Early transcriptomic changes in peripheral blood 7 days post-embryo transfer in dairy cattle

A common goal of the dairy industry is to shorten the calving interval to reap several benefits associated with improved fertility. Early pregnancy detection is crucial to shorten this interval, allowing for prompt re-insemination of cows that failed to conceive after the first service. Currently, the industry lacks a method to accurately predict pregnancy within the first 3 weeks. The polypeptide cytokine interferon-tau (IFNT) is the primary signal for maternal recognition of pregnancy in ruminants. As IFNT is released from the early conceptus, it initiates a cascade of effects, including upregulation of interferon-stimulated genes (ISGs). Expression of ISGs can be detected in the peripheral blood. The present study aimed to characterize peripheral transcriptomic changes, including the ISGs, as early as d 7 post embryo transfer. A total of 170 Holstein heifers received in vitro-produced embryos. Whole blood was collected from these heifers within 24 h of the embryo transfer (d 0), d 7, and d 14 post embryo transfer. The heifers were divided into 2 groups, pregnant and nonpregnant, based on pregnancy diagnosis on d 28 via ultrasound. Total RNA was extracted from the peripheral blood of pregnant and nonpregnant heifers, pooled and sequenced. Expression analysis on d 7 heifers resulted in 13 significantly differentially expressed genes mostly related to innate immunity. Differential expression analysis comparing pregnant heifers on d 0 to the same heifers on d 14 showed 51 significantly differentially expressed genes. Eight genes were further quantified through RT-qPCR for biological validation. On d 7 post embryo transfer, mRNA transcriptions of EDN1 , CXCL3 , CCL4 , and IL1A were significantly upregulated in pregnant heifers (n = 14) compared with nonpregnant heifers (n = 14), with respective fold changes of 8.10, 18.12, 29.60, and 29.97. While


INTRODUCTION
In the dairy industry, early detection of cows that failed to conceive after the first service and their rapid re-insemination is crucial to minimize the days open, which reaps several economic benefits such as increasing milk yield and the number of calves, and reducing breeding costs and culling rates (Balhara et al., 2013;Temesgen et al., 2022).There are several direct and invasive methods of pregnancy detection, such as ultrasound and doppler ultrasound (Pohler et al., 2016), and other indirect, noninvasive methods, such as pregnancyassociated glycoproteins (PAGs) and preimplantation factor (PIF) (Zoli et al., 1992;Gajewski et al., 2009;Ramu et al., 2013;Wallace et al., 2015;Wonfor et al., 2020).Ultrasound, doppler ultrasound, PAGs, and PIF can detect pregnancy by d 28, 25, 24, and 20 of gestation, respectively (Balhara et al., 2013;Ramu et al., 2013;Pohler et al., 2016;Filho et al., 2020).However, a reliable method to accurately predict pregnancy outcomes within the first 3 weeks is still nonexistent.Investigating the detailed embryo-maternal interactions might be a promising approach to developing specific biomarkers for early pregnancy in cattle.
Maternal recognition and establishment of pregnancy require proper crosstalk between the early embryo and maternal innate immunity either locally at the embryo-maternal interface (Talukder et al., 2020) or systemically at the circulating immune cells for uterine receptivity and embryo tolerance (Oliveira et al., 2012;Gifford et al., 2019).The polypeptide hormone interferon-tau (IFNT) is the primary mediator for maternal recognition of pregnancy in ruminants (Hansen et al., 2017).It is secreted from the mononuclear trophectoderm of the conceptus around d 8, peaking at d 20 post-artificial insemination (AI) (Hirayama et al., 2014).IFNT specifically controls luteotropic and immune mechanisms, allowing for successful embryo implantation (Kowalczyk et al., 2021).It acts on the uterine endometrium to inhibit uterine pulse release of prostaglandin F2A (PGF2A), an inhibition critical for maintaining the corpus luteum (Basavaraja et al., 2021).Once released by the conceptus, IFNT binds to interferon α and β receptors (IFNR) subunits 1 and 2, initiating a cascade of effects, specifically through the classical type-1 interferon pathway (Toji et al., 2017;Basavaraja et al., 2019;Basavaraja et al., 2021).
Several studies have investigated peripheral blood transcriptomic changes during early pregnancy.It has been demonstrated that the expression of classical ISGs (ISG15, MX1, and MX2) in circulating immune cells of pregnant cows was upregulated compared with the nonpregnant animals on d 21 of gestation in embryotransferred Japanese Black cattle (Yoshino et al., 2023).Moreover, analyzing the transcriptomic profiles of peripheral immune cells from pregnant cows 21 d after embryo transfer revealed that over 600 genes were differentially expressed between pregnant and nonpregnant cows (De Los Santos et al., 2023).Those differentially expressed genes were related to immune tolerance, angiogenesis, inflammatory response, and cytokine secretion (De Los Santos et al., 2023).However, the earlier transcriptomic changes in circulating immune cells occurring within one week of embryo transfer have yet to be investigated.This study aimed to characterize the transcriptomic changes in peripheral blood leukocytes substantially earlier, on d 7 and 14 after embryo transfer.We hypothesized that immune-related genes are differentially expressed between pregnant and nonpregnant heifers and that these genes may provide insight into the mechanisms that promote immune tolerance to the embryo.These genes can potentially be biomarkers to predict pregnancy within the first 2 weeks after embryo transfer.

