**2. Culture methods**

Culture based methods are still the most widely used detection techniques and remain the gold standard for the detection of *Salmonella* due to their selectivity and sensitivity. For instance, the US Food and Drug Administration (FDA) and the Food Safety and Inspection Service (FSIS) of the US Department of Agriculture (USDA), requires an isolated organism as unambiguous proof of contamination (Alocilja and Radke 2003). Depending on the approach, standard culture methods typically require 5–7 days to obtain a result as they rely on the ability of *Salmonella* to multiply to visible colonies, which can then be characterized by performing additional biochemical and or serological tests. Due to their widespread use, numerous and varied bacteriological media (selective enrichment broths and selective agar plates) are applied to best monitor for *Salmonella* in food and food ingredients. The media may contain inhibitors in order to stop or delay the growth of non-target organisms, or particular substrates that only the target bacteria can degrade, or that confer a particular colour to the growing colonies (Manafi 2000).

Cultural methods typically involve the enrichment of a portion of the food sample to recover sub-lethally injured cells due to heat, cold, acid, or osmotic shock (Sandel et al. 2003; Gracias and McKillip 2004) in a non-selective pre-enrichment media, such as Buffered Peptone Water (BPW), and to increase the number of target cells as these are generally not uniformly distributed in foods, typically occur in low numbers, and may be present in a mixed microbial population. Next, primary enrichment cultures are typically inoculated into secondary selective enrichment broths, such as Selenite Cystine broth (SC), Rappaport Vasiliadis Soy broth (RVS), Tetrathionate Broth (TT), or Muller-Kauffmann Tetrathionate-Novobiocin broth (MKTTn) and incubated at elevated temperatures (37°C or 42°C for 18-24 hours) before being struck onto selective agars such as Xylose Lysine Deoxycholate agar (XLD agar), Bismuth Sulphite agar (BIS), Brilliant Green agar (BG) with or without the addition of sulfadiazine or sulfapyridine (BGS), modified semisolid Rappaport Vasiliadis (MSRV), Salmonella Shigella Agar, or Hektoen Enteric agar. There are several published standard methods utilizing combinations of media such as the current ISO horizontal method, ISO 6579:2002 (updated in 2007) for the detection *Salmonella*, including *Salmonella*  Typhi and *Salmonella* Paratyphi applicable to products intended for human consumption and the feeding of animals, and to environmental samples in the area of food production and food handling. Similar standard methods have been published elsewhere, most notably in the FDA Bacteriological Analytical Manual (BAM).

Typical *Salmonella* colonies based on morphology and or indicative biochemical reactions on selective agars are then cultured onto non-selective media prior to confirmatory testing. There are well-established confirmations and identification procedures for *Salmonella*. Preliminary identification is traditionally performed using classical biochemical and serological tests. Key biochemical tests include the fermentation of glucose, negative urease

advances introduced for the improvement of conventional culture methods, the use of Polymerase Chain Reaction (PCR) technology, immunology-based methods, and bacteriophage based assays. Whenever possible, examples from the academic literature as well as from commercial applications will be considered. The importance of sample preparation will be examined throughout as it relates to its impact on sensitivity and turnaround time for detection. Specific *Salmonella* serovars will be named according to the nomenclature of Leminor and Popoff (2001), e.g. *Salmonella* Enteritidis or *S.* Enteritidis.

Culture based methods are still the most widely used detection techniques and remain the gold standard for the detection of *Salmonella* due to their selectivity and sensitivity. For instance, the US Food and Drug Administration (FDA) and the Food Safety and Inspection Service (FSIS) of the US Department of Agriculture (USDA), requires an isolated organism as unambiguous proof of contamination (Alocilja and Radke 2003). Depending on the approach, standard culture methods typically require 5–7 days to obtain a result as they rely on the ability of *Salmonella* to multiply to visible colonies, which can then be characterized by performing additional biochemical and or serological tests. Due to their widespread use, numerous and varied bacteriological media (selective enrichment broths and selective agar plates) are applied to best monitor for *Salmonella* in food and food ingredients. The media may contain inhibitors in order to stop or delay the growth of non-target organisms, or particular substrates that only the target bacteria can degrade, or that confer a particular

