Genome-wide DNA methylation patterns of bovine blastocysts derived from in vivo embryos subjected to in vitro culture before, during or after embryonic genome activation

Genome-wide DNA methylation patterns of bovine blastocysts derived from in vivo embryos subjected... Background: Aberrant DNA methylation patterns of genes required for development are common in in vitro produced embryos. In this regard, we previously identified altered DNA methylation patterns of in vivo developed blastocysts from embryos which spent different stages of development in vitro, indicating carryover effects of suboptimal culture conditions on epigenetic signatures of preimplantation embryos. However, epigenetic responses of in vivo originated embryos to suboptimal culture conditions are not fully understood. Therefore, here we investigated DNA methylation patterns of in vivo derived bovine embryos subjected to in vitro culture condition before, during or after major embryonic genome activation (EGA). For this, in vivo produced 2-, 8- and 16-cell stage embryos were cultured in vitro until the blastocyst stage and blastocysts were used for genome-wide DNA methylation analysis. Results: The 2- and 8-cell flushed embryo groups showed lower blastocyst rates compared to the 16-cell flush group. This was further accompanied by increased numbers of differentially methylated genomic regions (DMRs) in blastocysts of the 2- and 8-cell flush groups compared to the complete in vivo control ones. Moreover, 1623 genomic loci including imprinted genes were hypermethylated in blastocyst of 2-, 8- and 16-cell flushed groups, indicating the presence of genomic regions which are sensitive to the in vitro culture at any stage of embryonic development. Furthermore, hypermethylated genomic loci outnumbered hypomethylated ones in blastocysts of 2- and 16-cell flushed embryo groups, but the opposite occurred in the 8-cell group. Moreover, DMRs which were unique to blastocysts of the 2-cell flushed group and inversely correlated with corresponding mRNA expression levels were involved in plasma membrane lactate transport, amino acid transport and phosphorus metabolic processes, whereas DMRs which were specific to the 8-cell group and inversely correlated with corresponding mRNA expression levels were involved in several biological processes including regulation of fatty acids and steroid biosynthesis processes. Conclusion: In vivo embryos subjected to in vitro culture before and during major embryonic genome activation (EGA) are prone to changes in DNA methylationmarks andexposureofinvivoembryos to in vitroculture during thetimeof EGA increased hypomethylated genomic loci in blastocysts. Keywords: DNA methylation, Gene expression, Culture condition, Bovine embryo * Correspondence: dsal@itw.uni-bonn.de Institute of Animal Science, Animal Breeding and Husbandry Group, University of Bonn, 53115 Bonn, Germany Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 2 of 19 Background responses of embryos derived from oocytes matured and Following fertilization, the mammalian zygote undergoes fertilized in vivo when subjected to in vitro culture condi- successive cleavage divisions in the oviduct and eventually, tions is essential to unravel the epigenetic signatures at the blastocyst stage, the embryo enters to the uterus, induced by environmental challenges during embryo hatched from the zona pellucida and attach to the luminal culture. Therefore, this study aimed to investigate epithelium to initiate implantation [1]. During the early genome-wide DNA methylation patterns of bovine blasto- stage of successive developmental processes, the develop- cysts derived from in vivo embryos subjected to the in ing embryo undergoes signal communication with the ma- vitro culture condition before, during or after embryonic ternal environment [2–4] by secreting signals which genome activation. subsequently initiates the establishment of persistent bidir- ectional interaction with the maternal environment [5–7]. Methods Effects of environmental insults on the survival and de- Animal handling, management and estrus synchronization velopment of the embryo are highly noticeable at early Thirty-eight healthy Simmental heifers aged between 15 stages of development when the culture condition induces and 20 months and weighing between 380 and 500 kg in alterations in genome reprogramming [8, 9] which could the FrankenForst teaching and research station, University even persist until adult life [10]. Several studies indicated of Bonn, were selected for this study for generation of in theeffects of culture conditiononthe DNAmethylation vivo embryos. The animals were housed in a free-stall patterns of embryos. For instance, the DNA methylation barn with slotted floors and cubicles, lined with rubber patterns of key genes such as insulin-induced gene (INSIG) mats and they were fed a total mixed ration. Animal hand- and sterol regulatory element-binding protein (SREBP) ling and management of the experimental animals were were found to be altered in the fetus and placenta in human performed according to the rules and regulations of the pregnancies established after in vitro fertilization [11]. Simi- German law of animal protection. The Animal Welfare larly, aberrant hypomethylation of the maternal imprinting Committee of the University of Bonn approved the experi- control region of the KvDMR1 gene in clinically normal ment with proposition number 84–02.05.20.12.075. children conceived by in vitro fertilization (IVF) and intra- cytoplasmic sperm injection (ICSI) was also identified [12]. In vivo embryo production and in vitro culture Moreover, alteration in the global DNA methylation pat- In this study, different stages of in vivo derived embryos terns of human placenta derived from pregnancies estab- were further cultured in vitro and the resulted blastocysts lished from embryos cultured under high oxygen tension were used for DNA methylation analysis. For this, Simmen- [13], and alterations in the allelic methylation patterns of tal heifers at the FrankenFrost teaching and research sta- maternally or paternally expressed genes in mice embryonic tion, University of Bonn, were estrous synchronized, stem cell fetuses [14] are additional evidences indicating super-ovulated and artificially inseminated as described pre- how suboptimal culture condition could affect the genome viously [18]. A total of 11, 6, 12 and 13 animals were ran- methylation at the later stage of embryo development. In domly assigned for 2-, 8–16-cell and blastocyst stage porcine, the DNA methylation level was increased in in embryo production, respectively. Afterwards, 2-, 8- and vitro embryos compared to in vivo ones suggesting the ad- 16-cell stage embryos were flushed at 48, 72 and 96 h post verse effect of in vitro culture on the DNA methylome of insemination, respectively using tubal embryo recovery and embryos [15]. In bovine, oocyte maturation media was con- transfer methods [19]. The experimental design used for firmed to cause hypermethylation in the liver of fetuses at sample collection is indicated in Fig. 1. These embryo day 50 of the gestation period [16] indicating the long term groups represent the embryonic developmental stage be- effect of culture condition on the DNA methylome of the fore, during and after embryonic genome activation, re- embryo or the fetal genome. spectively. All embryo groups were then in vitro cultured Previously, we have reported a stage-specific alter- until the blastocyst stage and the developmental compe- ations of DNA methylation patterns of in vivo devel- tence of the 2-cell (2C_Flush), 8-cell (8C_Flush) and 16-cell oped blastocysts from embryos developed in vitro up to (16C_Flush) were recorded. In addition, the in vivo blasto- zygote, 4- or 16-cell stages [17]. In that study, we found cyst group (vivo) was obtained from embryos completely that DNA methylation patterns in the resulting blasto- developed under in vivo condition. cysts was compensated in a stage specific manner compared to blastocysts completely developed under in DNA fragmentation, adapter ligation and methyl sensitive vitro condition. Nevertheless, in that study, it was not enzymes digestion possible to predict whether the DNA methylation dys- Genomic DNA (gDNA) from four replicates (10 blasto- regulation was derived from the epigenetic memory cysts per replicate) of the 2C_Flush, 8C_Flush, 16C_Flush caused by the in vitro oocyte maturation, fertilization and vivo blastocyst groups was isolated using the Allprep and culture condition. Therefore, understanding epigenetic DNA/RNA micro kit (Qiagen, Germany) following the Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 3 of 19 Fig. 1 Experimental design used to generate different blastocyst groups for DNA methylation analysis. IVC; In vitro culutre manufacturer’s recommendations. The gDNA was then or 16C_Flush and 1 μg Cy5 or Cy3 labeled amplified fragmented, adapter ligated and cleaved by gDNA of the vivo blastocyst groups were mixed with methyl-sensitive restriction endonucleases using the simi- 157.6 μl hybridization cocktail and incubated at 95 °C for lar methodology that was used previously ([17]. Briefly, 3 min and at 37 °C for 30 min. At the end of incubation, the gDNA of each group was subjected to the MseI en- 65 μl of Agilent-CGHBlock was added and the mix was zyme digestion and the MseLig12 (100 μM) (5′-TAA CTA then transferred onto the EmbryoGENE DNA Methyla- GCA TGC-3′) and 0.5 μl MseLig21 (100 μM) (5-AGT tion Array slides [20]. The slides were then placed in a GGG ATT CCG CAT GCT AGT-3′)adapterswereli- hybridization oven (Shel Lab) for 40 h at 65 °C at with a gated to the fragmented DNA using T4 DNA ligase in the speed of 20 RPM. Four biological replicates of 2C_Flush, presence of 10× One-phor-All buffer PLUS and ATP. 8C_Flush and 16C_Flush blastocyst groups were hybrid- Non-methylated genomic regions were then cleaved using ized to four biological replicates of the in vivo group. A FastDigest™ methyl-sensitive restriction endonucleases total of 12 arrays were hybridized and biological [HpaII (C/CGG), HinP1I (GC/GC) and Aci1I (C/CGC) dye-swaps were performed to avoid any false positive re- (Fermentas, Thermo Fisher Scientific, Germany)]. The ef- sults. The arrays were then washed and scanned using ficiency of methyl-sensitive restriction endonucleases PowerScanner (Tecan) which integrates the Array-Pro cleavage was evaluated based on the qPCR amplification Analyzer 6.3 software (MediaCybernetics). Agilent’sfea- results of the spike-in controls added during the gDNA ture extraction software (Agilent Technologies, USA) was fragmentation step [17, 20]. Samples which appeared after used to extract the array features. 5 cycles compared to undigested controls were used for downstream treatments, otherwise the samples were Array data and downstream analysis digested one more time using specific methyl sensitive re- The array data analysis methodsusedfor this studyare ex- striction enzyme that did not work in the first run. After- tensively described previously [17, 20]. Briefly, Loess wards, unmethylated genomic regions were enriched by inter-array normalization method was used to correct the performing a second round ligation-mediated polymerase dye effects, while inter-array scale normalization was used to chain reaction as indicated previously [20]. balance the distribution differences among experiments. Linear models for microarray data (limma) [21]was used to DNA labeling and hybridization identify differentially methylated regions (DMRs). Probes About 2 μg of amplified gDNA from the 2C_Flush, with absolute log (fold-change) ≥ 1.5 differences between 8C_Flush, 16C_Flush or vivo blastocyst groups was mixed samples to be compared with significant (p <0.05) were with 1 μl Cy-3 or Cy-5 dyes (Kreatech Biotechnology) and considered as differentially methylated regions (DMRs). heated at 85 °C for 30 min. After the end of incubation, Thus, probes which showed a significant increase or de- the samples were purified using the Qiagen PCR purifica- crease in signal intensity by ≥1.5 folds in 2C_Flush, tion kit (Qiagen, Germany). Afterwards, 1 μg of Cy5 or 8C_Flush or 16C_Flush compared to the vivo group were Cy3 labeled gDNA sample from the 2C_Flush, 8C_Flush considered as hypermethylated and hypomethylated probes, Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 4 of 19 respectively. The pathways enriched by the differentially SYBR Green Supermix with ROX (Bio-Rad laboratories, methylated genes were analyzed using Gprofiler (http:// Germany), the cDNA samples of each treatment group biit.cs.ut.ee/gprofiler/). For comparative analysis of the and the specific forward and reverse primer in the Ste- methylation profile and gene expression data, the genes with pOnePlus™ Real-Time PCR Systems (Applied Biosys- absolute log (fold-change) ≥1.5, p value < 0.05 and false dis- tems, USA). The qPCR thermal cycling parameters were covery rate (FDR) < 0.3 were selected from our previous data set to 95 °C for 3 min followed by 40 cycles of 95 °C for (GSE33314 and GSE111990). The heat maps of differentially 15 s and 60 °C for 1 min. Following this, dissociation methylated regions and differentially expressed genes were curve was generated by starting fluorescence acquisition constructed using PermutMatrix (http://www.atgc-montpel- at 60 °C and measurements were taken in every 7 s lier.fr/permutmatrix/) and matrix2png interface (https:// interval until the temperature reached 95 °C. The speci- matrix2png.msl.ubc.ca/bin/matrix2png.cgi). The network of ficity of amplification was evaluated by monitoring the genes was visualized using the networkanalyst tool (http:// dissociation (melting) curve of each candidate gene. The www.networkanalyst.ca/). qPCR data was generated from three independent bio- logical replicates and the data was analyzed using the −ΔΔCt Confirmation of differentially methylated genomic delta Ct method (2 ). The expression level of Glyc- regions using bisulfite sequencing eraldehyde 3-phosphate dehydrogenase (GAPDH) was The DNA methylation levels of selected candidate genes used as an endogenous normalizer. were validated using bisulfite sequencing. For this, the sequences of the candidate gene/probe were retrieved Results using the Site Search tool http://emb-bioinfo.fsaa.ula- DNA methylation marks sensitive to in vitro culture val.ca/bioinfo/html/SiteSearch.html. Afterwards, gene/ condition at any stage of embryonic development sequence specific primers (Additional file 1: Table S1) In this study, the DNA methylation patterns of blastocysts were designed using MethPrimer (http://www.urogen- derived from embryos developed in vivo followed by in e.org/cgi-bin/methprimer/methprimer.cgi) and used for vitro culture were investigated. For this, in vivo developed amplification of the gene of interest. To do this, first the 2-cell (2C_Flush), 8-cell (8C_Flush) and 16-cell genomic DNA of each blastocyst group was modified (16C_Flush) embryos were subjected to in vitro culture with sodium bisulfite converted using EZ DNA methyla- condition until the blastocyst stage. Although the day 9 tion direct kit (ZymoResearch). Afterwards, bisulfite blastocyst rate was similar in the 2C_Flush and 8C_Flush converted gDNA was subjected to PCR amplification groups, the day 7 and day 8 blastocyst rates were relatively using Zymo Taq™ DNA polymerase (ZymoResearch). lower in 2C_Flush embryos. The day 7–9 blastocyst rates The PCR product was purified and ligated to the in the 16C_Flush group were relatively higher compared pGEM®-T Easy Vector System (Promega, USA) and to both the 2C_Flush and 8C_Flush groups (Table 1). transformed to E. coli competent cells. The bacterial cul- Therefore, to get an overview of the epigenome repro- ture was plated onto LB agar/ampicillin/IPTG/X-gal gramming that occurred in these embryo groups, we in- plate and incubated overnight at 37 °C. After overnight vestigated DNA methylation patterns in each of the culture independent white colonies were selected and se- blastocysts group using EmbryoGENE DNA methylation quenced with GenomeLab™ GeXP Genetic Analysis Sys- array platform. Aberrant hyper or hypomethylated gen- tem (Beckman Coulter). The bisulfite sequencing data omic loci were increased in blastocysts derived from em- were then analyzed using the quantification tool for bryos that were exposed to in vitro conditions at 2- and methylation analysis (http://quma.cdb.riken.jp/). 