Evaluation of culture enrichment procedures for use with Salmonella detection immunoassay

International Journal of Food Microbiology 51 (1999) 85–94 www.elsevier.nl / locate / ijfoodmicro

Evaluation of culture enrichment procedures for use with Salmonella detection immunoassay a, a a a Hongsheng Huang *, Manuel M. Garcia , Brian W. Brooks , Klaus Nielsen , b Sze-Park Ng a

Animal Diseases Research Institute (ADRI), Canadian Food Inspection Agency, 3851 Fallowfield Road, Nepean, Ontario, K2 H 8 P9 Canada b Department of Microbiology, University of Hong Kong, Queen Mary Hospital, Pokfulam, Hong Kong Received 27 January 1999; received in revised form 6 May 1999; accepted 20 June 1999

Abstract To design efficient culture strategies for use with immunoassays to detect Salmonella in food, the growth of these organisms was investigated according to the Bacteriological Analytical Manual (BAM) and enrichment-immunoassay (EI) culture procedures. The cultures were further evaluated using a commercial enzyme-linked immunosorbent assay (ELISA) kit. The BAM procedure includes pre-enrichment in nutrient broth (NB) for 16 h followed by selective enrichment in either Rappaport–Vassiliadis (RV) or tetrathionate brilliant green (TBG) broth for 16 h. The EI procedure includes pre-enrichment in NB for 4 h, selective enrichment in RV for 16 h and post-enrichment in NB for 4 h. The effects of different incubation times for pre- and post-enrichment, and different culture media for selective enrichment (TBG and RV) and post-enrichment in NB and Brain Heart Infusion broth (BHI) on the growth of the bacteria and ELISA titers in the EI procedure were also investigated. Salmonella enteritidis and S. typhimurium inoculated at different initial concentrations between 0.1 and 35 CFU / ml grew to similar concentrations of 10 7 to 10 8 colony forming unit (CFU) / ml in pure culture and generally 2 to 4 fold lower concentrations (P , 0.05) in mixed culture using spiked chicken rinse. In the BAM procedure, the concentration of Salmonella cultured in RV was higher (P , 0.01) than that in TBG. The cultures in TBG showed positive results for ELISA, but those in RV were generally negative. In the EI procedure, the ELISA titers from cultures post-enriched in NB or BHI were higher (P , 0.01) when TBG, as compared to RV, was used for selective enrichment. Post-enrichment in BHI yielded higher numbers of Salmonella and higher ELISA titers than those in NB (P , 0.05) for post-enrichment. This study demonstrated that in both culture procedures small numbers of Salmonella could be increased to at least 10 7 CFU / ml which is detectable by most ELISAs, and that the type of the culture media used may have a significant impact on ELISA results. Crown copyright  1999 Published by Elsevier Science B.V. Keywords: Salmonella; Enrichment; ELISA

*Corresponding author. Tel.: 1 1-613-228-6698; fax: 1 1-613-228-6670. E-mail address: [email protected] (H. Huang) 0168-1605 / 99 / $ – see front matter PII: S0168-1605( 99 )00102-6

Crown copyright  1999 Published by Elsevier Science B.V.

86

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

1. Introduction Salmonella, with more than 2300 serovars identified, has persisted as one of the major foodborne pathogens (D’Aoust, 1997). The conventional procedure for detection of Salmonella is culture. However, cultural isolation and identification requires three to five days for completion and includes preenrichment, selective enrichment, selective agar plating, biochemical screening, and serological identification (Andrews et al., 1992). The need by the food industry to reduce testing costs and shorten the time to obtain results has led to the development of rapid immunoassays for Salmonella detection (Feng, 1992). Several enzymelinked immunosorbent assays (ELISAs) have been reported which can detect Salmonella within one to two days (Jay and Comar, 1988; Prusak-Sochaczewski and Luong, 1989; Lee et al., 1990; Wyatt et al., 1993; Ng et al., 1996). The detection limit of most ELISAs is between 10 4 and 10 7 colony forming units (CFU) per ml in the enrichment broth (PrusakSochaczewski and Luong, 1989; Lee et al., 1990; Wyatt et al., 1993; Ng et al., 1996; Mattingly and Gehle, 1984; Choi et al., 1992). However, many samples contain less than 10 4 CFU / ml, and in some outbreaks as few as several to 100 Salmonella cells have been able to bring about disease (Lipson, 1976; Fontaine et al., 1978; D’Aoust et al., 1975; D’Aoust, 1985). Therefore, it is important for a culture enrichment step to be included in the assay procedure to allow multiplication of small numbers of Salmonella organisms to ELISA detectable levels. In some ELISA tests, Salmonella are enriched in either nonselective or selective medium (Lee et al., 1990; Wyatt et al., 1993). Ng et al. (1996) demonstrated that post-enrichment, following pre-enrichment and selective enrichment, is important to increase ELISA sensitivity, particularly with samples such as poultry carcasses which contain high bacterial loads. This enrichment-immunoassay procedure is able to achieve a result within 30 hours by eliminating the need for isolation (Ng et al., 1996; Sinoclone Ltd., 1996; Ng et al., 1997). Knowledge of the growth of Salmonella in various media would be useful in designing enrichment procedures to meet the detection limits of different ELISAs. In the present study the growth of S. enteritidis

