X-chromosome Inactivation, Epigenetics and the Transcriptome

The X chromosome is a cornerstone to the pathogenesis of a number of syndromes, whereof some are Turner syndrome (45,X), Klinefelter syndrome (47,XXY), triple X syndrome (47,XXX) and double Y syndrome (47,XYY). Despite this importance to clinical disease, only one gene on the X chromosome has so far been implicated in the wide spectra of phenotypic traits seen in these and other X-related syndromes. The one known gene is the SHOX (the short stature homeobox) gene and encodes a transcription factor that has brain natriuretic peptide (BNP) and fibroblast growth factor receptor gene (FGFR3) as transcriptional targets. It is located at the pseudoautosomal region of the X and Y chromosomes. This gene has been shown to be involved in short stature in Turner syndrome, Leri-Weill syndrome and idiopathic short stature. It also causes the increased stature in Klinefelter syndrome, triple X syndrome and XYY syndrome.

A number of traits and diseases are seen frequently in X-chromosomal syndromes that cannot be explained by this SHOX gene. The best characterized of these syndromes is Turner syndrome, where these phenotype traits can be divided into:

1. Autoimmune predilection, which leads to an increased risk of virtually all autoimmune diseases of unknown pathogenesis such as diabetes and hypothyroidism.

2. Decreased intrauterine viability. Here haploinsufficiency of X-linked pseudoautosomal genes operating in the placenta has been suggested to be involved (STS and CSF2RA).

3. Ovarian dysgenesis, leading to ovarian insufficiency and the need for long term sex hormone replacement therapy.

4. Congenital cardiovascular malformations of unresolved pathogeneses.

5. Brain development, especially social-cognitive development, which is altered in many cases, often in a more "male-like" direction.

6. Increased prevalence of the metabolic syndrome and osteoporosis. In healthy women's cells, with two X-chromosomes, random X inactivation takes place (13). The process is governed by the X inactivation center (XIC) and initiated by Xist that is a gene encoding a long intervening non-coding RNA (lincRNA). The Xist gene is located close to the centromere on the long arm of the X chromosome, where from it orchestrates repressive histone modifications (recruiting PRC2) along the X chromosome leading to inactivation. In the remaining active X chromosome PRC2 is titrated away by Tsix, which effectively leaves all females as mosaics for the X chromosome with one of maternal and one of paternal origin. However, a great number of genes that are spread out on the X chromosome escape this X-inactivation by unknown mechanisms and dosage compensation takes place, so that expression between males and females are comparable for many genes (15, 16). Approximately 65% of genes are fully silenced, while 15% completely escape X-inactivation, and 20% show variable expression, depending on tissue cell origin (17).

LincRNAs are pervasively transcribed in the genome, although their role in health and disease is poorly understood. Studies of dosage compensation, imprinting and homeotic gene expression suggest that lincRNAs function at the interface between DNA and chromatin remodeling with further involvement in reprogramming of chromatin to promote cancer metastasis. To date a range of different interactions have been hypothesized for lincRNAs in transcriptional regulation, and they may function both as intact interacting molecules as well as Dicer processed molecules that are chopped into small interfering RNAs that degrade other RNAs.

Chromatin remodeling can be analyzed by the marks left by histones on the DNA strand, which can be of either permissive or repressive nature, depending on the acetylation or methylation taking place of the histones. As an example, trimethylation of lysine 4 on histone H3 (H3K4me3) is enriched at transcriptionally active gene promoters, whereas trimethylation of H3K9 (H3Kme3) and H3K27 (H3K27me3) are present at gene promoters that are transcriptionally repressed. By use of chromatin immunoprecipitation coupled with deep sequencing (chIPseq) one can obtain these marks along the whole X chromosome in one assay.

The epigenetic alterations of histone modifications can be studied by a new methodology, enabling the use of relatively old pathological specimens. This opens new prospects for expansion of our knowledge of the role of the X-chromosomal permission and inactivation to different diseases, where X-chromosomal syndromes may serve as the initial model to understand such processes that are highly likely to be important to diseases (e.g. diabetes and hypothyroidism) beyond these syndromes. As another example, congenital malformations of the heart are frequent in Turner syndrome and often lead to early aortic dilatation and dissection. In these patients and in controls, we collect paraffin- embedded blocks of tissue from the aortic wall, which can now be assessed using this frontline methodology with a potential to identify novel marks on Turner patients DNA compared to the DNA of non-Turner patients.

Imprinting is another important aspect of sex chromosome action. Imprinting refers to the process where a gene (or more genes) may be imprinted depending on parental origin. Put another way, a gene can be "turned on or off" depending on its maternal of paternal origin. Furthermore, mouse studies show that clusters of genes on the X chromosome are imprinted and are independent of X chromosome inactivation.

The importance of the biological inheritance is apparent for the major cardiovascular morbidities affecting the population, where a hereditary trait clearly prevails in certain families. Despite a promise for targeting the prevention and treatment of cardiovascular morbidity, the specific parts of the genome that potentially trigger the pathologies largely remain to be defined, and could bring important knowledge of the pathophysiology.

The major body of knowledge on the implications of genome aberrations originates from diseases with obvious and severe manifestations resulting from clear modes of transmission that allow identification of the causative regions of the genome. Such genetic disorders hold the potential for understanding the role of a specific locus of the genome, if this can be identified, as large chromosomal regions often are involved. In the case of the X- chromosomal phenotypes we expect the causative agent to be on the X chromosome, and will use various novel technologies to identify this agent.

Currently, our limited knowledge of the importance of the X-chromosome to cardiovascular pathology comes from single-gene disorders, and more non-specific gender differences in addition to the sex chromosomal anomalies. In contrast, no single-gene disorder on the Y- chromosome has been established to be related to cardiovascular morbidity.

Appropriate human models for improved understanding of the role played by the sex chromosomes are available. Here, deviations from normality not only occur at a reasonable prevalence but also associate with readily identifiable phenotypes and adverse prognosis. Turner and Klinefelter syndromes constitute such models; females with a reduction in X- chromosomal material and males with an increase in X-chromosomal material, respectively. These anomalies of the sex-chromosomes associate with excess morbidity and mortality from both congenital and acquired cardiovascular as well as diabetes, ovarian insufficiency and other diseases.

The cardiovascular phenotypes and the expression and activation of genes are investigated in healthy females and males with a comparison between Turner and Klinefelter syndromes in a cross-sectional descriptive design. These studies have already been performed and a precise characterization is established. The hypothesis is that the significance of the X-chromosome will manifest as altered levels of expression and activation in association with different cardiovascular phenotypes. Secondarily, basic analogous knowledge is provided of the Y-chromosome. The project is expected to generate further hypotheses on the role played by the genome to morbidity in both the population having a normal karyotype as well as in abnormal karyotypes.

In this project we will provide a unique combination of front line molecular technologies and well defined patient cohorts. The hypotheses we will test are the following:

1. Non-coding transcripts from the X chromosome play a fundamental role in sex chromosome abnormalities, and may work through regulation of epigenetic mechanisms and through mRNA destabilization

2. The regulation of non-coding RNA expression on X-chromosomes is based on epigenetic mechanisms that lead to different histone marks, and different DNA methylation in e.g., Turner and Klinefelter syndrome persons when compared to healthy gender-matched controls.

3. The gene expression pattern resulting from these mechanisms is different in sex chromosome abnormalities in comparison with healthy males and females, and this difference can be studied in diseased tissues from Turner syndrome women and compared to normal control tissue.

4. It may be possible to identify one or a few driver molecules in diseased tissues from Turner and Klinefelter syndrome persons, that can be validated in vitro and in vivo and that may explain the disease processes, giving important pathophysiological information.

Expected findings. We expect to be able to define the epigenetic changes at the X-chromosomes at a single base resolution, thus identifying CpG methylation at the DNA strands as well as permissive and repressive histone marks in histones.

We expect to identify the transcriptome both regarding mRNA and non-coding RNAs (long as well as microRNAs) for RNAs generated from the X-chromosome.

We expect to be able to provide an Atlas of the epigenetic events specific for Turner syndromes and the effects of these on the transcriptome.

Using bioinformatic methods this will hopefully lead to identification of novel dysregulated molecules that may explain various properties of these patients. These molecules will then be subject to validation in separate patient cohorts using PCR or IHC technology.

In diseased tissue we will study the tissue specific alterations of the epigenome and transcriptome of the X chromosomes and compare this to normal tissues from the control samples. We hope this will lead to identification of the drivers of the disease process and a pathophysiological understanding of the disease process.