Available online at www.sciencedirect.com
ScienceDirect Zygotic genome activation and imprinting: parent-of-origin gene regulation in plant embryogenesis Marcelina Garcı´a-Aguilar and C Stewart Gillmor Parent-of-origin dependent gene expression refers to differential activity of alleles inherited from the egg and sperm. In plants, zygotic genome activation (ZGA) and gene imprinting are two examples of this phenomenon, both of which occur during seed development. As its name implies, ZGA is a genome-wide process that occurs in embryos during the first few days after fertilization. Evidence exists that maternal alleles initially predominate during ZGA, although most genes also show some paternal activity. By contrast, imprinting can be defined as a bias in gene expression that lasts beyond the first few days of seed development. Hundreds of imprinted genes have been discovered in the endosperm, and a few have been described in the embryo. This review discusses recent advances in our understanding of the phenomena and mechanisms of ZGA and imprinting in seeds, with an emphasis on embryo development. Important unanswered questions and areas for future research are highlighted. Address Laboratorio Nacional de Geno´mica para la Biodiversidad (Langebio), Unidad de Geno´mica Avanzada, Centro de Investigacio´n y de Estudios Avanzados (CINVESTAV), Irapuato, Guanajuato 36821, Me´xico Corresponding author: Gillmor, C Stewart ([email protected])
Current Opinion in Plant Biology 2015, 27:29–35 This review comes from a themed issue on Cell signalling and gene regulation Edited by Xiaofeng Cao and Blake C Meyers
http://dx.doi.org/10.1016/j.pbi.2015.05.020 1369-5266/# 2015 Elsevier Ltd. All rights reserved.
Introduction Seed development initiates with a double fertilization of the haploid egg and diploid central cell, producing the diploid embryo and triploid endosperm. In the embryo, zygotic genome activation (ZGA) occurs following a period of transcriptional quiescence after fertilization (reviewed in [1,2]). With few exceptions, ZGA in animals is thought to occur in a bi-allelic manner [3,4], though most studies of animal ZGA have used methods that would not be able to detect parent-of-origin effects on zygotic transcription (reviewed in [5], see [6] for an exception). In contrast, studies on ZGA in plants have www.sciencedirect.com
relied on expression of reporter genes and allele-specific RNA-seq profiling [7,8,9,10,11], as well as genetic tests for allele function [7,12–14,15]. These experiments have uncovered many cases of parent-of-origin effects on gene expression in early plant embryogenesis [7,9,10,11,15], along with examples of equal parental contributions [8,11,12–14]. In addition, one maternal effect and one paternal effect gene are known to be transcribed in gametes, yet act in the embryo [16,17]. Imprinting causes the expression of a gene to be dependent on its parent of origin, due to differential epigenetic modifications established during male and female gametogenesis (reviewed in [18]). According to expression bias, imprinted genes are classified as maternally expressed genes (MEGs) or paternally expressed genes (PEGs). Genome scale studies of imprinted genes in Arabidopsis, maize, rice and other plants have identified over 500 MEGs and about 100 PEGs in the endosperm [19–25,26,27,28]. Most of these genes show transcriptional bias in favor of one allele or the other, as opposed to uniparental expression. At least two mechanisms act to control imprinting. One involves genome-wide DNA methylation in the gametes, followed by maternal allele-specific demethylation in the central cell, allowing transcription of maternal alleles in the endosperm after fertilization [29–34]. The second mechanism relies on POLYCOMB REPRESSIVE COMPLEX 2 (PRC2) to repress transcription of maternal alleles of paternally expressed genes (PEGs), as well as paternal alleles of maternally expressed genes (MEGs) (reviewed in [18]) [20,24,31,35,36]. In mammals, imprinting is known to play an important role in development of the placenta (a tissue somewhat analogous to the endosperm in flowering plants) (reviewed in [37]). Imprinted genes also play important roles in mammalian embryo development, while gene imprinting in plant embryogenesis is just beginning to be studied (reviewed in [37,38]). Below, we summarize recent work on zygotic genome activation and imprinting, two types of parent-of-origin dependent gene expression during embryo development. Since both result in preferential expression of one parental allele, we outline how ZGA and imprinting can be distinguished by both their phenomena and mechanisms. We also point out important unresolved questions for future research.
Zygotic genome activation In an extensive study published several years ago, Autran et al. revisited the issue of paternal allele activation in Current Opinion in Plant Biology 2015, 27:29–35
30 Cell signalling and gene regulation
early embryogenesis [9]. They found that numerous GUS reporters showed incomplete paternal expression at the earliest stages of embryogenesis, increasing to full penetrance within 3–4 days after pollination (dap). At the genome scale, RNA transcriptome analysis of Landsberg erecta (Ler) Columbia (Col) hybrid embryos at the 2–4 cell stage (2 dap) showed 1% of genes were strictly paternal, 13% were biallelic, and 86% with partial or total maternal bias; by the early globular stage (3 dap), the biallelic class increased to 47%. Pathways that repress or promote paternal transcription in early embryogenesis were also identified. Maternal mutations in genes encoding H3K9 methylation, non-CG DNA methylation, and RNA directed DNA Methylation (RdDM) caused earlier paternal activation at the single gene and genome level, while mutations in CG DNA methylation pathways had no effect. By contrast, maternal mutations in several members of the CAF1 nucleosome assembly complex delayed paternal allele activation. Autran et al. concluded that the paternal genome makes initially limited contributions to the transcriptome, which increase over the first few days of embryogenesis [9]. A subsequent experiment on parent-of-origin transcription in hybrid embryos came to a different conclusion: that the maternal and paternal genomes make equivalent transcriptional contributions at the 1–2 cell, 8 cell and 32 cell embryo stages (corresponding to approximately 2, 2.5 and 3 dap) [11]. Nodine and Bartel used embryos hybrid for the Cape Verde Islands (Cvi) and Col strains, resulting in a transcriptome resolution 20 times higher than the previous study [9]. Transcriptomes were also generated from crosses in both directions (Cvi Col and Col Cvi), allowing identification of about 100 potentially imprinted genes, as well as hundreds of genes whose differential expression was dependent on the strain, regardless of the parent-of-origin. Use of bi-directional crosses also allowed exclusion of the possibility of endosperm or maternal tissue contamination of embryo transcriptome results [11]. Recently, Del Toro-De Leo´n et al. assayed gene activity in early embryogenesis by testing the ability of wild type paternal alleles to complement early embryo phenotypes conditioned by mutant maternal alleles [15]. This experiment was performed on 49 genes in isogenic Col, avoiding the hybrid genetic backgrounds that are required for RNA sequence profiling of maternally and paternally derived transcripts [9,11]. In isogenic embryos, the onset of paternal activity was found to vary by gene, from before 2 dap, to 3, 4, or 5 dap [15]. A second experiment demonstrated large effects on the onset of paternal gene activity when hybrid embryos were used in the assay. Notably, the Col Ler and Col Cvi hybrids used in [9] and [11] were found to be significantly different from isogenic Col embryos, and also significantly different from each other. Of all the hybrids tested, Current Opinion in Plant Biology 2015, 27:29–35
Col Cvi had the earliest paternal gene activation, consistent with [11]. Natural epigenetic variation between Cvi and other ecotypes may explain the earlier activation of paternal gene expression in Cvi hybrids compared with other hybrids [15]. Compared to Ler, Cvi shows reduced chromatin compaction, reduced H3K9me2 at chromocenters, and decreased HISTONE DEACETYLASE 6 (HDA6) activity, all of which are characteristic of transcriptionally permissive chromatin [39,40]. Cvi was also shown to have global CG hypomethylation in vegetative tissues, appearing as the outgroup in a recent methylome study of 152 ecotypes [41]. Likewise, Cvi has much lower levels of CG methylation in genes in embryos compared to Col and Ler, though non-CG methylation levels between Cvi, Ler and Col embryos are not particularly different [27]. Since only non-CG methylation was shown to regulate paternal allele activation [9], the increase in paternal allele activity in Cvi hybrids is more likely due to differences in histone modifications, or other, unknown factors. Thus, paternal alleles for most genes initially show decreased activity or transcription compared to their maternal counterparts. Delayed paternal allele activation partly reflects a maternal requirement for nucleosome assembly and histone H3.3 turnover via the CAF1 complex, to promote paternal transcription [9]. Replacement of parentally derived histone H3 variants occurs before the first division of the zygote [42], long before the release of paternal silencing by the activity of numerous maternal epigenetic pathways [9]. Therefore, silencing must act on epigenetic marks derived from the sperm (Figure 1). Why silence paternal alleles in early embryogenesis? One possibility is that precocious paternal transcription causes defects in development. Indeed, mutations in KYP and CHROMOMETHYLASE 3 (CMT3), encoding two enzymes demonstrated to repress paternal transcription [9], cause morphological defects during early embryogenesis [9,43]. In animals, the delay in zygotic genome activation after fertilization is thought to allow time for remodeling of chromatin from a gametic to zygotic state, and to avoid excessive paternal control of early embryogenesis (reviewed in [44]). According to theories of parental conflict in seed development (reviewed in [45]), avoiding paternal control might also be an important reason for delayed paternal genome activation in seed development.
