1022 Journal of Food Protection, Vol. 67, No. 5, 2004, Pages 1022–1026 Copyright q, International Association for Food Protection
Research Note
Prevalence and Growth of Listeria on Naturally Contaminated Smoked Salmon over 28 Days of Storage at 48 C† VICTORIA R. LAPPI,1 ALPHINA HO, 1 KEN GALL,2 1Department
AND
MARTIN WIEDMANN 1*
of Food Science, Cornell University, Ithaca, New York 14853; and 2New York Sea Grant and Cornell Cooperative Extension, Stony Brook, New York 11794, USA MS 03-441: Received 29 September 2003/Accepted 4 January 2004
ABSTRACT Only limited data are available on the growth characteristics of Listeria in naturally contaminated ready-to-eat foods. To evaluate Listeria contamination patterns and growth in smoked salmon, 72 smoked salmon product samples from two processing plants were tested for Listeria spp. and L. monocytogenes. Samples were divided into four approximately equal portions: one portion was tested on receipt, and the other three were vacuum sealed and stored at 48C for 7, 14, and 28 days. Listeria testing was performed using both an enrichment procedure and direct plating to enumerate Listeria in samples that contained .2 to 10 CFU/g. Five samples were positive for Listeria spp., including one sample that was positive for L. monocytogenes. Most samples yielded only sporadic positive results among the portions tested on days 0, 7, 14, and 28. Only one sample contained Listeria spp. in numbers above the detection limit for enumeration. For this sample, the portions tested on days 7 and 28 contained 46 and 52 CFU/g, respectively, whereas the portion tested on day 14 was negative. Overall, our data indicate that there is considerable heterogeneity in Listeria spp. distribution within a single positive smoked sh sample. Even with refrigerated storage for 28 days, none of the naturally contaminated samples reached Listeria spp. numbers .100 CFU/g, which indicates that Listeria growth was limited within a 4-week storage period. However, because of the apparent heterogeneity of Listeria distribution within samples, the interpretation of growth data collected on naturally contaminated samples is dif cult.
Contamination of ready-to-eat (RTE) food products with Listeria monocytogenes is of considerable concern, because this bacterium is a foodborne pathogen that causes a severe human disease (9, 19, 28). L. monocytogenes has been isolated from many environments, including soil, water, sh, and farm animals (2, 10–14, 31), and has the ability to survive and grow at refrigeration temperature. The incidence of L. monocytogenes in cold-smoked salmon and cooked sh products has been reported to range from 4.3% to 36%, and a recent draft risk assessment by the U.S. Food and Drug Administration (FDA) and the U.S. Department of Agriculture (USDA) estimated that 15% of all smoked sh is contaminated with L. monocytogenes (1, 2, 11). Understanding the growth rates and contamination prevalence of Listeria in RTE food products is critical to help determine the safety-based shelf life of a product. Although some inoculation studies have shown that Listeria spp. and L. monocytogenes can grow to high levels in smoked seafoods even if they are stored at 48C (13, 22–24, 27), studies that have used smoked sh samples inoculated with cultured L. monocytogenes cells may not be representative for products naturally contaminated with Listeria (5, * Author for correspondence. Tel: 607-254-2838; Fax: 607-254-4868; E-mail:
[email protected] † Any opinions, ndings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily re ect the view of the U.S. Department of Agriculture.
