The X. oryzae pv. oryzae (Xoo) strain AS29 was isolated from susceptible rice cultivar BLB Pusa Basmati 1 (India) and further identified using 16S-rDNA gene sequence. The identified strain was submitted to GenBank database (Accession Number KX010418) and routinely grown on PME (Peptone Malt Extract) (HiMedia, India) plates (0.5% peptone, 3% malt extract, 1.5% bacto agar) at 26 ± 2°C.

The protocol for measuring biofilm production was carried out following the method described by Dunger et al. (2014). Overnight grown XooAS29 culture was inoculated into 2 mL PME broth with 1:1000 dilution so that OD₆₀₀ reached 0.6. To the 96-well U-plastic titre plate (SPL Life Sciences, Co., Ltd; South Korea), 1 μL culture was added to 99 μL PME broth in each well. For evaluating the consequence of THY oil, 100–1000 ppm were initially used to assess their effect on biofilm formation. The titre plate was incubated without shaking at 28°C for 10 h. The incubated cells were then stained with crystal violet (CV) dye for 20 min. The unbound dye was removed by rinsing twice with sterile distilled water (SDW). Finally, the well-bound dye was solubilized in 200 μL dimethyl sulfoxide (100% DMSO) and quantified spectrophotometrically by recording absorbance at 595 nm.

Since, 500 ppm oil was most efficient in the aforementioned method; the outcome of this concentration on biofilm (microscopic visualization) in contrast to XooAS29 alone (control) was measured using polyvinyl chloride (PVC) assay (Krzyściak et al., 2014). The XooAS29 culture (OD₆₀₀ = 0.6) was supplemented into 1 mL of PME broth consisting of 500 ppm THY oil for 14 h at 28°C. The PVC plate wells were washed twice with SDW to eliminate the buoyant planktonic cells after the incubation time. The bacterial cells adhering to the surface were stained with 0.5% CV and 0.2% methylene blue (MB) dyes, respectively and incubated for 5 min at room temperature. The surplus dye was eliminated by pipetting out and the plates were again kept for air drying at room temperature. The dried plates were further imaged at a magnification of 20X (Olympus BH2, Japan) under a light microscope. The XooAS29 culture was allowed to grow on the cover glass submerged in PME medium in the absence and presence of specific concentration of oil and incubated for 10–12 h. The biofilm formed were further stained with green fluorescent dye 20 mM SYTO-9 and observed under fluorescent microscope (Leica DMR epifluorescence microscope, Germany).

In the metabolic activity assay, a stock solution of 0.002% (w/v) resazurin reagent (Sigma, United Kingdom; filter sterilized) was prepared (Mariscal et al., 2009) and stored at -20°C. XooAS29 culture was grown with and without 500 ppm THY oil in PME broth and incubated for 24 h. The culture was centrifuged and the cells were washed with saline. Individually XooAS29 (100 μL) was added to each well of 96 well titre plate followed by addition of 20 μL resazurin solution. The resazurin fluorescence activity (λ_ex: 560 nm and λ_em: 590 nm) was measured after 60 min of incubation at 37°C. To test the growth inhibitory effect of THY oil, XooAS29 was grown in PME broth with and without effective concentrations of THY oil (500 ppm) using a shaking incubator for 8–13 h at 25 ± 2°C and the absorbance was recorded at 600 nm in triplicate samples collected hourly. Xoo cells treated with oil served as treatment while cells grown in the absence of oil was taken as control.

XooAS29 inoculum was grown as mentioned above and prior to carrying out the assay the final density was adjusted to 1 × 10⁸ CFU mL^-1. For conducting the motility assays, XooAS29 cells mixed with 500 ppm THY oil were gently inoculated in the center of the solidified PME agar plates. 1 μL suspension droplets were carefully placed in the center of soft swimming plates (3% malt extract, 0.5% peptone, 0.3% agar), and (2) swarm plates (3% malt extract, 0.5% peptone, 0.5% agar), and kept for incubation at 25 ± 2°C. The swimming and swarming traits were evaluated after 24 h by calculating the diameter of the bacterial zone. An equivalent quantity of only XooAS29 inoculum without oil in the respective media served as control.

The wetness assay of the colonies on swarm plates was calculated by the capillary-drop method as delineated by Wang et al. (2005). Capillary tubes of 0.5 mL capacity were mildly positioned on the treatment and control colonies for 30 s and the length of fluid entered was measured in millimeters. Xoo cells treated with oil served as treatment while cells grown in the absence of oil was taken as control. Five tubes were placed on different points on swarm plates for calculating the average and SD values.

