示例图片

Detective Engineering Bacteria: Rendering Pseudomonas aeruginosa Undetectable in Water


Abstract

Based on synthetic biology strategies, this study successfully constructed an engineered Escherichia coli biosensor and established a high-throughput detection method using a 96-well quantitative plate for the specific detection of the opportunistic pathogen Pseudomonas aeruginosa in water. The sensor achieves high specificity through a triple-verification mechanism: recognition of the Pseudomonas aeruginosa-specific quorum-sensing molecule PQS, detection of the characteristic virulence factor pyocyanin (PYO), and selective enrichment via triclosan resistance. The reporting system employs the lacZ and gus genes, which catalyze the formation of yellow chromogenic and blue fluorogenic signals from the substrates ONPG and MUG, respectively. Results demonstrated that the engineered bacterial chassis exhibits a clean background with highly specific sensing elements capable of responding with high sensitivity to both PQS and PYO. In practical water sample detection, the method utilizes a 96-well quantitative plate format combined with the MPN (Most Probable Number) technique for accurate quantification. The results are in excellent agreement with those obtained using the national standard method, while the detection time is reduced from 48–72 hours to 6–12 hours. This study provides an accurate and practical new strategy for the rapid, low-cost on-site monitoring and high-throughput screening of Pseudomonas aeruginosa.


1. Validation of the gus Enzyme Substrate Detection Platform

The quantitative detection platform is based on a well-established enzyme substrate methodology, in which specific enzymatic activity of the target bacterium catalyzes colorless substrates to generate measurable chromogenic or fluorogenic products. To confirm the platform's applicability, we initially tested various water samples using a commercial enzyme substrate kit.

Samples included sterile water, wild-type E. coli suspension, environmental water samples, and commercially bottled water. Experiments were performed according to standard protocols using β-galactosidase (lacZ)- and β-glucuronidase (gus)-based commercial reagents [2]. Results demonstrated successful qualitative detection of E. coli in all sample types. The wild-type E. coli suspension and river water sample exhibited high E. coli counts, whereas sterile water and bottled water samples showed no detectable E. coli (Figure 1). This preliminary validation confirmed the stability and reliability of the enzyme substrate method across different water matrices, providing a robust methodological foundation for our detection system.

示例图片

Figure 1. Validation of the enzyme substrate method applicability in various water samples.

This figure presents the results of Escherichia coli detection in various water samples using a commercial enzyme substrate assay kit.


2. Genetic background engineering and biosafety containment validation of the EC1000 chassis strain

To ensure reporter signals originate exclusively from engineered plasmids rather than endogenous host activity, we constructed a clean chassis lacking both lacZ and gus. Additionally, to prevent plasmid dissemination into the environment, we incorporated a stringent physical containment system. E. coli EC1000 was selected as the parental strain due to its natural lacZ deficiency and genomic encoding of the RepA protein required for pORI280 replication, making it ideal for biosensor development. We then performed systematic genetic modifications and comprehensive functional validation.

We validated the endogenous β-galactosidase activity of EC1000 and its derivatives using both qualitative and quantitative approaches. Initially, blue–white screening was performed, which revealed that wild-type Escherichia coli ATCC 25922 (positive control) formed blue colonies, whereas EC1000 and EC1000 Δgus colonies remained white (Figure 2A), indicating a lack of X-gal hydrolysis capability in the latter. For more precise confirmation, β-galactosidase activity was assessed using ONPG (o-nitrophenyl-β-D-galactopyranoside) as the substrate. Hydrolysis of ONPG by β-galactosidase yields yellow o-nitrophenol. As shown in Figure 2B, ATCC 25922 exhibited a distinct yellow coloration, whereas EC1000 and EC1000 Δgus showed no significant color change.

