Anti-infective potential of hydroalcoholic extract of Punica granatum peel against gram-negative bacterial pathogens

Materials

All the growth media/ media ingredients/ assay reagents, and antibiotics were procured from Himedia, unless specified otherwise. TLC plates, and all organic solvents used in this study were procured from Merck. Whatman paper, and bacteriological filters were from Axiva (Haryana). Catechin was from Sigma Aldrich (USA). Triton X-100 was from CDH (New Delhi). All experiments were performed with the same set of reagents.

Test organisms

C. violaceum (MTCC 2656), and Lactobacillus plantarum (MTCC 2621), were procured from MTCC (Microbial Type Culture Collection, Chandigarh), whereas Bifidobacterium bifidum (NCDC 255) was procured from NCDC (National Collection of Dairy Cultures), NDRI (National Dairy Research Institute, Karnal). C. violaceum was grown in nutrient broth (HiMedia, Mumbai), L. plantarum was grown in Lactobacillus MRS medium (HiMedia, Mumbai), and B. bifidum was grown on MRS with 0.05% cysteine. Incubation temperature and time for C. violaceum, L. plantarum, and B. bifidum was 37°C, and 22–24 h. Antibiotic susceptibility profile of the bacterial strains used in this study was generated using the antibiotic discs-Dodeca Universal – I, Dodeca G - III – Plus, and Icosa Universal -2 (HiMedia, Mumbai). C. violaceum was found to be resistant to cefadroxil (30 µg), ampicillin (10 µg), cloxacillin (1 µg), and penicillin (10 µg). Culture of P. aeruginosa was obtained from Microbiology Department, M.G. Science Institute, Ahmedabad. Pseudomonas agar (HiMedia, Mumbai) was used for the maintenance of the culture. Strain of P. aeruginosa was found to be resistant to amoxicillin (30 µg), cefadroxil (30 µg), ampicillin (10 µg), cloxacillin (1 µg), penicillin (10 µg), chloramphenicol (30 µg), cefixime (5 µg), clindamycin (2 µg), and nitrofurantoin (300 µg).

Plant material

Peels of P. granatum were procured from the fruits purchased from local market in the city of Ahmedabad. Plant material (ref no: GU/Bot/29/4/2015) was authenticated for its unambiguous identity by Dr Archana Mankad (Deparment of Botany, Gujarat University, Ahmedabad). The plant name was checked against http://www.theplantlist.org on April 20, 2018. Collected peels were shade-dried, before being used for extract preparation. We also analysed another pomegranate fruit extract as ‘Pomella’ by Pharmanza Herbal Pvt. Ltd., which contained not less than 30% punicalagin

Extract preparation

A hydroalcoholic extract of the plant material was prepared in 50% ethanol using the microwave-assisted extraction (MAE) method (Kothari et al., 2009). Dry peel powder (total 5 g; 1 g in each vessel) was soaked into the solvent in a ratio of 1:50, and subjected to microwave heating (Electrolux EM30EC90SS) at 720 W. Total heating time was 120 seconds, with intermittent cooling. This was followed by centrifugation (at 10,000 rpm for 15 min.), and filtration with Whatman paper #1 (Axiva, Haryana). Solvent was evaporated from the filtered extract and then the dried extract was reconstituted in dimethyl sulfoxide [DMSO (absolute); Merck] for broth dilution assay. Reconstituted extract was stored under refrigeration for further use. Extraction efficiency was calculated as percentage weight of the starting dried plant material, and its value was 42% w/w.

Broth dilution assay

Assessment of QS-regulated pigment production by test pathogens in presence or absence of the test formulation, was done using broth dilution assay (Patel et al., 2017). Organisms were challenged with different concentrations (0.5-500 μg/ml) of P. granatum peel extract (PGPE). Nutrient broth or Pseudomonas broth (peptic digest of animal tissue 20 g/l, potassium sulphate 10 g/l, magnesium chloride 1.4 g/l, pH 7.0 ± 0.2) was used as a growth medium. Inoculum standardized to 0.5 McFarland turbidity standard was added at 10% v/v, to the media supplemented with required concentration of PGPE, followed by incubation at appropriate temp for each organism. Appropriate vehicle control containing DMSO was also included in the experiment, along with abiotic control (containing extract and growth medium, but no inoculum). Catechin (50 μg/ml; Sigma Aldrich, USA) was used as positive control, since it is already known QS inhibitor (Nazzaro et al., 2013; Vandeputte et al., 2010).

Measurement of bacterial growth and pigment production

At the end of the incubation, bacterial growth was quantified at 764 nm (Joshi et al., 2016). This was followed by pigment extraction and quantification, as per the method described below for each of the pigments. Purity of each of the extracted pigment was confirmed by running a UV-vis scan (Agilent Cary 60 UV-visible spectrophotometer). Appearance of single major peak (at the λ_max reported in literature) was taken as indication of purity.

Violacein extraction

Extraction of violacein from C. violaceum culture was executed as described by Choo et al. (2006). A total of 1 ml of the culture broth was centrifuged (Eppendorf 5417 R) at 15,300g for 10 min at room temperature, and the resulting supernatant was discarded. The remaining cell pellet was resuspended into 1 ml of DMSO, and vortexed, followed by centrifugation at 15,300g for 10 min. The purple coloured violacein was extracted in the supernatant; OD was measured at 585 nm. Violacein unit was calculated as OD₅₈₅/OD₇₆₄. This parameter was calculated to nullify the effect of any change in cell density on pigment production.

Pyoverdine and pyocyanin extraction

Extraction of pyoverdine and pyocyanin from P. aeruginosa culture was achieved as described in El-Fouly et al. (2015) and Unni et al. (2014), respectively. A total of 1 ml of the culture broth was mixed with chloroform (Merck, Mumbai) in 2:1 proportion followed by centrifugation at 12,000 rpm (13,520g; REMI CPR-24 Plus) for 10 min. This resulted in formation of two immiscible layers. OD of the upper water-soluble phase containing yellow-green fluorescent pigment pyoverdine was measured at 405 nm. Pyoverdine Unit was calculated as OD₄₀₅/OD₇₆₄.

