3D Structure Elucidation and Appraisal of Mode of Action of a Bacteriocin BaCf3 with Anticancer Potential Produced by Marine Bacillus amyloliquefaciens BTSS3

Abstract

In the world of intense usage of antibiotics, the emergence of antimicrobial resistance necessitates research on alternative forms of antibiotics, antimicrobial peptides (AMPs), which are least known to induce resistance. The partial sequence of bacteriocin BaCf3, produced by marine Bacillus amyloliquefaciens BTSS3, derived from Matrix Assisted Laser Desorption Ionisation - Time of Flight Tandem Mass Spectroscopy (MALDI-ToF MS/MS) data was analyzed for amino acid composition and modelled in silico using TrRosetta. The mechanism of action of BaCf3 was studied in vitro on B. circulans NCIM2107 cell wall using microscopic techniques, such as confocal laser scanning microscopy, scanning electron microscopy, and high resolution transmission electron microscopy. Docking studies with cancer markers, glucose transporter protein, and mesenchymal-epithelial transition factor (MET) receptor tyrosine kinase were also conducted. BaCf3 was found to be rich in glycine and hydrophobic in nature, a characteristic property of cell wall acting AMPs. The structure of BaCf3 obtained from TrRosetta had antiparallel b-sheets resembling Laterosporulin. The bacteriocin BaCf3 has been found to act on the cell membrane of opportunistic pathogen Bacillus circulans, causing permeabilization and pore formation by dissipating the membrane potential. The microscopic examination also proved the mode of action of BaCf3 as cell wall acting. In silico docking studies with anticancer target proved that bacteriocin BaCf3 is also a possible anticancer drug candidate. In vitro anticancer assays such as 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay and acridine orange/ethidium bromide dual staining on lung carcinorma cell line A549 further prove the anticancer activity of the bacteriocin.

Introduction

Antimicrobial resistance due to overuse and misuse of antibiotics causes untreatable infections to persist, thereby increasing the risk of contagion. In May 2017, World Health Organization adopted a global action plan to increase investment in new medicines, diagnostic tools, vaccines, and other interventions. Antimicrobial peptides (AMPs) are considered as good alternative for artificial antibiotics as they are more natural and biodegradable and are not known to induce resistance by biomagnifications.1 Most AMPs are small in size, which kill bacteria using a mode-of-action (MOA) different from traditional antibiotics. There are two mechanisms by which AMPs act on target cells: (1) AMPs induce membrane disruption, cell lysis, and death and (2) AMPs enter the cells without membrane disruption and inhibit intracellular functions.2 A clear understanding of the MOA is useful to reduce toxicity to the host cells. This could also help to improve the potency of the AMPs.3 Biochemically simple and highly efficient killing mechanism of AMPs could effectively prevent the evolution of resistance to AMPs.4 Bacteriocins are AMPs produced by bacteria against the same or related species.

Bacterial cells incubated with sublethal and/or lethal concentrations of bacteriocin are imaged to identify whether the membrane surface is intact, becomes wrinkly, has blebs, or is lysed. The biological properties of proteins are dependent on their tertiary structures. The primary structure can be converted to tertiary structure or their spatial models using modern in silico platforms. These models can be used for virtual screening of the molecule for bioactivities by homology modeling.5

Here, the effect of bacteriocin BaCf3 was studied on opportunistic pathogen B. circulans. Even though B. circulans is mainly considered as an opportunistic pathogen in immunocompromised patients, very few cases of pathogenic B. circulans were reported untill now; many of the reported pathogenicity were fatal. The first case of fatal meningitis due to B. circulans was reported in a 5-day old infant in 1948.6 Another case of fatal sepsis was also reported in immunosuppressed patient who had a history of kidney graft surgery.7 Many other infections have also been reported due to B. circulans, which include bacteremia, abscesses, endophthalmitis, and wound infections. The first case of vertebral osteomyelitis caused by B. circulans was reported by Russo et al.8 In all these cases, B. circulans was resistant to broad spectrum antibiotics. The above-mentioned facts necessitated the need for a natural antibacterial compound against B. circulans with a clear understanding of the mechanism of action.

Bacteriocins, in addition to antibacterial action, exhibit other bioactivities like anticancer action. In silico approaches like docking on possible target molecules are an easy and convenient method of screening for anticancer activity.

In this study, the bacteriocin BaCf3 whose primary sequence was identified from its MALDI-MS/MS data in our previous report9 was investigated in silico to predict its secondary and tertiary structures. The mode of action was proved by in vitro microscopic methods like confocal laser scanning microscopy (CLSM), scanning electron microscopy (SEM), and high-resolution transmission electron microscopy (HR TEM) on B. circulans NCIM 2107. The bacteriocin BaCf3 is also used to study the anticancer activity in silico.

Methodology

Amino acid composition and instability index of bacteriocin

The amino acid composition and instability index of BaCf3 sequence, GNHDCCMGQLPKCMLGPGQGAATCQGK, was calculated using the ProtParam protein analysis tool in ExPASy (http://web.expasy.org/protparam/).10 ProtParam is a tool that allows the computation of various physical and chemical parameters for a given protein stored in Swiss-Prot or TrEMBL or for a user entered protein sequence.

AMP classification by CAMPR3

The derived amino acid sequence of BaCf3 was predicted for AMP status using CAMPR3 available online at www.camp3.bicnirrh.res.in. CAMPR3 is an experimentally validated database of AMPs, which could predict the given amino acid sequence as AMPs or not. CAMPR3 includes AMP prediction tools based on random forests,11 support vector machines (SVMs),12 artificial neural network (ANN), and discriminant analysis.13 The analysis is done by converting the input sequence to physicochemical properties and structural characteristics of amino acids, as well as dipeptide and tripeptide frequencies of the reduced alphabets.

