Inhibition of Pseudomonas aeruginosa quorum sensing by methyl gallate from Mangifera indica

Antipathogenic drugs are a potential source of therapeutics, particularly following the emergence of multiple drug-resistant pathogenic microorganisms in the last decade. The inhibition of quorum sensing (QS) is an advanced antipathogenic approach for suppression of bacterial virulence and dissemination. This study aimed to investigate the inhibitory effect of some Egyptian medicinal plants on the QS signaling system of Pseudomonas aeruginosa. Among the tested plants, Mangifera indica exhibited the highest quorum sensing inhibition (QSI) activity against Chromobacterium violaceum ATCC 12472. Four pure compounds were extracted and identified; of these, methyl gallate (MG) showed the most potent QSI. MG had a minimum inhibitory concentration (MIC) of 512 g/mL against P. aeruginosa strains PAO1, PA14, Pa21, Pa22, Pa23, Pa24, and PAO-JP2. The virulence factors of PAO1, PA14, Pa21, Pa22, Pa23, and Pa24 were significantly inhibited by MG at 1/4 and 1/2 sub-MICs without affecting bacterial viability. Computational insights were performed by docking the MG compound on the LasR receptor, and the QSI behavior of MG was found to be mediated by three hydrogen bonds: Trp60, Arg61, and Thr75. This study indicates the importance of M. indica and MG in the inhibition and modulation of QS and QS-related virulence factors in P. aeruginosa.


Structure clarification and fractionation of isolated compounds from M. indica
The chromatographic analysis of the M. indica ethyl acetate extract led to the identification of four known phenolic components.By examining NMR and HRMS data and comparing it to spectroscopic data previously published in the literature, compound structural identities were confirmed.

Determination of the QSI activity of M. indica isolated compounds
The QSI assay of the four purified compounds isolated from M. indica was evaluated by the reporter strain C. violaceum ATCC 12472.MG showed potent QSI activity, with inhibition of violacein pigment formation at 33 mm diameter.The other three compounds did not exhibit any QSI activity.As a result, MG was chosen for the assessment of antipathogenic activity against clinical isolates and standard strains of P. aeruginosa.

LasR and ligand binding affinity analysis using molecular docking
The autoinducer 3-oxo-C12-HSL was docked at the LasR binding site.As reported previously, this revealed three hydrogen bonds with Ser129, Trp60, and Asp73 and an ICM score of − 107.47.The active site of LasR requires all three of these amino acids.With Trp60, Arg61, and Thr75, MG formed three H-bonds with an ICM score of − 57.18 (Table 2, 3) (Fig. 5A,B).2) (Fig. 5C).On the other hand, MG also bound to zinc metal active sites and formed H-bonds with Gly397, ser398, and His399 (Fig. 5D).