Ethics Statement
This study is exempt from the approval of the Institutional Animal Care and Use Committee as animals were not handled at our institution.Animals used in oocyte retrieval and embryo transfer are owned by ST Genetics, located in Vienna, Wisconsin, and Fond du Lac, Wisconsin.

Ovum Pickup (OPU)
Superovulated Holstein heifers housed together in one facility were used as oocyte donors.Ovum pickup (OPU) technique, or the transvaginal removal of the oocytes, was performed by guiding an ultrasound probe to locate and aspirate ovarian follicles (Callesen et al., 1987;Pieterse et al., 1988;Pieterse et al., 1991).OPUs were performed by the same veterinarian.The aspirated follicular fluid was searched for oocytes immediately.

In Vitro Production of Embryos
Embryos were generated through the in vitro fertilization (IVF) system at ST Genetics from May 2021 to August 2021.In brief, oocytes were aspirated from the follicles of donor heifers.The aspirated fluid was searched for cumulus-oocyte complexes (COCs) immediately following aspiration and scored as either viable or nonviable.Viable oocytes with at least one layer of cumulus cells were washed and placed into a maturation medium covered by mineral oil with cohorts grouped by the donor heifer.Following an 18-23 h incubation, the COCs were removed and placed into a fertiliza- Strangstalien et al.: Early transcriptomic changes… tion medium.There, complexes were stripped of the cumulus cells by pipetting.Conventional semen was prepared using the PureSperm 40/80 (https: / / nidacon .com/product/ puresperm -40 -80 -2x20 -ml/ ) method with slight adjustments to the protocol.Sperm was added to fresh fertilization media with the stripped oocytes.Gametes were co-incubated for 8 h.Presumptive zygotes were washed in a culture medium, transferred to a fresh culture medium, and cultured until d 7 of development.At d 7, embryos were morphologically assessed.Those that developed into the blastocyst stage were noted with a pronounced differentiation of the outer trophoblast layer and inner cell mass, as well as a prominent blastocele.Mid-blastocysts, expanded blastocysts, and hatched blastocysts of quality grades 1-2 (Bo and Mapletoft, 2013) were transferred into straws in preparation for the transfer into a recipient heifer.

Embryo Transfer
Estrus synchronized recipient cattle chosen for embryo transfer were all Holstein heifers (average 2 years old) housed in the same facility.The embryos produced were randomly assigned to a recipient heifer.A total of 170 heifers received embryos.Pregnancy was confirmed on d 28 via ultrasound.Heifers that experienced early embryonic loss before d 28 were recognized as nonpregnant.

Blood Collection
Within 24 h of the embryo transfer, blood samples were collected from the tail (coccygeal) vein in 9 mL EDTA blood collection tubes.This collection is denoted as Day 0 (D0).Blood was collected from the same heifers on d 7 (D7) and 14 (D14) post embryo transfer.The whole blood samples were centrifuged at 2,000 rpm for 20 min at 4°C within 24 h of the collection.The resulting buffy coat layer was transferred into fresh 1.5 mL microcentrifuge tubes.A 500 μL RNA-Later (ThermoFisher Scientific, Waltham, MA) was added to each buffy coat, and samples were stored at −20°C until further processing.