Secondary prediction from the derived partial sequence

The secondary and tertiary structure prediction of the de novo sequence of BaCf3 was done using I-TASSER (Iterative Threading ASSEmbly Refinement, https://zhanglab.ccmb.med.umich.edu/I-TASSER/), a hierarchal approach to predict protein structure and function.14 This server provides the most accurate structural and functional predictions for small peptide sequences.15,16

Tertiary structure prediction of BaCf3

The tertiary structure of the partial sequence was derived using TrRosetta available in the RObetta protein structure prediction service.17 This is an interactive submission interface that allows modeling, constraints, local fragments, and multichain complexes. This used PDB100 as template database. The results can also predict the formation of disulfide bridges. The FASTA format of the protein sequence was uploaded in the submission form. The results obtained can be viewed and downloaded. A multiple sequence alignment was conducted in Clustal Omega with similar sequences and obtained the percentage identity.

Prediction of disulphide bridge

Disulfide bridges are involved in the stabilization of tertiary structures. DiAminoacid Neural Network Application (DiANNA) is a web server that provides disulfide connectivity prediction. DiANNA 1.1 determines the disulfide connectivity involving a novel architecture neural network.18 The neural network input includes evolutionary as well as secondary structure information.

Experimental Validation of Mode of Action of BaCf3

Materials

Purified bacteriocin BaCf3 produced by Bacillus amyloliquefaciens BTSS3 (MCC 2981) was used for this study. The purification steps involved collection of supernatant, precipitation, dialysis, and gel filtration chromatography as described in our previous study.9 The test organism B. circulans NCIM2107 was procured from NCIM, Pune, India. The organism was maintained on nutrient agar slant at room temperature and long-term storage as glycerol stock at –80°C.

Minimum inhibitory concentration

Minimum inhibitory concentration (MIC) of BaCf3 was determined according to Sarker et al.19 To each well of a sterile 96-well U-shaped microtiter plate, a 100 µL of double strength Luria–Bertani (LB) broth was added. Then, a 100 µL of purified bacteriocin of known concentration (1 mg/mL) was added to the first well, mixed well, and a 100 µL was taken from the first well and serially diluted. The B. circulans NCIM2107 suspension (absorbance at 600 nm = 1) prepared in the LB broth (10 μL) was added to each well. The plate was incubated at 37°C for 18 h. A 2 µL of rezasurin was added to each well. Wells with live organisms were converted from blue rezasurin to pink color. The lowest concentration of bacteriocin that inhibited the growth of the B. circulans NCIM2107 was considered as the MIC. Thus, MIC was calculated for B. circulans NCIM2107. The experiment was repeated thrice independently.

Bactericidal/static mode of action

To get 1 mL of B. circulans NCIM2107 culture (OD₆₀₀ = 0.1), purified bacteriocin was added at its MIC concentration. The tube was then incubated at 37°C at 120 rpm for 6 h, and absorbance was noted at 600 nm. Tube without bacteriocin acted as control. The experiment was repeated thrice, and standard deviation was determined. A time-dependent inhibition experiment was also conducted for 3 h, and a graph of time vs percentage inhibition was plotted. $% Inhibition = [\frac{OD of control - OD of test}{OD of control}] 100$

Action of bacteriocin on bacterial cell wall

Confocal laser scanning microscopy

B. circulans NCIM2107 cells (1 × 10⁴ CFU/mL) were allowed to grow on StarFrost^® glass slides (Knittel, Germany) with activated surface for 30 min. The slide was then flooded with BaCf3 at its MIC for 30 min. After the treatment, the slides were washed thrice with phosphate buffered saline (PBS), fixed with 2.5% gluteraldehyde in PBS, and stained with To-Pro-3 stain at a dilution of 1:1000 in PBS for 20 min in the dark at room temperature. The slides were observed and photographed using confocal imaging system (Leica TCS SP 5).20 The pixel intensities of the red spots in the images before (Control) and after (Treated) treatment with bacteriocin BaCf3 were quantified using the ImageJ software.21

Scanning electron microscopy

B. circulans NCIM2107 was allowed to grow on three sterile coverslips by aseptically immersing them in LB broth inoculated with the bacteria. The two coverslips were then treated with MIC concentration of BaCf3 for 1 and 2 h separately at 37°C. The coverslip culture without BaCf3 served as the control. After fixation using 2.5% gluteraldehyde in PBS, the coverslips were washed twice with PBS and dehydrated in a graded series of ethanol (50, 70, 80, 90, and 100%). These were dried in desiccators and used for SEM imaging (Tescan VEGA3 SB) after Gold-Palladium coating.

High-resolution transmission electron microscopy

B. circulans NCIM2107 broth culture of OD₆₀₀ = 1 was mixed with BaCf3 at its MIC and incubated at 37°C for 1 h. This was mixed with equal volume of 2% phosphotungstic acid in PBS, loaded on the grid, dried, and analyzed by transmission electron microscopy (TEM) (Jeol/JEM 2100). The culture without bacteriocin served as control. The gold-plated cover slips were observed under 15,000× magnification.

In silico analysis of anticancer activity of BaCf3

Two target proteins, glucose transporter protein (GLUT1) and MET receptor tyrosine kinase, which could be considered as the possible anticancer targets of the bacteriocin, were selected randomly from literature. PDB IDs of the cancer target proteins were secured, and these were docked against the model of the bacteriocin predicted by TrRosetta using the ClusPro 2.0 server and obtained the results.22

The visual representation, the complex interactions, and bond distances between interacting amino acids were assessed using the Discovery Studio^® 2016 software (BIOVIA, San Diego, CA, USA).

In vitro anticancer activity of BaCf3 in cell culture by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay

The effect of bacteriocin BaCf3 on the proliferative capacity of A549 cells (adenocarcenomic human alveolar epithelial cell line) was determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The cell line was propagated in Dulbecco's Modified Eagle Medium (DMEM) (Invitrogen, Carlsbad, CA, USA), supplemented with 10% heat inactivated fetal calf serum (PAN-Biotech GmbH, Aidenbach, Germany) and antibiotics (100 IU/mL penicillin and 100 μg/mL streptomycin), and maintained at 37°C and 5% CO₂ atmosphere as described previously.23 The A549 cell line was procured from the National Centre for Cell Science, Pune, India. The cells were seeded (5,000 cells/well) in 96-well, flat-bottom microtiter plates along with different concentrations of bacteriocin and incubated for 24, 48, and 72 h at 37°C in 5% CO₂ atmosphere. After incubation, the medium was removed, and wells were washed with PBS; 100 µL of the working MTT dye in DMEM media was added and again incubated for 2 h. MTT lysis buffer (100 µL) was added and continued incubation for 4 h. The absorbance was measured at 570 nm, and the proliferation rate (PR) was calculated using the following formula: $PR = \{\frac{Absorbance of Test - Absorbance of Blank}{Absorbance of Control - Absorbance of Blank}\} \times 100$

All the experiments were repeated thrice for cell line. The results were expressed as mean ± standard deviation. Statistical analysis was done by analysis of variance test in GraphPad Prism 6.0. P-Values less than 0.05 were considered significant.