and S. typhimurium was investigated using two methods based on the American Association of Analytical Chemists (AOAC) International recommended Bacteriological Analytical Manual (BAM) procedure (Andrews et al., 1992) which is similar to that of International Standard Organization (ISO 6579:1993) and European Standard procedure with minor modifications of enrichment media, and in an enrichment-immunassay procedure recommended for use with a commercially available ELISA (EI) (Sinoclone Ltd., 1996). The culture results were compared with those of the ELISA. Because poultry has been indicated as a major source of Salmonella for humans (D’Aoust, 1997), the present study examined the growth of Salmonella in chicken rinse samples and pure culture.

2. Materials and methods

2.1. Bacterial strains and preparation of cell suspensions Three S. enteritidis strains, two nalidixic acid sensitive (A600-P1-1.625, Health of Animals Laboratory or HAL, Health Canada, Guelph, Canada, and 13076, American Tissue Culture Collection or ATCC, Manassa, USA) and one nalidixic acid resistant (SE257(517N), HAL), and three S. typhimurium strains, two nalidixic acid sensitive (S3392, HAL, and LN 203688, Queen Mary Hospital, Hong Kong) and one nalidixic acid resistant (S324, HAL), were used. Strains, A600-P1-1.625, S3392, SE257(517N) and S324, were used in most of the experiments except those described in section 2.6. Following overnight incubation on brilliant green sulfa (BGS) agar plates (Difco, Ottawa, Canada) at 358C, cells were harvested, suspended in sterile saline, and the concentration adjusted to the density of a MacFarland nephelometer tube [5 (approximately 1.5x10 9 CFU / ml) (Hendrickson and Krenz, 1991). Ten fold serial dilutions of the suspensions from 10 9 to 10 1 CFU / ml were prepared in sterile saline. The bacterial concentrations were also determined by plating 100 ml of appropriate dilutions of the cell suspensions on BGS agar in duplicate, incubating overnight at 358C, and then counting the number of CFU.

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

2.2. Chicken rinse preparation Randomly selected chicken parts (wings and legs) were purchased from local food stores. Chicken parts were weighed and placed aseptically in individual sterile plastic bags with nutrient broth (NB) (Difco, Ottawa, Canada) (1 ml / g of poultry parts). The parts were shaken vigorously for 30 seconds. The rinses were then collected and tested for the presence of Salmonella by the culture method of Lammerding et al. (1988). The rinse samples were stored at 2 208C. Only Salmonella-negative rinses were used for spiking in the experimental procedures described below.

2.3. Enrichment procedures Two enrichment procedures, BAM and EI, were investigated for both pure cultures and mixed cultures using spiked chicken rinses. In the BAM procedure, pre-enrichment and selective enrichment were used. Salmonella at initial concentrations of approximately 0.1, 1 or 10 CFU / ml were pre-enriched in 10 ml of NB. After incubation for 16–18 h at 358C, the above cultures were selectively enriched by transferring 1 ml into 9 ml of tetrathionate brilliant green (TBG) broth (Difco, Ottawa, Canada) or 0.1 ml into 10 ml of Rappaport–Vassiliadis (RV) broth (Oxoid, Ottawa, Canada), and incubated for 24 h at 428C. In the EI procedure, pre-enrichment, selective enrichment, and post-enrichment were used. Salmonella at initial concentrations of approximately 0.1, 1 and 10 CFU / ml were pre-enriched in 10 ml of NB for 6 h at 358C. The cultures were then selectively enriched by transferring 1 ml of the above culture into 9 ml of TBG or 0.1 ml into 10 ml of RV. After 16 h incubation at 428C, the above cultures were post-enriched by transferring 1 ml of the culture into 9 ml of NB and incubating for 4 h at 358C. In addition, the effects of 6 and 8 h pre-enrichment in NB and the effects of 1, 2, 4 or 6 h post-enrichment in NB or Brain Heart Infusion (BHI) broth (Difco, Detroit, USA) after 18 h selective enrichment in TBG or RV were compared. The bacterial concentrations after various culture steps in each experiment were determined by dilution plating. The cultures were diluted in saline, and 100 ml was spread in duplicate on BGS (Difco) plates for

87

pure cultures using nalidixic acid sensitive strains. In order to determine the growth of the nalidixic acid resistant strains of Salmonella spiked in the mixed cultures using chicken rinses, 100 ml of the mixed cultures diluted as above were spread on the BGS plates supplemented with nalidixic acid (25 g / ml, Sigma-Aldrich, Oakville, Canada). After overnight incubation at 358C, the number of CFU were determined.