6). Speci cally, observed growth patterns in inoculated products are likely in uenced by the type and number of L. monocytogenes strains used, the physiological state of the bacterial inoculum, the level of inoculum, and the natural micro ora of the inoculated product (5, 13, 27). To further probe the potential of smoked salmon to support Listeria growth, we analyzed Listeria prevalence and contamination levels over 28 days of storage at 48C for 72 smoked salmon products collected from two smoked sh processing plants with a history of Listeria contamination. MATERIALS AND METHODS Plant pro les and cold-smoked salmon samples. Two smoked sh processing plants (plant 1 and plant 3) were selected for the collection of RTE smoked salmon products on the basis of previously documented Listeria contamination of nished products produced in these plants (29). Over a 6-month period, two nished product samples from each of three different lots (six samples total) were collected monthly from each plant, to yield a total of 36 samples per plant (36 cold-smoked products from plant 1 and 20 cold- and 16 hot-smoked products from plant 3). Smoked salmon from plant 1 was produced from farm-raised Atlantic salmon (Salmo salar) and wild-caught King salmon (Oncorhynchus tshawytscha), whereas products from plant 3 were produced from farm-raised Atlantic salmon and wild-caught King salmon, Chum salmon (Oncorhynchus keta), and Sockeye salmon (Oncorhynchus nerka). The production process for both plants included
J. Food Prot., Vol. 67, No. 5
LISTERIA PREVALENCE AND GROWTH ON SMOKED SALMON
wet brining and smoking in a smoking chamber. No physical and chemical analyses of the sh were performed. Product samples were collected aseptically in sterile WhirlPak (Nasco, Fort Atkinson, Wis.) bags for whole sh samples or as commercial vacuum packages for sliced product. Samples were shipped to the laboratory overnight on ice in an insulated container and processed immediately on arrival. Each 500-g nished product sample (whole or sliced) was divided approximately evenly into four portions. One portion was tested for Listeria on the day of arrival in the laboratory, which was designated as day 0. The remaining three portions were placed aseptically into individual vacuum pouches (supplied by each plant to represent the same packaging material used in each plant), vacuum sealed, and stored at 48C for further testing on days 7, 14, and 28 after packaging. Bacteriological analysis. For enumeration by direct plating, each nished product portion was processed on days 0, 7, 14, and 28 by use of the following method. Twenty- ve-gram samples of nished product were stomached for 60 s with 225 ml (1:10) or 25 ml (1:2) of Listeria enrichment broth (LEB; Difco, Becton Dickinson, Sparks, Md.) in a Stomacher 400 (Seward Medical, London, UK). A total of 1.0 ml of the homogenized sample was spread plated onto four Oxford plates that contained Oxford supplement (Difco, Becton Dickinson), and 250 ml of the homogenate was plated on each Oxford plate. Oxford plates were incubated for 48 h at 308C. Listeria-like colonies on Oxford plates that contained the directly plated sample dilutions were counted and recorded as CFU per gram. Initial tests by direct plating used a 1:10 sample dilution with a calculated Listeria detection limit of 10 CFU/g. To increase detection sensitivity, samples in the later part of the study were diluted 1:2 (instead of 1:10) for direct plating, giving a calculated detection limit of 2 CFU/g. If Listerialike colonies were observed on Oxford agar, 20 representative colonies ( ve colonies per plate) from the direct plating plates were streaked onto L. monocytogenes plating medium (LMPM; Biosynth Biochemica & Synthetica, Naperville, Ill.) to identify L. monocytogenes (26). L. monocytogenes and L. ivanovii show characteristic colony morphology and turquoise blue color on LMPM, because of bacterial hydrolysis of a colorimetric phospholipase substrate (26). Presumptive L. monocytogenes isolates were con rmed by a PCR assay that targeted the listeriolysin O gene (hlyA), as described by Norton et al. (21). Detection of low levels of Listeria spp. was achieved by an enrichment protocol that used a modi cation of the FDA method for the isolation of L. monocytogenes (18). After plating 1:2 sample dilutions, samples were further diluted in LEB to a nal dilution of 1:10. Enrichments were plated onto differential media after 24 and 48 h of incubation at 308C. At each time point, 0.05 ml of enrichment was streaked onto both an Oxford and LMPM plates. LMPM and Oxford plates were incubated for 48 h at 358C and 308C, respectively. Putative Listeria spp. were identi ed on the basis of the occurrence of typical colonies on Oxford. Typical turquoise blue color colonies of characteristic morphology on LMPM indicated the presence of L. monocytogenes or L. ivanovii (26). Selected typical colonies on LMPM were con rmed as L. monocytogenes using the PCR assay described above. Ribotyping. One randomly selected L. monocytogenes isolate from each culture-positive sample was subtyped by automated EcoRI ribotyping using the DuPont Qualicon RiboPrinter, as described elsewhere (21). Images were acquired with a charge-coupled device camera and processed using the Riboprinter’s custom software. This software normalizes fragment pattern data for band intensity and relative band position (3). Ribotype patterns were automatically assigned a DuPont ID (e.g., DUP-1039) by the Ri-
1023
boprinter, which was con rmed by visual inspection. If visual inspection found that a given DuPont ID included more than one distinct ribotype pattern, which generally differed by position of only a single weak band, then each pattern was designated with an additional alphabetized letter (e.g., DUP-1039A and DUP1039B). Ribotype patterns for isolates in the present study are available for comparison through Pathogen Tracker (available at: http://www.pathogentracker.net). Speciation using API Listeria. For samples that were negative for L. monocytogenes, one presumptive Listeria isolate from an Oxford plate was streaked onto brain heart infusion (BHI) agar (Difco, Becton Dickinson) and incubated for 24 h at 358C. A single colony from the BHI agar plate was used for species identi cation with the API Listeria test (bioMe´rieux, Inc., Hazelwood, Mo.), according to the manufacturer’s instructions. The test strips were incubated for 18 to 24 h at 358C and then visually scored against the reading table supplied by the manufacturer.
