Genetic engineering with plant viruses, and their potential as vectors

Strategies for expression of foreign genes in plants. Potential use of engineered viruses

Advances in gene transfer techniques for higher plants have already permitted important achievements towards crop protection and improvement using recombinant DNA technology. Besides plant genetic engineering, the possible use of plant viruses to express foreign genes could be of considerable interest to plant biotechnology. However, insuring containment of engineered viruses for environmental use is an important safety issue that must be addressed.

Detection of CaMV gene I and gene VI protein products in vivo using antisera raised to COOH-terminal β-galactosidase fusion proteins

Specific antisera were prepared to the inclusion body protein (gene VI product) and the gene I product of cauliflower mosaic virus (CaMV). Translational fusions between the lacZ gene and gene VI or gene I were constructed by cloning the relevant DNA fragments into the expression vectors pUR290, pUR291 or pUR292. Large amounts of fusion protein were synthesized when the inserted DNA fragment was in frame with the lacZ gene of the expression vector. These fusion proteins were used to raise specific antisera to gene VI and gene I proteins of CaMV. Antiserum to the gene VI product detected a range of proteins in crude extracts and in a subcellular fraction enriched for virus inclusion bodies. This range of proteins was further shown to be related to gene VI by Staphylococcus aureus V8 partial proteolysis. Antiserum to the gene I product detected viral specific proteins of 46, 42 and 38 K in preparations of CaMV replication complexes from infected plants but not in any other subcellular fraction.

Gene Transfer in Crop Improvement

Transfer of genes between plant species has played an important role in crop improvement for many decades. Useful traits such as resistance to disease, insects, and stress have been transferred to crop varieties from noncultivated plants. Recombinant DNA methods greatly extend (even outside the plant kingdom) the sources from which genetic information can be obtained for crop improvement. Gene transfer systems based on recombinant DNA are available for several crop species and are under development for others. The concerted use of traditional and more recent methods for plant genetic manipulation will contribute to crop improvement.

Synthesis of viral DNA forms in Nicotiana plumbaginifolia protoplasts inoculated with cassava latent virus (CLV); evidence for the independent replication of one component of the CLV genome

Totipotent leaf mesophyll protoplasts of Nicotiana plumbaginifolia, Viviani were inoculated with cassava latent virus (CLV) or with full length copies of CLV genomic DNAs 1 and 2 excised from replicative forms of M13 clones. Virus specific DNAs began to appear 48-72h after inoculation with virus or cloned DNAs, coincident with the onset of host cell division. Infected cells accumulated supercoiled forms of DNAs 1 and 2 as well as progeny single-stranded (ss) virion (+) sense DNAs representing each component of the genome. Both supercoiled and ss molecules were synthesised by cells inoculated with cloned DNA 1 alone but DNA 2 failed to replicate independently.

Bidirectional transcription of maize streak virus DNA and identification of the coat protein gene

Three RNA transcripts encoded by maize streak virus DNA were detected in polyadenylated RNA from virus-infected maize leaves. Two of the transcripts, a major 0.9 kb and a minor 1.05 kb RNA, were mapped on the virion (+) sense DNA and the other minor transcript of 1.2 kb was mapped on the complementary (-) sense DNA, demonstrating that transcription of MSV DNA was bidirectional. The two virion sense transcripts were 3' coterminal at nucleotide 1114 but had 5' termini at nucleotides 2682 and 163 respectively. Virus-specific polyadenylated RNA translated in vitro to produce a 28,000 MW polypeptide, specifically immunoprecipitable by antiserum raised against whole virus. The mRNA for this protein was mapped by hybrid-arrested translation to the long open reading frame in virion sense DNA whose potential amino acid composition, calculated from nucleotide sequence data, closely agreed with that determined experimentally for the coat protein.

Plant-virus-based vectors for gene transfer will be of limited use because of the high error frequency during viral RNA synthesis

The error frequency during the RNA replication of alfalfa mosaic virus (AMV) was calculated to be significantly higher than 10(-5). It may be expected that RNA synthesis in general will have low fidelity compared to DNA synthesis. The low fidelity of RNA replication will severely restrict the usefulness of vectors for genetic engineering which are based on RNA viruses, viroids or DNA viruses which are replicated via an RNA intermediate (e.g. caulimoviruses). Spontaneous mutants selected by host shift were found to be much less stable than UV-induced mutants. This difference points to variations in fidelity during RNA synthesis, probably due to the local sequence of the template.

Sequence homology among the coat proteins of gemini viruses

The coat proteins of the gemini viruses - African cassava latent virus, tomato golden mosaic virus and maize streak virus - are shown to have reasonable to good amino acid sequence homology. It is suggested that the maize streak virus genome is ancestral and the bipartite genomes of the other viruses evolved from it.

Physical structure and genetic organisation of the genome of maize streak virus (Kenyan isolate)

The structure of the maize streak virus genome (Kenyan isolate, MSV-K), as determined from the sequence of clones obtained from DNA isolated from virus particles, is composed of one major DNA component of about 2.6 kb. MSV virion DNA is partially double-stranded, composed of a full-length virion (V) strand and a short (70-80b) primer (P) strand. The primer strand has a fixed 5'-end capped with alkaline labile material, presumably 1-2 ribonucleotides. The MSV genome has two major coding regions oriented on opposite strands and flanked by two small intergenic regions. The coding region on the P strand is composed of two major open reading frames (ORFs), arranged in tandem and in the same reading frame. Because the predicted protein derived from a composite of these two ORFs closely corresponds to the product from a single ORF in the cassava latent virus genome, it is likely that this region encodes two proteins with common amino-termini, one a read-through product of the amber codon terminator in the first ORF. The intergenic regions contain potential transcription start and stop signals oriented in the direction of the two opposing coding regions. Considerable DNA sequence heterogeneity was observed, mostly silent or conservative third base substitutions in coding regions and base substitutions, small insertions and small, close-range transpositions in intergenic regions.