.^6g The QS system of P. aeruginosa-lasB-gfp was used for screening. Firstly, the QS system of P. aeruginosa-lasB-gfp was cultured at 37 °C for 20 h, with a shaking speed of 180 rpm. After leaving overnight, it was diluted to give the final OD₄₅₀ of 0.1. All the synthesized compounds, reference compound, were added to Black Isoplate 96-well microtiter dishes (PerkinElmer, Waltham, MA, USA) containing growth media, together with reporter strains N-(3-oxohexanoyl)-l-homoserine lactone (OHHL). The media was kept at 34 °C and the growth was recorded every 15 min over 20 h using a Victor X4 multilabel plate reader (PerkinElmer). A green fluorescent protein (GFP) expression assay (PerkinElmer, Waltham, MA, USA) was used, and the fluorescence was recorded at 485 nm and 535 nm.

2.2.2. Anti-biofilm biomass

The anti-biofilm biomass assay was adapted from the previously described method using P. aeruginosa PA14.^6g In this work, the synthesized compounds were tested at a concentration of 30 μM. The negative control was DMSO, and the positive control was 4NPO. Briefly, P. aeruginosa PA14 culture was diluted in M63 minimal medium, to give a final OD₆₀₀ of 0.02. Then 200 μL of P. aeruginosa PA14 culture was added to the 24-well imaging plate. All the synthesized compounds and reference compounds were added to the 24-well imaging plate and incubated for 48 h at 30 °C. The biofilm biomass was monitored using the crystal violet (CV) staining method.¹⁵ The CV solution (190 μL of 0.01%, Sigma-Aldrich) was added to three wells of the 96-well biofilm microplate, and then incubated at room temperature for 30 min. Then the CV solution was removed, the resulting biomass was washed with sterile water (3 times) and dried at 50 °C. Finally, ethanol (96%) was added dropwise to each well to detach the biofilm. The absorbances were measured at 570 nm. If a negative value for optical density (OD) was obtained, it was presented as zero. The experiment was performed with three replicates.

2.2.3. CviR inhibition of Chromobacterium violaceum

Previously, we have identified the relationship between the quorum sensing inhibitory activity of our designed compound in P. aeruginosa and the inhibition of CviR of C. violaceum.^6g Therefore, we decided to screen the library compounds for the CviR of C. violaceum. The anti-CviR assay was adapted from the previously described method^6g using C. violaceum ATCC 31532. The strain was grown at 30 °C overnight, and then the culture with or without compounds (30 μM) at 30 °C for an additional 24 h. The absorbance of the soluble violacein was read at 585 nm using a microplate reader.

2.3. Docking studies

Compound structures were created using ChemDraw version 9.0 and then optimized with MOE 2015 to minimize energy to a rms gradient of 0.1 kcal⁻¹ mol⁻¹. The Merck molecular force field's MMFF94s variant was selected for the force field's configuration. The X-ray crystallographic structure of LasR, a quorum sensing protein in P. aeruginosa, bound to its autoinducer was sourced from the Protein Data Bank (ID: 2UV0).¹⁶ The structure was prepared using the MOE QuickPrep tool by adding hydrogen, protonating, removing water, modifying atom types, and applying AMBER FF99 charges in accordance with earlier techniques.^17,18 Using the MOE Triangle matcher placement method, docking simulations were run with a flexible ligand and a rigid protein, keeping 30 poses for analysis. For the purpose of determining the free binding energy between the ligands and proteins, the chosen conformations were scored using the London dG functions. All the selected poses were visualized with the BIOVIA Discovery Studio v3.5 program.

3. Results and discussion

3.1. Synthesis of a compound library

The library compounds were prepared from a key intermediate (2S)-2-amino-3-phenyl-N-((tetrahydro-2H-pyran-2-yl)oxy)propanamide (2) (Scheme 1). Compound 2 was prepared with an excellent yield via a two-step reaction of Fmoc-phenylalanine with NH₂OTHP followed by Fmoc-deprotection (removal of Fmoc group). In this scheme, compounds 3a–3i were prepared in situ and then directly converted into target compounds 4a–4i under acidic conditions.

Scheme 1. Synthetic scheme for the synthesis of a library of compounds.

