The mechanism for the replication of animal RNA viruses has been the subject of intense investigation. With few exceptions (Gomatos and Tamm, 1963; Sonnabend et al., 1964), however, these studies have been on picornaviruses, specifically poliovirus (Baltimore, Becker, and Darnell, 1964) or members of the encephalomyo-(arditis group (Montagnier and Saunders, 1963). This report will present studies on the replication pattern of Semliki Forest virus, a group A arbovirus. Semliki Forest virus has some properties advantageous to the study of animal virus replication, including a particle-to-infectious-virus ratio close to 1 (Cheng, 1961), wide host range, rapid growth to high titers in tissue culture, and little or no pathogenicity for man (Andrews, 1964). In this paper the forms of viral RNA produced in the cytoplasm of infected chick embryo fibroblasts are considered. Materials and Methods.-Virus: The Kumba strain of Semliki Forest virus (SFV) employed in these studies was originally obtained from Dr. Alick Isaccs, Mill Hill, London. Pools of virus were prepared by infection of actinomycin D (ACM)-treated (0.5 ,g/ml) chick embryo fibroblasts with low multiplicities of virus. Virus was grown in Eagle's medium, harvested after 14 hr by the methods of Taylor (1965), and assayed by a plaque titration method (Friedman, 1964). Radioactive virus was grown in chick cells treated with ACM (0.5 ,gg/ml) in the presence of H3uridine and H3-adenosine. Fluids were harvested after 18 hr and virus was purified by the method of Cheng (1961) with the addition of a final dialysis to remove small tritiated products. The specific activity of the final preparation was usually about 1 cpm/2000 infectious units. The RNA extracted from this virus preparation by the method of Cheng (1958) had a sedimentation value of 42S and was ribonuclease-sensitive. Sedimentation velocities of the tritiated virus performed on sucrose gradients indicated a peak of radioactivity at about 350S which corresponded to the peak of virus infectivity (cf. Cheng, 1961). Radioactivity and infectivity were also present in the pellet, possibly due to viral aggregation. Cells: One-day-old primary chick embryo monolayers (CE) were prepared in 100-mm plastic Petri plates by previously described methods (Taylor, 1965); these cells were employed in all studies. When confluent, these monolayers contained about 2.5 X 107 cells/plate. Pulse-labeling and viral RNA extraction: Pulse-labeling was carried out by adding to the cell cultures 1 ml of tissue culture fluid containing the isotope. After an appropriate pulse period, the monolayers were washed five times with cold phosphate buffered (0.01 M, pH = 7.2) 0.85% saline and once with potassium phosphate buffer (0.1 M, pH = 7.1). Cells were removed from plastic Petri dishes with a rubber stopper and suspended in 1 ml of potassium phosphate buffer. The cells were disrupted with a Dounce homogenizer, and the homogenate was centrifuged at 600 g for 10 min. The remaining cytoplasmic fraction was extracted with 5% sodium dodecyl

Actinomycin D-treated chick fibroblasts were infected with purified 32P-labeled Semliki forest virus, and ribonucleic acid (RNA) was extracted after 1 or 2 hr. Within 1 hr, viral RNA forms sedimenting in sucrose gradients at 42S, 30S, and 16S were present. The 42S form corresponded to the RNA of the virion. The 16S form appeared to be a double-stranded template for the formation of new viral RNA, since nascent RNA was associated with it and the molecule could be heat-denatured and subsequently reannealed by slow cooling. Interferon treatment before infection, or puromycin (50 ,g/ml) or cycloheximide (200 ,g/ml) added at the time of virus infection, had no effect on the formation of the 30S RNA but inhibited the production of the 16S form. Several findings made it unlikely that these results were due to breakdown of parental RNA and reincorporation of 32P into progeny structures. The results suggested that the mechanism of interferon action involves inhibition of protein synthesis by parental viral RNA, since a specific viral RNA polymerase had previously been demonstrated to be necessary for production of 16S RNA. No protein synthesis appears necessary for formation of 30S RNA from parental virus RNA.on July 6, 2020 by guest