Parental imprinting studied by allele-specific primer extension after PCR: paternal X chromosome-linked genes are transcribed prior to preferential paternal X chromosome inactivation

Analysis of allele-specific gene expression using a target-oriented tiling microarray assay

Advances in molecular and computational biology in recent years have led to the development or the improvement of methods for analyzing global gene expression. In most of these efforts, it is assumed that alleles of different origins contribute equally to the transcript pool. However, accumulating evidence suggests that many genes are not equally expressed from the paternal and maternal chromosomes. In addition to imprinting, the phenomenon of imbalanced allelic expression is widespread in heterozygous individuals. To distinguish transcript pools derived from different alleles present in the same organism, a number of methods have been developed. Here, we describe an oligonucleotide tiling microarray-based assay for analyzing allele-specific gene expression. Specifically targeting single-nucleotide polymorphisms, this two-step assay offers a high-throughput and multiplex method for detecting and quantifying unequal allelic expression that is readily applicable to many experimental systems.

Next-generation DNA sequencing-based assay for measuring allelic expression imbalance (AEI) of candidate neuropsychiatric disorder genes in human brain

BACKGROUND

Common genetic variants that regulate gene expression are widely suspected to contribute to the etiology and phenotypic variability of complex diseases. Although high-throughput, microarray-based assays have been developed to measure differences in mRNA expression among independent samples, these assays often lack the sensitivity to detect rare mRNAs and the reproducibility to quantify small changes in mRNA expression. By contrast, PCR-based allelic expression imbalance (AEI) assays, which use a "marker" single nucleotide polymorphism (mSNP) in the mRNA to distinguish expression from pairs of genetic alleles in individual samples, have high sensitivity and accuracy, allowing differences in mRNA expression greater than 1.2-fold to be quantified with high reproducibility. In this paper, we describe the use of an efficient PCR/next-generation DNA sequencing-based assay to analyze allele-specific differences in mRNA expression for candidate neuropsychiatric disorder genes in human brain.

RESULTS

Using our assay, we successfully analyzed AEI for 70 candidate neuropsychiatric disorder genes in 52 independent human brain samples. Among these genes, 62/70 (89%) showed AEI ratios greater than 1 ± 0.2 in at least one sample and 8/70 (11%) showed no AEI. Arranging log2AEI ratios in increasing order from negative-to-positive values revealed highly reproducible distributions of log2AEI ratios that are distinct for each gene/marker SNP combination. Mathematical modeling suggests that these log2AEI distributions can provide important clues concerning the number, location and contributions of cis-acting regulatory variants to mRNA expression.

CONCLUSIONS

We have developed a highly sensitive and reproducible method for quantifying AEI of mRNA expressed in human brain. Importantly, this assay allowed quantification of differential mRNA expression for many candidate disease genes entirely missed in previously published microarray-based studies of mRNA expression in human brain. Given the ability of next-generation sequencing technology to generate large numbers of independent sequencing reads, our method should be suitable for analyzing from 100- to 200-candidate genes in 100 samples in a single experiment. We believe that this is the appropriate scale for investigating variation in mRNA expression for defined sets candidate disorder genes, allowing, for example, comprehensive coverage of genes that function within biological pathways implicated in specific disorders. The combination of AEI measurements and mathematical modeling described in this study can assist in identifying SNPs that correlate with mRNA expression. Alleles of these SNPs (individually or as sets) that accurately predict high- or low-mRNA expression should be useful as markers in genetic association studies aimed at linking candidate genes to specific neuropsychiatric disorders.

Cis-acting mutation and duplication: History of molecular evolution in a P450 haplotype responsible for insecticide resistance in Culex quinquefasciatus

A cytochrome P450 gene, Cyp9m10, is more than 200-fold overexpressed in a pyrethroid resistant strain of Culex quinquefasciatus, JPal-per. The haplotype of this strain contains two copies of Cyp9m10 resulted from recent tandem duplication. In this study, we discovered and isolated a Cyp9m10 haplotype closely related to this duplicated Cyp9m10 haplotype from JHB, a strain used for the recent genome project for this mosquito species. The isolated haplotype (JHB-NIID-B haplotype) shared the same insertion of a transposable element upstream of the coding region with JPal-per strain but not duplicated. The JHB-NIID-B haplotype was considered to have diverged from the JPal-per lineage just before the duplication event. Cyp9m10 was moderately overexpressed in larvae with the JHB-NIID-B haplotype. The overexpressions in JHB-NIID-B and JPal-per haplotypes were developmentally regulated in similar pattern indicating both haplotypes share a common cis-acting mutation responsible for the overexpressions. The isolated moderately overexpressed haplotype conferred resistance, however, its efficacy was relatively small. We hypothesized that the first cis-acting mutation modified the consequence of the subsequent duplication in JPal-per lineage to confer stronger phenotypic effect than that if it occurred before the first cis-acting mutation.

Animal models of middle ear cholesteatoma

Middle ear acquired cholesteatoma is a pathological condition associated with otitis media, which may be associated with temporal bone resorption, otorrhea and hearing loss, and occasionally various other complications. Cholesteatoma is characterized by the enhanced proliferation of epithelial cells with aberrant morphologic characteristics. Unfortunately, our understanding of the mechanism underlying its pathogenesis is limited. To investigate its pathogenesis, different animal models have been used. This paper provides a brief overview of the current status of research in the field of pathogenesis of middle ear acquired cholesteatoma, four types of animal models previously reported on, up-to-date cholesteatoma research using these animal models, our current studies of the local hybrid ear model, and the future prospect of new animal models of middle ear cholesteatoma.

Development of a versatile, target-oriented tiling microarray assay for measuring allele-specific gene expression

In the study of gene expression, it is often desirable to distinguish transcript pools derived from different alleles present in the same organism. We report here an oligonucleotide tiling microarray designed to specifically target 518 single nucleotide polymorphisms (SNPs) between the two sequenced rice (Oryza sativa) subspecies indica and japonica. The tiling array included all 25-mer probes interrogating each SNP by placing the polymorphic site at all 25 possible positions within the probe. Through hybridization to a titration series in which the japonica- and indica-derived cDNA templates were mixed with altering proportions, a regression model was used to screen for diagnostic probe sets for each SNP. Our result indicates that 284 (55%) SNPs have at least one diagnostic probe pair suitable for distinguishing and quantifying the relative abundance of allele-specific transcripts. As a proof-of-concept, we analyzed allele-specific expression in reciprocal indica×japonica F(1) hybrids and detected imbalanced expression at approximately one third of the SNPs. These results were validated by RNA-sequencing and allele-specific real-time PCR experiments. Together, our work demonstrates the utility and advantages of the tiling array method in interrogating large numbers of SNPs for quantifying allele-specific gene expression.

