This week we profile a recent publication in Nature Communications from
Dr. Matthew Lorincz (fourth from left) at the UBC Life Sciences Institute.
Can you provide a brief overview of your lab’s current research focus?
Research in the Lorincz lab is directed towards understanding the interplay between transcription, DNA methylation and histone modifications in early development and in the germline, using the mouse as a model system. Employing conventional genetic knockouts and CRISPR/Cas9 genome editing, we are currently studying the roles of several histone modifying enzymes (methyltransferases and demethylases implicated in developmental disorders as well as cancer) in directing DNA methylation and/or regulating gene expression in the early embryo and male germ cells. We are also employing CRISPR/Cas9 to study the role of specific retroelements sequences – so called “junk DNA” – in the transcriptional regulation of nearby genes. These studies are made possible by the application of low-cell input methods for whole genome analyses, including of chromatin marks using ultra-low input ChIPseq (ULI-ChIP-seq), developed in the lab, DNA methylation (PBAT) and transcription (RNAseq), as well as bioinformatic pipelines developed in house to integrate the analyses of these epigenomic datasets, including at an allele-specific level. Ultimately, we hope that these studies will shed light on the basic principles of gene regulation during normal development, with clear implications for disease states in which such gene regulatory pathways are disrupted, including in cancer and genetic disorders associated with intellectual disability.
What is the significance of the findings in this publication?
In this study, we characterized the role of a specific class of repetitive elements in the genome, known as long-terminal repeat (LTR) retrotransposons, in directing the establishment of DNA methylation in oocytes, and how they may influence the expression of nearby genes in the offspring. Notably, a subset of LTR elements are particularly active, ie transcribed at high levels, in both rodent and human oocytes. To study the implications of expression of such parasitic elements and their relics, we utilized genome-wide approaches to study the relationship between such LTR-initiated transcription units (LITs), the transcription-coupled histone mark H3K36me3, DNA methylation and transcription of nearby genes in mouse, rat and human oocytes, as well as between distantly related strains of mice. Comparison of these datasets revealed a surprising level of species-specific transcription of LITs and associated DNA methylation, including over many nearby genes and their regulatory elements, known as CpG islands. We show that such DNA methylation may persist on the maternal genome after fertilization, inhibiting the expression of nearby genes exclusively on the maternal allele in the early embryo in mice and placental tissues in human and for a subset of genes in both species in adult tissues. Expression exclusively from the paternal allele likely influences the protein levels expressed from these genes, which likely influences the phenotype of the offspring. As many of the observed LITs are species-specific, this phenomenon may explain phenotypic differences observed between species and even with a species, such as in mice, where these LTR retrotransposons are particularly active and occupy distinct genomic regions depending on the mouse strain.
What are the next steps for this research?
The phenomenon of allele-specific expression in mammals has been studied for some time, with a particular focus on “imprinted” genes, which by definition are expressed exclusively from one parental allele, generally in association with DNA methylation of the promoter region of the silent parental allele. For a subset of such imprinted genes, DNA methylation is established in the gametes and maintained in somatic cells. Disruption of this process is associated with many diseases, including developmental disorders such as Beckwith-Wiedemann and Angelman syndromes, as well as in cancers, such as Wilms’ tumor. We have found several cases of genes imprinted exclusively in primates likely as a consequence of primate-specific LITs. In collaboration with Louis Lefebvre’s lab at UBC, we are currently characterizing the role of a specific set of LTR-initiated transcripts in the establishment of imprinting in mouse oocytes, using CRISPR/Cas9 genome editing to delete the LTRs implicated at two candidate rodent-specific imprinted genes. As the transcriptionally active LTRs upstream of these genes are rodent specific, we hypothesize that integration of such LTRs played a critical role in the conversion of these genes from biallelically expressed to imprinted in the rodent lineage. This study will directly address whether these maternally imprinted genes require the transcription of upstream retrotransposons in the female germ line, with clear implications for the genesis imprinting in mammals.
This research was funded by:
Work this study was supported by grants from CIHR, Genome BC and NSERC. Julie Brind’Amour, lead author of the manuscript, was supported by a postdoctoral fellowship from the MSFHR.
