High levels of background flora inhibits growth of Escherichia coli O157:H7 in ground beef

International Journal of Food Microbiology 56 (2000) 219–225 www.elsevier.nl / locate / ijfoodmicro

High levels of background flora inhibits growth of Escherichia coli O157:H7 in ground beef a b a b, L. Vold , A. Holck , Y. Wasteson , H. Nissen * a

Department of Pharmacology, Microbiology and Food Hygiene, The Norwegian School of Veterinary Science, P.O. Box 8146, N-0033 Oslo, Norway b ˚ , Norway MATFORSK, The Norwegian Food Research Institute, Osloveien 1, N-1430 Oslovn As Received 24 February 1999; received in revised form 26 November 1999; accepted 11 January 2000

Abstract The influence of natural background flora under aerobic and anaerobic incubation on the growth of Escherichia coli O157:H7 in ground beef was investigated. The background flora from eight different commercial ground beef were added to ground beef spiked with E. coli O157:H7 and stored either aerobically or anaerobically at 128C. The results showed that the presence of a large number of background bacteria in the ground meat inhibited the growth of E. coli O157:H7 both aerobically and anaerobically. Inhibition was more pronounced under anaerobic conditions. The background floras consisted mainly of lactic acid bacteria of which approximately 80% were Lactobacillus sakei. These results show the importance of the natural background flora in meat for inhibition of growth of E. coli O157:H7.  2000 Elsevier Science B.V. All rights reserved. Keywords: E. coli O157:H7; Background flora; Ground beef

1. Introduction Cattle, especially young animals, have been implicated as a principal reservoir of the emerging foodborne pathogen Escherichia coli O157:H7 (Paton and Paton, 1998). Consequently, many of the outbreaks caused by this bacterium have been associated with eating undercooked ground beef (Bell et al., 1994). In response to the potential health hazard *Corresponding author. Tel.: 1 47-64-970-100; fax: 1 47-64970-333. E-mail address: [email protected] (H. Nissen)

posed by E. coli O157:H7-contaminated meat, the industry has reinforced its efforts to lower levels of E. coli O157:H7 and other pathogens in meat by monitoring food production more closely to improve the safety of the food supply (MacDonald and Osterholm, 1993). This focus on slaughter hygiene has already resulted in a reduction of the total bacterial load in the meat. While microbial numbers in the range 10 5 –10 7 colony forming units (cfu) / g meat were considered normal for many decades, a recent survey of 563 samples of fresh ground beef in the USA revealed a mean of only 10 3.9 cfu / g meat (United States Department of Agriculture, 1994).

0168-1605 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0168-1605( 00 )00215-4

220

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225

However, the employed pathogen-reduction measures are not selective for pathogens, and as a result, the harmless background microflora is also reduced. Concerns have now arisen as to the effects such measures may have on the microbial ecology of the products in question, and regarding their influence on the potential for growth of pathogens (Hintlian and Hotchkiss, 1987; Farber, 1991; Hotchkiss and Banco, 1992; Hao and Brackett, 1993; Sofos, 1993). Although the degree to which the background flora influences the growth of pathogens is a controversial issue, Jay (1996) suggested that a possible explanation for the large hamburger-associated E. coli O157:H7 outbreak in USA in 1992–1993 was that the raw ground beef contained such a low level of background micro-organisms that the few contaminating E. coli O157:H7 had no antagonists to hinder their survival and growth. He further claimed that a background flora of 10 5 –10 6 cfu / g would inhibit growth of low numbers of E. coli O157:H7, but does not refer to any experimental data in support of this. The aim of the current study was to investigate the influence of a high and low level of background flora on the growth of E. coli O157:H7 in ground beef. We chose storage at 128C during the experiments as an example of worst case in commercial food storage. We also wanted to study whether the effect of the background flora was of a quantitative nature only, or if any effect also depended on the specific dominant bacterial species in the background flora. As growth will also vary with the conditions, i.e. aerobic and anaerobic, both samples of ground beef packed in oxygen transmissible film and samples packed under vacuum were investigated.

2. Materials and methods

2.1. Bacteria and growth conditions An antibiotic-resistant mutant of the E. coli O157:H7 strain NCTC 1200 (National Collection of Type Cultures, Collindale, London, UK) was used to spike different batches of ground beef. This strain does not harbour the shiga-toxin genes, and is resistant to nalidixic acid (ICN, Biomedical INC, Auroras, Ohio, USA) (50 mg / ml) and streptomycin sulphate (Sigma, Steinham, Germany) (1000 mg /

ml). Before inoculation to the ground beef, the strain was grown in TSYB (Tryptone Soya Broth CM129, Oxoid, Ltd., Basingstoke, UK) with 1% yeast extract (Oxoid, CM 121), incubated overnight at 308C and kept refrigerated for 2 days before use. The growth rate of the antibiotic resistant mutant was comparable to the regular strain at 128C.

2.2. Isolation of background flora for inoculation Four different batches of fresh ground beef (aerobic) and four from vacuum / gas-packed (60% CO 2 , 40% N 2 , 0.4% CO, anaerobic) were bought from different stores. Samples weighing 25 g were diluted 1:10 in peptone water (PW) (1 g / l Bacto peptone (Difco, Detroit, MI, USA) 8.5 g / l NaCl (Merck, Darmstadt, Germany)), homogenised in a stomacher bag for 1 min and spread in 10-fold dilutions on TSY agar plates which were incubated aerobically and anaerobically, using the anaerobic system, AnaerobemE (Oxoid, Ltd., Basingstoke, Hampshire, England) respectively at 258C for 3 days. Bacteria used as background flora were collected from plates with 50–100 colonies by washing the plates with 5 ml PW to obtain a suspension of about 10 9 cfu / ml. The background flora suspensions were kept at 48C for approximately 5 h, and then added to the experimental ground beef.

2.3. Preparation of experimental ground beef Beef trimmings with 14% fat were obtained at a commercial abattoir 2–3 days after slaughter of the source animals. The beef was ground in a small meat grinder, using strict hygienic procedures to obtain a low level of background flora. The four aerobic and the four anaerobic background flora prepared as described above were added to eight sets (A–H) of 0.7 kg ground beef at a concentration of 10 5 –10 6 cfu / g (determined by plating on TSYA agar). Sets labelled A, B, C and D received each one of the four suspensions of background flora (isolated from four different aerobic ground beefs). Sets labelled E, F, G and H received each one of the four different suspension of background flora from vacuumpacked / gas-packed (anaerobic) ground beef. To secure a uniform distribution of the bacteria in the sample, 5 ml of the appropriate dilution of the background flora suspension were mixed into the

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225

221

and 20 ml was spread on MRS agar (de Man, Rogosa, Sharpe, CM 359, Oxoid, LTD, Basingstoke, UK) and TSY agar to monitor the background flora. An amount of 100 ml was spread on TSY agar with nalidixic acid (50 mg / ml) and streptomycin (1000 mg / ml), to quantify the number of E. coli O157:H7. The MRS and TSY plates were incubated at 258C for 3 days, while the TSY plates with nalidixic acid and streptomycin were incubated at 378C for 3 days.

ground meat in a food processor. Six other control sets were also prepared to which no background flora was added. These were labelled I, J, K, L, M and N. An overview of the experimental conditions to which the ground beef sets A–N were exposed, is presented in Table 1.

2.4. Spiking and storage of the experimental ground beef All sets, except M and N, were spiked with about 10 3 cfu / g E. coli O157:H7 (the concentration of the stock solution was determined by spreading on TSYA agar plates). After spiking, the ground beef sets A–N were divided to samples of 50 g. Two parallel samples from batches A, B, C, D, K, L and M were packed aerobically in a small tray covered with food grade cling film (PVC, O 2 transmission . 10 000 cm 3 / m 2 per 24 h / atm.), while two parallel samples from batches E, F, G, H, I, J and N were vacuum-packed on an Intervac, IN30, chamber machine (Intervac Verpackungsmaschinen, Wallenhurst, Germany) (Table 1). The bags were made from polyamide / polyethylene (Halvorsen and Larsen A / S, Oslo, Norway), transmission 25 cm 3 / m 2 per 24 h / atm. All samples were stored at 128C.

2.6. Characterisation of lactic acid bacteria background flora that may have an inhibiting effect on growth of E. coli O157: H7 About 60 isolates from the lactic acid bacteria (LAB) background flora growing on MRS agar plates were characterised. Gram-staining, gas-production, oxidase reactions, catalase reactions and growth on Gibson medium, were performed according to standard microbiological procedures ¨ (Ternstrom, 1983). Colony blots were hybridised with 16S rRNA oligo nucleotide probes specific for Lactobacillus sakei, L. curvatus, and Leuconostoc sp. as described by Nissen et al. (1994) and Nissen and Dainty (1995). The agar spot tests for detection of inhibition zones, were performed by spotting LAB strains on MRS plates and using an overlay of E. coli O157:H7 in soft agar TSA (Axelsson et al., 1993).

2.5. Detection of background flora and E. coli O157: H7

2.7. Analysis of data

Sampling for bacterial growth was done in duplicate at the start of the experiment, and after storage for 2, 4, 7, 10 and 14 days. The samples were diluted 1:5 in PW and homogenised in a stomacher bag for 1 min. Serial 10-fold dilutions were prepared in PW,

Statistical analyses were performed using the ANOVA procedure in JMP version 3.2.2. (SAS Institute Inc.).

Table 1 Experimental conditions to which the different ground beef sets A–N were exposed during the challenge experiment Experimental conditions to which ground beef sets A–N were exposed

A

B

C

D

E

F

G

H

I

J

K

L

Spiked with E. coli O157:H7 Extra background flora from aerobically stored ground beef added Extra background flora from anaerobically stored ground beef added Stored under aerobic conditions Stored under anaerobic conditions

x x

x x

x x

x x

x

x

x

x

x

x

x

x

x

x

x

x x

x

x

x

x

x

x

x

x

x

x

x

M

N

x x

222

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225

3. Results and discussion The results from the challenge experiment are presented in Fig. 1. Fig. 1a shows that growth of E. coli O157:H7, inoculated at a concentration of about 10 3 cfu / g, was inhibited by the background flora added to sets A, B, and C. The background flora in set D allowed E. coli O157:H7 to grow to a maximum concentration of about 10 6 cfu / g after 10 days storage at 128C under aerobic conditions. In comparison, when no background flora was added to the aerobically stored meat, growth of E. coli O157:H7 reached a concentration of 10 6 cfu / g after only 4 days (sets K and L; Fig. 1e). Differences in inhibitory ability may be due to differences in competitive properties of the various background flora added to sets A, B, C, and D. In contrast, the growth of E. coli O157:H7 in the anaerobicallystored sets was generally inhibited regardless of whether an external background flora was added or not (Fig. 1c, sets E, F, G, H, Fig. 1e, sets I, J). For sets E, F, G, H, no significant difference in the growth of E. coli O157:H7 was observed (T 5 4.55, P , 0.001). Based on the colony types on the TSYA, the added background flora consisted of a mixture of different bacteria, but during storage, Lactobacillus spp. (LAB) soon became the dominating species, growing to a level of about 10 8 cfu / g. The growth of LAB as detected on MRS agar from aerobically and anaerobically stored ground beef is shown in Fig. 1b and d, respectively. In order to investigate the difference in growth of E. coli O157:H7 in samples stored aerobically and anaerobically without any externally added background flora, growth rate in Sets I and J was compared to Sets K and L (Fig. 1e). In the aerobically stored sets K and L the E. coli O157:H7 counts after four days of storage had risen to 4.3 ? 10 5 and 5.15 ? 10 6 cfu / g, respectively. In the anaerobicallystored sets I and J, the corresponding figures were 1.25 ? 10 3 and 9 ? 10 2 cfu / g. The magnitude of this difference between aerobic and anaerobic growth was verified by ANOVA analysis (F 5 61.64 and P , 0.0001). The growth of endogenous LAB was, however, more pronounced in the anaerobicallystored samples, reaching levels of 10 7 cfu / g after 4 days (Fig. 1f). This indicates that both the back-

ground flora and the availability of air exert influence on the outgrowth of E. coli O157:H7. Monitoring of pH throughout the experiments revealed only minor variations in the beef samples, levels varying between 5.4 and 5.6 (results not shown). This indicates that pH had only minor, if any, influence on the observed differences in growth rate. In another set of experiments ground beef containing low levels of endogenous background flora was spiked with 10 3 cfu / g E. coli O157:H7 and subjected to both aerobic and anaerobic storage (not shown). Again growth of E. coli O157:H7 was observed under aerobic conditions, but in addition, in some cases where low growth of LAB was observed, E. coli O157:H7 would also grow under anaerobic conditions (not shown). Ground beef stored at 88C showed no growth of E. coli O157:H7 when samples were stored anaerobically, while aerobically stored samples reached levels of 10 4 cfu / g E. coli O157:H7 by the 7th day of storage (results not shown). Growth of E. coli O157:H7 at 88C in ground beef was also found by Palumbo et al. (1997) when samples were exposed to air. The dominating species during storage consisted of LAB, and we chose to isolate dominating colony types of LAB from the samples which showed total inhibition of E. coli O157:H7 although the inhibiting background flora may also contain antagonistic Gram negative bacteria. The LAB strains isolated were mostly homofermentative lactic acid bacteria and about 80% of the LABs were L. sakei. Agar spot tests (Axelsson et al., 1993) with E. coli O157:H7 showed distinct inhibition zones of most, but not all, of the isolated lactic acid bacteria (results not shown). Inhibition by LAB bacteriocins has recently been shown also for Gram negative bacteria like E. coli O157:H7, but mostly in a food matrix with low ¨ pH or high sodium chloride concentration (Ganzle et al., 1999). In ground beef, antibacterial compounds like lactic acid, alcohols and H 2 O 2 may be more important. The results indicate that L. sakei could be used as a competitive flora in such products. However, different L. sakei strains will differ in their competitive abilities due to factors like difference in ability to grow at low temperatures and lactic acid production. The use of competitive microflora has been earlier proposed by a number of authors (Gombas, 1989; Holzapfel et al., 1995; Stiles, 1996)

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225

223

Fig. 1. Growth of E. coli O157:H7 and lactic acid bacteria (LAB) in ground beef with externally added background flora under aerobic and anaerobic conditions. (a) Growth of E. coli O157:H7 in ground beef sets A, B, C and D, sets inoculated with background flora isolated from four different aerobically stored ground beef, and incubated under aerobic conditions. (b) Growth of LAB in ground beef sets A, B, C and D. (c) Growth of E. coli O157:H7 in ground beef sets E, F, G and H, sets inoculated with background flora isolated from four different anaerobically stored ground beef, and incubated under anaerobic conditions. (d) Growth of LAB in ground beef sets E, F, G and H. (e) Growth of E. coli O157:H7 in ground beef sets I, J, K and L, sets are not inoculated with external background flora. I and J are stored anaerobically, K and L are stored aerobically. (f) Growth of LAB in sets I, J, K and L.

224

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225

and was reviewed by Hurst as early as 1973 (Hurst, 1973). The current trend in the food industry is to extend the shelf-life of fresh foods by reducing the microbial load through washing or other sanitising procedures, and through modified-atmosphere packaging (Breidt and Fleming, 1998). Jay (1997) has expressed concern that pathogen reduction strategies will make the products potentially more hazardous due to the low background flora. He suggested that immediately after pathogen reduction strategies have been applied, beef carcasses should be sprayed with a background flora (Jay, 1995). In this way, a harmless microflora would be restored, but, on the other hand, this may cause a shorter product shelflife. Several publications have supported the concept of microbial interference and its relation to the safety of fresh food. Gilliland and Speck (1975) found that a lactic acid bacterium, Carnobacterium piscicola, completely inhibited the growth of Listeria monocytogenes in various foods, and Wang et al. (1997) found that the background organisms in unpasteurized milk were probably responsible for reducing growth of E. coli O157:H7 in unpasteurized milk compared to the growth that occurred in pasteurized milk. On the other hand, it might be argued that application of LAB as protective cultures will compromise improvement in hygiene, and there is also concern that LAB may contribute to undesirable sensory changes of meat during aerobic storage (Leisner et al., 1995). The results of the present study support the theory that interfering normal micro-flora are important in controlling the growth of a food-borne pathogen such as E. coli O157:H7. It should, however, be emphasised, that, like Palumbo et al. (1997), under no circumstances did we observe death of E. coli O157:H7. Due to the low infectious dose of this pathogen, even survival without growth in ground beef may pose a public health risk if the meat is insufficiently heat-treated. Thus, regardless of the background flora level, measures to reduce contamination of meat with E. coli O157:H7 are absolutely essential. A suitable approach could therefore be to employ good hygiene in all stages of food production in order to keep the number of pathogens low. As an additional measure to inhibit growth, the use of appropriate LAB as competitive or ‘protective’ cultures could be used.

Acknowledgements The authors thank Janina Berg for technical assistance, and Eystein Skjerve for statistical discussions and valuable advice. This research work was financially supported by The Research Council of Norway, grant no. 107898 / 112.

