A

Sa

b

ARRAA

KDAFLF

1

tbviw2pwive

tadgcums

1

0d

Journal of Virological Methods 167 (2010) 172–177

Contents lists available at ScienceDirect

Journal of Virological Methods

journa l homepage: www.e lsev ier .com/ locate / jv i romet

simple method for Alexa Fluor dye labelling of dengue virus

ummer Li-Xin Zhanga,b, Hwee-Cheng Tanb, Brendon J. Hansona, Eng Eong Ooia,b,∗

Defence Medical and Environmental Research Institute, DSO National Laboratories, 27 Medical Drive, 117510 Singapore, SingaporeDuke-NUS Graduate Medical School, 8 College Road, 169857 Singapore, Singapore

rticle history:eceived 6 March 2010eceived in revised form 31 March 2010ccepted 8 April 2010vailable online 23 April 2010

a b s t r a c t

Dengue virus causes frequent and cyclical epidemics throughout the tropics, resulting in significant mor-bidity and mortality rates. There is neither a specific antiviral treatment nor a vaccine to prevent epidemictransmission. The lack of a detailed understanding of the pathogenesis of the disease complicates theseefforts. The development of methods to probe the interaction between the virus and host cells would thusbe useful. Direct fluorescence labelling of virus would facilitate the visualization of the early events in

eywords:engue viruslexa Fluorluorescent virusabellingluorescence

virus–cell interaction. This report describes a simple method of labelling of dengue virus with Alexa Fluorsuccinimidyl ester dye dissolved directly in the sodium bicarbonate buffer that yielded highly viable virusafter labelling. Alexa Fluor dyes have superior photostability and are less pH-sensitive than the commondyes, such as fluorescein and rhodamine, making them ideal for studies on cellular uptake and endosomaltransport of the virus. The conjugation of Alexa Fluor dye did not affect the recognition of labelled denguevirus by virus-specific antibody and its putative receptors in host cells. This method could have useful

l stud

applications in virologica

. Introduction

Dengue is a significant disease globally. With rapid urbaniza-ion, lack of vector control and increase in air travel, dengue hasecome the most common and rapidly spreading arthropod-borneiral disease (Gubler, 1998). An estimated 50–100 million denguenfections occur annually, and more are at risk of being infected

ith 2.5 billion people living in dengue endemic countries (WHO,007). Despite research efforts, however, there remains an incom-lete understanding on the mechanism underlying pathogenesis,hich limits antiviral and vaccine development. Among these gaps

n knowledge are the specific cell-surface binding motifs and earlyiral entry processes. The availability of tools to visualize these earlyvents can accelerate this area of research.

Dengue virus (DENV) is an enveloped, positive strand RNA virushat belongs to the family of Flaviviridae. Its structure consists ofn external icosahedral scaffold of 90 envelope glycoprotein (E)imers protecting the nucleocapsid shell, which contains the RNAenome (Kuhn et al., 2002). Dengue virus surface has several identi-

al protein subunits and can therefore be labelled at multiple sitessing an amine reactive dye; the resulting fluorescence intensityay be sufficient to track the virus under a fluorescence micro-

cope.

∗ Corresponding author at: Duke-NUS Graduate Medical School, 8 College Road,69857 Singapore, Singapore. Tel.: +65 6516 8594; fax: +65 6221 2529.

E-mail address: engeong.ooi@duke-nus.edu.sg (E.E. Ooi).

166-0934/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.jviromet.2010.04.001

ies.© 2010 Elsevier B.V. All rights reserved.

A simple procedure of labelling dengue virus with a fluores-cent Alexa Fluor succinimidyl ester that results in minimal lossof infectivity post-labelling is described. Alexa Fluor dye was cho-sen as it has better photostability and signal intensity comparedto common fluorophores such as fluorescein isothiocyanate (FITC)or cyanine 3 bihexanoic acid (Cy3). In addition, Alexa Fluor dyesare stable at a lower pH compared to the common fluorophores,which could be useful for visualizing labelled viral particles asthey move through the acidic compartments of endosomes post-infection. Furthermore, the permanence of covalent conjugationof the fluorophores to the surface proteins allows for storage ofbatch-labelled virus at −80 ◦C, providing uniformity across multi-ple experiments. It also opens up the possibility of performing livecell imaging to compliment the ‘snapshot’ imaging of conventionalfluorescence microscopy.

