Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
APPLIED MICROBIOLOGY, Feb. 1970, p. 224-231Copyright @ 1970 American Society for Microbiology
Vol. 19, No. 2Printed in U.S.A.
Assay of Chikungunya Virus in Cell Monolayersby Immunofluorescence
NICHOLAS HAHON AND WILLIAM A. HANKINS
Aerobiology & Evaluationi Laboratories, Fort Detrick, Frederick, Maryland 21701
Received for publication 5 November 1969
Chikungunya virus was quantitatively assayed by counting immunofluorescentfoci after infection of BHK21/C13 cell monolayers. The speed and efficiency ofvirus attachment to cells were markedly enhanced when augmented by centrifugalforce. By this procedure, a proportionality was obtained between the number ofimmunofluorescent foci and the volume of inoculum. Virus penetration into cellswas linear and complete within 15 min at 35 C. From observations on the sequentialdevelopment of viral antigen within cells and immunofluorescent focus counts, fociof infected cells may be enumerated as early as 16 hr after inoculation of cell mono-layers. A linear function was demonstrated between immunofluorescent focuscounts and relative virus concentration. The immunofluorescent assay was com-parable in sensitivity but more precise and rapid than virus assays based on theintracerebral inoculation of suckling mice or on plaque counting. By the immuno-fluorescent procedure, the 50% neutralizing end point of antiviral serum was rapidlyand quantitatively determined.
Quantitative assays of virus infectivity byenumeration of cells containing immunofluores-cent viral antigens have been established foragents representative of almost all major animalvirus groups and, recently, the procedure hasbeen successfully employed to assay a plant virus(2). However, immunofluorescent assays havebeen developed for only four viruses of the largenumber of agents that comprise the differentarbovirus groups, i.e., yellow fever (4), Vene-zuelan equine encephalomyelitis (VEE; 6), Sem-liki Forest (1), and Rift Valley fever (RVF)virus (5). The feasibility of extending the immu-nofluorescent technique for the infectivity assess-ment of another arbovirus, chikungunya, wasinvestigated because it has been demonstratedthat cells infected with the virus are amenableto immunofluorescent staining (7; J. D. Whiteand L. L. Kupferberg, Bacteriol. Proc., p. 128,1966). This is a report on the development andevaluation of this technique for the quantitativeassay of chikungunya virus.
MATERILLS AND METHODSVirus strain. The Banganike strain of chikungunya
virus (12) used in this study was in the form of a10% suckling mouse brain suspension. Assayed insuckling mice, the suspension had a titer of 108.6LDo per ml by intracerebral inoculation.
Cell lines and cultivation. The principal cell lineused for assay of the virus was baby hamster kidney
(BHK21/C13) obtained from the American TypeCulture Collection, Rockville, Md. This cell linemorphologically consists of elongated fibroblasticcells. Nutrient medium consisted of Earle minimalessential medium (MEM) supplemented with 1%glutamine (200 mM), 10% tryptose phosphate broth,10% fetal calf serum, 0.005% neomycin, and 75,ug of kanamycin per ml. Maintenance medium wasequal parts of nutrient medium and MEM. Cellswere cultivated on circular cover slips (15-mmdiameter) inserted in flat-bottomed glass vials (19by 65 mm). One milliliter of cell suspension con-taining from 105 to 3 X 105 cells was introduced ontocover slips which were then incubated at 35 C for24 hr or until a complete cell was formed.
Immunofluorescent virus assay. Determinationswere usually made in triplicate. Initially, virus dilu-tions were prepared in phosphate-buffered saline(PBS) solution (pH 7.1) free of calcium and mag-nesium ions (6), but later heart infusion broth wasused. Inoculum in a 0.2-ml volume was introduceddirectly onto cover slip cell monolayers after theirtransfer from glass vials to rotor chamber inserts (4).The latter were used because they withstood the highcentrifugal force required to sediment the virus.Rotor chamber inserts placed in a swinging-bucket(SW 25.1) rotor were centrifuged in a model L Pre-parative Ultracentrifuge (Beckman Instruments, Inc.,Spinco Div., Palo Alto, Calif.) at 19,642 to 29,432 Xg for 15 min at 30 C. The residual inoculum wasremoved after centrifugation, the cover slip cellmonolayers were replaced into glass vials, and 1 mlof maintenance medium was then added to each vial.After incubation at 35 C from 0.5 to 1 hr, the main-
224
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
ASSAY OF CHIKUNGUNYA VIRUS
tenance medium was replaced with 1 ml of a 1/15dilution of monkey viral antiserum in maintenancemedium. After further incubation at 35 C from 20to 22 hr, cover-slip cell monolayers were rinsed twicewith PBS solution, fixed with cold (-60 C) acetone,and either prepared immediately for immunofluo-rescent staining and focus counting or stored at -60C. Fluorescence of viral antigens in fixed cell cultureswas not diminished after storage for several weeks.
