- Email: [email protected]
Plaque assay and replication of Tipula iridescent virus in Spodoptera frugiperda ovarian cells
© INSTITUTPASTEURJ~LSEVIER Paris 1994
Res. Virol. 1994, 145, 319-330
Plaque assay and replication of Tipula iridescent virus in Spodopterafrugiperda ovarian cells M. Czuba (i), S. Tajbakhsh (l. 2), T. Walker (1), M.J. Dove (1), B.F. Johnson (1, 3) and V.L. Seligy (1, 3, 4) (*) (I) Institute of Biological Sciences, National Research Council, Ottawa, (2) Departement de Biologic Moldculaire, Institute Pasteur, Paris, (3) Department of Biology, Carleton University, Ottawa, and (4) Mutagenesis Section, Environmental Health Centre, Health Canada, Ottawa
SUMMARY
A plaque assay was developed for the study of 77pula iridescent virus (TIV) replication using a cell line derived from the fall army worm Spodoptera frugiperda (Sf9). Infection and plaque formation were monitored with time by phase contrast microscopy, video and fluorescent light microscopy. Structure of virions, viroplasmic centres and organelles of infected cells were examined by transmission electron microscopy (TEM). After 4 h postinfection, plaques were visibly detected within the cell monolayer by the presence of localized cell damage and production of numerous vesicular-like cytoplasmic structures. Quantitation of virions present per A~o unit of TIV preparation was determined by TEM. The number of visible plaques corresponded to virus concentration and 1 A2s0 produced = 10s plaques. DNA hybridization analysis revealed no gross differences in genomic DNA from TIV propagated in either Sf9 cells or wax moth Galle~ ria mellonella larvae. These findings indicate that Sf9 is permissive for replication of TIV and superior by some parameters to other cell lines currently in use for the study of host cell/TIV interactions.
Key-words: Replication, Iridoviridae, TIV, Permissivity; Ovarian cells, Plaque assay, Ultrastructure.
INTRODUCTION Tipula iridescent virus (TIV), of the family Iridoviridae, was originally isolated from the dipteran Tipula p a l u d o s a (Xeros, 1954) and shown to have a broad host range, covering three insect orders: Lepidoptera, Diptera and Coleoptera (Smith et al., 1961). TIV's potential use as a
biopesticide has not been critically examined. Some aspects of its cytopathology have been studied at the cell and molecular biology levels (Mathieson and Lee, 1981; Kelly, 1985; Tajbakhsh et al., 1986; Tajbakhsh and Seligy, 1990; Tajbakhsh et al., 1990a,b; Jahagirdar and Seligy, 1991). TIV derived from North America is icosahedral in shape and contains =25-30 proteins, and
Submitted January 18, 1992, accepted April 23, 1994. (*) Correspondingauthor: Dr. V.L. Seligy, Environmental& OccupationalToxicologyDivision,EnvironmentalHealth Centre, Health Canada,Tunney'sPasture,Ottawa,Ontario,KIA 0L2.
320
M. CZUBA ET AL.
two d o u b l e - s t r a n d e d D N A c o m p o n e n t s , L (178 kbp) and $1 (10.8 kbp) (Tajbakhsh et al., 1986; Tajbakhsh and Seligy, 1990). TIV propagates well in larvae of the wax moth Galleria mellonella, but poorly in cell lines such as those derived from the salt marsh caterpillar Estigmene acrea, BTI-EAA (Tajbakhsh and Seligy, 1990). Attempts to plaque TIV for molecular studies using monolayer cultures of the C6/36 cell line from mosquito A e d e s albopictus, were unsuccessful and shown to be semipermissive for TIV replication (Tajbakhsh et al., 1990a). Since insect iridovirus type 22 (IV22), a TIV-related virus (Cameron, 1990), replicates in cells of A. albopictus as well as in S]9 ceils derived from ovarian tissue of the fall army worm Spodoptera frugiperda (Kelly, 1976; B r o w n et al., 1977), we investigated the use of the latter to plaque TIV. Here we show that Sf9 ceils are permissive for TIV propagation and exhibit pronounced cytopathic features which precede plaque formation and should be useful for characterization of host cell/virns interactions.
MATERIALS AND METHODS Insect cell culture Fall army worm SJ9 cells were obtained from ATCC and cultured at 28°C with TNM-FI-I composed of Grace's insect tissue culture medium (Gibco) with 20 lxg/ml yeastolate (Dffco), 20 lxg/ml lactalbumin hydrolysate (Difco), 50 l~g/ml garamycin and 10% foetal bovine serum (FBS, from Oxoid) at pH 6.25. Passaging of cells was done every three days by disruption of the monolayer and transfer of cells into fresh medium at a density of 4.25 x 106 cells/ml. For the plaque assay, ceils were grown in 3.8-cm wells Nunc-Gibco). Growth curves and cell dilutions for TIV infection were derived by adapting
c.p.m. FBS m.o.i. nt PB PBS PCR
= = = = = = =
counts per minute. foetal bovine serum. multiplicity o f infection. nucleotide. phosphate buffer. phosphate-buffered saline. polymerase chain reaction.
microscopic methods (Hall et al., 1935) for counting in Sedgewick-Rafter chambers. Accuracy per count in these chambers is better than 5 % if the count exceeds 500. Infection
with TIV
and
plaque
formation
Bulk stocks of TIV were prepared from infected G. mellonella larvae as described earlier (Tajbakhsh et al., 1986). Monolayers of approximately 2 x 107 log-phase cells (48-h cultures) were infected using 1-ml overlays of TIV in TNM-FI-I medium without FBS at concentrations ranging from 1 pg to 360 I.tg of virus. Concentrations of fresh and frozen stocks of TIV in phosphate-buffered saline (PBS) pH 6.8 were estimated by measuring the optical density at 260 nm after suspending aliquots of TIV in a solution containing 2% sodium dodocyl sulphate (SDS) in 10 mM Tris-HCl and 1 mM EDTA at pH 8.0. One A 2 0 unit o f TIV yields a p p r o x i m a t e l y 11.0+'0.~ ttg of TIV DNA which corresponds to < 4 . 6 x 10 l° virus particles/lxg TIV (assuming a genome size =200 kbp, Tajbakhsh et al., 1986). Controls consisted of overlays without virus. To reduce the possibility of TIV aggregation, each overlay was vortexed immediately prior to coating the cell monolayers. Overlays were removed 1.5 to 3 h later by aspiration and immediated replacely with 2 ml of 1.5 % agarose (Sea Plaque, FMC Corporation, Marine Colloids Div., Rockland). The agarose was dissolved in TNM-FH (minus FBS) with heat and immediately mixed with TNM-FH at a ratio of 1:2 before cooling to 33°C and application. After the agarose solidified, 2 ml of TNM-FH were added, and each dish was sealed with parafilm. Ceils were incubated for up to 5 days without disturbance. Plaques were photographed with and without addition of contrasting solution (0.01% neutral red). As an alternate approach, cells were grown to confluency and infected with 100 pg of virus without agarose. After incubation of plates for 3 to 5 days at 280C, plaques were visualized by staining monolayers with neutral red or " H o e c h s t 33258" dye (1 ~tg/ml PBS, American Hoechst Corp., USA). In some experiments, imprints of cell surfaces containing plaques were made by absorption to nylon filters
PFU p.i. SDS
TEM TIC VC
= plaque-formingunit. = post-infection. = sodiumdodecylsulphate. = transmissionelectronmicroscopy. = Tipula iridescentvirus. = viroplasmiccentre.
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
(2.5 cm dia., Zeta-probe, Biorad). These filters were subjected to DNA-hybridization analysis using the capsid gene as probe (see "DNA purification and analysis").