RESULTS AND DISCUSSION Although L. monocytogenes has been isolated from many foods (8, 11, 28, 30), the recent FDA/USDA draft Listeria risk assessment ranked smoked seafood (which included both hot- and cold-smoked products) sixth in terms of its relative risk for causing listeriosis on a per-annum basis in intermediate-aged individuals (1). Additional data on the prevalence and growth of L. monocytogenes in smoked seafoods are clearly needed to further re ne our understanding of the speci c risk of acquiring listeriosis associated with smoked seafoods. We thus tested naturally contaminated RTE smoked salmon from two processing facilities for Listeria spp. and/or L. monocytogenes to determine the (i) incidence of contamination, (ii) level of contamination, and (iii) Listeria growth patterns in vacuumsealed samples stored at 48C over 28 days to simulate retail storage conditions. For the year before the initiation of the present study, Listeria spp. prevalence for nished product obtained from the two plants selected for our study was 12.0% (n 5 78) and 14.0% (n 5 89) for plants 1 and 3, respectively (29). All samples from plant 3, including the 16 hot smoked salmon samples, tested negative for any Listeria spp. Among the 36 cold-smoked samples collected from plant 1, Listeria spp. were isolated from at least one portion of ve samples (14%; Table 1). Four of the ve Listeria-positive samples were products that were sliced and vacuum packaged in the processing plant; among all 72 samples tested, only 12 samples were products that were sliced during manufacturing, whereas 60 samples were whole smoked sh. Among the ve samples positive for Listeria spp., one sample was positive for L. monocytogenes, and L. seeligeri isolates were identi ed in three samples (Table 1). No isolates from the other positive sample (sample P1-1; Table 1) were available for speciation. For two of the ve Listeria spp.–positive samples, only the portion tested after 14 days of storage at 48C was positive for Listeria spp. Although the portion tested on day 0 was positive for Listeria for three samples, some or all of the other portions that were derived from these samples and tested at days 7, 14, and 28 were negative (Table 1). The
1024
LAPPI ET AL.
J. Food Prot., Vol. 67, No. 5
TABLE 1. Listeria spp.–positive smoked salmon samples among a total of 72 samples tested and Listeria contamination patterns over 28 days of storage at 48C Results after enrichment for: 24 h
48 h
Enumeration in CFU/g (species isolated)
Sample P1-1 Day 0 Day 7 Day 14 Day 28
— — — —
— — Listeria spp.a —
,10 ,10 ,10 ,10
Sample P1-8 Day 0 Day 7 Day 14 Day 28
L. seeligeri L. seeligeri — L. seeligeri
L. seeligeri — — L. seeligeri
,10 46 (L. seeligeri) ,10 52 (L. seeligeri)
L. monocytogenes (DUP-1039C) L. monocytogenes (DUP-1039C) — —
L. monocytogenes (DUP-1039C) L. monocytogenes (DUP-1039C) — —
,2
Sample P1-25 Day 0 Day 7 Day 14 Day 28
— — L. seeligeri —
— — L. seeligeri —
,2 ,2 ,2 ,2
Sample P1-31b Day 0 Day 7 Day 14 Day 28
— — — —
L. seeligeri — — —
,2 ,2 ,2 ,2
Sample and day
Sample P1-19 Day 0 Day 7 Day 14 Day 28
a b
,2 ,2 ,2
Isolate was not available for speciation. Sample P1-31 was a whole smoked sh, whereas all other samples were sliced and vacuum-packaged products.
