Growth and Survival of Listeria monocytogenes in Market Cheeses ...

662 Journal of Food Protection, Vol. 54, No. 9 Pages 662-668 (September 1991) Copyright© International Association of Milk, Food and Environmental Sanitarians

Growth and Survival of Listeria monocytogenes in Market Cheeses Stored at 4 to 30°C CONSTANTIN GENIGEORGIS*, MARIUS CARNICIU, DAN DUTULESCU, and THOMAS B. FARVER Department of Epidemiology and Preventive Medicine, School of Veterinary Medicine, University of California, Davis, California 95616 (Received for publication October 15, 1990)

ABSTRACT

Forty-nine market cheeses representing 24 types and 28 brands were purchased from local supermarkets. Pieces of cheeses of approximately 1.5 x 0.5 cm were surface inoculated with log10 3.95 to 4.36 cells of a Listeria monocytogenes pool made up of five strains (Scott A, V7, RM-1, VPH1, VPH2) and placed in petri dishes. After wrapping with cellophane, the dishes were stored at 4, 8, and 30°C for up to 36 d. Of the cheeses, 36.7% supported growth equivalent to a mean inoculum increase of 1.4 log10 (range 0.21 to 3.58) in at least one storage temperature. They included soft Hispanic type (Queso Fresco, Panela Ranchero, pH 6.2-6.6), Ricotta (pH 5.9-6.1), Teleme (pH 5.9), Brie (pH 7.2-7.7), Camembert (pH 7.3), and cottage (pH 4.9-5.1) cheeses. Ricotta was the best and cottage the worst substrate for growth. Cheeses not supporting Listeria growth but causing gradual death at all temperatures include: Cotija (Hispanic hard cheese), cream, blue, Tillamook, Cracker Barrel, Monterey Jack, Swiss, Cheddar, Colby, string, Provolone, Muenster, Feta, and Kasseri with values of pH 4.3-5.6, process (American, Monterey Jack, Piedmont, pH 5.76.4), and Limburger (pH 7.2) cheeses. A highly significant (P5.5 and absence of starter cultures during the cheese manufacturing was observed. Overall, the study demonstrated that crosscontamination of certain cheeses with L. monocytogenes originating from raw foods (meat, poultry, fish, vegetables), after opening of packages, may lead to significant growth of the pathogen during refrigerated storage.

For many years Listeria monocytogenes was considered an important veterinary pathogen associated with animal encephalitis, abortion, and mastitis (6,17). Sporadic human infections were manifested mostly as meningitis, septicemia, abortion, and perinatal disease (34). In recent years food contamination with L. monocytogenes became a major concern for the food industry because of several outbreaks and cases associated with consumption of foods including milk and dairy products (2,3,5, 12,13,19). Outbreaks and cases of listeriosis due to the consumption of soft cheeses have been recorded from the *Author to whom correspondence should be sent.

United States, Canada, the United Kingdom, and Switzerland (2,3,5,12,19). The epidemiology of sporadic cases of listeriosis remains obscure (13). For many years it was thought that listeriosis was a zoonotic disease only (16). There is now sufficient evidence to indicate that L. monocytogenes is a widely distributed environmental bacterium (9,17). L. monocytogenes is very prevalent in red meat, poultry, fish, and fresh market produce (7,14,15,24,38). Listeria-canymg raw foods may present a major source from which other foods, including cooked foods, can be cross-contaminated in plants, supermarkets, restaurants, or homes because of inappropriate handling (9). Studies on the prevalence of L. monocytogenes in cheeses have been reported from a number of countries including USA, UK, West Germany, Italy, Netherlands, and Switzerland (4,7,21,24,35) and the prevalence has ranged from 0.55 to 14.5%. The behavior of L. monocytogenes during the making and storage of a variety of cheeses including Cottage, Camembert, Brick, Colby, Cheddar, Feta, blue, and cold-pack cheeses has been evaluated (22,23,2630,37). Potential growth of Listeria during processing and aging was demonstrated for Brick, Feta, and Camembert. The present study was undertaken in order to determine whether market cheeses contaminated with Listeria after the opening of the package could support aerobic growth of the pathogen at refrigeration or abuse temperatures. MATERIALS AND METHODS

