Antibacterial Activity of AI-Hemocidin 2, a Novel N-Terminal Peptide of Hemoglobin Purified from ...

17 Pages·2017·2.12 MB·English

marine drugs
Article
Antibacterial Activity of AI-Hemocidin 2, a Novel
N-Terminal Peptide of Hemoglobin Purified from
Arca inflata
ChunleiLi1,†,JianhuaZhu1,†,YanqingWang2,†,YuyanChen1,LiyanSong3,WeimingZheng3,
JingjingLi3andRongminYu1,3,*
1 BiotechnologicalInstituteofChineseMateriaMedica,JinanUniversity,Guangzhou510632,China;
[email protected](C.L.);[email protected](J.Z.);[email protected](Y.C.)
2 NationalEngineeringResearchCenterofMicrosphereTechnologyforControlled-ReleaseDrugDelivery,
Zhuhai519090,China;[email protected]
3 DepartmentofPharmacology,JinanUniversity,Guangzhou510632,China;[email protected](L.S.);
[email protected](W.Z.);[email protected](J.L.)
* Correspondence:[email protected];Tel.:[phone];Fax:[phone]
† Theseauthorscontributedequallytothiswork.
Received:23April2017;Accepted:27June2017;Published:29June2017
Abstract:Thecontinuedemergenceofantibioticresistantbacteriainrecentyearsisofgreatconcern.
Thesearchfornewclassesofantibacterialagentshasexpandedtonon-traditionalsourcessuchas
shellfish.Anantibacterialsubunitofhemoglobin(Hb-I)waspurifiedfromthemantleofArcainflataby
phosphateextractionandionexchangechromatography.Anovelantibacterialpeptide,AI-hemocidin2,
derivedfromHb-I,wasdiscoveredusingbioinformaticsanalysis. Itdisplayedantibacterialactivity
acrossabroadspectrumofmicroorganisms,includingseveralGram-positiveandGram-negative
bacteria,withminimalinhibitoryconcentration(MIC)valuesrangingfrom37.5to300µg/mL,and
itexhibitedminimalhemolyticorcytotoxicactivities. TheantibacterialactivityofAI-hemocidin2
wasthermostable(25–100◦C)andpHresistant(pH3–10). Thecellularintegritywasdetermined
byflowcytometry. AI-hemocidin2wascapableofpermeatingthecellularmembrane. Changesin
the cell morphology were observed with a scanning electron microscope. Circular dichroism
spectra suggested that AI-hemocidin 2 formed an α-helix structure in the membrane mimetic
environment. Theresultsindicatedthattheanti-bacterialmechanismforAI-hemocidin2occurred
throughdisruptingthecellmembrane. AI-hemocidin2mightbeapotentialcandidatefortackling
antibioticresistantbacteria.
Keywords:Arcainflata;peptide;hemoglobin;purification;structuralelucidation;antibacterialactivity
1. Introduction
The emergence and rapid spread of bacterial infections caused by antibiotic resistant strains
has led to a health crisis in recent years [1–6]. The threat of drug-resistant bacterial infections has
been attributed to the abuse and misuse of antibiotics as well as a lack of development of new
antibacterial agents [1,3,5,6]. A recent investigation demonstrated that premature deaths caused
by antibiotic resistant bacteria will rise to 1,000,000 lives per year and cost the global economy
$100trillionby2050[7]. Theurgentneedfornewtypesofantibacterialagentshasledtoinnovative
multidisciplinaryresearch.
Antimicrobialpeptides(AMPs)areproducedbyhostsaspartoftheinnateimmuneresponseand
constitutethefirstlineofdefenseagainstinvadingpathogenicmicrobes[8]. AMPstypicallycomprise
10–50 amino acids, including basic residues such as arginine or lysine, that possess a positive net
Mar.Drugs2017,15,205;doi:10.3390/md15070205 www.mdpi.com/journal/marinedrugs
Mar.Drugs2017,15,205 2of17
charge and amphipathic secondary structure such as an α-helix or β-sheet [9–11]. Based on these
properties, AMPs target and perturb the microbial cytoplasmic membranes that typically present
anionicsurfacesviaanionicinteraction. SomeinvestigationshavedemonstratedthatmostAMPs
permeabilizemicrobialcytoplasmicmembranes. AlthoughmostAMPsaffecttheintegrityofmicrobial
membranes,itremainstobedeterminedwhetherthemicrobialmembraneistheonlysiteofactionor
whethermembranepermeabilizationisalwaysfatal. SomeAMPsthatdonotextensivelypermeabilize
microbialmembranesmightinteractwithintracellularcomponentsandthusaffectmicrobeviability.
ThemechanismofactionofAMPsremainstobefullyelucidated[12]. Comparedwithantibiotics,
AMPshavedrawnincreasingattentionasnovelantibacterialagentstocombatbacterialinvasiondueto
theirspecialmechanismsofaction,non-induciblebacterialresistanceandlackofdetrimentaleffectson
humans[13]. AMPsderivedfrommarineorganismsarepromisingcandidatesfordrugdevelopment
andasnutritionalfoodadditives. Bycontrastwithon-landcompounds,marinepeptideshavediverse
sequences,uniquestructuresandprominentbioactivitiesasaresultofparticularadaptationstooceanic
environments[14]. Manypeptidesandderivatives(suchasmytilin,defensin,mytimycinandmyticins)
derivedfromshellfishhavebeenreportedtoexhibitantimicrobialactivity[15]andhavepotential
applicationsinthehealthcareandfisheryindustries[16,17].
Hemoglobin, which is widely distributed among organisms, has previously been regarded
to function exclusively as an endogenous respiratory protein. However, studies have shown that
hemoglobinexertsinhibitoryeffectsonbacteriaandisasourceofantibacterialpeptides,inadditionto
itsroleasanoxygencarrier,anditsinvolvementintransportingsulfideandregulatingtheacid–base
balance[18]. Antibacterialpeptidefragmentsderivedfromhemoglobinsubunits(hemocidins)have
beenconsideredasimportanteffectorsoftheinnateimmuneresponseagainstpathogeninfections[19].
Hemocidins in humans, bovine and channel catfish were found to possess potent antimicrobial
activities against pathogenic bacteria [20–24]. P3, a novel hemoglobin-related peptide derived
from bovine erythrocytes, exhibited significant antimicrobial activity [25]. In addition, Hb137-141,
a peptide isolated from the enzyme digestion of bovine hemoglobin alpha-chain showed potent
antibacterialactivityagainstEscherichiacoli,Salmonellaenteritidis,Listeriainnocua,Micrococcusluteus
andStaphylococcusaureus[26].Althoughstudiesonhemocidinshavebeenperformedinawiderangeof
species,theyhavebeenlimitedtovertebrates. Insomeinvertebrates,suchasarthropodsandmollusks,
hemolymph is analogous to the blood in vertebrates. There are three types of oxygen-transport
proteins in the hemolymph of invertebrates: hemoglobin, hemocyanin and hemerythrin [27].
Among these, hemocyanins are regarded as the major, multifunctional proteins in invertebrate
hemolymph, responsible for oxygen transport and contributing to innate immunity in resisting
invasionbypathogenicmicrobes[27,28]. Recently,hemocyanin-derivedantimicrobialpeptideshave
been identified from abalone, a marine mollusk [28]. However, few studies have focused on the
hemoglobinsandhemocidinsofinvertebrates.
A.inflataisaspecialtypeofbivalveshellfishthatcontainshemoglobin.A.inflata,whichbelongsto
theArcidaefamilyunderthephylumMolluscaandtheclassLamellibranchiata,iswidelydistributed
inthewesternPacificOcean. Ithasbeenusedasanutritiousfoodmaterialaroundtheworld,and
a traditional medicine for treating blood deficiency, epigastric pain and indigestion for centuries
in China. A polypeptide from A. inflata with impressive antitumor activity was reported by our
researchgroup[29]. However,littleattentionhasbeendevotedtoantibacterialpeptidesinA.inflata.
Asaresultoftheirlackofaspecificimmunesystem,invertebratessuchasA.inflatarelyontheinnate
immuneresponsecomprisinghemocytesandhumoralfactorstoresistinvasionfrompathogensin
theenvironment[30]. Therefore,itmaybepossibletodiscovernovelantibacterialagentsinmarine
invertebratessuchasA.inflata.
