INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Human cytomegalovirus (HCMV) is a ubiquitous herpesvirus which causes mild or subclinical diseases in immunocompetent adults but may lead to severe morbidity or mortality in neonates and immunocompromised individuals (2, 24). Disseminated HCMV infection, common in AIDS patients and organ transplant recipients, is usually associated with gastroenteritis, pneumonia, and retinitis (12, 29). Studies on the functions of viral genes in HCMV replication in vivo are essential for understanding viral pathogenesis and developing new strategies to combat the viral infection. However, there are currently no suitable animal models for HCMV infection. HCMV only propagates in human cells and grows slowly due to a long lytic replication cycle (24). These properties of HCMV have hampered the studies of HCMV pathogenesis and gene function.

Infection of the mouse with murine CMV (MCMV) provides a valuable in vivo model for studying the biology of CMV infection. This is because infection of mice by MCMV resembles in many ways its human counterpart with respect to pathogenesis during acute infection, establishment of latency, and reactivation after immunosuppression, transfusion, or transplantation (2, 15, 17, 24). Its genome of 230 kb is predicted to encode more than 170 open reading frames, 78 of which have extensive homology to those of HCMV (5, 32). However, many of these MCMV genes remained uncharacterized and their functions in viral pathogenesis have not been investigated.

One of the most powerful approaches to study the function of virus-encoded genes is to introduce mutations into the viral genome and to screen viral mutants in both tissue culture and animals for possible growth defects in vitro and in vivo. The construction of herpesvirus mutants was first reported using site-directed homologous recombination and then using transposon-mediated insertional mutagenesis (16, 25, 31, 34, 48). Methods using overlapping cosmid DNA fragments to generate mutants of HCMV and other herpesviruses have also been reported (7, 9, 18, 44, 45). More recently, the MCMV genome as well as the genomes of other herpesviruses have been cloned into a bacterial artificial chromosome (BAC), and MCMV mutants were successfully generated from the BAC-based viral genome by both site-directed homologous recombination and transposon-mediated insertional mutagenesis (3, 11, 23, 37, 41-43, 47). It has been shown that the BAC sequence in a BAC-based pseudorabies virus does not affect viral pathogenesis in vivo in animals (41, 42). Moreover, two different approaches to excise the BAC sequence from the viral genome have been described (42, 47). These studies have greatly facilitated the identification of the functions of viral genes in tissue culture and in animals.

We have recently applied a Tn3 transposon-mediated shuttle mutagenesis system to generate MCMV mutants (49). In this approach, the transposon is inserted into a plasmid library of MCMV genomic DNA in Escherichia coli. Regions bearing an insertion mutation are then transferred to the MCMV genome by homologous recombination. In the study reported here, we have successfully used this approach to generate MCMV mutants containing transpositional insertions in open reading frames m09 and M83. Our results provide the first direct evidence to suggest that the m09 open reading frame is not essential for viral replication in NIH 3T3 cells. Studies in BALB/c-Byj mice and CB17 severe combined immunodeficient (SCID) mice indicated that m09 is dispensable for viral growth in the salivary glands, lungs, livers, spleens, and kidneys of these animals that were infected intraperitoneally with the viral mutant. Moreover, our data suggest that the presence of the transposon sequence in the viral genome does not significantly affect viral growth in both the BALB/c and SCID mice. These results demonstrate the feasibility of using this Tn3-based system to study the functions of MCMV genes in both immunocompetent and immunodeficient hosts.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cells and viruses. The wild-type Smith strain of MCMV was obtained from the American Tissue Culture Collection (ATCC) (Rockville, Md.). NIH 3T3 cells (ATCC CRL1658) were cultured in a humidified incubator at 37°C and in the presence of 5% CO₂. Cells were maintained in complete medium containing Dulbecco modified Eagle medium (DMEM) supplemented with 10% NuSerum (Becton Dickinson, Mass.). The Smith strain and the viral mutants, Rvm09 and RvM83, were propagated in NIH 3T3 cells as described previously (49).

