Characterization and application of monoclonal antibodies specific to West Nile virus envelope protein.

Abstract

Recent epidemics of West Nile virus (WNV) around the world have been associated with significant rates of mortality and morbidity in humans. To develop standard WNV diagnostic tools that can differentiate WNV from Japanese encephalitis virus (JEV), four monoclonal antibodies (MAbs) specific to WNV envelope (E) protein were produced and characterized by isotyping, reactivity with denatured and native antigens, affinity assay, immunofluorescence assay (IFA), and epitope competition, as well as cross-reactivity with JEV. Two of the MAbs (6A11 and 4B3) showed stronger reactivity with E protein than the others (2F5 and 6H7) in Western blot analysis. 4B3 could bind with denatured antigen, as well as native antigens in indirect ELISA, flow cytometry analysis, and IFA; whereas 2F5 showed highest affinity with native antigen. 4B3 and 2F5 were therefore used to establish an antigen capture-ELISA (AC-ELISA) detection system. The sensitivity of this AC-ELISA was 3.95 TCID(50)/0.1 ml for WNV-infected cell culture supernatant. Notably, these MAbs showed no cross-reactivity with JEV, which suggests that they are useful for further development of highly sensitive, easy handling, and less time-consuming detection kits/tools in WNV surveillance in areas where JEV is epidemic.

Full text

Page 1

Please cite this article in press as: Liu, J., et al., Characterization and application of monoclonal antibodies specific to West Nile virus envelope
protein. J. Virol. Methods (2008), doi:10.1016/j.jviromet.2008.09.019
ARTICLE IN PRESS
G Model
VIRMET-10677;No.of Pages7
Journal of Virological Methods xxx (2008) xxx–xxx
Contents lists available at ScienceDirect
Journal of Virological Methods
journal homepage: www.elsevier.com/locate/jviromet
Characterization and application of monoclonal antibodies specific
to West Nile virus envelope protein
June Liua,f, Bohua Liub, Zhen Caoc, Shingo Inoued, Kouichi Moritad,
Kegong Tianc, Qingyu Zhub, George F. Gaoa,e,∗
aCenter for Molecular Immunology and Center for Molecular Virology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China
bInstitute of Microbiology and Epidemiology, Academy of Military Medical Sciences, Beijing 100071, China
cChina Animal Disease Control Center, Beijing 100094, China
dDepartment of Virology, Institute of Tropical Medicine, Nagasaki University, Japan
eChina–Japan Joint Laboratory of Molecular Immunology and Molecular Microbiology, Institute of Microbiology,
Chinese Academy of Sciences, Beijing 100101, China
fGraduate School, Chinese Academy of Sciences, Beijing 100080, China
Article history:
Received 22 July 2008
Received in revised form 5 September 2008
Accepted 11 September 2008
Available online xxx
Keywords:
West Nile virus
Monoclonal antibody
Envelope protein domain III
Antigen capture-ELISA
a b s t r a c t
Recent epidemics of West Nile virus (WNV) around the world have been associated with significant rates
of mortality and morbidity in humans. To develop standard WNV diagnostic tools that can differentiate
WNVfromJapaneseencephalitisvirus(JEV),fourmonoclonalantibodies(MAbs)specifictoWNVenvelope
(E) protein were produced and characterized by isotyping, reactivity with denatured and native antigens,
affinity assay, immunofluorescence assay (IFA), and epitope competition, as well as cross-reactivity with
JEV. Two of the MAbs (6A11 and 4B3) showed stronger reactivity with E protein than the others (2F5 and
6H7)inWesternblotanalysis.4B3couldbindwithdenaturedantigen,aswellasnativeantigensinindirect
ELISA,flowcytometryanalysis,andIFA;whereas2F5showedhighestaffinitywithnativeantigen.4B3and
2F5werethereforeusedtoestablishanantigencapture-ELISA(AC-ELISA)detectionsystem.Thesensitivity
of this AC-ELISA was 3.95 TCID50/0.1ml for WNV-infected cell culture supernatant. Notably, these MAbs
showednocross-reactivitywithJEV,whichsuggeststhattheyareusefulforfurtherdevelopmentofhighly
sensitive,easyhandling,andlesstime-consumingdetectionkits/toolsinWNVsurveillanceinareaswhere
JEV is epidemic.
