106
Journal o f Food Protection, Vol. 77, No. 1, 2014, Pages 106-111 doi: 10.4315/0362-028X.JFP-13-049 Copyright © , International Association for Food Protection
Research Note
P re v a le n c e o f Salmonella S e ro v a rs , Listeria monocytogenes, and Escherichia coli 0 1 5 7 :H 7 in M e d ite rra n e a n R e a d y -to -E a t M eat P ro d u c ts in J o rd a n TAREQ M. OSAILI,1* ANAS A. AL-NABULSI, 1 REYAD R. SHAKER, 1 ZIAD W. JARADAT,12 MOHAMMAD TAHA, 1 MOHAMMED AL-KHERASHA,3 MERVET MEHERAT,4 A N D RICHARD HOLLEY5 1Department o f Nutrition and Food Technology; 2Department o f Biotechnology and Genetic Engineering, Jordan University o f Science and Technology, Irbid 22110, Jordan;3Jordan Food and Drug Administration, Amman, Jordan;4Municipality o f Greater Amman, Amman, Jordan; and 5Department o f Food Science, University o f Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 MS 13-049: Received 6 February 2013/Accepted 26 June 2013
ABSTRACT The presence of Salmonella, Listeria monocytogenes, and Escherichia coli 0157:H7 in ready-to-eat (RTE) meat products is considered a major concern for food control authorities worldwide. The aims of this study were to determine (i) the prevalence of Salmonella, L. monocytogenes, and E. coli 0157:H7 in Mediterranean RTE chicken and beef (CB) products sold in Jordanian restaurants and (ii) the susceptibility of the isolates to antibiotics. A total of 1,028 samples of various types of RTE CB products (550 RTE chicken and 478 RTE beef products) were analyzed by methods described by the International Organization for Standardization followed by molecular confirmation of the isolates. The VITEK2 automated system was used for testing antibiotic susceptibility of the isolates. The overall prevalence of Salmonella serovars in RTE CB products was 0.5%, with 0.8 and 0.2% in RTE chicken and RTE beef, respectively. The overall prevalence of L. monocytogenes in RTE CB products was 2%, with 2.7 and 1.5% in RTE chicken and RTE beef products, respectively. E. coli 0157:H7 was not isolated from any of the tested samples. Multidrug-resistant Salmonella and L. monocytogenes isolates were found. The majority of Salmonella isolates were sensitive to most of the tested antibiotics, and all of the isolates were resistant to more than one antibiotic. Similarly, more than 85% of L. monocytogenes isolates were sensitive to nine antibiotics, and the majority of L. monocytogenes isolates were resistant to fosfomycin and oxacillin.
Ready-to-eat (RTE) food products are those foods that do not require further heat treatment to significantly reduce the microbial load before consumption (8). Examples of traditional RTE meat products in Jordan and in the Mediterranean region are shawirma (beef or chicken meat, sometimes known by other names such as gyro, donair, doner, Doner kebab, and chawarma), beef pastry (a baked food composed of a pastry filled with ground beef meat seasoned with spices), beef kubba (a fried food composed of bulgur wheat and ground beef dough stuffed with ground beef meat seasoned with spices and onion), meat kebab (beef and/or lamb), and beef in pita bread (a baked food composed of a pita bread filled with ground beef meat seasoned with spices). Other popular RTE meat products are roasted chicken and beef or chicken burgers. The presence of a microbiological hazard such as Salmonella, Listeria monocytogenes, and Escherichia coli 0157:H7 in RTE meat products is a major concern to food control authorities worldwide. These foodbome pathogens can cause severe illnesses or death to humans, especially high-risk individuals. Major foodborne pathogens (31 * Author for correspondence. Tel: + 962-2-7201000, Ext 22278; Fax: + 962-2-7201078; E-mail: [email protected].
