Host Cell Range and Growth Characteristics of

INFECTION AND IMMUNITY, Mar. 1973, p. 398-4 Copyright 1973 Americau Society for Microbiology Vol. 7, No. 3 Printed in U.S.A. Host Cell Range and Growth Characteristics of Bovine Parvoviruses' R. C. BATES' AND J. STORZ Department of Microbiology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado 861 Rteceived for publicationi 5 September 197 In assessing the host cell range of bovine parvoviruses, these viruses were found to replicate optimally in actively dividing bovine fetal lung and spleen cells. Other primary bovine fetal cells supported growth to a lesser extent, but bovine line cells and line cells of other animal species tested did not. Minimal infectivity remained after passage of bovine parvovirus in cells from chicken embryos and guinea pig fetuses. During bovine parvovirus replication in bovine fetal lung and spleen cells, production kinetics of infectious virus and hemagglutinins were determined. An eclipse period of 16 h occurred, and viral release from cells was not detected until 3 h after inoculation of bovine fetal lung cells and 36 h after inoculation of bovine fetal spleen cells. Cell-associated virus titers were always higher than extracellular virus titers. Hemagglutinins were detected in parallel to infectious virus. The HADEN virus of Abinanti and Warfield (1) was recently identified as a bovine parvovirus (6), and additional bovine parvovirus isolates were recovered and identified (R. C. Bates, J. Storz, and D. E. Reed, J. Infect. Dis., in press). Cultivation of parvoviruses is difficult, since the host cell range and conditions for.optim=. propagation of parvoviruses are not well defined. The host cell range of Kilham rat virus is limited. to rat and hamster cells (7). The H-viruses grew extensively in a variety of human and simian cell lines as well as in primary cells of whole rat and hamster embryos (7). Adeno-associated viruses can be propagated in a wide variety of cell types, but they require adenovirus as a helper (4). To study bovine parvoviral infections, optimal conditions for isolation and propagation had to be defined. The purpose of this investigation was to identify cell types cultured in vitro from cattle and laboratory animals which supported optimal replication of bovine parvoviruses. (This investigation is from a thesis submitted by R. C. Bates to Colorado State University in partial fulfillment of the requirements for the Ph.D. degree.) MATERIALS AND METHODS Virus strains. Bovine parvovirus 1 (1) in the 11th passage and bovine parvovirus 71-1-W I This publication is scientific paper no. 1799 of the Colorado Agricultural Experiment Station. Present address: Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Va. 461. 398 (Bates, et al., in press) in the 6thpassage were used. Preparation of cell cultures. Primary cell cultures were prepared by the method of Youngner (8) from organs of bovine and guinea pig fetuses and from chicken embryos. Trypsin solution at a concentration of.5% was used to disperse fetal lung tissue. Bovine and guinea pig fetal cells were grown in Earle balanced salt solution containing.5%, lactalbuipin and lambbsdrum. Bovine fetal lung (BFL) cells were grown in Eagle HeLa medium supplemented with bovine fetal serum in the first passage and lamb serum in the second and third passage. Chicken embryonic cells were grown in medium 199 and bovine fetal serum. Recommended media and serum (ATCC) supporting optimal growth were used for culture of different cell lines. All media contained 5 U per ml of penicillin G and 5 ;g of streptomycin per ml. For parvovirus propagation, cells were seeded at a concentration of 15, to 5, cells per ml of medium, and cell cultures were inoculated 18 to 4 h after seeding. Such cell cultures are referred to as "actively dividing." Host cell range study. Three passages of bovine parvovirus 1 with an input of 15-' mean tissue culture infective dose (TCIDro) per.1 ml were made with each cell type grown in tubes. The inoculated cell cultures were observed for 1 days, and the cultures were frozen at each passage when cytopathic changes were complete or when 1 days had elapsed. These cell cultures were frozen and thawed three times and then tested for hemagglutinins (Bates et al., in press). Uninoculated control cultures for each cell type were processed identically. Following the third passage in each cell type,

