AA

MSVa

b

c

d

ARRAA

KDYRG

1

twh(ascnNo(

m

h0

Journal of Virological Methods 226 (2015) 40–51

Contents lists available at ScienceDirect

Journal of Virological Methods

j o ur nal ho me pag e: www.elsev ier .com/ locate / jv i romet

nalysis of a Reverse Transcription Loop-mediated Isothermalmplification (RT-LAMP) for yellow fever diagnostic

arcio R.T. Nunesa,∗,1, João Lídio Vianez Jr. a,1, Keley N.B. Nunesa,andro Patroca da Silvaa, Clayton P.S. Limaa, Hilda Guzmanb, Lívia C. Martinsc,aléria L. Carvalhoc, Robert B. Teshb, Pedro F.C. Vasconcelosc,d

Center for Technological Innovation, Evandro Chagas Institute, Brazilian Ministry of Health, Ananindeua, Pará, BrazilDepartment of Pathology, University of Texas Medical Branch, Galveston, TX, USADepartment of Arbovirology and Hemorrhagic Fevers, Evandro Chagas Institute, Brazilian Ministry of Health, Ananindeua, Pará, BrazilDepartment of Pathology, University of Para State, Belem, Brazil

rticle history:eceived 22 July 2015eceived in revised form 6 October 2015ccepted 7 October 2015vailable online 13 October 2015

eywords:iagnostic

a b s t r a c t

Yellow Fever virus (YFV) is an important human pathogen in tropical areas of Africa and South America.Although an efficient vaccine is available and has been used since the early 1940s, sylvatic YFV transmis-sion still occurs in forested areas where anthropogenic actions are present, such as mineral extraction,rearing livestock and agriculture, and ecological tourism. In this context, two distinct techniques basedon the RT-PCR derived method have been previously developed, however both methods are expensivedue to the use of thermo cyclers and labeled probes. We developed isothermal genome amplification,which is a rapid, sensitive, specific and low cost molecular approach for YFV genome detection. This assay

ellow fever virusT-LAMPenome detection

used a set of degenerate primers designed for the NS1 gene and was able to amplify, within 30 min inisothermal conditions, the YFV 17D vaccine strain derived from an African wild prototype strain (Asibi),as well as field strains from Brazil, other endemic countries from South and Central America, and theCaribbean. The generic RT-LAMP assay could be helpful for YFV surveillance in field and rapid responseduring outbreaks in endemic areas.

. Introduction

Yellow fever virus (YFV) is an arthropod-borne virus belongingo the Family Flaviviridae, genus Flavivirus (Dietzgen et al., 2012),here it is the prototype species. It is one of the most importantuman pathogens in tropical areas of Africa and South AmericaFerguson et al., 2010; Jentes et al., 2011; Monath, 2001; Monathnd Vasconcelos, 2015). The viral genome is composed of a singletranded, positive sense RNA with approximately 11 kb in lengthomposed by 10 genes, three structural (E, prM and E) and sevenon structural (NS) namely NS1, NS2a, NS2b,NS3, NS4, NS4b andS5, which encodes ten proteins with the same name, and the

pen reading frame is flanked by two non-coding regions (NCR)Dietzgen et al., 2012).

∗ Corresponding author.E-mail addresses: marcionunes@iec.pa.gov.br,

arcionunesbrasil@yahoo.com.br (M.R.T. Nunes).1 These authors contributed equally.

ttp://dx.doi.org/10.1016/j.jviromet.2015.10.003166-0934/© 2015 Elsevier B.V. All rights reserved.

© 2015 Elsevier B.V. All rights reserved.

YFV is the etiological agent of yellow fever, an ancient arbovi-ral disease that originated in Africa and was introduced into theAmericas during the slave’s trade in the 17th century (Bryantet al., 2007; Nunes et al., 2012). YFV is maintained in nature inAfrica through three distinct cycles (urban, sylvatic and interme-diate) and in Americas through two different cycles (urban andsylvatic). In Africa, YFV is mainly transmitted by the Aedes sp.mosquitoes, viz. Aedes africanus, Aedes furcifur and Aedes simpsoni.Aedes aegypti, Haemagogus leucocelaenus and Haemagogus janthi-nomys are the main vectors in the Americas (Ferguson et al.,2010; Monath, 2001; Gardner and Ryman, 2010; Cardoso et al.,2010).

Approximately 85% of YFV infections are asymptomatic or mild,and are characterized by sudden onset of fever, chills, severeheadache, back pain, general arthralgia, nausea, vomiting, fatigue,and weakness with a clinical picture that resembles dengue orinfluenza. The classic severe picture is detected in about 15%

of infected individuals; these patients present with fever, jaun-dice, hemorrhagic diathesis, and eventually shock, and failure ofmultiple organs especially liver and renal failure that result in case-fatality rates (CFR) of about 50%. The overall CFR of yellow fever is

irolog

1tvlofditlrcDtuWfe

ssemdaJ(euctt(wRcvat

2

2

Coa#crNAwrgM

2

6no4

M.R.T. Nunes et al. / Journal of V

0–20% (Monath and Vasconcelos, 2015; Vasconcelos, 2003) Withhe strong vector control measures during introduction of the YFVaccine in 1942 (Frierson, 2010; Norrby, 2007), the disease wasargely controlled in urban areas of tropical America. However casesr limited outbreaks of sylvatic yellow fever are still detected inorested areas of the Amazon region and central region of Brazilue to exposure of susceptible (non immunized) people to the virus

n endemic areas, and also to disruption of natural ecosystems inhese areas (Vasconcelos et al., 2001). Cases or clusters of urban yel-ow fever were detected in Bolivia and Paraguay in 1997 and 2009espectively, after several decades of absence of yellow fever urbanases (Ferguson et al., 2010; Monath and Vasconcelos, 2015; Vaner Stuyft et al., 1999). Although of great public health importance,

he majority of molecular test protocols developed for yellow feverse probes, thermo cyclers (Dash et al., 2012; Nunes et al., 2011;eidmann et al., 2010) or have been developed for African Yellow

ever strains (Domingo et al., 2012; Escadafal et al., 2014; Kwallaht al., 2013).

The isothermal approach is a simple method that combines aet of non-labeled primers, conventional deoxynucleotides and apecific polymerase able to unfold the double DNA strand (Notomit al., 2000). The Reverse Transcription Loop-mediated Isother-al Amplification (RT-LAMP) has been previously developed to

etect other viruses including influenza and various other distinctrboviruses viz. dengue (DENV-1 to 4), West Nile, Chikungunya andapanese Encephalitis viruses, as well as for African YFV strainsKalvatchev et al., 2010; Lu et al., 2012; Parida et al., 2004; Poont al., 2005; Toriniwa and Komiya, 2006). This approach has beensed as an alternative method for genome detection. It is low-ost, rapid, accurate and sensitive (Mori and Notomi, 2009). Inhis work, a total of 43 YFV complete genomes retrieved fromhe GenBank database were used to develop the RT-LAMP assayRT-dLAMP) with degenerate primers. This approach was comparedith previous specific African strain RT-LAMP method (RT-sLAMP),T-PCR, RT-hemi-nested-PCR, RT-qPCR, and virus isolation. Theurrent method using degenerate primers was able to detect alliral isolates recovered from human cases, non-human primatesnd mosquitoes, as well as YFV samples obtained from experimen-ally infected hamsters.

