Possible Production of Genome-Edited Animals Using Gene-Engineered Sperm

CRISPR/Cas9 is widely used for genome editing in a variety of organisms, including mammals, fishes, and plants. In mammals, zygotes are considered an appropriate target for gene delivery of CRISPR/Cas9 components [Cas9 endonuclease and a single-guide (sgRNA)] via microinjection or in vitro electroporation. However, these approaches require ex vivo handling of zygotes, which is necessary for egg transfer to recipient females to allow the treated zygotes to develop full-term. These procedures are often laborious, time-consuming, and use numerous mice. In our previous experiments, the plasmid DNA encapsulated by liposomal reagent introduced into the internal portion of a testis can be transferred to the mature sperm present in the epididymal ducts, and is finally transferred to oocytes via fertilization. Although it was not integrated into their genome, this approach would be useful for creating genome-edited animals, since CRISPR/Cas9 can be performed by transient interaction of Cas9 and sgRNA, whereby chromosomal integration of the CRISPR components is not a prerequisite. Here, we will review past achievements concerning in vivo transfection of immature/mature sperm and present experimental proposals for possible genome editing via gene-engineered sperm based on recent findings.


Introduction
Transgenesis is a method to induce genetic change in an organism by delivering exogenous DNA (also called transgenes) to early embryos (i.e., zygotes), and is now considered an important technique to examine gene function in vivo and for creating animal models of human disease [1,2]. In 1980, Gordon et al. [3] first demonstrated that microinjection of purified DNA fragments into the pronuclei of zygotes led to the production of mice carrying the transgenes, which are generally referred to as transgenic (Tg) or genetically modified (GM) mice. When the injected transgenes are successfully integrated into the host chromosomes of the zygotes, they are transmitted to the next generation through mating with normal mice in a Mendelian ratio, and gene expression derived from the integrated transgenes will occur in the Tg offspring depending on the property of the promoter used. For the production of Tg animals through zygote microinjection, several steps are required for the "ex vivo handling of embryos," including: collection of zygotes, DNA microinjection using an expensive micromanipulator, temporal incubation of the injected zygotes, and egg transfer (ET) to the oviducts of the pseudo-pregnant females to allow full-term development of the injected eggs [4,5]. Furthermore, all of this requires highly specialized and skilled personnel for the preparation of pseudo-pregnant females and vasectomized males, which is time-consuming and tedious, and requires a large number of mice.
Since 1980, several methods for bypassing microinjection-based transgenesis have been provided, which include infection of zygotes with viral vectors like a retrovirus [6,7], embryonic stem (ES) cell-based gene transfer [8][9][10], transgenesis via somatic cell nuclear transfer [11][12][13], and intracytoplasmic sperm injection (ICSI) using sperm associated with the transgene (TransICSI) [14,15]. All of these methods deal with zygotes and require ex vivo handling of embryos, although a micromanipulator system is not used in almost the cases.
Genome-editing via Oviductal Nucleic Acids Delivery (GONAD) is a recently developed method for creating GM mice and rats without ex vivo handling of embryos [16][17][18][19][20][21]. It can be simply performed by injecting a solution containing nucleic acids into the oviductal lumen of pregnant females at zygote to two-cell stages and subsequent in vivo electroporation (EP) to enhance DNA uptake by early embryos in situ. In this case, there is no need for the large number of mice that is normally required for the traditional microinjection-based transgenesis: only four to five pregnant females are required for modifying an endogenous gene [18]. Thus, creation of GM animals is simplified with the development of GONAD, but an expensive apparatus electroporator is still required for the technique.
In 1989, Lavitrano et al. [22] reported the simplest, convenient, and cost-effective method to create GM animals, which was called sperm-mediated gene transfer (SMGT), where isolated sperm were incubated in the presence of naked plasmid DNA for a short period and these DNA-associated sperm were subjected to in vitro fertilization (IVF) with normal oocytes. The resulting progeny are later judged as those carrying the exogenous DNA in their genome. Since the report, there has been controversy over its reproducibility among researchers [23][24][25]. However, several recent improvements were made on this SMGT system by several researchers who employed reagents capable of enhancing gene delivery towards isolated sperm [26][27][28]. For example, Shen et al. [29] incubated mouse sperm in a solution containing 3% dimethyl sulfoxide (DMSO) and plasmid DNA for 10-15 min at 4°C prior to IVF. Embryos (42%; 25/60) obtained from this experiment showed bright enhanced green fluorescent protein (EGFP)-derived fluorescence. Furthermore, Kim et al. [30] reported that nanoparticles, such as magnetic nanoparticles, can be a vehicle for delivering exogenous DNA to sperm from various animals such as boar. They incubated boar sperm in the presence of 0.5% (v/v) of magnetic nanoparticles (MNPs) and plasmid DNA coding for green fluorescent protein (GFP) on the magnetic field for 90 min, and the magnetofected sperm were subjected to IVF with normal oocytes. As a result, they obtained fertilized eggs expressing GFP. Unfortunately, for further development of IVF-derived embryos, it still required ET towards recipient females, which is laborious and requires specialized skill. Notably, it is possible to perform artificial insemination (AI) using in vitro-transfected sperm, which can be simply done by injecting those sperm into the uterine horn or lumen of the oviducts of recipient females showing oocyte ovulation. This method, called "SMGT-based AI" (SMGT-AI), has already been performed by several laboratories and will be discussed in more detail in the last part of this paper.
Testis-mediated gene transfer (TMGT) is the in vivo version of SMGT, in which sperm is transfected in vivo. This technology was first developed by Sato et al. [31], who performed intratesticular injection of calcium phosphate-precipitated plasmid DNA using a glass micropipette in mice. The injected exogenous DNA is transferred to epididymal sperm or spermatogenic cells within a seminiferous tubule (ST) of the testis, and those transfected sperm will transmit it to an oocyte through fertilization (Figure 1). They could detect the injected DNA in isolates of sperm from the epididymis and from the uteri of females mated with the injected males, but the DNA could not be detected in embryos [31]. In contrast with SMGT, TMGT does not require ex vivo handling of embryos such as collection of oocytes, IVF and ET. In this context, TMGT appears to be more convenient and simpler than SMGT for the purpose of GM animal production. Since the report of Sato et al. [31], several in vivo gene delivery approaches targeted to male reproductive systems have been reported: gene delivery to spermatogenic cells by intratubular injection of STs (ST-mediated gene transfer, STGT) (Figure 2a), to epididymal sperm present in the epididymal ducts (epididymis-mediated gene transfer, EpiGT) (Figure 2b), and to sperm present in the vas deferens (vas deferens-mediated gene transfer, VDGT) (Figure 2c).
Based on this background, the previous terminology TMGT appears to be now recognized as "direct in vivo gene delivery approach towards male reproductive system." In this context, it may be better to re-name TMGT as "intratesticular injection-based gene transfer" (IIGT), which involves direct injection of genetic materials into the interstitial space of a testis. Thus, IIGT, STGT, EpiGT and VDGT can be considered as TMGT-related experiments. In Table 1, a summary of previous studies on TMGT-related experiments is listed. Furthermore, there are several IIGT in mice. To perform IIGT, at least three different ways to inject a DNA-containing solution into the testis have been employed. The first way (shown in a) is to perform IIGT towards a testis (that is exposed outside after surgery) under anesthesia [31,32]. The second way (shown in b) is to perform IIGT through insertion of a needle via scrotum under anesthesia [33]. The third way (shown in c) is to insert a needle at three times to different sites [34]. In these latter two cases, no surgery is required. Three to five days after IIGT, the IIGTtreated males are subjected to mating with normal estrous females. The in vivo transfected sperm may fertilize ovulated oocytes, leading to creation of offspring carrying the introduced exogenous DNA. excellent papers reviewing the SMGT/TMGT-related studies [28,[90][91][92][93], which provide a helpful survey of this field. In the following sections, the TMGT-related experiments will be mentioned in more detail.
