PMSG Pretreatment Before Fixed-Time Articial Insemination Improves the Reproductive Performance in Replacement Gilts

Background: Fixed-time articial insemination (FTAI) uses exogenous reproductive hormones to regulate the sexual cycle of sows and realize the mating of pigs at a xed-time without estrus identication. There is a great difference in litter size following FTAI, which may be due to the ovarian status before treatment. However, the specic underlying mechanism remains unclear. Results: We selected replacement gilts (n=104) and divided them into four groups: CON (n=35), EST (n=16), pregnant mare serum gonadotrophin (PMSG)-15D (n=26), and PMSG-20D (n=27). The PMSG-15D and PMSG-20D groups were pretreated with PMSG for 15 and 20 days, respectively. The four groups were then subjected to the same FTAI treatment. Pretreatment with PMSG causes ovulation in the gilts, to articially promote an estrus cycle. At 42 hours after ALT feeding stopped, The follicular diameter of PMSG-15D group (2.94±0.24 mm) was signicantly higher (P<0.05) than the other groups.(cid:0)and the proportion of gilts with follicular diameter greater than 3 mm in the PMSG-15D group was signicantly higher than that in the CON group (37.50% vs 5.88%). The maximum follicular diameter before ovulation in the PMSG-15D group (6.28±0.23 mm) was signicantly larger (P<0.05) than that of the PMSG-20D group (5.98±0.59 mm) and EST group (5.60±0.47 mm), indicating that the follicular development of the PMSG-15D group was better than that of the other groups. The ovulation time of the PMSG-15D group was concentrated on the 25 th to 26 th day of FTAI. The ovulation rate was as high as 94.5% in the optimal insemination window period and was more suitable for FTAI. Compared with the CON (52.70±12.71 h) and the PMSG-20D (47.88±13.98 h) groups, the GnRH to ovulation interval (42.33±5.87 h) was signicantly shortened in the PMSG-15D group. The level of follicle-stimulating hormone of gilts in the PMSG-15D group was more uniform and signicantly lower than that in the CON group (P<0.01) at 42 h after Altrenogest feeding. The pregnancy rate (80.77%), total litter size (10.44±2.83), and live litter size (10.00±2.61) of the PMSG-15D group were higher than those of the other groups.


Background
With the large-scale development of the pig industry towards batch production, problems, such as high recessive estrus rate, low breeding utilization rate, and inadequate sterilization of piggery, in the traditional breeding modes have become increasingly concerning. The development of new breeding technologies that improve the breeding utilization rate, optimize pig farm management, and reduce labor costs have been the recent focus of research efforts.
Fixed-time arti cial insemination (FTAI) regulates the sexual cycle of sows and controls their estrus and ovulation times, using exogenous reproductive hormones, to realize population mating at a xed-time without estrus identi cation [1] . The estrus cycle is successfully synchronized after administering 15-gonadotrophin (eCG ) alone or in combination with Human chorionic gonadotrophin (hCG) stimulates follicular growth and improves estrus and ovulation rates in sows and gilts [4,5] . Ovulation induction via the gonadorelin (GnRH) analog, buserelin after weaning shortens the interval between weaning and ovulation (WOI) and the interval between estrus and ovulation (EOI) in sows [6] . As a result, FTAI technology effectively improves the estrus and ovulation rates of sows and gilts, shortens the nonproductive days, reduces labor costs, improves production e ciency, and plays an important role in promoting the batch production of the pig industry.
However, the litter size of replacement gilts is signi cantly different when employing FTAI. A metaanalysis of the effects of ALT on the reproductive performance of gilts showed that the litter size and number of live births signi cantly improved [7] . Vangroenweghe found that the estrus rate and birth weight of piglets signi cantly improved after treatment of gilts with ALT and GnRH analogs; however, the litter size was not affected [8] . In a recent study, after discontinuation of ALT treatment, eCG and Porcine luteinizing hormone (pLH) intramuscular injection in gilts resulted in shorter delivery time; however, it resulted in a lower litter size than that of the control group [9] .
The prerequisite for successful ovulation induction is the presence of a dominant group of follicles in ovulating ovaries [10] . Previous studies in our laboratory found that under the same FTAI procedure, the estrus rate and pregnancy rate of gilts at the initial stage were all higher than those without puberty, indicating that the complete stage of estrus can promote the estrus, ovulation, and pregnancy of gilts. Therefore, the ovarian state of gilts may greatly in uence the FTAI effects. However, this process is resource-intensive and time-consuming during practical application, as it is necessary to select or adjust the ovary status of the replacement gilts before the FTAI procedure. The serum gonadotropin of pregnant equine (PMSG) exerts a dual activity of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) [11] . It also has the most effective effect on the follicular development of replacement gilts at a concentration of 800-1000 IU. PMSG treatment may be an effective way to adjust the ovarian status of replacement gilts [12] .
The purpose of this study was to investigate the effect of ovarian state on FTAI. Before feeding the ALT, replacement gilts were injected with 1000 IU PMSG to promote follicular development and ovulation as well as to stimulate the next estrus. After PMSG treatment for varying durations, follicular development varied across groups. We then carried out FTAI and subsequently monitored and analyzed the follicular development, ovulation, and reproductive hormone level changes to lay a foundation for optimizing the FTAI program in replacement gilts.