. Material and methods

.1. Ethics statements

The study protocol was approved by the Animal Care Ethicsommittee of the Evandro Chagas Institute. The experimentsbserved the principles of the Brazilian Laws and Regulation,nd the authorization for use was obtained through the protocol061/2009. VERO cells used in the study were provided by the cellulture collection of the Department of Arbovirology and Hemor-hagic Fevers, World Health Organization Collaborating Center andational Reference Center for Training and Research in Arbovirus,nanindeua, Para, Brazil. All procedures involving virus strainsere performed in BSL2/BSL3 facilities under the National Biosafety

ules and the Institutional Biocontainment manual (Evandro Cha-as Institute, Brazilian Ministry of Health, Brasília; Editora doinistério da Saúde, 2015. 134 p, protocol no. BR275.1; PCIEC2015).

.2. Panel of RNA samples

A total of 130 RNA samples were prepared, corresponding to

5 samples obtained from low passage YFV strains isolated inewborn mice and passed once in VERO cells, 25 RNA samplesbtained from DENV (5 DENV-1, 5 DENV-2, 5 DENV-3, and 5 DENV-), infected C6/36 cells, 5 RNAs from other flavivirus (Bussuquara

ical Methods 226 (2015) 40–51 41

virus, Cacipacoré, virus, Ilhéus virus, Saint Louis encephalitis virus,and Naranjal-like virus), 15 RNA samples from supernatants VEROcells infected with Oropouche virus (OROV), 15 RNAs obtainedfrom supernatants of VERO cells infected with Mayaro virus, andfive negative controls corresponding to RNA extracted from noninfected VERO cells. YFV strains from South American countries(except Brazil) and Caribbean region (Table 1) were cultured atthe Department of Pathology, University of Texas Medical Branch,Galveston, and used for RNA extraction using the Trizol-Qiagen pro-tocol (Qiagen, 2003; Rio et al., 2010). Total RNA samples were sentto the Center for Technological Innovation (CTI), Evandro ChagasInstitute, Ministry of Health, Brazil. YFV Brazilian strains were cul-tured in the Department of Arbovirology and Hemorrhagic Fevers,Evandro Chagas Institute using Biosafety Level 3 (BSL3) facilities,and RNA samples were obtained at the CTI, Genomic Core, as previ-ously described (Qiagen, 2003; Rio et al., 2010). RNA samples wereused for RT-LAMP using degenerate primers (RT-dLAMP) evalua-tion in comparison to four other methods: RT-Hemi-Nested-PCR,RT-PCR, RT-qPCR (Nunes et al., 2011) and with the current RT-LAMP assay based on NS1 gene sequence of YFV 17D strain hereafternamed as RT-specific LAMP (RT-sLAMP) (Khunthong et al., 2013).

2.3. Hamster (Mesocricetus auratus) samples

A total of 120 young hamsters (10-days old) were divided intwo main groups named YFV group and non YFV group, composedof 60 animals per group. The YFV group was used for experimen-tal infection with the BeH 111 YFV strain (Brazilian prototypestrain) provided by the Department of Arbovirology and Hemor-rhagic Fevers, while the non YFV group was used as a control group.The experiments were conducted in BSL3 facilities according to theInstitutional Biosafety rules and approved by the Ethics Committeeon experiments with animals (CEUA/IEC). Hamsters were infectedby intraperitoneal route with a virus dose of 1.9 × 105 in a volumeof 0.2 mL, separated in cages with six animals each and observeddaily for a period of 10 days. Blood samples were collected dailyfrom both groups (YFV and non YFV groups) between days 1–10post infection. Blood samples were used to obtain sera used forRNA extraction and for testing the RT-dLAMP and other molecularmethods.

2.4. Primer design

Specific set of primers were designed based on 43 YFV com-plete genomes obtained from YFV isolates from Brazil (n = 12),Africa (n = 30), and Trinidad (n = 01), available at the GenBankdatabase (http://www.ncbi.org.nih.com) using the online opensource Primer Explores V.4 software (http://primerexplorer.jp/e/). The sequences were grouped into three distinct clusters based ongeographic locations in South American group, Central Americanand the Caribbean group, and African group, and were aligned usingthe Geneious v. R-7 (Kearse et al., 2012). Consensus sequences foreach of the established YFV groups were used for re-alignment andproduction of a unique consensus of YFV sequences. Degenerate setof primers were designed and denominated YFVF3 outer primer(TCCACACCYTGGAGRCAYTR), YFV B3 reverse primer (GYCCAT-CACAGYYGCCRTCA), YFV Fic (GRCCTCCGATTGAYCTCGGC+TTT),YFV F2 forward inner primer (ARTGTGARTGGCCRCTGAC), BIPreverse inner primer YFV B2 (GGTYCAGACRAACGGACCTTGG+TTT),

BIP reverse inner primer B1c (YCCTGGGCAAGCTTCTCT), YFV LFforward loop primer (CTTCAACTGATGTTCCAATCGTATG), YFV LBreverse loop primer (ATGCAGGTRCCACTAGAAGTGA). Primers wereused for RT-dLAMP assays (Fig. 1)

42 M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51

Table 1Yellow fever virus strains and other arboviruses RNAs used for evaluation of RT-dLAMP-PCR assay in comparison to other molecular methods.

Group Reference sample no. Strain Place of isolation Host association No. of samples

YFV group

YFV1 JR 35 Bolivia humanYFV2 FVB-0196 Bolivia humanYFV3 92-99 Bolivia humanYFV4 OBS 8027 Bolivia humanYFV5 OBS 7549 Bolivia humanYFV6 OBS 8026 Bolivia humanYFV7 JSS Brazil humanYFV8 BeAR378600 Brazil Haemagogus spYFV9 BeH394880 Brazil humanYFV10 BeH413820 Brazil humanYFV11 BeH422973 Brazil humanYFV12 BeH423602 Brazil humanYFV13 BeH463676 Brazil humanYFV14 BeAR513008 Brazil Sabethes chloropterusYFV15 BeH526722 Brazil humanYFV16 BeH622205 Brazil humanYFV17 BeH622493 Brazil humanYFV18 BeAR646536 Brazil (Haemagogus leucocelaenus)YFV19 BeH655417 Brazil humanYFV20 INS 347613 Colombia humanYFV21 V 528A Colombia humanYFV22 INS 382060 Colombia humanYFV23 1337 Ecuador humanYFV24 OBS 5026 Ecuador humanYFV25 1345 Ecuador humanYFV26 OBS 5041 Ecuador humanYFV27 690 French Guyana humanYFV28 Jimenez GML Panama humanYFV29 Jimenez Panama humanYFV30 OBS 7904 Peru humanYFV31 OBS 2240 Peru humanYFV32 HEB 4248 Peru humanYFV33 OBS 6745 Peru humanYFV34 1362/77 Peru humanYFV35 HEB 4245 Peru humanYFV36 HEB 4224 Peru humanYFV37 IQD 8393 Peru humanYFV38 FMD 1240 Peru humanYFV39 1368-77 Peru humanYFV40 OBS 2250 Peru humanYFV41 ARV 0544 Peru humanYFV42 06-15094-99 Peru humanYFV43 HEB 4246 Peru humanYFV44 1371-77 Peru humanYFV45 ARV 0548 Peru humanYFV46 HEB 4240 Peru humanYFV47 03-5350-98 Peru humanYFV48 1914-81 Peru monkeyYFV49 HEB 4236 Peru humanYFV50 06-13015-99 Peru humanYFV51 #1 Peru humanYFV52 TRVL 4205 Trinidad Monkey (Alouatta seniculus)YFV53 CAREC 790882 Trinidad mosquito (Haemagogus janthinomys)YFV54 CAREC 9514929 Trinidad mosquito (Haemagogus janthinomys)YFV55 SVM 3-18-09 Trinidad Monkey (Alouatta sp)YFV56 CAREC 891957 Trinidad monkeyYFV57 CAREC M2-09 [09-00-122] Trinidad monkey (Alouatta sp)YFV58 CAREC 797984 Trinidad humanYFV59 CAREC M3-09 Trinidad monkey (Alouatta sp)YFV60 Trinidad monkey (Alouatta sp)YFV61 PHO 42 8? H Venezuela humanYFV62 P128 MC Venezuela Monkey (Alouatta seniculus)YFV63 INH 35708 Venezuela humanYFV64 INH-35720 Venezuela humanYFV 65 17DD ASIBI derived vaccine strain NA 65

