antimicrobial peptides as therapeutic...

Guest Editors: Mathew Upton, Paul Cotter, and John Tagg

International Journal of Microbiology

Antimicrobial Peptides as Therapeutic Agents

International Journal of Microbiology


Guest Editors: Mathew Upton, Paul Cotter, and John Tagg

Copyright © 2012 Hindawi Publishing Corporation. All rights reserved.

This is a special issue published in “International Journal of Microbiology.” All articles are open access articles distributed under theCreative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.

Editorial Board

Vasco Azevedo, BrazilArvind A. Bhagwat, USADulal Borthakur, USATodd R. Callaway, USAMichael L. Chikindas, USAP. Patrick Cleary, USALuca Simone Cocolin, ItalyPeter Coloe, AustraliaGiuseppe Comi, ItalyMichael A. Cotta, USADaniele Daffonchio, ItalyEduardo Dei-Cas, FranceJoseph Falkinham, USAPaula J. Fedorka-Cray, USAArsenio M. Fialho, PortugalMarco Gobbetti, Italy

Robert P. Gunsalus, USAAkira Hiraishi, JapanPo-Ren Hsueh, TaiwanJ. Hugenholtz, The NetherlandsBarbara H. Iglewski, USAVijay K. Juneja, USAThomas L. Kieft, USASandra Macfarlane, UKMichael McClelland, USAMichael J. McInerney, USASusana Merino, SpainTimothy A. Mietzner, USAHugh W. Morgan, New ZealandSudarsan Mukhopadhyay, USAIngolf Figved Nes, NorwayJames D. Oliver, USA

Toni L. Poole, USACarla Pruzzo, ItalyRosana Puccia, BrazilR. M. Roop, USAKenneth S. Rosenthal, USAIsabel Sa-Correia, PortugalKaarina Sivonen, FinlandJ. Glenn Songer, USAA. J. Stams, The NetherlandsDavid C. Straus, USAJohn Tagg, New ZealandMichael M. Tunney, NorwayJ. Wiegel, USAYanjin Zhang, USAM. H. Zwietering, The Netherlands

Contents

Antimicrobial Peptides as Therapeutic Agents, Mathew Upton, Paul Cotter, and John TaggVolume 2012, Article ID 326503, 2 pages

Salivaricin G32, a Homolog of the Prototype Streptococcus pyogenes Nisin-Like Lantibiotic SA-FF22,Produced by the Commensal Species Streptococcus salivarius, Philip A. Wescombe, Kristin H. Dyet,Karen P. Dierksen, Daniel A. Power, Ralph W. Jack, Jeremy P. Burton, Megan A. Inglis, Anna L. Wescombe,and John R. TaggVolume 2012, Article ID 738503, 10 pages

Coleopteran Antimicrobial Peptides: Prospects for Clinical Applications, Monde Ntwasa, Akira Goto,and Shoichiro KurataVolume 2012, Article ID 101989, 8 pages

The Design and Construction of K11: A Novel α-Helical Antimicrobial Peptide, Huang Jin-Jiang,Lu Jin-Chun, Lu Min, Huang Qing-Shan, and Li Guo-DongVolume 2012, Article ID 764834, 6 pages

The Lantibiotic Lacticin 3147 Prevents Systemic Spread of Staphylococcus aureus in a Murine InfectionModel, Clare Piper, Pat G. Casey, Colin Hill, Paul D. Cotter, and R. Paul RossVolume 2012, Article ID 806230, 6 pages

Development of Class IIa Bacteriocins as Therapeutic Agents, Christopher T. Lohans and John C. VederasVolume 2012, Article ID 386410, 13 pages

Hindawi Publishing CorporationInternational Journal of MicrobiologyVolume 2012, Article ID 326503, 2 pagesdoi:10.1155/2012/326503

Editorial


Mathew Upton,1 Paul Cotter,2 and John Tagg3

1 Microbiology and Virology Unit, School of Medicine, University of Manchester, Oxford Road, Manchester M13 9PL, UK2 Teagasc Food Research Centre, Alimentary Pharmabiotic Centre, Moorepark, Fermoy, Cork, Ireland3 Department of Microbiology and Immunology, University of Otago, P.O. Box 56, Dunedin, Otago, New Zealand

Correspondence should be addressed to Mathew Upton, [email protected]

Received 5 April 2012; Accepted 5 April 2012

Copyright © 2012 Mathew Upton et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

In developing this special issue, we sought to raise awarenessof the potential of antimicrobial peptides (AMPs) for clinicaluse and to highlight recent developments in the field. Theissue contains research articles covering the full spectrumof contemporary work, ranging from the rational design ofsynthetic peptides and discovery of bacteriocins, to testing ofthe efficacy of AMPs in animal models. We have included twoexpert reviews reflecting on the potential of both eukaryoticand prokaryotic peptide antibiotics for clinical use.

The antibiotic development pipeline is devoid of immi-nent solutions to the problems caused by resistant gram-negative bacteria and only a small number of new agentsexist for therapy of infections caused by gram-positives. Thepathway for development of new antibiotics is both long andextremely expensive and, therefore, “big-pharma” has notinvested significantly in this area for decades.

The scarcity of antibiotics to combat infectious diseaseis commonly acknowledged and the specter of untreat-able infections has been widely publicized. However, someadditional potential impacts of the absence of efficaciousantibiotics are not as widely discussed. The inability to reli-ably treat infections could potentially mean that all electivesurgery ceases, and, moreover, there could be significantrestrictions placed on the use of cancer chemotherapy. Thus,in order to maintain anything like the level of medical carethat many of us now enjoy, we must devote considerableeffort to find solutions to this steadily-evolving clinicaldilemma. AMPs appear to merit a prominent position inthe catalogue of pharmaceuticals that should be more ex-haustively investigated in our search for new antimicrobialagents.

Many AMPs have potent activity against bacteria, includ-ing those that are resistant to conventional antibiotics. Their

activity is often relatively-specifically directed against certaingenera or groups of bacteria, which could limit damagingeffects on a patient’s commensal flora. AMPs generallyexhibit high stability to wide ranges in pH and temperature,characteristics that may be beneficial for their scaled-up pro-duction and formulation into deliverable products. Also, dueto their rather specific modes of action, many AMPs exhibitlow toxicity for eukaryotic cells, providing an opportunity fora wide therapeutic window. Their typical mode of action isalso linked to a low propensity for resistance developmentin target bacteria. These unique features differentiate AMPsfrom many conventional antibiotics.

Outside of the potential use of AMPs as alternativesto antibiotics with respect to the treatment/prevention ofdisease, numerous recent reports have also described theactivity of AMPs against bacteria growing in biofilms, that is,communities encased in structured matrices of extracellularmacromolecules, on medical devices. Medical devices canbecome colonized with bacteria growing in biofilms, whichare resistant to elements of the host immune response andto extremely high concentrations of conventional antibiotics.Consequently, resolution of medical device-related infec-tions usually requires the removal of the device. Notably,low concentrations of some AMPs have been shown to haveinhibitory and disruptive properties that can eliminate evenwell-established biofilms.

It is also important to consider the fact that some AMPshave been shown to exhibit immunomodulatory properties;that is, they can reduce the tissue-damaging inflamma-tory response to infection, whilst stimulating antimicrobialimmune activities. Peptides exhibiting modulatory features,as well as overt antimicrobial activity, are now being selectedas favoured candidates for further evaluation in clinical trials.

2 International Journal of Microbiology

Recently, the search for AMPs has benefited from anumber of advances in informatics and peptide engineer-ing strategies. This, coupled with renewed interest frombig-pharma, manifested by some recent acquisitions ofcompanies using natural product screening approaches, isincreasing the prominence of AMPs. This current specialissue includes research articles and reviews that togetherdescribe the process of discovery of AMPs, their testing inin vivo models and outlines steps taken in the developmentof clinically-relevant agents.

The paper by H. Jin-Jiang and colleagues describes thedesign and synthesis of a range of AMPs and reports onthe activity of K11, a 20 mer highly cationic helical pep-tide with activity against gram-positive and gram-negativebacteria, including drug resistant clinical isolates. K11 has ahigh therapeutic index/window and clearly warrants furtherinvestigation. Thus, the paper demonstrates the utility ofsynthetic approaches to the development of novel chimericAMPs.

Also included in the special issue is a report by P. A.Wescombe and colleagues, which focuses on the character-ization of a bacteriocin from the SA-FF22 family producedby members of the Streptococcaceae. The plasmid-encodedpeptide G32 was purified from a culture of Streptococcussalivarius, a species known to include numerous bacteriocinproducers, and is active against the pathogen Streptococcuspyogenes. An interesting feature of this lantibiotic peptideis the apparent absence of an auto-induction capability.The work supports previous suggestions that Streptococcussalivarius mega-plasmids are repositories for lantibiotic lociand that bacteria producing these lantibiotics can be effectiveprobiotics.

C. Piper and colleagues demonstrate the utility of wholeanimal bioimaging approaches to demonstrate that thelactococcal lantibiotic lacticin 3147 could prevent systemicspread of Staphylococcus aureus following subcutaneousdelivery of the peptide. This is important work as it addsto the growing evidence base that suggests that lantibioticsand other AMPs could potentially be used to control systemicinfections, most previous approaches having been restrictedto topical applications.

The two reviews included in this issue elegantly precisrecent advances that have been made with AMPs derivedfrom both bacteria and insects. M. Ntwasa and colleaguesdescribe the medical potential of immune system peptides ofthe coleopteran insects. The authors report that these ancienteukaryotes have a highly developed and robust immunesystem that has contributed to their survival. Exploitationof the immune effectors from these organisms holds greatpotential for use in clinical medicine. The review alsodescribes the utility of databases and full genome sequencedetermination in AMP discovery.

The paper by C. T. Lohans and J. C. Vederas is an exten-sive review of the potential utility of the unmodified, or classIIa, bacteriocins. The paper describes peptide engineeringapproaches for increasing the value of bacteriocins, methodsfor improved in vitro production and efficient purification,various in vivo delivery options and, finally, gives some atten-tion to the important issue of the development of resistance

to bacteriocins. The paper concludes with a positive outlookfor the use of bacteriocins in medical therapy.

We agree that there is great potential for the use of AMPsin clinical therapy for infectious disease, either throughdirect antimicrobial activity or modulation of the immuneresponse to infection. Contributors to this special issue alsodiscuss the possibility for use of some AMPs in antitumortherapy. AMPs offer some significant advantages over con-ventional antibiotics and we are now making importantadvances in the discovery, optimization, and activity testingof these agents. In parallel, publications in this field areattracting the attention of increasing numbers of academicand industrial researchers. It is genuinely an exciting time tobe involved in the development of AMPs and we trust thatthis special issue leaves readers similarly enthused about thesubject.

Mathew UptonPaul Cotter

John Tagg


Research Article

Salivaricin G32, a Homolog of the Prototype Streptococcuspyogenes Nisin-Like Lantibiotic SA-FF22, Produced by theCommensal Species Streptococcus salivarius

Philip A. Wescombe,1 Kristin H. Dyet,2 Karen P. Dierksen,2 Daniel A. Power,2 Ralph W. Jack,2

Jeremy P. Burton,1 Megan A. Inglis,2 Anna L. Wescombe,2 and John R. Tagg1, 2

1 BLIS Technologies Ltd., Centre for Innovation, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand2 Department of Microbiology and Immunology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand

Correspondence should be addressed to Philip A. Wescombe, [email protected]

Received 14 October 2011; Accepted 5 December 2011

Academic Editor: Paul Cotter

Copyright © 2012 Philip A. Wescombe et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Salivaricin G32, a 2667 Da novel member of the SA-FF22 cluster of lantibiotics, has been purified and characterized fromStreptococcus salivarius strain G32. The inhibitory peptide differs from the Streptococcus pyogenes—produced SA-FF22 in theabsence of lysine in position 2. The salivaricin G32 locus was widely distributed in BLIS-producing S. salivarius, with 6 (23%)of 26 strains PCR-positive for the structural gene, slnA. As for most other lantibiotics produced by S. salivarius, the salivaricinG32 locus can be megaplasmid encoded. Another member of the SA-FF22 family was detected in two Streptococcus dysgalactiaeof bovine origin, an observation supportive of widespread distribution of this lantibiotic within the genus Streptococcus. Since theinhibitory spectrum of salivaricin G32 includes Streptococcus pyogenes, its production by S. salivarius, either as a member of thenormal oral microflora or as a commercial probiotic, could serve to enhance protection of the human host against S. pyogenesinfection.

1. Introduction

The Streptococcus pyogenes (Lancefield group A strepto-coccus)-derived streptococcin A-FF22 (SA-FF22) was thefirst of the streptococcal bacteriocins to be isolated [1] andthen characterized both chemically [2, 3] and genetically[4, 5]. Prior to its complete characterization, it was alreadyapparent that SA-FF22 was closely similar to the well-knownLactococcus lactis bacteriocin, nisin [1]. Nisin, a widely-used biopreservative agent, is regarded as the prototype ofthe lantibiotics, a heterogeneous group of lanthionine-containing bacteriocins produced by a wide variety of Gram-positive bacteria. Strain FF22, the producer of the 26-aminoacid 2791 Da SA-FF22, gives a characteristic bacteriocinproduction ([P]-type) inhibitory pattern referred to as 436when tested against a set of nine standard indicator bacteria

in a standardized deferred antagonism bacteriocin-typingprotocol [6].

The SA-FF22 genetic locus comprises nine open readingframes arranged in three operons responsible, respectively,for SA-FF22 regulation, biosynthesis, and immunity [7].Interestingly, immediately downstream of the SA-FF22 struc-tural gene (scnA), there is another open reading frame-designated scnA′ which has close similarity to scnA. Althoughthis reading frame has been shown to be transcribed [7], itsfunction has not been established. The inducing factorfor upregulation of SA-FF22 production has also not beenidentified. Many lantibiotics, such as nisin [8] and salivaricinA [9], have been shown to function as the signal peptidesfor their own upregulation. In the present study we haveattempted to identify the signal for the regulation of SA-FF22transcription and show that this is not the SA-FF22 peptide.


Relatively recently a lantibiotic peptide identical to SA-FF22 was shown to be produced by the food-grade speciesStreptococcus macedonicus and named macedocin [10]. Inscreening tests of streptococcal strains for their productionof bacteriocin-like inhibitory substances (BLISs), we havedetected a number of P-type 436 strains in the species Strep-tococcus salivarius, Streptococcus mutans, and Streptococcusdysgalactiae [11]. In the present study we show that the agentresponsible for the P-type 436 BLIS activity of S. salivariusG32 is a SA-FF22-like peptide named salivaricin G32.

2. Materials and Methods

2.1. Bacterial Strains and Culture Conditions. All strains weremaintained on Columbia blood agar base (Difco, Sparks,MD) supplemented with 5% human blood and 0.1% wt/vol.CaCO3 (BACa) with incubation in 5% CO2 in air at 37◦C.Liquid culture media were Todd Hewitt broth (THB) (Difco)and THB supplemented with CaCO3 (4 μmol·L−1) and glu-cose (0.1% wt/vol.) (THBCaGlu). S. pyogenes strain FF22 [4]and S. salivarius strain G32 were originally sourced fromhuman throat and human saliva, respectively. S. dysgalactiaestrains 61 and 67 were from bovine mastitis (courtesy ofDr. B. Jayarao, University of Tennessee). The details of otherstrains are listed in the tables in which they appear.

2.2. Screening for Antimicrobial Activity. BLIS activity wasdetected using the deferred antagonism method essentially asoriginally described by Tagg and Bannister [6]. Briefly, aBACa plate was seeded from one side to the other with a 1 cmwide inoculum of the test strain from an 18 hr BACa cultureusing a cotton swab. Following incubation for 18 hr at 37◦Cin 5% CO2 in air, the 1 cm wide culture growth was removedusing a glass slide and the agar surface sterilised by exposureto chloroform vapours for 30 min, followed by airing for afurther 30 min. Indicator bacteria (from 18 hr THB cultures)were then inoculated with a swab at right angles across theline of the original diametric streak culture of the test strainand the plate re-incubated for 18 hr at 37◦C in 5% CO2 inair. Zones of inhibition were scored as − for no inhibitionor + if definite interference with the growth of the indicatorwas evident. Variations to the deferred antagonism methodincluded supplementation of the agar with either 2 mg·mL−1

Trypan Blue or 1% MgCl2, to test for interference of theseagents with BLIS activity.

2.3. Purification of Salivaricin G32. Crude preparations ofsalivaricin G32 were obtained by freeze-thaw extraction of80 lawn cultures of S. salivarius G32 that had been grown for18 hr at 37◦C in 5% CO2 in air on tryptic soy broth (BBL)supplemented with 2% yeast extract (Difco), 1% CaCO3,and 0.7% Bacto agar adjusted to pH 6.5 before autoclaving(TsYECa). The freeze thaw extraction process consisted ofthe agar cultures (entire agar plate including bacterial lawn)being frozen at −70◦C and then subsequently thawed.The exudate on thawing was collected and clarified bycentrifugation (15300× g for 25 min). Salivaricin G32 purifi-cation from this preparation (volume 2 L) was essentially as

described previously for SA-FF22 [3]. The supernatant waspassed through an XAD-2 column (bed volume 330 mLServa, Heidelburg) followed by washing with MQ water,then one bed volume of each of 50% and 70% methanol.The active peptide was eluted in one bed volume of 95%methanol (pH 2). Rotary evaporation removed the methanoland the aqueous phase was clarified by centrifugation at5000× g for 15 min prior to loading 500 μL aliquots ontoa Brownlee C8 reversed phase (RP-300, Aquapore Octyl,300 A, 7 U) column. Elution of the inhibitory peptide wasachieved in a gradient of 18–30% acetonitrile over 60 minat 1 mL/min pump speed using a Pharmacia FPLC system.The active fractions (detected by spot test assay on alawn of Micrococcus luteus indicator I1) were pooled andfurther fractionated by HPLC using a reversed phase C18

column (Phenomenex Jupiter C18, 5 μm, 300 A, 250 ×4.6 mm) equilibrated in 0.1% trifluoroacetic acid. Elutionwas in isocratic 45% acetonitrile over 60 min. The activefractions were pooled and analysed by Matrix Assisted LaserDesorption Time of Flight Mass Spectrometry (MALDI-TOFMS). A sample of purified salivaricin G32 was subjected toautomated sequential Edman degradation on a gas-phaseprotein sequencer equipped with an on-line microbore PTH-amino acid analyser (Applied Microsystems).

2.4. Cloning and Sequencing of svnA. To take advantage of theapparent close similarity of the salivaricin G32 peptidesequence to that of the previously-identified SA-FF22, theSA-FF22 primer pair scnF (5′-GCACCTATCCTTCTGAAG-AAAG) and scnR (5′-GCACCTAGGCACATTTTTTCT-TCC) was used to amplify the SA-FF22 structural gene,and this was used to probe a series of chromosomal digestsof strain G32 (results not shown). A 1.9-Kb HindIII-EcoRIderived restriction fragment containing the salivaricin G32structural gene (slnA) was subsequently cloned into pUC19using standard techniques [12] and sequenced. The LanMuniversal primers [13] were used to sequence part of theslnM gene in strain G32, and then PCR was used to close thegap between slnA and slnM in order to obtain the completesequence of slnA1 (Figure 3).

2.5. Distribution of scnA and slnA in S. pyogenes and S.salivarius. The species distribution of the SA-FF22 and sali-varicin G32 structural genes was determined using PCR. Allamplifications were carried out using Hotmaster taq poly-merase (Eppendorf, Hamburg, Germany). Typical PCR re-actions consisted of 40.5 μL PCR grade water (Eppendorf),5 μL of 10 × Buffer (Hotmaster, Eppendorf), 1 μL nucleotidemix (Roche), 1 μL each of the appropriate forward andreverse primers (primer stocks at 0.01 ng·μL−1), 0.5 μL Taq(Hotmaster 5 U·μL−1), and 1 μL of template DNA. Initialdenaturation at 95◦C for 2 min was followed by 30 cycles of95◦C for 30 sec, annealing at 60◦C for 30 sec and elongationfor 30 sec at 65◦C. Two percent agarose gels were used toanalyse the PCR products.

2.6. Megaplasmid DNA Detection Using Pulsed Field GelElectrophoresis. Using previously described methods [14],

International Journal of Microbiology 3

pulsed field gel electrophoresis (PFGE) was used to detect themegaplasmid DNA content of S. salivarius G32, S. salivarius20P3, S. dysgalactiae 61, S. dysgalactiae 67, and S. pyogenesFF22. One mL of 18 hr THB cultures of the test strains wasused to inoculate 20 mL THBs, which in turn were incubatedat 37◦C in 5% CO2 in air to an optical density (650 nm) of0.5. The cells were then embedded in low-melting-pointagarose (SeaPlaque, FMC BioProducts, Rockland, ME, USA)and lysed using 40 mg lysozyme mL−1 (Roche), followed bytreatment with proteinase K (1 mg·mL−1) (Sigma). The pro-teinase K was removed by six 1 hr TE (10 mM Tris, 1 mMEDTA pH 7.5) washes with shaking (120 rpm) at roomtemperature. The DNA was separated using a CHEF-DR IIIPulsed Field Electrophoresis System (Bio-Rad) over 20 hrin a 1% agarose gel (Pulsed Field Certified Agarose, Bio-Rad Laboratories). An initial pulse time of 6 sec and final of18 sec was used. The included angle (change in orientationof the field) was 120◦, and the gel was run at 4.5 V·cm−1

with the buffer maintained at 14◦C. Gels were stainedwith ethidium bromide (5 mg·mL−1) and examined by UVtransillumination.

2.7. Induction of SA-FF22 and Salivaricin G32 Production.The lantibiotic-producer strain (either S. pyogenes FF22 or S.salivarius G32) was grown for 18 hr in THB. Putative inducerpreparations (freeze thaw extracts of lawn cultures of strainsFF22, G32, EB1 (SA-FF22-negative derivative of strain FF22)or strains of various other P-type designations) were addedas 20 μL spots on the “test” side of a BACa plate and allowedto dry into the agar. The lantibiotic producer strain wasthen inoculated (by swabbing from the 18 hr THB culture)over the whole plate and incubated for a specified time.Approximately 1-2 hr before the bacteriocin produced by theuninduced bacteria in the lawn culture was expected to attaininhibitory levels; the bacterial growth was scraped from theagar surface. On the “control” side of the plate, 20 μL dropsof inducer extracts equivalent to those previously depositedon the “test side” were added and allowed to dry. The agarsurface was then sterilized with chloroform vapour. Afterairing, an indicator strain culture was inoculated over theentire agar surface and incubated. Zone sizes (diameter inmm) were compared between the test and control halvesof the plate and induction was considered to have occurredwhen, at a particular time, the test zone was at least twicethe width of the corresponding control zone. Differenttimes for exposure to the putative inducer preparations wereevaluated in each experiment, depending on the anticipatedbacteriocin production characteristics of the strain underinvestigation.

2.8. Genbank Accession Number. The GenBank accession nu-mber for the partial salivaricin G32 (sln) locus from S. sali-varius strain G32 is JN831266.

3. Results/Discussion

3.1. Inhibitory Activity of S. salivarius G32. Application of theP-typing scheme allows for a preliminary categorization of

bacteriocin-producing strains on the basis of their inhibitoryprofile against nine standard indicator bacteria [6]. In thistest, S. salivarius G32 and S. dysgalactiae 61 both gave P-type 436 inhibitory profiles identical to that of S. pyogenesFF22 (Table 1). This can be considered to be a good pre-liminary indication that all three strains produce closely-similar inhibitory agents. A further indication of inhibitorsimilarity was the apparent cross-immunity displayed wheneach of these producers was tested for sensitivity to thehomologous (same strain) and heterologous (other twostrains) inhibitory products (Table 1) and also the failureof all three strains to produce detectable inhibitory activitywhen the P-typing medium was supplemented with either2 mg·mL−1 Trypan blue or 1% MgCl2, two agents previouslyshown to specifically suppress SA-FF22 production by S.pyogenes FF22 [22].

3.2. Purification of Salivaricin G32. Salivaricin G32 waspurified to apparent homogeneity from TsYECa agar culturesusing XAD-2 and a combination of C8 and C18 reversedphase chromatography (Figure 1). MALDI-TOF MS anal-ysis of the purified fraction identified a mass of 2667 Da(Figure 1), and Edman N-terminal sequencing revealeda partial peptide sequence GNGVFKXIXHEXXLNXXAFL,where X corresponds to an unidentifiable residue or blankcycle. This amino acid sequence corresponds closely to thatof SA-FF22, the only difference being the absence of alysine residue at position two of the salivaricin G32 peptidecompared to the SA-FF22 peptide (Figure 2). On the basisof the apparently closely-similar inhibitory spectra of strainsG32 and FF22, it appears that this lysine residue difference,although changing the isoelectric point of the molecule, mayhave relatively minimal impact on the target range of thepeptide.

3.3. Identification of the Salivaricin G32 Structural Gene.Attempts to amplify the salivaricin G32 structural gene usingprimers designed to the SA-FF22 locus were unsuccessful andso a cloning approach was taken using the SA-FF22 structuralgene scnA as a probe for Southern blotting. Followingsequencing of the region surrounding slnA, the salivaricinG32 structural gene (slnA) was shown to be present as twoalmost identical copies in strain G32 (Figure 3). An align-ment of scnA, mcdA, and slnA and some other knownvariants is shown in Figure 2. The predicted propeptideresulting from translation of slnA differs from SA-FF22 inonly the absence of a Lys residue in position 2 of thesalivaricin G32 propeptide. However, there are also fivedifferent amino acids in the predicted leader sequenceencoded by slnA and six differences in slnA′ when comparedto the scnA-encoded leader region (Figure 2). The first 483 bpof slnR was sequenced and the predicted amino acid sequenceshowed 85% identity to the corresponding region encoded byscnR. Similarly, translation of the first 903 bp of slnM showed63% identity to that encoded by scnM. Based on thesecomparisons, it appears that there is significant heterogeneitywithin the processing genes for the SA-FF22-like lantibiotics


Ta

ble

1:B

acte

rioc

inP-

typi

ng

ofSA

-FF2

2-lik

est

rain

s.

Indi

cato

rst

rain

Des

crip

tion

ofst

rain

/tes

tIn

hib

itio

nof

indi

cato

rby

prod

uce

rst

rain∗

Indi

cato

rst

rain

sou

rce/

refe

ren

ceS.

saliv

ariu

sG

32S.

pyog

enes

FF22

∗S.

dysg

alac

tiae

61M

.lut

eus

T18

Indi

cato

r1

++

+[6

]S.

pyog

enes

FF22

Indi

cato

r2:

self

/cro

sste

st−

−−

[6]

S.co

nste

llatu

sT

29In

dica

tor

3−

−−

[6]

S.ub

eris

AT

CC

2795

8In

dica

tor

4−

−−

[6]

S.py

ogen

es71

-679

Indi

cato

r5

++

+[6

]L.

lact

isT

-21

Indi

cato

r6

++

+[6

]S.

pyog

enes

71-6

98In

dica

tor

7+

++

[6]

S.py

ogen

esW

-1In

dica

tor

8+

++

[6]

S.dy

sgal

acti

aesu

bsp.

equi

sim

ilis

T14

8In

dica

tor

9−

−−

[6]

S.py

ogen

esE

B1

Cu

red

deri

vati

veof

FF22

++

+J.

R.T

agg∗

S.dy

sgal

acti

ae61

Self

/cro

sste

st−

−−

B.J

ayar

aoS.

saliv

ariu

sG

32Se

lf/c

ross

test

−−

−J.

R.T

agg

∗La

bora

tory

colle

ctio

nof

J.R

.Tag

g.


Rel

ativ

e ab

sorb

ance

at

214

nm

10

00

0

1

90

18

30

80

Time (min)

Ace

ton

itri

le (

%)

(a)

1200

1600

1400

2000

1800

2400

2200

2800

2600

3200

3000

1200

1600

1400

2000

1800

2400

2200

2600

00

2

7

8

9

10

11

12

8065

Mass (m/z )

Rel

ativ

e ab

sorb

ance

at

214

nm

10

80

45

8

6

4

2

0

Ace

ton

itri

le (

%)

−2

Rel

ativ

e ab

un

dan

ce

Calibrant

2667 Da

Time (min)

(b)

Figure 1: Reversed phase fractionation of salivaricin G32 preparations. (a) Representative C8 reversed phase fractionation of concentratedXAD-2 fractions containing inhibitory activity against indicator organism Micrococcus luteus. Elution of the column was with a gradient of18–30% acetonitrile at a flow rate of 1 mL·min−1 with detection of absorbance at 214 nm. (b) C18 reversed phase fractionation of pooled C8

reversed phase fractionated samples having inhibitory activity as indicated in panel A by the solid bar labelled 1. Elution was with isocratic45% acetonitrile (containing 0.1% TFA) over 60 minutes at a flow rate of 0.7 mL·min−1. Active fractions (solid bar 2) from multiple runswere pooled then analysed by MALDI-TOF MS as indicated by the grey insert.

