Abstract
Surface plasmon resonance (SPR) biosensors were used to develop a rapid and simultaneous detection method for three important foodborne pathogens, i.e., Escherichia coli O157:H7 (O157), Salmonella Enteritidis (SE), and Listeria monocytogenes (LM). Bacterial homogenates prepared by sonication from bacterial suspensions at various cell concentrations were analyzed using SPR biosensors and sensor chips with polyclonal antibodies specific to each of the target pathogens. The precipitates from the homogenates were demonstrated to be suitable for specific and simultaneous detection of O157, SE, and LM. By using the precipitates and a custom-built multichannel SPR biosensor, the lower detection limit for O157, SE, and LM was determined to be 0.6 × 106, 1.8 × 106, and 0.7 × 107 CFU/mL, respectively, in the presence of non-target pathogens at concentrations of 105 to 108 CFU/mL.
Introduction
Food safety is a global health goal and food poisoning is a major health risk (Velusamy et al., 2010). There are approximately 76 million instances of food poisoning, accounting for 324,000 hospitalizations, 5200 deaths and a $23 billion financial loss annually in the United States (Mead et al., 2000). Although the safety of food has dramatically improved overall in the past several years, food poisoning still occurs. According to the Statistics of Food Poisoning published by the Food Safety Division, the Ministry of Health, Labor and Welfare of Japan, 1100 incidents of food poisoning occurred in Japan in 2012, causing 26,699 hospitalizations and 11 deaths (i). Foodborne pathogens are a major cause of food poisoning, thus the rapid, sensitive and reliable detection of foodborne pathogens is critical for the effective prevention of outbreaks of food poisoning.
At present, culture-based conventional methods are commonly used in the food industry to detect foodborne pathogens. The conventional methods are considered to be the “gold-standard” and are well known for their cost effectiveness, sensitivity, ability to confirm cell viability, and ease of standardization (Dwivedi and Jaykus, 2011). However, because they have complicated procedures, they are time consuming, taking at least 2 to 3 days for positive presumptive results and 7 to 10 days for confirmation, and are also labor intensive (Bai et al., 2010). The rapid growth of the food industry means that these methods can no longer meet the demands of the industry. Thus, the need for rapid and simple detection methods for foodborne pathogens is growing. Several new methods, such as enzyme-linked immunosorbent assay (ELISA) (Park et al., 2012; Zhong et al., 2012) and polymerase chain reaction (PCR) (Liu et al., 2012; Wang and Mustapha, 2010), have been developed for the rapid detection of foodborne pathogens. While PCR and ELISA have improved detection time, biosensors can perform faster detection, maintaining high sensitivity and specificity (Taylor et al., 2006).
In recent years, there has been much research activity in the area of developing biosensors for detecting foodborne pathogens (Velusamy et al., 2010). Surface plasmon resonance (SPR)-based optical biosensors, which allow for real-time and label-free detection, are of great utility. SPR biosensor can detect minute changes in the refractive index, which occur when an analyte binds to the ligand immobilized on the transducer surface. The sensor measures the change in the angle of the reflected light due to the change in the density of the medium with time (Fig. 1) (Rich and Myszka, 2001). SPR biosensors have been used by many research groups for the detection of foodborne pathogens (Meeusen et al., 2005; Oh et al., 2004; Subramanian et al., 2006; Tawil et al., 2012; Waswa et al., 2007; Wei et al., 2007). However, all these studies detected only one species of target pathogen in one detection period. In this study, we used a multichannel SPR biosensor to develop a method for simultaneously detecting three important foodborne pathogens, i.e., Escherichia coli O157:H7, Salmonella Enteritidis, and Listeria monocytogenes.
Materials and Methods
Bacteria and cultivation Escherichia coli O157:H7 (O157) and Listeria monocytogenes LIS 21 (LM) were obtained from the Fukuoka City Institute for Hygiene and the Environment, Fukuoka, Japan. Salmonella Enteritidis (SE) IFO 3313 was purchased from the Institute for Fermentation, Osaka, Japan. Each of the three bacterial strains was cultured in 5 mL tryptic soy broth (TSB; Becton, Dickinson and Company, Sparks, MD, USA) at 37°C for 18 h with shaking at 130 rpm. Cultures were used for the following sample preparations.