M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51 43

Table 1 (Continued)

Group Reference sample no. Strain Place of isolation Host association No. of samples

Non YFV group

NYFV1 DENV-1 (H547625) Brazil humanNYFV2 DENV-1 (H550175) Brazil humanNYFV3 DENV-1 (H551022) Brazil humanNYFV4 DENV-1 (H611377) Brazil humanNYFV5 DENV-1 (H622822) Brazil humanNYFV6 DENV-2 (H 674704) Brazil humanNYFV7 DENV-2 (H 676618) Brazil humanNYFV8 DENV-2 (H 666995) Brazil humanNYFV9 DENV-2 (H 660059) Brazil humanNYFV10 DENV-2 (H 655259) Brazil humanNYFV11 DENV-3 (H 702980) Brazil humanNYFV12 DENV-3 (H 704582) Brazil humanNYFV13 DENV-3 (H 707629) Brazil humanNYFV14 DENV-3 (H 712120) Brazil humanNYFV15 DENV-3 (H 707877) Brazil humanNYFV16 DENV-4 (H 774846) Brazil humanNYFV17 DENV-4 (H 780090) Brazil humanNYFV18 DENV-4 (H 780120) Brazil humanNYFV19 DENV-4 (H 772846) Brazil humanNYFV20 DENV-4 (H 772852) Brazil humanNYFV21 BSQV (AN 4116) Brazil Alouatta belzebulNYFV22 CPCV (AN 327600) Brazil Percnostola rufrifronsNYFV23 ILHV (H 7445) Brazil humanNYFV24 SLEV (AR263600) Brazil Culex declaratorNYFV25 NRLV (AN70170) Brazil Methachirus opossumNYFV26 OROV (H 389865) Brazil humanNYFV27 OROV (H 390233) Brazil humanNYFV28 OROV (H 390242) Brazil humanNYFV29 OROV (H 472433) Brazil humanNYFV30 OROV (H 472435) Brazil humanNYFV31 OROV (H 472200) Brazil humanNYFV32 OROV (H 472204) Brazil humanNYFV33 OROV (H 475248) Brazil humanNYFV34 OROV (H 498913) Brazil humanNYFV35 OROV (H 505442) Brazil humanNYFV36 OROV (H 505663) Brazil humanNYFV37 OROV (H 505764) Brazil humanNYFV38 OROV (H 505768) Brazil humanNYFV39 OROV (H 505805) Brazil humanNYFV40 OROV (H 504514) Brazil humanNYFV41 MAYV (H 505372) Brazil humanNYFV42 MAYV (H 505465) Brazil humanNYFV43 MAYV (H 394874) Brazil humanNYFV44 MAYV (H 342906) Brazil humanNYFV45 MAYV (H 342925) Brazil humanNYFV46 MAYV (H 342915) Brazil humanNYFV47 MAYV (H 342916) Brazil humanNYFV48 MAYV (AR745587) Brazil Haemagogus janthinomysNYFV49 MAYV (AR301133) Brazil Haemagogus sppNYFV50 MAYV (AR327360) Brazil Haemagogus sppNYFV51 MAYV (AR 344910) Brazil Haemagogus janthinomysNYFV52 MAYV (H744141) Brazil humanNYFV53 MAYV (H743921) Brazil humanNYFV54 MAYV (H744173) Brazil humanNYFV55 MAYV (AN 345007) Brazil Monkey (Callithrix sp)NYFV56 NC NA NANYFV57 NC NA NANYFV58 NC NA NANYFV59 NC NA NANYFV60 NC NA NANYFV61 NC NA NANYFV62 NC NA NANYFV63 NC NA NANYFV64 NC NA NANYFV65 NC NA NA 65

Total no. of samples 130

YFV: Yellow fever virus; DENV: Dengue virus; DENV-1: Dengue virus type 1; DENV-2: Dengue virus type 2; DENV-3: Dengue virus type 3; DENV-4: Dengue virus type 4;BSQV: Bussuquara virus; CPQV: Cacipacoré virus; ILHV: Ilhéus virus; SLEV: Saint Louis Encephalitis virus; NRLV: Narajal-like virus; OROV: Oropouche virus; MAYV: Mayarovirus; NYFV: non-Yellow fever virus; NC: negative control (RNA from non-infected VERO cells); NA: not applied; H: human; AN: animal; AR: arthropod.

44 M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51

F NS1 ga bases

t .)

2A

aucdYp1m2aigs1gpmawe

2

Yt11tdilcs

ig. 1. Schematic representation of RT-LAMP-PCR degenerate primers targeting thend downstream (blue) arrows indicate the primer position in virus RNA. Colored

o color in this figure legend, the reader is referred to the web version of this article

.5. Reverse Transcription Loop-mediated Isothermalmplification using degenerate primers (RT-dLAMP)

The one step RT-dLAMP approach was used for YFV genomemplification. In this case, a set of degenerate primers (Fig. 1) wassed with the OmniAmpTM RNA & DNA LAMP Kit (http://lucigen.om/OmniAmp-RNA-and-DNA-LAMP-Kit/) in the following con-itions: 5 uL of target RNA, 40 pmol of inner primers (YFFIP andFBIP), 20 pmol of loop primers (YFLF and YFLB), 5 pmol of outerrimers (YFF3 and YFB3), 800 uM of each dNTP, 12 mM of MgSO4,

M of Betaine, 1X DNA Polymerase Buffer, 2X OmniAmp DNA Poly-erase, and DNAse/RNAse free water adjusted to a final volume of

5 uL. Reverse transciption was carried out for 30 min followed bydditional 30 min LAMP step run in isothermal condition (60 ◦C)n a thermoblock heat system (Thermo Scientific). The amplifiedene products were separated by electrophoresis in 2% agarose geltained with 10,000 × SYBR safe® (Invitrogen) diluted 1:10,000 in

× TAE (Tris Acetate EDTA) gel and of SYBR® Green I nucleic acidel stain (10,000X concentrate in DMSO) added to the samples in aroportion of 1:25. Cross contamination among samples was mini-ized using individual disposable tips with barrier for each sample,

nd also by pipetting the samples in a class II-B2 biosafety hood,hich maintains the internal environment in balanced pressure

liminating the aerosol production.

