Magnetophoretic Chromatography for the Detection of Pathogenic

Article pubs.acs.org/ac

Magnetophoretic Chromatography for the Detection of Pathogenic Bacteria with the Naked Eye Donghoon Kwon,†,§ Jinmyoung Joo,†,§ Jaejin Lee,† Ki-Hwan Park,‡ and Sangmin Jeon*,† †

Department of Chemical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, Republic of Korea Department of Food Science and Technology, Chung-Ang University, Gyeonggi 456-756, Republic of Korea

‡

S Supporting Information *

ABSTRACT: A facile and sensitive analytical method that uses gold-coated magnetic nanoparticle clusters (Au/MNCs) and magnetophoretic chromatography with a precision pipet has been developed for the detection of Salmonella bacteria. Antibody-conjugated Au/MNCs are used to capture the Salmonella bacteria in milk and are then separated from the milk by applying an external magnetic field. The Salmonellacontaining solution is sucked into a precision pipet tip to which a viscous polymer solution is then added. Once the magnetophoretic chromatography process has been carried out for 10 min, the presence of 100 cfu/mL Salmonella bacteria can be detected with the naked eye because the bacteria have become concentrated at the narrow pipet tip. The performance of this method was evaluated by using dynamic light scattering and light absorption spectroscopy.

F

dyes,13 semiconductor nanoparticles,14−16 or gold nanoparticles.12,17,18 To address these problems, we have developed a novel method for the detection of pathogenic bacteria that uses simple magnetic separation in a precision pipet. The Salmonella bacteria are magnetically separated from milk by using antibody-conjugated magnetic nanoparticle clusters (MNCs). The Salmonella-containing solution is sucked into a precision pipet tip to which a viscous polymer solution is added, and then the pipet is placed above a permanent magnet. The MNC− Salmonella complexes are attracted to the bottom of the tip whereas most of the free particles are trapped at the interface between the solutions. Once the separation is completed, the presence or otherwise of Salmonella bacteria can be determined with the naked eye because of the magnetic enrichment of the concentration of the complexes at the narrow pipet tip. The detection limit of this assay is 100 cfu/mL, which is comparable to results obtained with light absorption measurements.

ood poisoning is usually caused by pathogenic bacteria and is a major public health issue. Since microorganisms tend to proliferate over time, the detection of pathogenic bacteria at an early stage is crucial for preventing food-borne illnesses.1 Although the microbial culture technique is the most popular method for the detection of pathogenic bacteria,2 it does not provide on-site feedback because such assays typically require cell culture times of a few days. A number of rapid detection methods based on immunoassays that use fluorescent and radioactive tags3−5 or the polymerase chain reaction (PCR)6−8 have been developed as replacements for the time-consuming culture-based detection method, but they still have drawbacks, such as cost inefficiency and the use of hazardous materials or complex pretreatment procedures. A breakthrough in the rapid and cost-effective detection of food poisoning bacteria has been realized through the use of functional nanoparticles. Antibody-conjugated magnetic nanoparticles can be used to capture and separate such bacteria under an external magnetic field.9−12 To determine the concentration of the bacteria, the particle-bacteria complexes must be separated from the free magnetic particles. A number of size sorting methods including size exclusion chromatography, filtration, centrifugation, and electrophoresis can be used to achieve this separation because particle-bacteria complexes are much larger than free nanoparticles. However, these methods require extensive instrumentation and recovery of the separated complex from the filter, which results in the loss of analytes and degrades detection sensitivity. Hence, instead of direct separation, most conventional techniques adopt timeconsuming and cost-ineffective methods, particularly the conjugation of the particle-bacteria complexes with fluorescent © XXXX American Chemical Society

■

EXPERIMENTAL SECTION Materials. Iron(III) chloride hexahydrate, sodium citrate, polyacrylamide, urea, 3-amino-propyltriethoxysilane (APTES), 16-mercaptohexadecanoic acid (MHDA), 1-ethyl-3-[3dimethylaminopropyl]carbodiimide (EDC), N-hydroxysuccinimide (NHS), bovine serum albumin, and poly(ethylene glycol) (av MW = 8000 g/mol) were purchased from Aldrich (St. Louis, MO) and used without further purification. Deionized water (18.3 M cm−1) was obtained with a reverse osmosis water

Received: June 8, 2013 Accepted: July 5, 2013

A

dx.doi.org/10.1021/ac401717f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