Study of transcriptional effects in Cis at the IFIH1 locus

BACKGROUND

The Thr allele at the non-synonymous single-nucleotide polymorphism (nsSNP) Thr946Ala in the IFIH1 gene confers risk for Type 1 diabetes (T1D). The SNP is embedded in a 236 kb linkage disequilibrium (LD) block that includes four genes: IFIH1, GCA, FAP and KCNH7. The absence of common nsSNPs in the other genes makes the IFIH1 SNP the strongest functional candidate, but it could be merely a marker of association, due to LD with a variant regulating expression levels of IFIH1 or neighboring genes.

METHODOLOGY/PRINCIPAL FINDINGS

We investigated the effect of the T1D-associated variation on mRNA transcript expression of these genes. Heterozygous mRNA from lymphoblastoid cell lines (LCLs), pancreas and thymus was examined by allelic expression imbalance, to detect effects in cis on mRNA expression. Using single-nucleotide primer extension, we found no difference between mRNA transcripts in 9 LCLs, 6 pancreas and 13 thymus samples, suggesting that GCA and FAP are not involved. On the other hand, KCNH7 was not expressed at a detectable level in all tissues examined. Moreover, the association of the Thr946Ala SNP with T1D is not due to modulation of IFIH1 expression in organs involved in the disease, pointing to the IFIH1 nsSNP as the causal variant.

CONCLUSIONS/SIGNIFICANCE

The mechanism of the association of the nsSNP with T1D remains to be determined, but does not involve mRNA modulation. It becomes necessary to study differential function of the IFIH1 protein alleles at Thr946Ala to confirm that it is responsible for the disease association.

Difference between random and imprinted X inactivation in common voles

During early development in female mammals, most genes on one of the two X-chromosomes undergo transcriptional silencing. In the extraembryonic lineages of some eutherian species, imprinted X-inactivation of the paternal X-chromosome occurs. In the cells of the embryo proper, the choice of the future inactive X-chromosome is random. We mapped several genes on the X-chromosomes of five common vole species and compared their expression and methylation patterns in somatic and extraembryonic tissues, where random and imprinted X-inactivation occurs, respectively. In extraembryonic tissues, more genes were expressed on the inactive X-chromosome than in somatic tissues. We also found that the methylation status of the X-linked genes was always in accordance with their expression pattern in somatic, but not in extraembryonic tissues. The data provide new evidence that imprinted X-inactivation is less complete and/or stable than the random form and DNA methylation contributes less to its maintenance.

Pathogenesis of middle ear cholesteatoma: a new model of experimentally induced cholesteatoma in Mongolian gerbils

Middle ear cholesteatoma is characterized by enhanced proliferation of epithelial cells with aberrant morphological characteristics. To investigate the origin of the cholesteatoma cells, we analyzed spontaneously occurring cholesteatomas associated with a new transplantation model in Mongolian gerbils (gerbils). Cholesteatomas were induced in gerbils with a transplanted tympanic membrane by using the external auditory canal (EAC) ligation method. After the pars flaccida of the tympanic membranes were completely removed from male gerbils, corresponding portions of tympanic membranes of female gerbils were transplanted to the area of defect, and then we ligated the EAC (hybrid-model group). As a control group, the EAC of normal male and female gerbils was ligated without myringoplasty. In all ears of each group, the induced cholesteatomas were seen. In situ PCR was then performed to detect the mouse X chromosome-linked phosphoglycerate kinase-1 (pgk-1) gene on the paraffin sections. One pgk-1 spot in the epithelial nuclei was detected in male cholesteatoma, and two pgk-1 spots were detected in female cholesteatoma, respectively. On the other hand, in the hybrid-model group, we detected not only one but also two pgk-1 spots in the epithelial nuclei of cholesteatoma. These results strengthened the evidence that the origin of epithelial cells in cholesteatoma is the tympanic membrane in this model, but not the residential middle ear epithelial cells or the skin of the EAC.

Evidence of Xist RNA-independent initiation of mouse imprinted X-chromosome inactivation

XX female mammals undergo transcriptional silencing of most genes on one of their two X-chromosomes to equalize X-linked gene dosage with XY males in a process referred to as X-chromosome inactivation (XCI). XCI is a paradigm of epigenetic regulation1. Once enacted in individual cells of the early female embryo, XCI is stably transmitted such that most descendant cells maintain silencing of that X-chromosome2. In eutherian mammals, XCI is thought to be triggered by the expression of the non-coding Xist RNA from the future inactive-X (Xi)3,4,5; Xist RNA in turn is proposed to recruit protein complexes that bring about heterochromatinization of the Xi6,7. Here we test whether imprinted XCI, which results in preferential inactivation of the paternal X-chromosome (Xp), occurs in mouse embryos inheriting an Xp lacking Xist. We find that silencing of Xp-linked genes can initiate in the absence of paternal Xist; Xist is, however, required to stabilize silencing of the Xp. Xp-linked gene silencing associated with mouse imprinted XCI, therefore, can initiate in the embryo independently of Xist RNA.

Improvement of in situ PCR by optimization of PCR cycle number and proteinase k concentration: localization of x chromosome-linked phosphoglycerate kinase-1 gene in mouse reproductive organs

In situ polymerase chain reaction (in situ PCR), which can detect a few copies of genes within a cell by amplifying the target gene, was developed to better understand the biological functions of tissues. In this study, we optimized the protocol conditions for the detection of X chromosome-linked phosphoglycerate kinase-1 (pgk-1) gene in paraffin-embedded sections of mouse reproductive organs. The effects of various concentrations of proteinase K (PK) and PCR cycle numbers were examined. To label the amplified DNA, we used digoxigenin-dUTP (Dig), Cy-3-dUTP (Cy-3), or FluorX-dCTP (FluorX). The optimal concentration of PK was 50 µg/ml for the ovary and 10 µg/ml for the testis. Ten PCR cycles were optimal for Dig and 25 cycles were optimal for FluorX and Cy-3 in the ovary and testis. The signal-to-noise ratio of FluorX and Cy-3 for ovarian tissue was better than that of Dig. Using the above conditions, we detected 1–4 and 1–2 spots of pgk-1 in the nuclei of granulosa and germ cells, respectively. Our results indicate that in situ PCR is useful for detecting a specific gene in paraffin-embedded sections under optimized conditions of both PCR cycle number and PK concentration.

