Intracytoplasmic Sperm Injection-Revolution in Human and Animal Assisted Reproduction: A Review

INTRODUCTION

During normal fertilization, the sperm nucleus is incorporated into an oocyte through membrane fusion between the two gametes. The ICSI is the mechanical insertion of a single sperm directly into the ooplasm of a matured oocyte using a microscopic needle. It is a variation of in vitro fertilization (IVF) that has been a standard clinical technique for many years (Committee, 2004). Its efficacy and short-term safety made the technique reliable to use in the clinical setting to alleviate severe Male Factor Infertility (MFI) in human. The MFI is a term used to describe an individual with one or more sperm abnormalities that may impair their fertilizing capability (Gordon and Talansky, 1986; Ng et al., 1988; Palermo et al., 1992). Problems with the production and maturation of sperm are the most common causes of MFI. Sperm may be immature, abnormally shaped, or unable to move properly. Or, normal sperm may be produced in abnormally low numbers (oligospermia) or seemingly not at all (azoospermia) due to various reasons. Therefore, the Practice Committee of the American Society for Reproductive Medicine has approved ICSI as a clinical technique rather than experimental in human medicine (Committee, 2004). This procedure can bypass many barriers of fertilization including the process of sperm penetration through cumulus-corona cells, zona pellucida (ZP) and oolemma during fertilization by directly depositing the sperm into the ooplasm. The introduction of ICSI in assisted reproduction has helped to investigate and understand the early events of fertilization such as sperm capacitation, the Acrosome Reaction (AR) and pronucleus (PN) formation (Keskintepe et al., 1997).

In natural mating, genetically important but biologically inferior (e.g., low sperm production, poor sperm motility and sperm abnormalities) males cannot be used. In IVF, severe MFI could not be treated. However, ICSI is a technique by which these situations can be overcome. Therefore, ICSI would be an effective technique to use on genetically important but biologically inferior male gametes for procreating domestic and wild livestock species (Keskintepe et al., 1997) and thereby facilitating the production of large number of embryos and offspring from a single genetically valuable animal. Development of an ideal ART like ICSI revolutionizes the management of infertile or sub-fertile males. Nowadays the ICSI became an efficient and reliable clinical technique in human. Because of its higher success rate, this technique is expanding very rapidly in clinical as well as research fields both in human and animal reproduction. In animals, especially in livestock, the ICSI technique could be an alternative approach to generate transgenic livestock, for example goats, using sperm as the DNA carrier (Perry et al., 1999, 2001), for propagation of useful genetics (Wang et al., 2003) or gender pre-selection (Hamano et al., 1999). The progeny produced by this technique could be used as founder animals for the production of recombinant protein in their milk such as pharmaceutical proteins for the treatment or prevention of human diseases or biomaterials for medical use (Niemann and Kues, 2003; Keefer, 2004). Therefore, engaging in ICSI appears to be profitable and beneficial to livestock industry. The present study will discuss the ICSI technique with special reference to history and gradual development of micromanipulation, ICSI success, remodeling of reproductive process, oocyte activation and various factors affecting the outcomes of ICSI.

HISTORY AND GRADUAL DEVELOPMENT OF MICROMANIPULATION AND ICSI

Micromanipulation is comparatively a recent term of biological micro-dissection that replaced the old term micrurgy. As cited in Timson and McDermott (1994), the term micrurgy was first coined by Peterfi (1923) for biological micro-dissection, which was also known as micrurgery. The term micromanipulation is used to include all operations performed in the microscopic field of vision with the aid of mechanical devices that guide the operating tools (Timson and McDermott, 1994). Although, the invention of microscope made it possible to perform all sorts of micro-dissection under the microscopic field, however, the invention of mechanical devices which guide the operating tools proper that heralded the advent of micromanipulators. Micromanipulation techniques are being used to obtain fertilization and pregnancies in cases, which would have been considered hopeless only one and a half decades ago. The gradual developments of micromanipulation and ICSI techniques have been reviewed elsewhere (Hiramoto, 1962; Iritani, 1991, 1994; Timson and McDermott, 1994; Goto, 1997; Payne, 1995; Mansour, 1998; Horiuchi et al., 2002; Horiuchi, 2006).

Development of micromanipulation tools and techniques: The ICSI has been the major breakthrough in the field of assisted reproduction among the all micromanipulation techniques developed in the eve of twentieth century. All the micromanipulation techniques now used for assisted reproduction basically originated from a pre-exiting technology of micromanipulation, which was developed for various purposes other than assisted fertilization (Mansour, 1998). It is to be mentioned here that the micromanipulative techniques were first conceived and applied by the microbiologist (Timson and McDermott, 1994). As cited in Timson and McDermott (1994) and Schmidt (1859) seemed to have been the first to describe the use of a microscopic dissector. However, the use of micromanipulation tools was first described by Chabry (1987) and Timson and McDermott (1994) in the study of the embryos of marine animals. Chabry (1987) developed modified equipment and was one of the earliest to comment on the properties of micromanipulated embryonic blastomeres. As mentioned by Timson and McDermott (1994), Hiramoto (1962) that Barber (1904, 1911) and G.L. Kite (never published cited in Lillie, 1934), two early microbiologists, extended the use of micromanipulation, respectively, in bacteriology and basic cell biology. Subsequent scientists broadened and fine-tuned the tool-making technology and micromanipulation-dissection developed with enormous potential and considerable accuracy (Timson and McDermott, 1994). The micromanipulative techniques have attained a high degree of perfection since that time. In 1928, Emmerson (El Badry, 1963) described the first micromanipulator with a joystick design that could directly transmit the movements required by the operator. The development of the varied types of exceedingly delicate microtools went hand in hand with the development of the micromanipulators and the significant advances in microscopy aided greatly in that direction (Timson and McDermott, 1994). As mentioned in (Mansour, 1998), hydraulic principles that are now used for the joystick manipulators had been described by De Fonbrune (1934) who also invented the first microforge and microtool-making techniques that are still in use today. The timeline of early developmental history of micromanipulation tools and techniques has been depicted in the Table 1.