Isolation of viral DNA and construction of a MCMV DNA subclone pool. NIH 3T3 cells were infected at a multiplicity of infection (MOI) of 0.1 PFU per cell. Approximately 5 to 7 days postinfection, almost all of the cells showed cytopathic effects (CPE) and were harvested. Viral particles and DNA were purified as described previously (6, 22). To generate the MCMV genomic pool, the DNA was partially digested with the restriction enzyme Sau3A. The first two nucleotides of the 3' overhang sequence of the digested DNA fragments were then filled in with dGTP and dATP. The digested DNA fragments were separated on 0.8% agarose gels. DNA fragments in the size range of 1.6 to 4 kb were purified and cloned into pHSS6-SalI (a gift from Michael Snyder of Yale University) (4, 14, 40).

Transposon shuttle mutagenesis to generate the MCMV DNA pool that contained the transposon sequence. The MCMV genomic fragment pool was first transformed into B211 {RDP146 [F recAI(lac-pro)rpsE; spectinomycin resistant] with plasmid pLB101} and colonies were selected on plates that contained kanamycin and chloramphenicol. These colonies were then mated with strain XZ95 which contained the transposon (Tn3-gpt) sequence (49). The Tn3-gpt construct was derived from a E. coli Tn3 transposon and contained the gpt gene (49). The mixture was then allowed to grow on plates that contained tetracycline, kanamycin, and chloramphenicol for 2 days at 30°C. At this time, cointegration occurred. Finally, the cointegrates were resolved to generate the plasmid pHSS6::MCMV fragments containing a Tn3 insertion by mating the bacteria with strain E. coli 70 (NG135 [K-12 recA56 gal-delS165 strA; streptomycin resistant] with plasmid pNG54). The plasmid DNA that contained the viral DNA fragments with a randomly inserted transposon was isolated and used to transform into E. coli DH5 for long-term storage.

Construction of recombinant MCMV. The transposon-MCMV DNA constructs were isolated and digested with NotI in order to release the genomic fragments containing the transposon (Fig. 1B). The excised fragments (1 to 3 µg) and full-length intact viral genomic DNA (Smith strain, 8 to 12 µg) were subsequently cotransfected into mouse NIH 3T3 fibroblasts using a calcium phosphate precipitation protocol (Gibco BRL, Grand Island, N.Y.). The recombinant virus was purified by six rounds of amplification and plaque purification in the presence of 25 µg of mycophenolic acid (Gibco BRL) and 50 µg of xanthine (Sigma, St. Louis, Mo.) per ml, as described previously (46, 49). For each cotransfection, several viral plaques were picked and expanded. Virus stocks were prepared by growing the viruses in T150 flasks of NIH 3T3 cells. In order to determine the location of the transposon insertion within the viral genome, the junctions of the transposition in the recombinant viral DNA were directly sequenced with primer FL110PRIM using the fmol Cycle Sequencing Kit (Promega, Inc., Madison, Wis.).

View larger version (21K): [in this window] [in a new window]   FIG. 1.   Schematic representation of the structure of the transposon construct used for mutagenesis (A) and the procedure for the construction of MCMV mutants that contained random transposon insertions (B). TR, terminal repeat; Tet, tetracycline resistance gene; gpt, gene that encodes guanine phosphoribosyltransferase (gpt); poly(A), transcription termination signal.

Southern and Northern analyses of recombinant viruses. Viral genomic DNA was isolated from NIH 3T3 cells as described previously (20-22). Briefly, cells that exhibited 100% CPE were washed with phosphate-buffered saline (PBS) and then subjected to proteolysis by a solution that contained sodium dodecyl sulfate and proteinase K. The genomic DNA was purified by extraction with phenol-chloroform, followed by precipitation with 2-propanol (22). Southern analyses were carried out to detect the presence of the transposon within the viral genome. Briefly, genomic DNA was digested with either HindIII, NotI, or EcoRI; separated on a 0.8% agarose gel; transferred to a Zeta-Probe nylon membrane (Bio-Rad, Hercules, Calif.); hybridized with the ³²P-radiolabeled DNA probes that contain the transposon and the MCMV sequences; and finally analyzed with a STORM840 phosphorimager (49). The labeled DNA probes were prepared by random primer synthesis (Boehringer Mannheim, Indianapolis, Ind.).

Growth kinetics of recombinant viruses. The analyses of the growth of the recombinant viruses in vitro were carried out as described previously (49). In brief, 5 × 10⁵ NIH 3T3 cells were infected at an MOI of either 0.5 or 5.0 PFU per cell. The cells and medium were harvested at 0, 1, 2, 4, and 7 days postinfection, and viral stocks were prepared by adding an equal volume of 10% skim milk, followed by sonication. The titers of the viral stocks were determined by plaque assays in triplicate experiments.

Viral growth studies in animals. Three-week-old male BALB/c-Byj mice (Jackson Laboratory, Bar Harbor, Maine) or six-week-old CB17 SCID mice (National Cancer Institute, Bethesda, Md.) were infected intraperitoneally with 10³ to 10⁴ PFU of each virus. The animals were sacrificed at different time points (e.g., 3 days) postinoculation as specified in Results. For each time point, at least three animals were used as a group and infected with the same virus. The whole salivary gland, about 0.1 to 0.2 g of the liver, one-quarter of the lungs, the whole spleen, and one of the kidneys were collected individually into 3-ml sterile tubes. To avoid cross-contamination of viruses between organs and different recombinants, surgical tools (forceps and scissors) were rinsed once in PBS, rinsed three times in 70% ethanol, and flamed after each rinse in ethanol. Each sample was suspended in a mixture of DMEM and 10% skim milk (50% [vol/vol]) at 0.1 g/ml. The organs were then sonicated on ice using a 550 Sonic Dismembrator (Fisher Scientific, Pittsburgh, Pa.) until the organ became homogenized. The samples were stored at 80°C until titers of the viruses in these samples were determined.

1

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of MCMV mutants containing the transposon insertions at open reading frames m09 and M83. We recently constructed an E. coli Tn3-based transposon (Fig. 1A) which contained the expression cassette encoding guanine phosphoribosyltransferase (i.e., gpt) and, in addition, a transcription termination site (49). The gpt gene was used as a selectable marker in the construction of MCMV recombinant viruses (46). The gpt expression cassette was inserted such that its transcription termination site functioned in the opposite direction as the other poly(A) signal presented in the transposon (Fig. 1A). Such a design would assure that the transcription of the targeted gene is truncated without altering the expression of nearby genes that may share a common poly(A) signal with the disrupted gene. To generate a pool of MCMV DNA fragments that contained a randomly inserted transposon, virus DNA was first purified and partially digested with Sau3A. Digested fragments in the size range of 1.6 to 4 kb were then cloned into vector pHSS6-SalI (4, 14, 40) to generate a MCMV genomic library. The transposon was introduced into this MCMV DNA library through a shuttle mutagenesis protocol, as described previously (see Materials and Methods) (49), to generate a pool of MCMV genomic sequences that contained the transposon sequence (Fig. 1B).

View larger version (35K): [in this window] [in a new window]   FIG. 2.   (A) The locations of the transposon insertions in the recombinant viruses. The transposon sequence is shown as a filled bar, while the coding sequence of each open reading frame is represented by an open arrow. The orientation of the arrow represents the direction of the translation and transcription predicted based on the nucleotide sequence (32). The numbers represent the sizes of the DNA fragments of the mutant viruses that contained the transposon sequence and were generated by digestion with HindIII (H), NotI (N), or EcoRI (E). (B) Southern blot analyses of the viral mutants. The DNA fractions were isolated from cells infected with the wild-type (WT) virus and different MCMV mutants. The DNA samples (20 µg) were digested with HindIII (H), NotI (N), or EcoRI (E); separated on 0.8% agarose gels; transferred to a Zeta-Probe membrane; and hybridized to a DNA probe. The probes used for the analyses were the plasmids that contained the MCMV DNA fragments inserted with the transposon sequence.

In vitro characterization of MCMV mutants in tissue culture. Southern hybridization analyses with DNA probes containing the transposon and the viral sequence were used to examine the genomic structure of the recombinant viruses and determine if the mutants contained the transposon insertion (Fig. 2B). Analysis of the HindIII-digested DNA of each recombinant virus clearly showed a small fragment of 1.8 kb representing the gpt gene fragment, indicating the presence of the transposon sequence within the viral genome (Fig. 2B, lanes 1 and 5). This conclusion is further supported by the results of Southern analyses of the mutant DNAs digested with another restriction enzyme (i.e., EcoRI for Rvm09 and NotI for RvM83) (lanes 3 and 4 and lanes 7 and 8). In each case, the genomic fragment from the transposon-containing viruses should be bigger than that of the wild-type virus by 3.6 kb, which is the size of the transposon. The stocks of these recombinant viruses appeared to be pure and free of the wild-type strain, since the hybridizing DNA fragments from the mutants did not comigrate with those of the wild-type Smith strain (Fig. 2B, lanes 1 to 8). For example, the hybridization patterns of the Rvm09 and Smith strain DNAs digested with HindIII gave rise to three DNA bands (28.9, 6.1, and 1.8 kb) and one DNA band (33.1 kb), respectively (Fig. 2B, lanes 1 and 2). Meanwhile, the hybridized species (13.6 kb) of the EcoRI-digested Rvm09 DNA migrated differently from that (10 kb) of the wild-type viral DNA digested with the same enzyme (Fig. 2B, lanes 3 and 4). The sizes of the hybridized DNA fragments (Fig. 2B) were consistent with the predicted digestion patterns of the recombinant viruses based on the MCMV genomic sequence (32) and the location of the transposon insertion in the viral genome as determined by sequence analysis (Fig. 2A). The restriction enzyme digestion patterns of the regions of the mutant genomic DNAs other than the transposon insertional site appeared to be identical to those of the parental Smith strain, as indicated by ethidium bromide staining of the digested DNAs (data not shown). This observation suggested that regions of the viral genome other than those containing the transposon insertion remained intact in these MCMV mutants.

View larger version (122K): [in this window] [in a new window]   FIG. 3.   Northern blot analyses of the RNA fractions isolated from cells that were mock infected (lanes 1, 4, and 7) or infected with the wild-type virus (WT) (lanes 2, 5, and 8) and the mutant viruses (lanes 3, 6, and 9). A total of 107 NIH 3T3 cells were infected with each virus at an MOI of 10 PFU/cell, and cells were harvested 24 h postinfection. Equal amounts of RNA samples (30 µg) were separated on agarose gels that contained formaldehyde, transferred to a nitrocellulose membrane, and hybridized to a 32P-radiolabeled probe that contained the sequence of m09 (lanes 1 to 3), M83 (lanes 4 to 6), or M25 (lanes 7 to 9).

View larger version (11K): [in this window] [in a new window]   FIG. 4.   In vitro growth of MCMV mutants in tissue culture. Mouse NIH 3T3 cells were infected with each virus at an MOI of either 0.5 PFU (A) or 5 PFU (B) per cell. At 0, 1, 2, 4, and 7 days postinfection, cells and culture media were harvested and sonicated. The viral titers were determined by plaque assays on NIH 3T3 cells. The values of the viral titer represent the average obtained from triplicate experiments. The standard deviation is indicated by the error bars.

Replication of the transposon-containing viral mutants in immunocompetent animals. In order to use the Tn3-based transposon system to generate MCMV mutants and study the phenotypes of the mutants in vivo, it is necessary to determine whether the presence of the transposon sequence in the viral genome does not significantly affect viral replication in animals. To determine accurately the capability of the viral mutants to grow in animals, a low dose of viruses was used for inoculation. BALB/c-Byj mice were injected intraperitoneally with 10³ PFU of Rvm09, RvM83, and Smith strain. At 1, 3, 7, and 14 days postinfection, salivary glands, lungs, spleens, livers, and kidneys were harvested, and the virus titers in these five organs were determined. These organs are among the major targets for MCMV infection (2, 15, 17, 24). At 14 days postinfection, the titers of Rvm09 in the salivary glands were similar to those of the Smith strain, while a reduction of 500-fold in the virus titer was found in the salivary glands of the RvM83-infected animals (Fig. 5). At 7 days postinfection, the titers of Rvm09 in the lungs, spleens, livers, and kidneys were also similar to those of the Smith strain. In contrast, a reduction of at least 20-fold in the virus titer was found in the lungs, spleens, livers, and kidneys of the animals infected with Rvm83 (Fig. 5). These results are consistent with previous observations that a MCMV mutant with a deletion at M83 was attenuated in replication in these organs (26). Since Rvm09 replicated as equally well as the Smith strain in all the organs examined, the presence of the transposon sequence per se within the viral genome appears to have no significant effect on viral replication in BALB/c mice, at least in these organs. These results further suggest that the m09 open reading frame is dispensable for viral replication in these organs of BALB/c mice.

View larger version (19K): [in this window] [in a new window]   FIG. 5.   Titers of MCMV mutants in salivary glands (A), lungs (B), spleens (C), livers (D), and kidneys (E) of the infected BALB/c mice. BALB/c-Byj mice were infected intraperitoneally with 103 PFU of each virus. At 1, 3, 7, and 14 days postinfection, the animals (three mice per group) were sacrificed. The salivary glands, lungs, spleens, livers, and kidneys were collected and sonicated. The viral titers in the tissue homogenates were determined by standard plaque assays in NIH 3T3 cells. The limit of detection was 10 PFU/ml of the tissue homogenate. The viral titers represent the average obtained from triplicate experiments. The error bars indicate the standard deviation. Error bars that are not evident indicate that the standard deviation was less than or equal to the height of the symbols.

Replication of the transposon-containing viral mutants in immunodeficient animals. Previously studies have indicated that immunodeficient animals are extremely susceptible to MCMV infection (13, 28, 30, 33). For example, the CB17 SCID mice, which lacks functional T and B lymphocytes, are sensitive to an extremely low level of viral replication since these animals succumb to as little as 10 PFU of MCMV (28, 30). These animals have served as an excellent model in determining the virulence of different MCMV strains and mutants and in studying the mechanism of how they cause opportunistic infections in immunocompromised hosts. In order to use the Tn3-based transposon system to generate MCMV mutants and study the phenotypes of the mutants in vivo in an immunodeficient host, it is also necessary to determine whether the presence of the transposon sequence in the viral genome does not significantly affect viral replication in these animals. Two sets of experiments with the SCID mice were carried out to address this issue. First, the survival rates of the animals infected with Rvm09 and RvM83 were determined and compared with those infected with the Smith strain. The SCID mice (five animals per group) were injected intraperitoneally with 10⁴ PFU of Rvm09, RvM83, and Smith strain. As shown in Fig. 6, the survival rates of the animals infected with Rvm09 were similar to those of the animals infected with the Smith strain. Half of the animals infected with Rvm09 and the Smith strain died within 22 and 23 days postinfection, respectively. In contrast, no animals infected with RvM83 were found dead until 40 days postinfection.

View larger version (16K): [in this window] [in a new window]   FIG. 6.   Mortality of the SCID mice infected with the Smith strain, Rvm09, and RvM83. CB17 SCID mice (five animals per group) were infected intraperitoneally with 104 PFU of each virus. Mortality of mice was monitored for 46 days postinfection, and survival rates were determined.

View larger version (20K): [in this window] [in a new window]   FIG. 7.   Titers of MCMV mutants in the salivary glands (A), lungs (B), spleens (C), livers (D), and kidneys (E) of the infected SCID mice. CB17 SCID mice were infected intraperitoneally with 104 PFU of each virus. At 1, 3, 7, 10, 14, and 21 days postinfection, the animals (three mice per group) were sacrificed. The salivary glands, lungs, spleens, livers, and kidneys were collected and sonicated. The viral titers in the tissue homogenates were determined by standard plaque assays in NIH 3T3 cells. The limit of detection was 10 PFU/ml of the tissue homogenate. The viral titers represent the average obtained from triplicate experiments. The error bars indicate the standard deviation. Error bars that are not evident indicate that the standard deviation was less than or equal to the height of the symbols.

