What type of genes bypass reprogramming

Oswald, J. Lane, N. Genesis , 35 , 88 — Rougier, N. Olek, A. Young, L. USA , , — Shi, W. Chen, T. FEBS Lett. Aoki, F. Memili, E. Zygote , 8 , 87 — Jabbari, K. Gene , , — Cowell, I. Chromosoma , , 22 — B Biol. Liu, H. Lepikhov, K. BMC Dev. Erhardt, S. Kourmouli, N. Cell Sci. Arney, K. Lehnertz, B. Wang, H.

Cell , 8 , — Tachibana, M. Dodge, J. Lewis, A. Umlauf, D. Mak, W. Monk, M. Development , 99 , — Howlett, S. Carlson, L. Bestor, T. Howell, C. Cell , , — Sarmento, O. Nowak, S.

Trends Genet. Fujimori, T. Johnson, M. Rossant, J. Chapman, V. Sado, T. Manes, C. Watanabe, D. Patnaik, D. Substrate specificity and kinetic mechanism of mammalian G9a histone H3 methyltransferase. Cell , 69 , — Okano, M. Cell , 99 , — Okamoto, I. McLaren, A. Szabo, P. Hajkova, P. Lee, J. Yamazaki, Y. Morgan, H. Rakyan, V. Davis, T.

Genomics , 58 , 18 — Lucifero, D. Zhang, S. Humpherys, D. Science , , 95 — Eggan, K. Xue, F. Inoue, K. Science , , Mann, M. Bourc'his, D. USA , 99 , — Boiani, M. Bortvin, A. EMBO J. Chung, Y. Gao, S. Tada, M. Byrne, J. Hansis, C. Landsverk, H. EMBO Rep. Kimura, H. Simonsson, S. Stancheva, I. Collas, P. Tamaru, H. Jackson, J. Tariq, M. Fuks, F. Bhattacharya, S.

Ng, H. Duncan, T. Hardeland, U. Nucleic Acids Res. Zhu, B. Lindahl, T. Weiss, A. Cell , 86 , — [Erratum Cell , 95 , ]. Zeng, F. Once a cell is fully differentiated, this state is strikingly stable. Regardless of how different a neuron is from a hepatocyte, most cells retain an intact genome with the full complement of genes that are present at the beginning in the zygote.

This simple concept of profound significance for development had its origin in the work of Spemann. The distinguishing features of cells arise from an orderly selection of genes that are expressed while the rest are switched off. The genetic network that controls developmental decisions is beginning to be defined. The ability to acquire and inherit gene-expression patterns efficiently is also crucial to the individual history of cell differentiation. There are potential mechanisms that can allow differentiated cells to perpetuate the 'molecular memory' of the developmental decisions that created it.

We know that this occurs without alterations or deletion of any DNA sequences, but rather by epigenetic mechanisms, which propagate appropriate patterns of gene expression Fig. These mechanisms involve heritable but potentially reversible modifications of DNA, primarily methylation of CpG cytosine—guanine dinucleotide 1.

The binding of specific protein complexes to DNA also occurs to form stable and heritable chromatin structures that ensure efficient silencing of genes 2 , 3 that are no longer required for determination of cell fate, allowing expression of only those genes that define properties of specific, differentiated cell types. Most cells contain the same set of genes, but their phenotype can vary according to which genes are expressed and repressed.

Alterations in gene-expression patterns, without changes in DNA sequences, are referred to as epigenetic mechanisms. Epigenetic mechanisms make it possible to restore pluripotency to a differentiated cell, and a differentiated cell can also undergo transdifferentiation resulting in a pronounced change in its appearance and function. Mammalian genomes contain an additional layer of epigenetic information referred to as parental 'imprints'.

These imprints are erased and re-initiated normally in the germ line, and passed on to the offspring in which they survive into adulthood. Parental imprints also regulate gene expression and confer functional differences on parental genomes during development. Parental imprints can undergo changes without affecting the fundamental property of pluripotency. Although the mechanisms that perpetuate cell memory are naturally robust and reliable, they can be erased under some circumstances.

The most pronounced manifestation of this erasure occurs when a differentiated somatic nucleus is transplanted back into an oocyte, which results in the restoration of totipotency 4 , 5 , 6 , 7. The reconstituted egg can then progress forward to generate a new organism that is a genetic copy or a clone of the individual nuclear donor.

