3-Benzyl-Hexahydro-Pyrrolo[1,2-a]Pyrazine-1,4-Dione Extracted From Exiguobacterium indicum Showed Anti-biofilm Activity Against Pseudomonas aeruginosa by Attenuating Quorum Sensing

Bacterial cell-to-cell communication promotes biofilm formation and can potentially lead to multidrug resistance development. Quorum sensing inhibition (QSI) is an effective and widely employed strategy against biofilm formation. The extract from Exiguobacterium indicum SJ16, a gram-positive bacterium, isolated from the rhizosphere of Cyperus laevigatus showed significant anti-quorum sensing activity (about 99%) against the reference Chromobacterium violaceum CV026 strain without exerting any antibacterial effect. The potentially active QSI compound identified in the SJ16 extract was 3-Benzyl-hexahydro-pyrrolo[1, 2-a]pyrazine-1,4-dione. The SJ16 extract containing this active compound showed significant anti-quorum sensing activity against a model quorum sensing bacterium strain Pseudomonas aeruginosa PAO1 and a clinical isolate P. aeruginosa PAH by preventing biofilm formation without attenuating the cell growth within the biofilm. More specifically, the SJ16 extract changed the topography and architecture of the biofilm, thus preventing bacterial adherence and further development of the biofilm. Furthermore, it decreased virulence factors (rhamnolipid and pyocyanin), the bacterial motility, as well as the elastase, and protease activities in P. aeruginosa. Microarray analysis revealed the differential expression of quorum sensing regulatory genes. Based on these results, we herein propose a hypothetical model, characterizing the role of this QSI agent in the transcriptional regulation of quorum sensing in P. aeruginosa PAO1, demonstrating that this compound has significant drug-development potential. Further research is required to delineate its possible applications in therapeutics in the context of biofilm forming bacterial infections.


INTRODUCTION
The diseases caused by pathogenic bacteria are controlled, prevented, and treated with a number of antibiotics which inhibit essential bacterial processes, such as cell wall synthesis, DNA replication, or protein biosynthesis. Antibiotics have long become the commonplace in our effort to tackle infectious diseases (Chu et al., 2014). However, extensive use of these agents creates an evolutionary pressure on bacteria which, in many cases, leads to antibiotics resistance. Therefore, alternative strategies with low potential for resistance emergence are required in order to combat pathogenic multidrug resistant bacteria (Chu et al., 2014;Singh et al., 2016b). Quorum sensing (QS) is directly involved in pathogenesis of infectious disease, by regulating the biofilm formation, the production of multiple virulence factors, as well as the motility of bacteria (Singh et al., 2017). In QS system, molecular cascades regulate the gene expression and determine the fate of bacterial biofilms (Ganin et al., 2015). In this context, gram-negative bacteria communicate through small diffusible molecules, such as acyl homoserine lactones, whereas gram-positive bacteria use autoinducer peptides for their communication (Galloway et al., 2011).
Biofilms are comprised of single or multiple microbial species and develop on different biotic and abiotic surfaces. In most of the cases, mixed species biofilms are predominant. However, single species biofilms are commonly associated with medical equipment-related infections, especially dental plaque and medical implants (O'Toole et al., 2000). Pseudomonas aeruginosa, a gram-negative bacterium, is considered to be the most important species of biofilm-forming bacteria. It can develop biofilms on a variety of biotic and abiotic surfaces; especially in immunocompromised patients (Driscoll et al., 2007). Moreover, this bacterium shows resistance to most of the conventional antibiotics, because it can form a biofilm matrix, which protects bacterial cells. Therefore, eradicating such infections poses a significant challenge (Driscoll et al., 2007;Lee and Zhang, 2015). QS regulatory networks control the virulence factors and biofilm formation in P. aeruginosa (Lee and Zhang, 2015). Therefore, utilization of anti-QS strategies could prove to be a promising way to tackle P. aeruginosa infections.
The rhizosphere is a region of soil adjacent to the plant roots that inhabits a number of microbes and facilitates various plant-microbe and microbe-microbe interactions. Many rhizospheric bacteria prevent the development of soilborne pathogens, while at the same time they protect the associated plants by activating the induced systemic resistance (Berendsen et al., 2012). Importantly, the rhizosphere supports different bacterial communities that exert QS and quorum quenching activities. Christiaen et al. (2011) isolated 59 bacterial species from 16 different environmental samples including plant rhizospheres and water, and found that 41 of them had anti-QS properties. Furthermore, Stenotrophomonas maltophilia and Delftia tsuruhatensis, isolated from the rhizosphere of Cyperus laevigatus showed anti-QS and anti-biofilm activities against P. aeruginosa (Singh et al., 2013(Singh et al., , 2017. Moreover, a bacterium from the family of the Acinetobacter spp., isolated from the cucumber rhizosphere, inhibited the growth of Pseudomonas chlororaphis and Burkholderia glumae during co-cultivation by degrading acyl-homoserine lactones (AHLs), produced by these bacteria (Kang et al., 2004). Other AHL producing and degrading bacteria were isolated from the tobacco rhizosphere, including two newly identified species, Sphingopyxis witflariensis and Bosea thiooxidans, belong to the Bacillus α-proteobacteria family (D' Angelo-Picard et al., 2005).
A three bacterial consortium comprised of Acinetobacter sp., Burkholderia sp. and Klebsiella sp. was isolated from the ginger rhizosphere, demonstrating significant AHL degrading activities and growth-inhibiting capabilities against P. aeruginosa and the plant pathogen Erwinia carotovora without affecting their planktonic growth (Chan et al., 2011). Last but not least, the AHL degrading bacterial consortia isolated from the potato rhizosphere showed significant biocontrol activity against Pectobacterium atrosepticum (Cirou et al., 2007). In most of these earlier studies, the above mentioned bacteria were isolated and characterized for their anti-QS properties; however, to the best of our knowledge there is no report, so far, on the isolation and characterization of the active compounds from these bacteria.
In the present study, a bacterium Exiguobacterium indicum SJ16 was isolated from the rhizosphere of a monocot C. laevigatus, amply growing in the coastal saline area and was tested for its anti-QS and anti-biofilm potential. The active fraction was collected, identified, and regulatory key genes were studied to elucidate a possible QSI mechanism.