pathogens) cause an estimated 9.4 million cases of foodbome illness, 55,961 hospitalizations, and 1,351 deaths each year in the United States. Fifty percent of the deaths result from consumption of foods contaminated with Salmonella, L. monocytogenes, or E. coli 0157:H7 (34). In Jordan, during 2006 and 2007 Salmonella in shawirma from local restaurants caused a total of 1,600 hospitaliza tions and one death (20). Salmonella, L. monocytogenes, and E. coli 0157:H7 have been isolated from various types of RTE meat products in the Mediterranean region (3, 15, 22, 29, 38). In Turkey, Ulukanli et al. (38) isolated E. coli 0157:H7 from cooked doner (11.3% of samples) and Kayisoglu et al. (22) isolated Salmonella from cooked beef doner (40% of samples) and cooked chicken doner (80% of samples). In Lebanon, Harakeh et al. (15) isolated Salmonella (7.4% of samples) and E. coli 0157:H7 (7.4% of samples) from meat pies and shawirma. In Amman, Jordan, Osaili et al. (29) isolated L. monocytogenes from shawirma (13.3% of samples) and precooked frozen chicken burgers (76.7% of samples). The increased occurrence of multidrag resistance among foodbome pathogens and their resistance to clinically important antibiotics are currently growing global public health concerns (5,10, 40). Harakeh et al. (15) found
J. Food Prot., Vol. 77, No. 1
107
PREVALENCE OF FOODBORNE PATHOGENS IN RTE MEAT PRODUCTS
that E. coli 0157:H7 and Salmonella isolates from meat pies and shawirma were resistant to one or more tested antibiotics. Osaili et al. (29) found that 30% of L. monocytogenes isolates from popular and traditional RTE chicken meat products were resistant to tilmicosin and tetracycline. Assessment of the food safety status and microbiolog ical quality of RTE foods available to consumers in Jordan is one of the major priorities for the Jordan Food and Drug Administration (JFDA), but few studies on the prevalence of foodbome pathogens in RTE food products in Jordan have been published (1, 2, 29, 32). The objectives of the current study were to determine the prevalence Salmonella, L. monocytogenes, and E. coli 0157:H7 in Mediterranean RTE meat products sold in restaurants in Jordan and to determine the susceptibility of the isolates to selected antibiotics. MATERIALS AND METHODS Sample collection. One thousand twenty-eight samples of various types of RTE meat products (550 samples RTE chicken and 478 samples of RTE beef products), including chicken or beef shawimia, beef pastry, beef kubba, meat kebab, beef in pita bread, roasted chicken, and chicken or beef burgers, were collected from both large and small restaurants in all 12 Jordan govemorates from April through August 2009. The samples were transferred in an insulated container with ice packs to the food laboratories at the Jordan University of Science and Technology, the JFDA, and the Municipality of Greater Amman and analyzed within 24 h. In samples where meat was easily separated from bread, these ingredients were tested separately. W hen sauce was used, it was tested with the meat. No vegetable samples were collected. Methods o f the International Organization for Standardization (ISO) were followed for detection and isolation o f Salmonella, L. monocytogenes, and E. coli 0157:H 7 from all of the samples. Isolation, identification, and confirmation of Salmonella isolates by PCR. Salmonella strains were isolated using the ISO 6579 procedure (18). DNA was extracted from a Salmonella reference strain (ATCC 14028) and suspect Salmonella cells with
TABLE 1. Prevalence o/Salm onella and Listeria monocytogenes
in ready-to-eat chicken and beef products in Jordan No. of samples positive for:
Product
No. of samples
Salmonella
Listeria monocytogenes
301 157 20 478
3 1 0 4
12 1 0 13
42 163 115 92 90 48 550 1,028
0 0 1 0 0 0 1 5
1 5 0 2 0 0 8 21
Chicken Shawirma Roasted Burger Total Beef Shawirma Pastry Kubba Kebab In pita bread Burger Total Grand total
7644) and suspect L. monocytogenes cells with the Wizard DNA extraction kit as per the manufacturer’s instructions. L. monocytogenes- specific primers for the PCR assay were selected based on published nucleotide sequences o f the inlA, inlB, inlC, inLI, actA, hylA, and lap genes (4,13, 24, 25). DNA amplification o f the inlA, inlB, inlC, and inU genes was conducted following the procedure of Liu et al. (24). DNA amplification of the actA, hylA, and lap genes was performed following the protocols o f Cai et al. (4), Mengaud et al. (25), and Furrer et al. (13), respectively. All amplifications were performed in a Veriti thermocycler. The PCR products were separated on 2% agarose gels and photographed with the Gel Doc 2000 system.
Isolation of E. coli 0157:H7. The ISO 16654 method (17) was followed for isolation of E. coli 0157:H 7 strains.
the Wizard DNA extraction kit (Promega, Madison, WI) according to the manufacturer’s instructions. To confirm the identity o f the presumptive Salmonella isolates, a primer pair (5'-GTGAAATT ATCGCC ACGTTCGGGC A A-3', 5 1-TC ATCGC ACCGTC AA AGGAACC-3') that targets the invA gene (present in all Salmonella strains) was used (34). DNA amplification was performed according to Rahn et al. (34). All amplifications were conducted in a Veriti thermocycler (Applied Biosystems, Foster City, CA). The reaction products were separated on 2% agarose gels and photographed with a gel documentation system (Gel Doc 2000, Bio-Rad, Hercules, CA). All confirmed Salmonella isolates were sent to Macrogen (Seoul, Korea) for 16S rRNA analysis to identify Salmonella them to serotype.
Antibiotic susceptibility tests. Isolated pathogens were tested for their susceptibility to antibiotics using the VITEK2 automated system (bioMerieux S.A., Marcy l’Etoile, France). The AST-P592 card for Staphylococcus aureus was used to test the antibiotic susceptibility of L. monocytogenes isolates because no specific card for Listeria exists, and the AST-GN24 card for gram negative bacteria was used to test the antibiotic susceptibility of Salmonella isolates. To prepare the microbial inocula for antibiotic susceptibility tests, adequate numbers o f colonies were mixed in a sterile saline solution (0.45%) as per the manufacturer’s instruc tions. The cards were injected with cell suspensions with an integrated vacuum apparatus, sealed, and loaded into the apparatus for incubation (36°C) and analysis. The results were interpreted according to breakpoints recommended by the Clinical and Laboratory Standards Institute (CLSI) (6, 7) and Soussy et al. (36).
Isolation and identification of L. monocytogenes. L. monocytogenes strains were isolated using the ISO 11290-1 procedure (16, 19). All presumptive isolates were subjected to
RESULTS
Gram staining, to oxidase, catalase, hemolysis, motility, and CAMP tests, and to biochemical identification with the Microbact Listeria 12L kit system (Oxoid, Basingstoke, UK).