The lower chloroform layer containing pyocyanin was mixed with 0.1 N HCl (Merck; at the rate of 20%v/v), resulting in a colour change from blue to pink. Absorbance of this pyocyanin in acidic form was measured at 520 nm. Pyocyanin Unit was calculated as OD₅₂₀/OD₇₆₄.

N- acyl-homoserine lactone (AHL) augmentation assay

AHL extraction: AHL was extracted from the bacterial culture broth as described by Chang et al. (2014). OD of overnight grown bacterial culture was standardized to 1.00 at 764 nm. It was centrifuged at 5000g for 5 min. Cell free supernatant was filter sterilized using 0.45 µm filter (Axiva, Haryana), and was mixed with equal volume of acidified ethyl acetate [0.1% formic acid (Merck) in Ethyl acetate (Merck)]. Ethyl acetate layer was collected, and evaporated at 55°C, followed by reconstitution of the dried crystals in 100 µl phosphate buffer saline (pH 6.8). Identity of thus extracted AHL was confirmed by thin-layer chromatography (TLC). R_f value of purified AHL from C. violaceum culture, while performing TLC [Methanol (60): Water (40); TLC Silica gel 60 F₂₅₄ plates; Merck] McClean et al., (1997) was found to be 0.70, near to that (0.68) reported for N-hexonyl homoserine lactone (C6-HSL) (Shaw et al., 1997). TLC of AHL extracted from P. aeruginosa resulted in three spots corresponding to R_f values of 0.43, 0.68, and 0.92 near to those (0.41, 0.68, 0.84) for C8-HSL, C6-HSL, and C4-HSL respectively.

Bacterial culture growing in presence of test formulation was supplemented with 2%v/v AHL after 6 hours of incubation, and at the end of a total 24-hour incubation pigments were extracted from AHL-supplemented experimental tubes, as well as AHL-non-supplemented control tubes. If the QS-regulated pigment production is found inhibited in both these tubes in comparison to bacteria growing in absence of extract as well as AHL, then the effect of test extract was interpreted as signal-response inhibitor, because if the test formulation would have acted as a signal-supply inhibitor, then exogenous supply of AHL should restore pigment formation by the bacteria.

Hemolysis assay

This assay was done as described by Neun et al. (2015). OD₇₆₄ of overnight grown culture was standardized to 1.00. Cell free supernatant was prepared by centrifugation at 15,300 g for 10 min. A total of 10 µl human blood (collected in heparinized vial) was incubated with this cell free supernatant for 2 h at 37°C, followed by centrifugation at 800 g for 15 min; 1% Triton X-100 (CDH, New Delhi) was used as a positive control. Phosphate buffer saline was used as a negative control. OD of the supernatant was read at 540 nm, to quantify the amount of haemoglobin released.

Assay of bacterial susceptibility to lysis in presence of human serum

This assay was performed as described by Ferro et al. (2016). Serum was separated by centrifuging blood at 1,500 rpm (800g) for 10 min. Bacterial culture grown in media with and without PGPE was centrifuged, and the cell pellet was reconstituted in PBS, so that the resulting suspension attains OD₇₆₄=1. A total of 200 µl of this bacterial suspension from control or experimental tubes was mixed with 740 µl of PBS and 60 µl of serum. After incubation for 24 h at 37°C, absorbance was read at 764 nm. DMSO -treated cells (0.5%v/v) suspended in PBS served as control, against which OD of the PGPE-treated cells (serum-exposed) was compared. Tubes containing bacterial cells exposed neither to DMSO nor serum were also included in the experimental set-up, to nullify any interference from autolysis.

Human blood was obtained from the authors, who each gave their written informed consent. The use of this blood was approved by the Institutional Ethics Committee of the Institute of Science, Nirma University (approval no: EC/NU/18/IS/4).

Catalase assay

The OD₇₆₄ of the culture was adjusted to 1.00. Next, 400 µl of phosphate buffer was added into a 2 ml vial followed by 400 µl H₂O_2. To this 200 µl of the bacterial culture was added, and the mixture was incubated for 10 min. Then 10 µM of sodium azide was added to stop the reaction (Iwase et al., 2013), followed by centrifugation at 12,000 rpm for 10 min. OD of the supernatant was measured at 240 nm to quantify remaining H₂O₂ (Weydert & Cullen, 2010), with phosphate buffer as blank. Disappearance of the substrate H₂O₂ was taken as the measure of the catalase activity.

Assay for biofilm formation, eradication and viability

In this assay, the control and experimental groups contained nine test tubes. In each group, three subgroups were made. First subgroup of three test tubes in the experimental group contained Pseudomonas broth supplemented with PGPE, whereas remaining six tubes contained Pseudomonas broth with no PGPE on first day of experiment. All these tubes were inoculated with inoculum (10%v/v) standardized to 0.5 McFarland turbidity standard (making the total volume in the tube 1 ml), followed by incubation at 37°C for 24 h under static condition, which resulted in formation of biofilm as a ring on walls of the glass tubes. This biofilm was quantified by crystal violet assay (Patel et al., 2013), preceded by quantification of bacterial cell density and pigment. Content from the remaining six test tubes from rest of the two subgroups were discarded following cell density and pigment estimation, and then the biofilms remaining on inner surface of these tubes were washed with phosphate buffer saline (PBS; pH 7) to remove loosely attached cells. Now, 2 ml of minimal media (Sucrose 15 g/l, K₂HPO₄ 5.0 g/l, NH₄Cl 2 g/l, NaCl 1 g/l, MgSO₄ 0.1 g/l, yeast extract 0.1 g/l, pH 7.4±0.2) containing PGPE, was added into each of these tubes, so as to cover the biofilm completely, and tubes incubated for 24 h at 37°C. At the end of incubation, one subgroup of 3 tubes was subjected to crystal violet assay to know whether any eradication of the pre-formed biofilm has occurred under the influence of PGPE, and the last subgroup of 3 tubes was subjected to viability assessment through MTT assay. For the crystal violet assay, the biofilm- containing tubes (after discarding the inside liquid) were washed with PBS in order to remove all non-adherent (planktonic) bacteria, and air-dried for 15 min. Then, each of the washed tubes was stained with 1.5 ml of 0.4% aqueous crystal violet solution for 30 min. Afterwards, each tube was washed twice with 2 ml sterile distilled water and immediately de-stained with 1500 µl of 95% ethanol. After 45 min of de-staining, 1 ml of de-staining solution was transferred into separate tubes, and read at 580 nm. For the MTT assay (Trafny et al., 2013), the biofilm-containing tubes (after discarding the inside liquid) were washed with PBS in order to remove all non-adherent (planktonic) bacteria, and air-dried for 15 min. Then 900 µl of minimal media was added into each tube, followed by addition of 100 µl of 0.3% MTT [3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide, HiMedia]. After 2 h incubation at 37°C, resulting purple formazan derivatives were dissolved in DMSO and measured at 560 nm.