3.1.1. Synthesis of (2S)-2-amino-3-phenyl-N-((tetrahydro-2H-pyran-2-yl)oxy)propanamide (2)

In the solution of Fmoc-l-phenylalanine (1.5 mmol, 1.0 equivalent) in DCM, TEA (0.313 mL, 2.25 mmol, 1.5 equivalent) was added and then cooled down to 0 °C. Then EDC (429 mg, 2.25 mmol, 1.5 equivalent), HOBt (303 mg, 1.95 mmol, 1.5 equivalent) were added to the mixture and stirred for 1 min, followed by NH₂OTHP (210 mg, 1.8 mmol, 1.2 equivalent) and stirred at room temperature for 2 h. After the reaction was completed (by TLC), the mixture was diluted with 5 mL of DCM, then washed with 5 mL of aqueous 5% NaHCO₃ and brine, respectively. The organic layer was dried (Na₂SO₄) and then concentrated under vacuum. The residue was used for the next step without purification.

The above residue was added to 5 mL of RBF. Then 2 mL of 20% piperidine in DMF was added and stirred for 30 min at rt. The crude residue was purified by column chromatography (eluent: CH₂Cl₂/MeOH = 95 : 5) to give 2 with an excellent yield of 91%.

3.1.2. Synthesis of compounds 4a–4i

Preparation of 3a–3i

Commercially available carboxylic acids were used for the preparation of 3a–3c, and 3i. For 3d–3h, the carboxylic acid intermediates were prepared from the corresponding methyl benzoate derivatives (see ESI†). The procedure for the synthesis of 3a–3i from the corresponding carboxylic acids is as follows: 20 mL RBF and carboxylic acid (1 equivalent) was dissolved in 5 mL of DCM, and then EDC·HCl (1 equivalent), HOBt (1 equivalent), and TEA (1.5 equivalent) were added. The mixture was stirred for 10 min at room temperature. Then compound 2 (1 equivalent, pre-dissolved in DCM) was added dropwise, and stirred at room temperature overnight. The mixture was then diluted to 15 mL with DCM. Water was added, and the organic layer was washed three times with 5% NaHCO₃ in brine, and then concentrated under vacuum to give crude 3a–3i. The crude compounds were used directly in the next step without additional purification.

Preparation of 4a–4i

The crude products of 3a–3i were diluted in 10 mL of DCM. Then, 3.0 mL of HCl (4 M in dioxane) was added dropwise at 0 °C. Then the mixture was stirred for 6 h at room temperature. The precipitate formed was filtered, then washed with DCM, and then dried to give the final products 4a–4i.

3.1.3. Compound characterization

Key intermediates

(2S)-2-Amino-3-phenyl-N-((tetrahydro-2H-pyran-2-yl)oxy)propanamide (2)

Yield 91%, off-white solid, ¹H-NMR (400 MHz, CDCl₃) δ 7.19 (m, 5H), 4.86 (br, 1H), 3.88 (br, 1H), 3.56 (m, 1H), 3.23 (m, 1H), 2.69–2.89 (m, 1H), 1.52–1.78 (m, 7H).

For carboxylic acid intermediates: see ESI.†

Final products

(S)-5-Chloro-2-fluoro-N-(1-(hydroxyamino)-1-oxo-3-phenylpropan-2-yl)benzamide (4a)

Yield 67%, 2 steps from 2, mp 156.9–157.2 °C, ¹H-NMR (500 MHz, DMSO-d₆) δ 8.55 (s, 1H), 7.57 (s, 1H), 7.47 (s, 1H), 7.09–7.34 (m, 8H) (s, 1H), 2.90–3.04 (m, 2H); ¹³C-NMR (151 MHz, DMSO-d₆) δ 167.6, 163.7, 149.9, 146.4, 138.5, 137.8, 131.2, 129.8, (d, J = 251.8, 13.5 Hz); HRMS (ESI) m/z [M + H]⁺ calcd for C₁₆H₁₃ClFN₂O₃⁻ 335.0604, found 335.0601.

(S)-3,4-Dichloro-N-(1-(hydroxyamino)-1-oxo-3-phenylpropan-2-yl)benzamide (4b)

Yield 49%, 2 steps from 2, mp 150.1–151.2 °C; ¹H-NMR (500 MHz, DMSO-d₆) δ 10.86 (s, 1H), 8.93 (m, 3H), 8.08 (s, 1H), 7.75–7.78 (dd, J = 20, 10 Hz, 2H), 7.26–7.33 (m, 4H), 7.19–7.20 (t, J = 8.5 Hz, 1H), 4.62 (br, 1H), 2.99–3.08 (m, 2H); ¹³C-NMR (151 MHz, DMSO-d₆) δ 168.1, 164.4, 138.5, 134.8, 134.5, 131.6, 131.1, 129.9, 129.6 (2C), 128.6 (2C), 128.2, 126.8, 53.4, 37.8; HRMS (ESI) m/z [M–H]⁻ calcd for C₁₆H₁₃Cl₂N₂O₃⁻ 351.0309, found 351.0311.