Polymorphisms affecting gene transcription and mRNA processing in pharmacogenetic candidate genes: detection through allelic expression imbalance in human target tissues

BACKGROUND

Genetic variation in mRNA expression plays a critical role in human phenotypic diversity, but it has proven difficult to detect regulatory polymorphisms - mostly single nucleotide polymorphisms (rSNPs). Additionally, variants in the transcribed region, termed here 'structural RNA SNPs' (srSNPs), can affect mRNA processing and turnover. Both rSNPs and srSNPs cause allelic mRNA expression imbalance (AEI) in heterozygous individuals. We have used AEI to discover and characterize regulatory polymorphisms in OPRM1, TPH2, MDR1, DRD2, and VKORC1. The objective of this study was to use AEI to determine the extent of cis-regulatory factors in pharmacogenetic genes.

METHODS

We applied a rapid and accurate AEI methodology for testing 42 genes implicated in cardiovascular and central nervous system diseases, and affecting drug metabolism and transport. Each gene was analyzed in physiologically relevant human autopsy tissues, including brain, heart, liver, intestines, and lymphocytes.

RESULTS

Substantial AEI was observed in approximately 55% of the surveyed genes. Focusing on cardiovascular candidate genes in human hearts, AEI analysis revealed frequent cis-acting regulatory factors in ACE and SOD2 mRNA expression, having potential clinical significance. SNP scanning to locate regulatory polymorphisms in a number of genes failed to support several previously proposed promoter SNPs discovered with use of reporter gene assays in heterologous tissues, while srSNPs appear more frequent than expected. Computational analysis of mRNA folding indicates that approximately 90% of srSNPs affect mRNA folding, and hence potentially function.

CONCLUSION

Our results indicate that both rSNPs and srSNPs represent a still largely untapped reservoir of variants that contribute to human phenotypic diversity.

Biological effects of FoxJ2 over-expression

As reported previously, we have extensively studied FoxJ2, a member of the Fork Head transcription factors family. While the biochemical and functional structures of this transcription factor are well understood, its biological function remains unknown. Here, we present data that address this point using transgenic mouse technology. We found that the birth rate and the number of transgenic animals obtained when transferring embryos over-expressing the FoxJ2 protein were lower than those obtained with embryos over-expressing a control protein, suggesting FoxJ2 overexpression has a negative effect on embryonic development. Transient FoxJ2 transgenesis experiments have confirmed that FoxJ2 over-expression has a lethal effect on embryonic development from E10.5. Moreover, in vitro culture of FoxJ2-microinjected embryos demonstrated a significant developmental blockage, indicating that FoxJ2 could also have an effect on pre-implantation stages. Most probably, these negative effects of FoxJ2 over-expression during development also explain the low percentage of adult transgenic mice obtained. Furthermore, most of the transgenic mice that lived to adulthood did not show transgene expression. In fact, the only two adult transgenic animals (one male and one female) in which FoxJ2 transgene expression was detected showed a mosaic expression and died prematurely as a result of cardio-respiratory failure. Postmortem analysis of these animals revealed a hypertrophic heart and abnormal testes in the male. In order to identify genes regulated by FoxJ2 consistent with the phenotypes observed for FoxJ2 transgenic mice, EMSA assays and co-transfection experiments were carried out. Our data indicate that the genes coding for the gap junction protein Connexin-43 and the cell-cell contact protein E-Cadherin, may be good candidates for FoxJ2-regulated genes. Interestingly, Connexin-43 and E-Cadherin show expression patterns similar to FoxJ2, and the phenotypes of Connexin-43 and E-Cadherin mutants resemble those of our FoxJ2 transgenic animals. These data suggest that the lethal effect on embryonic development of FoxJ2 overexpression, as well as the alterations observed in the heart and testes of adult transgenic mice, could be determined by changes in the transcription of genes such as Connexin-43 and/or E-Cadherin.

Expression partitioning between genes duplicated by polyploidy under abiotic stress and during organ development

Allopolyploidy has been a prominent mode of speciation and a recurrent process during plant evolution and has contributed greatly to the large number of duplicated genes in plant genomes [1-4]. Polyploidy often leads to changes in genome organization and gene expression [5-9]. The expression of genes that are duplicated by polyploidy (termed homeologs) can be partitioned between the duplicates so that one copy is expressed and functions only in some organs and the other copy is expressed only in other organs, indicative of subfunctionalization [10]. To determine how homeologous-gene expression patterns change during organ development and in response to abiotic stress conditions, we have examined expression of the alcohol dehydrogenase gene AdhA in allopolyploid cotton (Gossypium hirsutum). Expression ratios of the two homeologs vary considerably during the development of organs from seedlings and fruits. Abiotic stress treatments, including cold, dark, and water submersion, altered homeologous-gene expression. Most notably, only one copy is expressed in hypocotyls during a water-submersion treatment, and only the other copy is expressed during cold stress. These results imply that subfunctionalization of genes duplicated by polyploidy has occurred in response to abiotic stress conditions. Partitioning of duplicate gene expression in response to environmental stress may lead to duplicate gene retention during subsequent evolution.

Functional analysis of regulatory single-nucleotide polymorphisms

PURPOSE OF REVIEW

The identification of regulatory polymorphisms has become a key problem in human genetics. In the past few years there has been a conceptual change in the way in which regulatory single-nucleotide polymorphisms are studied. We revise the new approaches and discuss how gene expression studies can contribute to a better knowledge of the genetics of common diseases.

RECENT FINDINGS

New techniques for the association of single-nucleotide polymorphisms with changes in gene expression have been recently developed. This, together with a more comprehensive use of the old in-vitro methods, has produced a great amount of genetic information. When added to current databases, it will help to design better tools for the detection of regulatory single-nucleotide polymorphisms.

SUMMARY

The identification of functional regulatory single-nucleotide polymorphisms cannot be done by the simple inspection of DNA sequence. In-vivo techniques, based on primer-extension, and the more recently developed 'haploChIP' allow the association of gene variants to changes in gene expression. Gene expression analysis by conventional in-vitro techniques is the only way to identify the functional consequences of regulatory single-nucleotide polymorphisms. The amount of information produced in the last few years will help to refine the tools for the future analysis of regulatory gene variants.

