Search

Female reproductive performance is essential for economic profitability of swine operations, and sex steroid hormones, such as progesterone (P4), are necessary for proper reproductive function. Recent research targeting the reproductive microbiome in beef cattle found that as P4 concentrations fluctuate during the estrous cycle, bacterial communities in reproductive tissues also change. However, characterizing the relationship between P4 concentrations and bacterial communities within the reproductive tract has yet to be fully determined in swine. Therefore, the objective of this project was to correlate the abundance of reproductive tract bacterial communities with plasma P4 concentrations from pregnant gilts. Duroc X Landrace X Yorkshire gilts (n = 25), free of any physical or health related issues, were euthanized and hysterectomized on days (d) 11 (n = 7), 15 (n = 6), 60 (n = 6) or 90 (n = 6) of gestation. For microbiome evaluation, sterile swabs were individually rotated 8 times on each tissue and immediately placed in sterile microcentrifuge tubes for storage (-80°C) until sequencing. Allantoic and amniotic fluid collected on d 60 and d 90 of gestation was placed immediately in a sterile 10 mL conical tube and stored (-80°C) for subsequent analyses. Blood was collected concurrently, centrifuged at 2,500 x g for 25 min at 4°C, and the resulting plasma stored (-20°C). Concentrations of P4 in plasma were quantified using a double-antibody RIA kit. Bacterial sequencing of the V4 hypervariable region of 16S rRNA gene was completed by FERA Diagnostics and Biologicals Corp. The data were analyzed statistically using PROC CORR in SAS 9.4. In the cervix, the relative abundance of the phylum Firmicutes was positively correlated with P4 concentrations (r = 0.45, P = 0.03), and within this phylum, genus Lactobacillus was positively correlated with P4 concentrations (r = 0.54, P = 0.01). Phylum Bacteroidetes was negatively correlated with P4 concentrations (r = 0.45, P = 0.03), and within this phylum, genera Prevotella and Bacteroides were also negatively correlated with P4 concentrations (r = -0.49, P = 0.01; r = -0.05, P = 0.01, respectively). In the uterus, plasma P4 tended to be negatively correlated with phylum Proteobacteria (r = -0.37; P = 0.07) yet positively correlated with Phylum Firmicutes (r = 0.40; P = 0.05). Within Firmicutes, genus Streptococcus in the uterus was positively correlated with P4 concentrations (r = 0.50, P = 0.01), and the relative abundances of genus Lactobacillus in both the uterus and allantoic fluid were positively correlated with concentrations of P4 in plasma (r = 0.43, P = 0.04; r = 0.70, P = 0.02, respectively). There were no significant correlations between concentrations of P4 in plasma and abundances of bacteria within the vagina or amniotic fluid. These results indicate that shifts plasma P4 influence microbiomes in reproductive tissues of gilts during gestation. Further investigation is necessary to understand the interplay between P4 and microbiome fluctuations within reproductive tissues in swine.

