Journal of Virological Methods, 39 (1992) 269-278 0 1992 Elsevier Science Publishers B.V. / All rights reserved / 0166-0934/92/%05.00
269
VIRMET 01385
Comparison of eight different procedures for harvesting avian reoviruses grown in Vero cells Yatri Drastini, Frederick S.B. Kibenge, Patricia K. McKenna and Alfonso Lopez Atlantic Veterinary College, University of Prince Edward Island, Charlottetown. Prince Edward Island (Canada) (Accepted 29 April 1992)
Summary
14 avian reovirus isolates adapted to replicate in an African green monkey (Vera) cell line were studied for the nature of their replication. The growth curves of 5 viruses showed them to be highly cell-associated in Vero cells. Different procedures were examined for releasing the cell-associated virus following propagation in Vero cells, including several freeze-thaw cycles, treatment with sterile distilled deionized water (ddHzO), freon extraction, and trypsin treatment. Treatment of virus infected cultures with ddHz0 was the most effective, and trypsin treatment was the least effective procedure for dissociation of virus from cells. Treatment of virus infected cultures with ddH20 is a simple and effective procedure which can be used where large amounts of virus are required for experimental purposes. Avian reovirus; Vero cell; Virus replication; Cell-associated virus; Virus harvest Reoviruses belong to the family Reoviridae which has an extremely wide host range among animals and plants (Samuel, 1988). Members of the Reoviridae family found in vertebrates form five genera: Orthoreovirus, Rotavirus, Orbivirus, Coltivirus, and Aquareovirus (Francki et al., 1991). The members of the Urthoreovirus genus are distinguishable into two distinct groups: those of avian origin, and those of mammalian origin (Kawamura and Tsubahara, 1966). Both groups of viruses have a genome consisting of 10 segments of double-stranded RNA (dsRNA), separable into three size classes, which are enclosed within a double protein capsid shell 70 to 80 nm in diameter Correspondence IO: F.S.B. Kibenge, Department of Pathology and Microbiology, Atlantic Veterinary College, UPEI, 550 University Avenue, Charlottetown, PEI, CIA 4P3, Canada.
270
with a similar protein composition and distribution (Joklik, 1983; Schnitzer et al., 1982; Spandidos and Graham, 1976). However, avian reoviruses, unlike their mammalian counterparts, exhibit a wide heterogeneity in their neutralizing antigens (Robertson and Wilcox, 1986), possess a different group-specific antigen (Petek et al., 1967), cause no haemagglutination (Glass et al., 1973), are associated with disease in their natural hosts (Kibenge and Wilcox, 1983), and cause syncytium formation both in infected cell cultures and in susceptible chickens (Kibenge and Wilcox, 1983; Wilcox and Compans, 1982). Avian reoviruses can be readily grown in the laboratory in primary cultures of cells of avian origin (Barta et al., 1984; Guneratne et al., 1982). However, among the established mammalian cell lines tested, only Vero cells have been reported to support replication of certain strains of avian reovirus (Barta et al., 1984; Wilcox et al., 1985; Jones and Alfaleq, 1990; Nwajei et al., 1988). The avian reoviruses in the present study were adapted to Vero cells which, in addition to being readily available, allowed for consistent titration of infectious virus because of the uniform nature of cells of the continuous cell line. However, the only report on the nature of avian reovirus replication in Vero cells found the virus to be highly cell-associated, requiring several freeze-thaw cycles to release virus from infected cells (Wilcox et al., 1985). Since that study examined only one avian reovirus, strain RAM-l, because other strains could not replicate in Vero cells, it was not clear whether the cell-associated character of RAM-l in Vero cells was strain specific. This paper reports the growth curves of an additional live avian reoviruses in Vero cells. Furthermore, several procedures evaluated for recovery of virus grown in Vero cells are described. Virus samples, including 4 reference strains (avian reovirus Type 59, Type 24, WVU2937, and Crawley), and 10 local isolates (SK suffix) from the Health of Animals Laboratory, Sackville, New Brunswick, Canada were used. All samples were received as chorioalantoic membrane material except for three samples (SK125, SK73a, and SK25a) which were chick kidney cell culture supernatants. All virus samples were passaged once in primary chicken embryo liver (CELi) cell monolayers