Cell surface hydrophobicity (CSH) assay

Bacterial surface hydrophobicity was measured using the bacterial adhesion to hydrocarbon (BATH) assay as described in Hui & Dykes (2012). P. aeruginosa culture was collected at stationary phase and pelleted by centrifugation (NF800R, NUVE, Belgium) at 7,000 rpm for 10 min. This pellet was washed twice in phosphate buffer saline (PBS; pH 7.4), and then resuspended in PBS with PGPE (25 and 50 μg/ml) to OD₇₆₄= 1.00. Same procedure was repeated with P. aeruginosa culture without PGPE, as a control. Each bacterial suspension was then incubated for 1 h at room temperature. A 2-ml sample of each suspension was collected and absorbance (A) at 764 nm was measured, using PBS as the blank. Next, 1 ml xylene (HiMedia, Mumbai) was added to the 2-ml cell suspension, and this mixture was vortexed for 2 min. The phases were then allowed to separate for 1 h. The absorbance of the aqueous phase (A₀) was again determined. Results were expressed as % attachment to xylene = (1-A/A₀) × 100.

Determination of the effect of PGPE on antibiotic susceptibility of the test organism

After in vitro assessment of QS modulatory (QSM) property of the test formulation, effect of this PGPE on antibiotic susceptibility of the test pathogen was investigated. The bacterial cells pre-treated with PGPE were subsequently challenged with sub-MIC concentration of antibiotic. All the antibiotics were procured from HiMedia, Mumbai. Names and concentrations of the antibiotics used can be seen in Figure 1c, Figure 2c, Figure 5c and Figure 7e.

In vivo assay

In vivo efficacy of the PGPE was evaluated using the nematode worm Caenorhabditis elegans as the model host, using the method described by Eng & Nathan (2015) with some modification. This worm was maintained on Nematode Growing Medium (NGM; 3 g/l NaCl, 2.5 g/l peptone, 1 M CaCl₂, 1 M MgSO₄, 5 mg/ml cholesterol, 1 M phosphate buffer of pH 6, 17 g/l agar-agar) with Escherichia coli OP50 (procured from LabTIE B.V., JR Rosmalen, the Netherlands) as the feed. Worm population to be used for the in vivo assay was kept on NGM plates not seeded with E. coli OP50 for three days, before being challenged with the test pathogen.

Test bacterium was incubated with the PGPE for 22–24h. Following incubation, OD₇₆₄ of the culture suspension was equalized to that of the DMSO control. 100 µl of this bacterial suspension was mixed with 900 µl of the M9 buffer containing 10 worms (L3–L4 stage). This experiment was performed in 24-well (sterile, non-treated) polystyrene plates (HiMedia), and incubation was carried out at 22°C. Number of live vs. lead worms was counted everyday till five days by putting the plate (with lid) under light microscope (4X). Standard antibiotic- (ampicillin 500 µg/ml; gentamicin 0.1 µg/ml) and catechin- (50 µg/ml) treated bacterial suspension were used as positive control. Straight worms were considered to be dead. On last day of the experiment, when plates could be opened, their death was also confirmed by touching them with a straight wire, wherein no movement was taken as confirmation of death.

Investigating whether the bacterium becomes resistant to PGPE upon repeated exposure

P. aeruginosa was subcultured on PGPE (500 μg/ml)-containing agar medium 10 times, and this PGPE-habituated culture was then challenged with PGPE (500 μg/ml) in liquid media for broth dilution assay, wherein bacterial cell density and pigment production were quantified as detailed in preceding sections. Virulence of this PGPE-habituated culture towards C. elegans was also assessed using the same methodology as described above under the heading ‘In vivo assay’.

Chromatographic analysis

Chromatographic fingerprint of PGPE was generated through HPLC (Shimandzu Lab Solutions) by running the extract through a C18 column (Shimandzu) for 75 min. Mobile phase consisted of a mixture of trichloroacetic acid (0.05%, w/v) in water, and trichloroacetic acid (0.05%, w/v) in methanol, wherein proportion of latter was varied over a gradient of 10–80%. Flow rate was 1 ml/min. Detection wavelength of the UV detector was set at 258 nm. Punicalagin A and punicalagin B (both at 100 ppm; Sigma Aldrich) were used as standard compounds.

Molecular docking

Structures of the pure compounds were downloaded from PubChem: catechin (PubChem CID: 73160), gallic acid (GA) (PubChem CID: 370), quercetin (PubChem CID: 5280343), ellagic acid (PubChem CID: 5281855), cinnamic acid (PubChem CID: 444539), and chlorogenic acid (PubChem CID: 1794427). Structure of the bacterial target proteins was downloaded from the Protein Data Bank (PDB): CviR of C. violaceum (PDB ID 3QP4), LasR (PDB ID 2UV0), and PqsR (PDB ID 4JVC) of P. aeruginosa. AutoDock Vina (Trott & Olson, 2010) was used for docking studies, Discovery Studio Visualizer 4.0.1 was used to study protein-ligand interactions, Pymol viewer (version v1.3r1) was used to convert protein-ligand docked structures from .pdbqt to .pdb format, and MglTools was used to prepare protein and ligand files for docking.