8-cell stages compared to the 16-cell stage indicating that the suboptimal culture condition induced aberrant DNA Quantification of candidate genes using quantitative real methylation pattern more strongly before and during the time PCR embryonic genome activation than later (Fig. 2). Since the DNA methyltion profile data was supperim- posed to previously generated expression data [[22], Table 1 Developmental competence of in vivo developed 2-, 8- GSE33314 and GSE111990]] using the same experimen- and 16-cell stage embryos under in vitro culture condition tal design, expression patterns of selected differentially Group Heifers Total Blastocyst development expressed genes were further quantified using quantita- embryo Day 7 Day 8 Day 9 tive polymerase chain reaction (qPCR) using gene spec- flushed fic primers (Additional file 2: Table S2) designed using (n) (n) (n) (%) (n) (%) (n) (%) a a a primer3 (http://bioinfo.ut.ee/primer3-0.4.0/). For this, 2-cell flush 11 69 15 21.7 42 60.9 57 82.6 the total RNA isolated from the experimental blastocyst b b a 8-cell flush 6 118 63 53.4 93 78.8 97 82.2 groups were reverse transcribed. The qPCR was then c c b 16-cell flush 12 45 36 80.0 42 93.3 45 100.0 performed in 20 μl reaction volume containing iTaq Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 5 of 19 Fig. 2 Global DNA methylation patterns of differentially methylated genomic regions in different blastocyst groups. a Volcano plots displaying the distribution of hypermethylated (positive log fold changes) and hypomethylated (negative log fold changes) genomic loci in each 2 2 blastocysts group. b The total number of differentially methylated genomic regions in each blastocyst group A detailed analysis of DMRs including 160 CpG 2C_Flush, 8C_Flush and 16C_Flush, respectively. Among islands indicated that the DNA methylation patterns of these, 88.5% (n = 1623) of the DMRs including the 1832 genomic loci (containing 29,166 CpG sites) were imprinted genes (PEG3, IGF2R, ASB4, SFMBT2 and differentially methylated in all the three blastocyst GRB10) and DNMT1 were found to be hypermethylated. groups, despite the stage specific exposure to the in vitro The majority of commonly altered genes including the culture (Fig. 3a). These genomic loci represented 17.6, solute carrier clusters, collagens, ADAM metallopepti- 19.6 and 26.7% of the total DMRs identified in the dases, ATPases, zink finger proteins, kinesins were Fig. 3 Commonly differentially methylated genomic regions in all blastocyst groups. a The proportion of commonly detected hypermethylated and hypomethylated genomic loci. b Genomic localization of commonly differentially methylated genomic regions and the list of representative genes in each category Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 6 of 19 differentially methylated in the exonic and/or intronic Network analysis of these genes indicated that PAK4 and gene body regions (Fig. 3b). These differentially methyl- RPS15 genes which were downregulated but hypermethy- ated genes were found to be involved in 16 biological lated in all blastocysts group were found to be the center pathways including the calcium signaling pathway and genes in lipid biosynthesis and nucleosome assembly, re- focal adhesion (Fig. 4a). In addition, PAK4, COL4A1, spectively while TP53BP2 was the center gene of inflamma- COL1A2, EGFR, SMAD3, ITPKA and MYLK were tory response and intracellular transport pathways (Fig. 5b). among the commonly differentially methylated genes that were foundtobeinvolvedintwo or more path- Aberrant DNA methylation patterns induced in both ways (Fig. 4b). blastocysts developed in vitro from embryos flushed After identification of genomic regions commonly differ- before and during embryonic genomic activation entially methylated in all blastocyst groups, we investigated Unlike the 16C_Flush blastocysts group, both the the effects of aberrant DNA methylation patterns on the 2C_Flush and 8C_Flush blastocysts were obtained from blastocyst gene expression by superimposing the DNA embryos that spent several cleavages in vitro including methylation profile with the transcriptome data [[22]], the major embryonic genome activation (EGA). There- GSE33314 and GSE111990]]. The results of this analysis in- fore, here we investigated the DMRs commonly dysregu- dicated that only 2% (n = 28) of the annotated differentially lated only in the 2C_Flush and 8C_Flush groups, but not methylated genes were differentially expressed in all three in the 16C_Flush group. The results showed that groups. Among which, the expression level of 11 genes in- including DNA (cytosine-5)-methyltransferase 3 beta cluding BAHD1, RPS15, PAK4 and PISD were negatively (DNMT3B), the methylation patterns of 2988 genomic correlated with the DNA methylation patterns (Fig. 5a). loci were altered only in these two blastocyst groups of Fig. 4 a List of pathways enriched by commonly differentially methylated genomic regions in all blastocyst groups and b list of representative differentially methylated genomic regions involved in more than one pathway Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 7 of 19 Fig. 5 a The heatmap depicting the methylation and expression patterns of DMRs commonly detected in all blastocyst groups. DM; differentially methylated, DE; differentially expressed. The red color indicates hypermethylation of the DNA methylation or upregulation of the mRNA expression whereas the green color indicates hypomethylation of the DNA methylation or downregulation of the mRNA expression. b The gene network and biological processes enriched by commonly detected DMRs and which showed inverse correlation with their corresponding mRNA expression levels which 72% were hypomethylated in both blastocyst correlated to their corresponding expression levels. groups (Fig. 6a). Genomic localization of these DMRs Among these, the expression level of 61 hypomethylated showed that about 67% of the DMRs are localized in genes including MLH1, SMARCA5, NFYB, ATR, ROCK1 and around the gene body regions (Fig. 6b) and the and ACVR1 was upregulated while the mRNA level of hypermethylated loci were more common in the exonic 11 hypermethylated genes including ADCY5 and IGF2R regions and the hypomethylated loci were common in was downregulated in both 2C_Flush and 8C_Flush the promoter and intronic regions (Fig. 6c). blastocyst groups (Fig. 7a). Furthermore, the gene net- Next, we investigated the function of these genomic work analysis showed that SMARCA5, NFYB and regions by performing gene enrichment analysis and the ACVR1 were the center genes regulating cell maturation results have shown that Pi3-Akt signaling pathway, Ras and ATR, ROCK1 and ADCY5 were found to be the cen- signaling pathway, peroxisome pathway, EGFR tyrosine ter genes of the lipid biosynthesis processes (Fig. 7b). kinase inhibitor resistance and metabolic pathways were among the biological pathways altered by the differen- DNA methylation patterns induced exclusively in tially methylated genes (Fig. 6d). blastocysts developed in vitro from embryos flushed Comparative analysis of the methylation and mRNA before, during or after embryonic genomic activation expression patterns indicated that the methylation levels We considered that in the 2C_Flush blastocysts group, of 72 out of 753 annotated DMRs were inversely both the major and minor EGA occurred under the in Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 8 of 19 Fig. 6 Commonly differentially methylated genomic regions in both the 2C_Flush and 8C_Flush blastocyst groups. a The proportion of hypermethylated and hypomethylated DMRs. b The distribution of DMRs in different genomic regions. c The proportion of hypermethylated and hypomethylated DMRs in different genomic regions. Log FC: the ratio of hypermethylated and hypomethylated genomic loci described in Log 2 2 scale. d Pathways enriched by differentially methylated genomic regions in both the 2C_Flush and 8C_Flush blastocyst groups. DMRs:TGS, the ratio of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in pathway vitro condition, while in the 8C_Flush blastocyst group, EIF4B) were found to be unique to the 16C_Flush minor EGA was completed under in vivo condition blastocysts group. When these genomic loci were whereas the major EGA was partly occurred under in partitioned into hypermethylated and hypomethylated vivo condition. In the 16C_Flush blastocyst group, both categories, the hypermethylated loci exceeded the minor and major EGA was completed under in vivo con- number of hypomethylated genomic loci in the dition. Therefore, we further analyzed the differentially 2C_Flush and 16C_Flush groups. However, the reverse methylated genomic regions that were specific to the wasobservedin the 8C_Flush group(Fig. 8a). 2C_Flush, 8C_Flush or 16C_Flush blastocyst groups. Our findings revealed that a total of 4599 DMRs including kruppel-like factors (KIF1, − 4, − 6, − 7, − 8, − 11, − 16, and Chromosomal distribution and genomic localization of KLF22) were unique to the 2C_Flush blastocysts exclusively differentially methylated loci in blastocysts whereas a total of 3410 DMRs including ankyrin repeat developed in vitro from embryos flushed before, during domain-containing protein (ANKRD6,-16, − 24, − 28, or after embryonic genomic activation − 35, − 39 and ANKRD60) were unique to the To further identify chromosomal regions sensitive to the in 8C_Flush group. Conversely, a total of 2889 DMRs vitro culture condition, we looked into the chromosomal including elongation factors (EIF2AK1, EIF3C and distribution of the DMRs induced exclusively before, during Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 9 of 19 Fig. 7 a The scatter plot displaying expression patterns and the methylation levels of DMRs commonly detected in the 2C_Flush and 8C_Flush groups in blastocysts of the 2C_Flush. The DNA methylation and mRNA expression levels are indicated in the X and Y axis, respectively. b The gene network and biological processes enriched by these DMRs whose expression levels arenegatively correlated with mRNA expression levels or after EGA. To perform this, we first arranged the Uniquely differentially methylated genomic loci were chromosome numbers based on the total probe and DMR subjected to the genomic localization and genomic en- densities (Additional file 3:FigureS1).The frequencyofthe richment analysis. The results indicated that 72.7, 71.4 DMRs in the chromosomes tended to follow the probe and 74.2% of the DMRs in the 2C_Flush, 8C_Flush and density available in the EDMA with minor deviations. 16C_Flush blastocysts, respectively were located in and However, the number of differentially methylated genomic around the gene body regions, whereas the rest were lo- loci at X chromosome was 1.8 and 2.6 times higher in the calized in the intergenic regions (Additional file 5: Figure 2C_Flush compared to the 8C_Flush and 16C_Flush S3). Within the gene body regions, higher proportions of groups, respectively. This has been also evidenced in the unique DMRs were located in the intronic followed by circular circos plot which shows the distribution of signifi- exonic regions whereas the proportion of DMRs on the cantly differentially methylated genomic loci across the plain and proximal promoter was relatively lower in all chromosomes (Additional file 4: Figure S3). In addition, blastocyst groups. However, the hypermethylated gen- chromosomal enrichment analysis (analysis of hyper- omic loci exceeded the hypomethylated ones in the ex- methylated probes with reference to hypomethylated onic regions of all blastocyst groups and the probes) of the DMRs revealed that unlike the 8C_Flush hypermethylated genomic loci surpassed the hypomethy- group, the proportion of hypermethylated loci exceeded lated ones in the plain promoter, proximal promoter, the hypomethylated ones in all chromosomes in the distal promoter and intronic regions of the 2C_Flush 2C_Flush and 16C_Flush groups. However, in the and 16C_Flush blastocyst groups (Fig. 9). 8C_Flush, unlike other groups, four times higher hypo- In order to investigate the biological relevance of the methylated probes than hypermethylated loci were de- DMRs exclusively induced in blastocysts derived from tected on X chromosome (Fig. 8b). embryos subjected to the in vitro culture condition before, Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 10 of 19 Fig. 8 a the number and b chromosomal enrichment of genomic loci differentially methylated exclusively in 2C_Flush, 8C_Flush or 16C_Flush blastocyst group. Chr, Chromosome, Log FC: the ratio of hypermethylated and hypomethylated genomic loci described in Log scale 2 2 during or after major EGA, annotated DMRs were differentially expressed genes and the methylation pat- subjected to pathway analysis. Results showed that differ- terns of 94 DMRs were inversely correlated with their cor- entially methylated genes only in the 2C_Flush were sig- responding mRNA expression patterns (Fig. 11a). Among nificantly enriched in 15 biological pathways including these, the hypomethylated genes including SLC16A1, Hippo signaling pathway. On the other hand, the genes ABHD12 and ACOT4 were upregulated while hyper- differentially methylated only in the 8C_Flush group were methylated genes including SLC16A13, ABAT, ALDH4A1 significantly enriched in steroid biosynthesis, axon guid- and PPP2R2B were downregulated. ance, calcium signaling and focal adhesion pathways Functional classification of these genes showed that (Fig. 10). However, exclusively differentially methylated the DMRs which exhibited opposite patterns to the genes in the 16C_Flush group didn’t display any signifi- mRNA expression levels were found to be significantly cant enrichment in any of the biological pathways. enriched in 7 biological processes (Fig. 11b) including, plasma membrane lactate transport (SLC16A13), ala- The expression patterns of exclusively differentially nine, aspartate and glutamate metabolism (ABAT, methylated genomic loci in blastocysts developed in vitro ALDH4A1) and phosphorus metabolic processes from embryos flushed before, during or after embryonic (GLIPR2, CTNS, ABHD12, PPP2R2B, PROM2, SERINC3 genomic activation and ACOT4). Comparative analysis of DNA methylation and gene ex- In the 8C_Flush blastocyst group, a total of 67 pression showed that a total of 179 exclusively differen- genes including those differentially methylated in pro- tially methylated genes in 2C_Flush were mapped to 167 moter regions (PBK, TXNL4A, UPK1A and ZFYVE16), Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 11 of 19 Fig. 9 Genomic enrichment of the DMRs which are unique to each blastocyst group. Log FC: the ratio of hypermethylated and hypomethylated genomic regions described in Log scale in intronic region (ATP6V0A4, NUDT3 and TNPO3) inversely related to the gene expression patterns and exonic region (NR1H2, TSR1 and EIF6) were in- (Fig. 13). versely correlated with the gene expression (Fig. 12a). Functional annotation of these genes indicated that Quantification of the DNA methylation patterns using biological processes including ribosome biogenesis, bisulfite sequencing protein ubiquitination, vacuolar acidification, regula- The methylation patterns and the expression level of selected tion of fatty acid biosynthetic process, steroid biosyn- candidate differentially methylated imprinted (GRB10, PEG3 thesis and diphosphoinositol polyphosphate catabolic and IGF2R) and non-imprinted genes (ITPK1, COL4A1 and process were found to be altered (Fig. 12b). However, TRAPPC9) were analyzed using bisulfite sequencing (Fig. 14). in the 16C_Flush group, only 15 differentially methyl- Among genes analyzed, GRB10 which was hypermethylated ated genes including PGAP1, and PRDX4 were in 2C_Flush, 8C_Flush and 16C_Flush blastocyst groups Fig. 10 Pathways enriched by the DMRs exclusively detected only in the 2C_Flush (a) and 8C_Flush (b) blastocyst groups. DMRs:TGS, the ration of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 12 of 19 Fig. 11 a The expression patterns of the DMRs exclusively detected in the 2C_Flush blastocysts group. DM; differentially methylated, DE, differentially expressed. b The pathways enriched by DMRs whose DNA methylation patterns were inversely correlated with the mRNA expression levels. DMRs:TGS, the ration of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway compared to the in vivo counterparts, and the bisulfite se- qPCR using the total RNA isolated from independent quencing also confirmed hypermethylation of this gene in samples. The results indicated that of 15 out 18 compar- the 8C_Flush and 16C_Flush blastocyst groups. In addition, isons were found to be in line to the array data (Fig. 15). according to the array result TRAPPC9, COL4A1 and ITPK1 Furthermore, the qPCR data indicated that the expres- were hypermethylated in 2C_Flush, 8C_Flush and the bisul- sion levels of the hypermethylated genes, namely fite sequencing indicated relatively higher methylation of TRAPPC9, COL4A1, PEG3, IGFR2 in 2C_Flush and these genes in those samples relative to the vivo blastocysts. 