Confirmation of L. monocytogenes isolates by PCR. DNA was extracted from an L. monocytogenes reference strain (ATCC
All of the RTE chicken and beef (CB) samples were analyzed for the presence of Salmonella, L. monocytogenes, and E. coli 0157:H7. Five Salmonella enterica serovars were isolated from three types of RTE CB products (Table 1). The overall prevalence of S. enterica in RTE CB products was 0.5%. Two isolates of Salmonella
108
OSAILI ET AL.
J. Food Prot., Vol. 77, No. 1
TABLE 2. Prevalence o f Listeria species other than L. monocytogenes in ready-to-eat chicken and beef products in Jordan No. of samples positive for:
Product
L. grayi
L. innocua
L. welshimeri
L. seeligeri
L. ivanovii
Chicken Shawirma Roasted Burger
12 3 0
3 0 0
10 0 0
17 2 0
204 23 0
Beef Shawirma Pastry Kubba Kebab In pita bread Burger Total
5 6 2 0 0 0 28
0 1 0 0 0 0 4
2 5 5 0 0 0 22
1 24 5 1 0 0 50
7 66 47 5 0 1 353
Typhimurium and one isolate of Salmonella Typhi were purified from chicken shawirma, Salmonella Typhimurium was isolated from roasted chicken, and Salmonella Para typhi was isolated from beef kubba. The prevalence of Salmonella serovars in RTE chicken and RTE beef products was 0.8 and 0.2%, respectively. Salmonella serovars were isolated from 3 (1%) of 301 chicken shawirma samples, 1 (0.6%) of 157 roasted chicken samples, and 1 (0.9%) of 115 beef kubba samples (Table 1). L. monocytogenes was isolated from five types of RTE CB products. The overall prevalence of L. monocytogenes in RTE CB products was 2%. The prevalence of L. monocytogenes in RTE chicken and RTE beef products was 2.7 and 1.5%, respectively. L. monocytogenes was isolated from 12 (4%) of 301 chicken shawirma samples, 1 (0.6%) of 157 roasted chicken samples, 1 (2.4%) of 42 beef shawirma samples, 5 (3.1%) of 163 beef pastry samples, and 2 (2.2%) of 92 meat kebab samples. Listeria grayi and Listeria welshimeri were isolated from RTE CB products at similar rates (2.7 and 2.1%). In contrast, Listeria ivanovii was isolated from RTE CB products at far higher rates than was L. monocytogenes (Table 2). L. ivanovii was isolated from 34.3% of the RTE CB products; it was found in 67.8, 40.9, 40.5, and 14.6% of samples of chicken shawirma, beef kubba, beef pastry, and roasted chicken, respectively. E. coli 0157:H7 was not isolated from any of the 1,028 RTE CB products tested. However, non-0157:H7 E. coli was isolated from beef kubba (2% of samples) and beef pastry (1% of samples). The majority of Salmonella serovars isolated were sensitive to most of the tested antibiotics with the exception of nitrofurantoin and trimethoprim-sulfamethoxazole, to which all Salmonella isolates were resistant. Two isolates each of Salmonella Typhimurium from shawirma were resistant to ampicillin, ampicillin-sulbactam, and gentami cin, and one Salmonella Typhi isolate was resistant to ampicillin and ampicillin-sulbactam (Table 3). All of the Salmonella isolates were resistant to more than one antibiotic; Salmonella Typhi was resistant to four antibiotics, and a
shawirma isolate of Salmonella Typhimurium was resistant to five antibiotics. More than 85% of L. monocytogenes isolates were sensitive to nine antibiotics: imipenem, gentamicin, cipro floxacin, linezolid, teicoplanin, vancomycin, tetracycline, rifampin, and trimethoprim-sulfamethoxazole (Table 4). The majority of L. monocytogenes isolates were resistant to fosfomycin (95% of isolates) and oxacillin (90% of isolates), and 62% of the isolates were resistant to clindamycin or fusidic acid. Multidrug resistance patterns were detected among L. monocytogenes isolates; 57, 14.3, 19, 4.8, and 4.8% of the L. monocytogenes isolates were resistant to three, four, five, six, and eight antibiotics, respectively. DISCUSSION In this study, Salmonella serovars, L. monocytogenes, and other Listeria species were isolated from different types of RTE CB products that are popular in Jordan and East Mediterranean countries. Surprisingly, E. coli 0157:H7 was not isolated from any of these products, although this pathogen was isolated previously from raw beef in Jordan at a rate of 7.8% (30). The findings in the current study are promising in comparison with earlier reports on the prevalence of L. monocytogenes in RTE products in Jordan (29, 32). In Amman, Osaili et al. (29) found L. monocytogenes prevalences of 13.3, 76.7, and 30% in chicken shawirma, chicken burger, and chicken sausage, respectively. Further, Osaili et al. (32) found that the prevalence of L. monocytogenes in different types of brined white cheese (soft to semi-hard) was 11.1%. In comparison with data from other Mediterranean countries, the data from the current study suggest that the food safety risk associated with RTE CB products in Jordanian restaurants is lower. In Turkey, Ulukanli et al. (38) found an E. coli 0157:H7 prevalence of 11.25% in beef doner kebab, and Kayisoglu et al. (22) found Salmonella prevalences of 40% (12 of 30 samples) in cooked beef doner and 80% (24 of 30 samples) in cooked chicken doner. In Lebanon, Harakeh et al. (15) reported prevalences of 9.5% for E. coli 0157:H7 and 7.4% for Salmonella in meat pies and shawirma. In contrast,