Imprinted gene expression in embryos The first imprinted gene to be discovered in plant embryos was mee1 of maize. mee1 shows almost exclusive maternal expression from 3 to 8 dap, after which it is not expressed. Interestingly, mee1 shows high levels of CG and CHG methylation in both gametes, but the maternal allele is actively demethylated in zygotes, indicating that an asymmetric epigenetic mark exists to recruit demethylation of the maternal allele in the diploid zygote [46]. In rice, the Os10g05750 gene was also demonstrated to show www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 31
Figure 1
sperm
zygote (1 dap)
2-4 cell (2 dap)
dermatogen (2.5 dap)
globular (3 dap)
heart (5 dap)
torpedo (6 dap)
+ egg
Maternal CAF1, H3.3
(a) Zygotic Genome Activation
(b) Imprinting
Maternal H3K9me, non-CGme, RdDM
thousands of genes ? Maternal PRC2
tens or hundreds of genes ?
Current Opinion in Plant Biology
Zygotic genome activation and Imprinting in early embryogenesis. A summary of the work on Zygotic genome activation (ZGA) and imprinting discussed in this review. (a) ZGA is a genome-wide phenomenon that involves thousands of genes. Paternal transcription (green) varies by gene, but for the majority of genes is probably initially less than maternal transcription (blue). At the genome level, paternal transcription increases during the first few days after fertilization, reaching levels similar to maternal transcription by about the heart stage (5 dap). The maternal CAF1 complex promotes paternal transcription through histone H3 exchange and nucleosome assembly. Maternal H3K9me, non-CGme and RdDM activity recognize epigenetic marks in the sperm (red) in order to silence paternal alleles. The question mark in the blue maternal transcript bar indicates that, as yet, there are no experiments that have definitively assessed maternal transcription, due to the difficulties in distinguishing between maternal gametophytic and zygotic transcripts in the zygote. (b) Imprinting affects a much smaller number of genes in the embryo — tens or perhaps hundreds. Imprinting results in parentally biased transcription that lasts throughout embryogenesis. Maternal PRC2 activity is required for silencing of some imprinted genes. Epigenetic marks from the sperm (red) are required for silencing paternal alleles of maternally expressed genes (MEGs), while similar marks in the egg would be required to silence maternal alleles of paternally expressed genes (PEGs).
maternal expression up to 8 dap, after which its expression became bi-allelic [21]. A genomic level study to characterize parent-of-origin expression in embryos of Arabidopsis identified 77 potential MEGs and 45 potential PEGs; expression of three MEGs and two PEGs was verified by RT-PCR up to the globular stage (3 dap) [11]. A number of other genome-wide studies whose primary focus was discovery of endosperm-imprinted genes also looked for imprinting in the embryo (Table 1) [19,20,23,27]. In total, these four studies discovered approximately 100 candidate embryo imprinted genes. The imprinting status for most of these genes was not pursued further, either because the few genes that were validated in the embryo proved to be false positives [20], or due to concerns that embryo RNA-seq data were contaminated by endosperm transcripts (most potential embryo imprinted genes were MEGs) [19]. Nonetheless, many genes which are potentially imprinted in the embryo exist, and await validation. By contrast, a recent study in Arabidopsis specifically sought to identify imprinted genes in embryos. Starting www.sciencedirect.com
with 80 candidate imprinted genes, 24 genes were studied in detail, using allele-specific RT-PCR, and GUS reporter lines. This analysis identified 11 MEGs and 1 PEG, the majority of which showed uniparental expression. Imprinting marks were lost after embryogenesis, as all 12 genes were found to be expressed from both alleles during seedling development [47]. Since the primary mechanisms controlling imprinting of genes in the endosperm involve CG methylation by METHYLTRANSFERASE 1 (MET1), and H3K27me3 methylation by the POLYCOMB REPRESSIVE COMPLEX 2 (PRC2), Raissig et al. tested the effect of paternal mutation in MET1 and maternal mutation in the PRC2 component FERTILIZATION INDEPENDENT ENDOSPERM (FIE) on imprinted expression. met1 was found to have no effect on silencing of paternal alleles, while fie mutants affected allele-specific expression for only 3 of the 12 imprinted genes [47]. These experiments and those on mee1 suggest that numerous mechanisms regulate imprinting in the embryo, and that some of these mechanisms are different than those for imprinting in the endosperm. Current Opinion in Plant Biology 2015, 27:29–35
32 Cell signalling and gene regulation
Table 1 Parent-of-origin studies of imprinted genes in plant embryos Cross
Species Arabidopsis thaliana
Ler Col
Zea mays
B73 H99 TX303 W22 Mo17 A188 Col Ler Col Ler Col Cvi Ler Cvi Col Cvi B73 Mo17 Nip 93-11 SR1 Hamayan
Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis
thaliana thaliana thaliana thaliana thaliana
Zea mays Oriza sativa Nicotiana tabacum
MEGs (n) 11
1
1
0
17 37 14 15 77
1 1 1 1 45
29 1 1
9 0 4
What is the functional importance of imprinting? Mutation of genes involved in the mechanism of imprinting has strong phenotypic effects on seeds, indicating that imprinting is indeed globally important for development (reviewed in [18]). However, there are only two examples for the importance of imprinting per se in plants. The Maternally expressed gene 1 (Meg1) of maize controls nutrient transfer from the endosperm to embryo; changes in dosage of Meg1 affect kernel size by changing nutrient allocation to the embryo [48]. The PEG ADMETOS (ADM) of Arabidopsis has been shown to be responsible for seed abortion resulting from crosses that produce unbalanced endosperms (i.e. ratios other than 2m:1p) [49]. This seed abortion phenotype is dosage sensitive, indicating a functional role for imprinting of ADM. Whether imprinting of the hundreds of other genes in the endosperm or embryo has functional importance remains to be seen. One possibility is that imprinted expression of transposable elements (TEs) in the endosperm could result in production of siRNAs that non-cell autonomously silence TE expression in the embryo [50,51]. Indeed, siRNAs produced by the endosperm are derived from the maternal genome [52,53]. Since the endosperm does not contribute to the adult plant or the germline, this would be an effective way of silencing TE expression in the embryo, similar to mechanisms proposed for TE silencing in the gametes by siRNA production in companion cells [54,55]. TEs are often targets of the deglycosylase DEMETER (DME), leading to differentially methylated regions (DMRs) that cause imprinted expression of the TEs in the endosperm. Genes that are located near these DMRs can show imprinted expression, an indirect effect of imprinted expression of nearby TEs [55]. Thus, in at least some cases, imprinted gene expression could be a secondary effect of transcriptional unmasking of endosperm TEs that participate in embryo TE silencing. Current Opinion in Plant Biology 2015, 27:29–35
PEGs (n)
Verified?
Reference
RT-PCR & restriction digest; RT-PCR & sequencing; GUS markers RT-PCR & restriction digest
[47]
Considered false positives Considered false positives Not verified Not verified 3 MEGs and 2 PEGs by RT-PCR & restriction digest Not verified RT-PCR seq RT-PCR & restriction digest
[19] [20] [27] [27] [11]
[46]
[23] [21] [10]
Conclusions An important goal for future research on parent-oforigin dependent gene expression in plant embryogenesis is to clearly differentiate between ZGA and imprinting. Since there is considerable evidence that ZGA occurs in a parent-of-origin specific manner in plants [7,9,10,15], ZGA could be defined as a transient form of imprinting. Despite this, the limited research done so far points both to phenomena and mechanisms that differentiate between the two processes (Figure 1). In terms of phenomena, ZGA is defined as a genomewide process affecting thousands of genes, while imprinting need affect only a handful of genes (as in the embryo), or hundreds of genes (as in the endosperm) [47]. It will be necessary to perform additional experiments to demonstrate that ZGA in Arabidopsis does indeed occur in a parent-of-origin specific manner at the genome level, ideally using hybrid combinations demonstrated by Del Toro-De Leo´n et al. to show paternal allele activation patterns similar to isogenic Col [15]. Phylogenetic depth in studies of ZGA could also indicate if delayed paternal (or zygotic) genome activation is widespread in the plant kingdom. Thus far, the majority of parent-of-origin studies in plants have been conducted with Arabidopsis or maize. Studies in other monocots, basal angiosperms such as Amborella trichopoda, or gymnosperms would be useful in determining if delayed paternal genome activation is a conserved or derived trait [56]. It would also be interesting to look at post-fertilization transcriptional activation in a bryophyte such as Physcomitrella patens or Marchantia polymorpha. The study of plants with a transient sporophytic phase might indicate whether ZGA results from genome reprogramming after fertilization, or from a developmental requirement for early maternal control of pattern formation for the formation of complex sporophyte body plants. www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 33
For imprinting, it is important to have greater resolution on the timing of this process in the embryo. Both studies that examined imprinting in Arabidopsis embryos assayed gene expression only in the first few days of embryogenesis, when ZGA also occurs, and making it formally impossible to distinguish between ZGA and imprinting [11,47]. Determining allele-specific expression patterns at 6 dap and after could easily differentiate between ZGA and imprinting, as ZGA is essentially finished for most genes at 5 dap [9,15]. Indeed, allele-specific profiling to determine imprinting in the endosperm has also typically been performed at only a single timepoint [19,20,23,27]. By profiling whole seed at 0, 3, and 5 dap, as well as dissected endosperm at 7, 10, and 15 dap, Xin et al. [26] were able to achieve the highest temporal resolution of any study of imprinting. Their sampling of early timepoints suggested that the paternal genome in the endosperm undergoes gradual transcriptional activation, similar to that seen in embryos of Arabidopsis. ZGA and imprinting can also be distinguished at the mechanistic level. Paternal allele activation during ZGA was shown to be repressed by H3K9 methylation, non-CG DNA methylation, and the RdDM pathway, but not to be affected by CG DNA methylation or PRC2 [9] By contrast, CG DNA methylation and PRC2 were correlated with imprinting of certain genes [46,47] (Figure 1). Further mechanistic studies will be required to better understand how parent-of-origin specific marks of silencing or activation are deposited in the gametes, and acted upon in the zygote. A clearer understanding of ZGA and imprinting will allow us to better distinguish between the two processes, and to recognize how they interact to regulate gene expression in early embryogenesis.
