A simple, sensitive and colorimetric assay for Pseudomonas aeruginosa infection analysis

A variety of infections are brought on by the opportunistic Gram-negative bacteria Pseudomonas aeruginosa [1,2], such as infections in postpartum care. Immunocompromised postpartum women may experience a high rate of P. aeruginosa-related infections [3]. Early termination and quantification of P. aeruginosa is significant for reducing complications during nursing and improving the prognosis of patients with severe diseases, such as patients receiving chemotherapy. However, the sensitive and direct analysis of P. aeruginosa remains a challenge.

Many efforts have been made in the past few decades to detect and identify organisms and in vitro methodologies account for the majority of the established approaches [4–7]. Currently, colony counting and culturing remain the most traditional approach for pathogenic bacteria detection due to its unique capability in directly culturing and plating the organisms [5,8]. Colony counting and culturing is accurate and has been widely utilized in clinical applications. However, the method requires lengthy experimental time (>72 h) for bacteria culturing, which is time-consuming, and an incorrect sampling period can cause false evaluation of organisms and their growth responses [9–11]. Apart from colony counting and culturing, immunoassays have also been widely exploited in the identification of pathogenic bacteria [12,13]. However, immunoassays lack enough sensitivity and are time-consuming. Aptamers are nucleic acid sequences with functional secondary structures that can serve as novel recognition ligands with high specificity and affinity for their targets [14–16]. In recent years, a variety of biosensors have been proposed by utilizing aptamers to specifically identify pathogenic bacteria and convert bacteria signals to nucleic acid signals [17–19]. For example, Xu et al. proposed a fluorescent assay for accurate methicillin-resistant Staphylococcus aureus identification through dual-functional aptamer- and CRISPR-Cas12a-assisted rolling circle amplification [8]. Fluorescent assays are sensitive but require cumbersome equipment for result readout, which limits their further applications in resource-limited scenarios. Because the results of colorimetric methods can be seen with the naked eye, they have gained a lot of interest. However, colorimetric assays are not sensitive enough for the detection of trace amounts of bacteria. Therefore, there is an urgent demand to develop a simple, sensitive and colorimetric method for pathogenic bacteria detection.

Herein, the authors depicted a novel and simple colorimetric approach for sensitive P. aeruginosa detection by employing an allosteric probe to specifically bind with P. aeruginosa and to induce proximity ligation-assisted multiple-signal recycles. In this method, an allosteric probe is composed of a protein A aptamer and a linker sequence. The recognition of P. aeruginosa by an allosteric probe exposes the linker sequence. The linker sequence mediates the ligation of the hairpin probe (HP) and template sequence to initiate subsequent multiple-signal recycles, in which numerous G-rich sequences are produced to induce a color reaction. Based on its elegant design, this approach can be successfully applied for pathogenic bacteria detection with high sensitivity and reliability, showing promising prospects in early diagnosis of P. aeruginosa infection and, thus, reducing mortality and infection rates in nursing.

Materials & methods

Materials & reagents

The HPLC-purified DNA oligonucleotides used in this work are listed in Supplementary Table 1 and were bought from Shanghai Generay Biotech Co., Ltd. (Shanghai, China). DNA ligase (100 U), DNA polymerase (250 U) and Nb.BbvCI (1000 U) were purchased from Vazyme (Nanjing, China); 2,2′-azino-bis (3-ethylben-zothiazoline-6-sulfonic acid) (ABTS), hemin, MgCl₂·6H₂O and NaClO₄ were obtained from Sigma-Aldrich (MO, USA); H₂SO₄ and H₂O₂ were provided by Lingfengshiji Co., Ltd. (Shanghai, China). The bacteria strains used in this research, inducing Escherichia coli (ATCC 25922), Pseudomonas aeruginosa (ATCC 27853, P. aeruginosa) and Staphylococcus aureus (ATCC29213) were obtained from the laboratory of Shanghai Ruijin Hospital. Procedures for culturing the bacteria are in the Supplementary Experimental section. UV spectrophotometry (UV-2600, SHIMADZU, Japan) was used to record the absorbance and an Edinburgh FLSP920 spectrophotometer (Edinburgh, UK) was used to monitor the fluorescent spectrum.

Assembly & verification of allosteric probe

The synthesized allosteric probe (10 μM; 10 μl) was heated to 90°C and incubated for 10 min. Then, the solution was gradually cooled to room temperature. To explore the assembly of allosteric probes, the fluorescent dye FAM (carboxyfluorescein) and quenching dye BHQ (Black Hole Quencher) were labeled on both ends of the allosteric probes, respectively, and fluorescence changes before and after assembly of the allosteric probe were detected to assess whether the allosteric probes were assembled into a hairpin structure.

