If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Direct genomic value daughter pregnancy rate and services per conception are associated with characteristics of day-16 conceptuses and hormone signaling for maternal recognition of pregnancy in lactating Holstein cows
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins 80523-1680;Department of Animal Sciences, College of Agricultural Sciences, Colorado State University, Fort Collins 80523-1680
Our objective was to test the hypothesis that sorting Holstein dairy cows with high direct genomic value for daughter pregnancy rate (DGV-DPR) and low services per conception records improves interferon tau production and endometrial signaling for maternal recognition of pregnancy.
Materials and Methods
Lactating Holstein cows (n = 86) were sorted by DGV-DPR into high- and low-fertility groups with the Clarifide (https://www.zoetisus.com/animal-genetics/dairy/clarifide/clarifide.aspx#) DNA test and low and high services per conception from farm records. High-fertility nonpregnant (NP) cows were not inseminated (n = 7) and served as controls. High-fertility pregnant (HP; n = 7) and low-fertility pregnant (LP; n = 6) cows received timed AI. Conceptuses, uterine flushings, endometrial biopsies, and peripheral blood mononuclear cells were collected on d 16 of pregnancy or the estrous cycle.
Results and Discussion
Conceptuses from HP cows were longer (P < 0.05) than those from LP conceptuses. Concentrations of interferon tau in uterine flushings were greater (P < 0.05) in HP compared with LP and NP cows, positively correlated with DGV-DPR (r = +0.68; P < 0.05), and negatively correlated with services per conception (r = −0.59; P < 0.05). Endometrial ISG15 mRNA and protein were upregulated (P < 0.05) in HP compared with NP. Low-fertility pregnant cows had intermediate endometrial levels of ISG15 mRNA between HP and NP, and ISG15 protein tended (P < 0.10) to be greater in LP compared with NP cows. This tendency was also observed for increased ISG15 in HP compared with NP in peripheral blood mononuclear cells.
Implications and Applications
Results supported the hypothesis and provided evidence to suggest that Holstein cows with greater DNA-based DGV-DPR values and fewer services per conception have increased conceptus length and enhanced signaling for maternal recognition of pregnancy.
Decades of decreased reproductive efficiency in Holstein cattle in the United States were multifactorial and strongly associated with increasing milk yield; however, development of fertility trait breeding values, their use in economic selection indexes, and improving the reliability of breeding values with genomic selection has resulted in a modest reversal of this decline in fertility (
). During early pregnancy, the spherical blastocyst hatches from the zona pellucida (d 9 to 10), grows into an ovoid shape by d 12 to 14, and doubles in length every day between d 9 and 16 (
). During elongation, the trophectoderm of the conceptus secretes interferon tau (IFNT), which suppresses paracrine release of endometrial prostaglandin F2α (PGF2α) to initiate maternal recognition of pregnancy. The cytokine IFNT also induces a cascade of interferon-stimulated genes (ISG), which, in concert with progesterone, regulate cell-signal transduction to initiate maternal recognition of pregnancy in the endometrium (
Expression of interferon (IFN)-stimulated genes in extrauterine tissues during early pregnancy in sheep is the consequence of endocrine IFN-tau release from the uterine vein..
). However, recently, results from early-life genomic testing procedures were associated with later-life measures of reproduction [i.e., traits such as daughter pregnancy rate (DPR) measured in mature cows;
]. Also, when heifers were sorted into high- and low-fertility groups based on embryo transfer success, high-fertility heifers had greater pregnancy rates with more differentially expressed endometrial genes (
Differentially expressed genes in endometrium and corpus luteum of Holstein cows selected for high and low fertility are enriched for sequence variants associated with fertility..
). Consequently, we reasoned that combinations of farm-level data, such as services per conception (SPC), and genomic testing results could be used to identify cows that produced conceptuses with robust IFNT signaling with the dam. We hypothesized that sorting dairy cows for high direct genomic value (DGV-DPR) and low SPC improves IFNT production and endometrial signaling of maternal recognition of pregnancy on d 16.