on 24-multiwell tissue culture plates (Jones and Kibenge, 1984). Cytopathic effect (CPE)-positive cultures were then serially passaged at 7-day intervals in Vero cells (Kibenge et al., 1988) 2 to 6 times prior to preparation of stock cultures. For viral growth curves, 25-cm* tissue culture flasks containing monolayers of Vero cells were inoculated with lo5 TCIDso of virus in 0.1 ml of medium per flask, considered sufficient to produce a ‘onestep’ growth curve. At intervals after infection (relative to the end of the adsorption period), two flasks were removed and harvested essentially as previously described (Kibenge et al., 1988). Three fractions, namely extracellular virus harvest, cell-associated virus harvest, and total virus harvest were stored at -70°C and later assayed for virus infectivity. The assay for the growth curves of one selected virus was repeated three times. For harvesting of virus, infected Vero cell monolayers in 25-cm* tissue culture flasks with approximately 75% CPE were used. The medium of one flask was
271
discarded, and was replaced with 2 ml of sterile distilled deionized water (ddH*O). The flask was stored at -70°C and was later assayed for virus infectivity. A second flask was frozen and thawed three times, the cell lysate was extracted once with 1,1,2-trichlorotrifluoroethane (Freon) (J.T. Baker Chemical Co., Phillipsburg, NJ, USA), and the aqueous phase was stored at -70°C until when assayed for virus infectivity. The medium of a third flask was discarded, and the infected cell monolayer was treated with 0.25% trypsin at 37°C for 10-15 min. Following trypsinization, 2 ml of growth medium were added and the flask was stored at - 70°C and later assayed for virus infectivity. The fourth, fifth, sixth, seventh and eighth flasks were frozen and thawed 3, 5, 7, 10 and 15 times, respectively, and following low speed centrifugation to remove cellular debris, the supernatants were titrated to determine virus infectivity. Flasks treated with sterile ddHz0 or with trypsin were also frozen and thawed three times prior to titration of the viral supernatants. Each assay was repeated three times. Titration of virus was done in quadruplicates in 96well microtiter tissue culture plates. Plates were incubated at 37°C in a humidified 5% CO2 atmosphere for 7 days and were then examined microscopically prior to fixation and staining with 1% crystal violet in 70% ethanol. The virus titer was determined from the CPE using the procedure described by Reed and Muench (1938). All 14 avian reoviruses isolated in CELi cells induced serially sustained CPE during successive passages in Vero cells indicating that they were adapted to grow in the Vero cell line. Hussian et al. (198 1) previously reported successful adaptation of three Australian isolates examined although Wilcox et al. (1985) could only adapt one strain, designated RAM-l, of six other Australian avian reovirus strains to Vero cells. The RAM1 strain had been passaged 4 to 6 more times in chicken cells than the unadaptable strains. In the present study, prior to the single passage in CELi cell cultures, most isolates had only been propagated in chicken embryonated eggs 1 to 2 times, suggesting that cell passage history of the reovirus strain is probably not a major factor in adaptation to Vero cells. It is probable that variation either in culture conditions or between batches of Vero cell line (Jones and Alfaleq, 1990), and individual virus strain variation may he responsible for the different success rates of adaptation of avian reoviruses to Vero cells. It has been recently demonstrated that avian reovirus strain S1133 can replicate in Mouse L cells at pH 6.4 and pH 7.2 but not pH 8.2, suggesting that a basic pH in the culture medium is responsible for the inhibition in viral replication (Mall0 et al., 1991a). These authors also showed that the presence of Actinomycin D during infection increased viral replication in L cells by eliminating host mRNA competition for the translation machinery (Mall0 et al., 1991b). This suggests that virus replication in L cells is also dependent on the multiplicity of virus infection. The same may be true for avian reovirus replication in Vero cells. The growth curves of 5 avian reovirus strains (Crawley, SK125, SK98a, SK103a, and SK73a) were determined to further characterize the nature of avian reovirus replication in Vero cells. In all five virus strains the growth curve
io
20
I
50
A
I
100
~
100
TOTAL VIRUS
CELL&sOClAlEO
I
Fig. 1. a and b. (see page 273 for legend.)