Statistical analysis

All the in vitro experiments were performed in triplicate, and measurements are reported as mean ± standard deviation (SD) of three independent experiments. In vivo experiments were run in four replicates. Statistical significance of the data was evaluated by applying t-test using Microsoft Excel. p values ≤0.05 were considered to be statistically significant.

C. violaceum

C. violaceum was challenged with the PGPE at 0.5–500 μg/ml (Figure 1A). PGPE had a negative effect on bacterial growth only at concentrations ≤25 μg/ml; however, an increase in concentration could not inhibit growth to any greater extent. In fact, statistically all the concentrations in the range 50–500 μg/ml had equivalent effect on C. violaceum growth. On the other hand, production of the QS-regulated pigment violacein was sensitive to PGPE at lesser concentrations (i.e. at ≤5 μg/ml), and the inhibitory effect increased with increase in concentration of the extract. PGPE can be said to exert a purely QSM effect at 5–10 μg/ml, as at these concentrations, it could inhibit pigment production significantly without inhibiting C. violaceum growth.

Figure 1. Effect of PGPE on C. violaceum.

‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and violacein production of C. violaceum. (A) Effect of PGPE on growth and QS- regulated violacein production in C. violaceum: Bacterial growth was measured as OD₇₆₄; OD of violacein was measured at 585 nm, and Violacein Unit was calculated as the ratio OD₅₈₅/OD₇₆₄ (an indication of violacein production per unit of growth); Catechin (50 µg/ml) did not exert any effect on growth of C. violaceum, but inhibited violacein production by 47.69±0.03%. (B) PGPE acts as a signal-response inhibitor against C. violaceum. (C) PGPE-pre-treatment enhances susceptibility of C. violaceum to different antibiotics. (D) PGPE-treatment attenuates virulence of C. violaceum towards C. elegans: Catechin (50 μg/ml) and ampicillin (500 μg/ml) employed as positive controls conferred 100% protection on worm population, and pre-treatment of bacteria with PGPE at concentrations (10, 25, 50 and 100 μg/ml) other than those shown in figure allowed 75%, 77.5%, 80%, and 75% worm survival, respectively; DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans; DMSO (0.5%v/v) and PGPE at tested concentrations showed no toxicity towards the worm. *p<0.05, **p<0.01, ***p<0.001; AS, antibiotic susceptibility; QS, quorum sensing; PGPE, Punica granatum peel extract.

Once the QS-inhibitory potential of PGPE against C. violaceum was confirmed, we experimented to know whether it acts as a signal-supply inhibitor or signal-response inhibitor. To know this, we added AHL (2% v/v; 6 h post-inoculation) into quorum inhibited culture of C. violaceum growing in the presence of PGPE (10–100 μg/ml). Following AHL (signal) augmentation, the quorum inhibitory effect of PGPE on violacein production was not reversed (Figure 1B), indicating that PGPE acts as a signal-response inhibitor.

When PGPE-treated C. violaceum culture was subsequently challenged with sub-MIC concentrations of four antibiotics belonging to four different classes, extract pre-treatment was found to enhance susceptibility of this bacterium to all these antibiotics. However, pre-treatment with higher (25 μg/ml) PGPE concentration was found to be no better than the lesser (10 μg/ml) concentration, with respect to making bacteria more susceptible to given antibiotic, except in the case of ampicillin (Figure 1C).

PGPE was found to curb haemolytic potential of C. violaceum in a dose-dependent fashion, wherein this extract at 100 μg/ml could reduce haemolysis by ~36% (Table 1). This extract could also reduce the catalase activity of C. violaceum marginally; however, the effect on this oxidative stress response enzyme activity did not increase with increase in PGPE concentration (Table 1).

In vitro demonstration of the QSM potential of PGPE was followed by assessment of its anti-infective potential in vivo, employing C. elegans as the model host, wherein PGPE was found to confer survival benefit on C. elegans upon challenge with C. violaceum at all the concentrations (0.5–100 μg/ml) tested (Figure 1D). Notably, even the concentrations (0.5–2.5 μg/ml) having no significant in vitro effect on C. violaceum growth and pigment production, were found to be effective in vivo. Maximum in vivo benefit was obtained at 5 μg/ml PGPE concentration, and worm survival percentage at any of the higher concentrations were statistically no better than that obtained at 5 μg/ml. Onset of death in the worm population was also delayed by 2 days (i.e. death started on first day-post infection in control wells, as against on third day in experimental wells) when the challenger bacteria were pre-treated with PGPE. Movement of the surviving worm in ‘vehicle control’ wells was slowed down, whereas worms surviving in the ‘experimental’ wells showed normal movement. Further, worms in the experimental well corresponding to 100 μg/ml PGPE could also generate progeny worms, which were not observed in any other well.

P. aeruginosa

P. aeruginosa was challenged with PGPE at 0.5–500 μg/ml, wherein growth of this bacterium got notably inhibited 50 μg/ml onwards. Pyocyanin production was affected negatively from 5 μg/ml onwards in a concentration-dependent manner, whereas pyoverdine production was enhanced from 2.5 μg/ml onwards in a largely concentration-dependent fashion except at 500 μg/ml (Figure 2A). When AHL was added exogenously to the quorum-modulated P. aeruginosa culture, the inhibitory effect of PGPE on pyocyanin and its stimulatory effect on pyoverdine production was not reversed (Figure 2B) indicating that PGPE acts as a signal-response inhibitor against this pathogen.

Figure 2. Effect of PGPE on various traits of P. aeruginosa, in vitro.

‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and pigment production of P. aeruginosa. (A) Effect of PGPE on growth and QS-regulated pigment production in P. aeruginosa: Bacterial growth was measured as OD₇₆₄; OD of pyoverdine was measured at 405 nm, and Pyoverdine Unit was calculated as the ratio OD₄₀₅/OD₇₆₄ (an indication of pyoverdine production per unit of growth), Pyocyanin Unit was calculated as the ratio OD₅₂₀/OD₇₆₄ (an indication of pyocyanin production per unit of growth); Catechin (50 µg/ml) inhibited pyoverdine and pyocyanin production by 17.13±0.06% and 23.65±0.04% respectively, without affecting the bacterial growth. (B) PGPE acts as a signal-response inhibitor against P. aeruginosa. (C) P. aeruginosa challenged with antibiotics following pre-treatment with PGPE. (D) PGPE modulates susceptibility of P. aeruginosa to lysis by human serum: ‘Control’ in serum-dependent lysis assay was PGPE-unexposed cells of P. aeruginosa incubated with human serum. (E) Effect of PGPE on P. aeruginosa biofilm formation, eradication, and viability. Crystal violet assay was performed to measure biofilm formation, and biofilm eradication, followed by the measurement of OD at 580 nm; Cell viability in biofilm was estimated through MTT assay, wherein OD was measured at 560 nm. (F) PGPE-treatment reduces the virulence of P. aeruginosa towards C. elegans: Catechin (50 μg/ml) and gentamicin (0.1 μg/ml) employed as a positive controls conferred 100% and 80% protection on worm population respectively. Pre-treatment of bacteria with PGPE at concentrations (2.5, 5, 10, 25, and 100 μg/ml) other than shown in figure allowed 73.75, 75, 77.5, 77.5, and 75% worm survival respectively; DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans; DMSO (0.5%v/v) and PGPE at tested concentrations showed no toxicity towards the worm. (G) Effect of PGPE on P. aeruginosa growth, Pyoverdine Unit, and Pyocyanin Unit remained unaltered after repeated exposure to PGPE (500 μg/ml). *p<0.05, **p<0.01, ***p<0.001. AS, antibiotic susceptibility; QS, quorum sensing; PGPE, Punica granatum peel extract.

Effect of PGPE pre-treatment on antibiotic susceptibility of P. aeruginosa was also investigated. However, no major changes were observed in bacterial susceptibility to four different test antibiotics, when challenged with these antibiotics following PGPE exposure (Figure 2C). This extract was not found to alter catalase activity much; however, haemolytic potential of P. aeruginosa was notably affected 25 μg/ml onwards (Table 1). PGPE at 50 μg/ml could enhance (by 13.55%; p=0.03) susceptibility of this bacterial pathogen to lysis by human serum (Figure 2D).

PGPE was able to reduce P. aeruginosa biofilm formation up to 17-30%, when tested at 10–50 μg/ml (Figure 2E). When this extract was applied on pre-formed biofilm, it could eradicate the biofilm by ~15–19%; and biofilm viability (metabolic activity as measured by MTT assay) was reduced by ~13–22%. However, whether this observed reduction in viability is due to lesser number of cells, or some other reason, remains to be investigated. Reduced biofilm formation by P. aeruginosa in presence of PGPE may partly be due to this extract’s ability to reduce CSH marginally, which is an important determinant of biofilm forming ability in bacteria (Hui and Dykes, 2012). Xylene adherence of P. aeruginosa treated with PGPE at 25 and 50 μg/ml was found to be reduced by 14.19% (p<0.05; % adherence=44.74) and 7.82% (p<0.05; % adherence =38.37) respectively, against 30.55% adherence of control.

In vivo assay did confirm the anti-infective potential of PGPE against P. aeruginosa. At concentrations as low as 0.5–1 μg/ml, it could support overall survival of the worm population up to 56-71% in face of P. aeruginosa challenge, as against only 12.5% worm survival in control population. Higher PGPE concentrations did not offer any better protection to C. elegans against P. aeruginosa (Figure 2F). Notably, in vitro experiments showed PGPE to have no effect on P. aeruginosa growth below 5 μg/ml, and no effect on its QS-regulated pigments below 2.5 μg/ml; whereas PGPE at 0.5–1 μg/ml was able to confer significant survival benefit on C. elegans against P. aeruginosa challenge. Onset of death in worm population was also delayed by 1–3 days, when P. aeruginosa was exposed to PGPE before being allowed to infect C. elegans.

After confirming the anti-infective potential of PGPE against P. aeruginosa, we investigated whether this bacterium can develop resistance to PGPE upon repeated exposure to this extract. We sub-cultured P. aeruginosa on PGPE (500 μg/ml) containing agar medium 10 times, and this PGPE-habituated culture was then challenged with PGPE (500 μg/ml) in liquid media for broth dilution assay, wherein effect of PGPE on growth and pigment production in P. aeruginosa was not found to be much different than that on P. aeruginosa with no previous PGPE exposure (Figure 2G). Similarly, there was no difference in the virulence of P. aeruginosa towards C. elegans, irrespective of whether it was previously habituated to PGPE. (Figure 2F).

With P. aeruginosa, we did an additional experiment to investigate whether the daughter cells of PGPE-exposed P. aeruginosa bear any attenuation of their virulence owing to their parent cells being exposed to PGPE. When the PGPE-exposed P. aeruginosa was subsequently subcultured on PGPE-free media, pigment production in cells obtained after first such subculturing was still altered in a pattern similar to the PGPE-exposed parent culture (Figure 3A). Though QS-regulated pigment production remained deviated from ‘control’ upon second subculturing too, it did not follow the same pattern, as observed with cells obtained after first subculturing on PGPE-free media. With each subculturing, virulence of P. aeruginosa towards C. elegans approached nearer to that of ‘control’, but it still did not got equivalent to ‘control’ (Figure 3B), clearly suggesting that effect of PGPE-treatment was not completely absent even from the progeny cells (who never were directly exposed to PGPE), whose parent cells received single PGPE-exposure. This phenomenon of long-lasting effect after single-time exposure to the antimicrobial agent is described as ‘post-antimicrobial effect’. Here, in case of a plant extract, we prefer to call it post-extract effect (PEE), which refers to the persistent suppression of one or more bacterial traits (e.g. growth/virulence, etc.) after one-time-exposure to antimicrobial agents, and may last for many hours depending on the concentration of test agent and the susceptibility of the target bacterium (Pfaller et al., 2004; Ramadan et al., 1995; Ramanuj et al., 2012). During actual animal/human infections, such phenomenon can be of high significance, as though the infectious bacteria may multiply inside the host system, their progenies may not have the virulence at par with the parent cells.

