hyperendemic dengue transmission and identification of a ... · limiting febrile illness (dengue...

in subsequent years, although severe disease was not identified in PNG. Additional studies

need to be undertaken to understand dengue epidemiology and burden of disease in PNG.

Author summary

Denguevirus (DENV) wasfirst identified in PapuaNewGuinea(PNG) in 1944.Dengueiscurrentlyassumedto beanendemicdiseasein PNGalthoughthereis little incidenceorprevalencedata,andtheevidenceconsensusfor denguepresenceis low.Routinesurveil-lanceisnot undertakenanddengueisnot anotifiabledisease.Severedengueis rarelyidentifiedby localcliniciansandthereasonsfor thisareunclearbut mayberelatedtopoor recognitionof dengueanda low indexof suspicion,despitehigh incidenceandprev-alenceratesin neighbouringcountries.Forexample,Indonesiasharesborderswith PNGandregularlyreportsoutbreaksof severedengueandtransmissionof multiple DENVserotypes.DENV infection is identified in travellersfrom PNGhoweverthereareno dataon locallycirculatingstrainsandhow theymaycompareto virusesassociatedwith severedengueepidemicsin othercountriesin theAsiaPacificregion.We identifiedevidenceforpreviousinfectionwith all four DENV serotypesamongpeopleliving on thenortherncoastof PNG,in Madang,andon Lihir Islandin theBismarckArchipelagooff thenorth-easterncoast.WealsodetectedDENV-1,DENV-2,andDENV-3virus in febrilepatients,andwedescribethe first wholegenomesequencesof endemicallycirculatingDENV sincetheprototype1944DENV-2New GuineaC strainwascharacterized.Of note,severedenguewasnot diagnosedin anypatientinfectedwith thesevirusesin PNGalthoughintroduc-tion of thePNGDENV-3strain into theSolomonIslandsfiveyearslaterresultedin alargeoutbreakof severedenguewith hospitalizationsanddeathsin thatcountry.Dengueepidemiologyandburdenof diseaseshouldbeinvestigatedin PNG.

IntroductionThedengueviruses(DENV) arethemostimportant arboviralpathogensof humanscausinganestimated390million infectionsannually,of whichapproximatelyonequarteraresymp-tomatic[1]. Infectionwith DENV causesaspectrumof clinicaloutcomesrangingfrom self-limiting febrileillness(denguefever,DF) to potentiallyfatalseveredengue,characterizedbyplasmaleakage,thrombocytopenia,andhypovolemicshock.Dengueisendemicin morethan100tropicalandsubtropicalcountries,wheretheprincipalmosquitovectorsAedes aegypti andAedes albopictus arefound [2, 3].

DENV isasingle-strandedpositive-senseRNA virusof theFlaviviridae family.LikeotherRNA viruses,DENV displaysconsiderablegeneticdiversityand isgroupedinto four antigeni-callydistinctserotypes(DENV-1-DENV-4)whichmaybedistinguishedon thebasisof serumneutralizationtests.Thefour serotypesaremorepreciselyclassified,usingphylogeneticapproaches,into distinctgenotypeswhichhavebeendefinedasclusterswith nucleotidesequencedivergenceof not morethan6%[4]; lineageswithin thegenotypesmayrepresentstrainswith similargeographicorigins [5]. Certaingenotypeshavebeenassociatedwith more[6,7] or less[8] virulent disease,andthereissomeevidencefor humoralandcellularimmuneselectionfocusedon viral B- andT-cellepitopes[9,10].DENV geneticdiversitythusappearsto impacthostmechanismsshownto mediatepathogenesis[11] andultimately,diseaseseverity.

Dengue in Papua New Guinea

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006254 March 1, 2018 2 / 15

design,datacollectionandanalysis,decisiontopublish,orpreparationof themanuscript.

Competinginterests: Theauthorshavedeclaredthatnocompetinginterestsexist.

https://doi.org/10.1371/journal.pntd.0006254

Denguewasfirst identified in PapuaNewGuinea(PNG)whenSabinisolatedtheprototypeDENV-2New GuineaC strainfrom febrilesoldiersdeployedon thenortherncoastof NewGuineain 1944[12]. A DENV strainwith similarbiologicfeaturesastheprototypeDENV-1Hawaii thatSabinhadrecentlyisolatedfrom febrilesoldiersin Hawaii,wasalsoisolatedfrom soldiersinthesameareaof NewGuineain 1944[13], suggestingthatat leasttwo serotypesmayhavecir-culatedin thenorth of PNG in the1940s.More recently,DENV-1-3geneticdataderivedfromviremic travellersreturning to northernAustraliafrom PNGbetween1999–2010[14] indicatethatdengueisendemicin PNG,supportinganearlierserologicalstudydemonstratingasero-prevalencerateof 8%amongpatientspresentingto clinicsin Madangwith acutefebrileillness[15]. Despitethe likely transmissionof multiple DENV serotypesandthepotentialfor intro-ductionof DENV from endemicneighbouringcountrieswhichexperiencelarge-scaleepidem-icsof severedengue[16], little isknownabouttheepidemiologyandtransmissiondynamicsofdenguein PNG,wheredenguesurveillanceisnot undertaken,individualswith acutefebrileill-nessarenot routinely testedfor DENV infection,andwhereseverediseaseis rarelyreported.

Wesoughtto determineDENV serotypeandgenotypeprevalencein localpopulationspre-sentingwith febrileillness,or with a rangeof non-febrileconditions.Wesequencedwholegenomesof DENV andconductedaphylogeneticanalysisto determinetheevolutionaryoriginof PNGDENV. In additionwedeterminedanti-DENV-1-4neutralizingantibodyprofilesinadultsandchildren in order to assessprevalenceandserotypediversity.