Sample preparation Cultures of O157, SE, and LM were centrifuged at 5,800 × g for 5 min. The pellets were resuspended in HBS-EP buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% (v/v) surfactant P20), centrifuged at 5,800 × g for 5 min, and the pellets were resuspended again in HBS-EP buffer. The bacterial suspensions in an ice bath were sonicated until the bacteria were disrupted thoroughly by using a TOMY Ultrasonic Disruptor UD-201 (TOMY SEIKO CO., LTD, Tokyo, Japan) at an output of 50 W. The whole homogenates were directly used as whole sonicated samples. After centrifuging 1 mL of the homogenates at 15,000 × g and 4°C for 10 min, 850 µL of the supernatant was withdrawn and mixed with 150 µL of HBS-EP buffer. The mixtures were used as the supernatants of the sonicated samples. HBS-EP buffer (850 µL) was added to and mixed with the remaining precipitate and supernatant (150 µL). These mixtures were used as the precipitates of the sonicated samples. These samples were serially diluted and mixed to produce samples for SPR analysis and stored at −20°C until use.
Previous studies showed that the lower detection limit for bacteria using an SPR biosensor is not only dependent on the sensitivity of the instrument and the specificity and affinity of the surface chemistry, but also on the sample preparation method (Taylor et al., 2005; Zhang et al., 2013). Compared with whole bacterial cells, small pieces of cells are easier to detect using an SPR biosensor (Bhunia et al., 2004). Each bacterium contains many antigens. Breaking a bacterium into small pieces will not only increase the number of analytes available for detection, but also improve the diffusion of the analytes. Both factors will affect the ability of the analytes to bind to the ligand immobilized on the sensing surface (Taylor et al., 2005). Sonication, which appears to weaken microbial membranes through cavitation induced by ultrasonic shock waves (Wong et al., 2012), is often used to break cells into small pieces and release cellular contents. Sonication has been proven to be an effective way to improve the lower detection limit for bacteria using an SPR biosensor (Zhang et al., 2013). Thus, the whole sonicated samples were used here for bacterial detection. However, for the specific detection of bacteria, the cross-reactivity of antibody against non-target bacteria should be minimized. The whole sonicated sample was thus separated into supernatant and precipitate by centrifugation and used for SPR detection to minimize the cross-reactivity of antibodies against non-target bacteria.
Viable counts Viable bacterial counts were determined by plating 100 µL of 10-fold serial dilutions of the bacterial suspensions in phosphate buffered saline (PBS; 0.1 M phosphate buffer containing 0.15 M NaCl, pH 7.2) on tryptic soy agar (TSA; Becton, Dickinson and Company) plates. After cultivation at 37°C for 24 h, formed colonies were counted.
SPR instrumentation Two SPR biosensors were used for the detection of foodborne pathogens. One was a Biacore J (GE Healthcare Bio-Sciences AB, Uppsala, Sweden), and the other one was a custom-built multichannel SPR biosensor (Kyushu Keisokki Co., Ltd., Fukuoka, Japan). Biacore J was equipped with two flow channels: flow channel 1 for detection and flow channel 2 for reference. The multichannel SPR biosensor was equipped with five flow channels. Two flow channels served as reference channels and the other three flow channels were used for detection.
Surface modification of sensor chips The gold surfaces of Sensor Chips Au (GE Healthcare Bio-Sciences AB and Kyushu Keisokki Co., Ltd.) were cleaned by washing in acetone for 10 min, ethanol for 2 min, and 2-propanol for 2 min with sonication, washing with deionized water, immersing in a mixed solution (deionized water: 28% ammonia solution: 30% H2O2 = 5:1:1) for 30 min at 90°C, and washing again with deionized water. The surfaces were subsequently modified with the formation of a mixed self-assembled monolayer (SAM) by immersing in ethanol with 0.1 mM carboxy-EG6-undecanethiol (PEG6; DOJINDO Laboratories, Kamimashiki, Kumamoto, Japan) and 0.9 mM hydroxy-EG3-undecanethiol (PEG3; DOJINDO Laboratories) for 24 h at room temperature (Fig. 2). The carboxyl group of PEG6 was used for subsequent immobilization of antibody, while PEG3 was used to create a non-fouling background. The 1:9 ratio of PEG6:PEG3 maximized antibody binding while minimizing non-specific binding, based on our preliminary study. After the formation of the mixed SAM, the chips were washed in ethanol for 5 min with sonication and then washed with deionized water.