Acknowledgements Thanks to Wolfgang Lukowitz and Gerardo Del Toro De Leo´n for comments on the manuscript. Research on zygotic genome activation in the authors’ laboratory is supported by CONACyT Ciencia Ba´sica No. 237480, and by CINVESTAV institutional funds. C.S.G. is a DuPont Young Professor.
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
Baroux C, Autran D, Gillmor CS, Grimanelli D, Grossniklaus U: The maternal to zygotic transition in animals and plants. Cold Spring Harb Symp Quant Biol 2008, 73:89-100.
2.
Lee MT, Bonneau AR, Giraldez AJ: Zygotic genome activation during the maternal-to-zygotic transition. Annu Rev Cell Dev Biol 2014, 30:581-613.
3.
Aoki F, Worrad DM, Schultz RM: Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev Biol 1997, 181: 296-307.
www.sciencedirect.com
4.
Hamatani T, Carter MG, Sharov AA, Ko MSH: Dynamics of global gene expression changes during mouse preimplantation development. Dev Cell 2004, 6:117-131.
5.
Tadros W, Lipschitz HD: The maternal-to-zygotic transition: a play in two acts. Development 2009, 136:3033-3042.
6.
Harvey SA, Sealy I, Kettleborough R, Fenyes F, White R, Stemple D, Smith JC: Identification of the zebrafish maternal and paternal transcriptomes. Development 2013, 140:27032710.
7.
Vielle-Calzada JP, Baskar R, Grossniklaus U: Delayed activation of the paternal genome during early seed development. Nature 2000, 404:91-94.
8.
Meyer S, Scholten S: Equivalent parental contribution to early plant zygotic development. Curr Biol 2007, 17:1686-1691.
Autran D, Baroux C, Raissig MT, Lenormand T, Wittig M, Grob S, Steimer A, Barann M, Klostermeier UC, Leblanc O, VielleCalzada J-P, Rosenstiel P, Grimanelli D, Grossniklaus U: Maternal epigenetic pathways control parental contributions to Arabidopsis early embryogenesis. Cell 2011, 145:707-719. A tour-de-force study characterizing the phenomena and mechanisms of paternal allele activation at the individual gene and genome level. The paternal genome was shown to be gradually activated after fertilization, resulting in early maternal bias (but with limited paternal contribution from most genes). Maternally encoded DNA and histone methylation were shown to repress paternal transcription, whereas maternal activity of the histone chaperone complex CAF1 was discovered to promote paternal transcription.
9.
10. Zhang J-E, Luo A, Xin H-P, Zhao J, Li S-S, Qu L-H, Ma L-G, Scholten S, Sun M-X: Genes of both parental origins are differentially involved in early embryogenesis of a tobacco interspecies hybrid. PLoS ONE 2011, 6:e23153. 11. Nodine MD, Bartel DP: Maternal and paternal genomes contribute equally to the transcriptome of early plant embryos. Nature 2012, 482:94-97. This work used Cvi–Col hybrid embryos to uncover approximately 100 potentially embryo-imprinted genes, and pointed the way to discovery of the effects of intraspecific hybridization on transcriptional activation in early embryos. 12. Weijers D, Geldner N, Offringa R, Ju¨rgens G: Seed development: early paternal gene activity in Arabidopsis. Nature 2001, 414:709-710. 13. Lukowitz W, Roeder A, Parmenter D, Somerville C: A MAPKK kinase gene regulates extra-embryonic cell fate in Arabidopsis. Cell 2004, 116:109-119. 14. Ueda M, Zhang Z, Laux T: Transcriptional activation of Arabidopsis axis patterning genes WOX8/9 links zygote polarity to embryo development. Dev Cell 2011, 20:264-270. 15. Del Toro-De Leo´n G, Garcı´a-Aguilar M, Gillmor CS: Non equivalent contributions of maternal and paternal genomes to early plant embryogenesis. Nature 2014, 514:624-627. Using a genetic test for paternal allele function in early embryogenesis, this study demonstrated that paternal allele activity in early embryogenesis varies widely between genes. In addition, an unexpectedly large effect of certain hybrid combinations on early gene activity was demonstrated, providing a plausible explanation for differences in transcriptional activation observed between [9] and [11]. 16. Bayer M, Nawy T, Giglione C, Galli M, Meinnel T, Lukowitz W: Paternal control of embryonic patterning in Arabidopsis thaliana. Science 2009, 323:1485-1488. 17. Costa LM, Marshall E, Tesfaye M, Silverstein KAT, Mori M, Umetsu Y, Otterbach SL, Papareddy R, Dickinson HG, Boutiller K, VandenBosch KA, Ohki S, Gutie´rrez-Marcos J: Central cellderived peptides regulate early embryo patterning in flowering plants. Science 2014, 344:168-172. This work uncovered a family of peptides expressed from the endosperm which promote suspensor elongation through the YODA MAPKKK pathway. This was the first demonstration of a direct effect of the central cell and endosperm on embryo patterning. 18. Gehring M: Genomic imprinting: insights from plants. Annu Rev Genet 2013, 47:187-208. A comprehensive and thoughtful review of imprinting in plants. Current Opinion in Plant Biology 2015, 27:29–35
34 Cell signalling and gene regulation
19. Gehring M, Missirian V, Henikoff S: Genomic analysis of parentof-origin allelic expression in Arabidopsis thaliana seeds. PLoS ONE 2011, 6:e23687.
34. Hsieh T-F, Ibarra CA, Silva P, Zemach A, Eshed-Williams L, Fischer RL, Zilberman D: Genome-wide demethylation of Arabidopsis endosperm. Science 2009, 324:1451-1454.
20. Hsieh T-F, Shin J, Uzawa R, Silva P, Cohen S, Bauer MJ, Hashimoto M, Kirkbride RC, Harada JJ, Zilberman D, Fischer RL: Regulation of imprinted gene expression in Arabidopsis endosperm. Proc Natl Acad Sci USA 2011, 108:1755-1762.
35. Makarevitch I, Eichten SR, Briskine R, Waters AJ, Danilevskaya ON, Meeley RB, Myers CL, Vaughn MW, Springer NM: Genomic distribution of maize facultative heterochromatin marked by trimethylation of H3K27. Plant Cell 2013, 25:780-793.
