Bacterial strains used for antimicrobial activity testing were provided in-kind by Bath University. Strains include methicillin sensitive Staphylococcus aureus (MSSA) RN4220, Escherichia coli K12, Salmonella enterica serovar Typhimurium SL1344, Listeria monocytogenes NCTC 11994 (serovar 4b) and Enterococcus faecalis JH2-2. E. coli TOP10 (Invitrogen, Carlsbad CA, USA) was used for cloning (to host expression vectors for protein expression). The pTrcHis TOPO® vector (Invitrogen, Carlsbad, CA, USA) was used to clone polymerase chain reaction (PCR) products for protein expression.

Mueller Hinton (MH) (Sigma-Aldrich UK) and Luria Bertani (LB) broth and agar (Fisher Scientific Leicestershire, UK) were used as growth media. When leakage assays were performed under buffered conditions, 5 mM HEPES (pH 7.2) supplemented with 5 mM glucose was used (Wu and Hancock, 1999). Media were prepared and sterilized according to the manufacturers' instructions. Bacterial strains were grown using standard conditions unless otherwise specified. Broth cultures were incubated at 37°C for 18–20 h with aeration and cultures on solid media were incubated at 37°C for 18–24 h.

Antimicrobial genes were identified from the fosmid metagenomic library as previously described (Oyama, 2015). Briefly, sterile pin replicators (Molecular Devices Ltd., Berkshire UK) were used to transfer 2 μl metagenomic clones onto LB agar plates that had been plated before with 500 μl (OD_{600 nm} = 1) of pathogens such as S. aureus, E. coli, Sal. Typhimurium, E. faecalis and L. monocytogenes. Plates were incubated at appropriate temperatures for 24 h and zones of clearing around the clones were used to identify clones with inserts encoding antimicrobials. Putative antimicrobial positive fosmid clones were sequenced using Roche's 454 pyrosequencing platform. BLASTN (v2.2.28) on NCBI and BioEdit (version 7.1.11) (Hall, 1999) were used to edit and trim the vector sequence from the contigs. VecScreen on NCBI was used to search the sequences for vector contamination. Open reading frames (ORFs) were determined using the NCBI ORF finder program (Wheeler et al., 2003) and all ORFs with homology to antimicrobial genes and/or peptides were collated (Table 1). Based on expression levels and final protein yield, an ORF composed of 71 amino acids and named buwchitin was further investigated. Here, we report the activity of buwchitin.

Extracted fosmid DNA (1 μl) from a metagenomic clone containing the buwchitin insert was used as template for PCR amplification. The buwchitin sequence was deposited in the GenBank database with accession number KY823515 and predicted to contain a signal peptide, when analyzed on the SignalP 4.1 server (Petersen et al., 2011). Primers were designed to start and stop at the first predicted methionine and at the last stop codon, respectively in order to conserve the reading frame and take account of the entire gene of interest. The primers used for the amplification of buwchitin gene were 5′-ATGAGGCTGTCACACGTTTG-3′ (forward primer) and 5′-TCACCAATCTGTATGGCACCG-3′ (reverse primer). Primers were diluted to a stock concentration of 100 μM and a total volume of 50 μl PCR reaction was set up as follows: 2 μl DNA template, 1 μl each of the forward and reverse primers (2 μM final concentration), 39.5 μl molecular grade water and 1 μl Titanium® Taq DNA Polymerase (Clonetech- Takara Bio Europe/SAS, France). Taq polymerase was activated for 1 min at 95°C, followed by 30 cycles of 95°C for 30 s, 68°C for 1.5 min, followed by a final extension step at 68°C for 1.5 min. PCR products were verified by electrophoresis on a 1.5% agarose gel using a 1 kb DNA ladder. Gel image was taken after exposure to UV using the Gel Doc™ XR⁺ system (BIO-RAD Hertfordshire, UK). Subsequently, the band of interest was excised with a sterile scalpel under a Dark Reader blue transilluminator (Clare Chemical Research Inc. USA) and DNA was purified and eluted using the QIAquick Gel Extraction Kit (Qiagen, Crawley, UK) according to manufacturer guidelines.