Morphological evaluation of apoptosis

Cells seeded in a 96-well plate were treated with 100 µg/mL bacteriocin BaCf3 and incubated for 48 h at 37°C in a 5% CO₂ incubator.24 After incubation, medium was discarded and 25 μL acridine orange (7.5 μg)/ethidium bromide (25 μg) stain was added, mixed well, and observed under fluorescent microscope with the FITC filter. More than 300 cells were examined in a fluorescence phase contrast microscope (Olympus IX51) using a fluorescein filter and a 40× objective.

Results and Discussion

Amino acid composition of BaCf3

The analysis of sequence composition by the ProtParam tool shows that the most prominent amino acid in the bacteriocin was glycine, 22.5% in BaCf3. Cysteine and glutamine form the next predominant amino acid of BaCf3. The results of amino acid composition of bacteriocins are detailed in Table 1. The instability index (II) was computed to be 28.69 and classified the protein as stable. From the AMP database, APD3 and AMPs, respectively, have an average charge of +3.2 and 37.2 average amino acids with glycine (G), lysine (K), and leucine (L) being the most abundant amino acids, and α-helices and β-sheets being the most common secondary structures.25

Table 1.

Amino acid composition of bacteriocins as calculated by the ProtParam tool

Amino acid	BaCf3 (%)
Ala (A)	7.4
Arg (R)	0
Asn (N)	3.7
Asp (D)	3.7
Cys (C)	11.1
Gln (Q)	11.1
Glu (E)	0
Gly (G)	22.2
His (H)	3.7
Ile (I)	0
Leu (L)	7.4
Lys (K)	7.4
Met (M)	7.4
Phe (F)	0
Pro (P)	7.4
Ser (S)	0
Thr (T)	3.7
Trp (W)	0
Tyr (Y)	0
Val (V)	0
Pyl (O)	0
Sec (U)	0
X	3.7

Considering the amino acid composition, the net charge, and hydrophobicity of the amino acids, BaCf3 was found to be hydrophobic. Hydrophobic AMPs isolated from fish epidermal mucus were reported to have pore forming action on target cells.26

AMP Classification CAMPR3

Bacteriocin BaCf3 was classified as AMP by different classifiers of CAMPR3. Detailed probabilities are given in Table 2.

Table 2.

AMP probability of BaCf3 predicted by different classifiers

Classifier	AMP probability
Support vector machine (SVM) classifier	0.865
Random forest classifier	0.5715
Artificial neural network (ANN) classifier	AMP
Discriminant analysis classifier	0.972

Secondary structure prediction from the derived partial sequence

The secondary structure of BaCf3 as predicted by I-TASSER shows that the sequence contains helix, coils, and strands. The high score shows more confidence in the secondary structure prediction ( Fig. 1). Though the confidence scores for helix (H) and strands (S) are 1 and 0, the stability of the structure is more accurately predicted with the help of B-factor. The normalized B-factor (called B-factor profile, BFP) is predicted using a combination of both template-based assignment and profile-based prediction. Based on the distributions and predictions of the BFP, residues with BFP values higher than 0 are less stable in experimental structures, and negative values are more stable.27 The B-factor values for the region are –0.15, –0.4, –0.34, and –0.29 (Supplementary material).

FIG. 1.

Secondary structure of BaCf3.

Tertiary structure prediction of BaCf3

The tertiary structure of BaCf3 predicted by TrRosetta had triple stranded β-sheets with distinctive folds ( Fig. 2) and a disulfide bridge. A similarity alignment matrix with Laterosporulin shows that the sequences have more than 20% similarity.

FIG. 2.

(A) Three-dimensional structure of bacteriocin BaCf3 with antiparallel β-sheets and disulfide bridge. (B) Multiple sequence alignment of BaCf3 with laterosporulin and Laterosporulin10 using Clustal Omega. Refer Singh et al.29 and Baindara et al.30 for sequences of Laterosporulin and Laterosporulin10, respectively.

Three antiparallel beta sheets were predicted in the model, which gives the structural resemblance to Laterosporulin. In this model, the beta sheets were observed from C5 to M7, K12 to L15, and A22 to C24. A disulfide bridge was proposed in the model between C5 and C13. Usually, an SS bridge was proposed since the distance was less than 3 Å. Bacteriocins that induce visible damage by acting on the membrane are classified as Class II bacteriocins. Class II bacteriocins are considered to have diverse sequence and structure whose mechanism of action is through interaction with the cell membrane. They are diverse in sequence with some of them having β-sheets and defensin like structure.28 The structure of BaCf3 was also similar to that of Laterosporulin (PDB ID: 4OZK), Laterosporulin10 (PDB ID: 6LWZ) produced by Brevibacillus sp., and many other cysteine rich AMPs.29,30 All these AMPs are known to have pore forming action on target cell wall further confirming the proposed action of BaCf3 on cell wall.

Prediction of disulphide bridge

Many of the biologically active peptides contain one or multiple disulfifide bridges, which are believed to contribute to the peptides' stability and activity. In BaCf3, a disulfifide linkage can be formed among C5–C6, C5–C13, and C6–C13 cysteine as predicted by DiANNA 1.1 ( Fig. 3). The model predicted by TrRosetta also supports the disulphide bond between C5 and C13 for which the score was comparatively high. Disulfide bonds or the ability to form dimers by forming disulfide bond accounts for the heat stability of bacteriocins. Disulfide bonds help in the stabilization of tertiary structures by reducing the number of possible unfolded structures (entropic effect).

FIG. 3.