.6. Testing the limit of detection, sensitivity and specificity

To test the limit of detection of the RT-dLAMP assay forFV, a standard sample obtained from VERO cells infected withhe prototype YFV strain H111 was used at a concentration of.9 × 105 PFU/mL. The standard sample was diluted from 10−1 to0−7 and used for establishment of a standard curve to estimatehe number of viruses in the samples and the limit of viral particlesetected. Each dilution was used to infect six different hamsters via

ntraperitoneal route. Blood samples from each hamster were col-ected and tested daily between days 0 and 10. The detection wasonsidered successful in a given day if at least 5 of 6 (83.3%) bloodamples were positive for RT-dLAMP using CT = 38.

ene of YFV from South America, Central/Caribe region and Africa. Upstream (green)indicate polymorphism or base degeneration. (For interpretation of the references

Sensitivity and specificity were determined using the resultsobtained from the test panels. True Positive Rates (TPR) and FalsePositive Rates (FPR) were calculated according to the following for-mulas: TPR = TP/(TP + FN) and FPR = FP/(TN + FP) (or 1-specificity),where TP is the number of true positive results; FN is the number offalse negative results; FP is the number of false-positive results andTN is the number true negative results. All calculations were per-formed using the caret Package (Kuhn, 2008) in the R environment(R Core Team, 2014).

2.7. Statistical comparison of genome amplification assays

For comparison with the RT-dLAMP assay, two other methodswere selected. The methods corresponded to the RT-Hemi-NESTED-PCR and the SYBR green one step real time PCR assayswhich target the E gene of the YFV genome (Nunes et al., 2011) andthe recently described RT-sLAMP for YFV 17D strain (Khunthonget al., 2013). The predictive power RT-sLAMP, RT-dLAMP andqRT-PCR were also evaluated using the Receiver Operating Char-acteristic (ROC) curve analysis (Kumar and Indrayan, 2011; Pepe,2000) with an 95% confidence interval (CI) using pROC (Robin et al.,2011), an open-source R (R Core Team, 2014). The procedureswhere TPR and FPR values are closer to 1 and zero, respectively,represent ideal performance with an AUC = 1 (or 100%) in a ROCgraph.

3. Results

3.1. RT-dLAMP assay and visual analysis

The products of the RT-dLAMP were visualized in both elec-trophoresis gel (EPG) and by naked eye. Positive results by EPG

were interpreted as ladder bands. For naked eye visualization, pos-itive samples were observed as fluorescent reactions in the tubes.Fig. 2 illustrates the positive reactions for the RT-dLAMP PCR assayusing the degenerate YFV NS1 primers.

M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51 45

Fig. 2. Visualization of RT-dLAMP-PCR products for YFV genome detection. (A) Electrophoresis in 2% agarose gel stained with SYBR® safe. MW: molecular weight ladder; slots1–10 correspond to YFV RNAs obtained from wild strains isolated in Bolivia (FVB-0196), Brazil-1 (BeAR378600), Brazil-2 (BeH394880), Colombia (INS 347613), Ecuador (OBS5026), French Guyana (690), Panama (Jimenez GML), Peru (OBS 2240), Trinidad (TRVL 4205) and Venezuela (P128 MC), respectively. Slots 11 and 12 correspond to YFV RNAsobtained from YFV 17D vaccine and BeH 111 strains (positive control), respectively. Slot 13: RNA extracted from non-infected VERO cells (negative control). MW marquersizes are represented in base pairs (bp); (B) visualization using SYBR® green stained samples under UV irradiation. MW: molecular weight marker; slots 1–6 correspondto RNA extracted from strains isolated in Bolivia (FVB-0196), Brazil (BeH394880), Colombia (INS 347613), Ecuador (OBS 5026), Panama (Jimenez GML) and Trinidad (TRVL4 n; slota ) durint

3A

Rs(attma

Yp

RFsf

fAA

in the endemic areas (Africa and South America). Poor vectorcontrol and anthropogenic actions in the forest environment, asso-

205); slot 7: 17D YFV vaccine strain; slot 8: BeH 111 YFV Brazilian prototype strairrows indicate the positive pattern (light green) and negative pattern (light orangehe reader is referred to the web version of this article.)

.2. Determination of true positive and false positive rates andUCs (area under curve)

The positivity rate for RT-PCR, RT-Hemi Nested-PCR, qRT-PCR,T-dLAMP and RT-sLAMP were determined using the panel ofamples (Table 1) The numbers of positives in the YFV grouptrue positive samples) were 40 (61.5%), 61 (93.8%), 65 (100%)nd 65 (100%), respectively, in 65 tested samples for conven-ional RT-PCR, RT-Hemi Nested-PCR, qRT-PCR and RT-dLAMP. Forhe RT-sLAMP assay, 41 of 65 (63%) were positive. Table 2 sum-

arizes the results for genome detection involving all testedssays.

For all tested assays, no false positives were detected in the nonFV group, except for the RT-Hemi-Nested-PCR, where two falseositives were observed (3.07%).

Based on these results the TPR and FPR were estimated as follow:T-PCR (TPR = 0.615, FPR = 0), RT-Hemi-Nested PCR (TPR = 0.938,PR = 0.031); qRT-PCR (TPR = 1, FPR = 0); RT-sLAMP for 17D vaccinetrain (TPR = 0, FPR = 0.631); RT-dLAMP using degenerate primersor YFV (TPR = 1, FPR = 0). Table 3 shows the matrix for all assays.

Comparing the areas under the ROC curves (AUC), the per-ormances of qRT-PCR and RT-dLAMP were equivalent, with an

UC = 100%, and significantly higher then RT-sLAMP, which had anUC of 91.1% (95% CI, 86.1–96.1%, Fig. 4).

9: RNA obtained from non-infected VERO cells (negative control). Double pointedg UV irradiation. (For interpretation of the references to color in this figure legend,

3.3. Limit of detection

The RT-dLAMP showed distinct limits of detection, based on thesample dilution and days after YFV infection. Undiluted sampleswere detectable from day 1 to day 6 with a viral concentration of1.9 × 105 PFU/mL and CTs ranging from 19 (Day 1) to 36 (Day 6).For samples diluted from 10−1 to 10−5 (1.9 × 104 to 1.9 PFU/mL),detections were respectively observed with CTs from 20 (Day 1) to36 (day 4), CT 31 (day 1) to CT 38 (day 2), CT 33 (day 1) to CT 35(day 3), CT 35 (day 1) to CT 38 (day 2) and CT 38 (day 1). Samplesdiluted from 10−6 (0.19 PFU/ml) to 10−7 (0.019 PFU/mL), reactionwere detected in less than 66% of the samples, and only in day 1(CT > 38) (Fig. 3). On day 0, none of the dilutions could be detected.Fig. 3 shows the progression of the chance of detection along thedays.

4. Discussion

Despite an efficient vaccine for control of yellow fever sincethe 1940s, the disease still represents a public health concern

ciated with unplanned, rapid urbanization of cities (Frierson, 2010;Vasconcelos, 2003; Vasconcelos et al., 2001; Walther et al., 2002),

46 M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51

Table 2RNA samples used for comparative analysis of RT-dLAMP-PCR and other molecular methods. YFV: Yellow fever virus; CT: cut-off threshold value (CT = 38).