8C_Flush (Fig. 14) were found to be downregulated but Therefore, results of the bisulfite sequencing were similar to the expression level of GRB10 was upregulated (Fig. 13). the array data indicating the validity of the array analysis. Validation of differentially expressed genes using Discussion quantitative polymerase chain reaction (qPCR) In vitro embryo production is one of one of the most For comparative analysis of the DNA methylation and important innovations and one of the sustainable strat- expression profile, we have selected genes which showed egies to meet the global milk and meat demands [23]. a difference of log fold change ≥0.56, or log fold Moreover, in vitro embryo production has been one of 2 2 change ≤− 0.56 with p value < 0.05 in the previous data- the innovation used to foster farm animal breeding and set. Therefore, the expression patterns of nine candidate to make a significant progress in human and veterinary differentially expressed genes were validated using the medicine [24]. However, several evidences indicated that Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 13 of 19 Fig. 12 a The expression patterns of DMRs exclusively detected only in the 8C_Flush blastocyst group. DM; differentially methylated, DE, differentially expressed. b Pathways enriched by DMRs which were exclusively detected only in the 8C_Flush blastocyst group and whose methylation patterns were inversely correlated with the expression levels. DMRs:TGS, the ratio of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway Fig. 13 The gene expression patterns of the DMRs unique to 16C_Flush blastocyst group. DM; differentially methylated, DE, differentially expressed Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 14 of 19 Fig. 14 Bisulfite sequencing results of the DMRs identified by the EmbryoGENE DNA Methylation Array in vitro culture condition are not exactly mimicking or in embryonic stages subjected to in vitro culture condi- replacing the in vivo situation. As the result of this, only tion before or during the onset embryonic genome acti- 30–40% of the in vitro produced embryos are developing vation compared to those exposed after the onset of to the blastocysts stage [25]. In some cases, embryos de- embryonic genome activation. Interestingly, the blasto- veloped under in vitro culture condition can be of low cysts from these embryo groups exhibited a higher DNA quality compared to those developed in vivo and these methylation dysregulation. Indeed, a detailed analysis of differences are further manifested by the darker cyto- the DMRs indicated the presence of commonly affected plasm, low cryotolerance and lower pregnancy rate after genomic regions in all or at least in the two blastocyst transfer to recipients [26, 27]. Several evidences have groups irrespective of stage-specific exposure to in vitro demonstrated that these phenotypic differences are crit- culture condition. Accordingly, the DNA methylation ically affected by the post- fertilization embryo culture patterns of several arrays of genomic loci including those condition which can be subsequently manifested by involved in the focal adhesion and metabolic pathways aberrant expression and DNA methylation pattern of were commonly altered in all blastocyst groups suggest- developmentally related genes [11–14, 17]. However, ing that these genomic regions are sensitive to the envir- identification of the embryonic stages which are sensitive onmental insults at any stages of preimplantation to the environmental insults remains a challenge. In- Table 2 List of commonly differentially methylated imprinted deed, establishing an appropriate model which mimics genes and DNMTs the optimal and suboptimal embryo culture condition is Gene symbol 2C_Flush 8C_Flush 16C_Flush Genomic region indispensable to unravel the phenotypic and molecular PEG3 ▲▲▲ Gene body abnormalities induced in embryo subjected to environ- mental challenges. Therefore, embryos developed in vivo GRB10 ▲▲▲ Gene body until 2-, 8- or 16-cell stages were subjected to in vitro IGF2R ▲▲▲ Gene body culture until the blastocyst stage. These embryo groups DNMT3B ▲▲ - Gene body are believed to represent the developmental stages of be- DNMT1 ▼▲▲ Gene body fore, during and after the onset of major embryonic gen- ASB4 ▲▼▼ Gene body ome activation, respectively. In fact, developmental data SFMBT2 ▲▼▼ Gene body indicated that the proportion of embryos developed to the blastocyst stage, 7 days post-insemination was lower ▲ Hypermethylated, ▼ Hypomethylated Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 15 of 19 Fig. 15 The qPCR validation results of the selected differentially expressed genes generated by EmbryoGENE transcriptome microarray embryonic development. For instance, the imprinted during embryonic development was also resulted in over- genes such as PEG3, IGF2R, SNRRP and GRB10 (Table 2) growth of both the embryo and placenta in human [32]. were aberrantly methylated in all blastocysts and inde- In addition to imprinted genes, cluster of genes such as pendent bisulfite sequencing. We also observed differ- the solute carriers and the zinc finger proteins were com- ences in the DNA methylation level of PEG3, IGF2R and monly differentially methylated in all blastocyst groups. GRB10 genes between the blastocyst groups. Culture The solute carriers are believed to be essential for trans- condition induced aberrant methylations of the porting cellular materials across the cell membrane and al- imprinted genes in preimplantation embryos have been tered DNA methylation of the solute carrier genes in the highlighted in many instances. For instance, culture developing embryos could negatively affect the proper func- condition-induced aberrant DNA methylation pattern of tioning of protein transportation across the cell membrane. SNRRP in bovine embryos [28] and PEG3 gene in mouse In addition to these, aberrant DNA methylation of the placenta [29] have been indicated. Moreover, aberrant zinc-finger proteins such as ZNF234, ZNF618, ZNF688 and DNA methylation pattern of PEG3, IGF2R and GRB10 ZNF862 are interesting as the zinc proteins are found to in children conceived by IVF/ICSI [30] also described further regulate the DNA methylation patterns of some the long term effect of suboptimal culture condition on other genes by targeting the M.CviPI (GC methylation) or the DNA methylation pattern of imprinted genes. This, M.SssI (CG methylation) sites [33]. On the other hand, could in turn, reduce embryonic developmental compe- extracellular-matrix related genes including COL4A1 and tence and/or result in fetal abnormalities. For instance, ab- COL1A2 were also differentially methylated in all blastocyst errant methylation of IGF2R could cause large offspring groups of which the COL4A1 gene was hypermethylated in syndrome in sheep [31] and functional loss of GRB10, all blastocyst groups and its expression pattern was Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 16 of 19 downregulated. The DNA methylation and expression pat- beginning of de novo methylation appears to overlap the terns of COL4A1 were further confirmed by bisulfite se- genome activation occurring at the same cell stage. quencing and qPCR, respectively using independent Nevertheless, whether this pattern remains similar while blastocyst samples. Accordingly, COL4A1 gene was embryos are exposed to the suboptimal environmental hypermethylated, but its expression was downregulated in condition is not clear. However, the study in rabbit em- blastocyst obtained from 2- and 8-cell flush embryos com- bryo showed that DNA demethylation in in vitro pro- pared to completely in vivo developed blastocysts. The duced embryos occurs between the 2- and 8-cell stages functional relevance of these gene in bovine embryo de- while in in vivo ones, demethylation occurs between 4- velopment is barely understood, but one study showed and 16-cell stages and remains constant from 16-cell that functional loss of COL4A1 and COL4A2 resulted in until the morula stage suggesting that demethylation embryonic lethality in mouse embryo due to impaired and remethylation of the genome occurs faster in in basement membrane stability [34]. Moreover aberrant vitro embryos than the in vivo ones. Similarly, the study DNA methylation pattern of COL4A1 gene in the gene by Desmet [28] also indicated differences in the DNA body was associated with cancer [35]. methylation changes when early stage embryos are ex- Apart from the DNA marks altered in all blastocyst posed to suboptimal culture suggesting that early-stage groups; we have also analyzed the DNA methylation embryos could be more susceptible to adverse culture marks, which are specific to each of the blastocyst condition. group. Unlike the blastocysts of the 8-cell flush, the blas- In addition to the holistic approaches, we have also tocysts of the 2- and 16-cell flush groups were associated looked into the developmental relevance of uniquely dif- with more hypermethylated genomic loci than the hypo- ferentially methylated individual genes in each blastocyst methylated ones, suggesting that exposure of embryos group. For instance, some of the Krüppel-like transcrip- before or after embryonic genome activation could in- tion factors (KLFs) were differentially methylated in blas- duce DNA methylation, while exposing at the time of tocysts of the 2-cell flush group. The KLFs are believed to genome activation could foster demethylation of the be evolutionarily conserved and potentially involved in resulting blastocysts. Indeed, the interdependence be- several cellular activities such as proliferation, differenti- tween the onset of genome activation and DNA methy- ation and cellular development [46]. Thesefamiliesof lation patterns during mammalian embryogenesis is genes are consisting of three Krüppel-like zinc fingers barely understood. In fact, the genome DNA undergoes which bind to CACCC elements and GC-rich regions of multiples epigenetic modifications immediately after DNA for mediating the gene transcription [47]. Among fertilization which may in turn affect the gene expression the KLFs, KIF4 was among the differentially methylated patterns that could be established during and after em- KLFs in blastocysts of the 2-cell flush. KLF4 is involved in bryonic genome activation. For instance, active and pas- cell proliferation, apoptosis and invasion [48] and believed sive demethylation of the pronuclei occurs during to be negatively regulating the expression of endodermal mouse, bovine and human embryonic development, but markers which may be involved in the transformation of the timing of re-methylation is different depending on pluripotent stem cells to endoderm differentiation [49]. In the species [8, 36, 37]. In mice and humans, massive de addition, KLF4 regulates the NANOG and OCT3/4 tran- novo methylation occurs at the blastocyst stage [8, 38], scription during embryonic stem cell self-renewal and whereas in bovine, the major embryo genome activation subsequently block the embryonic stem cell differentiation is believed to occur between the 8- and 16-cell stage [50, 51]. Therefore, developmental retardation of the [39] although the bovine embryo has been found to be 2-cell flush may partly be induced by aberrant DNA transcriptionally competent at the 2-cell stage [40]. Con- methylation pattern of the KLFs. ceptually, for genes to be expressed, the DNA sequence Overall, in this study we have identified several DMRs must be accessible for the transcriptional machinery and which are sensitive to culture condition at any stage of any blockage of the transcription machinery could po- embryonic development. Hypermethylated or hypomethy- tentially shutdown the gene expression [41]. Among lated genomic loci are generally believed to play a role in these, hypomethylation (removal of the DNA methyla- regulating cellular development and function by inhibiting tion marks) around the promoter region is believed to or inducing the gene expression. Therefore, in this study, increase the transcriptional activity of the genes [42–44]. apart from exploring the molecular pathways enriched by During normal embryo development, the fertilized egg the DMRs, we also investigated the effects of those DMRs undergoes multiple epigenetic modifications, which sub- on the gene expression patterns. Accordingly, we have de- sequently affect the resulting gene expression profiles. In tected about 2–9% of annotated DMRs of in the promoter mice, while the paternal pronucleus undergoes active de- or gene body regions were inversely correlated with the methylation, the maternal genome passively demethy- corresponding gene expression (Figs. 5, 7, 11, 12 & 13). lated after the 2-cell stage [45]. However, in bovine, the Interestingly, these genes were found to be involved in key Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 17 of 19 pathways that are believed to be essential for embryogen- Additional file 4: Figure S2. The circos plot representing the overall esis such as fatty acid biosynthetic process, steroid biosyn- methylation levels in the in different blastocyst groups. Positive and negative fold-changes represent the level of hypermethylation and hypomethylation. thesis, energy metabolism and JAK-STAT cascade (PDF 19412 kb) pathways. Indeed, although the proportion of DMRs that Additional file 5: Figure S3. Genomic distribution of exclusively inversely correlated with the gene expression seems to be differentially methylated regions in the 2C_Flush, 8C_Flush or 16C_Flush lower while considering the total number of genomic loci blastocyst group. (PDF 47 kb) differentially methylated, it should be noted that there were several DMRs whose expression was positively corre- Abbreviations lated with gene expression (data not shown). On