DNA Misfolding and the Maintenance of Genome Stability: an Integrated Molecular, Cellular and Genomic Investigation of DNA Double-Strand Break Repair

DNA is essential for all living organisms as it encodes the information needed for growth, survival and inheritance. Chromosomes contain this genetic information in very long molecules of DNA (often more that several million molecular building blocks, known as base-pairs, in length). One consequence of this great length is fragility of the molecule. Therefore, cells need efficient systems to repair broken DNA. One single unrepaired DNA break in a chromosome leads to cell death. Not only does a break need to be repaired efficiently but it must also be repaired accurately. Inaccurate repair leads to genetic alterations that cause cancer or genetic disease. Furthermore, DNA breakage is used to kill cancer cells in treatments such as radiotherapy and chemotherapy. The break repair pathways also offer targets for the development of new antibiotics. Understanding how DNA molecules are broken and how they are repaired is fundamentally important to inform future clinical practice.

Interestingly, not all DNA sequences are equally susceptible to breakage. Due to their inverted-repeat nature, palindromic DNA sequences can misfold in chromosomes and these misfolded structures have an elevated probability of breakage. DNA palindromes are known to be locations in chromosomes associated with human genetic disease in foetuses and cancer in adults. We know also that this association with disease is likely to be caused by DNA breakage. In bacteria (such as E. coli), we have discovered that a specific protein, known as SbcCD (also present in humans where it is known as Rad50/Mre11), is responsible for generating DNA breaks at palindromic sequences. However, these breaks are very efficiently and accurately repaired in a reaction called homologous recombination, using the genetic information on a second unbroken copy of the chromosome present in the same cells.

We will use this system of DNA misfolding and cleavage (and another system that uses molecular scissors) to generate repairable DNA breaks at a specific chromosomal location and study their repair by homologous recombination. We will concentrate on the mechanism of repair of DNA breaks in living cells. This is the area where the least is known and where our research can have the greatest impact. Many years of genetic investigation have identified the genes involved in homologous recombination and the proteins that they encode have been studied in the test tube. However, the experimental systems to study this reaction in living cells and the methods to carry out this analysis have recently progressed to a stage where substantial progress can be made. Using microscopy, we will actually look at the repair reactions, as they are happening in live cells. We will also isolate and analyse the DNA molecules "caught in the act" of repairing breaks. Finally we will observe the indirect consequences of DNA break repair on the duplication of chromosomes and their distribution into the next generation of cells.

Our sophisticated systems for making DNA breaks and analysing their repair, combined with the wealth of knowledge of E. coli as an experimental organism and its intrinsic advantages (such as small size and rapid growth rate) will be used to make more rapid progress than it is possible with more complex organisms. Not only will this work elucidate how DNA breaks are actually repaired in living E. coli (an important bacterial pathogen) but it will be a first step towards implementing similar experimental strategies in human cells. An example will help to illustrate this. We intend for the first time to isolate the DNA from a single DNA repair site on the chromosome and analyse this DNA by electron microscopy. This will not be trivial even for a chromosome of 4.6 million base pairs in length. However if we succeed, we will have pioneered the way towards a similar goal for the DNA in a human cell where a specific repair event will have to be isolated from over 4 billion base pairs of DNA.

We will:

1. Compare and contrast recombination between sister chromatids in replication-dependent DSBR with recombination between sister chromatids in replication-independent DSBR.

2. Determine the molecular structures implicated in chromosomal DSBR by recombination and to relate these structures to the genomic consequences of DSBR.

We will use three experimental approaches to integrate our understanding of DSBR at cellular, molecular and genomic levels. We will:

A. Use live cell imaging to visualise DSBR. This will include determining the cellular location of events, dynamic movements of proteins and DNA and timing of reactions.

B. Isolate and determine the structures of DSBR intermediates at a single chromosomal locus. This will involve multidimensional gel electrophoresis and electron microscopy of DNA molecules.

C. Use genomic technologies such as ChIP-seq and Solexa sequencing to obtain quantitative data on the consequences of DSBR at a genomic scale. We will couple these data with stochastic modelling in collaboration with Dr El Karoui and Prof. Danos to derive mechanistic information from the data. Throughout these studies, we will use mutants that inactivate or modify the activities of key recombination proteins.

Our previous MRC programme funding places us in the unique position of having developed sophisticated E. coil systems for the accomplishment of these aims. These include a way of generating replication-dependent DSBs at the site of a chromosomal palindrome cleaved by SbcCD and a way of generating replication-independent DSBs based on I-SceI cleavage. The breaks generated by both system are fully repairable. We have also developed systems for the analysis of DSBR including: live cell imaging, pulsed field gel electrophoresis, 2D native-native gel electrophoresis, ChIP-seq and Solexa DNA sequencing. In the course of the new programme, we will add 2D native-denaturing gel electrophoresis and electron microscopy to our toolkit.

Progress in science requires fundamental understanding, and our work is aimed at pushing back the frontiers of knowledge. It is a privilege and joy to work with E. coli. Important contributions of E. coli research include the discoveries of key genes in DNA repair pathways and the biochemical properties of the proteins they encode. Two examples are the discovery of the recA gene and the elucidation of the function of the RecA protein that mediates homologous recombination and the discoveries of the genes implicated in mismatch repair (mutS, mutL etc.) and the activities of their cognate proteins. In both cases, human homologues exist that are implicated in cancer and genetic disease. In our own work, the discovery that the sbcCD genes encode a hairpin endonuclease opened up a field of work on the activities and functions of Rad50/Mre11, the human homologue that plays key roles in DNA repair (e.g. hairpin cleavage, removal of covalent protein-DNA complexes, initiation of resection in recombination, checkpoint signalling, DNA end-tethering etc.). Through this influence of fundamental discovery on the progress of science, I am confident that our new work will lead to understanding that will illuminate similar processes happening in human cells. Given that our work focuses on DNA repair, the beneficiaries will include individuals who currently suffer from genetic diseases or cancer where DNA repair plays an important role in the disease or treatment.

In addition to being a model system for fundamental science, strains of E. coli exist that cause serious human disease and death, with devastating economic impact in the event of outbreaks. At present, antibiotics are becoming increasingly ineffective at treating bacterial diseases as resistance to current drugs increases. It is important to develop new drugs that kill bacteria or that can be used in combination with "old" antibiotics to enhance their efficacy. The pathway of DSBR by recombination offers one such line of attack. For example, we have preliminary evidence that inactivation of recombination can enhance the action of antibiotics that do not specifically target DSBR (e.g. antibiotics such as streptomycin and chloramphenicol). Furthermore, sub-lethal doses of several different antibiotics induce recombination and mutagenesis via the SOS response and thereby accelerate the evolution of antibiotic resistance (Andersson and Hughes, 2014, Nature Review, Microbiology). Limiting the recombination signals that induce the SOS response by drug treatment may slow the evolution of antibiotic resistance. If inactivation of recombination can both enhance the action of other antibiotics and limit the evolution of resistance, drugs targeting recombination may become more important than generally appreciated. Clearly in order to design effective therapies, it is firstly important to understand the mechanisms of DNA repair. Our work is part of an essential maintenance of knowledge in basic bacterial microbiology upon which the development of new antibacterial therapies is critically dependent. In this way, our work will benefit individuals suffering from bacterial infections and economic productivity through minimising the devastating consequences of epidemic outbreaks.

Since our work is at the fundamental end of the scientific spectrum, impacts are long-term and often the biggest impacts come from unexpected discoveries that are by definition impossible to predict. Who could have predicted that host-mediated restriction of bacteriophages would lead to the dawn of genetic engineering or that a more recently discovered form of bacterial immunity to infection would lead to the CRISPR revolution in gene targeting in human cells that we are witnessing today? What can be said for certain is that fundamental research does have major impacts on society, medicine and economic productivity. Without fundamental research, the march of science would slowly grind to a halt.