**6.2 Homologous recombination**

The development of techniques for precise gene KO by homologous recombination has been adapted for site-specific gene insertion or gene knock-in (KI). Similar to the KO vectors described above, the gene to be introduced is flanked by DNA sequences that are identical to regions of the genome to be targeted. After introduction of the heterologous DNA vector, the regions of DNA sequence identity are aligned with the target genome sequence, after which the homologous recombination machinery catalyzes reactions which swap the endogenous genomic DNA with the heterologous DNA within the construct. If the recombination event occurs with high fidelity, the gene of interest will functionally replace the gene that was removed [24]. Similar to gene KO, this approach is much less efficient than random TG integration. Therefore, vectors carrying the TG are often designed to incorporate selectable markers, similar to those used for gene KO, to allow the identification of cells carrying the desired TG in the genome. In this case, both targeted and randomly integrated TGs may be selected, requiring additional assays, such as PCR or Southern blotting, to distinguish between sequence-specific and random insertion events [30].

As described above for gene KO, specific targeting is more efficient in murine ES cells, which express the enzymatic machinery necessary for homologous recombination, than is currently possible for pig primary cells. Insertion of heterologous DNA into the mammalian genome is believed to be driven by endogenous DNA repair mechanisms, presumably in response to DNA breaks, whether randomly via NHEJ, or specifically via homologous recombination [66]. The deliberate introduction of double-stranded DNA breaks at the desired integration site should therefore improve the efficiency of heterologous DNA insertion by activating and recruiting the cellular repair machinery. Application of ZFN, TALEN and CRISPR technologies have shown that homologous recombination efficiencies are improved when one or more doublestrand DNA breaks are introduced into the genome at the desired site of insertion [67] with CRISPR exhibiting bi-allelic targeting rates as high as 90% [68]. The use of these more advanced genome engineering tools has greatly improved the rates and specificity of both gene deletion and gene insertion in genomes.

Gene insertion by homologous recombination for xenotransplantation has not advanced as rapidly as other approaches, in part due to the challenges of using primary porcine cells, for which the efficiencies can be extremely low, particularly with large DNA constructs. Use of improved genome engineering tools with increased targeting efficiencies have already been applied in pig and will continue to grow in impact [69]. However, even with improved efficiencies of gene insertion, breeding pigs with multiple, independently segregating loci is challenging. The number of litters required to produce animals bearing all of the genetic modifications greatly increases with each additional locus, which can be impractical for large animals such as pigs. Therefore, the ability to insert the maximal amount of genetic information into the genome in the minimal number of steps, as discussed below, is highly valuable.

#### **6.3 Multigenic insertion**

The most straightforward example of multigenic targeting at a single site takes advantage of the observation that multiple DNA vectors co-transfected into cells will tend to insert together at a given genome site. This approach was used successfully to simultaneously introduce as many as five large transgenes in a single step into porcine cells [70]. Although difficulties in producing mature cloned pigs limited the study to animals with fewer integrated genes, the study demonstrated the feasibility of rapidly making animals with multiple TG.

Another, relatively less complicated, way to introduce multiple TG is to generate large DNA constructs bearing multiple TGs for integration at random into the genome [71]. This greatly reduces the complexity of screening, while increasing the efficiency of insertion, but still relies upon random integrants which can have variable TG copy numbers and expression levels.

Multiple genomic regions have been defined, such as Rosa26, which allow expression of heterologous genes at similar levels regardless of cell type. These "safe harbor" regions are believed to have a chromatin structure that is more easily accessible to the gene expression machinery, regardless of cell type. Targeting at porcine homologs of murine safe harbor sites such as Rosa26 has been described and demonstrates the utility of this approach [70, 72]. On a practical level, the use of safe harbor sites yields more reproducible gene expression than random TG insertions, so fewer lineages are required to select animals with desired TG levels. Furthermore, the defined location and copy number of TG inserted at a safe harbor site makes breeding and genotyping more straightforward, and is expected to provide a less complicated regulatory pathway for clinical use.

