Wound Healing Activity and Mechanisms of Action of an Antibacterial Protein from the Venom of

16 Pages·2014·4.8 MB·English

Wound Healing Activity and Mechanisms of Action of an
Antibacterial Protein from the Venom of the Eastern
Diamondback Rattlesnake (Crotalusadamanteus)
Ramar Perumal Samy1,2*, Matheswaran Kandasamy3, Ponnampalam Gopalakrishnakone1,
Bradley G. Stiles4¤, Edward G. Rowan5, David Becker6, Muthu K. Shanmugam7, Gautam Sethi7,
Vincent T. K. Chow2
1VenomandToxinResearchProgramme,DepartmentofAnatomy,YongLooLinSchoolofMedicine,NationalUniversityofSingapore,Singapore,Singapore,2Infectious
DiseasesProgramme,DepartmentofMicrobiology,YongLooLinSchoolofMedicine,NationalUniversityofSingapore,Singapore,Singapore,3Infection&Immunity
Programme, Singapore Institute for Clinical Sciences, Agency for Science,Technology and Research, Brenner Centre for Molecular Medicine, Singapore, Singapore,
4IntegratedToxicologyDivision,USArmyMedicalResearchInstituteofInfectiousDiseases,FortDetrick,Maryland,UnitedStatesofAmerica,5StrathclydeInstituteof
PharmacyandBiomedicalSciences,UniversityofStrathclyde,Glasgow,UnitedKingdom,6DepartmentofAnatomyandDevelopmentalBiology,University College
London, London, United Kingdom, 7Department of Pharmacology, Clinical Research Centre, Yong Loo Lin School of Medicine, National University of Singapore,
Singapore,Singapore
Abstract
BasicphospholipaseA wasidentifiedfromthevenomoftheeasterndiamondbackrattlesnake.TheCrotalusadamanteus
2
toxin-II (CaTx-II) induced bactericidal effects (7.8mg/ml) on Staphylococcus aureus, while on Burkholderia pseudomallei
(KHW), and Enterobacter aerogenes were killed at 15.6 mg/ml. CaTx-II caused pore formation and membrane damaging
effectsonthebacterialcellwall.CaTx-IIwasnotcytotoxiconlung(MRC-5),skinfibroblast(HEPK)cellsandinmice.CaTx-II-
treated mice showed significant wound closure and complete healing by 16days as compared to untreated controls
(**P,0.01). Histological examination revealed enhanced collagen synthesis and neovascularization after treatment with
CaTx-IIversus2%FusidicAcidointment(FAO)treatedcontrols.Measurementoftissuecytokinesrevealedthatinterleukin-1
beta(IL-1b)expressioninCaTx-IItreatedmicewassignificantlysuppressedversusuntreatedcontrols.Incontrast,cytokines
involvedinwoundhealingandcellmigrationi.e.,monocytechemotacticprotein-1(MCP-1),fibroblastgrowthfactor-basic
(FGF-b),chemokine(KC),granulocyte-macrophagecolony-stimulatingfactor(GM-CSF)weresignificantlyenhancedinCaTx-
IItreatedmice,butnotinthecontrols.CaTx-IIalsomodulatednuclearfactor-kappaB(NF-kB)activationduringskinwound
healing. The CaTx-II protein highlights distinct snake proteins as a potential source of novel antimicrobial agents with
significant therapeuticapplication forbacterial skin infections.
Citation:SamyRP,KandasamyM,GopalakrishnakoneP,StilesBG,RowanEG,etal.(2014)WoundHealingActivityandMechanismsofActionofanAntibacterial
ProteinfromtheVenomoftheEasternDiamondbackRattlesnake(Crotalusadamanteus).PLoSONE9(2):e80199.doi:10.1371/journal.pone.0080199
Editor:StefanBereswill,Charite´-UniversityMedicineBerlin,Germany
ReceivedJuly3,2013;AcceptedSeptember30,2013;PublishedFebruary14,2014
Copyright: (cid:2) 2014 Samy etal. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited.
Funding:TheauthorsthankDr.Naiduforprovidingcells(skinfibroblast)andtechnicalsupportinperformingthecytotoxicityassay.Theauthorsarethankfulto
theEconomicDevelopmentBoard(EDB),Singapore,forfinancialsupportbyProofofConcept(GrantNo:R-181-000-110-144).Thisworkwasalsosupportedin
partbyagrantfromMOETierItoDr.GautamSethi.Thefundershadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparationof
themanuscript.
CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist.
*E-mail:[email protected]
¤Currentaddress:DepartmentofBiology,WilsonCollege,Chambersburg,Pennsylvania,UnitedStatesofAmerica
Introduction infectedsurgicalwoundsandotherskinrelated infectionsinclude
methicillin [9], vancomycin [10], glycopeptides [11], and beta-
Antimicrobialproteinsandpeptidesplayamajorroleininnate lactams [12]. Following S. aureus infection, wound healing after
immunity by interacting directly with bacteria and killing them treatment with some of these antibiotics has been reported to be
[1]. These multifunctional proteins have wound healing activity onlymoderately successfulin manycases [13].
and receptor-mediated effects on eukaryotic cells [2]. Previously
However,theprevalenceofbacterialresistancetoconventional
reported mammalian proteinsandpeptidessuchasdefensins [3],
antibiotics has prompted an intensive search for new therapeutic
cathelicidins [4], and peptide LL-37 [5] can cause hemolysis of
agents including various antimicrobial peptides of animal origin
human erythrocytes and are cytotoxic at higher doses that might
[14]. Proteins/peptides with potent antimicrobial activity have
adverselyaffectwoundhealing[6],[7].Inaddition,emergencesof
beenfoundinawidevarietyoforganisms,includingsnakes.Their
multi-drug resistant (MDR) bacteria generated by specific
venom, particularly crotalidae venoms [15], is a rich natural
mutationsonantibiotictargets,andacquiredbytransposongenes
source for discovery and development of novel antimicrobial
[8], are one of the important healthcare concerns for long-term
agents.Crotapotin,asecretoryphospholipaseA isolatedfromthe
2
management.AntibioticsusedfortreatmentofStaphylococcusaureus
venomofCrotalusdurissusterrificus,showsantibacterialactivity[16]
PLOSONE | www.plosone.org 1 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
as well as antiviral activity against the human immunodeficiency California, USA). Protein concentrations were determined by
virus[17],[18].CrotamineisasmallbasicmyotoxinfromSouth Lowry’s method[29].