Stability of the transposon mutations in the recombinant viruses. Our previous studies have indicated that a transposon sequence inserted at the 3'-terminal region of the MCMV genome (i.e., the m155 open reading frame) is stable during viral replication in NIH 3T3 cells and in BALB/c mice (49). However, it is not known whether the transposon sequence inserted at the 5'-terminal sequence of the viral genome is also stable during viral propagation. Equally unclear is whether the insertional mutation is stable during viral propagation in immunodeficient animals, in which the viral replication may achieve a higher level than that in immunocompetent mice. To address these issues, two sets of experiments were carried out. First, recombinant virus Rvm09 and RvM83 were used to infect NIH 3T3 cells at an MOI of <0.01 and allowed to grow for five generations (60 days) in the absence of gpt selection. Second, 10⁴ PFU of viruses were used to infect both the BALB/c and SCID mice. At 14 days postinfection, salivary glands and lungs were harvested from the infected animals and sonicated to release the virus. Viruses were recovered by infecting NIH 3T3 cells with the sonicated tissue homogenates. Viral DNAs were purified from the infected cells, and their restriction digestion patterns were analyzed in agarose gels. Figure 8 shows a Southern analysis of the Rvm09 viral DNAs with a DNA probe that contained the transposon and m09 open reading frame sequence. These results indicated that no change in the hybridization patterns of Rvm09 occurred as a result of viral growth for five generations (60 days) in cultured cells (Fig. 8, lane 3) or in animals for 14 days (lanes 4 to 7). Moreover, the overall HindIII-digestion patterns of Rvm09 DNA isolated from either infected cultured cells or animals were identical to those of the original recombinant virus, as visualized by ethidium bromide staining of the viral DNAs (data not shown). Similar results were also observed in experiments with RvM83 (data not shown). Thus, the transposon insertion in Rvm09 and RvM83 appeared to be stable in tissue culture and in both BALB/c and SCID mice.

View larger version (60K): [in this window] [in a new window]   FIG. 8.   The stability of the transposon mutations in tissue cultured cells and in BALB/c and SCID mice. Viral DNAs were either isolated from cells that were infected with Rvm09 (MOI = <0.01) and allowed to grow in culture for 5 days (P0) (lane 2) or five generations (60 days) (P5) (lane 3) or from cells that were infected with the virus collected from the salivary glands (SG, lanes 4 and 6) and lungs (LU, lanes 5 and 7) of either BALB/c (BALB/c, lanes 4 and 5) or SCID mice (SCID, lanes 6 and 7) 14 days after intraperitoneal inoculation with 104 PFU of Rvm09. Southern blot analyses of the viral DNA fractions digested with HindIII are shown. The DNA of the wild-type virus (WT) is shown in lane 1. The 32P-radiolabeled probe was derived from the same plasmid which was used for Southern analyses of Rvm09 in Fig. 2 and contained the transposon and the m09 open reading frame sequence.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Analyses of herpesviral genome by transposon-mediated insertional mutagenesis. Transposon-mediated mutagenesis has been widely used to study gene function in viruses, bacteria, yeast, and mammalian cells. Previously, Tn5 and phage-mu-based transposons have been used in the genetic analysis of herpes simplex virus 1 (34, 48). More recently, transposons derived from Tn1721 and Tn5 have also been used in the mutagenesis of the BAC-based genomes of MCMV and pseudorabies virus, respectively (3, 41). The system we used is based on an E. coli Tn3 transposon which has been employed in the mutagenesis studies of the genomes of Saccharomyces cerevisiae and other organisms such as Salmonella enterica serovar Typhimurium and Neisseria gonorrhoeae (35, 36, 39, 40). One of the major advantages of the Tn3 system is its transposition immunity. A plasmid already containing a copy of Tn3 is immune to further insertions of the transposon. This immunity is due to the presence of a 38-nucleotide sequence, which is also found in the E. coli chromosome (19). Therefore, Tn3 mutagenesis is simple, is straightforward, and yields little background since most of the transposition occurs in the target sequence (e.g., MCMV DNA) rather than in the E. coli chromosome sequence (14). In this study, the constructed Tn3-gpt transposon efficiently transposed into the MCMV DNA fragments, and viral mutants that contained transposon mutations at open reading frames m09 and M83 were successfully constructed. Moreover, preliminary characterization of 20 different mutants among a library of ca. 500 isolates revealed that most of these mutants contained the transposon insertion at different locations of the genome (data not shown). These results further demonstrate that the Tn3 system can be used for simultaneous isolation of multiple viral mutants and may have advantages over traditional homologous recombination system for systematic generation of viral mutants.

Potential function of open reading frames m09 and M83 in viral replication. In this study, recombinant viruses that contained the insertional mutations at open reading frames m09 and M83 were generated. Both viruses were able to replicate in NIH 3T3 cells. These results are consistent with the previous observation that M83 is not essential for MCMV replication in vitro (26). Moreover, our results provide the first direct evidence to suggest that m09 is not essential for viral replication in NIH 3T3 cells in vitro.