MATERIALS AND METHODS
Animal Care and Sorting of Lactating Dairy Cows into High- and Low-Fertility Groups
Cattle handling, housing, and tissue sampling procedures were reviewed and approved by the Colorado State University Animal Care and Use Committee (protocol #14-5190). Eighty-six freshening Holstein cows entering their second and third lactation on a commercial dairy in southern Wyoming were used in this study. These cows were managed in a single pen as per the dairy’s relationship with Bovine Reproduction Specialist LLC. To initiate the study, 10 mL of whole blood was collected from the tail of each cow, placed in a purple-top EDTA tube (Vacutainer, Becton Dickinson), and then centrifuged at 1,500 × g at 4°C for 15 min. Buffy coat (peripheral blood mononuclear cells, PBMC) was collected and placed in 5 mL of ammonium-chloride-potassium (ACK) red blood cell lysing buffer. The pellet was washed again in ACK but centrifuged at 300 × g at 4°C for 10 min. The pellet was then washed in 1 mL of 1× PBS, centrifuged at 300 × g at 4°C for 10 min, resuspended in 1 mL of 1× PBS, and processed with the Qiagen DNeasy Blood and Tissue Kit according to the manufacturer’s instructions. The 100 ng of DNA was then shipped to Zoetis Inc. for genotyping by what is known as the Clarifide system (https://www.zoetisus.com/animal-genetics/dairy/clarifide/clarifide.aspx#). Specifically, extracted DNA from each cow was analyzed with the LD-Max single nucleotide polymorphism (SNP) chip. This chip contained approximately 63,000 SNP, which included the genotype results of BovineSNP50 (Version II) that are used in the Clarifide program, which in lay terms is called genomic testing. The trait of interest for this study was DPR, which was the percentage of nonpregnant cows that become pregnant during each 21-d period. A DPR of 1 implies that daughters from this bull are 1% more likely to become pregnant during that estrous cycle when compared with that of a bull with an evaluation of 0 (
). The DGV-DPR estimate was a summation of the effects of 54,001 SNP and their haplotypes. The SNP effects were estimated from a training population (i.e., cows and phenotypes used to estimate SNP effects) that included the reproductive performance records and breeding values of more than 15,000 lactating US Holstein cows. The Holstein cows used in this study were slightly below average for the trait of DPR (−1.7 ± 0.07), with a range of −3.2 to −0.3. This PTA typically has a range of −5 to +5 (
After identifying the top 20% and bottom 10% of the 86 cows based on DGV-DPR, we further ranked for fertility based on the number of SPC from the previous lactation. For a cow to be classified as high-fertility pregnant (HP), she had to have both high DGV-DPR and low SPC, whereas a cow classified as low-fertility pregnant (LP) had to have low DGV-DPR and high SPC. Fourteen cows with the highest DGV-DPR and lowest number of SPC were selected for the high-fertility group and randomly assigned to nonpregnant and pregnant groups [n = 7 nonpregnant (NP) and 7 pregnant (HP); Table 1]. Cows with the lowest DGV-DPR and greatest number of SPC were selected for the low-fertility group (n = 7 pregnant, LP). The SPC values for sorting cattle in the present study were based on the previous lactation and were not based on the values during the lactation in the present study. To focus on specific sorting of DGV-DPR combined with SPC, cows with conflicting DGV-DPR and SPC values, as well as unhealthy cows (lameness, mastitis, and so on), were omitted from the study.
Table 1Trait and lactation information (mean ± SE) in 20 Holstein cows classified as low or high fertility
High-fertility NP cows were designated as the control group because it was the intent of this study to compare LP and HP with NP, so that pregnancy responses such as IFNT and ISG could be studied. Adding a second low-fertility NP control group was considered unnecessary in the context of the primary interest in pregnancy responses. Because this was a study with such high labor, large financial investment, and tedious sample collection, 2 control groups were deemed excessive in the context that important dependent variables would likely be similar between the 2 nonpregnant groups (i.e., no IFNT and no upregulation of ISG15 in either group). Also, we considered other endocrine variables, such as serum concentrations of progesterone, to help understand differences between HP and LP groups; however, we observed that blood steroid hormones were similar between LP and HP groups and were within normal ranges that were consistent with previous studies.
Cows were managed together in a single pen of 86 cows before selection to minimize differences in environmental factors and effects. After selection, cows were managed as a group of 20 cows in a single pen at the dairy owned and managed by Bovine Reproduction Specialist, where facilities existed for intensive reproductive management. The study was initiated in winter, in Colorado and Wyoming, to eliminate the effect of heat stress. The 20 cows were daughters of 6 sires. One LP cow was removed from the study and all subsequent analysis because of development of a uterine abscess and inability to produce an embryo. Because these cows received Clarifide scores, they were also issued genome-enhance (g)PTA and reliability predictions for other production traits, such as milk yield, through the cooperative program of Zoetis Inc. and the Council for Dairy Cattle Breeding (https://www.uscdcb.com).
Estrous Synchronization, Timed AI, and Conceptus Flush
The HP and LP cows were bred by timed AI (TAI) by the theriogenologist involved in this study, following a 60-d voluntary wait period (60 d after calving). Estrous synchronization procedures were applied to each cow based on the production calendar to make sure that each cow had completed the 60-d voluntary wait period before AI. Estrous cycles were synchronized using a variation of the OvSynch program: 2 mL of gonadotropin-releasing hormone (Factrel Gonadorelin-Gonadotropin Releasing Hormone; Zoetis) on d 0, followed by 5 mL of PGF2α (Lutalyse Dinoprost Tromethamine-Prostaglandin; Zoetis) on d 7, with gonadotropin-releasing hormone 52 h later and TAI at 16 to 18 h. Cows with short estrous cycles that did not have a corpus luteum (CL) for 2 wk in a row based on ultrasound before estrous synchronization were enrolled in a 7-d CIDR-Synch (controlled internal drug-release insert; Zoetis) program with the same intervals of treatments as the standard OvSynch program (
The cows that were bred to standing estrus were inseminated on an “a.m.–p.m.” schedule under the same general management so that this was consistent across treatment groups. Each cow had the same opportunity to become pregnant to a single insemination for each estrous cycle. All cows assigned to the pregnant groups were inseminated using semen from the same high-fertility bull (Ronelee Gold Digger; Accelerated Genetics). The relevant PTA for this bull were DPR (+1.8), heifer conception rate (+2.9), and cow conception rate (+3.5). For this reason, sire was not included in the statistical model. Reliabilities of these PTA were ≤93%. Furthermore, this bull was not a carrier for any of the known detrimental fertility haplotypes. All NP cows were not inseminated and considered d 0 at the onset of estrus.