Americanrattlesnake(Crotalusdurissusterrificus)killedviamembrane
permeabilizationofEscherichiacoliwiththeMICsrangingfrom25– Phospholipase A (PLA ) Activity
2 2
100mg/ml at 1–2h and fail to show any hemolytic activity to The PLA activity of Crotalus adamanteus venom, fractions, and
2
erythrocytes. Crotamine antimicrobial peptides (AMP) structure purified proteins were determined as described [1], with slight
similartothatmanyclassofbeta-defensin[19].AcidicPLA2[20], modifications. Each of the above sample solutions (containing
both Asp49 and Lys49 PLA2 homologues [21], have previously 20mgproteinconcentration)of180mlwasaddedtothesubstrate
beenshowntopossessbactericidalactivity[22].Acationicprotein solution containing 5 mM Triton X 100, 5 mM phosphatidyl-
isolated from venom of the inland taipan (Oxyuranus microlepidotus) choline (Sigma Co, St. Louis, Missouri, USA), 2 mM HEPES,
selectively and dose-dependently kills Gram-positive bacteria
10mM calcium chloride (Merck & Co., Inc, New Jersey, USA)
through membrane disruption [23]. These molecules are very
plus 0.124% bromothymol blue dye (wt/vol) in water (pH 7.5).
attractive because of their biochemical diversity, broad spectrum
Afterincubatingfor10minat37uC,absorbancewasdetermined
of activity against enveloped viruses, bacteria, fungi, protozoa,
at 590nm in a microtiter spectrophotometer (Research Instru-
parasites,andpromotionofwoundhealing[25],[26].Theeastern
ment, Virginia, USA). Results were expressed in nmoles of acid
diamondbackrattlesnake(Crotalusadamanteus)belongstothefamily
per minute per mg ofvenom protein.
Viperidae (subfamily Crotalinae) and is widely distributed in the
United States [27]. Many venom enzymes from various species,
Mass Spectrometry, Amino Acid Composition, and
including PLA , hyaluronidase, L-amino acid oxidase, metallopro-
2 Sequencing Analyses
teinases, as well as myoprotein-CAM-protein and hemorrhagic
The molecular mass of CaTx-II was analyzed using matrix-
protein possess potent pharmacological effects such as myotoxic,
assistedlaserdesorptiontimeofflightmassspectrometry(MALDI-
neurotoxic and edematogenic activities [26]. However, relatively
very little is known regarding the antimicrobial activity of C. TOF/MS) in positive ionization mode. Alpha-cyano-4-hydroxy-
adamanteusvenom[15]oritscomponents.Theaimofthepresent cinnamicacidusedtoperformwithaVoyager-DESTR(Applied
finding was to evaluate the antimicrobial and wound healing biosystems, Farmingham, Massachusetts, USA). Amino acid
activity of a C. adamanteus venom protein (CaTx-II) in a mouse compositionandsequencinganalyseswereperformedasdescribed
model for S.aureus wounds. previously [1]. The terminal sequence of CaTx-II samples were
performed at the Protein and Proteomic Center, Department of
Biological Science, NUS, Singapore using a fully automated ABI
Materials and Methods
Procise 494 protein sequencer (Applied Biosystem,) system that
Venom employsEdman’sdegradationreactionforsequentialseparationof
Required quantities (10g) of the crude venom of Crotalus N-terminal aminoacids.
adamanteus(Viperidae:subfamilyCrotalinae)werepurchasedfrom
VenomSuppliesPteLtd,Tanuda,SouthernAustralia,andstored In vitro Antimicrobial Assays
desiccated at 4uC in individual containers at the Department of Bacterial cultures of Burkholderia pseudomallei (TES), Burk-
Anatomy, NUS, Singapore. holderia pseudomallei (KHW), Escherichia coli, Enterobacter
aerogenes, Klebsiella pneumoniae, Streptococcus pneumoniae,
Purification of Protein (CaTx-II) Proteusvulgaris,Proteusmirabilis,Pseudomonasaeruginosa,and
Crotalus adamanteus venom (1 g) was dissolved completely in Staphylococcus aureus were obtained from the Department of
10ml of 1 M Tris-HCl (pH 7.4), and after centrifugation at Microbiology, NUS, Singapore. Venom fractions and purified
18006g for 15min, the clear yellowish solution was sterilized by CaTx-II were tested for antimicrobial activity. Bacteria were
filtration through a 0.22mm membrane filter (Nalge Nunc grown in Mueller Hinton (MH) and Tryptic Soya (TS) broth
International, Rochester, New York, USA). Venom samples (Oxoid Limited, Basingstoke, Hampshire, UK) to exponential
(1 ml) diluted in 9 ml of Tris-HCl was then injected into a phase (0.5–1.56106 colony forming units (CFU per ml)) with an
Superdex G-75 column attached to a high performance liquid A600of0.8,representing12,000 cfu/ml.20mlofMHorTSagar
chromatography system (HPLC, AKTA GE Healthcare, Den- medium was poured into each petri dish (140 mm620mm)
mark). Eluents (1 ml fractions) were monitored at 280nm and obtained from Sterilin Limited, UK. After solidification, bacteria
collected in 1.5 ml eppendorf tubes. All the fractions were tested were swabbeduniformlyontotheagarsurface bya sterile cotton
for enzymatic and antibacterial effects, and the fraction that swab.20mlofeachtestsample(250 mg/ml)wasappliedtoasterile
showed the highest enzymatic and antibacterial effects were then blank disc (6 millimeter in diameter, BD Bioscience, New Jersey,
subjected to reverse-phased liquid chromatography (RP-HPLC) USA)placedonthesurfaceoftheMHandTSagarmedium,and
viaaC18column(4.66150mm,Phenomenex,Upsala,Sweden), incubated at 37uC for 24h. The blank disc containing sterile
using a linear gradient of aqueous acetonitrile (80%) in 0.1% water(20 ml)servedasanegativecontrol,whilethecommercially
trifluoroaceticacid(SigmaAldrichCo,StLouis,Missouri,USA)at available antibiotic discs of chloramphenicol (30 mg/disc) and
a flow rate of 1.0ml per min. Subsequent purification of active streptomycin (30 mg/disc) served as positive controls. Three
fractions were derived by using a C8 RP-HPLC column (100 A˚, replications were used for each assay and effects were recorded
4.66250mm) [1]. as a clear zone of inhibition, measured diameter in mm, after
incubation[1], [30].