2. Materials and methods

2.1. Cells and antibody

All cells and hybridomas were obtained from The AmericanType Culture Collection (ATCC, Manassas, VA), and all cell culturemedia and supplements were purchased from Gibco (Invitrogen,

Singapore). Baby hamster kidney BHK-21 cells were maintained inRoswell Park Memorial Institute 1640 medium (RPMI-1640), sup-plemented with 10% fetal bovine serum, penicillin (100 U/ml) andstreptomycin (100 �g/ml) at 37 ◦C with 5% CO2. African green mon-key kidney Vero cells were maintained in Medium-199 (M199)

ologic

satmsSBRFM

2

caiotcl(ifswf

2

2iw230f

2

m5w(35dws(gla

bDHwfl

2

c

S.L.-X. Zhang et al. / Journal of Vir

upplemented with 10% fetal bovine serum and 4 mM l-glutaminet 37 ◦C with 5% CO2. Mouse anti-dengue virus serotype 2 E pro-ein monoclonal antibody hybridoma, 3H5 (ATCC: HB46), was

aintained in RPMI-1640 supplemented with 10% fetal bovineerum. Dendritic cell-specific ICAM3-grabbing non-integrin (DC-IGN) transfected-Raji B cell line was a kind gift from Timothy H.urgess (Naval Medical Research Centre, US) and maintained inPMI medium supplemented with 10% fetal bovine serum. Alexaluor 488 (AF488) anti-mouse IgG antibody was purchased fromolecular Probes, Invitrogen and used at 1:100 dilution.

.2. Virus culture and purification

DENV serotype 2 (DENV-2) (ST) strain was first isolated from alinical sample in the Singapore General Hospital using an Aedeslbopictus mosquito cell line C6/36 and subsequently propagatedn Vero cells. The culture supernatant was harvested when 75%f the cells showed cytopathic effect and clarified by centrifuga-ion at 1000 × g for 30 min at 4 ◦C. Virus in the supernatant wasoncentrated by centrifugation at 30,000 × g for 2 h at 4 ◦C. The pel-et was resuspended in 5 mM Hepes, 150 mM NaCl, 0.1 mM EDTAHNE buffer, pH 7.4) and purified further on a 30% sucrose cush-on by ultracentrifugation in a Beckman SW41Ti rotor at 80,000 × gor 15 h at 4 ◦C. Purified virus was resuspended in HNE buffer andtored in 100 �l aliquots at −80 ◦C. A limiting dilution plaque assayas performed on BHK cell line to determine the viral titre in plaque

orming units per millilitre (pfu/ml).

.3. Plaque assay

Tenfold serial dilutions of virus were added to BHK-21 cells in4 well plates and incubated for 1 h at 37 ◦C, with gentle rock-

ng every 15 min. The medium was then aspirated and replacedith 0.8% methyl-cellulose in maintenance medium (RPMI-1640,

% fetal bovine serum, penicillin and streptomycin). After 4 days at7 ◦C, the cells were fixed with 25% formaldehyde and stained with.5% crystal violet. The plates were washed, dried and the plaqueorming units per ml (pfu/ml) calculated.

.4. Virus labelling

For the initial experiment aimed at determining the opti-al concentration of dye needed to label DENV, Alexa Fluor

94 succinimidyl ester (AF594SE, Molecular Probes, Invitrogen)as reconstituted in 0.2 M sodium bicarbonate buffer, pH 8.3

SBB) immediately before labelling, and added to approximately× 107 pfu DENV diluted in SBB at final concentrations of 10,0,100, 200, 500 and 1000 �M of dye, while stirring gently. Theye-virus mixture was incubated at room temperature for 1 hith gentle inversions every 15 min. The labelling reaction was

topped by adding freshly prepared 1.5 M hydroxylamine, pH 8.5Sigma–Aldrich) and incubated at room temperature for 1 h withentle inversions every 15 min. Labelled DENV was re-titrated afterabelling and tested for fluorescence using immunofluorescencessay.