On d 16 of pregnancy, conceptuses were collected from LP (n = 6) and HP (n = 7) dairy cows using a nonsurgical flush technique (
). Before initiating these procedures, ovulation and the subsequent CL were confirmed with palpation and ultrasound (IBEX-Pro, EI Technologies). The flush medium was Dulbecco’s Phosphate Buffered Saline (PBS; Sigma) containing 0.1% polyvinyl acetate (Sigma) and Bioniche Vigro Complete Flush media (MWI Veterinary Supply). The flushing medium was administered with 20 French silicone catheters with 30-mL balloons and a single opening. The uteri of the 7 NP cows were flushed to serve as negative controls. Initially, uteri were flushed with 50 mL of medium with recovery of about 40 mL for analysis of IFNT and the conceptus. Conceptuses that were not recovered on the first uterine flushing (UF) were immediately flushed for a second time using larger volumes (up to 1 L) of flush medium. If no conceptus was recovered from HP or LP groups, cows were resynchronized, inseminated, and flushed again on d 16 following TAI. Embryos that were recovered in multiple pieces (2 or more) were pieced together and then measured. The average number of UF per group was 1 for NP cows, 2.33 (14 flushes for 6 conceptuses) for LP cows, and 1.86 (13 flushes for 7 conceptuses) for HP cows.
Conceptuses were examined for the following morphologic criteria: color, length, and tubular structure using a stereo-microscope (Stereo Star Zoom, American Optical) at 7× magnification. Conceptuses were photographed beside a ruler to determine their length. Following the evaluation of previously mentioned criteria, conceptuses were then typed as either viable/normal (clear, translucent, tubular) or undergoing EM (pink, opaque, collapsed). One-half of each conceptus was placed into a microcentrifuge tube, snap frozen in liquid nitrogen, and stored at −80°C until processing for RNA analysis. The other half was stored at −80°C for DNA and protein extraction. The UF were snap frozen in liquid nitrogen and stored at −80°C for protein analysis.
Collection of Endometrium, Serum, and Peripheral Blood Mononuclear Cells
After collection of the conceptus, endometrial biopsies were obtained using transcervical Jackson uterine biopsy forceps with a cutting area of 4 mm × 28 mm (Jorvet, Jorgensen Laboratories Inc.). Each endometrial biopsy was trimmed from any connective tissue or myometrium and immediately frozen in liquid nitrogen and stored at −80°C until processed for RNA, DNA, and protein analysis.
Blood from the jugular vein was collected into serum collection tubes and centrifuged at 1,500 × g for 15 min at 4°C, and the serum was aliquoted and frozen at −20°C. Peripheral blood mononuclear cells (PBMC) were isolated from whole blood that was collected in 10-mL EDTA blood collection tubes (Becton, Dickinson and Company) and centrifuged at 1,500 ×gfor 15 min at 4°C. The buffy coat was transferred to a 15-mL centrifuge tube containing 5 mL of ACK (150 mM ammonium chloride, 10 mM potassium bicarbonate, 0.1 mM EDTA, pH 7.2 to 7.4) lysis buffer, inverted 10 times, and then incubated at room temperature for 10 min. The PBMC were pelleted by centrifugation at 300 ×gfor 10 min at 4°C, decanted, resuspended in 2.5 mL of ACK lysing buffer, incubated at room temperature for 5 min, pelleted at 300 ×gfor 10 min, decanted, resuspended in 2 mL PBS, transferred to a 2-mL microcentrifuge tube, and pelleted by centrifugation at 300 ×gat 10 min (4°C). The PBMC were then frozen in liquid nitrogen and stored at −80°C until processing for RNA, DNA, and protein analysis.
RNA, DNA, and Protein Extraction
TRIzol Reagent (Life Technologies) was used to extract RNA and DNA from endometrium and PBMC samples following the manufacturer’s instructions. Endometrium and PBMC pellets (50–100 mg) were homogenized in 1 mL of TRIzol Reagent. The RNA pellets were dissolved in 87.5 μl nuclease-free water. Remaining DNA was removed from the RNA fractions by treatment with RNase-Free DNase (Qiagen) and RNeasy MinElute Cleanup Kit (Qiagen). The RNA was quantified and the A260/280 and A260/230 determined using NanoDrop (Thermo Fisher Scientific); A260/280 ratios greater than 1.7 and A260/230 ratios greater than 1.8 were considered free of DNA contamination. Protein pellets were dissolved in 200 to 400 μl of 1% SDS. The insoluble material was sedimented via centrifugation at 10,000 ×gfor 10 min at 4°C and the supernatant diluted 1:10 in PBS for quantification. Protein was quantified using a Pierce BCA Protein Assay Kit (Thermo Scientific) as per the manufacturer’s instructions.
