**4.1 Factors affecting SCNT efficiency**

SCNT is a technique that requires precise skills in micromanipulation. In addition to technical skills, numerous other factors also impact SCNT efficiency, including donor cell types, quality of the recipient oocytes, cell cycle synchronization of both donor cell and recipient cytoplasm, the epigenetic status of the donor cell, method of reconstructed zygote activation, epigenetic reprogramming following activation, in vitro culture conditions of reconstructed embryos and embryo transfer into surrogate mothers. Here, we will focus on the cell cycle synchronization and methods being used to improve nuclear remodeling and reprogramming during SCNT.

### **4.1.1 Type of cell and synchronization of cell and recipient oocyte**

Different types of donor cells and the different origins of recipient oocyte have been successfully used to produce the cloned animals in domestic species (Table 2). Generally, nuclei from less differentiated donor cells demonstrate greater SCNT efficiency than those

the targeting experiment. Also, it is important to perform southern blotting analysis using both probes to confirm the double-crossover event since homologous recombination can occur at only one end of the targeting vector (Hasty et al., 1991b; Moens et al., 1992). Of the methods of screening for gene targeting in somatic cells, southern blotting is considered to be the golden approach to identify correctly targeted colonies despite being time-consuming

Fluorescence in situ hybridization (FISH) is most commonly used to identify the number of integration sites following random integration of a transgene. FISH has been successfully used to detect the number of bacterial artificial chromosome (BAC) sequences integrated and chromosomal location(s) in mouse ES cells (Yang and Seed, 2003). Although the chromosomal location can be shown directly by FISH, Gómez-Rodríguez et al. (2008) found that screening of BAC-based constructs by FISH can be prone to false positives because of small pieces of

Once the cell line containing the appropriate genetic modification has been identified, those cells can then be used for SCNT as described in Figure 1. SCNT is a process in which the nucleus of a somatic cell is transferred into the cytoplasm of an enucleated oocyte. The hereby reconstructed SCNT embryos can be activated to initiate development and then are transferred into a synchronized surrogate mother immediately or after short-term *in vitro* culture. In addition to the technical factors involved in the SCNT process, the status of the donor cells and the quality of unfertilized oocytes are considered to largely affect the overall

integrated DNA that are below the limits of detection by fluorescent hybridization.

**4. Improvement of the efficiency of cell-based transgenesis via SCNT** 

Since cell-based transgenesis via SCNT relies extensively on both transgenic and SCNT techniques, the efficiency could be improved by increasing SCNT efficiency and/or gene

SCNT is a technique that requires precise skills in micromanipulation. In addition to technical skills, numerous other factors also impact SCNT efficiency, including donor cell types, quality of the recipient oocytes, cell cycle synchronization of both donor cell and recipient cytoplasm, the epigenetic status of the donor cell, method of reconstructed zygote activation, epigenetic reprogramming following activation, in vitro culture conditions of reconstructed embryos and embryo transfer into surrogate mothers. Here, we will focus on the cell cycle synchronization and methods being used to improve nuclear remodeling and

Different types of donor cells and the different origins of recipient oocyte have been successfully used to produce the cloned animals in domestic species (Table 2). Generally, nuclei from less differentiated donor cells demonstrate greater SCNT efficiency than those

and relatively expensive compared to PCR approaches.

**3.4 SCNT using genetically modified somatic cells** 

efficiency of the SCNT. This point will be discussed more below.

**4.1.1 Type of cell and synchronization of cell and recipient oocyte** 

**3.3.3 Fluorescence In situ hybridization** 

transfer efficiency.

**4.1 Factors affecting SCNT efficiency** 

reprogramming during SCNT.

from differentiated donor cells (Hill et al., 2000; Lee et al., 2003). SCNT embryos created from *in vivo*-derived MII oocytes have better developmental potential than *in vitro*maturated MII oocytes; and *in vitro*-maturated oocytes from sexually mature animals have been demonstrated to more efficiently produce cloned embryos than oocytes *in vitro*matured from prepubertal animals (Lai et al., 2002a; Lee et al., 2003; Hyun et al., 2003b).

