Application of Rapid Dot Blot Immunoassay for Detection of Salmonella enterica Serovar Enteritidis in Eggs, Poultry, and Other Foods

Abstract

Salmonella enterica serovar Enteritidis was detected in artificially inoculated eggs within 24 h through a rapid monoclonal antibody-based dot blot immunoassay. Detection in poultry and other products required 28 h. Samples were directly enriched in homogenized egg without the need for pre- or postenrichment steps. Serovar Enteritidis was detected in the presence of other bacteria when outcompeted 1:400.

Conventional methods for detection of Salmonella in foods are labor-intensive, time-consuming, and expensive. It has also been found that some of the routinely used selective enrichment broths are inhibitory towards Salmonella enterica serovar Enteritidis (28). Rapid methods based on principles such as membrane technology (11), latex agglutination (27), immunoassays (6, 18, 30), and immunomagnetic separation (8, 9, 19) have been developed. Methods employing PCR in combination with preenrichment broths (24, 31, 32), immunomagnetic separation (25, 26), or centrifugation (21, 22) are currently being developed.

Immunologically based methods specific for serovar Enteritidis suffer from the same drawbacks as the above methods: one or more enrichment broths, or in some cases, postenrichment broths, are required. Cross-reactivity has also been observed with most monoclonal antibodies produced against serovar Enteritidis (17, 18, 27). This report describes the development of a rapid dot blot immunoassay for the detection of serovar Enteritidis in eggs, poultry, and other products.

Large grade A eggs were scrubbed with 70% ethanol and opened aseptically. Eggs were mixed for 30 s using a stomacher lab blender 400 (Seward Laboratory, London, England) and were inoculated with either serovar Enteritidis phage type 1, 4, 8, 13, or 13a. After incubation, 25-ml portions were placed into 50-ml polypropylene tubes, and 0.1 volume of 15% sodium cholate in phosphate-buffered saline (PBS) (pH 7.2) was added. After being mixed, tubes were placed in boiling water for 10 min and cooled for 30 min at 4°C. Through heating, the egg mixture was solidified, forming a solid egg gel. After cooling, the gel was removed from the tube, and a sterile core borer (10-mm diameter) was used to create small cylindrical gels. The cylindrical gels were then cut into disks of 2 mm in thickness.

Nitrocellulose strips (8.5 by 2.5 cm) were prewetted with PBS prior to use. Egg disks were placed on the membrane for 5 min, removed, and washed for 2 min in PBS. Strips were blocked for 45 min in 5% skim milk powder in Tris-buffered saline (pH 7.5) and incubated with a murine monoclonal antibody solution (tissue culture supernatant) for 45 min, followed by 1 h of incubation in biotinylated goat anti-mouse immunoglobulin G. Strips were incubated with streptavidin-alkaline phosphatase for 1 h and developed with a BCIP (5-bromo-4-chloro-3-indolylphosphate)-nitroblue tetrazolium chloride substrate solution. Membranes were washed twice with Tris-buffered saline containing 0.05% Tween 20 for 2 min between each step. All steps were performed at room temperature.

Bacterial cultures were serially diluted to appropriate inoculum levels and confirmed by plating on standard plate count agar in triplicate. Negative controls were inoculated with 0.1% peptone water. When the specificity of the assay was evaluated, the negative control contained all bacterial species except S. enterica serovar Enteritidis. Artificially inoculated samples were also tested through conventional culture methods (1) with serological confirmation (Salmonella O-9 antiserum). For samples containing a mixed population of bacteria, Salmonella O-4 and O-5 antisera were used to differentiate between B and D₁ serogroups.

For detection of serovar Enteritidis in eggs, cultures were enriched directly in homogenized eggs without the need for preenrichment or selective enrichment steps. This method demonstrated that only 20 h of incubation, without the need to isolate the organism, was required to enrich serovar Enteritidis to detectable levels (Fig. 1). After incubation, initial inocula of 1, 5, 10, 50, and 500 CFU per 25 g of homogenized egg multiplied to approximately 10⁶, 10⁷, 10⁸, 10⁸, and 10⁹ CFU/ml, respectively. For detection of 500 CFU/25 g, a distinct circular pattern was not observed. Due to the high bacterial concentration after incubation, it is believed that the lipopolysaccharide (LPS) antigen saturated the strip, causing complete coloration.