Enzymatichydrolysisisafeasiblemethodwithwhichtodiscoverpotentantibacterialpeptides.
Peptideshydrolyzedbytrypsin,preferablycontainingarginineandlysine,usuallypossesspositive
charges and exhibit antibacterial activity [31,32]. However, it is difficult to identify the bioactive
peptides among the huge number of peptides produced by hydrolysis. In addition, the bioactive
Mar.Drugs2017,15,205 3of17
Mpaerp. Dtirdugess 2m017a,y 15b, e20c5o me inactive during the enzymatic hydrolysis process. Since bioactive pep3 toidf 1e7s 
possessingsimilardomainandphysiochemicalparametersexhibitsimilarbioactivities,bioinformatic
bioinformatic analysis is a rapid and convenient method for discovering novel bioactive peptide 
analysisisarapidandconvenientmethodfordiscoveringnovelbioactivepeptidesequences. Hence,
sequences. Hence, enzymatic hydrolysis simulation combined with bioinformatic analysis is an 
enzymatichydrolysissimulationcombinedwithbioinformaticanalysisisanefficientandinnovative
efficient and innovative approach to discover potential antibacterial agents. 
approachtodiscoverpotentialantibacterialagents.
The aim of this study was to purify and characterize antibacterial agents from A. inflata. A novel 
TheaimofthisstudywastopurifyandcharacterizeantibacterialagentsfromA.inflata. Anovel
antibacterial peptide, AI‐hemocidin 2, was discovered from A. inflata and the physicochemical and 
antibacterialpeptide,AI-hemocidin2,wasdiscoveredfromA.inflataandthephysicochemicaland
structural properties were determined. Moreover, the preliminary mechanism of antibacterial action 
structuralpropertiesweredetermined. Moreover,thepreliminarymechanismofantibacterialactionof
of AI‐hemocidin 2 was revealed. 
AI-hemocidin2wasrevealed.
2. Results and Discussion 
2. ResultsandDiscussion
22.1.1. .PPuurrifiificcaattioionn ooff HHbb‐-II 
FFoolllloowwiinngg aann aannttiibbaacctteerriiaall aassssaayy,, HHbb‐-II wwaass ppuurriiffiieedd ffrroomm tthhee mmaannttlleess ooff AA.. iinnffllaattaa bbyy aammmmoonniiuumm 
ssuullffaattee pprreecciippiittaattiioonn aanndd ccoolluummnn cchhrroommaattooggrraapphhyy..T Thheer reessuultlstso offt htheep puurirfiifciactaitoinonp rporcoecsessas raerseh sohwonwinn 
iFni gFuigruer1e a1n adnTda Tbalebl1e. 1. 

FFiigguurree 11.. PPuurriiffiiccaattiioonn ooffh heemmoogglolobbininH Hb-bI:‐I(:A ()As)e psaerpaatrioatnioonf aocft iavcetfivraec tfiroancstiionntsh eins uthpee rsnuaptaenrntabtyaDntE bAyE 
DFEasAtEF lFoawsta nFiloown eaxnchioann geexcchharnomgea tcohgrroampahtyo;g(rBa)pphuyr; ifi(Bc)a tpiounriofifcaatcitoivne offr aacctitoivne J1frSa1ctwiohni cJh1Sco1 nwtahinicehd 
cHonbt-aIionbetda iHnebd‐If roobmta(iAne)db yfrSoPmF a(Ast) Fbloyw SPan Fioanste Fxclohwan agneicohnr oemxcahtaonggraep chhyr;o(mC)aEtoSgIr-MapShmy; a(sCs)s pEeScIt‐rMumS 
mofasHs bs-pI;eactnrdum(D o)ft hHebfi‐In; aalnpdu (rDifi)c tahteio fninoafl Hpubr-iIf.ication of Hb‐I. 
TTaabbllee 11. .AAnntitbibaaccteterriaial laaccttivivitityy ooff eeaacchh ppuurrififiieedd ffrraaccttiioonn aaggaaiinnsstt EE.. ccoollii AATTCCCC2255992222.. 
Fractions  J1  J2  J3 J4 J1S1 J1S2 J1S3  Cipro
Fractions J1 J2 J3 J4 J1S1 J1S2 J1S3 Cipro
IZ (mm)  10.18 ± 0.24  9.22 ± 0.21  nd  nd  16.19 ± 0.62  nd  6.74 ± 0.18  31.66 ± 0.22 
IZ(mm) 10.18±0.24 9.22±0.21 nd nd 16.19±0.62 nd 6.74±0.18 31.66±0.22
IZ = Inhibition zone (mm). nd = not detected. Cipro = Ciprofloxacin. Sample amount: 20 μL. The 
IZ=Inhibitionzone(mm).nd=notdetected.Cipro=Ciprofloxacin.Sampleamount:20µL.Theconcentrationof
csoanmcpelnetsrwataiso2n. 0o0f msagm/mpLle.sE wacahsa 2ss.0a0y wmags/mpeLrf.o Ermacehd ainsstraiyp lwicaatse ,paenrdfothrmereedsu ilnts tarriepplirceasteen,t aedndas tthhee rmeseuanlts± aSrDe .
presented as the mean ± SD. 
Thehighestantimicrobialactivitypeak,J1S1,wasobtainedbyatwo-steppurificationprocedureof
The  highest  antimicrobial  activity  peak,  J1S1,  was  obtained  by  a  two‐step  purification 
ion-exchangechromatography(Figure1A,B).J1S1exhibitedthehighestantimicrobialactivityagainst
procedure of ion‐exchange chromatography (Figure 1A,B). J1S1 exhibited the highest antimicrobial 
E.coliwithinhibitionzonediametersof16.19±0.62mm. ToevaluatethepurityoffractionJ1S1,tricine
activity against E. coli with inhibition zone diameters of 16.19 ± 0.62 mm. To evaluate the purity of 
sodiumdodecylsulfatepolyacrylamidegelelectrophoresis(Tricine-SDS-PAGE),nativepolyacrylamide
fraction  J1S1,  tricine  sodium  dodecyl  sulfate  polyacrylamide  gel  electrophoresis 
gelelectrophoresis(Native-PAGE)andRP-HPLCwereapplied,asdescribedinapreviousstudy[33].
(Tricine‐SDS‐PAGE), native polyacrylamide gel electrophoresis (Native‐PAGE) and RP‐HPLC were 
applied, as described in a previous study [33]. A single band in both Tricine‐SDS‐PAGE and 
Native‐PAGE revealed that J1S1 was a pure polypeptide with an approximate molecular weight of 
15 kDa (Figure S1A,B). Furthermore, J1S1 was loaded onto a reversed‐phase C18 column to confirm
Mar.Drugs2017,15,205 4of17
AsinglebandinbothTricine-SDS-PAGEandNative-PAGErevealedthatJ1S1wasapurepolypeptide
with an approximate molecular weight of 15 kDa (Figure S1A,B). Furthermore, J1S1 was loaded
onto a reversed-phase C18 column to confirm the purity. A single peak was found at 10.43 min
Mar. Drugs 2017, 15, 205  4 of 17 
(Figure1D),highlightingthehighpurityofJ1S1. Electrospray-ionizationmassspectrometry(ESI-MS)
analysis of purified J1S1 revealed that its accurate molecular weight was 15,859.0 Da (Figure 1C).
the purity. A single peak was found at 10.43 min (Figure 1D), highlighting the high purity of J1S1. 