presence of Listeria spp. at .10 CFU/g was only observed in one sample (P1-8; Table 1). The portions of this sample that were tested on days 7 and 28 yielded 46 and 52 CFU Listeria per g, respectively, whereas the portion tested on day 14 was negative by both direct plating and enrichment. For the one L. monocytogenes–positive sample, L. monocytogenes was isolated only from the portions tested on days 0 and 7 but not from the portions tested on days 14 and 28 (Table 1). Molecular subtyping showed that representative L. monocytogenes isolates from both positive samples were ribotype DUP-1039C (Table 1). On the basis of our data, which showed that not all portions of a given sample were found to be consistently positive for Listeria (Table 1), we conclude that Listeria contamination was not ‘‘evenly’’ distributed throughout the smoked sh samples tested. Other researchers, which used direct plating and most probable number approaches for the detection and quanti cation of Listeria, also observed variability of contamination in samples of frankfurters (30), vacuum-packed smoked salmon (5), pate, sliced ham, turkey breast, and wieners (8) when samples were tested at multiple time points throughout shelf life. Overall, there appears to be considerable heterogeneity in Listeria spp.
distribution within positive food samples, a fact that considerably complicates the interpretation of growth data for naturally Listeria contaminated food samples. Future studies might consider testing larger samples sizes (e.g., 125 g), to overcome some of these problems. As shown in Table 1, none of the positive samples contained .100 CFU/g of any Listeria species, which indicates that limited Listeria growth occurred during a 4week storage period at 48C in the ve cold smoked salmon products that tested positive for Listeria. Although some other studies have shown that Listeria can grow on inoculated smoked sh products, even when the products were stored at 48C (27), our data appear to be consistent with the few other studies that were conducted using naturally contaminated smoked seafoods. Speci cally, Cortesi et al. (5) did not nd any indication of the growth of L. monocytogenes in three lots of naturally contaminated vacuumpacked sliced smoked salmon as long as the product was stored at 28C. Jorgensen and Huss (17) also reported little to no indications of the growth of L. monocytogenes in naturally contaminated cold-smoked sh stored at 58C. Even some studies on arti cially contaminated smoked seafoods found that the growth of L. monocytogenes may be
J. Food Prot., Vol. 67, No. 5
LISTERIA PREVALENCE AND GROWTH ON SMOKED SALMON
limited in some product batches; for example, in one of three trials with arti cially contaminated cold-smoked salmon, no growth of L. monocytogenes was observed by Guyer and Jemmi (13). On the other hand, even in naturally Listeria-contaminated smoked seafoods, apparent Listeria growth is observed if the product is stored at temperatures above 48C to 58C (5). Of interest, the growth patterns of Listeria and L. monocytogenes seem to differ between cold smoked and hotsmoked seafood products. L. monocytogenes growth on cold-smoked salmon appears to be slower than on hotsmoked salmon (15, 17). Hot-smoked salmon processing temperatures reach levels that can kill L. monocytogenes, whereas, in cold-smoked salmon, bacterial contamination and growth are controlled by a combination of salt brining, competitive bacteria, liquid or forced air smoke concentrations, and smoking time and temperatures. These inhibitory factors may be bene cial by inhibiting Listeria growth on the product (7, 20, 25). In particular, the reduced competitive micro ora on hot smoked products may facilitate faster growth of L. monocytogenes, compared with coldsmoked products, where no suf cient heat treatment has occurred to signi cantly reduce the bacterial micro ora. It is thus important to bear in mind that the data shown here were based on naturally contaminated cold-smoked salmon; similar growth patterns cannot be assumed for hot-smoked products. Although our data only represent a small number of Listeria-positive cold-smoked salmon samples, they are of considerable interest because of the very limited data on Listeria growth patterns available for naturally contaminated smoked seafood samples. Our data add to the emerging evidence that Listeria growth patterns for naturally contaminated smoked seafoods differ from those obtained with arti cially inoculated samples or those predicted by microbial growth models (4–6, 8, 13, 16, 17). Indeed, at least two other studies found data that support that the growth of L. monocytogenes on naturally contaminated cold-smoked salmon stored under appropriate refrigeration temperatures (,48C to 58C) for fewer than 30 to 40 days may be extremely limited (5, 17). These data further support that appropriate refrigeration throughout distribution and storage are critical to minimizing the potential public health risk associated with high levels of L. monocytogenes in smoked seafoods.