Figure 3. Post extract effect of PGPE (10 μg/ml) on P. aeruginosa.

‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and pigment production of P. aeruginosa. (A) Effect of PGPE on growth and QS-regulated pigment production in P. aeruginosa after subculturing of cells in PGPE-free media. Bacterial growth was measured as OD₇₆₄; OD of pyoverdine was measured at 405 nm, and Pyoverdine Unit was calculated as the ratio OD₄₀₅/OD₇₆₄ (an indication of pyoverdine production per unit of growth), Pyocyanin Unit was calculated as the ratio OD₅₂₀/OD₇₆₄ (an indication of pyocyanin production per unit of growth). (B) PGPE-treatment reduces the virulence of P. aeruginosa towards C. elegans even after subculturing of cells in PGPE-free media. Catechin (50 μg/ml) and gentamicin (0.1 μg/ml) employed as a positive controls conferred 100% and 80% protection on worm population respectively; DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans; DMSO (0.5%v/v) and PGPE at tested concentrations showed no toxicity towards the worm. *p<0.05, **p<0.01, ***p<0.001; AS, antibiotic susceptibility; QS, quorum sensing; PGPE, Punica granatum peel extract.

Characterization of PGPE

Once in vitro and in vivo anti-pathogenic potential of PGPE was confirmed, we proceeded to chromatographic fingerprinting of this extract. The resultant chromatogram is shown in Figure 4. Since punicalagin is one of the most widely reported bioactive metabolite from P. granatum, we quantified its amount in our extract, which was found to be present at 6.6%. Earlier we had analyzed another pomegranate fruit extract marketed as ‘Pomella’ by Pharmanza Herbal Pvt. Ltd., which contained not less than 30% punicalagin for its QS-modulatory potential (extended data, Figure S1A) (Joshi et al., 2018). Though there were no major differences with respect to the in vitro effect on growth and pyocyanin production by P. aeruginosa, pyoverdine production was affected much more by Pomella than PGPE. However, in vivo efficacy of PGPE was bit better than that of Pomella; PGPE at 0.5-50 μg/ml allowed 56–80% worm survival, as against 55–65% worm survival supported by ‘Pomella’ at identical concentrations (extended data, Figure S1B) (Joshi et al., 2018). From this, we may speculate punicalagin not to be the major constituent responsible for P. granatum extract’s anti-pathogenic activity. Against C. violaceum too, except at 250 μg/ml, effect of PGPE on violacein production was statistically not different from that of Pomella, with latter exerting a bit more inhibitory effect on growth of this bacterium (extended data, Figure S1C) (Joshi et al., 2018). However, role of punicalagins cannot be completely ruled out, as they are known to be metabolized inside human system to urolithins, and latter has been reported to be capable of exerting anti-QS effect (Giménez-Bastida et al., 2012). Larrosa et al. (2010) suggested the possibility of urolithin-A being the most active anti-inflammatory compound derived from pomegranate ingestion in healthy subjects.

Figure 4. HPLC profile of PGPE.

Peak 11 and 14 occupied 24.61% and 27.48% of total area, which corresponds to Punicalagin A and B, respectively. PGPE, Punica granatum peel extract.

QSM potential of P. granatum peel phytocompounds in pure form

Other than punicalagin, pomegranate peel extracts have been reported to contain pyrogallol, catechin, rutin, cinnamic acid, benzoic acid, chlorogenic acid, acacetin, ferulic acid, kampferol, genistein (Ali et al., 2014), coumaric acid (Khan et al., 2017) tannins (punicalin, pedunculagin, GA and ellagic acid) (Ismail et al., 2012; Pagliarulo et al., 2016), and quercetin (Saxena et al., 2017). To gain more insight, we docked all these reported compounds against LuxR analogues of C. violaceum and P. aeruginosa, against which our extract was found to act as signal-response inhibitor. Results of this molecular docking exercise are presented in Table 3. The amino acid residues involved in this on-screen ligand-receptor binding, for compounds used in wet-lab study, are shown in extended data, Table S1 (Joshi et al., 2018).

Pure compounds against C. violaceum. Against the LuxR analogue of C. violaceum, docking scores higher than that of its natural ligand (C6-HSL) were obtained for acacetin, catechin, and quercetin. However, for further in vitro experiments, we did not set any cut-off value with respect to docking score, instead we experimented with all those pure compounds from those listed in Table 2, which were available in our lab viz. catechin, quercetin, chlorogenic acid, cinnamic acid, and GA. Cinnamic acid had no effect on C. violaceum growth and pigment production in the concentration range 5–100 μg/ml, except 26% reduction in violacein production at 100 μg/ml (Figure 5A). Hence all subsequent experiments with cinnamic acid were performed at 100 μg/ml. Cinnamic acid was able to enhance C. violaceum susceptibility to streptomycin by 11.94%, but had no effect on its susceptibility to cephalexin and tetracycline (Figure 5C). Haemolytic potential and catalase activity of C. violaceum were curbed to a smaller (8.47% and 2.68%, respectively), albeit statistically significant extent (Table 3). This phenylpropanoid metabolite was also able to confer some survival benefit (24% lesser worm death) on C. elegans (Figure 5D). Cinnamic acid induced violacein inhibition was not reversed upon exogenous supply of AHL (extended data, Figure S2) (Joshi et al., 2018), indicating its mode of QS-inhibition to be signal-response inhibition.