Comparedwithorganicsolventextraction, ammoniumsulfateprecipitationcanbeusedtoobtain
Electrospray‐ionization mass spectrometry (ESI‐MS) analysis of purified J1S1 revealed that its 
moreapcrcoutreaitnes manoldecpuelaprt iwdeeisgwhti twhabs i1o5a,8ct5i9v.0it yD.aM (Faingyurset u1dCi)e. sChomavpeauresde dwaitmh moroganniuicm sosluvelfnatt eexptrraecctiipoint,a tion
foranatimbmacotenriuiamlp seuplftaidtee pirseocliaptiitoanti,oinn cclaund bien gustehde tpou orbifitaciant mioonroe fpprolatenitnasr iacnidn p16ep3t[i3d4e]s, wpiltahn btaioriaccitnivAity-1. [35]
andsaMkaancyin  sLtuSdJ6ie1s8  h[3a6v]e. Eu.secdol iaismomnoenoiufmth esumlfoatset cpormecmipiotantipoant hfoorg eannitcibbaacctetreiraila  paenpdtiadef reisqoulaetniotnc,a use
ofnosinoccloumdiinagl tahned pucorimficmatiuonni toyf -palcaqnutairriecdin b1a6c3t [e3r4i]a, lpilnanfetacrtiicoinns A.‐I1n [3re5]c eanndt ysaekaarcsi,ni nLfSeJc6t1i8o n[3s6]w. Eit. hcoEli .coli
is  one  of  the  most  common  pathogenic  bacteria  and  a  frequent  cause  of  nosocomial  and 
havebeenincreasinglyreported,andthenumberofE.coliisolatesfoundtoberesistanttoantibiotics,
community‐acquired  bacterial  infections.  In  recent  years,  infections  with  E.  coli  have  been 
especiallytocephalosporinsandfluoroquinolones, hasincreasedsignificantlyduringthelasttwo
increasingly reported, and the number of E. coli isolates found to be resistant to antibiotics, especially 
decades[37]. ThepolypeptidefractionJ1S1,purifiedfromA.inflata,whichiscapableofinhibiting
to cephalosporins and fluoroquinolones, has increased significantly during the last two decades [37]. 
E.coli,isapromisingcandidateforthedevelopmentofantibacterialagents.
The polypeptide fraction J1S1, purified from A. inflata, which is capable of inhibiting E. coli, is a 
promising candidate for the development of antibacterial agents. 
2.2. IdentificationofHb-I
2.2. Identification of Hb‐I 
J1S1 was subjected to amino acid sequence analysis by automated Edman degradation.
The N-terJm1Si1n awlaas msuinbjoecatecdid tos eaqmuineno caeciwd asseqdueentecrem ainnaelydsitso byb eauPtSoVmQatGedA EAdQmQanL TdAegDraVdKatio(Fni.g Tuhree  S2).
ThereNsu‐tletrsmoifnBalL aAmSiTnos eaacricdh seesqsuheonwcee dwaosn ldyetoenrmemineadtc hto tobet hPeSVheQmGoAgAloQbQinLTsuAbDuVnKit (IF(iHgubr-eI )So2f).S Tcahpeh arca
broughretosunlitis( BofA BMLA63S3T2 s3e.1ar)cwheisth shaohwigedh odnelgyr oeneeo mfsaitmchi ltaor itthye (h9e3m%o)g.lWobiitnh stuhbeupnoitl yI m(Hebr‐aI)s eofc hScaaipnhraercaac tion
broughtonii (BAM63323.1) with a high degree of similarity (93%). With the polymerase chain reaction 
(PCR)technique,J1S1thatcontainedanopenreadingframeof441nucleotideswasfoundtobeidentical
(PCR) technique, J1S1 that contained an open reading frame of 441 nucleotides was found to be 
toHb-IinA.inflata(GenBank: AB713934.1). ByBLASTsearch,Hb-IinA.inflatashowedhighsimilarity
identical to Hb‐I in A. inflata (GenBank: AB713934.1). By BLAST search, Hb‐I in A. inflata showed 
withthehemoglobinsubunitIofScapharcainaequivalvis,ArcasubcrenataandTegillarcagranosa,butlittle
high similarity with the hemoglobin subunit I of Scapharca inaequivalvis, Arca subcrenata and Tegillarca 
similaritywiththehemoglobin-relatedantibacterialpeptidesofvertebrates(Figure2). Somestudies
granosa, but little similarity with the hemoglobin‐related antibacterial peptides of vertebrates (Figure 
havedemonstratedclearantibacterialactivityinthehemoglobinofvertebrates[19–27].However,there
2). Some studies have demonstrated clear antibacterial activity in the hemoglobin of vertebrates 
hasbe[1e9n–l2i7tt]l.e Hfoowcuesveorn, tthheerea nhtaisb abceteenr ilaitltleef ffeocctusso ofni nthvee ratnetbibraatcetehrieaml eofgfelcotbs ionf ainnvderhteembroatcei dhienms.oCgloombinp ared
withtahnedh heemmoogcliodbinins. sCuobmupnaitr,ehd ewmiothc itdhien shehmavogelostbrionn sguebruannitt,i bhaecmteorciiadlinasc thivaivtey satnrodngstearb ailnittiyba[3ct8e,r3i9al] and
therefaocrteivhitayv aentdh setapboiltietyn t[i3a8l,3t9o] baendp othteernetfoarnet hibaavcet tehreia ploatgenetniatsl .to be potent antibacterial agents. 

FigurFei2g.uSree q2u. eSnecqeuecnocme pcaormispoanriosofnH obf- IHwbi‐tIh wthiteh htehme ohgemloobginlofbrionm froomth eorthinevr eirntveebrrtaetbersa.teIsd. eIndteicnatilcaalm  ino
acidraemsiidnuoe ascaidr erehsiigdhuleisg ahrtee dhiignhlyigehlltoewd ,ina nydellsoiwm,i laanrda smiminiloara acmidinreos aidciude rsesairdeuhesig ahreli ghhigtheldigihntebdl uine and
greenb,lruees paencdt igvreeleyn., respectively.
Mar.Drugs2017,15,205 5of17
Mar. Drugs 2017, 15, 205  5 of 17 
22..33.. BBiiooiinnffoorrmmaattiicc AAnnaallyyssiiss aanndd AAnnttiibbaacctteerriiaall FFuunnccttiioonn SSccrreeeenniinngg 
PPrrootteeiinn hhyyddrroollyyssiissb byye neznyzmymeseiss aisf eaa sfiebalesiwblaey wtoayd istcoo vdeisrcpoovteern tpaontetinbta catnertiibalapcteepritaidl epsehpidtiddeens 
hinidpdreonte iinns .pPreoptetiidness. hPyedprtoidlyezse dhybdyrtorlyypzseidn ,bpyre fterryapbsliyn,c opnrteafienrianbglya rgcoinnitnaeinainndg lyasrignien,iunesu aanlldy plyosssineses, 
upsousaitlilvye pcohsasregsess paonsditievxeh icbhiatragnetsib aancdte reixahliabcitti vanittyib[a3c1t,3er2i]a.lF aocrtievxiatym [p3l1e,,3i2n]d. Foolirc iedxianm,wphlei,c hincdoonlitcaiidniend, 
warhgiicnhin ceoonrtalyinsiende aartgitisniCn-et eormr linysuisn,ed iastp liatsy eCd‐tpeormtenintuasn, tidbiascptlearyiaelda cptoivteitnyt [4a0n]t.ibHayctderroialyl saisctsiivmituyl a[t4i0o]n. 
Hcoymdbroinlyesdisw siimthublaiotiionnfo crommabtiicneadn awlyitshis biisoainnfoerffimcaietinct aannadlyisnisn oisv aanti vefefiacpiepnrto aanchd tinondoivscaotivveer appoptreonatciahl 
taon dtiibsaccotveerria ploatgeenntitasl. aTnhtirboaucgtehritarly apgsiennthsy. Tdhrorloyusgish stirmypuslainti ohnydarnodlybsiiso isnimfourmlaatitoicn aannadl ybsiiosinufsoirnmgatthice 
aannatilmysiicsr oubsiinalg pthepe taidnetimdaictraobbaisael p(AepPtDid),e AdaI-thaebmasoec (iAdiPnDs,),i nAcIl‐uhdeimngocsidevinesr,a ilnpcelupdtiidneg fsreavgemraeln ptespwtiidthe 
fproatgemnteinaltsa nwtiitbha cptoetreianltifauln acntitoibna,cwteerrieali dfuennctitfiioend, (wFiegrue riede3nAt)i.fied (Figure 3A). 

FFiigguurree 33.. IInn ssiilliiccoo aannaallyyssiiss ooff HHbb‐-II:: ((AA)) TTrryyppssiinn hhyyddrroollyyttiicc ssiitteess aanndd ppeeppttiiddee ffrraaggmmeennttss aalliiggnneedd wwiitthh 
APD;and(B)SchifferandEdmundsonhelicalwheelprojectionofAI-hemocidins.Thehydrophobic
APD; and (B) Schiffer and Edmundson helical wheel projection of AI‐hemocidins. The hydrophobic 
residuesareyellow,positivelychargedhydrophilicresiduesareblue,negativelychargedhydrophilic
residues  are  yellow,  positively  charged  hydrophilic  residues  are  blue,  negatively  charged 
residuesarered,asparagineandglutamineareinpinkcolor,smallneutralresiduesaregray,andsmall
hydrophilic residues are red, asparagine and glutamine are in pink color, small neutral residues are 
polarresiduesarepurple.
gray, and small polar residues are purple. 