(S)-4-Cyclohexyl-N-(1-(hydroxyamino)-1-oxo-3-phenylpropan-2-yl)benzamide (4c)

Yield 44%, 2 steps from 2, mp 139.2–140.0 °C; ¹H-NMR (400 MHz, DMSO-d₆) δ 10.76 (s, 1H), 8.50 (d, J = 4.0 Hz, 1H), 7.71 (br, 2H), 7.14–7.29 (m, 7H), 4.57 (br, 1H), 2.99 (br, 2H), 1.75 (br, 6H), 1.22–1.37 (br, 5H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.7, 166.6, 151.6, 138.6, 132.2, 129.7 (2C), 128.6 (2C), 128.1 (2C), 126.9 (2C), 126.8 (2C), 53.2, 44.2, 38.0, 34.3, 26.8 (2C), 26.1; HRMS (ESI) m/z [M–H]⁻ calcd for C₂₂H₂₅N₂O₃⁻ 365.1871, found 365.1871.

(S)-N-(1-(Hydroxyamino)-1-oxo-3-phenylpropan-2-yl)-4-(phenoxymethyl)benzamide (4d)

Yield 47%, 2 steps from 2, mp 176.1–176.6 °C; ¹H-NMR (500 MHz, DMSO-d₆) δ 10.81 (s, 1H), 8.91 (s, 1H), 8.67 (d, J = 10.5 Hz, 1H), 7.82 (d, J = 10.5 Hz, 1H), 7.50 (d, J = 10.5 Hz, 1H), 7.24–7.34 (m, 6H), 7.16 (t, J = 9.0 Hz, 1H), 7.0 (d, 2H, J = 9.5 Hz), 6.95 (t, J = 9.0 Hz, 1H), 5.15 (s, 1H), 4.57–4.63 (m, 1H), 3.03 (m, 2H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.6, 166.5, 158.7, 141.0, 138.8, 134.0, 130.1, 129.7 (2C), 128.7 (2C), 128.2 (2C), 127.7 (2C), 126.9 (2C), 126.8, 121.4, 115.4 (2C), 69.1, 53.3, 38.0; HRMS (EI−) m/z calcd for C₂₃H₂₁N₂O₄⁻ [M–H]⁻ 389.1507, found 389.1506.

(S)-N¹-(1-(Hydroxyamino)-1-oxo-3-phenylpropan-2-yl)-N⁴-phenylterephthalamide (4e)

Yield 51%, 2 steps from 2, mp 254.1–255.5 °C; ¹H-NMR (500 MHz, DMSO-d₆) δ 10.87 (s, 1H), 10.35 (s, 1H), 8.95 (s, 1H), 8.86 (d, J = 8.5 Hz, 1H), 8.02 (d, J = 8.5 Hz, 2H), 7.95 (d, J = 8.5 Hz, 2H), 7.81 (d, J = 8.0 Hz, 2H), 7.37 (m, 4H), 7.29 (t, J = 7.5 Hz, 2H), 7.20 (t, J = 7.5 Hz, 1H), 7.14 (t, J = 7.5 Hz, 1H), 4.63–4.68 (q, J = 15.5, 8.0 Hz, 1H), 3.08 (d, J = 7.5 Hz, 2H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.4, 165.9, 165.4, 139.5, 138.6, 137.7, 137.0, 129.7 (2C), 129.1 (2C), 128.6 (2C), 127.98 (2C), 127.95 (2C), 126.8, 124.3, 120.9 (2C), 53.3, 37.9; HRMS (EI−) m/z calcd for C₂₃H₂₀N₃O₄ [M–H]⁻ 402.1448, found 402.1459.

