Myristica lowiana Phytochemicals as Inhibitor of Plasmid Conjugation in Escherichia coli

Hexane extract and methanol fraction from the stem bark of Myristica lowiana specifically and significantly inhibited the conjugal transfer of the IncW plasmid R7K, a plasmid which harbors ampicillin-, streptomycin-, and spectinomycin-resistant genes. The transfer of this plasmid via the conjugative pilli of Escherichia coli was reduced by 76.5 ± 2.0% and 79.0 ± 1.2% by hexane extract and methanol fraction of M. lowiana, respectively. The hexane extract exhibited significant anti-conjugant activity at a non-cytotoxic concentration of 100 mg/L as assessed against adult human dermal fibroblast cells. The hexane extract and methanol fraction were screened using phytochemical tests, NMR spectroscopy, IR spectroscopy, and high-resolution electrospray ionization mass spectrometry (HRESIMS) and were found to contain terpenoids, sterols, and fatty acids.


Introduction
Currently, antibiotic resistance is a major public health concern, especially since to date there is no known universal method for reversing drug resistance in microbes and the world is running out of effective antimicrobials [1,2]. ere is considerable evidence of resistance against almost all existing classes of antimicrobial agents, namely, the β-lactams, aminoglycosides, chloramphenicol, glycopeptides, quinolones, oxalidinones, sulfonamides, tetracyclines, macrolides, ansamycins, streptogramins, and lipopeptides [3]. Additionally, it has been predicted that if no antimicrobials or means of reversing antimicrobial resistance are found soon, the death toll for antimicrobial-resistant associated infections may rise to 300 million by 2050 [4]. is illustrates the gravity of antibiotic resistance and the urgent need to address the problem.
All existing antimicrobial agents interfere with essential bacterial growth and metabolic processes or the integrity of the cell. Unfortunately, targeting such vital functions introduces selection pressure for resistant strains and promotes evolution of resistance. e evidence for this is apparent with the rise of antibiotic resistance in hospitals and communities, most notably with organisms such as methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus, carbapenem-resistant Enterobacteriaceae (including Escherichia coli), multidrug-resistant Acinetobacter, drug-resistant Campylobacter, multidrug-resistant Pseudomonas aeruginosa, drug-resistant Streptococcus pneumoniae, drug-resistant Neisseria gonorrhoeae, and multidrug-resistant and extensive drug-resistant Mycobacterium tuberculosis [5].
A complementary approach of tackling antibiotic or multidrug resistance is to target bacterial adaptation and persistence mechanisms such as horizontal DNA transfer, mutation, antibiotic tolerance, and the production of virulence factors. Such mechanisms are known to be nonessential to the growth of bacteria and as a consequence of this, they are not expected to extensively promote evolution of bacterial resistance against such complementary mediators [6]. e approach of inhibiting bacterial adaptation and persistence mechanisms and the progress made so far have been reported by Fernandez-Lopez et. al. [7], Smith and Romesberg [6], Getino et al. [8], and Getino et al. [9].
In this study, extract and fraction from Myristica lowiana, a dioecious flowering tree of the family Myristicaceae, were investigated for their capacities to inhibit plasmid conjugation in E. coli a process that utilizes bacterial type-IV secretion systems to transfer genetic material, including antimicrobial resistance genes, between donor and recipient cells [10]. In addition, the extract and fraction were investigated for their antibacterial activities, cytotoxicity, and their phytochemical constituents.
M. lowiana is commonly found in Southeast Asian riverine forests. It usually grows up to 7-25 meters in height and has a dark chocolate or blackish coloured stem bark, buttress roots, lanceolate olivaceous or brown leaves, cream rusty coloured flowers, and ellipsoidal-shaped fruits and seeds [11]. M. lowiana has a characteristic watery red sap, which oozes out upon cutting its bark and wood. ere is no known medicinal use of this plant in the literature, and so far, it is used mainly for construction purposes. ere are also no phytochemical reports on this species, but Myristicaceae plants are well known for essential oils, coumarins, flavonoids, lignans, acylphenols, alkaloids, polyketides, and terpenoids [12][13][14][15]. Herein, we report on the hexane extract and methanol fraction of M. lowiana and their phytochemistry and biological activities.

Plant Material.
e stem bark of M. lowiana was collected from St. Matang, Malaysia, in May 2014. Stephen Teo, a member of the Forest Department Sarawak identified the species. A voucher specimen (ST001/14) of this tree has been deposited within the herbarium at the UCL School of Pharmacy.