Chikungunya antiserum conjugate and immuno-fluorescence staining. Antiviral serum was obtainedfrom rhesus monkeys that had been injected intra-venously with 107 immunofluorescent focus units(IFU) of virus in a 1-ml volume. Animals wereinjected twice during a period of 1 month. Twoweeks after the last injection, the animals were bled.Antiserum was conjugated with fluorescein isothio-cyanate by the method of Riggs et al. (9). Conju-gated globulin was passed through a column ofSephadex G-50 (Pharmacia Fine Chemicals, Inc.,Piscataway, N.J.) to remove unbound dye. To reducenonspecific fluorescence, 5 ml of conjugated globulinwas diluted with an equal volume of PBS solutionand adsorbed twice with 200 mg of acetone-driedmouse liver powder by the procedure of Coons andKaplan (3).The direct fluorescent-antibody method was used
to demonstrate immunofluorescence of viral antigensin infected cells. Fixed cell monolayers were washedonce with PBS solution and stained with conjugatedglobulin for 30 min at room temperature. Cover-slipcell monolayers were then rinsed in two changes ofPBS solution and mounted in a semipermanentmedium (10).
Fluorescence microscopy and focus counting. Cover-slip cell cultures were examined with an AmericanOptical microscope equipped with a Fluorolume il-luminator (model 645), Corning no. 5849 and SchottBG-12 exciter filters, and an E. K. no. 2A barrierfilter. With this optical system at a magnification of430X, the number of microscopic fields contained inthe area of a 15-mm cover slip was 1,064. On occasionwhen cell monolayers contained numerous immuno-fluorescent foci, 200X magnification was employed;the number of fields per cover slip was 226. For eachcover slip cell monolayer, 50 microscopic fields wereexamined for immunofluorescent foci. To calculatethe number of IFU of virus per milliliter, the averagenumber of IFU per field was multiplied by the numberof fields per cover slip, the reciprocal of the dilution ofvirus inoculum, and a volume factor (for conversionto milliliters).
Suckling mouse virus assay. One suckling mouselitter was inoculated for each virus dilution; eachlitter consisted of eight 1- to 2-day-old mice. Appro-priate virus dilutions were inoculated intracerebrallyin 0.02-ml volumes into each mouse. Deaths occurringwithin 24 hr after inoculation were considered non-specific. Mice were observed for lethality daily for 10days. Virus titers were calculated by the Reed andMuench (8) formula and expressed as suckling mouseintracerebral median lethal dose (SMICLD60).
Plaque virus assay. Monolayers of guinea pig lung,BHK21/C13, or chick embryo fibroblast cells grown
in 25-cm2 flasks (Falcon Plastics, Los Angeles, Calif.)were inoculated with 0.1 ml of appropriate virusdilutions and then incubated at 37 C for 1 hr. Fivemilliliters of overlay medium was added to eachflask. This medium consisted of basal medium Eagle(Earle base), 2% fetal calf serum, 2% Noble agar(Difco), 100 pg of diethylaminoethyl (DEAE)-dextran (Pharmacia Fine Chemicals, Inc., Piscataway,N.J.), 100 units of streptomycin, and 100 pg of peni-cillin per ml. Guinea pig lung cell monolayers wereincubated at 33 C for 72 hr; BHK21/C13 and chickembryo fibroblast cells were incubated at 33 C for 48hr. The second overlay was then applied; it wassimilar to the first except that it contained 0.01%neutral red but no DEAE-dextran. After additionatincubation of cell monolayers at 33 C from 18 to 24hr, plaques were counted.