TThhee sseeqquueenncceessa nadndp hpyhsyicsoiccohcehmeimcailcpalr oppreorptieerstioefsA oI-fh AemI‐ohceimdioncsidariensli satreed ilnistTeadb lein2 .TAalbllpe e2p.t idAelsl 
pheapdtiadnese thpaods iati vneetc hpaorsgiteivaen dchwaregree panrodp woseerde tporobpeocsaetdio ntoic bpee pcatitdioensi.cT pheeprtaidtieoss. oTfhpeo rlaatrioress oidf upeoslator 
rneosnidpuoelasr troe sniodnupesolware rreesaipdpureosx iwmeartee layp5p0r%ox.iAmmatoenlyg 5th0e%m. ,AAmI-ohnemg othceidmin, A2Is‐hhoewmeodcitdhienh 2ig shheoswtreadti othoef 
hpioglharesrte sriadtuioe sotfo pnoolnapr orlaesrirdeusiedsu teos (n7o1n.4p3o%la),r wrehseirdeuaessA (I7-h1.e4m3%oc)i,d winh2erheaads tAheI‐lhoewmeostcihdyidn ro2p hhaodb icthitey 
loofw0e.0s4t 3h.yTdhroepphroebdiiccitteyd ohf e0l.i0x4s3t.r uTchteu rpersedoifctthede pheeplitxi dsetrsuacrteursehso owf nthine pFeigputirdee3sB a.rAe lslhpoewpnti dine sFcigouurlde 
3foBr. mAlhl epliexpetsid,tehso cuoguhldo nfolyrmA hI-ehleixmeso,c tihdoinusg2h wonalsyc AapI‐ahbelmeoofcifdoirnms i2n gwaass tcaabpleabalme pofh ifporamthiincgs tar usctatubrlee 
athmapthcoipnatathinice dsatruhyctdurroep hthoabti ccsoidnetaainndeda pao shityivderolypchhoabrigc edsisduer faanced. Gae npeorsailtliyv,ealnyt ibchacatregreiadl pseuprtfiadcees. 
Gpoesnseersasllsye,v aenratilbfaecatteurriaels ipnedpitciadteivs epoofstsheessir saenvteibraalc tfeeraitaulraecst ivinitdyic[a4t1iv,4e2 ]o.fO tnheeifre aatnutriebaiscttehreiairl paoctsiivtiivtye 
[c4h1a,4rg2e]., Oannde afenaotuthreer isfe tahtueirre pisostihtievep rcehsaerngcee, oafndat alenaostthe5r0 %feaotfuhrey disr othpeh opbriecsernescied oufe sa.t Tlehaestp 5o0s%iti voef 
hchyadrrgoephisocboicn druesciidvueetso. aTnheel epctorsoistitvaeti cchinatregrea citsio cnowndituhctivhee btoa catenr iaellemctermosbtartainc ei,nwtehriacchtiiosnn ewgiathti vtehley 
bchacatregreiadl [m41e]m. Tbrhaenhe,i gwhhriacthio iso fnhegyadtriovpelhyo bchicarrgeseidd u[4e1s]i. sTthheo uhgighht troatcioon otrfi bhuytderotophdoisbriucp rteiosinduoefst hise 
tbhaocutegrhiatl tmo ecmonbtrraibnue.teH toow deivsreur,patimono roef ctrhuec ibaalcfeteartiuarle mofemthberpaenpet. idHeoswpeovsseer,s sai nmgoarnet icbraucctiearli afleaactutiroen oifs 
tthhee pfoerpmtiadteios npoofssaenssoirndge raendtiabnadctestraiabll eacatmiopnh iisp aththei cfosrtmruacttiuorne .oTf haen hoyrddreorpedh oabnidc ssitdaebloef athmepshtriupacttuhrice 
sistrtuhcotuurgeh. tTthoee hnytedrroinpthooabnicd sdidiser uopf tththee sbtrauccteturirae mise tmhoburagnhet ,two heenrteear sinthtoe ahnydd rdoipshruilpict rtehsei dbuaecstearirae 
mcoenmsibdrearneed, twohperroevaisd ethteh ehyodprpooprhtuilnicit ryetsoidcuoems bairnee coornasiddheerreedt otot hperomveidmeb trhaen oepspuorfratucen[it4y2 ]t.oA clotmhobuinghe 
oArI -ahdehmeroec itdoi nth2e cmonemtaibnreadnea shuirgfhacrea t[i4o2]o. fAplothlaorurgehs idAuI‐ehseamnodcildoiwn h2 ycdornotpaihnoebdi cait yh,igthhe raamtiop hofip paothlairc 
rsetrsuidcutuerse atnhda tlofwor mhyeddrobpehtwobeiecnityth, tehhe yadmropphhipoabtihcics isdtreuacntudrea thpaots iftoivrmeleydc bheatrwgeedens uthrefa hcyedcroonpfheorrbeidc 
saindteib aancdte ari paloascittiivveitlyy. charged surface conferred antibacterial activity.
Mar.Drugs2017,15,205 6of17
Table2.SequencesandphysicochemicalpropertiesofAI-hemocidins.
Peptide Sequence Length pI Mw NC H PR/n%
1 PSVQGAAAQLTADVKK 16 9.01 1583.81 1 0.196 8/50%
2 DLRDSWKVIGSDKK 14 8.50 1646.86 1 0.043 10/71.43%
3 ISAAEFGKINGPIKK 15 9.70 1572.87 2 0.285 8/53.33%
4 VLASKNFGDK 10 8.56 1078.23 1 0.163 6/60%
pI:isoelectricpoint.Mw:molecularweight.NC:netcharge.H:hydrophobicity.PR/n%:polarresiduesandratio
ofpolarresidues.
TheantibacterialactivitiesofHb-IandAI-hemocidinsareshowninTable3. Hb-Idemonstrated
adequateantibacterialactivityagainstE.coliandS.aureus. AmongtheAI-hemocidins,AI-hemocidin2
showedthehighactivityagainsttheteststrains,withtheexceptionofEnterococcusfaecalis. S.aureus
exhibited the highest sensitivity to AI-hemocidin 2 with a MIC value of 37.5 µg/mL (22.77 µM).
ComparedwithHb-I,theantibacterialeffectofAI-hemocidin2wasenhanced. AlthoughHb-Iwas
initiallyscreenedwithE.coli,theantibacterialactivityofAI-hemocidin2againstseveralGram-positive
bacteriawasobserved. TheresultssuggestedthatAI-hemocidin2wasanantibacterialpeptidewith
abroadantimicrobialspectrum. Inaddition,AI-hemocidin1displayedantibacterialactivityselectively
against Gram-negative bacteria Pseudomonas aeruginosa and E. coli with a MIC value of 75 µg/mL
(47.35µM).Peptidesderivedfromtheerythrocytesofvariouslivingorganismshavealsobeenreported
toexhibitantibacterialactivity. Recently,hydrolytefractionsofbovinehemoglobinwerereportedto
haveantibacterialactivityagainstbacteriainvasion. Anovelhemoglobinpeptide,P3derivedfrom
bovineerythrocytes,assumedanα-helicalconformationandexhibitedmodestantimicrobialactivity
invitro. P3killedbacteriarapidlybydisruptingthebacterialcytoplasmicmembraneanddisturbing
theintracellularcalciumbalance. Hemoglobin-relatedpeptidesaretherebybecominganewsourceof
antibacterialagentsthatarecapableofinhibitingpathogenicbacteria.
Table3.AntibacterialactivitiesofHb-IandAI-hemocidins.
MICValue(µM)
Bacteria
1 2 3 4 Hb-I Cipro
E.coliATCC25922 47.35 45.54 >381.47 >556.47 75.66 24.14
P.aeruginosaATCC27853 47.35 91.08 95.37 >556.47 >100.88 48.28
P.aeruginosaclinicalisolate 47.35 91.08 >381.47 >556.47 >100.88 48.28
S.aureusATCC25923 >378.83 22.77 >381.47 >556.47 75.66 24.14
B.subtilisATCC6633 >378.83 182.16 >381.47 69.55 >100.88 12.07
B.subtilisATCC6633 >378.83 182.16 >381.47 >556.47 >100.88 12.07
E.faecalisATCC29212 >378.83 >364.33 >381.47 >556.47 >100.88 24.14
Eachassaywasperformedintriplicate.Cipro=Ciprofloxacin.
