Collection of plant material, drying and storage
After obtaining the necessary permit, leaves of two selected plants namely Eugenia zeyheri (Harv.) Harv. and Syzygium legatii Burtt Davy & Greenway were harvested in April, 2017, at the Lowveld National Botanical Garden in Nelspruit, Mpumalanga, South Africa. Herbarium specimens were prepared and deposited in the HGWJ Schweickerdt Herbarium of the University of Pretoria for authentication while herbarium specimen identity number (PRU) were obtained. These were 123617 and 123619 for E. zeyheri and S. legatii respectively. The leaves were placed in open mesh loosely woven bags and dried under room temperature with adequate ventilation. Using a Janke and Künkel Model A10 mill, the dried leaves were ground to a fine powder, weighed and stored in at room temperature in closed jam jars .
Two grams of dried powdered plant material was extracted with 20 mL of acetone technical grade, Merck) in 50 mL centrifuge tubes. Acetone is widely considered a solvent of choice because it can extract compounds with a wide range of polarities, it is relatively easy to remove from extracts and it is non-toxic to bioassays systems . The tube containing the mixture was vigorously shaken and sonicated for 20 min followed by centrifugation for 10 min at 4000 X g. The supernatant was then filtered using a Whatman No. 1 filter paper into a pre-weighed glass container and then dried under a cold stream of air in a fume hood at room temperature to obtain a dried extract. The dried extracts for were dissolved in the required volume of acetone for the bioassays.
An enterotoxigenic E. coli (possessing the STA and F6 virulence genes) isolated from a diarrhoeic piglet was obtained from the Department of Veterinary Tropical Diseases, Faculty of Veterinary Sciences, University of Pretoria. The organism was maintained on Tryptic Soy Agar (TSA, Oxoid) at 4oC.
Preparation of E. coli culture for electron microscopy
Escherichia coli was grown in TSA for 18 h after which a single colony was inoculated into Tryptic Soy Broth aseptically and incubated at 37oC on a shaker for 18 h. After this, an inoculum equivalent to a McFarland No 1 standard (3.6 × 108 cfu/mL) was prepared from the 18 h culture. Prior to this, TSA was prepared, and 5 mL of the molten agar was gently poured into 35 mm diameter sterile tissue culture plates to form a smooth and evenly spread surface and allowed to solidify in a sterile environment. The 35 mm agar plates were inoculated with the appropriately adjusted E. coli suspension which were spread evenly on the agar surface with a sterile glass spreader. The plates were then incubated at 350C for 12 h under aerobic conditions. The plates were divided into two groups. Plates in group one was flooded with 100 µl of E. zeyheri extract while those in group two were flooded with the extract of S. legatii, both at 0.04 mg/mL in acetone. This concentration represented the minimum inhibitory concentration of the extracts on the test bacterium in our previous study. One plate was flooded with 100 µl of 50% acetone to represent the solvent control. One plate also served as the untreated control, while another was treated with 100 µl gentamicin (1 µg/mL) as positive control. The plates were then incubated under aerobic conditions at 370C. After 0, 3, 6, 12, and 24 h, separate plates were removed from the incubator and flooded with 1 mL of 2.5% glutaraldehyde in 0.075 M phosphate buffer solution (pH 7.4) to fix the samples for 60 min. The bacterial biofilms were then collected from each plate using sterile loops and transferred into 2 mL microcentrifuge tubes containing 1.5 mL of 0.5% glutaraldehyde in order to fix the cells for 1 h. Glutaraldehyde was removed with a pipette and the cells were washed thrice with 0.075 M sodium buffer for 10 min each. The samples were then fixed with osmium tetroxide (OsO4, Merck, Darmstadt, Germany) in a fume hood for 30 min. Osmium tetroxide was removed and the cells were rinsed three times with the sodium phosphate buffer. Dehydration was done with increasing ethanol concentrations of 50%, 70%, 90% and 100% for 15 min each. The 100% ethanol step was repeated three times before preparation for scanning and transmission electron microscopy.