Proximity ligation of HP & template sequence

To test the proximity ligation of the HP and template sequence, the 3′ terminus of the template sequence and the 5′ terminus of the HP were labeled with FAM and BHQ, respectively. A 10-μl assembled allosteric probe (10 μM) was firstly incubated with 10 μl P. aeruginosa solution. The mixture was incubated at room temperature for 30 min. Afterward, 2 μl HP and 2 μl template sequence were added to the mixture. After being incubated for 30 min, the fluorescent signal was detected by a fluorescence spectrophotometer.

Feasibility analysis

A total of 10 μl allosteric probe was mixed with 2 μl target for 20 min, and 10 μl HP and template sequence were added into the system for 20 min. Then, 10 μl of T4 DNA ligase (0.5 U/l) was added, and the DNA ligase was inactivated by heating to 65°C for 10 min after 30 min incubation at room temperature. Then, phi29 DNA polymerase (2 μl; 2.5 U/l) and Nb.BbvCI endonuclease (2 μl’ 2.5 U/l) were added to the coculture for 60 min. Potassium ions (2 μl) and 2 μl hemin were added. Peroxidase activity was determined by mixing 180 μl ATBS-H₂O₂ substrate solution with 20 μl G-DNAzyme solution. Finally, the solution was analyzed by UV-visible spectroscopy and the peroxidase activity of these G-DNAzymes was evaluated by absorbance at 418 nm. Following the same procedures, the signals produced by the method were evaluated when DNA polymerase, endonuclease and hemin were absent.

Condition optimization

According to this procedure, different concentrations of ligase (0.1 U/l, 0.3 U/l, 0.5 U/l, 0.7 U/l and 0.9 U/l) were used to conduct the approach and absorbance was compared. Different concentrations of DNA polymerase (0.5 U/l, 1 U/l, 1.5 U/l, 2 U/l, 2.5 U/l and 3 U/l) and hemin (20 nM, 40 nM, 60 nM, 80 nM and 100 nM) were also compared.

Results & discussion

Working mechanism of established approach for P. aeruginosa detection

The working mechanism of the established approach is illustrated in Figure 1. In this method, an allosteric probe is designed to be composed of an F23 aptamer (an aptamer of the surface protein of P. aeruginosa) and a linker sequence. Particularly, the linker sequence is partially complementary to the F23 aptamer. In its original state, the allosteric probe is a hairpin structure, and the linker sequence is blocked by the aptamer from initiating the following signal amplification process: in the presence of P. aeruginosa, the F23 aptamer specifically binds with the target protein and is transformed, exposing linker sequences. The linker sequence is partially complementary to the 5′ terminus of the designed HP and 3′ terminus of the template sequence. The HP contains a primer in the 3′ terminus and an endonuclease enzyme recognizing site; the template sequence is C-rich. The proximity between the HP and the template sequence under the assistance of the linker sequence makes it possible to connect the gap by ligase. With the assistance of DNA polymerase, a sequence containing a G-rich section and nicking site is attached to the end of the primer, in which process the F23 aptamer P. aeruginosa is liberated from the complex to initiate the next signal cycle. The endonuclease recognizes the recognizing site and generates a gap in the nicking site. By cooperating with the DNA polymerase, numerous G-rich sequences are produced. The G-rich sequence then induces a color reaction. In detail, the G-rich sequence folds to the G-quadruplex structure with the assistance of K⁺, which cooperates with hemin to catalyze the ABTS-based color reaction.

Investigation of target recognition, proximity ligation & color reaction

The allosteric probe is capable of accurate identification of P. aeruginosa and assists in the proximity of the HP and template sequence. Therefore, we first tested the assembly of the allosteric probe and evaluated its capability in target recognition by a fluorescent assay. The mechanism of the fluorescent assay is shown in Figure 2A. In the linear state, the fluorescent signal of FAM was not quenched by BHQ, thus a high FAM signal was recorded in the system. After heating the linear probe to 90°C for 10 min and annealing it to room temperature, the allosteric probe was assembled to a hairpin structure, which was demonstrated by the significantly decreased FAM signal (Figure 2B). In the presence of target bacteria, the F23 aptamer was transformed and the linker sequence was released, leading to the recovery of the FAM signal, indicating that the allosteric probe can specifically bind with the target bacteria. To test the proximity ligation of the HP and template sequence, the 5′ terminus of the HP and the 3′ terminus of the template sequence were labeled with FAM and BHQ, respectively. When the F23 aptamer–P. aeruginosa complex was present, the FAM signal greatly decreased, indicating the proximity of the HP and template (Figure 2C). Meanwhile, the allosteric probe showed high stability in complicated experimental conditions (phosphate-buffered saline, serum, bovine serum albumin), indicating its potential in clinical applications. The feasibility of the whole sensing system was evaluated by comparing the color change. The result in Figure 2D shows no significant color change when the target bacteria, ligase, DNA polymerase, endonuclease and hemin were absent in the sensing system. Only when all components existed in the system could a color reaction be observed, indicating that the target bacteria, ligase, DNA polymerase, endonuclease and hemin were essential for performing the approach.