(S)-N¹-(4-Chlorophenyl)-N⁴-(1-(hydroxyamino)-1-oxo-3-phenylpropan-2-yl)terephthalamide (4f)

Yield 62%, 2 steps from 2, mp 270.3–271.0 °C; ¹H-NMR (400 MHz, DMSO-d₆) δ 10.87 (s, 1H), 10.48 (s, 1H), 8.96 (s, 1H), 8.86 (d, J = 8.4 Hz, 1H), 8.01 (d, J = 8.4 Hz, 2H), 7.95 (d, J = 8.4 Hz, 2H), 7.84 (d, J = 8.8 Hz, 2H), 7.44 (d, J = 8.8 Hz, 2H), 7.35 (d, J = 8.8 Hz, 2H), 7.28 (t, J = 7.6 Hz, 2H), 7.15 (t, J = 7.6 Hz, 1H), 4.61–4.67 (q, J = 8.0, 7.6 Hz, 1H), 3.06 (d, J = 7.6 Hz, 2H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.4, 166.0, 165.5, 138.7, 138.5, 137.5, 137.2, 129.7 (2C), 129.1 (2C), 128.7 (2C), 128.1 (2C), 128.1 (2C), 128.0, 126.9, 122.5 (2C), 53.4, 38.0; HRMS (EI−) m/z calcd for C₂₃H₁₈ClN₃O₄⁻ [M–H]⁻ 436.1069, found 436.1069.

(S)-N-(1-(Hydroxyamino)-1-oxo-3-phenylpropan-2-yl)-4-(3-phenylureido)benzamide (4g)

Yield 48%, 2 steps from 2, mp 200.0–202.1 °C; ¹H-NMR (500 MHz, DMSO-d₆) δ 10.78 (s, 1H), 8.92 (s, 1H), 8.89 (s, 1H), 8.75 (s, 1H), 8.47 (d, J = 8.5 Hz, 1H), 7.76 (d, J = 8.5 Hz, 2H), 7.48 (dd, J = 17.5, 8.5 Hz, 4H), 7.33 (t, J = 7.0 Hz, 2H), 7.30 (dd, J = 15.5, 7.5 Hz, 4H), 7.19 (t, J = 7.5 Hz, 1H), 7.01 (t, J = 7.5 Hz, 1H), 4.57–4.62 (m, 1H), 3.07 (m, 2H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.7, 166.1, 152.8, 143.0, 139.9, 138.8, 129.6 (2C), 129.3 (2C), 129.0 (2C), 128.6 (2C), 127.5, 126.7, 122.6, 118.8 (2C), 117.4 (2C), 53.1, 37.9; HRMS (EI−) m/z calcd for C₂₃H₂₁N₄O₄ [M–H]⁻ 417.1568, found 417.1564.

(S)-N-(1-(Hydroxyamino)-1-oxo-3-phenylpropan-2-yl)-4-((4-methylphenyl)sulfonamido)benzamide (4h)

Yield 21%, 2 steps from 2, mp 126.0–126.6 °C, ¹H-NMR (500 MHz, DMSO-d₆) δ 10.75 (s, 1H), 10.56 (s, 1H), 8.49 (d, J = 8.5 Hz, 1H), 7.65–7.69 (m, 4H), 7.36 (d, J = 8.0 Hz, 2H), 7.30 (d, J = 7.5 Hz, 2H), 7.26 (t, J = 7.5 Hz, 2H), 7.17 (t, J = 7.5 Hz, 1H), 7.12 (d, J = 9.0 Hz, 2H), 4.50–4.55 (q, J = 8.0, 7.5 Hz, 1H), 2.98 (d, J = 7.5 Hz, 2H), 2.33 (s, 3H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 168.5, 166.0, 144.0, 141.0, 138.7, 136.9, 130.3 (2C), 129.6 (2C), 129.2 (2C), 128.7 (2C), 128.5, 127.2 (2C), 126.7, 118.6 (2C), 53.2, 37.8, 21.4; HRMS (EI−) m/z calcd for C₂₃H₂₂N₃O₅S [M–H]⁻ 452.1358, found 452.1360.