Scanning electron microscopy (SEM) sample processing
Following the last 100% ethanol dehydration from the step above, hexamethyldisilazane (HMDS) was added at 50% in ethanol for 30 min. Hexamethyldisilazane/ethanol was replaced with two changes of pure HMDS for 1 h each. A small droplet (0.05 mL) containing sample was placed on highly polished carbon discs and left open in a fume hood to dry overnight. These carbon discs were stuck using double-sided carbon tape onto aluminium stubs. Samples were made conductive by exposure to ruthenium tetroxide (RuO4) for 45 min . Samples were then viewed with a Zeiss Ultra Plus Field Emission Gun Scanning Electron Microscope (FEGSEM) at the Electron Microscope Unit of the University of Pretoria.
Transmission electron microscopy (TEM) sample processing
After the third dehydration step in 100%, ethanol was removed and replaced with propylene for 2 h. The propylene oxide was replaced with an Epon type epoxy resin (TAAB 812) and infiltrated for 5 h. Pellets were removed from the microcentrifuge tubes, placed in embedding moulds and polymerised for 48 h. After this, sections were made with a Reichert Ultracut E ultramicrotome using a diamond knife and picked up onto copper grids. The sections on the grid were stained with 2% uranyl acetate followed by 2 min staining in Reynold’s lead citrate. Sections were viewed and photographed with a Philips EM 10 transmission electron microscope (Eindhoven, Netherlands) in the Electron Microscopy Unit, Department of Anatomy and Physiology, Faculty of Veterinary Sciences, University of Pretoria.
Propidium iodide (PI) uptake assay: cell viability assessment
This was done as previously reported  with slight modifications. Briefly, the E. coli cells were grown in TSB for 18 h, washed twice with phosphate buffer saline (PBS) and then adjusted to 106 CFU/mL with PBS. The cells were then incubated at 370C with the extracts of E. zeyheri and S. legatii at the MIC concentrations (0.04 mg/mL respectively) for 3 h. Gentamicin (1 µg/mL) served as positive control, untreated cells served as untreated negative control while heat-killed (70oC for 60 min) bacteria served as treated negative control. Following incubation, cells were washed twice with PBS and fixed with 3.7% paraformaldehyde in 0.1 M cacodylate buffer for 20 min at room temperature. Cells were then washed twice with PBS and incubated with PI at 370C for 30 min in the dark. Cells were spread on a glass slide, covered by a cover slip and observed using a Nikon Eclipse TS 100-U inverted fluorescence microscope (Nikon, Champigny sur Marne, France) fitted with a Nikon Intensilight C-HGFI lamp unit at excitation and emission wavelengths of 536 and 617 nm respectively. Images were captured at 20 x magnification.
Gas chromatography-mass spectrometry (GC-MS) of plant extracts
Analysis of chemical constituents of plant extracts of S. legatii and E. zeyheri were carried out in a LECO Pegasus 4D GC-TOFMS (LECO Africa (Pty) Ltd, Kempton Park, South Africa) on an apolar Rxi-5SilMS 30 m x 0.25 mm ID x 0.2 μm film thickness (Restek, Bellefonte, PA, USA) gas chromatography capillary column. Compound spectra were detected by electron ionization system (70 eV). Ultra-high purity grade carrier gas, Pure helium gas (Afrox, South Africa) was set at a constant flow rate of 1 mL/min. Oven temperature was held for 3 min with 5 min solvent delay programmed at 40°C and held isothermally at 300 °C for 5 min. A 1 μL of acetone solution of the sample was injected in a splitless mode (splitless time 30s) with the injector temperature at 250 ºC. Ion source temperature was maintained at 280 ºC. A scan interval of 0.5 seconds and fragments from 40-550 Da was maintained. Relative quantity of the compounds in extracts was expressed as a percentage based on the peak area produced in the chromatogram. Tentative identification of the bioactive constituents was done by comparison of retention times with standard samples and by matching the spectral fragmentation patterns against commercial library mass spectra. Analysis was done at the Department of Chemistry, University of Pretoria, South Africa.