VOL. 7, 1973 BOVINE PARVOVIRUSES 399 Type of bovine fetal cell the tissue culture fluid was titered in BFL cells by determining cytopathic and hemadsorbing end points and levels of hemagglutinins. Guinea pig red blood cells were used in the hemagglutination and hemadsorption tests. One-step growth experiments. One million BFL and bovine fetal spleen (BFS) cells were seeded into 3-ml plastic tissue culture flasks. Between 18 to 4 h after seeding, the cell cultures were washed once with 5 ml of Dulbecco phosphate buffer, and 1 ml of a high multiplicity (8:1) of bovine parvovirus 71-1-W was adsorbed for h at 37 C. The unadsorbed inoculum was removed by washing the cell layers twice, and 5 ml of growth medium with 5% lamb serum was added. The amount of unadsorbed virus was determined by plaque assay. At selected intervals postinfection, the growth medium was removed from the culture flask and the cell layer was washed three times with ml of Dulbecco phosphate buffer. The culture medium and washing fluids were combined and centrifuged at 1, X g for 1 min to sediment cells, nuclei, and cellular debris. The supernatant fraction was saved for titration of extracellular virus. The pellet was resuspended in 5 ml of medium and added to the cell layer. These cell culture flasks were frozen and thawed three times, and the resultant cell debris was sonically treated for min at a power setting of 85 W. Virus released by this treatment was designated as cell-associated virus. Plaque assay. The plaque technique was used to determine viral infectivity in the extracellular virus and cell-associated virus fractions. BFS cells (1,5,) were seeded in plastic petri dishes (6 mm). Between 18 to 4 h after seeding, the cell cultures were washed with 5 ml of Dulbecco phosphate buffer, and.5 ml of serial 1-fold dilutions of extracellular virus and cellassociated virus samples were adsorbed for 1 h at 37 C. After adsorption, 8 ml of overlay consisting of Lavit (lactalbumin-vitamin) medium and.85% Noble agar, and 5% inactivated lamb serum was added to each plate. The cultures were incubated for 5 to 7 days at 37 C in a CO incubator. The plaques were observed by staining with neutral red or crystal violet. RESULTS Replication in bovine fetal cells. Bovine parvovirus 1 replicated in all primary bovine fetal cells tested (Table 1). High levels of hemagglutinins were present in tissue culture fluids after each passage in the different cell types. The virus reached cytopathic and hemadsorption titers of 16 and 16.8 TCID5o/ml on final assay after three passages in BFL cells. These were the highest titers observed for replication of bovine parvovirus 1 in any type of bovine fetal cells. The least efficient replication of bovine parvovirus 1 occurred in bovine fetal kidney and bovine fetal intestinal cells with cytopathic end points of 14. and 13 5 TCID5/ ml, respectively. Replication of bovine parvovirus 1 occurred to approximately similar titers in bovine fetal testicle, adrenal, and BFS cells. In all cases, hemadsorption end points were consistently higher than cytopathic end points. The cytopathic effect resulting from bovine parvovirus replication in actively dividing BFL cells was distinctive and reproducible. At to 4 h after inoculation, infected cells were swollen, refractile, and had a stellate appearance. These cells were scattered over the cell layer and ultimately became uniformly round and highly refractile. Soon after the cells became round, they detached from the glass surface. These cytopathic changes continued to appear until all cells were involved, usually by 7 h after inoculation. Similar cytopathic changes were observed after infection of bovine fetal testicle, adrenal, and BFS cells, but no discernable cytopathic effect was seen in inoculated bovine fetal intestinal and bovine fetal kidney cells. Replication in established cell lines. Cell lines from a variety of animal species were TABLE 1. Replication of bovinie parvovirus 1 in bovine fetal cells" HA titer in pasage Titer of final passage in BFL cells Lung 51 51,48 6. 6.8 4,96 Kidney 51 56 18 4. 4.8 56 Testicle,48 51 18 5. 5.5 4,96 Spleen 56 1,4 51 4.5 4.8 4,96 Intestine 4,96 4,96 51 3.5 5. 3 Adrenal 4,96 18 4,96 4.5 5. 56 a Abbreviations: HA, hemagglutination titer; BFL, bovine fetal lung; CPE, cytopathic end point; HAD, hemadsorption end point. b Titers expressed as -log1o/ml.