Through these works, the following information became available: i. Transgene expression in a testis after IIGT: IIGT using an adenoviral vector resulted in preferential transgene expression in Leydig cells [46]. Mogas et al. [43] reported that transgene-derived protein is mainly present in the interstitial tissue of a testis, and in STs 7-16 days after inoculation. Li et al. [51] demonstrated that plasmid DNA mixed with PEI can be transferred within STs after IIGT and expressed in primary spermatocytes, but not Sertoli cells.
ii. Detection of transgenes and expressed products on the epididymal or ejaculated sperm: According to Yonezawa et al. [40], foreign DNA encapsulated by DMRIE-C and SuperFect can be detected by PCR in the cauda epididymisderived sperm isolated 1, 4, and 14 days after IIGT. The exogenous DNA was also detectable in the ejaculated sperm collected from the uterine horn of the females in the morning after mating with the IIGT-treated males [31,38].
iii. Transgene expression in the F0 offspring obtained after mating with the IIGT-treated males: PCR analysis revealed that detection of the transgenes in the mid-gestational fetuses was found at frequencies ranging about 50-85% [39]. However, the copy number of the transgenes in those fetal DNA are estimated to be <1 copy per diploid cell [32]. PCR/slot blot analyses revealed that 41% of F0 offspring had the transgenes [53]. Ogawa et al. [34] reported that 80% of blastocysts exhibited the transgene-derived lacZ gene coding for β-galactosidase, one of the key enzymes consisting of lactose operon. Sato et al. [39] reported that lacZ expression was evident in most (72%) of the blastocysts, but neither expression of the LacZ gene nor its mRNA was found in the mid-gestational fetuses. More than 80% of morulae expressed EGFP, but the ratio of animals carrying the foreign DNA decreased as they developed, and only some of the progeny were foreign DNA-positive with high incidence of mosaicism [40].
iv. Transgene transmission to the next generation: Gene transmission to F1 offspring was at frequencies ranging about 16-23%, but when F2 offspring was obtained from mating normal female with F1 male offspring, about 94% of F2 offspring had the transgenes, suggesting that the introduced DNA was chimeric in these F0 mouse testes [38]. On the contrary, He et al. [53] demonstrated that the transgene transmission rate from F0 to F1 generation was 37%, suggesting that the transgene transmission rate was similar to the Mendelian law of inheritance. In 2007 and onward, attempts to improve IIGT systems were made by several laboratories to enhance the gene delivery efficiency [54][55][56][57][58][59][60][61][62][63][64][65][66][67]. In the following sections, we will describe several examples [(v) to (x)] about the improvement of IIGT, in vitro assessment for gene expression after IIGT or possible mechanism underlying IIGT.
v. IIGT at a young stage: In mammalian testis, spermatogenesis occurs in the STs of a testis (Figure 3d). In the ST, there are spermatogenic cells called spermatogonia (spermatogonial cells) that can be further matured into spermatocytes and spermatids. Spermatogonia are largely divided into two types, type A and type B cells. The former undergoes active mitosis and divide to produce type B cells. The type B cells divide to give rise to spermatocytes and spermatids, which move towards the lumen of the ST as they mature. According to Hui-ming et al. [59], type A spermatogonia first appear between 3 and 7 days postnatally in mice and are the only immortalized diploid cells. They considered that if these type A spermatogonia are stably transfected with the exogenous DNA, the transfected cells would be able to produce mature Tg sperm leading to production of Tg mice through natural fertilization. Based on this hypothesis, they performed IIGT using GFP-expressing plasmid DNA encapsulated by the ExGen500 transfection reagent on 7-day old male ICR mice. When these treated mice reached different stages of sexual maturity (6, 12, and 24 weeks of age), they were mated with normal females. The resulting pups were identified as Tg, with efficiencies of 11-14%. They observed GFP expression in sperm cells isolated from F0 and F1 pups. They referred to this technology as "type A spermatogonia-mediated gene transfer" (TASMGT).
vi. EP-based IIGT: Majumdar's group developed a method for generating Tg mice by directly injecting the desired gene into the testis followed by in vivo EP [54]. In this method, one of the testis was surgically exposed for DNA injection (Figure 1a) and the other contra-lateral testis was removed. An improved version of this EP-based IIGT was provided by Usmani et al. [60], who introduced the transgenes to both testes of mice directly from the outside (Figure 3a), prior to in vivo EP. They employed a two-step EP in which four 60 V electric pulses (50 ms each with an inter-pulse interval of 1 s) in one direction (forward direction) (Figure 3b) and four more pulses after changing the sides of the electrodes (reverse direction) (Figure 3c). The EP-based IIGT-treated testis expressed EGFP from the introduced transgenes even at 80 days of age. Furthermore, fluorescent germ cells were discernible in the STs of those mice (Figure 3d). F1 offspring that were generated after mating the IIGT-EP-treated males with normal females showed tissue-specific expression of transgenes. This improved procedure is based on non-surgical gene delivery using a two-step EP, which appears to be a user-friendly technique for a person who is less experienced in performing surgery.
vii. In vitro EP-based IIGT: It is difficult to apply IIGT in the testis of larger animals such as goats, because the size of the testis is much bigger than that of smaller animals such as mice and rats. Raina et al. [63] hypothesized that IIGT may be possible when in vitro gene delivery is performed towards the testis dissected from goats. They slowly injected GFP-expressing plasmid DNA into the interstitial space at eight different sites of the isolated testis using a 1-mL syringe (Figure 4a). Then, the testis was subjected to in vitro EP using a pair of tweezer-type electrodes (Figure 4b). After that, ST was partially dissociated and placed under in vitro cultivation for checking GFP expression at regular intervals (Figure 4c, d). A strong green fluorescence signal was observed 24 h after EP and its expression was continuously observed for as long as 23 days post-EP. The authors mention that the results of this study cannot be applied straightforwardly for in vivo studies, but the in vitro transfection of ST using EP will provide valuable baseline information, prior to IIGT in vivo.
viii. IIGT using a simple hypotonic solution: Majumdar's group have developed a method to generate Tg mice by directly injecting the desired gene in the testis followed by in vivo EP, as mentioned above [54,60]. This technique is less complicated for small animal like mice but appears to not be feasible for transgenesis in large animals, because these animals have larger testes and greater scrotal thickness, to which it is difficult to standardize voltage parameters. To overcome this difficulty, they developed an alternative technique for making Tg mice by hypotonic shocking male germ cells for gene delivery. According to Usmani et al. [65], treatment with hypotonic Tris-HCl solution reduced osmolarity and led to hypotonic-swelling of germ cells. The hypotonic-swelling eventually killed the cells with increased hypotonicity, but led to the uptake of surrounding molecules such as nucleosides inside the cell. The authors hypothesized that a hypotonic Tris-HCl solution at a certain hypotonic concentration might allow the germ cells to internalize the surrounding solutes like DNA in vivo without being killed and the sperm produced from transfected germ cells may carry a desired DNA fragment (transgene) to generate Tg animals. Usmani et al. [65] suspended the linearized plasmid DNA (transgenes) in hypotonic Tris-HCl solution (pH 7.0) and simply performed IIGT to internalize the injected transgenes in the genome of spermatogonia residing at basal compartment of tubules. As a result, such males successfully generated Tg progeny by natural mating. This technique is easy and simple and does not require expensive apparatuses like electroporators. Usmani et al. [65] proposed that such a procedure enables researchers to generate their