Isolation and Molecular Identification of Bacteria
Bacteria from the rhizosphere of C. laevigatus L. from the natural habitat of Bhavnagar, India (Latitude N 21 • 45.124", Longitude E 72 • 13.579") were isolated. Bacteria were screened by plate based bioassay, and two were found positive for anti-QS activity (Singh et al., 2017). The bacterial isolate SJ16, which showed anti-QS (but not antibacterial) activity was selected for further characterization. Genomic DNA of the bacterium was isolated, and the 16S rRNA gene was amplified by universal primers fD1-5 -AGA GTT TGA TCC TGG CTC AG -3 and rP2-5 -ACG GCT ACC TTG TTA CGA CTT -3 using optimized PCR conditions (Weisburg et al., 1991). The amplified PCR product was purified, sequenced (at Macrogen Inc., South Korea), analyzed and submitted to the NCBI GenBank. The phylogenetic analysis was performed using MEGA version 6.0 (Tamura et al., 2013), and a phylogenetic tree was inferred by neighbor-joining methods (Saitou and Nei, 1987). Bootstrap analysis was performed, and maximum composite likelihood algorithms were used for the determination of the evolutionary distances (Felsenstein, 1985;Tamura et al., 2004).

Fatty Acid Methyl Ester Profiling of Bacteria
Fatty acid methyl ester (FAME) profiling of E. indicum SJ16 was carried out using Microbial Identification System coupled with gas chromatography (MIDI, Microbial ID; GC system-6850, Agilent Technologies, United States). The bacterium was grown for 24 h at 30 • C on Tryptic soy yeast agar, FAMEs were prepared using MIDI manual, and peaks were identified with RTSBA6 6.10 database (Jha et al., 2015).

Preparation of the Bacterial Extract, Fractionation, and Identification of the Active Compound
Bacterial culture (E. indicum SJ16) was grown for 48 h at 30 • C in 500 ml of nutrient broth, kept in an incubator shaker with agitation at180 rpm. The culture was centrifuged for 15 min at 10,000 × g (4 • C), the supernatant was collected, and filtered through 0.22 µm filter for the complete removal of remaining bacterial cells. The supernatant was extracted with ethyl acetate (equal volume), evaporated to dryness under vacuum and the dried residue finally dissolved in methanol.
The methanol extract of SJ16 was further fractionated by solid phase extraction method using different cartridges including anion exchanger DAE, cation mixed Plexa PCX, polar SI, and non-polar C18 (Agilent, United States). Each fraction was tested for quorum sensing inhibition (QSI) activity (Singh et al., 2017). The positive fraction (collected through the C18 cartridge with 40% methanol) showing a maximum zone of QSI was used for the subsequent studies and was also subjected to GC-MS (GC-2010, Shimadzu, Japan) analysis. The probable active compound was identified by comparing the mass spectra with the reference spectra library. The mass of the probable active compound was further confirmed by electrospray ionization mass spectrometry (ESI-MS; Q-Tof micro TM, Micromass, United Kingdom).