While not an efficient process, it is remarkable that it occurs at all; importantly, however, this establishes the principle that epigenetic states are reversible. Mammalian genomes have an additional layer of epigenetic information referred to as genomic imprints, so called because they carry a molecular memory of their parental origin that is acquired in the germ line Fig.

All our genomes therefore contain these distinct maternal and paternal 'imprints' that are inherited after fertilization by embryos and endure thereafter into adulthood 8 , 9. These modifications, which are recognized as differential methylation of specific DNA sequences in sperm and oocytes, regulate expression of imprinted genes, which confer functional differences between parental genomes during development.

Thus parental genomes exhibit an epigenetic asymmetry at fertilization, which persists throughout life. Furthermore, while the overall epigenetic state of the genome changes markedly during development and differentiation of cells, the parental imprints remain relatively stable.

Reprogramming of genomes during imprinting in the germ line requires a stepwise cycle of erasure and re-initiation of imprints. A less well explored but a highly significant consequence of genomic imprinting is that the oocyte cytoplasmic factors have apparently evolved and acquired complex properties in mammals that are required to enhance and maintain the epigenetic asymmetry between parental genomes in the zygote refs 8 , 10 — 13 , and K.

Arney et al. These factors could have important consequences for reprogramming of a somatic nucleus to totipotency when transplanted into the oocyte. One key objective in this field is to gain a detailed knowledge of the mechanisms involved in the erasure of existing epigenetic states and establishment of new modifications for totipotency and during imprinting.

These studies will allow us to assess more precisely events associated with reprogramming of somatic nuclei to a pluripotent or a totipotent state. The analysis of epigenetic mechanisms involved is also crucial for our ability to manipulate pluripotent stem cells and for the derivation of a range of differentiated cell types from pluripotent embryonic stem cells.

One of the main consequences of genomic imprinting and epigenetic asymmetry is that, whereas oocytes are potentially totipotent in many organisms, this is not so in mammals. This is because the maternal genome is epigenetically modified in the germ line to contain only the maternal 'imprints', which will normally result in the repression of certain maternally inherited imprinted genes.

A paternal genome is essential to 'rescue' the oocyte, as the maternal genes are imprinted reciprocally to paternal imprints 14 , So both parental genomes are needed for normal development — the paternal genome is relatively more important for development of the extraembryonic tissues, such as the trophectoderm, whereas the maternal genome apparently has a greater influence on development of the embryo proper.

So far, about 45 imprinted genes have been identified in mice and humans 8 , 9 , Imprinted genes may regulate some of the crucial aspects of mammalian physiology associated with reproduction, placentation, energy homeostasis, lactation and behaviour 16 , 17 , For example, Igf2 , which encodes a fetal insulin-like growth factor 2, is repressed in the maternal genome and active only in the paternal genome Some of the anomalies encountered in cloned embryos suggest disruption of imprinted gene expression Imprinted genes are often organized in clusters, sometimes in the megabase-range chromosomal regions containing key control elements — the differentially methylated regions DMRs 8 , 9 , DMRs are CpG rich and subject to epigenetic modifications.

These imprinting control regions are often complex with multiple functions acting to repress genes when methylated, or serving as boundary elements when unmethylated the boundary element 21 , 22 indirectly affects expression of neighbouring genes.

Some DMRs also function as silencer elements when unmethylated 23 , a function that is apparently abolished when the DMR is methylated. In other instances, a DMR is associated with the expression of an antisense transcript whose expression in turn ensures repression of the upstream gene The result, in all cases, is to ensure monoallelic expression of imprinted genes.

Such complex organization of imprinted genes means that any disruption of such clusters through chromosomal translocations or epigenetic mutations can cause anomalous expression of many genes within the cluster, and this accounts for some of the diseases associated with growth, neurogenetic disorders and diabetes 8 , 9 , Genomic imprinting probably accompanied mammalian evolution 25 , and the evolution of placentation and viviparity in mammals resulted in significant changes in early development.

One striking aspect is the emergence of trophectoderm cells, which are essential for blastocyst implantation and as such are the first differentiated cell type to form during development.