RNA Extraction and RNA Pooling
The buffy coat samples stored in RNA-Later were thawed at room temperature, then diluted 1:1 in PBS before RNA extraction (500 μL buffy coat: 500 μL PBS).The 1:1 diluted samples were spun at 5,000 x g for 5 min at room temperature to pellet the cells.The supernatant was discarded, and the pellet was resuspended in 1 mL TRIzol, mixed by pipetting, and incubated at room temperature for 5 min.Then, 200 μL chloroform was added to each sample and mixed by vigorously shaking.The samples were incubated at room temperature for 3 min and spun at 12,000 x g at 4°C for 15 min.Following this phase separation, the upper aqueous phase was transferred to a new 1.5 mL microcentrifuge tube.After, 500 μL of isopropanol was added, and samples were then incubated at 4°C for 10 min.The samples were spun at 12,000 x g at 4°C for 10 min.The supernatant was discarded and 1 mL freshly made 75% molecular grade ethanol was added.The samples were spun at 7,500 x g at 4°C for 5 min and the supernatant was discarded.The samples were dried at room temperature for 5 to 10 min to remove excess ethanol.The RNA was resuspended in 20 μL nucleasefree water (NFW).
The concentration of extracted RNA was quantified using the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, Delaware).The RNA samples were normalized to 60 ng/μL using NFW and pooled for RNA sequencing.Pooled RNA samples were subjected to a DNase treatment using Ambion DNA-free DNase Treatment (Thermo Fisher Scientific, Waltham, MA).Samples were pooled based on the blood collection day and the pregnancy outcome.Samples were also pooled by the sire of the embryo transferred ensuring equal representation of the sires in the pregnant and nonpregnant groups.Each pool contained 6 to 7 individual RNA samples from different heifers.A total of 64 heifers were included in the sequencing pools (Supplementary Figure 1).A total of 30 pools were sequenced, where 15 represented the pregnant heifers and 15 were the nonpregnant heifers.Of the 15 pools from the pregnant heifers, 5 consisted of D0 collections, 5 were D7 collections, and 5 were D14 collections.Likewise, of the 15 pools from the nonpregnant heifers, 5 consisted of D0 collections, 5 were D7 collections, and 5 were D14 collections.The bulls used to fertilize the embryos for transfer were equally included in the pregnant and nonpregnant pools.The pools of RNA from pregnant heifers are denoted as PR, while the pools of RNA samples from nonpregnant heifers are denoted as NP.

Illumina Sequencing and Data Analysis
DNase-treated pooled RNA samples were sequenced by Admera Health (South Plainfield, NJ).Before library preparation, RNA quantity was checked using the Qubit RNA HS assay (Thermo Fisher Scientific).RNA quality was analyzed using the Bioanalyzer 2100 Eukaryote Total RNA Nano (Agilent Technologies, CA).RNA library preparation was performed using the Quant-seq 3′ RNaseq FWD kit (Lexogen, Vienna, Austria).Library concentration and quality were checked Sequencing was done with Illumina HiSeq X, generating 100 single-end reads to make 30 million single-end reads per sample.For each sample, FastQC (http: / / www .bioinformatics.babraham.ac.uk/projects/ fastqc/ ) and Trimmomatic (Bolger et al., 2014) were used to check the quality of the raw reads and to trim adapter sequences and low-quality reads, respectively.Trimmed reads were then aligned to the bovine reference genome (Btau 5.0.1) using STAR (Dobin et al., 2013), including the "-quantMode GeneCounts" option to estimate gene counts.Only expressed genes with at least 15 counts in more than 5 samples were considered for each analysis.The R package "edgeR" was used to normalize the gene counts based on the trimmed mean of M-values (TMM) method (Robinson et al., 2010).Differential expression analyses were performed based on a negative binomial generalized linear model, including the sire groups as a blocking factor, using the edgeR package (Robinson et al., 2010).The statistical tests were corrected for multiple testing, so only genes with a false discovery rate (FDR) of less than 0.05 were considered significant (Benjamini and Hochberg, 1995).Fold difference in expression was calculated by the following: Fold Change = 2 log2FC , log 2 FC = log 2 fold change between the groups.The D0 and D14 pools were sequenced together in the same sequencing run.However, the D7 pools were sequenced within a different sequencing run, eliminating the opportunity for differential expression analysis of D0 and D7 since the source of variation, from the sequencing data or from biological differences, could not be determined.

