Antibacterial and Antibiofilm Activities of Nonpolar Extracts of Allium stipitatum Regel. against Multidrug Resistant Bacteria

The present study assessed the in vitro antibacterial and antibiofilm potential of hexane (ASHE) and dichloromethane (ASDE) extracts of Allium stipitatum (Persian shallot) against planktonic cells and biofilm structures of clinically significant antibiotic resistant pathogens, with a special emphasis on methicillin-sensitive Staphylococcus aureus (MSSA), methicillin-resistant S. aureus (MRSA), and emerging pathogens, Acinetobacter baumannii and Stenotrophomonas maltophilia. Antibacterial activities were determined through disk diffusion, minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC), time-kill kinetics, and electron microscopy. Antibiofilm activity was assessed by XTT [2,3-bis(2-methoxy-4-nitro-5-sulfo-phenyl)-2H-tetrazolium-5-carboxanilide] reduction assay and by confocal laser scanning microscopy (CLSM). The zone of inhibition ranged from 13 to 33 mm, while the MICs and MBCs ranged from 16 to 1024 μg mL−1. Both ASHE and ASDE completely eradicated overnight cultures of the test microorganisms, including antibiotic resistant strains. Time-kill studies showed that the extracts were strongly bactericidal against planktonic cultures of S. aureus, MRSA, Acinetobacter baumannii, and S. maltophilia as early as 4 hours postinoculation (hpi). ASHE and ASDE were shown to inhibit preformed biofilms of the four biofilm phenotypes tested. Our results demonstrate the potential therapeutic application of ASHE and ASDE to inhibit the growth of gram-positive and gram-negative biofilms of clinical significance and warrant further investigation of the potential of A. stipitatum bulbs against biofilm-related drug resistance.


Introduction
Biofilms are ubiquitous in nature which can be found at the bottom of streams to extremely hot waters of hot springs. In human health, biofilms pose a great threat due to their resistant nature to antibiotics, contamination of indwelling medical devices, forming dental plaques causing tooth decay, chronicity in wounds, and association with several illnesses from cystic fibrosis to otitis media. Biofilm forms of microbes are up to 1000-fold more resistant to antibiotics than their planktonic counterparts [1,2].
Infectious diseases caused by multidrug resistant (MDR) pathogens are the biggest challenge in healthcare. The conditions are made more worrisome by the biofilm associated infections. Such diseases are most frequently caused by S. aureus, S. epidermidis, Pseudomonas aeruginosa, A. baumannii, and Enterobacteria such as Escherichia coli [3][4][5].
Few novel antibacterial agents have been developed in recent years and their bacteriostatic or bactericidal activity results in selective pressure, with antimicrobial resistance as an unavoidable aftermath of their use [6]. In addition, effective therapies which target bacterial biofilms are scarce due to their implicit resistance to antibiotics [7]. An extensive review by Wu et al. [8] has shared the clinical experiences/challenges in managing biofilm infections. If the infections are not involving foreign body, a high dose of antibiotics or combinational therapy may sometimes eliminate the infections. But in cases where foreign bodies such as catheters, shunts, heart valves, pacemakers, stents, breast implants, intrauterine devices, contact lenses, etc. are used, the removal of material, followed by treatment is often necessary.
To date, several agents from natural products to synthetic compounds have been routinely used or evaluated for treating biofilm infections [9][10][11][12][13]; however, the integrity of the cells in the biofilm matrix and reduced permeability to antibiotics limit the complete detachment or destruction of biofilms. Hence, the search for new antimicrobials and biofilm dissolving agents is always vital in the human medicine.
Allium stipitatum that belongs to the Amaryllidaceae family is a wild edible plant mostly found in the cold mountains of central, south, and western Iran as well as some provinces of Turkey and central Asia [14,15]. Persian shallot is locally known as "Mooser" in Iran and widely used as a spice and flavoring agent in Persian foods, especially as added ingredients to yogurt, salads, and pickling mixtures. The appearance, shape, color, weight, texture, and storage tissues differentiate A. stipitatum from the common shallot, A. ascalonicum [15,16]. Unlike other Allium members, Persian shallot usually consists of a single bulb and rarely two bulbs [15]. Dried bulbs of A. stipitatum are frequently used in Persian folklore medicine for various ailments, like anti-inflammatory disorders, diarrhoea, gout, haemorrhoids, psoriasis, rheumatic arthritis, stomach pain, etc. [17]. Several studies performed on A. stipitatum elucidated its antibacterial, antiproliferative, anthelmintic, antiprotozoal, immunomodulatory, and wound healing properties [18][19][20][21][22][23]. Moreover, in our recent study, we have shown that A. stipitatum exerts anti-MRSA and wound healing activity in a mouse model of burn wound and infection [24]. We have also reported that both the extracts were relatively noncytotoxic towards mammalian cells and effective in eradicating MRSA colonized in thermal wounds. However, the effect of A. stipitatum on bacterial biofilm formation is still not clearly defined, requiring further investigation. In this study, we investigated the effects of ASHE and ASDE against a panel of medically important gram-positive and gram-negative bacteria followed by SEM and TEM examination of the in vitro effects of ASHE and ASDE on bacterial cells at different concentrations. We further provide evidence that ASHE and ASDE can increase the susceptibility of bacterial biofilms with emphasis on MSSA, MRSA, A. baumannii, and S. maltophilia biofilms. were obtained from the Medical Microbiology and Parasitology Laboratory at Universiti Putra Malaysia (UPM). All strains were confirmed by cultural and biochemical characteristics and maintained in glycerol stock cultures at -80 ∘ C prior to use. Bacterial cultures were propagated by streaking onto tryptic soy agar (TSA) or nutrient agar (NA). Single colonies of bacteria from the overnight cultures were inoculated into Luria-Bertani (LB) broth or brain heart infusion broth (BHI) and incubated in a shaking incubator at 37 ∘ C.

