Conformational Dynamics of Two-Way DNA Junctions

Our genetic information is contained within a molecule called DNA, which is a polymer of nucleotides. DNA is normally double-stranded (ds) and each strand can act as a template for formation of the other. During this process of copying genes, which is known as replication, protein catalysts have to cut DNA in very precise ways. These DNA cutting protein catalysts are called nucleases and can be thought of as DNA scissors designed to cut in a particular place. Similarly, because damage to DNA can be so catastrophic to life, cells have evolved ways of removing DNA damage and restoring genome integrity. These DNA repair pathways also require nucleases. However, cutting DNA in the wrong place is potentially life-threatening, so nucleases have to be very specific.

Most of the intermediate structures and some of the damaged ones that need cutting by nucleases have only very subtle differences from dsDNA. They can have a cut known as a nick on one strand, or they may have parts of one strand missing called a gap. Alternatively, they may also have single-stranded protrusions known as flaps at the nick. Each of these DNA structures has one strand that is discontinuous. This means that unlike normal dsDNA the structure can bend at the nick into two dsDNA parts with a junction between them. Bendability is therefore an important way of distinguishing between replication intermediates and damaged DNA that needs cutting and dsDNA that doesn't.

In this work we want to understand how the fundamental properties of the intermediate and damaged DNAs are used to distinguish them from dsDNA. To do this we need to understand what the junction DNAs look like. We will find out how they bend and how fast they bend by studying them using complementary techniques that allow us to understand the spatial relationships of parts of the DNAs. We will find out what features of the DNA junctions make them bend most; these could be where the flaps are or the size of the gaps or the particular identity of the nucleotides. We will also find out how the environment changes the bending properties; DNA is negatively charged so its conformation can be altered by positively charged metal ions.

It has been suggested that structure-sensing nucleases find the site where they need to act by sliding along dsDNA until they reach one of the junctions. We will test whether this is possible, asking whether they can interact with dsDNA that cannot bend. We will also explore an alternative mechanism of damage detection, where another protein that the nuclease binds to pauses at the junction site. Together these studies will allow us to understand the interplay of the fundamental properties of two important very important biological molecules, DNA and protein, as they go about protecting our genomes.

Structure-sensing nucleases are essential to life in all organisms, but require exquisite specificity for their respective aberrant DNA substrates. These enzymes have the formidable task of recognizing and processing two-way duplex-duplex junctions in DNA such as nicks, flaps and gaps, which only appear to deviate modestly from the duplex form. Typically these junction DNAs bind to proteins with close to a 90 degree angle between constituent duplex arms. In current models for the action of structure-sensing nucleases the proteins slide along duplex DNA until they encounter a DNA junction, which they then bend. These models originate from the assumption that protein-free DNA junctions exist in a coaxially stacked form. We propose to test current models for the action of structure-specific nucleases. Firstly, we will characterise the dynamic properties of exemplar DNA two-way junction DNAs using NMR techniques, quantifying the extent to which and the rate at which the junction DNAs transiently adopt bent forms in solution. Spectroscopic studies of the extent of stacking interactions within and across junctions will allow a more comprehensive comparison of the properties of various junctions. We will learn about how the number, position and size of flaps, the sizes of gaps, the presence or absence of phosphate monoester and variation in sequence alter the junction arrangements. We will also quantify the extent to which structures are altered by the presence on mono- and di-valent ions. Finally, we will test the current model for reaction site location by DNA structure-sensing enzymes by assessing whether they can bind to non-junction duplex DNAs. Alternatively, we will investigate the possibility that protein interaction partners such as PCNA play a role in junction location by response to junction properties. Together these studies will shed light on the essential principles of protein-DNA junction recognition that are fundamental to DNA metabolism in all life forms.

The immediate beneficiary will be the researcher CoI who will further develop their research skills.

Pharmaceutical and biotechnology companies seeking to inhibit DNA replication and repair nucleases will directly benefit form this research. Several structure-sensing nucleases are already a target for therapeutic intervention for a number of companies, including some located within the UK. Ultimately such efforts rely on molecular level understanding of structure and mechanism.

Our study will also impact other academic groups seeking to understand DNA replication and repair as it will define the features that give rise to specificity.

Longer term should our efforts result in a treatment it will increase UK pharmaceutical and biotechnology companies competitiveness and benefit society.