Animals and management
This study was conducted in Large White × Landrace crossing breed replacement gilts from the Shuangdingshan Animal Farm (Cicheng, Ningbo, China). Healthy gilts aged 190-220 days (n = 53) and 210-240 days (n = 51) were selected. During the experiment, different groups of gilts were divided into different crates for breeding. All gilts adopted the same feeding and management methods. All were fed a regular diet, and the pig farm was equipped with an independent drinking water nipple to provide free drinking water.
The boar semen used in the mating was purchased from a local boar breeding station, and the breed of the boar was Landrace. Before mating, all the semen was examined, and only semen with sperm motility up to 70%, malformation rate less than 10%, and medium density with microscopic examination were selected. Gilts were examined for pregnancy using ultrasound diagnosis on the 25th day after mating, and gilts that could not be determined for pregnancy were examined for the second time 10 days after the rst pregnancy examination.

Experimental design
Our experimental design is depicted in Fig. 1. The replacement gilts of 190-220 days were randomly selected and treated with 1000 IU PMSG for 15 days (n = 26) and 20 days (n = 27), and were divided into two groups, PMSG-15D and PMSG-20D, respectively. FTAI was conducted after 15 or 20 days of PMSG use. Ultrasonic monitoring was conducted at 0-6 days, and blood was collected at 0, 1, 3, and 5 days of PMSG treatment. In addition, replacement gilts aged 210-240 days were randomly divided into two groups, EST (n = 16) group, in which FTAI processing was started for those in estrus and CON (n = 35) group, which was used as a blank control and no processing was done before FTAI programs were performed.
The FTAI procedure is also depicted in Fig. 1. Oral administration of ALT (20 mg per pig) every day for 18 days. 42 h after stopping feeding, the pigs were injected with 1000 IU PMSG, and 80 h later, injected with 100 µg GnRH. Insemination for the rst time was conducted 24 h after GnRH injection and the second insemination was conducted 16 h later. The gilts that still had a static reaction at 24 h after the second time were inseminated one more time. On the 20th day of the FTAI (42 h after ALT was stopped), blood was collected. Follicular development of gilts from the four groups was ultrasonically monitored after one to three and 20 days until the end of ovulation fed with ALT. All the hormones used in this study, including ALT (N190403), PMSG (B190707) and GnRH (C190806) were purchased from Ningbo Sansheng Biological Technology Co., Ltd.

Follicular development and ovulation detection
Transabdominal ultrasonography (HS-1600, Honda Electronics Co. Ltd., Toyohashi, Japan) was performed twice a day (8:00 a.m. and 2:30 p.m.) with a 5 MHz multi-convex transducer for follicular development detection on both gilt ovaries. The diameters of the three to seven largest follicles were measured and averaged. Ovulation was presumed to have occurred when previously present preovulatory sized follicles (6-9 mm) disappeared [13] . Follicles larger than 12 mm in diameter and visible for several consecutive days were de ned as cystic follicles [14] . Data from gilts with multiple cyst follicles were excluded from subsequent analyses.

Blood collection and hormone analysis
After the gilts were secured, the blood samples collected from the anterior jugular vein with a 10 mL syringe were transferred to a 5 mL coagulation tube for 30 min. After the serum was precipitated, the blood samples were centrifuged at 3000 rpm for 4 min, and the serum was absorbed into the centrifuge tube and stored in the refrigerator for 20 min for testing.