21. Luo M, Taylor JM, Spriggs A, Zhang H, Wu X, Russell S, Singh M, Koltunow A: A genome-wide survey of imprinted genes in rice seeds reveals imprinting primarily occurs in the endosperm. PLoS Genet 2011, 7:e1002125. 22. McKeown PC, Laouielle-Duprat S, Prins P, Wolff P, Schmid MW et al.: Identification of imprinted genes subject to parent-oforigin specific expression in Arabidopsis thaliana seeds. BMC Plant Biol 2011, 11:113. 23. Waters AJ, Makarevitch I, Eichten SR, Swanson-Wagner RA, Yeh C-T, Xu W, Schnable PS, Vaughn MW, Gehring M, Springer NM: Parent-of-origin effects on gene expression and DNA methylation in the maize endosperm. Plant Cell 2011, 23:4221-4233. 24. Wolff P, Weinhofer I, Seguin J, Roszak P, Beisel C, Donoghue MTA, Spillane C, Nordborg M, Rehmsmeier M, Ko¨hler C: High-resolution analysis of parent-of-origin allelic expression in the Arabidopsis endosperm. PLoS Genet 2011, 7:e1002126. 25. Waters AJ, Bilinski P, Eichten SR, Vaughn MW, Ross-Ibarra J, Gehring M, Springer NM: Comprehensive analysis of imprinted genes in maize reveals allelic variation for imprinting and limited conservation with other species. Proc Natl Acad Sci USA 2013, 110:19639-19644. 26. Xin M, Yan R, Li G, Chen H, Laurie J, Ma C, Wang D, Yao Y, Larkins B, Sun Q, Yadegari R, Wang X, Ni Z: Dynamic expression of imprinted genes associates with maternally controlled nutrient allocation during maize endosperm. Plant Cell 2013, 25:3212-3227. Multiple sampling timepoints were used to study of parent-of-origin contributions to transcription in early seeds and endosperm of maize, capturing gradual activation of paternal alleles in the seed. This study provides the first potential link between post-fertilization genome activation and imprinting. 27. Pignatta D, Erdmann RM, Scheer E, Picard CL, Bell GW, Gehring M: Natural epigenetic polymorphisms lead to intraspecific variation in Arabidopsis gene imprinting. eLife 2014, 3:e03198. This study describes intraspecific variation in imprinting between hybrids of Col, Ler and Cvi. Compared to other ecotypes, Cvi was found to show CG hypomethylation in embryos. 28. Xu W, Dai M, Li F, Liu A: Genomic imprinting, methylation and parent-of-origin effects in reciprocal hybrid endosperm of castor bean. Nucleic Acids Res 2014, 42:6987-6998. 29. Choi Y, Gehring M, Johnson L, Hannon M, Harada JJ, Goldberg RB, Jacobsen SE, Fischer RL: DEMETER, a DNA glycosylase domain protein, is required for endosperm gene imprinting and seed viability in Arabidopsis. Cell 2002, 110: 33-42. 30. Xiao W, Gehring M, Choi Y, Margossian L, Pu H, Harada JJ, Goldberg RB, Pennell RI, Fischer RL: Imprinting of the MEA Polycomb gene is controlled by antagonism between MET1 methyltransferase and DME glycosylase. Dev Cell 2003, 5: 891-901. 31. Gehring M, Huh JH, Hsieh T-F, Penterman J, Choi Y, Harada JJ, Goldberg RB, Fischer RL: DEMETER DNA glycosylase establishes MEDEA polycomb gene self-imprinting by allelespecific demethylation. Cell 2006, 124:495-506. 32. Jullien PE, Mosquna A, Ingouff M, Sakata T, Ohad N, Berger F: Retinoblastoma and its binding partner MSI1 control imprinting in Arabidopsis. PLoS Biol 2008, 6:e194. 33. Gehring M, Bubb KL, Henikoff S: Extensive demethylation of repetitive elements during seed development underlies gene imprinting. Science 2009, 324:1447-1451. Current Opinion in Plant Biology 2015, 27:29–35
36. Zhang M, Xie S, Dong X, Zhao X, Zeng B, Chen J, Li H, Yang W, Zhao H, Wang G, Chen Z, Sun S, Hauck A, Jin W, Lai J: Genomewide high resolution parental-specific DNA and histone methylation maps uncover patterns of imprinting regulation in maize. Genome Res 2014, 24:167-176. 37. Barlow DP, Bartolomei MS: Genomic imprinting in mammals. Cold Spring Harb Perspect Biol 2014, 6:a018382. 38. Dickinson H, Scholten S: And baby makes three: genomic imprinting in plant embryos. PLoS Genet 2013, 9:e1003981. 39. Tessadori F, van Zanten M, Pavlova P, Clifton R, Pontvianne F, Snoek LB, Millenaar FF, Schulkes RK, van Driel R, Voesenek LACJ, Spillane C, Pikaard CS, Fransz P, Peeters AJM: PHYTOCHROME B and HISTONE DEACETYLASE 6 Control Light-Induced Chromatin Compaction in Arabidopsis thaliana. PLoS Genet 2009, 5:e1000638. 40. Baroux C, Autran D, Raissig MT, Grimanelli D, Grossniklaus U: Parental contributions to the transcriptome of early plant embryos. Curr Opin Genet Dev 2013, 23:72-74. 41. Schmitz RJ, Schultz MD, Urich MA, Nery JR, Pelizzola M, Libiger O, Alix A, McCosh RB, Chen H, Schork NJ, Ecker JR: Patterns of population epigenomic diversity. Nature 2013, 495:193-198. 42. Ingouff M, Rademacher S, Holec S, Soljic’ L, Xin N, Readshaw A, Foo SH, Lahouze B, Sprunck S, Berger F: Zygotic resetting of the HISTONE 3 variant repertoire participates in epigenetic reprogramming in Arabidopsis. Curr Biol 2010, 20:2137-2143. 43. Pillot M, Baroux C, Arteaga-Va´zquez M, Autran D, Leblanc O, Vielle-Calzada J-P, Grossniklaus U, Grimanelli D: Embryo and endosperm inherit distinct chromatin and transcriptional states from the female gametes in Arabidopsis. Plant Cell 2010, 22:307-320. 44. Blaxter M: Development: the maternal-zygotic transition revisited. Curr Biol 2014, 24:R72-R75. 45. Haig D: Kin conflict in seed development: an interdependent but fractious collective. Annu Rev Cell Dev Biol 2013, 29: 189-211. 46. Jahnke S, Scholten S: Epigenetic resetting of a gene imprinted in plant embryos. Curr Biol 2009, 19:1677-1681. 47. Raissig MT, Bemer M, Baroux C, Grossniklaus U: Genomic imprinting in the Arabidopsis embryo is partly regulated by PRC2. PLoS Genet 2013, 9:e1003862. This study discovered 12 genes with imprinted expression in the Arabidopsis embryo, and demonstrated that PRC2 regulates the imprinted expression of a subset of these genes. Together with mee1 of maize, this work provides the only thorough documentation of imprinted gene expression in plant embryos. 48. Costa LM, Yuan J, Rouster J, Paul W, Dickinson H, Gutie´rrez Marcos JF: Maternal control of nutrient allocation in plant seeds by genomic imprinting. Curr Biol 2012, 22:160-165. This work showed that the endosperm imprinted Meg1 gene of maize plays an important role in nutrient allocation from the endosperm to the embryo, and demonstrated that Meg1 is one of only two imprinted genes in plants whose dosage has been shown to be functionally important. 49. Kradolfer D, Wolff P, Jiang H, Siretskiy A, Ko¨hler C: An imprinted gene underlies postzygotic reproductive isolation in Arabidopsis thaliana. Dev Cell 2013, 26:525-535. The imprinted ADM gene of Arabidopsis was shown to be required for growth failure of unbalanced endosperms (maternal to paternal ratios other than 2m:1p). ADM is a PEG whose increased dosage leads to triploid seed arrest. Along with Meg1 of maize, ADM is one of only two imprinted genes of functional importance. www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 35
50. Van Ex F, Jacob Y, Martienssen RA: Multiple roles for small RNAs during plant reproduction. Curr Opin Plant Biol 2011, 14:588-593. 51. Calarco JP, Borges F, Donoghue MTA, Van Ex F, Jullien PE et al.: Reprogramming of DNA methylation in pollen guides epigenetic inheritance via small RNA. Cell 2012, 151:194-205. 52. Mosher RA, Melnyk CW, Kelly KA, Dunn RM, Studholme DJ, Baulcombe DC: Uniparental expression of PolIV-dependent siRNAs in developing endosperm of Arabidopsis. Nature 2009, 460:283-286. 53. Lu J, Zhang C, Baulcombe DC, Chen ZJ: Maternal siRNAs as regulators of parental genome imbalance and gene expression in endosperm of Arabidopsis seeds. Proc Natl Acad Sci USA 2012, 109:5529-5534.
www.sciencedirect.com
54. Slotkin RK, Vaughn M, Borges F, Tanurdzic’ M, Becker JD, Feijo´ JA, Martienssen RA: Epigenetic reprogramming and small RNA silencing of transposable elements in pollen. Cell 2009, 136:461-472. 55. Ibarra CA, Feng X, Schoft VK, Hsieh TF, Uzawa R, Rodrigues JA, Zemach A, Chumak N, Machlicova A, Nishimura T, Rojas D, Fischer RL, Tamaru H, Zilberman D: Active DNA demethylation in plant companion cells reinforces transposon methylation in gametes. Science 2012, 337:1360-1364. 56. Finet C, Timme RE, Delwiche CF, Marle´taz F: Multigene phylogeny of the green lineage reveals the origin and diversification of land plants. Curr Biol 2010, 20: 2217-2222.