For subsequent AF594SE or AF488SE labelling involving largeratches of virus, the same approach was employed and the labelledENV was purified using Sephadex G-25 columns (Amersham, GEealthcare, Singapore) to remove the excess dye. The labelled virusas stored in 100 �l aliquots at −80 ◦C, re-titrated and tested foruorescence before use.

.5. Immunofluorescence assay

Equal volume of AF594-labelled DENV was incubated with Veroells plated on coverslips for 10 min at 37 ◦C, washed, fixed with 3%

al Methods 167 (2010) 172–177 173

paraformaldehyde (Sigma–Aldrich, Singapore) and permeabilizedwith 0.1% saponin (Sigma–Aldrich, Singapore). The cells were thenincubated for 1 h with undiluted centrifugation-clarified super-natant of 3H5 monoclonal antibody hybridoma culture, at roomtemperature. The cells were washed three times in wash buffer (PBScontaining 1 mM calcium chloride, 1 mM magnesium chloride and0.1% saponin), followed by incubation with AF488 anti-mouse IgGantibody for 45 min at room temperature. Cells were then washedthree times with the wash buffer, rinsed once with deionized waterand mounted on glass slides with Mowiol 4-88 (Calbiochem, SanDiego, CA) with 2.5% Dabco (Sigma–Aldrich, Singapore). Cells werevisualized at 63× magnification using a Leica Microsystem TCS SP5confocal microscope and merged images exported for processing.Processing of the images involved adjustment of the contrast of theimages on a whole for clarity and exported in individual colours forunmerged images using Adobe Photoshop CS3 version 10.

2.6. Flow cytometry determination of labelled DENV

Raji B cells transfected with DC-SIGN were counted, aliquoted,centrifuged and resuspended in AF488-labelled DENV at multiplic-ity of infection (MOI) of 1 and incubated for 10 min at 37 ◦C, thenfixed with 3% paraformaldehyde. The cells were then washed twicewith FACS buffer consisting of PBS with 0.1% fetal bovine serum andpermeabilized with 0.1% saponin in PBS. Cells were stained subse-quently for the presence of DENV using anti-E protein monoclonalantibody, 3H5, and PE anti-mouse anti-IgG antibody before readingon BD FACS Calibur machine and analysed with CellQuest software,version 3.3 (Becton Dickinson).

2.7. Detection by SYBR green-based real-time PCR

Real-time detection of viral RNA was conducted as describedpreviously (Lai et al., 2007) using pan-dengue forward and reverseprimers with some modifications. Briefly, RNA extraction was car-ried out using QIAamp viral RNA mini kit for purified non-labelledand AF594-labelled DENV (AF594-DENV), or RNeasy mini kit (bothfrom QIAGEN, Hilden, Germany) for total RNA extraction from cellsaccording to the manufacturer’s instructions. Complementary DNAwas synthesized using SuperScript III First Strand Synthesis Sys-tem (Invitrogen, Singapore) with random hexamers and accordingto the manufacturer’s instructions. RNA copy number was thendetermined by SYBR green-based real-time PCR on LightCycler 480Real-Time PCR System (Roche Diagnostics, Penzberg, Germany).

2.8. Growth kinetics

Vero cells were seeded at 2 × 105 per well in 24 well plates andincubated at 37 ◦C for 4 h before infecting at a MOI of 0.1 with eitherpurified DENV or AF594-DENV for 1 h at 37 ◦C, with gentle rockingevery 15 min. The cells were then washed twice with 1 ml per wellPBS and replaced with maintenance medium (M199 supplementedwith 2% FBS). Supernatant was harvested from wells and stored at−80 ◦C for plaque assay, and cells were washed twice in PBS andlysed with RLT buffer provided in RNeasy mini kit at 0, 6, 8, 12, 24, 48and 72 h post-adsorption. The cell lysates were stored frozen untilall samples were collected and subsequently extracted accordingto the kit’s protocol. Viral RNA was detected using real-time PCR.