© 2008 Elsevier B.V. All rights reserved.
1. Introduction
West Nile virus (WNV) is a member of the Japanese encephalitis
virus (JEV) serocomplex of the genus Flavivirus, family Flaviviridae.
Recent epidemics of WNV around the world have been associ-
ated with significant rates of mortality and morbidity in humans
(Lanciotti et al., 1999; Gea-Banacloche et al., 2004; Gubler, 2007;
Murgueetal.,2002).However,neitheraspecifictreatmentforWNV
infection nor a preventive vaccine is available at present. In nature,
WNV exists in an enzootic cycle between mosquitoes and birds,
with birds being the principal amplifying host (Glaser, 2004). The
rapid spread of WNV is most likely caused by the migration of
infected wild birds after contact with pools of Culex mosquitoes
∗Corresponding author at: Institute of Microbiology, Chinese Academy of Sci-
ences, No. 3A, Da Tun Road, Chao Yang District, Beijing 100101, China.
Tel.: +86 10 64807688; fax: +86 10 64807882.
E-mail address: gaof@im.ac.cn (G.F. Gao).
(Malkinson et al., 2002; Rappole et al., 2000). As the clinical symp-
toms of WNV infection are non-specific compared to those of other
encephalitis viruses, diagnosis relies mainly on laboratory tests.
Serological testing is the primary method of diagnosing WNV
infection. The plaque reduction neutralization tests for type-
specific diagnosis are laborious, expensive, and require live virus,
which limits their application in large-scale surveillance. ELISA-
baseddetectionforIgM,IgGorIgAhasbeendeveloped,andsomeof
theseassaysarecommerciallyavailable(Hogrefeetal.,2004;Levett
et al., 2005; Martin et al., 2000; Prince and Lape-Nixon, 2005).
However, the serological cross-reactions and cross-neutralizations
found in the JEV serocomplex viruses limit the specificity of sero-
logical tests (Hogrefe et al., 2004; Martin et al., 2000; Niedrig et al.,
2007).
WNV viremia can serve as a clear indicator of recent infection
and is suitable for early detection because it begins within a few
days after infection and is short-lived. WNV-infected mosquitoes
can be easily detected by various virus-detection methods (Hunt et
al., 2002; Marfin and Gubler, 2001). Viral isolation depends heav-
0166-0934/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jviromet.2008.09.019

Page 2

Please cite this article in press as: Liu, J., et al., Characterization and application of monoclonal antibodies specific to West Nile virus envelope
protein. J. Virol. Methods (2008), doi:10.1016/j.jviromet.2008.09.019
ARTICLE IN PRESS
G Model
VIRMET-10677;No.of Pages7
2
J. Liu et al. / Journal of Virological Methods xxx (2008) xxx–xxx
ily on the quality of samples and requires the use of cell culture
and a BSL-3 laboratory, with 6-day delay. Reverse-transcriptase
polymerase chain reaction (RT-PCR) is expensive and prone to
contamination.Indirectimmunofluorescenceassay(IFA)withwell-
identifiedspecificmonoclonalantibodies(MAbs)canconfirmvirus
infection. WNV antigen detection tests with specific MAbs have
been used for dead birds and mosquito surveillance programs in
North America (Dauphin and Zientara, 2007). An MAb-based anti-
gen capture-ELISA (AC-ELISA) that can differentiate WNV from St
Louisencephalitisvirushasalsobeendeveloped(Huntetal.,2002).
