Effects of luteal blood flow on endometrial progesterone concentrations and gene expression in Japanese Black cows

Abstract

Luteal blood flow (LBF) is essential for progesterone (P₄) biosynthesis in the corpus luteum (CL) and affects bovine fertility. However, the mechanism by which LBF affects fertility remains unclear. This study was conducted to investigate the effects of LBF on endometrial P₄ concentrations and gene expression. Endometrial biopsies and blood samples were collected from 13 Japanese Black cows after ultrasound examination on Day 7 (Day 0 = day of estrus). Based on LBF, the cows were divided into low- (LV; n = 5), medium- (MV; n = 2), and high-vascularity (HV; n = 6) groups. Plasma and endometrial P₄ concentrations were measured using enzyme immunoassays. RNA sequencing was performed to compare the endometrial gene expression profiles from three cows in each of the LV and HV groups. Reverse transcription-quantitative PCR was performed for genes selected from the differentially expressed genes (DEGs), P₄ receptors (PGR, PGRMC1, and PGRMC2), and P₄-regulated genes (ANPEP, DGAT2, DKK1, and LTF). No differences were observed in plasma or endometrial P₄ concentrations between the HV and LV groups. CCN3 was identified as a DEG between the HV and LV groups and was upregulated in the HV group. Compared to those of the LV group, the HV group exhibited higher CCN3 and PGR mRNA expression levels and lower ANPEP, DGAT2, and DKK1 mRNA expression levels. In conclusion, LBF affects endometrial gene expression without changing plasma or endometrial P₄ concentrations on Day 7.

Angiogenesis in the corpus luteum (CL) is crucial for the CL development and maintenance of luteal function in mammals [1, 2]. After ovulation, the vascular network surrounding the CL forms rapidly during CL growth [1, 2]. In the mature CL, luteal cells are in contact with one or more capillaries, making the CL a highly vascularized organ in the body [1, 2]. Endothelial cells and pericytes, which form and stabilize blood vessels, account for ˃50% of the total cell population in mature bovine CL [3]. Neovascularization is pivotal for the biosynthesis, secretion, and systemic release of progesterone (P₄), which is required for the establishment and maintenance of gestation [1, 2]. Thus, impaired luteal vascularization may be associated with inadequate luteal function [1, 2].

Recently, color Doppler ultrasonography has enabled the noninvasive assessment of luteal blood flow (LBF) in cattle [4]. Cyclic changes in LBF have been documented, with increases in LBF during the early and mid-luteal phases and decreases during luteal regression, coinciding with changes in circulating P₄ concentrations [4]. Numerous recent studies have focused on using LBF for early pregnancy diagnosis [5, 6] and recipient selection for embryo transfer (ET) [6,7,8]. From day 7 after artificial insemination (AI), pregnant cows maintain a higher LBF than that of non-pregnant cows [9]. LBF is a reliable diagnostic marker for detecting non-pregnant cattle 17–21 days post-AI [5, 6]. Furthermore, the conception rate of recipients with high LBF on the day of ET (7 days after estrus) is higher than that of recipients with low LBF [6, 7]. Therefore, evaluation of LBF at the time of ET is a reliable method for selecting recipients with high fertility. Although high LBF after AI or at ET evidently has a positive effect on fertility, the underlying mechanism remains unclear.

Successful pregnancy establishment requires a suitable uterine environment for embryo growth and implantation, in addition to the presence of a functional CL [10]. Embryo survival and growth before implantation in the endometrium depend on histotrophs supplied by the endometrium (particularly the endometrial glands) to the uterine luminal fluid (ULF) [10, 11]. The composition of the ULF is affected by several factors, such as the endometrial transcriptome and circulating P₄ concentrations; therefore, variations in these factors are associated with uterine receptivity [10,11,12]. In addition, P₄ indirectly contributes to successful embryonic growth and elongation by modifying endometrial gene expression and ULF composition [11, 12]. Therefore, we hypothesized that LBF levels alter the endometrial P₄ concentrations and/or gene expression, resulting in different fertility. However, the relationship between LBF and endometrial P₄ concentrations or gene expression remains unclear.

This study aimed to elucidate the effects of LBF on endometrial P₄ concentrations and gene expression on Day 7 of the estrous cycle in Japanese Black (JB) cows.

Materials and Methods

This study was approved by the Iwate University Laboratory Animal Care and Use Committee (approval number: A202044) and the Animal Ethics Committee of the Tohoku Agricultural Research Center, NARO (approval number: 20C116TARC), and was conducted under ethical principles and animal welfare standards.