In vitro EP-based IIGT in goats. The isolated goat testis is subjected to IIGT with a total of eight repeated injections from different directions (a). Then, the entire testis is hold by a pair of electrodes prior to EP (b). After that, the EP-treated STs are partially dissected and subjected to in vitro cultivation (c). By this, the transgene expression on the spermatogenic cells of STs can be monitored continuously (d).
Possible Production of Genome-Edited Animals Using Gene-Engineered own Tg animals, instead of outsourcing, and would drastically minimize the time required for studies on functional genomics.
ix. IIGT using nanoparticles: It has previously been reported that nanoparticles such as magnetic nanoparticles, mesoporous silica nanoparticles, and halloysite clay nanoparticle can be used as a vehicle for delivering exogenous DNA to sperm of various animals, such as boar and bovine [30,[94][95][96]. However, these early reports are confined to in vitro events. Wang et al. [97] applied this technology to IIGT using bacterial magnetic particles (BMPs)/ PEI complex-conjugated foreign DNAs (BMP-PEI-DNA complex). According to Wang et al. [97], BMPs help to reduce the toxicity of the PEI, an efficient gene transfer agent, and assist gene delivery in vivo. After performing IIGT in mice, the testis was returned to the original position, and a neodymium-ironboron magnet with an intensity of 600 milliteslas (mT) was placed onto the surface of the abdomen to facilitate gene delivery towards the spermatogenic cells. The authors report that this procedure is not harmful to the functioning of the testis. They reported that the clusters of BMP-PEI-DNA complex successfully reached the cytoplasm and the nucleus of spermatogenic cell and expressed in the testes of F0 mice. The resulting F0 mice could transmit the introduced transgene to the offspring with efficiencies of 88%.
x. Possible mechanism for IIGT-mediated gene delivery to sperm: When intratesticular injection of a DNA-containing solution into the interstitial space of a testis is performed, the fate of the injected solution is largely divided into two routes. This was first assessed by Sato et al. [42] who employed trypan blue (TB) as dye to visualize the transferring solution. One route is when a solution is transferred to the epididymal ducts, and the other route is when a solution is transferred to the lumen of the STs (Figure 5).
After the intratesticular injection of 30 μL TB, the dye was rapidly transferred via rete testis to the proximal segment of caput epididymis [42]. One day after the injection, the dye was observed in the middle segment of caput epididymis. This was also confirmed by a previous observation using PCR analysis, which revealed the presence of the exogenous DNA in the spermatozoa isolated from caput epididymis [31]. Furthermore, the exogenous DNA can be detected in the ejaculated sperm collected from the uterine horn 1 day after mating with females [31]. Thus, it is reasonable to consider that the exogenous DNA injected into the interstitial space of a testis is rapidly transferred to the epididymal ducts and taken up by epididymal sperm in situ.
Only a minor portion of the solution injected into the interstitial space of a testis is transferred inside the STs. This may be elicited by mechanical shearing of STs upon insertion of a needle or glass capillary. To increase the transfection efficiency in the spermatogenic cells existing within STs, repeated needle insertions (over at least three times from different sites) have been employed by some research groups [34,46,53,59,63]. At present, it remains unknown how many STs are indeed transfected by this treatment. Usmani et al. [65] reported that spermatogenic cells, including spermatogonia present within STs, are transfected after IIGT and subsequent in vivo EP using fluorescent marker genes.

