- Split View
-
Views
-
Cite
Cite
Sho Nakamura, Yoshihiro Wakabayashi, Takashi Yamamura, Satoshi Ohkura, Shuichi Matsuyama, A neurokinin 3 receptor-selective agonist accelerates pulsatile luteinizing hormone secretion in lactating cattle, Biology of Reproduction, Volume 97, Issue 1, July 2017, Pages 81–90, https://doi.org/10.1093/biolre/iox068
- Share Icon Share
Abstract
Pulsatile gonadotropin-releasing hormone (GnRH) secretion, which is indispensable for follicular development, is suppressed in lactating dairy and beef cattle. Neurokinin B (NKB) neurons in the arcuate nucleus of the hypothalamus are considered to play an essential role in generating the pulsatile mode of GnRH/luteinizing hormone (LH) secretion. The present study aimed to clarify the role of NKB-neurokinin 3 receptor (NK3R) signaling in the pulsatile pattern of GnRH/gonadotropin secretion in postpartum lactating cattle. We examined the effects of the administration of an NK3R-selective agonist, senktide, on gonadotropin secretion in lactating cattle. The lactating cattle, at approximately 7 days postpartum, were intravenously infused with senktide (30 or 300 nmol/min) or vehicle for 24 h. The administration of 30 or 300 nmol/min senktide significantly increased LH pulse frequency compared to in the control group during 0–4 or 20–24 h after infusion, respectively. Moreover, LH and follicle-stimulating hormone levels were gradually increased by 300 nmol/min administration of senktide during the 0–4-h sampling period. Ultrasonography of the ovaries was performed to identify the first postpartum ovulation in senktide-administered lactating cattle. The interval from calving to first postpartum ovulation was significantly shorter in the 300 nmol/min senktide-administered group than in the control group. Taken together, these findings suggest that senktide infusion elicits an increase in LH pulse frequency that may stimulate follicular development and, in turn, induce the first postpartum ovulation in lactating cattle.
Introduction
In cattle, the duration from parturition to first postpartum ovulation is directly associated with reproductive efficiency, while extended calving intervals cause economic losses. Murphy et al. reported that suckled beef cows show low ovulation rate (∼11%) of the first postpartum dominant follicle [1], and another review showed that 20 to 35% of these cows ovulate the first dominant follicle [2]. A resumption of the ovarian cycle is required to advance fertility in postpartum cows. Ovarian cyclicity is strongly dependent on the pulsatile pattern of gonadotropin-releasing hormone (GnRH) secretion followed by luteinizing hormone (LH) pulses. Luteinizing hormone pulse frequency decides the fate of the dominant follicle in the first postpartum follicular wave as in the three types described by Beam and Butler [3]: (1) ovulation, (2) failure of ovulation followed by regression, (3) nonovulation followed by a cystic follicle. Insufficient LH pulses due to the suckling stimulus and negative energy balance cause anovulation of the dominant follicle in cows in the early postpartum period [3] and prolongation of calving intervals.
In mammals, the reproductive function is controlled by the hypothalamic-pituitary-gonadal axis. It is well accepted that pulsatile LH secretion from the pituitary gland corresponds to pulsatile GnRH release to the hypophyseal portal system. Various studies have been performed to identify the GnRH pulse generator that is responsible for the episodic pattern of GnRH secretion [4]. KNDy neurons, which produce kisspeptin, neurokinin B (NKB) and dynorphin A, are considered to be fundamental elements of the GnRH pulse generator [4, 5]. The presence of KNDy neurons of the arcuate nucleus (ARC) has been reported in various species, such as mice [6], rats [7, 8], goats [9], sheep [10], and cattle [11], within the mediobasal hypothalamus, where the GnRH pulse generator has been postulated to be located by hypothalamic deafferentation [12] and electrophysiological studies [13–15]. Ezzat et al. provided that KNDy neurons expressing c-Fos were observed at the time when LH pulse appears in the ewe [16]. Recent studies recording multiple-unit activities (MUA) from electrodes placed at the vicinity of KNDy neurons demonstrated that the periodic bursts of MUA (MUA volleys) were accompanied by LH secretion in goats [9, 17]. Central or intravenous administration of NKB or neurokinin 3 receptor (NK3R)-selective agonist, senktide, facilitates LH secretion synchronized with MUA volleys in ovariectomized goats treated with estradiol [9, 17]. The acceleration of LH secretion by NKB or senktide has also been reported in rats [18], sheep [19], and monkeys [20]. Furthermore, in vitro studies have confirmed that NKB or senktide directly stimulate KNDy neurons [21, 22]. On the other hand, it has been also reported that NKB showed inhibitory effects on pulsatile LH secretion in ovariectomized rats [23] and goats [9]. These studies suggest that the GnRH/LH pulse generation could receive inhibitory/stimulatory modulation of NKB-NK3R signaling in KNDy neurons, which may be dependent on the animal model or the steroid milieu of the animal. Given that kisspeptin is known to be a potent stimulator of GnRH/LH secretion [24] and has no effect on the activity of KNDy neurons [22, 25], it is suggested that NKB is involved in the process of generating the pulsatile discharge of kisspeptin from KNDy neurons followed by episodic GnRH/LH secretion.
