Simple, specific molecular typing of dengue virus isolates using one-step RT-PCR and restriction fragment length polymorphism

G

V

Sa

AYa

b

c

ARRAA

KDFPDPPR

1

smaFv

delY

AP

i(gdP

P

0h

ARTICLE IN PRESS Model

IRMET-11868; No. of Pages 7

Journal of Virological Methods xxx (2012) xxx– xxx

Contents lists available at SciVerse ScienceDirect

Journal of Virological Methods

jou rn al h om epage: www.elsev ier .com/ locate / jv i romet

imple, specific molecular typing of dengue virus isolates using one-step RT-PCRnd restriction fragment length polymorphism

lma Ortiza,b,1, Zeuz Capitana,1, Yaxelis Mendozaa,1, Julio Cisnerosa,1, Brechla Morenoa,1,amitzel Zaldivara,1, Mariana Garciaa,1, Rebecca E. Smitha,1, Jorge Mottaa,1, Juan Miguel Pascalea,b,c,∗,2

Gorgas Memorial Institute for Health Studies, Panama City, PanamaMSc Program in Biotechnology, University of Panama, Panama City, PanamaSchool of Medicine, University of Panama, Panama City, Panama

rticle history:eceived 26 October 2011eceived in revised form 3 June 2012ccepted 12 June 2012vailable online xxx

eywords:

a b s t r a c t

A one-step RT-PCR and one-enzyme RFLP was used to detect and distinguish among flaviviruses, includ-ing the four serotypes of dengue and the St. Louis Encephalitis, West Nile and Yellow Fever viruses incultured virus samples or acute-phase human serum. Using a previously described RT-PCR, but novelRFLP procedure, results are obtained in 24 h with basic PCR and electrophoresis equipment. There is 95%agreement between RT-PCR/RFLP results and those achieved by indirect immunofluorescence assays,and 100% agreement between RT-PCR/RFLP results and gene sequencing. This method is more rapid thantests of cytopathic effect based on virus isolation in tissue culture, and simpler than real-time PCR. It

enguelavivirusanamaiagnostic techniques and proceduresolymerase chain reaction

does not require specialized equipment, radioisotopes or computer analysis and is a method that can beapplied widely in the developing world. It allows for prompt determination of whether a flavivirus is thecause of illness in a febrile patient, rapid identification of dengue serotypes in circulation, and improvedpatient management in cases where prior dengue exposure make dengue hemorrhagic fever or dengue

olymorphismestriction fragment length

shock syndrome a risk.

. Introduction

The dengue viruses (DENV) are human pathogens with fourerologically distinct types, DENV-1, -2, -3 and -4. They are trans-itted by Aedes aegypti and Aedes albopictus mosquitoes (Weaver

Please cite this article in press as: Ortiz, A., et al., Simple, specific molecularfragment length polymorphism. J. Virol. Methods (2012), http://dx.doi.org

nd Reisen, 2010) and belong to the family Flaviviridae and genuslavivirus (ICTV, 2005), comprising over 70 positive-sense RNAiruses, with 40 tick-borne or mosquito-borne pathogens among

Abbreviations: DENV, dengue virus; DHF, dengue hemorrhagic fever; DSS,engue shock syndrome; JE, Japanese encephalitis virus; MVE, Murray Valleyncephalitis virus; RT-PCR, reverse-transcriptase PCR; RFLP, restriction fragmentength polymorphism; SLE, St. Louis encephalitis virus; WN, West Nile virus; YF,ellow Fever virus.∗ Corresponding author at: Instituto Conmemorativo de Estudios de la Salud,venida Justo Arosemena y Calle 35, Apartado Postal 0816 – 02593, Bella Vista,anamá, Panama. Tel.: +507 527 4811; fax: +507 527 4889.

E-mail addresses: [email protected] (A. Ortiz), [email protected] (Z. Cap-tan), Yaxelis [email protected] (Y. Mendoza), [email protected]. Cisneros), [email protected] (B. Moreno), [email protected] (Y. Zaldivar),[email protected] (M. Garcia), [email protected] (R.E. Smith),[email protected] (J. Motta), [email protected], [email protected] (J.M.ascale).1 Address: Apartado Postal 0816 – 02593, Panama.2 Address: Ciudad Universitaria, Apartado Postal Estafeta Universitaria 3366,

anamá 4, Panama.

166-0934/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jviromet.2012.06.016

© 2012 Elsevier B.V. All rights reserved.

these (Kuno et al., 1998). Dengue causes up to 50 million cases eachyear worldwide and is the world’s most important mosquito-borneviral disease (WHO, 2011).