Also within the blastocyst are the pluripotent epiblast cells, the precursor of embryonic stem ES cells Fig. Gastrulation commences relatively late after implantation in response to signals that emanate from the trophectoderm and primary endoderm cells.

Another feature of mammalian development is that the oocytes are relatively small and lack the cytoplasmic determinants of development commonly encountered in other organisms, including those for the germ cell lineage.

As a result, there is relatively early activation of the embryonic genome, which in the mouse occurs at the two-cell stage Fig. Development commences with the totipotent zygote following fertilization with reciprocally 'imprinted' parental genomes. Imprinting confers developmental asymmetry on parental genomes, so that both are essential for normal development. The mouse embryonic genome is activated at the two-cell stage, and is followed by development of a blastocyst with an inner cell mass and trophectoderm cells.

The epiblast cells within the inner cell mass are pluripotent, and give rise to all the somatic cells in the fetus, as well as germ cells. PGCs retain pluripotency as shown by the ability to generate EG cells, which may lack parental imprints. The initiation of imprinting is confined to the germ line, first with the erasure of existing imprints in primordial germ cells PGCs 26 , 27 , 28 , 29 , followed by the initiation of a new set of imprints in the male and female germ lines.

It should be noted that whereas methylation of DMRs for some genes results in their repression, in other instances for example, the Igf2r gene , methylation is essential for gene activation 8 , 9 , In most instances, methylation of DMRs occurs predominantly in the female germ line. Nuclear transplantation studies between developing oocytes have shown that the maternal imprints are acquired at the time when oocytes resume growth from a quiescent state prior to ovulation 30 , 31 , At this time, the X chromosome also acquires an imprint for the non-random inactivation of the paternal X chromosome in trophectoderm and primary endoderm cells The precise mechanism by which de novo methylation of DMRs occurs is not yet known, but it should involve some germline-specific factors acting in conjunction with DNA methyltransferase enzymes.

Two important properties of PGCs are that they retain pluripotency reviewed by Donovan and Gearhart, pages 92—97 , and they are endowed with an exceptional capacity for epigenetic modifications of the genome. A unique feature is their ability to erase parental imprints, which shows that epigenetic modifications associated with imprinting can occur independently of the genomic status concerning pluripotency.

Elaborate transcriptional regulation is generally associated with the founding of germ cells to prevent them from acquiring a somatic cell fate. In mice, germ cells originate from the proximal epiblast cells of the embryonic day 6. It is the proximal location of the epiblast cells that is critical for germ cell fate, rather than any intrinsic properties of these cells The specification of germ cell lineage depends on signals, such as bone morphogenetic protein BMP -4 and BMP8b refs 36 , 37 , originating from the extraembryonic ectoderm in contact with the proximal epiblast.

PGCs can be generated in vitro by combining epiblast with extraembryonic tissues 38 and, in principle, it should also be possible to generate PGCs from ES or embryonic germ EG cells.

Oct4 , a gene expressed in all totipotent and pluripotent cells in mammals 39 , 40 , has an enhancer that is required for its expression predominantly in germ cells There are, however, additional genes involved in the specification of the germ cell fate, and these are also likely to be crucial for pluripotency in general, including pluripotent stem cells M. Saitou, S. Barton and M. Nevertheless, PGCs cannot participate in early development if re-introduced into blastocysts, unlike ES cells that can differentiate into a wide variety of somatic cells and germ cells.

It may be that PGCs do not respond to signalling molecules, or that they are transcriptionally repressed. That PGCs are pluripotent is illustrated by the ability to derive EG cells from them 41 , 42 , which share many common properties with ES cells.

When PGCs begin migration into the genital ridge at E9. Hajkova et al. Pronounced epigenetic modifications commence with the entry of PGCs into the genital ridge Fig. There is rapid and possibly active genome-wide demethylation in both male and female PGCs, resulting in the erasure of imprints refs 28 , 30 , 32 , 44 , and P.

This is accompanied by reactivation of the inactive X chromosome in female germ cells The same mechanism also erases any aberrant epigenetic modifications, so preventing the inheritance of epimutations, which consequently occurs very rarely The precise mechanism responsible for epigenetic erasure and demethylation in PGCs is as yet unclear. Primordial germ cells in E9. Upon the entry of PGCs into the genital ridge at E These epigenetic modifications include genome-wide demethylation, reactivation of the inactive X chromosome and erasure of imprints.

