Zinc oxide nanoparticles inhibits quorum sensing and virulence in Pseudomonas aeruginosa

Background Quorum sensing inhibitionis an advanced strategy that aims to interfere with bacterial cell-to-cell communication systems (quorum sensing), which regulate virulence factors production in Pseudomonas aeruginosa, in order to overcome the globalcrisis of antimicrobial resistance. Objectives Study the potential quorum sensing inhibitory effect of Zinc oxide (ZnO)nanoparticlesin Pseudomonas aeruginosa and the impact on production of virulence factors. Methods Quorum sensing inhibitory effect of ZnO was evaluated by assessing its ability to reducePseudomonas aeruginosa virulence factors production; rhamnolipids, pyocyanin, pyoverdin, hemolysins, elastase and proteases. Furthermore, qRT-PCR was performed to determine ZnO inhibitory effect onQS-regulatory geneslasI, lasR, rhlI, rhlR, pqsA and pqsR that control virulence factors secretion. Moreover, mice survival test was conducted to investigate the influence of ZnO on Pseudomonas aeruginosa-induced mortality in vivo. Results ZnO revealed a statistically significant reduction in the production of QS-controlled virulence factors rhamnolipids, pyocyanin, pyoverdin, hemolysins, elastase and proteases. Furthermore, ZnO exhibited a significant decrease in the relative expression of QS-regulatory geneslasI, lasR, rhlI, rhlR, pqsA and pqsR. Additionally, ZnO significantly reduced the pathogenesis of Pseudomonas aeruginosa in vivo Conclusion ZnO nanoparticles can be used as a quorum sensing inhibitor in Pseudomonas aeruginosa infections either as an adjuvant or alternative to conventional antimicrobials.

In Ps. aeruginosa, bacterial cells communicatewith each other through a process known as quorum sensing (QS) which plays a major role in bacterial pathogenesis 3 . In Ps. aeruginosa, there are three major QS systems, lasI/R system, rhlI/R system and pqsA/R system which are interconnected together through signaling chemical molecules (autoinducers); oxododecanoyl-homoserine lactone (C12-HSL), butyryl-homoserine lactone (C4-HSL) and Pseudomonas quinolone-based intracellular signal (PQS) produced by bacterial cells. When chemical autoinducers reach a certain threshold, the quorum, they trigger the genes that regulate the production of virulence factorssuch as pyocyanin, pyoverdin, hemolysins, elastase and proteases [4][5][6] . Quorum sensing inhibitors are agents that disrupt QS systems in bacterial cells leading to a reduction of virulence factors production and suppression of virulence without interrupting the bacterial growth and so no or low resistance is anticipated to arise against these agents 7 .
In the recent years, the advances accomplished in the field of nanotechnology resulted in an increase in the applications of nanoparticles in the medical sector andas a therapy for infectious diseases. Superior effectiveness on resistant strains of metal oxide nanoparticles such as Zinc oxide (ZnO) and silver has been reported. ZnO nanoparticles were found to exert a potent antimicrobial activity and significantly reduced skin infections and inflammation in mice [8][9] . The current study aimed to investigate the possible quorum sensing inhibiting activity of ZnO nanoparticles and their potential role in reducing QS-controlled virulence factors production and pathogenesis in Ps. aeruginosa.

Materials and methods Bacterial isolates and their identification
Ps.aeruginosa PAO1 wild-type standard strain and five clinical isolates(Ps1, Ps2, Ps3, Ps4 and Ps5) were used in this study. Ps.aeruginosa PAO1 was provided from the stock culture collection of Microbiology and Immunology Department, Faculty of Pharmacy, Zagazig University. Clinical isolates were isolated from patients with burn and surgical wound infections admitted to Port Said General Hospital, Egypt. Clinical isolates were identified by Gram-stain, production of green pigmentson nutrient agar, growth on MacConkey agar, oxidase test, motility, growth on selective mediumcetrimide agar and the ability to grow at 42°C as stated by Koneman et al 10 .