Western Blot
Uterine fluid samples were centrifuged for 20 min at 4°C and 1,000 ×gto reduce cellular debris. Samples were concentrated and desalted using 2 Amicon Ultra-4 Centrifugal filters (EMD Millipore). The protein concentration was quantified using a Pierce BCA Protein Assay Kit (Thermo Scientific). A total of 45 μg of protein from each UF sample was loaded into a lane of 12% SDS-PAGE gels and electrophoresed for 1.5 h at 200 V. Proteins were transferred to a nitrocellulose blotting membrane (Amersham Protran 0.2 μm NC; GE Healthcare) for 1 h at 100 V. Membranes were incubated for 1 h in 5% nonfat dry milk in Tris-buffered saline + Tween 20 with a pH of 7.5 (TBST) at room temperature, incubated with rabbit anti-bovine IFNT primary antibody (1:5,000; from Dr. Michael R. Roberts, University of Missouri, Columbia, MO) or mouse monoclonal anti-bovine interferon stimulated gene 15 (ISG15) primary antibody (1:1,000;
) diluted in TBST with 5% nonfat dry milk overnight at 4°C. Membranes were washed 3 times in TBST for 5 min and incubated with donkey anti-rabbit or donkey anti-mouse secondary antibody diluted in TBST with 1% nonfat dry milk (1:2000; Santa Cruz Biotechnology Inc.) for 1 h at room temperature. Following 3 more washes in TBST for 5 min each, proteins were detected using the Amersham ECL Prime Western Blotting Detection Reagent Kit (GE Healthcare). Quantification was performed using the optical densitometry program, Image Lab 4.1, on a ChemiDoc XRS+ System with Image Lab Software (Bio-Rad Life Science).
Semiquantitative RT-qPCR and Primers for ISG15
Single-stranded complementary DNA (cDNA) was synthesized from 1 μg of RNA using the iScript cDNA synthesis kit (Bio-Rad Life Science). Synthesized cDNA was diluted 5-fold with RNase-free water for the reverse-transcription quantitative PCR (RT-qPCR) reaction. Primer3 software was used to design primers that were optimized for a 61°C melting point and lacked self-annealing or folding at high temperatures. Housekeeping genes, target genes, and primers are listed in Table 2. Primers generated amplicons ranging from 80 to 120 bp in PCR reactions. Amplicons were sequenced to confirm identity to the target gene. The RT-qPCR reactions were performed in duplicate using 2 μl of cDNA, 5 μl of iQ SYBR green supermix (Bio-Rad Life Science), 1.5 μl of RNase-free water, and 1.5 μl of 7.5 nM solution of each primer set per well of a 384-well plate and amplified on a LightCycler 480 (Roche). Amplification of qPCR products was performed at 95°C for 3 min for denaturation, followed by 40 cycles of 95°C for 30 s, 61°C for 30 s, and 72°C for 15 s. The reaction products were assessed for quality by melting curve analysis on the LightCycler 480 SW 1.5.1.62 program (Roche). The RT-qPCR results were analyzed using the comparative Ct (ΔCt) method described by
Whole blood was centrifuged at 896 ×gfor 20 min at 4°C to separate clotted blood from serum. Concentrations of progesterone (P4) in serum were determined by radioimmunoassay as previously described (
). All samples were analyzed in a single assay. The average sensitivity (limit of detection) of the assay was 14.7 pg/mL; the intra-assay coefficient of variation was 4.64%. The estradiol radioimmunoassay was completed as previously described (
). The limit of detection for the estradiol assay was 0.92 pg/mL with an intra-assay coefficient of variation of 11.3%.
Statistics
The statistical analyses were performed using R statistics (version R.3.2.3; R Foundation for Statistical Computing; Team RC, 2015). Correlations between variables, such as DGV-DPR, SPC, and IFNT, were completed using the lm and aov procedures to fit the regression and test for ANOVA. Groups NP, HP, and LP were entered as equal main effects and analyzed for ANOVA using the aov (ANOVA), lsmeans, and cld (compact letter display of pairwise comparisons in R Software) procedures in the lsmeans (
). Differences between groups were considered a tendency when P < 0.10 and significant when P < 0.05. Other than correlation data, values represent the arithmetic mean ± SE.
RESULTS AND DISCUSSION
Relationship Between gPTA-DPR, SPC, and Fertility Classifications
Conventional breeding approaches to improve pregnancy rates are difficult in dairy cattle because most reproductive traits are polygenic and lowly heritable. Mechanisms such as pleiotropic gene effects, linkage, and physiological associations, and the time it takes to acquire data, also contribute to this complex problem (
). Therefore, the use of genomic tools to understand parental contributions to an offspring’s genome could help genetic selection of fertility traits. Most importantly, the advancement and use of these types of tools has great industry applicability in DNA-based early-life screening for herd replacements.
These types of advancements are occurring in the North American dairy industry as early-life genomic testing procedures were recently reported to be associated with later-life measures of reproduction (i.e., traits such as DPR and cow conception rate;
). These processes are simple for farmers as they collect a sample (i.e., blood, hair, tissue punch) for DNA extraction and send the sample to the laboratory service provider, and subsequently, genotype scores, such as direct genomic values, are returned. These data can also be combined with the genetic evaluation system and returned as genome-enhanced PTA values (i.e., more correctly, the acronym gPTA) providing breeding value tools for dairy farmers to decide whether to keep or cull females.
