Diagnostic Microbiology and Infectious Disease 77 (2013) 133–137

Contents lists available at ScienceDirect

Diagnostic Microbiology and Infectious Disease

j ourna l homepage: www.e lsev ie r .com/ locate /d iagmicrob io

Real-time polymerase chain reaction for diagnosis and quantitation of negativestrand of chikungunya virus☆

Chun Wei Chiam a, Yoke Fun Chan a, Shih Keng Loong a, Sara Su Jin Yong a, Poh Sim Hooi b, I-Ching Sam a,b,⁎a Department of Medical Microbiology, Faculty of Medicine, University of Malaya, 50603, Kuala Lumpur, Malaysiab Diagnostic Virology Laboratory, University Malaya Medical Centre, Lembah Pantai, 59100, Kuala Lumpur, Malaysia

a b s t r a c ta r t i c l e i n f o

☆ Preliminary findings of this study were presented asCongress of Medical Virology, held in Adelaide, Australi⁎ Corresponding author. Tel.: +60-3-79492184; fax:

E-mail address: jicsam@ummc.edu.my (I.-C. Sam).

0732-8893/$ – see front matter © 2013 Elsevier Inc. Alhttp://dx.doi.org/10.1016/j.diagmicrobio.2013.06.018

Article history:Received 1 February 2013Received in revised form 10 June 2013Accepted 13 June 2013Available online 23 July 2013

Keywords:Chikungunya virusMolecular diagnosisReal-time PCRViral loadArthralgia

Quantitative real-time polymerase chain reaction (qRT-PCR) is useful for diagnosis and studying virusreplication. We developed positive- and negative-strand qRT-PCR assays to detect nsP3 of chikungunya virus(CHIKV), a positive-strand RNA alphavirus that causes epidemic fever, rash, and arthritis. The positive- andnegative-strand qRT-PCR assays had limits of quantification of 1 and 3 log10 RNA copies/reaction, respectively.Compared to a published E1 diagnostic assay using 30 laboratory-confirmed clinical samples, the positive-strand nsP3 qRT-PCR assay had higher R2 and efficiency and detectedmore positive samples. Peak viral load of12.9 log10 RNA copies/mL was reached on day 2 of illness, and RNA was detectable up to day 9, even in thepresence of anti-CHIKV IgM. There was no correlation between viral load and persistent arthralgia. Thepositive-strand nsP3 assay is suitable for diagnosis, while the negative-strand nsP3 assay, which uses taggedprimers to increase specificity, is useful for study of active viral replication kinetics.

a poster at the 9th Asian Pacifica, on June 6–8, 2012.+60-3-79675752.

l rights reserved.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Chikungunya virus (CHIKV) is a mosquito-borne alphavirus fromthe Togaviridae family that causes epidemic fever, rash, and arthritis.It is an enveloped, single-stranded, linear, positive-sense RNA viruswith a genome size of approximately 11.8 kb, and 2 open readingframes encoding the nonstructural (nsP1-nsP2-nsP3-nsP4) andstructural polyproteins (C-E3-E2-6K-E1) (Powers and Logue, 2007).In 2005, CHIKV caused major outbreaks spreading from East Africa tothe Indian Ocean (Schuffenecker et al., 2006) and India (Arankalleet al., 2007), and then to Europe (Rezza et al., 2007) and Asia (Ng et al.,2009; Sam et al., 2009), affecting millions.

Conventional methods of laboratory diagnosis, such as virus cul-ture and serology, are time-consuming, non-specific, or less sen-sitive. The advent of conventional reverse transcriptase polymerasechain reaction (PCR) improved rapidity, specificity, and sensitivity(Hasebe et al., 2002). This was further improved with quantitativereal-time PCR (qRT-PCR) technology, which is faster, highlysensitive, specific and reproducible, and also provides potentiallyinformative viral loads (Mackay et al., 2002). Several CHIKV diagnosticqRT-PCR assays using different gene targets such as E1, nsP1, and nsP2have been described (Carletti et al., 2007; Edwards et al., 2007;Laurent et al., 2007; Panning et al., 2009; Pastorino et al., 2005; Paridaet al., 2007).

