Beyond sense: the role of antisense RNA in controlling Xist expression

Transcriptome Encyclopedia of Early Human Development

Our understanding of human pre-implantation development is limited by the availability of human embryos and cannot completely rely on mouse studies. Petropoulos et al. now provide an extensive transcriptome analysis of a large number of human pre-implantation embryos at single-cell resolution, revealing previously unrecognized features unique to early human development.

To Know How a Gene Works, We Need to Redefine It First but then, More Importantly, to Let the Cell Itself Decide How to Transcribe and Process Its RNAs

Recent genomic and ribonomic research reveals that our genome produces a stupendous amount of non-coding RNAs (ncRNAs), including antisense RNAs, and that many genes contain other gene(s) in their introns. Since ncRNAs either regulate the transcription, translation or stability of mRNAs or directly exert cellular functions, they should be regarded as the fourth category of RNAs, after ribosomal, messenger and transfer RNAs. These and other research advances challenge the current concept of gene and raise a question as to how we should redefine gene. We can either consider each tiny part of the classically-defined gene, such as each mRNA variant, as a “gene”, or, alternatively and oppositely, regard a whole genomic locus as a “gene” that may contain intron-embedded genes and produce different types of RNAs and proteins. Each of the two ways to redefine gene not only has its strengths and weaknesses but also has its particular concern on the methodology for the determination of the gene's function: Ectopic expression of complementary DNA (cDNA) in cells has in the past decades provided us with great deal of detail about the functions of individual mRNA variants, and will make the data less conflicting with each other if just a small part of a classically-defined gene is considered as a “gene”. On the other hand, genomic DNA (gDNA) will better help us in understanding the collective function of a genomic locus. In our opinion, we need to be more cautious in the use of cDNA and in the explanation of data resulting from cDNA, and, instead, should make delivery of gDNA into cells routine in determination of genes' functions, although this demands some technology renovation.

Chromatin-remodeling factors mediate the balance of sense-antisense transcription at the FGF2 locus

Antisense transcription is prevalent in mammalian genomes, yet the function of many antisense transcripts remains elusive. We have previously shown that the fibroblast growth factor 2 (FGF2) gene is regulated endogenously by an overlapping antisense gene called Nudix-type motif 6 (NUDT6). However, the molecular mechanisms that determine the balance of FGF2 and NUDT6 transcripts are not yet well understood. Here we demonstrate that there is a strong negative correlation between FGF2 and NUDT6 across 7 different cell lines. Small interfering RNA-mediated knockdown of NUDT6 causes an increase in nascent FGF2 transcripts, including a short FGF2 variant that lacks sequence complementarity with NUDT6, indicating the involvement of transcriptional mechanisms. In support of this, we show that changes in histone acetylation by trichostatin A treatment, histone deacetylase inhibition, or small interfering RNA knockdown of the histone acetyltransferase CSRP2BP, oppositely affect NUDT6 and FGF2 mRNA levels. A significant increase in histone acetylation with trichostatin A treatment was only detected at the genomic region where the 2 genes overlap, suggesting that this may be an important regulatory region for determining the balance of NUDT6 and FGF2. Knockdown of the histone demethylase KDM4A similarly causes a shift in the balance of NUDT6 and FGF2 transcripts. Expression of CSRP2BP and KDM4A correlates positively with NUDT6 expression and negatively with FGF2 expression. The results presented here indicate that histone acetylation and additional chromatin modifiers are important in determining the relative levels of FGF2 and NUDT6 and support a model in which epigenetic remodeling contributes to their relative expression levels.