Mild or moderate infections with dengue present with fever,headache, rash, muscle and joint pain, and lethargy (Nielsen, 2009).Dengue hemorrhagic fever (DHF) and dengue shock syndrome(DSS) are more severe, complicated forms of dengue infection,manifesting in coagulopathy and increased vascular permeabil-ity, which results from a complex and only partially understoodimmunological process (Martina et al., 2009). Preexisting immu-nity to a related flavivirus, however, is known to be an importantfactor in the progression of DHF/DSS; 90% of DHF/DSS cases occurin patients with a secondary infection to a heterologous dengueserotype (Mathew and Rothman, 2008).

For this reason, in regions where more than one dengue serotypeis known, simple and rapid methods of identifying the responsi-ble serotype for a disease outbreak are important for the purposesof public health and vector control programs, risk mitigation forDHF/DSS, and diagnosis and management of individual patients.Current methods for serotype identification range from sophisti-cated techniques, which are sensitive but not widely applicable in

typing of dengue virus isolates using one-step RT-PCR and restriction/10.1016/j.jviromet.2012.06.016

diagnostic laboratories of the developing world (Nascimento et al.,2011); to viral culture techniques requiring more than 1 week fora result (Kuno et al., 1985); and a range of ELISA-based (Puttikhuntet al., 2011) and real-time (Hue et al., 2011), nested (Gomes et al.,

dx.doi.org/10.1016/j.jviromet.2012.06.016


http://www.sciencedirect.com/science/journal/01660934

http://www.elsevier.com/locate/jviromet

mailto:[email protected]












ING Model

V

2 ogical

2t

ocsPbRdsdvtdogu

F(YtUbp1banamaeb

2

2

wtlia2uImstt

2i

stawadc

ARTICLEIRMET-11868; No. of Pages 7

A. Ortiz et al. / Journal of Virol

007) and multiplex (Das et al., 2008; Saxena et al., 2008) PCRechniques.

A simple and rapid method is described for the identificationf Flavivirus in the acute-phase sera of febrile patients, or fromultured virus samples, and the specific identification of dengueerotypes using a one-step, standard RT-PCR and RFLP. Other RT-CR/RFLP techniques for the detection of dengue or Flavivirus haveeen described previously: Vorndam et al. (1994) described anT-PCR/RFLP technique which identifies geographic subgroups ofengue, but which requires 2 days and amplification of the entiretructural region of the flavivirus genome; Sudiro et al. (1997)escribed an RT-PCR method that is rapid and distinguishes dengueiruses from other flaviviruses in human sera, but does not iden-ify definitively dengue serotype; and Gaunt and Gould (2005)escribed an RT-PCR/RFLP method which discriminates among 90%f known vector-borne flaviviruses with published full length E-ene sequences within 10 h, but have not yet published on thetility of this method with serum samples.

The RT-PCR primers used in this investigation, FLAVI-1 andLAVI-2, amplify a region of the non-structural coding gene 5NS5) from the four most important American flaviviruses (dengue,ellow Fever, West Nile and St. Louis encephalitis viruses), andhe Japanese Encephalitis, Murray Valley Encephalitis, Kunjin andsutu viruses (Ayers et al., 2006). The NS5 gene has alreadyeen established as a useful region of the genome for creatinghylogenies and distinguishing among flaviviruses (Kuno et al.,998), which allows confirmatory identification of the pathogeny sequencing if desired. The RT-PCR requires only one round ofmplification after cDNA synthesis – no semi-nested PCR step isecessary – and production of an amplicon of 854–863 bp immedi-tely allows determination of whether the fever-causing agent is aember of the Flavivirus. PCR products are then cut by HaeIII and

mplicons from DENV-1, -2, -3 and -4, WN and YF each yield differ-nt restriction patterns. Confirmatory identification of DENV-1 cane performed with an additional restriction digestion using MspI.

. Materials and methods

.1. Collection of samples

Virus isolates from 364 febrile patients, collected 1994–2004ere obtained from the archives of the Virology Department of

he Gorgas Memorial Institute for Health Studies (GMI). These iso-ates had tested positive previously for dengue infection by indirectmmunofluorescence assays (IFA; Section 2.2). An additional 37cute-phase sera were collected from patients in 1999, 2005 and007–2010. Approval of the local ethics committee was deemednnecessary as samples were collected for diagnostic purposes.

nformed patient consent was obtained upon the patient’s agree-ent for sample collection. Six of the acute-phase patients could

pecify the number of days of fever they had experienced prior tohe serum collection date. Five serum samples from healthy volun-eers were cultured and subject to PCR-RFLP as negative controls.

.2. Isolation of virus and characterization of serotypes byndirect immunofluorescence

Between 1994 and 2004, virus was isolated and cultured fromera obtained from febrile patients within the first days of symp-oms, just prior to the end of the fever period. Samples were storedt 4 ◦C during transport to the laboratory where individual sera


ere diluted 1/3 in PBS with gelatin plus 2% penicillin/streptomycinnd 25 �g/mL amphotericin B. A sample volume of 100 �L of serum,iluted to 1 mL, was used to inoculate C6/36 A. albopictus mosquitoell lines which were at 90% confluency after 2 days of plating.