Optimization of experimental parameters

We then optimized the experimental parameters for better detection performance. The ligase catalyzes the ligation of the HP and template sequence and is important in initiating subsequent signal amplification. In this section, the absorbance at 450 nm by UV absorption spectrometer was detected to compare the detection performance of the approach. Therefore, we first compared the absorbance at 418 nm of the approach when detecting 10⁴ CFU/l P. aeruginosa with different concentrations of ligase. The result in Figure 3A shows gradually increased absorbance with the concentration of ligase ranging from 0.1 to 0.5 U/l. When the approach was incubated with more than 0.5 U/l ligase, no significant fluorescent signal could be observed, indicating that the 0.5 U/l ligase could provide the most significant signal response. Thus, 0.5 U/l ligase was selected in the following experiments. In addition, we optimized the concentration of DNA polymerase. According to the result in Figure 3B, 2.5 U/l DNA polymerase shows the highest amplifying efficiency and was used to induce chain extension in the approach. Hemin can potentially cause background signals since it exhibits peroxidase activity. Hemin concentration was optimized to lower the noise level. Following examination of a range of hemin concentrations, the concentration of 80 nM was chosen as the best concentration for the following step (Figure 3C).

Analytical performance of approach for P. aeruginosa detection

We investigated the detective response of the approach toward P. aeruginosa by recording the color change and the absorbance at 450 nm of the approach when detecting different concentrations of P. aeruginosa. With an increase in the concentration of P. aeruginosa, the colorimetric reaction rose (Figure 4A). The allosteric-probe–P. aeruginosa complex assisted the proximity ligation of the HP and template sequence and induced signal amplification to produce G-quadruplex access to the peroxidase substrate and subsequently caused a color change. In Figure 4B, the colorimetric absorbance and P. aeruginosa concentrations between 10² and 10⁷ CFU/ml are linearly correlated (R² = 0.9314). The limit of detection of the proposed method was determined at 45 CFU/ml according to the three-times deviation method. The method could not detect lower concentrations than 45 CFU/ml because of interference from background signals. The selectivity of the biosensor was investigated for P. aeruginosa toward S. aureus, E. coli and a mixture of the three bacteria. Equal concentrations and identical growth conditions of the three bacteria were used. This simple and effective colorimetric approach can potentially be a selective method for P. aeruginosa detection because, as shown in Figure 4C, the experiment indicated that all of these interfering bacteria have a small increase in colorimetric absorbance compared with P. aeruginosa. We then tested the clinical application of the method by detecting P. aeruginosa from phosphate-buffered saline samples and commercial serum samples. The results in Supplementary Figure 1 show that the detection results by the method in phosphate-buffered saline buffer and in commercial serum samples were statistically consistent, implying the method could be applied in clinical practice.

Conclusion

In summary, we exploited allosteric probe-based target recognition to mediate proximity ligation of the HP and template sequence and, thus, to induce signal amplifications and G-quadruplex mediated ATBS^2- based color reaction. Therefore, results can be observed directly by the color change of ABTS^2-. Notably, the method can be used for on-site analysis without a complicated device. Additionally, this technique offers a general, label-free colorimetric approach for nucleic acid amplification-based detection with a straightforward, sequence-specific response in a variety of domains, including environmental monitoring, food safety detection and medical diagnostics.

Future perspective

Sensitive and reliable detection of P. aeruginosa is crucial for the early diagnosis of skin and soft tissue infections in women receiving postpartum care. The authors described a colorimetric approach using a designed allosteric probe to specifically identify target bacteria and to induce proximity ligation for multiple-signal amplification. The approach exhibits a wide detection range and a low limit of detection of 45 CFU/ml. The advantages of the approach could include the high affinity and specificity between the aptamer sequence and target endowing the method with high selectivity, and the multiple-signal amplification process produced numerous G-rich sequences to induce color reaction enabling the sensitive detection of P. aeruginosa. Thus, the newly developed ligation-initiated multiple-signal amplification method may be used as a universal biosensing approach for on-site testing of pathogenic bacteria and assist in the early diagnosis of postpartum skin infections.