Virus particle counts To relate virus particle concentration to OD260 measurements, TIV particles in freshly prepared TIV stocks were counted using the loop-drop phosphotungstate (PTA) negative staining technique (Watson et al., 1963) with - 0 . 2 × the reported concentration of PTA to stabilize the carbon-formvarcovered grids. A suspension of latex particles (Marivae) of known mean diameter (0.481 btm) and concentration (1.64× I0 t° m1-1) was used to obtain relative counts at electron microscope magnifications from 8,000 to 25,000. Ratios of virus particles to latex particles were established by counting > 1,000 particles from uniformly dense regions of grids. Morphology of Sf9 cells and TIV To detect the appearance of viroplasmic centres (VC) or inclusion bodies in cells, TIV-infected cells were resuspended in 2 ml of 95 % methanol precooled to -20°C, and fixed for 8 min at -20°C. Aliquots of 200 gl of fixed cells (=5×105 cells) were applied to coverslips and allowed to air-dry for 20 min. The fixed cells were washed in PBS. Cellular DNA was stained with Hoechst dye in PBS for 2 rain. Cells were then washed twice for 5 rain with PBS, mounted in 0.1% p-phenylene diamine in 50 % glycerol/PBS and examined with a "Reichert Zetopan" photomicroscope equipped with epifluorescence. This assay was also used routinely to determine the percentage of infected cells. Video microscopy was carried out by time-lapse photography using a "Nikon IM35" inverted microscope adapted to a "Sony" video recorder system. Continuous records were made of plaque formation, from the time of initial virus contact and infection, up to vesicle formation and ceil monolayer disruption. For electron microscopy, ceils from virusinfected and uninfected Sf9 cells were initially fixed on ice for 2 h with 3 % glutaraldehyde in a solution containing equal parts of TNM-FH and 50 mM p h o s p h a t e b u f f e r (PB) pH 6.8. Cells were thoroughly washed and then post-fixed for 1 h on ice in 1% osmium tetroxide in 25 mM PB. After dehydration in acetone and embedding in Epon, sections were cut with a diamond knife, stained for 30 min with uranyl acetate and 3 rain with lead citrate, before viewing in a "Phillips 420" electron microscope.
321
DNA purification and analysis L and S1 DNA components were coextracted from virus isolated from infected G. melloneUa larvae and Sf9 cells according to Tajbakhsh et al. (1990b). DNA samples from infected and uninfected Sf9 cells were digested with EcoRI, Hindlll and SalI restriction endonucleases (New England Biolabs, USA) at 5 units/Bg of DNA, fractionated by electrophoresis in 0.75 % agarose and transferred to nylon membranes (Biotrans, ICN) for hybridization analysis as described elsewhere (Tajbakhsh et al., 1986). A cloned 2.5-kbp fragment (HindlIl fragment 26b of L DNA, Tajbakhsh and Seligy, 1990; Tajbakhsh et aL, 1990b) containing the capsid gene, and a 0.75-kbp EcoRI fragment cloned from $1 DNA (Tajbakhsh et al., 1986) were used to distinguish L and $1 TIV DNA components. These fragments were purified from plasmid DNA by agarose gel e l e c t r o p h o r e s i s using " N A - 4 5 " paper (Schleicher and Schuell) as previously described (Tajbakhsh et al., 1990b). For plaque detection, a 372-nucleotide (nt) probe of the capsid gene was made by PCR amplification (Cetus-Perkin Elmer 9600) using primers located at codons 121 to 128 (TV-121, 5'-GGTTGCAGCTCGATTrGATAAC3') and codons 248 to 255 (TV-248 reverse 5'A C G A T G G C A T A C T T A G C C C A T A C - 3 ' ) (Tajb a k h s h et al., 1990b). D N A p r o b e s were 32p-labelled (specific activity, --109 c.p.m./~tg) using T4 DNA polymerase and random primers (Tajbakhsh et aL, 1990b). DNA filters were incubated in plastic freezer bags containing pre-bybridization solution (1 M NaCI, 1%,SDS, 100 lxg/ml denatured salmon sperm DNA) for at least 1 h at 65°C before replacement of the solution with one containing heat-denatured, radiolabelled capsid gene or SI DNA (105 to 2 x 106 cpngml). Hybridization was for 16 h at 65°C. Washes of filters and autoradiography were carried out as described earlier (Tajbakhsh et al., 1990b).
RESULTS Morphological properties of TIV-infected S.f9 cells Starting with = 5 . 9 × 106 o f S f 9 cells/plate, confluent monolayers o f - - 3 . 0 × 107 cells were produced in 4 days of culture. With these conditions and TIV-agarose overlays, plaques w e r e first d e t e c t a b l e in 4 - 6 h and r o u t i n e l y s e e n within 3 to 4 days post-infection (p.i.). At low magnification ( 1 0 × ) , the plaques appeared as
322
M. CZUBA ET AL.
0.03 to 2-mm diameter patches scattered within the cell monolayer. The plaques could be visually enumerated but were generally difficult to photograph. Plaque localizations could be partly detected by DNA hybridization and autoradiogr a p h y u s i n g the T I V c a p s i d g e n e as p r o b e (fig. 1A-D). However, this procedure was diffic u l t to r e p r o d u c e at m . o . i . < 10 a n d > 10 3 (TIV/cell) because hybridization signals were similar to what is seen with u n i n f e c t e d cells (fig. 1A, B) or were smeared together. Accurate quantification of plaques and information on early stages o f their formation were obtained using video microscopy and phase contrast light microscopy. At all times, uninfected St'9 cells remained as tight monolayers (fig. 1E). In contrast, infected cells at the site of a plaque (fig. 1F-I) appeared in disarray. The plaques were characterized by the appearance o f large numbers of cell-derived vesicles, ranging from 3 to 8 ~tm in diameter, as early as 4 h after infection. At 8 h p.i., infection production of these vesicles, loss of cell adhesion and formation of giant cells by way of cell-cell fusions (fig. 1G and H) were distinct features of the infection process. By 16 h p.i., large clearing zones in the cell monolayer were formed (fig. lI). As shown by phase contrast microscopy and by Hoechst 33258 dye binding to DNA within uninfected (fig. 2A, B) and infected cells (fig. 2C-H), infected Sf9 cells produced single or multiple VC o f variable size as well as giant, binucleated fused cells. These features are also typical of infected cells of E. acrea and A. albopictus (Tajbakhsh et al., 1990a).
A further comparison of morphological features o f uninfected and infected cells was carried out using transmission electron microscopy (TEM). The micrographs in figures 3A and 3B show nuclear and cytoplasmic structures typical o f u n i n f e c t e d cells. Nuclear h e t e r o c h r o m a t i n was generally dispersed and mitochondria had distinct cristae. Vesicles and membrane "blebb i n g " w e r e o b s e r v e d within the control cell population at low frequency. These structures m a y be related to the pseudopods o f various shapes often concentrated at cell-cell junctions (see arrows, fig. 3A and B). Infected cells produced cell blebs and pseudopodia (fig. 3C) but not as much as seen by phase contrast (fig. 1 EI). This is probably due to the loss o f these vesicles by the p r e t r e a t m e n t o f cells for T E M . T E M micrographs o f infected S f 9 cell nuclei (fig. 4) exhibited little or no change in amount or distribution o f heterochromatin in comparison to control cells or TIV-infected E. acrea cells, which showed significant decondensation of chromatin (Lee and Brownrigg, 1982). However, i n f e c t e d S f 9 c e l l s e x h i b i t e d u n u s u a l sequestering o f mitochondria, usually between VC and nuclei. The morphology o f these mitochondria was altered in terms o f size, shape, a b s e n c e o f c i s t e r n a e and i n c r e a s e d staining (fig. 3 and 4A). This putative organelle damage is supported by observations o f a substantial loss in metabolic activity which occurred by about 8 hours p.i., as indicated by significant decreases in trehalase activity (Jahagirdar and Seligy, 1991) and incorporation o f 35S-methio-
Fig. 1. TIV plaques in Sf9 cell monolaycrs. Monolayers of S.f9 cells were exposed to an overlay containing 1.5 % "Sea Plaque" agarose, culture medium and TIV (see "Materials and Methods"). A-D=autoradiographs of TIV capsid-gene hybridizations to nylon membrane imprints made from monolayers of ~-2x 107 uninfected (A) and infected cells at m.o.i.--3 (B), 102 (C), and 103 (D), for 3 days p.i. E=control cell morphology (3 days), indicating tightly packed monolayer with some perpendicular cell groupings. F=eells after 4 h p.i. (m.o.i.--3), showing formation of gap in monolaycr and appearance of variably sized "vesicles" which indicates earliest stage of plaque formation. G, H=eeUs at 8 h p.i., presence of large numbers of microvesieles ((3) and formation of giant fusion cells (H) typify clearing zones at location of a plaque. I: at 16 h p.i. plaque boundaries are delimited by large clearing zones, of "blebbing" cells and cellular debris. Mag.= 180×.