Animals

A total of 90 non-inseminated cyclic JB cows (5.9 ± 2.9 years [mean ± standard deviation]; 4.2 ± 2.3 times of calving; 92.0 ± 52.3 days postpartum) from two farms in Miyagi (farm 1) and Iwate Prefectures (farm 2), Japan, were used in this experiment. The cows were housed in free-barn facilities or semi-loose housing systems, fed twice daily, and provided free access to water and mineralized salt. Estrus was confirmed by observing standing behavior and ultrasonographic examination of the ovaries. The day of estrus was defined as Day 0.

Ultrasound examination

Ultrasound examinations of the ovaries were performed on Day 7 by a single operator using portable ultrasound equipment with a 10-MHz linear rectal transducer (MyLab One Vet; Esaote SpA, Genoa, Italy). The CL and central cavity areas (CCA) were measured at their maximum diameter using B-mode ultrasonography. The LBF area (BFA) was evaluated using the color-flow Doppler mapping (CFM) mode as previously described [5]. The CFM settings were standardized for all examinations as follows: 7.5 MHz B-mode frequency, 90% B-mode gain, 6.6 MHz CFM frequency, 45% CFM gain, and 1.5 kHz pulse repetition frequency. The CL area, CCA (if present), CL tissue area (calculated by subtracting the CCA from the CL area), and BFA were quantified using ImageJ software (v1.54c; National Institutes of Health, Bethesda, MD, USA) [5, 7]. Data analyses were performed using the mean values of three recorded images. The 90 cows were divided into tertiles based on BFA; cows in the 1st, 2nd, and 3rd tertiles were defined as low (LV), medium (MV), and high vascularity (HV) groups, respectively.

Blood and endometrial tissue collection

Blood samples from all JB cows were collected by jugular venipuncture into vacuum tubes containing sodium heparin (TERUMO, Tokyo, Japan) immediately before the ultrasound examination on Day 7. After collection, the blood samples were immediately placed on ice and centrifuged at 1,450 × g for 60 min at 4°C. The plasma was stored at −30°C until the P₄ assay.

Endometrial samples were collected transcervically from the tip of the uterine horn ipsilateral to the CL (approximately 5 cm from the uterotubal junction) using a sterile endometrial biopsy device (Fujihira Kogyo Co., Ltd., Tokyo, Japan) from 13 of the 90 cows after ultrasound examination. These 13 cows were randomly selected from those maintained on farm 2. Before this procedure, the cows received caudal epidural anesthesia using 3 ml of 2% lidocaine (Xylocaine® Injection 2%, Sandoz K.K., Tokyo, Japan), and the genitalia were washed with benzalkonium chloride solutions (Osvan®, Fuji Pharma CO., Ltd., Tokyo, Japan). After collection, the samples were immediately divided into small pieces of approximately 30 mg, rinsed with physiological saline, and blotted using sterile paper towels. Each sample was placed in a cryotube, frozen in liquid nitrogen, and stored at −80°C until RNA extraction and the P₄ assay were performed.

These 13 cows were classified into LV, MV, and HV groups based on the BFA tertiles derived from the 90 cows.

Plasma and endometrial tissue P₄ assay

Endometrial tissue samples were homogenized on ice using a disposable homogenizer (Bio Masher® II, Nippi, Inc., Tokyo, Japan) in phosphate-buffered saline (PBS) at a ratio of 0.1 g tissue/ml, as previously described [13]. After the homogenates were centrifuged at 1,450 × g for 15 min at 4°C, the supernatants were collected. Extraction of P₄ from the supernatants and plasma, as well as all enzyme immunoassay (EIA) procedures were performed following the method described by Takenouchi et al. [14]. Briefly, P₄ was extracted twice with diethyl ether. The ether phase was transferred into a glass tube and evaporated under a stream of nitrogen gas. The residues were dissolved in PBS containing 1% (w/v) bovine serum albumin (BSA, Roche Diagnostics, Indianapolis, IN, USA). P₄ levels were measured using a double antibody EIA. Anti-rabbit IgG [H+L] goat serum (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA), was coated onto a 96-well microtiter plate (Thermo Fisher Scientific, Waltham, MA, USA) at 100 ng/well in 0.1 M carbonate buffer, serving as the priming antiserum. A rabbit antiserum raised against 11α-hydroxy-P₄-hemisuccinate-BSA (MAKOR Chemicals, Jerusalem, Israel) was diluted to 1:800,000 in PBS containing 1% (w/v) BSA. P₄-3(O-carboxymethyl)oxime-horseradish peroxidase (Sigma Chemicals Co., St. Louis, MO, USA), at 0.6 ng/well in PBS containing 1% (w/v) BSA, was used as the steroid-enzyme conjugate. The P₄ standard curve ranged from 0.02 to 10 ng/ml, with a median effective dose of 0.19 ng/ml. The intra- and inter-assay coefficients of variation were 4.6% and 6.6%, respectively.