STGT-related experiments
ST is a tubular structure packed in a testis, which contains spermatogenic cells, such as spermatogonia, a precursor for mature sperm, and maturing sperm cells (spermatocytes and spermatids), and Sertoli cells, which support the proliferation and differentiation of spermatogonia (Figure 3d). If the exogenous DNA is introduced within a lumen of the ST, it is possible to transfect those spermatogenic cells, from which transfected spermatozoa are transported via epididymal ducts and finally ejaculated upon mating with estrous females. As a result, the introduced exogenous DNA will be transmitted to oocytes via fertilization with the transfected sperm.
To our knowledge, STGT was first performed in 1997 by two groups: Blanchard and Boekelheide [35], with rats using adenoviral vector, and Kim et al. [68], with mouse and pigs using liposomally encapsulated plasmid DNA. The aim of the former group was to study transgene expression in the adult rat testis in vivo. They demonstrated that there was transgene expression in Sertoli cells and principal cells of the epididymis, and expression persisted for at least 10 days. The latter group demonstrated that in mice, 8.0-14.8% of STs expressed the introduced transgene, as evaluated by histochemical staining for the lacZ gene, and 7-13% of epididymal sperm had the exogenous DNA, as evaluated by PCR analysis. In pigs, 15-25% of the STs contained lacZ-positive germ cells. They suggested that STGT can be used as a powerful tool for producing Tg livestock.
In 1998, Yamazaki et al. [69] employed STGT for examining transcriptional regulatory elements of spermatogenic specific genes. After injecting DNA into STs, subsequent in vivo EP was used to enhance DNA uptake by spermatogenic cells. Based on these experiments, they suggested that spermatogenic stem cells and/or spermatogonia can incorporate foreign DNA, and that the transgene could be transmitted to the progenitor cells derived from a transfected proliferating germ cell. Later, the same group [70] examined the possibility to create Tg offspring using STGT coupled with in vivo EP (EP-based STGT). Although longlasting transgene expression could be detected in the spermatogenic cells even 2 months after EP, no Tg offspring were obtained after natural mating with normal adult females.
Huang et al. [71] used the EP-based STGT towards an entire mouse testis after intratubular injection of plasmid DNA that coded for fluorescent genes. To trace the fate of transfected spermatogenic cells, they obtained fluorescent sperm by fluorescence activated cell sorting (FACS), and performed ICSI to obtain their offspring. Almost all the individuals produced from fluorescent sperm were Tg. The authors