Current Opinion in Plant Biology 2015, 27:29–35
ScienceDirect Zygotic genome activation and imprinting: parent-of-origin gene regulation in plant embryogenesis Marcelina Garcı´a-Aguilar and C Stewart Gillmor Parent-of-origin dependent gene expression refers to differential activity of alleles inherited from the egg and sperm. In plants, zygotic genome activation (ZGA) and gene imprinting are two examples of this phenomenon, both of which occur during seed development. As its name implies, ZGA is a genome-wide process that occurs in embryos during the first few days after fertilization. Evidence exists that maternal alleles initially predominate during ZGA, although most genes also show some paternal activity. By contrast, imprinting can be defined as a bias in gene expression that lasts beyond the first few days of seed development. Hundreds of imprinted genes have been discovered in the endosperm, and a few have been described in the embryo. This review discusses recent advances in our understanding of the phenomena and mechanisms of ZGA and imprinting in seeds, with an emphasis on embryo development. Important unanswered questions and areas for future research are highlighted. Address Laboratorio Nacional de Geno´mica para la Biodiversidad (Langebio), Unidad de Geno´mica Avanzada, Centro de Investigacio´n y de Estudios Avanzados (CINVESTAV), Irapuato, Guanajuato 36821, Me´xico Corresponding author: Gillmor, C Stewart ([email protected])
Current Opinion in Plant Biology 2015, 27:29–35 This review comes from a themed issue on Cell signalling and gene regulation Edited by Xiaofeng Cao and Blake C Meyers
http://dx.doi.org/10.1016/j.pbi.2015.05.020 1369-5266/# 2015 Elsevier Ltd. All rights reserved.
Introduction Seed development initiates with a double fertilization of the haploid egg and diploid central cell, producing the diploid embryo and triploid endosperm. In the embryo, zygotic genome activation (ZGA) occurs following a period of transcriptional quiescence after fertilization (reviewed in [1,2]). With few exceptions, ZGA in animals is thought to occur in a bi-allelic manner [3,4], though most studies of animal ZGA have used methods that would not be able to detect parent-of-origin effects on zygotic transcription (reviewed in [5], see [6] for an exception). In contrast, studies on ZGA in plants have www.sciencedirect.com
relied on expression of reporter genes and allele-specific RNA-seq profiling [7,8,9,10,11], as well as genetic tests for allele function [7,12–14,15]. These experiments have uncovered many cases of parent-of-origin effects on gene expression in early plant embryogenesis [7,9,10,11,15], along with examples of equal parental contributions [8,11,12–14]. In addition, one maternal effect and one paternal effect gene are known to be transcribed in gametes, yet act in the embryo [16,17]. Imprinting causes the expression of a gene to be dependent on its parent of origin, due to differential epigenetic modifications established during male and female gametogenesis (reviewed in [18]). According to expression bias, imprinted genes are classified as maternally expressed genes (MEGs) or paternally expressed genes (PEGs). Genome scale studies of imprinted genes in Arabidopsis, maize, rice and other plants have identified over 500 MEGs and about 100 PEGs in the endosperm [19–25,26,27,28]. Most of these genes show transcriptional bias in favor of one allele or the other, as opposed to uniparental expression. At least two mechanisms act to control imprinting. One involves genome-wide DNA methylation in the gametes, followed by maternal allele-specific demethylation in the central cell, allowing transcription of maternal alleles in the endosperm after fertilization [29–34]. The second mechanism relies on POLYCOMB REPRESSIVE COMPLEX 2 (PRC2) to repress transcription of maternal alleles of paternally expressed genes (PEGs), as well as paternal alleles of maternally expressed genes (MEGs) (reviewed in [18]) [20,24,31,35,36]. In mammals, imprinting is known to play an important role in development of the placenta (a tissue somewhat analogous to the endosperm in flowering plants) (reviewed in [37]). Imprinted genes also play important roles in mammalian embryo development, while gene imprinting in plant embryogenesis is just beginning to be studied (reviewed in [37,38]). Below, we summarize recent work on zygotic genome activation and imprinting, two types of parent-of-origin dependent gene expression during embryo development. Since both result in preferential expression of one parental allele, we outline how ZGA and imprinting can be distinguished by both their phenomena and mechanisms. We also point out important unresolved questions for future research.
Zygotic genome activation In an extensive study published several years ago, Autran et al. revisited the issue of paternal allele activation in Current Opinion in Plant Biology 2015, 27:29–35
30 Cell signalling and gene regulation
early embryogenesis [9]. They found that numerous GUS reporters showed incomplete paternal expression at the earliest stages of embryogenesis, increasing to full penetrance within 3–4 days after pollination (dap). At the genome scale, RNA transcriptome analysis of Landsberg erecta (Ler) Columbia (Col) hybrid embryos at the 2–4 cell stage (2 dap) showed 1% of genes were strictly paternal, 13% were biallelic, and 86% with partial or total maternal bias; by the early globular stage (3 dap), the biallelic class increased to 47%. Pathways that repress or promote paternal transcription in early embryogenesis were also identified. Maternal mutations in genes encoding H3K9 methylation, non-CG DNA methylation, and RNA directed DNA Methylation (RdDM) caused earlier paternal activation at the single gene and genome level, while mutations in CG DNA methylation pathways had no effect. By contrast, maternal mutations in several members of the CAF1 nucleosome assembly complex delayed paternal allele activation. Autran et al. concluded that the paternal genome makes initially limited contributions to the transcriptome, which increase over the first few days of embryogenesis [9]. A subsequent experiment on parent-of-origin transcription in hybrid embryos came to a different conclusion: that the maternal and paternal genomes make equivalent transcriptional contributions at the 1–2 cell, 8 cell and 32 cell embryo stages (corresponding to approximately 2, 2.5 and 3 dap) [11]. Nodine and Bartel used embryos hybrid for the Cape Verde Islands (Cvi) and Col strains, resulting in a transcriptome resolution 20 times higher than the previous study [9]. Transcriptomes were also generated from crosses in both directions (Cvi Col and Col Cvi), allowing identification of about 100 potentially imprinted genes, as well as hundreds of genes whose differential expression was dependent on the strain, regardless of the parent-of-origin. Use of bi-directional crosses also allowed exclusion of the possibility of endosperm or maternal tissue contamination of embryo transcriptome results [11]. Recently, Del Toro-De Leo´n et al. assayed gene activity in early embryogenesis by testing the ability of wild type paternal alleles to complement early embryo phenotypes conditioned by mutant maternal alleles [15]. This experiment was performed on 49 genes in isogenic Col, avoiding the hybrid genetic backgrounds that are required for RNA sequence profiling of maternally and paternally derived transcripts [9,11]. In isogenic embryos, the onset of paternal activity was found to vary by gene, from before 2 dap, to 3, 4, or 5 dap [15]. A second experiment demonstrated large effects on the onset of paternal gene activity when hybrid embryos were used in the assay. Notably, the Col Ler and Col Cvi hybrids used in [9] and [11] were found to be significantly different from isogenic Col embryos, and also significantly different from each other. Of all the hybrids tested, Current Opinion in Plant Biology 2015, 27:29–35
Col Cvi had the earliest paternal gene activation, consistent with [11]. Natural epigenetic variation between Cvi and other ecotypes may explain the earlier activation of paternal gene expression in Cvi hybrids compared with other hybrids [15]. Compared to Ler, Cvi shows reduced chromatin compaction, reduced H3K9me2 at chromocenters, and decreased HISTONE DEACETYLASE 6 (HDA6) activity, all of which are characteristic of transcriptionally permissive chromatin [39,40]. Cvi was also shown to have global CG hypomethylation in vegetative tissues, appearing as the outgroup in a recent methylome study of 152 ecotypes [41]. Likewise, Cvi has much lower levels of CG methylation in genes in embryos compared to Col and Ler, though non-CG methylation levels between Cvi, Ler and Col embryos are not particularly different [27]. Since only non-CG methylation was shown to regulate paternal allele activation [9], the increase in paternal allele activity in Cvi hybrids is more likely due to differences in histone modifications, or other, unknown factors. Thus, paternal alleles for most genes initially show decreased activity or transcription compared to their maternal counterparts. Delayed paternal allele activation partly reflects a maternal requirement for nucleosome assembly and histone H3.3 turnover via the CAF1 complex, to promote paternal transcription [9]. Replacement of parentally derived histone H3 variants occurs before the first division of the zygote [42], long before the release of paternal silencing by the activity of numerous maternal epigenetic pathways [9]. Therefore, silencing must act on epigenetic marks derived from the sperm (Figure 1). Why silence paternal alleles in early embryogenesis? One possibility is that precocious paternal transcription causes defects in development. Indeed, mutations in KYP and CHROMOMETHYLASE 3 (CMT3), encoding two enzymes demonstrated to repress paternal transcription [9], cause morphological defects during early embryogenesis [9,43]. In animals, the delay in zygotic genome activation after fertilization is thought to allow time for remodeling of chromatin from a gametic to zygotic state, and to avoid excessive paternal control of early embryogenesis (reviewed in [44]). According to theories of parental conflict in seed development (reviewed in [45]), avoiding paternal control might also be an important reason for delayed paternal genome activation in seed development.