RNA extraction and sequencing

Total RNA was extracted from approximately 30 mg of endometrial tissue using either TRIzol reagent (Invitrogen, Waltham, MA, USA) or the RNeasy Mini Kit (Qiagen, Hilden, Germany) with on-column DNase treatment (RNase-Free DNase Set, Qiagen) according to the manufacturer’s instructions. The quantity and purity of RNA samples were assessed using the Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific), and RNA integrity was confirmed to with an RNA integrity number (RIN) ≥ 7.0 using the RNA 6000 Pico kit (Agilent Technologies, Santa Clara, CA, USA) on an Agilent 2100 Bioanalyzer (Agilent Technologies). cDNA library preparation, RNA sequencing (RNA-seq), and analysis were performed by Rhelixa Inc. (Tokyo, Japan) using three samples each from the LV and HV groups. Poly-A mRNA was isolated from total RNA and fragmented using the NEBNext® Poly(A) mRNA Magnetic Isolation Module (New England Biolabs®, Ipswich, MA, USA). Subsequently, the mRNA fragmented was reverse-transcribed into cDNA, and strand-specific cDNA libraries were prepared according to the NEBNext® Ultra™ II Directional RNA Library Prep Kit protocol (New England Biolabs®). Paired-end sequencing was performed using the Illumina® NovaSeq 6000 platform (Illumina, San Diego, CA, USA) with a read length of 2 × 150 bp (PE150). The quality of raw sequencing reads was estimated using FastQC software (v0.11.7). Trimmomatic software (v0.38) was used to filter out sequencing adapters, low-quality reads, and short reads based on quality scores calculated using FastQC software. The clean and high-quality reads were subsequently mapped to the Bos taurus reference genome (Bos taurus ARS-UCD1.2) using HISAT2 software (v2.1.0), and the number of reads mapped to each gene was counted using the FeatureCounts software (v1.6.3). Differentially expressed genes (DEGs) between the LV and HV groups were identified using the DESeq2 package (v1.24.0). P-values were adjusted using the Benjamini‒Hochberg false discovery rate (FDR) multiple testing correction. Genes with an FDR corrected P-value < 0.05 and |log₂ fold change| > 1 were considered significantly DEGs. The RNA-seq data have been deposited in the DNA Data Bank of Japan (DDBJ) BioProject database under accession number PRJDB35471.

Reverse transcription-quantitative PCR (RT-qPCR)

One microgram of total RNA from each sample was reverse-transcribed into cDNA using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturer’s instructions. The resulted cDNA was diluted 1:49 with TE buffer (NIPPON GENE, Tokyo, Japan), and 5 μl of this dilution was used for each RT-qPCR reaction. RT-qPCR analysis was performed on the genes selected from the DEGs, P₄ receptors (nuclear P₄ receptor [PGR], P₄ receptor membrane component 1 [PGRMC1], and P₄ receptor membrane component 2 [PGRMC2]), as well as other candidate genes whose expressions is altered by circulating P₄ concentrations and are presumed to influence the uterine environment. These P₄-regulated genes included membrane alanyl aminopeptidase (ANPEP), diacylglycerol O-acyltransferase 2 (DGAT2), Dickkopf homolog 1 (DKK1), and lactotransferrin (LTF). The primers for the target genes used in RT-qPCR were designed using the Primer Express™ software v3.0 (Applied Biosystems) and are listed in Table 1. RT-qPCR was performed using an ABI 7300 real-time PCR system (Applied Biosystems) with PowerUp SYBR Green Master Mix (Applied Biosystems). The final 20 μl PCR reaction mixture, consisted of 10 μl PowerUp SYBR Green Master Mix, 5 μl cDNA, 1.5 μl of each primer (5 μM), and 2.0 μl of RNase/DNase-free water (Invitrogen). The RT-qPCR conditions were as follows: 50°C for 2 min and 95°C for 2 min, followed by 40 cycles at 95°C for 15 sec and 60°C for 1 min. All PCR reactions were performed in duplicate, and dissociation curve analysis was performed to confirm the specificity of amplification. The mRNA copy number of each gene was determined using the standard curve method. The expression stability of candidate internal reference genes (glyceraldehyde 3-phosphate dehydrogenase [GAPDH], ribosomal protein L27 [RPL 27], and beta-actin [ACTB]) was evaluated using NormFinder (v0.935) [15]. These results confirmed that GAPDH was the most stable gene and could be used as a reference gene. Therefore, these values were normalized to those of GAPDH.