Possible mechanism of how the exogenous DNA introduced into the interstitial space of a testis is transmitted to epididymal sperm or to spermatogenic cells within a ST.
Possible Production of Genome-Edited Animals Using Gene-Engineered Sperm DOI: http://dx.doi.org /10.5772/intechopen.84859 claim that this is the first report of gene transfer into germ cells and subsequent production of Tg offspring.
STGT appears to be the direct approach to transfect spermatogenic cells in situ, but most experiments [46,49,51,[72][73][74][75][76][77][78][79][80][81][82][83][84] have been confined to successful transfection of spermatogenic cells and Sertoli cells for rescuing damaged/inactive Sertoli cells, in vivo testing of efficiency of RNA interference (RNAi), or establishment of an in vivo assay system to evaluate the promoter activity of the gene of interest. There have been no trials to create Tg offspring through STGT. Only some groups have tried to test the possibility of creating Tg animals by STGT. For example, Celebi et al. [74] performed STGT using circular plasmid carrying the lacZ reporter gene mixed with noncommercial cationic lipids. These injected males were mated with wild-type females and the progeny were analyzed by PCR and Southern blot assay. They demonstrated that the transgenes were transmitted to the offspring, but remained episomal, since it was found in the tail of the young animals and was lost at adulthood. Therefore, the plasmid seemed to be lost during the numerous germ cell divisions. This plasmid stayed in some tissues, such as in the skeletal and cardiac muscles. No integrative forms have yet been found with the use of circular DNA. Kanatsu-Shinohara et al. [75] described a novel approach for producing Tg animals by transducing spermatogonial stem cells in vivo using a retroviral vector by STGT. When these injected males were mated with wild-type females, Tg offspring were obtained with an efficiency of 2.8%. The transgene was transmitted stably and expressed in the next generation. The authors, thus, concluded that this technique will be useful as an alternative to the pre-exiting microinjection-based transgenesis, as well as provide a means for analyzing the self-renewal and differentiation processes of spermatogonial stem cells in vivo. Sehgal et al. [83] described a technique for the generation of Tg mice by infection of spermatogonial stem cells with recombinant lentiviruses expressing EGFP with a high rate of success. When the infected males were mated to normal females, over 60% of the delivered pups were found to be Tg. Li et al. [84] employed methods similar to those of Kanatsu-Shinohara et al. [75] and Sehgal et al. [83] and reported that the Tg efficiency is around 2.4%, which is similar to the previous report of Kanatsu-Shinohara et al. [75].

EpiGT-related experiments
Epididymal sperm present on the ducts of caput and cauda epididymides and epididymal epithelial cells can be targeted for transfection by the exogenous DNA. Kirby et al. [85] performed intraluminal injections (2-5 μL) of plasmid DNA into the lumen of an initial segment tubule of caput epididymis (Figure 2b), and subsequent in vivo EP towards the injected portion to examine the function of epididymal epithelial cells, which are thought to play critical role in sperm maturation during transport through epididymides. They concluded that this procedure is useful for elucidating the activity of promoter elements included in the injected plasmid that may not be identified when traditional in vitro methods are used. Esponda and Carballada [86] injected plasmid DNA mixed with the lipid FuGENE6 into the lumen of mouse cauda epididymis (Figure 2c). Successful transfection was observed in about 40% of cells after 2 days and in about 32% after 7 days, and then diminished progressively over time. Gene expression continued up to 15 days after gene injection and occupied about 22% of the area of the tubules. They concluded that intraluminal injections of exogenous DNA are effective for the study of epididymal physiology or to change the fertilizing ability of sperm. These studies are not aimed to create Tg animals, but they hold a potential to transfect epididymal sperm as a useful Tg tool.

VDGT-related experiments
In 1998, the Esponda's group [87] first attempted to examine whether exogenous plasmid DNA introduced into the lumen of the proximal region of the vas deferens could be taken up by mouse and rat sperm (Figure 2d). They demonstrated that 60-70% of sperm recovered 6 h after DNA injection had positive signal for successful transfection in their sperm nucleus, which was not affected by DNase treatment. This was also confirmed by PCR and slot blot analyses. They concluded that sperm within the vas deferens had the ability to incorporate exogenous DNA, which can be transferred to their nuclei, and vas deferens secretions do not block these capacities. In 2000, the same group [88] showed that this VDGT is useful for production of Tg mice. They injected plasmid DNA encoding GFP into the lumen of mouse vas deferens. The night after injections, males were mated with normal estrous females, and the offspring were analyzed. About 8% (4/53) of the newborns delivered expressed the GFP gene. They concluded that VDGT is a simple alternative to the pre-existing microinjection-based production of Tg animals and can be used for species in which the microinjection procedure is not feasible. This technology was later found to also be useful for transfection of epithelial cells of the vas deferens using a direct injection of DNA-liposome complexes, which could modify vas fluid contents [89].