J. Food Prot., Vol. 77, No. 1
PREVALENCE OF FOODBORNE PATHOGENS IN RTE MEAT PRODUCTS
109
TABLE 3. MICs and antibiotic sensitivity o f Salmonella isolates from ready-to-eat chicken and beef products in Jordana MIC (pg/ml) Sensitivity, no. (%) of isolates
Breakpoint7’ Antibiotic Ampicillin
Ampicillin-sulbactam
Amikacin Cefazolin Cefoxitin Ceftazidime Ciprofloxacin Cefepime Gentamicin Ertapenem Imipenem Levofloxacin Nitrofurantoin Piperacillin-tazobactam Tobramycin Trimethoprim-sulfamethoxazole
S
R
Range
32
2-32
16 > 4 , 76
S
i
3 (60) (2 Salmonella Typhimurium and 1 Salmonella Typhi isolate) 3 (60) (2 Salmonella Typhimurium and 1 Salmonella Typhi isolate)
2 (40)
2 (40)
2 4-1 6 4-1 6 1 0.25-2 1 1-16
5 4 3 5 4 5 3
0.5 1M 0.12-4 128-512 4 -8 1 20-320
5 (100) 5 (100) 4 (80)
(100) (80) (60) (100) (80) (100) (60)
R
1 (20) 2 (4 0 ) 1 (20) 2 (40) (2 Salmonella Typhimurium isolates)
1 (20) 5 (100)
3 (60) 5 (100)
2 (4 0 ) 5 (100)
a S, sensitive; I, intermediate; R, resistant. h From CLSI (6,7) and Soussy et al. (36).
Vazgecer et al. (39) did not detect Salmonella in 72 chicken doner kebab samples in Turkey. The isolation of Salmonella Typhi and Salmonella Paratyphi from a chicken shawirma and beef kubba sample, respectively, from two different restaurants in the current study is problematic because these organisms are linked to human reservoirs and contaminate food through poor personal hygiene or the use of unsafe water (27). In a 10-year prevalence study of Salmonella, L. monocytogenes, and E. coli 0157:H7 in RTE meat and poultry products in the United States, Levine et al. (23) reported Salmonella and L. monocytogenes prevalences similar to those found in the current study, and none of the tested samples were positive for E. coli 0157:H7. Meyer et al. (26) reported an L. monocytogenes prevalence of 3% in RTE poultry products in Germany. The presence of foodbome pathogens in RTE CB foods may be attributed to insufficient heat treatment during cooking or to postcooking contamination (14). The absence of E. coli 0157:H7 from the tested samples may indicate that the cooking step used in the restaurants was effective for eradicating this bacterium from the products but was not sufficient to eradicate the other two pathogens. E. coli 0157:H7 is a relatively heat-sensitive bacterium compared with Salmonella and L. monocytogenes (33). In the current study, Listeria species other than L. monocytogenes were isolated from RTE CB products, especially shawirma, at a very high rate. L. ivanovii, which is rarely considered pathogenic for humans, was the most common Listeria species isolated from the tested products.
Similar to these findings, Osaili et al. (29) found that L. ivanovii was the most common Listeria species isolated from RTE poultry products sold in Jordan; it was found in 67% of chicken shawirma samples. The high prevalence of Listeria in RTE CB products may be due to the ubiquitous nature of this microbe, which can contaminate products in slaughterhouses, before or after cooking, or through food contact surfaces and food handlers (12). In a parallel project (31), we tested two strains of Salmonella Typhimurium isolated from chicken shawirma for their thermal resistance in this product. These strains were more sensitive than expected at 54 to 64°C with D-values of 0.52 to 9.15 min. These results suggest that the risk of Salmonella in shawirma more likely results from inadequate cooking rather than from postcooking contamination, although the isolation of Salmonella Typhi from chicken shawirma in the current study indicates that postcooking contamination may contribute to the risk. Inadequately cooked shawirma itself may be a health risk to customers, and partially cooked slices could transfer Salmonella to cooked slices. Most Salmonella isolates in the current study were sensitive to the majority of antibiotics tested; however, resistance was noted, and some isolates were resistant to more than two antibiotics, especially the Salmonella Typhi and Salmonella Typhimurium isolates from chicken sha wirma. Similar resistance patterns have been previously reported for Salmonella. Harakeh et al. (15) found that 86 and 57% of the Salmonella isolates from meat-based fast food in Lebanon were resistant to trimethoprim-sulfameth oxazole and gentamicin, respectively. Thai et al. (37) found
no
J. Food Prot., Vol. 77, No. 1
OSAILI ET AL.