Cloning of buwchitin was carried out using the pTrcHis TOPO® TA Expression kit as described by the manufacturer. Five positive colonies from the transformation were analyzed for correct size, sequence and orientation of the insert. Selected colonies were cultured overnight in LB medium containing 100 μg/ml ampicillin and 0.5% glucose, and analyzed by PCR. Briefly, aliquots (1 ml) were lysed by heating for 10 min at 95°C in sterile 1.5 ml microcentrifuge tubes. The cell debris was pelleted by centrifugation at 13,000 × g for 2 min. The supernatant was used as template for the subsequent PCR. The PCR was set up in a total volume of 50 μl as follows: 2 μl of template DNA, 1 μl of gene specific forward primer (5′-ATGAGGCTGTCACACGTTTG-3′) and vector specific reverse primer (5′-GATTTAATCTGTATCAGG-3′), 21 μl molecular grade water and 25 μl MyTaq™ Red Mix (Bioline, UK Ltd., London UK). Initial Taq activation was performed at 95°C for 1 min, followed by 35 cycles of 95°C for 15 s, at insert specific annealing temperature for 15 s with an extension step at 72°C for 10 s, and a final extension step at 72°C for 7 min and holding at 4°C. PCR products were verified by electrophoresis on a 1.5% agarose gel using a 500 bp DNA ladder. A positive PCR control was also prepared using the control PCR template (expected size of 750 bp) and primers provided with the pTrcHis-TOPO® expression kit. Positive transformants were further analyzed by Sanger sequencing using plasmid DNA from extracted positive transformants as templates. The Xpress™ Forward sequencing primer for pTrcHis-TOPO® (5′-TATGGCTAGCATGACTGGT-3′) was then used to sequence the insert and alignments to original sequence orientation was confirmed using BioEdit (Hall, 1999).

A single recombinant E. coli colony from a clone confirmed as containing the buwchitin gene was inoculated into LB broth containing 100 μg/ml ampicillin and grown overnight at 37°C with aeration and agitation (225–250 rpm). The following day, 1 L of LB broth containing 100 μg/ml ampicillin was inoculated with 20 ml of the overnight culture and incubated at 37°C under aeration (225–250 rpm). Gene expression was induced at OD_{600 nm} = 0.6 with 1 mM IPTG. Cells were harvested after 4 h by centrifugation (3,000 × g for 10 min at 4°C) and cell pellets were stored at −80°C for subsequent protein purification. Simultaneous purification and concentration of the buwchitin protein was carried out under native conditions using the Amicon® Pro Purification System (Merck Millipore Ltd Carrigtwohill, Ireland) following the manufacturer's protocol. Protein concentration was calculated as the ratio of absorbance at 280 nm [BioTek's Epoch™ Multi-Volume Spectrophotometer, (BioTek Instruments, Inc. Vermont, USA)] to the extinction coefficient absorbance (Abs 0.1% = 1 g/l calculated using the ExPASy ProtParam tool) (Gasteiger et al., 2005).

Vancomycin, Polymyxin B sulfate and ciprofloxacin were purchased from Sigma-Aldrich (Poole, Dorset, United Kingdom). All stock solutions were dissolved in the appropriate solvent prior to dilution in sterile distilled water (Andrews, 2001). MICs of buwchitin was measured by broth microdilution method using two-fold serial dilutions of antimicrobial agents in MH broth (CLSI, 2012). Buwchitin or comparator agents, vancomycin hydrochloride, polymyxin B sulfate and ciprofloxacin were added to the wells of a 96-well plate containing bacteria from overnight culture (adjusted to 1 × 10⁸ CFU/ml) to achieve a final inoculum concentration of 5 × 10⁵ CFU/ml (Cherkasov et al., 2008; Wiegand et al., 2008). MIC was defined as the lowest concentration of test agent that inhibited visible growth of the organism after 18–24 h of incubation at 37°C.