2.4. CircularDichroismSpectraand3DSecondaryStructurePrediction
ThesecondarystructureofAI-hemocidin2wasinvestigatedindifferentenvironments,including
10mMPBSsolutionand50%trifluoroethanolsolutionbycirculardichroism(CD)spectroscopy[43].
As illustrated in Figure 4A, peptides displayed random coil conformations in PBS. However, the
spectraofAI-hemocidin2featuredanα-helixconformationinthepresenceof50%TFE,whichwas
indicatedbythepresenceoftwonegativedichroicbandsat208and222nm[44]. TheresultsoftheCD
spectruminAI-hemocidin2wereaccordanttothesecondarystructurepredictedbythehelicalwheel
projection(Figure3B).ThemodelingstructureofAI-hemocidin2wasproposedbyIterativeThreading
ASSEmblyRefinement(ITASSER)(Figure4B).Thestructureoftheantibacterialpeptidesallowsfor
theirinsertionintobacterialmembranesformingavoltage-dependentionchannel,thereafteraltering
thepermeabilityofthebacterialcell,leadingtodisruptionofcytoplasmicprocessesandtargetcell
lysis[11]. Thehighlevelofα-helixinAI-hemocidin2contributedtothesensitiveantibacterialactivity.
Mar. Drugs 2017, 15, 205  7 of 17 
2.5. Assays for Hemolysis and Cytotoxicity 
AI‐hemocidin 2 showed low hemolytic activity on murine red blood cells. The hemolysis of 
AI‐hemocidin 2 reached a maximum of 10.27%, even at a high concentration of 500 μg/mL, which 
was  13  times  higher  than  its  MIC  for  S.  aureus  (Table  3).  The  HEK293  cell  viability  was 
approximately 88% after treatment with AI‐hemocidin 2 at 250 μg/mL. The half maximal inhibitory 
concentration of cell viability was larger than 1000 μg/mL (Table 4). The cytotoxic concentrations for 
HEK cells were higher than the MIC values. These results indicated that AI‐hemocidin 2 displayed 
excellent bacteria selectivity (Table 4). Piscidin 1, a peptide purified from hybrid striped bass, 
possesses significant antibacterial action as well as cytotoxic activity [45]. The mechanism of action 
of its bactericidal and fungicidal effects was operated via targeting and disrupting the bacterial and 
fungal membranes [46,47]. However, the application of piscidin 1 has been limited on account of its 
cytotoxic effects. On the contrary, hemoglobin‐derived peptides exhibit antibacterial activity with low 
cytotoxicity. SHβAP and P3 were both hemoglobin‐related antibacterial peptides derived from the 
liver of skipjack tuna and bovine, respectively [25,45]. SHβAP and P3 both showed low hemolytic 
activity and cytotoxicity, although their mechanisms of antibacterial action were similar to Piscidin 1. 
The finding of low cytotoxicity and low hemolytic activity of AI‐hemocidin 2 is comparable to SHβAP 
and P3. The antimicrobial activity of AI‐hemocidin 2 was not greatly affected by pH or heat treatment 
(Figure 5). As shown in Figure 5, compared with 0.01% HAc (pH 3) and sterile water (pH 7), the 
antibacterial activity of AI‐hemocidin 2 was significantly decreased (p < 0.05) in 0.01% NaOH (pH 10), 
but still retained 60.76% antibacterial activity. The net charge of AI‐hemocidin 2 (p.I. 8.5) was negative 
in 0.01% NaOH (pH 10) and this negatively charged state was not optimal for the association of 
AI‐hemocidin 2 with the negatively charged bacterial membrane for antibacterial activity. The 
binding ability of AI‐hemocidin 2 would have been decreased in 0.01% NaOH (pH 10), thus 
reducing the antibacterial activity. However, various amino acids display different isoelectric points 
and different charges in different solutions. AI‐hemocidin 2 contained three aspartic acids at the 
hydrophilic side. The net charge of the hydrophilic portion with aspartic acids might be positive in 
0.01% NaOH (pH 10), providing the possibility of an interaction between AI‐hemocidin 2 and the 
negatively charged bacterial membrane. Therefore, AI‐hemocidin 2 retained antibacterial activity 
Mar.Drugs2017,15,205 7of17
even in 0.01% NaOH (pH 10) showing its potential as a candidate antibacterial agent. 

FFiigguurree 44.. SSeeccoonnddaarryys tsrturcutcutruerech cahraacrtaecrtiezraitziaotnioonf AofI -hAeIm‐hoecmidoincid2:in(A 2): ci(rAcu) lacirrdcuiclharr odisimch(rCoiDsm)s p(eCcDtr)a 
sopfeActIr-ah eomf oAciId‐hinem2;oacniddi(nB )23; Danstdr u(cBtu) r3aDlp rsetrduiccttiuornalo fpAreI-dhiecmtioonc idoifn A2.I‐Thheemcoocloidrsina r2e. mTehaen icnoglloersss aanred 
monelayniunsgeldestso apnedrf oonrmly tuhseedse tcoo npderafroyrmstr tuhcet usreecoinndaanriyc es-tlrouocktuinrge iwn aay .nice‐looking way. 
2.5. AssaysforHemolysisandCytotoxicity
AI-hemocidin 2 showed low hemolytic activity on murine red blood cells. The hemolysis of
AI-hemocidin2reachedamaximumof10.27%,evenatahighconcentrationof500µg/mL,which
was13timeshigherthanitsMICforS.aureus(Table3). TheHEK293cellviabilitywasapproximately
88%aftertreatmentwithAI-hemocidin2at250µg/mL.Thehalfmaximalinhibitoryconcentration
ofcell viabilitywas largerthan 1000 µg/mL(Table 4). The cytotoxicconcentrations forHEKcells
werehigherthantheMICvalues. TheseresultsindicatedthatAI-hemocidin2displayedexcellent

bacteria selectivity (Table 4). Piscidin 1, a peptide purified from hybrid striped bass, possesses
significant antibacterial action as well as cytotoxic activity [45]. The mechanism of action of its
bactericidal and fungicidal effects was operated via targeting and disrupting the bacterial and
fungal membranes[46,47]. However, the application of piscidin 1 has been limited on account of
itscytotoxiceffects. Onthecontrary,hemoglobin-derivedpeptidesexhibitantibacterialactivitywith
lowcytotoxicity. SHβAPandP3werebothhemoglobin-relatedantibacterialpeptidesderivedfrom
theliverofskipjacktunaandbovine,respectively[25,45]. SHβAPandP3bothshowedlowhemolytic
activityandcytotoxicity,althoughtheirmechanismsofantibacterialactionweresimilartoPiscidin1.
ThefindingoflowcytotoxicityandlowhemolyticactivityofAI-hemocidin2iscomparabletoSHβAP
andP3. TheantimicrobialactivityofAI-hemocidin2wasnotgreatlyaffectedbypHorheattreatment
(Figure 5). As shown in Figure 5, compared with 0.01% HAc (pH 3) and sterile water (pH 7), the
antibacterialactivityofAI-hemocidin2wassignificantlydecreased(p<0.05)in0.01%NaOH(pH10),
butstillretained60.76%antibacterialactivity. ThenetchargeofAI-hemocidin2(p.I.8.5)wasnegative
in 0.01% NaOH (pH 10) and this negatively charged state was not optimal for the association of
AI-hemocidin2withthenegativelychargedbacterialmembraneforantibacterialactivity. Thebinding
ability of AI-hemocidin 2 would have been decreased in 0.01% NaOH (pH 10), thus reducing the
antibacterialactivity. However,variousaminoacidsdisplaydifferentisoelectricpointsanddifferent
chargesindifferentsolutions. AI-hemocidin2containedthreeasparticacidsatthehydrophilicside.
Thenetchargeofthehydrophilicportionwithasparticacidsmightbepositivein0.01%NaOH(pH10),
providingthepossibilityofaninteractionbetweenAI-hemocidin2andthenegativelychargedbacterial
membrane. Therefore,AI-hemocidin2retainedantibacterialactivityevenin0.01%NaOH(pH10)
showingitspotentialasacandidateantibacterialagent.