Female sex steroid hormones have been shown to interact with cytokines within the immune system to help influence fertility. Previous research has indicated that shifts in commensal bacteria are associated with compromised reproductive performance during the postpartum period, yet evaluation of how the female sex hormone progesterone (P4) influences these microbial shifts is not fully understood. The objectives of this study were to determine 1) the relationship between P4 concentrations and cytokine concentrations, and 2) the relationship between P4 concentrations and the reproductive microbiome. Twenty Bos indicus-influenced beef cows (~60 days postpartum and free of health issues) were weighed, body condition scored and then subjected to the Bos indicus prostaglandin 5 day (d) + CIDR protocol beginning on d -8 and timed artificially inseminated on d 0. Blood samples were collected on d-3, -1, 0, and 28. Cytokines and reproductive microbiome were evaluated using vaginal and uterine flushes collected on d -3 and d-1. Experimental groups were assigned by d 28 pregnancy status, determined by transrectal ultrasonography, resulting in Open (n = 13) and Pregnant (n = 7) females. Bacterial community analyses were conducted targeting the V4 hypervariable region of the 16S rRNA gene. Plasma P4 concentrations on d -3 and d -1 were quantified per manufacturers’ instructions using a double-antibody RIA kit. Cytokine analyses of plasma and flushes were conducted using the RayBiotech Quantibody Bovine Cytokine Array Q1 kit per the manufacturer’s instructions. Statistical analyses were conducted using the GLM procedure and CORR procedure in SAS. Plasma P4 concentrations did not differ by pregnancy status (P > 0.05) but did decrease from d -3 to d -1 (0.89 vs. 0.08 ± 0.23 ng/mL, respectively; P = 0.02). There were no significant correlations between plasma P4 and plasma cytokines (P > 0.05). Interferon-alpha (IFNα) and P4 concentrations were positively correlated in Open and Pregnant uterine flushes (r = 0.41; P = 0.01). Concentrations of P4 interleukin (IL)-1F5 were negatively correlated in Open and Pregnant vaginal flushes (r = -0.32; P = 0.04). The relative abundance of genus Methanobrevibacter in uterine flushes was positively correlated with P4 concentrations in both experimental groups (r = 0.35; P = 0.02). Within vaginal flushes, there was a tendency for a negative correlation between the relative abundance of genus Bacteroides and P4 concentrations (r = -0.31; P = 0.05). Alternatively, there is a positive correlation between P4 concentration and relative abundance of genus Streptococcus (r = 0.47; P = 0.01) and genus Methanobrevibacter (r = 0.37; P = 0.02) in vaginal flushes. These results suggest there may be relationships between several cytokines and circulating P4 concentrations; therefore, potentially suggesting immunomodulatory activities of P4. Overall, this research alludes to P4 having a vital role in reproductive microbial shifts and associated immune responses. Further investigation on the local effects of P4 within the reproductive tract on immune responses and bacterial shifts could provide greater insight into mechanisms to improve fertility in beef cattle.

The differences between the microbial communities in reproductive tissues, and amniotic and allantoic fluids of cattle have previously been investigated. Our objective was to provide insight into microbial differences across reproductive tissues in gilts during mid- and late-gestation. Specifically, we characterized the microbiomes of the vagina, cervix, endometrium, chorion, and allantoic and amniotic fluids. Duroc x Landrace x Yorkshire gilts (n = 12) free of physical, health or reproductive-related issues were euthanized and hysterectomized at either mid-gestation (d 60, n = 6) or late-gestation (d 90, n = 6). A sterile swab was rotated 8 times to collect samples from the mucosal surface of individual tissues and immediately placed in microcentrifuge tubes for storage (-80°C) until sequencing. To sample the endometrium and cervix, a 1 cm incision was made to expose the endometrium and interdigitating prominences, whereas for vaginal sampling, sterile swabs were inserted 6 inches past the vulva and then rotated. For collection of samples of chorion, an incision was made in the endometrium, exposing the chorioallantois to allow sampling of both the chorion-endometrial and chorion-allantois sides. For allantoic and amniotic fluids, 5 mL samples were collected and placed immediately in a sterile 10 mL conical tube and stored (-80°C) for subsequent analyses. Bacterial DNA was extracted and genome sequencing targeting the V4 hypervariable region of the 16S rRNA gene was conducted by FERA Diagnostics and Biologicals Corp. Data were analyzed statistically using the GLM procedure in SAS. Across all samples, abundances of the phylum Firmicutes was affected by stage of pregnancy with greater relative abundance at d 90 compared with d 60 (52.23 ± 2.85% vs 42.60 ± 3.14%, respectively; P = 0.03). Within this phylum and across all samples, the relative abundance of the genus Lactobacillus differed due to stage of pregnancy with a greater relative abundance at d 90 compared with d 60 (19.33 ± 2.37% vs 12.14 ± 2.61%, respectively; P = 0.05). Lactobacillus was most abundant in allantoic fluid when compared with all other samples (P < 0.01). The relative abundance of the genus Streptococcus, within the phylum Firmicutes, was less in amniotic and allantoic fluid compared with the cervix and vagina (P < 0.01). The phylum Bacteroidetes tended to differ by sample type (P = 0.06) with amniotic fluid having the lowest relative abundance. Within this phylum, the relative abundance of the genus Prevotella was also least in amniotic fluid (P < 0.01). These results suggest that microbial communities of reproductive tissues and fluids differ between mid- and late-gestation in pregnant gilts. Additional research on reproductive microbiomes in swine, and other livestock species, is required to develop methods to modulate bacteria and improve reproductive performance and successful outcomes of pregnancy.