Groups Virus RNA RT-PCR RT-HemiNested PCR RT-qPCR CT RT-dLAMP PCR CT RT-sLAMP PCR CT Virus isolationResult Result Result Result Result Result

YFVG

YFV 1 pos pos pos 28 pos 25 pos 34 posYFV 2 pos pos pos 25 pos 31 neg 40 posYFV 3 pos pos pos 22 pos 22 neg 39 posYFV 4 neg pos pos 32 pos 22 neg 38 posYFV 5 neg pos pos 31 pos 23 pos 29 posYFV 6 pos pos pos 33 pos 26 pos 34 posYFV 7 neg pos pos 21 pos 34 neg 39 posYFV 8 neg pos pos 21 pos 33 neg 38 posYFV 9 pos pos pos 35 pos 31 neg 38 posYFV 10 neg pos pos 32 pos 32 neg 38 posYFV 11 pos pos pos 29 pos 21 neg 39 posYFV 12 pos pos pos 22 pos 24 neg 40 posYFV 13 neg pos pos 27 pos 22 neg 40 posYFV 14 neg pos pos 22 pos 23 neg 38 posYFV 15 neg pos pos 29 pos 21 neg 39 posYFV 16 pos pos pos 30 pos 22 neg 40 posYFV 17 pos pos pos 31 pos 22 neg 40 posYFV 18 neg pos pos 33 pos 30 neg 40 posYFV 19 pos pos pos 31 pos 34 neg 38 posYFV 20 pos pos pos 28 pos 34 pos 33 posYFV 21 pos pos pos 27 pos 35 pos 34 posYFV 22 pos pos pos 26 pos 31 pos 37 posYFV 23 pos pos pos 31 pos 32 pos 33 posYFV 24 neg pos pos 33 pos 32 pos 34 posYFV 25 neg neg pos 29 pos 33 pos 37 posYFV 26 pos pos pos 27 pos 29 pos 35 posYFV 27 pos pos pos 26 pos 29 pos 37 posYFV 28 pos pos pos 25 pos 23 pos 32 posYFV 29 neg pos pos 22 pos 24 pos 30 posYFV 30 neg pos pos 22 pos 23 pos 31 posYFV 31 neg pos pos 24 pos 21 pos 33 posYFV 32 pos pos pos 31 pos 25 pos 32 posYFV 33 pos pos pos 33 pos 34 neg 38 posYFV 34 pos pos pos 36 pos 33 neg 38 posYFV 35 pos pos pos 33 pos 37 neg 40 posYFV 36 neg pos pos 22 pos 31 neg 38 posYFV 37 neg neg pos 21 pos 21 pos 33 posYFV 38 pos pos pos 20 pos 24 pos 32 posYFV 39 pos pos pos 29 pos 26 pos 32 posYFV 40 neg neg pos 32 pos 25 pos 32 posYFV 41 neg pos pos 22 pos 29 pos 31 posYFV 42 pos pos pos 21 pos 31 pos 33 posYFV 43 pos pos pos 26 pos 24 pos 34 posYFV 44 pos pos pos 27 pos 24 neg 39 posYFV 45 pos pos pos 21 pos 21 neg 38 posYFV 46 neg pos pos 27 pos 22 neg 38 posYFV 47 pos pos pos 22 pos 22 neg 40 posYFV 48 neg pos pos 32 pos 34 pos 37 posYFV 49 pos pos pos 33 pos 35 pos 36 posYFV 50 neg pos pos 28 pos 24 pos 33 posYFV 51 neg pos pos 27 pos 25 pos 32 posYFV 52 pos pos pos 27 pos 26 pos 31 posYFV 53 pos pos pos 26 pos 34 pos 37 posYFV 54 neg neg pos 32 pos 35 pos 33 posYFV 55 neg pos pos 33 pos 34 pos 30 posYFV 56 pos pos pos 33 pos 34 pos 29 posYFV 57 pos pos pos 32 pos 32 pos 36 posYFV 58 pos pos pos 26 pos 21 pos 34 posYFV 59 pos pos pos 24 pos 29 pos 29 posYFV 60 pos pos pos 24 pos 22 pos 28 posYFV 61 pos pos pos 23 pos 21 pos 32 posYFV 62 neg pos pos 21 pos 22 pos 32 posYFV 63 pos pos pos 27 pos 31 pos 37 posYFV 64 pos pos pos 30 pos 28 pos 33 posYFV 65 pos pos pos 31 pos 30 pos 34 pos

M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51 47

Table 2 (Continued)

Groups Virus RNA RT-PCR RT-HemiNested PCR RT-qPCR CT RT-dLAMP PCR CT RT-sLAMP PCR CT Virus isolationResult Result Result Result Result Result

NYFVG

NYFV1 neg neg neg 40 neg 39 neg 40 posNYFV2 neg neg neg 40 neg 40 neg 40 posNYFV3 neg neg neg 40 neg 40 neg 39 posNYFV4 neg pos neg 39 neg 38 neg 39 posNYFV5 neg neg neg 40 neg 40 neg 39 posNYFV6 neg neg neg 40 neg 40 neg 40 posNYFV7 neg neg neg 40 neg 40 neg 40 posNYFV8 neg neg neg 39 neg 40 neg 40 posNYFV9 neg neg neg 38 neg 40 neg 39 posNYFV10 neg neg neg 40 neg 40 neg 40 posNYFV11 neg neg neg 40 neg 40 neg 40 posNYFV12 neg neg neg 40 neg 38 neg 40 posNYFV13 neg neg neg 40 neg 40 neg 40 posNYFV14 neg neg neg 40 neg 40 neg 40 posNYFV15 neg neg neg 40 neg 40 neg 39 posNYFV16 neg neg neg 40 neg 40 neg 40 posNYFV17 neg neg neg 40 neg 40 neg 40 posNYFV18 neg neg neg 38 neg 40 neg 40 posNYFV19 neg neg neg 40 neg 40 neg 40 posNYFV20 neg neg neg 40 neg 40 neg 40 posNYFV21 neg neg neg 40 neg 40 neg 40 posNYFV22 neg neg neg 40 neg 39 neg 40 posNYFV23 neg neg neg 40 neg 40 neg 40 posNYFV24 neg pos neg 40 neg 40 neg 40 posNYFV25 neg neg neg 40 neg 40 neg 38 posNYFV26 neg neg neg 40 neg 40 neg 40 posNYFV27 neg neg neg 40 neg 40 neg 38 posNYFV28 neg neg neg 40 neg 40 neg 40 posNYFV29 neg neg neg 40 neg 40 neg 40 posNYFV30 neg neg neg 40 neg 40 neg 38 posNYFV31 neg neg neg 40 neg 40 neg 40 posNYFV32 neg neg neg 40 neg 40 neg 40 posNYFV33 neg neg neg 39 neg 40 neg 40 posNYFV34 neg neg neg 40 neg 39 neg 40 posNYFV35 neg neg neg 39 neg 40 neg 40 posNYFV36 neg neg neg 39 neg 40 neg 40 posNYFV37 neg neg neg 40 neg 40 neg 40 posNYFV38 neg neg neg 40 neg 40 neg 40 posNYFV39 neg neg neg 40 neg 40 neg 40 posNYFV40 neg neg neg 38 neg 38 neg 40 posNYFV41 neg neg neg 39 neg 38 neg 39 posNYFV42 neg neg neg 40 neg 40 neg 39 posNYFV43 neg neg neg 40 neg 40 neg 40 posNYFV44 neg neg neg 40 neg 40 neg 40 posNYFV45 neg neg neg 40 neg 40 neg 40 posNYFV46 neg neg neg 40 neg 40 neg 40 posNYFV47 neg neg neg 40 neg 40 neg 40 posNYFV48 neg neg neg 40 neg 40 neg 40 posNYFV49 neg neg neg 40 neg 40 neg 38 posNYFV50 neg neg neg 40 neg 39 neg 40 posNYFV51 neg neg neg 40 neg 40 neg 40 posNYFV52 neg neg neg 39 neg 40 neg 40 posNYFV53 neg neg neg 40 neg 40 neg 40 posNYFV54 neg neg neg 40 neg 40 neg 40 posNYFV55 neg neg neg 40 neg 40 neg 39 posNYFV56 neg neg neg 40 neg 40 neg 40 posNYFV57 neg neg neg 40 neg 40 neg 40 posNYFV58 neg neg neg 40 neg 40 neg 40 posNYFV59 neg neg neg 39 neg 38 neg 40 posNYFV60 neg neg neg 38 neg 40 neg 40 posNYFV61 neg neg neg 40 neg 40 neg 40 posNYFV62 neg neg neg 40 neg 40 neg 40 posNYFV63 neg neg neg 40 neg 40 neg 40 posNYFV64 neg neg neg 40 neg 38 neg 39 pos