History and gradual development of sperm injection experiments: Although ICSI in the field of assisted reproduction is a recent innovation, research on sperm injection into the oocyte in animals had already started a century ago. For more than half a century, sperm injection experiments had been conducted with lower animals especially with marine echinoderms. Scientists were devoted themselves to develop various techniques for direct micromanipulation of living cells from the beginning of the twentieth century. As mentioned by Hiramoto (1962), the first sperm injection experiment was attempted in 1911 by G.L. Kite (Lillie, 1914), one of the pioneers in the field of microsurgery or micrurgy.

Table 1:	Timeline of development of micromanipulation tools and techniques

He used micro-needles and microinjected starfish sperm into the starfish egg. With the aid of micro-needles G.L. Kite in his nuclear dissection experiment was able to separate the PN in newly fertilized zygotes (Chambers, 1940). However, due to interruption of his experiment, unfortunately the complete result of his work was never been published and the study became inconclusive (Lillie, 1914). In the 1940s and 1950s, micromanipulative or microsurgical procedures were first attempted in several species of mammals to study the living eggs of the rat, the human, the mouse and the rabbit although the results of these studies were not found (Timson and McDermott, 1994). In his book on fertilization Rothschild (1956) mentioned that live sperm injection into the cytoplasm of sea urchin was possible although no fertilization could be found, however, he did not cite any reference for his comment. It is to be mentioned here that T.C. Lin (Fishel and Symonds, 1993), a pioneer mammalian embryologist, performed numerous studies on the technical and experimental nature of egg micrurgery or micromanipulation during the 1950s and 1960s.

The first detailed and successful sperm injection experiment was conducted by Hiramoto (1962) who injected live sea urchin sperm into the protoplasm of an unfertilized egg of the same species that had been activated and the injected sperm nuclei participated in the meiotic process. At that time, it was not performed for the purpose of assisted fertilization, but to prove that sperm nuclear decondensation and male pronucleus (MPN) formation did not require prior interaction between the sperm and the oolemma. This work inspired the scientist to study the technique in higher species of animals. In amphibians, the technique was first documented in frogs by Graham (1966) followed by Brun (1974).

After these achievements in lower animals, sperm injection experiments were conducted in mammals. Uehara and Yanagimachi (1976) investigated the possibility of hamster and human sperm to develop into MPN when injected into hamster oocytes. It was found that isolated hamster sperm nuclei developed into MPN in hamster oocytes. The nuclei of fresh, frozen-dried and frozen-thawed human sperm were also able to develop into MPN in hamster oocytes. In another experiment, hamster, mouse, rabbit and fish sperm nuclei, or hamster hepatocyte nuclei were microinjected into hamster oocytes were able to develop into MPN for each type of nuclei and began DNA synthesis (Naish et al., 1987). These findings suggested that the ooplasmic factors controlling the transformation of sperm nuclei into MPN are not species-specific. Injection of sperm head, detached from tail, induced normal fertilization of unfertilized mouse oocyte (Markert, 1983). Injection of grossly defective and immotile mouse sperm into the oocyte exhibited the same reactions found in the fertilization with healthy sperm.

Among domestic animals, sperm injection was first attempted in bovine oocytes with MPN formation (Westhusin et al., 1984). Laws-King et al. (1987) reported the insertion of a single sperm under the ZP of human oocytes with a high rate of fertilization. However, improvements in microinjection using a rabbit model not only led to MPN formation following ICSI-induced fertilization, but also resulted in further embryonic cleavage and the first live birth of normal offspring (Hosoi et al., 1988). A historical picture of sperm injection attempts (first report only) has been illustrated in Table 2 before the first successful live birth.

Development of other micromanipulation techniques related to ICSI: In most of the early research activities on micro-fertilization, ooplasmic sperm injections or ICSI were exclusively conducted. However, ooplasmic microinjection bypasses all natural barriers to sperm penetration and removes all biological selectivity from the process, such as sperm binding to the ZP and fusion of the sperm with the oolemma (Gordon and Laufer, 1988). Because of this reason, clinicians preferred other micromanipulation methods without disruption of normal fertilization process. Therefore, attempts were made to develop other micromanipulation techniques to assist fertilization which include Zona Drilling (ZD), Partial Zona Dissection (PZD) and Sub-Zonal Sperm Injection (SUZI). Although these techniques were successful in producing live offspring in laboratory animals as well as in human, however, they were not able to become the method of choice due to a number of reasons. Therefore, the principal breakthrough and method of choice in the field of assisted fertilization was ICSI.

Table 2:	History of sperm injection attempts in different animals and human

Fig. 1:	Diagrammatic representation of Zona Drilling (ZD). Acid Tyrodes solution is used to make a hole in the ZP to allow sperm access to the oocyte. Drawn by the Author from Catt (1996)

The salient features of other micromanipulative techniques are described briefly in the following subsections.

Zona drilling: The ZD technique was first described by Gordon and Talansky (1986). This technique involves the use of a fine stream of acid Tyrodes solution (pH 2-3) to dissolve a small hole in the ZP of a cumulus-free and mature oocyte held with a micromanipulator (Fig. 1). The piercing of ZP can also be done with a fine needle after prior softening with chymotrypsin (Iritani, 1991, 1994). Like IVF, the sperm must be capacitated, acrosome reacted and progressively motile in ZD. The fact that fertilization occurred at increased rates after ZD suggests that the ZP is a significant barrier to the fertilizing sperm and that sperm do not necessarily have to come in contact with the outer surface of the ZP for the acrosome reaction to occur. Although, ZD could be utilized for oligospermia with weak sperm motility, this cannot be applied to immotile sperm and the oocytes are also damaged to certain extent (Iritani, 1991, 1994). Fertilization rates with ZD in human ranged from 6 to 35% compared with 3 to 13% in control oocytes (Payne, 1995). However, polyspermy and immature hatching due to big hole (approximately 15.2 μm in diameter) (Payne, 1995) are major problems. Live births after ZD were reported in mouse (Gordon and Talansky, 1986) and in human (Jean et al., 1992). In clinical practice ZD method was discontinued when it was discovered that the human oocyte is particularly sensitive to the low pH of the acidified Tyrodes and that high oocyte degeneration and poor embryonic development occurred as a consequence of its use (Payne, 1995).