Imprinted gene expression in embryos The first imprinted gene to be discovered in plant embryos was mee1 of maize. mee1 shows almost exclusive maternal expression from 3 to 8 dap, after which it is not expressed. Interestingly, mee1 shows high levels of CG and CHG methylation in both gametes, but the maternal allele is actively demethylated in zygotes, indicating that an asymmetric epigenetic mark exists to recruit demethylation of the maternal allele in the diploid zygote [46]. In rice, the Os10g05750 gene was also demonstrated to show www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 31
Figure 1
sperm
zygote (1 dap)
2-4 cell (2 dap)
dermatogen (2.5 dap)
globular (3 dap)
heart (5 dap)
torpedo (6 dap)
+ egg
Maternal CAF1, H3.3
(a) Zygotic Genome Activation
(b) Imprinting
Maternal H3K9me, non-CGme, RdDM
thousands of genes ? Maternal PRC2
tens or hundreds of genes ?
Current Opinion in Plant Biology
Zygotic genome activation and Imprinting in early embryogenesis. A summary of the work on Zygotic genome activation (ZGA) and imprinting discussed in this review. (a) ZGA is a genome-wide phenomenon that involves thousands of genes. Paternal transcription (green) varies by gene, but for the majority of genes is probably initially less than maternal transcription (blue). At the genome level, paternal transcription increases during the first few days after fertilization, reaching levels similar to maternal transcription by about the heart stage (5 dap). The maternal CAF1 complex promotes paternal transcription through histone H3 exchange and nucleosome assembly. Maternal H3K9me, non-CGme and RdDM activity recognize epigenetic marks in the sperm (red) in order to silence paternal alleles. The question mark in the blue maternal transcript bar indicates that, as yet, there are no experiments that have definitively assessed maternal transcription, due to the difficulties in distinguishing between maternal gametophytic and zygotic transcripts in the zygote. (b) Imprinting affects a much smaller number of genes in the embryo — tens or perhaps hundreds. Imprinting results in parentally biased transcription that lasts throughout embryogenesis. Maternal PRC2 activity is required for silencing of some imprinted genes. Epigenetic marks from the sperm (red) are required for silencing paternal alleles of maternally expressed genes (MEGs), while similar marks in the egg would be required to silence maternal alleles of paternally expressed genes (PEGs).
maternal expression up to 8 dap, after which its expression became bi-allelic [21]. A genomic level study to characterize parent-of-origin expression in embryos of Arabidopsis identified 77 potential MEGs and 45 potential PEGs; expression of three MEGs and two PEGs was verified by RT-PCR up to the globular stage (3 dap) [11]. A number of other genome-wide studies whose primary focus was discovery of endosperm-imprinted genes also looked for imprinting in the embryo (Table 1) [19,20,23,27]. In total, these four studies discovered approximately 100 candidate embryo imprinted genes. The imprinting status for most of these genes was not pursued further, either because the few genes that were validated in the embryo proved to be false positives [20], or due to concerns that embryo RNA-seq data were contaminated by endosperm transcripts (most potential embryo imprinted genes were MEGs) [19]. Nonetheless, many genes which are potentially imprinted in the embryo exist, and await validation. By contrast, a recent study in Arabidopsis specifically sought to identify imprinted genes in embryos. Starting www.sciencedirect.com
with 80 candidate imprinted genes, 24 genes were studied in detail, using allele-specific RT-PCR, and GUS reporter lines. This analysis identified 11 MEGs and 1 PEG, the majority of which showed uniparental expression. Imprinting marks were lost after embryogenesis, as all 12 genes were found to be expressed from both alleles during seedling development [47]. Since the primary mechanisms controlling imprinting of genes in the endosperm involve CG methylation by METHYLTRANSFERASE 1 (MET1), and H3K27me3 methylation by the POLYCOMB REPRESSIVE COMPLEX 2 (PRC2), Raissig et al. tested the effect of paternal mutation in MET1 and maternal mutation in the PRC2 component FERTILIZATION INDEPENDENT ENDOSPERM (FIE) on imprinted expression. met1 was found to have no effect on silencing of paternal alleles, while fie mutants affected allele-specific expression for only 3 of the 12 imprinted genes [47]. These experiments and those on mee1 suggest that numerous mechanisms regulate imprinting in the embryo, and that some of these mechanisms are different than those for imprinting in the endosperm. Current Opinion in Plant Biology 2015, 27:29–35
32 Cell signalling and gene regulation
Table 1 Parent-of-origin studies of imprinted genes in plant embryos Cross
Species Arabidopsis thaliana
Ler Col
Zea mays
B73 H99 TX303 W22 Mo17 A188 Col Ler Col Ler Col Cvi Ler Cvi Col Cvi B73 Mo17 Nip 93-11 SR1 Hamayan
Arabidopsis Arabidopsis Arabidopsis Arabidopsis Arabidopsis
thaliana thaliana thaliana thaliana thaliana
Zea mays Oriza sativa Nicotiana tabacum
MEGs (n) 11
1
1
0
17 37 14 15 77
1 1 1 1 45
29 1 1
9 0 4
What is the functional importance of imprinting? Mutation of genes involved in the mechanism of imprinting has strong phenotypic effects on seeds, indicating that imprinting is indeed globally important for development (reviewed in [18]). However, there are only two examples for the importance of imprinting per se in plants. The Maternally expressed gene 1 (Meg1) of maize controls nutrient transfer from the endosperm to embryo; changes in dosage of Meg1 affect kernel size by changing nutrient allocation to the embryo [48]. The PEG ADMETOS (ADM) of Arabidopsis has been shown to be responsible for seed abortion resulting from crosses that produce unbalanced endosperms (i.e. ratios other than 2m:1p) [49]. This seed abortion phenotype is dosage sensitive, indicating a functional role for imprinting of ADM. Whether imprinting of the hundreds of other genes in the endosperm or embryo has functional importance remains to be seen. One possibility is that imprinted expression of transposable elements (TEs) in the endosperm could result in production of siRNAs that non-cell autonomously silence TE expression in the embryo [50,51]. Indeed, siRNAs produced by the endosperm are derived from the maternal genome [52,53]. Since the endosperm does not contribute to the adult plant or the germline, this would be an effective way of silencing TE expression in the embryo, similar to mechanisms proposed for TE silencing in the gametes by siRNA production in companion cells [54,55]. TEs are often targets of the deglycosylase DEMETER (DME), leading to differentially methylated regions (DMRs) that cause imprinted expression of the TEs in the endosperm. Genes that are located near these DMRs can show imprinted expression, an indirect effect of imprinted expression of nearby TEs [55]. Thus, in at least some cases, imprinted gene expression could be a secondary effect of transcriptional unmasking of endosperm TEs that participate in embryo TE silencing. Current Opinion in Plant Biology 2015, 27:29–35
PEGs (n)
Verified?
Reference
RT-PCR & restriction digest; RT-PCR & sequencing; GUS markers RT-PCR & restriction digest
[47]
Considered false positives Considered false positives Not verified Not verified 3 MEGs and 2 PEGs by RT-PCR & restriction digest Not verified RT-PCR seq RT-PCR & restriction digest
[19] [20] [27] [27] [11]
[46]
[23] [21] [10]
Conclusions An important goal for future research on parent-oforigin dependent gene expression in plant embryogenesis is to clearly differentiate between ZGA and imprinting. Since there is considerable evidence that ZGA occurs in a parent-of-origin specific manner in plants [7,9,10,15], ZGA could be defined as a transient form of imprinting. Despite this, the limited research done so far points both to phenomena and mechanisms that differentiate between the two processes (Figure 1). In terms of phenomena, ZGA is defined as a genomewide process affecting thousands of genes, while imprinting need affect only a handful of genes (as in the embryo), or hundreds of genes (as in the endosperm) [47]. It will be necessary to perform additional experiments to demonstrate that ZGA in Arabidopsis does indeed occur in a parent-of-origin specific manner at the genome level, ideally using hybrid combinations demonstrated by Del Toro-De Leo´n et al. to show paternal allele activation patterns similar to isogenic Col [15]. Phylogenetic depth in studies of ZGA could also indicate if delayed paternal (or zygotic) genome activation is widespread in the plant kingdom. Thus far, the majority of parent-of-origin studies in plants have been conducted with Arabidopsis or maize. Studies in other monocots, basal angiosperms such as Amborella trichopoda, or gymnosperms would be useful in determining if delayed paternal genome activation is a conserved or derived trait [56]. It would also be interesting to look at post-fertilization transcriptional activation in a bryophyte such as Physcomitrella patens or Marchantia polymorpha. The study of plants with a transient sporophytic phase might indicate whether ZGA results from genome reprogramming after fertilization, or from a developmental requirement for early maternal control of pattern formation for the formation of complex sporophyte body plants. www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 33
For imprinting, it is important to have greater resolution on the timing of this process in the embryo. Both studies that examined imprinting in Arabidopsis embryos assayed gene expression only in the first few days of embryogenesis, when ZGA also occurs, and making it formally impossible to distinguish between ZGA and imprinting [11,47]. Determining allele-specific expression patterns at 6 dap and after could easily differentiate between ZGA and imprinting, as ZGA is essentially finished for most genes at 5 dap [9,15]. Indeed, allele-specific profiling to determine imprinting in the endosperm has also typically been performed at only a single timepoint [19,20,23,27]. By profiling whole seed at 0, 3, and 5 dap, as well as dissected endosperm at 7, 10, and 15 dap, Xin et al. [26] were able to achieve the highest temporal resolution of any study of imprinting. Their sampling of early timepoints suggested that the paternal genome in the endosperm undergoes gradual transcriptional activation, similar to that seen in embryos of Arabidopsis. ZGA and imprinting can also be distinguished at the mechanistic level. Paternal allele activation during ZGA was shown to be repressed by H3K9 methylation, non-CG DNA methylation, and the RdDM pathway, but not to be affected by CG DNA methylation or PRC2 [9] By contrast, CG DNA methylation and PRC2 were correlated with imprinting of certain genes [46,47] (Figure 1). Further mechanistic studies will be required to better understand how parent-of-origin specific marks of silencing or activation are deposited in the gametes, and acted upon in the zygote. A clearer understanding of ZGA and imprinting will allow us to better distinguish between the two processes, and to recognize how they interact to regulate gene expression in early embryogenesis.
