Nucleotide sequence data was entered and analysed on a PDP 11/4 computer using a package of programs obtained from R. Staden, MRC Laboratory of Molecular Biology, Cambridge, UK (Staden, 1980).
At any rate, the truth can now be told:
Title: Studies on the Genome Structure of Neurovirulent and Attenuated Polioviruses
Author: Cann, Alan James
Publisher: University of Leicester
Description: Thesis submitted for the degree of Doctor of Philosophy at the University of Leicester, 1984.
And if you want to read the full thing, here it is.
Abstract: The RNA genomes of neurovirulent and attenuated type 3 polioviruses have been cloned in E. coli using an efficient RNA.cDNA hybrid technique. The complete nucleotide sequence of the vaccine-associated neurovirulent revertant P3/119 and, in collaboration with others, the attenuated vaccine strain P3/Leon 12 a1b, have been determined. These have been compared with that of the neurovirulent parent strain P3/Leon/37. Ten nucleotide sequence differences were observed between the parent P3/Leon/37 and the vaccine P3/Leon 12 a1b, three of which resulted in amino acid substitutions. Between the vaccine and the revertant P3/119, seven nucleotide sequence differences were observed. Three of these resulted in amino acid substitutions. The possible significance of individual nucleotide sequence differences to the attenuation of and reversion to neurovirulence in poliovirus type 3 is discussed. The nucleotide sequence of P3/Leon 12 a1b was the first to be determined for a type 3 poliovirus. Comparison of this sequence with published type 1 sequences has demonstrated the extent of the molecular homology between them.
The major objective of this study was to identify the nucleotide sequence differences which account for the neurovirulent or attenuated phenotype of three closely related strains of poliovirus type 3. To achieve this, an efficient RNA.cDNA hybrid cloning method was devised. Although this was not the first report of hybrid cloning, previous unfavourable comments on its efficiency (Wood and Lee, 1976; Zain et al., 1979) and suggestions that the method given rise to cloning artefacts (Okayama and Berg, 1982) have almost certainly discouraged widespread use. Thorough investigation of each of the manipulative steps involved has shown that these problems can be overcome (Cann et al, 1983). Comparable in efficiency to the more used double-stranded cDNA cloning technique, the hybrid method has the advantages of experimental simplicity and that cDNA clones corresponding to entire virus genome can be obtained from a single experiment. The method has proved to be ideally suited to the molecular cloning of picornavirus genomes. It is possible that the transformation efficiency of RNA.cDNA hybrids could be further increased by treatment with E. coli DNA ligase and DNA polymerase I before ligation, thus carrying out repair of the hybrid molecule in vitro, as in the method of Okayama and Berg (1982). This modification has not yet been tested.
Together with the work of Dr G. Stanway on the neurovirulent strains P3/Leon/37 and P3/119, hundreds of cDNA clones were examined and more than 22 kbp of nucleotide sequence determined. These experiments resulted in the identification of a mall number of mutations in the genomes of the strains studied which must be responsible for their differences in neurovirulence. However, it has not yet been possible to identify the individual mutations involved in attenuation and reversion and further experiments are currently in progress. These experiments represent a number of different approaches. Firstly, the sequence of other neurovirulent vaccine revertants are being determined, to ascertain whether the mutations observed in P3/119 are shared by other strains. Secondly, the work of Racaniello and Baltimore (1981a) has demonstrated that the construction of recombinant virus genomes in vitro, at the level of cloned cDNA, is possible. Transfection of susceptible cells with these recombinant genomes gives rise to new, viable viruses with a defined set of mutations which can then be examined phenotypically. Initial experiments with recombinants between the neurovirulent strain P3/Leon/37 and the vaccine strain P3/Leon 12 a1b and also between the vaccine strain and the neurovirulent revertant P3/119 are in progress. Final proof that the mutations involved in attenuation and reversion have been accurately identified could be provided by the construction of an attenuated strain by recombination between P3/Leon/37 and P3/119. Parallel studies on neurovirulent and attenuated type 1 strains should help to explain the different stabilities of the type 1 and type 3 vaccines and may suggest how the type 3 vaccine can be modified to improve Stability. Finally, it is hoped to use site-directed mutagenesis of cloned virus genomes ln vitro to produce strains with specific biological properties.
Although the main aim of the work presented here has been the investigation of the molecular basis of attenuation in poliovirus, the information obtained has wider significance. The complete nucleotide sequence of P3/Leon 12 a1b was the first to be determined from a type 3 poliovirus. This has been compares with that of type 1 and the extent of the molecular homology between the demonstrated (Stanway et al, 1983a). As part of a larger study based on the analysis of monoclonal antibody resistant mutants, the major neutralizing antibody binding site of poliovirus type 3 has been identified (Minor et al, 1983). The nucleotide sequence information obtained is also being used currently in the design of synthetic antigenic peptides, a development which may hold many advantages for the prevention and perhaps treatment of poliomyelitis and other related picornaviral infections. At the outset of the work described in this dissertation, it was difficult to envisage that improved alternative to the Sabin vaccines would ever be a realistic proposition. It now seems that the immediate future holds just such a prospect.