During the lactation period, LH pulse frequency is suppressed in mammals [26–28]. It is clear that the interval to estrus onset is longer in lactating cows than in nonlactating postpartum cows and that suckling prolongs this interval [29]. In addition, both the mean LH concentration and LH pulse frequency are kept lower than that in nonlactating postpartum cows [29]. Calf removal showed the resumption of gonadotropin secretion and hence advanced follicular development in postpartum cows [30]. Similarly in rats, the removal of the pups restores the LH pulses that are strongly suppressed by the suckling stimulus [28]. Accordingly, it is accepted that the suckling stimulus is one of the major inhibitory factors of pulsatile LH secretion. Furthermore, both gene expression and immunoreactivity of NKB were suppressed in the ARC of lactating rats [7, 31]. These findings present the possibility that the lowered frequency of pulsatile LH secretion would be due to the lack of NKB-NK3R signaling during postpartum period. The role of NKB-NK3R signaling in gonadotropin secretion under the steroidal [9, 17, 18], seasonal [19], and prepubertal [20, 32] milieu has been described, whereas the effect of NKB-NK3R signaling on gonadotropin secretion in postpartum lactating animals remains unclear. The objective of the present study is to examine whether (1) senktide treatment accelerates pulsatile gonadotropin secretion in postpartum early period when the first postpartum follicular wave initiates, and (2) the restored gonadotropin secretion contributes to the follicular development followed by ovulation.
Materials and methods
Animals and experimental procedure
Lactating Japanese Black or crossbred Japanese Black × Holstein cattle were used (n = 15). The information on age, body weight, body condition score, and parity of experimental animals is shown in Table 1. All animals were housed under natural conditions at the Institute of Livestock and Grassland Science, National Agriculture, and Food Research Organization (NARO) and provided with grass silage and water ad libitum. All procedures involved in the animal experiments were approved by the Committee of the Care and Use of Experimental Animals of Institute of Livestock and Grassland Science, NARO, Japan.
Treatment . | No. of animals . | Age of month . | Body weight (kg) . | BCS . | Parity . |
---|---|---|---|---|---|
Vehicle | 5 | 67.4 ± 21.8 | 578.6 ± 44.9 | 5.3 ± 0.2 | 2.8 ± 1.1 |
Senktide 30 nmol/min | 5 | 51.0 ± 8.0 | 531.0 ± 54.9 | 5.6 ± 0.1 | 2.2 ± 0.5 |
Senktide 300 nmol/min | 5 | 71.1 ± 26.2 | 566.8 ± 65.0 | 5.8 ± 0.2 | 2.6 ± 1.0 |
Treatment . | No. of animals . | Age of month . | Body weight (kg) . | BCS . | Parity . |
---|---|---|---|---|---|
Vehicle | 5 | 67.4 ± 21.8 | 578.6 ± 44.9 | 5.3 ± 0.2 | 2.8 ± 1.1 |
Senktide 30 nmol/min | 5 | 51.0 ± 8.0 | 531.0 ± 54.9 | 5.6 ± 0.1 | 2.2 ± 0.5 |
Senktide 300 nmol/min | 5 | 71.1 ± 26.2 | 566.8 ± 65.0 | 5.8 ± 0.2 | 2.6 ± 1.0 |
Data are mean ± SEM. BCS, body condition score; scale of 1 (thin) to 9 (obese).