0
4
/R1/---
1
0
TIME (HOURS)
VIRUS
nME (HOURS)
I~--
GROWTH CURVE OF REOVIRUS STRAIN CRAWLEY IN VERO CELLS
3
5
0
0
0
20
20
20
/‘
b
/
n--
50
00
50
nME (HOURS)
100
TOTAL VIRUS
100
nm
(HOURS)
TINE (HOURS)
CELL-ASSOC~A~EO VIRUS
100
EXTRACELLULAR VIRUS
GROWTH CURVE OF AVIAN REOVIRUS STRAIN SK125 IN VERO CELLS
273
of cell-associated virus resembled that of total virus (Fig. 1) indicating that these viruses were highly cell-associated. This may be true for all avian reoviruses adapted to replicate in Vero cells. The titers of total and cellassociated virus peaked at 30 to 96 h PI; whereas the maximum titers of extracellular virus were steadily rising throughout the 144 h observation period. However, this cell-associated character of avian reovirus replication in Vero cells could not be confirmed by statistical analysis of the growth curves. For this analysis, the amount of virus, A, at time t was modelled using an adaptation of the Gompertz model for growth curves (Draper et al., 1981). A= aexp{-ble-k’f-_bze-k2’}. GROWTH CURVE OF AVIhN REOVIRUS STRAIN %W#
IN VERO w
C
9
BTRACEUUIAR
VIRUS
lI
0
I
I
I
I
I
100
So
20
I
TIME (HOURS)
.f
-
0
I
I
I
60
20
I
100
I
I
TIME (HOURS)
TOTAL VIRUS 9
I 0
1
20
I
sb
I
l&
‘iTME (HbURS)
Fig. 1. (c above.) Growth curves of avian reovirus strains Crawley (a), SK125 (b), SK98a (c), SK103a (d), and SK73a (e). Vero cell monolayers were infected with lo5 TCIDH, of virus in 0.1 ml of medium per 25cm’ tissue culture flask. After a l-h adsorption period, cells were washed twice with phosphate buffered saline, and following the addition of fresh growth medium, they were maintained at 37°C in a humidified 5% CO2 atmosphere. Infected cells were harvested at the indicated time intervals, and viral progeny from the infected cells was titrated by TCIDs,,.
274
,.
II \ I ii I
5 VI
CI
f
275
GROWTH CURVE OF AVIAN REOVIRUS
STRAIN SK725 IN VERO CELLS
Fig. 2. Growth curve of avian reovirus strain SK73a in Vero cells, analysed using the Gompertz model for growth curves (Draper et al., 1981).
In this model, ki indicates the magnitude of the ascending slope and kZ, the magnitude of the descending slope. For strain SK73a, t tests were used to compare the descending slope of the total virus curve with that of the cellassociated curve, and the descending slope of the total virus curve with that of the extracellular virus curve (Fig. 2). The t values were 0.518 and 0.919, respectively. The fact that neither of these values is statistically significant (P>O.O5) indicates that the amount of total virus declines at the same rate as the amounts of cell-associated and extracellular virus. The traditional method of harvesting stock cultures of reoviruses involves freezing and thawing of infected monolayers three times, and clarification of the supernatant by low speed centrifugation. However, this method would result in poor virus recovery if the virus were predominantly cell-associated, as is the case with the Vero-propagated avian reoviruses. It was our experience
276 TABLE 1 The effect of different propagation
treatments
on dissociation
of avian reoviru?
from Vero cells following
Treatment
Viral titerb
Distilled deionized water Freon extraction Trypsin 3 x freeze-thaw cycles 5 x freeze-thaw cycles 7 x freeze-thaw cycles 10 x freeze-thaw cycles 15 x freeze-thaw cycles
5.78 4.80 3.55 4.63 4.05 3.55 3.97 4.13
“Avian reovirus strain SK125. bMean titer of released virus expressed as negative loglo TCIDX/ml
f 0.61 k 0.63 + 0.63 _+ 0.63 k 0.38 + 0.76 + 0.46 k 0.60
k SEM.
that strain SK125 rapidly produced strong CPE in Vero cells, but stock cultures had low viral titers upon titration. This could be because SK125 makes a lot of fusion protein resulting in strong CPE, but only either immature virions or defective interfering particles are produced which are not detectable by titration. However, based on the viral titers in Fig. lb, this virus also appeared to be more cell-associated in Vero cells than the other 4 strains. The SK125 strain was therefore chosen for analysis of different procedures for releasing of cell-associated virus propagated in Vero cells. In the present study, treatment of infected cell monolayers with sterile ddHzO was the most effective for releasing cell-associated virus (Table 1). In fact this procedure was significantly better than the traditional method of repeated freezing and thawing to lyse cells (Wilcox et al., 1985). Distilled water is routinely used to lyse virus-infected cells when preparing specimens for electron microscopy, and is probably more effective than repeated thawing and freezing of cultures. Trypsin treatment was the least effective procedure for harvesting virus propagated in Vero cells. It was considered that trypsin treatment had a deleterious effect on SK125 because it resulted in a viral titer which was less than that obtained with the standard three freeze-thaw cycles. This finding was surprising since all avian reoviruses had previously been reported to be resistant to trypsin (Kawamura et al., 1965; Petek et al., 1967; Jones et al., 1975). It is concluded that all 14 avian reoviruses isolated in CELi cells induced serially sustained CPE during propagation in Vero cells. The growth curves of 5 of the viruses examined showed them to be highly cell-associated, and this may be true for all avian reoviruses adapted to replicate in Vero cells. Of eight procedures tested with SK125 strain, treatment of virus-infected cultures with ddHz0 was the most effective, and treatment with trypsin was the least effective for recovering cell-associated virus. It is considered that sterile ddH20, as demonstrated for SK125, may prove useful for release of cell-associated virus particularly where large amounts of virus are needed for experimental purposes.