Microarray analysis of gene expression during early development: a cautionary overview

The rise of the 'omics' technologies started nearly a decade ago and, among them, transcriptomics has been used successfully to contrast gene expression in mammalian oocytes and early embryos. The scarcity of biological material that early developmental stages provide is the prime reason why the field of transcriptomics is becoming more and more popular with reproductive biologists. The potential to amplify scarce mRNA samples and generate the necessary amounts of starting material enables the relative measurement of RNA abundance of thousands of candidates simultaneously. So far, microarrays have been the most commonly used high-throughput method in this field. Microarray platforms can be found in a wide variety of formats, from cDNA collections to long or short oligo probe sets. These platforms generate large amounts of data that require the integration of comparative RNA abundance values in the physiological context of early development for their full benefit to be appreciated. Unfortunately, significant discrepancies between datasets suggest that direct comparison between studies is difficult and often not possible. We have investigated the sample-handling steps leading to the generation of microarray data produced from prehatching embryo samples and have identified key steps that significantly impact the downstream results. This review provides a discussion on the best methods for the preparation of samples from early embryos for microarray analysis and focuses on the challenges that impede dataset comparisons from different platforms and the reasons why methodological benchmarking performed using somatic cells may not apply to the atypical nature of prehatching development.

Genomic landscape of developing male germ cells

Spermatogenesis is a highly orchestrated developmental process by which spermatogonia develop into mature spermatozoa. This process involves many testis- or male germ cell-specific gene products whose expressions are strictly regulated. In the past decade the advent of high-throughput gene expression analytical techniques has made functional genomic studies of this process, particularly in model animals such as mice and rats, feasible and practical. These studies have just begun to reveal the complexity of the genomic landscape of the developing male germ cells. Over 50% of the mouse and rat genome are expressed during testicular development. Among transcripts present in germ cells, 40% - 60% are uncharacterized. A number of genes, and consequently their associated biological pathways, are differentially expressed at different stages of spermatogenesis. Developing male germ cells present a rich repertoire of genetic processes. Tissue-specific as well as spermatogenesis stage-specific alternative splicing of genes exemplifies the complexity of genome expression. In addition to this layer of control, discoveries of abundant presence of antisense transcripts, expressed psuedogenes, non-coding RNAs (ncRNA) including long ncRNAs, microRNAs (miRNAs) and Piwi-interacting RNAs (piRNAs), and retrogenes all point to the presence of multiple layers of expression and functional regulation in male germ cells. It is anticipated that application of systems biology approaches will further our understanding of the regulatory mechanism of spermatogenesis.

X chromosome dosage compensation: how mammals keep the balance

The development of genetic sex determination and cytologically distinct sex chromosomes leads to the potential problem of gene dosage imbalances between autosomes and sex chromosomes and also between males and females. To circumvent these imbalances, mammals have developed an elaborate system of dosage compensation that includes both upregulation and repression of the X chromosome. Recent advances have provided insights into the evolutionary history of how both the imprinted and random forms of X chromosome inactivation have come about. Furthermore, our understanding of the epigenetic switch at the X-inactivation center and the molecular aspects of chromosome-wide silencing has greatly improved recently. Here, we review various facets of the ever-expanding field of mammalian dosage compensation and discuss its evolutionary, developmental, and mechanistic components.

Genomic imprinting mechanisms in mammals

Genomic imprinting is a form of epigenetic gene regulation that results in expression from a single allele in a parent-of-origin-dependent manner. This form of monoallelic expression affects a small but growing number of genes and is essential to normal mammalian development. Despite extensive studies and some major breakthroughs regarding this intriguing phenomenon, we have not yet fully characterized the underlying molecular mechanisms of genomic imprinting. This is in part due to the complexity of the system in that the epigenetic markings required for proper imprinting must be established in the germline, maintained throughout development, and then erased before being re-established in the next generation's germline. Furthermore, imprinted gene expression is often tissue or stage-specific. It has also become clear that while imprinted loci across the genome seem to rely consistently on epigenetic markings of DNA methylation and/or histone modifications to discern parental alleles, the regulatory activities underlying these markings vary among loci. Here, we discuss different modes of imprinting regulation in mammals and how perturbations of these systems result in human disease. We focus on the mechanism of genomic imprinting mediated by insulators as is present at the H19/Igf2 locus, and by non-coding RNA present at the Igf2r and Kcnq1 loci. In addition to imprinting mechanisms at autosomal loci, what is known about imprinted X-chromosome inactivation and how it compares to autosomal imprinting is also discussed. Overall, this review summarizes many years of imprinting research, while pointing out exciting new discoveries that further elucidate the mechanism of genomic imprinting, and speculating on areas that require further investigation.