Discussion
Pseudomonas aeruginosa is a nosocomial and opportunistic human pathogen that can be isolated from a variety of environments.QS controls pathogenesis by regulating gene expression and the release of virulence factors such as proteases, hemolysin, pyocyanin, and biofilm.The capability of P. aeruginosa to create a biofilm that can resist antibiotics has recently made treatment of Pseudomonas infections more challenging 28 .Antimicrobial resistance in P. aeruginosa is associated with its ability to produce degrading enzymes that can inactivate antimicrobial agents and efflux pumps that have antibiotic resistance genes 29 .As a result, finding new approaches for combating Pseudomonas infection is critical; one of these is inhibiting QS and bacterial cross-talk, which can manage bacterial virulence factors.
In this study, the anti-QS activities of several Egyptian medicinal plants were evaluated.C. violaceum ATCC 12472 is the most used strain for determining the QSI of compounds by inhibition of its distinctive violet pigment.The production of violacein pigment by reporter strain C. violaceum ATCC 12472 is under the control of QS signals (AHL).Therefore, inhibition of violacein pigment without affecting bacterial growth implies potential QSI 30,31 .
The pulp, bark, leaf, fruit peel, heartwood, and seed of M. indica have numerous medicinal uses 32 for treatment of a wide range of human diseases including diarrhea, cough, anemia, gastric disorders, jaundice, malaria, hepatic disorders, hemorrhage, diabetes, and ulcers 33 .
Among the tested extracts, M. indica had the greatest ability to inhibit this pigment without any effect on growth.The total extracts of M. indica demonstrated potential QSI activity 34 .Hence, fractionation of M. indica extract was performed.The highest QSI was produced by a MG compound.Similarly, MG isolated from other plants like Piper betle (betel), Anacardium occidentale (cashew), and Anacardium occidentale L. (cashew) exhibited anti-Qs activity against C. violaceum 12472 35,36 .Additionally, MG displayed QSI activity against other pathogenic strains like Salmonella Typhimurium alone and in combination with marbofloxacin at 30 μg/mL concentration 37 .
Furthermore, MG is a well-known antibacterial compound and it has been reported to have antibacterial activity against many bacterial strains.For example, it inhibited the growth of plant-pathogenic bacteria Ralstonia solanacearum 38 , and some oral bacteria such as Actinomyces viscosus, Streptococcus mutans, and S. sobrinus 39 .Additionally, it inhibited the growth of some Salmonella spp.clinical isolates 40 and it approved a synergistic effect against nalidixic acid-resistant bacteria when combined with nalidixic acid 41 .Hence, we tested its effectiveness as QSI on virulence factors of P. aeruginosa.
Pseudomonas aeruginosa causes many serious nosocomial infections, which are frequently linked to the development of biofilm and are often resistant to the majority of antibacterial drugs 28 .MG inhibited biofilm formation in all tested strains.Therefore, it could be an effective tool for inhibiting biofilm formation and making the biofilm more vulnerable to neutrophil phagocytosis and immune system responses 42 .P. aeruginosa produces a green pigment called pyocyanin, which is controlled by the RhlI/R and PQS signaling systems.The removal of this green pigment implies lower levels of pyocyanin content as compared with untreated cultures.Protease and hemolysin are Las-regulated virulence genes in P. aeruginosa and are affected by MG to various degrees.They are hydrolytic enzymes that are formed to assist bacteria to infect host tissues and escape the host's defenses 43 .The QSI activity of MG was obtained at sub-minimum inhibitory concentrations of the pure isolated compound (MG) from M. indica.As a result, pathogenicity and virulence factors were reduced without affecting viability.Our results are in line with previous work that emphasized the ability of MG to inhibit virulence factors regulated by QS in P. aeruginosa 44 .
Similarly, M. indica leaf extracts exhibit QSI with a decrease in total protease production of 43.8-56% and a 72% reduction in biofilm formation in P. aeruginosa at sub-MIC 800 mg/mL 34 .Furthermore, MG has been reported to disrupt biofilm formation in P. aeruginosa by 70% 45 and in S. mutans 46 .Also, MG was reported to inhibit pyocyanin production in P. aeruginosa PAO1 by 80% 44 , by 65% at the concentration of 12.5 μg/mL after 18 h of incubation 47 , by 64% at the concentration of 256 μg/mL 45 .Additionally, MG could eliminate protease production in P. aeruginosa by 51% at 256 μg/mL 45 and Gallic acid also eliminates protease production in P. aeruginosa 48 .
Hence, utilization of natural compound MG with elimination of QS-related virulence factors could be an appropriate alternative therapy for inhibition of microbial dissemination without development of microbial resistance.
On a structural basis, virtual screening and protein docking against LasR receptor was performed by using Molsoft ICM 3.4-8C software to determine the QS inhibitory potential of the MG compound 49 .The Protein Data Bank was used to obtain the PDB structure of the receptor protein LasR (PDB ID: 2UV0).Calculating the scoring and hydrogen bonds with the surrounding amino acids at the active site of LasR revealed the binding mode, affinity, and orientation of MG.According to PDBsum, the amino acids at the LasR active site (PDB ID: 2UV0) were Trp60, Asp73, Cys79, Tyr64, Gly126, Ala50, Trp88, Tyr56, Thr75, Tyr93, Ala105, Leu110, and Ser129.The polar groups of AHL and the residues Asp73, Trp60, Tyr56, and Ser129 of LasR form H-bond interactions that are responsible for the correct folding of the LuxR type of protein 30,50,51 .It has been reported that 3oxo-C 12 -HSL forms three H-bonds with the amino acids Asp73, Trp60, and Tyr56 31 .In agreement with the MG docking results, many H-bonds are formed between LasR and the HSL of autoinducer 3-oxo-C 12 -HSL, such as Ser 129, Tyr 56, Trp 60, Arg 61, Asp 73, and Thr 75, which is similar to findings on pyridoxal lactohydrazone 52 .Surface rendering of LasI revealed that the tunnel in the acyl-chain binding region is formed by the following residues: Trp33, Trp69, Met79, Leu102, Phe105, Thr121, Leu122, Met125, Leu140, Thr142, Thr144, Val148, Met151, Met152, Ala155, Leu157, Ile178, and Leu188.Several of these residues, including Met79, Phe105, Thr142, and Thr144, are well-conserved among AHL-synthases 53 .MG bound to Arg30 sulfate, the ligand of the enzyme, and leu102, Arg104, Thr 142, and Thr144, well-conserved amino acids.Moreover, MG also bound to a zinc ion binding site in the enzyme.Hence, it acts as a competitive inhibitor by a dual mechanism, either competing with sulfate for its binding site or binding to Gly397, ser398, and His399 amino acids to inhibit the catalytic role of the Zn ion into LasI enzyme.On the other hand, NHQ redocked into the PqsR active site had an ICM score of − 57.78 and formed H-bonds with Gln194, Leu208, and Ile236, which are amino acids at the active sites of the PqsR enzyme 54 .When MG was docked into the active site of PqsR, it had an ICM score of − 44.48 and formed H-bonds with amino acids Ile136, Gln138, Asp139, Glu142, Ile143, and Asn272.None of these were reported as amino acids at the active site, which suggests that the activity of MG is due to its binding with LasR and LasI rather than PqsR.