Validation of RNA-Seq Using Quantitative Real-Time PCR (RT-qPCR)
Individual RNA samples, not included in the sequencing pools, were used for biological validation via RT-qPCR.Extracted RNAs were reverse transcribed into cDNA by the iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA).The reaction volume was 10 μL.Reverse transcriptase-PCR reactions were incubated at 25°C for 5 min, 46°C for 20 min, and then at 95°C for 1 min.
Gene expression analysis was done using the iTaq Universal SYBR Green Supermix (BioRad).cDNA was diluted 1:6 with NFW, then combined with iTaq Universal SYBR Green supermix and the forward and reverse primers (Table 1).Specific forward and reverse primers were obtained from Integrated DNA Technologies (IDT, Coralville, IA).The RT-qPCR reactions were carried out in the Bio-Rad iCycler real-time PCR machine with the following cycling conditions: 95°C incubation for 30 s followed by 40 cycles of 95°C for 5 s and 59°C for 30 s, ending with an incubation at 65°C for 5 s and an incubation at 95°C for 5 s.

Statistical Analysis of RT-qPCR
To evaluate the RNA expression, cycle threshold (Ct) values were normalized to the endogenous housekeeping β-actin gene.The β-actin gene control was selected using NormFinder, which calculates expression stability.The mean and the range of the fold change for each RNA were calculated as 2 -ΔΔCT using the estimated ΔΔCt value ± standard error following the 2 -ΔΔCT method by Livak and Schmittgen (2001).To compare the gene expression levels of pregnant and nonpregnant heifers on D7, data were analyzed with a t-test on the ΔCt values for each gene.To compare the gene expression levels of D0 and D14 collections from the same heifer, data were analyzed using a paired t-test on the ΔCt values for each gene.

Analysis of RNA-Seq Data
Differential expression analyses were performed in 5 different comparisons (Table 2).Analysis 1 compared the nonpregnant and pregnant heifers on D7, resulting in 13 differentially expressed genes (Supplementary Table 1).Analysis 2 compared the pregnant heifers on D0 to the same heifers on D14, where 51 genes were significantly differentially expressed (Supplementary Table 2).Differential expression analysis 3 compared the nonpregnant and pregnant animals on D14, showing zero differentially expressed genes.Differential expression analysis 4 compared the nonpregnant and pregnant groups on D0 or the day of the embryo transfer.This revealed zero differentially expressed genes.This comparison served as a control to ensure no significant expression differences between the 2 groups of heifers on the day of the embryo transfer.Differential expression analysis 5 compared the nonpregnant animals on D0 to D14, revealing 26 differentially expressed genes (Supplementary Table 3).This analysis also acted as a control to reduce the chance of identifying false positives or genes that may be differentially expressed on D14 unrelated to pregnancy.
To validate the differentially expressed genes obtained by RNA sequencing, the 8 genes selected were further tested through RT-qPCR in new samples not included in the sequencing pools.For the comparison of pregnant and nonpregnant heifers on D7, the RNA of 28 individual heifers (14 pregnant and 14 nonpregnant) was analyzed (Figure 1A).CCL4 expression was found to be upregulated in 13 of the 14 pregnant heifers compared with the nonpregnant heifers, with an average fold change of 29.60 (P < 0.001).IL1A expression was upregulated in 12 of the 14 pregnant heifers compared with the nonpregnant heifers, with an average fold change of 29.97 (P < 0.001).The expression of CXCL3 was upregulated in all 14 pregnant heifers compared with the nonpregnant heifers with an average fold change of 18.12 (P = 0.002).For EDN1, expression was upregulated in 11 of the 14 pregnant heifers compared with the nonpregnant heifers with an average fold change of 8.10 (P = 0.043).Thus, the RT-qPCR results of these genes reported upregulation in the pregnant animals, supporting the RNA-sequencing results.
For the comparison of pregnant heifers on D0 collection compared with the same heifers on D14 after embryo transfer, the RNA of 14 heifers was analyzed (Figure 1B).IFI6 was significantly upregulated in 13 of the 14 heifers on D14 compared with D0 with an average fold change of 3.08 (P = 0.002).MX2 expression was significantly upregulated in 13 of the 14 heifers on D14 compared with D0 with an average fold change of 2.59 (P < 0.001).OAS1Y expression was significantly upregulated in all 14 of the heifers on D14 compared with D0 with an average fold change of 3.89 (P < 0.001), and ISG15 expression was significantly upregulated in all 14 heifers on D14 compared with D0 with an average fold change of 5.09 (P < 0.001).The RT-qPCR results of all 4 genes supported the RNA-sequencing results for D14 heifers.