Materials and methods

Study area

Samplesanalysedin thisstudywerecollectedfrom Madang,on thenortherncoastof PNG,andfrom Lihir Islandin NewIrelandProvince.Madangisa town of about30,000peoplewithaseaport that isamajorhub for domesticand internationalshipping,andanairport wheredomesticflights from throughoutPNGarriveseveraltimeseachday.Lihir Islandis800kilo-metresnortheastof Madangin theBismarckArchipelago,in thewesternPacificOcean.Theisland’spopulationdoubledto morethan12,000peopleafterestablishmentof agoldmine in1997andalthoughmanyresidentsstill liveapredominantlytraditionalsubsistencelifestyle,inrecentyearstherehasbeenan influx of PNG-nationalandexpatriatemineworkersanddevel-opmentof an internationalairport andseaport.

Patients and samples

Madangserawerecollectedfrom febrilepatientspresentingto theoutpatientclinic of Yagaumrural hospitalor to Jombatown clinic from September2007throughJune2008,andwhowereenrolledin amalariastudy[15]. Seraexcludedfor malariaantigensweretestedfor anti-DENVIgGandIgM, andNS1antigen;8%(46/578)wereidentifiedasprobableacuteDENV infectionie.NS1antigen-positiveand/oranti-DENV IgM-positive.A totalof 55acutephasesera(46seraidentifiedserologicallyasprobableacuteDENV infectionplus9 additionalfebrileserathatwerenot testedfor DENV), wereassessedin thepresentstudyfor thepresenceof DENVbyDENV EgeneRT-PCRandvirus isolationwasattemptedon RT-PCRpositivesamples.Wholegenomesof isolatedvirusesweresequencedusingIllumina. Convalescentserafrom119patientsexcludedfor acuteDENV infectionandwhopresentedfor recollectionwerecol-lectedanaverageof 29.6days(range5–159days)afterthe first patientvisit andtestedfor thepresenceof anti-DENV neutralizingantibody(NAb) to all four serotypessimultaneouslyusingamicroneutralizationassayoptimizedfor smallsamplevolumes[17].

Lihir sera(55 in total) werecollectedfrom patientspresentingto theoutpatientclinic ofLihir rural healthcentrefrom May throughNovember2010during pre-employmentmedical




visits,atantenatalscreeningvisits,or from patientspresentingfor a rangeof conditionsincluding joint pain,diabetesandfever.Theserawerealsoassessedfor DENV genomes(11/55serafrom febrilepatients)andfor anti-DENV NAb (44/55seafrom patientswith non-febrileconditions).

Patientdataaresummarizedin Table1.

Ethical considerations

Ethicsapprovalfor thisstudywasgrantedby theMedicalResearchAdvisoryCommittee,Min-istry of Health,Governmentof PapuaNewGuinea(2010)andtheHumanResearchEthicsCommittee,Universityof WesternAustralia(2010).All dataanalysedwereanonymized.

Virus isolation

DENV wasisolatedfrom serumby inoculationonto monolayersof Verocells[5]. Briefly,100μlof acutephaseserumwasinoculatedonto aVerocellmonolayerin minimal mediain a2.5ml culturetube,and incubatedovernight.On the followingdaythe inoculumwasremovedand3ml of DMEM with 2%FBS(supplementedwith L-glutamineandantibiotics)wasaddedto thecells,andtheculturewasmaintainedat37˚Cwith 5%CO2 for 7 daysor until cytopathiceffect(CPE)wasobserved;for mostsamplesablind passageinto asecond2.5ml culturetubewasrequiredin sorderto isolatevirus.Successfulvirus isolationwasidentifiedbyNS1antigenELISA(PlateliaDengueNS1AntigenELISA;Bio-Rad,Australia).

RNA extraction and RT-PCR

Viral RNA wasextractedfrom 140μl of culturesupernatantusingQIAmp viral RNA Mini kits(Qiagen),accordingto themanufacturer’sinstructions.cDNA wassynthesizedfrom extractedRNA usingSuperScriptIII First-StrandSynthesisSystemfor RT-PCR(Invitrogen)asper themanufacturer’sinstructions.DENV serotypewasidentifiedbyRT-PCRusingserotype-specificprimers[5], andtheLongRangePCRKit (Qiagen)(thermocyclingconditionsareavailableonrequest).

Whole genome sequence analysis

RNA extraction and library preparation. DENV NS1-positivecultureswereexpandedbypassagein 25ml tissuecultureflasksin avolumeof 15ml. Supernatantswereharvestedandclarifiedbycentrifugationat1400rpm for 10minutesat4degC to removecellulardebris.Theclarifiedsupernatantwasthen transferredto aMillipore centrifugalfilter unit (100,000kDa)andcentrifugedat4000x g for 20minutes.RNA wasextractedfrom 150–200μl concentratedvirussupernatantusingtheRocheHigh PureRNA isolationkit asper themanufacturer’s

Table 1. Patient clinical and demographic data.

Category Location Samples collected No. sera Age

Median (range)

Sex

M F NS�

AcuteDENV infection Madang September2007–June2008 55 3 (0.5–50) 26 24 5

Lihir May-November2010 11 39.5(32–47) 1 1 9

Convalescent/non-infectious/non-febrile Madang October2007–July2008 119 14.2(0.5–60) 53 66

Lihir September-October2010 44 25.5(12–49) 5 39

� Not Specified

https://doi.org/10.1371/journal.pntd.0006254.t001



https://doi.org/10.1371/journal.pntd.0006254.t001https://doi.org/10.1371/journal.pntd.0006254

instructions.RNA mass/concentrationwasthendeterminedby fluorescentdetectionusingtheQubit systemwith Qubit RNA HSAssaykit (Invitrogen).All RNA wasstoredat -80˚Cuntilused.DNA librarieswerepreparedwith theTruSeqStrandedmRNA kit (Illumina), with theuseof 200ngof totalstartingRNA andtheexclusionof thepolyA selectionstep.TheDNAlibrary sizeswerevalidatedwith aHigh SensitivityDNA kit (Agilent) on theAgilent2100Bioanalyzer.Theconcentrationof the librarieswerenormalisedto 4nM usingHT1 buffersup-pliedby Illumina andpooledtogether.Thefinal concentrationof thepooledDNA library wasthenre-examinedusingtheQubit ssDNAAssaykit (Invitrogen).ThepooledDNA library wassequencedusingaMiSeqReagentNanoKit, v2(Illumina).