By E New sex-specific imprints are introduced later during gametogenesis, and are detected in mature sperm and oocytes. Concerning the timing of epigenetic modifications in PGCs, there are at least two possibilities.

First, these epigenetic modifications may be triggered in PGCs by a signal from somatic cells when they enter the genital ridge at E Alternatively, the erasure of imprints might occur at a specific time and be regulated by a developmental clock. Whatever determines the timing of these events, PGCs by E The erasure of imprints is also observed in EG cells Here it occurs precociously, being found in EG cells derived from PGCs before their entry into the genital ridge, and could be due to culture of PGCs in vitro.

It is important to note that ES cells do not show the same property for the erasure of imprints see below. When the imprint-free EG cells are introduced into blastocysts to generate chimaeras, they can cause developmental anomalies, such as aberrant growth and skeletal abnormalities The initiation of new imprints occurs subsequently during gametogenesis, and primarily during oogenesis.

Other forms of genomic modifications probably occur in PGCs, including restoration of telomeres to their optimum size and DNA repair generally In single-cell organisms such as yeast, these functions reside within all cells, but in multicellular organisms, some of these functions occur either optimally or exclusively in the germ line. For example, the levels of telomerase are low in somatic cells, and this contributes to their ageing and senescence.

Further work is needed to determine the capacity of the mammalian germ line to deal with DNA damage. Following imprinting in the germ line, the parental genomes exhibit epigenetic asymmetry at fertilization. These epigenetic differences are both maintained and enhanced in the zygote. During 0—5 hours post fertilization h. Thereafter, the pronuclear membrane forms which can regulate access of oocyte cytoplasmic factors to the parental genomes Fig.

During the initial 0—5 h. The paternal genome undergoes marked demethylation while the maternal genome, which contains most of the methylation marks associated with imprints, undergoes further de novo methylation 11 , 12 , Species where imprinting is unknown do not show such differential methylation of parental genomes during early development. Demethylation of the paternal genome may be essential to make it compatible for early activation of the embryonic genome.

Alternatively, demethylation of the paternal genome may have accompanied evolution of developmental asymmetry between parental genomes The TFs with z-score between and are highlighted in gray because Figure 2B suggests these TFs predictivity may be prone to extra noise induced by the data discretization.

C Inset of ESC positive predictivity and gene expression. Zfp42 Rex1 [40] and Nr0b1 Dax1 [41] are pluripotency markers that are not necessary to overexpress for reprogramming, while combinations of the remaining labeled TFs have been successfully used in reprogramming protocols [8]. D Heart induced cardiomyocytes, iCM [3]. E Liver induced hepatocytes, iHep. There are two published protocols. F Thyroid [7]. H Neurons induced neuron, iN [2]. The reprogramming protocol used a combination of factors that were known to be important to ether mature neurons Myt1l or NPCs Pou3f2 , Ascl1.

While ESC have been studied in the most detail, recent experiments have reprogrammed aka direct conversion to other cell fates such as cardiomyocytes [3] Figure 3D , liver [4] , [5] Figure 3E , and thyroid [7] Figure 3F. Once again we have explicitly labeled the TFs that have been successfully used for direct conversion.

Notice that all of these TFs except Mef2c are highly predictive and highly expressed. Note that p19Arf [5] used in the direct conversion to liver was not differentially expressed in our microarrays and therefore was not included in our model.

We also examined TFs used in direct conversion to neural lineages. As discussed in [2] , these TFs were chosen because they were known to be important in either neurons or neural progenitor cells NPC.

Consistent with their experimental design, we find that Myt1l is highly predictive for mature neurons, while the remaining TFs Pou3f2 , Ascl1 are predictive for NPCs. While it is not possible to perform statistical tests to test our examples due to the scarcity of reprogramming protocols, we performed a simple numerical exercise to gauge the predictive power of our model.