Extraction and Fractionation.
Powdered M. lowiana stem bark (592 g) was subjected to ultrasound-assisted extraction and extracted successively with hexane, chloroform, and methanol. e powdered plant material was placed in a 5 L glass beaker and covered with the extracting solvent (2.5 L), hexane. is was then placed in an ultrasonic bath for 4 h. e resulting extract was filtered and dried under vacuum. e marc was then dried to remove all residual solvents. e procedure was repeated with the dried marc and chloroform and finally with methanol to obtain extracts of increasing polarity. is yielded a 0.7% hexane extract, a 1.1% chloroform extract, and a 2.2% methanol extract. e extracts were further fractionated using the solid phase extraction. Nonpolar extracts were subjected to normalphase solid phase extraction (Phenomenex ® Strata Silica S1) using a step gradient system, 10% chloroform increments in hexane. Polar extracts were subjected to reverse-phase solid phase extraction (Phenomenex ® Strata Silica C18) using a step gradient system, 10% methanol increments in water.

Phytochemical Screening.
e extract and fraction of M. lowiana were subjected to qualitative phytochemical screening as per the standard methods. For the test of sterols, the test samples were treated with chloroform and concentrated sulphuric acid. A red colouration at the upper layer and a yellow with green fluorescence at the sulphuric acid layer indicated the presence of sterols. For the test of terpenoids, the test samples were treated with acetic anhydride and concentrated sulphuric acid; formation of blue green rings indicated the presence of terpenoids. e biuret test was used to test for the presence of amide bonds; aqueous solutions of the test samples were treated with 2 drops of 2% copper (II) sulphate solution and 1 mL of 5% sodium hydroxide solution. A violet or bluish violet or pink colouration indicated the presence of amide bonds. For flavonoids, the test samples were treated with a few drops of 1% lead acetate; a yellowish precipitate indicated the presence of flavonoids. For saponins, the froth test was used. e test samples were diluted with distilled water and shaken; formation of foam indicated the presence of saponins. For reducing sugars, Benedict's test was used. For tannins, the gelatin test was used where the test samples were treated with 1% gelatin solution containing sodium chloride. Formation of a white precipitate indicated the presence of tannins.

General Experimental Procedures.
e 1D-and 2D-NMR spectra were recorded on a Bruker Avance 500 MHz spectrometer. IR spectra were recorded as a thin-film on a PerkinElmer Spectrum 100 FT-IR Spectrometer. Mass spectra were recorded on a Micromass Q-ToF Premier Tandem Mass Spectrometer. Bertani broth (LB), MacConkey agar, nutrient agar, Sabouraud dextrose agar, sodium chloride, sodium hydroxide, phosphate buffered saline (PBS) and human dermal fibroblasts, and adult (HDFa) cells were obtained from Fischer Scientific UK Ltd. Dulbecco's Modified Eagle's Medium (DMEM), Fetal Bovine Serum (FBS), and nonessential amino acids were obtained from Gibco. Staphylococcus aureus XU212, which possesses the TetK tetracycline efflux pump and mecA gene, was provided by Udo [16]. S. aureus 1199B, which possesses the NorA efflux pump, was provided by Kaatz [17]. Dr. Paul Stapleton, UCL School of Pharmacy, provided all other bacterial and fungal strains used in this study. 2 Evidence-Based Complementary and Alternative Medicine

Antibacterial Activity.
e antibacterial activity was determined using the broth microdilution method as described previously [18]. Bacterial strains were cultured on nutrient agar (Fluka Analytical) and incubated at 37°C for 18 hours. A bacterial suspension equivalent to 5 × 10 5 cfu/mL was added to MHB and the test sample, which had been serially diluted across the 96 well microtitre plate. MICs were determined after 18 hours of incubation at 37°C. With the addition of 20 μL of a 1 mg/mL methanolic solution of MTT, a colour change of the dye to purple indicated the presence of growth. e MIC was defined as the lowest concentration at which no bacterial growth was observed. MIC values were determined in duplicate per plate and repeated in at least two independent experiments.

Antifungal Activity.
e antifungal activity was determined using the broth dilution method as described previously [19]. Fungal strains were cultured on Sabouraud dextrose agar plates. A fungal suspension equivalent to 1-5 × 10 5 cfu/mL was added to supplemented RPMI-1640 and test sample, which had been serially diluted across the 96 well microtitre plate. Control compounds included the known antifungal agents itraconazole and amphotericin B.
e MIC was visually determined under light after incubation at 37°C for 24 hours and 48 hours for C. albicans and C. tropicalis, respectively. MIC values were determined in duplicate per plate and repeated in at least two independent experiments.