Since the first SCNT animal was produced by transferring the nuclei from quiescent cells (G0), synchronization of the donor nuclei into G0/G1 is thought to be crucial to the success of SCNT (Wilmut et al., 1997). The G0 cells, with lower transcription activity and different chromatin configuration in contrast to the cells at other stages of the cell cycle, may be more responsive to factors inside the recipient oocyte cytoplasm that impact the nuclear remodeling and reprogramming process following SCNT. The cell cycle of cultured somatic cells can be synchronized by serum starvation, contact inhibition and chemical treatments (Cho et al., 2005; Gibbons et al., 2002). Serum starvation to induce the donor cells into G0 is widely used for the production of cloned transgenic animals (Table 2). However, recent studies have demonstrated that synchronizing the cell cycle of donor cells by serum starvation can cause apoptosis (Dalman et al., 2010) and reduce blastocyst production in cattle (Miranda Mdos et al., 2009). The production of the cloned transgenic calves from nonquiescent fetal fibroblasts demonstrated that the synchronization of the donor cells is not an absolute requirement for SCNT success (Cibelli et al., 1998). However, the relative cell cycle combination of both donor cells and recipient oocyte, not just donor cells, is thought to be important for the maintenance of correct ploidy and the subsequent development of reconstructed embryos (reviewed by Campbell, 1999).

### **4.1.2 Methods to improve nuclear remodeling and reprogramming**

Following transfer of a donor nucleus into a recipient oocyte cytoplasm and subsequent activation, nuclear remodeling events of the chromatin structure, such as changes of DNA methylation patterns and histone modifications, result in the reprogramming of gene expression to recapitulate developmental patterns observed in a normal fertilized embryo (Whitworth and Prather, 2010). In contrast to less differentiated donor nuclei, the relatively high level of DNA methylation and low histone acetylation exist in the chromatin of the highly differentiated nuclei. These epigenetic modifications are used to maintain the temporal and spatial patterns of gene expression specific to the cell type or developmental stage. When differentiated nuclei are transferred into the enucleated oocyte cytoplasm, correctly establishing normal patterns of zygotic gene expression is crucial to the full term development of SCNT animals. However, numerous studies have demonstrated improper reprogramming of genes in embryos and tissues of domestic animals following SCNT (Wrenzycki et al., 2001; Pfister-Genskow et al., 2005; Aston et al., 2010; Ross et al., 2010b). Thus, various strategies are under development to facilitate and promote appropriate nuclear reprogramming of the transferred nucleus following nuclear transfer or to preprogram the genome of the donor nucleus prior to SCNT. Of these strategies, one widely used is to treat reconstructed SCNT embryos, but not the donor cells, with histone deacetylase inhibitors (HDACi), such as trichostatin A (TSA), 6-(1,3-dioxo-1H, 3H-benzo [de] isoquinolin-2-yl)-hexanoic acid hydroxyamide (Scriptaid), sodium butyrate and valproic acid. It has been demonstrated that histone deacetylase inhibitor treatment after SCNT can improve both *in vitro* development of SCNT embryos to the blastocyst stage and *in vivo* development to term following embryo transfer (Li et al., 2008; Cervera et al., 2009; Zhao et al., 2009a, 2010; Das et al., 2010; Himaki et al., 2010a; Miyoshi et al., 2010). However, these treatments may be with some level of toxicity, as one group has reported that offspring from TSA-treated rabbit embryos did not survive to adulthood (Meng et al., 2009). The exact mechanism by which the HDACi treatment significantly improves the cloning efficiency remains largely unknown, although it has been shown to increase levels of global histone acetylation after HDACi treatment which may subsequently change the structure of chromatin and improve nuclear reprogramming (Shi et al., 2008; Iager et al., 2008; Zhao et al., 2010; Das et al., 2010).

Accruing investigations have demonstrated that abnormal DNA methylation patterns contribute to the lower developmental competency of SCNT derived embryos (Kang et al., 2001, 2002; Bourc'his et al., 2001; Santos et al., 2003; Wrenzycki et al., 2006), suggesting the inability of oocyte to fully restore the DNA methylation pattern of differentiated donor nuclei to that of normal totipotent 1-cell stage embryos. Thus, the second method used to assist epigenetic reprogramming is to reduce the DNA methylation level in donor cells or reconstructed embryos by treating them with the DNA methyl-transferase inhibitor such as 5-aza-2′-deoxycytidine (5-aza-dC). Unfortunately, previous studies have demonstrated that 5-aza-dC treatment of donor cells or cloned embryos does not improve the *in vitro* and *in vivo* development of SCNT derived embryos (Enright et al., 2005; Tsuji et al., 2009). However, combining the treatment of donor cells and embryos with both TSA and 5-aza-dC resulted in improved blastocyst development (Ding et al., 2008). Furthermore, an additional study has demonstrated enhanced gene targeting frequency in ES cells with low genomic methylation levels, suggesting the epigenetic status of targeted loci may influence the efficiency of gene targeting by affecting the accessibility for the homologous recombination machinery (Domínguez-Bendala and McWhir, 2004).