The bactericidal or bacteriostatic activity of buwchitin against E. faecalis was measured at MIC concentration using optical density measurements. An increase in both cell mass and cell number can readily be estimated by measuring the turbidity of a cell suspension using a spectrophotometer, thereby offering a rapid and sensitive alternative to cell counting (Dalgaard and Koutsoumanis, 2001; Madrid and Felice, 2005). This method has been shown to produce comparable results to plate counting, flow cytometric and green fluorescence viability analyses methods (Lehtinen et al., 2006). In a 96 well plate, buwchitin was added to cells in mid-logarithmic phase (1 × 10⁶ CFU/ml, OD_{600 nm} of ≤ 0.2) in MH broth and serially diluted as previously described. Plates were incubated at 37°C in a microplate incubator shaker. Wells without antimicrobial agents were used as growth control while wells with MH broth alone served as negative control. The rate of kill was calculated as a percentage (OD_{600 nm}) of surviving cells over a 24 h period (Lehtinen et al., 2006; Hazan et al., 2012). The percentage of viable cells was normalized to 100% for the growth control (cells without antibiotic treatment).

The ability of buwchitin to lyse red blood cells was assessed in a 96 well plate using defibrinated sheep blood (Oxoid Ltd Hampshire, UK). Sheep red blood cells (RBC) washed and diluted (4%) in phosphate buffered saline (35 mM PBS) (pH 7.3) were treated with buwchitin at different concentrations and incubated at 37°C for 1 h. Triton X-100 (0.1% causes 100% cell lysis) served as a positive control. Absorbance (OD_{450 nm}) of the supernatant (70 μl) from each well of the plate was measured to detect hemoglobin leakage from the erythrocyte cytoplasm and obtained results were used to determine the percentage hemolysis given that the 0.1% Triton X-100 represented 100% lysis after normalizing auto-hemolysis (PBS only treatment).

The ability of buwchitin to disrupt the electrochemical potential across the bacterial cytoplasmic membrane was measured by determining the amount of the membrane-associated probe, 3,3′-dipropylthiadicarbocyanine iodide [diSC₃(5)] released from the cytoplasm (Wu et al., 1999; Lee et al., 2004). Briefly, mid-logarithmic phase (OD_{600 nm} = 0.2) E. faecalis cells were washed and resuspended to an OD_{600 nm} of 0.05 in 5 mm HEPES-glucose buffer, pH 7.2. In a 96-well plate, the cell suspension was incubated with 100 mM potassium chloride (KCl) and 0.4 mM 3,3′-dipropylthiadicarbocyanine iodide [diSC₃(5)] until a stable reduction of fluorescence (excitation λ 622 nm, emission λ 670 nm) was achieved (~1 h). The KCl was added to equilibrate the cytoplasmic and external K⁺. After 1 h, buwchitin, positive control agent (0.1% Triton X-100) or negative control agent (untreated cells) were added to the cells in the wells. The plate was further incubated at 37°C with shaking while fluorescence was continuously monitored (excitation λ 622 nm, emission λ 670 nm) upon addition of peptide at 2–5 min intervals for 2 h.

Exponential phase cultures of E. faecalis grown in MH broth were washed and resuspended to an OD_{600 nm} of 0.2 in 10 mM PBS. The cell suspensions (1 ml) were incubated at 37°C with buwchitin at 1 × MIC concentration in microcentrifuge tubes. To investigate possible changes in cell morphology following exposure to buwchitin, samples were removed at 1 and 24 h after exposure and prepared for TEM as previously described (Huws et al., 2013). Briefly, samples were fixed with 2.5% (v/v) glutaraldehyde, after which they were post-fixed with 1% (w/v) osmium tetroxide. Fixed samples were then stained with 2% (w/v) uranyl acetate and Reynold's lead citrate and observed using a JEOL JEM1010 transmission electron microscope (JEOL Ltd, Tokyo, Japan) at 80 kV.

Structural modeling of buwchitin was completed using the PHYRE2 web portal (Kelley et al., 2015). Results were visualized using the PyMOL v1.7.6 program (Schrödinger, 2010). The biophysical properties of buwchitin were predicted on the antimicrobial peptide database (APD2) (Wang et al., 2009).