Nucleic acid sequencing and phylogenetic analysis. Sequencesweretrimmed,qualitycheckedandassembledagainstDENV referencegenomesinto contigsusingCLCGenomicsWorkbench(Ver.8.5.1).Completeassembledcontigswerecheckedfor chimerasandalignedagainstpubliclyavailablesequencesusingMACSE[18]. Alignmentswereverifiedin Se-Al2.0(http://tree.bio.ed.ac.uk/software/seal/)andanalyzedusingRAxML to generateamaximum-likelihood(ML) treewith ML bootstrapsupportbasedon theoptimalnumberof ML bootstrapreplicates[19], usingageneral-time-reversiblesubstitutionmodelwith among-siteratehetero-geneityaccountedfor by thegammamodel(GTR+ gamma,bestfit asindicatedindatamonkey.org).Treesweremid-point rooted.To identify nucleotidesubstitutionsthathaveoccurredon the internalbranchesof lineagesformedduring theevolutionof DENV, particu-larly in PNG,wemappedchangesonto theML phylogenetictreeusingaparsimony-basedmethodimplementedin MacClade4.08[20].

Microneutralisation assay

A serummicroneutralization(MN) assaywasusedto measureserumanti-DENV antibodies[17]. Thisapproachwasselectedto allowsimultaneousassessmentof antibodyto all fourDENV serotypesin sampleswith limited volumes.Standardanti-DENV-1-4seraNIBSC05/248(NationalInstitute for BiologicalStandardsandControl [NIBSC],Potter’sBar,Hertford-shire,United Kingdom)wereassayedagainstthehomologousDENV prototypestrainsDENV-1Hawaii2001; DENV-2NGC;DENV-3H-87 andDENV-4H-241andconsistentlyproducedMN titresof 10–20andthus,thecut-off valuefor apositivetestresultwasareciprocalserumdilution of 10.Subjectresultsweresummarizedandpresentedasgeometricmeantitres(GMT). Cross-neutralizationexperimentsin whichStandardDENV-1-4serawereeachtestedagainstheterologousprototypeDENV consistentlyproducednegativeresults.Standardanti-JEserum(NIBSC02/182)andserumsamplesfrom individualswith diagnosedother flavivirusinfection (JEV,MVEV, andKUNJV) weretestedagainstDENV-1-4andwerealwaysnegative.

Accession numbers

All sequenceshavebeendepositedin GenBankandassignedaccessionnumbersKY794785-KY794790.

Results

Transmission of multiple DENV serotypes, PNG

ThreecirculatingDENV serotypes:DENV-1,DENV-2andDENV-3wereidentifiedby fulllengthEgenePCRof acutephaseserumsamplesfrom 17febrilepatientssampledin Madangbetween2007–2008andLihir in 2010(Table2).

DENV-1,DENV-2,andDENV-3 infectionswereidentified in adultsandchildrenfromMadangtown or from villagesandrural settlementsaroundthe town. In severalinstances,two



http://tree.bio.ed.ac.uk/software/seal/http://datamonkey.orghttps://doi.org/10.1371/journal.pntd.0006254

serotypeswereidentified in individualssampledwithin aperiodof 10days.Two DENV casesoriginatedin Lihir in MayandOctober2010–a47year-oldmaleAustraliantravellerinfectedwith DENV-2and32year-oldfemaleresidentof Lihir infectedwith DENV-3.Thesedataclearlyidentify hyperendemicDENV transmissionon thenortherncoastof PNG,in Madang,andon Lihir Islandin theBismarckArchipaleago.

Genome analysis identifies an endemic DENV-3 lineage that has evolved

locally

Wholegenomesweresequencedfrom oneDENV-2andfiveDENV-3 isolates.Phylogeneticanalysisof the fiveDENV-3 isolatesindicatedthat the four from Madanggrouptogethertoform apreviouslyunidentifiedlineagewithin GenotypeI (Fig1).TheMadanglineageincludedotherDENV-3 from theregion,includingagroupexportedto SolomonIslandsandFiji andcollectedin 2013[21] and2014,respectively.TheDENV-3strainoriginating in Lihirin 2010alsogroupedwithin GenotypeI andclusteredwith a lineageformedbyDENV-3 firstfound in travellersbetweenPNGandnorthernAustralia[14]. Collectively,thesedataidentifya lineageof DENV-3endemicto thenortherncoastof PNG,closelyrelatedto DENV-3circu-lating in neighbouringIndonesia,andthesubsequentintroduction of this lineageinto thewesternandsouthPacific.Theentire lineageiswellsupportedanddistinguishedfrom otherviruseswithin GenotypeI bysevenaminoacidsubstitutionsspanningseveralgenes[prM, E,NS2A,NS3,NS5](Table3) andareindicatedby thehighlightedbranchin FigureI. Severalofthesesubstitutionswerenon-conservative,involving thereplacementof oneencodedaminoacidwith anotherof verydifferentproperties.At leasttwo of thesesubstitutionswerefixedthroughtheprocessof positiveselectionaccordingto at leastonestatisticaltestimplementedin HyPhy[22] andtheonlineversionDataMonkey(http://www.datamonkey.org/ [23,24].

TheLihir 2010DENV-2virusgroupedwith theCosmopolitangenotype(Fig2) andclus-teredwith DENV-2originating in Makassar,Indonesiain 2007[25], and identified in travel-lersbetweenPNGandnorthernAustralia[14]. CosmopolitanDENV-2 circulateswidelyin

Table 2. DENV identified in febrile patients, PNG.

Sample Sex, Age (Years) Location Date collected DENV Serotype Strain ID GenBank Accession No.