Disk Diffusion Assay and Determination of MICs and
MBCs. The antibacterial activities of ASHE and ASDE were assessed by disk diffusion method as previously described [25]. For disk diffusion assay, ASHE and ASDE were prepared freshly at a concentration of 10 mg mL −1 in 10% DMSO. Sterile antibiotic assay filter paper discs of 6 mm diameter were placed on MHA plates and 20 L (10 mg mL −1 corresponding to 200 g extract) of ASHE or ASDE was loaded onto the filter paper discs. Appropriate antibiotics were included as positive controls, while filter paper disc loaded with 20 L of DMSO (10%) was included as a negative control (diluent control).
The plates were incubated at 37 ∘ C and the inhibition zones were measured after 24 h incubation. The MICs and MBCs of ASHE and ASDE were determined by broth microdilution method and spread plate technique, respectively, by following the previously described methods [24,26].

Time-Kill
Assay. Time-kill assay was carried out on selected bacterial strains (MSSA, MRSA, A. baumannii, and S. maltophilia) according to the method described previously with minor modifications [26]. Bacterial suspensions were diluted to 1 × 10 6 CFU mL −1 . ASHE and ASDE concentrations were adjusted to 1x, 2x, and 4x MICs. Bacterial cultures treated with varying concentrations of the extracts were incubated at 37 ∘ C for 0, 0.5, 1, 2, 4, 8, 12, and 24 h. Aliquots of 100 L were pipetted out from each tube at each time point, serially diluted in PBS, and spread-plated onto MHA plates. Tubes without ASHE and ASDE served as growth controls (0x). The plates were incubated at 37 ∘ C for 24 h followed by the enumeration of bacterial colonies. Killing curves were constructed by plotting the log 10 CFU mL −1 versus time over a 24 h time period. Bactericidal activity (99.9% kill) was defined as a ≥3-log 10 CFU mL −1 reduction in colony count from the initial inoculum.

Ultrastructure Microscopical Analysis.
The ultrastructural changes in the bacteria treated with ASHE/ASDE were examined by SEM and TEM. Bacterial samples for electron microscopy were prepared by following the previously described method [27] with minor modifications. Cultures of S. aureus, MRSA, A. baumannii, and S. maltophilia were grown in TSB or BHI broth for 24 h at 37 ∘ C. The cultures were diluted to a final concentration of 5 × 10 7 CFU mL −1 in broth (1:10) and aliquots of 5 mL were placed into each well of a 6-well polystyrene tissue culture plate. Samples were treated with varying concentrations of ASHE and ASDE (1x, 2x, and 4x MICs) for 4 h (based on the results obtained in time-kill studies) and cells were harvested by centrifugation and washed twice with 0.1 M PBS (pH 7.2) before proceeding for SEM and TEM analysis.