277
Acknowledgement We thank Dr. Alan Donald
for help with the statistical
analysis.
References Barta, V., Springer, W.T. and Millar, D.L. (1984) A comparison of avian and mammalian cell cultures for the propagation of avian reovirus WVU2937. Avian Dis. 28, 216-223. Draper, N.R. and Smith, H. (1981) Applied regression analysis, second edition. John Wiley and Sons, New York. Francki, R.I.B., Fauquet, C.M., Knudson, D.L. and Brown, F. (1991) Classification and nomenclature of viruses. 5th Report of the International Committee on Taxonomy of Viruses. Arch Virol., Suppl 2, Springer-Verlag Vienna, New York, pp. 186. Glass, S.E., Naqi, S.A., Hall, CF. and Kerr, K.M. (1973) Isolation and characterisation of a virus associated with arthritis of chickens. Avian Dis. 17, 415424. Guneratne, J.M. and Jones, R.C. (1982) Some observations on the isolation and cultivation of avian reoviruses. Avian Pathol. 11, 453462. Hussian, M. and Spradbrow, P.B. (1981) Avian adenoviruses and avian reoviruses isolated from diseased chickens. Aust. Vet. J. 57, 436437. Joklik, W.K. (1983) The members of the family reoviridae. In: W.K. Joklik (Ed.), The Reoviridae. Plenum, New York/London, pp. 11. Jones, R.C. and Al Afaleq, AI. (1990) Different sensitivities of Vero cells from two sources to avian reoviruses. Res. Vet. Sci. 48, 379-380. Jones, R.C. and Kibenge, F.S.B. (1984) Reovirus-induced tenosynovitis in chickens: the effect of breed. Avian Pathol. 13, 173-189. Jones, R.C., Jordan, F.T.W. and Lioupis, S. (1975) Characterization of reovirus isolated from ruptured gastrocnemius tendons of chickens. Vet. Rec. 96, 153-154. Kawamura, H. and Tsubahara, H. (1966) Common antigenicity of avian reoviruses. Natl. Inst. Animal Hlth. Q. Tokyo. 6, 187-195. Kawamura, H., Shimizu, F. Maeda, M. and Tsubahara, H. (1965) Avian reovirus: its properties and serological classification. Natl. Inst. Animal Hlth. Q. Tokyo. 5, 115-124. Kibenge, F.S.B. and Wilcox, G.E. (1983) Tenosynovitis in chickens. Vet. Bull. Weybridge 53, 431443. Kibenge, F.S.B., Dhillon, A.S. and Russell, R.G. (1988) Growth of serotypes I and II and variant strains of infectious bursal disease virus in Vero cells. Avian Dis. 32, 298-303. Mallo, M., Martinez-Costas, J. and Benavente, J. (1991a) Avian reovirus S1133 can replicate in mouse L cells: effect of pH and cell attachment status on viral infection. J. Virol. 65, 5499-5505. Mallo, M., Martinez-Costas, J. and Benavente, J. (1991 b) The stimulatory effect of Actinomycin D on avian reovirus replication in L cells suggests that translational competition dictates the fate of the infection. J. Virol. 65, 55065512. Nwajei, B.N.C., Al Afaleq, AI. and Jones, R.C. (1988) Comparison of chick embryo liver and Vero cell cultures for the isolation and growth of avian reoviruses. Avian Pathol. 17, 759-766. Petek, M., Felluga, B., Borghi, G. and Baroni, A. (1967) The crawley agent: an avian reovirus. Arch. Ges. Virusf. 21, 413424. Reed, L.J. and Muench, H. (1938) A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27, 493497. Robertson, M.D. and Wilcox, G.E. (1986) Avian reovirus. Vet. Bull. Weybridge 56, 155174. Samuel, C.E. (1988) Reoviruses. In: G. Darai (Ed), Virus Diseases in Laboratory and Captive Animals. Martinus Nijhoff Publ., Boston, pp. 497-520. Schnitzer, T.J., Ramos, T. and Gouvea, V. (1982) Avian reovirus polypeptides: analysis of intracellular virus-specified products, virions, top component, and cores. J. Virol. 43, 1006-1014.
278 Spandidos, D.A. and Graham, A.F. (1976) Physical and chemical characterization of an avian reovirus. J. Virol. 19, 968-976. Wilcox, G.E. and Compans, R.W. (1982) Cell fusion induced by Nelson Bay virus. Virology 123, 312-322. Wilcox, G.E., Robertson, M.D. and Lines, A.D. (1985) Adaptation and characteristics of replication of a strain of avian reovirus in Vero cells. Avian Pathol. 14, 321-328.