The four Yamanaka factors are all in the top 50 when ranked by their reprogramming score for ESCs where the reprogramming score of a TF is defined as the product of the expression and predictivity scores of a TF. We randomly permuted TF labels and asked how often all four Yamanaka factors remained in the top For a million independent permutations, this occurred only once, confirming that our model is capturing many essential aspects of cellular reprogramming. Our results suggest that epigenetic landscapes may be useful for rationally-designing reprogramming protocols to novel cell fates.

To this end, we have used our model to identify candidate TFs for reprogramming, see File S5 for the top 50 candidates for overexpression for all cell fates and File S6 for top 50 candidates for knockouts for all cell fates. Despite decades of biological innovation, the large number of genes and their complex interactions has prevented the quantitative modeling of a global epigenetic landscape.

To meet this challenge, we have developed a new quantitative framework of cellular identity to directly model the global, high-dimensional epigenetic landscape. Using whole genome expression data, we constructed an epigenetic landscape based on techniques from spin glass physics and neural networks. Our landscape only depends on the experimentally determined gene expression of natural cell fates. Yet, it explains the existence of spurious cell fates known as partially-reprogrammed cells and can reproduce known reprogramming protocols to embryonic stem cells, heart, liver, thyroid, neural progenitor cells, and neurons.

More importantly, our model can be used to identify candidate transcription factors for reprogramming to novel cell fates. An interesting question is if spurious attractors are ubiquitous throughout the landscape, why does standard development not produce partially reprogrammed cells?

The key is the difference in the dynamics. In cellular reprogramming, the starting cell fate is forced to express a small number of TF and this leads to a stochastic conversion to the desired cell fate Figure 1A.

During this stochastic exploration of the landscape, there is only a weak bias towards the final state, so it is easy for the cells to get trapped in a metastable state. However, during standard development, the external signals actively reshape the landscape and open up low energy valleys between cell fates Figure 1C. This strong bias towards the final cell state results in a deterministic switch during which the spurious attractors are only a small road bump on the path to the final cell state.

Therefore, it is not a surprise that partially reprogrammed cells are only found during cellular reprogramming and not during standard development. Epigenetic landscapes can also be used to identify important, or predictive, TFs for cell fates.

The predictivity of a TF for a cell fate generalizes the idea of specificity. A TF is specific to a cell fate if it is expressed only on in a small subset of cell fates. In contrast with specificity, predictivity weighs the global correlations amongst cell fates when assessing the importance of a TF for a cell fate. Thus, the predictivity not only picks out important specific TFs, but also TFs that are lineage markers. For example, Brachyury T [42] is a general marker of mesodermal lineages.

Since it is highly expressed in large a number of cell fates, it is not specific to any given cell fate. However, it is predictive because its expression is a strong indicator that a given cell fate is a mesodermal lineage. The concept of predictivity also yields new insights into the Yamanaka protocol. In contrast, the role of the other two TFs, Klf4 and Myc , was not well understood [43]. It was quickly shown that Myc was was not essential to reprogramming Oct4 , Sox2 , and Klf4 can reprogram alone , but nonetheless enhanced the efficacy of reprogramming [44].

The importance of Klf4 was surprising given that it is neither highly expressed nor specific for ESC. For this reason, our model actually explains why Klf4 is a prime candidate for reprogramming to ESCs. We make several experimentally verifiable predictions. First, our model predicts the partially reprogrammed cells should be hybrids of existing natural cell fates.

As more partially reprogrammed cells are studied, if they are found to either have high projection on only one cell fate for one or no projections on any cell fates for all , this would call into question whether partially reprogrammed cells are truly the spurious attractors of an attractor neural network. Second, our model can be used to identify important, or predictive, TFs for cell fates.

TFs with large positive negative predictivity should be positive negative markers for a cell fate. Additionally, for cellular reprogramming we predict that TFs with large positive negative predictivity and expression could be over expressed knocked out to reprogram to a desired cell fate.

Therefore, our model has several predictions that can be tested against future experimental progress in the field. Our model has several limitations. First, a generic limitation for any method relying on microarrays to define gene expression is that one cannot distinguish between direct, causal, interactions and indirect, correlative, interactions.

Therefore, predictivity can establish the importance of a gene, but further experiments are needed to determine if the predictive gene is the controller of the cell type or just a passive indicator of a cell type. Second, it fails to accurately capture the dynamics of reprogramming. Simulations of reprogramming with known protocols, such as the Yamanaka protocol, lead to rates of reprogramming that are comparable to the rates from a reprogramming simulation with a randomly selected protocol.