M K K D - V V I E S I K E V S L E E L D Q I I G AM K K D - A I I E S I K E V S L E E L D Q I I G AM E K N N E V I N S I Q E V S L E E L D Q I I G AM E K E T T

(a)

I I E S I Q E V S L E E L D Q I I G AM E K E T T I I E S I Q E V S L E E L D Q I I G AM E K N N E V I N S I Q E V S L E E L D Q I I G AM T K E H E I I N S I Q E V S L E E L D Q I I G A

24 Salivaricin G32 (slnA)

25 StreptococcinA-FF22 (scnA)25 Macedocin (mcdA)

24 Salivaricin G32 (sln ) A

25 Macedocin (mcd ) A25 Streptococcin A-M49 (scn ) A25 Streptococcin A-M49 (scn ) A

G - N G V F K T I S H E C H L N T W A F L A T C C SG - N G V F K T I S H E C H L N T W A F L A T C C SG K N G V F K T I S H E C H L N T W A F L A T C C SG K N G V F K T I S H E C H L N T W A F L A T C C SG K N G V F K T I S H E C H L N T W A F L A T C C SG K N G V F K T I S H E C H L N T W A F L A T C C SG K N G V F K T I S H E C H L N T W A F L A T C C S

G - N G V F K X I X H E X X L N X X A F L Salivaricin G32 (N-terminalpeptide sequence)

25 Salivaricin G32 (slnA)

26 Streptococcin A-FF22 (scnA)26 Macedocin (mcdA)

25 Salivaricin G32 (sln ) A

26 Macedocin (mcd ) A26 Streptococcin A-M49 (scn ) A26 Streptococcin A-M49 (scn ) A

(b)

(1)

M K N N - K I H L E A L E A L Q E L K M E E I D N L L G G AM E E - - K M C L G A L N A L Q E F Q I E E L D N L L G GM K N N N K I C Q D A L E S L Q E L K L E E V D E L L G G A

29 Salivaricin G32 (slnA1)27 Streptococcin A-FF22 (scnA1)30 Macedocin (mcdA1)

(a)

- G H G V N T I S A E C R W N S L Q A I F T C CR G H G V N T I S A E C R W N S L Q A I F T C C- G H G V N T I S A E C R W N S L Q A I F S C C

23 Salivaricin G32 (slnA1)24 Streptococcin A-FF22 (scnA1)23 Macedocin (mcdA1)

(b)

(2)

Figure 2: (1) Alignment of (a) the leader sequences and (b) the propeptide sequences of salivaricin G32, SA-FF22, and macedocin (- indicatesa spacer introduced to aid the alignment). The N-terminal sequence derived from the purified G32 preparation is shown below, where Xindicates a blank cycle in the Edman reaction. (2) Alignment of (a) the predicted translated leader sequences and (b) the predicted translatedpropeptide sequences for slnA1, scnA1, and mcdA1. The number of amino acids in each sequence is indicated in front of each sequence name.

indicating a considerable degree of compositional flexibilityin a locus that appears to have dispersed widely amongsta variety of streptococcal species.

3.4. Distribution of slnA and scnA in S. salivarius and S.pyogenes. PCR detected slnA in 7 (23%) of 27 BLIS-pro-

ducing S. salivarius (Table 2). In S. pyogenes, only thirteen(9%) of 144 tested strains of a wide variety of serotypes werePCR-positive for scnA (Table 3). By contrast, 125 of the 144S. pyogenes were determined to be positive for the salivaricinA structural gene (salA) (results not shown). Interestingly,scnA-like genes were detected in some S. pyogenes strains


Table 2: Distribution of S. salivarius lantibiotic structural genes in S. salivarius strains of different P-type designations.

S. salivarius strain P-typePresence of salivaricin gene

Strain source/referenceslnA salA sboB sivA srtA-like∗

193 777 − − − − − [14]

20P3 677 + + − − − [15]

220 634 − − − + − Laboratory collection of J. R.Tagg

36 777 − − − − − [16]

5 677 + + − + − [16]

6 777 − + + − − [16]

9 676 − + − + − [16]

DC135 B 677 + + − + +Laboratory collection of J. R.

Tagg

DC 156A 636 − − − + − Laboratory collection of J. R.Tagg

G32 436 + − − − +Laboratory collection of J. R.

Tagg

GR 677 − + − − − Laboratory collection of J. R.Tagg

H7f 677 + + − − +Laboratory collection of J. R.

Tagg

H21 777 − + + − − [17]

H25 777 − + + − − Laboratory collection of J. R.Tagg

JH 677 + + − − + [18]

JIM8777 226 − − − − − [19]

JIM8780 (CCHSS3) 636 − − − + − [20]

JO-1 777 + − − + − Laboratory collection of J. R.Tagg

K12 777 − + + − − [21]

K30 777 − + + − − [17]

K-8P 226 − − − + − Laboratory collection of J. R.Tagg

M18 677 − + − + − [13]

Min5 777 − + + + − [17]

MPS 636 − + − + +Laboratory collection of J. R.

Tagg

NR 777 − − + − − [17]

Pirie 777 − + − − +Laboratory collection of J. R.

Tagg

Strong SA 777 − + + − − [17]

Total positive/total tested 7/27 17/27 8/27 11/27 6/27∗

A lantibiotic structural gene having close homology with the streptin lantibiotic gene srtA (O. Hyink, unpublished 2009).

that did not appear to produce SA-FF22-like P-type activitysuch as the P-type 614 strain M57-71724, the P-type 324strains M58-78234, M58-71726, M77-79305 and emm109-99454, the P-type 577 strains emm83-60173 and emm113-99458, the P-type 400 strain ST 2037-99448 and the BLIS-negative (P-type 000) strains emm88-60183 and emm105-99449. This indicates that in many S. pyogenes the SA-FF22operon is incomplete, as has previously been shown to be thecase in S. pyogenes for the streptin and salivaricin A loci[23, 24]. Interestingly, a selection of the S. pyogenes positive

for scnA but apparently not producing SA-FF22 was found tobe resistant to SA-FF22 (strains M58-78234, M77-79305, ST2037-99448, emm105-99449, and emm109-99454), indicat-ing that the immunity component of the SA-FF22 locus wasfunctional for these strains and may provide an ecologicaladvantage in the presence of S. salivarius or S. pyogenesproducing SA-FF22-like inhibitors. In addition, it appearsthat the salivaricin G32 locus can be expressed as part of abattery of antimicrobial peptides in certain strains such asS. salivarius strain T32 [14]. Four S. pyogenes strains A1020,


Ta

ble

3:D

istr

ibu

tion

ofsc

nAin

S.py

ogen

esof

vari

ous

M-t

ype/

emm

-typ

ede

sign

atio

ns.

scnA

cate

gory

Tota

lM

-or

emm

-typ

ean

dst

rain

nam

e(P

-typ

e)

Posi

tive

for

scnA

13/1

44M

57-7

1724

(614

),M

57-8

6152

,M58

-782

34(3

24),

M58

-717

26,M

77-7

9305

(324

),em

m83

-601

73(5

77),

emm

88-6

0183

(000

),em

m97

-994

40(7

77),

emm

105-

9944

9(0

00),

emm

109-

9945

4(3

24),

emm

113-

9945

8(5

77),

Trin

idad

(777

),ST

2037

-994

48(4

00),

Neg

ativ

efo

rsc

nA13

1/14

4

M1-

7167

5(4

00),

M2-

7167

6(2

04),

M3-

7167

5(3

24),

M4-

8534

8(6

57),

M5-

7168

0(0

00),

M6-

7168

1(0

00),

M8-

7168

2(3

24),

M9-

7168

3(3

24),

M11

-853

38(7

74),

M11

-716

84,M

12-7

1685

(774

),M

13-7

1686

(400

),M

14-8

5339

(204

),M

15-7

1688

(234

),M

17-8

5340

(000

),M

18-7

1690

(000

),M

19-7

1691

(000

),M

22-8

6331

(000

),M

23-8

5342

(400

),M

24-7

1694

(000

),M

25-7

1695

(774

),M

26-7

1696

(400

),M

27-8

5350

(774

),M

28-7

1698

(776

),M

29-7

1699

(400

),M

30-7

1700

(000

),M

31-7

1701

(324

),M

32-7

1702

(000

),M

33-8

7249

(324

),M

34-8

5451

(000

),M

36-7

1705

(400

),M

37-7

1706

(000

),M

38-7

1707

,M39

-717

08(0

00),

M40

-717

09(0

00),

M41

-717

10(4

00),

M42

-717

11,M

42-8

7335

(400

),M

43-8

6332

(000

),M

44-9

7418

(204

),M

46-7

1713

,M

47-8

5344

(204

),M

48-8

5345

(324

),M

48-7

1715

,M49

-717

16(3

24),

M50

-717

17(0

00),

M51

-717

18(4

00),

M52

-853

46(0

00),

M53

-717

20(4

00),

M54

-717

21(4

00),

M55

-922

00,M

55-8

5347

(400

),M

55-7

1722

,M

56-7

1723

(400

),M

57-9

9152

,M57

-994

35,M

59-9

1286

(324

),M

60-7

1727

(777

),M

61-7

1728

(234

),M

61-9

9190

,M62

-853

49(3

24),

M63

-724

53(3

24),

M64

-723

92,M

64-7

5411

(000

),M

65-7

5412

(324

),M

66-7

6182

(774

),M

67-7

5414

(400

),M

68-8

7331

(324

),M

69-7

5416

(324

),M

70-7

5417

(326

),M

71-7

5418

(777

),M

72-7

9300

(000

),M

73-8

7332

(324

),M

73-7

9301

,M74

-854

54(2

26),

M75

-863

34(2

04),

M76

-873

33(7

77),

M78

-874

57(4

00),

M79

-793

07(3

24),

M80

-793

08(4

00),

M81

-854

58(0

00),

M81

-793

09,e

mm

76-9

9436

,em

m82

-601

85(0

00),

emm

82-2

0060

182

(324

),em

m84

-601

74(6

74),

emm

85-6

0175

(204

),em

m86

-601

176

(657

),em

m87

-601

77(0

00),

emm

89-6

0186

(400

),em

m89

-618

2,em

m90

-601

78(0

00),

emm

91-6

0179

(000

),em

m92

-601

80(0

00),

emm

93-6

0187

(000

),em

m94

-994

37(3

24),

emm

95-9

9438

(000

),em

m96

-994

39(3

04),

emm

98-9

9441

(324

),em

m99

-994

42(6

54),

emm

100-

9944

3(4

54),

emm

101-

9944

4(4

00),

emm

102-

9944

5(4

00),

emm

103-

9944

6(0

00),

emm

104-

9944

7(0

00),

emm

106-

9945

0(0

00),

emm

107-

9945

1(2

24),

emm

108-

9945

3(2

04),

emm

110-

9945

5(7

26),

emm

111-

9945

6(7

26),

emm

112-

9945

7(7

26),

emm

114-

2000

0172

(226

),em

m11

5-20

0001

73(2

04),

emm

116-

2000

0174

(204

),em

m11

7-20

0001

76(0

00),

emm

118-

2000

0177

(226

),em

m11

9-20

0001

78(0

00),

emm

120-

2000

0183

(204

),em

m12

1-20

0001

85(0

00),

emm

122-

2000

0186

(626

),em

m12

3-20

0001

87(6

00),

emm

124-

2000

0184

(204

),M

28-6

-127

,M-T

14A

lask

a788

409

(234

),M

-T14

Ala

ska7

8051

,M49

Ala

ska7

8056

(324

),M

neg

Ala

ska7

8444

(400

),st

g454

5(G

pG

)(0

00),

DK

1,JT

1(0

00)


Table 4: Screen of extracts for their potential to induce inhibitor production by members of the SA-FF22 family.

Inhibitor-positive preparationtested for inducing activity

P-type of strain used as the sourceof inhibitor preparation

Preparation induces inhibitor production in

S. pyogenesFF22

S. salivariusG32

S. dysgalactiae 61

S. pyogenes FF22 436 Yes Yes Yes

S. pyogenes EB1 000 No No No

S. salivarius G32 436 Yes Yes Yes

S. dysgalactiae 61 436 Yes Yes Yes

L. lactis C2102 636 No No No

S. mutans K24 636 No No No

S. mutans H10 777 No No No

S. mitis SK653 777 No No No

K. varians NCC1482 777 No No No

Salivaricin G32 (pure) 436 No No No

slnR slnA slnA1 slnM

scnK scnR scnA scnM scnT scnE scnE scnGscn A

sln A

Figure 3: Graphical representation of the SA-FF22 locus illustratingthe section of the salivaricin G32 locus that has been sequenced.Black-filled arrows indicate genes that have been completelysequenced. Unfilled arrows indicate genes for which the completesequence has not yet been determined.

FF45, Min19, and Min52 were shown to have two copies ofscnA, an arrangement similar to that of the scnA locus in S.pyogenes strain M49 and for the slnA locus in S. salivariusstrains G32 and JH.

These observations indicate that the production of sali-varicin G32 could be a useful component of the anti-S. pyo-genes armoury of S. salivarius probiotics. The antibacterialspectrum of the SA-FF22 family of lantibiotics is known toextend beyond streptococci to also include a broad variety ofpotential pathogens and food spoilage bacteria such as Liste-ria, Leuconostoc, and Enterococcus [10]. As has been found fornisin-producing strains of the food grade species Lactococcuslactis, BLIS-producing S. salivarius also has potential efficacyand applicability when added to foods as preservatives,since they are known to be nonpathogenic [25]. Indeed thewidely-used BLIS-producing probiotic, S. salivarius strainK12, has recently received self-affirmed generally regardedas safe (GRAS) status for addition to food in the USA.One additional potential benefit for the human consumer offood containing salivaricin G32- (or macedocin-) producingbacteria is that the coproduced induction peptide may boostthe BLIS activity of salivaricin G32-positive S. salivariuspresent in the consumer’s oral microflora, thereby potentially

providing increased protection against S. pyogenes, as hasbeen previously reported for the stimulation of salivaricinA production by the host’s indigenous microflora uponingestion of inducer peptide-containing milk [26].

3.5. Amplification of an scnA-Like Gene in S. dysgalactiae . AscnA-like gene was also detected in S. dysgalactiae strain 61by PCR using primers designed to conserved regions of theSA-FF22 and salivaricin G32 structural genes. The structuralgene in S. dysgalactiae strain 61 appeared identical to that instrain FF22, although it is possible that the regions definedby these scnA-internal primers may differ in strain 61. S.dysgalactiae strain 61 was shown to produce a P-type of 436and freeze-thaw extracts cross-induced production of bothSA-FF22 and G32 (Table 4).

3.6. Megaplasmid Detection Using Pulsed Field Gel Elec-trophoresis. Analysis of S. salivarius G32, S. dysgalactiae 61,and S. pyogenes FF22 showed megaplasmid DNA (ca. 170 kb)to be present only in strain G32 (Figure 4). An apparentlyintact salivaricin G32 locus has also been detected on a220 kb megaplasmid in S. salivarius strain JH, where itcomprises part of a lantibiotic island adjacent to loci encod-ing salivaricin A and streptin [14] (N. Heng unpublished,2011). It is interesting to note that S. pyogenes strain FF22does not appear to contain megaplasmid DNA; however,both S. dysgalactiae 61 and 67 have similar-sized (48 kb)exogenous DNA fragments (which may be bacteriophage orplasmid). The S. salivarius megaplasmids do not therefore inthemselves appear to comprise the structural entity wherebylantibiotic determinants translocate to other streptococcalspecies, but rather, may act as repositories for lantibiotic-encoding DNA acquired from other bacteria.

3.7. Induction of SA-FF22 and Salivaricin G32 Production.The bacteriocin production of strains FF22, G32, and 61was increased (induced) following the addition of freezethaw extracts from 18 hr TsYECa cultures of either S. sali-varius G32 or S. dysgalactiae 61 and also by 18 hr THB


1 2 3 54 6 7

48.5

145

97

194(kb)

Figure 4: PFGE Analysis of megaplasmid content of total DNAof S. salivarius strains G32 (lane 1), 20P3 (lane 2), S. dysgalactiaestrains 67 (lane 3), and 61 (lane 4), and S. pyogenes strain FF22(lane 5). Lane 6 was left blank and lane 7 contained low range PFGmarker (New England Biolabs). Megaplasmid bands are indicatedwith white arrows and marker sizes are indicated to the right of thegel.

cultures of S. pyogenes strain FF22. No enhanced productionfollowed addition of extracts from strains having differentP-type profiles or preparations of purified salivaricin G32(Table 4). The inability of pure salivaricin G32 to induceproduction of any of this family of SA-FF22 lantibioticsindicates that the signal for upregulation of their productionis not the antimicrobial peptide itself, but rather, someother molecule formed by the lantibiotic producer strain.In an attempt to identify the inducing molecule, the SA-FF22-like peptides from a strain FF22 THB culture wereseparated by XAD-2 chromatography followed by anionicexchange then reversed phase HPLC using a C18 column.Individual fractions were assayed for induction of SA-FF22production in the liquid induction assay. The inducing agentinitially copurified with fractions also exhibiting inhibitoryactivity, but during C18 separation it eluted earlier (resultsnot shown). MALDI-TOF MS analysis of inducer activefractions however, only detected peptide having the massof SA-FF22, indicating that the molecule responsible (evenwhen in very small amounts) for the induction of SA-FF22 production may have copurified with residual (butnevertheless subinhibitory) quantities of SA-FF22 present inthat fraction. This is consistent with reports that the amountsof inducer peptide required to induce other lantibioticsystems are also extremely small [8]. Furthermore, the recentreport that macedocin is not an autoinducing molecule, butrather that an αS1-casein medium component can serve toinduce macedocin production in laboratory-grown cultures[27], lends further support to our contention that the SA-FF22 family of lantibiotics may not be autoinducible.

References

[1] J. R. Tagg, R. S. Read, and A. R. McGiven, “Bacteriocin of agroup A streptococcus: partial purification and properties,”Antimicrobial Agents and Chemotherapy, vol. 4, no. 3, pp. 214–221, 1973.

[2] J. R. Tagg, A. S. Dajani, L. W. Wannamaker, and E. D. Gray,“Group A streptococcal bacteriocin: production, purification,

and mode of action,” Journal of Experimental Medicine, vol.138, no. 5, pp. 1168–1183, 1973.

[3] R. W. Jack, A. Carne, J. Metzger et al., “Elucidation of thestructure of SA-FF22, a lanthionine-containing antibacterialpeptide produced by Streptococcus pyogenes strain FF22,”European Journal of Biochemistry, vol. 220, no. 2, pp. 455–462,1994.

[4] J. R. Tagg and L. W. Wannamaker, “Genetic basis of strep-tococcin A FF22 production,” Antimicrobial Agents andChemotherapy, vol. 10, no. 2, pp. 299–306, 1976.

[5] W. L. Hynes, J. J. Ferretti, and J. R. Tagg, “Cloning of the geneencoding streptococcin A-FF22, a novel lantibiotic producedby Streptococcus pyogenes, and determination of its nucleotidesequence,” Applied and Environmental Microbiology, vol. 59,no. 6, pp. 1969–1971, 1993.

[6] J. R. Tagg and L. V. Bannister, “‘Fingerprinting’ β-haemolyticstreptococci by their production of and sensitivity tobacteriocine-like inhibitors,” Journal of Medical Microbiology,vol. 12, no. 4, pp. 397–411, 1979.

[7] R. E. McLaughlin, J. J. Ferretti, and W. L. Hynes, “Nucleotidesequence of the streptococcin A-FF22 lantibiotic regulon:model for production of the lantibiotic SA-FF22 by strains ofStreptococcus pyogenes,” FEMS Microbiology Letters, vol. 175,no. 2, pp. 171–177, 1999.

[8] O. P. Kuipers, M. M. Beerthuyzen, P. G. G. A. De Ruyter,E. J. Luesink, and W. M. De Vos, “Autoregulation of nisinbiosynthesis in Lactococcus lactis by signal transduction,”Journal of Biological Chemistry, vol. 270, no. 45, pp. 27299–27304, 1995.

[9] M. Upton, J. R. Tagg, P. Wescombe, and H. F. Jenkinson,“Intra- and interspecies signaling between Streptococcus sali-varius and Streptococcus pyogenes mediated by SalA and SalA1lantibiotic peptides,” Journal of Bacteriology, vol. 183, no. 13,pp. 3931–3938, 2001.

[10] M. D. Georgalaki, E. Van den Berghe, D. Kritikos et al.,“Macedocin, a food-grade lantibiotic produced by Streptococ-cus macedonicus ACA-DC 198,” Applied and EnvironmentalMicrobiology, vol. 68, no. 12, pp. 5891–5903, 2002.

[11] K. H. Dyet, “Subgrouping strains that produce inhibitoryactivity similar to that of the lantibiotic Streptococcin A-FF22,” Honours Dissertation, University of Otago, Dunedin,New Zealand, 1999.

[12] J. Sambrook, E. F. Fritsch, and T. Maniatis, Molecular Cloning:A Laboratory Manual, Cold Spring Harbor Laboratory, NewYork, NY, USA, 2nd edition, 1989.

[13] P. A. Wescombe, M. Upton, P. Renault et al., “Salivaricin9, a new lantibiotic produced by Streptococcus salivarius,”Microbiology, vol. 157, no. 5, pp. 1290–1299, 2011.

[14] P. A. Wescombe, J. P. Burton, P. A. Cadieux et al., “Megaplas-mids encode differing combinations of lantibiotics in Strepto-coccus salivarius,” International Journal of General and Molecu-lar Microbiology, vol. 90, no. 3, pp. 269–280, 2006.

[15] K. F. Ross, C. W. Ronson, and J. R. Tagg, “Isolation and char-acterization of the lantibiotic salivaricin A and its structuralgene salA from Streptococcus salivarius 20P3,” Applied andEnvironmental Microbiology, vol. 59, no. 7, pp. 2014–2021,1993.

[16] R. P. Dempster and J. R. Tagg, “The production of bacteriocin-like substances by the oral bacterium Streptococcus salivarius,”Archives of Oral Biology, vol. 27, no. 2, pp. 151–157, 1982.

[17] O. Hyink, P. A. Wescombe, M. Upton, N. Ragland, J. P. Burton,and J. R. Tagg, “Salivaricin A2 and the novel lantibiotic sali-varicin B are encoded at adjacent loci on a 190-kilobase trans-missible megaplasmid in the oral probiotic strain Streptococcus


salivanus K12,” Applied and Environmental Microbiology, vol.73, no. 4, pp. 1107–1113, 2007.

[18] G. R. Tompkins and J. R. Tagg, “The ecology of bacteriocin-producing strains of Streptococcus salivarius,” Microbial Ecol-ogy in Health and Disease, vol. 2, pp. 19–28, 1989.

[19] C. Delorme, C. Poyart, S. D. Ehrlich, and P. Renault, “Extentof horizontal gene transfer in evolution of streptococci of thesalivarius group,” Journal of Bacteriology, vol. 189, no. 4, pp.1330–1341, 2007.

[20] L. De Vuyst and E. Vandamme, Bacteriocins of Lactic AcidBacteria: Microbiology, Genetics and Application, Chapmanand Hall, Hampshire, England, 1994.

[21] J. P. Burton, P. A. Wescombe, C. J. Moore, C. N. Chilcott,and J. R. Tagg, “Safety assessment of the oral cavity probi-otic Streptococcus salivarius K12,” Applied and EnvironmentalMicrobiology, vol. 72, no. 4, pp. 3050–3053, 2006.

[22] R. W. Jack and J. R. Tagg, “Factors affecting production ofthe group A streptococcus bacteriocin SA-FF22,” Journal ofMedical Microbiology, vol. 36, no. 2, pp. 132–138, 1992.

[23] P. A. Wescombe and J. R. Tagg, “Purification and characteriza-tion of streptin, a type a1 lantibiotic produced by streptococ-cus pyogenes,” Applied and Environmental Microbiology, vol.69, no. 5, pp. 2737–2747, 2003.

[24] P. A. Wescombe, M. Upton, K. P. Dierksen et al., “Productionof the lantibiotic salivaricin A and its variants by oralstreptococci and use of a specific induction assay to detecttheir presence in human saliva,” Applied and EnvironmentalMicrobiology, vol. 72, no. 2, pp. 1459–1466, 2006.

[25] J. P. Burton, S. Cowley, R. R. Simon, J. McKinney, P. A.Wescombe, and J. R. Tagg, “Evaluation of safety and humantolerance of the oral probiotic Streptococcus salivarius K12:A randomized, placebo-controlled, double-blind study,” Foodand Chemical Toxicology, vol. 49, no. 9, pp. 2356–2364, 2011.

[26] K. P. Dierksen, C. J. Moore, M. Inglis, P. A. Wescombe, and J.R. Tagg, “The effect of ingestion of milk supplemented withsalivaricin A-producing Streptococcus salivarius on the bac-teriocin-like inhibitory activity of streptococcal populationson the tongue,” FEMS Microbiology Ecology, vol. 59, no. 3, pp.584–591, 2007.

[27] M. Georgalaki, M. Papadelli, E. Chassioti et al., “Milk proteinfragments induce the biosynthesis of macedocin, the lantibi-otic produced by Streptococcus macedonicus ACA-DC 198,”Applied and Environmental Microbiology, vol. 76, no. 4, pp.1143–1151, 2010.


Review Article

Coleopteran Antimicrobial Peptides:Prospects for Clinical Applications

Monde Ntwasa,1 Akira Goto,2 and Shoichiro Kurata2

1 School of Molecular and Cell Biology, University of the Witwatersrand, Wits 2050, South Africa2 Graduate School of Pharmaceutical Sciences, Tohoku University, Aoba 6-3, Aramaki, Aoba-ku, Sendai 980-8578, Japan

Correspondence should be addressed to Shoichiro Kurata, [email protected]

Received 8 August 2011; Revised 2 November 2011; Accepted 5 December 2011


Copyright © 2012 Monde Ntwasa et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Antimicrobial peptides (AMPs) are activated in response to septic injury and have important roles in vertebrate and invertebrateimmune systems. AMPs act directly against pathogens and have both wound healing and antitumor activities. Althoughcoleopterans comprise the largest and most diverse order of eukaryotes and occupy an earlier branch than Drosophila in theholometabolous lineage of insects, their immune system has not been studied extensively. Initial research reports, however, indicatethat coleopterans possess unique immune response mechanisms, and studies of these novel mechanisms may help to furtherelucidate innate immunity. Recently, the complete genome sequence of Tribolium was published, boosting research on coleopteranimmunity and leading to the identification of Tribolium AMPs that are shared by Drosophila and mammals, as well as otherAMPs that are unique. AMPs have potential applicability in the development of vaccines. Here, we review coleopteran AMPs, theirpotential impact on clinical medicine, and the molecular basis of immune defense.

1. Overview

Research on innate immunity has led to an accumulation ofinformation that offers prospects for the development of an-timicrobial therapeutic drugs and vaccines. The low rate ofdiscovery of new antibiotics, the emergence of multiple-drugresistance, and the alarming death rate due to infection indi-cate a clear need for the development of alternative means tocombat infections. A highlight of the 20th century was thediscovery of vaccines that led to the eradication of diseasessuch as polio, small pox, and others. Even after more thantwo decades, however, a vaccine against the highly mutablehuman immune-deficiency virus remains to be developed,illustrating the need for new strategies to produce vaccines.A better understanding of innate immunity has revealed im-portant links between innate and adaptive immune systemsthat could lead to effective approaches in vaccine develop-ment.