Disulfide bond prediction of BaCf3 by DiANNA 1.1.

Minimum inhibitory concentration

MIC of BaCf3 against B. circulans NCIM2107 was determined by microtiter plate assay incorporating resazurin. The MIC of BaCf3 required for complete inhibition of B. circulans NCIM2107 was 1.84 ± 0.005 μg/mL.

Bactericidal/static mode of action

The effect of bacteriocin on the growth of B. circulans NCIM2107 was studied by incubating 0.1 OD₆₀₀ culture of test organism with bacteriocin BaCf3 at its MIC concentration for 6 h. The percentage inhibition was calculated as 95.66 ± 0.302%, which means that BaCf3 had a bactericidal mode of action on B. circulans NCIM2107. In the time vs percentage inhibition curve (Supplementary material), more than 90% inhibition was observed after 3 h of incubation with the bacteriocin. According to Balouiri et al.,31 >90% inhibition after 6 h of incubation can be considered bactericidal.

Most bacteriocins reportedly have bactericidal mode of action. Mesentericin Y105, Plantaricin A, and Subtilosin A were bactericidal toward Listeria monocytogenes, Gardnerella vaginalis, and Streptococcus agalactiae and also against Gram-negative bacteria.32 –34 Since B. circulans is an opportunistic pathogen, bactericidal action is the preferred antibacterial mode of action. The organism is ubiquitous and, hence, is a cause of bacterial sepsis in immunocompetent patients.8

Action of bacteriocins on bacterial membrane

Confocal laser scanning microscopy

ToPro-3 is a nuclear stain; hence, membrane compromised cells can be seen as red spots due to the far-red fluorescence of the dye. CLSM showed that the membrane damage occurred after 30 min of treatment indicated by an increase in the pixel intensity of treated samples ( Table 3). Red fluorescence was clearly visible in treated cells indicating membrane damage and ToPro-3 binding to DNA ( Fig. 4).

FIG. 4.

Confocal imaging to study the action of bacteriocins on membranes of B. circulans NCIM2107. (A) Untreated B. circulans NCIM2107. (B) BaCf3 treated B. circulans NCIM2107. Membrane disruption occurred after 30 min of incubation with bacteriocin. This can be observed by increase in pixel intensity of treated samples due to the binding of To-Pro-3 on DNA of membrane compromised cells (scale bar—250 µm).

Table 3.

The different interaction scores of BaCf3 with the ligands

Receptor	Receptor ID	Score (balanced)	Score (electrostatic)	Score (hydrophobicity)	Score (vdw)
Glucose transporter protein (GLUT1)	1SUK	–903.1	–899.1	–1197.9	–137.5
MET receptor tyrosine kinase (TK)	3DKC	–703.6	–594.9	–744.8	–115.3

According to Torrent et al.,35 confocal microscopy is a novel tool to assess bacterial cell viability and to characterize action of bacteriocins. Major advantage of CLSM is that it allows statistical analysis of the pixel intensity measurements to derive quantitative measurements about viability. Viable cells with intact membrane prevent penetration of ToPro-3 and remain unstained, whereas dead cells or cells with permeable and disrupted membrane give fluorescent signal due to penetration of the dye into the cells and its intercalation into cellular DNA or RNA. Consequently, pixel intensity is a measure of permeability of target cells and extent of cell wall damage. CLSM was used to study the effect of biocidal compounds on the viability of Escherichia coli. 36 Membrane permeabilization of L. monocytogenes 54002 by plantaricin LPL-1 was investigated by Wang et al.37

Scanning electron microscopy

Changes in cell morphology of treated cells were compared to untreated control cells using SEM analysis ( Fig. 5A–C). The untreated control cells of B. circulans NCIM2107 appeared well in shape with intact cell wall (Fig. 5A). Observed misshaped or distorted cell wall of organism in treated samples can be attributed to the action of bacteriocin on the cell wall (Fig. 5B and C). It was noted that cell wall deformation is clearly visible after 1 h incubation, whereas it was completely ruptured after 2 h. Cells seemed collapsed, flat, and empty after bacteriocin treatment for 2 h (Fig. 5C). It was also noticed that deformation of target cells occurred in a time-dependent manner.

FIG. 5.

Action of BaCf3 on test organism by SEM. (A) Control cells of B. circulans NCIM2107. (B) and (C) After treatment with bacteriocin for 1 and 2 h, respectively. Images were captured at 15,000× magnification.

Changes observed in SEMs reflect the efflux of cell contents through pores created in the cell membrane. Similar observations were noted following the action of nisin and pediocin mixture on L. monocytogenes. 38 The morphological changes visualized by SEM proved that BaCf3 acted as bactericidal bacteriocin on B. circulans NCIM2107.

Subtilin exhibited bactericidal activity based on pore formation in the cytoplasmic membrane. Sublancin168, a lantibiotic produced by B. subtilis 168, exhibited bactericidal activity against other Gram-positive bacteria, including pathogens such as Bacillus cereus, Streptococcus pyogenes, and Staphylococcus aureus. 39,40 Sonorensin, a bacteriocin from Bacillus sonorensis MT93 isolated from marine soil sample, induced an increase in the permeability of S. aureus cytoplasmic membrane over time. SEM image of sonorensin treated S. aureus displayed roughened cell surface with accumulation of cell debris.41 Subtilin, Sublancin168, and Sonorensin are thermostable and pH tolerant bacteriocins produced by Bacillus sp. SEM of E. coli treated with lethal dose of bacteriocin facilitated comprehension of bactericidal action of laterosporulin on indicator strain,29 with detection of roughened cell surface and accumulation of cell debris.

High-resolution transmission electron microscopy

Morphological changes created by bacteriocin action were better observed with TEM micrographs ( Fig. 6). Changes in cell wall integrity and ghost cell formation were clearly visualized. Accumulation of membranous structures in the cytoplasm of treated cells was visible. Cell wall disruption with the release of intracellular material concomitant with cells losing their cytoplasm (empty and flaccid cells) was also noted.

FIG. 6.