Anti-quorum Sensing and Anti-biofilm Activities
The anti-QS activity was tested (i) by a plate-based bioassay using Chromobacterium violaceum CV026 as a tester strain, methanol as a negative control and cinnamaldehyde (Sigma-Aldrich, United States) as a positive control, and (ii) by quantifying the violacein production. Both the plate-based bioassay and the violacein quantification, were performed as per our previously optimized methods (Singh et al., 2013).
The anti-biofilm formation assay was performed with different concentrations of bacterial (strain SJ16) extracts (0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 mg/ml) using two P. aeruginosa strains, PAO1 as well as the PAH clinical isolate (provided as a courtesy from the Government Medical College, Bhavnagar; Goswami et al., 2011). Briefly, 200 µl bacterial culture (OD 600 nm = 0.1) dilute from overnight grown PAO1 and PAH cultures were added to a 96-well polystyrene microtiter plate with different concentrations of the SJ16 extracts. The plates were allowed to grow for 24 h (at 37 • C with 100 rpm), and growth was measured at 600 nm. The biofilm formation was assayed using our previously optimized crystal violet staining method (Kavita et al., 2014). The experiments were performed three times with five replicates each time.

Fluorescence Microscopy
Pseudomonas aeruginosa (strains PAO1 and PAH) were grown in a 24 well polystyrene plate with or without SJ16 extract to assess the development of biofilm on a sterilized glass coverslip (11 mm). In a 24-well polystyrene plate 1 ml of NB media containing bacterial culture at OD 600 nm = 0.1 (diluted from overnight grown culture) was added to each well with or without (control) 1.0 mg/ml of SJ16 extract. One glass coverslip/well was submerged and the plate was incubated in static condition at 37 • C. The effect of the SJ16 extract (1.0 mg/ml) on the P. aeruginosa cell viability within biofilm was examined at different time points (24, 48, and 72 h) under a fluorescence microscope using FilmTracer LIVE/DEAD Biofilm Viability Kit (Invitrogen, United States) (Singh et al., 2013). The bacterial cells within the biofilm were labeled with a fluorescent dye (propidium iodide and SYTO 9), were further processed according to the manufacturer's instructions and were visualized under an epifluorescence microscope (Axio Imager, Carl Zeiss AG, Germany).

Scanning Electron and Atomic Force Microscopy
Scanning electron microscopy (SEM) and atomic force microscopy (AFM) were used to visualize the effect of the SJ16 extract on the topology of the biofilm developed by the P. aeruginosa strains. For the SEM, a previously described protocol was used (Andersson et al., 2009;Singh et al., 2013). In brief, biofilms of P. aeruginosa PAO1 and P. aeruginosa PAH were grown on glass coverslips submerged in nutrient broth in a 24-well polystyrene plate with (1.0 mg/ml) or without extract (control). The plate containing the culture and the coverslips were kept at 37 • C for 24 h. After incubation, the planktonic culture was removed and coverslips were gently washed with 0.9% NaCl. The samples were treated with 2.5% glutaraldehyde for 20 min followed by 4% OsO 4 for 30 min and dehydrated using ethanol gradient (10 to 95%) treatment for 10 min for each concentration. The dehydrated and dried biofilms were coated with gold and observed under a scanning electron microscope (SEM, LEO series VP1430, Germany). For the AFM, overnight grown cultures of P. aeruginosa PAO1 and P. aeruginosa PAH were diluted to reach an OD 600 nm = 0.1 in NB broth, sterile glass cover slips were submerged in 24-well polystyrene plate with (1.0 mg/ml) or without extract (control). The plate was kept for 24 h at 37 • C. Following this incubation period, the biofilm that developed on the glass coverslip was rinsed gently with phosphate buffer saline (pH 7.4) and was kept under the desiccators until completely dry. Finally, the biofilm was scanned under AFM (NT-MDT, Russia) in a semi-contact mode (Oh et al., 2009;Singh et al., 2013).

Swarming Motility Assay
The swarming motility of PAO1 and PAH were tested in the presence (1.0 mg/ml) and absence of SJ16 extract. Overnight grown culture of P. aeruginosa PAO1 and P. aeruginosa PAH were diluted to OD 600 nm = 1.0 and spotted on a plate containing BM2 swarming medium (62 mM PBS at pH 7, 2 mM MgSO 4 , 10 µM FeSO 4 , 0.4% glucose, 0.1% casamino acids and 0.5% agar) supplemented with 1.0 mg/ml SJ16 extract or without extract supplementation (control) (Overhage et al., 2007). The plates were incubated at 37 • C for 24 h and swarming zones were observed.

