Application of F0F1‑ATPase immuno‑biosensors for detecting Escherichia coli O157:H7

Introduction

Escherichia coli (E. coli) O157:H7 is a type of pathogenic bacterium that infects humans and livestock primarily through contaminated food. It can cause abdominal pain, hemorrhagic fever or bloody stools, and can induce secondary haemolytic uraemic syndrome in infants, preschool children or weak, elderly individuals. In addition, due to its strong drug resistance, it is very hard to eliminate O157:H7 from contaminated food sources. O157:H7 contamination has now become an international food security concern (1). The American Centers for Disease Control has revealed that E. coli O157:H7 is one of the major pathogenic bacteria causing food-borne diseases; thus, poses a serious threat to public health. Furthermore, this strain has been detected in pork, beef and mutton in China (2).

There are several detection methods currently used for pathogenic bacteria, including culture-based, enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR)-based methods (3–5). However, these methods are usually time-consuming, expensive and insensitive, which makes them unsuitable for the detection of this pathogen. Therefore, it is necessary to develop more efficient detection apparatus.

F0F1-ATPase, located in the mitochondria and/or the chloroplast thylakoid of eukaryotic organisms and the bacterial plasma membrane, catalyzes the synthesis of ATP using the transmembrane proton gradient. In E. coli, the soluble F1 and transmembrane F0 parts are comprised of the α3β3γδε and ab2cn subunits, respectively. These two parts are connected by the stalks of γε in the centre and b2 δ on the outside. When the downhill proton passes through F0, the c and γε subunits are rotated leading to conformational changes in F1, which promotes the formation of ATP from ADP and inorganic phosphate, and vice versa. As a result, F0F1-ATPase forms a molecule size motor, which can transform the electric chemical potential energy into chemical energy. If this process is disturbed by other factors, the rate of ATP synthesis and proton flux maybe altered; this phenomenon maybe reflected by pH sensitive substances (6–7). F1300 is a pH sensitive fluorescent probe that can be used as an indicator of changes in the pH of F0F1-ATPase. During ATP synthesis, protons are pumped out of the chromatophore and this transfer of protons is detected by F1300 (4–5,8). This concept was used to construct the immuno-rotary biosensor (IRB) for detecting specific targets, which achieved great success (4–5,9–10). Although this type of biosensor has been used to detect a virus (3,6–7), detection of much larger antigen such as a single bacterium has not been reported. The aim of the present study was to investigate the potential use of this method for the detection of E. coli O157:H7.

Materials and methods

Bacterial strains

E. coli O157:H7 (strain no. ATCC35150) was obtained from the Guangdong Microbial Culture Collection Center (Guangzhou, China) and was incubated in nutrient broth medium at 37°C for 24 h. The bacterial suspension was then diluted to 10−4, 10−5 and 10−6 in bacteria-free PBS. A total of 100 µl was transferred to a panel for further cultivation and each dilution gradient sample was tripled. The bacterial clone in the incubated sample was counted 24 h later. Surplus bacteria were in activated by heating to 80°C for 1 h, then 10 ml was centrifuged for 30 min at 4,000 × g and 4°C. The supernatant was discarded and the precipitate was resuspended with sterile normal saline (NS) to the original volume. This process was repeated twice to remove medium complex components and the precipitate was resuspended with sterile NS to the original volume following the third centrifugation.

Salmonella (strain no. ATCC14028; American Type Culture Collection, Manassas, VA, USA) and Escherichia coli (E. coli; strain no. CMCC-44101; China Medical Culture Collection, Beijing, China) were also subjected to the same procedure as O157:H7; they were tested with the same methodology to estimate the specificity of O157:H7.

Preparation of ‘signal into components’

‘Signal into components’ is a chromatophore with the pH sensitive fluorescent probe, F1300.

Preparation of chromatophores

Thermomicrobium-roseumwa0073 (ATCC27502) were purchased from the American Type Culture Collection and incubated at 60°C for 24 h. The cells were harvested by centrifugation at 4,000 × g for 30 min at 4°C and resuspended in buffer (20 mM Tris-HCl, 100 mM NaCl, 1 mM DTT, 0.1 mM PMSF and 2 mM MgCl2, pH 8.0) followed by ultrasonication for 10 min in an ice bath. The suspension was centrifuged at 10,000 × g for 30 min at 4°C to remove unbroken cells and cell fractions. The cell-free supernatant was centrifuged at 40,000 × g for 1 h at 4°C to collect membrane vesicles, termed chromatophores. The chromatophores were stored in 50% glycerol at −80°C. The chromatophore structure is presented in Fig. 1.