Fig. 2:	Diagrammatic representation of Partial Zona Dissection (PZD). A cumulus-free oocyte is held by suction onto a holding pipette and a fine glass needle is pushed through both sides of the ZP; the oocyte is then released and the glass needle is pushed against the holding pipette until a slit in the ZP is produced. Drawn by the Author from Payne (1995)

Partial zona dissection: The PZD technique is also known as zona cutting or zona opening. The oocyte is held by a holding pipette and a fine glass pipette or metal microprobe is pushed through the ZP and the oocyte is released from the holding pipette and rubbed against the cutting pipette until a slit is produced (Fig. 2). The hole produced in PZD is much smaller (approximately 8.9 μm in diameter) (Payne, 1995) than that produced in ZD. Like ZD, the sperm must be capacitated and acrosome reacted for fusion with oolemma in PZD. Instead of using glass microneedles or metal microprobe, laser technology has been investigated as an alternative method for creating opening in the ZP (Feichtinger et al., 1992; El-Danasouri et al., 1993). Using this technique, fertilization and live birth had been reported in human (Malter and Cohen, 1989; Feichtinger et al., 1992). However, the technique has not been widely used because it requires relatively high number of progressively motile sperm and moreover, the incidence of polyspermy was high even when few sperm were used (Goto, 1997). Compared with IVF, the results of PZD were also found to be unsatisfactory and were associated with erratic but generally low normal fertilization rates (Cohen, 1992).

Sub-zonal injection: In the next step after ZD and PZD, the microinjection of sperm into the perivitelline space (PVS) or under the ZP was developed which is known as Sub-Zonal Injection (SZI) or SUZI. In SUZI, 5 to 8 sperm are generally injected into the PVS of cumulus-free and matured oocytes (Fig. 3), so that selection of biologically normal sperm is expected to a certain extent during the process of membrane fusion (Iritani, 1991, 1994; Goto, 1997). Although the technology for SUZI had been in existence before the clinical use of ZD and PZD (Metka et al., 1985; Laws-King et al., 1987), it was not used clinically until the limited application and success of ZD and PZD were demonstrated (Payne, 1995). The technique circumvents the process of sperm penetration through the ZP, so that sperm with extremely poor motility can be injected. However, fertilization rates after SUZI were similar to those using the previous techniques, yet polyspermy was a particular problem, with some IVF units reporting polyspermy in more than 30% of the fertilized oocytes (Payne, 1995). Although SUZI led to the birth of live offspring in the mouse (Mann, 1988) and human (Ng et al., 1988); the incidence of polyspermy limits its use and thus, like ZD and PZD, SUZI could not be popularized (Goto, 1997).

In general, the greatest weakness of ZD, PZD and SUZI methods was their inefficiency, as they fail to achieve normal fertilization rates beyond 20% and naturally produce high rate of polyspermic fertilization (Cohen, 1992). Whereas using only one sperm achievements in ICSI are far better than the other micromanipulative techniques. The advantage of ICSI over other micromanipulative techniques are that any type of sperm (sperm head, immotile, dead or grossly defective sperm) can be injected and both capacitation and acrosome reaction appear to be unnecessary for fertilization (Iritani, 1991, 1994). Mechanical pricking of the oocytes and the infusion of reagents such as polyvinylpyrrolidone (PVP) in the injection medium appear to be effective for oocyte activation (Markert, 1983; Iritani and Hosoi, 1989).

REPORTS OF LIVE BIRTHS THROUGH ICSI

The first report of live birth through ICSI came from Japan with rabbit (Hosoi et al., 1988). Since then, many reports have been published on live birth in a number of mammalian species, even in insects and birds. However, ICSI technique had revolutionized in 1992 when human babies born in Belgium using this technique (Palermo et al., 1992). The ICSI proved its efficacy and successfulness to overcome MFI in human and well accepted by the clinicians due its higher success rates. Since its first success in 1992, thousands of human babies had born using ICSI technique that represented a major breakthrough in the field of assisted fertilization. This way, micromanipulation techniques have gradually developed over the century and have reached such a precision that skilled scientists are able to micro-dissect one single chromosome. The ICSI procedure is illustrated in the Fig. 4. The chronological report (first report only) of live offspring production through ICSI in different species of animals and human has been depicted in Table 3.

Fig. 3:	Diagrammatic representation of sub-zonal injection (SUZI). Several sperm are selected and injected under the ZP in SUZI . Drawn by the Author from Payne (1995)

Fig. 4:	Diagrammatic representation of ICSI. The ICSI involves the injection of a single, live, immobilized sperm into the ooplasm of the oocyte. Drawn by Author

Table 3:	Chronological reports of live offspring production through ICSI in animals and human

REMODELING OF FERTILIZATION PROCESS AND OOCYTE ACTIVATION IN ICSI

The ICSI is the latest and by far the most efficient, variant of micromanipulated-assisted fertilization, whereby a single sperm is selected, aspirated into a microinjection needle and injected into ooplasm of a MII oocyte. The development of this technique is mainly linked to application in human assisted reproduction for which it enables fertilization with defective sperm that would not otherwise be able to penetrate an oocyte by proper means. Because ICSI bypasses many steps of natural fertilization process, therefore, remodeling of fertilization process that occurs in ICSI is different from that of fertilization in vivo or in vitro. The ICSI has contributed greatly to the knowledge concerning the different stages of fertilization, such as sperm capacitation, AR, membrane fusion, oocyte activation and ooplasmic factors involved in the transformation of sperm nucleus into a MPN. The factors involved in sperm pronuclear development are not species-specific. It was demonstrated that sperm of some species (e.g., human sperm) could develop into normal MPN in another species (e.g., hamster oocyte) and form a normal chromosome complement (Elder and Dale, 2000). Sperm injection from a number of species, for example, frog, hamster and human was able to activate hamster oocyte and subsequently developed to MPN (Perreault et al., 1988). This result suggests that sperm interaction with the ooplasm is less species specific than the interaction with the oolemma (Perreault et al., 1988). The ooplasmic factors controlling the transformation of sperm nucleus to MPN are also not species specific (Iritani, 1991).