Antibody immobilization on sensor chips Polyclonal goat antibody (PAb) for O157, Salmonella and Listeria were purchased from Kirkegaard & Perry Laboratories, Inc. (KPL, Inc., Gaithersburg, Maryland, USA). Antibodies were immobilized onto the surface of sensor chips using the amine-coupling method (Löfås and Johnsson, 1990). The surfaces of the flow channels were activated for 10 min with a 1:1 mixture of 0.1 M N-hydroxysuccinimide (NHS) and 0.4 M 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) at medium flow rate (approximately 30 µL/min). In the case of Biacore J, antibody was immobilized on flow channel 1 at a concentration of 100 µg/mL in 10 mM acetate buffer with the corresponding pH (pH 4.0 for anti-Salmonella PAb, pH 5.0 for anti-O157 PAb and anti-Listeria PAb) at medium flow rate (approximately 30 µL/min) for 12 min. Flow channel 2 was left blank to serve as a reference channel. In the case of the multichannel SPR biosensor, anti-O157 PAb, anti-Salmonella PAb and anti-Listeria PAb were immobilized on flow channel 1 (CH1), flow channel 3 (CH3) and flow channel 5 (CH5), respectively. Flow channel 2 (CH2) and flow channel 4 (CH4) were left blank to serve as the reference channels (Fig. 3). After antibody immobilization, the surfaces of all flow channels were blocked with 1.0 M ethanolamine-HCl (pH 8.5) for 10 min and 1% bovine serum albumin in HBS-EP buffer for 6 min at medium flow rate (approximately 30 µL/min) to minimize non-specific adsorption (Rich and Myszka, 2001).
SPR detection of samples Sensor chips with antibodies were docked into Biacore J and the multichannel SPR biosensor. HBS-EP buffer was used as a running buffer and run at medium flow rate (approximately 30 µL/min) and 25°C. HBS-EP buffer was first injected for 5 min to establish a baseline. Then, the corresponding sample was injected for 5 min for antigen-antibody binding. The signal change for the sample was obtained by subtracting the signal recorded at 30 s before the start of the sample injection from the signal recorded at 60 s after the end of the sample injection. After measurement of each sample, regeneration solution (10 mM glycine-HCl, pH 1.7) was injected for 1 min to dissociate antigen-antibody binding. After regeneration, sensor chip surfaces were equilibrated with running buffer (HBS-EP buffer).
The unit of signal change is the resonance unit (RU). For instance, 1,000 RU represents a mass change of 1 ng per mm2 on the surface of the sensor chip and also represents a resonance angle change of 0.1 degree. To compensate the detection channel for variations in sample composition and non-specific adsorption, the signal of the reference channel was subtracted from that of the detection channel. This is important for determining the accurate lower limit of detection (Taylor et al., 2005). Moreover, the accuracy of detection results could be improved by setting the baseline. For this purpose, the signal of HBS-EP buffer was subtracted from the signals of samples. This compensation removed the sensor responses of baseline drift and measurement noise of the instrument. This is particularly important when the value of the signal change derived from the binding of target antigen in the sample is small or the capture method is used for SPR analysis.
Statistical analysis Triplicate experiments were run for every Biacore J experiment and standard deviation (SD) was calculated for the results. Student's t-test (one-tailed distribution, paired type) was performed to determine significant (p < 0.05 and p < 0.01) differences between means. All statistical analyses were performed using Microsoft Excel for Mac 2011 (V 14.3.1, Microsoft Corporation, Redmond, WA, USA).