PRESS Methods xxx (2012) xxx– xxx

Inoculated cells were incubated at 33 ◦C. Cytopathic effect wasobserved within 12–14 days. Supernatants from each sampledemonstrating cytopathic effects were extracted in two aliquotsto cryogenic vials for storage at −70 ◦C, and these were used forsubsequent extraction of viral RNA for RT-PCR (Section 2.3.1).

A sample of the culture was taken for immunofluorescence usingstandard methods (2003; PAHO, 2002). FITC-conjugated MABs,used at 1/10 dilution, were a kind gift from the Center for Dis-ease Control. They were obtained from mouse ascites fluid andwere specific for dengue virus types DENV-1, DENV-2, DENV-3 orDENV-4.

2.3. Preparation of viral RNA

2.3.1. Extraction of RNA from viral isolatesA 250 �L aliquot of each viral isolate was extracted using

TRIzol® LS Reagent (Invitrogen, Carlsbad, California, USA), fol-lowing manufacturer’s instructions and the modified method ofChomczynski and Sacchi (1987). Extracted RNA was rehydrated in60 �L molecular-grade H2O and stored at −70 ◦C.

2.3.2. Extraction of RNA from acute-phase seraAcute-phase sera were centrifuged for 2 h at 14,000 × g for

sample concentration and 140 �L of the precipitate was used inextracting viral RNA using the QIAamp Viral RNA Mini Kit (Qiagen,Valencia, California, USA), according to manufacturer’s recommen-dations.

2.4. RT-PCR/RFLP

2.4.1. RT-PCR for FlavivirusThe protocol of Ayers et al. (2006) was used, with primers

FLAVI-1 (5′-AATGTACGCTGATGACACAGCTGGCTGGGACAC-3′) andFLAVI-2 (5′-TCCAGACCTTCAGCATGTCTTCTGTTGTCATCCA-3′) to a854–863 bp region of the non-structural 5 gene (NS5), highly con-served among the flaviviruses. One-Step RT-PCR (Qiagen) wasperformed with 16 �l water, 10 �l 10× Buffer, 10 �l Q-Solution,2 �l dNTPs (10 mM each dNTP), 2 �l Enzyme Mix, 1 �l each(25 pmol) of primers FLAVI-1 and FLAVI-2, and 8 �l of extractedRNA in a 2470 Thermal Applied Biosystem thermocycler (Carlsbad,California, USA). Genetic material for the YFV control sample wasobtained from vaccine isolate 17D RKI with 100% identity to Gen-bank sequence JN628279.1. SLE material, 99% identical to strainGML 903797, Genbank sequence EF158060.1, was obtained froma clinical sample and WNV RNA was obtained from the CARECCaribbean Epidemiology Center and showed 99% identity at thenucleotide level to strain 68856, represented by Genbank sequenceEU249803.1. The programmed cycling was 30 min at 50 ◦C, 15 minat 94 ◦C and then 40 cycles of 94 ◦C for 1 min, 58 ◦C for 1 minand 72 ◦C for 1 min, followed by a final 10 min extension step at72 ◦C. Electrophoresis was performed with 1 �l of the amplifica-tion product and 2.0% agarose gels, 0.5× TBE and ethidium bromide.Amplicons were analyzed on a UV transilluminator.

2.4.2. Restriction enzyme incubationFor restriction enzyme analysis, 3 �l of the unpurified amplifica-

tion product was added to 14.3 �l water, 2 �l Buffer C (10×), 0.2 �lBSA (10 mg/ml) and 0.5 �l HaeIII (10 U/�l) (Promega, Madison, Wis-consin, USA). The reaction was incubated at 37 ◦C for 2 h, after whichrestriction products were subject to 2% agarose gel electrophore-


sis. Distinctive restriction patterns were obtained at this stage forYF, WN and SLE viruses, and DENV-2, DENV-3 and DENV-4. If nodigestion was observed for the amplicon of DENV-1, the samplewas subject to another reaction using the reagents and incubation


ARTICLE IN PRESSG Model

VIRMET-11868; No. of Pages 7

A. Ortiz et al. / Journal of Virological Methods xxx (2012) xxx– xxx 3

Table 1Comparative efficacy of IFA, RT-PCR/RFLP and sequencing to identify dengue serotype in cultured virus.

