AMH-INH-RFRP DNA vaccine has previously been constructed and preserved in our laboratory. Briefly, the recombinant plasmid AMH-INH-RFRP contained three fused fragments including S/AMH, S/INH, and S/RFRP, which were linked with 2A peptide ( Figure 1 ). Subsequently, AMH-INH-RFRP plasmid was transformed into attenuated salmonella choleraesuis (C500) with asd and crp double deletion, producing AMH-INH-RFRP DNA vaccine delivered by C500 strain.

AMH-INH-RFRP DNA vaccine was added (1: 100) to LB liquid medium (200 ml) and cultured in 37°C shaker at 220 rpm/min for 16 h. After harvesting through 4°C centrifugation at 6000 rpm for 10 min, the vaccine was re-suspended by phosphate-buffered saline (PBS). Before immunization, LB agar plates were used to place the bacteria in triplicate to determine the number of colony-forming units and acquire adjusted roughly 3 × 10¹⁰ CFU/ml by PBS.

Experimental buffaloes (n = 84) of 3-6 years old with first to third lactations were from local buffalo farms (Hubei JinNiu Co., Ltd. Shayang, China) and enrolled in this study, which has the body condition score (BCS) between 2.5 to 3 (1-5 score) and are free from reproductive problems. Experimental animals were fed a total mixed ration (TMR) containing forages and concentrates in routine. Forages were comprised of corn silage, rice straw, peanut, and concentrates included corn (38.0%), soybean meal (16.0%), linen (6.0%), cottonseed cake (6.0%), cornmeal (17.5%), vinasse (10.0%), baking soda (0.5%) and premixed material (6.0%). The experimental animals had free access to fresh and clean water during the experimental period. A guideline followed for using animals and experimental procedures was from the Animal Care and Use Committee of Huazhong Agricultural University (HZAUBU-2019-001).

A total of 84 buffaloes was randomly divided into four groups: groups T1 (3 × 10¹⁰ CFU/ml, n = 21), T2 (3 × 10⁹ CFU/ml, n = 21), and T3 (3 × 10⁸ CFU/ml, n = 21) were nasally immunized twice a day with 10 ml of AMH-INH-RFRP DNA vaccine for 3 days, respectively. The control group (n = 21) was nasally immunized twice daily with 10 ml PBS for 3 days. At 2 weeks after primary immunization, all animals were boosted with the same procedures as primary immunization. After a booster immunization, buffaloes were subjected to the classic Ovsynch-TAI protocol (GPG). During the experiment, a total of 10 buffaloes did not participate in the complete GPG procedure due to buffalo injuries and management negligence. The details of the treatment procedure are presented in Figure 2 . During the experiment, blood samples from the jugular vein of buffaloes were collected on days 14 and 28. The blood was then centrifuged at 3000 rpm for 10 min to separate the serum. The serum samples were stored at -80°C for further analysis.

From day 34 (a day earlier than PGF2α treatment) to day 40 (72 h post 2^nd GnRH injection), all buffaloes were scanned using B-type ultrasound (Shenzhen Well.D Medical Electronics Co., Ltd. China) to evaluate follicles and ovulation at 12 h intervals. Oestrus was observed and confirmed by obvious vaginal mucous discharge twice a day (8:00 and 16:00). Ovulation was confirmed at the subsequent ultrasonographic examination by the abrupt disappearance of the previously recorded dominant follicle. At 24 h intervals, the prominent follicles were compared to determine their growth rate (mm/day). A pregnancy diagnosis was conducted on the 35th day following artificial insemination (AI) to assess the rate of conception.

The Ovsynch-TAI program was used in this study. Briefly, all buffaloes were injected with 200 μg of GnRH analog (Sansheng Pharmaceutical Co., Ltd., Ningbo, China) on day 28. Further, after an intramuscular injection of 0.4 mg PGF2α (Sansheng Pharmaceutical Co., Ltd., Ningbo, China), the buffaloes received another shot of 200 μg GnRH on day 37. From 18 to 24 h after the second GnRH injection, TAI was performed for all buffaloes.

Anti-AMH, INH, and RFRP antibodies in serum were measured by indirect ELISA. Briefly, each 96-well microtiter plate was coated 100 ng/100 μl AMH, INH, and RFRP antigens, which were synthesized by Sangon-Peptide Biotech CO., Ltd. (Shanghai, China), and then incubated overnight at 4°C. After discarding the reaction reagent, 300 μl of PBS with Tween 20 (Tween 20, 1247ML100, Bioroxx, Germany) (PBST) was added into each well and washed for 5 times. Next, 200 μl of 1% bovine serum albumin (BSA, 4240GR100, Bioroxx, Germany) was added into each well for blocking and then incubated at 37°C for 90 min. After washing with PBST, serum samples (100 μl) diluted with PBS (1:25, 1:50, 1:100, 1:200, 1:400, 1:800, 1:1600, and 1:3200) were added into each well and incubated at 37°C for 1 h. Following PBST wash, goat anti-bovine IgG-HRP (1:5000; 6030-05, Southern Biotechnology Associates, Inc. USA) was added (100 μl/well) and incubated at 37°C for 1 h. After washing with PBST, 150 μl of tetramethylbenzidine (TMB, ATO00001, AtaGenix, Wuhan, China) were added into each well and incubated at 37°C for 15 min. Then 50 μl of stop solution (C1058, Solarbio Science & Technology Co., Ltd. China) were added to stop the reaction. Finally, the absorbance at 450 nm was measured by a Microplate Reader (PerkinElmer, USA). Under the same dilution factor, if the OD450 value of the experimental group was greater than the mean OD450 value in the control group + 2SD (standard deviation), it was judged as a positive value (18, 25).