2.4. Colony blot ELISA To confirm that the bacterial colonies on the plates from both pure cultures and spiked chicken rinses were Salmonella, the colony blot ELISA was conducted. The method was based on the one described by Nielsen et al. (1996) with modification. Briefly, the colonies on the plates were transferred to nitrocellulose membranes (0.45 mm pore size) (Millipore Corporation, Bedford, USA) by allowing the membranes to remain in contact with the colonies until the membranes were completely wet. Then, the membranes were lifted, air dried in glass petri dishes, and immersed in chloroform (Anachemia, Rouses Point, USA) for 15 min and air dried. The membranes were then incubated for 45 min at room temperature with blocking-digestion buffer containing 0.15 M NaCl, 0.02 M Tris, 0.3% Tween-20 (Sigma-Aldrich, Oakville, Canada), 80 mg / ml lysozyme (Sigma-Aldrich) and 1% casein (Fisher Scientific, Nepean, Canada) (Hoszowski et al. 1996). After washing with PBS (0.1 M, pH 7.2) containing 0.3% Tween 20, ascites fluid of monoclonal antibody (Mab), M04 against serogroup B including S. typhimurium (Tsang et al., 1991) or M09 against serogroup D1 including S. enteritidis (Lu et al., 1991) was diluted 1:800 in PBS, applied to the membranes with homogenous antigen, and incubated at room temperature for 1 h with constant shaking. As a control for the specificity of the Mabs, M04 and M09 were also applied to membranes with S. enteritidis and S. typhimurium respectively. Following two washes with PBS containing 0.3% Tween 20 (PBST), the membranes were incubated with the goat anti-mouse IgG antiserum conjugated with peroxidase (1:1000 in PBST) (Jackson ImmunoResearch Laboratories, Inc., distributed by Bio / Can Scientific, Mississauga, Canada) for 45 min at room

88

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

temperature. After two washes with PBST, the membranes were incubated with 0.6 mg / ml of 4chloro-1-naphthol (Sigma-Aldrich, Oakville, Canada) and 0.3% H 2 O 2 (Fisher Scientific Ltd., Nepean, Canada) in the dark and the reaction was stopped by washing with distilled water.

media at the same time point. The effect was considered significant at P , 0.05.

3. Results

3.1. Colony blot ELISA 2.5. Microtiter plate ELISA The ELISA for detection of Salmonella was performed using a commercial capture ELISA kit, SinocloneE Salmonella Test 1 (Sinoclone Ltd, Hong Kong). This kit used Mab T6 for both capture and detection of antigen. Monoclonal antibody T6 is specific for the core region of Salmonella LPS (Tsang et al., 1987) and reacts with approximately 98% of all naturally occurring Salmonella known to affect humans, with a detection limit between 10 5 to 10 7 CFU / ml (Ng et al., 1996). The samples were prepared and tested according to the instructions provided with the kit.

2.6. Investigation of the effects of different culture media on the detection of salmonella by ELISA Based on the observations in the present study that the organisms grown in different culture media showed different results in ELISA, we investigated the effects of different culture media on the detection of Salmonella by ELISA. S. typhimurium (LN 203688) and S. enteritidis (ATCC 13076) were grown in BHI and RV for 48 h. The cultures were then boiled in the presence of 10 mM EDTA for 15 min. The bacterial pellets were washed in PBS containing 10 mM EDTA and 0.5% formaldehyde by centrifugation, twice for BHI cultures and three times for RV cultures. After the final wash in PBS containing 10 mM EDTA, three equal portions from each culture were suspended in either RV, BHI or PBS to their original concentration followed by further 1:5 serial dilutions in PBS containing 10 mM EDTA. The bacterial suspensions were tested by ELISA (Sinoclone Ltd).

M04 and M09 specifically detected S. typhimurium and S. enteritidis respectively (Fig. 1). The Colony Blot ELISA confirmed that the colonies on the plates inoculated with pure cultures were the expected S. enteritidis or S. typhimurium. The culture plates inoculated with mixed cultures using spiked chicken rinses showed mixed bacterial colonies, only the typical Salmonella colonies confirmed by Colony Blot ELISA were counted.

3.2. Bacterial growth and ELISA results 3.2.1. BAM procedure After pre-enrichment for 16 h in NB, pure cultures of both S. enteritidis and S. typhimurium increased from initial concentrations of 0.3, 3 and 30 CFU / ml to approximately 10 8 CFU / ml (Table 1). After selective enrichment, both S. enteritidis and S. typhimurium increased to approximately 10 7 to 10 8 CFU / ml in TBG, and above 10 8 CFU / ml in RV which was significantly (P , 0.01) higher than those in TBG (Table 1). S. typhimurium in spiked chicken

2.7. Statistical analysis Student’s t-test (GraphPad Instat, 1993, GraphPad Software, San Diego, USA) was used to compare pooled data from cultures using various culture

Fig. 1. Colony Blot ELISA. Specificities of M04 and M09 Mabs. Mab M09 detected S. enteritidis (A), but not S. typhimurium (B); Mab M04 detected S. typhimurium (C), but not S. enteritidis (D).