Transmission Electron Microscopy (TEM).
Bacterial cells were fixed with glutaraldehyde and postfixed similar to the sample preparation as described for SEM in Section 2.5.1. Samples were dehydrated with a series of acetone grade (35%, 50%, 75%, 95%, and 100%) for 10-15 min each and infiltrated with increasing concentrations of acetone:resin mixture. Epoxy resin embedded samples were subjected to ultramicrotome and the ultrathin sections were doublestained with uranyl acetate and lead citrate. The specimens were examined under a Hitachi H-7100 TEM (Hitachi, Ibaragi) at 120 kV.

Antibiofilm Effect of ASHE and ASDE
2.6.1. Antibiofilm Assay. Biofilms of S. aureus, MRSA, A. baumannii, and S. maltophilia were produced by microtitre plate method [28]. Briefly, overnight broth cultures of the bacterial samples were grown in BHI broth to a turbidity equivalent to 0.5 McFarland standard (10 8 CFU mL −1 ). Each biofilm phenotype was added to 24 wells of a sterile microtitre plate and incubated at 37 ∘ C for 6 h under static conditions. After 6 h of adhesion, nonadherent cells were removed from each well and the wells were rinsed with 100 L of physiological saline, and subsequently 100 L of fresh medium was added to each well and incubated for 24 h. After 24 h of adhesion and biofilm formation, the supernatant was again removed and the wells were rinsed with physiological saline and 100 L of ASHE and ASDE at concentrations of 1x, 2x, and 4x MICs were added. Wells without antibiotics or extracts were considered "untreated". Untreated cells/negative controls were incubated with 100 L of DMSO (5%) for an additional 12 h at 37 ∘ C with gentle shaking and the viability of biofilms was quantified by XTTcalorimetric assay.

XTT Reduction Assay.
The antibiofilm potential of ASHE and ASDE to inactivate biofilms was assessed by microtitre plate assay [29]. Before each assay, fresh XTT solutions were prepared by dissolving 4 mg of XTT (Sigma) in 10 mL prewarmed (37 ∘ C) PBS. This solution was supplemented with 100 L menadione stock solution, containing 55 mg menadione (Sigma) in 100 mL acetone. The effect of ASHE and ASDE was tested at concentrations of 0x (untreated/negative control), 1x, 2x, and 4x MICs against 24 h old biofilms formed earlier. Posttreated biofilm plates were washed thrice with 200 L of sterile PBS. The wells were dried and 100 L of XTT-menadione solution was added to each well and incubated at 37 ∘ C in dark for 5 h. The contents of the wells were pipetted into fresh 1.5 mL Eppendorf tubes and centrifuged at 15,000 g for 4 min. One hundred microlitres of the clear supernatant from each well was transferred to a sterile 96-well flat-bottomed microtitre plate and the absorbance of the adherent biofilm was read at 490 nm in a microplate reader (BioTek EL808, USA).

In Situ Visualization of the Antibiofilm Effects of ASHE and ASDE by Confocal Laser Scanning Microscopy (CLSM).
In situ antibiofilm potential of ASHE and ASDE on preformed biofilms of S. aureus, MRSA, A. baumannii, and S. maltophilia was visualized under a CLSM (Olympus FV1000-IX81). Biofilms were carefully washed with PBS and stained with LIVE/DEAD5BacLight Bacterial Viability Kit (Invitrogen, Grand Island, NY, USA). Equal volumes of live stain (green fluorescence) and dead stain (red fluorescence) were mixed and diluted to a working solution of 0.3% in sterile distilled water. The staining procedure was carried out according to the method described previously [30]. CLSM images were captured and processed using a Fluoview5 FV1000 live cell imaging software.