Table 1. Primers used in RT-qPCR.

Gene symbol	Accession No.	Sequence (5′–3′)
CCN family members
CCN1/CYR61	NM_001034340.2	F: TGTGGACAGCCGGTGTACAG
		R: GGACTTCTTGGTCTTGCTGCAT
CCN2/CTGF	NM_174030.2	F: CGTGTGCACCGCTAAAGATG
		R: TCCGCTCTGGTACACAGTTCCT
CCN3/NOV	NM_001102382.1	F: GTAGCGATGGCCGATGCT
		R: TGGAACTCCACCTGGATGGT
CCN4/WISP-1	XM_005215320.4	F: CTACAAGTCCAAGACCATCAGCAT
		R: CAGATTGCAGAAGCAAGCGTTA
CCN5/WISP-2	NM_001102176.1	F: ATCTCCACGTCTGTGACCCC
		R: TCATCTTCTCCCCAGAGGCAC
CCN6/WISP-3	XM_010808451.3	F: CCTAGGTACGAGACCGGAGTGT
		R: GCACCCCGTTGAACTCACAT
Progesterone receptors
PGRMC1	NM_001075133.1	F: CGCTTCGACGGCGTACA
		R: CGAACACCTTGCCATTGATG
PGRMC2	NM_001099060.1	F: CCTGCTTGCGGTCAATGG
		R: CGCCGGGCCGTAGAA
PGR	NM_001205356.1	F: CGGGCACTGAGTGTTGAATTT
		R: TCTTGGGTAACTGTGCAGCAA
Progesterone-regulated genes
ANPEP	NM_001075144.1	F: CCAGCCGCAAGAACACTACTG
		R: AGTCGTCTGCTGCGGCTTT
DGAT2	NM_205793.2	F: TCATTGCCGTGCTCTACTTCA
		R: GACCTCCTGCCACCTTTCTTG
DKK1	NM_001205544.1	F: GGATGGGTACTCCAGAAGAACTACA
		R: GCACAGTCTGATGAGCGAAGAC
LTF	NM_180998.2	F: CAGCTGTGTTCCCTGCATTG
		R: CCCCCTTGCACAGTTGACA
Internal reference genes
ACTB	NM_173979.3	F: TCCTGACGGGCAGGTCAT
		R: GTGAATGCCGCAGGATTCC
RPL27	NM_001034051.2	F: CGCTACTCCGGACGCAAA
		R: GTCTGAGGTGCCGTCATCAA
GAPDH	NM_001034034.2	F: AAGGCCATCACCATCTTCCA
		R: CCACCACATACTCAGCACCAGCAT

F, forward; R, reverse; RT-qPCR, reverse transcription-quantitative polymerase chain reaction.

Data analysis

Statistical analyses were performed using BellCurve for Excel (Social Survey Research Information Co., Ltd., Tokyo, Japan) and R software (v4.0.4). Fertility differed considerably between the HV and LV groups, whereas the conception rate of ET in the MV group does not significantly differ from that in either the HV or LV group [16]. We compared all the results between the HV and LV groups to clarify the impact of LBF on CL area, CCA, CL tissue area, BFA, plasma/endometrial P₄ concentrations, and endometrial gene expression. For cows from which endometrial tissue was collected, we analyzed differences in the mean values of CL area, CCA (for samples with a central cavity), CL tissue area, BFA, and plasma/endometrial P₄ concentrations between the LV and HV groups using Student’s t-test, after confirming normality and homogeneity. None of the gene expression levels were normally distributed according to the Shapiro–Wilk test. Therefore, the differences in gene expression between the LV and HV groups were analyzed using the Brunner‒Munzel test. A P-value < 0.05 was considered statistically significant, whereas 0.05 < P < 0.10 was considered a trend toward significance.

Results

Animals and grouping

To establish criteria for identifying low or high luteal vascularity, 90 cows were divided into tertiles based on BFA: 1st tertile (LV), ≤ 51.29 mm²; 2nd tertile (MV), 52.71–70.34 mm²; and 3rd tertile (HV), ≥ 70.74 mm² (Table 2). Based on these criteria, the 13 cows from which endometrial tissue was collected were classified into LV (n = 5), MV (n = 2), or HV (n = 6) groups.

Table 2. Features of the corpus luteum (CL) and plasma progesterone (P₄) concentrations in groups divided according to BFA tertiles on Day 7.