S-adenosylhomocysteine treatment of adult female fibroblasts alters X-chromosome inactivation and improves in vitro embryo development after somatic cell nuclear transfer

The poor outcome of somatic cell nuclear transfer (SCNT) is thought to be a consequence of incomplete reprogramming of the donor cell. The objective of this study was to investigate the effects of treatment withS-adenosylhomocysteine (SAH) a DNA demethylation agent, on DNA methylation levels and X-chromosome inactivation status of bovine female fibroblast donor cells and the subsequent impact on developmental potential after SCNT. Compared with non-treated controls, the cells treated with SAH revealed (i) significantly (P<0.05) reduced global DNA methylation, (ii) significantly (∼1.5-fold) increased telomerase activity, (iii) diminished distribution signals of methylated histones H3-3mK9 and H3-3mK27 on the presumptive inactive X-chromosome (Xi), (iv) alteration in the replication pattern of the Xi, and (v) elevation of transcript levels for X-chromosome linked genes,ANT3,MECP2,XIAP,XIST, andHPRT. SCNT embryos produced with SAH-treated donor cells compared with those derived from untreated donor cells revealed (i) similar cleavage frequencies, (ii) significant elevation in the frequencies of development of cleaved embryos to hatched blastocyst stage, and (iii) 1.5-fold increase in telomerase activity. We concluded that SAH induces global DNA demethylation that partially reactivates the Xi, and that a hypomethylated genome may facilitate the nuclear reprogramming process.

Highly expressed genes are associated with inverse antisense transcription in mouse

There is a growing evidence, that antisense transcription might have a key role in a range of human diseases. Although predefined sense-antisense pairs were extensively studied, the antisense expression of the known sense genes is rarely investigated. We retrieved and correlated the expression of sense and antisense sequences of 1182 mouse transcripts to assess the prevalence and to find the characteristic pattern of antisense transcription. We contrasted three Affymetrix MGU74A version 1 mouse genome chips to six MGU74A version 2 chips. For these 1182 transcripts, the version 1 chips contain the antisense sequences of the transcripts presented on the version 2 chips. The original data was taken from the GEO database (GDS431 and GDS432). As the Affymetrix data are semiquantitative, the relative expression levels of antisense partners were analysed. We detected antisense transcription, although the average antisense expression is shifted towards smaller expression values (MGU74A version 1, 516; version 2, 1688). An inverse direct correlation between sense and antisense expression values could be observed at high expression values. At a very high relative expression--above 40,000--the Pearson correlation coefficient is getting closer to -1. Transcripts with high inverse expression ratio may be correlated to the investigated gene (major histocompatibility complex class II trans activator). The ratio of sense to antisense transcripts varied among different chromosomes; on chromosomes 14 and 1 the level of antisense expression was higher than that of sense. We conclude that antisense transcription is a common phenomenon in the mouse genome. The hypothesis of regulatory role of antisense transcripts is supported by the inverse antisense gene expression of highly expressed genes.

Computational analysis reveals 43% antisense transcription in 1182 transcripts in mouse muscle

It is increasingly evident that there is a widespread antisense transcription in the human and other eukaryotic genomes. However, the concept of general antisense expression is rarely investigated. We retrieved and correlated the expression of sense and antisense sequences of 1182 mouse transcripts to assess the prevalence of antisense transcription. We contrasted 20 Affymetrix MGU74A version 1 mouse genome chips to 12 MGU74A version 2 chips. For these 1182 transcripts, the version 1 chips contained the antisense sequences of the transcripts presented on the version 2 chips. The original data was taken from the GEO database. As the Affymetrix data is semi quantitative, the relative levels of antisense partners were analysed. We detected antisense transcription with an overall magnitude of 43% relative to sense transcription in the investigated transcripts. The average MGU74A version 1 expression is shifted towards smaller expression values (MGU74A version 1: 525.1; version 2: 1219.1; t-test: p < 0.001). A direct correlation between sense and antisense expression values could not be observed. Genes with high inverse expression values may be correlated to the investigated condition: genes where sense/control and control/antisense ratios were above two may be included in the pathogenetic pathways associated with dystrophin deficiency. The ratio of sense to antisense transcription varied between different chromosomes. We conclude that antisense transcription is a common phenomenon in the mouse genome and may have indirect regulatory functions.

The complexity of antisense transcription revealed by the study of developing male germ cells

Computational analyses have identified the widespread occurrence of antisense transcripts in the human and the mouse genome. However, the structure and the origin of the majority of the antisense transcripts are unknown. The presence of antisense transcripts for 19 of 64 differentially expressed genes during mouse spermatogenesis was demonstrated with orientation-specific RT-PCR. These antisense transcripts were derived from a wide variety of origins, including processed sense transcripts, intronic and exonic sequences of a single gene or multiple genes, intergenic sequences, and pseudogenes. They underwent normal and alternative splicing, 5' capping, and 3' polyadenylation, similar to the sense transcripts. There were also antisense transcripts that were not capped and/or polyadenylated. The testicular levels of the sense transcripts were higher than those of the antisense transcripts in all cases, while the relative expression in nontesticular tissues was variable. Thus antisense transcripts have complex origins and structures and the sense and antisense transcripts can be regulated independently.

SILENCING OF THE MAMMALIAN X CHROMOSOME

Mammalian X chromosome inactivation is one of the most striking examples of epigenetic gene regulation. Early in development one of the pair of approximately 160-Mb X chromosomes is chosen to be silenced, and this silencing is then stably inherited through subsequent somatic cell divisions. Recent advances have revealed many of the chromatin changes that underlie this stable silencing of an entire chromosome. The key initiator of these changes is a functional RNA, XIST, which is transcribed from, and associates with, the inactive X chromosome, although the mechanism of association with the inactive X and recruitment of facultative heterochromatin remain to be elucidated. This review describes the unique evolutionary history and resulting genomic structure of the X chromosome as well as the current understanding of the factors and events involved in silencing an X chromosome in mammals.

Tsix Silences Xist through Modification of Chromatin Structure

X inactivation is controlled by Xist and its antisense gene, Tsix, neither of which encodes a protein. Xist is essential for X inactivation to occur in cis, and its differential expression is a key event in the initiation of X inactivation. Xist and Tsix are imprinted in the extraembryonic tissues of mouse embryos so that they are expressed from the paternal and maternal X, respectively, resulting in the preferential inactivation of the paternal X. Targeted disruption of Tsix causes ectopic expression of Xist, suggesting that Tsix negatively regulates Xist in cis. However, the molecular mechanism of this antisense regulation remains unknown. Here, we demonstrate that Tsix transcriptionally silences Xist in both embryonic and extraembryonic tissues of mouse embryos. Moreover, we show that disruption of Tsix impairs establishment of repressive epigenetic modifications and chromatin structure at the Xist locus. We propose that Tsix silences Xist through modification of the chromatin structure.