However, the gus gene in the EC1000 genome remains active, which would interfere with fluorescence detection using gus as the reporter. To address this, we precisely deleted the gus gene from the EC1000 genome using a homologous recombination system, successfully generating the double-deletion strain EC1000 Δgus. To confirm the deletion, EC1000 wild-type, EC1000 Δgus, and ATCC 25922 were cultured in liquid medium containing MUG substrate and visualized under a 365 nm UV lamp. As shown in Figure 2C, EC1000 wild-type and ATCC 25922 emitted strong blue fluorescence, whereas the EC1000 Δgus mutant produced no fluorescence, confirming successful gus deletion and providing a clean chassis with no background interference for the fluorescence reporting system.

To ensure that the genetic modifications did not adversely affect the basic physiological functions of the engineered bacteria, we determined the growth curves of the three strains. The results (Figure 2D) showed no significant differences in growth rate or final cell density between EC1000 Δgus and its parental strain EC1000 or ATCC 25922, indicating that deletion of the gus gene did not impact normal bacterial growth and confirming the suitability of this chassis for subsequent sensor circuit integration and application.

To evaluate the effectiveness of the physical containment strategy based on the pORI280 conditional origin of replication, we electrotransformed test plasmids carrying the pORI280 origin of replication [3] (and containing triclosan resistance) into E. coli TG1, E. coli ATCC 25922, EC1000, and EC1000 Δgus, using a vector with the pBlue origin of replication as a control. Spot plating results (Figure 2E) showed that the plasmid could only form colonies in EC1000 and its derivative EC1000 Δgus, which provide the RepA replication protein, but failed to grow in TG1 and ATCC 25922. This result strongly demonstrates that our "lock-and-key" plasmid replication system strictly restricts plasmid replication to the engineered host strain, effectively preventing horizontal gene transfer to other environmental bacteria and thereby enhancing biosafety at the physical level.

In conclusion, we successfully obtained an engineered chassis strain EC1000 Δgus with a clean genetic background (ΔlacZ Δgus), normal growth characteristics, and stringent biosafety containment, providing a solid foundation for the construction of high-performance biosensors.

示例图片
Figure 2. Validation of genetic engineering and biosafety containment of the chassis strain.

(A) Blue–white screening results showed that wild-type Escherichia coli ATCC 25922 (positive control) formed blue colonies, whereas the engineered chassis strains EC1000 and EC1000 Δgus formed white colonies, indicating the absence of endogenous β-galactosidase activity. (B) The ONPG liquid colorimetric assay further confirmed that only ATCC 25922 could hydrolyze ONPG to produce a yellow product, whereas the EC1000 series strains exhibited no such activity. (C) The MUG fluorescence assay, observed under ultraviolet light, showed that EC1000 wild-type and ATCC 25922 emitted blue fluorescence, while the EC1000 Δgus mutant produced no fluorescence, confirming successful deletion of the gus gene. (D) Growth curve analysis indicated no significant difference in growth rate between the engineered strain EC1000 Δgus and its parental strain, demonstrating that the genetic modification did not affect normal physiological functions. (E) Physical containment validation experiments showed that the test plasmid based on the pORI280 conditional origin of replication could only grow in engineered host strains (EC1000 and its derivatives) that provide the RepA protein, confirming that this system effectively prevents horizontal plasmid transfer.


3. Selective Pressure System Validation

To ensure high specificity of the detection system in complex microbial environments, we developed a selective pressure system based on dual antibiotics (triclosan and kanamycin). The primary objectives were: first, to verify that the triclosan resistance gene fabV had been successfully introduced into the chassis strain and was functioning effectively; and second, to demonstrate that triclosan could efficiently eliminate the vast majority of common environmental bacteria, allowing only the engineered strain and the target bacterium—Pseudomonas aeruginosa—to survive.