Sperm capacitation might not be necessary to attain fertilization through ICSI as oocyte activation was induced with injection of both capacitated and non-capacitated sperm (Thadani, 1979). Although, in goat ICSI, generally capacitated sperm are used (Keskintepe et al., 1997; Jimenez-Macedo et al., 2005; Rahman et al., 2007, 2009; Abdullah et al., 2008), a higher fertilization rate was also obtained with un-capacitated washed sperm (Wang et al., 2003). During fertilization in vivo (or in vitro) sperm penetrates the ZP without its plasma membrane over the acrosomal cap, together with the underlying outer acrosomal membrane and acrosomal contents disperse. The remaining plasma membrane over the equatorial segment fuses with the oolemma. Thus, all the remaining portions of the sperm including the nucleus, perinuclear material, centriole and tail are engulfed by the ooplasm in a phagocytic manner (Yanagimachi, 1994, 1997). In contrast, during ICSI whole sperm is incorporated into the ooplasm including the acrosomal content and membranes. In this case, although sperm binding to ZP is bypassed, the AR still could occur in the ooplasm of a mature Metaphase II (MII) oocyte (Ben-Yosef and Shalgi, 1998, 2001). The AR is preceded by acrosomal swelling followed by vesiculation of surface membranes, exposing the inner acrosomal membrane, similar to that observed on the surface of the ZP during IVF (Sathananthan et al., 1997). It is demonstrated in the ICSI studies that barriers penetrated by sperm during fertilization in vivo or in vitro are probably not biologically necessary and can be bypassed by direct sperm injection into the ooplasm.

Other important features of in vivo fertilization or IVF, for example capacitation and AR - the functional changes that the sperm undergoes in order to acquire the fertilization potential - are also not a prerequisite to ICSI (Yanagimachi, 1994, 1997). From different ICSI studies, it is now known that all the sperm components are not required for successful oocyte activation at fertilization. In mouse, injection of isolated sperm head is enough to activate oocyte (Kimura et al., 1998) and sperm centriole is not necessary for fertilization (Schatten, 1994). However, in other mammals, for example human, cattle, sheep or goat sperm centriole, a structure later serving as an organizer of zygotic centrosome, is very important (Le Guen and Crozet, 1989; Crozet, 1990; Navara et al., 1994; Schatten, 1994; Sathananthan et al., 1997). The sperm centriole plays a key role in the formation of microtubule within the zygote. A radial array of microtubules, called the sperm aster, is formed, which emanates from the sperm centrioles following its exposure to the ooplasm. The sperm aster is responsible for bringing the MPN and female pronucleus (FPN) into close contact to enable syngamy (Yanagimachi, 1994). Therefore, in mammals excluding rodent species sperm head and neck where centrioles are located, must be injected. In goat, high fertilization and embryo development rates were achieved after injection of tail-cut sperm near the mid-piece (Wang et al., 2003).

The ability of the sperm to induce oocyte activation following injection is a crucial requirement for successful fertilization. Two hypotheses have been proposed to explain the mechanisms by which the sperm induces the activation of the mammalian oocyte: the receptor hypothesis model and the sperm factor hypothesis. In the receptor hypothesis, an oocyte surface receptor is coupled to a G-protein (Miyazaki et al., 1993) or a tyrosine-kinase-mediated (Ben-Yosef and Shalgi, 1998) signaling pathway and when activated by a sperm, leads to the release of Ca⁺² from intracellular stores. These sperm induced Ca⁺² oscillations usually last several hours after the initial Ca⁺² increases and the frequency shows considerable variation between species (Jones, 1998). The sperm factor hypothesis, on the other hand, suggests that egg activation is initiated only after sperm-oocyte fusion, by introducing a sperm-borne-oocyte-activating-factor (SOAF) or Sperm Factor (SF) into the ooplasm (Stice and Robl, 1990; Swann, 1990). When homogenized rabbit sperm was microinjected into rabbit oocytes, activation rates were higher than in oocytes injected with medium alone; activated oocytes exhibited cortical granule exocytosis, MPN formation and cleavage (Stice and Robl, 1990). Additionally, sperm extracts do not appear to be species-specific as rabbit sperm fractions activated both rabbit and mouse oocytes following injection into the ooplasm (Stice and Robl, 1990). Finally, the ICSI procedure itself bypasses sperm-oocyte binding and fusion and its success also supports the soluble sperm factor hypothesis. Recently, a hypothesis combining the two major hypotheses was proposed (Tesarik, 1998). In this trigger and oscillator hypothesis, the role of the trigger or initiation of Ca⁺² oscillations, normally released by sperm-oocyte cell surface interactions in normal fertilization, is substituted for ICSI by an artificial Ca⁺² influx generated by the ICSI procedure itself. In this hypothesis, the sperm cytosolic factors play the role of the oscillator.