TABLE 4. MICs and antibiotic sensitivity of Listeria monocytogenes isolates from ready-to-eat chicken and beef products in Jordana MIC (pg/ml)
Sensitivity
Breakpoint*
Sample isolates, no. (%) of isolates L. monocytogenes ------------------------------------------------------
Antibiotic
S
R
Range
(ATCC 7644)
s
I
R
Benzylpenicillin Oxacillin Imipenem Gentamicin Ciprofloxacin Erythromycin Clindamycin Linezolid Teicoplanin Vancomycin Tetracycline Fosfomycin Fusidic acid Rifampin Trimethoprim-sulfamethoxazole
Journal o f Food Protection, Vol. 77, No. 1, 2014, Pages 106-111 doi: 10.4315/0362-028X.JFP-13-049 Copyright © , International Association for Food Protection
Research Note
P re v a le n c e o f Salmonella S e ro v a rs , Listeria monocytogenes, and Escherichia coli 0 1 5 7 :H 7 in M e d ite rra n e a n R e a d y -to -E a t M eat P ro d u c ts in J o rd a n TAREQ M. OSAILI,1* ANAS A. AL-NABULSI, 1 REYAD R. SHAKER, 1 ZIAD W. JARADAT,12 MOHAMMAD TAHA, 1 MOHAMMED AL-KHERASHA,3 MERVET MEHERAT,4 A N D RICHARD HOLLEY5 1Department o f Nutrition and Food Technology; 2Department o f Biotechnology and Genetic Engineering, Jordan University o f Science and Technology, Irbid 22110, Jordan;3Jordan Food and Drug Administration, Amman, Jordan;4Municipality o f Greater Amman, Amman, Jordan; and 5Department o f Food Science, University o f Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 MS 13-049: Received 6 February 2013/Accepted 26 June 2013
ABSTRACT The presence of Salmonella, Listeria monocytogenes, and Escherichia coli 0157:H7 in ready-to-eat (RTE) meat products is considered a major concern for food control authorities worldwide. The aims of this study were to determine (i) the prevalence of Salmonella, L. monocytogenes, and E. coli 0157:H7 in Mediterranean RTE chicken and beef (CB) products sold in Jordanian restaurants and (ii) the susceptibility of the isolates to antibiotics. A total of 1,028 samples of various types of RTE CB products (550 RTE chicken and 478 RTE beef products) were analyzed by methods described by the International Organization for Standardization followed by molecular confirmation of the isolates. The VITEK2 automated system was used for testing antibiotic susceptibility of the isolates. The overall prevalence of Salmonella serovars in RTE CB products was 0.5%, with 0.8 and 0.2% in RTE chicken and RTE beef, respectively. The overall prevalence of L. monocytogenes in RTE CB products was 2%, with 2.7 and 1.5% in RTE chicken and RTE beef products, respectively. E. coli 0157:H7 was not isolated from any of the tested samples. Multidrug-resistant Salmonella and L. monocytogenes isolates were found. The majority of Salmonella isolates were sensitive to most of the tested antibiotics, and all of the isolates were resistant to more than one antibiotic. Similarly, more than 85% of L. monocytogenes isolates were sensitive to nine antibiotics, and the majority of L. monocytogenes isolates were resistant to fosfomycin and oxacillin.
Ready-to-eat (RTE) food products are those foods that do not require further heat treatment to significantly reduce the microbial load before consumption (8). Examples of traditional RTE meat products in Jordan and in the Mediterranean region are shawirma (beef or chicken meat, sometimes known by other names such as gyro, donair, doner, Doner kebab, and chawarma), beef pastry (a baked food composed of a pastry filled with ground beef meat seasoned with spices), beef kubba (a fried food composed of bulgur wheat and ground beef dough stuffed with ground beef meat seasoned with spices and onion), meat kebab (beef and/or lamb), and beef in pita bread (a baked food composed of a pita bread filled with ground beef meat seasoned with spices). Other popular RTE meat products are roasted chicken and beef or chicken burgers. The presence of a microbiological hazard such as Salmonella, Listeria monocytogenes, and Escherichia coli 0157:H7 in RTE meat products is a major concern to food control authorities worldwide. These foodbome pathogens can cause severe illnesses or death to humans, especially high-risk individuals. Major foodborne pathogens (31 * Author for correspondence. Tel: + 962-2-7201000, Ext 22278; Fax: + 962-2-7201078; E-mail: [email protected].