Conclusions
This study demonstrated that MG purified from M. indica medicinal plants leaves significantly reduced the pathogenicity and virulence factors including elastase, protease, pyocyanin production, hemolysin and distrusted biofilm of P. aeruginosa without affecting its growth rate for the first time.This could play a critical role in combating MDR, which has been identified by WHO as one of the top 10 global public health risks especially in immune-compromised patients.

Bacterial strains, culture conditions, and media
The reporter strain Chromobacterium violaceum ATCC 12472 55 was used to screen the QSI activity of the natural extracts.The strain was inoculated on Luria-Bertani (LB) media containing 1% (w/v) tryptone, 0.5% (w/v) yeast extract, and 1% (w/v) NaCl at pH 7, solidified with 2% (w/v) agar, and grown overnight at 28 °C for 48 h 56 .The clinical isolates of P. aeruginosa, Pa21, Pa22, Pa23, and Pa24, were obtained from urine samples according to the ethical committee standards of the Faculty of Medicine, Alexandria University, Egypt (Institutional Review Board (IRB), No. 0201472, 18 March 2021, Faculty of Medicine, Alexandria University, Egypt).The identification of P. aeruginosa was based on laboratory biochemical standards 57 .
The bacterial isolates Pa21, Pa22, Pa23, and Pa24 identified as P. aeruginosa (collection number QSCC6422 as Quorum Sensing Culture Collection, Mansoura University, Egypt).Standard strains including P. aeruginosa PA14 (DSM 19882), P. aeruginosa PAO1 (ATCC47085), are kindly provided by Prof. Keller, UW, USA 58 , and the QS-negative control strain P. aeruginosa PAO-JP2, are kindly provided by Prof. Martin Schuster, Department of Microbiology, Oregon State University, Nash Hall, Corvallis, OR 97331 59 .All P. aeruginosa strains were grown in LB broth media and kept in an incubator at 37 °C overnight.

