Introduction
It is well known that oocyte activation is induced when fertilizing spermatozoa inactivate metaphase-promoting factor (MPF), which blocks the oocyte cell cycle during metaphase of the second meiotic division (MII) and then initiates normal embryonic development (including exocytosis of cortical granules for blocking polyspermic fertilization), pronucleus formation and cleavage division progression. In all mammalian species studied to date, it is also known that repetitive intracellular [Ca2+]i elevations (called Ca2+ oscillation), which are released from the oocyte's endoplasmic reticulum, occur. This Ca2+ oscillation is not essential for normal embryonic development to term, but plays an important role in effective embryonic development (Yazawa et al., Reference Yazawa, Yanagida and Sato2001).
In previous studies using microinjection of immature spermatogenic cells from experimental animals and human (Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000, Reference Yazawa, Yanagida and Sato2001, Reference Yazawa, Yanagida and Sato2007; Ogonuki et al., Reference Ogonuki, Sankai, Yagami, Shikano, Oda, Miyazaki and Ogura2001), it was demonstrated that sperm-borne oocyte-activating factor (SOAF) becomes biologically active during the maturation of spermatogenic cells, but the timing of the acquisition of SOAF activity differed among species. For example, mouse round spermatids (ROSs) do not have any SOAF activity, hamster and rabbit ROSs have intermediate levels of SOAF activity and monkey and human ROSs have some SOAF activity. Moreover, except for rare cases of fertilization failure with ICSI in mice and humans (Tesarik & Sousa, Reference Tesaric and Sousa1995; Yanagida et al., Reference Yanagida, Katayose, Yazawa and Sato1999, Reference Yanagida, Morozumi, Katayose, Hayashi and Sato2006; Elder-Geva et al., Reference Eldar-Geva, Brooks, Camnar, Marqalioth, Zylber-Haran and Silber2003), live mature spermatozoa possess sufficient SOAF activity.
To examine the fertilizing ability of immobilized (killed) mouse and human spermatozoa, we sought to determine when and how their SOAF activity declined after spermatozoa immobilization by analyzing their ability to activate oocytes and induce Ca2+ responses following injection into mouse oocytes.
Materials and Methods
Preparation of mouse oocytes
Six- to 8-week-old B6D2F1 female mice were superovulated by consecutive i.p. injections (48 h apart) of 8 IU pregnant mare's serum gonadotropin (PMSG; Teikokuzouki Co.) and 8 IU human chorionic gonadotropin (HCG; Mochida Pharmaceutical Co.). Mature oocytes were collected from the oviducts 15–16 h after hCG injection and freed from cumulus cells by treatment with 0.1% hyaluronidase in HEPES-buffered human tubal fluid medium (mHTF; Irvine Scientific) supplemented with 10% synthetic serum substitute (SSS; Irvine Scientific). The cumulus-free oocytes were rinsed thoroughly, placed in drops of human tubal fluid medium (HTF; Irvine Scientific) containing 10% SSS and covered with mineral oil for up to 2 h at 37°C under 5% CO2, 5%O2 and 90% N2 before ICSI.
Preparation of mature spermatozoa of mouse and human
Preparation of mouse epididymal spermatozoa
The cauda epididymis was isolated from a mature 8–10-week-old male B6D2F1 mouse. Mature spermatozoa were obtained by puncturing the epididymal tubes and incubating them in 3 ml of mHTF for 10–15 min at 25°C, to allow the spermatozoa to disperse evenly in the medium. Next, this suspension was lightly sonicated for 5 s at 5 W power output using an ultrasonic sonicator (Microson; Misonic Inc.) and washed using HTF medium by centrifugation for 5 min at 1000 g. With this treatment, >95% of spermatozoa were immobilized and decapitated. All isolated sperm heads were diagnosed as ‘dead’ after assessment with a sperm viability kit (Live/Dead Fertilight Sperm Viability Kit; Molecular Probes Inc.) (Kuretake et al., Reference Kuretake, Kimura, Hoshi and Yanagimachi1996). As a control, the sonicated (dead) spermatozoa were injected immediately into mouse oocytes. For the experiments, the sonicated spermatozoa were kept in HTF medium for up to 72 h at 37°C under 5% CO2, 5% O2 and 90% N2 before injection. After varying maintenance intervals (1.5–72 h), the sonicated/incubated spermatozoa were injected into mouse oocytes to examine their oocyte activating and Ca2+ oscillation-inducing abilities and the developmental capacities of those injected oocytes.