the other 16C_Flush: blastocysts developed in vitro from embryos flushed at 16-cell stage; 2C_Flush: blastocysts developed in vitro from embryos flushed at 2- hand, the DMRs whose mRNA was not changed does not cell stage; 8C_Flush: blastocysts developed in vitro from embryos flushed at necessarily imply that their methylation pattern is not 8-cell stage; CpG: Cytosine phosphate guanine; DE: Differentially expressed; essential for mRNA expression, rather these DMRs DM: Differentially methylated; DMRs: Differentially methylated genomic regions; EDMA: Embryogene DNA methylation array; EGA: Embryonic could be involved in regulating the embryonic devel- genome activation; EGFR: Epidermal growth factor receptor; FDR: False opment by regulating the downstream genes or by discovery rate; ICSI: Intracytoplasmic sperm injection; IVF: In vitro fertilization; fine-tuning the expression pattern of other gene regu- qPCR: Quantitative real time polymerase chain reaction latory agents such as miRNAs. Funding This study was funded by EmbryoGENE. The funding body has no role in the Conclusion designing the experiment, collection of data, analysis interpretation of data, This study identified culture-induced stage-specific and writing the manuscript. non-stage-specific aberrant DNA methylation patterns Availability of data and materials of paternally or maternally imprinted genes and several All data used and/or analyzed during the present study are available from arrays of genomic regions in in vivo originated bovine the corresponding author if requested. embryos subjected to the in vitro culture in stage spe- cific manner. Thus, embryos exposed to in vitro culture Authors’ contributions condition before the onset or during embryonic genome DSW has performed experiment, performed sample processing and array hybridization, data analysis, interpreting the results, drafting and writing the activation are more sensitive to the DNA methylation manuscript and serve as the corresponding author. MH was involved in marks changes that could be further displayed in the endoscopic transfer & flushing, design of the experiment. ET involved in resulting blastocysts compared to those exposed to in designing the experiment and revised manuscript. MSZ performed qPCR analysis and critically read the manuscript. EH collected the samples and vitro culture condition after embryonic genome activa- critically revised the manuscript. SG, MP and HOP and CN critically revised the tion. Thus, while exposure of in vivo embryos to in vitro manuscript. FR did in vitro culture embryo and sample processing. UB & VH culture condition before embryonic genome activation performed endoscopic flushing of embryos and designed the experiment. EF performed statistical analysis of differentially methylated regions and favored the initiation of DNA methylation, subjecting of interpreted the data. DG performed array hybridization and involved in in vivo embryos to the in vitro culture at the time of interpretation of the results. AG did transcriptome profile analysis. MA and CR genome activation increased the hypomethylated gen- involved in experimental design, supervised during array hybridization and revised the manuscript. KS designed the experiment, supervised the omic loci in the resulting blastocysts. Nevertheless, a de- experiments and reviewed the manuscript. DT was responsible for designing tailed study is required to understand whether these the experiment, interpretation of the results and reviewing the manuscript. All altered DNA methylation patterns induced in the em- authors approved the final version of the paper. bryo by suboptimal culture stimuli are initiated as means Ethics approval of coping mechanisms of the embryo or an indicator of The experiment was approved by the Animal Welfare committee of the abnormal embryo development. Moreover, future studies University of Bonn with proposition number 84–02.05.20.12.075. could further investigate whether or not the suboptimal culture-induced epigenetic modifications occurred dur- Competing interests ing bovine preimplantation embryo at the blastocysts The authors declare that they have no competing interests. stage could have a long-lasting consequence on the off- spring epigenome. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional files Author details Additional file 1: Table S1. The list of primers used to validate DMRs in Institute of Animal Science, Animal Breeding and Husbandry Group, blastocysts of different groups using bisulfite sequencing. (DOCX 15 kb) University of Bonn, 53115 Bonn, Germany. Institute of Animal Breeding and Genetics, University of Veterinary Medicine Vienna, A-1210 Vienna, Austria. Additional file 2: Table S2. The list of primers used for validation of the Centre de recherche en biologie de la reproduction, Faculté des sciences differentially expressed genes using qPCR. (DOC 33 kb) de l’agriculture et de l’alimentation, INAF, Pavillon des services, Université Additional file 3: Figure S1. The order of chromosomes based their Laval, Québec G1V 0A6, Canada. Department of Animal Production, Faculty probe density and differentially methylated probes (TIF 4242 kb) of Agriculture, Cairo University, Giza 12613, Egypt. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 18 of 19 Received: 18 September 2017 Accepted: 25 May 2018 21. Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004;3:Article3. 22. Gad A, Hoelker M, Besenfelder U, Havlicek V, Cinar U, Rings F, et al. Molecular mechanisms and pathways involved in bovine embryonic References genome activation and their regulation by alternative in vivo and in vitro 1. Bedzhov I, Leung CY, Bialecka M, Zernicka-Goetz M. In vitro culture of culture conditions. Biol Reprod. 2012;87:100. mouse blastocysts beyond the implantation stages. Nat Protoc. 2014;9: 23. Blondin P. Status of embryo production in the world. Animal Reprod. 2015; 2732–9. 12:356–8. 2. Feuer S, Rinaudo P. Preimplantation stress and development. Birth defects 24. Duszewska AM, Trzeciak P, Compa A, Rąpała Ł. Selected issues concerning Res C Embryo Today. 2012;96:299–314. biotechnology of farm animals breeding - a review. Anim Sci Pap Rep. 2010; 3. Maillo V, Lopera-Vasquez R, Hamdi M, Gutierrez-Adan A, Lonergan P, Rizos 28:295–306. D. Maternal-embryo interaction in the bovine oviduct: evidence from in 25. Camargo LSDA, Viana JHM, Sá WFD, Ferreira ADM, Ramos ADA. Factors vivo and in vitro studies. Theriogenology. 2016;86:443–50. influencing in vitro embryo production. Animal Reprod. 2006;3:19–28. 4. Zheng LL, Tan XW, Cui XZ, Yuan HJ, Li H, Jiao GZ, et al. Preimplantation 26. Rizos D, Ward F, Boland MP, Lonergan P. Effect of culture system on the maternal stress impairs embryo development by inducing oviductal apoptosis yield and quality of bovine blastocysts as assessed by survival after with activation of the Fas system. Mol Hum Reprod. 2016;22:778–90. vitrification. Theriogenology. 2001;56:1–16. 5. Duc-Goiran P, Mignot TM, Bourgeois C, Ferre F. Embryo-maternal 27. Lonergan P, Rizos D, Gutierrez-Adan A, Fair T, Boland MP. Effect of culture interactions at the implantation site: a delicate equilibrium. Eur J Obstet environment on embryo quality and gene expression - experience from Gynecol Reprod Biol. 1999;83:85–100. animal studies. Reprod BioMed Online. 2003;7:657–63. 6. Herrler A, von Rango U, Beier HM. Embryo-maternal signalling: how the 28. Desmet K, van Hoeck V, Gagne D, Fournier E, Thakur A, O'Doherty AM, et al. embryo starts talking to its mother to accomplish implantation. Reprod Exposure of bovine oocytes and embryos to elevated non-esterified fatty BioMed Online. 2003;6:244–56. acid concentrations: integration of epigenetic and transcriptomic signatures 7. Lonergan P, Forde N. Maternal-embryo interaction leading up to the initiation in resultant blastocysts. BMC Genomics. 2016;17:1004. of implantation of pregnancy in cattle. Animal. 2014;8(Suppl 1):64–9. 29. de Waal E, Mak W, Calhoun S, Stein P, Ord T, Krapp C, et al. In vitro culture 8. Dean W, Santos F, Stojkovic M, Zakhartchenko V, Walter J, Wolf E, Reik W. increases the frequency of stochastic epigenetic errors at imprinted genes Conservation of methylation reprogramming in mammalian development. in placental tissues from mouse concepti produced through assisted Aberrant reprogramming in cloned embryos. PNAS. 2001;98:13734–8. reproductive technologies. Biol Reprod. 2014;90:22. 9. Santos F, Hyslop L, Stojkovic P, Leary C, Murdoch A, Reik W, Stojkovic M, 30. Lazaraviciute G, Kauser M, Bhattacharya S, Haggarty P, Bhattacharya S. A Herbert M, Dean W. Evaluation of epigenetic marks in human embryos systematic review and meta-analysis of DNA methylation levels and derived from IVF and ICSI. Hum Reprod. 2010;25:2387–95. imprinting disorders in children conceived by IVF/ICSI compared with 10. Calle A, Fernandez-Gonzalez R, Ramos-Ibeas P, Laguna-Barraza R, Perez- children conceived spontaneously. Hum Reprod Update. 2015;21:555–7. Cerezales S, Bermejo-Alvarez P, et al. Long-term and transgenerational effects 31. Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG, Carolan C, of in vitro culture on mouse embryos. Theriogenology. 2012;77:785–93. et al. Epigenetic change in IGF2R is associated with fetal overgrowth after 11. Lou H, Le F, Zheng Y, Li L, Wang L, Wang N, et al. Assisted reproductive sheep embryo culture. Nat Genet. 2001;27:153–4. technologies impair the expression and methylation of insulin-induced 32. Charalambous M, Smith FM, Bennett WR, Crew TE, Mackenzie F, Ward A. gene 1 and sterol regulatory element-binding factor 1 in the fetus and Disruption of the imprinted Grb10 gene leads to disproportionate placenta. Fertil Steril. 2014;101:974–80. overgrowth by an Igf2-independent mechanism. Proc Natl Acad Sci. 2003; 12. Gomes MV, Huber J, Ferriani RA, Amaral Neto AM, Ramos ES. Abnormal 100:8292–7. methylation at the KvDMR1 imprinting control region in clinically normal 33. Carvin CD, Parr RD, Kladde MP. Site-selective in vivo targeting of children conceived by assisted reproductive technologies. Mol Hum Reprod. cytosine-5 DNA methylation by zinc-finger proteins. Nucleic Acids Res. 2009;15:471–7. 2003;31:6493–501. 13. Ghosh J, Coutifaris C, Sapienza C, Mainigi M. Global DNA methylation levels 34. Pöschl E, Schlötzer-Schrehardt U, Brachvogel B, Saito K, Ninomiya Y, Mayer are altered by modifiable clinical manipulations in assisted reproductive U. Collagen IV is essential for basement membrane stability but dispensable technologies. Clin Epigenetics. 2017;9:14. for initiation of its assembly during early development. Development. 2004; 14. Dean W, Bowden L, Aitchison A, Klose J, Moore T, Meneses JJ, et al. Altered 131:1619–28. imprinted gene methylation and expression in completely ES cell-derived 35. Maekawa R, Sato S, Yamagata Y, Asada H, Tamura I, Lee L, et al. Genome-wide mouse fetuses: association with aberrant phenotypes. Development. 1998; DNA methylation analysis reveals a potential mechanism for the pathogenesis 125:2273–82. and development of uterine leiomyomas. PLoS One. 2013;8:e66632. 15. Deshmukh RS, Østrup O, Østrup E, Vejlsted M, Niemann H, Lucas-Hahn A, et 36. Mayer W, Niveleau A, Walter J, Fundele R, Haaf T. Demethylation of the al. DNA methylation in porcine preimplantation embryos developed in vivo zygotic paternal genome. Nature. 2000;403:501–2. and produced by in vitro fertilization, parthenogenetic activation and 37. Beaujean N, Hartshorne G, Cavilla J, Taylor J, Gardner J, Wilmut I, Meehan R, somatic cell nuclear transfer. Epigenetics. 2011;6:177–87. Young L. Non-conservation of mammalian preimplantation methylation 16. Hiendleder S, Wirtz M, Mund C, Klempt M, Reichenbach HD, Stojkovic M, et dynamics. Curr Biol. 2004;14:R266–7. al. Tissue-specific effects of in vitro fertilization procedures on genomic cytosine methylation levels in overgrown and normal sized bovine fetuses. 38. Guo F, Li X, Liang D, Li T, Zhu P, Guo H, Wu X, Wen L, Gu T-P, Hu B, Walsh Biol Reprod. 2006;75:17–23. CP, Li J, Tang F, Xu G-L. Active and passive demethylation of male and 17. Salilew-Wondim D, Fournier E, Hoelker M, Saeed-Zidane M, Tholen E, Looft female pronuclear DNA in the mammalian zygote. Cell Stem Cell. 2014;15:447–59. C, et al. Genome-wide DNA methylation patterns of bovine blastocysts 39. Barnes FL, First NL. Embryonic transcription in in vitro cultured bovine developed in vivo from embryos completed different stages of embryos. Mol Reprod Dev. 1991;29:117–23. development in vitro. PLoS One. 2015;10:e0140467. 40. Memili E, Dominko T, First NL. Onset of transcription in bovine oocytes and preimplantation embryos. Mol Reprod Dev. 1998;51:36–41. 18. Gad A, Besenfelder U, Rings F, Ghanem N, Salilew-Wondim D, Hossain MM, 41. Frith MC, Valen E, Krogh A, Hayashizaki Y, Carninci P, Sandelin A. A code for et al. Effect of reproductive tract environment following controlled ovarian transcription initiation in mammalian genomes. Genome Res. 2008;18:1–12. hyperstimulation treatment on embryo development and global transcriptome profile of blastocysts: implications for animal breeding and 42. Sizemore ST, Sizemore GM, Booth CN, Thompson CL, Silverman P, Bebek G, human assisted reproduction. Hum Reprod. 2011;26:1693–707. Abdul-Karim FW, Avril S, Keri RA. Hypomethylation of the MMP7 promoter 19. Besenfelder U, Brem G. Tubal transfer of bovine embryos: a simple and increased expression of MMP7 distinguishes the basal-like breast cancer endoscopic method reducing long-term exposure of in vitro produced subtype from other triple-negative tumors. Breast Cancer Res Treat. 2014; embryos. Theriogenology. 1998;50:739–45. 146:25–40. 20. Shojaei Saadi HA, O'Doherty AM, Gagne D, Fournier E, Grant JR, Sirard M-A, 43. Haney SL, Upchurch GM, Opavska J, Klinkebiel D, Hlady RA, Suresh A, et al. Promoter Robert C. An integrated platform for bovine DNA methylome analysis Hypomethylation and expression is conserved in mouse chronic lymphocytic suitable for small samples. BMC Genomics. 2014;15:451. leukemia induced by decreased or inactivated Dnmt3a. Cell Rep. 2016;15:1190–201. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 19 of 19 44. Kobayashi T, Ishida K, Yoshie H. Increased expression of interleukin-6 (IL-6) gene transcript in relation to IL-6 promoter hypomethylation in gingival tissue from patients with chronic periodontitis. Arch Oral Biol. 2016;69:89–94. 45. Haaf T. Methylation dynamics in the early mammalian embryo: implications of genome reprogramming defects for development. Curr Top Microbiol Immunol. 2006;310:13–22. 46. McConnell BB, Yang VW. Mammalian Kruppel-like factors in health and diseases. Physiol Rev. 2010;90:1337–81. 47. Pearson R, Fleetwood J, Eaton S, Crossley M, Bao S. Kruppel-like transcription factors: a functional family. Int J Biochem Cell Biol. 2008;40:1996–2001. 48. Yang Y, Goldstein BG, Chao H-H, Katz JP. KLF4 and KLF5 regulate proliferation, apoptosis and invasion in esophageal cancer cells. Cancer Biol Ther. 2005;4:1216–21. 49. Aksoy I, Giudice V, Delahaye E, Wianny F, Aubry M, Mure M, et al. Klf4 and Klf5 differentially inhibit mesoderm and endoderm differentiation in embryonic stem cells. Nat Commun. 2014;5:3719. 50. Parisi S, Passaro F, Aloia L, Manabe I, Nagai R, Pastore L, Russo T. Klf5 is involved in self-renewal of mouse embryonic stem cells. J Cell Sci. 2008;121: 2629–34. 51. Zhang P, Andrianakos R, Yang Y, Liu C, Lu W. Kruppel-like factor 4 (Klf4) prevents embryonic stem (ES) cell differentiation by regulating Nanog gene expression. J Biol Chem. 2010;285:9180–9. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Genomics Springer Journals

Genome-wide DNA methylation patterns of bovine blastocysts derived from in vivo embryos subjected to in vitro culture before, during or after embryonic genome activation

Free
19 pages
Loading next page...
