Sample preparation
The animal experiment has been described in Wang et al [60]. All experiments involving animals were carried out in accordance with regulations (No. 5 proclamation of the Standing Committee of Hubei People’s Congress) approved by the Standing Committee of the Hubei People's Congress in P. R. China. All animal sample collection procedures were approved by the Ethics Committee of Huazhong Agricultural University. The animals were housed at Jingpin farm of Huazhong Agricultural University (Wuhan, China) and were slaughtered at the slaughterhouse of Huazhong Agricultural University. Before every gilt was slaughtered, a warm shower to relax the pigs, then they were stunned with low-voltage electric shock to reduce the pain and were exsanguinated by puncturing carotid artery to death.
Nine purebred Yorkshire gilts with similar weights (120 ± 10 kg), ages (8 months) and genetic background were selected in this study. Gilts were artificially inseminated using extended semen from one boar (the breeding pig farm of Huazhong Agricultural University) at the onset of estrus (day 0) and again 12 h after. Uteri were obtained from pigs slaughtered on days 10, 13 and 18, and uterine flushing fluids (UFs) was collected by flushing with 30 mL phosphate buffer saline (PBS; pH 7.4, catalog number: 02-024-1ACS, Biological Industries). The presence of filamentous conceptuses in the flush of the uterine horns was used to confirm pregnancy. The UFs sample from pregnancy gilts were used for subsequent sequencing analysis. For characterization of extracellular vesicles from porcine UFs, three additional nonpregnant gilts (day13 of estrous cycle) with similar weights, ages and genetic background (see above) were slaughtered and their uteri were collected on day 13 after estrus. The UFs were obtained by using the above method. At the same time, these uteri were used for the separation of endometrial epithelial cells.
UFs exosome isolation
The UFs was clarified by centrifugation (2000 g for 30 min at 4°C), filtered through a 0.22 μm filter (MILLEX-GP, USA) to remove impurity. Then, these samples were ultracentrifuged at 130,000 g for 2 h at 4°C (Beckman Optima XE-90, SW32 Ti rotor, Beckman Coulter. USA). Extracellular vesicles (EVs) were re-suspended in 200 μL PBS (catalog number: 02-024-1ACS, Biological Industries) after washing with PBS (130,000 g for 2 h). Extracellular vesicles were stored at –80°C until the RNA was isolated.
Western blot (WB)
The exosomal and cellular proteins were extracted by DNA/RNA/protein Isolation Kit (catalog number: R6734-02, OMEGA). The sample was denatured by heating, separated by SDS-PAGE and transferred to a PVDF (polyvinylidene fluoride) membrane. Next, the membranes were blocked with 5% skimmed milk powder and separately probed with rabbit anti CD9 (catalog number: 20597-1-AP, Proteintech), rabbit anti HSP70 (catalog number: GB11241, servicebio), rabbit anti CD63 (catalog number: ab231975, abcam) , rabbit anti calnexin (catalog number: 10427-2-AP, Proteintech)and rabbit anti cytochrome c (catalog number: sc:13156, santa cruz) overnight at 4°C with a final dilution of 1:1000 (v/v). After three times of washing, the membranes were incubated with goat anti-rabbit secondary antibodies (catalog number: GB23303, Sevicebio) with 1:2000 dilution (v/v) at 37°C for 1.5 h. The images of membranes treated with ECL (enhance chemiluminescence) were captured by Western Blotting Detection System (Tiangen, Beijing, China).
Transmission electron microscope (TEM)
The morphology of isolated extracellular vesicles was visualized with high-resolution transmission electron microscope (Hitachi HT7700) based on a previous method [61]. In brief, the resuspended extracellular vesicles were placed on carbon-coated copper grid, and then subjected to 2% phosphotungstic acid (catalog number: DZ0035, Leagene) staining for 2 min. The excess liquid was blotted off by filter paper, and grids were allowed to dry overnight.
Nanoparticle tracking analysis (NTA)
The size distribution and concentration of isolated extracellular vesicles were analyzed by NTA with Zeta View (Particle Metrix). Isolated extracellular vesicles were diluted with particle-free PBS at the ratio of 1:200 and added into the chamber. The results of size distribution and vesicle concentrations were assessed with the software Zeta View.