40

Y ; pos:

aei22h

NYFV65 neg neg neg

FVG: Yellow fever virus group; NYFV: non-Yellow fever virus group; neg: negative

nd human movements (Woolhouse et al., 2012) from yellow feverndemic to non-endemic urban areas have contributed to thencreased risk of yellow fever urbanization (Gardner and Ryman,

010; Monath, 2001; Vasconcelos, 2003; Monath and Vasconcelos,015). In this context, new approaches for detection of yellow feverave been developed in recent years (Bae et al., 2003; Dash et al.,

neg 40 neg 40 pos

positive; CT: cut-off threshold value (CT = 38).

2012; Mantel et al., 2008; Weidmann et al., 2010; Nunes et al.,2011). However, all of them requires the need for a thermal cycleror real time device, which increases the cost and limits the use of the

test to laboratories with good facilities, equipment, and funding.

More recently, Weidmann et al. (2010), Domingo et al. (2012)and Escadafal et al. (2014) have developed specific assays for

48 M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51

Table 3Test sensitivity and specificity for RT-PCR, RT-Nested PCR, RT-qPCR, RT-dLAMP-PCR and RT-sLAMP-PCR assays from 130 RNA samples used in this study, with cutoff CT = 38.

RT-PCR RT-Nested PCR RT-qPCR RT-dLAMP-PCR RT-sLAMP-PCR

True positives 40 61 65 65 41False negatives 25 4 0 0 24True negatives 65 63 65 65 65False positives 0 2 0 0 0Specificity 1 0.97 1 1 1

doKfoprcs

Fsd

Sensitivity 0.61 0.94

etection of YFV genomes. Although robust and easy, the meth-ds still need the use of labeled probes or specific equipment.wallah and co-workers (2013) developed an isothermal approach

or YFV genome detection. This was an important improvementn field assay since the LAMP methodology does not require com-lex devices (Mori and Notomi, 2009) such as thermo cyclers or

eal time PCR machines which needs mains power supply. In thisase, the LAMP method requires a simple thermo block heatingupplied by low voltage battery easily and economically obtained

ig. 3. Detection of YFV genome in hamster serum samples using the RT-dLAMP PCR. (A)ample dilution; (B) chance in percentage of YFV genome detection in hamster serum samilution. CT: cut-off threshold value (CT = 38).

1 1 0.63

in general hardware shops. The developed assay for YFV genomedetection described by Kwallah et al. used a set of primers specificfor the NS1 gene of the YFV 17D vaccine strain which was derivedfrom the wild YFV ASIBI strain isolated in Ghana in 1927 (Frierson,2010). This approach succeeded in detecting the YFV genomesof strains isolated in Africa including the 17D strain. However

YFV strains from other countries were not tested, and there-fore, the capacity to detect New World YFV strains remains to beinvestigated.

Evaluation of limit of detection according to the day post-infection, CT values andples using the RT-dLAMP-PCR assay according to days post-infection and sample

M.R.T. Nunes et al. / Journal of Virological Methods 226 (2015) 40–51 49

Fig. 4. ROC curves computed for RT-qPCR, RT-dLAMP-PCR (blue) and RT-sLAMP-PCR assays (red). Confidence interval for RT-sLAMP-PCR ROC curve is shown in salmon color.C nces t

2AeoIrVts(tudC

tfssdOtdfh

omputed AUCs values are presented as percentage. (For interpretation of the refere

Partial sequences of YFV prM, E, and NS5 genes (Bryant et al.,007), as well as the complete genome sequences of Brazilian,frican and Trinidadian YFV strains (Nunes et al., 2012; Pisanot al., 1999; Stock et al., 2013) have demonstrated a high degreef genetic variation among YFV isolated in Africa and the Americas.ndeed, five and two distinct genetic lineages have been describedespectively in these YFV endemic regions (Bryant et al., 2007;asconcelos et al., 2004). Due to this genetic variability, we decided

o redesign a set of primers in order to increase the sensitivity andpecificity to those used by Kwallah et al. (2013) and other authorsDomingo et al., 2012; Escadafal et al., 2014) including degenera-ions in the consensus sequence (Fig. 1). These degenerations weresed in order to target all YFV strains available in the Genbankatabase which included isolates from African, South American,entral American and the Caribbean regions (Table 1).

The analysis of the RT-dLAMP products demonstrated that theest succeeded in detecting a wide range of YFV strains obtainedrom new world and one African derived strain (17DD vaccinetrain) which were distinctly visualized in both 2% electrophore-is gel (Fig. 2(a)) and during UV irradiation (Fig. 2(b)), where a veryistinct pattern for positive and negative reactions was observed.btaining African strains was a limitation for the study, however

he positivity for the 17DD vaccine strain suggests that the RT-LAMP is also able to detect other YFV African strains. In this case,urther evaluation using YFV strains isolated in Africa should beelpful to support this hypothesis.

o color in this figure legend, the reader is referred to the web version of this article.)

The usefulness of a diagnostic test is generally assessed by cal-culating the sensitivity (TPR) and specificity (1-FPR), or the positivepredictive value and negative predictive value of the test (DeLonget al., 1985). We have assessed the sensitivity and specificity of fivemolecular methods and found 100% positivity for the RT-dLAMP,indicating that the test was able to detect all true positive sam-ples (Table 1) included in our assay (Table 2). In comparison withother tests (conventional RT-PCR, RT-Hemi-Nested-PCR, qRT-PCRand RT-sLAMP), all tests were specific. However, the RT-dLAMPsensitivity was 100% compared to the RT-PCR (61.5%), RT-sLAMP(63.1%), and RT-Hemi-Nested-PCR (93.8%) (Table 2).

In the case of the conventional RT-PCR, the low sensitivity(61.5%) may be explained by the use of primers that amplify a1 kb long fragment. Efficiency on amplification of large genomicfragments could be affected by storage conditions and enzymaticdegradation of RNA or DNA, in particular target sequence (Nuneset al., 2011). In the case of the RT-sLAMP, analysis of the targetprimer binding sites at the NS1 gene proposed by Kwallah andcolleagues (2013) demonstrated that, especially in the F3 outerprimer, F2 Forward inner primer and B3 reverse outer primer bindsites, three to five degeneration were found and could be responsi-ble for the lower test sensitivity (Fig. 1). The efficiency with African

strains is explained by the primer specificity for the YFV 17D vac-cine strains that was derived from the African wild type Asibi strain.