As a result of the antigenic cross-reaction in the JEV serocom-
plex flaviviruses, it is critical to distinguish between WNV and JEV
in areas such as China and Japan where JEV is endemic. Molecu-
lar diagnostic methods that simultaneously discriminate between
WNVandJEVusingRT-PCRanalyseshavepreviouslybeenreported
(Shiratoetal.,2003,2005).MAbisthemostattractiveoptionforthe
development of standardized viral diagnostic assays. In this study,
four MAbs against WNV envelope protein domain III (EDIII) were
characterizedbyisotyping,affinityassay,reactivitywithdenatured
and native antigens, and epitope competition, as well as cross-
reactivitywithJEV.TheresultssuggesttheapplicabilityoftheMAbs
to various analytical methods, such as immunoblotting, IFA, and
AC-ELISA, for detection and pathogenic study of WNV.
2. Materials and methods
2.1. Preparation of recombinant WNV EDIII protein
The EDIII (residues 298–415) of WNV bird 5810 strain was
expressed, purified and refolded as described previously (Yuan
et al., 2005). Briefly, the recombinant protein was expressed in
Escherichia coli as an inclusion body and refolded in an appro-
priate buffer. The refolded protein was purified by gel-filtration
chromatography.
2.2. Production and purification of MAbs
Six BALB/c mice (from National Institute for the Control of Phar-
maceutical and Biological Products, Beijing, China), aged 8 weeks,
were primed intraperitoneally with 50?g recombinant EDIII pro-
tein, mixed with complete Freund’s adjuvant (Sigma–Aldrich). Two
boosts were given at days 14 and 28 with 50?g EDIII mixed with
incomplete adjuvant (Sigma–Aldrich). Three days after the last
boost, the titer of polyclonal antiserum was assessed using indirect
ELISA (described below) with EDIII as antigen. The mouse with the
highest titer was chosen to harvest splenocytes. Separated spleno-
cytes were fused with SP2/0 myeloma cells at a ratio of 5:1 using
50% (w/v) polyethylene glycol, according to a previously described
protocol (Kohler and Milstein, 1975). The hybridoma cells were
obtained and subsequently cloned by limiting dilution. The cell
linesthatproducedspecificantibodiesweresubclonedsuccessively
3–5 times by limiting dilution to ensure monoclonality and stabil-
ity.Positiveclonesthatsecretedhigh-titerEDIII-specificantibodies
in indirect ELISA were further identified. The immunoglobulin
subclass was determined using SBA Clonotyping System/AP kit
(Southern Biotechnology Associates). Four positive cell lines (6A11,
4B3, 2F5 and 6H7) were used to generate ascites in BALB/c mice
and MAbs were purified by protein A or protein G chromatography,
according to manufacturer’s protocols (Pharmacia). The concentra-
tion of purified MAb was determined by bicinchoninic acid protein
assay (Pierce Biotechnology).
2.3. Western blot analysis
ToexaminewhethertheascitesMAbsrecognizedthelinearepi-
tope of EDIII protein, Western blot analysis was performed under
denaturing conditions. EDIII protein was run on 12% SDS-PAGE,
then electrotransferred onto a nitrocellulose membrane (Amer-
sham Biosciences UK) and blocked with 5% non-fat dry milk in
Tris-buffered saline (TBS). Membranes were incubated for 2h at
room temperature with four ascites MAbs (1:2000), respectively,
and then washed 3 times with 0.05% Tween-20 in TBS (TBST),
and incubated with horseradish peroxidase (HRP)-conjugated goat
anti-mouse IgG (1:5000 dilution; Santa Cruz) for 1h at room
temperature, and detected by SuperSignal West Pico Chemilumi-
nescent substrate solution (Pierce Biotechnology). In the control
experiment, EDIII protein was incubated with an irrelevant MAb
H5, which is an anti-influenza antibody (1:2000).