Statistical analyses
Data were analyzed by repeated measures analysis of variance (split-plot ANOVA, Gill & Hafs, 1971) using the general linear model with repeated measures. The SPSS 21.0 statistical software was used for one-way ANOVA with different replicates. Duncan's test was used for multiple comparisons and signi cance analysis. The estrus rate and pregnancy rate of gilts were measured using the Chi-square test. Except for χ 2 test indexes, they were all expressed and calculated as mean ± standard deviation (mean ± SD). All results were considered statistically signi cant at P < 0.05.

Follicular development and hormone levels in replacement gilts preconditioned with PMSG
The follicle development trend of the replacement gilts after PMSG treatment is shown in Fig. 2a. After PMSG treatment, two of the 24 gilts monitored by ultrasound were not in estrus, and thus data of the two gilts were not considered. After PMSG treatment, the average follicular diameter (3.23 mm) exceeded 3 mm on the afternoon of the second day, and the average follicular diameter reached a maximum of 4.16 mm on the morning of the fth day. It then began to decrease until the average follicular diameter of 2.89 mm dropped to less than 3 mm on the afternoon of the sixth day. The distribution of ovulation time of gilts treated with PMSG is shown in Fig. 2b. The results revealed that 91% of gilts ovulated on the fth to sixth day after PMSG treatment.
After PMSG pretreatment, the contents of E 2 , LH, and FSH in gilts increased with time and reached the highest level on day 5. The contents of E2 and LH in PMSG pretreatment were signi cantly higher on day 5 than those on day 0, day 1 and day 3(P < 0.05). However, there was no signi cant difference in the contents of FSH (Fig. 3). The content of P 4 did not change signi cantly within 0-5 days after PMSG treatment (P > 0.05), and the concentration uctuated within 150-200 pg/mL (Fig. 3).

Effect on follicular development and hormone levels in FTAI replacement gilts
The average follicular diameter (2.59 mm) of the CON group and the average follicular diameter (2.62 mm) of the PMSG-15D group were all lower than 3 mm when the FTAI started, namely began to feed ALT. The average follicular diameter of the PMSG-20D group was 3.18 mm, which tended to be that of the EST group (3.83 mm). The average follicular diameter of the gilts in the EST group decreased to 2.98 mm in the afternoon of the second day after FTAI, and the follicular diameter of all the replacement gilts was synchronized to the same level on the third day, as shown in Fig. 4.
After PMSG treatment on the afternoon of the 20th day, the follicles of the four groups of gilts began to develop, and there was a signi cant trend of follicular enlargement. The average follicular diameters of the four groups of gilts at different time points of D20-PM (PMSG treatment) and D24-PM (the rst arti cial insemination) was, from the highest to the lowest, PMSG-15D, CON, PMSG-20D, and EST groups.
In the afternoon of the 24th day, the follicle diameter of the four groups reached the maximum, which was 6.08 mm, 5.76 mm, 5.72 mm, and 5.00 mm, respectively. After that, the average follicular diameter began to decrease, and on the 26th day, the follicular diameter decreased to less than 3 mm (Fig. 4). At 42 hours after ALT feeding was stopped, the maximum observed values of preovulation follicular diameter of the gilts in different treatment groups, from the highest to the lowest, were the PMSG-15D, CON, PMSG-20D, and EST groups ( Table 1). The follicular diameter of PMSG-15D group was signi cantly higher than that in CON group (2.94 ± 0.24 mm vs 2.56 ± 0.30, P < 0.05), EST group (2.94 ± 0.24 mm vs 2.72 ± 0.44 mm, P < 0.05) and PMSG-20D (2.94 ± 0.24 mm vs 2.60 ± 0.24, P < 0.05).There was no signi cant difference between CON group, EST group and PMSG-20D group. At PMSG treatment, the gilts proportion of follicle diameter greater than 3 mm was different in different treatment groups, ranging from the largest to the lowest, were EST group, PMSG-15D group, PMSG-20D group and CON group respectively( Table 1).The proportion of gilts with follicular diameter greater than 3 mm in the PMSG-15D group was signi cantly higher than that in the CON group (37.50% vs 5.88%), and there was no signi cant difference among the CON group, the PMSG-15D group and the PMSG-20D group. The maximum observed values of preovulation follicular diameter of the gilts in different treatment groups, from the highest to the lowest, were the PMSG-15D, CON, PMSG-20D, and EST groups ( Table 2). Preovulation maximum follicular diameter in the PMSG-15D (6.28 ± 0.23 mm vs. 5.60 ± 0.47 mm, P < 0.01), PMSG-20D (6.06 ± 0.38 mm vs. 5.60 ± 0.47 mm, P < 0.01), and CON (5.98 ± 0.59 mm vs. 5.60 ± 0.47 mm, P < 0.05) groups were signi cantly higher than that of the EST group. Moreover, the preovulation maximum follicular diameter in the PMSG-15D group was signi cantly higher than that of the PMSG-20D group (6.28 ± 0.23. vs. 5.98 ± 0.59 mm, P < 0.05; Table 2). a, b, c, A, B means with different letters in the same column and within a category differed (P < 0.05) At 42 hours after ALT feeding was stopped, hormone levels of E2, LH, and P4 of gilts in the CON, EST, PMSG-15D, and PMSG-20D groups showed no signi cant difference (P > 0.05, Fig. 5a, b, d). Analysis of FSH content showed that the EST, PMSG-15D, and PMSG-20D groups were signi cantly lower than those of the CON group (P < 0.01). Furthermore, the consistency of FSH levels in the PMSG-15D group was higher than that in the EST and PMSG-20D groups (Fig. 5c).