Bacterial Plasmid Conjugation Inhibition Assay (Liquid Conjugation).
e donor cells with plasmids pKM101, TP114, and pUB307 were mated with the recipient ER1793, while plasmid R7K donor cells were mated with the recipient JM109 in the assay. Conjugation was performed as described previously [20], with minor modifications. Twenty microliters of predetermined inocula (cfu/mL; Supplementary Table 1) of donor and recipient was mixed with 160 μL Luria Bertani (LB) broth and extract or control drug. is was incubated at 37°C for 18 hours after which the number of transconjugants and donor cells were determined using appropriate antibiotic-containing MacConkey agar plates. Linoleic acid, a known inhibitor of IncW plasmid conjugation, was used as a control in this experiment. e extract and fraction were evaluated for anti-conjugant activity at subinhibitory concentration (one-quarter of their MIC against E. coli NCTC 10418). Antibiotics and the concentrations used in MacConkey agar for positive identification of donors, recipients, and transconjugants (μg/mL) were amoxicillin (30), streptomycin sulphate (20), nalidixic acid (30), and kanamycin sulphate (20 and 30). Conjugation frequency was calculated as the ratio of total number of transconjugants (cfu/mL) to the total number of donor (cfu/ mL) and expressed as a percentage relative to the negative control. is experiment was performed as duplicate in three independent experiments and anticonjugation activity has been reported as mean ± standard deviation.

Cytotoxicity Activity.
e sulforhodamine B (SRB) assay as described previously [21] was adapted for the cytotoxicity screening. Adult human dermal fibroblast cells (HDFa; C-013-5C) were grown at 37°C in a humidified atmosphere of 5% in a culture flask (75 cm 2 ) which contained Dulbecco's Modified Eagle's Medium, modified with 10% FBS, 1% nonessential amino acids, and 0.1% gentamicin and amphotericin B. e grown cells were seeded in a 96 well microtitre plate, and a determined quantity of test samples was added. e samples were then incubated for 72 h at 37°C in 5% CO 2 . Afterwards, 50 μL of cold 40% W / V trichloroacetic acid (TCA) solution was added and incubated for an hour at 4°C and washed four times with distilled water. Cultured cells were then stained with 0.4% W / V SRB solution and left for an hour at room temperature. Afterwards, the plate was rinsed four times with 1% acetic acid and left for 24 h to dry. Prior to optical density (OD) determination at 510 nm using a microtitre plate reader (Tecan Infinite ® M200), 100 μL of 10 mM Tris buffer solution was dispensed into the wells and agitated in an orbital shaker for 5 min, to allow solubilisation of SRB-protein complex. e percentage of viable cell was calculated using the following formula: is experiment was performed as triplicate in three independent experiments, and cytotoxicity has been reported as mean ± standard deviation.

Statistical Analyses.
e statistical analyses were carried out using Excel Data Analysis and GraphPad Prism. Student's t-test was used to evaluate the difference between the control conjugal transfer frequency and the test compounds. Results with p < 0.05 were considered statistically significant.

