The Development, Implementation and Transfer of a Reproducible Methodology for the Generation of High Quality Data for Part Characterisation

Synthetic Biology aims to apply an engineering approach to bio design. This involves breaking down biological function into components, Parts or modules. One of the key elements of this approach is that it uses standard Parts. Is relies on the ability to accurately characterise parts and to reproduce the characterisation process. Reproducibility is at the heart of engineering biology, but this can be challenging. Consequently a collaborative project has been undertaken between Imperial College London and Nanyang Technological University (NTU) in Singapore to determine the transferability and reproducibility of our characterisation process. 

The Kitney Lab at Imperial College has developed a robust and automated method to measure the function of Parts. Because we are interested in reproducibility, the joint project with NTU has focused primarily on constitutive promoters. The methodology has been designed and optimised to ensure reproducibility, with the workflow comprising a number of stages. This includes an initial sample preparation with an overnight growth, dilution step in a chemical defined media, followed by a 90 minute outgrowth and with a second dilution step to a specified cell concentration (absorbance). The cell absorbance and fluorescence data are collected at regular (15 minute) intervals during a total of 6 hours - to calculate the promoter activity. The method was employed in triplicates and over 3 different characterisation runs (i.e. 3 different days). The promoter activity was shown to be highly reproducible using this approach.

The methodology was originally developed using an automated liquid handling system (TheOnyx;ALS, Germany). To test the robustness of the workflow, the same set of promoters were also characterised on a different automated liquid handling system (a FeliX; AnalytikJena) in our centre at Imperial College. The two automated systems have differences in their pipetting capabilities, but by tightly following the workflow the characterisation data was demonstrated to be closely matched.

The end goal of this project was to show the transferability of the methodology and that difference in the site, users and/or hardware would not significantly impact the comparability of the results. To test this method was utilised in the Poh Lab at NTU. Using the workflow developed in the Kitney Lab (but performing the dilution steps manually), similar results were obtained at NTU for the characterisation of an identical set of promoters. The next stage in the project, which will also be reported, is to undertake automated characterisation in Singapore on the same set of promoters.

In conclusion, we have developed an automated, reproducible and transferable characterisation methodology. This has initially been applied to the characterisation of a set on constitutive promoters. The workflow is now being modified and applied to other Part families. A lack of reproducibility is often cited as weakness in biological experimentation more generally. In synthetic biology it is essential. The development of our automated workflow is a step towards the standardisation of characterisation protocols and the production of large numbers of Parts with associated high quality, reproducible characterisation.