Item		BFA
		1st tertile	2nd tertile	3rd tertile
	All (n = 90)	(n = 30)	(n = 30)	(n = 30)
CL area (mm2)	356.52 (250.79–559.89)	354.84 (250.79–508.44)	358.03 (266.03–559.89)	357.90 (262.56–513.40)
CCA (mm2) 1)	22.64 (2.96–194.91)	22.79 (6.65–178.91)	28.14 (2.96–194.91)	22.49 (4.02–122.17)
CL tissue area (mm2)	333.70 (207.40–513.40)	316.35 (235.67–440.22)	331.59 (207.40–473.31)	343.50 (238.26–513.40)
BFA (mm2)	61.44 (26.79–124.81)	42.08 (26.79–51.29)	61.44 (52.71–70.34)	81.02(70.74–124.81)
Plasma P4 concentrations (ng/ml)	2.42 (1.25–4.99)	2.31 (1.25–4.39)	2.38 (1.39–4.99)	2.48 (1.28–4.28)

All the data are represented as medians with ranges. BFA, blood flow area; CCA, central cavity area. ¹⁾ Median number of cows with a central cavity in each group (all, n = 60; 1st tertile, n = 17; 2nd tertile, n = 24; 3rd tertile, n = 19).

Morphological characteristics, LBF, and plasma/endometrial tissue P₄ concentrations in the LV and HV groups

Morphological characteristics, including CL area, CCA, and CL tissue area, were similar among the groups (P > 0.1; Table 3). BFA was significantly greater in the HV group than in the LV group (P = 0.001; Table 3). The plasma and endometrial tissue P₄ concentrations did not differ between the LV and HV groups (P > 0.1; Table 3). Among the six cows selected for the RNA-seq analysis, no differences were observed in any items other than BFA (LV group, n = 3; HV group, n = 3; Supplementary Table 1).

Table 3. Features of the corpus luteum (CL) and plasma/endometrial tissue progesterone (P₄) concentrations in the low (LV) and high vascularity (HV) groups.

Item	LV (n = 5)	HV (n = 6)	P-value
CL area (mm2)	359.52 ± 26.72	368.24 ± 20.08	0.796
CCA (mm2) 1)	19.01 ± 4.23	31.23 ± 15.32	0.544
CL tissue area (mm2)	348.11 ± 23.24	352.62 ± 27.78	0.906
BFA (mm2)	40.09 ± 2.50	84.71 ± 8.17	0.001
P4 concentrations
Plasma (ng/ml)	2.56 ± 0.25	2.57 ± 0.13	0.977
Endometrial tissue (ng/g)	7.70 ± 1.05	6.54 ± 0.52	0.317

All the data are represented as the means ± SEMs. BFA, blood flow area; CCA, central cavity area. ¹⁾ Mean of three cows with a central cavity in the LV (n = 3) and HV (n = 3) groups.

Quality control of RNA-seq data and read alignment

After quality control, including trimming of sequencing adapters, low-quality reads, and short reads, we obtained an average of 32.9 million clean reads per sample (˃ 98.54% of raw reads; Supplementary Table 2). The clean reads for each sample were mapped to the Bos taurus reference genome, and the mapping rates ranged from 98.04% to 98.28%. In total, 13,727 genes were annotated from the Bos taurus reference genome in the endometrial tissue. All subsequent analyses were based on high-quality data.

DEGs between the LV and HV groups

The DEGs between the LV (n = 3) and HV (n = 3) groups, identified using the DESeq2 package, are presented in a volcano plot (Fig. 1). In the comparison between the LV and HV groups, CCN3 exhibited a significant difference in expression levels with a threshold of P-value <0.05, |log₂ fold change| > 1, and its expression was upregulated in the HV group.

Fig. 1.

Volcano plot of differential gene expression between the LV and HV groups. Red point represents the upregulated gene in the HV group. Gray points represent the genes whose expression levels did not differ between the two groups. Numbers in parentheses indicate the number of genes. LV, low vascularity; HV, high vascularity.

RT-qPCR validation and mRNA expression of the candidate genes

CCN3, which was identified as a DEG by RNA-seq analysis, was validated by RT-qPCR. Additionally, we determined the mRNA expression levels of other CCN family members, P₄ receptors, and four P₄-regulated genes. The expression levels of these genes are presented in Fig. 2 and Fig. 3. The expression levels of CCN3 (Fig. 2C) and PGR (Fig. 3A) were higher in the HV group than those in the LV group. ANPEP (Fig. 3D), DGAT2 (Fig. 3E), and DKK1 (Fig. 3F) mRNA levels were lower in the HV group than those in the LV group. However, no differences were observed in CCN1 (Fig. 2A), CCN2 (Fig. 2B), CCN4 (Fig. 2D), PGRMC1 (Fig. 3B), PGRMC2 (Fig. 3C), or LTF (Fig. 3G) mRNA expression between the groups. The expression of CCN5 and CCN6 mRNA was nearly undetectable in both groups.