Transmission electron microscopy showing different stages of bacteriocin BaCf3 action. (A) Untreated cell of B. circulans NCIM2107. (B)–(D) Bacf3 treated cells of B. circulans NCIM2107. The change in cell wall integrity is clearly visible. (B) Shrinkage of intact cells, (C) the release of cell contents, and (D) the ghost cell.

The cell membrane separated from the cell wall and part of the cytoplasm appeared concentrated and dispersed (Fig. 6D). After treatment, the cell surface collapsed with obvious debris around the cell. Microscopy showed that the treatment of bacteriocin affected membrane integrity of target cells.

TEM was adopted to study the mechanism of action of BLS P34 on L. monocytogenes-indicated cytoplasmic membrane alteration.42 TEM also exposed cell lysis of B. cereus and L. monocytogenes after treatment with cerein 8A.43 Similarly, nisin, pediocin, enterocin, epidermin, and other amphiphilic antimicrobial Type A lantibiotics exert their activity by disrupting the functional barrier of microbial cytoplasmic membranes.44 –46 Cerein 7 produced by B. cereus Bc7 is a hydrophobic membrane-active compound.47 The empty and flaccid cells, called cell ghost, with intact cell envelope structure devoid of cytoplasmic content including genetic material were clearly visualized in the TEM image. This distortion of physical structure is caused by expansion and destabilization of membrane and increased membrane fluidity, which, in turn, increases passive permeability, manifested as leakage of various vital intracellular constituents, such as ions, ATP, nucleic acids, sugars, enzymes, and amino acids.

A time-dependent morphological shift was demonstrated in sensitive L. monocytogenes by action of pediocin from P. acidilactici. Bacteriocin-treated L. monocytogenes V7 was destroyed after 6 h. After 1 and 3 h of treatment, breaches in the cell wall and plasma membrane were more evident with more cell content escaping. Cell wall was completely irregular and damaged after 6 h.48 The treatment of B. circulans NCIM2107 with the bacteriocin BaCf3 leads to major morphological changes in the cell wall and cell membrane, causing increased permeability even after 0.5 h of incubation, pore formation after1 h and complete cell damage after 2 h of incubation. This validates the mechanism of BaCf3 as a membrane acting agent as proved by the microscopic imaging techniques.

In silico analysis of anticancer activity of BaCf3

ClusPro 2.0 generated hundreds of conformational models by minimizing their energy with at least one near native model and was clustered. These clusters were ranked based on their size. All clusters that were unstable or having high energy after optimization were discarded. They were then scored for interaction of ligand with receptor-like balanced, hydrophobic, electrostatic, and Van Der Waals (vdw) scores (Table 3). The scores were expressed in terms of energy (kcal/mol). For template-based docking without specified binding sites, the balanced score is more important. Lower the scores and energy, the greater the interaction.

Docking studies revealed that BaCf3 interacts with Ala405, Ala402, Pro401, Arg400, Gln397, Leu262, Glu261, Val257, Glu254, Arg253, Met252, Met251, Gln250, Lys225, His160, Phe152, Met142, Pro141, Gly138, Val87, Leu85, and Phe81 amino acids ( Fig. 7) of glucose transporter protein, GLUT1 (PDB ID-1SUK). Some of the interacting amino acids are known to be involved in glucose transport.

FIG. 7.

Docking interaction of glucose transporter, GLUT1 (PDB ID-1SUK) with bacteriocin BaCf3. (A) Complete docked model, (B) interaction with the receptor, and (C) a closer view of interacting amino acids of the ligand; the bacteriocin BaCf3 is depicted in the stick model.

GLUT1 cell surface marker is over expressed in many tumor cells such as hepatic, pancreatic, breast, esophageal, brain, renal, lung, cutaneous, colorectal, head and neck, endometrial, ovarian, and cervical cancers.49,50 The increased expression of GLUT1 indicates an increased intake of glucose there by supplying the energy necessary for tumor cell proliferation. Positron emission tomography scans evaluated glucose uptake by cancer cells.51 Thus, in order to properly target glucose transporters, it is crucial that GLUT inhibitors are identified.

Quercetin, a GLUT1 inhibitor, binds to Glu254 and Gln397 by hydrogen bonding.52 The docking study with BaCf3 also gives an idea about the involvement of these amino acids in the nonbond interaction. This indicates that BaCf3 could inhibit glucose transport to the cells, thereby reducing the metabolic rate of cancer tissue. Docking of D-glucose and quercetin on the GLUT1 receptor revealed that uncharged quercetin also binds one of the glucose binding sites in the inner vestibule of GLUT1, which is negatively charged.52 Studies show that this site likely involved Glu254 and Lys256 that hydrogen-bond to quercetin and glucose. Now considering the net charge and spatial conformation of bacteriocin BaCf3, the interaction with the negatively charged inner vestibule of GLUT1 is certain.

The docking sites of BaCf3 on tyrosine kinase (PDB ID-3DKC) included 14 residues such as Pro1158, Asp1222, Tyr1159, Lys1110, Met1211, Met1160, Arg1208, His1088, Val1092, His1094, Arg1166, Asn1167, Val1083, and Arg1170 ( Fig. 8).

FIG. 8.

Docking interaction of tyrosine kinase (PDB ID-3DKC) with bacteriocin BaCf3. (A) Complete docked model, (B) interaction with the receptor, and (C) a closer view of interacting amino acids of the ligand; the bacteriocin BaCf3 is depicted in the stick model.

MET receptor tyrosine kinases are frequently deregulated in cancer and promote cell proliferation, progression, metastasis, and therapeutic resistance. Activation of MET by mutation or gene amplification has been linked to kidney, gastric, and lung cancers, while autocrine activation was demonstrated in glioblastoma. Since MET is a target in so many aspects of cancer development and progression, the MET receptor tyrosine kinase has emerged as an important target for the development of novel cancer therapeutics.53 Tyrosine kinases are frequently deregulated in cancer, and this promotes tumor formation, progression, and metastasis. Numerous agents have been developed that are able to target MET expression and/or function.