This is likely due to the fact that cell fates are extremely stable and hence reprogramming is extremely rare. Third, our model does not directly explain the importance of the non-specific transcription factor Myc. Many protocols use Myc [8] , but it can be replaced with no deleterious effect by short hairpin RNAs shRNAs [45] , or dropped completely from protocols at the expense of speed and less efficient reprogramming [44]. This suggests that Myc may have an alternative role and instead of being a biasing field, , it may instead raise the effective noise of the system i.

Another limitation is that based on the currently available experimental data, our landscape construction cannot definitively be distinguished from alternative constructions. For example, the interaction network could be constructed by such that it does not weigh each cell fate equally as is currently done.

This would have the effect of changing the relative stability of cell fates. Therefore, in the absence of more experimental data, our landscape and a weighted landscape cannot be distinguished.

This method has been used to model firing neurons [46] , protein configurations [47] , [48] , and antibody diversity [49]. The Maximum Entropy approach takes as input large samples of biological data and a set of constraints and outputs a landscape that maximizes the entropy.

While Maximum Entropy models can be used to infer landscapes with basins of attraction [50] , it can quickly become a computationally challenging problem. Our approach differs from Maximum Entropy models in the following way. Since our goal is to model a landscape with basins of attractions, we make the ansatz that the landscape can be described by a Hopfield neural network.

Then we insert real biological data, , to construct the landscape exactly. Our method requires no computational inference of parameters. There are several natural extensions of the model discussed in this paper.

The landscape could be constructed with additional biological input such as other genes, microRNAs, or histone modification data. This opens up possibilities of improving upon the high reprogramming rates achieved by overexpressing microRNAs [51] or synthetic mRNAs [52].

Another attractive element of the framework presented here is that it allows for a quantitative analysis of whole genome-wide expression states see Table 2. This is likely to yield a more accurate classification of reprogrammed cells. Finally, directed differentiation protocols [53] attempt to mimic standard development in vitro and have proven to have high efficiency and fidelity. Future work will try to use our landscape to predict the necessary signaling factors for rationally designing more efficient directed differentiation protocols.

Overall, epigenetic landscapes provide a unifying framework for cell identity, reprogramming, and directed differentiation, and our results suggest these landscapes can provide crucial insight into the molecular circuitry and dynamics that gives rise to cell fate.

Here we present the details of the dataset. All data used in this paper are available in the online Supplementary Information and is organized as follows:. An older version of this manuscript, Arxiv v3 [54] , has additional microarrays available that are unused in this version of the text. All microarrays used in this paper were taken from the public databases ArrayExpress www. See File S1: Microarray Sources for details on where to obtain raw, pre-normalized and pre-averaged data.

There are two datasets, the natural cell fates and the partially reprogrammed cells. There is limited data on partially reprogrammed cells so we used microarrays from Affymetrix GeneChip Mouse Genome 2. The raw microarray data was converted to an expression level as follows. Microarray probe-to-gene map was created with Bioconductor 2. We did additional processing of this output for two reasons. First, we need to compare microarrays from multiple platforms, but the standard RMA output can vary significantly from platform to platform.

Second, since gene expression is a set of positive definite numbers, the minimal assumption model of gene expression is a log-normal distribution. Therefore, to make robust comparisons across platforms, we used order statistics [55]. The RMA output was converted to a rank order. Next, we want to convert this rank order to the z-score of a log-normal distribution. We converting the rank to a percentile for genes, divide by , and then this percentile into a normal z-score.

For later mathematical convenience, we used a biased estimator normalize by not since then the Euclidean norm of each microarray gene expression is. At this point, the natural dataset consisted of microarrays with genes. Since we were interested in cellular identity, only transcription factors, transcription factor co-factors, or chromatin remodeling genes were kept for short hand, referred to as transcription factors TF throughout the text [56] , leaving TFs.

As explained in the main text, since continuous sigmoidal input attractor neural networks and discrete attractor neural networks are known to have the same stable fixed points [57] , we used the binarized gene expression.