Mar.Drugs2017,15,205 8of17
Mar. Drugs 2017, 15, 205  8 of 17 
Table 4. Hemolysis activity and cytotoxicity of AI‐hemocidin 2. 
Table4.HemolysisactivityandcytotoxicityofAI-hemocidin2.
Sample  AI‐Hemocidin‐2
MIC a (μSga/mmpLl)e  AI-H37em.5–o3c0id0i n-2
HC1M0 b I(Cμga/(mµgL/) mL) 37>.55–0300 0
HHCm10ax bc (µg/mL) 10.2>75 0±0 0.42 
Cell viability (%H) amt a2x5c0 μg/mL  1808..2178 ±± 90..4428 
Cellviability(%)at250µg/mL 88.18±9.48
IC50I dC(μgd/m(µLg/) mL) >>11000000 
SI v5a0lue e  3.33–26.67 
SIvaluee 3.33–26.67
Note:  a Minimum inhibitory concentration of AI‐hemocidin 2 against E. coli, P. aeruginosa, P. 
Note: aMinimuminhibitoryconcentrationofAI-hemocidin2againstE.coli,P.aeruginosa,P.aeruginosaclinical
isoaleartue,gSin.oasuar eculisn,iSc.aelp iisdoelramteid, iSs.a anudreBu.ss,u Sb.t ielpisi.debrmHiCd1i0s aisntdh eBc. osunbcteinlitsr.a bt iHonCo10f ips etphtei dceoncacuensitnrgat1io0n% ohfe pmeopltyidsies on
hucmaaunsienrgyt 1h0ro%cy hteesm.coHlymsiasx iosnt hheupmeracnen etraygteh(r%o)chyetemso. lcy Hsismaaxt itsh ethheig pheersctetensttaegdep (e%pt)i dheemcoonlcyesnitsr aatti otnhe(5 h00igµhge/smt L).
dItCe5s0teisdd pefienpetiddaes cthoenccoennctreanttiroanti o(n50a0t wμhgi/cmh5L0)%. do IfCg5r0o wist hdiesfiinnheidb itaesd .theeS ecloecntciveintytriantdioenx (aint vwithroic)h:I C5500%in oHf EK
p29re3gsrceoenwltlestd/hM aissI Ctihn,ehICmib5ei0taetnadk±. ee nSSDeals.eactmiviintyim inudme,x1 (0i0n0 vµitgr/om): LIC.A50 sisna HysEwKe 2re93p ecreflolsr/mMeIdCi,n ICtr5i0p tlaickaeten, aasn da mthienriemsuulmts, are
1000 μg/mL. Assays were performed in triplicate, and the results are presented as the mean ± SD. 

FigFuigruer5e. 5S. tSatbaibliitliytyo offt htheea anntitbibaacctteerriiaall aaccttiivviittyy ooff AAII‐-hheemmooccididinin 22 aaggaianinsts Et.E c.oclio ldiudruinrgin cghcahnagnesg eins pinHp H
and heat, ** p < 0.05. 
andheat,**p<0.05.
2.6. Effects of AI‐Hemocidin 2 on the Permeability of the Bacterial Membrane 
2.6. EffectsofAI-Hemocidin2onthePermeabilityoftheBacterialMembrane
Increasing the permeability of the bacterial membrane is a possible mechanism by which 
Increasing the permeability of the bacterial membrane is a possible mechanism by which
peptides exert antibacterial function. Outer membrane permeabilization by AI‐hemocidin 2 was 
peptides exert antibacterial function. Outer membrane permeabilization by AI-hemocidin 2 was
assessed by an NPN uptake assay. Upon destabilization of the cellular membrane by AMPs, NPN 
assessedbyanNPNuptakeassay. UpondestabilizationofthecellularmembranebyAMPs,NPN
inserted into the damaged membrane and emitted a strong fluorescence signal in the hydrophobic 
insertedintothedamagedmembraneandemittedastrongfluorescencesignalinthehydrophobic
interior of the cell membrane [48]. As illustrated in Figure 6A, 45.54 μM (MIC) of AI‐hemocidin 2 
intcearuioserdo 5f7t.h2%e cNePllNm uepmtabkrea nbye E[4. 8c]o.li Aats 36il0lu ss atrftaetre dsaminpFlei gtrueraetm6Aen,t4, w5.5h4erµeaMs 1(3M6.I6C2 )μoMf A(3I ×-h MemICo) coifd in
2 AcaIu‐hseemdo5ci7d.2in% 2 NinPdNuceudp htaigkheerb yNPEN.  cuoplitaakte 3(8610.6s6%a)f.t eTrhes armespullets tirnedaitcmateendt ,thwath AerI‐ehaesm1o3c6id.6in2 2µ M
(3w×aMs ICca)poabfleA I-ohfe mpeorcmideianbi2liziinndgu ctehde hoiguhteerr  NmPemNburapntea kein( 8a1 .6d6%os)e.‐dTepheendreesnut ltsmianndniecra. teTdhteh at
AIp-heremmeoacbiidliiznat2iown aosf ctahpea ibnlneeor fcpyetorpmlaesambiilci zminegmtbhreanoeu twerasm memeabsurarnede ibny aa dβo‐sgea-ldacetpoesniddaesne tamssaanyn. er.
ThOe‐npietrrompehaebniylilz‐βa‐tDio‐gnaloafctothpeyriannnoesridcey t(oOpNlaPsGm) icis mheymdrborlyazneed winatso moe‐naisturorpedhenboyl ainβ r-gesaplaocntsoes itdoa se
assβa‐gy.alOac-tnoistirdoapshee rneylel-aβse-Dd -fgraolmac ttohpe ycryatnoposlaidsme (wOhNePnG th)eis chyytodprloalsymzeicd minetmobor-anniter oisp hpeernmoleainblree s[4p9o]n. se
toTβh-egraelfaocrteo, siOdNasPeGr elweaass edchforsoemn tahse  cay tporpolbaes mtow  dheetnermthiencey  tthoep lapsemrmiceambeilmityb roafn ethise pceyrtmopelaabslmei[c4 9].
Thmereemfobrrea,nOeN bPy Gspwecatsrocphhoosetonmaestrayp arto b4e05t ondme.t eβr‐mgailnaecttohseidpaeserm leeaakbaigliet ywoafst hoebsceyrtvoepdl afsrommic Sm. eaumrebursa ne
by(Fspigeucrtreo 6pBh)o atnodm Eet. rcyolai t(F4i0g5unrem 6.Cβ) -igna tlahcet opsriedsaensecele oafk AagI‐ehwemasocoibdsiner 2v eind afr doomseS‐. aanudre tuims(eF‐idgeupreen6dBe)nat nd
manner. There was no change over time in the control cell medium. As shown in Figure 6B,C, the 
E.coli(Figure6C)inthepresenceofAI-hemocidin2inadose-andtime-dependentmanner. Therewas
absorbance values with 3 × MIC of AI‐hemocidin 2 were not much higher than the values with 1 × 
nochangeovertimeinthecontrolcellmedium. AsshowninFigure6B,C,theabsorbancevalueswith
MIC of AI‐hemocidin 2. This may be because the 1 × MIC concentration permeabilized the bacterial 
3×MICofAI-hemocidin2werenotmuchhigherthanthevalueswith1×MICofAI-hemocidin2.
cytoplasmic membrane to a sufficient degree and that the 3 × MIC concentration just accelerated this 
Thismaybebecausethe1×MICconcentrationpermeabilizedthebacterialcytoplasmicmembraneto
process. In addition, the absorbance values measured by a microplate reader (Figure 6B,C) showed 
asufficientdegreeandthatthe3×MICconcentrationjustacceleratedthisprocess. Inaddition,the
little difference between the 1 × MIC and 3 × MIC treatments. As a result of its positive net charge and 
absorbancevaluesmeasuredbyamicroplatereader(Figure6B,C)showedlittledifferencebetween
the 1 × MIC and 3 × MIC treatments. As a result of its positive net charge and α-helix structure,
Mar. Drugs 2017, 15, 205  9 of 17 
αM‐haer.lDixr usgtsru20c1t7u,r1e5,, A20I5‐hemocidin 2 possesses bacterial membrane permeability activity. AI‐hemocid9inof 21 7
Mar. Drugs 2017, 15, 205  9 of 17 
functions in a similar manner to bactenecin, permeabilizing the bacterial membrane in a rapid, reliable 
way  [49].  The  results  of  our  experiments  suggested  that  AI‐hemocidin  2  could  increase  the 
AI-αh‐ehmeloixc isdtriunc2tupreo,s AseI‐shseemsboacicdtienr i2a lpmosesemssbersa bnaectpeeriraml meaebmilbitryanaec tpievrimtye.aAbiIl-ihtye macoticviidtyin. A2I‐fhuenmctoicoindsini n2 a
permeability of the cytoplasmic membrane of pathogens, thereby exerting antibacterial activity. 
simfiulanrctmionans nine ra tsoimbialacrt emnaencnine,r ptoe brmaceteanbeicliizni,n pgertmheeabbailcitzeinriga lthme ebmacbterraianle minemabrraapnied i,nr ae lriaapbilde, wrealiyab[l4e9 ].