Two-way analysis of variance (ANOVA) with factors “antimicrobial treatments” and “time” was performed to determine whether there were significant changes in cell viability and membrane depolarization before and after treatment (Harmon, 2011). This was followed by post-hoc multiple comparisons using Tukey's HSD (Honestly Significant Difference) test (Bender and Lange, 2001; Feise, 2002; Harmon, 2011). Alpha (α) levels were set at P < 0.05.

The buwchitin gene was successfully PCR amplified using DNA from the fosmid clone, SAB PL27 L10/66. Bands of the correct size (expected size of 216 bp) were excised from the gel before proceeding to cloning. Electrophoresis results confirmed that the transformants carried the gene of the correct size, which was also confirmed by Sanger sequencing. The antimicrobial protein was expressed with an N-terminal 6xHis-Tag in E. coli to facilitate purification and investigation of its biochemical properties. Preliminary protein expression assay indicated that protein expression was optimal 4 h after induction (data not shown). SDS PAGE analysis of negative expression control (E. coli Top10 cells without plasmid) showed no protein expression bands while positive expression control (E. coli Top10 cells with pTrcHis-TOPO/lacZ) showed expression of the protein with a correct size of 40 kDa (data not shown). Cultivation of buwchitin transformants were scaled up to a total volume of 1 L to produce cell pellets for protein purification. Recombinant proteins were purified in their native conditions to preserve their activity (Karakus et al., 2016). Figure 1 shows the SDS-PAGE analysis of the purification fractions for buwchitin. The purification protocol reproducibly yielded a total of ~0.8 mg of purified protein per liter of culture.

Buwchitin was active against E. faecalis with an MIC of 100–200 μg/ml (Table 2). It also showed some inhibition of E. coli growth (observed in growth curves), but no detectable MIC at the highest concentration tested. This may account for the low level of expression of buwchitin in the E. coli expression host. The highest concentration of buwchitin tested was 400 μg/ml due to low levels of protein expression and/or yield of purified protein. The killing activity of buwchitin against E. faecalis was calculated as a percentage (OD_{600 nm}) of surviving cells compared to the growth control. Only about 30 ± 1.4% surviving E. faecalis cells remained after a 24 h incubation period (P < 0.05). It would seem that buwchitin had a bacteriostatic effect against E. faecalis cells (Figure 2) as no change in E. faecalis cell density was observed over an incubation period of 24 h. Very little hemolytic effect (12.81 ± 0.02%) was observed when sheep red blood cells were treated with buwchitin at a concentration twice as high as the MIC determined for E. faecalis (Table 3).

Buwchitin did not induce membrane depolarization in E. faecalis in the first 2 h of treatment. To determine whether the loss of viability in E. faecalis following exposure to buwchitin was accompanied by or was a result of changes in cell morphology and cell wall ultrastructure, TEM was performed. Electron micrographs of untreated E. faecalis at 1 and 24 h reveal intact healthy cells. Electron micrographs of buwchitin treated E. faecalis cells at 1 h showed intact outer membranes with blebbing but no major damaging effects and cell morphology changes. In contrast, micrographs of buwchitin treated E. faecalis cells at 24 h revealed several changes in cell morphology including cell lysis and detachment of the cell interior from the cell envelope (Figure 3).

Modeling and visualization of the 3D conformation of buwchitin using PHYRE2 (Kelley et al., 2015) and PyMOL v1.7.6 (Schrödinger, 2010), respectively, suggested that buwchitin is composed of a compact, all-helical, structure with major amphipathic helix connecting two smaller helices (Figure 4). The amphipathic helix agrees with a common structural feature of AMPs as the dual hydrophilic/hydrophobic nature allows the interaction and embedding of cellular membranes (Hancock and Sahl, 2006). As predicted by the APD2 database, buwchitin (71AA) is positively charged (+9), has a total hydrophobicity ratio of 29% and total Arginine and Lysine ratio of 19%.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.