Investigating cis-acting regulatory variation using assays of relative allelic expression

Assays of relative allelic expression offer a powerful means of exploring the influence of cis-acting regulatory variation. In this article, we provide a general assay protocol and discuss practical and theoretical considerations for the design and interpretation of relative allelic expression analysis.

Evolution of mammalian X chromosome-linked imprinting

We analyse the evolution of X chromosome-linked imprinting by modifying our previous model of imprinting of autosomal genes that influence the trade-off between maternal fecundity and offspring viability through alterations in maternal investment (Mills and Moore, 2004). Unlike previous genetic models, we analyse X-linked imprinting in the context of populations at equilibrium for either autosomal or X-linked biallelically expressed alleles at loci that influence the fecundity/viability trade-off. We show that selection under parental conflict over maternal investment in offspring can parsimoniously explain the occurrence of sex-specific gene expression patterns, without a requirement to postulate direct selection for sexual dimorphism mediated through imprinting. We note that sex chromosome imprinting causes a small distortion of the post-weaning sex ratio, providing a possible selection pressure against the evolution of X-linked imprints. We discuss our conclusions in the context of recent reports of imprinting of mouse X-linked Xlr genes.

The dynamics of imprinted X inactivation during preimplantation development in mice

In the mouse, there are two forms of X chromosome inactivation (XCI), random XCI in the fetus and imprinted paternal XCI, which is limited to the extraembryonic tissues. While the mechanism of random XCI has been studied extensively using the in vitro XX ES cell differentiation system, imprinted XCI during early embryonic development has been less well characterized. Recent studies of early embryos have reported unexpected findings for the paternal X chromosome (Xp). Imprinted XCI may not be linked to meiotic silencing in the male germ line but rather to the imprinted status of the Xist gene. Furthermore, the Xp becomes inactivated in all cells of cleavage-stage embryos and then reactivated in the cells of the inner cell mass (ICM) that form the epiblast, where random XCI ensues.

Genetics, epigenetics and gene silencing in differentiating mammalian embryos

A highly complex pattern of differentiation involving maternal and embryonic factors characterizes the early development of mammalian embryos. These complex genetic and proteonomic patterns of early growth also involve various forms of gene silencing and tissue reprogramming. Understanding the nature of fundamental developmental events is hence essential to appreciate the significance of natural and induced forms of remodelling, damaged forms of gene expression and gene silencing during the initial stages of growth. Natural forms of remodelling include subtle genetic events involved in, for example, the changing nature of imprinting from before fertilization or the inactivation of one X chromosome in female blastocysts. Induced forms include the consequences of nuclear transfer and embryo cloning or the immediate effects of placing embryos in culture media. Animal and human studies are described in this paper, relating reprogramming to detailed embryological and clinical knowledge gained through the use of IVF, preimplantation genetic diagnosis and the establishment in vitro of stem cells. Attention concentrates on the consequences of variations in all growth stages from the formation of oocytes, through fertilization, the differentiation of blastocysts and early haemopoietic stages in mammalian species. Unique features of gene expression or gene modification are described for each developmental stage.

Challenges for identifying functionally important genetic variation: the promise of combining complementary research strategies

Strategies for the identification of functional genetic variation underlying phenotypic traits of ecological and evolutionary importance have received considerable attention in the literature recently. This paper aims to bring together and compare the relative strengths and limitations of various potentially useful research strategies for dissecting functionally important genetic variation in a wide range of organisms. We briefly explore the relative strengths and limitations of traditional and emerging approaches and evaluate their potential use in free-living populations. While it is likely that much of the progress in functional genetic analyses will rely on progress in traditional model species, it is clear that with prudent choices of methods and appropriate sampling designs, much headway can be also made in a diverse range of species. We suggest that combining research approaches targeting different functional and biological levels can potentially increase understanding the genetic basis of ecological and evolutionary processes both in model and non-model organisms.

X-chromosome inactivation: a hypothesis linking ontogeny and phylogeny

In mammals, sex is determined by differential inheritance of a pair of dimorphic chromosomes: the gene-rich X chromosome and the gene-poor Y chromosome. To balance the unequal X-chromosome dosage between the XX female and XY male, mammals have adopted a unique form of dosage compensation in which one of the two X chromosomes is inactivated in the female. This mechanism involves a complex, highly coordinated sequence of events and is a very different strategy from those used by other organisms, such as the fruitfly and the worm. Why did mammals choose an inactivation mechanism when other, perhaps simpler, means could have been used? Recent data offer a compelling link between ontogeny and phylogeny. Here, we propose that X-chromosome inactivation and imprinting might have evolved from an ancient genome-defence mechanism that silences unpaired DNA.

Haplotypes at the dystrobrevin binding protein 1 (DTNBP1) gene locus mediate risk for schizophrenia through reduced DTNBP1 expression

The DTNBP1 gene, encoding dysbindin, is now generally considered to be a susceptibility gene for schizophrenia. However, the confidence with which this hypothesis can be held has to be tempered by the poor reproducibility between studies in terms of the exact nature of the associated haplotypes, by the failure so far to identify any specific susceptibility variants and by the absence of any demonstrated function associated with any of the risk haplotypes. In the present study, we show that a defined schizophrenia risk haplotype tags one or more cis-acting variants that results in a relative reduction in DTNBP1 mRNA expression in human cerebral cortex. Subsidiary analyses suggest that risk haplotypes identified in other sample groups of white European ancestry also index lower DTNBP1 expression, whereas putative 'protective' haplotypes index high DTNBP1 expression. Our data indicate that variation in the DTNBP1 gene confers susceptibility to schizophrenia through reduced expression, and that this, therefore, represents a primary aetiological mechanism in the disorder.