Figure 1. Schematic view of F1300 chromatophores. The arrow indicates the F0F1-ATPase carrier. The chromatophore provides H+ for ATP-synthesis. The chromatophore and the F0F1-ATPase are the organic combination. Cn, subunit c; a, subunit a; b, subunit b.

Labeling of chromatophores with F1300

The chromatophores were centrifuged at 12,000 × g at 4°C for 30 min to remove glycerol, then they were resuspended with buffer (pH 6.0, 0.1 mM tricine, 5 mM MgCl2 and 5 mM KCl). A total of 1–2 µl F1300 (1 mg/ml) was added to 600 µl chromatophores prior to ultrasonication for 3 min in an ice bath, to incorporate the probe into the inner part of the chromatophores. The free F1300 fraction was purified by centrifugation at 12,000 × g for 30 min at 4°C. The purification process was repeated three times to remove free F1300, and aliquots of the supernatant were collected to assess the level of purification. The precipitate was resuspended in tricine-NaOH buffer (0.1 mM tricine, 5 mM MgCl2 and 5 mM KCl, pH 6.5) and then the relative fluorescence signal was detected using the Varioskan Flash spectral scanning multimode reader (excitation, 485 nm; emission, 538 nm; Thermo Fisher Scientific, Inc., Waltham, MA, USA). When the relative fluorescence signal did not decrease further, the free F1300 was removed completely. The ATP hydrolysis activity of the labeled chromatophores was assayed using the enzyme coupling method with pyruvate kinase and lactate dehydrogenase, as described previously (4). The unit of enzyme activity was defined by hydrolyzing 1 µmol ATP per minute with 1 mg chromatophores (Fig. 2).

Figure 2. Schematic view of the F1300-labelled inner chromatophores (green dots). Cn, subunit c; a, subunit a; b, subunit b.

Constructing the O157:H7 detector

A ATPase β-subunit antibody [Homemade, as previously described (10)], biotin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and an antibody targeted against O157:H7 (cat. no. AB81131; Abcam, Cambridge, UK) were used for the following: 2 µl of 10 mM biotin was added to 500 µl of 3 mg/ml β-subunit antibody for 15–30 min at room temperature, then free biotin was removed via dialysis to produce the β-subunit antibody-biotin complex. In addition, 2 µl of 10 mM biotin was added to 500 µl of 3 mg/ml O157:H7 antibody for 15–30 min at room temperature, then free biotin was removed via dialysis to produce the O157:H7 antibody-biotin complex. To create the capture system complex, a reaction was set up that contained equal amounts ofβ-subunit antibody-biotin complex (200 µl, 50 mM) and O157:H7 antibody-biotin complex (200 µl, 50 mM). Subsequently, 200 µl 55 mM streptavidin (Sigma-Aldrich; Merck KGaA) was added and incubated for 15–30 min at room temperature. This produced the capture system complex: the β-subunit antibody-biotin-Streptavidin-biotin-O157:H7 antibody (Fig. 3). The capture system complexes and chromatophores labeled with F1300 (F1300-ch) were then mixed to a 4:1 dilution, and were incubated at 37°C for 1 h. This produced the biosensor used in the present study (Fig. 4). Three different reactions were set up, Group A, Group Band Group C, to demonstrate that the construction of the immuno-biosensor was successful, based on the fluorescence of F1300-ch with different loads: Group A, F1300-ch control; Group B, F1300-ch-β-subunit antibody-biotin-Streptavidin complex; Group C, F1300-ch-β-subunit antibody-biotin-Streptavidin-biotin-O157:H7 antibody complex.

Figure 3. Schematic view of the process applied to establish the capture system complex.

Figure 4. Schematic view of the biosensor constructed based on F0F1-ATPase. 1, detection carrier; 2, chromatophore complex; 3, ATPase; F1300-ch; 5, β-subunit antibody; 6, biotin; 7, streptavidin; 8, O157:H7 antibody; Cn, subunit c; a, subunit a; b, subunit b.