Plant crude extracts
Fresh leaves of Mangifera indica L. were collected from Mansoura University gardens (permitted for researchers) in May 2017 and complied with institutional guidelines and legislation.The plant was verified by staff members of the Pomology Department, Faculty of Agriculture, Mansoura University.Aerial parts of Atriplex halimus L., Euphorbia paralias L., Retama raetam (Forssk.)Webb & Berthel., Eichhornia azurea, and Pituranthos tortuosus (Desf.)Benth.& Hook.f.ex Asch.& Schweinf.plants were collected from Borg El-Arab, Egypt in November 2016 after the permit of the landowners and according to institutional guidelines and legislation.The plants were identified by the staff members of the Botany and Microbiology Department, Alexandria University, Alexandria (Prof.Salama El-Darier Professor of botany, Botany and Microbiology Department, Alexandria University, Alexandria and Prof. Sania Kamal, Professor of botany, Botany and Microbiology Department, Alexandria University, Alexandria).
Voucher samples of the plants were preserved under the codes MI-L-17, AH-A-16, EP-A-16, RR-A-16, EA-A-16, and PT-A-16 in the Herbarium of Pharmacognosy Department, Faculty of Pharmacy.The plants were rinsed in tap water, allowed to dry in the shade for six weeks, and then pulverized into a fine powder.Each dry powder (50 g) was extracted with 70% ethanol and incubated overnight at 30 °C and 200 rpm.The plants were filtered, and extracts were concentrated using a rotary evaporator at 40 °C under vacuum.

QSI assay of crude extracts
The LB agar plates were prepared, and after solidification, they were overlaid with soft LB medium (1% w/v agar) inoculated with C. violaceum ATCC 12472.A sterile cork borer was used to carve out wells (10 mm diameter) in the double-layered plates.A 100 µL aliquot of the concentrated plant ethanolic extract was added to the corresponding wells.Plates were incubated at 28 °C for 48 h, and the inhibition of violacein pigment around each well was measured 55 .In the assay plates, a well with ethanol was used as a negative control.

Isolation of the active fraction of M. indica
The shade-dried leaves of M. indica, which showed the highest QSI activity, were pulverized to a powder (940 g).Next, 4 L of ethanol was used to extract the entire amount of dry powder for 48 h at room temperature, until it was completely used up.The combined ethanol extracts were dried in a vacuum using a rotary evaporator at 40 °C to yield 69.3 g of crude extract.The extract was dissolved in 300 mL water and divided in a polarity-based manner with petroleum ether, ethyl acetate, methylene chloride, and butanol.The fractions were collected, dried, and weighed (21.7, 18.9, and 19 g, respectively).

Characterization of isolated compounds by NMR spectroscopy
Before biological evaluation, NMR experiments (1D and 2D) were performed on a Varian Dual Broadband Probe 400 MHz, Bruker Avance III 600, or Bruker DRX-500 and 400 MHz spectrometer using pyridine-d 5 , CD 3 OD, or DMSO-d 6 as solvents to identify the isolated compounds.The internal index for modification was the solvent peaks.On a mass spectrometer (Agilent Technologies 6200 series), negative and positive ion modes of mass spectra were observed.To measure the precise rotations, an automatic polarimeter IV was used.UV spectra were recorded by a Varian Cary 50 Bio UV-visible spectrophotometer.Column chromatography was performed using a flash silica gel and reversed-phase C-18.Analytical TLC was performed on a silica gel F254 aluminum sheet (20 × 20 cm, Fluka) or a Silica 60 RP-18 F254S aluminum sheet (20 × 20 cm, Merck), and detection was performed using UV-254 nm.Next, 1% vanillin in concentrated H 2 SO 4 -EtOH (10:90) was sprayed and then heated with a heat gun to visualize the spots.Analytical-grade solvents (Fischer Chemicals) were used in the purification and isolation procedures.