The RT-dLAMP using degenerate primers were able to amplifyall tested YFV strains isolated in distinct geographic areas and from

5 irolog

daNewpa8dohtoid

bpoobYc

A

PoiYbt4(

R

B

B

C

D

D

D

D

E

F

FG

0 M.R.T. Nunes et al. / Journal of V

istinct hosts (humans, mosquitoes and monkeys; Tables 1 and 2),nd had equivalent sensitivity in comparison to the RT-Hemi-ested-PCR and qRT-PCR (Nunes et al., 2011). In addition, thevaluation of the assay using hamster serum samples infectedith the YFV Brazilian prototype strain indicates that degeneraterimers succeeded in detection of YFV genome in low viral titersnd a higher percentage of detection ranging between 100% and3% from day 1 to day 4 (Fig. 3(a)). More specifically, the limit ofetection for the assay was 19 PFU/mL which is equivalent to thosebserved for other methods (Nunes et al., 2011). The ROC methodas been used to evaluate and indicate the performance of a givenest in comparison to other (Kumar and Indrayan, 2011). In casef the RT-dLAMP, the ROC analysis (Fig. 4) clearly indicates thatt is comparable in performance with RT-qPCR – widely used foretection of RNA viruses.

The current assay does not use complex equipment, and cane easily handled by technicians with basic knowledge in sam-le handling. The absence of thermocyclers and eye visualizationf test results are possible factors that contribute to the low-costf the test, facilitating wider range of use in laboratories withasic facilities. Thus, the RT-dLAMP could be a useful approach forFV detection during ecologic/epidemiologic investigations in fieldonditions.

cknowledgements

We are grateful to Drs. Sueli Guerreiro Rodrigues and Elianainto Vieira from the Department of Arbovirology and Hem-rrhagic Fevers at the Evandro Chagas Institute for the virussolation procedures and antigenic identification of the Brazilianellow Fever virus strains. This study was partially supportedy CNPq (Conselho Nacional para o Desenvolvimento Cien-ífico e Tecnológico) projects 401558/2013-4; 457664/2013-4,86069/2012-5 and 301641/2010-2), and CNPq/CAPES/FAPESPAproject 573739-2008-0).

eferences

ae, H.-G., Nitsche, A., Teichmann, A., Biel, S.S., Niedrig, M., 2003. Detection ofyellow fever virus: a comparison of quantitative real-time PCR and plaqueassay. J. Virol. Methods 110, 185–191, http://dx.doi.org/10.1016/S0166-0934(03)00129-0.

ryant, J.E., Holmes, E.C., Barrett, A.D.T., 2007. Out of Africa: a molecularperspective on the introduction of yellow fever virus into the Americas. PLoSPathog. 3, e75, http://dx.doi.org/10.1371/journal.ppat.0030075.

ardoso, J.C., Almeida, M.A.B., Santos, E., Fonseca, D.F., Sallum, M.A.M., Noll, C.A.,Monteiro, H.A.O., Cruz, A.C.R., Carvalho, V.L., Pinto, E.V.P., Vasconcelos, P.F.C.,2010. Yellow fever virus isolation from Haemagogus leucocelaenus and Aedesserratus mosquitoes, southern Brazil, 2008. Emerg. Infect. Dis. 16, 1918–1924.

ash, P., Boutonnier, A., Prina, E., Sharma, S., Reiter, P., 2012. Development of aSYBR green I based RT-PCR assay for yellow fever virus: application inassessment of YFV infection in Aedes aegypti. Virol. J., http://dx.doi.org/10.1186/1743-422X-9-27.

eLong, E.R., Vernon, W.B., Bollinger, R.R., 1985. Sensitivity and specificity of amonitoring test. Biometrics 41, 947–958.

ietzgen, R.G., Calisher, C.H., Kurath, G., Kuzman, I.V., Rodriguez, L.L., Stone, D.M.,Tesh, R.B., Tordo, N., Walker, P.J., Wetzel, T., Whitfield, A.E., 2012. VirusTaxonomy. In: King, A.M.Q., Adams, M.J., Carstens, E.B., Lefkowitz, E.J. (Eds.),Ninth Report of the International Committe on Taxonomy of Viruses. Elsevier,San Diego, pp. 654–681.

omingo, C., Patel, P., Yillah, J., Weidmann, M., Méndez, J.A., Nakouné, E.R., Niedrig,M., 2012. Advanced yellow fever virus genome detection in point-of-carefacilities and reference laboratories. J. Clin. Microbiol. 50, 4054–4060, http://dx.doi.org/10.1128/JCM.01799-12.

scadafal, C., Faye, O., Sall, A.A., Faye, O., Weidmann, M., Strohmeier, O., vonStetten, F., Drexler, J., Eberhard, M., Niedrig, M., Patel, P., 2014. Rapid molecularassays for the detection of yellow fever virus in low-resource settings. PLoSNegl. Trop. Dis. 8, http://dx.doi.org/10.1371/journal.pntd.0002730.

erguson, M., Shin, J., Knezevic, I., Minor, P., Barrett, A., 2010. WHO Working Group

on Technical Specifications for Manufacture and Evaluation of Yellow FeverVaccines, Geneva, Switzerland, 13–14 May 2009. Vaccine.

rierson, J.G., 2010. The yellow fever vaccine: a history. Yale J. Biol. Med. 83, 77–85.ardner, C.L., Ryman, K.D., 2010. Yellow fever: a reemerging threat. Clin. Lab. Med.,

http://dx.doi.org/10.1016/j.cll.2010.01.001.

ical Methods 226 (2015) 40–51

Jentes, E.S., Poumerol, G., Gershman, M.D., Hill, D.R., Lemarchand, J., Lewis, R.F.,Staples, J.E., Tomori, O., Wilder-Smith, A., Monath, T.P., 2011. The revised globalyellow fever risk map and recommendations for vaccination, 2010: consensusof the informal WHO working group on geographic risk for yellow fever.Lancet Infect. Dis., http://dx.doi.org/10.1016/S1473-3099(11)70147-5.

Kalvatchev, Z., Tsekov, I., Kalvatchev, N., 2010. Loop-mediated amplification forsensitive and specific detection of viruses. Biotechnol. Biotechnol. Equip.,http://dx.doi.org/10.2478/V10133-010-0004-8.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton,S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P.,Drummond, A., 2012. Geneious Basic: an integrated and extendable desktopsoftware platform for the organization and analysis of sequence data.Bioinformatics 28, 1647–1649, http://dx.doi.org/10.1093/bioinformatics/bts199.

Khunthong, S., Jaroenram, W., Arunrut, N., Suebsing, R., Mungsantisuk, I.,Kiatpathomchai, W., 2013. Rapid and sensitive detection of shrimp yellow headvirus by loop-mediated isothermal amplification combined with a lateral flowdipstick. J. Virol. Methods, http://dx.doi.org/10.1016/j.jviromet.2012.11.041.

Kuhn, Max, 2008. Building predictive models in R using the caret package. J. Stat.Softw., http://dx.doi.org/10.18637/jss.v028.i05.

Kumar, R., Indrayan, A., 2011. Receiver operating characteristic (ROC) curve formedical researchers. Indian Pediatr., http://dx.doi.org/10.1007/s13312-011-0055-4.

Kwallah, A. ole, Inoue, S., Muigai, A.W.T., Kubo, T., Sang, R., Morita, K., Mwau, M.,2013. A real-time reverse transcription loop-mediated isothermalamplification assay for the rapid detection of yellow fever virus. J. Virol.Methods 193, 23–27, http://dx.doi.org/10.1016/j.jviromet.2013.05.004.

