Oncolytic viruses can and have been used to deliver exogenous genetic material, giving them much needed flexibility and complexity as a cancer treatment and augmenting their potential as viable cancer therapies [
21]. The ability to engineer the genome of a replicating virus to carry foreign material is referred to as “arming”. This strategy has been used in a variety of ways to alter viral selectivity, potency, safety, and utility [
22,
23].
Like nonreplicating viruses, early armed oncolytic viruses controlled expression of foreign genes by using an exogenous constitutive or conditional promoter. This approach allowed for high levels of expression and, in the case of conditional promoters, some level of specificity. As the field advanced, researchers demonstrated that expression of exogenous genes could be achieved by using the endogenous expression machinery of the virus itself [
24–
26]. This latter method proved to have some advantages, the first being that it decreased the size of the foreign DNA that could be inserted into the viral genome. This outcome was beneficial given that viruses have genomic size constraints that greatly effect packaging efficiency. The second advantage of utilizing endogenous expression machinery is that depending on the endogenous promoter used, the timing and level of expression of the foreign gene could be controlled [
24–
26]. This not only allowed for efficient expression of the exogenous gene but could be used as an added measure of safety. Because the level of expression could be modulated depending on the endogenous promoter used, researchers and clinicians could have greater control over the levels of the armed therapeutic/exogenous gene. Furthermore, by utilizing an endogenous late promoter, it is possible to limit the expression of the gene to late in the viral cycle which, presumably, for an oncolytic virus would never be reached in a nonpermissive cell (normal cell) and, therefore, confine expression only to target cancer cells. Importantly, an armed therapeutic virus can express multiple genes [
27] that potentially increase the therapeutic benefit of the virus and/or enable the treating physician with the opportunity to track the virus location (e.g., via imaging [
28]) and viability (e.g., through the expression of a protein that can be detected in the blood or is secreted from the treated patient), as a means to better understand when additional treatments are warranted.
Arming is traditionally carried out by manipulating the viral genome and usually calls for a well characterized system. The novel viruses resulting from directed evolution will make rational insertion of therapeutic moieties extremely challenging, especially when one considers that they would not want to negatively impact the replication capabilities of these viruses which are critical to their clinical benefit. Clearly, innovative approaches will need to be taken in order to manipulate viruses derived in this manner to ensure that this does not occur. One such approach taken uses a transposon-based method to “scan” a viral genome for insertion sites which are compatible with the viral lifecycle [
29]. This approach makes it possible to arm these novel viruses with exogenous genes expressed from a foreign promoter or by the inclusion of a splice acceptor from an endogenous promoter [
30]. Although this method, like many advances in the oncolytic field, was originally developed on Adenoviruses, it can be extended to any number of viral species whose genome can be cloned into a plasmid.
Given the potential to derive potent novel viruses from the directed evolution approach, it is very important to keep safety in mind. If these novel viruses are ever to advance as therapeutics, having the ability to control their replication in the treated patient is paramount. To this end, two approaches were taken with ColoAd1. The first was to study the effect on ColoAd1 of two clinically approved antivirals, ribovirin (RBV), and cidofovir (CDV), and the second was to genetically introduce drug sensitivity to the virus by inserting the HSV TK gene into the ColoAd1 genome using the transposon method [
31].
Although there are no approved anti-Ad treatments, both RBV and CDV have been used to experimentally treat Adenoviral infections [
32–
34]. From these studies, RBV was found to be effective on Group C Ads but less so on Group B Ads [
35,
36]. Not surprisingly, ColoAd1, being derived from 2 group B viruses was refractory to RBV treatment. CDV has been shown to have better Anti-Ad activity than RBV and when used to treat ColoAd1 was effective at inhibiting viral replication and spread on both tumor and normal cell lines. Interestingly, ColoAd1 was more sensitive to CDV treatment than either of its 2 parental strains, Ad11p or Ad3 suggesting that this increased CDV sensitivity was an outcome of the directed evolution process and not simply an inherited trait [
31]. An alternative approach to the use of approved antivirals is the insertion of the HSV TK gene into the viral genome to create sensitivity to the approved drug ganciclovir (GCV) [
37]. This approach, made possible by the transposon method of arming was successful in inhibiting ColoAd1 infection of both tumor and normal cells [
29,
31]. The potential for using TK expression to track the virus also makes this approach appealing [
38]. From these studies, it was demonstrated that ColoAd1 could be controlled through outside intervention.
The value gleaned from the ColoAd1 experience goes beyond a promising therapeutic, as it validates a new approach for the oncolytic virus field. The directed evolution method that resulted in ColoAd1 could be applied to other cancer types and other viral families. By utilizing this method, researchers are not longer limited to well characterized systems. Moreover, the ease by which viruses are molecularly manipulated need not be the deciding factor for proceeding. Although Adenoviruses have been covered in this paper, this approach is amendable to all viral families and opens up the possibility of harnessing inherent and novel oncolytic properties from a multitude of human viruses. Unlike more deliberately designed approaches, directed evolution capitalizes on the complexity of the tumor and can be directed towards an outcome that depends predominantly on the selective pressure applied by the researcher. Similar to Ad5-based oncolytic viruses, it is possible to arm these novel viruses without interfering with their lifecycle, thus unlocking the potential to modulate their characteristics or utility. Because these viruses are potentially highly potent, developing ways to control their replication should be a priority.