The specificity of purified MAbs for WNV E protein was also
evaluated by Western blot analysis. The recombinant E proteins
of WNV (bird 5810 strain) and JEV (Beijing-1 strain) with a His
tag were expressed on the membrane of 293T cells, by transiently
transfecting pcDNA4-WNV E or pcDNA4-JEV E plasmids into 293T
cells. The cell lysate and inactivated WNV (Chin01 strain) or JEV
(Beijing-1 strain) were separated by 10% SDS-PAGE, and were then
electrotransferred onto a nitrocellulose membrane and blocked.
Membrane was incubated for 2h with purified MAb (1?g/ml) or
anti-His MAb (0.5?g/ml; Santa Cruz), as a positive control for the
expression of JEV E protein. The membrane was washed 3 times
with TBST, and incubated with HRP-conjugated goat anti-mouse
IgG secondary antibody (1:5000) for 1h, and detected by substrate
solution.
2.4. Indirect ELISA
All ELISAs were carried out in 96-well microtiter ELISA Plates
(Greiner Bio-One). Titers of hybridoma-cell-secreted MAbs were
detected by indirect ELISA. Briefly, the wells were coated overnight
at 4◦C with 20ng/well of purified EDIII, or an equal amount of
bovine serum albumin (BSA; Sigma–Aldrich), as a negative con-
trol, and diluted in 50mM carbonate saline (pH 9.6). After blocked
for 1h at 37◦C with PBS containing 3% BSA (PBSA), the wells were
washed 4 times with PBS containing 0.05% Tween-20 (PBST). Seri-
ally diluted MAbs in PBSA (100?l) were added to each well in
triplicate and incubated for 1h at 37◦C. After wells were washed
4 times with PBST, HRP-conjugated anti-mouse IgG (1:2000) was
added to each well and incubated at 37◦C for 40min, then washed
again. Antibody binding was visualized by addition of the mix-
ture of H2O2and 3,3?,5,5?-tetramethyl-benzidene substrate (TMB;
Sigma–Aldrich). After incubation for 15min at 37◦C, the reaction
was stopped by addition of 0.1M H2SO4, and absorbance was read
at 450nm with a reference wavelength of 595nm on a model Sun-
rise plate reader (Tecan). The endpoint titers of purified MAbs were
also determined by 10-fold serial dilution with indirect ELISA. In
all ELISAs, the irrelevant MAb H5 was used as an antibody control.
The positive cutoff ratio was set at 2 (ratio of OD value coated with
EDIII/OD value coated with BSA). This value is comparable to “pos-
itive to negative” cutoff ratios used in other WNV diagnostic assays
(Davidson et al., 2005; Estrada-Franco et al., 2003).
2.5. Cell surface staining detection by flow cytometry analysis
The percentage of 293T cells expressing WNV E protein was
determined by cell surface staining with MAbs. A FACSCalibur flow
cytometer (BD Biosciences) was used for flow cytometry analysis.
WNV E protein was expressed on the membrane of 293T cells by
transfection of pcDNA4-WNV E plasmids into 293T cells and cul-
turedfor48h.Single-cellsuspensionswerepreparedandincubated
with ascites (1:2000) at 4◦C for 1h in 100?l PBSA buffer, then
washed 3 times with PBS buffer. Cells were adsorbed with FITC-
conjugated anti-mouse IgG (1:500; Santa Cruz) at 4◦C for 1h, and

Page 3

Please cite this article in press as: Liu, J., et al., Characterization and application of monoclonal antibodies specific to West Nile virus envelope
protein. J. Virol. Methods (2008), doi:10.1016/j.jviromet.2008.09.019
ARTICLE IN PRESS
G Model
VIRMET-10677;No.of Pages7
J. Liu et al. / Journal of Virological Methods xxx (2008) xxx–xxx
3
washedagain.Fluorescentsignalsonthecellsurfaceweredetected
and the percentage of positive cells was counted among 3×104
cells.ControlsincludedcellswithoutadditionofprimaryMAb,cells
with H5 MAb and normal mouse IgG.