In this study, DGV-DPR and SPC data were used to sort Holstein cows on a commercial dairy into HP and LP groups. Specifically, no differences were observed in gPTA-DPR, gPTA-milk, DIM, number of lactations, and Julian calving date among NP, HP, and LP groups (Table 1). However, the DGV-DPR was greater (P < 0.05) and SPC from the previous lactation was lower (P < 0.05) in the high-fertility (NP and HP) groups compared with the low-fertility (LP) group (Table 1). The DGV-DPR was negatively correlated with SPC (r = −0.57; R2 = 0.33; P < 0.05), confirming the inverse relationship for these fertility classifications and the experimental groups (Figure 1). It should also be noted that the current study used cows sorted by these types of genomic tools (DGV-DPR) and the farm’s SPC to obtain conceptuses, uterine fluid, and endometrial tissues as well as PBMC to help understand major components and processes of maternal recognition of pregnancy (i.e., IFNT and ISG15). The physiological measures from the cows of this study may be limited relative to the traits used in quantitative and population genetic analyses (
); however, this strategic approach allowed sampling of cows that helped provide very valuable tissues and important physiological information underpinning genetic improvement of the trait DPR.
Figure 1Relationship between direct genomic value for daughter pregnancy rate (DGV-DPR) and services per conception (SPC) in lactating dairy cows. The coefficient of determination (R2) is the percent of variance explained by the model. NP = high-fertility nonpregnant cows; LP = low-fertility pregnant cows; HP = high-fertility pregnant cows.
Conceptuses were classified based on morphology and length (Figure 2A), and their phenotypes were confirmed by the detection of IFNT mRNA using RNA-Seq (RNA-Seq normalized counts; T. R. Hansen and M. G. Thomas, both of Colorado State University, Fort Collins, personal communication). Healthy conceptuses were translucent, long, and did not collapse during handling, whereas EM conceptuses were darker, red to pink in appearance, and shorter. Embryo mortality conceptuses were sticky and often collapsed during manipulation with a plastic pipette. The HP cows (n = 7) yielded 5 normal conceptuses and 2 EM conceptuses, whereas the LP cows (n = 6) produced 3 normal conceptuses and 3 EM conceptuses. There were no differences observed in the length of all LP (normal and EM) compared with all HP conceptuses; however, normal conceptuses from HP (124.5 ± 5.6 mm) cows tended (P < 0.10) to be longer than normal LP (94.5 ± 17.5 mm) conceptuses (Figure 2B) and were more than double the length (114.5 ± 8.5 mm) of the EM conceptuses (50.4 ± 11.5 mm, P < 0.05). There was no difference in length between HP and LP EM conceptuses.
Figure 2Embryo classification (A) and numbers (B) of embryos from low-fertility pregnant (LP) and high-fertility pregnant (HP) dairy cows. Panel A shows a collapsed, narrow, short-and-dark-pink embryo mortality (EM) conceptus and a viable, normal (N) long and translucent conceptus. Note the guides on the metric ruler are aligned in both photos to provide perspective regarding the size of these conceptuses. Embryo lengths from all embryos classified as N or EM are presented in panel C. Values represent mean ± SE. Means marked with different letters (a, b) differ (P < 0.05). *Represents a tendency to differ (P < 0.10).
describing greater pregnancy rates and longer conceptuses in high- compared with low-fertility heifers that were sorted using a serial embryo transfer success research model. These authors also described increased prostaglandin (PGE2 and PGF2α) concentrations and increased lipid metabolism in high- compared with low-fertility heifers (
) in high- compared with low-fertility heifers. In the present study and described more in depth in the next section, an increase in IFNT in UF was associated with an increase in endometrial ISG15 mRNA and protein concentrations in HP compared with NP cows, with LP cows having ISG15 concentrations that were not different from NP or HP (Figures 3 and 4). The difference between the 2 studies regarding the endometrial ISG response may be caused by the timing of the endometrial sample collection on d 16 in the present study versus d 14 in the study by
). It is plausible to suggest that by d 16 of gestation, there would be greater exposure to IFNT (i.e., both greater concentrations and increased duration) to induce the ISG gene expression response in the endometrium compared with the d-14 observations described by
), which may contribute to the difference in endometrial ISG response.
Figure 3Interferon tau (IFNT) protein detected from uterine flushing in lactating dairy cows. (A) Image of IFNT Western blot. (B) Quantitation of uterine flushing IFNT release in NP, LP, and HP cows. NP = high-fertility nonpregnant cows; LP = low-fertility pregnant cows; HP = high-fertility pregnant cows. Values represent mean ± SE. Means marked with different letters (a, b) differ (P < 0.05). Mr = relative molecular mass.
Figure 4Relationship between interferon tau (IFNT) detected in uterine flushing and direct genomic value for daughter pregnancy rate (DGV-DPR; A) or services per conception (SPC; B). LP = low-fertility pregnant cows; HP = high-fertility pregnant cows. The coefficient of determination (R2) is the percentage of variance explained by the model.
The IFNT protein in UF appeared as a doublet migrating at about 22 to 24 kDa on Western blots (Figure 3A), which reflects differential glycosylation. This glycosylation may contribute to differing bioactivity of IFNT proteins (
Synthesis and processing of ovine trophoblast protein-1 and bovine trophoblast protein-1, conceptus secretory proteins involved in the maternal recognition of pregnancy..
Molecular cloning and characterization of complementary deoxyribonucleic acids corresponding to bovine trophoblast protein-1: A comparison with ovine trophoblast protein-1 and bovine interferon-alpha II..
). Even though both glycosylated forms were present (see Figure 3), we were unable to discern structural or functional differences in IFNT in the UF from HP and LP cows in this study using the anti-IFNT rabbit polyclonal antibody. Nonetheless, greater concentrations of IFNT were discovered in UF from all HP compared with LP cows (P < 0.05, Figure 3B), which supports the hypothesis that sorting dairy cows for high DGV-DPR and low SPC improves d-16 IFNT production and endometrial signaling of maternal recognition of pregnancy.