Determination of virus attachment. Virus attach-ment was measured by observing and recording thedisappearance of virus from inoculum after its ad-dition to cell monolayers. Cover-slip cell cultures tobe subjected to centrifugation or stationary incubationat 35 C received 0.2 ml and 0.4 ml of inoculum,respectively. After designated incubation intervals,residual inoculum was removed, and the cell mono-layers were immediately washed twice with PBSsolution. Residual inoculum was then introducedonto fresh cell monolayers to measure unattachedvirus. For this, the residual inoculum was attachedonto cell cultures by centrifugation at 19,642 to 29,432X g for 15 min. Cover-slip cell monolayers exposedto initial or residual inocula were then treated in themanner described earlier for assay of virus. Theamount of virus that was attached to cells at agiven time was expressed as a percentage of the virusinput. The latter was the sum of the amounts of at-tached and residual virus.
Detemination of virus penetration. Virus penetra-tion into cells was measured by the insensitivity ofattached virus to antiserum. Cell cultures were washedtwice with PBS solution, overlaid at designated timeswith 1 ml of a 1/15 dilution of viral antiserum, andthen incubated at 35 C for 20 hr. The quantity ofvirus that penetrated into cells at a given time wasexpressed as a percentage of the input virus.
Calculation of attachment and penetration constants.The attachment and penetration rate constants (K)were calculated from the relationship, 2.3 log(Vo/Vt)/nt, where Vo = the input virus concen-tration, V: = unattached or unpenetrated virusat time t, and n = the number of cells per cubiccentimeter determined by resuspension of cover-slipcell cultures after trypsin treatment and enumerationof cells in a hemocytometer.
RESULTSVirus attachment. The rate of chikungunya
virus attachment to BHK21 /C13 cell monolayerswas determined for both centrifugation andstationary incubation (35 C). Virus inputs forthe former and the latter were 740 and 1,480IFU, respectively. The concentration of cells percover-slip culture was 4.0 X 105 per ml. Aided
225VOL. 19, 1970
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
HAHON AND HANKINS
by centrifugal force, approximately 95% of thevirus inoculum was attached within 10 min; afterstationary incubation for 2 hr, only 14% of viruswas attached (Fig. 1). The attachment rate con-
;
c
0.
u
-
0
5
010
Centrifugation, minutes
5 10 15
Centrifugation
Stationary Incubation,
Stationary Incubation, minutes
FIG. 1. Attachrnent of chikungunya virus onto cover-
slip BHK2J/C13 cell monolayers by centrifugation(19,642 to 29,432 X g at 25 C) and stationary incuba-tion (35 C).
TABLE 1. Relationship between volume of inoculumand immunofluorescent focus units (IFU) ofchikungunya virus: centrifugation versus
stationary incubation
IFU per 50 microscopic fields
VolCentrifugationa Stationary
ml
0.1 33 11
0.2 60 230.5 159 191.0 332 29
a Virus inoculum centrifuged at 19,642 to
29,432 X g for 15 min into BHK21/C13 cell mono-layers at 25 C.
bStationary incubation of virus inoculum at
35 C for 2 hr.
TABLE 2. Effect of different virus diluents onattachment of chikungunya virus onto
BHK2I/C13 cell monolayers
IFU4 per 50Virus diluent microscopic Virus titerfieldspl
IFU/mlPBSb...................... 43 4.2 X 108PBS + 0.0009 M CaCl2 +0.0005 M MgC2.......... 42 4.2 X 108
Cell maintenance mediumc. 42 4.2 X 108Heart infusion broth. 56 5.8 X 10'
a Immunofluorescent focus units of virus.bNaCl (0.1 M) buffered by 0.01 M phosphate
buffer.cMinimal essential medium, 5% fetal calf
serum, and 5% Tryptose phosphate broth.
stants (K) were 3.4 X 1O-7 cm'/min with cen-trifugation and 2.8 X 109 cm3/min with station-ary incubation. The former attachment rate con-stant is comparable to those reported for VEEand RVF viruses under similar experimentalconditions (5, 6).The efficiency of centrifugation and stationary
incubation for infecting cell monolayers fromdifferent volumes of inoculum is shown in Table1. A proportionality between the number ofimmunofluorescent foci and volume of inoculumwas obtained when centrifugation was employed.With stationary incubation, a similar relationshipwas not evident.
