Mechanistic characterisation of enhancer hijacking: identifying essential and targetable chromatin interactions
Lead Research Organisation:
Newcastle University
Department Name: Biosciences Institute
Abstract
Our genetic code is like a recipe book: we need instructions and ingredients to produce the cells within our body. The ingredients are called genes, and in healthy cells, enhancers provide the instructions that say which genes are to be switched on and when. In patients with blood cancers, some enhancers get moved to new locations within the genetic code, and because of this they end up switching on the wrong gene. This is known as 'enhancer hijacking' and has been shown to contribute to the development of cancer. To understand how enhancers instruct the wrong genes we need to learn about the communication between enhancers and genes in both healthy and cancer cells. We also need to shut down these enhancers to prevent communication and record the response of the cells. By stopping enhancers communicating with the wrong genes, we hope to kill the cancer cells.
We can edit the genomes of cancer cells in the laboratory, making alterations to regions within enhancers to shut them down. This can help us understand how enhancer hijacking works, and develop new treatments to reverse the effects of the hijacking (i.e., to stop the wrong genes from being switched on). However, identifying the most important parts of the enhancers is difficult, and the laboratory experiments are time consuming and expensive. In this project, we will develop and test a new computer programme which can identify these important sites, and predict the effects of the genome editing experiments. It will be possible to quickly mimic an experiment using the computer programme, trying many different scenarios before going into the lab. This will accelerate and improve the targeting of the experiments, saving time and money, and revolutionise the way we design our experiments. As well as this, the computer models will provide new insight and understanding of how enhancers switch on the wrong genes to cause disease, and how genes and enhancers communicate more generally.
This work is important because the changes to the genetic code that we are interested in are found in patients who do not respond well to current treatments. By understanding how enhancers communicate with genes, we can look to block this interaction. In the longer term, this will enable us to develop more specific treatments that only target the cancer cells, reducing side-effects of treatment and improving the lives of those living with cancer.
Our team brings together researchers with very different skill sets - laboratory-based cancer biology and computational biophysics. Having both of these aspects will be crucial for taking this exciting work forward. We have successfully worked together for over three years and have already published new findings for the scientific community. The experimental protocols that we will use are already established within the group and the computer model that we plan to build on has already proved successful in multiple projects. By continuing this successful collaboration, together with our project partners, we hope to make significant contributions to knowledge about how hijacked enhancers communicate with genes. If successful this approach could be applied to many different cancers which involve enhancer hijacking.
We can edit the genomes of cancer cells in the laboratory, making alterations to regions within enhancers to shut them down. This can help us understand how enhancer hijacking works, and develop new treatments to reverse the effects of the hijacking (i.e., to stop the wrong genes from being switched on). However, identifying the most important parts of the enhancers is difficult, and the laboratory experiments are time consuming and expensive. In this project, we will develop and test a new computer programme which can identify these important sites, and predict the effects of the genome editing experiments. It will be possible to quickly mimic an experiment using the computer programme, trying many different scenarios before going into the lab. This will accelerate and improve the targeting of the experiments, saving time and money, and revolutionise the way we design our experiments. As well as this, the computer models will provide new insight and understanding of how enhancers switch on the wrong genes to cause disease, and how genes and enhancers communicate more generally.
This work is important because the changes to the genetic code that we are interested in are found in patients who do not respond well to current treatments. By understanding how enhancers communicate with genes, we can look to block this interaction. In the longer term, this will enable us to develop more specific treatments that only target the cancer cells, reducing side-effects of treatment and improving the lives of those living with cancer.
Our team brings together researchers with very different skill sets - laboratory-based cancer biology and computational biophysics. Having both of these aspects will be crucial for taking this exciting work forward. We have successfully worked together for over three years and have already published new findings for the scientific community. The experimental protocols that we will use are already established within the group and the computer model that we plan to build on has already proved successful in multiple projects. By continuing this successful collaboration, together with our project partners, we hope to make significant contributions to knowledge about how hijacked enhancers communicate with genes. If successful this approach could be applied to many different cancers which involve enhancer hijacking.
Technical Summary
Annually in the UK, over 7300 people are diagnosed with haematological malignancies where, after genome rearrangements, super-enhancers (SEs) regulating immunoglobulin or T-cell receptor expression mistakenly pair with proto-oncogenes, converting them to oncogenes. Such rearrangements are accompanied by local changes to the chromatin and epigenome, with expression driven by looping between gene promoters and specific 'expression driving regions' (EDRs) within SEs. This 'enhancer hijacking' is poorly understood.
EDRs can be characterised via CRISPR silencing, however these experiments can only target ~100bp sites, while SEs can be up to 100,000 bp long. Systematic targeting of many candidate sites is prohibitively expensive and time consuming. Chromosome-conformation-capture (3C) experiments provide data on looping interactions, but require large high-quality samples that are challenging to obtain from cancer patients.
In this project we will use a combined experimental and computer modelling approach to study enhancer hijacking. Our polymer physics-based model was previously shown to give good predictions of chromatin interactions and gene expression based on minimal data. Using the model, we will predict EDR positions in malignancy-derived cell lines and PDX models, confirm these via targeted 3C experiments, and use that data to improve the model. We will extend the model to include epigenomic dynamics, and then simulate CRISPR experiments we will undertake to silence EDRs. This will give insight on chromatin 3D-structural and epigenetic dynamics that would be difficult to obtain from experiments alone, providing new mechanistic understanding of enhancer hijacking and its broader effect on chromatin organisation and regulation. We aim to develop a model to simulate the evolution of the epigenome from a healthy to malignant state after a genome rearrangement: in principle this could predict the effects of any rearrangement using only data from healthy cells.
EDRs can be characterised via CRISPR silencing, however these experiments can only target ~100bp sites, while SEs can be up to 100,000 bp long. Systematic targeting of many candidate sites is prohibitively expensive and time consuming. Chromosome-conformation-capture (3C) experiments provide data on looping interactions, but require large high-quality samples that are challenging to obtain from cancer patients.
In this project we will use a combined experimental and computer modelling approach to study enhancer hijacking. Our polymer physics-based model was previously shown to give good predictions of chromatin interactions and gene expression based on minimal data. Using the model, we will predict EDR positions in malignancy-derived cell lines and PDX models, confirm these via targeted 3C experiments, and use that data to improve the model. We will extend the model to include epigenomic dynamics, and then simulate CRISPR experiments we will undertake to silence EDRs. This will give insight on chromatin 3D-structural and epigenetic dynamics that would be difficult to obtain from experiments alone, providing new mechanistic understanding of enhancer hijacking and its broader effect on chromatin organisation and regulation. We aim to develop a model to simulate the evolution of the epigenome from a healthy to malignant state after a genome rearrangement: in principle this could predict the effects of any rearrangement using only data from healthy cells.