From Fig. 8C, we can observe that Arg1208, His1088, Arg1166, Val1083, and Arg1170 are the five amino acids of the tyrosine kinase receptor that the bacteriocin interacts with. The inhibition of ATP binding site of the tyrosine kinases may inhibit it, and hence, prevent cancer metastasis. Thus, many ATP analogues can act as cancer therapeutics.54 His1088, Lys1110, Asp1204, Pro1158, Asn1209, Arg1208, Met1160, Met1211, and Val1092 are the ATP binding sites on MET receptor tyrosine kinase.54 According to the docking study, the binding site of bacteriocin BaCf3 includes the ATP binding sites Arg1208, Pro1158, Asp1222, and His1088 on tyrosine kinase.

The antimicrobial and anticancer activities of AMP KL15 obtained by in silico sequence modification of two bacteriocins m2163 and m2386 from Lactobacillus casei ATCC 334 were studied.55 Laterosporulin10 (LS10), produced by Brevibacillus sp. strain SKDU10, exhibited cytotoxicity against cancer cell lines such as MCF-7, HEK293T, HT1080, HeLa, and H1299 at less than 10 μM concentration.56

In vitro anticancer activity of BaCf3 in cell culture by MTT assay

The proliferation rate of BaCf3-treated cells was more significant after 48 h of incubation. The proliferation rate was less than 50% for all the concentrations after 72 h of incubation, showing that it is a slow acting anticancer compound ( Fig. 9). The tests were statistically significant with p < 0.05.

FIG. 9.

Anticancer activity of increasing concentration of bacteriocin BaCf3 on A549 cell line by MTT assay. The rate of proliferation decreased with increasing concentration (p < 0.05).

Morphological evaluation of apoptosis

There were several significant morphological changes in the cancer cells that were bacteriocins induced, including cell shrinkage and apoptotic body formation. Untreated cells showed green fluorescence indicating no apoptosis. The intact nuclei of control cells could be visualized in Fig. 10A. Treatment with bacteriocins led to the appearance of red/orange fluorescence indicating onset of apoptosis. The condensed body inside the nucleus was clearly visible (Fig. 10B). The dead cells were visible as orange color in the fluorescent microscopic image (Fig. 10B). All these indicated that there were some bacteriocin-induced apoptotic pathways in A549 cells.

FIG. 10.

Cytochemical staining of cancer cells treated with BaCf3 using fluorescent dyes reveals characteristic features of apoptosis. Changes associated with apoptosis such as nuclear condensation and cell shrinkage were noted as obtained by the acridine-orange/ethidium bromide dual staining (A—control and B—treated cells) (original magniﬁcation 40× for both the images).

Acridine orange/ethidium bromide dual staining is commonly used to detect apoptosis based on the differential uptake of the two fluorescent DNA binding dyes to determine the viable and non-viable cells in a population. Acridine orange is a vital dye and will stain both live and dead cells. Ethidium bromide will stain only cells that have lost their membrane integrity. Live cells appear uniformly green. Early apoptotic cells will stain green and contain bright green dots in the nuclei as a consequence of chromatin condensation and nuclear fragmentation. Late apoptotic cells will incorporate ethidium bromide, and therefore, stain orange, but, in contrast to necrotic cells, the late apoptotic cells show condensed and often fragmented nuclei. Necrotic cells stain orange, but have a nuclear morphology resembling that of viable cells, without condensed chromatin.

Several researchers have studied anticancer activity of bacteriocins. The design and research on AMPs with antitumor properties have high medical value as tumor cells are selectively killed. The cytotoxicity of bacteriocins and their ability to differentially target cancer cells depend on their structural properties like number of positively charged amino acids, hydrophobicity, and ability to form amphipathic structures or oligomerization as in the case of other AMPs.57

Conclusion

Bacteriocin BaCf3 is a potent antibacterial peptide with marked anticancer activity. This bacteriocin was proved to be membrane acting. The docking study supports the anticancer activity of the bacteriocin. In vitro experiments on A549 cell line also support anticancer action of BaCf3. Thus, BaCf3 could be considered as a lead protein for future research.

Author Disclosures: All authors have read and approved the manuscript. Authors declare that they have no conflict of interest.

Footnotes

Acknowledgments

The authors acknowledge project grants from Centre for Marine Living Resources & Ecology-Ministry of Earth Sciences, Government of India (MOES/10-MLR/2/2007 and MOES/10-MLR-TD/03/2013) given to Dr. Sarita G. Bhat, Dept. of Biotechnology, Cochin University of Science and Technology, Kochi, India. The first author also acknowledges Cochin University of Science and Technology, Kochi, India, for the postdoctoral fellowship. A portion of this paper was previously submitted in the thesis of E.S. Bindiya, which can be found at "> https://dyuthi.cusat.ac.in/xmlui/bitstream/handle/purl/5402/T-2447.pdf?sequence=1.

References

Lei

, Sun

, Huang

, et al. The antimicrobial peptides and their potential clinical applications. Am J Translational Res. 2019; 11(7): 3919–3931.

Pirtskhalava

, Vishnepolsky

, Grigolava

, et al. Physicochemical features and peculiarities of interaction of AMP with the membrane. Pharmaceuticals. 2021; 14: 471. DOI: 10.3390/ph14050471

Benfield

, Henriques

Mode-of-action of antimicrobial peptides: Membrane disruption vs. intracellular mechanisms. Front Med Technol. 2020; 2. DOI: 10.3389/fmedt.2020.610997

Lazzaro

, Zasloff

, Rolff

Antimicrobial peptides: Application informed by evolution. Science. 2020; 368: eaau5480. DOI: 10.1126/science.aau5480

Maccari

, Di Luca

, Nifosì

In silico design of antimicrobial peptides. Methods Mol Biol. 2014; 195–219. DOI: 10.1007/978-1-4939-2285-7_9.

Boyette

Heretofore undescribed aerobic spore forming bacillus in child with meningitis. JAMA. 1952; 148: 1223. DOI: 10.1001/jama.1952.62930140008015c

Alebouyeh

, GooranOrimi

, Azimi-Rad

, et al. Fatal sepsis by Bacillus circulans in an immunocompromised patient. Iranian J Microbiol. 2011; 3(3): 156–158.