Treatment . | No. of animals . | Age of month . | Body weight (kg) . | BCS . | Parity . |
---|---|---|---|---|---|
Vehicle | 5 | 67.4 ± 21.8 | 578.6 ± 44.9 | 5.3 ± 0.2 | 2.8 ± 1.1 |
Senktide 30 nmol/min | 5 | 51.0 ± 8.0 | 531.0 ± 54.9 | 5.6 ± 0.1 | 2.2 ± 0.5 |
Senktide 300 nmol/min | 5 | 71.1 ± 26.2 | 566.8 ± 65.0 | 5.8 ± 0.2 | 2.6 ± 1.0 |
Treatment . | No. of animals . | Age of month . | Body weight (kg) . | BCS . | Parity . |
---|---|---|---|---|---|
Vehicle | 5 | 67.4 ± 21.8 | 578.6 ± 44.9 | 5.3 ± 0.2 | 2.8 ± 1.1 |
Senktide 30 nmol/min | 5 | 51.0 ± 8.0 | 531.0 ± 54.9 | 5.6 ± 0.1 | 2.2 ± 0.5 |
Senktide 300 nmol/min | 5 | 71.1 ± 26.2 | 566.8 ± 65.0 | 5.8 ± 0.2 | 2.6 ± 1.0 |
Data are mean ± SEM. BCS, body condition score; scale of 1 (thin) to 9 (obese).
Intravenous administration of neurokinin 3 receptor-selective agonist and blood sampling
The day of parturition was defined as day 0. Serial blood sampling was performed at any one of the following: day 6 to 7, 7 to 8, or 8 to 9. Sixteen-gauge catheters (UA-1627-S, NIPRO, Osaka, Japan) were bilaterally inserted 1 day prior to the onset of blood sampling into the jugular veins for serial blood sampling and senktide administration. Blood samples were collected every 10 min for 8 h (0800 to 1600 h) in two consecutive days. Senktide (WuXi AppTec, Shanghai, China) was dissolved in 0.04 N sodium bicarbonate to prepare the concentration of 0.6 or 6 mM and intravenously infused using a syringe pump (FP-1000, MELQUEST, Toyama, Japan) with a flow rate of 50 μl/min (30 or 300 nmol/min) for 24 h from 1200 h at the first day. The control group was administered vehicle (0.04 N sodium bicarbonate) at the same flow rate. Heparinized blood samples were centrifuged and, then, plasma samples were stored at –30˚C until assayed for LH and follicle-stimulating hormone (FSH).
Ultrasonography of ovaries
Ultrasonography of the ovaries and one-point blood collection were performed every 2 to 3 days after serial blood sampling until detection of the first postpartum ovulation in lactating cattle. The ovaries were monitored by transrectal ultrasonography using a transrectal linear probe (S6V, SonoScape, Shenzhen, China). The blood samples were stored at –30˚C until assayed for progesterone. Ovulation was confirmed by both the appearance of a corpus luteum and plasma progesterone levels above 1 ng/ml, as has been previously defined [29, 33].
Assays
Plasma LH and FSH concentrations in 100 μl plasma samples were measured by a double-antibody radioimmunoassay (RIA) with a bovine LH and FSH RIA kit provided by the National Hormone and Peptide Program (Baltimore, MD), respectively, as previously described [34]. The least detectable level of LH assay was 0.10 ng/ml, and the intra- and interassay of coefficients of variation were 6.9% at 0.30 ng/ml and 6.2% at 0.31 ng/ml, respectively. The least detectable level of FSH assay was 0.20 ng/ml, and the intra- and interassay of coefficients of variation were 6.8% at 2.47 ng/ml and 1.2% at 2.45 ng/ml, respectively.
Plasma progesterone concentrations after parturition were measured by a double-antibody enzyme immunoassay as previously described [35]. Assay sensitivity was 0.10 ng/ml for 100 μl plasma samples. The intra- and interassay coefficients of variation were 2.8% at 2.68 ng/ml and 8.6% at 2.91 ng/ml, respectively.
Data analysis
Luteinizing hormone and FSH pulses were identified by the PULSAR computer program [36]. Mean LH and FSH pulse frequency, pulse amplitude, and concentrations were calculated for each individual every 4 h during the sampling period (–4 to 0, 0 to 4, 20 to 24, and 24 to 28 h after the initiation of infusion). Statistical differences (P < 0.05) in LH and FSH pulse frequency, pulse amplitude, and concentrations in each sampling period between the senktide-administered (30 or 300 nmol/min) and control groups were analyzed by one-way ANOVA followed by Tukey post hoc tests. Statistical differences (P < 0.05) in days for first postpartum ovulation between the groups were determined by one-way ANOVA followed by Tukey post hoc tests.