Acknowledgements Thanks to Wolfgang Lukowitz and Gerardo Del Toro De Leo´n for comments on the manuscript. Research on zygotic genome activation in the authors’ laboratory is supported by CONACyT Ciencia Ba´sica No. 237480, and by CINVESTAV institutional funds. C.S.G. is a DuPont Young Professor.
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
Baroux C, Autran D, Gillmor CS, Grimanelli D, Grossniklaus U: The maternal to zygotic transition in animals and plants. Cold Spring Harb Symp Quant Biol 2008, 73:89-100.
2.
Lee MT, Bonneau AR, Giraldez AJ: Zygotic genome activation during the maternal-to-zygotic transition. Annu Rev Cell Dev Biol 2014, 30:581-613.
3.
Aoki F, Worrad DM, Schultz RM: Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev Biol 1997, 181: 296-307.
www.sciencedirect.com
4.
Hamatani T, Carter MG, Sharov AA, Ko MSH: Dynamics of global gene expression changes during mouse preimplantation development. Dev Cell 2004, 6:117-131.
5.
Tadros W, Lipschitz HD: The maternal-to-zygotic transition: a play in two acts. Development 2009, 136:3033-3042.
6.
Harvey SA, Sealy I, Kettleborough R, Fenyes F, White R, Stemple D, Smith JC: Identification of the zebrafish maternal and paternal transcriptomes. Development 2013, 140:27032710.
7.
Vielle-Calzada JP, Baskar R, Grossniklaus U: Delayed activation of the paternal genome during early seed development. Nature 2000, 404:91-94.
8.
Meyer S, Scholten S: Equivalent parental contribution to early plant zygotic development. Curr Biol 2007, 17:1686-1691.
Autran D, Baroux C, Raissig MT, Lenormand T, Wittig M, Grob S, Steimer A, Barann M, Klostermeier UC, Leblanc O, VielleCalzada J-P, Rosenstiel P, Grimanelli D, Grossniklaus U: Maternal epigenetic pathways control parental contributions to Arabidopsis early embryogenesis. Cell 2011, 145:707-719. A tour-de-force study characterizing the phenomena and mechanisms of paternal allele activation at the individual gene and genome level. The paternal genome was shown to be gradually activated after fertilization, resulting in early maternal bias (but with limited paternal contribution from most genes). Maternally encoded DNA and histone methylation were shown to repress paternal transcription, whereas maternal activity of the histone chaperone complex CAF1 was discovered to promote paternal transcription.
9.
10. Zhang J-E, Luo A, Xin H-P, Zhao J, Li S-S, Qu L-H, Ma L-G, Scholten S, Sun M-X: Genes of both parental origins are differentially involved in early embryogenesis of a tobacco interspecies hybrid. PLoS ONE 2011, 6:e23153. 11. Nodine MD, Bartel DP: Maternal and paternal genomes contribute equally to the transcriptome of early plant embryos. Nature 2012, 482:94-97. This work used Cvi–Col hybrid embryos to uncover approximately 100 potentially embryo-imprinted genes, and pointed the way to discovery of the effects of intraspecific hybridization on transcriptional activation in early embryos. 12. Weijers D, Geldner N, Offringa R, Ju¨rgens G: Seed development: early paternal gene activity in Arabidopsis. Nature 2001, 414:709-710. 13. Lukowitz W, Roeder A, Parmenter D, Somerville C: A MAPKK kinase gene regulates extra-embryonic cell fate in Arabidopsis. Cell 2004, 116:109-119. 14. Ueda M, Zhang Z, Laux T: Transcriptional activation of Arabidopsis axis patterning genes WOX8/9 links zygote polarity to embryo development. Dev Cell 2011, 20:264-270. 15. Del Toro-De Leo´n G, Garcı´a-Aguilar M, Gillmor CS: Non equivalent contributions of maternal and paternal genomes to early plant embryogenesis. Nature 2014, 514:624-627. Using a genetic test for paternal allele function in early embryogenesis, this study demonstrated that paternal allele activity in early embryogenesis varies widely between genes. In addition, an unexpectedly large effect of certain hybrid combinations on early gene activity was demonstrated, providing a plausible explanation for differences in transcriptional activation observed between [9] and [11]. 16. Bayer M, Nawy T, Giglione C, Galli M, Meinnel T, Lukowitz W: Paternal control of embryonic patterning in Arabidopsis thaliana. Science 2009, 323:1485-1488. 17. Costa LM, Marshall E, Tesfaye M, Silverstein KAT, Mori M, Umetsu Y, Otterbach SL, Papareddy R, Dickinson HG, Boutiller K, VandenBosch KA, Ohki S, Gutie´rrez-Marcos J: Central cellderived peptides regulate early embryo patterning in flowering plants. Science 2014, 344:168-172. This work uncovered a family of peptides expressed from the endosperm which promote suspensor elongation through the YODA MAPKKK pathway. This was the first demonstration of a direct effect of the central cell and endosperm on embryo patterning. 18. Gehring M: Genomic imprinting: insights from plants. Annu Rev Genet 2013, 47:187-208. A comprehensive and thoughtful review of imprinting in plants. Current Opinion in Plant Biology 2015, 27:29–35
34 Cell signalling and gene regulation
19. Gehring M, Missirian V, Henikoff S: Genomic analysis of parentof-origin allelic expression in Arabidopsis thaliana seeds. PLoS ONE 2011, 6:e23687.
34. Hsieh T-F, Ibarra CA, Silva P, Zemach A, Eshed-Williams L, Fischer RL, Zilberman D: Genome-wide demethylation of Arabidopsis endosperm. Science 2009, 324:1451-1454.
20. Hsieh T-F, Shin J, Uzawa R, Silva P, Cohen S, Bauer MJ, Hashimoto M, Kirkbride RC, Harada JJ, Zilberman D, Fischer RL: Regulation of imprinted gene expression in Arabidopsis endosperm. Proc Natl Acad Sci USA 2011, 108:1755-1762.
35. Makarevitch I, Eichten SR, Briskine R, Waters AJ, Danilevskaya ON, Meeley RB, Myers CL, Vaughn MW, Springer NM: Genomic distribution of maize facultative heterochromatin marked by trimethylation of H3K27. Plant Cell 2013, 25:780-793.