 
/lp/springer_journal/genome-wide-dna-methylation-patterns-of-bovine-blastocysts-derived-sms9meLWK1
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Life Sciences; Life Sciences, general; Microarrays; Proteomics; Animal Genetics and Genomics; Microbial Genetics and Genomics; Plant Genetics and Genomics
eISSN
1471-2164
D.O.I.
10.1186/s12864-018-4826-3
Publisher site
See Article on Publisher Site

Abstract

Background: Aberrant DNA methylation patterns of genes required for development are common in in vitro produced embryos. In this regard, we previously identified altered DNA methylation patterns of in vivo developed blastocysts from embryos which spent different stages of development in vitro, indicating carryover effects of suboptimal culture conditions on epigenetic signatures of preimplantation embryos. However, epigenetic responses of in vivo originated embryos to suboptimal culture conditions are not fully understood. Therefore, here we investigated DNA methylation patterns of in vivo derived bovine embryos subjected to in vitro culture condition before, during or after major embryonic genome activation (EGA). For this, in vivo produced 2-, 8- and 16-cell stage embryos were cultured in vitro until the blastocyst stage and blastocysts were used for genome-wide DNA methylation analysis. Results: The 2- and 8-cell flushed embryo groups showed lower blastocyst rates compared to the 16-cell flush group. This was further accompanied by increased numbers of differentially methylated genomic regions (DMRs) in blastocysts of the 2- and 8-cell flush groups compared to the complete in vivo control ones. Moreover, 1623 genomic loci including imprinted genes were hypermethylated in blastocyst of 2-, 8- and 16-cell flushed groups, indicating the presence of genomic regions which are sensitive to the in vitro culture at any stage of embryonic development. Furthermore, hypermethylated genomic loci outnumbered hypomethylated ones in blastocysts of 2- and 16-cell flushed embryo groups, but the opposite occurred in the 8-cell group. Moreover, DMRs which were unique to blastocysts of the 2-cell flushed group and inversely correlated with corresponding mRNA expression levels were involved in plasma membrane lactate transport, amino acid transport and phosphorus metabolic processes, whereas DMRs which were specific to the 8-cell group and inversely correlated with corresponding mRNA expression levels were involved in several biological processes including regulation of fatty acids and steroid biosynthesis processes. Conclusion: In vivo embryos subjected to in vitro culture before and during major embryonic genome activation (EGA) are prone to changes in DNA methylationmarks andexposureofinvivoembryos to in vitroculture during thetimeof EGA increased hypomethylated genomic loci in blastocysts. Keywords: DNA methylation, Gene expression, Culture condition, Bovine embryo * Correspondence: dsal@itw.uni-bonn.de Institute of Animal Science, Animal Breeding and Husbandry Group, University of Bonn, 53115 Bonn, Germany Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 2 of 19 Background responses of embryos derived from oocytes matured and Following fertilization, the mammalian zygote undergoes fertilized in vivo when subjected to in vitro culture condi- successive cleavage divisions in the oviduct and eventually, tions is essential to unravel the epigenetic signatures at the blastocyst stage, the embryo enters to the uterus, induced by environmental challenges during embryo hatched from the zona pellucida and attach to the luminal culture. Therefore, this study aimed to investigate epithelium to initiate implantation [1]. During the early genome-wide DNA methylation patterns of bovine blasto- stage of successive developmental processes, the develop- cysts derived from in vivo embryos subjected to the in ing embryo undergoes signal communication with the ma- vitro culture condition before, during or after embryonic ternal environment [2–4] by secreting signals which genome activation. subsequently initiates the establishment of persistent bidir- ectional interaction with the maternal environment [5–7]. Methods Effects of environmental insults on the survival and de- Animal handling, management and estrus synchronization velopment of the embryo are highly noticeable at early Thirty-eight healthy Simmental heifers aged between 15 stages of development when the culture condition induces and 20 months and weighing between 380 and 500 kg in alterations in genome reprogramming [8, 9] which could the FrankenForst teaching and research station, University even persist until adult life [10]. Several studies indicated of Bonn, were selected for this study for generation of in theeffects of culture conditiononthe DNAmethylation vivo embryos. The animals were housed in a free-stall patterns of embryos. For instance, the DNA methylation barn with slotted floors and cubicles, lined with rubber patterns of key genes such as insulin-induced gene (INSIG) mats and they were fed a total mixed ration. Animal hand- and sterol regulatory element-binding protein (SREBP) ling and management of the experimental animals were were found to be altered in the fetus and placenta in human performed according to the rules and regulations of the pregnancies established after in vitro fertilization [11]. Simi- German law of animal protection. The Animal Welfare larly, aberrant hypomethylation of the maternal imprinting Committee of the University of Bonn approved the experi- control region of the KvDMR1 gene in clinically normal ment with proposition number 84–02.05.20.12.075. children conceived by in vitro fertilization (IVF) and intra- cytoplasmic sperm injection (ICSI) was also identified [12]. In vivo embryo production and in vitro culture Moreover, alteration in the global DNA methylation pat- In this study, different stages of in vivo derived embryos terns of human placenta derived from pregnancies estab- were further cultured in vitro and the resulted blastocysts lished from embryos cultured under high oxygen tension were used for DNA methylation analysis. For this, Simmen- [13], and alterations in the allelic methylation patterns of tal heifers at the FrankenFrost teaching and research sta- maternally or paternally expressed genes in mice embryonic tion, University of Bonn, were estrous synchronized, stem cell fetuses [14] are additional evidences indicating super-ovulated and artificially inseminated as described pre- how suboptimal culture condition could affect the genome viously [18]. A total of 11, 6, 12 and 13 animals were ran- methylation at the later stage of embryo development. In domly assigned for 2-, 8–16-cell and blastocyst stage porcine, the DNA methylation level was increased in in embryo production, respectively. Afterwards, 2-, 8- and vitro embryos compared to in vivo ones suggesting the ad- 16-cell stage embryos were flushed at 48, 72 and 96 h post verse effect of in vitro culture on the DNA methylome of insemination, respectively using tubal embryo recovery and embryos [15]. In bovine, oocyte maturation media was con- transfer methods [19]. The experimental design used for firmed to cause hypermethylation in the liver of fetuses at sample collection is indicated in Fig. 1. These embryo day 50 of the gestation period [16] indicating the long term groups represent the embryonic developmental stage be- effect of culture condition on the DNA methylome of the fore, during and after embryonic genome activation, re- embryo or the fetal genome. spectively. All embryo groups were then in vitro cultured Previously, we have reported a stage-specific alter- until the blastocyst stage and the developmental compe- ations of DNA methylation patterns of in vivo devel- tence of the 2-cell (2C_Flush), 8-cell (8C_Flush) and 16-cell oped blastocysts from embryos developed in vitro up to (16C_Flush) were recorded. In addition, the in vivo blasto- zygote, 4- or 16-cell stages [17]. In that study, we found cyst group (vivo) was obtained from embryos completely that DNA methylation patterns in the resulting blasto- developed under in vivo condition. cysts was compensated in a stage specific manner compared to blastocysts completely developed under in DNA fragmentation, adapter ligation and methyl sensitive vitro condition. Nevertheless, in that study, it was not enzymes digestion possible to predict whether the DNA methylation dys- Genomic DNA (gDNA) from four replicates (10 blasto- regulation was derived from the epigenetic memory cysts per replicate) of the 2C_Flush, 8C_Flush, 16C_Flush caused by the in vitro oocyte maturation, fertilization and vivo blastocyst groups was isolated using the Allprep and culture condition. Therefore, understanding epigenetic DNA/RNA micro kit (Qiagen, Germany) following the Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 3 of 19 Fig. 1 Experimental design used to generate different blastocyst groups for DNA methylation analysis. IVC; In vitro culutre manufacturer’s recommendations. The gDNA was then or 16C_Flush and 1 μg Cy5 or Cy3 labeled amplified fragmented, adapter ligated and cleaved by gDNA of the vivo blastocyst groups were mixed with methyl-sensitive restriction endonucleases using the simi- 157.6 μl hybridization cocktail and incubated at 95 °C for lar methodology that was used previously ([17]. Briefly, 3 min and at 37 °C for 30 min. At the end of incubation, the gDNA of each group was subjected to the MseI en- 65 μl of Agilent-CGHBlock was added and the mix was zyme digestion and the MseLig12 (100 μM) (5′-TAA CTA then transferred onto the EmbryoGENE DNA Methyla- GCA TGC-3′) and 0.5 μl MseLig21 (100 μM) (5-AGT tion Array slides [20]. The slides were then placed in a GGG ATT CCG CAT GCT AGT-3′)adapterswereli- hybridization oven (Shel Lab) for 40 h at 65 °C at with a gated to the fragmented DNA using T4 DNA ligase in the speed of 20 RPM. Four biological replicates of 2C_Flush, presence of 10× One-phor-All buffer PLUS and ATP. 8C_Flush and 16C_Flush blastocyst groups were hybrid- Non-methylated genomic regions were then cleaved using ized to four biological replicates of the in vivo group. A FastDigest™ methyl-sensitive restriction endonucleases total of 12 arrays were hybridized and biological [HpaII (C/CGG), HinP1I (GC/GC) and Aci1I (C/CGC) dye-swaps were performed to avoid any false positive re- (Fermentas, Thermo Fisher Scientific, Germany)]. The ef- sults. The arrays were then washed and scanned using ficiency of methyl-sensitive restriction endonucleases PowerScanner (Tecan) which integrates the Array-Pro cleavage was evaluated based on the qPCR amplification Analyzer 6.3 software (MediaCybernetics). Agilent’sfea- results of the spike-in controls added during the gDNA ture extraction software (Agilent Technologies, USA) was fragmentation step [17, 20]. Samples which appeared after used to extract the array features. 5 cycles compared to undigested controls were used for downstream treatments, otherwise the samples were Array data and downstream analysis digested one more time using specific methyl sensitive re- The array data analysis methodsusedfor this studyare ex- striction enzyme that did not work in the first run. After- tensively described previously [17, 20]. Briefly, Loess wards, unmethylated genomic regions were enriched by inter-array normalization method was used to correct the performing a second round ligation-mediated polymerase dye effects, while inter-array scale normalization was used to chain reaction as indicated previously [20]. balance the distribution differences among experiments. Linear models for microarray data (limma) [21]was used to DNA labeling and hybridization identify differentially methylated regions (DMRs). Probes About 2 μg of amplified gDNA from the 2C_Flush, with absolute log (fold-change) ≥ 1.5 differences between 8C_Flush, 16C_Flush or vivo blastocyst groups was mixed samples to be compared with significant (p <0.05) were with 1 μl Cy-3 or Cy-5 dyes (Kreatech Biotechnology) and considered as differentially methylated regions (DMRs). heated at 85 °C for 30 min. After the end of incubation, Thus, probes which showed a significant increase or de- the samples were purified using the Qiagen PCR purifica- crease in signal intensity by ≥1.5 folds in 2C_Flush, tion kit (Qiagen, Germany). Afterwards, 1 μg of Cy5 or 8C_Flush or 16C_Flush compared to the vivo group were Cy3 labeled gDNA sample from the 2C_Flush, 8C_Flush considered as hypermethylated and hypomethylated probes, Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 4 of 19 respectively. The pathways enriched by the differentially SYBR Green Supermix with ROX (Bio-Rad laboratories, methylated genes were analyzed using Gprofiler (http:// Germany), the cDNA samples of each treatment group biit.cs.ut.ee/gprofiler/). For comparative analysis of the and the specific forward and reverse primer in the Ste- methylation profile and gene expression data, the genes with pOnePlus™ Real-Time PCR Systems (Applied Biosys- absolute log (fold-change) ≥1.5, p value < 0.05 and false dis- tems, USA). The qPCR thermal cycling parameters were covery rate (FDR) < 0.3 were selected from our previous data set to 95 °C for 3 min followed by 40 cycles of 95 °C for (GSE33314 and GSE111990). The heat maps of differentially 15 s and 60 °C for 1 min. Following this, dissociation methylated regions and differentially expressed genes were curve was generated by starting fluorescence acquisition constructed using PermutMatrix (http://www.atgc-montpel- at 60 °C and measurements were taken in every 7 s lier.fr/permutmatrix/) and matrix2png interface (https:// interval until the temperature reached 95 °C. The speci- matrix2png.msl.ubc.ca/bin/matrix2png.cgi). The network of ficity of amplification was evaluated by monitoring the genes was visualized using the networkanalyst tool (http:// dissociation (melting) curve of each candidate gene. The www.networkanalyst.ca/). qPCR data was generated from three independent bio- logical replicates and the data was analyzed using the −ΔΔCt Confirmation of differentially methylated genomic delta Ct method (2 ). The expression level of Glyc- regions using bisulfite sequencing eraldehyde 3-phosphate dehydrogenase (GAPDH) was The DNA methylation levels of selected candidate genes used as an endogenous normalizer. were validated using bisulfite sequencing. For this, the sequences of the candidate gene/probe were retrieved Results using the Site Search tool http://emb-bioinfo.fsaa.ula- DNA methylation marks sensitive to in vitro culture val.ca/bioinfo/html/SiteSearch.html. Afterwards, gene/ condition at any stage of embryonic development sequence specific primers (Additional file 1: Table S1) In this study, the DNA methylation patterns of blastocysts were designed using MethPrimer (http://www.urogen- derived from embryos developed in vivo followed by in e.org/cgi-bin/methprimer/methprimer.cgi) and used for vitro culture were investigated. For this, in vivo developed amplification of the gene of interest. To do this, first the 2-cell (2C_Flush), 8-cell (8C_Flush) and 16-cell genomic DNA of each blastocyst group was modified (16C_Flush) embryos were subjected to in vitro culture with sodium bisulfite converted using EZ DNA methyla- condition until the blastocyst stage. Although the day 9 tion direct kit (ZymoResearch). Afterwards, bisulfite blastocyst rate was similar in the 2C_Flush and 8C_Flush converted gDNA was subjected to PCR amplification groups, the day 7 and day 8 blastocyst rates were relatively using Zymo Taq™ DNA polymerase (ZymoResearch). lower in 2C_Flush embryos. The day 7–9 blastocyst rates The PCR product was purified and ligated to the in the 16C_Flush group were relatively higher compared pGEM®-T Easy Vector System (Promega, USA) and to both the 2C_Flush and 8C_Flush groups (Table 1). transformed to E. coli competent cells. The bacterial cul- Therefore, to get an overview of the epigenome repro- ture was plated onto LB agar/ampicillin/IPTG/X-gal gramming that occurred in these embryo groups, we in- plate and incubated overnight at 37 °C. After overnight vestigated DNA methylation patterns in each of the culture independent white colonies were selected and se- blastocysts group using EmbryoGENE DNA methylation quenced with GenomeLab™ GeXP Genetic Analysis Sys- array platform. Aberrant hyper or hypomethylated gen- tem (Beckman Coulter). The bisulfite sequencing data omic loci were increased in blastocysts derived from em- were then analyzed using the quantification tool for bryos that were exposed to in vitro conditions at 2- and methylation analysis (http://quma.cdb.riken.jp/). 