Cultural methods typically involve the enrichment of a portion of the food sample to recover sub-lethally injured cells due to heat, cold, acid, or osmotic shock (Sandel et al. 2003; Gracias and McKillip 2004) in a non-selective pre-enrichment media, such as Buffered Peptone Water (BPW), and to increase the number of target cells as these are generally not uniformly distributed in foods, typically occur in low numbers, and may be present in a mixed microbial population. Next, primary enrichment cultures are typically inoculated into secondary selective enrichment broths, such as Selenite Cystine broth (SC), Rappaport Vasiliadis Soy broth (RVS), Tetrathionate Broth (TT), or Muller-Kauffmann Tetrathionate-Novobiocin broth (MKTTn) and incubated at elevated temperatures (37°C or 42°C for 18-24 hours) before being struck onto selective agars such as Xylose Lysine Deoxycholate agar (XLD agar), Bismuth Sulphite agar (BIS), Brilliant Green agar (BG) with or without the addition of sulfadiazine or sulfapyridine (BGS), modified semisolid Rappaport Vasiliadis (MSRV), Salmonella Shigella Agar, or Hektoen Enteric agar. There are several published standard methods utilizing combinations of media such as the current ISO horizontal method, ISO 6579:2002 (updated in 2007) for the detection *Salmonella*, including *Salmonella*  Typhi and *Salmonella* Paratyphi applicable to products intended for human consumption and the feeding of animals, and to environmental samples in the area of food production and food handling. Similar standard methods have been published elsewhere, most notably

Typical *Salmonella* colonies based on morphology and or indicative biochemical reactions on selective agars are then cultured onto non-selective media prior to confirmatory testing. There are well-established confirmations and identification procedures for *Salmonella*. Preliminary identification is traditionally performed using classical biochemical and serological tests. Key biochemical tests include the fermentation of glucose, negative urease

**2. Culture methods** 

colour to the growing colonies (Manafi 2000).

in the FDA Bacteriological Analytical Manual (BAM).

reaction, lysine decarboxylase, negative indole test, H2S production, and fermentation of dulcitol. Serological confirmation tests typically utilize polyvalent antisera for flagellar (H) and somatic (O) antigens. Isolates with a typical biochemical profile, which agglutinate with both H and O antisera are identified as *Salmonella* species. Where results are inconclusive, it may be necessary to perform additional biochemical tests. Positive isolates are often sent for further serotyping to identify the serovar using specific antisera as per the Kauffman-White (KW) typing scheme recognizing 46 O antigens, and 119 H antigens, thereby permitting the characterization of 2,541 serotypes (Shipp and Rowe 1980). Serotyping is a useful epidemiological tool in identifying circulating serotypes and to characterize outbreaks. The antigenic formulae of Le Minor and Popoff (2001) is a standard method for naming the serovars. However, serotyping is normally undertaken at reference laboratories and is rarely performed in routine food or clinical laboratories. Reference laboratories are also able to further type isolates using techniques such a phage typing (Anderson and Williams 1956; Callow 1959; Anderson 1964; Anderson et al. 1977), antibiotic susceptibility (Bauer et al. 1966), pulsed-field gel electrophoresis (PFGE), or other emerging genetic typing technologies such Multiple-Locus Variable Number Tandem Repeat Analysis (MLVA) and Multilocus sequence typing (MLST) (Kruy et al. 2011).