(S)-N-(1-(Hydroxyamino)-1-oxo-3-phenylpropan-2-yl)quinoline-2-carboxamide (4i)

Yield 28%, 2 steps from 2, mp 121.0–123.9 °C; ¹H-NMR (500 MHz, DMSO-d₆) δ 10.89 (br, 1H), 8.76 (br, 1H), 8.58 (br, 1H), 8.10–8.17 (m, 3H), 7.90 (br, 1H), 7.75 (br, 1H), 7.19–7.28 (m, 6H), 4.72 (br, 1H), 3.12 (s, 1H); ¹³C-NMR (125 MHz, DMSO-d₆) δ 167.6, 163.7, 149.9, 146.4, 138.6, 137.9, 131.2, 129.8 (2C), 129.5, 128.8 (2C), 128.7, 128.7, 128.6, 127.0, 119.0, 52.5, 38.8; EIMS+ m/z 337 (M + 2, 25), 336 (M + 1, 100); HRMS (EI−) m/z calcd for C₁₉H₁₅N₃O₃⁻ [M–H]⁻ 334.1197, found 334.1199.

3.2. Biology

3.2.1. QSI activities of the library compounds

The synthesized compounds 4a–4i were screened for use in the quorum sensing system of P. aeruginosa-lasB-gfp. As a result, six compounds 4a, 4c–4e, 4h–4i showed a good quorum sensing inhibitory activity with an IC₅₀ range from 7.12 ± 2.11 to 92.34 ± 2.09 μM. Compound 4h (IC₅₀ of 7.12 ± 2.11 μM) is the strongest in this series, which is four-fold better than that of reference compound 4NPO (IC₅₀ of 29.13 ± 0.88 μM). No growth inhibition was observed under the conditions tested for the quorum sensing inhibitory activity (data not shown) (Table 1).

The IC₅₀ of library compounds for P. aeruginosa-lasB-gfp.

Compound		IC50 (μM)
4a	R = 5-chloro-2-fluorophenyl	10.03 ± 1.04
4b	R = 3,4-dichlorophenyl	>100
4c	R = 4-cyclohexylphenyl	9.44 ± 1.44
4d	R = 4-(phenoxymethyl)phenyl	83.54 ± 1.11
4e	R = 4-(phenylcarbamoyl)phenyl	92.34 ± 2.09
4f	R = 4-((4-chlorophenyl)carbamoyl)phenyl	>100
4g	R = 4-(3-phenylureido)phenyl	>100
4h	R = 4-((4-methylphenyl)sulfonamido)phenyl	7.12 ± 2.11
4i	R = quinolin-2-yl	50.23 ± 2.14
4NPO		29.13 ± 0.88

3.2.2. Anti-biofilm formation

The quorum sensing system plays a crucial role in the formation of biofilm, which was demonstrated previously.^6–8 Therefore, the library compounds were tested for their anti-biofilm formation via the quantification of the biofilm biomass. The results are shown in Fig. 2 and Table 2.

Fig. 2. Quantification of the biofilm biomass. The anti-biofilm formation (biomass) of active compounds 4a–4i and the reference compound 4NPO towards P. aeruginosa after 48 h growth with the concentration of 30 μM. The DMSO is a negative control. The results were obtained from three independent assays.

Inhibition (%) of the biofilm biomass of P. aeruginosa at 30 μM (subtracted median values from Fig. 2).

Cpds	Inhibition (%)
4a	40%
4b	<1%
4c	55%
4d	20%
4e	11%
4f	<1%
4g	<1%
4h	75%
4i	<1%
4NPO	74%

Interestingly, the inhibition of the biofilm biomass of P. aeruginosa by the library compounds is in agreement with their ability to show quorum sensing inhibition. In detail, the most active quorum sensing inhibitor compounds (4a, 4c, 4h) are the most active anti-biofilm formation agents. 4h, the strongest quorum sensing inhibitor, and also the strongest anti-biofilm reagent (75% inhibition), was equal to the reference compound, 4NPO. Other compounds show moderate to no activity towards the biofilm biomass (Fig. 2, Table 2).

3.2.3. CviR inhibition

Inhibition of the CviR receptor is one of the main mechanisms for quorum sensing inhibitory activity. Several studies have identified that compounds that inhibit the CviR receptor could reduce the formation of biofilm.¹⁹ Therefore, in order to understand the mechanism of action, the synthesized compounds were screened toward the quorum sensing receptor (CviR) of C. violaceum. In addition, the results also revealed the anti-violacein activities of the compounds. The overall results are shown in Fig. 3 and Table 3.

Fig. 3. The activities of 4a–4i and reference 4NPO on the violacein production of C. violaceum ATCC 31532 at 30 μM, with DMSO used as a negative control.

The CviR inhibition (%) of C. violaceum ATCC 31532 at 30 μM (subtracted median values from Fig. 3).