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

89

Table 1 Quantitation of S. typhimurium and S. enteritidis by Standard Method in Bacteriological Analytical Manual (BAM) Culture a

Pre-enrichment in NB b CFU / ml c

S. typhimurium Pure culture

Spiked rinse

S. enteritidis Pure culture

Selective Enrichment in TBG

Selective Enrichment in RV

ELISAd

CFU / ml

ELISA

CFU / ml

ELISA

Initial

Final

Final

Initial

Final

Final

Initial

Final

Final

2 0.4 0.6 1.6 2 0.7 0.3 1.3

8.3 8.3 8.4 7.7 8.0 8.0

1.1 1.1 1.0 0.2 0.3 0.4

7.2 7.2 7.4 6.7 6.9 7.0

7.9 8.0 7.6 7.2 7.2 7.3

1.2 0.9 1.2 1.0 1.7 1.3

6.3 6.3 6.4 5.7 6.0 6.0

8.3 8.1 8.3 6.8 7.0 7.4

0.3 0.2 0.2 0.2 0.2 0.2

2 0.5 0.5 1.5

8.3 8.4 8.4

1.0 0.7 0.8

7.2 7.3 7.3

7.8 7.9 8.0

0.6 0.9 1.7

6.9 6.4 6.4

8.1 8.0 8.0

0.3 0.5 0.4

a

Bacteria used for pure culture in NB was nalidixic acid sensitive strain, for chicken rinse with NB was nalidixic acid resistant strain. Bacteria were grown in NB (pure culture) or chicken rinse in NB (spiked rinse) for 16 h, then transferred to TBG or RV for further 16 h incubation. c Numbers represent the mean CFU / ml (Log 10 ). d ELISA O.D.; cutoff value is 0.259. b

rinses increased from initial concentrations of 0.18, 1.8 and 18 CFU / ml to approximately 10 8 CFU / ml after pre-enrichment in NB and approximately above 10 7 CFU / ml after selective enrichment in both TBG and RV (Table 1). The numbers of Salmonella in spiked chicken rinses were approximately 2–4 fold lower (P , 0.001) than those in pure culture after both pre-enrichment and selective enrichment (Table 1). Using the cutoff value of 0.259 as calculated according to the instruction with the kit, ELISA results (Table 1) were positive for both S. enteritidis and S. typhimurium pre-enriched in NB after 16 h incubation (O.D. approximately 1.0), and positive for S. typhimurium pre-enriched in chicken rinse (O.D. approximately 0.3). Selective enrichment in TBG generally further increased the ELISA values in both pure culture and the culture using spiked chicken rinse without increasing the numbers of CFU / ml. However, selective enrichment in RV greatly decreased the ELISA values to generally negative for S. enteritidis although total Salmonella numbers were higher (P , 0.01) than those in TBG (Table 1).

3.2.2. EI procedure Both S. enteritidis and S. typhimurium showed similar results with the EI procedure and only the results for S. typhimurium are shown (Table 2). S. typhimurium in initial concentrations of 0.35, 3.5 and

35 CFU / ml increased to approximately 10 2 , 10 3 and 10 4 CFU / ml respectively after 6 h pre-enrichment in NB, and then increased to 10 8 CFU / ml after 16 h selective enrichment in RV (Table 2). There was only a 4 fold increase to approximately 10 7 CFU / ml after transferring to NB for 4 h post-enrichment (Table 2). The nalidixic acid resistant strain of S. typhimurium multiplied from initial concentrations of 0.18, 1.8 and 18 CFU / ml to approximately 10 2 , 10 2 and 10 3 respectively in chicken rinse with NB for 6 h pre-enrichment, and then increased to 10 6 , 10 7 , 10 7 CFU / ml respectively after 16 h selective enrichment in RV (Table 2). After 4 h post-enrichment in NB, there was about a 10 fold increase to approximately 10 7 CFU / ml (Table 2). The ELISA results showed that 6 h pre-enrichment in NB gave positive results with pure culture, but not with chicken rinses. However, the ELISA results were negative when the bacteria were further cultured in RV for 16 h selective enrichment and subsequently cultured in NB for 4 h post-enrichment. An experiment was performed to compare the bacterial growth rates in NB for 6 and 8 h for pre-enrichment. The experiment (Table 3) showed that S. enteritidis and S. typhimurium with initial concentrations of 0.15, 1.5 and 15, or 0.35, 3.5 and 35 CFU / ml respectively were increased to approximately 10 2 to 10 4 CFU / ml in 6 h, and over 10 4