The pioneer microbiome is the initial and foundational colonization of microbiological organisms within the neonatal animal and has lifelong implications for health and performance. The objective of this project was to quantify various maternal and environmental sources contributing to the pioneer microbiome of the piglet. Five pregnant gilts farrowed at the Texas A&M University Swine Center in April 2022. Litters were a Duroc x Landrace x Yorkshire composite and averaged 13.4 total number born and 11.4 number weaned. At birth, all piglets were individually identified and weighed and 5 piglets per litter weighing ≥1,200 grams were selected for microbiome analysis. Sterile swabs were used to collect, in duplicate, samples from the farrowing crate before gilt introduction (EMPTYCRATE) and after gilts were moved in (FULLCRATE), from the birth canal during farrowing (BIRTHCANAL), and from the rectum of the piglet on days 0 (prior to suckling), 3, and 10 post-farrowing and at weaning (21.6 ±.0 days post-farrowing). Colostrum was collected from each gilt during farrowing from a representative sample of teats into a single sterile collection cup (COLOSTRUM). Swabs and colostrum were stored in sterile microcentrifuge tubes at -80°C until sequencing. Bacterial DNA was extracted, and sequencing targeted the V4 hypervariable region of the6S rRNA gene. Statistical analyses were conducted using PROC GLM and PROC REG in SAS 9.4. The phyla Firmicutes and Proteobacteria were not different between maternal (BIRTHCANAL, COLOSTRUM) and environmental (EMPTYCRATE, FULLCRATE) sources (Table 1). The piglet microbiome shifted from birth to weaning. The relative abundance of phylum Firmicutes was less on day 3 compared with day 0 and at weaning but did not differ from day0 (Table 1). Within the phylum Firmicutes, Lactobacillus and Clostridium genera were greater on day 3 compared with day 0, 10, and at weaning (Table 1). The relative abundance of phylum Proteobacteria, and the relative abundance of genus Escherichia within phylum Proteobacteria, were greater on day 3 compared with day 0, 10, and at weaning (Table 1). Multiple regression analyses indicated that BIRTHCANAL explained 51.6% of the variation observed in piglet day 0 microbiome (P < 0.0001), and 6.5% of the variation in the piglet day 10 microbiome (P = 0.01). The piglet day 0 (pre-suckle) microbiome explained 10.0% of the variation observed in day 3 microbiome and 15.6% of the day 10 microbiome (P < 0.05). Finally, day 10 microbiome explained 58.6% of the variation observed in the piglet microbiome at weaning (P < 0.0001). The microbiome of COLOSTRUM and the farrowing crate did not impact piglet microbiome (P > 0.10). Results indicate that the piglet pioneer microbiome is largely influenced by the microbiome of the birth canal and that the microbiome of the animal at weaning may be largely established by 10 days of age.