Statistical Analysis.
All assays were performed in triplicate (at a minimum) and repeated three times. Statistical analyses were performed using GraphPad Prism 6.0 software (GraphPad, San Diego, CA, USA) by one-way analysis of variance (ANOVA) with Dunnett's multiple comparison test. Values are expressed as ± SD. * * * < 0.001 compared with the control.

Antibacterial Activities of ASHE and ASDE.
Both ASHE and ASDE showed promising antibacterial activity against all the bacteria tested and the inhibition zones ranged from 13 to 33 mm for ASHE and 15 to 32 mm for ASDE at 10 mg mL −1 concentration (Table 1). Both gram-positive and gram-negative organisms were susceptible to ASHE and ASDE. Of the 12 organisms tested, MDR pathogens, namely, A. baumannii, MRSA, and S. maltophilia, were highly susceptible to ASHE and ASDE with zone sizes ranging from 23 to 28 mm. The antibacterial results of MRSA were represented in this manuscript in order to facilitate better understanding for the readers. Meanwhile the diameter zone of inhibition for VRE and P. aeruginosa ranged from 13 to 15 mm ( Figure 1). The largest zone of inhibition was observed in A. lwoffii exerted by ASHE (33 mm) at a concentration of 200 g/disc. The MICs of ASHE and ASDE ranged from 16 to 1024 g mL −1 , while the MBCs ranged between 32 and 4096 g mL −1 (Table 1). MSSA, MRSA, and A. lwoffii were found to be highly susceptible, especially at very low concentrations.

Analysis of Bacterial Killing Kinetics.
Studies on the bacterial killing kinetics of ASHE and ASDE on MSSA, MRSA, A. baumannii, and S. maltophilia showed that the growth controls in MHB without any antibiotics (untreated) maintained their viability for 24 h. However, MSSA, MRSA, A. baumannii, and S. maltophilia treated with ASHE and ASDE at 1x, 2x, and 4x MICs showed a significant reduction in growth, indicating that both extracts were strongly bactericidal by killing >90% of the cells. Bactericidal endpoints were achieved at 2 h for MRSA (Figures 2(b) and 3(b)) and 4 h for MSSA (Figures 2(a) and 3(a)), A. baumannii (Figures 2(c) and 3(c)), and S. maltophilia (Figures 2(d) and 3(d)). At 1× and 4× MICs, the average reduction in CFUs was found to be ∼1.9 and ∼2.5 log units for ASHE and ASDE, respectively (99.9%).  Figure 7 displays a series of CLSM images of the biofilms, before and after ASHE and ASDE treatments at varying concentrations. The controls (biofilms and DMSO treated) images on the 1 st and 2 nd columns as shown by the typical two-dimensional biofilm architectures of MSSA (Figures 7(a) and 7(b)), MRSA (Figures 7(i) and 7(j)), A. baumannii (Figures 7(q) and 7(r)), and S. maltophilia (Figures 7(y) and 7(z)) displayed well-developed biofilms (Figure 7), while bacterial samples treated with ASHE and ASDE at 1x MICs showed significant reduction in biofilms (Figures 7(a), 7(f), 7(k), 7(n), 7(s), 7(v), Table 1: Zone of inhibition, MIC, and MBC of ASHE and ASDE against test microorganisms (20 L corresponding to 200 g/disc).