Because previous studies with another groupA arbovirus (VEE) indicated that the efficiencyof virus attachment to cells was markedly influ-enced by the menstruum used to suspend thevirus (6), this factor was also investigated forattachment of chikungunya virus. Virus wasdiluted in each of four different suspendingmedia, pH 7.1 to 7.3. Before virus inoculum wasadded, each cell monolayer was washed with itsrespective test diluent. Virus inoculum was cen-trifuged onto cell cultures; they were then treatedaccording to the described assay procedure. Thecompositon of the medium used for attachmentof chikungunya virus to cell monolayers did notsignificantly affect assay values (Table 2).Slightly more virus appeared to be attached, how-ever, in the presence of heart infusion broth.
Virus penetration. The rate of virus penetrationinto cells was followed by determining the insen-sitivity of attached virus to antiviral serum atdesignated times. A virus input of 168 IFU wasintroduced onto cell monolayers. They were thentreated as described. Virus penetration at 35 Cproceeded at a linear rate without evidence ofan initial lag period (Fig. 2). Approximately 50%
226 APPL. MICROBIOL.
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
ASSAY OF CHIKUNGUNYA VIRUS
of attached virus penetrated into cells within 7.5min; the process was complete within 15 min.The penetration rate constant (K) of 2.3 X107 cm8/min was faster than that noted withVEE and RVF viruses (5, 6).
Because an antiviral serum overlay is employedin this assay to prevent the extracellular spreadof virus from infected cells, this information onthe rate of virus penetration is highly relevant.The earliest time that the antiviral serum overlaymay be added without neutralizing attachedvirus was fixed by the results obtained from the
100 -* . _
90
80
70
X 60
> 50
40_
Minutes
FIG. 2. Rate of chikungunya virus penetration intoBHK21/C13 cells at 35 C as measured by insensitivityof attached virus to antiviral serum.
TABLE 3. Assay of chikungunya virus in differentcell lines
Cell line Virus titer
mlBHK21/C13 ............ 3.1 X 10C IFUaBHK21 ........... 2.1 X 107 IFUL-929.................. 3.3 X 107 CIUbGuinea pig lung ...... 1.8 X 107 CIUMcCoy................. 5.8 X 104 CIUKB (Eagle) ......... 1.0 X 104 CIUHEp-2.................. 2.1 X 103 CIU
a Immunofluorescent focus units of virus.b Cell-infecting units of virus.
preceding experiment. For routine assay of virus,the antiviral serum overlay was added after cellmonolayers were incubated at 35 C for 0.5 to 1hr after virus attachment.
Susceptibility of cell lines. Chikungunya viruswas assayed by the immunofluorescence proce-dure in different cell lines to determine theirsusceptibility to infection. The cell lines testedwere BHK21/C13, BHK21 (short fibroblasticcells obtained from Microbiological Associates,Bethesda, Md.), KB (Eagle), HEp-2, L-929,guinea pig lung, and McCoy. Virus was inocu-lated onto cell monolayers in the usual mannerand cell cultures were then incubated in thepresence of antiviral serum overlay. Virus as-sayed in the BHK21/C13 cell line was approxi-mately 1.0 log10 unit higher in titer than thatobtained in the BHK21 cell line (Table 3). Inboth cell lines, immunofluorescent foci appearedin the presence of antiviral serum; focus countswere the basis of assay values. In the other celllines, the assay was based on individual infectedcell counts, because immunofluorescent foci didnot appear in the presence of the antiviral serumoverlay. The BHK21/C13 cell line was the mostsusceptible to virus infection of the cell linestested and was subsequently employed for thestandardization of the assay.