Swimming Motility Assay
The overnight grown culture of PAO1 and PAH were diluted to OD 600 nm = 1.0 and spotted on a plate containing tryptone broth (10 g/l tryptone, 5 g/l NaCl and 0.3% agar) supplemented with 1.0 mg/ml SJ16 extract or without extract supplementation (control) (Rashid and Kornberg, 2000). The plates were incubated at 37 • C and analyzed after 24 h.

Virulence Factor Analysis
The effect of the SJ16 bacterial extracts (1.0 mg/ml) on the production of virulence factors by reference P. aeruginosa strains (PAO1 and PAH) was studied by measuring the levels of pyocyanin and rhamnolipid, and by analyzing the elastase and protease activities. For pyocyanin assay, starter cultures of P. aeruginosa strains (PAO1 and PAH) were grown at 37 • C in an incubator shaker until OD 600 nm = 3.0 and diluted to OD 600 nm = 0.1 in PB medium (5 ml; 20 g/l peptone, 1.4 g/l MgCl 2, and 10 g/l K 2 SO 4 ). The tubes containing 5 ml PB medium (OD 600 nm = 0.1) supplemented with 1.0 mg/ml SJ16 extract or without extract supplementation (control). Supernatants were collected by centrifuging cultures at 10,000 × g for 10 min; pyocyanin was extracted in 3 ml of chloroform followed by 1 ml of 0.2 N HCl and quantified spectrophotometrically at 520 nm (Essar et al., 1990). For rhamnolipid assay, P. aeruginosa strains PAO1 and PAH (OD 600 nm = 0.1) were grown overnight at 37 • C in NB medium, supplemented with 1.0 mg/ml SJ16 extract or without extract supplementation (control). The culture was centrifuged (10,000 × g for 10 min), the supernatant was collected and acidified to pH 2 (with HCl), and absorbance was measured at 570 nm (McClure and Schiller, 1992;Sarabhai et al., 2013).

Protease Assay
For the protease activity, 2% azocasein solution was prepared in 50 mM phosphate buffer saline (pH 7). The supernatant (400 µl) from overnight grown (with or without SJ16 extract) cultures of P. aeruginosa (PAO1 and PAH) were incubated with an equal volume of azocasein solution (2%) at 37 • C for 1 h. A measure of 500 µl of 10% trichloroacetic acid was added to stop the reaction. Reaction-mixtures were centrifuged (8,000 × g for 5 min) to remove residual azocasein and the absorbance of the supernatant was measured at 400 nm (Adonizio et al., 2008).

Microarray
A single colony of the reference strain P. aeruginosa PAO1 was inoculated in a 5 ml NB tubes, was grown until OD 600 nm 3.0 and was then diluted to OD 600 nm = 0.1. Following dilution, P. aeruginosa PAO1 was grown in a tube containing 5 ml starter culture (OD 600 nm = 0.1) with or without bacterial extracts (1.0 mg/ml) at 37 • C for 24 h in a shaker incubator at 200 rpm (Singh et al., 2017). Planktonic cells were harvested by centrifuge at 12,000 × g for 5 min and total RNA was isolated using TRI reagent (Sigma, United States). Total RNA was analyzed on 2% agoras gel and quantified using ND-1000 spectrophotometer (Nanodrop Technologies, United States). Ten µg RNA, extracted from control and treated samples were converted to cDNA, fragmented and labeled using previously optimized method (Singh et al., 2016a). Labeled cDNAs were hybridized with genome array gene chip (Gene Chip P. aeruginosa PAO1 containing total 5,886 gene probes), washed, stained, and scanned (Scanner 3000 7G, Affymetrix, United States) (Singh et al., 2017). Scanned chips were processed and analyzed using the expression console and the transcriptome analysis console (Affymetrix, United States). Microarray analysis was performed in duplicate and the genes showing significant differences in fold-change expression (ANOVA p-value < 0.05) were considered for the study.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism software. One-way ANOVA followed by Tukey post hoc that was applied for the comparison of test samples and controls.