Results and Discussion
Detection of pathogens using whole sonicated samples and Biacore J The whole sonicated samples were applied to Biacore J equipped with sensor chip functionalized with the corresponding PAb. Fig. 4 shows the signal changes determined on the whole sonicated samples. Two kinds of samples, samples containing the target pathogen only and samples containing both target and non-target pathogens, were used for detection. Compared with the signal changes of samples containing target pathogen only, samples containing the same level of target pathogen in the presence of non-target pathogens produced higher positive signal changes. In the case of the detection of O157 using anti-O157 PAb (Fig. 4a), the signal change (84 RU) of the sample containing only O157 at 1.8 × 106 CFU/mL was not significantly different from the signal change (100 RU) of the sample containing O157 at the same concentration, and same level of SE and LM, at 1.2 × 106 and 1.8 × 106 CFU/mL, respectively. However, in the presence of SE and LM at 1.2 × 108 and 1.8 × 108 CFU/mL, respectively, the signal change (148 RU) was significantly different from that (84 RU) of O157 alone at 1.8 × 106 CFU/mL. The same results were obtained on the whole sonicated samples of O157 at 1.8 × 105 CFU/mL. For detection of SE by anti-Salmonella PAb, the presence of O157 and LM at a cell concentration of about 100-fold of SE significantly affected the detection of SE (Fig. 4b). For the detection of LM, the presence of O157 and SE at a cell concentration of about 10-fold of LM significantly affected the detection of LM (Fig. 4c). The significant differences between the signal changes of samples containing target pathogen in the presence and absence of non-target pathogens seem to be related to cross-reactivity of antibody against non-target pathogens.
Cross-reactivity test on supernatants and precipitates of sonicated samples using Biacore J Fig. 5 shows the signal changes obtained when the supernatants and precipitates of sonicated samples containing non-target pathogens at various cell concentrations were run across the sensor chip surfaces functionalized with the corresponding PAbs. No significant positive signal changes were obtained with both the supernatants and precipitates of sonicated samples containing non-target pathogens at low cell concentrations (below 106 CFU/mL) (Fig. 5a, b, and c). In contrast, at a high cell concentration of 108 CFU/mL, both the supernatants and precipitates of sonicated samples containing non-target pathogens produced positive signal changes of less than 25 RU for anti-O157 PAb and anti-Salmonella PAb (Fig. 5a and b). For anti-Listeria PAb, the supernatant of the sonicated sample containing non-target pathogens produced a positive signal change of 7 RU while the precipitate of the sonicated sample produced no signal change (Fig. 5c). Compared to the signal changes of the supernatants of sonicated samples, the signal changes of the precipitates of sonicated samples were much lower in the case of anti-O157 PAb and anti-Listeria PAb (Fig. 5a and c). For anti-Salmonella PAb, the signal change of the precipitate of the sonicated sample was slightly higher than that of the supernatant of the sonicated sample (Fig. 5b). Hence, for specific and simultaneous detection of the three pathogens using an SPR biosensor, the precipitate of the sonicated sample seems more suitable than the supernatant of the sonicated sample as it minimized the overall cross-reactivity of PAbs against non-target pathogens.
Cross-reactivity arises because the non-target pathogen shares an epitope in common with the target pathogen or because it has an epitope that is structurally similar to that of the target pathogen (ii). For anti-O157 PAb and anti-Listeria PAb, the concentration of shared epitope or similar epitope of the non-target pathogens in the supernatant of the sonicated sample appears to be higher than that in the precipitate of the sonicated sample. On the other hand, the situation is opposite for anti-Salmonella PAb.
Specific detection of each pathogen using the precipitates of sonicated samples and SPR biosensors (Biacore J and multichannel SPR biosensor) Fig. 6 shows the signal changes for target pathogens at various cell concentrations (105, 106, and 107 CFU/mL) and control samples containing the other two non-target pathogens at 108 CFU/mL. These results were obtained by using the precipitates of sonicated samples and Biacore J. O157 at 1.8 × 106 CFU/mL, SE at 0.9 × 107 CFU/mL and LM at 1.8 × 107 CFU/mL produced positive signal changes of 39, 65, and 22 RU, respectively, and were significantly higher than those of the control samples. It has been reported that the cell concentration of O157, SE, and LM was less than 108 CFU/mL after the simultaneous enrichment of six foodborne pathogens, including these three pathogens (Kobayashi et al., 2009). Hence, the lower limit of detection using an SPR biosensor with specific PAbs for O157, SE, and LM was determined to be 1.8 × 106, 0.9 × 107, and 1.8 × 107 CFU/mL, respectively.