4 BATES AND STORZ INFECT. IMMUNITY tested for their susceptibility to bovine parvovirus infection. The results clearly indicate that the line cells examined did not support significant replication of bovine parvovirus 1 (Table ). Except for decreasing titers observed in the bovine bone marrow cell strain (FB4BM), hemagglutinins were not detected in these tissue cultures. No significant infectivity was detected after three passages in FB4BM cells. A very low level of infectivity and hemagglutinins was detected in BFL cells after three passages of bovine parvovirus 1 in Madin Darby bovine kidney and BHK-1 cells. Replication in chicken embryonic and fetal guinea pig cells. A variable number of passages of bovine parvovirus 1 were made in chicken embryonic fibroblast, luilg, and kidney cells, and in guinea pig fetal lung, kidney, and testicle cells before final assay in BFL cells. TABLE. Replication of bovine parvovirus I in line cellsa Type of cell HA titer in passage Titer of final passage in BFL cells FB4BM 56 3 4. 1. < MDBK 3 < <.5 1.5 56 LLC-MK < < <.. < Mouse L < < <. 1.5 < HeLa < < <..5 < BHK-1 < < <.5.5 3 CHO < < <.. < a Abbreviations: FB4BM, bovine bone marrow; MDBK, Madin Darby bovine kidney; LLC-MK, rhesus monkey kidney; Mouse L, mouse fibroblast; HeLa, human cervical carcinoma; BHK-1, baby hamster kidney; CHO, Chinese hamster ovary; see Table 1 for other abbreviatioins. b Titers expressed as -logio/ml. Type of cell Both guinea pig and chicken cells supported limited replication of bovine parvovirus 1 (Table 3). Hemagglutinins were not present in chicken fibroblast cultures after different passages, but a cytopathic titer of 13"1 TCID5o/ml and a hemagglutination titer of 64 were detected in BFL cells inoculated with material of the third passage in chicken fibroblasts. Minimal infectivity remained after three passages in chicken embryonic lung and kidney cells. One or two passages of bovine parvovirus 1 were made in guinea pig embryonic cells. Low levels of hemagglutinins were detected after each passage of virus. Cytopathic titers ranging from.8 to 4. TCID5o/ml and hemagglutination titers of 56 to 4,96 were detected in BFL cells after passage in guinea pig cells. Distinctive cytopathic changes were not induced by bovine parvovirus 1 in guinea pig and chicken cells. Plaque characteristics. Plaques resulting from bovine parvovirus 1 and 71-1-W replication in BFS cells were detected within 5 days after inoculation. The plaque sizes ranged from 1 to 3 mm in diameter at 5 days and increased to 3 to 5 mm at 7 days postinfection. All plaques, irrespective of the size, had fuzzy edges (Fig. 1). One-step growth curve. Results of growth curve experiments with bovine parvovirus 71-1- W in BFL cells are summarized in Fig.. Cell cultures were inoculated with 1 ml of virus stock with a titer of 8 X 16 plaque-forming units (PFU) /ml. The titer of the unadsorbed inoculum was X 16 PFU/ml. The adsorbed multiplicity was 6 PFU per cell. As this virus is very stable, residual infectivity was detected at each sampling time plrior to the appearance of progeny virus. Approximately 16 h elapsed after infection before progeny virus was detected intracellularly. AIn additional 14 h elapsed before progeny virus appeared extracellularly. A TABLE 3. Replicationb of bovine parvovirus 1 in chicketn anla guinea pig cellsa HA titer in passage Titer of final passage in BFL cells CE fibroblasts < < < 3.8 4.8 64 CE kidney 16 <..5 CE lung < < < 1.5 1.8 16 GP kidney 3 4 NT.8 3.8 56 GP lung 3 NT NT 3.5 4.5 4,96 GP testis 3 NT 4. 4.5 4,96 a Abbreviations: NT, not tested; CE, chickeni embryo; GP, guiniea pig; see Table 1 for other abbreviations. b Titers expressed as -logio/ml.

VOL. 7, 1973 BOVINE PARVOVIRUSES 41 FIG. 1. Plaques induced by bovine parvovirus 1 in bovine fetal spleent cells. Plaques (1-3 mm) were detected 5 days after inoculation following stainintg with crystal violet. -J CD 6 *ACELL-ASSOCIATED VIRUS o *EXTRACELLULAR VIRUS A -CELL-ASSOCIATED HEMAGGLUTININ * -EXTRACELLULAR HEMAUULUTININ 4l *~~~~~~~~~ I 196 _ D4 c- 4. ^: o ^ 56 4 Z - 6 1 18 4 3 36 4 48 54 6 66 7 78 84 HOURS AFTER INFECTION Fiw;.. Onte-step growth curve of bovine parvovirus 71-1-W replication in bovinte fetal luntg cells. logarithmic increase in cell-associated virus occurred between 16 anid 48 h after inoculation. A maximum cell-associated titer of.1 X 17 PFU/ml was reached at 84 h after inoculation. A logarithmic increase in extracellular virus was observed between 3 anid 6 h after inoculationi. The peak titer of 1.5 X 16 PFU/ml of extracellular virus was observed at 84 h. Cell-associated virus titers were consistently higher thail extracellular virus titers. The production of cell-associated hemaggluti- 64-16 4S 1 *t nins (Fig. ) paralleled infectivity. Cell-associated hemagglutinins began to increase 16 h after inoculation and reached a titer of 4,96 at 48 h. Hemagglutinins were not detected in the extracellular medium until 36 h, and they reached a titer of 51 at 6 h after inoculation. Cell-associated hemagglutinin titers always were higher. Results of growth curve