Serotypeby IFA

No. samples No. tested byRT-PCR/RFLP

Serotype byRT-PCR/RFLPand (n)

No. sequenced Serotype bysequencingand (n)

RT-PCR/RFLPresults confirmingIFA [n (%)]

Sequencing resultsconfirming IFA [n(%)]

Sequencing resultsconfirmingRT-PCR/RFLP [n (%)]

1 235 49 1 (48), 2 (1) 23 1 (22), 2(1) 48/49 (98%) 22/23 (96%) 23/23 (100%)2 59 34 2 (34) 14 1 (1), 2 (13) 33/34 (97%) 13/14 (93%) 14/14 (100%)3 19 18 3 (18) 12 12 18/18 (100%) 12/12 (100%) 12/12 (100%)4 15 12 4 (12) 9 9 12/12 (100%) 9/9 (100%) 9/9 (100%)1-2a 2 1 1 (1) 0 – NAb NAb –1-3a 3 1 1 (1) 0 – NAb NAb –1-2-3-4a 31 8 1 (8) 7 1 (7) NAb NAb 7/7 (100%)

Totals 364 123 65 111/113 (98%) 56/58 (97%) 65/65 (100%)

( classible to

c(

2

Qg(irSllipAtSp

2

lfNg(FlFve

3

oo(dpaRsctms

−) Not tested; NA: not applicable.a Some samples were positive for more than one serotype by IFA and a definitiveb Some IFA results were positive for more than one serotype and it was not possi

onditions described above, but enzyme MspI and Buffer B (10×)Promega), to confirm the specific amplification of the flavivirus.

.5. Confirmatory sequencing

RT-PCR products (Section 2.4.1) were purified using theIAquick PCR Purification kit (Qiagen) and quantified by agaroseel electrophoresis and comparison to the DNA-Quanti LadderOrigene, Rockville, Maryland, USA). Virus from 102 samples (65solates and 37 acute-phase sera) was sequenced. Sequencingeactions were performed with Big Dye 3.1 Terminator Cycleequencing Ready Reaction reagent (Applied Biosystems), fol-owing the manufacturer’s protocol, with consideration for theength and concentration (ng/�l) of the RT-PCR product. Sequenc-ng primers used are described in Ayers et al. (2006). Sequencingroducts were purified by ethanol/EDTA precipitation according topplied Biosystem’s method and sequenced using capillary elec-

rophoresis on a 3130xl Genetic Analyzer (Applied Biosystems).equences were identified using the BLASTn tool with defaultarameters at NCBI.

.6. In silico analysis of other Flavivirus

Several other flaviviruses of emerging importance were ana-yzed in silico. Nucleotide sequences for these viruses were obtainedrom NCBI: Ilheus, NC 009028 (complete genome); Bussuquara,C 009026 (complete genome); Modoc, NC 003635 (completeenome); Aroa, AF013362 (partial CDS of NS5); Naranjal, AF013390partial CDS of NS5) and Jutiapa, AF013379 (partial CDS of NS5).LAVI-1/2 primer binding sites were compared with those pub-ished by Ayers et al. (2006). Analysis was also made of theLAVI-1/2 amplicons generated theoretically for each of theseiruses to determine how many HaeIII and MspI restriction sitesxisted.

. Results

Three hundred and sixty-four isolates reported positive previ-usly for dengue infection by IFA were positive for the presencef flavivirus by RT-PCR amplification with FLAVI-1/2 primersTable 1). Of these samples, 123 were subject to restriction enzymeigestion and 65 were sequenced. Thirty-seven samples of acute-hase sera were also subject to IFA (n = 30), RT-PCR/RFLP (n = 37)nd sequencing (n = 37) (Table 2). Using the simple and rapid PCR-FLP procedure described in this paper, the four dengue viruserotypes, Yellow Fever, West Nile and St. Louis encephalitis viruses


ould be discriminated among in less than 24 h, based on the restric-ion fragment patterns observed after electrophoresis (Fig. 1). No

ajor differences among the 4 dengue serotypes in sensitivity orpecificity of the RT-PCR/RFLP test was observed (Tables 1 and 2).

fication was not possible. confirm their identity by RT-PCR/RFLP or sequencing.

Viruses from both culture and acute-phase sera were testedby the three methods under comparison and yielded similarresults. Using RNA extracted from cultured virus or from virus inacute-phase sera, the technique described in this paper was 98%and 97% specific, respectively, when compared to results obtainedby IFA. For cultured virus, 113 samples which had had a serotypeidentified definitively by IFA were assayed by RT-PCR/RFLP and 111(98%) of the RT-PCR/RFLP results confirmed the serotype identifiedby IFA (Table 1). For acute-phase human sera, 29 samples whichhad had a serotype identified definitively by IFA were assayed byRT-PCR/RFLP and 28 (97%) of the RT-PCR/RFLP results confirmedthe serotype identified by IFA (Table 2).