Thewraeys u[l4ts9]o. fTohuer erxepsuelrtism  oefn tsousur gegxepsetreidmtehnatst AsIu-ghgeemstoecdi dtihna2t cAoIu‐hldeminoccriedainse  2th ecopuelrdm  ienacbrielaitsye otfhteh e
cytopperlamsemaibcilmitye mofb trhaen ceyotofpplaasthmoigc emnesm,tbhrearnebe yofe pxearthtionggenans,t itbhaercetebryi aelxaecrttiinvgit ya.ntibacterial activity. 

Figure  6.  Membrane  permeabilizing  activity  of  AI‐hemocidin  2:  (A)  the  outer  membr ane 
permeabilization of E. coli by AI‐hemocidin 2, NPN uptake (%) is relative to polymyxin B, a positive 
Figure6.MembranepermeabilizingactivityofAI-hemocidin2:(A)theoutermembranepermeabilization
Figure  6.  Membrane  permeabilizing  activity  of  AI‐hemocidin  2:  (A)  the  outer  membrane 
control with a strong outer  membrane permeabilizing property, ** p < 0.01; (B) cytoplasmic 
ofEp.ecromliebaybAiliIz-ahteiomno ocifd Ein. c2ol,iN bPy NAIu‐hpetamkoec(i%di)ni s2,r eNlaPtNiv euptotapkoel y(%m)y ixs irnelBa,tiavpe otosi tpivoleycmoynxtrinol Bw, iat hpoassittirvoen g
membrane  permeabilization  of  S. aureus  by  AI‐hemocidin  2;  and  (C)  cytoplasmic  membrane 
outceornmtreoml bwriatnhe ap esrtmroenagb ioliuzitnerg pmreompebrrtayn,e* *ppe<rm0e.0a1b;il(iBzi)ncgy tporpolpaesmrtyic, m** emp b<r a0n.0e1p; e(rBm) ecaybtiolipzlaatsimonico f
permeabilization of E. coli by AI‐hemocidin 2. Each assay was performed in triplicate, and the results 
S.aumreeumsbbryanAeI -hpeemrmoceiadbiinliz2a;atinodn (Cof) cSy.t oapulraesums icbmy eAmIb‐hraemneopciedrmin ea2b; ilaiznadt io(nC)o fcEy.tcooplilabsymAicI -hmeemmobcriadnine 2.
arEea cpphreearsmsesneatayebdwil iaazssa ttpihoeenr f moofre mEan.e c d±ol iiS nbDyt.r  AipIl‐ihceamteo,cainddint h2.e Eraecshu latsssaarye wparse speenrtfeodrmaesdt hine tmriepalinca±te,S aDn.d the results 
are presented as the mean ± SD. 
2.7. Propidium Iodide Uptake Experiment 
2.7. PropidiumIodideUptakeExperiment
2.7. Propidium Iodide Uptake Experiment 
Propidium iodide (PI), a DNA dye that passes through damaged cell membranes rather than 
Propidiumiodide(PI),aDNAdyethatpassesthroughdamagedcellmembranesratherthan
integratinPgro ipnitdoi uthme  icoedlild me e(PmI)b, raa nDeN, wA adsy ue stehda ta ps aas sseesn tshitriovueg ihn ddaicmataogre dto c ienllv mesetmigbatrea nceesl lr matheemr btrhaanne  
integratingintothecellmembrane,wasusedasasensitiveindicatortoinvestigatecellmembrane
integinritteyg raantdin gce ilnl tdoe tahteh  cbeyll f lmowem cbyrtaonme,e wtrays [ 5u0s]e.d I na st hae s aebnsseitnivcee  oinfd piceapttoidr eto,  tihnev aesmtioguatnet  coefl lP mI uemptbarkaen ien  
integrityandcelldeathbyflowcytometry[50]. Intheabsenceofpeptide,theamountofPIuptake
E. coilni teagnrdit yS a. nadu rceeulsl  dceealltsh  wbya fsl olwow c,y tionmdiectaryti n[5g0 ]a. nIn  itnhtea catb sceenllc em oef mpebprtaindee.,  tEh.e  caomli oaunndt  oSf .P aIu urpeutask cee ilnls  
in E. coli and S. aureus cells was low, indicating an intact cell membrane. E. coli and S. aureus
treatEe.d c owlii tahn d1  ×S .M auICre uosr  c3e l×ls M wIaCs  AloIw‐h, eimndoicciadtiinng 2  adni sipnltaaycte dce dlli smtienmctblyra innec.r eEa. sceodli  laenvde lSs . oafu PreIu us pcteallkse  
cells treated with 1 × MIC or 3 × MIC AI-hemocidin 2 displayed distinctly increased levels of
relattirveea tteod c weliltsh i n1c ×u bMaItCed o irn 3 t h× eM aIbCs eAnIc‐eh eomf AocIi‐dhienm 2o dciisdpinla y2 e(dF idgiusrtien 7ct)l.y A i ndcoreseas‐dedep leevnedles notf  iPnIc ruepatsaek ien  
PIuptakerelativetocellsincubatedintheabsenceofAI-hemocidin2(Figure7). Adose-dependent
PI flrueolaretisvcee ntoc ec eilnlsd iinccautebda ttehda itn A thI‐eh aebmsoencicde ionf  2A wI‐haesm caopciadbilne  2o (fF digaumrea g7i)n. Ag  tdhoes eE‐.d ceoplie nandden St .i naucrreeausse c ienl l 
increase in PI fluorescence indicated that AI-hemocidin 2 was capable of damaging the E. coli
PI fluorescence indicated that AI‐hemocidin 2 was capable of damaging the E. coli and S. aureus cell 
membranes.  Cell  integrity  damage  assays  were  carried  out  concurrently  with  membrane 
and S. aureus cell membranes. Cell integrity damage assays were carried out concurrently with
membranes.  Cell  integrity  damage  assays  were  carried  out  concurrently  with  membrane 
permeabilization tests of antibacterial agents to establish whether these two effects correlated. For 
membrane permeabilization tests of antibacterial agents to establish whether these two effects
permeabilization tests of antibacterial agents to establish whether these two effects correlated. For 
example, the proline‐rich peptide Bac7(1‐35) rapidly killed E. coli at micromolar concentrations, 
correlated. For example, the proline-rich peptide Bac7(1-35) rapidly killed E. coli at micromolar
example, the proline‐rich peptide Bac7(1‐35) rapidly killed E. coli at micromolar concentrations, 
while PI remained outside the cells [50]. These results indicated that AI‐hemocidin 2 was capable of 
conwcehnilter aPtiIo rnems,awinheidle oPuItsriedme athinee cdelolsu [t5si0d].e Tthheescee rlelssu[5lt0s] i.nTdhiceasteedre tshualtt sAiIn‐hdeicmaotecdidtihn a2t wAaI-sh ceampaobclied oinf 2
damaging cell integrity and improving the permeability of the cell membrane. 
wasdcaampaagbilnego cfedlla imntaeggrinitgy caenldl iinmtepgrorivtiynagn tdhei mpeprrmoevainbiglittyh eofp tehrem ceeallb milietmyobrfatnhee. cellmembrane.

