On the expression strategy of the tospoviral genome

The work described in this thesis was aimed at the unravelling of the molecular biology of tospoviruses, with special emphasis on the process of replication of the tripartite RNA genome. At the onset of the research the complete genome sequence of tomato spotted wilt virus (TSWV), type species of the genus Tospovirus, became available. These sequence data indicated that the tospoviruses represent plant-infecting members of the large family of the arthropod-born Bunyaviridae. Genome sequence comparisons indicated however that the L RNA segment of TSWV would encode a much larger viral polymerase (331.5 kDa) than, as far as known, its animal-infecting counterparts (reported sizes of 241 to 259 kDa). To verify whether a large polymerase represents a characteristic i.e. genus-specific property of tospoviruses the complete sequence of the L RNA segment of a second tospovirus, impatiens necrotis spot virus (INSV), was elucidated (Chapter 2). These sequence data revealed that the L RNA of INSV appeared to be comparable in size to that of TSWV (8675 nucleotides versus 8897 for TSWV), containing an open reading frame with a predicted size of 330.3 kDa of the INSV polymerase. Therefore the next question to be answered was whether the large primary translation product of the tospoviral L RNA acts as an unprocessed polymerase or whether this protein would undergo some cleavages to obtain smaller, functional replication proteins. Answering this question was even more necessary since the theoretical size of the TSWV L RNA ORF greatly exceeded previously determined sizes (110 to 220 kDa) for a large protein reported to copurify with TSWV particles. To this end both the 5'-terminal and 3'-terminal parts of the ORF in the TSWV L RNA were expressed in Escherichia coli and antibodies raised against these regions. Using these tools it could be established that the polymerase (L protein) of TSWV, though significantly larger than that of other bunyaviruses, is present in virus particles (10 to 20 copies per virion) in an unprocessed, full length form (Chapter 3). To allow further analyses of the TSWV polymerase, attempts were made to clone and express the complete L RNA ORF in the baculovirus/insect cell system. In spite of all efforts, only a shorter translation product of 67 kDa was obtained from a baculovirus recombinant containing a complete DNA copy of the TSWV L RNA (Chapter 3). Sequence analysis of the cloned copy revealed a 80 basepairs deletion, resulting in two premature stop codons, which most likely have led to the resulting truncated L protein. To gain more insight in the "cap-snatching" event which takes place during initiation of tospovirus transcription, nucleoprotein (N) mRNAs were partially purified from TSWV-infected N. rustica leaves and cloned (Chapter 4). Sequence analysis of the cloned, 5'-proximate regions of 20 cloned mRNAs showed the presence of extra, non-templated sequences, ranging in length from 12 to 21 nucleotides, confirming our earlier primer extension studies. As these sequences were of non-viral origin a cap-snatching mechanism for tospoviral transcription initiation could thus be definitively identified. None of the hostderived leader sequences analyzed were identical and only limited sequence specificity at the endonucleolytic site was observed (some preference for cleavage at a U residue). During the course of this Ph.D. research, Adkins et al. (1995) reported that in vitro transcriptase activity was associated with freshly isolated TSWV particles. It was investigated (Chapter 5) whether the reported levels of in vitro activity could be further improved and whether this system would lend itself for analysis of the viral proteins involved by e.g. inhibition studies using specific antibodies. Trichloroacetic acid-precipitable products could consistently be obtained after incubation of detergent-disrupted TSWV virions under the assay conditions reported by Adkins et al. (1995) and using (α- 32 P)CTP. No significant improvement in CMP incorporation levels could be achieved by testing variable conditions and varying concentrations of assay components. The reaction products obtained hybridized with clones from all three genomic RNA segments. No discrimination between transcription and replication could be made however, and since none of the available specific antibodies directed against any viral protein had an inhibitory effect, it was concluded that the current in vitro system will be of limited value for unravelling the RNA synthesizing process and the role of the individual viral proteins therein. As a first step towards a manipulatable transcription/replication system, a hybrid baculovirus/bacteriophage T7 vector system was developed for transient expression in insect cells of all factors involved in TSWV genome transcription and replication. The results obtained (Chapter 6) illustrate the potential of the system. Although various foreign genes could successfully be expressed to measurable amounts, the reconstitution of a TSWV transcription/replication complex was hampered due to the apparent impossibility (Chapter 3) to clone the complete polymerase gene. Finally, in Chapter 7 (General discussion and concluding remarks), the results obtained are compared with the data reported for animalinfecting bunyaviruses, leading to a discussion of some evolutionary aspects. Furthermore, suggestions are made to circumvent some of the problems encountered during the course of the studies presented in this thesis.