269
VIRMET 01385
Comparison of eight different procedures for harvesting avian reoviruses grown in Vero cells Yatri Drastini, Frederick S.B. Kibenge, Patricia K. McKenna and Alfonso Lopez Atlantic Veterinary College, University of Prince Edward Island, Charlottetown. Prince Edward Island (Canada) (Accepted 29 April 1992)
Summary
14 avian reovirus isolates adapted to replicate in an African green monkey (Vera) cell line were studied for the nature of their replication. The growth curves of 5 viruses showed them to be highly cell-associated in Vero cells. Different procedures were examined for releasing the cell-associated virus following propagation in Vero cells, including several freeze-thaw cycles, treatment with sterile distilled deionized water (ddHzO), freon extraction, and trypsin treatment. Treatment of virus infected cultures with ddHz0 was the most effective, and trypsin treatment was the least effective procedure for dissociation of virus from cells. Treatment of virus infected cultures with ddH20 is a simple and effective procedure which can be used where large amounts of virus are required for experimental purposes. Avian reovirus; Vero cell; Virus replication; Cell-associated virus; Virus harvest Reoviruses belong to the family Reoviridae which has an extremely wide host range among animals and plants (Samuel, 1988). Members of the Reoviridae family found in vertebrates form five genera: Orthoreovirus, Rotavirus, Orbivirus, Coltivirus, and Aquareovirus (Francki et al., 1991). The members of the Urthoreovirus genus are distinguishable into two distinct groups: those of avian origin, and those of mammalian origin (Kawamura and Tsubahara, 1966). Both groups of viruses have a genome consisting of 10 segments of double-stranded RNA (dsRNA), separable into three size classes, which are enclosed within a double protein capsid shell 70 to 80 nm in diameter Correspondence IO: F.S.B. Kibenge, Department of Pathology and Microbiology, Atlantic Veterinary College, UPEI, 550 University Avenue, Charlottetown, PEI, CIA 4P3, Canada.
270
with a similar protein composition and distribution (Joklik, 1983; Schnitzer et al., 1982; Spandidos and Graham, 1976). However, avian reoviruses, unlike their mammalian counterparts, exhibit a wide heterogeneity in their neutralizing antigens (Robertson and Wilcox, 1986), possess a different group-specific antigen (Petek et al., 1967), cause no haemagglutination (Glass et al., 1973), are associated with disease in their natural hosts (Kibenge and Wilcox, 1983), and cause syncytium formation both in infected cell cultures and in susceptible chickens (Kibenge and Wilcox, 1983; Wilcox and Compans, 1982). Avian reoviruses can be readily grown in the laboratory in primary cultures of cells of avian origin (Barta et al., 1984; Guneratne et al., 1982). However, among the established mammalian cell lines tested, only Vero cells have been reported to support replication of certain strains of avian reovirus (Barta et al., 1984; Wilcox et al., 1985; Jones and Alfaleq, 1990; Nwajei et al., 1988). The avian reoviruses in the present study were adapted to Vero cells which, in addition to being readily available, allowed for consistent titration of infectious virus because of the uniform nature of cells of the continuous cell line. However, the only report on the nature of avian reovirus replication in Vero cells found the virus to be highly cell-associated, requiring several freeze-thaw cycles to release virus from infected cells (Wilcox et al., 1985). Since that study examined only one avian reovirus, strain RAM-l, because other strains could not replicate in Vero cells, it was not clear whether the cell-associated character of RAM-l in Vero cells was strain specific. This paper reports the growth curves of an additional live avian reoviruses in Vero cells. Furthermore, several procedures evaluated for recovery of virus grown in Vero cells are described. Virus samples, including 4 reference strains (avian reovirus Type 59, Type 24, WVU2937, and Crawley), and 10 local isolates (SK suffix) from the Health of Animals Laboratory, Sackville, New Brunswick, Canada were used. All samples were received as chorioalantoic membrane material except for three samples (SK125, SK73a, and SK25a) which were chick kidney cell culture supernatants. All virus samples were passaged once in primary chicken embryo liver (CELi) cell monolayers on 24-multiwell tissue culture plates (Jones and Kibenge, 