Bovine ELISA Kits (Ruixin Biological Technology Co., Ltd.; Quanzhou, China) were used to measure concentrations of IL-4 (RX1600854B), IFN-γ (RX1600863B), E2 (RXJ1600845B), and P4 (RXJ1600748B) concentrations, according to the manufacturer’s instructions. The intra- and inter-assay coefficients of variation were less than 10.0% and 15.0% for IL-4, 15.0% and 15.0% for IFN-γ, 15.0% and 15.0% for E2, and 15.0% and 15.0% for P4, respectively.

The data were analysed using the MIXED procedure of SAS 9.4 software (SAS Institute, Inc., Cary, NC, USA). The mean values were compared using Tukey’ multiple range test with a significance level of p < 0.05. Nonparametric data were analysed using the chisquared test. Variability in the data was expressed as the standard error of means (SEM), and p < 0.05 was considered statistically significant.

As shown in Figure 3 , buffaloes immunized with AMH-INH-RFRP DNA vaccine could induce anti-AMH, anti-INH, and anti-RFRP antibodies on day 14 after primary immunization (PI) and booster immunization (BI). After primary immunization, T2 group showed remarkably higher anti-AMH and anti-INH antibody levels than T3 group (P < 0.01), and T1 and T2 groups exhibited significantly higher anti-RFRP antibody titer than T3 group (P < 0.05). After booster immunization, there was an increasing trend in anti-AMH, anti-INH, and anti-RFRP antibody titers. Anti-AMH and anti-INH antibody titers in T2 group were significantly greater than that in T3 group (P < 0.05), and anti-RFRP antibody titers were significantly greater in T1 and T2 groups compared to that in T3 group (P < 0.01).

On the other hand, we also calculated the antibody positive rate in different treatment groups ( Table 1 ). After primary immunization, the anti-AMH antibody positive rate in T1 and T2 groups was significantly higher than that in T3 group (85.71% and 85.71% vs 47.62%, P < 0.01). In addition, the antibody positive rates for anti-INH, and anti-RFRP in T2 group were significantly increased by 38.09% compared to T3 group (P < 0.05). After booster immunization, the antibody positive rates increased compared to primary immunization, even though there were no significant changes among different groups.

The levels of steroid hormones (P4 and E2) and cytokines (IL-4 and IFN-γ) were detected by ELISA kits ( Figure 4 ). Regarding serum P4 concentration, no significant difference was observed between the different groups (P > 0.05, Figure 4A ). Compared with the control (Ctrl) group, the E2 concentration in T1 group significantly increased by 432.65 pg/mL (P < 0.05, Figure 4B ). In addition, we observed that the levels of IL-4 and IFN-γ were significantly greater in T1 and T2 groups than that in Ctrl group (P < 0.05, Figures 4C, D ). Furthermore, we compared the differences in steroid hormones and cytokines between the antibody-positive (P) and antibody-negative (N) buffaloes. The results showed that the concentrations of IL-4, IFN-γ and E2 in antibody-positive buffaloes were significantly increased compared to those in antibody-negative buffaloes (P < 0.01), while there was no significant difference in the concentration of P4 (P > 0.05, Figure 4E ).

All buffaloes were monitored by ultrasonography to record the diameter of follicle and the growth rate of the dominant follicle (mm/day). As shown in Table 2 , the buffaloes in T1 (12.98 ± 0.46), T2 (13.55 ± 0.44) and T3 (12.56 ± 0.46) groups had a significantly greater diameter of ovulatory follicle compared to the Ctrl group (10.74 ± 0.46, P < 0.05). In addition, there was a significant increase in the growth speed of dominant follicle in T1 (1.34 ± 0.07) and T2 (1.32 ± 0.08) groups compared to Ctrl group (1.10 ± 0.08, P < 0.05). However, there were no significant differences in diameter of ovulatory follicles and the growth speed of dominant follicle between DNA vaccine immunization groups (P > 0.05).

Moreover, we further compared the changes in diameter of ovulatory follicles and the growth speed of dominant follicle between antibody-positive and antibody-negative buffaloes. As shown in Table 3 , the diameter of ovulatory follicles in the P group was highly significantly increased by 2.02 mm compared to the N group (P < 0.01), and growth speed of dominant follicle in the P group was much faster than that in the N group (1.33 ± 0.04 vs 1.13 ± 0.06, P < 0.05).

After immunization with AMH-INH-RFRP DNA vaccine, the oestrus, ovulation, and conception rates of buffaloes were observed and recorded. As shown in Table 4 , oestrus and conception rates of buffaloes in groups T1, T2, and T3 tended to be higher than the Ctrl group, although the difference didn’t reach significance (P > 0.05). Furthermore, the proportion of ovulation in T1, T2, and T3 groups was significantly superior to that of the Ctrl group by 35.67%, 36.84%, and 35.67%, respectively (P < 0.05).

Notably, there were extremely significant differences in oestrus and ovulation rates between antibody-positive and antibody-negative buffaloes in Table 5 (P < 0.05). Likewise, the conception rate in P group was also significantly higher by 24.03% than that in N group (P < 0.05).