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
323
324
M. CZUBA E T AL.
Fig. 2. Detection of viroplasmic centres in TIV-infected Sf9 cells. S/9 ceils (Mag.=960x) viewed under phase contrast (A, C, E, G) or fluorescence optics (Zeiss) after Hoechst 33258 dye treatment to localize cell nuclei (N) and viroplasmic centres (VC) (B, D, F, I-I). Control Sf9 cells (A) and nuclei (13). C=single St9 infected cells (m.o.i.= 1) with extended pseudopods and large viroplasmic centre (D). E = a pair of infected Sf9 cells with multiple VC (F). G=cells in possible fusion after infection showing multiple VC (H). Arrows indicate proximity of nuclei and VC.
nine and 3H-thymidine into Sf9 cell proteins and DNA, respectively (S. Tajbakhsh, data not shown). Analysis of several micrographs indicated that TIV vixions were produced and that they tended to cluster within the VC cytoplasm (fig. 3C and 4A-C) rather than in the electrondense VC matrix. The variation in morphology and staining of virions within clusters (fig. 4B and C) is consistent with them being empty and full virions (Tajbakhsh et al., 1986).
Efficiency of TIV infection Production of plaques, or plaque-forming units (PFU) as a function of TIV DNA content in purified virions was found to be approximately linear (fig. 5) at m.o.i.=0.1-103. Under such conditions, the enumeration of PFU determined by using cytopathic morphology indicators such as production of cell blebbing and staining with "Hoechst 33258" dye were reli-
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
..'
325
- ,~,. , .--i;
e
8
Fig. 3. TEM of S/9 cells. Panel A shows normal nucleus, even distribution of mitochondria discernible cristae and cytoplasmic peripheral "blebs", indicated by arrow (bar=l.5 ttm). B=magnification of membranous structures formed at interface layer between two cells (bar= 1.2 gin). C=high magnification of a 12-h infected cell (condition as described in fig. 1) showing excessive blebbing of cytoplasm (bar=0.5 ltm).
able. F r o m these experiments as well as use o f D N A - D N A hybridization (fig. 1A-D), one A260 unit o f T I V was e s t i m a t e d to g e n e r a t e --10 5 PFU. As an a l t e r n a t e m e t h o d f o r e s t i m a t i n g
T I V virion n u m b e r based on T I V D N A c o n tent, estimates o f TIV-virion c o n t e n t per A26o unit were carried out using 0.412 lxm diameter l a t e x b e a d s as r e f e r e n c e m a r k e r s f o r T E M .
326
M. CZUBA E T AL.
Fig. 4. TEM of TIV-infected Sf9 ceils. Cells were harvested 12 h after infection. A=4,300x magnification showing morphology of a cell resulting from a TIV-induced fusion of two cells; note presence of two nuclei (N), large crescent-shaped viroplasmie centre (VC), aggregation of morphologically altered and deeply stained mitochondria (M) and putative blebbed vesicles (B); neighbouring cells are at different stages of disorganization due to infection." B =24,000x enlargement, showing characteristic "honeycomb" clustering of TIV particles in a VC cytoplasm. C=45,000x enlargement, showing a cluster of TIV particles at different stages of maturation: full (light stain) or empty (dark stain) capsids.
Estimates obtained by using three different ratios o f latex b e a d and T I V r a n g e d f r o m 1.8× 1011 to 2.7× 1011 viral particles/A260 unit. This is about 4 to 6 times the number of particles estimated by TIV genomic DNA content. This result is not unexpected because m a n y particles contain little or no DNA (Tajbakhsh et al., 1986). As determined by Hoechst 33258 dye staining of VC p r e s e n t in i n d i v i d u a l l y
i n f e c t e d S f 9 cells (fig. 2), a p p r o x i m a t e l y 24.8 % o f the cells in c o n f l u e n t plates were infected at an m.o.i. > 100. This amount could increase to 30-40 % when freshly prepared TIV was u s e d at the s a m e A 2 6 0 c o n c e n t r a t i o n c o m p a r e d to T I V stock that was frozen and t h a w e d at least once. Based on 11 separate experiments of this type, we estimate that 0.1 to 0.3 % of TIV particles actually leads to pro-
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS 10 a
¢/}
102
°m e-
t~
°~t-" E
101
/ ~ l
O LL ID -1 t:/" 13-
I
10 ° ' ~
10.1 ......... 10 s 109 TIV
........ | ....... .I ........ i 10 r
10 a
109
Particles/Plate
Fig. 5. Plaque formation (PFU) as a function of TIV concentration. Plaque assay was carried out u s i n g d i l u t i o n s of 2.26x 10 I° TIV particlesdml and ceils as described in legend of figure 1 and "Materials and Methods". Data are from freshly prepared (triangles) and 1 × frozen (circles) TIV stocks.
327
tion with EcoR1, Hindlll and SalI fragments of TIV DNA. Only L DNA hybridized with the capsid gene located on the 2.47-kbp HindIII fragment, referred to as 26b, and the 0.75-kbp EcoR1 cloned sequence from S1 DNA recognized the two sequences which are known to map on L DNA and S1 DNA, respectively. As found earlier using two other cell lines (Tajbakhsh et al., 1989; Tajbakhsh and Sehgy, 1990), the S1 DNA component is produced in Sf9 cells at levels =5 to 10 times higher than that found in TIN particles obtained from wax moth larvae. The mechanism by which this overproduction takes place in culturod cells is not known; it may be related to replication defects or packaging of TIV DNA. PartiaUy assembled particles were found to contain more S1 than L DNA (Tajbakhsh et al., 1986). These observations collectively indicate that there are no gross differences in the organization of these two sources of TIV DNA. In addition, we have found that de novo synthesis of the capsid protein, as monitored by in vivo labelling with 35S-methionine and Western blot analysis with anti-capsid antibody revealed the presence of a 46-kDa polypeptide which corresponds to mature capsid protein (S. Tajbaldash, unpublished data). The expression of capsid protein was found to be in yields comparable to or better than those from E. acrea cells (Tajbakhsh et aL, 1~990b).
DISCUSSION
duction of visible VC and that < 1% of these i n f e c t i o n s l e a d s to p r o d u c t i o n o f v i s i b l e plaques (= 10-5). Molecular analysis Total DNA extracted from TIV virions propagated in G. melloneUa larvae and infected Sf9 cells were compared by restriction endonuclease mapping and DNA-DNA hybridization analysis. As shown in figure 6, the TIV capsid gene (specific only for L DNA) and a 0.75-kbp cloned segment of $1 DNA, present as a 1.1-kbp E c o R I fragment in L DNA (Tajbakhsh and Seligy, 1990), gave the expected patterns of hybridiza-
Here we report for the first time the successful plaquing of TIV using Sf9 cells, as well as modifications made to methods described for use with type 22 iridovirus (Brown et al., 1977), baculovirus (Wood, 1977; Brown and Faulkner, 1978; Summers and Smith, 1988) and black beetle virus (Selling and Rueckert, 1984). In combination with changes in the preparation of the culture medium, virus and cells, and the use of visual aids to detect plaques and count virions, we show that infectivity and production of TIV are generally very low and are not practical for scale-up production of TIV from single plaques. This is a feature common to insect iridoviruses in general, and as a consequence, production in insect larvae
328
M. CZUBA E T AL.
!1
Fig. 6. Comparison of TIV DNA from G. mellonella and SJ:9. DNA was harvested from TIV produced in either G. mellonella (lanes a, e, g and i) or infected SJ9 cells (lanes b, f, h and j) and uninfected SJ9 ceils (lanes c and d). DNA, 2 I.tg of undigested (lanes a, b and c) or digested (EcoR1, lanes d, e and f; HindlH, lanes g and h; and SalI, lanes i and j), were fractionated by electrophoresis in 0.75 % agarose before transfer to nylon membranes and hybridization with denatured, 32p-labelled, $1 DNA (0.75-kbp EcoRI fragment from plasmid pSE3) (panel A). This filter was then reused after removal of $1 DNA probe and hybridized with 32p-labelled, TIV-capsid gene located on a 2.47-kbp Hindlll DNA fragment, 26b, which is present only in the L DNA component (panel B).