Preparation of human spermatozoa
Human spermatozoa were obtained from volunteers with proven fertility. Semen samples were allowed to liquefy for 30 min at room temperature and spermatozoa were collected by the swim-up method using HTF medium. This suspension was then lightly sonicated in the same manner as described above. This sonicated sperm were kept in HTF medium for up to 30 h at 37°C under 5% CO2, 5% O2 and 90% N2 before injection (Fig. 1a).
Human sonicated sperm injection into mouse oocytes using a piezo micromanipulator. Sonicated and decapitated human spermatozoa are suspended in medium (a). A sonicated human spermatozoon was sucked into an injection pipette attached to a piezo-electric driving unit (b). A mature unfertilized mouse oocyte was stabilized using a holding pipette and its zona pellucida was penetrated by applying several piezo pulses (c, d). When the needle had advanced deep enough into the ooplasm, the oolemma was punctured with a single piezo pulse and the spermatozoon was slowly released into the ooplasm (e), before the pipette was gently removed (f). Arrows indicate the positions of spermatozoa.
Microinjection of mouse and human sonicated (dead) spermatozoa into mouse oocytes
Mouse and human sperm injections were preformed using a micromanipulator with piezo-electric elements (Kimura & Yanagimachi, Reference Kasai, Hoshi and Yanagimachi1995; Yanagimachi, Reference Yanagimachi1998). The method of intracytoplasmic sperm injection (ICSI) has been described previously (Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000, Reference Yazawa, Yanagida and Sato2001). Briefly, a single spermatozoon was sucked into an injection pipette (5–7 μm inner diameter at the tip) that was attached to a piezo-electric driving unit (model PMM-MB-A; Prime Tech Ltd) (Fig. 1b). A mature unfertilized mouse oocyte was stabilized using a holding pipette and its zona pellucida was penetrated by applying several piezo pulses (Fig. 1c, d). When the needle had advanced deep enough into the ooplasm, the oolemma was punctured with a single piezo pulse and the spermatozoa were slowly released into the ooplasm, before the pipette was gently removed (Fig. 1e, f). All ICSI procedures were performed in 3 μl of mHTF on a stage cooled to 17–18°C (Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995a; Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000). After injection, oocytes were washed three times in HTF and incubated under 5% CO2, 5% O2 and 90% N2 at 37°C.
Examination of oocyte activation and embryo development
After 4–5 h incubation, the sonicated sperm-injected oocytes were placed on a slide coverslip and fixed and stained with acetocarmine to examine the chromatin configurations of the sperm and oocytes chromosomes (Yanagida et al., Reference Yanagida, Yanagimachi, Perreault and Kleinfeld1991) (Fig. 2). Oocytes with a second polar body and two pronuclei were considered to be ‘normally activated’. In other experiments, the sonicated mouse sperm injected oocytes were cultured continuously in THF (up to 96 h) to examine embryo development to the morula and blastocyst stages.
The sonicated sperm-injected oocytes were fixed and stained to examine their oocyte activating ability. Normally activated oocyte with two pronuclei (PN) (a) and an unactivated oocyte with decondensed sperm head (white arrow) and MII plate of an oocyte (black arrow) (b).
Measurement of [Ca2+]i of sperm injected oocytes
The Ca2+ responses of sonicated sperm injected oocytes were investigated using a Ca2+-imaging method and a confocal laser scanning microscope system (Bio-Rad MRC-600, Nippon Bio-Rad Laboratories). Before injection, oocytes were loaded with the Ca2+-sensitive fluorescent dye fluo-3-acetoxymethyl ester (Fluo-3/AM; Molecular Probes Inc.) in dimethyl sulfoxide (final concentration 44 μmol/ml in HTF) with 0.02% Pluronic F-127 for 30 min at 37°C. The loaded and injected oocytes were placed in a droplet (approx. 3 μl) of mHTF and covered with mineral oil on the chambered coverglass (Lab-Tec, Nunc Inc.). The dish was mounted on the stage of a phase-contrast inverted microscope equipped with a confocal laser scanning microscope system and Ca2+ changes were sampled at 20 s intervals for approx. 60 min.