1 M, 9 Madang 26-Sept-2007 1 - -

2 F,21 Madang 05-Oct-2007 3 - -

3 F,29 Madang 9-Oct-2007 3 D3_PNG_Madang_071009_2007 KY794787

4 F,9 Madang 19-Dec-2007 2 - -

5 F,

SouthandSoutheastAsia,Africa, theMiddle East,andnorthernAustralia,andthesedataillus-trateintroduction of DENV-2 into theWesternPacificregionfrom SoutheastAsia.

Previous DENV infection

Overalldengueseroprevalenceat the two studysiteswas85.3%.Assessmentof convalescentserafrom residentsof MadangidentifiedpreviousDENV infection in thegreatmajority ofsamplestested(101/119sera;84.9%),mostof whichshowedmultitypic NAb responsesto allfour DENV serotypes(Table4).DENV-2andDENV-3GMTswereof greatestmagnitude.Lihir serashowedasimilarprofile (Table5) to thatseenin Madangwherethemajority of serawereseropositive(38/44;86.4%),multitypic DENV-1-DENV-4NAb responsespredominatedandamajority neutralizedall four serotypes.In both locations,monotypicNAb responseswereidentified in 11%of seraandoverallweredirectedagainsteachof the four serotypes.Seroprevalenceincreasedwith age(Table6,Fig3),apatternthat is typicallyseenin dengue-endemicareas.

DiscussionAlthoughdenguewasfirst describedin PapuaNewGuineamorethan70yearsagowhenSabinisolatedtheprototypeDENV-2strainNewGuinea-Cfrom USsoldiersdeployedalongthenorthernPNGcoastduring theSecondWorld War [12], the lackof reportedcasedataandDENV transmissiondatasincethat time hasmeantthat thedistribution of denguein PNG isnot understood.In thisstudyweidentified infectionwith threeDENV serotypesamongfebrilepatientspresentingto healthclinicson thenortherncoastof PNG,in Madang,overaninemonth period in 2007–2008,and infectionwith two serotypeswithin 6 monthsin 2010,inLihir Island.Thesefindingsareconsistentwith hyperendemicDENV transmissionbetween2007–2010andconfirm thatdengueisendemicin thiscountry.

Fig 1. Phylogenetic tree of DENV-3 from Madang (2007–2008) and Lihir (2010). PNGDENV-3 (indicatedin red)comparedwith referencevirusesobtainedfrom GenBank. ThePNGvirusesgroupedwith genotype1of DENV-3.Virusesisolatedfrom Madangclusteredtogetherandweredistinct from laterstrains,clusteringwith theLihir isolate,identified in travellersto northernAustraliaoriginating in PNG.Thetreewasderivedbymaximum likelihoodmethodsusingwholegenomesequences.Most referencevirussequenceswerecompleteenvelopegene.Bootstrapsupportvaluesareshownatnodes.

https://doi.org/10.1371/journal.pntd.0006254.g001

Table 3. Substitutions along the DENV-3 PNG lineage formation.

Substitution Gene Region First Type Replacement Change

F242L membrane Phenylalanine;large(MW 165),nonpolar,aromatic Leucine;MW 131,nonpolar, aliphatic Nonconservative

L404S envelope Leucine;MW 131,nonpolar Serine;MW 105,polar,phosphorylated,O-glycosylated Nonconservative

V1305M1 NS2A Valine;MW 117,nonpolar Methionine;MW 149,nonpolar Nonconservative

T1592I NS3 Threonine;MW 119,polar,phosphorylated,O-glycosylated

Isoleucine; MW 131,nonpolar Nonconservative

T1866S NS3 Threonine;MW 119,polar,phosphorylated,O-glycosylated

Serine;MW 105,polar,phosphorylated,O-glycosylated Conservative

A2021S NS3 Alanine;MW 89,nonpolar Serine;MW 105,polar,phosphorylated,O-glycosylated Nonconservative

F3257Y2 NS5 Phenylalanine;large(MW 165),nonpolar,aromatic Tyrosine;MW 181,polar,phosphorylated,O-glycosylated

Nonconservative

1 underpositiveselectionaccording to theFEL,iFEL,andToggletestsfor selection2 underpositiveselectionaccording to theToggletestfor selection

https://doi.org/10.1371/journal.pntd.0006254.t003



https://doi.org/10.1371/journal.pntd.0006254.g001https://doi.org/10.1371/journal.pntd.0006254.t003https://doi.org/10.1371/journal.pntd.0006254

Our understandingof DENV transmissionin PNGhaspreviouslybeenlargelylimited todetectionof DENV in febriletravellersto northernAustralia[14]. TheMadangDENV-3Genotype1 lineageevolvedlocallyandcirculatedovertheyearthestudywasconducted.Ourphylogeneticanalysisshowedthat it wasmostcloselyrelatedto DENV-3 detected3.5yearslaterin theSolomonIslandsin January2103andthen in theFiji Islandsin 2014,likelyexportedby travellersandhavingmovedoverdistancesof thousandsof kilometres.PNGsharesgeographicborderswith Indonesia,wheredengueepidemicsoccurregularlyandendemic,locallyevolvedDENV-3 lineageshavebeendescribed[5, 25].TheMadanglineageismostcloselyrelatedto DENV-3 originating in Indonesia,asis theDENV-3originating inLihir Islandwhichclusteredwith DENV-3 identified in travellersto northernAustralia.

Similarly,theDENV-2virus identified in thisstudyin a localresidentof Lihir Islandin 2010is representativeof strainsknown to circulatein theregion,clusteringwith ahighlysimilarstrainoriginating in Makassar,Indonesia,in 2007;bothvirusesbelongto aDENV-2 (Cosmopolitangenotype)cladealsocirculatingin neighbouringSingaporein 2008andidentified in travellerstonorthernAustraliain 2004and2006.Our resultsprovidefurther evidenceof DENV movementbetweenendemiccountriesin theAsiaPacificregionthathasbeendescribedbyourselvesandothers[26–29],andalsoillustratethevalueof sequencedataasameansof understandingvirusdispersal.PNGmaybeasource,or at leastaplaceof transit,for dengueto enterthePacificandfurther disseminateto otherPacificIslandnations.