The replication of CHIKV involves the formation of a negative-strand RNA intermediate, which is then used as a template for pro-duction of positive-strand RNA that will be incorporated in maturevirions (Jose et al., 2009). Quantification of negative-strand RNAduring the course of infection is useful in assessing the kinetics ofactive viral replication (Bannister et al., 2010; Richardson et al., 2006;Shirako and Strauss, 1994).

The nsP3 protein is involved in negative-strand and subgenomicRNA synthesis (Jose et al., 2009). Hence, nsP3would be a suitable genetarget to quantify the virus load during replication.

Using the nsP3 gene of CHIKV as the target, the objective of thisstudy was to design 2 qRT-PCR assays, firstly, to detect the positive-strand RNA for diagnosis, and the secondly, to measure negative-strand RNA as a research tool for determining active replication ofCHIKV. The positive-strand qRT-PCR was compared to a publisheddiagnostic E1 assay (Edwards et al., 2007) by testing on clinical sam-ples from Kuala Lumpur, Malaysia. Both assays were then used todetermine in vitro replication of CHIKV in cell culture.

2. Materials and methods

2.1. Patients' samples

This study used 30 acute-phase serum samples collected in 2008–2009 from patients in Kuala Lumpur, Malaysia, confirmed to haveCHIKV by at least 1 of virus culture, conventional PCR (Hasebe et al.,2002), IgM detection by immunofluorescence (Lam et al., 2001), orIgM/IgG seroconversion in a subsequent sample. The samples werecollected between 1 and 9 days after onset of illness. The study also

134 C.W. Chiam et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 133–137

included serum samples from 20 patients not suspected of CHIKVinfection, which were negative by conventional PCR and negative forculture in Vero cells. Sensitivity, specificity, positive predictive value(PPV), and negative predictive value (NPV) were calculated.

2.2. Virus isolates

The CHIKV strain used was MY/08/065 (GenBank accessionnumber FN295485). Dengue virus (DENV) strain New Guinea C(ATCC VR-1584) and Sindbis virus (SINV) strain Ar-339 (ATCC VR-1248) were used to test specificity, as they are viruses present inMalaysia, which cause similar clinical illness to CHIKV. Isolates werepropagated in Vero cells (African greenmonkey kidney, ATCC CCL-81).

2.3. Infection of Vero cells

Vero cells were seeded in 24-well plates at 1 × 105 cells/well with500-μL growth media. After overnight incubation, cells were infectedwith CHIKV at multiplicity of infection of 0.1 and rocked at roomtemperature for 1 hour. Virus inoculums were removed, cells rinsedtwice, and medium supplemented with 2% fetal bovine serum wasadded before incubation at 37 °C with 5% CO2. Supernatant wascollected at selected time-points for CHIKV positive-strand RNAquantification and virus titration. Cell lysates were collected for thenegative-strand assay. All the collected samples were stored at−80 °C.

2.4. Virus titration assay

Vero cells were seeded in 96-well plates at 1 × 104 cells/wellwith 100-μL growth media. After overnight incubation, cells wereinfected with CHIKV supernatant and incubated as above. Plates wereread on day 7 to determine the median tissue culture infective dose/mL values, using the Reed and Muench formula.

2.5. Viral RNA and total RNA extraction

For the positive-strand RNA assay, CHIKV RNA was extracted frompatients' serumwith the QIAamp Viral Mini kit (QIAGEN, Valencia, CA,USA). CHIKV, DENV, and SINV RNA were extracted from infected Verocell supernatants. For the negative-strand RNA assay, total CHIKV RNAwas extracted from Vero cell lysates using the RNeasy Mini Kit(QIAGEN). Extracted RNA was eluted in 40 μL of Ambion RNA StorageSolution (Life Technologies, Carlsbad, CA, USA) and stored at −80°C.

