Team:British Columbia/Approach
Introduction
In response to the issues that saxitoxin and other PSTs bring, we are aiming to discover a biosensor system that is capable of reliably indicating the presence of saxitoxin above a specific concentration. As far as we know, there are no PST detection kits that can provide cheap, accessible, and reliable semi-quantitative to quantitative data on shellfish samples. Other issues include relatively high false-positive/false-negative rates and the difficulty to perform such tests in a non-laboratory environment. Furthermore, pre-existing test kits mostly rely on immunoglobulins which contribute to their fairly costly price points.
Design
Transcription-based biosensorsWith this in mind, we decided to move towards a more novel approach and use transcription factor-based biosensors. Transcription-based biosensors have become increasingly popular in biotechnology due to their easy implementation into genetic circuits and their high modularity.(1) Additionally, because they use DNA sequences and protein products, they can be modified to change specificity, dynamic range, and sensitivity.(2) This appealed to us due to its potential and flexibility, which may allow us to augment and optimize our biosensor, giving birth to a larger range of possibilities compared to other biosensor types (i.e. Antibody based biosensors). Moreover, downstream applications may be cheaper and easier to produce, such as whole-cell biosensors.
While we parsed the literature, we did not find documented regulatory units induced by saxitoxin, which drove the project towards gene discovery. We decided to screen for the transcription factor in samples coming from locations and times of a harmful algal bloom producing saxitoxin. The metagenomic DNA we possessed were acquired from the Saanich Inlet through a time-series being run by the Hallam lab. With data from the CFIA we were able to pinpoint samples collected during high saxitoxin production in the periods. The theory behind working with these samples to find a saxitoxin induced regulatory unit, is the knowledge that these biotoxins may be used in signalling (3, 4). Furthermore, with how microbial dynamics change throughout algal blooms and the close knit relationships between species (5), and the wide distribution/long existence of STX genes (which is found even across kingdoms) (6), means it is likely that other microorganisms may respond to the presence of saxitoxin at the transcription level.
This is based on knowledge that catabolic gene expression is generally induced by relevant compounds (substrates or metabolites) and, in many cases, controlled by regulatory elements situated proximate to catabolic genes
The method we chose for gene discovery is substrate-induced gene expression (SIGEX), a procedure pioneered by Uchiyama and Watanabe. This technique allows for the screening of metagenomic libraries without the need to isolate bacteria and overcomes many of the limitations of current screening methods (7). SIGEX also lends itself well to high-throughput screening when paired to FACS or other methods of cell sorting. The general workflow can be seen below or more specifically in the experiments page. We modified this method to better fit our own applications, including building our own plasmid, to increase the overall efficiency of our experiments. Additionally, we were fortunate enough to use cutting edge microfluidics (utilizing microfluidic chips) for cell sorting with the On-chip cell sorter, instead of a more conventional FACs machine (however both ways could be suitable for this project).
The vector-trap plasmid was designed to ensure the maximum possibility of acquiring a fragment that may contain the promoter-transcription factor combination that we want. We were inspired by Uchiyama and Watanabe’s plasmid vector design, with multiple changes (8). By using a blunt end restriction site (NruI), we ensured that a higher variety of DNA would be ligated. We compensated for the lower ligation efficiency through fast-link DNA ligase and blunt-end fixing for the prepped metagenomic DNA. As per the paper, a GFP gene was added for possible FACs screening, but due to blunt ends ligate with different orientations, a reverse RFP was added as well. The reporter genes allowed high-throughput screening through FACS, if a promoter and associated transcription factor was ligated into the vector.
On top of SIGEX, we decided to utilize a library of around ~2300 known E. coli promoters (9) and screen it for saxitoxin-responsive transcriptional system. We did this to have some comparable results to SIGEX and as a backup if SIGEX were to fail in finding a useable promoter. This process was automated through the Labcyte Echo liquid handler for more reliable and reproducible results.
To top everything off, we decided to also produce a basic model of what our hardware would be if our biosensor were to be implemented in one of the downstream applications. Our design revolves around ease of use and affordability, to improve on the current choices for on-site PSP testing. Go to our hardware page to learn more about it.
Overall, with more testing, we can say that we have improved on the SIGEX, developing a simpler method to screening metagenomic library that can be, theoretically, applied to the discovery of other biosensors from the environment. With these tools, hopefully, we can continue to progress the use of transcription-factor based biosensors in biotechnology to aid solutions in other areas as well.
References
[1]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4700088/
[2]https://onlinelibrary-wiley-com.ezproxy.library.ubc.ca/doi/full/10.1002/biot.201700648
[3]https://www.ncbi.nlm.nih.gov/pubmed/21076133/
[4]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3705384/
[5]https://www.frontiersin.org/articles/10.3389/fmicb.2018.01201/full
[6]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3766867/
[7]https://www.tandfonline.com/doi/pdf/10.1080/02648725.2007.10648094
[8]https://www.ncbi.nlm.nih.gov/pubmed/18600226
[9]https://www.ncbi.nlm.nih.gov/pubmed/16862137?dopt=Abstract