Historical background of SMGT-based AI
As mentioned previously, AI of transfected sperm with exogenous DNA through SMGT is a highly convenient route for producing Tg animals. To our knowledge, Sperandio et al. [98] was the first to demonstrate its usefulness in domestic animals, such as bovine and swine. They performed AI towards ten sows with boar sperm cells that had been preincubated with plasmid DNA and obtained 82 offspring. Southern blot analysis of the DNA extracted from the animal tails showed that five animals were Tg and contained sequences complementary to the exogenous plasmid DNA that appeared to be rearranged compared to the original plasmid. From this study, it was suggested that SMGT-AI can be successfully adapted for the generation of Tg livestock. Yonezawa et al. [99] tested whether liposome-peptide (derived from human protamine)-DNA complex (LPD), a new reagent known to stabilize transfection in cultured cells, was useful to increase the efficiency of SMGT. They performed AI using rat epididymal sperm that had been incubated in a solution containing GFP expressing plasmid DNA and LPD complex. Expression of GFP was detectable in the morulae isolated from the treated animals. Furthermore, the AI-treated animals produced pups carrying foreign DNA.
This SMGT-AI is applicable to avian species. Yang et al. [100] performed AI using freshly-ejaculated chicken sperm that had been incubated in the presence of plasmid DNA and liposome, and found that about 4% (2/53) newly hatched chicks was identified as Tg. Harel-Markowits et al. [101] employed restriction enzymemediated insertion (REMI) to increase the efficacy of the transfection towards the isolated chicken sperm. REMI was used to insert exogenous DNA linearized with a restriction enzyme that cuts the genomic DNA at sites that enable the exogenous DNA to integrate via its matching cohesive ends [102,103]. Following insemination with sperm transfected with linearized DNA, restriction enzyme, and liposome, they obtained Tg offspring. Furthermore, when chicken sperm are incubated in a solution containing plasmid DNA and DMSO or N,N-dimethylacetamide (DMAc) and subsequently subjected to AI, the resultant newborn chicks have the transgene, with efficiencies of 38% (for the DMSO-treated group) and 19% (for the Possible Production of Genome-Edited Animals Using Gene-Engineered Sperm DOI: http://dx.doi.org /10.5772/intechopen.84859 DMAc-treated group) [104]. However, Chaparian et al. [105] recently reported that they were unable to create Tg chicks by SMGT-AI.

Exosomes as a possible carrier to deliver genetic materials to sperm
Exosomes, membrane-enclosed sub-cellular microvesicles shed from most cell types, are present in a wide variety of body fluids [106]. Recently, it was found that they can mediate various effects on the behavior of recipient cells, since they contain cytokines, growth factors, and membrane proteins [107]. Furthermore, they contain a substantial amount of small and functional RNA molecules, called microRNAs (<100 nucleotides in length) [108], which could potentially control gene expression of various endogenous genes. It has recently been shown that these exosomes are (1) found in human semen [109], (2) involved in sperm maturation process during the transit along the male epididymal tracts [110], (3) accumulated in mature spermatozoa nuclei [111], and (4) delivered to oocytes through fertilization [112].
Notably, there are some reports describing non-Mendelian germline-independent inheritance of phenotypes in the absence of any classically identifiable mutation or predisposing genetic lesion in the genome of individuals who develop the disease [113][114][115]. For example, Cossetti et al. [116] performed subcutaneous inoculation of EGFP-expressing human melanoma cells into an immunocompromised mouse, from which EGFP RNA was released from the grafted melanoma cells, delivered to the bloodstream, and finally brought to sperm. When epididymal sperm isolated from these tumor-bearing males were examined carefully, the EGFP RNA was found to be tightly associated with the extracellular fraction of these mature sperm. They termed this phenomenon "soma-to-germ line transmission of information," and thought that exosomes may be involved in this phenomenon as the carrier to deliver EGFP RNA. The findings of Cossetti et al. [116] appear to be well correlated with those obtained from the TMGT-related experiments done at earlier stages of IIGT development, which include (1) non-Mendelian transmission of the exogenous DNA in the offspring obtained [33], (2) extreme low copy number of the exogenous DNA (<1 copy per diploid cell) transmitted to these offspring [32], (3) mosaic expression of the exogenous DNA in the offspring (blastocysts) obtained [33,40], and (4) reduction in the number of offspring carrying the exogenous DNA during development [40]. As mentioned in Section 2.1 (x), parts of a solution introduced into the interstitial space of a testis is transferred to the excurrent ducts of epididymides, and the exogenous DNA may be taken up by the extracellular fraction of epididymal sperm, possibly through exosomes. We detected the presence of exogenous DNA in the DNase I-treated epididymal sperm, which have been isolated from the IIGT-treated males [42]. This may be due to the fact that exosomes can protect its exogenous DNA against DNase I-mediated digestion.
Notably, in their review article, Jiang and Gao [117] demonstrated that exosomes can be used as naturally occurring cell-to-cell transporters or as novel biocarriers for gene and drug delivery. These exosomes are naturally secreted by the cells and pass through additional biological barriers. They are more biocompatible and biodegradable and can avoid immune response which is most likely due to the surface expression of the complement regulatory proteins, such as CD55 and CD59. For these natural characteristics, exosomes are being extensively explored as gene delivery vehicles. For example, in 2011, Alvarez-Erviti et al. [118] first demonstrated that exosomes are useful for delivering short interfering (si)RNA to the mouse brain. They engineered dendritic cells to express lysosome-associated membrane protein 2 (Lamp2) isoform (Lamp2b), an exosomal membrane protein fused to a neuron-specific peptide. Following that, exosomes were isolated from the gene-engineered dendritic cells and loaded with siRNA using electroporation and were administered intravenously to mice. As a result, the targeting peptide was shown to be successfully delivered to the brain. The concomitantly delivered siRNA caused reduced expression of a target protein associated with the pathogenesis of Alzheimer's disease. Furthermore, Lin et al. [119] proposed that exosomes can be a good carrier to introduce various cargoes, including plasmid DNA, into a cell. They prepared a mixture composed of purified exosomes isolated from HEK293FT cell line, pEGFP-C1 plasmid DNA, and Lipofectamine 2000 liposomes in vitro. During the incubation at 37°C for 12 h, exomes and liposomes are fused together and the exogenous plasmid DNA becomes incorporated into exosome-liposome hybrid nanoparticles. Transfecting mesenchymal stem cells (MSCs), which cannot be transfected by the liposome alone, with this complex resulted in successful generation of fluorescent cells when evaluated by FACS. Now, an exosome-based transfection kit, possibly based on this principle, is commercially available: Exo-Fect Exosome Transfection Kit (System Biosciences). We confirmed the usefulness of this kit by in vivo transfecting oviductal epithelial cells through intraoviductal instillation of a solution prepared using this kit. Some oviductal epithelial cells were found to be fluorescent after transfection with a plasmid expressing GFP (unpublished results). Thus, it may be possible to transfect isolated sperm by incubating plasmid DNA and exosome/liposome hybrid vesicles provided from the Exo-Fect Exosome Transfection Kit, prior to AI as mentioned below.