2.6. Affinity analysis by surface plasmon resonance (SPR)
The affinity between MAb and purified EDIII was determined by
SPR on a Biacore 3000 (Biacore, Inc). Firstly, EDIII was immobilized
on the surface of a CM5 chip by amine coupling and then used to
capture purified MAb. Analysis was performed at 25◦C at a con-
stant flow rate of 30?l/min, using HBS-EP buffer [10mM HEPES
(pH 7.4), 150mM NaCl, 3.4mM EDTA, 0.005% surfactant P20] as a
running buffer. To determine the association rate, dissociation rate
andaffinityconstant(KD),aconcentrationseriesfrom0.4to400nM
of purified MAb was injected (240?l, associated for 8min and
then dissociated over 10min). The EDIII surface was regenerated
by injection of 50mM NaOH before each EDIII injection. Binding
curves and kinetic parameters were analyzed with a global fit 1:1
bindingalgorithmwithdriftingbaselinebyBIAevaluationsoftware
version 3.2 (Biacore). The affinity constant KDwas determined as
koff/kon, using data from three independent experiments.
2.7. Immunofluorescence assay
Binding of mouse ascites MAbs with WNV- or JEV-infected
cells was determined by IFA. Sub-confluent BHK-21 cells, which
were grown in 24-well microplates with slides, were infected with
WNV (Chin-01 strain) or JEV (Beijing-1 strain) at a multiplicity
of infection of 0.1. After incubation for 3 days, serially diluted
MAbs were added to virus-infected BHK-21 cells. After incuba-
tion at room temperature for 2h, slides were washed 3 times with
PBST, and FITC-labeled anti-mouse IgG was added at dilution of
1:1000. Slides were washed again after 1h incubation, stained
with Evans blue, and observed under fluorescence microscope at
200× magnification. Cells showing strong green fluorescence were
recorded as positive. The highest dilution of mouse ascites MAb
that showed a strong positive fluorescence signal was recorded as
the IFA titer. The uninfected cells were used as a negative con-
trol at each dilution, and JEV-infected BHK-21 cells were used
to evaluate the cross-reactivity of MAbs with JEV. The experi-
ments that involved the use of WNV were performed in a BSL-3
laboratory.
2.8. Competitive-binding ELISA and AC-ELISA
The detector MAb was labeled with biotin using an EZ-link
Sulfo-NHS-LC-Biotinylationkit(PierceBiotechnology)accordingto
the manufacturer’s instructions. Experiments on epitope competi-
tion of the three purified MAbs (6A11, 4B3 and 2F5) were carried
out using competitive-binding ELISA. The wells were coated and
blocked as described in Section 2.4. After 100?l unlabeled MAbs
(5?g/ml) were added and incubated for 1h at 37◦C, wells were
washed 3 times, followed by incubation with an equal amount of
another biotin-labeled MAb for 1h at 37◦C. Plates were washed
again and incubated with HRP-conjugated streptavidin (diluted
1:2000 in PBS; Zhongshan Goldenbridge Biotechnology). After
washing, the color was developed with the addition of 100?l
freshly prepared substrate solution (1:1 mixture of TMB and H2O2
solution) for 15min at 37◦C. The color reaction was stopped by
100?l 0.1M H2SO4, and absorbance was read at 450nm, with a
reference wavelength of 595nm. Wells with addition of the irrele-
vantly unlabeled MAb (H5) were used as a negative control, and
wells with addition of unlabeled MAb (the same as the biotin-
labeled MAb) were a positive control. Each pair of MAbs was
assayed in triplicate. Results were expressed as a percentage of
inhibitionandderivedbythefollowingformula:percentageofinhi-
bition(PI)=[(negativecontrolOD−testMAbOD)/(negativecontrol
OD−positive control OD)]×100%.
FortheAC-ELISA,thepurifiedMAb(1?g/well),dilutedin50mM
carbonate saline (pH 9.6), was coated on wells overnight at 4◦C.