The technique described above was also 100% specific whencompared to results obtained by gene sequencing. Of the total 65samples of cultured virus that were gene sequenced and for whichRT-PCR/RFLP results existed, all sequencing results confirmed theserotype identified by RT-PCR/RFLP (Table 1). A similar result wasachieved when using acute-phase sera: all 37 sequencing resultsconfirmed the serotype identified by RT-PCR/RFLP (Table 2). Bycomparison, 58 cultured virus samples and 29 acute-phase serumsamples which had had a serotype identified definitively by IFAwere gene sequenced and 56 (97%) for cultured virus, and 28 (97%)for sera, of the sequencing results confirmed the serotype identi-fied by IFA (Tables 1 and 2). The few mismatches are thought tobe due to errors in the IFA analysis, which occurred relatively earlyin the study (1999, 2001 and 2004), and not in the RT-PCR/RFLPtechnique, which was repeated for samples which demonstratedresults at variance with IFA.

Using acute-phase sera from 2009 and 2010, a level of sensitivityfor the RT-PCR/RFLP procedure was observed that also demon-strates the clinical utility of this method. Five samples collected onday 3 of fever (2009, n = 4; 2010, n = 1) and one sample collected onday 5 of fever (2009) yielded clear RFLP results: adequate amountsof virus were present in the serum to be amplified by the methoddescribed in this paper, and to provide a physician with a moleculardiagnosis of the febrile illness within 4 days of its onset, withoutneeding to perform gene sequencing to identify the virus (Table 2).

For regional interest, the dengue serotypes observed to be incirculation in Panama for the years 1994–2005 and 2007–2010were also analyzed (data not shown). The results obtained byRT-PCR/RFLP and sequencing are mostly in agreement with theserotypes found to be circulating in Panama published previouslyfor 1994–2005 (Larrú-Martínez et al., 2006; PAHO, 2000), but due tothe small sample size used in this study, not all the same serotypesfor each year are represented in this study. For the first time, thedengue serotypes in circulation in Panama for 2007–2010 (Table 2)


were investigated, with the findings that serotype 3 has been preva-lent until 2010.

In silico analysis of the Ilheus, Bussuquara, Modoc, Aroa, Naran-jal and Jutiapa flaviviruses for the FLAVI-1/2 primer binding




4 A. Ortiz et al. / Journal of Virological Methods xxx (2012) xxx– xxx

Table 2Comparative efficacy of IFA, RT-PCR/RFLP and sequencing to identify dengue serotype in acute-phase human sera.

Serotypeby IFA

No. samples No. tested byRT-PCR/RFLP

Serotype byRT-PCR/RFLPand (n)

No. sequenced Serotype bysequencingand (n)

RT-PCR/RFLPresults confirmingIFA [n (%)]

Sequencing resultsconfirming IFA [n(%)]

Sequencing resultsconfirmingRT-PCR/RFLP [n (%)]

1 7 7 6(2), 4(1) 7 6(2), 4(1) 6/7 (86%) 6/7 (86%) 7/7 (100%)2 2 2 2 (2) 2 2 (2) 2/2 (100%) 2/2 (100%) 2/2 (100%)3 19 19 3 (19) 19 3 (19) 19/19 (100%) 19/19 (100%) 19/19 (100%)4 1 1 4 (1) 1 4 (1) 1/1 (100%) 1/1 (100%) 1/1 (100%)– 1 8 3 (8) 8 3 (8) NA NA 8/8 (100%)

( o confi

sdoFfltt2vt(

F11

F(Cta

Totals 30 37 37

–) Not tested; NA: not applicable: IFA was not performed and it was not possible t

ites of Ayers et al. (2006) in the NS5 gene sequences availableemonstrated that some of these viruses would be amplified the-retically under the conditions described in this paper (Appendixig. A1). Analysis of the theoretical FLAVI-1/2 amplicons of theseaviviruses for HaeIII and MspI restriction sites showed thathe Aroa, Jutiapa, Ilheus and Modoc viruses yielded distinc-ive high-molecular weight bands when compared to DENV-1,, 3 and 4, Yellow Fever, West Nile or St. Louis encephalitis


iruses, but that the Naranjal and Bussuquara viruses gave restric-ion patterns that are difficult to distinguish from one anotherAppendix Figs. A2 and A3).

ig. 1. Representative restriction fragments observed for each virus subtype following cD000 bp to 100 bp in 100 bp increments; 1: DENV-1: 750 bp; 2: DENV-2: 430 bp, 180 bp, 1270 bp, 100 bp; 6: WN: 350 bp, 150 bp, 120 bp; 7: YF: 420 bp; 8: Negative PCR control. (B)

ig. A1. FLAVI-1/2 primer binding sites for emerging American flaviviruses. NCBI accecomplete genome); Modoc, NC 003635 (complete genome); Aroa, AF013362 (partial CDS of NS5). Note: no sequence data for the FLAVI-2 primer binding site was availablehe end of the sequence available is given in parentheses, indicating how much of the FLpproximately 25–34 bp missing; Naranjal (832 bp), approximately 22–31 bp missing; or

28/29 (97%) 28/29 (97%) 37/37 (100%)

rm identity by RT-PCR/RFLP or sequencing.

