A Single-Step Enrichment Medium for Nonchromogenic Isolation of Healthy and Cold-Injured Salmonella spp. from Fresh Vegetables

Abstract

Culture-based detection of nontyphoidal Salmonella spp. in foods requires at least four working days; therefore, new detection methods that shorten the test time are needed. In this study, we developed a novel single-step Salmonella enrichment broth, SSE-1, and compared its detection capability with that of commercial single-step ONE broth-Salmonella (OBS) medium and a conventional two-step enrichment method using buffered peptone water and Rappaport-Vassiliadis soy broth (BPW-RVS). Minimally processed lettuce samples were artificially inoculated with low levels of healthy and cold-injured Salmonella Enteritidis (10⁰ or 10¹ colony-forming unit/25 g), incubated in OBS, BPW-RVS, and SSE-1 broths, and streaked on xylose lysine deoxycholate (XLD) agar. Salmonella recoverability was significantly higher in BPW-RVS (79.2%) and SSE-1 (83.3%) compared to OBS (39.3%) (p < 0.05). Our data suggest that the SSE-1 single-step enrichment broth could completely replace two-step enrichment with reduced enrichment time from 48 to 24 h, performing better than commercial single-step enrichment medium in the conventional nonchromogenic Salmonella detection, thus saving time, labor, and cost.

Introduction

Foodborne diseases caused by nontyphoidal Salmonella pose a significant public health problem and worldwide economic burden (Gomez et al., 1997; Scharff, 2012; McLinden et al., 2014). According to the Foodborne Diseases Active Surveillance Network, FoodNet, Salmonella-linked infections account for 15% of foodborne disease incidents, 28% of hospitalizations, and 0.4% of deaths in the U.S. (CDC, 2014). Salmonellosis is most commonly transmitted to humans by consumption of contaminated foods, especially under-cooked products such as fresh vegetables and fruits (Harris et al., 2003).

Novel rapid methods for Salmonella detection have been constantly developed and reported; however, the enrichment step is still necessary to increase cell numbers in the samples with low Salmonella concentration and resuscitate bacteria sublethally injured by inhibitory food components (Taskila et al., 2012). The current standard isolation method for Salmonella involves two enrichment stages, a pre-enrichment step and selective enrichment step, which require two working days before the enriched population can be analyzed for growth on selective media (ISO 6579, 2002; BAM, 2011). To circumvent the long enrichment process, new Salmonella detection methods such as Rapid Salmonella (Bio-Rad, Marnes-la-Coquette, France), IBISA Salmonella (AES Chemunex, BRUZ Cedex, France), and Precis™ Salmonella (Oxoid, Hampshire, United Kingdom), which use single-step enrichment broths and no longer require a secondary enrichment step, have been developed (Margot et al., 2013). However, these commercial methods include the use of self-developed chromogenic agars after the enrichment, which improve selectivity, but are relatively expensive, hampering the routine use of these methods in an average research laboratory. Therefore, it is useful to develop a single-step enrichment broth medium that provides equivalent or better selectivity than conventional or commercial pre-enrichment broths when used with regular Salmonella selective media such as xylose lysine deoxycholate (XLD) agar, but shortens the enrichment time.

In this study, minimally processed lettuces were used to develop a novel single-step enrichment medium named SSE-1, which can replace conventional two-step selective enrichment of Salmonella spp. The effectiveness of the newly developed medium was compared to that of a reference enrichment method based on ISO 6579 (2002) and of the ONE Broth-Salmonella (OBS) medium manufactured by Oxoid Thermo Fisher, using XLD agar.

Materials and Methods

Preparation of the media

The enrichment broths and plating agars used in this study included Rappaport-Vassiliadis soy broth (RVS; bioMérieux, Marcy l'Etoile, France), buffered peptone water (BPW), OBS, XLD, and tryptic soy agar (TSA) (all from Oxoid); they were prepared according to the manufacturers' recommendations. Novobiocin and nalidixic acid were purchased from Oxoid and Sigma (St. Louis, MO), respectively.

Formulation of single-step Salmonella enrichment broth for nonchromogenic detection, SSE-1

The final composition of the SSE-1 medium is presented in Table 1. Potassium phosphate monobasic (KH₂PO₄) and sodium phosphate dibasic dihydrate (Na₂HPO₄) used as buffering agents and sodium chloride (NaCl), an osmotic component, were the same as in the BPW base. d-mannitol and yeast extract, 3,3′-thiodipropionic acid (TDP) and sodium pyruvate, and sodium citrate (all from Sigma) and novobiocin were used as growth substrates, antioxidants, and selective agents, respectively, in the concentrations based on previous studies (Taylor, 1961; Chen et al., 1993; Baylis et al., 2000; Jensen et al., 2003; Gurtler and Kornacki, 2009; Kobayashi et al., 2009; Xiao et al., 2010) and our pilot experiments (data not shown).

Table 1.

Composition of the Novel Medium (SSE-1)

Source	Ingredient	Concentration(g/L)
Buffer (same as BPW)	Potassium phosphate monobasic	1.5
	Sodium phosphate dibasic dihydrate	3.5
	Sodium chloride	5.0
Growth promoters	D-mannitol	5.0
	Yeast extract	6.0
Antioxidants	3,3′-thiodipropionic acid	1.0
	Sodium pyruvate	2.0
Inhibitors	Sodium citrate	10.0
	Novobiocin	0.005

BPW, buffered peptone water.

Bacterial strains

The growth of Salmonella strains in the SSE-1 medium was evaluated using stock cultures of nine Salmonella serovars (Salmonella Heidelberg, Salmonella Infantis, Salmonella Agona, Salmonella Typhimurium, Salmonella Ohio, Salmonella Montevideo, Salmonella Hartford, Salmonella Enteritidis, and Salmonella Newport) obtained from the Food and Drug Administration (FDA, College Park, MD). Nalidixic acid-resistant Salmonella Enteritidis strain no. 106, also obtained from FDA, was used in artificial inoculation tests to exclude false-positive results.

Growth of pure Salmonella cultures in BPW, OBS, and SSE-1

Less than 20 colony-forming unit (CFU) of nine Salmonella strains were suspended in 100 μL of phosphate-buffered saline (PBS, pH 7.4; Sigma), inoculated each into 10 mL of BPW, OBS, or SSE-1, and cultured at 37°C (BPW) or 42°C (OBS and SSE-1). To determine maximal growth capacity, stationary-phase cultures enriched overnight were 10-fold serially diluted, and 0.1 mL of each dilution was spread on XLD agar plates in duplicate and incubated for 20 ± 2 at 37°C followed by colony counting.

Detection of Salmonella Enteritidis in artificially contaminated lettuce

Minimally processed lettuces were purchased from a local supermarket and used for the experiments within 1 h. To determine the Salmonella recoverability in cases of low-level contamination, overnight cultures of nalidixic acid-resistant Salmonella Enteritidis were 10-fold serially diluted in PBS and a 1-mL aliquot of bacterial suspension was inoculated at a low level of 10⁰ or 10¹ CFU/25 g on the surface of lettuces, which were then placed in stomacher bags at 7°C overnight to simulate lettuce storage conditions. After that, 225 mL of each enrichment broth was poured into the stomacher bags containing inoculated lettuces, blended for 1 min, and incubated for 20 ± 2 h at 37°C (BPW) or 42°C (OBS and SSE-1). A loopful of each OBS and SSE-1 enrichment culture was streaked onto XLD agar and incubated for 24 h at 37°C. In case of BPW enrichment cultures, 0.1 mL was transferred into 10 mL of RVS broth, and incubated for 24 h at 42°C; a loopful of RVS broth was then streaked onto XLD agar and incubated for 24 h at 37°C. A maximum of five suspected colonies grown on XLD agar were further subcultured on XLD plates containing 200 ppm nalidixic acid to exclude false-positive results (Gurtler and Kornacki, 2009). The identity of Salmonella Enteritidis colonies was then confirmed by real-time PCR as described by O'Regan et al. (2008).