RNA extraction and RNase treatment
The RNAs of EVs were extracted by miRNeasy Serum/plasma Kit (catalog number: 217184, Qiagen, German) according to manufacturer’s protocol. The quantity and quality of RNA were assessed using a Small RNA kit and a Small RNA assay on an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Extracted RNA was treated with 10ng/ul of RNase A (Sigma-Aldrich) at room temperature for 30 min. and then treated RNA was assessed as above.
Cell culture
The uterus of non-pregnant gilts was opened longitudinally on a sterile clean bench. The endometrium was separated and shredded with a sterile scissor. After washing twice with PBS, the tissue pieces were incubated at 37°C for 2.5 h with collagenase I (catalog number: 17100017, Gibco) and shaken vigorously every half hour. Undigested tissue pieces were removed by screen filtration. Then, the filtrate was centrifuged at 500 g for 10 min to remove supernatant (fraction rich of endometrial stromal cells). The pellet (epithelial‐rich fraction) was resuspended twice in PBS and recentrifuged (500 g, 10 min) twice. The resultant pellets containing endometrial epithelial cells (EECs) were suspended in Dulbecco’s modified Eagle’s medium/F-12 (DMEM/F12; 1:1) medium (Gibco, NY, USA) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin-streptomycin (Gibco, NY, USA) and cultured in 37°C and 5% CO2 incubator. Endometrial stromal cells were further removed by 0.25% trypsin without edetic acid disodium salt (EDTA) after 2–4 d. The epithelial cells were then trypsinized with 0.25% trypsin-EDTA and placed on Cell Culture Flask (Corning, NY, USA) for subsequent experiments. Isolation and culture of porcine primary EECs referred to pervious research [60, 62].
Porcine trophectoderm cells (PTr2) were kindly provided by Mr. Jiang zongyong, Guangdong Academy of Agricultural Sciences. PTr2 cells were established from dispersed cell culture of day 12 filamentous conceptus obtained from pigs. These cells have been previously characterized for SN1/38 (porcine trophectoderm-specific monoclonal antibody) positive, cytokeratin 7 positive, vimentin negative, express fibronectin and many of the integrin subunits present in porcine trophectoderm in vivo. Detailed methods have been published [63].These cells were cultured in DMEM-F12 (Gibco) supplemented with insulin (0.1 Units/mL; Sigma-Aldrich, German), glutamine (2 mM, Sigma-Aldrich), 1% penicillin-streptomycin and 5% fetal bovine serum (Gibco).
In order to ensure that EECs and PTr2 cells are not contaminated with mycoplasma, their spent culture media were assessed mycoplasma contamination with MycAway™ -Color One-Step Mycoplasma Detection Kit (catalog number: 40611ES25, yeasen) following the manufacturer’s instructions.
PKH67 staining and Confocal fluorescence microscopy
Isolated extracellular vesicles were stained with PKH67 green fluorescent mini kit (catalog number: MINI67-1KT, Sigma-Aldrich) following the manufacturer’s instructions. Briefly, extracellular vesicles and PBS (negative control) were respectively added to 1 mL Diluent C, and then incubated with 4 ul PKH67 for 4 min. At last, the mixture was closed with 2 mL 0.5% bovine serum albumin (BSA), washed once with PBS, and then ultracentrifuged at 130,000 g for 2 h at 4°C.
For visual inspection of the uptake of EVs by cells, the cells were placed on glass bottom cell culture dish (catalog number: 801001, NEST) and incubated at 37°C for 12h or 24h with 20 μg/mL of PKH67 labelled EVs. At the indicated time points, cells were fixed with 4% paraformaldehyde (catalog number: BL539A, biosharp) for 10 min followed by PBS wash three times for 5 min, and cells were permeabilized with 0.5% Triton X-100 (catalog number: ST795, beyotime) solution followed by PBS wash three times for 5 min. Next, 200 μL TRITC Phalloidin dye (catalog number: 40734ES75, yeasen) was added to the cells and allowed to incubate for 30 min followed by PBS wash three times for 5 min, and then 200 µL DAPI (catalog number: C1006, biosharp) solution was added to the cells and allowed to incubate for 10 min followed by PBS wash three times for 5 min. Finally, cells were visualized with a 63× oil immersion confocal microscope (LSM 800, Zeiss). The acquired images were further processed with ZEN 2.3 software.
