Portable nuclear magnetic resonance biosensor and assay for a highly sensitive and rapid detection of foodborne bacteria in complex matrices
© The Author(s). 2017
Received: 2 December 2016
Accepted: 17 February 2017
Published: 27 March 2017
Nuclear magnetic resonance (NMR) technique is a powerful analytical tool in determining the presence of bacterial contaminants in complex biological samples. In this paper, a portable NMR-based (pNMR) biosensor and assay to detect the foodborne bacteria Escherichia coli O157:H7 is reported. It uses antibody-functionalized polymer-coated magnetic nanoparticles as proximity biomarker of the bacteria which accelerates NMR resonance signal decay.
The pNMR biosensor operates at 0.47 Tesla of magnetic strength and consists of a high-power pulsed RF transmitter and an ultra-low noise sensing circuitry capable of detecting weak NMR signal at 0.1 μV. The pNMR biosensor assay and sensing mechanism is used in detecting E. coli O157:H7 bacteria in drinking water and milk samples. Experimental results demonstrate that by adding a filtration step in the assay, the pNMR biosensor is able to detect E. coli O157:H7 as low as 76 CFU/mL in water samples and as low as 92 CFU/mL in milk samples in about one min.
The pNMR biosensor assay and sensing system is innovative for foodborne bacterial detection in food matrices. The lowest detection level for E. coli O157:H7 in water and milk samples is essentially 101 CFU/mL. Although the linear range of detection is only from 101 to 104 CFU/mL, the wider detection range spans from 101 CFU/mL to 107 CFU/mL. Existing pNMR biosensors have detection limits at 103-104 CFU/mL only. The detection technique can be extended to other microbial or viral organisms by merely changing the specificity of the antibodies. Besides food safety, the pNMR biosensor described in this paper has potential to be applied as a rapid detection device in biodefense and healthcare diagnostic applications.
Nuclear magnetic resonance
Nuclear magnetic resonance (NMR) is a physical phenomenon where certain atomic nuclei absorb and re-emit electromagnetic radiation when placed in a magnetic field. NMR technique exploits this phenomenon to analyze the magnetic properties of atomic nuclei and determine chemical properties of bio-molecules. Since its signal is able to penetrate turbid raw samples, NMR techniques have wide applications in non-destructive food and biomedical diagnoses, which can simplify sample preparation processes and save analysis time . During the early stages of development, NMR analyses were performed in laboratories by placing the sample of interest inside a strong stationary magnet . Although the configuration produced accurate results, the instruments were rather large, heavy, not portable, and expensive [2, 3]. This has made the NMR technique less practical and less useful for on-site applications. Continuous development however resulted in the advancement of portable NMR (pNMR) sensor hardware and detection performance. pNMR sensors use magnetic nanoparticles and microparticles as proximity sensors to amplify molecular interactions . They are based on the reversible self-assembly of dispersed magnetic particles into stable nano-assemblies . When a few magnetic nanoparticles bind their intended molecular target through affinity ligands, they form magnetic clusters which leads to a corresponding decrease in the bulk spin-spin relaxation time (T 2) of the surrounding water molecules . Furthermore, measurements can be performed in turbid samples with minimal sample preparation and the sensing method is faster than those with surface-based techniques relying on molecular diffusion of targets to the sensing elements. These advantages render the proximity assay ideal for fast, simple and high-throughput sensing operations, especially in miniaturized device format. This class of sensors is more flexible, smaller in size, and suited for on-site and field-based measurements. They are also cheaper to develop and less costly to maintain as compared to bulky conventional NMR instruments.
A portable NMR system was reported in 2000 for a non-invasive spin-echo imaging of living plants in their natural environment . Recent advances in micro-fabrication technologies have accelerated the development of palm-sized NMR and high-throughput NMR spectrometers. The palm-sized NMR used integrated circuit technique and was tested in detecting biological molecules and cancer cells by measuring the sample’s NMR relaxation time which was proportional to the immunological clustering of magnetic nanoparticles . The reported sensitivity for cancer cell detection was in the range of 103-104 cells/mL . Other portable NMR developments include various designs [2–4, 6, 8, 9]. The development of high-throughput NMR spectrometers using complementary metal oxide semiconductor (CMOS) technology consisted of an array of high sensitivity micro-coils integrated with interfacing radio-frequency circuits on the same chip . A micro-nuclear magnetic resonance (μNMR) relaxometer miniaturized to palm-size and electronically automated for multi-step and multi-sample chemical/biological diagnosis was also developed . The μNMR relaxometer integrated microfluidic and microelectronic technologies to enable the association between the droplet management and μNMR assay inside a portable sub-Tesla magnet. Targets in unprocessed biological samples, captured by specific probe-decorated magnetic nanoparticles, were sequentially quantified by their spin–spin relaxation time (T2) via multiplexed μNMR screening. An NMR that can fit a 2 mm by 2 mm silicon chip was also reported .
Food safety is a major concern with so many microbial contaminations in recent years. Escherichia coli O157:H7 is one of the major bacterial pathogens that has caused foodborne illness outbreaks in the US and around the world. In 2009, E. coli food contamination in beef caused a foodborne illness outbreak across 8 US states for a total of 26 identified cases and 2 deaths, a recall of more than half a million pounds of ground beef products, and caused the Centers for Disease Control and Prevention (CDC) to issue a health alert [12, 13]. As recent as September 2016, E. coli O157:H7 contaminated beef caused a multi-state illness outbreak . The organism has also been implicated in various other types of food, such as spinach, cookie dough, cheese, sausage, and hazelnut. The aggregated annual cost of foodborne illness in the US alone is estimated at $77.7 billion .
The detection of E. coli O157:H7 is time consuming and requires complex instruments and extensive training. The procedure involves several steps including selective enrichment, filtration, incubation, confirmation, biochemical, and serological techniques . Polymerase chain reaction (PCR) based detection assays are sensitive and accurate however, they require complex sample preparation steps, such as DNA extraction and amplification, which increase additional diagnostic time [17, 18]. Fast detection methods have been reported based on immunological detection. However, most of these diagnostic systems have sensitivity limit greater than 102 CFU/mL, detection time of greater than 1 h, and are not applicable for on-field applications . In order to minimize the spread of infection and costly product recall, a rapid, sensitive, and portable detection of E. coli O157:H7 is essential in the food supply and healthcare applications.
Learning from the basic NMR design by Fukushima and Roeder , we developed a portable NMR-based biosensor and assay for the rapid and sensitive detection of foodborne pathogens. The novelty of our pNMR biosensor includes filtration assay, use of inexpensive NMR probe, and RF transceiver for microbial detection in complex matrices. The proximity biomarker uses an antibody-functionalized magnetic nanoparticles (Ab-MNP). The system is low cost, small in size, and highly portable for foodborne pathogen detection.
Design of pNMR biosensor
Test pathogen and antibodies
E. coli O157:H7 was obtained from the collection of the Nano-Biosensors Laboratory at Michigan State University. E. coli O157:H7 test strains were inoculated using sterile loop into 10 mL of Tryptic soy nutrient broth from Difco Laboratories (Detroit, MI) and incubated for 24 h at 37 °C to make a stock culture. The stock culture was then serially diluted in 0.1% peptone water in logarithmic scale to obtain different concentrations. All of the experiments were performed in a certified Biological Safety Level II laboratory. The antibody used for NMR biosensor was purified mouse anti-E. coli O157:H7 monoclonal antibody (Meridian Life Science, Inc. Saco, ME).
Antibody functionalization of magnetic nanoparticles
Magnetic nanoparticles (MNPs) were synthesized from aniline monomer and gamma iron (III) oxide (γ-Fe2O3) nanoparticles with weight ratio of 0.6:1 using a sol-gel chemical solution deposition method according to Pal et al. . The MNPs were filtered using 0.22 μm syringe filter (Millipore, MA, USA) to remove large particles. Pal et al.  had previously shown that the different ratios of iron oxide to aniline (1:0, 1:0.1, 1:0.4, 1:0.6, 1:0.8) resulted in various coercivity, retentivity, saturation magnetization, and temperature conductivity values. Electron diffraction micrographs of these ratios were also studied. The ratio of 1:0.6 was shown to be a better material in terms of these parameters.
To conjugate the antibodies to MNPs (Ab-MNP), 2.5 mg of MNPs were added into a 3 mL test tube. Then, 150 μL of 0.1 M phosphate buffer solution (PBS) was added to the test tube and sonicated for 15 min to evenly disperse the particles. Then, 100 μL of 2.5 mg/mL monoclonal anti E. coli O157:H7 antibody in PBS was added to the test tube and incubated in a hybridization oven at 37 °C for 1 h . The sample was rotated in the oven at 30 revolutions per minute (rpm) to facilitate antibody functionalization. The incubated solution was purified by magnetic separation process for three times using 0.01 M Tris buffer with 0.01% casein solution, and finally suspended in 2.5 mL of 0.01 M PBS. This prepared Ab-MNP solution was stored at 4 °C in the refrigerator until needed.
Magnetic pathogen interaction and filtration assay
Sensor architecture and detection principle
Detection and data analysis
The pNMR biosensor signal was measured using a digital oscilloscope Model Agilent DSO1024A (Agilent Technologies, Santa Clara, CA) that was connected to the pNMR’s signal output using BNC cable. For the biosensor measurement, a volume of 100 μL of the test solution was placed in the NMR tube. The pNMR spin-echo relaxation signal was recorded by the oscilloscope, which was controlled by FPGA synchronization signal. The whole process of pNMR relaxation was less than 1 min. For data analysis, a minimum of three replications were performed for each experiment. All measurements were calibrated using a control sample without the pathogen. Standard deviations and mean values for the data were calculated using Excel.
Results and discussion
The antibodies were conjugated onto MNP by physical non-covalent adsorption. The Ab-MNP conjugation is mainly formed by electrostatic interaction between the positively charged polyaniline surface and the negatively charged Fc portion of the antibodies, along with other factors, such as van der Waals interaction, electrostatic interaction, hydrophobic effect and hydrogen bonding [21, 23]. The successful functionalization was earlier confirmed by spectrophotometric studies . From UV-VIS scanning spectrophotometer measurement, pure antibody solution showed a characteristic peak of protein molecules at 280 nm. After the functionalization process and three times of magnetic separation, the supernatant solution showed no peak at 280 nm, demonstrating that antibodies were effectively conjugated onto the MNPs.
In this case, 1/T2 is equal to 17.97 resulting in T2 being equal to 0.0556 s or 55.6 ms. T2 of the spiked sample is 54.1% (55.6/102.7) shorter than that of the blank sample. These data show that T2 decays faster in samples with bacteria than in samples with no bacteria due to the formation of magnetic clusters around the bacterial cell walls and correspondingly decreasing the bulk spin-spin relaxation time of the surrounding water molecules.
The NMR biosensor signal for whole milk is comparable with that to water. The filtration assay has basically cleaned-up the matrix. Overall, the lowest detection for milk and water are in the order of 101 CFU/mL or specifically, 76 CFU/mL for water and 92 CFU/mL for milk. Bacteria can be detected from 101 CFU/mL to 107 CFU/mL. The wide detection range is highly relevant to food safety because certain organisms have varied infectious doses. For example, studies have shown that the infectious dose for some Shigella spp. is less than 10 organisms while the infectious dose for toxigenic V. cholera is 104 organisms . The infective dose for Salmonella is 103 organisms .
Results from the NMR biosensor show that the sensitivity is better than those in the existing literature. A chip-NMR biosensor designed by Lee et al. has a detection sensitivity of only 103 CFU/mL for Staphylococcus aureus . This level of detection limit is consistent with our results if filtration is not applied (103 CFU/mL of E. coli O157:H7, data not shown).
This paper has described a portable NMR biosensor that has been applied to detect E. coli O157:H7 in food matrices. The bacteria are labeled with the MNP through antibody-pathogen interaction. The detection of the biosensor system is fast, which includes a filtration assay for 20-30 min followed by signal detection of 1 min. The average detection limit for water and milk is 84 CFU/mL, much lower than those reported in the literature. The linear range of detection is from 101 to 104 CFU/mL while the detection range spans from 101 CFU/mL to 107 CFU/mL. The detection technique can be extended to other microbial or viral organisms with a change in the specificity of the antibodies in the Ab-MNP system. Besides food safety, the NMR biosensor described in this paper has potential to be applied as a rapid detection device in biodefense and health diagnostic applications.
Funding for this project was provided by the state of Michigan’s 21st Century Jobs Fund Program.
Availability of data and materials
Please contact author for data requests.
All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Wiley: NMR Spectroscopy: Basic Principles, Concepts, and Applications in Chemistry, 2nd Edition - Harald Günther. [Online]. Available: http://www.wiley.com/WileyCDA/WileyTitle/productCd-047195201X.html. [Accessed 14 Oct 2016].
- Ng SM. Chapter 3 - Portable NMR-Based Sensors in Medical Diagnosis A2 - ur-Rahman, Atta. In Applications of NMR Spectroscopy, M. I. Choudhary, Ed. Bentham Science Publishers, 2015, pp. 121–146.Google Scholar
- Yu D, Garcia N, Xu S. Toward portable nuclear magnetic resonance devices using atomic magnetometers. Concepts Magn Reson Part A. 2009;34A(2):124–32.View ArticleGoogle Scholar
- Perez JM, Josephson L, O’Loughlin T, Högemann D, Weissleder R. Magnetic relaxation switches capable of sensing molecular interactions. Nat Biotechnol. 2002;20(8):816–20.View ArticleGoogle Scholar
- Lee H, Sun E, Ham D, Weissleder R. Chip–NMR biosensor for detection and molecular analysis of cells. Nat Med. 2008;14(8):869–74.View ArticleGoogle Scholar
- Rokitta M, Rommel E, Zimmermann U, Haase A. Portable nuclear magnetic resonance imaging system. Rev Sci Instrum. 2000;71(11):4257–62.View ArticleGoogle Scholar
- Lee H, Yoon T-J, Figueiredo J-L, Swirski FK, Weissleder R. Rapid detection and profiling of cancer cells in fine-needle aspirates. Proc Natl Acad Sci. 2009;106(30):12459–64.View ArticleGoogle Scholar
- Zalesskiy SS, Danieli E, Blümich B, Ananikov VP. Miniaturization of NMR Systems: Desktop Spectrometers, Microcoil Spectroscopy, and “NMR on a Chip” for Chemistry, Biochemistry, and Industry. Chem Rev. 2014;114(11):5641–94.View ArticleGoogle Scholar
- Patel P. An NMR Chip The Size of a Seed. IEEE Spectrum: Technology, Engineering, and Science News, 05-Aug-2014. [Online]. Available: http://spectrum.ieee.org/tech-talk/biomedical/devices/an-nmr-device-the-size-of-a-seed. [Accessed 30 Sep 2016].
- Pourmodheji H, Ghafar-Zadeh E, Magierowski S. A Multidisciplinary Approach to High Throughput Nuclear Magnetic Resonance Spectroscopy. Sensors. 2016;16 no. 6.Google Scholar
- Lei K-M, Mak P-I, Law M-K, Martins RP. A palm-size μNMR relaxometer using a digital microfluidic device and a semiconductor transceiver for chemical/biological diagnosis. Analyst. 2015;140(15):5129–37.View ArticleGoogle Scholar
- July 1, 2009: Multistate Outbreak of E. coli O157 Infections Associated with Beef | E. coli CDC. [Online]. Available: http://www.cdc.gov/ecoli/2009/beef-jbs-swift-7-1-2009.html. [Accessed 14 Oct 2016].
- November 24, 2009: Multistate Outbreak of E. coli O157:H7 Infections | E. coli CDC. [Online]. Available: http://www.cdc.gov/ecoli/2009/. [Accessed: 14-Oct-2016].
- Multistate Outbreak of Shiga toxin-producing < em > Escherichia coli</em > O157:H7 Infections Linked to Beef Products Produced by Adams Farm. [Online]. Available: http://www.cdc.gov/ecoli/2016/o157h7-09-16/index.html. [Accessed: 14-Oct-2016].
- Scharff RL. Economic burden from health losses due to foodborne illness in the United States. J Food Prot. 2012;75(1):123–31.View ArticleGoogle Scholar
- Doyle MP, Schoeni JL. Isolation of Escherichia coli O157:H7 from retail fresh meats and poultry. Appl Environ Microbiol. 1987;53(10):2394–6.Google Scholar
- Botkin DJ, Galli L, Sankarapani V, Soler M, Rivas M, Torres AG. Development of a Multiplex PCR Assay for Detection of Shiga Toxin-Producing Escherichia coli, Enterohemorrhagic E. coli, and Enteropathogenic E. coli Strains. Front Cell Infect. Microbiol. 2012;2.Google Scholar
- Fratamico PM, Sackitey SK, Wiedmann M, Deng MY. Detection of Escherichia coli O157:H7 by multiplex PCR. J Clin Microbiol. 1995;33(8):2188–91.Google Scholar
- Zhang Y, et al. Sensitive chemiluminescence immunoassay for E. coli O157:H7 detection with signal dual-amplification using glucose oxidase and laccase. Anal Chem. 2014;86(2):1115–22.View ArticleGoogle Scholar
- Fukushima E, Roeder SBW. Experimental pulse NMR: a nuts and bolts approach. Reading, Mass.: Addison-Wesley Pub. Co., Advanced Book Program, 1981.Google Scholar
- Pal S, Setterington E, Alocilja EC. Electrically active magnetic nanoparticles for concentrating and detecting Bacillus anthracis spores in a direct-charge transfer biosensor. IEEE Sensors J. 2008;8(6):647–54.View ArticleGoogle Scholar
- Luo Y, et al. Novel Biosensor Based on Electrospun Nanofiber and Magnetic Nanoparticles for the Detection of E. coli O157:H7. IEEE Trans Nanotechnol. 2012;11(4):676–81.View ArticleGoogle Scholar
- Sharma S, Byrne H, O’Kennedy RJ. Antibodies and Antibody-Derived Analytical Biosensors. Ed. Pedro Estrela. Essays Biochem. 2016;60(1):9–18.View ArticleGoogle Scholar
- Kothary MH, Babu US. Infective Dose of Foodborne Pathogens in Volunteers: A Review. J Food Saf. 2001;21(1):49–68.View ArticleGoogle Scholar
- Loynachan AT, Harris DL. Dose Determination for Acute Salmonella Infection in Pigs. Appl Environ Microbiol. 2005;71(5):2753–5.View ArticleGoogle Scholar