Correlation between estrus and ovulation interval and estrus duration
The ovulation time distributions of the four groups were counted and are shown in Fig. 6 Table 3). The interval between estrus and ovulation (24.78 ± 12.30, P < 0.01) and the duration of estrus (41.63 ± 17.45, P < 0.05) in the EST group were signi cantly lower than those in the other three groups. However, there was no signi cant difference between the CON, PMSG-15D, and PMSG-20D groups (P > 0.05, Table 3). In the CON (r = 0.71), PMSG-15D (r = 0.65), and PMSG-20D (r = 0.63) groups, there was a signi cant correlation between the interval between estrus and ovulation and the duration of estrus (P < 0.01); however, no signi cant correlation was found in the EST group (P > 0.05, Fig. 7).

Effect on reproduction performance in FTAI replacement gilts
The estrus rate of gilts in the EST group (87.50%) was the lowest after the FTAI procedure. Compared with the PMSG-15D group (97.12%), the estrus rate in the EST group was not signi cant difference (P > 0.05), however, it was signi cantly lower than that of the PMSG-20D (100%) and CON (100%; P < 0.05) groups. The analysis of pregnancy rate of gilts in different treatment groups showed that it was highest in the PMSG-15D group (80.77%), which was 8.55%, 12.02%, and 2.99% higher than in the CON (72.22%), EST (68.75%), and PMSG-20D (77.78%) groups, respectively. However, the difference was not signi cant (P < 0.05). The total litter size and live litter size of the PMSG-15D group were the highest (10.44 ± 2.83 and 10.00 ± 2.61) among the four groups, which were 0.49 and 0.57 heads higher than those of the CON group (9.95 ± 2.44 and 9.43 ± 2.60) and live litter size of the PMSG-15D group, respectively. However, the difference was not signi cant (P > 0.05, Table 4).