Identification of the Active Fraction/Compound
All fractions, collected from the various SPE cartridges (anion exchanger DAE, cation mixed Plexa PCX, polar SI, and non-polar C18) were individually screened for anti-QS activity using a biosensor plate containing the tester strain C. violaceum CV026. The fraction collected from the C18 cartridge (with 40% methanol) showed a maximum zone of QSI. This fraction was subjected to GC-MS analysis, and the chromatogram showed a single predominant peak at the retention time 16.01 min (Figure 1). The detected peak resembled the 3-Benzyl-hexahydro-pyrrolo[1, 2-a]pyrazine-1,4dione (C 14 H 16 N 2 O 2 ; expected molecular mass 244.28) from the GC-MS library. The active fraction was also subjected to ESI-MS analysis, which showed a dominating spectral peak at m/z 249.29 (Figure 1). Thus, the experimental molecular mass of the active compound was determined to be 249.29, which is corresponding to the theoretical mass of the active fraction.
Exiguobacterium indicum SJ16 Showed Anti-quorum Sensing and Anti-biofilm Activities Without Inhibiting the Planktonic Growth A clear white opaque zone was observed with the axenic culture and crude extract of the E. indicum strain SJ16 in the biosensor plate containing the reference strain C. violaceum CV026, result that was comparative to the positive control with cinnamaldehyde, a well-known QS inhibitor (Figure 2). The zone of inhibition was not detected with the extraction solvent methanol that was used as a negative control. As inhibition of violacein production is an indicator of QS activity, different concentrations (0.1-0.6 mg/ml) of the SJ16 extract were used to quantify the levels of this QSI marker. Indeed, the violacein production decreased concomitantly with increasing concentrations of the extract, and about 99% inhibition was achieved with 0.6 mg/ml of the SJ16 extract (Figure 2). Furthermore, the antibacterial assay, performed by using the disc diffusion technique, confirmed that the bacterial isolate SJ16 did not halt the growth or the viability of the reference strain C. violaceum CV026 (Supplementary Figure S3).
The anti-biofilm activity of E. indicum SJ16 extract was performed with a different concentration (0.2-1.2 mg/ml) against PAO1 and PAH strains. The biofilm formation capability of both strains significantly decreased with increasing concentrations of the extract (Figure 3). About 50% inhibition of biofilm formation was noticed with 1.0 mg/ml extract. No significant effect of this compound was observed on the planktonic growth of either PAO1 or PAH even with highest extract concentrations used (0.2-1.2 mg/ml) (Figure 3).
The viability of the reference strains (P. aeruginosa PAO1 and PAH) within the biofilm was further analyzed with the SJ16 extract at 24, 48, and 72 h. The live cells of P. aeruginosa were labeled with SYTO 9, whereas dead cells were stained with propidium iodide using a Live/Dead staining kit. Live and dead cells produced green and red fluorescence, respectively, when visualized under an epi-fluorescence microscope ( Figure 4A)   1-0.6 mg/ml) were used to quantify the inhibition of violacein production, an indicator of QS activity. Culture without extract was considered as a control. " * * " and " * * * " Indicates significant differences from the control at P < 0.01 and P < 0.001, respectively. extract were not tightly attached to the biofilm surface compared to the control. There was no difference between the control and treated biofilm when considering the cell viability whithin biofilm, and insignificant numbers of dead cells were detected in the biofilm at different time points ( Figure 4A). Overall, these results indicate that the SJ16 extract does not exert any toxic effect on the growth or viability of the aforementioned reference strains, even after a longer duration, or higher concentrations of the compound, while at the same time it effectively inhibits the bacterial biofilm formation (Figure 4).

Exiguobacterium indicum SJ16 Extract Disrupts the Topology of the Biofilm
The effect of the SJ16 extract on the topology of the biofilm developed by the PAO1 and PAH was visualized under SEM and AFM. A biofilm along with adhering bacterial cells was developed by P. aeruginosa in the absence of SJ16 extract (control), whereas an immature biofilm with dispersed bacterial cells was observed when P. aeruginosa was grown with SJ16 extract (Figure 4B).
Similarly, AFM micrographs showed changes in the topology of the biofilm developed by P. aeruginosa in the presence of SJ16 extract compared to normal (control) biofilm ( Figure 4C). AFM showed poor biofilm adherence on the surface of the glass cover slip, while height distribution profile showed the average thickness of the biofilm developed in the presence of SJ16 extract was significantly reduced compared to the control biofilm ( Figure 4C).

Exiguobacterium indicum SJ16 Extract
Downregulates the Motility of P. aeruginosa Bacterial motility and initial attachment of bacterial cells to surfaces are key prerequisites for biofilm formation. Therefore, the effect of the E. indicum SJ16 extract on the motility of reference P. aeruginosa strains (PAO1 and PAH) was studied. Our results show that the swarming and swimming motility of PAO1 and PAH are significantly inhibited in the presence of the SJ16 extract ( Figure 5A).

