Jugular vein blood and litter size data of 890 healthy non-pregnant ewes aged approximately 3 years were collected randomly from a herd in southwest Shandong Province, China. These ewes were genotyped for the FecB gene by the TaqMan method. Primer sequences of FecB-TaqMan and its probe sequences are listed in Table 1. The 6-μl reaction system contained 1 μl of DNA, 3 μl of 2× TaqMan Universal Master Mix II (Thermo Fisher Scientific Inc., Waltham, MA, USA), 0.3 μl of each FecB-TaqMan primer, 0.15 μl of two probe primers, and 1.9 μl of H₂O. Polymerase chain reaction (PCR) amplification was performed on a Roche LightCycler 480 II System (Roche, Basel, Switzerland) under the following PCR conditions: incubation at 95°C for 10 min, followed by 40 cycles of 95°C for 30 s and 60°C for 1 min. The allelic discrimination data were analyzed using endpoint genotyping with the hydrolysis probe protocol.

Pluriparous ewes (n = 53) with similar body conditions were selected and divided into three groups based on the FecB genotype (23 wild-type ++ ewes, 16 heterozygote +B ewes, and 14 homozygous mutant BB ewes) for subsequent experiments.

During the spring, estrus synchronization in these three groups was performed using a controlled internal drug release device (CIDR) (EAZI-BREED® CIDR® Sheep and Goat Device, Pfizer Animal Health, Auckland, New Zealand) with 300 mg of progesterone in all ewes for 12 days without any use of exogenous hormones for superovulation. Only 250,000-IU vitamin A and 25,000-IU vitamin D were injected to protect the vaginal epithelium when the CIDR was inserted. Estrus synchronization was performed on ewes twice with an interval of 14 days. In the first synchronization of estrus in 53 sheep of the three genotypes, estrus response, estrus onset, estrus duration, ovulation time, ovulation rate, and hormonal patterns in these estrous cycles were recorded. In the second experiment, cumulus cells were collected from 12 ewes. All ewes were maintained in a barn with food and water ad libitum.

After removal of the CIDR, eight rams with proven high serving capacity were selected as teasers wearing an apron around their hypogastrium to detect estrus in the ewes. The 53 experimental ewes (23 ++, 16 +B, 14 BB) and rams were raised separately. CIDRs were removed at 6:00 a.m. Teasing was initially performed every 6 h (at 6:00, 12:00, 18:00, and 24:00) for 96 h (19–22). Teasing was subsequently performed only once a day at 6:00 a.m. until the initiation of the next estrus. Following established definition and procedures, traits of estrus response, estrus onset, and estrus duration of ewes of the three groups in the first estrous cycle were recorded and calculated (22). The first estrus was defined from the ewes' first accepted mount, and the estrous cycle was defined as the interval from the last refused mount to the next one. Details of the experimental process of estrus determination are presented in Figure 1.

We randomly selected 6 ++ ewes, 7 +B ewes, and 6 BB ewes from among the 53 ewes of the three experimental groups to determine the ovulation time and rate by a modified version of the method reported by Romano et al. (22, 23). Laparoscopy was performed to determine ovulation at 48, 54, 60, 66, and 72 h after CIDR removal. The intervals from CIDR removal and estrus onset to first ovulation were respectively defined as the times elapsed from CIDR removal and estrus onset to the time halfway between the first appearance of preovulatory follicles and their disappearance. The intervals from CIDR removal and estrus onset to the last ovulation were respectively defined as the times elapsed from CIDR removal and estrus onset to the time halfway between the last appearance of preovulatory follicles and their disappearance. The ovulation rate (the number of corpora lutea on both ovaries) was also recorded for all 53 sheep in the three groups using laparoscopy on day 8 after CIDR removal (Figure 1).

Blood samples for testing hormone levels were collected from 7 ++, 8 +B, and 8 BB ewes from among the 53 ewes at the same time to determine the time of ovulation in the same batch. After removal of the CIDR at 06:00 a.m., 10 ml of jugular vein blood was collected in BD SST tubes (BD, San Diego, CA, USA) every 3 h for a 36-h period (24). From 36 to 120 h, blood was collected every 6 h. Jugular vein blood was kept at room temperature for 30 min, followed by centrifugation at 1,500× g for 10 min (25). The concentration of follicle-stimulating hormone (FSH) was determined using a commercial FSH [¹²⁵I]-labeled radioimmunoassay (RIA) kit (B03-FSH; Beijing North Institute of Biological Technology, BNIBT, Beijing, P. R. China). The sensitivity of the kit was <1.0 mIU/ml. The coefficient of variation (CV) in each batch was <10%, and the CV across the batch was <15%. Luteinizing hormone (LH) was quantified using a commercial LH [¹²⁵I]-labeled RIA kit (B04-LH; BNIBT). The kit sensitivity was 1.0 mIU/ml, and intra- and inter-batch CVs were <10 and <15%, respectively. The estradiol (E₂) concentration was tested using an E₂ RIA kit (B05-E2; BNIBT), the sensitivity of which was <2 pg/ml. The CVs for each batch and between the batches were <10 and 15%, respectively. Progesterone (P₄) was quantified using RIA kits (B08-P; BNIBT) with a sensitivity of <0.1 ng/ml, intra-batch CV of <10%, and inter-batch CV of <15%. To ensure the accuracy of hormone detection, triplicate assays were performed for each sample. The details are presented in Figure 1.