The above results demonstrate that the precipitate of the sonicated sample is more suitable for the specific and simultaneous detection of pathogens than the supernatant of the sonicated sample in cross-reactivity test using Biacore J and specific PAbs. Similar results were obtained in cross-reactivity tests using the multichannel SPR biosensor and specific PAbs (data not shown). Hence, the precipitates of the sonicated samples were applied to the multichannel SPR biosensor for pathogens detection.
Fig. 7 shows the signal changes obtained by using the multichannel SPR biosensor for the precipitates of sonicated samples prepared from each pathogen at about 105 to 108 CFU/mL. For CH1 (anti-O157 PAb), non-target pathogens of SE and LM produced positive signal changes ranging from 4 to 25 RU and 4 to 28 RU, respectively, which indicated that the cut-off value for the detection of O157 with CH1 should be 25 + 28 = 53 RU. Similarly, the cut-off value for the detection of SE with CH3 is 42 RU, and for the detection of LM with CH5 is 13 RU. The cut-off value was used as a reference when the lower detection limit for each species of pathogen was calculated. The lower detection limit was calculated from the data shown in Fig. 7 by regression analysis based on three times the standard deviation of the baseline noise (Taylor et al., 2006; Thomsen et al., 2003), which is 4 RU of signal change for the multichannel SPR biosensor. The calculated lower detection limit for O157, SE and LM is 1.1 × 106, 1.9 × 106, and 1.3 × 107 CFU/mL, respectively. Signal changes obtained for target pathogens at various cell concentrations using the multichannel SPR biosensor were similar to those obtained by using Biacore J. This indicated that the custom-built multichannel SPR biosensor used in this study has the same detection sensitivity as that of Biacore J.
Many researchers have applied SPR biosensors for the detection of O157 (Meeusen et al., 2005; Subramanian et al., 2006; Taylor et al., 2005), SE (Bhunia et al., 2004; Bokken et al., 2003), and LM (Hearty et al., 2006; Leonard et al., 2004). The lower detection limit was reported to be 105 to 107 CFU/mL for O157, 106 to 107 CFU/mL for SE, and 105 to 107 CFU/mL for LM, respectively. The different lower detection limits from these studies may be attributed to the use of different SPR biosensors, surface chemistries, antibodies, assay types (direct, sandwich, and subtractive inhibition assay), sample preparation methods, antibody immobilization methods, and linker molecules.
Simultaneous and specific detection of pathogens using the precipitates of sonicated samples of a mixture of three pathogens and the multichannel SPR biosensor Successful biosensor detection must discriminate a target analyte from coexisting complex constituents in the sample (Taylor et al., 2006). The multichannel SPR biosensor used in this work could discriminate multiple pathogens simultaneously from a complex bacterial mixture in the sample. Samples containing the three pathogens at cell concentrations of 105, 106, and 107 CFU/mL were applied to a multichannel SPR biosensor equipped with sensor chip functionalized with PAbs on corresponding flow channels. Fig 8 shows the corresponding signal changes of samples for CH1 (anti-O157 PAb), CH3 (anti-Salmonella PAb) and CH5 (anti-Listeria PAb). Cell concentration-dependent increases in signal were observed for all three channels. The lower detection limit for O157, SE and LM calculated from the data shown in Fig. 8 by regression analysis was 0.6 × 106, 1.8 × 106, and 0.7 × 107 CFU/mL, respectively.
It has been reported that an eight-channel SPR biosensor based on wavelength division multiplexing achieved a lower detection limit for O157 and LM of 1.4 × 104 and 3.5 × 103 CFU/mL, respectively (Taylor et al., 2006). The study adopted the sandwich assay method for bacterial detection, in which secondary antibody was used to amplify the signal change of detection. However, the use of secondary antibody increased the cost of detection significantly because of the high price of antibody. Moreover, the use of secondary antibody also increased the detection time up to 30 min for one sample. In contrast, the detection time for one sample was only 5 min in our study since we used the direct assay method without using secondary antibody.