2.9. Statistical analysis

Statistical analyses were performed using unpaired Student’s ttest (Graphpad Prism 5.0). P value < 0.05 was considered significant.

1 ological Methods 167 (2010) 172–177

3

3

AoDieo

3

Afioca

dopcaDAhAo

3

luCmEaocaA

3

fotattedtydrPt

Fig. 1. Dengue virus viability post AF594 labelling. Purified dengue virus waslabelled with various concentrations of AF594SE dye for 1 h at room temperature(described in Section 2.4). The labelled virus was then re-titrated post-labelling byplaque assay. Bar diagram shows average virus titre of four determinations, pre- andpost-labelling (a). Error bars indicate standard deviation. The labelled virus was alsotested for fluorescence intensity by immunofluorescence assay (b). Vero cells seededon coverslips the day prior are infected with labelled DENV for 10 min at 37 ◦C. Thecells were subsequently fixed and labelled with anti-E antibody, and examined forco-localization of E protein (green) and AF594 labelling (red). Fluorescent signals

74 S.L.-X. Zhang et al. / Journal of Vir

. Results

.1. Infectivity of DENV post-labelling

DENV was re-titrated by plaque assay post-labelling withF594SE (Fig. 1a). The virus titre of DENV labelled with 10–100 �Mf AF594SE dye showed approximately 15-fold reduction, andENV labelled with 500 and 200 �M AF594SE dye showed approx-

mately 26- to 29-fold reductions, respectively, compared toquivalent number of non-labelled virus. DENV labelled with 1 mMf AF594SE dye showed an 82-fold reduction.

.2. Intensity of AF594 labelling

Equal volume of DENV labelled with various concentrations ofF594SE dye was incubated with Vero cells seeded on coverslips,xed in 3% paraformaldehyde and stained with mouse anti-E mon-clonal antibody, 3H5, and AF488 anti-mouse IgG antibody. Theoverslips were then mounted on glass slides and examined usingconfocal microscope.

Cells infected with DENV labelled with 10 �M to 1 mM AF594SEye showed decreased staining for E protein with concentrationf dye above 200 �M, indicating a reduction in virus infectivityost-labelling with higher concentrations of dye (Fig. 1b). This wasonsistent with the plaque assay results. DENV labelled with 10nd 50 �M of dye showed no discernable AF594 fluorescence, whileENV labelled with 100 and 200 �M of AF594SE dye displayed highF594 fluorescence which co-localized well with the E protein asighlighted by the 3H5 monoclonal antibody. Fluorescence for bothF594 and E protein diminished with AF594SE dye concentrationsf 500 �M and 1 mM.

.3. Efficiency of Alexa Fluor dye labelling

DC-SIGN transfected-Raji B cells were infected with AF488-abelled DENV at a MOI of 1, stained for the presence of E proteinsing 3H5 monoclonal antibody and analysed by flow cytometry.ompared to the uninfected Raji B cells with DC-SIGN, approxi-ately 81.4% of the cells following infection were positive for theprotein. Of these cells, 58.4% were positive for both E protein

nd AF488 (Fig. 2a), suggesting that approximately 71% of the viri-ns were labelled with the Alexa Fluor dye. Similar conclusionsan be drawn from the Pearson’s correlation values obtained bynalysis of confocal microscopy images showing co-localization ofF594-DENV with E protein (Fig. 2b).

.4. Reproducibility of labelling

Three independent AF594SE labelling experiments were per-ormed with the same approximate starting titre of 2 × 108 pfuf purified DENV using the same procedure as described in Sec-ion 2. The titre post-labelling was then determined by plaquessay. All three experiments showed three- to eightfold reduc-ion in titre post-labelling and column purification, from a meanitre of 2.21 × 108 (SD = 3.68 × 107) to 3.78 × 107 (SD = 1.5 × 107). Toxclude the possibility that the reduction in titre post-labelling wasue to a loss in infectivity of the virus, the ratio of RNA copy numbero plaque forming unit of DENV pre- and post-labelling was anal-

sed (the same batch of purified virus was used). No significantifference in the viral RNA copy number to plaque forming unitatio was observed between pre- and post-labelled virus (Table 1,= 0.9507), indicating that the fall in titre was not due to inactiva-

ion of the virions but rather due to loss during column purification.

are visualized under 63× magnification using Leica Microsystem TCS SP5 confo-cal microscope. Scale bar is 10 �m. White arrows indicate areas of co-localization(yellow when merged). (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of the article.)