DISCUSSION
During the first 2 weeks after embryo transfer, the bovine blastocyst hatches escaping from the zona pellucida, undergoing trophectoderm growth, elongation, and implants into the endometrial tissue toward the end of the second week after embryo transfer (Thomas, 2013;Brooks et al., 2014;Spencer et al., 2016).Our in vivo study characterizes the transcriptomic profiles of peripheral leukocytes of pregnant heifers during these first 2 weeks after embryo transfer.Our novel results showed that several genes were differentially expressed in the peripheral leukocytes of pregnant heifers as early as D7 and D14 post embryo transfer.Most of those differentially expressed genes were immune-related genes and ISGs.These findings suggest that the bovine embryo interacts with the circulating immune cells as early as D7 after embryo transfer to induce a state of tolerance essential for embryo development and survival.
Our results showed that CCL4, IL1A, CXCL3, and EDN1, and mRNA expressions were significantly upregulated in pregnant heifers at D7 post embryo transfer.CCL4 and CXCL3 are chemokines that regulate several biological functions, including angiogenesis, embryogenesis, and initiation of an immune response (Borroni et al., 2008).They play a crucial role in the communication at the conceptus-maternal interface, allowing the allogenic embryo to survive and ultimately implant into the endometrium (Borroni et al., 2008;Du et al., 2014;Zhang et al., 2022).In porcine, CCL4 is vital in preparing the endometrium for pregnancy by activating implantation-related genes and recruiting macrophages to the implantation site (Borroni et al., 2008;Du et al., 2014).It has also been shown that CCL4 stimulates trophoblast migration during implantation (Hannan et al., 2006).CXCL3 is also reported to be associated with trophoblast migration, invasion, and proliferation (Gui et al., 2014;Wang et al., 2018).Studies have shown that both CCL4 and CXCL3 expressions were   abnormal in women with preeclampsia, indicating their importance for proper implantation (Gui et al., 2014;Garrido-Gomez et al., 2017;Wang et al., 2018;Ren et al., 2022).Interleukin IL1A is a cytokine, a major immune regulator during pregnancy, regulating uterine receptivity, implantation, and fetal development (Dutta and Sengupta, 2017;Malik and Kanneganti, 2018).EDN1 is a pleiotropic peptide that regulates vascularity and cell growth (Korth et al., 1999).It is known that inflammatory or immune factors in maternal circulation activate the endothelium, which increases the expression of EDN1 in the peripheral blood (George and Granger, 2012;Galaviz-Hernandez et al., 2016).
The ability of circulating immune cells of pregnant heifers to express such immune-related genes on D7 post embryo transfer (onset of initiation of uterine receptivity to implantation; Brooks et al., 2014), together with their above-mentioned roles in the induction of implantation suggests that circulating immune cells recognize and react to the immune network to ensure the embryo implants in the uterus.Additionally, the lack of expression of these genes in circulating immune cells potentially serves as an early signal for implantation failure and subsequent early embryonic death.However, it is still obscure how such a minuscule embryo resident in the uterus can induce systemic immune responses in circulating blood cells during the first week after embryo transfer.It has been shown that the pre-hatching bovine embryo induces anti-inflammatory responses in the endometrial epithelium and local immune cells at the embryo-maternal interface (Talukder et al., 2020;Rashid et al., 2018).Additionally, immune cells, particularly PMNs, can amplify and transfer embryo signals into the extrauterine environment via a cell-tocell communication mechanism (Fiorenza et al., 2021).Therefore, our results suggest that the presence of an embryo triggers anti-inflammatory responses locally at the uterine milieu and that the resident immune cells amplify and transfer embryo signals into the general circulation to tolerate this semi-allogeneic embryo.
Extending to the second week post embryo transfer (D14), our results showed that IFI6, MX2, OAS1Y, and ISG15 genes were significantly upregulated in pregnant heifers compared with their levels in the same heifers on D0.Notably, these genes were also not significantly upregulated in the D14 samples of nonpregnant animals.ISG15, MX2, and OAS1Y are known as classical ISGs, where IFNT binds to receptors in the uterine epithelium, activating the JAK-STAT pathway, ultimately leading to the expression of these genes (Platanias, 2005;Chaney et al., 2021).The D7 bovine embryo secretes minute amounts of IFNT into the uterine lumen, which triggers ISG expression in endometrial epithelium for embryo tolerance at the embryo-maternal interface (Ta-lukder et al., 2020;Rashid et al., 2018).Later, IFNT infiltrates into the circulation during the conceptus elongation period (d 12-15) and reaches peak production around d 20 of pregnancy (Hansen et al., 2017), when apparent effects on circulating immune cells have been reported (Green et al., 2010;Rocha et al., 2020;Ferraz et al., 2021) as confirmed with the present study's data.Similarly, Yoshino et al. found that expression levels of ISG15 and MX2 were considerably higher in peripheral blood leukocytes of pregnant cattle compared with the nonpregnant animals on D21 of pregnancy (Yoshino et al., 2020(Yoshino et al., , 2023)).Ferraz et al. found increased expression of ISG15 in the PBMCs of pregnant Holstein cows at D20 post-AI (Ferraz et al., 2021).OAS1 expression has also been reported to be upregulated during d 15 to 18 in pregnant heifers (Green et al., 2010).IFI6 is also an ISG that functions in immunomodulation (Sajid et al., 2021;Villamayor et al., 2023).It has been previously reported that IFI6 was upregulated in the peripheral circulation of pregnant women (Wright et al., 2023).Additionally, Rocha et al. (2020) identified IFI6 expression to be significantly upregulated in the peripheral bovine immune cells (PBMC and PMN) as early as D18 post-AI in beef heifers.Previously, it has been reported that ISGs were upregulated in the circulating PBMCs and PMNs as early as D7 of pregnancy in cows with multiple embryos (Talukder et al., 2019).Together, it seems that the ability of bovine embryos to trigger immune responses in circulating immune cells is dependent on embryo proliferation and subsequent levels of embryo-derived IFNT infiltrated into general circulation.
However, our differential expression analysis 3 revealed that ISGs were not differentially expressed in nonpregnant heifers compared with pregnant heifers on D14 post embryo transfer.Similarly, Yoshino et al. reported no differences in ISG expression in pregnant cows on D21 compared with nonpregnant cows that experienced late embryonic death (Yoshino et al., 2020;Yoshino et al., 2023).Therefore, we suggest that analyzing ISG expression in blood cells on D14 post embryo transfer serves as a potential biomarker for early detection of those animals that failed to conceive as well as the relative progress of gestation in pregnant animals.This is the first reported study to characterize circulating transcriptomic changes as early as one week after embryo transfer.However, one limitation of this study was the inability to compare the D7 collections to the D0 and D14 collections.The ability to analyze the pregnancy-related transcriptomic changes from heifers on the day of embryo transfer (D0) to one week later (D7) can provide additional information on how circulating leukocyte gene expression changes within the first week of embryonic development post-transfer.