We binarized the gene expression by setting a positive z-score to and a negative z-score to. While this was mainly done for mathematical convenience, this is potentially biologically justified. Histone modifications HM either leave chromatin in an open, accessible configuration or a closed, inaccessible state [35]. Consequently, we used the global HM data for these three cell fates and compared them to microarray TF expression levels.

This allowed us to create a conditional probability distribution of each HM for a given TF expression level Figure 2B. This shows that our mathematical assumption is justified by the HM data. After the binarization of TF expression, all TFs that were not differentially expressed across cell fates i. The binarized TF expression for the 63 cell fates was found by first binarizing all microarrays and then taking the majority vote for each cell state with ties broken by averaging the continuous data.

The final result was the binary expression state for 63 cell fates. The same procedure was used to convert raw microarray data to z-score expression. Several self-consistency checks were performed on the data. First, the correlation matrix explained in main text and below was calculated for the original continuous data and for the binarized data Figure S1. Pericentromeric repeats These elements show latent transcriptional activity, which is repressed by DNA methylation to achieve proper chromosome alignment and segregation Fig.

X-chromosome inactivation DNA methylation mediates gene dosage control via inactivation of the second X chromosome in females Fig. Imprinted genes DNA methylation is vital to the controlled expression of imprinted genes Fig. Methylation maintenance A requisite for a maintenance mechanism is a high affinity for hemimethylated CpGs i. De novo methylation Dnmt1 deletion in mouse embryonic stem cells mESCs causes dramatic DNA hypomethylation but not a complete loss, predicting the presence of further enzymes with methyltransferase activity.

DNA demethylation In contrast to the enzymatically controlled, straightforward methylating mechanism, a direct DNA demethylase capable of breaking carbon—carbon bonds has not yet been identified.

Passive, replication-dependent dilution Loss of 5mC in mitotic cells can be achieved by down-regulation or exclusion of the DNA methylation maintenance machinery DNMT1 or its recruitment factors, such as NP95 from the nucleus Fig.

Figure 2. Figure 3. Figure 4. The fate of paternal 5hmC The bulk of 5mCs in the paternal genome is hydroxylated in the late zygote, but only a few regions were shown to completely revert to unmodified cytosine before the first cleavage division. Distinction of parental pronuclei How is TET3 targeted to the paternal genome without affecting the maternal genome? Is active DNA demethylation of the paternal genome required? Passive DNA demethylation and maintenance of parental imprints The maternal genome is, at least globally, resistant to hydroxylation by TET3 yet nonetheless loses the bulk of its oocyte-specific DNA methylation pattern by replication-mediated 5mC dilution during preimplantation development Fig.

Noncanonical targeting of DNMT1 to imprinted regions in preimplantation embryos DNMT1 is required to maintain imprints, yet, at the same time, its nuclear protein levels are drastically reduced to allow global demethylation.

Previous Section. Active demethylation of paternal genome in mammalian zygotes. J Reprod Dev 55 : — CrossRef Medline Google Scholar. Human epigenome data reveal increased CpG methylation in alternatively spliced sites and putative exonic splicing enhancers. DNA Cell Biol 30 : — DNA methylation of distal regulatory sites characterizes dysregulation of cancer genes.

Genome Biol 14 : R PLoS Genet 8 : e Genomic imprinting: employing and avoiding epigenetic processes. Genes Dev 23 : — Cloning of a mammalian DNA methyltransferase. The DNA methyltransferases of mammals.

Hum Mol Genet 9 : — Targets and dynamics of promoter DNA methylation during early mouse development. Nature Publishing Group 42 : — Google Scholar. Science : — Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L. Nature : 96 — Dnmt3L and the establishment of maternal genomic imprints. Safeguarding parental identity: Dnmt1 maintains imprints during epigenetic reprogramming in early embryogenesis. Genes Dev 22 : — Sp1 elements protect a CpG island from de novo methylation.

Nature : — Properties and localization of DNA methyltransferase in preimplantation mouse embryos: implications for genomic imprinting. Genes Dev 6 : — Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals. Cell : — Proc Natl Acad Sci 99 : — Structure and function of eukaryotic DNA methyltransferases. Curr Top Dev Biol 60 : 55 — X chromosome activity in mouse XX primordial germ cells. PLoS Genet 4 : e Preimplantation expression of the somatic form of Dnmt1 suggests a role in the inheritance of genomic imprints.