Fig. 2.

Relative mRNA expression levels of CCN family genes in the endometrium. Relative expression levels of CCN1 (A), CCN2 (B), CCN3 (C), and CCN4 (D) were normalized to those of GAPDH. Black circles represent the low vascularity group (LV, n = 5). Open circles represent the high vascularity group (HV, n = 6). The open and black triangles represent the mean of each group. ** P < 0.01.

Fig. 3.

Relative mRNA expression levels of progesterone receptor and progesterone-regulated genes in the endometrium. The relative expression levels of PGR (A), PGRMC1 (B), PGRMC2 (C), ANPEP (D), DGAT2 (E), DKK1 (F), and LTF (G) were normalized to those of GAPDH. Black circles represent the low vascularity group (LV, n = 5). Open circles represent the high vascularity group (HV, n = 6). The open and black triangles represent the mean of each group. † P < 0.1, * P < 0.05, ** P < 0.01.

Discussion

The major findings of this study are as follows: (i) plasma and endometrial tissue P₄ concentrations did not differ between the LV and HV groups; (ii) only CCN3 was identified as a DEG among the groups and was upregulated in the HV group; and (iii) LBF affected the mRNA expression of CCN3, PGR, ANPEP, DGAT2, and DKK1 in the endometrium.

Although circulating P₄ concentrations are correlated with LBF throughout the estrous cycle [17], during the CL development period (Days 7–8), P₄ levels depend more on CL size rather than on LBF [18, 19]. Consistent with these findings, plasma P₄ concentrations on Day 7 did not differ between the LV and HV groups in this study, due to the similar CL sizes (area and tissue area). Notably, endometrial tissue P₄ concentrations vary between different parts of the same uterine horn (tip vs. middle) as well as between the ipsilateral and contralateral uterine horns of the ovary containing a CL [13, 20]. In addition, P₄ is locally delivered from the ovary to the ipsilateral uterine horn through a countercurrent transfer mechanism from the ovarian vein to the ovarian artery, which is connected to the uterine artery [21]. Therefore, we hypothesized that LBF may affect P₄ concentrations locally in the uterine horn ipsilateral to the CL rather than systemically. However, associations between LBF and P₄ concentrations in the ULF or endometrial tissue, which are collected from the horn ipsilateral to the CL, have not been reported. In this study, no difference in endometrial tissue P₄ concentrations was observed between the LV and HV groups. These findings indicate that neither systemic nor local P₄ is involved in the mechanism by which LBF affects fertility in cows.

To investigate the effects of LBF on endometrial gene expression, we compared the RNA-seq data from endometrial tissues between the LV and HV groups. Only CCN3 was identified as a DEG. The CCN3 mRNA expression level was high in the HV group, whereas the expression levels of other CCN family members did not differ between the two groups. CCN3, additionally referred to as nephroblastoma overexpressed (NOV), is a member of the CCN family, including cysteine-rich 61(CYR61/CCN1), connective tissue growth factor (CTGF/CCN2), Wnt-induced secreted protein-1 (WISP-1/CCN4), WISP-2/CCN5, and WISP-3/CCN6 [22, 23]. The CCN family members are matricellular proteins involved in various biological processes that modulate cell-extracellular matrix (ECM) signaling [22, 23]. These proteins regulate cell differentiation, proliferation, adhesion, migration, apoptosis, angiogenesis, and ECM production in various organs as well as promote tumor growth and fibrosis [22, 23]. In the reproductive tract, CCN family members play important roles in angiogenesis, embryo adhesion, implantation, placentation, and embryogenesis across species including mice [24], pigs [25], cattle [26], and humans [27]. CCN3 functions as a potent angiogenic agent by promoting endothelial cell proliferation, adhesion, and migration [28]. CCN3 is important for placental angiogenesis and embryonic vascular development in both mice and humans, with reduced placental and serum CCN3 levels being associated with fetal death and preeclampsia [24, 27, 28]. Therefore, the increased CCN3 mRNA expression observed in the HV group may enhance endometrial and embryonic angiogenesis and improve fertility. CCN family members regulate the production and activity of angiogenic factors such as vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), placental growth factor (PIGF), and insulin-like growth factor 1 (IGF1) [27]. In addition, transforming growth factor-β3 (TGF-β3) increases the expression of CCN3 protein in human trophoblasts [29]. TGF-β acts on vascular endothelial cells, vascular smooth muscle cells, and pericytes to regulate angiogenesis [30]. Therefore, the increased CCN3 in the HV group may interact with various angiogenic factors to promote endometrial angiogenesis; however, further studies are required to confirm these hypotheses. In this study, apart from CNN3, no differences were observed in the expression of other CCN family members between the HV and LV groups. Additionally, the expressions of CCN5 and CCN6 were nearly undetectable in both groups. CCN1 and CNN2 are involved in endometrial angiogenesis, placental and embryonic vascularization, and ECM remodeling during embryo implantation [23]. Mice lacking these genes experience fetal mortality owing to impaired placental vascularization and skeletal malformations [24, 28]. Furthermore, uterine expressions of CCN1 and CCN2 are regulated by the steroid hormones such as estrogen and P₄ [22, 23]. In cattle, CCN2 mRNA expression in the endometrium increases from day 13 of gestation and is induced by P₄ and interferon-tau, suggesting that it is involved in embryonic attachment and conceptus trophectoderm growth [26, 31]. Additionally, exogenous P₄ supplementation increases the expression of CCN2 mRNA in cattle [31]. In this study, no difference was observed in the plasma or endometrial tissue P₄ concentrations between the two groups, which may explain the lack of differences in CCN2 mRNA expression. Studies on the role of CCN4–CCN6 in the endometrium and placenta are limited. CCN4–CCN6 have been suggested to be involved in cell adhesion to the endometrium, placental development, regulation of vascular structure, and remodeling of glandular and endometrial cells in pigs [25] and rats [32]. Therefore, it is presumed that the CCN family members positively affect the maintenance of normal endometrial function and pregnancy in cattle. The results of this study revealed that among CCN family members, only CCN3 mRNA expression was affected by LBF, representing the first such observation reported in cattle.