■

Article

RESULTS AND DISCUSSION Characterization of the Gold-Coated Fe3O4 Magnetic Nanoparticle Clusters (Au/MNCs). Figures 1a and b show

purification system and used to prepare phosphate buffer (PB) and poly(ethylene glycol) (PEG) aqueous solutions. Monoclonal salmonella antibody was purchased from Abcam Inc. (Cambridge, MA). Preparation of Gold-Coated Fe3O4 Magnetic Nanoparticle Clusters (Au/MNCs). The Fe3O4 magnetic nanoparticle clusters (MNCs) were synthesized by using a one-pot hydrothermal method.19 In brief, 4 mmol of FeCl3, 12 mmol urea, and 8 mmol sodium citrate were dissolved in 80 mL of DI water and then 0.6 mmol polyacrylamide was added to the solution. The solution was maintained at 200 °C in a Teflonlined autoclave for 10 h. After it was cooled to room temperature, the precipitate was separated and rinsed several times with water and absolute ethanol. The resulting Fe3O4 MNCs were dispersed in absolute ethanol and treated with 3amino-propyltriethoxysilane (APTES) to produce a selfassembled monolayer of amine groups on the surface. Two nanometer gold nanoseeds were synthesized as described elsewhere.20 In brief, 1 mL of 1 wt % HAuCl4 and 2 mL of 38.8 mM sodium citrate were dissolved in 90 mL of DI water and then 0.075 wt % NaBH4 was added to the solution under vigorous stirring. The synthesized gold nanoseeds were attached to the APTES-coated Fe3O4 MNCs. Then, the reduction reaction of gold ions to metallic gold on the gold nanoseed was carried out to increase the size of the gold nanoparticles.21 Antibody Immobilization on the Gold-Coated Fe3O4 Magnetic Nanoparticle Clusters (Au/MNCs). Au/MNCs were functionalized sequentially with 16-mercaptohexadecanoic acid (MHDA), 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide (EDC), and N-hydroxysuccinimide (NHS) to produce stable amine-reactive NHS ester linkers on their surfaces, and then monoclonal Salmonella antibodies were immobilized onto the Au/MNCs. Magnetophoretic Chromatography Technique for Salmonella Detection. Salmonella-spiked samples were prepared at concentrations in the range 10−105 cfu/mL in milk. 100 μL of the antibody-functionalized Au/MNCs were added to 10 mL of the Salmonella-spiked milk, and the solutions were incubated for 1 h with gentle shaking. After incubation, the mixture of Au/MNCs-Salmonella complexes and free Au/MNCs was extracted from the milk sample by applying an external magnetic field, and then redispersed in 50 μL of buffer solution for subsequent magnetophoretic chromatography. The buffer solution containing Au/MNCsSalmonella complexes and free Au/MNCs was sucked into a precision pipet tip, to which 80 μL of a polyethylene glycol (PEG) solution was then added. To avoid abrupt suction, the sucking was conducted by adjusting the volume setting wheel of the pipet. The magnetophoretic chromatography separation was carried out by placing a permanent magnet underneath the pipet tip for 10 min. The particles at the end of the pipet tip were dispersed in 50 μL of buffer solution and the light absorbance was measured by using a UV−vis spectrometer (Ocean Optics, Dunedin, FL) for quantitative analysis. Size Distribution Measurements. The size distributions of free Au/MNCs, Au/MNCs-Salmonella complexes, and the mixture were determined with a dynamic light scattering technique. Scattered light was analyzed by using a Zetasizer (Malvern Instruments Ltd., England) with a He−Ne laser at a detection angle of 173°.

Figure 1. Characterization of gold-coated magnetic nanoparticle clusters. SEM images of (a) Fe3O4 magnetic nanoparticle clusters (MNCs) and (b) gold-coated magnetic nanoparticle clusters (Au/ MNCs). (c) The absorption spectra of MNCs (red) and Au/MNCs (blue) in buffer solutions. (d) TEM image of a Au/MNCs-bound Salmonella bacterium.