pathogens) cause an estimated 9.4 million cases of foodbome illness, 55,961 hospitalizations, and 1,351 deaths each year in the United States. Fifty percent of the deaths result from consumption of foods contaminated with Salmonella, L. monocytogenes, or E. coli 0157:H7 (34). In Jordan, during 2006 and 2007 Salmonella in shawirma from local restaurants caused a total of 1,600 hospitaliza tions and one death (20). Salmonella, L. monocytogenes, and E. coli 0157:H7 have been isolated from various types of RTE meat products in the Mediterranean region (3, 15, 22, 29, 38). In Turkey, Ulukanli et al. (38) isolated E. coli 0157:H7 from cooked doner (11.3% of samples) and Kayisoglu et al. (22) isolated Salmonella from cooked beef doner (40% of samples) and cooked chicken doner (80% of samples). In Lebanon, Harakeh et al. (15) isolated Salmonella (7.4% of samples) and E. coli 0157:H7 (7.4% of samples) from meat pies and shawirma. In Amman, Jordan, Osaili et al. (29) isolated L. monocytogenes from shawirma (13.3% of samples) and precooked frozen chicken burgers (76.7% of samples). The increased occurrence of multidrag resistance among foodbome pathogens and their resistance to clinically important antibiotics are currently growing global public health concerns (5,10, 40). Harakeh et al. (15) found
J. Food Prot., Vol. 77, No. 1
107
PREVALENCE OF FOODBORNE PATHOGENS IN RTE MEAT PRODUCTS
that E. coli 0157:H7 and Salmonella isolates from meat pies and shawirma were resistant to one or more tested antibiotics. Osaili et al. (29) found that 30% of L. monocytogenes isolates from popular and traditional RTE chicken meat products were resistant to tilmicosin and tetracycline. Assessment of the food safety status and microbiolog ical quality of RTE foods available to consumers in Jordan is one of the major priorities for the Jordan Food and Drug Administration (JFDA), but few studies on the prevalence of foodbome pathogens in RTE food products in Jordan have been published (1, 2, 29, 32). The objectives of the current study were to determine the prevalence Salmonella, L. monocytogenes, and E. coli 0157:H7 in Mediterranean RTE meat products sold in restaurants in Jordan and to determine the susceptibility of the isolates to selected antibiotics. MATERIALS AND METHODS Sample collection. One thousand twenty-eight samples of various types of RTE meat products (550 samples RTE chicken and 478 samples of RTE beef products), including chicken or beef shawimia, beef pastry, beef kubba, meat kebab, beef in pita bread, roasted chicken, and chicken or beef burgers, were collected from both large and small restaurants in all 12 Jordan govemorates from April through August 2009. The samples were transferred in an insulated container with ice packs to the food laboratories at the Jordan University of Science and Technology, the JFDA, and the Municipality of Greater Amman and analyzed within 24 h. In samples where meat was easily separated from bread, these ingredients were tested separately. W hen sauce was used, it was tested with the meat. No vegetable samples were collected. Methods o f the International Organization for Standardization (ISO) were followed for detection and isolation o f Salmonella, L. monocytogenes, and E. coli 0157:H 7 from all of the samples. Isolation, identification, and confirmation of Salmonella isolates by PCR. Salmonella strains were isolated using the ISO 6579 procedure (18). DNA was extracted from a Salmonella reference strain (ATCC 14028) and suspect Salmonella cells with
TABLE 1. Prevalence o/Salm onella and Listeria monocytogenes
in ready-to-eat chicken and beef products in Jordan No. of samples positive for:
Product
No. of samples
Salmonella
Listeria monocytogenes
301 157 20 478
3 1 0 4
12 1 0 13
42 163 115 92 90 48 550 1,028
0 0 1 0 0 0 1 5
1 5 0 2 0 0 8 21
Chicken Shawirma Roasted Burger Total Beef Shawirma Pastry Kubba Kebab In pita bread Burger Total Grand total
7644) and suspect L. monocytogenes cells with the Wizard DNA extraction kit as per the manufacturer’s instructions. L. monocytogenes- specific primers for the PCR assay were selected based on published nucleotide sequences o f the inlA, inlB, inlC, inLI, actA, hylA, and lap genes (4,13, 24, 25). DNA amplification o f the inlA, inlB, inlC, and inU genes was conducted following the procedure of Liu et al. (24). DNA amplification of the actA, hylA, and lap genes was performed following the protocols o f Cai et al. (4), Mengaud et al. (25), and Furrer et al. (13), respectively. All amplifications were performed in a Veriti thermocycler. The PCR products were separated on 2% agarose gels and photographed with the Gel Doc 2000 system.
Isolation of E. coli 0157:H7. The ISO 16654 method (17) was followed for isolation of E. coli 0157:H 7 strains.
the Wizard DNA extraction kit (Promega, Madison, WI) according to the manufacturer’s instructions. To confirm the identity o f the presumptive Salmonella isolates, a primer pair (5'-GTGAAATT ATCGCC ACGTTCGGGC A A-3', 5 1-TC ATCGC ACCGTC AA AGGAACC-3') that targets the invA gene (present in all Salmonella strains) was used (34). DNA amplification was performed according to Rahn et al. (34). All amplifications were conducted in a Veriti thermocycler (Applied Biosystems, Foster City, CA). The reaction products were separated on 2% agarose gels and photographed with a gel documentation system (Gel Doc 2000, Bio-Rad, Hercules, CA). All confirmed Salmonella isolates were sent to Macrogen (Seoul, Korea) for 16S rRNA analysis to identify Salmonella them to serotype.
Antibiotic susceptibility tests. Isolated pathogens were tested for their susceptibility to antibiotics using the VITEK2 automated system (bioMerieux S.A., Marcy l’Etoile, France). The AST-P592 card for Staphylococcus aureus was used to test the antibiotic susceptibility of L. monocytogenes isolates because no specific card for Listeria exists, and the AST-GN24 card for gram negative bacteria was used to test the antibiotic susceptibility of Salmonella isolates. To prepare the microbial inocula for antibiotic susceptibility tests, adequate numbers o f colonies were mixed in a sterile saline solution (0.45%) as per the manufacturer’s instruc tions. The cards were injected with cell suspensions with an integrated vacuum apparatus, sealed, and loaded into the apparatus for incubation (36°C) and analysis. The results were interpreted according to breakpoints recommended by the Clinical and Laboratory Standards Institute (CLSI) (6, 7) and Soussy et al. (36).
Isolation and identification of L. monocytogenes. L. monocytogenes strains were isolated using the ISO 11290-1 procedure (16, 19). All presumptive isolates were subjected to
RESULTS
Gram staining, to oxidase, catalase, hemolysis, motility, and CAMP tests, and to biochemical identification with the Microbact Listeria 12L kit system (Oxoid, Basingstoke, UK).