Fig 2. Phylogenetic tree of DENV-2 from Lihir, October 2010. PNGDENV-2 (indicatedin red)comparedwithreferencevirusesobtainedfrom GenBank. TheLihir virusgroupedwith theCosmopolitangenotypeof DENV-2,andclusteredwith DENV-2 identified in travellersenteringnorthernAustralia from PNG2009–2010. Thetreewasderivedbymaximum likelihoodmethodsusingwholegenomesequences.Most referencevirussequenceswerecompleteenvelopegene.Bootstrapsupportvaluesareshownatnodes.


Table 4. Anti-DENV neutralizing antibody profiles, Madang.

Anti-DENV NAb-positive Anti-DENV NAb Titer�

Serotype(s) No. positive (%) DENV-1 DENV-2 DENV-3 DENV-4

DENV-1 1 (1) 10 - - -

DENV-2 7 (6.9) - 14.9(10–40) - -

DENV-3 1 (1) - - 320 -

DENV-4 2 (2) - - - 10(10)

Sub-total, monotypic^ 11 (10.9)DENV-1,-2 2 (2) 20(20) 20(20) - -

DENV-2,-3 6 (5.9) - 26.9(10–80) 40(10–320) -

DENV-2,-4 2 (2) - 20(20) 20(10–40) -

DENV-3,-4 2 (2) - - 20(10–40) 10(10)

DENV-1,-2,-3 2 (2) 113.1(40–320) 20(10–40) 28.3(20–40) -

DENV-1,-2,-4 6 (5.9) 17.8(10–40) 25.2(10–80) - 15.9(10–40)

DENV-1,-3,-4 5 (5) 13.2(10–20) - 60.6(10–320) 30.3(20–40)

DENV-2,-3,-4 4 (4) - 14.1(10–20) 20(10–80) 14.1(10–40)

DENV-1,-2,-3,-4 61(60.4) 45.4(10–640) 83.8(10–640) 80.9(10–640) 63.6(10–640)

Sub-total, multiple 90 (89.1)Total 101(100)

OverallGMT 37.1(10–640) 61.5(10–640) 72.8(10–640) 46.4(10–640)

� NAb expressedasgeometricmeantiter (range)

^ Monotypic responseswerelargelyin children(medianage,2 years)

- denotestiter

DENV epidemicvirulencehasbeenlinked to introduction andtransmissionof specificserotypesand lineages[6,7].TheDENV-3 lineageweidentified to becirculatingin Madangin2007–2008andwhichour analysisshowedto haveevolvedlocallywassubsequentlyassociatedwith hospitalizationsanddeathswhenit wasintroducedinto theSolomonIslandsin 2013[30], althoughseverediseasewasnot identifiedamongpatientsin Madang.Thisentire lineagewasdistinguishedbyseveralaminoacidsubstitutionsthatmayinfluencevirusphenotypic

Table 5. Anti-DENV neutralizing antibody profiles, Lihir.

Anti-DENV NAb-positive Anti-DENV NAb Titer �

Serotype(s) No. positive (%) DENV-1 DENV-2 DENV-3 DENV-4

DENV-1 2 (5.4) 10 - - -

DENV-2 1 (2.7) - 10 - -

DENV-3 1 (2.7) - - 10 -

DENV-4 0 (0) - - - -

Sub-total, monotypic 4 (10.5)DENV-1,-2 2 (5.3) 20 20(20–160) - -

DENV-1,-4 1 (2.7) 20 - - 10

DENV-2,-4 2 (5.3) - 20(10–40) - 10

DENV-1,-2,-3 3 (7.9) 40(10–160) 127(80–320) 15.9(10–20) -

DENV-2,-3,-4 4 (10.5) - 33.6(10–80) 11.9(10–20) 10

DENV-1,-2,-3,-4 22(59.5) 25.7(10–640) 49.9(10–640) 30.1(10–160) 22.7(10–640)

Sub-total, multiple 34 (91.9)Total 38(100)

OverallGMT 26.3(10–640) 49.4(10–640) 24.8(10–160) 18.6(10–160)

� NAb expressedasgeometricmeantiter (range)

- denotestiter

characteristicsandultimately,diseaseoutcome.Severedenguehasalsobeenlinked to geneticpolymorphismsincludingHLA type[31]; PNGandtheSolomonIslandsbothbelongto theMelanesiansubgroupof PacificIslandersandalthoughrelativelyfewHLA dataareavailablefor thispopulationgroup,certainallelefrequenciesareknown to beshared.Studiesto assessthe immunopathogenesisof denguein PNGwouldbeinformativeandshouldbeundertakenin future.

Dengueseroprevalencewasveryhighwith anoverallrateof 85.3%,and increasedwith age.More thanhalfof seropositiveindividualsfrom Madangweregreaterthan11yearsof ageandabout40%wereolder than20years.Thisagedistribution reflectsopportunitiesfor multipleexposuresoverthe lifetimeof the individual in asettingwheredengueisendemicor is regu-larly introducedand indeed,anti-DENV neutralisingantibodyprofileswerepredominantlymultitypic. A majority of individualsin Madangand in Lihir demonstratedresponsesto twoor moreDENV serotypesandmostseraneutralizedall four DENV serotypes.Serumsamplesincludedin thisanalysiswerefrom two maingroups–convalescentpatientsexcludedfor acuteDENV infection in Madang,andclinic attendeesin Lihir (predominantlywomenattendingantenatalclinics)thereforetheseneutralizationdatalikely reflectmultiple DENV infectionsintheyearsprior to sampling,andcorroboratethegeneticevidencefor circulationof multipleserotypes.Studiesin rural Haiti andNicaragua[32,33]haveshownthat in endemicpopula-tionsanti-DENV antibodyprevalenceincreaseswith ageandbeginsto plateauin adolescencereflectinglong-termexposureto, and infectionwith, endemicallycirculatingDENV. In thepresentstudyfurther supportfor endemicDENV transmissionin PNG is indicatedby theyoungage(median2 years)of individualsin Madangwith monotypicanti-DENV NAbresponsesto DENV-1,DENV-2,DENV-3or DENV-4whereasthemajority of multitypicinfectionswerein adults.Threeindividualswith monotypicNAb profileswerebabieslessthan1yearold andanti-DENV NAb werelikely passivelytransferredmaternalantibody.Wedonot know theageof themothersbut it ispossibletheywererelativelyyoungand/orhadexperi-encedasingleDENV infection in thepast,or that theyhadexperiencedmultiple infectionsbut thatonly antibodiesto asingleserotypewerepresentatsufficientlyhigh levelsin the infantto bedetectablein our assay.