scanning electron microscopy (SEM) images of the Fe3O4 MNCs and Au/MNCs, respectively. The average sizes of the MNCs and Au/MNCs were found to be ∼150 and ∼160 nm, respectively, and both particle samples have narrow size distributions. The average size of a single Fe3O4 nanoparticle is 15 nm, so each Au/MNC particle consists of a few hundred superparamagnetic Fe3O4 particles. The magnetic force experienced by magnetic particles is proportional to the particle volume, and thus the MNCs (or Au/MNCs) can be separated from analyte solutions more effectively than small Fe3O4 nanoparticles. UV−vis absorption spectra have been used to show that >99% separation is achieved within 1 min. Figure 1c compares the absorption spectra of the MNCs and Au/MNCs. Greater absorption of visible light is observed for the Au/MNC solution than for the MNC solution, which enhances the sensitivity of the light absorption measurements. In addition, the choice of gold nanoparticles enables the use of wellestablished thiol chemistry to improve the orientation of the antibodies on the Au/MNCs. Figure 1d shows a TEM image of a Salmonella bacterium exposed to the antibody-conjugated Au/MNCs in a phosphate buffer solution. Magnetophoretic Separation of Free Au/MNCs and Au/MNCs-Salmonella Complexes. Figures 2a and b show schematic illustrations and the corresponding optical images of the magnetophoretic chromatography setup before and after the separation of Au/MNC−Salmonella complexes from free Au/MNCs respectively. The formation of a complex including a bacterial cell with a number of antibody-conjugated Au/ MNCs results in a different response from that of free Au/ MNCs to an external magnetic field. The magnetophoretic mobility of magnetic particles under an external magnetic field depends on the intrinsic properties of the particles and the medium, such as the particle size, the viscosity of the medium, and the magnetic susceptibilities of both the medium and the B



Article

viscosity has a larger surface tension. The relationship between surface tension and viscosity is22

γ = αe − β / η

(1)

where γ is the surface tension of the solution, η is the viscosity, and α, β are empirical constants. Most of the free nanoparticles with lower settling velocities are captured at the interface between the two solutions whereas the complexes with higher settling velocities pass through the interface. As a result, only the Au/MNCs-Salmonella complexes reach the bottom of the tip during the magnetophoretic chromatography procedure. To evaluate the magnetic separation technique, a dynamic light scattering (DLS) analysis was performed. Figure 3a shows the number mean size distribution curves obtained from the sample in Box 1 of Figure 2a. The two peaks at 150 and 800 nm correspond to free Au/MNCs and the Au/MNCs-Salmonella complexes respectively. The size of the Au/MNCs-Salmonella complexes is smaller than the typical size of Salmonella bacteria due to the multiple light scattering between Au/MNCs bound to a Salmonella bacterium. The multiple scattering affects the cross correlation and leads to underestimate the hydrodynamic diameter of particles.23 The concentration of free Au/MNCs is higher than that of the complexes, so the peak at 150 nm is larger than that at 800 nm. The presence of the peak at 800 nm is more evident in the intensity mean size distribution (see the inset in Figure 3a) because the scattering intensity of a particle is proportional to the sixth power of its diameter. After 10 min of magnetophoretic chromatography, DLS measurements were conducted for the samples in Boxes 2 and 3 of Figure 2b and the results are shown in Figures 3b and c respectively. In contrast to the results in Figure 3a, there is only a single peak in each of Figures 3b and c. The peak in Figure 3b corresponds to free Au/MNCs, which indicates that most of the free Au/ MNCs are trapped at the interface between the buffer and the PEG solution and that only free Au/MNCs remain in the sample in Box 2. In contrast, the single peak at 800 nm in Figure 3c indicates that only Au/MNCs-Salmonella complexes have reached the end of the pipet tip, which demonstrates that magnetophoretic chromatography is an efficient method for the separation of these particles with different sizes. Detection of Salmonella Using Magnetophoretic Chromatography. After incubation of the antibody-conjugated Au/MNCs in 10 mL Salmonella-spiked milk samples, a mixture of Au/MNC−Salmonella complexes and free Au/ MNCs was extracted from the milk sample by applying an external magnetic field and then redispersed in 50 μL of buffer solution for subsequent magnetophoretic chromatography. Note that the concentration of Salmonella bacteria in the

Figure 2. Schematic illustrations and the corresponding optical images of the magnetophoretic chromatography process (a) before and (b) after magnetophoretic chromatography.

particles. The forces determining the movement of the free Au/ MNCs and Au/MNC−Salmonella complexes through a liquid medium are the magnetic force (Fm), buoyant force (Fb), gravitational force (Fg), and drag force (Fd). Although it is not straightforward to quantify these forces because of the lack of information about the number and distribution of Au/MNCs bound to each bacterium, it is safe to conclude that the magnetic force dominates the other forces because no separation of particles was observed in a control experiment without an external magnetic field (see the Supporting Information). Note that the magnetic force is proportional to the volume of a magnetic particle and that small particles are more susceptible to Brownian motion, which means that the settling velocity under a magnetic field of the larger particles is higher than that of the smaller particles. A PEG aqueous solution was used as the viscous medium for the magnetophoretic chromatography separation process because PEG increases the viscosity of the solution substantially without noticeably affecting the density. The best results were obtained when the magnetophoretic chromatography separation was carried out using a 25 wt % PEG solution for 10 min (see the Supporting Information). The viscosity of a solution affects the mobility of particles: particles move more slowly in a more viscous solution. More importantly, a liquid with higher

Figure 3. DLS number mean size distribution curves of the solutions in the boxes in Figure 2: (a) box 1, (b) box 2, and (c) box 3. Each inset shows the intensity mean size distribution curve of the sample. C



Article

Figure 4. Salmonella detection with the naked eye. Optical images of pipet tips after the magnetic separation of milk samples spiked with various concentrations of Salmonella bacteria.

MNCs and the Au/MNC−Salmonella complexes was achieved with this magnetophoretic chromatography technique. Further, negligible changes in intensity were observed for samples without Salmonella bacteria (0 cfu/mL), which confirms the high performance of this technique. Figure 5b shows the variation with Salmonella concentration in the absorption peak intensity of the sample in Box A. The limit of detection in milk is ∼100 cfu/mL, which is comparable to that obtained with the naked eye, as shown in Figure 4. The light absorption measurements require the redispersion of the Salmonella bacteria extracted from the pipet tip into 50 μL of buffer solution and this dilution degrades the sensitivity of the measurements. To determine the selectivity of this assay, a control experiment was conducted. After incubation of the Salmonella antibody-conjugated Au/MNCs in a milk sample containing 108 cfu/mL of Escherichia coli, the magnetophoretic chromatography procedure was carried out. Negligible change in the absorption spectra was observed for the samples, which indicates that the specific binding between the antibodyconjugated nanoparticles and Salmonella results in the desired selectivity.

resulting solution is 200 times higher than that in the original sample because of the magnetic enrichment. Figure 4 shows optical images of the pipet tips after 10 min of magnetophoretic chromatography. As the concentration of Salmonella bacteria increases, the color at the end of the tip becomes dark whereas the color of the interface of the solutions becomes pale. The Au/MNCs-Salmonella complexes become concentrated at the narrow tip of the pipet, so the presence of Salmonella bacteria can be confirmed with the naked eye for concentrations down to 100 cfu/mL. This limit of detection is comparable to the results obtained by using UV−vis absorption spectroscopy and magnetic enrichment. To determine the concentration in each solution of Salmonella bacteria quantitatively, the solution in the pipet tip was divided into three 15 μL samples (A−C) and one 45 μL sample (D), as shown in the inset in Figure 5a. The solution in

■

CONCLUSION

We have developed a magnetophoretic chromatography separation technique that uses a precision pipet and applied this method to the detection of Salmonella bacteria in milk with the naked eye. This technique was evaluated by using UV−vis spectroscopy and dynamic light scattering and found to enable detection of Salmonella at concentrations down to 100 cfu/mL. The detection sensitivity can easily be improved by impregnating the magnetic nanoparticle clusters with fluorescent tags. Because the experimental setup requires only a permanent magnet and a precision pipet, the developed method is simple and cost-efficient, which means that it can be used in the rapid on-site detection of bacteria at restaurants, hospitals, and even homes for food safety monitoring. In particular, this assay is expected to play a key role in preventing the outbreak of foodborne diseases caused by microorganisms in developing countries.

Figure 5. Variations in the light absorption intensities of the solution in (a) each box and (b) box A with the concentration of Salmonella bacteria.

each sample was dispensed into a cuvette and the light absorption intensity at 594 nm was measured by using a portable UV−vis spectrometer. Figure 5a shows that the light absorption intensity of the sample in box A increases with the concentration of Salmonella bacteria because more Au/MNC− Salmonella complexes have reached the end of the tip. The increase in the light absorption intensity of the sample in box A corresponds to the decrease in the light absorption intensity of the sample in box D. Whereas substantial changes are evident for the samples in boxes A and D, only slight changes in light absorption were observed for the samples in boxes B and C, which indicates that almost complete separation of free Au/ D



■

ASSOCIATED CONTENT

■

AUTHOR INFORMATION

Article

(22) Schonhorn, H. J. Chem. Eng. Data 1967, 12 (4), 524−525. (23) Meyer, W. V.; Cannell, D. S.; Smart, A. E.; Taylor, T. W.; Tin, P. Appl. Opt. 1997, 36 (30), 7551−7558.

S Supporting Information *

Additional figures and table for the effects of the density and viscosity of PEG solutions on the performance of magnetophoretic chromatography. This material is available free of charge via the Internet at http://pubs.acs.org.

Corresponding Author

*E-mail: [email protected]. Author Contributions §

Donghoon Kwon and Jinmyoung Joo contributed equally to this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank H. Cha for helpful discussions. This research was supported by a grant (10162MFDS995) from Ministry of Food and Drug Safety in 2013.

■

REFERENCES

(1) Jones, K. E.; Patel, N. G.; Levy, M. A.; Storeygard, A.; Balk, D.; Gittleman, J. L.; Daszak, P. Nature 2008, 451 (7181), 990−993. (2) Hobson, N. S.; Tothill, I.; Turner, A. P. F. Biosens. Bioelectron. 1996, 11 (5), 455−477. (3) Carrillo-Carrión, C.; Simonet, B. M.; Valcárcel, M. Biosens. Bioelectron. 2011, 26 (11), 4368−4374. (4) Rattray, N. J. W.; Zalloum, W. A.; Mansell, D.; Latimer, J.; Schwalbe, C. H.; Blake, A. J.; Bichenkova, E. V.; Freeman, S. Chem. Commun. 2012, 48 (51), 6393−6395. (5) Zhu, C.; Yang, Q.; Liu, L.; Wang, S. Angew. Chem., Int. Ed. 2011, 50 (41), 9607−9610. (6) Rodríguez-Lázaro, D.; D’Agostino, M.; Herrewegh, A.; Pla, M.; Cook, N.; Ikonomopoulos, J. Int. J. Food Microbiol. 2005, 101 (1), 93− 104. (7) Jofré, A.; Martin, B.; Garriga, M.; Hugas, M.; Pla, M.; RodríguezLázaro, D.; Aymerich, T. Food Microbiol. 2005, 22 (1), 109−115. (8) Deisingh, A. K.; Thompson, M. J. Appl. Microbiol. 2004, 96 (3), 419−429. (9) Gu, H.; Xu, K.; Xu, C.; Xu, B. Chem. Commun. 2006, No. 9, 941− 949. (10) Liébana, S.; Lermo, A.; Campoy, S.; Cortés, M. P.; Alegret, S.; Pividori, M. I. Biosens. Bioelectron. 2009, 25 (2), 510−513. (11) Mouffouk, F.; Rosa da Costa, A. M.; Martins, J.; Zourob, M.; Abu-Salah, K. M.; Alrokayan, S. A. Biosens. Bioelectron. 2011, 26 (8), 3517−3523. (12) Wang, C.; Irudayaraj, J. Small 2010, 6 (2), 283−289. (13) Zhu, P.; Shelton, D. R.; Li, S.; Adams, D. L.; Karns, J. S.; Amstutz, P.; Tang, C.-M. Biosens. Bioelectron. 2011, 30 (1), 337−341. (14) Su, X.-L.; Li, Y. Anal. Chem. 2004, 76 (16), 4806−4810. (15) Hahn, M. A.; Tabb, J. S.; Krauss, T. D. Anal. Chem. 2005, 77 (15), 4861−4869. (16) Joo, J.; Yim, C.; Kwon, D.; Lee, J.; Shin, H. H.; Cha, H. J.; Jeon, S. Analyst 2012, 137 (16), 3609−3612. (17) Singh, A. K.; Senapati, D.; Wang, S.; Griffin, J.; Neely, A.; Candice, P.; Naylor, K. M.; Varisli, B.; Kalluri, J. R.; Ray, P. C. ACS Nano 2009, 3 (7), 1906−1912. (18) Khan, S. A.; Singh, A. K.; Senapati, D.; Fan, Z.; Ray, P. C. Chem. Commun. 2011, 47 (33), 9444−9446. (19) Cheng, W.; Tang, K.; Qi, Y.; Sheng, J.; Liu, Z. J. Mater. Chem. 2010, 20 (9), 1799−1805. (20) Chun, C.; Joo, J.; Kwon, D.; Kim, C. S.; Cha, H. J.; Chung, M.S.; Jeon, S. Chem. Commun. 2011, 47 (39), 11047−11049. (21) Wang, L.; Bai, J.; Li, Y.; Huang, Y. Angew. Chem., Int. Ed. 2008, 47 (13), 2439−2442. E


Magnetophoretic Chromatography for the Detection of Pathogenic

Recommend Documents