is still the preferred method of generating large amounts of TIV. Nevertheless, Sf9 cell monolayers provide us with a potentially practical system for studying details of TIV replication and gene expression, in comparison to use of E. a c r e a ceils, which attach poorly in culture, or A. albopictus ceils, which grow as a monolayer but are semipermissive for TIV replication (Tajbakhsh et aL, I990a, 1990b). Morphological studies using various types of microscopy, combined with analysis of G. mellonella- and Sf9-derived TIV DNA, revealed no gross differences in TIV biology, including levels of production o f the $1 T I V DNA component. These results, in general, compare favorably with earlier analyses using E. acrea as a host cell (Tajbakhsh et al., 1990a). However, the microscopy studies reported here provide details concerning early events in Sf9 infection and plaque formation. One feature is the pronounced change in morphology and staining o f mitochondria, an indicator of cytotoxicity. A distinct macro-feature
which is seen in infected ceils, located within plaques early and during their formation, is the production of one or more giant ceils containing multiple nuclei and VC. Multiply fused ceils of this type have been seen in cultures of other ceil lines infected with either TIV (Tajbakhsh and Seligy, 1990) or other 'insect viruses (Cerutti and D e v a u c h e l l e , 1979). A m o r e striking feature which is seen at the earliest stage o f plaque formarion is the appearance o f large quantifies of c e l l - f r e e vesicles w i t h i n the S f 9 m o n o l a y e r (fig. 1E and 3C). The early appearance of these vesicles, demonstrated by continuous (video) microscopy, and their increase in number with time contribute significantly to the visibility of plaques, which are detectable even without staining. These vesicles vary in diameter from 3 to 8 lxm on average, and m a y or may not contain D N A f r a g m e n t s (as o b s e r v e d using H o e c h s t staining, data not shown). Attempts to characterize these vesicles by TEM proved unsuccessful, probably because these vesicles are highly mobile
REPLICATION OF T I P U L A IRIDESCENT VIRUS IN Sf9 CELLS a n d v e r y f r a g i l e . V e s i c l e s a p p e a r i n g e a r l y in infection m a y originate f r o m budding or blebbing o f the Sf9 cell m e m b r a n e as a result o f virally induced metabolic stress. However, at late stages o f infection, vesicular debris m a y contain virus particles, released b y lysis or exocytosis, as sugg e s t e d b y M a t h i e s o n and L e e (1981). F u r t h e r experimentation will be necessary to determine to w h a t e x t e n t c h a n g e s in surface structure a f f e c t S f 9 cell r e c e p t i v i t y to the virus and p r o m o t e intercellular f u s i o n and vesicle f o r m a t i o n after viral entry into cells.
Acknowledgements We thank T. Devesceri for photographic support.
Plages de lyse et r~plication de l'iridovirus Tipula dans les cellules ovariennes de Spodoptera frugiperda Un test de quantification par plages de lyse permettant l ' r t u d e de la r r p l i c a t i o n de l ' i r i d o v i r u s Tipula (TIV) a 6t6 mis au point sur une lignre cellulaire d r r i v r e de Spodoptera frugiperda (Sf9). L'infection et la formation des ptages ont 6t6 suivies par microscopie ~ contraste de phase, vidrom i c r o s c o p i c et m i c r o s c o p i e ~ f l u o r e s c e n c e . L a structure des virions, des centres viroplasmiques et des organites cellulaires infectrs a 6t6 examinre par microscopie 61ectronique ~ transmission (TEM). Quatre heures apr~s l'infection, la drtrrioration des cellules ainsi que la prrsence de nombreuses structures cytoplasmiques d'aspect vrsiculaire, permettent la visualisation de plages sur le tapis cellulaire. La quantification des virions prrsents par absorption ~t 260 n m d ' u n e p r r p a r a t i o n de T I V a 6t6 drterminre par TEM. Le nombre de plages visibles correspond ~ la concentration de virus, et 1 unit6 A 2 produit ca 105 plages L'analyse de I ' A D N p a r h y b n d a t t o n n ' a pas p e r m l s d e m e t t r e en 6vidence de d i f f r r e n c e significative entre I ' A D N g r n o m i q u e p r o v e n a n t du T I V propag6 dans la lignre Sf9 ou celui du TIV propag6 chez les larves de Galleria mellonella. Ces rrsultats indiquent que la lignre Sf9 est permissive pour la rrplication du TIV et prrsente plusieurs avantages si on la compare aux autres l i g n r e s cellulaires a c t u e l l e m e n t u t i l i s r e s p o u r l ' r t u d e des i n t e r a c t i o n s c e l l u l e hrte/TIV. 60
.
.
°
"
.
329
Mots-clds : R r p l i e a t i o n , Iridoviridae, T I V , P e r m i s s i v i t r ; Cellules o v a r i e n n e s , D o s a g e par plaques, Ultrastructure.
References Brown, D.A., Lescott, T., Harrap, K.A. & Kelly, D.C. (1977), The replication and titration of iridescent virus type 22 in Spodoptera frugiperda cells. J. Gen. ViroL, 38, 175-178. Brown, M. & Faulkner, P. (1978), Plaque assay of nuclear polyhedrosis viruses in cell culture. Appl. Environ. Microbiol., 36, 13-35. Cameron, I.R. (1990), Identification and characterization of the gene encoding the major structural protein of insect iridescent virus type 22. Virology, 178, 3542. Cerutti, M. & Devauchelle, G. (1979), Cell fusion induced by an invertebrate virus. Arch. Virol., 61, 149-155. Hall, R.P., Johnson, D.F. & Loefer, J.B. (1935), A method for counting protozoa in the measurement of growth under experimental conditions. Trans. Am. Microscop. Soc., 54, 298-300. Jahagirdar, A.P. & Seligy, V.L. (1991), Modulation of trehalase activity in two insect cell lines by virus infection and trehalose. Biochem. Inter., 23, 10491054. Kelly, D.C. (1985), Insect iridescent virus. Curr. Top. Microbiol. ImmunoL, 116, 23-25. Lee, P.E. & Brownrigg, S. (1982), Effect of virus inactivation on Tipula iridescent virus-cell relationships. J. Ultrastruct. Res., 79, 189-197. Mathieson, W.B. & Lee, P.E. (1981), C~ytologyand autoradiography of Tipula iridescent virus infection of insect suspension cell cultures. J. Ultrastruct. Res., 74, 59-68. Selling, B.H. & Rueckert, R.R. (1984), Plaque assay for black beetle virus. J. ViroL, 51, 251-253. Smith, K.M., Hills, G.J. & Rivers, C.F. (1961), Studies on the cross-inoculation of the Tipula iridescent virus. Virology, 13, 233-241. Summers, M.D. & Smith, G.E. (1988), A manual of methods for Baculovirus vectors and insect cell culture procedures. Texas Agric. Exper. St. Bull., n° 1555, 29-30. Tajbakhsh, S., Dove, M.J., Lee, P.E. & Seligy, V.L. (1986), DNA components of Tipula iridescent virus. Biochera. Cell BioL, 64, 495-503. Tajbakhsh, S., Lee, P.E. & Seligy, V.L. (1989), Comparative studies on Tipula iridescent virus DNA derived from whole insects and ceils in culture, in "Invertebrate cell systems applications". Chapter 6 (pp. 5361). CRC Press, Inc., Boca Raton, FL. Tajbakhsh, S. & Seligy, V.L. (1990), Molecular biology of Tipula iridescent virus, in "Molecular biology of iridoviruses" (G. Daral). Developments in molecular virology. Chapter 5. Martinus Nijhoff Publ., Boston, MA. Tajbakhsh, S., Kiss, G., l.e~, P.E. & Seligy, V.L. (1990a), Semipermissive replication of Tipula iridescent virus
330
M. CZUBA E T AL.
in Aedes albopictus C6/36 cells. Virology, 174, 264275. Tajbakhsh, S., Lee, P.E., Watson, D.C. & Seligy, V.L. (1990b), Molecular cloning, characterization, and expression of the Tipula iridescent virus eapsid gene. J. Virol., 64, 125-136. Watson, D.H., Russell, W.C. & Wildy, P. (1963), Electron microscopic particle counts on Herpes virus using the
phosphotungstate negative staining technique. Virology, 19, 250-260. Wood, H.A. (1977), An agar overlay plaque assay method for Autographa californica nuclear-polyhedrosis virus. J. Invert. Pathol., 29, 304-307. Xeros, N. (1954), A second viruss disease of the leatherjacket, Tipula paludosa. Nature CLond.), 174, 562563.