Determination of the QSI activity of the purified compounds
Four isolated compounds from M. indica were dissolved in DMSO and tested for their QSI activity using C. violaceum ATCC 12472.Each well (10 mm diameter) was filled with 50 µL of each compound.The inhibition of the violet zone around the well of induced violacein production was measured in millimeters.DMSO was used as the control.

Determination of minimum inhibitory concentrations
Using the microtiter plate assay method, the minimum inhibitory concentration (MIC) of methyl gallate (MG) was determined.MH broth was distributed at 100 µL in each well.The MG compound was serially diluted 1:1 in 10 dilutions (0.5-512 µg/mL).P. aeruginosa cultures PAO1, PA14, Pa21, Pa22, Pa23, Pa24, and PAO-JP2 were added to each well to a final concentration of 1 × 10 5 CFU/well 60 .Wells containing media alone were considered as the negative control, while wells containing media inoculated with each of the tested strains were considered as the positive control.Microbial growth was detected in each well visually after overnight incubation and compared with the positive control 60 .

Effect on bacterial growth
Pseudomonas aeruginosa strains that were untreated or treated with MG at 1/2 MIC (256 µg/mL) were incubated overnight at 37 °C.Treated and untreated samples (1 mL) were taken every hour and diluted at 1:10 for bacterial counting.Using the pour plate method, the vitality of the treated P. aeruginosa samples was determined and compared to the untreated culture.Each diluted sample was added to melted nutrient agar medium (20 mL) at 50 °C, mixed, and poured into 9 cm sterile plates.The plates were incubated overnight at 37 °C, and the colonies were counted.The growth curves of P. aeruginosa (treated and untreated) were estimated 61,62 .

Effects of MG sub-MICs on Pseudomonas virulence factors
Pseudomonas aeruginosa clinical isolates Pa21, Pa22, Pa23, and Pa24 were treated with 1/4 and 1/2 MICs of MG.Cultures with and without MG were incubated overnight at 37 °C.Cells were removed by centrifugation, and the cell-free supernatant was used for the detection of Pseudomonas virulence factors (hemolysin, elastase, protease, and biofilm) in both untreated and treated cultures in triplicate 62,63 .The same conditions were used to evaluate the standard strains PAO-JP2, PA14, and PAO1 64 .

Determination of hemolysin activity
The hemolysin test of treated and untreated P. aeruginosa supernatants was performed by mixing bacteria supernatants with washed RBC suspension (2% v/v) 65 .The mixture of 500 of cell-free supernatant was mixed with 700 µL erythrocytes.The mixture was incubated at 37 °C for 2 h and then centrifuged at 4000 rpm for 10 min at 4 °C.The absorbance of the supernatant at OD 540 nm, which indicates the level of hemolysin activity, was measured 66 .

Biofilm assay
The development of biofilm by P. aeruginosa strains was assessed using the microtiter plate crystal violet staining method.Wells were inoculated with 100 µL of treated and untreated cultures, and the plates were incubated for 24 h at 37 °C.The wells were rinsed with saline and set for 15 min in 150 µL of absolute methanol.Crystal violet (1% w/v) was used to stain the bacterial cells, and any excess stain was removed by washing the plate with water.The plates were then allowed to air dry.A volume of 150 µL of glacial acetic acid; 33% (v/v) was used to dissolve the dye that bound to the formed biofilm, and the absorbance was determined at OD 490 67 .

Determination of protease production
The total protease production of P. aeruginosa cultures with or without 1/4 and 1/2 MICs of MG was assessed using the skim milk method 68 .The assay was carried out by mixing 500 µL of supernatants with 1 mL of 1.25% A reduction in OD 600 indicated that the skim milk was cleared and that proteolysis activity had increased 69 .

Pyocyanin assay
Pyocyanin pigment production was assessed by the cultivation of P. aeruginosa strains in King A broth media (MgCl 2 0.14% (w/v); peptone, 2% (w/v), and K 2 SO 4 , 1.0% (w/v)) with or without 1/4 and 1/2 MICs of MG 70 , and incubated at 37 °C for 48 h at 200 rpm.The cultures were centrifuged at 4 °C for 10 min at 3000 rpm, and chloroform was used to extract the pyocyanin.The mixture was vortexed after the addition of chloroform until the color turned greenish-blue and then centrifuged for 10 min at 3000 rpm.The chloroform layer was separated and a volume of 1 mL of 0.2 M HCL was added to the mixture with the development of pink color.The absorbance was evaluated at OD 520 nm.Pyocyanin concentration (μg/mL) was calculated by multiplying the OD 520 nm by 17.072 70,71 .

Molecular docking
To determine the binding affinities, binding modes, and inhibition behavior at the LasR active site, MG was docked into the binding site of three QS systems in P. aeruginosa: LasR (PDB ID: 2UV0) 72 , LasI (PDB ID: 1RO5) 53 , and PqsR (PDB ID: 4JVD) 54 to evaluate their binding affinities, inhibitory activities, and binding modes.The crystal structure was obtained using the protein data bank.Many of the protein's water ligands were extracted, and the components were created on ChemBioDraw using ChemOffice (PerkinElmer Informatics).The energy was reduced using MM2.PDB file 2UV0 was then transformed into an internal coordinate mechanics (ICM) object using Molsoft software to conduct docking experiments.ICM were as described by 49 .ICM seeks to identify the global minimum energy that most accurately characterizes the ligand-receptor interaction.The autoinducer molecule 3oxo-C12-HSL interacted with 2UV0 as a typical docked model.

Data analysis and statistics
The standard deviations and means of the three different measurements were computed using an Excel data sheet, and the experiments were performed in triplicate.Differences between strains that were untreated or treated with MG were considered significant when the probability value was *P ≤ 0.05, **P ≤ 0.01, or ***P ≤ 0.001.The results of the statistical analysis were evaluated by Welch's t-test.

Ethical approval and consent to participate
Institutional Review Board (IRB), No. 0201472, Faculty of Medicine, Alexandria University, Egypt, assessed this study and found it to be exempt on March 18, 2021.

Figure 1 .
Figure 1.Structures of four isolated compounds from M. indica.

Figure 3 .
Figure 3.Effect of sub-MICs of methyl gallate on hemolysin activity (A) and biofilm formation (B) of P. aeruginosa strains.

Table 2 .
Internal coordinate mechanics ratings, the number of H-bonds and amino acid residues.

Table 3 .
Molecular docking results of methyl gallate with interacting amino acids with LasI of P. aeruginosa.

Amino acid residues Atom of amino acid Atom of compound length Å
Figure 5. Binding modes of 3-oxo-C12-HSL (A) and MG (B) into P. aeruginosa active site; sulphate ligand (C) and MG (D) into LasI synthase of P. aeruginosa; NHQ ligand (E) and MG (F) into PqsR coinducer of P. aeruginosa.

PqsR and ligand binding affinity analysis NHQ
redocked into the active site of its receptor, PqsR, which was bound to it by amino acids Gln194, Leu208, and Ile236, with an ICM score of − 57.78.When MG was docked at the active site of PqsR, it had an ICM score of − 44.48 and formed six H-bonds with Ile136, Gln138, Asp139, Glu142, Ile143, and Asn272 (Table2) (Fig.5E,F).
w/v) skim milk.The mixture was incubated for 1 h at 37 °C, and the absorbance was determined at OD 600 nm.