Regulation of histone H3 lysine 9 methylation in oocytes and early pre-implantation embryos

Epigenetic modifications of the genome, such as covalent modification of histone residues, ensure appropriate gene activation during pre-implantation development, and are probably involved in the asymmetric reprogramming of the parental genomes after fertilization. We investigated the methylation patterns of histone H3 at lysine 9 (H3/K9), and the regulatory mechanism involved in the asymmetric remodeling of parental genomes during early preimplantation development in mice. Immunocytochemistry with an antibody that specifically recognizes methylated H3/K9 showed a very weak or absent methylation signal in the male pronucleus, whereas a distinct methylation signal was detected in the female pronucleus. This asymmetric H3/K9 methylation pattern in the different parental genomes persisted until the two-cell stage. However, de novo methylation of H3/K9 occurred and the asymmetry was lost during the four-cell stage. The unmethylated male pronucleus underwent de novo methylation when it was transferred into enucleated GV- or MII-stage oocytes, which suggests that histone H3 methylase is active before fertilization, but not afterwards, and that the asymmetric methylation pattern is generated by this change in methylase activity in the cytoplasm after fertilization. Thus, histone H3 is methylated only in the maternal chromosomes, which are present in the oocytes before fertilization, and is not methylated in the paternal chromosomes, which are absent. The maintenance of asymmetric H3/K9 methylation patterns in early embryos is an active process that depends on protein synthesis and zygotic transcription, as de novo methylation in the male pronucleus occurred when either protein synthesis or gene expression was inhibited by cycloheximide orα-amanitin, respectively. In addition, corresponding de novo methylation of H3/K9 and DNA occurred when the male pronucleus was transferred to an enucleated GV oocyte. Our results suggest that H3/K9 methylation is an epigenetic marker of parental genome origin during early preimplantation development.

Reactivation of the Paternal X Chromosome in Early Mouse Embryos

It is generally accepted that paternally imprinted X inactivation occurs exclusively in extraembryonic lineages of mouse embryos, whereas cells of the embryo proper, derived from the inner cell mass (ICM), undergo only random X inactivation. Here we show that imprinted X inactivation, in fact, occurs in all cells of early embryos and that the paternal X is then selectively reactivated in cells allocated to the ICM. This contrasts with more differentiated cell types where X inactivation is highly stable and generally irreversible. Our observations illustrate that an important component of genome plasticity in early development is the capacity to reverse heritable gene silencing decisions.

The serotonin-2A receptor gene locus does not contain common polymorphism affecting mRNA levels in adult brain

The serotonin-2A (HTR2A) receptor is a molecule of particular interest in biological psychiatry, as it is an important target for psychotropic drugs, and altered HTR2A expression has been found in several neuropsychiatric conditions, including depression and schizophrenia. Genetic association has been reported between a synonymous 102T/C polymorphism in the gene encoding HTR2A and a number of clinical phenotypes, including schizophrenia, clozapine response, psychotic symptoms in Alzheimer's disease and certain features of depression. Given that there are no known effects of the 102T/C polymorphism on the structure of the receptor, attention has switched to the possibility that the observations of both altered expression and genetic association point to functional sequence variants that alter expression of the HTR2A gene. Moreover, data have been presented recently suggesting that mRNAs containing the 102T- and C-alleles are differentially expressed. This suggests a direct effect of the variant itself on mRNA levels, or the influence of a distinct regulatory variant, such as the -1438A/G promoter polymorphism, with which it is in perfect linkage disequilibrium. The present study tested this hypothesis by employing a highly accurate quantitative allele- specific primer extension assay to measure the relative expression of brain mRNAs carrying each allele in 23 individuals heterozygous for the 102T/C polymorphism. Comparison between allele ratios derived from genomic DNA and mRNA from several cortical regions revealed that the 102C- and T-alleles are expressed identically. Furthermore, the absence of any interindividual variability in relative mRNA allele ratio suggests that the HTR2A locus is unlikely to contain common polymorphisms or epigenetic modification that alter HTR2A mRNA levels in adult brain, and essentially exclude such phenomena as a potential explanation for the altered expression and genetic associations that have been reported to date.

Epigenetic dynamics of imprinted X inactivation during early mouse development

The initiation of X-chromosome inactivation is thought to be tightly correlated with early differentiation events during mouse development. Here, we show that although initially active, the paternal X chromosome undergoes imprinted inactivation from the cleavage stages, well before cellular differentiation. A reversal of the inactive state, with a loss of epigenetic marks such as histone modifications and polycomb proteins, subsequently occurs in cells of the inner cell mass (ICM), which give rise to the embryo-proper in which random X inactivation is known to occur. This reveals the remarkable plasticity of the X-inactivation process during preimplantation development and underlines the importance of the ICM in global reprogramming of epigenetic marks in the early embryo.

Cis-acting variation in the expression of a high proportion of genes in human brain

Much of the genetic component of human phenotypic diversity, including susceptibility to disease, is proposed to be the result of cis-acting influences on gene expression. If this hypothesis is correct, it implies that cis-acting regulatory variation is a common phenomenon. However, direct evidence in support of this view is scarce. We have applied highly quantitative measures of allele-specific expression in order to screen an average of 19 informative subjects (range 8-26) for the presence of common cis-acting influences on the expression of 15 genes by using RNA derived from human brain. We found that, in seven of the 15 assayed genes, at least one individual exhibited relative differences in allelic expression of 20% or more and, in one gene (DTNBP1), allelic expression differences exceeded 50%. These results suggest that cis-acting variation in gene expression commonly occurs in native tissue and hence provide empirical support for the hypothesis that this is potentially an important mechanism underlying human phenotypic diversity.

A haplotype implicated in schizophrenia susceptibility is associated with reduced COMT expression in human brain

The gene encoding catechol-O-methyltransferase (COMT) is a strong candidate for schizophrenia susceptibility, owing to the role of COMT in dopamine metabolism, and the location of the gene within the deleted region in velocardiofacial syndrome, a disorder associated with high rates of schizophrenia. Recently, a highly significant association was reported between schizophrenia and a COMT haplotype in a large case-control sample (Shifman et al. 2002). In addition to a functional valine-->methionine (Val/Met) polymorphism, this haplotype included two noncoding single-nucleotide polymorphisms (SNPs) at either end of the COMT gene. Given the role of COMT in dopamine catabolism and that deletion of 22q11 (containing COMT) is associated with schizophrenia, we postulated that the susceptibility COMT haplotype is associated with low COMT expression. To test this hypothesis, we have applied quantitative measures of allele-specific expression using mRNA from human brain. We demonstrate that COMT is subject to allelic differences in expression in human brain and that the COMT haplotype implicated in schizophrenia (Shifman et al. 2002) is associated with lower expression of COMT mRNA. We also show that the 3' flanking region SNP that gave greatest evidence for association with schizophrenia in that study is transcribed in human brain and exhibits significant differences in allelic expression, with lower relative expression of the associated allele. Our results indicate that COMT variants other than the Val/Met change are of functional importance in human brain and that the haplotype implicated in schizophrenia susceptibility is likely to exert its effect, directly or indirectly, by down-regulating COMT expression.