2.
3. 4.
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
ACKNOWLEDGMENTS
18.
This project was supported by the Cooperative State Research Education, and Extension Service, USDA, under agreement 00-51110-9769. Victoria Lappi was partially supported by a fellowship from the National Fisheries Institute. We thank Katy Windham and Esther Fortes for performing automated ribotyping analyses. Our speci c appreciation goes to the plants that participated in the study; their consistent cooperation and their openness in allowing us to conduct the study provided an outstanding example and model for the type of industry collaboration that is needed to address the challenge of controlling L. monocytogenes in the food system.
19. 20.
21.
22.
REFERENCES 1.
Anonymous. 2001. Draft assessment of the relative risk to public health from foodborne Listeria monocytogenes among selected cat-
23.
1025
egories of ready-to-eat foods. Available at: http:// www.foodsafety.gov/;dms/lmrisk.html. Accessed 29 March 2004. Ben Embarek, P. K. 1994. Presence, detection and growth of Listeria monocytogenes in seafoods: a review. Int. J. Food Microbiol. 23: 17–34. Bruce, J. 1996. Automated system rapidly identi es and characterizes microorganisms in food. Food Technol. 50:77–81. Capita R., and C. Alonso-Calleja. 2003. Comparison of different most-probable-number methods for enumeration of Listeria in poultry. J. Food Prot. 66:65–71. Cortesi, M. L., S. Sarli, A. Santoro, N. Murru, and T. Pepe. 1997. Distribution and behavior of Listeria monocytogenes in three lots of naturally-contaminated vacuum-packed salmon stored at 28C and 108C. Int. J. Food Microbiol. 37:209–214. Dalgaard, P., and L. V. Jørgensen. 1998. Predicted and observed growth of Listeria monocytogenes in seafood challenge tests and in naturally contaminated cold-smoked salmon. Int. J. Food Microbiol. 40:105–115. Duffes, F., C. Corre, F. Leroi, X. Dousset, and P. Boyaval. 1999. Inhibition of Listeria monocytogenes by in situ produced and semipuri ed bacteriocins of Carnobacterium spp. on vacuum-packed, refrigerated cold-smoked salmon. J. Food Prot. 62:1394–1403. Farber, J. M., and E. Daley. 1994. Presence and growth of Listeria monocytogenes in naturally-contaminated meats. Int. J. Food Microbiol. 22:33–42. Farber, J. M., and P. I. Peterkin. 1991. Listeria monocytogenes, a food-borne pathogen. Microbiol. Rev. 55:476–511. Fenlon, D. R. 1999. Listeria monocytogenes in the natural environment. p. 21–37. In E. T. Ryser and E. H. Marth (ed.), Listeria, listeriosis and food safety, 2nd ed. Marcel Dekker, Inc., New York. Gombas, D. E., Y. Chen, R. S. Clavero, and V. N. Scott. 2003. Survey of Listeria monocytogenes in ready-to-eat foods. J. Food Prot. 66:559–569. Gram, L. 2001. Potential hazards in cold-smoked sh: Listeria monocytogenes. In Processing parameters needed to control pathogens in cold smoked sh. J. Food Sci. Suppl. 66S:1067–1071. Guyer, S., and T. Jemmi. 1991. Behavior of Listeria monocytogenes during fabrication and storage of experimentally contaminated smoked salmon. Appl. Env. Microbiol. 57:1523–1527. Hoffman, A., K. L. Gall, and M. Wiedmann. 