experiments with bovine parvovirus 71-1-W in BFS cells are presented in Fig. 3. Cell cultures were inoculated with 1 ml of virus stock with a titer of 8 X 16 PFU/ml. The titer of the unadsorbed inoculum was 16 PFU/ml. The adsorbed multiplicity was calculated to be 7 PFU per cell. Progeny cellassociated virus was detected 16 h after inoculation. Cell-associated virus increased exponentially between 16 and 48 h. At 7 h a maximal titer of 4.4 X 17 PFU/ml was reached. Extracellular virus was detected at 36 h after inoculation and increased logarithmically to a titer of 7.7 X 15 PFU/ml at 6 h. A maximal extracellular virus titer of 8.4 X 15 PFU/ml was observed at 7 h. Cell-associated virus titers were consistently higher thani extracellular virus titers at each sampling time. Cell-associated hemagglutinins increased after 36 h p)ostinfection, h later than detection of cell-associated infectivity (Fig. 3). A maximum cell-associated hemagglutinin titer of 18 was

4 BATES AND STORZ INFECT. IMMUNITY -I a Ui. CD *.CELL z ASSOCIATED 56 HEMAGGLUTINK P.EXETRACELLULAR <6,.16 6 1 W 6 1 18 4 3 36 48 54 6 66 7 HOURS AFTER INFECTION FIG. 3. One-step growth curve of bovine parvovirus 71-1-8OW replication in bovine fetal spleen cells. observed at 6 h. Extracellular hemagglutinins increased after 36 h postinfection and reached a titer of 56 at 7 h. Extracellular hemagglutinins paralleled extracellular infectivity. DISCUSSION The results presented here demonstrate that bovine parvoviruses replicate mainly in primary bovine fetal cells. Surprisingly, cell lines of bovine origin did not support significant replication. Considering the limited studies on host cell ranges of other parvoviruses, it appears that they also replicate primarily in cell types of the host species from which they were originally isolated. For example, H-1 virus replicates l)referentially in cell lines of human and simian origin (7). Similarly, the canine parvovirus grew only in a canine cell line (3). The porcine parvovirus can be cultivated in primary cells and cell lines of porcine origin (5). A 1-h eclipse period for H-1 virus replication in NB cells was observed (), but in this study, cell-associated infectivity was not determined separately from extracellular infectivity. Therefore, the time of virus release after onset of viral maturation was not known. Production kinetics for other parvoviruses have not been reported. An eclipse period of 16 h occurred after BFL and BFS cells were inoculated with bovine parvovirus. Intracellular progeny virus was detected at this time, but virus release did not occur until 3 h after inoculation of BFL cells and 36 h after inoculation of BFS cells. Cellassociated virus titers were always higher than extracellular virus titers. The slow release of virus and the observation of large numbers of virions in ultrathin sections of nuclei examined late in infection suggest that this virus remains closely cell associated during its replicative cycle (R. C. Bates, Ph.D. thesis, Colorado State University, 197). Equal infectivity titers were obtained after bovine parvovirus infection of BFL and BFS cells. However, hemagglutinin titers were higher in BFL cells than in BFS cells. In addition to complete virions, more hemagglutinin and thus empty viral capsids were produced in BFL cells. This became evident from examinations of thin sections of infected cells. The nuclei of such cells always contained a greater number of empty capsids than complete virions (Bates, Ph.D. thesis). ACKNOWLEDGMENTS This investigation was made possible by Public Health Service research grant AI 84 from the National Institute of Allergy and Infectious Diseases, by Public Health Service general research support grant 5S 1 RR5458 from the National Institute of General Medical Sciences, and by funds from the Colorado Agricultural Experiment Station and Regional Research Funds of the Western Regional Projects W-88 and W-11. LITERATURE CITED 1. Abinanti, F. R., and M. S. Warfield. 1961. Recovery of a hemadsorbing virus (HADEN) from the gastrointestinal tract of calves. Virology 14:88-89.. Al-Lami, F., N. Ledinko, and H. Toolan. 1969. Electron microscope study of human NB and SMH cells infected with the parvovirus, H-1: involvement of the nucleolus. J. Gen. Virol. 5: 485-49. 3. Binn, L. N., E. C. Lazar, G. A. Eddy, and M. Kajima. 197. Recovery and characterization of a minute virus of canines. Infect. Immunity 1: 53-58. 4. Hoggan, M. D. 197. Adenovirus associated viruses. Progr. Med. Virol. 1:11-39. 5. Mayr, A., P. A. Bachmann, G. Siegl, H. Mahnel, and B. E. Sheffy. 1968. Characterization of a small porcine DNA virus. Arch. Gesamte Virusforsch. 5:38-51. 6. Storz, J., and G. S. Warren. 197. Effect of antimetabolites and actinomycin D on the replication of HADEN, a bovine parvovirus. Arch. Gesamte Virusforsch. 3:71-74. 7. Toolan, H. W. 1968. The picodna viruses: H, RV, and AAV. Int. Rev. Exp. Pathol. 6:135-18. 8. Youngner, J. S. 1954. Monolayer tissue cultures. I. Preparation and standardization of suspensions of trypsin-dispersed monkey kidney cells. Proc. Soc. Exp. Biol. Med. 85:-5.