Although standard culture methods are excessively time-consuming, there is potential for further improvements, and thus many attempts have been made to maximize their efficiency by introducing new technologies, making reliability of detection more convenient, user friendly, as well as by reducing the costs of materials and labour (de Boer and Beumer 1999; Weenk, 1992). For example, biochemical confirmatory tests may be easily replaced by commercial identification kits such as the API 20E (BioMérieux) or other commercially available bacterial identification kits. The detection of sub-lethally damaged cells is of utmost importance as these may still pose a risk to human health and may lead to false negative results. Strategies for the recovery of injured bacteria are based on overlay methods such as tryptic soy agar (TSA) overlayed on XLD selective agar (Kang and Fung 2000) and other approaches also include the development of single enrichment broths where multiple step enrichments are usually required (Baylis et al. 2000). Other novel approaches include the addition of bacteriophages for the elimination of background microflora that may out-compete the target organism. For example, RapidChek® SELECTTM Salmonella (Strategic Diagnostics Inc.) employs a primary enrichment media supplemented with a bacteriophage cocktail as a selective agent, which reduces the level of background flora in high burden samples allowing *Salmonella* to grow with minimal competition. In addition, there is also the development of enrichment broths for the concurrent enrichment of pathogens thereby reducing laboratory workloads with respect to the preparation of sample homogenates since different enrichment broths would no longer be required, and multiple analyses could be performed from a single universal enrichment culture (Kim and Bhunia 2008). Amendments to media have also been performed such as the addition of novobiocin (Restaino et al. 1977; Devenish et al. 1986), and cycloheximide to decrease fungal overgrowth (Ricke et al. 1998). Lastly, and perhaps the most important advancement is the use of chromogenic or fluorogenic substrates in selective agars, permitting identification to be performed directly on the isolation plate, thereby expediting or eliminating the use of subculture media or additional biochemical tests as these media provide highly specific reactions, and help reduce the workload for unnecessary examination of suspect colonies arising from poor specificity of conventional agars (Manafi 1996; Manafi 2000). A number of selective chromogenic agar media

*Salmonella* Detection Methods for Food and Food Ingredients 377

automated immunoconcentration prior to analysis by an automated ELISA instrument for the detection of *Salmonella* from food and food ingredients. Another IMS variation was also developed by Pathatrix (Matrix MicroScience Ltd) combining IMS and a recirculation step (Flow Through Immunocapture or FTI), to further increase the sensitivity of detection since larger enrichment volumes can be reacted with IMS beads. For example Warren et al. (2007) investigated FTI, using the Pathatrix device, followed by plating on XLD agar (FTI-XLD) or analysis by real-time PCR (FTI-PCR) for the detection of *Salmonella* on smooth tomato surfaces and in potato salad and ground beef. The FTI-XLD method demonstrated the ability to isolate presumptive *Salmonella* colonies up to 48 h faster than did the standard modified BAM

Among the problems associated with IMS is non-target carryover where non-target organisms adhere to the walls of glass test tubes (Meadows 1971). Protamine as well as the use of mild detergents is commonly used to minimize non-target carryover since it adheres to the glass and to the bacteria in the sample reducing the net negative charge to prevent adherence. IMS also suffers in that it requires small sample sizes, organisms may be lost from beads during separation from samples with high fat content, and non specific binding of *Citrobacter freundii*

Several rapid latex agglutination assay tests are widely used for the rapid detection of *Salmonella*. These assays however, are primarily used as a confirmation screen for presumptive *Salmonella* colonies after culture isolation from selective agar plates, with further confirmation and identification work carried out on those organisms giving a positive latex reaction. An aliquot of a colony suspension or enrichment broth is simply mixed with the latex reagent and after a few minutes rotation, the results are clearly visible. If the test is negative, the latex remains in smooth suspension and retains its original colour. A positive result is indicated by distinct colour agglutination against an altered background. By reducing the number of samples requiring further confirmatory testing, these tests save time and resources and allow negative results to be reported at least 24 hours earlier than by conventional culture methods. However, depending on the antibodies used they may lack specificity due to non-specific agglutination of some organisms (Cheesbrough and Donnelly, 1996). Some commercial kits include Remel Wellcolex Colour tests for the presumptive identification of *Salmonella* serogroups A, B, C, D, E, and G, and the Vi antigen using just two reagents. Similar tests include Oxoid Salmonella latex test, Microgen Salmonella Latex test,

Enzyme-linked immunosorbent assay (ELISA) also known as an enzyme immunoassay (EIA), is a biochemical technique used to detect the presence of an antibody or an antigen in a sample. In the context of *Salmonella* detection, a sample with an unknown amount of antigen is immobilized on a solid support (usually a polystyrene microtitre plate) either non-specifically (via adsorption to the surface) or specifically (via capture by another antibody specific to the same antigen, in a "Sandwich" ELISA). After the antigen is immobilized, a detection antibody linked to an enzyme such as Horse Radish Peroxidase (HRP) is added, forming a complex with the antigen. Between each step, the plate is

*Salmonella* culture method and the FTI-PCR was able to detect *Salmonella* within 8h.

and coliforms with mucoid layers has also been observed (Coleman et al 1995).