Genome-editing sperm
Gene modification based on recently developed techniques such as zinc-finger nucleases (ZFNs), TAL effector nucleases (TALENs), and clustered regularly interspersed short palindromic repeats/CRISPR-associated protein 9 (CRISPR/ Cas9) are now recognized as a revolutionary genetic engineering tool in vitro and in vivo [120][121][122][123][124]. Three types of endonucleases from ZFNs, TALENs, and CRISPR/ Cas9 have been developed to promote precise genome editing at a target gene. All these enzymes have a DNA-binding ability and an ability to elicit double-strand DNA break (DSB) at a target genomic locus. Subsequently, in the absence of a homologous template to repair, nonhomologous end joining (NHEJ) occurs and causes small insertions or deletions (termed "indels"). In the presence of a template donor DNA, site-specific recombination through homology-directed repair (HDR) occurs. Generally, the frequency of NHEJ is thought to be higher than that of HDR in most of the cell types [125]. Table 2 shows comparison among these three technologies.
ZFNs are the first engineered endonucleases [127] that combine the DNA recognition ability of zinc-finger protein (called zinc-finger motifs) and restriction enzyme Fok I to introduce DSB [120][121][122][123][124]. In 2005, Urnov et al. [128] first demonstrated that ZFNs are effective as a genome editing system in the human cells.
TALENs are similar to ZFNs and require a string of TALEN motif (consisting of a series of 33-35 amino acid repeats) to bind to the specific sequence of a target gene and Fok I enzyme to introduce DSB [120][121][122][123][124]. TALENs provide more flexibility to the target sequences since ZFNs are known to be more active towards GC-rich region, whereas TALENs can be assembled to the target AT-rich regions [120][121][122][123][124].
Since both ZFNs and TALENs require assembling of an array to build each set, which is a complex and time-consuming process [124], CRISPR/Cas9 has become the favorite because of its easy application. CRISPR/Cas9 requires only two components, Possible Production of Genome-Edited Animals Using Gene-Engineered Sperm DOI: http://dx.doi.org /10.5772/intechopen.84859 the Cas9 nuclease and a single-guide RNA (sgRNA), which is a short sequence to guide the Cas9 protein to a target site. More importantly, these events are performed by transient interaction of Cas9 and sgRNA, whereby chromosomal integration of the CRISPR components is not a prerequisite [129]. There is a concern of off-target cleavage activity from the endonuclease from CRISPR/Cas9 because the system requires recognition of only 20 bp target sequence and allows up to 5 bp mismatches for the formation of DSB [124]. Several strategies for minimizing the off-target cleavage have been employed including use of double nickase mutant form of Cas9, which induces a single-strand break instead of DSB [130]; use of Cas9-sgRNA ribonucleoprotein (RNP) complex, whose half-life is shorter than that the time in which plasmid or viral nucleic acid is transcribed [131]; or use of fusions of catalytically inactive Cas9 with Fok I nuclease domain (fCas9) to improve the DNA cleavage specificity [132].
In the case of producing GM animals using SMGT or TMGT, it is better for the exogenous DNA (transgenes) to be integrated into the chromosomes of sperm. This event appears to occur more frequently in immature sperm cells present in the ST of a testis than in the mature epididymal sperm because the chromosomal DNA in the latter cells is tightly packed in the head region of a sperm. In this context, STGT is a preferable system to create GM animals because it is targeted to transfection of spermatogenic cells present within the STs. However, it takes about 4 weeks for mature sperm to reach the epididymal portion for fertilization. If a researcher wants to generate GM animals within a short period of time, direct transfection of mature sperm present in epididymides or vas deferens is recommended. In this case, as mentioned above, the introduced exogenous DNA may be associated to the extracellular fraction of a sperm, as episomal DNA. Notably, CRISPR/Cas9based genome editing does not always require chromosomal integration of its components; it can be performed by transient expression of their components after transfection [129]. In this sense, an attempt to transfect mature sperm would be a useful alternative for GM animal production.

Proposal of new experimental systems for simple creation of genome-edited animals using in vivo or in vitro transfected sperm
In the following section, we propose two experimental plans to create genomeedited animals using VDGT or AI-based systems, all of which are simpler and more convenient than the previously described systems.