After blocking for 3h at 37◦C with PBS containing 5% non-fat dry
milk,wellswerewashed3timeswithPBST.Allthefollowingwash-
ing procedures were the same as described above. Virus culture
supernatant (103.5TCID50/ml) or recombinant EDIII protein, seri-
ally diluted in PBS containing 1% non-fat dry milk (PBSM) was
added to the wells (100?l/well) and incubated for 3h. Cell cul-
ture supernatant or BSA was used as a negative control. After
washing, 100?l per well biotin-labeled detector MAb (2?g/well,
diluted in PBSM) was added and incubated for 1h at 37◦C. After
washing, the wells were incubated for 30min at 37◦C with 100?l
per well HRP-conjugated streptavidin and detected as above. In
this ELISA test, the positive cutoff was also set at 2 (ratio of
positive/negative).
3. Results
3.1. Generation and purification of MAbs against WNV EDIII
protein
The positive-fused cell clones were screened using indirect
ELISA with recombinant EDIII as antigen. The hybridomas with
higher ELISA titers were selected for screening, and four MAbs
(6A11, 4B3, 2F5 and 6H7) were finally isolated and cloned. Ascites
was produced in BALB/c mouse by hybridomas. The heavy chain
subclassesofMAbsweredeterminedasIgG2a(6A11)andIgG1(4B3,
2F5and6H7),andthelightchainsofallofthesewerekappaisotype.
6A11wasefficientlypurifiedbyproteinAchromatography,and4B3,
2F5 and 6H7 by protein G chromatography. The concentrations of
purified MAbs were determined as 10–18mg/ml.
3.2. Western blot analysis
The binding specificity and cross-reactivity of the MAbs against
denatured EDIII protein and E protein were determined by West-
ern blot analysis. Four MAbs reacted with both EDIII and E proteins.
Two of the ascites MAbs (6A11 and 4B3) showed stronger reactiv-
ity with recombinant EDIII protein (12.5kDa) than the others (2F5
and 6H7), and the irrelevant MAb (H5) against influenza virus did
not bind to EDIII (Fig. 1A). WNV E protein, which was expressed
on the surface of 293T cells, as well as that from inactivated WNV,
showed specific binding to 2F5, and there was no cross-reactivity
with recombinant E protein or inactivated JEV (Fig. 1B). MAb 2F5
Fig. 1. Western blot analysis of anti-EDIII MAbs with denatured antigen. (A) Reac-
tivity of four MAbs with recombinant EDIII protein, using irrelevant MAb against
influenza virus (H5) as a negative control. (B) Reactivity of MAb 2F5 with E pro-
teins from WNV and JEV. Lanes 1–3: lysates of 293T cells transfected with pcDNA4,
pcDNA4-WNVEandpcDNA4-JEVEplasmids,respectively;lane4:inactivatedWNV;
lane 5: inactivated JEV; lane 6: cell culture supernatant of BHK-21 cells; lanes 7 and
8: the same as lanes 3 and 1, respectively. M: protein molecular weight markers. Left
panel (lanes 1–6) was detected with MAb 2F5 and right panel (lanes 7 and 8) was
detected with anti-His antibody.

Page 4

Please cite this article in press as: Liu, J., et al., Characterization and application of monoclonal antibodies specific to West Nile virus envelope
protein. J. Virol. Methods (2008), doi:10.1016/j.jviromet.2008.09.019
ARTICLE IN PRESS
G Model
VIRMET-10677;No.of Pages7
4
J. Liu et al. / Journal of Virological Methods xxx (2008) xxx–xxx
Fig. 2. Absorbance ratio of MAbs binding to EDIII and BSA. The broken line indicates
the absorbance ratio cutoff value, which was set at 2.
did not show any non-specific binding to negative controls, which
were 293T cells transfected with pcDNA4 vector and cultured
BHK-21 cells. The binding analysis with other MAbs (6A11, 4B3 and
6H7) yielded similar results (data not shown). The JEV E protein
was fused with a His tag at its C terminus, and expression of JEV
E on 293T cells was confirmed by anti-His antibody (Fig. 1B). This
indicated that the four MAbs recognized denatured WNV EDIII and
E protein with different binding affinity and did not show cross-
reactivity with E protein from JEV.