Sodium Dodecyl Sulphate Polyacrylamide Gel
Electrophoresis (SDS-PAGE) In vitro Bactericidal Activities of CaTx-II
C. adamantetus RP-HPLC fractions, and the final purified MinimumInhibitoryConcentrations(MICs)ofCaTx-IIagainst
fraction of protein (CaTx-II), were analyzed by SDS-PAGE different bacteria were determined by the MH and TS broth
according to Laemmli [28]. The molecular weights of protein dilution method [31]. A single bacterial colony was used to
bands were determined using standard markers (Bio-Rad, inoculate each MH and TS broth (5 ml/tube) subsequently
PLOSONE | www.plosone.org 2 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
incubated at 37uC for 6–8h (exponential-phase cells) before cytotoxiceffects,CaTx-IIwasappliedtoculturedfibroblaststested
adjusting with sterile medium to a 0.5 McFarland turbidity at different concentrations (15.6–2000 mg/ml) in a 96 well plate.
standard. The adjusted bacterial cultures were diluted to Theassayplateswereincubatedat24and48htimeintervals.Cell
approximately 105–106CFU/ml before using in the MIC assays. proliferation was spectrometrically quantified using an ELISA
Bacteria(50 ml)wereintroducedintoeachwell(containing100ml platereaderat490nm(Bio-Rad,Hercules,California,USA).All
of medium) of a sterile 96-well plate, followed by addition of assays wereperformed in triplicate(n=3).
another 50ml of serially diluted CaTx-II (250, 125, 62.5, 31.25,
15.6, 7.8 and 3.9mg/ml). 100ml of medium containing bacteria Cytolysis Determination by Lactate Dehydrogenase
servedasacontrol,and100mlofmediumaloneservedasablank. (LDH) Release
Plateswereincubatedfor24hat37uC,andinhibitionofbacterial
The cytotoxicity of CaTx-II upon human cells (lung and skin
growth was determined by measuring turbidity at 560nm. Each
fibroblasts) was evaluated by measuring LDH release using a
MIC was determined from three independent experiments. The
cytotoxicity detection kit (Roche Mannheim, Germany). Cells
lowest concentrationsrequiredforinhibitingthebacterial growth (106) were cultured in 96-well micro titer plates with RPMI and
(MICs) were calculated by statistical analysis [32]. Minimum
DMEM medium (NUMI, Singapore) supplemented with 10%
bactericidalconcentrations(MBCs)weredeterminedasdescribed
(vol/vol) FBS and 1 ml of penicillin/streptomycin (1 x). CaTx-II
by Perumal Samy etal, [1]. Briefly, log-phase bacteria were
wasaddedintothreewellsperdose(15.6–2000mg/ml),whilecells
obtainedfromMHandTSbrothcultures,andtheirconcentration
withoutanytreatmentservedasacontrol.Thecellswerefurther
adjusted to 105–106CFU/ml. Bacterial suspensions (50ml) were
incubated at 37uC in the presence of 5% CO for 24 and 48h.
incubatedat37uCfor24hwith50mlofCaTx-II(250–3.9 mg/ml) 2
Afterincubation,theLDHreleasedintotheculturemediumfrom
and 100ml of MH and TS broth in a 96-well plate. To assess
eachwellwascollectedfollowingcentrifugationat10,0006g,4uC
bacterialviability,thesebrothcultureswereplatedonMHandTS
for5 min.A200mlaliquotwasusedforthequantificationofcell
agar and counted following 24h of growth at 37uC. The
death and lysis, based on LDH released from the cytosol of
percentage of bactericidal effects was calculated versus bacterial
damaged cells into the supernatant. The assay was performed in
control.
triplicate [34].
Scanning Electron Microscope (SEM)
Mice
BacterialcellsweregrowninMHandTSbroth(5mlpertube)
Pathogen free, 8–12 weeks old male Swiss Albino mice (body
to an exponential phase. Approximately 26106 CFU/ml were
weight 25–30gm) were obtained from the Animal Breeding
then centrifuged at 12,0006g for 10min at 4uC, washed twice
Center,Sembawang,Singapore.Theanimalsweremaintainedfor
with2mlof10mMsodiumphosphatebuffer,andresuspendedin
12hunderlight/darkconditions,andweregivenappropriatediet
2 mlofthesamebuffer.Thebacteriawereincubatedfor30min
plus water ad libitum. All experimental animals were approved by
withdifferentconcentrationsofCaTx-II,dependingontheirpre-
the ethics committee, NUS, Singapore (Approval number 055/
determinedMICvaluesforeachstrain(i.e.,7.8mg/mlforS.aureus
2007).
and B. pseudomallei (KHW), 15.6mg/ml for B. pseudomallei (TES)
and31.25 mg/mlforE.aerogenes,P.vulgaris,and62.5mg/mlforP.