It is generally accepted that intracellular increase in Ca²⁺ level, which mimic fertilization very closely (Jones et al., 1995), is the universal signal for triggering oocyte activation and subsequent embryo development in all mammals studied so far (Vitullo and Ozil, 1992; Homa et al., 1993; Tesarik, 1995). Moreover, it is a histone kinase inhibitor and prevents the re-accumulation of M-phase or maturation promoting factor and this yields an improvement in the efficiency of oocyte activation (Susko-Parrish et al., 1994). As observed in fertilization in vivo or in vitro, Ca²⁺ oscillations are characteristics of activated ICSI oocytes which continue for several hours after the ICSI procedure (Sousa et al., 1996), but the dynamics of the response is different from IVF. In contrast, no Ca²⁺ oscillations are observed after sham injection of oocytes with sperm-free culture medium (Tesarik et al., 1994), demonstrating the decisive role of SF in this process (Tesarik and Mendoza, 1999). Although, no studies related to Ca²⁺ oscillations pattern after ICSI in goat was reported, one study reported that a typical oscillatory Ca²⁺ release pattern does occur following IVF in the goat, as previously been shown in other mammalian species (Jellerette et al., 2006). Analyses of human unfertilized oocytes by ICSI indicated that this caused by a failure of oocyte activation (Sousa and Tesarik, 1994; Tesarik, 1995). In fact, a high fertilization rate is reported in human oocytes after ICSI and artificial induction activation with calcium ionophore (Ca²⁺ ionophore) or electric pulse (Tesarik, 1995; Abe et al., 1996). Furthermore, normal offspring were born after the artificial activation of ICSI oocytes in cattle (Goto et al., 1990), sawfly (Hatakeyama et al., 1994) and human (Hoshi et al., 1994). However, abnormalities of Ca⁺² signals driving oocyte activation after ICSI are suspected to be at the origin of various abnormalities in the resulting embryos including chromosomal abnormalities (Tesarik, 1995). Therefore, a proper method of oocyte activation that mimics a sperm must be developed.

The exact timing of fertilization events in vivo or the IVF procedure is difficult to evaluate, as the exact moment of gamete fusion cannot be determined. Several studies have monitored the ultrastructure of sperm-oocyte interaction and the timing of the fertilization events after ICSI in human (Nagy et al., 1994; Bourgain et al., 1998). The sign of AR observed 15 minutes after ICSI procedure, whereas in IVF or in vivo fertilization it occurs before gamete fusion. Cortical Reaction (CR) that occurs in IVF within a second after sperm entry takes 90 min to occur after ICSI. Decondensation of sperm head chromatin occurred from 30 min to 3 h after ICSI. Relocation of MII spindle with a paired set of chromosome near the oolemma in close apposition to the polar body-1 (PB-1) occurs up to 3 h of ICSI. Extrusion of PB-2 started after 4 h of ICSI. The MPN is formed before the FPN and around 8 h after ICSI approximately 80% PN formation is completed; 100% PN formation complete by 16 h (Nagy et al., 1994). Although, timing of extrusion of PB-2 and PN formation are similar in both IVF and ICSI procedure, but ICSI embryos start to cleave earlier (approximately 20 h) than IVF. Therefore, all the events which are bypassed by ICSI procedure occur in the first few hours after sperm injection within the oocyte (Ben-Yosef and Shalgi, 1998, 2001). However, chromatin remodeling is more asynchronous in ICSI than in IVF. In their study, Ajduk et al. (2006) compared sperm chromatin remodeling in IVF and ICSI in the mouse. Their results demonstrated that sperm capacitation prior to injection enhanced remodeling asynchrony and resulted in delayed PN formation and DNA synthesis. They also found that injection of Ca²⁺ ionophore into the mouse oocyte after removal of acrosome allowed more synchronous chromatin remodeling, timely DNA synthesis and good embryo development than that of detergent Triton X-100. Although the dynamics and ultrastructure of sperm interaction within ooplasm of an oocyte occur somewhat differently in ICSI than fertilization in vivo or in vitro, most of the cellular events responsible for mediating oocyte activation are stimulated in the same manner (Ben-Yosef and Shalgi, 2001).

FACTORS AFFECTING THE OUTCOMES OF ICSI

As in all micromanipulation techniques, ICSI involves intricate and precise maneuvers of microtools on the oocytes and sperm under an inverted microscope. Although the skill and experience of the embryologist in performing ICSI is important, the outcome of the procedure depends largely on the quality of the microtools such as the microinjection needles, sperm immobilization and injection technique. Other factors such as oolemma characteristic, orientation of PB, use of PVP, oocyte in vitro maturation (IVM) conditions, oocyte morphology and IVC conditions are also important. Although, ICSI does not have strict requirements of sperm number, motility, morphology and quality to achieve pregnancies; some of the sperm parameters may affect the outcome of ICSI. Some of the important factors affecting ICSI outcomes are describes in the following subsections mostly based on human ICSI.

Injection needle: Injection of sperm into the ooplasm is the most invasive micromanipulation technique in assisted reproduction. Hence, the design and quality of the microtools especially the injection needle is of utmost important as the size and shape of the ICSI needle will determine the success or failure of a micromanipulation procedure. Generally, thick- and thin-walled borosilicate glass capillaries are used to make holding pipettes and ICSI needles, respectively (Payne, 1995). There are two features of the injection needle which are important for ICSI. First, the needle must be beveled and sharp, enough to perforate both the ZP and the oolemma. Second, the outer and inner diameter (OD and ID) of the needle should be perfect, just large enough to draw in a sperm head. Although no reports were published in goat, fertilization and embryo development from bovine oocytes after ICSI was reported to be affected by the OD and ID of injection needles (Tocharus et al., 1996). The percentage of embryo development was statistically higher with narrower diameter injection needles. The additional mechanical force and increased volume of injected solution with the wider diameter needles cause too much disruption of the ooplasm and arrest development (Tocharus et al., 1996). However, if too narrow needle is used it will be difficult to move sperm in and out of the needle (Payne, 1995). Providing that these criteria are met, the bevel angle can vary from 28 to 30° (Hamberger et al., 1995; Palermo et al., 1995) to about 45 to 50° (Van Steirteghem et al., 1993) and the needle can either be fitted with a sharp spike or they can be used without a spike (Catt et al., 1995). To achieve better results in goat ICSI, the ID of ICSI needle should be between 6 and 7 μm with a sharp spiked tip (Wang et al., 2003; Abdullah et al., 2008; Rahman et al., 2009).