3.3. Reactivity of MAbs with WNV EDIII protein in indirect ELISA
To examine the reactivity of the MAbs with EDIII protein under
non-denaturing conditions, the indirect ELISA was performed with
foldedEDIIIprotein.ThetitersoffourunpurifiedascitesMAbswere
higher than 107in indirect ELISA (data not shown). The reactivity
of purified MAbs with EDIII is shown in Fig. 2. Three MAbs (6A11,
4B3 and 2F5) showed strong positive binding with EDIII at concen-
trations of 0.02–20?g/ml (absorbance ratio >10), compared with
BSA-coatedwells.Meanwhile,theyshowedobviouspositivesignals
at the concentration of 0.002?g/ml (absorbance ratio >2). Among
these, MAb 2F5 showed highest sensitivity. The low absorbance
ratios at the highest concentration tested (200?g/ml) were due to
the high non-specific binding of MAbs with BSA at this concen-
tration. The irrelevant MAb H5 did not show specific binding to
EDIII. It was surprising that MAb 6H7 showed no specific reactivity
with EDIII after purification, so it was excluded from the following
ELISAs.
3.4. Flow cytometry analysis of MAbs
Cell surface expression of WNV E protein was detected by MAb
staining and determined by flow cytometry. When stained with
different MAbs, the percentage of fluorescent cells varied greatly.
Representative profiles are shown in Fig. 3. The percentage of flu-
orescent positive cells was 4.5±2.6%, 26.9±8.6%, 44.6±8.0%, and
8.9±1.2%, when stained with MAb 6A11, 4B3, 2F5 and 6H7, respec-
tively.Controls,whichincludedcellswithoutMAb,andcellsstained
with the normal mouse serum or H5, were all negative. Among the
fourMAbsagainstEDIII,2F5showedthestrongestbindingtonative
E protein.
3.5. Binding affinity between purified MAbs and recombinant
EDIII protein
The binding affinity between recombinant EDIII protein and
purified MAbs (6A11, 4B3 and 2F5) was analyzed by SPR in the solid
phase. MAb 6H7 did not bind to immobilized EDIII protein under
experimental conditions, so the binding affinity between 6H7 and
EDIII was undetectable. MAb 2F5 bound to EDIII with an affin-
ity of 1.8±0.3nM, which was the highest among the three MAbs.
The affinity of 4B3 and 6A11 was similar, with a KDrange from
407.1±96.3 to 692.5±112.1nM (Fig. 4).
3.6. IFA
IFA was performed to further analyze whether the MAbs rec-
ognized the endogenously produced E protein in WNV-infected
BHK-21 cells. Both normal mouse serum and three MAbs did not
show non-specific binding to uninfected cells (data not shown).
6A11, 4B3 and 2F5 showed strong reactivity with WNV-infected
cells, whereas normal mouse serum did not bind to infected
Fig. 3. Flow cytometry analysis of cells that expressed WNV E proteins. The profile of transfected 293T cells without antibody staining was used to define the background
of fluorescent intensity (A–H, red). The white profiles in (B–H) were compared with reference to the red profile. (B) Stained directly with FITC-conjugated anti-mouse IgG,
withoutadditionofprimaryantibody;(C)stainedwithnormalmouseserum;(D–H)stainedwithH5,6A11,4B3,2F5and6H7,respectively.(Forinterpretationofthereferences
to color in this figure legend, the reader is referred to the web version of the article.)

Page 5

Please cite this article in press as: Liu, J., et al., Characterization and application of monoclonal antibodies specific to West Nile virus envelope
protein. J. Virol. Methods (2008), doi:10.1016/j.jviromet.2008.09.019
ARTICLE IN PRESS
G Model
VIRMET-10677;No.of Pages7
J. Liu et al. / Journal of Virological Methods xxx (2008) xxx–xxx
5
Fig. 4. Biacore binding curves of purified MAbs 4B3 with immobilized EDIII protein.
