Overview
Many outstanding iGEM teams in the past have developed various detection systems to identify a range of biomarkers, which can include environmental heavy metals, proteins related to human health, or small molecules linked to food safety. The mechanisms of action generally fall into two categories:
- Direct mechanism: The biomarker can directly cross the cell membrane and act on a specific sensitive promoter, thereby starting or stopping the expression of downstream proteins.
- Indirect mechanism: The biomarker cannot directly cross the membrane or may need to interact with a protein first, forming a complex that then activates or terminates downstream expression.
We summarized the designs of these teams and developed our own system. In our system, we designed two plasmids, dividing them into two types and abstracting two functional modules:
- A (Auxiliary system) module:
- This module is primarily responsible for encoding auxiliary proteins, such as transport proteins and intermediate proteins. Since some biomarkers cannot directly cross the cell membrane, transport proteins are required to help these biomarkers enter the E. coli cells.
- Additionally, these biomarkers may need to first form a complex with certain proteins before they can act on the sensitive promoter and initiate downstream gene expression.
- To achieve these functions, we used a strong promoter to express auxiliary proteins, while also designing standard Golden Gate interfaces to allow future replacement of auxiliary proteins, enhancing the flexibility and adaptability of the module.
- R (Reporter assay) module:
- This module is mainly responsible for generating the signal, typically achieved by encoding a fluorescent protein downstream of a sensitive promoter. By detecting the fluorescent signal, the impact of the biomarker on the bacteria can be evaluated.
- Standard Golden Gate interfaces were also designed upstream and downstream of the sensitive promoter, facilitating other research teams in constructing their own sensitive promoters, thus promoting further research.
We conducted the following engineering validations based on this design:
- To validate the performance of the system, we tested it against three different biomarkers to ensure their feasibility and effectiveness under preset conditions.
- To enable simultaneous detection, we loaded three different fluorescent proteins (mTagBF2, sfGFP, and mKate2) into the R modules for the three biomarkers, and verified that their signals could be accurately detected.
- To ensure that there was no crosstalk between these three fluorescent proteins in the designed hardware system, we designed and conducted a fluorescence crosstalk experiment, specifically testing their crosstalk at different excitation-emission wavelengths.
- To enhance the system's flexibility, we designed a standard interface and, through modular assembly via this interface, successfully validated the ability to test a fourth biomarker, further expanding the system’s range of applications.
Module I: Uric Acid
We cultured the strains harboring the uric acid-responsive plasmids on LB agar plates corresponding to their resistance, and after obtaining monoclonal growth, we selected three monoclonal bacterial strains. These were then inoculated into LB broth with the appropriate resistance and cultured overnight. On the following day, the cultures were diluted into the logarithmic growth phase and supplemented with PBS solutions containing varying concentrations of uric acid, after which the cultures were further incubated for 16 hours. Post incubation, 150 μL of the culture medium was transferred to a TECAN plate reader to measure the fluorescence signal of mTagBF2 as well as the optical density (OD). The experimental results indicated that as the concentration of uric acid increased, the relative fluorescence values of the strains also increased, suggesting that the presence of uric acid can activate the expression of fluorescent proteins within the system. To more accurately obtain the response dynamics parameters of the system, we performed a Hill reaction fitting on the response curve.
Figure 1. The response of the HucR-YgfU-phucR system to uric acid solutions of different concentrations (0, 5, 10, 20, 50, 100, 200 µM).
Figure 2. The Hill equation fitting curve representing the response of the HucR-YgfU-phucR system to uric acid solutions of varying concentrations (0, 5, 10, 20, 50, 100, 200 µM).
1.Fitting Parameters:
Vmin=699.79
Vmax=12,395.12
K=72.97 mol-1L
n=3.17
R2=0.9893
2.Analysis:
High R2 value: The R2 value of 0.9893 indicates a very high degree of fit between the experimental data and the activated Hill equation model, suggesting that the model can effectively describe the relationship between uric acid concentration and the biosensor response.
Hill coefficient (n): The value of (n = 3.17) implies positive cooperativity in the binding process of uric acid. This means that the binding of one uric acid molecule enhances the binding ability of subsequent molecules, leading to a more sensitive response of the biosensor within certain concentration ranges.
Half-maximum effective concentration (K): The K value of 72.97 mol-1L indicates that the biosensor's response reaches half of its maximum at a uric acid concentration of 72.97 mol-1L. This reflects the sensitivity and optimal working range of the biosensor for uric acid.
Dynamic range: The values of Vmax and Vmin are 12,395.12 and 699.79, respectively, demonstrating that the biosensor has a wide detection range for uric acid, capable of detecting from low to high concentrations.
Conclusion
The results indicate that our system exhibits the highest sensitivity to uric acid at a concentration of 72.97 mol·L⁻¹ and has a broad detection range, capable of detecting uric acid from low to high concentrations. Additionally, we found that the binding process of uric acid exhibits positive cooperativity, which means that the binding of uric acid molecules enhances the binding ability of subsequent molecules, leading to a more sensitive response of our system within certain concentration ranges.