Confirmation of L. monocytogenes isolates by PCR. DNA was extracted from an L. monocytogenes reference strain (ATCC
All of the RTE chicken and beef (CB) samples were analyzed for the presence of Salmonella, L. monocytogenes, and E. coli 0157:H7. Five Salmonella enterica serovars were isolated from three types of RTE CB products (Table 1). The overall prevalence of S. enterica in RTE CB products was 0.5%. Two isolates of Salmonella
108
OSAILI ET AL.
J. Food Prot., Vol. 77, No. 1
TABLE 2. Prevalence o f Listeria species other than L. monocytogenes in ready-to-eat chicken and beef products in Jordan No. of samples positive for:
Product
L. grayi
L. innocua
L. welshimeri
L. seeligeri
L. ivanovii
Chicken Shawirma Roasted Burger
12 3 0
3 0 0
10 0 0
17 2 0
204 23 0
Beef Shawirma Pastry Kubba Kebab In pita bread Burger Total
5 6 2 0 0 0 28
0 1 0 0 0 0 4
2 5 5 0 0 0 22
1 24 5 1 0 0 50
7 66 47 5 0 1 353
Typhimurium and one isolate of Salmonella Typhi were purified from chicken shawirma, Salmonella Typhimurium was isolated from roasted chicken, and Salmonella Para typhi was isolated from beef kubba. The prevalence of Salmonella serovars in RTE chicken and RTE beef products was 0.8 and 0.2%, respectively. Salmonella serovars were isolated from 3 (1%) of 301 chicken shawirma samples, 1 (0.6%) of 157 roasted chicken samples, and 1 (0.9%) of 115 beef kubba samples (Table 1). L. monocytogenes was isolated from five types of RTE CB products. The overall prevalence of L. monocytogenes in RTE CB products was 2%. The prevalence of L. monocytogenes in RTE chicken and RTE beef products was 2.7 and 1.5%, respectively. L. monocytogenes was isolated from 12 (4%) of 301 chicken shawirma samples, 1 (0.6%) of 157 roasted chicken samples, 1 (2.4%) of 42 beef shawirma samples, 5 (3.1%) of 163 beef pastry samples, and 2 (2.2%) of 92 meat kebab samples. Listeria grayi and Listeria welshimeri were isolated from RTE CB products at similar rates (2.7 and 2.1%). In contrast, Listeria ivanovii was isolated from RTE CB products at far higher rates than was L. monocytogenes (Table 2). L. ivanovii was isolated from 34.3% of the RTE CB products; it was found in 67.8, 40.9, 40.5, and 14.6% of samples of chicken shawirma, beef kubba, beef pastry, and roasted chicken, respectively. E. coli 0157:H7 was not isolated from any of the 1,028 RTE CB products tested. However, non-0157:H7 E. coli was isolated from beef kubba (2% of samples) and beef pastry (1% of samples). The majority of Salmonella serovars isolated were sensitive to most of the tested antibiotics with the exception of nitrofurantoin and trimethoprim-sulfamethoxazole, to which all Salmonella isolates were resistant. Two isolates each of Salmonella Typhimurium from shawirma were resistant to ampicillin, ampicillin-sulbactam, and gentami cin, and one Salmonella Typhi isolate was resistant to ampicillin and ampicillin-sulbactam (Table 3). All of the Salmonella isolates were resistant to more than one antibiotic; Salmonella Typhi was resistant to four antibiotics, and a
shawirma isolate of Salmonella Typhimurium was resistant to five antibiotics. More than 85% of L. monocytogenes isolates were sensitive to nine antibiotics: imipenem, gentamicin, cipro floxacin, linezolid, teicoplanin, vancomycin, tetracycline, rifampin, and trimethoprim-sulfamethoxazole (Table 4). The majority of L. monocytogenes isolates were resistant to fosfomycin (95% of isolates) and oxacillin (90% of isolates), and 62% of the isolates were resistant to clindamycin or fusidic acid. Multidrug resistance patterns were detected among L. monocytogenes isolates; 57, 14.3, 19, 4.8, and 4.8% of the L. monocytogenes isolates were resistant to three, four, five, six, and eight antibiotics, respectively. DISCUSSION In this study, Salmonella serovars, L. monocytogenes, and other Listeria species were isolated from different types of RTE CB products that are popular in Jordan and East Mediterranean countries. Surprisingly, E. coli 0157:H7 was not isolated from any of these products, although this pathogen was isolated previously from raw beef in Jordan at a rate of 7.8% (30). The findings in the current study are promising in comparison with earlier reports on the prevalence of L. monocytogenes in RTE products in Jordan (29, 32). In Amman, Osaili et al. (29) found L. monocytogenes prevalences of 13.3, 76.7, and 30% in chicken shawirma, chicken burger, and chicken sausage, respectively. Further, Osaili et al. (32) found that the prevalence of L. monocytogenes in different types of brined white cheese (soft to semi-hard) was 11.1%. In comparison with data from other Mediterranean countries, the data from the current study suggest that the food safety risk associated with RTE CB products in Jordanian restaurants is lower. In Turkey, Ulukanli et al. (38) found an E. coli 0157:H7 prevalence of 11.25% in beef doner kebab, and Kayisoglu et al. (22) found Salmonella prevalences of 40% (12 of 30 samples) in cooked beef doner and 80% (24 of 30 samples) in cooked chicken doner. In Lebanon, Harakeh et al. (15) reported prevalences of 9.5% for E. coli 0157:H7 and 7.4% for Salmonella in meat pies and shawirma. In contrast,