2.6. Primer design

For the positive-strand qRT-PCR assay, nsP3 primers (136 bp)were designed (Table 1). The E1 primers (127 bp) were as describedby Edwards et al. (2007). For the negative-strand nsP3 assay, a taggedprimer system was used, as this has clearly been shown to improvespecificity of negative qRT-PCR assays (Komurian-Pradel et al., 2004;Plaskon et al., 2009). During the reverse transcription step of viralRNA, false priming may occur even in the absence of specific primer,

Table 1Primers used in this study.

Primer Sequence (5′-3′)

nsP3-F GCGCGTAAGTCCAAGGGAATnsP3-R AGCATCCAGGTCTGACGGGCHIK E1-F TCGACGCGCCCTCTTTAACHIK E1-R ATCGAATGCACCGCACACTTag CCTCCGCGGCCGTCATGGTGGCGATag nsP3-F CCTCCGCGGCCGTCATGGTGGCGAGCGCGTAAGTCCAAGGGAATT7 TAATACGACTCACTATAGGG

a Numbering based on CHIKV prototype S27 sequence (GenBank accession number AF36

leading to generation of both strands of RNA. The presence ofcompeting falsely-primed complementary DNA (cDNA) can affectaccurate quantification of negative RNA (Plaskon et al., 2009). Tominimise this, a tagged forward primer (in this study, Tag nsP3-F) wasused during reverse transcription, which generates cDNA with anadded tag sequence unrelated to the target RNA. During the quan-titative PCR step, use of a tag-specific primer (Tag) allows accuratequantification of the correct cDNAs and not the falsely-primed cDNAslacking the complementary tag sequence. Primers were designedbased on alignments of publicly-available CHIKV sequences, withPrimer Express version 3.0 (Life Technologies). The primers were alsoaligned with available full-genome sequences of the related alpha-viruses Ross River virus, O'nyong-nyong virus, and Semliki Forestvirus to ensure minimal matching. All primers were synthesised andpurified with high-performance liquid chromatography.

2.7. Generation of cDNA

The nsP3-R and CHIK E1-R primers were used to generate nsP3 andE1 positive-strand cDNA, respectively. Negative-strand cDNA wasgenerated with Tag nsP3-F primer. The mixture of 500 nmol/L ofprimers, 50 nmol/L of dNTPmix (Promega, Madison,WI, USA), and 1 μLof RNA (equivalent to 3.5 μL of patient serum) was incubated at 65 °Cfor 5 min and placed on ice for 4 min. The cDNA was synthesised with200 U of Superscript III Reverse Transcriptase (Life Technologies),0.1 mol/L DTT (Life Technologies), 40 U of RNaseOUT (Life Technolo-gies), and 1× First Strand buffer (Life Technologies) at 50 °C for 60min.RT enzyme was inactivated at 70 °C for 15 min, then unincorporatedprimers were digested with 20 U of Exonuclease I (New EnglandBiolabs, Ipswich, MA, USA). The cDNA were stored at −80 °C.

2.8. Generation of in vitro RNA transcripts for standard curve

The nsP3 and E1 DNA inserts plus T7 promoter sequence wereamplified, gel purified with GeneAll Expin GEL SV (GeneAll, Seoul,Korea), then in vitro transcribed using MEGAshortscript Kit (LifeTechnologies), and purified with MEGAclear Kit (Life Technologies).The in vitro RNA transcripts were eluted in 100 μL of elution bufferpreheated to 95 °C. Concentrations of in vitro RNA transcripts weredetermined with a nanospectrophotometer (Implen, Munich, Ger-many). For standards curves, cDNA was synthesised with in vitro RNAtranscripts as templates, quantified, serially diluted 10-fold, thenstored at −80 °C.