Status of bacterial diversity within the male seminal microbiome has previously been linked to fertility status in mammals. However, minimal research has focused on the reproductive microbiome in boar semen. The objective of this study was to analyze the diversity of bacterial communities within fresh extended boar semen samples 1) between studs, 2) over 5 consecutive days, and 3) between pooled and single-sire doses. Eight single-sire (n = 4 per stud) and eight pooled (n = 4 per stud) fresh extended semen doses (80 mL) were obtained from two boar studs (A vs. B). Pooled doses were the composite of boars in single-sire doses. Doses were subsampled (16 mL) for 5 days post-collection. Negative controls remained unopened until the last day. Each day, 5 mL of each subsample were flash frozen and stored at -80°C for bacterial analysis targeting the V4 hypervariable region of the 16S rRNA gene for Illumina sequencing. Quality of all sequences was determined by FastQC before being processed for taxonomic analysis using the qiime2 pipeline. Alpha diversity (difference within sample metric) was analyzed using Kruskal-Wallis test, whereas PERMANOVA was performed for beta diversity analysis (differences between sample metric). According to alpha diversity metrics, Chao1 (P < 0.01) and Shannon’s index (P < 0.01) indicated contrasts in diversity among all days and the negative controls. This was further confirmed with a pairwise comparison between individual groups, in which day 1 exhibited lower alpha diversity when compared with days 2, 3, 4, and 5 using Chao1 (P < 0.01) and Shannon’s index (P < 0.01). Interestingly, reduced diversity was observed for day 1 compared with the negative controls with Chao1 (57.4 ± 6.6 vs. 132.4 ± 22.1, P < 0.01, Q < 0.01; respectively) and Shannon’s index (4.8 ± 0.2 vs. 6.1 ± 0.3, P < 0.01, Q < 0.01; respectively). No differences between boar stud or sire-type were detected for alpha diversity metrics. For beta diversity analyses, both PCoA plots and PERMANOVA for weighted unifrac metric demonstrated no significant clustering between sampling days. However, distinctions between all days and negative controls were identified using PERMANOVA for unweighted unifrac metric. Specifically, differences were discovered between day 1 and days 3, 4, and 5 (P < 0.01, Q < 0.01). PERMANOVA for both weighted (P < 0.01) and unweighted (P < 0.01) unifrac metrics revealed distinct clustering by boar stud. However, there were no differences in beta diversity between pooled and single-sire samples for both weighted and unweighted unifrac metrics. Based on these results, bacterial diversity within a semen sample may be altered over time and could occur between boar stud operations. Additional studies analyzing the reproductive microbiome of semen could help improve the characterization of fertility status in boars.

Shifts in commensal bacteria (i.e., Firmicutes) in the human microbiome are associated with compromised fertility. The objectives of this study were to characterize 1) the reproductive microbiome and 2) the cytokine concentrations prior to artificial insemination (AI) in vaginal and uterine flushes from postpartum beef cows. Twenty Bos indicus-influenced beef cows (~60 days postpartum and free of health issues) were weighed, body condition scored (BCS) then subjected to the Bee Synch II synchronization protocol on d-8 and timed AI on d0. Blood samples were collected on d-3, -1, 0, and 28. Vaginal and uterine flushes were collected on d-3 and d-1. On d28, pregnancy was determined by transrectal ultrasonography (Open, n=13 and Pregnant, n=7). Bacterial community analyses were conducted targeting the V4 hypervariable region of the 16S rRNA gene. Cytokine analyses were conducted using the RayBiotech Quantibody® Bovine Cytokine Array Q1 kit per manufacturer's instructions. Weight, BCS, and vaginal bacterial communities by phyla were not different (P >0.05). In open cows, the uterine relative abundance of Firmicutes decreased between d-3 and -1 (60.4% ± 0.9% vs. 48.5% ± 3.2%, respectively; P< 0.05). In pregnant cows, the uterine relative abundance of Tenericutes significantly increased between d-3 and -1 (1.0% ± 0.1% vs. 7.6% ± 4.1%, respectively; P< 0.05). Additionally, the genus Ureaplasma tended to increase in relative abundance from d-3 to -1 (0.1% ± 0.1% vs. 7.3% ± 4.9%, respectively; P=0.06). Uterine concentrations of Interferon (IFN)-γ, Interleukin (IL)-1α and IL-21 were greater in open females compared to pregnant. Vaginal concentrations for IFN-α and chemokine ligand (CXCL)-10 were greater in open females compared to pregnant. Regardless of treatment, IL-13 increased from d-3 to d-1. These results suggest a distinct difference in the uterine microbiome and cytokine profiles prior to AI for resulting pregnant and open cows.