We previously classified the Ca2+ response patterns of sperm-injected oocytes into four groups (Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000). In this study, the classification was modified slightly (Fig. 3). A normal oscillation pattern consisted of repetitive spike-shaped [Ca2+]i rises. A transient pattern had only a few transient rises [Ca2+]i (1–4/60 min). The no-response pattern did not have any [Ca2+]i rises during the observation period. Atypical patterns had irregular increases in [Ca2+]i that did not fit into any of the above three patterns.
Intracellular calcium ([Ca2+]i) patterns of sperm-injected oocytes. The normal oscillation pattern (a) consisted of regular repetitive spike-shaped peaks in [Ca2+]i at intervals of 2–10 min. The transient pattern (b) was composed of several (1–4) transient peaks in [Ca2+]i. The no-response pattern (c) lacked any [Ca2+]i peaks. The atypical patterns had irregular rises in [Ca2+]i that did not fit into any of the above three patterns.
Results
Oocyte activation and [Ca2+]i responses following the injection of mouse sonicated spermatozoa
Tables 1 and 2 summarize the activation rates and [Ca2+]i responses of mouse oocytes injected with sonicated (dead) mouse spermatozoa, which had various maintenance intervals between sonication and injection. More than 90% of sonicated mouse sperm-injected oocytes were normally activated (2PN) with an approx. 10 h maintenance interval between sonication and ICSI. The rate of normal activations gradually declined when the maintenance interval was extended to longer than 20 h. When the maintenance interval was prolonged to 72 h, the rate of normal activations became extremely poor (14%).
The oocyte activation-inducing ability of mouse sonicated (killed) spermatozoa with increasing time after sonication and storage in THF medium
aOocyte as percentage of injected oocytes.
bOocytes as percentage of surviving oocytes.
The Ca oscillation-inducing ability of mouse sonicated (killed) spermatozoa with increasing time after sonication and storage in THF medium
The [Ca2+]i responses of the sperm-injected oocytes were more sensitive to incubation period than activation rate. The rates of sonicated mouse sperm-injected oocytes with normal oscillation patterns remained relatively high with maintenance intervals up to 1.5 h, but the rates of injected oocytes with normal oscillation patterns began to decline with intervals longer than 3 h.
Embryo development following injection of mouse sonicated spermatozoa
When mouse spermatozoa were injected into oocytes immediately after sonication and cultured continuously for up to 96 h, 46% of the fertilized oocytes developed into blastocysts. The rate of blastocyst development was 3–35% when the maintenance interval between sonication and ICSI was between 3 to 5 h. However, the rates declined substantially when the maintenance interval was prolonged to longer than 10 h (Table 3).
Embryo development of mouse eggs injected with mouse sonicated spermatozoa with increasing time after sonication and storage in THF medium
aOocytes as percentage of injected oocytes.
bOocytes as percentage of surviving oocytes.
cOocytes as percentage of fertilized oocytes.
Injected oocytes were cultured in THF medium for 96 h and evaluated for their stage of development.
Injection of human sonicated sperm into mouse oocytes
Tables 4 and 5 summarize the activation and [Ca2+]i responses of mouse oocytes injected with sonicated (dead) human spermatozoa after various maintenance intervals between sonication and injection. Ninety-seven per cent of the injected oocytes were normally activated (2PN) when the human spermatozoa were injected immediately after sonication. However, the rate of normal activations gradually declined when the maintenance intervals were prolonged to over 1.5 h. Moreover, no oocytes were activated when the maintenance interval was longer than 30 h.
The oocyte activation-inducing ability of human sonicated (killed) spermatozoa with increasing time after sonication and storage in THF medium
aOocyte as percentage of injected oocytes.
bOocytes as percentage of surviving oocytes.