As expected, IFNT was not detected in UF from any NP cows. Increased IFNT release from the conceptus was associated with increased DGV-DPR (Figure 4A) and decreased SPC (Figure 4B) in HP compared with LP cows. Direct genomic value-DPR was positively correlated (r = 0.68; P < 0.05) with IFNT concentrations in UF and explained approximately 46% (R2 = 0.46) of the variation in IFNT concentrations. Furthermore, IFNT concentrations in UF were negatively correlated with SPC (r = −0.59; R2 = 0.34; P < 0.05). These results support the premise that genomic testing results are associated with physiological measures of factors involved in maternal recognition of pregnancy.
Because of data recording systems on commercial dairy farms and the advancements of genomic testing procedures, genotype-to-phenotype validation analyses are possible and of great interest for genetic improvement strategies. Validation studies with large numbers of cattle are needed in these processes (
The relationships among sire’s predicted transmitting ability for daughter pregnancy rate and cow conception rate and daughter’s reproductive performance in Canadian Holstein cows..
); however, intensive physiological measurements and associations with fertility traits, such as in the current study, provide much needed information and confidence for the use of genomic tools to select for cow fertility and the on-farm decision of retaining and culling replacement females.
The lower concentrations of IFNT in LP compared with HP UF may have been caused by LP embryos that were dying. Preliminary analysis of the EM embryos and gene expression of this study revealed upregulation of several cell death genes {i.e., caspase 7 [fold change (FC) = 17.1], caspase 8 (FC = 22.8), receptor interacting serine/threonine kinase (FC = 39.0), E3 ubiquitn ligase-associated factor 1 (FC = 28.2)} that were highly upregulated (P < 0.05) in EM compared with normal conceptuses, which supports this idea (T. R. Hansen and M. G. Thomas, both of Colorado State University, Fort Collins, personal communication). The cytokine IFNT was not detected in 4 out of 5 EM UF, suggesting that these conceptuses ceased production of IFNT. Also, one EM conceptus secreted some IFNT into the UF media, and we suspect that this embryo was in the process of dying. The causes of early EM and features of early dying embryos are the focus of future studies. Nonetheless, even though cow numbers in the current study were limited, this study again documents the challenges of high rate of EM in lactating Holstein cows (
For healthy conceptuses, it should also be noted that IFNT synthesis and secretion by trophoblasts may be influenced by (1) the numbers of IFNT genes in the genome (
), (2) increased IFNT gene expression by the trophoblast, (3) preferential expression of one IFNT gene subtype, (4) increased expression of IFNT genes that encode an IFNT protein with greater activity compared with other IFNT gene products (
The cross-species antiviral activities of different IFN-tau subtypes on bovine, murine, and human cells: contradictory evidence for therapeutic potential..
). Delineating which IFNT genes contribute to the robust production of IFNT, or if specific genes are attenuated or modified epigenetically in LP versus HP cows, remains to be studied.
Effect of IFNT on ISG15 Expression in Endometrium and PBMC
There are many endometrial and PBMC ISG that are upregulated by the presence of a conceptus (
). These ISG are critical when acting in cooperation with progesterone to prepare the uterine lining for adhesion, impairing pulsatile release of PGF2α, and facilitating nutrient delivery to the conceptus (
). One of these genes is ISG15, a tandem ubiquitin-like repeat that conjugates to proteins in the cytosol of endometrial epithelial cells to aid in maternal recognition of pregnancy (
Deletion of the Isg15 gene results in up-regulation of decidual cell survival genes and down-regulation of adhesion genes: Implication for regulation by IL-1beta..
). Herein, ISG15 was used as a marker for IFNT action and also an indicator of maternal (endometrial and PBMC) response to the d-16 elongating conceptus.
In the current study, ISG15 mRNA concentrations were greater (P < 0.05) in the endometrium from HP cows compared with NP cows (Figure 5). However, in LP cows, endometrial ISG15 mRNA concentrations were intermediate and did not differ from that of HP or NP cows. The protein fraction of endometrial tissues was examined by Western blot to determine ISG15 protein concentrations. Free ISG15 was not detected in the endometrial proteins from any of the cows. Likewise, conjugated ISG15 was not detected in endometrium from NP cows, tended (P < 0.10) to increase in concentration in LP cows, and increased significantly (P < 0.05) in HP cows when compared with control NP cows. The expression pattern of conjugated ISG15 protein was also very similar to the ISG15 mRNA concentrations in endometrium from NP, LP, and HP cows. In PBMC, ISG15 mRNA concentrations in HP cows tended to be upregulated (P < 0.10) compared with LP and NP cows (Figure 6). This latter result is very encouraging as it suggests that there may be potential blood biomarkers detectable on d 16 of pregnancy in cattle; although, the search for blood biomarkers warrants more in-depth study of the days of early pregnancy.
Figure 5Relative concentration of endometrial ISG15 mRNA (A), conjugated ISG15 protein in a Western blot (B), and quantitation of ISG15 protein (C) in endometrium from lactating NP, LP, and HP dairy cows. NP = high-fertility nonpregnant cows; LP = low-fertility pregnant cows; HP = high-fertility pregnant cows. All ISG15 protein detected was conjugated to targeted proteins. The recombinant bovine ISG15 (rbISG15) blot indicates where free ISG15 migrates. Values represent mean ± SE. Means marked with different letters (a, b) differ (P < 0.05). *Represents a tendency to differ (P < 0.10). Mr = relative molecular mass.