2003. The microbial safety of minimally processed seafood with respect to Listeria monocytogenes, p. 53–75. In J. S. Novak, G. M. Sapers, and V. K. Juneja (ed.). Microbial safety of minimally processed foods. CRC Press, Boca Raton, Fla. Jemmi, T., and A. Keusch, 1992. Behavior of Listeria monocytogenes during processing and storage of experimentally contaminated hot-smoked trout. Int. J. Food Microbiol. 15:339–346. Johansson, T., 1998. Enhanced detection and enumeration of Listeria monocytogenes from foodstuffs and food processing environments. Int. J. Food Microbiol. 40:77–85. Jørgensen, L. V., and H. H. Huss. 1998. Prevalence and growth of Listeria monocytogenes in naturally contaminated seafood. Int. J. Food Microbiol. 42:127–131. Lovett, J. 1988. Isolation and identi cation of Listeria monocytogenes in dairy products. J. AOAC. 71:658–660. McLauchlin, J. 1997. The relationship between Listeria and listeriosis. Food Control 7:187–193. Nilsson, L., L. Gram, and H. H. Huss. 1999. Growth control of Listeria monocytogenes on cold smoked salmon using a competitive lactic acid bacteria ora. J. Food Prot. 62:336–342. Norton, D. M., M. A. McCamey, K. L. Gall, J. M. Scarlett, K. J. Boor, and M. Wiedmann. 2001. Molecular studies on the ecology of Listeria monocytogenes in the smoked sh processing industry. Appl. Env. Microbiol. 67:198–205. Pelroy, G., M. E. Peterson, P. J. Holland, and M. W. Eklund. 1994. Inhibition of Listeria monocytogenes in cold-process (smoked) salmon by sodium lactate. J. Food Prot. 57:108–113. Pelroy, G., M. Peterson, R. Paranjpye, J. Almond, and M. Eklund. 1994. Inhibition of Listeria monocytogenes in cold-process (smoked)
1026
24.
25.
26.
27.
LAPPI ET AL.
salmon by sodium nitrite and packaging method. J. Food Prot. 57: 114–119. Peterson, M. E., G. A. Pelroy, R. N. Paranjpye, F. T. Poysky, J. S. Almond, and M. W. Eklund. 1993. Parameters for control of Listeria monocytogenes in smoked shery products: sodium chloride and packaging method. J. Food Prot. 56:938–943. Poysky, F. T., R. N. Paranjpye, M. E. Peterson, G. A. Pelroy, A. E. Guttman, and M. W. Eklund. 1997. Inactivation of Listeria monocytogenes on hot smoked salmon by the interaction of heat and smoke or liquid smoke. J. Food Prot. 60:649–654. Restaino, L., E. W. Frampton, R. M. Irbe, G. Schabert, and H. Spitz. 1999. Isolation and detection of Listeria monocytogenes using uorogenic and chromogenic substrates for phosphatidylinositol-specic phospholipase C. J. Food Prot. 62:224–251. Rørvik, L. M., M. Yndestad, and E. Skjerve. 1991. Growth of Lis-
J. Food Prot., Vol. 67, No. 5
28.
29.
30.
31.
teria monocytogenes in vacuum-packed, smoked salmon during storage at 48C. Int. J. Food Microbiol. 14:111–118. Ryser, E. T. 1999. Food borne listeriosis, p. 299–358. In E. T. Ryser and E. H. Marth (ed.), Listeria, listeriosis and food safety, 2nd ed. Marcel Dekker, Inc., New York. Thimothe, J., K. K. Nightingale, K. Gall, V. N. Scott, and M. Wiedmann. 2004. Tracking of Listeria monocytogenes in smoked sh processing plants. J. Food Prot. 67:328–341. Wallace, F. M., J. E. Call, A. C. S. Porto, G. J. Cocoma, the ERRC Special Projects Team, and J. B. Luchansky. 2003. Recovery rate of Listeria monocytogenes from commercially prepared frankfurters during extended refrigerated storage. J Food Prot. 66:584–591. Wesley, I. V. 1999. Listeriosis in animals, p. 39–73. In E. T. Ryser and E. H. Marth (ed.), Listeria, listeriosis and food safety, 2nd ed. Marcel Dekker, Inc., New York.