In vivo Toxicity (Mice)
mirabilis.Afterincubation,eachsamplewasspreadontoapolyL-
lysine(SigmaAldrich,StLouis,Missouri,USA)coatedglassslide The toxic effect of purified CaTx-II was studied using a
toimmobilize bacteriaat 37uCfor24h.Thebacteriawerefixed modified method previously reported [1]. The toxic effect of
with 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer, rattlesnakevenomandCaTx-IIwasassessedinSwissAlbinomale
extensivelywashedwiththesamebuffer,anddehydratedwithan and female mice (1–14mg/kg, body weight) by intraperitoneal
ethanol gradient (50–100%). After critical-point drying and gold (i.p.) injection in 200ml of PBS. Six animals per dose were used,
coating, the samples were observed with a Philips XL 30 andthemortality recorded.
microscope[1].
Antibodies
Cell Culture Monoclonalorpolyclonalantibodies(anti-mousetypeIcollagen
Human lung (MRC-5) and skin fibroblast (HEPK) cells were and anti-rat CD68, Santa Cruz, California, USA) and secondary
purchasedfromtheAmericanTypeCultureCollection(Virginia, antibodies labeled with FITC (Sigma Co., St. Louis, Missouri,
USA), and the cell proliferation Kit II was from Roche Applied USA) or Cy3 (Santa Cruz, California, USA), and DAPI stain
Sciences, Indianapolis, Indiana, USA. Sterile Roswell Park (Daiko,Japan).Allotherreagentswereofanalyticalgrade(Roche,
Memorial Institute (RPMI) medium, fetal bovine serum (FBS), Germany).
10mM phosphate buffered saline (PBS, pH 7.4), and 10mM
HEPES were purchased from the National University Medical Skin Wound Preparation and Macroscopic Observation
Institute(NUMI),Singapore.Allchemicalswereofanalyticaland Mice (n=6 per group) were anesthetized with meditorin and
cell culturegrade. ketamin injections (i.p., 20 and 100mg/kg body weight,
respectively), and their skin hair from the back shaved with a
In vitro Cytotoxicity sterile razor blade. After wiping the shaved area with 70% ethyl
TheMRC-5andHEPKcellswereculturedin72cm2flaskstoa alcohol, an incision 2cm (full thickness of the wound
density of 106 cells per flask in RPMI as well as Dulbecco’s 4 mm65 mm in length and width) was made. Fifty microliters
ModifiedEagleMedium(DMEM)culturemedium,supplemented of S. aureus (106cfu/ml) suspended in growth medium in sterile
with 10% FBS and 1 ml of HEPES. Cell viability was measured PBS were applied onto the wound region. Mice were separated
usingtetrazoliumsalts(XTT)asdescribedpreviously[33].Briefly, into individual cages and kept for 2 days prior to topical
the cells were allowed to adhere to the flask bottom overnight at application of CaTx-II (10 mg/kg, body weight), antibiotic (2%
37uCinahumidifiedatmosphereof5%CO and95%air.Cells FAO)anduntreatedwoundcontrol(WCtrl).Woundhealingofthe
2
were split and reseeded into new flasks, etc. To analyze the animals was monitored daily by measuring the area of wound
PLOSONE | www.plosone.org 3 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
closurewithatracingpaperat0–16days.Thewoundhealingwas from Thermo Fisher Scientific (Rockford, Illinois, USA) and
then analyzed and the percent reduction of original wound area stored at 280uC prior to analysis. After dilution of the
calculated. Animals were killed after 16days by excess CO homogenates with MH and TS broth, 20ml of each sample was
2
inhalation,andtheskinbiopsysamplescollectedfromthebisected swabbedontheMHandTSagarplatesandincubatedat37uCfor
center from the skin wound for biochemical studies [35]. For 24h.BacterialcolonieswerecountedandCFU/mlcalculatedfor
analysisofcollagenandpro-inflammatorycytokines,woundtissues thedifferent treatment groups.
were frozen immediately in liquid nitrogen and stored at 280uC
until assay time. Analysis of Proinflammatory Cytokines
Skin wound tissues were collected and extracted by standard
Immunohistochemistry protocols.Totalproteinoftheskinhomogenatesweredetermined
For H & E staining, sections of skin were fixed in 4% by using a Bio-Rad Protein Assay (Bio-Rad Laboratories,
paraformaldehyde in PBS for 16h before embedding into Hercules, California, USA). The concentrations of proinflamma-
paraffin. Thin sections (7mm ) were cut and processed in an tory cytokines interleukin-1 beta (IL-1b), IL-6, IL-10, chemokine
alcoholseries(50–100%)for5 min,stainedwithH&Efor3 min, (KC), tumor necrosis factor-alpha (TNF-alpha), fibroblast growth
andthenrinsedintapwaterfor3 min,beforemountingonglass factor-basic (FGF-b), granulocyte-macrophage colony-stimulating
slideswithanembeddingmediaofHisto-Clear(Merck&Co,New factor (GM-CSF), monocyte chemotactic protein-1 (MCP-1) and
Jersey, USA). The sections were examined under the microscope macrophage inflammatory protein-1 beta (MIP-1b) levels were
(Carl Zeiss Inc., Microimaging, LLC, Thornwood, New York, measuredbyusingacommerciallyavailablemultiplexcytokinekit
USA).Masson’sTrichromestainingwasperformedforidentifica- (Mouse Inflammation Kits, BD Pharmingen, San Diego, Califor-
tion of collagen [36]. For immunohistochemical staining, excess nia, USA)according tothemanufacturer’s instructions.