CONCLUSIONS
This study provides in vivo evidence for the transcriptomic changes in circulating leukocytes that occur 1-2 weeks after embryo transfer in dairy cattle.Several immune-related genes and ISGs were activated during this period, which may provide clues to how the immune system accommodates the allogenic conceptus to survive and evade immune system rejection.To further investigate the diagnostic potentials of the genes identified in this study, additional studies are warranted to analyze the specificity and sensitivity of these biomarkers to predict early pregnancy in individual cows.
Strangstalien et al.:  Early transcriptomic changes… using the Tapestation High Sensitivity D1000 Assay (Agilent Technologies).RNA integrity (RIN) numbers of all pools were from 7.4 to 9.4.

Table 1 .
Strangstalien et al.:Early transcriptomic changes… RT-qPCR primer sequences of the differentially expressed genes *Housekeeping internal control.

Table 3 .
Differential expression analysis of RNA-seq data comparing PR D7 and NP D7 of genes selected for further validation 1 PR = pregnant, NP = nonpregnant, D7 = d 7 blood collection. 1 RNA-sequencing data from pooled RNA extracted from peripheral blood leukocytes of pregnant and nonpregnant heifers on day seven after embryo transfer.

Table 4 .
Differential expression analysis of RNA-seq data comparing PR D0 and PR 14 of genes selected for further validation 1