Cpds	Inhibition (%)
4a	58%
4b	35%
4c	65%
4d	11%
4e	<1%
4f	<1%
4g	<1%
4h	79%
4i	<1%
4NPO	65%

Compounds 4a–4c (35–58% inhibition) showed good CviR inhibitory activity which is equal to that of the reference compound 4NPO (65%). In this series, 4h (79% inhibition) showed excellent activity which is almost 1.5 fold stronger than that of 4NPO. Interestingly, according to our studies, the most potent quorum sensing inhibitor compounds (4a, 4c, and 4h) are also the most potent CviR inhibitors, indicating a relationship between quorum sensing inhibitors (QSI) and CviR inhibitors. Taken together with the biological results, we envisioned that the ability to inhibit the CviR receptor could be one of the main mechanisms for the QSI of the library compound. These results were also in agreement with their anti-biofilm formation activity.

3.2.4. Preliminary structure–activity relationship

In our series of compounds, increasing the polarity, length and introduction of the aromatic ring in the phenyl moiety seems to reduce the QSI activity, as evidenced by the gradual decreases of activity from 4c (R = 4-cyclohexylphenyl, IC₅₀ of 9.44 ± 1.44) to 4d (R = 4-(phenoxymethyl)phenyl, IC₅₀ of 83.54 ± 1.11), 4e (R = 4-(phenylcarbamoyl)phenyl, IC₅₀ of 92.34 ± 2.09), 4f (R = 4-((4-chlorophenyl)carbamoyl)phenyl, IC₅₀ > 100), 4g (R = 4-(3-phenylureido)phenyl, IC₅₀ > 100). However, the presence of the phenylsulfonamido moiety (compound 4h, IC₅₀ of 7.12 ± 2.11) is good for biological activity, including QSI activity, anti-biofilm formation, and inhibition of the CviR receptor. This phenomenon can be explained by the role of sulfonamide in the quorum sensing activity which was reported previously.^20,21

However, in order to fully understand the structure–activity relationship of these novel structures, the expansion of the library compounds is needed. The preliminary structure–activity relationship of the compounds in this work is shown in Fig. 4.

Fig. 4. Preliminary structure–activity relationship of the compounds in this work.

3.2.5. Docking studies

The most active compounds: 4a, 4c, 4h, and reference compound 4NPO were selected for the computational study for a better understanding of the mechanism of action. As reported previously, LasR is one of the notable target proteins for the QSIs of our assays. Therefore, the docking experiments have been performed to briefly understand the mechanism of action. Firstly, in order to validate the docking procedures, we conducted redocking experiments using the native ligands and the crystal structures of the LasR protein from P. aeruginosa (PDB code: 2UV0). The N-acyl homoserine lactone (AHL) exhibited a superposition to the co-crystal ligands located in the protein active sites, as indicated by an RMSD value of 1.0604. Key hydrogen bonds with residues such as Tyr56, Trp60, Asp73, and Ser129 or a pi-stacking interaction with Trp88 were preserved and have been described before (Fig. 5).¹⁶

Fig. 5. A) Docking results of AHL, 4a, 4c, 4h and 4NPO in the target (LasR ligand, PDB ID: 2UV0). B) Interactions of docked poses with the binding pocket.

Three compounds, 4a, 4c, and 4h, together with the reference compound 4NPO, were then docked into the active site of P. aeruginosa quorum-sensing LasR. The docking results are presented in Fig. 5. As a result, all the compounds bound perfectly to the active site of the target protein. Compound 4h exhibited the best docking score of −18.87 kcal mol⁻¹, whereas compounds 4a and 4c showed scores of approximately −17.54 and −15.55 kcal mol⁻¹, respectively. In comparison to the reference compound 4NPO, which scored only −9.05 kcal mol⁻¹, all three compounds displayed superior docking scores (Table 4). Furthermore, all the compounds exhibited a high degree of overlap (see Fig. 5), indicating there was a common binding motif in their interactions with the active site.

Prediction of binding energy.

Compound	Free binding energy scorea (kcal mol−1)
4a	−15.5510
	−15.5489
	−15.2869
4c	−17.5400
	−17.2109
	−16.9203
4h	−18.8719
	−18.8366
	−18.7966
4NPO	−9.0484
	−8.7662
	−8.1083

^aThree best energies.