To achieve these goals, nine representative strains were selected, including the engineered strain, multiple control strains, and common environmental bacteria (EC1000Δgus Pcat-fabV [engineered strain], S17-1, TG1, E. coli ATCC 25922, EC1000Δgus, EC1000, Salmonella enterica, Staphylococcus aureus, and Pseudomonas aeruginosa PAO1). Spot plating experiments were performed on LB agar plates with and without triclosan (6 μg/mL). The results, as shown in Figure 3, demonstrated that on triclosan-containing plates, only the engineered strain carrying the fabV gene (EC1000Δgus Pcat-fabV) and the naturally tolerant Pseudomonas aeruginosa PAO1 were able to grow. In striking contrast, the parental strains EC1000Δgus and EC1000, as well as all other tested strains, were completely inhibited. This result conclusively confirmed that the fabV resistance gene had been successfully integrated into the engineered strain and was functioning as intended, thereby achieving the genetic modification objectives for the chassis. Furthermore, all other tested strains, including common pathogens such as Salmonella enterica and Staphylococcus aureus, as well as other Escherichia coli strains that may be present in the environment (e.g., S17-1, TG1, ATCC 25922)—were effectively eliminated.

In conclusion, this experiment not only successfully validated the functional expression of triclosan resistance in the engineered strain but also confirmed that the antibiotic-based strategy could create a highly specific selection environment, minimizing false-positive interference and providing a fundamental guarantee for the specificity and reliability of subsequent biosensing detection.

示例图片
Figure 3. Validation of the triclosan-based selective pressure system.

Spot plating experiments compared the growth of nine representative strains on LB agar plates without antibiotics (left) and with triclosan (6 μg/mL, right). The results clearly showed that on triclosan-containing plates, only the engineered strain carrying the fabV resistance gene (EC1000Δgus Pcat-fabV) and the naturally tolerant Pseudomonas aeruginosa PAO1 were able to grow, while all other tested strains were completely inhibited. This demonstrates the effectiveness and high specificity of the selective pressure system.


4. Vector Element Validation

To ensure that the function of each genetic element in the biosensor system meets the design expectations, we conducted functional validation of both the promoter elements in the sensing circuit and the inducible resistance system.

4.1 Sensor Element Specificity

In addition to the resistance elements validated in section 1.3, the core of the sensing circuit lies in the promoters that specifically respond to PQS or PYO. To ensure the correct functionality of the promoters (PpqsA, PmexG, PphzA1, and PbrlR) selected from Pseudomonas aeruginosa, we first verified their activity within the native P. aeruginosa host.

The aforementioned promoters were individually ligated to a promoterless lacZ reporter gene to construct recombinant plasmids, which were then electrotransformed into P. aeruginosa PAO1. Subsequently, exogenous PQS or PYO was added to the culture medium for induction, and β-galactosidase activity (Miller Units) was measured to quantify promoter strength. The results showed that the PpqsA promoter activity was significantly upregulated in the presence of PQS, while PmexG, PphzA1, and PbrlR exhibited varying degrees of activation under PYO induction (Figures 4A–D). This experiment confirmed the correct functionality and induction specificity of these promoter elements in their native host, thereby providing reliable components for subsequent sensing circuit construction in engineered bacteria.

示例图片
Figure 4. Functional specificity validation of sensor elements (promoters) in the native host.

Candidate promoters (PpqsA, PmexG, PphzA1, and PbrlR) derived from Pseudomonas aeruginosa were ligated to the lacZ reporter gene and introduced into P. aeruginosa PAO1. The response specificity was quantified by measuring β-galactosidase activity (Miller Units). (A) PpqsA promoter activity was significantly upregulated following the addition of its signaling molecule PQS. (B–D) PmexG, PphzA1, and PbrlR promoter activities were all effectively activated upon the addition of their signaling molecule PYO. This experiment confirmed the correct functionality of these promoters in their native host.

4.2 Functional Validation of the Inducible Resistance System

To achieve precise control of the engineered bacteria and enhance their reliability in practical applications, we introduced a triclosan resistance gene (fabV) regulated by an arabinose-inducible promoter (pBAD) into the plasmid system. The design is such that fabV is expressed only in the presence of arabinose, conferring triclosan resistance; in the absence of arabinose, no resistance is expressed, and the bacteria cannot grow in a triclosan-containing environment.