S.L.-X. Zhang et al. / Journal of Virological Methods 167 (2010) 172–177 175

Fig. 2. Efficiency of Alexa Fluor dye labelling as a function of percentage infected. Raji B cells transfected with DC-SIGN were infected at a MOI of 1 with AF488-DENV for 10 minat 37 ◦C, fixed and stained for the presence of E protein. Fluorescence was then measured using BD FACS Calibur (a). Close to 100% of uninfected Raji B cells with DC-SIGNwere double negative for both Alexa Fluor and anti-E antibody signals, whereas approximately 81% of the infected cells stained positive for E protein, and approximately7 tely 71c ean Pe colou

3

ovAwawRs

TV

Vcb(cp

1% of which are positive for Alexa Fluor fluorescence, suggesting that approximaells infected with AF594-DENV (red) and stained for E protein (green) showed a mfficiency (yellow indicates co-localization). (For interpretation of the references to

.5. Growth kinetics of labelled DENV

In order to ascertain further that labelling of the surface proteinsf DENV did not interfere with infectivity and growth kinetics of theirus, Vero cells were infected with either purified non-labelled orF594-DENV at a MOI of 0.1. Culture supernatant and cell lysatesere collected up to 72 h and assayed for virus titre in supernatant

s well as viral RNA in the cell lysates. No significant differenceas observed between labelled and unlabelled virus for both viralNA copy number and the number of infectious particles in theupernatant of the culture (Fig. 3) (P > 0.8).

able 1iral RNA copy number to infectious particle ratio.

Titre (pfu/ml) RNA copynumber (/ml)

RNA copy numberto pfu ratio

AF594-DENV 3.5 × 107 6.27 × 108 17.9Purified DENV 1.0 × 109 1.67 × 1010 16.7

iral RNA was extracted from purified non-labelled DENV or AF594-DENV and RNAopy number quantitated by real-time PCR. The RNA copy number was then dividedy the titre to obtain a ratio for both non-labelled DENV (16.7) and AF594-DENV17.9). The similar ratios indicated that both labelled and non-labelled DENV haveomparable infectious/non-infectious proportions, suggesting that the labellingrocess did not inactivate the virions.

% of the virions were labelled with the dye. Confocal microscope analysis of Veroearson’s correlation of 0.677 ± 0.053 (mean ± SD) (b), suggesting similar labellingr in this figure legend, the reader is referred to the web version of the article.)

4. Discussion

Direct fluorescent labelling of virus allows for real-time track-ing of post-internalization events and requires few labelling steps,removing the possibility of non-specific staining that could beintroduced with the use of indirect antiviral antibodies. The track-ing of single-virus entry into cell by labelling DENV with lipophilicDiD (1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine, 4-chorobenzenesulfonate salt) was described recently (van derSchaar et al., 2007, 2008). However, as DiD fluorescence is largelyquenched when the fluorophores are in close proximity to eachother (e.g. on virion surface) and will only fluoresce brightly fol-lowing fusion with cellular membranes, the DiD-labelled virionsare not traceable easily prior to fusion. This hampers the visualiza-tion of early virus–cell interaction. In addition, the DiD fluorescencepost-fusion tends to be diffused, possibly due to rapid lipid recy-cling, thus complicating co-localization studies. Furthermore, thelabelling is not stable and labelled virus can only be stored in thecold for up to 3 days. Hence, a fluorophore, such as Alexa Fluor, that

is unquenched extracellularly and stably conjugates to the virussurface proteins would enable the tracking of virus within a cellrather than visualizing lipid movement.