As shown in Fig. 5, several strong bindings were observed between the benzamide and hydroxamic moieties of compounds 4a, 4c, and 4h with the protein pocket. Compound 4a has two halogen atoms at positions 2- and 5-, which could be steric hindrances which prevent the formation of hydrogen bonds with the surrounding residues, such as Ser129. Only one strong hydrogen bond was formed between residue Arg61 and the hydroxamic moiety in compound 4a, suggesting it had the lowest QSI activity among the three compounds. In contrast, compounds 4c and 4h exhibited several strong hydrogen bonds with the benzamide groups and residues Arg61, Thr115, and Ser129. In the case of compound 4h, there was an additional hydrogen bond between the sulfonamide group and Gly126, resulting in the best binding energy. The computational results are in agreement with the biological activity of this series of compounds.

As stated in the preliminary structure–activity relationship, the physicochemical properties and pharmacokinetics of the compounds might play a crucial role in bioactivity. Therefore, the QSI compounds 4a, 4c–4e, 4h, and 4i were calculated for their drug-likeness, and absorption, distribution, metabolism, and excretion (ADME) properties (Table 5).

The ADME prediction of the quorum sensing inhibitor compounds 4a, 4c–4e, 4h, 4i, and 4NPO.

Cpd	Formula	MW	#Rota	#H-ab	#H-dc	MRd	TPSAe	Solubilityf (mg mL)	Sol. classg	GI abs.h	Bio.i	P-gp subs.j
4a	C16H14ClFN2O3	336.75	7	4	3	82.24	78.43	8.52 × 10−2	Soluble	High	0.55	No
4c	C22H26N2O3	366.45	8	3	3	104.16	78.43	5.83 × 10−3	Moderately soluble	High	0.55	No
4d	C23H22N2O4	390.43	10	4	3	108.25	87.66	2.09 × 10−2	Moderately soluble	High	0.55	No
4e	C23H21N3O4	403.43	10	4	4	111.48	107.53	4.64 × 10−2	Soluble	High	0.55	No
4h	C23H23N3O5S	453.51	10	5	4	119.35	132.98	2.53 × 10−2	Moderately soluble	Low	0.55	No
4i	C19H17N3O3	335.36	7	4	3	92.57	91.32	8.42 × 10−2	Soluble	High	0.55	No
4NPO	C5H5NO	95.1	0	1	0	27.6	25.46	4.68 × 101	Very soluble	High	0.55	No

^aNumber of rotation bonds.

^bNumber of H-bond acceptors.

^cNumber of H-bond donors.

^dMolecular refractivity.

^eTopological polar surface area.

^fIntrinsic solubility at 25 °C calculated by Delaney's ESOL equation.

^gSolubility class, ESOL class.

^hGastrointestinal absorption: according to the white of the BOILED-Egg, SwissADME.

ⁱAbbott bioavailability score: probability of F > 10% in rat calculated by SwissADME.

^jP-glycoprotein substrate: SVM model built on 1033 molecules (training set) and tested on 415 molecules (test set) 10-fold CV: ACC = 0.72/AUC = 0.77 external: ACC = 0.88/AUC = 0.94.

As a result, all the compounds meet the criteria of the Lipinski rules (drug-likeness) with similarities in physicochemical properties and pharmacokinetics. In addition, computational calculation suggests that all the compounds showed promising pharmacokinetic properties, which are suitable for development as orally administered drugs.

4. Conclusion

In this work, we have applied the fragment-based drug design method to the design of phenylalanine derivatives bearing hydroxamic acid as quorum sensing inhibitors. Three compounds 4a, 4c, and 4h, show good quorum sensing inhibitory activity, anti-biofilm formation, and inhibition of the CviR receptor. In this series, 4h, which contains the sulfonamido group, showed excellent biological activities in all assays. Docking studies also confirmed that compounds 4c and 4h exhibited several strong hydrogen bonds with the benzamide groups and residues: Arg61, Thr115, and Ser129. In particular, there was an additional hydrogen bond between the sulfonamide group of 4h and Gly126, resulting in the best binding energy. A preliminary structure–activity relationship revealed that increasing the polarity, length, and introduction of the aromatic ring in the phenyl moiety seemed to reduce the quorum sensing inhibitory activity. In order to fully understand the structure–activity relationship of this novel structure, the expansion of this library compound is needed, which is ongoing in our laboratory. The results will be published in due course.

Conflicts of interest

There are no conflicts to declare.