In addition, to compare medium performance with sublethally injured bacteria, the 10-fold diluted overnight cultures (ca. 10⁷ CFU/mL) were treated at 4°C for 24 h to achieve ∼50% injured cells. Prepared cold-injured cells were inoculated and tested as described above. The rate of sublethal injury was determined by comparing bacterial counts on TSA as nonselective agar to those on XLD selective agar (Uyttendaele et al., 2008).

Data analysis

The SPSS software (version 19.0; SPSS, Inc., Chicago, IL) was used for statistical analysis. The differences in the number of Salmonella cells recovered from three different broths were assessed using a t-test. For the comparison of recoverability and selectivity, the Fisher's exact test was used. The selectivity was evaluated based on the growth index of competing microflora (1, growth of a few colonies; 2, growth of colonies approximately half of the plates; and 3, growth on most of the plate), according to our previous study (Chon et al., 2013). The growth index was determined only for the medium contaminated with competing microflora. The difference at the probability level less than 0.05 (p < 0.05) was considered statistically significant.

Results and Discussion

Although antibiotics could suppress overgrowth of the background microflora, it has been reported that antibiotic addition at the early stage of enrichment can compromise the recovery of Salmonella cells (Chen et al., 1993). To increase the proportion of Salmonella cells in the SSE-1 medium, we focused on the optimization of its composition by using the constituents that provide Salmonella recovery and selective enrichment with minimal antibiotic concentration.

In this study, we used only OBS as a commercial control because there was no difference in performance among three commercially available single-enrichment Salmonella detection methods described in the Introduction section (Margot et al., 2013). In addition, it is a newly designed complete selective enrichment medium, similar to SSE-1 developed in this study, while other two broths are modifications of BPW.

The buffering agents formulated in the SSE-1 broth were the same as in the BPW medium manufactured by Oxoid, which has been proven beneficial for the recovery of injured Salmonella cells (Baylis et al., 2000). Both peptone and yeast extract support microbial growth, but it has been reported that 0.6% yeast extract can recover injured Salmonella cells (Gurtler and Kornacki, 2009) and prevent pH decrease due to the growth of Bacillus cereus, which frequently contaminate unprocessed food, especially vegetables (Kobayashi et al., 2009); therefore, we replaced peptone with yeast extract. TDP and sodium pyruvate were used as antioxidants or reducing agents. Gurtler and Kornacki (2009) reported that TDP (0.5–1 g/L) and sodium pyruvate (1.0–3.5 g/L) efficiently neutralize or prevent the formation of reactive oxygen species known to injure bacterial cells and hamper their resuscitation.

In previous studies, mannitol has been used as a carbon source for selective enrichment of Salmonella (Taylor, 1961), and it has been reported that 1.25–5.0 g/L mannitol can promote Salmonella growth (Xiao et al., 2010). Sodium citrate used in some selective media such as Gram-negative (GN) broth and deoxycholate citrate agar, not only inhibits competing microflora but also, at the concentration of 10 g/L, slightly promotes Salmonella growth (Xiao et al., 2010).

The optimal concentration of novobiocin, the only antibiotic in the SSE-1 medium, was determined in our pilot study, in which antibiotic-free SSE-1 could not recover Salmonella artificially contaminated in lettuce samples (data not shown). The concentration of novobiocin was minimized to 5.0 mg/L, which is 2.4 and 4.4 times lower than that in OBS and modified semisolid RVS (Jensen et al., 2003), respectively.

Nine different Salmonella serotypes were analyzed for cell numbers after the enrichment in three different media: BPW, OBS, and the newly developed single-step enrichment broth SSE-1 (Table 2). BPW showed significantly (p < 0.05) higher average numbers (8.98 ± 0.09 log CFU/mL) than SSE-1 (8.68 ± 0.17 log CFU/mL) and OBS (8.60 ± 0.25 log CFU/mL). Although cell numbers in the SSE-1 broth were lower than those in the nonselective pre-enrichment BPW broth, all the tested strains displayed even growth capacity above 8 log CFU/mL, and the capability of Salmonella growth did not differ significantly between SSE-1 and OBS. These results indicate that the SSE-1 medium may be suitable as an enrichment broth for the detection of Salmonella spp..

Table 2.

Growth of Pure Cultures of Nine Salmonella Strains in Three Enrichment Media

	Salmonella growth after 24-h enrichment (log CFU/mL)
Strain	BPW	SSE-1	OBS
Salmonella Heidelberg	8.98 ± 0.11	8.76 ± 0.13	8.52 ± 0.37
Salmonella Infantis	9.08 ± 0.06	8.54 ± 0.36	8.62 ± 0.29
Salmonella Agona	9.01 ± 0.03	8.80 ± 0.04	8.72 ± 0.09
Salmonella Typhimurium	8.95 ± 0.07	8.47 ± 0.05	8.33 ± 0.48
Salmonella Ohio	9.02 ± 0.08	8.80 ± 0.08	8.74 ± 0.12
Salmonella Montevideo	9.02 ± 0.03	8.63 ± 0.08	8.61 ± 0.26
Salmonella Hartford	8.87 ± 0.11	8.71 ± 0.04	8.49 ± 0.14
Salmonella Enteritidis	8.95 ± 0.07	8.69 ± 0.09	8.67 ± 0.17
Salmonella Newport	9.01 ± 0.07	8.84 ± 0.14	8.75 ± 0.05
Mean ± SD (log CFU/mL)^a	8.98 ± 0.09 A	8.69 ± 0.17 B	8.60 ± 0.25 B

Different letters (A, B) within a row indicate significant differences (p < 0.05).

BPW, buffered peptone water; OBS, ONE Broth-Salmonella; SD, standard deviation; SSE-1, the new medium developed in this study.

Tables 3 and 4 show the performance of three enrichment methods in combination with XLD selective agar for the isolation of Salmonella Enteritidis from artificially inoculated fresh lettuces. Through the pilot small-scale study performed on four types of vegetable samples such as chicory, red cabbage, parsley, and lettuce, BPW as a single-enrichment broth was excluded because it could not recover target bacteria in all tested samples (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/fpd). Lettuce was selected for further tests because of its highest level of competing microflora.

Table 3.

The Recoverability of Healthy and Cold-Injured Nalidixic Acid-Adapted Salmonella Enteritidis Artificially Inoculated Lettuce Using Three Enrichment Methods

	No. of positive plates/total no. of lettuce samples ^a
Inoculum level (CFU/25 g)	BPW-RVS	SSE-1	OBS
Healthy
10¹	12/12	11/12	7/12
10⁰	9/12	8/12	1/12
Total	21/24 A	19/24 A	8/24 B
50% Injured
10¹	9/12	11/12	8/12
10⁰	8/12	10/12	3/12
Total	17/24 AB	21/24 A	11/24 B

Different letters (A, B) within a row indicate significant differences (p < 0.05).

BPW-RVS, two-step enrichment method using buffered peptone water and Rappaport-Vassiliadis Soy broth; OBS, ONE Broth-Salmonella; SSE-1, the new medium developed in this study.

Table 4.

Comparison of Three Enrichment Methods for the Elimination of Competing Microflora

	BPW-RVS	SSE-1	OBS
No. of plates with competing flora/total number of lettuce tested^a	44/48 A	40/48 A	44/48 A
Average growth index of the competing flora^b	1.44	1.74	2.24

Different letters (A, B) within a row indicate significant differences (p < 0.05) in the number of contaminated plates.

1, growth of a few colonies; 2, growth of colonies on approximately half of the plate; 3, growth on most of the plate.

BPW-RVS, two-step enrichment method using buffered peptone water and Rappaport-Vassiliadis Soy broth; OBS, ONE Broth-Salmonella; SSE-1, the new medium developed in this study.

The recoverability of healthy Salmonella was significantly higher in BPW-RVS and SSE-1 (21 of 24 and 19 of 24, respectively) than in OBS (8 of 24; p < 0.05; Table 3). When lettuce was inoculated with less than 10² CFU/25 g, OBS failed to recover Salmonella in almost half of the samples (5 of 12), while BPW-RVS and SSE-1 exhibited the recovery failure in none or only 1, respectively, of the 12 samples. In the case of lettuce inoculated with less than 10 CFU/25 g, OBS demonstrated the recovery in only 1 of 12 samples, whereas BPW-RVS and SSE-1 recovered the cells in 9 and 8, respectively, of 12 samples (Table 3).

For samples containing ca. 50% of cold-injured cells, SSE-1 (21 of 24) still showed significant difference with OBS (11 of 24; p < 0.05) in the recoverability, while BPW-RVS failed (17 of 24; p > 0.05), although its recoverability was higher compared with OBS (Table 3). As with the results on healthy cells, the performance of OBS (3 of 12) further declined at a lower inoculum level compared to those of BPW-RVS (8 of 12) and SSE-1 (10 of 12); it might result from exposure of target cells to relatively high novobiocin concentration in OBS at an early stage because the growth-inhibitory effect by the stress factors tends to increase at a low concentration of target cells (Taskila et al., 2012). Compared to healthy cells, the number of Salmonella-positive samples decreased on BPW-RVS (from 21 to 17), whereas slightly increased on SSE-1 (from 19 to 21) and OBS (from 8 to 11) despite inoculation of stressed cells, although there was no significant difference in the recoverability between the results of healthy and cold-injured cells in each broth (19 vs. 21 on SSE-1 and 8 vs. 11 on OBS). With the results of healthy cells, it can be concluded that SSE-1 maintained the recovery capacity comparable to that of BPW-RVS even at the lowest inoculum, in sublethal conditions, and after a 1-day shorter incubation period.

Most plates inoculated with the three enriched cultures exhibited growth of typical non-Salmonella colonies (Table 4). Although the average growth index of competing microflora in the SSE-1 broth (1.74) was higher than that in the BPW-RVS two-step enrichment method (1.44), it was lower compared to a similar single-step enrichment broth, OBS (2.24).

It appears that combined effect of constituents, including the antibiotic in the SSE-1 medium, resulted in a better performance compared with OBS by providing a more favorable environment for the inoculated Salmonella cells and sufficiently suppressing the overgrowth of competing microflora (Tables 3 and 4).

Conclusion

This study provides a proof-of-concept application of a single enrichment in combination with nonchromogenic agar for the detection of foodborne pathogens generally requiring two-step enrichment. The SSE-1 single-enrichment broth effectively detected low numbers of Salmonella in lettuce, which is a complicated product for bacterial analysis due to high levels of background microflora (Seow et al., 2012). The data presented in this study suggest that SSE-1 as a complete single-step enrichment broth could replace two-step conventional enrichment, performing better than a commercial single-step enrichment medium in the detection of Salmonella based on conventional nonchromogenic agar; thus, SSE-1 can save time, labor, and cost. However, to commercialize the SSE-1 medium, further optimization of each component concentration is required to provide the effective Salmonella detection, including more serotypes, different stress conditions, and various food types.

Footnotes

Acknowledgments

This work was supported by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries (IPET) through Agriculture, Food and Rural Affairs Research Center Support Program, funded by Ministry of Agriculture, Food and Rural Affairs (MAFRA) (716002-7), by the Export Promotion Technology Development Program of IPET (No. 313010-3) funded by the Ministry for Food, Agriculture, Forestry, and Fisheries, and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2015R1A2A2A05001288).

Disclosure Statement

No competing financial interests exist.

References

Bacteriological Analytical Manual (BAM). 2011. Chapter 5 Salmonella November 2011 version US FDA. Available at: http://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm070149.htm, accessed December 9, 2016 .

Baylis

, MacPhee

, Betts

. Comparison of two commercial preparations of buffered peptone water for the recovery and growth of Salmonella bacteria from foods. J Appl Microbiol, 2000; 89:501–510.

Centers for Disease Control Prevention (CDC). Incidence and trends of infection with pathogens transmitted commonly through food—Foodborne Diseases Active Surveillance Network, 10 U.S. Sites, 2006–2013. MMWR Morb Mortal Wkly Rep, 2014; 63:328–332.