Res. Virol. 1994, 145, 319-330
Plaque assay and replication of Tipula iridescent virus in Spodopterafrugiperda ovarian cells M. Czuba (i), S. Tajbakhsh (l. 2), T. Walker (1), M.J. Dove (1), B.F. Johnson (1, 3) and V.L. Seligy (1, 3, 4) (*) (I) Institute of Biological Sciences, National Research Council, Ottawa, (2) Departement de Biologic Moldculaire, Institute Pasteur, Paris, (3) Department of Biology, Carleton University, Ottawa, and (4) Mutagenesis Section, Environmental Health Centre, Health Canada, Ottawa
SUMMARY
A plaque assay was developed for the study of 77pula iridescent virus (TIV) replication using a cell line derived from the fall army worm Spodoptera frugiperda (Sf9). Infection and plaque formation were monitored with time by phase contrast microscopy, video and fluorescent light microscopy. Structure of virions, viroplasmic centres and organelles of infected cells were examined by transmission electron microscopy (TEM). After 4 h postinfection, plaques were visibly detected within the cell monolayer by the presence of localized cell damage and production of numerous vesicular-like cytoplasmic structures. Quantitation of virions present per A~o unit of TIV preparation was determined by TEM. The number of visible plaques corresponded to virus concentration and 1 A2s0 produced = 10s plaques. DNA hybridization analysis revealed no gross differences in genomic DNA from TIV propagated in either Sf9 cells or wax moth Galle~ ria mellonella larvae. These findings indicate that Sf9 is permissive for replication of TIV and superior by some parameters to other cell lines currently in use for the study of host cell/TIV interactions.
Key-words: Replication, Iridoviridae, TIV, Permissivity; Ovarian cells, Plaque assay, Ultrastructure.
INTRODUCTION Tipula iridescent virus (TIV), of the family Iridoviridae, was originally isolated from the dipteran Tipula p a l u d o s a (Xeros, 1954) and shown to have a broad host range, covering three insect orders: Lepidoptera, Diptera and Coleoptera (Smith et al., 1961). TIV's potential use as a
biopesticide has not been critically examined. Some aspects of its cytopathology have been studied at the cell and molecular biology levels (Mathieson and Lee, 1981; Kelly, 1985; Tajbakhsh et al., 1986; Tajbakhsh and Seligy, 1990; Tajbakhsh et al., 1990a,b; Jahagirdar and Seligy, 1991). TIV derived from North America is icosahedral in shape and contains =25-30 proteins, and
Submitted January 18, 1992, accepted April 23, 1994. (*) Correspondingauthor: Dr. V.L. Seligy, Environmental& OccupationalToxicologyDivision,EnvironmentalHealth Centre, Health Canada,Tunney'sPasture,Ottawa,Ontario,KIA 0L2.
320
M. CZUBA ET AL.
two d o u b l e - s t r a n d e d D N A c o m p o n e n t s , L (178 kbp) and $1 (10.8 kbp) (Tajbakhsh et al., 1986; Tajbakhsh and Seligy, 1990). TIV propagates well in larvae of the wax moth Galleria mellonella, but poorly in cell lines such as those derived from the salt marsh caterpillar Estigmene acrea, BTI-EAA (Tajbakhsh and Seligy, 1990). Attempts to plaque TIV for molecular studies using monolayer cultures of the C6/36 cell line from mosquito A e d e s albopictus, were unsuccessful and shown to be semipermissive for TIV replication (Tajbakhsh et al., 1990a). Since insect iridovirus type 22 (IV22), a TIV-related virus (Cameron, 1990), replicates in cells of A. albopictus as well as in S]9 ceils derived from ovarian tissue of the fall army worm Spodoptera frugiperda (Kelly, 1976; B r o w n et al., 1977), we investigated the use of the latter to plaque TIV. Here we show that Sf9 ceils are permissive for TIV propagation and exhibit pronounced cytopathic features which precede plaque formation and should be useful for characterization of host cell/virns interactions.
MATERIALS AND METHODS Insect cell culture Fall army worm SJ9 cells were obtained from ATCC and cultured at 28°C with TNM-FI-I composed of Grace's insect tissue culture medium (Gibco) with 20 lxg/ml yeastolate (Dffco), 20 lxg/ml lactalbumin hydrolysate (Difco), 50 l~g/ml garamycin and 10% foetal bovine serum (FBS, from Oxoid) at pH 6.25. Passaging of cells was done every three days by disruption of the monolayer and transfer of cells into fresh medium at a density of 4.25 x 106 cells/ml. For the plaque assay, ceils were grown in 3.8-cm wells Nunc-Gibco). Growth curves and cell dilutions for TIV infection were derived by adapting
c.p.m. FBS m.o.i. nt PB PBS PCR
= = = = = = =
counts per minute. foetal bovine serum. multiplicity o f infection. nucleotide. phosphate buffer. phosphate-buffered saline. polymerase chain reaction.
microscopic methods (Hall et al., 1935) for counting in Sedgewick-Rafter chambers. Accuracy per count in these chambers is better than 5 % if the count exceeds 500. Infection
with TIV
and
plaque
formation
Bulk stocks of TIV were prepared from infected G. mellonella larvae as described earlier (Tajbakhsh et al., 1986). Monolayers of approximately 2 x 107 log-phase cells (48-h cultures) were infected using 1-ml overlays of TIV in TNM-FI-I medium without FBS at concentrations ranging from 1 pg to 360 I.tg of virus. Concentrations of fresh and frozen stocks of TIV in phosphate-buffered saline (PBS) pH 6.8 were estimated by measuring the optical density at 260 nm after suspending aliquots of TIV in a solution containing 2% sodium dodocyl sulphate (SDS) in 10 mM Tris-HCl and 1 mM EDTA at pH 8.0. One A 2 0 unit o f TIV yields a p p r o x i m a t e l y 11.0+'0.~ ttg of TIV DNA which corresponds to < 4 . 6 x 10 l° virus particles/lxg TIV (assuming a genome size =200 kbp, Tajbakhsh et al., 1986). Controls consisted of overlays without virus. To reduce the possibility of TIV aggregation, each overlay was vortexed immediately prior to coating the cell monolayers. Overlays were removed 1.5 to 3 h later by aspiration and immediated replacely with 2 ml of 1.5 % agarose (Sea Plaque, FMC Corporation, Marine Colloids Div., Rockland). The agarose was dissolved in TNM-FH (minus FBS) with heat and immediately mixed with TNM-FH at a ratio of 1:2 before cooling to 33°C and application. After the agarose solidified, 2 ml of TNM-FH were added, and each dish was sealed with parafilm. Ceils were incubated for up to 5 days without disturbance. Plaques were photographed with and without addition of contrasting solution (0.01% neutral red). As an alternate approach, cells were grown to confluency and infected with 100 pg of virus without agarose. After incubation of plates for 3 to 5 days at 280C, plaques were visualized by staining monolayers with neutral red or " H o e c h s t 33258" dye (1 ~tg/ml PBS, American Hoechst Corp., USA). In some experiments, imprints of cell surfaces containing plaques were made by absorption to nylon filters
PFU p.i. SDS
TEM TIC VC
= plaque-formingunit. = post-infection. = sodiumdodecylsulphate. = transmissionelectronmicroscopy. = Tipula iridescentvirus. = viroplasmiccentre.
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
(2.5 cm dia., Zeta-probe, Biorad). These filters were subjected to DNA-hybridization analysis using the capsid gene as probe (see "DNA purification and analysis").
