Aptamer-Based Biosensing System

Although we initially planned to only implement a degradation system for our target pyrethroid pesticide, λ-cyhalothrin, we thought it would be beneficial to further administer and implement a sensing system during the latter stages of our project. We transitioned into developing an integrated system involving sensing and degradation. This sensing part is vital, as by locating specific areas with high pesticide content, the release of our engineered E. coli can be controlled, and therefore also be more efficient, whilst also reducing concerns surrounding biocontainment, as our engineered E. coli is effectively contained to specific areas.

Our biosensing system for λ-cyhalothrin was devised after a thorough literature review. We wanted to develop an accurate and efficient sensing system, that is also cheap and rapid to use, with the potential for it to be deployed for in-field testing. Our research introduced us to the concept of the implementation of gold nanoparticles (AuNP) and specific aptamers within a red to blue colorimetric sensing system^[1], with the colour change indicating the presence of our target λ-cyhalothrin, and for our partnership with Concordia University, their pesticide target fenitrothion.

This sensing system involves the utilisation of gold nanoparticles (AuNPs) aggregation in the presence of poly(diallyldimethylammonium chloride) (PDDA) polymer to form AuNP-PDDA complexes, which is visualised by a colour change from red to blue. Due to AuNPs special optical property from localised surface plasmon resonance (LSPR); PDDA enables aggregation of AuNP under intense LSPR, shifting from the optimal absorbance wavelength of approximately ~520 nm (red), to peak absorbance at approximately ~650 nm (blue). A colour change is observed, as it shifts from red to blue, with intensity on the colour change correlating to the level aggregation of AuNP, and amount of AuNP-PDDA complexes formed^[2].

This colourimetric aggregating mechanism (Figure 1) is central to our aptamer-based biosensing system. Specific aptamer sequences that are complementary to our target λ-cyhalothrin pesticide, retrieved from Yang et al., 2021, are integrated into an AuNP and PDDA system^[3]. In the absence of λ-cyhalothrin, aptamers are not able to bind to the complementary pesticide, binding instead to PDDA polymer. As free PDDA is used up to form aptamer-PDDA complexes, AuNP can no longer bind to PDDA, therefore inhibiting the formation of AuNP-PDDA complexes - there is no colour change, as there is no AuNP aggregation, and thus the system stays red. However, in the presence of λ-cyhalothrin, complementary aptamers are now able to bind to the pesticide, enabling the formation of aptamer-pesticide complexes. This enables free PDDA within the system to bind free AuNP, promoting AuNP aggregation, and inducing colour change. The system turns blue and visually dictates the presence of the pesticide within our sensing system.

Whole-cell Biodegradation System

As the biodegradation system is central to our project, it was crucial that we spend most of our time focusing on identifying the design considerations surrounding the biodegradation system.

Although the deployment of GMO is banned in the UK, we decided during the early stages against the cell-free alternative for three reasons:

• Cell-free systems require a further purification step after expression
• Living cells are self-propagating, thus decrease the number of spraying instances, therefore reducing overall farmer burden
• Enzyme efficiency would fluctuate as purified enzymes are subject to environmental conditions, for example pH and temperature

We spoke to Dr Alan Goddard, a researcher at Aston University specialising in membrane proteins and lipid membranes, to investigate the challenges associated with our endeavour. Dr Goddard reaffirmed that choosing surface expression over cytoplasmic expression was the within right school of thought. He further mentioned that incorporating a repeater sequence after the carboxylesterase gene would confer a higher rate of membrane insertion. Additionally, he advised that the incubation should be done at an ideal temperature of 30°C rather than at 37°C.

Design of Pyre1
Pyre1 is a fusion protein formed from a truncated IPN domain isolated from Pseudomonas syringae and carboxylesterase CarCB2 from Bacillus velezensis sd. The IPN protein was truncated to only include its N-domain, which reduces the size of the protein from 4303 bp to 537 bp, This reduction changes the proteins interaction with the membrane, it is no longer a transdomain protein, but rather acts as an anchor to the membrane. This also successfully reduced the metabolic burden by 8-fold, with no adverse effects on function^[4].

Regardless, it was clear that the membrane interaction of any protein was a large challenge, as we had been reminded of by multiple experts. However, we believed that this was truly the best way to maximise degradation. In an attempt to improve our chances of integration we chose to add in a specific repeated sequence identified from literature^[5]. This repeated sequence was combined with a glycine-serine linker, in which this extra flexibility would allow us to hopefully avoid any steric clashes between the anchor, enzyme, and the lipid membrane. We believed this linker choice to be viable for any future teams considering membrane integration of enzymes and converted them into freely available basic parts. Find out more about it in Parts Overview.

Initially, we used computational modelling using AlphaFold, Clustal Omega and Autodock Vina to identify the affinity of λ-cyhalothrin to CarCB2, as seen in Figure. 2. The affinity was predicted to be -7.0kcal/mol, this is within expected values. We used Clustal Omega to conduct multiple sequence analysis and identified the catalytic triad consisting of aspartic acid, histidine and serine. This was used to guide the active site prediction for Autodock Vina on the protein structure predicted by AlphaFold.

Although this was a great start to better understand CarCB2, the same strategy cannot be applied once our protein is attached to the IPN(N) and linker sequences. This is due to the limitations present in AlphaFold as it does not account for the presence and effect of the membrane on the enzyme, which would lead to greater disorder and steric clashes which would not occur in a real system. This means we are not able to characterise the effect that the addition of linker and integration into the membrane would have on docking affinity and had to be elucidated experimentally.

To ensure we were cognizant of the physical limitations of the cell membrane, we modelled in silico the factors that a affect protein expression, including ribosome availability and lipid capacity of E. coli. More information can be found here. With some confidence in our expression system, we moved ahead with the experimental design for engineering our bacteria.

Plasmid Construction

We began with the cloning of our protein into the expression plasmid pYTK001 (Parts Overview). We opted for golden gate assembly, due to its capacity to produce scarless constructs. Additionally, this technique removes the possibility for the gene to be inserted backwards into the plasmid during incubation.

Pyre1 was designed with BsmBI flanking sites to ensure compatibility with pYTK001 (Figure 4). Successful transfects would knockout the GFP gene within pYTK001, and the chloramphenicol resistance gene would confer resistance to competent E. coli that have successfully ingested pYRE001. The results of our transformation can be seen here.

Although plasmids are known to have a limited life cycle within a host, and genome-level edits such as CRISPR confer greater stability for a transgenic gene, we decided against CRISPR as the extra stability would also lead to increased biosafety concerns as a result of horizontal gene transfer of unrefined GMO release into the environment. CRISPR is also known for making off-target modifications and producing a customised protein is extremely expensive.

Design of λ-cyhalothrin Degradation Assay

The design of our enzyme assay was primarily guided by the limitations of our ability to detect λ-cyhalothrin and 3-phenoxybenzaldehyde using HPLC. Find out more here. We designed our assay with 500 µL triplicates of both control and the Pyre1-expressing strain in a 24-well plate. Each replicate consisted of 2 wells; with 0 mg/mL and 1.25 mg/mL of λ-cyhalothrin being added to the system. This gives us a total of 12 samples, which served well as an initial guiding assay. Though we would have liked to test at varying concentrations, we were unfortunately limited by iGEM time constraints.

We had to make the following design considerations:

• Initial concentrations - Our initial concentration of choice was 1.25mg/mL. Although this was higher than legal limits, we could not test with lower concentrations without access to a GC/MS. Regardless, this served as a good proof of concept.
• Cell concentrations - In the interest of designing a rigorous assay, the cell OD was normalised from overnight cultures to 0.2 before addition to well plates. It was then allowed to grow for 3 hours to around 0.5 before λ-cyhalothrin was added. This ensured even OD across all replicants and strains.
• Delivery of λ-cyhalothrin to cells - The first challenge was delivery of the compound to cells. Our stock solution was dissolved in acetone, which would not be viable in HPLC. This led us to use an evaporator to remove acetone before resuspending in DMSO. This solution of DMSO was at a concentration of 100 mg/mL, which allowed the addition of 5 µL to 495 µL of cell culture. This kept the total organic solvent to below 2%.
• Time - We chose to run the assay with λ-cyhalothrin overnight. Although this does not provide data for degradation over time, we were aiming to use this as a guiding assay, which would be further optimised later
• Solvent Extraction - After incubation, the cells would be centrifuged (5 mins at 6000 rpm) to separate the solvent and E. coli. We chose to work with both the supernatant and pelleted samples, as there was prior literature for where the compound would be present. 50% acetonitrile was added to both samples. This would push our non-polar compounds out of the cell pellet and/or water content of the supernatant, and into the respective organic layer. This layer can then be spin filtered for testing in HPLC.