Figure 5. Effect of cinnamic acid and chlorogenic acid on C. violaceum.

‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and violacein production of C. violaceum. (A) Effect of cinnamic acid on growth and violacein production in C. violaceum. Bacterial growth was measured as OD₇₆₄; OD of violacein was measured at 585 nm, and Violacein Unit was calculated as the ratio OD₅₈₅/OD₇₆₄ (an indication of violacein production per unit of growth). (B) Effect of chlorogenic acid on growth and violacein production in C. violaceum. Bacterial growth was measured as OD₇₆₄; OD of violacein was measured at 585 nm, and violacein unit was calculated as the ratio OD₅₈₅/OD₇₆₄ (an indication of violacein production per unit of growth). (C) Pre-treatment of C. violaceum with chlorogenic acid and cinnamic acid enhances its susceptibility to streptomycin and cephalexin. (D) Cinnamic acid, chlorogenic acid, and catechin reduced virulence of C. violaceum towards C. elegans. Catechin (50 μg/ml) and ampicillin (500 μg/ml) employed as positive controls conferred 100% protection on worm population. DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans; DMSO (0.5%v/v) and compounds at tested concentrations showed no toxicity towards the worm. *p<0.05, **p<0.01, ***p<0.001. AS, antibiotic susceptibility; QS, quorum sensing.

Chlorogenic acid could not affect C. violaceum growth and pigment production at any of the test concentrations (5–100 μg/ml), except 16.21% inhibition of violacein production at 75 μg/ml (Figure 5B). Notably, violacein production was not affected at higher concentration (i.e. 100 μg/ml). Hence, all subsequent experiments with chlorogenic acid were performed at 75 μg/ml. At this concentration chlorogenic acid could make C. violaceum 8.91% more susceptible to streptomycin; however, it did not alter this bacterium’s susceptibility to cephalexin and tetracycline (Figure 5C). Chlorogenic acid was also able to inhibit catalase activity and haemolytic activity of C. violaceum by 3.84% and 10.16% respectively (Table 3). It could also offer a sizable (49%) survival benefit to C. elegans, when challenged with C. violaceum (Figure 5D). AHL augmentation of the C. violaceum culture experiencing chlorogenic acid induced QS inhibition, did not result in reversal of this inhibition (extended data, Figure S2) (Joshi et al., 2018), indicating this plant metabolite to act as a signal-response inhibitor.

GA was tested at 10-200 μg/ml for its potential quorum modulatory action against C. violaceum, and it was found to inhibit the growth of this bacterium in a dose-dependent fashion (Figure 6A). Violacein production was negatively affected from 100 μg/ml onwards, and this inhibition was reversed upon external AHL supply (Figure 6B). Since GA was indicated to act as a signal-supply inhibitor, we docked it against the LuxI analogue of C. violaceum i.e. CviI, the enzyme responsible for AHL synthesis, wherein the docking score was found to be -6.9 kcal/mol. At all test concentrations GA was able to induce catalase activity marginally.

Figure 6. Effect of GA, quercetin, and catechin on C. violaceum.

Bacterial growth was measured as OD₇₆₄; OD of violacein was measured at 585 nm, and Violacein Unit was calculated as the ratio OD₅₈₅/OD₇₆₄ (an indication of violacein production per unit of growth) ‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and violacein production of C. violaceum. (A) Effect of gallic acid on growth and violacein production in C. violaceum. (B) GA acts as a signal-supply inhibitor. (C) Effect of quercetin on growth and violacein production in C. violaceum. (D) Quercetin acts as a signal-response inhibitor. (E) Effect of catechin on growth and violacein production in C. violaceum. (F) Catechin acts as a signal-response inhibitor. *p<0.05, **p<0.01, ***p<0.001. QS, quorum sensing; GA, gallic acid.

When C. violaceum was challenged with quercetin (10–200 μg/ml), its growth was negatively affected, and violacein production even more so. Concentration of 150 μg/ml quercetin was able to achieve almost complete inhibition of violacein production (Figure 6C), and this inhibition was not reversible in response to exogenous AHL augmentation (Figure 6D), indicating quercetin to act as a signal-response inhibitor. This observation also matches with the docking score (-8.5 Kcal/mol) of quercetin against CviR, which is second highest amongst all the docked compounds. Quercetin was also able to curb the catalase activity of C. violaceum, maximum inhibition being observed at 150 μg/ml.

Catechin had no effect on C. violaceum growth till 200 μg/ml, whereas its quorum-inhibitory effect started from 10 μg/ml onwards. Hence, catechin can be said to have a purely quorum-inhibitory effect on C. violaceum (Figure 6E), and its inhibitory effect on violacein production was not reversed upon AHL augmentation, indicating it to be acting as a signal-response inhibitor (Figure 6F). This compound could notably reduce haemolytic potential of C. violaceum, and also had little but statistically significant effect on catalase activity (Table 3). Irrespective of its presence in PGPE, we employed catechin in all our QS experiments as a positive control, since it is a known QS inhibitor (Vandeputte et al., 2010). Information on its in vivo efficacy has been provided in respective figure legends (Figure 1D, Figure 2F, Figure 5D).