Conclusion
In this work, SPR biosensors were used to develop a rapid and simultaneous detection method for three important foodborne pathogens, O157, SE, and LM. It was demonstrated that the precipitates of sonicated samples are more suitable for the simultaneous and specific detection of the three pathogens than the supernatants of sonicated samples. The three pathogens were detected simultaneously by using the precipitates of sonicated samples and a custom-built multichannel SPR biosensor with sensor chip functionalized with polyclonal antibodies specific to the target pathogens. The presence of non-target pathogens interfered with the detection of target pathogens due to cross-reactivity of the immobilized PAbs against non-target pathogens. This interference was eliminated by setting an appropriate cut-off value for each detection channel.
The three important foodborne pathogens were detected simultaneously and specifically from a sample containing pathogens in buffer. Detection limits and/or signal changes for pathogens in buffer may be different from that for pathogens in food samples, as the components of food may affect the detection of pathogens. Further work is required to develop a simultaneous detection method for multiple foodborne pathogens in food samples.
References
- Bai, S., Zhao, J., Zhang, Y., Huang, W., Xu, S., Chen, H., Fan, L., Chen, Y., and Deng, X.W. (2010). Rapid and reliable detection of 11 food-borne pathogens using thin-film biosensor chips. Appl. Microbiol. Biotechnol., 86, 983-990.
- Bhunia, A.K., Geng, T., Lathrop, A., Valadez, A., and Morgan, M.T. (2004). Optical immunosensors for detection of Listeria monocytogenes and Salmonella Enteritidis from food. Proc. of SPIE, Bellingham, WA, pp. 1-6.
- Bokken, G.C.A.M., Corbee, R.J., Van Knapen, F., and Bergwerff, A.A. (2003). Immunochemical detection of Salmonella group B, D and E using an optical surface plasmon resonance biosensor. FEMS Microbiol. Lett., 222, 75-82.
- Dwivedi, H.P. and Jaykus, L.A. (2011). Detection of pathogens in foods: The current state-of-the-art and future directions. Crit. Rev. Microbiol., 37, 40-63.
- Hearty, S., Leonard, P., Quinn, J., and O'Kennedy, R. (2006). Production, characterisation and potential application of a novel monoclonal antibody for rapid identification of virulent Listeria monocytogenes. J. Microbiol. Methods, 66, 294-312.
- Kobayashi, H., Kubota, J., Fujihara, K., Honjoh, K., Iio, M., Fujiki, N., Nakabe, M., Oda, S., Takasu, K., Nakanishi, H., and Miyamoto, T. (2009). Simultaneous enrichment of Salmonella spp, Escherichia coli O157:H7, Vibrio parahaemolyticus, Staphylococcus aureus, Bacillus cereus, and Listeria monocytogenes by single broth and screening of the pathogens by multiplex real-time PCR. Food Sci. Technol. Res., 15, 427-438.
- Leonard, P., Hearty, S., Quinn, J., and O'Kennedy, R. (2004). A generic approach for the detection of whole Listeria monocytogenes cells in contaminated samples using surface plasmon resonance. Biosens. Bioelectron., 19, 1331-1335.
- Liu, P., Mizue, H., Fujihara, K., Kobayashi, H., Kamikado, H., Tanaka, T., Honjoh, K., and Miyamoto, T. (2012). A new rapid real-time PCR method for detection of Listeria monocytogenes targeting the hlyA gene. Food Sci. Technol. Res., 18, 47-57.
- Löfås, S. and Johnsson, B. (1990). A novel hydrogel matrix on gold surfaces in surface plasmon resonance sensors for fast and efficient covalent immobilization of ligands. J. Chem. Soc., Chem. Commun., 21, 1526-1528.
- Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M., and Tauxe, R.V. (2000). Food-related illness and death in the United States. J. Environ. Health, 62, 9-18.
- Meeusen, C.A., Alocilja, E.C., and Osburn, W.N. (2005). Detection of E. coli O157:H7 using a miniaturized surface plasmon resonance biosensor. Trans. ASAE, 48, 2409-2416.