Discussion
PMSG stimulates follicular growth and ovulation [15] . Some experimental studies showed that the estrus rate of gilts injected with 725-1000 IU PMSG was 70-100% [16] , whereas those injected with 363-600 IU PMSG was only 25-52% [17,18] . Therefore, a dose of 1000 IU PMSG was selected for pretreatment in this study. The follicle average diameter of gilts that were pretreated with PMSG before FTAI reached a maximum after ve days. At the same time, E2, LH, and FSH also increased sharply on the fth day, which led to ovulation between 5-6 days after PMSG treatment. Thus, PMSG pretreatment was shown to induce an estrus cycle in gilts and synchronize follicular development and ovulation time preliminarily.
Based on the reproductive cycle of gilts [19] , we assumed that the ovaries of gilts were in the luteal phase on the 10th day after ovulation (15-16 days after PMSG pretreatment), and in the follicular phase on the 15th day after ovulation (20-21 days after PMSG pretreatment). Therefore, FTAI treatment was selected after 15 and 20 days of PMSG pretreatment to explore the in uence of different ovarian states on the FTAI effect in replacement gilts.
When the FTAI procedure was started, the average size of follicles in the PMSG-15D group was below 3 mm, and no developing or mature follicles were present, indicating that the ovarian state of gilts in the PMSG-15D group was in line with the expectation that the ovarian follicle would be in the luteal phase after 15 days of PMSG treatment. The follicle sizes of the gilts in the PMSG-20D and EST groups were similar and larger than those in the PMSG-15D and CON groups, indicating that developing follicles appeared on the ovaries of gilts in the PMSG-20D group and that the ovarian state in the PMSG-20D group was in the follicular stage.On the third day after ALT treatment, the follicles of the four groups were synchronized to the same level and the diameter of the follicles was less than 3 mm, indicating that ALT played a role in effectively preventing the development of small and medium follicles, to control synchronous estrus [20] . PMSG has follicle-stimulating activity and promotes follicular development [21] . After the PMSG injection, follicular diameter tended to increase with time. GnRH effectively improves LH secretion and promotes the nal maturation and ovulation of dominant follicles. After intramuscular injection of GnRH, the follicular diameter of gilts in the four groups increased to the maximum, and the follicle reached maturity. On the second day after mating, the follicular diameter of all gilts dropped below 3 mm again, marking the end of ovulation.
Although the follicular development trend of the four groups was similar, the follicular development level was different at different time points. The study found that PMSG-15D had the largest follicle diameter at the time of PMSG injected compared with other groups. This is consistent with the results of Soede [22] . When gilts were fed with Regumate® (trade name of ATL) for 18 days during the pre-follicular phase, the average follicular size of the treated gilts increased by 0.5 mm compared to the control group, indicating that follicular selection and development had been differentiated after treatment with ALT. There is a continuous growth and atresia of ovarian follicles during days 7 to 15 of the estrous cycle, without evidence of follicular dominance, resulting in a stable size and number of small follicles on the ovaries. Moreover, ALT was found to have less inhibitory effect on follicular growth than endogenous progesterone (with or without ALT) [23,24] . Thus the application of ALT may not adversely affect the follicles during luteal phase.
After the effects of PMSG and GnRH, the differences in follicular development between the gilts in each group were signi cantly increased. The maximum follicular diameter before ovulation in the PMSG-15D group was signi cantly larger than that in the EST group and the PMSG-20D group (Table 1). It can be seen that the gilts in the PMSG-15D group (that is, the initial state is in the luteal phase) undergoing FTAI, showed the best growth ability in terms of the degree of follicular development. Soede found that the size of follicles before ovulation was positively correlated with luteal weight on the fth day after ovulation [25] .Similarly, a positive correlation was also observed in cows and ewes between the size of the follicle before ovulation and the size of the corpus luteum [26] . In cattle, the oocytes from larger follicles showed faster embryonic developmental ability than those from smaller follicles, which demonstrated the relationship between follicular size and oocyte capacity [27][28][29] . From these results, it can be inferred that larger follicles may have more capable oocytes, and may form larger or more functional luteal bodies after ovulation, potentially improving the subsequent fertility.
The results showed that the variation range of follicle size in PMSG-15D group at PMSG treatment and pre-ovulation was smaller than that in other groups, and the difference value was only 0.94 mm and 0.81 mm, indicating that the gilts in PMSG-15D group had a better synchronism in follicle development after FTAI treatment. The results also showed that the distribution of ovulation time of gilts in the four groups was different. All the gilts in PMSG-15D group completed ovulation within 36 hours, and the distribution of ovulation time was more concentrated than that in CON group, EST group and PMSG-20D group, indicating that the ovulation synchrony was better in gilts pretreated with PMSG for 15 days. The reason for the difference in follicular development and ovulation time was closely related to the hormone level in gilts. To verify the hormone level change, blood samples were collected at 42 hours after stopping feeding ALT.The level of FSH in the PMSG-15D group was the lowest, and it had the highest uniformity (Fig. 9).This may be due to that FSH of gilts in PMSG-15D group were not only inhibited by exogenous ATL supplementation, but also inhibited by endogenous progesterone secreted in luteal stage [24] .Subsequently, follicular development and ovulation in gilts were further improved by PMSG and GnRH treatment.