Pure compounds against P. aeruginosa. With respect to docking score against lasR receptor of P. aeruginosa, catechin, querectin and genistein exhibited highest affinity towards this receptor; whereas punicalagin, punicalin, rutin, and pedunculagin exhibited maximum affinity for the pqsR receptor of this bacterium. Preliminary in vitro experiments revealed cinnamic acid, chlorogenic acid, and catechin to exert their maximum effect on P. aeruginosa QS-regulated pigment production at different concentrations. Data for the most effective concentration for each of these three compounds is presented in Figure 7A–C. Production of both pigments was affected the most by chlorogenic acid, and this compound was also able to make P. aeruginosa more susceptible to cephalexin (Figure 7E); however, it could not alter this bacterium’s susceptibility to ofloxacin and gentamicin. With respect to antibiotic susceptibility result obtained with cinnamic acid (Figure 7E) were identical to those with chlorogenic acid. AHL augmentation was able to reverse effect of chlorogenic as well as cinnamic acid on production of both the QS-regulated pigments of P. aeruginosa (Figure 7D), suggesting these compounds to disturb the QS machinery of this pathogen by acting as signal-supply inhibitors. Both these compounds were able to negate haemolytic ability of P. aeruginosa notably (19.11%; p =0.04 and 35.29%; p= 8.31363E-05 reduction respectively by cinnamic acid and chlorogenic acid), whereas catalase activity was enhanced to a marginal (1.47%) extent only by chlorogenic acid. In vivo efficacy of chlorogenic acid was found to be better than that of cinnamic acid (Figure 7F).

Figure 7. Effect of pure compounds on QS-regulated traits of P. aeruginosa.

Bacterial growth was measured as OD₇₆₄; OD of pyoverdine was measured at 405 nm, and Pyoverdine Unit was calculated as the ratio OD₄₀₅/OD₇₆₄ (an indication of pyoverdine production per unit of growth), Pyocyanin Unit was calculated as the ratio OD₅₂₀/OD₇₆₄ (an indication of pyocyanin production per unit of growth). ‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and pigment production of P. aeruginosa. (A) Effect of cinnamic acid on growth and QS-regulated pigment production in P. aeruginosa. (B) Effect of chlorogenic acid on growth and QS-regulated pigment production in P. aeruginosa. (C) Effect of catechin on growth and QS-regulated pigment production in P. aeruginosa. (D) Cinnamic acid and chlorogenic acid act as signal-supply inhibitors against P. aeruginosa, whereas catechin acts as a signal-response inhibitor. (E) P. aeruginosa receiving pre-incubation with cinnamic acid or chlorogenic acid becomes more susceptible to cephalexin. (F) Pre-treatment of P. aeruginosa with cinnamic acid or chlorogenic acid reduces its virulence towards C. elegans: Catechin (50 μg/ml) and gentamicin (0.1 μg/ml) employed as positive controls conferred 100% and 80% protection on worm population respectively; DMSO present in the ‘vehicle control’ at 0.5%v/v did not affect virulence of the bacterium towards C. elegans; DMSO (0.5%v/v) and compounds at tested concentrations showed no toxicity towards the worm. *p<0.05, **p<0.01, ***p<0.001; AS, Antibiotic susceptibility; QS, Quorum sensing.

Catechin could reduce production of pyocyanin and pyoverdine both, in P. aeruginosa, without affecting its growth (Figure 7C). However, there was not much difference in the effects exerted by different concentrations of catechin. For example, the effect on pyoverdine production exerted by all concentrations in the range 10–100 μg/ml was statistically same; and so was the case for the effect exerted by catechin on pyocyanin production in the concentration range of 100-200 μg/ml. Catalase and haemolytic activity of P. aeruginosa were both negatively affected by catechin (Table 3). Catechin seemed to act as a signal-response inhibitor against P. aeruginosa (Figure 7D).

GA and quercetin were tested against P. aeruginosa at 10–200 μg/ml. Both these compounds were able to inhibit the bacterial growth, and this growth inhibitory effect was more profound in case of quercetin. Production of both the QS-regulated pigments in P. aeruginosa was enhanced by GA. Quercetin had a negative effect on pyocyanin production, whereas pyoverdine production was enhanced by it from 10 μg/ml onwards. Catalase activity remained unaffected in GA-treated P. aeruginosa culture, whereas it was affected to a minor extent in quercetin-treated culture. Both of these plant compounds were able to reduce biofilm-forming capacity of P. aeruginosa notably (Figure 8A, B).

Figure 8. Effect of GA and quercetin on various traits of P. aeruginosa, in vitro.

‘Control’ in this figure is the vehicle control (0.5%v/v DMSO), which did not exert any effect on growth and pigment production of P. aeruginosa; Bacterial growth was measured as OD₇₆₄; OD of pyoverdine was measured at 405 nm, and Pyoverdine Unit was calculated as the ratio OD₄₀₅/OD₇₆₄ (an indication of pyoverdine production per unit of growth), Pyocyanin Unit was calculated as the ratio OD₅₂₀/OD₇₆₄ (an indication of pyocyanin production per unit of growth); Crystal violet assay was performed to measure biofilm formation, and biofilm eradication, followed by the measurement of OD at 580 nm. (A) Effect of GA on growth and QS-regulated pigment production in P. aeruginosa. (B) Effect of quercetin on growth and QS-regulated pigment production in P. aeruginosa. *p<0.05, **p<0.01, ***p<0.001; AS, antibiotic susceptibility; QS, quorum sensing; GA, gallic acid.

Growth promoting effect of PGPE on probiotic strains. PGPE was able to enhance growth of L. plantarum and B. bifidum at 10–50 μg/ml. Higher concentration (100 μg/ml) did not seem to promote their growth further (Figure 9).

Figure 9. Growth promoting effect of Punica granatum peel extract on probiotic strains.

Bacterial growth was measured as OD₆₆₀. *p<0.05, **p<0.01, ***p<0.001. PGPE: Punica granatum peel extract.

Underlying data

All raw data containing the underlying data behind each figure and table are available on figshare. DOI: https://doi.org/10.6084/m9.figshare.7471979 (Joshi et al., 2018).

Extended data

Extended data are available on figshare. DOI: https://doi.org/10.6084/m9.figshare.7471979 (Joshi et al., 2018).

Figure S1. Antipathogenic effect of Pomella on P. aeruginosa.

Figure S2. Quorum-inhibitory effect of cinnamic and chlorogenic acid on violacein production in C. violaceum is not reversed following augmentation of the quorum-inhibited culture by AHL

Table S1. Binding affinity of different phytocompounds when docked against LuxR homologue of test pathogens.

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).