Discussion
This is the first report on two different extracts from the same botanical product with antibacterial activity against MSSA, MRSA, A. baumannii, and S. maltophilia biofilms and their planktonic counterparts. The presence of allicin in the bulbs of A. hirtifolium (also known as Persian shallot locally) has been demonstrated by thin layer chromatography (TLC) methods [20]. In our earlier investigation on the in vivo antibacterial and burn wound healing properties of A. stipitatum, both ASHE and ASDE were subjected to gas chromatography mass spectrometer (GC-MS) analysis [24]. Although the antibacterial compound allicin was not detected in the GC-MS analysis, several sulfur-containing compounds including S-methyl methanethiosulfonate, 2,4,5-trithiahexane, 2,4-dithiapentane, 2-pyridinethione, and methane (chloromethylthio) (methylthio) were detected in our recently published work [24]. Similar components were also reported in the hydromethanolic extract of A. hirtifolium as detected by GMCS [21]. In their analysis, the hydromethanolic extract of A. hirtifolium was reported to be effective against MSSA, MRSA, S. typhi, E. coli, and K.
pneumoniae with zone sizes ranging from 10 to 18 mm at 1.2 mg/disc concentration [21]. However, the use of nonpolar extracts (ASHE and ASDE) in this study resulted in a slightly higher activity. The strong antibacterial activity of ASHE and ASHE against the test pathogens can be attributed to the presence of volatile sulfur-containing compounds in the extracts as compared to the amount detected in the hydromethanolic extract reported elsewhere [21]. Another significance of the present study is the relatively equal potency exerted by ASHE and ASDE especially in killing gram-negative pathogens. Preliminary antibacterial data on the anti-MRSA activity of ASHE and ASDE based on disk diffusion and MIC experiments has been published recently by our research group [24]. Therefore, the disk diffusion plate picture for MRSA was not shown in Figure 1. However, the time-kill assay, electron microscopy images, and antibiofilm data of MRSA were included in the present study. Ismail et al. [21] reported that the inhibition zones exerted by hydromethanolic extract at 60 mg mL −1 towards grampositive and gram-negative pathogens were in the ranges of 11-15 mm and 10-14 mm diameter, respectively. However, in the present study, ASHE at 10 mg mL −1 exerted zone sizes of 13-27 mm towards gram-positive bacteria and 13-33 mm towards gram-negative bacteria. Meanwhile, ASDE exerted zone sizes of 15-30 mm towards gram-positive bacteria and 15-32 mm towards gram-negative bacteria. Another significance of the present study is the use of emerging antibiotic resistant pathogens such as A. baumannii, S. maltophilia, and VRE which were highly susceptible to ASHE and ASDE. Based on the above data, it is evident that both ASHE and ASDE were equally potent in inhibiting both gram-positive and emerging gram-negative bacterial pathogens even at very low concentrations. According to a previous study, the MIC values of hydromethanolic extract ranged from 1.88 to 7.50 mg mL for gram-positive and gram-negative pathogens [21]. In the present study, ASHE and ASDE were strongly antibacterial and the MICs were as low as 32 g mL −1 against epidemiologically important superbug like MRSA. The MICs of ASHE and ASDE were in the range of 16-1024 g mL −1 , while the MBCs ranged between 32 and 4096 g mL −1 . Such low levels of MICs for a natural product extract against emerging antibiotic resistant pathogens are noteworthy and advantageous compared with compounds of synthetic origin. Few reports are available on the nontoxic antibacterial and cytoprotective effects of chloroformic and hydromethanolic extracts of A. hirtifolium [19,21]. Moreover, in our earlier investigation on the anti-MRSA and burn wound healing properties of A. stipitatum, both extracts (ASHE and ASDE) were proven to be noncytotoxic and safe to Vero cells at <400 g mL −1 [24]. In terms of antibacterial activity, the MICs of ASHE and ASDE against all grampositive (except VRE) and gram-negative bacteria (except Enterobacter spp. and P. aeruginosa) were in the ranges of 16-256 g mL −1 and 16-128 g mL −1 , respectively. The above MIC values were comparatively lesser than the CC 50 concentrations on Vero cells (383.4 g mL −1 ; 390.4 g mL −1 ) which is therapeutically acceptable for a nontoxic edible natural extract. The cytotoxicity results as reported in our previous study [24] and the strong antibacterial activity of ASHE and ASDE at low concentrations clearly underline that nonpolar extracts of A. stipitatum can be successfully used as promising alternatives in managing gram-positive and gramnegative pathogens.
Several studies have reported on the preliminary antibacterial activity of A. hirtifolium [21,31,32]; however, investigation on the bactericidal effect of Persian shallot extract has never been reported elsewhere. Determination of the minimum time required by ASHE/ASDE in killing a pathogen will help in investigating the mechanism of action which involves studies like transcriptomic, gene expression analysis, and enzyme assays. Based on the data on time-kill assay, cell counts were found to be either too few to count (TFTC) or zero in MHA plates with bacterial inoculum plated after 4 h postinoculation. The colony counts of test bacteria treated with ASHE and ASDE at 2x and 4x MICs varied, but interestingly the killing time was the same for all the concentrations tested. The killing efficiency of both ASHE and ASDE towards S. aureus, MRSA, A. baumannii, and S. maltophilia was comparatively similar with slight variations in colony counts which could be due to the solvent's (dichloromethane) nature as a chlorinated solvent. No evidence of recurrence or growth was observed among the 4 pathogens tested which imply that the extracts are strongly bactericidal by completely killing the test pathogens in 2-4 hpi with an ∼ 2 to 2.5 log reduction in the inoculum. The above findings are partly in agreement with Nidadavolu et al. [33] who reported ∼ 7 log reduction in A. baumannii, ∼ 8 log reduction in S. aureus, and ∼ 2 log reduction in Enterococcus faecalis biofilms upon treatment with garlic oil. In addition, MDR strains of A. baumannii have been reported to be highly susceptible to chloroform extract of garlic [34]. Increasing concentrations of the extracts (2x and 4x MICs) did not show significant changes in the killing time, but the colony counts varied slightly which implies that the killing efficacy of ASHE and ASDE was concentration dependent. Biofilm-related casualties due to S. aureus, MRSA, A. baumannii, and S. maltophilia are becoming more persistent in susceptible hospitalized patients with indwelling catheters or artificial grafts [35][36][37]. In the present study, S. aureus and S. maltophilia biofilms were more susceptible to ASHE and ASDE especially at 1x MIC levels (p< 0.001, p< 0.001) as compared to MRSA and A. baumannii (p< 0.001, p< 0.001), but statistically significant reduction in biofilm reduction was evident. One limitation of the present study is that the antibiofilm activities of ASHE and ASDE were tested only on mature biofilms (24-hour-old biofilm) and not on early    biofilms (6-hour-old partial biofilm). In fact, mature biofilms are dense, thicker, and increasingly resistant to antibiotics as compared to early biofilms at 6 h. Both extracts were equally potential in penetrating deeper layers of the mature biofilms and effective in killing bacterial biofilms. Based on these results, it is evident that in addition to the broad-spectrum antibacterial activity towards planktonic bacteria, ASHE and ASDE possess antibiofilm activity on gram-positive and gram-negative biofilms. In an earlier investigation, Lee et al. [38] reported that garlic extract increases the biofilm formation by the oral biofilm colonizer Streptococcus mutans to orthodontic wire. However, the findings of the present study are contradictory, where A. stipitatum being a member of the Allium family disrupted matured biofilms of the MDR pathogens. According to Lee et al. [38], garlic oil (GarO) was shown to prevent biofilm development in burn wound pathogens such as S. aureus and A. baumannii [33]. These results were in agreement with the present study results implying that ASHE and ASDE could be possibly used as a prophylactic therapy to prevent biofilm associated infections upon further investigations.