8-cell stages compared to the 16-cell stage indicating that the suboptimal culture condition induced aberrant DNA Quantification of candidate genes using quantitative real methylation pattern more strongly before and during the time PCR embryonic genome activation than later (Fig. 2). Since the DNA methyltion profile data was supperim- posed to previously generated expression data [[22], Table 1 Developmental competence of in vivo developed 2-, 8- GSE33314 and GSE111990]] using the same experimen- and 16-cell stage embryos under in vitro culture condition tal design, expression patterns of selected differentially Group Heifers Total Blastocyst development expressed genes were further quantified using quantita- embryo Day 7 Day 8 Day 9 tive polymerase chain reaction (qPCR) using gene spec- flushed fic primers (Additional file 2: Table S2) designed using (n) (n) (n) (%) (n) (%) (n) (%) a a a primer3 (http://bioinfo.ut.ee/primer3-0.4.0/). For this, 2-cell flush 11 69 15 21.7 42 60.9 57 82.6 the total RNA isolated from the experimental blastocyst b b a 8-cell flush 6 118 63 53.4 93 78.8 97 82.2 groups were reverse transcribed. The qPCR was then c c b 16-cell flush 12 45 36 80.0 42 93.3 45 100.0 performed in 20 μl reaction volume containing iTaq Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 5 of 19 Fig. 2 Global DNA methylation patterns of differentially methylated genomic regions in different blastocyst groups. a Volcano plots displaying the distribution of hypermethylated (positive log fold changes) and hypomethylated (negative log fold changes) genomic loci in each 2 2 blastocysts group. b The total number of differentially methylated genomic regions in each blastocyst group A detailed analysis of DMRs including 160 CpG 2C_Flush, 8C_Flush and 16C_Flush, respectively. Among islands indicated that the DNA methylation patterns of these, 88.5% (n = 1623) of the DMRs including the 1832 genomic loci (containing 29,166 CpG sites) were imprinted genes (PEG3, IGF2R, ASB4, SFMBT2 and differentially methylated in all the three blastocyst GRB10) and DNMT1 were found to be hypermethylated. groups, despite the stage specific exposure to the in vitro The majority of commonly altered genes including the culture (Fig. 3a). These genomic loci represented 17.6, solute carrier clusters, collagens, ADAM metallopepti- 19.6 and 26.7% of the total DMRs identified in the dases, ATPases, zink finger proteins, kinesins were Fig. 3 Commonly differentially methylated genomic regions in all blastocyst groups. a The proportion of commonly detected hypermethylated and hypomethylated genomic loci. b Genomic localization of commonly differentially methylated genomic regions and the list of representative genes in each category Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 6 of 19 differentially methylated in the exonic and/or intronic Network analysis of these genes indicated that PAK4 and gene body regions (Fig. 3b). These differentially methyl- RPS15 genes which were downregulated but hypermethy- ated genes were found to be involved in 16 biological lated in all blastocysts group were found to be the center pathways including the calcium signaling pathway and genes in lipid biosynthesis and nucleosome assembly, re- focal adhesion (Fig. 4a). In addition, PAK4, COL4A1, spectively while TP53BP2 was the center gene of inflamma- COL1A2, EGFR, SMAD3, ITPKA and MYLK were tory response and intracellular transport pathways (Fig. 5b). among the commonly differentially methylated genes that were foundtobeinvolvedintwo or more path- Aberrant DNA methylation patterns induced in both ways (Fig. 4b). blastocysts developed in vitro from embryos flushed After identification of genomic regions commonly differ- before and during embryonic genomic activation entially methylated in all blastocyst groups, we investigated Unlike the 16C_Flush blastocysts group, both the the effects of aberrant DNA methylation patterns on the 2C_Flush and 8C_Flush blastocysts were obtained from blastocyst gene expression by superimposing the DNA embryos that spent several cleavages in vitro including methylation profile with the transcriptome data [[22]], the major embryonic genome activation (EGA). There- GSE33314 and GSE111990]]. The results of this analysis in- fore, here we investigated the DMRs commonly dysregu- dicated that only 2% (n = 28) of the annotated differentially lated only in the 2C_Flush and 8C_Flush groups, but not methylated genes were differentially expressed in all three in the 16C_Flush group. The results showed that groups. Among which, the expression level of 11 genes in- including DNA (cytosine-5)-methyltransferase 3 beta cluding BAHD1, RPS15, PAK4 and PISD were negatively (DNMT3B), the methylation patterns of 2988 genomic correlated with the DNA methylation patterns (Fig. 5a). loci were altered only in these two blastocyst groups of Fig. 4 a List of pathways enriched by commonly differentially methylated genomic regions in all blastocyst groups and b list of representative differentially methylated genomic regions involved in more than one pathway Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 7 of 19 Fig. 5 a The heatmap depicting the methylation and expression patterns of DMRs commonly detected in all blastocyst groups. DM; differentially methylated, DE; differentially expressed. The red color indicates hypermethylation of the DNA methylation or upregulation of the mRNA expression whereas the green color indicates hypomethylation of the DNA methylation or downregulation of the mRNA expression. b The gene network and biological processes enriched by commonly detected DMRs and which showed inverse correlation with their corresponding mRNA expression levels which 72% were hypomethylated in both blastocyst correlated to their corresponding expression levels. groups (Fig. 6a). Genomic localization of these DMRs Among these, the expression level of 61 hypomethylated showed that about 67% of the DMRs are localized in genes including MLH1, SMARCA5, NFYB, ATR, ROCK1 and around the gene body regions (Fig. 6b) and the and ACVR1 was upregulated while the mRNA level of hypermethylated loci were more common in the exonic 11 hypermethylated genes including ADCY5 and IGF2R regions and the hypomethylated loci were common in was downregulated in both 2C_Flush and 8C_Flush the promoter and intronic regions (Fig. 6c). blastocyst groups (Fig. 7a). Furthermore, the gene net- Next, we investigated the function of these genomic work analysis showed that SMARCA5, NFYB and regions by performing gene enrichment analysis and the ACVR1 were the center genes regulating cell maturation results have shown that Pi3-Akt signaling pathway, Ras and ATR, ROCK1 and ADCY5 were found to be the cen- signaling pathway, peroxisome pathway, EGFR tyrosine ter genes of the lipid biosynthesis processes (Fig. 7b). kinase inhibitor resistance and metabolic pathways were among the biological pathways altered by the differen- DNA methylation patterns induced exclusively in tially methylated genes (Fig. 6d). blastocysts developed in vitro from embryos flushed Comparative analysis of the methylation and mRNA before, during or after embryonic genomic activation expression patterns indicated that the methylation levels We considered that in the 2C_Flush blastocysts group, of 72 out of 753 annotated DMRs were inversely both the major and minor EGA occurred under the in Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 8 of 19 Fig. 6 Commonly differentially methylated genomic regions in both the 2C_Flush and 8C_Flush blastocyst groups. a The proportion of hypermethylated and hypomethylated DMRs. b The distribution of DMRs in different genomic regions. c The proportion of hypermethylated and hypomethylated DMRs in different genomic regions. Log FC: the ratio of hypermethylated and hypomethylated genomic loci described in Log 2 2 scale. d Pathways enriched by differentially methylated genomic regions in both the 2C_Flush and 8C_Flush blastocyst groups. DMRs:TGS, the ratio of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in pathway vitro condition, while in the 8C_Flush blastocyst group, EIF4B) were found to be unique to the 16C_Flush minor EGA was completed under in vivo condition blastocysts group. When these genomic loci were whereas the major EGA was partly occurred under in partitioned into hypermethylated and hypomethylated vivo condition. In the 16C_Flush blastocyst group, both categories, the hypermethylated loci exceeded the minor and major EGA was completed under in vivo con- number of hypomethylated genomic loci in the dition. Therefore, we further analyzed the differentially 2C_Flush and 16C_Flush groups. However, the reverse methylated genomic regions that were specific to the wasobservedin the 8C_Flush group(Fig. 8a). 2C_Flush, 8C_Flush or 16C_Flush blastocyst groups. Our findings revealed that a total of 4599 DMRs including kruppel-like factors (KIF1, − 4, − 6, − 7, − 8, − 11, − 16, and Chromosomal distribution and genomic localization of KLF22) were unique to the 2C_Flush blastocysts exclusively differentially methylated loci in blastocysts whereas a total of 3410 DMRs including ankyrin repeat developed in vitro from embryos flushed before, during domain-containing protein (ANKRD6,-16, − 24, − 28, or after embryonic genomic activation − 35, − 39 and ANKRD60) were unique to the To further identify chromosomal regions sensitive to the in 8C_Flush group. Conversely, a total of 2889 DMRs vitro culture condition, we looked into the chromosomal including elongation factors (EIF2AK1, EIF3C and distribution of the DMRs induced exclusively before, during Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 9 of 19 Fig. 7 a The scatter plot displaying expression patterns and the methylation levels of DMRs commonly detected in the 2C_Flush and 8C_Flush groups in blastocysts of the 2C_Flush. The DNA methylation and mRNA expression levels are indicated in the X and Y axis, respectively. b The gene network and biological processes enriched by these DMRs whose expression levels arenegatively correlated with mRNA expression levels or after EGA. To perform this, we first arranged the Uniquely differentially methylated genomic loci were chromosome numbers based on the total probe and DMR subjected to the genomic localization and genomic en- densities (Additional file 3:FigureS1).The frequencyofthe richment analysis. The results indicated that 72.7, 71.4 DMRs in the chromosomes tended to follow the probe and 74.2% of the DMRs in the 2C_Flush, 8C_Flush and density available in the EDMA with minor deviations. 16C_Flush blastocysts, respectively were located in and However, the number of differentially methylated genomic around the gene body regions, whereas the rest were lo- loci at X chromosome was 1.8 and 2.6 times higher in the calized in the intergenic regions (Additional file 5: Figure 2C_Flush compared to the 8C_Flush and 16C_Flush S3). Within the gene body regions, higher proportions of groups, respectively. This has been also evidenced in the unique DMRs were located in the intronic followed by circular circos plot which shows the distribution of signifi- exonic regions whereas the proportion of DMRs on the cantly differentially methylated genomic loci across the plain and proximal promoter was relatively lower in all chromosomes (Additional file 4: Figure S3). In addition, blastocyst groups. However, the hypermethylated gen- chromosomal enrichment analysis (analysis of hyper- omic loci exceeded the hypomethylated ones in the ex- methylated probes with reference to hypomethylated onic regions of all blastocyst groups and the probes) of the DMRs revealed that unlike the 8C_Flush hypermethylated genomic loci surpassed the hypomethy- group, the proportion of hypermethylated loci exceeded lated ones in the plain promoter, proximal promoter, the hypomethylated ones in all chromosomes in the distal promoter and intronic regions of the 2C_Flush 2C_Flush and 16C_Flush groups. However, in the and 16C_Flush blastocyst groups (Fig. 9). 8C_Flush, unlike other groups, four times higher hypo- In order to investigate the biological relevance of the methylated probes than hypermethylated loci were de- DMRs exclusively induced in blastocysts derived from tected on X chromosome (Fig. 8b). embryos subjected to the in vitro culture condition before, Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 10 of 19 Fig. 8 a the number and b chromosomal enrichment of genomic loci differentially methylated exclusively in 2C_Flush, 8C_Flush or 16C_Flush blastocyst group. Chr, Chromosome, Log FC: the ratio of hypermethylated and hypomethylated genomic loci described in Log scale 2 2 during or after major EGA, annotated DMRs were differentially expressed genes and the methylation pat- subjected to pathway analysis. Results showed that differ- terns of 94 DMRs were inversely correlated with their cor- entially methylated genes only in the 2C_Flush were sig- responding mRNA expression patterns (Fig. 11a). Among nificantly enriched in 15 biological pathways including these, the hypomethylated genes including SLC16A1, Hippo signaling pathway. On the other hand, the genes ABHD12 and ACOT4 were upregulated while hyper- differentially methylated only in the 8C_Flush group were methylated genes including SLC16A13, ABAT, ALDH4A1 significantly enriched in steroid biosynthesis, axon guid- and PPP2R2B were downregulated. ance, calcium signaling and focal adhesion pathways Functional classification of these genes showed that (Fig. 10). However, exclusively differentially methylated the DMRs which exhibited opposite patterns to the genes in the 16C_Flush group didn’t display any signifi- mRNA expression levels were found to be significantly cant enrichment in any of the biological pathways. enriched in 7 biological processes (Fig. 11b) including, plasma membrane lactate transport (SLC16A13), ala- The expression patterns of exclusively differentially nine, aspartate and glutamate metabolism (ABAT, methylated genomic loci in blastocysts developed in vitro ALDH4A1) and phosphorus metabolic processes from embryos flushed before, during or after embryonic (GLIPR2, CTNS, ABHD12, PPP2R2B, PROM2, SERINC3 genomic activation and ACOT4). Comparative analysis of DNA methylation and gene ex- In the 8C_Flush blastocyst group, a total of 67 pression showed that a total of 179 exclusively differen- genes including those differentially methylated in pro- tially methylated genes in 2C_Flush were mapped to 167 moter regions (PBK, TXNL4A, UPK1A and ZFYVE16), Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 11 of 19 Fig. 9 Genomic enrichment of the DMRs which are unique to each blastocyst group. Log FC: the ratio of hypermethylated and hypomethylated genomic regions described in Log scale in intronic region (ATP6V0A4, NUDT3 and TNPO3) inversely related to the gene expression patterns and exonic region (NR1H2, TSR1 and EIF6) were in- (Fig. 13). versely correlated with the gene expression (Fig. 12a). Functional annotation of these genes indicated that Quantification of the DNA methylation patterns using biological processes including ribosome biogenesis, bisulfite sequencing protein ubiquitination, vacuolar acidification, regula- The methylation patterns and the expression level of selected tion of fatty acid biosynthetic process, steroid biosyn- candidate differentially methylated imprinted (GRB10, PEG3 thesis and diphosphoinositol polyphosphate catabolic and IGF2R) and non-imprinted genes (ITPK1, COL4A1 and process were found to be altered (Fig. 12b). However, TRAPPC9) were analyzed using bisulfite sequencing (Fig. 14). in the 16C_Flush group, only 15 differentially methyl- Among genes analyzed, GRB10 which was hypermethylated ated genes including PGAP1, and PRDX4 were in 2C_Flush, 8C_Flush and 16C_Flush blastocyst groups Fig. 10 Pathways enriched by the DMRs exclusively detected only in the 2C_Flush (a) and 8C_Flush (b) blastocyst groups. DMRs:TGS, the ration of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 12 of 19 Fig. 11 a The expression patterns of the DMRs exclusively detected in the 2C_Flush blastocysts group. DM; differentially methylated, DE, differentially expressed. b The pathways enriched by DMRs whose DNA methylation patterns were inversely correlated with the mRNA expression levels. DMRs:TGS, the ration of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway compared to the in vivo counterparts, and the bisulfite se- qPCR using the total RNA isolated from independent quencing also confirmed hypermethylation of this gene in samples. The results indicated that of 15 out 18 compar- the 8C_Flush and 16C_Flush blastocyst groups. In addition, isons were found to be in line to the array data (Fig. 15). according to the array result TRAPPC9, COL4A1 and ITPK1 Furthermore, the qPCR data indicated that the expres- were hypermethylated in 2C_Flush, 8C_Flush and the bisul- sion levels of the hypermethylated genes, namely fite sequencing indicated relatively higher methylation of TRAPPC9, COL4A1, PEG3, IGFR2 in 2C_Flush and these genes in those samples relative to the vivo blastocysts. 