1984). Cytopathic effect (CPE)-positive cultures were then serially passaged at 7-day intervals in Vero cells (Kibenge et al., 1988) 2 to 6 times prior to preparation of stock cultures. For viral growth curves, 25-cm* tissue culture flasks containing monolayers of Vero cells were inoculated with lo5 TCIDso of virus in 0.1 ml of medium per flask, considered sufficient to produce a ‘onestep’ growth curve. At intervals after infection (relative to the end of the adsorption period), two flasks were removed and harvested essentially as previously described (Kibenge et al., 1988). Three fractions, namely extracellular virus harvest, cell-associated virus harvest, and total virus harvest were stored at -70°C and later assayed for virus infectivity. The assay for the growth curves of one selected virus was repeated three times. For harvesting of virus, infected Vero cell monolayers in 25-cm* tissue culture flasks with approximately 75% CPE were used. The medium of one flask was
271
discarded, and was replaced with 2 ml of sterile distilled deionized water (ddH*O). The flask was stored at -70°C and was later assayed for virus infectivity. A second flask was frozen and thawed three times, the cell lysate was extracted once with 1,1,2-trichlorotrifluoroethane (Freon) (J.T. Baker Chemical Co., Phillipsburg, NJ, USA), and the aqueous phase was stored at -70°C until when assayed for virus infectivity. The medium of a third flask was discarded, and the infected cell monolayer was treated with 0.25% trypsin at 37°C for 10-15 min. Following trypsinization, 2 ml of growth medium were added and the flask was stored at - 70°C and later assayed for virus infectivity. The fourth, fifth, sixth, seventh and eighth flasks were frozen and thawed 3, 5, 7, 10 and 15 times, respectively, and following low speed centrifugation to remove cellular debris, the supernatants were titrated to determine virus infectivity. Flasks treated with sterile ddHz0 or with trypsin were also frozen and thawed three times prior to titration of the viral supernatants. Each assay was repeated three times. Titration of virus was done in quadruplicates in 96well microtiter tissue culture plates. Plates were incubated at 37°C in a humidified 5% CO2 atmosphere for 7 days and were then examined microscopically prior to fixation and staining with 1% crystal violet in 70% ethanol. The virus titer was determined from the CPE using the procedure described by Reed and Muench (1938). All 14 avian reoviruses isolated in CELi cells induced serially sustained CPE during successive passages in Vero cells indicating that they were adapted to grow in the Vero cell line. Hussian et al. (198 1) previously reported successful adaptation of three Australian isolates examined although Wilcox et al. (1985) could only adapt one strain, designated RAM-l, of six other Australian avian reovirus strains to Vero cells. The RAM1 strain had been passaged 4 to 6 more times in chicken cells than the unadaptable strains. In the present study, prior to the single passage in CELi cell cultures, most isolates had only been propagated in chicken embryonated eggs 1 to 2 times, suggesting that cell passage history of the reovirus strain is probably not a major factor in adaptation to Vero cells. It is probable that variation either in culture conditions or between batches of Vero cell line (Jones and Alfaleq, 1990), and individual virus strain variation may he responsible for the different success rates of adaptation of avian reoviruses to Vero cells. It has been recently demonstrated that avian reovirus strain S1133 can replicate in Mouse L cells at pH 6.4 and pH 7.2 but not pH 8.2, suggesting that a basic pH in the culture medium is responsible for the inhibition in viral replication (Mall0 et al., 1991a). These authors also showed that the presence of Actinomycin D during infection increased viral replication in L cells by eliminating host mRNA competition for the translation machinery (Mall0 et al., 1991b). This suggests that virus replication in L cells is also dependent on the multiplicity of virus infection. The same may be true for avian reovirus replication in Vero cells. The growth curves of 5 avian reovirus strains (Crawley, SK125, SK98a, SK103a, and SK73a) were determined to further characterize the nature of avian reovirus replication in Vero cells. In all five virus strains the growth curve
io
20
I
50
A
I
100
~
100
TOTAL VIRUS
CELL&sOClAlEO
I
Fig. 1. a and b. (see page 273 for legend.)