VDGT-based genome editing
As previously described in Section 2.4, VDGT enables transfer of exogenous DNA to oocytes via fertilization by mature sperm transfected within vas deferens [87,88]. Injecting a solution containing genome editing components (e.g., sgRNA + DNA/ mRNA/protein for Cas9) into the lumen of vas deferens of anesthetized males and subsequent mating between the VDGT-treated males and normal estrous females the day (night) after the surgery may result in production of genome-edited offspring.
In Figure 6a-c, experiments obtained after TB injection into the lumen of mouse vas deferens is shown (unpublished results). Under anesthesia, cauda epididymis and vas deferens were pulled out and a small slit was made at the proximal region of vas deferens using micro scissors (Figure 6a). Then, a glass micropipette containing TB was inserted into the lumen of vas deferens under observation using a dissecting microscope and about 15 μL of the solution is slowly injected (Figure 6b). It is easily discernible that the injected TB still remains within the proximal portion of vas deferens immediately after the injection (arrows in Figure 6c). However, the injected TB moved to the distal portion of vas deferens the next day (arrow in Figure 6d). Thus, to produce genome-edited animals by VDGT, a solution containing CRISPR/Cas9 components (sgRNA + DNA/mRNA/protein for Cas9), gene delivery enhancing reagents (such as DMSO, liposomes, microparticles, etc.) and fluorescent marker expression plasmid DNA has to be prepared first (Figure 6e).  After short incubation period, this solution is injected into the lumen of vas deferens. Then, the VDGT-treated males are mated to normal estrous females on the day (night) or next day. Later, cleavage stage embryos were collected from the VDGTtreated females to examine the presence and expression of the transgene (plasmid) (Figure 6e, bottom) and occurrence of mutations in a target locus. In some cases, the SMGT-AI-treated females were allowed to deliver their pups to see whether genome editing is induced in their chromosomes.

SMGT-AI-based genome editing
AI is one of the assisted reproduction technologies that is based on the introduction of isolated sperm into the female reproductive tracts, such as uterine horn or oviductal lumen, to in vivo fertilize ovulated oocytes. As previously described in Section 3, isolated sperm are incubated in a solution containing exogenous DNA and gene delivery enhancing reagents such as DMSO, liposomes, and microparticles, for a short period (SMGT), and then the transfected sperm are subjected to AI, called "SMGT-based AI" (SMGT-AI). During this process, the exogenous DNA should be transmitted to oocytes via fertilization resulting in Tg embryos.
We previously reported that transfer of sperm into a space near the infundibulum between the ovary and ovarian bursa enables in vivo fertilization of ovulated oocytes in the ampulla region of the oviduct [133,134]. In more detail, 2 μL of fresh epididymal B6C3F1 (F1 hybrid mice between C57BL/6 and C3H) sperm (containing 2 x 10 5 spermatozoa) were intrabursally injected 7 h after human chorionic gonadotropin (hCG) administration to B6C3F1 females that had been administrated with pregnant mare serum gonadotropin (PMSG) 48 h before. At 1.7 days after AI, normal cleaving embryos were recovered at rates of 40-100%. We called this AI technology "intrabursal transfer of sperm" (ITS) [133]. In Figure 7a, the ITS procedure is schematically illustrated. In Figure 7b and c, photographs before (b) and after (c) ITS are shown by using TB as a dye to visualize the process of AI. It is clear that the injected solution is present between the ovary and ovarian bursa (arrow in Figure 7c). In Figure 7d, an example for SGMT-AI-mediated genome editing in embryos is schematically shown. First, sperm isolated from the vas deferens are treated with CRISPR/Cas9 components (sgRNA + DNA/mRNA/protein for Cas9), gene delivery enhancing reagents (such as DMSO, liposomes, microparticles, etc.), and fluorescent marker expression plasmid DNA for a short period. Then, the solution containing the transfected sperm is subjected to AI towards females 7 h after hCG administration. The next day, 2-cell embryos are collected from the AI-treated females to examine the presence and expression of the transgene (plasmid) and occurrence of mutations in a target locus. In some cases, the SMGT-AI-treated females are allowed to deliver their pups to see whether genome editing is induced in their chromosomes. Notably, the selection of a successfully genome-edited sperm prior to AI may accelerate the production efficiency of genome edited offspring, although the practical approach for this remains unknown at present. Therefore, the success or failure of genome editing performed in this system may depend on the molecular analysis of the offspring (e.g. blastocysts or fetuses) generated after AI of the SMGT-treated sperm.

Conclusion
TMGT, based on direct in vivo gene delivery towards interstitial space of a testis, ST within a testis, or excurrent ducts of epididymides and vas deferens, is less labor intensive and less time consuming for the production of GM animals. This testicular route is also ethically superior since fewer mice are required than existing alternative methods of transgenesis. The TMGT-treated males can be used to mate with estrous females, through which the exogenous genetic materials are transferred to oocytes at fertilization. During this process, there is no need for ex vivo handling of embryos, which is strictly required for zygote-based gene modification such as microinjection, EP, viral infection, and TransICSI. SMGT-AI, based on AI of sperm that have been transfected in vitro with the exogenous DNA, is also a convenient system for production of Tg animals, like TMGT. The CRISPR/Cas9 system, one of the recently developed genome editing technologies, is now recognized as a powerful and simple tool to create GM animals. More importantly, in this system, chromosomal integration of the genome editing components is not the prerequisite. Coupling this genome editing system with TMGT or SMGT-AI would accelerate creation of genome-edited animals in a more convenient manner. Furthermore, TMGT/SMGT-AI will be particularly useful for other animals that are difficult to manipulate as early embryos in vitro.

Conflicts of interest
The founding sponsors had no role in the design of the study, collection, analyses, or interpretation of data, writing of the manuscript, and decision to publish the results.