Results
Effects of intravenous administration of senktide on gonadotropin secretion
Luteinizing hormone secretion was stimulated by both 30 and 300 nmol/min of senktide administration (Figure 1). The frequency of LH pulses was significantly higher in the 30 nmol/min senktide-administered group (2.2 ± 0.2 pulses/4 h) than in the vehicle-administered control group (0.8 ± 0.2 pulses/4 h) during 0 to 4 h after initiation of senktide infusion (Figure 2A). During 20 to 24 h after the infusion, senktide administration resulted in a dose-dependent increase in LH pulse frequency. The group treated with 300 nmol/min senktide (2.8 ± 0.7 pulses/4 h) showed significantly higher pulse frequency compared to the control group (0.8 ± 0.4 pulses/4 h) (Figure 2A). The frequency in the 30 nmol/min senktide-administered group was higher (2.0 ± 0.3 pulses/4 h) than that in the control group, although the difference did not reach statistical significance (Figure 2A). The LH pulse amplitude in the 300 nmol/min senktide-administered group (0.19 ± 0.02 ng/ml) was significantly lower than that in the control group (0.34 ± 0.04 ng/ml) during the 20 to 24 h sampling period (Figure 2B). The dose of 300 nmol/min senktide infusion gradually augmented LH secretion during 0 to 4 h after the initiation of infusion (Figure 1), resulting in mean LH concentrations in the 300 nmol/min senktide-administered group (0.85 ± 0.10 ng/ml) that were significantly higher than those in the control group (0.26 ± 0.03 ng/ml) and the 30 nmol/min senktide-administered group (0.39 ± 0.04 ng/ml, Figure 2C). There was no significant difference in the LH pulse frequency, amplitude, and mean LH concentrations between the groups before and after senktide infusion (Figure 2).
Plasma FSH secretion was also stimulated by senktide infusion (Figure 3), resulting in mean FSH concentrations in the 300 nmol/min senktide-treated group (3.46 ± 0.17 ng/ml) that were significantly higher than those in the control group (2.49 ± 0.25 ng/ml) and the 30 nmol/min senktide-treated group (2.74 ± 0.13 ng/ml) within 0 to 4 h after administration (Figure 4C). The FSH pulse frequency in the 30 nmol/min senktide-administered group was higher (1.6 ± 0.2 pulses/4 h) than that in the control group (0.8 ± 0.4 pulses/4 h) within 0 to 4 h after infusion (Figure 4A), but no significant difference was found between the groups. The FSH pulse amplitude was not significantly different between the groups in any sampling period (Figure 4B).
Effects of intravenous administration of senktide on first postpartum ovulation
A corpus luteum was identified in the ovary by ultrasonography at 27–76 days after parturition in the control group, whereas 30 or 300 nmol/min senktide administration induced the appearance of the corpus luteum in the ovary at 25–36 or 15–26 postpartum days, respectively (Figure 5A). The mean interval from calving to luteinization was significantly shorter in the 300 nmol/min senktide-administered group (20 ± 2.4 days postpartum) than in the vehicle-administered control group (45 ± 11 days postpartum, Figure 5B). The dose of 30 nmol/min senktide infusion showed an intermediate interval (29 ± 2.2 days postpartum), namely shorter than that of the control group and longer than that of the high-dose group, although no significant difference was found. The luteinization was confirmed by plasma progesterone levels above 1 ng/ml (Supplemental Figure S1).
Discussion
The present study demonstrates that the administration of senktide, an NK3R-selective agonist, stimulated gonadotropin secretion, and induced significantly earlier first postpartum ovulation in lactating cattle. In general, a transient elevation of FSH, which is shown within a week after parturition due to the lack of the negative feedback effects of progesterone and estradiol [37], induces the first follicular wave within 10–15 days in postpartum cows [38]. The fate of the dominant follicles in the first postpartum follicular wave is dependent on the LH pulse frequency. The dominant follicle receiving high LH pulse frequencies (3.4–4.5 pulses/6 h) develops into an ovulatory dominant follicle and the dominant follicle receiving low LH pulse frequencies (1.7–2.2 pulses/6 h) induces a regression of the wave in postpartum cows [39]. This study showed that senktide infusion increased LH pulse frequency, at an approximately 2-fold level compared to the control group. Thus, it is suggested that an increase in LH pulse frequency caused by senktide administration stimulated postpartum follicular development that could reach the first postpartum ovulation in lactating cattle.