HyperendemicDENV transmissionis associatedwith symptomaticdengueinfectionandwith greaterincidenceof severedengue[3]. No febrilepatientswerediagnosedwith

Fig 3. Dengue seroprevalence in PNG. Columnsrepresentagegroups,andtheproportion of NAb-positiveseraisshownfor eachgroup.Seroprevalenceincreaseswith age,plateauingafteradultsreach20yearsof age.Monotypicinfections(blackline) declineto closeto zerowith increasingage.




https://doi.org/10.1371/journal.pntd.0006254.g003https://doi.org/10.1371/journal.pntd.0006254

severedenguein Madang[15] or Lihir and indeed,severedengueis rarelyidentified inPNGdespitethehyperendemictransmissionwehaveidentifiedandwhichhaslikely beenoccurringfor asignificantperiodof time. In previousstudieswedemonstratedhigh sero-prevalenceandpredominantlymultitypic NAb responsesto all four serotypesin seracol-lectedin NewGuineabetween1959–1963[17] indicatingDENV transmissionin thedecadesprior to sampling.Thereasonsfor the rarity of severediseaseareunclearbut mayberelatedto poor recognitionof dengueand the lackof routine denguesurveillance.It isalsopossiblethatundefinedDENV-specificimmunemechanismsmaycontributeto theapparentrarity of severedenguedisease.

In summarywehaveidentifiedhyperendemicdenguetransmissionin PNG in theperiodup to 2010,whichhaslikely beenoccurringfor manyyears.Wealsodemonstratecirculationof DENV whichhasevolvedlocally,andshownbyothersto havebeenintroducedinto thewesternandsouthPacificin subsequentyears.Theabsenceof regularsamplingfor DENV inPNGandthepotentialfor misdiagnosisof febrileindividualshasmeantthat theevidencecon-sensusfor denguepresenceis low [2] andthe trueburdenof diseaseisunknown.A dengueoutbreakin PortMoresbywasconfirmedby thePNGDepartmentof Healthin 2016[34]highlightingtheneedfor additionalstudiesto beundertakento understandtheepidemiologyand impactof denguein thiscountry.An important aspectof this is to understandtheoriginandtransmissionpatternsof PNGDENV, defineendemicand introducedgenotypesand line-ages,andto characterizeepidemicvirulenceassociatedwith circulationof thesevirusesamongthePNGpopulationandwithin theAsiaPacificregion.

Acknowledgmen tsWe thankMichaelAlpersandRichardYanagiharafor helpfuldiscussionsandcritical reviewof themanuscript,andVictor EfflerandJamesBurgessfor helpwith graphics.

Author ContributionsConceptualization: Allison Imrie.

Data curation: ShannonBennett,Allison Imrie.

Formal analysis: DagwinLuang-Suarkia,Oriol Mitja, Timo Ernst,ShannonBennett,AlfredTay,DavidW. Smith,Allison Imrie.

Funding acquisition: DavidW. Smith,Allison Imrie.

Investigation: DagwinLuang-Suarkia,Oriol Mitja, DavidW. Smith,Allison Imrie.

Methodology: DagwinLuang-Suarkia,Timo Ernst,ShannonBennett,Alfred Tay,AllisonImrie.

Project administration: Allison Imrie.

Resources: Oriol Mitja, RussellHays,DavidW. Smith,Allison Imrie.

Supervision: DavidW. Smith,Allison Imrie.

Writing – original draft: DagwinLuang-Suarkia,ShannonBennett,Alfred Tay,AllisonImrie.

Writing – review & editing: DagwinLuang-Suarkia,Oriol Mitja, Timo Ernst,ShannonBen-nett,Alfred Tay,RussellHays,DavidW. Smith,Allison Imrie.




References1. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL et.al. The global distribution and

burden of dengue. Nature 2013; 496:504–507. https://doi.org/10.1038/nature12060 PMID: 23563266

2. Brady OJ, Gething PW, Bhatt S, Messina JP, Brownstein JS, Hoen AG et al. Refining the global spatiallimits of dengue virus transmission by evidence-based consensus. PLoS Negl Trop Dis. 2012; 6:e1760.

https://doi.org/10.1371/journal.pntd.0001760 PMID: 22880140

3. World Health Organization. Dengue and severe dengue. Fact sheet 117, updated July 2016. http://www.who.int/mediacentre/factsheets/fs117/en/

4. Rico-Hesse R. Molecular evolution and distribution of dengue virus types 1 and 2 in nature. Virology1990; 174:479–493. PMID: 2129562

5. Ernst T, McCarthy S, Chidlow G, Luang-Suarkia D, Holmes EC, Smith DW, Imrie A. Emergence of anew lineage of dengue virus type 2 identified in travellers entering Western Australia from Indonesia,

2010–2012. PLoS Negl Trop Dis. 2015; 9: e0003442. https://doi.org/10.1371/journal.pntd.0003442

PMID: 25635775

6. Rico-Hesse R, Harrison LM, Salas RA, Tovar D, Nisalak A, Ramos C, et al. Origins of dengue type 2viruses associated with increased pathogenicity in the Americas. Virology 1997; 230(2); 244–251.