Russo

, Tarantino

, d’Ettorre

, et al. First report of spondylodiscitis caused by Bacillus circulans in an immunocompetent patient: Clinical case and review of the literature. IDCases. 2021; 23: E01058. DOI: 10.1016/j.idcr.2021.e01058

Bindiya

, Tina

, Sasidharan

, et al. BaCf3: Highly thermostable bacteriocin from Bacillus amyloliquefaciens BTSS3 antagonistic on food-borne pathogens. 3 Biotech. 2019; 9. DOI: 10.1007/s13205-019-1639-2

10.

Gasteiger

, Hoogland

, Gattiker

, et al. Protein identification and analysis tools on the ExPASy server. In Walker

(ed.): The Proteomics Protocols Handbook. Totowa, NJ: Humana Press, 2005: 571–607.

11.

Breiman

Random forests. Machine Learning. 2001; 45: 5–32. DOI: 10.1023/A:1010933404324

12.

Vapnik

The Nature of Statistical Learning Theory. New York: Springer, 1995. DOI: 10.1007/978-1-4757-3264-1.

13.

Karatzoglou

, Smola

, Hornik

, et al. kernlab-AnS4Package for kernel methods in R. J Stat Softw. 2004; 11. DOI: 10.18637/jss.v011.i09.

14.

Yang

, Yan

, Roy

, et al. The I-TASSER suite: Protein structure and function prediction. Nat Methods. 2015; 12: 7–8. DOI: 10.1038/nmeth.3213

15.

Heider

, Barnekow

DNA watermarks: A proof of concept. BMC Mol Biol. 2008; 9: 40. DOI: 10.1186/1471-2199-9-40

16.

Zhang

I-TASSER server for protein 3D structure prediction. BMC Bioinform. 2008; 9. DOI: 10.1186/1471-2105-9-40

17.

Available at https://robetta.bakerlab.org/.

18.

Ferre

, Clote

DiANNA 1.1: An extension of the DiANNA web server for ternary cysteine classification. Nucleic Acids Res. 2006; 34: W182–W185. DOI: 10.1093/nar/gkl189

19.

Sarker

, Nahar

, Kumarasamy

Microtitre plate-based antibacterial assay incorporating resazurin as an indicator of cell growth, and its application in the in vitro antibacterial screening of phytochemicals. Methods. 2007; 42: 321–324. DOI: 10.1016/j.ymeth.2007.01.006

20.

Anju

, Ajayan

, Paulose

Disruption of cerebellar cholinergic system in hypoxic neonatal rats and its regulation with glucose, oxygen and epinephrine resuscitations. Neuroscience. 2013; 236: 253–261. DOI: 10.1016/j.neuroscience.2012.12.056

21.

Schneider

, Rasband

, Eliceiri

NIH image to ImageJ: 25 years of image analysis. Nat Methods. 2012; 9: 671–675. DOI: 10.1038/nmeth.2089

22.

Kozakov

, Hall

, Xia

, et al. The ClusPro web server for protein–protein docking. Nat Protoc. 2017; 12: 255–278. DOI: 10.1038/nprot.2016.169

23.

Alley

, Scudiero

, Monks

, et al. Feasibility of drug screening with panels of human tumor cell lines using a microculture tetrazolium assay. Cancer Res. 1988; 48: 589–601. [PMC][3335022

24.

McGahon

, Martin

, Bissonnette

, et al. The end of the (cell) line: Methods for the study of apoptosis in vitro. Methods Cell Biol. 1995; 46: 153–185. DOI: 10.1016/s0091-679x(08)61929-9.

25.

Gong

, Zhang

, Hu

, et al. Hydrophobic control of the bioactivity and cytotoxicity of de novo-designed antimicrobial peptides. ACS Appl Mater Interfaces. 2019; 11: 34609–34620. DOI: 10.1021/acsami.9b10028

26.

Ebran

, Julien

, Orange

, et al. Isolation and characterization of novel glycoproteins from fish epidermal mucus: Correlation between their pore-forming properties and their antibacterial activities. BBA-Biomembranes. 2000; 1467: 271–280. DOI: 10.1016/S0005-2736(00)00225-X

27.

Yang

, Wang

, Zhang

ResQ: An approach to unified estimation of B-factor and residue-specific error in protein structure prediction. J Mol Biol. 2016; 428(4): 693–701. DOI: 10.1016/j.jmb.2015.09.024

28.

Sugrue

, O’Connor

, Hill

, et al. Actinomyces produces defensin-like bacteriocins (actifensins) with a highly degenerate structure and broad antimicrobial activity. J Bacteriol. 2020; 202. DOI: 10.1128/jb.00529-19.

29.

Singh

, Chittpurna

, Sharma

, et al. Identification, purification and characterization of laterosporulin, a novel bacteriocin produced by Brevibacillus sp. strain GI-9. PLoS One. 2012; 7: E31498. DOI: 10.1371/journal.pone.0031498

30.

Baindara

, Kapoor

, Korpole

, et al. Cysteine-rich low molecular weight antimicrobial peptides from Brevibacillus and related genera for biotechnological applications. World J Microbiol Biotechnol. 2017; 3. DOI: 10.1007/s11274-017-2291-9.

31.

Balouiri

, Sadiki

, Ibnsouda

SK.

Methods for in vitro evaluating antimicrobial activity: A review. J Pharm Anal. 2016; 6: 71–79. DOI: 10.1016/j.jpha.2015.11.005

32.

Maftah

, Renault

, Vignoles

, et al. Membrane permeabilization of Listeria monocytogenes and mitochondria by the bacteriocin mesentericin Y105. J Bacteriol. 1993; 175: 3232–3235. DOI: 10.1128/jb.175.10.3232-3235.1993

33.

Hauge

, Mantzilas

, Moll

, et al. Plantaricin a is an amphiphilic α-helical bacteriocin-like pheromone which exerts antimicrobial and pheromone activities through different mechanisms. Biochemistry. 1998; 37: 16026–16032. DOI: 10.1021/bi981532j

34.

Sutyak

, Wirawan

, Aroutcheva

, et al. Isolation of the Bacillus subtilis antimicrobial peptide subtilosin from the dairy product-derived Bacillus amyloliquefaciens . J Appl Microbiol. 2008; 104: 1067–1074. DOI: 10.1111/j.1365-2672.2007.03626.x

35.

