A universal pipeline for genetically-encoded fluorescent biosensor production

Advances in high-resolution bio-imaging have revolutionised research in the life sciences and medicine allowing the localisation and movement of proteins within living cells to be visualised and quantified. However, in addition to proteins, there are many other important biomolecules such as metabolites and signal molecules which have not proved so easy to quantify in the same way. The problem is that one has to design a fluorescent sensor protein for each biomolecule. The fluorescence properties of these proteins change when they bind the target molecule, providing a readout of its concentration in the cell. These fluorescent sensors are difficult to make, partly because they rely on identifying a suitable protein that both binds the target molecule and, upon doing so, changes shape substantially - it is this shape change that causes the detectable change in properties of a linked fluorescent protein such as green fluorescent protein (GFP).

We have devised an alternative method of constructing fluorescent sensor proteins that does not require the receptor protein to undergo a large shape-change. This massively increases the range of potentially useful receptor proteins and also allows receptor proteins to be identified from antibody libraries. This proposal is to assess the new sensor design by making two test sensors, one for a peptide and one for the hormone estrogen. If successful, our design would represent a universal platform from which biosensors for any molecule could be constructed. We would make this platform available to academic researchers and anticipate that it would dramatically speed up the collecting of critically important data about small molecule dynamics in living organisms.

Advances in high-resolution bio-imaging have revolutionised research in the life sciences and medicine. However, although it is now possible to observe the localisation and dynamics of any protein in living cells, the same is not true for the huge numbers of important small molecules such as metabolites, signal molecules, ions and xenobiotics. This is because it is technically challenging to design biosensors for such molecules and once a sensor is built for one molecule, it cannot readily be transformed into a sensor for other molecules. Moreover, current approaches for biosensor design are limited by the availability of naturally occurring receptor-transducer proteins i.e. those that can specifically bind the target molecule and also undergo a large-scale conformational change upon binding that can be detected through altered properties of a linked fluorescent protein.

This proposal aims to address this bottleneck in biosensor construction by testing a revolutionary new sensor design that would, from a single molecular platform, allow sensors to be built for any target molecule. Moreover, our design removes the dependence on naturally occurring receptor-transducer proteins - there is no requirement for the receptor protein to undergo conformational change. This substantially increases the number of potentially useful naturally occurring receptor proteins and there is also the possibility to screen single chain antibody libraries for appropriate binding motifs.
This project will assess the novel sensor design by constructing test sensors, one for a peptide and one for the hormone estrogen. If successful, our design would represent a universal platform from which biosensors for any molecule could be constructed and would dramatically speed up the collecting of critically important data about small molecule dynamics in living organisms.

From a commercial standpoint, biosensors are increasingly being used for routine medical diagnostics. The global biosensors market is estimated to be worth $12 billion by 2015. Blood and urine tests for glucose and cholesterol dominate, but there is massive demand for other diagnostic tests both for point-of-care and consumer use. Similarly, biosensors are increasingly being exploited by the environmental monitoring industry to detect pollutants in watercourses and other sites. Our sensor platform could be commercially exploited in both of these growing markets and businesses in these sectors could therefore be major beneficiaries of the research. These companies would benefit from a streamlined R&D timeline for the development of new biosensors and from a massively increased range of potential molecules that sensors could be designed for. This could allow a substantial expansion of the biosensor industry and this would benefit the nation economically through the profit and employment generated by these activities as well as potentially improving national health and well-being through improved medical diagnostic tests and environmental monitoring.

We will seek to transfer knowledge generated in this research by discussing the formation of partnerships with appropriate companies in these sectors, or by establishing a University spin-out company. Realistically, this initial phase of work can be considered a proof-of-principle project and further research will be required to optimise the design of the sensor cassette (for sensitivity, stability and operating dynamic range) and to develop biosensors for specific purposes. This is likely to require an additional 2-3 years of work but could lead directly to marketable products at the end of this period.