BMC Dev Biol 8 : 9. Primate CpG islands are maintained by heterogeneous evolutionary regimes involving minimal selection. Cell : 67 — The H19 methylation imprint is erased and re-established differentially on the parental alleles during male germ cell development. Tet1 is dispensable for maintaining pluripotency and its loss is compatible with embryonic and postnatal development. Cell Stem Cell 9 : — Combined deficiency of Tet1 and Tet2 causes epigenetic abnormalities but is compatible with postnatal development.

Dev Cell 24 : — CpG islands and the regulation of transcription. Genes Dev 25 : — In vivo stabilization of the Dnmt1 cytosine-5 - methyltransferase protein. Proc Natl Acad Sci : — Global mapping of DNA methylation in mouse promoters reveals epigenetic reprogramming of pluripotency genes.

Epigenetic reprogramming in plant and animal development. Genomic imprinting: the emergence of an epigenetic paradigm. Nat Rev Genet 12 : — DNA methylation pattern in human zygotes and developing embryos.

Reproduction : — A short DNA methyltransferase isoform restores methylation in vivo. J Biol Chem : — Hum Mol Genet 18 : — The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes. Epigenetics: erase for a new start. Germline DNA demethylation dynamics and imprint erasure through 5-hydroxymethylcytosine. Hajkova P Hajkova P. Epigenetic reprogramming—taking a lesson from the embryo. Curr Opin Cell Biol 22 : — Epigenetic reprogramming in mouse primordial germ cells.

Mech Dev : 15 — Chromatin dynamics during epigenetic reprogramming in the mouse germ line. Genome-wide reprogramming in the mouse germ line entails the base excision repair pathway. Science : 78 — Recognition and potential mechanisms for replication and erasure of cytosine hydroxymethylation.

Nucleic Acids Res 40 : — Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice. Development : — Medline Web of Science Google Scholar. Germ cell specification in mice. Gene body-specific methylation on the active X chromosome.

Histone methylation is mechanistically linked to DNA methylation at imprinting control regions in mammals. Maternal and zygotic Dnmt1 are necessary and sufficient for the maintenance of DNA methylation imprints during preimplantation development. High definition profiling of mammalian DNA methylation by array capture and single molecule bisulfite sequencing.

Genome Res 19 : — DNA modification mechanisms and gene activity during development. Genomic imprinting disrupted by a maternal effect mutation in the Dnmt1 gene.

Replication-dependent loss of 5-hydroxymethylcytosine in mouse preimplantation embryos. Science : Generation and replication-dependent dilution of 5fC and 5caC during mouse preimplantation development. Cell Res 21 : — Reprogramming of the paternal genome upon fertilization involves genome-wide oxidation of 5-methylcytosine. Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification.

Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcytosine. KAP1 protein: an enigmatic master regulator of the genome. Hum Mol Genet 17 : — Replication-coupled passive DNA demethylation for the erasure of genome imprints in mice. EMBO J 32 : — Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting.

Role of the Dnmt3 family in de novo methylation of imprinted and repetitive sequences during male germ cell development in the mouse. Hum Mol Genet 16 : — New insights into establishment and maintenance of DNA methylation imprints in mammals. Control elements of Dnmt1 gene are regulated in cell-cycle dependent manner. Nucleic Acids Res 3 : — CrossRef Google Scholar. Contribution of intragenic DNA methylation in mouse gametic DNA methylomes to establish oocyte-specific heritable marks.

High-resolution DNA methylome analysis of primordial germ cells identifies gender-specific reprogramming in mice.

Genome Res 23 : — DNA methylation and methylcytosine oxidation in cell fate decisions. Curr Opin Cell Biol 25 : — Epigenetic transitionsin germ cell development and meiosis. Dev Cell 19 : — Maintenance of genomic methylation patterns during preimplantation development requires the somatic form of DNA methyltransferase 1. Dev Biol : — Complex genome-wide transcription dynamics orchestrated by Blimp1 for the specification of the germ cell lineage in mice.

Initial sequencing and analysis of the human genome. Erasing genomic imprinting memory in mouse clone embryos produced from day De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells.