lable at ScienceDirect

Food Control 47 (2015) 264e276

Contents lists avai

Food Control

journal homepage: www.elsevier .com/locate/ foodcont

Review

Review of Salmonella detection and identification methods: Aspects ofrapid emergency response and food safety

Kyung-Min Lee a, *, Mick Runyon a, Timothy J. Herrman a, Robert Phillips b, John Hsieh a

a Office of the Texas State Chemist, Texas AgriLife Research, Texas A&M University System, College Station, TX 77841, United Statesb Food Safety and Inspection Service, U.S. Department of Agriculture, Athens, GA 30605, United States

a r t i c l e i n f o

Article history:Received 14 May 2014Received in revised form30 June 2014Accepted 3 July 2014Available online 11 July 2014

Keywords:Salmonella detection methodFood emergency responseFood safetyFoodborne illnessPublic health

* Corresponding author. Tel.: þ1 979 845 1121; faxE-mail address: kml@otsc.tamu.edu (K.-M. Lee).

http://dx.doi.org/10.1016/j.foodcont.2014.07.0110956-7135/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Salmonella has been recognized as a major and important foodborne pathogen for humans and animalsover more than a century, causing human foodborne illness as well as high medical and economical cost.Accordingly, the effort to develop efficient and reliable Salmonella detection methods continues. Thispaper reviews and describes the development and application of commercially available Salmonelladetection methods. These are categorized into several groups based on the principle applied: conven-tional culture methods, immunology-based assays, nucleic acid-based assays, miniaturized biochemicalassays, and biosensors. Conventional culture methods serve as the basis in food testing laboratoriesdespite rather laborious and time-consuming protocols. Considerable progress in rapid methods usingemerging technologies yield faster answers and higher throughput of samples. This paper also shows andanalyzes Salmonella test results and summarizes the features and limitations of the studies involvingSalmonella detection methods developed for emergency response, mainly by food emergency responselaboratories participating during recent fiscal years. The emergency response laboratories utilize Sal-monella detection methods possessing properties that include simplicity, versatility, high sensitivity,good specificity, and cost efficiency. Collaboration of the food emergency response laboratories in thedevelopment of these technologies is important essentially to compare for the purpose of continuallyimproving Salmonella detection methods.

© 2014 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2652. Conventional Salmonella detection methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2653. Rapid Salmonella detection methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266

3.1. Immunology-based assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2683.1.1. Enzyme-linked immunoabsorbent assay (ELISA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2683.1.2. Latex agglutination assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2683.1.3. Immunodiffusion assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2683.1.4. Immunochromatography (dipstick) assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

3.2. Nucleic acid-based assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2693.2.1. Polymerase chain reaction (PCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2693.2.2. DNA probe hybridization assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

3.3. Miniaturized biochemical assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2703.4. Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270

3.4.1. Enzyme biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2703.4.2. Nucleic acid biosensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2703.4.3. Antibody-based biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2713.4.4. Micro- and nano-biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

: þ1 979 845 1389.

K.-M. Lee et al. / Food Control 47 (2015) 264e276 265

4. Salmonella detection methods for food emergency response laboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2715. Selection of Salmonella detection methods for rapid emergency response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

5.1. Matrix effect and sample preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2735.2. Selectivity and specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2735.3. Analysis time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2735.4. Cost efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

6. Future aspects of Salmonella detection methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2737. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

1. Introduction

Salmonella bacteria represent the most common and primarycause of food poisoning in many countries for at least over 100years (Alakomi & Saarela, 2009; Chalker & Blaser, 1988; Coburn,Grassl, & Finlay, 2007). Despite well-established instructions andmeasures for preventing salmonellosis (Salmonella food poisoning),the incidence and severity of human salmonellosis have signifi-cantly increased. Most salmonellosis cases are self-limiting(approximately 80%), but large outbreaks caused in schools, hos-pitals, and restaurants are not very common (Guiney et al., 1995). Atpresent, over 2500 serotypes of Salmonella have been reported(Fierer & Guiney, 2001). Of these, the most common serotypesassociated with human illness are Salmonella enterica serovartyphimurium (S. Typhimurium) and S. enterica serovar Enteritidis(S. Enteritidis) in the United States and European countries.

Approximately, 1.4 million human Salmonella infections occurannually in the United States with assumption of only 2% casesreported to Center for Disease Control and Prevention (CDC),resulting in about 16,000 hospitalizations with nearly 600 deaths(Cummings et al., 2010; Mead et al., 1999; Turner, 2010). The esti-mated total cost associated with Salmonella incidences may be upto several billion dollars annually (Frenzen et al.,1999;WHO, 2005).An estimated annual cost for a Salmonella control program hasincreased in some countries (WHO, 2005). Much higher incidencesrelated to Salmonella may occur in some developing countrieswhere relevant data is not readily available.

Salmonella is not considered to be fatal to healthy people or as abioweapon agent. However, efforts have been made to develop andimprove detection technologies for this organism because Salmo-nellamay cause devastating foodborne illness. Salmonella detectionmethods are desirable to have sensitivity enough to detect one cellin a defined sample. The analysis time of conventional and rapidmethods can vary with cell enrichment steps to reach minimal cellconcentration enough for Salmonella detection. The cell enrichmentprocess is typically lengthy in a conventional method whereas therapid detection method generally requires at least 104 cells ml�1 ofSalmonella concentration for detection.

Salmonella surveillance and monitoring should be based onreliable and efficient detection methods, which should helpimprove the food safety (Rodríguez-L�azaro et al., 2007). It isessential that surveillance and monitoring should cover the entirefood chain, preferably starting from investigation of feed and feedingredients for Salmonella contamination (Alakomi & Saarela,2009; Mead et al., 2010). Standardization, regulation, and interna-tional surveillance networks are also necessary to effectively pre-vent and control Salmonella pathogens. International Standards forthe Salmonella tests such as ISO 6579:2002 (Amd 1:2007 Annex D)and the Nordic Committee on Food Analysis (NMKL) contain in-formation about sample storage, sampling, and other critical stepsin Salmonella analysis (Feldsine et al., 2003; L€ofstr€om, Hansen, &Hoorfar, 2010; NordVal, 2009). Such standards are critical to

ensure that the results are consistent and comparable among lab-oratories. Several countries have built regulations and guidelines,such as Regulation (EC) No 2160/2003, the EU Zoonoses MonitoringDirective (2003/99/EC), and the Food and Drug Administration(FDA) Food Code (EC, 2003a, 2003b; FDA, 2009). These regulationsand guidelines provide or require information on Salmonella andcontrol methods as well as help develop the food safety rules andregulatory policies, in order to reduce Salmonella pathogens at anystage of feed and food production (de Jong & Ekdahl, 2006; Tietjen& Fung, 1995).

Bioterror events, such as anthrax spores in contaminated letters,have greatly altered the view of the public and the scientific com-munity about bacterial pathogens which could be used as weaponsof biological terrorism, and about the needs and requirements ofthe biotechnology for detection of potential bioterrorism-employed pathogens (Cannons, Amuso, & Anderson., 2006). Sal-monella spp. is one of such bacterial pathogen which has beenpurposely spread mainly using food as a carrier (Ashford et al.,2003; T€or€ok et al., 1997). For example, S. Typhimurium was usedto infect residents by contaminating salad bars at the local res-taurants in Oregon. The United States capability to respond to abioterrorism event has been enhanced by integrated laboratorynetworks including the Food Emergency Response Network (FERN)and Laboratory Response Network (LRN) of Centers for DiseaseControl and Prevention (CDC) as well as to emerging infectiousdiseases.

The purpose of this paper is to review Salmonella detectionmethods including detection level, sensitivity, specificity, andsample matrices. Included in the discussion are a summary of Sal-monella testing results and information from food emergencyresponse laboratories available in a web-based database portal forrecent fiscal years.

2. Conventional Salmonella detection methods

Traditional isolation of Salmonella spp. involves a nonselectivepre-enrichment of a defined weight or volume of the sample, fol-lowed by a selective enrichment step, plating onto selective agars,and biochemical and serological confirmation of suspect colonies(Fig. 1). Different approaches of Salmonella enrichment using theunique biochemical physical properties of the organisms have beenstandardized by several regulatory agencies, such as InternationalOrganization for Standardization (ISO), Association of OfficialAnalytical Chemists (AOAC), FDA, and Food Safety and InspectionService (FSIS) of USDA. The current ISO horizontal method, ISO6579:2002, consists of a pre-enrichment of samples in bufferedpeptone water (BPW) followed by a selective enrichment in Rap-paporteVassiliadis (soya base) (RVS) and Muller-Kauffmann Tet-rathionate-Novobiocin (MKTTn) (ISO, 2002). Standard methodspublished for detection of Salmonella by other regulatory agenciesare essentially similar to ISO 6579:2002. The conventional cultural

Fig. 1. Scheme of Food Emergency Response Network (FERN)/Food and Drug Admin-istration (FDA) BAM and Salmonella culture method. RV, RappaporteVassiliadis me-dium; BS, bismuth sulfite agar; HE, Hektoen enteric agar; XLD, xylose lysinedesoxycholate agar; TT, tetrathionate broth, TSI, Triple sugar iron agar, and LIA, Lysineiron agar.

K.-M. Lee et al. / Food Control 47 (2015) 264e276266

methods for Salmonella recovery basically include the followingessential steps.

Pre-enrichment uses a nutritious nonselective medium torecover sublethally injured Salmonella cells while inhibiting thegrowth of competing flora (Tietjen & Fung, 1995). Most commonlyused media in pre-enrichment step are BPW and lactose broth. Thepre-enrichment process and formulas have been tested to increasethe sensitivity of Salmonella detection. Antimicrobial resistant Sal-monella spp. could be highly recovered by using appropriate anti-biotics in pre-enrichment and selective media (Carrique-Mas &Davies, 2008). The incubated pre-enrichment media are inocu-lated into selective media containing two or more inhibitory re-agents such as bile salts, brilliant green, thiosulphate, deoxycholate,malachite green, novobiocin, tetrathionate, cycloheximide, nitro-furantoin, and sulphacetamide (Arroyo & Arroyo, 1995; Ha, Pillai, &Ricke, 1995; Ricke, Pillai, Norto, Maciorowski, & Jones, 1998). Theuse of inhibitors in a selective media allows continuous growth ofSalmonella while suppressing the propagation of other bacteria(Tietjen & Fung, 1995). RappaporteVassiliadis (RV) medium andtetrathionate (TT) broth has been used as official Salmonellaenrichment media in approved standard methods such as FDABacteriological Analytical Manual (BAM) and FERN Salmonellamethods. Enrichment media has been modified to enhance Sal-monella growth and thus to increase the sensitivity and selectivity.

The selective enrichment media containing Salmonella at thelevel of above 104 cells ml�1 are streaked on solid selectivemedia toisolate presumptive positive Salmonella colonies, simultaneouslyinhibiting the growth of other bacteria (Carrique-Mas & Davies,2008; Ruiz et al., 1996; Tietjen & Fung, 1995). Commonly usedplating media include Salmonella-Shigella agar (SS), brilliant greenagar (BGA), bismuth-sulfite agar (BSA), Hektoen enteric (HE), andxylose-lysine-deoxycholate agar (XLD). The colonies of differentSalmonella serotypes are differentiated by colors with the coliforms

on these media (Mallinson et al., 2000; Ruiz et al., 1996). Forexample, S. Typhi on SS may appear as colorless colonies with blackcenter. Lactose-fermenting S. Arizonae serotype producing lightpink-red halo around colonies can be also identified on XLD, butother lactose-fermenting serotypes such as S. Montevideo and S.Virchow may not be detected on the same agar medium. However,some serotypes are not distinctive and evenmissed on thosemedia,yielding false negatives and increasing cost for additional tests(Carrique-Mas& Davies, 2008; Gruenewald, Henderson,& Yappow,1991; Manafi, 2000; Reid, Porter,& Ball, 1993; Ruiz et al., 1996). Therecovery and selectivity of Salmonella may be further enhanced byusing two or more selective media or adding appropriate supple-ments to plating media such as novobiocin, malachite green, thio-sulphate, and sulphamethazine (Arroyo & Arroyo, 1995; Carrique-Mas & Davies, 2008; Kang & Fung, 2000; Tietjen & Fung, 1995).

Presumptive Salmonella colonies isolated on plating media aretypically incubated in triple sugar iron agar (TSI) and lysine ironagar (LIA) followed by urease test and additional tests for urease-negative cultures. TSI and LIA media use glucose-fermentationand lysine decarboxylase reactions for screening Salmonella spp,respectively. The cultures giving typical reactions for Salmonella aresubmitted for biochemical and serological identification tests.Serotyping identifies the serogroups of the cultures by antigenicanalysis using agglutination reaction subject to antigenic structure.However, the same Salmonella serotype can vary with antigenicitydue to change and loss of the surface antigens, reducing thesensitivity of serological methods (Hoorfar, Aherns, & Rådstr€om,1999; Sørensen, Alban, Nielsen, & Dahl, 2004). If Salmonella iso-lates cannot be serotyped under certain conditions, PFGE charac-terization might be useful in place of serological assays.

The conventional microbiological methods serve as the basis foranalysis in many food safety and public health laboratories due tothe ease of use, reliability of results, high sensitivity and specificity,and lower cost compared to emerging molecular-based technolo-gies (Gracias & McKillip, 2004; Maciorowski, Herrera, Jones, Pillai,& Ricke, 2006). However, these procedures need to prepare mul-tiple subcultures required for several identification steps, takingmore than 5 days for complete isolation and confirmation. Inaddition, false positive results may occur due to competitive flora(e.g. proteus) (Naravaneni & Jamil, 2005; Swaminathan & Feng,1994). Under circumstances in which high throughput screeningis required for a large number of samples, the laborious and time-consuming culture-based techniques may not properly addresssuch a requirement.

The conventional methods have been improved to reduce costand labor and to offer faster detection and identification of Sal-monella. For example, chromogenic and fluorogenic growth media(e.g. SM-ID agar, Rambach agar, and BBL CHROMagar Salmonella)used for detection, enumeration, and identification directly on theisolation plate have shown to be convenient, reliable, and morespecific and selective than conventional media (Alakomi & Saarela,2009; Maciorowski et al., 2006; Manafi, 2000; Perry & Freydi�ere,2007). The test result using these selective media is typicallyavailable 1 day earlier than conventional methods but is not fastenough to respond to bioterrorism events, Salmonella outbreak, andproduct recall.

3. Rapid Salmonella detection methods

Several rapid methods using molecular cloning and recombi-nant DNA have been developed, validated, and are available on themarket. These accelerated Salmonella detection procedures enablequick answers, reduce storage space of conventional supplies andsamples, and increase the throughput of samples (Alakomi &Saarela, 2009; Blivet, Soumet, Ermel, & Colin, 1998; Tietjen &

Table 1A list of Salmonella detection methods commercially available for detection of Sal-monella spp.

Method and Formata Assay Manufacturer

MediaSelective culture BBLTM CHROMagar™ BD Diag.

EF-18 Agar (ISO-GRIDMethod)

Neogen

HGMF/EF-18 QA Life Sci.OSRT (Oxoid SalmonellaRapid Test)

Oxoid

Oxoid SRT OxoidRambach agar TechnogramRAPID' Salmonella Bio-Rad Lab.S.P.R.I.N.T. Salmonella OxoidSalmonella Contamination

Sciences LLCSalmonella Precis OxoidSalmonella Rapid Test OxoidSESAME SalmonellaTest

Biokar Diag.

Simple Mehtod forSalmonella (SMS)

AES Chemunex

Suncoli Test Paper Sun Chem.Immunology-basedELISA Assurance GDS™ for

SalmonellaBioControl

BacTrace KPLSalmonella ELISA TestSELECTA/OPTIMA

Bioline

CHEKIT Bommeli Diag.Colortrix Matrix

\MicroscienceEIAFoss FOSSEquate BinaxFlockChek IDEXXImmuno-magnetic-separation

Golden

Locate Rhone-PoulencMASTAZYME MASTMicroELISA Dynatech Lab.Rapidyme Salmonella BIO ART SA/NVRidascreen Salmonella R-BiopharmSalmonella GEMSalmonella Dioline/Mast Diag.Salmonella-TEK Organon TeknikaSalmotype Labor Diag LeipzigTAG 24 Salmonella BiocontrolTecra® SalmonellaVIA™

Tecra

TRANSIA Salmonella Diffchamb ABVIDAS bioM�erieux

LA ANI Salmonella ANI BiotechBactigen Wampole Lab.DuPont Lateral FlowSystem

DuPont Qualicon

Microscreen MicrogenRapidTest UnipathRapidChek SELECTSalmonella

SDIX

Salmonella Seiken Denka SeikenSalmonella Verify VICAMSerobact REMELSinglepath Salmonella EMDSalmonella Latex test OxoidSalmonella Screen/Salmonella Verify

Vicam

Slidex bioM�erieuxSpectate May & BakerVIP® Salmonella BioControlWellcolex Wellcolex

Immunochromatography Salmonella 1-2 SE BioControlClearview UnipathPATH-STIK CelsisReveal® for Salmonella NeogenSinglepath MerckSMART-II New Horizon

Table 1 (continued )

Method and Formata Assay Manufacturer

Tecra® SalmonellaUNIQUE™

Tecra

Biosensor DIA/PRO UMEDIKRBD3000 AATIDETEX Moluecular

CircuitryIMS Dynabeads® anti-

SalmonellaDynal

M1M analyser BioVerisPATHATRIX Matrix

MicroScienceSalmonella Screen VICAM

Agglutination TUBEX IDL BiotechBlot CheckPoint Colony Lift

Immunoassay KitKirkegaard &Perry

Dot-ELISA TyphiDot™ MoodyECL PATHIGEN BioVerisImmunodiffusion Salmonella 1-2 Test BioControlLateral flow Multi-Test Dip-S-Ticks PANBIO INDXTube agglutination Sanofi qualitative

agglutinationBio-Rad

Nucleic acid-basedPCR ABI Prism 7500 Applied

BiosystemsADIAFOOD Salmonella AES ChemunexAssurance GDSSalmonella

BiocontrolSystems

BAX Salmonella PCR DuPontQualicon

AmpliSensor BiotronicsFoodproof Salmonelladetection kit

Merck KGaA

foodproof® SalmonellaDetection Kit

BIOTECONDiagnostics

GeneDisc Salmonellaspp.

PallGeneSystems

GeneSTAR GullLaboratories

Genevision WarnexGene-Trak NeogenHQS Salmonella ADNucleisiQ-Check™ PCR BioRad

LaboratoriesLightCycler Roche

DiagnosticsLuminex Luminex

CorporationMicroSEQ Salmonellaspp

AppliedBiosystems

Probelia SanofiDiagnosticsPasteur

SmartCycler CepheidTagMan PE-Applied

BiosystemsDNA hybridization DNAH GeneTrak

GeneQuence™ NeogenCorporation

Gene-Trak® NeogenCorporation

RABIT DonWhite SciMiniaturized testsBiochemical API 20E bioM�erieux sa

Enterobacteriaceae Set II BBLEnterotube II Roche Diag.EPS (Enteric PathogensScreen Card)

Vitek

GNI (Gram NegativeIdentification Card)

Vitek

MICRO-ID Organon TeknikaMicroscan Baxter Diag.Mnitek I1 BBLQuad Enteric Panel Micro-MediaQuantum I1 Abbott Lab.

(continued on next page)

K.-M. Lee et al. / Food Control 47 (2015) 264e276 267

Table 1 (continued )

Method and Formata Assay Manufacturer

Sensititre SensititreTri Panel DifcoPasco Lab.Vitek 2 bioM�erieux

a ELISA, enzyme-linked immunoabsorbent assay; LA, latex agglutination; IMS,immunomagnetic separation; ECL, electrochemiluminescence, PCR, Polymerasechain reaction.

K.-M. Lee et al. / Food Control 47 (2015) 264e276268

Fung, 1995). The rapid method may be defined as one which allowsthe detection of Salmonella spp. in samples and delivers reliableresults within a few hours to a day (Ferretti, Mannazzu, Cocolin,Comi, & Clementi, 2001; Tietjen & Fung, 1995).

Commercially available rapid methods for Salmonella detectioncan be divided into several categories including new selectivemedia, modified or adapted conventional procedures,immunology-based assays, and nucleic acid-based assays (Alakomi& Saarela, 2009; Blivet et al., 1998; Eijkelkamp, Aarts, & van derFels-Klerx, 2009; Iqbal et al., 2000; Tietjen & Fung, 1995)(Table 1). Of these methods, ELISA and PCR procedures showcomparable specificity and sensitivity to conventional methods.ELISA assays are able to detect Salmonella concentration at the levelof 104e105 ml�1 while PCR-based assays provide the level ofsensitivity of 104 ml�1 after enrichment. The sensitivity and spec-ificity of these methods largely depend on the background micro-flora, sample matrix, presence of non-culturable cells, andinhibitory substances (e.g. fats, proteins, polysaccharides, heavymetals, antibiotics, and organic compounds) (Alakomi & Saarela,2009; Mozola, 2006; Naravaneni & Jamil, 2005). It has been re-ported that the sensitivity and detection limits of themethodswereimproved by additional sample preparation and purification tech-niques such as modification of pre-enrichment and enrichmentmedia, dilution, centrifugation, filtration, flow injection, chroma-tography, organic solvent extract, and fluorescence hybridization(Kutter, Hartmann, & Schmid, 2006; Mozola, 2006; Polaczyk et al.,2008; Wolffs, Glencross, Thibaudeau, & Griffiths, 2006).

3.1. Immunology-based assays

Immunology-based assays which employ specific mono- orpolyclonal antibodies binding with somatic or flagella antigenshave been broadly used for detection of Salmonella spp. in a varietyof sample matrices (Betts, 2002; Blivet et al., 1998; Iqbal et al.,2000; Maciorowski et al., 2006). The immunology-based assaysinclude enzyme-linked immunoabsorbent assay (ELISA), latexagglutination tests, immunodiffusion, and immunochromatog-raphy (dipstick). As routine analysis techniques, these assaysremain a valuable option due to the ability to detect viable non-culturable Salmonella cells (Maciorowski et al., 2006). The previ-ous studies showed that the immunology-based assays are com-parable to and more rapid and specific than cultural methods,feasible to isolate Salmonella spp. in conjunction with immuno-magnetic separation (IMS) techniques to handle large samples, andreadily automated to reduce time and labor (Keith, 1997; Tietjen &Fung, 1995; Uyttendaele, Vanwildemeersch, & Debevere, 2003).However, these methods have some potential shortcomings fordetection of Salmonella including a longer enrichment time toobtain the appropriate number of cells, cross-reactions with closelyrelated antigens, antigen variation, limits in sensitivity for somesample matrices and stressed cells, and high cost for automation(Blivet et al., 1998; Bolton, Fritz, & Poynton, 2000; Dill, Stanker, &Young, 1999; Uyttendaele et al., 2003).

3.1.1. Enzyme-linked immunoabsorbent assay (ELISA)Of immunology-based assays, ELISA is the most commonly used

assay for the detection of antigens or products of Salmonella spp.The different ELISA systems have been developed and commer-cially available in kit form. In the ELISA assay, an antigen specific toSalmonella spp. is bound to the appropriate antibody linked to asolid matrix. After forming the antigeneantibody complex, theconcentration of the antigen and the presence of Salmonella can bemeasured through the change in color caused by the enzymaticcleavage of a chromogenic substrate (Blivet et al., 1998; Tietjen &Fung, 1995). Alternatively, the presence of antibodies in samplesinfected with Salmonella spp. can be detected using antigenscoupled to the solid phase of ELISA (Wiuff et al., 2000). ELISAs havealso been used to detect antibodies for development of vaccinesagainst Salmonella infections (Meenakshi et al., 1999). Severalcommercial validated ELISA systems designed for rapid detection ofSalmonella in raw or processed products are available: AssuranceGDS™ for Salmonella (BioControl Systems, Inc., Bellevue, WA),TECRA Salmonella (Tecra International Pty Ltd, French Forest, NewSouth Wales, Australia), Salmonella ELISA Test SELECTA/OPTIMA(Bioline APS, Denmark), and Vitek Immuno Diagnostic Assay Sys-tem (VIDAS) (BioMerieus, Hazelwood, MO) (Table 1). The aggluti-nation technique employs latex particles coated with antibodieswhich react with antigens on the surface of Salmonella cells to formvisible aggregates for identification of Salmonella positive samples(Thorns, McLaren, & Sojka, 1994; Tietjen & Fung, 1995). The assaysare specific, uncomplicated, and reliable so that generally, theyhave been used as a confirmatory analysis technique, ratherscreening test for Salmonella organisms (Eijkelkamp et al., 2009;Love & Sobsey, 2007). There are several commercial kits availableon the market (Table 1).

3.1.2. Latex agglutination assayThe agglutination technique employs latex particles coated with

antibodies which react with antigens on the surface of Salmonellacells to form visible aggregates for identification of Salmonellapositive samples (Thorns et al., 1994; Tietjen & Fung, 1995). Theassays are specific, uncomplicated, and reliable so that generally,they have been used as a confirmatory analysis technique, ratherscreening test for Salmonella organisms (Eijkelkamp et al., 2009;Love & Sobsey, 2007). There are several commercial kits availableon the market, including Spectate (May & Baker Diagnostics Ltd.,Glasgow, Scotland, UK), Wellcolex color Salmonella (Wellcolex,Merseyside, UK), Salmonella Latex test (Oxoid, Basingstoke, UK),Bactigen (Wampole Laboratories, Cranbury, NJ), and Slidex (Bio-M�erieux, Marcy L'Etoile, France).

3.1.3. Immunodiffusion assaysThe Salmonella 1-2 Test system (BioControl Systems, Inc.,

Bothell, WA) for detection of motile Salmonella is a commercial kitthat operates on the basis of immunodiffusion reaction (D'Aoust &Sewell, 1988; Flowers & Klatt, 1989; Nath, Neidert, & Randall, 1989;Tietjen & Fung, 1995). Before inoculation into the system unitwhich consists of two connected chambers, the sample is pre-enriched for 24 h. The enriched sample is inoculated to a tetrathi-onate brilliant green broth in the inoculation chamber. Salmonellathen moves out of the inoculation chamber into the mobilitychamber in which antibody has been added onto a distal surface ofa semisolid medium. Salmonella in the mobility chamber isimmobilized by forming an antigeneantibody complex. After in-cubation for 14 h, the readable three-dimensional immunodiffusionband is produced. Modifications in an enrichment step beforeinoculation and an increase of incubation time improved theeffectiveness of detection of Salmonella spp. (Nath et al., 1989). The

K.-M. Lee et al. / Food Control 47 (2015) 264e276 269

modified and original assays have shown to be equivalent with thereference culture method.

3.1.4. Immunochromatography (dipstick) assaysThe commercial Salmonella test kits, based on the dipstick

format, include the Tecra® Salmonella Unique™ test (Tecra Inter-national Pty Ltd, French Forest, New SouthWales, Australia) and thePATH-STIK (Celsis, Inc., Edison, NJ). The Tecra Salmonella Unique™test which consists of the immunoenrichment and detection stepsis a screening procedure for Salmonella detection. The PATH-STIKSalmonella test is an immunological dipstick assay for one-stepdetection of Salmonella. The dipstick contains an assay controland all reagents used for Salmonella detection. Prior to the dipstickassay, a sample is pre-enriched and selectively enriched in media.The enriched sample is applied to the test unit and Salmonella in thesample is captured onto the dipstick. Unlike the Tecra® SalmonellaUnique™ test, the PATH-STIK doesn't include a washing step,requiring only 30 min for analysis (van Beurden, 1992; Brinkman,van Beurden, Mackintosh, & Beumer, 1995).

3.2. Nucleic acid-based assays

The nucleic acid-based assays are Salmonella detection tests thatutilize a specific nucleic acid target sequence within the organism.The assays have been most intensively explored and developed forthe past decade among Salmonella detection methods because theyoffer some advantages of sensitivity, specificity, and inclusivity overother methods, rapidly identifying Salmonella without obtainingpure cultures (Glynn et al., 2006; Mozola, 2006). Two major tech-niques of the assays are direct hybridization (DNA probe) andamplification (PCR) methods. The great progress of the assays al-lows the detection of very low numbers of organisms in the sampleand high throughput of a large number of samples for routineanalysis (Cohen et al., 1993; Mozola, 2006; Rasmussen, Rasmussen,Larsen, Hoff-Jorgensen, & Cano, 1994).

3.2.1. Polymerase chain reaction (PCR)The PCR assay is the preferred non-cultural technique for in vitro

primer-meditated enzymatic amplification of specific segments ofDNA for the detection of Salmonella pathogens (Cocolin, Manzano,Cantoni,& Comi, 1998; McKillip & Drake, 2004; Wolcott, 1991). ThePCR assay can exponentially amplify a single specific sequence toone million-fold of DNA fragment within 2 or 3 h using a ther-mostable DNA polymerase. The amplified products can be detectedby several means (e.g. gel-based systems and real-time PCR)(Eijkelkamp et al., 2009; Mozola, 2006). The development andadvancement of the technique for amplifying specific segment ofDNA improve the specificity and sensitivity enough for detectingeven one molecule of target DNA in a defined sample. Due to thecapability to detect such a low concentration of Salmonella,enrichment times are considerably shorter to reach Salmonellaconcentration needed for reliable detection by the PCR compared toother assays. The equivalency was reported between the PCR assayand the standard culture methods in detecting Salmonella in foodproducts (Aabo, Andersen, & Olsen, 1995; Cano, Rasmussen,Sanchez Fraga, & Palomares, 1993; Eyigor, Carli, & Unal, 2002;Stone, Oberst, Hays, McVey, & Chengappa, 1994). The PCR assayshave been developed, validated, and standardized according to thegeneral guidelines and requirements by ISO (IOS 20838:2006, ISO/DIS 22119, ISO 16140:2003, ISO 6579:2002, ISO 22174).

Commercial kits based on the PCR technology have beendeveloped, and successfully used for routine Salmonella screeningin the industry and emergency response laboratories. The PCR-based commercial kits and systems with different specificity andsensitivity include ABI Prism 7500 (Applied Biosystems,

Warrington, UK), Probelia (Sanofi -Diagnostics Pasteur, Marnes-la-Coquette, France), BAX system (DuPont Qualicon, Wilmington, DE),TagMan (PE-Applied Biosystems, Foster City, CA), Gene-Trak (Neo-gen Corporation, Lansing, MI), iQ-Check™ PCR (BioRad Labora-tories, Hercules, CA), LightCycler (Roche Diagnostics, Manheim,Germany), and SmartCycler (Cepheid Inc., Sunnyville, CA). The BAXsystem is the first commercially available PCR method using asingle tablet combining all reagents (primers, enzyme, and de-oxyribonucleotides) needed for PCR to rapidly amplify and detectSalmonella spp. (Bennett,., Greenwood, Tennant, Banks, & Betts,1998; Hoorfar et al., 1999; Maciorowski et al., 2006).

The real-time PCR (RT-PCR) system detects accumulated PCRproduct by monitoring the increase fluorescence signal using theintegrated real-time thermocycler and fluorescence detectorwithin a system. This helped overcome the problem of false positiveresults posed by amplicon contamination (Maurer, 2011). The falsepositive results can be further excluded by an internal standardused for ensuring the absence of inhibitors. The RT-PCR systememploying automated DNA extraction, amplification, and detectionhas improved the ability of industry and emergency response lab-oratories in identifying Salmonella from a variety of samplematrices. The RT-PCR methods available on the market can beclassified into several categories based on the principle andapproach of the methods. Several fluorescent techniques have beendeveloped and incorporated into the RT-PCR systems for simplicityand cost saving in detecting the target Salmonella sequences.

3.2.2. DNA probe hybridization assayA DNA probe hybridization assay uses a DNA probe with a

sequence complementary to the target sequence of a DNA or RNAmolecule in the target organism (de Boer & Beumer, 1999; Fung,2002; Mozola, 2006). In this assay, the target cells are first lysedand the produced nucleic acids are purified prior to hybridizationwith a DNA probe. The unbound probe to a target sequence iseliminated by washing. The stable hybrid with the labeled DNA canbe detected using different detection techniques, such as radio-isotopes and enzymatic reactions. The presumptive Salmonella re-sults are typically confirmed by the culture methods.

The DNA probe hybridization assays have some advantages inspecificity and sensitivity of identifying Salmonella in samplescompared toothermethods. Theassays areeffective to rapidly screenthe large numbers of samples and highly specific to detect the targetcells without cross-reactions with the other organism (Agron et al.,2001; de Boer & Beumer, 1999; Dunbar & Jacobson, 2007; Mozola,2006). This may allow more rapid distribution of Salmonella-freeproducts and more frequent monitoring at important investigationsites. However, the required sensitivity may be achieved only in thepresence of sufficient concentration of target organisms after pre-enrichment (Jones, Law, & Bej, 1993). The amplification of targetnucleic acid sequence andDNA probemight be an alternativeway toincrease the sensitivity of the assays (Tietjen & Fung, 1995).

The Gene-Trak® Salmonella Assay (Neogen Corporation, Lansing,MI) is the first introduced commercial assay. The second generationof this assay is based on a hybridization format of two probes, apolyadenylic acid (poly dA) tail on the capture probe and a fluo-rescein labeled detector probe, to the Salmonella ribosomal RNA(rRNA) and enzyme-mediated colorimetric detection. The Gene-Trak® Salmonella Assay has been modified to the direct-labeledformat which replaces a fluorescein labeled detector probe withHRP. This assay is equivalent to the original dipstick format assayand greatly simplifies the procedure by eliminating the enzymeincubation and washing steps and by lowering a hybridizationtemperature (Eijkelkamp et al., 2009; Mozola, 2006). Themicrowellformat assay was also introduced and available on the market un-der the name GeneQuence™ Salmonella.

K.-M. Lee et al. / Food Control 47 (2015) 264e276270

The different colorimetric DNA hybridization assays available inthe market have less than 48 h of a test time frame. Since the assaysrequire higher detection limits of enrichment culture for positivesignal, they need a longer enrichment time to increase the con-centration of Salmonella in the final culture. However, the total testtime with certain samples can be shortened to a significant extent.GeneQuence® Salmonella showed a sensitivity of 97.1% using a new2-stage procedures including 24e27 h pre-enrichment and selec-tive enrichment followed by a 2 h testing time in selected foods(Alles, Peng, Wendorf, & Mozola, 2007). The assays based on suchrRNA hybridization approach have certain benefits in designing aprobe because rRNA sequences are highly conserved, stable, andabundant in a bacteria cell, providing various targets and increasingthe assay sensitivity (Mozola, Peng, & Wendorf, 2007).

3.3. Miniaturized biochemical assay

Miniaturized biochemical assay employs the reduced volumesof reagents, media, and vessels for rapid characterization of Sal-monella to replace conventional biochemical tests and to providemore results in the same span of time (Fung, 2002; Ge & Meng,2009; Tietjen & Fung, 1995). Many microbiologists have devel-oped and used diverse miniaturized biochemical systems to rapidlyconfirm isolated organisms from a larger number of samples. Mostminiaturized biochemical systems consist of disposable sterilizedmicrotiter plates, a multiple inoculation device, and 10e20 mediaor substrates specially designed to target organisms. Salmonellaspp. are identified based on the assimilation and utilization ofspecial substrate and automated colony and cell density measure-ment. The test procedures include placing pure cultures into wellsof a microtiter plate each of which represents independent inocu-lation in the conventional method, holding up to 96 different cul-tures. After 16e24 h incubation at a desirable temperature, thepresence or absence of Salmonella on solid media or liquid mediacan be determined by reading color changes caused by the reactionof specific compositions andmetabolites with reagents with the aidof a manual identification code or an automatic reader interfacedwith a computer.

Diverse miniaturized kits for rapid biochemical characteriza-tion of Salmonella are commercially available, including API 20E(bioM�erieux sa, Marcy-I'Etoile, France), Enterotube II (BD Di-agnostics, Sparks, MD), Enterobacteriaceae Set II (previouslyknown as Mnitek II, BBL Microbiology Systems, Cockeysville, MD),MICRO-ID (Remel, Lenexa, KS), EPS (Enteric Pathogens ScreenCard) (Vitek Systems, Hazelwood, MO), GNI (Gram NegativeIdentification Card) (Vitek Systems, Hazelwood, MO), Microscan(Baxter Diagnostics, West Sacramento, CA), Quad Enteric Panel(Micro-Media Systems, San Jose, CA), Quantum II (Abbott Labo-ratories, Diagnostic Division, Abbott Park, IL), Sensititre (Sensititre,Salem, NH), Tri Panel (DifcoPasco Laboratories, Wheat Ridge, CA),Vitek 2 (bioM�erieux sa, Marcy-I'Etoile, France), and Walk Away(Dade MicroScan, West Sacramento, CA). The majority of thesecommercial kits were first used for comparative analyses of clin-ical samples and later applied for food microbiological samples(Fung, 2002; Stager & Davis, 1992). These systems use similarapproaches and procedures ideal for analyzing large number ofsamples and isolates. The sensitivity and specificity of these sys-tems, particularly the latest ones, easily exceed 95% with thehigher sample throughput, but the analysis cost may significantlyincrease if a sophisticated automated identification system is used.Most miniaturized biochemical assay kits were initially designedfor identification of gram-negative bacteria including Salmonella,Shigella, and Enterobacteriaceae, and later expanded to grampositive bacteria, anaerobes, and even yeasts and molds (Gracias &McKillip, 2004).

The previous studies reported that miniaturized biochemicalsystems are accurate, efficient, convenient, labor-saving, space-saving, and more economical compared to the conventional culturemethods (Cox, Fung, Bailey, Hartman, & Vasavada, 1987; Feng,1997; Holmes, 1989; Kalamaki, Prioce, & Fung, 1997; Stager &Davis, 1992). Except for the MICRO-ID and APE 20E systemrequiring 4 h of incubation, the results are typically obtained after24 h of incubation at an appropriate temperature in other systems.These miniaturized biochemical systems and kits have significantlycontributed to improve accuracy, efficiency, and capacity indetecting Salmonella spp. It is anticipated that in the future theywill continue to play an important role in the food, industrial, orenvironmental microbiology areas.

3.4. Biosensors

Biosensor technology encompasses an applied microbiologyapproach to bacterial detection (Alonso-Lomillo, Domínguez-Renedo, & Arcos-Martínez, 2010; Baeumner, 2003; Fung, 2002;Lazcka, Campo, & Mu�noz, 2007) that defines a molecule or agroup of molecules by incorporating materials immobilized on thesurface of a physicochemical transducer or transducing micro-system. A recognition signal is generated when a specific analytebinds to the biological recognition element. The signal can bechanges in mass, oxygen consumption, potential difference,refractive index, pH, current, and other parameters (Baeumner,2003; Eijkelkamp et al., 2009). It is anticipated that a biosensortechnique may replace existing immunology- and nucleic acid-based assays (Alocilja & Radke, 2003; Lazcka et al., 2007; VanDorst et al., 2010).

Biological recognition elements used in biosensor applicationinclude enzymes, antibodies, nucleic acids, whole cells, tissue/whole organisms, and biomimetic material. The signal recognitionand transduction in the biosensor is achieved by different types oftransducers: electrochemical, optical, thermometric, micro-mechanical, and other miscellaneous transducers according towhich the biosensors have been often classified by several authors(Goepel & Heiduschka, 1995; Ivnitski, Abdel-Hamid, Atanasov,Wilkins, & Stricker, 2000; Leonard et al., 2003).

3.4.1. Enzyme biosensorAn enzyme biosensor is an analytical device that integrates an

enzyme with a transducer to produce a signal resulting fromdifferent physicochemical changes caused by enzyme-catalyzedreactions. The signal is converted into measurable responses suchas current, temperature change, and light absorption to determinetarget analyte concentration (Baeumner, 2003). Enzyme biosensorsare highly selective, rapid, and easy to use while they are expensiveand sometimes lose their activities when immobilized on a trans-ducer. Enzymes used in this type of biosensor include glucose ox-idase, galactosidase, glucoamlyase, lactate oxidase, and otherenzymes (Fung, 2002).

3.4.2. Nucleic acid biosensorA nucleic acid biosensor which consists of a nucleic acid as a

biological recognition element and a signal transducer is based onthe hybridization of complementary strands of DNA or RNA mole-cules for detection of organisms (Baeumner, 2003; Eijkelkampet al., 2009; Lazcka et al., 2007). In addition to detection of bacte-ria and other pathogens, nucleic acid biosensors have been appliedto food and environmental samples for detection of toxic com-pounds and biotechnology products using different signal trans-duction systems such as quartz crystal microbalances, surfaceplasmon resonance, surface acoustic wave sensor, and evanescentwave biosensors (Deisingh & Thompson, 2004; Feriotto, Borgatti,

K.-M. Lee et al. / Food Control 47 (2015) 264e276 271

Mischiati, Bianchi, & Gambari, 2002; Mascini, Palchetti, &Marrazza, 2001).

3.4.3. Antibody-based biosensorsAntibody-based biosensors are analytical devices which use an

antibody as a biological recognition element attached to a signaltransducer to quantify the target analyte concentration (Bilitewski,2000; Rogers, 2000). Antibodies are the most commonly usedbiological recognition element due to their high specificity andaffinity to antigen, diversity, and availability on the market.Antibody-based biosensors can be divided into three main assayformats including an immobilized antibody, immobilized antigen,and non-immobilized antibody or antigen whose characteristicsare largely determined by the type of signal transduction mecha-nisms (Rogers, 2000).

3.4.4. Micro- and nano-biosensorsThe field of micro- and nano-biosensors has been drawingmuch

attention from microbiologist due to their great potential to solvemany current problems including real time detection of a variety ofpathogens, and believed as one of the future directions to detectSalmonella spp. (Chen et al., 2005; Mozola, 2006; Scaria et al.,2008). However, actual application to Salmonella detection israther limited so far because of the instrumental complexity andimmature stage of the technology development (Alakomi &Saarela, 2009). However, the techniques are rather restricted inapplication to pathogen detection because of their incompatibilitywith some biological samples. Despite some obstacles and chal-lenges in the research and application of the micro- and nano-biosensors, rapid development and integration of associated coretechnologies seems to provide no doubt that in the future theywould be one of important technologies in microbiology and have agreat value in the market.

4. Salmonella detection methods for food emergencyresponse laboratories

Salmonella detection is critical for emergency responders whorequire robust and preferably transportable laboratory-baseddetection systems. The methods for Salmonella detection typicallyrecommended by the FDA and the FSIS of the USDA are based oncultural, serological, and biochemical properties using selectivemedia. However, rapid methods for Salmonella detection havebecome increasingly important for effectively monitoring foodproducts and are considered desirable as a future approach andnational strategy. Rapid test kits recommended by the US FoodEmergency Response Network for Salmonella include the VIDASSalmonella (SLM) and immuno-concentration Salmonella (ICS)methods, the Tecra Unique Salmonella test, and the polymerasechain reaction (PCR)-based BAX™ system, which were chosenwiththe expectation of providing results within 24 h. These test kits areapproved for screening purposes by the AOAC and have beenwidely used by food testing laboratories.

During the past four fiscal years, food emergency responselaboratories have performed single lab and/or multi-lab validationsinvolving rapid Salmonella detection platforms. Several studieshave focused on the recovery of Salmonella spp. from a variety ofsample matrices spiked with Salmonella strains. According to theresults of these studies, detection levels varied with sample matrixand spiking concentration. For example, when spiked at the samelow concentration Salmonella was easily identified in cat food andvegetable burgers, while its detection in other matrices such asbaby oatmeal, peanut butter, liquid eggs, and cantaloupe was moredifficult. Several food matrices, including milk, strawberries, andcreamed corn, displayed an inverse relationship between

detectability and concentration. Furthermore, wild-type Salmonellaenteriditis showed much lower recoveries from liquid egg productsrelative to the American Type Culture Collection (ATCC) strains.

In comparison studies of recovery of Salmonella from culturemedia, Yersinia pestis enrichment (YPE) broth and buffered peptonewater (BPW) were as effective as lactose broth, a standard enrich-ment medium, in some sample matrices. Overall, xylose lysinedeoxycholate (XLD) agar exhibited a higher recovery of Salmonellaspp. than Hektoen enteric (HE) and bismuth sulfite (BS) agars,which may be explained by the selectivity of the media and thephysiological properties of the strains used in the studies. Samplepreparation techniques also appeared to affect the recovery of thespiked samples. For example, results obtained with the VIDAS® SLMand BAX® Q7 Salmonella PCR assays indicated that washing withsoaking and soaking alone were more effective for extraction ofSalmonella than surface washing only.

Comparative and collaborative studies of the rapid methods forscreening of Salmonella have been performed by many foodemergency response laboratories by using a variety of spikedsample matrices. However, as seen in Table 2, only limited numbersof methods and platforms were extensively evaluated andcompared within integrated laboratory networks during this spe-cific period. In the validation studies, a suspension of Salmonellacells at the appropriate turbidity was serially diluted for spikinginto samples prior to assessing the test kits. Enrichment using BPWor YPE broth for the PCR-based methods offered a rapid alternativeto the standard culture method for identifying Salmonella spp. Ofthe PCR methods tested, the BAX™ system showed comparablesensitivity and specificity to conventional culture methods in mostsample matrices spiked at low and high concentrations of Salmo-nella cells, with the exception of some false results due to cross-contamination of PCR products. No differences in identifying Sal-monella spp. were found between the standard BAX™ system andthe BAX™ Q7 system.

Overall, there was little or no difference between the rapidmethods, which displayed equally high accuracies in identificationof the target organisms regardless of the matrix. Comparable re-sults were found between BAX and SmartCycler PCR platforms insingle laboratory testing. Some laboratories reported that the per-formance and accuracy of the SmartCycler PCR system was some-what dependent on the characteristics of the sample matrix. It wasalso reported that the sensitivity of the systems was furtherimproved by increasing enrichment time. Taken together, theseobservations seem to imply that the difference between thesemethods may be mainly due to the matrix properties and thepossible presence of PCR inhibitors, rather than the inherenteffectiveness of the assay systems.

Several laboratories have compared the BAX™ Q7 PCR andVIDAS Salmonella (SLM) assays. The results showed no significantdifference between the two assays in detecting Salmonella spp. inmost food matrices. Indeed, most failures of the two assays inSalmonella detection appeared to be related to the matrix effect.Several laboratories have indicated that the VIDAS assay might bemore reliable than the BAX system based on the study results andon the discussions moderated through a web-based database por-tal. However, the VIDAS assay was insufficiently sensitive forcertain matrices (e.g., bean sprouts), and requires a secondaryenrichment procedure to reduce the effects of background flora.Furthermore, there is only limited information available on how theBAX™ and VIDAS systems interact with matrices.

Several immunomagnetic separation (IMS) systems such as theDynal BeadRetriever, Pathatrix, and BioVeris M1M Platform werevalidated and assessed for their efficiency in recovering andisolating different strains of Salmonella from sample matrices(Table 2). Typically, after an overnight incubation, a sample aliquot

Table 2Salmonella detection methods for screening of Salmonella and sample matrices usedfor the studies in food emergency response laboratories during the past fiscal years.

Methoda Sample matrix

PCR assaysABI 7500 Fast Cheese, cookie, cookie, ground beef, ice

cream, liquid egg, peanut butter, potatosalad, cracker, snack bar

BAX® PCR system Baby carrot, beef stick, black pepper,cilantro, cilantro flakes-dried, cookedham, cooked ham, deli ham, groundbeef, ground beef, hamburger, hotpepper, hot dog, infant formula,jalapeno pepper, liquid egg, milk,orange juice, peanut butter, potatosalad, powdered formula, precut ready-to-eat (RTE) lettuce, serrano pepper,spinach, sponge sicle, tomato, vegetablejuice, whole head romaine lettuce

LightCycler Alfalfa sprout, baby oatmeal, broccoli,cantaloupe, cat food, chicken, creamedcorn, frozen spinach, ground beef, hotdog, jalapeno pepper, lettuce, liquidegg, milk, orange juice, peanut butter,strawberry, tomato, vegetable burger,watermelon

SmartCycler Baby carrot, cheese, chicken breast delimeat, cookie, deli ham, ground beef, icecream, liquid egg, milk, orange juice,peanut butter, potato salad, raisin, rawpork sausage, cracker, snack bar,spinach, vegetable juice

Luminex Alfalfa sprout, chocolate, bean sproutELISA assayVIDAS Cilantro, cilantro flakes-dried, jalapeno

pepper, precut ready-to-eat (RTE)lettuce, serrano pepper, spinach, wholehead romaine lettuce

IMS assaysDynal BeadRetriever Chicken breast deli meat, cooked ham,

ground beef, hamburger, liquid egg,milk, orange juice, peanut butter, potatosalad, raisin, raw pork sausage, spinach

M1M analyzer Alfalfa sprout, baby oatmeal, broccoli,cantaloupe, cat food, chicken, creamedcorn, frozen spinach, ground beef, hotdog, jalapeno pepper, lettuce, liquidegg, milk, orange juice, peanut butter,strawberry, tomato, vegetable burger,watermelon

PATHATRIX Hot pepper, peanut butter, tomatoMiniaturized biochemical assaysAPI 20E Chicken breast deli meat, raisin, raw

pork sausageEnterotubes Liquid egg, milk, orange juice, peanut

butter, spinachVitek® Black pepper, ground beef, peanut

butter, potato salad, powdered formulaMediaCHROMagar Ground beef, liquid egg, potato salad

a PCR, Polymerase chain reaction; ELISA, enzyme-linked immunoabsorbent assay;IMS, immunomagnetic separation.

K.-M. Lee et al. / Food Control 47 (2015) 264e276272

was subjected to immunomagnetic separation and the two bacte-rial fractions (immuno-bound and unbound) were plated ontoagarifiedmedium and/or processed using the Roche LightCycler 2.0RT-PCR, BAX System Q7 PCR, or Applied Biosystems 7500. Irre-spective of manufacturer and concentration of Salmonella, the IMSsystems were generally effective for improving the recovery fromsample matrices and allowing for shorter turnaround times.However, the recovery from some food matrices (e.g., bean sprouts,cooked ham, ground beef, and peanut butter) was identical with orwithout the use of an IMS system at all spiking levels, indicating

that the system efficiency is highly dependent on sample matrix. Inaddition, the IMS systems appeared to be not fully functional whenthe complexity of the sample matrices increased. Compared tomanual methods, the automated IMS systems further increased thespeed and productivity, and minimized cross-contamination bysignificantly reducing sample transfer time and pipetting steps.

Apart from the rapid methods, CHROMagar™ Salmonellaappeared to be useful for colony isolation, in particular, in highlycontaminated matrices. This medium could conceivably replaceMacConkey agar as the growth agar of choice because it was easierto read and more quickly differentiated Salmonella from other or-ganisms. Commercial biochemical assays used by food emergencyresponse laboratories for confirmation of typical Salmonella col-onies isolated on selective plating agars such as CHROMagar™include API 20E, Enterotubes, and Vitek II Compact. One potentialadvantage of the automated Vitek II Compact system is that it couldprovide higher sample throughput compared to the API 20E, whichwas limited to testing smaller sample volumes.

As stated above, diverse and rapid commercial assays for iden-tifying Salmonella in foods have been evaluated and validated byfood emergency response laboratories. However, most studies haveused artificially spiked samples, which may not truly reflect thephysiological state and acclimation of naturally contaminatedsamples. Previous studies have indicated that the currently avail-able rapid methods might produce a relatively high rate of falsenegative results for naturally contaminated samples. Although therapid methods require much shorter detection times compared toconventional methods, some of them still appear to suffer from lackof sensitivity and specificity for certain sample matrices. Addi-tionally, the PCR-based detection platforms require extensivesample preparation steps for some complex matrices and inter-pretation of the data can be relatively complicated. Finally, asimplied by the outcomes of the studies, participation of the testinglaboratories and their collaborative efforts is crucial to improve themethods and technologies needed to achieve the goals of inte-grated laboratory networks and to benefit the laboratories involvedin Salmonella investigations in future.

5. Selection of Salmonella detection methods for rapidemergency response

Salmonella detection methods to quickly identify Salmonellacontamination sources are vital for immediate actions of emer-gency responders. Considerable progress has been made to shortenanalysis time of the methods while maintaining or increasingsensitivity and specificity for detection of Salmonella. However,despite significant progress in development of a variety of assays todetect and control Salmonella, the routine use of the assays hasbeen limited due to some reasons. First of all, the assays have notbeen properly selected for sample matrices to be tested and thesampling and sample preparation have not been optimized for themethods and samples. Besides, the detection limits of the methodshave not been accurately considered in evaluating the test results.Another reason may be a long and inappropriate enrichment time.These limitations are believed to influence the efficacy of the assaysand further mislead the assessment of food safety.

Although the rapid methods are equally efficacious for theintended use as discussed in the present study, the testing labo-ratories should consider the following parameters to choose themost acceptable method, or a combination of themethods, for theirown needs or routine use regardless of the method's principle:accuracy (specificity and sensitivity), precision, detection limit,throughput, ease of use, the nature of samples, cost acceptability,laboratory space, training of laboratory personnel, and quality ofsale services (Blivet et al., 1998; Eijkelkamp et al., 2009; Ivnitski

K.-M. Lee et al. / Food Control 47 (2015) 264e276 273

et al., 2000; Mozola, 2006; Tietjen & Fung, 1995). Multiplexing tosimultaneously detect more than one target organism and theability to differentiate between naturally contaminated and artifi-cially spiked samples are also desirable parameters to be includedinto consideration in selection of the method. Currently, any singlemethodmay not fulfill all these requirements (Cannons et al., 2006;Mahon, Murphy, Jones, & Barrow, 1994; Settanni & Corsetti, 2007).The following parameters are particularly important to be carefullyconsidered in analytical methods for identification of Salmonella.

5.1. Matrix effect and sample preparation

The efficacy of a rapid method is significantly influenced bysample matrix (Mozola, 2006; Ricke et al., 1998; Ricci, Volpe,Micheli, & Palleschi, 2007; Stevens & Jaykus, 2004). Therefore, itis critical to verify if the method is fitting for the sample matrix tobe tested. Regardless of the detection methods, the sample prepa-ration is a critical and challenging step that should be optimized tobetter separate the target organisms from inhibitory substances ina sample, whichmay improve signal to noise ratios and lower false-positive results (Koyuncu, Andersson, & H€aggblom, 2010; Rickeet al., 1998; Saroj, Shashidhar, Karani, & Bandekar, 2008). Immu-nomagnetic separation and molecular imprints systems are excel-lent tools for sample concentration and purification. The rapidmethods that can be used for crude samples and require minimalsample preparation will be desirable, particularly for field and on-site usage.

5.2. Selectivity and specificity

The sensitivity and specificity are extensively examined duringthe method development and independent studies. However, careshould be given to compare sensitivity and specificity rates of themethods presented in the published literature and on websitesbecause the study design and sample matrices used are differentbetween the studies. It is desirable that both sensitivity and spec-ificity are as high as possible and can be used to determinewhethera confirmation test will be conducted or not. According to theprevious studies and the manufacturer's claim, the sensitivity andspecificity rates of some commercial assays are more than 99%(Eijkelkamp et al., 2009). However, such a high level of theseproperties may not be achievable bymost of commercially availablerapid assays. Even the validation bodies such as AOAC, AFNOR, andMicroVal do not specify acceptance criteria and values for suchproperties. In the meantime, it should point out that the detectioncapability of the rapid methods needs to be determined byconsidering the bioavailability of toxins and assessment of foodsafety, rather than by simply measuring concentrations of patho-gens by the methods because all Salmonella isolates are not equallypathogenic for humans.

5.3. Analysis time

In emergency response situations, the rapid methods should beable to give the results preferably within a few hours to a day andperform on-site (Skottrup, Nicolaisen, & Justesen, 2008). However,the commercially available rapid methods which produce the testresults in the next day are rather limited. The presumptive Salmo-nella results need to be confirmed by culture and biochemicalidentification methods, requiring at least additional 24 h and morefor completion. The use of microplates with the rapid methods mayoffer the advantage of reducing analysis time and increasing thecapacity of the methods when processing a large number ofsamples.

5.4. Cost efficiency

The laboratorians seek the rapid methods with lower opera-tional and maintenance costs with comparable accuracy to avoidexpense and time consuming methods. However, the public healthrisk and cost associated with false test results might have muchgreater negative impacts than the benefits from the use of a low-cost method (Eijkelkamp et al., 2009; Tietjen & Fung, 1995).

6. Future aspects of Salmonella detection methods

In order to replace conventional procedures and to meet thefood emergency and safety needs, the rapid methods should besimple, fast, precise, sensitive, specific, stable, versatile, and cost-effective. These requirements are in part associated withimprovement of sample preparation and reduced time and laborfor analysis in the methods. According to the literature and previ-ous studies by food emergency response laboratories on testing andvalidating the methods, none of commercially available methodsfor detection of Salmonella seem to meet all the requirements foradequate emergency response to Salmonella incidence.

More reliable and efficient new assays are needed to replaceexisting conventional methods and to meet the food emergencyand safety needs although considerable works should be donebefore routine use. In addition, the sensitivity and specificity of thenew assays should be more uniformly assessed under a well-defined study design.

The development and improvement of separation and concen-tration methods that efficiently isolate, concentrate, and purifytarget Salmonella spp. and simultaneously eliminate interferingcomponents directly from food samples is needed to further in-crease sensitivity and selectivity of the rapid detection methodsand provide real-time screening of the samples for Salmonella insignificantly less detection time than that required for culturalenrichments. The removal of the need for cultural enrichmentwould allow food emergency response laboratories to detect lowlevels of Salmonella contamination in a sample matrix. Furtherresearch needs to focus on the development of separation andconcentration methods to be simple, rapid, low-cost, and appli-cable to a variety of sample matrices and background microflora.

The development of accurate field and on-site systems may alsohelp improve the efficiency of food emergency response networks,reducing analysis time and thework load for reference laboratories.The progress of microarray and automation technologies willfurther increase the capacity of the food emergency response lab-oratories for screening Salmonella. Since a single Salmonella bac-terium can be rapidly multiplied into more than a million in a fewhours, the methods should be capable of cultivating extremely lowconcentration of the target organisms in samples. The assay whichenables to simultaneously detect multiple pathogens will provideadditional benefits and advantages to the laboratories. Perhapsmore importantly, it is needed to develop the schemes and pro-cedures to validate the efficiency of new assays in food emergencyresponse laboratories for coordinating of the results usingdependable and representative control samples.

7. Conclusions

Presently, many commercial detection methods are availableand vary in detection limit, sensitivity, specificity, and cost effi-ciency with sample matrices to be tested. Despite of great advancesin technologies, current Salmonella detection methods are not fastand reliable enough for emergency responders to determine thedesired directions within the appropriate period of time, beingrequired for further improvements and higher efficiency of the

K.-M. Lee et al. / Food Control 47 (2015) 264e276274

methods. Food emergency response laboratories should choose themost acceptable method or a combination of methods having thedesired analytical properties to meet food emergency and safetyneeds, based on their own needs and requirements. The presentstudy indicates that validation and comparative studies for the testkits should be set up under the identical test conditions to bettercompare and evaluate the test results among different food emer-gency response laboratories. A study on the effect of sampling andsample size on the recovery of the target organism might also helpobtain more accurate test results.

Due to increasing population mobility and the wide distributionof feed/food products, Salmonella outbreaks and salmonellosiscases are occurring across state and national boundaries. Suchwidely dispersed outbreaks and salmonellosis may limit the publichealth resources for detection and investigation of Salmonellaoutbreaks. During the past years, some integrated laboratory net-works have effectively met demands from public and emergencyresponders, organized training and the inter-laboratory compara-tive studies, and coordinated the test results of the participatinglaboratories through online reporting and data sharing system.With years of experience, they should efficiently integrate theemergency networks for food safety and harmonization of Salmo-nella detection methods to provide timely investigation of a widelydispersed multistate Salmonella outbreak and respond to potentialfuture events of bioterrorism.

Acknowledgments

The authors acknowledge that this study was supported andmade possible by funds from USDA-FSIS Microbiology CooperativeAgreement Program.

References

Aabo, S., Andersen, J. K., & Olsen, J. E. (1995). Research note: detection of Salmonellain minced meat by the polymerase chain reaction method. Letters in AppliedMicrobiology, 21, 180e182.

Agron, P. G., Walker, R. L., Kinde, H., Sawyer, S. J., Hayes, D. C., Wollard, J., et al.(2001). Identification by subtractive hybridization of sequences specific forSalmonella enterica serovar Enteritidis. Applied and Environmental Microbiology,67, 4984e4991.

Alakomi, H. L., & Saarela, M. (2009). Salmonella importance and current status ofdetection and surveillance methods. Quality Assurance and Safety of Crops &Foods, 1, 142e152.

Alles, S., Peng, X., Wendorf, M., & Mozola, M. (2007). Precollaborative study of theGeneQuence Salmonella assay using 24-hour enrichment protocols for detec-tion of Salmonella spp. in select foods. Journal of AOAC International, 90,725e737.

Alocilja, E. C., & Radke, S. M. (2003). Market analysis of biosensors for food safety.Biosensors and Bioelectronics, 18, 841e846.

Alonso-Lomillo, M. A., Domínguez-Renedo, O., & Arcos-Martínez, M. J. (2010).Screen-printed biosensors in microbiology; a review. Talanta, 82, 1629e1636.

European Commission (EC). (2003a). Regulation no. 2160/2003 of the EuropeanParliament and of the Council of 17 November 2003 on the control of Salmo-nella and other specified food-borne zoonotic agents. Official Journal of theEuropean Union, L325, 1e15.

European Commission (EC). (2003b). Directive 2003/99/EC of the EuropeanParliament and of the Council of 17 November 2003 on the monitoring ofzoonoses and zoonotic agents, amending Council Decision 90/424/EEC andrepealing Council Directive 92/117/EEC. Official Journal of the European Union,L325, 31e40.

Arroyo, G., & Arroyo, J. A. (1995). Selective action of inhibitors used in differentculture media on the competitive microflora of Salmonella. Journal of AppliedBacteriology, 78, 281e289.

Ashford, D. A., Kaiser, R. M., Bales, M. E., Shutt, K., Patrawalla, A., McShan, A., et al.(2003). Planning against biological terrorism: lessons from outbreak in-vestigations. Emerging Infectious Diseases, 9, 515e519.

Baeumner, A. J. (2003). Biosensors for environmental pollutants and food con-taminants. Analytical and Bioanalytical Chemistry, 377, 434e445.

Bennett, A. R., Greenwood, D., Tennant, C., Banks, J. G., & Betts, R. P. (1998). Rapidand definitive detection of Salmonella in foods by PCR. Letters in AppliedMicrobiology, 26, 437e441.

Betts, R. (2002). Detecting pathogens in food. In C. W. Blackburn, & P. J. McClure(Eds.), Foodborn pathogens, hazard, risk analysis and control (pp. 13e52). Cam-bridge: Woodhead.

van Beurden, R. (1992). PATH-STIK rapid Salmonella test e a new immunoassay todetect foodborne pathogens. Scandinavian Dairy Information (Sweden), 6,52e54.

Bilitewski, U. (November 2000). Can affinity sensors be used to detect food con-taminants? Analytical Chemistry, 692Ae701A.

Blivet, D., Soumet, C., Ermel, G., & Colin, P. (1998). Rapid detection methods forpathogens, 3rd Karlsruhe Nutrition Symposium: European research towardsSafer and better food. In V. Gaukel, & W. E. L. Spieß (Eds.), Food safety andmonitoring of safety aspects (pp. 3e9). Karlsruhe, Germany: Review and TransferCongress, Congress Centre.

de Boer, E., & Beumer, R. (1999). Methodology for detection and typing of food-borne micro-organisms. International Journal of Food Microbiology, 50, 119e130.

Bolton, F. J., Fritz, E., & Poynton, S. (2000). Rapid enzyme-linked immunoassay fordetection of Salmonella in food and feed products: performance testing pro-gram. Journal of AOAC International, 83, 299e303.

Brinkman, E., van Beurden, R., Mackintosh, R., & Beumer, R. (1995). Evaluation of anew Dip-Stick test for the rapid detection of Salmonella in food. Journal of FoodProtection, 58, 1023e1027.

Cannons, A., Amuso, P., & Anderson, B. (2006). Biotechnology and the public healthresponse to bioterrorism. In B. Anderson, H. Friedman, & M. Bendinelli (Eds.),Microorganisms and bioterrorism (pp. 1e13). New York: Springer.

Cano, R. J., Rasmussen, S. R., Sanchez Fraga, G., & Palomares, J. C. (1993). Fluorescentdetection-polymerase chain reaction (FD-PCR) assay on microwell plates as ascreening test for salmonellas in foods. Journal of Applied Bacteriology, 75,247e253.

Carrique-Mas, J. J., & Davies, R. H. (2008). Sampling and bacteriological detection ofSalmonella in poultry and poultry premises: a review. Revue Scientifique etTechnique (International Office of Epizootics), 27, 665e677.

Chalker, R. B., & Blaser, M. J. (1988). A review of human salmonellosis: III. Magnitudeof Salmonella infection in the United States. Review of Infectious Diseases, 10,111e124.

Chen, S., Zhao, S., McDermott, P. F., Schoeder, C. M., White, D. G., & Meng, J. (2005).A DNA microarray for identification of virulence and antimicrobial resistancegenes in Salmonella serovars and Escherichia coli. Molecular and Cellular Probes,19, 195e201.

Coburn, B., Grassl, G. A., & Finlay, B. B. (2007). Salmonella, the host and disease: abrief review. Immunology and Cell Biology, 85, 112e118.

Cocolin, L., Manzano, M., Cantoni, C., & Comi, G. (1998). Use of polymerase chainreaction and restriction enzyme analysis to directly detect and identify Sal-monella typhimurium in food. Journal of Applied Microbiology, 85, 673e677.

Cohen, N. D., Neibergs, H. L., McGruder, E. D., Whitford, H. W., Behle, R. W.,Ray, P. M., et al. (1993). Genus-specific detection of salmonellae using the po-lymerase chain reaction (PCR). Journal of Veterinary Diagnostic Investigation, 5,368e371.

Cox, N. A., Fung, D. Y. C., Bailey, J. S., Hartman, P. A., & Vasavada, P. C. (1987).Miniaturized kits, immunoassays and DNA hybridization for recognition andidentification of foodborne bacteria. Dairy and Food Sanitation, 7, 628e631.

Cummings, K. J., Warnick, L. D., Elton, M., Gr€ohn, Y. T., Patrick, L. M., & Siler, J. D.(2010). The effect of clinical outbreaks of Salmonellosis on the prevalence offecal Salmonella shedding among dairy cattle in New York. Foodborne Pathogensand Disease, 7, 815e823.

D'Aoust, J. Y., & Sewell, A. M. (1988). Reliability of the immunodiffusion 1-2 Test™system for detection of Salmonella in foods. Journal of Food Protection, 51,853e856.

Deisingh, A. K., & Thompson, M. (2004). Strategies for the detection of Escherichiacoli O157:H7 in foods. Journal of Applied Microbiology, 96, 419e429.

Dill, K., Stanker, L. H., & Young, C. R. (1999). Detection of Salmonella in poultry usinga silicon chip-based sensor. Journal of Biochemical and Biophysical Methods, 41,61e67.

Dunbar, S. A., & Jacobson, J. W. (2007). Quantitative, multiplexed detection of Sal-monella and other pathogens by Luminex® xMAP™ suspension array. InH. Schatten, & A. Eisenstark (Eds.), Salmonella methods and protocols (pp. 1e19).Totwa: Human Press Inc.

Eijkelkamp, J. M., Aarts, H. J. M., & van der Fels-Klerx, H. J. (2009). Suitability of rapiddetection methods for Salmonella in poultry slaughterhouses. Food AnalyticalMethods, 2, 1e13.

Eyigor, A., Carli, K. T., & Unal, C. B. (2002). Implementation of real-time PCR totetrathionate broth enrichment step of Salmonella detection in poultry. Lettersin Applied Microbiology, 34, 37e41.

Feldsine, P. T., Lienau, A. H., Leung, S. C., Mui, L. A., Humbert, F., Bohnert, M., et al.(2003). Detection of Salmonella in fresh cheese, poultry products, and dried eggproducts by the ISO 6579 Salmonella culture procedure and the AOAC officialmethod: collaborative study. Journal of AOAC International, 86, 275e295.

Feng, P. (1997). Impact of molecular biology on the detection of foodborne patho-gens. Molecular Biotechnology, 7, 267e278.

Feriotto, G., Borgatti, M., Mischiati, C., Bianchi, N., & Gambari, R. (2002). Biosensortechnology and surface plasmon resonance for real-time detection of geneti-cally modified roundup ready soybean gene sequences. Journal of Agriculturaland Food Chemistry, 50, 955e962.

Ferretti, R., Mannazzu, I., Cocolin, L., Comi, G., & Clementi, F. (2001). Twelve-HourPCR-based method for detection of Salmonella spp. in food. Applied and Envi-ronmental Microbiology, 67, 977e978.

K.-M. Lee et al. / Food Control 47 (2015) 264e276 275

Fierer, J., & Guiney, D. G. (2001). Diverse virulence traits underlying differentclinical outcomes of Salmonella infection. Journal of Clinical Investigation, 107,775e780.

Flowers, R. S., & Klatt, M. J. (1989). Immunodiffusion screening method for detectionof motile Salmonella in foods: collaborative study. Journal of AOAC International,72, 303e311.

Food and Drug Administration (FDA). (2009). Food code 2009 Accessed February2014 http://www.fda.gov/downloads/Food/GuidanceRegulation/UCM189448.pdf.

Frenzen, P., Riggs, T., Buzby, J., Breuer, T., Roberts, T., Voetsch, D., et al., , FoodNetWorking Group. (1999). Salmonella cost estimate update using FoodNet data.Food Review, 22, 10e15.

Fung, D. Y. C. (2002). Rapid methods and automation in microbiology. Compre-hensive Reviews in Food Science and Food Safety, 1, 3e14.

Ge, B., & Meng, J. (2009). Advanced technologies for pathogen and toxin detectionin foods: current applications and future directions. Journal of the Association forLaboratory Automation, 14, 235e241.

Glynn, B., Lahiff, S., Wernecke, M., Barry, T., Smith, T. J., & Maher, M. (2006). Currentand emerging molecular diagnostic technologies applicable to bacterial foodsafety. International Journal of Dairy Technology, 59, 126e139.

Goepel, W., & Heiduschka, P. (1995). Interface analysis in biosensor design. Bio-sensors and Bioelectronics, 10, 853e883.

Gracias, K. S., & McKillip, J. L. (2004). A review of conventional detection andenumeration methods for pathogenic bacteria in food. Canadian Journal ofMicrobiology, 50, 883e890.

Gruenewald, R., Henderson, W., & Yappow, S. (1991). Use of Rambach propyleneglycol containing agar for identification of Salmonella spp. Journal of ClinicalMicrobiology, 29, 2354e2356.

Guiney, D. G., Fang, F. C., Krause, M., Libby, S., Buchmeier, N. A., & Fierer, J. (1995).Biology and clinical significance of virulence plasmids in Salmonella serovars.Clinical Infectious Diseases, 21, 146e151.

Ha, S. D., Pillai, S. D., & Ricke, S. C. (1995). Growth response of Salmonella spp. tocycloheximide amendment in media. Journal of Rapid Methods & Automation inMicrobiology, 4, 77e83.

Holmes, B. (1989). Comparative evaluation of the Roche Cobas IDA and Enterotube IIsystems for identifying members of the family Enterobacteriaceae. Journal ofClinical Microbiology, 27, 1027e1030.

Hoorfar, J., Baggesen, D. L., & Porting, P. H. (1999). A PCR-based strategy for simpleand rapid identification of rough presumptive Salmonella isolates. Journal ofMicrobiological Methods, 35, 77e84.

Iqbal, S. S., Mayo, M. W., Bruno, J. G., Bronk, B. V., Batt, C. A., & Chambers, J. P. (2000).A review of molecular recognition technologies for detection of biologicalthreat agents. Biosensors and Bioelectronics, 15, 549e578.

ISO. (2002). ISO 6579:2002. Microbiology of food and animal feeding stuffs. Horizontalmethod for the detection of Salmonella spp. Geneva, Switzerland: InternationalOrganization for Standardization.

Ivnitski, D., Abdel-Hamid, I., Atanasov, P., Wilkins, E., & Stricker, S. (2000). Appli-cation of electrochemical biosensors for detection of food pathogenic bacteria.Electroanalysis, 12, 317e325.

Jones, D. D., Law, R., & Bej, A. K. (1993). Detection of Salmonella spp. in oysters usingpolymerase chain reactions (PCR) and gene probes. Journal of Food Science, 58,1191e1197.

de Jong, B., & Ekdahl, K. (2006). The comparative burden of salmonellosis in theEuropean Union member states, associated and candidate countries. BMC PublicHealth, 6, 4.

Kalamaki, M., Prioce, R. J., & Fung, D. Y. C. (1997). Rapid methods for identifyingseafood microbial pathogens and toxins. Journal of Rapid Methods & Automationin Microbiology, 5, 87e137.

Kang, D. H., & Fung, D. Y. C. (2000). Application of thin agar layer method for re-covery of injured Salmonella typhimurium. International Journal of Food Micro-biology, 54, 127e132.

Keith, M. (1997). Evaluation of an automated enzyme-linked fluorescent immu-noassay system for the detection of Salmonellae in foods. Journal of Food Pro-tection, 60, 682e685.

Koyuncu, S., Andersson, M. G., & H€aggblom, P. (2010). Accuracy and sensitivity ofcommercial PCR-based methods for detection of Salmonella enterica in Feed.Applied and Environmental Microbiology, 76, 2815e2822.

Kutter, S., Hartmann, A., & Schmid, M. (2006). Colonization of barley (Hordeumvulgare) with Salmonella entrica and Listeria spp. FEMS Microbiology Ecology, 56,262e271.

Lazcka, O., Campo, F. J. D., & Mu�noz, F. X. (2007). Pathogen detection: a perspectiveof traditional methods and biosensors. Biosensors and Bioelectronics, 22,1205e1217.

Leonard, P., Hearty, S., Brennan, J., Dunne, L., Quinn, J., Chakraborty, T., et al. (2003).Advances in biosensors for detection of pathogens in food and water. Enzymeand Microbial Technology, 32, 3e13.

L€ofstr€om, C., Hansen, F., & Hoorfar, J. (2010). Validation of a 20-h real-time PCRmethod for screening of Salmonella in poultry faecal samples. VeterinaryMicrobiology, 144, 511e514.

Love, D. C., & Sobsey, M. D. (2007). Simple and rapid Fþ coliphage culture, latexagglutination, and typing assay to detect and source track fecal contamination.Applied and Environmental Microbiology, 73, 4110e4118.

Maciorowski, K. G., Herrera, P., Jones, F. T., Pillai, S. D., & Ricke, S. C. (2006). Culturaland immunological detection for Salmonella spp. in animal feeds e a review.Veterinary Research Communications, 30, 127e137.

Mahon, J., Murphy, C. K., Jones, P. W., & Barrow, P. A. (1994). Comparison ofmultiplex PCR and standard bacteriological methods of detecting Salmonella onchicken skin. Letters in Applied Microbiology, 19, 169e172.

Mallinson, E. T., Miller, R. G., de Rezende, C. E., Ferris, K. E., deGraft-Hanson, J., &Joseph, S. W. (2000). Improved plating media for the detection of Salmonellaspecies with typical and atypical hydrogen sulfide production. Journal of Vet-erinary Diagnostic Investigation, 12, 83e87.

Manafi, M. (2000). New developments in chromogenic and fluorogenic culturemedia. International Journal of Food Microbiology, 60, 205e218.

Mascini, M., Palchetti, I., & Marrazza, G. (2001). DNA electrochemical biosensors.Fresenius' Journal of Analytical Chemistry, 369, 15e22.

Maurer, J. J. (2011). Rapid detection and limitations of molecular techniques. AnnualReview of Food Science and Technology, 2, 259e279.

McKillip, J. L., & Drake, M. (2004). Real-time nucleic acid-based detection methodsfor pathogenic bacteria in food. Journal of Food Protection, 67, 823e832.

Mead, G., Lammerding, A. M., Nelson, C., Doyle, M. P., Humbert, F., Kulikovskiy, A.,et al. (2010). Scientific and technical factors affecting the setting of Salmonellacriteria for raw poultry: a global perspective. Journal of Food Protection, 73,1566e1590.

Mead, P. S., Slutsker, L., Dietz, V., McCaig, L. F., Bresee, J. S., Shapiro, C., et al. (1999).Food-related illness and death in the United States. Emerging Infectious Diseases,5, 607e625.

Meenakshi, M., Bakshi, C. S., Butchaiah, G., Bansal, M. P., Siddiqui, M. Z., & Singh, V. P.(1999). Adjuvanted outer membrane protein vaccine protects poultry againstinfection with Salmonella enteriditis. Veterinary Research Communications, 23,81e90.

Mozola, M. A. (2006). Genetics-based methods for detection of Salmonella spp. infoods. Journal of AOAC International, 89, 517e529.

Mozola, M. A., Peng, X., & Wendorf, M. (2007). Evaluation of the GeneQuence® DNAhybridization method in conjunction with 24-hour enrichment protocols fordetection of Salmonella spp. in select foods: collaborative study. Journal of AOACInternational, 90, 738e755.

Naravaneni, R., & Jamil, K. (2005). Rapid detection of food-borne pathogens by usingmolecular techniques. Journal of Medical Microbiology, 54, 51e54.

Nath, E. J., Neidert, E., & Randall, C. J. (1989). Evaluation of enrichment protocols forthe 1-2 Test for Salmonella detection in naturally contaminated foods and feeds.Journal of Food Protection, 52, 498e499.

Nordic Validation Organ (NordVal). (2009). Protocol for the validation of alternativemicrobiological methods. Accessed June 2012 http://www.nmkl.org/dokumenter/nordval/NordValProtocol.pdf.

Perry, J. D., & Freydi�ere, A. M. (2007). The application of chromogenic media inclinical microbiology. Journal of Applied Microbiology, 103, 2046e2055.

Polaczyk, A. L., Naryanan, J., Cromeans, T. I., Hahn, D., Roberts, J. M., Amburgey, J. E.,et al. (2008). Ultrafiltration-based techniques for rapid and simultaneous con-centration of multiple microbe classes from 100L tap water samples. Journal ofMicrobiological Methods, 73, 92e99.

Rasmussen, S. R., Rasmussen, H. B., Larsen, L. R., Hoff-Jorgensen, R., & Cano, R.(1994). Combined polymerase chain reaction-hybridization microplate assayused to detect bovine leukemia virus and Salmonella. Clinical Chemistry, 40,200e205.

Reid, R. L., Porter, R. C., & Ball, H. J. (1993). The isolation of sucrose-fermentingSalmonella mbandaka. Veterinary Microbiology, 37, 181e185.

Ricci, F., Volpe, G., Micheli, L., & Palleschi, G. (2007). A review on novel de-velopments and applications of immunosensors in food analysis. AnalyticaChimica Acta, 605, 111e129.

Ricke, S. C., Pillai, S. D., Norto, R. A., Maciorowski, K. G., & Jones, F. T. (1998).Applicability of rapid methods for detection of Salmonella spp. in poultry feeds:a review. Journal of Rapid Methods & Automation in Microbiology, 6, 239e258.

Rodríguez-L�azaro, D., Lombardc, B., Smith, H., Rzezutka, A., D'Agostino, M.,Helmuth, R., et al. (2007). Trends in analytical methodology in food safety andquality: monitoring microorganisms and genetically modified organisms.Trends in Food Science & Technology, 18, 306e319.

Rogers, K. R. (2000). Principles of affinity-based biosensors.Molecular Biotechnology,14, 109e129.

Ruiz, J., Nunez, M. L., Diaz, J., Sempere, M. A., Gomez, J., & Usera, M. A. (1996). Note:comparison of media for the isolation of lactose-positive Salmonella. Journal ofApplied Microbiology, 81, 571e574.

Saroj, S. D., Shashidhar, R., Karani, M., & Bandekar, J. R. (2008). Rapid, sensitive, andvalidated method for detection of Salmonella in food by an enrichment brothculture e Nested PCR combination assay. Molecular and Cellular Probes, 22,201e206.

Scaria, J., Palaniappan, U. M., Ckiu, D., Phan, J. A., Ponnala, L., McDonough, P., et al.(2008). Microarray for molecular typing of Salmonella enterica serovars. Mo-lecular and Cellular Probes, 22, 238e243.

Settanni, L., & Corsetti, A. (2007). The use of multiplex PCR to detect and differ-entiate food- and beverage-associated microorganisms: a review. Journal ofMicrobiological Methods, 69, 1e22.

Skottrup, P. D., Nicolaisen, M., & Justesen, A. F. (2008). Towards on-site pathogendetectionusingantibody-basedsensors.BiosensorsandBioelectronics, 24, 339e348.

Sørensen, L. L., Alban, L., Nielsen, B., & Dahl, J. (2004). The correlation betweenSalmonella serology and isolation of Salmonella in Danish pigs at slaughter.Veterinary Microbiology, 101, 131e141.

Stager, C. E., & Davis, J. R. (1992). Automated systems for identification of micro-organisms. Clinical Microbiology Reviews, 5, 302e327.

K.-M. Lee et al. / Food Control 47 (2015) 264e276276

Stevens, K. A., & Jaykus, L. A. (2004). Bacterial separation and concentrationfrom complex sample matrices: a review. Critical Reviews in Microbiology, 30,7e24.

Stone, G. G., Oberst, R. D., Hays, M. P., McVey, S., & Chengappa, M. M. (1994).Detection of Salmonella serovars from clinical samples by enrichment brothcultivation-PCR procedure. Journal of Clinical Microbiology, 32, 1742e1749.

Swaminathan, B., & Feng, P. (1994). Rapid detection of food-borne pathogenicbacteria. Annual Reviews in Microbiology, 48, 401e426.

Thorns, C. J., McLaren, I. M., & Sojka, M. G. (1994). The use of latex particle agglu-tination to specifically detect Salmonella enteritidis. International Journal of FoodMicrobiology, 21, 47e53.

Tietjen, M., & Fung, D. Y. C. (1995). Salmonellae and food safety. Critical Reviews inMicrobiology, 21, 53e83.

T€or€ok, T. J., Tauxe, R. V., Wise, R. P., Livengood, J. R., Sokolow, R. S., Mauvais, S., et al.(1997). A large community outbreak of salmonellosis caused by intentionalcontamination of restaurant salad bars. Journal of the American Medical Asso-ciation, 278, 389e395.

Turner, C. W. (2010). Microbiology of ready-to-eat poultry products. In I. Guerrero-Legarreta, & Y. H. Hui (Eds.), Handbook of poultry science and technology, volume2: Secondary processing (pp. 507e515). Hoboken: John Wiley & Sons, Inc.

Uyttendaele, M., Vanwildemeersch, K., & Debevere, J. (2003). Evaluation of real-time PCR vs automated ELISA and a conventional culture method using asemi-solid medium for detection of Salmonella. Letters in Applied Microbiology,37, 386e391.

Van Dorst, B., Mehta, J., Bekaert, K., Rouah-Martin, E., De Coen, W., Dubruel, P., et al.(2010). Recent advances in recognition elements of food and environmentalbiosensors: a review. Biosensors and Bioelectronics, 26, 1178e1194.

Wiuff, C., Jauho, E. S., Stryhn, H., Andresen, L. O., Thaulov, K., Boas, U., et al. (2000).Evaluation of a novel enzyme-linked immunosorbent assay for detection ofantibodies against Salmonella, employing a stable coating oflipopolysaccharide-derived antigens covalently attached to polystyrenemicrowells. Journal of Veterinary Diagnostic Investigation, 12, 130e135.

Wolcott, M. J. (1991). DNA-based rapid methods for the detection of foodbornepathogens. Journal of Food Protection, 54, 387e401.

Wolffs, P. F. G., Glencross, K., Thibaudeau, R., & Griffiths, M. (2006). Direct quanti-tation and detection of salmonellae in biological samples without enrichment,using two-step filtration and real-time PCR. Applied and Environmental Micro-biology, 72, 3896e3900.

World Health Organization (WHO). (2005). Drug-resistant Salmonella AccessedMarch 2013 http://www.who.int/mediacentre/factsheets/fs139/en/.