For EPS, XooAS29 culture with and without 500 ppm THY oil was grown on PME media for 24 h at 28 ± 2°C. To detect the production of EPS from bacterial cells, the cells were scraped off the plates and resuspended in 15 mL SDW. Centrifugation at 15,000 × g for 12 min was done and the supernatant was collected for further analysis. The collected supernatant was muddled up with 1% (w/v) potassium chloride and double volume of 100% ethanol and further incubated at -20°C overnight. Again centrifugation was done and the precipitated EPS was left overnight for drying at 55°C before determining the total dry weight. Expression of the results was done relative to the cell density.

Measurement of xanthomonadin pigment was done as mentioned in the method illustrated by Wang et al. (2015). The XooAS29 cells collected by centrifuging 4 mL broth suspension with and without 500 ppm THY oil was mixed with 1 mL 100% methanol. The concoctions were further incubated in darkness for 10 min kept on rotating shaker followed by centrifugation at 12,000 × g for 8 min to collect the supernatant. The xanthomonadin pigment was estimated by measuring the absorbance at OD₄₄₅ and the result was denoted relative to the cell density measured before the assay (OD₅₉₅).

The fresh colony of XooAS29 strain was grown in 10 mL of PME liquid medium in presence and absence of 500 ppm THY oil at a starting OD₆₀₀ of 0.05. After incubation for 24 h, the XooAS29 culture at an OD₆₀₀ of 1.8 was centrifuged at 12,000 rpm for 12 min and the supernatant obtained was used for enzymatic plate assays (Yang and Tseng, 1988). The extracellular protease activity was measured by a radial diffusion assay in agar plates containing skimmed milk as substrate with 0.5% skimmed milk and 2% (wt/vol) agar. The media was poured in the plates and wells of 6 mm diameter were cut out with the help of a cork borer. The cell suspension (with or without THY oil) was applied in the well of the plates. Clearing zones around the colony due to the utilization of the substrate were measured after 24 h of incubation at 28 ± 2°C.

Extracellular cellulase activity was calculated as explained previously (Maki et al., 2011) using 2% CMC (Sigma-Aldrich) as substrate. Xylanase activity was also assayed on 1% agarose plates containing 0.5% RBB-Xylan and PME agar media (He et al., 2010). Positive xylanase activity was indicated by production of a white halo around the well in the plate.

For endoglucanase activity, XooAS29 culture with and without oil was pipetted into 6 mm diameter wells cut into CMC agar plates (1% agar, 0.125% CMC in 0.05 M potassium phosphate buffer, pH 6.0). The plates were incubated for 48 h at 28°C and then developed with Congo red dye (Barber et al., 1997). Extracellular lipase/esterase activity was assayed on PME media containing 0.01% CaCl₂ and 1% Tween 80 (Rajeshwari et al., 2005). The white crystals surrounding colonies was measured in medium containing CaCl₂ and Tween 80 indicating positive lipase activity. All the assays were repeated three times, independently, in triplicates.

For RNA isolation, single celled colony of XooAS29 with and without 500 ppm THY oil was transferred to 10 mL PME medium at 28 ± 2°C and sampled when the OD₅₉₅ reached 2.0 (Xu et al., 2015). RNA was isolated by following the manufacturer’s instructions of Trizol (Invitrogen, Carlsbad, CA, United States) method and RNA purity was evaluated with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, United States). Total RNA was used for synthesizing first-stranded cDNA using c-DNA verso reverse transcriptase (Thermo Scientific) following the instructions mentioned in the kit. The realtime RT-PCR was carried out in 96-well plates using a 7500 fast real-time PCR system (Applied Biosystems) with a Sybr Green master mix (Fermentas) following the manufacturer’s instructions (Li and Wang, 2014). The gene expression was done in triplicate in a total volume of 25 μL including a passive reference dye (ROX) (Fermentas). The designing of gene specific primers (Table 1) was done on the basis of the available genome sequence of X. oryzae pv. oryzae strain KACC10331. The primers synthesized targeted 21 genes that were earlier recognized to be related to motility, chemotaxis, hydrolytic enzymes, virulence, and DSFs. As an endogenous control DNA gyrase subunit B encoding gene gyrB was used and the relative transcript level was calculated using the method 2^{-deltadelta CT} described by Livak and Schmittgen, 2001.

