Viruses are inherently capable of binding to mammalian cells and transferring genetic information into those cells. Each virus has evolved to utilize the host cell’s machinery in order to replicate itself. The amount of genetic material that can be packaged in each virus is determined, in part, by the structure and volume of its capsid (shell).
Using viruses for vectors takes advantage of this cell targeting and gene expression system. In designing gene vectors, the virus is generally engineered so it cannot reproduce (replication-deficient). This is accomplished by removing a gene from the virus genome that is critical for replication. Removing this gene also creates space to allow the insertion of the gene desired for expression (gene of interest). This vector can now be reproduced by incubating it with cells that can compensate for the gene that was deleted, allowing the virus to replicate within the cell (packaging cell line). In some cases, another virus can supply the missing replication machinery (helper virus). The goal is to end up with a large number of viral particles with the gene of interest, but to not allow the virus to exert any pathogenic properties associated with the whole or wild-type virus.
There are two basic biosafety concerns regarding research using viral vectors. First, it is impossible to completely control cellular processes to ensure that a replication-deficient virus will not naturally gain back genes that it requires for replication (become replication-competent). If a virus becomes replication-competent, it may re-acquire any pathogenic characteristics associated with the wild-type virus and could cause illness.
In addition, the gene of interest may be acquired by viruses or cells not expected to be associated with that gene. Because of these concerns, biosafety containment recommendations are made according to the properties of the wild type virus, and must also take into consideration the nature and necessary containment of the gene of interest.