Incubation period. Incubation period is definedas the time interval between virus inoculationand the development of recognizable quantitiesof fluorescent viral antigens in infected cells. Theperiod was established from sequential observa-tions of infected cell monolayers and immuno-fluorescent focus counts. Two groups of BHK21 /C13 cover-slip cell monolayers were inoculatedwith 2 X 103 IFU in 0.2-ml volume and treatedin the manner described for attachment ofinoculum. After the prescribed period for viruspenetration into cells, antiviral serum overlaywas added to one group of cell monolayers; theother group received maintenance medium. Thelatter group was included to determine whetherthere was an optimal time period for countingindividual infected cells before the appearance offoci. Both test groups were incubated at 35 C fortime intervals of 4, 6, 8, 10, 12, 16, 21, and 25hr. Representative cell monolayers from eachgroup were then processed for fluorescent-anti-body staining.The earliest sign of specific fluorescence of
chikungunya virus antigens in cells was seen at 6hr after infection. Fluorescence was localizedexclusively in the cell cytoplasm and was con-fined to individual cells. At 8 hr, an occasionalimmunofluorescent focus that consisted of threeor more infected cells was noted. Generally, thenumber of infected cells in a focus increased to
227VOL. 19, 1970
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
HAHON AND HANKINS
10 or more with prolonged incubation times(Fig. 3). Individual infected cells were mostlyseen in cell monolayers incubated with main-tenance medium; immunofluorescent foci werepredominate in cell cultures incubated withantiviral serum. The increase in numbers ofimmunofluorescent foci and individual infectedcells during incubation of the two test groups ofcell monolayers is shown in Fig. 4. In the pres-ence of antiviral serum, immunofluorescent foci
FIG. 3. Immunofluorescent focus of chikungunyavirus infection in BHK21/C13 cells in the presence ofantiviral serum. X 225.
.LI
r2u
-!wv
.2S
-0
2Ez
120a0
Hours
FIG. 4. Effect of the length of the incubation period(35 C) on the number of infected BHK21/C13 cellsin the presence and absence of antiviral serum.
increased during the first 12 hr of incubation,and thereafter the number remained constant.Individual infected cell counts in cell monolayersincubated with maintenance medium increasedalmost linearly throughout the 25-hr incubationperiod, which made it difficult to delineate theprimary from secondary cycle of infection. Inview of these findings, the assay of chikungunyavirus was based on immunofluorescent focuscounting. Foci could be enumerated as early as16 hr after inoculation of cell monolayers toobtain an estimate of virus titer.
Quantitative evaluations of the assay. A linearrelationship was obtained between the numberof immunofluorescent focus units and relativevirus concentration throughout the inoculumrange of 1.2 log1o units (Fig. 5). Each immuno-fluorescent focus appeared to result from infec-tion by a single infective virus particle or aggre-gate not divisible by dilution.Ten determinations were performed in a
single experiment to estimate the precision of theassay. Cover-slip cell monolayers were infectedwith 0.2 ml of a 10-5 virus dilution and thentreated in the prescribed manner. The number ofIFU per milliliter of virus inoculum ranged from5.2 X 108 to 6.3 X 108 with a mean of 5.8 X
36 -
323
2 8
24
20
16
8 2
8
z
Concentration Of Virus
FIG. 5. Linear function between the number ofimmunofluorescent focus units and relative concentra-tion of chikungunya virus.
228 APPL. MICROBIOL..
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
ASSAY OF CHIKUNGUNYA VIRUS
TABLE 4. Assay of chikungunya virus by different methods
Assay no. IFUa per ml SmCLDw, per mlb | BHK2l/C13 cells GPL cells(PFU/ml)c (PFU/milc)
1 5.3 X 108 9.5 X 10 7.6 X 108 3.6 X 1082 5.8 X 108 10.0 X 108 7.0 X 108 3.2 X 1083 5.3 X 108 12.0 X 108 7.6 X 108 3.6 X 1084 5.1 X 108 0.63 X 108 8.8 X 108 2.0 X 1085 5.1 X 108 10.0 X 108 10.0 x 108 3.6 X 1086 4.8 X 108 10.0 X 108 7.8 X 106 4.7 X 108
Mean 5.2 X 108 8.6 X 108 8.1 X 108 3.4 X 108SDd 4±0.34 X 10' ±44.0 X 10' :1:1.1 X 10J 4±0.87 X 10'SEe 40.13 X 108 ±1.63 X 108 ±0.45 X 10 ±0. 35 X 10'
a Immunofluorescent focus units of virus determined in 20 hr.b Suckling mouse intracerebral median lethal units of virus determined in 10 days.c Plaque-forming units (PFU) of virus in BHK21/C13 cell monolayers determined
guinea pig lung (GPL) cell monolayers at 4 days.d Standard deviation.a Standard error of mean.