Currently, there is no standardized procedure for fluorescencelabelling of viruses and protein labelling protocols by manufac-

176 S.L.-X. Zhang et al. / Journal of Virologic

Fig. 3. Comparing the growth kinetics of purified non-labelled and labelled DENV.Vero cells were plated and allowed to adhere before infecting at a MOI of 0.1 witheither AF594-DENV or non-labelled DENV. Viral RNA copy number in cell lysates wasdetermined to assess viral replication (a) and supernatant was assayed for infectiouspca

trvidlaFbTvmyt

gactvgbgl1aD

Edab

Martin, N.C., Pardo, J., Simmons, M., Tjaden, J.A., Widjaja, S., Marovich, M.A., Sun, W.,

articles by plaque assay (b). There is no significant difference between the growthurves of labelled and non-labelled dengue virus in RNA copy number (P = 0.8630)nd total pfu (P = 0.8191). Each point is an average of four determinations.

urers often call for the use of dimethyl sufoxide (DMSO) toeconstitute the dye. DMSO can block productive infection ofirus in cells, even when present in minute amounts, and maynduce cell cytotoxicity (Aguilar et al., 2002) which could affectownstream assays. An alternative method of producing infective

abelled viruses without the use of DMSO would therefore be desir-ble. This initiated the assessment of labelling of DENV with Alexaluor succinimidyl ester dye dissolved directly in the sodium bicar-onate buffer since the dye is highly soluble in aqueous solution.he results obtained indicate that this approach is able to produceiable labelled DENV that can be visualized easily by fluorescenceicroscopy. The findings were also reproducible and consistently

ielded fluorescently labelled virus with less than tenfold drop initre following column purification.

Alexa Fluor dyes are small molecules that react with free aminoroups, primarily arginine and lysine (Huang et al., 2004), usu-lly outward-facing amino acids. Despite its relatively small sizeompared to other fluorescent proteins, increasing the concen-ration of the dye used for labelling correlated with a reducediral titre. This was observed with concentrations of AF594SE dyereater than 200 �M. An optimal level of labelling should thusalance between the degree of labelling and the functional abro-ation (Freistadt and Eberle, 2006). Consequently, although DENVabelled with 200 �M of dye showed brighter fluorescence over00 �M of dye, 100 �M was chosen for subsequent experimentss it provided adequate labelling with minimal loss of viability ofENV.

The labelled DENV retained the ability to be recognized by anti-

antibody and DC-SIGN expressed on transfected-Raji B cells as

emonstrated by immunofluorescence assays performed on FACSnd confocal microscope. Raji B cells are not normally suscepti-le to DENV infection, but DC-SIGN transfection renders these cells

al Methods 167 (2010) 172–177

permissive to this virus (Martin et al., 2006; Tassaneetrithep et al.,2003; Wu et al., 2004). This suggests that conjugation of the AlexaFluor dye molecules to the surface of DENV did not alter nor blockimportant epitopes of the surface glycoproteins used for receptorrecognition. Additional assays were also conducted to determinewhether labelling of the viral proteins would interfere with itsinfection and replication in normally permissive cells. Study of thegrowth kinetics in Vero cells showed no significant difference in theinfectivity or replication rate between the labelled and the non-labelled virus as indicated by the viral RNA copy number in cellsand titre in the supernatant (P > 0.8).

To estimate the percentage of virus labelled with the fluores-cence dye using this method, flow cytometric studies were carriedout using DENV labelled with AF488SE dye. Approximately 58.4%of the cell population were positive for both AF488 and anti-Estaining, out of a total of 81.4% infected. This suggests that approx-imately 71.7% of the virus preparation was actually labelled withAF488 dye. Although AF594-DENV could not be assessed for effi-ciency of labelling on the flow cytometer used, due to the lack ofappropriate excitation laser, Pearson’s correlation values obtainedfrom co-localization studies using confocal microscopy revealedthat similar labelling efficiency can be achieved.