0
4
/R1/---
1
0
TIME (HOURS)
VIRUS
nME (HOURS)
I~--
GROWTH CURVE OF REOVIRUS STRAIN CRAWLEY IN VERO CELLS
3
5
0
0
0
20
20
20
/‘
b
/
n--
50
00
50
nME (HOURS)
100
TOTAL VIRUS
100
nm
(HOURS)
TINE (HOURS)
CELL-ASSOC~A~EO VIRUS
100
EXTRACELLULAR VIRUS
GROWTH CURVE OF AVIAN REOVIRUS STRAIN SK125 IN VERO CELLS
273
of cell-associated virus resembled that of total virus (Fig. 1) indicating that these viruses were highly cell-associated. This may be true for all avian reoviruses adapted to replicate in Vero cells. The titers of total and cellassociated virus peaked at 30 to 96 h PI; whereas the maximum titers of extracellular virus were steadily rising throughout the 144 h observation period. However, this cell-associated character of avian reovirus replication in Vero cells could not be confirmed by statistical analysis of the growth curves. For this analysis, the amount of virus, A, at time t was modelled using an adaptation of the Gompertz model for growth curves (Draper et al., 1981). A= aexp{-ble-k’f-_bze-k2’}. GROWTH CURVE OF AVIhN REOVIRUS STRAIN %W#
IN VERO w
C
9
BTRACEUUIAR
VIRUS
lI
0
I
I
I
I
I
100
So
20
I
TIME (HOURS)
.f
-
0
I
I
I
60
20
I
100
I
I
TIME (HOURS)
TOTAL VIRUS 9
I 0
1
20
I
sb
I
l&
‘iTME (HbURS)
Fig. 1. (c above.) Growth curves of avian reovirus strains Crawley (a), SK125 (b), SK98a (c), SK103a (d), and SK73a (e). Vero cell monolayers were infected with lo5 TCIDH, of virus in 0.1 ml of medium per 25cm’ tissue culture flask. After a l-h adsorption period, cells were washed twice with phosphate buffered saline, and following the addition of fresh growth medium, they were maintained at 37°C in a humidified 5% CO2 atmosphere. Infected cells were harvested at the indicated time intervals, and viral progeny from the infected cells was titrated by TCIDs,,.
274
,.
II \ I ii I
5 VI
CI
f
275
GROWTH CURVE OF AVIAN REOVIRUS
STRAIN SK725 IN VERO CELLS
Fig. 2. Growth curve of avian reovirus strain SK73a in Vero cells, analysed using the Gompertz model for growth curves (Draper et al., 1981).
In this model, ki indicates the magnitude of the ascending slope and kZ, the magnitude of the descending slope. For strain SK73a, t tests were used to compare the descending slope of the total virus curve with that of the cellassociated curve, and the descending slope of the total virus curve with that of the extracellular virus curve (Fig. 2). The t values were 0.518 and 0.919, respectively. The fact that neither of these values is statistically significant (P>O.O5) indicates that the amount of total virus declines at the same rate as the amounts of cell-associated and extracellular virus. The traditional method of harvesting stock cultures of reoviruses involves freezing and thawing of infected monolayers three times, and clarification of the supernatant by low speed centrifugation. However, this method would result in poor virus recovery if the virus were predominantly cell-associated, as is the case with the Vero-propagated avian reoviruses. It was our experience
276 TABLE 1 The effect of different propagation
treatments
on dissociation
of avian reoviru?
from Vero cells following
Treatment
Viral titerb
Distilled deionized water Freon extraction Trypsin 3 x freeze-thaw cycles 5 x freeze-thaw cycles 7 x freeze-thaw cycles 10 x freeze-thaw cycles 15 x freeze-thaw cycles
5.78 4.80 3.55 4.63 4.05 3.55 3.97 4.13
“Avian reovirus strain SK125. bMean titer of released virus expressed as negative loglo TCIDX/ml
f 0.61 k 0.63 + 0.63 _+ 0.63 k 0.38 + 0.76 + 0.46 k 0.60
k SEM.
that strain SK125 rapidly produced strong CPE in Vero cells, but stock cultures had low viral titers upon titration. This could be because SK125 makes a lot of fusion protein resulting in strong CPE, but only either immature virions or defective interfering particles are produced which are not detectable by titration. However, based on the viral titers in Fig. lb, this virus also appeared to be more cell-associated in Vero cells than the other 4 strains. The SK125 strain was therefore chosen for analysis of different procedures for releasing of cell-associated virus propagated in Vero cells. In the present study, treatment of infected cell monolayers with sterile ddHzO was the most effective for releasing cell-associated virus (Table 1). In fact this procedure was significantly better than the traditional method of repeated freezing and thawing to lyse cells (Wilcox et al., 1985). Distilled water is routinely used to lyse virus-infected cells when preparing specimens for electron microscopy, and is probably more effective than repeated thawing and freezing of cultures. Trypsin treatment was the least effective procedure for harvesting virus propagated in Vero cells. It was considered that trypsin treatment had a deleterious effect on SK125 because it resulted in a viral titer which was less than that obtained with the standard three freeze-thaw cycles. This finding was surprising since all avian reoviruses had previously been reported to be resistant to trypsin (Kawamura et al., 1965; Petek et al., 1967; Jones et al., 1975). It is concluded that all 14 avian reoviruses isolated in CELi cells induced serially sustained CPE during propagation in Vero cells. The growth curves of 5 of the viruses examined showed them to be highly cell-associated, and this may be true for all avian reoviruses adapted to replicate in Vero cells. Of eight procedures tested with SK125 strain, treatment of virus-infected cultures with ddHz0 was the most effective, and treatment with trypsin was the least effective for recovering cell-associated virus. It is considered that sterile ddH20, as demonstrated for SK125, may prove useful for release of cell-associated virus particularly where large amounts of virus are needed for experimental purposes.