2.9. Conventional PCR and qRT-PCR

Conventional PCR was performed using GoTaq Flexi DNA poly-merase (1× GoTaq Flexi Buffer, 1 mmol/L MgCl2, 0.2 mmol/L eachdNTP, 1.25 U GoTaq DNA Polymerase), 0.2 μmol/L of each primer(Table 1), 2 μL cDNA, and dH2O to a final volume of 50 μL. The PCRprogram was 95 °C for 4 min, 30 cycles of 95 °C for 1 min, 55 °C for 1min and 72 °C for 1 min, and 5 min extension at 72 °C. Products wereelectrophoresed on a 1.5% agarose gel.

Melting temperature (°C) Genome positiona Polarity

48.7 5,026–5,045 Sense50.3 5,143–5,161 Antisense45.2 10,865–10,882 Sense46.0 10,973–10,991 Antisense62.5 - Sense71.6 - Sense56.3 - Sense

9024).

Table 2Performance characteristics of positive-strand and negative-strand qRT-PCR assays.

Assays (strand) Sample detection(n = 29)a

LOQ (copies per reaction) Slope R2 Efficiency (%) Dynamic range(log10 copies)

nsP3 (+) 29/29 1 log10 −3.31 0.999 100.6 1–9E1 (+) 26/29 1 log10 −3.21 0.986 105.0 1–9nsP3 (−) - 3 log10 −3.43 1.000 95.7 3–9

a The denominator used was the number of samples for which at least 1 qRT-PCR assay was positive.

Fig. 1. CHIKV qRT-PCR standard curves. The qRT-PCR standard curves for nsP3 positive-strand (A), nsP3 negative-strand (B), and E1 positive-strand assays (C).

135C.W. Chiam et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 133–137

The qRT-PCR was performed in the StepOnePlus Real-Time PCRSystem (Life Technologies) using a 10 μL reaction volume containing 1×Power SYBR Green PCR Master Mix (Life Technologies), 1 μmol/L senseand antisense primers, 1 μL of cDNA (equivalent to 0.175 μL of serum),and 3.8 μL of nuclease-free water. For the positive-strand assays, nsP3-FandnsP3-R primerswere used to detect nsP3,while CHIKE1-F and CHIKE1-R primers were used for E1 detection. Tag and nsP3-R primers wereused for the nsP3 negative-strand assay. Cycling parameters were 95 °Cfor 10min, then 40 cycles of 95 °C for 15 s, and 60 °C for 1 min. Meltingcurve analysis was used to verify the amplified product by its specificmelting temperature. Experiments were repeated 3 times. Both nsP3and E1 positive-strand assays were tested on RNA extracted frompatient samples. The nsP3 positive- and negative-strand assays weretested on CHIKV-infected Vero cells.

The resulting number of RNA copies per reaction was multipliedby 5714 to obtain a final viral RNA copy number per milliliter ofinitial sample. PCR amplification efficiency was calculated using for-mula E = 100 (10−1/slope − 1).

2.10. Performance characteristics of qRT-PCR assays

To assess specificity, positive-strand assays were tested on DENVand SINV RNA. To confirm strand-specificity of the negative-strandnsP3 qRT-PCR assays, the assay was tested with positive-strand cDNAof up to 9 log10 RNA copies/reaction. The limit of quantification (LOQ)was determined as the lowest dilution of cDNA detectable by eachassay in all 3 replicates tested.

2.11. Patient follow-up

In an earlier study, we contacted laboratory-confirmed CHIKVpatients 15–24 months after discharge to determine if they had hadpersistent arthralgia lasting N4 months (Mohd Zim et al., 2013). Withthe viral loads now available from this study, the viral loads werecompared between patients with and without persistent arthralgia.

2.12. Statistical analysis

Comparison of patients' viral loads using the nsP3 and E1 assayswas performed with the Wilcoxon signed-rank test, using SPSSversion 20 (IBM, Armonk, NY, USA).