Bacterial contamination of non-frozen extended boar semen has been associated with a decrease in sperm viability and longevity. The objective of this study was to analyze the microbiome of extended boar semen and sperm quality parameters 1) between studs 2) over time and 3) between pooled and single-sire doses. Sixteen pooled (n=4 per stud) and single-sire (n=4 per stud) boar semen doses (80mL) were obtained from two studs (A vs. B). Pooled doses were the composite of boars in single-sire doses. Doses were stored at 16°C and subsampled 16mL daily for five days post-collection. Each day, 5mL were flash frozen and stored at -80°C for bacterial analysis targeting the V4 hypervariable region of the 16S rRNA gene. The remaining 11mL were used to assess sperm progressive motility and agglutination (0: no adhesion to 3: >50% adhesion) by two evaluators with 10 visual estimates per sample. Negative controls of each pooled (n=2 per stud) and single-sire (n=4 per stud) dose remained unopened until day 5. Statistical analyses were conducted using GLM procedure in SAS. Sperm agglutination was greater in stud A than B (1.6 vs. 1.0 ± 0.1, respectively; P< 0.01). Sperm motility tended (P=0.09) to be greater in stud B than A doses (67.4% vs. 61.5% ± 0.02%, respectively), and decreased over time (P< 0.01). Stud B had a greater relative abundance of Proteobacteria (60.0% vs. 47.2% ± 1.5%, respectively; P< 0.01) and lower abundance of Firmicutes (22.5% vs. 31.9% ± 1.4%, respectively; P< 0.01) compared to stud A. Bacterial load did not differ by stud or dose type but increased from day 1 to 3 (P< 0.01). Bacterial load between days 3, 4, and 5 post-collection was not different from the negative control. Results indicate differences in the microbiome between boar studs and its influence on semen over time.

Cytokines have a vital role in reproductive immune environment during the postpartum period. Previous data evaluated uterine and vaginal cytokine concentrations prior to insemination; however, was unable to correlate these data with plasma cytokine concentrations. The objective of this study was to determine relationships between uterine, vaginal, and plasma pro- and anti-inflammatory cytokine concentrations in postpartum beef cows prior to timed artificial insemination (TAI). Bos indicus-influenced beef cows (n=20) free of any physical, health or reproductive-related issues were subjected to the Bee Synch II protocol 8 days prior (d-8) to TAI (d0). Uterine and vaginal flushes and blood were collected on d-3 and d-1. Pregnancy was determined by transrectal ultrasonography on d28 (Pregnant, n=7; Open, n=13). Using the RayBiotech Quantibody Bovine Cytokine Array 1 the following cytokines concentrations were determined: interferon (IFN)-α, IFN-γ, interleukin (IL)-13, IL-1α, IL-1-F5, IL-21, tumor necrosis factor (TNF)-α, chemokine ligand (CXCL)-9, CXCL-10, and chemokine ligand 4 (CCL4). Concentration data were analyzed using PROC GLM and correlations using Pearson correlation in SAS. For plasma samples, 8 pro-inflammatory cytokines (IFN-α, IFN-γ, IL-13, IL-1α, IL-1-F5, IL-21, CXCL-9, and TNF-α) had greater (P< 0.05) concentrations in open cows compared with pregnant cows. For uterine flushes, 3 pro-inflammatory cytokines (IFN-γ, IL-1α, and IL-21) had greater (P< 0.05) concentrations in open cows compared with pregnant cows. For vaginal flushes, IFN-α concentrations were greater (P< 0.05) in open cows, while CXCL-10 concentrations were greater in pregnant cows. There were no significant correlations between plasma and uterine or vaginal samples (P>0.05). Interestingly, for open cows there were correlations for IFN-α (r=0.46; P=0.02), IFN-γ (r=0.58; P=0.002), and IL-21 (r=0.54; P=0.004) between uterine and vaginal samples; however, no correlations were observed for pregnant cows. These results suggest a greater abundance of pro-inflammatory cytokines within the uterus, vagina, and in peripheral circulation for resulting open cows prior to TAI.