Fluorescence assay

The concentration of the bacterial suspension was adjusted to 5×103, 5×104 and 5×105 cfu/ml. In order to generate 102, 103 and 104 cfu bacteria/well, respectively, four groups were set up: Group 1, control (sterile NS); Group 2, 102 cfu bacteria/well; Group 3, 103 cfu bacteria/well; Group 4, 104 cfu bacteria/well. To each well, 50 µl biosensor and 20 µl of the bacterial suspension were added. Following incubation for 30 min at 37°C, 70 µl ATP synthesis buffer (50 mmol/l tricine, 10% glycerol, 2 mmol/ADP, 5 mmol/l NaH2PO4 and 5 mmol/l MgCl2, pH 8.0) was added to each well for further incubation at 45°C for 15 min. The relative fluorescence signal was detected using the Varioskan Flash spectral scanning multimode reader (excitation, 485 nm; emission, 538 nm; Thermo Fisher Scientific, Inc.).

Specificity

ATCC14028 Salmonella and CMCC44101 E. coli were subjected to the same protocols in order to determine the specificity to O157:H7. The bacterial concentration for each group and pathogenic bacteria are presented in Table I.

Table I. Bacterial concentration in assay wells to estimate the specificity of the O157:H7 biosensor.

Table I.

Bacterial concentration in assay wells to estimate the specificity of the O157:H7 biosensor.

	Bacterial concentration (cfu/well)
Strain of pathogenic bacteria	Control	Group 2	Group 3	Group 4	Group 5
E. coli O157:H7	0	27	133	265	1,325
E. coli CMCC44101	0	35	175	350	1,750
Salmonella ATCC14028	0	26	128	255	1,275

[i] E. coli, Escherichia coli; cfu, colony forming units; Control, contained sterile normal saline and 0 cfu/well O157:H7.

Statistical analysis

Data are presented as the mean ± standard deviation. Statistical analysis was performed using SPSS 10 software (SPSS, Inc., Chicago, IL, USA). The correlation was assessed by linear regression and a Dunnett's T3 post hoc test was used for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference.

Results

Construction of the immuno-biosensor

The inner chromatophores were successfully labeled with the fluorescent pH indicator F1300 as a unidirectional label (Fig. 5; Table II). The fluorescence intensity of the control chromatophores was the lowest. The fluorescence intensity of the mixture of chromatophores and F1300 was much higher than the control; however, it decreased with ultrasonication and the four subsequent centrifugation steps. Initially, a part of F1300 entered the chromatophore with ultrasonication; however, the remaining free F1300 stayed out of the chromatophore, producing high fluorescence intensity. Following three centrifugation procedures, the fluorescence intensity decreased to 1.49. This indicated that the free F1300 were removed by centrifugation. The fluorescence intensity did not change following the fourth round of centrifugation; it was 2.8 times higher than that of the control. The results revealed that free F1300 was completely removed, while the fluorescent probe F1300 labeled the inner chromatophores. The synthetic activity of F1300-chisshown in Fig. 6; the enzyme activity was 106.4 µmol/mg/min.

Figure 5. Relative fluorescence value of the F1300-labeled inner chromatophores. Activity of F1300-ch in the control, fluorescence labelled and the 4 E. coli centrifugation groups. Following the incorporation of F1300 into the chromatophore the mix was centrifuged 4 times to produce the 4 E. coli centrifugation groups. This was performed to verify that the inner chromatophores were labeled with F1300 successfully.

Figure 6. Synthetic activity of the F1300-labelled chromatophores (F1300-ch).

Table II. Results of F1300 labelling in the inner chromatophores.

Table II.

Results of F1300 labelling in the inner chromatophores.

Group	Relative fluorescence value
Chromatophores	0.519
Fluorescence labeling	18.12
1st centrifugation	6.69
2nd centrifugation	2.82
3rd centrifugation	1.49
4th centrifugation	1.48

As presented in Fig. 7, the fluorescence intensity of group A was the highest, while group C was the lowest. This indicated that the capture system complex (β-subunit antibody-biotin-Streptavidin-biotin-substrate antibody) was successfully established. In addition, it revealed that the greater the load on F0F1-ATPase, the lower the enzyme activity and thus, the lower the relative fluorescence. Therefore, the F1300-labeled chromatophores were used in the present study.

Figure 7. Effects of the biosensor based on F0F1-ATPase for O157:H7. This experiment was performed to verify that the F0F1-ATPase-based biosensor for the detection of O157:H7 was constructed successfully. Group A, F1300-ch; Group B, F1300-ch-β-subunit antibody-biotin-Streptavidin complex; Group C, F1300-ch-β-subunit antibody-biotin-Streptavidin-biotin-O157:H7 antibody complex.