Chen

, Fraser

, Yamazaki

. Evaluation of the toxicity of Salmonella selective media for shortening the enrichment period. Intl J Food Microbiol, 1993; 18:151–159.

Chon

, Kim

, Seo

. Improvement of modified charcoal-cefoperazone-deoxycholate agar by addition of potassium clavulanate for detecting Campylobacter spp. in chicken carcass rinse. Intl J Food Microbiol, 2013; 165:7–10.

Gomez

, Motarjemi

, Miyagawa

, Käferstein

, Stöhr

. Foodborne salmonellosis. World Health Stat Q, 1997; 50:81–89.

Gurtler

, Kornacki

. Comparison of supplements to enhance recovery of heat-injured Salmonella from egg albumen. Lett Appl Microbiol, 2009; 49:503–509.

Harris

, Farber

, Beuchat

, Parish

, Suslow

, Garrett

, Busta

. Outbreaks associated with fresh produce: Incidence, growth, and survival of pathogens in fresh and fresh-cut produce. Compr Rev Food Sci Food Saf, 2003; 2:78–141.

ISO International Standard 6579, 2002. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for the Detection of Salmonella spp., 4 ^th ed. (07-15).

10.

Jensen

, Sørensen

, Baggesen

, Bødker

, Hoorfar

. Addition of Novobiocin in pre-enrichment step can improve Salmonella culture protocol of modified semisolid Rappaport-Vassiliadis. J Microbiol Methods, 2003; 55:249–255.

11.

Kobayashi

, Kubota

, Fujihara

, Honjoh

, Iio

, Fujiki

, Nakabe

, Oda

, Satoyama

, Takasu

, Nakanishi

, Miyamoto

. Simultaneous enrichment of Salmonella spp, Escherichia coli O157:H7, Vibrio parahaemolyticus, Staphylococcus aureus, Bacillus cereus, and Listeria monocytogenes by single broth and screening of the pathogens by mutiplex real-time PCR. Food Sci Technol Res, 2009; 15:427–438.

12.

Margot

, Stephan

, O'Mahony

, Iversen

. Comparison of rapid cultural methods for the detection of Salmonella species. Int J Food Microbiol, 2013; 163:47–50.

13.

McLinden

, Sargeant

, Thomas

, Papadopoulos

, Fazil

. Association between component costs, study methodologies, and foodborne illness-related factors with the cost of nontyphoidal Salmonella illness. Foodborne Pathog Dis, 2014; 11:718–726.

14.

O'Regan

, McCabe

, Burgess

, McGuinness

, Barry

, Duffy

, Whyte

, Fanning

. Development of a real-time multiplex PCR assay for the detection of multiple Salmonella serotypes in chicken samples. BMC Microbiol, 2008; 8:156.

15.

Scharff

. Economic burden from health losses due to foodborne illness in the United States. J Food Prot, 2012; 75:123–131.

16.

Seow

, Ágoston

, Phua

, Yuk

. Microbiological quality of fresh vegetables and fruits sold in Singapore. Food Control, 2012; 25:39–44.

17.

Taskila

, Tuomola

, Ojamo

. Enrichment cultivation in detection of food-borne Salmonella . Food Control, 2012; 26:369–377.

18.

Taylor

. Isolation of Salmonellae from foods samples: V. Determination of the method of choice for enumeration of Salmonella . Appl Environ Microbiol, 1961; 9:487–490.

19.

Uyttendaele

, Rajkovic

, Van Houteghem

, Boon

, Thas

, Debevere

, Devlieghere

. Multi-method approach indicates no presence of sub-lethally injured Listeria monocytogenes cells after mild heat treatment. Int J Food Microbiol, 2008; 123:262–268.

20.

Xiao

X-L

, Li

Y-J

, Qin

Y-Y

, Yu

Y-G

, Wu

. A multipathogen selective enrichment broth for simultaneous growth of Salmonella spp., Vibrio parahaemolyticus, and Vibrio cholerae . J Gen Appl Microbiol, 2010; 56:465–474.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB