The creation of artificial protein molecular switches

The ability of an organism to sense changes to its environment and react in a suitable manner is critical to its survival. Cells must respond to many stimuli, including chemical signals such as changes in nutrient levels and messenger molecules. Consequently, nature has evolved many different systems capable of responding to the signal. Generally, proteins act as the sensor of the stimulus by recognising and binding the chemicals responsible. On binding the chemical, the three-dimensional structure of the protein changes so altering its function. It is this change in function that allows the signal to be detected and reacted upon by activating the next link in the chain or allowing the protein to directly tackle the cause to the stimulus. Such proteins that recognise and bind chemical signals leading to a change in their structure and function are termed protein molecular switches. The ability to produce molecules that can change their properties (or output) in response to a desired input has potential for a wide variety of applications, including the creation of novel sensors and materials. As proteins are already known to have the properties suitable for a molecular switch, it would appear logical that they would be the ideal material from which to construct our desired switches. Although it might appear simplest to use natural sensing proteins, these have evolved to fulfil specific functions within a defined biological process and may not have the requisite properties for a particular application. Therefore, new proteins may need to be utilised and adapted. The natural diversity of protein structure illustrates that proteins are very flexible molecules capable of a wide range of functions. Our ability to increase this diversity by modifying the DNA sequence that encodes a protein, also broadens the variety of structural permutations that are open to proteins thus allowing new properties to be incorporated. The creation of proteins whose output changes in response to a desired chemical will provide a powerful tool for sensing changes in the cellular environment and eliciting a suitable response. Artificial protein switches could also have applications outside of the biological context, such as in the area of nanotechnology. As proteins work at the nanometre scale, their ability to act as a molecular switch could be applied to create novel sensors, transducers and intelligent materials that respond in a required manner rapidly and reversibly. Therefore, we hypothesise that artificial proteins can be created that can act as molecular switches controlled by an input of choice. The proposed research will address this hypothesis by creating a novel molecular switch that responds to the biologically important small molecule haem. To achieve this, we will link the structural changes that occur on haem binding to the protein cytochrome b562 (cyt b) to the catalytic activity of the enzyme TEM-1 beta-lactamase. Cyt b and TEM-1 have unrelated functions in nature and exist as separate proteins. In order to link their functions, a strategy called domain insertion will be used, in which one protein is inserted within another. In this case, cyt b will be inserted within TEM-1 using a recently developed genetic engineering tool. As it is difficult to predict sites within a protein that permit the insertion of another while retaining the functions of the individual proteins and allowing events on haem binding to be coupled to enzyme activity, many different insertion positions within TEM-1 will be sampled. The new chimeric proteins will be analysed to identify and isolate those whose TEM-1 activity is now dependent on haem binding. Those chimeric proteins with the desired switching attributes will be analysed in more detail to characterise their properties.

The ability to design molecules that can change their properties in response to a desired input will allow significant new possibilities for creating novel sensors, modulators and transducers for use in both natural and artificial contexts. The capacity of proteins to fulfil the function of a molecular switch is well established in nature, as demonstrated by their key roles in regulating many biological processes. The aim of this project is to create artificial protein molecular switches whose output is modulated by an input of choice. While it might appear easiest to modify natural protein switches, they have evolved to fulfil specific requirements within a certain biological context, making them difficult to adapt for new applications. Therefore, as a general approach for creating tailored protein switches, we hypothesise that unrelated proteins with the requisite properties can be coupled through ligand-dependent conformational events. To address this hypothesis, the protein TEM-1 beta-lactamase will be converted into an allosteric enzyme whose activity is regulated by the biologically important small molecule haem. A strategy called domain insertion will be used to link the enzyme activity of TEM-1 with the conformational changes associated with haem binding to cytochrome b562 (cyt b). As it is difficult to predict sites within a protein that will permit the insertion of another protein while retaining the function of both proteins and allowing the transmission of conformational events, a directed evolution approach will be used. Libraries of chimeric proteins will be created using a recently developed transposon method in which cyt b will be inserted randomly into TEM-1. The libraries will be screened to identify chimeric proteins whose TEM-1 activity is dependent on haem binding. Those chimeras with the desired switching properties will be analysed in more detail in vitro and in vivo to characterise their haem dependent properties.