Torrent

, Badia

, Moussaoui

, et al. Comparison of human RNase 3 and RNase 7 bactericidal action at the gram-negative and gram-positive bacterial cell wall. FEBS J. 2010; 277: 1713–1725. DOI: 10.1111/j.1742-4658.2010.07595.x

36.

Sharma

, Dang

, Aruna

, et al. Synergistic antibacterial and anti-biofilm activity of nisin like bacteriocin with curcumin and cinnamaldehyde against ESBL and MBL producing clinical strains. Biofouling. 2020; 36: 710–724. DOI: 10.1080/08927014.2020.1804553

37.

Wang

, Qin

, Xie

, et al. Purification and characterization of plantaricin LPL-1, a novel class IIa bacteriocin produced by Lactobacillus plantarum LPL-1 isolated from fermented fish. Front Microbiol. 2018; 9: 2276. DOI: 10.3389/fmicb.2018.02276.

38.

Kalchayanand

, Dunne

, Sikes

, et al. Viability loss and morphology change of foodborne pathogens following exposure to hydrostatic pressures in the presence and absence of bacteriocins. Int J Food Microbiol. 2004; 91: 91–98. DOI: 10.1016/S0168-1605(03)00324-6

39.

Paik

, Chakicherla

, Hansen

Identification and characterization of the structural and transporter genes for, and the chemical and biological properties of, sublancin 168, a novel lantibiotic produced by Bacillus subtilis 168. J Biol Chem. 1998; 273: 23134–23142. DOI: 10.1074/jbc.273.36.23134

40.

Stein

Bacillus subtilis antibiotics: Structures, syntheses and specific functions. Mol Microbiol. 2005; 56: 845–857. DOI: 10.1111/j.1365-2958.2005.04587.x

41.

Chopra

, Singh

, Kumar Jena

, et al. Sonorensin: A new bacteriocin with potential of an anti-biofilm agent and a food biopreservative. Sci Rep. 2015; 5. DOI: 10.1038/srep13412.

42.

Motta

, Flores

, Souto

, et al. Antibacterial activity of a bacteriocin-like substance produced by Bacillus sp. P34 that targets the bacterial cell envelope. Antonie Leeuwenhoek. 2008; 93: 275–284. DOI: 10.1007/s10482-007-9202-2

43.

Bizani

, Motta

, Morrissy

, et al. Antibacterial activity of cerein 8A, a bacteriocin-like peptide produced by Bacillus cereus . Int Microbiol. 2005; 8: 125–131. [PMC][16052461

44.

Dalmau

, Maier

, Mulet

, et al. Bacterial membrane injuries induced by lactacin F and nisin. Int Microbiol. 2002; 5: 73–80. DOI: 10.1007/s10123-002-0063-2

45.

Sonbol

, Abdel Aziz

, El-Banna

, et al. Antimicrobial activity of bacteriocins produced by enterococcus isolates recovered from Egyptian homemade dairy products against some foodborne pathogens. Int Microbiol. 2020; 23: 533–547. DOI: 10.1007/s10123-020-00127-z

46.

McCafferty

, Cudic

, Yu

, et al. Synergy and duality in peptide antibiotic mechanisms. Curr Opin Chem Biol. 1999; 3: 672–680. DOI: 10.1016/S1367-5931(99)00025-3

47.

Oscáriz

, Lasa

, Pisabarro

Detection and characterization of cerein 7, a new bacteriocin produced by Bacillus cereus with a broad spectrum of activity. FEMS Microbiol Lett. 1999; 178: 337–341. DOI: 10.1111/j.1574-6968.1999.tb08696.x

48.

Heo

, Lee

, et al. Morphological changes induced in Listeria monocytogenes V7 by a bacteriocin produced by Pediococcus acidilactici . J Microbiol Biotechnol. 2007; 17: 663–667. [PMC][18051280

49.

Yamamoto

, Seino

, Fukumoto

, et al. Over-expression of facilitative glucose transporter genes in human cancer. Biochem Biophys Res Commun. 1990; 170: 223–230. DOI: 10.1016/0006-291X(90)91263-R

50.

Rudlowski

, Becker

, Schroder

, et al. GLUT1 messenger RNA and protein production relates to the malignant transformation of cervical cancer. Am J Clin Pathol. 2003; 120: 691–698. DOI: 10.1309/4KYNQM5862JW2GD7

51.

Labak

, Wang

, Arora

, et al. Glucose transport: Meeting the metabolic demands of cancer, and applications in glioblastoma treatment. Am J Cancer Res. 2016; 6(8): 1599.

52.

Cunningham

, Afzal-Ahmed

, Naftalin

Docking studies show that D-glucose and quercetin slide through the transporter GLUT1. J Biol Chem. 2006; 281(9): 5797–5803. [CrossRef][10.1074/jbc.M509422200

53.

Stellrecht

, Gandhi

MET receptor tyrosine kinase as a therapeutic anticancer target. Cancer Lett. 2009; 280(1): 1–14. [CrossRef][10.1016/j.canlet.2008.10.045

54.

Buchanan

, Hendle

, Lee

, et al. SGX523 is an exquisitely selective, ATP-competitive inhibitor of the MET receptor tyrosine kinase with antitumor activity in vivo. Mol Cancer Ther. 2009; 8(12): 3181–3190. [CrossRef][10.1158/1535-7163.MCT-09-0477

55.

Tsai

, Li

, Chen

, et al. Antimicrobial peptide m2163 or m2386 identified from Lactobacillus casei ATCC 334 can trigger apoptosis in the human colorectal cancer cell line SW480. Tumor Biol. 2015; 36(5): 3775–3789. [CrossRef][10.1007/s13277-014-3018-2

56.

Baindara

, Gautam

, Raghava

, et al. Anticancer properties of a defensin like class IId bacteriocin Laterosporulin10. Sci Rep. 2017; 7. DOI: 10.1038/srep46541.

57.

Gaspar

, Veiga

, Castanho

From antimicrobial to anticancer peptides: A review. Front Microbiol. 2013; 1(4): 294. DOI:

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB