Genes and sequences involved in the replication of cowpea mosaic virus RNAs

The aim of the studies described in this thesis was to gain more insight in the complex molecular mechanisms underlying the RNA replication of the cowpea mosaic virus genome. Previously the replication of CPMV RNA has been examined extensively with crude membrane fractions prepared from CPMV infected cowpea leaves (Zabel, 1978; Dorssers, 1983). These studies resulted in the identification of a host-encoded RNA-dependent RNA polymerase (Mr 130K), with unknown biological function, and the virus-encoded replicase (Mr 110K). As integral constituent of a membrane bound replication complex (RCX) the viral replicase was only capable of elongating viral plus-sense RNA chains that have already been Initiated invivo . Furthermore, the invitro activity was not representative for the invivo in situation in that no single-stranded (as) progeny RNA production was observed. To resolve the mechanism of initiation of CP14V RNA synthesis and to examine the role of different virus-encoded proteins and specific genomic RNA sequences in this process a replication system that also accepts exogenously added RNA templates is required. Despite exhaustive attempts it has not yet been possible to reconstitute polymerase activity after the removal of the tightly bound template RNA from crude replication complexes isolated from CPMV infected cowpea leaves. In chapter 2 we have explained that CPMV RNA replication probably requires a primer and a template for the Initiation of RNA synthesis. Based on this Idea we have tried to reconstitute polymerase activity with a variety of mixtures of primers and templates. Wildtype CPMV-RNA and plus or minus-stranded transcript RNAs were tested as exogenously added templates. The primers used in the assays were oligo(U) or a short RNA stretch corresponding with the first 94 nucleotides of the viral B-RNA. In addition a mixture of poly(A) and oligo(U) as used in the research on RNA replication of poliovirus, has been tested in our system. Since the viral replicase activities, isolated from CPMV infected cowpea tissue, were low, incomplete and overshadowed by the strong host-encoded RNA-dependent RNA polymerase activity, the invitro RNA replication experiments were difficult to perform and it was troublesome to interprete the results. In an attempt to overcome these problems the invitro RNA replication assays and reconstitution experiments were extended to crude membrane fractions, prepared from another systemic host of CPMV, Chenopodiumamaranticolor . Indeed the host-encoded RNA-dependent RNA polymerase activity was less prominent In this plant and it was shown that due to the greater stability of the CPMV-RCX in this host,ss progeny RNA was produced. However, as a result of the more stable RCX too, It was very difficult to remove the endogenous RNA template and hence reconstitution of polymerase activity with exogenously added templates was not possible. As an alternative approach to analyse the function of viral proteins in theCPMV RNA replication, we have started to examine the possibility of reconstituting a system for viral RNA synthesis fromCPMV encoded proteins produced in Escherichiacoli . For that purpose fragments of cDNA clones containing the coding regions of the B-RNA encoded 87K putative core polymerase and the 110K viral replicase were used to produce sizable levels of these proteins in E.coli . Such an expression system may generate a template-dependent activity due toCPMVproteins which are not associated with membranes and hence could be more amenable to study the RNA replication process. This RNA synthetic activity has been demonstrated successfully for polioviral polymerase expressed in E.coli . Although the expectedCPMV encoded proteins were synthesized, they did not show any polymerase activity, contrasting with the highly active polio polymerase, simultaneously obtained in the same way. Since poliovirus andCPMV have analogous features of the genomic RNA and their expression strategy and show a comparable functional organization In their polyproteins, it has been assumed that both viruses probably would have a similar RNA replication mechanism. The results however contradict this Idea and indicate that the analogy betweenCPMVand pollovirus can not be extended to homologous mechanistical aspects of the virus multiplication. This hypothesis Is strengthened by the differences In the expression mechanisms (Wellink, 1988). Probably as a result of the split genome ofCPMV, whereas pollovirus has a genome consisting of a single RNA molecule, gene expression will be differently regulated. This makes sense since for the production of progeny virus a single RNA molecule provided with a single VPg is required versus 60 copies of each of the capsid proteins. The divided genome ofCPMV enables a regulated production of these proteins, whereas for poliovirus such regulation is not possible. Since the expression of virus-encoded proteins Is needed for viral RNA replication, it is plausible to assume that differences in the regulation of gene expression may result In different RNA replication mechanisms for poliovirus andCPMV. Although the 87K and 110K B-RNA-encoded proteins produced in E.coli appear to be not capable or not sufficient for polymerase activity, the expression system has shown to be very useful for the production of separate virusencoded proteins, among which the active viral protease. Based on both the expression experiments and the inability to prepare a template-dependent replication system, it seems a plausible hypothesis that the processing products87Kand110K,already folded in a certain way, are not able to (re)initiatie RNA replication on an added template, but that a larger precursor protein from which the polymerase is simultaneously cleaved and incorporated into an active RCX is needed. If this hypothesis Is true it would be impossible to observe complementation between two mutant B-RNA molecules, each encoding only a part of the open reading frame such that the two mutant RNA molecules together produce the complete set of B-RNA encoded proteins. These experiments still have to be performed. As polymerase activity is only expected upon binding of the proteins to a template and upon subsequent initiation of complementary RNA strand synthesis, it would be worthwhile to study these processes Independently. Firstly, one can examine the binding between specific polymerase proteins, which may be Isolated from the E.coli expression system, and templates, followed by reconstitution of a functional replication complex by the addition of suitable primers like specific oligonucleotides or VPg in various forms and membranes which can be isolated from plant or synthesized invitro . For the examination of specific nueleotide sequences involved in the viral RNA replication Infectious transcripts produced from full-length DNA copies of B- and M-RNA have been exploited(Vos, 1987).Both the infection procedure and the specific infectivity of the transcripts were improved, resulting in a system suitable to introduce site specific mutations in the viral RNAs and to analyse the effects of such mutations on the RNA replication invivo . The potential of the Infectious transcripts has been demonstrated by the analysis of genomic RNA sequences with putative RNA replication signals. By subtile modifications, a function In the RNA replication could be described to an homologous nucleotide-stretch which is present in the3'region of both B- and M-RNA. The Importance of these nucleotides in the viral RNA multiplication was strengthened by the reversion of the mutated stretches to the wildtype sequence in this area during replication cycles invivo . As the reversion was a stepwise process, the reversion probably occurs by the preferential multiplication of those RNA molecules which have advantageous point mutations in this nucleotide stretch. Alternatively, recombination with the homologous area in the wildtype M-RNA transcript could be the underlying mechanism for reversion. When Band M-RNA transcripts with identical mutations in this region will be used as inoculum, one can discriminate between these two possibilities. Whether the nucleotide stretch Indeed forms the supposed hairpin, which may protect the RNA against degradation or alternatively have a specific signal function In RNA replication or encapsidation, remains an interesting question to be solved. A major advantage for further studies on the RNA replication of CPMV Is that this virus has a genome consisting of two RNA molecules. B-RNA can replicate Independently and contains the genetic information needed for the RNA multiplication. M-RNA contains the information for the cell to cell transport of the virus and needs B-RNA encoded proteins for its multiplication. This dependency must be its multiplication. This dependency must be exploited In future experiments by modifying the M-RNA sequence at specific places while keeping the B-RNA encoded replication machinery intact. Such studies may considerably contribute to the fully understanding of the complex RNA replication mechanism of CPMV and related viruses.