Studies of nuclear transplantation of diploid imprint
Studies of nuclear transplantation of diploid imprint-free PGCs have shown that the reconstituted oocytes developed to day 9.5 of gestation, with severely retarded embryos and abnormal placentae (Kato et al., 1999). We also reported that MII oocyte injection of imprint-free phESCs without imprinted region deletion could not develop beyond E13.5 (Li et al., 2016). These results demonstrate that the PGC-like state alone is not sufficient to cross the uniparental development barrier without proper modifications.
The molecular mechanisms controlling imprinting at different regions are quite complex and have not been fully elucidated. For instance, activation of Rasgrf1 gene expression has been suggested to be regulated by the change of CTCF-dependent bazedoxifene interaction within the locus through methylation of the imprinted region (Yoon et al., 2005), similar to another CTCF regulation-dependent imprinting gene, H19 (Hark et al., 2000, Kurukuti et al., 2006). But this hypothesis for Rasgrf1 imprinting lacks functional evidence. In this study we showed that Rasgrf1 expression could be activated by Rasgrf1 imprinted region deletion in 3KO-bimaternal mice. This is in vivo evidence that supports the CTCF-dependent enhancer-blocking hypothesis of Rasgrf1 (Yoon et al., 2005). Interestingly, the recovery of Th and Xlr3b in 3KO-bimaternal mice also implied a possible regulating role of Rasgrf1 in the brain. However, several imprinted regions have been reported to function through CTCF-dependent enhancer blocking to date (Phillips and Corces, 2009). On the other hand, new mechanisms, such as DNA methylation-independent placenta imprinting, are also emerging (Inoue et al., 2017).
Phenotype analyses of uniparental mice revealed that two paternal genomes increased (Figures 5D–5F) whereas two maternal genomes decreased organ and body size (Figures S2Q and S2S), which supports the conflict theory, a hypothesis that assumes a resource acquisition competition relationship between mother and offspring in which the paternal inherited genome promoted but the maternal inherited genome hindered the growth of offspring (Wilkins and Haig, 2003). In addition, the distinct results of co-injecting ahESC with sperm (Figures S4A and S4B) and tetraploid complementation of 6KO-adESCs (Figures 4A and 4B) imply a crucial role of H3K27me3-dependent placenta imprinting in embryo development (Inoue et al., 2017, Lewis et al., 2004).
Global demethylation was previously found in mouse diploid ESCs under serum and leukemia inhibitory factor (LIF) (Zvetkova et al., 2005) or 2i and LIF culture (Choi et al., 2017, Ficz et al., 2013, Habibi et al., 2013, Yagi et al., 2017). We also confirmed DNA hypomethylation in diploid ESCs (Figures 1B, 1C, S1A, and S1B). Thus, DNA hypomethylation and the epigenomic similarity to PGCs is not a haploid-specific phenomenon. Notably, we found that both ahESCs and phESCs experienced demethylating processes with different dynamics. Although ahESCs still maintained global methylation and imprints at passage 20, phESCs exhibited hypomethylation and loss of imprints (Figures 1B and 1D). The bisulfite sequencing results also confirmed different loss-of-imprint speeds in phESCs and ahESCs (Figures 1E, S1D, and S1E). These findings suggested allele-related factors affecting ESC demethylation, which remains to be explored. Taking into consideration the successful generation of haploid ESCs in multiple species (Leeb and Wutz, 2011, Li et al., 2012, Li et al., 2014, Sagi et al., 2016, Yang et al., 2012, Yang et al., 2013), the results of the present study may also reveal many aspects of mammal reproduction.
Acknowledgments This study was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA16000000); the National Basic Research Program of China (2014CB964800); the National Key Research and Development Program (2017YFA0103803); the National High Technology R&D Program (2015AA020307); the National Natural Science Foundation of China (31621004, 31422038, and 31471395); the Key Research Projects of the Frontier Science of the Chinese Academy of Sciences (QYZDY-SSW-SMC002 and QYZDB-SSW-SMC022); the National Postdoctoral Program for Innovative Talents (BX201600161 and BX201700243); and the Ferring Institute of Reproductive Medicine, a strategic collaborative research program of Ferring Pharmaceuticals and Chinese Academy of Sciences (FIRMD180304). We thank Shi-Wen Li, Xi-Li Zhu, and Qing Meng for help with fluorescence-activated cell sorting and confocal laser-scanning microscopy.