4. Discussion

The technique described in this paper distinguishes DENV-1,2, 3 and 4 from one another, and from YF, WN and SLE viruses.It gives a definitive result within 24 h and does not require genesequencing, radioactive isotopes, computer analysis or real-timePCR and its specialized reagents. It is a specific and rapid discrimi-natory test for the flaviviruses and is already proving useful in the


Central Reference Laboratory of the Gorgas Memorial Institute ofPanama, a region of endemicity known to harbor the four dengueserotypes.

NA digestion. (A) HaeIII digestion: MWM (molecular weight marker): 1500 bp and0 bp; 3: DENV-3: 460 bp, 390 bp; 4: DENV-4: 580 bp, 170 bp, 110 bp; 5: SLE: 520 bp,

MspI digestion. MWM (molecular weight marker): 1: DENV-1: 620 bp, 220 bp.

ssion numbers: Ilheus, NC 009028 (complete genome); Bussuquara, NC 009026DS of NS5); Naranjal, AF013390 (partial CDS of NS5); Jutiapa, AF013379 (partial

for three viruses. The sequence length from the 5′ end of the FLAVI-1 primer toAVI-1/2 amplicon, approximately 854–863 bp, is missing: NS5 gene: Aroa (829 bp)

Jutiapa (808 bp), approximately 46–55 bp missing.




A. Ortiz et al. / Journal of Virological Methods xxx (2012) xxx– xxx 5

Fig. A2. HaeIII restriction fragments for emerging American flaviviruses. A comparison of the restriction fragments known to be produced for dengue serotypes one to four,Yellow Fever, West Nile and St. Louis encephalitis viruses with predicted restriction patterns for emerging American flaviviruses based on the NS5 sequence data availablei ppenb er anN

db1tsgbdarvdpo

ptI

n the NCBI databases. The origins of the sequences are described in Fig. A1 of this aands that are indicated to exist over a range, 377–411 bp for example, are the lowS5 gene. All units are in base pairs.

A limited number of other protocols using RT-PCR/RFLP for theetection and identification of dengue or other flaviviruses haveeen described previously (Gaunt and Gould, 2005; Sudiro et al.,997; Vorndam et al., 1994), but the technique of RT-PCR/RFLPo detect and identify dengue has been largely superseded byerological methods or techniques such as real-time PCR andene sequencing. However, there is an important role for simple,ut rapid molecular methods such as RT-PCR/RFLP in diagnosingengue and other flaviviruses, especially in resource-poor settings,nd this work has demonstrated that the assay described fulfillsecently published specifications for an ideal dengue test – pro-iding rapid results; being inexpensive and easy to use; allowingistinction between dengue and other diseases with similar clinicalresentation, and showing high sensitivity during the acute stagef infection (Peeling et al., 2010).


In the Americas, the most significant of the human flavivirusathogens are dengue, Yellow Fever virus, St. Louis encephali-is virus and the more recent introduction, West Nile virus.n Panama, season-associated dengue outbreaks are an annual

dix. For Aroa, Naranjal and Jutiapa viruses, NS5 sequence data was incomplete andd upper limits of bands that could be produced from the incomplete 3′ end of the

problem, with all four dengue serotypes observed in this coun-try since 1993 (PAHO, 2000); WN has not yet been detected; SLEhas been isolated from mosquitoes, birds (Charrel et al., 1999)and humans (unpublished results, Julio Cisneros, Gorgas Memo-rial Institute of Health Studies, May 2011); and YF has not beenreported since prior to 1960, although it persists in neighboringColombia (PAHO, 2007).

The test presented in this paper can distinguish among the fourdengue serotypes, SLE, WN and YF in 24 h, using human sera andstandard PCR and agarose gel electrophoresis equipment. It alsois capable of detecting virus by days 3–5 of fever, which is usefulfor two reasons: firstly, at the earlier time of detection (3 days offever), the test allows diagnosis of the causative pathogen in febrileillness on only the fourth day of clinical disease and potentiallybefore the onset of DHF/DSS; secondly, at the later time of detection,


the test remains sufficiently sensitive to detect decreasing viremiaThe analysis of virus from both culture or acute-phase sera usingup to three methods and each viral source yielded similar results.This demonstrates the robustness of the viral extraction steps in