PMID: 9143280

7. Messer WB, Gubler DJ, Harris E, Sivananthan K, de Silva AM. Emergence and global spread of a den-gue serotype 3, subtype III virus. Emerg Infect Dis 2003; 9:800–809. https://doi.org/10.3201/eid0907.

030038 PMID: 12899133

8. Watts DM, Porter KR, Putvatana P, Vasquez B, Calampa C, Hayes CG, et al. Failure of secondaryinfection with American genotype dengue 2 to cause dengue haemorrhagic fever. Lancet 1999;

354:1431–1434. https://doi.org/10.1016/S0140-6736(99)04015-5 PMID: 10543670

9. Twiddy SS, Woelk CH,Holmes EC. Phylogenetic evidence for adaptive evolution of dengue viruses innature. J Gen Virol 2002; 83:1679–1689. https://doi.org/10.1099/0022-1317-83-7-1679 PMID:

12075087

10. Hughes AL. Evolutionary change of predicted cytotoxic T cell epitopes of dengue virus. Infect GenetEvol. 2001; 1:123–130. PMID: 12798027

11. Rothman AL. Immunity to dengue virus: a tale of original antigenic sin and tropical cytokine storms. NatRev Immunol 2011; 11:532–543. https://doi.org/10.1038/nri3014 PMID: 21760609

12. Sabin AB, Schlesinger RW. Production of immunity to dengue with virus modified by propagation inmice. Science 1945; 101:640–642. https://doi.org/10.1126/science.101.2634.640 PMID: 17844088

13. Sabin AB. The dengue group of viruses and its family relationships. Bact. Rev. 1950; 14:225–232.PMID: 14772197

14. Warrilow D, Northill JA, Pyke AT. Sources of dengue viruses imported into Queensland, Australia,2002–2010. Emerg Infect Dis. 2012; 11:1850–1857.

15. Senn N, Luang-Suarkia D, Manong D, Siba PM, McBride WJH. Contribution of dengue fever to the bur-den of acute febrile illness in Papua New Guinea: an age-specific prospective study. Am J Trop Med

Hyg 2011; 851:132–137.

16. Karyanti MR, Uiterwaal CS, Kusriastuti R, Hadinegoro SR, Rovers MM, Heesterbeek H, et al. Thechanging incidence of dengue haemorrhagic fever in Indonesia: a 45-year registry-based analysis.

BMC Infect Dis. 2014;14: https://doi.org/10.1186/1471-2334-14-14 PMID: 24405719

17. Luang-Suarkia D, Ernst T, Alpers MP, Garruto R, Smith D, Imrie A. Serological evidence for transmis-sion of multiple dengue virus serotypes in Papua New Guinea and West Papua prior to 1963. PLoS

Negl Trop Dis. 2017; 11(4):e0005488. https://doi.org/10.1371/journal.pntd.0005488 PMID: 28437465

18. Ranwez V, Harispe S, Delsuc F, Douzery EJP. MACSE: Multiple Alignment of Coding SequencesAccounting for Frameshifts and Stop Codons. PLoS ONE 2011; 6: e22594. https://doi.org/10.1371/

journal.pone.0022594 PMID: 21949676

19. Stamatakis A, Hoover P., Rougemont J. A Rapid Bootstrap Algorithm for the RAxML Web-Servers. Sys-tematic Biology 2008; 75:758–771.

20. Maddison WP, Maddison DR. Interactive analysis of phylogeny and character evolution using the com-puter program MacClade. Folia primatologica; Int. J of Primatology. 1989; 53:190–202.

21. Cao-Lormeau V-M, Roche C, Musso D, Mallet H-P, Dalipanda T, Dofai A, et al. Dengue virus type 3,South Pacific Islands, 2013. Emerg. Infect. Dis. 2014; 20:1034–1036. https://doi.org/10.3201/eid2006.

131413 PMID: 24856252

22. Pond SL, Frost SD, Muse SV. HyPhy: hypothesis testing using phylogenies. Bioinformatics 2005;21:676–679. https://doi.org/10.1093/bioinformatics/bti079 PMID: 15509596



https://doi.org/10.1038/nature12060http://www.ncbi.nlm.nih.gov/pubmed/23563266https://doi.org/10.1371/journal.pntd.0001760http://www.ncbi.nlm.nih.gov/pubmed/22880140http://www.who.int/mediacentre/factsheets/fs117/en/http://www.who.int/mediacentre/factsheets/fs117/en/http://www.ncbi.nlm.nih.gov/pubmed/2129562https://doi.org/10.1371/journal.pntd.0003442http://www.ncbi.nlm.nih.gov/pubmed/25635775http://www.ncbi.nlm.nih.gov/pubmed/9143280https://doi.org/10.3201/eid0907.030038https://doi.org/10.3201/eid0907.030038http://www.ncbi.nlm.nih.gov/pubmed/12899133https://doi.org/10.1016/S0140-6736(99)04015-5http://www.ncbi.nlm.nih.gov/pubmed/10543670https://doi.org/10.1099/0022-1317-83-7-1679http://www.ncbi.nlm.nih.gov/pubmed/12075087http://www.ncbi.nlm.nih.gov/pubmed/12798027https://doi.org/10.1038/nri3014http://www.ncbi.nlm.nih.gov/pubmed/21760609https://doi.org/10.1126/science.101.2634.640http://www.ncbi.nlm.nih.gov/pubmed/17844088http://www.ncbi.nlm.nih.gov/pubmed/14772197https://doi.org/10.1186/1471-2334-14-14http://www.ncbi.nlm.nih.gov/pubmed/24405719https://doi.org/10.1371/journal.pntd.0005488http://www.ncbi.nlm.nih.gov/pubmed/28437465https://doi.org/10.1371/journal.pone.0022594https://doi.org/10.1371/journal.pone.0022594http://www.ncbi.nlm.nih.gov/pubmed/21949676https://doi.org/10.3201/eid2006.131413https://doi.org/10.3201/eid2006.131413http://www.ncbi.nlm.nih.gov/pubmed/24856252https://doi.org/10.1093/bioinformatics/bti079http://www.ncbi.nlm.nih.gov/pubmed/15509596https://doi.org/10.1371/journal.pntd.0006254