To validate the system, the plasmid carrying the pBAD-fabV expression cassette was introduced into the EC1000 Δgus strain, and spot plating experiments were performed on LB agar plates containing triclosan (3 μg/mL) with or without arabinose (0.1%). The results, as shown in Figure 5, demonstrated that the engineered bacteria grew normally in the presence of arabinose, whereas no colonies formed in the absence of arabinose. This clearly indicates that the arabinose promoter effectively regulates fabV expression, thereby achieving inducible control of exogenous resistance.

The successful construction of this inducible resistance system not only provides a more flexible means of selectively enriching engineered bacteria in complex environments but also further reduces the environmental safety risks associated with gene leakage.

示例图片
Figure 5. Functional validation of the arabinose-inducible triclosan resistance system.

A: Arabinose-inducible triclosan resistance system.

B: The engineered bacteria carrying the pBAD-fabV system were spotted on LB agar plates containing triclosan (3 μg/mL) with or without 0.1% arabinose. The results showed that the engineered bacteria grew normally only on the plates supplemented with arabinose, whereas no growth was observed in the absence of arabinose, confirming the stringent regulatory function of the inducible resistance system.


5. Detection System Vector Construction

After completing the functional validation of each vector element (including promoters, reporter genes, and resistance systems), we proceeded to construct the vector backbone for the final detection system. This backbone was designed to integrate both the PQS-sensing module and the PYO-sensing module, ensuring their stable coexistence without mutual interference, while retaining essential selective pressure and biosafety control elements.

A modular assembly strategy was adopted to construct two core plasmids: pAZ2 and pBZ2. The pAZ2 plasmid utilizes the conditional origin of replication from pORI280, whose replication strictly depends on the RepA protein provided by the engineered host strain EC1000 Δgus, thereby effectively preventing the leakage of the plasmid into environmental strains.

The vector construction process is schematically illustrated in Figures 6A and 6B. Each functional module was sequentially assembled using restriction enzyme digestion and ligation, and the correctness and orientation of all elements were verified by DNA sequencing.

The successful construction of this dual-plasmid backbone provides a stable and controllable genetic platform for subsequent dual-signal detection of PQS and PYO, and lays the foundation for further optimization of detection performance.

示例图片
Figure 6. Schematic illustration of the construction of the dual-plasmid sensing system vectors.

A: Sensing plasmid pAZ2; B: Reporter plasmid pBZ2.


6. PQS Sensing System Validation

To achieve efficient detection of the Pseudomonas aeruginosa-specific quorum-sensing molecule PQS, we constructed a sensing circuit based on the PqsR protein and its cognate promoter PpqsA. This circuit uses lacZ as the reporter gene, whose expression product—β-galactosidase—catalyzes the hydrolysis of the colorless substrate ONPG to produce yellow o-nitrophenol, thereby enabling visual detection of the PQS signal (Figure 7A).

To validate the functionality of this circuit, the engineered bacteria harboring the PqsR–PpqsA sensing module were exposed to various concentrations of PQS, and the ONPG substrate was added to initiate the color reaction. The results, as shown in Figure 7B, demonstrated that in the PQS-treated groups, the solution developed a distinct yellow color over the course of incubation, whereas the control group without PQS remained colorless. This result clearly indicates that the PqsR–PpqsA sensing circuit can be specifically activated by PQS, successfully driving the expression of the lacZ reporter gene and ultimately generating an easily interpretable detection signal through a visible color change.

In conclusion, the PqsR–PpqsA sensing circuit efficiently and specifically responds to the PQS signal and achieves detection through an intuitive color change, providing a reliable PQS-sensing module for the subsequent construction of a complete dual-signal detection system.

示例图片
Figure 7. Colorimetric validation of the PQS sensing system.