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

90

Table 2 Quantitation of S. typhimurium by enrichment-immunoassay procedure Culture a

Pre-enrichment in NB b CFU / ml c

Pure culture

Spiked rinse

Selective Enrichment in RV

Post-enrichment in NB

ELISAd

CFU / ml

ELISA

CFU / ml

ELISA

Initial

Final

Final

Initial

Final

Final

Initial

Final

Final

2 0.5 0.5 1.5 2 0.7 0.3 1.3

2.3 3.4 4.7 2 2.5 3.6

0.2 0.3 0.5 0.2 0.2 0.4

0.3 1.4 2.8 0 0.5 1.6

8.4 8.2 8.1 6.6 7.0 7.7

0.2 0.2 0.2 0.1 0.1 0.2

7.4 7.2 7.1 5.6 5.9 6.7

7.8 7.8 7.8 6.9 7.0 8.0

0.2 0.2 0.6 0.1 0.1 0.1

a

Bacteria used for pure culture in NB was nalidixic acid sensitive strain, for chicken rinse with NB was nalidixic acid resistant strain. Bacteria were grown in NB (pure culture) or chicken rinse in NB (spiked rinse) for 6 h, then transferred to TBG or RV for 16 h incubation followed with further 4 h incubation in NB. c Bacteria used for pure culture in NB was nalidixic acid sensitive strain, for chicken rinse with NB was nalidixic acid resistant strain. d ELISA O.D.; cutoff value is 0.259. b

Table 3 Comparison of S. typhimurium and S. enteritidis growth after 6 and 8 h pre-enrichment in the EI procedure Culture a

S. typhimurium Pure culture

Spiked rinse

S. enteritidis Pure culture

Spiked rinse

6h

8h

CFU / ml

ELISAc

CFU

ELISA

Initial

Final

Final

Initial

Final

Final

2 0.6 b 0.4 1.4 2 0.1 0.9 1.9

2.7 3.7 4.6 0 2.8 3.5

0.3 0.2 0.2 0.5 0.5 0.5

2 0.6 0.4 1.4 2 0.1 0.9 1.9

4.4 4.8 5.3 0 4.0 5.1

0.2 0.3 0.7 0.3 0.4 0.5

2 0.4 0.6 1.6 2 0.7 0.3 1.3

4.1 3.3 4.6 –d 1.8 2.7

0.3 0.2 0.2 0.3 0.4 0.6

2 0.4 0.6 1.6 2 0.7 0.3 1.3

4.8 4.2 5.4 – 2.9 3.2

0.2 0.2 0.2 0.3 0.3 0.3

a

Bacteria used for pure culture in NB was nalidixic acid sensitive strain, for chicken rinse with NB was nalidixic acid resistant strain. Numbers represent the mean CFU / ml (Log 10 ). c ELISA O.D.; cutoff value is 0.6. d No colonies were observed on the plate. b

CFU / ml in 8 h in pure culture. The nalidixic acid resistant strains of S. enteritidis and S. typhimurium in chicken rinse were increased from 0.2, 2 and 20 or 0.8, 8 and 80 CFU / ml respectively to approximately 10 2 CFU / ml in 6 h and 10 2 to10 4 CFU / ml in 8 h. Due to the negative ELISA results in RV (16 h) and NB (4 h) after transferring from RV, an experi-

ment was conducted to determine the effects of post-enrichment incubation time (1, 2, 4 and 6 h), and types of media for selective enrichment (RV and TBG) and post-enrichment (NB or BHI) on bacterial concentrations and ELISA results. Table 4 shows that for both S. enteritidis and S. typhimurium, the longer the incubation time, the higher the bacterial

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

91

Table 4 Comparison of S. typhimurium and S. enteritidis concentrations with various incubation times and media in the post-enrichment of the EI procedure Culture a

Selective / Post-enrichment

Post-enrichment Incubation Time (h) 0 CFU

S. typhimurium Pure culture Spiked rinse Pure culture Spiked rinse Pure culture Spiked rinse Pure culture Spiked rinse S. enteritidis Pure culture Spiked rinse Pure culture Spiked rinse Pure culture Spiked rinse Pure culture Spiked rinse