It should also be noted that a different response was observed in ISG gene expression between the endometrium (Figure 5) and PBMC (Figure 6). The rationale for these responses is most likely that during elongation, the conceptus secretes IFNT, which attenuates the release of endometrial PGF2α in a paracrine manner to initiate maternal recognition of pregnancy; therefore, there is a local uterine effect of IFNT on stimulation of ISG15 gene expression in the endometrium (
). The LP group was similar and lacked significance compared with NP and HP in Figure 5; however, there does appear to be a linear relationship relative to stimulation by IFNT, particularly high levels of ISG15 in the HP group. An endocrine role for IFNT has been proposed in ruminants, where it may be involved with modulating maternal immune responses and conferring resistance of the CL to lytic effects of PGF2α (
). It must be noted that this is an endocrine effect of IFNT on PBMC, which are peripheral, meaning outside the uterus in the blood stream. Therefore, there are likely 2 reasons for a different observations in endometrium (Figure 5) compared with PBMC (Figure 6): (1) the effect of IFNT on PBMC is peripheral and likely of very low concentration in the blood relative to the uterine lumen and (2) a delayed response could be expected for a peripheral endocrine response due to the time is takes for IFNT to leave the uterine lumen via the blood stream and interact with the gene expression machinery within the cells of PBMC. In support of these statements,
Conceptus-induced changes in the gene expression of blood immune cells and the ultrasound-accessed luteal function in beef cattle: how early can we detect pregnancy?.
described maximal gene expression of ISG in PBMC in beef cows to be close to d 20, and the dairy cows in the current study were sampled on d 16 of pregnancy.
When the hypothesis of this study was developed, it was reasoned that if there is an increased number of active IFNT genes or increased IFNT released from the conceptus, then there should be more active cell signaling transduction downstream of the IFNT receptor and a greater increase in ISG. This would enhance maternal response to the conceptus and may also contribute to increased fertility in the HP group. Indeed, HP cows had increased ISG15 mRNA concentrations and isgylated proteins in endometrial tissues when compared with NP cows, whereas LP cows had ISG15 mRNA concentrations and conjugated proteins that were intermediate, but not different, from NP and HP cows. The lack of statistical difference in ISG15 concentrations between LP and HP endometrium may be caused by variable production of IFNT, as well as the timing of collection of conceptuses on d 16 of pregnancy and endometrial responses to IFNT. The production of IFNT by LP conceptuses may be attenuated or delayed compared with the HP conceptuses, as it may take some time for the endometrial ISG response to be observed on d 16. Because of this potential response, we considered waiting to collect tissues on d 17 of pregnancy, which may have been more aligned with anticipated changes in progesterone and estradiol hormones associated with onset of luteolysis and follicular recruitment in the LP and NP groups; however, when considering flushing uteri on d 17, the adhering and fragmenting peri-attached conceptuses would have limited our ability to collect intact conceptuses. Therefore, because of this concern, we decided the most optimal day of collection for this study was d 16.
reported that ISG15 was detectable in endometrium by d 15 of pregnancy in cows, increased to d 18, and then declined but was still detected until d 26 of pregnancy. In early responses to IFNT, almost all of the free ISG15 becomes conjugated to targeted proteins in the correct conditions (i.e., presence of conjugating enzymes) in a mechanism that is similar, but not identical, to ubiquitin. Isgylated proteins are typically thought to be regulated, rather than being targeted to degradation, as is the case for many ubiquitinylated proteins (
). In cells, ISG15 exists in free form (nonconjugated) at about 15 kDa on 1D-PAGE gels, whereas conjugated ISG15 can range in size from 40 kDa to >250 kDa (Figure 6;
). It is also possible in the present study that all free ISG15 was incorporated into conjugated forms in response to conceptus-derived IFNT during this early stage of pregnancy.
An endocrine role for IFNT has been proposed in ruminants where it may be involved with modulating maternal immune responses (
Expression of interferon (IFN)-stimulated genes in extrauterine tissues during early pregnancy in sheep is the consequence of endocrine IFN-tau release from the uterine vein..
). Herein, ISG15 mRNA concentrations tended to be upregulated in the PBMC of HP cows compared with NP and LP cows (Figure 6). This upregulation of ISG15 expression in maternal PBMC by conceptus-derived IFNT may also reflect upregulation of other ISG in maternal tissues peripheral to the uterus, such as the CL. These findings support previous studies that have described the endocrine action of IFNT and its upregulation of ISG in circulating blood cells. For example,
reported that ISG15 mRNA concentrations increased by d 16 of pregnancy and peaked by d 20 in blood cells. Furthermore, the abundance of ISG in PBMC increased from d 15 to 20 after AI and then dramatically decreased from d 20 to 22 of gestation in beef cattle (
Conceptus-induced changes in the gene expression of blood immune cells and the ultrasound-accessed luteal function in beef cattle: how early can we detect pregnancy?.
reported that ISG15 mRNA concentrations from peripheral blood leukocytes were not significantly upregulated until d 18 of pregnancy when ISG15 mRNA concentrations were significantly greater in PBMC from pregnant compared with nonpregnant cattle. For some multiparous cows, the increase in PBMC ISG were almost undetectable in response to early pregnancy, whereas the ISG response was very prominent in primiparous cows (
Measurement of interferon-tau (IFN-tau) stimulated gene expression in blood leukocytes for pregnancy diagnosis within 18–20d after insemination in dairy cattle..