- Oh, B.K., Lee, W., Kim, Y.K., Lee, W.H., and Choi, J.W. (2004). Surface plasmon resonance immunosensor using self-assembled protein G for the detection of Salmonella paratyphi. J. Biotechnol., 111, 1-8.
- Park, S., Min, J., and Kim, Y.K. (2012). Chemiluminescent enzyme-linked immunosorbent assay on a strip to detect Escherichia coli O157:H7. Int. J. Environ. Anal. Chem., 92, 655-664.
- Rich, R.L. and Myszka, D.G. (2001). BIACORE J: A new platform for routine biomolecular interaction analysis. J. Mol. Recognit., 14, 223-228.
- Subramanian, A., Irudayaraj, J., and Ryan, T. (2006). A mixed self-assembled monolayer-based surface plasmon immunosensor for detection of E. coli O157:H7. Biosens. Bioelectron., 21, 998-1006.
- Tawil, N., Sacher, E., Mandeville, R., and Meunier, M. (2012). Surface plasmon resonance detection of E. coli and methicillin-resistant S. aureus using bacteriophages. Biosens. Bioelectron., 37, 24-29.
- Taylor, A.D., Ladd, J., Yu, Q., Chen, S., Homola, J., and Jiang, S. (2006). Quantitative and simultaneous detection of four foodborne bacterial pathogens with a multi-channel SPR sensor. Biosens. Bioelectron., 22, 752-758.
- Taylor, A.D., Yu, Q., Chen, S., Homola, J., and Jiang, S. (2005). Comparison of E. coli O157:H7 preparation methods used for detection with surface plasmon resonance sensor. Sens. Actuators, B, 107, 202-208.
- Thomsen, V., Schatzlein, D., and Mercuro, D. (2003). Limits of detection in spectroscopy. Spectroscopy (Santa Monica), 18, 112-114.
- Velusamy, V., Arshak, K., Korostynska, O., Oliwa, K., and Adley, C. (2010). An overview of foodborne pathogen detection: In the perspective of biosensors. Biotechnol. Adv., 28, 232-254.
- Wang, L. and Mustapha, A. (2010). Ema-real-time PCR as a reliable method for detection of viable Salmonella in chicken and eggs. J. Food Sci., 75, 134-139.
- Waswa, J., Irudayaraj, J., and DebRoy, C. (2007). Direct detection of E. Coli O157:H7 in selected food systems by a surface plasmon resonance biosensor. LWT--Food Sci. Technol., 40, 187-192.
- Wei, D., Oyarzabal, O.A., Huang, T.S., Balasubramanian, S., Sista, S., and Simonian, A.L. (2007). Development of a surface plasmon resonance biosensor for the identification of Campylobacter jejuni. J. Microbiol. Methods, 69, 78-85.
- Wong, E., Vaillant-Barka, F., and Chaves-Olarte, E. (2012). Synergistic effect of sonication and high osmotic pressure enhances membrane damage and viability loss of Salmonella in orange juice. Food Res. Int., 45, 1072-1079.
- Zhang, X., Tsuji, S., Kitaoka, H., Tamai, M., Kobayashi, H., Honjoh, K., and Miyamoto, T. (2013). Effects of pretreatments on detection of E. coli O157:H7 by SPR biosensor. J. Fac. Agric. Kyushu Univ., In press.
- Zhong, Q., Dong, Y., Wang, L., and He, B. (2012). Development of IgY-based sandwich enzyme-linked immunosorbent assay for the detection of Vibrio parahaemolyticus. Advances in Intelligent and Soft Computing, 134, 517-524.
- i) http://www.mhlw.go.jp/topics/syokuchu/04.html (Jul. 19, 2013)
- ii) http://pathmicro.med.sc.edu/mayer/ab-ag-rx.htm (Jul. 17, 2013)
) and precipitates (□) were prepared from the whole sonicated samples as described in the text. The viable counts in samples are shown in the figure. Signal changes of the samples were determined using Biacore J equipped with sensor chip functionalized with (a) anti-O157 PAb, (b) anti-Salmonella PAb, and (c) anti-Listeria PAb. Results are shown as mean ± SD (n = 3).