An additional strategy to promote developmental reprogramming of cloned embryos is to treat them with latrunculin A (LatA), an actin polymerization inhibitor. One group has reported that post-activation treatment with LatA is effective to improve *in vitro* developmental capacity of gene-modified cloned miniature pig embryos and embryos treated with LatA have the ability to develop into fetuses (Himaki et al., 2010b). Prereprogramming donor nuclei prior to SCNT has also been attempted. Rathbone et al (2010) reported that pretreatment of permeabilized ovine fetal fibroblasts with a cytoplasmic extract produced from germinal vesicle (GV) stage *Xenopus laevis* oocytes improves the live birth rate, but not development to blastocyst stage or pregnancy rate following embryo transfer.

### **4.2 Methods to improve gene transfer efficiency**

Production of transgenic domestic animals has been widely accomplished, however, several limitations remain. Typically, plasmid based DNA constructs are limiting in the size of exogenous DNA to transfer and creating animals with targeted genetic modifications has been significantly more challenging. Thus, it is important to continue development of new strategies that broaden application and increase the efficiency of creating targeted genetic modifications in domestic animals.

### **4.2.1 Artificial chromosomes as DNA transfer vector**

In contrast to plasmid based vector, artificial chromosomes have the capacity to carry Megabase-sized pieces of DNA that are maintained as autonomous, replicating chromosomes. Artificial chromosome vectors include a centromere, two telomeres and origins of replication (Robl et al., 2007). A 10 Mb human artificial chromosome (HAC) vector

these treatments may be with some level of toxicity, as one group has reported that offspring from TSA-treated rabbit embryos did not survive to adulthood (Meng et al., 2009). The exact mechanism by which the HDACi treatment significantly improves the cloning efficiency remains largely unknown, although it has been shown to increase levels of global histone acetylation after HDACi treatment which may subsequently change the structure of chromatin and improve nuclear reprogramming (Shi et al., 2008; Iager et al., 2008; Zhao et

Accruing investigations have demonstrated that abnormal DNA methylation patterns contribute to the lower developmental competency of SCNT derived embryos (Kang et al., 2001, 2002; Bourc'his et al., 2001; Santos et al., 2003; Wrenzycki et al., 2006), suggesting the inability of oocyte to fully restore the DNA methylation pattern of differentiated donor nuclei to that of normal totipotent 1-cell stage embryos. Thus, the second method used to assist epigenetic reprogramming is to reduce the DNA methylation level in donor cells or reconstructed embryos by treating them with the DNA methyl-transferase inhibitor such as 5-aza-2′-deoxycytidine (5-aza-dC). Unfortunately, previous studies have demonstrated that 5-aza-dC treatment of donor cells or cloned embryos does not improve the *in vitro* and *in vivo* development of SCNT derived embryos (Enright et al., 2005; Tsuji et al., 2009). However, combining the treatment of donor cells and embryos with both TSA and 5-aza-dC resulted in improved blastocyst development (Ding et al., 2008). Furthermore, an additional study has demonstrated enhanced gene targeting frequency in ES cells with low genomic methylation levels, suggesting the epigenetic status of targeted loci may influence the efficiency of gene targeting by affecting the accessibility for the homologous recombination

An additional strategy to promote developmental reprogramming of cloned embryos is to treat them with latrunculin A (LatA), an actin polymerization inhibitor. One group has reported that post-activation treatment with LatA is effective to improve *in vitro* developmental capacity of gene-modified cloned miniature pig embryos and embryos treated with LatA have the ability to develop into fetuses (Himaki et al., 2010b). Prereprogramming donor nuclei prior to SCNT has also been attempted. Rathbone et al (2010) reported that pretreatment of permeabilized ovine fetal fibroblasts with a cytoplasmic extract produced from germinal vesicle (GV) stage *Xenopus laevis* oocytes improves the live birth rate, but not development to blastocyst stage or pregnancy rate following embryo

Production of transgenic domestic animals has been widely accomplished, however, several limitations remain. Typically, plasmid based DNA constructs are limiting in the size of exogenous DNA to transfer and creating animals with targeted genetic modifications has been significantly more challenging. Thus, it is important to continue development of new strategies that broaden application and increase the efficiency of creating targeted genetic

In contrast to plasmid based vector, artificial chromosomes have the capacity to carry Megabase-sized pieces of DNA that are maintained as autonomous, replicating chromosomes. Artificial chromosome vectors include a centromere, two telomeres and origins of replication (Robl et al., 2007). A 10 Mb human artificial chromosome (HAC) vector

al., 2010; Das et al., 2010).

transfer.

machinery (Domínguez-Bendala and McWhir, 2004).