Ewes (4 ++, 4 +B, and 4 BB) were used for estrus synchronization. At 45 h after CIDR removal, cumulus–oocyte complexes were aspirated from follicles > 3 mm in diameter with a syringe containing Dulbecco's phosphate-buffered saline (Thermo Fisher Scientific Inc.). Then, the cumulus–oocyte complexes were washed with Dulbecco's phosphate-buffered saline and digested with hyaluronidase. After 5 min of centrifugation at 2,000 rpm, the granulosa cells were gathered and lysed in 1 ml of Trizol (Thermo Fisher Scientific Inc.). The lysate was immediately frozen in liquid nitrogen and stored at −80°C for further RNA extraction.

Total RNA from cumulus cells was isolated following the Trizol protocol. One microgram of RNA was reverse-transcribed with PrimeScript RT Reagent Kit (TaKaRa, Shiga, Japan). Real-time quantitative PCR was performed on the LightCycler 480 Real-Time PCR system (Roche, Switzerland) with the SYBR Premix Ex Taq kit (TaKaRa), as described (26). Primer sequences for quantitative PCR are listed in Table 1. Ribosomal protein L19 (RPL19) was used as an internal reference gene, and the 2^−ΔΔCt method was used to determine the relative expression levels in messenger RNA analysis (27).

Statistical analysis was performed using the generalized linear model Statistical Analysis System release 8.12 (SAS Institute Inc., NC, USA). All data are presented as the mean ± standard error of the mean, and P ≤ 0.05 was considered significant. P ≤ 0.01 was considered extremely significant. The statistical analyses of genotype frequency, allele frequency, and polymorphism information content were performed with reference to the formulas reported by Guo et al. (28). Traits in estrus determination, ovulation time, hormone concentrations in serum, and gene expression were continuous data. The normality of these continuous data was analyzed using the Shapiro–Wilk normality test. One-way analysis of variance was used to analyze the effect of FecB mutations on these normally distributed data. The data of the first estrus and estrus cycle were not normally distributed, so the Kruskal–Wallis nonparametric test was applied. Repeated-measures analysis of variance was adopted to analyze how hormonal (FSH, LH, E₂, and P₄) concentrations changed in different FecB genotypes over time. The estrus response was a binary variable, so Pearson's Chi-squared test was adopted to analyze the estrus response. Litter size and ovulation, traits measured as counts, are usually modeled by the Poisson distribution, so we analyzed the genotypic effects on litter size and ovulation rate through a generalized Poisson regression with log–link function using the generalized linear model GENMOD procedure of SAS. For litter size data, we set Y_j = 0, 1, …, ∞ to be the litter size of individual sheep. The expectation (E) and the variance of litter size were calculated as E(Y_j) = var(Y_j) = μ_j, where μ_j = exp(α₀ + X_jβ + Z_jγ), α₀ is the intercept, β is a 3 × 1 vector for parity, and γ is a 3 × 1 vector for genotype. The ovulation data from 53 and 19 sheep were analyzed separately, letting Y_j = 0, 1, …, ∞ be the ovulation rate of an individual. The expectation and the variance of ovulation rate were equivalent, E(Y_j) = var(Y_j) = μ_j, where μ_j = exp(α₀ + X_jβ), α₀ is the intercept, and β is a 3 × 1 vector for genotyping.

Using a TaqMan probe-based method, the allele and genotype frequencies of the FecB gene in experimental STH ewes were determined, as presented in Table 2. The frequencies of ++, +B, and BB genotypes were 0.16, 0.46, and 0.38, respectively; accordingly, the frequencies of the + and B alleles were 0.4 and 0.6, respectively. Chi-squared fitness testing showed that FecB gene mutations were under Hardy–Weinberg equilibrium in the randomly selected ewes (P > 0.05). The polymorphism information content of the FecB gene exceeded 0.3, reflecting moderate polymorphism, implying that the selection potential of the sites was considerable. For the polymorphic sites of FecB, the results of variance analysis as shown in Table 3 indicated that the litter size increased with increasing copy numbers of the B allele. Genotype and parity were shown to have significant effects on litter size. In each parity group, the mean litter size of the ++ genotype ewes was significantly smaller than in the other two genotypes. With increasing parity, the differences in litter size between BB and +B genotypes decreased. There was no significant difference in litter size between the BB ewes and +B ewes in terms of parity above two.

A total of 53 STH ewes of similar ages (2.9 ± 0.1 years) and weights (72.5 ± 1.9 kg) were selected for an estrus determination experiment based on the estrus synchronization technique. After CIDR removal, apart from the estrus duration (32.6 ± 1.4 h), all other estrous cycle indexes were significantly affected by FecB (Table 4). The first estrus for +B STH ewes was 23.6 ± 1.9 h from CIDR removal, which was earlier than in the ++ and BB genotypes at 32.6 ± 2.4 and 30.4 ± 1.6 h, respectively (P < 0.05). The cumulative estrus response is shown in Figure 2; all ewes presented estrus. The STH ewes with the +B genotype had an earlier response to estrus synchronization treatment, whereas the response of BB STH ewes had a narrower distribution.