Trophectoderm-specific expression of the X-linked Bex1/Rex3 gene in preimplantation stage mouse embryos

The Bex1/Rex3 gene was recently identified as an X-linked gene that is differentially expressed between parthenogenetic and normal fertilized, preimplantation stage mouse embryos. The Bex1/Rex3 gene appears to be expressed preferentially from the maternal X chromosome in blastocysts, but from either X chromosome in later stage embryonic tissues and adult tissues. To investigate whether differential expression of the Bex1/Rex3 gene between normal and parthenogenetic blastocyst stage embryos reflects genomic imprinting at the Bex1/Rex3 locus itself, or instead is the result of preferential inactivation of the paternal X chromosome or differences in timing of cellular differentiation, we examined in detail the expression pattern of the Bex1/Rex3 mRNA in normal preimplantation stage embryos, and compared its expression between androgenetic, gynogenetic, and normal fertilized embryos. Expression data reveal that the Bex1/Rex3 gene is initially transcribed at the 2-cell stage, transiently induced at the 8-cell stage, and then increases in expression again at the blastocyst stage. Very little expression is observed in isolated inner cell masses, indicating selective expression in the trophectoderm. Comparisons of Bex1/Rex3 mRNA expression between male and female androgenetic and control embryos and gynogenetic embros failed to reveal any significant difference in expression between the different classes of embryos at the 8-cell stage, or the expanding blastocyst stage (121 hr post-hCG). At the late blastocyst stage (141 hr post-hCG), expression was significantly lower in XY control embryos as compared with XX controls. Bex1/Rex3 mRNA expression did not differ between XX and XY androgenones at the blastocyst stage or between gynogenones and XX control embryos. Thus, the Bex1/Rex3 gene does not appear to be regulated directly by genomic imprinting during the preimplantation period, just as it is not regulated by imprinting at later stages. Apparent differences in gene expression may arise through the effects of trophectoderm-specific expression coupled with differences in timing of trophectoderm differentiation between the different classes of embryos and effects of preferential paternal X chromosome inactivation (XCI).

Regulatory mechanisms at the mouse Igf2/H19 locus

The closely linked H19 and Igf2 genes show highly similar patterns of gene expression but are reciprocally imprinted. H19 is expressed almost exclusively from the maternally inherited chromosome, while Igf2 expression is mostly from the paternal chromosome. In humans, loss of imprinting at this locus is associated with tumors and with developmental disorders. Monoallelic expression at the imprinted Igf2/H19 locus occurs by at least two distinct mechanisms: a developmentally regulated silencing of the paternal H19 promoter, and transcriptional insulation of the maternal Igf2 promoters. Both mechanisms of allele-specific silencing are ultimately dependent on a common cis-acting element located just upstream of the H19 promoter. The coordinated expression patterns and some experimental data support the idea that positive regulatory elements are also shared by the two genes. To clarify the organization and function of positive and negative regulatory elements at the H19/Igf2 locus, we analyzed two mouse mutations. First, we generated a deletion allele to localize enhancers used in vivo for expression of both H19 and Igf2 in mesodermal tissues to sequences downstream of the H19 gene. Coincidentally, we demonstrated that some expression of Igf2 is independent of the shared enhancer element. Second, we used this new information to further characterize an ectopic H19 differentially regulated region and the associated insulator. We demonstrated that its activity is parent-of-origin dependent. In contrast to recent results from Drosophila model systems; we showed that this duplication of a mammalian insulator does not interfere with its normal function. Implications of these findings for current models for monoallelic gene expression at this locus are discussed.

Fluorescence-based DHPLC for allelic quantification by single-nucleotide primer extension

We have investigated the possibility of determining quantitatively the alleles of binary DNA polymorphisms by single-nucleotide primer extension (SNuPE) and fluorescence-based DHPLC. Using a polymorphism of interest to our group, ROX-labeled dideoxy CTP (ROX-ddCTP) was incorporated at the 3' end of the primer annealed to the template adjacent to the polymorphic site. The primer extension product was then resolved from the unincorporated dye terminator by ion-pair reversed-phase liquid chromatography. The signal intensity of incorporated ROX-ddCTP correlated well over one order of magnitude with the relative amount of the C-allele present in the genomic DNA template. We conclude that SNuPE, when combined with fluorescence-based DHPLC, can accurately determine the relative molar proportion of one allele in total DNA.

A transcriptional insulator at the imprinted H19/Igf2 locus

Igf2 and H19 exhibit parent-of-origin-specific monoallelic expression. H19 is expressed from the maternal chromosome and Igf2 from the paternal. The two genes share enhancer elements and monoallelic expression of both genes is dependent on cis-acting sequences upstream of the H19 promoter. In this work we examine the mechanisms by which this region silences the maternal Igf2 allele and we demonstrate that deletion of this region can result in high levels of activation of both H19 and Igf2 from a single chromosome. Moreover, by inserting this cis element between a promoter and its enhancer at a heterologous position, we demonstrate that the sequences carry both insulator activity and the ability to be stably imprinted. We also characterize the insulator in vitro and show that it is neither enhancer nor promoter specific.

A transcriptional insulator at the imprinted H19/Igf2 locus

Igf2 and H19 exhibit parent-of-origin-specific monoallelic expression. H19 is expressed from the maternal chromosome and Igf2 from the paternal. The two genes share enhancer elements and monoallelic expression of both genes is dependent on cis-acting sequences upstream of the H19 promoter. In this work we examine the mechanisms by which this region silences the maternal Igf2 allele and we demonstrate that deletion of this region can result in high levels of activation of both H19and Igf2 from a single chromosome. Moreover, by inserting thiscis element between a promoter and its enhancer at a heterologous position, we demonstrate that the sequences carry both insulator activity and the ability to be stably imprinted. We also characterize the insulator in vitro and show that it is neither enhancer nor promoter specific.

Effects of X chromosome number and parental origin on X-linked gene expression in preimplantation mouse embryos

Diploid androgenetic mouse embryos, possessing two sets of paternally inherited chromosomes, and control fertilized embryos were used to examine the relative effects of X chromosome number and parental chromosome origin on androgenone viability and X-linked gene expression. A significant difference in efficiency of blastocyst formation was observed between XX and XY androgenones in some experiments, but this difference was not uniformly observed. Significant effects of both X chromosome number and parental origin on X-linked gene expression were observed. Male and female control embryos expressed the XIST: RNA initially. This expression was followed by a preferential reduction in XIST: RNA abundance in male embryos, indicating that dosage compensation for the X chromosome may normally require the downregulation of XIST: RNA expression in male embryos, in conjunction with the production of stable XIST: transcripts in female embryos. By the late blastocyst stage, XX control embryos expressed significantly more XIST: RNA than did XY embryos. Unlike their normal counterparts, XX androgenones did not express significantly more XIST: RNA than did XY androgenones at the late blastocyst stage. Androgenones exhibited severe repression of the Pgk1 gene, but during development to the late blastocyst stage Pgk1 mRNA expression increased in XX androgenones and decreased in XY androgenones. Thus, the initial repression of the Pgk1 gene in XX androgenones was lost as the XIST: RNA declined in abundance, and this loss was correlated with a failure of XX androgenones to express significantly more XIST: RNA than did XY androgenones. These results indicate that androgenones may lack a factor that is expressed from the maternal genome and required for dosage compensation in preimplantation embryos. The results also indicate that early dosage compensation in preimplantation embryos may normally be reversible, thus providing flexibility to meet different developmental requirements of the embryonic and extraembryonic lineages.

Histone macroH2A1 is concentrated in the inactive X chromosome of female preimplantation mouse embryos

MacroH2As are core histone proteins with a hybrid structure consisting of a domain that closely resembles a full-length histone H2A followed by a large nonhistone domain. We recently showed that one of the macroH2A subtypes, macroH2A1.2, is concentrated in the inactive X chromosome in adult female mammals. Here we examine the timing of the association of macroH2A1.2 with the inactive X chromosome during preimplantation mouse development in order to assess the possibility that macroH2A1 participates in the initiation of X inactivation. The association of macroH2A1.2 with one of the X chromosomes was observed in 50% of blastocysts, occurring mostly, if not exclusively, in extraembryonic cells as was expected from previous studies, which indicated that X inactivation in embryonic lineages happens after implantation. Examination of earlier embryonic stages indicates that the association of macroH2A1 with the inactive X chromosome begins between the 8- and 16-cell stages. Of the changes that are known to happen during X inactivation in preimplantation embryos, the accumulation of macroH2A1 appears to be the earliest marker of the inactive X chromosome and is the only change that has been shown to occur during the period when transcriptional silencing is initiated.

Translation of maternal messenger ribonucleic acids encoding transcription factors during genome activation in early mouse embryos

Embryonic genome activation (EGA) in mice is sensitive to treatment with cycloheximide, indicating that protein synthesis plays an important role in mediating EGA. We hypothesized that regulated maternal mRNA recruitment may control the time of EGA by controlling the time of appearance of certain transcription factors (TFs). We also hypothesized that synthesis of other TFs may contribute to EGA independently of controlling the timing of EGA. To test these hypotheses, we used sucrose density gradient fractionation coupled to a quantitative reverse transcription-polymerase chain reaction method to compare polysomal mRNA abundances of specific TF mRNAs between metaphase II oocytes, 1-cell-stage embryos, and 2-cell-stage embryos. We observed a 2-cell-stage-specific increase in polysomal abundance of mouse TEA DNA binding domain 2 (mTEAD-2) mRNA, coincident with the first appearance of mTEAD activity in the early embryo. The mRNAs encoding Sp1, TATA binding protein, and cyclic AMP response element binding protein did not undergo translational recruitment, but exhibited differences in polysomal abundance. We also observed a continuous, high proportion in the polysomal fraction for the mRNA encoding ribosomal protein L23 mRNA, which contrasted with the patterns observed for other maternal transcripts. These observations are consistent with the hypothesis that regulated recruitment of maternal TF mRNAs may control the time of activation of some genes during EGA, and that continuous synthesis of other TFs, like Sp1, may facilitate EGA.

A hemizygous short tandem repeat polymorphism 3' to the human phosphoglycerate kinase gene

The human phosphoglycerate kinase (PGK) gene is located within Xq11-Xq13, a region implicated in genitourinary diseases including: prostate cancer, androgen insensitivity, perineal hypospadias, and other genetic abnormalities. The PGK gene and the androgen receptor gene are in linkage disequilibrium. PGK has been mapped extensively for nuclease-sensitive sites, methylation sites, and flanking DNA sequences. A PGK-associated BstXI polymorphism has been used to determine clonality of neoplastic tissues. Using fluorescent PCR product analysis and DNA sequencing, we discovered that a short tandem repeat (STR) in the 3' flanking region of the PGK gene is polymorphic. Among 231 individuals, there were nine distinct alleles, including eight based on variations in the number of TATC repeats. The PGK STR demonstrated hemizygosity, consistent with its X-chromosomal location and with an absence of cross-hybridizing autosomal homologs. The polymorphic PGK STR shows promise for rapid investigation of neoplastic clonality, for personal identification, and for studies of inherited predisposition to urologic disorders.

Bex1, a gene with increased expression in parthenogenetic embryos, is a member of a novel gene family on the mouse X chromosome

Parthenogenetic and normal blastocysts were compared using differential display analysis as a means to identify new imprinted genes. A single gene was identified with increased expression in parthenogenetic blastocysts, suggesting it might be an imprinted gene expressed from the maternally inherited allele. The gene, named Bex1 (brainexpressedX-linked gene), maps near Plp on the mouse X chromosome and to Xq22 in humans. Database homology searches revealed two additional uncharacterized cDNAs similar to Bex1 that were named Bex2 and Bex3. Allele-specific expression analysis of Bex1 using F1 blastocysts indicated an excess of transcript expressed from the maternally inherited allele compared with the paternally inherited allele. This excess level of transcript derived from the maternally inherited allele may be due to imprinted X inactivation of the paternally inherited allele in the extraembryonic lineages of female embryos rather than a result of genomic imprinting.

Xist and X chromosome inactivation

X inactivation acts in female mammals to equalise X-linked gene dosage between XX females and XY males. X inactivation is controlled by a single X-linked cis-acting locus called the X inactivation centre (Xic). In 1991 the Xist gene was identified as a candidate for the Xic. Xist is expressed in all adult female tissues, but only from the allele on the inactive X. The Xist transcript does not encode a protein but remains sequestered within the nucleus and co-localises with the inactive X chromosome. Transgenic and knockout studies have shown that a genomic region covering only a few kilobases either side of Xist carries all of the functions attributed to the Xic. The major questions currently occupying researchers studying X inactivation are: how do cells count their number of X chromosomes to determine whether X inactivation is necessary, and how does the Xist transcript inactivate all genes on the X chromosome?