Lu, X., Li, X., Mo, Z., Jin, F., Wang, B., Zhao, H., Shan, X., Shi, L., 2012. Rapididentification of chikungunya and dengue virus by a real-time reversetranscription-loop-mediated isothermal amplification method. Am. J. Trop.Med. Hyg. 87, 947–953, http://dx.doi.org/10.4269/ajtmh.2012.11-0721.

Mantel, N., Aguirre, M., Gulia, S., Girerd-Chambaz, Y., Colombani, S., Moste, C.,Barban, V., 2008. Standardized quantitative RT-PCR assays for quantitation ofyellow fever and chimeric yellow fever-dengue vaccines. J. Virol. Methods 151,40–46, http://dx.doi.org/10.1016/j.jviromet.2008.03.026.

Monath, T.P., 2001. Reviews yellow fever: an update. Lancet 1, 11–20, http://dx.doi.org/10.1016/S1473-3099(01)00016-0.

Monath, T.P., Vasconcelos, P.F.C., 2015. Yellow fever. J. Clin. Virol. 64,160–173.

Mori, Y., Notomi, T., 2009. Loop-mediated isothermal amplification (LAMP): arapid, accurate, and cost-effective diagnostic method for infectious diseases. J.Infect. Chemother., http://dx.doi.org/10.1007/s10156-009-0669-9.

Norrby, E., 2007. Yellow fever and Max Theiler: the only Nobel Prize for a virusvaccine. J. Exp. Med. 204, 2779–2784, http://dx.doi.org/10.1084/jem.20072290.

Notomi, T., Okayama, H., Masubuchi, H., Yonekawa, T., Watanabe, K., Amino, N.,Hase, T., 2000. Loop-mediated isothermal amplification of DNA. Nucleic AcidsRes. 28, E63, http://dx.doi.org/10.1093/nar/28.12.e63.

Nunes, M.R.T., Palacios, G., Cardoso, J.F., Martins, L.C., Sousa, E.C., De Lima, C.P.S.,Medeiros, D.B.A., Savji, N., Desai, A., Rodrigues, S.G., Carvalho, V.L., Lipkin, W.I.,Vasconcelos, P.F.C., 2012. Genomic and phylogenetic characterization ofbrazilian yellow fever virus strains. J. Virol. 86, 13263–13271, http://dx.doi.org/10.1128/JVI.00565-12.

Nunes, M.R.T., Palacios, G., Nunes, K.N.B., Casseb, S.M.M., Martins, L.C., Quaresma,J.A.S., Savji, N., Lipkin, W.I., Vasconcelos, P.F.C., 2011. Evaluation of twomolecular methods for the detection of yellow fever virus genome. J. Virol.Methods 174, 29–34, http://dx.doi.org/10.1016/j.jviromet.2011.02.025.

Parida, M., Posadas, G., Inoue, S., Hasebe, F., Morita, K., 2004. Real-time reversetranscription loop-mediated isothermal amplification for rapid detection ofWest Nile virus. J. Clin. Microbiol. 42, 257–263, http://dx.doi.org/10.1128/JCM.42.1.257-263.2004.

Pepe, M.S., 2000. Receiver operating characteristic methodology. J. Am. Stat. Assoc.95, 308–311, http://dx.doi.org/10.2307/2669554.

Pisano, M.R., Mercier, V., Deubel, V., Tolou, H., 1999. Complete nucleotide sequenceand phylogeny of an American strain of yellow fever virus, TRINID79A. Arch.Virol. 144, 1837–1843, http://dx.doi.org/10.1007/s007050050708.

Poon, L.L.M., Leung, C.S.W., Chan, K.H., Lee, J.H.C., Yuen, K.Y., Guan, Y., Peiris, J.S.M.,2005. Detection of human influenza A viruses by loop-mediated isothermalamplification. J. Clin. Microbiol. 43, 427–430, http://dx.doi.org/10.1128/JCM.43.1.427-430.2005.

Qiagen, 2003. RNA Purification: Qiagen RNeasy Micro Handbook. Prod. Lit. Qiagen.,pp. 1–76.

R Core Team, 2014. R: A Language and Environment for Statistical Computing. RFoundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/.

Rio, D.C., Ares, M., Hannon, G.J., Nilsen, T.W., 2010. Purification of RNA using TRIzol(TRI reagent). Cold Spring Harb. Protoc. 5, http://dx.doi.org/10.1101/pdb.prot5439.

Robin, X., Turck, N., Hainard, A., Tiberti, N., Lisacek, F., Sanchez, J.-C., Müller, M.,2011. pROC: an open-source package for R and S+ to analyze and compare ROCcurves. BMC Bioinform. 12, 77, http://dx.doi.org/10.1186/1471-2105-12-77.

Stock, N.K., Laraway, H., Faye, O., Diallo, M., Niedrig, M., Sall, A.a., 2013. Biological

and phylogenetic characteristics of yellow fever virus lineages from WestAfrica. J. Virol. 87, 2895–2907, http://dx.doi.org/10.1128/JVI.01116-12.

Toriniwa, H., Komiya, T., 2006. Rapid detection and quantification of Japaneseencephalitis virus by real-time reverse transcription loop-mediated isothermalamplification. Microbiol. Immunol. 50, 379–387, pii:JST.JSTAGE/mandi/50.379.

irolog

V

V

V

V

M.R.T. Nunes et al. / Journal of V

an Der Stuyft, P., Gianella, A., Pirard, M., Cespedes, J., Lora, J., Peredo, C., Pelegrino,J.L., Vorndam, V., Boelaert, M., 1999. Urbanisation of yellow fever in Santa Cruz,Bolivia. Lancet 353, 1558–1562, http://dx.doi.org/10.1016/S0140-6736(99)03291-2.

asconcelos, P.F., 2003. Yellow fever. Rev. Soc. Bras. Med. Trop. 36, 275–293,pii:S0037-86822003000200012.

asconcelos, P.F.C., Bryant, J.E., Travassos da Rosa, A.P.A., Tesh, R.B., Rodrigues, S.G.,

Barrett, A.D.T., 2004. Genetic divergence and dispersal of yellow fever virus,Brazil. Emerg. Infect. Dis. 10 (9), 1578–1584.

asconcelos, P.F., Travassos da Rosa, A.P., Rodrigues, S.G., Travassos da Rosa, E.S.,Dégallier, N., Travassos da Rosa, J.F., 2001. Inadequate management of naturalecosystem in the Brazilian Amazon region results in the emergence and

ical Methods 226 (2015) 40–51 51

reemergence of arboviruses. Cad. Saude Publica 17 (Suppl.), 155–164, http://dx.doi.org/10.1590/S0102-311X2001000700025.

Walther, G.-R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T.J.C.,Fromentin, J.-M., Hoegh-Guldberg, O., Bairlein, F., 2002. Ecological responses torecent climate change. Nature 416, 389–395.

Weidmann, M., Faye, O., Faye, O., Kranaster, R., Marx, A., Nunes, M.R.T.,Vasconcelos, P.F.C., Hufert, F.T., Sall, A.A., 2010. Improved LNA probe-based

assay for the detection of African and South American yellow fever virusstrains. J. Clin. Virol. 48, 187–192, http://dx.doi.org/10.1016/j.jcv.2010.04.013.

Woolhouse, M., Scott, F., Hudson, Z., Howey, R., Chase-Topping, M., 2012. Humanviruses: discovery and emergence. Philos. Trans. R. Soc. B Biol. Sci., http://dx.doi.org/10.1098/rstb.2011.0354.