Analysis of the Xist RNA isoforms suggests two distinctly different forms of regulation

The noncoding RNA Xist has been shown to direct the mammalian dosage compensation pathway. Expression of the Xist RNA is regulated through an uncharacterized post-transcriptional mechanism, thought to involve Xist RNA stability. We have previously demonstrated that Xist RNA isoforms contain different 3' ends. In this report we analyze the expression patterns of Xist RNA isoforms and show the Xist RNA long form (L-isoform) is the predominant form in early development. Significant amounts of both the short form (S-isoform) and the L-isoform were found in the female soma. We also define the precise sequence structure of the Xist RNA isoforms 3' ends and show the S-isoform and the L-isoform are structurally dissimilar. Our data show both the S-isoform and L-isoform are cleaved from the same primary transcript. However, the S-isoform is subsequently post-transcriptionally polyadenylated, while the L-isoform is not post-transcriptionally polyadenylated. Sequence organization of the L-isoform shows that there are at least five different nonadenylated L-isoforms in the female soma and only one in embryonic stem (ES) cells. This stem cell-and somatic cell-specific processing may suggest a role for Xist RNA processing in the regulation of Xist RNA expression.

A conserved noncoding intronic transcript at the mouse Dnm3 locus

A 6-kb antisense transcript (Dnm3os) contained within an intron of the mouse Dnm3 gene has been identified in a screen for genes that may be regulated by the basic helix-loop-helix transcription factor Twist during mouse development. The antisense transcript is highly conserved between vertebrate species, but does not appear to encode a protein. We show that expression of the Dnm3 and Dnm3os transcripts overlaps during embryogenesis and in adult tissues, except that Dnm3 is most highly expressed in adult brain and testis and expressed at lower levels in embryos, whereas the antisense transcript is most strongly expressed in embryos and gravid uterus. Both Dnm3 and Dnm3os are downregulated in branchial arch tissue of Twist-null embryos. The conservation and restricted expression pattern of this noncoding transcript suggest that it may perform an important function during embryonic development and further suggest a connection between the regulation of Dnm3 and this novel intronic, antisense transcript.

Epigenetic mechanisms in memory formation

Discoveries concerning the molecular mechanisms of cell differentiation and development have dictated the definition of a new sub-discipline of genetics known as epigenetics. Epigenetics refers to a set of self-perpetuating, post-translational modifications of DNA and nuclear proteins that produce lasting alterations in chromatin structure as a direct consequence, and lasting alterations in patterns of gene expression as an indirect consequence. The area of epigenetics is a burgeoning subfield of genetics in which there is considerable enthusiasm driving new discoveries. Neurobiologists have only recently begun to investigate the possible roles of epigenetic mechanisms in behaviour, physiology and neuropathology. Strikingly, the relevant data from the few extant neurobiology-related studies have already indicated a theme - epigenetic mechanisms probably have an important role in synaptic plasticity and memory formation.

Recent advances in X-chromosome inactivation

X inactivation is the silencing one of the two X chromosomes in XX female mammals. Initiation of this process during early development is controlled by the X-inactivation centre, a complex locus that determines how many, and which, X chromosomes will be inactivated. It also produces the Xist transcript, a remarkable RNA that coats the X chromosome in cis and triggers its silencing. Xist RNA coating induces a cascade of chromatin changes on the X chromosome, including the recruitment of Polycomb group proteins. This results in an inactive state that is initially labile, but may be further locked in by epigenetic marks such as DNA methylation. In mice, X inactivation has recently been found to be much more dynamic than previously thought during early pre-implantation development. The paternal X chromosome is initially inactivated in all cells of cleavage-stage embryos and then selectively reactivated in the subset of cells that will form the embryo, with random X inactivation occurring thereafter.

Barring gene expression after XIST: maintaining facultative heterochromatin on the inactive X

X chromosome inactivation refers to the developmentally regulated process of silencing gene expression from all but one X chromosome per cell in female mammals in order to equalize the levels of X chromosome derived gene expression between the sexes. While much attention has focused on the genetic and epigenetic events early in development that initiate the inactivation process, it is also important to understand the events that ensure maintenance of the inactive state through subsequent cell divisions. Gene silencing at the inactive X chromosome is irreversible in somatic cells and is achieved through the formation of facultative heterochromatin (visible as the Barr body) that is remarkably stable and faithfully preserved. Here we review the many features of inactive X chromatin in terminally differentiated cells and address the highly redundant mechanisms of maintaining the inactive X chromatin.