The current study showed that senktide induced elevation of LH and FSH secretion and increased LH pulse frequency in lactating cattle. Previously, NKB or senktide elicited an obvious increase in LH secretion in seasonally anestrus ewe [19] and juvenile monkeys [20], whose pulsatile GnRH secretions are suppressed, while LH secretions are inhibited by NKB or senktide in the absence or very low levels of sex steroid condition [6, 9, 23, 40]. Therefore, the present study and previous reports suggest that GnRH/LH secretion was stimulated by the enhancement of NKB-NK3R signaling under the inhibitory influence on LH secretions, such as the suckling stimulus, nonreproductive season, and prepubertal status. The accumulated evidence has also shown that NKB-NK3R signaling is a prerequisite for the regulation of GnRH/LH secretion in mammals. The loss-of-function mutation of the NKB or NK3R gene induced hypogonadotropic hypogonadism in humans [41–43]. The NK3R expression was found in a majority of KNDy neurons in rats [8], mice [6], and sheep [44], whereas few GnRH neurons coexpress NK3R in rats [45] and sheep [44]. The increased LH response to senktide was totally suppressed in Gpr54 deficient mice, which lack kisspeptin receptor [46], suggesting that the main action site for NKB to elicit LH secretion would be upstream of GPR54. These studies suggest that the increase in LH pulse frequency following senktide infusion in postpartum suckled cows would be caused by the stimulation of NKB-NK3R signaling in KNDy neurons which elicit kisspeptin release from KNDy neurons, thereby inducing GnRH and, subsequently, LH secretion. On the other hand, we cannot rule out the possibility that NKB directly acts on ovaries. Increase in NKB gene expression levels was detected during estrous period in rats [47]. NKB induced CYP11A and CYP19A1 gene expression, involved in steroidogenesis, in the human granulosa cells [48]. These data suggest that NKB-NK3R signaling in the ovary may be associated with the regulation of ovarian functions such as estrogen secretion and the following ovulation.
The present study demonstrates that continuous administration of 300 nmol/min senktide initially induced gradual increase in LH secretion, whereas long-term infusion finally elicited high frequency of pulsatile LH secretion in lactating cattle. An escalation of LH secretion by high dose of senktide is consistent with previous studies in which senktide administration induced a gradual increase in LH concentrations in sheep [49] and goats [50]. In goats, NKB administration drastically shortened the interval between MUA volleys, which could reflect the activity of KNDy neurons [9]. Microimplantation of NKB into the ARC decreased LH pulse intervals in ewes [51]. These findings suggest the possibility that the gradual increase in LH secretion by high-dose senktide infusion seems to be caused by the additive LH secretion with indistinguishable frequent pulses. Thus, the preceding pulse being covered by the next pulse might induce gradual elevation of LH secretion after continuous senktide administration. Alternatively, Porter et al. reported that local administration of senktide into either the retrochiasmatic or preoptic area induced nonepisodic LH elevation in ewes [52]. This would provide the other possibility that high-dose senktide infusion might activate NKB-NK3R signaling in those regions, resulting in producing gradual LH elevation during 0 to 4 h after the administration. Furthermore, the present study also demonstrated that long-term senktide infusion (300 nmol/min) elicited pulsatile LH secretion during 20 to 24 h after the administration of senktide. In that period, LH pulse amplitude in the 300 nmol/min senktide-administered group was significantly lower than that in control group, suggesting that a low amplitude per pulse allowed the generation of distinguishable high frequency of pulsatile LH release. Previously, chronic administration of a kisspeptin analog, TAK-448, reduced hypothalamic GnRH protein contents, although the analog initially induced an activation of GnRH neurons and subsequent gonadotropin secretions [53]. It is plausible that the low amplitude of LH pulses may be caused by decreased hypothalamic contents of GnRH, due to frequently releasing GnRH from the hypothalamus in response to continuously administered senktide. On the other hand, Ramaswamy et al. reported that juvenile monkeys treated with continuous infusion of senktide for 48 h showed desensitization of NK3R, because LH secretion was elicited by either kisspeptin or GnRH administration but not induced by an additional senktide dosing during last 4 h of the infusion [54]. This suggests that the low amplitude of LH secretion would be a result of the desensitization of NK3R. Although senktide certainly activates the pulsatile GnRH secretions, further studies are required to clarify what causes the various LH responses to senktide, by investigating the status of GnRH and/or KNDy neurons.