XooAS29 culture with and without 500 ppm THY oil was individually allowed to grow as mentioned above and sampled when the OD₅₉₅ reached 2.0. DSF (cis-11-methyl-2-dodecenoic acid) and BDSF (cis-2-dodecenoic acid) were quantified as illustrated by He et al. (2010) and Xu et al. (2015). Five liters of XooAS29 supernatant with and without 500 ppm THY oil was individually collected by centrifugation at 9,000 rpm for 40 min at 4°C. Equal volume of ethyl acetate was used for extracting the obtained supernatant whose pH was adjusted to 4.0. The removal of organic phase was done by rotary evaporator at 40°C, and the residue obtained was dissolved in 20 mL of methanol. The final dried extract obtained was dissolved in HPLC grade methanol. The signaling molecules present in ethyl acetate extracted samples were detected through prewashed aluminum HPTLC plates (10 cm × 10 cm), coated with silica gel 60- GF254 (Merck). 2 μl of each sample was applied in triplicate on preactivated TLC plate using the Camag Linomat 5 automated TLC applicator equipped with a 100-μl Hamilton syringe. Hexane: diethylether: acetic acid (8:2:0.1) was implemented as mobile phase to develop TLC plate in Camag twin glass chamber. After developing the TLC plate to 90 mm distance, the plate was removed and dried at room temperature for 1 h. Developed plate was documented under 254 and 366 nm, and then transferred into previously saturated iodine chamber to visualize the spot under day light. The plates were then scanned at 270 nm with a Camag TLC Scanner with winCATS 3 software, using the deuterium lamp. The densitograms was recorded and peak area of both samples was observed.

The leave samples were taken from 40- to 60-day-old susceptible rice cultivar BLB Pusa Basmati 1 plants grown in greenhouse conditions. The leaves were washed twice with SDW followed by clipping tips of leaves with sterile scissors dipped in saturated XOOAS29 (10⁹ cells/mL) grown with and without 500 ppm THY oil (Kauffman et al., 1973). As per the adopted method, approximately 10⁶ cells are expected to be deposited at the site of pathogen inoculation. The leaves were then placed on the water agar (0.5% (wt/vol) agar) plates and left for 3 days at 28°C. The lesion lengths were measured by comparing leaves clipped with scissors dipped in XooAS29 alone and with THY oil. The experiment was repeated three times in triplicates.

The crystal structure of the targeted protein, dsf synthase RpfF (PDB: 3M6M) was selected as receptor and retrieved from the RCSB protein data bank. The elimination of water molecules and selection of single peptide chain of receptor was performed using chimera software. All ligand structures were extracted from NCBI PubChem database in sdf format and converted into pdb format using online tool “Smile Converter.” The graphical user interface program “Auto-Dock Tool 4.2” was used in docking simulation. The ligand energy was also optimized by defining the rotable bond, torsion angle and submerging the non-polar hydrogen bonds. Autodock tool demands the precalculated grid-box which was set at 70, 70, and 70 A⁰ (x, y, and z) with 0.375 A⁰ grid spacing to include all the presented amino acid residues of ligand binding pocket of receptor. Lamarckian genetic algorithm method was applied to perform docking. The default settings were used for all the other parameters. Final docked conformations were clustured using a tolerance of 2 A⁰ RMSD and the docking log (dlg) files were analyzed. The best ligand-receptor structure from the docked structures was opted based on the binding energy and number of H-bonds formed between the target and ligand. The results were visualized using visualization tool “Discovery studio 4.5”.

After getting the best interaction with thymol and targeted protein, screening was done to find out the effective concentration of thymol (0, 10, 25, 50, 75, 100, 200, 300, 400, and 500 μM). The selection for the effective concentration was done on the basis of the concentration which showed no growth inhibitory effect on Xoo along with inhibitory outcome on the biofilm formation. The experiment was repeated three times in triplicates.

Also, RT-qPCR was performed to quantify and compare the levels of gene expression for gumC, motA, estA, virulence gene acvB cheD, flgA, cheY, pilA, pglA along with DSF related genes rpfF, rpfC, and rpfG in Xoo when exposed to thymol. qRT-PCR was repeated twice, with four independent biological replicates each time.

Each experiment was done in three technical replicates and three biological replicates. Mean significant values were determined by Student’s t-test using SPSS package (SPSS V16.0, SPSS, Inc., Chicago, IL, United States). Analysis of variance (ANOVA) followed by the post-test (Duncan’s Multiple Comparison Test) was also used to analyze significance between more than two treatments. ^∗P < 0.05 and ^∗∗P < 0.01 was considered statistically significant.

In the present study, it was attempted to probe whether biofilm formation was reduced in XOOAS29, when treated with THY oil or not. The Xoo strain AS29 (KX010418) was routinely used throughout the experiment. The result of micro titre plate based CV staining revealed a significant decrease in biofilm formation in a concentration dependent manner after 24 h of incubation at 28°C in THY oil treated XooAS29 cells. As shown in Figure 1 and Supplementary Figure S1, THY oil at concentrations of 1000 and 500 ppm significantly reduced (1.64 and 1.50 fold respectively) biofilm forming ability of XooAS29.

To further confirm the result obtained from quantitative assay microscopic visualization was done only for the lower concentration of THY oil (500 ppm) significantly inhibiting biofilm formation under light microscope. The images obtained from three different staining methods namely SYTO-9, CV and MB dye further reconfirmed the observation that biofilm production was significantly lessened when XooAS29 cells were treated with 500 ppm of THY oil than the alone Xoo cultures (control) (Figure 2).

Since, the aim was to find an anti-virulence agent that only affected the secretion of virulence factors and did not kill the cells, the effect on XooAS29 on both metabolic activity and growth kinetics in the presence and absence of THY oil was investigated. The XooAS29 culture grown in the presence of both the 1000 and 500 ppm concentrations of THY oil had little/no significantly reduced Xoo metabolic activity (Figure 3A). Similarly, XooAS29 treated with THY oil (500 ppm) also showed no significant effect on growth of bacterial cells in comparison to the control set (Figure 3B).

Since, 500 ppm THY oil didn’t have any bactericidal effect on the XooAS29 cells along with significant reduction in biofilm formation; we tested the ability of XooAS29 to swim in low percentage agar medium in the presence of THY oil. After incubation at 28 ± 2°C for 24 h, Xoo AS29 cultured on plates with 500 ppm oil showed (P < 0.05) considerable reduction in swimming and swarming motility (1.19 and 1.24 fold, respectively) as compared to XooAS29 in the control plate. Small swimming and swarming diameters were observed in THY oil treatment which was in sharp contrast with the control sets where large diffused colony was observed on PME plates (Figures 4a–d).

In general, it is believed that bacterial cells need a certain amount of wetness for swarming movement on sticky surfaces. Thus, with the capillary-drop method, we evaluated the wetness of XooAS29 colonies grown on swarm plates with and without 500 ppm THY oil. As shown in Supplementary Figure S2, the wetness of the XooAS29 colony was significantly reduced (1.24 fold) in the presence of THY oil, suggesting that swarming defects by THY oil may be due to insufficient production of wetting agents or extracellular materials.

Production of EPS is a key determining factor of biofilm formation and is also correlated with the chemotactic movement in Xoo. Since the results revealed significant inhibition in biofilm formation of XooAS29 treated with 500 ppm of THY oil, we sought to extend our study by evaluating the effect of THY oil on EPS synthesis and xanthomonadin production by Xoo. The XooAS29 culture grown in the presence of 500 ppm THY oil significantly (P < 0.05) reduced (1.35 fold) EPS than the control treatment after 24 h of inoculation (Figure 5A). Interestingly, XooAS29 treated with 500 ppm THY oil also significantly (P < 0.05) reduced xanthomonadin which was about 1.23 fold less compared with the control set (Figure 5B).

Several other virulence related phenotypes were assayed by assessing the extracellular enzyme production of XooAS29 both in the presence and absence of THY oil. As shown in Table 2, 500 ppm of THY oil was found to significantly reduce endogluconase and xylanase activity (1.55 and 1.94 fold, respectively). However, interestingly non-significant affect was observed in the case of protease, cellulase, and lipase activity as compared to control plate inoculated with only XooAS29.

Previously, we showed that 500 ppm THY oil showed inhibitory effect on biofilm formation, EPS, xanthomonadin, and some of the extracellular hydrolytic enzymes produced by XooAS29. To obtain insight into the pathway by which THY oil exhibited these activities, we assessed the effect of oil on expression of genes significant for motility, EPS, lytic enzymes, virulence factors and DSF signaling in XooAS29 using RT-qPCR. The selected genes included the gum genes gumC, lytic enymes genes (xylanase, extracellular protease, endoglucanase, cellulase, 1,4-beta cellobiosidase, lytic enzyme, lipase/esterase, and polygalacturonase), DSF related biosynthetic genes rpfF, chemotaxis and motility genes cheD, cheY, motA, and motB, fimbrial protein gene pilA, flagellar genes flgA and flgE, and virulence gene acvB. The results showed that THY oil significantly down- regulated the expression of lytic enzymes genes such as extracellular protease (12%), lipase/esterase (12%), acvB (14%), chemotaxis and motility genes motB (19%), cheD (28%), cheY (11%), and gumC (18%), rpfC (38%), genes. Genes strongly suppressed were related to motility and flagellar biosynthesis genes such as motA (76%), flagellar gene flgE (60%), hydrolytic enzymes such as endoglucanase (75%), xylanase (64%), 1, 4-beta cellobiosidase (67%), and polygalacturonase (65%), along with DSF related gene rpfF (78%). Contrarily, the expression of cellulase, lytic enzyme, flgA and fimbrial protein pilA genes was not found to be significantly affected (Figure 6).

In Xoo, DSF and BDSF are key signaling molecules controlling virulence in response to cell density. The inhibitory effect of 500 ppm THY on production of both the molecules was also confirmed by HPTLC analysis (Figure 7). Separation of DSF and BDSF was attained using a mobile phase comprising of hexane–ethylether–acetic acid (8: 2: 1). Photography of the TLC plate was done at λmax 254 and 366 nm wavelengths in UV absorbance mode using the HPTLC system (CAMAG, Switzerland). The corresponding bands of DSF and BDSF were observed at Rf 5.9 and Rf 7.8, respectively.

In order to gain insight into whether the decrement in the above mentioned genes and traits impaired XooAS29 virulence and disease inciting potential, the virulence assay by leaf clip method was conducted. The results showed that the effective concentration of 500 ppm THY oil enervated the virulence potential of XooAS29 in comparison to the control leaves infected with only Xoo 3 days after pathogen inoculation (Figure 8). Water agar plates having rice leaves revealed significant decrement in lesion length (1.57 fold) in THY oil treated rice leaves as compared to only Xoo treated control leaves.

After THY oil was found to be potential in curtailing the pathogencity of XooAS29, our next step was to identify the individual components present in this oil. For this, THY oil was subjected to GC-FID apparatus. Thymol, gamma-terpinene, and para-cymene were found as major components of THY oil being 47.88, 29.61, and 20.15%, respectively. A representative image of the chromatogram is shown in Supplementary Figure S3.

After identification of the individual components of THY oil as thymol, gamma-terpinene, and para-cymene, we expanded our finding toward to identification of the active component responsible for inhibition of signaling molecules using molecular docking. Silencing of the DSF synthase RpfF protein in X. oryzae pv. oryzae, could be a potential target protein to curb the infection. To study the possible inhibitory action of individual components of THY oil on DSF synthase RpfF protein, we used information available from the RpfF (PDB: 3M6M) to model the three-dimensional (3D) structure of the RpfF protein catalytic domain. Docking of thymol, gamma-terpinene, and para-cymene in the putative RpfF binding pocket indicated that these compounds formed potential hydrogen bonds with some residues of catalytic importance (Figure 9 and Supplementary Figure S4). However, this structural model revealed a substantial region of interaction between thymol with RpfF protein, a positive regulator of DSF synthesis. The calculated binding energy of thymol, gamma-terpinene, and para-cymene was found to be -6.94, -5.99, and -5.79 kcal/mol, respectively.

XooAS29 biofilm formation with thymol was evaluated using a static biofilm assay. Biofilm formation was reduced by thymol in a concentration dependent manner as shown in Figure 10A. The effect of thymol on XooAS29 growth was also evaluated by monitoring the OD of cells cultures at 600 nm (Figure 10B). The pattern of XooAS29 growth was not significantly different between the cultures with 0 and 75 μM thymol during the lag, exponential and stationary growth phases (Figure 10B).

Previously, we showed that 75 μM thymol inhibited XooAS29 biofilm formation without effective cell growth. As shown in supplementary data, the effect of thymol was evaluated on the expression of various genes responsible for motility, virulence factors and DSF signaling in XooAS29 using RT-qPCR. The results showed that thymol (75 μM) significantly down-regulated the expression of gumC (30%), motA (17%), estA (22%), virulence gene acvB (12%), and pglA (54%)along with rpfF (68%) while other genes such as cheD, flgA, cheY, and pilA, rpfG, and rpfC were not significantly affected by thymol (Figure 11).