Exiguobacterium indicum SJ16 Extract Shows Inhibitory Effect on Virulence Activities
Virulence activities of the P. aeruginosa strains (PAO1 and PAH), including elastase and protease activities, as well as the production of virulence factors (pyocyanin and rhamnolipid) accelerate biofilm formation. Our data indicate that the SJ16 extract considerably reduced the production of virulence factors, with pyocyanin production decreasing about 50% in both PAO1 and PAH compared to the untreated samples ( Figure 5B). Similarly, rhamnolipid production was also significantly decreased by 55% in PAO1 and 37% in PAH in the presence of the SJ16 extract compared to control ( Figure 5B). Furthermore, the SJ16 extract led to a drastic inhibition of the elastase and protease activities in the above-mentioned P. aeruginosa strains. Specifically, elastase activity was decreased by 62% in PAO1 and by 31% PAH compared to untreated samples ( Figure 5B). Similarly, protease activity was reduced by about 40% in PAO1 and by 28% in PAH in the presence of the SJ16 extract compared to control ( Figure 5B). Overall, our results (Figures 3-5) strongly demonstrate that the SJ16 extract inhibits the cell-to-cell communication and thus prevents P. aeruginosa from developing a robust biofilm.

Differential Expression of Quorum Sensing Regulatory Genes
A P. aeruginosa PAO1 genome array chip, containing 5,886 gene-probe sets, was used to study the effect of 3-Benzylhexahydro-pyrrolo[1, 2-a]pyrazine-1,4-dione on QS regulatory genes. Scattered plot analysis showed that 1,237 out of 5,886 genes (Array-Express accession E-MTAB-6829) were differentially expressed (Figure 6), with a two-fold (p < 0.05) up-(>2) or down-(<−2) regulation at minimum (Supplementary Table S1). More specifically, a total of 868 genes were down-regulated, whereas 369 genes were up-regulated in the presence of the compound. Among these, we identified a plethora of genes

D C B A
FIGURE 3 | Effect of E. indicum SJ16 extract on biofilm formation and planktonic growth of P. aeruginosa. Different concentration of bacterial extracts (0.2-1.2 mg/ml) ware tested against the biofilm forming reference P. aeruginosa strains PAO1 (A) and PAH (B). The same concentrations of bacterial extracts were also tested for the inhibitory effect on the planktonic growth of PAO1 (C) and PAH (D) strains. Tests without extract and with methanol were considered as control and negative control, respectively. " * ", " * * " and " * * * " Indicates significant differences from the control at P < 0.05, P < 0.01, and P < 0.001, respectively. that are key players in bacterial QS network of P. aeruginosa and exert crucial functions in general metabolic pathways in bacteria (Table 1).