RV/ BHI TBG / BHI RV/ NB TBG / NB

RV/ BHI TBG / BHI RV/ NB TBG / NB

2 b

c

4

6

ELISA

CFU

ELISA

CFU

ELISA

CFU

ELISA

6.8 7.0 6.7 6.9 6.8 7.0 6.7 6.9

0.4 0.4 0.5 0.5 0.3 0.3 0.6 0.9

7.5 7.2 7.4 8.3 7.3 8.0 7.0 7.9

0.3 0.5 0.5 0.5 0.4 0.3 0.8 1.2

8.5 8.8 7.6 8.5 7.8 8.0 7.6 8.0

0.5 0.7 0.6 0.9 0.4 0.4 0.9 1.1

9.2 9.0 9.3 8.8 7.9 8.2 8.0 8.3

0.6 0.8 1.4 1.6 0.6 0.4 1.2 1.0

6.7 7.1 6.9 7.0 6.7 7.1 6.9 6.9

0.3 0.3 0.5 0.4 0.3 0.3 0.5 0.5

7.4 8.2 7.3 8.2 7.3 8.0 6.9 7.5

0.3 0.3 0.4 0.3 0.3 0.2 0.4 1.2

8.2 8.7 8.2 8.5 7.1 8.0 8.1 8.2

0.5 0.8 0.9 0.7 0.4 0.4 1.0 0.9

9.1 8.9 9.1 8.9 8.2 8.4 8.3 8.3

0.7 0.9 1.0 1.2 0.4 0.4 0.9 0.8

a Bacteria were grown in NB (pure culture) or chicken rinse in NB (spiked rinse) for 6 h, then transferred to RV or TBG selective medium for 16 h incubation followed with further 1, 2, 4 and 6 h incubation in NB or BHI post-enrichment medium. b Numbers represent CFU / ml (Log 10 ) from one determination. c ELISA O.D.; cutoff value is 0.6.

concentration. The ELISA titers from cultures postenriched in NB or BHI were higher (P , 0.01) when TBG as compared to RV was used for selective enrichment. Overall, post-enrichment in BHI after selective enrichment in either TBG or RV yielded higher numbers of Salmonella and higher ELISA titers than those in NB (P , 0.05).

3.3. Effects of RV medium on ELISA An experiment was carried out to investigate why the ELISA showed low O.D. for the cultures in RV medium or in NB after transfer from RV. In this experiment, the 48 h cultures in RV (3x10 8 CFU / ml) or BHI (12x10 8 CFU / ml) were washed and resuspended in either PBS, RV or BHI broth to the original concentration followed by subsequent 1:5 dilutions in PBS. Table 5 shows that after washing and resuspending in PBS, BHI or RV, all of the samples from RV culture showed much lower O.D. than those of the cultures from BHI. There was no

significant difference in O.D. between the samples resuspended in PBS, BHI and RV and further diluted in PBS.

4. Discussion In general, the limit of ELISAs for detecting Salmonella spp. has been reported to be approximately 10 4 to 10 7 CFU / ml Salmonella in the enrichment broth, thus the knowledge of the growth of Salmonella in different media at different culture stages (pre-enrichment, selective enrichment and post-enrichment) is useful for designing efficient enrichment. The present study has shown that S. enteritidis and S. typhimurium could increase from 0.18 CFU / ml to over 10 7 CFU / ml in pure culture media and to approximately 10 6 CFU / ml in spiked chicken rinse by the BAM and EI procedures. At these levels, Salmonella can be detected by ELISAs for Salmonella (Mattingly and Gehle, 1984; Prusak-

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

92

Table 5 Comparison of the effects of RV and BHI broth on ELISA Dilution a

S. enteritidis 1/1 1/5 1 / 25 1 / 125 1 / 625 Blank c S. typhimurium 1/1 1/5 1 / 25 1 / 125 1 / 625 Blank

RV culture

BHI culture

PBS

BHI

RV

PBS

BHI

RV

0.141 b 0.098 0.113 0.111 0.104 0.088

0.092 0.099 0.112 0.112 0.105 0.086

0.090 0.162 0.148 0.115 0.109 0.060

0.864 0.363 0.139 0.113 0.112

0.730 0.369 0.127 0.114 0.103

0.785 0.370 0.145 0.101 0.093

0.432 0.279 0.165 0.093 0.071 0.085

0.267 0.246 0.165 0.098 0.074 0.067

0.194 0.425 0.228 0.098 0.075 0.054

1.456 0.508 0.290 0.138 0.094

0.741 0.330 0.185 0.105 0.083

2.324 over 0.785 0.091 0.088

a

The cultures in RV or BHI were resuspended in PBS, BHI or RV after washing, and further 1:5 serially diluted in PBS. The value represents ELISA O.D. c Blank was the control without bacteria. b

Sochaczewski and Luong, 1989; Lee et al., 1990; Wyatt et al., 1993; Choi et al., 1992; Ng et al., 1996). The present study has shown that Salmonella grew to various concentrations by different pre-enrichment and post-enrichment time, and in the different selective and enrichment media used (Tables 1–3). However, S. enteritidis and S. typhimurium increased up to similar concentrations in BAM and EI procedures by the end of incubation procedure (Tables 1 and 2). By changing the medium from NB to BHI for post-enrichment in the EI procedure, it was possible to further increase the bacterial concentration to10 9 CFU / ml (Table 4). The present study also shows that the media used for culture have a significant influence on ELISA results. In this study, the numbers of Salmonella selectively enriched in RV broth following pre-enrichment in spiked chicken rinse were similar to that in TBG, and approximately 10 fold higher following pre-enrichment in pure culture than that in TBG broth (Table 1). However, ELISA titers were lower or even negative with RV samples than those in TBG samples, and remained negative despite a further 4 to 6 h post-enrichment in NB (Tables 2 and 4). Postenrichment in BHI after transferring from RV significantly improved the ELISA detection limit when compared with NB (Table 4), which is in agreement with the results of Ng et al. (1996). Wyatt et al.