Results and Discussion
In this study, extracts and fractions from Myristica lowiana were investigated for their capacities to inhibit plasmid conjugation in E. coli. Extract and fraction that showed potent anti-conjugant activity, that is, at least 70% reduction in the transfer frequency of the any of the test plasmids Evidence-Based Complementary and Alternative Medicine (pKM101, TP114, pUB304 and R7K) was further investigated for their antibacterial activities, cytotoxicity, and phytochemical constituents. e hexane fractions, chloroform extract and its fractions, and methanol extract and its other fractions did not exhibit potent anti-conjugant activity; hence, these products were not further investigated. Some of the aforementioned extracts and fractions also showed no conjugal inhibition, while others showed increase in the conjugation frequency of the test plasmids. e hexane extract and methanol fraction eluted with methanol and water (9 : 1, v/v) were the only products that exhibited potent anti-conjugant activity. e hexane extract (4.14 g) gave rise to an olive green liquid. Phytochemical screening of this extract revealed the presence of sterols, terpenoids, and amide bonds. However, the extract tested negative for tannins, saponins, reducing sugars, and flavonoids. e NMR spectra ( Supplementary  Figures 1 and 2), HRESIMS (Supplementary Figure 3), and IR spectrum (Supplementary Figure 4) of the hexane extract indicated the presence of aromatic and aliphatic groups attributable to aromatic compounds, terpenoids, and fatty acids. e methanol fraction (24.1 mg) was eluted with methanol and water (9 : 1, v/v) to yield a brown solid. Phytochemical screening of this fraction revealed the presence of tannins and amide bonds. is fraction tested negative for saponins, sterols, reducing sugars, flavonoids, and terpenoids. e NMR spectra (Supplementary Figures 5  and 6), HRESIMS (Supplementary Figure 7), and IR spectrum (Supplementary Figure 8) of the methanol fraction indicated the presence of aromatic and aliphatic groups attributable to aromatic compounds and fatty acids. e antibacterial activities (Table 1) revealed that both hexane extract and methanol fraction had moderate antibacterial activity (32 to 64 mg/L) against Gram-positive bacteria but were not active against Gram-negative organisms. We suggest that the difference in activities against Gram-positive and Gram-negative bacteria may be due to the differences in the cell wall and the additional outer membrane in Gram-negative organisms restricting cell entry or accumulation of components within the extract and fraction [22]. e presence of efflux pumps in Staphylococcus aureus strains SA-1199B (NorA pump) and XU212 (TetK pump) did not have any effect on the antibacterial activities of the hexane extract and methanol fraction. is indicated that antimicrobial compounds within the extract and fraction were not susceptible to these prevalent efflux pumps or that the target(s) lies externally to the cell. e extract and fraction were further screened at a subinhibitory concentration (100 mg/L) for anti-conjugal activity in Escherichia coli. is test concentration was used to ensure that any observed anti-conjugal activity may not be due to growth inhibition. Both hexane extract and methanol fraction showed significant anti-conjugal activity against the transfer of the IncW plasmid R7K, reducing the transfer frequency by 76.5 ± 2.0% and 79 ± 1.2%, respectively (Figure 1). e anti-conjugation activities of the extract and fraction were specific to the IncW plasmid R7K; they did not significantly inhibit the conjugal transfer of IncN plasmid pKM101, IncI 2 plasmid TP114, and IncP pUB307, rather the extract and fraction increased the conjugation frequency of plasmids pKM101 and TP114 by a further 20% or more. We therefore suspect that the extract and fraction might be targeting different sites on the conjugation machineries, which may be plasmid incompatibility group specific; hence, the difference in activities.
is is plausible, as published literature has shown component variations in the conjugation machineries and the mobilizable (MOB) and mating pair formation (Mpf ) systems between different plasmid incompatibility groups [8,9,23,24]. e cytotoxicity studies of hexane extract and methanol fraction against adult human dermal fibroblast cells (Supplementary Table 2) revealed that their cytotoxic-IC 50s were 154.0 mg/L and  Evidence-Based Complementary and Alternative Medicine 92.6 mg/L, respectively. A comparison of cytotoxicity of the extract and fraction with effective anti-conjugative activities revealed that the cytotoxic-IC 50 level of the hexane extract was above the concentration (100 mg/L) needed to reduce conjugal transfer of the IncW plasmid R7K by 76.5 ± 2.0%. is suggested that the hexane extract showed anti-conjugal activity at a non-toxic concentration. For the methanol fraction, the cytotoxic-IC 50 level was slightly below the effective anti-conjugation concentration (100 mg/L), which indicates that cytotoxicity may be an issue for one or more components in the fraction, but whether there is an association with inhibition of plasmid transfer is yet to be determined. No antibacterial activity was observed at a concentration of 512 mg/L for the methanol fraction against Escherichia coli. We propose that the activities of the extract and fraction might be through disruption of the conjugative machineries in a highly specific manner, and the slight variation in their anti-conjugal activities against IncW plasmid R7K may be due to the variations in their chemical structure. e hexane extract, which exhibited good anti-conjugant activity at a non-cytotoxic concentration was further evaluated against plasmid R7K (IncW) at sub-inhibitory concentrations, 16 to 256 mg/L. e results (Figure 2) showed that the activity of the hexane extract was concentration dependent and its anti-conjugant activity was comparable to the known IncW plasmid R7K conjugal inhibitor [7], linoleic acid, although the activity of the linoleic acid was slightly more potent. e anti-conjugant IC 50 of the hexane extract and linoleic acid was found to be 77.8 and 60.5 mg/L, respectively, against the conjugal transfer of IncW plasmid R7K. Statistically, the activities of both the M. lowiana hexane extract and linoleic acid were found to be similar since the corresponding p value (0.68) indicated that their IC 50 's were not significantly different.   e development a potent anti-conjugant molecule and/or new antimicrobial molecule with a new mechanism of action has the potential in reducing transfer and spread of resistance and virulence within important Gram-negative organisms such as E. coli.

Data Availability
All data generated or analysed during this study are included in this published article and its supplementary information file.

Conflicts of Interest
e authors declare that they have no conflicts of interest.