Nuclear progesterone receptor (PGR) and membrane progesterone receptors are present in the bovine uterus [13, 33, 34]. P₄ regulates uterine function through a genomic mechanism via PGRs, which are expressed in the endometrium during the early- and mid-luteal phases, thereby exerting positive effects on pregnancy establishment [33, 35,36,37,38,39]. Two isoforms of PGR (PGR-A and PGR-B) are encoded by the same gene (PGR) [33, 35,36,37]. In this study, PGR mRNA expression was higher in the HV group than in the LV group. Therefore, the action of P₄ through PGRs may have been enhanced in the HV group, thereby promoting the establishment of a uterine environment conducive to embryonic and conceptus development. Temporal changes in PGR mRNA expression have been demonstrated, with high expression from estrus through the early luteal phase, followed by a decline beginning on Days 8–10 due to continuous exposure to P_4, reaching baseline levels by Day 13 [35, 38]. Furthermore, PGR mRNA levels in pregnant cows on day 13 post-AI were lower than those in non-pregnant cows. This preimplantation downregulation of PGR is a prerequisite for maternal recognition of pregnancy and conceptus development [35, 37, 38]. Circulating P₄ concentrations affect endometrial PGR mRNA expression, and manipulation of circulating P₄ concentrations during the early luteal phase results in differences in PGR mRNA expression [35, 40, 41]. PGR mRNA expression on Day 7 increased in the low P₄ model induced by multiple administrations of PGF_2α [41] and decreased in the high P₄ model caused by exogenous supplementation with P₄ during the early luteal phase (Days 3–7) [35, 40], compared with that in cows with normal P₄ concentrations. Therefore, an early increase in circulating P₄ concentrations advances the normal temporal changes that initially occur before implantation, that is, the loss of PGR in the endometrium, whereas a decrease in circulating P₄ concentrations delays this change. In contrast to these studies, no difference was observed in plasma or endometrial tissue P₄ concentrations between the two groups on Day 7; however, PGR mRNA expression was high in the HV group. The LBF of the low P₄ model was reduced compared to that of normal cows [41], and the CL area or LBF was decreased in the high P₄ model [42, 43]. In this study, we compared two groups with similar CL areas and circulating P₄ concentrations but different LBFs, which may have led to different results from those of previous studies.

Membrane progesterone receptors include progesterone receptor membrane components (PGRMC1 and PGRMC2) and membrane progestin receptors (mPRs; mPRα, mPRβ, and mPRγ). These receptors have been proposed to mediate the non-genomic effects of P₄, such as regulating intracellular signaling pathways and promoting glycogenolysis in the endometrial epithelium [33, 34]; however, the specific mechanisms underlying their functions remain unclear. In contrast to PGR, PGRMC1 and PGRMC2 mRNA expression did not change throughout the estrous cycle despite the fluctuations in plasma and endometrial P₄ concentrations; however, their expression increased at 3–12 weeks of pregnancy [44]. In contrast, PGRMC1 mRNA expression is positively correlated with endometrial tissue P₄ concentrations only during the early luteal phase (Days 5–6) [13]. In this study, no difference was observed in PGRMC1 and PGRMC2 mRNA expression levels between the two groups. Therefore, LBF may affect uterine function via a genomic mechanism through PGR, rather than a non-genomic mechanism.