As DNA synthesis and assembly has improved, increasingly large DNA constructs encoding a variety of TG are possible, however, as the size of the DNA increases, the rate of insertion decreases. Considering the lower insertion rates observed for large animals such as pig, alternate approaches are necessary to incorporate larger DNAs into the genome.

#### **6.4 Site-specific recombination**

Bacteriophage- and yeast-derived site-specific recombinases are, as the name suggests, proteins which catalyze recombination between two specific DNA recognition sites, small (<50 base pair) sequences that are unique to the recombinase being

**41**

**8. Conclusions**

*Genome Engineering for Xenotransplantation DOI: http://dx.doi.org/10.5772/intechopen.84782*

**7. Future needs**

used. The recombination event is highly efficient, in some cases eliminating the need for selection genes, and allowing large DNA constructs to be inserted at a much

One limitation for the use of site-specific recombination is the need for a recognition sequence to be present in the target genome at the desired locus. This requires a preceding step in which the recognition site is engineered into the genome using less efficient homologous recombination. Therefore, the gain in efficiency for introduction of large DNA constructs may be offset by the need for insertion of the recombinase recognition site into the genome. Despite this constraint, the potential for site-specific recombination into a defined locus has been demonstrated in pigs

The advancements in genome engineering, both in general and in their application to xenotransplantation, have been significant, but many needs remain to be addressed. As new genome engineering tools are identified and further refined, improvement of targeting efficiencies will allow more sophisticated modifications of the pig genome. Ideally, the pig genome will become as readily manipulated as the mouse, allowing researchers to further leverage approaches shown to be effective in murine models. One major technological difference in the genetic modification of mice and pigs (and many other mammalian species) is the lack of ES cells possessing significant rates of homologous recombination that can be grown in culture for extended periods and subjected to multiple manipulations without losing the ability to produce viable pigs. Efforts to identify natural or induced pluripotent stem cells (iPSC) suitable for these purposes have been described, but have yet to demonstrate practical application for porcine genome engineering [74]. Ongoing work will be required to

The function of the TGs themselves can also be further improved. The majority of TG constructs used in pigs have used constitutive promoters to drive high level expression of the proteins encoded by the TGs. In some cases, such as CRPs, this approach may be useful, however, overexpression of TGs which inhibit critical immune processes may create risks of immune deficiency and infections. For this reason, use of expression control elements which can turn on and off TG activity is of increasing interest in xenotransplantation. There are multiple examples of inducible promoters employed in mice which can be controlled by exogenously applied small molecules (such as the tetracycline repressor system), or by endogenous signals (such as promoters for innate immune response genes) [75]. Advanced DNA synthesis and assembly methods also allow synthetic biology approaches to create novel signaling pathways and networks not present in nature.

Immune tolerance is another very active area in xenotransplantation research. As the molecular mechanisms controlling the balance between immunity and tolerance are further elucidated, manipulation of the human immune system itself to specifically reduce or eliminate responses to porcine targets, while leaving intact immunity to infectious diseases, will help overcome xenorejection. Multiple approaches are currently being tested and genome targets identified to encourage human immune

The speed and ease of genome engineering technologies has helped to overcome many of the limitations for the use of pig organs for xenotransplantation. Despite recent

higher frequency than possible for homologous recombination [73].

[70] and provides a route for more rapid complex genetic modifications.

identify and validate cells which meet these needs.

tolerance of porcine cells and tissues [76, 77].

#### *Genome Engineering for Xenotransplantation DOI: http://dx.doi.org/10.5772/intechopen.84782*

used. The recombination event is highly efficient, in some cases eliminating the need for selection genes, and allowing large DNA constructs to be inserted at a much higher frequency than possible for homologous recombination [73].

One limitation for the use of site-specific recombination is the need for a recognition sequence to be present in the target genome at the desired locus. This requires a preceding step in which the recognition site is engineered into the genome using less efficient homologous recombination. Therefore, the gain in efficiency for introduction of large DNA constructs may be offset by the need for insertion of the recombinase recognition site into the genome. Despite this constraint, the potential for site-specific recombination into a defined locus has been demonstrated in pigs [70] and provides a route for more rapid complex genetic modifications.
