How does the expression of one gene affect that of its neighbour?

A strand of DNA can contain several genes. The expression of one gene may affect the expression of its neighbouring genes. We see this occurring in many circumstances in normal and disease biology. Understanding exactly how this happens is important for appreciating the knock-on effects of genome rearrangements (i.e when DNA is broken and repaired with the result that genes are repositioned) and transgene insertions (as in gene therapy, where a new DNA sequence is inserted into the genome), and hence for understanding disease progression. It will also increase our understanding of how viruses that integrate into the genome and lie dormant can be reactivated through expression changes in nearby genes.

In this proposal we address how such non-autonomous gene expression occurs. As examples we are using the DIRAS3 gene and its neighbour GNG12-AS1. These two genes face each other on the DNA and they are a considerable genomic distance apart, yet the expression of GNG12-AS1 interferes with the expression of DIRAS3, with the result that DIRAS3 expression is lowered. This process is known as "transcription interference". We have recently found that we can epigenetically manipulate GNG12-AS1 expression without mutating or modifying the underlying DNA sequence. This provides us with the means to test two models for transcription interference. The first model considers the two enzymes which enable transcription (polymerases) travelling towards one another and because they cannot pass each other, one or both need to pause to prevent a polymerase collision. The second model considers the expression of one gene simply blocking the accessibility of the other gene to polymerases.

In our experiments we will take into consideration the fact that DNA is folded into looping structures, so that two genes separated by a big genomic distance may in fact be in close proximity, and thus more likely to compete for polymerase when the DNA is folded into loops. We will also track the progress of polymerases between these genes to map pausing positions.

The results of this study will give a deeper understanding of how the genome functions to regulate itself. Emerging technologies for gene editing are providing unprecedented opportunities for gene therapy to treat many diseases. Epigenetic drugs already exist that change gene expression without changing the underlying DNA sequence. The opportunities for epigenetic- and gene therapy make it imperative to better understand the mechanisms of non-autonomous gene expression. The models that we are testing have different implications for epigenetic and gene therapies and will allow us to predict how the genome environment will respond to these therapies.

Classically it has been presumed that genes are autonomous in their expression. However, there is increasing evidence that in both health and disease, gene expression is not autonomous. In all eukaryotic genomes highly co-expressed genes and housekeeping genes cluster. In cell fate specification, blocks of up or down regulated genes are observed and in cancers we see unusual blocks of genes downregulated. Importantly, the first gene therapy trials had to be stopped as the expression of the introduced gene affected the expression of an oncogenic neighbour. What is less clear is what the underlying mechanisms are for such non-autonomous expression.

Here we take advantage of an emerging model system for the analysis of gene-gene interactions, the DIRAS3/GNG12-AS1 expression in humans. We have discovered that DIRAS3 transcription is modulated through transcriptional interference (TI) from an lncRNA, GNG12-AS1. We have found that we can inhibit GNG12-AS1 either at its transcription level or post-transcriptionally by selective siRNA targeting. We intend to use this system to test two models for TI, one that considers polymerase collision/pausing, and another that transcription of one locus alters promoter accessibility at the other. The techniques that we will use will be allele-specific capture chromatin conformation (C-HiC) and Nuclear run assays (GROseq), R-loop and RNAPII ChIP in combination with gene editing to examine the effects of transcription through specific genomic features on TI.

The results of these studies will lead a better understanding of non-autonomous gene expression, which is key to understanding the potential pitfalls of gene therapy, the effects of genome rearrangements (in cancers or in evolutionary time) as well as guide epigenetic therapies aimed at reversibly modulating gene expression.

Other than the academic beneficiaries listed above, potential users of this research will ultimately be clinicians and pharma with an interest in gene therapy, epigenetics and diagnosing patients with genomic imprinting disorders. The patients that are likely to finally benefit include those with congenital gene defects and imprinting defects, as well patients with acquired polyfactorial disease (type 2 diabetes, cancer, cardio vascular disease and degenerative neurological diseases) who would be candidates for gene or epigenetic therapies.

We will reach these clinical stakeholders by targeting specific professional organisation bodies (British Society for Gene and Cell Therapy BSGCT, European Network for Congenital Imprinting Disorders; Wellbeing of Women charity) and contribute to their meetings to publicise our project.

We will also include specific Biotech companies that interact with these professional organisations in the conferences we organise (Via CR@B and the Biochemical Society) and arrange for them to participate in our student placement schemes.

Our research topics (genome evolution, genomic imprinting, the clashing of polymerases) are very accessible to members of the public with an interest in science and its application to disease. An ideal venue for public engagement locally is the Bath Royal Literary and Science Institute (BRLSI). The Department has strong links with the BRLSI and have participated in several of their public events. In addition we participate in Science Cafe and a MOOC on "Inside Cancer"