Virus particle counts To relate virus particle concentration to OD260 measurements, TIV particles in freshly prepared TIV stocks were counted using the loop-drop phosphotungstate (PTA) negative staining technique (Watson et al., 1963) with - 0 . 2 × the reported concentration of PTA to stabilize the carbon-formvarcovered grids. A suspension of latex particles (Marivae) of known mean diameter (0.481 btm) and concentration (1.64× I0 t° m1-1) was used to obtain relative counts at electron microscope magnifications from 8,000 to 25,000. Ratios of virus particles to latex particles were established by counting > 1,000 particles from uniformly dense regions of grids. Morphology of Sf9 cells and TIV To detect the appearance of viroplasmic centres (VC) or inclusion bodies in cells, TIV-infected cells were resuspended in 2 ml of 95 % methanol precooled to -20°C, and fixed for 8 min at -20°C. Aliquots of 200 gl of fixed cells (=5×105 cells) were applied to coverslips and allowed to air-dry for 20 min. The fixed cells were washed in PBS. Cellular DNA was stained with Hoechst dye in PBS for 2 rain. Cells were then washed twice for 5 rain with PBS, mounted in 0.1% p-phenylene diamine in 50 % glycerol/PBS and examined with a "Reichert Zetopan" photomicroscope equipped with epifluorescence. This assay was also used routinely to determine the percentage of infected cells. Video microscopy was carried out by time-lapse photography using a "Nikon IM35" inverted microscope adapted to a "Sony" video recorder system. Continuous records were made of plaque formation, from the time of initial virus contact and infection, up to vesicle formation and ceil monolayer disruption. For electron microscopy, ceils from virusinfected and uninfected Sf9 cells were initially fixed on ice for 2 h with 3 % glutaraldehyde in a solution containing equal parts of TNM-FH and 50 mM p h o s p h a t e b u f f e r (PB) pH 6.8. Cells were thoroughly washed and then post-fixed for 1 h on ice in 1% osmium tetroxide in 25 mM PB. After dehydration in acetone and embedding in Epon, sections were cut with a diamond knife, stained for 30 min with uranyl acetate and 3 rain with lead citrate, before viewing in a "Phillips 420" electron microscope.
321
DNA purification and analysis L and S1 DNA components were coextracted from virus isolated from infected G. melloneUa larvae and Sf9 cells according to Tajbakhsh et al. (1990b). DNA samples from infected and uninfected Sf9 cells were digested with EcoRI, Hindlll and SalI restriction endonucleases (New England Biolabs, USA) at 5 units/Bg of DNA, fractionated by electrophoresis in 0.75 % agarose and transferred to nylon membranes (Biotrans, ICN) for hybridization analysis as described elsewhere (Tajbakhsh et al., 1986). A cloned 2.5-kbp fragment (HindlIl fragment 26b of L DNA, Tajbakhsh and Seligy, 1990; Tajbakhsh et aL, 1990b) containing the capsid gene, and a 0.75-kbp EcoRI fragment cloned from $1 DNA (Tajbakhsh et al., 1986) were used to distinguish L and $1 TIV DNA components. These fragments were purified from plasmid DNA by agarose gel e l e c t r o p h o r e s i s using " N A - 4 5 " paper (Schleicher and Schuell) as previously described (Tajbakhsh et al., 1990b). For plaque detection, a 372-nucleotide (nt) probe of the capsid gene was made by PCR amplification (Cetus-Perkin Elmer 9600) using primers located at codons 121 to 128 (TV-121, 5'-GGTTGCAGCTCGATTrGATAAC3') and codons 248 to 255 (TV-248 reverse 5'A C G A T G G C A T A C T T A G C C C A T A C - 3 ' ) (Tajb a k h s h et al., 1990b). D N A p r o b e s were 32p-labelled (specific activity, --109 c.p.m./~tg) using T4 DNA polymerase and random primers (Tajbakhsh et aL, 1990b). DNA filters were incubated in plastic freezer bags containing pre-bybridization solution (1 M NaCI, 1%,SDS, 100 lxg/ml denatured salmon sperm DNA) for at least 1 h at 65°C before replacement of the solution with one containing heat-denatured, radiolabelled capsid gene or SI DNA (105 to 2 x 106 cpngml). Hybridization was for 16 h at 65°C. Washes of filters and autoradiography were carried out as described earlier (Tajbakhsh et al., 1990b).
RESULTS Morphological properties of TIV-infected S.f9 cells Starting with = 5 . 9 × 106 o f S f 9 cells/plate, confluent monolayers o f - - 3 . 0 × 107 cells were produced in 4 days of culture. With these conditions and TIV-agarose overlays, plaques w e r e first d e t e c t a b l e in 4 - 6 h and r o u t i n e l y s e e n within 3 to 4 days post-infection (p.i.). At low magnification ( 1 0 × ) , the plaques appeared as
322
M. CZUBA ET AL.
0.03 to 2-mm diameter patches scattered within the cell monolayer. The plaques could be visually enumerated but were generally difficult to photograph. Plaque localizations could be partly detected by DNA hybridization and autoradiogr a p h y u s i n g the T I V c a p s i d g e n e as p r o b e (fig. 1A-D). However, this procedure was diffic u l t to r e p r o d u c e at m . o . i . < 10 a n d > 10 3 (TIV/cell) because hybridization signals were similar to what is seen with u n i n f e c t e d cells (fig. 1A, B) or were smeared together. Accurate quantification of plaques and information on early stages o f their formation were obtained using video microscopy and phase contrast light microscopy. At all times, uninfected St'9 cells remained as tight monolayers (fig. 1E). In contrast, infected cells at the site of a plaque (fig. 1F-I) appeared in disarray. The plaques were characterized by the appearance o f large numbers of cell-derived vesicles, ranging from 3 to 8 ~tm in diameter, as early as 4 h after infection. At 8 h p.i., infection production of these vesicles, loss of cell adhesion and formation of giant cells by way of cell-cell fusions (fig. 1G and H) were distinct features of the infection process. By 16 h p.i., large clearing zones in the cell monolayer were formed (fig. lI). As shown by phase contrast microscopy and by Hoechst 33258 dye binding to DNA within uninfected (fig. 2A, B) and infected cells (fig. 2C-H), infected Sf9 cells produced single or multiple VC o f variable size as well as giant, binucleated fused cells. These features are also typical of infected cells of E. acrea and A. albopictus (Tajbakhsh et al., 1990a).
A further comparison of morphological features o f uninfected and infected cells was carried out using transmission electron microscopy (TEM). The micrographs in figures 3A and 3B show nuclear and cytoplasmic structures typical o f u n i n f e c t e d cells. Nuclear h e t e r o c h r o m a t i n was generally dispersed and mitochondria had distinct cristae. Vesicles and membrane "blebb i n g " w e r e o b s e r v e d within the control cell population at low frequency. These structures m a y be related to the pseudopods o f various shapes often concentrated at cell-cell junctions (see arrows, fig. 3A and B). Infected cells produced cell blebs and pseudopodia (fig. 3C) but not as much as seen by phase contrast (fig. 1 EI). This is probably due to the loss o f these vesicles by the p r e t r e a t m e n t o f cells for T E M . T E M micrographs o f infected S f 9 cell nuclei (fig. 4) exhibited little or no change in amount or distribution o f heterochromatin in comparison to control cells or TIV-infected E. acrea cells, which showed significant decondensation of chromatin (Lee and Brownrigg, 1982). However, i n f e c t e d S f 9 c e l l s e x h i b i t e d u n u s u a l sequestering o f mitochondria, usually between VC and nuclei. The morphology o f these mitochondria was altered in terms o f size, shape, a b s e n c e o f c i s t e r n a e and i n c r e a s e d staining (fig. 3 and 4A). This putative organelle damage is supported by observations o f a substantial loss in metabolic activity which occurred by about 8 hours p.i., as indicated by significant decreases in trehalase activity (Jahagirdar and Seligy, 1991) and incorporation o f 35S-methio-
Fig. 1. TIV plaques in Sf9 cell monolaycrs. Monolayers of S.f9 cells were exposed to an overlay containing 1.5 % "Sea Plaque" agarose, culture medium and TIV (see "Materials and Methods"). A-D=autoradiographs of TIV capsid-gene hybridizations to nylon membrane imprints made from monolayers of ~-2x 107 uninfected (A) and infected cells at m.o.i.--3 (B), 102 (C), and 103 (D), for 3 days p.i. E=control cell morphology (3 days), indicating tightly packed monolayer with some perpendicular cell groupings. F=eells after 4 h p.i. (m.o.i.--3), showing formation of gap in monolaycr and appearance of variably sized "vesicles" which indicates earliest stage of plaque formation. G, H=eeUs at 8 h p.i., presence of large numbers of microvesieles ((3) and formation of giant fusion cells (H) typify clearing zones at location of a plaque. I: at 16 h p.i. plaque boundaries are delimited by large clearing zones, of "blebbing" cells and cellular debris. Mag.= 180×.