8C_Flush (Fig. 14) were found to be downregulated but Therefore, results of the bisulfite sequencing were similar to the expression level of GRB10 was upregulated (Fig. 13). the array data indicating the validity of the array analysis. Validation of differentially expressed genes using Discussion quantitative polymerase chain reaction (qPCR) In vitro embryo production is one of one of the most For comparative analysis of the DNA methylation and important innovations and one of the sustainable strat- expression profile, we have selected genes which showed egies to meet the global milk and meat demands [23]. a difference of log fold change ≥0.56, or log fold Moreover, in vitro embryo production has been one of 2 2 change ≤− 0.56 with p value < 0.05 in the previous data- the innovation used to foster farm animal breeding and set. Therefore, the expression patterns of nine candidate to make a significant progress in human and veterinary differentially expressed genes were validated using the medicine [24]. However, several evidences indicated that Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 13 of 19 Fig. 12 a The expression patterns of DMRs exclusively detected only in the 8C_Flush blastocyst group. DM; differentially methylated, DE, differentially expressed. b Pathways enriched by DMRs which were exclusively detected only in the 8C_Flush blastocyst group and whose methylation patterns were inversely correlated with the expression levels. DMRs:TGS, the ratio of differentially methylated genomic regions (DMRs) and the total number of genes (TGs) involved in the pathway Fig. 13 The gene expression patterns of the DMRs unique to 16C_Flush blastocyst group. DM; differentially methylated, DE, differentially expressed Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 14 of 19 Fig. 14 Bisulfite sequencing results of the DMRs identified by the EmbryoGENE DNA Methylation Array in vitro culture condition are not exactly mimicking or in embryonic stages subjected to in vitro culture condi- replacing the in vivo situation. As the result of this, only tion before or during the onset embryonic genome acti- 30–40% of the in vitro produced embryos are developing vation compared to those exposed after the onset of to the blastocysts stage [25]. In some cases, embryos de- embryonic genome activation. Interestingly, the blasto- veloped under in vitro culture condition can be of low cysts from these embryo groups exhibited a higher DNA quality compared to those developed in vivo and these methylation dysregulation. Indeed, a detailed analysis of differences are further manifested by the darker cyto- the DMRs indicated the presence of commonly affected plasm, low cryotolerance and lower pregnancy rate after genomic regions in all or at least in the two blastocyst transfer to recipients [26, 27]. Several evidences have groups irrespective of stage-specific exposure to in vitro demonstrated that these phenotypic differences are crit- culture condition. Accordingly, the DNA methylation ically affected by the post- fertilization embryo culture patterns of several arrays of genomic loci including those condition which can be subsequently manifested by involved in the focal adhesion and metabolic pathways aberrant expression and DNA methylation pattern of were commonly altered in all blastocyst groups suggest- developmentally related genes [11–14, 17]. However, ing that these genomic regions are sensitive to the envir- identification of the embryonic stages which are sensitive onmental insults at any stages of preimplantation to the environmental insults remains a challenge. In- Table 2 List of commonly differentially methylated imprinted deed, establishing an appropriate model which mimics genes and DNMTs the optimal and suboptimal embryo culture condition is Gene symbol 2C_Flush 8C_Flush 16C_Flush Genomic region indispensable to unravel the phenotypic and molecular PEG3 ▲▲▲ Gene body abnormalities induced in embryo subjected to environ- mental challenges. Therefore, embryos developed in vivo GRB10 ▲▲▲ Gene body until 2-, 8- or 16-cell stages were subjected to in vitro IGF2R ▲▲▲ Gene body culture until the blastocyst stage. These embryo groups DNMT3B ▲▲ - Gene body are believed to represent the developmental stages of be- DNMT1 ▼▲▲ Gene body fore, during and after the onset of major embryonic gen- ASB4 ▲▼▼ Gene body ome activation, respectively. In fact, developmental data SFMBT2 ▲▼▼ Gene body indicated that the proportion of embryos developed to the blastocyst stage, 7 days post-insemination was lower ▲ Hypermethylated, ▼ Hypomethylated Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 15 of 19 Fig. 15 The qPCR validation results of the selected differentially expressed genes generated by EmbryoGENE transcriptome microarray embryonic development. For instance, the imprinted during embryonic development was also resulted in over- genes such as PEG3, IGF2R, SNRRP and GRB10 (Table 2) growth of both the embryo and placenta in human [32]. were aberrantly methylated in all blastocysts and inde- In addition to imprinted genes, cluster of genes such as pendent bisulfite sequencing. We also observed differ- the solute carriers and the zinc finger proteins were com- ences in the DNA methylation level of PEG3, IGF2R and monly differentially methylated in all blastocyst groups. GRB10 genes between the blastocyst groups. Culture The solute carriers are believed to be essential for trans- condition induced aberrant methylations of the porting cellular materials across the cell membrane and al- imprinted genes in preimplantation embryos have been tered DNA methylation of the solute carrier genes in the highlighted in many instances. For instance, culture developing embryos could negatively affect the proper func- condition-induced aberrant DNA methylation pattern of tioning of protein transportation across the cell membrane. SNRRP in bovine embryos [28] and PEG3 gene in mouse In addition to these, aberrant DNA methylation of the placenta [29] have been indicated. Moreover, aberrant zinc-finger proteins such as ZNF234, ZNF618, ZNF688 and DNA methylation pattern of PEG3, IGF2R and GRB10 ZNF862 are interesting as the zinc proteins are found to in children conceived by IVF/ICSI [30] also described further regulate the DNA methylation patterns of some the long term effect of suboptimal culture condition on other genes by targeting the M.CviPI (GC methylation) or the DNA methylation pattern of imprinted genes. This, M.SssI (CG methylation) sites [33]. On the other hand, could in turn, reduce embryonic developmental compe- extracellular-matrix related genes including COL4A1 and tence and/or result in fetal abnormalities. For instance, ab- COL1A2 were also differentially methylated in all blastocyst errant methylation of IGF2R could cause large offspring groups of which the COL4A1 gene was hypermethylated in syndrome in sheep [31] and functional loss of GRB10, all blastocyst groups and its expression pattern was Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 16 of 19 downregulated. The DNA methylation and expression pat- beginning of de novo methylation appears to overlap the terns of COL4A1 were further confirmed by bisulfite se- genome activation occurring at the same cell stage. quencing and qPCR, respectively using independent Nevertheless, whether this pattern remains similar while blastocyst samples. Accordingly, COL4A1 gene was embryos are exposed to the suboptimal environmental hypermethylated, but its expression was downregulated in condition is not clear. However, the study in rabbit em- blastocyst obtained from 2- and 8-cell flush embryos com- bryo showed that DNA demethylation in in vitro pro- pared to completely in vivo developed blastocysts. The duced embryos occurs between the 2- and 8-cell stages functional relevance of these gene in bovine embryo de- while in in vivo ones, demethylation occurs between 4- velopment is barely understood, but one study showed and 16-cell stages and remains constant from 16-cell that functional loss of COL4A1 and COL4A2 resulted in until the morula stage suggesting that demethylation embryonic lethality in mouse embryo due to impaired and remethylation of the genome occurs faster in in basement membrane stability [34]. Moreover aberrant vitro embryos than the in vivo ones. Similarly, the study DNA methylation pattern of COL4A1 gene in the gene by Desmet [28] also indicated differences in the DNA body was associated with cancer [35]. methylation changes when early stage embryos are ex- Apart from the DNA marks altered in all blastocyst posed to suboptimal culture suggesting that early-stage groups; we have also analyzed the DNA methylation embryos could be more susceptible to adverse culture marks, which are specific to each of the blastocyst condition. group. Unlike the blastocysts of the 8-cell flush, the blas- In addition to the holistic approaches, we have also tocysts of the 2- and 16-cell flush groups were associated looked into the developmental relevance of uniquely dif- with more hypermethylated genomic loci than the hypo- ferentially methylated individual genes in each blastocyst methylated ones, suggesting that exposure of embryos group. For instance, some of the Krüppel-like transcrip- before or after embryonic genome activation could in- tion factors (KLFs) were differentially methylated in blas- duce DNA methylation, while exposing at the time of tocysts of the 2-cell flush group. The KLFs are believed to genome activation could foster demethylation of the be evolutionarily conserved and potentially involved in resulting blastocysts. Indeed, the interdependence be- several cellular activities such as proliferation, differenti- tween the onset of genome activation and DNA methy- ation and cellular development [46]. Thesefamiliesof lation patterns during mammalian embryogenesis is genes are consisting of three Krüppel-like zinc fingers barely understood. In fact, the genome DNA undergoes which bind to CACCC elements and GC-rich regions of multiples epigenetic modifications immediately after DNA for mediating the gene transcription [47]. Among fertilization which may in turn affect the gene expression the KLFs, KIF4 was among the differentially methylated patterns that could be established during and after em- KLFs in blastocysts of the 2-cell flush. KLF4 is involved in bryonic genome activation. For instance, active and pas- cell proliferation, apoptosis and invasion [48] and believed sive demethylation of the pronuclei occurs during to be negatively regulating the expression of endodermal mouse, bovine and human embryonic development, but markers which may be involved in the transformation of the timing of re-methylation is different depending on pluripotent stem cells to endoderm differentiation [49]. In the species [8, 36, 37]. In mice and humans, massive de addition, KLF4 regulates the NANOG and OCT3/4 tran- novo methylation occurs at the blastocyst stage [8, 38], scription during embryonic stem cell self-renewal and whereas in bovine, the major embryo genome activation subsequently block the embryonic stem cell differentiation is believed to occur between the 8- and 16-cell stage [50, 51]. Therefore, developmental retardation of the [39] although the bovine embryo has been found to be 2-cell flush may partly be induced by aberrant DNA transcriptionally competent at the 2-cell stage [40]. Con- methylation pattern of the KLFs. ceptually, for genes to be expressed, the DNA sequence Overall, in this study we have identified several DMRs must be accessible for the transcriptional machinery and which are sensitive to culture condition at any stage of any blockage of the transcription machinery could po- embryonic development. Hypermethylated or hypomethy- tentially shutdown the gene expression [41]. Among lated genomic loci are generally believed to play a role in these, hypomethylation (removal of the DNA methyla- regulating cellular development and function by inhibiting tion marks) around the promoter region is believed to or inducing the gene expression. Therefore, in this study, increase the transcriptional activity of the genes [42–44]. apart from exploring the molecular pathways enriched by During normal embryo development, the fertilized egg the DMRs, we also investigated the effects of those DMRs undergoes multiple epigenetic modifications, which sub- on the gene expression patterns. Accordingly, we have de- sequently affect the resulting gene expression profiles. In tected about 2–9% of annotated DMRs of in the promoter mice, while the paternal pronucleus undergoes active de- or gene body regions were inversely correlated with the methylation, the maternal genome passively demethy- corresponding gene expression (Figs. 5, 7, 11, 12 & 13). lated after the 2-cell stage [45]. However, in bovine, the Interestingly, these genes were found to be involved in key Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 17 of 19 pathways that are believed to be essential for embryogen- Additional file 4: Figure S2. The circos plot representing the overall esis such as fatty acid biosynthetic process, steroid biosyn- methylation levels in the in different blastocyst groups. Positive and negative fold-changes represent the level of hypermethylation and hypomethylation. thesis, energy metabolism and JAK-STAT cascade (PDF 19412 kb) pathways. Indeed, although the proportion of DMRs that Additional file 5: Figure S3. Genomic distribution of exclusively inversely correlated with the gene expression seems to be differentially methylated regions in the 2C_Flush, 8C_Flush or 16C_Flush lower while considering the total number of genomic loci blastocyst group. (PDF 47 kb) differentially methylated, it should be noted that there were several DMRs whose expression was positively corre- Abbreviations lated with gene expression (data not shown). On the other 16C_Flush: blastocysts developed in vitro from embryos flushed at 16-cell stage; 2C_Flush: blastocysts developed in vitro from embryos flushed at 2- hand, the DMRs whose mRNA was not changed does not cell stage; 8C_Flush: blastocysts developed in vitro from embryos flushed at necessarily imply that their methylation pattern is not 8-cell stage; CpG: Cytosine phosphate guanine; DE: Differentially expressed; essential for mRNA expression, rather these DMRs DM: Differentially methylated; DMRs: Differentially methylated genomic regions; EDMA: Embryogene DNA methylation array; EGA: Embryonic could be involved in regulating the embryonic devel- genome activation; EGFR: Epidermal growth factor receptor; FDR: False opment by regulating the downstream genes or by discovery rate; ICSI: Intracytoplasmic sperm injection; IVF: In vitro fertilization; fine-tuning the expression pattern of other gene regu- qPCR: Quantitative real time polymerase chain reaction latory agents such as miRNAs. Funding This study was funded by EmbryoGENE. The funding body has no role in the Conclusion designing the experiment, collection of data, analysis interpretation of data, This study identified culture-induced stage-specific and writing the manuscript. non-stage-specific aberrant DNA methylation patterns Availability of data and materials of paternally or maternally imprinted genes and several All data used and/or analyzed during the present study are available from arrays of genomic regions in in vivo originated bovine the corresponding author if requested. embryos subjected to the in vitro culture in stage spe- cific manner. Thus, embryos exposed to in vitro culture Authors’ contributions condition before the onset or during embryonic genome DSW has performed experiment, performed sample processing and array hybridization, data analysis, interpreting the results, drafting and writing the activation are more sensitive to the DNA methylation manuscript and serve as the corresponding author. MH was involved in marks changes that could be further displayed in the endoscopic transfer & flushing, design of the experiment. ET involved in resulting blastocysts compared to those exposed to in designing the experiment and revised manuscript. MSZ performed qPCR analysis and critically read the manuscript. EH collected the samples and vitro culture condition after embryonic genome activa- critically revised the manuscript. SG, MP and HOP and CN critically revised the tion. Thus, while exposure of in vivo embryos to in vitro manuscript. FR did in vitro culture embryo and sample processing. UB & VH culture condition before embryonic genome activation performed endoscopic flushing of embryos and designed the experiment. EF performed statistical analysis of differentially methylated regions and favored the initiation of DNA methylation, subjecting of interpreted the data. DG performed array hybridization and involved in in vivo embryos to the in vitro culture at the time of interpretation of the results. AG did transcriptome profile analysis. MA and CR genome activation increased the hypomethylated gen- involved in experimental design, supervised during array hybridization and revised the manuscript. KS designed the experiment, supervised the omic loci in the resulting blastocysts. Nevertheless, a de- experiments and reviewed the manuscript. DT was responsible for designing tailed study is required to understand whether these the experiment, interpretation of the results and reviewing the manuscript. All altered DNA methylation patterns induced in the em- authors approved the final version of the paper. bryo by suboptimal culture stimuli are initiated as means Ethics approval of coping mechanisms of the embryo or an indicator of The experiment was approved by the Animal Welfare committee of the abnormal embryo development. Moreover, future studies University of Bonn with proposition number 84–02.05.20.12.075. could further investigate whether or not the suboptimal culture-induced epigenetic modifications occurred dur- Competing interests ing bovine preimplantation embryo at the blastocysts The authors declare that they have no competing interests. stage could have a long-lasting consequence on the off- spring epigenome. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional files Author details Additional file 1: Table S1. The list of primers used to validate DMRs in Institute of Animal Science, Animal Breeding and Husbandry Group, blastocysts of different groups using bisulfite sequencing. (DOCX 15 kb) University of Bonn, 53115 Bonn, Germany. Institute of Animal Breeding and Genetics, University of Veterinary Medicine Vienna, A-1210 Vienna, Austria. Additional file 2: Table S2. The list of primers used for validation of the Centre de recherche en biologie de la reproduction, Faculté des sciences differentially expressed genes using qPCR. (DOC 33 kb) de l’agriculture et de l’alimentation, INAF, Pavillon des services, Université Additional file 3: Figure S1. The order of chromosomes based their Laval, Québec G1V 0A6, Canada. Department of Animal Production, Faculty probe density and differentially methylated probes (TIF 4242 kb) of Agriculture, Cairo University, Giza 12613, Egypt. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 18 of 19 Received: 18 September 2017 Accepted: 25 May 2018 21. Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004;3:Article3. 22. Gad A, Hoelker M, Besenfelder U, Havlicek V, Cinar U, Rings F, et al. Molecular mechanisms and pathways involved in bovine embryonic References genome activation and their regulation by alternative in vivo and in vitro 1. Bedzhov I, Leung CY, Bialecka M, Zernicka-Goetz M. In vitro culture of culture conditions. Biol Reprod. 2012;87:100. mouse blastocysts beyond the implantation stages. Nat Protoc. 2014;9: 23. Blondin P. Status of embryo production in the world. Animal Reprod. 2015; 2732–9. 12:356–8. 2. Feuer S, Rinaudo P. Preimplantation stress and development. Birth defects 24. Duszewska AM, Trzeciak P, Compa A, Rąpała Ł. Selected issues concerning Res C Embryo Today. 2012;96:299–314. biotechnology of farm animals breeding - a review. Anim Sci Pap Rep. 2010; 3. Maillo V, Lopera-Vasquez R, Hamdi M, Gutierrez-Adan A, Lonergan P, Rizos 28:295–306. D. Maternal-embryo interaction in the bovine oviduct: evidence from in 25. Camargo LSDA, Viana JHM, Sá WFD, Ferreira ADM, Ramos ADA. Factors vivo and in vitro studies. Theriogenology. 2016;86:443–50. influencing in vitro embryo production. Animal Reprod. 2006;3:19–28. 4. Zheng LL, Tan XW, Cui XZ, Yuan HJ, Li H, Jiao GZ, et al. Preimplantation 26. Rizos D, Ward F, Boland MP, Lonergan P. Effect of culture system on the maternal stress impairs embryo development by inducing oviductal apoptosis yield and quality of bovine blastocysts as assessed by survival after with activation of the Fas system. Mol Hum Reprod. 2016;22:778–90. vitrification. Theriogenology. 2001;56:1–16. 5. Duc-Goiran P, Mignot TM, Bourgeois C, Ferre F. Embryo-maternal 27. Lonergan P, Rizos D, Gutierrez-Adan A, Fair T, Boland MP. Effect of culture interactions at the implantation site: a delicate equilibrium. Eur J Obstet environment on embryo quality and gene expression - experience from Gynecol Reprod Biol. 1999;83:85–100. animal studies. Reprod BioMed Online. 2003;7:657–63. 6. Herrler A, von Rango U, Beier HM. Embryo-maternal signalling: how the 28. Desmet K, van Hoeck V, Gagne D, Fournier E, Thakur A, O'Doherty AM, et al. embryo starts talking to its mother to accomplish implantation. Reprod Exposure of bovine oocytes and embryos to elevated non-esterified fatty BioMed Online. 2003;6:244–56. acid concentrations: integration of epigenetic and transcriptomic signatures 7. Lonergan P, Forde N. Maternal-embryo interaction leading up to the initiation in resultant blastocysts. BMC Genomics. 2016;17:1004. of implantation of pregnancy in cattle. Animal. 2014;8(Suppl 1):64–9. 29. de Waal E, Mak W, Calhoun S, Stein P, Ord T, Krapp C, et al. In vitro culture 8. Dean W, Santos F, Stojkovic M, Zakhartchenko V, Walter J, Wolf E, Reik W. increases the frequency of stochastic epigenetic errors at imprinted genes Conservation of methylation reprogramming in mammalian development. in placental tissues from mouse concepti produced through assisted Aberrant reprogramming in cloned embryos. PNAS. 2001;98:13734–8. reproductive technologies. Biol Reprod. 2014;90:22. 9. Santos F, Hyslop L, Stojkovic P, Leary C, Murdoch A, Reik W, Stojkovic M, 30. Lazaraviciute G, Kauser M, Bhattacharya S, Haggarty P, Bhattacharya S. A Herbert M, Dean W. Evaluation of epigenetic marks in human embryos systematic review and meta-analysis of DNA methylation levels and derived from IVF and ICSI. Hum Reprod. 2010;25:2387–95. imprinting disorders in children conceived by IVF/ICSI compared with 10. Calle A, Fernandez-Gonzalez R, Ramos-Ibeas P, Laguna-Barraza R, Perez- children conceived spontaneously. Hum Reprod Update. 2015;21:555–7. Cerezales S, Bermejo-Alvarez P, et al. Long-term and transgenerational effects 31. Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG, Carolan C, of in vitro culture on mouse embryos. Theriogenology. 2012;77:785–93. et al. Epigenetic change in IGF2R is associated with fetal overgrowth after 11. Lou H, Le F, Zheng Y, Li L, Wang L, Wang N, et al. Assisted reproductive sheep embryo culture. Nat Genet. 2001;27:153–4. technologies impair the expression and methylation of insulin-induced 32. Charalambous M, Smith FM, Bennett WR, Crew TE, Mackenzie F, Ward A. gene 1 and sterol regulatory element-binding factor 1 in the fetus and Disruption of the imprinted Grb10 gene leads to disproportionate placenta. Fertil Steril. 2014;101:974–80. overgrowth by an Igf2-independent mechanism. Proc Natl Acad Sci. 2003; 12. Gomes MV, Huber J, Ferriani RA, Amaral Neto AM, Ramos ES. Abnormal 100:8292–7. methylation at the KvDMR1 imprinting control region in clinically normal 33. Carvin CD, Parr RD, Kladde MP. Site-selective in vivo targeting of children conceived by assisted reproductive technologies. Mol Hum Reprod. cytosine-5 DNA methylation by zinc-finger proteins. Nucleic Acids Res. 2009;15:471–7. 2003;31:6493–501. 13. Ghosh J, Coutifaris C, Sapienza C, Mainigi M. Global DNA methylation levels 34. Pöschl E, Schlötzer-Schrehardt U, Brachvogel B, Saito K, Ninomiya Y, Mayer are altered by modifiable clinical manipulations in assisted reproductive U. Collagen IV is essential for basement membrane stability but dispensable technologies. Clin Epigenetics. 2017;9:14. for initiation of its assembly during early development. Development. 2004; 14. Dean W, Bowden L, Aitchison A, Klose J, Moore T, Meneses JJ, et al. Altered 131:1619–28. imprinted gene methylation and expression in completely ES cell-derived 35. Maekawa R, Sato S, Yamagata Y, Asada H, Tamura I, Lee L, et al. Genome-wide mouse fetuses: association with aberrant phenotypes. Development. 1998; DNA methylation analysis reveals a potential mechanism for the pathogenesis 125:2273–82. and development of uterine leiomyomas. PLoS One. 2013;8:e66632. 15. Deshmukh RS, Østrup O, Østrup E, Vejlsted M, Niemann H, Lucas-Hahn A, et 36. Mayer W, Niveleau A, Walter J, Fundele R, Haaf T. Demethylation of the al. DNA methylation in porcine preimplantation embryos developed in vivo zygotic paternal genome. Nature. 2000;403:501–2. and produced by in vitro fertilization, parthenogenetic activation and 37. Beaujean N, Hartshorne G, Cavilla J, Taylor J, Gardner J, Wilmut I, Meehan R, somatic cell nuclear transfer. Epigenetics. 2011;6:177–87. Young L. Non-conservation of mammalian preimplantation methylation 16. Hiendleder S, Wirtz M, Mund C, Klempt M, Reichenbach HD, Stojkovic M, et dynamics. Curr Biol. 2004;14:R266–7. al. Tissue-specific effects of in vitro fertilization procedures on genomic cytosine methylation levels in overgrown and normal sized bovine fetuses. 38. Guo F, Li X, Liang D, Li T, Zhu P, Guo H, Wu X, Wen L, Gu T-P, Hu B, Walsh Biol Reprod. 2006;75:17–23. CP, Li J, Tang F, Xu G-L. Active and passive demethylation of male and 17. Salilew-Wondim D, Fournier E, Hoelker M, Saeed-Zidane M, Tholen E, Looft female pronuclear DNA in the mammalian zygote. Cell Stem Cell. 2014;15:447–59. C, et al. Genome-wide DNA methylation patterns of bovine blastocysts 39. Barnes FL, First NL. Embryonic transcription in in vitro cultured bovine developed in vivo from embryos completed different stages of embryos. Mol Reprod Dev. 1991;29:117–23. development in vitro. PLoS One. 2015;10:e0140467. 40. Memili E, Dominko T, First NL. Onset of transcription in bovine oocytes and preimplantation embryos. Mol Reprod Dev. 1998;51:36–41. 18. Gad A, Besenfelder U, Rings F, Ghanem N, Salilew-Wondim D, Hossain MM, 41. Frith MC, Valen E, Krogh A, Hayashizaki Y, Carninci P, Sandelin A. A code for et al. Effect of reproductive tract environment following controlled ovarian transcription initiation in mammalian genomes. Genome Res. 2008;18:1–12. hyperstimulation treatment on embryo development and global transcriptome profile of blastocysts: implications for animal breeding and 42. Sizemore ST, Sizemore GM, Booth CN, Thompson CL, Silverman P, Bebek G, human assisted reproduction. Hum Reprod. 2011;26:1693–707. Abdul-Karim FW, Avril S, Keri RA. Hypomethylation of the MMP7 promoter 19. Besenfelder U, Brem G. Tubal transfer of bovine embryos: a simple and increased expression of MMP7 distinguishes the basal-like breast cancer endoscopic method reducing long-term exposure of in vitro produced subtype from other triple-negative tumors. Breast Cancer Res Treat. 2014; embryos. Theriogenology. 1998;50:739–45. 146:25–40. 20. Shojaei Saadi HA, O'Doherty AM, Gagne D, Fournier E, Grant JR, Sirard M-A, 43. Haney SL, Upchurch GM, Opavska J, Klinkebiel D, Hlady RA, Suresh A, et al. Promoter Robert C. An integrated platform for bovine DNA methylome analysis Hypomethylation and expression is conserved in mouse chronic lymphocytic suitable for small samples. BMC Genomics. 2014;15:451. leukemia induced by decreased or inactivated Dnmt3a. Cell Rep. 2016;15:1190–201. Salilew-Wondim et al. BMC Genomics (2018) 19:424 Page 19 of 19 44. Kobayashi T, Ishida K, Yoshie H. Increased expression of interleukin-6 (IL-6) gene transcript in relation to IL-6 promoter hypomethylation in gingival tissue from patients with chronic periodontitis. Arch Oral Biol. 2016;69:89–94. 45. Haaf T. Methylation dynamics in the early mammalian embryo: implications of genome reprogramming defects for development. Curr Top Microbiol Immunol. 2006;310:13–22. 46. McConnell BB, Yang VW. Mammalian Kruppel-like factors in health and diseases. Physiol Rev. 2010;90:1337–81. 47. Pearson R, Fleetwood J, Eaton S, Crossley M, Bao S. Kruppel-like transcription factors: a functional family. Int J Biochem Cell Biol. 2008;40:1996–2001. 48. Yang Y, Goldstein BG, Chao H-H, Katz JP. KLF4 and KLF5 regulate proliferation, apoptosis and invasion in esophageal cancer cells. Cancer Biol Ther. 2005;4:1216–21. 49. Aksoy I, Giudice V, Delahaye E, Wianny F, Aubry M, Mure M, et al. Klf4 and Klf5 differentially inhibit mesoderm and endoderm differentiation in embryonic stem cells. Nat Commun. 2014;5:3719. 50. Parisi S, Passaro F, Aloia L, Manabe I, Nagai R, Pastore L, Russo T. Klf5 is involved in self-renewal of mouse embryonic stem cells. J Cell Sci. 2008;121: 2629–34. 51. Zhang P, Andrianakos R, Yang Y, Liu C, Lu W. Kruppel-like factor 4 (Klf4) prevents embryonic stem (ES) cell differentiation by regulating Nanog gene expression. J Biol Chem. 2010;285:9180–9.

Journal

BMC GenomicsSpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off