The onset of estrus in +B STH ewes was 42.9 ± 2.2 h from CIDR removal, which was also earlier than in ++ and BB genotypes with 51.5 ± 2.0 and 50.0 ± 1.7 h (P < 0.05). Compared with 17.9 ± 0.1 and 17.9 ± 0.2 days in the STH ewes with ++ and BB genotypes, a shorter estrous cycle of 17.2 ± 0.2 days was observed in the ewes with the +B genotype. The findings suggested that STH ewes with FecB heterozygosity (+B) may exhibit an earlier first estrus, earlier estrus onset, and a shorter estrous cycle.

Ovulation times and rates in STH ewes among the three genotype groups were observed laparoscopically. The mean interval from CIDR removal to first ovulation in STH ewes was 58.1 ± 2.7 h (Table 5). We noted that the interval from CIDR removal to first ovulation for +B STH ewes appeared to be shorter (52.2 ± 4.4 h), albeit not significantly (P = 0.30). The mean intervals from first estrus and estrus onset to first ovulation in STH ewes were 26.3 ± 2.0 and 9.7 ± 1.2 h, respectively. The three FecB genotype groups did not differ significantly in any of the variables related to ovulation time. For the ewes with FecB mutation, the ovaries did not expel eggs simultaneously. Endoscopy performed every 6 h to estimate the time of ovulation indicated that all ovulations occurred in a period no longer than 6 h. The ovulation rate was measured successfully in the 53 STH ewes (Table 4). The ovulation rate of BB ewes (3.07 ± 0.35) was significantly higher than that of +B ewes (2.38 ± 0.13; P ≤ 0.01). Meanwhile, the ovulation rate of +B ewes was also highly significantly higher than that of ++ ewes (1.09 ± 0.02; P ≤ 0.01). The FecB mutation significantly influenced the ovulation rate of STH ewes.

The serum concentrations of FSH, LH, E₂, and P₄ were measured during an intact synchronized estrous cycle in 7 ++, 8 +B, and 8 BB ewes for 5 days after CIDR removal. Data presented in Figure 3 show the mean endocrine profiles of these ewes. Upon CIDR withdrawal, the P₄ concentration rapidly dropped to a low level. E₂ and FSH concentrations started to increase and peaked at approximately 27 h, which was aligned with the first estrus time. At 60 h after CIDR removal, the peaks of E₂ and P₄ dropped before the LH peak. FSH concentration peaked at 72 h after the LH peak. These four hormone concentrations peaked over 60–72 h.

The time of the first estrus of +B STH ewes was earlier than those of the other two genotypes. To investigate the changes of endocrine characteristics in serum before and after estrus, mean FSH, LH, E₂, and P₄ concentrations were calculated in relation to their levels of each individual's time of the first estrus (hour 0; see Figure 4). In the three genotypes, P₄ dropped at around hour 0, whereas P₄ concentrations in BB ewes were higher than those in +B and ++ ewes. E₂ increased from hour 0 to 3 in all three genotypes. In BB ewes, E₂ exhibited two peaks (−6 and 3 h), at which the E₂ concentrations were higher (P ≤ 0.05) than in the ++ genotypes. In ++ and +B ewes, the FSH concentrations dropped before hour 0. The FSH concentrations of ++ and BB ewes increased after hour 0 and peaked at hour 3. BB ewes had significantly higher (P ≤ 0.05) FSH concentrations than ++ ewes. LH concentrations increased at −3 h and decreased at 3 h, particularly in BB ewes. LH concentrations at 3 h were lower (P ≤ 0.05) than in the ++ ewes, whereas those in +B ewes were intermediate.

Taking together the results of the temporal trends in reproductive hormone levels and estrus and ovulation times, granulosa cells were derived from preovulatory dominant follicles approximately 45 h after CIDR removal. Total RNA was extracted and subjected to quantitative real-time PCR to analyze the expression of steroidogenic genes, cytochrome P450, family 17, subfamily A, polypeptide 1 (CYP17A1) and cytochrome P450, family 19, subfamily A, polypeptide 1 (CYP19A1); gonadotropin receptors, follicle-stimulating hormone receptor (FSHR) and luteinizing hormone/choriogonadotropin receptor (LHCGR); estrogen receptor 1 (ESR1); and the inhibin subunit alpha (INHA) (Figure 5). Expressions of the genes for the FSH and LH receptors were detected during the preovulatory stage, whereas no significant differences in the expression of these two genes were observed across the three genotypes. The expression of ESR1 in the +B genotype was higher (p ≤ 0.05) than in the wild-type (++) genotype. ESR1 expression in the BB genotype was higher than that in the wild type, but this did not reach significance. There was a trend for an increase of INHA expression with increasing FecB mutant copy number (P = 0.08). The expression of CYP17A1 and CYP19A1, markers of preovulatory follicles, also appeared at 45 h after CIDR removal when we collected the sample.