108, standard deviation of ±0.17, and coefficientof variation of 2.9%.To determine the reproducibility of the im-
munofluorescent focus-counting assay, a vial ofvirus was randomly selected at approximately2-week intervals from the standard pool storedat -60 C and assayed. In four such trials, assayvalues ranged from 5.2 X 108 to 6.3 X 108 IFUper ml with a mean of 5.8 X 108.The sensitivity of the immunofluorescent
focus-counting assay of chikungunya virus wascompared with that of the methods of intra-cerebral inoculation of suckling mice and virusplaque counting. Chick embryo fibroblasts,guinea pig lung, and BHK21/C13 cell mono-layers were each employed for virus assay by theplaque procedure. Six determinations were madewith each assay; virus from a standard pool wasused. Virus assays in either suckling mice or bycounting virus-induced plaques in BHK21/C13cell monolayers appeared to be comparable insensitivity (Table 4). The mean assay valueobtained by immunofluorescent focus countingwas only 0.2 logio unit less than that obtained bythe former procedures. The mean assay valuederived from plaque counts using guinea pig lungcell monolayers was 0.2 log1l unit less than thatobtained by the immunofluorescent focus-count-ing method. In chick embryo fibroblast cellmonolayers, plaques were not formed with thevirus dilutions inoculated. The immunofluores-cent focus counting assay, however, exhibitedless variability than that of the other virus assaymethods tested and, in addition, had the singularadvantage of rapidity (less than 24 hr).Serum neutralization test. A preliminary test
was made to determine whether the immuno-fluorescent focus counting assay of chikungunya
at 3 days, in
Percentage Virus Neutralized
10
-510c
E2
.-
Ana
-410
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
Probits
FIG. 6. Estimation of chikungunya serum-neutralizingantibodies by the immunofluorescent virus assay pro-cedure. Fifty per cent serum neutralizing end point wasinterpolated from the percentage reduction of immuno-fluorescent foci of infected cells by antiviral serumdilutions.
virus could be adapted for estimating serum neu-tralizing antibodies against the virus. Appropri-ate dilutions of monkey antiviral or normalserum (control) were mixed with equal volumes
VOL. 19, 1970 229
1-3L
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
HAHON AND HANKINS
of a constant quantity of virus (5.0 X 104 IFUper ml). Virus and serum dilutions were pre-pared in PBS. After test mixtures were incubatedat 35 C for 2 hr, 0.2 ml of each mixture wasintroduced onto one of three cover-slip BHK21/C13 cell monolayers for assay of residual virusinfectivity in the manner described previously.To determine the 50% serum-neutralizing endpoint, the percentage of reduction of immuno-fluorescent focus counts for each antiviral serumdilution was computed from the control counts.Reduction percentages were then plotted againstthe logarithm of the corresponding final dilutionsof antiviral serum on probability paper. Results(Fig. 6) show that a linear relationship wasobtained over a critical range. By interpolation,the dilution of antiviral serum that neutralized50% of virus was determined.
DISCUSSION
The feasibility of employing the immunofluo-rescence procedure for the quantitative assess-ment of chikungunya virus infectivity was estab-lished by the studies described here. The assaywas highly precise, reproducible, and comparablein sensitivity to other established methods ofchikungunya virus assessment. The outstandingfeature of the method, in common with immuno-fluorescent assays described for other arboviruses(1, 4-6), was the ability to assess virus infectivitywithin 24 hr after inoculation of cell monolayers.The time required to assay chikungunya virus byintracerebral injection of suckling mice or plaquecounting was 10 days and from 3 to 4 days,respectively.The use of centrifugal force for attaching
virus onto cell monolayers offers several advan-tages over stationary incubation. By the formerprocedure, virus attachment is efficient andrapid, thermal inactivation of virus inoculum isminimized, and the proportionality demon-strated between immunofluorescent focus countsand volume of inoculum facilitates the detectionof virus particles from dilute suspensions. Withthe use of centrifugal force, almost synchronousinfection of cells may be achieved, and thestages of virus attachment onto and penetrationinto cells may be readily delineated. These arehighly desirable advantages in studies concernedwith the early stages of virus replication.For attachment of chikungunya virus to cells,
the requirements were less demanding thanthose for VEE virus, another group A arbovirus.Maximal attachment of chikungunya virusoccurred in the presence of menstruums ofvaried composition. With VEE virus, however,
maximal attachment to cells occurred only in thepresence of PBS free of calcium and magnesiumions. In contrast to the findings with chikungunyavirus, heart infusion broth markedly inhibitedVEE virus attachment (6). This indicates thatthere is an inherent difference in the require-ments of these two group A arboviruses foreffecting maximal attachment to cells.An attempt to base the assay of chikungunya
virus on counts of individual infected cells washampered by the appearance of foci of infectedcells in BHK21 /C13 cell monolayers. These focioccurred even when inoculated cell monolayerswere incubated in the presence of a potent anti-viral serum overlay. When different establishedcell lines were inoculated and incubated withantiviral serum, virus infection was confined toindividual cells. In earlier studies, direct evidencewas obtained to show that chikungunya virusinfection of BHK21 /C13 cell monolayers occursby cell to cell transmission in the presence ofantiviral serum overlay and, also, extracellularlyin the absence of the overlay (Hahon and Zim-merman, in preparation). Under similar circum-stances, foci of VEE virus-infected cells inBHK21/C13 cell cultures also occurred in thepresence of antiviral serum, but infection waslimited to individual cells when different celllines were used. These observations suggest thatBHK21 cell lines possess some unique biologicalmembrane structure or physiology that is con-ducive for cell to cell transmission of virus par-ticles. It may be of some relevance that the cellline has been extensively employed for neoplastictransformation by viruses (11). Because BHK21 /C13 cells were the most susceptible of the celllines tested to chikungunya virus infection, it wasthe host cell line selected for assay of the virus.The assay was based, therefore, on the appear-ance of immunofluorescent foci that could beenumerated as early as 16 hr after inoculation ofcell monolayers.