X-chromosome inactivation in mammals

The inactive X chromosome differs from the active X in a number of ways; some of these, such as allocyclic replication and altered histone acetylation, are associated with all types of epigenetic silencing, whereas others, such as DNA methylation, are of more restricted use. These features are acquired progressively by the inactive X after onset of initiation. Initiation of X-inactivation is controlled by the X-inactivation center (Xic) and influenced by the X chromosome controlling element (Xce), which causes primary nonrandom X-inactivation. Other examples of nonrandom X-inactivation are also presented in this review. The definition of a major role for Xist, a noncoding RNA, in X-inactivation has enabled investigation of the mechanism leading to establishment of the heterochromatinized X-chromosome and also of the interactions between X-inactivation and imprinting as well as between X-inactivation and developmental processes in the early embryo.

Epigenetic changes at the insulin-like growth factor II/H19 locus in developing kidney is an early event in Wilms tumorigenesis

Relaxation of imprinting at the insulin-like growth factor II (IFG-II)/H19 locus is a major mechanism involved in the onset of sporadic Wilms tumor and several other embryonal tumors. The high prevalence of histologically abnormal foci in kidney adjacent to Wilms tumors suggests that tumor-predisposing genetic/epigenetic lesion might also be found at high frequency in Wilms tumor-bearing kidneys. Focusing on Wilms tumors with relaxation of IFG-II imprinting, we determined the frequency of epigenetic change at the IFG-II/H19 locus in adjacent kidney. In all kidneys adjacent to these Wilms tumors, we detected substantial mosaicism for a population of cells with relaxation of IFG-II imprinting and biallelic H19 methylation, regardless of whether the patient had a tumor-predisposing syndrome or not. The high proportion of epigenetically modified cells among "normal" tissue indicates that the epigenetic error occurred very early in development, before the onset of Wilms tumor. Not only does this suggest that the major Wilms tumor-predisposing event occurs within the first few days of development, but it also suggests that sporadic Wilms tumor may represent one end of a spectrum of overgrowth disorders characterized by mosaic epigenetic change at the IFG-II/H19 locus.

Theoretical uncertainty of measurements using quantitative polymerase chain reaction

Current quantitative polymerase chain reaction (PCR) protocols are only indicative of the quantity of a target sequence relative to a standard, because no means of estimating the amplification rate is yet available. The variability of PCR performed on isolated cells has already been reported by several authors, but it could not be extensively studied, because of lack of a system for doing kinetic data acquisition and of statistical methods suitable for analyzing this type of data. We used the branching process theory to simulate and analyze quantitative kinetic PCR data. We computed the probability distribution of the offspring of a single molecule. We demonstrated that the rate of amplication has a severe influence on the shape of this distribution. For high values of the amplification rate, the distribution has several maxima of probability. A single amplification trajectory is used to estimate the initial copy number of the target sequence as well as its confidence interval, provided that the amplification is done over more than 20 cycles. The consequence of possible molecular fluctuations in the early stage of amplification is that small copy numbers result in relatively larger intervals than large initial copy numbers. The confidence interval amplitude is the theoretical uncertainty of measurements using quantitative PCR. We expect these results to be applicable to the data produced by the next generation of thermocyclers for quantitative applications.

Single nucleotide primer extension: quantitative range, variability, and multiplex analysis

The quantitative measurement of transcription products from homologous alleles at a diploid locus has broad application for the study of mammalian gene expression. Single nucleotide primer extension (SNuPE) analysis is a simple and sensitive method for allelic transcript discrimination requiring only 1 bp difference between alleles. In this study the effective range, experimental variation, and the influences of poly(dT)-primed and gene-specific reverse transcriptions are characterized. The ability to analyze several genes from a single reverse transcription reaction is assessed as well. For the genes examined, the maximum range of detection is reached when the minor transcript represents 1/250 of the major allele. Relatively little error is seen within or between assays and linearity of response is maintained over an approximately thousandfold range.

Requirement for Xist in X chromosome inactivation

The Xist gene has been proposed as a candidate for the X inactivation centre, the master regulatory switch locus that controls X chromosome inactivation. So far this hypothesis has been supported solely by indirect evidence. Here we describe gene targeting of Xist, and provide evidence for its absolute requirement in the process of X chromosome inactivation.

Allele-specific expression and total expression levels of imprinted genes during early mouse development: implications for imprinting mechanisms

Genomic imprinting determines the monoallelic expression of a small number of genes during at least later stages of development. To obtain information necessary for the elucidation of imprinting mechanisms, we assessed the allele-specific expression and total expression level of four imprinted genes during early stages of development of normal F1 hybrid mice utilizing quantitative allele-specific reverse transcription-PCR (RT-PCR) single-nucleotide primer extension assays. The Igf2r and Snrpn genes were activated by the early 4-cell stage and exhibited biallelic and monoallelic expression, respectively, throughout preimplantation development. Thus, with respect to different imprinted genes, epigenetic systems determining monoallelic expression are not uniform in their time of establishment. Biallelic expression of Igf2r was observed in single blastomeres, discounting the possibility of random allelic inactivation at this stage. The closely linked H19 and Igf2 genes were activated after the blastocyst stage and often exhibited biallelic and monoallelic expression respectively in tissues of pregastrulation postimplantation-stage embryos, rather than reciprocal monoallelic modes as observed at later stages. This raises the possibility that imprinting of H19 is involved only in the maintenance and not in the initiation of monoallelic expression of Igf2. Monoallelic expression of Snrpn was observed in each blastomere at the 4-cell stage, demonstrating that the germ line, which exhibits biallelic expression of imprinted genes, must be derived from cells in which imprinting was once manifest.

Stage-specific and cell type-specific aspects of genomic imprinting effects in mammals

Recent studies have revealed that maternal and paternal alleles of some imprinted genes are differentially expressed from the earliest time of expression, with virtually no expression from one of the two alleles, while for other imprinted genes the normally silent allele can be transcribed during early development. In addition, a number of imprinted genes manifest their imprints only in select tissues. These observations indicate that the marks that denote parental chromosome origin need not directly determine allele expression, but rather bias later epigenetic modifications toward a particular allele. Thus, factors expressed at specific stages or in specific cell types are required to silence one parental allele or another. Stage-dependent and tissue-specific epigenetic modifications include the progressive establishment of the mature adult parental allele-specific DNA methylation patterns. These changes resemble and may share a common mechanistic basis with other epigenetic modifications that occur during development. Understanding the mechanisms by which these post-fertilization epigenetic modifications are mediated and regulated will be essential for understanding how genomic imprinting leads to differences in parental allele expression.