References Axelsson, L., Holck, A., Birkeland, S.E., Aukrust, T., Blom, H., 1993. Cloning and nucleotide-sequence of a gene from Lactobacillus sake LB706 necessary for sakacin-A production and immunity. Appl. Environ. Microbiol. 59, 2868–2875. Bell, B.P., Goldoft, M., Griffin, P.M., Davis, M.A., Gordon, D.C., Tarr, P.I., Bartleson, C.A., Lewis, J.H., Barrett, T.J., Wells, J.G., 1994. A multistate outbreak of Escherichia coli O157:H7associated bloody diarrhea and hemolytic uremic syndrome from hamburgers. The Washington experience. J. Am. Med. Assoc. 272, 1349–1353. Breidt, F., Fleming, H.P., 1998. Modeling of the competitive growth of Listeria monocytogenes and Lactococcus lactis in vegetable. Appl. Environ. Microbiol. 64, 3159–3165. Farber, J.M., 1991. Microbiological aspects of modified-atmosphere packaging technology a review. J. Food Prot. 54, 58–70. ¨ Ganzle, M.G., Weber, S., Hammes, W.P., 1999. Effect of ecological factors on the inhibitory spectrum and activity of bacteriocins. Int. J. Food Microbiol. 46, 207–217. Gilliland, S.E., Speck, M.L., 1975. Inhibition of psycrotrophic bacteria by Lactobacilli and Pediococci in nonfermented refrigerated foods. J. Food Sci. 40, 903–905. Gombas, D.E., 1989. Biological competition as a preserving mechanism. J. Food Safety 10, 107–117. Hao, Y.Y., Brackett, R.E., 1993. Influence of modified atmosphere on growth of vegetable spoilage bacteria in media. J. Food Prot. 56, 223–228. Hintlian, C.B., Hotchkiss, J.H., 1987. Comparative growth of spoilage and pathogenic organisms on modified atmospherepackaged cooked beef. J. Food Prot. 50, 218–223. Holzapfel, W.H., Geisen, R., Schillinger, U., 1995. Biological preservation of foods with reference to protective cultures, bacteriocins, and food-grade enzymes. Int. J. Food Microbiol. 24, 343–362. Hotchkiss, J.H., Banco, M.J., 1992. Influence of new packaging technologies on the growth of microorganisms in produce. J. Food Prot. 55, 815–820. Hurst, A., 1973. Microbial antagonism in foods. Can. Inst. Food Sci. Technol. J. 6, 80–90. Jay, J.M., 1995. Foods with low numbers of microorganisms may not be the safest foods, or why did human listeriosis and hemorrhagic colitis become foodborne diseases? Dairy Food Environ. Sanit. 15, 674–677. Jay, J.M., 1996. Microorganisms in fresh ground meats: The relative safety of products with low versus high numbers. Meat Sci. 43, S59–S66.

L. Vold et al. / International Journal of Food Microbiology 56 (2000) 219 – 225 Jay, J.M., 1997. Do background microorganisms play a role in the safety of fresh foods. Trends Food Sci. Technol. 8, 421–424. Leisner, J.J., Greer, G.G., Dilts, B.D., Stiles, M.E., 1995. Effect of growth of selected lactic acid bacteria on storage life of beef stored under vacuum and in air. Int. J. Food Microbiol. 26, 231–243. MacDonald, K.L., Osterholm, M.T., 1993. The emergence of Escherichia coli O157:H7 infection in the United States. The changing epidemiology of foodborne disease. J. Am. Med. Assoc. 269, 2264–2266. Nissen, H., Dainty, R., 1995. Comparison of the use of rRNA probes and conventional methods in identifying strains of Lactobacillus sakei and L. curvatus isolated from meat. Int. J. Food Microbiol. 25, 311–315. Nissen, H., Holck, A., Dainty, R.H., 1994. Identification of Carnobacterium spp. and Leuconostoc spp. in meat by genusspecific 16S rRNA probes. Lett. Appl. Microbiol. 19, 165–168. Palumbo, S.A., Pickard, A., Call, J.E., 1997. Population changes and verotoxin production of enterohemorrhagic Escherichia

225

coli strains inoculated in milk and ground beef held at low temperatures. J. Food Prot. 60, 746–750. Paton, J.C., Paton, A.W., 1998. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin. Microbiol. Rev. 11, 450. Sofos, J.N., 1993. Current microbiological considerations in food preservation. Int. J. Food Microbiol. 19, 87–108. Stiles, M.E., 1996. Biopreservation by lactic acid bacteria. Antonie van Leeuwenhoek. 70, 331–345. ¨ ¨ Ternstrom, A. 1983. Livsmedelsbakteriologi. Kottforskingsin¨ ¨ stituttet, Kjavlinge, Kepa-Tryck, Kjavlinge, pp. 129–145. United States Department of Agriculture, 1994. Nationwide Federal Plant Raw Ground Beef Microbiological Survey, United States Department of Agriculture, Food Safety and Inspection Service. Washington DC, USA. Wang, G.D., Zhao, T., Doyle, M.P., 1997. Survival and growth of Escherichia coli O157:H7 in unpasteurized and pasteurized milk. J. Food Prot. 60, 610–613.