Introduction
The idea of ‘epigenesis and development’ coined by Waddington as early as 1957 (Patra et al., Reference Patra, Patra, Rizzi, Ghosh and Bettuzzi2008) and which made the connection of epigenetics and development for the first time, suggested the important role of epigenetics in development. As for gametogenesis, this process involves overall demethylation followed by de novo methylation from primordial germ cells to mature gametes (Hajkova et al., Reference Hajkova, Erhardt, Lane, Haaf, El-Maarri, Reik, Walter and Surani2002; Swales & Spears, Reference Swales and Spears2005). Furthermore, it involves a complicated and delicate gene regulation network, in which stem cell marker genes exert important roles (Chambers et al., Reference Chambers, Silva, Colby, Nichols, Nijmeijer, Robertson, Vrana, Jones, Grotewold and Smith2007; Geijsen & Jones, Reference Geijsen and Jones2008; Kristensen et al., Reference Kristensen, Nielsen, Skakkebaek, Graem, Jacobsen, Meyts and Leffers2008; Pesce et al., Reference Pesce, Wang, Wolgemuth and Scholer1998). Given that DNA methylation of the pluripotency-associated genes harbouring CpG islands in mature gametes could have an influence on embryogenesis, it is necessary to investigate their DNA methylation profiles in mature gametes. However, to date, there are limited data available on these genes in mature gametes of bovine. Here we chose five pluripotency-associated genes, namely Oct4, Sox2, Nanog, Rex1 as well as Fgf4, as representatives and showed their methylation profiles in mature oocytes and sperm of bovine, based on the reasoning that the five genes harbour CpG islands in their own 5′ terminal regions, which are frequently the targets of DNA methylation.
Materials and Methods
In vitro maturation of oocytes
Holstein cow ovaries were collected from a local abattoir. The procedure of in vitro maturation of oocytes was carried out according to established methods in our laboratory (Hua et al., Reference Hua, Zhang, Song, Song, Zhang, Zhang, Zhang, Cao and Ma2008). All the collected oocytes used for future work were of high quality in morphology and obtained from the same production line to eliminate the potential interference factors.
Collection of sperm
Bovine frozen–thawed semen was purchased from Keyuan Co., Ltd, Yangling, China. The preparation of sperm followed the previous method described (Wrenzycki et al., Reference Wrenzycki, Herrmann, Keskintepe, Martins, Sirisathien, Brackett and Niemann2001). Capacitated sperm were used for DNA extraction.
Sodium bisulfite genomic sequencing
Extraction of genomic DNA from oocytes and sodium bisulfite treatment was combined using the EZ DNA Methylation-Direct™ Kit (Zymo Research). Genomic DNA from sperm was extracted with the E.Z.N.A.R Forensic DNA Kit (Omega), followed by quantitation using a NanoDrop™ ND-1000 spectrophotometer (Thermo Finnigan), and finally by sodium bisulfite treatment with the EZ DNA Methylation-Gold™ Kit (Zymo Research). All the procedures above were carried out following the manufacture's instructions strictly. The amplification of bisulfite-modified DNA was performed with ZymoTaq™ DNA polymerase (Zymo Research) in a reaction volume of 50 μl. Cycling conditions were 95 °C for 10 min followed by 40 cycles of 94 °C for 30 s, at an annealing temperature (Tm) for 40 s, then 72 °C for 30 s and a final extension of 7 min at 72 °C. The primer sets for the five genes were designed according to online software (http://www.urogene.org/methprimer/) with the exception of the Oct4 primers described previously (Lin et al., Reference Lin, Li, Zhang, Zhao, Dai and Li2008). Details of primer sequences and Tm values are listed in Table 1. The locations of the CpG islands are indicated in Fig. 1. Given the sampling bias of PCR, three independent PCR reactions were performed. Next, the PCR products were purified using the Zymoclean™ Gel DNA Recovery Kit (Zymo Research). Then, PCR products of three reactions were mixed together and cloned into a pMD18-T vector (TaKaRa), followed by verification using PCR. Finally, 10 colonies for each sample were sequenced.
Table 1 Primer sequences
Figure 1 The locations of the CpG islands and amplified regions in the neighbourhood of transcription start sites (TSS) of five pluripotency-related genes (Oct4, Sox2, Nanog, Rex1 and Fgf4). The black line represents the sequence surrounding TSS and spanning from –1 kb to +1 kb. The black box and the grey box indicate the CpG islands and the amplified regions for bisulfite analysis, respectively. The amplified regions of Oct4, Sox2, Nanog, Rex1 and Fgf4 are +399 to +691, −22 to −382, +450 to +737, +560 to +858 and −62 to +161, respectively.
Statistic analysis
DNA methylation levels of five genes were calculated in sperm and oocytes by BiQ-Analyzer software. Significant differences were determined using the chi-squared test with statistical significance being accepted at p < 0.05.
Results and Discussion
Methylation profiles of five pluripotency-related genes
DNA methylation levels of Oct4, Sox2, Nanog, Rex1 and Fgf4 are shown in Fig. 2. Results showed Oct4 and Fgf4 exhibited significant hypermethylation in sperm compared with that in oocytes (p < 0.01) (Fig. 3), while Sox2 and Nanog displayed relatively similar methylation levels between sperm and oocytes (p > 0.05) (Fig. 3). Additionally, Rex1 showed a relatively high methylation level in sperm compared with oocytes, although no significant differences were found (p > 0.05) (Fig. 3).
Figure 2 Methylation profiles of five pluripotency-related genes in mature gametes. (A), (B), (C), (D) and (E) respectively represent Oct4, Sox2, Nanog, Rex1 and Fgf4. They respectively harbour 23, 24, 13, 22 and 11 CpG sites in the amplified CpG islands. Methylation levels are labelled below right. Each line and circle represents a sequencing result and a CpG site, respectively. Open and closed circles indicate unmethylated and methylated CpGs, respectively.
Figure 3 The diagram shows differences in DNA methylation levels of the five pluripotency-related genes between sperm and oocytes. **Extremely significant differences (p < 0.01).
With the completion of de novo methylation, the mature gametes harbour maximum overall DNA methylation level (Reik et al., Reference Reik, Santos and Dean2003). However, our results showed hypermethylation of Oct4, Rex1 as well as Fgf4 and hypomethylation of Sox2 and Nanog in sperm. According to a more recent report, 4% of the haploid genome in sperm is histone bound rather than protamine bound, with very significant overlap with hypomethylation regions and promoters for some developmental transcription factors (Hammoud et al., Reference Hammoud, Nix, Zhang, Purwar, Carrell and Cairns2009), suggesting that Sox2 and Nanog may be within the histone-enriched regions, while Oct4, Rex1 and Fgf4 may be within the protamine-enriched ones. Expectedly, the methylation levels of Oct4 and Sox2 in sperm of bovine were similar to that of human. However, surprisingly, the methylation level of Nanog showed the opposite result in sperm of bovine compared with that of human, which may be associated with differences between species. On the other hand, Oct4 and Rex1 exhibited the average methylation status (49.6%) and hypermethylation, respectively, while Sox2, Nanog and Fgf4 displayed hypomethylation in oocytes, which may be due to the regulation of chromatin configuration to a certain extent. This aspect remains to be investigated. The above results suggested that it seems to be necessary for the given genes to maintain hypomethylation in mature gametes regardless of the global DNA methylation level, which may be beneficial for early embryo development and is consistent with the report on mouse that site-specific demethylation events occurred in mature germ cells (Kafri et al., Reference Kafri, Ariel, Brandeis, Shemer, Urven, McCarrey, Cedar and Razin1992). Taken together, these results generally showed two methylation patterns, namely consistency and opposition, for the five genes we analysed in mature gametes, with the former being non-sex-specific and the latter sex-specific.
Through comparison between sperm and oocytes, it was observed that the methylation level of Oct4 in oocytes significantly fails to reach the same as that in sperm (p < 0.01), although the average methylation status (49.6%) and hypermethylation existed in oocytes and sperm, respectively. As for Rex1, this gene also showed a relatively high methylation level in sperm than in oocytes, although no significant differences were found (p > 0.05). This finding may be explained by incomplete maturation of oocytes. However, the more reasonable explanation for the differences of methylation levels of Oct4 and Rex1 between sperm and oocytes is that the distinctions may reflect the time delay of demethylation. This idea is based on the fact that DNA in sperm and oocytes respectively is surrounded by protamines and histones. Differences in methylation may also affects the transition of protamines to histones in the male pronucleus after fertilization (Rousseaux et al., Reference Rousseaux, Reynoird, Escoffier, Thevenon, Caron and Khochbin2008), providing a relatively relaxed environment – namely from highly packaged to less packaged status, during which time demethylation of the paternal genome takes place prior to that of maternal genome and that both genomes reach the same minimum methylation level at the 8-cell or morula stage, depending on the species (Dean et al., Reference Dean, Santos and Reik2003). As for Sox2, this gene remained unmethylated either in oocytes or in sperm, which coincided with the situations in fetal fibroblasts and in in vitro fertilized 8-cell embryos (data not shown), seeming that this analysed CpG island is not associated with the differentiation status; gametes, fetal fibroblasts and 8-cell embryos respectively represent three distinct differentiation situations. Additionally, Nanog possessed the similar low methylation levels not only in oocytes and sperm but also in IVF 8-cell embryos (data not shown), indicating that it may escape drastic demethylation action both in maternal and in paternal genomes from fertilization to the 8-cell stage (Dean et al., Reference Dean, Santos and Reik2003). Finally, in contrast to the genes mentioned above, Fgf4 exhibited another pattern, with hypomethylation in oocytes and hypermethylation in sperm, which is similar to the pattern of paternal imprinted genes, indicating that only the copy from paternal genome requires demethylation after fertilization. However, it should be taken into account that the present study was made only on in vitro-derived gametes and it is well known that in vitro conditions have profound effects on the epigenetic make-up of gametes. Therefore, results of this study would have to be confirmed in the future by analysis of in vivo-derived gametes.
Given that the methylation levels of gametes could have an effect on embryogenesis (Benchaib et al., Reference Benchaib, Braun, Ressnikof, Lornage, Durand, Niveleau and Guerin2005; Rousseaux et al., Reference Rousseaux, Reynoird, Escoffier, Thevenon, Caron and Khochbin2008; Hammoud et al., Reference Hammoud, Nix, Zhang, Purwar, Carrell and Cairns2009), the contribution of the present study could serve as a theoretical basis for future work on bovine embryo development in assisted reproductive technologies, and provide a reference for methylation levels of donor cells used for nuclear transfer.
Acknowledgements
This work is supported by the Key Scientific and Technological Special Program for the Culture of Disease-resistance Transgenic Cattle Species (2008ZX08007–004), Government of China.
