The origins of defective interfering particles of the negative-strand RNA viruses

Abstract

Robert A. Laztarini,* Jack D. Keenet and Manfred Schubert* * Laboratory of Molecular Genetics National Institute of Neurological and Communicative Disorders and Stroke Bethesda, Maryland 20205 t Department of Microbiology and Immunology Duke University Medical Center Durham, North Carolina 27710 The negative-strand RNA viruses have deceptively diverse properties but are grouped together because they use a negative-strand (antisense) RNA as their genetic material and consequently are likely to share common strategies of replication. The group consists of the myxoviruses, paramyxoviruses, rhabdoviruses, bunyaviruses and arenaviruses, and thus includes pleiomorphic as well as bacilliform viruses, and vi- ruses with segmented as well as nonsegmented ge- nomes (Table 1). Nonetheless, these viruses share a number of unifying structural features: their RNA ge- nomes are firmly associated with protein in the form of nucleocapsid (ribonucleoprotein, or RNP) struc- tures, they are enclosed by membrane envelopes and it is likely that each contains an RNA polymerase capable of using the nucleocapsid as a template. Recent biochemical studies have illuminated various aspects of the replication and gene expression of these viruses, showing in many cases that the antici- pated similarities do exist. We will summarize our present understanding of negative-strand RNA virus replication and examine the replicative event leading to the generation of defective interfering (DI) particles. The Replicative Strategy Although the anticipated similarity in replicative strat- egies has been observed among the paramyxoviruses and rhabdoviruses, it remains an article of faith that a// negative-strand RNA viruses share a common rep- licative strategy. Some differences do exist between the replicative cycles of other negative-strand RNA viruses. For example, influenza virus replication has a nuclear involvement and is sensitive to actinomycin D, while the replication of the rhabdoviruses does not share these properties. At present, however, there is no clear indication that these differences signal a departure from the general scheme. Consequently, the details of replication may differ, but the overall scheme of each is likely to be only a minor variation on the canonical scheme. In this scheme, the genome of the infecting virus is first transcribed by the virion- bound RNA polymerase, producing the monocistronic mRNAs. This process, which employs the infecting virion template and polymerase, has been termed primary transcription. It does not require concomitant host or viral protein synthesis. The monocistronic transcripts are translated into viral proteins necessary for the synthesis of new, secondary viral RNP tem- plates and RNA polymerases. Although protein syn- thesis is necessary for the synthesis of the catalytic units of secondary transcription, the process mRNA synthesis on secondary templates, like that the primary template, does not require concomitant pro- tein synthesis. The synthesis of full-length genomic and antigen- omit RNA (replicative RNA synthesis) requires contin- ued protein synthesis, and protein synthesis inhibitors arrest replicative synthesis in the infected cell. This stringent requirement for protein synthesis may be traced, in part, to the need for a continuous supply of the nucleocapsid proteins that are tightly and spe- cifically associated with the full-length genomic and antigenomic RNAs, forming RNP structures called nu- cleocapsids. Thus unencapsidated genome-length RNA does not accumulate in vivo when protein syn- thesis is blocked. The possibility that RNA synthesis