The Ca-oscillation-inducing ability of human sonicated (killed) spermatozoa with increasing time after sonication and storage in THF medium
Ninety per cent of the injected oocytes had normal [Ca2+]i oscillation patterns when injected with human spermatozoa immediately after sonication. However, when the maintenance interval was increased to 3 h, the rate of oocytes with normal oscillation pattern declined to 27%. No oocytes exhibited a normal [Ca2+]i oscillation pattern when the maintenance interval was prolonged to 20 h.
Discussion
In previous reports, we had examined whether the spermatids from some experimental rodents and humans possessed oocyte-activating and Ca2+ oscillation-inducing abilities by using a mouse oocyte activation assay method (Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000, Reference Yazawa, Yanagida and Sato2001, Reference Yazawa, Yanagida and Sato2007). These abilities of ROSs and ELSs differed between species. For example, hamster, rabbit and human ROSs could activate mouse oocytes, while mouse and rat ROSs could not. ELS from all of the above species could activate mouse oocytes, but normal [Ca2+]i oscillation was induced only by injection of hamster, rabbit and human ELSs. Additionally, we confirmed that the timing of the acquisition oocyte activation-inducing and Ca2+ oscillation-inducing abilities in spermatogenic cells was dissociated. For example, mouse ELSs could activate oocytes but could not induce Ca2+ oscillation. Similar results were obtained from hamster and rabbit ROSs. Based on these findings, we estimated that the sperm-borne oocyte-activating and Ca2+ oscillation-inducing factor (SOA-COIF, which is presumably identical to the so-called SOAF or sperm factor) appears (or becomes active) gradually when the spermatogenic cells mature. When these factors (SOAF or sperm factor) become fully active, injected oocytes can be activated and exhibit Ca2+ oscillation, while no activation occurs when spermatogenic cells that lack SOA-COIF are injected. When the SOA-COIF becomes intermediately active during spermiogenesis, the injected oocytes are activated without normal Ca2+ oscillation (only several transient [Ca2+]i rises are observed).
In our experiments, we analysed the activation rates and [Ca2+]i responses of mouse oocytes that were injected with sonicated mature spermatozoa to evaluate the biological SOA-COIF activities of dead spermatozoa, after various lengths of time between their death (by sonication) and injection into oocytes.
In the ICSI procedures, it is well known that sperm immobilization (killing) prior to injection is very important for obtaining successful results. Immobilization is commonly achieved by scoring the sperm tail with an injection pipette on the bottom of the dish, applying a Piezo pulse to sperm tail, or by sonication. These procedures disintegrate the sperm plasma membrane and facilitate the immediate mixing of the intracellular components of the sperm (such as the nucleus and the sperm factor) with the oocyte cytoplasm, leading to effective oocyte activation and embryo development (Yanagimachi, Reference Yanagimachi2005). The more extensively the sperm plasma membrane is damaged, the faster oocyte activation is induced (Kasai et al., Reference Kasai, Hoshi and Yanagimachi1999). In clinical ICSI treatments, immotile spermatozoa must be used in the cases of severe asthenozoospermia or necrozoospermia (Dozortsev et al., Reference Dozortsev, Rybouchikin, De Sutter, Qian and Dhout1995; Nagy et al., Reference Nagy, Liu, Joris, Verheyeu, Tournaye, Camus, Derde, Devroey and Van Stierteghem1995). Most of the ejaculated immotile spermatozoa were found to be dead (Eliasson, Reference Elliasson, Mossberg, Camnar and Afzelius1977), but it is unclear when the spermatozoa died, i.e. before or after ejaculation. Moreover, pregnancies have not been reported using dead ejaculated spermatozoa. Another important part of ICSI procedure is the timing of the immobilization (Yanagimachi, Reference Yanagimachi2005). Once the sperm plasma membrane is disrupted, irreversible damage to the sperm nucleus occurs within a certain interval. Disintegrations of the chromosomes of the dead spermatozoa (Rybouchkin et al., Reference Rybouchkin, Benijts, Sutter and Dhont1997) and the spermatozoa killed by sonication and storage (Tateno et al., Reference Tateno, Kimura and Yanagimachi2000) was demonstrated by their injection into mouse oocytes. This may be due to the influx of Na+-rich extracellular medium, which is extremely different from the intracellular environment, because the incidence of structural chromosomal aberrations was decreased (delayed) when sonicated spermatozoa were stored in K+-rich NIM medium prior to injection (Tateno et al., Reference Tateno, Kimura and Yanagimachi2000). Additionally, improved embryo development was seen when dead (killed) sperm or spermatids were stored in NIM medium for certain periods prior to injection (Suzuki et al., Reference Suzuki, Yanagida and Yanagimachi1998; Mizuno et al., Reference Mizuno, Hoshi and Huang2002).