Sperm immobilization prior to ICSI: Immobilization of motile sperm prior to ICSI, by touching the tail with the needle is mandatory even with immotile sperm (Fishel et al., 1995; Van den Bergh et al., 1995; Vanderzwalmen et al., 1996). Sperm immobilization induces permeabilization of the sperm membrane which may result in the release of a cytosolic factor that diffuses into the ooplasm with subsequent intracellular Ca²⁺ oscillations and hyperpolarization of the oocyte (Homa and Swann, 1994; Dale et al., 1999). This phenomenon is known as oocyte activation. As already mentioned earlier, the activation of the oocyte triggers a series of biochemical processes in the ooplasm that would eventually lead to sperm nuclear decondensation (Tesarik and Kopecny, 1989; Montag et al., 1992), PB-2 emission, PN formation and exocytosis (Stice and Robl, 1990; Swann, 1990). At the edge of the sperm-PVP droplet, a morphologically normal motile sperm is immobilized by touching its tail with the needle. The sperm tail is then crushed by sliding the needle on the mid-piece of the tail until a kink is seen. It is then aspirated tail-first into the needle. The sperm neck contains the centriole that contributes a major role in fertilization in mammals other than rodents; therefore, care should be taken not to damage the neck. Proper immobilization technique is necessary to ensure leakage of SOAF or SF into the ooplasm which have significant function in the fertilization process.

The role of polyvinylpyrrolidone in ICSI: The PVP, a large molecular weight polymer (molecular weight 360,000), has been used since the beginning of assisted fertilization programs (Lanzendorf et al., 1988). The PVP when dissolved in a solution becomes very viscous which allows better manipulation of sperm, control of the ooplasmic aspiration and sperm injection procedures. It also prevents adhesion of sperm to the inner wall of the injection needle (Uehara and Yanagimachi, 1976). The use of PVP also allows the three-dimensional motility patterns of the sperm to be carefully observed (Cohen et al., 1994). Control of the volume of ICSI medium injected into the oocyte is more difficult without using PVP (Joris et al., 1998). The concentration of PVP is an important factor affecting the outcome of ICSI and usually 10% (w/v) PVP is used (Palermo et al., 1992). Gordon (1997) examined the effect of different PVP concentrations as a sperm retardant for ICSI in bovine oocytes matured in vitro, as measured by cleavage rate and early embryo development. Although the damage rate in oocytes following ICSI was affected significantly by the level of PVP in the injection medium, there was no difference in the cleavage rates in all groups studied. However, the efficacy and safety of sperm immobilization process and use of PVP has been questioned as it is accused of having a potential harmful, mutagenic or carcinogenic effect (Ashwood-Smith, 1971; Feichtinger et al., 1995; Jean et al., 1996) and may also interfere with sperm nuclear decondensation (Dozortsev et al., 1995b). Using PVP, mouse sperm are found susceptible to alterations that affect both fertilization and development (Mizuno et al., 2002). In their study, Motoishi et al. (1996) evaluated the safety of ICSI procedures on bovine zygotes by using PVP and concluded that the ICSI procedures with PVP currently used for animal and human are neither detrimental to embryo development nor to embryo quality. There are reports of better fertilization rates with PVP-free medium (Butler and Mason, 1996; Jean et al., 1996), even though embryo quality and implantation rates remain the same (Hlinka et al., 1998; Tsai et al., 2000). In goats, no reports have been published on the effect of sperm immobilization with PVP or use of PVP in ICSI medium on the embryo development. Until now all the researchers dealing with goat ICSI used PVP (Wang et al., 2003; Jimenez-Macedo et al., 2006; Keskintepe et al., 1997; Abdullah et al., 2008; Rahman et al., 2008c, 2009). Therefore, PVP will still be the substance of choice in ICSI unless its potential harmful effects are proven to be true or a safer and better alternative is found.

Polar body orientations of the oocyte for ICSI: During ICSI, the MII oocyte is held under minimal suction by the holding pipette with orientation of the PB-1 at either 6 or 12 o'clock. This orientation has been used in the first ICSI pregnancy in human (Palermo et al., 1992) and thereafter continued to be used in human and animal ICSI because, damage to the MII spindle could be avoided. Although there was no significant difference in the survival or normal fertilization rates when injection of oocytes were compared between 6 and 12 oclock PB orientation, injection of oocytes at 6 oclock PB orientation resulted in higher number of good quality embryos (Nagy et al., 1995) and possible pregnancy rates (Van Der Westerlaken et al., 1999). Except for the injection of the oocyte at 9 oclock, normal fertilization can occur irrespective of the other PB orientations. However, the PB orientation that could result in the sperm deposited near the MII spindle subsequently could result in better development of the embryo. In a retrospective analysis of oocyte fertilization and embryo development with respect to sperm deposition during ICSI (Blake et al., 2000) demonstrated that injection of human oocytes with the PB oriented at 7 or 11 oclock position resulted in higher number of high quality embryos, significantly more than the conventional 6 or 12 oclock PB orientation. However, till now, goat oocytes are injected either at 6 or 12 oclock PB orientation (Keskintepe et al., 1997; Wang et al., 2003; Jimenez-Macedo et al., 2007; Rahman et al., 2007; Abdullah et al., 2008).