In a preliminary test, the 50% neutralizing endpoint of antiviral serum was determined byreacting chikungunya virus with antiserum. Overa critical range, a linear relationship was ob-tained between reduction percentages of immu-nofluorescent focus counts and dilutions of anti-viral serum. This finding augurs well for thefeasibility of estimating, rapidly and quantita-tively, chikungunya serum neutralizing anti-bodies by the immunofluorescence virus assayprocedure.
ACKNOWLEDGMENT
The excellent technical assistance of W. Douglas Zimmernanand James Booth is gratefully acknowledged.
230 APPL. MICROBIOL.
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.
ASSAY OF CHIKUNGUNYA VIRUS
LrITRATURE CIED
1. Carter, G. G. 1969. The rapid titration of Semliki Forestvirus in cell monolayers by immunofluorescence. J. Gen.Virol. 4:139-143.
2. Chiu, R. J., and L. M. Black. 1969. Assay of wound tumorvirus by the fluorescent cell counting technique. Virology37:667-677.
3. Coons, A. H., and M. H. Kaplan. 1950. Localization ofantigen in tissue cells. II. Improvements in a method forthe detection of antigen by means of fluorescent antibody.J. Exp. Med. 91:1-13.
4. Hahon, N. 1966. Fluorescent cell-counting assay of yellowfever virus. J. Infect. Dis. 116:33-40.
5. Hahon, N. 1969. Immunoflourescent cell-counting assay ofRift Valley fever virus. Amer. J. Vet. Res. 30:1007-1014.
6. Hahon, N., and K. 0. Cooke. 1967. Primary virus-cellinteractions in the immunofluorescence assay of Venezuelanequine encephalomyelitis virus. J. Virol. 1:317-326.
7. Mutani, M., A. Igarashi, P. Tuchinda, and S. Kato. 1967.Cytoplasmic RNA synthesis and viral antigen in FL cellsinfected with chikungunya virus. Biken J. 10:203-218.
8. Reed, L. J., and H. Muench. 1938. Simple method for esti-mating 50 per cent end points. Amer. J. Hyg. 27:493-497.
9. Riggs, J. L., R. J. Seiwald, J. H. Burckhalter, C. M. Downs,and T. G. Metcalf. 1958. Isothiocyanate compounds as
fluorescent labeling agents for immune serum. Amer. J.Pathol. 34:1081-1097.
10. Rodriguez, J., and F. Deinhardt. 1960. Preparation of a
semi-permanent mounting medium for fluorescent anti-body studies. Virology 12:316-317.
11. Stoker, M., and L. Macpherson. 1964. Syrian hamster fibro-blast cell line BHK21 and its derivatives. Nature (London)203:1355-1357.
12. Weinbren, M. P., A. J. Haddow, and N. C. Williams. 1958.The occurrence of chikungunya virus in Uganda. Trans.Roy. Soc. Trop. Med. Hyg. 52 253-262.
VOL. 19, 1970 231
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/a
m o
n 13
Jan
uary
202
2 by
14.
39.1
30.1
41.