Figure 4 shows the activation rates and [Ca2+]i responses of mouse oocytes injected with mouse sonicated mature spermatozoa and mouse immature spermatogenic cells (ROSs or ELSs), examined in this study and in our previous studies (Yazawa et al., Reference Yazawa, Yanagida, Katayose, Hayashi and Sato2000, Reference Yazawa, Yanagida and Sato2001, Reference Yazawa, Yanagida and Sato2007). We confirmed in this study that the SOAF (sperm factor) activities of dead spermatozoa gradually disappeared, in accordance with the maintenance interval between sonication and injection. As the incubation time after sonication was increased, normal Ca2+ oscillations disappeared first (in the earliest interval assayed, approx. 3 h) and then blastocyst development (approx. 10 h) was gradually obstructed at the earlier interval assayed. In contrast, the oocyte-activating ability was maintained for a relatively long period, but declined after the spermatozoa were incubated for 20 h. Interestingly, when the incubation period for the killed spermatozoa was prolonged from 3 to 10 h, most of the injected oocytes were normally activated (>90%) without normal Ca2+ oscillations. This same phenomenon was observed in the maturation process of immature spermatogenic cells (in the case of mouse ELSs) and occurs when the SOAF activity is sufficient for the induction of oocyte activation, but is insufficient to induce Ca2+ oscillations. The transient Ca2+ response patterns were least affected by increasing the incubation time intervals after sonication, as 78% of the oocytes that were injected with 72 h-incubated sonicated spermatozoa had transient responses, even though their activation rate was decreased to 14%. These results indicated that Ca2+ oscillation was the most sensitive indicator for evaluating the SOAF activities of injected spermatozoa or spermatogenic cells. Moreover, although ELS-injected oocytes had only transient Ca2+ response patterns with high rates of activation, they maintained a reasonable rate of blastocyst development when compared with controls. However, oocytes injected with 10–20 h-incubated sonicated spermatozoa (that had the same pattern of Ca2+ response and activation rate) had disappointing rates of blastocyst development.
The activation rates and Ca2+ responses of mouse oocytes injected with mouse immature spermatogenic cell (ROS, ELS) and mouse sonicated spermatozoa.
Figure 5 shows the activation rates and [Ca2+]i responses of mouse oocytes injected with human sonicated mature spermatozoa and human immature spermatogenic cells (ROSs or ELSs). Compared with mouse sperm or spermatogenic cells, SOAF activity began to appear in the more immature stages of the spermatogenic cells, but began to disappear after shorter incubation intervals prior to sperm injection
The activation rates and Ca2+ responses of mouse oocytes injected with human immature spermatogenic cell (ROS, ELS) and human sonicated spermatozoa.
In the mouse, although ROSs could not activate oocytes, normal offspring were obtained by ROSI or by using electrofusion, with the induction of oocyte activation by an electric pulse (Ogura et al., Reference Ogura, Matsuda and Yanagimachi1994; Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995b). Mouse ELS-injected oocytes, which did not exhibit normal [Ca2+]i oscillation patterns, also produced normal offspring (Yazawa et al., Reference Yazawa, Yanagida and Sato2001). Despite the fact that immature spermatogenic cells lack SOAF activity, injection and supplementation of SOAF activities (with electric pulse or Ca2+ ionophore) can produce normal development to term, because these cells possess genetically normal nuclei, similar to that for mature spermatozoa. On the other hand, irreversible damage to the sperm nucleus occurs during prolonged incubation of dead spermatozoa and supplementation of SOAF activities by any method may not be enough to rescue their function.