6 A. Ortiz et al. / Journal of Virological Methods xxx (2012) xxx– xxx

Fig. A3. MspI restriction fragments for emerging American flaviviruses. A comparison of the restriction fragments known to be produced for dengue serotypes one to four,Yellow Fever, West Nile and St. Louis encephalitis viruses with predicted restriction patterns for emerging American flaviviruses based on the NS5 sequence data availablein the NCBI databases. The origins of the sequences are described in Fig. A1 of this appendix. For Aroa, Naranjal and Jutiapa viruses, NS5 sequence data was incomplete andb er an ′

N

peplcado

mstvtan

ands that are indicated to exist over a range, 757–766 bp for example, are the lowS5 gene. All units are in base pairs.

reparing RNA from different origins for RT-PCR and indicates thatxtending the assay for use in field surveillance of the flavivirusesrevalent in mosquitoes is possible, a future goal for this work. Pre-

iminary results demonstrate that the technique described abovean detect dengue virus in field-collected pools of A. albopictus,

vector of increasing importance in the transmission of urbanengue in Panama (unpublished results, Indira Espino, Universityf Panama).

This assay is under continuing development in order to deter-ine the end-point sensitivity for detecting virus from acute-phase

erum, its specificity in mixed infections, and its use for the detec-ion of other members of the Flavivirus genus. In addition to the


iruses tested with the assay described, several poorly charac-erized flaviviruses which are emerging diseases of humans andnimals in South and Central America and for which Panama alsoeeds to be vigilant were also analyzed in silico: Aroa species from

d upper limits of bands that could be produced from the incomplete 3 end of the

Venezuela and the Aroa group viruses, Naranjal from the Ecuador,and South American Bussuquara; Jutiapa virus from the Modocsub-genus in Guatemala; and Ilheus virus from Central and SouthAmerica, and Trinidad (Kuno et al., 1998) Further work on thisaspect of the assay is required.

Conflicts of interest

The authors have no conflicts of interest to declare.

Acknowledgements


This research was supported by SENACYT (National Science andTechnology Council) and the Instituto Conmemorativo Gorgas Estu-dios de la Salud under grants FID05-152 and FID10-127, and FondoNacional de Inversión, project 1.11.1.3.501.01.01 (Improvement of


ING Model

V

ogical

eibpsadsM

A

R

2

A

C

C

D

G

G

H

I

Methods 48, 237–244.Weaver, S.C., Reisen, W.K., 2010. Present and future arboviral threats. Antiviral

ARTICLEIRMET-11868; No. of Pages 7

A. Ortiz et al. / Journal of Virol

pidemiological surveillance of emerging and re-emerging diseasesn human populations on the Colombia-Panama Border). Fundingodies had no role in study design; collection, analysis or inter-retation of the data; in writing this manuscript; or in deciding toubmit the paper for publication. The Center for Disease Controlnd Prevention (Denver, Colorado and Puerto Rico) was the kindonor of FITC-conjugated MABs for the detection of specific denguetrains. The authors thank Alexander A. Martinez and Juan Castillo. for their assistance with gene sequencing.

ppendix A.

See Figs. A1–A3.

eferences

003. Instituto de Pedro Kourí. Departmento de Virología. Manual detécnicas de laboratorio para el estudio de los virus del dengue,http://www.ipk.sld.cu/biblioweb/monographs.htm.

yers, M., Adachi, D., Johnson, G., Andonova, M., Drebot, M., Tellier, R., 2006. A singletube RT-PCR assay for the detection of mosquito-borne flaviviruses. Journal ofVirological Methods 135, 235–239.

harrel, R.N., Levy, N., Tesh, R.B., Chandler, L.J., 1999. Use of base excision sequencescanning for detection of genetic variations in St. Louis encephalitis virus iso-lates. Journal of Clinical Microbiology 37, 1935–1940.

homczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochem-istry 162, 156–159.

as, S., Pingle, M.R., Munoz-Jordan, J., Rundell, M.S., Rondini, S., Granger, K., Chang,G.J., Kelly, E., Spier, E.G., Larone, D., Spitzer, E., Barany, F., Golightly, L.M., 2008.Detection and serotyping of dengue virus in serum samples by multiplex reversetranscriptase PCR-ligase detection reaction assay. Journal of Clinical Microbiol-ogy 46, 3276–3284.

aunt, M.W., Gould, E.A., 2005. Rapid subgroup identification of the flavivirusesusing degenerate primer E-gene RT-PCR and site specific restriction enzymeanalysis. Journal of Virological Methods 128, 113–127.

omes, A.L., Silva, A.M., Cordeiro, M.T., Guimaraes, G.F., Marques Jr., E.T., Abath,F.G., 2007. Single-tube nested PCR using immobilized internal primers for theidentification of dengue virus serotypes. Journal of Virological Methods 145,76–79.

ue, K.D., Tuan, T.V., Thi, H.T., Bich, C.T., Anh, H.H., Wills, B.A., Simmons, C.P., 2011.Validation of an internally controlled one-step real-time multiplex RT-PCR assay


for the detection and quantitation of dengue virus RNA in plasma. Journal ofVirological Methods 177, 168–173.

