Detection of Salmonella Typhimurium in Romaine Lettuce Using a Surface Plasmon Resonance Biosensor.

biosensors Article Detection of Salmonella Typhimurium in Romaine Lettuce Using a Surface Plasmon Resonance Biosensor Devendra Bhandari 1 , Fur-Chi Chen 2, * 1 2 3 * and Roger C. Bridgman 3 Department of Agricultural and Environmental Sciences, Tennessee State University, Nashville, TN 37209, USA Department of Human Sciences, Tennessee State University, Nashville, TN 37209, USA Hybridoma Facility, Auburn University, Auburn, AL 36830, USA Correspondence: ; Tel.: +1-615-963-5410



 Received: 20 June 2019; Accepted: 22 July 2019; Published: 28 July 2019 Abstract: Leafy vegetables have been associated with high-profile outbreaks causing severe illnesses. Timely and accurate identification of potential contamination is essential to ensure food safety. A surface plasmon resonance (SPR) assay has been developed for the detection of Salmonella Typhimurium in leafy vegetables. The assay utilizes a pair of well characterized monoclonal antibodies specific to the flagellin of S. Typhimurium. Samples of romaine lettuce contaminated with S. Typhimurium at different levels (between 0.9 and 5.9 log cfu/g) were pre-enriched in buffered peptone water. Three SPR assay formats, direct assay, sequential two-step sandwich assay, and pre-incubation one-step sandwich assay were evaluated. All three assay formats detect well even at a low level of contamination (0.9 log cfu/g). The SPR assay showed a high specificity for the detection of S. Typhimurium in the presence of other commensal bacteria in the romaine lettuce samples. The results also suggested that further purification of flagellin from the sample preparation using immunomagnetic separation did not improve the detection sensitivity of the SPR assay. The functional protocol developed in this study can be readily used for the detection of S. Typhimurium in leafy vegetables with high sensitivity and specificity. Keywords: biosensor; monoclonal antibodies surface plasmon resonance; Salmonella Typhimurium; flagellin; 1. Introduction Non-typhoidal Salmonella is one of the leading causes of foodborne illnesses in the world [1]. In the United States, Salmonella is responsible for 1,027,561 cases of illnesses, 19,336 hospitalizations, and 378 deaths annually [2]; which result in a direct and indirect economic cost of $3.3 billion [3]. Salmonella enterica serovar Typhimurium (S. Typhimurium) is the second most common serotype (after Salmonella enterica serovar Enteritidis) that causes foodborne illnesses [4]. These public health issues and enormous economic costs associated with illnesses mandate the need for rapid, sensitive, and specific S. Typhimurium detection methods. The culture based detection methods require intensive labor and take 3–4 days for the preliminary identification and 5–7 days for the confirmation [5]. Use of chromogenic and fluorogenic growth media in the culture based methods had reduced the detection time by days, but this was not fast enough to respond to disease outbreaks and product recalls [6]. Immunological methods like enzyme-linked immunosorbent assay (ELISA) are faster than the cultural methods [7], and are commonly used in the detection of S. Typhimurium [8–11]. However, lower sensitivity, requirement of pre-enrichment step and need for sample pre-treatment in ELISA still left the avenues to develop faster and more sensitive detection methods. Polymerase chain reaction (PCR) based molecular methods are faster and Biosensors 2019, 9, 94; doi:10.3390/bios9030094 www.mdpi.com/journal/biosensors Biosensors 2019, 9, 94 2 of 15 more sensitive than ELISA, and have been extensively studied for the detection of S. Typhimurium in foods [12–16]. Major drawbacks of the PCR detection methods are difficulty in automation, requirement of sample pre-enrichment [17,18] and false-negative results due to PCR inhibitors in the samples [19–21]. Recently the surface plasmon resonance (SPR) biosensor has gained attentions in foodborne pathogen detection because SPR assays are fast, label-free, and allow real-time monitoring of the biomolecular interactions with high sensitivity and specificity [22]. There have been limited studies on the detection of S. Typhimurium in food matrixes using SPR biosensors [23–27]; yet there is no report of the SPR assay for the detection of S. Typhimurium in leafy vegetables such as romaine lettuce. Most of these studies have used polyclonal antibodies to directly detect bacterial cells from food matrixes. However, polyclonal antibodies (mixture of various antibodies) have the tendency to bind with non-target antigens. This cross reactivity (nonspecific binding) is a major concern in many detection methods. In addition, use of bulk sized bacteria (0.7-1.5 × 2.0-5.0 µm) may hinder the binding of antibody with antigen in the SPR assay and result in the reduction of detection sensitivity. The problem of nonspecific binding of polyclonal antibodies associated with S. Typhimurium in the SPR assay can be circumvented by using monoclonal antibodies specific to the targeted antigens. Monoclonal antibodies specific to flagellin of S. Typhimurium have been produced [28] and characterized in terms of their binding kinetics and epitope maps, as described in our previous work [29]. Use of the well characterized monoclonal antibodies either in direct or sandwich assays will render the SPR assay more specific and sensitive. The purpose of this study is to develop a sensitive and specific SPR assay for the detection of S. Typhimurium in romaine lettuce. Different assay formats and their sensitivity and specificity were evaluated and a functional protocol for the routine application was developed. 2. Materials and Methods 2.1. Materials and Instrument S. Typhimurium (ATCC 13311) was acquired from American Type Culture Collection (Manassas, VA, USA) and stored at −80 ◦ C before use. Enterobacter cloacae, Pseudomonas fluorescens, Pseudomonas aeruginosa, Aeromonas salmonicida, Photobacterium damselae, Serratia spp., and Brucella spp. were isolated in our laboratory from the romaine lettuce purchased from a local grocery store. Tryptic soy agar (TSA), xylose-lysine-tergitol 4 (XLT-4) agar, and buffered peptone water (BPW) were supplied by Thermo Fisher Scientific Inc. (Lenexa, KS, USA). Difco plate count agar (PCA) and BBL lactose broth (LB) were purchased from Becton, Dickinson and company (Sparks, MD, USA). Bovine serum albumin (BSA), 10X phosphate buffered saline (PBS), and TWEEN 20 were obtained from Fisher Scientific (Fair Lawn, NJ, USA). PBST (1X PBS with 0.05% TWEEN 20) was prepared in our laboratory and used as working buffer and SPR running buffer. API-20E identification kits were purchased from bioMérieux, Inc. (Durham, NC, USA). N-(3-Dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC), N-Hydroxysuccinimide (NHS), ethanolamine hydrochloride, sodium acetate, and glycine were acquired from Sigma-Aldrich Inc (St. Louis, MO, USA). Deionized water was purified with a Millipor (...truncated)