Recently, a sperm-specific zeta isoform of phospholipase C (PLCζ) in mouse and human has been identified and demonstrated as a powerful candidate to be a sperm factor (Cox et al., Reference Cox, Larman, Saunders, Hashimoto, Swann and Lai2002; Saunders et al., Reference Saunders, Larman, Parrington, Cox, Royse, Blayney, Swamm and Lai2002). Microinjection of PLCζ cRNA triggered Ca2+ oscillation in mouse and human oocytes, similar to that observed during fertilization, and evoked embryo development to blastocysts (Saunders et al., Reference Saunders, Larman, Parrington, Cox, Royse, Blayney, Swamm and Lai2002; Roger et al., Reference Rogers, Hobson, Pickering, Lai, Braude and Swann2004). Similarly, mouse oocytes injected with anti-PLCζ antibody-treated sperm extract did not exhibit Ca2+ responses. The activity of PLCζ is conserved across mammalian and non-mammalian vertebrates (Cox et al., Reference Cox, Larman, Saunders, Hashimoto, Swann and Lai2002; Coward et al., Reference Coward, Ponting, Chang, Hibbitt, Savolainen, Jones and Parrington2005). In spermatozoa, PLCζ is located mainly at perinuclear matrix of post-acrosome lesion, where the acrosome-reacted spermatozoa first fuse with the oocyte plasma membrane and enter the oocyte at fertilization (Fujimoto et al., Reference Fujimoto, Yoshida, Fukui, Amanai, Isobe, Itagaki, Izumi and Perry2004). In the activated egg, exogenously expressed PLCζ is distributed throughout the cytoplasm before pronucleus formation and then accumulates in pronuclei after their formation (Larman et al., Reference Larman, Saunders, Carroll, Lai and Swann2004; Yoda et al., Reference Yoda, Oda, Shikano, Kouchi, Awaji, Shirakawa, Kinoshita and Miyazaki2004; Sone et al., Reference Sone, Ito, Shirakawa, Shikano, Takaeuchi, Kinoshita and Miyazaki2005; Oda, Reference Oda2006) and sperm factor is distributed in a similar manner (Kono et al., Reference Kono, Jones, Swann and Whittingham1995; Ogonuki et al., Reference Ogonuki, Sankai, Yagami, Shikano, Oda, Miyazaki and Ogura2001). Together, these data indicate that PLCζ is the sperm factor that triggers egg activation and Ca2+ oscillations during fertilization. Recently, Yu et al., clearly demonstrated the relationship between injected PLCζ levels, the pattern of Ca2+ oscillation and embryo development (Yu et al., Reference Yu, Saunders, Lai and Swann2007). The more PLCζ protein expressed, the earlier the first Ca2+ oscillation occurred and the higher the frequency of transient Ca2+ spikes. However, higher levels of PLCζ resulted in the earlier cessation of the Ca2+ responses (shorter duration of Ca2+ oscillation), arrest of most embryos around the 2-cell stage and failure to develop to blastocysts. This confirmed that, despite the fact that oocyte activation and Ca2+ oscillations occurs over a wide range of PLCζ levels, successful development of an embryo into blastocyst could only be achieved by a relatively narrow and appropriate level of PLCζ expression. Additionally, low levels of PLCζ induced high rates of oocyte activation (i.e. induced relatively low frequent Ca2+ oscillations), but this did not lead to effective development into blastocysts.
Once the sperm plasma membrane is disrupted with immobilization procedure, irreversible damage of SOAF activity and the sperm nucleus occurs during a certain interval. In these experiments, we confirmed that the normal oscillation pattern disappeared within 3 h of incubation and the Ca2+ oscillation is the most sensitive indicator of the SOAF activity of dead spermatozoa. We also reconfirmed that using live spermatozoa and immobilization immediately prior to injection is important for obtaining successful results from the ICSI procedure.
