Lean bio-manufacture of surface-binding proteins for multi-functional coatings

There are numerous biomedical and biotechnological applications for recombinant proteins, some of which, requires them to be stable under non-biological conditions. This is leading the demand for the immobilisation of proteins on surfaces. However, problems can arise due to the requirement for the surface material properties being preserved, as well as retaining the protein activity throughout the whole process of production, purification and surface binding. Additionally, when the phases of production are separated, this allows for unforeseen errors to occur.

This project proposes to use a Pichia pastoris modular cloning (MoClo) toolkit as well as a Yeast MoClo toolkit, in combination with a statistical Design of Experiment (DoE) approach to produce a rapid, automatable system for protein-based multi-functional materials.

The project has four main aims:
I. Development of Pichia pastoris as an expression host to produce surface-binding proteins.
II. Validate Pichia pastoris for the production of surface-binding proteins of relevance to the manufacture of multi-functional coatings.
III. Address the problem of "one-factor-at-a-time" optimisation in standard bio-design practices.
IV. Extend the range of enzymes and surface materials. This will include the use of metal nanoparticles as a surface because they have been demonstrated to be directly coated with proteins leading to stable metal-NPs which retain protein activity.

The aims and objectives would be achieved through the following procedures. Firstly, using the MoClo Yeast Toolkit, in combination with the MoClo Pichia Toolkit, cassette plasmids would be constructed containing different promoters, secretion tags and fluorescent reporters and transformed into E. coli before being transformed into P. pastoris by electroporation. The plasmid cassette would be integrated into the genome by homologous recombination, increasing the stability compared to plasmid-based expression due to the lack of a stable plasmid system in P pastoris. Secretion efficiency of the cell would be assessed by measuring the total fluorescence, then by measuring fluorescence of the supernatant post cell removal by centrifugation. Cellulose-binding proteins (CBP), from the iGEM registry of characterised cellulose-binding domains, would be fused to a fluorescent reporter. The ability of this to bind to cellulose would be analysed with increasing levels of purity, such as with the cells in the media, cells removed via centrifugation, ion exchange chromatography and gel filtration chromatography.
Streptomyces sp. chaplins would be tested as there are species, such as Streptomyces reticuli, containing genes encoding for proteins involved in cellulose binding. C2-chitin binding protein of Polistes dominula would be tested as the protein shows similarity to cellulose binding proteins and has the ability to bind to cellulose.

A DoE approach would be taken where all relevant parameters of the production, purification and surface binding properties will be changed simultaneously. The benefit of DoE is that optimisation is more effectively searched with the effect of interactions identified, and assessed, independently.

After a proof-of-principle DoE model for automated production of cellulose-binding proteins, the range of enzymes and surface materials can be explored. Metal-NP's would be interesting to investigate because they have high surface-area to volume ratios along with plasmonic or magnetic properties, allowing for high surface binding efficiency and simple detection and separation. We would produce metal-binding proteins fused to a library of metal binding peptides, recovery would be achieved by centrifugation or magnetically and evaluated as mentioned previously.