FIG. 1.

Detection of serovar Enteritidis in artificially inoculated eggs by the dot blot immunoassay. Values are the initial inocula (CFU) per 25 g of egg.

Contamination by transshell transmission frequently involves a mixed infection dominated by gram-negative bacteria (5). In order to assess the ability of the assay to detect serovar Enteritidis among a mixed population of bacteria, six species were selected. S. enterica serovar Heidelberg has been frequently isolated from poultry and egg shells (3, 16, 23, 29), while Escherichia coli, Proteus vulgaris, Citrobacter freundii (ATCC 8090), Alcanigenes faecalis, and Pseudomonas fluorescens are common spoilage organisms associated with eggs (5). Samples were inoculated with 33, 50, 67, 100, or 133 CFU of each bacterial species per ml. When all six bacterial species at these inoculum levels were combined with 2 CFU of serovar Enteritidis, the ratios of serovar Enteritidis to competing bacteria were 1 to 100, 1 to 150, 1 to 200, 1 to 300, and 1 to 400. After 20 h of incubation at 37°C, serovar Enteritidis was detected in all samples (Fig. 2).

FIG. 2.

Detection of serovar Enteritidis in artificially inoculated eggs in the presence of a mixed population of bacteria (serovar Heidelberg, P. vulgaris, P. fluorescens, E. coli, C. freundii, and A. faecalis). Values are the ratios of serovar Enteritidis to the mixed population of bacteria per 50 g of egg.

Eggs contain an adequate amount of nutrients to support the growth of serovar Enteritidis (2). Although antimicrobial agents such as conalbumen and lysozyme are present in the albumen, both are neutralized when the yolk and white are homogenized, thus allowing microbial growth (7, 12). However, Cudjoe and coworkers (8) found that yolk-albumen mixtures still had inhibitory effects on the growth of bacteria. Gast and Holt (14) also suggested that the ability of serovar Enteritidis to grow rapidly in liquid whole eggs is a characteristic of various strains. This could explain why serovar Enteritidis is able to rapidly multiply in homogenized eggs despite the inhibitory effects. This could also explain why serovar Enteritidis was able to survive and proliferate in the presence of competing bacteria. Dolman and Board (10) also reported that serovar Enteritidis phage type 4 was able to outcompete other gram-negative bacteria at incubation temperatures near 37°C.

Since the number of Salmonella-positive eggs laid by infected hens is small, with small numbers of serovar Enteritidis present, detection methods must be highly sensitive. Gast (13) pooled the contents of eggs, which permits a statistically meaningful sample size; however, some form of preenrichment must be employed. Incubation of 1, 5, or 10 CFU of serovar Enteritidis per 500 g of homogenized egg supplemented with ferrous sulfate for 20 h at 37°C allowed CFU to multiply to detectable levels (results not shown). Ferrous sulfate has been found to promote the growth of Salmonella in egg contents due to the limited availability of iron when eggs are pooled (15). When egg pools were inoculated with competing bacterial species and incubated for 24 h at 37°C, serovar Enteritidis was detected when outcompeted 1:300 (results not shown).

Detection in poultry, ice cream, skim milk powder, and poultry feed was conducted by direct enrichment in homogenized egg followed by immunoassay. Incubation of 1, 5, or 10 CFU for 24 h at 37°C was sufficient for detection of serovar Enteritidis (results not shown).

This assay employed the use of a monoclonal antibody (ATCC HB-11891) specific to the LPS O-9 (factor 9) antigen of serovar Enteritidis. Factor 9 is part of the D₁ Salmonella O antigen and is composed of two monosaccharides, tyvelose and mannose. The tyvelose residue is a side sugar attached to the trisaccharide backbone through α1,3 linkage. This particular sugar residue and linkage, specific to D₁ Salmonella, is believed to play a significant role in the specificity of the antibody (20).

One of the other important features of the immunoassay is based on the distribution of LPS in the gelled egg matrix formed upon heating. Through the addition of sodium cholate and the application of heat, the LPS antigen of serovar Enteritidis was released from the bacterial membrane. Through diffusional forces, the antigen was able to move through the porous egg sample and onto the solid support for detection. Other detergents have been used for extraction of LPS antigens (4); however, Wang and coworkers (30) found that a 15% sodium cholate solution was the most efficient.