During the estrous cycle and pregnancy, P₄ induces the expression of various genes in the endometrium to establish uterine receptivity [12, 45]. The expression levels of these P₄-regulated genes change over time, and their products alter the composition of histotrophs to promote embryonic growth, conceptus elongation, implantation, and placentation [11, 12, 41, 45, 46]. Several genes are classified as P₄-regulated genes; therefore, in this study, we investigated ANPEP, DGAT2, DKK1, and LTF. Alanyl (membrane) aminopeptidase (ANPEP), known as CD13, is a membrane-bound zinc-dependent peptidase that cleaves neutral amino acids from the amino terminus of an oligopeptide [47]. ANPEP is present in various organs, tissues, and cells, and is involved in angiogenesis, tumor cell invasion and metastasis, immune responses, and ECM remodeling [47, 48]. In addition, ANPEP may modify the intrauterine environment by regulating the concentration of bioactive peptides in the endometrium [31, 48, 49]. ANPEP mRNA expression on Day 7 was lower in the low P₄ model than in the normal cows, whereas no differences were observed between the high P₄ model and the normal cows [31, 41]. Diacylglycerol O-acyltransferase 2 (DGAT2) is an enzyme involved in lipid metabolism that catalyzes the final synthesis step of triacylglyceride [50], the main energy source for bovine blastocysts. DGAT2 mRNA expression in the endometrium increases with the progression of the estrous cycle (that is, as P₄ concentrations increase) [51, 52] and decreases by Day 16 [46]. Additionally, DGAT2 mRNA expression was higher in high P₄ model cows than in normal cows on Day 7 [46]. Dickkopf homolog 1 (DKK1), an inhibitor of the canonical WNT signaling pathway, is an endometrial secretory protein produced during the estrous cycle and early pregnancy [53] and is believed to be an embryokine that regulates embryo development [54]. DKK1 positively affects trophectoderm development [53] and trophoblast elongation [55]. Some studies have reported high conception rates after the transfer of embryos treated with DKK1 [53], while others have observed no improvement [56, 57]. Furthermore, DKK1 mRNA expression on Day 7 was lower in low P₄ model cows than that in normal cows [41]. In this study, the expression levels of ANPEP (P < 0.1), DGAT2 (P < 0.01), and DKK1 (P < 0.01) mRNA were lower in the HV group than in the LV group. The expression of these genes correlates with changes in the expression level of PGR mRNA [41, 46]. Increased PGR mRNA expression on Day 7 is associated with decreased ANPEP and DKK1 mRNA expression [41], whereas decreased PGR mRNA expression is associated with increased DGAT2 mRNA expression [46]. In this study, no difference was observed in plasma or endometrial tissue P₄ concentrations; however, PGR mRNA expression was higher in the HV group than in the LV group. Therefore, the reduced ANPEP, DKK1, and DGAT2 mRNA expression observed in the HV group in the present study may be attributable to high PGR mRNA expression.

LTF is an iron-binding glycoprotein found in mammalian exocrine secretions that has various physiological functions, including antibacterial, antiviral, antifungal, anti-inflammatory, and immunomodulatory properties [58]. In the mouse and horse endometrium, LTF expression is induced by estrogen and suppressed by P₄ [59, 60]; therefore, LTF is most abundant during proestrus and estrus [60, 61]. In the bovine endometrium, LTF mRNA expression is high up to Day 7, but decreases to basal levels by Day 13, when P₄ peaks [31]. However, the manipulation of circulating P₄ concentrations during the early luteal phase and differential PGR mRNA expression did not alter LTF mRNA expression on Day 7 [31, 41]. In contrast to the other three genes, LTF was not affected by the expression levels of PGR mRNA on Day 7; therefore, we speculated that no differences were observed between the two groups in this study.

In conclusion, we demonstrated that the LBF level on Day 7 altered endometrial gene expression (CCN3, PGR, ANPEP, DGAT2, and DKK1), whereas it did not affect circulating or endometrial tissue P₄ concentrations in JB cows. These findings suggest that LBF affects fertility by regulating endometrial function through a pathway that regulates PGRs in the endometrium. However, the mechanism by which LBF affects endometrial gene expression remains unclear, and further studies are required to clarify these pathways.

Conflicts of interests

The authors declare no conflicts of interest associated with this manuscript.