General ICSI technique: Although the timing of sperm injection does not significantly affect the survival and fertilization rates of the oocytes and embryo quality (Van de Velde et al., 1998), higher number of good quality human embryos can be obtained when ICSI is performed between 1 and 9 h after oocyte recovery (Yanagida et al., 1998). However, unlike human, in goats when oocytes collected by LOPU 36 h (Baldassarre et al., 2003), which is a general practice for OR from live donors, or even 60 h (Abdullah et al., 2008; Rahman et al., 2008b) after eCG treatment are still at the immature stage and require IVM for 27 h before being able to acquire meiotic competence. The ICSI technique performed by most human IVF centers require that the oolemma is broken at the point of injection before the sperm is delivered into the ooplasm (Dozortsev et al., 1995a; Palermo et al., 1996; Vanderzwalmen et al., 1996; Joris et al., 1998). Ooplasmic aspiration to induce oolemma breakage has been considered as an integral part of the ICSI procedure as it ensures that the sperm is in the oocyte and in intimate contact with the ooplasm. Factors that are present in the ooplasm promote the formation of MPN (Tesarik and Kopecny, 1989; (Montag et al., 1992). Ooplasmic aspiration is also proposed to trigger the initial steps in inducing oocyte activation by an artificial Ca⁺² influx generated by ICSI Tesarik et al. (1994) and Tesarik (1998). It is to be remembered that ooplasmic aspiration is a delicate part of the technique. Gentle ooplasmic aspiration within the needle would result in higher survival rate of the oocytes and embryo quality (Vanderzwalmen et al., 1996; Carrillo et al., 1998). Vigorous ooplasmic aspiration as advocated by Tesarik and Sousa (1995) as a crucial step to the success of ICSI would instead increase the rate of oocyte damage and did not improve fertilization rate. The amount of ooplasm aspirated depends on the size of the needle and the speed in which it is performed. Minimal amount of ooplasm aspirated with a slow and gentle aspiration ensures that the oocyte is not unduly traumatized. On the other hand, large amount of ooplasm aspirated with or without a sudden aspiration will alter the integrity of the oocyte resulting in degeneration, failure to fertilize or poor embryo quality. Likewise, gentle re-injection of sperm with minimal amount of PVP is preferred. This can be achieved by placing the sperm at the tip of the needle that has already been positioned close to the ZP prior to injection. The sperm should be injected through the oocyte at 9 oclock position. Similar procedure is also followed during ICSI of farm animal oocytes including goat (Keskintepe et al., 1997; Wang et al., 2003; Jimenez-Macedo et al., 2007; Abdullah et al., 2008; Rahman et al., 2009). The average time taken to inject an oocyte is about one minute. It is to be remembered that oocytes are not recommended to expose on the micromanipulation microscope for more than 10 minutes (Keskintepe et al., 1997); less exposure time will provide better results. To maintain proper temperature, a transparent heating plate or stage warmer is generally used on the ICSI microscope. The temperature of the stage should correspond to the temperature that is maintained inside the CO₂ incubator and must be specific for species. For goat, 38.5 to 39°C temperatures is generally maintained (Keskintepe et al., 1997; Wang et al., 2003; Jimenez-Macedo et al., 2007; Abdullah et al., 2008; Rahman et al., 2009). To keep the pH of the ICSI medium at 7.2 to 7.4, HEPES is supplemented in the ICSI medium (Keskintepe et al., 1997; Wang et al., 2003; Jimenez-Macedo et al., 2007; Abdullah et al., 2008; Rahman et al., 2009).

ICSI technique in relation to flexibility of the oolemma: The oolemma characteristic is also important for efficiency in ICSI. The oolemma of oocyte is flexible or elastic; however, in some cases flexibility or elasticity of oolemma is very high or low. As ICSI begins with the insertion of a needle, damage to the oocyte, especially those having very high or low elasticity, cannot be avoided. Although advances in ICSI technology have improved oocyte survival rates of 50 to 70% to current level of 85 to 90% or even better (Tucker et al., 1996; Van Steirteghem et al., 1996; Devroey, 1998; Yanagida et al., 1999), but ICSI technicians acknowledge the high probability of oocyte damage sometimes, even if the ICSI is performed carefully and properly with oolemma of normal elasticity. When the oolemma is less elastic, the ICSI needle can penetrate easily as soon as inserted into oolemma and when it is more elastic become very difficult to penetrate. Therefore, needle with a sharp spiked tip is generally used to minimize damage to the oolemma during ICSI (Nagy et al., 1995; Palermo et al., 1996). However, the processing of needle tips requires skill and it is difficult to obtain tips of consistent sharpness (Yanagida et al., 2001). The size of the sperm head of buck is bigger than that of human and as a result the pore size of the injection needle in goat ICSI is larger than that used for human ICSI. Therefore, like in bovine ICSI, breakage of oolemma by aspiration results in remarkable damage in goat (Rahman et al., 2007). Also, oolemma of goat oocyte is elastic like that of cattle and mouse and therefore difficult to puncture them by conventional ICSI (Horiuchi and Numabe, 1999). To overcome this difficulty, avoid damage and improve the efficiency of ICSI, a Piezo-driven procedure has been developed (Kimura and Yanagimachi, 1995) by which oolemma breakage can be done using completely flattened tips. Using a Piezo-driven micromanipulator, ICSI efficiencies were dramatically improved in mice (Kimura and Yanagimachi, 1995), cattle (Katayose et al., 1999), human (Huang et al., 1996), rats (Hirabayash et al., 2002), horses (Choi et al., 2002) and goat (Wang et al., 2003). This technique is far less traumatic to mouse oocytes than the conventional method using a mechanically-driven pipette when performing ICSI (Kimura and Yanagimachi, 1995). The Piezo-electric effect is used to produce a stabbing, punctuate movement to puncture the oolemma with minimal distortion of the oocyte. In Piezo-ICSI aspiration of ooplasm is not necessary which can reduce the chance of oolemma damage. The incidence of normal fertilization was about 10 times higher in Piezo-operated oocyte than in oocytes injected using the conventional method (Kimura and Yanagimachi, 1995). This technique is now extensively used in human and, in some extent, in animal ICSI. In goat, using Piezo-driven ICSI with a sharp spiked needle tip, high fertilization and embryo developmental rates were obtained without any activation treatment (Wang et al., 2003). However, this technique has not been popularized in goat ICSI yet. Therefore, goat ICSI is still done conventionally without the used of a Piezo-driven micromanipulator and usually sperm or oocytes are treated to get better fertilization and embryo developmental rates (Keskintepe et al., 1997; Zhou et al., 2004; Jimenez-Macedo et al., 2005-2007).