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
323
324
M. CZUBA E T AL.
Fig. 2. Detection of viroplasmic centres in TIV-infected Sf9 cells. S/9 ceils (Mag.=960x) viewed under phase contrast (A, C, E, G) or fluorescence optics (Zeiss) after Hoechst 33258 dye treatment to localize cell nuclei (N) and viroplasmic centres (VC) (B, D, F, I-I). Control Sf9 cells (A) and nuclei (13). C=single St9 infected cells (m.o.i.= 1) with extended pseudopods and large viroplasmic centre (D). E = a pair of infected Sf9 cells with multiple VC (F). G=cells in possible fusion after infection showing multiple VC (H). Arrows indicate proximity of nuclei and VC.
nine and 3H-thymidine into Sf9 cell proteins and DNA, respectively (S. Tajbakhsh, data not shown). Analysis of several micrographs indicated that TIV vixions were produced and that they tended to cluster within the VC cytoplasm (fig. 3C and 4A-C) rather than in the electrondense VC matrix. The variation in morphology and staining of virions within clusters (fig. 4B and C) is consistent with them being empty and full virions (Tajbakhsh et al., 1986).
Efficiency of TIV infection Production of plaques, or plaque-forming units (PFU) as a function of TIV DNA content in purified virions was found to be approximately linear (fig. 5) at m.o.i.=0.1-103. Under such conditions, the enumeration of PFU determined by using cytopathic morphology indicators such as production of cell blebbing and staining with "Hoechst 33258" dye were reli-
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS
..'
325
- ,~,. , .--i;
e
8
Fig. 3. TEM of S/9 cells. Panel A shows normal nucleus, even distribution of mitochondria discernible cristae and cytoplasmic peripheral "blebs", indicated by arrow (bar=l.5 ttm). B=magnification of membranous structures formed at interface layer between two cells (bar= 1.2 gin). C=high magnification of a 12-h infected cell (condition as described in fig. 1) showing excessive blebbing of cytoplasm (bar=0.5 ltm).
able. F r o m these experiments as well as use o f D N A - D N A hybridization (fig. 1A-D), one A260 unit o f T I V was e s t i m a t e d to g e n e r a t e --10 5 PFU. As an a l t e r n a t e m e t h o d f o r e s t i m a t i n g
T I V virion n u m b e r based on T I V D N A c o n tent, estimates o f TIV-virion c o n t e n t per A26o unit were carried out using 0.412 lxm diameter l a t e x b e a d s as r e f e r e n c e m a r k e r s f o r T E M .
326
M. CZUBA E T AL.
Fig. 4. TEM of TIV-infected Sf9 ceils. Cells were harvested 12 h after infection. A=4,300x magnification showing morphology of a cell resulting from a TIV-induced fusion of two cells; note presence of two nuclei (N), large crescent-shaped viroplasmie centre (VC), aggregation of morphologically altered and deeply stained mitochondria (M) and putative blebbed vesicles (B); neighbouring cells are at different stages of disorganization due to infection." B =24,000x enlargement, showing characteristic "honeycomb" clustering of TIV particles in a VC cytoplasm. C=45,000x enlargement, showing a cluster of TIV particles at different stages of maturation: full (light stain) or empty (dark stain) capsids.
Estimates obtained by using three different ratios o f latex b e a d and T I V r a n g e d f r o m 1.8× 1011 to 2.7× 1011 viral particles/A260 unit. This is about 4 to 6 times the number of particles estimated by TIV genomic DNA content. This result is not unexpected because m a n y particles contain little or no DNA (Tajbakhsh et al., 1986). As determined by Hoechst 33258 dye staining of VC p r e s e n t in i n d i v i d u a l l y
i n f e c t e d S f 9 cells (fig. 2), a p p r o x i m a t e l y 24.8 % o f the cells in c o n f l u e n t plates were infected at an m.o.i. > 100. This amount could increase to 30-40 % when freshly prepared TIV was u s e d at the s a m e A 2 6 0 c o n c e n t r a t i o n c o m p a r e d to T I V stock that was frozen and t h a w e d at least once. Based on 11 separate experiments of this type, we estimate that 0.1 to 0.3 % of TIV particles actually leads to pro-
REPLICATION OF TIPULA IRIDESCENT VIRUS IN Sf9 CELLS 10 a
¢/}
102
°m e-
t~
°~t-" E
101
/ ~ l
O LL ID -1 t:/" 13-
I
10 ° ' ~
10.1 ......... 10 s 109 TIV
........ | ....... .I ........ i 10 r
10 a
109
Particles/Plate
Fig. 5. Plaque formation (PFU) as a function of TIV concentration. Plaque assay was carried out u s i n g d i l u t i o n s of 2.26x 10 I° TIV particlesdml and ceils as described in legend of figure 1 and "Materials and Methods". Data are from freshly prepared (triangles) and 1 × frozen (circles) TIV stocks.
327
tion with EcoR1, Hindlll and SalI fragments of TIV DNA. Only L DNA hybridized with the capsid gene located on the 2.47-kbp HindIII fragment, referred to as 26b, and the 0.75-kbp EcoR1 cloned sequence from S1 DNA recognized the two sequences which are known to map on L DNA and S1 DNA, respectively. As found earlier using two other cell lines (Tajbakhsh et al., 1989; Tajbakhsh and Sehgy, 1990), the S1 DNA component is produced in Sf9 cells at levels =5 to 10 times higher than that found in TIN particles obtained from wax moth larvae. The mechanism by which this overproduction takes place in culturod cells is not known; it may be related to replication defects or packaging of TIV DNA. PartiaUy assembled particles were found to contain more S1 than L DNA (Tajbakhsh et al., 1986). These observations collectively indicate that there are no gross differences in the organization of these two sources of TIV DNA. In addition, we have found that de novo synthesis of the capsid protein, as monitored by in vivo labelling with 35S-methionine and Western blot analysis with anti-capsid antibody revealed the presence of a 46-kDa polypeptide which corresponds to mature capsid protein (S. Tajbaldash, unpublished data). The expression of capsid protein was found to be in yields comparable to or better than those from E. acrea cells (Tajbakhsh et aL, 1~990b).
DISCUSSION
duction of visible VC and that < 1% of these i n f e c t i o n s l e a d s to p r o d u c t i o n o f v i s i b l e plaques (= 10-5). Molecular analysis Total DNA extracted from TIV virions propagated in G. melloneUa larvae and infected Sf9 cells were compared by restriction endonuclease mapping and DNA-DNA hybridization analysis. As shown in figure 6, the TIV capsid gene (specific only for L DNA) and a 0.75-kbp cloned segment of $1 DNA, present as a 1.1-kbp E c o R I fragment in L DNA (Tajbakhsh and Seligy, 1990), gave the expected patterns of hybridiza-
Here we report for the first time the successful plaquing of TIV using Sf9 cells, as well as modifications made to methods described for use with type 22 iridovirus (Brown et al., 1977), baculovirus (Wood, 1977; Brown and Faulkner, 1978; Summers and Smith, 1988) and black beetle virus (Selling and Rueckert, 1984). In combination with changes in the preparation of the culture medium, virus and cells, and the use of visual aids to detect plaques and count virions, we show that infectivity and production of TIV are generally very low and are not practical for scale-up production of TIV from single plaques. This is a feature common to insect iridoviruses in general, and as a consequence, production in insect larvae
328
M. CZUBA E T AL.
!1
Fig. 6. Comparison of TIV DNA from G. mellonella and SJ:9. DNA was harvested from TIV produced in either G. mellonella (lanes a, e, g and i) or infected SJ9 cells (lanes b, f, h and j) and uninfected SJ9 ceils (lanes c and d). DNA, 2 I.tg of undigested (lanes a, b and c) or digested (EcoR1, lanes d, e and f; HindlH, lanes g and h; and SalI, lanes i and j), were fractionated by electrophoresis in 0.75 % agarose before transfer to nylon membranes and hybridization with denatured, 32p-labelled, $1 DNA (0.75-kbp EcoRI fragment from plasmid pSE3) (panel A). This filter was then reused after removal of $1 DNA probe and hybridized with 32p-labelled, TIV-capsid gene located on a 2.47-kbp Hindlll DNA fragment, 26b, which is present only in the L DNA component (panel B).