A concentration series from 0.4 to 400nM of purified 4B3 was injected (240?l,
associated for 8min and then dissociated over 10min). The affinity constant KDwas
determined as koff/kon.
Fig. 5. Detection of WNV-infected BHK-21 cells by immunofluorescence assay. The
slides were stained with Evans blue and then observed by fluorescence microscopy.
Scale bar: 20?m. (A) WNV-infected cells stained with normal mouse serum. (B)
JEV-infected cells stained with MAb 2F5. (C) WNV-infected cells stained with MAb
4B3. (D) WNV-infected cells stained with MAb 2F5.
cells (Fig. 5). The IFA titer of each MAb was determined, based
on the highest dilution of ascites that gave a strong signal on
WNV-infectedBHK-21cells.WhenBHK-21cellswereinfectedwith
WNV Chin-01 strain, the IFA titers of 6A11, 4B3, 2F5 and 6H7 were
2560, 7680, 5120 and 40, respectively. Notably, these MAbs showed
no cross-reactivity with cells infected with JEV (Fig. 5). Validity of
Table 1
Properties of MAbs against WNV EDIII protein in different assays.
AssayMAbs
6A11 4B32F56H7
Indirect ELISA
Western blot
Flow cytometry
Binding affinity
Immunofluorescence
Antigen capture-ELISA
++a
+++
+
++
+
+
+++
++++
++
++
+++
+++
++++
+
++++
++++
++
+++
–b
++
+
–
–
–
a+: Weak positive; ++++: strong positive.
b–: Not detectable.
the MAbs used in IFA was confirmed in an independent labora-
tory using BHK-21 cells infected with WNV strain NY99 (data not
shown).
3.7. Epitope competitions of purified MAbs and AC-ELISA
After biotinylation, the epitope competitions of three MAbs
(6A11, 4B3 and 2F5) were assayed by competitive-binding indirect
ELISA. The biotin-labeled 6A11 inhibited the binding of 4B3 to EDIII
(57.2±6.1%)andviceversa,whichindicatedthat6A11and4B3rec-
ognized overlapping epitopes. 2F5 showed no competitive binding
with 6A11 or 4B3, which meant that 2F5 recognized a different
epitope from 6A11 or 4B3.
In order to establish a sensitive AC-ELISA for WNV detec-
tion, each pair of the three MAbs was evaluated. The highest
sensitivity was obtained by using 2F5 as capture antibody, and
biotin-conjugated 4B3 as detector antibody. To determine the
detection limit of AC-ELISA, a serial dilution of EDIII protein and
WNV culture supernatant (103.5TCID50/ml) were used to construct
the binding curve (Fig. 6). Cell culture supernatant without WNV
infection was used as a negative control. According to the cutoff
threshold(2,whichistheratioofpositive/negative),itwasdeduced
that as little as 10ng/0.1ml of recombinant EDIII protein and 3.95
TCID50/0.1mlofvirusculturesupernatantcouldbedetected(Fig.6).
This result revealed that MAbs 4B3 and 2F5 could be used to detect
WNV in cell culture supernatant.
3.8. Properties of MAbs against WNV EDIII protein
Four MAbs against WNV EDIII protein were identified by isotyp-
ing, reactivity with denatured and native antigens, affinity assay
and epitope competition ELISA, and IFA, and the results of these
analyses were used to design the AC-ELISA. The properties of these
MAbs are summarized in Table 1.
Fig. 6. Sensitivity of AC-ELISA using MAbs. (A) Quantitative analysis using recombinant EDIII protein. 10ng/0.1ml was the detection limit. (B) Quantitative analysis using
WNV cell culture supernatant. 80-fold dilution of cell culture supernatant was the detection limit, which was 3.95 TCID50/0.1ml. The broken line indicates the absorbance
ratio cutoff value, which was set at 2.