3. Results

3.1. Assay performance characteristics

The LOQs for positive-strand nsP3 and E1 assays were 1 log10 RNAcopies/reaction, while the LOQ of the negative-strand nsP3 assay was3 log10 RNA copies/reaction (Table 2, Fig. 1). The conventional PCRfor the nsP3 and E1 positive-strand assays and the nsP3 negative-strand assay were less sensitive, with LOQs of 4 log10 RNA copies/reaction (Supplementary Figure 1). Compared to E1, the positive-strand nsP3 assay had more optimum R2, slope, and efficiency.

The 2 positive-strand assays were specific for CHIKV and did notdetect RNA from negative control serum and DENV RNA (at 163.6ng/uL) and SINDV RNA (at 114.5 ng/uL). The negative-strand nsP3

assays were also highly strand-specific and did not detect the pres-ence of the complementary RNA at concentrations of up to 9 log10RNA copies/reaction. Melting curve analysis showed that the spe-cific melting temperatures for nsP3 positive- and negative-strandand E1 positive-strand assays were 81.7 ± 0.1 °C, 82.8 ± 0.2 °C, and82.5 ± 0.1 °C, respectively (Supplementary Figure 2).

3.2. Quantification of clinical samples

At least 1 of the qRT-PCR assays was positive for 29/30 (96.7%) ofthe laboratory-confirmed samples. The single CHIKV sample, whichwas negative for both qRT-PCR assays, came from an IgM-positivepatient at day 7 of illness, who was most likely no longer viraemic or

Fig. 3. RNA viral loads and persistent arthralgia. The viral loads from acute-phaseclinical samples (n = 17) were compared between patients with persistent arthralgialasting N4 months (○) and those without (●). The lines indicate medians. Samples aregrouped into 1–3, 4–6, and 7–9 days after onset of illness. The dashed line indicates theLOQ (1 log10 RNA copies/reaction or 4.76 log10 RNA/mL) of the positive assay.

Fig. 4. Virus titration and positive- and negative-strand RNA quantification of CHIKV-infected Vero cells. Positive-strand (●) and negative-strand (○) RNA quantification(left axis) and virus titration (▲, right axis) were plotted as medians ± interquartilerange from 2 independent experiments. The dashed lines indicate the LOQs of thepositive-strand assay (1 log10 RNA copies/reaction or 4.76 log10 RNA/mL) and thenegative-strand assay (3 log10 RNA copies/reaction or 6.76 log10 RNA/mL).

Fig. 2. RNA quantification of clinical samples and comparison between positive-strandassays. qRT-PCR was performed using primers amplifying nsP3 (A) and E1 (B). Circledpoints represent samples, which also had positive CHIKV IgM (n= 15). The dashed lineindicates the LOQ (1 log10 RNA copies/reaction or 4.76 log10 RNA/mL) of the positive-strand assays. (C) Spearman's correlation between the RNA copies detected using thensP3 and E1 positive-strand assays (R2 = 0.963, P b 0.001).

136 C.W. Chiam et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 133–137

had viraemia below the limit of detection. Of these 29 samples, thensP3 assay was positive in all, and the E1 assay was positive in 26. Theclinical viral loads for nsP3 and E1 genes showed similar patterns overdays 1–9 of onset of illness (Figs. 2A and 2B). The viral loads peaked atday 2 at 10.5–12.9 log10 RNA copies/mL and then declined steadily to5.6–7.2 log10 RNA copies/mL on day 6. The positive qRT-PCR assayscould detect RNA in 12/14 (85.7%) samples even in the presence ofIgM, collected from days 5 to day 9. Viral loads ranged from 5.2 to 12.1log10 RNA copies/mL for nsP3, and 4.0 to 12.9 log10 RNA copies/mL forE1. The viral loads detected using nsP3 positive-strand and E1positive-strand assays showed significant correlation (Spearman'sR2 = 0.963, P b 0.001) (Fig. 2C).

When tested on both CHIKV-confirmed and CHIKV-negativesamples, the sensitivity, specificity, PPV, and NPV rates for the E1assay were 86.7%, 100%, 100%, and 83.3%, respectively. The nsP3 assayhad higher rates of 96.7%, 100%, 100%, and 95.2%, respectively.