). The endocrine effects of IFNT on both PBCM and the CL warrant further study as deciphering these mechanism-related relationships with ISG gene expression may enhance basic understanding that could lead to development of pregnancy test as early as d 16 to 20 in cattle.
Serum P4 and Estradiol Concentrations
High concentrations of P4 produced by the CL are necessary for creating a uterine environment that is conducive to embryo survival and the maintenance of pregnancy. The uterine epithelia secrete AA, glucose, cytokines, enzymes, growth factors, lymphokines, transport proteins for vitamins and minerals, and extracellular matrix molecules that constitute histotrophe, which supports the early expanding blastocyst (for review:
A mathematical model of the interaction between bovine blastocyst developmental stage and progesterone-stimulated uterine factors on differential embryonic development observed on day 15 of gestation..
). Unfortunately, d-7 blood was not collected for P4 analysis in the present study; however, on d 16 of pregnancy or the estrous cycle, serum P4 (overall average: 5.6 ± 0.2 ng/mL) and estradiol concentrations (overall average: 0.5 ± 0.1 pg/mL) did not differ in HP compared with LP cows. The values were within normal ranges for dairy cows at this stage of gestation or the estrous cycle (
Alteration of oestrous cycle length, ovarian function and oxytocin-induced release of prostaglandin F-2α by intrauterine and intramuscular administration of recombinant bovine interferon-α to cows..
Inverse associations observed between concentrations of IFNT and DGV-DPR and SPC suggested that combining multiple types of data may be useful in developing strategies to improve fertility on dairy farms. We clearly recognize that the number of Holstein cows examined in the present study was limiting; however, results support the hypothesis that sorting dairy cows for high DGV-DPR and low SPC improves IFNT production and endometrial signaling of maternal recognition of pregnancy on d 16. Specifically, although low-fertility cows may have similar fertilization rates as high-fertility cows, some low-fertility conceptuses may experience retardation in elongation, a decrease in the amount of IFNT produced by the conceptus, or both. Also, if IFNT does not fully stimulate maternal responses (such as endometrial or blood ISG) within the time window for recognition of pregnancy, then the embryo may enter apoptosis and die, potentially explaining the greater SPC in low-fertility Holstein cows.
ACKNOWLEDGMENTS
The authors thank Zella Brink for her animal assistance and technical expertise, Graham Opie and Michael Stine for their help with conceptus measurements, and Ann Hess, Department of Statistics, Colorado State University, for consultation on statistical analysis. This research was supported by Zoetis Inc. (Zoetis Study #: 7AGNN00000 EPM AC-168; CSU Award #: 002888-00002) and a USDA (WAAESD) regional project: W3112 Reproductive Performance in Ruminants.
LITERATURE CITED
Abdollahi-Arpanahi R.
Carvalho M.R.
Ribeiro E.S.
Peñagaricano F.
Association of lipid-related genes implicated in conceptus elongation with female fertility traits in dairy cattle..
The cross-species antiviral activities of different IFN-tau subtypes on bovine, murine, and human cells: contradictory evidence for therapeutic potential..
Synthesis and processing of ovine trophoblast protein-1 and bovine trophoblast protein-1, conceptus secretory proteins involved in the maternal recognition of pregnancy..
Deletion of the Isg15 gene results in up-regulation of decidual cell survival genes and down-regulation of adhesion genes: Implication for regulation by IL-1beta..
The relationships among sire’s predicted transmitting ability for daughter pregnancy rate and cow conception rate and daughter’s reproductive performance in Canadian Holstein cows..
Measurement of interferon-tau (IFN-tau) stimulated gene expression in blood leukocytes for pregnancy diagnosis within 18–20d after insemination in dairy cattle..
Molecular cloning and characterization of complementary deoxyribonucleic acids corresponding to bovine trophoblast protein-1: A comparison with ovine trophoblast protein-1 and bovine interferon-alpha II..
Differentially expressed genes in endometrium and corpus luteum of Holstein cows selected for high and low fertility are enriched for sequence variants associated with fertility..
Expression of interferon (IFN)-stimulated genes in extrauterine tissues during early pregnancy in sheep is the consequence of endocrine IFN-tau release from the uterine vein..
Alteration of oestrous cycle length, ovarian function and oxytocin-induced release of prostaglandin F-2α by intrauterine and intramuscular administration of recombinant bovine interferon-α to cows..
Conceptus-induced changes in the gene expression of blood immune cells and the ultrasound-accessed luteal function in beef cattle: how early can we detect pregnancy?.
A mathematical model of the interaction between bovine blastocyst developmental stage and progesterone-stimulated uterine factors on differential embryonic development observed on day 15 of gestation..
00020-9&id=gr1.jpg){kind=link}
00020-9&id=gr2.jpg){kind=link}
00020-9&id=gr3.jpg){kind=link}
00020-9&id=gr4.jpg){kind=link}
00020-9&id=gr5.jpg){kind=link}
00020-9&id=gr6.jpg){kind=link}