RESEARCH ARTICLE

Identification and characterisation of

capidermicin, a novel bacteriocin produced by
Staphylococcus capitis
David Lynch1, Paula M. O’Connor2,3, Paul D. Cotter2,3, Colin Hill3,4, Des Field3,4*,
Máire Begley ID1,3*
1 Department of Biological Sciences, Cork Institute of Technology, Cork, Ireland, 2 Teagasc Food Research
Centre, Moorepark, Fermoy, Cork, Ireland, 3 APC Microbiome Ireland, University College Cork, Cork,
a1111111111 Ireland, 4 School of Microbiology, University College Cork, Cork, Ireland
a1111111111
a1111111111 * [email protected] (MB); [email protected] (DF)
a1111111111
a1111111111
Abstract
One hundred human-derived coagulase negative staphylococci (CoNS) were screened for
antimicrobial activity using agar-based deferred antagonism assays with a range of indicator
OPEN ACCESS
bacteria. Based on the findings of the screen and subsequent well assays with cell free
Citation: Lynch D, O’Connor PM, Cotter PD, Hill C, supernatants and whole cell extracts, one strain, designated CIT060, was selected for fur-
Field D, Begley M (2019) Identification and
ther investigation. It was identified as Staphylococcus capitis and herein we describe the
characterisation of capidermicin, a novel
bacteriocin produced by Staphylococcus capitis. purification and characterisation of the novel bacteriocin that the strain produces. This bac-
PLoS ONE 14(10): e0223541. https://doi.org/ teriocin which we have named capidermicin was extracted from the cell-free supernatant of
10.1371/journal.pone.0223541
S. capitis CIT060 and purified to homogeneity using reversed-phase high performance liq-
Editor: Herman Tse, Department of Health, Hong uid chromatography (RP-HPLC). Matrix-assisted laser desorption/ionization time-of-flight
Kong, HONG KONG
(MALDI-TOF) mass spectrometric (MS) analysis revealed that the capidermicin peptide has
Received: May 28, 2019 a mass of 5,464 Da. Minimal inhibitory concentration (MIC) experiments showed that capi-
Accepted: September 22, 2019 dermicin was active in the micro-molar range against all the Gram-positive bacteria that
were tested. Antimicrobial activity was retained over a range of pHs (2–11) and tempera-
Published: October 16, 2019
tures (10–121˚C x 15 mins). The draft genome sequence of S. capitis CIT060 was deter-
Copyright: © 2019 Lynch et al. This is an open
mined and the genes predicted to be involved in the biosynthesis of capidermicin were
access article distributed under the terms of the
Creative Commons Attribution License, which identified. These genes included the predicted capidermicin precursor gene, and genes that
permits unrestricted use, distribution, and are predicted to encode a membrane transporter, an immunity protein and a transcriptional
reproduction in any medium, provided the original
regulator. Homology searches suggest that capidermicin is a novel member of the family of
author and source are credited.
class II leaderless bacteriocins.
Data Availability Statement: The sequence of the
S. capitis CIT060 genomic region that was
analysed in this study was submitted to Genbank
(accession number MN234131).

Funding: This work was supported by a Cork Introduction

Institute of Technology RISAM PhD scholarship to
DL. The funder had no role in study design, data The rise and spread of multidrug-resistant bacterial pathogens, coupled with a diminishing
collection and analysis, decision to publish, or repertoire of effective antibiotics has necessitated the search for new alternative antimicrobial
preparation of the manuscript. agents. Over the past decade, ribosomally synthesized natural peptides produced by a diverse
Competing interests: The authors have declared group of bacterial species have received attention [1]. Compared to non-ribosomally synthe-
that no competing interests exist. sized antimicrobials, the ribosomally produced peptides are attractive for pharmaceutical

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 1 / 17

 Bacteriocin production by Staphylococcus capitis

applications as they could potentially be bioengineered to improve characteristics such as

potency, stability, solubility etc. [2]. One group of compounds, namely the bacteriocins, have
attracted great interest due to their high potency (often active in the nanomolar range) and
heat stability [3].
It has been suggested that most bacteria produce at least one bacteriocin [4]. While the
exact ecological function of bacteriocins is unknown, they may play a role in competition by
directly killing competing bacteria, function as colonizing peptides or function as signalling
molecules to communicate with other bacteria or the host [5]. Bacteriocin-producing bacteria
have been isolated from a wide variety of sources including food [6], soil [7] and the intestines
of fish and animals [8, 9]. In addition, several members of the human microbiota have been
shown to produce bacteriocins. For example we have previously employed both functional
and in silico approaches in our laboratory to identify novel bacteriocins from the human gut
microbiome [10–15]. Skin-derived bacteria have also been shown to produce, or shown poten-
tial to produce, bacteriocins [16–21].
For the current study, we have focused on examining the antimicrobial ability of coagulase
negative staphylococci (CoNS) that were isolated from human skin or human blood with the aim
of identifying novel bacteriocin-producers. CoNS are considered part of the normal commensal
bacteria of the skin and are thought to act as host guardians by targeting pathogens [22]. We pos-
tulate that bacteriocin production plays an important role in this function and in support of this
hypothesis Nakatsuji and colleagues [23] have shown that antimicrobial-producing CoNS are
deficient in subjects with atopic dermatitis and reintroduction of these strains decreased coloniza-
tion with Staphylococcus aureus. While numerous lantibiotics have been identified from CoNS,
including epidermin from Staphylococcus epidermidis, gallidermin from Staphylococcus gallinar-
ium and hominicin from Staphylococcus hominis [24], analysis of the entire genome sequences of
publicly available CoNS genomes suggests that there are potentially many novel CoNS-derived
bacteriocins that have not yet been functionally characterized (M. Begley; unpublished data).
The aim of this study was to screen one hundred human-derived CoNS for antimicrobial
activity using agar-based deferred antagonism assays with a range of indicator bacteria. One
CoNS strain (CIT060) was selected for further investigation and it was identified as Staphylo-
coccus capitis and herein we describe the purification and characterisation of the novel bacteri-
ocin that the strain produces.

Materials and methods

Bacterial strains and growth conditions
The bacterial indicator strains used in this study are listed in Table 1. Strains were grown in
either Brain Heart Infusion (BHI) broth at 37˚C or in M17 broth supplemented with 0.5% glu-
cose (GM17) at 30˚C as indicated in the Table 1. All media was purchased from Oxoid and
prepared according to the manufacturer’s recommendations.

Assembly of a bank of coagulase negative Staphylococci (CoNS)

The 100 CoNS strains were originally isolated from human skin swabs (obtained by swabbing
the retro auricular crease i.e. behind the ear, the alar crease i.e. the side of the nose or the wrist),
or from human blood samples at Cork University Hospital (CUH). Aliquots from archived
stocks that are stored at -80˚C at Cork Institute of Technology (CIT) were plated onto Mannitol
Salt Agar (MSA) and incubated at 37˚C for 16–18 hours. Single colonies were selected from the
MSA plates and re-streaked onto BHI agar for purity determination. Strains were identified
using Matrix-Assisted Laser Desorption/Ionization Time-Of-Flight (MALDI-TOF) at CUH.
Briefly, single fresh colonies (from BHI agar plates incubated at 37˚C for 16–18 hours) were

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 2 / 17

 Bacteriocin production by Staphylococcus capitis

directly applied to the MALDI-TOF stainless steel target plate. After application, each bacterial
colony was covered with 0.8 μL of matrix solution (10 mg/mL α-cyano-4-hydroxycinnamic
acid [HCAA] in 50% acetonitrile-2.5% trifluoroacetic acid) (Bruker Daltonik, GmbH, Ger-
many). The data collected was classified in accordance to Bruker Taxonomy database of CUH.
The 100 strains were re-stocked in two separate master 96 well plates (50 in each plate).
Overnight cultures of the 100 strains were prepared by selecting a single pure colony from BHI
agar and adding it to 10ml of BHI broth and incubating at 37˚C for 16–20 hours. After incuba-
tion, 100μl of each fresh overnight culture was added to a specific well in the 96 well plate, and
100μl of sterile 80% glycerol was added to each well. The master stock 96 well plates were
stored at -80˚C. Prior to use, the plates were thawed at room temperature.

Screening of a bank of CoNS for antimicrobial activity

Agar-based deferred antagonism assays were carried out with a selection of indicator bacteria
(listed in Table 1). The 100 CoNS strains were replicated onto BHI agar from the 96 well stock

Table 1. Bacterial strains used in this study.

Bacterial strains Culture medium and temperature Source
100 CoNS isolates BHI at 37˚C CIT Culture Collection
Staphylococcus epidermidis TU3298 BHI at 37˚C Teagasc, Culture Collection
Positive control (epidermin producer), used during antimicrobial screening
Staphylococcus capitis CIT060 BHI at 37˚C CIT Culture Collection
Indicator Strains
Bacillus cereus DPC 6089 BHI at 37˚C UCC culture Collection
Enterococcus faecailis MR103 BHI at 37˚C UCC culture Collection
Geobacillus kaustophilus DSM 7263 BHI at 55˚C UCC culture Collection
Geobacillus stearothermophilus ATCC 12930 BHI at 55˚C UCC culture Collection
Lactococcus cremoris IP5 GM17 at 30˚C UCC culture Collection
Lactococcus lactis HP GM17 at 30˚C UCC culture Collection
Lactococcus lactis MG1363 GM17 at 30˚C UCC culture Collection
Lactococcus lactis NZ9800 GM17 at 30˚C UCC culture Collection
Micrococcus luteus DSM1790 BHI at 37˚C UCC culture Collection
Staphylococcus aureus 5971 BHI at 37˚C UCC culture Collection
Staphylococcus aureus DPC 5243 BHI at 37˚C UCC culture Collection
Staphylococcus aureus DPC5297 BHI at 37˚C UCC culture Collection
Staphylococcus aureus NCDO 1499 BHI at 37˚C UCC culture Collection
Staphylococcus aureus Newman BHI at 37˚C UCC culture Collection
Staphylococcus aureus RF122 BHI at 37˚C UCC culture Collection
Staphylococcus epidermidis (UCC strain) BHI at 37˚C UCC culture Collection
Staphylococcus gallinarium 4616 BHI at 37˚C UCC culture Collection
Staphylococcus intermedius DSM 20373 BHI at 37˚C UCC culture Collection
Staphylococcus lugdunesis BHI at 37˚C UCC culture Collection
Staphylococcus pseudintermedius DSM 21284 BHI at 37˚C UCC culture Collection
Streptococcus agalactiae ATCC 13813 BHI at 37˚C UCC culture Collection
Streptococcus dysgalactiae ATCC43078 BHI at 37˚C UCC culture Collection
Streptococcus pneumoniae (UCC strain) BHI at 37˚C UCC culture Collection
Streptococcus pyogenes NCDO 2381 BHI at 37˚C UCC culture Collection

BHI = Brain heart infusion; GM17 = M17 broth + 0.5% glucose; CIT = Cork Institute of Technology; UCC = University College Cork.

https://doi.org/10.1371/journal.pone.0223541.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 3 / 17

 Bacteriocin production by Staphylococcus capitis

plates using a 96-pin replicator (Boekel). Plates were incubated at 37˚C for 16–18 hours after
which the surface of the agar plate was subjected to UV treatment for 30 minutes (High perfor-
mance UV transmitter, Upland, Ca, USA). 30μL of fresh overnight cultures of indicator strains
were added to 20ml of relevant sloppy/soft agar (BHI/GM17 broth supplemented with 0.75%
w/v agar) and poured over the replicated plates. Plates were incubated at 37˚C for 12–16 hours
after which they were examined for zones of inhibition.

Investigation of the antimicrobial activity of cell-free supernatants and

crude whole cell extracts
Short-listed CoNS were grown overnight in BHI broth at 37˚C under vigorous continued
shaking (130 rpm), and a 1% inoculum of the cultures were added to 50 ml clarified BHI broth
(prepared by passing BHI broth through XAD-16N beads (Sigma-Aldrich) prior to autoclav-
ing) and incubated with vigorous continued shaking at 37˚C. Following incubation, 50 ml of
bacterial cells were centrifuged at 7,000 rpm for 20 minutes, supernatant removed and
retained, i.e., cell-free supernatant (CFS). The cell pellets were resuspended in 7 ml 70% iso-
propanol (IPA) 0.1% trifluoroacetic acid (TFA) and stirred vigorously for 3 hours. Cell debris
was removed through centrifugation and the supernatants were retained and referred to as
whole cell extracts (WCE). CFS and WCE were examined for antimicrobial activity in an agar
well diffusion assay using L. lactis HP as the indicator strain. 50ml of molten agar was seeded
with 100 μL of L. lactis HP that was grown overnight at 30˚C, and once solidified, 4.6mm holes
were bored with a sterile glass pipette. 50 μL of CFS and WCE were added to separate wells.
Plates were incubated at 30˚C for 16–18 hours after which they were examined for zones of
inhibition.

Purification of capidermicin from S. capitis CIT060

Capidermicin was purified from S. capitis CIT060 using a method described by Field et al. [25]
with modifications. Three litres of clarified BHI (cBHI) broth was inoculated (1%) with S. capi-
tis CIT060 and incubated for 18–20 hours at 37˚C under vigorous continued shaking (130
rpm). The culture was centrifuged at 7,000 rpm for 15 minutes. The cell pellet was removed,
and the supernatant was retained and passed through 60g of Amberlite XAD16N beads (Sigma
Aldrich). The beads were washed with 30% ethanol, and the peptide was eluted in 500ml 70%
isopropanol (IPA) containing 0.1% trifluoroacetic acid (TFA). The IPA was evaporated using a
rotary evaporator (Buchi) and the sample pH adjusted to 4 before applying to a 60 ml Strata C-
18 E column (Phenomenex) that was previously pre-equilibrated with 60 ml methanol (Fisher
Scientific, UK) and 60 ml H2O. 100 ml of 30% ethanol was used to wash the column and the
peptide was eluted in 60 ml of 70% IPA, 0.1% TFA. 10 ml aliquots were concentrated to 2 ml
through the removal of IPA by rotary evaporation. 2.0 ml aliquots were applied to a Phenom-
enex (Phenomenex, Cheshire, UK) C12 reverse phase (RP)-HPLC column (Jupiter 4u proteo
90 Å, 250 × 10.0 mm, 4 μm) previously equilibrated with 25% acetonitrile containing 0.1%
TFA. The column was subsequently developed in a gradient of 25% acetonitrile containing
0.1% TFA to 50% acetonitrile containing 0.1% TFA from 10 to 45 minutes at a flow rate of 2ml
min-1. The relevant active fractions were collected and pooled, subjected to rotary-evaporation
to remove the acetonitrile and freeze-dried. The purity of the peptide was analysed by MAL-
DI-TOF Mass Spectrometry [26].

Minimum inhibitory concentration (MIC) assays

MIC determinations were carried out in triplicate in 96 well microtitre plates pre-treated with
bovine serum albumin (BSA) as described by [27]. 200μL of 1X phosphate buffered saline

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 4 / 17

 Bacteriocin production by Staphylococcus capitis

(PBS) containing 1% (w/v) bovine serum albumin (PBS/BSA) was added to each well and
incubated for 30 minutes at 37˚C. The wells were washed with 200μL PBS and allowed to dry.
Target strains were grown overnight in the appropriate medium and temperature conditions,
sub-cultured into fresh broth and allowed to grow to an OD600 of approximately 0.5, and
diluted to a final concentration of 105 CFU/ml in a volume of 200μL broth. Lyophilised capi-
dermicin was resuspended in cation adjusted BHI broth to a desired concentration, and a 2—
fold dilution of the peptide was made in the 96 well plate. The target strain was then added and
after incubation at 30˚C or 37˚C for 16 hours the MIC was read as the lowest peptide concen-
tration causing inhibition of visible growth.

Stability assays with capidermicin

The susceptibility of purified capidermicin peptide to temperature, pH and protease enzymes
was investigated through well diffusion assays. To determine temperature stability the purified
peptide was subjected to 10, 30, 40, 50, 80, 90, 121˚C for 15 minutes. Bioactivity was then
determined by carrying out an agar well diffusion assay with L. lactis HP as the bacterial indi-
cator. To evaluate the susceptibility of the peptide to varying pH values, the purified peptide
solution was adjusted to pH 2–11 using 1M HCl or 1M NaOH, respectively. After a brief vor-
tex, the peptide was incubated at room temperature for 15 minutes, and the bioactivity was
again determined with the well diffusion assay with L. lactis HP. The susceptibility of the pep-
tide to proteolytic cleavage was analysed using trypsin, α-chymotrypsin, pepsin and proteinase
K (Sigma-Aldrich). Protease enzymes were dissolved in 100 mM Tris-HCl– 10 mM CaCl2, to a
final concentration of 100 μg/ml. Preparations of capidermicin were incubated with the vari-
ous enzymes at 37˚C for 4 hours and the bioactivity of capidermicin was reassessed using the
well diffusion assay described above.

Sequencing of S. capitis CIT060 genome

DNA extracted from S. capitis CIT060 was quantified using a Qubit high sensitivity assay
(Invitrogen), and diluted to 0.2ng/μl. Genomic libraries were then prepared using the Nextera
XT Library preparation kit (Illumina) essentially as described in the manufacturer’s protocol
with the following exceptions. Firstly, the tagmentation time was extended to 7min. Following
addition of indices, products were cleaned using AMPure XP magnetic bead-based purifica-
tion, as described in the manufacturer’s protocol and then secondly, in place of the bead based
normalisation, the products were run on an Agilent Bioanalyser to determine average frag-
ment size (Agilent) and quantified again by Qubit. Cleaned genomic fragments were then
pooled equimolarly. The sample pool (4nM) was denatured with 0.2N NaOH, then diluted to
6pM and combined with 10% (v/v) denatured 6pM PhiX, prepared following Illumina guide-
lines. Samples were sequenced on the MiSeq sequencing platform in the Teagasc sequencing
facility, Moorepark, Fermoy, using a 2 x 300 cycle V3 kit, following standard Illumina sequenc-
ing protocols.

In silico analyses of the predicted capidermicin gene cluster and encoded

proteins
Following sequencing, the reads were assembled using Spades v. 3.5.0 [28]. Opening reading
frames (ORFs) were predicted using Prodigal V.1.20 [29] and assigned a putative function
based on BLASTp analysis at NCBI (http://www.ncbi.nlm.nih.gov/) and Pfam matches
(EMBL-EBI) [30]. Any genomic regions which were identified as potentially containing anti-
microbial-encoding genes were visualised using Snapgene Viewer (GSL Biotech; available at
www.snapgene.com). These regions were manually annotated, and BLAST searches were

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 5 / 17

 Bacteriocin production by Staphylococcus capitis

performed with ORFs. Phyre2 (http://www.sbg.bio.ic.ac.uk/~phyre2) [31]) was used to gener-

ate a putative three-dimensional structure of the capidermicin peptide using two homologous
peptides as templates—aureocin A53 (accession number AAN71834.1) and lacticin Q (acces-
sion number BAM66973.1). The sequence alignment of peptides of interest was performed
using CLUSTAL OMEGA (https://www.ebi.ac.uk/Tools/msa/clustalo/).

Results
Screening a bank of CoNS strains for antibacterial activity
A bank of 100 CoNS strains was assembled. All strains were phenotypically characterised
using MSA agar plates and identified by MALDI-TOF analysis. The bank consisted of various
staphylococcal species including 83 S. epidermidis, 7 S. capitis, 3 S. haemolyticus, 3 S. hominis, 2
S. warneri, 1 S. saprophyticus and 1 S. simulans. The 100 strains were examined using agar-
based deferred antagonism assays for their ability to inhibit a selection of indicator strains (24
in total), including enterococci, lactococci, Micrococcus, streptococci and other Staphylococcus
strains. Zones of inhibition were observed for 94 strains; representative images are shown in
Fig 1. Six of the S. epidermidis strains did not demonstrate antimicrobial activity under the
conditions tested. 15 of the 94 strains displayed a broad spectrum of activity, inhibiting 10 or
more indicator strains. One strain, namely S. capitis CIT060, was capable of inhibiting 14 of
the 24 bacterial indicators (B. cereus DPC6089, E. faecailis MR103, G. kaustophilius DSM7263,
L. lactis subsp cremonis IP5, L. lactis HP, M. luteus DSM1790, S. aureus NCDO1499,
DPC5297, Newman, and RF122, S. lugdunensis, S. pseudintermedius DSM21284, S. intermedius
DSM 20373 and S. dysgalactiae ATCC43078).

Purification of capidermicin from S. capitis CIT060

Initial experiments with cell-free supernatant and crude whole cell extracts prepared from S.
capitis CIT060 suggested that the antimicrobial produced by the strain is primarily in the super-
natant (data not shown). Consequently, efforts focused on purifying an antimicrobial peptide
from culture supernatants as described in the Methods section. MALDI-TOF MS analysis
revealed that the purified peptide, that we named capidermicin, had a mass of 5,464 Da (Fig 2).
The antimicrobial activity of purified capidermicin, as determined by agar well diffusion
assays with L. lactis HP, remained unaffected following heat treatment at 10, 30, 40, 50, 80, 90
or 121˚C for 15 minutes. The peptide was also able to maintain full bioactivity following expo-
sure to pH 2–8. A 70% decrease in zone of inhibition observed at pH 9–11. Exposing capider-
micin to α-chymotrypsin and pepsin had no effect on the bioactivity of the peptide. However,
treatment with trypsin or proteinase K resulted in a 50% reduction in activity.
The specific activity of capidermicin was assessed using standard MIC broth-based assay
and results are presented in Table 2. It was observed that capidermicin was active against the
selected target bacteria in the nano- and micromolar range.

In silico analyses of the predicted capidermicin gene cluster and encoded

proteins
The draft genome of S. capitis CIT060 was analysed using a variety of in silico tools for the
presence of potential antimicrobial peptide encoding genes. Three areas of interest were iden-
tified which included a lantibiotic gene cluster, a phenol soluble modulin (PSM) gene cluster
and an aureocin-like gene cluster.
The first area of interest contains a gene that is predicted to encode a 44 amino acid peptide
that is homologous to the lantibiotic gallidermin of Staphylococcus gallinarum (accession

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 6 / 17

 Bacteriocin production by Staphylococcus capitis

Fig 1. Representative images of the results obtained during the screen of 100 CoNS strains for antimicrobial activity. CoNS were replicated from master stock 96
well plates onto BHI agar using a 96-pin replicator. Plates were incubated at 37˚C overnight after which they were overlaid with sloppy agar containing relevant
indicator bacteria. For the plates shown the indicators used were (A) M. luteus DSM1790, (B) S. aureus NCDO1499 and (C) S. pseudintermedius DSM21284. The arrows
indicate the position of S. capitis CIT060 on the plates.
https://doi.org/10.1371/journal.pone.0223541.g001

number U61158.1). The 44 amino acid peptide is predicted to encode a peptide with a mass of
5 kDa. However, a corresponding mass could not be detected from our bioactive HPLC frac-
tions or purified capidermicin preparations. The second area of interest contains four genes
that are predicted to encode PSMβ peptides. Phenol-soluble modulins (PSMs) are a recently
discovered group of amphipathic peptides that have multiple roles in staphylococcal pathogen-
esis. They have been shown to exhibit antimicrobial activity [32]. All four predicted peptides
contain the conserved domain of staph_haemo superfamilies (pfam05480), their amino acid
sequences are identical to the PSMβ peptides previously reported in the literature [16], [21]
and they are predicted to have masses of 4.57 kDa, 4.54 kDa, 4.62 kDa and 4.79 kDa. Again,
corresponding masses could not be detected from our bioactive HPLC fractions or purified
capidermicin preparations. The genetic organization of the final area of interest is shown in
Fig 3A and the predicted functions of the putative gene products are shown in Table 3. Orf4
was predicted to encode a peptide that is homologous to a number of previously characterised
bacteriocins including lacticin Z produced by Lactococcus lactis QU14 (46% identity; accession
number BAF75975), aureocin A53 produced by Corynebacterium jeikeium (41% identity;
accession number WP010976360) and an aureocin-like bacteriocin produced by Lactococcus
ruminis (57% identity; accession number SEM89646). More distant homologues include
BacSp222 (32% identity; accession number A0A0P0C3P7) and Lactolisterin BU (44% identity;
accession number SDR48784). A Clustal Omega alignment was carried out to compare the

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 7 / 17

 Bacteriocin production by Staphylococcus capitis

Fig 2. (A) Reversed–phase high performance liquid chromatography (RP-HPLC) profile for the purification of capidermicin using a Phenomenex C12 reverse-phase
column, at a flow rate of 2ml/min. (B) MALDI-TOF Mass Spectrometry of lyophilized capidermicin revealed a mass of 5,464 Da. MALDI TOF MS chromatogram above
indicates the presence of capidermicin (5464.39) and the K+ adduct ion (5502.61) and the doubly charged ion (2732.69) (2732 x 2 = 5464). (C) Antimicrobial activity of
HPLC bioactive fractions was determined using well diffusion assay using L. lactis HP as the indicator strain. GM17 agar was seeded with L. lactis HP, wells were bored,
50μL of the HPLC fraction was added to the well, and plates were incubated at 30˚C for 16 hours.
https://doi.org/10.1371/journal.pone.0223541.g002

predicted peptide encoded by orf4 and previously characterised bacteriocins, and the related-
ness of the peptides is depicted in Fig 4.
The mass of the putative orf4-encoded peptide was predicted to be 5,438 Da by in silico
tools (under Genbank accession MN234131). However, as the homologue aureocin A53 con-
tains an N-formylated methionine [33]), and the start codon of orf4 was noted to be TTG (Fig
3B), a revised theoretical mass of 5,466 Da was predicted. This mass is virtually identical to the
mass of the capidermicin peptide that we purified from S. capitis CIT060 (Fig 2).

Table 2. Minimum inhibitory concentration (MIC) values of purified capidermicin against a range of Gram posi-
tive indicators. Identical MICs values were obtained in three independent determinations.
Species and Strain
Lactococcus lactis HP 19 μg/ml 3.4 μM
Staphylococcus aureus NCDO 1499 3.1 μg/ml 0.6 μM
Staphylococcus aureus SA113 10 μg/ml 1.8 μM
Staphylococcus intermedius DSM20373 40 μg/ml 7.3 μM
Staphylococcus pseudintermedius DSM21284 10 μg/ml 1.8 μM
Staphylococcus pseudintermedius DK729 10 μg/ml 1.8 μM
Micrococcus luteus DSM1790 100μg/ml 18 μM
https://doi.org/10.1371/journal.pone.0223541.t002

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 8 / 17

 Bacteriocin production by Staphylococcus capitis

Fig 3. (A) Organisation of the genomic region that is predicted to encode capidermicin. Open reading frames (ORFs)/genes are coloured according to the predicted
function. (B) The nucleotide sequence of the 153 bp ORF4 that is predicted to encode capidermicin. The deduced amino acid sequence is shown under the DNA
sequence. The start and stop codon, TTG and TAA, respectively, are shown in bold and underlined.
https://doi.org/10.1371/journal.pone.0223541.g003

Table 3. In silico analysis of the genes predicted to be involved in the biosynthesis of capidermicin.

ORF 1 60_02323 738 Plasmid replication initiator protein of Listeria monocytogenes 98% Plasmid replication
(WP_096929472.1)
ORF 2 60_02322 1407 YdbT-like protein of S. aureus 29% Self-immunity
(WP_032072953.1)
ORF 3 60_02321 405 YdbS-like protein of S. intermedius 38% Self-immunity
(COG3402)
ORF 4 60_02320 153 Aureocin A53 –like protein 46% Structural gene
(AF447813)
ORF 5 60_02319 558 Membrane protein of Staphylococcus sp.TE8 76% Unknown
WP_082243241.1)
ORF 6 60_02318 243 Hypothetical protein of S. epidermidis 95% Unknown
(WP_049397332.1)
ORF 7 60_02317 624 Transposase of Staphylococcus 100% Transposon
(WP_017176851.1)
ORF 8 60_02316 516 Hypothetical protein 99% Unknown
(WP_020368224.1)
ORF 9 60_02315 996 Putative sulfate exporter family transporter 100% Transport of peptide
(WP_070441690.1)
ORF 10 60_02314 822 Transcriptional Regulator 100% Gene Regulation
(WP_070441693.1)
https://doi.org/10.1371/journal.pone.0223541.t003

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 9 / 17

 Bacteriocin production by Staphylococcus capitis

Fig 4. Alignment of the capidermicin amino acid sequence with homologous bacteriocins; epidermicin NI01 from
S. epidermidis strain 224 (JQ025383), mutacin BHT-B from Streptococcus ratti BHT (DQ145753), lactolisterin BU
from L. lactis subsp. lactis bv. diacetylactis BGBU1-4 (SDR48784), lacticin Q from L. lactis QU5 (BAF57910),
lacticin Z from L. lactis QU14 (BAF75975), aureocin-like produced by numerous bacteria (SEM89646), and
aureocin A53 from S. aureus A53 (WP_010976360). Highlighted residues indicate conserved amino acid sequences.
https://doi.org/10.1371/journal.pone.0223541.g004

Capidermicin is predicted to be cationic with a putative net charge of 5.34 (theoretical

pI = 10.22) and is rich in lysine residues (14%). A 3D structural model is presented in Fig 5.
The peptide is predicted to be composed of four α-helices.

Discussion
The aim of the current study was to screen 100 human-derived CoNS for antimicrobial activ-
ity. Agar-based deferred antagonism assays revealed that 94 of the strains reproducibly inhib-
ited at least some of the tested indicator bacteria. Six strains did not show antimicrobial
activity under the test conditions employed but it is possible that these strains may demon-
strate activity under other conditions, such as different growth medium or different indicator
bacteria. Janek and colleagues [18] carried out a similar agar-based screen to determine the fre-
quency of antimicrobial production by 89 human nasal Staphylococcus isolates. 77 of the 89
isolates (86.5%) exhibited antimicrobial activity. When taken together, the findings of both
studies suggest that antimicrobial production is a common phenotype among CoNS isolates.
In contrast, a recent study by O’ Sullivan et al. [19], which describes a similar screen with
human skin-derived staphylococci reports that only 101 possible antimicrobial-producers
were identified from over 90,000 colonies that were screened for antimicrobial activity. The

Fig 5. Putative three-dimensional structure of capidermicin. A rainbow colour scheme is used to indicate the N-
terminus in blue, and the C-terminus in red. The structure was generated using Phyre2.
https://doi.org/10.1371/journal.pone.0223541.g005

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 10 / 17

 Bacteriocin production by Staphylococcus capitis

difference in the frequency of isolation of antimicrobial producers between that study

(0.112%), the study by Janek et al. (86.5%) and our study (94%) may be because O’ Sullivan
et al. used only one indicator organism in their screen (Lactobacillus delbrueckii subsp. bulgari-
cus) while a variety of indicators were used in the other two studies. It is possible that antimi-
crobials produced by CoNS may not inhibit L. delbruecki or activity may be too low to be
detected in deferred antagonism assays.
Interestingly, functional screens of human intestine-derived bacteria report isolation of
antimicrobial producers at a low frequency. Lakshminarayanan et al. [11] screened over
70,000 faecal bacteria for their ability to inhibit the indicators Lactobacillus bulgaricus
LMG6901 and Listeria innocua DPC3572 and only identified 55 antimicrobial producers (a
frequency of 0.08%). The in silico-based investigations of Zheng and colleagues [34] revealed
that the human gut microbiome had the lowest frequency of putative bacteriocin genes of all
the human body sites investigated. It is possible that production of antimicrobials by CoNS
will increase their fitness in order to compete and survive on human skin. The limited avail-
ability of nutrients and water in this environment compared to the gastrointestinal tract may
mean that the skin is a more competitive environment hence explaining the higher frequency
of antimicrobial producers in screens with skin-derived bacteria compared to gut-derived bac-
teria. As humans and their microbes have co-evolved it is likely that production of antimicro-
bials by CoNS plays a beneficial role for the host, perhaps by contributing to the role of the
skin as the body’s first line of defence by protecting against pathogenic bacteria. It is also possi-
ble that the antimicrobials may act as signalling molecules and interact with the human
immune system [5].
The 94 CoNS that demonstrated antimicrobial activity in the initial screen were shortlisted
to 15 based on their inhibition spectra and the activity of cell-free supernatants and whole cell
extracts in well assays. S. capitis CIT060 was selected for further investigation as it demon-
strated a broad inhibition spectrum and the results of well assays suggested that the antimicro-
bial could potentially be purified using methods that we currently use for bacteriocins in our
lab [25]. We also noted that there are very few reports of functionally characterized S. capitis
antimicrobials in the literature. While antimicrobial production by S. capitis strains has been
shown by agar-based studies [18, 19] and antimicrobial genes have been identified in S. capitis
genomes by in silico based methods [16, 35], to our knowledge the only two functional charac-
terization studies in the literature are those of Sugai et al. [36], who reported the purification of
the glycylglycine endopeptidase ALE-1 from S. capitis EPK1 that is similar to the bacteriocin
lysostaphin, and Kumar et al. [21] who chemically synthesized and characterized phenol solu-
ble modulins.
A 5,464 Da peptide, that we named capidermicin, was purified from S. capitis CIT060 cell
free supernatants. The antimicrobial activity of the peptide was confirmed, and it was shown
to retain its activity over a range of pH and temperatures. Analysis of the S. capitis CIT060
genome revealed the presence of 6 potential antimicrobial encoding genes (2 bacteriocins, 4
PSMs) but based on mass we deduced that the antimicrobial peptide that we purified was the
product of the aureocin-like gene cluster. A gene encoding a potential 44 amino acid lantibio-
tic similar to gallidermin was also identified in the genome of S. capitis CIT060. Carson et al.
[35] reported the presence of lanthipeptide gene clusters in the genomes of the 12 S. capitis iso-
lates analysed (S. capitis 1319, 3379, 3769, 4275, 6079, 807, 1187, 1642, 2477, 2643, 4830 and
5871). Three of these strains were also shown to contain a sactipeptide gene cluster (S. capitis
1319, 2487 and 3379) [35]). Kumar et al. [21] identified four distinct gene clusters with the
ability to encode antimicrobial peptides (epidermicin, gallidermin and phenol-soluble modu-
lins) in S. capitis TE8. While the lantibiotic did not seem to be produced by S. capitis CIT060
under the experimental conditions used, the production of more than one bacteriocin by a

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 11 / 17

 Bacteriocin production by Staphylococcus capitis

bacterium is not uncommon. Lactococci commonly produce more than one bacteriocin and
Lactococcus lactis subsp. lactis bv. diacetylactis BGBU1-4 strain produces at least two bacterio-
cins [37, 38]. Similarly, Staphylococcus aureus 4185, a bovine mastitis isolate, was shown to
produce five antimicrobial peptides (named peptides A–E) [39].
In silico analyses suggest that capidermicin is a novel member of the class II leaderless bacte-
riocin family. It shows most similarity to members of the aureocin 53-like sub-group of the fam-
ily which includes aureocin A53 produced by S. aureus A53 [33], lacticin Z produced by L. lactis
QU14 [40], lacticin Q produced by L. lactis QU5 [41] and epidermicin NI01 produced by S. epi-
dermidis strain 224 [42]. All of the bacteriocins within the group are 34–53 amino acid peptides,
are highly cationic and are characterised by their lack of an N-terminal leader sequence during
biosynthesis meaning that they do not undergo any post-translational modifications and
become active shortly after translation [43]. While the genes encoding the immunity and secre-
tion machinery have been experimentally determined for lacticin Q and Z [44], [45], aureocin
A53 [46] and aureocin A70 [47]; [48], compared to other the classes of bacteriocins there is little
known about the biosynthesis of leaderless bacteriocins and they have been referred to as the
most enigmatic and poorly understood group of bacteriocins [43]. Similar to lacticin Q (48%)
and aureocin A53 [49], capidermicin is predicted to be α-helical globular molecule.
MIC assays revealed that capidermicin was active against Gram-positive bacteria at low
concentrations (μM/nM). Similar findings have been reported for epidermicin NI01 [42] and
lacticin Q [41]. While capidermicin is insensitive to α-chymotrypsin and pepsin, a 70% reduc-
tion in activity was observed when it was treated with proteinase K or trypsin. A similar
decrease in activity has previously been observed for epidermicin NI01, which displayed a 75%
and 50% reduction in activity for proteinase K and trypsin, respectively [42]. Resistance to pro-
teases has been reported for other staphylococcal bacteriocins, including aureocin A53 and
BacCH91 [33]; [50]. Capidermicin showed high stability under acidic, alkaline and neutral
conditions, which has also been reported for aureocin A53 [33], lacticin Z [40] and lacticin Q
[41]). It has previously been noted that the high stability of leaderless bacteriocins together
with the simplicity of their biosynthesis may make them more attractive from a commercial
view point compared to other bacteriocins [43].
A 3D structural model is presented in Fig 5. The peptide is predicted to be composed of
four α-helices, and exhibits a recurring three dimensional structural motif found among many
linear leaderless bacteriocins (lacticin Q, aureocin A53) and similar to that found in a larger
superfamily of proteins known as saposin-like peptides [51]. All the helices of capidermicin
are amphipathic whereby hydrophobic residues are oriented inward that pack to give a hydro-
phobic core and the hydrophilic residues are exposed on the surface in the same manner as
LnqQ (data not shown). Similarly, both peptides have highly cationic surfaces. Capidermicin
has 7 lysine residues which are well distributed throughout its primary structure. Indeed, all
lysine residues are situated on the surface and are in fact found in the exact same locations to
LnqQ (in capidermicin K3, K10, K14, K23, K44 and K50). Notably, although LnqQ and the
closely related aureocin A53 are composed of four distinct α-helical structures that are struc-
turally identical to each other, both have varying antimicrobial activity spectrums against
Gram-positive bacteria, [49]. Studies have shown that LnqQ permeates target membranes by
forming toroidal pores (4.6–6.6 nm in diameter), which facilitate the leakage of cellular con-
tents [52]. However, a more recent study demonstrated that LnqQ was able to induce cell
death even without the formation of pores [53]. AucA was also proposed to permeabilize cell
membranes causing the leakage of essential molecules, dissipation of membrane potential, and
cessation of macromolecular synthesis but without the formation of discrete pores [54]. Given
the structural similarity of capidermicin to LnqQ and A53, it is likely that capidermicin can
also permeabilize cell membranes resulting in cell death.

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 12 / 17

 Bacteriocin production by Staphylococcus capitis

In conclusion, we report that our screening experiment revealed a large frequency of anti-
microbial production by human CoNS isolates and we describe the subsequent identification
and characterization of a novel bacteriocin from S. capitis. Future work in our laboratory will
include the chemical synthesis of the capidermicin peptide for additional in vitro experiments
to confirm activity and antimicrobial spectrum, similar to studies conducted on other lead-
erless bacteriocins including garvicin KS and epidermicin NI01 [55, 56]. In the case of epider-
micin NI01, chemical synthesis permitted the generation of sufficient peptide for further
analyses including in vivo studies [57, 58]. Moreover, accessibility to capidermicin by chemical
synthesis would provide a means for peptide engineering investigations to be carried out i.e.
molecular engineering to enhance the potency, improve pharmacological properties, increase
peptide stability and potentially modify the spectrum of activity. It may also provide more
detail regarding the importance of the formylated methionine at the N-terminus for the anti-
microbial activity of capidermicin. Future experiments will also include mutational analysis of
all of the genes that are predicted to be involved in capidermicin production. A comprehensive
gene disruption of the capidermicin biosynthetic cluster will be carried out, similar to that of
Iwatani and colleagues [56] whereby each gene of the lacticin Q operon (lnqQBCDEF) was
individually deleted and the impact on production and immunity evaluated.

Acknowledgments
We are grateful to Dr. Fiona Crispie (Teagasc, Moorepark) for kindly giving us access to
genome sequencing platforms and experimental assistance. We would like to acknowledge the
guidance of Dr. Alan Lucid in the annotation of the Staphylococcus capitis CIT060 genome.

Author Contributions
Conceptualization: Máire Begley.
Formal analysis: David Lynch, Máire Begley.
Funding acquisition: Máire Begley.
Investigation: David Lynch, Paula M. O’Connor.
Methodology: Máire Begley.
Project administration: Des Field, Máire Begley.
Resources: Paul D. Cotter, Colin Hill.
Supervision: Des Field, Máire Begley.
Visualization: David Lynch, Máire Begley.
Writing – original draft: David Lynch, Des Field, Máire Begley.
Writing – review & editing: David Lynch, Paula M. O’Connor, Paul D. Cotter, Colin Hill, Des
Field, Máire Begley.

References
1. Arnison PG, Bibb MJ, Bierbaum G, Bowers AA, Bugni TS, Bulaj G, et al. Ribosomally synthesized and
post-translationally modified peptide natural products: overview and recommendations for a universal
nomenclature. Nat Prod Rep. 2013; 30(1):108–60. https://doi.org/10.1039/c2np20085f PMID:
23165928
2. Perez RH, Zendo T, Sonomoto K. Novel bacteriocins from lactic acid bacteria (LAB): various structures
and applications. Microbial Cell Factories. 2014; 13(Suppl 1):S3–S. https://doi.org/10.1186/1475-2859-
13-S1-S3 PMID: 25186038

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 13 / 17

 Bacteriocin production by Staphylococcus capitis

3. Cotter PD, Ross RP, Hill C. Bacteriocins—a viable alternative to antibiotics? Nature Reviews Microbiol-
ogy. 2013; 11(2):95–105. https://doi.org/10.1038/nrmicro2937 PMID: 23268227
4. Riley MA, Wertz JE. Bacteriocins: Evolution, Ecology, and Application. Annual Review of Microbiology.
2002; 56(1):117–37. https://doi.org/10.1146/annurev.micro.56.012302.161024 PMID: 12142491
5. Dobson A, Cotter PD, Ross RP, Hill C. Bacteriocin Production: a Probiotic Trait? Applied and Environ-
mental Microbiology. 2012; 78(1):1–6. https://doi.org/10.1128/AEM.05576-11 PMID: 22038602
6. Kelly WJ, Asmundson RV, Huang CM. Isolation and characterization of bacteriocin-producing lactic
acid bacteria from ready-to-eat food products. International journal of food microbiology. 1996; 33(2–
3):209–18. https://doi.org/10.1016/0168-1605(96)01157-9 PMID: 8930706
7. Yanagida F, Chen YS, Shinohara T. Searching for bacteriocin-producing lactic acid bacteria in soil. J
Gen Appl Microbiol. 2006; 52(1):21–8. Epub 2006/04/07. PMID: 16598155.
8. Collins FWJ, O’Connor PM, O’Sullivan O, Rea MC, Hill C, Ross RP. Formicin–a novel broad-spectrum
two-component lantibiotic produced by Bacillus paralicheniformis APC 1576. Microbiology. 2016; 162
(9):1662–71. https://doi.org/10.1099/mic.0.000340 PMID: 27450592
9. Ringø E, Hoseinifar SH, Ghosh K, Doan HV, Beck BR, Song SK. Lactic Acid Bacteria in Finfish—An
Update. Frontiers in Microbiology. 2018; 9:1818–. https://doi.org/10.3389/fmicb.2018.01818 PMID:
30147679
10. Begley M, Cotter PD, Hill C, Ross RP. Identification of a Novel Two-Peptide Lantibiotic, Lichenicidin, fol-
lowing Rational Genome Mining for LanM Proteins. Applied and Environmental Microbiology. 2009; 75
(17):5451–60. https://doi.org/10.1128/AEM.00730-09 PMID: 19561184
11. Lakshminarayanan B, Guinane CM, O’Connor PM, Coakley M, Hill C, Stanton C, et al. Isolation and
characterization of bacteriocin-producing bacteria from the intestinal microbiota of elderly Irish subjects.
Journal of Applied Microbiology. 2013; 114(3):886–98. https://doi.org/10.1111/jam.12085 PMID:
23181509
12. O’Connor PM, O’Shea EF, Guinane CM, O’Sullivan O, Cotter PD, Ross RP, et al. Nisin H Is a New
Nisin Variant Produced by the Gut-Derived Strain Streptococcus hyointestinalis DPC6484. Applied and
Environmental Microbiology. 2015; 81(12):3953–60. https://doi.org/10.1128/AEM.00212-15 PMID:
25841003
13. O’Sullivan O, Begley M, Ross RP, Cotter PD, Hill C. Further Identification of Novel Lantibiotic Operons
Using LanM-Based Genome Mining. Probiotics and Antimicrobial Proteins. 2011; 3(1):27–40. https://
doi.org/10.1007/s12602-011-9062-y PMID: 26781497
14. Collins FWJ, O’Connor PM, O’Sullivan O, Gómez-Sala B, Rea MC, Hill C, et al. Bacteriocin Gene-Trait
matching across the complete Lactobacillus Pan-genome. Scientific Reports. 2017; 7(1):3481–. https://
doi.org/10.1038/s41598-017-03339-y PMID: 28615683
15. Walsh CJ, Guinane CM, Hill C, Ross RP, O’Toole PW, Cotter PD. In silico identification of bacteriocin
gene clusters in the gastrointestinal tract, based on the Human Microbiome Project’s reference genome
database. BMC Microbiology. 2015; 15(1):183. https://doi.org/10.1186/s12866-015-0515-4 PMID:
26377179
16. Cameron DR, Jiang J-H, Hassan KA, Elbourne LDH, Tuck KL, Paulsen IT, et al. Insights on virulence
from the complete genome of Staphylococcus capitis. Frontiers in microbiology. 2015; 6:980–. https://
doi.org/10.3389/fmicb.2015.00980 PMID: 26441910
17. Zipperer A, Konnerth MC, Laux C, Berscheid A, Janek D, Weidenmaier C, et al. Human commensals
producing a novel antibiotic impair pathogen colonization. Nature. 2016; 535(7613):511–6. https://doi.
org/10.1038/nature18634 PMID: 27466123
18. Janek D, Zipperer A, Kulik A, Krismer B, Peschel A, Bertram R. High Frequency and Diversity of Antimi-
crobial Activities Produced by Nasal Staphylococcus Strains against Bacterial Competitors. PLOS
Pathogens. 2016; 12(8):e1005812–e. https://doi.org/10.1371/journal.ppat.1005812 PMID: 27490492
19. O’Sullivan JN, Rea MC, O’Connor PM, Hill C, Ross RP. Human skin microbiota is a rich source of bacte-
riocin-producing staphylococci that kill human pathogens. FEMS Microbiology Ecology. 2019; 95(2).
https://doi.org/10.1093/femsec/fiy241 PMID: 30590567
20. Cogen AL, Nizet V, Gallo RL. Skin microbiota: a source of disease or defence? British Journal of Derma-
tology. 2008; 158(3):442–55. https://doi.org/10.1111/j.1365-2133.2008.08437.x PMID: 18275522
21. Kumar R, Jangir PK, Das J, Taneja B, Sharma R. Genome Analysis of Staphylococcus capitis TE8
Reveals Repertoire of Antimicrobial Peptides and Adaptation Strategies for Growth on Human Skin.
Scientific reports. 2017; 7(1):10447–. https://doi.org/10.1038/s41598-017-11020-7 PMID: 28874737
22. Christensen GJ, Bruggemann H. Bacterial skin commensals and their role as host guardians. Benef
Microbes. 2014; 5(2):201–15. Epub 2013/12/11. https://doi.org/10.3920/BM2012.0062 PMID:
24322878.

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 14 / 17

 Bacteriocin production by Staphylococcus capitis

23. Nakatsuji T, Chen TH, Narala S, Chun KA, Two AM, Yun T, et al. Antimicrobials from human skin com-
mensal bacteria protect against Staphylococcus aureus and are deficient in atopic dermatitis. Science
translational medicine. 2017; 9(378):eaah4680–eaah. https://doi.org/10.1126/scitranslmed.aah4680
PMID: 28228596
24. Götz F, Perconti S, Popella P, Werner R, Schlag M. Epidermin and gallidermin: Staphylococcal lantibio-
tics. International Journal of Medical Microbiology. 2014; 304(1):63–71. https://doi.org/10.1016/j.ijmm.
2013.08.012 PMID: 24119540
25. Field D, Gaudin N, Lyons F, O’Connor PM, Cotter PD, Hill C, et al. A Bioengineered Nisin Derivative to
Control Biofilms of Staphylococcus pseudintermedius. PLOS ONE. 2015; 10(3):e0119684–e. https://
doi.org/10.1371/journal.pone.0119684 PMID: 25789988
26. Cotter PD, Draper LA, Lawton EM, McAuliffe O, Hill C, Ross RP. Overproduction of wild-type and bioen-
gineered derivatives of the lantibiotic lacticin 3147. Applied and environmental microbiology. 2006; 72
(6):4492–6. https://doi.org/10.1128/AEM.02543-05 PMID: 16751576
27. Field D, Begley M, O’Connor PM, Daly KM, Hugenholtz F, Cotter PD, et al. Bioengineered Nisin A Deriv-
atives with Enhanced Activity against Both Gram Positive and Gram Negative Pathogens. PLoS ONE.
2012; 7(10):e46884–e. https://doi.org/10.1371/journal.pone.0046884 PMID: 23056510
28. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: A New Genome
Assembly Algorithm and Its Applications to Single-Cell Sequencing. Journal of Computational Biology.
2012; 19(5):455–77. https://doi.org/10.1089/cmb.2012.0021 PMID: 22506599
29. Hyatt D, Chen G-L, LoCascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recogni-
tion and translation initiation site identification. BMC Bioinformatics. 2010; 11(1):119–. https://doi.org/
10.1186/1471-2105-11-119 PMID: 20211023
30. Finn RD, Coggill P, Eberhardt RY, Eddy SR, Mistry J, Mitchell AL, et al. The Pfam protein families data-
base: towards a more sustainable future. Nucleic Acids Research. 2016; 44(D1):D279–D85. https://doi.
org/10.1093/nar/gkv1344 PMID: 26673716
31. Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein model-
ing, prediction and analysis. Nature Protocols. 2015; 10(6):845–58. https://doi.org/10.1038/nprot.2015.
053 PMID: 25950237
32. Cheung GY, Joo HS, Chatterjee SS, Otto M. Phenol-soluble modulins—critical determinants of staphy-
lococcal virulence. FEMS microbiology reviews. 2014; 38(4):698–719. Epub 2014/01/01. https://doi.
org/10.1111/1574-6976.12057 PMID: 24372362; PubMed Central PMCID: PMC4072763.
33. Netz DJA, Pohl R, Beck-Sickinger AG, Selmer T, Pierik AJ, Bastos MdCdF, et al. Biochemical Charac-
terisation and Genetic Analysis of Aureocin A53, a New, Atypical Bacteriocin from Staphylococcus
aureus. Journal of Molecular Biology. 2002; 319(3):745–56. https://doi.org/10.1016/S0022-2836(02)
00368-6 PMID: 12054867
34. Zheng J, Gänzle MG, Lin XB, Ruan L, Sun M. Diversity and dynamics of bacteriocins from human
microbiome. Environmental Microbiology. 2015; 17(6):2133–43. https://doi.org/10.1111/1462-2920.
12662 PMID: 25346017
35. Carson DA, Barkema HW, Naushad S, De Buck J. Bacteriocins of non-aureus staphylococci isolated
from bovine milk. Applied and environmental microbiology. 2017:AEM.01015-17. https://doi.org/10.
1128/AEM.01015-17 PMID: 28667105
36. Sugai M, Fujiwara T, Akiyama T, Ohara M, Komatsuzawa H, Inoue S, et al. Purification and molecular
characterization of glycylglycine endopeptidase produced by Staphylococcus capitis EPK1. Journal of
bacteriology. 1997; 179(4):1193–202. https://doi.org/10.1128/jb.179.4.1193-1202.1997 PMID:
9023202
37. Miljkovic M, Uzelac G, Mirkovic N, Devescovi G, Diep DB, Venturi V, et al. LsbB Bacteriocin Interacts
with the Third Transmembrane Domain of the YvjB Receptor. Applied and Environmental Microbiology.
2016; 82(17):5364–74. https://doi.org/10.1128/AEM.01293-16 PMID: 27342562
38. Kojic M, Strahinic I, Fira D, Jovcic B, Topisirovic L. Plasmid content and bacteriocin production by five
strains of Lactococcus lactis isolated from semi-hard homemade cheese. Can J Microbiol. 2006; 52
(11):1110–20. Epub 2007/01/12. https://doi.org/10.1139/w06-072 PMID: 17215903.
39. Ceotto H, Brede D, Salehian Z, Nascimento Jdos S, Fagundes PC, Nes IF, et al. Aureocins 4185, bacte-
riocins produced by Staphylococcus aureus 4185: potential application in food preservation. Foodborne
pathogens and disease. 2010; 7(10):1255–62. Epub 2010/07/14. https://doi.org/10.1089/fpd.2010.
0578 PMID: 20618078.
40. Iwatani S, Zendo T, Yoneyama F, Nakayama J, Sonomoto K. Characterization and Structure Analysis
of a Novel Bacteriocin, Lacticin Z, Produced by Lactococcus lactis QU 14. Bioscience, Biotechnology,
and Biochemistry. 2007; 71(8):1984–92. https://doi.org/10.1271/bbb.70169 PMID: 17690480
41. Fujita K, Ichimasa S, Zendo T, Koga S, Yoneyama F, Nakayama J, et al. Structural analysis and charac-
terization of lacticin Q, a novel bacteriocin belonging to a new family of unmodified bacteriocins of gram-

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 15 / 17

 Bacteriocin production by Staphylococcus capitis

positive bacteria. Applied and environmental microbiology. 2007; 73(9):2871–7. https://doi.org/10.