Even though it has not been demonstrated in this report, thisapproach of labelling DENV should apply to the other serotypes. Asthe labelling chemistry is the same, different fluorophores can alsobe used (AF594 and AF488 demonstrated here). This could increasethe range of applications of the labelled virus (e.g. flow cytometry,confocal microscopy). In addition, conjugation of dye to viral pro-teins is stable and the labelled virus could be stored frozen untiluse. Experience in this laboratory indicates that labelled virus canbe stored at −80 ◦C for up to 8 months without noticeable loss influorescence or viral titre. Thus, a large batch of virus can be labelledand used for consecutive experiments to minimize inter-assay vari-ations.

In conclusion, Alexa Fluor labelled virus provides a usefulmethod for tracking the early events in virus–cell interaction.

Competing interests

The authors declare that they have no competing interests.

Acknowledgement

This work was funded by the Duke-NUS Graduate MedicalSchool Program in Emerging Infectious Diseases.

References

Aguilar, J.S., Roy, D., Ghazal, P., Wagner, E.K., 2002. Dimethyl sulfoxide blocks her-pes simplex virus-1 productive infection in vitro acting at different stages withpositive cooperativity. Application of micro-array analysis. BMC Infect. Dis. 2, 9.

Freistadt, M.S., Eberle, K.E., 2006. Fluorescent poliovirus for flow cytometric cellsurface binding studies. J. Virol. Methods 134, 1–7.

Gubler, D.J., 1998. Dengue and dengue hemorrhagic fever. Clin. Microbiol. Rev. 11,480–496.

Huang, S., Wang, H., Carroll, C.A., Hayes, S.J., Weintraub, S.T., Serwer, P., 2004. Anal-ysis of proteins stained by Alexa dyes. Electrophoresis 25, 779–784.

Kuhn, R.J., Zhang, W., Rossmann, M.G., Pletnev, S.V., Corver, J., Lenches, E., Jones, C.T.,Mukhopadhyay, S., Chipman, P.R., Strauss, E.G., Baker, T.S., Strauss, J.H., 2002.Structure of dengue virus: implications for flavivirus organization, maturation,and fusion. Cell 108, 717–725.

Lai, Y.L., Chung, Y.K., Tan, H.C., Yap, H.F., Yap, G., Ooi, E.E., Ng, L.C., 2007. Cost-effectivereal-time reverse transcriptase PCR (RT-PCR) to screen for Dengue virus fol-lowed by rapid single-tube multiplex RT-PCR for serotyping of the virus. J. Clin.Microbiol. 45, 935–941.

Porter, K.R., Burgess, T.H., 2006. An immunocytometric assay based on dengueinfection via DC-SIGN permits rapid measurement of anti-dengue neutralizingantibodies. J. Virol. Methods 134, 74–85.

Tassaneetrithep, B., Burgess, T.H., Granelli-Piperno, A., Trumpfheller, C., Finke, J., Sun,W., Eller, M.A., Pattanapanyasat, K., Sarasombath, S., Birx, D.L., Steinman, R.M.,

ologic

v

v

dengue virus cell entry by biochemical assays and single-virus tracking. J. Virol.

S.L.-X. Zhang et al. / Journal of Vir

Schlesinger, S., Marovich, M.A., 2003. DC-SIGN (CD209) mediates dengue virus

infection of human dendritic cells. J. Exp. Med. 197, 823–829.

an der Schaar, H.M., Rust, M.J., Chen, C., van der Ende-Metselaar, H., Wilschut, J.,Zhuang, X., Smit, J.M., 2008. Dissecting the cell entry pathway of dengue virusby single-particle tracking in living cells. PLoS Pathog. 4, e1000244.

an der Schaar, H.M., Rust, M.J., Waarts, B.L., van der Ende-Metselaar, H., Kuhn, R.J.,Wilschut, J., Zhuang, X., Smit, J.M., 2007. Characterization of the early events in

al Methods 167 (2010) 172–177 177

81, 12019–12028.WHO, 2007. Report of Dengue Scientific Working Group. TDR/SWG/08, Geneva.Wu, L., Martin, T.D., Carrington, M., KewalRamani, V.N., 2004. Raji B cells, misiden-

tified as THP-1 cells, stimulate DC-SIGN-mediated HIV transmission. Virology318, 17–23.