J. Food Prot., Vol. 77, No. 1
PREVALENCE OF FOODBORNE PATHOGENS IN RTE MEAT PRODUCTS
109
TABLE 3. MICs and antibiotic sensitivity o f Salmonella isolates from ready-to-eat chicken and beef products in Jordana MIC (pg/ml) Sensitivity, no. (%) of isolates
Breakpoint7’ Antibiotic Ampicillin
Ampicillin-sulbactam
Amikacin Cefazolin Cefoxitin Ceftazidime Ciprofloxacin Cefepime Gentamicin Ertapenem Imipenem Levofloxacin Nitrofurantoin Piperacillin-tazobactam Tobramycin Trimethoprim-sulfamethoxazole
S
R
Range
32
2-32
16 > 4 , 76
S
i
3 (60) (2 Salmonella Typhimurium and 1 Salmonella Typhi isolate) 3 (60) (2 Salmonella Typhimurium and 1 Salmonella Typhi isolate)
2 (40)
2 (40)
2 4-1 6 4-1 6 1 0.25-2 1 1-16
5 4 3 5 4 5 3
0.5 1M 0.12-4 128-512 4 -8 1 20-320
5 (100) 5 (100) 4 (80)
(100) (80) (60) (100) (80) (100) (60)
R
1 (20) 2 (4 0 ) 1 (20) 2 (40) (2 Salmonella Typhimurium isolates)
1 (20) 5 (100)
3 (60) 5 (100)
2 (4 0 ) 5 (100)
a S, sensitive; I, intermediate; R, resistant. h From CLSI (6,7) and Soussy et al. (36).
Vazgecer et al. (39) did not detect Salmonella in 72 chicken doner kebab samples in Turkey. The isolation of Salmonella Typhi and Salmonella Paratyphi from a chicken shawirma and beef kubba sample, respectively, from two different restaurants in the current study is problematic because these organisms are linked to human reservoirs and contaminate food through poor personal hygiene or the use of unsafe water (27). In a 10-year prevalence study of Salmonella, L. monocytogenes, and E. coli 0157:H7 in RTE meat and poultry products in the United States, Levine et al. (23) reported Salmonella and L. monocytogenes prevalences similar to those found in the current study, and none of the tested samples were positive for E. coli 0157:H7. Meyer et al. (26) reported an L. monocytogenes prevalence of 3% in RTE poultry products in Germany. The presence of foodbome pathogens in RTE CB foods may be attributed to insufficient heat treatment during cooking or to postcooking contamination (14). The absence of E. coli 0157:H7 from the tested samples may indicate that the cooking step used in the restaurants was effective for eradicating this bacterium from the products but was not sufficient to eradicate the other two pathogens. E. coli 0157:H7 is a relatively heat-sensitive bacterium compared with Salmonella and L. monocytogenes (33). In the current study, Listeria species other than L. monocytogenes were isolated from RTE CB products, especially shawirma, at a very high rate. L. ivanovii, which is rarely considered pathogenic for humans, was the most common Listeria species isolated from the tested products.
Similar to these findings, Osaili et al. (29) found that L. ivanovii was the most common Listeria species isolated from RTE poultry products sold in Jordan; it was found in 67% of chicken shawirma samples. The high prevalence of Listeria in RTE CB products may be due to the ubiquitous nature of this microbe, which can contaminate products in slaughterhouses, before or after cooking, or through food contact surfaces and food handlers (12). In a parallel project (31), we tested two strains of Salmonella Typhimurium isolated from chicken shawirma for their thermal resistance in this product. These strains were more sensitive than expected at 54 to 64°C with D-values of 0.52 to 9.15 min. These results suggest that the risk of Salmonella in shawirma more likely results from inadequate cooking rather than from postcooking contamination, although the isolation of Salmonella Typhi from chicken shawirma in the current study indicates that postcooking contamination may contribute to the risk. Inadequately cooked shawirma itself may be a health risk to customers, and partially cooked slices could transfer Salmonella to cooked slices. Most Salmonella isolates in the current study were sensitive to the majority of antibiotics tested; however, resistance was noted, and some isolates were resistant to more than two antibiotics, especially the Salmonella Typhi and Salmonella Typhimurium isolates from chicken sha wirma. Similar resistance patterns have been previously reported for Salmonella. Harakeh et al. (15) found that 86 and 57% of the Salmonella isolates from meat-based fast food in Lebanon were resistant to trimethoprim-sulfameth oxazole and gentamicin, respectively. Thai et al. (37) found
no
J. Food Prot., Vol. 77, No. 1
OSAILI ET AL.
TABLE 4. MICs and antibiotic sensitivity of Listeria monocytogenes isolates from ready-to-eat chicken and beef products in Jordana MIC (pg/ml)
Sensitivity
Breakpoint*
Sample isolates, no. (%) of isolates L. monocytogenes ------------------------------------------------------
Antibiotic
S
R
Range
(ATCC 7644)
s
I
R
Benzylpenicillin Oxacillin Imipenem Gentamicin Ciprofloxacin Erythromycin Clindamycin Linezolid Teicoplanin Vancomycin Tetracycline Fosfomycin Fusidic acid Rifampin Trimethoprim-sulfamethoxazole