**4.2 Methods to improve gene transfer efficiency** 

**4.2.1 Artificial chromosomes as DNA transfer vector** 

modifications in domestic animals.

containing the entire unarranged sequences of the human immunoglobulin heavy and light chain loci (1-1.5 Megabase for each locus) have been transferred into bovine fibroblasts using a microcell-mediated chromosome transfer approach. Following SCNT using selected cells, trans-chromosomal cloned bovine offspring were produced that expressed human immunoglobulin proteins in the blood. HAC was retained as an independent chromosome with the proportion of cells ranged from 78 to 100% in most animals and HAC retention rate has not changed over several years. This system provides a useful tool to produce human therapeutic polyclonal antibodies using trans-chromosomal cloned domestic animals (Kuroiwa et al., 2002; Robl et al., 2007). In domestic animals, the swine artificial chromosome (SAC) (about 310 kb) containing pig centromeric DNA and the neomycin resistance gene was constructed and introduced into pig cell lines, and one positive clone was characterized, showing the possibility for producing transgenic pigs for xenotransplantation and other purposes (Poggiali et al., 2002).

### **4.2.2 Zinc finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALNs)**

The low frequency of homologous recombination hampers rapid progress and wide application of gene targeting in domestic animals. The generation of a site-specific doublestranded DNA break (DSB) within the desired locus can facilitate gene targeting by resulting in additions and deletions causing inactivation of gene function. Naturally occurring DNA-binding proteins, including zinc finger proteins (ZFPs) and meganucleases, have been engineered to bind site-specific DNA sequence. Zinc finger nucleases (ZFNs) combine the specific DNA-binding domain (ZFP) with the non-specific cleavage domain of the restriction endonuclease *FokI* and offer powerful tools to create a site-specific DSB to facilitate local homologous recombination. The ZFN induced DSB can lead to incorporating exogenous DNA in a site specific manner by utilization of a homologous recombination targeting vector that overlaps with the DSB region. Additionally, DSBs repaired by nonhomologous end-joining (NHEJ) can result in loss of a single nucleotide, multiple nucleotides, or small regions; all capable of rendering a targeted gene dysfunctional. For increasing the specificity of DNA binding, multiple zinc fingers, each recognizing and binding to a 3-bp sequence of DNA nucleotides can be linked in tandem to recognize a unique genomic locus. In human somatic cells, custom-designed ZFNs yielded more than 18% targeting efficiency at the X-linked interleukin-2 receptor gamma gene locus and about 7% of the cells possessed a bi-allelic gene modification (Urnov et al., 2005). ZFNs can also promote the addition of novel DNA sequence into a targeted endogenous locus of human cells at a frequency ranging from 5% to 15% depending on the size of extra-chromosomal DNA (Moehle et al., 2007). Thus, the efficiency of gene targeting using ZFNs offers the possibility of gene therapy for human genetic disease and an approach for improvement of genetic engineering in domestic animals. Recently, targeted gene disruption of exogenous EGFP gene was achieved in porcine somatic cells using ZFNs (Watanabe et al., 2010) and transgenic EGFP knockout pigs were produced using ZFNs followed by SCNT (Whyte et al., 2011).

The appropriate design of a site-specific ZFN is critical to successfully introduce ZFNmediated genetic modifications. Of the available zinc-finger engineering methods, modular assembly is the most easily performed method, but is also associated with high failure rates to yield a functional three–zinc finger array for the majority of potentially targetable sites (Ramirez et al, 2008). The second method is to combine selection-based methods with ZFNs made using modular assembly. While effective, this method is labor intensive and requires additional expertise. The third method, referred to as OPEN (Oligomerized Pool ENgineering) is based on bacterial 2-hybrid (B2H) selection and has been proven to be a rapid platform for plant and human cells with high targeting efficiency (ranging from 1 to 50% at different loci) and less toxicity compared to modular assembly system. However, utilization of OPEN requires an archive of pre-selected zinc-finger pools and *E. coli* selection (Maeder et al., 2008). Furthermore, due to the challenge of engineering the endonucleases, orthophenanthroline (OP, a DNA cleaving molecule) was conjugated with triplex-forming oligonucleotides (TFOs, sequence-specific binding capacity) to induce targeted DSBs and stimulate mutations at the target site in approximately 10% of treated human cells (Cannata et al., 2008). TFO conjugating to OP or other DNA cleaving molecules may provide a useful tool to induce targeted gene modification because triplex-forming sequences are frequent in mammalian genes. While ZFN-driven gene targeting can be much more efficient than homologous recombination-based methods, the design and development of highly specific ZFNs remain difficult because of the lack of a simple correspondence between amino acid sequence and DNA recognition sequence.