A: Schematic illustration of the PqsR–PpqsA sensing circuit. B: The engineered bacteria carrying the PqsR–PpqsA sensing circuit were incubated in culture medium containing the ONPG substrate and exposed to PQS (right) or DMSO as a control (left). The results showed that only the PQS-treated group developed a yellow color, indicating successful activation of the β-galactosidase reporter pathway and confirming the chromogenic detection of the PQS signal.


7. PYO Sensing System Optimization and Validation

To optimize the detection performance for pyocyanin (PYO), a characteristic virulence factor of Pseudomonas aeruginosa, we implemented the “choose-one-from-multiple” strategy proposed during the design phase and conducted a systematic comparative analysis of three parallel PYO-sensing circuits. These circuits were constructed using different transcriptional regulatory proteins (SoxR, MexL, BrlR) and their corresponding specific promoters (PmexG, PphzA1, PbrlR). Through molecular docking simulations and experimental validation, our objective was to identify the sensing module with the best response performance.

7.1 Molecular Docking Analysis

We first employed molecular docking to simulate the binding modes and affinities between PYO and the key regulatory proteins (SoxR, MexL, BrlR) in the three circuits. The docking results, as shown in Figure 8, indicated that the regulatory protein BrlR exhibited the lowest binding energy with PYO (−7.26 kcal/mol), suggesting the strongest theoretical binding affinity. The binding energies of MexL and SoxR with PYO were −7.10 kcal/mol and −5.79 kcal/mol, respectively, showing a gradual decrease in theoretical affinity.

Although molecular docking suggests that BrlR may have an advantage at the protein–ligand binding level, transcriptional activation efficiency is influenced by multiple factors, including protein expression levels, the intracellular environment, and the intrinsic strength of the promoter. 

示例图片
Figure 8. Molecular docking analysis of the binding modes and affinities between PYO and various regulatory proteins.

7.2 Experimental Detection Performance Comparison

To evaluate the performance of the three circuits in the actual sensing system, we introduced the complete sensing modules—SoxR–PmexG, MexL–PphzA1, and BrlR–PbrlR—into the engineered bacteria. Under induction with the same concentration of PYO, their performance was assessed by quantitatively measuring the fluorescence intensity generated by the gus reporter gene. The working principles are illustrated in Figures 9A–C.

The results, as shown in Figure 9D, indicate that the SoxR–PmexG circuit exhibited the highest fluorescence signal intensity and the lowest background noise in practical detection, demonstrating the best sensitivity and reliability. The MexL–PphzA1 circuit displayed basal fluorescence in the absence of PYO, and this fluorescence was significantly reduced upon the addition of PYO (dissolved in DMSO), representing a typical repressive response pattern; however, the extent of reduction was not particularly pronounced. Although the BrlR–PbrlR circuit showed the most favorable binding energy in molecular docking, it produced the weakest reporter signal in the actual system, possibly due to insufficient protein expression or lower transcriptional activation efficiency.

示例图片
Figure 9. Performance comparison of different PYO sensing circuits in actual detection.

A: Working principle of the SoxR–PmexG sensing circuit. B: Working principle of the MexL–PphzA1 sensing circuit. C: Working principle of the BrlR–PbrlR sensing circuit. D: Fluorescence intensity generated by the gus reporter gene in different sensing circuits under induction with the same concentration of PYO.

Integrating the results of molecular docking and experimental validation, we conclude that although the BrlR protein exhibits the highest theoretical binding capacity, the actual performance of its associated circuit is suboptimal. The MexL–PphzA1 circuit demonstrates a repression-type response profile, making it suitable for negative verification under specific contexts. In contrast, the circuit composed of SoxR–PmexG exhibits the most excellent sensitivity and reliability in practical detection. Therefore, the SoxR–PmexG module is selected as the core component for PYO sensing in subsequent sensing system development.