Coleopterans comprise 40% of the 360,000 currentlyknown insect species and are therefore the largest and mostdiverse order of eukaryotic organisms [1]. Tribolium, the co-leopteran model, is proposed to be a better model than

Drosophila, especially for evolutionary studies, as it is ac-knowledged to be the most evolutionarily successful meta-zoan and to be more representative of insects in general thanDrosophila [1, 2]. Coleopterans, with no adaptive immunity,thrive on this planet. Studies of the molecular basis of cole-opteran immunity could therefore lead to a better under-standing of the evolution of the innate immune system.Much of the work on innate immunity and studies of thefunctional aspects of antimicrobial peptides (AMPs) hasbeen performed using Drosophila, which represents dipter-ans, while studies on coleopterans lag behind. Insects andhumans share innate immunity, but humans also have adap-tive immunity. Some of the conserved molecular signalingpathways that are used by insects and humans for immunedefense are also used for early embryonic development ininsects, but there are notable differences, probably due tothe fact that the innate immune systems of invertebrates andvertebrates diverged some 800 million years ago, and adap-tive immunity appeared in the vertebrate branch only about500 million years ago [3, 4]. The divergence of dipteransand coleoptera occurred some 284 million years ago, andDrosophila, in the dipteran branch, exhibits a remarkably


accelerated protein evolution [5]. Furthermore, despite theseseparate evolutionary paths, molecular coevolution couldhave occurred between coleoptera and mammals due to in-terdependence, that is, sharing common habitats and resour-ces.

While the majority of the work on immunity has beenconducted using Drosophila as a model, there is evidencethat coleoptera has retained many ancestral vertebrate genes,suggesting that studies of coleoptera could provide moreinsight into the properties and evolution of innate immunity.For example, Tribolium has many ancestral genes that arepresent in vertebrates and absent in Drosophila [6]. Similarly,the sequenced Tribolium genome revealed that ancestralgenes involved in cell-cell communication and developmentare retained in Tribolium, but not in Drosophila [2]. Further-more, in homology searches, human genes compare signifi-cantly better with Tribolium than Drosophila [5].

AMPs are small peptides characterized by an overall pos-itive charge (cationic), hydrophobicity, and amphipathicity.Structurally, they fall into two broad groups: linear α-helicaland cysteine-containing forms with one or more disulfidebridges and β-hairpin-like, β-sheet, or mixed α-helical/β-sheet structures. The peptides assume these conformationsupon contact with the target membranes [7–9]. Their char-acteristic physicochemical properties facilitate interactionswith the phospholipid bilayer in the cell membranes ofpathogens [10–12]. AMPs have been shown to kill pathogensdirectly by disrupting their membranes using mechanismsthat are not fully understood. Several models, however,have been proposed. First, there is the “barrel-stave” modelwhereby a transmembrane pore is created by amphipathic α-helical peptides, disrupting the cell membrane of a pathogen.Second, the “carpet” model proposes that the peptidessolubilize the membrane by interacting with the lipid headgroups on the pathogen cell surface. This model was alsoproposed for viral killing [13]. Another is the aggregationmodel that is exhibited by sapecin from Sarcophaga pereg-rina, based on the existence of hydrophobic and hydrophilicdomains on the AMPs. These structural features allowthe peptides to form pores with hydrophilic walls andhydrophobic regions facing the acyl side chains of pathogenmembrane phospholipids, thus facilitating movement ofhydrophilic molecules through the pore [14]. Finally, thetoroidal model, a subtle variation of the aggregation model,involves the formation of a dynamic lipid-layer core byhydrophilic regions of the peptide and lipid head groupsand is induced by magainins, melittin, and protegrins [15–17]. While the indispensability of the structural featuresof cationic peptides in pathogen killing is under debate,charge differences between cationic peptides and lipids onthe membrane are considered crucial. This may be the basisfor their selective activity as nonhemolytic peptides have ahigh net positive charge distributed along the peptide length,whereas hemolytic peptides have a low negative charge [10,11]. Evidence suggests that AMPs have intracellular targets.This is exemplified by elafin, a cationic and α-helical humaninnate defense AMP that does not lyse the bacterial mem-brane and is translocated into the cytoplasm. In vitro analysisusing a mobility shift assay revealed that elafin binds DNA

[18]. The histone-derived peptide buforin II binds nucleicacids in gel retardation assays and rapidly kills Escherichiacoli by translocating into the cytoplasm of the pathogen andprobably interfering with the functions of DNA or RNA.The structurally similar magainin 2 also kills E. coli but doesnot enter the cytoplasm [19]. Similarly, cationic antibacterialpeptides enter the cytoplasm of Aspergillus nidulans andkill the fungus by targeting intracellular molecules whoseidentity has not been verified [20]. An excellent review ofthe intracellular targets of AMPs was recently published [21].More studies are required, however, to confirm the existenceand actual mode of action of AMPs with intracellular targets.

Insects produce AMPs constitutively at local sites or theAMPs are released systemically upon pathogenic infection toinitiate pathogen-killing activities. In addition to the well-characterized Drosophila and mouse innate immune signal-ing pathways, the sequencing of the Tribolium genome hasboosted research progress because bioinformatics analysesrevealed putative immune-related genes based on compar-isons with the genomes of other species [22].

AMPs are multifunctional molecules that, in addition totheir well-known role as effectors of the innate immune sys-tem, are involved in several biologic processes and pathologicconditions, such as immune modulation, angiogenesis, andcytokine and histamine release [23–27]. Probably due to thenegative charge in the plasma membrane of many cancercells, some cationic peptides also have anticancer activity [28,29]. These properties can be potentially exploited for clinicalpurposes [12, 30]. Cecropins are selectively cytotoxic tocancer cells, preventing their proliferation in bladder cancer,and are therefore likely candidates in strategies for the devel-opment of anticancer drugs [31]. In addition to antimicro-bial activity, defensins facilitate the induction of adaptiveimmunity and promote cell proliferation and wound healing.Defensins show chemotactic activity whereby dendritic cells,monocytes, and T cells are recruited to the site of infection.Moreover, human β-defensins and the cathelicidin LL-37stimulate the production of pruritogenic cytokines, such asinterleukin-31, leukotrienes, prostaglandin E2, and others,suggesting an important role in allergic reactions [32–34].AMPs also form the basis of the potentially lucrative com-mercial area of “cosmeceuticals”-products with beneficialtopical activities that are delivered by rubbing, sprinkling,spraying, and so forth [35].

Here, we review the progress made in discovery of co-leopteran AMPs, the molecular basis of Tribolium innateimmunity, and prospects for the application of antimicrobialpeptides in medicine.

2. The Discovery Process

2.1. Antimicrobial Peptides in Tribolium . The first wide-scalestudy of Tribolium immunity was conducted by Zou et al.in 2007 [22]. Taking advantage of the fully sequenced Tri-bolium genome to predict putative immune genes using bi-oinformatics techniques and real-time polymerase chainreaction (PCR), Zou et al. [22] predicted 12 AMPs in Tri-bolium compared to 20 in Drosophila, the most studied


invertebrate. Another study using suppression subtractivehybridization led to the addition of a few more AMPs to thislist [36] (see Table 1). Both studies identified four defensinsin Tribolium, and phylogenetic analysis indicated that threeof these are found in the evolutionary branch comprisingonly coleopterans. The fourth defensin (Def4) is found ina mixed branch that includes hymenopterans. A search ofthe Defensins Knowledgebase [37] revealed that the sequenceinformation of this defensin is not available in the publicdomain, although its existence has been reported [22]. Atta-cins, which were identified in lepidopterans, were found in acluster of three genes. Attacins are rich in glycine and proline,are structurally similar to coleoptericins, and are inducibleby bacteria. Furthermore, Drosophila studies demonstratedthat the induction of attacin is reduced in both imd andTl− mutants [38]. Coleoptericins were first isolated from thelarvae of Allomyrina dichotoma beetles immunized with E.coli. Coleoptericins also show activity against Staphylococcusaureus, methicillin-resistant S. aureus, and Bacillus subtilis.Like attacins, but unlike cecropins, coleoptericins do notform pores on the bacterial membrane, but do cause defectsin cell division, as liposomes containing E. coli or S. aureusmembrane constituents do not leak upon treatment with therecombinant form of coleoptericin, but instead form chains[39].

Tribolium cecropins are predicted to be pseudogenes be-cause of a shift in the open reading frame; some cecropin-related proteins with an unusual structure, however, havebeen reported [22]. A cecropin has been reported in at leastone coleopteran, Acalolepta luxuriosa [44].

Four thaumatin-like genes were found in Tribolium usingsuppression subtractive hybridization and genome search.Experimentally, septic injury induces thaumatin-1 anddefensins in Tribolium [36]. Sterile wounding also inducesthaumatin-1 and defensin-2. Furthermore, recombinantthaumatin-1 heterologously overexpressed in E. coli is activeagainst fungi [36]. Coleopteran cationic peptides might beremarkably different from other known peptides and aretherefore not readily identified by homology searches. A clearhomolog of the Drosophila antifungal drosomycin could notbe found in the Tribolium genome, but a weakly homologousprotein with a cysteine-rich sequence was detected [22].An overview of Tribolium AMPs indicates similarities withother coleopterans, but some differences with Drosophila.The work reported by these groups provides a good basis foradvancing research on coleopteran AMPs.

2.2. Other Antimicrobial Peptides Identified in Coleopterans.A number of AMPs present in certain coleopterans have notyet been identified in Tribolium (see Table 2). One of these isan interesting class of insect peptides that adopts the knottinfold and was first identified in 2003 from the harlequin bee-tle, Acrocinus longimanus. Members of this class include Alo-1, Alo-2, and Alo3 [42]. Psacotheasin from the yellow starlonghorn beetle Psacothea hilaris has also been identified asa member of this class [43, 45]. Alo-3 is active against fungi,while psacotheasin is active against bacteria and fungi. Theknottin fold is characterized by a disulfide topology of the

“abcabc” type, in which disulfide bridges are formed betweenthe first cysteine and the fourth, second, and fifth cysteines,and the third and sixth cysteines [46]. Disulfide bridge for-mation may confer important properties to the peptides,such as stability and resistance to protease cleavage. Membersof the knottin family in general have low sequence simi-larity, reducing their chances of identification by homologysearches [46]. In contrast, however, the coleopteran knottinfold AMPs share sequence similarities with several plantantifungal peptides [42]. Although the mechanism by whichthese peptides function is not fully understood, psacotheasinkills Candida albicans by inducing apoptosis [47]. This hasclinical significance as C. albicans can cause mild superficialto severe infections in immunocompromised patients. Abetter understanding of the molecular events that are criticalto the induction of apoptosis by cationic peptides could leadto new targets for antifungal drug development. Alarmingly,candidemia, a systemic Candida infection, is on the increaseand is accompanied by the reemergence of resistance againstcommon drugs, pointing to the urgency of finding alterna-tive means of treating fungal infections [48, 49].

2.3. Databases. The Antimicrobial Peptides database, a com-prehensive and searchable database for AMPs was establishedbased on information from literature surveys [50, 51]. Cur-rently, an updated version on the website indicates that thereare 1773 cationic peptides in the database, including antiviral(5.8%), antibacterial (78.56%), antifungal (31.19%), andantitumor (6.14%) peptides. Some of these peptides functionagainst more than one type of pathogens. The structures of231 of these peptides have been determined by nuclear mag-netic resonance and X-ray diffraction studies. Another usefuldatabase is the Defensins Knowledgebase, which allows text-based searches for information on this large family of AMPs[37]. It is a manually curated and specialized database sim-ilar to the shrimp penaeidin database, PenBase [52]. We havealso started molecular studies of another coleopteran, thedung beetle Euoniticellus intermedius, and sequenced theadult transcriptome with a view to study its immune system[53]. These databases serve as useful tools for the discoveryand design of new peptides. Indeed, key features upon whichantimicrobial activity is based have been studied using theAntimicrobial Peptides database [54, 55]. Such analyses gen-erate an important information pool for drug design.

3. Regulation of AMP Expressionby Coleopterans

The signaling pathways that mediate the immune responsein Tribolium castaneum were initially predicted based on acombination of in silico studies and experimental work byZou et al. [22] and more recently another study involvingthe burying beetle Nicrophorus vespilloides [56]. In addition,studies using adult beetles exposed to E. coli, M. luteus,C. albicans, and S. cerevisiae have provided information onthe signaling pathways. Accordingly, large-scale studies usingreal-time PCR revealed the presence of innate immunegenes, such as PGRP-LA, PGRP-LE, PGRP-SA, PGRP-SB,


Table 1: Antimicrobial peptides currently predicted or identified in Tribolium.

Antimicrobial peptide Accession number Reference Target Method of identification

Attacin1 GLEAN 07737 [22] Homology searches



Cecropin1 GLEAN 00499 [22, 31] Antibacterial, antitumor Homology searches

Cecropin2 Cec2 [22, 31] Antibacterial, antitumor Homology searches

Cecropin3 GLEAN 00500 [22, 31] Antibacterial, antitumor Homology searches

Defensin1GLEAN 06250;

XM 962101[22, 36] Antibacterial

Homology searches and suppressionsubtractive hybridization







Defensin4 Def4 [22] Homology searches

Coleoptericin1 GLEAN 05093 [22] Antibacterial Homology searches

Coleoptericin2 GLEAN 05096 [22] Antibacterial Homology searches

Similar to thaumatin family XM 963631 [36] Antifungal Suppression subtractive hybridization

Probable antimicrobial peptide Tc11324 [22] Homology searches

Putative antimicrobial peptide AM712902 [36] Suppression subtractive hybridization

Table 2: Antimicrobial peptides expressed in other coleopterans not yet identified in Tribolium.

Antimicrobial peptide Organism Accession no. Reference

Diptericin A S. zeamis, (G. morsitans) Q8WTD5 [40]

Acaloleptin A S. zeamis (A. luxuriosa) Q76K70 [40]

Sarcotoxin II-1 S. zeamis, (S. peregrina) P24491 [40]

Tenecin-1 S. zeamis, (T. molitor) Q27023 [40, 41]

Tenecin-2 T. molitor [41]

Luxuriosin S. zeamis, (A. luxuriosa) Q60FC9 [40]

Alo-3 (knottin type) A. longimanus P83653 [42]

Psacotheasin (Knottintype)

P. hilaris [43]

several Toll proteins, and the immune deficiency (IMD) pro-tein. Notably, some of the PGRPs had no orthologs in Dro-sophila, indicating a diversity of specificity. Recent biochem-ical studies using the large beetles Tenebrio molitor and Hol-otrichia diomphalia further elucidated the extracellular sig-naling network involved in responses to fungal and bacterialinfections [41, 57]. Overall, coleopteran signaling appears tooccur via the Toll and IMD pathways (Figure 1).

The Toll pathway is activated by PAMPS such as β-1,3-glucans, found in fungi, and by Lys-type peptidoglycans(PGN), found primarily in Gram-positive bacteria. A com-plex of the PAMPS and pathogen recognition receptors(PRRs) activates an apical protease, leading to a three-stepserine protease cascade that culminates in the generation ofactive spaetzle, the ligand of the transmembrane receptorToll. Subsequent intracellular signaling leads to the transcrip-tional activation of genes that encode antimicrobial peptides.

Activation of the immune response by DAP-type PGNfound primarily in Gram-negative bacteria and Gram-positive bacilli is still poorly understood in flies and beetles.

Generally, it is understood that Gram-negative bacteriarequire the IMD pathway because imd− mutants cannot ex-press antimicrobial peptides against Gram-negative bacte-ria. In Drosophila, candidates for the signal transduction-activated Gram-negative bacteria are the transmembranereceptor PGRP-LC and PGRPP-LE. Both molecules canactivate the IMD pathway [3, 58]. Because these moleculesare present in beetles and PGRP-LE is orthologous to theDrosophila protein, it is likely that the corresponding path-ways are conserved. In Tribolium, PGRP-LA and PGRP-LEare activated by bacterial infection, but poorly activated byC. albicans and M. luteus [22]. Other Tribolium studies showthat the IMD pathway is activated by two Gram-negativebacteria, Xenorhabdus nematophila and E. coli, inducing 12AMPs of which 5 are significantly dependent on the IMDpathway as demonstrated by RNA interference studies [59].The same study, however, demonstrated that two Gram-positive bacteria with different peptidoglycans expressed thesame AMPs with only defensin-1 being dependent on Toll.Taken together, these studies show that while the pathways


GNBP3

MSP (apical)

SAE

SPE

PGRP-LC?

Anti-Gram-negative bacterial(Tenascin2)

Spzpro-Spz

Toll receptor

AMPs AMPs

Antifungal and antibacterial(Tenascin 1 and Tenascin 2)

Activation mechanisms in coleopteran immune system

β-1,3-glucans

Pathogens

PAMPS

PRRs

Proteasecascades

Ligands

Receptors

?

LRP

LPSDAP-type PG(monomeric)

Gram-negativebacteria

Gram-positivebacteria

Fungi

Lys-type or DAP-typePG (polymeric)

PGRP-SAGNBP1

Figure 1: Activation mechanisms in the coleopteran immune system. Immune response pathways activated by bacteria and fungi showing apathogen-associated recognition pattern (PAMP), pattern recognition receptors (PRRs), and downstream signaling molecules. The proteasecascade in the Toll pathway involves the modular apical modular serine protease (MSP), the Spz-processing enzyme-activating enzyme(SAE), and the spaetzle processing enzyme (SPE). GNBP3: glucan binding protein 3; PGRP: peptidoglycan recognition protein.

may be conserved, differences in PAMPS recognition and sig-nal transduction exist between Tribolium and Drosophila.

The discovery of another PRR known as the LPS recog-nition protein (LRP) based on its E. coli agglutinating prop-erties suggests the existence of an LPS pathway. LRP circu-lates in the hemolymph and does not agglutinate S. aureusor C. albicans. Interestingly, LRP comprises six repeats of anepidermal-growth-factor- (EGF)-like domain, an unusualstructural feature for PRRs [60]. The downstream events inthis pathway remain unclear.

4. Antimicrobial Peptides in Clinical Medicine

Cationic peptides have emerged as important targets for thedevelopment of therapeutics against bacteria, fungi, viruses,and parasites. They are key effector molecules in host de-fense through direct and indirect antimicrobial activity. Fur-thermore, in vertebrates, these peptides mediate a variety ofcellular processes such as immunomodulation, wound heal-ing, and tumorigenesis. These roles provide opportunitiesfor the development of therapeutic products and vaccines.

AMPs are attractive molecules for the development of clinicaland veterinary therapeutics because they are fast acting andeffective against susceptible pathogens, are less likely to causethe emergence of resistance compared to traditional antibi-otics, have low toxicity to mammalian cells, and their modeof action tends to be more physical rather than targeted atmetabolic pathways. A search of the FreePatentsOnline data-base using the word “antimicrobial peptide” produced morethan 66.000 hits, and a number of AMPs have undergoneclinical development [30]. A recent review of cationic pep-tides lists the peptides that are in various stages of clinicaltrials [29].

As mentioned above, the predicted Tribolium AMPsinclude defensins, attacin, coleoptericin, thaumatin, and ce-cropin. Defensins exhibit a broad spectrum of antimicrobialactivity directed at bacteria, fungi, and viruses and are prob-ably the most studied class of AMPs. Many therapeutic pro-ducts have been modeled on them. The different types ofdefensins are either expressed constitutively or induced byinfections to control the composition of microorganisms onsurfaces such as the small and large intestines [61].


Many challenges remain that hamper the developmentof commercially viable peptides. The pressing issues concernpharmacokinetics (how the body deals with peptide drugs).When peptides are administered orally, the gastrointestinaltract may prevent their reabsorption into the systemic cir-culation. Furthermore, peptides may elicit an antigenic re-sponse when injected directly in the blood. This leavestopical medication the most feasible formulation while moreresearch is being pursued to address the remaining obstacles.Despite these obstacles, the prospects for AMPs are not bleakbecause some have proceeded to clinical application. Thereis some optimism that these obstacles may soon be overcomeby new strategies that combine natural cationic peptides andstable synthetic immunomodulatory peptides [29]. In thisregard, peptide drugs such as Polymyxin B and gramicidinthat are used for the treatment of Gram-negative bacterialinfections are reported to be safe and effective, and peptidessuch as the indolicidin-derived CLS001 (previously knownas MX594AN) have reached phase III clinical trials withpromising prospects [62–64]. Because of their evolutionarydistance, during which their survival against microbes hasbeen solely dependent on innate immunity, insects provideinteresting models for novel AMP drug design [65, 66].

5. Conclusions

The emergence of multidrug-resistant pathogens threatenshuman health globally and presents an urgent need to findantimicrobials with a reduced chance of inducing resistance.Cationic peptides for which the mechanism of actioninvolves targeting the plasma membrane in a nonspecificmanner, but does not involve specific proteins, offer goodprospects. Admittedly, more work is needed to elucidate themechanism of action of these peptides, as there is someevidence for intracellular targets. The importance of cationicpeptides is further highlighted by their emerging prospects inother aspects of medicine, such as cancer treatment and vac-cine development. Coleopterans are the most evolutionarilysuccessful group of insects and are more representative ofinsects than Drosophila. In addition, human genes are morecomparable to those of Tribolium than those of Drosophila.Thus, coleopterans are emerging as an important speciesfor study as, like vertebrates, they have retained ancestralgenes that are not present in Drosophila. Indeed, there isoverwhelming evidence that coleopterans are more suitablefor comparative studies between phyla than the commonlyused dipterans. Here, we suggest that perhaps the outstand-ing evolutionary success of coleopterans is consistent with arobust immune system that warrants more attention than ithas received to date.

Acknowledgments

All the authors are supported by the South Africa/Japan JointScience and Technology Research Collaboration through theNational Research Foundation and the Japan Society forthe Promotion of Science. This work was supported by theStrategic International Cooperative program from the Japan

Science and Technology Agency; Grants-in-Aid for ScientificResearch from the Ministry of Education, Culture, Sports,Science, and Technology of Japan; the Japan Society for thePromotion of Science; the Program for the Promotion ofBasic Research Activities for Innovative Biosciences (PRO-BRAIN); the National Institutes of Health (AI07495); theTakeda Science Foundation; the Mitsubishi Foundation; theNaito Foundation; Astellas Foundation for Research onMetabolic Disorders; a Global COE Research Grant (TohokuUniversity Ecosystem Adaptability); the National ResearchFoundation (NRF), South Africa.

References

[1] D. Tautz, “Insects on the rise,” Trends in Genetics, vol. 18, no.4, pp. 179–180, 2002.

[2] S. Richards, R. A. Gibbs, G. M. Weinstock et al., “The genomeof the model beetle and pest Tribolium castaneum,” Nature,vol. 452, no. 7190, pp. 949–955, 2008.

[3] J. A. Hoffmann, “The immune response of Drosophila,” Na-ture, vol. 426, no. 6962, pp. 33–38, 2003.

[4] J. A. Hoffmann, “Primitive immune systems,” ImmunologicalReviews, vol. 198, pp. 5–9, 2004.

[5] J. Savard, D. Tautz, and M. J. Lercher, “Genome-wide accelera-tion of protein evolution in flies (Diptera),” BMC EvolutionaryBiology, vol. 6, p. 7, 2006.

[6] M. van der Zee, R. N. da Fonseca, and S. Roth, “TGFβ sig-naling in Tribolium: vertebrate-like components in a beetle,”Development Genes and Evolution, vol. 218, no. 3-4, pp. 203–213, 2008.

[7] P. Bulet and R. Stocklin, “Insect antimicrobial peptides: struc-tures, properties and gene regulation,” Protein and PeptideLetters, vol. 12, no. 1, pp. 3–11, 2005.

[8] P. Bulet, R. Stocklin, and L. Menin, “Anti-microbial peptides:from invertebrates to vertebrates,” Immunological Reviews, vol.198, no. 1, pp. 169–184, 2004.

[9] A. Tossi, L. Sandri, and A. Giangaspero, “Amphipathic, α-helical antimicrobial peptides,” Peptide Science, vol. 55, no. 1,pp. 4–30, 2000.

[10] Y. Shai, “Mechanism of the binding, insertion and destabili-zation of phospholipid bilayer membranes by α-helical anti-microbial and cell non-selective membrane-lytic peptides,”Biochimica et Biophysica Acta, vol. 1462, no. 1-2, pp. 55–70,1999.

[11] Z. Oren and Y. Shai, “Mode of action of linear amphipathicα-helical antimicrobial peptides,” Peptide Science, vol. 47, no.6, pp. 451–463, 1998.

[12] K. V. R. Reddy, R. D. Yedery, and C. Aranha, “Antimicrobialpeptides: premises and promises,” International Journal ofAntimicrobial Agents, vol. 24, no. 6, pp. 536–547, 2004.

[13] R. E. Dean, L. M. O’Brien, J. E. Thwaite, M. A. Fox, H. Atkins,and D. O. Ulaeto, “A carpet-based mechanism for directantimicrobial peptide activity against vaccinia virus mem-branes,” Peptides, vol. 31, no. 11, pp. 1966–1972, 2010.

[14] K. Takeuchi, H. Takahashi, M. Sugai et al., “Channel-formingmembrane permeabilization by an antibacterial protein,sapecin,” Journal of Biological Chemistry, vol. 279, no. 6, pp.4981–4987, 2004.

[15] K. Matsuzaki, “Magainins as paradigm for the mode of actionof pore forming polypeptides,” Biochimica et Biophysica Acta,vol. 1376, no. 3, pp. 391–400, 1998.


[16] K. A. Brogden, “Antimicrobial peptides: pore formers ormetabolic inhibitors in bacteria?” Nature Reviews Microbiol-ogy, vol. 3, no. 3, pp. 238–250, 2005.

[17] R. Mani, S. D. Cady, M. Tang, A. J. Waring, R. I. Lehrer, andM. Hong, “Membrane-dependent oligomeric structure andpore formation of a β-hairpin antimicrobial peptide in lipidbilayers from solid-state NMR,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 103,no. 44, pp. 16242–16247, 2006.

[18] A. Bellemare, N. Vernoux, S. Morin, S. M. Gagne, and Y.Bourbonnais, “Structural and antimicrobial properties of hu-man pre-elafin/trappin-2 and derived peptides against Pseu-domonas aeruginosa,” BMC Microbiology, vol. 10, no. 1, p. 253,2010.

[19] C. B. Park, H. S. Kim, and S. C. Kim, “Mechanism of actionof the antimicrobial peptide buforin II: buforin II kills micro-organisms by penetrating the cell membrane and inhibitingcellular functions,” Biochemical and Biophysical Research Com-munications, vol. 244, no. 1, pp. 253–257, 1998.

[20] D. Mania, K. Hilpert, S. Ruden, R. Fischer, and N. Takeshita,“Screening for antifungal peptides and their modes of actionin Aspergillus nidulans,” Applied and Environmental Microbiol-ogy, vol. 76, no. 21, pp. 7102–7108, 2010.

[21] J. D. F. Hale and R. E. W. Hancock, “Alternative mechanisms ofaction of cationic antimicrobial peptides on bacteria,” ExpertReview of Anti-Infective Therapy, vol. 5, no. 6, pp. 951–959,2007.

[22] Z. Zou, J. D. Evans, Z. Lu et al., “Comparative genomic analysisof the Tribolium immune system,” Genome Biology, vol. 8, no.8, p. R177, 2007.

[23] R. Bals and J. M. Wilson, “Cathelicidins—a family of multi-functional antimicrobial peptides,” Cellular and Molecular LifeSciences, vol. 60, no. 4, pp. 711–720, 2003.

[24] M. G. Scott and R. E. W. Hancock, “Cationic antimicrobialpeptides and their multifunctional role in the immune sys-tem,” Critical Reviews in Immunology, vol. 20, no. 5, pp. 407–431, 2000.

[25] J. E. Meyer and J. Harder, “Antimicrobial peptides in oralcancer,” Current Pharmaceutical Design, vol. 13, no. 30, pp.3119–3130, 2007.

[26] A. F. Gombart, N. Borregaard, and H. P. Koeffler, “Humancathelicidin antimicrobial peptide (CAMP) gene is a directtarget of the vitamin D receptor and is strongly up-regulated inmyeloid cells by 1,25-dihydroxyvitamin D3,” FASEB Journal,vol. 19, no. 9, pp. 1067–1077, 2005.

[27] B. de Yang, Q. Chen, A. P. Schmidt et al., “LL-37, the neu-trophil granule-and epithelial cell-derived cathelicidin, utilizesformyl peptide receptor-like 1 (FPRL1) as a receptor to che-moattract human peripheral blood neutrophils, monocytes,and T cells,” Journal of Experimental Medicine, vol. 192, no.7, pp. 1069–1074, 2000.

[28] F. Schweizer, “Cationic amphiphilic peptides with cancer-selective toxicity,” European Journal of Pharmacology, vol. 625,no. 1–3, pp. 190–194, 2009.