CTV, 2005. Virus Taxonomy. Classification and Nomenclature of Viruses – EighthReport of the International Committee on Taxonomy of Viruses. Elsevier Aca-demic Press, San Diego.

PRESS Methods xxx (2012) xxx– xxx 7

Kuno, G., Chang, G.J., Tsuchiya, K.R., Karabatsos, N., Cropp, C.B., 1998. Phylogeny ofthe genus Flavivirus. Journal of Virology 72, 73–83.

Kuno, G., Gubler, D.J., Velez, M., Oliver, A., 1985. Comparative sensitivity of threemosquito cell lines for isolation of dengue viruses. Bulletin of World HealthOrganisation 63, 279–286.

Larrú-Martínez, B., Quiroz, E., Bellón, J.M., Esquivel, R., Guevara, J.N., Sáez-Llorens,X., 2006. Dengue pediátrico en Panamá. Annales de Pediatrie (Barcelona) 64,517–522.

Martina, B.E., Koraka, P., Osterhaus, A.D., 2009. Dengue virus pathogenesis: an inte-grated view. Clinical Microbiology Reviews 22, 564–581.

Mathew, A., Rothman, A.L., 2008. Understanding the contribution of cellu-lar immunity to dengue disease pathogenesis. Immunological Reviews 225,300–313.

Nascimento, H.P., Oliveira, M.D., de Melo, C.P., Silva, G.J., Cordeiro, M.T., Andrade, C.A.,2011. An impedimetric biosensor for detection of dengue serotype at picomo-lar concentration based on gold nanoparticles-polyaniline hybrid composites.Colloids and Surfaces B: Biointerfaces 86, 414–419.

Nielsen, D.G., 2009. The relationship of interacting immunological components indengue pathogenesis. Virology Journal 6, 211.

PAHO, 2000. Dengue in the Americas: the epidemics of 2000. Epidemiological Bul-letin 21, 4–8.

PAHO, 2002. Manual de procedimientos de técnicas para el diagnóstico del Dengue.,http://www.bvsde.paho.org/bvsacd/cd20/dengue.pdf.

PAHO, 2007. Yellow Fever: Number of Cases and Deaths, 1960–2007. PanAmerican Health Organization, http://www.paho.org/English/AD/DPC/CD/yf-fa-1960-2007.pdf (22 April 2011).

Peeling, R.W., Artsob, H., Pelegrino, J.L., Buchy, P., Cardosa, M.J., Devi, S., Enria, D.A.,Farrar, J., Gubler, D.J., Guzman, M.G., Halstead, S.B., Hunsperger, E., Kliks, S.,Margolis, H.S., Nathanson, C.M., Nguyen, V.C., Rizzo, N., Vazquez, S., Yoksan, S.,2010. Evaluation of diagnostic tests: dengue. Nature Reviews Microbiology 8,S30–S38.

Puttikhunt, C., Prommool, T., Ong-ajchaowlerd, N.U.t., Yoosook, P., Tawilert, K.,Duangchinda, C., Jairangsri, T., Tangthawornchaikul, A., Malasit, N., Kasinrerk W,P., 2011. The development of a novel serotyping-NS1-ELISA to identify serotypesof dengue virus. Journal of Clinical Virology 50, 314–319.

Saxena, P., Dash, P.K., Santhosh, S.R., Shrivastava, A., Parida, M., Rao, P.L., 2008.Development and evaluation of one step single tube multiplex RT-PCR for rapiddetection and typing of dengue viruses. Virology Journal 5, 20.

Sudiro, T.M., Ishiko, H., Green, S., Vaughn, D.W., Nisalak, A., Kalayanarooj, S., Roth-man, A.L., Raengsakulrach, B., Janus, J., Kurane, I., Ennis, F.A., 1997. Rapiddiagnosis of dengue viremia by reverse transcriptase-polymerase chain reac-tion using 3′-noncoding region universal primers. American Journal of TropicalMedicine and Hygiene 56, 424–429.

Vorndam, V., Kuno, G., Rosado, N., 1994. A PCR-restriction enzyme technique fordetermining dengue virus subgroups within serotypes. Journal of Virological


Research 85, 328–345.WHO, 2011. Impact of Dengue: World Health Organization., http://www.who.int/

csr/disease/dengue/impact/en/index.html (19 April 2011).


http://www.ipk.sld.cu/biblioweb/monographs.htm

Simple, specific molecular typing of dengue virus isolates using one-step RT-PCR and restriction fragment length polymorphism

Documents