**3. Immunological based methods** 

**3.1 Rapid agglutination assays** 

and Denka-Seiken, among others.

**3.2 Enzyme-linked immunosorbent assay (ELISA)** 

specifically designed for the differentiation of *Salmonella* colonies are commercially available with varying success of adoption by regulatory agencies such as: Salmonella SMS (AES Chemunex), BBL CHROMagar (CHROMagar), RAPID'Salmonella (Bio-Rad Laboratories, S.A.), chromID Salmonella (BioMerieux), Harlequin Salmonella ABC (Lab M), Oxoid Brilliance Salmonella Agar (Oxoid), and Rambach Agar (Merck), among others.

It is evident that the multitude of options for isolation of *Salmonella* and the lack of interlaboratory consistency make *Salmonella* isolation one of the most variable procedures in laboratories with new media available every year, promising to be more sensitive, specific, and rapid (Hyatt and Weese 2004). With this myriad of choice, laboratories must chose culture approaches which efficiently and accurately provide timely results via the development of standard methods and participation in proficiency quality assurance programs.

#### **2.1 Immunomagnetic separation**

In an attempt to reduce the length of routine microbiological analysis and to minimize the problems associated with rapid detection systems such as interference from foods and food ingredients debris, background micro-organisms, and lack of sensitivity, there has been a lot of interest in the development of separation and concentration techniques prior to detection. Various techniques have been utilized for this purpose including: centrifugation (Basel et al. 1983), filtration (Farber and Sharpe 1984), and lectin-based biosorbents (Payne et al. 1992). However, the most successful of approaches for the separation and concentration of target organisms has been the use of immunomagnetic separation (IMS). The advantages of IMS are that it reduces the total analysis time and improves the sensitivity of detection. IMS is rapid, technically simple, and specific method for the isolation of the target organisms (Shaw et al. 1998). Paramagnetic particles are coated with antibodies specific to the target organism and added to a post enrichment culture. The target organism is captured onto the magnetic particles and the whole complex is then removed from the system by the application of a magnetic field. Target organisms are thus removed from food debris and competing microorganisms, which may otherwise interfere with the detection system. If required, the isolated complex may be re-suspended in an enrichment broth so that cell numbers can be rapidly increased to improve the sensitivity of detection assays. In addition, IMS by design can be used in conjunction with other rapid detection methods, including ELISA, conductance microbiology, electrochemiluminescence, and polymerase chain reaction (PCR) to further increase its analytical sensitivity (Fluit et al. 1993; Cudjoe et al. 1994; Cudjoe et al. 1995; Sapanova et al. 2000). It has been reported that IMS is more sensitive than conventional culture methods and is able to reduce the total culture analysis time by one to two days (Lynch et al. 2004; Ten Bosch et al. 1992).

The most commonly used commercial IMS bead for the recovery of *Salmonella* from food samples is Dynabeads anti *Salmonella* (Invitrogen). Similar magnetic beads specific for *Salmonella* are available such as Captivate Salmonella (Lab M), Tecra Salmonella Unique (3M), as wells as for specific serovars such as *S.* Enteritidis, via Rapidchek Confirm *S.* Enteritidis IMS kit (SDIX). IMS can also be automated using automated IMS separators such as the BeadRetriever (Invitrogen) capable of processing up to fifteen 1 mL enrichments volumes per cycle (23 minutes), to larger scale instruments such as the Kingfisher IMS separator (Thermofisher) or Mag Max (Life Technologies) capable of processing up to 100 samples with the capability of re-suspending the IMS target complex in microtitre plates for further testing by PCR, or ELISA. For instance, the VIDAS ICS test (BioMérieux) uses

specifically designed for the differentiation of *Salmonella* colonies are commercially available with varying success of adoption by regulatory agencies such as: Salmonella SMS (AES Chemunex), BBL CHROMagar (CHROMagar), RAPID'Salmonella (Bio-Rad Laboratories, S.A.), chromID Salmonella (BioMerieux), Harlequin Salmonella ABC (Lab M), Oxoid

It is evident that the multitude of options for isolation of *Salmonella* and the lack of interlaboratory consistency make *Salmonella* isolation one of the most variable procedures in laboratories with new media available every year, promising to be more sensitive, specific, and rapid (Hyatt and Weese 2004). With this myriad of choice, laboratories must chose culture approaches which efficiently and accurately provide timely results via the development of