8. Screening of Sealing Films for 97-Well Quantitative Plates Based on Oxygen Supply Optimization

The production of the quorum-sensing molecule PQS and the virulence factor pyocyanin (PYO) by Pseudomonas aeruginosa occurs during the late growth phase at high cell density and is strictly aerobic. However, under conventional sealed culture conditions, dissolved oxygen in the medium is rapidly depleted before the late growth phase, severely limiting the synthesis of PQS and PYO and thereby compromising detection reliability. To address the insufficient dissolved oxygen supply during stationary culture—particularly in the late growth phase—which restricts the production of the key signaling molecules PQS and PYO, we screened and optimized the sealing film materials for 97-well quantitative plates.

In this study, five materials were compared for their performance: EPTFE, nanoscale breathable film for architectural doors and windows, PVDF, PTFE, and TPU. The results showed that EPTFE film exhibited the best performance in maintaining a sterile environment while supporting efficient oxygen transfer. Cultures sealed with EPTFE developed the characteristic green coloration indicative of PYO production. PVDF film supported partial PYO synthesis but resulted in unstable culture conditions due to water evaporation. In contrast, PTFE and TPU films failed to support PYO production effectively due to insufficient gas permeability (Figure 10).

In conclusion, by adopting the highly breathable EPTFE sealing film, we successfully established a reliable system suitable for small-volume, high-throughput cultivation, providing critical technical support for subsequent PYO-based sensing and detection.

示例图片
Figure 10. Screening of sealing membrane materials for 96-well quantitative plates based on oxygen supply optimization.

CK: Control group using the 97-well plate's original heat-sealable aluminum foil film.

a–e: Experimental groups using polyvinylidene fluoride (PVDF) membrane, polytetrafluoroethylene (PTFE) membrane, expanded polytetrafluoroethylene (EPTFE) membrane, nanoscale breathable film for architectural doors and windows, and thermoplastic polyurethane (TPU) film, respectively.


9. Construction of the Engineered Bacteria and Validation of Dual-Response to Target Signaling Molecules

Following the successful validation of each individual component, we constructed the complete dual-plasmid sensing vectors, designated pAZ2-1 and pBZ2-1, which contain the PQS-sensing circuit and the PYO-sensing circuit, respectively (Figure 11). 

示例图片
Figure 11. Schematic diagram of the dual-plasmid sensing vector

A: Schematic representation of the pAZ2 vector map.

B: Schematic representation of the pBZ2 vector map.

Since both plasmids carry the triclosan resistance gene fabV, sequential transformation into the EC1000 Δgus host presented a conflict in antibiotic selection. To resolve this issue, we leveraged the arabinose-inducible expression of fabV in pAZ2. First, under arabinose-supplemented conditions, we successfully introduced pAZ2-1 into EC1000 Δgus via triclosan resistance selection. Subsequently, in the absence of arabinose, we utilized the constitutively expressed fabV in pBZ2-1 to perform a second triclosan resistance selection, introducing pBZ2-1 into the engineered strain already harboring pAZ2-1. This process yielded the final engineered strain carrying the complete dual-plasmid system. Following the introduction of the complete dual-plasmid system (comprising the PQS-sensing circuit and the PYO-sensing circuit) into the EC1000 Δgus host, the final detection strain, designated EC1001, was successfully constructed. Its working principle is illustrated in Figure 12.

示例图片
Figure 12. Schematic diagram of the working principle of the EC1001 detection system

To verify whether EC1001 functions according to the designed logic—i.e., responding to PQS and PYO separately and producing corresponding signals—mid-logarithmic phase cultures of the engineered strain were aliquoted into three groups of test tubes: one supplemented with DMSO, one with PQS (dissolved in DMSO), and one with PYO. All tubes were simultaneously added with the substrates ONPG and MUG, followed by incubation at 37°C in the dark. Observations were made after 10 hours of incubation.

The results, as shown in Figure 13, demonstrated that the PQS-treated group developed a yellow color (indicating the hydrolysis of ONPG by β-galactosidase) but exhibited no significant fluorescence under UV light. In contrast, the PYO-treated group showed no color change but emitted strong blue fluorescence under UV light (indicating the hydrolysis of MUG by β-glucuronidase). The DMSO control group showed neither color nor fluorescence changes throughout the experiment.

These results confirm that the engineered strain can specifically distinguish between and respond to PQS and PYO signals, producing observable colorimetric and fluorescent outputs through independent reporter pathways. This successfully achieves the "dual verification" logic as designed.

示例图片
Figure 13. Dual-response validation of the engineered bacteria to target signal molecules.

The engineered bacteria were exposed to DMSO (solvent control), PQS, and PYO, respectively, with the simultaneous addition of ONPG (chromogenic substrate) and MUG (fluorescent substrate) for incubation. Upper panel (visible light): Only the PQS-treated group developed a yellow color, indicating that the β-galactosidase reporter pathway was specifically activated. Lower panel (UV light): Only the PYO-treated group emitted strong blue fluorescence, indicating that the β-glucuronidase reporter pathway was specifically activated. The DMSO control group showed no signal throughout the experiment. These results directly confirm that the engineered strain can distinguish between PQS and PYO and performing "dual verification" as designed.

10. Performance Evaluation in Actual Water Samples: Comparison with the National Standard Method

The ultimate goal of this study was to develop a highly accurate biosensor system. To evaluate the engineered bacterial detection method, we selected five representative water samples: laboratory tap water, office drinking water, swimming pool water, residential building tap water, and sterile water. Each sample was divided into two aliquots. One aliquot was tested using the 97-well quantitative plate method developed in this project (containing engineered bacteria, substrates, and triclosan), incubated at 37°C for 12 hours, and the number of dual-signal positive wells was recorded (Figure 14A). The concentration was then calculated using the Most Probable Number (MPN) method. Figure 14B presents the MPN lookup table for converting the number of positive wells into bacterial concentration. The other aliquot was tested in strict accordance with the national standard method (GB/T 8538-2016, membrane filtration method).

The calculation of results using the national standard method included the enumeration and confirmation of both typical and atypical colonies. The formula for calculating the total number (N) of confirmed Pseudomonas aeruginosa colonies on the filter membrane is:

N = P + F(cF / nF) + R(cR / nR)

where:

P = number of confirmed positive typical colonies

F and R = total number of typical and atypical colonies on the filter membrane, respectively

cF and cR = number of colonies selected for confirmation from the typical and atypical colonies, respectively

nF and nR = number of confirmed positive colonies among the selected typical and atypical colonies, respectively

The final concentration is reported as CFU/250 mL, calculated using the formula:

(N × 250) / Volume of filtered water (mL)

Table 1 presents the detailed raw data (P, F, cF, nF, R, cR, nR) and final calculated results for each water sample.

The quantitative comparison between the two methods is shown in Figure 14C. Statistical analysis indicated no significant difference between the detection results of the engineered bacterial method and the national standard method (p > 0.05). This demonstrates that the biosensor system maintains the same level of accuracy as the "gold standard" method while reducing the detection time from 48–72 hours to less than 12 hours, thereby exhibiting excellent potential for on-site applications.

Table 1: Raw data and quantitative results for Pseudomonas aeruginosa detection in different water samples using the national standard method (GB/T 8538-2016)

Water body The colony number of Bacteria P F cF nF R cR nR The colony number of PA CFU/250ml
Laboratory tap water 81 2 12 3 12 27 6 18 7.25
Office drinking water 4 0 0 0 0 0 0 0 0
Swimming pool water 63 1 12 1 6 43 1 20 5.15
Residential building tap water 112 1 18 2 6 83 2 20 15.3
Sterile Water 2 0 0 0 0 0 0 0 0

示例图片
Figure 14. Comparison of actual water sample detection performance with the National Standard Method.

