Engineering
The development of PestiGuard was guided by the iGEM engineering cycle. Through repeated rounds of Design, Build, Test, and Learn, we systematically transformed a conceptual genetic circuit into a reliable and sensitive biosensor. Each iteration addressed critical challenges, leading to the optimized system we present today.
Our initial design established the fundamental "sense-and-response" logic. The circuit consisted of a strong T7 promoter, a lac operator (lacO), a glyphosate-binding aptamer, an RBS, and an EGFP reporter gene.
[BBa_2561FDBM: Glyphosate-sensing EGFP device (w/o spacer)]
Figure 1: The original version of gene design
We constructed this genetic circuit in a plasmid and transformed it into E. coli BL21(DE3).
We assayed for leaky expression without IPTG and for a fluorescence decrease upon glyphosate addition with IPTG
Figure 2: Colonies from glyphosate-sensitive aptamer construct glowing on LB agar plates in the absence of IPTG
The constitutive fluorescence observed in the absence of IPTG indicated a critical failure in the system's regulatory control. This "leaky expression" demonstrated that the lac operator was ineffective at repressing the exceptionally strong T7 promoter in our specific genetic context. We hypothesized that the proximity of the aptamer sequence to the lacO site was potentially interfering with the repressor's binding or the formation of a complete transcriptional terminator loop. This insight directly informed our next design iteration, where we focused on modifying the genetic context to restore proper control, moving from a non-functional to a controllable system.
The initial design lacked sufficient transcriptional control. The genetic context of the regulatory elements needed optimization.
The biosensor showed high baseline fluorescence even without IPTG, indicating constitutive "leaky expression." The lacO switch was insufficient to fully repress the powerful T7 promoter.
To improve regulatory control, we hypothesized that the lacO and aptamer sequences were interfering with each other. We introduced a spacer sequence between them to ensure independent function.
[BBa_25EK3998: T7-Glyphosate-sensing EGFP device]
Figure 3: 2nd version of gene design (adding a spacer)
The modified plasmid with the spacer was constructed and transformed.
We repeated the induction and glyphosate sensing assays. Initial induction tests using 0.1 mM and 0.5 mM IPTG with a T7 promoter system failed to produce a proportional decrease in fluorescence. The standard curves generated across 12, 24, and 36-hour time points were unreliable, as indicated by low R-squared values.
The spacer successfully eliminated leaky expression. However, the fluorescence output in response to different glyphosate concentrations was inconsistent and not proportional to the pesticide dose.
While the introduction of a spacer successfully resolved the leaky expression by insulating the lacO operator, it revealed a more fundamental limitation of our initial design. The inconsistent and non-proportional fluorescence output indicated that the system's dynamic range was saturated. The extremely strong T7 promoter was driving such a high level of transcription that it overwhelmed the subtle, dose-dependent regulatory effect of the aptamer. The signal was perpetually maxed out, leaving no room to observe a quantifiable decrease. This led to a pivotal insight: the performance of a biosensor is critically dependent on matching promoter strength to the sensing mechanism. A promoter that is too strong can render a dose-response relationship unobservable, not due to a failure of the sensing element itself, but due to a mismatch in the system's overall expression dynamics. This understanding forced a paradigm shift in our design strategy, moving beyond simple genetic insulation to a holistic re-engineering of expression levels, which became the focus of the next cycle.
We had solved the control issue but uncovered a dynamic range problem. The signal output was not suitable for quantitative detection.
We diagnosed that the extremely strong T7 promoter was causing signal saturation. We replaced it with the moderately-tuned PlacUV5-MB7 promoter (Part:BBa_K4941056) to reduce metabolic burden and achieve a measurable dynamic range.
The new plasmid with the PlacUV5-MB7 promoter was built and transformed. [BBa_253OHGFW: Glyphosate-sensing EGFP device]
Figure 4: 3rd gene design
First, we systematically tested the new system under various IPTG concentrations and expression durations. Furthermore, a gradient of the target pesticide (glyphosate) across concentrations of 0, 0.3, 1, 3, 10, 30, 100, 300, and 1000 mg/L were tested.