S. maltophilia
Bacterial cell wall and cell membrane play a vital role in the survival, biofilm formation, and antibiotic resistance of a pathogen. Hence, a slight damage induced by a natural/synthetic antimicrobial compound could result in metabolic dysfunction/cell death. SEM and TEM experiments are highly essential techniques [27] in order to reveal any possible cell surface effects and/or intracellular alterations induced by ASHE/ASDE. It is evident from the SEM experiments that both gram-positive and gram-negative organisms tested showed signs of deep craters, burst, rough and damaged surfaces with depression, rupture lines, and cell debris. TEM experiments also showed clear disruption of bacterial membrane and leakage, which emphasize the bactericidal potentials of A. stipitatum.
Effective disruption and reduction of microcolonies were evidenced by CLSM images (Figure 7). The exact molecular mechanism underlying the growth and biofilm inhibition by ASHE and ASDE in MDR pathogens is not known; however, a recent investigation on the antibacterial effects of garlic (Allium sativum) concentrate and garlic-derived organosulfur compounds targets the cell membrane of Campylobacter jejuni [39]. Diallyl thiosulfinate (allicin) an active ingredient in garlic is known to inhibit RNA synthesis in S. typhimurium [40]. The use of crude plant extracts like ASHE/ASDE may be a valuable tool for future developments, also keeping in