Small RNA sequencing
Nine small RNA libraries were generated under standard procedures at Beijing Genomics Institute (BGI, China) according to a previous study [64]. Sequencing of the libraries was carried out with the BGISEQ-500 platform (BGI, China; http://www.seq500.com/en/).
Prediction of novel miRNAs, novel piRNAs and miRNA target genes
Before data analysis, clean reads were obtained by eliminating low-quality reads, adaptors, and other contaminants. Clean reads were then mapped to reference genome and other sRNA databases by Bowtie2 [65]. Because each unique small RNA was supposed to be mapped to only one category, we followed a precedence rule of MiRbase > piRBase > snoRNA (human/plant) > Rfam > other sRNA. Piano [66] and miRNADeeps2 [67] were used to predict novel piRNAs and miRNAs, respectively. This algorithm of Piano is based on the Support Vector Machine (SVM) algorithm and transposon interaction information. The miRNA expression profiles were normalized by transcript per million (TPM), as previously reported [68]. The miRNAs with expression values of >28 TPM in at least one of the 10 libraries were selected for all the following analyses.
Prediction of the target genes of miRNAs was carried out by TargetScan, RNAhybrid and miRnada. the intersection of miRnada, RNAhybird and Targetscan predictions were selected for further analysis. The default parameters are as follows: “-en -20 -strict” (miRanda), “ -b 100 -c -f 2,8 -m 100000 -v 3 -u 3 -e -20 -p 1 -s 3utr_human” (RNAhybrid) and “the 2-8nt of 5 'end of small RNA as seed sequences and 3′-UTR region of porcine transcript were used for prediction” (Targetscan).
Differentially expressed (DE) small RNAs
The expression levels of small RNAs were calculated by using Transcript Per Kilobase Million (TPM) as described in a previous report [69]. Small RNAs with expression value >28 TPM in at least in one of the nine libraries were selected for subsequent data analysis [68]. Raw P-value were convert to adjusted P-value using the Benjamin-Hochberg false discovery rate [70]. To improve the accuracy of the identification of DE small RNAs, we defined a miRNA as a DE miRNA when reads number foldchange ≥2 and P-value ≤ 0.05.
Real time quantitative PCR (RT-qPCR)
First-strand cDNAs of small RNAs were synthesized by Mir-X™ miRNA First Strand Synthesis kit (TaKaRa, Dalian, China) according to the manufacturer’s instructions. The qPCR was carried out by using SYBR Green PCR Master Mix (TaKaRa, Dalian, China) and LightCycler 480II Real-Time PCR System. The forward primers of small RNAs were designed based on the mature sequences of small RNAs (Additional file 12: Table S10), and the reverse primer was a universal primer supplied by the above kit. The expression levels of small RNAs were normalized with U6 to obtain the relative expression by comparative CT method.
Hierarchical cluster analysis (HCA)
The expression values of the miRNAs (>28 TPM in at least one library) were normalized by using the Z-score. HCA was conducted using the R package.
Principal component analysis (PCA)
The expression values of the small RNAs (> 28 TPM in at least one library) were normalized by using the lg (TPM). PCA was performed using the imageGP (http://www.ehbio.com/ImageGP/index.php/Home/Index/PCAplot.html).
Functional annotation of the predicted target genes
Functional analyses of the predicted target genes were conducted on the basis of Kyoto Encyclopedia of Genes and Genomes database (KEGG) and Gene Ontology slim database (GO-slim) by using KOBAS (http://kobas.cbi. pku.edu.cn/) and PANTHER (http://www.pantherdb.org/), respectively. An enriched functional class with a P-value <0.05 was defined as a significantly enriched target gene candidate. The network of GO terms based on the GO database was constituted by ClueGO, which is a plugin in Cytoscape (http://www.cytoscape.org/).
Time-series expression analysis
The profiles of DE miRNAs over time were visualized by Short Time-series Expression Miner v 1.3.8 (STEM, http://www.cs.cmu.edu/~jernst/stem/). The maximum number of model profiles and the maximum unit change in model profiles between time point were adjusted to 20 and 1, respectively. The STEM clustering method was selected as clustering method and other parameters were set as default. The miRNA expression profiles were ordered by significance.
Statistical analysis
Calculate the significance of relative fluorescence was assessed by two sets of comparisons using the Bonferroni t-test in SAS 9.1, and shown as * p < 0.05 and ** p < 0.01. Correlation tests were performed to calculate the correlation between the results of RNA-seq and qPCR.