Regulated expression and developmental functions of the H19 long noncoding RNA

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications & trials

Abstract

Imprinting represents a curious defiance of normal Mendelian genetics. Mammals inherit two complete sets of chromosomes, one from the mother and one from the father, and most autosomal genes will be expressed equally from maternal and paternal alleles. Imprinted genes, however, are expressed from only one chromosome in a parent-of-origin dependent manner. Because silent and active promoters are present in a single nucleus, the differences in activity cannot be explained by transcription factor abundance. Thus, the transcription of imprinted genes represents a clear situation in which epigenetic mechanisms restrict gene expression. Therefore, imprinted genes are good models for understanding the role of DNA modifications and chromatin structure in maintaining appropriate patterns of gene expression. Further, because of parent-of-origin restricted expression, phenotypes determined by imprinted genes are not only susceptible to mutations of the genes themselves but also to disruptions in the epigenetic programs controlling regulation. Thus, imprinted genes are frequently associated with human diseases, including disorders affecting cell growth, development, and behavior. Our Section is investigating a cluster of genes on the distal end of mouse chromosome 7. The syntenic region in humans on chromosome 11p15.5 is conserved in genomic organization and in monoallelic expression patterns. Especially, we are focusing on the molecular basis for the maternal specific expression of the H19 gene and the paternal specific expression of the Igf2 gene. Loss of imprinting mutations in these two genes is associated with developmental disorders (including Beckwith Wiedemann Syndrome (BWS) and Russell Silver Syndrome (RSS)), with pediatric cancers (including Wilms tumor and rhabdosarcoma), with cardiomyopathies, and with many adult cancers. Expression of both H19 and Igf2 is dependent upon a shared set of enhancer elements downstream of both genes and upon a 2.4 kb Imprinting Control Region (ICR) that lies just upstream of the H19 promoter. Using conditional deletion and insertional mutagenesis we have identified three functions associated with the ICR. First, this element acts to distinguish the parental origin of any chromosome into which it is inserted. Specifically, the CpGs within this region become hypermethylated upon paternal inheritance. Second, this element functions as a CTCF-dependent, methylation-sensitive transcriptional insulator. By reorganizing the long-range interactions of nearby promoter and enhancer elements, this insulator can direct parental-specific activation of nearby genes. Finally, this ICR also acts as a developmentally regulated silencer element when paternally inherited. Specifically, the methylated ICR induces changes in chromatin structure of neighboring sequences that impacts gene expression. Our current goals are to identify and characterize the protein factors and non-coding RNAs that interact with the ICR and establish the chromatin structures associated with the maternal and paternal chromosomes. We are addressing these issues both in germ cells, where the imprints are established, and in somatic tissues where expression of Igf2 and H19 are most critical for normal, healthy cell function. We are also working to establish mouse models that mimic the Beckwith Wiedemann syndrome phenotypes associated with maternal loss of imprinting at the Igf2/H19 locus in humans. We have demonstrated defects in muscle cell differentiation and in muscle regeneration in cells where Igf2/H19 imprinting is disrupted. We have demonstrated that even a <2-fold increase in Igf2 expression will result in large-scale disruption in cell cycle regulation by hyperactivation of the MAPK pathway. In addition, decreased expression of H19 disrupts normal regulation of p53 in muscle cells so that they can no longer respond to Wnt stimulation and therefore do no undergo normal hypertrophy. Thus, loss of imprinting of both H19 and Igf2 genes are relevant to overgrowth phenotypes in BWS More recently we have characterized cardiac dysfunction phenotypes in these maternal loss of imprinting mice. During early development, extra expression of Igf2 results in physiologic hypertrophy. However, hypertrophy diminishes after birth (when Igf2 expression stops) and there are no long-term health consequences. However, loss of the H19 lncRNA results in pathological hypertrophy and reduced cardiac function that progresses in the postnatal heart. Genetic analyses indicate that H19 prevents premature endothelial to mesenchymal transition. In the absence of H19, endothelial cells mis-express mesenchymal markers and adult mice show significant fibrosis. Using CRISPR-Cas9 technologies, we have generated novel mouse strains that carry mutations in specific H19 domains. These analyses demonstrate that H19 sequences that interact with let7 microRNAs are necessary to prevent cardiac fibrosis and functional defects. The primary product of the H19 gene is a 2.2 kb long noncoding RNA (lncRNA). A top goal of our research tis to understand the molecular and biochemical functions of this RNA. Using in vitro models, we discovered a critical for H19 RNA in mediating cellular stress responses through physical interactions with p21 mRNA molecules that regulate p21s stability and translation efficiency. In brief summary, mice lacking H19 are more likely to respond to stress by activating senescence pathways. In addition to our work regarding genomic imprinting, a secondary research goal is to generate mouse models for cardiac arrhythmias. Most recently, we have generated mouse models for Calsequestrin2 deficiency. We demonstrated that calsequestrin2 is not essential for cardiac calcium ion storage. Rather, the primary function of calsequestrin appears to be the regulation of the SR calcium ion release channel during conditions of beta-adrenergic stimulation. The loss of calsequestrin2 thus results in premature calcium ion release from the SR, leading to voltage changes that result in premature contraction of cardiomyocytes and thus arrhythmia. The validity of this mouse model has been recently confirmed by demonstration that drugs that we used to successfully ameliorate the mouse arrhythmias were highly effective in pilot studies on human patients. In the past several years, we have demonstrated that mouse arrhythmias associated with calsequestrin2-deficiency worsen significantly with age. This age-dependent increase in cardiac phenotypes had already been known to occur in humans. We are now completing genomic analyses to identify genes and pathways that are dysregulated specifically in older mice where arrhythmia phenotypes are strongest.