The single polypeptide type I restriction enzymes - minimal multifunctional molecular motors

To make life more straightforward, Mankind has developed machines to undertake complex tasks. But Nature got there first; life on Earth is reliant on the action of microscopic machines and motors made not from metal gears and cogs but from proteins. Miniaturised factories are working in your body right now! Take for example your genetic material, DNA. There are a great many 'molecular motor enzymes' whose task is to recognise a specific part of the DNA (a sequence made up of chemical bases called A, G, C and T) and then to convert chemical energy (in the form of a fuel compound called ATP) into mechanical events. For example, the proteins we study move along DNA many thousands of bases at a time, much like a train along a track. Once these motors reach a defined point in the DNA they stop and cut the track using an additional enzyme activity. Whilst this may seem a destructive event for the motor, it is vital to the health of the cell because these cuts are not made in the host genome but are targeted to invading viral DNA. A remarkable feature of the enzymes we wish to study is that all this complexity is contained within a single protein, a bit like a mobile phone which is also a camera, a games console and an MP3 player. How can all the multiple enzyme activities be contained in one protein unit and how are the activities coordinated so that the motor can move along DNA in a controlled fashion? This is what we intend to find out.

The ATP-dependent Restriction-Modification (RM) systems LlaGI and LlaBIII consist of a single polypeptide with motifs characteristic of an endonuclease, a helicase and a methyltransferase. They represent prototypes of a new RM sub-group that are most closely related at an amino acid level to the Type I RM enzymes. Initial exploration of their enzyme activities suggests 1-D DNA translocation and DNA cleavage activities that are also related to the Type I enzymes. The goal of this proposal is to elucidate how LlaGI and LlaBIII couple ATP hydrolysis to DNA cleavage and how a single polypeptide motor controls this process. Our first aim is measure rate of initiation and DNA translocation using a continuous triplex displacement assay developed in this lab that is amenable to rapid mixing technology. This assay can also reveal the processivity (the distance moved per binding event) and the directionality of motion. These parameters can be investigated under a range of conditions, with mutant enzymes and with different DNA substrates. In parallel, ATP hydrolysis can be measured in continuous assays to obtain kinetic and thermodynamic information. By marrying the translocation and ATP hydrolysis rates we can start to address the fundamental question of the efficiency of mechanochemical coupling. We can also vary the order of mixing of our reaction components or challenge with mutant proteins to address how the enzymes terminate translocation and how (or if) they re-initiate motion. For instance, if a mutant enzyme that lacks translocase activity is mixed with a translocating wild type enzyme, the mutant will only interfere with re-initiation if the wild type enzyme dissociates completely from the DNA. To delve more deeply into the structural basis for the motor activity we will probe the subunit and domain organisation of LlaGI and LlaBIII. Subunit stoichiometry will be examined using gel filtration and analytical ultracentrifugation for free proteins, for complexes with nucleotide and cofactor analogues and for complexes with oligoduplexes. The assembly state is unknown at present but the answer will in part be related to the directionality of the motor (bi-directional translocation would require at least a dimer whilst unidirectional translocation would only need a monomer). The protein architecture will be probed using protelytic mapping to identify domain boundaries. Domains will then be purified and analysed for enzyme activity both in isolation and in mixtures to try to recapitulate the activity of the intact protein. Ultimately, DNA translocation is likely to be used to allow the interaction of enzymes bound at distant DNA sites that will then produce dsDNA breaks. This will be investigated by measuring the location and stoichiometry of DNA hydrolysis as a function of different substrates (varying the number of recognition sites, DNA size, or topology). One useful feature of using LlaBIII and LlaGI is that the proteins are virtually identical (~96% identity) except in their target recognition domains. We can therefore generate DNA substrates with recognition sites for each enzyme and then mix a wild type version of one enzyme with mutant versions of the other. From these studies we can deduce how DNA cleavage sites are selected and the role of each enzyme subunit in the reaction. To date there has not been any success in crystallising a complete hetero-oligomeric Type I enzyme complex. The LlaGI and LlaBIII offer an alternative route due to their minimal subunit architecture. This will be an ongoing strand to the project which will form the basis for additional funding should diffracting crystals be generated.