The effect of senktide on FSH secretion is still unclear. Although this study showed that FSH pulse frequencies seem to be stimulated by the administration of 30 nmol/min senktide, the effect of senktide on pulsatile FSH secretions were unremarkable compared to the effect on LH pulse frequencies. In the previous study, pulsatile FSH secretion would be affected by some other factors other than pulsatile GnRH input, because FSH secretion pattern was nearly random and most FSH pulses (87%) does not synchronize LH release [55]. Accordingly, inhibin, an FSH-specific inhibitor secreted from ovarian follicle, or FSHβ-subunit sugar chain that is involved in circulating half-life, might cause different responses to senktide administration.
The effects of GnRH or its analog on the first postpartum ovulation have been investigated. A single injection of GnRH at 15 days postpartum [56] or multiple low-dose injections of GnRH at 1 or 2 h intervals for 2–4 days around 30 days postpartum [57] had no effects on ovulation in a majority of postpartum beef cows. Conversely, single administration of a GnRH analog, buserelin, during the growing phase of the first postpartum follicular wave (10–15 days postpartum) resulted in ovulation in all subjects immediately after the administration [38]. The reason for this discrepancy might be attributable to the variation of the dominant follicle status when receiving the effects of GnRH, as these GnRH treatments induce surge-like LH secretion that could stimulate ovulation rather than follicular developments [38]. Exogenous pulsatile injection of GnRH (every 2 h for 2 days or every 1–3 h for 3 days) stimulated follicular estrogen secretion and induced subsequent endogenous LH surge after the administration followed by an ovulation in 3–6 weeks postpartum beef cows [58] and ewes [59]. Pulsatile LH administrations at 1 h apart for 3 days from 10 days postpartum also resulted in the ovulation in suckled cows [60]. Since a long-term infusion of senktide stimulated endogenous pulsatile LH secretions, which is required to enhance the development of ovarian follicles, it is implied that the NK3R agonist targeting KNDy neurons may have potential as a therapeutic drug to improve or advance gonadal function, via its ability to activate the GnRH/LH pulse generation in postpartum cows. Further studies are necessary in order to fully understand the basement mechanism of NKB-NK3R signaling in decreasing the interval from calving to first ovulation. The drug regimen should be also investigated in the future study to make NK3R agonist treatments to be a convenient and practical methods in the field.
In conclusion, the present study demonstrated that the enhancement of NKB-NK3R signaling, via neuronal pathways involving KNDy neurons, facilitates pulsatile LH secretion albeit the activity of the GnRH/LH pulse generator is suppressed by the suckling stimulus during the lactation period. An increase in pulsatile LH secretion by senktide could stimulate dominant follicular development, which may result in the first postpartum ovulation in lactating cattle. Continuous studies are required to investigate the effect of long-term senktide infusion during the early postpartum period on the subsequent conception rate in lactating cattle.
Supplementary data
Supplementary data are available at BIOLRE online.
Supplemental Information contains Supplemental Experimental Procedures, two figures, and one table that are available online.
Supplemental Figure S1. Plasma progesterone profiles of postpartum individual cows receiving the NK3R antagonist, senktide (30 or 300 nmol/min). Horizontal dotted line on the graph indicates the lowest progesterone level to confirm luteinization (1 ng/ml).
Acknowledgments
We are grateful to Dr Naoki Takenouchi (Kyusyu Okinawa Agricultural Research Center, NARO, Kumamoto, Japan) for the antisera against progesterone. We thank Ms Hatsumi Suzuki and the staff (Institute of Livestock and Grassland Science, NARO, Nasushiobara, Japan) for their technical assistance and careful animal care.
References
Author notes
Grant Support: This study was supported by the Research Program on Innovative Technologies for Animal breeding, Reproduction and Vaccine Development (REP2005 to S.O.) from Ministry of Agriculture, Forestry and Fisheries of Japan.