Determination of minimum inhibitory concentration and investigating the effect of sub-inhibitory concentration of ZnO nanoparticles on bacterial growth
The minimum inhibitory concentration (MIC) of ZnO nanoparticles was determinedby using the agar dilution method according to(CLSI) 11 . Briefly, overnight bacterial cultures of the tested isolateswere diluted, each with Mueller-Hinton broth to reach a turbidity matching that of 0.5 MacFarland Standard and then with sterile saline to achieve a final concentration of 10 7 CFU/ml. Nutrient agar plates with different concentrations of ZnO (1,2,4,8,16, 32 and 64mg/ml) were prepared in addition to control plates without ZnO. The plates' surfaces were inoculated with 1µl of the suspensions of the tested isolates and incubated overnight at 37°C. The MIC was calculated as the least concentration of ZnO that prevented the visible growth of bacteria. To ensure that ZnO sub-MIC that would be used in further experiments had no influence on bacterial viability, the effect of ¼ MIC of ZnO on bacterial growth was assessed following Nalca et al 12 . The tested isolates were incubated in Luria-Bertani (LB) broth (tryptone 10 g, yeast extract 5g and 10 g sodium chloride in 1000 ml distilled H2O) with and without ¼ MIC of ZnO under the same conditions. After 24h of incubation at 37°C, the optical densities of ZnO-treated and untreated cultures were measured at OD600 using spectrofluorometer (Biotek, USA). The phenotypic effect of ZnO nanoparticles on QS-controlled virulence factors production Effect on rhamnolipids Rhamnolipids production, in the presence and absence of ZnO, was assessed by oil spreading method according to Morikawa et al 13 . A thin oily layer was formed on the surface of water by addition of 20µl of crude oil to 15 ml of distilled H 2 O in a Petri dish. Ten µl of cell-free supernatants of the tested isolates with and without ¼ MIC of ZnO was added to the center of the oily layer. The diameters of the clear zones formed that are related to the biosurfactant activity and the amounts of rhamnolipids produced by the tested isolates were measured and compared.

Effect on pyocyanin
Pyocyanin determination was performed in King A medium (peptone 20 g, MgCl 2 1.4 g and 10 g K 2 SO 4 in 1000 ml distilled H 2 O)according to Essar et al 14 . The tested isolates were grown in King A mediumwith and without¼ MIC of ZnO for 48 h at 37° C. Pyocyanin was extracted by addition of aliquots of 2.5 ml of bacterial supernatants to 3 ml chloroform followed by mixing with 1 ml of 0.2 N HCl. The pigment in chloroform layer was measured at OD520 using spectrofluorometer (Biotek, USA).

Effect on pyoverdin
In order to estimate pyoverdin, the method of Coxan-dAdams 15 was used. Overnight cultures of the tested isolatesin LB broth were prepared in the presence and absence of ¼ MIC of ZnO and then centrifuged at 10000 rpm for 10 min. The cell-free supernatants were diluted to 1/10 with 50 mM Tris-HCl and pH adjusted to 7.4. The pyoverdin fluorescence in supernatants was measured at 460 nm,where the samples were excited at 400 nm using spectrofluorometer (Biotek, USA).

Effect on hemolysins
Production of hemolysin was determined following the modified method ofDacheux et al 16 . Aliquots of 0.5 ml of cell-free supernatants of tested isolates in LB broth with and without ¼ MIC of ZnO were mixed with 0.7 ml of 2% sheep RBCs in saline followed by incubation at 37°C for 2 h. After centrifugation of assay mixtures at 2500 rpm for 5 min to remove any cells, the released hemoglobin was measured at OD540 nm. Percentage lysis was calculated from the formula: [X-B/T-B]x 100, where B is the negative control corresponding to RBCs in LB broth, T is the positive control corresponding to completely lysed RBCs with 0.1%SDS and X is the ZnO-treated or untreated isolates. The hemolytic activity percentage produced byZnO-treated isolates was compared to that produced by untreated isolates.

Effect on elastase
Elastase assay was evaluated according to Ohman et al 17 using ECR. Briefly, an aliquot of 0.5 ml of ECR solution (10 mg/ml) in Tris buffer (pH 7.0) was inoculated with 0.25ml of each cell-free supernatant of the tested isolates prepared in the presence and absence of ¼ MIC of ZnO. The mixtures were left at 37°C for 6 h and then centrifuged at 10000 rpm for 10 min to remove insoluble ECR pellets. The color of released ECR in supernatantswas measured at OD495 nm using spectrofluorometer(Biotek, USA).

Effect on proteases
The total proteases were estimated by the modified skim milk assayas described by El-Mowafy et al 18 . An aliquot of 0.5 ml cell-free supernatant of each of the testedisolates prepared with and without ¼ MIC of ZnOwasmixed with 1 ml of skim milk solution (1.25% in distilled H2O) and incubated at 37°C for 30 minutes. The turbidities of assay mixtures were measured at OD600using spectrofluorometer as a measure of the proteolytic activity.

