Molecular characterisation of leader sequences which can also mediate ribosome 'skipping'

Normally one gene encodes one protein. Many processes within the cell require multiple, different, proteins to present within each individual cell - these proteins may interact with each other ,or, act sequentially (in a 'pathway' to perform a specific task. The research in my laboratory has been concerned with how viruses - with just one 'gene' - can generate multiple, different, proteins. We discovered one method - which we called 'ribosome skipping' - by which viruses can accomplish this feat. This method has proved to be very adaptable for many biomedical and biotechnological purposes when one needs to express multiple, different, proteins within the same cell. In effect, one can concatenate multiple genes into one by linking them with the short sequence ('2A') we characterised. At each 2A sequence, the new protein we have designed (a 'polyprotein') is 'cleaved' apart into it's constituent individual proteins. Fortunately, this method works in all animal and plant cell-types, which makes treatment of diseases, the genetic manipulation of crops etc. much easier than trying to introduce multiple genes into the patient (human gene therapy) or in the case of plants, for example, to make drought-resistant potatoes. We have discovered a new type of 2A sequence that can not only cleave itself, but has another function to direct the protein to a different site within the cell. If the sequence cleaves the protein becomes located in one part of the cell: if it doesn't cleave the protein goes to a different part of the cell. In fact, we have discovered 144 new sequences - but many are closely related so we don't need to study all 144 - just ~35 representative sequences. This is a new discovery and some of the work we want to do will characterise the biology of this system: how the sequence of these different 2As relates to the two different properties (self-cleaving / protein targeting within the cell) they have. The other aspect of the work is to see if these sequences work in plants and yeast. If this is so, then the potential uses of these sequences is dramatically expanded. We want to develop these sequences as tools for other scientists to use, to enable them to conduct their research into new cures for disease, for new biotechnological products and processes.

'2A' is a ~20aa tract occurring in a wide range of viruses. When ribosomes encounter 2A, they become 'stalled'. These are 'rescued' by the activity of release (termination) factors 1 and 3, releasing the nascent protein from the complex. Two outcomes may ensue: either translation terminates, or, elongation may continue - without peptide bond formation. Sequences downstream of 2A synthesised as a discrete ('cleaved') product. In this manner multiple, discrete, products can be generated from a single ORF. Recently the genome of the sea urchin Strongylocentrotus purpuratus was published and we detected many 2A-like sequences ('SP2As') in two classes of genes. Firstly, non-LTR retrotransposons. Secondly, forming the N-terminus of over half of the CATERPILLER innate immunity proteins (~144/204) of this organism. Bioinformatics suggested many were signal sequences. The preliminary data we present shows these SP2As mediate co-translational cleavage (like virus 2As) but, importantly, they may also function as signals: a novel method of dual protein targeting. If the signal cleaves itself away the nascent protein localises to the cytoplasm. If it does not cleave, the protein is targeted to the exocytic pathway. We wish to catalogue the activity of a representative sub-set of these SP2As in terms of cleavage / protein targeting and to determine what type(s) of signal sequences these are, the sub-cellular localisation(s) they specify, the relative efficiencies (partition between the cytosol / exocytic pathway) of these signals and if they are cleaved by signal peptidase. The other aspect of the application is to explore the biomedical / biotechnological utilities of the system: can new combinations of activity (cleavage/targeting) be generated by site-directed mutagenesis? Do these sequences function in plants and yeast both in terms of cleavage (the viral 2As do function in these organisms) and as signal sequences?

The impact of this proposed program of work will profit from, and build upon, the profile we have established on our work on virus 2As as biotechnological tools for the co-expression of multiple proteins. Using 'FMDV 2A' as the probe in Google or Freepatentsonline reveals the impact of our work. Over 240 papers have been published citing the use of 2A in a broad range of biomedical and biotechnological areas. Products are coming onto the market using 2A as an essential element in their technologies: System Biosciences market a range of products using 2A technology. Human (immuno-)gene therapies using 2A (co-expression of A+B chains of the T-cell receptor: anti-metastatic melanoma) are in stage II clinical trials in the US. Many recent advances in pluripotent stem cell production rely upon 2A to co-express the multiple transcription factors required. The 'awareness' of 2As is increasing rapidly, and can be harnessed for the impact of SP2As. We have created web pages to explain and to 'showcase/advertise' 2A: subject to limitations imposed by patent applications, we will use these pages to increase the impact of SP2As by explaining the new dual protein localisation system and indicate the properties of the sequences we have characterised. The utilities of SP2As. We know that SP2As function (cleave / target) within mammalian cells, and it seems quite possible that this will be case in plants and yeast. Naturally, the utility - and, therefore, the impact - will depend upon how 'malleable' the system is with regards developing (and permuting) different cleavage / targeting functions. Indeed, using the knowledge gained from this research, a specific cellular / virus signal sequences of choice could be engineered to gain an additional self-cleaving function. For example, this system could be of use in; (i) high-throughput screening technologies, (ii) identification of protein/peptide secretor cells - in that a proportion of the effector protein would be retained within the cell, (iii) antigen expression systems - antigens would be presented in a wider range of contexts, (iv) the simultaneous expression of membrane-bound and soluble (secreted?) proteins, or, (v) the attenuation of viruses encoding membrane-bound virus particle surface glycoproteins (e.g. influenza haemagglutinin / neuraminidase), etc. The future applications may be manifold. Preliminary experiments that do not constitute a full 'objective' (and have not been included in the 'case for support') will be conducted - where appropriate as collaborations - to explore and extend the utility of the system to address the potential functions mentioned above. Dundee Cell Products are raising an anti-2A monoclonal antibody (to be jointly marketed with ourselves) which I strongly suspect will cross-react with SP2As. If this proves to be the case, the marketing / publicity material sent out by them will include data on SP2As. This Dundee University spin-out company is located very near St. Andrews. I have excellent links with this company and in the first instance I will approach them about marketing a 'bundled' cDNA vector / anti-SP2A monoclonal antibody product. The utility - and therefore the range of potential end-users - will be subject to the outcomes of objectives 5 and 6. Subject to restrictions imposed by the process of patent applications, potential industrial partners (e.g. GSK, Pasteur-Merieux) will be contacted to explore the possibility of joint projects to take the technology into the market place. System Biosciences (USA) already markets a range of products using virus 2As (T2A) and may be a future industrial partner.