Module II: Glucose
Strains containing the glucose-responsive element were cultured on SOB agar plates corresponding to their resistance, and monoclonal colonies were selected. Three of these monoclonal colonies were then inoculated into M9 medium supplemented with the appropriate resistance and incubated overnight. On the subsequent day, the cultures were diluted to the logarithmic growth phase and treated with glucose solutions of varying concentrations, followed by a further 16 hours of incubation. Post incubation, 150 μL of the culture medium was transferred to a TECAN plate reader to measure the fluorescence signal of mKate2 as well as the optical density (OD). The experimental results indicated that as the concentration of glucose increased, the relative fluorescence values of the strains also increased, demonstrating that the presence of glucose can activate the expression of fluorescent proteins within the system. To more accurately determine the system's response dynamics parameters, a Hill reaction fitting was applied to the response curve.
Figure3. Response of the PI system to glucose solutions of varying concentrations (0, 1, 2, 4, 8, 16, 32, 64 mM).
Figure 4. Hill equation fitting curve for the PI system's response to glucose solutions at varying concentrations.
Fitting Parameters:
Vmin=4,023.77
Vmax=45,740.73
K=6.09 mol-1L
n=0.56
R2=0.9711
Analysis:
High R2 Value: An R2 value of 0.9711 indicates a high degree of fit between the experimental data and the activated Hill equation model, suggesting that the model provides a good description of the relationship between glucose concentration and the biosensor response.
Hill Coefficient (n): A value of n=0.56 implies negative cooperativity in the binding process of glucose. This indicates that the binding of one glucose molecule diminishes the binding affinity for subsequent molecules, resulting in a more gradual response curve. This could be attributed to the fact that glucose, as the primary energy source for cells, may influence the biosensor's response due to its metabolic processes.
Half-Maximum Effective Concentration (K): The K value of 6.09 mol-1L signifies that the biosensor's response reaches half of its maximum at a glucose concentration of 6.09 mol-1L. This reflects the biosensor's sensitivity and optimal operational range for glucose.
Dynamic Range: The values of Vmax and Vmin being 45,740.73 and 4,023.77, respectively, demonstrate a broad detection range for glucose by the biosensor. However, further optimization is required to enhance sensitivity at lower concentrations.
Conclusion:
The results show that our system exhibits the highest sensitivity to glucose at a concentration of 6.09 mol-1L and has a wide detection range. However, there is a need for further optimization to improve sensitivity at lower glucose concentrations. Additionally, the negative cooperativity observed in the binding process suggests that the binding of glucose molecules weakens the binding capacity for subsequent molecules, leading to a more gradual response curve. This could be due to the impact of glucose metabolism on the biosensor's response, given that glucose serves as the main energy source for cells.
Module III: Lactic Acid
Bacterial strains equipped with a lactate-responsive element were cultured on LB agar plates with appropriate antibiotics, yielding monoclonal colonies. Three such colonies were selected and inoculated into LB medium with corresponding antibiotics for overnight cultivation. On the subsequent day, the cultures were diluted to the logarithmic growth phase and exposed to lactate solutions of varying concentrations for a further 16 hours of incubation. After the incubation period, 150 μL aliquots of the culture medium were assessed for mKate2 fluorescence and optical density (OD) using a TECAN plate reader. The experimental data revealed a correlation between increased lactate concentration and elevated relative fluorescence values, suggesting that lactate presence triggers the expression of fluorescent proteins within the system. To accurately determine the kinetic parameters of the system's response, the response curves were subjected to Hill equation fitting.
Figure 5. Hill equation fitting curve for the ALPaGA system's response to lactic acid solutions at varying concentrations (0, 0.25, 0.5, 1, 2.5, 5, 10, 20, 40 µM).
Figure 6. Hill equation fitting curve for the ALPaGA system's response to lactic acid solutions at varying concentrations (0, 0.25, 0.5, 1, 2.5, 5, 10, 20, 40 µM).
1.Fitting Parameters:
Vmin=443.89
Vmax=1,992.40
K=4.14 mol-1L
n=1.85
R2=0.9939
2.Analysis:
High R2 Value: An R2 value of 0.9939 indicates an extremely high degree of fit between the experimental data and the activated Hill equation model, suggesting that the model can accurately describe the relationship between lactic acid concentration and the biosensor response.
Hill Coefficient (n): A value of n=1.85 implies positive cooperativity in the binding process of lactic acid. This indicates that the binding of one lactic acid molecule enhances the binding affinity for subsequent molecules, leading to a more sensitive response of the biosensor within certain concentration ranges.
Half-Maximum Effective Concentration (K): The K value of 4.14 mol-1L signifies that the biosensor's response reaches half of its maximum at a lactic acid concentration of 4.14 mol-1L. This reflects the biosensor's sensitivity and optimal operational range for lactic acid.
Dynamic Range: The values of Vmax and Vmin being 1,992.40 and 443.89, respectively, demonstrate a relatively narrow detection range for lactic acid by the biosensor. However, the response curve is quite steep, making it suitable for high-sensitivity detection within a specific concentration range.