Sensitivity for O157:H7

The fluorescence value gradually decreased with the increasing concentration of O157:H7 (Fig. 8; Table III). Following statistical analysis, the results revealed that there were significant differences between the control and 104 cfu group (P<0.01), and between the 102 and 104 cfu groups (P<0.01). The results of O157:H7 detection using this method identified a strong positive gradient between 102−104 cfu (R2=0.9818).

Figure 8. Sensitivity for O157:H7 in the 0 (control), 102, 103 and 104 cfu/well groups. R2=0.9787, P=0.009163. **P<0.01 vs. 0 cfu/well (control).

Table III. Results of the comparison between the concentration groups.

Table III.

Results of the comparison between the concentration groups.

Concentration group (cfu/well)		P-value
0	102	0.19239
0	103	−0.06347
0	104	0.00327a
102	103	0.63584
102	104	0.00015b
103	104	0.60539

{ label (or @symbol) needed for fn[@id='tfn2-mmr-17-01-0870'] } Comparisons were performed using Dunnett's T3 test.

a P<0.01, 0 vs. 10⁴

b P<0.01, 10² vs. 10⁴ cfu, colony forming units.

Specificity for O157:H7

The curve of 101−103 cfu identified a good separation, which is consistent with the positive threshold value for O157:H7, 0.063–0.075. The relative fluorescence value of the CMCC44101 E. coli groups was 0.035–0.043 and 0.035–0.052 for Salmonella. These results demonstrated that this biosensor has specificity for O157:H7 (Fig. 9; Table IV).

Figure 9. Specificity for O157:H7 in Groups 1 to 5, comparing O157:H7, Salmonella and E. coli. **P<0.01 vs. O157:H7. E. coli, Escherichia coli; Group 1, control (sterile normal saline); Group 2, 27, 35 and 26 cfu/well for O157:H7, E. coli and Salmonella, respectively; Group 3,133, 175 and 128 cfu/well for O157:H7, E. coli and Salmonella, respectively; Group 4, 265, 350 and 255 cfu/well for O157:H7, E. coli and Salmonella, respectively; Group 5, 1,325, 1,750 and 1,275 cfu/well for O157:H7, E. coli and Salmonella, respectively.

Table IV. Results of the comparison between the O157:H7, E. coli and salmonella groups.

Table IV.

Results of the comparison between the O157:H7, E. coli and salmonella groups.

Group comparison	P-value
O157:H7 vs. E. coli
Group1	0.69
Group 2	2.81×10−5a
Group 3	1.03×10−4a
Group 4	1.89×10−7a
Group 5	9.82×10−7a
O157:H7 vs. Salmonella
Group 1	0.36
Group 2	2.15×10−5b
Group 3	2.00×10−4b
Group 4	1.16×10−6b
Group 5	4.92×10−5b

a P<0.01, vs. E. coli

b P<0.01, vs. salmonella.

Time of detection

This method is short and includes only four steps: bacterium solution treatment, preparation of biosensors, loading and testing. The time required for each step was 2, 1.75, 0.5 and 0.25 h, respectively, thus 4.5 h in total. Though this method requires separation and enrichment of target bacteria when testing samples, the limit of detection was 100 cfu. In addition, by combining it with the immune magnetic separation technique, the time required for sample pretreatment was 8.5 h and the total time for testing samples was <16 h.

Discussion

F0F1-ATPase has the following two characteristics: i) it can use the H+ gradient between the inside and outside of chromatophores to produce ATP and it can also hydrolyze ATP by reverse transporting H+. During ATP synthesis, protons are pumped to the outside from the inside of chromatophores, which leads to a change in proton concentration inside the chromatophores; ii) F0F1-ATPase rotation speed and the loads on its subunits are positively correlated. Based on its two enzymology characteristics, the F0F1-ATPase nano-biosensor is labeled by pH sensitive F1300, a fluorospectrophotometric probe, to produce the functional unit, F1300-ch. The F1300-ch combined with the capture system achieves the molecular motor nano-biosensor, which can be used as a fast detection technique (7,11–13). Liu et al (8) directly observed the mechanically driven proton influx or efflux in F0 coupled with rotation of F1 at a single molecular level; the specific underlying mechanism will be studied further in the future.