is still the preferred method of generating large amounts of TIV. Nevertheless, Sf9 cell monolayers provide us with a potentially practical system for studying details of TIV replication and gene expression, in comparison to use of E. a c r e a ceils, which attach poorly in culture, or A. albopictus ceils, which grow as a monolayer but are semipermissive for TIV replication (Tajbakhsh et aL, I990a, 1990b). Morphological studies using various types of microscopy, combined with analysis of G. mellonella- and Sf9-derived TIV DNA, revealed no gross differences in TIV biology, including levels of production o f the $1 T I V DNA component. These results, in general, compare favorably with earlier analyses using E. acrea as a host cell (Tajbakhsh et al., 1990a). However, the microscopy studies reported here provide details concerning early events in Sf9 infection and plaque formation. One feature is the pronounced change in morphology and staining o f mitochondria, an indicator of cytotoxicity. A distinct macro-feature
which is seen in infected ceils, located within plaques early and during their formation, is the production of one or more giant ceils containing multiple nuclei and VC. Multiply fused ceils of this type have been seen in cultures of other ceil lines infected with either TIV (Tajbakhsh and Seligy, 1990) or other 'insect viruses (Cerutti and D e v a u c h e l l e , 1979). A m o r e striking feature which is seen at the earliest stage o f plaque formarion is the appearance o f large quantifies of c e l l - f r e e vesicles w i t h i n the S f 9 m o n o l a y e r (fig. 1E and 3C). The early appearance of these vesicles, demonstrated by continuous (video) microscopy, and their increase in number with time contribute significantly to the visibility of plaques, which are detectable even without staining. These vesicles vary in diameter from 3 to 8 lxm on average, and m a y or may not contain D N A f r a g m e n t s (as o b s e r v e d using H o e c h s t staining, data not shown). Attempts to characterize these vesicles by TEM proved unsuccessful, probably because these vesicles are highly mobile
REPLICATION OF T I P U L A IRIDESCENT VIRUS IN Sf9 CELLS a n d v e r y f r a g i l e . V e s i c l e s a p p e a r i n g e a r l y in infection m a y originate f r o m budding or blebbing o f the Sf9 cell m e m b r a n e as a result o f virally induced metabolic stress. However, at late stages o f infection, vesicular debris m a y contain virus particles, released b y lysis or exocytosis, as sugg e s t e d b y M a t h i e s o n and L e e (1981). F u r t h e r experimentation will be necessary to determine to w h a t e x t e n t c h a n g e s in surface structure a f f e c t S f 9 cell r e c e p t i v i t y to the virus and p r o m o t e intercellular f u s i o n and vesicle f o r m a t i o n after viral entry into cells.
Acknowledgements We thank T. Devesceri for photographic support.
Plages de lyse et r~plication de l'iridovirus Tipula dans les cellules ovariennes de Spodoptera frugiperda Un test de quantification par plages de lyse permettant l ' r t u d e de la r r p l i c a t i o n de l ' i r i d o v i r u s Tipula (TIV) a 6t6 mis au point sur une lignre cellulaire d r r i v r e de Spodoptera frugiperda (Sf9). L'infection et la formation des ptages ont 6t6 suivies par microscopie ~ contraste de phase, vidrom i c r o s c o p i c et m i c r o s c o p i e ~ f l u o r e s c e n c e . L a structure des virions, des centres viroplasmiques et des organites cellulaires infectrs a 6t6 examinre par microscopie 61ectronique ~ transmission (TEM). Quatre heures apr~s l'infection, la drtrrioration des cellules ainsi que la prrsence de nombreuses structures cytoplasmiques d'aspect vrsiculaire, permettent la visualisation de plages sur le tapis cellulaire. La quantification des virions prrsents par absorption ~t 260 n m d ' u n e p r r p a r a t i o n de T I V a 6t6 drterminre par TEM. Le nombre de plages visibles correspond ~ la concentration de virus, et 1 unit6 A 2 produit ca 105 plages L'analyse de I ' A D N p a r h y b n d a t t o n n ' a pas p e r m l s d e m e t t r e en 6vidence de d i f f r r e n c e significative entre I ' A D N g r n o m i q u e p r o v e n a n t du T I V propag6 dans la lignre Sf9 ou celui du TIV propag6 chez les larves de Galleria mellonella. Ces rrsultats indiquent que la lignre Sf9 est permissive pour la rrplication du TIV et prrsente plusieurs avantages si on la compare aux autres l i g n r e s cellulaires a c t u e l l e m e n t u t i l i s r e s p o u r l ' r t u d e des i n t e r a c t i o n s c e l l u l e hrte/TIV. 60
.
.
°
"
.
329
Mots-clds : R r p l i e a t i o n , Iridoviridae, T I V , P e r m i s s i v i t r ; Cellules o v a r i e n n e s , D o s a g e par plaques, Ultrastructure.
References Brown, D.A., Lescott, T., Harrap, K.A. & Kelly, D.C. (1977), The replication and titration of iridescent virus type 22 in Spodoptera frugiperda cells. J. Gen. ViroL, 38, 175-178. Brown, M. & Faulkner, P. (1978), Plaque assay of nuclear polyhedrosis viruses in cell culture. Appl. Environ. Microbiol., 36, 13-35. Cameron, I.R. (1990), Identification and characterization of the gene encoding the major structural protein of insect iridescent virus type 22. Virology, 178, 3542. Cerutti, M. & Devauchelle, G. (1979), Cell fusion induced by an invertebrate virus. Arch. Virol., 61, 149-155. Hall, R.P., Johnson, D.F. & Loefer, J.B. (1935), A method for counting protozoa in the measurement of growth under experimental conditions. Trans. Am. Microscop. Soc., 54, 298-300. Jahagirdar, A.P. & Seligy, V.L. (1991), Modulation of trehalase activity in two insect cell lines by virus infection and trehalose. Biochem. Inter., 23, 10491054. Kelly, D.C. (1985), Insect iridescent virus. Curr. Top. Microbiol. ImmunoL, 116, 23-25. Lee, P.E. & Brownrigg, S. (1982), Effect of virus inactivation on Tipula iridescent virus-cell relationships. J. Ultrastruct. Res., 79, 189-197. Mathieson, W.B. & Lee, P.E. (1981), C~ytologyand autoradiography of Tipula iridescent virus infection of insect suspension cell cultures. J. Ultrastruct. Res., 74, 59-68. Selling, B.H. & Rueckert, R.R. (1984), Plaque assay for black beetle virus. J. ViroL, 51, 251-253. Smith, K.M., Hills, G.J. & Rivers, C.F. (1961), Studies on the cross-inoculation of the Tipula iridescent virus. Virology, 13, 233-241. Summers, M.D. & Smith, G.E. (1988), A manual of methods for Baculovirus vectors and insect cell culture procedures. Texas Agric. Exper. St. Bull., n° 1555, 29-30. Tajbakhsh, S., Dove, M.J., Lee, P.E. & Seligy, V.L. (1986), DNA components of Tipula iridescent virus. Biochera. Cell BioL, 64, 495-503. Tajbakhsh, S., Lee, P.E. & Seligy, V.L. (1989), Comparative studies on Tipula iridescent virus DNA derived from whole insects and ceils in culture, in "Invertebrate cell systems applications". Chapter 6 (pp. 5361). CRC Press, Inc., Boca Raton, FL. Tajbakhsh, S. & Seligy, V.L. (1990), Molecular biology of Tipula iridescent virus, in "Molecular biology of iridoviruses" (G. Daral). Developments in molecular virology. Chapter 5. Martinus Nijhoff Publ., Boston, MA. Tajbakhsh, S., Kiss, G., l.e~, P.E. & Seligy, V.L. (1990a), Semipermissive replication of Tipula iridescent virus
330
M. CZUBA E T AL.
in Aedes albopictus C6/36 cells. Virology, 174, 264275. Tajbakhsh, S., Lee, P.E., Watson, D.C. & Seligy, V.L. (1990b), Molecular cloning, characterization, and expression of the Tipula iridescent virus eapsid gene. J. Virol., 64, 125-136. Watson, D.H., Russell, W.C. & Wildy, P. (1963), Electron microscopic particle counts on Herpes virus using the
phosphotungstate negative staining technique. Virology, 19, 250-260. Wood, H.A. (1977), An agar overlay plaque assay method for Autographa californica nuclear-polyhedrosis virus. J. Invert. Pathol., 29, 304-307. Xeros, N. (1954), A second viruss disease of the leatherjacket, Tipula paludosa. Nature CLond.), 174, 562563.