1128/AEM.02286-06 PMID: 17351096
42. Sandiford S, Upton M. Identification, characterization, and recombinant expression of epidermicin NI01,
a novel unmodified bacteriocin produced by Staphylococcus epidermidis that displays potent activity
against Staphylococci. Antimicrobial agents and chemotherapy. 2012; 56(3):1539–47. https://doi.org/
10.1128/AAC.05397-11 PMID: 22155816
43. Perez RH, Zendo T, Sonomoto K. Circular and Leaderless Bacteriocins: Biosynthesis, Mode of Action,
Applications, and Prospects. Frontiers in microbiology. 2018; 9:2085–. https://doi.org/10.3389/fmicb.
2018.02085 PMID: 30233551
44. Iwatani S, Yoneyama F, Miyashita S, Zendo T, Nakayama J, Sonomoto K. Identification of the genes
involved in the secretion and self-immunity of lacticin Q, an unmodified leaderless bacteriocin from Lac-
tococcus lactis QU 5. Microbiology. 2012; 158(Pt_12):2927–35. https://doi.org/10.1099/mic.0.062943-0
PMID: 23103973
45. Iwatani S, Horikiri Y, Zendo T, Nakayama J, Sonomoto K. Bifunctional Gene Cluster lnqBCDEF
Mediates Bacteriocin Production and Immunity with Differential Genetic Requirements. Applied and
Environmental Microbiology. 2013; 79(7):2446–9. https://doi.org/10.1128/AEM.03783-12 PMID:
23335763
46. Nascimento JdS Coelho MLV, Ceotto H, Potter A, Fleming LR, Salehian Z, et al. Genes Involved in
Immunity to and Secretion of Aureocin A53, an Atypical Class II Bacteriocin Produced by Staphylococ-
cus aureus A53. Journal of Bacteriology. 2012; 194(4):875–83. https://doi.org/10.1128/JB.06203-11
PMID: 22155775
47. Netz DJA, Sahl H-G, Marcolino R, dos Santos Nascimento Jn, de Oliveira SS, Soares MB, et al. Molec-
ular characterisation of aureocin A70, a multi-peptide bacteriocin isolated from Staphylococcus aureus
11 Edited by M. Yaniv. Journal of Molecular Biology. 2001; 311(5):939–49. https://doi.org/10.1006/jmbi.
2001.4885 PMID: 11531330
48. Coelho MLV, Coutinho BG, Cabral da Silva Santos O, Nes IF, Bastos MdCdF. Immunity to the Staphy-
lococcus aureus leaderless four-peptide bacteriocin aureocin A70 is conferred by AurI, an integral
membrane protein. Research in Microbiology. 2014; 165(1):50–9. https://doi.org/10.1016/j.resmic.
2013.11.001 PMID: 24239961
49. Acedo JZ, van Belkum MJ, Lohans CT, Towle KM, Miskolzie M, Vederas JC. Nuclear Magnetic Reso-
nance Solution Structures of Lacticin Q and Aureocin A53 Reveal a Structural Motif Conserved among
Leaderless Bacteriocins with Broad-Spectrum Activity. Biochemistry. 2016; 55(4):733–42. Epub 2016/
01/16. https://doi.org/10.1021/acs.biochem.5b01306 PMID: 26771761.
50. Wladyka B, Wielebska K, Wloka M, Bochenska O, Dubin G, Dubin A, et al. Isolation, biochemical char-
acterization, and cloning of a bacteriocin from the poultry-associated Staphylococcus aureus strain CH-
91. Applied Microbiology and Biotechnology. 2013; 97(16):7229–39. https://doi.org/10.1007/s00253-
012-4578-y PMID: 23196985
51. Towle KM, Vederas JC. Structural features of many circular and leaderless bacteriocins are similar to
those in saposins and saposin-like peptides. MedChemComm. 2017; 8:276–85. https://doi.org/10.
1039/c6md00607h PMID: 30108744.
52. Yoneyama F, Imura Y, Ohno K, Zendo T, Nakayama J, Matsuzaki K, et al. Peptide-lipid huge toroidal
pore, a new antimicrobial mechanism mediated by a lactococcal bacteriocin, lacticin Q. Antimicrobial
agents and chemotherapy. 2009; 53:3211–7. https://doi.org/10.1128/AAC.00209-09 PMID: 19470516
53. Li M, Yoneyama F, Toshimitsu N, Zendo T, Nakayama J, Sonomoto K. Lethal hydroxyl radical accumu-
lation by a lactococcal bacteriocin, lacticin Q. Antimicrobial agents and chemotherapy. 2013; 57
(8):3897–902. Epub 2013/06/05. https://doi.org/10.1128/AAC.00638-13 PMID: 23733459; PubMed
Central PMCID: PMC3719721.
54. Netz DJ, Bastos Mdo C, Sahl HG. Mode of action of the antimicrobial peptide aureocin A53 from Staph-
ylococcus aureus. Appl Environ Microbiol. 2002; 68(11):5274–80. Epub 2002/10/31. https://doi.org/10.
1128/AEM.68.11.5274-5280.2002 PMID: 12406714; PubMed Central PMCID: PMC129900.
55. Ovchinnikov KV, Chi H, Mehmeti I, Holo H, Nes IF, Diep DB. Novel group of leaderless multipeptide
bacteriocins from Gram-positive bacteria. Applied and Environmental Microbiology. 2016; 82:5216–24.
https://doi.org/10.1128/AEM.01094-16 PMID: 27316965
56. Iwatani S, Horikiri Y, Zendo T, Nakayama J, Sonomoto K. Bifunctional Gene Cluster lnqBCDEF Medi-
ates Bacteriocin Production and Immunity with Differential Genetic Requirements. Applied and Environ-
mental Microbiology. 2013; 79:2446–9. https://doi.org/10.1128/AEM.03783-12 PMID: 23335763.
57. Gibreel TM, Upton M. Synthetic epidermicin NI01 can protect Galleria mellonella larvae from infection
with Staphylococcus aureus. Journal of Antimicrobial Chemotherapy. 2013; 68:2269–73. https://doi.
org/10.1093/jac/dkt195 PMID: 23711896

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 16 / 17

 Bacteriocin production by Staphylococcus capitis

58. Halliwell S, Warn P, Sattar A, Derrick JP, Upton M. A single dose of epidermicin NI01 is sufficient to
eradicate MRSA from the nares of cotton rats. The Journal of antimicrobial chemotherapy. 2017;
72:778–81. https://doi.org/10.1093/jac/dkw457 PMID: 27999015.

PLOS ONE | https://doi.org/10.1371/journal.pone.0223541 October 16, 2019 17 / 17