Conclusion:
The results show that our system exhibits the highest sensitivity to lactic acid at a concentration of 4.14 mol-1L and has a narrow detection range but with a steep response curve, making it suitable for high-sensitivity detection within a specific concentration range. Additionally, the positive cooperativity observed in the binding process suggests that the binding of lactic acid molecules enhances the binding capacity for subsequent molecules, leading to a more sensitive response of our system to lactic acid within certain concentration ranges.
Cycle IV: Tryptophan
Strains containing the tryptophan-responsive element were streaked on LB agar plates with corresponding antibiotic resistance to obtain monoclonal colonies. Three of these colonies were selected and inoculated into LB medium with the appropriate antibiotics for overnight culture. On the next day, the cultures were diluted to the logarithmic growth phase and supplemented with tryptophan solutions of varying concentrations for an additional 16 hours of incubation. After incubation, 150 μL of the culture medium was measured for mKate2 fluorescence and optical density (OD) using a TECAN plate reader. The results showed a decrease in relative fluorescence values with increasing tryptophan concentration, indicating that tryptophan represses the expression of fluorescent proteins in the system. To accurately determine the kinetic parameters of the system's response, the response curves were fitted with the Hill equation.
Figure 7. Response of the TrpR-pTrp system to tryptophan solutions at varying concentrations (0, 0.1, 0.25, 0.5, 1, 2, 4 mM).
Figure 8. Hill equation fitting curve for the TrpR-pTrp system's response to tryptophan solutions at varying concentrations (0, 0.1, 0.25, 0.5, 1, 2, 4 mM).
1.Fitting Parameters:
Vmin=186.39
Vmax=134,991.95
K=0.03 mol-1L
n=1.11
R2=0.9594
2.Analysis:
High R2 value: The R2 value of 0.9594 indicates a high degree of fit between the experimental data and the inhibitory Hill equation model, suggesting that the model can effectively describe the relationship between tryptophan concentration and the biosensor response.
Hill Coefficient (n): A value of n=1.11 implies that the binding process of tryptophan is close to non-cooperative. This means that the binding of one tryptophan molecule has a minimal effect on the binding capacity of subsequent molecules, resulting in a more gradual response curve.
Half-maximum effective concentration (K): The K value of 0.03 mol-1L indicates that the biosensor's response reaches half of its maximum at a tryptophan concentration of 0.03 mol-1L. This reflects the biosensor's high sensitivity and optimal working range for tryptophan.
Dynamic Range: The values of Vmax and Vmin being 134,991.95 and 186.39, respectively, demonstrate a very wide detection range for tryptophan by the biosensor, capable of detecting from extremely low to high concentrations of tryptophan.
Conclusion:
The results show that our system exhibits the highest sensitivity to tryptophan at a concentration of 0.03 mol-1L and has an extremely wide detection range, capable of detecting tryptophan from very low to high concentrations. Additionally, the binding process of tryptophan is close to non-cooperative. This means that the binding of tryptophan molecules has a minimal impact on the binding capacity of subsequent molecules, leading to a more gradual response curve.
Fluorescence Crosstalk
The fluorescent proteins—mTagBF2, sfGFP, and mKate2—were cloned into the pUC57 kanamycin-resistant vector, with the same promoter and RBS utilized to regulate their expression. The plasmids were transformed into DH5α bacterial strains and monoclonal colonies were selected after cultivation on agar plates containing the appropriate antibiotics. These colonies were grown overnight in 5 mL of LB broth with corresponding antibiotics. Fluorescence measurements were then conducted using a TECAN Spark microplate reader, with the excitation spectra recorded for each protein at their specific wavelengths, resulting in the targeted protein fluorescing without interference from the others. These results demonstrate the absence of crosstalk between the different fluorescent proteins, providing a theoretical basis for the simultaneous triple-channel detection capability of our hardware.
Figure 9. The Impact of Three Types of Excitation Light on Three Fluorescent Proteins. Blue Fluorescence Channel: Excitation at 399 nm, Emission at 455 nm. Green Fluorescence Channel: Excitation at 465 nm, Emission at 525 nm. Red Fluorescence Channel: Excitation at 588 nm, Emission at 635 nm.
When different excitation lights are used to excite the three fluorescent proteins, the crosstalk effect between them is observed to be weak:
-
When excited with the excitation light for sfGFP, the emission intensity of sfGFP is significantly higher than the other two fluorescent proteins, with an emission intensity more than 100 times that of the others.
-
When excited with the excitation light for mTagBF2, the emission intensity of mTagBF2 is also markedly higher than the other two fluorescent proteins, with an emission intensity more than 20 times that of the others.
-
3. When excited with the excitation light for mKate2, the emission intensity of mKate2 is the most prominent, with an emission intensity more than 800 times that of the other fluorescent proteins.
These results indicate that when excited by their specific excitation lights, each fluorescent protein primarily exhibits its own emission characteristics, with minimal crosstalk between the three. Therefore, for future use in the same hardware, we can utilize these three proteins.