21. Luo M, Taylor JM, Spriggs A, Zhang H, Wu X, Russell S, Singh M, Koltunow A: A genome-wide survey of imprinted genes in rice seeds reveals imprinting primarily occurs in the endosperm. PLoS Genet 2011, 7:e1002125. 22. McKeown PC, Laouielle-Duprat S, Prins P, Wolff P, Schmid MW et al.: Identification of imprinted genes subject to parent-oforigin specific expression in Arabidopsis thaliana seeds. BMC Plant Biol 2011, 11:113. 23. Waters AJ, Makarevitch I, Eichten SR, Swanson-Wagner RA, Yeh C-T, Xu W, Schnable PS, Vaughn MW, Gehring M, Springer NM: Parent-of-origin effects on gene expression and DNA methylation in the maize endosperm. Plant Cell 2011, 23:4221-4233. 24. Wolff P, Weinhofer I, Seguin J, Roszak P, Beisel C, Donoghue MTA, Spillane C, Nordborg M, Rehmsmeier M, Ko¨hler C: High-resolution analysis of parent-of-origin allelic expression in the Arabidopsis endosperm. PLoS Genet 2011, 7:e1002126. 25. Waters AJ, Bilinski P, Eichten SR, Vaughn MW, Ross-Ibarra J, Gehring M, Springer NM: Comprehensive analysis of imprinted genes in maize reveals allelic variation for imprinting and limited conservation with other species. Proc Natl Acad Sci USA 2013, 110:19639-19644. 26. Xin M, Yan R, Li G, Chen H, Laurie J, Ma C, Wang D, Yao Y, Larkins B, Sun Q, Yadegari R, Wang X, Ni Z: Dynamic expression of imprinted genes associates with maternally controlled nutrient allocation during maize endosperm. Plant Cell 2013, 25:3212-3227. Multiple sampling timepoints were used to study of parent-of-origin contributions to transcription in early seeds and endosperm of maize, capturing gradual activation of paternal alleles in the seed. This study provides the first potential link between post-fertilization genome activation and imprinting. 27. Pignatta D, Erdmann RM, Scheer E, Picard CL, Bell GW, Gehring M: Natural epigenetic polymorphisms lead to intraspecific variation in Arabidopsis gene imprinting. eLife 2014, 3:e03198. This study describes intraspecific variation in imprinting between hybrids of Col, Ler and Cvi. Compared to other ecotypes, Cvi was found to show CG hypomethylation in embryos. 28. Xu W, Dai M, Li F, Liu A: Genomic imprinting, methylation and parent-of-origin effects in reciprocal hybrid endosperm of castor bean. Nucleic Acids Res 2014, 42:6987-6998. 29. Choi Y, Gehring M, Johnson L, Hannon M, Harada JJ, Goldberg RB, Jacobsen SE, Fischer RL: DEMETER, a DNA glycosylase domain protein, is required for endosperm gene imprinting and seed viability in Arabidopsis. Cell 2002, 110: 33-42. 30. Xiao W, Gehring M, Choi Y, Margossian L, Pu H, Harada JJ, Goldberg RB, Pennell RI, Fischer RL: Imprinting of the MEA Polycomb gene is controlled by antagonism between MET1 methyltransferase and DME glycosylase. Dev Cell 2003, 5: 891-901. 31. Gehring M, Huh JH, Hsieh T-F, Penterman J, Choi Y, Harada JJ, Goldberg RB, Fischer RL: DEMETER DNA glycosylase establishes MEDEA polycomb gene self-imprinting by allelespecific demethylation. Cell 2006, 124:495-506. 32. Jullien PE, Mosquna A, Ingouff M, Sakata T, Ohad N, Berger F: Retinoblastoma and its binding partner MSI1 control imprinting in Arabidopsis. PLoS Biol 2008, 6:e194. 33. Gehring M, Bubb KL, Henikoff S: Extensive demethylation of repetitive elements during seed development underlies gene imprinting. Science 2009, 324:1447-1451. Current Opinion in Plant Biology 2015, 27:29–35
36. Zhang M, Xie S, Dong X, Zhao X, Zeng B, Chen J, Li H, Yang W, Zhao H, Wang G, Chen Z, Sun S, Hauck A, Jin W, Lai J: Genomewide high resolution parental-specific DNA and histone methylation maps uncover patterns of imprinting regulation in maize. Genome Res 2014, 24:167-176. 37. Barlow DP, Bartolomei MS: Genomic imprinting in mammals. Cold Spring Harb Perspect Biol 2014, 6:a018382. 38. Dickinson H, Scholten S: And baby makes three: genomic imprinting in plant embryos. PLoS Genet 2013, 9:e1003981. 39. Tessadori F, van Zanten M, Pavlova P, Clifton R, Pontvianne F, Snoek LB, Millenaar FF, Schulkes RK, van Driel R, Voesenek LACJ, Spillane C, Pikaard CS, Fransz P, Peeters AJM: PHYTOCHROME B and HISTONE DEACETYLASE 6 Control Light-Induced Chromatin Compaction in Arabidopsis thaliana. PLoS Genet 2009, 5:e1000638. 40. Baroux C, Autran D, Raissig MT, Grimanelli D, Grossniklaus U: Parental contributions to the transcriptome of early plant embryos. Curr Opin Genet Dev 2013, 23:72-74. 41. Schmitz RJ, Schultz MD, Urich MA, Nery JR, Pelizzola M, Libiger O, Alix A, McCosh RB, Chen H, Schork NJ, Ecker JR: Patterns of population epigenomic diversity. Nature 2013, 495:193-198. 42. Ingouff M, Rademacher S, Holec S, Soljic’ L, Xin N, Readshaw A, Foo SH, Lahouze B, Sprunck S, Berger F: Zygotic resetting of the HISTONE 3 variant repertoire participates in epigenetic reprogramming in Arabidopsis. Curr Biol 2010, 20:2137-2143. 43. Pillot M, Baroux C, Arteaga-Va´zquez M, Autran D, Leblanc O, Vielle-Calzada J-P, Grossniklaus U, Grimanelli D: Embryo and endosperm inherit distinct chromatin and transcriptional states from the female gametes in Arabidopsis. Plant Cell 2010, 22:307-320. 44. Blaxter M: Development: the maternal-zygotic transition revisited. Curr Biol 2014, 24:R72-R75. 45. Haig D: Kin conflict in seed development: an interdependent but fractious collective. Annu Rev Cell Dev Biol 2013, 29: 189-211. 46. Jahnke S, Scholten S: Epigenetic resetting of a gene imprinted in plant embryos. Curr Biol 2009, 19:1677-1681. 47. Raissig MT, Bemer M, Baroux C, Grossniklaus U: Genomic imprinting in the Arabidopsis embryo is partly regulated by PRC2. PLoS Genet 2013, 9:e1003862. This study discovered 12 genes with imprinted expression in the Arabidopsis embryo, and demonstrated that PRC2 regulates the imprinted expression of a subset of these genes. Together with mee1 of maize, this work provides the only thorough documentation of imprinted gene expression in plant embryos. 48. Costa LM, Yuan J, Rouster J, Paul W, Dickinson H, Gutie´rrez Marcos JF: Maternal control of nutrient allocation in plant seeds by genomic imprinting. Curr Biol 2012, 22:160-165. This work showed that the endosperm imprinted Meg1 gene of maize plays an important role in nutrient allocation from the endosperm to the embryo, and demonstrated that Meg1 is one of only two imprinted genes in plants whose dosage has been shown to be functionally important. 49. Kradolfer D, Wolff P, Jiang H, Siretskiy A, Ko¨hler C: An imprinted gene underlies postzygotic reproductive isolation in Arabidopsis thaliana. Dev Cell 2013, 26:525-535. The imprinted ADM gene of Arabidopsis was shown to be required for growth failure of unbalanced endosperms (maternal to paternal ratios other than 2m:1p). ADM is a PEG whose increased dosage leads to triploid seed arrest. Along with Meg1 of maize, ADM is one of only two imprinted genes of functional importance. www.sciencedirect.com
Parent-of-origin gene expression in plant embryos Garcı´a-Aguilar and Gillmor 35
50. Van Ex F, Jacob Y, Martienssen RA: Multiple roles for small RNAs during plant reproduction. Curr Opin Plant Biol 2011, 14:588-593. 51. Calarco JP, Borges F, Donoghue MTA, Van Ex F, Jullien PE et al.: Reprogramming of DNA methylation in pollen guides epigenetic inheritance via small RNA. Cell 2012, 151:194-205. 52. Mosher RA, Melnyk CW, Kelly KA, Dunn RM, Studholme DJ, Baulcombe DC: Uniparental expression of PolIV-dependent siRNAs in developing endosperm of Arabidopsis. Nature 2009, 460:283-286. 53. Lu J, Zhang C, Baulcombe DC, Chen ZJ: Maternal siRNAs as regulators of parental genome imbalance and gene expression in endosperm of Arabidopsis seeds. Proc Natl Acad Sci USA 2012, 109:5529-5534.
www.sciencedirect.com
54. Slotkin RK, Vaughn M, Borges F, Tanurdzic’ M, Becker JD, Feijo´ JA, Martienssen RA: Epigenetic reprogramming and small RNA silencing of transposable elements in pollen. Cell 2009, 136:461-472. 55. Ibarra CA, Feng X, Schoft VK, Hsieh TF, Uzawa R, Rodrigues JA, Zemach A, Chumak N, Machlicova A, Nishimura T, Rojas D, Fischer RL, Tamaru H, Zilberman D: Active DNA demethylation in plant companion cells reinforces transposon methylation in gametes. Science 2012, 337:1360-1364. 56. Finet C, Timme RE, Delwiche CF, Marle´taz F: Multigene phylogeny of the green lineage reveals the origin and diversification of land plants. Curr Biol 2010, 20: 2217-2222.
Current Opinion in Plant Biology 2015, 27:29–35