(1993) also found that culture in RV media reduced the ELISA sensitivity compared with a chemically defined medium, which is thought due to the reduced expression of the antigens rather than reduction in the numbers of bacteria. The present study has also shown that after washing and resuspending in PBS, BHI or RV, all of the samples from RV culture showed much lower O.D. than those of the cultures from BHI, and there was no significant difference of O.D. between the samples diluted in PBS, BHI and RV (Table 5). These results indicate that component(s) in RV broth may not directly inhibit the ELISA. Therefore it is most likely that component(s) in RV broth may inhibit the expression of the antigenic epitope(s) in LPS detected by the capture ELISA. Further identification of the substance(s) in RV broth which may influence ELISA will be useful to enhance ELISA sensitivity for RV samples. In the present study, S. enteritidis and S. typhimurium grew in pure culture to 10 8 CFU / ml after 16 h pre-enrichment in NB in the BAM procedure or after 6 h pre-enrichment in NB followed by 16 h selective enrichment in RV in the EI procedure. However, the numbers of bacteria did not significantly increase after transferring to TBG or RV for 16 h selective enrichment in the BAM procedure or to NB for 4 to 6 h post-enrichment in the EI procedure (Tables 1 and 2). S. typhimurium in

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

chicken rinse also showed similar results (Table 1). However, the ELISA showed higher O.D. when TBG was used for selective enrichment as compared to NB used for pre-enrichment in the BAM procedure although the bacteria concentrations were similar. It has also been shown that post-enrichment is important for increasing the limit of ELISA (Ng et al., 1996). Thus the post-enrichment may enhance the expression of specific antigens without significantly increasing the numbers of bacteria. The present study has shown that in the BAM procedure, the numbers of Salmonella in RV broth were more than that in TBG for selective enrichment. Thus, RV may be a better selective enrichment for conventional differentiating agar because more Salmonella colonies will be present and less chance to be missed. When used for ELISA, TBG broth showed more positive results than RV. Therefore, different selective media could be used for different purposes. However, a combination of RV as selective enrichment and BHI as post-enrichment would detect more species of Salmonella than the combination of TBG / BHI (Ng et al., 1996; Ng, pers. comm.). A combination of RV as selective enrichment with an appropriate post-enrichment medium such as BHI may be used for both conventional cultural isolation and identification, as well as ELISA. Therefore, the EI procedure, which utilizes less time than the BAM procedure and a combination of selective enrichment and post-enrichment, may have advantages in terms of time and sensitivity. In conclusion, the present study has shown that several Salmonella could increase to 10 8 CFU / ml in pure culture media for both culture procedures used in this study, with several fold lower levels in spiked chicken rinse cultures. These concentration levels are above the detection limits of most ELISA tests reported to date. It has also been shown that different media affect the growth of Salmonella and ELISA. The results obtained in this study may help in designing an appropriate culture procedure for different ELISAs.

Acknowledgements Dr. Huang was a recipient of a Natural Sciences and Engineering Research Council of Canada Visiting Fellowship. Sincere thanks also go to Sinoclone

93

Ltd. for financial support and to Cheryl Lutze-Wallace and Omar Abubakar, for laboratory help.

References Andrews, W.H., Bruce, V.R., June, G., Satchell, F., Sherrod, P., 1992. Chapter 5. Salmonella. In: F.D.A. Bacteriological Analytical Manual, 7th edn, Association of Official Analytical Chemists, Arlington, USA, pp. 51–69. Choi, D., Tsang, R.S.W., Ng, M.H., 1992. Sandwich capture ELISA by a murine monoclonal antibody against a genusspecific LPS epitope for the detection of different common serotypes of salmonellas. J. Appl. Bacteriol. 72, 134–138. D’Aoust, J.-Y., 1985. Infective dose of Salmonella typhimurium in cheddar cheese. Am. J. Epidemiol. 122, 717–720. D’Aoust, J.-Y., 1997. In: Doyle, M.P., Beuchat, L.R., Montville, T.J. (Eds.), Food Microbiology, Fundamentals and Frontiers. Salmonella species, ASM Press, Washington D.C, pp. 129– 158. D’Aoust, J.Y., Aris, B.J., Thisdele, P., Durante, A., Brisson, N., Dragon, D., Lachapelle, G., Johnston, M., Laidley, R., 1975. Salmonella eastbourne outbreak associated with chocolate. J. Inst. Can. Sci. Technol. Alim. 8, 181–184. Feng, P., 1992. Appendix 1. Rapid methods for detecting foodborne pathogens. In: F.D.A. Bacteriological Analytical Manual, 7th edn, Association of Official Analytical Chemists, Arlington, USA. Fontaine, R.E., Arnon, S., Martin, W.T., Vernon, Jr. T.M., Gangarosa, E.J., Farmer, J.J., Moran, A.B., Silliker, J.H., Decker, D.L., 1978. Raw hamburger: an interstate common source of human salmonellosis. Am. J. Epidemiol. 107, 36–45. Hendrickson, D.A., Krenz, M.M., 1991. Reagents and stains. In: Balows, A., Hausler, W.J., Herrmann, K.L., Isenberg, H.D., Shadowy, H.J. (Eds.), Manual of Clinical Microbiology, American Society for Microbiology, Washington, D.C, pp. 1289– 1314. Hoszowski, A., Fraser, A.D.E., Brooks, B.W., Riche, E.M E.M., 1996. Rapid detection and enumeration of Salmonella in chicken carcass rinses using filtration, enrichment and colony blot immunoassay. Int. J. Food Microbiol. 28, 341–350. Jay, L.S., Comar, D., 1988. Comparative study of TECRAESalmonella visual immunoassay and Australian standard cultural methods for analysis of salmonellae in foods. Food Technol. Aust. 40, 186–191. Lammerding, A.M., Garcia, M.M., Mann, E.D., Robinson, Y., Dorward, W.J., Truscott, R.B., Tittiger, F., 1988. Prevalence of Salmonella and thermophilic Campylobacter in fresh pork, beef, veal and poultry in Canada. J. Food Prot. 51, 47–52. Lee, H.A., Wyatt, G.M., Bramham, S., Morgan, M.R.A., 1990. Enzyme-linked immunosorbent assay for Salmonella typhimurium in food: feasibility of 1-day Salmonella detection. Appl. Environ. Microbiol. 56, 1541–1546. Lipson, A., 1976. Infectious dose of Salmonella. Lancet. 1, 969. Lu, G.Z., Tsang, R.S.W., Chau, P.Y., Choi, D., Law, D., Ng, M.H., 1991. Characterization of murine monoclonal antibodies that react specifically with the serogroup D1 Salmonella. FEMS Microbiol. Lett. 80, 135–140.

94

H. Huang et al. / International Journal of Food Microbiology 51 (1999) 85 – 94

Mattingly, J.A., Gehle, W.D., 1984. An improved enzyme immunoassay for the detection of Salmonella. J. Food Sci. 49, 807–809. Ng, S.P., Tsui, C.O., Roberts, D., Chau, P.Y., Ng, M.H., 1996. Detection and serogroup differentiation of Salmonella spp. in food within 30 hours by enrichment-immunoassay with a T6 monoclonal antibody capture enzyme-linked immunosorbent assay. Appl. Environ. Microbiol. 62, 2294–2302. Ng, S.P., Li, Z.G., Chen, B.W., Qin, Z.K., Garcia, M.M., IM, S.W.K., Ng, M.H., 1997. Rapid screening for Salmonella in fishmeal by enrichment-immunoassay. J. Rap. Meth. Auto. Microbiol. 5, 285–294. Nielsen, K., Gall, D., Kelly, W., Vigliocco, A., Henning, D., Garcia, M., 1996. Immunoassay development. Application to enzyme immunoassay for the diagnosis of Brucellosis, Agriculture and Agri-Food Canada, Canada. Prusak-Sochaczewski, E., Luong, J.H.T., 1989. An improved

ELISA method for the detection of Salmonella typhimurium. J. Appl. Bacteriol. 66, 127–135. SinocloneE Salmonella Test 1 & 2, ELISA kit, Instruction, Sinoclone Ltd, 4th Floor, 131 Wing Lok Street, Hong Kong. Tsang, R.S.W., Chan, K.H., Chau, P.Y., Wan, K.C., Ng, M.H., Schlecht, S., 1987. A murine monoclonal antibody specific for the outer core oligosaccharide of Salmonella lipopolysaccharide. Infect. Immun. 55, 211–216. Tsang, R.S.W., Chan, K.H., Lau, N.W.H., Choi, D.K.W., Law, D.K.S., Ng, M.H., 1991. Characterization of murine monoclonal antibodies against serogroup B salmonellae and application of serotyping reagents. J. Clin. Microbiol. 29, 1899–1903. Wyatt, G.M., Langley, M.N., Lee, H.A., Morgan, M.R.A., 1993. Further studies on the feasibility of one-day Salmonella detection by enzyme-linked immunosorbent assay. Appl. Environ. Microbiol. 59, 1383–1390.