Follow-up data were available for 17/29 (58.6%) of the patientswith acute-phase serum tested. There were no significant differencesbetween the viral loads of patients with persistent arthralgia andthose without (Fig. 3).

3.3. CHIKV quantification of infected Vero cells

Virus titre, positive-strand, and negative-strand nsP3 RNA werequantified in CHIKV-infected Vero cells (Fig. 4). CHIKV negative-strand RNA increased exponentially in the first 16 hours to a plateauof 12.9–13.6 log10 RNA copies/mL. The increases of positive-strandRNA and viral titres were more gradual, reaching their peaks intandem at 40–48 hours.

4. Discussion

In this study, we designed positive- and negative-strand qRT-PCRassays for CHIKV nsP3. The LOQ of the positive-strand nsP3 assay was 1log10 RNA copies/reaction, which is comparable to the range of 0.5–1.5log10 copies/reaction reported in other studies (Carletti et al., 2007;Edwards et al., 2007; Grivard et al., 2007; Laurent et al., 2007; Nazeet al., 2009; Panning et al., 2009; Parida et al., 2007; Pastorino et al.,2005). The 2 positive-strand assays were 1000 times more sensitivethan conventional PCR. Both nsP3 and E1 positive-strand assaysshowed high coefficients of determination (R2), while the nsP3 assayhad the slope closest to−3.3 and the greater efficiency. The nsP3 assayalso had the highest positive detection rate for the 29 clinical samples.Thus, the positive-strand nsP3 assay appears to be the more suitable.

The highest viral load was detected on the second day of illness innsP3 and E1 positive-strand assays. This is consistent with otherreports that the highest viral loads are detected in the first to thirddays after onset of illness, with viremia usually lasting up to 5–7 days

137C.W. Chiam et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 133–137

(Laurent et al., 2007; Parola et al., 2006; Rezza et al., 2007). Levels of5.8–7.1 log10 RNA copies/mL have been reported up to 12 days afterclinical onset (Laurent et al., 2007). Our assays could detect RNA atlevels of 8.7 log10 RNA copies/mL up to the 9th day, even in thepresence of IgM. Therefore, the qRT-PCR assays are useful fordiagnosis even in the early convalescent period, as they are sensitiveenough to detect low levels of residual viral RNA.

Measurement of the acute viral load may have clinical andprognostic implications, although the existing data for CHIKV areinconsistent. The disease has been associated with arthralgiapersisting after acute infection in many settings, including 45% ofour Kuala Lumpur cohort (Mohd Zim et al., 2013). It has been shownthat higher viral loads were associated with more severe acutedisease in both humans (Chow et al., 2011) and a non-humanprimate model (Labadie et al., 2010). However, higher viral loadsalso appear to induce IgG3 production, leading to more effectiveviral clearance, and reduced chances of persistent arthralgia (Kamet al., 2012). Although there were limited numbers of patients in ourstudy for which both viral load and follow-up data were available,we found no significant association between viral load and per-sistent arthralgia, as did a prospective study from Réunion (Schilteet al., 2013). This is an important issue to address in future cohorts,in order to identify those at risk of long-term arthralgia who maybenefit from therapeutic intervention.

The tagged-primer system has been shown to improve specificityand accuracy in the quantification of negative-strand RNA (Komurian-Pradel et al., 2004; Plaskon et al., 2009). Our negative-strand nsP3 assaywas able to retain specificity even in the presence of up to 9 log10 RNAcopies/reaction of positive RNA. The LOQ of the negative-strand nsP3assay was 3 log10 RNA copies/reaction, which is comparable to thesensitivity of between 2 and 3 log10 RNA copies/reaction described for apreviously reported negative-strand assay for CHIKV nsP1 (Plaskonet al., 2009). We show that quantification of both positive and negativeRNA strands in CHIKV-infected Vero cells was consistent with theknown replication cycle of alphaviruses (Jose et al., 2009). An early,rapid increase of negative-strand RNA within the first 16 hours wasshown, following CHIKV entry during early infection. The negative-strand then functions as the template for subsequent positive-strandRNA replication and CHIKV virion formation and release. This wasshown as a more gradual rise in positive-strand RNA and CHIKV titre,with a later peak at 40–48 hours.