The results showed a dramatic improvement. Induction with 0.5 mM IPTG for 24 hours provided the optimal balance: a high R-squared value (~0.95) and large, unambiguous signal steps, enabling quantifiable detection. Most critically, within the agriculturally relevant 0-1 mg/L range, the dose-response curve achieved an R-squared value of ≥ 0.95, passing our pre-set reliability threshold for a robust standard curve.
The successful function of the final design confirmed a fundamental principle of biosensor engineering: the sensing element and the expression system must be functionally balanced. While the aptamer provided the specificity, its regulatory capacity was only observable within a suitable range of transcriptional activity. The moderate-strength PlacUV5-MB7 promoter created an expression window where the aptamer’s conformational change could effectively compete with translation initiation, translating ligand binding into a measurable, dose-dependent fluorescence output.
Promoter strength is a critical parameter for setting the dynamic range of a whole-cell biosensor. The switch to a moderate promoter was essential to prevent signal saturation and to unmask the aptamer's dose-dependent activity, thereby transforming a qualitative sensor into a quantitative analytical tool.
Table 1: RGB values of bacterial cultures at 0.1mM and 0.5mM IPTG, 12, 24 and 36 hours and different glyphosate concentrations
Figure 5: Standard curve of bacterial cultures at 0.1mM and 0.5mM IPTG, 12, 24 and 36 hours and different glyphosate concentrations
Table 2: Fluorescence of bacterial cultures at 0.5mM IPTG, 24 hours and different pesticide concentrations
Table 3: RGB value of different types of aptamer at different concentrations of pesticides
Figure 6: Standard curves of glyphosate at different ranges of concentration
Parallel to our wet-lab cycles, we employed computational modeling to address a fundamental design risk: the aptamer sequences were derived from DNA, but their functional form in our system is RNA.
We used in silico tools (mFold, molecular docking, MD simulations) to predict if the DNA-derived RNA aptamers would fold correctly and bind glyphosate.
To evaluate the computational predictions, we conducted parallel in silico and in vitro analyses. For each candidate aptamer (glyphosate, acephate, malathion, chlorpyrifos, and acetamiprid), we performed:
These computational results were directly compared with experimental wet-lab data from fluorescence-based dose-response assays, allowing us to correlate predicted binding stability with observed biosensor performance.
Our models confirmed that the RNA and DNA forms have different secondary structures. Crucially, MD simulations indicated that the glyphosate RNA aptamer formed a stable complex with its target, providing a theoretical foundation for our wet-lab approach. This pre-validation helped us prioritize resources.
However, we observed discrepancies where modeling predicted weak binding for some aptamers that showed function in the lab. Our analysis revealed:
The most compelling explanation for our success is that the functional, glyphosate-binding structure of our RNA aptamer is stabilized by the unique in vivo environment and/or is a kinetically trapped intermediate. Our working biosensor is the ultimate validation, demonstrating that biological function emerges from the complex, dynamic environment of the cell.
To further bridge this gap and rigorously prove our biosensor's mechanism:
We designed a new construct for Cell-free Protein Synthesis (CFPS). [Parts: BBa_25IONSGX: Non-cell protein synthesis - glyphosate-sensing EGFP device]
The CFPS-optimized plasmid was constructed with the following features:
Figure 7: The cell-free synthesis version of gene design
We will test this construct in a commercial CFPS kit (BioSharp Life Science) to isolate the RNA-pesticide interaction and confirm the intrinsic function of our genetic circuit.
A positive result in CFPS would powerfully narrow the cause of the modeling discrepancy to folding and affinity, independent of cellular metabolism.
This structured, problem-solving approach—embracing both wet-lab and dry-lab cycles—exemplifies the power of the engineering cycle in synthetic biology and provides a robust framework for future biosensor development.