In an attempt to reduce the length of routine microbiological analysis and to minimize the problems associated with rapid detection systems such as interference from foods and food ingredients debris, background micro-organisms, and lack of sensitivity, there has been a lot of interest in the development of separation and concentration techniques prior to detection. Various techniques have been utilized for this purpose including: centrifugation (Basel et al. 1983), filtration (Farber and Sharpe 1984), and lectin-based biosorbents (Payne et al. 1992). However, the most successful of approaches for the separation and concentration of target organisms has been the use of immunomagnetic separation (IMS). The advantages of IMS are that it reduces the total analysis time and improves the sensitivity of detection. IMS is rapid, technically simple, and specific method for the isolation of the target organisms (Shaw et al. 1998). Paramagnetic particles are coated with antibodies specific to the target organism and added to a post enrichment culture. The target organism is captured onto the magnetic particles and the whole complex is then removed from the system by the application of a magnetic field. Target organisms are thus removed from food debris and competing microorganisms, which may otherwise interfere with the detection system. If required, the isolated complex may be re-suspended in an enrichment broth so that cell numbers can be rapidly increased to improve the sensitivity of detection assays. In addition, IMS by design can be used in conjunction with other rapid detection methods, including ELISA, conductance microbiology, electrochemiluminescence, and polymerase chain reaction (PCR) to further increase its analytical sensitivity (Fluit et al. 1993; Cudjoe et al. 1994; Cudjoe et al. 1995; Sapanova et al. 2000). It has been reported that IMS is more sensitive than conventional culture methods and is able to reduce the total culture analysis time by one to two days

The most commonly used commercial IMS bead for the recovery of *Salmonella* from food samples is Dynabeads anti *Salmonella* (Invitrogen). Similar magnetic beads specific for *Salmonella* are available such as Captivate Salmonella (Lab M), Tecra Salmonella Unique (3M), as wells as for specific serovars such as *S.* Enteritidis, via Rapidchek Confirm *S.* Enteritidis IMS kit (SDIX). IMS can also be automated using automated IMS separators such as the BeadRetriever (Invitrogen) capable of processing up to fifteen 1 mL enrichments volumes per cycle (23 minutes), to larger scale instruments such as the Kingfisher IMS separator (Thermofisher) or Mag Max (Life Technologies) capable of processing up to 100 samples with the capability of re-suspending the IMS target complex in microtitre plates for further testing by PCR, or ELISA. For instance, the VIDAS ICS test (BioMérieux) uses

Brilliance Salmonella Agar (Oxoid), and Rambach Agar (Merck), among others.

standard methods and participation in proficiency quality assurance programs.

**2.1 Immunomagnetic separation** 

(Lynch et al. 2004; Ten Bosch et al. 1992).

automated immunoconcentration prior to analysis by an automated ELISA instrument for the detection of *Salmonella* from food and food ingredients. Another IMS variation was also developed by Pathatrix (Matrix MicroScience Ltd) combining IMS and a recirculation step (Flow Through Immunocapture or FTI), to further increase the sensitivity of detection since larger enrichment volumes can be reacted with IMS beads. For example Warren et al. (2007) investigated FTI, using the Pathatrix device, followed by plating on XLD agar (FTI-XLD) or analysis by real-time PCR (FTI-PCR) for the detection of *Salmonella* on smooth tomato surfaces and in potato salad and ground beef. The FTI-XLD method demonstrated the ability to isolate presumptive *Salmonella* colonies up to 48 h faster than did the standard modified BAM *Salmonella* culture method and the FTI-PCR was able to detect *Salmonella* within 8h.

Among the problems associated with IMS is non-target carryover where non-target organisms adhere to the walls of glass test tubes (Meadows 1971). Protamine as well as the use of mild detergents is commonly used to minimize non-target carryover since it adheres to the glass and to the bacteria in the sample reducing the net negative charge to prevent adherence. IMS also suffers in that it requires small sample sizes, organisms may be lost from beads during separation from samples with high fat content, and non specific binding of *Citrobacter freundii* and coliforms with mucoid layers has also been observed (Coleman et al 1995).