Successful insemination of gilts depends primarily on the ovulation and inseminationtime [30] . In theory, sperm show optimal fertility at 24 hours after insemination [31,32] . However, if insemination is done too late, the oocytes would be degraded [33] , which negatively impacts the sow fertility, including reductions in pregnancies, fertilization rates [34] , and embryo survival [35] . Since the arti cial insemination of FTAI was performed in the afternoon and morning on the 24th and 25th day, respectively, the optimal ovulation time was concentrated on the 25th day. This experiment found that the interval between GnRH injection to ovulation of PMSG-15D group was signi cantly shortened, and the ovulation rate (94.5%) was the highest on the 25th day, indicating that gilts in PMSG-15D group ovulated earlier than the other groups, and were more suitable for the current FTAI procedure. However, we also found that 18.9% of gilts in the PMSG-20D group ovulated after 25 days (Fig. 6), indicating that the ovulation time was too late.
In the EST group, the average follicle diameter was not the largest when treated with PMSG, but the percentage of gilts with an average follicle diameter above 3 mm was the highest. Therefore, the time from follicular development to ovulation in the EST group of gilts did not need to be too long after stoppong feeding of ATL, leading to a signi cant shorter intrval between GnRH injection to ovulationthan the other groups. However, gilts in the EST group had smaller preovulatory follicles, poor ovulation synchronization, and low reproductive e ciency.The gilts in EST group was in the follicular stage at the beginning of ATL treatment, and the follicles on the ovary were mainly composed of large follicles [23] .
Previous study had found a relatively low but constant concentration of LH during ALT treatment may cause the growth of large follicles [36] because large follicles are more sensitive to LH than small follicles [37] . In cows, ALT treatment for synchronization the estrous cycle had been found to prolong the lifespan of the dominant follicle, preventing it from becoming atretic [36] . Thus, if the large follicle was not atresia and degraded during ALT treatment, a portion of the follicle will continue to develop until ovulate after stopping feeding ALT. This may lead to excessive maturation of pig oocytes, which may negatively affect the maturation and fertilization of oocytes [38] .
In addition,Almeida found that the ovulation time depends on the duration of estrus, and there was a strong correlation between estrus duration and ovulation interval (r = 0.57, P = 0.0001) [39] . In this experiment, the CON, PMSG-15D, and PMSG-20D groups all showed a positive correlation between estrus duration and the interval from estrus to ovulation, while only the EST group showed no correlation (Fig. 7). Therefore, although the EST group ovulated earlier, it did not follow the correlation rule and presented with the worst follicular development, poor synchronization of ovulation time, and other factors explained the poor reproductive performance of the gilts in the EST group in this experiment.
Ultrasonic monitoring results of the PMSG-15D group showed that follicular development and ovulation time were superior to those of other groups since the initial state ovary was in the luteal phase. In subsequent FTAI, the reproductive performance of the PMSG-15D group improved, including the pregnancy rate (80.77%) and litter size (10.44 ± 2.83) (Table 3). However, the pregnancy rate and litter size of the PMSG-15D group did not increase signi cantly. This may be mainly because litter size is a compound variable in uenced by many factors, such as follicular mass [26] , ovulation number [40] , and uterine volume [41] . Hazeleger found that sows with feed restriction had smaller follicles and lower ovulation rates, indicating that follicle diameter before ovulation could re ect the number of ovulations on the ovaries [42] . Furthermore, the ovulation rate is moderately heritable and positively correlated with the litter size of gilts. Therefore, the lack of signi cant improvement in the litter rate in the PMSG-15D group was likely due to the increased ovulation rate and the embryo crowding caused by the limited uterine volume of gilts [43][44][45] .

Conclusions
We analyzed the follicular development, ovulation, and reproductive performance of replacement gilts in the luteal stage, pretreated with PMSG before undergoing FTAI and found that those pretreated with PMSG for 15 days showed better reproduction performance than those in the control group. It was feasible to optimize the FTAI program with PMSG pretreatment.

Declarations
Authors' contributions JJL JZP designed the study and revised manuscript. QQZ performed all the experiments, analyzed the data and wrote the manuscript. CYT participated in the discussion and wrote the manuscript. QLW helped solve the operational problems and data collection of the experiment. MZL and LYW helped collect the samples. ZWZ JH was involved in the selection of sows. XYC worked on the ELISA experiment. FXY LZ gives constructive analysis through discussion. All authors read and approved the nal manuscript. Availability of data and materials The data analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval
All experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals prepared by the Institutional Animal Care and Use Committee of Hebei Agricultural University, China.

Consent for publication
Not applicable.

Figure 1
Schematic of experimental design