[29] A. T. Y. Yeung, S. L. Gellatly, and R. E. W. Hancock, “Mul-tifunctional cationic host defence peptides and their clinicalapplications,” Cellular and Molecular Life Sciences, vol. 68, no.13, pp. 2161–2176, 2011.

[30] E. Andres and J. L. Dimarcq, “Clinical development ofantimicrobial peptides,” International Journal of AntimicrobialAgents, vol. 25, no. 5, pp. 448–449, 2005.

[31] H. Suttmann, M. Retz, F. Paulsen et al., “Antimicrobial pep-tides of the Cecropin-family show potent antitumor activity

against bladder cancer cells,” BMC Urology, vol. 8, no. 1, p. 5,2008.

[32] D. Yang, O. Chertov, S. N. Bykovskaia et al., “β-Defensins:linking innate and adaptive immunity through dendritic andT cell CCR6,” Science, vol. 286, no. 5439, pp. 525–528, 1999.

[33] F. Niyonsaba, H. Ushio, M. Hara et al., “Antimicrobial pep-tides human β-defensins and cathelicidin LL-37 induce the se-cretion of a pruritogenic cytokine IL-31 by human mast cells,”Journal of Immunology, vol. 184, no. 7, pp. 3526–3534, 2010.

[34] S. Dressel, J. Harder, J. Cordes et al., “Differential expression ofantimicrobial peptides in margins of chronic wounds,” Exper-imental Dermatology, vol. 19, no. 7, pp. 628–632, 2010.

[35] L. Zhang and T. J. Falla, “Cosmeceuticals and peptides,” Clinicsin Dermatology, vol. 27, no. 5, pp. 485–494, 2009.

[36] B. Altincicek, E. Knorr, and A. Vilcinskas, “Beetle immunity:identification of immune-inducible genes from the model in-sect Tribolium castaneum,” Developmental and ComparativeImmunology, vol. 32, no. 5, pp. 585–595, 2008.

[37] S. Seebah, A. Suresh, S. Zhuo et al., “Defensins knowledgebase:a manually curated database and information source focusedon the defensins family of antimicrobial peptides,” NucleicAcids Research, vol. 35, supplement 1, pp. D265–D268, 2007.

[38] M. Hedengren, K. Borge, and D. Hultmark, “Expression andevolution of the Drosophila Attacin/Diptericin gene family,”Biochemical and Biophysical Research Communications, vol.279, no. 2, pp. 574–581, 2000.

[39] A. Sagisaka, A. Miyanoshita, J. Ishibashi, and M. Yamakawa,“Purification, characterization and gene expression of a gly-cine and proline-rich antibacterial protein family from larvaeof a beetle, Allomyrina dichotoma,” Insect Molecular Biology,vol. 10, no. 4, pp. 293–302, 2001.

[40] C. Anselme, V. Perez-Brocal, A. Vallier et al., “Identification ofthe weevil immune genes and their expression in the bacteri-ome tissue,” BMC Biology, vol. 6, no. 1, p. 43, 2008.

[41] Y. Yu, J.-W. Park, H.-M. Kwon et al., “Diversity of innate im-mune recognition mechanism for bacterial polymeric meso-diaminopimelic acid-type peptidoglycan in insects,” Journal ofBiological Chemistry, vol. 285, no. 43, pp. 32937–32945, 2010.

[42] F. Barbault, C. Landon, M. Guenneugues et al., “Solutionstructure of Alo-3: a new knottin-type antifungal peptide fromthe insect Acrocinus longimanus,” Biochemistry, vol. 42, no. 49,pp. 14434–14442, 2003.

[43] J. S. Hwang, J. Lee, B. Hwang et al., “Isolation and charac-terization of psacotheasin, a novel knottin-type antimicrobialpeptide, from Psacothea hilaris,” Journal of Microbiology andBiotechnology, vol. 20, no. 4, pp. 708–711, 2010.

[44] A. Saito, K. Ueda, M. Imamura et al., “Purification and cDNAcloning of a cecropin from the longicorn beetle, Acaloleptaluxuriosa,” Comparative Biochemistry and Physiology B, vol.142, no. 3, pp. 317–323, 2005.

[45] B. Hwang, J. S. Hwang, J. Lee, and D. G. Lee, “Antifungal prop-erties and mode of action of psacotheasin, a novel knottin-type peptide derived from Psacothea hilaris,” Biochemical andBiophysical Research Communications, vol. 400, no. 3, pp. 352–357, 2010.

[46] S. Lu, P. Deng, X. Liu et al., “Solution structure of the majorα-amylase inhibitor of the crop plant amaranth,” Journal ofBiological Chemistry, vol. 274, no. 29, pp. 20473–20478, 1999.

[47] B. Hwang, J. S. Hwang, J. Lee, and D. G. Lee, “The antimi-crobial peptide, psacotheasin induces reactive oxygen speciesand triggers apoptosis in Candida albicans,” Biochemical andBiophysical Research Communications, vol. 405, no. 2, pp. 267–271, 2011.


[48] R. E. Lewis and M. E. Klepser, “The changing face of noso-comial candidemia: epidemiology, resistance, and drug ther-apy,” American Journal of Health-System Pharmacy, vol. 56, no.6, pp. 525–533, 1999.

[49] M. H. Nguyen, J. E. Peacock, A. J. Morris et al., “The chang-ing face of candidemia: emergence of non-Candida albicansspecies and antifungal resistance,” American Journal of Medi-cine, vol. 100, no. 6, pp. 617–623, 1996.

[50] Z. Wang and G. Wang, “APD: the antimicrobial peptidedatabase,” Nucleic Acids Research, vol. 32, supplement 1, pp.D590–D592, 2004.

[51] G. Wang, X. Li, and Z. Wang, “APD2: the updated antimicro-bial peptide database and its application in peptide design,”Nucleic Acids Research, vol. 37, supplement 1, pp. D933–D937,2009.

[52] Y. Gueguen, J. Garnier, L. Robert et al., “PenBase, the shrimpantimicrobial peptide penaeidin database: sequence-basedclassification and recommended nomenclature,” Developmen-tal and Comparative Immunology, vol. 30, no. 3, pp. 283–288,2006.

[53] L. M. Khanyile, R. Hull, and M. Ntwasa, “Dung beetle data-base: comparison with other invertebrate transcriptomes,” Bi-oinformation, vol. 3, no. 4, pp. 159–161, 2008.

[54] S. Lata, N. K. Mishra, and G. P. S. Raghava, “AntiBP2: im-proved version of antibacterial peptide prediction,” BMCBioinformatics, vol. 11, no. 1, p. S19, 2010.

[55] S. Lata, B. K. Sharma, and G. P. S. Raghava, “Analysis andprediction of antibacterial peptides,” BMC Bioinformatics, vol.8, p. 263, 2007.

[56] H. Vogel, C. Badapanda, and A. Vilcinskas, “Identification ofimmunity-related genes in the burying beetle Nicrophorusvespilloides by suppression subtractive hybridization,” InsectMolecular Biology, vol. 20, no. 6, pp. 787–800, 2011.

[57] J.-W. Park, C.-H. Kim, J. Rui et al., Beetle Immunity Inverte-brate Immunity, Springer, New York, NY, USA, 2010.

[58] A. Takehana, T. Katsuyama, T. Yano et al., “Overexpression ofa pattern-recognition receptor, peptidoglycan-recognitionpro-tein-LE, activates imd/relish-mediated antibacterial de-fense and the prophenoloxidase cascade in Drosophila larvae,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 99, no. 21, pp. 13705–13710, 2002.

[59] S. Shrestha and Y. Kim, “Activation of immune-associatedphospholipase A2 is functionally linked to Toll/Imd signalpathways in the red flour beetle, Tribolium castaneum,” Devel-opmental and Comparative Immunology, vol. 34, no. 5, pp.530–537, 2010.

[60] J. S. Ju, M. H. Cho, L. Brade et al., “A novel 40-kDaprotein containing six repeats of an epidermal growth factor-like domain functions as a pattern recognition protein forlipopolysaccharide,” Journal of Immunology, vol. 177, no. 3, pp.1838–1845, 2006.

[61] J. Wehkamp, E. F. Stange, and K. Fellermann, “Defensin-im-munology in inflammatory bowel disease,” GastroenterologieClinique et Biologique, vol. 33, supplement 3, pp. S137–S144,2009.

[62] M. E. Evans, D. J. Feola, and R. P. Rapp, “Polymyxin B sulfateand colistin: old antibiotics for emerging multiresistant gram-negative bacteria,” Annals of Pharmacotherapy, vol. 33, no. 9,pp. 960–967, 1999.

[63] A. K. Marr, W. J. Gooderham, and R. E. Hancock, “Antibac-terial peptides for therapeutic use: obstacles and realistic out-look,” Current Opinion in Pharmacology, vol. 6, no. 5, pp. 468–472, 2006.

[64] S. A. Okorochenkov, G. A. Zheltukhina, and V. E. Nebol’sin,“Antimicrobial peptides: the mode of action and perspectivesof practical application,” Biochemistry (Moscow) SupplementSeries B, vol. 5, no. 2, pp. 95–102, 2011.

[65] M. Zasloff, “Antimicrobial peptides of multicellular organ-isms,” Nature, vol. 415, no. 6870, pp. 389–395, 2002.

[66] R. E. W. Hancock, K. L. Brown, and N. Mookherjee, “Host de-fence peptides from invertebrates—emerging antimicrobialstrategies,” Immunobiology, vol. 211, no. 4, pp. 315–322, 2006.


Research Article

The Design and Construction of K11: A Novel α-HelicalAntimicrobial Peptide

Huang Jin-Jiang,1 Lu Jin-Chun,2 Lu Min,2 Huang Qing-Shan,1, 2 and Li Guo-Dong2

1 State Key Laboratory of Genetic Engineering, Institute of Genetics, College of Life Science, Fudan University,Shanghai 200433, China

2 Shanghai Hi-Tech United Bio-Technological Research & Development Co., Ltd, Shanghai 201206, China

Correspondence should be addressed to Huang Qing-Shan, [email protected] and Li Guo-Dong, [email protected]

Received 12 August 2011; Revised 20 October 2011; Accepted 5 December 2011


Copyright © 2012 Huang Jin-Jiang et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Amphipathic α-helical antimicrobial peptides comprise a class of broad-spectrum agents that are used against pathogens. Wedesigned a series of antimicrobial peptides, CP-P (KWKSFIKKLTSKFLHLAKKF) and its derivatives, and determined theirminimum inhibitory concentrations (MICs) against Pseudomonas aeruginosa, their minimum hemolytic concentrations (MHCs)for human erythrocytes, and the Therapeutic Index (MHC/MIC ratio). We selected the derivative peptide K11, which had thehighest therapeutic index (320) among the tested peptides, to determine the MICs against Gram-positive and Gram-negativebacteria and 22 clinical isolates including Acinetobacter baumannii, methicillin-resistant Staphylococcus aureus, Pseudomonasaeruginosa, Staphylococcus epidermidis, and Klebsiella pneumonia. K11 exhibited low MICs (less than 10 μg/mL) and broad-spectrum antimicrobial activity, especially against clinically isolated drug-resistant pathogens. Therefore, these results indicatethat K11 is a promising candidate antimicrobial peptide for further studies.

1. Introduction

The extensive and intensive use of antibiotics has led to theemergence of resistant strains of bacteria. During the lastfew years, only three new types of antibiotics have beendeveloped for clinical use [1]. Antimicrobial peptides havebeen identified from a wide variety of sources, includingbacteria, insects, plants, and animals [2, 3]. Most nativeantimicrobial peptides are effective against a broad spectrumof pathogens but can also be toxic to normal cells [4].

The intensive study of the structure and function ofantimicrobial peptides has led to the development andclinical application of many peptides with enhanced activityand low toxicity; these peptides have been developed throughsequence splicing, amino acid substitution, and changing theratio of hydrophobic amino acids [5–7]. Among them, α-helical antimicrobial peptides are one of the most studiedtypes [8]. CP26, composed of the N-terminal 8 amino acidresidues of cecropin A1 and the N-terminal 18 amino acidresidues of melittin, has high antimicrobial activity and

low toxicity [9]. P18, composed of 8 amino acid residuesof the N-terminus of cecropin A1 and the N-terminal 12amino acid residues of magainin 2, showed satisfactoryantimicrobial activity and no toxicity [10].

In this study, CP26 and P18 were chosen as templatepeptides for the synthesis of CP-P, which is composed ofamino acids 1 to 11 at the N-terminus (KWKSFIKKLTS)of CP26 and amino acids 10 to 18 at the C-terminus(KFLHLAKKF) of P18. CP-P was modified to yield a numberof peptides with single amino acid substitutions at differentsites on the polar or nonpolar faces, and their therapeuticindices (MIC/MHC) were evaluated against Pseudomonasaeruginosa. The in vitro activities of the selected peptideswith high therapeutic indices were evaluated against a setof multiresistant clinical isolates of both Gram-positiveand Gram-negative pathogens. The results indicate thatthe modifications changed the antimicrobial activities andhemolytic toxicities. Among the derivatives, K11 exhibiteda high therapeutic index and may have potential for use infurther studies.



2.1. Bacterial Strains. Twenty-two clinical isolates and nightstandard laboratory strains were used in this study: 4 Acine-tobacter baumannii, 4 Pseudomonas aeruginosa, 3 Staphylo-coccus epidermidis, and 6 Klebsiella pneumonia (ChanghaiHospital, Shanghai, China); 5 methicillin-resistance Staphy-lococcus aureus (Ruijin Hospital, Shanghai, China); Staphy-lococcus aureus CMCC26003; Bacillus subtilis DB430; Bacil-lus pumilus CMCC63202; Micrococcus luteus CMCC28001;E. coli ATCC8099; Klebsiella pneumoniae CMCC46117;Salmonella paratyphi B CMCC50094; Pseudomonas aerug-inosa CMCC10104; Micrococcus S1.634. In addition, Pseu-domonas aeruginosa CMCC10104 was used during the designof the peptides for evaluating antimicrobial activities. Allstrains were cultured on Mueller-Hinton medium.

2.2. Design of Peptides. CP-P was designed by splicing thesequences of CP26 and P18. S16, which has a single aminoacid change at the 16th site of CP-P, served as a templatepeptide from which several peptides were derived by singleamino acid substitution at different sites. All of the peptidesequences are listed in Table 1.

2.3. Synthesis and Purification of Peptides. The syntheses ofthe peptides CP26, P18, CP-P, S16, and the S16 derivatives (atotal of 23) were carried out by solid-phase peptide synthesis(SPPS) [11]. The crude peptides were purified by RP-HPLCusing a Kromasil C18-5 column at a flow rate of 1 mL/minwith a linear AB gradient (1% acetonitrile/min); mobilephase A was 0.1% trifluoroacetic in 100% water, and mobilephase B was 0.1% trifluoroacetic in 100% acetonitrile. Theidentities of the purified peptides were confirmed by massspectrometry.

2.4. Circular Dichroism (CD). CD spectra were obtainedwith a Jasco J-710 instrument (Jasco, Tokyo, Japan) utilizingquartz cells with a 2 mm path length and peptide concentra-tions of 100 μM in a 10 mM sodium phosphate buffer witha pH of 7.5 containing 50% trifluoroethanol. Each spectrumwas obtained from an average of 5 pairs of duplicates, andthe percentage helicity of each peptide was determined [12].

2.5. Measurement of the Antibacterial Activity. The antibac-terial activities of the different peptides were evaluated usingthe minimum inhibitory concentration (MIC). The processfollowed the standard microtiter dilution method in LBmedium without salt. Briefly, bacteria were grown overnightin LB at 37◦C and diluted in the same medium. Then,100 μL of the medium was dispensed into each well of a96-well plate. The number of bacteria was between 104 and105 CFU/mL. Serial dilutions of peptides were performedconsecutively in each well in 10 μL. Plates were incubated at37◦C for 24 h; then, the OD620 was measured to determinethe MIC of each peptide [13].

2.6. Measurement of Hemolytic Activity (MHC). The MHCswere measured by the method as described in Chen et

KWKSFIKKLTSKFLHLAKKF

W

K

K

K

K

K

KK

S

S

IL

T

F

F

F

L

L

H

A

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

181–20

Hydrophilic

Hydrophobic

Figure 1: Helical wheel structure and sequence of peptide CP-P.

al. [7]: peptides were added to 1% human erythrocytesin phosphate-buffered saline (0.08 mol/L NaCl, 0.043 mol/LNa2HPO4, 0.011 mol/L KH2PO4) and then kept at 37◦C for1 h in microtiter plates. The supernatants were collected bycentrifugation (800 g), and the OD562 of each supernatantwas determined. We chose a 0.1% Triton X-100 solution asthe positive control because this solution causes the totalrelease of hemoglobin. We defined the MHC as the highestpeptide concentration that caused no detectable release ofhemoglobin (the ratio of the OD562 for the peptide solutionto the OD562 for the positive control should be less than 1%).

3. Results

3.1. Design and Construction of Peptides. All of the peptideswere designed based on a CP-P template, which had a highertherapeutic index than P18 and P26. S16, a derivative ofCP-P with only one amino acid substitution at the 16thsite, exhibited an 8-fold on the therapeutic index. Thisresult indicates that a polar amino acid substitution onthe nonpolar face could decrease the amphipathicity andinfluence the antimicrobial activity to some extent. Figure 1[14] shows the helical wheel structure of CP-P, whichillustrates the polar and non-polar faces of CP-P clearly.We chose specific sites for single amino acid substitutions,constructed a series of peptides and then calculated thetherapeutic index of all of the derivatives.

3.2. The Purification and Determination of the MolecularConfirmation of the Peptides. After purification, the designed


Table 1: Peptide sequences.

Peptide Sequence MW (Calc)a MW (Deter)b

CP26 KWKSFIKKLTSAAKKVVTTAKPLISS-NH2 2859.7 2860.5

P18 KWKLFKKIPKFLHLAKKF-NH2 2299.4 2300.5

CP-P KWKSFIKKLTSKFLHLAKKF-NH2 2480.1 2479.1

S16 KWKSFIKKLTSKFLHSAKKF-NH2 2453 2452.1

F2 KFKSFIKKLTSKFLHSAKKF-NH2 2415 2414.1

N3 KWNSFIKKLTSKFLHSAKKF-NH2 2439.9 2439.1

K6 KWKSFKKKLTSKFLHSAKKF-NH2 2469 2468.1

N7 KWKSFINKLTSKFLHSAKKF-NH2 2439.9 2438.9

A9 KWKSFIKKAKTSFLHSAKKF-NH2 2411.9 2410.7

K9 KWKSFIKKKTSKFLHSAKKF-NH2 2469 2468.1

S9 KWKSFIKKSTSKFLHSAKKF-NH2 2427.9 2426.7

R9 KWKSFIKKRTSKFLHSAKKF-NH2 2497 2496.1

A10 KWKSFIKKLASKFLHSAKKF-NH2 2424 2424.1

L10 KWKSFIKKLLSKFLHSAKKF-NH2 2466 2465.1

D11 KWKSFIKKLTDKFLHSAKKF-NH2 2482 2481.7

K11 KWKSFIKKLTKKFLHSAKKF-NH2 2495.1 2494.4

L11 KWKSFIKKLTLKFLHSKKKF-NH2 2480.1 2479.2

A13 KWKSFIKKLTSKALHSAKKF-NH2 2377.9 2378.1

K13 KWKSFIKKLTSKKLHSAKKF-NH2 2435 2434.3

K17 KWKSFIKKLTSKFLHSKKKF-NH2 2511.1 2510.2

D18 KWKSFIKKLTSKFLHSADKF-NH2 2440.9 2440.2

N18 KWKSFIKKLTSKFLHSANKF-NH2 2439.9 2440

N20 KWKSFIKKLTSKFLHSAKKN-NH2 2420.9 2419.6aCalculated using BioPerl.

bDetermined using mass spectrometry.

Table 2: The MICs, MHCs, and therapeutic indices of the designed peptides.

Peptide MIC (μg/mL) MHC (μg/mL) Therapeutic index Hydrophobicity Hydrophobic moment %Helix Net charge

P18 25 250 10 0.486 0.405 N 7

CP-P 12.5 250 20 0.411 0.648 63.2 7

S16 3.125 >500 160 0.323 0.596 71.8 7

F2 12.5 250 20 0.3 0.574 N 7

N3 12.5 N N 0.343 0.599 72.5 6

K6 12.5 N N 0.184 0.471 N 8

N7 12.5 >500 40 0.343 0.586 72.1 6

K9 >50 N N 0.189 0.495 51.5 8

S9 50 >500 10 0.236 0.529 68.2 7

R9 >50 N N 0.188 0.494 N 8

A9 25 >500 20 0.254 0.496 64.6 7

L10 12.5 <125 10 0.395 0.632 72.1 7

A10 12.5 250 20 0.326 0.597 72.5 7

D11 25 >500 20 0.287 0.631 71.6 6

K11 1.6 >500 320 0.276 0.642 67.4 8

L11 >50 N N 0.411 0.513 N 7

A13 6.25 >500 80 0.249 0.522 45.9 7

K13 >50 N N 0.184 0.457 62.6 8

K17 12.5 >500 40 0.258 0.551 58.7 8

D18 25 >500 20 0.343 0.587 51.4 5

N18 6.25 >500 80 0.343 0.58 37.2 6

N20 6.25 >500 80 0.204 0.483 N 7

Note: “N” signifies that the data were not determined.


Det. 166

Results

Time Area Area (%) Height Height (%)

24.733 118475 7.74 10654 9.46

25.067 1412627 92.26 101914 90.54

Totals 1531102 100.00 112568 100.00

(a.u

.)

0.25

0.2

0.15

0.1

0 5 10 15 20 25 30 35 40 45 50

(minutes)

Det. 166

Plates/meter (USP)

Figure 2: Analysis of the purity of peptide S16 by HPLC

Table 3: MIC tests on four strains of clinical Acinetobacterbaumannii isolates.

PeptideMIC (μg /mL)

b-01 b-02 b-03 b-04

P18 6.25 25 12.5 12.5

S16 3.125 6.25 6.25 3.125

K11 0.8 3.125 3.25 1.6

peptides were analyzed by RP-HPLC, which showed thattheir purities were greater than 90%. Figure 2 shows thatpeptide S16 reached a purity of 92.3% after purification. TheS16 peak was collected and analyzed by mass spectrometry.Table 1 shows that the mass spectrometry results of thesynthetic peptides matched well with those calculated byBioPerl [15].

3.3. Evaluation of the Designed Peptides. After the mutationof the 16th site to serine (S16), the MIC decreased to 1/4of that of CP-P (Table 2). Table 2 also provides informationabout other peptides that are based on modifications ofS16 by amino acid substitutions. Among these 21 peptides,peptide K11 showed the highest efficacy, with an MICof 1.6 μg/mL, an MHC of more than 500 μg/mL and atherapeutic index of 320.

The hydrophobic moments were determined with theHeliQuest web server [16], which can be used to evaluateamphipathicity. The percent helix values were determinedbased on CD spectra.

3.4. Antimicrobial Activity Against Clinical Acinetobacter Bau-mannii Isolates. After comparison, three designed peptides,S16, K11, and P18, were chosen to use in MIC tests onfour strains of clinically isolated Acinetobacter baumannii.Table 3 illustrates that peptide K11 had the lowest MICs

Table 4: MICs of peptide K11 for different bacteria.

Bacteria MIC (μg/mL)

Gram-positive bacteria

Staphylococcus aureusCMCC26003

0.5

Bacillus subtilisDB430

2

Bacillus pumilusCMCC63202

0.5

Soluble wall MicrococcusS1.634

1

Micrococcus luteusCMCC28001

2

Gram-negative bacteria

E. coliATCC8099

0.5

Klebsiella pneumoniaeCMCC46117

2

Salmonella paratyphi BCMCC50094

2

Pseudomonas aeruginosaCMCC10104

4

Table 5: MICs of K11 for clinically isolated bacteria.

Bacteria MIC (μg/mL)

MRSA (5) 0.25−4

Pseudomonas aeruginosa (4) 1.0−8.0

Staphylococcus epidermidis (3) 0.5−8

Klebsiella pneumonia (6) 0.5−4.0

for the four strains of Acinetobacter baumannii; these MICswere between 0.8 and 3.25 μg/mL. Upon consideration of thisresult, we evaluated the antimicrobial activity of K11 againstclinical isolates of both Gram-positive and Gram-negativepathogens.

3.5. Antimicrobial Activity of Peptide K11 Against DifferentBacteria. Similar to the results above, peptide K11 had alower MIC than the other designed peptides. Table 4 showsthe MICs of K11 against different bacteria, revealing that K11exhibited broad-spectrum antimicrobial activity.

3.6. Antimicrobial Activities of Peptide K11 Against ClinicalIsolates. The results of MIC tests on different clinical bacte-rial isolates can be found in Table 5. The clinically isolatedbacteria included 5 strains of MRSA, 4 strains of Pseu-domonas aeruginosa, 3 strains of Staphylococcus epidermidisand 6 strains of Klebsiella pneumonia. Peptide K11 exhibiteda low MIC for almost all of the strains of bacteria, whichindicated that this peptide has a high antimicrobial activityagainst several clinically isolated drug-resistant bacteria.

4. Discussion

To design novel antimicrobial peptides with enhanced bio-logical properties, we searched native antimicrobial peptides


and selected CP26 and P18 as frameworks. CP26 andP18 have been reported to be two α-helical peptides withhigh antimicrobial activities. Starting with CP26 and P18,we designed and constructed a dozen novel peptides bysequence splicing and amino acid substitution. Several ofthese peptides had lower MICs and higher therapeuticindices than CP26 and P18. Among them, K11 showed themaximum antimicrobial activity and minimum eukaryoticcell toxicity, with an MIC 1/16 of that of P18 and a 32-foldhigher therapeutic index than P18.

Recent studies on α-helical antimicrobial peptides haverevealed that the amphipathic structure plays a primaryrole in the antimicrobial activity of these compounds.The reduction of antimicrobial activity resulting from theenhanced ability of self-aggregation caused by increasingamphipathicity has been reported [17]. The mechanismresponsible for this effect may be the reduction of theeffective molecular number after self-aggregation. In ourstudy, we used HeliQuest to calculate the hydrophobicmoment of peptides to evaluate the amphipathicity andcame to the conclusion that high amphipathicity reducedthe therapeutic index. S16 has a lower hydrophobic momentthan CP-P (0.596 to 0.648) and was found to have a highertherapeutic index (160 to 20). Peptides L10 and D11 havehigh hydrophobic moments (0.632 and 0.631, resp.) andexhibited low therapeutic indices (<20). The MICs increasedfrom 3.125 μg/mL for S16 to 12.5 μg/mL and 25 μg/mL,and the MHCs decreased from >500 μg/mL for S16 to<125 μg/mL. These results confirm the conclusion above,but high hydrophobic moments may also lead to increasedantibacterial activities and decreased hemolytic activities.The results agree with the results of other studies that there isa threshold hydrophobicity at which optimal antimicrobialactivity can be achieved [18]. In this study, K11 reacheda therapeutic index of 320 and has a high hydrophobicmoment (0.642).

We found that substitutions at certain specific sites (9, 11,and 13 from the N-terminus) can change the antimicrobialactivity of S16 more effectively. The 9th residue of S16 is L,which is on the nonpolar face. If a polar amino acid weresubstituted at this site, there could be an increase in the MIC.R9, S9, and K9 are three derivatives that showed high MICs(>50 μg/mL) and decreased hydrophobic moments (0.494,0.529, and 0.495, resp.). The substitutions at the 11th and13th sites with opposite polar amino acids can also leadto significant increases in the MIC and decreases in theamphipathicity. We think that this structural implicationshould be investigated more closely in future studies.

The net positive charge on the polar face is important forthe antimicrobial and hemolytic activities of antimicrobialpeptides [19]. We constructed K11 by changing S to K atposition 11 on the polar face of S16. This one addition tothe net charge resulted in K11 showing the best biologicalproperties among the tested peptides.

In conclusion, our results showed that the therapeuticindex of a peptide depends on several factors. Modificationof specific sites can change the amphipathicity to differentextents, which can be used to predict the antimicrobialactivity. In our study, we found a highly effective peptide,

S16, and used it as template to establish a series of peptideswith single amino acid substitutions. Among these peptides,K11 showed strong therapeutic action against antibiotic-resistant clinical isolates of both Gram-positive and Gram-negative bacteria, making it a promising antimicrobialpeptide candidate for further study, especially in vivo studiesand clinical tests.