277
Acknowledgement We thank Dr. Alan Donald
for help with the statistical
analysis.
References Barta, V., Springer, W.T. and Millar, D.L. (1984) A comparison of avian and mammalian cell cultures for the propagation of avian reovirus WVU2937. Avian Dis. 28, 216-223. Draper, N.R. and Smith, H. (1981) Applied regression analysis, second edition. John Wiley and Sons, New York. Francki, R.I.B., Fauquet, C.M., Knudson, D.L. and Brown, F. (1991) Classification and nomenclature of viruses. 5th Report of the International Committee on Taxonomy of Viruses. Arch Virol., Suppl 2, Springer-Verlag Vienna, New York, pp. 186. Glass, S.E., Naqi, S.A., Hall, CF. and Kerr, K.M. (1973) Isolation and characterisation of a virus associated with arthritis of chickens. Avian Dis. 17, 415424. Guneratne, J.M. and Jones, R.C. (1982) Some observations on the isolation and cultivation of avian reoviruses. Avian Pathol. 11, 453462. Hussian, M. and Spradbrow, P.B. (1981) Avian adenoviruses and avian reoviruses isolated from diseased chickens. Aust. Vet. J. 57, 436437. Joklik, W.K. (1983) The members of the family reoviridae. In: W.K. Joklik (Ed.), The Reoviridae. Plenum, New York/London, pp. 11. Jones, R.C. and Al Afaleq, AI. (1990) Different sensitivities of Vero cells from two sources to avian reoviruses. Res. Vet. Sci. 48, 379-380. Jones, R.C. and Kibenge, F.S.B. (1984) Reovirus-induced tenosynovitis in chickens: the effect of breed. Avian Pathol. 13, 173-189. Jones, R.C., Jordan, F.T.W. and Lioupis, S. (1975) Characterization of reovirus isolated from ruptured gastrocnemius tendons of chickens. Vet. Rec. 96, 153-154. Kawamura, H. and Tsubahara, H. (1966) Common antigenicity of avian reoviruses. Natl. Inst. Animal Hlth. Q. Tokyo. 6, 187-195. Kawamura, H., Shimizu, F. Maeda, M. and Tsubahara, H. (1965) Avian reovirus: its properties and serological classification. Natl. Inst. Animal Hlth. Q. Tokyo. 5, 115-124. Kibenge, F.S.B. and Wilcox, G.E. (1983) Tenosynovitis in chickens. Vet. Bull. Weybridge 53, 431443. Kibenge, F.S.B., Dhillon, A.S. and Russell, R.G. (1988) Growth of serotypes I and II and variant strains of infectious bursal disease virus in Vero cells. Avian Dis. 32, 298-303. Mallo, M., Martinez-Costas, J. and Benavente, J. (1991a) Avian reovirus S1133 can replicate in mouse L cells: effect of pH and cell attachment status on viral infection. J. Virol. 65, 5499-5505. Mallo, M., Martinez-Costas, J. and Benavente, J. (1991 b) The stimulatory effect of Actinomycin D on avian reovirus replication in L cells suggests that translational competition dictates the fate of the infection. J. Virol. 65, 55065512. Nwajei, B.N.C., Al Afaleq, AI. and Jones, R.C. (1988) Comparison of chick embryo liver and Vero cell cultures for the isolation and growth of avian reoviruses. Avian Pathol. 17, 759-766. Petek, M., Felluga, B., Borghi, G. and Baroni, A. (1967) The crawley agent: an avian reovirus. Arch. Ges. Virusf. 21, 413424. Reed, L.J. and Muench, H. (1938) A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27, 493497. Robertson, M.D. and Wilcox, G.E. (1986) Avian reovirus. Vet. Bull. Weybridge 56, 155174. Samuel, C.E. (1988) Reoviruses. In: G. Darai (Ed), Virus Diseases in Laboratory and Captive Animals. Martinus Nijhoff Publ., Boston, pp. 497-520. Schnitzer, T.J., Ramos, T. and Gouvea, V. (1982) Avian reovirus polypeptides: analysis of intracellular virus-specified products, virions, top component, and cores. J. Virol. 43, 1006-1014.
278 Spandidos, D.A. and Graham, A.F. (1976) Physical and chemical characterization of an avian reovirus. J. Virol. 19, 968-976. Wilcox, G.E. and Compans, R.W. (1982) Cell fusion induced by Nelson Bay virus. Virology 123, 312-322. Wilcox, G.E., Robertson, M.D. and Lines, A.D. (1985) Adaptation and characteristics of replication of a strain of avian reovirus in Vero cells. Avian Pathol. 14, 321-328.