Sample collection Cheeses commercially produced in the USA or imported were purchased from supermarkets in Davis, Woodland, and Sacramento, California. The cheeses were kept refrigerated at all times and used for inoculation and growth studies on the day following purchase. Cheese analyses All cheeses were analyzed for pH, water, NaCl, brine concentration, and total plate counts (TPC). The pH was determined by homogenizing 10 g cheese with 20 ml distilled water and measured on ALTEX Selection 2000 Ion Analyzer (Beckman

. JOURNAL OF FOOD PROTECTION, VOL. 54, SEPTEMBER 1991

GROWTH AND SURVIVAL OF L. MONOCYTOGENES IN CHEESES Instruments, Inc., Irvine, CA) and a combination electrode (Orion, gel-filled electrode, Model 91-05, Cambridge, MA). The water content of the cheeses was determined by drying 10 g at 125°C for 2 h. The amount of NaCl (w/w) was determined with Quantab titration strips (1). The percentage of brine was calculated from the formula: % Brine = (% NaCl/% NaCl + % water) x 100 Aerobic TPC were determined by homogenizing 5 g of cheese with 45 ml sterile warm (37°C) 0.1% peptone water with 0.1% Tween 80 and then plating 10-fold dilutions on APT agar (Difco Laboratories, Detroit, MI). The plates were incubated in a candle jar at 30°C for 2 to 3 d. Developed colonies were tested for catalase using 3% H202 and gram-viscosity reaction using 3% KOH (31). Gram-positive (G+), catalase-negative (C-) bacteria were counted as lactic acid bacteria. Preparation of Listeria inoculum L. monocytogenes strains Scott A, V7, RM-1, VPH-1, and VPH-2 from our collection were transferred from stock agar slant cultures to brain heart infusion (BHI) broths (Difco). After 24 h at 35°C, 0.1 ml of each was subcultured in 9 ml BHI broth and then was incubated at 35°C for 18 h. Just before cheese inoculation, a pool of all five strains was prepared by mixing 1 ml from each culture in a tube and vortexing for 1 min. Listeria cells in the pool were estimated by plating on BHI, Modified McBride (MMA) (20), and lithium chloride-phenylethanol-moxalatum (LPM) (18) agars and incubating at 35°C for 24 to 48 h. Cheese inoculation and storage Using aseptic techniques, solid cheese squares (1.5 x 1.5 x 0.5 cm) were cut and placed in sterile petri dishes. For soft cheese 2-3 g of cheese was placed in sterile plastic centrifuge tubes (50ml capacity) with screw caps. Volumes of 0.01 ml of appropriate inoculum dilution were placed and spread on the surface of each cheese sample. The petri dishes (wrapped with cellophane to minimize dehydratation) and tubes were placed in the refrigerator until all samples were inoculated. Approximately 1 h after inoculation was completed, multiple samples of each cheese were placed at 4, 8, and 30°C and incubated for a maximum of 36 d. Estimating Listeria growth Samples were taken at preselected intervals (depending on growth or death of inoculum) from each of the 4, 8, and 30°C incubators and analyzed for Listeria. Solid pieces of cheese were placed in a calibrated centrifuge tube containing 10 ml peptone water with Tween 80. After vortexing for 2 min, the suspended cells were diluted and plated on MMA and LPM agars. For soft cheeses 20 ml of diluent was added into each tube, the suspension was vortexed, the volume was recorded, and appropriate dilutions plated on MMA and LPM agars. The agars were incubated at 35°C and then observed in the dissecting microscope under Henry's 45° transillumination for typical Listeria colonies (18). At the beginning of the study, suspect colonies were checked for catalase reaction, esculin fermentation on esculin agar (14), morphology, motility, phase contrast microscopy, and other biochemical tests as described previously (14). With developed experience, counting of Listeria colonies became easier and was based mainly on microscopy (phase and dissecting) and esculin reaction. RESULTS AND DISCUSSION During the period of April to December 1989, we examined 49 market cheese samples representing 24 types and 28 brands for their ability to support aerobic growth of L. monocytogenes. Three cheeses were imported. The pH,

663

brine (%), and TPC as well as the level of inoculated L. monocytogenes and the maximum degree of growth (+) or reduction (-) in log10 at a certain temperature after a certain period of storage are shown in Table 1. Of the 49 cheeses inoculated with log]0 3.95 to 4.36/ sample, 36.7% supported Listeria growth in at least one storage temperature. Cheeses supporting growth included soft Hispanic type (Queso Fresco, Panela, Ranchero), Ricotta, Teleme, Brie, Camembert, and cottage. For reasons of maintaining a desirable texture, soft Hispanic cheeses have pH values of >6.0. Those we tested were made from pasteurized milk and had pH values ranging from 6.2 to 6.7. Of the seven soft cheeses we inoculated with Listeria (representing 3 companies and 3 types), only one declared on the label the use of lactic starters. Inoculated Listeria increased by an average of 0.87 (range 0 to 3.18 log]0) in 17/21 cheeses stored at 4, 8, and 30°C for 1 to 36 d depending on temperature. The best growth was observed in a cheese that had the highest pH and a flora made mostly of C+, G+ bacteria. In the other six cheeses, the flora was made exclusively of C-, G+ bacteria (considered as lactic acid bacteria). The cheese that indicated on its label that a starter culture was used had the lowest pH (6.2) but still supported growth of L. monocytogenes. The step of processing at which the starter was used remains unknown. Overall the findings indicated that in the absence of starter and the presence of a high pH these cheeses can support good Listeria growth during their shelf life even at 4°C. The reported minimal temperature supporting L. monocytogenes growth is -0.1 to -0.4°C (39). The presence of natural lactic contaminants cannot be trusted for their potential inhibiting effect on L. monocytogenes since their initial types and numbers are not controlled. Some lactic acid bacteria, especially those producing bacteriocins or bacteriocin-like substances, have been shown to inhibit L. monocytogenes growth in culture media and fermented milks (8,25,32). Ricotta cheeses (pH 5.9 to 6.1) supported Listeria growth (1.53 to 4.18 log10 increase in inoculum) at 4 to 30°C. All three samples contained vinegar. One of the three samples (TPC of -1.87 >-1.87 +0.34 >-1.87 +0.42 +0.41 +1.19 (+2.19)# + 1.13 (+ 1.19) +0.94 (+0.49) >-1.87 >-1.87 >-1.87

GROWTH AND SURVIVAL OF L. MONOCYTOGENES IN CHEESES

Competing C-, G+ microorganisms considered as the starter bacteria ranged from log10 7.24 to 8.17 CFU/g, and obviously did not have an impact on Listeria growth. Surface molds did not exhibit any inhibitory effect on Listeria growth either, as reported previously (28). The frequent isolation of Listeria from such cheeses in the past led Ryser and Marth (28) to study the behavior of the pathogen during the manufacturing and ripening of Camembert cheese. Their studies indicated that because of lactic starter and low pH effect, growth of contaminating Listeria was practically controlled during the first 17 d of processing-ripening. After this period, growth commenced and proceeded at a rate dependent on strain and rate of pH increase. The minimum pH values of the surface and wedge of cheeses needed for all test stains to increase 10-fold were 5.6 to 6.3 and 6.2 to 6.7, respectively. Terplan et al. (36) reported similar findings for soft mold-ripened cheeses. In both of the above studies, there was better growth on the surface/ wedge than in the interior of the cheeses. In our Brie and Camembert samples, mean log increases at 4, 8, and 30°C were 1.78 and 1.63 for surface and center inoculations, respectively. Ryser and Marth (28) inoculated the surface of 10-d-old Camembert cheese with four L. monocytogenes strains and stored the samples at 6°C for up 70 d to simulate postmanufacture contamination. Three of the four strains increased by 2 to 3 log 0, a smaller increase than when Listeria was present in milk before processing. In our studies mean log10 increases after 36 d at 4 and 8°C were 0.59 and 2.36, respectively. Based on the findings of this and other (28,36) studies, L. monocytogenes can grow in the outer or exposed (after cutting) surfaces as a result of cross-contamination and sufficient storage at temperatures >4°C. Five cottage cheeses representing three brands were challenged with log 3.95 to 4.17 of L. monocytogenes strain pool. One cheese was challenged also with a log10 2.11 inoculum. Three of the labels indicated the use of starter cultures, one the presence of acetic acid, and one with neither of these two variables. The pH ranged from 4.9 to 5.1. Listeria inoculum increased by 1.2 to 2.2 log10 in one of the cheeses (made from cultured skim milk) stored at 30°C for 2 to 4 d. The microbial flora of this cheese (log10 3.0 CFU/g) was made up of C-, G+ microorganisms. The flora of the other cheeses was made up of 65 to 100% C-, G+ microorganisms ranging from 5.0 to 7.12 CFU/g log10. The cheese containing the acetic acid did not support Listeria growth at any temperature. On the contrary, its environment was bactericidal and the numbers of Listeria decreased by >1.87 log after storage at >4° C for at least 8 d. Three of the five cheeses supported growth at 8° C, which occurred between the 8th and 24th d of storage. Ryser et al. (26) evaluated the survival of L. monocytogenes during manufacture and storage of cottage cheese. While inoculated in the milk, L. monocytogenes was isolated from 52.7% of the 112 cheeses (pH 5.0 to 5.4) stored at 3°C for up to 28 d; they did not observe any growth of the pathogen. Failure of growth could have been due to the inability of heat-stressed cells of L. monocytogenes to repair at 3°C in a low pH cheese. In the case of our experiment, nonstressed 18-h cell cultures were able to

665

overcome the inhospitable environment of cottage cheese and eventually grew at >4°C in 8/15 experimental settings (5 cheeses x 3 temperatures). The findings indicated that Listeria contaminating an opened cottage cheese package may grow by a factor of 10 within its shelf-life time (1520 d). L. monocytogenes at levels of log10 3.95-4.36 CFU/g was unable to initiate growth at 4 to 30°C in any of 16 types of cheeses which included: Cotija (Hispanic hard cheese), cream cheese, blue cheese, Tillamook, Cracker Barrel, Monterey Jack, Swiss, Cheddar, Colby, string, Provolone, Limburger, Muenster, process cheese (American, Monterey Jack, Piedmont), Feta, and Kasseri (Table 2). The highest number of decimal reductions of the inoculum (log10) was >1.26 to >2.36 after 4 to 18 d at 30°C, 0.06 to > 2.36 after 4 to 36 d at 8°C, and 0.18 to > 2.36 after 8 to 36 d at 4°C. With the exception of Limburger (pH 7.2) and Piedmont processed cheese (pH 6.4), the other 14 cheeses inhibiting growth of L. monocytogenes had pH values of 4.2 to 5.7. Inhibition of L. monocytogenes growth in the high pH Limburger cheese is probably due to the extremely high levels (log10 7.98 CFU/g) of competing bacteria. Inhibition of growth in Piedmont cheese was probably due to the combining effects of added organic acids (lactic, citric, sorbic) and antimicrobial compounds originating from the cultured milk and buttermilk ingredients. Statistical analysis (x2) of all collected data for the 49 inoculated cheeses demonstrated a significant (P5.5. Of the 16 types of cheeses which did not support growth, all but the processed cheeses indicated on their label the use of starter cultures. All three process cheeses contained ingredients which were made by processes incorporating use of starter cultures. In addition, they contained sorbic acid which has been shown to inhibit the growth of L. monocytogenes (11). Use of starters was significantly (P6.0*

>7.00* 8.60* (C-, G+ 100%) 5.93*

-2.00 -0.22 >-1.68 >-2.00 -0.70 >-2.00 >-2.00 >-2.00 >-2.00 >-2.00 >-2.00 >-2.00 >-1.95