In summary, the nsP3 positive-strand qRT-PCR assay is suitable fordiagnostic purposes up to 9 days post-onset of illness, with LOQ of1 log10 RNA copies/reaction. The nsP3 negative-strand qRT-PCR isuseful to study active replication of CHIKV.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.diagmicrobio.2013.06.018.

Acknowledgments

This study was funded by University of Malaya (HIR grant E000013-20001 and PPP grant PG030-2012B), the Ministry of Higher EducationMalaysia (FRGS grant FP036-2013A), and the EuropeanUnion’s SeventhFramework Programme (grant agreement no. 261202).

References

Arankalle VA, Shrivastava S, Cherian S, Gunjikar RS, Walimbe AM, Jadhav SM, et al.Genetic divergence of chikungunya viruses in India (1963-2006) with specialreference to the 2005-2006 explosive epidemic. J Gen Virol 2007;88:1967–76.

Bannister R, Rodrigues D, Murray EJ, Laxton C, Westby M, Bright H. Use of a highlysensitive strand-specific quantitative PCR to identify abortive replication in themouse model of respiratory syncytial virus disease. Virol J 2010;7:250.

Carletti F, Bordi L, Chiappini R, Ippolito G, Sciarrone MR, Capobianchi MR, et al. Shortreport: Rapid detection and quantification of chikungunya virus by a one-stepreverse transcription–polymerase chain reaction real-time assay. Am J Trop MedHyg 2007;77:521–4.

Chow A, Her ZS, Ong EKS, Chen JM, Dimatatac F, Kwek DJ, et al. Persistent arthralgiainduced by chikungunya virus infection is associated with interleukin-6 andgranulocyte macrophage colony-stimulating factor. J Infect Dis 2011;203:149–57.

Edwards CJ, Welch SR, Chamberlain J, Hewson R, Tolley H, Cane PA, et al. Moleculardiagnosis and analysis of chikungunya virus. J Clin Virol 2007;39:271–5.

Grivard P, Rouxa KL, Laurenta P, Fianub A, Perrau J, Gigan J, et al. Molecular andserological diagnosis of chikungunya virus infection. Pathol Biol 2007;55:490–4.

Hasebe F, Parquet MC, Pandey BD, Mathenge EGM, Morita K, BalasubramaniamV, et al. Combined detection and genotyping of chikungunya virus by aspecific reverse transcription-polymerase chain reaction. J Med Virol 2002;67:370–4.

Jose J, Snyder JE, Kuhn RJ. A structural and functional perspective of alphavirusreplication and assembly. Future Microbiol 2009;4:837–56.

Kam YW, Simarmata D, Chow A, Her ZS, Teng TS, Ong EKS, et al. Early appearanceof neutralizing immunoglobulin G3 antibodies is associated with chikungunyavirus clearance and long-term clinical protection. J Infect Dis 2012;205:1147–54.

Komurian-Pradel F, Perret M, Deiman B, Sodoyer M, Lotteau V, Paranhos-Baccalà G,et al. Strand specific quantitative real-time PCR to study replication of hepatitis Cvirus genome. J Virol Methods 2004;116:103–6.

Labadie K, Larcher T, Joubert C, Mannioui A, Delache B, Brochard P, et al. Chikungunyadisease in nonhuman primates involves long-term viral persistence in macro-phages. J Clin Invest 2010;120:894–906.

Lam SK, Chua KB, Hooi PS, RahimahMA, Kumari S, TharmaratnamM, et al. Chikungunyainfection—an emerging disease in Malaysia. Southeast Asian J Trop Med PublicHealth 2001;32:447–51.

Laurent P, Roux KL, Grivard P, Bertil G, Naze F, Picard M, et al. Development of asensitive real-time reverse transcriptase PCR assay with an internal control todetect and quantify chikungunya virus. Clin Chem 2007;53:1408–14.

Mackay IM, Arden KE, Nitsche A. Real-time PCR in virology. Nucl Acids Res 2002;30:1292–305.

Mohd Zim MA, Sam IC, Syed Omar SF, Chan YF, AbuBakar S, Kamarulzaman A.Chikungunya infection in Malaysia: comparison with dengue infection in adultsand predictors of persistent arthralgia. J Clin Virol 2013;56:141–5.

Naze F, Rouxa KL, Schuffenecker I, Zeller H, Staikowsky F, Grivard P, et al. Simultaneousdetection and quantitation of chikungunya, dengue and West Nile viruses bymultiplex RT-PCR assays and dengue virus typing using High Resolution Melting.J Virol Methods 2009;162:1–7.

Ng LC, Tan LK, Tan CH, Tan SS, Hapuarachchi HC, Pok KY, et al. Entomologic and virologicinvestigation of chikungunya, Singapore. Emerg Infect Dis 2009;15:1243–9.

Panning M, Hess M, Fischer W, Grywna K, Pfeffer M, Drosten C. Performance of theRealStar chikungunya virus real-time reverse transcription-PCR kit. J Clin Microbiol2009;47:3014–6.

Parida MM, Santhosh SR, Dash PK, Tripathi NK, Lakshmi V, Mamidi N, et al. Rapid andreal-time detection of chikungunya virus by reverse transcription loop-mediatedisothermal amplification assay. J Clin Microbiol 2007;45:351–7.

Parola P, de Lamballerie X, Jourdan J, Rovery C, Vaillant V, Minodier P, et al. Novelchikungunya virus variant in travelers returning from Indian Ocean Islands. EmergInfect Dis 2006;12:1493–9.

Pastorino B, Bessauda M, Grandadama M, Murric S, Toloua HJ, Peyrefitte CN.Development of a TaqMan RT-PCR assay without RNA extraction step for thedetection and quantification of African chikungunya viruses. J Virol Methods2005;124:65–71.

Plaskon NE, Adelman ZN, Myles KM. Accurate strand-specific quantification of viralRNA. PLoS One 2009;4:e7468.

Powers AM, Logue CH. Changing patterns of chikungunya virus: re-emergence of azoonotic arbovirus. J Gen Virol 2007;88:2363–77.

Rezza G, Nicoletti L, Angelini R, Romi R, Finarelli AC, Panning M, et al. Infection withchikungunya virus in Italy: an outbreak in a temperate region. Lancet 2007;370:1840–6.

Richardson J, Molina-Cruz A, Salazar MI, Black IV W. Quantitative analysis of dengue-2virus RNA during extrinsic incubation period in individual Aedes aegypti. Am J TropMed Hyg 2006;74:132–41.

Sam IC, Chan YF, Chan SY, Loong SK, Chin HK, Hooi PS, et al. Chikungunya virus of Asianand Central/East African genotypes in Malaysia. J Clin Virol 2009;46:180–3.

Schuffenecker I, Iteman I, Michault A, Murri S, Frangeul L, Vaney MC, et al. Genomemicroevolution of chikungunya viruses causing the Indian Ocean outbreak. PLoSMed 2006;3:1058–70.

Schilte C, Staikovsky F, Couderc T, Madec Y, Carpentier F, Kassab S, et al. Chikungunyavirus-associated long-term arthralgia: a 36-month prospective longitudinal study.PLoS Negl Trop Dis 2013;7:e2137.

Shirako Y, Strauss JH. Regulation of Sindbis virus RNA replication: uncleaved P123and nsP4 function in minus-strand RNA synthesis, whereas cleaved productsfrom P123 are required for efficient plus-strand RNA synthesis. J Virol 1994;68:1874–85.