Purification and properties of cowpea mosaic virus RNA replicase

This thesis concerns the partial purification and properties of an RNA-dependent RNA polymerase (RNA replicase) produced upon infection of Vigna unguiculata plants with Cowpea Mosaic Virus (CPMV). The enzyme is believed to be coded, at least in part, by the virus genome and to be responsible for the replication of the virus RNA.In chapter 1 we describe the scope of the investigations and the motives underlying this thesis.In chapter 2 a literature review is presented of the RNA replicases of viruses containing a single-stranded RNA genome of the plus type. With respect to the prokaryote virus RNA replicases, studies are described on the structure and properties of QB replicase, with special emphasis on the role the individual subunits of the enzyme are playing in the different stages of RNA synthesis. Reviewing the research on animal and plant virus RNA replicases had to be limited necessarily to a description of the isolation and properties of several crude enzyme preparations, since no purified replicases have been obtained and little progress is made with their purification.In chapter 3 we describe the detection of an RNA-dependent RNA polymerase activity which is present in Vigna unguiculata leaves infected with CPMV but not in uninfected leaves. It is shown, that this RNA polymerase activity, which is designated as CPMV replicase and is associated with a membrane fraction, becomes detectable one day after infection and then continues to increase until the fourth day. This membrane-bound replicase activity was found to require Mg 2+ -ions and all four ribonucleoside triphosphates and to be resistant to DNase and actinomycin D. Analysis of the in vitro synthesized RNA products by sucrose gradient centrifugation and treatment with RNases revealed, that the majority consisted of double-stranded RNA species sedimenting at 17S and 20S, probably representing the replicative forms of both virus RNAs. A minor part consists of two single-stranded RNA species, similar in sedimentation rate (26S and 34S) to the virion RNAs. From these results we concluded, that we were dealing with a bound replicase complex most likely representing the replicase involved in virus replication in vivo. Having the final purification of CPMV replicase in view, we were then faced with the solubilization of the enzyme required to continue the purification.In chapter 4 we describe a very gentle and easy method to release the replicase from the membranes without employing detergent. The method consists of a washing procedure involving a Mg 2+ -deficient buffer, and provides several advantages in comparison with other solubilization procedures. Firstly, the solubilized replicase is highly stable, thus facilitating the further purification. Secondly, the release of the replicase from the membranes is rather selective. The majority of proteins is retained in the membrane pellet and the specific activity of the solubilized replicase is increased about 2-3 fold with respect to the membrane-bound replicase. Thirdly, more than 80% of the replicase activity is detached from the membranes. The solubilized replicase can be further purified and freed of endogenous template RNA by DEAE-BioGel. column chromatography to provide a highly stable enzyme dependent on template.In chapter 5 we describe several properties of the DEAE-purified replicase preparation. Replicase activity is not inhibited by a- amanitin, rifampicin, cordycepin, actinomycin D, DNase and orthophosphate but is completely suppressed by pyrophosphate and RNase A plus RNase T 1 . The in vitro RNA synthesis is shown to proceed for at least 15 hours under the following optimal conditions: 8 mM Mg(OAc) 2 or 12 mM MgCl 2 ; 60 mM (NH 4 ) 2 SO 4 , up to 100 mM K(OAc), but KCl as low as possible; pH 8.2; 30 to 34°C; all four ribonucleoside triphosphates present and 5-10 μg of CPMV RNA as template per 15 μg of protein.Having established the optimal conditions for RNA synthesis, we have studied the template specificity using a variety of viral, nonviral and synthetic template RNAs. It is shown that the replicase readily accepts natural RNAs as templates but is unable to efficiently synthesize RNA complementary to the synthetic ribopolymers poly(C), poly(G) and poly(U); poly(A) is able to direct the incorporation of 3 H-UMP, but only at a high concentration (400 μg/ml) and inefficiently with respect to CPMV RNA. Several possibilities to account for the lack of template specificity displayed by CPMV replicase and many other eukaryote replicases, are discussed. It is argued that template specificity does not have to be an intrinsic property of, and a prerequisite for, eukaryote virus RNA replicases to function properly in vivo, taking into account the specific location of the replication process in the cell and the occurrence of host RNA molecules as ribonucleoprotein particles. Moreover, the loss of essential protein factor(s), the possible requirement for primer(s) and the use of non-specific reaction conditions are considered.Initial studies have been carried out on the binding of CM replicase to 32 P-CPMV RNA and, in addition on the size and nature of the in vitro synthesized RNA products. The binding experiments using a nitrocellulose filter technique to detect RNA-protein complexes, demonstrate that the DEAE-purified replicase, but not a corresponding protein preparation isolated from healthy leaves, binds to CM RNA. This binding can be abolished by synthetic poly(A) and poly(U) but not by poly(C), suggesting that the poly(A) on the CM RNA genome comprises a potential part of the replicase binding site. However, further experiments are needed to substantiate this hypothesis.The bulk of the in vitro synthesized RNA was found to consist of 16S RNA and a rather small amount of faster sedimenting RNA (20S-38S), the latter representing single-stranded RNA molecules still attached to their parental template strand. Although about 60% of the RNA products appears to be sensitive to treatment with RNase A plus RNase T 1 , no free, full-length size virus RNA molecules were formed, due to the presence of RNase(s) contaminating the replicase preparation.In chapter 6 we show that the DEAE-purified replicase can be purified further by glycerol gradient centrifugation. This step affords the removal of some proteins predominating in all earlier stages. A final purification of about 150-200 fold relative to the crude extract is achieved. From analysis by polyacrylamide gel electrophoresis of the replicase purified by glycerol gradient centrifugation and of a corresponding protein preparation from mock-infected leaves, we conclude that the replicase still needs additional purification steps to allow its identification. However, the stability of the enzyme seems to offer good prospects to achieve this aim.