DISCUSSION
The genus Exiguobacterium is comprised of 16 species, and E. indicum was isolated from meltwater of the Hamta glacier (Himalayan mountain ranges) of India (Chaturvedi and Shivaji, 2006). The present study is the second report on the isolation of E. indicum from a rhizosphere of the coastal area after Susilowati et al. (2015), who isolated this bacteria from rice rhizosphere and reported its plant growth promotion trait. To the best of our knowledge, there is no available study, so far reporting the anti-QS activity of the genus Exiguobacterium. We herein for the first time demonstrate the inhibitory effect of E. indicum on bacterial quorum-sensing biofilm formation. In this study, the E. indicum SJ16 extract showed QSI activity (without any antibacterial activity; Supplementary Figure S3) against the reference strain C. violaceum CV026 on biosensor plates. This was also demonstrated by reduced levels of violacein in the violacein quantification assay (99% inhibition with 0.6 mg/ml SJ16 extract), which was a dose-dependent (Figure 2). A high concentration of methanolic extract (4 mg/ml) of an edible plant, Melicope lunuankenda, inhibited the response of C. violaceum CV026 to N-hexanoyl homoserine lactone (Tan et al., 2012). Similarly, Rodrigues et al. (2016) have shown that another plant methanol extract (Eugenia uniflora) showed inhibition of violacein production in C. violaceum, reaching up to 96% at the highest concentration tested. We have previously demonstrated a 98% decrease of violacein production in a dosedependent manner with 3.7 mg/ml extract of S. maltophilia, as well as a 95% reduction with 0.1 mg/ml of D. tsuruhatensis extract, both isolated from the rhizosphere of C. laevigatus (Singh et al., 2013(Singh et al., , 2017.
The treatment of biofilm-associated infections is extremely challenging, as biofilm-forming bacteria are commonly resistant to a broad spectrum of antibiotics (Høiby et al., 2010). P. aeruginosa, an opportunistic pathogen that causes severe infections in immunocompromised patients possesses a regulatory gene cascade that controls quorum-sensing and thus, biofilm formation and virulence factors productions. Quorum-sensing inhibition with potent QSI compounds shows promise in tackling biofilm formation in the setting of bacterial infections (Kalia, 2013;Singh et al., 2017;Quecán et al., 2018). In the field of drug discovery, extracts from natural products, especially those of microbial origin, have always been a tremendous source effective novel therapeutics (Gillespie et al., 2002;Courtois et al., 2003). These compounds have the ability to interfere with the QS system, inhibit the expression of virulence factors and prevent biofilm formation. In the present study, the E. indicum SJ16 extract robustly inhibits the biofilm formation of two reference P. aeruginosa strains (PAO1 and PAH) by modulating virulence factors ( Figure 5), without affecting the planktonic growth of these bacteria (Figure 3). It is possible that these observations are the effect of the active compound within the extract that hinder the P. aeruginosa QS regulatory cascade (Table 1). These results are in accordance to our previous work describing the quorum quenching and anti-biofilm forming activities of the extracts from S. maltophilia BJ01 and D. tsuruhatensis SJ01 that we isolated from the rhizosphere of C. laevigatus (Singh et al., 2013(Singh et al., , 2017. In this study, the active compound that we herein describe was fractionated and identified as 3-Benzyl-hexahydropyrrolo[1,2-a]pyrazine-1,4-dione with a theoretical mass of 244.28 (RT-16.018) and an experimental mass of 249.28 (m/z) (Figure 1). Interestingly, a similar to this molecule, Pyrrolo[1,2-a]pyrazine-1,4-dione, hexahydro compound that was extracted from the newly identified species of Streptomyces mangrovisoli showed significant antioxidant and free-radical scavenging activities (Ser et al., 2015). Furthermore, the Hexahydropyrrolo[1,2-a]pyrazine-1,4-dione compound, isolated from the Shewanella sp. Lzh-2 possessed algicidal activity against several cyanobacterial and algal strains (Li et al., 2014). Moreover, The effect of the bacterial extract on the swarming and swimming motility of reference P. aeruginosa strains (PAO1 and PAH). P. aeruginosa was spotted on a plate supplemented with 1.0 mg/ml extract or without extract supplementation (control). Plates were analyzed after incubation of 24 h at 37 • C; (B) The effect of the bacterial extracts (SJ16; 1.0 mg/ml) was determined based on the production of virulence factors of the reference P. aeruginosa strains by quantifying pyocyanin and rhamnolipid, as well as by analyzing the elastase and protease activities. " * * * " Indicates significant differences from the control at P < 0.001.
As stated above, the SJ16 extract inhibits the biofilm formation of the reference P. aeruginosa strains PAO1 and PAH in a concentration-dependent manner without affecting their planktonic growth (Figure 3). Furthermore, epi-fluorescence microscopy confirmed the viability of P. aeruginosa cells within the biofilm ( Figure 4A). Thus, the possibility of an inhibitory effect of the SJ16 extract on the growth of these reference strains (PAO1 and PAH) was ruled out. The SEM and AFM images suggest an alteration in the topology of the biofilm in the samples supplemented with the extract, compared to their corresponding controls, while a growing biofilm with dispersed bacterial cells was developed by P. aeruginosa strains in the presence of the SJ16 extract (Figures 4B,C). Disruption of biofilm architecture is a promising strategy to inhibit biofilm formation of drug-resistant P. aeruginosa strains. Importantly, in order for the bacteria to form a biofilm, they need to attach to a surface or substratum. Bacterial motility is crucial in the effort of the microbes to reach the substratum. Once attached to the surface, the bacteria spread all around via swarming and swimming type of motility, ultimately leading to biofilm formation over the surface (O'May and Tufenkji, 2011). Inhibition of the swarming and swimming motility of PAO1 and PAH was observed in the presence of the SJ16 extract ( Figure 5A). It is possible that the SJ16 extract may also have the capability to block the initial attachment of P. aeruginosa by preventing the bacterial motility toward the surface. This strategy could potentially open new avenue in the effort to halt bacterial spreading and thus minimize the ability of microbes to form biofilms (Singh et al., 2017). The P. aeruginosa pathogenicity depends on the ability of this microbe to produce virulence factors. Pyocyanin, rhamnolipids, elastase, and protease are the key virulence factors which are highly expressed by P. aeruginosa during QS, infection and biofilm formation (Driscoll et al., 2007;Sarabhai et al., 2013). Elastase and protease are involved in the early steps of host colonization by the bacterial cells, pyocyanin is crucial for the demonstration of P. aeruginosa virulence, while rhamnolipids facilitate the bacterial motility. The combined effect of the afore-mentioned microbial factors eventually lead to biofilm formation (O'May and Tufenkji, 2011;Sarabhai et al., 2013). Elastase and protease activities, as well as the production of pyocyanin and rhamnolipid of both P. aeruginosa strains PAO1 and PAH were significantly reduced following SJ16 extract supplementation (Figure 5B), demonstrating the ability of the extract to attenuate the P. aeruginosa virulence functions, largely regulated by the las and rhl regulatory gene cascades (De Kievit and Iglewski, 2000). Similar results were obtained with the extract of Terminalia chebula and D. tsuruhatensis (Sarabhai et al., 2013, Singh et al., 2017. The microarray analysis revealed that at least 1,237 genes were differentially expressed following SJ16 extract supplementation, 868 out of which were down-regulated, whereas 369 were up-regulated (Supplementary Table S1 and Figure 6). These microarray results demonstrated that a plethora of genes affected by the SJ16 extract are involved in QS regulation and biofilm formation. More specifically, these genes are crucial in the control of bacterial QS, virulence, motility, cell metabolism, cell wall synthesis, and transcriptional regulation, while some others encode hypothetical proteins ( Table 1). The expression profile of cells treated with the SJ16 extract containing the active 3-Benzyl-hexahydro-pyrrolo[1,2-a]pyrazine-1,4-dione compound showed that this agent downregulates genes responsible for the synthesis of the flagellar protein, the type III export apparatus protein, the flagellar basal-body rod protein, the pyochelin biosynthetic protein, the phenazine-modifying enzyme, the type III export protein, the twitching motility protein, as well as some hypothetical proteins. The extract also represses the expression of gene(s) which are involved in the biosynthesis of transcriptional regulators/ activators of QS network. These genes are controlled by QS systems and are closely associated with the pathogenicity of P. aeruginosa (Wagner et al., 2003). Similar results were obtained when P. aeruginosa PAO1 cells were treated with 1,2benzenedicarboxylic acid, diisooctyl ester (Singh et al., 2017).
Based on the results that we herein report, we adopted a hypothetical model that elucidates the transcriptional regulation of P. aeruginosa virulence mediated by the by E. indicum SJ16 extract (Figure 7). We hypothesized that the identified active compound, 3-Benzyl-hexahydro-pyrrolo[1,2-a]pyrazine-1,4dione lowers the expression of key regulatory genes including, pqsA, pqsC, pscB, pstC, pqqA, and qor (Table 1), in addition to the differential expression of other genes (Supplementary Table S1). Alteration in the transcript expression of these genes leads to lower levels of HHQ. Thus, it is possible that 3-Benzylhexahydro-pyrrolo[1, 2-a]pyrazine-1,4-dione modulates the function of the pqs transcriptional regulatory system, hence controlling the production of MvfR (also known as PqsR). Inhibition of the MvfR system decreases the production of QS activators and signaling molecules, thus regulating the las and rhl transcription cascades. Subsequently, rhl leads lower levels of pyocyanin and rhamnolipid, decreased protease and elastase activities as well as reduced bacterial motility (Déziel et al., 2005). Our results indicate that the active compound present in the SJ16 extract decreases the virulence activity through modulating the pqs transcription regulation system. The proposed model is a schematic representation of the regulatory mechanism, illustrated based on the available literature. However, a comprehensive study is required to confirm the precise role of the identified compound, 3-Benzyl-hexahydro-pyrrolo[1,2-a]pyrazine-1,4-dione in the QSI regulation mechanism as well as its interacting partners.