tissuesweretrimmeddownand7 mmthickparaffinsectionswere
used to detect different antigens. Type I collagen deposits were Western Blot Analysis
immunodetected according to the described protocol. Tissue Vascular endothelial growth factor (VEGF) antibody was
sections were incubated overnight with the primary antibodies obtained from Santa Cruz Biotechnology (Santa Cruz, CA,
(1:50or1:100dilutionsforcollagen,and1:200forCD-68)at4uC, USA). Phospho-specific anti-p65 (Ser 536) was purchased from
washed three times in PBS, and then incubated with peroxidase- Cell Signaling (Beverly, MA, USA). b-actin antibody were
conjugated secondary antibodies (1:1000 in blocking solution) for obtained from Sigma Chemicals Co (St. Louis, MO, USA). To
5 min. Sections were counterstained with 4, 6-diamidino-2- determinethelevelsofvariousproteins,skinwoundtissueextracts
phenylindole (DAPI) (Invitrogen,Carlsbad, California, USA)and were prepared and fractionated by SDS-polyacrylamide gel
stored at 4uC. The sections were examined by a Laser Confocal electrophoresis (PAGE) as described previously [37]. After
microscope (Olympus Imaging America Inc., Center Valley, electrophoresis, the proteins were electro-transferred to a nitro-
Pennsylvania,USA)connectedtosoftware(Flowview,FV10.AS), cellulose membrane, blocked with 5% nonfat milk to minimize
version (2.0viewer). non-specificbinding,andprobedwithvariousantibodies(1:1000)
overnight at 4uC. The blot was washed, exposed to HRP-
Measurement of Collagen in Wound Tissues conjugated secondary antibodies for 2 h, and the expression of
CollagencontentofskinwoundsweredeterminedbyanELISA various proteins was detected by chemiluminescence emission
using an assay kit for mouse type I collagen (Chondrex, Inc., (ECL;GE Healthcare, Little Chalfont, Buckinghamshire,UK).
Redmond, Washington, USA) according to the manufacturer’s
instructions. Briefly, the microtiter plate was washed 3 times and Statistical Analysis
blockedwith100mlofblockingbuffersolutionAfor1 hatroom Eachexperimentalgroup,sixanimalswereusedforthewound
temperature. Skin tissues (100 mg) were cut into small pieces by healingstudy(n=6).Allassayswereperformedintriplicate.Data
sterilerazorblade,eachskinsamplewashomogenizedin0.5mlof are presented as mean 6 SD (standard deviation). Statistical
SDS-buffer, (pH 7.6) by a variable speed tissue homogenizer analysis was carried out by ANOVA for comparison of multiple
(Model 7000, Eberbach Corporation, Michigan, USA). Homog- groups,orbyStudent’st-testfortwo-groupcomparisons.Levelsof
enatewascentrifuged(10,0006g,at4uCfor5 min)toremovethe significanceare indicated as*p,0.01,**p,0.05.
supernatant containing the soluble protein and diluted in
1:1000 mg/ml of solution B. Sample, standard, or blank solution Results
B(100 ml)wasaddedtothewellsintriplicateandincubatedover
night at 4uC. After 24h, the plates were washed 3 times with Purification and Identification of Protein (CaTx- II)
16wash buffer, and the secondary antibody (100 ml) introduced Molecules
intoeachwell.Followingincubationatroomtemperaturefor2h, Purification of C. adamanteus venom by Superdex G-75 gel-
the plates were washed 3 times with wash buffer, incubated with filtration chromatography provided four (CA1–CA4) fractions
OPDsolution(100 ml/well)for30minatroomtemperature,and (Fig.1A).Uponfractionationoftheenzymaticallyandantimicro-
50ml of sulfuric acid (50 mM) added to stop the reaction before bially most active fraction (CA2) by C18 RP-HPLC chromatog-
reading theODat 490nm. raphy(Fig.1B),itwasfurtherresolvedintofourfractions(CA-F1
toCA-F4).Theantimicrobially-activefractionCA-F1wasfurther
Extraction of Wound Samples fractionated through a C8 column to obtain a purified protein
Skinwoundtissueswereexcisedbyusingasterilerazorbladeto identified as CaTx-II. Each fraction was purified in subsequent
include the surrounding wound tissues (100mg weight of tissue steps to homogeneity (Fig. 1 C–D). Preliminary mass analysis
taken per animal) 16days after treatment. The samples were (MALDI-TOF/MS) revealed a 13675.34kDa protein (Fig.1 E).
frozeninliquidnitrogenandstoredat280uCuntilanalysis.The SDS-PAGE(Fig.1F)showsasinglebandwithamolecularmassof
tissue samples were homogenized in tissue protein extraction 15kDa. Elucidating PLA activity of column fractions showed
2
reagent (T-PER containing 25mM bicine and 150mM sodium variablelevelsofPLA activity,withthefractioncontainingCaTx-
2
chloride, pH 7.6, including protease inhibitor cocktail purchased II displaying higher levels of PLA activity than that of other
2
PLOSONE | www.plosone.org 4 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
PLOSONE | www.plosone.org 5 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
Figure1.PurificationofCrotalusadamanteustoxin-II(CaTx-II)fromEasternDiamondbackRattlesnakevenom.(A)HighPerformance
Liquid Chromatography (HPLC) profiles of C. adamanteus crude venom from a Superdex G-75 column, (B–D) Reverse-phase (RP)-HPLC
chromatograms from Sepharose C18 and C8 columns, (E) Molecular mass of pure CaTx-II, (F) Sodium dodecyl sulphate-polyacrylamide gel
electrophoresis(SDS-PAGE)profileofRP-HPLCfractions,lanesindicate:CA-CVC.adamanteuscrudevenom(1–2),lane(4)homogeneityofCaTx-II
confirmedbySDS-PAGEas15kDaofCA-F1-reverse-phasefractionand(5)marker,25mgofproteinloadedperlane,respectively.
doi:10.1371/journal.pone.0080199.g001
fractions or crude venom (Fig. S1 A–D). In addition, CaTx-II CaTx-IIvarieddependinguponthebacterium.Althoughitsmain
possessed the highest activity of any PLA purified from C. effect is on bacterial cell membranes as the primary target, it is
2
adamanteus venom. unclear whether CaTx-II may also act on other intracellular
targets.