In vitro matured oocytes and ICSI outcomes: In vitro matured oocytes are more vulnerable to damage by in vitro manipulation than in vivo matured oocytes (Leibfried-Rutledge et al., 1987; Pivko et al., 2003). If the IVM conditions are not optimal, it can lead to decrease in inositol IP₃ and glutathione stored via Cumulus Cells (CCs) which may give rise to the possibility that this may adversely affect oocyte activation and MPN formation (Chian et al., 1994; Geshi et al., 1999). The amount of stored glutathione in the ooplasm of bovine oocytes was low that underwent IVM in unsuitable culture conditions (Geshi et al., 1999), that is depletion of glutathione in the oocytes block decondensation of the MPN during fertilization (Sutovsky et al., 1997; Sirard et al., 2006). Most of the oocytes from caprine ovarian follicles with diameter of 2 to 6 mm progress to MII stage after IVM (De Smedt et al., 1994; Crozet et al., 2000). However, these oocytes might not be effectively activated by artificial activation because IVM causes disturbances in nuclear, ooplasmic and CC maturation which would normally occur simultaneously in vivo (Sato, 1991; Kikuchi et al., 2000). It is found that bovine oocytes after 28 h of IVM can be easily activated, although the developmental competence of the embryos obtained was low (Long et al., 1994). Hence a period of 26 to 27 h after IVM is probably the optimum activation timing of non-aged oocytes as this time duration provided the highest number of viable embryos after ICSI (Wang et al., 2003; Rahman et al., 2008a). In bovine it is found that extended time after IVM adversely affect PN formation and subsequent embryo developmental competence of IVF or ICSI-derived embryos (Long et al., 1994; Balakier et al., 2004).

Oocyte morphology: Oocyte quality is an important prognostic factor as the nuclear and ooplasmic maturity of the oocyte may be directly related to the success rate of ICSI (Kahraman et al., 2000). A search of available literature could not generate any work that relates oocyte quality or morphology with the outcome of ICSI in farm animals especially in goat. However, there are a number of studies reported in human in this regard. Oocyte denudation prior to an ICSI procedure has revealed that a significant proportion of oocytes exhibit different morphological abnormalities, such as excessive granularity, vacuolarization, clustering of smooth endoplasmic reticulum, refractile bodies, large PVS and fragmented PB. Although, morphological appearance of goat oocyte is quite different from that of human, it is likely that a similarity in oocyte dysmorphism may exist. The extent to which these abnormalities can interfere with the normal process of fertilization following ICSI is controversial. Some authors did not find any correlation between oocyte morphology, fertilization rates and embryo quality (De Sutter et al., 1996; Balaban et al., 1998), others have reported a significant influence of oocyte morphology on fertilization rates and embryo quality after ICSI (Sehral et al., 1997; Xia, 1997; Rahman, 2008). In their study Lourtradis et al. (1999) found human oocytes with dark ooplasm; ooplasmic vacuoles and fragments led to poor-quality embryos. Although normal fertilization and embryo development were achieved in oocytes with abnormal ooplasm, however, the resulting embryos failed to demonstrate the same implantation potential as those derived from normal oocytes (Sehral et al., 1997). In contrast, no significant differences in fertilization, embryo quality, implantation and pregnancy rates were found after ICSI with normal and abnormal oocytes (Balaban et al., 1998) or oocytes with abnormal morphology or Centrally Located Granular Cytoplasm (CLGC) (Kahraman et al., 2000), although chromosomal abnormality was found with CLGC oocytes in preimplantation genetic diagnosis (Kahraman et al., 2000). Although, not with ICSI, a few reports on oocyte quality and embryo developmental competence after IVF in cattle have been published. Except one (Nagano et al., 1999), all were in the opinion that poor quality oocytes had lower developmental competence than those of good quality oocytes (Nagano et al., 2006, 2007). As till now, no reports published on ICSI outcome in relation to oocyte quality in farm animals.

Sperm parameters: ICSI results are not correlated with the total sperm count as high fertilization and pregnancy rates can be obtained in cases of very low sperm count (extreme oligozoospermia) (Nagy et al., 1995). Sperm morphology is also not much related with ICSI outcome. In their study Lee et al. (1996) reported that although some morphological abnormalities, namely amorphous, round and elongated heads are associated with chromosome defects; surprisingly, no increase in chromosome aberrations were found in sperm with large or small heads compared with normal ones. Burruel et al. (1996) injected sperm with grossly misshapen heads into mouse oocyte and obtained high fertilization and embryo development rates (>80%); some of the embryos developed to normal fertile adults. This indicated that a proportion of abnormal sperm carry all the genome and organelles necessary for embryonic development and growth to fertile maturity. Therefore, morphological abnormality of sperm might not necessarily reflect a genetic abnormality of the male gamete rather its inability to cross physiological barriers (Ben-Yosef and Shalgi, 2001). In contrast to other sperm parameters stated above, sperm motility and livability are correlated with ICSI outcomes as motility reflects sperm vitality. Normally sperm are considered dead when their plasma membranes are severely damaged or lost. Such sperm are no longer motile and will not fertilize oocytes; and obviously these are dead as cells (Goto, 1997). However, normal offspring were obtained from killed or dead sperm from a number of species by using frozen-thawed sperm without cryoprotectant (Goto et al., 1990; Hatakeyama et al., 1994; Hoshi et al., 1994; Wakayama et al., 1998). These reports indicated that sperm injected into the oocytes were dead as cells but the nuclei were alive in terms of reproduction and genetics (Goto, 1997). It was found that initially immotile sperm show some motility after 2 to 3 h of incubation (Nijs et al., 1996). Lower quality embryos tended to be produced from totally immotile (dead) sperm compared with initially immotile sperm. This is probably due to the fact that some of the totally immotile (dead) sperm DNA have already started degeneration process, although pregnancies were reported after ICSI in both cases (Goto, 1997). It was reported that even after freezing-thawing, a significant amount of SF is still present in the sperm because after this treatment sperm are still able to activate mouse oocyte, whereas immotile (dead) sperm failed to do so (Dozortsev et al., 1995b). However, in clinical situation, it is impossible to know which sperm has a normal DNA as some of the immotile sperm may have degenerating or degenerated DNA. Therefore, motile sperm should be selected for ICSI.