FFigiguFureirge 7u7.r .Ee Ef7ff.ef Ecetcfsfte socoft sfA oAIf‐ IhA-heIme‐hmoeomcicdoidciniidn 2in2 o 2on n otnhth eth emem emmeembmbrbarrananen eieni intnettgeegrgirrtiiytty yo oof ffd ddifiifffeffeerrereennntt tb bbaaacctcteteerririaaiall  lsststrrtaaraiinninss..s  T.ThThehe  eddadatataat a 
show the MFI (mean fluorescence intensity). * p < 0.05 and ** p < 0.01 versus control, and each assay 
shshooww ththe eMMFFI I(m(meaeann flfluuoorerescsecenncec einintetennsistiyty).) .* *p p<< 0.00.50 5anandd *** *p p<< 0.00.10 1vveresrusus scocnontrtorlo, la,nandd eaecahch asassasyay 
was performed in triplicate. The results are presented as the mean ± SD. 
wwaas sppeerfroformrmeedd inin trtirpiplilcicaatete. .TThhee reresusultlst saarere ppreresseenntetedd aass ththee mmeeaann ±± SDSD. .
Mar.Drugs2017,15,205 10of17
Mar. Drugs 2017, 15, 205  10 of 17 
22..88.. EExxaammiinnaattiioonn ooff tthhee MMoorrpphhoollooggiicc CChhaannggeess iinn CCeellllss bbyy SSccaannnniinngg EElleeccttrroonn MMiiccrroossccooppyy 
DDiirreecctt vviissuuaalliizzaattioionno foAf IA-hIe‐hmeomciodciindi2n- in2d‐iuncdeudcEed.c oEli. acnodli Sa.nadu reSu.s acuerlleuulsa rcdelalmulaagr edwaamsaagses eswseads 
absysescssaendn inbgye lescctarnonnimngic roelseccotproyn( SEmMic).roAsIc-ohpemy o(cSidEiMn2). trAeaIt‐mheemntoicnidduinc ed2 cetlrlematomrpehnot loignidcaulccehda ncgeelsl .
mCoomrpphaorleodgiwcaitlh cthhaencgoenst. roClo,m1×paMreIdC wAiIt-hh etmheo cciodnintr2olt,r e1a t×m MenItCf oArI1‐hheimndoucicdeidn p2a rttrieaaltdmaemnat gfeoor f1t hhe 
iSn.dauucreeuds pmaretmiabl rdaanme,agane dofs tehvee rSe. amueremubs rmanemewbrrainnke,l ianngda snedvetrhee mreelmeabsreaonfec welrliunlkalrincogn atnedn ttshoec rceuleraresde 
owf icthelilnu3lahr coofntrteenattms oecnctu(rFriegdu rwei8thAin–C 3) .h of treatment (Figure 8A–C). 

FFiigguurree 88.. SSccaannnniinngge elelectcrtoronnm micricorgorgarpahpshosf Sof. aSu.r eauusreaunsd aEn.dco lEi.t rceoalit etdrewatiethd AwI-ihthe mAoIc‐hideimno2:ci(dAi)nc o2:n t(rAo)l ,
cSo.nautrroelu, sSw. aiuthroeuust wthiethpoeuptt itdhee; p(Bep)tSi.daeu; r(eBu)s Str. eaautreedusw tirtehatAeId- hwemitho cAidI‐ihne2maotc1id×inM 2I aCt f1o ×r 1MhI;C( Cfo)rS 1. ahu; r(eCu)s 
Str. eaautreedusw tritehatAedI- hweimtho AciId‐ihnem2aotci1d×in M2 aICt 1f o×r M3IhC; (fDor) 3c ohn; t(rDo)l, cEo.nctorloilw, Eit.h coolui twpiethpotiudte p;(eEp)tiEd.ec;o (lEi)t rEe.a ctoeldi 
twreiathteAd Iw-hiethm AocIi‐dhienm2oactid1in× 2M aItC 1 f×o rM1IhC; faonrd 1( hF); aEn.dco (liFt)r Eea. tceodli wtrietahteAdI -wheitmh oAcIid‐hinemato1ci×dinM aItC 1f o×r M3IhC. 
for 3 h. 
TheeffectsofAI-hemocidin2onE.coliareshowninFigure8D–F.Asmoothandintactmembrane
The effects of AI‐hemocidin 2 on E. coli are shown in Figure 8D–F. A smooth and intact 
wasexhibitedinuntreatedE.colicells. Bycontrast,membranesurfaceroughnessandcorrugationwas
membrane was exhibited in untreated E. coli cells. By contrast, membrane surface roughness and 
observedafter1hoftreatmentwith1×MICAI-hemocidin2(Figure8E),andrupturedmembranes
corrugation was observed after 1 h of treatment with 1 × MIC AI‐hemocidin 2 (Figure 8E), and 
andleakageofcellularcontentswereobservedfollowing3hoftreatmentwith1×MICAI-hemocidin2
ruptured membranes and leakage of cellular contents were observed following 3 h of treatment with 
(Figure8F).TheseresultssuggestedthatAI-hemocidin2iscapableofdestroyingcellularintegrity
1 × MIC AI‐hemocidin 2 (Figure 8F). These results suggested that AI‐hemocidin 2 is capable of 
and inducing significant deformation in cellular morphology. The cell surface changes induced
destroying cellular integrity and inducing significant deformation in cellular morphology. The cell 
by AI-hemocidin 2 were similar to those previously reported [31,51]. Peptides interact with the
surface changes induced by AI‐hemocidin 2 were similar to those previously reported [31,51]. 
cellularmembraneviabindingtonegativelychargedconstituentsontheoutermembrane,suchas
Peptides interact with the cellular membrane via binding to negatively charged constituents on the 
lipopolysaccharide (LPS), which is followed by membrane destabilization [52]. In addition, it has
outer membrane, such as lipopolysaccharide (LPS), which is followed by membrane destabilization 
previouslybeenproposedthattheα-helicalconformationcontributestotheantibacterialactivityby
[52]. In addition, it has previously been proposed that the α‐helical conformation contributes to the 
disruptingmembraneintegrity[41].
antibacterial activity by disrupting membrane integrity [41]. 
3. MaterialsandMethods
3. Materials and Methods 
3.1. MaterialsandStrainCultureConditions
3.1. Materials and Strain Culture Conditions 
A.inflatamaterialswerepurchasedfromChengyangseafoodmarket,Qingdao,China,andwere
identAifi. eindflbayta Rmoantgemriainls Ywue(rJein panurUchnaisveedrs iftryo,mG uCahnegnzghyoaun,gC sheianfao)o.dT hmeamrkaent,t lQeimngadssaow, aCshdiinsas,e catnedd, 
wweerigeh ieddeanntidfisetdo rebdy aRto−n2g0m◦Cin. DYEuA E(JiSneapnh aUronsieveFrassittyF,l oGwuaanndgzShPoSue,p hCahrionsae)F. aTshtFe lomwawntelere mpuarscsh awseads 
dfriossmecGteEd, Hweeailgthhecdar ea.ndT rsitso,reSdD Sa,t C−2o0o m°Ca.s sDieEAbrEi lSlieapnhtabrlousee RFa-2s5t 0F,lodwim aenthdy SlPs uSlefpoxhiadreos(eD FMasStO F)l,oPwI, 
wNePrNe ,ppuorlcyhmasyexdin frBo,mO NGPEG H,e3a-(l4th,5c-adriem. eTtrhiys,l -S2D-tSh,i aCzooolyml)a-2s,s5ie-d biprihlleinanytl- b2Hlu-et eRtr‐a2z5o0l,i udmimbertohmyli dsuel(fMoxTidTe), 
(sDtrMepStOom), yPciIn, ,NanPdNp, epnoiclyilmlinyxwiner eBo, bOtaNinPeGd, fr3o‐(m4,5S‐igdmimaeCthhyelm‐2i‐ctahliCazoo.l(ySlt).‐2L,o5u‐disip,MheOn,yUl‐S2AH)‐.teRtPraMzoI-l1iu64m0, 
bfertoamlbidove in(MesTeTru),m st(rFeBpSto)manydciLnB, barnodt hpweneirceilpliunr cwhaesreed ofbrotaminGedIB fCroOmIn Sviigtrmogae CnhCeomrpicoarla tCioon. ((SSatn. LDoiuegiso,, 
MO, USA). RPMI‐1640, fetal bovine serum (FBS) and LB broth were purchased from GIBCO 
Invitrogen Corporation (San Diego, CA, USA). Ciprofloxacin was purchased from Guangzhou