(A) Schematic representation of the 96-well plate results for detecting different water samples using the engineered bacterial method developed in this project (photographed under visible and UV light). The color or fluorescence change in each well represents the result of one detection unit. (B) MPN (Most Probable Number) lookup table used to convert the number of positive wells into bacterial concentration. (C) Comparative graph of quantitative detection results for Pseudomonas aeruginosa in five water samples using the engineered bacterial method and the national standard method (GB/T 8538-2016). Statistical analysis indicates no significant difference between the results of the two methods (p > 0.05), while the engineered bacterial method reduces the detection time from 48–72 hours (national standard method) to less than 12 hours.

Conclusion and Discussion

In this study, we successfully constructed and systematically validated a highly specific, rapid, and quantitative whole-cell biosensor for the detection of Pseudomonas aeruginosa in water. Based on synthetic biology strategies, this sensor achieves highly specific identification of the target bacterium through a triple-verification mechanism—sensing the quorum-sensing molecule PQS unique to P. aeruginosa, detecting the characteristic virulence factor pyocyanin (PYO), and enabling selective enrichment through triclosan resistance. The reporting system employs the lacZ and gus genes to produce chromogenic and fluorescent signals, respectively, offering the advantages of intuitive, dual-channel verification.

Through rigorous genetic engineering, we obtained the engineered chassis strain EC1000 Δgus, which exhibits a clean background (ΔlacZ Δgus), normal growth, and strict biosafety containment capabilities. This effectively avoids endogenous enzyme interference and the risk of environmental dissemination of engineered plasmids. Validation of the selective pressure system further ensures high specificity and anti-interference capability of the detection system in complex microbial environments.

In actual water sample testing, the results obtained by this method were highly consistent with those of the national standard method (GB/T 8538-2016), showing no significant difference, while the detection time was dramatically reduced from 48–72 hours to 6–12 hours, significantly improving detection efficiency. Furthermore, this method requires no complex instrumentation, is cost-effective, and simple to operate, demonstrating excellent potential for on-site applications.

In conclusion, this study not only provides an accurate, rapid, and low-cost new strategy for detecting P. aeruginosa, but also offers a viable model for the application of synthetic biology in environmental microbial monitoring. In the future, further optimization of signal amplification circuits and the development of portable reading devices could promote the industrial application of this biosensor in practical water management, public health monitoring, and other scenarios. 

References

[1] Flores SS, Clop PD, Barra JL, Argaraña CE, Perillo MA, Nolan V, Sánchez JM. His-tag β-galactosidase supramolecular performance. Biophysical chemistry, 2022, 281: 106739.

[2] Sarhan HR, Foster HA. A rapid fluorogenic method for the detection of Escherichia coli by the production of beta-glucuronidase. The Journal of applied bacteriology, 1991, 70(5): 394-400.

[3] Leenhouts K, Buist G, Bolhuis A, ten Berge A, Kiel J, Mierau I, Dabrowska M, Venema G, Kok J. A general system for generating unlabelled gene replacements in bacterial chromosomes. Molecular & general genetics : MGG, 1996, 253(1-2): 217-224.

Despite the significant achievements of this study, certain limitations remain. For instance, although the engineered bacterial method demonstrates high consistency with the national standard method in actual water sample detection, its sensitivity for detecting extremely low concentrations of Pseudomonas aeruginosa still requires further improvement. Future research could consider optimizing the design of sensing elements to enhance their responsiveness to low concentrations of signaling molecules.

Furthermore, this study only verified the biosensor's performance in several common types of water samples. Its applicability to more complex aquatic environments, such as industrial wastewater containing high levels of organic matter or heavy metals, still needs to be further evaluated. Addressing these issues will require exploring the introduction of more efficient signal amplification mechanisms or implementing targeted modifications to the engineered bacteria to enhance their stability and detection performance in complex environments.

Additionally, more extensive testing with various types of water samples should be conducted to comprehensively assess the biosensor's application range. Through continuous improvement and optimization, it is expected that the application scenarios of this biosensor can be further expanded, providing more powerful technical support for environmental microbial monitoring.