The advent of genetic engineering facilitated the manipulation of DNA viruses such as phage viruses, insect viruses, and then mammalian viruses including vaccinia virus and adenovirus (2-4,72). This allowed virus genetics to be performed and even new vaccine and gene therapy concepts to be developed. In the mid 1990s it was reported that adenovirus lacking the E1B-55k gene could replicate in tumor cells lacking functional p53 (6). This event has encouraged a number of studies aimed at improving the oncolytic activity of a variety of DNA viruses through genetic engineering (10,14). However, the ability to genetically manipulate RNA viruses proved a little more difficult because a DNA copy of the entire RNA genome was first required. The development of retroviruses as gene therapy vectors then progressed. Retroviruses are single-stranded RNA viruses that contain a core structural protein (gag) an RNA-dependent RNA polymerase (pol), and viral envelope (env) (73). Usually, some of these genes are replaced with a therapeutic gene, such as a suicide gene, to prevent virus replication. One of the most frequently used retrovirus is the Moloney murine leukemia virus (MMLV). Retroviruses, which predominantly infect dividing cells, have also been psuedotyped (for example, using VSV G) to broaden their host range and have been used in a variety of clinical trials to combat cancer (74). It was then discovered that the genomic RNA of positive stranded viruses, such as poliovirus, could directly function as an mRNA when transfected into recipient cells (75). These genomes became translated to give rise to progeny virions. Further, plasmid DNA versions of the positive stranded viruses could also give rise to infectious virions following transfection of the cell. Despite this progress, a problem with generating recombinant negative-stranded viruses, such as VSV, was apparent because their genomic RNAs or their antigenome complements could serve as mRNAs and so neither could be used directly to recover infectious virus (76). The minimal infectious unit for these viruses is the genome complexed with nucleocapsid and RNA-dependent RNA polymerase proteins in a ribonucleoprotein complex. It was subsequently found that the segmented negative-stranded RNA influenza virus, which has eight small genomes, could be assembled with RNPs in vitro and used to transfect cells that were already infected with influenza virus (77,78). Some of these progeny viruses acquired a cloned gene through resortment and could be isolated and studied. Whiereas such approaches were useful for influenza virus analysis, it proved difficult to assemble the 11-kb genome of viruses such as VSV, into RNPs in vitro as a result of their large size. A breakthrough was then made on recovering infectious, cloned rabies virus, the prototype of the rhabdovirus family and relative of VSV (79). This was achieved by placing a cDNA version of the entire rabies genome under control of a vaccinia encoded T7 polymerase. The cDNA was transfected into cells along with vaccinia virus to generate a supply of T7, and the full length antigenomic RNA (positive strand) was generated, along with the subgenomic mRNAs encoding the viral proteins, which in turn further assist replication. This system was subsequently found to work well with VSV and has now been used to recover infectious segmented negative-stranded RNA viruses such as Bunyamwera (80-82). In addition, other oncolytic RNA viruses are now able to be cloned through similar "reverse genetic" approaches. Example include the paramyxoviruses NDV and measles (83,84).
Recovering infectious cloned VSV was first reported from the laboratories of John Rose and Gail Wertz (81,85). A T7 promoter directed synthesis of the full-length negative-stranded RNA and the polymerase was generated from vaccinia virus infection. However, infectious virus was not recoverable from this strategy, possibly because the subgenomic mRNAs generated from this strand hybridized with its parental genomic negative sense partner. Thus, constructs were redesigned to express the antigenomic RNA of VSV and the L, N, and P mRNAs (which were also transfected into cells to supply L, N, and P protein and help replication) were not able to hybridize to the encoded genomic template. Thus infectious VSV could be recovered from DNA and amplified for analysis (81,85).
It was then demonstrated that new transcription units could be inserted into the VSV genome, between established genes (such as the G and L gene products) (86). Nucleotide sequence analysis indicated a conserved 11-23 nucleotide motif present at the beginning and end of each gene. Insertion or addition of the conserved sequence in the 3'-noncoding region of a heterologous gene was found to be sufficient to terminate transcription of the preceding viral gene and promote transcription of the foreign gene. In preliminary studies using the chloramphenicol acetyltransferase (CAT) gene, highlevel expression of the foreign gene was reported (86). Of further importance was that the recombinant viruses were quite stable, and did not rapidly "lose" the gene. It was rapidly noticed in subsequent studies that recombinant VSV designed to express large gene inserts could be potentially created where such limitations associated with other viruses are evident (87). Indeed, VSV was found to increase its length to accommodate extension of its genome. Later studies indicated that recombinant VSV could be generated to contain more than one foreign gene and to increase its genome length to greater than 40% (88).
Was this article helpful?