30 8 4 30 8 4 30

3 36 36 8 30 24 4

>-2.09 >-2.09 >-2.09 >-1.95 -1.95 -1.17 -0.85

30 8 4 30

13 19 30 4

>-2.09 >-2.09 >-2.09 >-1.40

30 8 4 30 8 4 30

13 30 30 7 19 36 4

>-2.09 >-2.09 >-2.09 >-2.09 >-2.09 >-2.09 >-1.26

30 8 4 30 8 30 8 4 30 8 4 30 8 4 30 8 4 30 8 4 30 8 4 30 8 4

7 30 30 4 4 3 36 36 9 36 36 9 36 36 9 36 36 9 36 36 9 36 36 7 36 36

>-2.09 >-2.09 >-2.09 -0.38 -0.06 >-2.09 -1.31 -2.09 >-2.36 -1.06 -0.81 >-2.36 -2.36 -2.29 >-2.36 -2.36 -2.36 >-2.36 -2.26 -2.26 >-2.36 -2.36 -2.00 -2.09 -0.90 -0.18

JOURNAL OF FOOD PROTECTION, VOL. 54, SEPTEMBER 1991

667

GROWTH AND SURVIVAL OF L. MONOCYTOGENES IN CHEESES TABLE 2. com.

Cheese type company

pH

Cheese characteristics Brine %

TPC log]0

Temp. (°C) storage

Days storage

Maximum death (-) in log 10 CFU/sample

6.88

30 8 4 30 8 4 30 8 4 30 8 4 30 8 4 30 8 30 8 4 30 8 4

9 36 36 9 36 36 4 8 8 4 8 8 4 8 8 4 8 8 24 36 4 8 6

>-2.36 -2.36 -1.84 >-2.36 -2.06 -1.62 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04 >-2.04

Monterey Jack process cheese (11) Piedmont process cheese (22) Imported Feta (23)

5.7 4.42 (sorbic and citric acids) 6.4 5.06 (sorbic and citric acids) 7.0 4.3

Imported Feta (24)

4.2

7.4

7.07* (C-, G+ 100%)

Domestic Feta (25)

4.3

7.5

5.0* (C-, G+ 100%)

Domestic Feta (26) Imported Kasseri (27)

4.3

2.2

5.3

5.52

7.14* (C-, G+ 100%) 7.19* (C-, G+ 100%)

Domestic Kasseri

4.8

5.8

(28)

2

7.12* (C-, G+ 100%)

5.25* (C-, G+ 100%)

* = Started culture added; 17a = company 17, sample a; C-, C+ = Catalase-negative or positive; G-, G+ = Gram-negative or positive; R = Raw milk and aged.

ceased when the pH of cheese dropped to below 5.0 and the numbers decreased markedly (2.65 to 2.73 log10 CFU/g) during the first 50 d of ripening (23). An increase in cheese pH from 5 to about 6 during the period of 50 to 120 d of storage did not result in any growth of the pathogen. The gradual decrease in numbers of L. monocytogenes during the ripening and storage of Cheddar, Colby, Feta, and blue cheeses as observed in the above studies is in agreement with our findings for the same cheeses. The inability of L. monocytogenes to grow in challenged processed cheeses and its gradual decrease with storage time observed is in agreement with the research of Ryser and Marth (29) who studied the survival of L. monocytogenes in cold-pack cheese. Of the cheeses studied, the highest concentration of brine was observed in Cotija (9.6 to 12.5%). Obviously, the combination of high brine, low pH (5.5 to 5.6), and the use of starters in Cotija contributed to the creation of an environment very inhospitable for the growth of L. monocytogenes. Growth of the pathogens in Queso Fresco cheeses with brines of 6.15 to 6.6% but pHs of 6.00 to 6.34 is understandable. L. monocytogenes can tolerate high salt environments and can grow in broths with salt concentration as high as 12% if pH is sufficiently high (33). Overall, this study has demonstrated that nonsoft cheeses made with the use of starter cultures and at pH values of