Acknowledgments

This work was supported by the Yangtze River Delta JointScientific and Technological Project (no. 10495810600) andthe special fund of the Jiangsu Province for technologicalcommercialization (no. BA2010089). The authors thankProfessor S. U. Deming for reading this paper and W. U.Hong-yu for technical assistance.

References

[1] A. K. Marr, W. J. Gooderham, and R. E. Hancock, “Antibac-terial peptides for therapeutic use: obstacles and realisticoutlook,” Current Opinion in Pharmacology, vol. 6, no. 5, pp.468–472, 2006.

[2] T. Ganz and R. I. Lehrer, “Antimicrobial peptides of verte-brates,” Current Opinion in Immunology, vol. 10, no. 1, pp. 41–44, 1998.

[3] M. Zasloff, “Antimicrobial peptides of multicellular organ-isms,” Nature, vol. 415, no. 6870, pp. 389–395, 2002.

[4] I. Ginsburg and E. Koren, “Are cationic antimicrobial peptidesalso ‘double-edged swords’,” Expert Review of Anti-InfectiveTherapy, vol. 6, no. 4, pp. 453–462, 2008.

[5] S. Y. Shin, S. H. Lee, S. T. Yang et al., “Antibacterial, antitumorand hemolytic activities of α-helical antibiotic peptide, P18and its analogs,” Journal of Peptide Research, vol. 58, no. 6, pp.504–514, 2001.

[6] H. Sato and J. B. Feix, “Lysine-enriched cecropin-mellitinantimicrobial peptides with enhanced selectivity,” Antimicro-bial Agents and Chemotherapy, vol. 52, no. 12, pp. 4463–4465,2008.

[7] Y. Chen, C. T. Mant, S. W. Farmer, R. E. W. Hancock,M. L. Vasil, and R. S. Hodges, “Rational design of α-helical antimicrobial peptides with enhanced activities andspecificity/therapeutic index,” Journal of Biological Chemistry,vol. 280, no. 13, pp. 12316–12329, 2005.

[8] A. Giangaspero, L. Sandri, and A. Tossi, “Amphipathic α heli-cal antimicrobial peptides,” European Journal of Biochemistry,vol. 268, no. 21, pp. 5589–5600, 2001.

[9] M. G. Scott, H. Yan, and R. E. W. Hancock, “Biological prop-erties of structurally related α-helical cationic antimicrobialpeptides,” Infection and Immunity, vol. 67, no. 4, pp. 2005–2009, 1999.

[10] S. Y. Shin, M. K. Lee, K. L. Kim, and K. S. Hahm, “Structure-antitumor and hemolytic activity relationships of syntheticpeptides derived from cecropin A-magainin 2 and cecropin A-melittin hybrid peptides,” Journal of Peptide Research, vol. 50,no. 4, pp. 279–285, 1997.

[11] J. D. Fontenot, J. M. Ball, M. A. Miller, C. M. David, and R.C. Montelaro, “A survey of potential problems and qualitycontrol in peptide synthesis by the fluorenylmethoxycarbonylprocedure,” Peptide Research, vol. 4, no. 1, pp. 19–25, 1991.


[12] U. Pag, M. Oedenkoven, V. Sass et al., “Analysis of in vitroactivities and modes of action of synthetic antimicrobial pep-tides derived from an α-helical ‘sequence template’,” Journal ofAntimicrobial Chemotherapy, vol. 61, no. 2, pp. 341–352, 2008.

[13] I. Y. Park, C. B. Park, M. S. Kim, and S. C. Kim, “Parasin I, anantimicrobial peptide derived from histone H2A in the catfish,Parasilurus asotus,” FEBS Letters, vol. 437, no. 3, pp. 258–262,1998.

[14] G. Deleage, C. Combet, C. Blanchet, and C. Geourjon,“ANTHEPROT: an integrated protein sequence analysis soft-ware with client/server capabilities,” Computers in Biology andMedicine, vol. 31, no. 4, pp. 259–267, 2001.

[15] J. E. Stajich, D. Block, K. Boulez et al., “The Bioperl toolkit:Perl modules for the life sciences,” Genome Research, vol. 12,no. 10, pp. 1611–1618, 2002.

[16] R. Gautier, D. Douguet, B. Antonny, and G. Drin,“HELIQUEST: a web server to screen sequences withspecific α-helical properties,” Bioinformatics, vol. 24, no. 18,pp. 2101–2102, 2008.

[17] Y. Chen, M. T. Guarnieri, A. I. Vasil, M. L. Vasil, C. T.Mant, and R. S. Hodges, “Role of peptide hydrophobicity inthe mechanism of action of α-helical antimicrobial peptides,”Antimicrobial Agents and Chemotherapy, vol. 51, no. 4, pp.1398–1406, 2007.

[18] Z. Jiang, A. I. Vasil, L. Gera, M. L. Vasil, and R. S. Hodges,“Rational design of α-helical antimicrobial peptides to tar-get Gram-negative pathogens, Acinetobacter baumannii andPseudomonas aeruginosa: utilization of charge, “specificitydeterminants,” total hydrophobicity, hydrophobe type andlocation as design parameters to improve the therapeuticratio,” Chemical Biology and Drug Design, vol. 77, no. 4, pp.225–240, 2011.

[19] Z. Jiang, A. I. Vasil, J. D. Hale, R. E. W. Hancock, M. L. Vasil,and R. S. Hodges, “Effects of net charge and the number ofpositively charged residues on the biological activity of amphi-pathic α-helical cationic antimicrobial peptides,” Biopolymers,vol. 90, no. 3, pp. 369–383, 2008.


Research Article

The Lantibiotic Lacticin 3147 Prevents Systemic Spread ofStaphylococcus aureus in a Murine Infection Model

Clare Piper,1 Pat G. Casey,1, 2 Colin Hill,1, 2 Paul D. Cotter,2, 3 and R. Paul Ross2, 3

1 Department of Microbiology, University College Cork, College Road, Cork, Ireland2 Alimentary Pharmabiotic Centre, Cork, Ireland3 Food Biosciences Department, Teagasc Food Research Centre, Moorepark, Fermoy, Cork, Ireland

Correspondence should be addressed to Colin Hill, [email protected] and Paul D. Cotter, [email protected]

Received 12 August 2011; Accepted 3 October 2011

Academic Editor: John Tagg

Copyright © 2012 Clare Piper et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The objective of this study was to investigate the in vivo activity of the lantibiotic lacticin 3147 against the luminescentStaphylococcus aureus strain Xen 29 using a murine model. Female BALB/c mice (7 weeks old, 17 g) were divided into groups(n = 5) and infected with the Xen 29 strain via the intraperitoneal route at a dose of 1 × 106 cfu/animal. After 1.5 hr, the animalswere treated subcutaneously with doses of phosphate-buffered saline (PBS; negative control) or lacticin 3147. Luminescent imagingwas carried 3 and 5 hours postinfection. Mice were then sacrificed, and the levels of S. aureus Xen 29 in the liver, spleen, and kidneyswere quantified. Notably, photoluminescence and culture-based analysis both revealed that lacticin 3147 successfully controlledthe systemic spread of S. aureus in mice thus indicating that lacticin 3147 has potential as a chemotherapeutic agent for in vivoapplications.

1. Introduction

Staphylococcus aureus is one of the most significant bacterialpathogens and can cause diseases ranging from minor andsurgical site infections [1] to potentially life-threateningendocarditis [2–4] and bacteraemia [5–8]. It is a particularproblem in hospitals as a consequence of the emergenceand dissemination of multidrug-resistant forms such asmethicillin-resistant S. aureus (MRSA), vancomycin inter-mediate susceptibility S. aureus (VISA), and heterogenousVISA (hVISA). The prevalence of these antibiotic resistantforms means that the discovery of novel chemotherapeuticagents to combat these pathogens is of key importance[9, 10]. The lantibiotics (lanthionine-containing antibiotics[11]) are a group of posttranslationally modified antimicro-bial peptides of which nisin and lacticin 3147 are among themost extensively investigated. A number of lantibiotics havebeen noted to exhibit potent antimicrobial activity againststaphylococci of clinical relevance. In agar diffusion assays,the type I lantibiotics epidermin, Pep5, epicidin K7, andepilancin 280 display impressive levels of activity againstcoagulase negative staphylococci (CNS) [12], and it has

been suggested that their potential could be exploited toprevent the colonization of medical devices [12]. Nisin hasalso been shown on several occasions to possess significantanti-Staphylococcus activity. When tested against 20 MRSAstrains, one study revealed that the minimum inhibitoryconcentration (MIC) of nisin A ranged between 1.5 and16 mg/L [13], while a more recent investigation revealedMICs of 0.5–4.1 mg/L [14]. The in vitro activity of otherforms of nisin (nisin F, Q, and Z) against MRSA has alsorecently been highlighted [15]. The in vivo activity of anumber of lantibiotics against staphylococci has also beeninvestigated. The effectiveness of the epidermin-like mutacinB-Ny266 was tested on mice infected by intraperitoneal (IP)injection with 3.1× 107 cfu of S. aureus Smith/mouse. Imme-diately after injection, mutacin B-Ny266 was administered,also via the IP route, at concentrations of 1–10 mg/kg ofmouse and was found to be protective [16]. More recently,it has been established that microbisporicin, in additionto having potent in vitro activity (MIC ≤ 0.13μg/mL),effectively controls murine septicemia caused by S. aureusin female CD-1 mice (23–25 g). The mice were infected viathe IP route with 1 × 106 cfu of S. aureus Smith 819 ATCC


19636 in 0.5 mls gastric hog mucin. Microbisporicin wasthen administered intravenously or subcutaneously (SC) 10–15 mins after infection at final concentrations of 10–15 mg/L[17]. The effective dose 50 (ED50) of microbisporicinwas found to be 2.1 mg/kg regardless of whether it wasadministered via IV or SC. ED50 values were determinedon the bases of survival of the mice to the seventh day.Higher doses of microbisporicin (≥200 mg/kg) led to thesurvival of all animals treated and were nontoxic [16]. NisinF effectively controlled the MRSA strain, S. aureus K, inimmunocompromised Wistar rats following the introduc-tion of 4 × 105 S. aureus cells into the nostrils of the ratsfor 4 consecutive days before treating with 8192 arbitraryunits (AU) of nisin F intranasally for the subsequent 4 days[18]. In contrast, however, when 1 x108 S. aureus Xen 36cells were injected intraperitoneally, the administration ofa lower concentration of nisin F (640 AU) after 4 hourssucceeded in inhibiting the growth of the pathogen for only15 minutes after which time the pathogen reemerged [19].Finally, short- and long-term in vivo studies with mersacidinestablished that this lantibiotic quite effectively inhibitedMRSA introduced intranasally into immunocompromised(hydrocortisone-treated) BALB/C mice [20]. For the shortterm trial, the mice were infected on days 5, 7, and 9 with3 × 102–104 cfu of the S. aureus strain. The mice werethen treated intranasally with mersacidin (1.66 mg/kg pertreatment) twice a day on days 10, 11, and 12. For the longertrial, the mice were challenged with S. aureus on days 5, 7, 9,30, 32, and 34 and subsequently treated with mersacidin ondays 35, 36, and 37. In both cases the mersacidin treatmentsuccessfully inhibited MRSA-induced rhinitis [20]. Notably,a comparison of the in vitro and in vivo activity of mersacidinagainst a number of MRSA strains indicates that mersacidinmore effectively inhibits S. aureus in vivo [21].

Lacticin 3147 is the most extensively investigated ofthe two peptide lantibiotics. These peptides are activeas a consequence of the synergistic activity of twolanthionine-containing peptides [22, 23]. Lacticin 3147 hasbeen found to exhibit potent in vitro activity against arange of pathogenic bacteria including Clostridium difficile,vancomycin-resistant enterococci, Propionibacterium acne,penicillin-resistant Pneumococcus, and Streptococcus mutans[14, 24–26] as well as pathogenic mycobacteria such asMycobacterium avium subsp paratuberculosis and Mycobac-terium tuberculosis H37Ra [27]. Of greatest relevance tothis study is the fact that lacticin 3147 possesses anti-Staphylococcus activity. The lantibiotic itself, when incorpo-rated into a teat seal, protects against S. aureus-associatedbovine mastitis [28, 29], while use of a lacticin 3147-producing Lactococcus lactis DPC 3251 within a teat dipinhibits S. aureus both in vitro and also in vivo [30]. The invitro activity of lacticin 3147 against clinical MRSA isolateshas also been established with MICs ranging from 1.9 to15.4 mg/L [14].

Despite lacticin 3147 being one of the most extensivelystudied lantibiotics, its ability to control a systemic infectioncaused by S. aureus, or indeed any other pathogen, hasnot been investigated. Here we address this issue usingBALB/c mice infected via the IP route with S. aureus Xen

29, a strain of methicillin sensitive S. aureus (MSSA) thathas been genetically modified to express the Photorhabdusluminescens lux genes to facilitate in vivo imaging. The abilityof subcutaneously administered lacticin 3147 to controlinfection was assessed by in vivo imaging and microbiologicalanalysis of the organs of sacrificed animals.


2.1. Antimicrobial Activity Assays. The in vitro activity oflacticin 3147 and vancomycin (employed as a positivecontrol) against S. aureus Xen 29 was assessed through MICdetermination assays carried out in triplicate as describedpreviously [14] with purified lacticin 3147, prepared viaHPLC, again as described previously [14]. Vancomycin wasobtained from Sigma Aldrich.

2.2. Inoculum Preparation. S. aureus Xen 29 (derived fromthe parental pleural isolate S. aureus 12600; Xenogen Cor-poration, Almeda, CA) possesses a copy of the modifiedluxABCDE operon of P. luminescens integrated at a single siteon the chromosome. S. aureus Xen 29 was cultured overnightin brain heart infusion (BHI) broth aerobically at 37◦Cfrom an isolated colony growing on BHI agar containing200 μg/mL kanamycin. On the day of the trial, the overnightculture was subcultured (1 : 100 dilution) into fresh BHI andgrown to log phase (OD600nm of 0.5). This culture was dilutedto facilitate the ultimate administration of the culture in theform of a 1 × 106 cfu/100 μL dose in 0.5% hog gastric mucin(Sigma Aldrich).

2.3. Mouse Peritonitis Model. Mice were fed a standardrodent diet ad libitum and all animal studies were approvedby the Animal Experimentation Ethics Committee. 13BALB/c female mice (7 weeks old, 15 g ± 2 g in weight) weredivided into 3 groups (A, B, C; n = 3, 5, and 5, resp.). At T0

mice in groups A–C received the 1 × 106 cfu dose (100 μLvolume) via the IP route in 0.5% gastic hog mucin (SigmaAldrich). At T1.5 hrs, the mice in group C were administeredlacticin 3147 (50.85 mg/kg of Ltnα and 43.8 mg/kg of Ltnβ,corresponding to 30.76 mM lacticin 3147/kg) in a singledose and a second dose at T3hrs (25.425 mg/kg Ltnα and21.90 mg/kg Ltnβ; 15.382 mM lacticin 3147/kg). Vancomycin(50 mg/kg; 33.6 mM/kg) was administered at T1.5 hrs and atT3hrs to the mice in group B while the mice in group Areceived PBS (once) as a control. Both antimicrobials andPBS were adminisitered subcutaneously in 100 μL doses.In vivo imaging was carried out at two time points thatis, 3 hours and 5 hours postinfection. Mice were anaes-thetized for bioluminescent imaging via the inhalation ofaerosolized isoflurane mixed with oxygen. The mice werethen transferred to the IVIS chamber ventral side up, andluminescence was measured over a 3-to-5 mins exposuretime. The imaging system measures the number of photonsreaching each detector of the charged-couple device camera,and the IVIS software translates these data into false colorimages that display regions of intense luminescence withred, moderate luminescence in yellow and green and mild


luminescence in blue. The images contained herein arephotographic images with an overlay of bioluminescencethat uses this computer-generated color scale [31]. Themice were euthanized approximately 6 hours postinfection.Liver, kidneys, and spleen were extracted. These organs weremechanically disrupted and serial dilutions made which weresubsequently plated in 100 μL volumes on TSA−Kan200μg/mL

plates in order to enumerate the staphylococci present ineach organ.

2.4. Quantification of Luminescence. Luminescent imageswere quantified with IVIS imaging software. The total flux(number of photons/s/cm2) was calculated by a user definedarea (region of interest) covering the infection site. The fluxwas averaged across all mice from each respective group. Thereduction in luminescence was quantified and represents acomparison with the luminescence from mice administeredphosphate-buffered saline control at the same time point.

2.5. Statistical Analysis. The mean and standard error of themean (SEM) of the luminescence at the final time pointand bacterial counts for the mice were calculated for allgroups. Differences in the bioluminescence and bacterialcounts analyzed through a one-way analysis of variance,followed by the Holm-Sidak posttest (Sigma Stat, version3.5).

3. Results/Discussion

3.1. Assessment of the In Vivo Activity of Lacticin 3147 againstS. aureus Xen 29 Using a Murine Peritonitis Model. Theability of subcutaneously injected lacticin 3147 to control asystemic S. aureus infection following the introduction of thepathogen into the murine peritoneal cavity was investigated.This involved in vivo imaging to detect levels of light emittedby the pathogen within mice and through the postmortemmicrobiological analysis of organs. Negative and positivecontrols were employed in the form of mice treated with PBSand the glycopeptide antibiotic vancomycin, respectively.The target strain S. aureus Xen 29 is a methicillin sensitiveisolate which has been employed previously to facilitate aninvestigation of acute in vivo infections [32–36]. Prior tocommencement of the study, the in vitro sensitivity of Xen29 to lacticin 3147 was assessed. The corresponding MICvalues were 1.013 mg/L and 19.1 mg/L for vancomycin andlacticin 3147, respectively (Table 1). For in vivo studies, micereceived a dose of 1 × 106 cfu and, 1.5 hrs postinfection,were administered lacticin 3147 (50.85 mg/kg of Ltnα and43.8 mg/kg of Ltnβ), vancomycin (50 mg/kg), or PBS. AtT3hrs, the mice were subject to IVIS imaging, and seconddoses of lacticin 3147 (25.425 mg/kg and 21.90 mg/kg) andvancomycin (50 mg/kg) were administered to the relevantmice. IVIS analysis of the progression of the S. aureus Xen 29infection showed that the pathogen spreads systemically andeventually also occupies the thoracic cavity in mice injectedwith PBS 5 hrs (T5hrs) after injection of the pathogen. A sig-nificant (P = 0.000116) reduction in the RLU measurementscorresponding to the thoracic region of the lacticin 3147

Table 1: The standard deviation in all cases is 0 reflecting identicaltriplicate results.

S. aureus Xen 29 MIC (mg/L)

Vancomycin 1.013

Lacticin 3147 19.1

treated group was evident when compared to that of thePBS (negative) control group (Figure 1) at this time pointhighlighting the ability of the lantibiotic to prevent systemicspread of the S. aureus Xen 29 infection. In contrast, lacticin3147 does not significantly reduce RLU values correspondingto the peritoneal cavity relative to the control. It may bethat lacticin 3147 is deficient in penetrating the peritonealcavity (Figure 1). To further ascertain lacticin 3147 efficacy,culture-based analysis of staphylococcal levels in the organswas determined after the mice were sacrificed. This analysisfurther highlighted the success of lacticin 3147 in controllingsystemic infection. Lacticin 3147 treatment resulted in asignificant reduction (P < 0.05 in all cases; Figure 1) inpathogen numbers in the liver, spleen, and kidneys of themice treated relative to the PBS-treated controls (Figure 1).

As expected, vancomycin brought about a significantreduction in S. aureus levels relative to the PBS-treated con-trols as determined by both bioimaging and culture-basedanalysis. Notably, numbers of S. aureus in the spleens oflacticin 3147- and vancomycin-treated mice were statisticallyindifferent. However, vancomycin treatment more success-fully lowered S. aureus numbers in the liver and kidneys.While both lacticin 3147 and vancomycin bind lipid II,[10, 37, 38] differences exist with respect to their mechanismof action. Vancomycin binds to the C-terminal D-Ala-D-Ala motif of the pentapeptide of lipid II [10] whereas, onthe basis of similarities between Ltnα and mersacidin, it isproposed that lacticin 3147 binds to the sugar phosphatehead group of lipid II [39]. Furthermore, lacticin 3147 is alsocapable of forming pores in the membranes of target cells[37, 38]. It should be noted that while similar mg/kg doses oflacticin 3147 and vancomycin were employed in this study,our in vitro investigations established that vancomycin is 19times more potent than lacticin 3147 against Xen29 (MICvalues; 1.013 mg/L and 19.1 mg/L of vancomycin and lacticin3147, resp.). Thus the dose of vancomycin administeredin vivo corresponded to 100-fold that of the in vitro MICwhereas lacticin 3147 was administered at a level 8-foldgreater than its in vitro MIC. This may explain the enhancedability of vancomycin with respect to clearance of Xen 29from the peritoneal cavity. This is the first occasion uponwhich the impact of lacticin 3147 against a systemic infectionhas been assessed and thus it is also the first instance ofits administration subcutaneously. It may be that lacticin3147 cannot travel to the peritoneum to eradicate theinfection but can prevent the spread of infection throughoutthe blood stream. As stated previously, mersacidin hassuccessfully been shown to inhibit a systemic MRSA infectionin mice when administered via the subcutaneous route [20].However, mersacidin is a one-component lantibiotic and isalso globular which may provide facile delivery through the


0.126

0.00141

0.000116Spleen

Liver

Kidney

0.00432

0.000451

Vancomycin

Thoracic cavity

RLU measurements

0.000201

0.0000742

0.000000115

Spleen

Liver

Kidney

0.00000370

0.000339

Lacticin 3147

Thoracic cavity

Abdominal cavity

Abdominal cavity

Abdominal cavity

RLU measurements

p/s/cm2/sr

p/s/cm2/sr

p/s/cm2/sr

P values

P values P values

P values

P values P values

Control

Thoracic cavity

cfu/mLof organ

cfu/mLof organ

cfu/mLof organ

RLU measurements

Spleen

Liver

Kidney

N/A N/A

N/A

N/A

N/A

1.73E+03

2.14E+04

3.9E+05

9.4E+05

1.71E+04

1.53E+03

1.27E+032.38E+05

2.35E+05

1.38E+04

5.31E+06

4.6E+06

1.39E+06

9.81E+03

1000

800

400

600

200

2.96E+04

Min = 200Max= 1000

T3hrs T5hrs

Figure 1: Impact of lacticin 3147 on the systemic spread of S. aureus Xen 29 in mice. Images are of representative mice from each group 3and 5 hours postinfection. Values (i.e., RLU measurements and cfu/mL of organs) represent averages of data collected at T5hrs, and P valuesrefer to the significance of differences between the treated and untreated equivalents at T5hrs as determined by one-way analysis of variance,followed by the Holm-Sidak posttest. The imaging system depicts false-color images representative of different levels of total flux. False colorimaging represents intense luminescence in red, moderate luminescence in green, and low level luminescence in blue/purple.

skin. It, like vancomycin, is quite a small peptide with amolecular weight of 1, 825 Da [40]. Lacticin 3147 consistsof 2 peptides with molecular weights of 3305 Da (Ltnα)and 2847 Da (Ltnβ), and it may be that the larger sizeof the individual peptides or a specific difficulty relatingthe transport of one of the components to the peritonealcavity may be an issue. Mutacin B-Ny266 has also beenshown to protect against S. aureus in the peritoneum, butthis lantibiotic was administered intraperitoneally, and thustransfer to the site of infection was not an issue [16].

3.2. Conclusion. In conclusion, here we have providedevidence that lacticin 3147 could be employed to treatsystemic infections. Both culture- and bioluminescence-based analyses reveal that the lantibiotic significantly reducesnumbers of the S. aureus Xen29 relative to the negativecontrol by preventing the dissemination of the pathogen.Although these results are more promising than thosedescribed when nisin F was employed in a similar manner(19), differences with respect to the strains of S. aureusemployed, concentrations of lantibiotic, and other factorsmean that a direct comparison of outcomes is not possible.While further investigations are required, over longer periods

of time, to more extensively assess the clinical potentialof lacticin 3147, it is worth noting that lacticin 3147possesses many physicochemical properties that favour itsin vivo application. These include excellent activity overa broad pH range, especially at physiological pH (pH 7),the absence of cytotoxicity towards eukaryotic cells [41],its broad spectrum of activity at nanomolar concentrations[42], its alternative mode of action [38], and the presenceof (methyl)lanthionine bridges that confer structural rigidityto lantibiotics and reduce proteolytic attack [43]. Theseproperties, accompanied by its ability to inhibit a systemic S.aureus infection, make lacticin 3147 a promising candidatefor potential applications in human medicine.

Acknowledgment

The authors are grateful to Ian Monk for helpful discussions.This research was funded by a Health Research Board (HRB)Ireland Research Project Grant and by the Irish Governmentunder the National Development Plan, through a ScienceFoundation Ireland Investigator Award to C. Hill, R. PaulRoss, and P. Cotter (06/IN.1/B98).


References

[1] R. P. Rennie, R. N. Jones, and A. H. Mutnick, “Occurrenceand antimicrobial susceptibility patterns of pathogens iso-lated from skin and soft tissue infections: report from theSENTRY Antimicrobial Surveillance Program (United Statesand Canada, 2000),” Diagnostic Microbiology and InfectiousDisease, vol. 45, no. 4, pp. 287–293, 2003.

[2] A. Tugcu, O. Yildirimturk, C. Baytaroglu et al., “Clinicalspectrum, presentation, and risk factors for mortality ininfective endocarditis: a review of 68 cases at a tertiary carecenter in Turkey,” Turk Kardiyoloji Dernegi Arsivi, vol. 37, no.1, pp. 9–18, 2009.

[3] A. Nomura, F. Omata, and K. Furukawa, “Risk factors of mid-term mortality of patients with infective endocarditis,” Euro-pean Journal of Clinical Microbiology and Infectious Diseases,vol. 29, no. 11, pp. 1355–1360, 2010.

[4] A. M. Valente, R. Jain, M. Scheurer et al., “Frequencyof infective endocarditis among infants and children withStaphylococcus aureus bacteremia,” Pediatrics, vol. 115, no. 1,pp. e15–e19, 2005.

[5] J. M. Boyce, B. Cookson, K. Christiansen et al., “Meticillin-resistant Staphylococcus aureus,” Lancet Infectious Diseases, vol.5, no. 10, pp. 653–663, 2005.

[6] S. E. Cosgrove, G. Sakoulas, E. N. Perencevich, M. J. Schwaber,A. W. Karchmer, and Y. Carmeli, “Comparison of mor-tality associated with methicillin-resistant and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-analysis,”Clinical Infectious Diseases, vol. 36, no. 1, pp. 53–59, 2003.

[7] D. Pittet, D. Tarara, and R. P. Wenzel, “Nosocomial blood-stream infection in critically ill patients: excess length of stay,extra costs, and attributable mortality,” Journal of the AmericanMedical Association, vol. 271, no. 20, pp. 1598–1601, 1994.

[8] M. J. Richards, J. R. Edwards, D. H. Culver, and R. P.Gaynes, “Nosocomial infections in combined medical-surgicalintensive care units in the United States,” Infection Control andHospital Epidemiology, vol. 21, no. 8, pp. 510–515, 2000.

[9] L. Cui, A. Iwamoto, J. Q. Lian et al., “Novel mechanism ofantibiotic resistance originating in vancomycin-intermediateStaphylococcus aureus,” Antimicrobial Agents and Chemother-apy, vol. 50, no. 2, pp. 428–438, 2006.

[10] K. Hiramatsu, “Vancomycin resistance in staphylococci,” DrugResistance Updates, vol. 1, no. 2, pp. 135–150, 1998.

[11] N. Schnell, K. D. Entian, U. Schneider et al., “Prepeptidesequence of epidermin, a ribosomally synthesized antibioticwith four sulphide-rings,” Nature, vol. 333, no. 6170, pp. 276–278, 1988.

[12] M. B. C. Fontana, M. D. C. Freire De Bastos, and A. Brandelli,“Bacteriocins Pep5 and epidermin inhibit Staphylococcusepidermidis adhesion to catheters,” Current Microbiology, vol.52, no. 5, pp. 350–353, 2006.

[13] W. Brumfitt, M. R. J. Salton, and J. M. T. Hamilton-Miller,“Nisin, alone and combined with peptidoglycan-modulatingantibiotics: activity against methicillin-resistant Staphylococ-cus aureus and vancomycin-resistant enterococci,” Journal ofAntimicrobial Chemotherapy, vol. 50, no. 5, pp. 731–734, 2002.

[14] C. Piper, L. A. Draper, P. D. Cotter, R. P. Ross, and C. Hill, “Acomparison of the activities of lacticin 3147 and nisin againstdrug-resistant Staphylococcus aureus and Enterococcus species,”Journal of Antimicrobial Chemotherapy, vol. 64, no. 3, pp. 546–551, 2009.

[15] C. Piper, C. Hill, P. D. Cotter, and R. P. Ross, “Bioengineeringof a Nisin A-producing Lactococcus lactis to create isogenic

strains producing the natural variants Nisin F, Q and Z,”Microbial Biotechnology, vol. 4, no. 3, pp. 375–382, 2011.

[16] M. Mota-Meira, H. Morency, and M. C. Lavoie, “In vivo activ-ity of mutacin B-Ny266,” Journal of Antimicrobial Chemother-apy, vol. 56, no. 5, pp. 869–871, 2005.

[17] F. Castiglione, A. Lazzarini, L. Carrano et al., “Determiningthe structure and mode of action of microbisporicin, a potentlantibiotic active against multiresistant pathogens,” Chemistryand Biology, vol. 15, no. 1, pp. 22–31, 2008.

[18] M. De Kwaadsteniet, K. T. Doeschate, and L. M. T. Dicks,“Nisin F in the treatment of respiratory tract infections causedby Staphylococcus aureus,” Letters in Applied Microbiology, vol.48, no. 1, pp. 65–70, 2009.

[19] A. M. Brand, M. de Kwaadsteniet, and L. M. T. Dicks, “Theability of nisin F to control Staphylococcus aureus infection inthe peritoneal cavity, as studied in mice,” Letters in AppliedMicrobiology, vol. 51, no. 6, pp. 645–649, 2010.

[20] D. Kruszewska, H. G. Sahl, G. Bierbaum, U. Pag, S. O. Hynes,and A. Ljungh, “Mersacidin eradicates methicillin-resistantStaphylococcus aureus (MRSA) in a mouse rhinitis model,”Journal of Antimicrobial Chemotherapy, vol. 54, no. 3, pp. 648–653, 2004.

[21] S. Chatterjee, D. K. Chatterjee, R. H. Jani et al., “Mersacidin, anew antibiotic from bacillus in vitro and in vivo antibacterialactivity,” Journal of Antibiotics, vol. 45, no. 6, pp. 839–845,1992.

[22] E. M. Lawton, R. P. Ross, C. Hill, and P. D. Cotter, “Two-peptide lantibiotics: a medical perspective,” Mini-Reviews inMedicinal Chemistry, vol. 7, no. 12, pp. 1236–1247, 2007.

[23] S. Suda, P. D. Cotter, C. Hill, and R. P. Ross, “Lacticin 3147—biosynthesis, molecular analysis, immunity, bioengineeringand applications,” Current Protein and Peptide Science. Inpress.

[24] M. C. Rea, E. Clayton, P. M. O’Connor et al., “Antimicrobialactivity of lacticin 3147 against clinical Clostridium difficilestrains,” Journal of Medical Microbiology, vol. 56, no. 7, pp.940–946, 2007.

[25] M. Galvin, C. Hill, and R. P. Ross, “Lacticin 3147 displaysactivity in buffer against Gram-positive bacterial pathogenswhich appear insensitive in standard plate assays,” Letters inApplied Microbiology, vol. 28, no. 5, pp. 355–358, 1999.

[26] M. C. Rea, A. Dobson, O. O’Sullivan et al., “Effect of broad-and narrow-spectrum antimicrobials on Clostridium difficileand microbial diversity in a model of the distal colon,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 108, supplement 1, pp. 4639–4644, 2011.

[27] J. Carroll, L. A. Draper, P. M. O’Connor et al., “Comparison ofthe activities of the lantibiotics nisin and lacticin 3147 againstclinically significant mycobacteria,” International Journal ofAntimicrobial Agents, vol. 36, no. 2, pp. 132–136, 2010.

[28] D. P. Twomey, A. I. Wheelock, J. Flynn, W. J. Meaney,C. Hill, and R. P. Ross, “Protection against Staphylococcusaureus mastitis in dairy cows using a bismuth-based teat sealcontaining the bacteriocin, Lacticin 3147,” Journal of DairyScience, vol. 83, no. 9, pp. 1981–1988, 2000.

[29] M. P. Ryan, W. J. Meaney, R. P. Ross, and C. Hill, “Evaluationof lacticin 3147 and a teat seal containing this bacteriocin forinhibition of mastitis pathogens,” Applied and EnvironmentalMicrobiology, vol. 64, no. 6, pp. 2287–2290, 1998.

[30] K. Klostermann, F. Crispie, J. Flynn, W. J. Meaney, R. PaulRoss, and C. Hill, “Efficacy of a teat dip containing the bac-teriocin lacticin 3147 to eliminate Gram-positive pathogensassociated with bovine mastitis,” Journal of Dairy Research, vol.77, no. 2, pp. 231–238, 2010.


[31] L. I. Mortin, T. Li, A. D. G. Van Praagh, S. Zhang, X.X. Zhang, and J. D. Alder, “Rapid bactericidal activityof daptomycin against methicillin-resistant and methicillin-susceptible Staphylococcus aureus peritonitis in mice as mea-sured with bioluminescent bacteria,” Antimicrobial Agents andChemotherapy, vol. 51, no. 5, pp. 1787–1794, 2007.

[32] C. H. Contag, P. R. Contag, J. I. Mullins, S. D. Spilman, andD. A. Benaron, “Photonic detection of bacterial pathogens inliving hosts,” Molecular Microbiology, vol. 18, no. 4, pp. 593–603, 1995.

[33] C. H. Contag, S. D. Spilman, P. R. Contag et al., “Visualizinggene expression in living mammals using a bioluminescentreporter,” Photochemistry and Photobiology, vol. 66, no. 4, pp.523–531, 1997.

[34] K. P. Francis, D. Joh, C. Bellinger-Kawahara, M. J. Hawkinson,T. F. Purchio, and P. R. Contag, “Monitoring bioluminescentStaphylococcus aureus infections in living mice using a novelluxABCDE construct,” Infection and Immunity, vol. 68, no. 6,pp. 3594–3600, 2000.

[35] M. J. Hickey, T. M. Arain, R. M. Shawar et al., “Luciferase invivo expression technology: use of recombinant mycobacterialreporter strains to evaluate antimycobacterial activity in mice,”Antimicrobial Agents and Chemotherapy, vol. 40, no. 2, pp.400–407, 1996.

[36] H. L. Rocchetta, C. J. Boylan, J. W. Foley et al., “Validation ofa noninvasive, real-time imaging technology using biolumi-nescent Escherichia coli in the neutropenic mouse thigh modelof infection,” Antimicrobial Agents and Chemotherapy, vol. 45,no. 1, pp. 129–137, 2001.

[37] E. Breukink, “A lesson in efficient killing from two-componentlantibiotics,” Molecular Microbiology, vol. 61, no. 2, pp. 271–273, 2006.

[38] I. Wiedemann, T. Bottiger, R. R. Bonelli et al., “The mode ofaction of the lantibiotic lacticin 3147—a complex mechanisminvolving specific interaction of two peptides and the cell wallprecursor lipid II,” Molecular Microbiology, vol. 61, no. 2, pp.285–296, 2006.

[39] H. Brotz, G. Bierbaum, K. Leopold, P. E. Reynolds, and H.G. Sahl, “The lantibiotic mersacidin inhibits peptidoglycansynthesis by targeting lipid II,” Antimicrobial Agents andChemotherapy, vol. 42, no. 1, pp. 154–160, 1998.

[40] S. Chatterjee, S. Chatterjee, S. J. Lad et al., “Mersacidin, a newantibiotic from bacillus fermentation, isolation, purificationand chemical characterization,” Journal of Antibiotics, vol. 45,no. 6, pp. 832–838, 1992.

[41] M. P. Ryan, O. McAuliffe, R. P. Ross, and C. Hill, “Het-erologous expression of lacticin 3147 in Enterococcus faecalis:comparison of biological activity with cytolysin,” Letters inApplied Microbiology, vol. 32, no. 2, pp. 71–77, 2001.

[42] S. M. Morgan, P. M. O’Connor, P. D. Cotter, R. P. Ross, andC. Hill, “Sequential actions of the two component peptidesof the lantibiotic lacticin 3147 explain its antimicrobialactivity at nanomolar concentrations,” Antimicrobial Agentsand Chemotherapy, vol. 49, no. 7, pp. 2606–2611, 2005.

[43] S. Suda, A. Westerbeek, P. M. O’Connor, R. P. Ross, C.Hill, and P. D. Cotter, “Effect of bioengineering lacticin 3147lanthionine bridges on specific activity and resistance to heatand proteases,” Chemistry and Biology, vol. 17, no. 10, pp.1151–1160, 2010.


Review Article

Development of Class IIa Bacteriocins as Therapeutic Agents

Christopher T. Lohans and John C. Vederas

Department of Chemistry, University of Alberta, Edmonton, AB, Canada T6G 2G2

Correspondence should be addressed to John C. Vederas, [email protected]

Received 17 August 2011; Accepted 8 October 2011

Academic Editor: John Tagg

Copyright © 2012 C. T. Lohans and J. C. Vederas. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Class IIa bacteriocins have been primarily explored as natural food preservatives, but there is much interest in exploring the ap-plication of these peptides as therapeutic antimicrobial agents. Bacteriocins of this class possess antimicrobial activity against sever-al important human pathogens. Therefore, the therapeutic development of these bacteriocins will be reviewed. Biological andchemical modifications to both stabilize and increase the potency of bacteriocins are discussed, as well as the optimization of theirproduction and purification. The suitability of bacteriocins as pharmaceuticals is explored through determinations of cytotoxicity,effects on the natural microbiota, and in vivo efficacy in mouse models. Recent results suggest that class IIa bacteriocins show pro-mise as a class of therapeutic agents.

1. Introduction

Bacteriocins are natural peptides secreted by many varietiesof bacteria for the purpose of killing other bacteria. This pro-vides them with a competitive advantage in their environ-ment, eliminating competitors to gain resources. These pep-tides are ribosomally synthesized, although some are exten-sively posttranslationally modified.

The classification system for bacteriocins has been sub-ject to ongoing revision [1–3]. However, bacteriocins fromGram-positive bacteria are generally classified according tosize, structure, and modifications. Class I bacteriocins are thelantibiotics, which are highly posttranslationally modifiedpeptides containing lanthionine and methyllanthionine resi-dues. Class II consists of small peptides that do not containmodified residues. Cotter et al. suggested to divide class IIbacteriocins into several subclasses: class IIa (pediocin-likebacteriocins), class IIb (two-peptide bacteriocins), and classIIc (circular bacteriocins) [3]. However, others have suggest-ed to consider circular bacteriocins as a separate class [4].Nonbacteriocin lytic proteins, termed bacteriolysins (also re-ferred to as class III bacteriocins), are large and heat-labileproteins with a distinct mechanism of action from otherGram-positive bacteriocins [3].

Class IIa bacteriocins are generally from 37 to 48amino acids long, and are characterized by several features.

Although they do not have broad spectrum antimicrobialactivity compared to other antibiotics, they are particularlypotent inhibitors of Listeria species, showing activity at lownanomolar concentrations [5]. They are heat-stable, and notposttranslationally modified beyond the proteolytic removalof a leader peptide and the formation of a conserved N-ter-minal disulfide bridge (although some members contain anadditional C-terminal disulfide bridge). The N-terminal re-gion contains a characteristic YGNGV amino acid sequence,although variants with the alternate YGNGL sequence havebeen classified in class IIa [6]. A representative class IIa bac-teriocin is shown in Figure 1. There have been a number ofthorough reviews describing aspects of the genetics, biosyn-thesis, immunity, structure, mode of action, and the applica-tion of class IIa bacteriocins to foods [7–13].

Briefly, class IIa bacteriocins kill susceptible bacteria byforming pores in their membranes, resulting in the loss of theproton-motive force and depletion of ATP [14]. It is thoughtthat these cationic bacteriocins are drawn to bacterial cellsthrough an initial electrostatic interaction [15]. Then, theamphiphilic C-terminal α-helix inserts into the membrane,wherein the bacteriocin induces the formation of hydrophilicpores. This mechanism of action is reliant on a mannosephosphotransferase (MPT) protein complex found in themembranes of susceptible organisms, but the exact nature


Lys Tyr Tyr Gly Asn Gly Val His Cys Thr

Lys

Ser

Gly

His

CysSerValAsnTrpGlyGluAlaPheSerAla

Gly

Val Arg Leu TrpPheGlyAsnGlyGlyAsnAla

H2N

COOH

SS

Figure 1: A representation of class IIa bacteriocin leucocin A,with the YGNGV consensus sequence and an N-terminal disulfidebridge.

Figure 2: The NMR solution structure of leucocin A [20].

of this interaction is not yet clear [16–18]. This is covered inmore detail by Drider et al. [12] and Nissen-Meyer et al. [19].

Structurally, the N-termini of class IIa bacteriocins tendto exhibit a three-strand antiparallel beta-sheet structurerigidified by a disulfide bridge. The C-terminal region showsan amphiphilic helix terminating in a hairpin structure. Inaqueous conditions, class IIa bacteriocins are randomly stru-ctured. However, membrane-mimicking conditions such asdodecylphosphocholine micelles or trifluoroethanol inducestructure formation [20]. This is not unexpected as theirmode of action involves membrane permeabilization [14].The NMR solution structures of class IIa bacteriocins leuco-cin A (shown in Figure 2) [20], carnobacteriocin B2 [21] andits precursor precarnobacteriocin B2 [22], sakacin P [23],and curvacin P [24] have been solved to date.

Much of the research on class IIa bacteriocins has focus-ed on their application for food preservation. While they maybe well-suited for this purpose, there is a growing body ofresearch exploring the prospect of using these bacteriocins asin vivo therapeutic agents. Bacteriocins are a promising sub-stitute for conventional antibiotics for several reasons. Therestricted target specificity of some bacteriocins minimizestheir impact on commensal microbiota and may decrease thethreat of opportunistic pathogens. Furthermore, most bacte-riocins are active at low concentrations, and their degrada-tion products are easily metabolized by the body. With thedevelopment of resistance to many important antibiotics,another tool for fighting bacteria is invaluable.

Class IIa bacteriocins are active against several importanthuman pathogens. Perhaps most promising is their activityagainst the foodborne pathogen Listeria monocytogenes, thedeadliest bacterial source of food poisoning [25]. Up to 30%of foodborne infections by L. monocytogenes in high-risk in-dividuals are fatal. Other bacterial foodborne pathogens in-hibited by some class IIa bacteriocins include Bacillus cereus,Clostridium botulinum, and C. perfringens [5].

Beyond foodborne pathogens, class IIa bacteriocins arealso active against other human pathogens, such as vanco-mycin-resistant enterococci [26] and the opportunistic path-ogen Staphylococcus aureus [5]. Although bacteria sensitiveto class IIa bacteriocins are almost exclusively Gram-posi-tive, the Gram-negative opportunistic pathogen Aeromonashydrophila is also inhibited [27]. Bacteriocins from this classalso show other potentially therapeutic properties as antineo-plastic [28, 29] and antiviral [30] agents.

The potential of other groups of bacteriocins such as lan-tibiotics, colicins, and microcins as oral and gastrointestinalantibiotics has been reviewed by Kirkup [31]. Focusing onbacteriocins from Gram-positive bacteria, there have beensuccesses with the administration of either lantibiotic-pro-ducing bacteria [26] or purified lantibiotics [32–38] for thetreatment of infections by several different pathogens. How-ever, less is known about the in vivo use of class IIa bacterio-cins.

One method for the therapeutic use of bacteriocins is tointroduce bacteriocinogenic bacteria to the gastrointestinaltract as probiotics, which has yielded positive results. Fre-quently, the mechanisms by which probiotic bacteria benefitthe host are not well characterized, but convincing evidencehas been put forth by Corr et al. for the production of classIIb bacteriocin Abp118 in vivo [39]. Generally, the intro-duction of bacteriocinogenic bacteria prior to infection witha pathogen has been more effective [26, 39] than the con-comitant introduction of both species [40]. This suggeststhat probiotic strains may be valuable for prophylactic pur-poses, but less suited for treating preexisting infections.

Indeed, introduction of a bacteriocin either concomi-tantly with the infectious agent or postchallenge has proveneffective [32–38]. A variety of administration methods havebeen used successfully: subcutaneous, intravenous, intran-asal, intragastric, intraperitoneal, and topical [41]. The effi-cacy of the different methods has not been directly comparedand likely depends on the pathogen targeted. Furthermore,some of these methods may be unnecessarily invasive for usein humans, with oral administration being preferred.Although the possibility exists of using crude bacteriocin ex-tract instead of purified bacteriocin, the introduction ofcomplex mixtures into a human may be hazardous and lessreproducible. Instead, this paper will focus on the adminis-tration of purified class IIa bacteriocin.

Compared to other classes of Gram-positive bacteriocins,the engineering of improved class IIa bacteriocins is some-what simplified due to their unmodified nature. Creatinganalogues, by biological or chemical means, does not requireimplementation of the thioether bridges found in lantibioticsor the cyclization required for circular bacteriocins. Nor doesthe recombinant expression of class IIa bacteriocins requirethe biosynthetic machinery, such as dehydratases and cyclas-es, required for some other bacteriocins. This also allows forthe production of class IIa bacteriocins as fusion proteins, ameans of increasing production levels and simplifying puri-fications.

This paper will explore different aspects of the develop-ment of class IIa bacteriocins as therapeutic agents for in vivoutilization. The first section discusses attempts to design


bacteriocins and bacteriocin analogues with increased sta-bility and potency. Next, methods for improving productionand purification of large amounts of bacteriocin from fer-mentation and recombinant expression will be explored.Finally, the suitability of class IIa bacteriocins for therapeuticuse, based on studies testing cytotoxicity, stability, the de-velopment of resistance, and the in vivo potential of class IIabacteriocins will be examined.

2. Engineering Class IIa Bacteriocins forIncreased Stability and Potency

The structure-function relationship of class IIa bacteriocinshas been well studied, and its implications for their mode ofaction has been well reviewed [8, 12, 19]. This paper focuseson structure-function as it contributes to the development ofimproved therapeutics. Specifically, engineering bacteriocinsto increase their stability, potency, and spectrum of activity,such that they are more suitable for in vivo utilization andother applications will be discussed.

The introduction of an additional disulfide bridge likelyhas the effect of rigidifying a specific conformation and couldresult in improved bacteriocin activity. There is a subgroup ofclass IIa bacteriocins, including pediocin PA-1, that containan additional disulfide bridge near the C-terminus. The effectof introducing a C-terminal disulfide bridge into sakacin P,a bacteriocin containing only the conserved N-terminal di-sulfide bridge, was examined [42]. This modification broad-ened its spectrum of antimicrobial activity in addition to de-creasing the detrimental effect of increased temperature onpotency. The C-terminus has been otherwise associated withthe target specificity of class IIa bacteriocins [43]. Notably,this sakacin P mutant was found to retain much of its activityat 37◦C compared to the natural peptide, and thus is moreeffective at human physiological temperature [42].

The necessity of the N-terminal disulfide bridge for activ-ity in class IIa bacteriocins has also been explored. Removalof this disulfide bridge could render bacteriocins more stablein reductive environments. Substitution of cysteines 9 and 14of leucocin A [44] and mesentericin Y105 [45] with serinesresulted in a complete loss of activity. However, replacementwith hydrophobic residues such as allylglycine, norvaline,and phenylalanine resulted in retention of activity in leucocinA [46]. Furthermore, the replacement of the disulfide bridgewith a carbocycle also yielded a biologically active peptide,although the activity was decreased by an order of magnitude[44]. However, the substitution of cysteines 9 and 14 of ped-iocin PA-1 with allylglycine and phenylalanine residues re-sulted in no observable activity [46]. This work has been dis-cussed further in a mini-review by Sit and Vederas [47].

Class IIa bacteriocins may also be stabilized by simpleamino acid substitutions. Methionine-31 of pediocin PA-1was found to oxidize over time with an accompanying loss ofactivity [48]. Mutation of this residue to leucine, isoleucineor alanine resulted in only minor decreases in potency whilestabilizing the mutant [48]. Similarly, a 4- to 8-fold decreasein activity was reported for carnobacteriocin BM1 due to anoxidized methionine residue [49], but replacement with

a valine residue yielded a mutant with comparable activity[50]. However, in some cases substitution of only a singleamino acid residue in class IIa bacteriocins results in drama-tically decreased activity relative to their wild-type counter-parts [51].

Consideration of the mode of action of class IIa bacterio-cins may permit the rational design of mutants with increas-ed potency. Enhancing the net positive charge of a bacterio-cin may be expected to promote the initial electrostatic inter-action with the membrane of the target and thus result inan increase in activity. Support for this was found in the 44 K(with an additional lysine introduced to the C-terminus) andT20K mutants of sakacin P, which show increased cell bind-ing and potency relative to the wild-type peptide [52].

Approaches to stabilizing other classes of bacteriocinsmay have potential for use with class IIa bacteriocins. Due totheir composition, proteolytic cleavage of bacteriocins in thegastrointestinal tract represents a major hurdle for any at-tempts to control gastrointestinal infections. Careful alter-ation of trypsin recognition sites in class IIb bacteriocin sali-varicin P had only minor effects on activity [53]. Chemicallysynthesized peptides with incorporated d-amino acids maybe similarly expected to render the peptide less susceptible toproteolytic cleavage. Analogues of class IIb bacteriocin lacto-coccin G were synthesized with the N- and C-terminal resid-ues replaced with d-amino acids, which decreased their sus-ceptibility to exopeptidases without much effect on activity[54]. However, the extent of incorporation of d-amino acidshas limitations. The enantiomer of leucocin A was synthe-sized containing exclusively d-amino acids, but it was foundto be largely inactive [55]. This may be rationalized based ona chiral interaction between class IIa bacteriocins with theMPT complex [16]. Nonetheless, these methods may be val-uable for stabilizing class IIa bacteriocins.

Much work is focused on using biological means to createbacteriocin analogues. Mutagenesis of bacteriocins can bereadily achieved, and large quantities of a desired mutant arereadily available through recombinant expression. However,the biological production of analogues suffers the restrictionof the proteogenic amino acid library. As a contrast, chemicalpeptide synthesis offers a vast array of possibilities for theintroduction of nonproteogenic amino acids. Furthermore,unnatural structural features not found in class IIa bacter-iocins such as carbocyclic rings and d-amino acids are feasi-ble. However, chemical peptide synthesis is not trivial, and itis relatively time consuming and costly. For a chemically syn-thesized bacteriocin to be considered a viable therapeuticagent, it would have to be greatly superior to any biologicallyproducible bacteriocins.

The rational substitution of amino acids in class IIa bac-teriocins is one method of creating mutants, and this has pro-vided much information about the structure-function rela-tionship of these bacteriocins. However, for the most part,the mutants have had decreased activity relative to the wild-type bacteriocin. Another common approach uses error-prone PCR to randomly generate mutants in the hope offinding interesting or improved activity. However, approach-es such as DNA shuffling [51] of related bacteriocins andNNK scanning [56] have been used to randomly generate


vast numbers of mutants, greatly increasing the numberof variants produced without requiring a proportionateamount of labour.

NNK scanning allows for the systematic examination ofthe role of each residue in a peptide. The native codons arereplaced one by one with the NNK triplet oligonucleotide,replacing the amino acid coded for by that codon with anyof the 20 proteogenic amino acids. This allows for testing amuch larger number of variants without requiring the timeconsuming preparation of each mutant separately. Conseq-uently, the possibility of discovering a mutant with increasedpotency is greater. NNK scanning has been applied to pedio-cin PA-1 to examine the importance of each residue for bac-tericidal activity and was indeed successful in creating somemutants with increased activity [56].

Often, changing one amino acid at a time is not sufficientto create improved variants. It has been suggested that bacter-iocins have evolved to be as effective as possible, and so thecreation of improved bacteriocins requires greater modifica-tion [51]. This is possible using an alternate approach thatallows for the swapping of multiamino acid sequences bet-ween different class IIa bacteriocins to create a hybrid bacter-iocin. This approach has been used to create a DNA-shufflinglibrary in which regions of pediocin PA-1 have been shuffledwith 10 other class IIa bacteriocins [51]. Some of the hybridsdid indeed show increased activity relative to the wild-typebacteriocins from which they were derived [51].

Another approach explored for creating new analoguesis to mix the N-terminus of one bacteriocin with the C-ter-minus of another, thereby creating a chimera. Some chimerasof pediocin PA-1 with other class IIa bacteriocins showedeither comparable or greater bactericidal activities to the cor-responding natural bacteriocins against certain indicatorstrains [43, 51].

These approaches to randomly generate vast numbers ofmutants and hybrids may allow for simplified drug develop-ment, facilitating the discovery of novel potent bacteriocins.Furthermore, these approaches enable the development ofnew bacteriocins tailored towards different strains of patho-genic bacteria.

3. Methods for Improving Production ofClass IIa Bacteriocins

For any potential therapeutic use of class IIa bacteriocins, aninexpensive method for the production of large quantitiesmust be available. One possibility is to purify class IIa bacter-iocins from their natural producer strain, taking advantageof the cationic and hydrophobic characters of these peptides.However, these purifications typically yield only smallamounts of purified peptide, often consisting of less than amilligram per liter of culture [49, 57]. However, the outlookis not bleak, as optimizing culture conditions and improvingthe design of purifications maximizes bacteriocin recoveryand permits increased scale.

Of the class IIa bacteriocins, pediocin PA-1 is most wellcharacterized in terms of optimization of fermentation. Even

then, reported yields must be interpreted carefully, as the sen-sitivity of indicator strains varies and activity tests are per-formed differently. A variety of different cultivation methodshave been used, such as shake-flasks, batch cultures, andfed-batch cultures. Batch cultures in reactors generally allowfor greater control over conditions than shake-flask cultures, with precise control of stirring, aeration, and pH. Fed-batch cultures are similar to batch cultures, except a growth-limiting nutrient is added over time, allowing for higher celldensities.

For the large-scale production of class IIa bacteriocinsto be feasible, several conditions must be met. The yield ofthe fermentation must be satisfactory, otherwise productioncosts will be high. The growth media must also be inexpen-sive, although this must be balanced with the bacteriocinyield as the use of more expensive media has been related toimproved bacterial growth and bacteriocin production [58].

The highest reported volumetric productivity wasaccomplished by a repeated-cycle batch culture of Pedio-coccus acidilactici UL5 immobilized in κ-carrageenan/locustbean gum gel beads, reaching levels of 133 mg of pediocinPA-1 produced per liter per hour in complex de ManRogosa and Sharpe (MRS) media [58]. Using less expensivesupplemented whey permeate (SWP) media under otherwiseidentical conditions, 50 mg of pediocin PA-1 was producedper liter per hour. The production of bacteriocins hastended to be much superior in immobilized cell culturescompared to free cell cultures [59], as exemplified by thegreater than tenfold increase in production of pediocin PA-1under immobilized conditions [58]. Naghmouchi et al. havepublished an informative literature summary of recent workon fermentation yields of pediocin PA-1 [58].

Bacteriocin-producing fermentations have been testedin a large variety of media as an attempt to minimizeproduction costs. Waste from the food industry especially hasbeen investigated as an inexpensive alternative to complexgrowth media. Examples of this include mussel-processingwaste [60, 61], whey permeate [58, 62–64], trout and squidviscera, and swordfish muscle [65]. Complex growth mediatend to be composed of a mixture of nutrients tailored to cer-tain types of bacteria to meet their specific nutritional re-quirements, while industrial effluents are not so optimized[63, 66].

Although the production of large amounts of bacteriocinis feasible, the purification of these peptides is anothermatter. A review by Carolissen-Mackay et al. discusses pre-vious purification approaches for bacteriocins [57]. Manypurification protocols provide poor yields of bacteriocin withrecoveries of under 20% [57]. These poor yields are likely dueto unoptimized protocols requiring a large number of steps.More recently, several general protocols have been publishedspecifically for the purpose of purifying class IIa bacteriocins[67–69]. For the industrial-scale production of bacteriocinsrequired for therapeutic use, an efficient, inexpensive, andscalable purification scheme with high recovery is needed.

Commonly, the purification of class IIa bacteriocins re-quires precipitation and centrifugation steps. The latter re-presents a major bottleneck when attempts are made to in-crease the scale of production. Furthermore, ammonium


sulfate precipitations are frequently a source of loss of mater-ial, yielding only 40%± 20% for a reported pediocin PA-1purification [67]. Using an initial ion-exchange chromato-graphic step to concentrate the bacteriocin directly from theculture media is a possible solution [67].

More recent general purification schemes generally fol-low a similar sequence, taking advantage of the cationic andhydrophobic character of class IIa bacteriocins. First, the cul-ture supernatant is passed through a cation-exchange col-umn [67–69] although loading the whole bacterial culture toavoid centrifugation has been reported [67]. Following thisstep, the eluate is further purified using hydrophobic inter-action chromatography, yielding greater than 90% pure bac-teriocin in only two steps [67, 69]. HPLC may also be usedto further clean up the sample at this stage [68]. These puri-fications allow for the acquisition of purified bacteriocin inonly a few hours [67], with bacteriocin recovery rates report-ed ranging from 60% [68] to greater than 80% [67].

The development of antibodies capable of recognizingbacteriocins has allowed for an alternate approach to purifi-cation, namely, immunoaffinity chromatography [70–73].Indeed, this approach has been used to purify divercin V41,piscicocin V1b, enterocin P, and pediocin PA-1 from culturesupernatant in a single step. Although reported yields aresparse, the recovery of enterocin P was 44% [71], while 53%of pediocin PA-1 was retained [73]. Although pure bacterio-cin is obtained after a single step, superior yields have beenreported for lengthier procedures [67], and the immuno-affinity purification requires costly noncommercial anti-body-conjugated resins.

Another notable purification approach uses triton X-114phase partitioning, which has been applied to the purifica-tion of divercin V41 [74]. This approach does not require re-moval of bacterial cells from the culture, thereby enablingcollection of the bacteriocin normally lost adhered to the cellpellet. After the two phases partition, the detergent rich phaseis removed, diluted, and loaded on to an ion-exchange col-umn. Purified bacteriocin is simply eluted from the column,with a recovery of greater than 55% [74].

All of these reported purifications have unique advan-tages and drawbacks. However, the focus has shifted to thelarge-scale production of bacteriocins instead of purifyingonly enough for characterization. These approaches have fo-cused on attaining improved yields in fewer steps with mostlyscalable steps.

3.1. Heterologous Expression. As an alternative to purificationfrom the natural producer, the recombinant expression ofbacteriocins offers a promising means for producing the largeamounts of material required for any potential therapeuticuse. There have been many reports of the heterologous ex-pression of class IIa bacteriocins in many different hosts,although the focus has been on Gram-positive lactic acidbacteria phylogenetically similar to the producer strain.Gram-negative bacteria such as Escherichia coli and yeast ex-pression platforms such as Saccharomyces cerevisiae have alsobeen used as expression hosts.

The subject of the heterologous expression of maturebacteriocins in lactic acid bacteria has been summarized in

an excellent review by Rodrıguez et al. [89], which also dis-cusses heterologous bacteriocin production in E. coli andother bacterial strains. Although some of these expressionsystems allow for the secretion of active bacteriocin into theculture supernatant, the quantity of bacteriocin obtainedfrom these cultures tends to be lower than from the naturalproducer strain. As such, this is not yet suitable for the large-scale production of bacteriocins required for any potentialtherapeutic use. However, these heterologous producers maybe suitable for food preservation as many lactic acid bacteriaare generally recognized as safe. Furthermore, these organ-isms are capable of simultaneously producing multiple dif-ferent bacteriocins allowing for a greater spectrum of activityin addition to the possibility of overcoming the developmentof resistance [90–92]. However, the use of genetically modi-fied organisms in food products is still a contentious issue.

The heterologous expression of bacteriocins as fusionproteins in E. coli has been successfully used for the produc-tion of larger amounts of bacteriocin than obtained usingother approaches. In particular, the commercial strain E. coliOrigami (DE3) has been used extensively in this area. Muta-tions in the genes encoding the glutathione and thioredoxinreductases of this strain allow for the facile formation of theconserved disulfide bridge in the host cytosol.

Additionally, the heterologous expression of class IIa bac-teriocins in E. coli as fusion proteins offers many advantages.A summary of the reported use of fusion proteins partneredwith class IIa bacteriocins is presented in Table 1. Fusionswith affinity labels, such as hexahistidine tags, allow forsimplified purification protocols. Additionally, some fusionpartners help solubilize the bacteriocin and prevent the desir-ed peptide from forming inclusion bodies, allowing for in-creased bacteriocin production. The presence of a fusionpartner also decreases the antimicrobial activity, avoidingpossible toxic effects on the host cell [93–95], although thereare exceptions [96]. Thioredoxin in particular is useful asa fusion partner. Beyond circumventing the formation ofinclusion bodies, a thermostable thioredoxin fusion allowsfor a thermal coagulation purification step [97]. This hasbeen used to remove high molecular weight contaminantsduring the purification of carnobacteriocins BM1 and B2[50]. Furthermore, thioredoxin may even assist in the for-mation of the conserved N-terminal disulfide bridge [97].

Expression of a bacteriocin solely with a hexahistidine taghas been reported for pediocin PA-1 [104]. However, this re-combinant pediocin PA-1 was found to be toxic to the E. coliproducer. The purification was complicated by the require-ment for denaturing conditions to allow for immobilized-metal affinity chromatography, although the His-tagged pep-tide was antimicrobially active. Furthermore, expression ofsmall-sized recombinant peptides in E. coli is complicateddue to the presence of proteases [105]. The expression of bac-teriocins with a larger fusion partner is likely to be advanta-geous.

The conditions used for fermentations have a significantimpact on the amount of fusion protein produced. The finalyield of purified bacteriocin is influenced by the purificationprotocol as well as the method used for fusion proteincleavage. Simple shake-flask cultures have been reported


Table 1: Fusion partners used for class IIa bacteriocins.

Fusion partner Bacteriocin

Thioredoxin

Pediocin PA-1 [95, 98],carnobacteriocins BM1 andB2 [50], divercin V41 [96],enterocin P [72], piscicolin

126 [99]

Maltose-binding proteinCarnobacteriocin B2 [93]

and its precursor [22],Pediocin AcH [100]

Intein-chitin-binding domainPiscicolin 126, divercin

V41, enterocin P, pediocinPA-1 [101]

Dihydrofolate reductase Pediocin PA-1 [94]

Xpress tag Pediocin PA-1 [102]

Cellulose-binding domain Enterocin A [103]

Hexahistidine tag Pediocin PA-1 [104]

most, although many of the reported yields are admittedlynot optimized. Piscicolin 126 was cleaved from a thioredoxinfusion yielding 26 mg per liter [99], while a divercin RV41-thioredoxin fusion yielded between 18 and 23 mg of purifiedpeptide per liter of culture [96, 106].

High-cell density E. coli cultures have also been exploredas a means to further increase the production of bacteriocinfusion proteins [50, 106]. The level of production of a recom-binant divercin V41-thioredoxin fusion in batch and fed-batch cultivation has been compared to shake-flask cultures[106]. Compared to the yield of 18 ± 1 mg obtained per literin shake flask cultures, batch and fed-batch yielded 30 ± 2and 74 ± 5 mg per liter, respectively. However, the highestyields reported are for carnobacteriocins BM1 and B2. Thesebacteriocins were expressed as thioredoxin fusions in a fed-batch fermentation induced with lactose. The final yields re-ported are around 320 mg of carnobacteriocin BM1 and car-nobacteriocin B2 per liter of the culture, fourfold greaterthan previous reports [50].

A disadvantage of using bacteriocin fusion proteins is thenecessary cleavage and further purification required to getpure bacteriocin. Enzymatic cleavage methods are the mostcommon approach, while chemical methods have also beenused. Enzymatic approaches offer the advantage of morespecific recognition sites and are thus more compatible withmost bacteriocin sequences—although the enzyme recogni-tion is not always infallible [22].

Cyanogen bromide (CNBr) is a common chemical meansof cleaving fusion proteins, selectively cleaving on the C-ter-minal side of methionine residues. However, methionine isfound in many class IIa bacteriocins. This has been circum-vented with carnobacteriocin BM1, wherein methionine-41was substituted with a valine residue with some impact onactivity [50]. However, CNBr has significant advantages overproteases: cost and cleavage efficiency. Besides being muchless expensive, the cleavage efficiency of CNBr has been re-ported to be up to twofold higher than enterokinase [50, 95].

An alternative approach for fusion protein cleavage re-quires the presence of the amino acid sequence Asp-Pro justN-terminal to the desired sequence. This cleavage method re-quires heating under strongly acidic conditions, as has beenapplied for the cleavage of a divercin V41 thioredoxin fusion[106]. This offers an inexpensive method to remove the fu-sion tag, although these may seem like unsuitable conditionsfor a peptide. However, class IIa bacteriocins tend to be stableat elevated temperatures and in acidic conditions [49].

The use of the intein-chitin-binding domain as a fusionpartner allows for circumvention of several of the issues relat-ed to fusion proteins. Following the binding of the fusionprotein on a chitin resin, cleavage is induced with DTT, re-sulting in elution of purified bacteriocin without requiringpurification from the fusion partner. This has been success-fully applied for a variety of class IIa bacteriocins, althoughthe yields have not been very substantial [101].

Yeast expression platforms are another option for theproduction of class IIa bacteriocins. Saccharomyces cerevisiaehas been used as an expression host for pediocin PA-1 [107]and plantaricin 423 [108]. Antimicrobial activity was indeedobserved, and colonies of yeast growing on agar inoculatedwith Listeria showed zones of inhibition. However, very littleantimicrobial activity was observed in the supernatant [107,108]. This low level of activity may be attributed to the bac-teriocin remaining associated with the fungal cell wall [107].

The use of Pichia pastoris as an expression host is morepromising, showing much higher levels of activity. The levelsof enterocin P produced by P. pastoris reached levels up to28 mg/L, almost four-fold higher than that produced by thenatural producer strain, Enterococcus faecium P13 [109].However, the final purified yield of enterocin P fromE. faecium P13 was still superior, demonstrating that im-proved purification methods are required to take advantageof any increased production. Class IIa-like bacteriocin hira-cin JM79 has also been expressed in P. pastoris, with similarissues [91]. Although the quantified amount of bacteriocinexceeds that of the natural producer, the observed antimi-crobial activity was found to be relatively smaller. Neutralproteases have been suggested as a possible reason for thisdiscrepancy, and bacteriocin amounts may be overestimateddue to the nature of the quantitative techniques used [91,109]. Furthermore, the activity of pediocin PA-1 producedby P. pastoris was found to be inhibited by the presence ofa collagen-like material, which appeared to be covalentlybound to the pediocin [110].

4. In Vivo Utilization of Class IIa Bacteriocins

As previously discussed, most published work regarding thein vivo use of bacteriocins has focused on the introductionof probiotic bacteria to the gastrointestinal tract, where theywill potentially secrete bacteriocins. Considerably less re-search has been done on the administration of purified bac-teriocin. The use of probiotic strains may prove beneficial asa prophylactic, but the use of purified bacteriocins appears tobe superior for countering an established infection. This hasbeen demonstrated by the administration of either pediocin


PA-1 or Pediococcus acidilactici UL5, a producer of pediocinPA-1, to mice infected with L. monocytogenes [40].

An important concern regarding the use of antibiotics isthe effects they have on the microbiota of the body. The pre-sence of commensal bacteria offers an invaluable barrier toinfection by opportunistic pathogens. Ideally, an antimicro-bial agent should specifically target the pathogenic bacteriawith only minimal impact on the natural flora. In fact, thespectrum of activity for class IIa bacteriocins may be extrem-ely well suited for targeting specific pathogens such as L.monocytogenes in vivo. Pediocin PA-1 has been tested in vitroagainst screens of common human intestinal bacteria such asbifidobacteria [75, 76], and at the concentrations tested, noantagonistic activity was observed against any of the assayedorganisms. This differs from class I lantibiotics nisin A andnisin Z, both of which inhibited the majority of Gram-posi-tive strains tested [75, 76]. Similarly, culture supernatantcontaining pediocin PA-1 was found to only inhibit onestrain of a screen of common gut bacterial species [77]. Fur-thermore, an in vivo study of pediocin PA-1 in a mousemodel showed no effect on the composition of the mouse in-testinal flora. Likewise, purified pediocin PA-1 fed to rats didnot affect the majority of their microbiota [77]. As a contrast,antibiotics such as penicillin and tetracycline strongly inhib-ited most of the common intestinal microbiota tested [76].

Two different routes of bacteriocin administration tofight L. monocytogenes have been tested in mouse models: in-travenous [78, 79] and intragastric [40]. The effects of pedio-cin PA-1 have also been studied in uninfected mice [80], rab-bits [80], and rats [77]. The suitability of the route dependson the nature of the pathogen being targeted, as well as thestage of the infection. However, as peptides, bacteriocins facechallenges related to their structure not shared by many anti-biotics.

Piscicolin 126, recombinant divercin RV41 (DvnRV41),and structural variants of DvnRV41 were all administeredintravenously to mice previously or soon to be infected withL. monocytogenes [78, 79]. In the control, the intravenous andintraperitoneal injection of these bacteriocins into healthymice resulted in no visible ill effects [78, 79]. The efficacy ofintravenous administration of bacteriocin was tested bothprior to and after the intravenous introduction of Listeria.Injection of bacteriocins was effective both 15 minutes pre-challenge and 30 minutes postchallenge. However, adminis-tration of piscicolin 126 24 hours postchallenge showed nosignificant reduction in listerial counts. Both of these exper-iments used only 2 μg of purified bacteriocin. The intracel-lular nature of Listeria as a pathogen may explain the lack ofsensitivity observed following bacteriocin administration 24hours postchallenge [25].

A possible concern with the intravenous administrationof peptides is the possibility of an immune response. Foreignpeptides are often antigenic, and the introduction of thesepeptides could trigger an immune response. To test this,pediocin AcH was intraperitoneally introduced into miceand rabbits to determine its antigenic properties. However,it did not elicit an antibody response and appears to benonimmunogenic [80]. In fact, approaches to develop anti-bodies to class IIa bacteriocins have required conjugation to

polyacrylamide gel [81] or carrier proteins such as keyholelimpet hemocyanin [70, 71, 82, 83].

The intragastric administration of bacteriocins suffersfrom its own set of problems. Bacteriocins are subjected toharsh environments designed precisely for the proteolyticcleavage of peptides and proteins. Class IIa bacteriocins aresusceptible to common digestive proteases. Furthermore, thestomach is a highly acidic environment. However, class IIabacteriocins tend to be relatively stable to acidic conditions,and pediocin PA-1 was stable at pH 2.5 for at least two hours[84].

The stability of bacteriocins in the gastrointestinal tracthas been examined by passing purified pediocin PA-1through an artificial system mimicking the human stomachand small intestine [85]. Pediocin PA-1 retained some acti-vity after 90 minutes in the artificial gastric conditions, whileall activity was lost once the sample was in the duodenal com-partment. It was suggested that pancreatin in the duodenumwas responsible for the ultimate cleavage of the pediocinPA-1, while a combination of pepsin and low pH may beresponsible for the decrease in activity observed in the gastricchamber. This is in agreement with in vivo results, as pedio-cin PA-1 fed to rats was not detected in their fecal samples[77]. Despite this, the intragastric administration of pediocinPA-1 has been proven effective for decreasing the load ofL. monocytogenes in a mouse model [40]. Furthermore,enca-psulation may preserve bacteriocin potency in thegastrointestinal tract, although this has not been reported forclass IIa bacteriocins as of yet. However, encapsulating thelantibiotic nisin in liposomes has shown some success [86–88].

The intragastric administration of pediocin PA-1 to miceinfected with L. monocytogenes has been examined [40].Treatment with 250 μg of pediocin PA-1 a day for threeconsecutive days resulted in a 2-log reduction in fecal listerialcounts. L. monocytogenes generally crosses the epithelialbarrier once it enters the small intestine and then spreadsto the liver, spleen, and central nervous system [25]. Thisbacteriocin treatment was found to decrease the amount ofL. monocytogenes reaching the liver and spleen [40].

4.1. Toxicity. An advantage that bacteriocins hold over someother antimicrobial therapies is their composition. Thesepeptides can be easily broken down to simple nontoxicamino acids that are metabolized, although this also meansthat they may not be as long-lasting compared to antibiotics.However, information regarding the in vitro cytotoxicity ofclass IIa bacteriocins is relatively limited. The cytotoxicity ofpediocin PA-1 was tested against simian virus 40-transfectedhuman colon cells and Vero monkey kidney cells [111]. Atthe levels tested, pediocin PA-1 did show cytotoxic effectson both cell lines, with a bacteriocin dose of 700 AU/mL(likely around 10–20 mg/mL) causing a decrease of greaterthan 50% on the viable cell counts. Lower dosages alsoaffected the viable cell count, although this was not asdramatic. However, combinations of carnobacteriocins BM1and B2 at concentrations 100-fold higher than requiredfor antimicrobial activity displayed no significant cytotoxic


effects to the human gastrointestinal Caco-2 cell line [112].The means of bacteriocin production and purification mustalso be considered with respect to potential toxic effects.Although this paper focuses on the administration ofpurified bacteriocin only, there still may be the possibility oftoxic contaminants retained in the bacteriocin sample, whichcould confuse any toxicity results obtained.

Based on the differing results obtained from these twoin vitro studies, further work must be done to carefullyexamine what amounts of class IIa bacteriocins can be usedsafely without cytotoxic effects. However, it is promising thatmouse and rabbit models did not show detrimental effectsfrom bacteriocin introduction [40, 79, 80].

4.2. Resistance Mechanisms. As with all therapeutic antibi-otics, the development of resistance to class IIa bacteriocinsin pathogenic bacteria is a critical issue to consider. Thistopic has been the subject of a recent review by Kaur etal. [113]. Much evidence has shown that the sensitivityof a bacterial strain to class IIa bacteriocins is dependenton the presence of a mannose phosphotransferase (MPT)transporter system [16, 114–116]. Additionally, there isevidence that nonclass IIa bacteriocin lactococcin A alsorequires MPT as a receptor [17]. Decreased expression levelsof MPT have been implicated in resistance to class IIabacteriocins in many strains of L. monocytogenes insensitiveto bacteriocins [114].

Beyond decreased receptor expression, L. monocytogenesand other susceptible strains have developed other resistancemechanisms. Multiple mechanisms may be operative at oncecontributing to an overall resistant phenotype. Modificationsof the bacterial membrane have been implicated as anothersource of bacterial resistance. Alterations of the bacterialmembrane, such that the acyl chains of phosphotidylglyc-erols are shorter and more unsaturated, affect membranefluidity and the efficiency of bacteriocin insertion [117, 118].Several other observed cell surface adaptations have beenimplicated in resistance, such as increasing the net positivecharge on the membrane and lysinylation of membranephospholipids [119].

Of special concern is the cross-resistance that has beenobserved for bacteriocins from different classes. For example,a strain of L. monocytogenes has shown resistance to nisin,pediocin PA-1, and leuconocin S, bacteriocins from threeseparate classes [120]. Based on this, the prospect of usingmultiple bacteriocins to overcome resistant strains may notbe entirely feasible. Like other antibiotics, bacteriocins needto be used judiciously to minimize the spread of resistantphenotypes.

5. Conclusions

Class IIa bacteriocins are antagonistic to many importanthuman pathogens. These bacteriocins have the ability totarget a relatively narrow range of bacteria without affectingmuch of the natural microbiota of the body, which is animportant advantage, especially when compared to otherantibiotics. Although these bacteriocins do not target as

many pathogens as other antibiotics, they have the potentialto perform a very specific role. Having another tool tocombat infections is especially important with considerationof the ever-growing problem of antibiotic resistance.

Although relatively little has been published about theactual in vivo use of class IIa bacteriocins to control bacter-ial infections, what is known is promising. Preliminary ex-periments have shown these bacteriocins to be effective atfighting L. monocytogenes infections in mouse models.

Now, more is known about the mode of action of bac-teriocins, and attempts at engineering bacteriocins withgreater potency and stability have been successful. Comparedto some other classes of bacteriocins, class IIa bacteriocins areespecially suitable for facile recombinant production and thepreparation of analogues. Improved fermentation conditionsin combination with scalable efficient purifications are nowknown, allowing for the industrial-scale production of purebacteriocin. The recombinant production of class IIa bacte-riocins as a variety of fusion proteins in E. coli has also beensuccessful, allowing for the production of even greateramounts of bacteriocin.

The application of class IIa bacteriocins as therapeuticagents is a rapidly developing area, and there is still muchto investigate. In particular, determination of their efficacyagainst pathogens other than L. monocytogenes is open for ex-ploration and would further reveal their potential for thera-peutic use. In addition, it would be informative to test thesebacteriocins against a wider range of targets beyond Gram-positive bacteria, as they have displayed unexpected activity.

The methodology is now in place to produce and purifylarge amounts of class IIa bacteriocins. The preliminarycharacterization that has been done reveals that this class ofbacteriocins possesses several desirable and useful propertiesas in vivo antimicrobial agents. What remains now is to usethat knowledge to fully explore the suitability of these pep-tides as in vivo antibiotics.

Acknowledgments

This work was supported by the Natural Sciences and Engin-eering Research Council of Canada (NSERC) and the CanadaResearch Chair in Bioorganic and Medicinal Chemistry. Theauthors thank Marco van Belkum and Avena Ross for helpfulsuggestions.

References

[1] T. Klaenhammer, “Genetics of bacteriocins produced bylactic acid bacteria,” FEMS Microbiology Reviews, vol. 12, no.1–3, pp. 39–86, 1993.

[2] I. F. Nes, D. B. Diep, L. S. Havarstein, M. B. Brurberg, V.Eijsink, and H. Holo, “Biosynthesis of bacteriocins in lacticacid bacteria,” Antonie van Leeuwenhoek, vol. 70, no. 2–4, pp.113–128, 1996.

[3] P. D. Cotter, C. Hill, and R. P. Ross, “Bacteriocins: developinginnate immunity for food,” Nature Reviews Microbiology, vol.3, no. 10, pp. 777–788, 2005.

[4] M. J. van Belkum, L. A. Martin-Visscher, and J. C. Vederas,“Structure and genetics of circular bacteriocins,” Trends inMicrobiology, vol. 19, no. 8, pp. 411–418, 2011.


[5] L. M. Cintas, P. Casaus, M. F. Fernandez, and P. E. Hernan-dez, “Comparative antimicrobial activity of enterocin L50,pediocin PA-1, nisin A and lactocin S against spoilage andfoodborne pathogenic bacteria,” Food Microbiology, vol. 15,no. 3, pp. 289–298, 1998.

[6] K. Yamazaki, M. Suzuki, Y. Kawai, N. Inoue, and T. J. Mont-ville, “Purification and characterization of a novel class IIabacteriocin, piscicocin CS526, from surimi-associated Car-nobacterium piscicola CS526,” Applied and EnvironmentalMicrobiology, vol. 71, no. 1, pp. 554–557, 2005.

[7] M. J. van Belkum and M. E. Stiles, “Nonlantibiotic antibac-terial peptides from lactic acid bacteria,” Natural Product Re-ports, vol. 17, no. 4, pp. 323–335, 2000.

[8] S. Ennahar, T. Sashihara, K. Sonomoto, and A. Ishizaki,“Class IIa bacteriocins: biosynthesis, structure and activity,”FEMS Microbiology Reviews, vol. 24, no. 1, pp. 85–106, 2000.

[9] D. B. Diep and I. F. Nes, “Ribosomally synthesized antibacte-rial peptides in gram positive bacteria,” Current Drug Targets,vol. 3, no. 2, pp. 107–122, 2002.

[10] V. Eijsink, L. Axelsson, D. Diep, L. S. Havarstein, H. Holo,and I. F. Nes, “Production of class II bacteriocins by lacticacid bacteria; an example of biological warfare and com-munication,” Antonie van Leeuwenhoek, vol. 81, no. 1–4, pp.639–654, 2002.

[11] J. M. Rodrıguez, M. I. Martınez, and J. Kok, “Pediocin PA-1, awide-spectrum bacteriocin from lactic acid bacteria,” CriticalReviews in Food Science and Nutrition, vol. 42, no. 2, pp. 91–121, 2002.

[12] D. Drider, G. Fimland, Y. Hechard, L. M. McMullen, andH. Prevost, “The continuing story of class IIa bacteriocins,”Microbiology and Molecular Biology Reviews, vol. 70, no. 2,pp. 564–582, 2006.

[13] M. Papagianni and S. Anastasiadou, “Pediocins: the bacte-riocins of pediococci. Sources, production, properties andapplications,” Microbial Cell Factories, vol. 8, no. 3, 2009.

[14] M. L. Chikindas, M. J. Garcia-Garcera, A. Driessen et al.,“Pediocin PA-1, a bacteriocin from Pediococcus acidilacticiPAC1.0, forms hydrophilic pores in the cytoplasmic mem-brane of target cells,” Applied and Environmental Microbiol-ogy, vol. 59, no. 11, pp. 3577–3584, 1993.

[15] Y. Chen, R. Ludescher, and T. Montville, “Electrostatic inter-actions, but not the YGNGV consensus motif, govern thebinding of pediocin PA-1 and its fragments to phospholipidvesicles,” Applied and Environmental Microbiology, vol. 63,no. 12, pp. 4770–4777, 1997.

[16] K. Dalet, Y. Cenatiempo, P. Cossart et al., “A σ(54)-dependentPTS permease of the mannose family is responsible forsensitivity of Listeria monocytogenes to mesentericin Y105,”Microbiology, vol. 147, no. 12, pp. 3263–3269, 2001.

[17] D. B. Diep, M. Skaugen, Z. Salehian, H. Holo, and I. F.Nes, “Common mechanisms of target cell recognition andimmunity for class II bacteriocins,” Proceedings of the Nation-al Academy of Sciences of the United States of America, vol. 104,no. 7, pp. 2384–2389, 2007.

[18] M. Kjos, Z. Salehian, I. F. Nes, and D. B. Diep, “An extracel-lular loop of the mannose phosphotransferase system com-ponent IIC is responsible for specific targeting by class IIabacteriocins,” Journal of Bacteriology, vol. 192, no. 22, pp.5906–5913, 2010.

[19] J. Nissen-Meyer, P. Rogne, C. Oppergard, H. S. Haugen, andP. E. Kristiansen, “Structure-function relationships of thenon-lanthionine-containing peptide (class II) bacteriocinsproduced by gram-positive bacteria,” Current PharmaceuticalBiotechnology, vol. 10, no. 1, pp. 19–37, 2009.

[20] N. L. Gallagher, M. Sailer, W. P. Niemczura, T. T. Nakashima,M. E. Stiles, and J. C. Vederas, “Three-dimensional structureof leucocin a in trifluoroethanol and dodecylphosphocholinemicelles: spatial location of residues critical for biologicalactivity in type IIa bacteriocins from lactic acid bacteria,”Biochemistry, vol. 36, no. 49, pp. 15062–15072, 1997.

[21] Y. Wang, M. E. Henz, N. L. Gallagher et al., “Solution struc-ture of carnobacteriocin B2 and implications for structure-activity relationships among type IIa bacteriocins from lacticacid bacteria,” Biochemistry, vol. 38, no. 47, pp. 15438–15447,1999.

[22] T. Sprules, K. E. Kawulka, A. C. Gibbs, D. S. Wishart, and J.C. Vederas, “NMR solution structure of the precursor forcarnobacteriocin B2, an antimicrobial peptide from Carno-bacterium piscicola: implications of the α-helical leader sec-tion for export and inhibition of type IIa bacteriocin activity,”European Journal of Biochemistry, vol. 271, no. 9, pp. 1748–1756, 2004.

[23] M. Uteng, H. H. Hauge, P. Markwick et al., “Three-dimen-sional structure in lipid micelles of the pediocin-like antimi-crobial peptide sakacin P and a sakacin P variant that is struc-turally stabilized by an inserted C-terminal disulfide bridge,”Biochemistry, vol. 42, no. 39, pp. 11417–11426, 2003.

[24] H. S. Haugen, G. Fimland, J. Nissen-Meyer, and P. E. Kristi-ansen, “Three-dimensional structure in lipid micelles of thepediocin-like antimicrobial peptide curvacin A,” Biochem-istry, vol. 44, no. 49, pp. 16149–16157, 2005.

[25] V. Ramaswamy, V. M. Cresence, J. S. Rejitha et al., “Listeria—Review of epidemiology and pathogenesis,” Journal of Micro-biology, Immunology and Infection, vol. 40, no. 1, pp. 4–13,2007.

[26] M. Millette, G. Cornut, C. Dupont, F. Shareck, D. Archam-bault, and M. Lacroix, “Capacity of human nisin- and pedio-cin-producing lactic acid bacteria to reduce intestinal colo-nization by vancomycin-resistant enterococci,” Applied andEnvironmental Microbiology, vol. 74, no. 7, pp. 1997–2003,2008.

[27] F. B. Elegado, W. J. Kim, and D. Y. Kwon, “Rapid purification,partial characterization, and antimicrobial spectrum of thebacteriocin, Pediocin AcM, from Pediococcus acidilactici M,”International Journal of Food Microbiology, vol. 37, no. 1, pp.1–11, 1997.

[28] L. Beaulieu, Production, Purification et Caracterisation dela Pediocine PA-1 Naturelle et de ses Formes Recombiantes:Contribution a la Mise en Evidence d’une Nouvelle ActiviteBiologique, Universite Laval, Quebec, Canada, 2004.

[29] G. Cornut, C. Fortin, and D. Soulieres, “Antineoplastic pro-perties of bacteriocins revisiting potential active agents,”American Journal of Clinical Oncology, vol. 31, no. 4, pp. 399–404, 2008.

[30] S. D. Todorov, M. Wachsman, E. Tome et al., “Characterisa-tion of an antiviral pediocin-like bacteriocin produced byEnterococcus faecium,” Food Microbiology, vol. 27, no. 7, pp.869–879, 2010.

[31] B. C. Kirkup Jr, “Bacteriocins as oral and gastrointestinalantibiotics: theoretical considerations, applied research, andpractical applications,” Current Medicinal Chemistry, vol. 13,no. 27, pp. 3335–3350, 2006.

[32] S. Chatterjee, D. K. Chatterjee, R. H. Jani et al., “Mersacidin, anew antibiotic from Bacillus in vitro and in vivo antibacterialactivity,” Journal of Antibiotics, vol. 45, no. 6, pp. 839–845,1992.

[33] B. P. Goldstein, J. Wei, K. Greenberg, and R. Novick, “Acti-vity of nisin against Streptococcus pneumoniae, in vitro, and


in a mouse infection model,” Journal of Antimicrobial Chemo-therapy, vol. 42, no. 2, pp. 277–278, 1998.

[34] D. Kruszewska, H. G. Sahl, G. Bierbaum, U. Pag, S. O. Hynes,and A. Ljungh, “Mersacidin eradicates methicillin-resistantStaphylococcus aureus (MRSA) in a mouse rhinitis model,”Journal of Antimicrobial Chemotherapy, vol. 54, no. 3, pp.648–653, 2004.

[35] M. Mota-Meira, H. Morency, and M. C. Lavoie, “In vivo acti-vity of mutacin B-Ny266,” Journal of Antimicrobial Chemo-therapy, vol. 56, no. 5, pp. 869–871, 2005.

[36] A. M. Brand, M. de Kwaadsteniet, and L. M. T. Dicks, “Theability of nisin F to control Staphylococcus aureus infection inthe peritoneal cavity, as studied in mice,” Letters in AppliedMicrobiology, vol. 51, no. 6, pp. 645–649, 2010.

[37] S. Y. Kim, S. Shin, H. C. Koo, J. H. Youn, H. D. Paik, and Y.H. Park, “In vitro antimicrobial effect and in vivo preventiveand therapeutic effects of partially purified lantibiotic lacticinNK34 against infection by Staphylococcus species isolatedfrom bovine mastitis,” Journal of Dairy Science, vol. 93, no.8, pp. 3610–3615, 2010.

[38] D. Jabes, C. Brunati, G. Candiani, S. Riva, G. Romano, andS. Donadio, “Efficacy of the new lantibiotic NAI-107 in ex-perimental infections induced by MDR Gram positive patho-gensEfficacy of the new lantibiotic NAI-107 in experimentalinfections induced by MDR Gram positive pathogens,” Anti-microbial Agents and Chemotherapy, vol. 55, no. 4, pp. 1671–1676, 2011.

[39] S. C. Corr, Y. Li, C. U. Riedel, P. W. O’Toole, C. Hill, and C.G. M. Gahan, “Bacteriocin production as a mechanism forthe antiinfective activity of Lactobacillus salivarius UCC118,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 104, no. 18, pp. 7617–7621, 2007.

[40] N. Dabour, A. Zihler, E. Kheadr, C. Lacroix, and I. Fliss, “Invivo study on the effectiveness of pediocin PA-1 and Pedio-coccus acidilactici UL5 at inhibiting Listeria monocytogenes,”International Journal of Food Microbiology, vol. 133, no. 3, pp.225–233, 2009.

[41] C. Valenta, A. Bernkop-Schnurch, and H. P. Rigler, “The anti-staphylococcal effect of nisin in a suitable vehicle: a potentialtherapy for atopic dermatitis in man,” Journal of Phar-macy and Pharmacology, vol. 48, no. 9, pp. 988–991, 1996.

[42] G. Fimland, L. Johnsen, L. Axelsson et al., “A C-terminaldisulfide bridge in pediocin-like bacteriocins renders bacte-riocin activity less temperature dependent and is a majordeterminant of the antimicrobial spectrum,” Journal ofBacteriology, vol. 182, no. 9, pp. 2643–2648, 2000.

[43] L. Johnsen, G. Fimland, and J. N. Meyer, “The C-terminaldomain of pediocin-like antimicrobial peptides (class IIabacteriocins) is involved in specific recognition of the C-ter-minal part of cognate immunity proteins and in determiningthe antimicrobial spectrum,” Journal of Biological Chemistry,vol. 280, no. 10, pp. 9243–9250, 2005.

[44] D. J. Derksen, J. L. Stymiest, and J. C. Vederas, “Antimicrobialleucocin analogues with a disulfide bridge replaced by acarbocycle or by noncovalent interactions of allyl glycineresidues,” Journal of the American Chemical Society, vol. 128,no. 44, pp. 14252–14253, 2006.

[45] Y. Fleury, M. A. Dayem, J. J. Montagne et al., “Covalentstructure, synthesis, and structure-function studies of mes-entericin Y 105(37), a defensive peptide from gram-positivebacteria Leuconostoc mesenteroides,” Journal of BiologicalChemistry, vol. 271, no. 24, pp. 14421–14429, 1996.

[46] D. J. Derksen, M. A. Boudreau, and J. C. Vederas, “Hydro-phobic interactions as substitutes for a conserved disulfide

linkage in the type IIa bacteriocins, leucocin A and pediocinPA-1,” ChemBioChem, vol. 9, no. 12, pp. 1898–1901, 2008.

[47] C. S. Sit and J. C. Vederas, “Approaches to the discovery ofnew antibacterial agents based on bacteriocins,” Biochemistryand Cell Biology, vol. 86, no. 2, pp. 116–123, 2008.

[48] L. Johnsen, G. Fimland, V. Eijsink, and J. Nissen-Meyer, “En-gineering increased stability in the antimicrobial peptidepediocin PA-1,” Applied and Environmental Microbiology, vol.66, no. 11, pp. 4798–4802, 2000.

[49] L. Quadri, M. Sailer, K. L. Roy, J. C. Vederas, and M. E.Stiles, “Chemical and genetic characterization of bacteriocinsproduced by Carnobacterium piscicola LV17B,” Journal of Bio-logical Chemistry, vol. 269, no. 16, pp. 12204–12211, 1994.

[50] J. Jasniewski, C. Cailliez-Grimal, E. Gelhaye, and A. Revol-Junelles, “Optimization of the production and purificationprocesses of carnobacteriocins Cbn BM1 and Cbn B2from Carnobacterium maltaromaticum CP5 by heterologousexpression in Escherichia coli,” Journal of MicrobiologicalMethods, vol. 73, no. 1, pp. 41–48, 2008.

[51] T. Tominaga and Y. Hatakeyama, “Development of innova-tive pediocin PA-1 by DNA shuffling among class IIa bacter-iocins,” Applied and Environmental Microbiology, vol. 73, no.16, pp. 5292–5299, 2007.

[52] M. Kazazic, J. Nissen-Meyer, and G. Fimland, “Mutationalanalysis of the role of charged residues in target-cell binding,potency and specificity of the pediocin-like bacteriocin saka-cin P,” Microbiology, vol. 148, no. 7, pp. 2019–2027, 2002.

[53] E. F. O’Shea, P. M. O’Connor, P. D. Cotter, R. Ross, and C.Hill, “Synthesis of trypsin-resistant variants of the Listeriabacteriocin salivaricin P,” Applied and Environmental Micro-biology, vol. 76, no. 16, pp. 5356–5362, 2010.

[54] C. Oppegard, P. Rogne, P. E. Kristiansen, and J. Nissen-Meyer, “Structure analysis of the two-peptide bacteriocinlactococcin G by introducing D-amino acid residues,” Micro-biology, vol. 156, no. 6, pp. 1883–1889, 2010.

[55] L. Z. Yan, A. C. Gibbs, M. E. Stiles, D. S. Wishart, and J. C.Vederas, “Analogues of bacteriocins: antimicrobial specificityand interactions of leucocin a with its enantiomer, carnobac-teriocin B2, and truncated derivatives,” Journal of MedicinalChemistry, vol. 43, no. 24, pp. 4579–4581, 2000.

[56] T. Tominaga and Y. Hatakeyama, “Determination of essentialand variable residues in pediocin PA-1 by NNK scanning,”Applied and Environmental Microbiology, vol. 72, no. 2, pp.1141–1147, 2006.

[57] V. Carolissen-Mackay, G. Arendse, and J. W. Hastings, “Puri-fication of bacteriocins of lactic acid bacteria: problems andpointers,” International Journal of Food Microbiology, vol. 34,no. 1, pp. 1–16, 1997.

[58] K. Naghmouchi, I. Fliss, D. Drider, and C. Lacroix, “PediocinPA-1 production during repeated-cycle batch culture of im-mobilized Pediococcus acidilactici UL5 cells,” Journal of Bio-science and Bioengineering, vol. 105, no. 5, pp. 513–517, 2008.

[59] J. Huang, C. Lacroix, H. Daba, and R. E. Simard, “Pediocin5 production and plasmid stability during continuous freeand immobilized cell cultures of Pediococcus acidilactici UL5,”Journal of Applied Bacteriology, vol. 80, no. 6, pp. 635–644,1996.

[60] J. A. V. Alvarez, M. P. Gonzalez, and M. A. Murado, “Pediocinproduction by Pediococcus acidilactici in solid state cultureon a waste medium: process simulation and experimentalresults,” Biotechnology and Bioengineering, vol. 85, no. 6, pp.676–682, 2004.


[61] N. P. Guerra and L. P. Castro, “Production of bacteriocinsfrom Lactococcus lactis subsp. lactis CECT 539 and Pediococ-cus acidilactici NRRL B-5627 using mussel-processing wast-es,” Biotechnology and Applied Biochemistry, vol. 36, no. 2, pp.119–125, 2002.

[62] F. Goulhen, J. Meghrous, and C. Lacroix, “Production of a ni-sin Z/pediocin mixture by pH-controlled mixed-strain batchcultures in supplemented whey permeate,” Journal of AppliedMicrobiology, vol. 86, no. 3, pp. 399–406, 1999.

[63] N. P. Guerra, M. L. Rua, and L. Pastrana, “Nutritional factorsaffecting the production of two bacteriocins from lactic acidbacteria on whey,” International Journal of Food Microbiology,vol. 70, no. 3, pp. 267–281, 2001.

[64] N. P. Guerra, P. F. Bernardez, and L. P. Castro, “Fed-batchpediocin production on whey using different feeding media,”Enzyme and Microbial Technology, vol. 41, no. 3, pp. 397–406,2007.

[65] J. A. Vazquez, M. P. Gonzalez, and M. A. Murado, “Prelimi-nary tests on nisin and pediocin production using waste pro-tein sources: factorial and kinetic studies,” Bioresource Tech-nology, vol. 97, no. 4, pp. 605–613, 2006.

[66] R. Yang and B. Ray, “Factors influencing production of bac-teriocins by lactic acid bacteria,” Food Microbiology, vol. 11,no. 4, pp. 281–291, 1994.

[67] M. Uteng, H. H. Hauge, I. Brondz, J. Nissen-Meyer, and G.Fimland, “Rapid two-step procedure for large-scale purifica-tion of pediocin-like bacteriocins and other cationic antimi-crobial peptides from complex culture medium,” Applied andEnvironmental Microbiology, vol. 68, no. 2, pp. 952–956,2002.

[68] D. Guyonnet, C. Fremaux, Y. Cenatiempo, and J. M. Ber-jeaud, “Method for rapid purification of class IIa bacteriocinsand comparison of their activities,” Applied and Environmen-tal Microbiology, vol. 66, no. 4, pp. 1744–1748, 2000.

[69] L. Beaulieu, H. Aomari, D. Groleau, and M. Subirade, “Animproved and simplified method for the large-scale purifica-tion of pediocin PA-1 produced by Pediococcus acidilactici,”Biotechnology and Applied Biochemistry, vol. 43, no. 2, pp. 77–84, 2006.

[70] J. Gutierrez, R. Criado, R. Citti et al., “Performance andApplications of Polyclonal Antipeptide Antibodies Specificfor the Enterococcal Bacteriocin Enterocin P,” Journal of Agri-cultural and Food Chemistry, vol. 52, no. 8, pp. 2247–2255,2004.

[71] C. Richard, D. Drider, I. Fliss, S. Denery, and H. Prevost,“Generation and utilization of polyclonal antibodies to a syn-thetic C-terminal amino acid fragment of divercin V41, aclass IIa bacteriocin,” Applied and Environmental Microbiol-ogy, vol. 70, no. 1, pp. 248–254, 2004.

[72] S. Cuozzo, S. Calvez, H. Prevost, and D. Drider, “Improve-ment of enterocin P purification process,” Folia Microbiolog-ica, vol. 51, no. 5, pp. 401–405, 2006.

[73] K. Naghmouchi, D. Drider, and I. Fliss, “Purification of ped-iocin PA-1 by immunoaffinity chromatography,” Journal ofAOAC International, vol. 91, no. 4, pp. 828–832, 2008.

[74] A. Metivier, P. Boyaval, F. Duffes, X. Dousset, J. P. Compoint,and D. Marion, “Triton X-114 phase partitioning for the iso-lation of a pediocin-like bacteriocin from Carnobacteriumdivergens,” Letters in Applied Microbiology, vol. 30, no. 1, pp.42–46, 2000.

[75] E. Kheadr, N. Bernoussi, C. Lacroix, and I. Fliss, “Compari-son of the sensitivity of commercial strains and infant isolates

of bifidobacteria to antibiotics and bacteriocins,” Interna-tional Dairy Journal, vol. 14, no. 12, pp. 1041–1053, 2004.

[76] G. le Blay, C. Lacroix, A. Zihler, and I. Fliss, “In vitro inhibi-tion activity of nisin A, nisin Z, pediocin PA-1 and antibioticsagainst common intestinal bacteria,” Letters in Applied Micro-biology, vol. 45, no. 3, pp. 252–257, 2007.

[77] N. Bernbom, B. Jelle, C. H. Brogren, F. K. Vogensen, B.Nørrung, and T. R. Licht, “Pediocin PA-1 and a pediocinproducing Lactobacillus plantarum strain do not change theHMA rat microbiota,” International Journal of Food Microbi-ology, vol. 130, no. 3, pp. 251–257, 2009.

[78] A. Ingham, M. Ford, R. J. Moore, and M. Tizard, “The bac-teriocin piscicolin 126 retains antilisterial activity in vivo,”Journal of Antimicrobial Chemotherapy, vol. 51, no. 6, pp.1365–1371, 2003.

[79] J. Rihakova, J. M. Cappelier, I. Hue et al., “In vivo activities ofrecombinant divercin V41 and its structural variants againstListeria monocytogenes,” Antimicrobial Agents and Chemo-therapy, vol. 54, no. 1, pp. 563–564, 2010.

[80] A. K. Bhunia, M. C. Johnson, B. Ray, and E. L. Belden, “Anti-genic property of Pediocin AcH produced by Pediococcusacidilactici H,” Journal of Applied Bacteriology, vol. 69, no. 2,pp. 211–215, 1990.

[81] A. K. Bhunia, “Monoclonal antibody-based enzyme im-munoassay for pediocins of Pediococcus acidilactici,” Appliedand Environmental Microbiology, vol. 60, no. 8, pp. 2692–2696, 1994.

[82] M. I. Martinez, J. M. Rodriguez, A. Suarez, J. M. Martinez,J. I. Azcona, and P. E. Hernandez, “Generation of polyclonalantibodies against a chemically synthesized N-terminal frag-ment of the bacteriocin pediocin PA-1,” Letters in AppliedMicrobiology, vol. 24, no. 6, pp. 488–492, 1997.

[83] J. M. Martinez, J. Kok, J. W. Sanders, and P. E. Hernandez,“Heterologous coproduction of enterocin A and pediocinPA-1 by Lactococcus lactis: detection by specific peptide-dir-ected antibodies,” Applied and Environmental Microbiology,vol. 66, no. 8, pp. 3543–3549, 2000.

[84] A. K. Bhunia, M. C. Johnson, and B. Ray, “Purification,characterization and antimicrobial spectrum of a bacteriocinproduced by Pediococcus acidilactici,” Journal of Applied Bac-teriology, vol. 65, no. 4, pp. 261–268, 1988.

[85] E. Kheadr, A. Zihler, N. Dabour, C. Lacroix, G. le Blay,and I. Fliss, “Study of the physicochemical and biologicalstability of pediocin PA-1 in the upper gastrointestinal tractconditions using a dynamic in vitro model,” Journal ofApplied Microbiology, vol. 109, no. 1, pp. 54–64, 2010.

[86] R. O. Benech, E. E. Kheadr, R. Laridi, C. Lacroix, and I. Fliss,“Inhibition of Listeria innocua in cheddar cheese by additionof nisin Z in liposomes or by in situ production in mixedculture,” Applied and Environmental Microbiology, vol. 68, no.8, pp. 3683–3690, 2002.

[87] L. M. Were, B. D. Bruce, P. M. Davidson, and J. Weiss, “Size,stability, and entrapment efficiency of phospholipid nano-capsules containing polypeptide antimicrobials,” Journal ofAgricultural and Food Chemistry, vol. 51, no. 27, pp. 8073–8079, 2003.

[88] L. Were, B. Bruce, P. M. Davidson, and J. Weiss, “Encapsu-lation of nisin and lysozyme in liposomes enhances efficacyagainst Listeria monocytogenes,” Journal of Food Protection,vol. 67, no. 5, pp. 922–927, 2004.

[89] J. M. Rodrıguez, M. I. Martınez, N. Horn, and H. M. Dodd,“Heterologous production of bacteriocins by lactic acid


bacteria,” International Journal of Food Microbiology, vol. 80,no. 2, pp. 101–116, 2003.

[90] J. Gutierrez, R. Larsen, L. M. Cintas, J. Kok, and P. E. Her-nandez, “High-level heterologous production and functionalexpression of the sec-dependent enterocin P from Enterococ-cus faecium P13 in Lactococcus lactis,” Applied Microbiologyand Biotechnology, vol. 72, no. 1, pp. 41–51, 2006.

[91] J. Sanchez, J. Borrero, B. Gomez-Sala et al., “Cloning and het-erologous production of hiracin JM79, a Sec-dependent bac-teriocin produced by Enterococcus hirae DCH5, in lactic acidbacteria and Pichia pastoris,” Applied and EnvironmentalMicrobiology, vol. 74, no. 8, pp. 2471–2479, 2008.

[92] C. Reviriego, L. Fernandez, and J. M. Rodrıguez, “A food-grade system for production of pediocin PA-1 in nisin-pro-ducing and non-nisin-producing Lactococcus lactis strains:application to inhibit Listeria growth in a cheese model sys-tem,” Journal of Food Protection, vol. 70, no. 11, pp. 2512–2517, 2007.

[93] L. E. N. Quadri, L. Z. Yan, M. E. Stiles, and J. C. Vederas,“Effect of amino acid substitutions on the activity of carno-bacteriocin B2. Overproduction of the antimicrobial peptide,its engineered variants, and its precursor in Escherichia coli,”Journal of Biological Chemistry, vol. 272, no. 6, pp. 3384–3388, 1997.

[94] G. S. Moon, Y. R. Pyun, and W. J. Kim, “Expression andpurification of a fusion-typed pediocin PA-1 in Escherichiacoli and recovery of biologically active pediocin PA-1,” Inter-national Journal of Food Microbiology, vol. 108, no. 1, pp. 136–140, 2006.

[95] L. Beaulieu, D. Tolkatchev, J. F. Jette, D. Groleau, and M.Subirade, “Production of active pediocin PA-1 in Escherichiacoli using a thioredoxin gene fusion expression approach:cloning, expression, purification, and characterization,” Can-adian Journal of Microbiology, vol. 53, no. 11, pp. 1246–1258,2007.

[96] C. Richard, D. Drider, K. Elmorjani, D. Marion, and H.Prevost, “Heterologous expression and purification of activedivercin V41, a class IIa bacteriocin encoded by a syntheticgene in Escherichia coli,” Journal of Bacteriology, vol. 186, no.13, pp. 4276–4284, 2004.

[97] E. R. LaVallie, E. A. DiBlasio, S. Kovacic, K. L. Grant, P. F.Schendel, and J. M. McCoy, “A thioredoxin gene fusion ex-pression system that circumvents inclusion body formationin the E. coli cytoplasm,” Bio/Technology, vol. 11, no. 2, pp.187–193, 1993.

[98] S. N. Liu, Y. Han, and Z. J. Zhou, “Fusion expression of pedAgene to obtain biologically active pediocin PA-1 in Esch-erichia coli,” Journal of Zhejiang University: Science B, vol. 12,no. 1, pp. 65–71, 2011.

[99] G. M. Gibbs, B. E. Davidson, and A. J. Hillier, “Novel ex-pression system for large-scale production and purificationof recombinant class IIa bacteriocins and its application topiscicolin 126,” Applied and Environmental Microbiology, vol.70, no. 6, pp. 3292–3297, 2004.

[100] K. W. Miller, R. Schamber, Y. Chen, and B. Ray, “Productionof active chimeric pediocin AcH in Escherichia coil in theabsence of processing and secretion genes from the Pedio-coccus pap operon,” Applied and Environmental Microbiology,vol. 64, no. 1, pp. 14–20, 1998.

[101] A. B. Ingham, K. W. Sproat, M. L. V. Tizard, and R. J. Moore,“A versatile system for the expression of nonmodified bac-teriocins in Escherichia coli,” Journal of Applied Microbiology,vol. 98, no. 3, pp. 676–683, 2005.

[102] P. M. Halami and A. Chandrashekar, “Heterologous expres-sion, purification and refolding of an anti-listerial peptideproduced by Pediococcus acidilactici K7,” Electronic Journal ofBiotechnology, vol. 10, no. 4, pp. 563–569, 2007.

[103] M. Klocke, K. Mundt, F. Idler, S. Jung, and J. E. Backhausen,“Heterologous expression of enterocin A, a bacteriocin fromEnterococcus faecium, fused to a cellulose-binding domain inEscherichia coli results in a functional protein with inhibitoryactivity against Listeria,” Applied Microbiology and Biotech-nology, vol. 67, no. 4, pp. 532–538, 2005.

[104] G. S. Moon, Y. R. Pyun, and W. J. Kim, “Characterization ofthe pediocin operon of Pediococcus acidilactici K10 and ex-pression of his-tagged recombinant pediocin PA-1 in Escher-ichia coli,” Journal of Microbiology and Biotechnology, vol. 15,no. 2, pp. 403–411, 2005.

[105] S. C. Makrides, “Strategies for achieving high-level expres-sion of genes in Escherichia coli,” Microbiological Reviews, vol.60, no. 3, pp. 512–538, 1996.

[106] S. Yildirim, D. Konrad, S. Calvez, D. Drider, H. Prevost, andC. Lacroix, “Production of recombinant bacteriocin divercinV41 by high cell density Escherichia coli batch and fed-batchcultures,” Applied Microbiology and Biotechnology, vol. 77, no.3, pp. 525–531, 2007.

[107] H. Schoeman, M. A. Vivier, M. du Toit, L. M. T. Dicks, andI. S. Pretorius, “The development of bactericidal yeast strainsby expressing the Pediococcus acidilactici pediocin gene(pedA) in Saccharomyces cerevisiae,” Yeast, vol. 15, no. 8, pp.647–656, 1999.

[108] C. A. van Reenen, M. L. Chikindas, W. H. van Zyl, and L.M. T. Dicks, “Characterization and heterologous expressionof a class IIa bacteriocin, plantaricin 423 from Lactobacillusplantarum 423, in Saccharomyces cerevisiae,” InternationalJournal of Food Microbiology, vol. 81, no. 1, pp. 29–40, 2003.

[109] J. Gutierrez, R. Criado, M. Martin, C. Herranz, L. M. Cintas,and P. E. Hernandez, “Production of enterocin P, an antiliste-rial pediocin-like bacteriocin from Enterococcus faecium P13,in Pichia pastons,” Antimicrobial Agents and Chemotherapy,vol. 49, no. 7, pp. 3004–3008, 2005.

[110] L. Beaulieu, D. Groleau, C. B. Miguez, J. F. Jette, H. Aomari,and M. Subirade, “Production of pediocin PA-1 in the meth-ylotrophic yeast Pichia pastoris reveals unexpected inhibitionof its biological activity due to the presence of collagen-likematerial,” Protein Expression and Purification, vol. 43, no. 2,pp. 111–125, 2005.

[111] S. E. Murinda, K. A. Rashid, and R. F. Roberts, “In vitro as-sessment of the cytotoxicity of nisin, pediocin, and selectedcolicins on simian virus 40-transfected human colon andVero monkey kidney cells with trypan blue staining viabilityassays,” Journal of Food Protection, vol. 66, no. 5, pp. 847–853,2003.

[112] J. Jasniewski, C. Cailliez-Grimal, I. Chevalot, J. B. Milliere,and A. M. Revol-Junelles, “Interactions between two carno-bacteriocins Cbn BM1 and Cbn B2 from Carnobacteriummaltaromaticum CP5 on target bacteria and Caco-2 cells,”Food and Chemical Toxicology, vol. 47, no. 4, pp. 893–897,2009.

[113] G. Kaur, R. K. Malik, S. K. Mishra et al., “Nisin and class IIabacteriocin resistance among Listeria and other foodbornepathogens and spoilage bacteria,” Microbial Drug Resistance,vol. 17, no. 2, pp. 197–205, 2011.

[114] A. Gravesen, M. Ramnath, K. B. Rechinger et al., “High-levelresistance to class IIa bacteriocins is associated with onegeneral mechanism in Listeria monocytogenes,” Microbiology,vol. 148, no. 8, pp. 2361–2369, 2002.


[115] M. Ramnath, S. Arous, A. Gravesen, J. W. Hastings, and Y.Hechard, “Expression of mptC of Listeria monocytogenes in-duces sensitivity to class IIa bacteriocins in Lactococcus lactis,”Microbiology, vol. 150, no. 8, pp. 2663–2668, 2004.

[116] M. Kjos, I. F. Nes, and D. B. Diep, “Mechanisms of resistanceto bacteriocins targeting the mannose phosphotransferasesystem,” Applied and Environmental Microbiology, vol. 77, no.10, pp. 3335–3342, 2011.

[117] V. Vadyvaloo, J. W. Hastings, M. J. van der Merwe, and M.Rautenbach, “Membranes of class IIa bacteriocin-resistantListeria monocytogenes cells contain increased levels of desat-urated and short-acyl-chain phosphatidylglycerols,” Appliedand Environmental Microbiology, vol. 68, no. 11, pp. 5223–5230, 2002.

[118] Y. Chen, R. Ludescher, and T. J. Montville, “Influence oflipid composition on pediocin PA-1 binding to phospholipidvesicles,” Applied and Environmental Microbiology, vol. 64,no. 9, pp. 3530–3532, 1998.

[119] V. Vadyvaloo, S. Arous, A. Gravesen et al., “Cell-surface alter-ations in class IIa bacteriocin-resistant Listeria monocytogenesstrains,” Microbiology, vol. 150, no. 9, pp. 3025–3033, 2004.

[120] A. D. Crandall and T. J. Montville, “Nisin resistance in Listeriamonocytogenes ATCC 700302 is a complex phenotype,” Appl-ied and Environmental Microbiology, vol. 64, no. 1, pp. 231–237, 1998.

antimicrobial peptides as therapeutic...

Documents