Estimation of relative gene expression of QS-regulatory genes using qRT-PCR
For molecular determination of QS-regulatory genes,total bacterial RNA was extracted at the middle of the log phase,corresponding to OD600of 0.5-0.6,from the tested strains cultivated overnight in LB broth at 37°C in presence and absence of ¼ MIC of ZnO using Gene-JET RNA Purification Kit following the manufacturer instructions.Reverse transcription followed by qRT-PCR of QS-regulatory genes lasI, lasR, rhlI, rhlR, pqsA and pqsR was carried out using SensiFAST™ SYBR® Hi-ROX One-Step Kit.StepOne Real-Time PCR thermal cycler utilizing primers illustrated in Table 1 was used to setup the qRT-PCR analysis. The relative expression values of QS-regulatory genes were normalized to the housekeep-ing gene rpoD and agarose gel electrophoresis was used to confirm the specific PCR amplification. The relative gene expression in ZnO-treatedcultures was compared to their expression levels in untreated ones following the 2-∆∆Ct method 19 .

Mice survival test
The influence of ZnO on Ps. aeruginosa pathogenesis was assessed by the mice survival in vivo model following the method of Kim et al 20 . The ethical standards of Medical Research Center, Ain Shams University, Cairo, Egypt, where the experiment was conducted and the mice were provided, were followed in the animal study. An approximate cell density of 2.5 x 10 7 CFU/ml in phosphate-buffered saline (PBS) of Ps. aeruginosa PAO1 was prepared from overnight bacterial cultures in LB broth with and without ¼ MIC of ZnO. Four random groups of three-weeks-old healthy female albino mice (Mus musculus) with the same weight were used, each comprising 10 mice. In Group 1,mice were injected intraperitoneallywith 100 µl of ZnO-treated bacteria in sterile PBS, while group 2 was injected with 100 µl of untreated bacteria. Two negative control groups are included also;group 3 miceare injected with 100 µl of sterile PBS andgroup 4 mice were left uninoculated. All groups were kept with normal feeding and aeration at room temperature. The survival of mice in each group was recorded every day for 3 successive days.The results were calculated using Log-rank test,GraphPad Prism 5 and plotted using Kaplan-Meier method.

Statistical analysis
The influence of ZnO on Ps. aeruginosa QS-controlled virulence factors production was analyzed using Graph-Pad Prism 5 software package with One Way ANOVA according to Dunnet's or Tukey's Multiple Comparison Tests < 0.05 or P < 0.001 for significance. Results were calculated as the means ± standard errors of three biological experiments with three technical replicates each.

Identification of clinical isolates
Ps. aeruginosa clinical isolates were identified from the following characters: they were Gram-negative rods and grew as non-lactose fermenters on MacConkey agar, they grew on cetrimide agar at 42°C and showed green pigmentation on nutrient agar, they were motile and oxidase positive.

Antibiotic susceptibility and resistance pattern of clinical isolates
The tested clinical isolates of P s. aeruginos ashowed high resistance against the various antibiotics used in this study. They all were found to be multi-drug resistant (MDR). The full results of antibiotic susceptibility test are illustrated in Table 2.

Antibacterial activity of ZnO and growth inhibition assay
ZnO prevented the growth of tested Ps. aeruginosa isolates at a concentration of 8 mg/ml and ¼ MIC (2mg/ml) was selected to test the effect of ZnO against QS-controlled virulence factors.
The efficacy of ZnO on QS-controlled virulence factors could be due to its effect on the growth of Pseudomonas isolates. To exclude this possibility, the effect of ¼ MIC of ZnO on bacterial growth was assessed by measuring the optical density of overnight culturesin LB broth at 600 nm and no statistically significant difference in the growth rate was found in the presence or absence of ZnO (Fig. 1).

ZnO decreased pathogenesis of Ps. aeruginosa in vivo
In mice survival test shown in (Fig. 7), mice injected with positive control (untreated) Ps. aeruginosa began to die after 24 h and only 20% of mice in this group were still alive at the end of the experiment. Importantly, at an infectious dose of approximately 2.5 x 10 7 CFU, mice injected with bacteria treated with ZnO showed a significantly higher survival rate as compared to positive control bacteria as all mice in this group remained alive at the end of the experiment; a result similar to that of the negative control groups in which no mice in these groups died at the end of the experiment. Our resultssuggest a protective role of ZnO nanoparticles against Ps.aeruginosa pathogenesis and virulence in mice.

Discussion
Ps. aeruginosa has become a principal etiologic agent of nosocomial infections which necessitates the urgent and efficient application of proper infection control policies to combat its spread. It shows both natural resistance and acquired multi-drug resistance to antimicrobial agents bydifferentmechanisms 21 . As an alternative strategy to antimicrobial therapy, targeting of QS has become an attractive option due to its involvement in regulating virulence factors production and pathogenesis in Ps. aeruginosa 22 .
Anti-virulence therapy (as QS inhibitors) would be valuable in the management of microbial diseases due to the lack of pressures affecting bacterial growth upon using anti-virulence agents and so microbial resistance against them is not expected 18 .
In the present study, ZnO inhibited the growth of all the tested Pseudomonas isolates at 8mg/ml. Testing the effect of ¼ MIC of ZnO nanoparticles(2mg/ml) on microbial growth revealed the absence of statistically signifi-  cant difference between both ZnO-treated and untreated cultures. As a result, any possible effect on QS is not due to adverse impact on bacterial viability and growth. Instead, it may be attributed to disrupting essential bacterial functions.
Ps. aeruginosa produces an arsenal of virulence factors including pyocyanin, pyoverdin, proteases, elastase and rhamnolipids which take partinestablishing infections and making the dissemination and invasion of hosttissues easier 23 . Recently, nanotechnology has been used to develop new nanoparticles that can target QS and virulence factors [24][25] .
In the current study, ZnO nanoparticles at ¼ MIC showed a potent inhibitory effect on the production ofrhamnolipids,pyocyanin, pyoverdin, hemolysins,elastase and proteases. In accordance with our data,Singh et al 26 reported that another type of metal nanoparticles, silver nanoparticles, at sub-MIC exhibited a remarkable reduction in production of proteases, elastase, pyocyanin and rhamnolipids. LasI/R and rhlI/R are two principle QS systems that regulate virulence genes in Ps. aeruginosa. LasI andrhlI synthases are responsible for the production of C12-AHL and C4-AHL autoinducers, respectively. At a threshold concentration of autoinducers, C12-AHL binds with lasR and induces the expression of genes control production of elastase, exotoxin and proteases and also activates the rhlI/R system. In addition, C4-AHL binds with rhlR andplays role in controlling the expression of genes encoding production of elastase, and pyocyanin. IflasI/R and rhlI/R are interrupted, virulence factors will be inhibited [27][28] .
To further explore the potential quorum quenching effect of ZnO nanoparticles, relative expression of QS-regulatory genes that controlvirulence factors in Ps. aeruginosa was examined using qRT-PCR.Importantly, ZnO nanoparticles significantly down-regulate the relative expression of QS regulatory genes, lasI, lasR, rhlI, rhlR, pqsA and pqsRwhich confirm the phenotypic results. Similarly, it was proved by Singh et al 26 that silver nanoparticles down-regulated the expression of lasI, lasR, rhlI and rhlR at the molecular level by inhibiting lasR and rhlR. ZnO nanoparticles effect might be similar to that of silver nanoparticles by inhibiting both lasR and rhlR resulting in disruption of QS circuits with subsequent inhibition of virulence factors production. The molecular basis and full mechanism of nanoparticles impact as quorum sensing inhibitor need more investigations in the future. For more confirmation, the effect of ZnO nanoparticles on Ps. aeruginosa pathogenesis was determined in vivo. Interestingly, a significant increase in the survival rate of mice injected with ZnO-treated isolates was found in comparison to those injected with untreated ones. Previous reports also showed the protective effect of QSinhibitors against bacterial pathogenesisin mice injected with Ps. aeruginosa. This was reported for gingerol 19 and polyphenolic compounds of honey 29 . Our collective phenotypic, genotypic and in vivo results strongly potentiate the potential use of ZnO nanoparticles as a powerful QS inhibitor in Ps. aeruginosa.

Conclusion
The spread of MDR strains of Ps. aeruginosa is of a particular concern causing healthcare-associated infections and increasing the challenge both in the clinical treatment of patients and in the prevention of the cross-transmission of this problematic pathogen. QS controls the production of many of virulence factors in Ps. aeruginosa and plays an essential role in antimicrobial resistance. ZnO nanoparticles is a promising QS inhibitor and anti-virulence compound that can be used as an adjunct for the treatment of Ps. aeruginosa infections such as burns and surgical wound infectionsmainly those caused by MDR isolates.