Amino Acid Analysis
The N-terminal amino acid sequence of CaTx-II is shown in Cell Viability & Cytotoxicity
table S1. When aligned and compared with the N-terminal CytotoxiceffectsofCaTx-IIwereexaminedonhumanlungand
sequencesofothersnakevenomPLA2s,thesequenceofCaTx-IIis skin fibroblast cells. Cell viability was not affected even at higher
quitehomologouswithbasicPLA2sisolatedfromothercrotalidae concentrations of 500mg/ml (Fig. 3A–D). There were no
venoms,withasequencesimilarityof70–80%foundwiththeLys- morphological changes, indicating that the cells remain intact
49PLA2sequence of crotapotinfromC.adamanteus venom. with no swelling or death evidenced at higher (up to 500mg/ml)
concentrations of CaTx-II in a time-dependent manner on skin
Antimicrobial Activity of CaTx-II (Fig.3E–F)andlung (Fig.3K–L).
When screened against a wide range of Gram-negative and
Gram-positive bacteria using a disc-diffusion assay, CaTx-II Release of LDH from Cells
showed very strong antimicrobial effect against S. aureus, B. LDH results revealed that exposure of the lung and skin
pseudomallei (TES & KHW), E. aerogenes, P. vulgaris and P. fibroblastcellstoCaTx-II(upto500mg/ml)werenotcytotoxicup
mirabilis.(Fig.S2A).Itexertedmaximuminhibitionzonesagainst to48h(Fig.3G–J).However,ahigherdose(500 mg/ml)ofCaTx-
S. aureus and B. pseudomalleii (KHW) compared to those of II did not cause LDH release and induced cell death or lysis.
chloramphenicolorstreptomycin(30 mg/disc).Themostsensitive However,thegrowthoffibroblastcellswasnotaffected,especially
bacterium to CaTx-II was S. aureus, whereas against Gram- at thetreatment doseup to500mg/ml.
negative B. pseudomallii (strain KHW), CaTx-II was more active
even at a lower dose of 20 mmole range than the antibiotics Toxicity Study in Mice
(30 mg). Similar effects were also observed with the less purified
The CaTx-II did not show toxic effects or kill mice up to
fraction (CA-F1) (Fig. S2 B). At higher concentrations, the crude
14mg/kg,b.wfor2weeks,andinfacttheybehavedlikecontrols
venom showed more potent antimicrobial activity against the
givensalineonly.
multi-drug resistant B. pseudomallei (KHW) than P. aeruginosa
(Fig. S2 C). The CA-CV crude venom, CA-F1 isolated fractions,
Wound Healing
andthepurifiedCaTx-IIproteinexhibitedonlymoderateactivity
CaTx-II treated mice showed strong wound healing activity
against E. coli, P. aeruginosa, K. pneumoniae and S. pneumoniae at the
following S. aureus infection with a significant reduction in the
tested concentration. However, the growth inhibition curves
clearly illustrated that CaTx-II was the most active antimicrobial woundarea(60%)after8 days,comparedtothatofwoundcontrol
agent against S. aureus than other tested fractions or control mice (P,0.01). In contrast, the antibiotic fusidic acid ointment
antibiotics (Fig.S2D–E). (FAO) treated mice showed only 40% reduction of wound size
after 8 days. The CaTx-II treated wounds contracted faster and
were completely healed by 16days as compared with that of the
Bacteriostatic (MIC) and Bactericidal (MBC) Assays
FAO-treated mice. We provide definitive evidence that CaTx-II
MICs indicated that CaTx-II had an interesting inhibitory
influencescrucialrolesinthewound-healingprocessbyregulating
(bacteriostatic) effect even at the lowest dilutions against S. aureus
leukocyte infiltration, angiogenesis, neovascularization and colla-
(MIC value of 7.8 mg/ml), B. pseudomallei strain KHW and B.
gendeposition (Fig.4 A).
pseudomalleistrainTES(MICvalueof15.6 mg/ml).AhigherMIC
value (62.5 and 125mg/ml) was recorded against P. mirabilis, K.
Collagen Content in Wound Site
pneumoniae,andP.aerugionsaversusP.vulgaris(31.25mg/ml)inFig.
S3.Whereas,bactericidaleffectsofCaTx-IIagainstS.aureus,and The collagen content in wound sites was measured by ELISA.
B.pseudomallei(MBCvaluesof7.8–15.6mg/ml)wereverypowerful TheCaTx-IItreatedanimalsshowedsignificantlyincreasedlevels
andmuchmorepotentonamolebasisversusstandardantibiotics ofcollagencontent(p,0.05)ascomparedwiththewoundcontrol
(Fig.S4). (WCtrl)animals.CaTx-IItreatedmiceshoweda50%increasein
collagenlevels after8 daysversus woundcontrol orFAOtreated
Membrane Damaging Effects of CaTx-II mice. At 16days post-treatment, the collagen content was
Further studies were done with CaTx-II to determine what, if increased to 80% in the CaTx-II treated group (Fig. 4 B–C),
any,overteffectsmightbeseenonbacteriaviaSEManalysis.As and 70% in FAO treated group as compared to that of wound
observedon B.pseudomalliiKHW(Fig.2A–B),B.pseudomalliiTES control mice.
(Fig.2C–D),E.aerogenes(Fig.2E–H),P.vulgaris(Fig.2I–L),S.aureus
(Fig.2M–N),andP.mirabils(Fig.2O–P),theeffectsofthisprotein Histological Re-epithelialization
appeared quite rapidly (24 h) at the tested concentrations (3.9– Histopathologicalexaminationswerecarriedoutonthewounds
250mg/ml). CaTx-II waspotent, evenat thelow concentrations, of treated and untreated mice after 16days. When compared to
in killing the bacteria very efficiently through damaging the the CaTx-II treated wound, the untreated wound epidermis was
membrane. Pore formation and membrane-damaging effects of thin and disorganized; indicating delayed healing and its non-
PLOSONE | www.plosone.org 6 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
involvementintheregenerationandwoundhealingprocess(Fig.4 which is important for wound healing. CaTx-II treated mice
D).Also,thesignificantreductionofinfiltratedinflammatorycells, showed a significant accumulation of collagen on mice skin
an increase in the formation of blood vessels, and enhanced 16days after the injury (Fig. 4 F), whereas saline injected mice
proliferation of cells were observed in CaTx-II treated wounds. showedvery littledeposition ofcollagen (Fig.4 G).