23. Pond SL, Frost SD. Datamonkey: rapid detection of selective pressure on individual sites of codonalignments. Bioinformatics 2005; 21:2531–2533. https://doi.org/10.1093/bioinformatics/bti320 PMID:

15713735

24. Delport W, Scheffler K, Botha G, Gravenor MB, Muse SV, Kosakovsky Pond SL. CodonTest: Modellingamino acid substitution preferences in coding sequences. PLoS Comput Biol. 2010; 6:e1000885.

https://doi.org/10.1371/journal.pcbi.1000885 PMID: 20808876

25. Sasmono T, Wahid I, Yohan B, Trimarsanto H, Yusuf I, Wahyuni S, et al. Genomic analysis and growthcharacteristics of dengue viruses from Makassar, Indonesia. Infect. Genet. Evol. 2015; 32:165–177.

https://doi.org/10.1016/j.meegid.2015.03.006 PMID: 25784569

26. Cao-Lormeau VM, Roche C, Aubry M, Teissier A, Lastere S, Daudens E, et al. Recent emergence ofdengue virus serotype 4 in French Polynesia results from multiple introductions from other South Pacific

islands. PLoS One. 2011; 12: e29555.

27. Imrie A, Roche C, Zhao Z, Bennett S, Laille M, Effler P, et al. Homology of complete genome sequencesfor dengue virus type 1 from dengue fever and dengue haemorrghagic fever-associated epidemics in

Hawaii and French Polynesia. Ann Trop Med Parasitol 2010; 104:225–235. https://doi.org/10.1179/

136485910X12647085215570 PMID: 20507696

28. A-Nuegoonpipat A, Berlioz-Arthaud A, Chow V, Endy T, Lowry K, Mai le Q, et al. Sustained transmis-sion of dengue virus type 1 in the Pacific due to repeated introductions of different Asian strains. Virol-

ogy 2004; 329:505–512. https://doi.org/10.1016/j.virol.2004.08.029 PMID: 15518827

29. Aubry A, Teissier Y, Mapotoeke M, Tessier A, Giard M, Musso D, et al. High risk of dengue type 2 out-break in French Polynesia, 2017. Euro Surveill. 2017; 22:30505. https://doi.org/10.2807/1560-7917.ES.

2017.22.14.30505 PMID: 28422007

30. Nogareda F, Joshua C, Sio A, Shortus M, Dalipanda T, Durski K, et al. Ongoing outbreak of dengueserotype-3 in Solomon Islands, January to May 2013. Western Pac Surveill Response J. 2013; 4:28–

33. https://doi.org/10.5365/WPSAR.2013.4.2.013 PMID: 24319611

31. Stephens HA. HLA and other gene associations with dengue disease severity. Curr Top MicrobiolImmunol. 2010; 338:99–114. https://doi.org/10.1007/978-3-642-02215-9_8 PMID: 19802581

32. Rioth M, Beauharnais CA, Noel F, Ikizler MR, Mehta S, Zhu YZet al. Serologic imprint of dengue virus inurban Haiti: characterization of humoral immunity to dengue in infants and young children. Am J Trop

Med Hyg. 2011; 84:630–636. https://doi.org/10.4269/ajtmh.2011.10-0323 PMID: 21460022

33. Balmaseda A, Hammond SN, Tellez Y, Rodriguez Y, Saborio SI, Mercado JC et al. High seropreva-lence of antibodies against dengue virus in a prospective study of schoolchildren in Managua, Nicara-

gua. Trop Med Int Health 2006; 11:935–492. https://doi.org/10.1111/j.1365-3156.2006.01641.x PMID:

16772016

34. One PNG News Online: https://www.onepng.com/2016/04/dengue-fever-update-from-port-moresby.html


https://doi.org/10.1093/bioinformatics/bti320http://www.ncbi.nlm.nih.gov/pubmed/15713735https://doi.org/10.1371/journal.pcbi.1000885http://www.ncbi.nlm.nih.gov/pubmed/20808876https://doi.org/10.1016/j.meegid.2015.03.006http://www.ncbi.nlm.nih.gov/pubmed/25784569https://doi.org/10.1179/136485910X12647085215570https://doi.org/10.1179/136485910X12647085215570http://www.ncbi.nlm.nih.gov/pubmed/20507696https://doi.org/10.1016/j.virol.2004.08.029http://www.ncbi.nlm.nih.gov/pubmed/15518827https://doi.org/10.2807/1560-7917.ES.2017.22.14.30505https://doi.org/10.2807/1560-7917.ES.2017.22.14.30505http://www.ncbi.nlm.nih.gov/pubmed/28422007https://doi.org/10.5365/WPSAR.2013.4.2.013http://www.ncbi.nlm.nih.gov/pubmed/24319611https://doi.org/10.1007/978-3-642-02215-9_8http://www.ncbi.nlm.nih.gov/pubmed/19802581https://doi.org/10.4269/ajtmh.2011.10-0323http://www.ncbi.nlm.nih.gov/pubmed/21460022https://doi.org/10.1111/j.1365-3156.2006.01641.xhttp://www.ncbi.nlm.nih.gov/pubmed/16772016https://www.onepng.com/2016/04/dengue-fever-update-from-port-moresby.htmlhttps://www.onepng.com/2016/04/dengue-fever-update-from-port-moresby.htmlhttps://doi.org/10.1371/journal.pntd.0006254

hyperendemic dengue transmission and identification of a ... · limiting febrile illness (dengue...

Documents