Recently, several groups have shown that transcription activator–like effectors (TALEs) from the bacterial genus *Xanthomonas* contain a central domain of tandem repeats that can be readily engineered to bind virtually any DNA sequence (Boch et al., 2009; Christian et al., 2010; Morbitzer et al., 2010). The structure of the central protein domain, highly conserved in all the known TALEs, includes 17.5 tandem repeats with 34 amino acids per repeat. In each repeat monomer of a TALE, only amino acid positions 12 and 13 are hypervariable (repeat variable diresidues) (Boch et al., 2011), which can specifically recognize a single nucleotide in the target site (Boch et al., 2009; Moscou et al., 2009). Thus, the correspondence between each repeat variable diresidues and the binding nucleotide in DNA sequence opens the possibility to create novel sequence-specific DNA binding proteins by rearrangement of TALE repeats. The engineered hybrid TALE nucleases (TALNs), produced by fusion of the *FokI* endonuclease domain with the high-specificity DNA-binding domains of TALEs, can bind and create targeted DSBs in tobacco and yeast (Mahfouz et al., 2011; Li et al., 2011), showing the feasibility of engineering TALE-based hybrid nucleases capable of generating site-specific genome modification. Recently, Miller et al. (2011) reported the generation of discrete edits or small deletions within endogenous human NTF3 and CCR5 genes and the insertion of 46-bp sequence at CCR5 locus into the genome of human K562 cells using designed TALNs, demonstrating the effective application of TALNs to modify endogenous genes. While the simple DNA-binding code of TALEs enables easier design strategies as compared to ZFPs, the repetitive nature of TALE DNA-binding domains results in difficulty to efficiently synthesize new TALEs by currently used vector construction methods. To overcome this problem, Zhang et al. (2011) recently developed a new strategy to construct repeat domains of TALEs by hierarchical ligation.

### **4.3 Induced pluripotent stem (iPS) cells**

Owing to the lack of ES cells in domestic animals, it is difficult to replicate strategies routinely used to create genetically modified mice. As an alternative, cell-based transgenesis via SCNT is currently used to produce genetically modified domestic animals. Recent advancements in the ability to generate induced pluripotent stem (iPS) cells may open another potential strategy to improve the efficiency of SCNT in domestic animals. Induced pluripotent stem cells in mice and human were successfully generated by reprogramming somatic cells with viral delivery of a combination of four defined transcription factors including Sox2, Oct4, Klf4, and c-Myc or Sox2, Oct4, Nanog, and Lin28 (Takahashi and Yamanaka, 2006; Takahashi et al., 2007; Yu et al., 2007). The iPS cells are similar to ES cells in morphology, biochemistry, gene expression and the ability to differentiate into many cell types and self renew (Wernig et al., 2007; Lowry et al., 2008). Furthermore, subsequent studies have optimized existing procedures and discovered novel reprogramming protocols to generate the iPS cells, including non-integrating viruses (Stadtfeld et al., 2008), non-viral vectors (Okita et al., 2008), non-integration episomal vectors (Yu et al., 2009) and RNAinduced reprogramming (Warren et al., 2010), which greatly decrease the biosafety concerns associated with the application of iPS cells. The use of iPS cells can benefit animal transgenesis in several aspects. (i) Stable genetic modification in iPS cells may be more efficient compared to somatic cells as is the case with ES cells. (ii) Genetically modified iPS cells can be used to produce chimeric animals since it has been reported that iPS cells produce viable, live-born and fertile mice offspring through tetraploid complementation (Zhao et al., 2009b; Boland et al., 2009). (iii) The use of genetically modified iPS cells as donors may increase the efficiency of cell-based transgenesis via SCNT owing to the pluripotent status of iPS cells. (iv) In contrast to the limited lifespan of somatic cells, true iPS cells are immortalized. These advantages, coupled to successful derivation of iPS cells from domestic animals (Esteban et al., 2009; Ezashi et al., 2009; Wu et al., 2009) presents a new opportunity to produce transgenic animals using iPS cells.