There was thick epidermal regeneration that covered completely
the wound area with well-formed granular layers. In addition, CaTx-II Inhibits NF-kB Activation in Skin Wound Model
there was extensive epithelialization, vascularisation, and hair As NF-kB activation plays a major role in mediating and
follicleformationinCaTx-IItreatedmice.Theearlydermaland modulating gene responses associated with wound healing [38],
epidermalregenerationintreatedmicealsoconfirmedthatCaTx- theabilityofCaTx-IItomodulateNF-kBactivationinskinwound
IIhadapositiveeffectuponcellularproliferation,granulartissue model was also investigated. We found that the treatment with
formation and epithelialization (Fig. 4 E). Masson’s Trichrome CaTx-II substantially inhibited constitutive p65 phosphorylation
staining of control and CaTx-II treated mice after 16days also that was observed in wound control mice (Fig. 4 H). Also, the
evidentlysupportedthecollagenaccumulationanditsexpression, levels of NF-kB-regulated protein, VEGF, an endothelial cell
Figure2.CaTx-IIdamagesthebacterialsurfaceasdeterminedbySEM.(NC)BacteriatreatedwithPBSasanormalcontrol.(A–D)Exposureto
7.8and15.6mg/mlproteinsinducedmainlymembrane-damagingeffectsuponB.pseudomallei(KHW&TES).(E–H)E.aerogenes,(I–L)P.vulgaris,(M–N)
S.aureusincubatedwith7.8mg/mlofproteinshowaseverelydamagedcellmembrane(disorganization),formationofmembraneblebbing,and
release of cellular content (cytoplasmic eruption). (O–P) P. mirabilis exposed to 62.5mg/ml of CaTx-II have surface roughening and membrane
damagingeffectsafter24h.Allsampleswereprocessedat24hforSEManalysisasdescribedintheexperimentalprocedures.
doi:10.1371/journal.pone.0080199.g002
PLOSONE | www.plosone.org 7 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
Figure3.InvitrocytotoxicitywasevaluatedbyXTT-assay.(A–D)CaTx-IIincubatedwithhumanlung(MRC-5)andskinfibroblast(HEPK)cells
usingdifferentconcentrations(15.6,31.25,62.5,125,250,and500mg/ml)after24hand48hincubation.Thevaluesareexpressedasmean6SDof
threereplicates(n=3),Control(Ctrl)cellswithoutanytreatmentservedasacontrol,(E)Lightmicrographshowingthenormalarchitectureofskin
fibroblasts,(F)cellsexposedtoCaTx-IIatdifferentconcentrations15.6–500mg/ml.(K)Lungfibroblastswithoutanytreatment,(L)cellstreatedwith
500mg/mlconcentrationsofCaTx-IIdidnotcausemorphologicalchangesafterexposure(Magnification620),(G–J)Quantificationofcelldeathand
lysiswasbasedonlactatedehydrogenase(LDH)activityreleasedfromthecytosolofdamagedcellsintotheculturesupernatantofthefibroblastcells
(skinandlung)exposedtoproteinupto15.6–500mg/mlconcentrations.CellsexposedtoCaTx-IIat500mg/mlshowedverylowlevelofcytolysis
after48hexposure.
doi:10.1371/journal.pone.0080199.g003
specific mitogen that plays an important role in vascular Bacterial Counting and Positive Cells at Wound Site
development and wound healing [54] was also suppressed upon Woundtissuesoftreatedanduntreatedmicewerecollectedfor
treatment of mice withCaTx-II (Fig.4 H). determiningthebacterialloadasapercentageofCFU/woundof
PLOSONE | www.plosone.org 8 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
three observations (Fig. 5 A). Representative data of six mice in [41]. They are calcium-dependent enzymes that hydrolyze fatty
each group were shown. Vascular densities (re-epithelialization) acids at the sn-2 position of phospholipids [42], consisting of
within the wound bed at 16days after the injury of CaTx-II or catalyticallyactiveandinactiveforms[43].Thecatalyticactivityof
FAO treated mice and untreated mice with injured skin (control) PLA sdependsonthebindingofacalciumionandresidueAsp-49
2
weredeterminedusingPhotoShop.Allthevaluesrepresentmean [44]. However, the side-chain atom of lysine residues can play a
6 SD (n=6 animals); significance at *P,0.01, **P,0.05 similar role to that of calcium, thereby enabling Lys-49 PLA to
2
compared with untreatedmice wound servedasa control. displayinterestingcatalyticactivity[45].ThesetwotypesofPLA s
2
exert most of their biological properties through relatively large
Immunohistochemical Examination molecularsurfacesencompassingseveraldiscretedomainssuchas
Significantbacterialreduction(cfu/wound)wasfoundinwound theN-terminal helixandcalcium-loop [46].
treatedwithCaTx-IIversusFAOtreatedandwoundcontrolmice
(Fig. 5 A). There was a very high level of positive cells (CD68 Antimicrobial Activity
marker for keratinocytes) expressed in the protein treated wound Severalantimicrobialproteinsandpeptideshavebeenidentified
(Fig.5B),andacceleratedre-epithelializations(Fig.5C)compared from both amphibians [47] and reptiles. These molecules are
withuntreatedcontrol.ImmunostainedwoundsamplesofCaTx- attractive in research due to their broad spectrum of activity on
II or FAO treated and wound control mice after 16days are bacteria, viruses and fungi, as well as wound healing effects [1],
shown in Fig. 5 D–F. The sections were stained with collagen [24],[25].C.adamanteusvenomsarebroadlyactiveagainstGram-
(green), CD-68 antigen (red), or co-localization of collagen plus negative and Gram-positive bacteria [48]. The inhibitory and
CD-68. Wound sections of mouse skin topically applied with bactericidal activity of C. adamanteus venom determined against
CaTx-II were treated mice skin intensively stained with collagen aerobicandanaerobicmicroorganismsshowedvariableactivityat
(green), CD-68 antigen (red), and a merger of collagen plus CD- 5–20mg/ml and 480mg/ml, respectively [49]. This venom was
68,indicatinganinterestingover-expressionofcollagenaswellas reported to be equally effective as some conventional antibiotics
CD-68antigenmakerforkeratinocytesascomparedtotheWCtrl. like chloramphenicol and ceftazidime [47], and the bactericidal
The CaTx-II treated wounds expressed or up-regulated more action might be due mainly to the enzymatic activity of C. adamanteus
CD68 positive cells and migration of keratinocytes at the wound venom PLA . In this study, we report for the first time the broad-
2
edgethantheothergroups(Fig.5A–C).TheFAOtreatedsection spectrum antimicrobial activity of CaTx-II purified from C.
showed less intense staining of collagen and CD-68 antigen than adamanateusevenom.ThisPLA wasmosteffectiveagainstaGram-
2
thetreated groups. positive(S.aureus)andGram-negative(B.pseudomalliistrainKHW),
withanantimicrobialpotencysimilartothatofantibiotics.CaTx-
Expression of Inflammatory Cytokines II was the most active, showing inhibitory activity even at low
Levels of interleukin 1beta (IL-1b), IL-6, IL-10, TNF-alpha, concentrations of CaTx-II (20mg) as compared with 30mg for
macrophage inflammatory protein 1 beta (MIP 1 beta), MCP)-1, chloramphenicol andstreptomycin.
FGF-Basic, KC, and granulocyte colony-stimulating factor (G- CaTx-II showed a significant inhibitory effect against S. aureus
CSF)weremeasuredat16dayspost-treatmentinthewoundtissue and B. pseudomallei strains even at lower concentrations (7.8 and
homogenates.WiththeexceptionofIL-10,allcytokinesmeasured 15.6 mg/ml). These findings are, in contrast with several studies
were elevated significantly versus the control group (Fig. 6). involvingbactericidalactivityofsnakevenomPLA2enzymes[16],
Wounds from both CaTx-II and FAO treated groups contained [17]. Whereas, diverse PLA2 enzyme purified from snake venom
significantly low levels of the proinflammatory cytokines TNF- possessed highly potent antimicrobial activity against multi-drug
alpha, IL-1 and IL-6 versus wound controls. The basal layer resistantB.pseudomallei[15].Thereportedactivityisnotonlyduetoa
keratinocytes at the wound edge showed up-regulated expression PLA2, but also the basic Asp-49 PLA2subunit of C. durissus terrificus
of MIP-1 beta. As a result of this mediator increase, the venom and displays bactericidal activity upon E. coli [22]. In this
inflammatory response to injury might potentially disrupt other current study, CaTx-II had minimal effect upon E. coli. In
essential phases of wound repair. The reduction in cytokine addition, myotoxic Asp-49 PLA2s and Lys-49 PLA2 homologues
contentrangedfromapproximately75%forTNF-alpha,IL-1,IL- are potentmicrobicidal components insnakevenoms[74].
6andIL-10,withnodifferencesinthewoundcontentofMCP-1 TheultrastructuralassessmentofCaTx-IIbactericidaleffects,as
measured between the groups. MCP-1 was expressed exclusively reported herein,on Gram-positive (S. aureus andE. aerogenes),and
around 7 days post-injury, whereas FGF-Basic, KC, and G-CSF Gram-negative (B. pseudomalleii) bacteria suggest that CaTx-II
were expressed late (16days) during the recovery phase. FGF ruptures the membrane subsequently and causes cytoplasm loss.
basic and G-CSF play an important role in wound healing as These results clearly provide evidence that CaTx-II acts via its
componentsofthecytokinenetworkthatregulateco-operationof pore-formingactionat7.8and15.6mg/mldoses.Itwaspreviously
development anddifferentiation ofcells during repair. reported that thepeptide derivedfrom theC-terminus (115–129)
of Lys49 Bothrops asper PLA reproduced bactericidal activity
2
Discussion against both Gram-negative (Salmonella typhimurium) and Gram-
positive(S.aureus)bacteriaaseffectivelyasitsparentmolecule[50],
SnakevenomphospholipaseA (svPLA s)enzymesareclassified [51]. Besides, the bactericidal effect of Lys49 or Asp49PLA has
2 2 2
into 15 groups and 23 sub-groups. Most of the elapidae and been previously reported with the specific membrane-damaging
hydrophidae svPLA s belong to group I, whereas viperidae effects[52],[53].MostLys-49PLA scontainahighlevelofbasic
2 2
venoms belong to group II. Previously, a number of PLA s have and hydrophobic amino acids residues located within the C-
2
been isolated from snake venoms [1], [40]. In this study, we terminalregion[54].PLA homologuesbindwithhighaffinityto
2
purified a basic PLA (CaTx-II) from the eastern diamondback bacterial lipopolysaccharide (LPS) to block their effects on
2
rattlesnake(C.adamanteus).ItsN-terminalsequenceexhibitsahigh macrophage and other targets [51]. An earlier study has shown
degree(80%)ofhomologywithcrotapotin,theLys-49PLA from thatthePLA -derivedpeptideinteractswithLPSandspecifically
2 2
C. adamanteus venom. PLA homologues from Crotalus snake the lipid A component from diverse Gram-negative bacteria or
2
venomscontain120–130aminoacidswithsevendisulfidebridges with lipoteichoic acid from S. aureus, and relies on a membrane-
PLOSONE | www.plosone.org 9 February2014 | Volume 9 | Issue 2 | e80199
WoundHealingActionofVenomProtein
PLOSONE | www.plosone.org 10 February2014 | Volume 9 | Issue 2 | e80199