CONCLUSION
The QS regulates virulence factor activities and biofilm formation, and disrupting the QS mechanism is an important strategy to inhibit pathogenicity of P. aeruginosa strains. Anti-QS compounds provides a useful tool in the effort to tackle infections caused by pathogenic bacteria. E. indicum SJ16, a microbe isolated from the rhizosphere of C. laevigatus showed promising anti-QS and anti-biofilm activities, without exhibiting any anti-bacterial properties. The 3-Benzyl-hexahydro-pyrrolo[1, 2-a]pyrazine-1,4-dione compound present in the SJ16 extract was identified as a potentially active agent inhibiting the biofilm formation of two references P. aeruginosa strains, PAO1 and PAH by decreasing their swimming and swarming motility and by regulating the production of virulence factors such as pyocyanin, rhamnolipid, elastase, and protease. Furthermore, it is possible that this compound controls the pqs QS system, thus regulating the bacterial QS mechanism as indicated by our proposed inhibitory model. Overall, our results indicate that the SJ16 extract did not have any toxic effect on the growth and viability of the reference P. aeruginosa strains (PAO1 and PAH), even after longer incubation periods. On the contrary, it exhibits a strong inhibitory effects on the microbial motility, on the production of virulence factors as well as on biofilm formation. Importantly, our data indicate that the SJ16 extract is able to disrupt the cellto-cell communication (QSI) by modulating a key component of the molecular cascade regulating the P. aeruginosa of QS systems (las, rhl, and pqs). Therefore, the identified compound has great potential for drug development in our efforts to enrich our antimicrobial armamentarium. Further research is necessary to explore and determine its pharmaceutical applications.