F0F1-ATPase activity is regulated by external links on b subunits with different molecular weights. It is inhibited when anti-b subunit antibodies, streptavidin and H9 antibodies link on to the β subunits successively (7). The holoenzyme activity was inhibited as it linked to more external substances, including Chloramphenicol, Listeria monocytogenes, H9 virus, Clenbuterol and Deoxynivalenol (4–5,7,9,10). When the O157:H7 loads into the chromatophore, the chromatophore cannot move completely and there are no alterations in protons between the internal and external chromatophore, therefore the alteration in relative fluorescence intensity should be generated by those non-O157:H7-loaded chromatophores in each detection well. In another way, this is similar to the competition method in ELISA; the stronger the concentration, the lower the changing biosensors and so the smaller the change value (14,15).

The application of F0F1-ATPase immuno-biosensors for the detection of O157:H7 has not been reported previously. The present study used biosensors to detect O157:H7, demonstrating that this method is rapid, sensitive, simple and has a low cost.

When compared with other novel detection methods, this method is faster, more sensitive and easier to operate. In addition, the present study investigated its specificity, as well as the feasibility of this method using standard strains. The results demonstrated that it has a good specificity to E. coli. Table V presents a comparison of the results between the current different methods. At present, the sophisticated testing methods of pathogenic microorganisms include the microbial method, PCR and ELISA. The microbial method is time-consuming and involves complicated processes. PCR is the most mature method in the national testing methods of pathogenic microorganisms, however, its reagents are expensive and the procedure is complex. In addition, as this method is extremely sensitive, the experimental conditions, the exogenous DNA, improper controls, primer design and the target selection of sequence will all affect the results. The ELISA method is relatively time-efficient; however, it requires skilled operation and a detection limit of 106 cfu/l. Surface plasmon resonance, biosensors, capillary zone electrophoresis and other technologies are also being studied by the researchers, as they are quicker than the microbial method and simpler than PCR; however, they require expensive instruments and the low detection limit is 105−106 cfu/ml (16–23).

Table V. A comparison of the current methods used for detection.

Table V.

A comparison of the current methods used for detection.

Method	Time	Detection limit	Specificity	Cost
F0F1-ATPase immuno-biosensor	13 h	1–5 cfu/25 g	95%	Very low
Microbial method	5–7 d	1 cfu/25 g	100%	Low
PCR	>48 h	102−7×103 cfu/ml	99% (DNA easily polluted)	High
ELISA	36–48 h	106 cfu/ml	95%	Low
Reverse transcription PCR	48 h	1–5 cfu/25 g	99% (DNA easily polluted)	High
SPR biosensor	>24 h	2×106 cfu/ml	95%	High
CEZ	>24 h	105−106 cfu/ml	95%	Low

[i] PCR, polymerase chain reaction; SPR, Surface plasmon resonance; CEZ, capillary zone electrophoresis; cfu, colony forming units.

The present study performed preliminary research on the feasibility of applying F0F1-ATPase immuno-biosensors for O157:H7 detection. The detection techniques based on F0F1-ATPase can rapidly detect the disease markers in patient serum or feces, which will aid rapid clinical diagnosis. To promote its application, novel fluorescent material could be chosen as fluorescent probe to improve the sensitivity and accuracy of this method (21–25). In addition, high specificity immune magnetic beads could be used to enrich the target bacterial, which can minimize the interference of other bacterial (26–28). Due to the complexity of the sample and the variety of bacteria in different samples, the application of a biosensor for the detection of pathogens is rarely reported. Furthermore, the application of biosensors in pathogenic bacteria detection is rarely reported; therefore, sample pretreatment, and increasing bacteria and bacteria solution treatment, will require extensive research in practical sample testing.

Acknowledgements

The authors would like to thank Professor Jia-Chang Yue (Institute of Biophysics, Chinese Academy of Sciences, Beijing, China) for his technical assistance, as well as Professor Yan-Qun Li (South of China Normal University, Guangdong, China) for the revisions and modifications made to this paper. The present study was supported by the Ministry of Science and Technology of the People's Republic of China (grant no. 2008IM021600), the Beijing Academy of Science and Technology of China (grant no. IG200905N), the Beijing Municipal Party Organization of China (grant no. 2010D002022000009) and the National Key Foundation for Exploring Scientific Instrument (grant no. 2013YQ140405).

References